diff options
691 files changed, 27532 insertions, 17383 deletions
diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml index 246bbb509bea..9923b065323b 100644 --- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml +++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml @@ -38,6 +38,10 @@ properties: - qcom,sm8450-dp - qcom,sm8550-dp - const: qcom,sm8350-dp + - items: + - enum: + - qcom,sm8750-dp + - const: qcom,sm8650-dp reg: minItems: 4 diff --git a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml index 82fe95a6d959..d4bb65c660af 100644 --- a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml +++ b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml @@ -42,6 +42,7 @@ properties: - qcom,sm8450-dsi-ctrl - qcom,sm8550-dsi-ctrl - qcom,sm8650-dsi-ctrl + - qcom,sm8750-dsi-ctrl - const: qcom,mdss-dsi-ctrl - enum: - qcom,dsi-ctrl-6g-qcm2290 @@ -70,11 +71,11 @@ properties: - mnoc:: MNOC clock - pixel:: Display pixel clock. minItems: 3 - maxItems: 9 + maxItems: 12 clock-names: minItems: 3 - maxItems: 9 + maxItems: 12 phys: maxItems: 1 @@ -109,7 +110,8 @@ properties: minItems: 2 maxItems: 4 description: | - Parents of "byte" and "pixel" for the given platform. + For DSI on SM8650 and older: parents of "byte" and "pixel" for the given + platform. For DSIv2 platforms this should contain "byte", "esc", "src" and "pixel_src" clocks. @@ -218,8 +220,6 @@ required: - clocks - clock-names - phys - - assigned-clocks - - assigned-clock-parents - ports allOf: @@ -244,6 +244,9 @@ allOf: - const: byte - const: pixel - const: core + required: + - assigned-clocks + - assigned-clock-parents - if: properties: @@ -266,6 +269,9 @@ allOf: - const: byte - const: pixel - const: core + required: + - assigned-clocks + - assigned-clock-parents - if: properties: @@ -288,6 +294,9 @@ allOf: - const: pixel - const: core - const: core_mmss + required: + - assigned-clocks + - assigned-clock-parents - if: properties: @@ -309,6 +318,9 @@ allOf: - const: core_mmss - const: pixel - const: core + required: + - assigned-clocks + - assigned-clock-parents - if: properties: @@ -346,6 +358,35 @@ allOf: - const: core - const: iface - const: bus + required: + - assigned-clocks + - assigned-clock-parents + + - if: + properties: + compatible: + contains: + enum: + - qcom,sm8750-dsi-ctrl + then: + properties: + clocks: + minItems: 12 + maxItems: 12 + clock-names: + items: + - const: byte + - const: byte_intf + - const: pixel + - const: core + - const: iface + - const: bus + - const: dsi_pll_pixel + - const: dsi_pll_byte + - const: esync + - const: osc + - const: byte_src + - const: pixel_src - if: properties: @@ -369,6 +410,9 @@ allOf: - const: core_mmss - const: pixel - const: core + required: + - assigned-clocks + - assigned-clock-parents unevaluatedProperties: false diff --git a/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml b/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml index 3c75ff42999a..1ca820a500b7 100644 --- a/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml +++ b/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml @@ -25,6 +25,7 @@ properties: - qcom,sm8450-dsi-phy-5nm - qcom,sm8550-dsi-phy-4nm - qcom,sm8650-dsi-phy-4nm + - qcom,sm8750-dsi-phy-3nm reg: items: diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml index 01cf79bd754b..0a46120dd868 100644 --- a/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml +++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml @@ -16,6 +16,7 @@ properties: enum: - qcom,sa8775p-dpu - qcom,sm8650-dpu + - qcom,sm8750-dpu - qcom,x1e80100-dpu reg: diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm8750-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm8750-mdss.yaml new file mode 100644 index 000000000000..72c70edc1fb0 --- /dev/null +++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8750-mdss.yaml @@ -0,0 +1,470 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/msm/qcom,sm8750-mdss.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm SM8750 Display MDSS + +maintainers: + - Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> + +description: + SM8650 MSM Mobile Display Subsystem(MDSS), which encapsulates sub-blocks like + DPU display controller, DSI and DP interfaces etc. + +$ref: /schemas/display/msm/mdss-common.yaml# + +properties: + compatible: + const: qcom,sm8750-mdss + + clocks: + items: + - description: Display AHB + - description: Display hf AXI + - description: Display core + + iommus: + maxItems: 1 + + interconnects: + items: + - description: Interconnect path from mdp0 port to the data bus + - description: Interconnect path from CPU to the reg bus + + interconnect-names: + items: + - const: mdp0-mem + - const: cpu-cfg + +patternProperties: + "^display-controller@[0-9a-f]+$": + type: object + additionalProperties: true + properties: + compatible: + const: qcom,sm8750-dpu + + "^displayport-controller@[0-9a-f]+$": + type: object + additionalProperties: true + properties: + compatible: + contains: + const: qcom,sm8750-dp + + "^dsi@[0-9a-f]+$": + type: object + additionalProperties: true + properties: + compatible: + contains: + const: qcom,sm8750-dsi-ctrl + + "^phy@[0-9a-f]+$": + type: object + additionalProperties: true + properties: + compatible: + const: qcom,sm8750-dsi-phy-3nm + +required: + - compatible + +unevaluatedProperties: false + +examples: + - | + #include <dt-bindings/clock/qcom,rpmh.h> + #include <dt-bindings/interconnect/qcom,icc.h> + #include <dt-bindings/interconnect/qcom,sm8750-rpmh.h> + #include <dt-bindings/interrupt-controller/arm-gic.h> + #include <dt-bindings/phy/phy-qcom-qmp.h> + #include <dt-bindings/power/qcom,rpmhpd.h> + + display-subsystem@ae00000 { + compatible = "qcom,sm8750-mdss"; + reg = <0x0ae00000 0x1000>; + reg-names = "mdss"; + + interrupts = <GIC_SPI 83 IRQ_TYPE_LEVEL_HIGH>; + + clocks = <&disp_cc_mdss_ahb_clk>, + <&gcc_disp_hf_axi_clk>, + <&disp_cc_mdss_mdp_clk>; + + interconnects = <&mmss_noc MASTER_MDP QCOM_ICC_TAG_ALWAYS + &mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ALWAYS>, + <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY + &config_noc SLAVE_DISPLAY_CFG QCOM_ICC_TAG_ACTIVE_ONLY>; + interconnect-names = "mdp0-mem", + "cpu-cfg"; + + resets = <&disp_cc_mdss_core_bcr>; + + power-domains = <&mdss_gdsc>; + + iommus = <&apps_smmu 0x800 0x2>; + + interrupt-controller; + #interrupt-cells = <1>; + + #address-cells = <1>; + #size-cells = <1>; + ranges; + + display-controller@ae01000 { + compatible = "qcom,sm8750-dpu"; + reg = <0x0ae01000 0x93000>, + <0x0aeb0000 0x2008>; + reg-names = "mdp", + "vbif"; + + interrupts-extended = <&mdss 0>; + + clocks = <&gcc_disp_hf_axi_clk>, + <&disp_cc_mdss_ahb_clk>, + <&disp_cc_mdss_mdp_lut_clk>, + <&disp_cc_mdss_mdp_clk>, + <&disp_cc_mdss_vsync_clk>; + clock-names = "nrt_bus", + "iface", + "lut", + "core", + "vsync"; + + assigned-clocks = <&disp_cc_mdss_vsync_clk>; + assigned-clock-rates = <19200000>; + + operating-points-v2 = <&mdp_opp_table>; + + power-domains = <&rpmhpd RPMHPD_MMCX>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + dpu_intf1_out: endpoint { + remote-endpoint = <&mdss_dsi0_in>; + }; + }; + + port@1 { + reg = <1>; + + dpu_intf2_out: endpoint { + remote-endpoint = <&mdss_dsi1_in>; + }; + }; + + port@2 { + reg = <2>; + + dpu_intf0_out: endpoint { + remote-endpoint = <&mdss_dp0_in>; + }; + }; + }; + + mdp_opp_table: opp-table { + compatible = "operating-points-v2"; + + opp-207000000 { + opp-hz = /bits/ 64 <207000000>; + required-opps = <&rpmhpd_opp_low_svs>; + }; + + opp-337000000 { + opp-hz = /bits/ 64 <337000000>; + required-opps = <&rpmhpd_opp_svs>; + }; + + opp-417000000 { + opp-hz = /bits/ 64 <417000000>; + required-opps = <&rpmhpd_opp_svs_l1>; + }; + + opp-532000000 { + opp-hz = /bits/ 64 <532000000>; + required-opps = <&rpmhpd_opp_nom>; + }; + + opp-575000000 { + opp-hz = /bits/ 64 <575000000>; + required-opps = <&rpmhpd_opp_nom_l1>; + }; + }; + }; + + dsi@ae94000 { + compatible = "qcom,sm8750-dsi-ctrl", "qcom,mdss-dsi-ctrl"; + reg = <0x0ae94000 0x400>; + reg-names = "dsi_ctrl"; + + interrupts-extended = <&mdss 4>; + + clocks = <&disp_cc_mdss_byte0_clk>, + <&disp_cc_mdss_byte0_intf_clk>, + <&disp_cc_mdss_pclk0_clk>, + <&disp_cc_mdss_esc0_clk>, + <&disp_cc_mdss_ahb_clk>, + <&gcc_disp_hf_axi_clk>, + <&mdss_dsi0_phy 1>, + <&mdss_dsi0_phy 0>, + <&disp_cc_esync0_clk>, + <&disp_cc_osc_clk>, + <&disp_cc_mdss_byte0_clk_src>, + <&disp_cc_mdss_pclk0_clk_src>; + clock-names = "byte", + "byte_intf", + "pixel", + "core", + "iface", + "bus", + "dsi_pll_pixel", + "dsi_pll_byte", + "esync", + "osc", + "byte_src", + "pixel_src"; + + operating-points-v2 = <&mdss_dsi_opp_table>; + + power-domains = <&rpmhpd RPMHPD_MMCX>; + + phys = <&mdss_dsi0_phy>; + phy-names = "dsi"; + + vdda-supply = <&vreg_l3g_1p2>; + + #address-cells = <1>; + #size-cells = <0>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + mdss_dsi0_in: endpoint { + remote-endpoint = <&dpu_intf1_out>; + }; + }; + + port@1 { + reg = <1>; + + mdss_dsi0_out: endpoint { + remote-endpoint = <&panel0_in>; + data-lanes = <0 1 2 3>; + }; + }; + }; + + mdss_dsi_opp_table: opp-table { + compatible = "operating-points-v2"; + + opp-187500000 { + opp-hz = /bits/ 64 <187500000>; + required-opps = <&rpmhpd_opp_low_svs>; + }; + + opp-300000000 { + opp-hz = /bits/ 64 <300000000>; + required-opps = <&rpmhpd_opp_svs>; + }; + + opp-358000000 { + opp-hz = /bits/ 64 <358000000>; + required-opps = <&rpmhpd_opp_svs_l1>; + }; + }; + }; + + mdss_dsi0_phy: phy@ae95000 { + compatible = "qcom,sm8750-dsi-phy-3nm"; + reg = <0x0ae95000 0x200>, + <0x0ae95200 0x280>, + <0x0ae95500 0x400>; + reg-names = "dsi_phy", + "dsi_phy_lane", + "dsi_pll"; + + clocks = <&disp_cc_mdss_ahb_clk>, + <&rpmhcc RPMH_CXO_CLK>; + clock-names = "iface", + "ref"; + + vdds-supply = <&vreg_l3i_0p88>; + + #clock-cells = <1>; + #phy-cells = <0>; + }; + + dsi@ae96000 { + compatible = "qcom,sm8750-dsi-ctrl", "qcom,mdss-dsi-ctrl"; + reg = <0x0ae96000 0x400>; + reg-names = "dsi_ctrl"; + + interrupts-extended = <&mdss 5>; + + clocks = <&disp_cc_mdss_byte1_clk>, + <&disp_cc_mdss_byte1_intf_clk>, + <&disp_cc_mdss_pclk1_clk>, + <&disp_cc_mdss_esc1_clk>, + <&disp_cc_mdss_ahb_clk>, + <&gcc_disp_hf_axi_clk>, + <&mdss_dsi1_phy 1>, + <&mdss_dsi1_phy 0>, + <&disp_cc_esync1_clk>, + <&disp_cc_osc_clk>, + <&disp_cc_mdss_byte1_clk_src>, + <&disp_cc_mdss_pclk1_clk_src>; + clock-names = "byte", + "byte_intf", + "pixel", + "core", + "iface", + "bus", + "dsi_pll_pixel", + "dsi_pll_byte", + "esync", + "osc", + "byte_src", + "pixel_src"; + + operating-points-v2 = <&mdss_dsi_opp_table>; + + power-domains = <&rpmhpd RPMHPD_MMCX>; + + phys = <&mdss_dsi1_phy>; + phy-names = "dsi"; + + #address-cells = <1>; + #size-cells = <0>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + mdss_dsi1_in: endpoint { + remote-endpoint = <&dpu_intf2_out>; + }; + }; + + port@1 { + reg = <1>; + + mdss_dsi1_out: endpoint { + }; + }; + }; + }; + + mdss_dsi1_phy: phy@ae97000 { + compatible = "qcom,sm8750-dsi-phy-3nm"; + reg = <0x0ae97000 0x200>, + <0x0ae97200 0x280>, + <0x0ae97500 0x400>; + reg-names = "dsi_phy", + "dsi_phy_lane", + "dsi_pll"; + + clocks = <&disp_cc_mdss_ahb_clk>, + <&rpmhcc RPMH_CXO_CLK>; + clock-names = "iface", + "ref"; + + #clock-cells = <1>; + #phy-cells = <0>; + }; + + displayport-controller@af54000 { + compatible = "qcom,sm8750-dp", "qcom,sm8650-dp"; + reg = <0xaf54000 0x104>, + <0xaf54200 0xc0>, + <0xaf55000 0x770>, + <0xaf56000 0x9c>, + <0xaf57000 0x9c>; + + interrupts-extended = <&mdss 12>; + + clocks = <&disp_cc_mdss_ahb_clk>, + <&disp_cc_mdss_dptx0_aux_clk>, + <&disp_cc_mdss_dptx0_link_clk>, + <&disp_cc_mdss_dptx0_link_intf_clk>, + <&disp_cc_mdss_dptx0_pixel0_clk>; + clock-names = "core_iface", + "core_aux", + "ctrl_link", + "ctrl_link_iface", + "stream_pixel"; + + assigned-clocks = <&disp_cc_mdss_dptx0_link_clk_src>, + <&disp_cc_mdss_dptx0_pixel0_clk_src>; + assigned-clock-parents = <&usb_dp_qmpphy QMP_USB43DP_DP_LINK_CLK>, + <&usb_dp_qmpphy QMP_USB43DP_DP_VCO_DIV_CLK>; + + operating-points-v2 = <&dp_opp_table>; + + power-domains = <&rpmhpd RPMHPD_MMCX>; + + phys = <&usb_dp_qmpphy QMP_USB43DP_DP_PHY>; + phy-names = "dp"; + + #sound-dai-cells = <0>; + + dp_opp_table: opp-table { + compatible = "operating-points-v2"; + + opp-192000000 { + opp-hz = /bits/ 64 <192000000>; + required-opps = <&rpmhpd_opp_low_svs_d1>; + }; + + opp-270000000 { + opp-hz = /bits/ 64 <270000000>; + required-opps = <&rpmhpd_opp_low_svs>; + }; + + opp-540000000 { + opp-hz = /bits/ 64 <540000000>; + required-opps = <&rpmhpd_opp_svs_l1>; + }; + + opp-810000000 { + opp-hz = /bits/ 64 <810000000>; + required-opps = <&rpmhpd_opp_nom>; + }; + }; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + mdss_dp0_in: endpoint { + remote-endpoint = <&dpu_intf0_out>; + }; + }; + + port@1 { + reg = <1>; + + mdss_dp0_out: endpoint { + remote-endpoint = <&usb_dp_qmpphy_dp_in>; + }; + }; + }; + }; + }; diff --git a/Documentation/devicetree/bindings/display/panel/himax,hx83112b.yaml b/Documentation/devicetree/bindings/display/panel/himax,hx83112b.yaml new file mode 100644 index 000000000000..e58bb3d45331 --- /dev/null +++ b/Documentation/devicetree/bindings/display/panel/himax,hx83112b.yaml @@ -0,0 +1,73 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/panel/himax,hx83112b.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Himax HX83112B-based DSI display panels + +maintainers: + - Luca Weiss <luca@lucaweiss.eu> + +description: + The Himax HX83112B is a generic DSI Panel IC used to control + LCD panels. + +allOf: + - $ref: panel-common.yaml# + +properties: + compatible: + contains: + const: djn,98-03057-6598b-i + + reg: + maxItems: 1 + + iovcc-supply: + description: I/O voltage rail + + vsn-supply: + description: Positive source voltage rail + + vsp-supply: + description: Negative source voltage rail + +required: + - compatible + - reg + - reset-gpios + - iovcc-supply + - vsn-supply + - vsp-supply + - port + +unevaluatedProperties: false + +examples: + - | + #include <dt-bindings/gpio/gpio.h> + + dsi { + #address-cells = <1>; + #size-cells = <0>; + + panel@0 { + compatible = "djn,98-03057-6598b-i"; + reg = <0>; + + reset-gpios = <&tlmm 61 GPIO_ACTIVE_LOW>; + + iovcc-supply = <&pm8953_l6>; + vsn-supply = <&pmi632_lcdb_ncp>; + vsp-supply = <&pmi632_lcdb_ldo>; + + port { + panel_in_0: endpoint { + remote-endpoint = <&dsi0_out>; + }; + }; + }; + }; + +... diff --git a/Documentation/devicetree/bindings/display/panel/raydium,rm67200.yaml b/Documentation/devicetree/bindings/display/panel/raydium,rm67200.yaml index 54c9c0ef45ec..97b7fbe05c07 100644 --- a/Documentation/devicetree/bindings/display/panel/raydium,rm67200.yaml +++ b/Documentation/devicetree/bindings/display/panel/raydium,rm67200.yaml @@ -42,7 +42,6 @@ required: - compatible - port - reg - - reset-gpios additionalProperties: false diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip-vop2.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip-vop2.yaml index f546d481b7e5..93da1fb9adc4 100644 --- a/Documentation/devicetree/bindings/display/rockchip/rockchip-vop2.yaml +++ b/Documentation/devicetree/bindings/display/rockchip/rockchip-vop2.yaml @@ -64,10 +64,10 @@ properties: - description: Pixel clock for video port 0. - description: Pixel clock for video port 1. - description: Pixel clock for video port 2. - - description: Pixel clock for video port 3. - - description: Peripheral(vop grf/dsi) clock. - - description: Alternative pixel clock provided by HDMI0 PHY PLL. - - description: Alternative pixel clock provided by HDMI1 PHY PLL. + - {} + - {} + - {} + - {} clock-names: minItems: 5 @@ -77,10 +77,10 @@ properties: - const: dclk_vp0 - const: dclk_vp1 - const: dclk_vp2 - - const: dclk_vp3 - - const: pclk_vop - - const: pll_hdmiphy0 - - const: pll_hdmiphy1 + - {} + - {} + - {} + - {} rockchip,grf: $ref: /schemas/types.yaml#/definitions/phandle @@ -175,10 +175,24 @@ allOf: then: properties: clocks: - maxItems: 5 + minItems: 5 + items: + - {} + - {} + - {} + - {} + - {} + - description: Alternative pixel clock provided by HDMI PHY PLL. clock-names: - maxItems: 5 + minItems: 5 + items: + - {} + - {} + - {} + - {} + - {} + - const: pll_hdmiphy0 interrupts: minItems: 4 @@ -208,11 +222,29 @@ allOf: properties: clocks: minItems: 7 - maxItems: 9 + items: + - {} + - {} + - {} + - {} + - {} + - description: Pixel clock for video port 3. + - description: Peripheral(vop grf/dsi) clock. + - description: Alternative pixel clock provided by HDMI0 PHY PLL. + - description: Alternative pixel clock provided by HDMI1 PHY PLL. clock-names: minItems: 7 - maxItems: 9 + items: + - {} + - {} + - {} + - {} + - {} + - const: dclk_vp3 + - const: pclk_vop + - const: pll_hdmiphy0 + - const: pll_hdmiphy1 interrupts: maxItems: 1 diff --git a/Documentation/devicetree/bindings/display/ti/ti,am625-oldi.yaml b/Documentation/devicetree/bindings/display/ti/ti,am625-oldi.yaml new file mode 100644 index 000000000000..8203ec5e5bb3 --- /dev/null +++ b/Documentation/devicetree/bindings/display/ti/ti,am625-oldi.yaml @@ -0,0 +1,79 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/ti/ti,am625-oldi.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Texas Instruments AM625 OLDI Transmitter + +maintainers: + - Tomi Valkeinen <tomi.valkeinen@ideasonboard.com> + - Aradhya Bhatia <aradhya.bhatia@linux.dev> + +description: + The AM625 TI Keystone OpenLDI transmitter (OLDI TX) supports serialized RGB + pixel data transmission between host and flat panel display over LVDS (Low + Voltage Differential Sampling) interface. The OLDI TX consists of 7-to-1 data + serializers, and 4-data and 1-clock LVDS outputs. It supports the LVDS output + formats "jeida-18", "jeida-24" and "vesa-18", and can accept 24-bit RGB or + padded and un-padded 18-bit RGB bus formats as input. + +properties: + reg: + maxItems: 1 + + clocks: + maxItems: 1 + description: serial clock input for the OLDI transmitters + + clock-names: + const: serial + + ti,companion-oldi: + $ref: /schemas/types.yaml#/definitions/phandle + description: + phandle to companion OLDI transmitter. This property is required for both + the OLDI TXes if they are expected to work either in dual-lvds mode or in + clone mode. This property should point to the other OLDI TX's phandle. + + ti,secondary-oldi: + type: boolean + description: + Boolean property to mark the OLDI transmitter as the secondary one, when the + OLDI hardware is expected to run as a companion HW, in cases of dual-lvds + mode or clone mode. The primary OLDI hardware is responsible for all the + hardware configuration. + + ti,oldi-io-ctrl: + $ref: /schemas/types.yaml#/definitions/phandle + description: + phandle to syscon device node mapping OLDI IO_CTRL registers found in the + control MMR region. These registers are required to toggle the I/O lane + power, and control its electrical characteristics. + + ports: + $ref: /schemas/graph.yaml#/properties/ports + + properties: + port@0: + $ref: /schemas/graph.yaml#/properties/port + description: Parallel RGB input port + + port@1: + $ref: /schemas/graph.yaml#/properties/port + description: LVDS output port + + required: + - port@0 + - port@1 + +required: + - reg + - clocks + - clock-names + - ti,oldi-io-ctrl + - ports + +additionalProperties: false + +... diff --git a/Documentation/devicetree/bindings/display/ti/ti,am65x-dss.yaml b/Documentation/devicetree/bindings/display/ti/ti,am65x-dss.yaml index a5b13cb7bc73..361e9cae6896 100644 --- a/Documentation/devicetree/bindings/display/ti/ti,am65x-dss.yaml +++ b/Documentation/devicetree/bindings/display/ti/ti,am65x-dss.yaml @@ -100,6 +100,24 @@ properties: For AM62A7 DSS, the port is tied off inside the SoC. For AM62L DSS, the DSS DPI output port node from video port 1 or DSI Tx controller node connected to video port 1. + properties: + endpoint@0: + $ref: /schemas/graph.yaml#/properties/endpoint + description: + For AM625 DSS, VP Connection to OLDI0. + For AM65X DSS, OLDI output from the SoC. + + endpoint@1: + $ref: /schemas/graph.yaml#/properties/endpoint + description: + For AM625 DSS, VP Connection to OLDI1. + + anyOf: + - required: + - endpoint + - required: + - endpoint@0 + - endpoint@1 port@1: $ref: /schemas/graph.yaml#/properties/port @@ -121,6 +139,25 @@ properties: Input memory (from main memory to dispc) bandwidth limit in bytes per second + oldi-transmitters: + description: + Child node under the DSS, to describe all the OLDI transmitters connected + to the DSS videoports. + type: object + additionalProperties: false + + properties: + "#address-cells": + const: 1 + + "#size-cells": + const: 0 + + patternProperties: + '^oldi@[0-1]$': + $ref: ti,am625-oldi.yaml# + description: OLDI transmitters connected to the DSS VPs + allOf: - if: properties: @@ -129,6 +166,7 @@ allOf: const: ti,am62a7-dss then: properties: + oldi-transmitters: false ports: properties: port@0: false @@ -143,6 +181,22 @@ allOf: properties: port@1: false + - if: + properties: + compatible: + contains: + enum: + - ti,am62l-dss + - ti,am65x-dss + then: + properties: + oldi-transmitters: false + ports: + properties: + port@0: + properties: + endpoint@1: false + required: - compatible - reg @@ -161,32 +215,135 @@ examples: #include <dt-bindings/soc/ti,sci_pm_domain.h> dss: dss@4a00000 { - compatible = "ti,am65x-dss"; - reg = <0x04a00000 0x1000>, /* common */ - <0x04a02000 0x1000>, /* vidl1 */ - <0x04a06000 0x1000>, /* vid */ - <0x04a07000 0x1000>, /* ovr1 */ - <0x04a08000 0x1000>, /* ovr2 */ - <0x04a0a000 0x1000>, /* vp1 */ - <0x04a0b000 0x1000>, /* vp2 */ - <0x04a01000 0x1000>; /* common1 */ + compatible = "ti,am65x-dss"; + reg = <0x04a00000 0x1000>, /* common */ + <0x04a02000 0x1000>, /* vidl1 */ + <0x04a06000 0x1000>, /* vid */ + <0x04a07000 0x1000>, /* ovr1 */ + <0x04a08000 0x1000>, /* ovr2 */ + <0x04a0a000 0x1000>, /* vp1 */ + <0x04a0b000 0x1000>, /* vp2 */ + <0x04a01000 0x1000>; /* common1 */ + reg-names = "common", "vidl1", "vid", + "ovr1", "ovr2", "vp1", "vp2", "common1"; + ti,am65x-oldi-io-ctrl = <&dss_oldi_io_ctrl>; + power-domains = <&k3_pds 67 TI_SCI_PD_EXCLUSIVE>; + clocks = <&k3_clks 67 1>, + <&k3_clks 216 1>, + <&k3_clks 67 2>; + clock-names = "fck", "vp1", "vp2"; + interrupts = <GIC_SPI 166 IRQ_TYPE_EDGE_RISING>; + ports { + #address-cells = <1>; + #size-cells = <0>; + port@0 { + reg = <0>; + oldi_out0: endpoint { + remote-endpoint = <&lcd_in0>; + }; + }; + }; + }; + + - | + #include <dt-bindings/interrupt-controller/arm-gic.h> + #include <dt-bindings/interrupt-controller/irq.h> + #include <dt-bindings/soc/ti,sci_pm_domain.h> + + bus { + #address-cells = <2>; + #size-cells = <2>; + dss1: dss@30200000 { + compatible = "ti,am625-dss"; + reg = <0x00 0x30200000 0x00 0x1000>, /* common */ + <0x00 0x30202000 0x00 0x1000>, /* vidl1 */ + <0x00 0x30206000 0x00 0x1000>, /* vid */ + <0x00 0x30207000 0x00 0x1000>, /* ovr1 */ + <0x00 0x30208000 0x00 0x1000>, /* ovr2 */ + <0x00 0x3020a000 0x00 0x1000>, /* vp1 */ + <0x00 0x3020b000 0x00 0x1000>, /* vp2 */ + <0x00 0x30201000 0x00 0x1000>; /* common1 */ reg-names = "common", "vidl1", "vid", - "ovr1", "ovr2", "vp1", "vp2", "common1"; - ti,am65x-oldi-io-ctrl = <&dss_oldi_io_ctrl>; - power-domains = <&k3_pds 67 TI_SCI_PD_EXCLUSIVE>; - clocks = <&k3_clks 67 1>, - <&k3_clks 216 1>, - <&k3_clks 67 2>; + "ovr1", "ovr2", "vp1", "vp2", "common1"; + power-domains = <&k3_pds 186 TI_SCI_PD_EXCLUSIVE>; + clocks = <&k3_clks 186 6>, + <&vp1_clock>, + <&k3_clks 186 2>; clock-names = "fck", "vp1", "vp2"; - interrupts = <GIC_SPI 166 IRQ_TYPE_EDGE_RISING>; + interrupts = <GIC_SPI 84 IRQ_TYPE_LEVEL_HIGH>; + oldi-transmitters { + #address-cells = <1>; + #size-cells = <0>; + oldi0: oldi@0 { + reg = <0>; + clocks = <&k3_clks 186 0>; + clock-names = "serial"; + ti,companion-oldi = <&oldi1>; + ti,oldi-io-ctrl = <&dss_oldi_io_ctrl>; + ports { + #address-cells = <1>; + #size-cells = <0>; + port@0 { + reg = <0>; + oldi0_in: endpoint { + remote-endpoint = <&dpi0_out0>; + }; + }; + port@1 { + reg = <1>; + oldi0_out: endpoint { + remote-endpoint = <&panel_in0>; + }; + }; + }; + }; + oldi1: oldi@1 { + reg = <1>; + clocks = <&k3_clks 186 0>; + clock-names = "serial"; + ti,secondary-oldi; + ti,companion-oldi = <&oldi0>; + ti,oldi-io-ctrl = <&dss_oldi_io_ctrl>; + ports { + #address-cells = <1>; + #size-cells = <0>; + port@0 { + reg = <0>; + oldi1_in: endpoint { + remote-endpoint = <&dpi0_out1>; + }; + }; + port@1 { + reg = <1>; + oldi1_out: endpoint { + remote-endpoint = <&panel_in1>; + }; + }; + }; + }; + }; ports { + #address-cells = <1>; + #size-cells = <0>; + port@0 { #address-cells = <1>; #size-cells = <0>; - port@0 { - reg = <0>; - oldi_out0: endpoint { - remote-endpoint = <&lcd_in0>; - }; + reg = <0>; + dpi0_out0: endpoint@0 { + reg = <0>; + remote-endpoint = <&oldi0_in>; + }; + dpi0_out1: endpoint@1 { + reg = <1>; + remote-endpoint = <&oldi1_in>; + }; + }; + port@1 { + reg = <1>; + dpi1_out: endpoint { + remote-endpoint = <&hdmi_bridge>; }; + }; }; + }; }; diff --git a/Documentation/devicetree/bindings/gpu/arm,mali-bifrost.yaml b/Documentation/devicetree/bindings/gpu/arm,mali-bifrost.yaml index b8d659d272d0..7e947016dee9 100644 --- a/Documentation/devicetree/bindings/gpu/arm,mali-bifrost.yaml +++ b/Documentation/devicetree/bindings/gpu/arm,mali-bifrost.yaml @@ -42,6 +42,7 @@ properties: - enum: - mediatek,mt8188-mali - mediatek,mt8192-mali + - mediatek,mt8370-mali - const: arm,mali-valhall-jm # Mali Valhall GPU model/revision is fully discoverable reg: @@ -225,7 +226,9 @@ allOf: properties: compatible: contains: - const: mediatek,mt8186-mali + enum: + - mediatek,mt8186-mali + - mediatek,mt8370-mali then: properties: power-domains: diff --git a/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml b/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml index a27ba7b663d4..0bd7d6b69755 100644 --- a/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml +++ b/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml @@ -23,7 +23,7 @@ properties: const: operating-points-v2-adreno patternProperties: - '^opp-[0-9]+$': + '^opp(-[0-9]+){1,2}$': type: object additionalProperties: false diff --git a/Documentation/devicetree/bindings/vendor-prefixes.yaml b/Documentation/devicetree/bindings/vendor-prefixes.yaml index 7a264a372789..9d5253369c0a 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.yaml +++ b/Documentation/devicetree/bindings/vendor-prefixes.yaml @@ -398,6 +398,8 @@ patternProperties: description: Diodes, Inc. "^dioo,.*": description: Dioo Microcircuit Co., Ltd + "^djn,.*": + description: Shenzhen DJN Optronics Technology Co., Ltd "^dlc,.*": description: DLC Display Co., Ltd. "^dlg,.*": diff --git a/Documentation/gpu/amdgpu/debugging.rst b/Documentation/gpu/amdgpu/debugging.rst index 7cbfea0606e1..ac914d524741 100644 --- a/Documentation/gpu/amdgpu/debugging.rst +++ b/Documentation/gpu/amdgpu/debugging.rst @@ -85,3 +85,21 @@ UMR GPU debugging and diagnostics tool. Please see the umr `documentation <https://umr.readthedocs.io/en/main/>`_ for more information about its capabilities. + +Debugging backlight brightness +============================== +Default backlight brightness is intended to be set via the policy advertised +by the firmware. Firmware will often provide different defaults for AC or DC. +Furthermore, some userspace software will save backlight brightness during +the previous boot and attempt to restore it. + +Some firmware also has support for a feature called "Custom Backlight Curves" +where an input value for brightness is mapped along a linearly interpolated +curve of brightness values that better match display characteristics. + +In the event of problems happening with backlight, there is a trace event +that can be enabled at bootup to log every brightness change request. +This can help isolate where the problem is. To enable the trace event add +the following to the kernel command line: + + tp_printk trace_event=amdgpu_dm:amdgpu_dm_brightness:mod:amdgpu trace_buf_size=1M diff --git a/Documentation/gpu/rfc/gpusvm.rst b/Documentation/gpu/rfc/gpusvm.rst index bcf66a8137a6..469db1372f16 100644 --- a/Documentation/gpu/rfc/gpusvm.rst +++ b/Documentation/gpu/rfc/gpusvm.rst @@ -74,14 +74,20 @@ Overview of baseline design :doc: Locking .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c - :doc: Migration - -.. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Partial Unmapping of Ranges .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Examples +Overview of drm_pagemap design +============================== + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Overview + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Migration + Possible future design features =============================== diff --git a/Documentation/userspace-api/dma-buf-heaps.rst b/Documentation/userspace-api/dma-buf-heaps.rst index 535f49047ce6..1dfe5e7acd5a 100644 --- a/Documentation/userspace-api/dma-buf-heaps.rst +++ b/Documentation/userspace-api/dma-buf-heaps.rst @@ -19,7 +19,10 @@ following heaps: - The ``cma`` heap allocates physically contiguous, cacheable, buffers. Only present if a CMA region is present. Such a region is usually created either through the kernel commandline through the - `cma` parameter, a memory region Device-Tree node with the - `linux,cma-default` property set, or through the `CMA_SIZE_MBYTES` or - `CMA_SIZE_PERCENTAGE` Kconfig options. Depending on the platform, it - might be called ``reserved``, ``linux,cma``, or ``default-pool``. + ``cma`` parameter, a memory region Device-Tree node with the + ``linux,cma-default`` property set, or through the ``CMA_SIZE_MBYTES`` or + ``CMA_SIZE_PERCENTAGE`` Kconfig options. The heap's name in devtmpfs is + ``default_cma_region``. For backwards compatibility, when the + ``DMABUF_HEAPS_CMA_LEGACY`` Kconfig option is set, a duplicate node is + created following legacy naming conventions; the legacy name might be + ``reserved``, ``linux,cma``, or ``default-pool``. diff --git a/MAINTAINERS b/MAINTAINERS index 7e7515a412e9..9d2f65f41726 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -7499,10 +7499,12 @@ M: Javier Martinez Canillas <javierm@redhat.com> L: dri-devel@lists.freedesktop.org S: Maintained T: git https://gitlab.freedesktop.org/drm/misc/kernel.git +F: drivers/firmware/sysfb*.c F: drivers/gpu/drm/sysfb/ F: drivers/video/aperture.c F: drivers/video/nomodeset.c F: include/linux/aperture.h +F: include/linux/sysfb.h F: include/video/nomodeset.h DRM DRIVER FOR GENERIC EDP PANELS @@ -8232,6 +8234,7 @@ M: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com> L: dri-devel@lists.freedesktop.org S: Maintained T: git https://gitlab.freedesktop.org/drm/misc/kernel.git +F: Documentation/devicetree/bindings/display/ti/ti,am625-oldi.yaml F: Documentation/devicetree/bindings/display/ti/ti,am65x-dss.yaml F: Documentation/devicetree/bindings/display/ti/ti,j721e-dss.yaml F: Documentation/devicetree/bindings/display/ti/ti,k2g-dss.yaml @@ -12086,6 +12089,13 @@ L: linux-kernel@vger.kernel.org S: Supported F: arch/x86/include/asm/intel-family.h +INTEL DISCRETE GRAPHICS NVM MTD DRIVER +M: Alexander Usyskin <alexander.usyskin@intel.com> +L: linux-mtd@lists.infradead.org +S: Supported +F: drivers/mtd/devices/mtd_intel_dg.c +F: include/linux/intel_dg_nvm_aux.h + INTEL DRM DISPLAY FOR XE AND I915 DRIVERS M: Jani Nikula <jani.nikula@linux.intel.com> M: Rodrigo Vivi <rodrigo.vivi@intel.com> diff --git a/arch/arm64/boot/dts/mediatek/mt8370.dtsi b/arch/arm64/boot/dts/mediatek/mt8370.dtsi index cf1a3759451f..7ac8b8d03494 100644 --- a/arch/arm64/boot/dts/mediatek/mt8370.dtsi +++ b/arch/arm64/boot/dts/mediatek/mt8370.dtsi @@ -59,6 +59,22 @@ <&cpu3 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>; }; +/* + * Please note that overriding compatibles is a discouraged practice and is a + * clear indication of nodes not being, well, compatible! + * + * This is a special case, where the GPU is the same as MT8188, but with one + * of the cores fused out in this lower-binned SoC. + */ +&gpu { + compatible = "mediatek,mt8370-mali", "arm,mali-valhall-jm"; + + power-domains = <&spm MT8188_POWER_DOMAIN_MFG2>, + <&spm MT8188_POWER_DOMAIN_MFG3>; + + power-domain-names = "core0", "core1"; +}; + &ppi_cluster0 { affinity = <&cpu0 &cpu1 &cpu2 &cpu3>; }; diff --git a/drivers/Kconfig b/drivers/Kconfig index 7c556c5ac4fd..c5edbd2288a1 100644 --- a/drivers/Kconfig +++ b/drivers/Kconfig @@ -209,8 +209,6 @@ source "drivers/thunderbolt/Kconfig" source "drivers/android/Kconfig" -source "drivers/gpu/trace/Kconfig" - source "drivers/nvdimm/Kconfig" source "drivers/dax/Kconfig" diff --git a/drivers/accel/drm_accel.c b/drivers/accel/drm_accel.c index aa826033b0ce..ca3357acd127 100644 --- a/drivers/accel/drm_accel.c +++ b/drivers/accel/drm_accel.c @@ -20,8 +20,6 @@ DEFINE_XARRAY_ALLOC(accel_minors_xa); -static struct dentry *accel_debugfs_root; - static const struct device_type accel_sysfs_device_minor = { .name = "accel_minor" }; @@ -74,17 +72,6 @@ static const struct drm_info_list accel_debugfs_list[] = { #define ACCEL_DEBUGFS_ENTRIES ARRAY_SIZE(accel_debugfs_list) /** - * accel_debugfs_init() - Initialize debugfs for device - * @dev: Pointer to the device instance. - * - * This function creates a root directory for the device in debugfs. - */ -void accel_debugfs_init(struct drm_device *dev) -{ - drm_debugfs_dev_init(dev, accel_debugfs_root); -} - -/** * accel_debugfs_register() - Register debugfs for device * @dev: Pointer to the device instance. * @@ -194,7 +181,6 @@ static const struct file_operations accel_stub_fops = { void accel_core_exit(void) { unregister_chrdev(ACCEL_MAJOR, "accel"); - debugfs_remove(accel_debugfs_root); accel_sysfs_destroy(); WARN_ON(!xa_empty(&accel_minors_xa)); } @@ -209,8 +195,6 @@ int __init accel_core_init(void) goto error; } - accel_debugfs_root = debugfs_create_dir("accel", NULL); - ret = register_chrdev(ACCEL_MAJOR, "accel", &accel_stub_fops); if (ret < 0) DRM_ERROR("Cannot register ACCEL major: %d\n", ret); diff --git a/drivers/dma-buf/heaps/Kconfig b/drivers/dma-buf/heaps/Kconfig index a5eef06c4226..bb369b38b001 100644 --- a/drivers/dma-buf/heaps/Kconfig +++ b/drivers/dma-buf/heaps/Kconfig @@ -12,3 +12,13 @@ config DMABUF_HEAPS_CMA Choose this option to enable dma-buf CMA heap. This heap is backed by the Contiguous Memory Allocator (CMA). If your system has these regions, you should say Y here. + +config DMABUF_HEAPS_CMA_LEGACY + bool "Legacy DMA-BUF CMA Heap" + default y + depends on DMABUF_HEAPS_CMA + help + Add a duplicate CMA-backed dma-buf heap with legacy naming derived + from the CMA area's devicetree node, or "reserved" if the area is not + defined in the devicetree. This uses the same underlying allocator as + CONFIG_DMABUF_HEAPS_CMA. diff --git a/drivers/dma-buf/heaps/cma_heap.c b/drivers/dma-buf/heaps/cma_heap.c index 9512d050563a..0df007111975 100644 --- a/drivers/dma-buf/heaps/cma_heap.c +++ b/drivers/dma-buf/heaps/cma_heap.c @@ -9,6 +9,9 @@ * Copyright (C) 2019 Texas Instruments Incorporated - http://www.ti.com/ * Andrew F. Davis <afd@ti.com> */ + +#define pr_fmt(fmt) "cma_heap: " fmt + #include <linux/cma.h> #include <linux/dma-buf.h> #include <linux/dma-heap.h> @@ -22,6 +25,7 @@ #include <linux/slab.h> #include <linux/vmalloc.h> +#define DEFAULT_CMA_NAME "default_cma_region" struct cma_heap { struct dma_heap *heap; @@ -366,17 +370,17 @@ static const struct dma_heap_ops cma_heap_ops = { .allocate = cma_heap_allocate, }; -static int __init __add_cma_heap(struct cma *cma, void *data) +static int __init __add_cma_heap(struct cma *cma, const char *name) { - struct cma_heap *cma_heap; struct dma_heap_export_info exp_info; + struct cma_heap *cma_heap; cma_heap = kzalloc(sizeof(*cma_heap), GFP_KERNEL); if (!cma_heap) return -ENOMEM; cma_heap->cma = cma; - exp_info.name = cma_get_name(cma); + exp_info.name = name; exp_info.ops = &cma_heap_ops; exp_info.priv = cma_heap; @@ -394,12 +398,30 @@ static int __init __add_cma_heap(struct cma *cma, void *data) static int __init add_default_cma_heap(void) { struct cma *default_cma = dev_get_cma_area(NULL); - int ret = 0; + const char *legacy_cma_name; + int ret; - if (default_cma) - ret = __add_cma_heap(default_cma, NULL); + if (!default_cma) + return 0; - return ret; + ret = __add_cma_heap(default_cma, DEFAULT_CMA_NAME); + if (ret) + return ret; + + if (IS_ENABLED(CONFIG_DMABUF_HEAPS_CMA_LEGACY)) { + legacy_cma_name = cma_get_name(default_cma); + if (!strcmp(legacy_cma_name, DEFAULT_CMA_NAME)) { + pr_warn("legacy name and default name are the same, skipping legacy heap\n"); + return 0; + } + + ret = __add_cma_heap(default_cma, legacy_cma_name); + if (ret) + pr_warn("failed to add legacy heap: %pe\n", + ERR_PTR(ret)); + } + + return 0; } module_init(add_default_cma_heap); MODULE_DESCRIPTION("DMA-BUF CMA Heap"); diff --git a/drivers/dma-buf/heaps/system_heap.c b/drivers/dma-buf/heaps/system_heap.c index 82b1b714300d..bbe7881f1360 100644 --- a/drivers/dma-buf/heaps/system_heap.c +++ b/drivers/dma-buf/heaps/system_heap.c @@ -33,7 +33,7 @@ struct system_heap_buffer { struct dma_heap_attachment { struct device *dev; - struct sg_table *table; + struct sg_table table; struct list_head list; bool mapped; }; @@ -52,29 +52,22 @@ static gfp_t order_flags[] = {HIGH_ORDER_GFP, HIGH_ORDER_GFP, LOW_ORDER_GFP}; static const unsigned int orders[] = {8, 4, 0}; #define NUM_ORDERS ARRAY_SIZE(orders) -static struct sg_table *dup_sg_table(struct sg_table *table) +static int dup_sg_table(struct sg_table *from, struct sg_table *to) { - struct sg_table *new_table; - int ret, i; struct scatterlist *sg, *new_sg; + int ret, i; - new_table = kzalloc(sizeof(*new_table), GFP_KERNEL); - if (!new_table) - return ERR_PTR(-ENOMEM); - - ret = sg_alloc_table(new_table, table->orig_nents, GFP_KERNEL); - if (ret) { - kfree(new_table); - return ERR_PTR(-ENOMEM); - } + ret = sg_alloc_table(to, from->orig_nents, GFP_KERNEL); + if (ret) + return ret; - new_sg = new_table->sgl; - for_each_sgtable_sg(table, sg, i) { + new_sg = to->sgl; + for_each_sgtable_sg(from, sg, i) { sg_set_page(new_sg, sg_page(sg), sg->length, sg->offset); new_sg = sg_next(new_sg); } - return new_table; + return 0; } static int system_heap_attach(struct dma_buf *dmabuf, @@ -82,19 +75,18 @@ static int system_heap_attach(struct dma_buf *dmabuf, { struct system_heap_buffer *buffer = dmabuf->priv; struct dma_heap_attachment *a; - struct sg_table *table; + int ret; a = kzalloc(sizeof(*a), GFP_KERNEL); if (!a) return -ENOMEM; - table = dup_sg_table(&buffer->sg_table); - if (IS_ERR(table)) { + ret = dup_sg_table(&buffer->sg_table, &a->table); + if (ret) { kfree(a); - return -ENOMEM; + return ret; } - a->table = table; a->dev = attachment->dev; INIT_LIST_HEAD(&a->list); a->mapped = false; @@ -118,8 +110,7 @@ static void system_heap_detach(struct dma_buf *dmabuf, list_del(&a->list); mutex_unlock(&buffer->lock); - sg_free_table(a->table); - kfree(a->table); + sg_free_table(&a->table); kfree(a); } @@ -127,7 +118,7 @@ static struct sg_table *system_heap_map_dma_buf(struct dma_buf_attachment *attac enum dma_data_direction direction) { struct dma_heap_attachment *a = attachment->priv; - struct sg_table *table = a->table; + struct sg_table *table = &a->table; int ret; ret = dma_map_sgtable(attachment->dev, table, direction, 0); @@ -162,7 +153,7 @@ static int system_heap_dma_buf_begin_cpu_access(struct dma_buf *dmabuf, list_for_each_entry(a, &buffer->attachments, list) { if (!a->mapped) continue; - dma_sync_sgtable_for_cpu(a->dev, a->table, direction); + dma_sync_sgtable_for_cpu(a->dev, &a->table, direction); } mutex_unlock(&buffer->lock); @@ -183,7 +174,7 @@ static int system_heap_dma_buf_end_cpu_access(struct dma_buf *dmabuf, list_for_each_entry(a, &buffer->attachments, list) { if (!a->mapped) continue; - dma_sync_sgtable_for_device(a->dev, a->table, direction); + dma_sync_sgtable_for_device(a->dev, &a->table, direction); } mutex_unlock(&buffer->lock); diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index 5050ac32bba2..4dafbdc8f86a 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -104,7 +104,11 @@ obj-$(CONFIG_DRM_PANEL_BACKLIGHT_QUIRKS) += drm_panel_backlight_quirks.o # obj-$(CONFIG_DRM_EXEC) += drm_exec.o obj-$(CONFIG_DRM_GPUVM) += drm_gpuvm.o -obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm.o + +drm_gpusvm_helper-y := \ + drm_gpusvm.o\ + drm_pagemap.o +obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm_helper.o obj-$(CONFIG_DRM_BUDDY) += drm_buddy.o diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 87080c06e5fc..930de203d533 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -66,7 +66,7 @@ amdgpu-y += amdgpu_device.o amdgpu_doorbell_mgr.o amdgpu_kms.o \ amdgpu_fw_attestation.o amdgpu_securedisplay.o \ amdgpu_eeprom.o amdgpu_mca.o amdgpu_psp_ta.o amdgpu_lsdma.o \ amdgpu_ring_mux.o amdgpu_xcp.o amdgpu_seq64.o amdgpu_aca.o amdgpu_dev_coredump.o \ - amdgpu_cper.o amdgpu_userq_fence.o amdgpu_eviction_fence.o + amdgpu_cper.o amdgpu_userq_fence.o amdgpu_eviction_fence.o amdgpu_ip.o amdgpu-$(CONFIG_PROC_FS) += amdgpu_fdinfo.o diff --git a/drivers/gpu/drm/amd/amdgpu/aldebaran.c b/drivers/gpu/drm/amd/amdgpu/aldebaran.c index e13fbd974141..9569dc16dd3d 100644 --- a/drivers/gpu/drm/amd/amdgpu/aldebaran.c +++ b/drivers/gpu/drm/amd/amdgpu/aldebaran.c @@ -71,18 +71,29 @@ aldebaran_get_reset_handler(struct amdgpu_reset_control *reset_ctl, return NULL; } +static inline uint32_t aldebaran_get_ip_block_mask(struct amdgpu_device *adev) +{ + uint32_t ip_block_mask = BIT(AMD_IP_BLOCK_TYPE_GFX) | + BIT(AMD_IP_BLOCK_TYPE_SDMA); + + if (adev->aid_mask) + ip_block_mask |= BIT(AMD_IP_BLOCK_TYPE_IH); + + return ip_block_mask; +} + static int aldebaran_mode2_suspend_ip(struct amdgpu_device *adev) { + uint32_t ip_block_mask = aldebaran_get_ip_block_mask(adev); + uint32_t ip_block; int r, i; amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); for (i = adev->num_ip_blocks - 1; i >= 0; i--) { - if (!(adev->ip_blocks[i].version->type == - AMD_IP_BLOCK_TYPE_GFX || - adev->ip_blocks[i].version->type == - AMD_IP_BLOCK_TYPE_SDMA)) + ip_block = BIT(adev->ip_blocks[i].version->type); + if (!(ip_block_mask & ip_block)) continue; r = amdgpu_ip_block_suspend(&adev->ip_blocks[i]); @@ -200,8 +211,10 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl, static int aldebaran_mode2_restore_ip(struct amdgpu_device *adev) { struct amdgpu_firmware_info *ucode_list[AMDGPU_UCODE_ID_MAXIMUM]; + uint32_t ip_block_mask = aldebaran_get_ip_block_mask(adev); struct amdgpu_firmware_info *ucode; struct amdgpu_ip_block *cmn_block; + struct amdgpu_ip_block *ih_block; int ucode_count = 0; int i, r; @@ -243,6 +256,18 @@ static int aldebaran_mode2_restore_ip(struct amdgpu_device *adev) if (r) return r; + if (ip_block_mask & BIT(AMD_IP_BLOCK_TYPE_IH)) { + ih_block = amdgpu_device_ip_get_ip_block(adev, + AMD_IP_BLOCK_TYPE_IH); + if (unlikely(!ih_block)) { + dev_err(adev->dev, "Failed to get IH handle\n"); + return -EINVAL; + } + r = amdgpu_ip_block_resume(ih_block); + if (r) + return r; + } + /* Reinit GFXHUB */ adev->gfxhub.funcs->init(adev); r = adev->gfxhub.funcs->gart_enable(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index a5ccd0ada16a..7edb7ba91786 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -470,9 +470,6 @@ struct amdgpu_sa_manager { void *cpu_ptr; }; -int amdgpu_fence_slab_init(void); -void amdgpu_fence_slab_fini(void); - /* * IRQS. */ @@ -1282,6 +1279,7 @@ struct amdgpu_device { bool debug_exp_resets; bool debug_disable_gpu_ring_reset; bool debug_vm_userptr; + bool debug_disable_ce_logs; /* Protection for the following isolation structure */ struct mutex enforce_isolation_mutex; @@ -1336,6 +1334,11 @@ static inline struct amdgpu_device *amdgpu_ttm_adev(struct ttm_device *bdev) return container_of(bdev, struct amdgpu_device, mman.bdev); } +static inline bool amdgpu_is_multi_aid(struct amdgpu_device *adev) +{ + return !!adev->aid_mask; +} + int amdgpu_device_init(struct amdgpu_device *adev, uint32_t flags); void amdgpu_device_fini_hw(struct amdgpu_device *adev); @@ -1387,7 +1390,8 @@ void amdgpu_device_indirect_wreg64(struct amdgpu_device *adev, void amdgpu_device_indirect_wreg64_ext(struct amdgpu_device *adev, u64 reg_addr, u64 reg_data); u32 amdgpu_device_get_rev_id(struct amdgpu_device *adev); -bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type); +bool amdgpu_device_asic_has_dc_support(struct pci_dev *pdev, + enum amd_asic_type asic_type); bool amdgpu_device_has_dc_support(struct amdgpu_device *adev); void amdgpu_device_set_sriov_virtual_display(struct amdgpu_device *adev); @@ -1619,6 +1623,7 @@ void amdgpu_driver_release_kms(struct drm_device *dev); int amdgpu_device_ip_suspend(struct amdgpu_device *adev); int amdgpu_device_prepare(struct drm_device *dev); +void amdgpu_device_complete(struct drm_device *dev); int amdgpu_device_suspend(struct drm_device *dev, bool fbcon); int amdgpu_device_resume(struct drm_device *dev, bool fbcon); u32 amdgpu_get_vblank_counter_kms(struct drm_crtc *crtc); @@ -1760,4 +1765,19 @@ extern const struct attribute_group amdgpu_flash_attr_group; void amdgpu_set_init_level(struct amdgpu_device *adev, enum amdgpu_init_lvl_id lvl); + +static inline int amdgpu_device_bus_status_check(struct amdgpu_device *adev) +{ + u32 status; + int r; + + r = pci_read_config_dword(adev->pdev, PCI_COMMAND, &status); + if (r || PCI_POSSIBLE_ERROR(status)) { + dev_err(adev->dev, "device lost from bus!"); + return -ENODEV; + } + + return 0; +} + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c index 3835f2592914..cbc40cad581b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c @@ -115,6 +115,11 @@ static void aca_smu_bank_dump(struct amdgpu_device *adev, int idx, int total, st u64 event_id = qctx ? qctx->evid.event_id : RAS_EVENT_INVALID_ID; int i; + if (adev->debug_disable_ce_logs && + bank->smu_err_type == ACA_SMU_TYPE_CE && + !ACA_BANK_ERR_IS_DEFFERED(bank)) + return; + RAS_EVENT_LOG(adev, event_id, HW_ERR "Accelerator Check Architecture events logged\n"); /* plus 1 for output format, e.g: ACA[08/08]: xxxx */ for (i = 0; i < ARRAY_SIZE(aca_regs); i++) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 55161e5cdc30..fbe7616555c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -248,18 +248,34 @@ void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev, kgd2kfd_interrupt(adev->kfd.dev, ih_ring_entry); } -void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm) +void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool suspend_proc) { if (adev->kfd.dev) - kgd2kfd_suspend(adev->kfd.dev, run_pm); + kgd2kfd_suspend(adev->kfd.dev, suspend_proc); } -int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm) +int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool resume_proc) { int r = 0; if (adev->kfd.dev) - r = kgd2kfd_resume(adev->kfd.dev, run_pm); + r = kgd2kfd_resume(adev->kfd.dev, resume_proc); + + return r; +} + +void amdgpu_amdkfd_suspend_process(struct amdgpu_device *adev) +{ + if (adev->kfd.dev) + kgd2kfd_suspend_process(adev->kfd.dev); +} + +int amdgpu_amdkfd_resume_process(struct amdgpu_device *adev) +{ + int r = 0; + + if (adev->kfd.dev) + r = kgd2kfd_resume_process(adev->kfd.dev); return r; } @@ -749,12 +765,12 @@ int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev, int amdgpu_amdkfd_check_and_lock_kfd(struct amdgpu_device *adev) { - return kgd2kfd_check_and_lock_kfd(); + return kgd2kfd_check_and_lock_kfd(adev->kfd.dev); } void amdgpu_amdkfd_unlock_kfd(struct amdgpu_device *adev) { - kgd2kfd_unlock_kfd(); + kgd2kfd_unlock_kfd(adev->kfd.dev); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index b6ca41859b53..33eb4826b58b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -154,8 +154,10 @@ struct amdkfd_process_info { int amdgpu_amdkfd_init(void); void amdgpu_amdkfd_fini(void); -void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm); -int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm); +void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool suspend_proc); +int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool resume_proc); +void amdgpu_amdkfd_suspend_process(struct amdgpu_device *adev); +int amdgpu_amdkfd_resume_process(struct amdgpu_device *adev); void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev, const void *ih_ring_entry); void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev); @@ -411,16 +413,18 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf); bool kgd2kfd_device_init(struct kfd_dev *kfd, const struct kgd2kfd_shared_resources *gpu_resources); void kgd2kfd_device_exit(struct kfd_dev *kfd); -void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm); -int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm); +void kgd2kfd_suspend(struct kfd_dev *kfd, bool suspend_proc); +int kgd2kfd_resume(struct kfd_dev *kfd, bool resume_proc); +void kgd2kfd_suspend_process(struct kfd_dev *kfd); +int kgd2kfd_resume_process(struct kfd_dev *kfd); int kgd2kfd_pre_reset(struct kfd_dev *kfd, struct amdgpu_reset_context *reset_context); int kgd2kfd_post_reset(struct kfd_dev *kfd); void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry); void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd); void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask); -int kgd2kfd_check_and_lock_kfd(void); -void kgd2kfd_unlock_kfd(void); +int kgd2kfd_check_and_lock_kfd(struct kfd_dev *kfd); +void kgd2kfd_unlock_kfd(struct kfd_dev *kfd); int kgd2kfd_start_sched(struct kfd_dev *kfd, uint32_t node_id); int kgd2kfd_stop_sched(struct kfd_dev *kfd, uint32_t node_id); bool kgd2kfd_compute_active(struct kfd_dev *kfd, uint32_t node_id); @@ -454,11 +458,20 @@ static inline void kgd2kfd_device_exit(struct kfd_dev *kfd) { } -static inline void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) +static inline void kgd2kfd_suspend(struct kfd_dev *kfd, bool suspend_proc) { } -static inline int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) +static inline int kgd2kfd_resume(struct kfd_dev *kfd, bool resume_proc) +{ + return 0; +} + +static inline void kgd2kfd_suspend_process(struct kfd_dev *kfd) +{ +} + +static inline int kgd2kfd_resume_process(struct kfd_dev *kfd) { return 0; } @@ -489,12 +502,12 @@ void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask) { } -static inline int kgd2kfd_check_and_lock_kfd(void) +static inline int kgd2kfd_check_and_lock_kfd(struct kfd_dev *kfd) { return 0; } -static inline void kgd2kfd_unlock_kfd(void) +static inline void kgd2kfd_unlock_kfd(struct kfd_dev *kfd) { } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c index ffbaa8bc5eea..1105a09e55dc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c @@ -320,7 +320,7 @@ static void set_barrier_auto_waitcnt(struct amdgpu_device *adev, bool enable_wai if (!down_read_trylock(&adev->reset_domain->sem)) return; - amdgpu_amdkfd_suspend(adev, false); + amdgpu_amdkfd_suspend(adev, true); if (suspend_resume_compute_scheduler(adev, true)) goto out; @@ -333,7 +333,7 @@ static void set_barrier_auto_waitcnt(struct amdgpu_device *adev, bool enable_wai out: suspend_resume_compute_scheduler(adev, false); - amdgpu_amdkfd_resume(adev, false); + amdgpu_amdkfd_resume(adev, true); up_read(&adev->reset_domain->sem); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c index ca4a6b82817f..df77558e03ef 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -561,6 +561,13 @@ static uint32_t read_vmid_from_vmfault_reg(struct amdgpu_device *adev) return REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, VMID); } +static uint32_t kgd_hqd_sdma_get_doorbell(struct amdgpu_device *adev, + int engine, int queue) + +{ + return 0; +} + const struct kfd2kgd_calls gfx_v7_kfd2kgd = { .program_sh_mem_settings = kgd_program_sh_mem_settings, .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping, @@ -578,4 +585,5 @@ const struct kfd2kgd_calls gfx_v7_kfd2kgd = { .set_scratch_backing_va = set_scratch_backing_va, .set_vm_context_page_table_base = set_vm_context_page_table_base, .read_vmid_from_vmfault_reg = read_vmid_from_vmfault_reg, + .hqd_sdma_get_doorbell = kgd_hqd_sdma_get_doorbell, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c index 0f3e2944edd7..e68c0fa8d751 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c @@ -582,6 +582,13 @@ static void set_vm_context_page_table_base(struct amdgpu_device *adev, lower_32_bits(page_table_base)); } +static uint32_t kgd_hqd_sdma_get_doorbell(struct amdgpu_device *adev, + int engine, int queue) + +{ + return 0; +} + const struct kfd2kgd_calls gfx_v8_kfd2kgd = { .program_sh_mem_settings = kgd_program_sh_mem_settings, .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping, @@ -599,4 +606,5 @@ const struct kfd2kgd_calls gfx_v8_kfd2kgd = { get_atc_vmid_pasid_mapping_info, .set_scratch_backing_va = set_scratch_backing_va, .set_vm_context_page_table_base = set_vm_context_page_table_base, + .hqd_sdma_get_doorbell = kgd_hqd_sdma_get_doorbell, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c index 5a234eadae8b..15dde1f50328 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c @@ -212,7 +212,7 @@ int amdgpu_cper_entry_fill_bad_page_threshold_section(struct amdgpu_device *adev NONSTD_SEC_OFFSET(hdr->sec_cnt, idx)); amdgpu_cper_entry_fill_section_desc(adev, section_desc, true, false, - CPER_SEV_NUM, RUNTIME, NONSTD_SEC_LEN, + CPER_SEV_FATAL, RUNTIME, NONSTD_SEC_LEN, NONSTD_SEC_OFFSET(hdr->sec_cnt, idx)); section->hdr.valid_bits.err_info_cnt = 1; @@ -326,7 +326,9 @@ int amdgpu_cper_generate_bp_threshold_record(struct amdgpu_device *adev) return -ENOMEM; } - amdgpu_cper_entry_fill_hdr(adev, bp_threshold, AMDGPU_CPER_TYPE_BP_THRESHOLD, CPER_SEV_NUM); + amdgpu_cper_entry_fill_hdr(adev, bp_threshold, + AMDGPU_CPER_TYPE_BP_THRESHOLD, + CPER_SEV_FATAL); ret = amdgpu_cper_entry_fill_bad_page_threshold_section(adev, bp_threshold, 0); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 85567d0d9545..f5d5c45ddc0d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -944,6 +944,7 @@ static void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr) drm_sched_entity_fini(entity); } } + kref_put(&ctx->refcount, amdgpu_ctx_fini); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index dac4b926e7be..0e6e2e2acf5b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -1902,7 +1902,7 @@ no_preempt: continue; } job = to_amdgpu_job(s_job); - if (preempted && (&job->hw_fence) == fence) + if (preempted && (&job->hw_fence.base) == fence) /* mark the job as preempted */ job->preemption_status |= AMDGPU_IB_PREEMPTED; } @@ -2131,6 +2131,55 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) return 0; } +static int amdgpu_pt_info_read(struct seq_file *m, void *unused) +{ + struct drm_file *file; + struct amdgpu_fpriv *fpriv; + struct amdgpu_bo *root_bo; + int r; + + file = m->private; + if (!file) + return -EINVAL; + + fpriv = file->driver_priv; + if (!fpriv || !fpriv->vm.root.bo) + return -ENODEV; + + root_bo = amdgpu_bo_ref(fpriv->vm.root.bo); + r = amdgpu_bo_reserve(root_bo, true); + if (r) { + amdgpu_bo_unref(&root_bo); + return -EINVAL; + } + + seq_printf(m, "gpu_address: 0x%llx\n", amdgpu_bo_gpu_offset(fpriv->vm.root.bo)); + + amdgpu_bo_unreserve(root_bo); + amdgpu_bo_unref(&root_bo); + + return 0; +} + +static int amdgpu_pt_info_open(struct inode *inode, struct file *file) +{ + return single_open(file, amdgpu_pt_info_read, inode->i_private); +} + +static const struct file_operations amdgpu_pt_info_fops = { + .owner = THIS_MODULE, + .open = amdgpu_pt_info_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +void amdgpu_debugfs_vm_init(struct drm_file *file) +{ + debugfs_create_file("vm_pagetable_info", 0444, file->debugfs_client, file, + &amdgpu_pt_info_fops); +} + #else int amdgpu_debugfs_init(struct amdgpu_device *adev) { @@ -2140,4 +2189,7 @@ int amdgpu_debugfs_regs_init(struct amdgpu_device *adev) { return 0; } +void amdgpu_debugfs_vm_init(struct drm_file *file) +{ +} #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h index 0425432d8659..e7b3c38e5186 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.h @@ -33,4 +33,5 @@ void amdgpu_debugfs_fence_init(struct amdgpu_device *adev); void amdgpu_debugfs_firmware_init(struct amdgpu_device *adev); void amdgpu_debugfs_gem_init(struct amdgpu_device *adev); void amdgpu_debugfs_mes_event_log_init(struct amdgpu_device *adev); +void amdgpu_debugfs_vm_init(struct drm_file *file); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index a59f194e3360..d282c0753b14 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -232,7 +232,7 @@ static int amdgpu_device_attr_sysfs_init(struct amdgpu_device *adev) { int ret = 0; - if (!amdgpu_sriov_vf(adev)) + if (amdgpu_nbio_is_replay_cnt_supported(adev)) ret = sysfs_create_file(&adev->dev->kobj, &dev_attr_pcie_replay_count.attr); @@ -241,7 +241,7 @@ static int amdgpu_device_attr_sysfs_init(struct amdgpu_device *adev) static void amdgpu_device_attr_sysfs_fini(struct amdgpu_device *adev) { - if (!amdgpu_sriov_vf(adev)) + if (amdgpu_nbio_is_replay_cnt_supported(adev)) sysfs_remove_file(&adev->dev->kobj, &dev_attr_pcie_replay_count.attr); } @@ -1288,14 +1288,14 @@ u32 amdgpu_device_get_rev_id(struct amdgpu_device *adev) */ static uint32_t amdgpu_invalid_rreg(struct amdgpu_device *adev, uint32_t reg) { - DRM_ERROR("Invalid callback to read register 0x%04X\n", reg); + dev_err(adev->dev, "Invalid callback to read register 0x%04X\n", reg); BUG(); return 0; } static uint32_t amdgpu_invalid_rreg_ext(struct amdgpu_device *adev, uint64_t reg) { - DRM_ERROR("Invalid callback to read register 0x%llX\n", reg); + dev_err(adev->dev, "Invalid callback to read register 0x%llX\n", reg); BUG(); return 0; } @@ -1312,15 +1312,17 @@ static uint32_t amdgpu_invalid_rreg_ext(struct amdgpu_device *adev, uint64_t reg */ static void amdgpu_invalid_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v) { - DRM_ERROR("Invalid callback to write register 0x%04X with 0x%08X\n", - reg, v); + dev_err(adev->dev, + "Invalid callback to write register 0x%04X with 0x%08X\n", reg, + v); BUG(); } static void amdgpu_invalid_wreg_ext(struct amdgpu_device *adev, uint64_t reg, uint32_t v) { - DRM_ERROR("Invalid callback to write register 0x%llX with 0x%08X\n", - reg, v); + dev_err(adev->dev, + "Invalid callback to write register 0x%llX with 0x%08X\n", reg, + v); BUG(); } @@ -1336,14 +1338,15 @@ static void amdgpu_invalid_wreg_ext(struct amdgpu_device *adev, uint64_t reg, ui */ static uint64_t amdgpu_invalid_rreg64(struct amdgpu_device *adev, uint32_t reg) { - DRM_ERROR("Invalid callback to read 64 bit register 0x%04X\n", reg); + dev_err(adev->dev, "Invalid callback to read 64 bit register 0x%04X\n", + reg); BUG(); return 0; } static uint64_t amdgpu_invalid_rreg64_ext(struct amdgpu_device *adev, uint64_t reg) { - DRM_ERROR("Invalid callback to read register 0x%llX\n", reg); + dev_err(adev->dev, "Invalid callback to read register 0x%llX\n", reg); BUG(); return 0; } @@ -1360,15 +1363,17 @@ static uint64_t amdgpu_invalid_rreg64_ext(struct amdgpu_device *adev, uint64_t r */ static void amdgpu_invalid_wreg64(struct amdgpu_device *adev, uint32_t reg, uint64_t v) { - DRM_ERROR("Invalid callback to write 64 bit register 0x%04X with 0x%08llX\n", - reg, v); + dev_err(adev->dev, + "Invalid callback to write 64 bit register 0x%04X with 0x%08llX\n", + reg, v); BUG(); } static void amdgpu_invalid_wreg64_ext(struct amdgpu_device *adev, uint64_t reg, uint64_t v) { - DRM_ERROR("Invalid callback to write 64 bit register 0x%llX with 0x%08llX\n", - reg, v); + dev_err(adev->dev, + "Invalid callback to write 64 bit register 0x%llX with 0x%08llX\n", + reg, v); BUG(); } @@ -1386,8 +1391,9 @@ static void amdgpu_invalid_wreg64_ext(struct amdgpu_device *adev, uint64_t reg, static uint32_t amdgpu_block_invalid_rreg(struct amdgpu_device *adev, uint32_t block, uint32_t reg) { - DRM_ERROR("Invalid callback to read register 0x%04X in block 0x%04X\n", - reg, block); + dev_err(adev->dev, + "Invalid callback to read register 0x%04X in block 0x%04X\n", + reg, block); BUG(); return 0; } @@ -1407,8 +1413,9 @@ static void amdgpu_block_invalid_wreg(struct amdgpu_device *adev, uint32_t block, uint32_t reg, uint32_t v) { - DRM_ERROR("Invalid block callback to write register 0x%04X in block 0x%04X with 0x%08X\n", - reg, block, v); + dev_err(adev->dev, + "Invalid block callback to write register 0x%04X in block 0x%04X with 0x%08X\n", + reg, block, v); BUG(); } @@ -1694,7 +1701,9 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev) /* PCI_EXT_CAP_ID_VNDR extended capability is located at 0x100 */ if (!pci_find_ext_capability(adev->pdev, PCI_EXT_CAP_ID_VNDR)) - DRM_WARN("System can't access extended configuration space, please check!!\n"); + dev_warn( + adev->dev, + "System can't access extended configuration space, please check!!\n"); /* skip if the bios has already enabled large BAR */ if (adev->gmc.real_vram_size && @@ -1734,9 +1743,10 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev) r = pci_resize_resource(adev->pdev, 0, rbar_size); if (r == -ENOSPC) - DRM_INFO("Not enough PCI address space for a large BAR."); + dev_info(adev->dev, + "Not enough PCI address space for a large BAR."); else if (r && r != -ENOTSUPP) - DRM_ERROR("Problem resizing BAR0 (%d).", r); + dev_err(adev->dev, "Problem resizing BAR0 (%d).", r); pci_assign_unassigned_bus_resources(adev->pdev->bus); @@ -1838,8 +1848,8 @@ bool amdgpu_device_seamless_boot_supported(struct amdgpu_device *adev) case 0: return false; default: - DRM_ERROR("Invalid value for amdgpu.seamless: %d\n", - amdgpu_seamless); + dev_err(adev->dev, "Invalid value for amdgpu.seamless: %d\n", + amdgpu_seamless); return false; } @@ -2015,7 +2025,7 @@ static void amdgpu_device_check_smu_prv_buffer_size(struct amdgpu_device *adev) return; if (!is_os_64) { - DRM_WARN("Not 64-bit OS, feature not supported\n"); + dev_warn(adev->dev, "Not 64-bit OS, feature not supported\n"); goto def_value; } si_meminfo(&si); @@ -2030,7 +2040,7 @@ static void amdgpu_device_check_smu_prv_buffer_size(struct amdgpu_device *adev) if (total_memory < dram_size_seven_GB) goto def_value1; } else { - DRM_WARN("Smu memory pool size not supported\n"); + dev_warn(adev->dev, "Smu memory pool size not supported\n"); goto def_value; } adev->pm.smu_prv_buffer_size = amdgpu_smu_memory_pool_size << 28; @@ -2038,7 +2048,7 @@ static void amdgpu_device_check_smu_prv_buffer_size(struct amdgpu_device *adev) return; def_value1: - DRM_WARN("No enough system memory\n"); + dev_warn(adev->dev, "No enough system memory\n"); def_value: adev->pm.smu_prv_buffer_size = 0; } @@ -2202,12 +2212,13 @@ static void amdgpu_switcheroo_set_state(struct pci_dev *pdev, amdgpu_device_load_pci_state(pdev); r = pci_enable_device(pdev); if (r) - DRM_WARN("pci_enable_device failed (%d)\n", r); + dev_warn(&pdev->dev, "pci_enable_device failed (%d)\n", + r); amdgpu_device_resume(dev, true); dev->switch_power_state = DRM_SWITCH_POWER_ON; } else { - pr_info("switched off\n"); + dev_info(&pdev->dev, "switched off\n"); dev->switch_power_state = DRM_SWITCH_POWER_CHANGING; amdgpu_device_prepare(dev); amdgpu_device_suspend(dev, true); @@ -2274,8 +2285,9 @@ int amdgpu_device_ip_set_clockgating_state(void *dev, r = adev->ip_blocks[i].version->funcs->set_clockgating_state( &adev->ip_blocks[i], state); if (r) - DRM_ERROR("set_clockgating_state of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "set_clockgating_state of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); } return r; } @@ -2308,8 +2320,9 @@ int amdgpu_device_ip_set_powergating_state(void *dev, r = adev->ip_blocks[i].version->funcs->set_powergating_state( &adev->ip_blocks[i], state); if (r) - DRM_ERROR("set_powergating_state of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "set_powergating_state of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); } return r; } @@ -2525,9 +2538,11 @@ static void amdgpu_device_enable_virtual_display(struct amdgpu_device *adev) } } - DRM_INFO("virtual display string:%s, %s:virtual_display:%d, num_crtc:%d\n", - amdgpu_virtual_display, pci_address_name, - adev->enable_virtual_display, adev->mode_info.num_crtc); + dev_info( + adev->dev, + "virtual display string:%s, %s:virtual_display:%d, num_crtc:%d\n", + amdgpu_virtual_display, pci_address_name, + adev->enable_virtual_display, adev->mode_info.num_crtc); kfree(pciaddstr); } @@ -2538,8 +2553,9 @@ void amdgpu_device_set_sriov_virtual_display(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev) && !adev->enable_virtual_display) { adev->mode_info.num_crtc = 1; adev->enable_virtual_display = true; - DRM_INFO("virtual_display:%d, num_crtc:%d\n", - adev->enable_virtual_display, adev->mode_info.num_crtc); + dev_info(adev->dev, "virtual_display:%d, num_crtc:%d\n", + adev->enable_virtual_display, + adev->mode_info.num_crtc); } } @@ -2773,21 +2789,29 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) if (!amdgpu_device_pcie_dynamic_switching_supported(adev)) adev->pm.pp_feature &= ~PP_PCIE_DPM_MASK; + adev->virt.is_xgmi_node_migrate_enabled = false; + if (amdgpu_sriov_vf(adev)) { + adev->virt.is_xgmi_node_migrate_enabled = + amdgpu_ip_version((adev), GC_HWIP, 0) == IP_VERSION(9, 4, 4); + } + total = true; for (i = 0; i < adev->num_ip_blocks; i++) { ip_block = &adev->ip_blocks[i]; if ((amdgpu_ip_block_mask & (1 << i)) == 0) { - DRM_WARN("disabled ip block: %d <%s>\n", - i, adev->ip_blocks[i].version->funcs->name); + dev_warn(adev->dev, "disabled ip block: %d <%s>\n", i, + adev->ip_blocks[i].version->funcs->name); adev->ip_blocks[i].status.valid = false; } else if (ip_block->version->funcs->early_init) { r = ip_block->version->funcs->early_init(ip_block); if (r == -ENOENT) { adev->ip_blocks[i].status.valid = false; } else if (r) { - DRM_ERROR("early_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "early_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); total = false; } else { adev->ip_blocks[i].status.valid = true; @@ -2868,8 +2892,10 @@ static int amdgpu_device_ip_hw_init_phase1(struct amdgpu_device *adev) adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH) { r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("hw_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "hw_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); return r; } adev->ip_blocks[i].status.hw = true; @@ -2893,8 +2919,9 @@ static int amdgpu_device_ip_hw_init_phase2(struct amdgpu_device *adev) continue; r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("hw_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "hw_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); return r; } adev->ip_blocks[i].status.hw = true; @@ -2932,8 +2959,11 @@ static int amdgpu_device_fw_loading(struct amdgpu_device *adev) } else { r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("hw_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "hw_init of IP block <%s> failed %d\n", + adev->ip_blocks[i] + .version->funcs->name, + r); return r; } adev->ip_blocks[i].status.hw = true; @@ -2988,25 +3018,29 @@ static int amdgpu_device_init_schedulers(struct amdgpu_device *adev) r = drm_sched_init(&ring->sched, &args); if (r) { - DRM_ERROR("Failed to create scheduler on ring %s.\n", - ring->name); + dev_err(adev->dev, + "Failed to create scheduler on ring %s.\n", + ring->name); return r; } r = amdgpu_uvd_entity_init(adev, ring); if (r) { - DRM_ERROR("Failed to create UVD scheduling entity on ring %s.\n", - ring->name); + dev_err(adev->dev, + "Failed to create UVD scheduling entity on ring %s.\n", + ring->name); return r; } r = amdgpu_vce_entity_init(adev, ring); if (r) { - DRM_ERROR("Failed to create VCE scheduling entity on ring %s.\n", - ring->name); + dev_err(adev->dev, + "Failed to create VCE scheduling entity on ring %s.\n", + ring->name); return r; } } - amdgpu_xcp_update_partition_sched_list(adev); + if (adev->xcp_mgr) + amdgpu_xcp_update_partition_sched_list(adev); return 0; } @@ -3038,8 +3072,10 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->funcs->sw_init) { r = adev->ip_blocks[i].version->funcs->sw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("sw_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "sw_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); goto init_failed; } } @@ -3053,7 +3089,8 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) /* need to do common hw init early so everything is set up for gmc */ r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("hw_init %d failed %d\n", i, r); + dev_err(adev->dev, "hw_init %d failed %d\n", i, + r); goto init_failed; } adev->ip_blocks[i].status.hw = true; @@ -3065,17 +3102,21 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) r = amdgpu_device_mem_scratch_init(adev); if (r) { - DRM_ERROR("amdgpu_mem_scratch_init failed %d\n", r); + dev_err(adev->dev, + "amdgpu_mem_scratch_init failed %d\n", + r); goto init_failed; } r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("hw_init %d failed %d\n", i, r); + dev_err(adev->dev, "hw_init %d failed %d\n", i, + r); goto init_failed; } r = amdgpu_device_wb_init(adev); if (r) { - DRM_ERROR("amdgpu_device_wb_init failed %d\n", r); + dev_err(adev->dev, + "amdgpu_device_wb_init failed %d\n", r); goto init_failed; } adev->ip_blocks[i].status.hw = true; @@ -3087,14 +3128,16 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) AMDGPU_GEM_DOMAIN_GTT, AMDGPU_CSA_SIZE); if (r) { - DRM_ERROR("allocate CSA failed %d\n", r); + dev_err(adev->dev, + "allocate CSA failed %d\n", r); goto init_failed; } } r = amdgpu_seq64_init(adev); if (r) { - DRM_ERROR("allocate seq64 failed %d\n", r); + dev_err(adev->dev, "allocate seq64 failed %d\n", + r); goto init_failed; } } @@ -3284,8 +3327,10 @@ int amdgpu_device_set_cg_state(struct amdgpu_device *adev, r = adev->ip_blocks[i].version->funcs->set_clockgating_state(&adev->ip_blocks[i], state); if (r) { - DRM_ERROR("set_clockgating_state(gate) of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "set_clockgating_state(gate) of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); return r; } } @@ -3321,8 +3366,10 @@ int amdgpu_device_set_pg_state(struct amdgpu_device *adev, r = adev->ip_blocks[i].version->funcs->set_powergating_state(&adev->ip_blocks[i], state); if (r) { - DRM_ERROR("set_powergating_state(gate) of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "set_powergating_state(gate) of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); return r; } } @@ -3388,8 +3435,10 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->funcs->late_init) { r = adev->ip_blocks[i].version->funcs->late_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("late_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "late_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); return r; } } @@ -3398,7 +3447,7 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) r = amdgpu_ras_late_init(adev); if (r) { - DRM_ERROR("amdgpu_ras_late_init failed %d", r); + dev_err(adev->dev, "amdgpu_ras_late_init failed %d", r); return r; } @@ -3412,7 +3461,7 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) r = amdgpu_device_enable_mgpu_fan_boost(); if (r) - DRM_ERROR("enable mgpu fan boost failed (%d).\n", r); + dev_err(adev->dev, "enable mgpu fan boost failed (%d).\n", r); /* For passthrough configuration on arcturus and aldebaran, enable special handling SBR */ if (amdgpu_passthrough(adev) && @@ -3445,7 +3494,9 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) r = amdgpu_xgmi_set_pstate(gpu_instance->adev, AMDGPU_XGMI_PSTATE_MIN); if (r) { - DRM_ERROR("pstate setting failed (%d).\n", r); + dev_err(adev->dev, + "pstate setting failed (%d).\n", + r); break; } } @@ -3459,17 +3510,19 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) static void amdgpu_ip_block_hw_fini(struct amdgpu_ip_block *ip_block) { + struct amdgpu_device *adev = ip_block->adev; int r; if (!ip_block->version->funcs->hw_fini) { - DRM_ERROR("hw_fini of IP block <%s> not defined\n", - ip_block->version->funcs->name); + dev_err(adev->dev, "hw_fini of IP block <%s> not defined\n", + ip_block->version->funcs->name); } else { r = ip_block->version->funcs->hw_fini(ip_block); /* XXX handle errors */ if (r) { - DRM_DEBUG("hw_fini of IP block <%s> failed %d\n", - ip_block->version->funcs->name, r); + dev_dbg(adev->dev, + "hw_fini of IP block <%s> failed %d\n", + ip_block->version->funcs->name, r); } } @@ -3510,15 +3563,16 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev) r = adev->ip_blocks[i].version->funcs->early_fini(&adev->ip_blocks[i]); if (r) { - DRM_DEBUG("early_fini of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_dbg(adev->dev, + "early_fini of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); } } amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); - amdgpu_amdkfd_suspend(adev, false); + amdgpu_amdkfd_suspend(adev, true); amdgpu_userq_suspend(adev); /* Workaround for ASICs need to disable SMC first */ @@ -3533,7 +3587,8 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev)) { if (amdgpu_virt_release_full_gpu(adev, false)) - DRM_ERROR("failed to release exclusive mode on fini\n"); + dev_err(adev->dev, + "failed to release exclusive mode on fini\n"); } return 0; @@ -3581,8 +3636,10 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) r = adev->ip_blocks[i].version->funcs->sw_fini(&adev->ip_blocks[i]); /* XXX handle errors */ if (r) { - DRM_DEBUG("sw_fini of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_dbg(adev->dev, + "sw_fini of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); } } adev->ip_blocks[i].status.sw = false; @@ -3615,7 +3672,7 @@ static void amdgpu_device_delayed_init_work_handler(struct work_struct *work) r = amdgpu_ib_ring_tests(adev); if (r) - DRM_ERROR("ib ring test failed (%d).\n", r); + dev_err(adev->dev, "ib ring test failed (%d).\n", r); } static void amdgpu_device_delay_enable_gfx_off(struct work_struct *work) @@ -3756,8 +3813,9 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { r = amdgpu_dpm_set_mp1_state(adev, adev->mp1_state); if (r) { - DRM_ERROR("SMC failed to set mp1 state %d, %d\n", - adev->mp1_state, r); + dev_err(adev->dev, + "SMC failed to set mp1 state %d, %d\n", + adev->mp1_state, r); return r; } } @@ -4041,12 +4099,14 @@ static void amdgpu_device_detect_sriov_bios(struct amdgpu_device *adev) /** * amdgpu_device_asic_has_dc_support - determine if DC supports the asic * + * @pdev : pci device context * @asic_type: AMD asic type * * Check if there is DC (new modesetting infrastructre) support for an asic. * returns true if DC has support, false if not. */ -bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type) +bool amdgpu_device_asic_has_dc_support(struct pci_dev *pdev, + enum amd_asic_type asic_type) { switch (asic_type) { #ifdef CONFIG_DRM_AMDGPU_SI @@ -4089,7 +4149,9 @@ bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type) #else default: if (amdgpu_dc > 0) - DRM_INFO_ONCE("Display Core has been requested via kernel parameter but isn't supported by ASIC, ignoring\n"); + dev_info_once( + &pdev->dev, + "Display Core has been requested via kernel parameter but isn't supported by ASIC, ignoring\n"); return false; #endif } @@ -4108,7 +4170,7 @@ bool amdgpu_device_has_dc_support(struct amdgpu_device *adev) (adev->harvest_ip_mask & AMD_HARVEST_IP_DMU_MASK)) return false; - return amdgpu_device_asic_has_dc_support(adev->asic_type); + return amdgpu_device_asic_has_dc_support(adev->pdev, adev->asic_type); } static void amdgpu_device_xgmi_reset_func(struct work_struct *__work) @@ -4150,7 +4212,8 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work) fail: if (adev->asic_reset_res) - DRM_WARN("ASIC reset failed with error, %d for drm dev, %s", + dev_warn(adev->dev, + "ASIC reset failed with error, %d for drm dev, %s", adev->asic_reset_res, adev_to_drm(adev)->unique); amdgpu_put_xgmi_hive(hive); } @@ -4274,7 +4337,7 @@ static void amdgpu_device_set_mcbp(struct amdgpu_device *adev) adev->gfx.mcbp = true; if (adev->gfx.mcbp) - DRM_INFO("MCBP is enabled\n"); + dev_info(adev->dev, "MCBP is enabled\n"); } /** @@ -4342,9 +4405,11 @@ int amdgpu_device_init(struct amdgpu_device *adev, adev->audio_endpt_rreg = &amdgpu_block_invalid_rreg; adev->audio_endpt_wreg = &amdgpu_block_invalid_wreg; - DRM_INFO("initializing kernel modesetting (%s 0x%04X:0x%04X 0x%04X:0x%04X 0x%02X).\n", - amdgpu_asic_name[adev->asic_type], pdev->vendor, pdev->device, - pdev->subsystem_vendor, pdev->subsystem_device, pdev->revision); + dev_info( + adev->dev, + "initializing kernel modesetting (%s 0x%04X:0x%04X 0x%04X:0x%04X 0x%02X).\n", + amdgpu_asic_name[adev->asic_type], pdev->vendor, pdev->device, + pdev->subsystem_vendor, pdev->subsystem_device, pdev->revision); /* mutex initialization are all done here so we * can recall function without having locking issues @@ -4461,8 +4526,10 @@ int amdgpu_device_init(struct amdgpu_device *adev, if (!adev->rmmio) return -ENOMEM; - DRM_INFO("register mmio base: 0x%08X\n", (uint32_t)adev->rmmio_base); - DRM_INFO("register mmio size: %u\n", (unsigned int)adev->rmmio_size); + dev_info(adev->dev, "register mmio base: 0x%08X\n", + (uint32_t)adev->rmmio_base); + dev_info(adev->dev, "register mmio size: %u\n", + (unsigned int)adev->rmmio_size); /* * Reset domain needs to be present early, before XGMI hive discovered @@ -4599,7 +4666,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, r = -EINVAL; goto failed; } - DRM_INFO("GPU posting now...\n"); + dev_info(adev->dev, "GPU posting now...\n"); r = amdgpu_device_asic_init(adev); if (r) { dev_err(adev->dev, "gpu post error!\n"); @@ -4709,12 +4776,12 @@ fence_driver_init: r = amdgpu_pm_sysfs_init(adev); if (r) - DRM_ERROR("registering pm sysfs failed (%d).\n", r); + dev_err(adev->dev, "registering pm sysfs failed (%d).\n", r); r = amdgpu_ucode_sysfs_init(adev); if (r) { adev->ucode_sysfs_en = false; - DRM_ERROR("Creating firmware sysfs failed (%d).\n", r); + dev_err(adev->dev, "Creating firmware sysfs failed (%d).\n", r); } else adev->ucode_sysfs_en = true; @@ -4963,7 +5030,7 @@ static int amdgpu_device_evict_resources(struct amdgpu_device *adev) ret = amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM); if (ret) - DRM_WARN("evicting device resources failed\n"); + dev_warn(adev->dev, "evicting device resources failed\n"); return ret; } @@ -5035,6 +5102,28 @@ int amdgpu_device_prepare(struct drm_device *dev) } /** + * amdgpu_device_complete - complete power state transition + * + * @dev: drm dev pointer + * + * Undo the changes from amdgpu_device_prepare. This will be + * called on all resume transitions, including those that failed. + */ +void amdgpu_device_complete(struct drm_device *dev) +{ + struct amdgpu_device *adev = drm_to_adev(dev); + int i; + + for (i = 0; i < adev->num_ip_blocks; i++) { + if (!adev->ip_blocks[i].status.valid) + continue; + if (!adev->ip_blocks[i].version->funcs->complete) + continue; + adev->ip_blocks[i].version->funcs->complete(&adev->ip_blocks[i]); + } +} + +/** * amdgpu_device_suspend - initiate device suspend * * @dev: drm dev pointer @@ -5055,6 +5144,8 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients) adev->in_suspend = true; if (amdgpu_sriov_vf(adev)) { + if (!adev->in_s0ix && !adev->in_runpm) + amdgpu_amdkfd_suspend_process(adev); amdgpu_virt_fini_data_exchange(adev); r = amdgpu_virt_request_full_gpu(adev, false); if (r) @@ -5062,7 +5153,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients) } if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D3)) - DRM_WARN("smart shift update failed\n"); + dev_warn(adev->dev, "smart shift update failed\n"); if (notify_clients) drm_client_dev_suspend(adev_to_drm(adev), false); @@ -5074,7 +5165,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients) amdgpu_device_ip_suspend_phase1(adev); if (!adev->in_s0ix) { - amdgpu_amdkfd_suspend(adev, adev->in_runpm); + amdgpu_amdkfd_suspend(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm); amdgpu_userq_suspend(adev); } @@ -5098,6 +5189,32 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients) return 0; } +static inline int amdgpu_virt_resume(struct amdgpu_device *adev) +{ + int r; + unsigned int prev_physical_node_id = adev->gmc.xgmi.physical_node_id; + + /* During VM resume, QEMU programming of VF MSIX table (register GFXMSIX_VECT0_ADDR_LO) + * may not work. The access could be blocked by nBIF protection as VF isn't in + * exclusive access mode. Exclusive access is enabled now, disable/enable MSIX + * so that QEMU reprograms MSIX table. + */ + amdgpu_restore_msix(adev); + + r = adev->gfxhub.funcs->get_xgmi_info(adev); + if (r) + return r; + + dev_info(adev->dev, "xgmi node, old id %d, new id %d\n", + prev_physical_node_id, adev->gmc.xgmi.physical_node_id); + + adev->vm_manager.vram_base_offset = adev->gfxhub.funcs->get_mc_fb_offset(adev); + adev->vm_manager.vram_base_offset += + adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; + + return 0; +} + /** * amdgpu_device_resume - initiate device resume * @@ -5119,6 +5236,12 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients) return r; } + if (amdgpu_virt_xgmi_migrate_enabled(adev)) { + r = amdgpu_virt_resume(adev); + if (r) + goto exit; + } + if (dev->switch_power_state == DRM_SWITCH_POWER_OFF) return 0; @@ -5140,7 +5263,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients) } if (!adev->in_s0ix) { - r = amdgpu_amdkfd_resume(adev, adev->in_runpm); + r = amdgpu_amdkfd_resume(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm); if (r) goto exit; @@ -5159,6 +5282,9 @@ exit: if (amdgpu_sriov_vf(adev)) { amdgpu_virt_init_data_exchange(adev); amdgpu_virt_release_full_gpu(adev, true); + + if (!adev->in_s0ix && !r && !adev->in_runpm) + r = amdgpu_amdkfd_resume_process(adev); } if (r) @@ -5196,7 +5322,7 @@ exit: adev->in_suspend = false; if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D0)) - DRM_WARN("smart shift update failed\n"); + dev_warn(adev->dev, "smart shift update failed\n"); return 0; } @@ -5727,7 +5853,9 @@ int amdgpu_device_reinit_after_reset(struct amdgpu_reset_context *reset_context) amdgpu_coredump(tmp_adev, false, vram_lost, reset_context->job); if (vram_lost) { - DRM_INFO("VRAM is lost due to GPU reset!\n"); + dev_info( + tmp_adev->dev, + "VRAM is lost due to GPU reset!\n"); amdgpu_inc_vram_lost(tmp_adev); } @@ -6006,29 +6134,20 @@ static int amdgpu_device_health_check(struct list_head *device_list_handle) { struct amdgpu_device *tmp_adev; int ret = 0; - u32 status; list_for_each_entry(tmp_adev, device_list_handle, reset_list) { - pci_read_config_dword(tmp_adev->pdev, PCI_COMMAND, &status); - if (PCI_POSSIBLE_ERROR(status)) { - dev_err(tmp_adev->dev, "device lost from bus!"); - ret = -ENODEV; - } + ret |= amdgpu_device_bus_status_check(tmp_adev); } return ret; } -static int amdgpu_device_halt_activities(struct amdgpu_device *adev, - struct amdgpu_job *job, - struct amdgpu_reset_context *reset_context, - struct list_head *device_list, - struct amdgpu_hive_info *hive, - bool need_emergency_restart) +static int amdgpu_device_recovery_prepare(struct amdgpu_device *adev, + struct list_head *device_list, + struct amdgpu_hive_info *hive) { - struct list_head *device_list_handle = NULL; struct amdgpu_device *tmp_adev = NULL; - int i, r = 0; + int r; /* * Build list of devices to reset. @@ -6045,26 +6164,55 @@ static int amdgpu_device_halt_activities(struct amdgpu_device *adev, } if (!list_is_first(&adev->reset_list, device_list)) list_rotate_to_front(&adev->reset_list, device_list); - device_list_handle = device_list; } else { list_add_tail(&adev->reset_list, device_list); - device_list_handle = device_list; } if (!amdgpu_sriov_vf(adev) && (!adev->pcie_reset_ctx.occurs_dpc)) { - r = amdgpu_device_health_check(device_list_handle); + r = amdgpu_device_health_check(device_list); if (r) return r; } - /* We need to lock reset domain only once both for XGMI and single device */ - tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device, - reset_list); + return 0; +} + +static void amdgpu_device_recovery_get_reset_lock(struct amdgpu_device *adev, + struct list_head *device_list) +{ + struct amdgpu_device *tmp_adev = NULL; + + if (list_empty(device_list)) + return; + tmp_adev = + list_first_entry(device_list, struct amdgpu_device, reset_list); amdgpu_device_lock_reset_domain(tmp_adev->reset_domain); +} - /* block all schedulers and reset given job's ring */ - list_for_each_entry(tmp_adev, device_list_handle, reset_list) { +static void amdgpu_device_recovery_put_reset_lock(struct amdgpu_device *adev, + struct list_head *device_list) +{ + struct amdgpu_device *tmp_adev = NULL; + + if (list_empty(device_list)) + return; + tmp_adev = + list_first_entry(device_list, struct amdgpu_device, reset_list); + amdgpu_device_unlock_reset_domain(tmp_adev->reset_domain); +} +static void amdgpu_device_halt_activities(struct amdgpu_device *adev, + struct amdgpu_job *job, + struct amdgpu_reset_context *reset_context, + struct list_head *device_list, + struct amdgpu_hive_info *hive, + bool need_emergency_restart) +{ + struct amdgpu_device *tmp_adev = NULL; + int i; + + /* block all schedulers and reset given job's ring */ + list_for_each_entry(tmp_adev, device_list, reset_list) { amdgpu_device_set_mp1_state(tmp_adev); /* @@ -6113,8 +6261,6 @@ static int amdgpu_device_halt_activities(struct amdgpu_device *adev, } atomic_inc(&tmp_adev->gpu_reset_counter); } - - return r; } static int amdgpu_device_asic_reset(struct amdgpu_device *adev, @@ -6220,7 +6366,8 @@ static int amdgpu_device_sched_resume(struct list_head *device_list, } else { dev_info(tmp_adev->dev, "GPU reset(%d) succeeded!\n", atomic_read(&tmp_adev->gpu_reset_counter)); if (amdgpu_acpi_smart_shift_update(adev_to_drm(tmp_adev), AMDGPU_SS_DEV_D0)) - DRM_WARN("smart shift update failed\n"); + dev_warn(tmp_adev->dev, + "smart shift update failed\n"); } } @@ -6252,11 +6399,6 @@ static void amdgpu_device_gpu_resume(struct amdgpu_device *adev, amdgpu_ras_set_error_query_ready(tmp_adev, true); } - - tmp_adev = list_first_entry(device_list, struct amdgpu_device, - reset_list); - amdgpu_device_unlock_reset_domain(tmp_adev->reset_domain); - } @@ -6306,7 +6448,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, */ if (need_emergency_restart && amdgpu_ras_get_context(adev) && amdgpu_ras_get_context(adev)->reboot) { - DRM_WARN("Emergency reboot."); + dev_warn(adev->dev, "Emergency reboot."); ksys_sync_helper(); emergency_restart(); @@ -6324,11 +6466,14 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, reset_context->hive = hive; INIT_LIST_HEAD(&device_list); - r = amdgpu_device_halt_activities(adev, job, reset_context, &device_list, - hive, need_emergency_restart); - if (r) + if (amdgpu_device_recovery_prepare(adev, &device_list, hive)) goto end_reset; + /* We need to lock reset domain only once both for XGMI and single device */ + amdgpu_device_recovery_get_reset_lock(adev, &device_list); + + amdgpu_device_halt_activities(adev, job, reset_context, &device_list, + hive, need_emergency_restart); if (need_emergency_restart) goto skip_sched_resume; /* @@ -6337,7 +6482,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, * * job->base holds a reference to parent fence */ - if (job && dma_fence_is_signaled(&job->hw_fence)) { + if (job && dma_fence_is_signaled(&job->hw_fence.base)) { job_signaled = true; dev_info(adev->dev, "Guilty job already signaled, skipping HW reset"); goto skip_hw_reset; @@ -6345,13 +6490,15 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, r = amdgpu_device_asic_reset(adev, &device_list, reset_context); if (r) - goto end_reset; + goto reset_unlock; skip_hw_reset: r = amdgpu_device_sched_resume(&device_list, reset_context, job_signaled); if (r) - goto end_reset; + goto reset_unlock; skip_sched_resume: amdgpu_device_gpu_resume(adev, &device_list, need_emergency_restart); +reset_unlock: + amdgpu_device_recovery_put_reset_lock(adev, &device_list); end_reset: if (hive) { mutex_unlock(&hive->hive_lock); @@ -6747,7 +6894,6 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev); struct amdgpu_reset_context reset_context; struct list_head device_list; - int r = 0; dev_info(adev->dev, "PCI error: detected callback!!\n"); @@ -6772,14 +6918,14 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta memset(&reset_context, 0, sizeof(reset_context)); INIT_LIST_HEAD(&device_list); - r = amdgpu_device_halt_activities(adev, NULL, &reset_context, &device_list, - hive, false); + amdgpu_device_recovery_prepare(adev, &device_list, hive); + amdgpu_device_recovery_get_reset_lock(adev, &device_list); + amdgpu_device_halt_activities(adev, NULL, &reset_context, &device_list, + hive, false); if (hive) { mutex_unlock(&hive->hive_lock); amdgpu_put_xgmi_hive(hive); } - if (r) - return PCI_ERS_RESULT_DISCONNECT; return PCI_ERS_RESULT_NEED_RESET; case pci_channel_io_perm_failure: /* Permanent error, prepare for device removal */ @@ -6889,8 +7035,8 @@ out: if (hive) { list_for_each_entry(tmp_adev, &device_list, reset_list) amdgpu_device_unset_mp1_state(tmp_adev); - amdgpu_device_unlock_reset_domain(adev->reset_domain); } + amdgpu_device_recovery_put_reset_lock(adev, &device_list); } if (hive) { @@ -6936,6 +7082,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev) amdgpu_device_sched_resume(&device_list, NULL, NULL); amdgpu_device_gpu_resume(adev, &device_list, false); + amdgpu_device_recovery_put_reset_lock(adev, &device_list); adev->pcie_reset_ctx.occurs_dpc = false; if (hive) { @@ -6960,11 +7107,11 @@ bool amdgpu_device_cache_pci_state(struct pci_dev *pdev) adev->pci_state = pci_store_saved_state(pdev); if (!adev->pci_state) { - DRM_ERROR("Failed to store PCI saved state"); + dev_err(adev->dev, "Failed to store PCI saved state"); return false; } } else { - DRM_WARN("Failed to save PCI state, err:%d\n", r); + dev_warn(adev->dev, "Failed to save PCI state, err:%d\n", r); return false; } @@ -6985,7 +7132,7 @@ bool amdgpu_device_load_pci_state(struct pci_dev *pdev) if (!r) { pci_restore_state(pdev); } else { - DRM_WARN("Failed to load PCI state, err:%d\n", r); + dev_warn(adev->dev, "Failed to load PCI state, err:%d\n", r); return false; } @@ -7231,7 +7378,7 @@ struct dma_fence *amdgpu_device_enforce_isolation(struct amdgpu_device *adev, dep = amdgpu_sync_peek_fence(&isolation->prev, ring); r = amdgpu_sync_fence(&isolation->active, &f->finished, GFP_NOWAIT); if (r) - DRM_WARN("OOM tracking isolation\n"); + dev_warn(adev->dev, "OOM tracking isolation\n"); out_grab_ref: dma_fence_get(dep); @@ -7299,9 +7446,11 @@ uint32_t amdgpu_device_wait_on_rreg(struct amdgpu_device *adev, tmp_ = RREG32(reg_addr); loop--; if (!loop) { - DRM_WARN("Register(%d) [%s] failed to reach value 0x%08x != 0x%08xn", - inst, reg_name, (uint32_t)expected_value, - (uint32_t)(tmp_ & (mask))); + dev_warn( + adev->dev, + "Register(%d) [%s] failed to reach value 0x%08x != 0x%08xn", + inst, reg_name, (uint32_t)expected_value, + (uint32_t)(tmp_ & (mask))); ret = -ETIMEDOUT; break; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index a0e9bf9b2710..81b3443c8d7f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -321,10 +321,12 @@ static int amdgpu_discovery_read_binary_from_file(struct amdgpu_device *adev, const struct firmware *fw; int r; - r = request_firmware(&fw, fw_name, adev->dev); + r = firmware_request_nowarn(&fw, fw_name, adev->dev); if (r) { - dev_err(adev->dev, "can't load firmware \"%s\"\n", - fw_name); + if (amdgpu_discovery == 2) + dev_err(adev->dev, "can't load firmware \"%s\"\n", fw_name); + else + drm_info(&adev->ddev, "Optional firmware \"%s\" was not found\n", fw_name); return r; } @@ -459,16 +461,12 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev) /* Read from file if it is the preferred option */ fw_name = amdgpu_discovery_get_fw_name(adev); if (fw_name != NULL) { - dev_info(adev->dev, "use ip discovery information from file"); + drm_dbg(&adev->ddev, "use ip discovery information from file"); r = amdgpu_discovery_read_binary_from_file(adev, adev->mman.discovery_bin, fw_name); - - if (r) { - dev_err(adev->dev, "failed to read ip discovery binary from file\n"); - r = -EINVAL; + if (r) goto out; - } - } else { + drm_dbg(&adev->ddev, "use ip discovery information from memory"); r = amdgpu_discovery_read_binary_from_mem( adev, adev->mman.discovery_bin); if (r) @@ -1338,10 +1336,8 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) int r; r = amdgpu_discovery_init(adev); - if (r) { - DRM_ERROR("amdgpu_discovery_init failed\n"); + if (r) return r; - } wafl_ver = 0; adev->gfx.xcc_mask = 0; @@ -2579,8 +2575,10 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) break; default: r = amdgpu_discovery_reg_base_init(adev); - if (r) - return -EINVAL; + if (r) { + drm_err(&adev->ddev, "discovery failed: %d\n", r); + return r; + } amdgpu_discovery_harvest_ip(adev); amdgpu_discovery_get_gfx_info(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index 35c778426a7c..9e463d3ee927 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -1317,7 +1317,7 @@ amdgpu_display_user_framebuffer_create(struct drm_device *dev, /* Handle is imported dma-buf, so cannot be migrated to VRAM for scanout */ bo = gem_to_amdgpu_bo(obj); domains = amdgpu_display_supported_domains(drm_to_adev(dev), bo->flags); - if (obj->import_attach && !(domains & AMDGPU_GEM_DOMAIN_GTT)) { + if (drm_gem_is_imported(obj) && !(domains & AMDGPU_GEM_DOMAIN_GTT)) { drm_dbg_kms(dev, "Cannot create framebuffer from imported dma_buf\n"); drm_gem_object_put(obj); return ERR_PTR(-EINVAL); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c index 44e120f9f764..ff98c87b2e0b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c @@ -513,8 +513,8 @@ bool amdgpu_dmabuf_is_xgmi_accessible(struct amdgpu_device *adev, if (!adev) return false; - if (obj->import_attach) { - struct dma_buf *dma_buf = obj->import_attach->dmabuf; + if (drm_gem_is_imported(obj)) { + struct dma_buf *dma_buf = obj->dma_buf; if (dma_buf->ops != &amdgpu_dmabuf_ops) /* No XGMI with non AMD GPUs */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c index 3f3662e8b871..3040437d99c2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c @@ -41,7 +41,8 @@ u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index) if (index < adev->doorbell.num_kernel_doorbells) return readl(adev->doorbell.cpu_addr + index); - DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", index); + dev_err(adev->dev, "reading beyond doorbell aperture: 0x%08x!\n", + index); return 0; } @@ -63,7 +64,8 @@ void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v) if (index < adev->doorbell.num_kernel_doorbells) writel(v, adev->doorbell.cpu_addr + index); else - DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", index); + dev_err(adev->dev, + "writing beyond doorbell aperture: 0x%08x!\n", index); } /** @@ -83,7 +85,8 @@ u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index) if (index < adev->doorbell.num_kernel_doorbells) return atomic64_read((atomic64_t *)(adev->doorbell.cpu_addr + index)); - DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", index); + dev_err(adev->dev, "reading beyond doorbell aperture: 0x%08x!\n", + index); return 0; } @@ -105,7 +108,8 @@ void amdgpu_mm_wdoorbell64(struct amdgpu_device *adev, u32 index, u64 v) if (index < adev->doorbell.num_kernel_doorbells) atomic64_set((atomic64_t *)(adev->doorbell.cpu_addr + index), v); else - DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", index); + dev_err(adev->dev, + "writing beyond doorbell aperture: 0x%08x!\n", index); } /** @@ -166,7 +170,8 @@ int amdgpu_doorbell_create_kernel_doorbells(struct amdgpu_device *adev) NULL, (void **)&adev->doorbell.cpu_addr); if (r) { - DRM_ERROR("Failed to allocate kernel doorbells, err=%d\n", r); + dev_err(adev->dev, + "Failed to allocate kernel doorbells, err=%d\n", r); return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 4db92e0a60da..b299e15bb5e5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -144,6 +144,7 @@ enum AMDGPU_DEBUG_MASK { AMDGPU_DEBUG_DISABLE_GPU_RING_RESET = BIT(6), AMDGPU_DEBUG_SMU_POOL = BIT(7), AMDGPU_DEBUG_VM_USERPTR = BIT(8), + AMDGPU_DEBUG_DISABLE_RAS_CE_LOG = BIT(9) }; unsigned int amdgpu_vram_limit = UINT_MAX; @@ -2278,6 +2279,11 @@ static void amdgpu_init_debug_options(struct amdgpu_device *adev) pr_info("debug: VM mode debug for userptr is enabled\n"); adev->debug_vm_userptr = true; } + + if (amdgpu_debug_mask & AMDGPU_DEBUG_DISABLE_RAS_CE_LOG) { + pr_info("debug: disable kernel logs of correctable errors\n"); + adev->debug_disable_ce_logs = true; + } } static unsigned long amdgpu_fix_asic_type(struct pci_dev *pdev, unsigned long flags) @@ -2321,7 +2327,7 @@ static int amdgpu_pci_probe(struct pci_dev *pdev, amdgpu_aspm = 0; if (amdgpu_virtual_display || - amdgpu_device_asic_has_dc_support(flags & AMD_ASIC_MASK)) + amdgpu_device_asic_has_dc_support(pdev, flags & AMD_ASIC_MASK)) supports_atomic = true; if ((flags & AMD_EXP_HW_SUPPORT) && !amdgpu_exp_hw_support) { @@ -2570,7 +2576,7 @@ static int amdgpu_pmops_prepare(struct device *dev) static void amdgpu_pmops_complete(struct device *dev) { - /* nothing to do */ + amdgpu_device_complete(dev_get_drvdata(dev)); } static int amdgpu_pmops_suspend(struct device *dev) @@ -2906,20 +2912,6 @@ done: return ret; } -static int amdgpu_drm_release(struct inode *inode, struct file *filp) -{ - struct drm_file *file_priv = filp->private_data; - struct amdgpu_fpriv *fpriv = file_priv->driver_priv; - - if (fpriv) { - fpriv->evf_mgr.fd_closing = true; - amdgpu_eviction_fence_destroy(&fpriv->evf_mgr); - amdgpu_userq_mgr_fini(&fpriv->userq_mgr); - } - - return drm_release(inode, filp); -} - long amdgpu_drm_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { @@ -2971,7 +2963,7 @@ static const struct file_operations amdgpu_driver_kms_fops = { .owner = THIS_MODULE, .open = drm_open, .flush = amdgpu_flush, - .release = amdgpu_drm_release, + .release = drm_release, .unlocked_ioctl = amdgpu_drm_ioctl, .mmap = drm_gem_mmap, .poll = drm_poll, @@ -3107,10 +3099,6 @@ static int __init amdgpu_init(void) if (r) goto error_sync; - r = amdgpu_fence_slab_init(); - if (r) - goto error_fence; - r = amdgpu_userq_fence_slab_init(); if (r) goto error_fence; @@ -3145,7 +3133,6 @@ static void __exit amdgpu_exit(void) amdgpu_unregister_atpx_handler(); amdgpu_acpi_release(); amdgpu_sync_fini(); - amdgpu_fence_slab_fini(); amdgpu_userq_fence_slab_fini(); mmu_notifier_synchronize(); amdgpu_xcp_drv_release(); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 8cecf25996ed..2c3547f4cea4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -42,37 +42,6 @@ #include "amdgpu_reset.h" /* - * Fences mark an event in the GPUs pipeline and are used - * for GPU/CPU synchronization. When the fence is written, - * it is expected that all buffers associated with that fence - * are no longer in use by the associated ring on the GPU and - * that the relevant GPU caches have been flushed. - */ - -struct amdgpu_fence { - struct dma_fence base; - - /* RB, DMA, etc. */ - struct amdgpu_ring *ring; - ktime_t start_timestamp; -}; - -static struct kmem_cache *amdgpu_fence_slab; - -int amdgpu_fence_slab_init(void) -{ - amdgpu_fence_slab = KMEM_CACHE(amdgpu_fence, SLAB_HWCACHE_ALIGN); - if (!amdgpu_fence_slab) - return -ENOMEM; - return 0; -} - -void amdgpu_fence_slab_fini(void) -{ - rcu_barrier(); - kmem_cache_destroy(amdgpu_fence_slab); -} -/* * Cast helper */ static const struct dma_fence_ops amdgpu_fence_ops; @@ -130,14 +99,14 @@ static u32 amdgpu_fence_read(struct amdgpu_ring *ring) * * @ring: ring the fence is associated with * @f: resulting fence object - * @job: job the fence is embedded in + * @af: amdgpu fence input * @flags: flags to pass into the subordinate .emit_fence() call * * Emits a fence command on the requested ring (all asics). * Returns 0 on success, -ENOMEM on failure. */ -int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, struct amdgpu_job *job, - unsigned int flags) +int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, + struct amdgpu_fence *af, unsigned int flags) { struct amdgpu_device *adev = ring->adev; struct dma_fence *fence; @@ -146,36 +115,28 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, struct amd uint32_t seq; int r; - if (job == NULL) { - /* create a sperate hw fence */ - am_fence = kmem_cache_alloc(amdgpu_fence_slab, GFP_ATOMIC); - if (am_fence == NULL) + if (!af) { + /* create a separate hw fence */ + am_fence = kzalloc(sizeof(*am_fence), GFP_KERNEL); + if (!am_fence) return -ENOMEM; - fence = &am_fence->base; - am_fence->ring = ring; } else { - /* take use of job-embedded fence */ - fence = &job->hw_fence; + am_fence = af; } + fence = &am_fence->base; + am_fence->ring = ring; seq = ++ring->fence_drv.sync_seq; - if (job && job->job_run_counter) { - /* reinit seq for resubmitted jobs */ - fence->seqno = seq; - /* TO be inline with external fence creation and other drivers */ + if (af) { + dma_fence_init(fence, &amdgpu_job_fence_ops, + &ring->fence_drv.lock, + adev->fence_context + ring->idx, seq); + /* Against remove in amdgpu_job_{free, free_cb} */ dma_fence_get(fence); } else { - if (job) { - dma_fence_init(fence, &amdgpu_job_fence_ops, - &ring->fence_drv.lock, - adev->fence_context + ring->idx, seq); - /* Against remove in amdgpu_job_{free, free_cb} */ - dma_fence_get(fence); - } else { - dma_fence_init(fence, &amdgpu_fence_ops, - &ring->fence_drv.lock, - adev->fence_context + ring->idx, seq); - } + dma_fence_init(fence, &amdgpu_fence_ops, + &ring->fence_drv.lock, + adev->fence_context + ring->idx, seq); } amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr, @@ -326,7 +287,9 @@ static void amdgpu_fence_fallback(struct timer_list *t) fence_drv.fallback_timer); if (amdgpu_fence_process(ring)) - DRM_WARN("Fence fallback timer expired on ring %s\n", ring->name); + dev_warn(ring->adev->dev, + "Fence fallback timer expired on ring %s\n", + ring->name); } /** @@ -718,7 +681,7 @@ void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring) * it right here or we won't be able to track them in fence_drv * and they will remain unsignaled during sa_bo free. */ - job = container_of(old, struct amdgpu_job, hw_fence); + job = container_of(old, struct amdgpu_job, hw_fence.base); if (!job->base.s_fence && !dma_fence_is_signaled(old)) dma_fence_signal(old); RCU_INIT_POINTER(*ptr, NULL); @@ -780,7 +743,7 @@ static const char *amdgpu_fence_get_timeline_name(struct dma_fence *f) static const char *amdgpu_job_fence_get_timeline_name(struct dma_fence *f) { - struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence); + struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence.base); return (const char *)to_amdgpu_ring(job->base.sched)->name; } @@ -810,7 +773,7 @@ static bool amdgpu_fence_enable_signaling(struct dma_fence *f) */ static bool amdgpu_job_fence_enable_signaling(struct dma_fence *f) { - struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence); + struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence.base); if (!timer_pending(&to_amdgpu_ring(job->base.sched)->fence_drv.fallback_timer)) amdgpu_fence_schedule_fallback(to_amdgpu_ring(job->base.sched)); @@ -830,7 +793,7 @@ static void amdgpu_fence_free(struct rcu_head *rcu) struct dma_fence *f = container_of(rcu, struct dma_fence, rcu); /* free fence_slab if it's separated fence*/ - kmem_cache_free(amdgpu_fence_slab, to_amdgpu_fence(f)); + kfree(to_amdgpu_fence(f)); } /** @@ -845,7 +808,7 @@ static void amdgpu_job_fence_free(struct rcu_head *rcu) struct dma_fence *f = container_of(rcu, struct dma_fence, rcu); /* free job if fence has a parent job */ - kfree(container_of(f, struct amdgpu_job, hw_fence)); + kfree(container_of(f, struct amdgpu_job, hw_fence.base)); } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c index 1ae88c459da5..b0082aa7f3c6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c @@ -144,7 +144,8 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) /* If algo exists, it means that the i2c_adapter's initialized */ if (!adev->pm.fru_eeprom_i2c_bus || !adev->pm.fru_eeprom_i2c_bus->algo) { - DRM_WARN("Cannot access FRU, EEPROM accessor not initialized"); + dev_warn(adev->dev, + "Cannot access FRU, EEPROM accessor not initialized"); return -ENODEV; } @@ -152,19 +153,22 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) len = amdgpu_eeprom_read(adev->pm.fru_eeprom_i2c_bus, fru_addr, buf, sizeof(buf)); if (len != 8) { - DRM_ERROR("Couldn't read the IPMI Common Header: %d", len); + dev_err(adev->dev, "Couldn't read the IPMI Common Header: %d", + len); return len < 0 ? len : -EIO; } if (buf[0] != 1) { - DRM_ERROR("Bad IPMI Common Header version: 0x%02x", buf[0]); + dev_err(adev->dev, "Bad IPMI Common Header version: 0x%02x", + buf[0]); return -EIO; } for (csum = 0; len > 0; len--) csum += buf[len - 1]; if (csum) { - DRM_ERROR("Bad IPMI Common Header checksum: 0x%02x", csum); + dev_err(adev->dev, "Bad IPMI Common Header checksum: 0x%02x", + csum); return -EIO; } @@ -179,12 +183,14 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) /* Read the header of the PIA. */ len = amdgpu_eeprom_read(adev->pm.fru_eeprom_i2c_bus, addr, buf, 3); if (len != 3) { - DRM_ERROR("Couldn't read the Product Info Area header: %d", len); + dev_err(adev->dev, + "Couldn't read the Product Info Area header: %d", len); return len < 0 ? len : -EIO; } if (buf[0] != 1) { - DRM_ERROR("Bad IPMI Product Info Area version: 0x%02x", buf[0]); + dev_err(adev->dev, "Bad IPMI Product Info Area version: 0x%02x", + buf[0]); return -EIO; } @@ -197,14 +203,16 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) len = amdgpu_eeprom_read(adev->pm.fru_eeprom_i2c_bus, addr, pia, size); if (len != size) { kfree(pia); - DRM_ERROR("Couldn't read the Product Info Area: %d", len); + dev_err(adev->dev, "Couldn't read the Product Info Area: %d", + len); return len < 0 ? len : -EIO; } for (csum = 0; size > 0; size--) csum += pia[size - 1]; if (csum) { - DRM_ERROR("Bad Product Info Area checksum: 0x%02x", csum); + dev_err(adev->dev, "Bad Product Info Area checksum: 0x%02x", + csum); kfree(pia); return -EIO; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index e5e33a68d935..6626a6e64ff5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -317,8 +317,7 @@ static int amdgpu_gem_object_open(struct drm_gem_object *obj, */ if (!vm->is_compute_context || !vm->process_info) return 0; - if (!obj->import_attach || - !dma_buf_is_dynamic(obj->import_attach->dmabuf)) + if (!drm_gem_is_imported(obj) || !dma_buf_is_dynamic(obj->dma_buf)) return 0; mutex_lock_nested(&vm->process_info->lock, 1); if (!WARN_ON(!vm->process_info->eviction_fence)) { @@ -1024,7 +1023,7 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data, break; } case AMDGPU_GEM_OP_SET_PLACEMENT: - if (robj->tbo.base.import_attach && + if (drm_gem_is_imported(&robj->tbo.base) && args->value & AMDGPU_GEM_DOMAIN_VRAM) { r = -EINVAL; amdgpu_bo_unreserve(robj); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index c5646af055ab..c80c8f543532 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -149,7 +149,7 @@ static bool amdgpu_gfx_is_graphics_multipipe_capable(struct amdgpu_device *adev) static bool amdgpu_gfx_is_compute_multipipe_capable(struct amdgpu_device *adev) { if (amdgpu_compute_multipipe != -1) { - DRM_INFO("amdgpu: forcing compute pipe policy %d\n", + dev_info(adev->dev, "amdgpu: forcing compute pipe policy %d\n", amdgpu_compute_multipipe); return amdgpu_compute_multipipe == 1; } @@ -674,7 +674,7 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id) * generation exposes more than 64 queues. If so, the * definition of queue_mask needs updating */ if (WARN_ON(i > (sizeof(queue_mask)*8))) { - DRM_ERROR("Invalid KCQ enabled: %d\n", i); + dev_err(adev->dev, "Invalid KCQ enabled: %d\n", i); break; } @@ -683,15 +683,15 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id) amdgpu_device_flush_hdp(adev, NULL); - DRM_INFO("kiq ring mec %d pipe %d q %d\n", kiq_ring->me, kiq_ring->pipe, - kiq_ring->queue); + dev_info(adev->dev, "kiq ring mec %d pipe %d q %d\n", kiq_ring->me, + kiq_ring->pipe, kiq_ring->queue); spin_lock(&kiq->ring_lock); r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size * adev->gfx.num_compute_rings + kiq->pmf->set_resources_size); if (r) { - DRM_ERROR("Failed to lock KIQ (%d).\n", r); + dev_err(adev->dev, "Failed to lock KIQ (%d).\n", r); spin_unlock(&kiq->ring_lock); return r; } @@ -712,7 +712,7 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id) r = amdgpu_ring_test_helper(kiq_ring); spin_unlock(&kiq->ring_lock); if (r) - DRM_ERROR("KCQ enable failed\n"); + dev_err(adev->dev, "KCQ enable failed\n"); return r; } @@ -734,7 +734,7 @@ int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id) r = amdgpu_mes_map_legacy_queue(adev, &adev->gfx.gfx_ring[j]); if (r) { - DRM_ERROR("failed to map gfx queue\n"); + dev_err(adev->dev, "failed to map gfx queue\n"); return r; } } @@ -748,7 +748,7 @@ int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id) r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size * adev->gfx.num_gfx_rings); if (r) { - DRM_ERROR("Failed to lock KIQ (%d).\n", r); + dev_err(adev->dev, "Failed to lock KIQ (%d).\n", r); spin_unlock(&kiq->ring_lock); return r; } @@ -769,7 +769,7 @@ int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id) r = amdgpu_ring_test_helper(kiq_ring); spin_unlock(&kiq->ring_lock); if (r) - DRM_ERROR("KGQ enable failed\n"); + dev_err(adev->dev, "KGQ enable failed\n"); return r; } @@ -1030,7 +1030,7 @@ int amdgpu_gfx_cp_ecc_error_irq(struct amdgpu_device *adev, ih_data.head = *ras_if; - DRM_ERROR("CP ECC ERROR IRQ\n"); + dev_err(adev->dev, "CP ECC ERROR IRQ\n"); amdgpu_ras_interrupt_dispatch(adev, &ih_data); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 6b0fbbb91e57..97b562a79ea8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -38,6 +38,13 @@ #include <drm/drm_drv.h> #include <drm/ttm/ttm_tt.h> +static const u64 four_gb = 0x100000000ULL; + +bool amdgpu_gmc_is_pdb0_enabled(struct amdgpu_device *adev) +{ + return adev->gmc.xgmi.connected_to_cpu || amdgpu_virt_xgmi_migrate_enabled(adev); +} + /** * amdgpu_gmc_pdb0_alloc - allocate vram for pdb0 * @@ -251,10 +258,20 @@ void amdgpu_gmc_sysvm_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc u64 hive_vram_end = mc->xgmi.node_segment_size * mc->xgmi.num_physical_nodes - 1; mc->vram_start = mc->xgmi.node_segment_size * mc->xgmi.physical_node_id; mc->vram_end = mc->vram_start + mc->xgmi.node_segment_size - 1; - mc->gart_start = hive_vram_end + 1; + /* node_segment_size may not 4GB aligned on SRIOV, align up is needed. */ + mc->gart_start = ALIGN(hive_vram_end + 1, four_gb); mc->gart_end = mc->gart_start + mc->gart_size - 1; - mc->fb_start = hive_vram_start; - mc->fb_end = hive_vram_end; + if (amdgpu_virt_xgmi_migrate_enabled(adev)) { + /* set mc->vram_start to 0 to switch the returned GPU address of + * amdgpu_bo_create_reserved() from FB aperture to GART aperture. + */ + mc->vram_start = 0; + mc->vram_end = mc->vram_start + mc->mc_vram_size - 1; + mc->visible_vram_size = min(mc->visible_vram_size, mc->real_vram_size); + } else { + mc->fb_start = hive_vram_start; + mc->fb_end = hive_vram_end; + } dev_info(adev->dev, "VRAM: %lluM 0x%016llX - 0x%016llX (%lluM used)\n", mc->mc_vram_size >> 20, mc->vram_start, mc->vram_end, mc->real_vram_size >> 20); @@ -276,7 +293,6 @@ void amdgpu_gmc_sysvm_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc void amdgpu_gmc_gart_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc, enum amdgpu_gart_placement gart_placement) { - const uint64_t four_gb = 0x100000000ULL; u64 size_af, size_bf; /*To avoid the hole, limit the max mc address to AMDGPU_GMC_HOLE_START*/ u64 max_mc_address = min(adev->gmc.mc_mask, AMDGPU_GMC_HOLE_START - 1); @@ -1041,9 +1057,7 @@ void amdgpu_gmc_init_pdb0(struct amdgpu_device *adev) */ u64 vram_size = adev->gmc.xgmi.node_segment_size * adev->gmc.xgmi.num_physical_nodes; u64 pde0_page_size = (1ULL<<adev->gmc.vmid0_page_table_block_size)<<21; - u64 vram_addr = adev->vm_manager.vram_base_offset - - adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; - u64 vram_end = vram_addr + vram_size; + u64 vram_addr, vram_end; u64 gart_ptb_gpu_pa = amdgpu_gmc_vram_pa(adev, adev->gart.bo); int idx; @@ -1056,6 +1070,11 @@ void amdgpu_gmc_init_pdb0(struct amdgpu_device *adev) flags |= AMDGPU_PTE_FRAG((adev->gmc.vmid0_page_table_block_size + 9*1)); flags |= AMDGPU_PDE_PTE_FLAG(adev); + vram_addr = adev->vm_manager.vram_base_offset; + if (!amdgpu_virt_xgmi_migrate_enabled(adev)) + vram_addr -= adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; + vram_end = vram_addr + vram_size; + /* The first n PDE0 entries are used as PTE, * pointing to vram */ @@ -1429,3 +1448,232 @@ bool amdgpu_gmc_need_reset_on_init(struct amdgpu_device *adev) return false; } + +enum amdgpu_memory_partition +amdgpu_gmc_get_vf_memory_partition(struct amdgpu_device *adev) +{ + switch (adev->gmc.num_mem_partitions) { + case 0: + return UNKNOWN_MEMORY_PARTITION_MODE; + case 1: + return AMDGPU_NPS1_PARTITION_MODE; + case 2: + return AMDGPU_NPS2_PARTITION_MODE; + case 4: + return AMDGPU_NPS4_PARTITION_MODE; + case 8: + return AMDGPU_NPS8_PARTITION_MODE; + default: + return AMDGPU_NPS1_PARTITION_MODE; + } +} + +enum amdgpu_memory_partition +amdgpu_gmc_get_memory_partition(struct amdgpu_device *adev, u32 *supp_modes) +{ + enum amdgpu_memory_partition mode = UNKNOWN_MEMORY_PARTITION_MODE; + + if (adev->nbio.funcs && + adev->nbio.funcs->get_memory_partition_mode) + mode = adev->nbio.funcs->get_memory_partition_mode(adev, + supp_modes); + else + dev_warn(adev->dev, "memory partition mode query is not supported\n"); + + return mode; +} + +enum amdgpu_memory_partition +amdgpu_gmc_query_memory_partition(struct amdgpu_device *adev) +{ + if (amdgpu_sriov_vf(adev)) + return amdgpu_gmc_get_vf_memory_partition(adev); + else + return amdgpu_gmc_get_memory_partition(adev, NULL); +} + +static bool amdgpu_gmc_validate_partition_info(struct amdgpu_device *adev) +{ + enum amdgpu_memory_partition mode; + u32 supp_modes; + bool valid; + + mode = amdgpu_gmc_get_memory_partition(adev, &supp_modes); + + /* Mode detected by hardware not present in supported modes */ + if ((mode != UNKNOWN_MEMORY_PARTITION_MODE) && + !(BIT(mode - 1) & supp_modes)) + return false; + + switch (mode) { + case UNKNOWN_MEMORY_PARTITION_MODE: + case AMDGPU_NPS1_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 1); + break; + case AMDGPU_NPS2_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 2); + break; + case AMDGPU_NPS4_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 3 || + adev->gmc.num_mem_partitions == 4); + break; + case AMDGPU_NPS8_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 8); + break; + default: + valid = false; + } + + return valid; +} + +static bool amdgpu_gmc_is_node_present(int *node_ids, int num_ids, int nid) +{ + int i; + + /* Check if node with id 'nid' is present in 'node_ids' array */ + for (i = 0; i < num_ids; ++i) + if (node_ids[i] == nid) + return true; + + return false; +} + +static void +amdgpu_gmc_init_acpi_mem_ranges(struct amdgpu_device *adev, + struct amdgpu_mem_partition_info *mem_ranges) +{ + struct amdgpu_numa_info numa_info; + int node_ids[AMDGPU_MAX_MEM_RANGES]; + int num_ranges = 0, ret; + int num_xcc, xcc_id; + uint32_t xcc_mask; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + xcc_mask = (1U << num_xcc) - 1; + + for_each_inst(xcc_id, xcc_mask) { + ret = amdgpu_acpi_get_mem_info(adev, xcc_id, &numa_info); + if (ret) + continue; + + if (numa_info.nid == NUMA_NO_NODE) { + mem_ranges[0].size = numa_info.size; + mem_ranges[0].numa.node = numa_info.nid; + num_ranges = 1; + break; + } + + if (amdgpu_gmc_is_node_present(node_ids, num_ranges, + numa_info.nid)) + continue; + + node_ids[num_ranges] = numa_info.nid; + mem_ranges[num_ranges].numa.node = numa_info.nid; + mem_ranges[num_ranges].size = numa_info.size; + ++num_ranges; + } + + adev->gmc.num_mem_partitions = num_ranges; +} + +void amdgpu_gmc_init_sw_mem_ranges(struct amdgpu_device *adev, + struct amdgpu_mem_partition_info *mem_ranges) +{ + enum amdgpu_memory_partition mode; + u32 start_addr = 0, size; + int i, r, l; + + mode = amdgpu_gmc_query_memory_partition(adev); + + switch (mode) { + case UNKNOWN_MEMORY_PARTITION_MODE: + adev->gmc.num_mem_partitions = 0; + break; + case AMDGPU_NPS1_PARTITION_MODE: + adev->gmc.num_mem_partitions = 1; + break; + case AMDGPU_NPS2_PARTITION_MODE: + adev->gmc.num_mem_partitions = 2; + break; + case AMDGPU_NPS4_PARTITION_MODE: + if (adev->flags & AMD_IS_APU) + adev->gmc.num_mem_partitions = 3; + else + adev->gmc.num_mem_partitions = 4; + break; + case AMDGPU_NPS8_PARTITION_MODE: + adev->gmc.num_mem_partitions = 8; + break; + default: + adev->gmc.num_mem_partitions = 1; + break; + } + + /* Use NPS range info, if populated */ + r = amdgpu_gmc_get_nps_memranges(adev, mem_ranges, + &adev->gmc.num_mem_partitions); + if (!r) { + l = 0; + for (i = 1; i < adev->gmc.num_mem_partitions; ++i) { + if (mem_ranges[i].range.lpfn > + mem_ranges[i - 1].range.lpfn) + l = i; + } + + } else { + if (!adev->gmc.num_mem_partitions) { + dev_warn(adev->dev, + "Not able to detect NPS mode, fall back to NPS1\n"); + adev->gmc.num_mem_partitions = 1; + } + /* Fallback to sw based calculation */ + size = (adev->gmc.real_vram_size + SZ_16M) >> AMDGPU_GPU_PAGE_SHIFT; + size /= adev->gmc.num_mem_partitions; + + for (i = 0; i < adev->gmc.num_mem_partitions; ++i) { + mem_ranges[i].range.fpfn = start_addr; + mem_ranges[i].size = + ((u64)size << AMDGPU_GPU_PAGE_SHIFT); + mem_ranges[i].range.lpfn = start_addr + size - 1; + start_addr += size; + } + + l = adev->gmc.num_mem_partitions - 1; + } + + /* Adjust the last one */ + mem_ranges[l].range.lpfn = + (adev->gmc.real_vram_size >> AMDGPU_GPU_PAGE_SHIFT) - 1; + mem_ranges[l].size = + adev->gmc.real_vram_size - + ((u64)mem_ranges[l].range.fpfn << AMDGPU_GPU_PAGE_SHIFT); +} + +int amdgpu_gmc_init_mem_ranges(struct amdgpu_device *adev) +{ + bool valid; + + adev->gmc.mem_partitions = kcalloc(AMDGPU_MAX_MEM_RANGES, + sizeof(struct amdgpu_mem_partition_info), + GFP_KERNEL); + if (!adev->gmc.mem_partitions) + return -ENOMEM; + + if (adev->gmc.is_app_apu) + amdgpu_gmc_init_acpi_mem_ranges(adev, adev->gmc.mem_partitions); + else + amdgpu_gmc_init_sw_mem_ranges(adev, adev->gmc.mem_partitions); + + if (amdgpu_sriov_vf(adev)) + valid = true; + else + valid = amdgpu_gmc_validate_partition_info(adev); + if (!valid) { + /* TODO: handle invalid case */ + dev_warn(adev->dev, + "Mem ranges not matching with hardware config\n"); + } + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 80fa29c26e9e..397c6ccdb903 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -84,6 +84,8 @@ enum amdgpu_memory_partition { #define AMDGPU_GMC_INIT_RESET_NPS BIT(0) +#define AMDGPU_MAX_MEM_RANGES 8 + /* * GMC page fault information */ @@ -394,6 +396,7 @@ static inline uint64_t amdgpu_gmc_sign_extend(uint64_t addr) return addr; } +bool amdgpu_gmc_is_pdb0_enabled(struct amdgpu_device *adev); int amdgpu_gmc_pdb0_alloc(struct amdgpu_device *adev); void amdgpu_gmc_get_pde_for_bo(struct amdgpu_bo *bo, int level, uint64_t *addr, uint64_t *flags); @@ -455,5 +458,13 @@ int amdgpu_gmc_request_memory_partition(struct amdgpu_device *adev, int nps_mode); void amdgpu_gmc_prepare_nps_mode_change(struct amdgpu_device *adev); bool amdgpu_gmc_need_reset_on_init(struct amdgpu_device *adev); - +enum amdgpu_memory_partition +amdgpu_gmc_get_vf_memory_partition(struct amdgpu_device *adev); +enum amdgpu_memory_partition +amdgpu_gmc_get_memory_partition(struct amdgpu_device *adev, u32 *supp_modes); +enum amdgpu_memory_partition +amdgpu_gmc_query_memory_partition(struct amdgpu_device *adev); +int amdgpu_gmc_init_mem_ranges(struct amdgpu_device *adev); +void amdgpu_gmc_init_sw_mem_ranges(struct amdgpu_device *adev, + struct amdgpu_mem_partition_info *mem_ranges); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c index 8179d0814db9..57101d24422f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c @@ -24,7 +24,6 @@ * Alex Deucher */ -#include <linux/export.h> #include <linux/pci.h> #include <drm/drm_edid.h> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index 802743efa3b3..206b70acb29a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -128,6 +128,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, struct amdgpu_device *adev = ring->adev; struct amdgpu_ib *ib = &ibs[0]; struct dma_fence *tmp = NULL; + struct amdgpu_fence *af; bool need_ctx_switch; struct amdgpu_vm *vm; uint64_t fence_ctx; @@ -154,6 +155,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, csa_va = job->csa_va; gds_va = job->gds_va; init_shadow = job->init_shadow; + af = &job->hw_fence; } else { vm = NULL; fence_ctx = 0; @@ -161,6 +163,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, csa_va = 0; gds_va = 0; init_shadow = false; + af = NULL; } if (!ring->sched.ready) { @@ -282,7 +285,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, amdgpu_ring_init_cond_exec(ring, ring->cond_exe_gpu_addr); } - r = amdgpu_fence_emit(ring, f, job, fence_flags); + r = amdgpu_fence_emit(ring, f, af, fence_flags); if (r) { dev_err(adev->dev, "failed to emit fence (%d)\n", r); if (job && job->vmid) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c index 30f16968b578..a6419246e9c2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c @@ -218,7 +218,7 @@ int amdgpu_ih_process(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih) restart_ih: count = AMDGPU_IH_MAX_NUM_IVS; - DRM_DEBUG("%s: rptr %d, wptr %d\n", __func__, ih->rptr, wptr); + dev_dbg(adev->dev, "%s: rptr %d, wptr %d\n", __func__, ih->rptr, wptr); /* Order reading of wptr vs. reading of IH ring data */ rmb(); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.c new file mode 100644 index 000000000000..99e1cf4fc955 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.c @@ -0,0 +1,96 @@ +/* + * Copyright 2025 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#include "amdgpu.h" +#include "amdgpu_ip.h" + +static int8_t amdgpu_logical_to_dev_inst(struct amdgpu_device *adev, + enum amd_hw_ip_block_type block, + int8_t inst) +{ + int8_t dev_inst; + + switch (block) { + case GC_HWIP: + case SDMA0_HWIP: + /* Both JPEG and VCN as JPEG is only alias of VCN */ + case VCN_HWIP: + dev_inst = adev->ip_map.dev_inst[block][inst]; + break; + default: + /* For rest of the IPs, no look up required. + * Assume 'logical instance == physical instance' for all configs. */ + dev_inst = inst; + break; + } + + return dev_inst; +} + +static uint32_t amdgpu_logical_to_dev_mask(struct amdgpu_device *adev, + enum amd_hw_ip_block_type block, + uint32_t mask) +{ + uint32_t dev_mask = 0; + int8_t log_inst, dev_inst; + + while (mask) { + log_inst = ffs(mask) - 1; + dev_inst = amdgpu_logical_to_dev_inst(adev, block, log_inst); + dev_mask |= (1 << dev_inst); + mask &= ~(1 << log_inst); + } + + return dev_mask; +} + +static void amdgpu_populate_ip_map(struct amdgpu_device *adev, + enum amd_hw_ip_block_type ip_block, + uint32_t inst_mask) +{ + int l = 0, i; + + while (inst_mask) { + i = ffs(inst_mask) - 1; + adev->ip_map.dev_inst[ip_block][l++] = i; + inst_mask &= ~(1 << i); + } + for (; l < HWIP_MAX_INSTANCE; l++) + adev->ip_map.dev_inst[ip_block][l] = -1; +} + +void amdgpu_ip_map_init(struct amdgpu_device *adev) +{ + u32 ip_map[][2] = { + { GC_HWIP, adev->gfx.xcc_mask }, + { SDMA0_HWIP, adev->sdma.sdma_mask }, + { VCN_HWIP, adev->vcn.inst_mask }, + }; + int i; + + for (i = 0; i < ARRAY_SIZE(ip_map); ++i) + amdgpu_populate_ip_map(adev, ip_map[i][0], ip_map[i][1]); + + adev->ip_map.logical_to_dev_inst = amdgpu_logical_to_dev_inst; + adev->ip_map.logical_to_dev_mask = amdgpu_logical_to_dev_mask; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.h new file mode 100644 index 000000000000..2490fd322aec --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.h @@ -0,0 +1,29 @@ +/* + * Copyright 2025 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef __AMDGPU_IP_H__ +#define __AMDGPU_IP_H__ + +void amdgpu_ip_map_init(struct amdgpu_device *adev); + +#endif /* __AMDGPU_IP_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index 13c60cac4261..183fa33c2434 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -142,8 +142,9 @@ void amdgpu_irq_disable_all(struct amdgpu_device *adev) r = src->funcs->set(adev, src, k, AMDGPU_IRQ_STATE_DISABLE); if (r) - DRM_ERROR("error disabling interrupt (%d)\n", - r); + dev_err(adev->dev, + "error disabling interrupt (%d)\n", + r); } } } @@ -242,7 +243,7 @@ static bool amdgpu_msi_ok(struct amdgpu_device *adev) return true; } -static void amdgpu_restore_msix(struct amdgpu_device *adev) +void amdgpu_restore_msix(struct amdgpu_device *adev) { u16 ctrl; @@ -315,7 +316,7 @@ int amdgpu_irq_init(struct amdgpu_device *adev) adev->irq.irq = irq; adev_to_drm(adev)->max_vblank_count = 0x00ffffff; - DRM_DEBUG("amdgpu: irq initialized.\n"); + dev_dbg(adev->dev, "amdgpu: irq initialized.\n"); return 0; free_vectors: @@ -461,10 +462,10 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev, src_id = entry.src_id; if (client_id >= AMDGPU_IRQ_CLIENTID_MAX) { - DRM_DEBUG("Invalid client_id in IV: %d\n", client_id); + dev_dbg(adev->dev, "Invalid client_id in IV: %d\n", client_id); } else if (src_id >= AMDGPU_MAX_IRQ_SRC_ID) { - DRM_DEBUG("Invalid src_id in IV: %d\n", src_id); + dev_dbg(adev->dev, "Invalid src_id in IV: %d\n", src_id); } else if (((client_id == AMDGPU_IRQ_CLIENTID_LEGACY) || (client_id == SOC15_IH_CLIENTID_ISP)) && @@ -472,18 +473,21 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev, generic_handle_domain_irq(adev->irq.domain, src_id); } else if (!adev->irq.client[client_id].sources) { - DRM_DEBUG("Unregistered interrupt client_id: %d src_id: %d\n", - client_id, src_id); + dev_dbg(adev->dev, + "Unregistered interrupt client_id: %d src_id: %d\n", + client_id, src_id); } else if ((src = adev->irq.client[client_id].sources[src_id])) { r = src->funcs->process(adev, src, &entry); if (r < 0) - DRM_ERROR("error processing interrupt (%d)\n", r); + dev_err(adev->dev, "error processing interrupt (%d)\n", + r); else if (r) handled = true; } else { - DRM_DEBUG("Unregistered interrupt src_id: %d of client_id:%d\n", + dev_dbg(adev->dev, + "Unregistered interrupt src_id: %d of client_id:%d\n", src_id, client_id); } @@ -732,7 +736,7 @@ int amdgpu_irq_add_domain(struct amdgpu_device *adev) adev->irq.domain = irq_domain_create_linear(NULL, AMDGPU_MAX_IRQ_SRC_ID, &amdgpu_hw_irqdomain_ops, adev); if (!adev->irq.domain) { - DRM_ERROR("GPU irq add domain failed\n"); + dev_err(adev->dev, "GPU irq add domain failed\n"); return -ENODEV; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h index 04c0b4fa17a4..9f0417456abd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h @@ -146,5 +146,6 @@ void amdgpu_irq_gpu_reset_resume_helper(struct amdgpu_device *adev); int amdgpu_irq_add_domain(struct amdgpu_device *adev); void amdgpu_irq_remove_domain(struct amdgpu_device *adev); unsigned amdgpu_irq_create_mapping(struct amdgpu_device *adev, unsigned src_id); +void amdgpu_restore_msix(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.h index 4f3b7b5d9c1f..1d1c4b1ec7e7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.h @@ -28,6 +28,8 @@ #ifndef __AMDGPU_ISP_H__ #define __AMDGPU_ISP_H__ +#include <linux/pm_domain.h> + #define ISP_REGS_OFFSET_END 0x629A4 struct amdgpu_isp; @@ -54,6 +56,7 @@ struct amdgpu_isp { struct isp_platform_data *isp_pdata; unsigned int harvest_config; const struct firmware *fw; + struct generic_pm_domain ispgpd; }; extern const struct amdgpu_ip_block_version isp_v4_1_0_ip_block; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index 1e24590ae144..2b58e353cca1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -92,8 +92,7 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job) struct drm_wedge_task_info *info = NULL; struct amdgpu_task_info *ti; struct amdgpu_device *adev = ring->adev; - int idx; - int r; + int idx, r; if (!drm_dev_enter(adev_to_drm(adev), &idx)) { dev_info(adev->dev, "%s - device unplugged skipping recovery on scheduler:%s", @@ -133,43 +132,20 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job) if (unlikely(adev->debug_disable_gpu_ring_reset)) { dev_err(adev->dev, "Ring reset disabled by debug mask\n"); } else if (amdgpu_gpu_recovery && ring->funcs->reset) { - bool is_guilty; - - dev_err(adev->dev, "Starting %s ring reset\n", s_job->sched->name); - /* stop the scheduler, but don't mess with the - * bad job yet because if ring reset fails - * we'll fall back to full GPU reset. - */ - drm_sched_wqueue_stop(&ring->sched); - - /* for engine resets, we need to reset the engine, - * but individual queues may be unaffected. - * check here to make sure the accounting is correct. - */ - if (ring->funcs->is_guilty) - is_guilty = ring->funcs->is_guilty(ring); - else - is_guilty = true; - - if (is_guilty) - dma_fence_set_error(&s_job->s_fence->finished, -ETIME); - - r = amdgpu_ring_reset(ring, job->vmid); + dev_err(adev->dev, "Starting %s ring reset\n", + s_job->sched->name); + r = amdgpu_ring_reset(ring, job->vmid, NULL); if (!r) { - if (amdgpu_ring_sched_ready(ring)) - drm_sched_stop(&ring->sched, s_job); - if (is_guilty) { - atomic_inc(&ring->adev->gpu_reset_counter); - amdgpu_fence_driver_force_completion(ring); - } - if (amdgpu_ring_sched_ready(ring)) - drm_sched_start(&ring->sched, 0); - dev_err(adev->dev, "Ring %s reset succeeded\n", ring->sched.name); - drm_dev_wedged_event(adev_to_drm(adev), DRM_WEDGE_RECOVERY_NONE, info); + atomic_inc(&ring->adev->gpu_reset_counter); + dev_err(adev->dev, "Ring %s reset succeeded\n", + ring->sched.name); + drm_dev_wedged_event(adev_to_drm(adev), + DRM_WEDGE_RECOVERY_NONE, info); goto exit; } - dev_err(adev->dev, "Ring %s reset failure\n", ring->sched.name); + dev_err(adev->dev, "Ring %s reset failed\n", ring->sched.name); } + dma_fence_set_error(&s_job->s_fence->finished, -ETIME); amdgpu_vm_put_task_info(ti); @@ -275,8 +251,8 @@ void amdgpu_job_free_resources(struct amdgpu_job *job) /* Check if any fences where initialized */ if (job->base.s_fence && job->base.s_fence->finished.ops) f = &job->base.s_fence->finished; - else if (job->hw_fence.ops) - f = &job->hw_fence; + else if (job->hw_fence.base.ops) + f = &job->hw_fence.base; else f = NULL; @@ -293,10 +269,10 @@ static void amdgpu_job_free_cb(struct drm_sched_job *s_job) amdgpu_sync_free(&job->explicit_sync); /* only put the hw fence if has embedded fence */ - if (!job->hw_fence.ops) + if (!job->hw_fence.base.ops) kfree(job); else - dma_fence_put(&job->hw_fence); + dma_fence_put(&job->hw_fence.base); } void amdgpu_job_set_gang_leader(struct amdgpu_job *job, @@ -325,10 +301,10 @@ void amdgpu_job_free(struct amdgpu_job *job) if (job->gang_submit != &job->base.s_fence->scheduled) dma_fence_put(job->gang_submit); - if (!job->hw_fence.ops) + if (!job->hw_fence.base.ops) kfree(job); else - dma_fence_put(&job->hw_fence); + dma_fence_put(&job->hw_fence.base); } struct dma_fence *amdgpu_job_submit(struct amdgpu_job *job) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h index 529045d60412..2f302266662b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h @@ -48,7 +48,7 @@ struct amdgpu_job { struct drm_sched_job base; struct amdgpu_vm *vm; struct amdgpu_sync explicit_sync; - struct dma_fence hw_fence; + struct amdgpu_fence hw_fence; struct dma_fence *gang_submit; uint32_t preamble_status; uint32_t preemption_status; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c index dda29132dfb2..82d58ac7afb0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c @@ -463,7 +463,8 @@ int amdgpu_jpeg_reg_dump_init(struct amdgpu_device *adev, adev->jpeg.ip_dump = kcalloc(adev->jpeg.num_jpeg_inst * count, sizeof(uint32_t), GFP_KERNEL); if (!adev->jpeg.ip_dump) { - DRM_ERROR("Failed to allocate memory for JPEG IP Dump\n"); + dev_err(adev->dev, + "Failed to allocate memory for JPEG IP Dump\n"); return -ENOMEM; } adev->jpeg.reg_list = reg; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index d2ce7d86dbc8..4c072451a8a1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -1395,6 +1395,8 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) if (r) goto error_pasid; + amdgpu_debugfs_vm_init(file_priv); + r = amdgpu_vm_init(adev, &fpriv->vm, fpriv->xcp_id); if (r) goto error_pasid; @@ -1501,6 +1503,9 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev, amdgpu_vm_bo_del(adev, fpriv->prt_va); amdgpu_bo_unreserve(pd); } + fpriv->evf_mgr.fd_closing = true; + amdgpu_eviction_fence_destroy(&fpriv->evf_mgr); + amdgpu_userq_mgr_fini(&fpriv->userq_mgr); amdgpu_ctx_mgr_fini(&fpriv->ctx_mgr); amdgpu_vm_fini(adev, &fpriv->vm); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c index 6fa9fa11c8f3..135598502c8d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c @@ -47,7 +47,7 @@ static int amdgpu_mes_doorbell_init(struct amdgpu_device *adev) /* Bitmap for dynamic allocation of kernel doorbells */ mes->doorbell_bitmap = bitmap_zalloc(PAGE_SIZE / sizeof(u32), GFP_KERNEL); if (!mes->doorbell_bitmap) { - DRM_ERROR("Failed to allocate MES doorbell bitmap\n"); + dev_err(adev->dev, "Failed to allocate MES doorbell bitmap\n"); return -ENOMEM; } @@ -256,7 +256,7 @@ int amdgpu_mes_suspend(struct amdgpu_device *adev) r = adev->mes.funcs->suspend_gang(&adev->mes, &input); amdgpu_mes_unlock(&adev->mes); if (r) - DRM_ERROR("failed to suspend all gangs"); + dev_err(adev->dev, "failed to suspend all gangs"); return r; } @@ -280,7 +280,7 @@ int amdgpu_mes_resume(struct amdgpu_device *adev) r = adev->mes.funcs->resume_gang(&adev->mes, &input); amdgpu_mes_unlock(&adev->mes); if (r) - DRM_ERROR("failed to resume all gangs"); + dev_err(adev->dev, "failed to resume all gangs"); return r; } @@ -304,7 +304,7 @@ int amdgpu_mes_map_legacy_queue(struct amdgpu_device *adev, r = adev->mes.funcs->map_legacy_queue(&adev->mes, &queue_input); amdgpu_mes_unlock(&adev->mes); if (r) - DRM_ERROR("failed to map legacy queue\n"); + dev_err(adev->dev, "failed to map legacy queue\n"); return r; } @@ -329,7 +329,7 @@ int amdgpu_mes_unmap_legacy_queue(struct amdgpu_device *adev, r = adev->mes.funcs->unmap_legacy_queue(&adev->mes, &queue_input); amdgpu_mes_unlock(&adev->mes); if (r) - DRM_ERROR("failed to unmap legacy queue\n"); + dev_err(adev->dev, "failed to unmap legacy queue\n"); return r; } @@ -361,7 +361,7 @@ int amdgpu_mes_reset_legacy_queue(struct amdgpu_device *adev, r = adev->mes.funcs->reset_hw_queue(&adev->mes, &queue_input); amdgpu_mes_unlock(&adev->mes); if (r) - DRM_ERROR("failed to reset legacy queue\n"); + dev_err(adev->dev, "failed to reset legacy queue\n"); return r; } @@ -469,7 +469,8 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev, int r; if (!adev->mes.funcs->misc_op) { - DRM_ERROR("mes set shader debugger is not supported!\n"); + dev_err(adev->dev, + "mes set shader debugger is not supported!\n"); return -EINVAL; } @@ -493,7 +494,7 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev, r = adev->mes.funcs->misc_op(&adev->mes, &op_input); if (r) - DRM_ERROR("failed to set_shader_debugger\n"); + dev_err(adev->dev, "failed to set_shader_debugger\n"); amdgpu_mes_unlock(&adev->mes); @@ -507,7 +508,8 @@ int amdgpu_mes_flush_shader_debugger(struct amdgpu_device *adev, int r; if (!adev->mes.funcs->misc_op) { - DRM_ERROR("mes flush shader debugger is not supported!\n"); + dev_err(adev->dev, + "mes flush shader debugger is not supported!\n"); return -EINVAL; } @@ -519,7 +521,7 @@ int amdgpu_mes_flush_shader_debugger(struct amdgpu_device *adev, r = adev->mes.funcs->misc_op(&adev->mes, &op_input); if (r) - DRM_ERROR("failed to set_shader_debugger\n"); + dev_err(adev->dev, "failed to set_shader_debugger\n"); amdgpu_mes_unlock(&adev->mes); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c index d085687a47ea..e56ba93a8df6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c @@ -53,6 +53,15 @@ u64 amdgpu_nbio_get_pcie_replay_count(struct amdgpu_device *adev) return 0; } +bool amdgpu_nbio_is_replay_cnt_supported(struct amdgpu_device *adev) +{ + if (amdgpu_sriov_vf(adev) || !adev->asic_funcs->get_pcie_replay_count || + (!adev->nbio.funcs || !adev->nbio.funcs->get_pcie_replay_count)) + return false; + + return true; +} + int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) { int r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h index 79c2f807b9fe..b528de6a01f6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h @@ -119,4 +119,6 @@ int amdgpu_nbio_ras_sw_init(struct amdgpu_device *adev); int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); u64 amdgpu_nbio_get_pcie_replay_count(struct amdgpu_device *adev); +bool amdgpu_nbio_is_replay_cnt_supported(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 73403744331a..c5fda18967c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -32,6 +32,7 @@ #include <linux/list.h> #include <linux/slab.h> #include <linux/dma-buf.h> +#include <linux/export.h> #include <drm/drm_drv.h> #include <drm/amdgpu_drm.h> @@ -62,7 +63,7 @@ static void amdgpu_bo_destroy(struct ttm_buffer_object *tbo) amdgpu_bo_kunmap(bo); - if (bo->tbo.base.import_attach) + if (drm_gem_is_imported(&bo->tbo.base)) drm_prime_gem_destroy(&bo->tbo.base, bo->tbo.sg); drm_gem_object_release(&bo->tbo.base); amdgpu_bo_unref(&bo->parent); @@ -939,7 +940,7 @@ int amdgpu_bo_pin(struct amdgpu_bo *bo, u32 domain) domain = bo->preferred_domains & domain; /* A shared bo cannot be migrated to VRAM */ - if (bo->tbo.base.import_attach) { + if (drm_gem_is_imported(&bo->tbo.base)) { if (domain & AMDGPU_GEM_DOMAIN_GTT) domain = AMDGPU_GEM_DOMAIN_GTT; else @@ -967,7 +968,7 @@ int amdgpu_bo_pin(struct amdgpu_bo *bo, u32 domain) */ domain = amdgpu_bo_get_preferred_domain(adev, domain); - if (bo->tbo.base.import_attach) + if (drm_gem_is_imported(&bo->tbo.base)) dma_buf_pin(bo->tbo.base.import_attach); /* force to pin into visible video ram */ @@ -1018,7 +1019,7 @@ void amdgpu_bo_unpin(struct amdgpu_bo *bo) if (bo->tbo.pin_count) return; - if (bo->tbo.base.import_attach) + if (drm_gem_is_imported(&bo->tbo.base)) dma_buf_unpin(bo->tbo.base.import_attach); if (bo->tbo.resource->mem_type == TTM_PL_VRAM) { @@ -1263,7 +1264,7 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, amdgpu_bo_kunmap(abo); - if (abo->tbo.base.dma_buf && !abo->tbo.base.import_attach && + if (abo->tbo.base.dma_buf && !drm_gem_is_imported(&abo->tbo.base) && old_mem && old_mem->mem_type != TTM_PL_SYSTEM) dma_buf_move_notify(abo->tbo.base.dma_buf); @@ -1473,6 +1474,26 @@ u64 amdgpu_bo_gpu_offset(struct amdgpu_bo *bo) } /** + * amdgpu_bo_fb_aper_addr - return FB aperture GPU offset of the VRAM bo + * @bo: amdgpu VRAM buffer object for which we query the offset + * + * Returns: + * current FB aperture GPU offset of the object. + */ +u64 amdgpu_bo_fb_aper_addr(struct amdgpu_bo *bo) +{ + struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); + uint64_t offset, fb_base; + + WARN_ON_ONCE(bo->tbo.resource->mem_type != TTM_PL_VRAM); + + fb_base = adev->gmc.fb_start; + fb_base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; + offset = (bo->tbo.resource->start << PAGE_SHIFT) + fb_base; + return amdgpu_gmc_sign_extend(offset); +} + +/** * amdgpu_bo_gpu_offset_no_check - return GPU offset of bo * @bo: amdgpu object for which we query the offset * diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 375448627f7b..c316920f3450 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -304,6 +304,7 @@ int amdgpu_bo_sync_wait_resv(struct amdgpu_device *adev, struct dma_resv *resv, bool intr); int amdgpu_bo_sync_wait(struct amdgpu_bo *bo, void *owner, bool intr); u64 amdgpu_bo_gpu_offset(struct amdgpu_bo *bo); +u64 amdgpu_bo_fb_aper_addr(struct amdgpu_bo *bo); u64 amdgpu_bo_gpu_offset_no_check(struct amdgpu_bo *bo); uint32_t amdgpu_bo_mem_stats_placement(struct amdgpu_bo *bo); uint32_t amdgpu_bo_get_preferred_domain(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index e6f0b035e20b..a4bede28df17 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -252,6 +252,7 @@ static int psp_early_init(struct amdgpu_ip_block *ip_block) break; case IP_VERSION(14, 0, 2): case IP_VERSION(14, 0, 3): + adev->psp.sup_ifwi_up = !amdgpu_sriov_vf(adev); psp_v14_0_set_psp_funcs(psp); break; case IP_VERSION(14, 0, 5): @@ -596,6 +597,10 @@ int psp_wait_for(struct psp_context *psp, uint32_t reg_index, udelay(1); } + dev_err(adev->dev, + "psp reg (0x%x) wait timed out, mask: %x, read: %x exp: %x", + reg_index, mask, val, reg_val); + return -ETIME; } @@ -654,6 +659,10 @@ static const char *psp_gfx_cmd_name(enum psp_gfx_cmd_id cmd_id) return "BOOT_CFG"; case GFX_CMD_ID_CONFIG_SQ_PERFMON: return "CONFIG_SQ_PERFMON"; + case GFX_CMD_ID_FB_FW_RESERV_ADDR: + return "FB_FW_RESERV_ADDR"; + case GFX_CMD_ID_FB_FW_RESERV_EXT_ADDR: + return "FB_FW_RESERV_EXT_ADDR"; default: return "UNKNOWN CMD"; } @@ -871,6 +880,8 @@ static int psp_tmr_init(struct psp_context *psp) &psp->tmr_bo, &psp->tmr_mc_addr, pptr); } + if (amdgpu_virt_xgmi_migrate_enabled(psp->adev) && psp->tmr_bo) + psp->tmr_mc_addr = amdgpu_bo_fb_aper_addr(psp->tmr_bo); return ret; } @@ -984,6 +995,93 @@ int psp_get_fw_attestation_records_addr(struct psp_context *psp, return ret; } +static int psp_get_fw_reservation_info(struct psp_context *psp, + uint32_t cmd_id, + uint64_t *addr, + uint32_t *size) +{ + int ret; + uint32_t status; + struct psp_gfx_cmd_resp *cmd; + + cmd = acquire_psp_cmd_buf(psp); + + cmd->cmd_id = cmd_id; + + ret = psp_cmd_submit_buf(psp, NULL, cmd, + psp->fence_buf_mc_addr); + if (ret) { + release_psp_cmd_buf(psp); + return ret; + } + + status = cmd->resp.status; + if (status == PSP_ERR_UNKNOWN_COMMAND) { + release_psp_cmd_buf(psp); + *addr = 0; + *size = 0; + return 0; + } + + *addr = (uint64_t)cmd->resp.uresp.fw_reserve_info.reserve_base_address_hi << 32 | + cmd->resp.uresp.fw_reserve_info.reserve_base_address_lo; + *size = cmd->resp.uresp.fw_reserve_info.reserve_size; + + release_psp_cmd_buf(psp); + + return 0; +} + +int psp_update_fw_reservation(struct psp_context *psp) +{ + int ret; + uint64_t reserv_addr, reserv_addr_ext; + uint32_t reserv_size, reserv_size_ext; + struct amdgpu_device *adev = psp->adev; + + if (amdgpu_sriov_vf(psp->adev)) + return 0; + + if ((amdgpu_ip_version(adev, MP0_HWIP, 0) != IP_VERSION(14, 0, 2)) && + (amdgpu_ip_version(adev, MP0_HWIP, 0) != IP_VERSION(14, 0, 3))) + return 0; + + ret = psp_get_fw_reservation_info(psp, GFX_CMD_ID_FB_FW_RESERV_ADDR, &reserv_addr, &reserv_size); + if (ret) + return ret; + ret = psp_get_fw_reservation_info(psp, GFX_CMD_ID_FB_FW_RESERV_EXT_ADDR, &reserv_addr_ext, &reserv_size_ext); + if (ret) + return ret; + + if (reserv_addr != adev->gmc.real_vram_size - reserv_size) { + dev_warn(adev->dev, "reserve fw region is not valid!\n"); + return 0; + } + + amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, NULL, NULL); + + reserv_size = roundup(reserv_size, SZ_1M); + + ret = amdgpu_bo_create_kernel_at(adev, reserv_addr, reserv_size, &adev->mman.fw_reserved_memory, NULL); + if (ret) { + dev_err(adev->dev, "reserve fw region failed(%d)!\n", ret); + amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, NULL, NULL); + return ret; + } + + reserv_size_ext = roundup(reserv_size_ext, SZ_1M); + + ret = amdgpu_bo_create_kernel_at(adev, reserv_addr_ext, reserv_size_ext, + &adev->mman.fw_reserved_memory_extend, NULL); + if (ret) { + dev_err(adev->dev, "reserve extend fw region failed(%d)!\n", ret); + amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory_extend, NULL, NULL); + return ret; + } + + return 0; +} + static int psp_boot_config_get(struct amdgpu_device *adev, uint32_t *boot_cfg) { struct psp_context *psp = &adev->psp; @@ -1270,6 +1368,11 @@ int psp_ta_load(struct psp_context *psp, struct ta_context *context) psp_copy_fw(psp, context->bin_desc.start_addr, context->bin_desc.size_bytes); + if (amdgpu_virt_xgmi_migrate_enabled(psp->adev) && + context->mem_context.shared_bo) + context->mem_context.shared_mc_addr = + amdgpu_bo_fb_aper_addr(context->mem_context.shared_bo); + psp_prep_ta_load_cmd_buf(cmd, psp->fw_pri_mc_addr, context); ret = psp_cmd_submit_buf(psp, NULL, cmd, @@ -2337,11 +2440,27 @@ bool amdgpu_psp_tos_reload_needed(struct amdgpu_device *adev) return false; } +static void psp_update_gpu_addresses(struct amdgpu_device *adev) +{ + struct psp_context *psp = &adev->psp; + + if (psp->cmd_buf_bo && psp->cmd_buf_mem) { + psp->fw_pri_mc_addr = amdgpu_bo_fb_aper_addr(psp->fw_pri_bo); + psp->fence_buf_mc_addr = amdgpu_bo_fb_aper_addr(psp->fence_buf_bo); + psp->cmd_buf_mc_addr = amdgpu_bo_fb_aper_addr(psp->cmd_buf_bo); + } + if (adev->firmware.rbuf && psp->km_ring.ring_mem) + psp->km_ring.ring_mem_mc_addr = amdgpu_bo_fb_aper_addr(adev->firmware.rbuf); +} + static int psp_hw_start(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; int ret; + if (amdgpu_virt_xgmi_migrate_enabled(adev)) + psp_update_gpu_addresses(adev); + if (!amdgpu_sriov_vf(adev)) { if ((is_psp_fw_valid(psp->kdb)) && (psp->funcs->bootloader_load_kdb != NULL)) { @@ -2440,6 +2559,14 @@ static int psp_hw_start(struct psp_context *psp) return ret; } + if (!amdgpu_in_reset(adev) && !adev->in_suspend) { + ret = psp_update_fw_reservation(psp); + if (ret) { + dev_err(adev->dev, "update fw reservation failed!\n"); + return ret; + } + } + if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) goto skip_pin_bo; @@ -3522,8 +3649,12 @@ int psp_init_sos_microcode(struct psp_context *psp, const char *chip_name) uint8_t *ucode_array_start_addr; int err = 0; - err = amdgpu_ucode_request(adev, &adev->psp.sos_fw, AMDGPU_UCODE_REQUIRED, - "amdgpu/%s_sos.bin", chip_name); + if (amdgpu_is_kicker_fw(adev)) + err = amdgpu_ucode_request(adev, &adev->psp.sos_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_sos_kicker.bin", chip_name); + else + err = amdgpu_ucode_request(adev, &adev->psp.sos_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_sos.bin", chip_name); if (err) goto out; @@ -3799,8 +3930,12 @@ int psp_init_ta_microcode(struct psp_context *psp, const char *chip_name) struct amdgpu_device *adev = psp->adev; int err; - err = amdgpu_ucode_request(adev, &adev->psp.ta_fw, AMDGPU_UCODE_REQUIRED, - "amdgpu/%s_ta.bin", chip_name); + if (amdgpu_is_kicker_fw(adev)) + err = amdgpu_ucode_request(adev, &adev->psp.ta_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_ta_kicker.bin", chip_name); + else + err = amdgpu_ucode_request(adev, &adev->psp.ta_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_ta.bin", chip_name); if (err) return err; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index 428adc7f741d..4bc0ec49d2e9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -51,6 +51,17 @@ #define C2PMSG_CMD_SPI_GET_ROM_IMAGE_ADDR_HI 0x10 #define C2PMSG_CMD_SPI_GET_FLASH_IMAGE 0x11 +/* Command register bit 31 set to indicate readiness */ +#define MBOX_TOS_READY_FLAG (GFX_FLAG_RESPONSE) +#define MBOX_TOS_READY_MASK (GFX_CMD_RESPONSE_MASK | GFX_CMD_STATUS_MASK) + +/* Values to check for a successful GFX_CMD response wait. Check against + * both status bits and response state - helps to detect a command failure + * or other unexpected cases like a device drop reading all 0xFFs + */ +#define MBOX_TOS_RESP_FLAG (GFX_FLAG_RESPONSE) +#define MBOX_TOS_RESP_MASK (GFX_CMD_RESPONSE_MASK | GFX_CMD_STATUS_MASK) + extern const struct attribute_group amdgpu_flash_attr_group; enum psp_shared_mem_size { @@ -588,7 +599,7 @@ int psp_init_cap_microcode(struct psp_context *psp, const char *chip_name); int psp_get_fw_attestation_records_addr(struct psp_context *psp, uint64_t *output_ptr); - +int psp_update_fw_reservation(struct psp_context *psp); int psp_load_fw_list(struct psp_context *psp, struct amdgpu_firmware_info **ucode_list, int ucode_count); void psp_copy_fw(struct psp_context *psp, uint8_t *start_addr, uint32_t bin_size); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index de0944947eaf..c508697a3412 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1107,6 +1107,9 @@ static void amdgpu_ras_error_print_error_data(struct amdgpu_device *adev, err_info->de_count, blk_name); } } else { + if (adev->debug_disable_ce_logs) + return; + for_each_ras_error(err_node, err_data) { err_info = &err_node->err_info; mcm_info = &err_info->mcm_info; @@ -3003,6 +3006,15 @@ int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev, return 0; } + if (!con->eeprom_control.is_eeprom_valid) { + dev_warn(adev->dev, + "Failed to save EEPROM table data because of EEPROM data corruption!"); + if (new_cnt) + *new_cnt = 0; + + return 0; + } + mutex_lock(&con->recovery_lock); control = &con->eeprom_control; data = con->eh_data; @@ -3294,7 +3306,6 @@ static int amdgpu_ras_poison_creation_handler(struct amdgpu_device *adev, uint64_t de_queried_count; uint32_t new_detect_count, total_detect_count; uint32_t need_query_count = poison_creation_count; - bool query_data_timeout = false; enum ras_event_type type = RAS_EVENT_TYPE_POISON_CREATION; memset(&info, 0, sizeof(info)); @@ -3323,21 +3334,13 @@ static int amdgpu_ras_poison_creation_handler(struct amdgpu_device *adev, timeout = MAX_UMC_POISON_POLLING_TIME_ASYNC; if (timeout) { - if (!--timeout) { - query_data_timeout = true; + if (!--timeout) break; - } msleep(1); } } } while (total_detect_count < need_query_count); - if (query_data_timeout) { - dev_warn(adev->dev, "Can't find deferred error! count: %u\n", - (need_query_count - total_detect_count)); - return -ENOENT; - } - if (total_detect_count) schedule_delayed_work(&ras->page_retirement_dwork, 0); @@ -3488,8 +3491,7 @@ int amdgpu_ras_init_badpage_info(struct amdgpu_device *adev) control = &con->eeprom_control; ret = amdgpu_ras_eeprom_init(control); - if (ret) - return ret; + control->is_eeprom_valid = !ret; if (!adev->umc.ras || !adev->umc.ras->convert_ras_err_addr) control->ras_num_pa_recs = control->ras_num_recs; @@ -3498,10 +3500,12 @@ int amdgpu_ras_init_badpage_info(struct amdgpu_device *adev) adev->umc.ras->get_retire_flip_bits) adev->umc.ras->get_retire_flip_bits(adev); - if (control->ras_num_recs) { + if (control->ras_num_recs && control->is_eeprom_valid) { ret = amdgpu_ras_load_bad_pages(adev); - if (ret) - return ret; + if (ret) { + control->is_eeprom_valid = false; + return 0; + } amdgpu_dpm_send_hbm_bad_pages_num( adev, control->ras_num_bad_pages); @@ -3520,7 +3524,7 @@ int amdgpu_ras_init_badpage_info(struct amdgpu_device *adev) dev_warn(adev->dev, "Failed to format RAS EEPROM data in V3 version!\n"); } - return ret; + return 0; } int amdgpu_ras_recovery_init(struct amdgpu_device *adev, bool init_bp_info) @@ -4414,8 +4418,10 @@ void amdgpu_ras_clear_err_state(struct amdgpu_device *adev) struct amdgpu_ras *ras; ras = amdgpu_ras_get_context(adev); - if (ras) + if (ras) { ras->ras_err_state = 0; + ras->gpu_reset_flags = 0; + } } void amdgpu_ras_set_err_poison(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index 2c58e09e56f9..54838746f97d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -277,10 +277,11 @@ static int __write_table_header(struct amdgpu_ras_eeprom_control *control) up_read(&adev->reset_domain->sem); if (res < 0) { - DRM_ERROR("Failed to write EEPROM table header:%d", res); + dev_err(adev->dev, "Failed to write EEPROM table header:%d", + res); } else if (res < RAS_TABLE_HEADER_SIZE) { - DRM_ERROR("Short write:%d out of %d\n", - res, RAS_TABLE_HEADER_SIZE); + dev_err(adev->dev, "Short write:%d out of %d\n", res, + RAS_TABLE_HEADER_SIZE); res = -EIO; } else { res = 0; @@ -323,7 +324,8 @@ static int __write_table_ras_info(struct amdgpu_ras_eeprom_control *control) buf = kzalloc(RAS_TABLE_V2_1_INFO_SIZE, GFP_KERNEL); if (!buf) { - DRM_ERROR("Failed to alloc buf to write table ras info\n"); + dev_err(adev->dev, + "Failed to alloc buf to write table ras info\n"); return -ENOMEM; } @@ -338,10 +340,11 @@ static int __write_table_ras_info(struct amdgpu_ras_eeprom_control *control) up_read(&adev->reset_domain->sem); if (res < 0) { - DRM_ERROR("Failed to write EEPROM table ras info:%d", res); + dev_err(adev->dev, "Failed to write EEPROM table ras info:%d", + res); } else if (res < RAS_TABLE_V2_1_INFO_SIZE) { - DRM_ERROR("Short write:%d out of %d\n", - res, RAS_TABLE_V2_1_INFO_SIZE); + dev_err(adev->dev, "Short write:%d out of %d\n", res, + RAS_TABLE_V2_1_INFO_SIZE); res = -EIO; } else { res = 0; @@ -476,6 +479,8 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control) control->ras_num_recs = 0; control->ras_num_bad_pages = 0; + control->ras_num_mca_recs = 0; + control->ras_num_pa_recs = 0; control->ras_fri = 0; amdgpu_dpm_send_hbm_bad_pages_num(adev, control->ras_num_bad_pages); @@ -607,13 +612,13 @@ static int __amdgpu_ras_eeprom_write(struct amdgpu_ras_eeprom_control *control, buf, buf_size); up_read(&adev->reset_domain->sem); if (res < 0) { - DRM_ERROR("Writing %d EEPROM table records error:%d", - num, res); + dev_err(adev->dev, "Writing %d EEPROM table records error:%d", + num, res); } else if (res < buf_size) { /* Short write, return error. */ - DRM_ERROR("Wrote %d records out of %d", - res / RAS_TABLE_RECORD_SIZE, num); + dev_err(adev->dev, "Wrote %d records out of %d", + res / RAS_TABLE_RECORD_SIZE, num); res = -EIO; } else { res = 0; @@ -761,18 +766,17 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) dev_warn(adev->dev, "Saved bad pages %d reaches threshold value %d\n", control->ras_num_bad_pages, ras->bad_page_cnt_threshold); - control->tbl_hdr.header = RAS_TABLE_HDR_BAD; - if (control->tbl_hdr.version >= RAS_TABLE_VER_V2_1) { - control->tbl_rai.rma_status = GPU_RETIRED__ECC_REACH_THRESHOLD; - control->tbl_rai.health_percent = 0; - } - if ((amdgpu_bad_page_threshold != -1) && - (amdgpu_bad_page_threshold != -2)) + (amdgpu_bad_page_threshold != -2)) { + control->tbl_hdr.header = RAS_TABLE_HDR_BAD; + if (control->tbl_hdr.version >= RAS_TABLE_VER_V2_1) { + control->tbl_rai.rma_status = GPU_RETIRED__ECC_REACH_THRESHOLD; + control->tbl_rai.health_percent = 0; + } ras->is_rma = true; - - /* ignore the -ENOTSUPP return value */ - amdgpu_dpm_send_rma_reason(adev); + /* ignore the -ENOTSUPP return value */ + amdgpu_dpm_send_rma_reason(adev); + } } if (control->tbl_hdr.version >= RAS_TABLE_VER_V2_1) @@ -787,8 +791,9 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) buf_size = control->ras_num_recs * RAS_TABLE_RECORD_SIZE; buf = kcalloc(control->ras_num_recs, RAS_TABLE_RECORD_SIZE, GFP_KERNEL); if (!buf) { - DRM_ERROR("allocating memory for table of size %d bytes failed\n", - control->tbl_hdr.tbl_size); + dev_err(adev->dev, + "allocating memory for table of size %d bytes failed\n", + control->tbl_hdr.tbl_size); res = -ENOMEM; goto Out; } @@ -800,12 +805,11 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) buf, buf_size); up_read(&adev->reset_domain->sem); if (res < 0) { - DRM_ERROR("EEPROM failed reading records:%d\n", - res); + dev_err(adev->dev, "EEPROM failed reading records:%d\n", res); goto Out; } else if (res < buf_size) { - DRM_ERROR("EEPROM read %d out of %d bytes\n", - res, buf_size); + dev_err(adev->dev, "EEPROM read %d out of %d bytes\n", res, + buf_size); res = -EIO; goto Out; } @@ -866,11 +870,12 @@ int amdgpu_ras_eeprom_append(struct amdgpu_ras_eeprom_control *control, return 0; if (num == 0) { - DRM_ERROR("will not append 0 records\n"); + dev_err(adev->dev, "will not append 0 records\n"); return -EINVAL; } else if (num > control->ras_max_record_count) { - DRM_ERROR("cannot append %d records than the size of table %d\n", - num, control->ras_max_record_count); + dev_err(adev->dev, + "cannot append %d records than the size of table %d\n", + num, control->ras_max_record_count); return -EINVAL; } @@ -924,13 +929,13 @@ static int __amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control, buf, buf_size); up_read(&adev->reset_domain->sem); if (res < 0) { - DRM_ERROR("Reading %d EEPROM table records error:%d", - num, res); + dev_err(adev->dev, "Reading %d EEPROM table records error:%d", + num, res); } else if (res < buf_size) { /* Short read, return error. */ - DRM_ERROR("Read %d records out of %d", - res / RAS_TABLE_RECORD_SIZE, num); + dev_err(adev->dev, "Read %d records out of %d", + res / RAS_TABLE_RECORD_SIZE, num); res = -EIO; } else { res = 0; @@ -964,11 +969,11 @@ int amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control, return 0; if (num == 0) { - DRM_ERROR("will not read 0 records\n"); + dev_err(adev->dev, "will not read 0 records\n"); return -EINVAL; } else if (num > control->ras_num_recs) { - DRM_ERROR("too many records to read:%d available:%d\n", - num, control->ras_num_recs); + dev_err(adev->dev, "too many records to read:%d available:%d\n", + num, control->ras_num_recs); return -EINVAL; } @@ -1300,7 +1305,8 @@ static int __verify_ras_table_checksum(struct amdgpu_ras_eeprom_control *control buf = kzalloc(buf_size, GFP_KERNEL); if (!buf) { - DRM_ERROR("Out of memory checking RAS table checksum.\n"); + dev_err(adev->dev, + "Out of memory checking RAS table checksum.\n"); return -ENOMEM; } @@ -1309,7 +1315,7 @@ static int __verify_ras_table_checksum(struct amdgpu_ras_eeprom_control *control control->ras_header_offset, buf, buf_size); if (res < buf_size) { - DRM_ERROR("Partial read for checksum, res:%d\n", res); + dev_err(adev->dev, "Partial read for checksum, res:%d\n", res); /* On partial reads, return -EIO. */ if (res >= 0) @@ -1334,7 +1340,8 @@ static int __read_table_ras_info(struct amdgpu_ras_eeprom_control *control) buf = kzalloc(RAS_TABLE_V2_1_INFO_SIZE, GFP_KERNEL); if (!buf) { - DRM_ERROR("Failed to alloc buf to read EEPROM table ras info\n"); + dev_err(adev->dev, + "Failed to alloc buf to read EEPROM table ras info\n"); return -ENOMEM; } @@ -1346,7 +1353,8 @@ static int __read_table_ras_info(struct amdgpu_ras_eeprom_control *control) control->i2c_address + control->ras_info_offset, buf, RAS_TABLE_V2_1_INFO_SIZE); if (res < RAS_TABLE_V2_1_INFO_SIZE) { - DRM_ERROR("Failed to read EEPROM table ras info, res:%d", res); + dev_err(adev->dev, + "Failed to read EEPROM table ras info, res:%d", res); res = res >= 0 ? -EIO : res; goto Out; } @@ -1387,7 +1395,8 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control) control->i2c_address + control->ras_header_offset, buf, RAS_TABLE_HEADER_SIZE); if (res < RAS_TABLE_HEADER_SIZE) { - DRM_ERROR("Failed to read EEPROM table header, res:%d", res); + dev_err(adev->dev, "Failed to read EEPROM table header, res:%d", + res); return res >= 0 ? -EIO : res; } @@ -1452,8 +1461,9 @@ int amdgpu_ras_eeprom_check(struct amdgpu_ras_eeprom_control *control) control->ras_num_mca_recs * adev->umc.retire_unit; if (hdr->header == RAS_TABLE_HDR_VAL) { - DRM_DEBUG_DRIVER("Found existing EEPROM table with %d records", - control->ras_num_bad_pages); + dev_dbg(adev->dev, + "Found existing EEPROM table with %d records", + control->ras_num_bad_pages); if (hdr->version >= RAS_TABLE_VER_V2_1) { res = __read_table_ras_info(control); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h index ec6d7ea37ad0..35c69ac3dbeb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h @@ -114,6 +114,8 @@ struct amdgpu_ras_eeprom_control { /* Record channel info which occurred bad pages */ u32 bad_channel_bitmap; + + bool is_eeprom_valid; }; /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index b95b47110769..784ba2ec354c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -127,6 +127,22 @@ struct amdgpu_fence_driver { struct dma_fence **fences; }; +/* + * Fences mark an event in the GPUs pipeline and are used + * for GPU/CPU synchronization. When the fence is written, + * it is expected that all buffers associated with that fence + * are no longer in use by the associated ring on the GPU and + * that the relevant GPU caches have been flushed. + */ + +struct amdgpu_fence { + struct dma_fence base; + + /* RB, DMA, etc. */ + struct amdgpu_ring *ring; + ktime_t start_timestamp; +}; + extern const struct drm_sched_backend_ops amdgpu_sched_ops; void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring); @@ -141,8 +157,8 @@ void amdgpu_fence_driver_hw_init(struct amdgpu_device *adev); void amdgpu_fence_driver_hw_fini(struct amdgpu_device *adev); int amdgpu_fence_driver_sw_init(struct amdgpu_device *adev); void amdgpu_fence_driver_sw_fini(struct amdgpu_device *adev); -int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **fence, struct amdgpu_job *job, - unsigned flags); +int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, + struct amdgpu_fence *af, unsigned int flags); int amdgpu_fence_emit_polling(struct amdgpu_ring *ring, uint32_t *s, uint32_t timeout); bool amdgpu_fence_process(struct amdgpu_ring *ring); @@ -252,9 +268,9 @@ struct amdgpu_ring_funcs { void (*patch_cntl)(struct amdgpu_ring *ring, unsigned offset); void (*patch_ce)(struct amdgpu_ring *ring, unsigned offset); void (*patch_de)(struct amdgpu_ring *ring, unsigned offset); - int (*reset)(struct amdgpu_ring *ring, unsigned int vmid); + int (*reset)(struct amdgpu_ring *ring, unsigned int vmid, + struct amdgpu_fence *timedout_fence); void (*emit_cleaner_shader)(struct amdgpu_ring *ring); - bool (*is_guilty)(struct amdgpu_ring *ring); }; /** @@ -409,7 +425,7 @@ struct amdgpu_ring { #define amdgpu_ring_patch_cntl(r, o) ((r)->funcs->patch_cntl((r), (o))) #define amdgpu_ring_patch_ce(r, o) ((r)->funcs->patch_ce((r), (o))) #define amdgpu_ring_patch_de(r, o) ((r)->funcs->patch_de((r), (o))) -#define amdgpu_ring_reset(r, v) (r)->funcs->reset((r), (v)) +#define amdgpu_ring_reset(r, v, f) (r)->funcs->reset((r), (v), (f)) unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type); int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned ndw); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index 6716ac281c49..7e26a44dcc1f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -534,41 +534,17 @@ bool amdgpu_sdma_is_shared_inv_eng(struct amdgpu_device *adev, struct amdgpu_rin static int amdgpu_sdma_soft_reset(struct amdgpu_device *adev, u32 instance_id) { struct amdgpu_sdma_instance *sdma_instance = &adev->sdma.instance[instance_id]; - int r = -EOPNOTSUPP; - - switch (amdgpu_ip_version(adev, SDMA0_HWIP, 0)) { - case IP_VERSION(4, 4, 2): - case IP_VERSION(4, 4, 4): - case IP_VERSION(4, 4, 5): - /* For SDMA 4.x, use the existing DPM interface for backward compatibility */ - r = amdgpu_dpm_reset_sdma(adev, 1 << instance_id); - break; - case IP_VERSION(5, 0, 0): - case IP_VERSION(5, 0, 1): - case IP_VERSION(5, 0, 2): - case IP_VERSION(5, 0, 5): - case IP_VERSION(5, 2, 0): - case IP_VERSION(5, 2, 2): - case IP_VERSION(5, 2, 4): - case IP_VERSION(5, 2, 5): - case IP_VERSION(5, 2, 6): - case IP_VERSION(5, 2, 3): - case IP_VERSION(5, 2, 1): - case IP_VERSION(5, 2, 7): - if (sdma_instance->funcs->soft_reset_kernel_queue) - r = sdma_instance->funcs->soft_reset_kernel_queue(adev, instance_id); - break; - default: - break; - } - return r; + if (sdma_instance->funcs->soft_reset_kernel_queue) + return sdma_instance->funcs->soft_reset_kernel_queue(adev, instance_id); + + return -EOPNOTSUPP; } /** * amdgpu_sdma_reset_engine - Reset a specific SDMA engine * @adev: Pointer to the AMDGPU device - * @instance_id: ID of the SDMA engine instance to reset + * @instance_id: Logical ID of the SDMA engine instance to reset * * Returns: 0 on success, or a negative error code on failure. */ @@ -578,35 +554,35 @@ int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id) struct amdgpu_sdma_instance *sdma_instance = &adev->sdma.instance[instance_id]; struct amdgpu_ring *gfx_ring = &sdma_instance->ring; struct amdgpu_ring *page_ring = &sdma_instance->page; - bool gfx_sched_stopped = false, page_sched_stopped = false; mutex_lock(&sdma_instance->engine_reset_mutex); /* Stop the scheduler's work queue for the GFX and page rings if they are running. * This ensures that no new tasks are submitted to the queues while * the reset is in progress. */ - if (!amdgpu_ring_sched_ready(gfx_ring)) { - drm_sched_wqueue_stop(&gfx_ring->sched); - gfx_sched_stopped = true; - } + drm_sched_wqueue_stop(&gfx_ring->sched); - if (adev->sdma.has_page_queue && !amdgpu_ring_sched_ready(page_ring)) { + if (adev->sdma.has_page_queue) drm_sched_wqueue_stop(&page_ring->sched); - page_sched_stopped = true; - } - if (sdma_instance->funcs->stop_kernel_queue) + if (sdma_instance->funcs->stop_kernel_queue) { sdma_instance->funcs->stop_kernel_queue(gfx_ring); + if (adev->sdma.has_page_queue) + sdma_instance->funcs->stop_kernel_queue(page_ring); + } /* Perform the SDMA reset for the specified instance */ ret = amdgpu_sdma_soft_reset(adev, instance_id); if (ret) { - dev_err(adev->dev, "Failed to reset SDMA instance %u\n", instance_id); + dev_err(adev->dev, "Failed to reset SDMA logical instance %u\n", instance_id); goto exit; } - if (sdma_instance->funcs->start_kernel_queue) + if (sdma_instance->funcs->start_kernel_queue) { sdma_instance->funcs->start_kernel_queue(gfx_ring); + if (adev->sdma.has_page_queue) + sdma_instance->funcs->start_kernel_queue(page_ring); + } exit: /* Restart the scheduler's work queue for the GFX and page rings @@ -614,12 +590,9 @@ exit: * to be submitted to the queues after the reset is complete. */ if (!ret) { - if (gfx_sched_stopped && amdgpu_ring_sched_ready(gfx_ring)) { - drm_sched_wqueue_start(&gfx_ring->sched); - } - if (page_sched_stopped && amdgpu_ring_sched_ready(page_ring)) { + drm_sched_wqueue_start(&gfx_ring->sched); + if (adev->sdma.has_page_queue) drm_sched_wqueue_start(&page_ring->sched); - } } mutex_unlock(&sdma_instance->engine_reset_mutex); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 9c5df35f05b7..27ab4e754b2a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -299,7 +299,8 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, struct amdgpu_bo *abo_src, *abo_dst; if (!adev->mman.buffer_funcs_enabled) { - DRM_ERROR("Trying to move memory with ring turned off.\n"); + dev_err(adev->dev, + "Trying to move memory with ring turned off.\n"); return -EINVAL; } @@ -934,7 +935,7 @@ static int amdgpu_ttm_backend_bind(struct ttm_device *bdev, if (gtt->userptr) { r = amdgpu_ttm_tt_pin_userptr(bdev, ttm); if (r) { - DRM_ERROR("failed to pin userptr\n"); + dev_err(adev->dev, "failed to pin userptr\n"); return r; } } else if (ttm->page_flags & TTM_TT_FLAG_EXTERNAL) { @@ -1060,7 +1061,7 @@ static void amdgpu_ttm_backend_unbind(struct ttm_device *bdev, /* if the pages have userptr pinning then clear that first */ if (gtt->userptr) { amdgpu_ttm_tt_unpin_userptr(bdev, ttm); - } else if (ttm->sg && gtt->gobj->import_attach) { + } else if (ttm->sg && drm_gem_is_imported(gtt->gobj)) { struct dma_buf_attachment *attach; attach = gtt->gobj->import_attach; @@ -1781,7 +1782,7 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev) &ctx->c2p_bo, NULL); if (ret) { - DRM_ERROR("alloc c2p_bo failed(%d)!\n", ret); + dev_err(adev->dev, "alloc c2p_bo failed(%d)!\n", ret); amdgpu_ttm_training_reserve_vram_fini(adev); return ret; } @@ -1793,7 +1794,7 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev) adev, adev->gmc.real_vram_size - reserve_size, reserve_size, &adev->mman.fw_reserved_memory, NULL); if (ret) { - DRM_ERROR("alloc tmr failed(%d)!\n", ret); + dev_err(adev->dev, "alloc tmr failed(%d)!\n", ret); amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, NULL, NULL); return ret; @@ -1864,13 +1865,14 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) adev->need_swiotlb, dma_addressing_limited(adev->dev)); if (r) { - DRM_ERROR("failed initializing buffer object driver(%d).\n", r); + dev_err(adev->dev, + "failed initializing buffer object driver(%d).\n", r); return r; } r = amdgpu_ttm_pools_init(adev); if (r) { - DRM_ERROR("failed to init ttm pools(%d).\n", r); + dev_err(adev->dev, "failed to init ttm pools(%d).\n", r); return r; } adev->mman.initialized = true; @@ -1878,7 +1880,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) /* Initialize VRAM pool with all of VRAM divided into pages */ r = amdgpu_vram_mgr_init(adev); if (r) { - DRM_ERROR("Failed initializing VRAM heap.\n"); + dev_err(adev->dev, "Failed initializing VRAM heap.\n"); return r; } @@ -1958,7 +1960,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) DRM_DEBUG_DRIVER("Skipped stolen memory reservation\n"); } - DRM_INFO("amdgpu: %uM of VRAM memory ready\n", + dev_info(adev->dev, "amdgpu: %uM of VRAM memory ready\n", (unsigned int)(adev->gmc.real_vram_size / (1024 * 1024))); /* Compute GTT size, either based on TTM limit @@ -1981,10 +1983,10 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) /* Initialize GTT memory pool */ r = amdgpu_gtt_mgr_init(adev, gtt_size); if (r) { - DRM_ERROR("Failed initializing GTT heap.\n"); + dev_err(adev->dev, "Failed initializing GTT heap.\n"); return r; } - DRM_INFO("amdgpu: %uM of GTT memory ready.\n", + dev_info(adev->dev, "amdgpu: %uM of GTT memory ready.\n", (unsigned int)(gtt_size / (1024 * 1024))); if (adev->flags & AMD_IS_APU) { @@ -1995,40 +1997,40 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) /* Initialize doorbell pool on PCI BAR */ r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_DOORBELL, adev->doorbell.size / PAGE_SIZE); if (r) { - DRM_ERROR("Failed initializing doorbell heap.\n"); + dev_err(adev->dev, "Failed initializing doorbell heap.\n"); return r; } /* Create a boorbell page for kernel usages */ r = amdgpu_doorbell_create_kernel_doorbells(adev); if (r) { - DRM_ERROR("Failed to initialize kernel doorbells.\n"); + dev_err(adev->dev, "Failed to initialize kernel doorbells.\n"); return r; } /* Initialize preemptible memory pool */ r = amdgpu_preempt_mgr_init(adev); if (r) { - DRM_ERROR("Failed initializing PREEMPT heap.\n"); + dev_err(adev->dev, "Failed initializing PREEMPT heap.\n"); return r; } /* Initialize various on-chip memory pools */ r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_GDS, adev->gds.gds_size); if (r) { - DRM_ERROR("Failed initializing GDS heap.\n"); + dev_err(adev->dev, "Failed initializing GDS heap.\n"); return r; } r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_GWS, adev->gds.gws_size); if (r) { - DRM_ERROR("Failed initializing gws heap.\n"); + dev_err(adev->dev, "Failed initializing gws heap.\n"); return r; } r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_OA, adev->gds.oa_size); if (r) { - DRM_ERROR("Failed initializing oa heap.\n"); + dev_err(adev->dev, "Failed initializing oa heap.\n"); return r; } if (amdgpu_bo_create_kernel(adev, PAGE_SIZE, PAGE_SIZE, @@ -2060,6 +2062,8 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev) /* return the FW reserved memory back to VRAM */ amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, NULL, NULL); + amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory_extend, NULL, + NULL); if (adev->mman.stolen_reserved_size) amdgpu_bo_free_kernel(&adev->mman.stolen_reserved_memory, NULL, NULL); @@ -2089,7 +2093,7 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev) ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_DOORBELL); ttm_device_fini(&adev->mman.bdev); adev->mman.initialized = false; - DRM_INFO("amdgpu: ttm finalized\n"); + dev_info(adev->dev, "amdgpu: ttm finalized\n"); } /** @@ -2121,8 +2125,9 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) DRM_SCHED_PRIORITY_KERNEL, &sched, 1, NULL); if (r) { - DRM_ERROR("Failed setting up TTM BO move entity (%d)\n", - r); + dev_err(adev->dev, + "Failed setting up TTM BO move entity (%d)\n", + r); return; } @@ -2130,8 +2135,9 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) DRM_SCHED_PRIORITY_NORMAL, &sched, 1, NULL); if (r) { - DRM_ERROR("Failed setting up TTM BO move entity (%d)\n", - r); + dev_err(adev->dev, + "Failed setting up TTM BO move entity (%d)\n", + r); goto error_free_entity; } } else { @@ -2202,7 +2208,8 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, int r; if (!direct_submit && !ring->sched.ready) { - DRM_ERROR("Trying to move memory with ring turned off.\n"); + dev_err(adev->dev, + "Trying to move memory with ring turned off.\n"); return -EINVAL; } @@ -2237,7 +2244,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, error_free: amdgpu_job_free(job); - DRM_ERROR("Error scheduling IBs (%d)\n", r); + dev_err(adev->dev, "Error scheduling IBs (%d)\n", r); return r; } @@ -2356,7 +2363,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo, int r; if (!adev->mman.buffer_funcs_enabled) { - DRM_ERROR("Trying to clear memory with ring turned off.\n"); + dev_err(adev->dev, + "Trying to clear memory with ring turned off.\n"); return -EINVAL; } @@ -2416,7 +2424,7 @@ int amdgpu_ttm_evict_resources(struct amdgpu_device *adev, int mem_type) man = ttm_manager_type(&adev->mman.bdev, mem_type); break; default: - DRM_ERROR("Trying to evict invalid memory type\n"); + dev_err(adev->dev, "Trying to evict invalid memory type\n"); return -EINVAL; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index 208b7d1d8a27..215c198e4aff 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -86,6 +86,7 @@ struct amdgpu_mman { uint32_t discovery_tmr_size; /* fw reserved memory */ struct amdgpu_bo *fw_reserved_memory; + struct amdgpu_bo *fw_reserved_memory_extend; /* firmware VRAM reservation */ u64 fw_vram_usage_start_offset; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index 2505c46a9c3d..a0b50a8ac9c4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -30,6 +30,10 @@ #define AMDGPU_UCODE_NAME_MAX (128) +static const struct kicker_device kicker_device_list[] = { + {0x744B, 0x00}, +}; + static void amdgpu_ucode_print_common_hdr(const struct common_firmware_header *hdr) { DRM_DEBUG("size_bytes: %u\n", le32_to_cpu(hdr->size_bytes)); @@ -1155,6 +1159,9 @@ int amdgpu_ucode_init_bo(struct amdgpu_device *adev) adev->firmware.max_ucodes = AMDGPU_UCODE_ID_MAXIMUM; } + if (amdgpu_virt_xgmi_migrate_enabled(adev) && adev->firmware.fw_buf) + adev->firmware.fw_buf_mc = amdgpu_bo_fb_aper_addr(adev->firmware.fw_buf); + for (i = 0; i < adev->firmware.max_ucodes; i++) { ucode = &adev->firmware.ucode[i]; if (ucode->fw) { @@ -1387,6 +1394,19 @@ static const char *amdgpu_ucode_legacy_naming(struct amdgpu_device *adev, int bl return NULL; } +bool amdgpu_is_kicker_fw(struct amdgpu_device *adev) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(kicker_device_list); i++) { + if (adev->pdev->device == kicker_device_list[i].device && + adev->pdev->revision == kicker_device_list[i].revision) + return true; + } + + return false; +} + void amdgpu_ucode_ip_version_decode(struct amdgpu_device *adev, int block_type, char *ucode_prefix, int len) { int maj, min, rev; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h index 9e89c3487be5..6349aad6da35 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h @@ -605,6 +605,11 @@ struct amdgpu_firmware { uint32_t pldm_version; }; +struct kicker_device{ + unsigned short device; + u8 revision; +}; + void amdgpu_ucode_print_mc_hdr(const struct common_firmware_header *hdr); void amdgpu_ucode_print_smc_hdr(const struct common_firmware_header *hdr); void amdgpu_ucode_print_imu_hdr(const struct common_firmware_header *hdr); @@ -632,5 +637,6 @@ amdgpu_ucode_get_load_type(struct amdgpu_device *adev, int load_type); const char *amdgpu_ucode_name(enum AMDGPU_UCODE_ID ucode_id); void amdgpu_ucode_ip_version_decode(struct amdgpu_device *adev, int block_type, char *ucode_prefix, int len); +bool amdgpu_is_kicker_fw(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 295e7186e156..b1b80efc7993 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -318,6 +318,10 @@ amdgpu_userq_destroy(struct drm_file *filp, int queue_id) amdgpu_bo_unreserve(queue->db_obj.obj); } amdgpu_bo_unref(&queue->db_obj.obj); + +#if defined(CONFIG_DEBUG_FS) + debugfs_remove_recursive(queue->debugfs_queue); +#endif r = amdgpu_userq_unmap_helper(uq_mgr, queue); amdgpu_userq_cleanup(uq_mgr, queue, queue_id); mutex_unlock(&uq_mgr->userq_mutex); @@ -343,6 +347,46 @@ static int amdgpu_userq_priority_permit(struct drm_file *filp, return -EACCES; } +#if defined(CONFIG_DEBUG_FS) +static int amdgpu_mqd_info_read(struct seq_file *m, void *unused) +{ + struct amdgpu_usermode_queue *queue = m->private; + struct amdgpu_bo *bo; + int r; + + if (!queue || !queue->mqd.obj) + return -EINVAL; + + bo = amdgpu_bo_ref(queue->mqd.obj); + r = amdgpu_bo_reserve(bo, true); + if (r) { + amdgpu_bo_unref(&bo); + return -EINVAL; + } + + seq_printf(m, "queue_type %d\n", queue->queue_type); + seq_printf(m, "mqd_gpu_address: 0x%llx\n", amdgpu_bo_gpu_offset(queue->mqd.obj)); + + amdgpu_bo_unreserve(bo); + amdgpu_bo_unref(&bo); + + return 0; +} + +static int amdgpu_mqd_info_open(struct inode *inode, struct file *file) +{ + return single_open(file, amdgpu_mqd_info_read, inode->i_private); +} + +static const struct file_operations amdgpu_mqd_info_fops = { + .owner = THIS_MODULE, + .open = amdgpu_mqd_info_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +#endif + static int amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) { @@ -352,6 +396,7 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) const struct amdgpu_userq_funcs *uq_funcs; struct amdgpu_usermode_queue *queue; struct amdgpu_db_info db_info; + char *queue_name; bool skip_map_queue; uint64_t index; int qid, r = 0; @@ -475,6 +520,16 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) } } + queue_name = kasprintf(GFP_KERNEL, "queue-%d", qid); + if (!queue_name) + return -ENOMEM; + +#if defined(CONFIG_DEBUG_FS) + /* Queue dentry per client to hold MQD information */ + queue->debugfs_queue = debugfs_create_dir(queue_name, filp->debugfs_client); + debugfs_create_file("mqd_info", 0444, queue->debugfs_queue, queue, &amdgpu_mqd_info_fops); +#endif + kfree(queue_name); args->out.queue_id = qid; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h index ec040c2fd6c9..b1ca91b7cda4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h @@ -65,6 +65,7 @@ struct amdgpu_usermode_queue { struct dma_fence *last_fence; u32 xcp_id; int priority; + struct dentry *debugfs_queue; }; struct amdgpu_userq_funcs { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index 577c6194db78..9320461bb486 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -301,6 +301,9 @@ struct amdgpu_virt { union amd_sriov_ras_caps ras_telemetry_en_caps; struct amdgpu_virt_ras ras; struct amd_sriov_ras_telemetry_error_count count_cache; + + /* hibernate and resume with different VF feature for xgmi enabled system */ + bool is_xgmi_node_migrate_enabled; }; struct amdgpu_video_codec_info; @@ -386,6 +389,10 @@ static inline bool is_virtual_machine(void) ((adev)->virt.gim_feature & AMDGIM_FEATURE_VCN_RB_DECOUPLE) #define amdgpu_sriov_is_mes_info_enable(adev) \ ((adev)->virt.gim_feature & AMDGIM_FEATURE_MES_INFO_ENABLE) + +#define amdgpu_virt_xgmi_migrate_enabled(adev) \ + ((adev)->virt.is_xgmi_node_migrate_enabled && (adev)->gmc.xgmi.node_segment_size != 0) + bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev); void amdgpu_virt_init_setting(struct amdgpu_device *adev); int amdgpu_virt_request_full_gpu(struct amdgpu_device *adev, bool init); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 0ff95a56c2ce..af0f655dfd5b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -1271,8 +1271,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, } else { struct drm_gem_object *obj = &bo->tbo.base; - if (obj->import_attach && bo_va->is_xgmi) { - struct dma_buf *dma_buf = obj->import_attach->dmabuf; + if (drm_gem_is_imported(obj) && bo_va->is_xgmi) { + struct dma_buf *dma_buf = obj->dma_buf; struct drm_gem_object *gobj = dma_buf->priv; struct amdgpu_bo *abo = gem_to_amdgpu_bo(gobj); @@ -1631,7 +1631,7 @@ int amdgpu_vm_handle_moved(struct amdgpu_device *adev, * validation */ if (vm->is_compute_context && - bo_va->base.bo->tbo.base.import_attach && + drm_gem_is_imported(&bo_va->base.bo->tbo.base) && (!bo_va->base.bo->tbo.resource || bo_va->base.bo->tbo.resource->mem_type == TTM_PL_SYSTEM)) amdgpu_vm_bo_evicted_user(&bo_va->base); @@ -2395,10 +2395,11 @@ void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint32_t min_vm_size, else adev->vm_manager.fragment_size = amdgpu_vm_fragment_size; - DRM_INFO("vm size is %u GB, %u levels, block size is %u-bit, fragment size is %u-bit\n", - vm_size, adev->vm_manager.num_level + 1, - adev->vm_manager.block_size, - adev->vm_manager.fragment_size); + dev_info( + adev->dev, + "vm size is %u GB, %u levels, block size is %u-bit, fragment size is %u-bit\n", + vm_size, adev->vm_manager.num_level + 1, + adev->vm_manager.block_size, adev->vm_manager.fragment_size); } /** @@ -2565,8 +2566,8 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode & AMDGPU_VM_USE_CPU_FOR_GFX); - DRM_DEBUG_DRIVER("VM update mode is %s\n", - vm->use_cpu_for_update ? "CPU" : "SDMA"); + dev_dbg(adev->dev, "VM update mode is %s\n", + vm->use_cpu_for_update ? "CPU" : "SDMA"); WARN_ONCE((vm->use_cpu_for_update && !amdgpu_gmc_vram_full_visible(&adev->gmc)), "CPU update of VM recommended only for large BAR system\n"); @@ -2608,7 +2609,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, r = amdgpu_vm_create_task_info(vm); if (r) - DRM_DEBUG("Failed to create task info for VM\n"); + dev_dbg(adev->dev, "Failed to create task info for VM\n"); amdgpu_bo_unreserve(vm->root.bo); amdgpu_bo_unref(&root_bo); @@ -2659,8 +2660,8 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm) /* Update VM state */ vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode & AMDGPU_VM_USE_CPU_FOR_COMPUTE); - DRM_DEBUG_DRIVER("VM update mode is %s\n", - vm->use_cpu_for_update ? "CPU" : "SDMA"); + dev_dbg(adev->dev, "VM update mode is %s\n", + vm->use_cpu_for_update ? "CPU" : "SDMA"); WARN_ONCE((vm->use_cpu_for_update && !amdgpu_gmc_vram_full_visible(&adev->gmc)), "CPU update of VM recommended only for large BAR system\n"); @@ -2983,7 +2984,7 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid, error_unlock: amdgpu_bo_unreserve(root); if (r < 0) - DRM_ERROR("Can't handle page fault (%d)\n", r); + dev_err(adev->dev, "Can't handle page fault (%d)\n", r); error_unref: amdgpu_bo_unref(&root); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c index 322816805bfb..c8fcafeb6864 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c @@ -445,6 +445,222 @@ void amdgpu_xcp_release_sched(struct amdgpu_device *adev, } } +int amdgpu_xcp_select_scheds(struct amdgpu_device *adev, + u32 hw_ip, u32 hw_prio, + struct amdgpu_fpriv *fpriv, + unsigned int *num_scheds, + struct drm_gpu_scheduler ***scheds) +{ + u32 sel_xcp_id; + int i; + struct amdgpu_xcp_mgr *xcp_mgr = adev->xcp_mgr; + + if (fpriv->xcp_id == AMDGPU_XCP_NO_PARTITION) { + u32 least_ref_cnt = ~0; + + fpriv->xcp_id = 0; + for (i = 0; i < xcp_mgr->num_xcps; i++) { + u32 total_ref_cnt; + + total_ref_cnt = atomic_read(&xcp_mgr->xcp[i].ref_cnt); + if (total_ref_cnt < least_ref_cnt) { + fpriv->xcp_id = i; + least_ref_cnt = total_ref_cnt; + } + } + } + sel_xcp_id = fpriv->xcp_id; + + if (xcp_mgr->xcp[sel_xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds) { + *num_scheds = + xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds; + *scheds = + xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].sched; + atomic_inc(&adev->xcp_mgr->xcp[sel_xcp_id].ref_cnt); + dev_dbg(adev->dev, "Selected partition #%d", sel_xcp_id); + } else { + dev_err(adev->dev, "Failed to schedule partition #%d.", sel_xcp_id); + return -ENOENT; + } + + return 0; +} + +static void amdgpu_set_xcp_id(struct amdgpu_device *adev, + uint32_t inst_idx, + struct amdgpu_ring *ring) +{ + int xcp_id; + enum AMDGPU_XCP_IP_BLOCK ip_blk; + uint32_t inst_mask; + + ring->xcp_id = AMDGPU_XCP_NO_PARTITION; + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) + adev->gfx.enforce_isolation[0].xcp_id = ring->xcp_id; + if ((adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) || + (ring->funcs->type == AMDGPU_RING_TYPE_CPER)) + return; + + inst_mask = 1 << inst_idx; + + switch (ring->funcs->type) { + case AMDGPU_HW_IP_GFX: + case AMDGPU_RING_TYPE_COMPUTE: + case AMDGPU_RING_TYPE_KIQ: + ip_blk = AMDGPU_XCP_GFX; + break; + case AMDGPU_RING_TYPE_SDMA: + ip_blk = AMDGPU_XCP_SDMA; + break; + case AMDGPU_RING_TYPE_VCN_ENC: + case AMDGPU_RING_TYPE_VCN_JPEG: + ip_blk = AMDGPU_XCP_VCN; + break; + default: + dev_err(adev->dev, "Not support ring type %d!", ring->funcs->type); + return; + } + + for (xcp_id = 0; xcp_id < adev->xcp_mgr->num_xcps; xcp_id++) { + if (adev->xcp_mgr->xcp[xcp_id].ip[ip_blk].inst_mask & inst_mask) { + ring->xcp_id = xcp_id; + dev_dbg(adev->dev, "ring:%s xcp_id :%u", ring->name, + ring->xcp_id); + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) + adev->gfx.enforce_isolation[xcp_id].xcp_id = xcp_id; + break; + } + } +} + +static void amdgpu_xcp_gpu_sched_update(struct amdgpu_device *adev, + struct amdgpu_ring *ring, + unsigned int sel_xcp_id) +{ + unsigned int *num_gpu_sched; + + num_gpu_sched = &adev->xcp_mgr->xcp[sel_xcp_id] + .gpu_sched[ring->funcs->type][ring->hw_prio].num_scheds; + adev->xcp_mgr->xcp[sel_xcp_id].gpu_sched[ring->funcs->type][ring->hw_prio] + .sched[(*num_gpu_sched)++] = &ring->sched; + dev_dbg(adev->dev, "%s :[%d] gpu_sched[%d][%d] = %d", + ring->name, sel_xcp_id, ring->funcs->type, + ring->hw_prio, *num_gpu_sched); +} + +static int amdgpu_xcp_sched_list_update(struct amdgpu_device *adev) +{ + struct amdgpu_ring *ring; + int i; + + for (i = 0; i < MAX_XCP; i++) { + atomic_set(&adev->xcp_mgr->xcp[i].ref_cnt, 0); + memset(adev->xcp_mgr->xcp[i].gpu_sched, 0, sizeof(adev->xcp_mgr->xcp->gpu_sched)); + } + + if (adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) + return 0; + + for (i = 0; i < AMDGPU_MAX_RINGS; i++) { + ring = adev->rings[i]; + if (!ring || !ring->sched.ready || ring->no_scheduler) + continue; + + amdgpu_xcp_gpu_sched_update(adev, ring, ring->xcp_id); + + /* VCN may be shared by two partitions under CPX MODE in certain + * configs. + */ + if ((ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC || + ring->funcs->type == AMDGPU_RING_TYPE_VCN_JPEG) && + (adev->xcp_mgr->num_xcps > adev->vcn.num_vcn_inst)) + amdgpu_xcp_gpu_sched_update(adev, ring, ring->xcp_id + 1); + } + + return 0; +} + +int amdgpu_xcp_update_partition_sched_list(struct amdgpu_device *adev) +{ + int i; + + for (i = 0; i < adev->num_rings; i++) { + struct amdgpu_ring *ring = adev->rings[i]; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE || + ring->funcs->type == AMDGPU_RING_TYPE_KIQ) + amdgpu_set_xcp_id(adev, ring->xcc_id, ring); + else + amdgpu_set_xcp_id(adev, ring->me, ring); + } + + return amdgpu_xcp_sched_list_update(adev); +} + +void amdgpu_xcp_update_supported_modes(struct amdgpu_xcp_mgr *xcp_mgr) +{ + struct amdgpu_device *adev = xcp_mgr->adev; + + xcp_mgr->supp_xcp_modes = 0; + + switch (NUM_XCC(adev->gfx.xcc_mask)) { + case 8: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_DPX_PARTITION_MODE) | + BIT(AMDGPU_QPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 6: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_TPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 4: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_DPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 2: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 1: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + + default: + break; + } +} + +int amdgpu_xcp_pre_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) +{ + /* TODO: + * Stop user queues and threads, and make sure GPU is empty of work. + */ + + if (flags & AMDGPU_XCP_OPS_KFD) + amdgpu_amdkfd_device_fini_sw(xcp_mgr->adev); + + return 0; +} + +int amdgpu_xcp_post_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) +{ + int ret = 0; + + if (flags & AMDGPU_XCP_OPS_KFD) { + amdgpu_amdkfd_device_probe(xcp_mgr->adev); + amdgpu_amdkfd_device_init(xcp_mgr->adev); + /* If KFD init failed, return failure */ + if (!xcp_mgr->adev->kfd.init_complete) + ret = -EIO; + } + + return ret; +} + /*====================== xcp sysfs - configuration ======================*/ #define XCP_CFG_SYSFS_RES_ATTR_SHOW(_name) \ static ssize_t amdgpu_xcp_res_sysfs_##_name##_show( \ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h index 454b33f889fb..70a0f8400b57 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h @@ -39,6 +39,8 @@ #define AMDGPU_XCP_NO_PARTITION (~0) +#define AMDGPU_XCP_OPS_KFD (1 << 0) + struct amdgpu_fpriv; enum AMDGPU_XCP_IP_BLOCK { @@ -144,10 +146,6 @@ struct amdgpu_xcp_mgr_funcs { int (*suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); int (*prepare_resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); int (*resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); - int (*select_scheds)(struct amdgpu_device *adev, - u32 hw_ip, u32 hw_prio, struct amdgpu_fpriv *fpriv, - unsigned int *num_scheds, struct drm_gpu_scheduler ***scheds); - int (*update_partition_sched_list)(struct amdgpu_device *adev); }; int amdgpu_xcp_prepare_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); @@ -176,19 +174,18 @@ int amdgpu_xcp_open_device(struct amdgpu_device *adev, struct drm_file *file_priv); void amdgpu_xcp_release_sched(struct amdgpu_device *adev, struct amdgpu_ctx_entity *entity); - +int amdgpu_xcp_select_scheds(struct amdgpu_device *adev, + u32 hw_ip, u32 hw_prio, + struct amdgpu_fpriv *fpriv, + unsigned int *num_scheds, + struct drm_gpu_scheduler ***scheds); +void amdgpu_xcp_update_supported_modes(struct amdgpu_xcp_mgr *xcp_mgr); +int amdgpu_xcp_update_partition_sched_list(struct amdgpu_device *adev); +int amdgpu_xcp_pre_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags); +int amdgpu_xcp_post_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags); void amdgpu_xcp_sysfs_init(struct amdgpu_device *adev); void amdgpu_xcp_sysfs_fini(struct amdgpu_device *adev); -#define amdgpu_xcp_select_scheds(adev, e, c, d, x, y) \ - ((adev)->xcp_mgr && (adev)->xcp_mgr->funcs && \ - (adev)->xcp_mgr->funcs->select_scheds ? \ - (adev)->xcp_mgr->funcs->select_scheds((adev), (e), (c), (d), (x), (y)) : -ENOENT) -#define amdgpu_xcp_update_partition_sched_list(adev) \ - ((adev)->xcp_mgr && (adev)->xcp_mgr->funcs && \ - (adev)->xcp_mgr->funcs->update_partition_sched_list ? \ - (adev)->xcp_mgr->funcs->update_partition_sched_list(adev) : 0) - static inline int amdgpu_xcp_get_num_xcp(struct amdgpu_xcp_mgr *xcp_mgr) { if (!xcp_mgr) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index d9ad37711c3e..1ede308a7c67 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -1771,16 +1771,25 @@ void amdgpu_xgmi_early_init(struct amdgpu_device *adev) case IP_VERSION(9, 4, 0): case IP_VERSION(9, 4, 1): case IP_VERSION(9, 4, 2): - adev->gmc.xgmi.max_speed = XGMI_SPEED_25GT; + /* 25 GT/s */ + adev->gmc.xgmi.max_speed = 25; adev->gmc.xgmi.max_width = 16; break; case IP_VERSION(9, 4, 3): case IP_VERSION(9, 4, 4): case IP_VERSION(9, 5, 0): - adev->gmc.xgmi.max_speed = XGMI_SPEED_32GT; + /* 32 GT/s */ + adev->gmc.xgmi.max_speed = 32; adev->gmc.xgmi.max_width = 16; break; default: break; } } + +void amgpu_xgmi_set_max_speed_width(struct amdgpu_device *adev, + uint16_t max_speed, uint8_t max_width) +{ + adev->gmc.xgmi.max_speed = max_speed; + adev->gmc.xgmi.max_width = max_width; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h index f994be985f42..bba0b26fee8f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h @@ -25,12 +25,6 @@ #include <drm/task_barrier.h> #include "amdgpu_ras.h" -enum amdgpu_xgmi_link_speed { - XGMI_SPEED_16GT = 16, - XGMI_SPEED_25GT = 25, - XGMI_SPEED_32GT = 32 -}; - struct amdgpu_hive_info { struct kobject kobj; uint64_t hive_id; @@ -97,7 +91,7 @@ struct amdgpu_xgmi { struct ras_common_if *ras_if; bool connected_to_cpu; struct amdgpu_xgmi_ras *ras; - enum amdgpu_xgmi_link_speed max_speed; + uint16_t max_speed; uint8_t max_width; }; @@ -130,4 +124,6 @@ int amdgpu_xgmi_get_ext_link(struct amdgpu_device *adev, int link_num); void amdgpu_xgmi_early_init(struct amdgpu_device *adev); uint32_t amdgpu_xgmi_get_max_bandwidth(struct amdgpu_device *adev); +void amgpu_xgmi_set_max_speed_width(struct amdgpu_device *adev, + uint16_t max_speed, uint8_t max_width); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c index 1c083304ae77..914cf4bfb033 100644 --- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c +++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c @@ -29,12 +29,11 @@ #include "gfx_v9_4_3.h" #include "gfxhub_v1_2.h" #include "sdma_v4_4_2.h" +#include "amdgpu_ip.h" #define XCP_INST_MASK(num_inst, xcp_id) \ (num_inst ? GENMASK(num_inst - 1, 0) << (xcp_id * num_inst) : 0) -#define AMDGPU_XCP_OPS_KFD (1 << 0) - void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev) { int i; @@ -62,234 +61,6 @@ void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev) adev->doorbell_index.max_assignment = AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT << 1; } -static bool aqua_vanjaram_xcp_vcn_shared(struct amdgpu_device *adev) -{ - return (adev->xcp_mgr->num_xcps > adev->vcn.num_vcn_inst); -} - -static void aqua_vanjaram_set_xcp_id(struct amdgpu_device *adev, - uint32_t inst_idx, struct amdgpu_ring *ring) -{ - int xcp_id; - enum AMDGPU_XCP_IP_BLOCK ip_blk; - uint32_t inst_mask; - - ring->xcp_id = AMDGPU_XCP_NO_PARTITION; - if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) - adev->gfx.enforce_isolation[0].xcp_id = ring->xcp_id; - if ((adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) || - (ring->funcs->type == AMDGPU_RING_TYPE_CPER)) - return; - - inst_mask = 1 << inst_idx; - - switch (ring->funcs->type) { - case AMDGPU_HW_IP_GFX: - case AMDGPU_RING_TYPE_COMPUTE: - case AMDGPU_RING_TYPE_KIQ: - ip_blk = AMDGPU_XCP_GFX; - break; - case AMDGPU_RING_TYPE_SDMA: - ip_blk = AMDGPU_XCP_SDMA; - break; - case AMDGPU_RING_TYPE_VCN_ENC: - case AMDGPU_RING_TYPE_VCN_JPEG: - ip_blk = AMDGPU_XCP_VCN; - break; - default: - DRM_ERROR("Not support ring type %d!", ring->funcs->type); - return; - } - - for (xcp_id = 0; xcp_id < adev->xcp_mgr->num_xcps; xcp_id++) { - if (adev->xcp_mgr->xcp[xcp_id].ip[ip_blk].inst_mask & inst_mask) { - ring->xcp_id = xcp_id; - dev_dbg(adev->dev, "ring:%s xcp_id :%u", ring->name, - ring->xcp_id); - if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) - adev->gfx.enforce_isolation[xcp_id].xcp_id = xcp_id; - break; - } - } -} - -static void aqua_vanjaram_xcp_gpu_sched_update( - struct amdgpu_device *adev, - struct amdgpu_ring *ring, - unsigned int sel_xcp_id) -{ - unsigned int *num_gpu_sched; - - num_gpu_sched = &adev->xcp_mgr->xcp[sel_xcp_id] - .gpu_sched[ring->funcs->type][ring->hw_prio].num_scheds; - adev->xcp_mgr->xcp[sel_xcp_id].gpu_sched[ring->funcs->type][ring->hw_prio] - .sched[(*num_gpu_sched)++] = &ring->sched; - DRM_DEBUG("%s :[%d] gpu_sched[%d][%d] = %d", ring->name, - sel_xcp_id, ring->funcs->type, - ring->hw_prio, *num_gpu_sched); -} - -static int aqua_vanjaram_xcp_sched_list_update( - struct amdgpu_device *adev) -{ - struct amdgpu_ring *ring; - int i; - - for (i = 0; i < MAX_XCP; i++) { - atomic_set(&adev->xcp_mgr->xcp[i].ref_cnt, 0); - memset(adev->xcp_mgr->xcp[i].gpu_sched, 0, sizeof(adev->xcp_mgr->xcp->gpu_sched)); - } - - if (adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) - return 0; - - for (i = 0; i < AMDGPU_MAX_RINGS; i++) { - ring = adev->rings[i]; - if (!ring || !ring->sched.ready || ring->no_scheduler) - continue; - - aqua_vanjaram_xcp_gpu_sched_update(adev, ring, ring->xcp_id); - - /* VCN may be shared by two partitions under CPX MODE in certain - * configs. - */ - if ((ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC || - ring->funcs->type == AMDGPU_RING_TYPE_VCN_JPEG) && - aqua_vanjaram_xcp_vcn_shared(adev)) - aqua_vanjaram_xcp_gpu_sched_update(adev, ring, ring->xcp_id + 1); - } - - return 0; -} - -static int aqua_vanjaram_update_partition_sched_list(struct amdgpu_device *adev) -{ - int i; - - for (i = 0; i < adev->num_rings; i++) { - struct amdgpu_ring *ring = adev->rings[i]; - - if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE || - ring->funcs->type == AMDGPU_RING_TYPE_KIQ) - aqua_vanjaram_set_xcp_id(adev, ring->xcc_id, ring); - else - aqua_vanjaram_set_xcp_id(adev, ring->me, ring); - } - - return aqua_vanjaram_xcp_sched_list_update(adev); -} - -static int aqua_vanjaram_select_scheds( - struct amdgpu_device *adev, - u32 hw_ip, - u32 hw_prio, - struct amdgpu_fpriv *fpriv, - unsigned int *num_scheds, - struct drm_gpu_scheduler ***scheds) -{ - u32 sel_xcp_id; - int i; - - if (fpriv->xcp_id == AMDGPU_XCP_NO_PARTITION) { - u32 least_ref_cnt = ~0; - - fpriv->xcp_id = 0; - for (i = 0; i < adev->xcp_mgr->num_xcps; i++) { - u32 total_ref_cnt; - - total_ref_cnt = atomic_read(&adev->xcp_mgr->xcp[i].ref_cnt); - if (total_ref_cnt < least_ref_cnt) { - fpriv->xcp_id = i; - least_ref_cnt = total_ref_cnt; - } - } - } - sel_xcp_id = fpriv->xcp_id; - - if (adev->xcp_mgr->xcp[sel_xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds) { - *num_scheds = adev->xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds; - *scheds = adev->xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].sched; - atomic_inc(&adev->xcp_mgr->xcp[sel_xcp_id].ref_cnt); - DRM_DEBUG("Selected partition #%d", sel_xcp_id); - } else { - DRM_ERROR("Failed to schedule partition #%d.", sel_xcp_id); - return -ENOENT; - } - - return 0; -} - -static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev, - enum amd_hw_ip_block_type block, - int8_t inst) -{ - int8_t dev_inst; - - switch (block) { - case GC_HWIP: - case SDMA0_HWIP: - /* Both JPEG and VCN as JPEG is only alias of VCN */ - case VCN_HWIP: - dev_inst = adev->ip_map.dev_inst[block][inst]; - break; - default: - /* For rest of the IPs, no look up required. - * Assume 'logical instance == physical instance' for all configs. */ - dev_inst = inst; - break; - } - - return dev_inst; -} - -static uint32_t aqua_vanjaram_logical_to_dev_mask(struct amdgpu_device *adev, - enum amd_hw_ip_block_type block, - uint32_t mask) -{ - uint32_t dev_mask = 0; - int8_t log_inst, dev_inst; - - while (mask) { - log_inst = ffs(mask) - 1; - dev_inst = aqua_vanjaram_logical_to_dev_inst(adev, block, log_inst); - dev_mask |= (1 << dev_inst); - mask &= ~(1 << log_inst); - } - - return dev_mask; -} - -static void aqua_vanjaram_populate_ip_map(struct amdgpu_device *adev, - enum amd_hw_ip_block_type ip_block, - uint32_t inst_mask) -{ - int l = 0, i; - - while (inst_mask) { - i = ffs(inst_mask) - 1; - adev->ip_map.dev_inst[ip_block][l++] = i; - inst_mask &= ~(1 << i); - } - for (; l < HWIP_MAX_INSTANCE; l++) - adev->ip_map.dev_inst[ip_block][l] = -1; -} - -void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev) -{ - u32 ip_map[][2] = { - { GC_HWIP, adev->gfx.xcc_mask }, - { SDMA0_HWIP, adev->sdma.sdma_mask }, - { VCN_HWIP, adev->vcn.inst_mask }, - }; - int i; - - for (i = 0; i < ARRAY_SIZE(ip_map); ++i) - aqua_vanjaram_populate_ip_map(adev, ip_map[i][0], ip_map[i][1]); - - adev->ip_map.logical_to_dev_inst = aqua_vanjaram_logical_to_dev_inst; - adev->ip_map.logical_to_dev_mask = aqua_vanjaram_logical_to_dev_mask; -} - /* Fixed pattern for smn addressing on different AIDs: * bit[34]: indicate cross AID access * bit[33:32]: indicate target AID id @@ -353,11 +124,14 @@ static int aqua_vanjaram_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr) if (adev->nbio.funcs->get_compute_partition_mode) { mode = adev->nbio.funcs->get_compute_partition_mode(adev); - if (mode != derv_mode) + if (mode != derv_mode) { dev_warn( adev->dev, "Mismatch in compute partition mode - reported : %d derived : %d", mode, derv_mode); + if (derv_mode == AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE) + amdgpu_device_bus_status_check(adev); + } } return mode; @@ -593,72 +367,6 @@ static bool __aqua_vanjaram_is_valid_mode(struct amdgpu_xcp_mgr *xcp_mgr, return false; } -static int __aqua_vanjaram_pre_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) -{ - /* TODO: - * Stop user queues and threads, and make sure GPU is empty of work. - */ - - if (flags & AMDGPU_XCP_OPS_KFD) - amdgpu_amdkfd_device_fini_sw(xcp_mgr->adev); - - return 0; -} - -static int __aqua_vanjaram_post_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) -{ - int ret = 0; - - if (flags & AMDGPU_XCP_OPS_KFD) { - amdgpu_amdkfd_device_probe(xcp_mgr->adev); - amdgpu_amdkfd_device_init(xcp_mgr->adev); - /* If KFD init failed, return failure */ - if (!xcp_mgr->adev->kfd.init_complete) - ret = -EIO; - } - - return ret; -} - -static void -__aqua_vanjaram_update_supported_modes(struct amdgpu_xcp_mgr *xcp_mgr) -{ - struct amdgpu_device *adev = xcp_mgr->adev; - - xcp_mgr->supp_xcp_modes = 0; - - switch (NUM_XCC(adev->gfx.xcc_mask)) { - case 8: - xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | - BIT(AMDGPU_DPX_PARTITION_MODE) | - BIT(AMDGPU_QPX_PARTITION_MODE) | - BIT(AMDGPU_CPX_PARTITION_MODE); - break; - case 6: - xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | - BIT(AMDGPU_TPX_PARTITION_MODE) | - BIT(AMDGPU_CPX_PARTITION_MODE); - break; - case 4: - xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | - BIT(AMDGPU_DPX_PARTITION_MODE) | - BIT(AMDGPU_CPX_PARTITION_MODE); - break; - /* this seems only existing in emulation phase */ - case 2: - xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | - BIT(AMDGPU_CPX_PARTITION_MODE); - break; - case 1: - xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | - BIT(AMDGPU_CPX_PARTITION_MODE); - break; - - default: - break; - } -} - static void __aqua_vanjaram_update_available_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr) { int mode; @@ -705,7 +413,7 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, goto out; } - ret = __aqua_vanjaram_pre_partition_switch(xcp_mgr, flags); + ret = amdgpu_xcp_pre_partition_switch(xcp_mgr, flags); if (ret) goto unlock; @@ -718,7 +426,7 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, *num_xcps = num_xcc / num_xcc_per_xcp; amdgpu_xcp_init(xcp_mgr, *num_xcps, mode); - ret = __aqua_vanjaram_post_partition_switch(xcp_mgr, flags); + ret = amdgpu_xcp_post_partition_switch(xcp_mgr, flags); if (!ret) __aqua_vanjaram_update_available_partition_mode(xcp_mgr); unlock: @@ -801,9 +509,6 @@ struct amdgpu_xcp_mgr_funcs aqua_vanjaram_xcp_funcs = { .get_ip_details = &aqua_vanjaram_get_xcp_ip_details, .get_xcp_res_info = &aqua_vanjaram_get_xcp_res_info, .get_xcp_mem_id = &aqua_vanjaram_get_xcp_mem_id, - .select_scheds = &aqua_vanjaram_select_scheds, - .update_partition_sched_list = - &aqua_vanjaram_update_partition_sched_list }; static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev) @@ -818,7 +523,7 @@ static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev) if (ret) return ret; - __aqua_vanjaram_update_supported_modes(adev->xcp_mgr); + amdgpu_xcp_update_supported_modes(adev->xcp_mgr); /* TODO: Default memory node affinity init */ return ret; @@ -858,7 +563,7 @@ int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev) if (ret) return ret; - aqua_vanjaram_ip_map_init(adev); + amdgpu_ip_map_init(adev); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 75ea071744eb..8c377ecbb8a7 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -9522,7 +9522,9 @@ static void gfx_v10_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) amdgpu_ring_insert_nop(ring, num_nop - 1); } -static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) +static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; @@ -9538,6 +9540,8 @@ static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + spin_lock_irqsave(&kiq->ring_lock, flags); if (amdgpu_ring_alloc(kiq_ring, 5 + 7 + 7 + kiq->pmf->map_queues_size)) { @@ -9575,11 +9579,17 @@ static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) return r; } - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static int gfx_v10_0_reset_kcq(struct amdgpu_ring *ring, - unsigned int vmid) + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; @@ -9593,6 +9603,8 @@ static int gfx_v10_0_reset_kcq(struct amdgpu_ring *ring, if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + spin_lock_irqsave(&kiq->ring_lock, flags); if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size)) { @@ -9647,7 +9659,12 @@ static int gfx_v10_0_reset_kcq(struct amdgpu_ring *ring, if (r) return r; - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static void gfx_v10_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index afd6d59164bf..37dcec2d0784 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -85,6 +85,7 @@ MODULE_FIRMWARE("amdgpu/gc_11_0_0_pfp.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_0_me.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_0_mec.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_0_rlc.bin"); +MODULE_FIRMWARE("amdgpu/gc_11_0_0_rlc_kicker.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_0_rlc_1.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_0_toc.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_1_pfp.bin"); @@ -759,6 +760,10 @@ static int gfx_v11_0_init_microcode(struct amdgpu_device *adev) err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, AMDGPU_UCODE_REQUIRED, "amdgpu/gc_11_0_0_rlc_1.bin"); + else if (amdgpu_is_kicker_fw(adev)) + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, + AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_rlc_kicker.bin", ucode_prefix); else err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, AMDGPU_UCODE_REQUIRED, @@ -6806,7 +6811,9 @@ static int gfx_v11_reset_gfx_pipe(struct amdgpu_ring *ring) return 0; } -static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) +static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int r; @@ -6814,6 +6821,8 @@ static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) if (amdgpu_sriov_vf(adev)) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, false); if (r) { @@ -6835,7 +6844,12 @@ static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) return r; } - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static int gfx_v11_0_reset_compute_pipe(struct amdgpu_ring *ring) @@ -6968,7 +6982,9 @@ static int gfx_v11_0_reset_compute_pipe(struct amdgpu_ring *ring) return 0; } -static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) +static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int r = 0; @@ -6976,6 +6992,8 @@ static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) if (amdgpu_sriov_vf(adev)) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, true); if (r) { dev_warn(adev->dev, "fail(%d) to reset kcq and try pipe reset\n", r); @@ -6995,7 +7013,12 @@ static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) return r; } - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static void gfx_v11_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c index 1234c8d64e20..e4fc42470cf3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c @@ -5307,7 +5307,9 @@ static int gfx_v12_reset_gfx_pipe(struct amdgpu_ring *ring) return 0; } -static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) +static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int r; @@ -5315,6 +5317,8 @@ static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) if (amdgpu_sriov_vf(adev)) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, false); if (r) { dev_warn(adev->dev, "reset via MES failed and try pipe reset %d\n", r); @@ -5335,7 +5339,12 @@ static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) return r; } - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static int gfx_v12_0_reset_compute_pipe(struct amdgpu_ring *ring) @@ -5421,7 +5430,9 @@ static int gfx_v12_0_reset_compute_pipe(struct amdgpu_ring *ring) return 0; } -static int gfx_v12_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) +static int gfx_v12_0_reset_kcq(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int r; @@ -5429,6 +5440,8 @@ static int gfx_v12_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) if (amdgpu_sriov_vf(adev)) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, true); if (r) { dev_warn(adev->dev, "fail(%d) to reset kcq and try pipe reset\n", r); @@ -5448,7 +5461,12 @@ static int gfx_v12_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) return r; } - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static void gfx_v12_0_ring_begin_use(struct amdgpu_ring *ring) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index da0534ff1271..2aa323dab34e 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -4884,76 +4884,6 @@ static void gfx_v7_0_emit_mem_sync_compute(struct amdgpu_ring *ring) amdgpu_ring_write(ring, 0x0000000A); /* poll interval */ } -static void gfx_v7_0_wait_reg_mem(struct amdgpu_ring *ring, int eng_sel, - int mem_space, int opt, uint32_t addr0, - uint32_t addr1, uint32_t ref, uint32_t mask, - uint32_t inv) -{ - amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); - amdgpu_ring_write(ring, - /* memory (1) or register (0) */ - (WAIT_REG_MEM_MEM_SPACE(mem_space) | - WAIT_REG_MEM_OPERATION(opt) | /* wait */ - WAIT_REG_MEM_FUNCTION(3) | /* equal */ - WAIT_REG_MEM_ENGINE(eng_sel))); - - if (mem_space) - BUG_ON(addr0 & 0x3); /* Dword align */ - amdgpu_ring_write(ring, addr0); - amdgpu_ring_write(ring, addr1); - amdgpu_ring_write(ring, ref); - amdgpu_ring_write(ring, mask); - amdgpu_ring_write(ring, inv); /* poll interval */ -} - -static void gfx_v7_0_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, - uint32_t val, uint32_t mask) -{ - gfx_v7_0_wait_reg_mem(ring, 0, 0, 0, reg, 0, val, mask, 0x20); -} - -static int gfx_v7_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) -{ - struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; - struct amdgpu_ring *kiq_ring = &kiq->ring; - unsigned long flags; - u32 tmp; - int r; - - if (amdgpu_sriov_vf(adev)) - return -EINVAL; - - if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) - return -EINVAL; - - spin_lock_irqsave(&kiq->ring_lock, flags); - - if (amdgpu_ring_alloc(kiq_ring, 5)) { - spin_unlock_irqrestore(&kiq->ring_lock, flags); - return -ENOMEM; - } - - tmp = REG_SET_FIELD(0, CP_VMID_RESET, RESET_REQUEST, 1 << vmid); - gfx_v7_0_ring_emit_wreg(kiq_ring, mmCP_VMID_RESET, tmp); - amdgpu_ring_commit(kiq_ring); - - spin_unlock_irqrestore(&kiq->ring_lock, flags); - - r = amdgpu_ring_test_ring(kiq_ring); - if (r) - return r; - - if (amdgpu_ring_alloc(ring, 7 + 12 + 5)) - return -ENOMEM; - gfx_v7_0_ring_emit_fence_gfx(ring, ring->fence_drv.gpu_addr, - ring->fence_drv.sync_seq, AMDGPU_FENCE_FLAG_EXEC); - gfx_v7_0_ring_emit_reg_wait(ring, mmCP_VMID_RESET, 0, 0xffff); - gfx_v7_0_ring_emit_wreg(ring, mmCP_VMID_RESET, 0); - - return amdgpu_ring_test_ring(ring); -} - static const struct amd_ip_funcs gfx_v7_0_ip_funcs = { .name = "gfx_v7_0", .early_init = gfx_v7_0_early_init, @@ -5003,7 +4933,6 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = { .emit_wreg = gfx_v7_0_ring_emit_wreg, .soft_recovery = gfx_v7_0_ring_soft_recovery, .emit_mem_sync = gfx_v7_0_emit_mem_sync, - .reset = gfx_v7_0_reset_kgq, }; static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 5ee2237d8ee8..68c401ecb3ec 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -6339,34 +6339,6 @@ static void gfx_v8_0_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, amdgpu_ring_write(ring, val); } -static void gfx_v8_0_wait_reg_mem(struct amdgpu_ring *ring, int eng_sel, - int mem_space, int opt, uint32_t addr0, - uint32_t addr1, uint32_t ref, uint32_t mask, - uint32_t inv) -{ - amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); - amdgpu_ring_write(ring, - /* memory (1) or register (0) */ - (WAIT_REG_MEM_MEM_SPACE(mem_space) | - WAIT_REG_MEM_OPERATION(opt) | /* wait */ - WAIT_REG_MEM_FUNCTION(3) | /* equal */ - WAIT_REG_MEM_ENGINE(eng_sel))); - - if (mem_space) - BUG_ON(addr0 & 0x3); /* Dword align */ - amdgpu_ring_write(ring, addr0); - amdgpu_ring_write(ring, addr1); - amdgpu_ring_write(ring, ref); - amdgpu_ring_write(ring, mask); - amdgpu_ring_write(ring, inv); /* poll interval */ -} - -static void gfx_v8_0_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, - uint32_t val, uint32_t mask) -{ - gfx_v8_0_wait_reg_mem(ring, 0, 0, 0, reg, 0, val, mask, 0x20); -} - static void gfx_v8_0_ring_soft_recovery(struct amdgpu_ring *ring, unsigned vmid) { struct amdgpu_device *adev = ring->adev; @@ -6843,48 +6815,6 @@ static void gfx_v8_0_emit_wave_limit(struct amdgpu_ring *ring, bool enable) } -static int gfx_v8_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) -{ - struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; - struct amdgpu_ring *kiq_ring = &kiq->ring; - unsigned long flags; - u32 tmp; - int r; - - if (amdgpu_sriov_vf(adev)) - return -EINVAL; - - if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) - return -EINVAL; - - spin_lock_irqsave(&kiq->ring_lock, flags); - - if (amdgpu_ring_alloc(kiq_ring, 5)) { - spin_unlock_irqrestore(&kiq->ring_lock, flags); - return -ENOMEM; - } - - tmp = REG_SET_FIELD(0, CP_VMID_RESET, RESET_REQUEST, 1 << vmid); - gfx_v8_0_ring_emit_wreg(kiq_ring, mmCP_VMID_RESET, tmp); - amdgpu_ring_commit(kiq_ring); - - spin_unlock_irqrestore(&kiq->ring_lock, flags); - - r = amdgpu_ring_test_ring(kiq_ring); - if (r) - return r; - - if (amdgpu_ring_alloc(ring, 7 + 12 + 5)) - return -ENOMEM; - gfx_v8_0_ring_emit_fence_gfx(ring, ring->fence_drv.gpu_addr, - ring->fence_drv.sync_seq, AMDGPU_FENCE_FLAG_EXEC); - gfx_v8_0_ring_emit_reg_wait(ring, mmCP_VMID_RESET, 0, 0xffff); - gfx_v8_0_ring_emit_wreg(ring, mmCP_VMID_RESET, 0); - - return amdgpu_ring_test_ring(ring); -} - static const struct amd_ip_funcs gfx_v8_0_ip_funcs = { .name = "gfx_v8_0", .early_init = gfx_v8_0_early_init, @@ -6950,7 +6880,6 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = { .emit_wreg = gfx_v8_0_ring_emit_wreg, .soft_recovery = gfx_v8_0_ring_soft_recovery, .emit_mem_sync = gfx_v8_0_emit_mem_sync, - .reset = gfx_v8_0_reset_kgq, }; static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index d377a7c57d5e..76ba664efecb 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -2235,6 +2235,25 @@ static int gfx_v9_0_sw_init(struct amdgpu_ip_block *ip_block) } switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { + case IP_VERSION(9, 0, 1): + case IP_VERSION(9, 2, 1): + case IP_VERSION(9, 4, 0): + case IP_VERSION(9, 2, 2): + case IP_VERSION(9, 1, 0): + case IP_VERSION(9, 3, 0): + adev->gfx.cleaner_shader_ptr = gfx_9_4_2_cleaner_shader_hex; + adev->gfx.cleaner_shader_size = sizeof(gfx_9_4_2_cleaner_shader_hex); + if (adev->gfx.me_fw_version >= 167 && + adev->gfx.pfp_fw_version >= 196 && + adev->gfx.mec_fw_version >= 474) { + adev->gfx.enable_cleaner_shader = true; + r = amdgpu_gfx_cleaner_shader_sw_init(adev, adev->gfx.cleaner_shader_size); + if (r) { + adev->gfx.enable_cleaner_shader = false; + dev_err(adev->dev, "Failed to initialize cleaner shader\n"); + } + } + break; case IP_VERSION(9, 4, 2): adev->gfx.cleaner_shader_ptr = gfx_9_4_2_cleaner_shader_hex; adev->gfx.cleaner_shader_size = sizeof(gfx_9_4_2_cleaner_shader_hex); @@ -7152,53 +7171,9 @@ static void gfx_v9_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) amdgpu_ring_insert_nop(ring, num_nop - 1); } -static int gfx_v9_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) -{ - struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; - struct amdgpu_ring *kiq_ring = &kiq->ring; - unsigned long flags; - u32 tmp; - int r; - - if (amdgpu_sriov_vf(adev)) - return -EINVAL; - - if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) - return -EINVAL; - - spin_lock_irqsave(&kiq->ring_lock, flags); - - if (amdgpu_ring_alloc(kiq_ring, 5)) { - spin_unlock_irqrestore(&kiq->ring_lock, flags); - return -ENOMEM; - } - - tmp = REG_SET_FIELD(0, CP_VMID_RESET, RESET_REQUEST, 1 << vmid); - gfx_v9_0_ring_emit_wreg(kiq_ring, - SOC15_REG_OFFSET(GC, 0, mmCP_VMID_RESET), tmp); - amdgpu_ring_commit(kiq_ring); - - spin_unlock_irqrestore(&kiq->ring_lock, flags); - - r = amdgpu_ring_test_ring(kiq_ring); - if (r) - return r; - - if (amdgpu_ring_alloc(ring, 7 + 7 + 5)) - return -ENOMEM; - gfx_v9_0_ring_emit_fence(ring, ring->fence_drv.gpu_addr, - ring->fence_drv.sync_seq, AMDGPU_FENCE_FLAG_EXEC); - gfx_v9_0_ring_emit_reg_wait(ring, - SOC15_REG_OFFSET(GC, 0, mmCP_VMID_RESET), 0, 0xffff); - gfx_v9_0_ring_emit_wreg(ring, - SOC15_REG_OFFSET(GC, 0, mmCP_VMID_RESET), 0); - - return amdgpu_ring_test_ring(ring); -} - static int gfx_v9_0_reset_kcq(struct amdgpu_ring *ring, - unsigned int vmid) + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; @@ -7212,6 +7187,8 @@ static int gfx_v9_0_reset_kcq(struct amdgpu_ring *ring, if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + spin_lock_irqsave(&kiq->ring_lock, flags); if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size)) { @@ -7267,7 +7244,13 @@ static int gfx_v9_0_reset_kcq(struct amdgpu_ring *ring, DRM_ERROR("fail to remap queue\n"); return r; } - return amdgpu_ring_test_ring(ring); + + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static void gfx_v9_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) @@ -7477,7 +7460,6 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_gfx = { .emit_reg_write_reg_wait = gfx_v9_0_ring_emit_reg_write_reg_wait, .soft_recovery = gfx_v9_0_ring_soft_recovery, .emit_mem_sync = gfx_v9_0_emit_mem_sync, - .reset = gfx_v9_0_reset_kgq, .emit_cleaner_shader = gfx_v9_0_ring_emit_cleaner_shader, .begin_use = amdgpu_gfx_enforce_isolation_ring_begin_use, .end_use = amdgpu_gfx_enforce_isolation_ring_end_use, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c index c233edf60569..daed0f187bda 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -3552,7 +3552,8 @@ static int gfx_v9_4_3_reset_hw_pipe(struct amdgpu_ring *ring) } static int gfx_v9_4_3_reset_kcq(struct amdgpu_ring *ring, - unsigned int vmid) + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_kiq *kiq = &adev->gfx.kiq[ring->xcc_id]; @@ -3566,6 +3567,8 @@ static int gfx_v9_4_3_reset_kcq(struct amdgpu_ring *ring, if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + spin_lock_irqsave(&kiq->ring_lock, flags); if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size)) { @@ -3619,7 +3622,13 @@ pipe_reset: dev_err(adev->dev, "fail to remap queue\n"); return r; } - return amdgpu_ring_test_ring(ring); + + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } enum amdgpu_gfx_cp_ras_mem_id { diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c index cb25f7f0dfc1..6c03bf9f1ae8 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c @@ -74,6 +74,8 @@ static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev, static void gfxhub_v1_2_xcc_init_gart_aperture_regs(struct amdgpu_device *adev, uint32_t xcc_mask) { + uint64_t gart_start = amdgpu_virt_xgmi_migrate_enabled(adev) ? + adev->gmc.vram_start : adev->gmc.fb_start; uint64_t pt_base; int i; @@ -91,10 +93,10 @@ static void gfxhub_v1_2_xcc_init_gart_aperture_regs(struct amdgpu_device *adev, if (adev->gmc.pdb0_bo) { WREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, - (u32)(adev->gmc.fb_start >> 12)); + (u32)(gart_start >> 12)); WREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, - (u32)(adev->gmc.fb_start >> 44)); + (u32)(gart_start >> 44)); WREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, @@ -180,7 +182,7 @@ gfxhub_v1_2_xcc_init_system_aperture_regs(struct amdgpu_device *adev, /* In the case squeezing vram into GART aperture, we don't use * FB aperture and AGP aperture. Disable them. */ - if (adev->gmc.pdb0_bo) { + if (adev->gmc.pdb0_bo && adev->gmc.xgmi.connected_to_cpu) { WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_TOP, 0); WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF); WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_AGP_TOP, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 78f65aea03f8..f73da518a6e8 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -78,8 +78,6 @@ #define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_DCN2 0x05ea #define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_DCN2_BASE_IDX 2 -#define MAX_MEM_RANGES 8 - static const char * const gfxhub_client_ids[] = { "CB", "DB", @@ -411,11 +409,6 @@ static const uint32_t ecc_umc_mcumc_ctrl_mask_addrs[] = { (0x001d43e0 + 0x00001800), }; -static inline bool gmc_v9_0_is_multi_chiplet(struct amdgpu_device *adev) -{ - return !!adev->aid_mask; -} - static int gmc_v9_0_ecc_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, unsigned int type, @@ -649,7 +642,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, addr, entry->client_id, soc15_ih_clientid_name[entry->client_id]); - if (gmc_v9_0_is_multi_chiplet(adev)) + if (amdgpu_is_multi_aid(adev)) dev_err(adev->dev, " cookie node_id %d fault from die %s%d%s\n", node_id, node_id % 4 == 3 ? "RSV" : "AID", node_id / 4, node_id % 4 == 1 ? ".XCD0" : node_id % 4 == 2 ? ".XCD1" : ""); @@ -798,7 +791,7 @@ static bool gmc_v9_0_use_invalidate_semaphore(struct amdgpu_device *adev, uint32_t vmhub) { if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 2) || - gmc_v9_0_is_multi_chiplet(adev)) + amdgpu_is_multi_aid(adev)) return false; return ((vmhub == AMDGPU_MMHUB0(0) || @@ -1382,46 +1375,6 @@ static unsigned int gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev) return size; } -static enum amdgpu_memory_partition -gmc_v9_0_get_memory_partition(struct amdgpu_device *adev, u32 *supp_modes) -{ - enum amdgpu_memory_partition mode = UNKNOWN_MEMORY_PARTITION_MODE; - - if (adev->nbio.funcs->get_memory_partition_mode) - mode = adev->nbio.funcs->get_memory_partition_mode(adev, - supp_modes); - - return mode; -} - -static enum amdgpu_memory_partition -gmc_v9_0_query_vf_memory_partition(struct amdgpu_device *adev) -{ - switch (adev->gmc.num_mem_partitions) { - case 0: - return UNKNOWN_MEMORY_PARTITION_MODE; - case 1: - return AMDGPU_NPS1_PARTITION_MODE; - case 2: - return AMDGPU_NPS2_PARTITION_MODE; - case 4: - return AMDGPU_NPS4_PARTITION_MODE; - default: - return AMDGPU_NPS1_PARTITION_MODE; - } - - return AMDGPU_NPS1_PARTITION_MODE; -} - -static enum amdgpu_memory_partition -gmc_v9_0_query_memory_partition(struct amdgpu_device *adev) -{ - if (amdgpu_sriov_vf(adev)) - return gmc_v9_0_query_vf_memory_partition(adev); - - return gmc_v9_0_get_memory_partition(adev, NULL); -} - static bool gmc_v9_0_need_reset_on_init(struct amdgpu_device *adev) { if (adev->nbio.funcs && adev->nbio.funcs->is_nps_switch_requested && @@ -1443,7 +1396,7 @@ static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = { .get_vm_pte = gmc_v9_0_get_vm_pte, .override_vm_pte_flags = gmc_v9_0_override_vm_pte_flags, .get_vbios_fb_size = gmc_v9_0_get_vbios_fb_size, - .query_mem_partition_mode = &gmc_v9_0_query_memory_partition, + .query_mem_partition_mode = &amdgpu_gmc_query_memory_partition, .request_mem_partition_mode = &amdgpu_gmc_request_memory_partition, .need_reset_on_init = &gmc_v9_0_need_reset_on_init, }; @@ -1550,7 +1503,7 @@ static void gmc_v9_0_set_mmhub_ras_funcs(struct amdgpu_device *adev) static void gmc_v9_0_set_gfxhub_funcs(struct amdgpu_device *adev) { - if (gmc_v9_0_is_multi_chiplet(adev)) + if (amdgpu_is_multi_aid(adev)) adev->gfxhub.funcs = &gfxhub_v1_2_funcs; else adev->gfxhub.funcs = &gfxhub_v1_0_funcs; @@ -1596,7 +1549,7 @@ static void gmc_v9_0_init_nps_details(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev) || (adev->flags & AMD_IS_APU)) return; - mode = gmc_v9_0_get_memory_partition(adev, &supp_modes); + mode = amdgpu_gmc_get_memory_partition(adev, &supp_modes); /* Mode detected by hardware and supported modes available */ if ((mode != UNKNOWN_MEMORY_PARTITION_MODE) && supp_modes) { @@ -1632,7 +1585,7 @@ static int gmc_v9_0_early_init(struct amdgpu_ip_block *ip_block) */ if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 0) || amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 1) || - gmc_v9_0_is_multi_chiplet(adev)) + amdgpu_is_multi_aid(adev)) adev->gmc.xgmi.supported = true; if (amdgpu_ip_version(adev, XGMI_HWIP, 0) == IP_VERSION(6, 1, 0)) { @@ -1719,7 +1672,7 @@ static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev, /* add the xgmi offset of the physical node */ base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; - if (adev->gmc.xgmi.connected_to_cpu) { + if (amdgpu_gmc_is_pdb0_enabled(adev)) { amdgpu_gmc_sysvm_location(adev, mc); } else { amdgpu_gmc_vram_location(adev, mc, base); @@ -1834,7 +1787,7 @@ static int gmc_v9_0_gart_init(struct amdgpu_device *adev) return 0; } - if (adev->gmc.xgmi.connected_to_cpu) { + if (amdgpu_gmc_is_pdb0_enabled(adev)) { adev->gmc.vmid0_page_table_depth = 1; adev->gmc.vmid0_page_table_block_size = 12; } else { @@ -1860,7 +1813,7 @@ static int gmc_v9_0_gart_init(struct amdgpu_device *adev) if (r) return r; - if (adev->gmc.xgmi.connected_to_cpu) + if (amdgpu_gmc_is_pdb0_enabled(adev)) r = amdgpu_gmc_pdb0_alloc(adev); } @@ -1882,188 +1835,6 @@ static void gmc_v9_0_save_registers(struct amdgpu_device *adev) adev->gmc.sdpif_register = RREG32_SOC15(DCE, 0, mmDCHUBBUB_SDPIF_MMIO_CNTRL_0); } -static bool gmc_v9_0_validate_partition_info(struct amdgpu_device *adev) -{ - enum amdgpu_memory_partition mode; - u32 supp_modes; - bool valid; - - mode = gmc_v9_0_get_memory_partition(adev, &supp_modes); - - /* Mode detected by hardware not present in supported modes */ - if ((mode != UNKNOWN_MEMORY_PARTITION_MODE) && - !(BIT(mode - 1) & supp_modes)) - return false; - - switch (mode) { - case UNKNOWN_MEMORY_PARTITION_MODE: - case AMDGPU_NPS1_PARTITION_MODE: - valid = (adev->gmc.num_mem_partitions == 1); - break; - case AMDGPU_NPS2_PARTITION_MODE: - valid = (adev->gmc.num_mem_partitions == 2); - break; - case AMDGPU_NPS4_PARTITION_MODE: - valid = (adev->gmc.num_mem_partitions == 3 || - adev->gmc.num_mem_partitions == 4); - break; - default: - valid = false; - } - - return valid; -} - -static bool gmc_v9_0_is_node_present(int *node_ids, int num_ids, int nid) -{ - int i; - - /* Check if node with id 'nid' is present in 'node_ids' array */ - for (i = 0; i < num_ids; ++i) - if (node_ids[i] == nid) - return true; - - return false; -} - -static void -gmc_v9_0_init_acpi_mem_ranges(struct amdgpu_device *adev, - struct amdgpu_mem_partition_info *mem_ranges) -{ - struct amdgpu_numa_info numa_info; - int node_ids[MAX_MEM_RANGES]; - int num_ranges = 0, ret; - int num_xcc, xcc_id; - uint32_t xcc_mask; - - num_xcc = NUM_XCC(adev->gfx.xcc_mask); - xcc_mask = (1U << num_xcc) - 1; - - for_each_inst(xcc_id, xcc_mask) { - ret = amdgpu_acpi_get_mem_info(adev, xcc_id, &numa_info); - if (ret) - continue; - - if (numa_info.nid == NUMA_NO_NODE) { - mem_ranges[0].size = numa_info.size; - mem_ranges[0].numa.node = numa_info.nid; - num_ranges = 1; - break; - } - - if (gmc_v9_0_is_node_present(node_ids, num_ranges, - numa_info.nid)) - continue; - - node_ids[num_ranges] = numa_info.nid; - mem_ranges[num_ranges].numa.node = numa_info.nid; - mem_ranges[num_ranges].size = numa_info.size; - ++num_ranges; - } - - adev->gmc.num_mem_partitions = num_ranges; -} - -static void -gmc_v9_0_init_sw_mem_ranges(struct amdgpu_device *adev, - struct amdgpu_mem_partition_info *mem_ranges) -{ - enum amdgpu_memory_partition mode; - u32 start_addr = 0, size; - int i, r, l; - - mode = gmc_v9_0_query_memory_partition(adev); - - switch (mode) { - case UNKNOWN_MEMORY_PARTITION_MODE: - adev->gmc.num_mem_partitions = 0; - break; - case AMDGPU_NPS1_PARTITION_MODE: - adev->gmc.num_mem_partitions = 1; - break; - case AMDGPU_NPS2_PARTITION_MODE: - adev->gmc.num_mem_partitions = 2; - break; - case AMDGPU_NPS4_PARTITION_MODE: - if (adev->flags & AMD_IS_APU) - adev->gmc.num_mem_partitions = 3; - else - adev->gmc.num_mem_partitions = 4; - break; - default: - adev->gmc.num_mem_partitions = 1; - break; - } - - /* Use NPS range info, if populated */ - r = amdgpu_gmc_get_nps_memranges(adev, mem_ranges, - &adev->gmc.num_mem_partitions); - if (!r) { - l = 0; - for (i = 1; i < adev->gmc.num_mem_partitions; ++i) { - if (mem_ranges[i].range.lpfn > - mem_ranges[i - 1].range.lpfn) - l = i; - } - - } else { - if (!adev->gmc.num_mem_partitions) { - dev_err(adev->dev, - "Not able to detect NPS mode, fall back to NPS1"); - adev->gmc.num_mem_partitions = 1; - } - /* Fallback to sw based calculation */ - size = (adev->gmc.real_vram_size + SZ_16M) >> AMDGPU_GPU_PAGE_SHIFT; - size /= adev->gmc.num_mem_partitions; - - for (i = 0; i < adev->gmc.num_mem_partitions; ++i) { - mem_ranges[i].range.fpfn = start_addr; - mem_ranges[i].size = - ((u64)size << AMDGPU_GPU_PAGE_SHIFT); - mem_ranges[i].range.lpfn = start_addr + size - 1; - start_addr += size; - } - - l = adev->gmc.num_mem_partitions - 1; - } - - /* Adjust the last one */ - mem_ranges[l].range.lpfn = - (adev->gmc.real_vram_size >> AMDGPU_GPU_PAGE_SHIFT) - 1; - mem_ranges[l].size = - adev->gmc.real_vram_size - - ((u64)mem_ranges[l].range.fpfn << AMDGPU_GPU_PAGE_SHIFT); -} - -static int gmc_v9_0_init_mem_ranges(struct amdgpu_device *adev) -{ - bool valid; - - adev->gmc.mem_partitions = kcalloc(MAX_MEM_RANGES, - sizeof(struct amdgpu_mem_partition_info), - GFP_KERNEL); - if (!adev->gmc.mem_partitions) - return -ENOMEM; - - /* TODO : Get the range from PSP/Discovery for dGPU */ - if (adev->gmc.is_app_apu) - gmc_v9_0_init_acpi_mem_ranges(adev, adev->gmc.mem_partitions); - else - gmc_v9_0_init_sw_mem_ranges(adev, adev->gmc.mem_partitions); - - if (amdgpu_sriov_vf(adev)) - valid = true; - else - valid = gmc_v9_0_validate_partition_info(adev); - if (!valid) { - /* TODO: handle invalid case */ - dev_WARN(adev->dev, - "Mem ranges not matching with hardware config"); - } - - return 0; -} - static void gmc_v9_4_3_init_vram_info(struct amdgpu_device *adev) { adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM; @@ -2085,7 +1856,7 @@ static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block) spin_lock_init(&adev->gmc.invalidate_lock); - if (gmc_v9_0_is_multi_chiplet(adev)) { + if (amdgpu_is_multi_aid(adev)) { gmc_v9_4_3_init_vram_info(adev); } else if (!adev->bios) { if (adev->flags & AMD_IS_APU) { @@ -2235,8 +2006,8 @@ static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block) amdgpu_gmc_get_vbios_allocations(adev); - if (gmc_v9_0_is_multi_chiplet(adev)) { - r = gmc_v9_0_init_mem_ranges(adev); + if (amdgpu_is_multi_aid(adev)) { + r = amdgpu_gmc_init_mem_ranges(adev); if (r) return r; } @@ -2264,7 +2035,7 @@ static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block) adev->vm_manager.first_kfd_vmid = (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 1) || amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 2) || - gmc_v9_0_is_multi_chiplet(adev)) ? + amdgpu_is_multi_aid(adev)) ? 3 : 8; @@ -2276,7 +2047,7 @@ static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block) if (r) return r; - if (gmc_v9_0_is_multi_chiplet(adev)) + if (amdgpu_is_multi_aid(adev)) amdgpu_gmc_sysfs_init(adev); return 0; @@ -2286,7 +2057,7 @@ static int gmc_v9_0_sw_fini(struct amdgpu_ip_block *ip_block) { struct amdgpu_device *adev = ip_block->adev; - if (gmc_v9_0_is_multi_chiplet(adev)) + if (amdgpu_is_multi_aid(adev)) amdgpu_gmc_sysfs_fini(adev); amdgpu_gmc_ras_fini(adev); @@ -2360,7 +2131,7 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev) { int r; - if (adev->gmc.xgmi.connected_to_cpu) + if (amdgpu_gmc_is_pdb0_enabled(adev)) amdgpu_gmc_init_pdb0(adev); if (adev->gart.bo == NULL) { @@ -2518,7 +2289,7 @@ static int gmc_v9_0_resume(struct amdgpu_ip_block *ip_block) * information again. */ if (adev->gmc.reset_flags & AMDGPU_GMC_INIT_RESET_NPS) { - gmc_v9_0_init_sw_mem_ranges(adev, adev->gmc.mem_partitions); + amdgpu_gmc_init_sw_mem_ranges(adev, adev->gmc.mem_partitions); adev->gmc.reset_flags &= ~AMDGPU_GMC_INIT_RESET_NPS; } diff --git a/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c b/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c index cfa91d709d49..cc626036ed9c 100644 --- a/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c @@ -32,6 +32,7 @@ #include "gc/gc_11_0_0_sh_mask.h" MODULE_FIRMWARE("amdgpu/gc_11_0_0_imu.bin"); +MODULE_FIRMWARE("amdgpu/gc_11_0_0_imu_kicker.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_1_imu.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_2_imu.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_3_imu.bin"); @@ -51,8 +52,12 @@ static int imu_v11_0_init_microcode(struct amdgpu_device *adev) DRM_DEBUG("\n"); amdgpu_ucode_ip_version_decode(adev, GC_HWIP, ucode_prefix, sizeof(ucode_prefix)); - err = amdgpu_ucode_request(adev, &adev->gfx.imu_fw, AMDGPU_UCODE_REQUIRED, - "amdgpu/%s_imu.bin", ucode_prefix); + if (amdgpu_is_kicker_fw(adev)) + err = amdgpu_ucode_request(adev, &adev->gfx.imu_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_imu_kicker.bin", ucode_prefix); + else + err = amdgpu_ucode_request(adev, &adev->gfx.imu_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_imu.bin", ucode_prefix); if (err) goto out; diff --git a/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c b/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c index 574880d67009..f857796f0297 100644 --- a/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c +++ b/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c @@ -29,6 +29,12 @@ #include "amdgpu.h" #include "isp_v4_1_1.h" +#define ISP_PERFORMANCE_STATE_LOW 0 +#define ISP_PERFORMANCE_STATE_HIGH 1 + +#define ISP_HIGH_PERFORMANC_XCLK 788 +#define ISP_HIGH_PERFORMANC_ICLK 788 + static const unsigned int isp_4_1_1_int_srcid[MAX_ISP411_INT_SRC] = { ISP_4_1__SRCID__ISP_RINGBUFFER_WPT9, ISP_4_1__SRCID__ISP_RINGBUFFER_WPT10, @@ -56,6 +62,125 @@ static struct gpiod_lookup_table isp_sensor_gpio_table = { }, }; +static int isp_poweroff(struct generic_pm_domain *genpd) +{ + struct amdgpu_isp *isp = container_of(genpd, struct amdgpu_isp, ispgpd); + struct amdgpu_device *adev = isp->adev; + + return amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ISP, true, 0); +} + +static int isp_poweron(struct generic_pm_domain *genpd) +{ + struct amdgpu_isp *isp = container_of(genpd, struct amdgpu_isp, ispgpd); + struct amdgpu_device *adev = isp->adev; + + return amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ISP, false, 0); +} + +static int isp_set_performance_state(struct generic_pm_domain *genpd, + unsigned int state) +{ + struct amdgpu_isp *isp = container_of(genpd, struct amdgpu_isp, ispgpd); + struct amdgpu_device *adev = isp->adev; + u32 iclk, xclk; + int ret; + + switch (state) { + case ISP_PERFORMANCE_STATE_HIGH: + xclk = ISP_HIGH_PERFORMANC_XCLK; + iclk = ISP_HIGH_PERFORMANC_ICLK; + break; + case ISP_PERFORMANCE_STATE_LOW: + /* isp runs at default lowest clock-rate on power-on, do nothing */ + return 0; + default: + return -EINVAL; + } + + ret = amdgpu_dpm_set_soft_freq_range(adev, PP_ISPXCLK, xclk, 0); + if (ret) { + drm_err(&adev->ddev, "failed to set xclk %u to %u: %d\n", + xclk, state, ret); + return ret; + } + + ret = amdgpu_dpm_set_soft_freq_range(adev, PP_ISPICLK, iclk, 0); + if (ret) { + drm_err(&adev->ddev, "failed to set iclk %u to %u: %d\n", + iclk, state, ret); + return ret; + } + + return 0; +} + +static int isp_genpd_add_device(struct device *dev, void *data) +{ + struct generic_pm_domain *gpd = data; + struct platform_device *pdev = container_of(dev, struct platform_device, dev); + struct amdgpu_isp *isp = container_of(gpd, struct amdgpu_isp, ispgpd); + struct amdgpu_device *adev = isp->adev; + int ret; + + if (!pdev) + return -EINVAL; + + if (!dev->type->name) { + drm_dbg(&adev->ddev, "Invalid device type to add\n"); + goto exit; + } + + if (strcmp(dev->type->name, "mfd_device")) { + drm_dbg(&adev->ddev, "Invalid isp mfd device %s to add\n", pdev->mfd_cell->name); + goto exit; + } + + ret = pm_genpd_add_device(gpd, dev); + if (ret) { + drm_err(&adev->ddev, "Failed to add dev %s to genpd %d\n", + pdev->mfd_cell->name, ret); + return -ENODEV; + } + +exit: + /* Continue to add */ + return 0; +} + +static int isp_genpd_remove_device(struct device *dev, void *data) +{ + struct generic_pm_domain *gpd = data; + struct platform_device *pdev = container_of(dev, struct platform_device, dev); + struct amdgpu_isp *isp = container_of(gpd, struct amdgpu_isp, ispgpd); + struct amdgpu_device *adev = isp->adev; + int ret; + + if (!pdev) + return -EINVAL; + + if (!dev->type->name) { + drm_dbg(&adev->ddev, "Invalid device type to remove\n"); + goto exit; + } + + if (strcmp(dev->type->name, "mfd_device")) { + drm_dbg(&adev->ddev, "Invalid isp mfd device %s to remove\n", + pdev->mfd_cell->name); + goto exit; + } + + ret = pm_genpd_remove_device(dev); + if (ret) { + drm_err(&adev->ddev, "Failed to remove dev from genpd %d\n", ret); + return -ENODEV; + } + +exit: + /* Continue to remove */ + return 0; +} + static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) { struct amdgpu_device *adev = isp->adev; @@ -81,11 +206,21 @@ static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) isp_base = adev->rmmio_base; + isp->ispgpd.name = "ISP_v_4_1_1"; + isp->ispgpd.power_off = isp_poweroff; + isp->ispgpd.power_on = isp_poweron; + isp->ispgpd.set_performance_state = isp_set_performance_state; + + r = pm_genpd_init(&isp->ispgpd, NULL, true); + if (r) { + drm_err(&adev->ddev, "failed to initialize genpd (%d)\n", r); + return -EINVAL; + } + isp->isp_cell = kcalloc(3, sizeof(struct mfd_cell), GFP_KERNEL); if (!isp->isp_cell) { r = -ENOMEM; - drm_err(&adev->ddev, - "%s: isp mfd cell alloc failed\n", __func__); + drm_err(&adev->ddev, "isp mfd cell alloc failed (%d)\n", r); goto failure; } @@ -95,16 +230,14 @@ static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) GFP_KERNEL); if (!isp->isp_res) { r = -ENOMEM; - drm_err(&adev->ddev, - "%s: isp mfd res alloc failed\n", __func__); + drm_err(&adev->ddev, "isp mfd resource alloc failed (%d)\n", r); goto failure; } isp->isp_pdata = kzalloc(sizeof(*isp->isp_pdata), GFP_KERNEL); if (!isp->isp_pdata) { r = -ENOMEM; - drm_err(&adev->ddev, - "%s: isp platform data alloc failed\n", __func__); + drm_err(&adev->ddev, "isp platform data alloc failed (%d)\n", r); goto failure; } @@ -142,8 +275,7 @@ static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) isp->isp_i2c_res = kcalloc(1, sizeof(struct resource), GFP_KERNEL); if (!isp->isp_i2c_res) { r = -ENOMEM; - drm_err(&adev->ddev, - "%s: isp mfd res alloc failed\n", __func__); + drm_err(&adev->ddev, "isp mfd res alloc failed (%d)\n", r); goto failure; } @@ -162,8 +294,7 @@ static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) isp->isp_gpio_res = kcalloc(1, sizeof(struct resource), GFP_KERNEL); if (!isp->isp_gpio_res) { r = -ENOMEM; - drm_err(&adev->ddev, - "%s: isp gpio res alloc failed\n", __func__); + drm_err(&adev->ddev, "isp gpio resource alloc failed (%d)\n", r); goto failure; } @@ -179,10 +310,23 @@ static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) isp->isp_cell[2].platform_data = isp->isp_pdata; isp->isp_cell[2].pdata_size = sizeof(struct isp_platform_data); - r = mfd_add_hotplug_devices(isp->parent, isp->isp_cell, 3); + /* add only amd_isp_capture and amd_isp_i2c_designware to genpd */ + r = mfd_add_hotplug_devices(isp->parent, isp->isp_cell, 2); + if (r) { + drm_err(&adev->ddev, "add mfd hotplug device failed (%d)\n", r); + goto failure; + } + + r = device_for_each_child(isp->parent, &isp->ispgpd, + isp_genpd_add_device); + if (r) { + drm_err(&adev->ddev, "failed to add devices to genpd (%d)\n", r); + goto failure; + } + + r = mfd_add_hotplug_devices(isp->parent, &isp->isp_cell[2], 1); if (r) { - drm_err(&adev->ddev, - "%s: add mfd hotplug device failed\n", __func__); + drm_err(&adev->ddev, "add pinctl hotplug device failed (%d)\n", r); goto failure; } @@ -201,6 +345,9 @@ failure: static int isp_v4_1_1_hw_fini(struct amdgpu_isp *isp) { + device_for_each_child(isp->parent, NULL, + isp_genpd_remove_device); + mfd_remove_devices(isp->parent); kfree(isp->isp_res); diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c index 4cde8a8bcc83..781a5a8a8361 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c @@ -764,11 +764,21 @@ static int jpeg_v2_0_process_interrupt(struct amdgpu_device *adev, return 0; } -static int jpeg_v2_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v2_0_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + + drm_sched_wqueue_stop(&ring->sched); jpeg_v2_0_stop(ring->adev); jpeg_v2_0_start(ring->adev); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v2_0_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c index 8b39e114f3be..5be9cdcae32c 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c @@ -643,11 +643,21 @@ static int jpeg_v2_5_process_interrupt(struct amdgpu_device *adev, return 0; } -static int jpeg_v2_5_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v2_5_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + + drm_sched_wqueue_stop(&ring->sched); jpeg_v2_5_stop_inst(ring->adev, ring->me); jpeg_v2_5_start_inst(ring->adev, ring->me); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v2_5_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c index 2f8510c2986b..a24bd833d644 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c @@ -555,11 +555,21 @@ static int jpeg_v3_0_process_interrupt(struct amdgpu_device *adev, return 0; } -static int jpeg_v3_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v3_0_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + + drm_sched_wqueue_stop(&ring->sched); jpeg_v3_0_stop(ring->adev); jpeg_v3_0_start(ring->adev); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v3_0_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c index f17ec5414fd6..1d4edd77837d 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c @@ -720,14 +720,24 @@ static int jpeg_v4_0_process_interrupt(struct amdgpu_device *adev, return 0; } -static int jpeg_v4_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v4_0_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + if (amdgpu_sriov_vf(ring->adev)) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); jpeg_v4_0_stop(ring->adev); jpeg_v4_0_start(ring->adev); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v4_0_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c index 79e342d5ab28..78441f8fce97 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c @@ -1143,14 +1143,24 @@ static void jpeg_v4_0_3_core_stall_reset(struct amdgpu_ring *ring) WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CORE_RST_CTRL, 0x00); } -static int jpeg_v4_0_3_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v4_0_3_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + if (amdgpu_sriov_vf(ring->adev)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); jpeg_v4_0_3_core_stall_reset(ring); jpeg_v4_0_3_start_jrbc(ring); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v4_0_3_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c index 3b6f65a25646..6f8a16da9d60 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c @@ -834,14 +834,24 @@ static void jpeg_v5_0_1_core_stall_reset(struct amdgpu_ring *ring) WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CORE_RST_CTRL, 0x00); } -static int jpeg_v5_0_1_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v5_0_1_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + if (amdgpu_sriov_vf(ring->adev)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); jpeg_v5_0_1_core_stall_reset(ring); jpeg_v5_0_1_init_jrbc(ring); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v5_0_1_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index c9eba537de09..28eb846280dd 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -1630,10 +1630,12 @@ static int mes_v11_0_hw_init(struct amdgpu_ip_block *ip_block) if (r) goto failure; - r = mes_v11_0_set_hw_resources_1(&adev->mes); - if (r) { - DRM_ERROR("failed mes_v11_0_set_hw_resources_1, r=%d\n", r); - goto failure; + if ((adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x50) { + r = mes_v11_0_set_hw_resources_1(&adev->mes); + if (r) { + DRM_ERROR("failed mes_v11_0_set_hw_resources_1, r=%d\n", r); + goto failure; + } } r = mes_v11_0_query_sched_status(&adev->mes); diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c index b4f17332d466..6b222630f3fa 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c @@ -1742,7 +1742,8 @@ static int mes_v12_0_hw_init(struct amdgpu_ip_block *ip_block) if (r) goto failure; - mes_v12_0_set_hw_resources_1(&adev->mes, AMDGPU_MES_SCHED_PIPE); + if ((adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x4b) + mes_v12_0_set_hw_resources_1(&adev->mes, AMDGPU_MES_SCHED_PIPE); mes_v12_0_init_aggregated_doorbell(&adev->mes); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c index 76167fadb292..cc688ae79e84 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c @@ -76,6 +76,8 @@ static void mmhub_v1_8_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmi static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev) { + uint64_t gart_start = amdgpu_virt_xgmi_migrate_enabled(adev) ? + adev->gmc.vram_start : adev->gmc.fb_start; uint64_t pt_base; u32 inst_mask; int i; @@ -95,10 +97,10 @@ static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev) if (adev->gmc.pdb0_bo) { WREG32_SOC15(MMHUB, i, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, - (u32)(adev->gmc.fb_start >> 12)); + (u32)(gart_start >> 12)); WREG32_SOC15(MMHUB, i, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, - (u32)(adev->gmc.fb_start >> 44)); + (u32)(gart_start >> 44)); WREG32_SOC15(MMHUB, i, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c index a376f072700d..1c22bc11c1f8 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c @@ -31,9 +31,6 @@ #define NPS_MODE_MASK 0x000000FFL -/* Core 0 Port 0 counter */ -#define smnPCIEP_NAK_COUNTER 0x1A340218 - static void nbio_v7_9_remap_hdp_registers(struct amdgpu_device *adev) { WREG32_SOC15(NBIO, 0, regBIF_BX0_REMAP_HDP_MEM_FLUSH_CNTL, @@ -467,22 +464,6 @@ static void nbio_v7_9_init_registers(struct amdgpu_device *adev) } } -static u64 nbio_v7_9_get_pcie_replay_count(struct amdgpu_device *adev) -{ - u32 val, nak_r, nak_g; - - if (adev->flags & AMD_IS_APU) - return 0; - - /* Get the number of NAKs received and generated */ - val = RREG32_PCIE(smnPCIEP_NAK_COUNTER); - nak_r = val & 0xFFFF; - nak_g = val >> 16; - - /* Add the total number of NAKs, i.e the number of replays */ - return (nak_r + nak_g); -} - #define MMIO_REG_HOLE_OFFSET 0x1A000 static void nbio_v7_9_set_reg_remap(struct amdgpu_device *adev) @@ -524,7 +505,6 @@ const struct amdgpu_nbio_funcs nbio_v7_9_funcs = { .get_memory_partition_mode = nbio_v7_9_get_memory_partition_mode, .is_nps_switch_requested = nbio_v7_9_is_nps_switch_requested, .init_registers = nbio_v7_9_init_registers, - .get_pcie_replay_count = nbio_v7_9_get_pcie_replay_count, .set_reg_remap = nbio_v7_9_set_reg_remap, }; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h index f4a91b126c73..73f87131a7e9 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h +++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h @@ -106,7 +106,9 @@ enum psp_gfx_cmd_id /*IDs of performance monitoring/profiling*/ GFX_CMD_ID_CONFIG_SQ_PERFMON = 0x00000046, /* Config CGTT_SQ_CLK_CTRL */ /* Dynamic memory partitioninig (NPS mode change)*/ - GFX_CMD_ID_FB_NPS_MODE = 0x00000048, /* Configure memory partitioning mode */ + GFX_CMD_ID_FB_NPS_MODE = 0x00000048, /* Configure memory partitioning mode */ + GFX_CMD_ID_FB_FW_RESERV_ADDR = 0x00000050, /* Query FW reservation addr */ + GFX_CMD_ID_FB_FW_RESERV_EXT_ADDR = 0x00000051, /* Query FW reservation extended addr */ }; /* PSP boot config sub-commands */ @@ -404,11 +406,19 @@ struct psp_gfx_uresp_bootcfg { uint32_t boot_cfg; /* boot config data */ }; +/* Command-specific response for fw reserve info */ +struct psp_gfx_uresp_fw_reserve_info { + uint32_t reserve_base_address_hi; + uint32_t reserve_base_address_lo; + uint32_t reserve_size; +}; + /* Union of command-specific responses for GPCOM ring. */ union psp_gfx_uresp { struct psp_gfx_uresp_reserved reserved; struct psp_gfx_uresp_bootcfg boot_cfg; struct psp_gfx_uresp_fwar_db_info fwar_db_info; + struct psp_gfx_uresp_fw_reserve_info fw_reserve_info; }; /* Structure of GFX Response buffer. diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c index 145186a1e48f..2c4ebd98927f 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c @@ -94,7 +94,7 @@ static int psp_v10_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); return ret; } @@ -115,7 +115,7 @@ static int psp_v10_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); return ret; } diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index 215543575f47..1a4a26e6ffd2 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -277,11 +277,13 @@ static int psp_v11_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ if (amdgpu_sriov_vf(adev)) - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); else - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); return ret; } @@ -317,13 +319,15 @@ static int psp_v11_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); if (ret) { DRM_ERROR("Failed to wait for sOS ready for ring creation\n"); return ret; @@ -347,8 +351,9 @@ static int psp_v11_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; @@ -381,7 +386,8 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64); - ret = psp_wait_for(psp, offset, 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for(psp, offset, MBOX_TOS_READY_FLAG, + MBOX_TOS_READY_MASK, false); if (ret) { DRM_INFO("psp is not working correctly before mode1 reset!\n"); @@ -395,7 +401,8 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); - ret = psp_wait_for(psp, offset, 0x80000000, 0x80000000, false); + ret = psp_wait_for(psp, offset, MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, + false); if (ret) { DRM_INFO("psp mode 1 reset failed!\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c index 5697760a819b..338d015c0f2e 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c @@ -41,8 +41,9 @@ static int psp_v11_0_8_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, @@ -50,8 +51,9 @@ static int psp_v11_0_8_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; @@ -87,13 +89,15 @@ static int psp_v11_0_8_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -117,8 +121,9 @@ static int psp_v11_0_8_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c index 80153f837470..d54b3e0fabaf 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c @@ -163,7 +163,7 @@ static int psp_v12_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); return ret; } @@ -184,11 +184,13 @@ static int psp_v12_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ if (amdgpu_sriov_vf(adev)) - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); else - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); return ret; } @@ -219,7 +221,8 @@ static int psp_v12_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64); - ret = psp_wait_for(psp, offset, 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for(psp, offset, MBOX_TOS_READY_FLAG, + MBOX_TOS_READY_MASK, false); if (ret) { DRM_INFO("psp is not working correctly before mode1 reset!\n"); @@ -233,7 +236,8 @@ static int psp_v12_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); - ret = psp_wait_for(psp, offset, 0x80000000, 0x80000000, false); + ret = psp_wait_for(psp, offset, MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, + false); if (ret) { DRM_INFO("psp mode 1 reset failed!\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c index df612fd9cc50..58b6b64dcd68 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c @@ -42,7 +42,9 @@ MODULE_FIRMWARE("amdgpu/psp_13_0_5_ta.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_8_toc.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_8_ta.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_0_sos.bin"); +MODULE_FIRMWARE("amdgpu/psp_13_0_0_sos_kicker.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_0_ta.bin"); +MODULE_FIRMWARE("amdgpu/psp_13_0_0_ta_kicker.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_7_sos.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_7_ta.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_10_sos.bin"); @@ -382,8 +384,9 @@ static int psp_v13_0_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_64, @@ -391,8 +394,9 @@ static int psp_v13_0_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; @@ -428,13 +432,15 @@ static int psp_v13_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -458,8 +464,9 @@ static int psp_v13_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c index eaa5512a21da..f65af52c1c19 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c @@ -204,8 +204,9 @@ static int psp_v13_0_4_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_64, @@ -213,8 +214,9 @@ static int psp_v13_0_4_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; @@ -250,13 +252,15 @@ static int psp_v13_0_4_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -280,8 +284,9 @@ static int psp_v13_0_4_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c index 256288c6cd78..ffa47c7d24c9 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c @@ -248,8 +248,9 @@ static int psp_v14_0_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_64, @@ -257,8 +258,9 @@ static int psp_v14_0_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; @@ -294,13 +296,15 @@ static int psp_v14_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_101), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -324,8 +328,9 @@ static int psp_v14_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c index bcde34e4e0a1..c05f3c1f50db 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c @@ -45,6 +45,7 @@ #include "amdgpu_ras.h" MODULE_FIRMWARE("amdgpu/sdma_4_4_2.bin"); +MODULE_FIRMWARE("amdgpu/sdma_4_4_4.bin"); MODULE_FIRMWARE("amdgpu/sdma_4_4_5.bin"); static const struct amdgpu_hwip_reg_entry sdma_reg_list_4_4_2[] = { @@ -109,6 +110,8 @@ static void sdma_v4_4_2_set_ras_funcs(struct amdgpu_device *adev); static void sdma_v4_4_2_update_reset_mask(struct amdgpu_device *adev); static int sdma_v4_4_2_stop_queue(struct amdgpu_ring *ring); static int sdma_v4_4_2_restore_queue(struct amdgpu_ring *ring); +static int sdma_v4_4_2_soft_reset_engine(struct amdgpu_device *adev, + u32 instance_id); static u32 sdma_v4_4_2_get_reg_offset(struct amdgpu_device *adev, u32 instance, u32 offset) @@ -490,7 +493,7 @@ static void sdma_v4_4_2_inst_gfx_stop(struct amdgpu_device *adev, { struct amdgpu_ring *sdma[AMDGPU_MAX_SDMA_INSTANCES]; u32 doorbell_offset, doorbell; - u32 rb_cntl, ib_cntl; + u32 rb_cntl, ib_cntl, sdma_cntl; int i; for_each_inst(i, inst_mask) { @@ -502,6 +505,9 @@ static void sdma_v4_4_2_inst_gfx_stop(struct amdgpu_device *adev, ib_cntl = RREG32_SDMA(i, regSDMA_GFX_IB_CNTL); ib_cntl = REG_SET_FIELD(ib_cntl, SDMA_GFX_IB_CNTL, IB_ENABLE, 0); WREG32_SDMA(i, regSDMA_GFX_IB_CNTL, ib_cntl); + sdma_cntl = RREG32_SDMA(i, regSDMA_CNTL); + sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA_CNTL, UTC_L1_ENABLE, 0); + WREG32_SDMA(i, regSDMA_CNTL, sdma_cntl); if (sdma[i]->use_doorbell) { doorbell = RREG32_SDMA(i, regSDMA_GFX_DOORBELL); @@ -995,6 +1001,7 @@ static int sdma_v4_4_2_inst_start(struct amdgpu_device *adev, /* set utc l1 enable flag always to 1 */ temp = RREG32_SDMA(i, regSDMA_CNTL); temp = REG_SET_FIELD(temp, SDMA_CNTL, UTC_L1_ENABLE, 1); + WREG32_SDMA(i, regSDMA_CNTL, temp); if (amdgpu_ip_version(adev, SDMA0_HWIP, 0) < IP_VERSION(4, 4, 5)) { /* enable context empty interrupt during initialization */ @@ -1337,6 +1344,7 @@ static bool sdma_v4_4_2_fw_support_paging_queue(struct amdgpu_device *adev) static const struct amdgpu_sdma_funcs sdma_v4_4_2_sdma_funcs = { .stop_kernel_queue = &sdma_v4_4_2_stop_queue, .start_kernel_queue = &sdma_v4_4_2_restore_queue, + .soft_reset_kernel_queue = &sdma_v4_4_2_soft_reset_engine, }; static int sdma_v4_4_2_early_init(struct amdgpu_ip_block *ip_block) @@ -1648,45 +1656,27 @@ static bool sdma_v4_4_2_is_queue_selected(struct amdgpu_device *adev, uint32_t i return (context_status & SDMA_GFX_CONTEXT_STATUS__SELECTED_MASK) != 0; } -static bool sdma_v4_4_2_ring_is_guilty(struct amdgpu_ring *ring) -{ - struct amdgpu_device *adev = ring->adev; - uint32_t instance_id = ring->me; - - return sdma_v4_4_2_is_queue_selected(adev, instance_id, false); -} - -static bool sdma_v4_4_2_page_ring_is_guilty(struct amdgpu_ring *ring) -{ - struct amdgpu_device *adev = ring->adev; - uint32_t instance_id = ring->me; - - if (!adev->sdma.has_page_queue) - return false; - - return sdma_v4_4_2_is_queue_selected(adev, instance_id, true); -} - -static int sdma_v4_4_2_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +static int sdma_v4_4_2_reset_queue(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; - u32 id = GET_INST(SDMA0, ring->me); + u32 id = ring->me; int r; if (!(adev->sdma.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) return -EOPNOTSUPP; - amdgpu_amdkfd_suspend(adev, false); + amdgpu_amdkfd_suspend(adev, true); r = amdgpu_sdma_reset_engine(adev, id); - amdgpu_amdkfd_resume(adev, false); - + amdgpu_amdkfd_resume(adev, true); return r; } static int sdma_v4_4_2_stop_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - u32 instance_id = GET_INST(SDMA0, ring->me); + u32 instance_id = ring->me; u32 inst_mask; uint64_t rptr; @@ -1724,8 +1714,8 @@ static int sdma_v4_4_2_stop_queue(struct amdgpu_ring *ring) static int sdma_v4_4_2_restore_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - u32 inst_mask; - int i; + u32 inst_mask, tmp_mask; + int i, r; inst_mask = 1 << ring->me; udelay(50); @@ -1742,7 +1732,33 @@ static int sdma_v4_4_2_restore_queue(struct amdgpu_ring *ring) return -ETIMEDOUT; } - return sdma_v4_4_2_inst_start(adev, inst_mask, true); + r = sdma_v4_4_2_inst_start(adev, inst_mask, true); + if (r) + return r; + + tmp_mask = inst_mask; + for_each_inst(i, tmp_mask) { + ring = &adev->sdma.instance[i].ring; + + amdgpu_fence_driver_force_completion(ring); + + if (adev->sdma.has_page_queue) { + struct amdgpu_ring *page = &adev->sdma.instance[i].page; + + amdgpu_fence_driver_force_completion(page); + } + } + + return r; +} + +static int sdma_v4_4_2_soft_reset_engine(struct amdgpu_device *adev, + u32 instance_id) +{ + /* For SDMA 4.x, use the existing DPM interface for backward compatibility + * we need to convert the logical instance ID to physical instance ID before reset. + */ + return amdgpu_dpm_reset_sdma(adev, 1 << GET_INST(SDMA0, instance_id)); } static int sdma_v4_4_2_set_trap_irq_state(struct amdgpu_device *adev, @@ -2139,7 +2155,6 @@ static const struct amdgpu_ring_funcs sdma_v4_4_2_ring_funcs = { .emit_reg_wait = sdma_v4_4_2_ring_emit_reg_wait, .emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper, .reset = sdma_v4_4_2_reset_queue, - .is_guilty = sdma_v4_4_2_ring_is_guilty, }; static const struct amdgpu_ring_funcs sdma_v4_4_2_page_ring_funcs = { @@ -2172,7 +2187,6 @@ static const struct amdgpu_ring_funcs sdma_v4_4_2_page_ring_funcs = { .emit_reg_wait = sdma_v4_4_2_ring_emit_reg_wait, .emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper, .reset = sdma_v4_4_2_reset_queue, - .is_guilty = sdma_v4_4_2_page_ring_is_guilty, }; static void sdma_v4_4_2_set_ring_funcs(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 9505ae96fbec..4d72b085b3dd 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -1399,6 +1399,7 @@ static int sdma_v5_0_sw_init(struct amdgpu_ip_block *ip_block) return r; for (i = 0; i < adev->sdma.num_instances; i++) { + mutex_init(&adev->sdma.instance[i].engine_reset_mutex); adev->sdma.instance[i].funcs = &sdma_v5_0_sdma_funcs; ring = &adev->sdma.instance[i].ring; ring->ring_obj = NULL; @@ -1538,12 +1539,19 @@ static int sdma_v5_0_soft_reset(struct amdgpu_ip_block *ip_block) return 0; } -static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; u32 inst_id = ring->me; + int r; + + amdgpu_amdkfd_suspend(adev, true); + r = amdgpu_sdma_reset_engine(adev, inst_id); + amdgpu_amdkfd_resume(adev, true); - return amdgpu_sdma_reset_engine(adev, inst_id); + return r; } static int sdma_v5_0_stop_queue(struct amdgpu_ring *ring) @@ -1610,7 +1618,10 @@ static int sdma_v5_0_restore_queue(struct amdgpu_ring *ring) r = sdma_v5_0_gfx_resume_instance(adev, inst_id, true); amdgpu_gfx_rlc_exit_safe_mode(adev, 0); - return r; + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + return 0; } static int sdma_v5_0_ring_preempt_ib(struct amdgpu_ring *ring) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index a6e612b4a892..42a25150f83a 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -1318,6 +1318,7 @@ static int sdma_v5_2_sw_init(struct amdgpu_ip_block *ip_block) } for (i = 0; i < adev->sdma.num_instances; i++) { + mutex_init(&adev->sdma.instance[i].engine_reset_mutex); adev->sdma.instance[i].funcs = &sdma_v5_2_sdma_funcs; ring = &adev->sdma.instance[i].ring; ring->ring_obj = NULL; @@ -1451,12 +1452,19 @@ static int sdma_v5_2_wait_for_idle(struct amdgpu_ip_block *ip_block) return -ETIMEDOUT; } -static int sdma_v5_2_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +static int sdma_v5_2_reset_queue(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; u32 inst_id = ring->me; + int r; + + amdgpu_amdkfd_suspend(adev, true); + r = amdgpu_sdma_reset_engine(adev, inst_id); + amdgpu_amdkfd_resume(adev, true); - return amdgpu_sdma_reset_engine(adev, inst_id); + return r; } static int sdma_v5_2_stop_queue(struct amdgpu_ring *ring) @@ -1526,7 +1534,10 @@ static int sdma_v5_2_restore_queue(struct amdgpu_ring *ring) r = sdma_v5_2_gfx_resume_instance(adev, inst_id, true); amdgpu_gfx_rlc_exit_safe_mode(adev, 0); - return r; + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + return 0; } static int sdma_v5_2_ring_preempt_ib(struct amdgpu_ring *ring) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c index 5a70ae17be04..d2effa531817 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c @@ -1374,9 +1374,42 @@ static int sdma_v6_0_sw_init(struct amdgpu_ip_block *ip_block) else DRM_ERROR("Failed to allocated memory for SDMA IP Dump\n"); - /* add firmware version checks here */ - if (0 && !adev->sdma.disable_uq) - adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + switch (amdgpu_ip_version(adev, SDMA0_HWIP, 0)) { + case IP_VERSION(6, 0, 0): + if ((adev->sdma.instance[0].fw_version >= 24) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 0, 1): + if ((adev->sdma.instance[0].fw_version >= 18) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 0, 2): + if ((adev->sdma.instance[0].fw_version >= 21) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 0, 3): + if ((adev->sdma.instance[0].fw_version >= 25) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 1, 0): + if ((adev->sdma.instance[0].fw_version >= 14) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 1, 1): + if ((adev->sdma.instance[0].fw_version >= 17) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 1, 2): + if ((adev->sdma.instance[0].fw_version >= 15) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 1, 3): + if ((adev->sdma.instance[0].fw_version >= 10) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + default: + break; + } r = amdgpu_sdma_sysfs_reset_mask_init(adev); if (r) @@ -1537,7 +1570,9 @@ static int sdma_v6_0_ring_preempt_ib(struct amdgpu_ring *ring) return r; } -static int sdma_v6_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +static int sdma_v6_0_reset_queue(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int i, r; @@ -1555,11 +1590,18 @@ static int sdma_v6_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) return -EINVAL; } + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(adev, ring, vmid, true); if (r) return r; - return sdma_v6_0_gfx_resume_instance(adev, i, true); + r = sdma_v6_0_gfx_resume_instance(adev, i, true); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static int sdma_v6_0_set_trap_irq_state(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c index ad47d0bdf777..99a080bad2a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c @@ -802,7 +802,9 @@ static bool sdma_v7_0_check_soft_reset(struct amdgpu_ip_block *ip_block) return false; } -static int sdma_v7_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +static int sdma_v7_0_reset_queue(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int i, r; @@ -820,11 +822,18 @@ static int sdma_v7_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) return -EINVAL; } + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(adev, ring, vmid, true); if (r) return r; - return sdma_v7_0_gfx_resume_instance(adev, i, true); + r = sdma_v7_0_gfx_resume_instance(adev, i, true); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } /** @@ -1349,9 +1358,15 @@ static int sdma_v7_0_sw_init(struct amdgpu_ip_block *ip_block) else DRM_ERROR("Failed to allocated memory for SDMA IP Dump\n"); - /* add firmware version checks here */ - if (0 && !adev->sdma.disable_uq) - adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + switch (amdgpu_ip_version(adev, SDMA0_HWIP, 0)) { + case IP_VERSION(7, 0, 0): + case IP_VERSION(7, 0, 1): + if ((adev->sdma.instance[0].fw_version >= 7836028) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + default: + break; + } return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h index ef7c603b50ae..c8ac11a9cdef 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15.h @@ -118,7 +118,6 @@ int vega10_reg_base_init(struct amdgpu_device *adev); int vega20_reg_base_init(struct amdgpu_device *adev); int arct_reg_base_init(struct amdgpu_device *adev); int aldebaran_reg_base_init(struct amdgpu_device *adev); -void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev); u64 aqua_vanjaram_encode_ext_smn_addressing(int ext_id); int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev); ssize_t aqua_vanjaram_get_reg_state(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index b5071f77f78d..eec9133e1b2c 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -1967,18 +1967,27 @@ static int vcn_v4_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p, return 0; } -static int vcn_v4_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int vcn_v4_0_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me]; + int r; if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); vcn_v4_0_stop(vinst); vcn_v4_0_start(vinst); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static struct amdgpu_ring_funcs vcn_v4_0_unified_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c index 5a33140f5723..d8fd32c1e38e 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c @@ -1594,7 +1594,9 @@ static void vcn_v4_0_3_unified_ring_set_wptr(struct amdgpu_ring *ring) } } -static int vcn_v4_0_3_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int vcn_v4_0_3_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { int r = 0; int vcn_inst; @@ -1607,6 +1609,8 @@ static int vcn_v4_0_3_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); + vcn_inst = GET_INST(VCN, ring->me); r = amdgpu_dpm_reset_vcn(adev, 1 << vcn_inst); @@ -1621,8 +1625,11 @@ static int vcn_v4_0_3_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) vcn_v4_0_3_hw_init_inst(vinst); vcn_v4_0_3_start_dpg_mode(vinst, adev->vcn.inst[ring->me].indirect_sram); r = amdgpu_ring_test_helper(ring); - - return r; + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amdgpu_ring_funcs vcn_v4_0_3_unified_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c index 16ade84facc7..7e37ddea6355 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c @@ -1465,18 +1465,27 @@ static void vcn_v4_0_5_unified_ring_set_wptr(struct amdgpu_ring *ring) } } -static int vcn_v4_0_5_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int vcn_v4_0_5_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me]; + int r; if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); vcn_v4_0_5_stop(vinst); vcn_v4_0_5_start(vinst); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static struct amdgpu_ring_funcs vcn_v4_0_5_unified_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c index f8e3f0b882da..47c0bcc9e7d8 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c @@ -1192,18 +1192,27 @@ static void vcn_v5_0_0_unified_ring_set_wptr(struct amdgpu_ring *ring) } } -static int vcn_v5_0_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int vcn_v5_0_0_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me]; + int r; if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); vcn_v5_0_0_stop(vinst); vcn_v5_0_0_start(vinst); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amdgpu_ring_funcs vcn_v5_0_0_unified_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c index 338cf43c45fe..cdefd7fcb0da 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c @@ -669,6 +669,9 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst, if (indirect) amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM); + /* resetting ring, fw should not check RB ring */ + fw_shared->sq.queue_mode |= FW_QUEUE_RING_RESET; + /* Pause dpg */ vcn_v5_0_1_pause_dpg_mode(vinst, &state); @@ -681,7 +684,7 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst, tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE); tmp &= ~(VCN_RB_ENABLE__RB1_EN_MASK); WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp); - fw_shared->sq.queue_mode |= FW_QUEUE_RING_RESET; + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_RPTR, 0); WREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR, 0); @@ -692,6 +695,7 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst, tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE); tmp |= VCN_RB_ENABLE__RB1_EN_MASK; WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp); + /* resetting done, fw can check RB ring */ fw_shared->sq.queue_mode &= ~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF); WREG32_SOC15(VCN, vcn_inst, regVCN_RB1_DB_CTRL, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index a2149afa5803..828a9ceef1e7 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -22,7 +22,6 @@ */ #include <linux/device.h> -#include <linux/export.h> #include <linux/err.h> #include <linux/fs.h> #include <linux/file.h> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index bf0854bd5555..7e749f9b6d69 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -971,7 +971,7 @@ int kgd2kfd_pre_reset(struct kfd_dev *kfd, kfd_smi_event_update_gpu_reset(node, false, reset_context); } - kgd2kfd_suspend(kfd, false); + kgd2kfd_suspend(kfd, true); for (i = 0; i < kfd->num_nodes; i++) kfd_signal_reset_event(kfd->nodes[i]); @@ -1013,13 +1013,33 @@ int kgd2kfd_post_reset(struct kfd_dev *kfd) return 0; } -bool kfd_is_locked(void) +bool kfd_is_locked(struct kfd_dev *kfd) { + uint8_t id = 0; + struct kfd_node *dev; + lockdep_assert_held(&kfd_processes_mutex); - return (kfd_locked > 0); + + /* check reset/suspend lock */ + if (kfd_locked > 0) + return true; + + if (kfd) + return kfd->kfd_dev_lock > 0; + + /* check lock on all cgroup accessible devices */ + while (kfd_topology_enum_kfd_devices(id++, &dev) == 0) { + if (!dev || kfd_devcgroup_check_permission(dev)) + continue; + + if (dev->kfd->kfd_dev_lock > 0) + return true; + } + + return false; } -void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) +void kgd2kfd_suspend(struct kfd_dev *kfd, bool suspend_proc) { struct kfd_node *node; int i; @@ -1027,14 +1047,8 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) if (!kfd->init_complete) return; - /* for runtime suspend, skip locking kfd */ - if (!run_pm) { - mutex_lock(&kfd_processes_mutex); - /* For first KFD device suspend all the KFD processes */ - if (++kfd_locked == 1) - kfd_suspend_all_processes(); - mutex_unlock(&kfd_processes_mutex); - } + if (suspend_proc) + kgd2kfd_suspend_process(kfd); for (i = 0; i < kfd->num_nodes; i++) { node = kfd->nodes[i]; @@ -1042,7 +1056,7 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) } } -int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) +int kgd2kfd_resume(struct kfd_dev *kfd, bool resume_proc) { int ret, i; @@ -1055,14 +1069,36 @@ int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) return ret; } - /* for runtime resume, skip unlocking kfd */ - if (!run_pm) { - mutex_lock(&kfd_processes_mutex); - if (--kfd_locked == 0) - ret = kfd_resume_all_processes(); - WARN_ONCE(kfd_locked < 0, "KFD suspend / resume ref. error"); - mutex_unlock(&kfd_processes_mutex); - } + if (resume_proc) + ret = kgd2kfd_resume_process(kfd); + + return ret; +} + +void kgd2kfd_suspend_process(struct kfd_dev *kfd) +{ + if (!kfd->init_complete) + return; + + mutex_lock(&kfd_processes_mutex); + /* For first KFD device suspend all the KFD processes */ + if (++kfd_locked == 1) + kfd_suspend_all_processes(); + mutex_unlock(&kfd_processes_mutex); +} + +int kgd2kfd_resume_process(struct kfd_dev *kfd) +{ + int ret = 0; + + if (!kfd->init_complete) + return 0; + + mutex_lock(&kfd_processes_mutex); + if (--kfd_locked == 0) + ret = kfd_resume_all_processes(); + WARN_ONCE(kfd_locked < 0, "KFD suspend / resume ref. error"); + mutex_unlock(&kfd_processes_mutex); return ret; } @@ -1442,24 +1478,53 @@ unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_node *node) kfd_get_num_sdma_engines(node); } -int kgd2kfd_check_and_lock_kfd(void) +int kgd2kfd_check_and_lock_kfd(struct kfd_dev *kfd) { + struct kfd_process *p; + int r = 0, temp, idx; + mutex_lock(&kfd_processes_mutex); - if (!hash_empty(kfd_processes_table) || kfd_is_locked()) { - mutex_unlock(&kfd_processes_mutex); - return -EBUSY; + + if (hash_empty(kfd_processes_table) && !kfd_is_locked(kfd)) + goto out; + + /* fail under system reset/resume or kfd device is partition switching. */ + if (kfd_is_locked(kfd)) { + r = -EBUSY; + goto out; + } + + /* + * ensure all running processes are cgroup excluded from device before mode switch. + * i.e. no pdd was created on the process socket. + */ + idx = srcu_read_lock(&kfd_processes_srcu); + hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) { + int i; + + for (i = 0; i < p->n_pdds; i++) { + if (p->pdds[i]->dev->kfd != kfd) + continue; + + r = -EBUSY; + goto proc_check_unlock; + } } - ++kfd_locked; +proc_check_unlock: + srcu_read_unlock(&kfd_processes_srcu, idx); +out: + if (!r) + ++kfd->kfd_dev_lock; mutex_unlock(&kfd_processes_mutex); - return 0; + return r; } -void kgd2kfd_unlock_kfd(void) +void kgd2kfd_unlock_kfd(struct kfd_dev *kfd) { mutex_lock(&kfd_processes_mutex); - --kfd_locked; + --kfd->kfd_dev_lock; mutex_unlock(&kfd_processes_mutex); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c index dbcb60eb54b2..1d170dc50df3 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c @@ -23,7 +23,6 @@ */ #include <linux/device.h> -#include <linux/export.h> #include <linux/err.h> #include <linux/fs.h> #include <linux/sched.h> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c index 8fa6489b6f5d..505036968a77 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c @@ -240,7 +240,7 @@ static int pm_map_queues_v9(struct packet_manager *pm, uint32_t *buffer, packet->bitfields2.engine_sel = engine_sel__mes_map_queues__compute_vi; - packet->bitfields2.gws_control_queue = q->gws ? 1 : 0; + packet->bitfields2.gws_control_queue = q->properties.is_gws ? 1 : 0; packet->bitfields2.extended_engine_sel = extended_engine_sel__mes_map_queues__legacy_engine_sel; packet->bitfields2.queue_type = diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index d221c58dccc3..67694bcd9464 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -372,6 +372,9 @@ struct kfd_dev { /* bitmap for dynamic doorbell allocation from doorbell object */ unsigned long *doorbell_bitmap; + + /* for dynamic partitioning */ + int kfd_dev_lock; }; enum kfd_mempool { @@ -1536,7 +1539,7 @@ static inline bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev) int kfd_send_exception_to_runtime(struct kfd_process *p, unsigned int queue_id, uint64_t error_reason); -bool kfd_is_locked(void); +bool kfd_is_locked(struct kfd_dev *kfd); /* Compute profile */ void kfd_inc_compute_active(struct kfd_node *dev); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 722ac1662bdc..5be28c6c4f6a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -854,7 +854,7 @@ struct kfd_process *kfd_create_process(struct task_struct *thread) */ mutex_lock(&kfd_processes_mutex); - if (kfd_is_locked()) { + if (kfd_is_locked(NULL)) { pr_debug("KFD is locked! Cannot create process"); process = ERR_PTR(-EINVAL); goto out; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index 865dca2547de..a0f22ea6d15a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -1171,13 +1171,12 @@ svm_range_split_head(struct svm_range *prange, uint64_t new_start, } static void -svm_range_add_child(struct svm_range *prange, struct mm_struct *mm, - struct svm_range *pchild, enum svm_work_list_ops op) +svm_range_add_child(struct svm_range *prange, struct svm_range *pchild, enum svm_work_list_ops op) { pr_debug("add child 0x%p [0x%lx 0x%lx] to prange 0x%p child list %d\n", pchild, pchild->start, pchild->last, prange, op); - pchild->work_item.mm = mm; + pchild->work_item.mm = NULL; pchild->work_item.op = op; list_add_tail(&pchild->child_list, &prange->child_list); } @@ -1278,7 +1277,7 @@ svm_range_get_pte_flags(struct kfd_node *node, mapping_flags |= ext_coherent ? AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC; /* system memory accessed by the dGPU */ } else { - if (gc_ip_version < IP_VERSION(9, 5, 0)) + if (gc_ip_version < IP_VERSION(9, 5, 0) || ext_coherent) mapping_flags |= AMDGPU_VM_MTYPE_UC; else mapping_flags |= AMDGPU_VM_MTYPE_NC; @@ -2394,15 +2393,17 @@ svm_range_add_list_work(struct svm_range_list *svms, struct svm_range *prange, prange->work_item.op != SVM_OP_UNMAP_RANGE) prange->work_item.op = op; } else { - prange->work_item.op = op; - - /* Pairs with mmput in deferred_list_work */ - mmget(mm); - prange->work_item.mm = mm; - list_add_tail(&prange->deferred_list, - &prange->svms->deferred_range_list); - pr_debug("add prange 0x%p [0x%lx 0x%lx] to work list op %d\n", - prange, prange->start, prange->last, op); + /* Pairs with mmput in deferred_list_work. + * If process is exiting and mm is gone, don't update mmu notifier. + */ + if (mmget_not_zero(mm)) { + prange->work_item.mm = mm; + prange->work_item.op = op; + list_add_tail(&prange->deferred_list, + &prange->svms->deferred_range_list); + pr_debug("add prange 0x%p [0x%lx 0x%lx] to work list op %d\n", + prange, prange->start, prange->last, op); + } } spin_unlock(&svms->deferred_list_lock); } @@ -2416,8 +2417,7 @@ void schedule_deferred_list_work(struct svm_range_list *svms) } static void -svm_range_unmap_split(struct mm_struct *mm, struct svm_range *parent, - struct svm_range *prange, unsigned long start, +svm_range_unmap_split(struct svm_range *parent, struct svm_range *prange, unsigned long start, unsigned long last) { struct svm_range *head; @@ -2438,12 +2438,12 @@ svm_range_unmap_split(struct mm_struct *mm, struct svm_range *parent, svm_range_split(tail, last + 1, tail->last, &head); if (head != prange && tail != prange) { - svm_range_add_child(parent, mm, head, SVM_OP_UNMAP_RANGE); - svm_range_add_child(parent, mm, tail, SVM_OP_ADD_RANGE); + svm_range_add_child(parent, head, SVM_OP_UNMAP_RANGE); + svm_range_add_child(parent, tail, SVM_OP_ADD_RANGE); } else if (tail != prange) { - svm_range_add_child(parent, mm, tail, SVM_OP_UNMAP_RANGE); + svm_range_add_child(parent, tail, SVM_OP_UNMAP_RANGE); } else if (head != prange) { - svm_range_add_child(parent, mm, head, SVM_OP_UNMAP_RANGE); + svm_range_add_child(parent, head, SVM_OP_UNMAP_RANGE); } else if (parent != prange) { prange->work_item.op = SVM_OP_UNMAP_RANGE; } @@ -2520,14 +2520,14 @@ svm_range_unmap_from_cpu(struct mm_struct *mm, struct svm_range *prange, l = min(last, pchild->last); if (l >= s) svm_range_unmap_from_gpus(pchild, s, l, trigger); - svm_range_unmap_split(mm, prange, pchild, start, last); + svm_range_unmap_split(prange, pchild, start, last); mutex_unlock(&pchild->lock); } s = max(start, prange->start); l = min(last, prange->last); if (l >= s) svm_range_unmap_from_gpus(prange, s, l, trigger); - svm_range_unmap_split(mm, prange, prange, start, last); + svm_range_unmap_split(prange, prange, start, last); if (unmap_parent) svm_range_add_list_work(svms, prange, mm, SVM_OP_UNMAP_RANGE); @@ -2570,8 +2570,6 @@ svm_range_cpu_invalidate_pagetables(struct mmu_interval_notifier *mni, if (range->event == MMU_NOTIFY_RELEASE) return true; - if (!mmget_not_zero(mni->mm)) - return true; start = mni->interval_tree.start; last = mni->interval_tree.last; @@ -2598,7 +2596,6 @@ svm_range_cpu_invalidate_pagetables(struct mmu_interval_notifier *mni, } svm_range_unlock(prange); - mmput(mni->mm); return true; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index baa2374acdeb..4ec73f33535e 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -510,6 +510,10 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, dev->node_props.capability |= HSA_CAP_AQL_QUEUE_DOUBLE_MAP; + if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(10, 0, 0) && + (dev->gpu->adev->sdma.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) + dev->node_props.capability2 |= HSA_CAP2_PER_SDMA_QUEUE_RESET_SUPPORTED; + sysfs_show_32bit_prop(buffer, offs, "max_engine_clk_fcompute", dev->node_props.max_engine_clk_fcompute); @@ -2008,8 +2012,6 @@ static void kfd_topology_set_capabilities(struct kfd_topology_device *dev) if (!amdgpu_sriov_vf(dev->gpu->adev)) dev->node_props.capability |= HSA_CAP_PER_QUEUE_RESET_SUPPORTED; - if (dev->gpu->adev->sdma.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE) - dev->node_props.capability2 |= HSA_CAP2_PER_SDMA_QUEUE_RESET_SUPPORTED; } else { dev->node_props.debug_prop |= HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX10 | HSA_DBG_WATCH_ADDR_MASK_HI_BIT; diff --git a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c index faed84172dd4..8bc36f04b1b7 100644 --- a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c +++ b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c @@ -21,6 +21,7 @@ * */ +#include <linux/export.h> #include <linux/init.h> #include <linux/module.h> #include <linux/platform_device.h> diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index d3100f641ac6..3dd4f9e9931d 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -1758,10 +1758,11 @@ dm_dmub_send_vbios_gpint_command(struct amdgpu_device *adev, return DMUB_STATUS_TIMEOUT; } -static struct dml2_soc_bb *dm_dmub_get_vbios_bounding_box(struct amdgpu_device *adev) +static void *dm_dmub_get_vbios_bounding_box(struct amdgpu_device *adev) { - struct dml2_soc_bb *bb; + void *bb; long long addr; + unsigned int bb_size; int i = 0; uint16_t chunk; enum dmub_gpint_command send_addrs[] = { @@ -1774,6 +1775,7 @@ static struct dml2_soc_bb *dm_dmub_get_vbios_bounding_box(struct amdgpu_device * switch (amdgpu_ip_version(adev, DCE_HWIP, 0)) { case IP_VERSION(4, 0, 1): + bb_size = sizeof(struct dml2_soc_bb); break; default: return NULL; @@ -1781,7 +1783,7 @@ static struct dml2_soc_bb *dm_dmub_get_vbios_bounding_box(struct amdgpu_device * bb = dm_allocate_gpu_mem(adev, DC_MEM_ALLOC_TYPE_GART, - sizeof(struct dml2_soc_bb), + bb_size, &addr); if (!bb) return NULL; @@ -1847,7 +1849,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) mutex_init(&adev->dm.audio_lock); if (amdgpu_dm_irq_init(adev)) { - drm_err(adev_to_drm(adev), "amdgpu: failed to initialize DM IRQ support.\n"); + drm_err(adev_to_drm(adev), "failed to initialize DM IRQ support.\n"); goto error; } @@ -2037,7 +2039,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) adev->dm.hpd_rx_offload_wq = hpd_rx_irq_create_workqueue(adev); if (!adev->dm.hpd_rx_offload_wq) { - drm_err(adev_to_drm(adev), "amdgpu: failed to create hpd rx offload workqueue.\n"); + drm_err(adev_to_drm(adev), "failed to create hpd rx offload workqueue.\n"); goto error; } @@ -2053,7 +2055,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) adev->dm.freesync_module = mod_freesync_create(adev->dm.dc); if (!adev->dm.freesync_module) { drm_err(adev_to_drm(adev), - "amdgpu: failed to initialize freesync_module.\n"); + "failed to initialize freesync_module.\n"); } else drm_dbg_driver(adev_to_drm(adev), "amdgpu: freesync_module init done %p.\n", adev->dm.freesync_module); @@ -2064,7 +2066,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) adev->dm.vblank_control_workqueue = create_singlethread_workqueue("dm_vblank_control_workqueue"); if (!adev->dm.vblank_control_workqueue) - drm_err(adev_to_drm(adev), "amdgpu: failed to initialize vblank_workqueue.\n"); + drm_err(adev_to_drm(adev), "failed to initialize vblank_workqueue.\n"); } if (adev->dm.dc->caps.ips_support && @@ -2075,7 +2077,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) adev->dm.hdcp_workqueue = hdcp_create_workqueue(adev, &init_params.cp_psp, adev->dm.dc); if (!adev->dm.hdcp_workqueue) - drm_err(adev_to_drm(adev), "amdgpu: failed to initialize hdcp_workqueue.\n"); + drm_err(adev_to_drm(adev), "failed to initialize hdcp_workqueue.\n"); else drm_dbg_driver(adev_to_drm(adev), "amdgpu: hdcp_workqueue init done %p.\n", adev->dm.hdcp_workqueue); @@ -2085,20 +2087,20 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) init_completion(&adev->dm.dmub_aux_transfer_done); adev->dm.dmub_notify = kzalloc(sizeof(struct dmub_notification), GFP_KERNEL); if (!adev->dm.dmub_notify) { - drm_info(adev_to_drm(adev), "amdgpu: fail to allocate adev->dm.dmub_notify"); + drm_info(adev_to_drm(adev), "fail to allocate adev->dm.dmub_notify"); goto error; } adev->dm.delayed_hpd_wq = create_singlethread_workqueue("amdgpu_dm_hpd_wq"); if (!adev->dm.delayed_hpd_wq) { - drm_err(adev_to_drm(adev), "amdgpu: failed to create hpd offload workqueue.\n"); + drm_err(adev_to_drm(adev), "failed to create hpd offload workqueue.\n"); goto error; } amdgpu_dm_outbox_init(adev); if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_AUX_REPLY, dmub_aux_setconfig_callback, false)) { - drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub aux callback"); + drm_err(adev_to_drm(adev), "fail to register dmub aux callback"); goto error; } @@ -2107,7 +2109,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_FUSED_IO, dmub_aux_fused_io_callback, false)) { - drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub fused io callback"); + drm_err(adev_to_drm(adev), "fail to register dmub fused io callback"); goto error; } /* Enable outbox notification only after IRQ handlers are registered and DMUB is alive. @@ -2125,7 +2127,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) if (amdgpu_dm_initialize_drm_device(adev)) { drm_err(adev_to_drm(adev), - "amdgpu: failed to initialize sw for display support.\n"); + "failed to initialize sw for display support.\n"); goto error; } @@ -2140,14 +2142,14 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) if (drm_vblank_init(adev_to_drm(adev), adev->dm.display_indexes_num)) { drm_err(adev_to_drm(adev), - "amdgpu: failed to initialize sw for display support.\n"); + "failed to initialize sw for display support.\n"); goto error; } #if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) amdgpu_dm_crtc_secure_display_create_contexts(adev); if (!adev->dm.secure_display_ctx.crtc_ctx) - drm_err(adev_to_drm(adev), "amdgpu: failed to initialize secure display contexts.\n"); + drm_err(adev_to_drm(adev), "failed to initialize secure display contexts.\n"); if (amdgpu_ip_version(adev, DCE_HWIP, 0) >= IP_VERSION(4, 0, 1)) adev->dm.secure_display_ctx.support_mul_roi = true; @@ -2404,6 +2406,7 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev) DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_5_TRACEBUFF DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_6_FW_STATE DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_7_SCRATCH_MEM + DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_IB_MEM DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_SHARED_STATE }; int r; @@ -2570,7 +2573,7 @@ static int dm_sw_init(struct amdgpu_ip_block *ip_block) adev->dm.cgs_device = amdgpu_cgs_create_device(adev); if (!adev->dm.cgs_device) { - drm_err(adev_to_drm(adev), "amdgpu: failed to create cgs device.\n"); + drm_err(adev_to_drm(adev), "failed to create cgs device.\n"); return -EINVAL; } @@ -3060,6 +3063,77 @@ static void hpd_rx_irq_work_suspend(struct amdgpu_display_manager *dm) } } +static int dm_cache_state(struct amdgpu_device *adev) +{ + int r; + + adev->dm.cached_state = drm_atomic_helper_suspend(adev_to_drm(adev)); + if (IS_ERR(adev->dm.cached_state)) { + r = PTR_ERR(adev->dm.cached_state); + adev->dm.cached_state = NULL; + } + + return adev->dm.cached_state ? 0 : r; +} + +static void dm_destroy_cached_state(struct amdgpu_device *adev) +{ + struct amdgpu_display_manager *dm = &adev->dm; + struct drm_device *ddev = adev_to_drm(adev); + struct dm_plane_state *dm_new_plane_state; + struct drm_plane_state *new_plane_state; + struct dm_crtc_state *dm_new_crtc_state; + struct drm_crtc_state *new_crtc_state; + struct drm_plane *plane; + struct drm_crtc *crtc; + int i; + + if (!dm->cached_state) + return; + + /* Force mode set in atomic commit */ + for_each_new_crtc_in_state(dm->cached_state, crtc, new_crtc_state, i) { + new_crtc_state->active_changed = true; + dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); + reset_freesync_config_for_crtc(dm_new_crtc_state); + } + + /* + * atomic_check is expected to create the dc states. We need to release + * them here, since they were duplicated as part of the suspend + * procedure. + */ + for_each_new_crtc_in_state(dm->cached_state, crtc, new_crtc_state, i) { + dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); + if (dm_new_crtc_state->stream) { + WARN_ON(kref_read(&dm_new_crtc_state->stream->refcount) > 1); + dc_stream_release(dm_new_crtc_state->stream); + dm_new_crtc_state->stream = NULL; + } + dm_new_crtc_state->base.color_mgmt_changed = true; + } + + for_each_new_plane_in_state(dm->cached_state, plane, new_plane_state, i) { + dm_new_plane_state = to_dm_plane_state(new_plane_state); + if (dm_new_plane_state->dc_state) { + WARN_ON(kref_read(&dm_new_plane_state->dc_state->refcount) > 1); + dc_plane_state_release(dm_new_plane_state->dc_state); + dm_new_plane_state->dc_state = NULL; + } + } + + drm_atomic_helper_resume(ddev, dm->cached_state); + + dm->cached_state = NULL; +} + +static void dm_complete(struct amdgpu_ip_block *ip_block) +{ + struct amdgpu_device *adev = ip_block->adev; + + dm_destroy_cached_state(adev); +} + static int dm_prepare_suspend(struct amdgpu_ip_block *ip_block) { struct amdgpu_device *adev = ip_block->adev; @@ -3068,11 +3142,8 @@ static int dm_prepare_suspend(struct amdgpu_ip_block *ip_block) return 0; WARN_ON(adev->dm.cached_state); - adev->dm.cached_state = drm_atomic_helper_suspend(adev_to_drm(adev)); - if (IS_ERR(adev->dm.cached_state)) - return PTR_ERR(adev->dm.cached_state); - return 0; + return dm_cache_state(adev); } static int dm_suspend(struct amdgpu_ip_block *ip_block) @@ -3106,9 +3177,10 @@ static int dm_suspend(struct amdgpu_ip_block *ip_block) } if (!adev->dm.cached_state) { - adev->dm.cached_state = drm_atomic_helper_suspend(adev_to_drm(adev)); - if (IS_ERR(adev->dm.cached_state)) - return PTR_ERR(adev->dm.cached_state); + int r = dm_cache_state(adev); + + if (r) + return r; } s3_handle_hdmi_cec(adev_to_drm(adev), true); @@ -3295,12 +3367,6 @@ static int dm_resume(struct amdgpu_ip_block *ip_block) struct amdgpu_dm_connector *aconnector; struct drm_connector *connector; struct drm_connector_list_iter iter; - struct drm_crtc *crtc; - struct drm_crtc_state *new_crtc_state; - struct dm_crtc_state *dm_new_crtc_state; - struct drm_plane *plane; - struct drm_plane_state *new_plane_state; - struct dm_plane_state *dm_new_plane_state; struct dm_atomic_state *dm_state = to_dm_atomic_state(dm->atomic_obj.state); enum dc_connection_type new_connection_type = dc_connection_none; struct dc_state *dc_state; @@ -3457,40 +3523,7 @@ static int dm_resume(struct amdgpu_ip_block *ip_block) } drm_connector_list_iter_end(&iter); - /* Force mode set in atomic commit */ - for_each_new_crtc_in_state(dm->cached_state, crtc, new_crtc_state, i) { - new_crtc_state->active_changed = true; - dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); - reset_freesync_config_for_crtc(dm_new_crtc_state); - } - - /* - * atomic_check is expected to create the dc states. We need to release - * them here, since they were duplicated as part of the suspend - * procedure. - */ - for_each_new_crtc_in_state(dm->cached_state, crtc, new_crtc_state, i) { - dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); - if (dm_new_crtc_state->stream) { - WARN_ON(kref_read(&dm_new_crtc_state->stream->refcount) > 1); - dc_stream_release(dm_new_crtc_state->stream); - dm_new_crtc_state->stream = NULL; - } - dm_new_crtc_state->base.color_mgmt_changed = true; - } - - for_each_new_plane_in_state(dm->cached_state, plane, new_plane_state, i) { - dm_new_plane_state = to_dm_plane_state(new_plane_state); - if (dm_new_plane_state->dc_state) { - WARN_ON(kref_read(&dm_new_plane_state->dc_state->refcount) > 1); - dc_plane_state_release(dm_new_plane_state->dc_state); - dm_new_plane_state->dc_state = NULL; - } - } - - drm_atomic_helper_resume(ddev, dm->cached_state); - - dm->cached_state = NULL; + dm_destroy_cached_state(adev); /* Do mst topology probing after resuming cached state*/ drm_connector_list_iter_begin(ddev, &iter); @@ -3539,6 +3572,7 @@ static const struct amd_ip_funcs amdgpu_dm_funcs = { .prepare_suspend = dm_prepare_suspend, .suspend = dm_suspend, .resume = dm_resume, + .complete = dm_complete, .is_idle = dm_is_idle, .wait_for_idle = dm_wait_for_idle, .check_soft_reset = dm_check_soft_reset, @@ -3610,13 +3644,15 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector) luminance_range = &conn_base->display_info.luminance_range; - if (luminance_range->max_luminance) { - caps->aux_min_input_signal = luminance_range->min_luminance; + if (luminance_range->max_luminance) caps->aux_max_input_signal = luminance_range->max_luminance; - } else { - caps->aux_min_input_signal = 0; + else caps->aux_max_input_signal = 512; - } + + if (luminance_range->min_luminance) + caps->aux_min_input_signal = luminance_range->min_luminance; + else + caps->aux_min_input_signal = 1; min_input_signal_override = drm_get_panel_min_brightness_quirk(aconnector->drm_edid); if (min_input_signal_override >= 0) @@ -4001,19 +4037,19 @@ static int register_hpd_handlers(struct amdgpu_device *adev) if (dc_is_dmub_outbox_supported(adev->dm.dc)) { if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_HPD, dmub_hpd_callback, true)) { - drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub hpd callback"); + drm_err(adev_to_drm(adev), "fail to register dmub hpd callback"); return -EINVAL; } if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_HPD_IRQ, dmub_hpd_callback, true)) { - drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub hpd callback"); + drm_err(adev_to_drm(adev), "fail to register dmub hpd callback"); return -EINVAL; } if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_HPD_SENSE_NOTIFY, dmub_hpd_sense_callback, true)) { - drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub hpd sense callback"); + drm_err(adev_to_drm(adev), "fail to register dmub hpd sense callback"); return -EINVAL; } } @@ -4718,9 +4754,23 @@ static int get_brightness_range(const struct amdgpu_dm_backlight_caps *caps, return 1; } +/* Rescale from [min..max] to [0..MAX_BACKLIGHT_LEVEL] */ +static inline u32 scale_input_to_fw(int min, int max, u64 input) +{ + return DIV_ROUND_CLOSEST_ULL(input * MAX_BACKLIGHT_LEVEL, max - min); +} + +/* Rescale from [0..MAX_BACKLIGHT_LEVEL] to [min..max] */ +static inline u32 scale_fw_to_input(int min, int max, u64 input) +{ + return min + DIV_ROUND_CLOSEST_ULL(input * (max - min), MAX_BACKLIGHT_LEVEL); +} + static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *caps, - uint32_t *brightness) + unsigned int min, unsigned int max, + uint32_t *user_brightness) { + u32 brightness = scale_input_to_fw(min, max, *user_brightness); u8 prev_signal = 0, prev_lum = 0; int i = 0; @@ -4731,7 +4781,7 @@ static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *cap return; /* choose start to run less interpolation steps */ - if (caps->luminance_data[caps->data_points/2].input_signal > *brightness) + if (caps->luminance_data[caps->data_points/2].input_signal > brightness) i = caps->data_points/2; do { u8 signal = caps->luminance_data[i].input_signal; @@ -4742,17 +4792,18 @@ static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *cap * brightness < signal: interpolate between previous and current luminance numerator * brightness > signal: find next data point */ - if (*brightness > signal) { + if (brightness > signal) { prev_signal = signal; prev_lum = lum; i++; continue; } - if (*brightness < signal) + if (brightness < signal) lum = prev_lum + DIV_ROUND_CLOSEST((lum - prev_lum) * - (*brightness - prev_signal), + (brightness - prev_signal), signal - prev_signal); - *brightness = DIV_ROUND_CLOSEST(lum * *brightness, 101); + *user_brightness = scale_fw_to_input(min, max, + DIV_ROUND_CLOSEST(lum * brightness, 101)); return; } while (i < caps->data_points); } @@ -4765,11 +4816,10 @@ static u32 convert_brightness_from_user(const struct amdgpu_dm_backlight_caps *c if (!get_brightness_range(caps, &min, &max)) return brightness; - convert_custom_brightness(caps, &brightness); + convert_custom_brightness(caps, min, max, &brightness); - // Rescale 0..255 to min..max - return min + DIV_ROUND_CLOSEST((max - min) * brightness, - AMDGPU_MAX_BL_LEVEL); + // Rescale 0..max to min..max + return min + DIV_ROUND_CLOSEST_ULL((u64)(max - min) * brightness, max); } static u32 convert_brightness_to_user(const struct amdgpu_dm_backlight_caps *caps, @@ -4782,8 +4832,8 @@ static u32 convert_brightness_to_user(const struct amdgpu_dm_backlight_caps *cap if (brightness < min) return 0; - // Rescale min..max to 0..255 - return DIV_ROUND_CLOSEST(AMDGPU_MAX_BL_LEVEL * (brightness - min), + // Rescale min..max to 0..max + return DIV_ROUND_CLOSEST_ULL((u64)max * (brightness - min), max - min); } @@ -4813,6 +4863,14 @@ static void amdgpu_dm_backlight_set_level(struct amdgpu_display_manager *dm, reallow_idle = true; } + if (trace_amdgpu_dm_brightness_enabled()) { + trace_amdgpu_dm_brightness(__builtin_return_address(0), + user_brightness, + brightness, + caps->aux_support, + power_supply_is_system_supplied() > 0); + } + if (caps->aux_support) { rc = dc_link_set_backlight_level_nits(link, true, brightness, AUX_BL_DEFAULT_TRANSITION_TIME_MS); @@ -4908,7 +4966,7 @@ amdgpu_dm_register_backlight_device(struct amdgpu_dm_connector *aconnector) struct drm_device *drm = aconnector->base.dev; struct amdgpu_display_manager *dm = &drm_to_adev(drm)->dm; struct backlight_properties props = { 0 }; - struct amdgpu_dm_backlight_caps caps = { 0 }; + struct amdgpu_dm_backlight_caps *caps; char bl_name[16]; int min, max; @@ -4922,22 +4980,21 @@ amdgpu_dm_register_backlight_device(struct amdgpu_dm_connector *aconnector) return; } - amdgpu_acpi_get_backlight_caps(&caps); - if (caps.caps_valid && get_brightness_range(&caps, &min, &max)) { + caps = &dm->backlight_caps[aconnector->bl_idx]; + if (get_brightness_range(caps, &min, &max)) { if (power_supply_is_system_supplied() > 0) - props.brightness = (max - min) * DIV_ROUND_CLOSEST(caps.ac_level, 100); + props.brightness = (max - min) * DIV_ROUND_CLOSEST(caps->ac_level, 100); else - props.brightness = (max - min) * DIV_ROUND_CLOSEST(caps.dc_level, 100); + props.brightness = (max - min) * DIV_ROUND_CLOSEST(caps->dc_level, 100); /* min is zero, so max needs to be adjusted */ props.max_brightness = max - min; drm_dbg(drm, "Backlight caps: min: %d, max: %d, ac %d, dc %d\n", min, max, - caps.ac_level, caps.dc_level); + caps->ac_level, caps->dc_level); } else - props.brightness = AMDGPU_MAX_BL_LEVEL; + props.brightness = props.max_brightness = MAX_BACKLIGHT_LEVEL; - if (caps.data_points && !(amdgpu_dc_debug_mask & DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE)) + if (caps->data_points && !(amdgpu_dc_debug_mask & DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE)) drm_info(drm, "Using custom brightness curve\n"); - props.max_brightness = AMDGPU_MAX_BL_LEVEL; props.type = BACKLIGHT_RAW; snprintf(bl_name, sizeof(bl_name), "amdgpu_bl%d", @@ -7519,7 +7576,7 @@ static enum dc_status dm_validate_stream_and_context(struct dc *dc, dc_result = DC_FAIL_ATTACH_SURFACES; if (dc_result == DC_OK) - dc_result = dc_validate_global_state(dc, dc_state, true); + dc_result = dc_validate_global_state(dc, dc_state, DC_VALIDATE_MODE_ONLY); cleanup: if (dc_state) @@ -7577,7 +7634,7 @@ create_validate_stream_for_sink(struct drm_connector *connector, dc_result = dm_validate_stream_and_context(adev->dm.dc, stream); if (dc_result != DC_OK) { - DRM_DEBUG_KMS("Mode %dx%d (clk %d) pixel_encoding:%s color_depth:%s failed validation -- %s\n", + DRM_DEBUG_KMS("Pruned mode %d x %d (clk %d) %s %s -- %s\n", drm_mode->hdisplay, drm_mode->vdisplay, drm_mode->clock, @@ -7844,6 +7901,22 @@ static int dm_encoder_helper_atomic_check(struct drm_encoder *encoder, int clock, bpp = 0; bool is_y420 = false; + if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) { + struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder); + struct drm_display_mode *native_mode = &amdgpu_encoder->native_mode; + enum drm_mode_status result; + + result = drm_crtc_helper_mode_valid_fixed(encoder->crtc, adjusted_mode, native_mode); + if (result != MODE_OK && dm_new_connector_state->scaling == RMX_OFF) { + drm_dbg_driver(encoder->dev, + "mode %dx%d@%dHz is not native, enabling scaling\n", + adjusted_mode->hdisplay, adjusted_mode->vdisplay, + drm_mode_vrefresh(adjusted_mode)); + dm_new_connector_state->scaling = RMX_FULL; + } + return 0; + } + if (!aconnector->mst_output_port) return 0; @@ -8301,7 +8374,7 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector) drm_add_modes_noedid(connector, 1920, 1080); } else { amdgpu_dm_connector_ddc_get_modes(connector, drm_edid); - if (encoder) + if (encoder && connector->connector_type != DRM_MODE_CONNECTOR_eDP) amdgpu_dm_connector_add_common_modes(encoder, connector); amdgpu_dm_connector_add_freesync_modes(connector, drm_edid); } @@ -12141,7 +12214,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, drm_dbg_atomic(dev, "MST drm_dp_mst_atomic_check() failed\n"); goto fail; } - status = dc_validate_global_state(dc, dm_state->context, true); + status = dc_validate_global_state(dc, dm_state->context, DC_VALIDATE_MODE_ONLY); if (status != DC_OK) { drm_dbg_atomic(dev, "DC global validation failure: %s (%d)", dc_status_to_str(status), status); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index d7d92f9911e4..b937da0a4e4a 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -636,8 +636,9 @@ struct amdgpu_display_manager { * @bb_from_dmub: * * Bounding box data read from dmub during early initialization for DCN4+ + * Data is stored as a byte array that should be casted to the appropriate bb struct */ - struct dml2_soc_bb *bb_from_dmub; + void *bb_from_dmub; /** * @oem_i2c: diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index d4395b92fb85..9e3e51a2dc49 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -1029,6 +1029,10 @@ enum dc_edid_status dm_helpers_read_local_edid( return EDID_NO_RESPONSE; edid = drm_edid_raw(drm_edid); // FIXME: Get rid of drm_edid_raw() + if (!edid || + edid->extensions >= sizeof(sink->dc_edid.raw_edid) / EDID_LENGTH) + return EDID_BAD_INPUT; + sink->dc_edid.length = EDID_LENGTH * (edid->extensions + 1); memmove(sink->dc_edid.raw_edid, (uint8_t *)edid, sink->dc_edid.length); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 25e8befbcc47..7187d5aedf0a 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -107,7 +107,7 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, if (payload.write && result >= 0) { if (result) { /*one byte indicating partially written bytes*/ - drm_dbg_dp(adev_to_drm(adev), "amdgpu: AUX partially written\n"); + drm_dbg_dp(adev_to_drm(adev), "AUX partially written\n"); result = payload.data[0]; } else if (!payload.reply[0]) /*I2C_ACK|AUX_ACK*/ @@ -133,11 +133,11 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, break; } - drm_dbg_dp(adev_to_drm(adev), "amdgpu: DP AUX transfer fail:%d\n", operation_result); + drm_dbg_dp(adev_to_drm(adev), "DP AUX transfer fail:%d\n", operation_result); } if (payload.reply[0]) - drm_dbg_dp(adev_to_drm(adev), "amdgpu: AUX reply command not ACK: 0x%02x.", + drm_dbg_dp(adev_to_drm(adev), "AUX reply command not ACK: 0x%02x.", payload.reply[0]); return result; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h index 4686d4b0cbad..95f890fda8aa 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h @@ -726,6 +726,32 @@ TRACE_EVENT(dcn_optc_lock_unlock_state, ) ); +TRACE_EVENT(amdgpu_dm_brightness, + TP_PROTO(void *function, u32 user_brightness, u32 converted_brightness, bool aux, bool ac), + TP_ARGS(function, user_brightness, converted_brightness, aux, ac), + TP_STRUCT__entry( + __field(void *, function) + __field(u32, user_brightness) + __field(u32, converted_brightness) + __field(bool, aux) + __field(bool, ac) + ), + TP_fast_assign( + __entry->function = function; + __entry->user_brightness = user_brightness; + __entry->converted_brightness = converted_brightness; + __entry->aux = aux; + __entry->ac = ac; + ), + TP_printk("%ps: brightness requested=%u converted=%u aux=%s power=%s", + (void *)__entry->function, + (u32)__entry->user_brightness, + (u32)__entry->converted_brightness, + (__entry->aux) ? "true" : "false", + (__entry->ac) ? "AC" : "DC" + ) +); + #endif /* _AMDGPU_DM_TRACE_H_ */ #undef TRACE_INCLUDE_PATH diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c index 2c645dffec18..f2b1720a6a66 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c @@ -396,6 +396,7 @@ static enum bp_result transmitter_control_v1_7( process_phy_transition_init_params.display_port_link_rate = link->cur_link_settings.link_rate; process_phy_transition_init_params.transition_bitmask = link->phy_transition_bitmask; } + dig_v1_7.skip_phy_ssc_reduction = link->wa_flags.skip_phy_ssc_reduction; } // Handle PRE_OFF_TO_ON: Process ACPI PHY Transition Interlock diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c index a3b8e3d4a429..514a5efda102 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c @@ -22,8 +22,6 @@ #include "dcn/dcn_4_1_0_offset.h" #include "dcn/dcn_4_1_0_sh_mask.h" -#include "dml/dcn401/dcn401_fpu.h" - #define DCN_BASE__INST0_SEG1 0x000000C0 #define mmCLK01_CLK0_CLK_PLL_REQ 0x16E37 @@ -183,43 +181,36 @@ static void dcn401_init_single_clock(struct clk_mgr_internal *clk_mgr, PPCLK_e c static void dcn401_build_wm_range_table(struct clk_mgr *clk_mgr) { - /* legacy */ - DC_FP_START(); - dcn401_build_wm_range_table_fpu(clk_mgr); - DC_FP_END(); - - if (clk_mgr->ctx->dc->debug.using_dml21) { - /* For min clocks use as reported by PM FW and report those as min */ - uint16_t min_uclk_mhz = clk_mgr->bw_params->clk_table.entries[0].memclk_mhz; - uint16_t min_dcfclk_mhz = clk_mgr->bw_params->clk_table.entries[0].dcfclk_mhz; - - /* Set A - Normal - default values */ - clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.wm_type = WATERMARKS_CLOCK_RANGE; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_uclk = min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_uclk = 0xFFFF; - - /* Set B - Unused on dcn4 */ - clk_mgr->bw_params->wm_table.nv_entries[WM_B].valid = false; - - /* Set 1A - Dummy P-State - P-State latency set to "dummy p-state" value */ - /* 'DalDummyClockChangeLatencyNs' registry key option set to 0x7FFFFFFF can be used to disable Set C for dummy p-state */ - if (clk_mgr->ctx->dc->bb_overrides.dummy_clock_change_latency_ns != 0x7FFFFFFF) { - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.wm_type = WATERMARKS_DUMMY_PSTATE; - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.min_uclk = min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.max_uclk = 0xFFFF; - } else { - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].valid = false; - } - - /* Set 1B - Unused on dcn4 */ - clk_mgr->bw_params->wm_table.nv_entries[WM_1B].valid = false; + /* For min clocks use as reported by PM FW and report those as min */ + uint16_t min_uclk_mhz = clk_mgr->bw_params->clk_table.entries[0].memclk_mhz; + uint16_t min_dcfclk_mhz = clk_mgr->bw_params->clk_table.entries[0].dcfclk_mhz; + + /* Set A - Normal - default values */ + clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid = true; + clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.wm_type = WATERMARKS_CLOCK_RANGE; + clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; + clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_dcfclk = 0xFFFF; + clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_uclk = min_uclk_mhz; + clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_uclk = 0xFFFF; + + /* Set B - Unused on dcn4 */ + clk_mgr->bw_params->wm_table.nv_entries[WM_B].valid = false; + + /* Set 1A - Dummy P-State - P-State latency set to "dummy p-state" value */ + /* 'DalDummyClockChangeLatencyNs' registry key option set to 0x7FFFFFFF can be used to disable Set C for dummy p-state */ + if (clk_mgr->ctx->dc->bb_overrides.dummy_clock_change_latency_ns != 0x7FFFFFFF) { + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].valid = true; + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.wm_type = WATERMARKS_DUMMY_PSTATE; + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.max_dcfclk = 0xFFFF; + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.min_uclk = min_uclk_mhz; + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.max_uclk = 0xFFFF; + } else { + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].valid = false; } + + /* Set 1B - Unused on dcn4 */ + clk_mgr->bw_params->wm_table.nv_entries[WM_1B].valid = false; } void dcn401_init_clocks(struct clk_mgr *clk_mgr_base) @@ -320,6 +311,25 @@ void dcn401_init_clocks(struct clk_mgr *clk_mgr_base) dcn401_build_wm_range_table(clk_mgr_base); } +bool dcn401_is_dc_mode_present(struct clk_mgr *clk_mgr_base) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + + return clk_mgr->smu_present && clk_mgr->dpm_present && + ((clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_dcfclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz) || + (clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_dispclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz) || + (clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_dtbclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz) || + (clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_fclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz) || + (clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_memclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz) || + (clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_socclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz)); +} + static void dcn401_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass, struct clk_mgr *clk_mgr_base, struct clk_log_info *log_info) { @@ -1505,6 +1515,7 @@ static struct clk_mgr_funcs dcn401_funcs = { .get_dispclk_from_dentist = dcn401_get_dispclk_from_dentist, .get_hard_min_memclk = dcn401_get_hard_min_memclk, .get_hard_min_fclk = dcn401_get_hard_min_fclk, + .is_dc_mode_present = dcn401_is_dc_mode_present, }; struct clk_mgr_internal *dcn401_clk_mgr_construct( diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.h index 6c9ae5ca2c7e..616e964df96d 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.h @@ -105,6 +105,7 @@ struct dcn401_clk_mgr { }; void dcn401_init_clocks(struct clk_mgr *clk_mgr_base); +bool dcn401_is_dc_mode_present(struct clk_mgr *clk_mgr_base); struct clk_mgr_internal *dcn401_clk_mgr_construct(struct dc_context *ctx, struct dccg *dccg); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 56d011a1323c..c744aa9d830f 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -241,6 +241,7 @@ static bool create_links( DC_LOG_DC("BIOS object table - end"); /* Create a link for each usb4 dpia port */ + dc->lowest_dpia_link_index = MAX_LINKS; for (i = 0; i < dc->res_pool->usb4_dpia_count; i++) { struct link_init_data link_init_params = {0}; struct dc_link *link; @@ -253,6 +254,9 @@ static bool create_links( link = dc->link_srv->create_link(&link_init_params); if (link) { + if (dc->lowest_dpia_link_index > dc->link_count) + dc->lowest_dpia_link_index = dc->link_count; + dc->links[dc->link_count] = link; link->dc = dc; ++dc->link_count; @@ -2377,7 +2381,7 @@ enum dc_status dc_commit_streams(struct dc *dc, struct dc_commit_streams_params context->power_source = params->power_source; - res = dc_validate_with_context(dc, set, params->stream_count, context, false); + res = dc_validate_with_context(dc, set, params->stream_count, context, DC_VALIDATE_MODE_AND_PROGRAMMING); /* * Only update link encoder to stream assignment after bandwidth validation passed. @@ -3300,7 +3304,8 @@ static void copy_stream_update_to_stream(struct dc *dc, if (dsc_validate_context) { stream->timing.dsc_cfg = *update->dsc_config; stream->timing.flags.DSC = enable_dsc; - if (dc->res_pool->funcs->validate_bandwidth(dc, dsc_validate_context, true) != DC_OK) { + if (dc->res_pool->funcs->validate_bandwidth(dc, dsc_validate_context, + DC_VALIDATE_MODE_ONLY) != DC_OK) { stream->timing.dsc_cfg = old_dsc_cfg; stream->timing.flags.DSC = old_dsc_enabled; update->dsc_config = NULL; @@ -3522,7 +3527,7 @@ static bool update_planes_and_stream_state(struct dc *dc, } if (update_type == UPDATE_TYPE_FULL) { - if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK) { + if (dc->res_pool->funcs->validate_bandwidth(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING) != DC_OK) { BREAK_TO_DEBUGGER(); goto fail; } @@ -4628,7 +4633,8 @@ static struct dc_state *create_minimal_transition_state(struct dc *dc, backup_and_set_minimal_pipe_split_policy(dc, base_context, policy); /* commit minimal state */ - if (dc->res_pool->funcs->validate_bandwidth(dc, minimal_transition_context, false) == DC_OK) { + if (dc->res_pool->funcs->validate_bandwidth(dc, minimal_transition_context, + DC_VALIDATE_MODE_AND_PROGRAMMING) == DC_OK) { /* prevent underflow and corruption when reconfiguring pipes */ force_vsync_flip_in_minimal_transition_context(minimal_transition_context); } else { @@ -5151,7 +5157,7 @@ static bool update_planes_and_stream_v1(struct dc *dc, copy_stream_update_to_stream(dc, context, stream, stream_update); if (update_type >= UPDATE_TYPE_FULL) { - if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK) { + if (dc->res_pool->funcs->validate_bandwidth(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING) != DC_OK) { DC_ERROR("Mode validation failed for stream update!\n"); dc_state_release(context); return false; @@ -6337,13 +6343,14 @@ void dc_set_edp_power(const struct dc *dc, struct dc_link *edp_link, edp_link->dc->link_srv->edp_set_panel_power(edp_link, powerOn); } -/* - ***************************************************************************** +/** * dc_get_power_profile_for_dc_state() - extracts power profile from dc state * * Called when DM wants to make power policy decisions based on dc_state * - ***************************************************************************** + * @context: Pointer to the dc_state from which the power profile is extracted. + * + * Return: The power profile structure containing the power level information. */ struct dc_power_profile dc_get_power_profile_for_dc_state(const struct dc_state *context) { @@ -6359,13 +6366,14 @@ struct dc_power_profile dc_get_power_profile_for_dc_state(const struct dc_state return profile; } -/* - ********************************************************************************** +/** * dc_get_det_buffer_size_from_state() - extracts detile buffer size from dc state * - * Called when DM wants to log detile buffer size from dc_state + * This function is called to log the detile buffer size from the dc_state. + * + * @context: a pointer to the dc_state from which the detile buffer size is extracted. * - ********************************************************************************** + * Return: the size of the detile buffer, or 0 if not available. */ unsigned int dc_get_det_buffer_size_from_state(const struct dc_state *context) { @@ -6377,6 +6385,36 @@ unsigned int dc_get_det_buffer_size_from_state(const struct dc_state *context) return 0; } +/** + * dc_get_host_router_index: Get index of host router from a dpia link + * + * This function return a host router index of the target link. If the target link is dpia link. + * + * @link: Pointer to the target link (input) + * @host_router_index: Pointer to store the host router index of the target link (output). + * + * Return: true if the host router index is found and valid. + * + */ +bool dc_get_host_router_index(const struct dc_link *link, unsigned int *host_router_index) +{ + struct dc *dc; + + if (!link || !host_router_index || link->ep_type != DISPLAY_ENDPOINT_USB4_DPIA) + return false; + + dc = link->ctx->dc; + + if (link->link_index < dc->lowest_dpia_link_index) + return false; + + *host_router_index = (link->link_index - dc->lowest_dpia_link_index) / dc->caps.num_of_dpias_per_host_router; + if (*host_router_index < dc->caps.num_of_host_routers) + return true; + else + return false; +} + bool dc_is_cursor_limit_pending(struct dc *dc) { uint32_t i; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c index 7551d0a3fe82..bbce751b485f 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c @@ -268,6 +268,8 @@ char *dc_status_to_str(enum dc_status status) return "Insufficient DP link bandwidth"; case DC_FAIL_HW_CURSOR_SUPPORT: return "HW Cursor not supported"; + case DC_FAIL_DP_TUNNEL_BW_VALIDATE: + return "Fail DP Tunnel BW validation"; case DC_ERROR_UNEXPECTED: return "Unexpected error"; } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c index 71e15da4bb69..130455f2802a 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c @@ -515,7 +515,8 @@ void dc_link_enable_hpd_filter(struct dc_link *link, bool enable) link->dc->link_srv->enable_hpd_filter(link, enable); } -bool dc_link_dp_dpia_validate(struct dc *dc, const struct dc_stream_state *streams, const unsigned int count) +enum dc_status dc_link_validate_dp_tunneling_bandwidth(const struct dc *dc, const struct dc_state *new_ctx) { - return dc->link_srv->validate_dpia_bandwidth(streams, count); + return dc->link_srv->validate_dp_tunnel_bandwidth(dc, new_ctx); } + diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 3da25bd8b578..854fc51f159c 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -4053,7 +4053,7 @@ static bool add_all_planes_for_stream( * @set: An array of dc_validation_set with all the current streams reference * @set_count: Total of streams * @context: New context - * @fast_validate: Enable or disable fast validation + * @validate_mode: identify the validation mode * * This function updates the potential new stream in the context object. It * creates multiple lists for the add, remove, and unchanged streams. In @@ -4068,7 +4068,7 @@ enum dc_status dc_validate_with_context(struct dc *dc, const struct dc_validation_set set[], int set_count, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { struct dc_stream_state *unchanged_streams[MAX_PIPES] = { 0 }; struct dc_stream_state *del_streams[MAX_PIPES] = { 0 }; @@ -4242,7 +4242,7 @@ enum dc_status dc_validate_with_context(struct dc *dc, dc_state_set_stream_subvp_cursor_limit(context->streams[i], context, false); } - res = dc_validate_global_state(dc, context, fast_validate); + res = dc_validate_global_state(dc, context, validate_mode); /* calculate pixel rate divider after deciding pxiel clock & odm combine */ if ((dc->hwss.calculate_pix_rate_divider) && (res == DC_OK)) { @@ -4299,7 +4299,7 @@ static void decide_hblank_borrow(struct pipe_ctx *pipe_ctx) * * @dc: dc struct for this driver * @new_ctx: state to be validated - * @fast_validate: set to true if only yes/no to support matters + * @validate_mode: identify the validation mode * * Checks hardware resource availability and bandwidth requirement. * @@ -4309,7 +4309,7 @@ static void decide_hblank_borrow(struct pipe_ctx *pipe_ctx) enum dc_status dc_validate_global_state( struct dc *dc, struct dc_state *new_ctx, - bool fast_validate) + enum dc_validate_mode validate_mode) { enum dc_status result = DC_ERROR_UNEXPECTED; int i, j; @@ -4368,7 +4368,7 @@ enum dc_status dc_validate_global_state( result = resource_build_scaling_params_for_context(dc, new_ctx); if (result == DC_OK) - result = dc->res_pool->funcs->validate_bandwidth(dc, new_ctx, fast_validate); + result = dc->res_pool->funcs->validate_bandwidth(dc, new_ctx, validate_mode); return result; } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_state.c b/drivers/gpu/drm/amd/display/dc/core/dc_state.c index 4db7383720fd..47712a4aec55 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_state.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_state.c @@ -194,11 +194,6 @@ static void init_state(struct dc *dc, struct dc_state *state) struct dc_state *dc_state_create(struct dc *dc, struct dc_state_create_params *params) { struct dc_state *state; -#ifdef CONFIG_DRM_AMD_DC_FP - struct dml2_configuration_options *dml2_opt = &dc->dml2_tmp; - - memcpy(dml2_opt, &dc->dml2_options, sizeof(dc->dml2_options)); -#endif state = kvzalloc(sizeof(struct dc_state), GFP_KERNEL); @@ -211,14 +206,12 @@ struct dc_state *dc_state_create(struct dc *dc, struct dc_state_create_params *p #ifdef CONFIG_DRM_AMD_DC_FP if (dc->debug.using_dml2) { - dml2_opt->use_clock_dc_limits = false; - if (!dml2_create(dc, dml2_opt, &state->bw_ctx.dml2)) { + if (!dml2_create(dc, &dc->dml2_options, &state->bw_ctx.dml2)) { dc_state_release(state); return NULL; } - dml2_opt->use_clock_dc_limits = true; - if (!dml2_create(dc, dml2_opt, &state->bw_ctx.dml2_dc_power_source)) { + if (!dml2_create(dc, &dc->dml2_dc_power_options, &state->bw_ctx.dml2_dc_power_source)) { dc_state_release(state); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 1d917be36fc4..ce1957c7862f 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -46,6 +46,8 @@ #include "dmub/inc/dmub_cmd.h" +#include "sspl/dc_spl_types.h" + struct abm_save_restore; /* forward declaration */ @@ -53,7 +55,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.334" +#define DC_VER "3.2.340" /** * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC @@ -66,7 +68,11 @@ struct dmub_notification; #define MAX_STREAMS 6 #define MIN_VIEWPORT_SIZE 12 #define MAX_NUM_EDP 2 -#define MAX_HOST_ROUTERS_NUM 2 +#define MAX_SUPPORTED_FORMATS 7 + +#define MAX_HOST_ROUTERS_NUM 3 +#define MAX_DPIA_PER_HOST_ROUTER 3 +#define MAX_DPIA_NUM (MAX_HOST_ROUTERS_NUM * MAX_DPIA_PER_HOST_ROUTER) /* Display Core Interfaces */ struct dc_versions { @@ -192,6 +198,34 @@ struct dpp_color_caps { struct rom_curve_caps ogam_rom_caps; }; +/* Below structure is to describe the HW support for mem layout, extend support + range to match what OS could handle in the roadmap */ +struct lut3d_caps { + uint32_t dma_3d_lut : 1; /*< DMA mode support for 3D LUT */ + struct { + uint32_t swizzle_3d_rgb : 1; + uint32_t swizzle_3d_bgr : 1; + uint32_t linear_1d : 1; + } mem_layout_support; + struct { + uint32_t unorm_12msb : 1; + uint32_t unorm_12lsb : 1; + uint32_t float_fp1_5_10 : 1; + } mem_format_support; + struct { + uint32_t order_rgba : 1; + uint32_t order_bgra : 1; + } mem_pixel_order_support; + /*< size options are 9, 17, 33, 45, 65 */ + struct { + uint32_t dim_9 : 1; /* 3D LUT support for 9x9x9 */ + uint32_t dim_17 : 1; /* 3D LUT support for 17x17x17 */ + uint32_t dim_33 : 1; /* 3D LUT support for 33x33x33 */ + uint32_t dim_45 : 1; /* 3D LUT support for 45x45x45 */ + uint32_t dim_65 : 1; /* 3D LUT support for 65x65x65 */ + } lut_dim_caps; +}; + /** * struct mpc_color_caps - color pipeline capabilities for multiple pipe and * plane combined blocks @@ -203,6 +237,9 @@ struct dpp_color_caps { * @shared_3d_lut: shared 3D LUT flag. Can be either DPP or MPC, but single * instance * @ogam_rom_caps: pre-definied curve caps for regamma 1D LUT + * @mcm_3d_lut_caps: HW support cap for MCM LUT memory + * @rmcm_3d_lut_caps: HW support cap for RMCM LUT memory + * @preblend: whether color manager supports preblend with MPC */ struct mpc_color_caps { uint16_t gamut_remap : 1; @@ -211,6 +248,9 @@ struct mpc_color_caps { uint16_t num_3dluts : 3; uint16_t shared_3d_lut:1; struct rom_curve_caps ogam_rom_caps; + struct lut3d_caps mcm_3d_lut_caps; + struct lut3d_caps rmcm_3d_lut_caps; + bool preblend; }; /** @@ -305,6 +345,8 @@ struct dc_caps { /* Conservative limit for DCC cases which require ODM4:1 to support*/ uint32_t dcc_plane_width_limit; struct dc_scl_caps scl_caps; + uint8_t num_of_host_routers; + uint8_t num_of_dpias_per_host_router; }; struct dc_bug_wa { @@ -481,6 +523,8 @@ struct dc_config { bool set_pipe_unlock_order; bool enable_dpia_pre_training; bool unify_link_enc_assignment; + struct spl_sharpness_range dcn_sharpness_range; + struct spl_sharpness_range dcn_override_sharpness_range; }; enum visual_confirm { @@ -786,10 +830,8 @@ union dpia_debug_options { uint32_t disable_mst_dsc_work_around:1; /* bit 3 */ uint32_t enable_force_tbt3_work_around:1; /* bit 4 */ uint32_t disable_usb4_pm_support:1; /* bit 5 */ - uint32_t enable_consolidated_dpia_dp_lt:1; /* bit 6 */ - uint32_t enable_dpia_pre_training:1; /* bit 7 */ - uint32_t unify_link_enc_assignment:1; /* bit 8 */ - uint32_t reserved:24; + uint32_t enable_usb4_bw_zero_alloc_patch:1; /* bit 6 */ + uint32_t reserved:25; } bits; uint32_t raw; }; @@ -1151,7 +1193,7 @@ struct dc_init_data { uint32_t *dcn_reg_offsets; uint32_t *nbio_reg_offsets; uint32_t *clk_reg_offsets; - struct dml2_soc_bb *bb_from_dmub; + void *bb_from_dmub; }; struct dc_callback_init { @@ -1389,6 +1431,8 @@ struct dc_plane_state { int sharpness_level; enum linear_light_scaling linear_light_scaling; unsigned int sdr_white_level_nits; + struct spl_sharpness_range sharpness_range; + enum sharpness_range_source sharpness_source; }; struct dc_plane_info { @@ -1570,6 +1614,7 @@ struct dc_scratch_space { bool blank_stream_on_ocs_change; bool read_dpcd204h_on_irq_hpd; bool force_dp_ffe_preset; + bool skip_phy_ssc_reduction; } wa_flags; union dc_dp_ffe_preset forced_dp_ffe_preset; struct link_mst_stream_allocation_table mst_stream_alloc_table; @@ -1603,6 +1648,7 @@ struct dc { uint8_t link_count; struct dc_link *links[MAX_LINKS]; + uint8_t lowest_dpia_link_index; struct link_service *link_srv; struct dc_state *current_state; @@ -1662,7 +1708,7 @@ struct dc { } scratch; struct dml2_configuration_options dml2_options; - struct dml2_configuration_options dml2_tmp; + struct dml2_configuration_options dml2_dc_power_options; enum dc_acpi_cm_power_state power_state; }; @@ -1767,19 +1813,15 @@ enum dc_status dc_validate_with_context(struct dc *dc, const struct dc_validation_set set[], int set_count, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); bool dc_set_generic_gpio_for_stereo(bool enable, struct gpio_service *gpio_service); -/* - * fast_validate: we return after determining if we can support the new state, - * but before we populate the programming info - */ enum dc_status dc_validate_global_state( struct dc *dc, struct dc_state *new_ctx, - bool fast_validate); + enum dc_validate_mode validate_mode); bool dc_acquire_release_mpc_3dlut( struct dc *dc, bool acquire, @@ -2375,17 +2417,12 @@ void dc_link_dp_dpia_handle_usb4_bandwidth_allocation_for_link( struct dc_link *link, int peak_bw); /* - * Validate the BW of all the valid DPIA links to make sure it doesn't exceed - * available BW for each host router - * - * @dc: pointer to dc struct - * @stream: pointer to all possible streams - * @count: number of valid DPIA streams + * Calculates the DP tunneling bandwidth required for the stream timing + * and aggregates the stream bandwidth for the respective DP tunneling link * - * return: TRUE if bw used by DPIAs doesn't exceed available BW else return FALSE + * return: dc_status */ -bool dc_link_dp_dpia_validate(struct dc *dc, const struct dc_stream_state *streams, - const unsigned int count); +enum dc_status dc_link_validate_dp_tunneling_bandwidth(const struct dc *dc, const struct dc_state *new_ctx); /* Sink Interfaces - A sink corresponds to a display output device */ @@ -2595,6 +2632,8 @@ struct dc_power_profile dc_get_power_profile_for_dc_state(const struct dc_state unsigned int dc_get_det_buffer_size_from_state(const struct dc_state *context); +bool dc_get_host_router_index(const struct dc_link *link, unsigned int *host_router_index); + /* DSC Interfaces */ #include "dc_dsc.h" diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c index afbcf866520e..0a47d1a3515b 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -1656,7 +1656,7 @@ bool dc_wake_and_execute_gpint(const struct dc_context *ctx, enum dmub_gpint_com return result; } -void dc_dmub_srv_fams2_update_config(struct dc *dc, +static void dc_dmub_srv_rb_based_fams2_update_config(struct dc *dc, struct dc_state *context, bool enable) { @@ -1722,6 +1722,63 @@ void dc_dmub_srv_fams2_update_config(struct dc *dc, dm_execute_dmub_cmd_list(dc->ctx, num_cmds, cmd, DM_DMUB_WAIT_TYPE_WAIT); } +static void dc_dmub_srv_ib_based_fams2_update_config(struct dc *dc, + struct dc_state *context, + bool enable) +{ + struct dmub_fams2_config_v2 *config = (struct dmub_fams2_config_v2 *)dc->ctx->dmub_srv->dmub->ib_mem_gart.cpu_addr; + union dmub_rb_cmd cmd; + uint32_t i; + + memset(config, 0, sizeof(*config)); + memset(&cmd, 0, sizeof(cmd)); + + cmd.ib_fams2_config.header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH; + cmd.ib_fams2_config.header.sub_type = DMUB_CMD__FAMS2_IB_CONFIG; + + cmd.ib_fams2_config.ib_data.src.quad_part = dc->ctx->dmub_srv->dmub->ib_mem_gart.gpu_addr; + cmd.ib_fams2_config.ib_data.size = sizeof(*config); + + if (enable && context->bw_ctx.bw.dcn.fams2_global_config.features.bits.enable) { + /* copy static feature configuration overrides */ + config->global.features.bits.enable_stall_recovery = dc->debug.fams2_config.bits.enable_stall_recovery; + config->global.features.bits.enable_offload_flip = dc->debug.fams2_config.bits.enable_offload_flip; + config->global.features.bits.enable_debug = dc->debug.fams2_config.bits.enable_debug; + + /* send global configuration parameters */ + memcpy(&config->global, &context->bw_ctx.bw.dcn.fams2_global_config, + sizeof(struct dmub_cmd_fams2_global_config)); + + /* construct per-stream configs */ + for (i = 0; i < context->bw_ctx.bw.dcn.fams2_global_config.num_streams; i++) { + /* copy stream static base state */ + memcpy(&config->stream_v1[i].base, + &context->bw_ctx.bw.dcn.fams2_stream_base_params[i], + sizeof(config->stream_v1[i].base)); + + /* copy stream static sub-state */ + memcpy(&config->stream_v1[i].sub_state, + &context->bw_ctx.bw.dcn.fams2_stream_sub_params_v2[i], + sizeof(config->stream_v1[i].sub_state)); + } + } + + config->global.features.bits.enable_visual_confirm = dc->debug.visual_confirm == VISUAL_CONFIRM_FAMS2; + config->global.features.bits.enable = enable; + + dm_execute_dmub_cmd_list(dc->ctx, 1, &cmd, DM_DMUB_WAIT_TYPE_WAIT); +} + +void dc_dmub_srv_fams2_update_config(struct dc *dc, + struct dc_state *context, + bool enable) +{ + if (dc->debug.fams_version.major == 2) + dc_dmub_srv_rb_based_fams2_update_config(dc, context, enable); + if (dc->debug.fams_version.major == 3) + dc_dmub_srv_ib_based_fams2_update_config(dc, context, enable); +} + void dc_dmub_srv_fams2_drr_update(struct dc *dc, uint32_t tg_inst, uint32_t vtotal_min, @@ -1847,83 +1904,250 @@ void dc_dmub_srv_fams2_passthrough_flip( } } -bool dc_dmub_srv_ips_residency_cntl(struct dc_dmub_srv *dc_dmub_srv, bool start_measurement) + +bool dc_dmub_srv_ips_residency_cntl(const struct dc_context *ctx, uint8_t panel_inst, bool start_measurement) { - bool result; + union dmub_rb_cmd cmd; - if (!dc_dmub_srv || !dc_dmub_srv->dmub) + memset(&cmd, 0, sizeof(cmd)); + + cmd.ips_residency_cntl.header.type = DMUB_CMD__IPS; + cmd.ips_residency_cntl.header.sub_type = DMUB_CMD__IPS_RESIDENCY_CNTL; + cmd.ips_residency_cntl.header.payload_bytes = sizeof(struct dmub_cmd_ips_residency_cntl_data); + + // only panel_inst=0 is supported at the moment + cmd.ips_residency_cntl.cntl_data.panel_inst = panel_inst; + cmd.ips_residency_cntl.cntl_data.start_measurement = start_measurement; + + if (!dc_wake_and_execute_dmub_cmd(ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) return false; - result = dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__IPS_RESIDENCY, - start_measurement, NULL, DM_DMUB_WAIT_TYPE_WAIT); + return true; +} + +bool dc_dmub_srv_ips_query_residency_info(const struct dc_context *ctx, uint8_t panel_inst, struct dmub_ips_residency_info *driver_info, + enum ips_residency_mode ips_mode) +{ + union dmub_rb_cmd cmd; + uint32_t bytes = sizeof(struct dmub_ips_residency_info); + + dmub_flush_buffer_mem(&ctx->dmub_srv->dmub->scratch_mem_fb); + memset(&cmd, 0, sizeof(cmd)); + + cmd.ips_query_residency_info.header.type = DMUB_CMD__IPS; + cmd.ips_query_residency_info.header.sub_type = DMUB_CMD__IPS_QUERY_RESIDENCY_INFO; + cmd.ips_query_residency_info.header.payload_bytes = sizeof(struct dmub_cmd_ips_query_residency_info_data); + + cmd.ips_query_residency_info.info_data.dest.quad_part = ctx->dmub_srv->dmub->scratch_mem_fb.gpu_addr; + cmd.ips_query_residency_info.info_data.size = bytes; + cmd.ips_query_residency_info.info_data.panel_inst = panel_inst; + cmd.ips_query_residency_info.info_data.ips_mode = (uint32_t)ips_mode; + + if (!dc_wake_and_execute_dmub_cmd(ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) || + cmd.ips_query_residency_info.header.ret_status == 0) + return false; + + // copy the result to the output since ret_status != 0 means the command returned data + memcpy(driver_info, ctx->dmub_srv->dmub->scratch_mem_fb.cpu_addr, bytes); + + return true; +} + +bool dmub_lsdma_init(struct dc_dmub_srv *dc_dmub_srv) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_INIT_CONFIG; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; + + lsdma_data->u.init_data.gpu_addr_base.quad_part = dc_ctx->dmub_srv->dmub->lsdma_rb_fb.gpu_addr; + lsdma_data->u.init_data.ring_size = dc_ctx->dmub_srv->dmub->lsdma_rb_fb.size; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA Init failed in DMUB"); return result; } -void dc_dmub_srv_ips_query_residency_info(struct dc_dmub_srv *dc_dmub_srv, struct ips_residency_info *output) +bool dmub_lsdma_send_linear_copy_packet( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t src_addr, + uint64_t dst_addr, + uint32_t count) { - uint32_t i; - enum dmub_gpint_command command_code; + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; - if (!dc_dmub_srv || !dc_dmub_srv->dmub) - return; + memset(&cmd, 0, sizeof(cmd)); - switch (output->ips_mode) { - case DMUB_IPS_MODE_IPS1_MAX: - command_code = DMUB_GPINT__GET_IPS1_HISTOGRAM_COUNTER; - break; - case DMUB_IPS_MODE_IPS2: - command_code = DMUB_GPINT__GET_IPS2_HISTOGRAM_COUNTER; - break; - case DMUB_IPS_MODE_IPS1_RCG: - command_code = DMUB_GPINT__GET_IPS1_RCG_HISTOGRAM_COUNTER; - break; - case DMUB_IPS_MODE_IPS1_ONO2_ON: - command_code = DMUB_GPINT__GET_IPS1_ONO2_ON_HISTOGRAM_COUNTER; - break; - default: - command_code = DMUB_GPINT__INVALID_COMMAND; - break; - } + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_LINEAR_COPY; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; - if (command_code == DMUB_GPINT__INVALID_COMMAND) - return; + lsdma_data->u.linear_copy_data.count = count - 1; // LSDMA controller expects bytes to copy -1 + lsdma_data->u.linear_copy_data.src_lo = src_addr & 0xFFFFFFFF; + lsdma_data->u.linear_copy_data.src_hi = (src_addr >> 32) & 0xFFFFFFFF; + lsdma_data->u.linear_copy_data.dst_lo = dst_addr & 0xFFFFFFFF; + lsdma_data->u.linear_copy_data.dst_hi = (dst_addr >> 32) & 0xFFFFFFFF; - for (i = 0; i < GPINT_RETRY_NUM; i++) { - // false could mean GPINT timeout, in which case we should retry - if (dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_PERCENT, - (uint16_t)(output->ips_mode), &output->residency_percent, - DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - break; - udelay(100); - } + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA Linear Copy failed in DMUB"); + + return result; +} + +bool dmub_lsdma_send_tiled_to_tiled_copy_command( + struct dc_dmub_srv *dc_dmub_srv, + struct lsdma_send_tiled_to_tiled_copy_command_params params) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_TILED_TO_TILED_COPY; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; + + lsdma_data->u.tiled_copy_data.src_addr_lo = params.src_addr & 0xFFFFFFFF; + lsdma_data->u.tiled_copy_data.src_addr_hi = (params.src_addr >> 32) & 0xFFFFFFFF; + lsdma_data->u.tiled_copy_data.dst_addr_lo = params.dst_addr & 0xFFFFFFFF; + lsdma_data->u.tiled_copy_data.dst_addr_hi = (params.dst_addr >> 32) & 0xFFFFFFFF; + lsdma_data->u.tiled_copy_data.src_x = params.src_x; + lsdma_data->u.tiled_copy_data.src_y = params.src_y; + lsdma_data->u.tiled_copy_data.dst_x = params.dst_x; + lsdma_data->u.tiled_copy_data.dst_y = params.dst_y; + lsdma_data->u.tiled_copy_data.src_width = params.src_width - 1; // LSDMA controller expects width -1 + lsdma_data->u.tiled_copy_data.dst_width = params.dst_width - 1; // LSDMA controller expects width -1 + lsdma_data->u.tiled_copy_data.src_swizzle_mode = params.swizzle_mode; + lsdma_data->u.tiled_copy_data.dst_swizzle_mode = params.swizzle_mode; + lsdma_data->u.tiled_copy_data.src_element_size = params.element_size; + lsdma_data->u.tiled_copy_data.dst_element_size = params.element_size; + lsdma_data->u.tiled_copy_data.rect_x = params.rect_x; + lsdma_data->u.tiled_copy_data.rect_y = params.rect_y; + lsdma_data->u.tiled_copy_data.dcc = params.dcc; + lsdma_data->u.tiled_copy_data.tmz = params.tmz; + lsdma_data->u.tiled_copy_data.read_compress = params.read_compress; + lsdma_data->u.tiled_copy_data.write_compress = params.write_compress; + lsdma_data->u.tiled_copy_data.src_height = params.src_height - 1; // LSDMA controller expects height -1 + lsdma_data->u.tiled_copy_data.dst_height = params.dst_height - 1; // LSDMA controller expects height -1 + lsdma_data->u.tiled_copy_data.data_format = params.data_format; + lsdma_data->u.tiled_copy_data.max_com = params.max_com; + lsdma_data->u.tiled_copy_data.max_uncom = params.max_uncom; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA Tiled to Tiled Copy failed in DMUB"); + + return result; +} + +bool dmub_lsdma_send_pio_copy_command( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t src_addr, + uint64_t dst_addr, + uint32_t byte_count, + uint32_t overlap_disable) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_PIO_COPY; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_ENTRY_COUNTER, - (uint16_t)(output->ips_mode), - &output->entry_counter, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->entry_counter = 0; - - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_DURATION_US_LO, - (uint16_t)(output->ips_mode), - &output->total_active_time_us[0], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->total_active_time_us[0] = 0; - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_DURATION_US_HI, - (uint16_t)(output->ips_mode), - &output->total_active_time_us[1], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->total_active_time_us[1] = 0; - - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_INACTIVE_RESIDENCY_DURATION_US_LO, - (uint16_t)(output->ips_mode), - &output->total_inactive_time_us[0], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->total_inactive_time_us[0] = 0; - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_INACTIVE_RESIDENCY_DURATION_US_HI, - (uint16_t)(output->ips_mode), - &output->total_inactive_time_us[1], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->total_inactive_time_us[1] = 0; - - // NUM_IPS_HISTOGRAM_BUCKETS = 16 - for (i = 0; i < 16; i++) - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, command_code, i, &output->histogram[i], - DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->histogram[i] = 0; + lsdma_data->u.pio_copy_data.packet.fields.byte_count = byte_count; + lsdma_data->u.pio_copy_data.packet.fields.overlap_disable = overlap_disable; + lsdma_data->u.pio_copy_data.src_lo = src_addr & 0xFFFFFFFF; + lsdma_data->u.pio_copy_data.src_hi = (src_addr >> 32) & 0xFFFFFFFF; + lsdma_data->u.pio_copy_data.dst_lo = dst_addr & 0xFFFFFFFF; + lsdma_data->u.pio_copy_data.dst_hi = (dst_addr >> 32) & 0xFFFFFFFF; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA PIO Copy failed in DMUB"); + + return result; } + +bool dmub_lsdma_send_pio_constfill_command( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t dst_addr, + uint32_t byte_count, + uint32_t data) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_PIO_CONSTFILL; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; + + lsdma_data->u.pio_constfill_data.packet.fields.constant_fill = 1; + lsdma_data->u.pio_constfill_data.packet.fields.byte_count = byte_count; + lsdma_data->u.pio_constfill_data.dst_lo = dst_addr & 0xFFFFFFFF; + lsdma_data->u.pio_constfill_data.dst_hi = (dst_addr >> 32) & 0xFFFFFFFF; + lsdma_data->u.pio_constfill_data.data = data; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA PIO Constfill failed in DMUB"); + + return result; +} + +bool dmub_lsdma_send_poll_reg_write_command(struct dc_dmub_srv *dc_dmub_srv, uint32_t reg_addr, uint32_t reg_data) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_POLL_REG_WRITE; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; + + lsdma_data->u.reg_write_data.reg_addr = reg_addr; + lsdma_data->u.reg_write_data.reg_data = reg_data; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA Poll Reg failed in DMUB"); + + return result; +} + diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h index ada5c2fb2db3..1f1c155be30e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h @@ -210,6 +210,60 @@ void dc_dmub_srv_fams2_passthrough_flip( struct dc_surface_update *srf_updates, int surface_count); +bool dmub_lsdma_init(struct dc_dmub_srv *dc_dmub_srv); +bool dmub_lsdma_send_linear_copy_packet( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t src_addr, + uint64_t dst_addr, + uint32_t count); +bool dmub_lsdma_send_pio_copy_command( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t src_addr, + uint64_t dst_addr, + uint32_t byte_count, + uint32_t overlap_disable); +bool dmub_lsdma_send_pio_constfill_command( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t dst_addr, + uint32_t byte_count, + uint32_t data); + +struct lsdma_send_tiled_to_tiled_copy_command_params { + uint64_t src_addr; + uint64_t dst_addr; + + uint32_t src_x : 16; + uint32_t src_y : 16; + + uint32_t dst_x : 16; + uint32_t dst_y : 16; + + uint32_t src_width : 16; + uint32_t dst_width : 16; + + uint32_t rect_x : 16; + uint32_t rect_y : 16; + + uint32_t src_height : 16; + uint32_t dst_height : 16; + + uint32_t data_format : 6; + uint32_t swizzle_mode : 5; + uint32_t element_size : 3; + uint32_t dcc : 1; + uint32_t tmz : 1; + uint32_t read_compress : 2; + uint32_t write_compress : 2; + uint32_t max_com : 2; + uint32_t max_uncom : 1; + uint32_t padding : 9; +}; + +bool dmub_lsdma_send_tiled_to_tiled_copy_command( + struct dc_dmub_srv *dc_dmub_srv, + struct lsdma_send_tiled_to_tiled_copy_command_params params); +bool dmub_lsdma_send_poll_reg_write_command(struct dc_dmub_srv *dc_dmub_srv, uint32_t reg_addr, uint32_t reg_data); + /** * struct ips_residency_info - struct containing info from dmub_ips_residency_stats * @@ -223,7 +277,7 @@ void dc_dmub_srv_fams2_passthrough_flip( * @histogram: Histogram of given IPS state durations - bucket definitions in dmub_ips.c */ struct ips_residency_info { - enum dmub_ips_mode ips_mode; + enum ips_residency_mode ips_mode; unsigned int residency_percent; unsigned int entry_counter; unsigned int total_active_time_us[2]; @@ -231,21 +285,10 @@ struct ips_residency_info { unsigned int histogram[16]; }; -/** - * bool dc_dmub_srv_ips_residency_cntl() - Controls IPS residency measurement status - * - * @dc_dmub_srv: The DC DMUB service pointer - * @start_measurement: Describes whether to start or stop measurement - * - * Return: true if GPINT was sent successfully, false otherwise - */ -bool dc_dmub_srv_ips_residency_cntl(struct dc_dmub_srv *dc_dmub_srv, bool start_measurement); +bool dc_dmub_srv_ips_residency_cntl(const struct dc_context *ctx, uint8_t panel_inst, bool start_measurement); + +bool dc_dmub_srv_ips_query_residency_info(const struct dc_context *ctx, uint8_t panel_inst, + struct dmub_ips_residency_info *driver_info, + enum ips_residency_mode ips_mode); -/** - * bool dc_dmub_srv_ips_query_residency_info() - Queries DMCUB for residency info - * - * @dc_dmub_srv: The DC DMUB service pointer - * @output: Output struct to copy the the residency info to - */ -void dc_dmub_srv_ips_query_residency_info(struct dc_dmub_srv *dc_dmub_srv, struct ips_residency_info *output); #endif /* _DMUB_DC_SRV_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index 0bad8304ccf6..5ce1be362534 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -162,6 +162,11 @@ struct dc_link_settings { struct dc_tunnel_settings { bool should_enable_dp_tunneling; bool should_use_dp_bw_allocation; + uint8_t cm_id; + uint8_t group_id; + uint32_t bw_granularity; + uint32_t estimated_bw; + uint32_t allocated_bw; }; union dc_dp_ffe_preset { @@ -957,11 +962,21 @@ union usb4_driver_bw_cap { uint8_t raw; }; +/* DPCD[0xE0021] DP_IN_ADAPTER_TUNNEL_INFORMATION register. */ +union dpia_tunnel_info { + struct { + uint8_t group_id :3; + uint8_t rsvd :5; + } bits; + uint8_t raw; +}; + /* DP Tunneling over USB4 */ struct dpcd_usb4_dp_tunneling_info { union dp_tun_cap_support dp_tun_cap; union dpia_info dpia_info; union usb4_driver_bw_cap driver_bw_cap; + union dpia_tunnel_info dpia_tunnel_info; uint8_t usb4_driver_id; uint8_t usb4_topology_id[DPCD_USB4_TOPOLOGY_ID_LEN]; }; @@ -1172,8 +1187,8 @@ struct dc_lttpr_caps { union dp_128b_132b_supported_lttpr_link_rates supported_128b_132b_rates; union dp_alpm_lttpr_cap alpm; uint8_t aux_rd_interval[MAX_REPEATER_CNT - 1]; - uint8_t lttpr_ieee_oui[3]; - uint8_t lttpr_device_id[6]; + uint8_t lttpr_ieee_oui[3]; // Always read from closest LTTPR to host + uint8_t lttpr_device_id[6]; // Always read from closest LTTPR to host }; struct dc_dongle_dfp_cap_ext { diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h index d562ddeca512..667852517246 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h @@ -68,7 +68,7 @@ enum dc_plane_addr_type { struct dc_plane_address { enum dc_plane_addr_type type; - bool tmz_surface; + uint8_t tmz_surface; union { struct{ PHYSICAL_ADDRESS_LOC addr; @@ -974,6 +974,7 @@ struct dc_crtc_timing { uint32_t pix_clk_100hz; uint32_t min_refresh_in_uhz; + uint32_t max_refresh_in_uhz; uint32_t vic; uint32_t hdmi_vic; @@ -1103,7 +1104,8 @@ enum mpcc_gamut_remap_mode_select { enum mpcc_gamut_remap_id { MPCC_OGAM_GAMUT_REMAP, MPCC_MCM_FIRST_GAMUT_REMAP, - MPCC_MCM_SECOND_GAMUT_REMAP + MPCC_MCM_SECOND_GAMUT_REMAP, + MPCC_RMCM_GAMUT_REMAP, }; enum cursor_matrix_mode { diff --git a/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c b/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c index e3a8283b4098..7f57661433eb 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c +++ b/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c @@ -156,15 +156,16 @@ void translate_SPL_in_params_from_pipe_ctx(struct pipe_ctx *pipe_ctx, struct spl spl_in->adaptive_sharpness.enable = true; spl_in->adaptive_sharpness.sharpness_level = 0; } else if (sharpness_setting == SHARPNESS_CUSTOM) { - spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_min = 0; - spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_max = 1750; - spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_mid = 750; - spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_min = 0; - spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_max = 3500; - spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_mid = 1500; - spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_min = 0; - spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_max = 2750; - spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_mid = 1500; + /* SAT: read harpness_range from dc_plane_state */ + spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_min = plane_state->sharpness_range.sdr_rgb_min; + spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_max = plane_state->sharpness_range.sdr_rgb_max; + spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_mid = plane_state->sharpness_range.sdr_rgb_mid; + spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_min = plane_state->sharpness_range.sdr_yuv_min; + spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_max = plane_state->sharpness_range.sdr_yuv_max; + spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_mid = plane_state->sharpness_range.sdr_yuv_mid; + spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_min = plane_state->sharpness_range.hdr_rgb_min; + spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_max = plane_state->sharpness_range.hdr_rgb_max; + spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_mid = plane_state->sharpness_range.hdr_rgb_mid; if (force_sharpness_level > 0) { if (force_sharpness_level > 10) diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index a4cd0eb39a3a..b203ed020cd5 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -1255,7 +1255,6 @@ enum dc_cm2_gpu_mem_layout { enum dc_cm2_gpu_mem_pixel_component_order { DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_RGBA, - DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_BGRA }; enum dc_cm2_gpu_mem_format { @@ -1277,7 +1276,6 @@ struct dc_cm2_gpu_mem_format_parameters { enum dc_cm2_gpu_mem_size { DC_CM2_GPU_MEM_SIZE_171717, - DC_CM2_GPU_MEM_SIZE_333333, DC_CM2_GPU_MEM_SIZE_TRANSFORMED, }; @@ -1372,4 +1370,19 @@ struct set_backlight_level_params { uint8_t aux_inst; }; +enum dc_validate_mode { + /* validate the mode and program HW */ + DC_VALIDATE_MODE_AND_PROGRAMMING = 0, + /* only validate the mode */ + DC_VALIDATE_MODE_ONLY = 1, + /* validate the mode and get the max state (voltage level) */ + DC_VALIDATE_MODE_AND_STATE_INDEX = 2, +}; + +struct dc_validation_dpia_set { + const struct dc_link *link; + const struct dc_tunnel_settings *tunnel_settings; + uint32_t required_bw; +}; + #endif /* DC_TYPES_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c index ffd172231fdf..668ee2d405fd 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c @@ -727,7 +727,7 @@ void dccg401_init(struct dccg *dccg) } } -void dccg401_set_dto_dscclk(struct dccg *dccg, uint32_t inst) +void dccg401_set_dto_dscclk(struct dccg *dccg, uint32_t inst, uint32_t num_slices_h) { struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.h b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.h index 55e8718aad22..5947a35363aa 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.h @@ -209,7 +209,7 @@ void dccg401_disable_symclk32_le( struct dccg *dccg, int hpo_le_inst); void dccg401_disable_dpstreamclk(struct dccg *dccg, int dp_hpo_inst); -void dccg401_set_dto_dscclk(struct dccg *dccg, uint32_t inst); +void dccg401_set_dto_dscclk(struct dccg *dccg, uint32_t inst, uint32_t num_slices_h); void dccg401_set_ref_dscclk(struct dccg *dccg, uint32_t dsc_inst); void dccg401_set_src_sel( @@ -230,7 +230,6 @@ void dccg401_set_dp_dto( const struct dp_dto_params *params); void dccg401_enable_symclk_se(struct dccg *dccg, uint32_t stream_enc_inst, uint32_t link_enc_inst); void dccg401_disable_symclk_se(struct dccg *dccg, uint32_t stream_enc_inst, uint32_t link_enc_inst); -void dccg401_set_dto_dscclk(struct dccg *dccg, uint32_t inst); void dccg401_set_dtbclk_p_src( struct dccg *dccg, enum streamclk_source src, diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index e1d500633dfa..b357683b4255 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -114,9 +114,6 @@ CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calcs.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_auto.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_rcflags) -CFLAGS_$(AMDDALPATH)/dc/dml/dcn401/dcn401_fpu.o := $(dml_ccflags) -CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn401/dcn401_fpu.o := $(dml_rcflags) - ifdef CONFIG_DRM_AMD_DC_FP DML += display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o DML += dcn10/dcn10_fpu.o @@ -137,7 +134,6 @@ DML += dcn303/dcn303_fpu.o DML += dcn314/dcn314_fpu.o DML += dcn35/dcn35_fpu.o DML += dcn351/dcn351_fpu.o -DML += dcn401/dcn401_fpu.o DML += dsc/rc_calc_fpu.o DML += calcs/dcn_calcs.o calcs/dcn_calc_math.o calcs/dcn_calc_auto.o endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c index f1235bf9a596..74962791302f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c @@ -748,7 +748,7 @@ static unsigned int get_highest_allowed_voltage_level(bool is_vmin_only_asic) bool dcn_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { /* * we want a breakdown of the various stages of validation, which the @@ -1119,7 +1119,7 @@ bool dcn_validate_bandwidth( BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (v->voltage_level != number_of_states_plus_one && !fast_validate) { + if (v->voltage_level != number_of_states_plus_one && validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING) { float bw_consumed = v->total_bandwidth_consumed_gbyte_per_second; if (bw_consumed < v->fabric_and_dram_bandwidth_vmin0p65) @@ -1286,7 +1286,7 @@ bool dcn_validate_bandwidth( } } else if (v->voltage_level == number_of_states_plus_one) { BW_VAL_TRACE_SKIP(fail); - } else if (fast_validate) { + } else if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c index e9fea9c2162e..2a2eaf6adf26 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c @@ -1315,7 +1315,7 @@ static void swizzle_to_dml_params( int dcn20_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int pipe_cnt, i; bool synchronized_vblank = true; @@ -1733,7 +1733,7 @@ void dcn20_calculate_wm(struct dc *dc, struct dc_state *context, int *out_pipe_cnt, int *pipe_split_from, int vlevel, - bool fast_validate) + enum dc_validate_mode validate_mode) { int pipe_cnt, i, pipe_idx; @@ -1780,10 +1780,10 @@ void dcn20_calculate_wm(struct dc *dc, struct dc_state *context, if (pipe_cnt != pipe_idx) { if (dc->res_pool->funcs->populate_dml_pipes) pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, - context, pipes, fast_validate); + context, pipes, validate_mode); else pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, - context, pipes, fast_validate); + context, pipes, validate_mode); } *out_pipe_cnt = pipe_cnt; @@ -2027,7 +2027,7 @@ void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st } static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *context, - bool fast_validate, display_e2e_pipe_params_st *pipes) + enum dc_validate_mode validate_mode, display_e2e_pipe_params_st *pipes) { bool out = false; @@ -2040,7 +2040,7 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co BW_VAL_TRACE_COUNT(); - out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, fast_validate); + out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, validate_mode); if (pipe_cnt == 0) goto validate_out; @@ -2050,12 +2050,12 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } - dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); + dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, validate_mode); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); BW_VAL_TRACE_END_WATERMARKS(); @@ -2077,7 +2077,7 @@ validate_out: } bool dcn20_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, - bool fast_validate, display_e2e_pipe_params_st *pipes) + enum dc_validate_mode validate_mode, display_e2e_pipe_params_st *pipes) { bool voltage_supported = false; bool full_pstate_supported = false; @@ -2095,12 +2095,11 @@ bool dcn20_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, /*Unsafe due to current pipe merge and split logic*/ ASSERT(context != dc->current_state); - if (fast_validate) { - return dcn20_validate_bandwidth_internal(dc, context, true, pipes); - } + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) + return dcn20_validate_bandwidth_internal(dc, context, validate_mode, pipes); // Best case, we support full UCLK switch latency - voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false, pipes); + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING, pipes); full_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; if (context->bw_ctx.dml.soc.dummy_pstate_latency_us == 0 || @@ -2113,7 +2112,7 @@ bool dcn20_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, context->bw_ctx.dml.soc.dram_clock_change_latency_us = context->bw_ctx.dml.soc.dummy_pstate_latency_us; memset(pipes, 0, dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st)); - voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false, pipes); + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING, pipes); dummy_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; if (voltage_supported && (dummy_pstate_supported || !(context->stream_count))) { @@ -2156,14 +2155,14 @@ void dcn20_fpu_adjust_dppclk(struct vba_vars_st *v, int dcn21_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { uint32_t pipe_cnt; int i; dc_assert_fp_enabled(); - pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); for (i = 0; i < pipe_cnt; i++) { @@ -2239,7 +2238,7 @@ static void dcn21_calculate_wm(struct dc *dc, struct dc_state *context, int *out_pipe_cnt, int *pipe_split_from, int vlevel_req, - bool fast_validate) + enum dc_validate_mode validate_mode) { int pipe_cnt, i, pipe_idx; int vlevel, vlevel_max; @@ -2281,10 +2280,10 @@ static void dcn21_calculate_wm(struct dc *dc, struct dc_state *context, if (pipe_cnt != pipe_idx) { if (dc->res_pool->funcs->populate_dml_pipes) pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, - context, pipes, fast_validate); + context, pipes, validate_mode); else pipe_cnt = dcn21_populate_dml_pipes_from_context(dc, - context, pipes, fast_validate); + context, pipes, validate_mode); } *out_pipe_cnt = pipe_cnt; @@ -2319,7 +2318,7 @@ static void dcn21_calculate_wm(struct dc *dc, struct dc_state *context, } bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, - bool fast_validate, display_e2e_pipe_params_st *pipes) + enum dc_validate_mode validate_mode, display_e2e_pipe_params_st *pipes) { bool out = false; @@ -2337,7 +2336,7 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, /*Unsafe due to current pipe merge and split logic*/ ASSERT(context != dc->current_state); - out = dcn21_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, fast_validate); + out = dcn21_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, validate_mode); if (pipe_cnt == 0) goto validate_out; @@ -2347,12 +2346,12 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } - dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); + dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, validate_mode); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); BW_VAL_TRACE_END_WATERMARKS(); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h index b6c34198ddc8..aed00039ca62 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h @@ -44,14 +44,14 @@ void dcn20_calculate_dlg_params(struct dc *dc, int dcn20_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn20_calculate_wm(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, int *out_pipe_cnt, int *pipe_split_from, int vlevel, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn20_cap_soc_clocks(struct _vcs_dpi_soc_bounding_box_st *bb, struct pp_smu_nv_clock_table max_clocks); void dcn20_update_bounding_box(struct dc *dc, @@ -62,7 +62,7 @@ void dcn20_update_bounding_box(struct dc *dc, void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb); bool dcn20_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, - bool fast_validate, display_e2e_pipe_params_st *pipes); + enum dc_validate_mode validate_mode, display_e2e_pipe_params_st *pipes); void dcn20_fpu_set_wm_ranges(int i, struct pp_smu_wm_range_sets *ranges, struct _vcs_dpi_soc_bounding_box_st *loaded_bb); @@ -75,9 +75,9 @@ void dcn20_fpu_adjust_dppclk(struct vba_vars_st *v, int dcn21_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); -bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, bool - fast_validate, display_e2e_pipe_params_st *pipes); + enum dc_validate_mode validate_mode); +bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, enum + dc_validate_mode, display_e2e_pipe_params_st *pipes); void dcn21_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); void dcn21_clk_mgr_set_bw_params_wm_table(struct clk_bw_params *bw_params); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c index 88789987bdbc..e5f5c0663750 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c @@ -339,7 +339,8 @@ void dcn30_fpu_calculate_wm_and_dlg( * newly found dummy_latency_index */ context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us; - dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false, true); + dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, + DC_VALIDATE_MODE_AND_PROGRAMMING, true); maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb]; pstate_en = context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] != dm_dram_clock_change_unsupported; @@ -630,7 +631,8 @@ int dcn30_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc, while (dummy_latency_index < max_latency_table_entries) { context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us; - dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false, true); + dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, + DC_VALIDATE_MODE_AND_PROGRAMMING, true); if (context->bw_ctx.dml.soc.allow_dram_self_refresh_or_dram_clock_change_in_vblank == dm_allow_self_refresh_and_mclk_switch) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h index d2ae43a82ba5..dfcc5d50071e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h @@ -55,5 +55,5 @@ int dcn_get_approx_det_segs_required_for_pstate( int dcn31x_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); #endif /* __DCN31_FPU_H__*/ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c index 5ed117e11aa2..df9d50b9b57c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c @@ -306,7 +306,7 @@ static unsigned int get_vertical_back_porch(struct dc_crtc_timing *timing) int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -316,7 +316,7 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c dc_assert_fp_enabled(); - dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn31x_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h index d32c5bb99f4c..362ac79184ea 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h @@ -35,6 +35,6 @@ void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params); int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c index b0fc1fd20208..6160952245b4 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c @@ -290,7 +290,7 @@ int dcn32_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc, vba->DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = temp_clock_change_support; context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us; - dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false); + dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, DC_VALIDATE_MODE_AND_PROGRAMMING); /* for subvp + DRR case, if subvp pipes are still present we support pstate */ if (vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported && @@ -1479,7 +1479,7 @@ static bool dcn32_full_validate_bw_helper(struct dc *dc, /* Conditions for setting up phantom pipes for SubVP: * 1. Not force disable SubVP - * 2. Full update (i.e. !fast_validate) + * 2. Full update (i.e. DC_VALIDATE_MODE_AND_PROGRAMMING) * 3. Enough pipes are available to support SubVP (TODO: Which pipes will use VACTIVE / VBLANK / SUBVP?) * 4. Display configuration passes validation * 5. (Config doesn't support MCLK in VACTIVE/VBLANK || dc->debug.force_subvp_mclk_switch) @@ -1517,7 +1517,8 @@ static bool dcn32_full_validate_bw_helper(struct dc *dc, dc->res_pool->funcs->add_phantom_pipes(dc, context, pipes, *pipe_cnt, dc_pipe_idx); - *pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, false); + *pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, + DC_VALIDATE_MODE_AND_PROGRAMMING); // Populate dppclk to trigger a recalculate in dml_get_voltage_level // so the phantom pipe DLG params can be assigned correctly. pipes[0].clks_cfg.dppclk_mhz = get_dppclk_calculated(&context->bw_ctx.dml, pipes, *pipe_cnt, 0); @@ -1560,7 +1561,8 @@ static bool dcn32_full_validate_bw_helper(struct dc *dc, dc_state_remove_phantom_streams_and_planes(dc, context); dc_state_release_phantom_streams_and_planes(dc, context); vba->DRAMClockChangeSupport[*vlevel][vba->maxMpcComb] = dm_dram_clock_change_unsupported; - *pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, false); + *pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, + DC_VALIDATE_MODE_AND_PROGRAMMING); *vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, *pipe_cnt); /* This may adjust vlevel and maxMpcComb */ @@ -2138,7 +2140,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, display_e2e_pipe_params_st *pipes, int *pipe_cnt_out, int *vlevel_out, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; bool repopulate_pipes = false; @@ -2162,7 +2164,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, for (i = 0; i < context->stream_count; i++) resource_update_pipes_for_stream_with_slice_count(context, dc->current_state, dc->res_pool, context->streams[i], 1); - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); if (!pipe_cnt) { out = true; @@ -2172,13 +2174,13 @@ bool dcn32_internal_validate_bw(struct dc *dc, dml_log_pipe_params(&context->bw_ctx.dml, pipes, pipe_cnt); context->bw_ctx.dml.soc.max_vratio_pre = dcn32_determine_max_vratio_prefetch(dc, context); - if (!fast_validate) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING) { if (!dcn32_full_validate_bw_helper(dc, context, pipes, &vlevel, split, merge, &pipe_cnt, &repopulate_pipes)) goto validate_fail; } - if (fast_validate || + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING || (dc->debug.dml_disallow_alternate_prefetch_modes && (vlevel == context->bw_ctx.dml.soc.num_states || vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported))) { @@ -2195,7 +2197,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final = dm_prefetch_support_none; - context->bw_ctx.dml.validate_max_state = fast_validate; + context->bw_ctx.dml.validate_max_state = (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING); vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt); context->bw_ctx.dml.validate_max_state = false; @@ -2247,7 +2249,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, int flag_vlevel = vlevel; int i; - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); if (!dc->config.enable_windowed_mpo_odm) dcn32_update_dml_pipes_odm_policy_based_on_context(dc, context, pipes); @@ -2343,7 +2345,7 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, } context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us; - dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false); + dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, DC_VALIDATE_MODE_AND_PROGRAMMING); maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; if (is_subvp_p_drr) { context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] = dm_dram_clock_change_vblank_w_mall_sub_vp; @@ -2389,7 +2391,8 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, context->bw_ctx.dml.soc.fclk_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us; } - dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel_temp, false); + dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel_temp, + DC_VALIDATE_MODE_AND_PROGRAMMING); if (vlevel_temp < vlevel) { vlevel = vlevel_temp; maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; @@ -2410,7 +2413,8 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, stream_status->fpo_in_use = false; } context->bw_ctx.dml.soc.fclk_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.fclk_change_latency_us; - dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false); + dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, + DC_VALIDATE_MODE_AND_PROGRAMMING); } } } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h index 276e90e4e0ce..273d2bd79d85 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h @@ -49,7 +49,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, display_e2e_pipe_params_st *pipes, int *pipe_cnt_out, int *vlevel_out, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c index 92f0a099d089..5d73efa2f0c9 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c @@ -437,7 +437,7 @@ static unsigned int get_vertical_back_porch(struct dc_crtc_timing *timing) int dcn35_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -446,7 +446,7 @@ int dcn35_populate_dml_pipes_from_context_fpu(struct dc *dc, const unsigned int max_allowed_vblank_nom = 1023; dcn31_populate_dml_pipes_from_context(dc, context, pipes, - fast_validate); + validate_mode); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.h index 067480fc3691..d121c5afce71 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.h @@ -37,7 +37,7 @@ void dcn35_update_bw_bounding_box_fpu(struct dc *dc, int dcn35_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn35_decide_zstate_support(struct dc *dc, struct dc_state *context); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c index 17d0b4923b0c..6f516af82956 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c @@ -470,7 +470,7 @@ static unsigned int get_vertical_back_porch(struct dc_crtc_timing *timing) int dcn351_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -479,7 +479,7 @@ int dcn351_populate_dml_pipes_from_context_fpu(struct dc *dc, const unsigned int max_allowed_vblank_nom = 1023; dcn31_populate_dml_pipes_from_context(dc, context, pipes, - fast_validate); + validate_mode); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.h index f93efab9a668..f71d9d8d0759 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.h @@ -12,7 +12,7 @@ void dcn351_update_bw_bounding_box_fpu(struct dc *dc, int dcn351_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn351_decide_zstate_support(struct dc *dc, struct dc_state *context); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.c deleted file mode 100644 index 4fbecb5ff349..000000000000 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.c +++ /dev/null @@ -1,239 +0,0 @@ -// SPDX-License-Identifier: MIT -// -// Copyright 2024 Advanced Micro Devices, Inc. - -#include "dcn401_fpu.h" -#include "dcn401/dcn401_resource.h" -// We need this includes for WATERMARKS_* defines -#include "clk_mgr/dcn401/dcn401_smu14_driver_if.h" -#include "link.h" - -#define DC_LOGGER_INIT(logger) - -void dcn401_build_wm_range_table_fpu(struct clk_mgr *clk_mgr) -{ - /* defaults */ - double pstate_latency_us = clk_mgr->ctx->dc->dml.soc.dram_clock_change_latency_us; - double fclk_change_latency_us = clk_mgr->ctx->dc->dml.soc.fclk_change_latency_us; - double sr_exit_time_us = clk_mgr->ctx->dc->dml.soc.sr_exit_time_us; - double sr_enter_plus_exit_time_us = clk_mgr->ctx->dc->dml.soc.sr_enter_plus_exit_time_us; - /* For min clocks use as reported by PM FW and report those as min */ - uint16_t min_uclk_mhz = clk_mgr->bw_params->clk_table.entries[0].memclk_mhz; - uint16_t min_dcfclk_mhz = clk_mgr->bw_params->clk_table.entries[0].dcfclk_mhz; - uint16_t setb_min_uclk_mhz = min_uclk_mhz; - uint16_t dcfclk_mhz_for_the_second_state = clk_mgr->ctx->dc->dml.soc.clock_limits[2].dcfclk_mhz; - - dc_assert_fp_enabled(); - - /* For Set B ranges use min clocks state 2 when available, and report those to PM FW */ - if (dcfclk_mhz_for_the_second_state) - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.min_dcfclk = dcfclk_mhz_for_the_second_state; - else - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.min_dcfclk = clk_mgr->bw_params->clk_table.entries[0].dcfclk_mhz; - - if (clk_mgr->bw_params->clk_table.entries[2].memclk_mhz) - setb_min_uclk_mhz = clk_mgr->bw_params->clk_table.entries[2].memclk_mhz; - - /* Set A - Normal - default values */ - clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us = pstate_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.fclk_change_latency_us = fclk_change_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us = sr_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.wm_type = WATERMARKS_CLOCK_RANGE; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_uclk = min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_uclk = 0xFFFF; - - /* Set B - Performance - higher clocks, using DPM[2] DCFCLK and UCLK */ - clk_mgr->bw_params->wm_table.nv_entries[WM_B].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.pstate_latency_us = pstate_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.fclk_change_latency_us = fclk_change_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.sr_exit_time_us = sr_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.wm_type = WATERMARKS_CLOCK_RANGE; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.min_uclk = setb_min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.max_uclk = 0xFFFF; - - /* Set C - Dummy P-State - P-State latency set to "dummy p-state" value */ - /* 'DalDummyClockChangeLatencyNs' registry key option set to 0x7FFFFFFF can be used to disable Set C for dummy p-state */ - if (clk_mgr->ctx->dc->bb_overrides.dummy_clock_change_latency_ns != 0x7FFFFFFF) { - clk_mgr->bw_params->wm_table.nv_entries[WM_C].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.pstate_latency_us = 50; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.fclk_change_latency_us = fclk_change_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us = sr_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.wm_type = WATERMARKS_DUMMY_PSTATE; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.min_uclk = min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.max_uclk = 0xFFFF; - clk_mgr->bw_params->dummy_pstate_table[0].dram_speed_mts = clk_mgr->bw_params->clk_table.entries[0].memclk_mhz * 16; - clk_mgr->bw_params->dummy_pstate_table[0].dummy_pstate_latency_us = 50; - clk_mgr->bw_params->dummy_pstate_table[1].dram_speed_mts = clk_mgr->bw_params->clk_table.entries[1].memclk_mhz * 16; - clk_mgr->bw_params->dummy_pstate_table[1].dummy_pstate_latency_us = 9; - clk_mgr->bw_params->dummy_pstate_table[2].dram_speed_mts = clk_mgr->bw_params->clk_table.entries[2].memclk_mhz * 16; - clk_mgr->bw_params->dummy_pstate_table[2].dummy_pstate_latency_us = 8; - clk_mgr->bw_params->dummy_pstate_table[3].dram_speed_mts = clk_mgr->bw_params->clk_table.entries[3].memclk_mhz * 16; - clk_mgr->bw_params->dummy_pstate_table[3].dummy_pstate_latency_us = 5; - } - /* Set D - MALL - SR enter and exit time specific to MALL, TBD after bringup or later phase for now use DRAM values / 2 */ - /* For MALL DRAM clock change latency is N/A, for watermak calculations use lowest value dummy P state latency */ - clk_mgr->bw_params->wm_table.nv_entries[WM_D].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.pstate_latency_us = clk_mgr->bw_params->dummy_pstate_table[3].dummy_pstate_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.fclk_change_latency_us = fclk_change_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_exit_time_us = sr_exit_time_us / 2; // TBD - clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us / 2; // TBD - clk_mgr->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.wm_type = WATERMARKS_MALL; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.min_uclk = min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.max_uclk = 0xFFFF; -} - -/* - * dcn401_update_bw_bounding_box - * - * This would override some dcn4_01 ip_or_soc initial parameters hardcoded from - * spreadsheet with actual values as per dGPU SKU: - * - with passed few options from dc->config - * - with dentist_vco_frequency from Clk Mgr (currently hardcoded, but might - * need to get it from PM FW) - * - with passed latency values (passed in ns units) in dc-> bb override for - * debugging purposes - * - with passed latencies from VBIOS (in 100_ns units) if available for - * certain dGPU SKU - * - with number of DRAM channels from VBIOS (which differ for certain dGPU SKU - * of the same ASIC) - * - clocks levels with passed clk_table entries from Clk Mgr as reported by PM - * FW for different clocks (which might differ for certain dGPU SKU of the - * same ASIC) - */ -void dcn401_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params) -{ - dc_assert_fp_enabled(); - - /* Override from passed dc->bb_overrides if available*/ - if (dc->bb_overrides.sr_exit_time_ns) - dc->dml2_options.bbox_overrides.sr_exit_latency_us = - dc->bb_overrides.sr_exit_time_ns / 1000.0; - - if (dc->bb_overrides.sr_enter_plus_exit_time_ns) - dc->dml2_options.bbox_overrides.sr_enter_plus_exit_latency_us = - dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; - - if (dc->bb_overrides.urgent_latency_ns) - dc->dml2_options.bbox_overrides.urgent_latency_us = - dc->bb_overrides.urgent_latency_ns / 1000.0; - - if (dc->bb_overrides.dram_clock_change_latency_ns) - dc->dml2_options.bbox_overrides.dram_clock_change_latency_us = - dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - - if (dc->bb_overrides.fclk_clock_change_latency_ns) - dc->dml2_options.bbox_overrides.fclk_change_latency_us = - dc->bb_overrides.fclk_clock_change_latency_ns / 1000; - - /* Override from VBIOS if VBIOS bb_info available */ - if (dc->ctx->dc_bios->funcs->get_soc_bb_info) { - struct bp_soc_bb_info bb_info = {0}; - if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) { - if (bb_info.dram_clock_change_latency_100ns > 0) - dc->dml2_options.bbox_overrides.dram_clock_change_latency_us = - bb_info.dram_clock_change_latency_100ns * 10; - - if (bb_info.dram_sr_enter_exit_latency_100ns > 0) - dc->dml2_options.bbox_overrides.sr_enter_plus_exit_latency_us = - bb_info.dram_sr_enter_exit_latency_100ns * 10; - - if (bb_info.dram_sr_exit_latency_100ns > 0) - dc->dml2_options.bbox_overrides.sr_exit_latency_us = - bb_info.dram_sr_exit_latency_100ns * 10; - } - } - - /* Override from VBIOS for num_chan */ - if (dc->ctx->dc_bios->vram_info.num_chans) { - dc->dml2_options.bbox_overrides.dram_num_chan = - dc->ctx->dc_bios->vram_info.num_chans; - - } - - if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) - dc->dml2_options.bbox_overrides.dram_chanel_width_bytes = - dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; - - dc->dml2_options.bbox_overrides.disp_pll_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; - dc->dml2_options.bbox_overrides.xtalclk_mhz = dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000.0; - dc->dml2_options.bbox_overrides.dchub_refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; - dc->dml2_options.bbox_overrides.dprefclk_mhz = dc->clk_mgr->dprefclk_khz / 1000.0; - - if (dc->clk_mgr->bw_params->clk_table.num_entries > 1) { - unsigned int i = 0; - - dc->dml2_options.bbox_overrides.clks_table.num_states = dc->clk_mgr->bw_params->clk_table.num_entries; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dcfclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dcfclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_fclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_fclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_memclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_socclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_socclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dtbclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dtbclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dispclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dispclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dppclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dppclk_levels; - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dcfclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].dcfclk_mhz) - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dcfclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].dcfclk_mhz; - } - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_fclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].fclk_mhz) - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].fclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].fclk_mhz; - } - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].memclk_mhz) - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].memclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].memclk_mhz; - } - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_socclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].socclk_mhz) - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].socclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].socclk_mhz; - } - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dtbclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].dtbclk_mhz) - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dtbclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].dtbclk_mhz; - } - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dispclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].dispclk_mhz) { - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dispclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].dispclk_mhz; - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dppclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].dispclk_mhz; - } - } - } -} - diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.h deleted file mode 100644 index 329f1788843c..000000000000 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.h +++ /dev/null @@ -1,14 +0,0 @@ -// SPDX-License-Identifier: MIT -// -// Copyright 2024 Advanced Micro Devices, Inc. - -#ifndef __DCN401_FPU_H__ -#define __DCN401_FPU_H__ - -#include "clk_mgr.h" - -void dcn401_build_wm_range_table_fpu(struct clk_mgr *clk_mgr); - -void dcn401_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params); - -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml2/Makefile b/drivers/gpu/drm/amd/display/dc/dml2/Makefile index 157ecf008d6c..4c21ce42054c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml2/Makefile @@ -81,10 +81,11 @@ AMD_DAL_DML2 = $(addprefix $(AMDDALPATH)/dc/dml2/,$(DML2)) AMD_DISPLAY_FILES += $(AMD_DAL_DML2) -CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_interfaces.o := $(dml2_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.o := $(dml2_ccflags) $(frame_warn_flag) +CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_utils.o := $(dml2_ccflags) $(frame_warn_flag) +CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_interfaces.o := $(dml2_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_factory.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.o := $(dml2_ccflags) @@ -94,17 +95,16 @@ CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn3.o := $(dml2_ccflag CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_standalone_libraries/lib_float_math.o := $(dml2_ccflags) - - CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml21_wrapper.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/dml21_translation_helper.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/dml21_utils.o := $(dml2_ccflags) -CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_interfaces.o := $(dml2_rcflags) -CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_factory.o := $(dml2_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_utils.o := $(dml2_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_interfaces.o := $(dml2_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_dcn4.o := $(dml2_rcflags) @@ -120,6 +120,7 @@ CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/dml21_utils.o := $(dml2_rcflags) DML21 := src/dml2_top/dml2_top_interfaces.o DML21 += src/dml2_top/dml2_top_soc15.o DML21 += src/dml2_core/dml2_core_dcn4.o +DML21 += src/dml2_core/dml2_core_utils.o DML21 += src/dml2_core/dml2_core_factory.o DML21 += src/dml2_core/dml2_core_dcn4_calcs.o DML21 += src/dml2_dpmm/dml2_dpmm_dcn4.o diff --git a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c index 7ae9c0ba0c9e..715f9019a33e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c @@ -10189,7 +10189,7 @@ dml_uint_t dml_mode_support_ex(struct dml_mode_support_ex_params_st *in_out_para result = mode_support_pwr_states(&in_out_params->out_lowest_state_idx, in_out_params->mode_lib, in_out_params->in_display_cfg, - 0, + in_out_params->in_start_state_idx, in_out_params->mode_lib->states.num_states - 1); if (result) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core_structs.h b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core_structs.h index 0670e4dc4fd9..dbeb08466092 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core_structs.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core_structs.h @@ -1917,6 +1917,7 @@ struct display_mode_lib_st { struct dml_mode_support_ex_params_st { struct display_mode_lib_st *mode_lib; const struct dml_display_cfg_st *in_display_cfg; + dml_uint_t in_start_state_idx; dml_uint_t out_lowest_state_idx; struct dml_mode_support_info_st *out_evaluation_info; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c index d47cacfdb695..a06217a9eef6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c @@ -2,7 +2,6 @@ // // Copyright 2024 Advanced Micro Devices, Inc. - #include "dml21_wrapper.h" #include "dml2_core_dcn4_calcs.h" #include "dml2_internal_shared_types.h" @@ -11,277 +10,263 @@ #include "dml21_translation_helper.h" #include "bounding_boxes/dcn4_soc_bb.h" -static void dml21_init_socbb_params(struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config, - const struct dc *in_dc) -{ - const struct dml2_soc_bb *soc_bb; - const struct dml2_soc_qos_parameters *qos_params; - - switch (in_dc->ctx->dce_version) { - case DCN_VERSION_4_01: - default: - if (config->bb_from_dmub) - soc_bb = config->bb_from_dmub; - else - soc_bb = &dml2_socbb_dcn401; - - qos_params = &dml_dcn4_variant_a_soc_qos_params; - } - - /* patch soc bb */ - memcpy(&dml_init->soc_bb, soc_bb, sizeof(struct dml2_soc_bb)); - - /* patch qos params */ - memcpy(&dml_init->soc_bb.qos_parameters, qos_params, sizeof(struct dml2_soc_qos_parameters)); -} - -static void dml21_external_socbb_params(struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config) -{ - memcpy(&dml_init->soc_bb, &config->external_socbb_ip_params->soc_bb, sizeof(struct dml2_soc_bb)); -} - -static void dml21_external_ip_params(struct dml2_initialize_instance_in_out *dml_init, +static void dml21_populate_pmo_options(struct dml2_pmo_options *pmo_options, + const struct dc *in_dc, const struct dml2_configuration_options *config) { - memcpy(&dml_init->ip_caps, &config->external_socbb_ip_params->ip_params, sizeof(struct dml2_ip_capabilities)); + bool disable_fams2 = !in_dc->debug.fams2_config.bits.enable; + + /* ODM options */ + pmo_options->disable_dyn_odm = !config->minimize_dispclk_using_odm; + pmo_options->disable_dyn_odm_for_multi_stream = true; + pmo_options->disable_dyn_odm_for_stream_with_svp = true; + + pmo_options->disable_vblank = ((in_dc->debug.dml21_disable_pstate_method_mask >> 1) & 1); + + /* NOTE: DRR and SubVP Require FAMS2 */ + pmo_options->disable_svp = ((in_dc->debug.dml21_disable_pstate_method_mask >> 2) & 1) || + in_dc->debug.force_disable_subvp || + disable_fams2; + pmo_options->disable_drr_clamped = ((in_dc->debug.dml21_disable_pstate_method_mask >> 3) & 1) || + disable_fams2; + pmo_options->disable_drr_var = ((in_dc->debug.dml21_disable_pstate_method_mask >> 4) & 1) || + disable_fams2; + pmo_options->disable_fams2 = disable_fams2; + + pmo_options->disable_drr_var_when_var_active = in_dc->debug.disable_fams_gaming == INGAME_FAMS_DISABLE || + in_dc->debug.disable_fams_gaming == INGAME_FAMS_MULTI_DISP_CLAMPED_ONLY; + pmo_options->disable_drr_clamped_when_var_active = in_dc->debug.disable_fams_gaming == INGAME_FAMS_DISABLE; } -static void dml21_init_ip_params(struct dml2_initialize_instance_in_out *dml_init, +/* + * Populate dml_init based on default static values in soc bb. The default + * values are for reference and support at least minimal operation of current + * SoC and DCN hardware. The values could be modifed by subsequent override + * functions to reflect our true hardware capability. + */ +static void populate_default_dml_init_params(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc) { - const struct dml2_ip_capabilities *ip_caps; - switch (in_dc->ctx->dce_version) { case DCN_VERSION_4_01: + dml_init->options.project_id = dml2_project_dcn4x_stage2_auto_drr_svp; + dml21_populate_pmo_options(&dml_init->options.pmo_options, in_dc, config); + dml_init->soc_bb = dml2_socbb_dcn401; + dml_init->soc_bb.qos_parameters = dml_dcn4_variant_a_soc_qos_params; + dml_init->ip_caps = dml2_dcn401_max_ip_caps; + break; default: - ip_caps = &dml2_dcn401_max_ip_caps; + memset(dml_init, 0, sizeof(*dml_init)); + DC_ERR("unsupported dcn version for DML21!"); + return; } - - memcpy(&dml_init->ip_caps, ip_caps, sizeof(struct dml2_ip_capabilities)); } -void dml21_initialize_soc_bb_params(struct dml2_initialize_instance_in_out *dml_init, +static void override_dml_init_with_values_from_hardware_default(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc) { - if (config->use_native_soc_bb_construction) - dml21_init_socbb_params(dml_init, config, in_dc); - else - dml21_external_socbb_params(dml_init, config); + dml_init->soc_bb.dchub_refclk_mhz = in_dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000; + dml_init->soc_bb.dprefclk_mhz = in_dc->clk_mgr->dprefclk_khz / 1000; + dml_init->soc_bb.dispclk_dppclk_vco_speed_mhz = in_dc->clk_mgr->dentist_vco_freq_khz / 1000.0; } -void dml21_initialize_ip_params(struct dml2_initialize_instance_in_out *dml_init, +/* + * SMU stands for System Management Unit. It is a power management processor. + * It owns the initialization of dc's clock table and programming of clock values + * based on dc's requests. + * Our clock values in base soc bb is a dummy placeholder. The real clock values + * are retrieved from SMU firmware to dc clock table at runtime. + * This function overrides our dummy placeholder values with real values in dc + * clock table. + */ +static void override_dml_init_with_values_from_smu( + struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc) { - if (config->use_native_soc_bb_construction) - dml21_init_ip_params(dml_init, config, in_dc); - else - dml21_external_ip_params(dml_init, config); -} - -void dml21_apply_soc_bb_overrides(struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config, const struct dc *in_dc) -{ int i; - const struct clk_bw_params *dc_bw_params = in_dc->clk_mgr->bw_params; const struct clk_limit_table *dc_clk_table = &dc_bw_params->clk_table; - struct dml2_soc_bb *dml_soc_bb = &dml_init->soc_bb; - struct dml2_soc_state_table *dml_clk_table = &dml_soc_bb->clk_table; - - /* override clocks if smu is present */ - if (in_dc->clk_mgr->funcs->is_smu_present && in_dc->clk_mgr->funcs->is_smu_present(in_dc->clk_mgr)) { - /* dcfclk */ - if (dc_clk_table->num_entries_per_clk.num_dcfclk_levels) { - dml_clk_table->dcfclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dcfclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dcfclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dcfclk_mhz && - dc_clk_table->entries[i].dcfclk_mhz > dc_bw_params->dc_mode_limit.dcfclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dcfclk_mhz < dc_bw_params->dc_mode_limit.dcfclk_mhz) { - dml_clk_table->dcfclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dcfclk_mhz * 1000; - dml_clk_table->dcfclk.num_clk_values = i + 1; - } else { - dml_clk_table->dcfclk.clk_values_khz[i] = 0; - dml_clk_table->dcfclk.num_clk_values = i; - } + struct dml2_soc_state_table *dml_clk_table = &dml_init->soc_bb.clk_table; + + if (!in_dc->clk_mgr->funcs->is_smu_present || + !in_dc->clk_mgr->funcs->is_smu_present(in_dc->clk_mgr)) + /* skip if smu is not present */ + return; + + /* dcfclk */ + if (dc_clk_table->num_entries_per_clk.num_dcfclk_levels) { + dml_clk_table->dcfclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dcfclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dcfclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dcfclk_mhz && + dc_clk_table->entries[i].dcfclk_mhz > dc_bw_params->dc_mode_limit.dcfclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dcfclk_mhz < dc_bw_params->dc_mode_limit.dcfclk_mhz) { + dml_clk_table->dcfclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dcfclk_mhz * 1000; + dml_clk_table->dcfclk.num_clk_values = i + 1; } else { - dml_clk_table->dcfclk.clk_values_khz[i] = dc_clk_table->entries[i].dcfclk_mhz * 1000; + dml_clk_table->dcfclk.clk_values_khz[i] = 0; + dml_clk_table->dcfclk.num_clk_values = i; } } else { - dml_clk_table->dcfclk.clk_values_khz[i] = 0; + dml_clk_table->dcfclk.clk_values_khz[i] = dc_clk_table->entries[i].dcfclk_mhz * 1000; } + } else { + dml_clk_table->dcfclk.clk_values_khz[i] = 0; } } + } - /* fclk */ - if (dc_clk_table->num_entries_per_clk.num_fclk_levels) { - dml_clk_table->fclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_fclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->fclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.fclk_mhz && - dc_clk_table->entries[i].fclk_mhz > dc_bw_params->dc_mode_limit.fclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].fclk_mhz < dc_bw_params->dc_mode_limit.fclk_mhz) { - dml_clk_table->fclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.fclk_mhz * 1000; - dml_clk_table->fclk.num_clk_values = i + 1; - } else { - dml_clk_table->fclk.clk_values_khz[i] = 0; - dml_clk_table->fclk.num_clk_values = i; - } + /* fclk */ + if (dc_clk_table->num_entries_per_clk.num_fclk_levels) { + dml_clk_table->fclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_fclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->fclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.fclk_mhz && + dc_clk_table->entries[i].fclk_mhz > dc_bw_params->dc_mode_limit.fclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].fclk_mhz < dc_bw_params->dc_mode_limit.fclk_mhz) { + dml_clk_table->fclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.fclk_mhz * 1000; + dml_clk_table->fclk.num_clk_values = i + 1; } else { - dml_clk_table->fclk.clk_values_khz[i] = dc_clk_table->entries[i].fclk_mhz * 1000; + dml_clk_table->fclk.clk_values_khz[i] = 0; + dml_clk_table->fclk.num_clk_values = i; } } else { - dml_clk_table->fclk.clk_values_khz[i] = 0; + dml_clk_table->fclk.clk_values_khz[i] = dc_clk_table->entries[i].fclk_mhz * 1000; } + } else { + dml_clk_table->fclk.clk_values_khz[i] = 0; } } + } - /* uclk */ - if (dc_clk_table->num_entries_per_clk.num_memclk_levels) { - dml_clk_table->uclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_memclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->uclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.memclk_mhz && - dc_clk_table->entries[i].memclk_mhz > dc_bw_params->dc_mode_limit.memclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].memclk_mhz < dc_bw_params->dc_mode_limit.memclk_mhz) { - dml_clk_table->uclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.memclk_mhz * 1000; - dml_clk_table->uclk.num_clk_values = i + 1; - } else { - dml_clk_table->uclk.clk_values_khz[i] = 0; - dml_clk_table->uclk.num_clk_values = i; - } + /* uclk */ + if (dc_clk_table->num_entries_per_clk.num_memclk_levels) { + dml_clk_table->uclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_memclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->uclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.memclk_mhz && + dc_clk_table->entries[i].memclk_mhz > dc_bw_params->dc_mode_limit.memclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].memclk_mhz < dc_bw_params->dc_mode_limit.memclk_mhz) { + dml_clk_table->uclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.memclk_mhz * 1000; + dml_clk_table->uclk.num_clk_values = i + 1; } else { - dml_clk_table->uclk.clk_values_khz[i] = dc_clk_table->entries[i].memclk_mhz * 1000; + dml_clk_table->uclk.clk_values_khz[i] = 0; + dml_clk_table->uclk.num_clk_values = i; } } else { - dml_clk_table->uclk.clk_values_khz[i] = 0; + dml_clk_table->uclk.clk_values_khz[i] = dc_clk_table->entries[i].memclk_mhz * 1000; } + } else { + dml_clk_table->uclk.clk_values_khz[i] = 0; } } + } - /* dispclk */ - if (dc_clk_table->num_entries_per_clk.num_dispclk_levels) { - dml_clk_table->dispclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dispclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dispclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dispclk_mhz && - dc_clk_table->entries[i].dispclk_mhz > dc_bw_params->dc_mode_limit.dispclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dispclk_mhz < dc_bw_params->dc_mode_limit.dispclk_mhz) { - dml_clk_table->dispclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dispclk_mhz * 1000; - dml_clk_table->dispclk.num_clk_values = i + 1; - } else { - dml_clk_table->dispclk.clk_values_khz[i] = 0; - dml_clk_table->dispclk.num_clk_values = i; - } + /* dispclk */ + if (dc_clk_table->num_entries_per_clk.num_dispclk_levels) { + dml_clk_table->dispclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dispclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dispclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dispclk_mhz && + dc_clk_table->entries[i].dispclk_mhz > dc_bw_params->dc_mode_limit.dispclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dispclk_mhz < dc_bw_params->dc_mode_limit.dispclk_mhz) { + dml_clk_table->dispclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dispclk_mhz * 1000; + dml_clk_table->dispclk.num_clk_values = i + 1; } else { - dml_clk_table->dispclk.clk_values_khz[i] = dc_clk_table->entries[i].dispclk_mhz * 1000; + dml_clk_table->dispclk.clk_values_khz[i] = 0; + dml_clk_table->dispclk.num_clk_values = i; } } else { - dml_clk_table->dispclk.clk_values_khz[i] = 0; + dml_clk_table->dispclk.clk_values_khz[i] = dc_clk_table->entries[i].dispclk_mhz * 1000; } + } else { + dml_clk_table->dispclk.clk_values_khz[i] = 0; } } + } - /* dppclk */ - if (dc_clk_table->num_entries_per_clk.num_dppclk_levels) { - dml_clk_table->dppclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dppclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dppclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dppclk_mhz && - dc_clk_table->entries[i].dppclk_mhz > dc_bw_params->dc_mode_limit.dppclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dppclk_mhz < dc_bw_params->dc_mode_limit.dppclk_mhz) { - dml_clk_table->dppclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dppclk_mhz * 1000; - dml_clk_table->dppclk.num_clk_values = i + 1; - } else { - dml_clk_table->dppclk.clk_values_khz[i] = 0; - dml_clk_table->dppclk.num_clk_values = i; - } + /* dppclk */ + if (dc_clk_table->num_entries_per_clk.num_dppclk_levels) { + dml_clk_table->dppclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dppclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dppclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dppclk_mhz && + dc_clk_table->entries[i].dppclk_mhz > dc_bw_params->dc_mode_limit.dppclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dppclk_mhz < dc_bw_params->dc_mode_limit.dppclk_mhz) { + dml_clk_table->dppclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dppclk_mhz * 1000; + dml_clk_table->dppclk.num_clk_values = i + 1; } else { - dml_clk_table->dppclk.clk_values_khz[i] = dc_clk_table->entries[i].dppclk_mhz * 1000; + dml_clk_table->dppclk.clk_values_khz[i] = 0; + dml_clk_table->dppclk.num_clk_values = i; } } else { - dml_clk_table->dppclk.clk_values_khz[i] = 0; + dml_clk_table->dppclk.clk_values_khz[i] = dc_clk_table->entries[i].dppclk_mhz * 1000; } + } else { + dml_clk_table->dppclk.clk_values_khz[i] = 0; } } + } - /* dtbclk */ - if (dc_clk_table->num_entries_per_clk.num_dtbclk_levels) { - dml_clk_table->dtbclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dtbclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dtbclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dtbclk_mhz && - dc_clk_table->entries[i].dtbclk_mhz > dc_bw_params->dc_mode_limit.dtbclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dtbclk_mhz < dc_bw_params->dc_mode_limit.dtbclk_mhz) { - dml_clk_table->dtbclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dtbclk_mhz * 1000; - dml_clk_table->dtbclk.num_clk_values = i + 1; - } else { - dml_clk_table->dtbclk.clk_values_khz[i] = 0; - dml_clk_table->dtbclk.num_clk_values = i; - } + /* dtbclk */ + if (dc_clk_table->num_entries_per_clk.num_dtbclk_levels) { + dml_clk_table->dtbclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dtbclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dtbclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dtbclk_mhz && + dc_clk_table->entries[i].dtbclk_mhz > dc_bw_params->dc_mode_limit.dtbclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dtbclk_mhz < dc_bw_params->dc_mode_limit.dtbclk_mhz) { + dml_clk_table->dtbclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dtbclk_mhz * 1000; + dml_clk_table->dtbclk.num_clk_values = i + 1; } else { - dml_clk_table->dtbclk.clk_values_khz[i] = dc_clk_table->entries[i].dtbclk_mhz * 1000; + dml_clk_table->dtbclk.clk_values_khz[i] = 0; + dml_clk_table->dtbclk.num_clk_values = i; } } else { - dml_clk_table->dtbclk.clk_values_khz[i] = 0; + dml_clk_table->dtbclk.clk_values_khz[i] = dc_clk_table->entries[i].dtbclk_mhz * 1000; } + } else { + dml_clk_table->dtbclk.clk_values_khz[i] = 0; } } + } - /* socclk */ - if (dc_clk_table->num_entries_per_clk.num_socclk_levels) { - dml_clk_table->socclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_socclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->socclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.socclk_mhz && - dc_clk_table->entries[i].socclk_mhz > dc_bw_params->dc_mode_limit.socclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].socclk_mhz < dc_bw_params->dc_mode_limit.socclk_mhz) { - dml_clk_table->socclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.socclk_mhz * 1000; - dml_clk_table->socclk.num_clk_values = i + 1; - } else { - dml_clk_table->socclk.clk_values_khz[i] = 0; - dml_clk_table->socclk.num_clk_values = i; - } + /* socclk */ + if (dc_clk_table->num_entries_per_clk.num_socclk_levels) { + dml_clk_table->socclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_socclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->socclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.socclk_mhz && + dc_clk_table->entries[i].socclk_mhz > dc_bw_params->dc_mode_limit.socclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].socclk_mhz < dc_bw_params->dc_mode_limit.socclk_mhz) { + dml_clk_table->socclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.socclk_mhz * 1000; + dml_clk_table->socclk.num_clk_values = i + 1; } else { - dml_clk_table->socclk.clk_values_khz[i] = dc_clk_table->entries[i].socclk_mhz * 1000; + dml_clk_table->socclk.clk_values_khz[i] = 0; + dml_clk_table->socclk.num_clk_values = i; } } else { - dml_clk_table->socclk.clk_values_khz[i] = 0; + dml_clk_table->socclk.clk_values_khz[i] = dc_clk_table->entries[i].socclk_mhz * 1000; } + } else { + dml_clk_table->socclk.clk_values_khz[i] = 0; } } - - /* do not override phyclks for now */ - /* phyclk */ - // dml_clk_table->phyclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_phyclk_levels; - // for (i = 0; i < DML_MAX_CLK_TABLE_SIZE; i++) { - // dml_clk_table->phyclk.clk_values_khz[i] = dc_clk_table->entries[i].phyclk_mhz * 1000; - // } - - /* phyclk_d18 */ - // dml_clk_table->phyclk_d18.num_clk_values = dc_clk_table->num_entries_per_clk.num_phyclk_d18_levels; - // for (i = 0; i < DML_MAX_CLK_TABLE_SIZE; i++) { - // dml_clk_table->phyclk_d18.clk_values_khz[i] = dc_clk_table->entries[i].phyclk_d18_mhz * 1000; - // } - - /* phyclk_d32 */ - // dml_clk_table->phyclk_d32.num_clk_values = dc_clk_table->num_entries_per_clk.num_phyclk_d32_levels; - // for (i = 0; i < DML_MAX_CLK_TABLE_SIZE; i++) { - // dml_clk_table->phyclk_d32.clk_values_khz[i] = dc_clk_table->entries[i].phyclk_d32_mhz * 1000; - // } } +} - dml_soc_bb->dchub_refclk_mhz = in_dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000; - dml_soc_bb->dprefclk_mhz = in_dc->clk_mgr->dprefclk_khz / 1000; - dml_soc_bb->xtalclk_mhz = in_dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000; - dml_soc_bb->dispclk_dppclk_vco_speed_mhz = in_dc->clk_mgr->dentist_vco_freq_khz / 1000.0; +static void override_dml_init_with_values_from_vbios( + struct dml2_initialize_instance_in_out *dml_init, + const struct dml2_configuration_options *config, + const struct dc *in_dc) +{ + const struct clk_bw_params *dc_bw_params = in_dc->clk_mgr->bw_params; + struct dml2_soc_bb *dml_soc_bb = &dml_init->soc_bb; + struct dml2_soc_state_table *dml_clk_table = &dml_init->soc_bb.clk_table; - /* override bounding box paramters from VBIOS */ if (in_dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns > 0) dml_soc_bb->power_management_parameters.dram_clk_change_blackout_us = (in_dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns + 9) / 10; @@ -308,32 +293,120 @@ void dml21_apply_soc_bb_overrides(struct dml2_initialize_instance_in_out *dml_in dml_clk_table->dram_config.channel_width_bytes = in_dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; } - /* override bounding box paramters from DC config */ - if (in_dc->bb_overrides.sr_exit_time_ns) { - dml_soc_bb->power_management_parameters.stutter_exit_latency_us = - in_dc->bb_overrides.sr_exit_time_ns / 1000.0; + dml_init->soc_bb.xtalclk_mhz = in_dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000; +} + + +static void override_dml_init_with_values_from_dmub(struct dml2_initialize_instance_in_out *dml_init, + const struct dml2_configuration_options *config, + const struct dc *in_dc) +{ + /* + * TODO - There seems to be overlaps between the values overriden from + * dmub and vbios. Investigate and identify the values that DMUB needs + * to own. + */ +// const struct dmub_soc_bb_params *dmub_bb_params = +// (const struct dmub_soc_bb_params *)config->bb_from_dmub; + +// if (dmub_bb_params == NULL) +// return; + +// if (dmub_bb_params->dram_clk_change_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.dram_clk_change_blackout_us = +// (double) dmub_bb_params->dram_clk_change_blackout_ns / 1000.0; +// if (dmub_bb_params->dram_clk_change_read_only_ns > 0) +// dml_init->soc_bb.power_management_parameters.dram_clk_change_read_only_us = +// (double) dmub_bb_params->dram_clk_change_read_only_ns / 1000.0; +// if (dmub_bb_params->dram_clk_change_write_only_ns > 0) +// dml_init->soc_bb.power_management_parameters.dram_clk_change_write_only_us = +// (double) dmub_bb_params->dram_clk_change_write_only_ns / 1000.0; +// if (dmub_bb_params->fclk_change_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.fclk_change_blackout_us = +// (double) dmub_bb_params->fclk_change_blackout_ns / 1000.0; +// if (dmub_bb_params->g7_ppt_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.g7_ppt_blackout_us = +// (double) dmub_bb_params->g7_ppt_blackout_ns / 1000.0; +// if (dmub_bb_params->stutter_enter_plus_exit_latency_ns > 0) +// dml_init->soc_bb.power_management_parameters.stutter_enter_plus_exit_latency_us = +// (double) dmub_bb_params->stutter_enter_plus_exit_latency_ns / 1000.0; +// if (dmub_bb_params->stutter_exit_latency_ns > 0) +// dml_init->soc_bb.power_management_parameters.stutter_exit_latency_us = +// (double) dmub_bb_params->stutter_exit_latency_ns / 1000.0; +// if (dmub_bb_params->z8_stutter_enter_plus_exit_latency_ns > 0) +// dml_init->soc_bb.power_management_parameters.z8_stutter_enter_plus_exit_latency_us = +// (double) dmub_bb_params->z8_stutter_enter_plus_exit_latency_ns / 1000.0; +// if (dmub_bb_params->z8_stutter_exit_latency_ns > 0) +// dml_init->soc_bb.power_management_parameters.z8_stutter_exit_latency_us = +// (double) dmub_bb_params->z8_stutter_exit_latency_ns / 1000.0; +// if (dmub_bb_params->z8_min_idle_time_ns > 0) +// dml_init->soc_bb.power_management_parameters.z8_min_idle_time = +// (double) dmub_bb_params->z8_min_idle_time_ns / 1000.0; +// #ifndef TRIM_DML2_DCN6B_IP_SENSITIVE +// if (dmub_bb_params->type_b_dram_clk_change_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.lpddr5_dram_clk_change_blackout_us = +// (double) dmub_bb_params->type_b_dram_clk_change_blackout_ns / 1000.0; +// if (dmub_bb_params->type_b_ppt_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.lpddr5_ppt_blackout_us = +// (double) dmub_bb_params->type_b_ppt_blackout_ns / 1000.0; +// #else +// if (dmub_bb_params->type_b_dram_clk_change_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.type_b_dram_clk_change_blackout_us = +// (double) dmub_bb_params->type_b_dram_clk_change_blackout_ns / 1000.0; +// if (dmub_bb_params->type_b_ppt_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.type_b_ppt_blackout_us = +// (double) dmub_bb_params->type_b_ppt_blackout_ns / 1000.0; +// #endif +// if (dmub_bb_params->vmin_limit_dispclk_khz > 0) +// dml_init->soc_bb.vmin_limit.dispclk_khz = dmub_bb_params->vmin_limit_dispclk_khz; +// if (dmub_bb_params->vmin_limit_dcfclk_khz > 0) +// dml_init->soc_bb.vmin_limit.dcfclk_khz = dmub_bb_params->vmin_limit_dcfclk_khz; +// if (dmub_bb_params->g7_temperature_read_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.g7_temperature_read_blackout_us = +// (double) dmub_bb_params->g7_temperature_read_blackout_ns / 1000.0; +} + +static void override_dml_init_with_values_from_software_policy(struct dml2_initialize_instance_in_out *dml_init, + const struct dml2_configuration_options *config, + const struct dc *in_dc) +{ + if (!config->use_native_soc_bb_construction) { + dml_init->soc_bb = config->external_socbb_ip_params->soc_bb; + dml_init->ip_caps = config->external_socbb_ip_params->ip_params; } - if (in_dc->bb_overrides.sr_enter_plus_exit_time_ns) { - dml_soc_bb->power_management_parameters.stutter_enter_plus_exit_latency_us = + if (in_dc->bb_overrides.sr_exit_time_ns) + dml_init->soc_bb.power_management_parameters.stutter_exit_latency_us = + in_dc->bb_overrides.sr_exit_time_ns / 1000.0; + + if (in_dc->bb_overrides.sr_enter_plus_exit_time_ns) + dml_init->soc_bb.power_management_parameters.stutter_enter_plus_exit_latency_us = in_dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; - } - if (in_dc->bb_overrides.dram_clock_change_latency_ns) { - dml_soc_bb->power_management_parameters.dram_clk_change_blackout_us = + if (in_dc->bb_overrides.dram_clock_change_latency_ns) + dml_init->soc_bb.power_management_parameters.dram_clk_change_blackout_us = in_dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - } - if (in_dc->bb_overrides.fclk_clock_change_latency_ns) { - dml_soc_bb->power_management_parameters.fclk_change_blackout_us = + if (in_dc->bb_overrides.fclk_clock_change_latency_ns) + dml_init->soc_bb.power_management_parameters.fclk_change_blackout_us = in_dc->bb_overrides.fclk_clock_change_latency_ns / 1000.0; - } +} - //TODO - // if (in_dc->bb_overrides.dummy_clock_change_latency_ns) { - // dml_soc_bb->power_management_parameters.dram_clk_change_blackout_us = - // in_dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - // } +void dml21_populate_dml_init_params(struct dml2_initialize_instance_in_out *dml_init, + const struct dml2_configuration_options *config, + const struct dc *in_dc) +{ + populate_default_dml_init_params(dml_init, config, in_dc); + + override_dml_init_with_values_from_hardware_default(dml_init, config, in_dc); + + override_dml_init_with_values_from_smu(dml_init, config, in_dc); + + override_dml_init_with_values_from_vbios(dml_init, config, in_dc); + + override_dml_init_with_values_from_dmub(dml_init, config, in_dc); + + override_dml_init_with_values_from_software_policy(dml_init, config, in_dc); } static unsigned int calc_max_hardware_v_total(const struct dc_stream_state *stream) @@ -726,7 +799,6 @@ static void populate_dml21_surface_config_from_plane_state( switch (plane_state->tiling_info.gfxversion) { case DcGfxVersion7: case DcGfxVersion8: - // Placeholder for programming the array_mode break; case DcGfxVersion9: case DcGfxVersion10: @@ -788,6 +860,7 @@ static void populate_dml21_plane_config_from_plane_state(struct dml2_context *dm plane->pixel_format = dml2_420_10; break; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F: case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F: plane->pixel_format = dml2_444_64; @@ -888,10 +961,8 @@ static void populate_dml21_plane_config_from_plane_state(struct dml2_context *dm case DC_CM2_GPU_MEM_SIZE_171717: plane->tdlut.tdlut_width_mode = dml2_tdlut_width_17_cube; break; - case DC_CM2_GPU_MEM_SIZE_333333: - plane->tdlut.tdlut_width_mode = dml2_tdlut_width_33_cube; - break; case DC_CM2_GPU_MEM_SIZE_TRANSFORMED: + default: //plane->tdlut.tdlut_width_mode = dml2_tdlut_width_flatten; // dml2_tdlut_width_flatten undefined break; } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h index 73a013be1e48..9880d3e0398e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h @@ -17,9 +17,7 @@ struct dml2_context; struct dml2_configuration_options; struct dml2_initialize_instance_in_out; -void dml21_apply_soc_bb_overrides(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc); -void dml21_initialize_soc_bb_params(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc); -void dml21_initialize_ip_params(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc); +void dml21_populate_dml_init_params(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc); bool dml21_map_dc_state_into_dml_display_cfg(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx); void dml21_copy_clocks_to_dc_state(struct dml2_context *in_ctx, struct dc_state *context); void dml21_extract_watermark_sets(const struct dc *in_dc, union dcn_watermark_set *watermarks, struct dml2_context *in_ctx); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_utils.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_utils.c index 930e86cdb88a..ee721606b883 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_utils.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_utils.c @@ -384,6 +384,7 @@ void dml21_build_fams2_programming(const struct dc *dc, /* reset fams2 data */ memset(&context->bw_ctx.bw.dcn.fams2_stream_base_params, 0, sizeof(union dmub_cmd_fams2_config) * DML2_MAX_PLANES); memset(&context->bw_ctx.bw.dcn.fams2_stream_sub_params, 0, sizeof(union dmub_cmd_fams2_config) * DML2_MAX_PLANES); + memset(&context->bw_ctx.bw.dcn.fams2_stream_sub_params_v2, 0, sizeof(union dmub_fams2_stream_static_sub_state_v2) * DML2_MAX_PLANES); memset(&context->bw_ctx.bw.dcn.fams2_global_config, 0, sizeof(struct dmub_cmd_fams2_global_config)); if (dml_ctx->v21.mode_programming.programming->fams2_required) { @@ -414,9 +415,16 @@ void dml21_build_fams2_programming(const struct dc *dc, memcpy(static_base_state, &dml_ctx->v21.mode_programming.programming->stream_programming[dml_stream_idx].fams2_base_params, sizeof(union dmub_cmd_fams2_config)); - memcpy(static_sub_state, - &dml_ctx->v21.mode_programming.programming->stream_programming[dml_stream_idx].fams2_sub_params, - sizeof(union dmub_cmd_fams2_config)); + + if (dc->debug.fams_version.major == 3) { + memcpy(&context->bw_ctx.bw.dcn.fams2_stream_sub_params_v2[num_fams2_streams], + &dml_ctx->v21.mode_programming.programming->stream_programming[dml_stream_idx].fams2_sub_params_v2, + sizeof(union dmub_fams2_stream_static_sub_state_v2)); + } else { + memcpy(static_sub_state, + &dml_ctx->v21.mode_programming.programming->stream_programming[dml_stream_idx].fams2_sub_params, + sizeof(union dmub_cmd_fams2_config)); + } switch (dc->debug.fams_version.minor) { case 1: diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c index 208d3651b6ba..03de3cf06ae5 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c @@ -2,8 +2,6 @@ // // Copyright 2024 Advanced Micro Devices, Inc. -#include <linux/vmalloc.h> - #include "dml2_internal_types.h" #include "dml_top.h" #include "dml2_core_dcn4_calcs.h" @@ -37,15 +35,11 @@ static bool dml21_allocate_memory(struct dml2_context **dml_ctx) return true; } -static void dml21_apply_debug_options(const struct dc *in_dc, struct dml2_context *dml_ctx, const struct dml2_configuration_options *config) +static void dml21_populate_configuration_options(const struct dc *in_dc, + struct dml2_context *dml_ctx, + const struct dml2_configuration_options *config) { - bool disable_fams2; - struct dml2_pmo_options *pmo_options = &dml_ctx->v21.dml_init.options.pmo_options; - - /* ODM options */ - pmo_options->disable_dyn_odm = !config->minimize_dispclk_using_odm; - pmo_options->disable_dyn_odm_for_multi_stream = true; - pmo_options->disable_dyn_odm_for_stream_with_svp = true; + dml_ctx->config = *config; /* UCLK P-State options */ if (in_dc->debug.dml21_force_pstate_method) { @@ -55,52 +49,20 @@ static void dml21_apply_debug_options(const struct dc *in_dc, struct dml2_contex } else { dml_ctx->config.pmo.force_pstate_method_enable = false; } - - pmo_options->disable_vblank = ((in_dc->debug.dml21_disable_pstate_method_mask >> 1) & 1); - - /* NOTE: DRR and SubVP Require FAMS2 */ - disable_fams2 = !in_dc->debug.fams2_config.bits.enable; - pmo_options->disable_svp = ((in_dc->debug.dml21_disable_pstate_method_mask >> 2) & 1) || - in_dc->debug.force_disable_subvp || - disable_fams2; - pmo_options->disable_drr_clamped = ((in_dc->debug.dml21_disable_pstate_method_mask >> 3) & 1) || - disable_fams2; - pmo_options->disable_drr_var = ((in_dc->debug.dml21_disable_pstate_method_mask >> 4) & 1) || - disable_fams2; - pmo_options->disable_fams2 = disable_fams2; - - pmo_options->disable_drr_var_when_var_active = in_dc->debug.disable_fams_gaming == INGAME_FAMS_DISABLE || - in_dc->debug.disable_fams_gaming == INGAME_FAMS_MULTI_DISP_CLAMPED_ONLY; - pmo_options->disable_drr_clamped_when_var_active = in_dc->debug.disable_fams_gaming == INGAME_FAMS_DISABLE; } -static void dml21_init(const struct dc *in_dc, struct dml2_context **dml_ctx, const struct dml2_configuration_options *config) +static void dml21_init(const struct dc *in_dc, struct dml2_context *dml_ctx, const struct dml2_configuration_options *config) { - switch (in_dc->ctx->dce_version) { - case DCN_VERSION_4_01: - (*dml_ctx)->v21.dml_init.options.project_id = dml2_project_dcn4x_stage2_auto_drr_svp; - break; - default: - (*dml_ctx)->v21.dml_init.options.project_id = dml2_project_invalid; - } - (*dml_ctx)->architecture = dml2_architecture_21; + dml_ctx->architecture = dml2_architecture_21; - /* Store configuration options */ - (*dml_ctx)->config = *config; + dml21_populate_configuration_options(in_dc, dml_ctx, config); DC_FP_START(); - /*Initialize SOCBB and DCNIP params */ - dml21_initialize_soc_bb_params(&(*dml_ctx)->v21.dml_init, config, in_dc); - dml21_initialize_ip_params(&(*dml_ctx)->v21.dml_init, config, in_dc); - dml21_apply_soc_bb_overrides(&(*dml_ctx)->v21.dml_init, config, in_dc); - - /* apply debug overrides */ - dml21_apply_debug_options(in_dc, *dml_ctx, config); + dml21_populate_dml_init_params(&dml_ctx->v21.dml_init, config, in_dc); - /*Initialize DML21 instance */ - dml2_initialize_instance(&(*dml_ctx)->v21.dml_init); + dml2_initialize_instance(&dml_ctx->v21.dml_init); DC_FP_END(); } @@ -111,7 +73,7 @@ bool dml21_create(const struct dc *in_dc, struct dml2_context **dml_ctx, const s if (!dml21_allocate_memory(dml_ctx)) return false; - dml21_init(in_dc, dml_ctx, config); + dml21_init(in_dc, *dml_ctx, config); return true; } @@ -328,12 +290,13 @@ static bool dml21_check_mode_support(const struct dc *in_dc, struct dc_state *co return true; } -bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx, bool fast_validate) +bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx, + enum dc_validate_mode validate_mode) { bool out = false; - /* Use dml_validate_only for fast_validate path */ - if (fast_validate) + /* Use dml21_check_mode_support for DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX path */ + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) out = dml21_check_mode_support(in_dc, context, dml_ctx); else out = dml21_mode_check_and_programming(in_dc, context, dml_ctx); @@ -496,7 +459,7 @@ bool dml21_create_copy(struct dml2_context **dst_dml_ctx, return true; } -void dml21_reinit(const struct dc *in_dc, struct dml2_context **dml_ctx, const struct dml2_configuration_options *config) +void dml21_reinit(const struct dc *in_dc, struct dml2_context *dml_ctx, const struct dml2_configuration_options *config) { dml21_init(in_dc, dml_ctx, config); } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h index 42e715024bc9..15f92029d2e5 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h @@ -14,6 +14,7 @@ struct dc; struct dc_state; struct dml2_configuration_options; struct dml2_context; +enum dc_validate_mode; /** * dml2_create - Creates dml21_context. @@ -33,22 +34,23 @@ void dml21_copy(struct dml2_context *dst_dml_ctx, struct dml2_context *src_dml_ctx); bool dml21_create_copy(struct dml2_context **dst_dml_ctx, struct dml2_context *src_dml_ctx); -void dml21_reinit(const struct dc *in_dc, struct dml2_context **dml_ctx, const struct dml2_configuration_options *config); +void dml21_reinit(const struct dc *in_dc, struct dml2_context *dml_ctx, const struct dml2_configuration_options *config); /** * dml21_validate - Determines if a display configuration is supported or not. * @in_dc: dc. * @context: dc_state to be validated. - * @fast_validate: Fast validate will not populate context.res_ctx. + * @validate_mode: DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX + * will not populate context.res_ctx. * * Based on fast_validate option internally would call: * - * -dml21_mode_check_and_programming - for non fast_validate option + * -dml21_mode_check_and_programming - for DC_VALIDATE_MODE_AND_PROGRAMMING option * Calculates if dc_state can be supported on the input display * configuration. If supported, generates the necessary HW * programming for the new dc_state. * - * -dml21_check_mode_support - for fast_validate option + * -dml21_check_mode_support - for DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX option * Calculates if dc_state can be supported for the input display * config. @@ -56,7 +58,8 @@ void dml21_reinit(const struct dc *in_dc, struct dml2_context **dml_ctx, const s * separate dc_states for validation. * Return: True if mode is supported, false otherwise. */ -bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx, bool fast_validate); +bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx, + enum dc_validate_mode validate_mode); /* Prepare hubp mcache_regs for hubp mcache ID and split coordinate programming */ void dml21_prepare_mcache_programming(struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h index c047d56527c4..a64ec4dcf11a 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h @@ -43,5 +43,4 @@ bool dml2_build_mode_programming(struct dml2_build_mode_programming_in_out *in_o */ bool dml2_build_mcache_programming(struct dml2_build_mcache_programming_in_out *in_out); - #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h index 84c90050668c..b05030926ce8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h @@ -46,6 +46,7 @@ struct dml2_display_dlg_regs { uint32_t dst_y_delta_drq_limit; uint32_t refcyc_per_vm_dmdata; uint32_t dmdata_dl_delta; + uint32_t dst_y_svp_drq_limit; // MRQ uint32_t refcyc_per_meta_chunk_vblank_l; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h index 255f05de362c..e8dc6471c0be 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h @@ -222,6 +222,7 @@ struct dml2_composition_cfg { struct { bool enabled; + bool upsp_enabled; struct { double h_ratio; double v_ratio; @@ -426,6 +427,7 @@ struct dml2_stream_parameters { struct dml2_display_cfg { bool gpuvm_enable; + bool ffbm_enable; bool hostvm_enable; // Allocate DET proportionally between streams based on pixel rate diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h index 5f0bc42d1d2f..8c9f414aa6bf 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h @@ -93,12 +93,15 @@ struct dml2_soc_power_management_parameters { double dram_clk_change_write_only_us; double fclk_change_blackout_us; double g7_ppt_blackout_us; + double g7_temperature_read_blackout_us; double stutter_enter_plus_exit_latency_us; double stutter_exit_latency_us; double z8_stutter_enter_plus_exit_latency_us; double z8_stutter_exit_latency_us; double z8_min_idle_time; double g6_temp_read_blackout_us[DML_MAX_CLK_TABLE_SIZE]; + double type_b_dram_clk_change_blackout_us; + double type_b_ppt_blackout_us; }; struct dml2_clk_table { @@ -130,6 +133,7 @@ struct dml2_soc_state_table { struct dml2_soc_vmin_clock_limits { unsigned long dispclk_khz; + unsigned long dcfclk_khz; }; struct dml2_soc_bb { @@ -138,6 +142,7 @@ struct dml2_soc_bb { struct dml2_soc_power_management_parameters power_management_parameters; struct dml2_soc_vmin_clock_limits vmin_limit; + double lower_bound_bandwidth_dchub; unsigned int dprefclk_mhz; unsigned int xtalclk_mhz; unsigned int pcie_refclk_mhz; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h index 0dbf886d8926..98c0234e2f47 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h @@ -53,7 +53,9 @@ enum dml2_output_type_and_rate__rate { dml2_output_rate_hdmi_rate_6x4 = 9, dml2_output_rate_hdmi_rate_8x4 = 10, dml2_output_rate_hdmi_rate_10x4 = 11, - dml2_output_rate_hdmi_rate_12x4 = 12 + dml2_output_rate_hdmi_rate_12x4 = 12, + dml2_output_rate_hdmi_rate_16x4 = 13, + dml2_output_rate_hdmi_rate_20x4 = 14 }; struct dml2_pmo_options { @@ -279,7 +281,10 @@ struct dml2_per_stream_programming { } phantom_stream; union dmub_cmd_fams2_config fams2_base_params; - union dmub_cmd_fams2_config fams2_sub_params; + union { + union dmub_cmd_fams2_config fams2_sub_params; + union dmub_fams2_stream_static_sub_state_v2 fams2_sub_params_v2; + }; }; //----------------- @@ -674,9 +679,14 @@ struct dml2_display_cfg_programming { // unlimited # of mcache struct dml2_mcache_surface_allocation non_optimized_mcache_allocation[DML2_MAX_PLANES]; + bool failed_prefetch; + bool failed_uclk_pstate; bool failed_mcache_validation; bool failed_dpmm; bool failed_mode_programming; + bool failed_mode_programming_dcfclk; + bool failed_mode_programming_prefetch; + bool failed_mode_programming_flip; bool failed_map_watermarks; } informative; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c index c4dad7164d31..b9cff2198511 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c @@ -4685,7 +4685,10 @@ static void calculate_tdlut_setting( //the tdlut is fetched during the 2 row times of prefetch. if (p->setup_for_tdlut) { *p->tdlut_groups_per_2row_ub = (unsigned int)math_ceil2((double) *p->tdlut_bytes_per_frame / *p->tdlut_bytes_per_group, 1); - *p->tdlut_opt_time = (*p->tdlut_bytes_per_frame - p->cursor_buffer_size * 1024) / tdlut_drain_rate; + if (*p->tdlut_bytes_per_frame > p->cursor_buffer_size * 1024) + *p->tdlut_opt_time = (*p->tdlut_bytes_per_frame - p->cursor_buffer_size * 1024) / tdlut_drain_rate; + else + *p->tdlut_opt_time = 0; *p->tdlut_drain_time = p->cursor_buffer_size * 1024 / tdlut_drain_rate; *p->tdlut_bytes_to_deliver = (unsigned int) (p->cursor_buffer_size * 1024.0); } @@ -4858,7 +4861,7 @@ static double get_urgent_bandwidth_required( double ReadBandwidthChroma[], double PrefetchBandwidthLuma[], double PrefetchBandwidthChroma[], - double PrefetchBandwidthOto[], + double PrefetchBandwidthMax[], double excess_vactive_fill_bw_l[], double excess_vactive_fill_bw_c[], double cursor_bw[], @@ -4922,9 +4925,9 @@ static double get_urgent_bandwidth_required( l->vm_row_bw = NumberOfDPP[k] * prefetch_vmrow_bw[k]; l->flip_and_active_bw = l->per_plane_flip_bw[k] + ReadBandwidthLuma[k] * l->adj_factor_p0 + ReadBandwidthChroma[k] * l->adj_factor_p1 + cursor_bw[k] * l->adj_factor_cur; l->flip_and_prefetch_bw = l->per_plane_flip_bw[k] + NumberOfDPP[k] * (PrefetchBandwidthLuma[k] * l->adj_factor_p0_pre + PrefetchBandwidthChroma[k] * l->adj_factor_p1_pre) + prefetch_cursor_bw[k] * l->adj_factor_cur_pre; - l->flip_and_prefetch_bw_oto = l->per_plane_flip_bw[k] + NumberOfDPP[k] * (PrefetchBandwidthOto[k] * l->adj_factor_p0_pre + PrefetchBandwidthChroma[k] * l->adj_factor_p1_pre) + prefetch_cursor_bw[k] * l->adj_factor_cur_pre; + l->flip_and_prefetch_bw_max = l->per_plane_flip_bw[k] + NumberOfDPP[k] * (PrefetchBandwidthMax[k] * l->adj_factor_p0_pre + PrefetchBandwidthChroma[k] * l->adj_factor_p1_pre) + prefetch_cursor_bw[k] * l->adj_factor_cur_pre; l->active_and_excess_bw = (ReadBandwidthLuma[k] + excess_vactive_fill_bw_l[k]) * l->tmp_nom_adj_factor_p0 + (ReadBandwidthChroma[k] + excess_vactive_fill_bw_c[k]) * l->tmp_nom_adj_factor_p1 + dpte_row_bw[k] + meta_row_bw[k]; - surface_required_bw[k] = math_max5(l->vm_row_bw, l->flip_and_active_bw, l->flip_and_prefetch_bw, l->active_and_excess_bw, l->flip_and_prefetch_bw_oto); + surface_required_bw[k] = math_max5(l->vm_row_bw, l->flip_and_active_bw, l->flip_and_prefetch_bw, l->active_and_excess_bw, l->flip_and_prefetch_bw_max); /* export peak required bandwidth for the surface */ surface_peak_required_bw[k] = math_max2(surface_required_bw[k], surface_peak_required_bw[k]); @@ -5122,7 +5125,7 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch s->Tsw_est3 = 0.0; s->cursor_prefetch_bytes = 0; *p->prefetch_cursor_bw = 0; - *p->RequiredPrefetchBWOTO = 0.0; + *p->RequiredPrefetchBWMax = 0.0; dcc_mrq_enable = (p->dcc_enable && p->mrq_present); @@ -5353,7 +5356,7 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch * mp will fail if ms decides to use equ schedule and mp decides to use oto schedule * and the required bandwidth increases when going from ms to mp */ - *p->RequiredPrefetchBWOTO = s->prefetch_bw_oto; + *p->RequiredPrefetchBWMax = s->prefetch_bw_oto; #ifdef __DML_VBA_DEBUG__ DML_LOG_VERBOSE("DML::%s: vactive_sw_bw_l = %f\n", __func__, p->vactive_sw_bw_l); @@ -5715,8 +5718,14 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch s->TimeForFetchingVM = s->Tvm_equ; s->TimeForFetchingRowInVBlank = s->Tr0_equ; - *p->dst_y_per_vm_vblank = math_ceil2(4.0 * s->TimeForFetchingVM / s->LineTime, 1.0) / 4.0; - *p->dst_y_per_row_vblank = math_ceil2(4.0 * s->TimeForFetchingRowInVBlank / s->LineTime, 1.0) / 4.0; + *p->dst_y_per_vm_vblank = math_ceil2(4.0 * s->TimeForFetchingVM / s->LineTime, 1.0) / 4.0; + *p->dst_y_per_row_vblank = math_ceil2(4.0 * s->TimeForFetchingRowInVBlank / s->LineTime, 1.0) / 4.0; + + /* equ bw should be propagated so a ceiling of the equ bw is accounted for prior to mode programming. + * Overall bandwidth may be lower when going from mode support to mode programming but final pixel data + * bandwidth may end up higher than what was calculated in mode support. + */ + *p->RequiredPrefetchBWMax = math_max2(s->prefetch_bw_equ, *p->RequiredPrefetchBWMax); #ifdef __DML_VBA_DEBUG__ DML_LOG_VERBOSE("DML::%s: Using equ bw scheduling for prefetch\n", __func__); @@ -6112,7 +6121,7 @@ static void calculate_peak_bandwidth_required( p->surface_read_bandwidth_c, l->zero_array, //PrefetchBandwidthLuma, l->zero_array, //PrefetchBandwidthChroma, - l->zero_array, //PrefetchBWOTO + l->zero_array, //PrefetchBWMax l->zero_array, l->zero_array, l->zero_array, @@ -6149,7 +6158,7 @@ static void calculate_peak_bandwidth_required( p->surface_read_bandwidth_c, l->zero_array, //PrefetchBandwidthLuma, l->zero_array, //PrefetchBandwidthChroma, - l->zero_array, //PrefetchBWOTO + l->zero_array, //PrefetchBWMax p->excess_vactive_fill_bw_l, p->excess_vactive_fill_bw_c, p->cursor_bw, @@ -6186,7 +6195,7 @@ static void calculate_peak_bandwidth_required( p->surface_read_bandwidth_c, p->prefetch_bandwidth_l, p->prefetch_bandwidth_c, - p->prefetch_bandwidth_oto, // to prevent ms/mp mismatch when oto bw > total vactive bw + p->prefetch_bandwidth_max, // to prevent ms/mp mismatches where mp prefetch bw > ms prefetch bw p->excess_vactive_fill_bw_l, p->excess_vactive_fill_bw_c, p->cursor_bw, @@ -6223,7 +6232,7 @@ static void calculate_peak_bandwidth_required( p->surface_read_bandwidth_c, p->prefetch_bandwidth_l, p->prefetch_bandwidth_c, - p->prefetch_bandwidth_oto, // to prevent ms/mp mismatch when oto bw > total vactive bw + p->prefetch_bandwidth_max, // to prevent ms/mp mismatch where mp prefetch bw > ms prefetch bw p->excess_vactive_fill_bw_l, p->excess_vactive_fill_bw_c, p->cursor_bw, @@ -6260,7 +6269,7 @@ static void calculate_peak_bandwidth_required( p->surface_read_bandwidth_c, p->prefetch_bandwidth_l, p->prefetch_bandwidth_c, - p->prefetch_bandwidth_oto, // to prevent ms/mp mismatch when oto bw > total vactive bw + p->prefetch_bandwidth_max, // to prevent ms/mp mismatches where mp prefetch bw > ms prefetch bw p->excess_vactive_fill_bw_l, p->excess_vactive_fill_bw_c, p->cursor_bw, @@ -7487,7 +7496,7 @@ static noinline_for_stack void dml_core_ms_prefetch_check(struct dml2_core_inter CalculatePrefetchSchedule_params->VRatioPrefetchC = &mode_lib->ms.VRatioPreC[k]; CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWLuma = &mode_lib->ms.RequiredPrefetchPixelDataBWLuma[k]; // prefetch_sw_bw_l CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWChroma = &mode_lib->ms.RequiredPrefetchPixelDataBWChroma[k]; // prefetch_sw_bw_c - CalculatePrefetchSchedule_params->RequiredPrefetchBWOTO = &mode_lib->ms.RequiredPrefetchBWOTO[k]; + CalculatePrefetchSchedule_params->RequiredPrefetchBWMax = &mode_lib->ms.RequiredPrefetchBWMax[k]; CalculatePrefetchSchedule_params->NotEnoughTimeForDynamicMetadata = &mode_lib->ms.NoTimeForDynamicMetadata[k]; CalculatePrefetchSchedule_params->Tno_bw = &mode_lib->ms.Tno_bw[k]; CalculatePrefetchSchedule_params->Tno_bw_flip = &mode_lib->ms.Tno_bw_flip[k]; @@ -7632,7 +7641,7 @@ static noinline_for_stack void dml_core_ms_prefetch_check(struct dml2_core_inter calculate_peak_bandwidth_params->surface_read_bandwidth_c = mode_lib->ms.vactive_sw_bw_c; calculate_peak_bandwidth_params->prefetch_bandwidth_l = mode_lib->ms.RequiredPrefetchPixelDataBWLuma; calculate_peak_bandwidth_params->prefetch_bandwidth_c = mode_lib->ms.RequiredPrefetchPixelDataBWChroma; - calculate_peak_bandwidth_params->prefetch_bandwidth_oto = mode_lib->ms.RequiredPrefetchBWOTO; + calculate_peak_bandwidth_params->prefetch_bandwidth_max = mode_lib->ms.RequiredPrefetchBWMax; calculate_peak_bandwidth_params->excess_vactive_fill_bw_l = mode_lib->ms.excess_vactive_fill_bw_l; calculate_peak_bandwidth_params->excess_vactive_fill_bw_c = mode_lib->ms.excess_vactive_fill_bw_c; calculate_peak_bandwidth_params->cursor_bw = mode_lib->ms.cursor_bw; @@ -7799,7 +7808,7 @@ static noinline_for_stack void dml_core_ms_prefetch_check(struct dml2_core_inter calculate_peak_bandwidth_params->surface_read_bandwidth_c = mode_lib->ms.vactive_sw_bw_c; calculate_peak_bandwidth_params->prefetch_bandwidth_l = mode_lib->ms.RequiredPrefetchPixelDataBWLuma; calculate_peak_bandwidth_params->prefetch_bandwidth_c = mode_lib->ms.RequiredPrefetchPixelDataBWChroma; - calculate_peak_bandwidth_params->prefetch_bandwidth_oto = mode_lib->ms.RequiredPrefetchBWOTO; + calculate_peak_bandwidth_params->prefetch_bandwidth_max = mode_lib->ms.RequiredPrefetchBWMax; calculate_peak_bandwidth_params->excess_vactive_fill_bw_l = mode_lib->ms.excess_vactive_fill_bw_l; calculate_peak_bandwidth_params->excess_vactive_fill_bw_c = mode_lib->ms.excess_vactive_fill_bw_c; calculate_peak_bandwidth_params->cursor_bw = mode_lib->ms.cursor_bw; @@ -7905,6 +7914,7 @@ static noinline_for_stack void dml_core_ms_prefetch_check(struct dml2_core_inter } + static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out_params) { struct dml2_core_internal_display_mode_lib *mode_lib = in_out_params->mode_lib; @@ -11253,7 +11263,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex CalculatePrefetchSchedule_params->VRatioPrefetchC = &mode_lib->mp.VRatioPrefetchC[k]; CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWLuma = &mode_lib->mp.RequiredPrefetchPixelDataBWLuma[k]; CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWChroma = &mode_lib->mp.RequiredPrefetchPixelDataBWChroma[k]; - CalculatePrefetchSchedule_params->RequiredPrefetchBWOTO = &s->dummy_single_array[0][k]; + CalculatePrefetchSchedule_params->RequiredPrefetchBWMax = &s->dummy_single_array[0][k]; CalculatePrefetchSchedule_params->NotEnoughTimeForDynamicMetadata = &mode_lib->mp.NotEnoughTimeForDynamicMetadata[k]; CalculatePrefetchSchedule_params->Tno_bw = &mode_lib->mp.Tno_bw[k]; CalculatePrefetchSchedule_params->Tno_bw_flip = &mode_lib->mp.Tno_bw_flip[k]; @@ -11396,7 +11406,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex calculate_peak_bandwidth_params->surface_read_bandwidth_c = mode_lib->mp.vactive_sw_bw_c; calculate_peak_bandwidth_params->prefetch_bandwidth_l = mode_lib->mp.RequiredPrefetchPixelDataBWLuma; calculate_peak_bandwidth_params->prefetch_bandwidth_c = mode_lib->mp.RequiredPrefetchPixelDataBWChroma; - calculate_peak_bandwidth_params->prefetch_bandwidth_oto = s->dummy_single_array[0]; + calculate_peak_bandwidth_params->prefetch_bandwidth_max = s->dummy_single_array[0]; calculate_peak_bandwidth_params->excess_vactive_fill_bw_l = mode_lib->mp.excess_vactive_fill_bw_l; calculate_peak_bandwidth_params->excess_vactive_fill_bw_c = mode_lib->mp.excess_vactive_fill_bw_c; calculate_peak_bandwidth_params->cursor_bw = mode_lib->mp.cursor_bw; @@ -11536,7 +11546,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex calculate_peak_bandwidth_params->meta_row_bw = mode_lib->mp.meta_row_bw; calculate_peak_bandwidth_params->prefetch_cursor_bw = mode_lib->mp.prefetch_cursor_bw; calculate_peak_bandwidth_params->prefetch_vmrow_bw = mode_lib->mp.prefetch_vmrow_bw; - calculate_peak_bandwidth_params->prefetch_bandwidth_oto = s->dummy_single_array[0]; + calculate_peak_bandwidth_params->prefetch_bandwidth_max = s->dummy_single_array[0]; calculate_peak_bandwidth_params->flip_bw = mode_lib->mp.final_flip_bw; calculate_peak_bandwidth_params->urgent_burst_factor_l = mode_lib->mp.UrgentBurstFactorLuma; calculate_peak_bandwidth_params->urgent_burst_factor_c = mode_lib->mp.UrgentBurstFactorChroma; @@ -11880,7 +11890,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex } //Maximum Bandwidth Used - s->TotalWRBandwidth = 0; + mode_lib->mp.TotalWRBandwidth = 0; for (k = 0; k < display_cfg->num_streams; ++k) { s->WRBandwidth = 0; if (display_cfg->stream_descriptors[k].writeback.active_writebacks_per_stream > 0) { @@ -11889,7 +11899,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex (display_cfg->stream_descriptors[k].timing.h_total * display_cfg->stream_descriptors[k].writeback.writeback_stream[0].input_height / ((double)display_cfg->stream_descriptors[k].timing.pixel_clock_khz / 1000)) * (display_cfg->stream_descriptors[k].writeback.writeback_stream[0].pixel_format == dml2_444_32 ? 4.0 : 8.0); - s->TotalWRBandwidth = s->TotalWRBandwidth + s->WRBandwidth; + mode_lib->mp.TotalWRBandwidth = mode_lib->mp.TotalWRBandwidth + s->WRBandwidth; } } @@ -13059,6 +13069,10 @@ void dml2_core_calcs_get_informative(const struct dml2_core_internal_display_mod out->informative.mode_support_info.OutputRate[k] = dml2_output_rate_hdmi_rate_10x4; else if (mode_lib->ms.support.OutputRate[k] == dml2_core_internal_output_rate_hdmi_rate_12x4) out->informative.mode_support_info.OutputRate[k] = dml2_output_rate_hdmi_rate_12x4; + else if (mode_lib->ms.support.OutputRate[k] == dml2_core_internal_output_rate_hdmi_rate_16x4) + out->informative.mode_support_info.OutputRate[k] = dml2_output_rate_hdmi_rate_16x4; + else if (mode_lib->ms.support.OutputRate[k] == dml2_core_internal_output_rate_hdmi_rate_20x4) + out->informative.mode_support_info.OutputRate[k] = dml2_output_rate_hdmi_rate_20x4; out->informative.mode_support_info.AlignedYPitch[k] = mode_lib->ms.support.AlignedYPitch[k]; out->informative.mode_support_info.AlignedCPitch[k] = mode_lib->ms.support.AlignedCPitch[k]; @@ -13243,7 +13257,7 @@ void dml2_core_calcs_get_informative(const struct dml2_core_internal_display_mod out->informative.misc.DisplayPipeLineDeliveryTimeLumaPrefetch[k] = mode_lib->mp.DisplayPipeLineDeliveryTimeLumaPrefetch[k]; out->informative.misc.DisplayPipeLineDeliveryTimeChromaPrefetch[k] = mode_lib->mp.DisplayPipeLineDeliveryTimeChromaPrefetch[k]; - out->informative.misc.WritebackRequiredBandwidth = mode_lib->scratch.dml_core_mode_programming_locals.TotalWRBandwidth / 1000.0; + out->informative.misc.WritebackRequiredBandwidth = mode_lib->mp.TotalWRBandwidth / 1000.0; out->informative.misc.WritebackAllowDRAMClockChangeEndPosition[k] = mode_lib->mp.WritebackAllowDRAMClockChangeEndPosition[k]; out->informative.misc.WritebackAllowFCLKChangeEndPosition[k] = mode_lib->mp.WritebackAllowFCLKChangeEndPosition[k]; out->informative.misc.DSCCLK_calculated[k] = mode_lib->mp.DSCCLK[k]; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h index bdee6ad7bc59..28687565ac22 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h @@ -102,6 +102,7 @@ struct dml2_core_internal_DmlPipe { double DCFClkDeepSleep; unsigned int DPPPerSurface; bool ScalerEnabled; + bool UPSPEnabled; enum dml2_rotation_angle RotationAngle; bool mirrored; unsigned int ViewportHeight; @@ -186,7 +187,9 @@ enum dml2_core_internal_output_type_rate { dml2_core_internal_output_rate_hdmi_rate_6x4 = 9, dml2_core_internal_output_rate_hdmi_rate_8x4 = 10, dml2_core_internal_output_rate_hdmi_rate_10x4 = 11, - dml2_core_internal_output_rate_hdmi_rate_12x4 = 12 + dml2_core_internal_output_rate_hdmi_rate_12x4 = 12, + dml2_core_internal_output_rate_hdmi_rate_16x4 = 13, + dml2_core_internal_output_rate_hdmi_rate_20x4 = 14 }; struct dml2_core_internal_watermarks { @@ -260,12 +263,14 @@ struct dml2_core_internal_mode_support_info { bool AvgBandwidthSupport; bool UrgVactiveBandwidthSupport; bool EnoughUrgentLatencyHidingSupport; + bool PrefetchScheduleSupported; bool PrefetchSupported; bool PrefetchBandwidthSupported; bool DynamicMetadataSupported; bool VRatioInPrefetchSupported; bool DISPCLK_DPPCLK_Support; bool TotalAvailablePipesSupport; + bool ODMSupport; bool ModeSupport; bool ViewportSizeSupport; @@ -314,9 +319,7 @@ struct dml2_core_internal_mode_support_info { double non_urg_bandwidth_required[dml2_core_internal_soc_state_max][dml2_core_internal_bw_max]; // same as urg_bandwidth, except not scaled by urg burst factor double non_urg_bandwidth_required_flip[dml2_core_internal_soc_state_max][dml2_core_internal_bw_max]; - bool avg_bandwidth_support_ok[dml2_core_internal_soc_state_max][dml2_core_internal_bw_max]; - double max_urgent_latency_us; double max_non_urgent_latency_us; double avg_non_urgent_latency_us; @@ -329,6 +332,8 @@ struct dml2_core_internal_mode_support_info { bool temp_read_or_ppt_support; struct dml2_core_internal_watermarks watermarks; + bool dcfclk_support; + bool qos_bandwidth_support; }; struct dml2_core_internal_mode_support { @@ -350,9 +355,11 @@ struct dml2_core_internal_mode_support { double SOCCLK; /// <brief Basically just the clock freq at the min (or given) state double DCFCLK; /// <brief Basically just the clock freq at the min (or given) state and max combine setting double GlobalDPPCLK; /// <brief the Max DPPCLK freq out of all pipes + double GlobalDTBCLK; /// <brief the Max DTBCLK freq out of all pipes double uclk_freq_mhz; double dram_bw_mbps; double max_dram_bw_mbps; + double min_available_urgent_bandwidth_MBps; /// <brief Minimum guaranteed available urgent return bandwidth in MBps double MaxFabricClock; /// <brief Basically just the clock freq at the min (or given) state double MaxDCFCLK; /// <brief Basically just the clock freq at the min (or given) state and max combine setting @@ -394,9 +401,13 @@ struct dml2_core_internal_mode_support { double TWait[DML2_MAX_PLANES]; bool UnboundedRequestEnabled; + unsigned int compbuf_reserved_space_64b; + bool hw_debug5; unsigned int CompressedBufferSizeInkByte; double VRatioPreY[DML2_MAX_PLANES]; double VRatioPreC[DML2_MAX_PLANES]; + unsigned int req_per_swath_ub_l[DML2_MAX_PLANES]; + unsigned int req_per_swath_ub_c[DML2_MAX_PLANES]; unsigned int swath_width_luma_ub[DML2_MAX_PLANES]; unsigned int swath_width_chroma_ub[DML2_MAX_PLANES]; unsigned int RequiredSlots[DML2_MAX_PLANES]; @@ -417,8 +428,8 @@ struct dml2_core_internal_mode_support { double dst_y_prefetch[DML2_MAX_PLANES]; double LinesForVM[DML2_MAX_PLANES]; double LinesForDPTERow[DML2_MAX_PLANES]; - double SwathWidthYSingleDPP[DML2_MAX_PLANES]; - double SwathWidthCSingleDPP[DML2_MAX_PLANES]; + unsigned int SwathWidthYSingleDPP[DML2_MAX_PLANES]; + unsigned int SwathWidthCSingleDPP[DML2_MAX_PLANES]; unsigned int BytePerPixelY[DML2_MAX_PLANES]; unsigned int BytePerPixelC[DML2_MAX_PLANES]; double BytePerPixelInDETY[DML2_MAX_PLANES]; @@ -469,13 +480,58 @@ struct dml2_core_internal_mode_support { double mall_prefetch_sdp_overhead_factor[DML2_MAX_PLANES]; // overhead to the imall or phantom pipe double mall_prefetch_dram_overhead_factor[DML2_MAX_PLANES]; + bool is_using_mall_for_ss[DML2_MAX_PLANES]; + unsigned int meta_row_width_chroma[DML2_MAX_PLANES]; + unsigned int PixelPTEReqHeightC[DML2_MAX_PLANES]; + bool PTE_BUFFER_MODE[DML2_MAX_PLANES]; + unsigned int meta_req_height_chroma[DML2_MAX_PLANES]; + unsigned int meta_pte_bytes_per_frame_ub_c[DML2_MAX_PLANES]; + unsigned int dpde0_bytes_per_frame_ub_c[DML2_MAX_PLANES]; + unsigned int dpte_row_width_luma_ub[DML2_MAX_PLANES]; + unsigned int meta_req_width[DML2_MAX_PLANES]; + unsigned int meta_row_width[DML2_MAX_PLANES]; + unsigned int PixelPTEReqWidthY[DML2_MAX_PLANES]; + unsigned int dpte_row_height_linear[DML2_MAX_PLANES]; + unsigned int PTERequestSizeY[DML2_MAX_PLANES]; + unsigned int dpte_row_width_chroma_ub[DML2_MAX_PLANES]; + unsigned int PixelPTEReqWidthC[DML2_MAX_PLANES]; + unsigned int meta_pte_bytes_per_frame_ub_l[DML2_MAX_PLANES]; + unsigned int dpte_row_height_linear_chroma[DML2_MAX_PLANES]; + unsigned int PTERequestSizeC[DML2_MAX_PLANES]; + unsigned int meta_req_height[DML2_MAX_PLANES]; + unsigned int dpde0_bytes_per_frame_ub_l[DML2_MAX_PLANES]; + unsigned int meta_req_width_chroma[DML2_MAX_PLANES]; + unsigned int PixelPTEReqHeightY[DML2_MAX_PLANES]; + unsigned int BIGK_FRAGMENT_SIZE[DML2_MAX_PLANES]; + unsigned int vm_group_bytes[DML2_MAX_PLANES]; + unsigned int VReadyOffsetPix[DML2_MAX_PLANES]; + unsigned int VUpdateOffsetPix[DML2_MAX_PLANES]; + unsigned int VUpdateWidthPix[DML2_MAX_PLANES]; + double TSetup[DML2_MAX_PLANES]; + double Tdmdl_vm_raw[DML2_MAX_PLANES]; + double Tdmdl_raw[DML2_MAX_PLANES]; + unsigned int VStartupMin[DML2_MAX_PLANES]; /// <brief Minimum vstartup to meet the prefetch schedule (i.e. the prefetch solution can be found at this vstartup time); not the actual global sync vstartup pos. + double MaxActiveDRAMClockChangeLatencySupported[DML2_MAX_PLANES]; + double MaxActiveFCLKChangeLatencySupported; + // Backend bool RequiresDSC[DML2_MAX_PLANES]; bool RequiresFEC[DML2_MAX_PLANES]; double OutputBpp[DML2_MAX_PLANES]; + double DesiredOutputBpp[DML2_MAX_PLANES]; + double PixelClockBackEnd[DML2_MAX_PLANES]; unsigned int DSCDelay[DML2_MAX_PLANES]; enum dml2_core_internal_output_type OutputType[DML2_MAX_PLANES]; enum dml2_core_internal_output_type_rate OutputRate[DML2_MAX_PLANES]; + bool TotalAvailablePipesSupportNoDSC; + bool TotalAvailablePipesSupportDSC; + unsigned int NumberOfDPPNoDSC; + unsigned int NumberOfDPPDSC; + enum dml2_odm_mode ODMModeNoDSC; + enum dml2_odm_mode ODMModeDSC; + double RequiredDISPCLKPerSurfaceNoDSC; + double RequiredDISPCLKPerSurfaceDSC; + unsigned int EstimatedNumberOfDSCSlices[DML2_MAX_PLANES]; // Bandwidth Related Info double BandwidthAvailableForImmediateFlip; @@ -484,8 +540,14 @@ struct dml2_core_internal_mode_support { double WriteBandwidth[DML2_MAX_PLANES][DML2_MAX_WRITEBACK]; double RequiredPrefetchPixelDataBWLuma[DML2_MAX_PLANES]; double RequiredPrefetchPixelDataBWChroma[DML2_MAX_PLANES]; - /* oto bw should also be considered when calculating peak urgent bw to avoid situations oto/equ mismatches between ms and mp */ - double RequiredPrefetchBWOTO[DML2_MAX_PLANES]; + /* Max bandwidth calculated from prefetch schedule should be considered in addition to the pixel data bw to avoid ms/mp mismatches. + * 1. oto bw should also be considered when calculating peak urgent bw to avoid situations oto/equ mismatches between ms and mp + * + * 2. equ bandwidth needs to be considered for calculating peak urgent bw when equ schedule is used in mode support. + * Some slight difference in variables may cause the pixel data bandwidth to be higher + * even though overall equ prefetch bandwidths can be lower going from ms to mp + */ + double RequiredPrefetchBWMax[DML2_MAX_PLANES]; double cursor_bw[DML2_MAX_PLANES]; double prefetch_cursor_bw[DML2_MAX_PLANES]; double prefetch_vmrow_bw[DML2_MAX_PLANES]; @@ -538,7 +600,44 @@ struct dml2_core_internal_mode_support { bool mall_comb_mcache_c[DML2_MAX_PLANES]; bool lc_comb_mcache[DML2_MAX_PLANES]; + unsigned int vmpg_width_y[DML2_MAX_PLANES]; + unsigned int vmpg_height_y[DML2_MAX_PLANES]; + unsigned int vmpg_width_c[DML2_MAX_PLANES]; + unsigned int vmpg_height_c[DML2_MAX_PLANES]; + + unsigned int meta_row_height_luma[DML2_MAX_PLANES]; + unsigned int meta_row_height_chroma[DML2_MAX_PLANES]; + unsigned int meta_row_bytes_per_row_ub_l[DML2_MAX_PLANES]; + unsigned int meta_row_bytes_per_row_ub_c[DML2_MAX_PLANES]; + unsigned int dpte_row_bytes_per_row_l[DML2_MAX_PLANES]; + unsigned int dpte_row_bytes_per_row_c[DML2_MAX_PLANES]; + + unsigned int pstate_bytes_required_l[DML2_MAX_PLANES]; + unsigned int pstate_bytes_required_c[DML2_MAX_PLANES]; + unsigned int cursor_bytes_per_chunk[DML2_MAX_PLANES]; + unsigned int cursor_bytes_per_line[DML2_MAX_PLANES]; + + unsigned int MaximumVStartup[DML2_MAX_PLANES]; + + double HostVMInefficiencyFactor; + double HostVMInefficiencyFactorPrefetch; + + unsigned int tdlut_pte_bytes_per_frame[DML2_MAX_PLANES]; + unsigned int tdlut_bytes_per_frame[DML2_MAX_PLANES]; + unsigned int tdlut_groups_per_2row_ub[DML2_MAX_PLANES]; + double tdlut_opt_time[DML2_MAX_PLANES]; + double tdlut_drain_time[DML2_MAX_PLANES]; + unsigned int tdlut_bytes_per_group[DML2_MAX_PLANES]; + + double Tvm_trips_flip[DML2_MAX_PLANES]; + double Tr0_trips_flip[DML2_MAX_PLANES]; + double Tvm_trips_flip_rounded[DML2_MAX_PLANES]; + double Tr0_trips_flip_rounded[DML2_MAX_PLANES]; + unsigned int DSTYAfterScaler[DML2_MAX_PLANES]; + unsigned int DSTXAfterScaler[DML2_MAX_PLANES]; + + enum dml2_pstate_method pstate_switch_modes[DML2_MAX_PLANES]; }; /// @brief A mega structure that houses various info for model programming step. @@ -548,6 +647,7 @@ struct dml2_core_internal_mode_program { double FabricClock; /// <brief Basically just the clock freq at the min (or given) state //double DCFCLK; /// <brief Basically just the clock freq at the min (or given) state and max combine setting double dram_bw_mbps; + double min_available_urgent_bandwidth_MBps; /// <brief Minimum guaranteed available urgent return bandwidth in MBps double uclk_freq_mhz; unsigned int NoOfDPP[DML2_MAX_PLANES]; enum dml2_odm_mode ODMMode[DML2_MAX_PLANES]; @@ -599,6 +699,8 @@ struct dml2_core_internal_mode_program { unsigned int MacroTileHeightC[DML2_MAX_PLANES]; unsigned int MacroTileWidthY[DML2_MAX_PLANES]; unsigned int MacroTileWidthC[DML2_MAX_PLANES]; + double MaximumSwathWidthLuma[DML2_MAX_PLANES]; + double MaximumSwathWidthChroma[DML2_MAX_PLANES]; bool surf_linear128_l[DML2_MAX_PLANES]; bool surf_linear128_c[DML2_MAX_PLANES]; @@ -631,6 +733,14 @@ struct dml2_core_internal_mode_program { double UrgentBurstFactorChroma[DML2_MAX_PLANES]; double UrgentBurstFactorChromaPre[DML2_MAX_PLANES]; + double MaximumSwathWidthInLineBufferLuma; + double MaximumSwathWidthInLineBufferChroma; + + unsigned int vmpg_width_y[DML2_MAX_PLANES]; + unsigned int vmpg_height_y[DML2_MAX_PLANES]; + unsigned int vmpg_width_c[DML2_MAX_PLANES]; + unsigned int vmpg_height_c[DML2_MAX_PLANES]; + double meta_row_bw[DML2_MAX_PLANES]; unsigned int meta_row_bytes[DML2_MAX_PLANES]; unsigned int meta_req_width[DML2_MAX_PLANES]; @@ -652,7 +762,9 @@ struct dml2_core_internal_mode_program { unsigned int PTERequestSizeC[DML2_MAX_PLANES]; double TWait[DML2_MAX_PLANES]; + double Tdmdl_vm_raw[DML2_MAX_PLANES]; double Tdmdl_vm[DML2_MAX_PLANES]; + double Tdmdl_raw[DML2_MAX_PLANES]; double Tdmdl[DML2_MAX_PLANES]; double TSetup[DML2_MAX_PLANES]; unsigned int dpde0_bytes_per_frame_ub_l[DML2_MAX_PLANES]; @@ -684,6 +796,38 @@ struct dml2_core_internal_mode_program { double TCalc; unsigned int TotImmediateFlipBytes; + unsigned int MaxTotalDETInKByte; + unsigned int NomDETInKByte; + unsigned int MinCompressedBufferSizeInKByte; + double PixelClockBackEnd[DML2_MAX_PLANES]; + double OutputBpp[DML2_MAX_PLANES]; + bool dsc_enable[DML2_MAX_PLANES]; + unsigned int num_dsc_slices[DML2_MAX_PLANES]; + unsigned int meta_row_bytes_per_row_ub_l[DML2_MAX_PLANES]; + unsigned int meta_row_bytes_per_row_ub_c[DML2_MAX_PLANES]; + unsigned int dpte_row_bytes_per_row_l[DML2_MAX_PLANES]; + unsigned int dpte_row_bytes_per_row_c[DML2_MAX_PLANES]; + unsigned int cursor_bytes_per_chunk[DML2_MAX_PLANES]; + unsigned int cursor_bytes_per_line[DML2_MAX_PLANES]; + unsigned int MaxVStartupLines[DML2_MAX_PLANES]; /// <brief more like vblank for the plane's OTG + double HostVMInefficiencyFactor; + double HostVMInefficiencyFactorPrefetch; + unsigned int tdlut_pte_bytes_per_frame[DML2_MAX_PLANES]; + unsigned int tdlut_bytes_per_frame[DML2_MAX_PLANES]; + unsigned int tdlut_groups_per_2row_ub[DML2_MAX_PLANES]; + double tdlut_opt_time[DML2_MAX_PLANES]; + double tdlut_drain_time[DML2_MAX_PLANES]; + unsigned int tdlut_bytes_per_group[DML2_MAX_PLANES]; + double Tvm_trips_flip[DML2_MAX_PLANES]; + double Tr0_trips_flip[DML2_MAX_PLANES]; + double Tvm_trips_flip_rounded[DML2_MAX_PLANES]; + double Tr0_trips_flip_rounded[DML2_MAX_PLANES]; + bool immediate_flip_required; // any pipes need immediate flip + double SOCCLK; /// <brief Basically just the clock freq at the min (or given) state + double TotalWRBandwidth; + double max_urgent_latency_us; + double df_response_time_us; + // ------------------- // Output // ------------------- @@ -694,9 +838,12 @@ struct dml2_core_internal_mode_program { // Support bool UrgVactiveBandwidthSupport; + bool PrefetchScheduleSupported; + bool UrgentBandwidthSupport; bool PrefetchModeSupported; // <brief Is the prefetch mode (bandwidth and latency) supported bool ImmediateFlipSupported; bool ImmediateFlipSupportedForPipe[DML2_MAX_PLANES]; + bool dcfclk_support; // Clock double Dcfclk; @@ -788,7 +935,7 @@ struct dml2_core_internal_mode_program { // RQ registers bool PTE_BUFFER_MODE[DML2_MAX_PLANES]; unsigned int BIGK_FRAGMENT_SIZE[DML2_MAX_PLANES]; - + double VActiveLatencyHidingUs[DML2_MAX_PLANES]; unsigned int SubViewportLinesNeededInMALL[DML2_MAX_PLANES]; bool is_using_mall_for_ss[DML2_MAX_PLANES]; @@ -1001,10 +1148,10 @@ struct dml2_core_calcs_mode_programming_locals { double dummy_bw[dml2_core_internal_soc_state_max][dml2_core_internal_bw_max]; double surface_dummy_bw[dml2_core_internal_soc_state_max][dml2_core_internal_bw_max][DML2_MAX_PLANES]; double surface_dummy_bw0[dml2_core_internal_soc_state_max][dml2_core_internal_bw_max][DML2_MAX_PLANES]; - unsigned int dummy_integer_array[2][DML2_MAX_PLANES]; + unsigned int dummy_integer_array[4][DML2_MAX_PLANES]; enum dml2_output_encoder_class dummy_output_encoder_array[DML2_MAX_PLANES]; double dummy_single_array[2][DML2_MAX_PLANES]; - unsigned int dummy_long_array[4][DML2_MAX_PLANES]; + unsigned int dummy_long_array[8][DML2_MAX_PLANES]; bool dummy_boolean_array[2][DML2_MAX_PLANES]; bool dummy_boolean[2]; double dummy_single[2]; @@ -1028,7 +1175,6 @@ struct dml2_core_calcs_mode_programming_locals { double dlg_vblank_start; double LSetup; double blank_lines_remaining; - double TotalWRBandwidth; double WRBandwidth; struct dml2_core_internal_DmlPipe myPipe; double PixelClockBackEndFactor; @@ -1153,6 +1299,7 @@ struct dml2_core_calcs_CalculateVMRowAndSwath_params { unsigned int HostVMMinPageSize; unsigned int DCCMetaBufferSizeBytes; bool mrq_present; + enum dml2_pstate_method pstate_switch_modes[DML2_MAX_PLANES]; // Output bool *PTEBufferSizeNotExceeded; @@ -1389,7 +1536,7 @@ struct dml2_core_shared_get_urgent_bandwidth_required_locals { double vm_row_bw; double flip_and_active_bw; double flip_and_prefetch_bw; - double flip_and_prefetch_bw_oto; + double flip_and_prefetch_bw_max; double active_and_excess_bw; }; @@ -1418,6 +1565,7 @@ struct dml2_core_shared_CalculateFlipSchedule_locals { struct dml2_core_shared_rq_dlg_get_dlg_reg_locals { unsigned int plane_idx; + unsigned int stream_idx; enum dml2_source_format_class source_format; const struct dml2_timing_cfg *timing; bool dual_plane; @@ -1625,6 +1773,9 @@ struct dml2_core_calcs_CalculateSwathAndDETConfiguration_params { double *BytePerPixDETC; unsigned int *DPPPerSurface; bool mrq_present; + unsigned int dummy[2][DML2_MAX_PLANES]; + unsigned int swath_width_luma_ub_single_dpp[DML2_MAX_PLANES]; + unsigned int swath_width_chroma_ub_single_dpp[DML2_MAX_PLANES]; // output unsigned int *req_per_swath_ub_l; @@ -1642,6 +1793,8 @@ struct dml2_core_calcs_CalculateSwathAndDETConfiguration_params { unsigned int *DETBufferSizeC; unsigned int *full_swath_bytes_l; unsigned int *full_swath_bytes_c; + unsigned int *full_swath_bytes_single_dpp_l; + unsigned int *full_swath_bytes_single_dpp_c; bool *UnboundedRequestEnabled; unsigned int *compbuf_reserved_space_64b; unsigned int *CompressedBufferSizeInkByte; @@ -1801,7 +1954,7 @@ struct dml2_core_calcs_CalculatePrefetchSchedule_params { double *VRatioPrefetchC; double *RequiredPrefetchPixelDataBWLuma; double *RequiredPrefetchPixelDataBWChroma; - double *RequiredPrefetchBWOTO; + double *RequiredPrefetchBWMax; bool *NotEnoughTimeForDynamicMetadata; double *Tno_bw; double *Tno_bw_flip; @@ -2038,7 +2191,7 @@ struct dml2_core_calcs_calculate_peak_bandwidth_required_params { double *surface_read_bandwidth_c; double *prefetch_bandwidth_l; double *prefetch_bandwidth_c; - double *prefetch_bandwidth_oto; + double *prefetch_bandwidth_max; double *excess_vactive_fill_bw_l; double *excess_vactive_fill_bw_c; double *cursor_bw; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c index 7a220c0141c2..5f301befed16 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c @@ -464,7 +464,7 @@ bool dml2_core_utils_get_segment_horizontal_contiguous(enum dml2_swizzle_mode sw bool dml2_core_utils_is_linear(enum dml2_swizzle_mode sw_mode) { - return (sw_mode == dml2_sw_linear || sw_mode == dml2_sw_linear_256b || sw_mode == dml2_linear_64elements); + return sw_mode == dml2_sw_linear; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.c index f486b090bbfc..22969a533a7b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.c @@ -389,9 +389,6 @@ static bool map_min_clocks_to_dpm(const struct dml2_core_mode_support_result *mo if (result) result = round_up_to_next_dpm(&display_cfg->min_clocks.dcn4x.dispclk_khz, &state_table->dispclk); - if (result) - result = round_up_to_next_dpm(&display_cfg->min_clocks.dcn4x.deepsleep_dcfclk_khz, &state_table->dcfclk); - for (i = 0; i < DML2_MAX_DCN_PIPES; i++) { if (result) result = round_up_to_next_dpm(&display_cfg->plane_programming[i].min_clocks.dcn4x.dppclk_khz, &state_table->dppclk); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h index b226225103c3..611c80f4f1bf 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h @@ -10,15 +10,74 @@ #define DML_LOG_LEVEL_DEFAULT DML_LOG_LEVEL_WARN #define DML_LOG_INTERNAL(fmt, ...) dm_output_to_console(fmt, ## __VA_ARGS__) -/* ASSERT with message output */ -#define DML_ASSERT_MSG(condition, fmt, ...) \ - do { \ - if (!(condition)) { \ - DML_LOG_ERROR("DML ASSERT hit in %s line %d\n", __func__, __LINE__); \ - DML_LOG_ERROR(fmt, ## __VA_ARGS__); \ - DML_ASSERT(condition); \ - } \ - } while (0) +/* private helper macros */ +#define _BOOL_FORMAT(field) "%s", field ? "true" : "false" +#define _UINT_FORMAT(field) "%u", field +#define _INT_FORMAT(field) "%d", field +#define _DOUBLE_FORMAT(field) "%lf", field +#define _ELEMENT_FUNC "function" +#define _ELEMENT_COMP_IF "component_interface" +#define _ELEMENT_TOP_IF "top_interface" +#define _LOG_ENTRY(element) do { \ + DML_LOG_INTERNAL("<"element" name=\""); \ + DML_LOG_INTERNAL(__func__); \ + DML_LOG_INTERNAL("\">\n"); \ +} while (0) +#define _LOG_EXIT(element) DML_LOG_INTERNAL("</"element">\n") +#define _LOG_SCALAR(field, format) do { \ + DML_LOG_INTERNAL(#field" = "format(field)); \ + DML_LOG_INTERNAL("\n"); \ +} while (0) +#define _LOG_ARRAY(field, size, format) do { \ + DML_LOG_INTERNAL(#field " = ["); \ + for (int _i = 0; _i < (int) size; _i++) { \ + DML_LOG_INTERNAL(format(field[_i])); \ + if (_i + 1 == (int) size) \ + DML_LOG_INTERNAL("]\n"); \ + else \ + DML_LOG_INTERNAL(", "); \ +}} while (0) +#define _LOG_2D_ARRAY(field, size0, size1, format) do { \ + DML_LOG_INTERNAL(#field" = ["); \ + for (int _i = 0; _i < (int) size0; _i++) { \ + DML_LOG_INTERNAL("\n\t["); \ + for (int _j = 0; _j < (int) size1; _j++) { \ + DML_LOG_INTERNAL(format(field[_i][_j])); \ + if (_j + 1 == (int) size1) \ + DML_LOG_INTERNAL("]"); \ + else \ + DML_LOG_INTERNAL(", "); \ + } \ + if (_i + 1 == (int) size0) \ + DML_LOG_INTERNAL("]\n"); \ + else \ + DML_LOG_INTERNAL(", "); \ + } \ +} while (0) +#define _LOG_3D_ARRAY(field, size0, size1, size2, format) do { \ + DML_LOG_INTERNAL(#field" = ["); \ + for (int _i = 0; _i < (int) size0; _i++) { \ + DML_LOG_INTERNAL("\n\t["); \ + for (int _j = 0; _j < (int) size1; _j++) { \ + DML_LOG_INTERNAL("["); \ + for (int _k = 0; _k < (int) size2; _k++) { \ + DML_LOG_INTERNAL(format(field[_i][_j][_k])); \ + if (_k + 1 == (int) size2) \ + DML_LOG_INTERNAL("]"); \ + else \ + DML_LOG_INTERNAL(", "); \ + } \ + if (_j + 1 == (int) size1) \ + DML_LOG_INTERNAL("]"); \ + else \ + DML_LOG_INTERNAL(", "); \ + } \ + if (_i + 1 == (int) size0) \ + DML_LOG_INTERNAL("]\n"); \ + else \ + DML_LOG_INTERNAL(", "); \ + } \ +} while (0) /* fatal errors for unrecoverable DML states until a full reset */ #define DML_LOG_LEVEL_FATAL 0 @@ -28,7 +87,7 @@ #define DML_LOG_LEVEL_WARN 2 /* high level tracing of DML interfaces */ #define DML_LOG_LEVEL_INFO 3 -/* detailed tracing of DML internal components */ +/* tracing of DML internal executions */ #define DML_LOG_LEVEL_DEBUG 4 /* detailed tracing of DML calculation procedure */ #define DML_LOG_LEVEL_VERBOSE 5 @@ -37,30 +96,94 @@ #define DML_LOG_LEVEL DML_LOG_LEVEL_DEFAULT #endif /* #ifndef DML_LOG_LEVEL */ +/* public macros for DML_LOG_LEVEL_FATAL and up */ #define DML_LOG_FATAL(fmt, ...) DML_LOG_INTERNAL("[DML FATAL] " fmt, ## __VA_ARGS__) + +/* public macros for DML_LOG_LEVEL_ERROR and up */ #if DML_LOG_LEVEL >= DML_LOG_LEVEL_ERROR #define DML_LOG_ERROR(fmt, ...) DML_LOG_INTERNAL("[DML ERROR] "fmt, ## __VA_ARGS__) +#define DML_ASSERT_MSG(condition, fmt, ...) \ + do { \ + if (!(condition)) { \ + DML_LOG_ERROR("ASSERT hit in %s line %d\n", __func__, __LINE__); \ + DML_LOG_ERROR(fmt, ## __VA_ARGS__); \ + DML_ASSERT(condition); \ + } \ + } while (0) #else #define DML_LOG_ERROR(fmt, ...) ((void)0) +#define DML_ASSERT_MSG(condition, fmt, ...) ((void)0) #endif + +/* public macros for DML_LOG_LEVEL_WARN and up */ #if DML_LOG_LEVEL >= DML_LOG_LEVEL_WARN #define DML_LOG_WARN(fmt, ...) DML_LOG_INTERNAL("[DML WARN] "fmt, ## __VA_ARGS__) #else #define DML_LOG_WARN(fmt, ...) ((void)0) #endif + +/* public macros for DML_LOG_LEVEL_INFO and up */ #if DML_LOG_LEVEL >= DML_LOG_LEVEL_INFO #define DML_LOG_INFO(fmt, ...) DML_LOG_INTERNAL("[DML INFO] "fmt, ## __VA_ARGS__) +#define DML_LOG_TOP_IF_ENTER() _LOG_ENTRY(_ELEMENT_TOP_IF) +#define DML_LOG_TOP_IF_EXIT() _LOG_EXIT(_ELEMENT_TOP_IF) #else #define DML_LOG_INFO(fmt, ...) ((void)0) +#define DML_LOG_TOP_IF_ENTER() ((void)0) +#define DML_LOG_TOP_IF_EXIT() ((void)0) #endif + +/* public macros for DML_LOG_LEVEL_DEBUG and up */ #if DML_LOG_LEVEL >= DML_LOG_LEVEL_DEBUG -#define DML_LOG_DEBUG(fmt, ...) DML_LOG_INTERNAL("[DML DEBUG] "fmt, ## __VA_ARGS__) +#define DML_LOG_DEBUG(fmt, ...) DML_LOG_INTERNAL(fmt, ## __VA_ARGS__) +#define DML_LOG_COMP_IF_ENTER() _LOG_ENTRY(_ELEMENT_COMP_IF) +#define DML_LOG_COMP_IF_EXIT() _LOG_EXIT(_ELEMENT_COMP_IF) +#define DML_LOG_FUNC_ENTER() _LOG_ENTRY(_ELEMENT_FUNC) +#define DML_LOG_FUNC_EXIT() _LOG_EXIT(_ELEMENT_FUNC) +#define DML_LOG_DEBUG_BOOL(field) _LOG_SCALAR(field, _BOOL_FORMAT) +#define DML_LOG_DEBUG_UINT(field) _LOG_SCALAR(field, _UINT_FORMAT) +#define DML_LOG_DEBUG_INT(field) _LOG_SCALAR(field, _INT_FORMAT) +#define DML_LOG_DEBUG_DOUBLE(field) _LOG_SCALAR(field, _DOUBLE_FORMAT) +#define DML_LOG_DEBUG_ARRAY_BOOL(field, size) _LOG_ARRAY(field, size, _BOOL_FORMAT) +#define DML_LOG_DEBUG_ARRAY_UINT(field, size) _LOG_ARRAY(field, size, _UINT_FORMAT) +#define DML_LOG_DEBUG_ARRAY_INT(field, size) _LOG_ARRAY(field, size, _INT_FORMAT) +#define DML_LOG_DEBUG_ARRAY_DOUBLE(field, size) _LOG_ARRAY(field, size, _DOUBLE_FORMAT) +#define DML_LOG_DEBUG_2D_ARRAY_BOOL(field, size0, size1) _LOG_2D_ARRAY(field, size0, size1, _BOOL_FORMAT) +#define DML_LOG_DEBUG_2D_ARRAY_UINT(field, size0, size1) _LOG_2D_ARRAY(field, size0, size1, _UINT_FORMAT) +#define DML_LOG_DEBUG_2D_ARRAY_INT(field, size0, size1) _LOG_2D_ARRAY(field, size0, size1, _INT_FORMAT) +#define DML_LOG_DEBUG_2D_ARRAY_DOUBLE(field, size0, size1) _LOG_2D_ARRAY(field, size0, size1, _DOUBLE_FORMAT) +#define DML_LOG_DEBUG_3D_ARRAY_BOOL(field, size0, size1, size2) _LOG_3D_ARRAY(field, size0, size1, size2, _BOOL_FORMAT) +#define DML_LOG_DEBUG_3D_ARRAY_UINT(field, size0, size1, size2) _LOG_3D_ARRAY(field, size0, size1, size2, _UINT_FORMAT) +#define DML_LOG_DEBUG_3D_ARRAY_INT(field, size0, size1, size2) _LOG_3D_ARRAY(field, size0, size1, size2, _INT_FORMAT) +#define DML_LOG_DEBUG_3D_ARRAY_DOUBLE(field, size0, size1, size2) _LOG_3D_ARRAY(field, size0, size1, size2, _DOUBLE_FORMAT) #else #define DML_LOG_DEBUG(fmt, ...) ((void)0) +#define DML_LOG_COMP_IF_ENTER() ((void)0) +#define DML_LOG_COMP_IF_EXIT() ((void)0) +#define DML_LOG_FUNC_ENTER() ((void)0) +#define DML_LOG_FUNC_EXIT() ((void)0) +#define DML_LOG_DEBUG_BOOL(field) ((void)0) +#define DML_LOG_DEBUG_UINT(field) ((void)0) +#define DML_LOG_DEBUG_INT(field) ((void)0) +#define DML_LOG_DEBUG_DOUBLE(field) ((void)0) +#define DML_LOG_DEBUG_ARRAY_BOOL(field, size) ((void)0) +#define DML_LOG_DEBUG_ARRAY_UINT(field, size) ((void)0) +#define DML_LOG_DEBUG_ARRAY_INT(field, size) ((void)0) +#define DML_LOG_DEBUG_ARRAY_DOUBLE(field, size) ((void)0) +#define DML_LOG_DEBUG_2D_ARRAY_BOOL(field, size0, size1) ((void)0) +#define DML_LOG_DEBUG_2D_ARRAY_UINT(field, size0, size1) ((void)0) +#define DML_LOG_DEBUG_2D_ARRAY_INT(field, size0, size1) ((void)0) +#define DML_LOG_DEBUG_2D_ARRAY_DOUBLE(field, size0, size1) ((void)0) +#define DML_LOG_DEBUG_3D_ARRAY_BOOL(field, size0, size1, size2) ((void)0) +#define DML_LOG_DEBUG_3D_ARRAY_UINT(field, size0, size1, size2) ((void)0) +#define DML_LOG_DEBUG_3D_ARRAY_INT(field, size0, size1, size2) ((void)0) +#define DML_LOG_DEBUG_3D_ARRAY_DOUBLE(field, size0, size1, size2) ((void)0) #endif + +/* public macros for DML_LOG_LEVEL_VERBOSE */ #if DML_LOG_LEVEL >= DML_LOG_LEVEL_VERBOSE -#define DML_LOG_VERBOSE(fmt, ...) DML_LOG_INTERNAL("[DML VERBOSE] "fmt, ## __VA_ARGS__) +#define DML_LOG_VERBOSE(fmt, ...) DML_LOG_INTERNAL(fmt, ## __VA_ARGS__) #else #define DML_LOG_VERBOSE(fmt, ...) ((void)0) -#endif +#endif /* #if DML_LOG_LEVEL >= DML_LOG_LEVEL_VERBOSE */ #endif /* __DML2_DEBUG_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h index 00688b9f1df4..d52aa82283b3 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h @@ -202,6 +202,8 @@ struct dml2_core_mode_support_result { } active; unsigned int dispclk_khz; + unsigned int dpprefclk_khz; + unsigned int dtbrefclk_khz; unsigned int dcfclk_deepsleep_khz; unsigned int socclk_khz; @@ -446,13 +448,17 @@ struct dml2_core_internal_state_intermediates { }; struct dml2_core_mode_support_locals { - struct dml2_core_calcs_mode_support_ex mode_support_ex_params; + union { + struct dml2_core_calcs_mode_support_ex mode_support_ex_params; + }; struct dml2_display_cfg svp_expanded_display_cfg; struct dml2_calculate_mcache_allocation_in_out calc_mcache_allocation_params; }; struct dml2_core_mode_programming_locals { - struct dml2_core_calcs_mode_programming_ex mode_programming_ex_params; + union { + struct dml2_core_calcs_mode_programming_ex mode_programming_ex_params; + }; struct dml2_display_cfg svp_expanded_display_cfg; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c index 6b3b8803e0ae..a56e75cdf712 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c @@ -868,7 +868,7 @@ bool dml2_svp_remove_all_phantom_pipes(struct dml2_context *ctx, struct dc_state /* Conditions for setting up phantom pipes for SubVP: * 1. Not force disable SubVP - * 2. Full update (i.e. !fast_validate) + * 2. Full update (i.e. DC_VALIDATE_MODE_AND_PROGRAMMING) * 3. Enough pipes are available to support SubVP (TODO: Which pipes will use VACTIVE / VBLANK / SUBVP?) * 4. Display configuration passes validation * 5. (Config doesn't support MCLK in VACTIVE/VBLANK || dc->debug.force_subvp_mclk_switch) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c index 5de775fd8fce..3b866e876bf4 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c @@ -953,6 +953,7 @@ static void populate_dml_surface_cfg_from_plane_state(enum dml_project_id dml2_p out->SourcePixelFormat[location] = dml_420_10; break; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F: case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F: out->SourcePixelFormat[location] = dml_444_64; @@ -1188,22 +1189,6 @@ static unsigned int map_plane_to_dml_display_cfg(const struct dml2_context *dml2 return location; } -static void apply_legacy_svp_drr_settings(struct dml2_context *dml2, const struct dc_state *state, struct dml_display_cfg_st *dml_dispcfg) -{ - int i; - - if (state->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) { - ASSERT(state->stream_count == 1); - dml_dispcfg->timing.DRRDisplay[0] = true; - } else if (state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index_valid) { - - for (i = 0; i < dml_dispcfg->num_timings; i++) { - if (dml2->v20.scratch.dml_to_dc_pipe_mapping.disp_cfg_to_stream_id[i] == state->streams[state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index]->stream_id) - dml_dispcfg->timing.DRRDisplay[i] = true; - } - } -} - static void dml2_populate_pipe_to_plane_index_mapping(struct dml2_context *dml2, struct dc_state *state) { unsigned int i; @@ -1436,9 +1421,6 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_stat } } } - - if (!dml2->config.use_native_pstate_optimization) - apply_legacy_svp_drr_settings(dml2, context, dml_dispcfg); } void dml2_update_pipe_ctx_dchub_regs(struct _vcs_dpi_dml_display_rq_regs_st *rq_regs, diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c index 525b7d04bf84..0318260370ed 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c @@ -24,8 +24,6 @@ * */ -#include <linux/vmalloc.h> - #include "display_mode_core.h" #include "dml2_internal_types.h" #include "dml2_utils.h" @@ -95,12 +93,17 @@ static void map_hw_resources(struct dml2_context *dml2, static unsigned int pack_and_call_dml_mode_support_ex(struct dml2_context *dml2, const struct dml_display_cfg_st *display_cfg, - struct dml_mode_support_info_st *evaluation_info) + struct dml_mode_support_info_st *evaluation_info, + enum dc_validate_mode validate_mode) { struct dml2_wrapper_scratch *s = &dml2->v20.scratch; s->mode_support_params.mode_lib = &dml2->v20.dml_core_ctx; s->mode_support_params.in_display_cfg = display_cfg; + if (validate_mode == DC_VALIDATE_MODE_ONLY) + s->mode_support_params.in_start_state_idx = dml2->v20.dml_core_ctx.states.num_states - 1; + else + s->mode_support_params.in_start_state_idx = 0; s->mode_support_params.out_evaluation_info = evaluation_info; memset(evaluation_info, 0, sizeof(struct dml_mode_support_info_st)); @@ -112,10 +115,8 @@ static unsigned int pack_and_call_dml_mode_support_ex(struct dml2_context *dml2, static bool optimize_configuration(struct dml2_context *dml2, struct dml2_wrapper_optimize_configuration_params *p) { int unused_dpps = p->ip_params->max_num_dpp; - int i, j; - int odms_needed, refresh_rate_hz, dpps_needed, subvp_height, pstate_width_fw_delay_lines, surface_count; - int subvp_timing_to_add, new_timing_index, subvp_surface_to_add, new_surface_index; - float frame_time_sec, max_frame_time_sec; + int i; + int odms_needed; int largest_blend_and_timing = 0; bool optimization_done = false; @@ -130,79 +131,6 @@ static bool optimize_configuration(struct dml2_context *dml2, struct dml2_wrappe if (p->new_display_config != p->cur_display_config) *p->new_display_config = *p->cur_display_config; - // Optimize P-State Support - if (dml2->config.use_native_pstate_optimization) { - if (p->cur_mode_support_info->DRAMClockChangeSupport[0] == dml_dram_clock_change_unsupported) { - // Find a display with < 120Hz refresh rate with maximal refresh rate that's not already subvp - subvp_timing_to_add = -1; - subvp_surface_to_add = -1; - max_frame_time_sec = 0; - surface_count = 0; - for (i = 0; i < (int) p->cur_display_config->num_timings; i++) { - refresh_rate_hz = (int)div_u64((unsigned long long) p->cur_display_config->timing.PixelClock[i] * 1000 * 1000, - (p->cur_display_config->timing.HTotal[i] * p->cur_display_config->timing.VTotal[i])); - if (refresh_rate_hz < 120) { - // Check its upstream surfaces to see if this one could be converted to subvp. - dpps_needed = 0; - for (j = 0; j < (int) p->cur_display_config->num_surfaces; j++) { - if (p->cur_display_config->plane.BlendingAndTiming[j] == i && - p->cur_display_config->plane.UseMALLForPStateChange[j] == dml_use_mall_pstate_change_disable) { - dpps_needed += p->cur_mode_support_info->DPPPerSurface[j]; - subvp_surface_to_add = j; - surface_count++; - } - } - - if (surface_count == 1 && dpps_needed > 0 && dpps_needed <= unused_dpps) { - frame_time_sec = (float)1 / refresh_rate_hz; - if (frame_time_sec > max_frame_time_sec) { - max_frame_time_sec = frame_time_sec; - subvp_timing_to_add = i; - } - } - } - } - if (subvp_timing_to_add >= 0) { - new_timing_index = p->new_display_config->num_timings++; - new_surface_index = p->new_display_config->num_surfaces++; - // Add a phantom pipe reflecting the main pipe's timing - dml2_util_copy_dml_timing(&p->new_display_config->timing, new_timing_index, subvp_timing_to_add); - - pstate_width_fw_delay_lines = (int)(((double)(p->config->svp_pstate.subvp_fw_processing_delay_us + - p->config->svp_pstate.subvp_pstate_allow_width_us) / 1000000) * - (p->new_display_config->timing.PixelClock[subvp_timing_to_add] * 1000 * 1000) / - (double)p->new_display_config->timing.HTotal[subvp_timing_to_add]); - - subvp_height = p->cur_mode_support_info->SubViewportLinesNeededInMALL[subvp_timing_to_add] + pstate_width_fw_delay_lines; - - p->new_display_config->timing.VActive[new_timing_index] = subvp_height; - p->new_display_config->timing.VTotal[new_timing_index] = subvp_height + - p->new_display_config->timing.VTotal[subvp_timing_to_add] - p->new_display_config->timing.VActive[subvp_timing_to_add]; - - p->new_display_config->output.OutputDisabled[new_timing_index] = true; - - p->new_display_config->plane.UseMALLForPStateChange[subvp_surface_to_add] = dml_use_mall_pstate_change_sub_viewport; - - dml2_util_copy_dml_plane(&p->new_display_config->plane, new_surface_index, subvp_surface_to_add); - dml2_util_copy_dml_surface(&p->new_display_config->surface, new_surface_index, subvp_surface_to_add); - - p->new_display_config->plane.ViewportHeight[new_surface_index] = subvp_height; - p->new_display_config->plane.ViewportHeightChroma[new_surface_index] = subvp_height; - p->new_display_config->plane.ViewportStationary[new_surface_index] = false; - - p->new_display_config->plane.UseMALLForStaticScreen[new_surface_index] = dml_use_mall_static_screen_disable; - p->new_display_config->plane.UseMALLForPStateChange[new_surface_index] = dml_use_mall_pstate_change_phantom_pipe; - - p->new_display_config->plane.NumberOfCursors[new_surface_index] = 0; - - p->new_policy->ImmediateFlipRequirement[new_surface_index] = dml_immediate_flip_not_required; - - p->new_display_config->plane.BlendingAndTiming[new_surface_index] = new_timing_index; - - optimization_done = true; - } - } - } // Optimize Clocks if (!optimization_done) { @@ -226,7 +154,8 @@ static bool optimize_configuration(struct dml2_context *dml2, struct dml2_wrappe return optimization_done; } -static int calculate_lowest_supported_state_for_temp_read(struct dml2_context *dml2, struct dc_state *display_state) +static int calculate_lowest_supported_state_for_temp_read(struct dml2_context *dml2, struct dc_state *display_state, + enum dc_validate_mode validate_mode) { struct dml2_calculate_lowest_supported_state_for_temp_read_scratch *s = &dml2->v20.scratch.dml2_calculate_lowest_supported_state_for_temp_read_scratch; struct dml2_wrapper_scratch *s_global = &dml2->v20.scratch; @@ -268,7 +197,8 @@ static int calculate_lowest_supported_state_for_temp_read(struct dml2_context *d dml2->v20.dml_core_ctx.states.state_array[j].dram_clock_change_latency_us = s_global->dummy_pstate_table[i].dummy_pstate_latency_us; } - dml_result = pack_and_call_dml_mode_support_ex(dml2, &s->cur_display_config, &s->evaluation_info); + dml_result = pack_and_call_dml_mode_support_ex(dml2, &s->cur_display_config, &s->evaluation_info, + validate_mode); if (dml_result && s->evaluation_info.DRAMClockChangeSupport[0] == dml_dram_clock_change_vactive) { map_hw_resources(dml2, &s->cur_display_config, &s->evaluation_info); @@ -333,7 +263,8 @@ static bool does_configuration_meet_sw_policies(struct dml2_context *ctx, const } static bool dml_mode_support_wrapper(struct dml2_context *dml2, - struct dc_state *display_state) + struct dc_state *display_state, + enum dc_validate_mode validate_mode) { struct dml2_wrapper_scratch *s = &dml2->v20.scratch; unsigned int result = 0, i; @@ -369,7 +300,8 @@ static bool dml_mode_support_wrapper(struct dml2_context *dml2, result = pack_and_call_dml_mode_support_ex(dml2, &s->cur_display_config, - &s->mode_support_info); + &s->mode_support_info, + validate_mode); if (result) result = does_configuration_meet_sw_policies(dml2, &s->cur_display_config, &s->mode_support_info); @@ -390,7 +322,8 @@ static bool dml_mode_support_wrapper(struct dml2_context *dml2, dml2->v20.dml_core_ctx.policy = s->new_policy; optimized_result = pack_and_call_dml_mode_support_ex(dml2, &s->new_display_config, - &s->mode_support_info); + &s->mode_support_info, + validate_mode); if (optimized_result) optimized_result = does_configuration_meet_sw_policies(dml2, &s->new_display_config, &s->mode_support_info); @@ -409,7 +342,8 @@ static bool dml_mode_support_wrapper(struct dml2_context *dml2, if (!optimized_result) { result = pack_and_call_dml_mode_support_ex(dml2, &s->cur_display_config, - &s->mode_support_info); + &s->mode_support_info, + validate_mode); } } @@ -419,118 +353,7 @@ static bool dml_mode_support_wrapper(struct dml2_context *dml2, return result; } -static int find_drr_eligible_stream(struct dc_state *display_state) -{ - int i; - - for (i = 0; i < display_state->stream_count; i++) { - if (dc_state_get_stream_subvp_type(display_state, display_state->streams[i]) == SUBVP_NONE - && display_state->streams[i]->ignore_msa_timing_param) { - // Use ignore_msa_timing_param flag to identify as DRR - return i; - } - } - - return -1; -} - -static bool optimize_pstate_with_svp_and_drr(struct dml2_context *dml2, struct dc_state *display_state) -{ - struct dml2_wrapper_scratch *s = &dml2->v20.scratch; - bool pstate_optimization_done = false; - bool pstate_optimization_success = false; - bool result = false; - int drr_display_index = 0, non_svp_streams = 0; - bool force_svp = dml2->config.svp_pstate.force_enable_subvp; - - display_state->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = false; - display_state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index_valid = false; - - result = dml_mode_support_wrapper(dml2, display_state); - - if (!result) { - pstate_optimization_done = true; - } else if (s->mode_support_info.DRAMClockChangeSupport[0] != dml_dram_clock_change_unsupported && !force_svp) { - pstate_optimization_success = true; - pstate_optimization_done = true; - } - - if (display_state->stream_count == 1 && dml2->config.callbacks.can_support_mclk_switch_using_fw_based_vblank_stretch(dml2->config.callbacks.dc, display_state)) { - display_state->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = true; - - result = dml_mode_support_wrapper(dml2, display_state); - } else { - non_svp_streams = display_state->stream_count; - - while (!pstate_optimization_done) { - result = dml_mode_programming(&dml2->v20.dml_core_ctx, s->mode_support_params.out_lowest_state_idx, &s->cur_display_config, true); - - // Always try adding SVP first - if (result) - result = dml2_svp_add_phantom_pipe_to_dc_state(dml2, display_state, &s->mode_support_info); - else - pstate_optimization_done = true; - - - if (result) { - result = dml_mode_support_wrapper(dml2, display_state); - } else { - pstate_optimization_done = true; - } - - if (result) { - non_svp_streams--; - - if (s->mode_support_info.DRAMClockChangeSupport[0] != dml_dram_clock_change_unsupported) { - if (dml2_svp_validate_static_schedulability(dml2, display_state, s->mode_support_info.DRAMClockChangeSupport[0])) { - pstate_optimization_success = true; - pstate_optimization_done = true; - } else { - pstate_optimization_success = false; - pstate_optimization_done = false; - } - } else { - drr_display_index = find_drr_eligible_stream(display_state); - - // If there is only 1 remaining non SubVP pipe that is DRR, check static - // schedulability for SubVP + DRR. - if (non_svp_streams == 1 && drr_display_index >= 0) { - if (dml2_svp_drr_schedulable(dml2, display_state, &display_state->streams[drr_display_index]->timing)) { - display_state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index_valid = true; - display_state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index = drr_display_index; - result = dml_mode_support_wrapper(dml2, display_state); - } - - if (result && s->mode_support_info.DRAMClockChangeSupport[0] != dml_dram_clock_change_unsupported) { - pstate_optimization_success = true; - pstate_optimization_done = true; - } else { - pstate_optimization_success = false; - pstate_optimization_done = false; - } - } - - if (pstate_optimization_success) { - pstate_optimization_done = true; - } else { - pstate_optimization_done = false; - } - } - } - } - } - - if (!pstate_optimization_success) { - dml2_svp_remove_all_phantom_pipes(dml2, display_state); - display_state->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = false; - display_state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index_valid = false; - result = dml_mode_support_wrapper(dml2, display_state); - } - - return result; -} - -static bool call_dml_mode_support_and_programming(struct dc_state *context) +static bool call_dml_mode_support_and_programming(struct dc_state *context, enum dc_validate_mode validate_mode) { unsigned int result = 0; unsigned int min_state = 0; @@ -544,16 +367,13 @@ static bool call_dml_mode_support_and_programming(struct dc_state *context) struct dml2_wrapper_scratch *s = &dml2->v20.scratch; if (!context->streams[0]->sink->link->dc->caps.is_apu) { - min_state_for_g6_temp_read = calculate_lowest_supported_state_for_temp_read(dml2, context); + min_state_for_g6_temp_read = calculate_lowest_supported_state_for_temp_read(dml2, context, + validate_mode); ASSERT(min_state_for_g6_temp_read >= 0); } - if (!dml2->config.use_native_pstate_optimization) { - result = optimize_pstate_with_svp_and_drr(dml2, context); - } else { - result = dml_mode_support_wrapper(dml2, context); - } + result = dml_mode_support_wrapper(dml2, context, validate_mode); /* Upon trying to sett certain frequencies in FRL, min_state_for_g6_temp_read is reported as -1. This leads to an invalid value of min_state causing crashes later on. * Use the default logic for min_state only when min_state_for_g6_temp_read is a valid value. In other cases, use the value calculated by the DML directly. @@ -575,7 +395,8 @@ static bool call_dml_mode_support_and_programming(struct dc_state *context) return result; } -static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_state *context) +static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_state *context, + enum dc_validate_mode validate_mode) { struct dml2_context *dml2 = context->bw_ctx.dml2; struct dml2_wrapper_scratch *s = &dml2->v20.scratch; @@ -611,7 +432,7 @@ static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_s copy_dummy_pstate_table(s->dummy_pstate_table, in_dc->clk_mgr->bw_params->dummy_pstate_table, 4); - result = call_dml_mode_support_and_programming(context); + result = call_dml_mode_support_and_programming(context, validate_mode); /* Call map dc pipes to map the pipes based on the DML output. For correctly determining if recalculation * is required or not, the resource context needs to correctly reflect the number of active pipes. We would * only know the correct number if active pipes after dml2_map_dc_pipes is called. @@ -628,7 +449,7 @@ static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_s need_recalculation = dml2_verify_det_buffer_configuration(dml2, context, &dml2->det_helper_scratch); if (need_recalculation) { /* Engage the DML again if recalculation is required. */ - call_dml_mode_support_and_programming(context); + call_dml_mode_support_and_programming(context, validate_mode); if (!dml2->config.skip_hw_state_mapping) { dml2_map_dc_pipes(dml2, context, &s->cur_display_config, &s->dml_to_dc_pipe_mapping, in_dc->current_state); } @@ -684,7 +505,7 @@ static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_s return result; } -static bool dml2_validate_only(struct dc_state *context) +static bool dml2_validate_only(struct dc_state *context, enum dc_validate_mode validate_mode) { struct dml2_context *dml2; unsigned int result = 0; @@ -708,7 +529,8 @@ static bool dml2_validate_only(struct dc_state *context) result = pack_and_call_dml_mode_support_ex(dml2, &dml2->v20.scratch.cur_display_config, - &dml2->v20.scratch.mode_support_info); + &dml2->v20.scratch.mode_support_info, + validate_mode); if (result) result = does_configuration_meet_sw_policies(dml2, &dml2->v20.scratch.cur_display_config, &dml2->v20.scratch.mode_support_info); @@ -723,7 +545,8 @@ static void dml2_apply_debug_options(const struct dc *dc, struct dml2_context *d } } -bool dml2_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml2, bool fast_validate) +bool dml2_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml2, + enum dc_validate_mode validate_mode) { bool out = false; @@ -733,17 +556,17 @@ bool dml2_validate(const struct dc *in_dc, struct dc_state *context, struct dml2 /* DML2.1 validation path */ if (dml2->architecture == dml2_architecture_21) { - out = dml21_validate(in_dc, context, dml2, fast_validate); + out = dml21_validate(in_dc, context, dml2, validate_mode); return out; } DC_FP_START(); - /* Use dml_validate_only for fast_validate path */ - if (fast_validate) - out = dml2_validate_only(context); + /* Use dml_validate_only for DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX path */ + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) + out = dml2_validate_only(context, validate_mode); else - out = dml2_validate_and_build_resource(in_dc, context); + out = dml2_validate_and_build_resource(in_dc, context, validate_mode); DC_FP_END(); @@ -757,8 +580,8 @@ static inline struct dml2_context *dml2_allocate_memory(void) static void dml2_init(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) { - if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version == DCN_VERSION_4_01)) { - dml21_reinit(in_dc, dml2, config); + if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version >= DCN_VERSION_4_01)) { + dml21_reinit(in_dc, *dml2, config); return; } @@ -803,9 +626,7 @@ static void dml2_init(const struct dc *in_dc, const struct dml2_configuration_op bool dml2_create(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) { // TODO : Temporarily add DCN_VERSION_3_2 for N-1 validation. Remove DCN_VERSION_3_2 after N-1 validation phase is complete. - if ((in_dc->debug.using_dml21) - && (in_dc->ctx->dce_version == DCN_VERSION_4_01 - )) + if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version >= DCN_VERSION_4_01)) return dml21_create(in_dc, dml2, config); // Allocate Mode Lib Ctx @@ -874,8 +695,8 @@ void dml2_reinit(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) { - if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version == DCN_VERSION_4_01)) { - dml21_reinit(in_dc, dml2, config); + if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version >= DCN_VERSION_4_01)) { + dml21_reinit(in_dc, *dml2, config); return; } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h index 5100f269368e..c384e141cebc 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h @@ -240,7 +240,7 @@ struct dml2_configuration_options { bool use_clock_dc_limits; bool gpuvm_enable; bool force_tdlut_enable; - struct dml2_soc_bb *bb_from_dmub; + void *bb_from_dmub; }; /* @@ -272,7 +272,7 @@ void dml2_reinit(const struct dc *in_dc, * dml2_validate - Determines if a display configuration is supported or not. * @in_dc: dc. * @context: dc_state to be validated. - * @fast_validate: Fast validate will not populate context.res_ctx. + * @validate_mode: DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX will not populate context.res_ctx. * * DML1.0 compatible interface for validation. * @@ -295,7 +295,7 @@ void dml2_reinit(const struct dc *in_dc, bool dml2_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml2, - bool fast_validate); + enum dc_validate_mode validate_mode); /* * dml2_extract_dram_and_fclk_change_support - Extracts the FCLK and UCLK change support info. diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.c index 97bf26fa3573..36187f890d5d 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.c @@ -231,7 +231,7 @@ static struct dpp_funcs dcn401_dpp_funcs = { .dpp_program_regamma_pwl = NULL, .dpp_set_pre_degam = dpp3_set_pre_degam, .dpp_program_input_lut = NULL, - .dpp_full_bypass = dpp401_full_bypass, + .dpp_full_bypass = NULL, .dpp_setup = dpp401_dpp_setup, .dpp_program_degamma_pwl = NULL, .dpp_program_cm_dealpha = dpp3_program_cm_dealpha, diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h index ecaa976e1f52..5a6a861402b3 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h @@ -641,6 +641,7 @@ uint32_t ISHARP_DELTA_DATA; \ uint32_t ISHARP_DELTA_INDEX; \ uint32_t ISHARP_NLDELTA_SOFT_CLIP + struct dcn401_dpp_registers { DPP_REG_VARIABLE_LIST_DCN401; }; @@ -683,8 +684,6 @@ void dpp401_dscl_set_scaler_manual_scale( struct dpp *dpp_base, const struct scaler_data *scl_data); -void dpp401_full_bypass(struct dpp *dpp_base); - void dpp401_dpp_setup( struct dpp *dpp_base, enum surface_pixel_format format, diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c index 712aff7e17f7..7aab77b58869 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c @@ -88,30 +88,6 @@ enum dscl_mode_sel { DSCL_MODE_DSCL_BYPASS = 6 }; -void dpp401_full_bypass(struct dpp *dpp_base) -{ - struct dcn401_dpp *dpp = TO_DCN401_DPP(dpp_base); - - /* Input pixel format: ARGB8888 */ - REG_SET(CNVC_SURFACE_PIXEL_FORMAT, 0, - CNVC_SURFACE_PIXEL_FORMAT, 0x8); - - /* Zero expansion */ - REG_SET_3(FORMAT_CONTROL, 0, - CNVC_BYPASS, 0, - FORMAT_CONTROL__ALPHA_EN, 0, - FORMAT_EXPANSION_MODE, 0); - - /* COLOR_KEYER_CONTROL.COLOR_KEYER_EN = 0 this should be default */ - if (dpp->tf_mask->CM_BYPASS_EN) - REG_SET(CM_CONTROL, 0, CM_BYPASS_EN, 1); - else - REG_SET(CM_CONTROL, 0, CM_BYPASS, 1); - - /* Setting degamma bypass for now */ - REG_SET(CM_DGAM_CONTROL, 0, CM_DGAM_LUT_MODE, 0); -} - void dpp401_set_cursor_attributes( struct dpp *dpp_base, struct dc_cursor_attributes *cursor_attributes) diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c index 4222679fd4c9..af5de564faec 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c @@ -17,8 +17,6 @@ static void dsc_write_to_registers(struct display_stream_compressor *dsc, const /* Object I/F functions */ //static void dsc401_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz); //static bool dsc401_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps); -static void dsc401_wait_disconnect_pending_clear(struct display_stream_compressor *dsc); -static void dsc401_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz); static const struct dsc_funcs dcn401_dsc_funcs = { .dsc_get_enc_caps = dsc401_get_enc_caps, @@ -66,7 +64,7 @@ void dsc401_construct(struct dcn401_dsc *dsc, dsc->max_image_width = 5184; } -static void dsc401_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz) +void dsc401_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz) { int min_dsc_unit_required = (pixel_clock_100Hz + MAX_THROUGHPUT_PER_DSC_100HZ - 1) / MAX_THROUGHPUT_PER_DSC_100HZ; @@ -191,7 +189,7 @@ void dsc401_disable(struct display_stream_compressor *dsc) DSC_CLOCK_EN, 0); } -static void dsc401_wait_disconnect_pending_clear(struct display_stream_compressor *dsc) +void dsc401_wait_disconnect_pending_clear(struct display_stream_compressor *dsc) { struct dcn401_dsc *dsc401 = TO_DCN401_DSC(dsc); diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.h b/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.h index e3ca70058e64..1d927d8e83bf 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.h +++ b/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.h @@ -341,5 +341,7 @@ void dsc401_set_config(struct display_stream_compressor *dsc, const struct dsc_c void dsc401_enable(struct display_stream_compressor *dsc, int opp_pipe); void dsc401_disable(struct display_stream_compressor *dsc); void dsc401_disconnect(struct display_stream_compressor *dsc); +void dsc401_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz); +void dsc401_wait_disconnect_pending_clear(struct display_stream_compressor *dsc); #endif diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h index c7765e6f09e6..f8f991785d4f 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h @@ -666,10 +666,29 @@ struct dcn_mi_mask { DCN_HUBP_REG_FIELD_LIST(uint32_t); }; +struct dcn_fl_regs_st { + uint32_t lut_enable; + uint32_t lut_done; + uint32_t lut_addr_mode; + uint32_t lut_width; + uint32_t lut_tmz; + uint32_t lut_crossbar_sel_r; + uint32_t lut_crossbar_sel_g; + uint32_t lut_crossbar_sel_b; + uint32_t lut_addr_hi; + uint32_t lut_addr_lo; + uint32_t refcyc_3dlut_group; + uint32_t lut_fl_bias; + uint32_t lut_fl_scale; + uint32_t lut_fl_mode; + uint32_t lut_fl_format; +}; + struct dcn_hubp_state { struct _vcs_dpi_display_dlg_regs_st dlg_attr; struct _vcs_dpi_display_ttu_regs_st ttu_attr; struct _vcs_dpi_display_rq_regs_st rq_regs; + struct dcn_fl_regs_st fl_regs; uint32_t pixel_format; uint32_t inuse_addr_hi; uint32_t inuse_addr_lo; diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c index e8730cc40edb..252e862449a2 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c @@ -1186,8 +1186,10 @@ void dce110_disable_stream(struct pipe_ctx *pipe_ctx) if (dccg) { dccg->funcs->disable_symclk32_se(dccg, dp_hpo_inst); dccg->funcs->set_dpstreamclk(dccg, REFCLK, tg->inst, dp_hpo_inst); - if (dccg && dccg->funcs->set_dtbclk_dto) - dccg->funcs->set_dtbclk_dto(dccg, &dto_params); + if (!(dc->ctx->dce_version >= DCN_VERSION_3_5)) { + if (dccg && dccg->funcs->set_dtbclk_dto) + dccg->funcs->set_dtbclk_dto(dccg, &dto_params); + } } } else if (dccg && dccg->funcs->disable_symclk_se) { dccg->funcs->disable_symclk_se(dccg, stream_enc->stream_enc_inst, @@ -1225,7 +1227,7 @@ void dce110_blank_stream(struct pipe_ctx *pipe_ctx) return; if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) { - if (!link->skip_implict_edp_power_control) + if (!link->skip_implict_edp_power_control && hws) hws->funcs.edp_backlight_control(link, false); link->dc->hwss.set_abm_immediate_disable(pipe_ctx); } @@ -1379,7 +1381,7 @@ static void populate_audio_dp_link_info( } } -static void build_audio_output( +void build_audio_output( struct dc_state *state, const struct pipe_ctx *pipe_ctx, struct audio_output *audio_output) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h index 06789ac3a224..7cd8c1576988 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h @@ -110,5 +110,9 @@ void dce110_enable_dp_link_output( enum signal_type signal, enum clock_source_id clock_source, const struct dc_link_settings *link_settings); +void build_audio_output( + struct dc_state *state, + const struct pipe_ctx *pipe_ctx, + struct audio_output *audio_output); #endif /* __DC_HWSS_DCE110_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c index f9ee55998b6b..9ad29be925e2 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c @@ -327,6 +327,35 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx) } } + DTN_INFO("\n=======HUBP FL======\n"); + DTN_INFO( + "HUBP FL: Enabled Done adr_mode width tmz xbar_sel_R xbar_sel_G xbar_sel_B adr_hi adr_low REFCYC Bias Scale Mode Format\n"); + for (i = 0; i < pool->pipe_count; i++) { + struct dcn_hubp_state *s = &(TO_DCN10_HUBP(pool->hubps[i])->state); + struct dcn_fl_regs_st *fl_regs = &s->fl_regs; + + if (!s->blank_en) { + DTN_INFO("[%2d]: %5xh %6xh %5d %6d %2xh %2xh %6xh %6d %8d %8d %7d %8xh %5x %5x %5x", + pool->hubps[i]->inst, + fl_regs->lut_enable, + fl_regs->lut_done, + fl_regs->lut_addr_mode, + fl_regs->lut_width, + fl_regs->lut_tmz, + fl_regs->lut_crossbar_sel_r, + fl_regs->lut_crossbar_sel_g, + fl_regs->lut_crossbar_sel_b, + fl_regs->lut_addr_hi, + fl_regs->lut_addr_lo, + fl_regs->refcyc_3dlut_group, + fl_regs->lut_fl_bias, + fl_regs->lut_fl_scale, + fl_regs->lut_fl_mode, + fl_regs->lut_fl_format); + DTN_INFO("\n"); + } + } + DTN_INFO("\n=========RQ========\n"); DTN_INFO("HUBP: drq_exp_m prq_exp_m mrq_exp_m crq_exp_m plane1_ba L:chunk_s min_chu_s meta_ch_s" " min_m_c_s dpte_gr_s mpte_gr_s swath_hei pte_row_h C:chunk_s min_chu_s meta_ch_s" @@ -511,6 +540,36 @@ static void dcn10_log_color_state(struct dc *dc, dc->caps.color.mpc.num_3dluts, dc->caps.color.mpc.ogam_ram, dc->caps.color.mpc.ocsc); + DTN_INFO("===== MPC RMCM 3DLUT =====\n"); + DTN_INFO("MPCC: SIZE MODE MODE_CUR RD_SEL 30BIT_EN WR_EN_MASK RAM_SEL OUT_NORM_FACTOR FL_SEL OUT_OFFSET OUT_SCALE FL_DONE SOFT_UNDERFLOW HARD_UNDERFLOW MEM_PWR_ST FORCE DIS MODE\n"); + for (i = 0; i < pool->mpcc_count; i++) { + struct mpcc_state s = {0}; + + pool->mpc->funcs->read_mpcc_state(pool->mpc, i, &s); + if (s.opp_id != 0xf) + DTN_INFO("[%2d]: %4xh %4xh %6xh %4x %4x %4x %4x %4x %4xh %4xh %6xh %4x %4x %4x %4x %4x %4x %4x\n", + i, s.rmcm_regs.rmcm_3dlut_size, s.rmcm_regs.rmcm_3dlut_mode, s.rmcm_regs.rmcm_3dlut_mode_cur, + s.rmcm_regs.rmcm_3dlut_read_sel, s.rmcm_regs.rmcm_3dlut_30bit_en, s.rmcm_regs.rmcm_3dlut_wr_en_mask, + s.rmcm_regs.rmcm_3dlut_ram_sel, s.rmcm_regs.rmcm_3dlut_out_norm_factor, s.rmcm_regs.rmcm_3dlut_fl_sel, + s.rmcm_regs.rmcm_3dlut_out_offset_r, s.rmcm_regs.rmcm_3dlut_out_scale_r, s.rmcm_regs.rmcm_3dlut_fl_done, + s.rmcm_regs.rmcm_3dlut_fl_soft_underflow, s.rmcm_regs.rmcm_3dlut_fl_hard_underflow, s.rmcm_regs.rmcm_3dlut_mem_pwr_state, + s.rmcm_regs.rmcm_3dlut_mem_pwr_force, s.rmcm_regs.rmcm_3dlut_mem_pwr_dis, s.rmcm_regs.rmcm_3dlut_mem_pwr_mode); + } + DTN_INFO("\n"); + DTN_INFO("===== MPC RMCM Shaper =====\n"); + DTN_INFO("MPCC: CNTL LUT_MODE MODE_CUR WR_EN_MASK WR_SEL OFFSET SCALE START_B START_SEG_B END_B END_BASE_B MEM_PWR_ST FORCE DIS MODE\n"); + for (i = 0; i < pool->mpcc_count; i++) { + struct mpcc_state s = {0}; + + pool->mpc->funcs->read_mpcc_state(pool->mpc, i, &s); + if (s.opp_id != 0xf) + DTN_INFO("[%2d]: %4xh %4xh %6xh %4x %4x %4x %4x %4x %4xh %4xh %6xh %4x %4x %4x %4x\n", + i, s.rmcm_regs.rmcm_cntl, s.rmcm_regs.rmcm_shaper_lut_mode, s.rmcm_regs.rmcm_shaper_mode_cur, + s.rmcm_regs.rmcm_shaper_lut_write_en_mask, s.rmcm_regs.rmcm_shaper_lut_write_sel, s.rmcm_regs.rmcm_shaper_offset_b, + s.rmcm_regs.rmcm_shaper_scale_b, s.rmcm_regs.rmcm_shaper_rama_exp_region_start_b, s.rmcm_regs.rmcm_shaper_rama_exp_region_start_seg_b, + s.rmcm_regs.rmcm_shaper_rama_exp_region_end_b, s.rmcm_regs.rmcm_shaper_rama_exp_region_end_base_b, s.rmcm_regs.rmcm_shaper_mem_pwr_state, + s.rmcm_regs.rmcm_shaper_mem_pwr_force, s.rmcm_regs.rmcm_shaper_mem_pwr_dis, s.rmcm_regs.rmcm_shaper_mem_pwr_mode); + } } void dcn10_log_hw_state(struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c index c277df12c817..3207addbd4eb 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c @@ -283,14 +283,13 @@ void dcn20_setup_gsl_group_as_lock( } /* at this point we want to program whether it's to enable or disable */ - if (pipe_ctx->stream_res.tg->funcs->set_gsl != NULL && - pipe_ctx->stream_res.tg->funcs->set_gsl_source_select != NULL) { + if (pipe_ctx->stream_res.tg->funcs->set_gsl != NULL) { pipe_ctx->stream_res.tg->funcs->set_gsl( pipe_ctx->stream_res.tg, &gsl); - - pipe_ctx->stream_res.tg->funcs->set_gsl_source_select( - pipe_ctx->stream_res.tg, group_idx, enable ? 4 : 0); + if (pipe_ctx->stream_res.tg->funcs->set_gsl_source_select != NULL) + pipe_ctx->stream_res.tg->funcs->set_gsl_source_select( + pipe_ctx->stream_res.tg, group_idx, enable ? 4 : 0); } else BREAK_TO_DEBUGGER(); } @@ -956,7 +955,7 @@ enum dc_status dcn20_enable_stream_timing( return DC_ERROR_UNEXPECTED; } - hws->funcs.wait_for_blank_complete(pipe_ctx->stream_res.opp); + fsleep(stream->timing.v_total * (stream->timing.h_total * 10000u / stream->timing.pix_clk_100hz)); params.vertical_total_min = stream->adjust.v_total_min; params.vertical_total_max = stream->adjust.v_total_max; @@ -1971,14 +1970,6 @@ static void dcn20_program_pipe( pipe_ctx->plane_state->update_flags.bits.hdr_mult)) hws->funcs.set_hdr_multiplier(pipe_ctx); - if (hws->funcs.populate_mcm_luts) { - if (pipe_ctx->plane_state) { - hws->funcs.populate_mcm_luts(dc, pipe_ctx, pipe_ctx->plane_state->mcm_luts, - pipe_ctx->plane_state->lut_bank_a); - pipe_ctx->plane_state->lut_bank_a = !pipe_ctx->plane_state->lut_bank_a; - } - } - if (pipe_ctx->plane_state && (pipe_ctx->plane_state->update_flags.bits.in_transfer_func_change || pipe_ctx->plane_state->update_flags.bits.gamma_change || @@ -2492,7 +2483,7 @@ bool dcn20_update_bandwidth( struct dce_hwseq *hws = dc->hwseq; /* recalculate DML parameters */ - if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK) + if (dc->res_pool->funcs->validate_bandwidth(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING) != DC_OK) return false; /* apply updated bandwidth parameters */ @@ -2816,6 +2807,8 @@ void dcn20_reset_back_end_for_pipe( { struct dc_link *link = pipe_ctx->stream->link; const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); + struct dccg *dccg = dc->res_pool->dccg; + struct dtbclk_dto_params dto_params = {0}; DC_LOGGER_INIT(dc->ctx->logger); if (pipe_ctx->stream_res.stream_enc == NULL) { @@ -2876,6 +2869,13 @@ void dcn20_reset_back_end_for_pipe( &pipe_ctx->link_res, pipe_ctx->stream->signal); link->phy_state.symclk_state = SYMCLK_OFF_TX_OFF; } + if (dc->link_srv->dp_is_128b_132b_signal(pipe_ctx) && dccg + && dc->ctx->dce_version >= DCN_VERSION_3_5) { + dto_params.otg_inst = pipe_ctx->stream_res.tg->inst; + dto_params.timing = &pipe_ctx->stream->timing; + if (dccg && dccg->funcs->set_dtbclk_dto) + dccg->funcs->set_dtbclk_dto(dccg, &dto_params); + } } /* diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c index e68f21fd5f0f..a40e119d8582 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c @@ -55,15 +55,15 @@ #include "dcn20/dcn20_optc.h" #include "dcn30/dcn30_cm_common.h" -#define DC_LOGGER_INIT(logger) +#define DC_LOGGER_INIT(logger) \ + struct dal_logger *dc_logger = logger #define CTX \ hws->ctx #define REG(reg)\ hws->regs->reg #define DC_LOGGER \ - stream->ctx->logger - + dc_logger #undef FN #define FN(reg_name, field_name) \ @@ -76,6 +76,8 @@ static void update_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) struct pipe_ctx *odm_pipe; int opp_cnt = 1; + DC_LOGGER_INIT(stream->ctx->logger); + ASSERT(dsc); for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) opp_cnt++; @@ -528,3 +530,32 @@ void dcn314_disable_link_output(struct dc_link *link, apply_symclk_on_tx_off_wa(link); } + + +void dcn314_plane_atomic_power_down(struct dc *dc, + struct dpp *dpp, + struct hubp *hubp) +{ + struct dce_hwseq *hws = dc->hwseq; + DC_LOGGER_INIT(dc->ctx->logger); + + if (REG(DC_IP_REQUEST_CNTL)) { + REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); + + if (hws->funcs.dpp_pg_control) { + hws->funcs.dpp_pg_control(hws, dpp->inst, false); + dpp->funcs->dpp_reset(dpp); + } + + if (hws->funcs.hubp_pg_control) { + hws->funcs.hubp_pg_control(hws, hubp->inst, false); + hubp->funcs->hubp_reset(hubp); + } + + REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); + DC_LOG_DEBUG("Power gated front end %d\n", hubp->inst); + } + + if (hws->funcs.dpp_root_clock_control) + hws->funcs.dpp_root_clock_control(hws, dpp->inst, false); +} diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h index 2305ad282f21..12a57b79edfb 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h @@ -47,4 +47,6 @@ void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, void dcn314_disable_link_output(struct dc_link *link, const struct link_resource *link_res, enum signal_type signal); +void dcn314_plane_atomic_power_down(struct dc *dc, struct dpp *dpp, struct hubp *hubp); + #endif /* __DC_HWSS_DCN314_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c index f5112742edf9..6963d25608ac 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c @@ -137,7 +137,7 @@ static const struct hwseq_private_funcs dcn314_private_funcs = { .disable_vga = dcn20_disable_vga, .bios_golden_init = dcn10_bios_golden_init, .plane_atomic_disable = dcn20_plane_atomic_disable, - .plane_atomic_power_down = dcn10_plane_atomic_power_down, + .plane_atomic_power_down = dcn314_plane_atomic_power_down, .enable_power_gating_plane = dcn314_enable_power_gating_plane, .dpp_root_clock_control = dcn314_dpp_root_clock_control, .hubp_pg_control = dcn31_hubp_pg_control, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c index a0b05b9ef660..416b1dca3dac 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c @@ -1063,15 +1063,17 @@ void dcn32_update_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; if (should_use_dto_dscclk) - dccg->funcs->set_dto_dscclk(dccg, dsc->inst); + dccg->funcs->set_dto_dscclk(dccg, dsc->inst, dsc_cfg.dc_dsc_cfg.num_slices_h); dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg); dsc->funcs->dsc_enable(dsc, pipe_ctx->stream_res.opp->inst); for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { struct display_stream_compressor *odm_dsc = odm_pipe->stream_res.dsc; ASSERT(odm_dsc); + if (!odm_dsc) + continue; if (should_use_dto_dscclk) - dccg->funcs->set_dto_dscclk(dccg, odm_dsc->inst); + dccg->funcs->set_dto_dscclk(dccg, odm_dsc->inst, dsc_cfg.dc_dsc_cfg.num_slices_h); odm_dsc->funcs->dsc_set_config(odm_dsc, &dsc_cfg, &dsc_optc_cfg); odm_dsc->funcs->dsc_enable(odm_dsc, odm_pipe->stream_res.opp->inst); } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c index c814d957305a..a267f574b619 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c @@ -1047,6 +1047,15 @@ void dcn35_calc_blocks_to_gate(struct dc *dc, struct dc_state *context, if (dc->caps.sequential_ono) { update_state->pg_pipe_res_update[PG_HUBP][pipe_ctx->stream_res.dsc->inst] = false; update_state->pg_pipe_res_update[PG_DPP][pipe_ctx->stream_res.dsc->inst] = false; + + /* All HUBP/DPP instances must be powered if the DSC inst != HUBP inst */ + if (!pipe_ctx->top_pipe && pipe_ctx->plane_res.hubp && + pipe_ctx->plane_res.hubp->inst != pipe_ctx->stream_res.dsc->inst) { + for (j = 0; j < dc->res_pool->pipe_count; ++j) { + update_state->pg_pipe_res_update[PG_HUBP][j] = false; + update_state->pg_pipe_res_update[PG_DPP][j] = false; + } + } } } @@ -1193,6 +1202,25 @@ void dcn35_calc_blocks_to_ungate(struct dc *dc, struct dc_state *context, update_state->pg_pipe_res_update[PG_HDMISTREAM][0] = true; if (dc->caps.sequential_ono) { + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i]; + + if (new_pipe->stream_res.dsc && !new_pipe->top_pipe && + update_state->pg_pipe_res_update[PG_DSC][new_pipe->stream_res.dsc->inst]) { + update_state->pg_pipe_res_update[PG_HUBP][new_pipe->stream_res.dsc->inst] = true; + update_state->pg_pipe_res_update[PG_DPP][new_pipe->stream_res.dsc->inst] = true; + + /* All HUBP/DPP instances must be powered if the DSC inst != HUBP inst */ + if (new_pipe->plane_res.hubp && + new_pipe->plane_res.hubp->inst != new_pipe->stream_res.dsc->inst) { + for (j = 0; j < dc->res_pool->pipe_count; ++j) { + update_state->pg_pipe_res_update[PG_HUBP][j] = true; + update_state->pg_pipe_res_update[PG_DPP][j] = true; + } + } + } + } + for (i = dc->res_pool->pipe_count - 1; i >= 0; i--) { if (update_state->pg_pipe_res_update[PG_HUBP][i] && update_state->pg_pipe_res_update[PG_DPP][i]) { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c index c4177a9a662f..a0d61df07f22 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c @@ -2,6 +2,8 @@ // // Copyright 2024 Advanced Micro Devices, Inc. + +#include "os_types.h" #include "dm_services.h" #include "basics/dc_common.h" #include "dm_helpers.h" @@ -143,13 +145,8 @@ void dcn401_init_hw(struct dc *dc) dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); // mark dcmode limits present if any clock has distinct AC and DC values from SMU - dc->caps.dcmode_power_limits_present = - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dcfclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.dcfclk_mhz) || - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dispclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.dispclk_mhz) || - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dtbclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.dtbclk_mhz) || - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_fclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.fclk_mhz) || - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.memclk_mhz) || - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_socclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.socclk_mhz); + dc->caps.dcmode_power_limits_present = dc->clk_mgr->funcs->is_dc_mode_present && + dc->clk_mgr->funcs->is_dc_mode_present(dc->clk_mgr); } // Initialize the dccg @@ -396,249 +393,6 @@ static void dcn401_get_mcm_lut_xable_from_pipe_ctx(struct dc *dc, struct pipe_ct } } -static void dcn401_set_mcm_location_post_blend(struct dc *dc, struct pipe_ctx *pipe_ctx, bool bPostBlend) -{ - struct mpc *mpc = dc->res_pool->mpc; - int mpcc_id = pipe_ctx->plane_res.hubp->inst; - - if (!pipe_ctx->plane_state) - return; - - mpc->funcs->set_movable_cm_location(mpc, MPCC_MOVABLE_CM_LOCATION_BEFORE, mpcc_id); - pipe_ctx->plane_state->mcm_location = (bPostBlend) ? - MPCC_MOVABLE_CM_LOCATION_AFTER : - MPCC_MOVABLE_CM_LOCATION_BEFORE; -} - -static void dc_get_lut_mode( - enum dc_cm2_gpu_mem_layout layout, - enum hubp_3dlut_fl_mode *mode, - enum hubp_3dlut_fl_addressing_mode *addr_mode) -{ - switch (layout) { - case DC_CM2_GPU_MEM_LAYOUT_3D_SWIZZLE_LINEAR_RGB: - *mode = hubp_3dlut_fl_mode_native_1; - *addr_mode = hubp_3dlut_fl_addressing_mode_sw_linear; - break; - case DC_CM2_GPU_MEM_LAYOUT_3D_SWIZZLE_LINEAR_BGR: - *mode = hubp_3dlut_fl_mode_native_2; - *addr_mode = hubp_3dlut_fl_addressing_mode_sw_linear; - break; - case DC_CM2_GPU_MEM_LAYOUT_1D_PACKED_LINEAR: - *mode = hubp_3dlut_fl_mode_transform; - *addr_mode = hubp_3dlut_fl_addressing_mode_simple_linear; - break; - default: - *mode = hubp_3dlut_fl_mode_disable; - *addr_mode = hubp_3dlut_fl_addressing_mode_sw_linear; - break; - } -} - -static void dc_get_lut_format( - enum dc_cm2_gpu_mem_format dc_format, - enum hubp_3dlut_fl_format *format) -{ - switch (dc_format) { - case DC_CM2_GPU_MEM_FORMAT_16161616_UNORM_12MSB: - *format = hubp_3dlut_fl_format_unorm_12msb_bitslice; - break; - case DC_CM2_GPU_MEM_FORMAT_16161616_UNORM_12LSB: - *format = hubp_3dlut_fl_format_unorm_12lsb_bitslice; - break; - case DC_CM2_GPU_MEM_FORMAT_16161616_FLOAT_FP1_5_10: - *format = hubp_3dlut_fl_format_float_fp1_5_10; - break; - } -} - -static void dc_get_lut_xbar( - enum dc_cm2_gpu_mem_pixel_component_order order, - enum hubp_3dlut_fl_crossbar_bit_slice *cr_r, - enum hubp_3dlut_fl_crossbar_bit_slice *y_g, - enum hubp_3dlut_fl_crossbar_bit_slice *cb_b) -{ - switch (order) { - case DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_RGBA: - *cr_r = hubp_3dlut_fl_crossbar_bit_slice_32_47; - *y_g = hubp_3dlut_fl_crossbar_bit_slice_16_31; - *cb_b = hubp_3dlut_fl_crossbar_bit_slice_0_15; - break; - case DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_BGRA: - *cr_r = hubp_3dlut_fl_crossbar_bit_slice_0_15; - *y_g = hubp_3dlut_fl_crossbar_bit_slice_16_31; - *cb_b = hubp_3dlut_fl_crossbar_bit_slice_32_47; - break; - } -} - -static void dc_get_lut_width( - enum dc_cm2_gpu_mem_size size, - enum hubp_3dlut_fl_width *width) -{ - switch (size) { - case DC_CM2_GPU_MEM_SIZE_333333: - *width = hubp_3dlut_fl_width_33; - break; - case DC_CM2_GPU_MEM_SIZE_171717: - *width = hubp_3dlut_fl_width_17; - break; - case DC_CM2_GPU_MEM_SIZE_TRANSFORMED: - *width = hubp_3dlut_fl_width_transformed; - break; - } -} -static bool dc_is_rmcm_3dlut_supported(struct hubp *hubp, struct mpc *mpc) -{ - if (mpc->funcs->rmcm.update_3dlut_fast_load_select && - mpc->funcs->rmcm.program_lut_read_write_control && - hubp->funcs->hubp_program_3dlut_fl_addr && - mpc->funcs->rmcm.program_bit_depth && - hubp->funcs->hubp_program_3dlut_fl_mode && - hubp->funcs->hubp_program_3dlut_fl_addressing_mode && - hubp->funcs->hubp_program_3dlut_fl_format && - hubp->funcs->hubp_update_3dlut_fl_bias_scale && - mpc->funcs->rmcm.program_bias_scale && - hubp->funcs->hubp_program_3dlut_fl_crossbar && - hubp->funcs->hubp_program_3dlut_fl_width && - mpc->funcs->rmcm.update_3dlut_fast_load_select && - mpc->funcs->rmcm.populate_lut && - mpc->funcs->rmcm.program_lut_mode && - hubp->funcs->hubp_enable_3dlut_fl && - mpc->funcs->rmcm.enable_3dlut_fl) - return true; - - return false; -} - -bool dcn401_program_rmcm_luts( - struct hubp *hubp, - struct pipe_ctx *pipe_ctx, - enum dc_cm2_transfer_func_source lut3d_src, - struct dc_cm2_func_luts *mcm_luts, - struct mpc *mpc, - bool lut_bank_a, - int mpcc_id) -{ - struct dpp *dpp_base = pipe_ctx->plane_res.dpp; - union mcm_lut_params m_lut_params; - enum MCM_LUT_XABLE shaper_xable, lut3d_xable = MCM_LUT_DISABLE, lut1d_xable; - enum hubp_3dlut_fl_mode mode; - enum hubp_3dlut_fl_addressing_mode addr_mode; - enum hubp_3dlut_fl_format format = 0; - enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_y_g = 0; - enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cb_b = 0; - enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cr_r = 0; - enum hubp_3dlut_fl_width width = 0; - struct dc *dc = hubp->ctx->dc; - - bool bypass_rmcm_3dlut = false; - bool bypass_rmcm_shaper = false; - - dcn401_get_mcm_lut_xable_from_pipe_ctx(dc, pipe_ctx, &shaper_xable, &lut3d_xable, &lut1d_xable); - - /* 3DLUT */ - switch (lut3d_src) { - case DC_CM2_TRANSFER_FUNC_SOURCE_SYSMEM: - memset(&m_lut_params, 0, sizeof(m_lut_params)); - // Don't know what to do in this case. - //case DC_CM2_TRANSFER_FUNC_SOURCE_SYSMEM: - break; - case DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM: - dc_get_lut_width(mcm_luts->lut3d_data.gpu_mem_params.size, &width); - if (!dc_is_rmcm_3dlut_supported(hubp, mpc) || - !mpc->funcs->rmcm.is_config_supported(width)) - return false; - - //0. disable fl on mpc - mpc->funcs->update_3dlut_fast_load_select(mpc, mpcc_id, 0xF); - - //1. power down the block - mpc->funcs->rmcm.power_on_shaper_3dlut(mpc, mpcc_id, false); - - //2. program RMCM - //2a. 3dlut reg programming - mpc->funcs->rmcm.program_lut_read_write_control(mpc, MCM_LUT_3DLUT, lut_bank_a, - (!bypass_rmcm_3dlut) && lut3d_xable != MCM_LUT_DISABLE, mpcc_id); - - hubp->funcs->hubp_program_3dlut_fl_addr(hubp, - mcm_luts->lut3d_data.gpu_mem_params.addr); - - mpc->funcs->rmcm.program_bit_depth(mpc, - mcm_luts->lut3d_data.gpu_mem_params.bit_depth, mpcc_id); - - // setting native or transformed mode, - dc_get_lut_mode(mcm_luts->lut3d_data.gpu_mem_params.layout, &mode, &addr_mode); - - //these program the mcm 3dlut - hubp->funcs->hubp_program_3dlut_fl_mode(hubp, mode); - - hubp->funcs->hubp_program_3dlut_fl_addressing_mode(hubp, addr_mode); - - //seems to be only for the MCM - dc_get_lut_format(mcm_luts->lut3d_data.gpu_mem_params.format_params.format, &format); - hubp->funcs->hubp_program_3dlut_fl_format(hubp, format); - - mpc->funcs->rmcm.program_bias_scale(mpc, - mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.bias, - mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.scale, - mpcc_id); - hubp->funcs->hubp_update_3dlut_fl_bias_scale(hubp, - mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.bias, - mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.scale); - - dc_get_lut_xbar( - mcm_luts->lut3d_data.gpu_mem_params.component_order, - &crossbar_bit_slice_cr_r, - &crossbar_bit_slice_y_g, - &crossbar_bit_slice_cb_b); - - hubp->funcs->hubp_program_3dlut_fl_crossbar(hubp, - crossbar_bit_slice_cr_r, - crossbar_bit_slice_y_g, - crossbar_bit_slice_cb_b); - - mpc->funcs->rmcm.program_3dlut_size(mpc, width, mpcc_id); - - mpc->funcs->update_3dlut_fast_load_select(mpc, mpcc_id, hubp->inst); - - //2b. shaper reg programming - memset(&m_lut_params, 0, sizeof(m_lut_params)); - - if (mcm_luts->shaper->type == TF_TYPE_HWPWL) { - m_lut_params.pwl = &mcm_luts->shaper->pwl; - } else if (mcm_luts->shaper->type == TF_TYPE_DISTRIBUTED_POINTS) { - ASSERT(false); - cm_helper_translate_curve_to_hw_format( - dc->ctx, - mcm_luts->shaper, - &dpp_base->regamma_params, true); - m_lut_params.pwl = &dpp_base->regamma_params; - } - if (m_lut_params.pwl) { - mpc->funcs->rmcm.populate_lut(mpc, m_lut_params, lut_bank_a, mpcc_id); - mpc->funcs->rmcm.program_lut_mode(mpc, !bypass_rmcm_shaper, lut_bank_a, mpcc_id); - } else { - //RMCM 3dlut won't work without its shaper - return false; - } - - //3. Select the hubp connected to this RMCM - hubp->funcs->hubp_enable_3dlut_fl(hubp, true); - mpc->funcs->rmcm.enable_3dlut_fl(mpc, true, mpcc_id); - - //4. power on the block - if (m_lut_params.pwl) - mpc->funcs->rmcm.power_on_shaper_3dlut(mpc, mpcc_id, true); - - break; - default: - return false; - } - - return true; -} - void dcn401_populate_mcm_luts(struct dc *dc, struct pipe_ctx *pipe_ctx, struct dc_cm2_func_luts mcm_luts, @@ -664,25 +418,6 @@ void dcn401_populate_mcm_luts(struct dc *dc, dcn401_get_mcm_lut_xable_from_pipe_ctx(dc, pipe_ctx, &shaper_xable, &lut3d_xable, &lut1d_xable); - //MCM - setting its location (Before/After) blender - //set to post blend (true) - dcn401_set_mcm_location_post_blend( - dc, - pipe_ctx, - mcm_luts.lut3d_data.mpc_mcm_post_blend); - - //RMCM - 3dLUT+Shaper - if (mcm_luts.lut3d_data.rmcm_3dlut_enable) { - dcn401_program_rmcm_luts( - hubp, - pipe_ctx, - lut3d_src, - &mcm_luts, - mpc, - lut_bank_a, - mpcc_id); - } - /* 1D LUT */ if (mcm_luts.lut1d_func) { memset(&m_lut_params, 0, sizeof(m_lut_params)); @@ -740,15 +475,15 @@ void dcn401_populate_mcm_luts(struct dc *dc, break; case DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM: switch (mcm_luts.lut3d_data.gpu_mem_params.size) { - case DC_CM2_GPU_MEM_SIZE_333333: - width = hubp_3dlut_fl_width_33; - break; case DC_CM2_GPU_MEM_SIZE_171717: width = hubp_3dlut_fl_width_17; break; case DC_CM2_GPU_MEM_SIZE_TRANSFORMED: width = hubp_3dlut_fl_width_transformed; break; + default: + //TODO: handle default case + break; } //check for support @@ -817,11 +552,14 @@ void dcn401_populate_mcm_luts(struct dc *dc, //navi 4x has a bug and r and blue are swapped and need to be worked around here in //TODO: need to make a method for get_xbar per asic OR do the workaround in program_crossbar for 4x - dc_get_lut_xbar( - mcm_luts.lut3d_data.gpu_mem_params.component_order, - &crossbar_bit_slice_cr_r, - &crossbar_bit_slice_y_g, - &crossbar_bit_slice_cb_b); + switch (mcm_luts.lut3d_data.gpu_mem_params.component_order) { + case DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_RGBA: + default: + crossbar_bit_slice_cr_r = hubp_3dlut_fl_crossbar_bit_slice_0_15; + crossbar_bit_slice_y_g = hubp_3dlut_fl_crossbar_bit_slice_16_31; + crossbar_bit_slice_cb_b = hubp_3dlut_fl_crossbar_bit_slice_32_47; + break; + } if (hubp->funcs->hubp_program_3dlut_fl_crossbar) hubp->funcs->hubp_program_3dlut_fl_crossbar(hubp, @@ -2269,14 +2007,6 @@ void dcn401_program_pipe( pipe_ctx->plane_state->update_flags.bits.hdr_mult)) hws->funcs.set_hdr_multiplier(pipe_ctx); - if (hws->funcs.populate_mcm_luts) { - if (pipe_ctx->plane_state) { - hws->funcs.populate_mcm_luts(dc, pipe_ctx, pipe_ctx->plane_state->mcm_luts, - pipe_ctx->plane_state->lut_bank_a); - pipe_ctx->plane_state->lut_bank_a = !pipe_ctx->plane_state->lut_bank_a; - } - } - if (pipe_ctx->plane_state && (pipe_ctx->plane_state->update_flags.bits.in_transfer_func_change || pipe_ctx->plane_state->update_flags.bits.gamma_change || @@ -2651,7 +2381,7 @@ bool dcn401_update_bandwidth( struct dce_hwseq *hws = dc->hwseq; /* recalculate DML parameters */ - if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK) + if (dc->res_pool->funcs->validate_bandwidth(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING) != DC_OK) return false; /* apply updated bandwidth parameters */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h index ce65b4f6c672..781cf0efccc6 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h @@ -109,12 +109,4 @@ void dcn401_detect_pipe_changes( void dcn401_plane_atomic_power_down(struct dc *dc, struct dpp *dpp, struct hubp *hubp); -bool dcn401_program_rmcm_luts( - struct hubp *hubp, - struct pipe_ctx *pipe_ctx, - enum dc_cm2_transfer_func_source lut3d_src, - struct dc_cm2_func_luts *mcm_luts, - struct mpc *mpc, - bool lut_bank_a, - int mpcc_id); #endif /* __DC_HWSS_DCN401_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_status.h b/drivers/gpu/drm/amd/display/dc/inc/core_status.h index f3696143590c..82085d9c3f40 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_status.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_status.h @@ -59,6 +59,7 @@ enum dc_status { DC_FAIL_DP_PAYLOAD_ALLOCATION = 27, DC_FAIL_DP_LINK_BANDWIDTH = 28, DC_FAIL_HW_CURSOR_SUPPORT = 29, + DC_FAIL_DP_TUNNEL_BW_VALIDATE = 30, DC_ERROR_UNEXPECTED = -1 }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index 0cf349cafb3e..c9454fe1cd05 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -82,7 +82,7 @@ struct resource_funcs { enum dc_status (*validate_bandwidth)( struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); void (*calculate_wm_and_dlg)( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, @@ -107,7 +107,7 @@ struct resource_funcs { struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); /* * Algorithm for assigning available link encoders to links. @@ -223,6 +223,11 @@ struct resource_funcs { const struct dc_stream_state *stream); bool (*program_mcache_pipe_config)(struct dc_state *context, const struct dc_mcache_params *mcache_params); + enum dc_status (*update_dc_state_for_encoder_switch)(struct dc_link *link, + struct dc_link_settings *link_setting, + uint8_t pipe_count, + struct pipe_ctx *pipes, + struct audio_output *audio_output); }; struct audio_support{ @@ -556,7 +561,10 @@ struct dcn_bw_output { struct dml2_mcache_surface_allocation mcache_allocations[DML2_MAX_PLANES]; struct dmub_cmd_fams2_global_config fams2_global_config; union dmub_cmd_fams2_config fams2_stream_base_params[DML2_MAX_PLANES]; - union dmub_cmd_fams2_config fams2_stream_sub_params[DML2_MAX_PLANES]; + union { + union dmub_cmd_fams2_config fams2_stream_sub_params[DML2_MAX_PLANES]; + union dmub_fams2_stream_static_sub_state_v2 fams2_stream_sub_params_v2[DML2_MAX_PLANES]; + }; struct dml2_display_arb_regs arb_regs; }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h b/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h index d19a595c2be4..134091d5842d 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h @@ -622,7 +622,7 @@ extern const struct dcn_ip_params dcn10_ip_defaults; bool dcn_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn_get_soc_clks( struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h index c14d64687a3d..3b736f4687a6 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h @@ -324,6 +324,8 @@ struct clk_mgr_funcs { int (*get_dispclk_from_dentist)(struct clk_mgr *clk_mgr_base); + bool (*is_dc_mode_present)(struct clk_mgr *clk_mgr); + }; struct clk_mgr { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h index e94e9ba60f55..61c4d2a7db1c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h @@ -211,7 +211,7 @@ struct dccg_funcs { struct dccg *dccg, enum streamclk_source src, uint32_t otg_inst); - void (*set_dto_dscclk)(struct dccg *dccg, uint32_t dsc_inst); + void (*set_dto_dscclk)(struct dccg *dccg, uint32_t dsc_inst, uint32_t num_slices_h); void (*set_ref_dscclk)(struct dccg *dccg, uint32_t dsc_inst); void (*dccg_root_gate_disable_control)(struct dccg *dccg, uint32_t pipe_idx, uint32_t disable_clock_gating); }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h index 6e303b81bfb0..7641439f6ca0 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h @@ -190,6 +190,42 @@ struct mpc_grph_gamut_adjustment { enum mpcc_gamut_remap_id mpcc_gamut_remap_block_id; }; +struct mpc_rmcm_regs { + uint32_t rmcm_3dlut_mem_pwr_state; + uint32_t rmcm_3dlut_mem_pwr_force; + uint32_t rmcm_3dlut_mem_pwr_dis; + uint32_t rmcm_3dlut_mem_pwr_mode; + uint32_t rmcm_3dlut_size; + uint32_t rmcm_3dlut_mode; + uint32_t rmcm_3dlut_mode_cur; + uint32_t rmcm_3dlut_read_sel; + uint32_t rmcm_3dlut_30bit_en; + uint32_t rmcm_3dlut_wr_en_mask; + uint32_t rmcm_3dlut_ram_sel; + uint32_t rmcm_3dlut_out_norm_factor; + uint32_t rmcm_3dlut_fl_sel; + uint32_t rmcm_3dlut_out_offset_r; + uint32_t rmcm_3dlut_out_scale_r; + uint32_t rmcm_3dlut_fl_done; + uint32_t rmcm_3dlut_fl_soft_underflow; + uint32_t rmcm_3dlut_fl_hard_underflow; + uint32_t rmcm_cntl; + uint32_t rmcm_shaper_mem_pwr_state; + uint32_t rmcm_shaper_mem_pwr_force; + uint32_t rmcm_shaper_mem_pwr_dis; + uint32_t rmcm_shaper_mem_pwr_mode; + uint32_t rmcm_shaper_lut_mode; + uint32_t rmcm_shaper_mode_cur; + uint32_t rmcm_shaper_lut_write_en_mask; + uint32_t rmcm_shaper_lut_write_sel; + uint32_t rmcm_shaper_offset_b; + uint32_t rmcm_shaper_scale_b; + uint32_t rmcm_shaper_rama_exp_region_start_b; + uint32_t rmcm_shaper_rama_exp_region_start_seg_b; + uint32_t rmcm_shaper_rama_exp_region_end_b; + uint32_t rmcm_shaper_rama_exp_region_end_base_b; +}; + struct mpcc_sm_cfg { bool enable; /* 0-single plane,2-row subsampling,4-column subsampling,6-checkboard subsampling */ @@ -301,6 +337,7 @@ struct mpcc_state { uint32_t rgam_mode; uint32_t rgam_lut; struct mpc_grph_gamut_adjustment gamut_remap; + struct mpc_rmcm_regs rmcm_regs; }; /** @@ -1038,6 +1075,11 @@ struct mpc_funcs { */ void (*program_3dlut_size)(struct mpc *mpc, bool is_17x17x17, int mpcc_id); + /** + * @mcm: + * + * MPC MCM new HW sequential programming functions + */ struct { void (*program_3dlut_size)(struct mpc *mpc, uint32_t width, int mpcc_id); void (*program_bias_scale)(struct mpc *mpc, uint16_t bias, uint16_t scale, int mpcc_id); @@ -1050,6 +1092,11 @@ struct mpc_funcs { bool lut_bank_a, int mpcc_id); } mcm; + /** + * @rmcm: + * + * MPC RMCM new HW sequential programming functions + */ struct { void (*enable_3dlut_fl)(struct mpc *mpc, bool enable, int mpcc_id); void (*update_3dlut_fast_load_select)(struct mpc *mpc, int mpcc_id, int hubp_idx); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h index fe7f3137f228..27f950ae45ee 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h @@ -117,6 +117,7 @@ struct stream_encoder { uint32_t stream_enc_inst; struct vpg *vpg; struct afmt *afmt; + struct apg *apg; }; struct enc_state { diff --git a/drivers/gpu/drm/amd/display/dc/inc/link.h b/drivers/gpu/drm/amd/display/dc/inc/link.h index 7d16351bba99..f2503402c10e 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link.h @@ -144,9 +144,9 @@ struct link_service { uint32_t (*dp_link_bandwidth_kbps)( const struct dc_link *link, const struct dc_link_settings *link_settings); - bool (*validate_dpia_bandwidth)( - const struct dc_stream_state *stream, - const unsigned int num_streams); + enum dc_status (*validate_dp_tunnel_bandwidth)( + const struct dc *dc, + const struct dc_state *new_ctx); uint32_t (*dp_required_hblank_size_bytes)( const struct dc_link *link, diff --git a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c index 96febabf464a..2956c2b3ad1a 100644 --- a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c +++ b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c @@ -34,6 +34,7 @@ #include "dm_helpers.h" #include "dc_dmub_srv.h" #include "dce/dmub_hw_lock_mgr.h" +#include "clk_mgr.h" #define DC_LOGGER \ link->ctx->logger @@ -67,10 +68,17 @@ static void dp_retrain_link_dp_test(struct dc_link *link, { struct pipe_ctx *pipes[MAX_PIPES]; struct dc_state *state = link->dc->current_state; + struct dc_stream_update stream_update = { 0 }; + bool dpms_off = false; + bool needs_divider_update = false; bool was_hpo_acquired = resource_is_hpo_acquired(link->dc->current_state); bool is_hpo_acquired; uint8_t count; int i; + struct audio_output audio_output[MAX_PIPES]; + + needs_divider_update = (link->dc->link_srv->dp_get_encoding_format(link_setting) != + link->dc->link_srv->dp_get_encoding_format((const struct dc_link_settings *) &link->cur_link_settings)); udelay(100); @@ -83,16 +91,59 @@ static void dp_retrain_link_dp_test(struct dc_link *link, link->dc, state, pipes[i]); + + // Disable OTG and re-enable after updating clocks + pipes[i]->stream_res.tg->funcs->disable_crtc(pipes[i]->stream_res.tg); } - if (link->dc->hwss.setup_hpo_hw_control) { - is_hpo_acquired = resource_is_hpo_acquired(state); - if (was_hpo_acquired != is_hpo_acquired) - link->dc->hwss.setup_hpo_hw_control(link->dc->hwseq, is_hpo_acquired); + if (needs_divider_update && link->dc->res_pool->funcs->update_dc_state_for_encoder_switch) { + link->dc->res_pool->funcs->update_dc_state_for_encoder_switch(link, + link_setting, count, + *pipes, &audio_output[0]); + for (i = 0; i < count; i++) { + pipes[i]->clock_source->funcs->program_pix_clk( + pipes[i]->clock_source, + &pipes[i]->stream_res.pix_clk_params, + link->dc->link_srv->dp_get_encoding_format(&pipes[i]->link_config.dp_link_settings), + &pipes[i]->pll_settings); + + if (pipes[i]->stream_res.audio != NULL) { + const struct link_hwss *link_hwss = get_link_hwss( + link, &pipes[i]->link_res); + + link_hwss->setup_audio_output(pipes[i], &audio_output[i], + pipes[i]->stream_res.audio->inst); + + pipes[i]->stream_res.audio->funcs->az_configure( + pipes[i]->stream_res.audio, + pipes[i]->stream->signal, + &audio_output[i].crtc_info, + &pipes[i]->stream->audio_info, + &audio_output[i].dp_link_info); + + if (link->dc->config.disable_hbr_audio_dp2 && + pipes[i]->stream_res.audio->funcs->az_disable_hbr_audio && + link->dc->link_srv->dp_is_128b_132b_signal(pipes[i])) + pipes[i]->stream_res.audio->funcs->az_disable_hbr_audio(pipes[i]->stream_res.audio); + } + } } - for (i = count-1; i >= 0; i--) - link_set_dpms_on(state, pipes[i]); + // Toggle on HPO I/O if necessary + is_hpo_acquired = resource_is_hpo_acquired(state); + if (was_hpo_acquired != is_hpo_acquired && link->dc->hwss.setup_hpo_hw_control) + link->dc->hwss.setup_hpo_hw_control(link->dc->hwseq, is_hpo_acquired); + + for (i = 0; i < count; i++) + pipes[i]->stream_res.tg->funcs->enable_crtc(pipes[i]->stream_res.tg); + + // Set DPMS on with stream update + for (i = 0; i < state->stream_count; i++) + if (state->streams[i] && state->streams[i]->link && state->streams[i]->link == link) { + stream_update.stream = state->streams[i]; + stream_update.dpms_off = &dpms_off; + dc_update_planes_and_stream(state->clk_mgr->ctx->dc, NULL, 0, state->streams[i], &stream_update); + } } static void dp_test_send_link_training(struct dc_link *link) diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c index 9655e6fa53a4..827b630daf49 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c @@ -593,8 +593,9 @@ static bool detect_dp(struct dc_link *link, if (sink_caps->transaction_type == DDC_TRANSACTION_TYPE_I2C_OVER_AUX) { sink_caps->signal = SIGNAL_TYPE_DISPLAY_PORT; - if (!detect_dp_sink_caps(link)) + if (!detect_dp_sink_caps(link)) { return false; + } if (is_dp_branch_device(link)) /* DP SST branch */ diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c index 273a3be6d593..bd51b279ad14 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c @@ -140,7 +140,8 @@ void link_blank_dp_stream(struct dc_link *link, bool hw_init) } } - if ((!link->wa_flags.dp_keep_receiver_powered) || hw_init) + if (((!link->wa_flags.dp_keep_receiver_powered) || hw_init) && + (link->type != dc_connection_none)) dpcd_write_rx_power_ctrl(link, false); } } @@ -842,14 +843,14 @@ void link_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; if (should_use_dto_dscclk) - dccg->funcs->set_dto_dscclk(dccg, dsc->inst); + dccg->funcs->set_dto_dscclk(dccg, dsc->inst, dsc_cfg.dc_dsc_cfg.num_slices_h); dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg); dsc->funcs->dsc_enable(dsc, pipe_ctx->stream_res.opp->inst); for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { struct display_stream_compressor *odm_dsc = odm_pipe->stream_res.dsc; if (should_use_dto_dscclk) - dccg->funcs->set_dto_dscclk(dccg, odm_dsc->inst); + dccg->funcs->set_dto_dscclk(dccg, odm_dsc->inst, dsc_cfg.dc_dsc_cfg.num_slices_h); odm_dsc->funcs->dsc_set_config(odm_dsc, &dsc_cfg, &dsc_optc_cfg); odm_dsc->funcs->dsc_enable(odm_dsc, odm_pipe->stream_res.opp->inst); } @@ -2296,8 +2297,7 @@ static bool allocate_usb4_bandwidth_for_stream(struct dc_stream_state *stream, i link->dpia_bw_alloc_config.remote_sink_req_bw[sink_index] = bw; } - /* get dp overhead for dp tunneling */ - link->dpia_bw_alloc_config.dp_overhead = link_dp_dpia_get_dp_overhead_in_dp_tunneling(link); + link->dpia_bw_alloc_config.dp_overhead = link_dpia_get_dp_overhead(link); req_bw += link->dpia_bw_alloc_config.dp_overhead; link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, req_bw); diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c index 1a04f4b74585..c5f4e803be84 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c @@ -100,7 +100,7 @@ static void construct_link_service_validation(struct link_service *link_srv) { link_srv->validate_mode_timing = link_validate_mode_timing; link_srv->dp_link_bandwidth_kbps = dp_link_bandwidth_kbps; - link_srv->validate_dpia_bandwidth = link_validate_dpia_bandwidth; + link_srv->validate_dp_tunnel_bandwidth = link_validate_dp_tunnel_bandwidth; link_srv->dp_required_hblank_size_bytes = dp_required_hblank_size_bytes; } diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.c b/drivers/gpu/drm/amd/display/dc/link/link_validation.c index 29606fda029d..aecaf37eee35 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_validation.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.c @@ -86,6 +86,10 @@ static bool dp_active_dongle_validate_timing( if (!dongle_caps->is_dp_hdmi_ycbcr420_pass_through) return false; break; + case PIXEL_ENCODING_UNDEFINED: + /* These color depths are currently not supported */ + ASSERT(false); + break; default: /* Invalid Pixel Encoding*/ return false; @@ -104,6 +108,10 @@ static bool dp_active_dongle_validate_timing( if (dongle_caps->dp_hdmi_max_bpc < 12) return false; break; + case COLOR_DEPTH_UNDEFINED: + /* These color depths are currently not supported */ + ASSERT(false); + break; case COLOR_DEPTH_141414: case COLOR_DEPTH_161616: default: @@ -255,6 +263,14 @@ uint32_t dp_link_bandwidth_kbps( return link_rate_per_lane_kbps * link_settings->lane_count / 10000 * total_data_bw_efficiency_x10000; } +static uint32_t dp_get_timing_bandwidth_kbps( + const struct dc_crtc_timing *timing, + const struct dc_link *link) +{ + return dc_bandwidth_in_kbps_from_timing(timing, + dc_link_get_highest_encoding_format(link)); +} + static bool dp_validate_mode_timing( struct dc_link *link, const struct dc_crtc_timing *timing) @@ -351,63 +367,81 @@ enum dc_status link_validate_mode_timing( return DC_OK; } +static const struct dc_tunnel_settings *get_dp_tunnel_settings(const struct dc_state *context, + const struct dc_stream_state *stream) +{ + int i; + const struct dc_tunnel_settings *dp_tunnel_settings = NULL; + + for (i = 0; i < MAX_PIPES; i++) { + if (context->res_ctx.pipe_ctx[i].stream && (context->res_ctx.pipe_ctx[i].stream == stream)) { + dp_tunnel_settings = &context->res_ctx.pipe_ctx[i].link_config.dp_tunnel_settings; + break; + } + } + + return dp_tunnel_settings; +} + /* - * This function calculates the bandwidth required for the stream timing - * and aggregates the stream bandwidth for the respective dpia link - * - * @stream: pointer to the dc_stream_state struct instance - * @num_streams: number of streams to be validated + * Calculates the DP tunneling bandwidth required for the stream timing + * and aggregates the stream bandwidth for the respective DP tunneling link * - * return: true if validation is succeeded + * return: dc_status */ -bool link_validate_dpia_bandwidth(const struct dc_stream_state *stream, const unsigned int num_streams) +enum dc_status link_validate_dp_tunnel_bandwidth(const struct dc *dc, const struct dc_state *new_ctx) { - int bw_needed[MAX_DPIA_NUM] = {0}; - struct dc_link *dpia_link[MAX_DPIA_NUM] = {0}; - int num_dpias = 0; - - for (unsigned int i = 0; i < num_streams; ++i) { - if (stream[i].signal == SIGNAL_TYPE_DISPLAY_PORT) { - /* new dpia sst stream, check whether it exceeds max dpia */ - if (num_dpias >= MAX_DPIA_NUM) - return false; + struct dc_validation_dpia_set dpia_link_sets[MAX_DPIA_NUM] = { 0 }; + uint8_t link_count = 0; + enum dc_status result = DC_OK; - dpia_link[num_dpias] = stream[i].link; - bw_needed[num_dpias] = dc_bandwidth_in_kbps_from_timing(&stream[i].timing, - dc_link_get_highest_encoding_format(dpia_link[num_dpias])); - num_dpias++; - } else if (stream[i].signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { - uint8_t j = 0; - /* check whether its a known dpia link */ - for (; j < num_dpias; ++j) { - if (dpia_link[j] == stream[i].link) - break; - } + // Iterate through streams in the new context + for (uint8_t i = 0; (i < MAX_PIPES && i < new_ctx->stream_count); i++) { + const struct dc_stream_state *stream = new_ctx->streams[i]; + const struct dc_link *link; + const struct dc_tunnel_settings *dp_tunnel_settings; + uint32_t timing_bw; + + if (stream == NULL) + continue; + + link = stream->link; + + if (!(link && (stream->signal == SIGNAL_TYPE_DISPLAY_PORT + || stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) + && link->hpd_status)) + continue; - if (j == num_dpias) { - /* new dpia mst stream, check whether it exceeds max dpia */ - if (num_dpias >= MAX_DPIA_NUM) - return false; - else { - dpia_link[j] = stream[i].link; - num_dpias++; - } + dp_tunnel_settings = get_dp_tunnel_settings(new_ctx, stream); + + if ((dp_tunnel_settings == NULL) || (dp_tunnel_settings->should_use_dp_bw_allocation == false)) + continue; + + timing_bw = dp_get_timing_bandwidth_kbps(&stream->timing, link); + + // Find an existing entry for this 'link' in 'dpia_link_sets' + for (uint8_t j = 0; j < MAX_DPIA_NUM; j++) { + bool is_new_slot = false; + + if (dpia_link_sets[j].link == NULL) { + is_new_slot = true; + link_count++; + dpia_link_sets[j].required_bw = 0; + dpia_link_sets[j].link = link; } - bw_needed[j] += dc_bandwidth_in_kbps_from_timing(&stream[i].timing, - dc_link_get_highest_encoding_format(dpia_link[j])); + if (is_new_slot || (dpia_link_sets[j].link == link)) { + dpia_link_sets[j].tunnel_settings = dp_tunnel_settings; + dpia_link_sets[j].required_bw += timing_bw; + break; + } } } - /* Include dp overheads */ - for (uint8_t i = 0; i < num_dpias; ++i) { - int dp_overhead = 0; - - dp_overhead = link_dp_dpia_get_dp_overhead_in_dp_tunneling(dpia_link[i]); - bw_needed[i] += dp_overhead; - } + if (link_count && link_dpia_validate_dp_tunnel_bandwidth(dpia_link_sets, link_count) == false) + result = DC_FAIL_DP_TUNNEL_BW_VALIDATE; - return dpia_validate_usb4_bw(dpia_link, bw_needed, num_dpias); + return result; } struct dp_audio_layout_config { diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.h b/drivers/gpu/drm/amd/display/dc/link/link_validation.h index bf398c49c3e8..9553c81053fe 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_validation.h +++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.h @@ -30,9 +30,9 @@ enum dc_status link_validate_mode_timing( const struct dc_stream_state *stream, struct dc_link *link, const struct dc_crtc_timing *timing); -bool link_validate_dpia_bandwidth( - const struct dc_stream_state *stream, - const unsigned int num_streams); +enum dc_status link_validate_dp_tunnel_bandwidth( + const struct dc *dc, + const struct dc_state *new_ctx); uint32_t dp_link_bandwidth_kbps( const struct dc_link *link, const struct dc_link_settings *link_settings); diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c index a5127c2d47ef..0f965380a9b4 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c @@ -385,9 +385,15 @@ bool dp_is_128b_132b_signal(struct pipe_ctx *pipe_ctx) bool dp_is_lttpr_present(struct dc_link *link) { /* Some sink devices report invalid LTTPR revision, so don't validate against that cap */ - return (dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt) != 0 && + uint32_t lttpr_count = dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + bool is_lttpr_present = (lttpr_count > 0 && link->dpcd_caps.lttpr_caps.max_lane_count > 0 && link->dpcd_caps.lttpr_caps.max_lane_count <= 4); + + if (lttpr_count > 0 && !is_lttpr_present) + DC_LOG_ERROR("LTTPR count is nonzero but invalid lane count reported. Assuming no LTTPR present.\n"); + + return is_lttpr_present; } /* in DP compliance test, DPR-120 may have @@ -1551,6 +1557,8 @@ enum dc_status dp_retrieve_lttpr_cap(struct dc_link *link) uint8_t lttpr_dpcd_data[10] = {0}; enum dc_status status; bool is_lttpr_present; + uint32_t lttpr_count; + uint32_t closest_lttpr_offset; /* Logic to determine LTTPR support*/ bool vbios_lttpr_interop = link->dc->caps.vbios_lttpr_aware; @@ -1602,20 +1610,22 @@ enum dc_status dp_retrieve_lttpr_cap(struct dc_link *link) lttpr_dpcd_data[DP_LTTPR_ALPM_CAPABILITIES - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + lttpr_count = dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + /* If this chip cap is set, at least one retimer must exist in the chain * Override count to 1 if we receive a known bad count (0 or an invalid value) */ if (((link->chip_caps & AMD_EXT_DISPLAY_PATH_CAPS__EXT_CHIP_MASK) == AMD_EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && - (dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt) == 0)) { + lttpr_count == 0) { /* If you see this message consistently, either the host platform has FIXED_VS flag * incorrectly configured or the sink device is returning an invalid count. */ DC_LOG_ERROR("lttpr_caps phy_repeater_cnt is 0x%x, forcing it to 0x80.", link->dpcd_caps.lttpr_caps.phy_repeater_cnt); link->dpcd_caps.lttpr_caps.phy_repeater_cnt = 0x80; + lttpr_count = 1; DC_LOG_DC("lttpr_caps forced phy_repeater_cnt = %d\n", link->dpcd_caps.lttpr_caps.phy_repeater_cnt); } - /* Attempt to train in LTTPR transparent mode if repeater count exceeds 8. */ is_lttpr_present = dp_is_lttpr_present(link); DC_LOG_DC("is_lttpr_present = %d\n", is_lttpr_present); @@ -1623,11 +1633,25 @@ enum dc_status dp_retrieve_lttpr_cap(struct dc_link *link) if (is_lttpr_present) { CONN_DATA_DETECT(link, lttpr_dpcd_data, sizeof(lttpr_dpcd_data), "LTTPR Caps: "); - core_link_read_dpcd(link, DP_LTTPR_IEEE_OUI, link->dpcd_caps.lttpr_caps.lttpr_ieee_oui, sizeof(link->dpcd_caps.lttpr_caps.lttpr_ieee_oui)); - CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_ieee_oui, sizeof(link->dpcd_caps.lttpr_caps.lttpr_ieee_oui), "LTTPR IEEE OUI: "); + // Identify closest LTTPR to determine if workarounds required for known embedded LTTPR + closest_lttpr_offset = dp_get_closest_lttpr_offset(lttpr_count); - core_link_read_dpcd(link, DP_LTTPR_DEVICE_ID, link->dpcd_caps.lttpr_caps.lttpr_device_id, sizeof(link->dpcd_caps.lttpr_caps.lttpr_device_id)); - CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_device_id, sizeof(link->dpcd_caps.lttpr_caps.lttpr_device_id), "LTTPR Device ID: "); + core_link_read_dpcd(link, (DP_LTTPR_IEEE_OUI + closest_lttpr_offset), + link->dpcd_caps.lttpr_caps.lttpr_ieee_oui, sizeof(link->dpcd_caps.lttpr_caps.lttpr_ieee_oui)); + core_link_read_dpcd(link, (DP_LTTPR_DEVICE_ID + closest_lttpr_offset), + link->dpcd_caps.lttpr_caps.lttpr_device_id, sizeof(link->dpcd_caps.lttpr_caps.lttpr_device_id)); + + if (lttpr_count > 1) { + CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_ieee_oui, sizeof(link->dpcd_caps.lttpr_caps.lttpr_ieee_oui), + "Closest LTTPR To Host's IEEE OUI: "); + CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_device_id, sizeof(link->dpcd_caps.lttpr_caps.lttpr_device_id), + "Closest LTTPR To Host's LTTPR Device ID: "); + } else { + CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_ieee_oui, sizeof(link->dpcd_caps.lttpr_caps.lttpr_ieee_oui), + "LTTPR IEEE OUI: "); + CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_device_id, sizeof(link->dpcd_caps.lttpr_caps.lttpr_device_id), + "LTTPR Device ID: "); + } } return status; diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c index 22bfdced64ab..9b2f1a7da1d1 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c @@ -75,12 +75,15 @@ enum dc_status dpcd_get_tunneling_device_data(struct dc_link *link) if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc) { status = core_link_read_dpcd(link, USB4_DRIVER_BW_CAPABILITY, - dpcd_dp_tun_data, 1); + dpcd_dp_tun_data, 2); if (status != DC_OK) goto err; - link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.raw = dpcd_dp_tun_data[0]; + link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.raw = + dpcd_dp_tun_data[USB4_DRIVER_BW_CAPABILITY - USB4_DRIVER_BW_CAPABILITY]; + link->dpcd_caps.usb4_dp_tun_info.dpia_tunnel_info.raw = + dpcd_dp_tun_data[DP_IN_ADAPTER_TUNNEL_INFO - USB4_DRIVER_BW_CAPABILITY]; } DC_LOG_DEBUG("%s: Link[%d] DP tunneling support (RouterId=%d AdapterId=%d) " @@ -155,8 +158,14 @@ void link_decide_dp_tunnel_settings(struct dc_stream_state *stream, link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling; if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc - && link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.bits.driver_bw_alloc_support) + && link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.bits.driver_bw_alloc_support) { dp_tunnel_setting->should_use_dp_bw_allocation = true; + dp_tunnel_setting->cm_id = link->dpcd_caps.usb4_dp_tun_info.usb4_driver_id & 0x0F; + dp_tunnel_setting->group_id = link->dpcd_caps.usb4_dp_tun_info.dpia_tunnel_info.bits.group_id; + dp_tunnel_setting->estimated_bw = link->dpia_bw_alloc_config.estimated_bw; + dp_tunnel_setting->allocated_bw = link->dpia_bw_alloc_config.allocated_bw; + dp_tunnel_setting->bw_granularity = link->dpia_bw_alloc_config.bw_granularity; + } } } diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c index 3af7564a84f1..819bf2d8ba53 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c @@ -35,6 +35,8 @@ #define Kbps_TO_Gbps (1000 * 1000) +#define MST_TIME_SLOT_COUNT 64 + // ------------------------------------------------------------------ // PRIVATE FUNCTIONS // ------------------------------------------------------------------ @@ -160,78 +162,6 @@ static void retrieve_usb4_dp_bw_allocation_info(struct dc_link *link) link->dpia_bw_alloc_config.nrd_max_lane_count); } -static uint8_t get_lowest_dpia_index(struct dc_link *link) -{ - const struct dc *dc_struct = link->dc; - uint8_t idx = 0xFF; - int i; - - for (i = 0; i < MAX_LINKS; ++i) { - - if (!dc_struct->links[i] || - dc_struct->links[i]->ep_type != DISPLAY_ENDPOINT_USB4_DPIA) - continue; - - if (idx > dc_struct->links[i]->link_index) { - idx = dc_struct->links[i]->link_index; - break; - } - } - - return idx; -} - -/* - * Get the maximum dp tunnel banwidth of host router - * - * @dc: pointer to the dc struct instance - * @hr_index: host router index - * - * return: host router maximum dp tunnel bandwidth - */ -static int get_host_router_total_dp_tunnel_bw(const struct dc *dc, uint8_t hr_index) -{ - uint8_t lowest_dpia_index = get_lowest_dpia_index(dc->links[0]); - uint8_t hr_index_temp = 0; - struct dc_link *link_dpia_primary, *link_dpia_secondary; - int total_bw = 0; - - for (uint8_t i = 0; i < MAX_LINKS - 1; ++i) { - - if (!dc->links[i] || dc->links[i]->ep_type != DISPLAY_ENDPOINT_USB4_DPIA) - continue; - - hr_index_temp = (dc->links[i]->link_index - lowest_dpia_index) / 2; - - if (hr_index_temp == hr_index) { - link_dpia_primary = dc->links[i]; - link_dpia_secondary = dc->links[i + 1]; - - /** - * If BW allocation enabled on both DPIAs, then - * HR BW = Estimated(dpia_primary) + Allocated(dpia_secondary) - * otherwise HR BW = Estimated(bw alloc enabled dpia) - */ - if ((link_dpia_primary->hpd_status && - link_dpia_primary->dpia_bw_alloc_config.bw_alloc_enabled) && - (link_dpia_secondary->hpd_status && - link_dpia_secondary->dpia_bw_alloc_config.bw_alloc_enabled)) { - total_bw += link_dpia_primary->dpia_bw_alloc_config.estimated_bw + - link_dpia_secondary->dpia_bw_alloc_config.allocated_bw; - } else if (link_dpia_primary->hpd_status && - link_dpia_primary->dpia_bw_alloc_config.bw_alloc_enabled) { - total_bw = link_dpia_primary->dpia_bw_alloc_config.estimated_bw; - } else if (link_dpia_secondary->hpd_status && - link_dpia_secondary->dpia_bw_alloc_config.bw_alloc_enabled) { - total_bw += link_dpia_secondary->dpia_bw_alloc_config.estimated_bw; - } - break; - } - } - - return total_bw; -} - /* * Cleanup function for when the dpia is unplugged to reset struct * and perform any required clean up @@ -251,32 +181,40 @@ static void dpia_bw_alloc_unplug(struct dc_link *link) static void link_dpia_send_bw_alloc_request(struct dc_link *link, int req_bw) { - uint8_t requested_bw; - uint32_t temp; + uint8_t request_reg_val; + uint32_t temp, request_bw; - /* Error check whether request bw greater than allocated */ - if (req_bw > link->dpia_bw_alloc_config.estimated_bw) { - DC_LOG_ERROR("%s: Request BW greater than estimated BW for link(%d)\n", - __func__, link->link_index); - req_bw = link->dpia_bw_alloc_config.estimated_bw; + if (link->dpia_bw_alloc_config.bw_granularity == 0) { + DC_LOG_ERROR("%s: Link[%d]: bw_granularity is zero!", __func__, link->link_index); + return; } temp = req_bw * link->dpia_bw_alloc_config.bw_granularity; - requested_bw = temp / Kbps_TO_Gbps; + request_reg_val = temp / Kbps_TO_Gbps; /* Always make sure to add more to account for floating points */ if (temp % Kbps_TO_Gbps) - ++requested_bw; + ++request_reg_val; - /* Error check whether requested and allocated are equal */ - req_bw = requested_bw * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity); - if (req_bw && (req_bw == link->dpia_bw_alloc_config.allocated_bw)) { - DC_LOG_ERROR("%s: Request BW equals to allocated BW for link(%d)\n", - __func__, link->link_index); + request_bw = request_reg_val * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity); + + if (request_bw > link->dpia_bw_alloc_config.estimated_bw) { + DC_LOG_ERROR("%s: Link[%d]: Request BW (%d --> %d) > Estimated BW (%d)... Set to Estimated BW!", + __func__, link->link_index, + req_bw, request_bw, link->dpia_bw_alloc_config.estimated_bw); + req_bw = link->dpia_bw_alloc_config.estimated_bw; + + temp = req_bw * link->dpia_bw_alloc_config.bw_granularity; + request_reg_val = temp / Kbps_TO_Gbps; + if (temp % Kbps_TO_Gbps) + ++request_reg_val; } + link->dpia_bw_alloc_config.allocated_bw = request_bw; + DC_LOG_DC("%s: Link[%d]: Request BW: %d", __func__, link->link_index, request_bw); + core_link_write_dpcd(link, REQUESTED_BW, - &requested_bw, + &request_reg_val, sizeof(uint8_t)); } @@ -304,14 +242,16 @@ bool link_dpia_enable_usb4_dp_bw_alloc_mode(struct dc_link *link) link->dpia_bw_alloc_config.bw_alloc_enabled = true; ret = true; - /* - * During DP tunnel creation, CM preallocates BW and reduces estimated BW of other - * DPIA. CM release preallocation only when allocation is complete. Do zero alloc - * to make the CM to release preallocation and update estimated BW correctly for - * all DPIAs per host router - */ - // TODO: Zero allocation can be removed once the MSFT CM fix has been released - link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, 0); + if (link->dc->debug.dpia_debug.bits.enable_usb4_bw_zero_alloc_patch) { + /* + * During DP tunnel creation, the CM preallocates BW + * and reduces the estimated BW of other DPIAs. + * The CM releases the preallocation only when the allocation is complete. + * Perform a zero allocation to make the CM release the preallocation + * and correctly update the estimated BW for all DPIAs per host router. + */ + link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, 0); + } } else DC_LOG_DEBUG("%s: link[%d] failed to enable DPTX BW allocation mode", __func__, link->link_index); } @@ -329,19 +269,17 @@ bool link_dpia_enable_usb4_dp_bw_alloc_mode(struct dc_link *link) */ void link_dp_dpia_handle_bw_alloc_status(struct dc_link *link, uint8_t status) { + link->dpia_bw_alloc_config.estimated_bw = get_estimated_bw(link); + if (status & DP_TUNNELING_BW_REQUEST_SUCCEEDED) { DC_LOG_DEBUG("%s: BW Allocation request succeeded on link(%d)", __func__, link->link_index); } else if (status & DP_TUNNELING_BW_REQUEST_FAILED) { - link->dpia_bw_alloc_config.estimated_bw = get_estimated_bw(link); - DC_LOG_DEBUG("%s: BW Allocation request failed on link(%d) allocated/estimated BW=%d", __func__, link->link_index, link->dpia_bw_alloc_config.estimated_bw); link_dpia_send_bw_alloc_request(link, link->dpia_bw_alloc_config.estimated_bw); } else if (status & DP_TUNNELING_ESTIMATED_BW_CHANGED) { - link->dpia_bw_alloc_config.estimated_bw = get_estimated_bw(link); - DC_LOG_DEBUG("%s: Estimated BW changed on link(%d) new estimated BW=%d", __func__, link->link_index, link->dpia_bw_alloc_config.estimated_bw); } @@ -374,9 +312,13 @@ void dpia_handle_usb4_bandwidth_allocation_for_link(struct dc_link *link, int pe void link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int req_bw) { - DC_LOG_DEBUG("%s: ENTER: link(%d), hpd_status(%d), current allocated_bw(%d), req_bw(%d)\n", + link->dpia_bw_alloc_config.estimated_bw = get_estimated_bw(link); + + DC_LOG_DEBUG("%s: ENTER: link[%d] hpd(%d) Allocated_BW: %d Estimated_BW: %d Req_BW: %d", __func__, link->link_index, link->hpd_status, - link->dpia_bw_alloc_config.allocated_bw, req_bw); + link->dpia_bw_alloc_config.allocated_bw, + link->dpia_bw_alloc_config.estimated_bw, + req_bw); if (link_dp_is_bw_alloc_available(link)) link_dpia_send_bw_alloc_request(link, req_bw); @@ -384,73 +326,116 @@ void link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int r DC_LOG_DEBUG("%s: BW Allocation mode not available", __func__); } -bool dpia_validate_usb4_bw(struct dc_link **link, int *bw_needed_per_dpia, const unsigned int num_dpias) +uint32_t link_dpia_get_dp_overhead(const struct dc_link *link) { - bool ret = true; - int bw_needed_per_hr[MAX_HR_NUM] = { 0, 0 }, host_router_total_dp_bw = 0; - uint8_t lowest_dpia_index, i, hr_index; + uint32_t link_dp_overhead = 0; - if (!num_dpias || num_dpias > MAX_DPIA_NUM) - return ret; + if ((link->type == dc_connection_mst_branch) && + !link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED) { + /* For 8b/10b encoding: MTP is 64 time slots long, slot 0 is used for MTPH + * MST overhead is 1/64 of link bandwidth (excluding any overhead) + */ + const struct dc_link_settings *link_cap = dc_link_get_link_cap(link); - lowest_dpia_index = get_lowest_dpia_index(link[0]); + if (link_cap) { + uint32_t link_bw_in_kbps = (uint32_t)link_cap->link_rate * + (uint32_t)link_cap->lane_count * + LINK_RATE_REF_FREQ_IN_KHZ * 8; + link_dp_overhead = (link_bw_in_kbps / MST_TIME_SLOT_COUNT) + + ((link_bw_in_kbps % MST_TIME_SLOT_COUNT) ? 1 : 0); + } + } - /* get total Host Router BW with granularity for the given modes */ - for (i = 0; i < num_dpias; ++i) { - int granularity_Gbps = 0; - int bw_granularity = 0; + return link_dp_overhead; +} - if (!link[i]->dpia_bw_alloc_config.bw_alloc_enabled) - continue; +/* + * Aggregates the DPIA bandwidth usage for the respective USB4 Router. + * And then validate if the required bandwidth is within the router's capacity. + * + * @dc_validation_dpia_set: pointer to the dc_validation_dpia_set + * @count: number of DPIA validation sets + * + * return: true if validation is succeeded + */ +bool link_dpia_validate_dp_tunnel_bandwidth(const struct dc_validation_dpia_set *dpia_link_sets, uint8_t count) +{ + uint32_t granularity_Gbps; + const struct dc_link *link; + uint32_t link_bw_granularity; + uint32_t link_required_bw; + struct usb4_router_validation_set router_sets[MAX_HOST_ROUTERS_NUM] = { 0 }; + uint8_t i; + bool is_success = true; + uint8_t router_count = 0; + + if ((dpia_link_sets == NULL) || (count == 0)) + return is_success; + + // Iterate through each DP tunneling link (DPIA). + // Aggregate its bandwidth requirements onto the respective USB4 router. + for (i = 0; i < count; i++) { + link = dpia_link_sets[i].link; + link_required_bw = dpia_link_sets[i].required_bw; + const struct dc_tunnel_settings *dp_tunnel_settings = dpia_link_sets[i].tunnel_settings; + + if ((link == NULL) || (dp_tunnel_settings == NULL) || dp_tunnel_settings->bw_granularity == 0) + break; - if (link[i]->link_index < lowest_dpia_index) - continue; + if (link->type == dc_connection_mst_branch) + link_required_bw += link_dpia_get_dp_overhead(link); - granularity_Gbps = (Kbps_TO_Gbps / link[i]->dpia_bw_alloc_config.bw_granularity); - bw_granularity = (bw_needed_per_dpia[i] / granularity_Gbps) * granularity_Gbps + - ((bw_needed_per_dpia[i] % granularity_Gbps) ? granularity_Gbps : 0); + granularity_Gbps = (Kbps_TO_Gbps / dp_tunnel_settings->bw_granularity); + link_bw_granularity = (link_required_bw / granularity_Gbps) * granularity_Gbps + + ((link_required_bw % granularity_Gbps) ? granularity_Gbps : 0); - hr_index = (link[i]->link_index - lowest_dpia_index) / 2; - bw_needed_per_hr[hr_index] += bw_granularity; - } + // Find or add the USB4 router associated with the current DPIA link + for (uint8_t j = 0; j < MAX_HOST_ROUTERS_NUM; j++) { + if (router_sets[j].is_valid == false) { + router_sets[j].is_valid = true; + router_sets[j].cm_id = dp_tunnel_settings->cm_id; + router_count++; + } - /* validate against each Host Router max BW */ - for (hr_index = 0; hr_index < MAX_HR_NUM; ++hr_index) { - if (bw_needed_per_hr[hr_index]) { - host_router_total_dp_bw = get_host_router_total_dp_tunnel_bw(link[0]->dc, hr_index); - if (bw_needed_per_hr[hr_index] > host_router_total_dp_bw) { - ret = false; + if (router_sets[j].cm_id == dp_tunnel_settings->cm_id) { + uint32_t remaining_bw = + dp_tunnel_settings->estimated_bw - dp_tunnel_settings->allocated_bw; + + router_sets[j].allocated_bw += dp_tunnel_settings->allocated_bw; + + if (remaining_bw > router_sets[j].remaining_bw) + router_sets[j].remaining_bw = remaining_bw; + + // Get the max estimated BW within the same CM_ID + if (dp_tunnel_settings->estimated_bw > router_sets[j].estimated_bw) + router_sets[j].estimated_bw = dp_tunnel_settings->estimated_bw; + + router_sets[j].required_bw += link_bw_granularity; + router_sets[j].dpia_count++; break; } } } - return ret; -} + // Validate bandwidth for each unique router found. + for (i = 0; i < router_count; i++) { + uint32_t total_bw = 0; -int link_dp_dpia_get_dp_overhead_in_dp_tunneling(struct dc_link *link) -{ - int dp_overhead = 0, link_mst_overhead = 0; + if (router_sets[i].is_valid == false) + break; - if (!link_dp_is_bw_alloc_available(link)) - return dp_overhead; + // Determine the total available bandwidth for the current router based on aggregated data + if ((router_sets[i].dpia_count == 1) || (router_sets[i].allocated_bw == 0)) + total_bw = router_sets[i].estimated_bw; + else + total_bw = router_sets[i].allocated_bw + router_sets[i].remaining_bw; - /* if its mst link, add MTPH overhead */ - if ((link->type == dc_connection_mst_branch) && - !link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED) { - /* For 8b/10b encoding: MTP is 64 time slots long, slot 0 is used for MTPH - * MST overhead is 1/64 of link bandwidth (excluding any overhead) - */ - const struct dc_link_settings *link_cap = - dc_link_get_link_cap(link); - uint32_t link_bw_in_kbps = (uint32_t)link_cap->link_rate * - (uint32_t)link_cap->lane_count * - LINK_RATE_REF_FREQ_IN_KHZ * 8; - link_mst_overhead = (link_bw_in_kbps / 64) + ((link_bw_in_kbps % 64) ? 1 : 0); + if (router_sets[i].required_bw > total_bw) { + is_success = false; + break; + } } - /* add all the overheads */ - dp_overhead = link_mst_overhead; - - return dp_overhead; + return is_success; } + diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h index 801965b5f9a4..41efcb3e44e2 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h @@ -28,10 +28,6 @@ #include "link.h" -/* Number of Host Routers per motherboard is 2 */ -#define MAX_HR_NUM 2 -/* Number of DPIA per host router is 2 */ -#define MAX_DPIA_NUM (MAX_HR_NUM * 2) /* * Host Router BW type @@ -42,6 +38,16 @@ enum bw_type { HOST_ROUTER_BW_INVALID, }; +struct usb4_router_validation_set { + bool is_valid; + uint8_t cm_id; + uint8_t dpia_count; + uint32_t required_bw; + uint32_t allocated_bw; + uint32_t estimated_bw; + uint32_t remaining_bw; +}; + /* * Enable USB4 DP BW allocation mode * @@ -74,25 +80,13 @@ void link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int r void dpia_handle_usb4_bandwidth_allocation_for_link(struct dc_link *link, int peak_bw); /* - * Handle the validation of total BW here and confirm that the bw used by each - * DPIA doesn't exceed available BW for each host router (HR) - * - * @link[]: array of link pointer to all possible DPIA links - * @bw_needed[]: bw needed for each DPIA link based on timing - * @num_dpias: Number of DPIAs for the above 2 arrays. Should always be <= MAX_DPIA_NUM - * - * return: TRUE if bw used by DPIAs doesn't exceed available BW else return FALSE - */ -bool dpia_validate_usb4_bw(struct dc_link **link, int *bw_needed, const unsigned int num_dpias); - -/* * Obtain all the DP overheads in dp tunneling for the dpia link * * @link: pointer to the dc_link struct instance * * return: DP overheads in DP tunneling */ -int link_dp_dpia_get_dp_overhead_in_dp_tunneling(struct dc_link *link); +uint32_t link_dpia_get_dp_overhead(const struct dc_link *link); /* * Handle DP BW allocation status register @@ -104,4 +98,15 @@ int link_dp_dpia_get_dp_overhead_in_dp_tunneling(struct dc_link *link); */ void link_dp_dpia_handle_bw_alloc_status(struct dc_link *link, uint8_t status); +/* + * Aggregates the DPIA bandwidth usage for the respective USB4 Router. + * + * @dc_validation_dpia_set: pointer to the dc_validation_dpia_set + * @count: number of DPIA validation sets + * + * return: true if validation is succeeded + */ +bool link_dpia_validate_dp_tunnel_bandwidth(const struct dc_validation_dpia_set *dpia_link_sets, uint8_t count); + #endif /* DC_INC_LINK_DP_DPIA_BW_H_ */ + diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c index da74c2b5854f..035795042a01 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c @@ -173,6 +173,15 @@ bool edp_set_backlight_level_nits(struct dc_link *link, target_luminance = (struct target_luminance_value *)&backlight_millinits; + //make sure we disable AMD ABC first. + core_link_read_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL, + &backlight_enable, sizeof(uint8_t)); + if (backlight_enable) { + backlight_enable = 0; + core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL, + &backlight_enable, 1); + } + core_link_read_dpcd(link, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, &backlight_enable, sizeof(uint8_t)); @@ -193,10 +202,22 @@ bool edp_set_backlight_level_nits(struct dc_link *link, *(uint16_t *)&dpcd_backlight_set.backlight_transition_time_ms = (uint16_t)transition_time_in_ms; uint8_t backlight_control = isHDR ? 1 : 0; + uint8_t backlight_enable = 0; + // OLEDs have no PWM, they can only use AUX if (link->dpcd_sink_ext_caps.bits.oled == 1) backlight_control = 1; + //make sure we disable VESA ABC first. + core_link_read_dpcd(link, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, + &backlight_enable, sizeof(uint8_t)); + + if (backlight_enable & DP_EDP_PANEL_LUMINANCE_CONTROL_ENABLE) { + backlight_enable &= ~DP_EDP_PANEL_LUMINANCE_CONTROL_ENABLE; + core_link_write_dpcd(link, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, + &backlight_enable, sizeof(backlight_enable)); + } + if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, (uint8_t *)(&dpcd_backlight_set), sizeof(dpcd_backlight_set)) != DC_OK) diff --git a/drivers/gpu/drm/amd/display/dc/mpc/Makefile b/drivers/gpu/drm/amd/display/dc/mpc/Makefile index 1e2e66508192..5402c3529f5e 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/mpc/Makefile @@ -68,5 +68,5 @@ MPC_DCN401 = dcn401_mpc.o AMD_DAL_MPC_DCN401 = $(addprefix $(AMDDALPATH)/dc/mpc/dcn401/,$(MPC_DCN401)) AMD_DISPLAY_FILES += $(AMD_DAL_MPC_DCN401) -endif +endif diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c b/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c index b4cea2b8cb2a..6f0e017a8ae2 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c @@ -30,7 +30,6 @@ #include "basics/conversion.h" #include "dcn10/dcn10_cm_common.h" #include "dc.h" -#include "dcn401/dcn401_mpc.h" #define REG(reg)\ mpc30->mpc_regs->reg @@ -879,7 +878,7 @@ void mpc32_set3dlut_ram10( } -static void mpc32_set_3dlut_mode( +void mpc32_set_3dlut_mode( struct mpc *mpc, enum dc_lut_mode mode, bool is_color_channel_12bits, @@ -1022,8 +1021,6 @@ static const struct mpc_funcs dcn32_mpc_funcs = { .power_on_mpc_mem_pwr = mpc3_power_on_ogam_lut, .get_mpc_out_mux = mpc1_get_mpc_out_mux, .set_bg_color = mpc1_set_bg_color, - .set_movable_cm_location = mpc401_set_movable_cm_location, - .populate_lut = mpc401_populate_lut, }; diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.h b/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.h index 9622518826c9..8c9b20bcca85 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.h @@ -391,4 +391,12 @@ void mpc32_select_3dlut_ram( enum dc_lut_mode mode, bool is_color_channel_12bits, uint32_t mpcc_id); + +void mpc32_set_3dlut_mode( + struct mpc *mpc, + enum dc_lut_mode mode, + bool is_color_channel_12bits, + bool is_lut_size17x17x17, + uint32_t mpcc_id); + #endif //__DC_MPCC_DCN32_H__ diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c index 98cf0cbd59ba..f3fb3fe13757 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c @@ -294,7 +294,7 @@ void mpc401_program_3dlut_size(struct mpc *mpc, bool is_17x17x17, int mpcc_id) REG_UPDATE(MPCC_MCM_3DLUT_MODE[mpcc_id], MPCC_MCM_3DLUT_SIZE, is_17x17x17 ? 0 : 1); } -static void program_gamut_remap( +void mpc_program_gamut_remap( struct mpc *mpc, unsigned int mpcc_id, const uint16_t *regval, @@ -426,7 +426,7 @@ void mpc401_set_gamut_remap( if (adjust->gamut_adjust_type != GRAPHICS_GAMUT_ADJUST_TYPE_SW) { /* Bypass / Disable if type is bypass or hw */ - program_gamut_remap(mpc, mpcc_id, NULL, + mpc_program_gamut_remap(mpc, mpcc_id, NULL, adjust->mpcc_gamut_remap_block_id, MPCC_GAMUT_REMAP_MODE_SELECT_0); } else { struct fixed31_32 arr_matrix[12]; @@ -460,12 +460,12 @@ void mpc401_set_gamut_remap( else mode_select = MPCC_GAMUT_REMAP_MODE_SELECT_2; - program_gamut_remap(mpc, mpcc_id, arr_reg_val, + mpc_program_gamut_remap(mpc, mpcc_id, arr_reg_val, adjust->mpcc_gamut_remap_block_id, mode_select); } } -static void read_gamut_remap(struct mpc *mpc, +void mpc_read_gamut_remap(struct mpc *mpc, int mpcc_id, uint16_t *regval, enum mpcc_gamut_remap_id gamut_remap_block_id, @@ -561,9 +561,9 @@ void mpc401_get_gamut_remap(struct mpc *mpc, struct mpc_grph_gamut_adjustment *adjust) { uint16_t arr_reg_val[12] = {0}; - uint32_t mode_select; + uint32_t mode_select = MPCC_GAMUT_REMAP_MODE_SELECT_0; - read_gamut_remap(mpc, mpcc_id, arr_reg_val, adjust->mpcc_gamut_remap_block_id, &mode_select); + mpc_read_gamut_remap(mpc, mpcc_id, arr_reg_val, adjust->mpcc_gamut_remap_block_id, &mode_select); if (mode_select == MPCC_GAMUT_REMAP_MODE_SELECT_0) { adjust->gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_BYPASS; diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h index 8e35ebc603a9..eb0c68d0b0c7 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h @@ -241,6 +241,19 @@ void mpc401_update_3dlut_fast_load_select( int mpcc_id, int hubp_idx); +void mpc_program_gamut_remap( + struct mpc *mpc, + unsigned int mpcc_id, + const uint16_t *regval, + enum mpcc_gamut_remap_id gamut_remap_block_id, + enum mpcc_gamut_remap_mode_select mode_select); + +void mpc_read_gamut_remap(struct mpc *mpc, + int mpcc_id, + uint16_t *regval, + enum mpcc_gamut_remap_id gamut_remap_block_id, + uint32_t *mode_select); + void mpc401_update_3dlut_fast_load_select( struct mpc *mpc, int mpcc_id, diff --git a/drivers/gpu/drm/amd/display/dc/os_types.h b/drivers/gpu/drm/amd/display/dc/os_types.h index f2ba76c1e0c0..782316348941 100644 --- a/drivers/gpu/drm/amd/display/dc/os_types.h +++ b/drivers/gpu/drm/amd/display/dc/os_types.h @@ -31,6 +31,7 @@ #include <linux/kgdb.h> #include <linux/delay.h> #include <linux/mm.h> +#include <linux/vmalloc.h> #include <asm/byteorder.h> diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c index 84f73fdb0f95..3a51be63f020 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c @@ -839,7 +839,7 @@ static enum dc_status build_mapped_resource( static enum dc_status dce100_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i; bool at_least_one_pipe = false; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c index f3d5baac11bf..cccde5a6f3cd 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c @@ -963,7 +963,7 @@ static enum dc_status build_mapped_resource( static enum dc_status dce110_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool result = false; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c index 4225cae68c10..164ba796f64c 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c @@ -886,7 +886,7 @@ static enum dc_status build_mapped_resource( enum dc_status dce112_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool result = false; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h index 6221d749246d..3efc4c55d2d2 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h @@ -45,7 +45,7 @@ enum dc_status dce112_validate_with_context( enum dc_status dce112_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); enum dc_status dce112_add_stream_to_ctx( struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c index d9ffdded5ce1..58b59d52dc9d 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c @@ -866,7 +866,7 @@ static void dce60_resource_destruct(struct dce110_resource_pool *pool) static enum dc_status dce60_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i; bool at_least_one_pipe = false; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c index bd5811f97531..3e8b0ac11d90 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c @@ -872,7 +872,7 @@ static void dce80_resource_destruct(struct dce110_resource_pool *pool) static enum dc_status dce80_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i; bool at_least_one_pipe = false; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c index be4ade0853e9..652c05c35494 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c @@ -1129,12 +1129,12 @@ static void dcn10_destroy_resource_pool(struct resource_pool **pool) static enum dc_status dcn10_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool voltage_supported; DC_FP_START(); - voltage_supported = dcn_validate_bandwidth(dc, context, fast_validate); + voltage_supported = dcn_validate_bandwidth(dc, context, validate_mode); DC_FP_END(); return voltage_supported ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c index 3405be07f5e3..067a93420a23 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c @@ -2007,7 +2007,7 @@ bool dcn20_fast_validate_bw( int *pipe_cnt_out, int *pipe_split_from, int *vlevel_out, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; int split[MAX_PIPES] = { 0 }; @@ -2021,7 +2021,7 @@ bool dcn20_fast_validate_bw( dcn20_merge_pipes_for_validate(dc, context); DC_FP_START(); - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); DC_FP_END(); *pipe_cnt_out = pipe_cnt; @@ -2125,7 +2125,7 @@ validate_out: } enum dc_status dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool voltage_supported; display_e2e_pipe_params_st *pipes; @@ -2135,7 +2135,7 @@ enum dc_status dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, return DC_FAIL_BANDWIDTH_VALIDATE; DC_FP_START(); - voltage_supported = dcn20_validate_bandwidth_fp(dc, context, fast_validate, pipes); + voltage_supported = dcn20_validate_bandwidth_fp(dc, context, validate_mode, pipes); DC_FP_END(); kfree(pipes); diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h index c0e062c7407d..e997d35a8b86 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h @@ -119,7 +119,7 @@ void dcn20_set_mcif_arb_params( struct dc_state *context, display_e2e_pipe_params_st *pipes, int pipe_cnt); -enum dc_status dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, bool fast_validate); +enum dc_status dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, enum dc_validate_mode validate_mode); void dcn20_merge_pipes_for_validate( struct dc *dc, struct dc_state *context); @@ -158,7 +158,7 @@ bool dcn20_fast_validate_bw( int *pipe_cnt_out, int *pipe_split_from, int *vlevel_out, - bool fast_validate); + enum dc_validate_mode validate_mode); enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state *context, struct dc_stream_state *stream); enum dc_status dcn20_add_stream_to_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c index 9ab01b65b177..238d7f8beb7c 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c @@ -769,7 +769,7 @@ bool dcn21_fast_validate_bw(struct dc *dc, int *pipe_cnt_out, int *pipe_split_from, int *vlevel_out, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; int split[MAX_PIPES] = { 0 }; @@ -783,7 +783,7 @@ bool dcn21_fast_validate_bw(struct dc *dc, dcn20_merge_pipes_for_validate(dc, context); DC_FP_START(); - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); DC_FP_END(); *pipe_cnt_out = pipe_cnt; @@ -924,7 +924,7 @@ validate_out: * dcn20_validate_bandwidth in dcn20_resource.c. */ static enum dc_status dcn21_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool voltage_supported; display_e2e_pipe_params_st *pipes; @@ -934,7 +934,7 @@ static enum dc_status dcn21_validate_bandwidth(struct dc *dc, struct dc_state *c return DC_FAIL_BANDWIDTH_VALIDATE; DC_FP_START(); - voltage_supported = dcn21_validate_bandwidth_fp(dc, context, fast_validate, pipes); + voltage_supported = dcn21_validate_bandwidth_fp(dc, context, validate_mode, pipes); DC_FP_END(); kfree(pipes); diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.h index f7ecc002c2f7..a017fd9854d1 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.h @@ -51,6 +51,6 @@ bool dcn21_fast_validate_bw( int *pipe_cnt_out, int *pipe_split_from, int *vlevel_out, - bool fast_validate); + enum dc_validate_mode validate_mode); #endif /* _DCN21_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c index f631ae34e320..4d4635e01eb6 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c @@ -1319,13 +1319,13 @@ static struct clock_source *dcn30_clock_source_create( int dcn30_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; DC_FP_START(); - dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn20_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); DC_FP_END(); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { @@ -1627,7 +1627,7 @@ noinline bool dcn30_internal_validate_bw( display_e2e_pipe_params_st *pipes, int *pipe_cnt_out, int *vlevel_out, - bool fast_validate, + enum dc_validate_mode validate_mode, bool allow_self_refresh_only) { bool out = false; @@ -1646,7 +1646,7 @@ noinline bool dcn30_internal_validate_bw( context->bw_ctx.dml.vba.VoltageLevel = 0; context->bw_ctx.dml.vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vactive; dc->res_pool->funcs->update_soc_for_wm_a(dc, context); - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); if (!pipe_cnt) { out = true; @@ -1655,7 +1655,7 @@ noinline bool dcn30_internal_validate_bw( dml_log_pipe_params(&context->bw_ctx.dml, pipes, pipe_cnt); - if (!fast_validate || !allow_self_refresh_only) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING || !allow_self_refresh_only) { /* * DML favors voltage over p-state, but we're more interested in * supporting p-state over voltage. We can't support p-state in @@ -1669,7 +1669,7 @@ noinline bool dcn30_internal_validate_bw( vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, vlevel, split, merge); } if (allow_self_refresh_only && - (fast_validate || vlevel == context->bw_ctx.dml.soc.num_states || + (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING || vlevel == context->bw_ctx.dml.soc.num_states || vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported)) { /* * If mode is unsupported or there's still no p-state support @@ -1678,7 +1678,7 @@ noinline bool dcn30_internal_validate_bw( * We don't actually support prefetch mode 2, so require that we * at least support prefetch mode 1. */ - context->bw_ctx.dml.validate_max_state = fast_validate; + context->bw_ctx.dml.validate_max_state = (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING); context->bw_ctx.dml.soc.allow_dram_self_refresh_or_dram_clock_change_in_vblank = dm_allow_self_refresh; @@ -1865,7 +1865,7 @@ noinline bool dcn30_internal_validate_bw( } if (repopulate_pipes) - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); context->bw_ctx.dml.vba.VoltageLevel = vlevel; *vlevel_out = vlevel; *pipe_cnt_out = pipe_cnt; @@ -2037,7 +2037,7 @@ void dcn30_calculate_wm_and_dlg( enum dc_status dcn30_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; @@ -2055,7 +2055,7 @@ enum dc_status dcn30_validate_bandwidth(struct dc *dc, goto validate_fail; DC_FP_START(); - out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate, true); + out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, validate_mode, true); DC_FP_END(); if (pipe_cnt == 0) @@ -2066,7 +2066,7 @@ enum dc_status dcn30_validate_bandwidth(struct dc *dc, BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h index 689d9bdace81..2c967fe55712 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h @@ -57,14 +57,14 @@ unsigned int dcn30_calc_max_scaled_time( unsigned int urgent_watermark); enum dc_status dcn30_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); bool dcn30_internal_validate_bw( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, int *pipe_cnt_out, int *vlevel_out, - bool fast_validate, + enum dc_validate_mode validate_mode, bool allow_self_refresh_only); void dcn30_calculate_wm_and_dlg( struct dc *dc, struct dc_state *context, @@ -78,7 +78,7 @@ void dcn30_populate_dml_writeback_from_context( int dcn30_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); bool dcn30_acquire_post_bldn_3dlut( struct resource_context *res_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c index 7e0af5297dc4..88afa59d17b0 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c @@ -1616,14 +1616,14 @@ static bool is_dual_plane(enum surface_pixel_format format) int dcn31x_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { uint32_t pipe_cnt; int i; dc_assert_fp_enabled(); - pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); for (i = 0; i < pipe_cnt; i++) { pipes[i].pipe.src.gpuvm = 1; @@ -1641,7 +1641,7 @@ int dcn31x_populate_dml_pipes_from_context(struct dc *dc, int dcn31_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -1649,7 +1649,7 @@ int dcn31_populate_dml_pipes_from_context( bool upscaled = false; DC_FP_START(); - dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn31x_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); DC_FP_END(); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { @@ -1760,7 +1760,7 @@ dcn31_set_mcif_arb_params(struct dc *dc, enum dc_status dcn31_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; @@ -1778,19 +1778,19 @@ enum dc_status dcn31_validate_bandwidth(struct dc *dc, goto validate_fail; DC_FP_START(); - out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate, true); + out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, validate_mode, true); DC_FP_END(); - // Disable fast_validate to set min dcfclk in calculate_wm_and_dlg + // Disable DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX to set min dcfclk in calculate_wm_and_dlg if (pipe_cnt == 0) - fast_validate = false; + validate_mode = DC_VALIDATE_MODE_AND_PROGRAMMING; if (!out) goto validate_fail; BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } @@ -1850,7 +1850,9 @@ static struct resource_funcs dcn31_res_pool_funcs = { .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .get_panel_config_defaults = dcn31_get_panel_config_defaults, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static struct clock_source *dcn30_clock_source_create( @@ -1954,6 +1956,9 @@ static bool dcn31_resource_construct( dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.num_of_host_routers = 2; + dc->caps.num_of_dpias_per_host_router = 2; + /* Use pipe context based otg sync logic */ dc->config.use_pipe_ctx_sync_logic = true; dc->config.disable_hbr_audio_dp2 = true; @@ -2228,3 +2233,35 @@ struct resource_pool *dcn31_create_resource_pool( kfree(pool); return NULL; } + +enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link, + struct dc_link_settings *link_setting, + uint8_t pipe_count, + struct pipe_ctx *pipes, + struct audio_output *audio_output) +{ + struct dc_state *state = link->dc->current_state; + int i; + +#if defined(CONFIG_DRM_AMD_DC_FP) + for (i = 0; i < state->stream_count; i++) + if (state->streams[i] && state->streams[i]->link && state->streams[i]->link == link) + link->dc->hwss.calculate_pix_rate_divider((struct dc *)link->dc, state, state->streams[i]); + + for (i = 0; i < pipe_count; i++) { + link->dc->res_pool->funcs->build_pipe_pix_clk_params(&pipes[i]); + + // Setup audio + if (pipes[i].stream_res.audio != NULL) + build_audio_output(state, &pipes[i], &audio_output[i]); + } +#else + /* This DCN requires rate divider updates and audio reprogramming to allow DP1<-->DP2 link rate switching, + * but the above will not compile on architectures without an FPU. + */ + DC_LOG_WARNING("%s: DP1<-->DP2 link retraining will not work on this DCN on non-FPU platforms", __func__); + ASSERT(0); +#endif + + return DC_OK; +} diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h index dd82815d7efe..c32c85ef0ba4 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h @@ -39,7 +39,7 @@ struct dcn31_resource_pool { enum dc_status dcn31_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn31_calculate_wm_and_dlg( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, @@ -48,7 +48,7 @@ void dcn31_calculate_wm_and_dlg( int dcn31_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn31_populate_dml_writeback_from_context(struct dc *dc, struct resource_context *res_ctx, @@ -66,6 +66,12 @@ struct resource_pool *dcn31_create_resource_pool( unsigned int dcn31_get_det_buffer_size( const struct dc_state *context); +enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link, + struct dc_link_settings *link_setting, + uint8_t pipe_count, + struct pipe_ctx *pipes, + struct audio_output *audio_output); + /*temp: B0 specific before switch to dcn313 headers*/ #ifndef regPHYPLLF_PIXCLK_RESYNC_CNTL #define regPHYPLLF_PIXCLK_RESYNC_CNTL 0x007e diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c index d96bc6cb73ad..dedf7bce6ece 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c @@ -1667,12 +1667,12 @@ static struct clock_source *dcn31_clock_source_create( static int dcn314_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int pipe_cnt; DC_FP_START(); - pipe_cnt = dcn314_populate_dml_pipes_from_context_fpu(dc, context, pipes, fast_validate); + pipe_cnt = dcn314_populate_dml_pipes_from_context_fpu(dc, context, pipes, validate_mode); DC_FP_END(); return pipe_cnt; @@ -1696,7 +1696,7 @@ static void dcn314_get_panel_config_defaults(struct dc_panel_config *panel_confi enum dc_status dcn314_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; @@ -1715,19 +1715,19 @@ enum dc_status dcn314_validate_bandwidth(struct dc *dc, DC_FP_START(); // do not support self refresh only - out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate, false); + out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, validate_mode, false); DC_FP_END(); - // Disable fast_validate to set min dcfclk in calculate_wm_and_dlg + // Disable DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX to set min dcfclk in calculate_wm_and_dlg if (pipe_cnt == 0) - fast_validate = false; + validate_mode = DC_VALIDATE_MODE_AND_PROGRAMMING; if (!out) goto validate_fail; BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } @@ -1779,7 +1779,9 @@ static struct resource_funcs dcn314_res_pool_funcs = { .get_panel_config_defaults = dcn314_get_panel_config_defaults, .get_preferred_eng_id_dpia = dcn314_get_preferred_eng_id_dpia, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static struct clock_source *dcn30_clock_source_create( @@ -1885,6 +1887,9 @@ static bool dcn314_resource_construct( dc->caps.max_disp_clock_khz_at_vmin = 650000; + dc->caps.num_of_host_routers = 2; + dc->caps.num_of_dpias_per_host_router = 2; + /* Use pipe context based otg sync logic */ dc->config.use_pipe_ctx_sync_logic = true; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h index f8ba531d6342..ac9bb7f097d5 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h @@ -41,7 +41,7 @@ struct dcn314_resource_pool { enum dc_status dcn314_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); struct resource_pool *dcn314_create_resource_pool( const struct dc_init_data *init_data, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c index 6c2bb3f63be1..d110be626bc2 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c @@ -1664,7 +1664,7 @@ static bool allow_pixel_rate_crb(struct dc *dc, struct dc_state *context) static int dcn315_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt, crb_idx, crb_pipes; struct resource_context *res_ctx = &context->res_ctx; @@ -1674,7 +1674,7 @@ static int dcn315_populate_dml_pipes_from_context( bool pixel_rate_crb = allow_pixel_rate_crb(dc, context); DC_FP_START(); - dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn31x_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); DC_FP_END(); for (i = 0, pipe_cnt = 0, crb_pipes = 0; i < dc->res_pool->pipe_count; i++) { @@ -1844,7 +1844,9 @@ static struct resource_funcs dcn315_res_pool_funcs = { .get_panel_config_defaults = dcn315_get_panel_config_defaults, .get_power_profile = dcn315_get_power_profile, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static bool dcn315_resource_construct( diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c index 568094827212..939811858ff7 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c @@ -1610,7 +1610,7 @@ static bool is_dual_plane(enum surface_pixel_format format) static int dcn316_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -1618,7 +1618,7 @@ static int dcn316_populate_dml_pipes_from_context( const int max_usable_det = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - DCN3_16_MIN_COMPBUF_SIZE_KB; DC_FP_START(); - dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn31x_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); DC_FP_END(); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { @@ -1720,7 +1720,9 @@ static struct resource_funcs dcn316_res_pool_funcs = { .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .get_panel_config_defaults = dcn316_get_panel_config_defaults, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static bool dcn316_resource_construct( diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c index bb0dae0be5b8..9ffa10189eee 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c @@ -1742,7 +1742,7 @@ void dcn32_add_phantom_pipes(struct dc *dc, struct dc_state *context, } } -static bool dml1_validate(struct dc *dc, struct dc_state *context, bool fast_validate) +static bool dml1_validate(struct dc *dc, struct dc_state *context, enum dc_validate_mode validate_mode) { bool out = false; @@ -1767,7 +1767,7 @@ static bool dml1_validate(struct dc *dc, struct dc_state *context, bool fast_val goto validate_fail; DC_FP_START(); - out = dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate); + out = dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, validate_mode); DC_FP_END(); if (pipe_cnt == 0) @@ -1778,7 +1778,7 @@ static bool dml1_validate(struct dc *dc, struct dc_state *context, bool fast_val BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } @@ -1809,7 +1809,7 @@ validate_out: enum dc_status dcn32_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { unsigned int i; enum dc_status status; @@ -1827,11 +1827,11 @@ enum dc_status dcn32_validate_bandwidth(struct dc *dc, if (dc->debug.using_dml2) status = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; else - status = dml1_validate(dc, context, fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + status = dml1_validate(dc, context, validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; - if (!fast_validate && status == DC_OK && dc_state_is_subvp_in_use(context)) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING && status == DC_OK && dc_state_is_subvp_in_use(context)) { /* check new stream configuration still supports cursor if subvp used */ for (i = 0; i < context->stream_count; i++) { stream = context->streams[i]; @@ -1846,14 +1846,14 @@ enum dc_status dcn32_validate_bandwidth(struct dc *dc, }; } - if (!fast_validate && status == DC_FAIL_HW_CURSOR_SUPPORT) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING && status == DC_FAIL_HW_CURSOR_SUPPORT) { /* attempt to validate again with subvp disabled due to cursor */ if (dc->debug.using_dml2) status = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; else - status = dml1_validate(dc, context, fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + status = dml1_validate(dc, context, validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; } return status; @@ -1862,7 +1862,7 @@ enum dc_status dcn32_validate_bandwidth(struct dc *dc, int dcn32_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -1878,7 +1878,7 @@ int dcn32_populate_dml_pipes_from_context( int num_subvp_none = 0; int odm_slice_count; - dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn20_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); /* For single display subvp, look for subvp main so if we have phantom * pipe, we can set odm policy to match main pipe @@ -1960,7 +1960,7 @@ int dcn32_populate_dml_pipes_from_context( /* Only populate DML input with subvp info for full updates. * This is just a workaround -- needs a proper fix. */ - if (!fast_validate) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING) { switch (dc_state_get_pipe_subvp_type(context, pipe)) { case SUBVP_MAIN: pipes[pipe_cnt].pipe.src.use_mall_for_pstate_change = dm_use_mall_pstate_change_sub_viewport; @@ -2061,21 +2061,15 @@ void dcn32_calculate_wm_and_dlg(struct dc *dc, struct dc_state *context, static void dcn32_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { - struct dml2_configuration_options *dml2_opt = &dc->dml2_tmp; - - memcpy(dml2_opt, &dc->dml2_options, sizeof(dc->dml2_options)); - DC_FP_START(); dcn32_update_bw_bounding_box_fpu(dc, bw_params); - dml2_opt->use_clock_dc_limits = false; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2); + dml2_reinit(dc, &dc->dml2_options, &dc->current_state->bw_ctx.dml2); - dml2_opt->use_clock_dc_limits = true; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2_dc_power_source) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2_dc_power_source); + dml2_reinit(dc, &dc->dml2_dc_power_options, &dc->current_state->bw_ctx.dml2_dc_power_source); DC_FP_END(); } @@ -2257,7 +2251,7 @@ static bool dcn32_resource_construct( dc->caps.color.dpp.gamma_corr = 1; dc->caps.color.dpp.dgam_rom_for_yuv = 0; - dc->caps.color.dpp.hw_3d_lut = 1; + dc->caps.color.dpp.hw_3d_lut = 0; dc->caps.color.dpp.ogam_ram = 0; // no OGAM in DPP since DCN1 // no OGAM ROM on DCN2 and later ASICs dc->caps.color.dpp.ogam_rom_caps.srgb = 0; @@ -2276,6 +2270,7 @@ static bool dcn32_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; /* Use pipe context based otg sync logic */ dc->config.use_pipe_ctx_sync_logic = true; @@ -2519,7 +2514,6 @@ static bool dcn32_resource_construct( } dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = false; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = true; @@ -2551,6 +2545,10 @@ static bool dcn32_resource_construct( if (ASICREV_IS_GC_11_0_3(dc->ctx->asic_id.hw_internal_rev) && (dc->config.sdpif_request_limit_words_per_umc == 0)) dc->config.sdpif_request_limit_words_per_umc = 16; + /* init DC limited DML2 options */ + memcpy(&dc->dml2_dc_power_options, &dc->dml2_options, sizeof(struct dml2_configuration_options)); + dc->dml2_dc_power_options.use_clock_dc_limits = true; + return true; create_fail: diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h index d60ed77eda80..82f966cf4ed2 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h @@ -100,12 +100,12 @@ void dcn32_add_phantom_pipes(struct dc *dc, enum dc_status dcn32_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); int dcn32_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn32_calculate_wm_and_dlg( struct dc *dc, struct dc_state *context, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c index 7db1f7a5613f..c53266e16c58 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c @@ -1580,21 +1580,15 @@ static struct dc_cap_funcs cap_funcs = { static void dcn321_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { - struct dml2_configuration_options *dml2_opt = &dc->dml2_tmp; - - memcpy(dml2_opt, &dc->dml2_options, sizeof(dc->dml2_options)); - DC_FP_START(); dcn321_update_bw_bounding_box_fpu(dc, bw_params); - dml2_opt->use_clock_dc_limits = false; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2); + dml2_reinit(dc, &dc->dml2_options, &dc->current_state->bw_ctx.dml2); - dml2_opt->use_clock_dc_limits = true; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2_dc_power_source) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2_dc_power_source); + dml2_reinit(dc, &dc->dml2_dc_power_options, &dc->current_state->bw_ctx.dml2_dc_power_source); DC_FP_END(); } @@ -1761,8 +1755,8 @@ static bool dcn321_resource_construct( dc->caps.color.dpp.gamma_corr = 1; dc->caps.color.dpp.dgam_rom_for_yuv = 0; - dc->caps.color.dpp.hw_3d_lut = 1; - dc->caps.color.dpp.ogam_ram = 1; + dc->caps.color.dpp.hw_3d_lut = 0; + dc->caps.color.dpp.ogam_ram = 0; // no OGAM ROM on DCN2 and later ASICs dc->caps.color.dpp.ogam_rom_caps.srgb = 0; dc->caps.color.dpp.ogam_rom_caps.bt2020 = 0; @@ -1780,6 +1774,7 @@ static bool dcn321_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; /* Use pipe context based otg sync logic */ dc->config.use_pipe_ctx_sync_logic = true; @@ -2018,7 +2013,6 @@ static bool dcn321_resource_construct( } dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = false; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = true; @@ -2046,6 +2040,10 @@ static bool dcn321_resource_construct( dc->dml2_options.max_segments_per_hubp = 18; dc->dml2_options.det_segment_size = DCN3_2_DET_SEG_SIZE; + /* init DC limited DML2 options */ + memcpy(&dc->dml2_dc_power_options, &dc->dml2_options, sizeof(struct dml2_configuration_options)); + dc->dml2_dc_power_options.use_clock_dc_limits = true; + return true; create_fail: diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c index 72c6cf047db0..e327dee9be21 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c @@ -1734,15 +1734,15 @@ static void dcn35_get_panel_config_defaults(struct dc_panel_config *panel_config static enum dc_status dcn35_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; out = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate); + validate_mode); - if (fast_validate) + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; DC_FP_START(); @@ -1786,7 +1786,9 @@ static struct resource_funcs dcn35_res_pool_funcs = { .get_panel_config_defaults = dcn35_get_panel_config_defaults, .get_preferred_eng_id_dpia = dcn35_get_preferred_eng_id_dpia, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static bool dcn35_resource_construct( @@ -1874,7 +1876,7 @@ static bool dcn35_resource_construct( dc->caps.color.dpp.gamma_corr = 1; dc->caps.color.dpp.dgam_rom_for_yuv = 0; - dc->caps.color.dpp.hw_3d_lut = 1; + dc->caps.color.dpp.hw_3d_lut = 0; dc->caps.color.dpp.ogam_ram = 0; // no OGAM in DPP since DCN1 // no OGAM ROM on DCN301 dc->caps.color.dpp.ogam_rom_caps.srgb = 0; @@ -1893,6 +1895,10 @@ static bool dcn35_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; + + dc->caps.num_of_host_routers = 2; + dc->caps.num_of_dpias_per_host_router = 2; /* max_disp_clock_khz_at_vmin is slightly lower than the STA value in order * to provide some margin. @@ -2153,7 +2159,6 @@ static bool dcn35_resource_construct( dc->dcn_ip->max_num_dpp = pool->base.pipe_count; dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = true; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = false; if (dc->config.EnableMinDispClkODM) diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c index 989a270f7dea..2f2976afc229 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c @@ -1714,15 +1714,15 @@ static void dcn35_get_panel_config_defaults(struct dc_panel_config *panel_config static enum dc_status dcn351_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; out = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate); + validate_mode); - if (fast_validate) + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; DC_FP_START(); @@ -1758,7 +1758,9 @@ static struct resource_funcs dcn351_res_pool_funcs = { .get_panel_config_defaults = dcn35_get_panel_config_defaults, .get_preferred_eng_id_dpia = dcn351_get_preferred_eng_id_dpia, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static bool dcn351_resource_construct( @@ -1846,7 +1848,7 @@ static bool dcn351_resource_construct( dc->caps.color.dpp.gamma_corr = 1; dc->caps.color.dpp.dgam_rom_for_yuv = 0; - dc->caps.color.dpp.hw_3d_lut = 1; + dc->caps.color.dpp.hw_3d_lut = 0; dc->caps.color.dpp.ogam_ram = 0; // no OGAM in DPP since DCN1 // no OGAM ROM on DCN301 dc->caps.color.dpp.ogam_rom_caps.srgb = 0; @@ -1865,6 +1867,10 @@ static bool dcn351_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; + + dc->caps.num_of_host_routers = 2; + dc->caps.num_of_dpias_per_host_router = 2; /* max_disp_clock_khz_at_vmin is slightly lower than the STA value in order * to provide some margin. @@ -2125,7 +2131,6 @@ static bool dcn351_resource_construct( dc->dcn_ip->max_num_dpp = pool->base.pipe_count; dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = true; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = false; if (dc->config.EnableMinDispClkODM) diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c index 48e1f234185f..5b7848496a70 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c @@ -1715,15 +1715,15 @@ static void dcn35_get_panel_config_defaults(struct dc_panel_config *panel_config static enum dc_status dcn35_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; out = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate); + validate_mode); - if (fast_validate) + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; DC_FP_START(); @@ -1759,7 +1759,9 @@ static struct resource_funcs dcn36_res_pool_funcs = { .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .get_panel_config_defaults = dcn35_get_panel_config_defaults, .get_preferred_eng_id_dpia = dcn36_get_preferred_eng_id_dpia, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static bool dcn36_resource_construct( @@ -1847,7 +1849,7 @@ static bool dcn36_resource_construct( dc->caps.color.dpp.gamma_corr = 1; dc->caps.color.dpp.dgam_rom_for_yuv = 0; - dc->caps.color.dpp.hw_3d_lut = 1; + dc->caps.color.dpp.hw_3d_lut = 0; dc->caps.color.dpp.ogam_ram = 0; // no OGAM in DPP since DCN1 // no OGAM ROM on DCN301 dc->caps.color.dpp.ogam_rom_caps.srgb = 0; @@ -1866,6 +1868,10 @@ static bool dcn36_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; + + dc->caps.num_of_host_routers = 2; + dc->caps.num_of_dpias_per_host_router = 2; /* max_disp_clock_khz_at_vmin is slightly lower than the STA value in order * to provide some margin. @@ -2126,7 +2132,6 @@ static bool dcn36_resource_construct( dc->dcn_ip->max_num_dpp = pool->base.pipe_count; dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = true; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = false; if (dc->config.EnableMinDispClkODM) diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c index f420c4dafa03..a9d989f20405 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c @@ -70,7 +70,6 @@ #include "dml/dcn30/display_mode_vba_30.h" #include "vm_helper.h" #include "dcn20/dcn20_vmid.h" -#include "dml/dcn401/dcn401_fpu.h" #include "dc_state_priv.h" @@ -1608,10 +1607,6 @@ static struct dc_cap_funcs cap_funcs = { static void dcn401_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { - struct dml2_configuration_options *dml2_opt = &dc->dml2_tmp; - - memcpy(dml2_opt, &dc->dml2_options, sizeof(dc->dml2_options)); - /* re-calculate the available MALL size if required */ if (bw_params->num_channels > 0) { dc->caps.max_cab_allocation_bytes = dcn401_calc_num_avail_chans_for_mall( @@ -1622,15 +1617,11 @@ static void dcn401_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *b DC_FP_START(); - dcn401_update_bw_bounding_box_fpu(dc, bw_params); - - dml2_opt->use_clock_dc_limits = false; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2); + dml2_reinit(dc, &dc->dml2_options, &dc->current_state->bw_ctx.dml2); - dml2_opt->use_clock_dc_limits = true; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2_dc_power_source) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2_dc_power_source); + dml2_reinit(dc, &dc->dml2_dc_power_options, &dc->current_state->bw_ctx.dml2_dc_power_source); DC_FP_END(); } @@ -1644,7 +1635,7 @@ enum dc_status dcn401_patch_unknown_plane_state(struct dc_plane_state *plane_sta enum dc_status dcn401_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { unsigned int i; enum dc_status status = DC_OK; @@ -1662,9 +1653,9 @@ enum dc_status dcn401_validate_bandwidth(struct dc *dc, if (dc->debug.using_dml2) status = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; - if (!fast_validate && status == DC_OK && dc_state_is_subvp_in_use(context)) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING && status == DC_OK && dc_state_is_subvp_in_use(context)) { /* check new stream configuration still supports cursor if subvp used */ for (i = 0; i < context->stream_count; i++) { stream = context->streams[i]; @@ -1679,12 +1670,12 @@ enum dc_status dcn401_validate_bandwidth(struct dc *dc, }; } - if (!fast_validate && status == DC_FAIL_HW_CURSOR_SUPPORT) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING && status == DC_FAIL_HW_CURSOR_SUPPORT) { /* attempt to validate again with subvp disabled due to cursor */ if (dc->debug.using_dml2) status = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; } return status; @@ -1957,8 +1948,30 @@ static bool dcn401_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; dc->config.use_spl = true; dc->config.prefer_easf = true; + + dc->config.dcn_sharpness_range.sdr_rgb_min = 0; + dc->config.dcn_sharpness_range.sdr_rgb_max = 1750; + dc->config.dcn_sharpness_range.sdr_rgb_mid = 750; + dc->config.dcn_sharpness_range.sdr_yuv_min = 0; + dc->config.dcn_sharpness_range.sdr_yuv_max = 3500; + dc->config.dcn_sharpness_range.sdr_yuv_mid = 1500; + dc->config.dcn_sharpness_range.hdr_rgb_min = 0; + dc->config.dcn_sharpness_range.hdr_rgb_max = 2750; + dc->config.dcn_sharpness_range.hdr_rgb_mid = 1500; + + dc->config.dcn_override_sharpness_range.sdr_rgb_min = 0; + dc->config.dcn_override_sharpness_range.sdr_rgb_max = 3250; + dc->config.dcn_override_sharpness_range.sdr_rgb_mid = 1250; + dc->config.dcn_override_sharpness_range.sdr_yuv_min = 0; + dc->config.dcn_override_sharpness_range.sdr_yuv_max = 3500; + dc->config.dcn_override_sharpness_range.sdr_yuv_mid = 1500; + dc->config.dcn_override_sharpness_range.hdr_rgb_min = 0; + dc->config.dcn_override_sharpness_range.hdr_rgb_max = 2750; + dc->config.dcn_override_sharpness_range.hdr_rgb_mid = 1500; + dc->config.dc_mode_clk_limit_support = true; dc->config.enable_windowed_mpo_odm = true; dc->config.set_pipe_unlock_order = true; /* Need to ensure DET gets freed before allocating */ @@ -2195,7 +2208,6 @@ static bool dcn401_resource_construct( dc->config.sdpif_request_limit_words_per_umc = 16; dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = false; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = true; dc->dml2_options.map_dc_pipes_with_callbacks = true; @@ -2228,6 +2240,10 @@ static bool dcn401_resource_construct( /* SPL */ dc->caps.scl_caps.sharpener_support = true; + /* init DC limited DML2 options */ + memcpy(&dc->dml2_dc_power_options, &dc->dml2_options, sizeof(struct dml2_configuration_options)); + dc->dml2_dc_power_options.use_clock_dc_limits = true; + return true; create_fail: diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h index dc52a30991af..2ae6831c31ef 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h @@ -24,7 +24,7 @@ enum dc_status dcn401_patch_unknown_plane_state(struct dc_plane_state *plane_sta enum dc_status dcn401_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn401_prepare_mcache_programming(struct dc *dc, struct dc_state *context); diff --git a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c index e0008c5f08ad..55b929ca7982 100644 --- a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c +++ b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c @@ -196,7 +196,12 @@ static struct spl_rect calculate_mpc_slice_in_timing_active( int epimo = mpc_slice_count - plane_clip_rec->width % mpc_slice_count - 1; struct spl_rect mpc_rec; - if (use_recout_width_aligned) { + if (spl_in->basic_in.custom_width != 0) { + mpc_rec.width = spl_in->basic_in.custom_width; + mpc_rec.x = spl_in->basic_in.custom_x; + mpc_rec.height = plane_clip_rec->height; + mpc_rec.y = plane_clip_rec->y; + } else if (use_recout_width_aligned) { mpc_rec.width = recout_width_align; if ((mpc_rec.width * (mpc_slice_idx + 1)) > plane_clip_rec->width) { mpc_rec.width = plane_clip_rec->width % recout_width_align; @@ -219,7 +224,7 @@ static struct spl_rect calculate_mpc_slice_in_timing_active( /* extra pixels in the division remainder need to go to pipes after * the extra pixel index minus one(epimo) defined here as: */ - if (mpc_slice_idx > epimo) { + if (mpc_slice_idx > epimo && spl_in->basic_in.custom_width == 0) { mpc_rec.x += mpc_slice_idx - epimo - 1; mpc_rec.width += 1; } @@ -252,10 +257,10 @@ static struct spl_rect calculate_odm_slice_in_timing_active(struct spl_in *spl_i odm_rec.x = odm_slice_width * odm_slice_idx; odm_rec.width = is_last_odm_slice ? - /* last slice width is the reminder of h_active */ - h_active - odm_slice_width * (odm_slice_count - 1) : - /* odm slice width is the floor of h_active / count */ - odm_slice_width; + /* last slice width is the reminder of h_active */ + h_active - odm_slice_width * (odm_slice_count - 1) : + /* odm slice width is the floor of h_active / count */ + odm_slice_width; odm_rec.y = 0; odm_rec.height = v_active; @@ -884,7 +889,9 @@ static bool spl_get_isharp_en(struct spl_in *spl_in, /* Calculate number of tap with adaptive scaling off */ static void spl_get_taps_non_adaptive_scaler( - struct spl_scratch *spl_scratch, const struct spl_taps *in_taps, bool always_scale) + struct spl_scratch *spl_scratch, + const struct spl_taps *in_taps, + bool is_subsampled) { bool check_max_downscale = false; @@ -945,14 +952,15 @@ static void spl_get_taps_non_adaptive_scaler( SPL_ASSERT(check_max_downscale); - if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz) && !always_scale) + if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz)) spl_scratch->scl_data.taps.h_taps = 1; - if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert) && !always_scale) + if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert)) spl_scratch->scl_data.taps.v_taps = 1; - if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz_c) && !always_scale) + if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz_c) && !is_subsampled) spl_scratch->scl_data.taps.h_taps_c = 1; - if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert_c) && !always_scale) + if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert_c) && !is_subsampled) spl_scratch->scl_data.taps.v_taps_c = 1; + } /* Calculate optimal number of taps */ @@ -965,15 +973,13 @@ static bool spl_get_optimal_number_of_taps( unsigned int max_taps_y, max_taps_c; unsigned int min_taps_y, min_taps_c; enum lb_memory_config lb_config; - bool skip_easf = false; - bool always_scale = spl_in->basic_out.always_scale; + bool skip_easf = false; bool is_subsampled = spl_is_subsampled_format(spl_in->basic_in.format); - if (spl_scratch->scl_data.viewport.width > spl_scratch->scl_data.h_active && max_downscale_src_width != 0 && spl_scratch->scl_data.viewport.width > max_downscale_src_width) { - spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, always_scale); + spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, is_subsampled); *enable_easf_v = false; *enable_easf_h = false; *enable_isharp = false; @@ -982,7 +988,7 @@ static bool spl_get_optimal_number_of_taps( /* Disable adaptive scaler and sharpener when integer scaling is enabled */ if (spl_in->scaling_quality.integer_scaling) { - spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, always_scale); + spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, is_subsampled); *enable_easf_v = false; *enable_easf_h = false; *enable_isharp = false; @@ -997,8 +1003,9 @@ static bool spl_get_optimal_number_of_taps( * From programming guide: taps = min{ ceil(2*H_RATIO,1), 8} for downscaling * taps = 4 for upscaling */ - if (skip_easf) - spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, always_scale); + if (skip_easf) { + spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, is_subsampled); + } else { if (spl_is_video_format(spl_in->basic_in.format)) { spl_scratch->scl_data.taps.h_taps = 6; @@ -1124,7 +1131,6 @@ static bool spl_get_optimal_number_of_taps( (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert))) { spl_scratch->scl_data.taps.h_taps = 1; spl_scratch->scl_data.taps.v_taps = 1; - if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz_c) && !is_subsampled) spl_scratch->scl_data.taps.h_taps_c = 1; @@ -1149,6 +1155,7 @@ static bool spl_get_optimal_number_of_taps( if ((!*enable_easf_v) && !is_subsampled && (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert_c))) spl_scratch->scl_data.taps.v_taps_c = 1; + } } return true; diff --git a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h index 36a284305a70..23d254dea18f 100644 --- a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h +++ b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h @@ -460,6 +460,8 @@ struct basic_in { enum spl_color_space color_space; // Color Space unsigned int max_luminance; // Max Luminance TODO: Is determined in dc_hw_sequencer.c is_sdr bool film_grain_applied; // Film Grain Applied // TODO: To check from where to get this? + int custom_width; // Width for non-standard segmentation - used when != 0 + int custom_x; // Start x for non-standard segmentation - used when custom_width != 0 }; // Basic output information diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h index 3f3fa1b6a69e..0bafb6710761 100644 --- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h @@ -129,7 +129,9 @@ enum dmub_window_id { DMUB_WINDOW_5_TRACEBUFF, DMUB_WINDOW_6_FW_STATE, DMUB_WINDOW_7_SCRATCH_MEM, + DMUB_WINDOW_IB_MEM, DMUB_WINDOW_SHARED_STATE, + DMUB_WINDOW_LSDMA_BUFFER, DMUB_WINDOW_TOTAL, }; @@ -355,6 +357,7 @@ struct dmub_diagnostic_data { uint8_t is_traceport_en : 1; uint8_t is_cw0_enabled : 1; uint8_t is_cw6_enabled : 1; + uint8_t is_pwait : 1; }; struct dmub_srv_inbox { @@ -539,6 +542,7 @@ struct dmub_srv { uint32_t fw_version; bool is_virtual; struct dmub_fb scratch_mem_fb; + struct dmub_fb ib_mem_gart; volatile struct dmub_shared_state_feature_block *shared_state; volatile const struct dmub_fw_state *fw_state; @@ -576,6 +580,7 @@ struct dmub_srv { enum dmub_srv_power_state_type power_state; struct dmub_diagnostic_data debug; + struct dmub_fb lsdma_rb_fb; }; /** @@ -602,14 +607,6 @@ struct dmub_notification { }; }; -/* enum dmub_ips_mode - IPS mode identifier */ -enum dmub_ips_mode { - DMUB_IPS_MODE_IPS1_MAX = 0, - DMUB_IPS_MODE_IPS2, - DMUB_IPS_MODE_IPS1_RCG, - DMUB_IPS_MODE_IPS1_ONO2_ON -}; - /** * DMUB firmware version helper macro - useful for checking if the version * of a firmware to know if feature or functionality is supported or present. diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index b66bd10cdc9b..5cf5dd5831fc 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -104,6 +104,14 @@ */ #define DMUB_MAX_FPO_STREAMS 4 +/* Define to ensure that the "common" members always appear in the same + * order in different structs for back compat purposes + */ +#define COMMON_STREAM_STATIC_SUB_STATE \ + struct dmub_fams2_cmd_legacy_stream_static_state legacy; \ + struct dmub_fams2_cmd_subvp_stream_static_state subvp; \ + struct dmub_fams2_cmd_drr_stream_static_state drr; + /* Maximum number of streams on any ASIC. */ #define DMUB_MAX_STREAMS 6 @@ -291,6 +299,31 @@ union dmub_addr { } u; /*<< Low/high bit access */ uint64_t quad_part; /*<< 64 bit address */ }; + +/* Flattened structure containing SOC BB parameters stored in the VBIOS + * It is not practical to store the entire bounding box in VBIOS since the bounding box struct can gain new parameters. + * This also prevents alighment issues when new parameters are added to the SoC BB. + * The following parameters should be added since these values can't be obtained elsewhere: + * -dml2_soc_power_management_parameters + * -dml2_soc_vmin_clock_limits + */ +struct dmub_soc_bb_params { + uint32_t dram_clk_change_blackout_ns; + uint32_t dram_clk_change_read_only_ns; + uint32_t dram_clk_change_write_only_ns; + uint32_t fclk_change_blackout_ns; + uint32_t g7_ppt_blackout_ns; + uint32_t stutter_enter_plus_exit_latency_ns; + uint32_t stutter_exit_latency_ns; + uint32_t z8_stutter_enter_plus_exit_latency_ns; + uint32_t z8_stutter_exit_latency_ns; + uint32_t z8_min_idle_time_ns; + uint32_t type_b_dram_clk_change_blackout_ns; + uint32_t type_b_ppt_blackout_ns; + uint32_t vmin_limit_dispclk_khz; + uint32_t vmin_limit_dcfclk_khz; + uint32_t g7_temperature_read_blackout_ns; +}; #pragma pack(pop) /** @@ -762,6 +795,17 @@ enum dmub_ips_rcg_disable_type { #define DMUB_IPS1_COMMIT_MASK 0x00000004 #define DMUB_IPS2_COMMIT_MASK 0x00000008 +enum dmub_ips_comand_type { + /** + * Start/stop IPS residency measurements for a given IPS mode + */ + DMUB_CMD__IPS_RESIDENCY_CNTL = 0, + /** + * Query IPS residency information for a given IPS mode + */ + DMUB_CMD__IPS_QUERY_RESIDENCY_INFO = 1, +}; + /** * union dmub_fw_boot_options - Boot option definitions for SCRATCH14 */ @@ -856,7 +900,9 @@ union dmub_shared_state_ips_driver_signals { uint32_t allow_ips0_rcg : 1; /**< 1 is IPS0 RCG is allowed */ uint32_t allow_ips1_rcg : 1; /**< 1 is IPS1 RCG is allowed */ uint32_t allow_ips1z8 : 1; /**< 1 is IPS1 Z8 Retention is allowed */ - uint32_t reserved_bits : 24; /**< Reversed bits */ + uint32_t allow_dynamic_ips1 : 1; /**< 1 if IPS1 is allowed in dynamic use cases such as VPB */ + uint32_t allow_dynamic_ips1_z8: 1; /**< 1 if IPS1 z8 ret is allowed in dynamic use cases such as VPB */ + uint32_t reserved_bits : 22; /**< Reversed bits */ } bits; uint32_t all; }; @@ -1508,6 +1554,16 @@ enum dmub_cmd_type { */ DMUB_CMD__FUSED_IO = 89, + /** + * Command type used for all LSDMA commands. + */ + DMUB_CMD__LSDMA = 90, + + /** + * Command type use for all IPS commands. + */ + DMUB_CMD__IPS = 91, + DMUB_CMD__VBIOS = 128, }; @@ -1918,6 +1974,121 @@ struct dmub_rb_cmd_fams2_flip { struct dmub_fams2_flip_info flip_info; }; +struct dmub_cmd_lsdma_data { + union { + struct lsdma_init_data { + union dmub_addr gpu_addr_base; + uint32_t ring_size; + } init_data; + struct lsdma_tiled_copy_data { + uint32_t src_addr_lo; + uint32_t src_addr_hi; + uint32_t dst_addr_lo; + uint32_t dst_addr_hi; + + uint32_t src_x : 16; + uint32_t src_y : 16; + + uint32_t src_width : 16; + uint32_t src_height : 16; + + uint32_t dst_x : 16; + uint32_t dst_y : 16; + + uint32_t dst_width : 16; + uint32_t dst_height : 16; + + uint32_t rect_x : 16; + uint32_t rect_y : 16; + + uint32_t src_swizzle_mode : 5; + uint32_t src_mip_max : 5; + uint32_t src_mip_id : 5; + uint32_t dst_mip_max : 5; + uint32_t dst_swizzle_mode : 5; + uint32_t dst_mip_id : 5; + uint32_t tmz : 1; + uint32_t dcc : 1; + + uint32_t data_format : 6; + uint32_t padding1 : 4; + uint32_t dst_element_size : 3; + uint32_t num_type : 3; + uint32_t src_element_size : 3; + uint32_t write_compress : 2; + uint32_t cache_policy_dst : 2; + uint32_t cache_policy_src : 2; + uint32_t read_compress : 2; + uint32_t src_dim : 2; + uint32_t dst_dim : 2; + uint32_t max_uncom : 1; + + uint32_t max_com : 2; + uint32_t padding : 30; + } tiled_copy_data; + struct lsdma_linear_copy_data { + uint32_t count : 30; + uint32_t cache_policy_dst : 2; + + uint32_t tmz : 1; + uint32_t cache_policy_src : 2; + uint32_t padding : 29; + + uint32_t src_lo; + uint32_t src_hi; + uint32_t dst_lo; + uint32_t dst_hi; + } linear_copy_data; + struct lsdma_reg_write_data { + uint32_t reg_addr; + uint32_t reg_data; + } reg_write_data; + struct lsdma_pio_copy_data { + union { + struct { + uint32_t byte_count : 26; + uint32_t src_loc : 1; + uint32_t dst_loc : 1; + uint32_t src_addr_inc : 1; + uint32_t dst_addr_inc : 1; + uint32_t overlap_disable : 1; + uint32_t constant_fill : 1; + } fields; + uint32_t raw; + } packet; + uint32_t src_lo; + uint32_t src_hi; + uint32_t dst_lo; + uint32_t dst_hi; + } pio_copy_data; + struct lsdma_pio_constfill_data { + union { + struct { + uint32_t byte_count : 26; + uint32_t src_loc : 1; + uint32_t dst_loc : 1; + uint32_t src_addr_inc : 1; + uint32_t dst_addr_inc : 1; + uint32_t overlap_disable : 1; + uint32_t constant_fill : 1; + } fields; + uint32_t raw; + } packet; + uint32_t dst_lo; + uint32_t dst_hi; + uint32_t data; + } pio_constfill_data; + + uint32_t all[14]; + } u; + +}; + +struct dmub_rb_cmd_lsdma { + struct dmub_cmd_header header; + struct dmub_cmd_lsdma_data lsdma_data; +}; + struct dmub_optc_state_v2 { uint32_t v_total_min; uint32_t v_total_max; @@ -1949,6 +2120,28 @@ enum fams2_stream_type { FAMS2_STREAM_TYPE_SUBVP = 4, }; +struct dmub_rect16 { + /** + * Dirty rect x offset. + */ + uint16_t x; + + /** + * Dirty rect y offset. + */ + uint16_t y; + + /** + * Dirty rect width. + */ + uint16_t width; + + /** + * Dirty rect height. + */ + uint16_t height; +}; + /* static stream state */ struct dmub_fams2_legacy_stream_static_state { uint8_t vactive_det_fill_delay_otg_vlines; @@ -2021,11 +2214,13 @@ union dmub_fams2_stream_static_sub_state { }; //v0 union dmub_fams2_cmd_stream_static_sub_state { - struct dmub_fams2_cmd_legacy_stream_static_state legacy; - struct dmub_fams2_cmd_subvp_stream_static_state subvp; - struct dmub_fams2_cmd_drr_stream_static_state drr; + COMMON_STREAM_STATIC_SUB_STATE }; //v1 +union dmub_fams2_stream_static_sub_state_v2 { + COMMON_STREAM_STATIC_SUB_STATE +}; //v2 + struct dmub_fams2_stream_static_state { enum fams2_stream_type type; uint32_t otg_vline_time_ns; @@ -2091,7 +2286,7 @@ struct dmub_fams2_cmd_stream_static_base_state { struct dmub_fams2_stream_static_state_v1 { struct dmub_fams2_cmd_stream_static_base_state base; - union dmub_fams2_cmd_stream_static_sub_state sub_state; + union dmub_fams2_stream_static_sub_state_v2 sub_state; }; //v1 /** @@ -2139,6 +2334,11 @@ union dmub_cmd_fams2_config { } stream_v1; //v1 }; +struct dmub_fams2_config_v2 { + struct dmub_cmd_fams2_global_config global; + struct dmub_fams2_stream_static_state_v1 stream_v1[DMUB_MAX_STREAMS]; //v1 +}; + /** * DMUB rb command definition for FAMS2 (merged SubVP, FPO, Legacy) */ @@ -2148,6 +2348,22 @@ struct dmub_rb_cmd_fams2 { }; /** + * Indirect buffer descriptor + */ +struct dmub_ib_data { + union dmub_addr src; // location of indirect buffer in memory + uint16_t size; // indirect buffer size in bytes +}; + +/** + * DMUB rb command definition for commands passed over indirect buffer + */ +struct dmub_rb_cmd_ib { + struct dmub_cmd_header header; + struct dmub_ib_data ib_data; +}; + +/** * enum dmub_cmd_idle_opt_type - Idle optimization command type. */ enum dmub_cmd_idle_opt_type { @@ -2170,6 +2386,11 @@ enum dmub_cmd_idle_opt_type { * DCN hardware notify power state. */ DMUB_CMD__IDLE_OPT_SET_DC_POWER_STATE = 3, + + /** + * DCN notify to release HW. + */ + DMUB_CMD__IDLE_OPT_RELEASE_HW = 4, }; /** @@ -2315,7 +2536,8 @@ struct dmub_dig_transmitter_control_data_v1_7 { uint8_t connobj_id; /**< Connector Object Id defined in ObjectId.h */ uint8_t HPO_instance; /**< HPO instance (0: inst0, 1: inst1) */ uint8_t reserved1; /**< For future use */ - uint8_t reserved2[3]; /**< For future use */ + uint8_t skip_phy_ssc_reduction; + uint8_t reserved2[2]; /**< For future use */ uint32_t reserved3[11]; /**< For future use */ }; @@ -2933,6 +3155,7 @@ enum dmub_cmd_fams_type { DMUB_CMD__FAMS2_CONFIG = 4, DMUB_CMD__FAMS2_DRR_UPDATE = 5, DMUB_CMD__FAMS2_FLIP = 6, + DMUB_CMD__FAMS2_IB_CONFIG = 7, }; /** @@ -4416,6 +4639,37 @@ enum dmub_cmd_abm_type { DMUB_CMD__ABM_GET_HISTOGRAM_DATA = 11, }; +/** + * LSDMA command sub-types. + */ +enum dmub_cmd_lsdma_type { + /** + * Initialize parameters for LSDMA. + * Ring buffer is mapped to the ring buffer + */ + DMUB_CMD__LSDMA_INIT_CONFIG = 0, + /** + * LSDMA copies data from source to destination linearly + */ + DMUB_CMD__LSDMA_LINEAR_COPY = 1, + /** + * Send the tiled-to-tiled copy command + */ + DMUB_CMD__LSDMA_TILED_TO_TILED_COPY = 2, + /** + * Send the poll reg write command + */ + DMUB_CMD__LSDMA_POLL_REG_WRITE = 3, + /** + * Send the pio copy command + */ + DMUB_CMD__LSDMA_PIO_COPY = 4, + /** + * Send the pio constfill command + */ + DMUB_CMD__LSDMA_PIO_CONSTFILL = 5, +}; + struct abm_ace_curve { /** * @offsets: ACE curve offsets. @@ -5621,6 +5875,59 @@ struct dmub_rb_cmd_assr_enable { }; /** + * Current definition of "ips_mode" from driver + */ +enum ips_residency_mode { + IPS_RESIDENCY__IPS1_MAX, + IPS_RESIDENCY__IPS2, + IPS_RESIDENCY__IPS1_RCG, + IPS_RESIDENCY__IPS1_ONO2_ON, +}; + +#define NUM_IPS_HISTOGRAM_BUCKETS 16 + +/** + * IPS residency statistics to be sent to driver - subset of struct dmub_ips_residency_stats + */ +struct dmub_ips_residency_info { + uint32_t residency_millipercent; + uint32_t entry_counter; + uint32_t histogram[NUM_IPS_HISTOGRAM_BUCKETS]; + uint64_t total_time_us; + uint64_t total_inactive_time_us; +}; + +/** + * Data passed from driver to FW in a DMUB_CMD__IPS_RESIDENCY_CNTL command. + */ +struct dmub_cmd_ips_residency_cntl_data { + uint8_t panel_inst; + uint8_t start_measurement; + uint8_t padding[2]; // align to 4-byte boundary +}; + +struct dmub_rb_cmd_ips_residency_cntl { + struct dmub_cmd_header header; + struct dmub_cmd_ips_residency_cntl_data cntl_data; +}; + +/** + * Data passed from FW to driver in a DMUB_CMD__IPS_QUERY_RESIDENCY_INFO command. + */ +struct dmub_cmd_ips_query_residency_info_data { + union dmub_addr dest; + uint32_t size; + uint32_t ips_mode; + uint8_t panel_inst; + uint8_t padding[3]; // align to 4-byte boundary +}; + +struct dmub_rb_cmd_ips_query_residency_info { + struct dmub_cmd_header header; + struct dmub_cmd_ips_query_residency_info_data info_data; +}; + +/** * union dmub_rb_cmd - DMUB inbox command. */ union dmub_rb_cmd { @@ -5926,13 +6233,25 @@ union dmub_rb_cmd { * Definition of a DMUB_CMD__PSP_ASSR_ENABLE command. */ struct dmub_rb_cmd_assr_enable assr_enable; + struct dmub_rb_cmd_fams2 fams2_config; + struct dmub_rb_cmd_ib ib_fams2_config; + struct dmub_rb_cmd_fams2_drr_update fams2_drr_update; struct dmub_rb_cmd_fams2_flip fams2_flip; struct dmub_rb_cmd_fused_io fused_io; + + /** + * Definition of a DMUB_CMD__LSDMA command. + */ + struct dmub_rb_cmd_lsdma lsdma; + + struct dmub_rb_cmd_ips_residency_cntl ips_residency_cntl; + + struct dmub_rb_cmd_ips_query_residency_info ips_query_residency_info; }; /** diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c index a308bd604677..3f38db752b84 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c @@ -416,7 +416,7 @@ uint32_t dmub_dcn31_get_current_time(struct dmub_srv *dmub) void dmub_dcn31_get_diagnostic_data(struct dmub_srv *dmub) { - uint32_t is_dmub_enabled, is_soft_reset, is_sec_reset; + uint32_t is_dmub_enabled, is_soft_reset, is_sec_reset, is_pwait; uint32_t is_traceport_enabled, is_cw0_enabled, is_cw6_enabled; struct dmub_timeout_info timeout = {0}; @@ -466,6 +466,9 @@ void dmub_dcn31_get_diagnostic_data(struct dmub_srv *dmub) REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_dmub_enabled); dmub->debug.is_dmcub_enabled = is_dmub_enabled; + REG_GET(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS, &is_pwait); + dmub->debug.is_pwait = is_pwait; + REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &is_soft_reset); dmub->debug.is_dmcub_soft_reset = is_soft_reset; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c index 72a0f078cd1a..2228d62adc7e 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c @@ -92,19 +92,15 @@ void dmub_dcn35_reset(struct dmub_srv *dmub) uint32_t in_reset, is_enabled, scratch, i, pwait_mode; REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &in_reset); + REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_enabled); - if (in_reset == 0) { + if (in_reset == 0 && is_enabled != 0) { cmd.bits.status = 1; cmd.bits.command_code = DMUB_GPINT__STOP_FW; cmd.bits.param = 0; dmub->hw_funcs.set_gpint(dmub, cmd); - /** - * Timeout covers both the ACK and the wait - * for remaining work to finish. - */ - for (i = 0; i < timeout; ++i) { if (dmub->hw_funcs.is_gpint_acked(dmub, cmd)) break; @@ -130,11 +126,9 @@ void dmub_dcn35_reset(struct dmub_srv *dmub) /* Force reset in case we timed out, DMCUB is likely hung. */ } - REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_enabled); - if (is_enabled) { REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 1); - REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); + udelay(1); REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0); } @@ -160,11 +154,7 @@ void dmub_dcn35_reset_release(struct dmub_srv *dmub) LONO_SOCCLK_GATE_DISABLE, 1, LONO_DMCUBCLK_GATE_DISABLE, 1); - REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); - udelay(1); REG_UPDATE_2(DMCUB_CNTL, DMCUB_ENABLE, 1, DMCUB_TRACEPORT_EN, 1); - REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 1); - udelay(1); REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 0); REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 0); } @@ -464,7 +454,7 @@ uint32_t dmub_dcn35_get_current_time(struct dmub_srv *dmub) void dmub_dcn35_get_diagnostic_data(struct dmub_srv *dmub) { - uint32_t is_dmub_enabled, is_soft_reset; + uint32_t is_dmub_enabled, is_soft_reset, is_pwait; uint32_t is_traceport_enabled, is_cw6_enabled; struct dmub_timeout_info timeout = {0}; @@ -515,6 +505,9 @@ void dmub_dcn35_get_diagnostic_data(struct dmub_srv *dmub) REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_dmub_enabled); dmub->debug.is_dmcub_enabled = is_dmub_enabled; + REG_GET(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS, &is_pwait); + dmub->debug.is_pwait = is_pwait; + REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &is_soft_reset); dmub->debug.is_dmcub_soft_reset = is_soft_reset; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c index 2575dbc448f7..b31adbd0d685 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c @@ -413,7 +413,7 @@ uint32_t dmub_dcn401_get_current_time(struct dmub_srv *dmub) void dmub_dcn401_get_diagnostic_data(struct dmub_srv *dmub) { - uint32_t is_dmub_enabled, is_soft_reset, is_sec_reset; + uint32_t is_dmub_enabled, is_soft_reset, is_sec_reset, is_pwait; uint32_t is_traceport_enabled, is_cw0_enabled, is_cw6_enabled; struct dmub_timeout_info timeout = {0}; @@ -464,6 +464,9 @@ void dmub_dcn401_get_diagnostic_data(struct dmub_srv *dmub) REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_dmub_enabled); dmub->debug.is_dmcub_enabled = is_dmub_enabled; + REG_GET(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS, &is_pwait); + dmub->debug.is_pwait = is_pwait; + REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &is_soft_reset); dmub->debug.is_dmcub_soft_reset = is_soft_reset; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c index acca7943a8c8..b17a19400c06 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c @@ -65,6 +65,12 @@ /* Default scratch mem size. */ #define DMUB_SCRATCH_MEM_SIZE (1024) +/* Default indirect buffer size. */ +#define DMUB_IB_MEM_SIZE (1280) + +/* Default LSDMA ring buffer size. */ +#define DMUB_LSDMA_RB_SIZE (64 * 1024) + /* Number of windows in use. */ #define DMUB_NUM_WINDOWS (DMUB_WINDOW_TOTAL) /* Base addresses. */ @@ -559,7 +565,9 @@ enum dmub_status window_sizes[DMUB_WINDOW_5_TRACEBUFF] = trace_buffer_size; window_sizes[DMUB_WINDOW_6_FW_STATE] = fw_state_size; window_sizes[DMUB_WINDOW_7_SCRATCH_MEM] = DMUB_SCRATCH_MEM_SIZE; + window_sizes[DMUB_WINDOW_IB_MEM] = DMUB_IB_MEM_SIZE; window_sizes[DMUB_WINDOW_SHARED_STATE] = max(DMUB_FW_HEADER_SHARED_STATE_SIZE, shared_state_size); + window_sizes[DMUB_WINDOW_LSDMA_BUFFER] = DMUB_LSDMA_RB_SIZE; out->fb_size = dmub_srv_calc_regions_for_memory_type(params, out, window_sizes, DMUB_WINDOW_MEMORY_TYPE_FB); @@ -645,6 +653,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, struct dmub_fb *tracebuff_fb = params->fb[DMUB_WINDOW_5_TRACEBUFF]; struct dmub_fb *fw_state_fb = params->fb[DMUB_WINDOW_6_FW_STATE]; struct dmub_fb *scratch_mem_fb = params->fb[DMUB_WINDOW_7_SCRATCH_MEM]; + struct dmub_fb *ib_mem_gart = params->fb[DMUB_WINDOW_IB_MEM]; struct dmub_fb *shared_state_fb = params->fb[DMUB_WINDOW_SHARED_STATE]; struct dmub_rb_init_params rb_params, outbox0_rb_params; @@ -655,7 +664,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, return DMUB_STATUS_INVALID; if (!inst_fb || !stack_fb || !data_fb || !bios_fb || !mail_fb || - !tracebuff_fb || !fw_state_fb || !scratch_mem_fb) { + !tracebuff_fb || !fw_state_fb || !scratch_mem_fb || !ib_mem_gart) { ASSERT(0); return DMUB_STATUS_INVALID; } @@ -741,6 +750,8 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, dmub->scratch_mem_fb = *scratch_mem_fb; + dmub->ib_mem_gart = *ib_mem_gart; + if (dmub->hw_funcs.setup_windows) dmub->hw_funcs.setup_windows(dmub, &cw2, &cw3, &cw4, &cw5, &cw6, ®ion6); diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index 3ba9b62ba70b..71efd2770c99 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -147,7 +147,7 @@ unsigned int mod_freesync_calc_v_total_from_refresh( ((unsigned int)(div64_u64((1000000000ULL * 1000000), refresh_in_uhz))); - if (MICRO_HZ_TO_HZ(refresh_in_uhz) <= stream->timing.min_refresh_in_uhz) { + if (refresh_in_uhz <= stream->timing.min_refresh_in_uhz) { /* When the target refresh rate is the minimum panel refresh rate, * round down the vtotal value to avoid stretching vblank over * panel's vtotal boundary. @@ -155,6 +155,14 @@ unsigned int mod_freesync_calc_v_total_from_refresh( v_total = div64_u64(div64_u64(((unsigned long long)( frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)), stream->timing.h_total), 1000000); + } else if (refresh_in_uhz >= stream->timing.max_refresh_in_uhz) { + /* When the target refresh rate is the maximum panel refresh rate + * round up the vtotal value to prevent off-by-one error causing + * v_total_min to be below the panel's lower bound + */ + v_total = div64_u64(div64_u64(((unsigned long long)( + frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)), + stream->timing.h_total) + (1000000 - 1), 1000000); } else { v_total = div64_u64(div64_u64(((unsigned long long)( frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)), diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index 11374a2cbab8..bfb446736ca8 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -396,6 +396,7 @@ enum amd_dpm_forced_level; * (such as allocating any required memory) * @suspend: handles IP specific hw/sw changes for suspend * @resume: handles IP specific hw/sw changes for resume + * @complete: handles IP specific changes after resume * @is_idle: returns current IP block idle status * @wait_for_idle: poll for idle * @check_soft_reset: check soft reset the IP block @@ -427,6 +428,7 @@ struct amd_ip_funcs { int (*prepare_suspend)(struct amdgpu_ip_block *ip_block); int (*suspend)(struct amdgpu_ip_block *ip_block); int (*resume)(struct amdgpu_ip_block *ip_block); + void (*complete)(struct amdgpu_ip_block *ip_block); bool (*is_idle)(struct amdgpu_ip_block *ip_block); int (*wait_for_idle)(struct amdgpu_ip_block *ip_block); bool (*check_soft_reset)(struct amdgpu_ip_block *ip_block); diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index f4d914dc731f..e2b1ea7467b0 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -108,6 +108,8 @@ enum pp_clock_type { PP_VCLK1, PP_DCLK, PP_DCLK1, + PP_ISPICLK, + PP_ISPXCLK, OD_SCLK, OD_MCLK, OD_VDDC_CURVE, diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c index 5c1cbdc122d2..71d986dd7a6e 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c @@ -98,6 +98,7 @@ int amdgpu_dpm_set_powergating_by_smu(struct amdgpu_device *adev, case AMD_IP_BLOCK_TYPE_GMC: case AMD_IP_BLOCK_TYPE_ACP: case AMD_IP_BLOCK_TYPE_VPE: + case AMD_IP_BLOCK_TYPE_ISP: if (pp_funcs && pp_funcs->set_powergating_by_smu) ret = (pp_funcs->set_powergating_by_smu( (adev)->powerplay.pp_handle, block_type, gate, 0)); @@ -852,22 +853,16 @@ int amdgpu_dpm_set_soft_freq_range(struct amdgpu_device *adev, uint32_t max) { struct smu_context *smu = adev->powerplay.pp_handle; - int ret = 0; - - if (type != PP_SCLK) - return -EINVAL; if (!is_support_sw_smu(adev)) return -EOPNOTSUPP; - mutex_lock(&adev->pm.mutex); - ret = smu_set_soft_freq_range(smu, - SMU_SCLK, + guard(mutex)(&adev->pm.mutex); + + return smu_set_soft_freq_range(smu, + type, min, max); - mutex_unlock(&adev->pm.mutex); - - return ret; } int amdgpu_dpm_write_watermarks_table(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c index 34e71727b27d..307ebf7e3226 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c @@ -1242,7 +1242,7 @@ static void kv_dpm_enable_bapm(void *handle, bool enable) if (pi->bapm_enable) { ret = amdgpu_kv_smc_bapm_enable(adev, enable); if (ret) - DRM_ERROR("amdgpu_kv_smc_bapm_enable failed\n"); + drm_err(adev_to_drm(adev), "amdgpu_kv_smc_bapm_enable failed\n"); } } @@ -1266,40 +1266,40 @@ static int kv_dpm_enable(struct amdgpu_device *adev) ret = kv_process_firmware_header(adev); if (ret) { - DRM_ERROR("kv_process_firmware_header failed\n"); + drm_err(adev_to_drm(adev), "kv_process_firmware_header failed\n"); return ret; } kv_init_fps_limits(adev); kv_init_graphics_levels(adev); ret = kv_program_bootup_state(adev); if (ret) { - DRM_ERROR("kv_program_bootup_state failed\n"); + drm_err(adev_to_drm(adev), "kv_program_bootup_state failed\n"); return ret; } kv_calculate_dfs_bypass_settings(adev); ret = kv_upload_dpm_settings(adev); if (ret) { - DRM_ERROR("kv_upload_dpm_settings failed\n"); + drm_err(adev_to_drm(adev), "kv_upload_dpm_settings failed\n"); return ret; } ret = kv_populate_uvd_table(adev); if (ret) { - DRM_ERROR("kv_populate_uvd_table failed\n"); + drm_err(adev_to_drm(adev), "kv_populate_uvd_table failed\n"); return ret; } ret = kv_populate_vce_table(adev); if (ret) { - DRM_ERROR("kv_populate_vce_table failed\n"); + drm_err(adev_to_drm(adev), "kv_populate_vce_table failed\n"); return ret; } ret = kv_populate_samu_table(adev); if (ret) { - DRM_ERROR("kv_populate_samu_table failed\n"); + drm_err(adev_to_drm(adev), "kv_populate_samu_table failed\n"); return ret; } ret = kv_populate_acp_table(adev); if (ret) { - DRM_ERROR("kv_populate_acp_table failed\n"); + drm_err(adev_to_drm(adev), "kv_populate_acp_table failed\n"); return ret; } kv_program_vc(adev); @@ -1310,39 +1310,39 @@ static int kv_dpm_enable(struct amdgpu_device *adev) if (pi->enable_auto_thermal_throttling) { ret = kv_enable_auto_thermal_throttling(adev); if (ret) { - DRM_ERROR("kv_enable_auto_thermal_throttling failed\n"); + drm_err(adev_to_drm(adev), "kv_enable_auto_thermal_throttling failed\n"); return ret; } } ret = kv_enable_dpm_voltage_scaling(adev); if (ret) { - DRM_ERROR("kv_enable_dpm_voltage_scaling failed\n"); + drm_err(adev_to_drm(adev), "kv_enable_dpm_voltage_scaling failed\n"); return ret; } ret = kv_set_dpm_interval(adev); if (ret) { - DRM_ERROR("kv_set_dpm_interval failed\n"); + drm_err(adev_to_drm(adev), "kv_set_dpm_interval failed\n"); return ret; } ret = kv_set_dpm_boot_state(adev); if (ret) { - DRM_ERROR("kv_set_dpm_boot_state failed\n"); + drm_err(adev_to_drm(adev), "kv_set_dpm_boot_state failed\n"); return ret; } ret = kv_enable_ulv(adev, true); if (ret) { - DRM_ERROR("kv_enable_ulv failed\n"); + drm_err(adev_to_drm(adev), "kv_enable_ulv failed\n"); return ret; } kv_start_dpm(adev); ret = kv_enable_didt(adev, true); if (ret) { - DRM_ERROR("kv_enable_didt failed\n"); + drm_err(adev_to_drm(adev), "kv_enable_didt failed\n"); return ret; } ret = kv_enable_smc_cac(adev, true); if (ret) { - DRM_ERROR("kv_enable_smc_cac failed\n"); + drm_err(adev_to_drm(adev), "kv_enable_smc_cac failed\n"); return ret; } @@ -1350,7 +1350,7 @@ static int kv_dpm_enable(struct amdgpu_device *adev) ret = amdgpu_kv_smc_bapm_enable(adev, false); if (ret) { - DRM_ERROR("amdgpu_kv_smc_bapm_enable failed\n"); + drm_err(adev_to_drm(adev), "amdgpu_kv_smc_bapm_enable failed\n"); return ret; } @@ -1358,7 +1358,7 @@ static int kv_dpm_enable(struct amdgpu_device *adev) kv_is_internal_thermal_sensor(adev->pm.int_thermal_type)) { ret = kv_set_thermal_temperature_range(adev, KV_TEMP_RANGE_MIN, KV_TEMP_RANGE_MAX); if (ret) { - DRM_ERROR("kv_set_thermal_temperature_range failed\n"); + drm_err(adev_to_drm(adev), "kv_set_thermal_temperature_range failed\n"); return ret; } amdgpu_irq_get(adev, &adev->pm.dpm.thermal.irq, @@ -1382,7 +1382,7 @@ static void kv_dpm_disable(struct amdgpu_device *adev) err = amdgpu_kv_smc_bapm_enable(adev, false); if (err) - DRM_ERROR("amdgpu_kv_smc_bapm_enable failed\n"); + drm_err(adev_to_drm(adev), "amdgpu_kv_smc_bapm_enable failed\n"); if (adev->asic_type == CHIP_MULLINS) kv_enable_nb_dpm(adev, false); @@ -1920,7 +1920,7 @@ static int kv_dpm_set_power_state(void *handle) if (pi->bapm_enable) { ret = amdgpu_kv_smc_bapm_enable(adev, adev->pm.ac_power); if (ret) { - DRM_ERROR("amdgpu_kv_smc_bapm_enable failed\n"); + drm_err(adev_to_drm(adev), "amdgpu_kv_smc_bapm_enable failed\n"); return ret; } } @@ -1931,7 +1931,7 @@ static int kv_dpm_set_power_state(void *handle) kv_update_dfs_bypass_settings(adev, new_ps); ret = kv_calculate_ds_divider(adev); if (ret) { - DRM_ERROR("kv_calculate_ds_divider failed\n"); + drm_err(adev_to_drm(adev), "kv_calculate_ds_divider failed\n"); return ret; } kv_calculate_nbps_level_settings(adev); @@ -1947,7 +1947,7 @@ static int kv_dpm_set_power_state(void *handle) ret = kv_update_vce_dpm(adev, new_ps, old_ps); if (ret) { - DRM_ERROR("kv_update_vce_dpm failed\n"); + drm_err(adev_to_drm(adev), "kv_update_vce_dpm failed\n"); return ret; } kv_update_sclk_t(adev); @@ -1960,7 +1960,7 @@ static int kv_dpm_set_power_state(void *handle) kv_update_dfs_bypass_settings(adev, new_ps); ret = kv_calculate_ds_divider(adev); if (ret) { - DRM_ERROR("kv_calculate_ds_divider failed\n"); + drm_err(adev_to_drm(adev), "kv_calculate_ds_divider failed\n"); return ret; } kv_calculate_nbps_level_settings(adev); @@ -1972,7 +1972,7 @@ static int kv_dpm_set_power_state(void *handle) kv_set_enabled_levels(adev); ret = kv_update_vce_dpm(adev, new_ps, old_ps); if (ret) { - DRM_ERROR("kv_update_vce_dpm failed\n"); + drm_err(adev_to_drm(adev), "kv_update_vce_dpm failed\n"); return ret; } kv_update_acp_boot_level(adev); @@ -2521,7 +2521,7 @@ static int kv_set_thermal_temperature_range(struct amdgpu_device *adev, if (high_temp > max_temp) high_temp = max_temp; if (high_temp < low_temp) { - DRM_ERROR("invalid thermal range: %d - %d\n", low_temp, high_temp); + drm_err(adev_to_drm(adev), "invalid thermal range: %d - %d\n", low_temp, high_temp); return -EINVAL; } @@ -2563,7 +2563,7 @@ static int kv_parse_sys_info_table(struct amdgpu_device *adev) data_offset); if (crev != 8) { - DRM_ERROR("Unsupported IGP table: %d %d\n", frev, crev); + drm_err(adev_to_drm(adev), "Unsupported IGP table: %d %d\n", frev, crev); return -EINVAL; } pi->sys_info.bootup_sclk = le32_to_cpu(igp_info->info_8.ulBootUpEngineClock); @@ -2579,7 +2579,7 @@ static int kv_parse_sys_info_table(struct amdgpu_device *adev) else pi->sys_info.htc_hyst_lmt = igp_info->info_8.ucHtcHystLmt; if (pi->sys_info.htc_tmp_lmt <= pi->sys_info.htc_hyst_lmt) { - DRM_ERROR("The htcTmpLmt should be larger than htcHystLmt.\n"); + drm_err(adev_to_drm(adev), "The htcTmpLmt should be larger than htcHystLmt.\n"); } if (le32_to_cpu(igp_info->info_8.ulSystemConfig) & (1 << 3)) @@ -2886,16 +2886,18 @@ kv_dpm_print_power_state(void *handle, void *request_ps) struct kv_ps *ps = kv_get_ps(rps); struct amdgpu_device *adev = (struct amdgpu_device *)handle; - amdgpu_dpm_print_class_info(rps->class, rps->class2); - amdgpu_dpm_print_cap_info(rps->caps); - printk("\tuvd vclk: %d dclk: %d\n", rps->vclk, rps->dclk); + amdgpu_dpm_dbg_print_class_info(adev, rps->class, rps->class2); + amdgpu_dpm_dbg_print_cap_info(adev, rps->caps); + drm_dbg(adev_to_drm(adev), "vclk: %d, dclk: %d\n", + rps->vclk, rps->dclk); for (i = 0; i < ps->num_levels; i++) { struct kv_pl *pl = &ps->levels[i]; - printk("\t\tpower level %d sclk: %u vddc: %u\n", - i, pl->sclk, - kv_convert_8bit_index_to_voltage(adev, pl->vddc_index)); + drm_dbg(adev_to_drm(adev), + "power level %d sclk: %u vddc: %u\n", + i, pl->sclk, + kv_convert_8bit_index_to_voltage(adev, pl->vddc_index)); } - amdgpu_dpm_print_ps_status(adev, rps); + amdgpu_dpm_dbg_print_ps_status(adev, rps); } static void kv_dpm_fini(struct amdgpu_device *adev) @@ -3013,13 +3015,13 @@ static int kv_dpm_sw_init(struct amdgpu_ip_block *ip_block) adev->pm.dpm.current_ps = adev->pm.dpm.requested_ps = adev->pm.dpm.boot_ps; if (amdgpu_dpm == 1) amdgpu_pm_print_power_states(adev); - DRM_INFO("amdgpu: dpm initialized\n"); + drm_info(adev_to_drm(adev), "dpm initialized\n"); return 0; dpm_failed: kv_dpm_fini(adev); - DRM_ERROR("amdgpu: dpm initialization failed\n"); + drm_err(adev_to_drm(adev), "dpm initialization failed: %d\n", ret); return ret; } diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c index c7518b13e787..ea3ace882a10 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c @@ -47,7 +47,7 @@ #define amdgpu_dpm_check_state_equal(adev, cps, rps, equal) \ ((adev)->powerplay.pp_funcs->check_state_equal((adev)->powerplay.pp_handle, (cps), (rps), (equal))) -void amdgpu_dpm_print_class_info(u32 class, u32 class2) +void amdgpu_dpm_dbg_print_class_info(struct amdgpu_device *adev, u32 class, u32 class2) { const char *s; @@ -66,71 +66,45 @@ void amdgpu_dpm_print_class_info(u32 class, u32 class2) s = "performance"; break; } - printk("\tui class: %s\n", s); - printk("\tinternal class:"); + drm_dbg(adev_to_drm(adev), "\tui class: %s\n", s); if (((class & ~ATOM_PPLIB_CLASSIFICATION_UI_MASK) == 0) && (class2 == 0)) - pr_cont(" none"); - else { - if (class & ATOM_PPLIB_CLASSIFICATION_BOOT) - pr_cont(" boot"); - if (class & ATOM_PPLIB_CLASSIFICATION_THERMAL) - pr_cont(" thermal"); - if (class & ATOM_PPLIB_CLASSIFICATION_LIMITEDPOWERSOURCE) - pr_cont(" limited_pwr"); - if (class & ATOM_PPLIB_CLASSIFICATION_REST) - pr_cont(" rest"); - if (class & ATOM_PPLIB_CLASSIFICATION_FORCED) - pr_cont(" forced"); - if (class & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE) - pr_cont(" 3d_perf"); - if (class & ATOM_PPLIB_CLASSIFICATION_OVERDRIVETEMPLATE) - pr_cont(" ovrdrv"); - if (class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE) - pr_cont(" uvd"); - if (class & ATOM_PPLIB_CLASSIFICATION_3DLOW) - pr_cont(" 3d_low"); - if (class & ATOM_PPLIB_CLASSIFICATION_ACPI) - pr_cont(" acpi"); - if (class & ATOM_PPLIB_CLASSIFICATION_HD2STATE) - pr_cont(" uvd_hd2"); - if (class & ATOM_PPLIB_CLASSIFICATION_HDSTATE) - pr_cont(" uvd_hd"); - if (class & ATOM_PPLIB_CLASSIFICATION_SDSTATE) - pr_cont(" uvd_sd"); - if (class2 & ATOM_PPLIB_CLASSIFICATION2_LIMITEDPOWERSOURCE_2) - pr_cont(" limited_pwr2"); - if (class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) - pr_cont(" ulv"); - if (class2 & ATOM_PPLIB_CLASSIFICATION2_MVC) - pr_cont(" uvd_mvc"); - } - pr_cont("\n"); + drm_dbg(adev_to_drm(adev), "\tinternal class: none\n"); + else + drm_dbg(adev_to_drm(adev), "\tinternal class: %s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s\n", + (class & ATOM_PPLIB_CLASSIFICATION_BOOT) ? " boot" : "", + (class & ATOM_PPLIB_CLASSIFICATION_THERMAL) ? " thermal" : "", + (class & ATOM_PPLIB_CLASSIFICATION_LIMITEDPOWERSOURCE) ? " limited_pwr" : "", + (class & ATOM_PPLIB_CLASSIFICATION_REST) ? " rest" : "", + (class & ATOM_PPLIB_CLASSIFICATION_FORCED) ? " forced" : "", + (class & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE) ? " 3d_perf" : "", + (class & ATOM_PPLIB_CLASSIFICATION_OVERDRIVETEMPLATE) ? " ovrdrv" : "", + (class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE) ? " uvd" : "", + (class & ATOM_PPLIB_CLASSIFICATION_3DLOW) ? " 3d_low" : "", + (class & ATOM_PPLIB_CLASSIFICATION_ACPI) ? " acpi" : "", + (class & ATOM_PPLIB_CLASSIFICATION_HD2STATE) ? " uvd_hd2" : "", + (class & ATOM_PPLIB_CLASSIFICATION_HDSTATE) ? " uvd_hd" : "", + (class & ATOM_PPLIB_CLASSIFICATION_SDSTATE) ? " uvd_sd" : "", + (class2 & ATOM_PPLIB_CLASSIFICATION2_LIMITEDPOWERSOURCE_2) ? " limited_pwr2" : "", + (class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) ? " ulv" : "", + (class2 & ATOM_PPLIB_CLASSIFICATION2_MVC) ? " uvd_mvc" : ""); } -void amdgpu_dpm_print_cap_info(u32 caps) +void amdgpu_dpm_dbg_print_cap_info(struct amdgpu_device *adev, u32 caps) { - printk("\tcaps:"); - if (caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) - pr_cont(" single_disp"); - if (caps & ATOM_PPLIB_SUPPORTS_VIDEO_PLAYBACK) - pr_cont(" video"); - if (caps & ATOM_PPLIB_DISALLOW_ON_DC) - pr_cont(" no_dc"); - pr_cont("\n"); + drm_dbg(adev_to_drm(adev), "\tcaps: %s%s%s\n", + (caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) ? " single_disp" : "", + (caps & ATOM_PPLIB_SUPPORTS_VIDEO_PLAYBACK) ? " video" : "", + (caps & ATOM_PPLIB_DISALLOW_ON_DC) ? " no_dc" : ""); } -void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev, +void amdgpu_dpm_dbg_print_ps_status(struct amdgpu_device *adev, struct amdgpu_ps *rps) { - printk("\tstatus:"); - if (rps == adev->pm.dpm.current_ps) - pr_cont(" c"); - if (rps == adev->pm.dpm.requested_ps) - pr_cont(" r"); - if (rps == adev->pm.dpm.boot_ps) - pr_cont(" b"); - pr_cont("\n"); + drm_dbg(adev_to_drm(adev), "\tstatus:%s%s%s\n", + rps == adev->pm.dpm.current_ps ? " c" : "", + rps == adev->pm.dpm.requested_ps ? " r" : "", + rps == adev->pm.dpm.boot_ps ? " b" : ""); } void amdgpu_pm_print_power_states(struct amdgpu_device *adev) @@ -699,64 +673,64 @@ void amdgpu_add_thermal_controller(struct amdgpu_device *adev) adev->pm.fan_max_rpm = controller->ucFanMaxRPM; } if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV6xx) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_RV6XX; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV770) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_RV770; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EVERGREEN) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_EVERGREEN; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_SUMO) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_SUMO; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_NISLANDS) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_NI; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_SISLANDS) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_SI; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_CISLANDS) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_CI; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_KAVERI) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_KV; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EXTERNAL_GPIO) { - DRM_INFO("External GPIO thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "External GPIO thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_EXTERNAL_GPIO; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_ADT7473_WITH_INTERNAL) { - DRM_INFO("ADT7473 with internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "ADT7473 with internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_ADT7473_WITH_INTERNAL; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EMC2103_WITH_INTERNAL) { - DRM_INFO("EMC2103 with internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "EMC2103 with internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_EMC2103_WITH_INTERNAL; } else if (controller->ucType < ARRAY_SIZE(pp_lib_thermal_controller_names)) { - DRM_INFO("Possible %s thermal controller at 0x%02x %s fan control\n", + drm_info(adev_to_drm(adev), "Possible %s thermal controller at 0x%02x %s fan control\n", pp_lib_thermal_controller_names[controller->ucType], controller->ucI2cAddress >> 1, (controller->ucFanParameters & @@ -772,7 +746,7 @@ void amdgpu_add_thermal_controller(struct amdgpu_device *adev) i2c_new_client_device(&adev->pm.i2c_bus->adapter, &info); } } else { - DRM_INFO("Unknown thermal controller type %d at 0x%02x %s fan control\n", + drm_info(adev_to_drm(adev), "Unknown thermal controller type %d at 0x%02x %s fan control\n", controller->ucType, controller->ucI2cAddress >> 1, (controller->ucFanParameters & @@ -943,9 +917,9 @@ static int amdgpu_dpm_change_power_state_locked(struct amdgpu_device *adev) return -EINVAL; if (amdgpu_dpm == 1 && pp_funcs->print_power_state) { - printk("switching from power state:\n"); + drm_dbg(adev_to_drm(adev), "switching from power state\n"); amdgpu_dpm_print_power_state(adev, adev->pm.dpm.current_ps); - printk("switching to power state:\n"); + drm_dbg(adev_to_drm(adev), "switching to power state\n"); amdgpu_dpm_print_power_state(adev, adev->pm.dpm.requested_ps); } diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h index 93bd3973330c..7120eef30509 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h @@ -23,10 +23,9 @@ #ifndef __LEGACY_DPM_H__ #define __LEGACY_DPM_H__ -void amdgpu_dpm_print_class_info(u32 class, u32 class2); -void amdgpu_dpm_print_cap_info(u32 caps); -void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev, - struct amdgpu_ps *rps); +void amdgpu_dpm_dbg_print_class_info(struct amdgpu_device *adev, u32 class, u32 class2); +void amdgpu_dpm_dbg_print_cap_info(struct amdgpu_device *adev, u32 caps); +void amdgpu_dpm_dbg_print_ps_status(struct amdgpu_device *adev, struct amdgpu_ps *rps); int amdgpu_get_platform_caps(struct amdgpu_device *adev); int amdgpu_parse_extended_power_table(struct amdgpu_device *adev); void amdgpu_free_extended_power_table(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c index 4c0e976004ba..52e732be59e3 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c @@ -7951,15 +7951,15 @@ static void si_dpm_print_power_state(void *handle, struct rv7xx_pl *pl; int i; - amdgpu_dpm_print_class_info(rps->class, rps->class2); - amdgpu_dpm_print_cap_info(rps->caps); - DRM_INFO("\tuvd vclk: %d dclk: %d\n", rps->vclk, rps->dclk); + amdgpu_dpm_dbg_print_class_info(adev, rps->class, rps->class2); + amdgpu_dpm_dbg_print_cap_info(adev, rps->caps); + drm_dbg(adev_to_drm(adev), "\tuvd vclk: %d dclk: %d\n", rps->vclk, rps->dclk); for (i = 0; i < ps->performance_level_count; i++) { pl = &ps->performance_levels[i]; - DRM_INFO("\t\tpower level %d sclk: %u mclk: %u vddc: %u vddci: %u pcie gen: %u\n", + drm_dbg(adev_to_drm(adev), "\t\tpower level %d sclk: %u mclk: %u vddc: %u vddci: %u pcie gen: %u\n", i, pl->sclk, pl->mclk, pl->vddc, pl->vddci, pl->pcie_gen + 1); } - amdgpu_dpm_print_ps_status(adev, rps); + amdgpu_dpm_dbg_print_ps_status(adev, rps); } static int si_dpm_early_init(struct amdgpu_ip_block *ip_block) diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index d79a1d94661a..756afe78a6e5 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -76,6 +76,7 @@ static void smu_power_profile_mode_get(struct smu_context *smu, enum PP_SMC_POWER_PROFILE profile_mode); static void smu_power_profile_mode_put(struct smu_context *smu, enum PP_SMC_POWER_PROFILE profile_mode); +static enum smu_clk_type smu_convert_to_smuclk(enum pp_clock_type type); static int smu_sys_get_pp_feature_mask(void *handle, char *buf) @@ -134,12 +135,17 @@ int smu_get_status_gfxoff(struct smu_context *smu, uint32_t *value) } int smu_set_soft_freq_range(struct smu_context *smu, - enum smu_clk_type clk_type, + enum pp_clock_type type, uint32_t min, uint32_t max) { + enum smu_clk_type clk_type; int ret = 0; + clk_type = smu_convert_to_smuclk(type); + if (clk_type == SMU_CLK_COUNT) + return -EINVAL; + if (smu->ppt_funcs->set_soft_freq_limited_range) ret = smu->ppt_funcs->set_soft_freq_limited_range(smu, clk_type, @@ -307,6 +313,26 @@ static int smu_dpm_set_vpe_enable(struct smu_context *smu, return ret; } +static int smu_dpm_set_isp_enable(struct smu_context *smu, + bool enable) +{ + struct smu_power_context *smu_power = &smu->smu_power; + struct smu_power_gate *power_gate = &smu_power->power_gate; + int ret; + + if (!smu->ppt_funcs->dpm_set_isp_enable) + return 0; + + if (atomic_read(&power_gate->isp_gated) ^ enable) + return 0; + + ret = smu->ppt_funcs->dpm_set_isp_enable(smu, enable); + if (!ret) + atomic_set(&power_gate->isp_gated, !enable); + + return ret; +} + static int smu_dpm_set_umsch_mm_enable(struct smu_context *smu, bool enable) { @@ -408,6 +434,12 @@ static int smu_dpm_set_power_gate(void *handle, dev_err(smu->adev->dev, "Failed to power %s VPE!\n", gate ? "gate" : "ungate"); break; + case AMD_IP_BLOCK_TYPE_ISP: + ret = smu_dpm_set_isp_enable(smu, !gate); + if (ret) + dev_err(smu->adev->dev, "Failed to power %s ISP!\n", + gate ? "gate" : "ungate"); + break; default: dev_err(smu->adev->dev, "Unsupported block type!\n"); return -EINVAL; @@ -1004,6 +1036,21 @@ static int smu_fini_fb_allocations(struct smu_context *smu) return 0; } +static void smu_update_gpu_addresses(struct smu_context *smu) +{ + struct smu_table_context *smu_table = &smu->smu_table; + struct smu_table *pm_status_table = smu_table->tables + SMU_TABLE_PMSTATUSLOG; + struct smu_table *driver_table = &(smu_table->driver_table); + struct smu_table *dummy_read_1_table = &smu_table->dummy_read_1_table; + + if (pm_status_table->bo) + pm_status_table->mc_address = amdgpu_bo_fb_aper_addr(pm_status_table->bo); + if (driver_table->bo) + driver_table->mc_address = amdgpu_bo_fb_aper_addr(driver_table->bo); + if (dummy_read_1_table->bo) + dummy_read_1_table->mc_address = amdgpu_bo_fb_aper_addr(dummy_read_1_table->bo); +} + /** * smu_alloc_memory_pool - allocate memory pool in the system memory * @@ -1285,6 +1332,7 @@ static int smu_sw_init(struct amdgpu_ip_block *ip_block) atomic_set(&smu->smu_power.power_gate.vcn_gated[i], 1); atomic_set(&smu->smu_power.power_gate.jpeg_gated, 1); atomic_set(&smu->smu_power.power_gate.vpe_gated, 1); + atomic_set(&smu->smu_power.power_gate.isp_gated, 1); atomic_set(&smu->smu_power.power_gate.umsch_mm_gated, 1); smu_init_power_profile(smu); @@ -1672,37 +1720,6 @@ static int smu_smc_hw_setup(struct smu_context *smu) } } - ret = smu_system_features_control(smu, true); - if (ret) { - dev_err(adev->dev, "Failed to enable requested dpm features!\n"); - return ret; - } - - smu_init_xgmi_plpd_mode(smu); - - ret = smu_feature_get_enabled_mask(smu, &features_supported); - if (ret) { - dev_err(adev->dev, "Failed to retrieve supported dpm features!\n"); - return ret; - } - bitmap_copy(feature->supported, - (unsigned long *)&features_supported, - feature->feature_num); - - if (!smu_is_dpm_running(smu)) - dev_info(adev->dev, "dpm has been disabled\n"); - - /* - * Set initialized values (get from vbios) to dpm tables context such as - * gfxclk, memclk, dcefclk, and etc. And enable the DPM feature for each - * type of clks. - */ - ret = smu_set_default_dpm_table(smu); - if (ret) { - dev_err(adev->dev, "Failed to setup default dpm clock tables!\n"); - return ret; - } - if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN5) pcie_gen = 4; else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4) @@ -1738,6 +1755,37 @@ static int smu_smc_hw_setup(struct smu_context *smu) return ret; } + ret = smu_system_features_control(smu, true); + if (ret) { + dev_err(adev->dev, "Failed to enable requested dpm features!\n"); + return ret; + } + + smu_init_xgmi_plpd_mode(smu); + + ret = smu_feature_get_enabled_mask(smu, &features_supported); + if (ret) { + dev_err(adev->dev, "Failed to retrieve supported dpm features!\n"); + return ret; + } + bitmap_copy(feature->supported, + (unsigned long *)&features_supported, + feature->feature_num); + + if (!smu_is_dpm_running(smu)) + dev_info(adev->dev, "dpm has been disabled\n"); + + /* + * Set initialized values (get from vbios) to dpm tables context such as + * gfxclk, memclk, dcefclk, and etc. And enable the DPM feature for each + * type of clks. + */ + ret = smu_set_default_dpm_table(smu); + if (ret) { + dev_err(adev->dev, "Failed to setup default dpm clock tables!\n"); + return ret; + } + ret = smu_get_thermal_temperature_range(smu); if (ret) { dev_err(adev->dev, "Failed to get thermal temperature ranges!\n"); @@ -1780,6 +1828,9 @@ static int smu_start_smc_engine(struct smu_context *smu) struct amdgpu_device *adev = smu->adev; int ret = 0; + if (amdgpu_virt_xgmi_migrate_enabled(adev)) + smu_update_gpu_addresses(smu); + smu->smc_fw_state = SMU_FW_INIT; if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { @@ -2935,6 +2986,12 @@ static enum smu_clk_type smu_convert_to_smuclk(enum pp_clock_type type) clk_type = SMU_DCLK; break; case PP_DCLK1: clk_type = SMU_DCLK1; break; + case PP_ISPICLK: + clk_type = SMU_ISPICLK; + break; + case PP_ISPXCLK: + clk_type = SMU_ISPXCLK; + break; case OD_SCLK: clk_type = SMU_OD_SCLK; break; case OD_MCLK: diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h index 9aacc7bc1c69..b52e194397e2 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h @@ -402,6 +402,7 @@ struct smu_power_gate { atomic_t vcn_gated[AMDGPU_MAX_VCN_INSTANCES]; atomic_t jpeg_gated; atomic_t vpe_gated; + atomic_t isp_gated; atomic_t umsch_mm_gated; }; @@ -1436,6 +1437,12 @@ struct pptable_funcs { int (*dpm_set_vpe_enable)(struct smu_context *smu, bool enable); /** + * @dpm_set_isp_enable: Enable/disable ISP engine dynamic power + * management. + */ + int (*dpm_set_isp_enable)(struct smu_context *smu, bool enable); + + /** * @dpm_set_umsch_mm_enable: Enable/disable UMSCH engine dynamic power * management. */ @@ -1635,7 +1642,7 @@ int smu_write_watermarks_table(struct smu_context *smu); int smu_get_dpm_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t *min, uint32_t *max); -int smu_set_soft_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, +int smu_set_soft_freq_range(struct smu_context *smu, enum pp_clock_type clk_type, uint32_t min, uint32_t max); int smu_set_gfx_power_up_by_imu(struct smu_context *smu); diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0_0.h index 1bc30db22f9c..cd44f4254134 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0_0.h @@ -106,6 +106,7 @@ typedef struct { #define NUM_FCLK_DPM_LEVELS 8 #define NUM_MEM_PSTATE_LEVELS 4 +#define ISP_ALL_TILES_MASK 0x7FF typedef struct { uint32_t UClk; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h index d7505cfc433a..0a2ca544f4e3 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h @@ -86,8 +86,10 @@ typedef enum { /*36*/ FEATURE_PIT = 36, /*37*/ FEATURE_DVO = 37, /*38*/ FEATURE_XVMINORPSM_CLKSTOP_DS = 38, +/*39*/ FEATURE_GLOBAL_DPM = 39, +/*40*/ FEATURE_NODE_POWER_MANAGER = 40, -/*39*/ NUM_FEATURES = 39 +/*41*/ NUM_FEATURES = 41 } FEATURE_LIST_e; //enum for MPIO PCIe gen speed msgs @@ -133,7 +135,7 @@ typedef enum { GFX_DVM_MARGIN_COUNT } GFX_DVM_MARGIN_e; -#define SMU_METRICS_TABLE_VERSION 0x12 +#define SMU_METRICS_TABLE_VERSION 0x13 typedef struct __attribute__((packed, aligned(4))) { uint64_t AccumulationCounter; @@ -275,6 +277,16 @@ typedef struct { //PSNs uint64_t PublicSerialNumber_AID[4]; uint64_t PublicSerialNumber_XCD[8]; + + //XGMI + uint32_t MaxXgmiWidth; + uint32_t MaxXgmiBitrate; + + // Telemetry + uint32_t InputTelemetryVoltageInmV; + + // General info + uint32_t pldmVersion[2]; } StaticMetricsTable_t; #pragma pack(pop) diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h index eefdaa0b5df6..d7a9e41820fa 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h @@ -305,6 +305,8 @@ enum smu_clk_type { SMU_MCLK, SMU_PCIE, SMU_LCLK, + SMU_ISPICLK, + SMU_ISPXCLK, SMU_OD_CCLK, SMU_OD_SCLK, SMU_OD_MCLK, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 7fad5dfb39c4..aac202d0c30e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -2444,7 +2444,8 @@ static int navi10_update_pcie_parameters(struct smu_context *smu, struct smu_11_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; PPTable_t *pptable = smu->smu_table.driver_pptable; uint32_t smu_pcie_arg; - int ret, i; + int ret = 0; + int i; /* lclk dpm table setup */ for (i = 0; i < MAX_PCIE_CONF; i++) { @@ -2453,25 +2454,27 @@ static int navi10_update_pcie_parameters(struct smu_context *smu, } for (i = 0; i < NUM_LINK_LEVELS; i++) { - smu_pcie_arg = (i << 16) | - ((pptable->PcieGenSpeed[i] <= pcie_gen_cap) ? (pptable->PcieGenSpeed[i] << 8) : - (pcie_gen_cap << 8)) | ((pptable->PcieLaneCount[i] <= pcie_width_cap) ? - pptable->PcieLaneCount[i] : pcie_width_cap); - ret = smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_OverridePcieParameters, - smu_pcie_arg, - NULL); - - if (ret) - return ret; - - if (pptable->PcieGenSpeed[i] > pcie_gen_cap) - dpm_context->dpm_tables.pcie_table.pcie_gen[i] = pcie_gen_cap; - if (pptable->PcieLaneCount[i] > pcie_width_cap) - dpm_context->dpm_tables.pcie_table.pcie_lane[i] = pcie_width_cap; + if (pptable->PcieGenSpeed[i] > pcie_gen_cap || + pptable->PcieLaneCount[i] > pcie_width_cap) { + dpm_context->dpm_tables.pcie_table.pcie_gen[i] = + pptable->PcieGenSpeed[i] > pcie_gen_cap ? + pcie_gen_cap : pptable->PcieGenSpeed[i]; + dpm_context->dpm_tables.pcie_table.pcie_lane[i] = + pptable->PcieLaneCount[i] > pcie_width_cap ? + pcie_width_cap : pptable->PcieLaneCount[i]; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_gen_cap << 8; + smu_pcie_arg |= pcie_width_cap; + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } } - return 0; + return ret; } static inline void navi10_dump_od_table(struct smu_context *smu, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index 115e3fa456bc..d57591509aed 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -2145,7 +2145,8 @@ static int sienna_cichlid_update_pcie_parameters(struct smu_context *smu, uint8_t min_gen_speed, max_gen_speed; uint8_t min_lane_width, max_lane_width; uint32_t smu_pcie_arg; - int ret, i; + int ret = 0; + int i; GET_PPTABLE_MEMBER(PcieGenSpeed, &table_member1); GET_PPTABLE_MEMBER(PcieLaneCount, &table_member2); @@ -2170,19 +2171,22 @@ static int sienna_cichlid_update_pcie_parameters(struct smu_context *smu, pcie_table->pcie_lane[1] = max_lane_width; for (i = 0; i < NUM_LINK_LEVELS; i++) { - smu_pcie_arg = (i << 16 | + if (!(smu->adev->pm.pp_feature & PP_PCIE_DPM_MASK) || + table_member1[i] > pcie_gen_cap || table_member2[i] > pcie_width_cap) { + smu_pcie_arg = (i << 16 | pcie_table->pcie_gen[i] << 8 | pcie_table->pcie_lane[i]); - ret = smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_OverridePcieParameters, - smu_pcie_arg, - NULL); - if (ret) - return ret; + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } } - return 0; + return ret; } static int sienna_cichlid_get_dpm_ultimate_freq(struct smu_context *smu, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c index a55ea76d7399..2c9869feba61 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c @@ -666,7 +666,6 @@ static int vangogh_print_clk_levels(struct smu_context *smu, { DpmClocks_t *clk_table = smu->smu_table.clocks_table; SmuMetrics_t metrics; - struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); int i, idx, size = 0, ret = 0; uint32_t cur_value = 0, value = 0, count = 0; bool cur_value_match_level = false; @@ -682,31 +681,25 @@ static int vangogh_print_clk_levels(struct smu_context *smu, switch (clk_type) { case SMU_OD_SCLK: - if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) { - size += sysfs_emit_at(buf, size, "%s:\n", "OD_SCLK"); - size += sysfs_emit_at(buf, size, "0: %10uMhz\n", - (smu->gfx_actual_hard_min_freq > 0) ? smu->gfx_actual_hard_min_freq : smu->gfx_default_hard_min_freq); - size += sysfs_emit_at(buf, size, "1: %10uMhz\n", - (smu->gfx_actual_soft_max_freq > 0) ? smu->gfx_actual_soft_max_freq : smu->gfx_default_soft_max_freq); - } + size += sysfs_emit_at(buf, size, "%s:\n", "OD_SCLK"); + size += sysfs_emit_at(buf, size, "0: %10uMhz\n", + (smu->gfx_actual_hard_min_freq > 0) ? smu->gfx_actual_hard_min_freq : smu->gfx_default_hard_min_freq); + size += sysfs_emit_at(buf, size, "1: %10uMhz\n", + (smu->gfx_actual_soft_max_freq > 0) ? smu->gfx_actual_soft_max_freq : smu->gfx_default_soft_max_freq); break; case SMU_OD_CCLK: - if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) { - size += sysfs_emit_at(buf, size, "CCLK_RANGE in Core%d:\n", smu->cpu_core_id_select); - size += sysfs_emit_at(buf, size, "0: %10uMhz\n", - (smu->cpu_actual_soft_min_freq > 0) ? smu->cpu_actual_soft_min_freq : smu->cpu_default_soft_min_freq); - size += sysfs_emit_at(buf, size, "1: %10uMhz\n", - (smu->cpu_actual_soft_max_freq > 0) ? smu->cpu_actual_soft_max_freq : smu->cpu_default_soft_max_freq); - } + size += sysfs_emit_at(buf, size, "CCLK_RANGE in Core%d:\n", smu->cpu_core_id_select); + size += sysfs_emit_at(buf, size, "0: %10uMhz\n", + (smu->cpu_actual_soft_min_freq > 0) ? smu->cpu_actual_soft_min_freq : smu->cpu_default_soft_min_freq); + size += sysfs_emit_at(buf, size, "1: %10uMhz\n", + (smu->cpu_actual_soft_max_freq > 0) ? smu->cpu_actual_soft_max_freq : smu->cpu_default_soft_max_freq); break; case SMU_OD_RANGE: - if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) { - size += sysfs_emit_at(buf, size, "%s:\n", "OD_RANGE"); - size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n", - smu->gfx_default_hard_min_freq, smu->gfx_default_soft_max_freq); - size += sysfs_emit_at(buf, size, "CCLK: %7uMhz %10uMhz\n", - smu->cpu_default_soft_min_freq, smu->cpu_default_soft_max_freq); - } + size += sysfs_emit_at(buf, size, "%s:\n", "OD_RANGE"); + size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n", + smu->gfx_default_hard_min_freq, smu->gfx_default_soft_max_freq); + size += sysfs_emit_at(buf, size, "CCLK: %7uMhz %10uMhz\n", + smu->cpu_default_soft_min_freq, smu->cpu_default_soft_max_freq); break; case SMU_SOCCLK: /* the level 3 ~ 6 of socclk use the same frequency for vangogh */ diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c index 9481f897432d..e97b0cf19197 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c @@ -497,7 +497,6 @@ static int renoir_print_clk_levels(struct smu_context *smu, int i, idx, size = 0, ret = 0; uint32_t cur_value = 0, value = 0, count = 0, min = 0, max = 0; SmuMetrics_t metrics; - struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); bool cur_value_match_level = false; memset(&metrics, 0, sizeof(metrics)); @@ -510,28 +509,24 @@ static int renoir_print_clk_levels(struct smu_context *smu, switch (clk_type) { case SMU_OD_RANGE: - if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) { - ret = smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_GetMinGfxclkFrequency, - 0, &min); - if (ret) - return ret; - ret = smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_GetMaxGfxclkFrequency, - 0, &max); - if (ret) - return ret; - size += sysfs_emit_at(buf, size, "OD_RANGE\nSCLK: %10uMhz %10uMhz\n", min, max); - } + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_GetMinGfxclkFrequency, + 0, &min); + if (ret) + return ret; + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_GetMaxGfxclkFrequency, + 0, &max); + if (ret) + return ret; + size += sysfs_emit_at(buf, size, "OD_RANGE\nSCLK: %10uMhz %10uMhz\n", min, max); break; case SMU_OD_SCLK: - if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) { - min = (smu->gfx_actual_hard_min_freq > 0) ? smu->gfx_actual_hard_min_freq : smu->gfx_default_hard_min_freq; - max = (smu->gfx_actual_soft_max_freq > 0) ? smu->gfx_actual_soft_max_freq : smu->gfx_default_soft_max_freq; - size += sysfs_emit_at(buf, size, "OD_SCLK\n"); - size += sysfs_emit_at(buf, size, "0:%10uMhz\n", min); - size += sysfs_emit_at(buf, size, "1:%10uMhz\n", max); - } + min = (smu->gfx_actual_hard_min_freq > 0) ? smu->gfx_actual_hard_min_freq : smu->gfx_default_hard_min_freq; + max = (smu->gfx_actual_soft_max_freq > 0) ? smu->gfx_actual_soft_max_freq : smu->gfx_default_soft_max_freq; + size += sysfs_emit_at(buf, size, "OD_SCLK\n"); + size += sysfs_emit_at(buf, size, "0:%10uMhz\n", min); + size += sysfs_emit_at(buf, size, "1:%10uMhz\n", max); break; case SMU_GFXCLK: case SMU_SCLK: diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index a7167668d189..1a1f2a6b2e52 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -58,6 +58,7 @@ MODULE_FIRMWARE("amdgpu/aldebaran_smc.bin"); MODULE_FIRMWARE("amdgpu/smu_13_0_0.bin"); +MODULE_FIRMWARE("amdgpu/smu_13_0_0_kicker.bin"); MODULE_FIRMWARE("amdgpu/smu_13_0_7.bin"); MODULE_FIRMWARE("amdgpu/smu_13_0_10.bin"); @@ -92,7 +93,7 @@ const int pmfw_decoded_link_width[7] = {0, 1, 2, 4, 8, 12, 16}; int smu_v13_0_init_microcode(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; - char ucode_prefix[15]; + char ucode_prefix[30]; int err = 0; const struct smc_firmware_header_v1_0 *hdr; const struct common_firmware_header *header; @@ -103,8 +104,13 @@ int smu_v13_0_init_microcode(struct smu_context *smu) return 0; amdgpu_ucode_ip_version_decode(adev, MP1_HWIP, ucode_prefix, sizeof(ucode_prefix)); - err = amdgpu_ucode_request(adev, &adev->pm.fw, AMDGPU_UCODE_REQUIRED, - "amdgpu/%s.bin", ucode_prefix); + + if (amdgpu_is_kicker_fw(adev)) + err = amdgpu_ucode_request(adev, &adev->pm.fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_kicker.bin", ucode_prefix); + else + err = amdgpu_ucode_request(adev, &adev->pm.fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s.bin", ucode_prefix); if (err) goto out; @@ -2380,7 +2386,8 @@ int smu_v13_0_update_pcie_parameters(struct smu_context *smu, &dpm_context->dpm_tables.pcie_table; int num_of_levels = pcie_table->num_of_link_levels; uint32_t smu_pcie_arg; - int ret, i; + int ret = 0; + int i; if (!num_of_levels) return 0; @@ -2396,30 +2403,38 @@ int smu_v13_0_update_pcie_parameters(struct smu_context *smu, for (i = 0; i < num_of_levels; i++) { pcie_table->pcie_gen[i] = pcie_gen_cap; pcie_table->pcie_lane[i] = pcie_width_cap; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; } } else { for (i = 0; i < num_of_levels; i++) { - if (pcie_table->pcie_gen[i] > pcie_gen_cap) + if (pcie_table->pcie_gen[i] > pcie_gen_cap || + pcie_table->pcie_lane[i] > pcie_width_cap) { pcie_table->pcie_gen[i] = pcie_gen_cap; - if (pcie_table->pcie_lane[i] > pcie_width_cap) pcie_table->pcie_lane[i] = pcie_width_cap; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } } } - for (i = 0; i < num_of_levels; i++) { - smu_pcie_arg = i << 16; - smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; - smu_pcie_arg |= pcie_table->pcie_lane[i]; - - ret = smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_OverridePcieParameters, - smu_pcie_arg, - NULL); - if (ret) - return ret; - } - - return 0; + return ret; } int smu_v13_0_disable_pmfw_state(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 5a9711e8cf68..e084ed99ec0e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -572,8 +572,6 @@ static int smu_v13_0_0_set_default_dpm_table(struct smu_context *smu) PPTable_t *pptable = table_context->driver_pptable; SkuTable_t *skutable = &pptable->SkuTable; struct smu_13_0_dpm_table *dpm_table; - struct smu_13_0_pcie_table *pcie_table; - uint32_t link_level; int ret = 0; /* socclk dpm table setup */ @@ -689,24 +687,6 @@ static int smu_v13_0_0_set_default_dpm_table(struct smu_context *smu) dpm_table->max = dpm_table->dpm_levels[0].value; } - /* lclk dpm table setup */ - pcie_table = &dpm_context->dpm_tables.pcie_table; - pcie_table->num_of_link_levels = 0; - for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { - if (!skutable->PcieGenSpeed[link_level] && - !skutable->PcieLaneCount[link_level] && - !skutable->LclkFreq[link_level]) - continue; - - pcie_table->pcie_gen[pcie_table->num_of_link_levels] = - skutable->PcieGenSpeed[link_level]; - pcie_table->pcie_lane[pcie_table->num_of_link_levels] = - skutable->PcieLaneCount[link_level]; - pcie_table->clk_freq[pcie_table->num_of_link_levels] = - skutable->LclkFreq[link_level]; - pcie_table->num_of_link_levels++; - } - /* dcefclk dpm table setup */ dpm_table = &dpm_context->dpm_tables.dcef_table; if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCN_BIT)) { @@ -3150,6 +3130,90 @@ static int smu_v13_0_0_set_power_limit(struct smu_context *smu, return 0; } +static int smu_v13_0_0_update_pcie_parameters(struct smu_context *smu, + uint8_t pcie_gen_cap, + uint8_t pcie_width_cap) +{ + struct smu_13_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; + struct smu_13_0_pcie_table *pcie_table = + &dpm_context->dpm_tables.pcie_table; + int num_of_levels; + uint32_t smu_pcie_arg; + uint32_t link_level; + struct smu_table_context *table_context = &smu->smu_table; + PPTable_t *pptable = table_context->driver_pptable; + SkuTable_t *skutable = &pptable->SkuTable; + int ret = 0; + int i; + + pcie_table->num_of_link_levels = 0; + + for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { + if (!skutable->PcieGenSpeed[link_level] && + !skutable->PcieLaneCount[link_level] && + !skutable->LclkFreq[link_level]) + continue; + + pcie_table->pcie_gen[pcie_table->num_of_link_levels] = + skutable->PcieGenSpeed[link_level]; + pcie_table->pcie_lane[pcie_table->num_of_link_levels] = + skutable->PcieLaneCount[link_level]; + pcie_table->clk_freq[pcie_table->num_of_link_levels] = + skutable->LclkFreq[link_level]; + pcie_table->num_of_link_levels++; + } + + num_of_levels = pcie_table->num_of_link_levels; + if (!num_of_levels) + return 0; + + if (!(smu->adev->pm.pp_feature & PP_PCIE_DPM_MASK)) { + if (pcie_table->pcie_gen[num_of_levels - 1] < pcie_gen_cap) + pcie_gen_cap = pcie_table->pcie_gen[num_of_levels - 1]; + + if (pcie_table->pcie_lane[num_of_levels - 1] < pcie_width_cap) + pcie_width_cap = pcie_table->pcie_lane[num_of_levels - 1]; + + /* Force all levels to use the same settings */ + for (i = 0; i < num_of_levels; i++) { + pcie_table->pcie_gen[i] = pcie_gen_cap; + pcie_table->pcie_lane[i] = pcie_width_cap; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } + } else { + for (i = 0; i < num_of_levels; i++) { + if (pcie_table->pcie_gen[i] > pcie_gen_cap || + pcie_table->pcie_lane[i] > pcie_width_cap) { + pcie_table->pcie_gen[i] = pcie_table->pcie_gen[i] > pcie_gen_cap ? + pcie_gen_cap : pcie_table->pcie_gen[i]; + pcie_table->pcie_lane[i] = pcie_table->pcie_lane[i] > pcie_width_cap ? + pcie_width_cap : pcie_table->pcie_lane[i]; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } + } + } + + return ret; +} + static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_0_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_0_set_default_dpm_table, @@ -3179,7 +3243,7 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .feature_is_enabled = smu_cmn_feature_is_enabled, .print_clk_levels = smu_v13_0_0_print_clk_levels, .force_clk_levels = smu_v13_0_0_force_clk_levels, - .update_pcie_parameters = smu_v13_0_update_pcie_parameters, + .update_pcie_parameters = smu_v13_0_0_update_pcie_parameters, .get_thermal_temperature_range = smu_v13_0_0_get_thermal_temperature_range, .register_irq_handler = smu_v13_0_register_irq_handler, .enable_thermal_alert = smu_v13_0_enable_thermal_alert, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c index e0d356f93ab0..b3adeb6e43a8 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c @@ -187,8 +187,34 @@ int smu_v13_0_12_get_max_metrics_size(void) return max(sizeof(StaticMetricsTable_t), sizeof(MetricsTable_t)); } +static void smu_v13_0_12_init_xgmi_data(struct smu_context *smu, + StaticMetricsTable_t *static_metrics) +{ + struct smu_table_context *smu_table = &smu->smu_table; + uint16_t max_speed; + uint8_t max_width; + int ret; + + if (smu_table->tables[SMU_TABLE_SMU_METRICS].version >= 0x13) { + max_width = (uint8_t)static_metrics->MaxXgmiWidth; + max_speed = (uint16_t)static_metrics->MaxXgmiBitrate; + ret = 0; + } else { + MetricsTable_t *metrics = (MetricsTable_t *)smu_table->metrics_table; + + ret = smu_v13_0_6_get_metrics_table(smu, NULL, true); + if (!ret) { + max_width = (uint8_t)metrics->XgmiWidth; + max_speed = (uint16_t)metrics->XgmiBitrate; + } + } + if (!ret) + amgpu_xgmi_set_max_speed_width(smu->adev, max_speed, max_width); +} + int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu) { + struct smu_13_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; struct smu_table_context *smu_table = &smu->smu_table; StaticMetricsTable_t *static_metrics = (StaticMetricsTable_t *)smu_table->metrics_table; struct PPTable_t *pptable = @@ -237,6 +263,18 @@ int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu) if (ret) return ret; + if (smu_v13_0_6_cap_supported(smu, SMU_CAP(BOARD_VOLTAGE))) { + if (!static_metrics->InputTelemetryVoltageInmV) { + dev_warn(smu->adev->dev, "Invalid board voltage %d\n", + static_metrics->InputTelemetryVoltageInmV); + } + dpm_context->board_volt = static_metrics->InputTelemetryVoltageInmV; + } + if (smu_v13_0_6_cap_supported(smu, SMU_CAP(PLDM_VERSION)) && + static_metrics->pldmVersion[0] != 0xFFFFFFFF) + smu->adev->firmware.pldm_version = + static_metrics->pldmVersion[0]; + smu_v13_0_12_init_xgmi_data(smu, static_metrics); pptable->Init = true; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c index f00ef7f3f355..68624afe7d83 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c @@ -345,6 +345,11 @@ static void smu_v13_0_12_init_caps(struct smu_context *smu) if (fw_ver >= 0x00562500) smu_v13_0_6_cap_set(smu, SMU_CAP(HST_LIMIT_METRICS)); + + if (fw_ver >= 0x04560100) { + smu_v13_0_6_cap_set(smu, SMU_CAP(BOARD_VOLTAGE)); + smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION)); + } } static void smu_v13_0_6_init_caps(struct smu_context *smu) @@ -685,8 +690,8 @@ static int smu_v13_0_6_get_allowed_feature_mask(struct smu_context *smu, return 0; } -static int smu_v13_0_6_get_metrics_table(struct smu_context *smu, - void *metrics_table, bool bypass_cache) +int smu_v13_0_6_get_metrics_table(struct smu_context *smu, void *metrics_table, + bool bypass_cache) { struct smu_table_context *smu_table = &smu->smu_table; uint32_t table_size = smu_table->tables[SMU_TABLE_SMU_METRICS].size; @@ -800,6 +805,8 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) int version = smu_v13_0_6_get_metrics_version(smu); int ret, i, retry = 100; uint32_t table_version; + uint16_t max_speed; + uint8_t max_width; if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12) && smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) @@ -835,6 +842,9 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) SMUQ10_ROUND(GET_METRIC_FIELD(MaxGfxclkFrequency, version)); pptable->MinGfxclkFrequency = SMUQ10_ROUND(GET_METRIC_FIELD(MinGfxclkFrequency, version)); + max_width = (uint8_t)GET_METRIC_FIELD(XgmiWidth, version); + max_speed = (uint16_t)GET_METRIC_FIELD(XgmiBitrate, version); + amgpu_xgmi_set_max_speed_width(smu->adev, max_speed, max_width); for (i = 0; i < 4; ++i) { pptable->FclkFrequencyTable[i] = @@ -1377,8 +1387,9 @@ static int smu_v13_0_6_print_clk_levels(struct smu_context *smu, return ret; } - min_clk = pstate_table->gfxclk_pstate.curr.min; - max_clk = pstate_table->gfxclk_pstate.curr.max; + single_dpm_table = &(dpm_context->dpm_tables.gfx_table); + min_clk = single_dpm_table->min; + max_clk = single_dpm_table->max; if (now < SMU_13_0_6_DSCLK_THRESHOLD) { size += sysfs_emit_at(buf, size, "S: %uMhz *\n", diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h index d38d6d76b1e7..67b30674fd31 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h @@ -74,6 +74,8 @@ enum smu_v13_0_6_caps { extern void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu); bool smu_v13_0_6_cap_supported(struct smu_context *smu, enum smu_v13_0_6_caps cap); int smu_v13_0_6_get_static_metrics_table(struct smu_context *smu); +int smu_v13_0_6_get_metrics_table(struct smu_context *smu, void *metrics_table, + bool bypass_cache); bool smu_v13_0_12_is_dpm_running(struct smu_context *smu); int smu_v13_0_12_get_max_metrics_size(void); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index c8f4f6fb4083..c96fa5e49ed6 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -579,8 +579,6 @@ static int smu_v13_0_7_set_default_dpm_table(struct smu_context *smu) PPTable_t *driver_ppt = smu->smu_table.driver_pptable; SkuTable_t *skutable = &driver_ppt->SkuTable; struct smu_13_0_dpm_table *dpm_table; - struct smu_13_0_pcie_table *pcie_table; - uint32_t link_level; int ret = 0; /* socclk dpm table setup */ @@ -687,24 +685,6 @@ static int smu_v13_0_7_set_default_dpm_table(struct smu_context *smu) dpm_table->max = dpm_table->dpm_levels[0].value; } - /* lclk dpm table setup */ - pcie_table = &dpm_context->dpm_tables.pcie_table; - pcie_table->num_of_link_levels = 0; - for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { - if (!skutable->PcieGenSpeed[link_level] && - !skutable->PcieLaneCount[link_level] && - !skutable->LclkFreq[link_level]) - continue; - - pcie_table->pcie_gen[pcie_table->num_of_link_levels] = - skutable->PcieGenSpeed[link_level]; - pcie_table->pcie_lane[pcie_table->num_of_link_levels] = - skutable->PcieLaneCount[link_level]; - pcie_table->clk_freq[pcie_table->num_of_link_levels] = - skutable->LclkFreq[link_level]; - pcie_table->num_of_link_levels++; - } - /* dcefclk dpm table setup */ dpm_table = &dpm_context->dpm_tables.dcef_table; if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCN_BIT)) { @@ -2739,6 +2719,89 @@ static int smu_v13_0_7_set_power_limit(struct smu_context *smu, return 0; } +static int smu_v13_0_7_update_pcie_parameters(struct smu_context *smu, + uint8_t pcie_gen_cap, + uint8_t pcie_width_cap) +{ + struct smu_13_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; + struct smu_13_0_pcie_table *pcie_table = + &dpm_context->dpm_tables.pcie_table; + int num_of_levels; + int link_level; + uint32_t smu_pcie_arg; + struct smu_table_context *table_context = &smu->smu_table; + PPTable_t *pptable = table_context->driver_pptable; + SkuTable_t *skutable = &pptable->SkuTable; + int ret = 0; + int i; + + pcie_table->num_of_link_levels = 0; + for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { + if (!skutable->PcieGenSpeed[link_level] && + !skutable->PcieLaneCount[link_level] && + !skutable->LclkFreq[link_level]) + continue; + + pcie_table->pcie_gen[pcie_table->num_of_link_levels] = + skutable->PcieGenSpeed[link_level]; + pcie_table->pcie_lane[pcie_table->num_of_link_levels] = + skutable->PcieLaneCount[link_level]; + pcie_table->clk_freq[pcie_table->num_of_link_levels] = + skutable->LclkFreq[link_level]; + pcie_table->num_of_link_levels++; + } + + num_of_levels = pcie_table->num_of_link_levels; + if (!num_of_levels) + return 0; + + if (!(smu->adev->pm.pp_feature & PP_PCIE_DPM_MASK)) { + if (pcie_table->pcie_gen[num_of_levels - 1] < pcie_gen_cap) + pcie_gen_cap = pcie_table->pcie_gen[num_of_levels - 1]; + + if (pcie_table->pcie_lane[num_of_levels - 1] < pcie_width_cap) + pcie_width_cap = pcie_table->pcie_lane[num_of_levels - 1]; + + /* Force all levels to use the same settings */ + for (i = 0; i < num_of_levels; i++) { + pcie_table->pcie_gen[i] = pcie_gen_cap; + pcie_table->pcie_lane[i] = pcie_width_cap; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } + } else { + for (i = 0; i < num_of_levels; i++) { + if (pcie_table->pcie_gen[i] > pcie_gen_cap || + pcie_table->pcie_lane[i] > pcie_width_cap) { + pcie_table->pcie_gen[i] = pcie_table->pcie_gen[i] > pcie_gen_cap ? + pcie_gen_cap : pcie_table->pcie_gen[i]; + pcie_table->pcie_lane[i] = pcie_table->pcie_lane[i] > pcie_width_cap ? + pcie_width_cap : pcie_table->pcie_lane[i]; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } + } + } + + return ret; +} + static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_7_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_7_set_default_dpm_table, @@ -2768,7 +2831,7 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .feature_is_enabled = smu_cmn_feature_is_enabled, .print_clk_levels = smu_v13_0_7_print_clk_levels, .force_clk_levels = smu_v13_0_7_force_clk_levels, - .update_pcie_parameters = smu_v13_0_update_pcie_parameters, + .update_pcie_parameters = smu_v13_0_7_update_pcie_parameters, .get_thermal_temperature_range = smu_v13_0_7_get_thermal_temperature_range, .register_irq_handler = smu_v13_0_register_irq_handler, .enable_thermal_alert = smu_v13_0_enable_thermal_alert, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c index 84f9b007b59f..fe00c84b1cc6 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c @@ -1207,11 +1207,13 @@ static int smu_v14_0_0_print_clk_levels(struct smu_context *smu, static int smu_v14_0_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, - uint32_t max) + u32 min, + u32 max, + bool __always_unused automatic) { - enum smu_message_type msg_set_min, msg_set_max; - int ret = 0; + enum smu_message_type msg_set_min = SMU_MSG_MAX_COUNT; + enum smu_message_type msg_set_max = SMU_MSG_MAX_COUNT; + int ret = -EINVAL; if (!smu_v14_0_0_clk_dpm_is_enabled(smu, clk_type)) return -EINVAL; @@ -1240,16 +1242,23 @@ static int smu_v14_0_0_set_soft_freq_limited_range(struct smu_context *smu, msg_set_min = SMU_MSG_SetHardMinVcn1; msg_set_max = SMU_MSG_SetSoftMaxVcn1; break; + case SMU_ISPICLK: + msg_set_min = SMU_MSG_SetHardMinIspiclkByFreq; + break; + case SMU_ISPXCLK: + msg_set_min = SMU_MSG_SetHardMinIspxclkByFreq; + break; default: return -EINVAL; } - ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min, NULL); - if (ret) - return ret; + if (min && msg_set_min != SMU_MSG_MAX_COUNT) + ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min, NULL); + + if (max && msg_set_max != SMU_MSG_MAX_COUNT) + ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_max, max, NULL); - return smu_cmn_send_smc_msg_with_param(smu, msg_set_max, - max, NULL); + return ret; } static int smu_v14_0_0_force_clk_levels(struct smu_context *smu, @@ -1278,7 +1287,7 @@ static int smu_v14_0_0_force_clk_levels(struct smu_context *smu, if (ret) break; - ret = smu_v14_0_0_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq); + ret = smu_v14_0_0_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq, false); break; default: ret = -EINVAL; @@ -1426,7 +1435,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_SCLK, sclk_min, - sclk_max); + sclk_max, + false); if (ret) return ret; @@ -1438,7 +1448,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_FCLK, fclk_min, - fclk_max); + fclk_max, + false); if (ret) return ret; } @@ -1447,7 +1458,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_SOCCLK, socclk_min, - socclk_max); + socclk_max, + false); if (ret) return ret; } @@ -1456,7 +1468,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_VCLK, vclk_min, - vclk_max); + vclk_max, + false); if (ret) return ret; } @@ -1465,7 +1478,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_VCLK1, vclk1_min, - vclk1_max); + vclk1_max, + false); if (ret) return ret; } @@ -1474,7 +1488,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_DCLK, dclk_min, - dclk_max); + dclk_max, + false); if (ret) return ret; } @@ -1483,7 +1498,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_DCLK1, dclk1_min, - dclk1_max); + dclk1_max, + false); if (ret) return ret; } @@ -1533,6 +1549,14 @@ static int smu_v14_0_0_set_vpe_enable(struct smu_context *smu, 0, NULL); } +static int smu_v14_0_0_set_isp_enable(struct smu_context *smu, + bool enable) +{ + return smu_cmn_send_smc_msg_with_param(smu, enable ? + SMU_MSG_PowerUpIspByTile : SMU_MSG_PowerDownIspByTile, + ISP_ALL_TILES_MASK, NULL); +} + static int smu_v14_0_0_set_umsch_mm_enable(struct smu_context *smu, bool enable) { @@ -1662,6 +1686,7 @@ static const struct pptable_funcs smu_v14_0_0_ppt_funcs = { .gfx_off_control = smu_v14_0_gfx_off_control, .mode2_reset = smu_v14_0_0_mode2_reset, .get_dpm_ultimate_freq = smu_v14_0_common_get_dpm_ultimate_freq, + .set_soft_freq_limited_range = smu_v14_0_0_set_soft_freq_limited_range, .od_edit_dpm_table = smu_v14_0_od_edit_dpm_table, .print_clk_levels = smu_v14_0_0_print_clk_levels, .force_clk_levels = smu_v14_0_0_force_clk_levels, @@ -1669,6 +1694,7 @@ static const struct pptable_funcs smu_v14_0_0_ppt_funcs = { .set_fine_grain_gfx_freq_parameters = smu_v14_0_common_set_fine_grain_gfx_freq_parameters, .set_gfx_power_up_by_imu = smu_v14_0_set_gfx_power_up_by_imu, .dpm_set_vpe_enable = smu_v14_0_0_set_vpe_enable, + .dpm_set_isp_enable = smu_v14_0_0_set_isp_enable, .dpm_set_umsch_mm_enable = smu_v14_0_0_set_umsch_mm_enable, .get_dpm_clock_table = smu_v14_0_common_get_dpm_table, .set_mall_enable = smu_v14_0_common_set_mall_enable, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c index 82c2db972491..3aea32baea3d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c @@ -502,8 +502,6 @@ static int smu_v14_0_2_set_default_dpm_table(struct smu_context *smu) PPTable_t *pptable = table_context->driver_pptable; SkuTable_t *skutable = &pptable->SkuTable; struct smu_14_0_dpm_table *dpm_table; - struct smu_14_0_pcie_table *pcie_table; - uint32_t link_level; int ret = 0; /* socclk dpm table setup */ @@ -619,27 +617,6 @@ static int smu_v14_0_2_set_default_dpm_table(struct smu_context *smu) dpm_table->max = dpm_table->dpm_levels[0].value; } - /* lclk dpm table setup */ - pcie_table = &dpm_context->dpm_tables.pcie_table; - pcie_table->num_of_link_levels = 0; - for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { - if (!skutable->PcieGenSpeed[link_level] && - !skutable->PcieLaneCount[link_level] && - !skutable->LclkFreq[link_level]) - continue; - - pcie_table->pcie_gen[pcie_table->num_of_link_levels] = - skutable->PcieGenSpeed[link_level]; - pcie_table->pcie_lane[pcie_table->num_of_link_levels] = - skutable->PcieLaneCount[link_level]; - pcie_table->clk_freq[pcie_table->num_of_link_levels] = - skutable->LclkFreq[link_level]; - pcie_table->num_of_link_levels++; - - if (link_level == 0) - link_level++; - } - /* dcefclk dpm table setup */ dpm_table = &dpm_context->dpm_tables.dcef_table; if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCN_BIT)) { @@ -1487,10 +1464,31 @@ static int smu_v14_0_2_update_pcie_parameters(struct smu_context *smu, struct smu_14_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; struct smu_14_0_pcie_table *pcie_table = &dpm_context->dpm_tables.pcie_table; - int num_of_levels = pcie_table->num_of_link_levels; + int num_of_levels; uint32_t smu_pcie_arg; - int ret, i; + uint32_t link_level; + struct smu_table_context *table_context = &smu->smu_table; + PPTable_t *pptable = table_context->driver_pptable; + SkuTable_t *skutable = &pptable->SkuTable; + int ret = 0; + int i; + + pcie_table->num_of_link_levels = 0; + for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { + if (!skutable->PcieGenSpeed[link_level] && + !skutable->PcieLaneCount[link_level] && + !skutable->LclkFreq[link_level]) + continue; + pcie_table->pcie_gen[pcie_table->num_of_link_levels] = + skutable->PcieGenSpeed[link_level]; + pcie_table->pcie_lane[pcie_table->num_of_link_levels] = + skutable->PcieLaneCount[link_level]; + pcie_table->clk_freq[pcie_table->num_of_link_levels] = + skutable->LclkFreq[link_level]; + pcie_table->num_of_link_levels++; + } + num_of_levels = pcie_table->num_of_link_levels; if (!num_of_levels) return 0; @@ -1505,30 +1503,40 @@ static int smu_v14_0_2_update_pcie_parameters(struct smu_context *smu, for (i = 0; i < num_of_levels; i++) { pcie_table->pcie_gen[i] = pcie_gen_cap; pcie_table->pcie_lane[i] = pcie_width_cap; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; } } else { for (i = 0; i < num_of_levels; i++) { - if (pcie_table->pcie_gen[i] > pcie_gen_cap) - pcie_table->pcie_gen[i] = pcie_gen_cap; - if (pcie_table->pcie_lane[i] > pcie_width_cap) - pcie_table->pcie_lane[i] = pcie_width_cap; - } - } - - for (i = 0; i < num_of_levels; i++) { - smu_pcie_arg = i << 16; - smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; - smu_pcie_arg |= pcie_table->pcie_lane[i]; - - ret = smu_cmn_send_smc_msg_with_param(smu, + if (pcie_table->pcie_gen[i] > pcie_gen_cap || + pcie_table->pcie_lane[i] > pcie_width_cap) { + pcie_table->pcie_gen[i] = pcie_table->pcie_gen[i] > pcie_gen_cap ? + pcie_gen_cap : pcie_table->pcie_gen[i]; + pcie_table->pcie_lane[i] = pcie_table->pcie_lane[i] > pcie_width_cap ? + pcie_width_cap : pcie_table->pcie_lane[i]; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_OverridePcieParameters, smu_pcie_arg, NULL); - if (ret) - return ret; + if (ret) + break; + } + } } - return 0; + return ret; } static const struct smu_temperature_range smu14_thermal_policy[] = { diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 7eaf58fd7f9a..59f9abd0f7b8 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -86,6 +86,7 @@ static void smu_cmn_read_arg(struct smu_context *smu, #define SMU_RESP_BUSY_OTHER 0xFC #define SMU_RESP_DEBUG_END 0xFB +#define SMU_RESP_UNEXP (~0U) /** * __smu_cmn_poll_stat -- poll for a status from the SMU * @smu: a pointer to SMU context @@ -171,6 +172,15 @@ static void __smu_cmn_reg_print_error(struct smu_context *smu, dev_err_ratelimited(adev->dev, "SMU: I'm debugging!"); break; + case SMU_RESP_UNEXP: + if (amdgpu_device_bus_status_check(smu->adev)) { + /* print error immediately if device is off the bus */ + dev_err(adev->dev, + "SMU: response:0x%08X for index:%d param:0x%08X message:%s?", + reg_c2pmsg_90, msg_index, param, message); + break; + } + fallthrough; default: dev_err_ratelimited(adev->dev, "SMU: response:0x%08X for index:%d param:0x%08X message:%s?", diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h index 7473672abd2a..a608cdbdada4 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h @@ -40,28 +40,29 @@ #define SMU_IH_INTERRUPT_CONTEXT_ID_FAN_ABNORMAL 0x8 #define SMU_IH_INTERRUPT_CONTEXT_ID_FAN_RECOVERY 0x9 -#define smu_cmn_init_soft_gpu_metrics(ptr, frev, crev) \ - do { \ - typecheck(struct gpu_metrics_v##frev##_##crev, \ - typeof(*(ptr))); \ - struct metrics_table_header *header = \ - (struct metrics_table_header *)(ptr); \ - memset(header, 0xFF, sizeof(*(ptr))); \ - header->format_revision = frev; \ - header->content_revision = crev; \ - header->structure_size = sizeof(*(ptr)); \ +#define smu_cmn_init_soft_gpu_metrics(ptr, frev, crev) \ + do { \ + typecheck(struct gpu_metrics_v##frev##_##crev *, (ptr)); \ + struct gpu_metrics_v##frev##_##crev *tmp = (ptr); \ + struct metrics_table_header *header = \ + (struct metrics_table_header *)tmp; \ + memset(header, 0xFF, sizeof(*tmp)); \ + header->format_revision = frev; \ + header->content_revision = crev; \ + header->structure_size = sizeof(*tmp); \ } while (0) -#define smu_cmn_init_partition_metrics(ptr, frev, crev) \ - do { \ - typecheck(struct amdgpu_partition_metrics_v##frev##_##crev, \ - typeof(*(ptr))); \ - struct metrics_table_header *header = \ - (struct metrics_table_header *)(ptr); \ - memset(header, 0xFF, sizeof(*(ptr))); \ - header->format_revision = frev; \ - header->content_revision = crev; \ - header->structure_size = sizeof(*(ptr)); \ +#define smu_cmn_init_partition_metrics(ptr, fr, cr) \ + do { \ + typecheck(struct amdgpu_partition_metrics_v##fr##_##cr *, \ + (ptr)); \ + struct amdgpu_partition_metrics_v##fr##_##cr *tmp = (ptr); \ + struct metrics_table_header *header = \ + (struct metrics_table_header *)tmp; \ + memset(header, 0xFF, sizeof(*tmp)); \ + header->format_revision = fr; \ + header->content_revision = cr; \ + header->structure_size = sizeof(*tmp); \ } while (0) extern const int link_speed[]; diff --git a/drivers/gpu/drm/ast/Makefile b/drivers/gpu/drm/ast/Makefile index 8d09ba5d5889..2547613155da 100644 --- a/drivers/gpu/drm/ast/Makefile +++ b/drivers/gpu/drm/ast/Makefile @@ -4,6 +4,11 @@ # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher. ast-y := \ + ast_2000.o \ + ast_2100.o \ + ast_2300.o \ + ast_2500.o \ + ast_2600.o \ ast_cursor.o \ ast_ddc.o \ ast_dp501.o \ diff --git a/drivers/gpu/drm/ast/ast_2000.c b/drivers/gpu/drm/ast/ast_2000.c new file mode 100644 index 000000000000..41c2aa1e425a --- /dev/null +++ b/drivers/gpu/drm/ast/ast_2000.c @@ -0,0 +1,149 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright 2012 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the + * "Software"), to deal in the Software without restriction, including + * without limitation the rights to use, copy, modify, merge, publish, + * distribute, sub license, and/or sell copies of the Software, and to + * permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, + * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR + * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE + * USE OR OTHER DEALINGS IN THE SOFTWARE. + * + * The above copyright notice and this permission notice (including the + * next paragraph) shall be included in all copies or substantial portions + * of the Software. + */ +/* + * Authors: Dave Airlie <airlied@redhat.com> + */ + +#include <linux/delay.h> + +#include "ast_drv.h" +#include "ast_post.h" + +/* + * POST + */ + +void ast_2000_set_def_ext_reg(struct ast_device *ast) +{ + static const u8 extreginfo[] = { 0x0f, 0x04, 0x1c, 0xff }; + u8 i, index, reg; + const u8 *ext_reg_info; + + /* reset scratch */ + for (i = 0x81; i <= 0x9f; i++) + ast_set_index_reg(ast, AST_IO_VGACRI, i, 0x00); + + ext_reg_info = extreginfo; + index = 0xa0; + while (*ext_reg_info != 0xff) { + ast_set_index_reg_mask(ast, AST_IO_VGACRI, index, 0x00, *ext_reg_info); + index++; + ext_reg_info++; + } + + /* disable standard IO/MEM decode if secondary */ + /* ast_set_index_reg-mask(ast, AST_IO_VGACRI, 0xa1, 0xff, 0x3); */ + + /* Set Ext. Default */ + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0x8c, 0x00, 0x01); + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xb7, 0x00, 0x00); + + /* Enable RAMDAC for A1 */ + reg = 0x04; + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xb6, 0xff, reg); +} + +static const struct ast_dramstruct ast2000_dram_table_data[] = { + { 0x0108, 0x00000000 }, + { 0x0120, 0x00004a21 }, + AST_DRAMSTRUCT_UDELAY(67u), + { 0x0000, 0xFFFFFFFF }, + AST_DRAMSTRUCT_INIT(DRAM_TYPE, 0x00000089), + { 0x0008, 0x22331353 }, + { 0x000C, 0x0d07000b }, + { 0x0010, 0x11113333 }, + { 0x0020, 0x00110350 }, + { 0x0028, 0x1e0828f0 }, + { 0x0024, 0x00000001 }, + { 0x001C, 0x00000000 }, + { 0x0014, 0x00000003 }, + AST_DRAMSTRUCT_UDELAY(67u), + { 0x0018, 0x00000131 }, + { 0x0014, 0x00000001 }, + AST_DRAMSTRUCT_UDELAY(67u), + { 0x0018, 0x00000031 }, + { 0x0014, 0x00000001 }, + AST_DRAMSTRUCT_UDELAY(67u), + { 0x0028, 0x1e0828f1 }, + { 0x0024, 0x00000003 }, + { 0x002C, 0x1f0f28fb }, + { 0x0030, 0xFFFFFE01 }, + AST_DRAMSTRUCT_INVALID, +}; + +static void ast_post_chip_2000(struct ast_device *ast) +{ + u8 j; + u32 temp, i; + const struct ast_dramstruct *dram_reg_info; + + j = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); + + if ((j & 0x80) == 0) { /* VGA only */ + dram_reg_info = ast2000_dram_table_data; + ast_write32(ast, 0xf004, 0x1e6e0000); + ast_write32(ast, 0xf000, 0x1); + ast_write32(ast, 0x10100, 0xa8); + + do { + ; + } while (ast_read32(ast, 0x10100) != 0xa8); + + while (!AST_DRAMSTRUCT_IS(dram_reg_info, INVALID)) { + if (AST_DRAMSTRUCT_IS(dram_reg_info, UDELAY)) { + for (i = 0; i < 15; i++) + udelay(dram_reg_info->data); + } else { + ast_write32(ast, 0x10000 + dram_reg_info->index, + dram_reg_info->data); + } + dram_reg_info++; + } + + temp = ast_read32(ast, 0x10140); + ast_write32(ast, 0x10140, temp | 0x40); + } + + /* wait ready */ + do { + j = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); + } while ((j & 0x40) == 0); +} + +int ast_2000_post(struct ast_device *ast) +{ + ast_2000_set_def_ext_reg(ast); + + if (ast->config_mode == ast_use_p2a) { + ast_post_chip_2000(ast); + } else { + if (ast->tx_chip == AST_TX_SIL164) { + /* Enable DVO */ + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); + } + } + + return 0; +} diff --git a/drivers/gpu/drm/ast/ast_2100.c b/drivers/gpu/drm/ast/ast_2100.c new file mode 100644 index 000000000000..477ee15eff5d --- /dev/null +++ b/drivers/gpu/drm/ast/ast_2100.c @@ -0,0 +1,348 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright 2012 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the + * "Software"), to deal in the Software without restriction, including + * without limitation the rights to use, copy, modify, merge, publish, + * distribute, sub license, and/or sell copies of the Software, and to + * permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, + * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR + * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE + * USE OR OTHER DEALINGS IN THE SOFTWARE. + * + * The above copyright notice and this permission notice (including the + * next paragraph) shall be included in all copies or substantial portions + * of the Software. + */ +/* + * Authors: Dave Airlie <airlied@redhat.com> + */ + +#include <linux/delay.h> + +#include "ast_drv.h" +#include "ast_post.h" + +/* + * POST + */ + +static const struct ast_dramstruct ast1100_dram_table_data[] = { + { 0x2000, 0x1688a8a8 }, + { 0x2020, 0x000041f0 }, + AST_DRAMSTRUCT_UDELAY(67u), + { 0x0000, 0xfc600309 }, + { 0x006C, 0x00909090 }, + { 0x0064, 0x00050000 }, + AST_DRAMSTRUCT_INIT(DRAM_TYPE, 0x00000585), + { 0x0008, 0x0011030f }, + { 0x0010, 0x22201724 }, + { 0x0018, 0x1e29011a }, + { 0x0020, 0x00c82222 }, + { 0x0014, 0x01001523 }, + { 0x001C, 0x1024010d }, + { 0x0024, 0x00cb2522 }, + { 0x0038, 0xffffff82 }, + { 0x003C, 0x00000000 }, + { 0x0040, 0x00000000 }, + { 0x0044, 0x00000000 }, + { 0x0048, 0x00000000 }, + { 0x004C, 0x00000000 }, + { 0x0050, 0x00000000 }, + { 0x0054, 0x00000000 }, + { 0x0058, 0x00000000 }, + { 0x005C, 0x00000000 }, + { 0x0060, 0x032aa02a }, + { 0x0064, 0x002d3000 }, + { 0x0068, 0x00000000 }, + { 0x0070, 0x00000000 }, + { 0x0074, 0x00000000 }, + { 0x0078, 0x00000000 }, + { 0x007C, 0x00000000 }, + { 0x0034, 0x00000001 }, + AST_DRAMSTRUCT_UDELAY(67u), + { 0x002C, 0x00000732 }, + { 0x0030, 0x00000040 }, + { 0x0028, 0x00000005 }, + { 0x0028, 0x00000007 }, + { 0x0028, 0x00000003 }, + { 0x0028, 0x00000001 }, + { 0x000C, 0x00005a08 }, + { 0x002C, 0x00000632 }, + { 0x0028, 0x00000001 }, + { 0x0030, 0x000003c0 }, + { 0x0028, 0x00000003 }, + { 0x0030, 0x00000040 }, + { 0x0028, 0x00000003 }, + { 0x000C, 0x00005a21 }, + { 0x0034, 0x00007c03 }, + { 0x0120, 0x00004c41 }, + AST_DRAMSTRUCT_INVALID, +}; + +static const struct ast_dramstruct ast2100_dram_table_data[] = { + { 0x2000, 0x1688a8a8 }, + { 0x2020, 0x00004120 }, + AST_DRAMSTRUCT_UDELAY(67u), + { 0x0000, 0xfc600309 }, + { 0x006C, 0x00909090 }, + { 0x0064, 0x00070000 }, + AST_DRAMSTRUCT_INIT(DRAM_TYPE, 0x00000489), + { 0x0008, 0x0011030f }, + { 0x0010, 0x32302926 }, + { 0x0018, 0x274c0122 }, + { 0x0020, 0x00ce2222 }, + { 0x0014, 0x01001523 }, + { 0x001C, 0x1024010d }, + { 0x0024, 0x00cb2522 }, + { 0x0038, 0xffffff82 }, + { 0x003C, 0x00000000 }, + { 0x0040, 0x00000000 }, + { 0x0044, 0x00000000 }, + { 0x0048, 0x00000000 }, + { 0x004C, 0x00000000 }, + { 0x0050, 0x00000000 }, + { 0x0054, 0x00000000 }, + { 0x0058, 0x00000000 }, + { 0x005C, 0x00000000 }, + { 0x0060, 0x0f2aa02a }, + { 0x0064, 0x003f3005 }, + { 0x0068, 0x02020202 }, + { 0x0070, 0x00000000 }, + { 0x0074, 0x00000000 }, + { 0x0078, 0x00000000 }, + { 0x007C, 0x00000000 }, + { 0x0034, 0x00000001 }, + AST_DRAMSTRUCT_UDELAY(67u), + { 0x002C, 0x00000942 }, + { 0x0030, 0x00000040 }, + { 0x0028, 0x00000005 }, + { 0x0028, 0x00000007 }, + { 0x0028, 0x00000003 }, + { 0x0028, 0x00000001 }, + { 0x000C, 0x00005a08 }, + { 0x002C, 0x00000842 }, + { 0x0028, 0x00000001 }, + { 0x0030, 0x000003c0 }, + { 0x0028, 0x00000003 }, + { 0x0030, 0x00000040 }, + { 0x0028, 0x00000003 }, + { 0x000C, 0x00005a21 }, + { 0x0034, 0x00007c03 }, + { 0x0120, 0x00005061 }, + AST_DRAMSTRUCT_INVALID, +}; + +/* + * AST2100/2150 DLL CBR Setting + */ +#define CBR_SIZE_AST2150 ((16 << 10) - 1) +#define CBR_PASSNUM_AST2150 5 +#define CBR_THRESHOLD_AST2150 10 +#define CBR_THRESHOLD2_AST2150 10 +#define TIMEOUT_AST2150 5000000 + +#define CBR_PATNUM_AST2150 8 + +static const u32 pattern_AST2150[14] = { + 0xFF00FF00, + 0xCC33CC33, + 0xAA55AA55, + 0xFFFE0001, + 0x683501FE, + 0x0F1929B0, + 0x2D0B4346, + 0x60767F02, + 0x6FBE36A6, + 0x3A253035, + 0x3019686D, + 0x41C6167E, + 0x620152BF, + 0x20F050E0 +}; + +static u32 mmctestburst2_ast2150(struct ast_device *ast, u32 datagen) +{ + u32 data, timeout; + + ast_moutdwm(ast, 0x1e6e0070, 0x00000000); + ast_moutdwm(ast, 0x1e6e0070, 0x00000001 | (datagen << 3)); + timeout = 0; + do { + data = ast_mindwm(ast, 0x1e6e0070) & 0x40; + if (++timeout > TIMEOUT_AST2150) { + ast_moutdwm(ast, 0x1e6e0070, 0x00000000); + return 0xffffffff; + } + } while (!data); + ast_moutdwm(ast, 0x1e6e0070, 0x00000000); + ast_moutdwm(ast, 0x1e6e0070, 0x00000003 | (datagen << 3)); + timeout = 0; + do { + data = ast_mindwm(ast, 0x1e6e0070) & 0x40; + if (++timeout > TIMEOUT_AST2150) { + ast_moutdwm(ast, 0x1e6e0070, 0x00000000); + return 0xffffffff; + } + } while (!data); + data = (ast_mindwm(ast, 0x1e6e0070) & 0x80) >> 7; + ast_moutdwm(ast, 0x1e6e0070, 0x00000000); + return data; +} + +static int cbrtest_ast2150(struct ast_device *ast) +{ + int i; + + for (i = 0; i < 8; i++) + if (mmctestburst2_ast2150(ast, i)) + return 0; + return 1; +} + +static int cbrscan_ast2150(struct ast_device *ast, int busw) +{ + u32 patcnt, loop; + + for (patcnt = 0; patcnt < CBR_PATNUM_AST2150; patcnt++) { + ast_moutdwm(ast, 0x1e6e007c, pattern_AST2150[patcnt]); + for (loop = 0; loop < CBR_PASSNUM_AST2150; loop++) { + if (cbrtest_ast2150(ast)) + break; + } + if (loop == CBR_PASSNUM_AST2150) + return 0; + } + return 1; +} + +static void cbrdlli_ast2150(struct ast_device *ast, int busw) +{ + u32 dll_min[4], dll_max[4], dlli, data, passcnt; + +cbr_start: + dll_min[0] = 0xff; + dll_min[1] = 0xff; + dll_min[2] = 0xff; + dll_min[3] = 0xff; + dll_max[0] = 0x00; + dll_max[1] = 0x00; + dll_max[2] = 0x00; + dll_max[3] = 0x00; + passcnt = 0; + + for (dlli = 0; dlli < 100; dlli++) { + ast_moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24)); + data = cbrscan_ast2150(ast, busw); + if (data != 0) { + if (data & 0x1) { + if (dll_min[0] > dlli) + dll_min[0] = dlli; + if (dll_max[0] < dlli) + dll_max[0] = dlli; + } + passcnt++; + } else if (passcnt >= CBR_THRESHOLD_AST2150) { + goto cbr_start; + } + } + if (dll_max[0] == 0 || (dll_max[0] - dll_min[0]) < CBR_THRESHOLD_AST2150) + goto cbr_start; + + dlli = dll_min[0] + (((dll_max[0] - dll_min[0]) * 7) >> 4); + ast_moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24)); +} + +static void ast_post_chip_2100(struct ast_device *ast) +{ + u8 j; + u32 data, temp, i; + const struct ast_dramstruct *dram_reg_info; + + j = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); + + if ((j & 0x80) == 0) { /* VGA only */ + if (ast->chip == AST2100 || ast->chip == AST2200) + dram_reg_info = ast2100_dram_table_data; + else + dram_reg_info = ast1100_dram_table_data; + + ast_write32(ast, 0xf004, 0x1e6e0000); + ast_write32(ast, 0xf000, 0x1); + ast_write32(ast, 0x12000, 0x1688A8A8); + do { + ; + } while (ast_read32(ast, 0x12000) != 0x01); + + ast_write32(ast, 0x10000, 0xfc600309); + do { + ; + } while (ast_read32(ast, 0x10000) != 0x01); + + while (!AST_DRAMSTRUCT_IS(dram_reg_info, INVALID)) { + if (AST_DRAMSTRUCT_IS(dram_reg_info, UDELAY)) { + for (i = 0; i < 15; i++) + udelay(dram_reg_info->data); + } else if (AST_DRAMSTRUCT_IS(dram_reg_info, DRAM_TYPE)) { + data = dram_reg_info->data; + if (ast->dram_type == AST_DRAM_1Gx16) + data = 0x00000d89; + else if (ast->dram_type == AST_DRAM_1Gx32) + data = 0x00000c8d; + + temp = ast_read32(ast, 0x12070); + temp &= 0xc; + temp <<= 2; + ast_write32(ast, 0x10000 + dram_reg_info->index, data | temp); + } else { + ast_write32(ast, 0x10000 + dram_reg_info->index, + dram_reg_info->data); + } + dram_reg_info++; + } + + /* AST 2100/2150 DRAM calibration */ + data = ast_read32(ast, 0x10120); + if (data == 0x5061) { /* 266Mhz */ + data = ast_read32(ast, 0x10004); + if (data & 0x40) + cbrdlli_ast2150(ast, 16); /* 16 bits */ + else + cbrdlli_ast2150(ast, 32); /* 32 bits */ + } + + temp = ast_read32(ast, 0x1200c); + ast_write32(ast, 0x1200c, temp & 0xfffffffd); + temp = ast_read32(ast, 0x12040); + ast_write32(ast, 0x12040, temp | 0x40); + } + + /* wait ready */ + do { + j = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); + } while ((j & 0x40) == 0); +} + +int ast_2100_post(struct ast_device *ast) +{ + ast_2000_set_def_ext_reg(ast); + + if (ast->config_mode == ast_use_p2a) { + ast_post_chip_2100(ast); + } else { + if (ast->tx_chip == AST_TX_SIL164) { + /* Enable DVO */ + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); + } + } + + return 0; +} diff --git a/drivers/gpu/drm/ast/ast_2300.c b/drivers/gpu/drm/ast/ast_2300.c new file mode 100644 index 000000000000..dc2a32244689 --- /dev/null +++ b/drivers/gpu/drm/ast/ast_2300.c @@ -0,0 +1,1328 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright 2012 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the + * "Software"), to deal in the Software without restriction, including + * without limitation the rights to use, copy, modify, merge, publish, + * distribute, sub license, and/or sell copies of the Software, and to + * permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, + * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR + * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE + * USE OR OTHER DEALINGS IN THE SOFTWARE. + * + * The above copyright notice and this permission notice (including the + * next paragraph) shall be included in all copies or substantial portions + * of the Software. + */ +/* + * Authors: Dave Airlie <airlied@redhat.com> + */ + +#include <linux/delay.h> + +#include "ast_drv.h" +#include "ast_post.h" + +/* + * POST + */ + +void ast_2300_set_def_ext_reg(struct ast_device *ast) +{ + static const u8 extreginfo[] = { 0x0f, 0x04, 0x1f, 0xff }; + u8 i, index, reg; + const u8 *ext_reg_info; + + /* reset scratch */ + for (i = 0x81; i <= 0x9f; i++) + ast_set_index_reg(ast, AST_IO_VGACRI, i, 0x00); + + ext_reg_info = extreginfo; + index = 0xa0; + while (*ext_reg_info != 0xff) { + ast_set_index_reg_mask(ast, AST_IO_VGACRI, index, 0x00, *ext_reg_info); + index++; + ext_reg_info++; + } + + /* disable standard IO/MEM decode if secondary */ + /* ast_set_index_reg-mask(ast, AST_IO_VGACRI, 0xa1, 0xff, 0x3); */ + + /* Set Ext. Default */ + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0x8c, 0x00, 0x01); + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xb7, 0x00, 0x00); + + /* Enable RAMDAC for A1 */ + reg = 0x04; + reg |= 0x20; + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xb6, 0xff, reg); +} + +/* AST 2300 DRAM settings */ +#define AST_DDR3 0 +#define AST_DDR2 1 + +struct ast2300_dram_param { + u32 dram_type; + u32 dram_chipid; + u32 dram_freq; + u32 vram_size; + u32 odt; + u32 wodt; + u32 rodt; + u32 dram_config; + u32 reg_PERIOD; + u32 reg_MADJ; + u32 reg_SADJ; + u32 reg_MRS; + u32 reg_EMRS; + u32 reg_AC1; + u32 reg_AC2; + u32 reg_DQSIC; + u32 reg_DRV; + u32 reg_IOZ; + u32 reg_DQIDLY; + u32 reg_FREQ; + u32 madj_max; + u32 dll2_finetune_step; +}; + +/* + * DQSI DLL CBR Setting + */ +#define CBR_SIZE0 ((1 << 10) - 1) +#define CBR_SIZE1 ((4 << 10) - 1) +#define CBR_SIZE2 ((64 << 10) - 1) +#define CBR_PASSNUM 5 +#define CBR_PASSNUM2 5 +#define CBR_THRESHOLD 10 +#define CBR_THRESHOLD2 10 +#define TIMEOUT 5000000 +#define CBR_PATNUM 8 + +static const u32 pattern[8] = { + 0xFF00FF00, + 0xCC33CC33, + 0xAA55AA55, + 0x88778877, + 0x92CC4D6E, + 0x543D3CDE, + 0xF1E843C7, + 0x7C61D253 +}; + +static u32 mmc_test2(struct ast_device *ast, u32 datagen, u8 test_ctl) +{ + u32 data, timeout; + + ast_moutdwm(ast, 0x1e6e0070, 0x00000000); + ast_moutdwm(ast, 0x1e6e0070, (datagen << 3) | test_ctl); + timeout = 0; + do { + data = ast_mindwm(ast, 0x1e6e0070) & 0x1000; + if (++timeout > TIMEOUT) { + ast_moutdwm(ast, 0x1e6e0070, 0x0); + return 0xffffffff; + } + } while (!data); + data = ast_mindwm(ast, 0x1e6e0078); + data = (data | (data >> 16)) & 0xffff; + ast_moutdwm(ast, 0x1e6e0070, 0x00000000); + return data; +} + +static u32 mmc_test_burst2(struct ast_device *ast, u32 datagen) +{ + return mmc_test2(ast, datagen, 0x41); +} + +static bool mmc_test_single(struct ast_device *ast, u32 datagen) +{ + return mmc_test(ast, datagen, 0xc5); +} + +static u32 mmc_test_single2(struct ast_device *ast, u32 datagen) +{ + return mmc_test2(ast, datagen, 0x05); +} + +static int cbr_test(struct ast_device *ast) +{ + u32 data; + int i; + + data = mmc_test_single2(ast, 0); + if ((data & 0xff) && (data & 0xff00)) + return 0; + for (i = 0; i < 8; i++) { + data = mmc_test_burst2(ast, i); + if ((data & 0xff) && (data & 0xff00)) + return 0; + } + if (!data) + return 3; + else if (data & 0xff) + return 2; + return 1; +} + +static int cbr_scan(struct ast_device *ast) +{ + u32 data, data2, patcnt, loop; + + data2 = 3; + for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) { + ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]); + for (loop = 0; loop < CBR_PASSNUM2; loop++) { + data = cbr_test(ast); + if (data != 0) { + data2 &= data; + if (!data2) + return 0; + break; + } + } + if (loop == CBR_PASSNUM2) + return 0; + } + return data2; +} + +static u32 cbr_test2(struct ast_device *ast) +{ + u32 data; + + data = mmc_test_burst2(ast, 0); + if (data == 0xffff) + return 0; + data |= mmc_test_single2(ast, 0); + if (data == 0xffff) + return 0; + + return ~data & 0xffff; +} + +static u32 cbr_scan2(struct ast_device *ast) +{ + u32 data, data2, patcnt, loop; + + data2 = 0xffff; + for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) { + ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]); + for (loop = 0; loop < CBR_PASSNUM2; loop++) { + data = cbr_test2(ast); + if (data != 0) { + data2 &= data; + if (!data2) + return 0; + break; + } + } + if (loop == CBR_PASSNUM2) + return 0; + } + return data2; +} + +static bool cbr_test3(struct ast_device *ast) +{ + if (!mmc_test_burst(ast, 0)) + return false; + if (!mmc_test_single(ast, 0)) + return false; + return true; +} + +static bool cbr_scan3(struct ast_device *ast) +{ + u32 patcnt, loop; + + for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) { + ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]); + for (loop = 0; loop < 2; loop++) { + if (cbr_test3(ast)) + break; + } + if (loop == 2) + return false; + } + return true; +} + +static bool finetuneDQI_L(struct ast_device *ast, struct ast2300_dram_param *param) +{ + u32 gold_sadj[2], dllmin[16], dllmax[16], dlli, data, cnt, mask, passcnt, retry = 0; + bool status = false; +FINETUNE_START: + for (cnt = 0; cnt < 16; cnt++) { + dllmin[cnt] = 0xff; + dllmax[cnt] = 0x0; + } + passcnt = 0; + for (dlli = 0; dlli < 76; dlli++) { + ast_moutdwm(ast, 0x1E6E0068, 0x00001400 | (dlli << 16) | (dlli << 24)); + ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE1); + data = cbr_scan2(ast); + if (data != 0) { + mask = 0x00010001; + for (cnt = 0; cnt < 16; cnt++) { + if (data & mask) { + if (dllmin[cnt] > dlli) + dllmin[cnt] = dlli; + if (dllmax[cnt] < dlli) + dllmax[cnt] = dlli; + } + mask <<= 1; + } + passcnt++; + } else if (passcnt >= CBR_THRESHOLD2) { + break; + } + } + gold_sadj[0] = 0x0; + passcnt = 0; + for (cnt = 0; cnt < 16; cnt++) { + if ((dllmax[cnt] > dllmin[cnt]) && + ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) { + gold_sadj[0] += dllmin[cnt]; + passcnt++; + } + } + if (retry++ > 10) + goto FINETUNE_DONE; + if (passcnt != 16) + goto FINETUNE_START; + status = true; +FINETUNE_DONE: + gold_sadj[0] = gold_sadj[0] >> 4; + gold_sadj[1] = gold_sadj[0]; + + data = 0; + for (cnt = 0; cnt < 8; cnt++) { + data >>= 3; + if ((dllmax[cnt] > dllmin[cnt]) && + ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) { + dlli = dllmin[cnt]; + if (gold_sadj[0] >= dlli) { + dlli = ((gold_sadj[0] - dlli) * 19) >> 5; + if (dlli > 3) + dlli = 3; + } else { + dlli = ((dlli - gold_sadj[0]) * 19) >> 5; + if (dlli > 4) + dlli = 4; + dlli = (8 - dlli) & 0x7; + } + data |= dlli << 21; + } + } + ast_moutdwm(ast, 0x1E6E0080, data); + + data = 0; + for (cnt = 8; cnt < 16; cnt++) { + data >>= 3; + if ((dllmax[cnt] > dllmin[cnt]) && + ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) { + dlli = dllmin[cnt]; + if (gold_sadj[1] >= dlli) { + dlli = ((gold_sadj[1] - dlli) * 19) >> 5; + if (dlli > 3) + dlli = 3; + else + dlli = (dlli - 1) & 0x7; + } else { + dlli = ((dlli - gold_sadj[1]) * 19) >> 5; + dlli += 1; + if (dlli > 4) + dlli = 4; + dlli = (8 - dlli) & 0x7; + } + data |= dlli << 21; + } + } + ast_moutdwm(ast, 0x1E6E0084, data); + return status; +} /* finetuneDQI_L */ + +static void finetuneDQSI(struct ast_device *ast) +{ + u32 dlli, dqsip, dqidly; + u32 reg_mcr18, reg_mcr0c, passcnt[2], diff; + u32 g_dqidly, g_dqsip, g_margin, g_side; + u16 pass[32][2][2]; + char tag[2][76]; + + /* Disable DQI CBR */ + reg_mcr0c = ast_mindwm(ast, 0x1E6E000C); + reg_mcr18 = ast_mindwm(ast, 0x1E6E0018); + reg_mcr18 &= 0x0000ffff; + ast_moutdwm(ast, 0x1E6E0018, reg_mcr18); + + for (dlli = 0; dlli < 76; dlli++) { + tag[0][dlli] = 0x0; + tag[1][dlli] = 0x0; + } + for (dqidly = 0; dqidly < 32; dqidly++) { + pass[dqidly][0][0] = 0xff; + pass[dqidly][0][1] = 0x0; + pass[dqidly][1][0] = 0xff; + pass[dqidly][1][1] = 0x0; + } + for (dqidly = 0; dqidly < 32; dqidly++) { + passcnt[0] = 0; + passcnt[1] = 0; + for (dqsip = 0; dqsip < 2; dqsip++) { + ast_moutdwm(ast, 0x1E6E000C, 0); + ast_moutdwm(ast, 0x1E6E0018, reg_mcr18 | (dqidly << 16) | (dqsip << 23)); + ast_moutdwm(ast, 0x1E6E000C, reg_mcr0c); + for (dlli = 0; dlli < 76; dlli++) { + ast_moutdwm(ast, 0x1E6E0068, + 0x00001300 | (dlli << 16) | (dlli << 24)); + ast_moutdwm(ast, 0x1E6E0070, 0); + ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE0); + if (cbr_scan3(ast)) { + if (dlli == 0) + break; + passcnt[dqsip]++; + tag[dqsip][dlli] = 'P'; + if (dlli < pass[dqidly][dqsip][0]) + pass[dqidly][dqsip][0] = (u16)dlli; + if (dlli > pass[dqidly][dqsip][1]) + pass[dqidly][dqsip][1] = (u16)dlli; + } else if (passcnt[dqsip] >= 5) { + break; + } else { + pass[dqidly][dqsip][0] = 0xff; + pass[dqidly][dqsip][1] = 0x0; + } + } + } + if (passcnt[0] == 0 && passcnt[1] == 0) + dqidly++; + } + /* Search margin */ + g_dqidly = 0; + g_dqsip = 0; + g_margin = 0; + g_side = 0; + + for (dqidly = 0; dqidly < 32; dqidly++) { + for (dqsip = 0; dqsip < 2; dqsip++) { + if (pass[dqidly][dqsip][0] > pass[dqidly][dqsip][1]) + continue; + diff = pass[dqidly][dqsip][1] - pass[dqidly][dqsip][0]; + if ((diff + 2) < g_margin) + continue; + passcnt[0] = 0; + passcnt[1] = 0; + for (dlli = pass[dqidly][dqsip][0]; + dlli > 0 && tag[dqsip][dlli] != 0; + dlli--, passcnt[0]++) { + } + for (dlli = pass[dqidly][dqsip][1]; + dlli < 76 && tag[dqsip][dlli] != 0; + dlli++, passcnt[1]++) { + } + if (passcnt[0] > passcnt[1]) + passcnt[0] = passcnt[1]; + passcnt[1] = 0; + if (passcnt[0] > g_side) + passcnt[1] = passcnt[0] - g_side; + if (diff > (g_margin + 1) && (passcnt[1] > 0 || passcnt[0] > 8)) { + g_margin = diff; + g_dqidly = dqidly; + g_dqsip = dqsip; + g_side = passcnt[0]; + } else if (passcnt[1] > 1 && g_side < 8) { + if (diff > g_margin) + g_margin = diff; + g_dqidly = dqidly; + g_dqsip = dqsip; + g_side = passcnt[0]; + } + } + } + reg_mcr18 = reg_mcr18 | (g_dqidly << 16) | (g_dqsip << 23); + ast_moutdwm(ast, 0x1E6E0018, reg_mcr18); +} + +static bool cbr_dll2(struct ast_device *ast, struct ast2300_dram_param *param) +{ + u32 dllmin[2], dllmax[2], dlli, data, passcnt, retry = 0; + bool status = false; + + finetuneDQSI(ast); + if (finetuneDQI_L(ast, param) == false) + return status; + +CBR_START2: + dllmin[0] = 0xff; + dllmin[1] = 0xff; + dllmax[0] = 0x0; + dllmax[1] = 0x0; + passcnt = 0; + for (dlli = 0; dlli < 76; dlli++) { + ast_moutdwm(ast, 0x1E6E0068, 0x00001300 | (dlli << 16) | (dlli << 24)); + ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE2); + data = cbr_scan(ast); + if (data != 0) { + if (data & 0x1) { + if (dllmin[0] > dlli) + dllmin[0] = dlli; + if (dllmax[0] < dlli) + dllmax[0] = dlli; + } + if (data & 0x2) { + if (dllmin[1] > dlli) + dllmin[1] = dlli; + if (dllmax[1] < dlli) + dllmax[1] = dlli; + } + passcnt++; + } else if (passcnt >= CBR_THRESHOLD) { + break; + } + } + if (retry++ > 10) + goto CBR_DONE2; + if (dllmax[0] == 0 || (dllmax[0] - dllmin[0]) < CBR_THRESHOLD) + goto CBR_START2; + if (dllmax[1] == 0 || (dllmax[1] - dllmin[1]) < CBR_THRESHOLD) + goto CBR_START2; + status = true; +CBR_DONE2: + dlli = (dllmin[1] + dllmax[1]) >> 1; + dlli <<= 8; + dlli += (dllmin[0] + dllmax[0]) >> 1; + ast_moutdwm(ast, 0x1E6E0068, ast_mindwm(ast, 0x1E720058) | (dlli << 16)); + return status; +} /* CBRDLL2 */ + +static void get_ddr3_info(struct ast_device *ast, struct ast2300_dram_param *param) +{ + u32 trap, trap_AC2, trap_MRS; + + ast_moutdwm(ast, 0x1E6E2000, 0x1688A8A8); + + /* Ger trap info */ + trap = (ast_mindwm(ast, 0x1E6E2070) >> 25) & 0x3; + trap_AC2 = 0x00020000 + (trap << 16); + trap_AC2 |= 0x00300000 + ((trap & 0x2) << 19); + trap_MRS = 0x00000010 + (trap << 4); + trap_MRS |= ((trap & 0x2) << 18); + + param->reg_MADJ = 0x00034C4C; + param->reg_SADJ = 0x00001800; + param->reg_DRV = 0x000000F0; + param->reg_PERIOD = param->dram_freq; + param->rodt = 0; + + switch (param->dram_freq) { + case 336: + ast_moutdwm(ast, 0x1E6E2020, 0x0190); + param->wodt = 0; + param->reg_AC1 = 0x22202725; + param->reg_AC2 = 0xAA007613 | trap_AC2; + param->reg_DQSIC = 0x000000BA; + param->reg_MRS = 0x04001400 | trap_MRS; + param->reg_EMRS = 0x00000000; + param->reg_IOZ = 0x00000023; + param->reg_DQIDLY = 0x00000074; + param->reg_FREQ = 0x00004DC0; + param->madj_max = 96; + param->dll2_finetune_step = 3; + switch (param->dram_chipid) { + default: + case AST_DRAM_512Mx16: + case AST_DRAM_1Gx16: + param->reg_AC2 = 0xAA007613 | trap_AC2; + break; + case AST_DRAM_2Gx16: + param->reg_AC2 = 0xAA00761C | trap_AC2; + break; + case AST_DRAM_4Gx16: + param->reg_AC2 = 0xAA007636 | trap_AC2; + break; + } + break; + default: + case 396: + ast_moutdwm(ast, 0x1E6E2020, 0x03F1); + param->wodt = 1; + param->reg_AC1 = 0x33302825; + param->reg_AC2 = 0xCC009617 | trap_AC2; + param->reg_DQSIC = 0x000000E2; + param->reg_MRS = 0x04001600 | trap_MRS; + param->reg_EMRS = 0x00000000; + param->reg_IOZ = 0x00000034; + param->reg_DRV = 0x000000FA; + param->reg_DQIDLY = 0x00000089; + param->reg_FREQ = 0x00005040; + param->madj_max = 96; + param->dll2_finetune_step = 4; + + switch (param->dram_chipid) { + default: + case AST_DRAM_512Mx16: + case AST_DRAM_1Gx16: + param->reg_AC2 = 0xCC009617 | trap_AC2; + break; + case AST_DRAM_2Gx16: + param->reg_AC2 = 0xCC009622 | trap_AC2; + break; + case AST_DRAM_4Gx16: + param->reg_AC2 = 0xCC00963F | trap_AC2; + break; + } + break; + + case 408: + ast_moutdwm(ast, 0x1E6E2020, 0x01F0); + param->wodt = 1; + param->reg_AC1 = 0x33302825; + param->reg_AC2 = 0xCC009617 | trap_AC2; + param->reg_DQSIC = 0x000000E2; + param->reg_MRS = 0x04001600 | trap_MRS; + param->reg_EMRS = 0x00000000; + param->reg_IOZ = 0x00000023; + param->reg_DRV = 0x000000FA; + param->reg_DQIDLY = 0x00000089; + param->reg_FREQ = 0x000050C0; + param->madj_max = 96; + param->dll2_finetune_step = 4; + + switch (param->dram_chipid) { + default: + case AST_DRAM_512Mx16: + case AST_DRAM_1Gx16: + param->reg_AC2 = 0xCC009617 | trap_AC2; + break; + case AST_DRAM_2Gx16: + param->reg_AC2 = 0xCC009622 | trap_AC2; + break; + case AST_DRAM_4Gx16: + param->reg_AC2 = 0xCC00963F | trap_AC2; + break; + } + + break; + case 456: + ast_moutdwm(ast, 0x1E6E2020, 0x0230); + param->wodt = 0; + param->reg_AC1 = 0x33302926; + param->reg_AC2 = 0xCD44961A; + param->reg_DQSIC = 0x000000FC; + param->reg_MRS = 0x00081830; + param->reg_EMRS = 0x00000000; + param->reg_IOZ = 0x00000045; + param->reg_DQIDLY = 0x00000097; + param->reg_FREQ = 0x000052C0; + param->madj_max = 88; + param->dll2_finetune_step = 4; + break; + case 504: + ast_moutdwm(ast, 0x1E6E2020, 0x0270); + param->wodt = 1; + param->reg_AC1 = 0x33302926; + param->reg_AC2 = 0xDE44A61D; + param->reg_DQSIC = 0x00000117; + param->reg_MRS = 0x00081A30; + param->reg_EMRS = 0x00000000; + param->reg_IOZ = 0x070000BB; + param->reg_DQIDLY = 0x000000A0; + param->reg_FREQ = 0x000054C0; + param->madj_max = 79; + param->dll2_finetune_step = 4; + break; + case 528: + ast_moutdwm(ast, 0x1E6E2020, 0x0290); + param->wodt = 1; + param->rodt = 1; + param->reg_AC1 = 0x33302926; + param->reg_AC2 = 0xEF44B61E; + param->reg_DQSIC = 0x00000125; + param->reg_MRS = 0x00081A30; + param->reg_EMRS = 0x00000040; + param->reg_DRV = 0x000000F5; + param->reg_IOZ = 0x00000023; + param->reg_DQIDLY = 0x00000088; + param->reg_FREQ = 0x000055C0; + param->madj_max = 76; + param->dll2_finetune_step = 3; + break; + case 576: + ast_moutdwm(ast, 0x1E6E2020, 0x0140); + param->reg_MADJ = 0x00136868; + param->reg_SADJ = 0x00004534; + param->wodt = 1; + param->rodt = 1; + param->reg_AC1 = 0x33302A37; + param->reg_AC2 = 0xEF56B61E; + param->reg_DQSIC = 0x0000013F; + param->reg_MRS = 0x00101A50; + param->reg_EMRS = 0x00000040; + param->reg_DRV = 0x000000FA; + param->reg_IOZ = 0x00000023; + param->reg_DQIDLY = 0x00000078; + param->reg_FREQ = 0x000057C0; + param->madj_max = 136; + param->dll2_finetune_step = 3; + break; + case 600: + ast_moutdwm(ast, 0x1E6E2020, 0x02E1); + param->reg_MADJ = 0x00136868; + param->reg_SADJ = 0x00004534; + param->wodt = 1; + param->rodt = 1; + param->reg_AC1 = 0x32302A37; + param->reg_AC2 = 0xDF56B61F; + param->reg_DQSIC = 0x0000014D; + param->reg_MRS = 0x00101A50; + param->reg_EMRS = 0x00000004; + param->reg_DRV = 0x000000F5; + param->reg_IOZ = 0x00000023; + param->reg_DQIDLY = 0x00000078; + param->reg_FREQ = 0x000058C0; + param->madj_max = 132; + param->dll2_finetune_step = 3; + break; + case 624: + ast_moutdwm(ast, 0x1E6E2020, 0x0160); + param->reg_MADJ = 0x00136868; + param->reg_SADJ = 0x00004534; + param->wodt = 1; + param->rodt = 1; + param->reg_AC1 = 0x32302A37; + param->reg_AC2 = 0xEF56B621; + param->reg_DQSIC = 0x0000015A; + param->reg_MRS = 0x02101A50; + param->reg_EMRS = 0x00000004; + param->reg_DRV = 0x000000F5; + param->reg_IOZ = 0x00000034; + param->reg_DQIDLY = 0x00000078; + param->reg_FREQ = 0x000059C0; + param->madj_max = 128; + param->dll2_finetune_step = 3; + break; + } /* switch freq */ + + switch (param->dram_chipid) { + case AST_DRAM_512Mx16: + param->dram_config = 0x130; + break; + default: + case AST_DRAM_1Gx16: + param->dram_config = 0x131; + break; + case AST_DRAM_2Gx16: + param->dram_config = 0x132; + break; + case AST_DRAM_4Gx16: + param->dram_config = 0x133; + break; + } /* switch size */ + + switch (param->vram_size) { + default: + case SZ_8M: + param->dram_config |= 0x00; + break; + case SZ_16M: + param->dram_config |= 0x04; + break; + case SZ_32M: + param->dram_config |= 0x08; + break; + case SZ_64M: + param->dram_config |= 0x0c; + break; + } +} + +static void ddr3_init(struct ast_device *ast, struct ast2300_dram_param *param) +{ + u32 data, data2, retry = 0; + +ddr3_init_start: + ast_moutdwm(ast, 0x1E6E0000, 0xFC600309); + ast_moutdwm(ast, 0x1E6E0018, 0x00000100); + ast_moutdwm(ast, 0x1E6E0024, 0x00000000); + ast_moutdwm(ast, 0x1E6E0034, 0x00000000); + udelay(10); + ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ); + ast_moutdwm(ast, 0x1E6E0068, param->reg_SADJ); + udelay(10); + ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000); + udelay(10); + + ast_moutdwm(ast, 0x1E6E0004, param->dram_config); + ast_moutdwm(ast, 0x1E6E0008, 0x90040f); + ast_moutdwm(ast, 0x1E6E0010, param->reg_AC1); + ast_moutdwm(ast, 0x1E6E0014, param->reg_AC2); + ast_moutdwm(ast, 0x1E6E0020, param->reg_DQSIC); + ast_moutdwm(ast, 0x1E6E0080, 0x00000000); + ast_moutdwm(ast, 0x1E6E0084, 0x00000000); + ast_moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY); + ast_moutdwm(ast, 0x1E6E0018, 0x4000A170); + ast_moutdwm(ast, 0x1E6E0018, 0x00002370); + ast_moutdwm(ast, 0x1E6E0038, 0x00000000); + ast_moutdwm(ast, 0x1E6E0040, 0xFF444444); + ast_moutdwm(ast, 0x1E6E0044, 0x22222222); + ast_moutdwm(ast, 0x1E6E0048, 0x22222222); + ast_moutdwm(ast, 0x1E6E004C, 0x00000002); + ast_moutdwm(ast, 0x1E6E0050, 0x80000000); + ast_moutdwm(ast, 0x1E6E0050, 0x00000000); + ast_moutdwm(ast, 0x1E6E0054, 0); + ast_moutdwm(ast, 0x1E6E0060, param->reg_DRV); + ast_moutdwm(ast, 0x1E6E006C, param->reg_IOZ); + ast_moutdwm(ast, 0x1E6E0070, 0x00000000); + ast_moutdwm(ast, 0x1E6E0074, 0x00000000); + ast_moutdwm(ast, 0x1E6E0078, 0x00000000); + ast_moutdwm(ast, 0x1E6E007C, 0x00000000); + /* Wait MCLK2X lock to MCLK */ + do { + data = ast_mindwm(ast, 0x1E6E001C); + } while (!(data & 0x08000000)); + data = ast_mindwm(ast, 0x1E6E001C); + data = (data >> 8) & 0xff; + while ((data & 0x08) || ((data & 0x7) < 2) || (data < 4)) { + data2 = (ast_mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4; + if ((data2 & 0xff) > param->madj_max) + break; + ast_moutdwm(ast, 0x1E6E0064, data2); + if (data2 & 0x00100000) + data2 = ((data2 & 0xff) >> 3) + 3; + else + data2 = ((data2 & 0xff) >> 2) + 5; + data = ast_mindwm(ast, 0x1E6E0068) & 0xffff00ff; + data2 += data & 0xff; + data = data | (data2 << 8); + ast_moutdwm(ast, 0x1E6E0068, data); + udelay(10); + ast_moutdwm(ast, 0x1E6E0064, ast_mindwm(ast, 0x1E6E0064) | 0xC0000); + udelay(10); + data = ast_mindwm(ast, 0x1E6E0018) & 0xfffff1ff; + ast_moutdwm(ast, 0x1E6E0018, data); + data = data | 0x200; + ast_moutdwm(ast, 0x1E6E0018, data); + do { + data = ast_mindwm(ast, 0x1E6E001C); + } while (!(data & 0x08000000)); + + data = ast_mindwm(ast, 0x1E6E001C); + data = (data >> 8) & 0xff; + } + ast_moutdwm(ast, 0x1E720058, ast_mindwm(ast, 0x1E6E0068) & 0xffff); + data = ast_mindwm(ast, 0x1E6E0018) | 0xC00; + ast_moutdwm(ast, 0x1E6E0018, data); + + ast_moutdwm(ast, 0x1E6E0034, 0x00000001); + ast_moutdwm(ast, 0x1E6E000C, 0x00000040); + udelay(50); + /* Mode Register Setting */ + ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100); + ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS); + ast_moutdwm(ast, 0x1E6E0028, 0x00000005); + ast_moutdwm(ast, 0x1E6E0028, 0x00000007); + ast_moutdwm(ast, 0x1E6E0028, 0x00000003); + ast_moutdwm(ast, 0x1E6E0028, 0x00000001); + ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS); + ast_moutdwm(ast, 0x1E6E000C, 0x00005C08); + ast_moutdwm(ast, 0x1E6E0028, 0x00000001); + + ast_moutdwm(ast, 0x1E6E000C, 0x00005C01); + data = 0; + if (param->wodt) + data = 0x300; + if (param->rodt) + data = data | 0x3000 | ((param->reg_AC2 & 0x60000) >> 3); + ast_moutdwm(ast, 0x1E6E0034, data | 0x3); + + /* Calibrate the DQSI delay */ + if ((cbr_dll2(ast, param) == false) && (retry++ < 10)) + goto ddr3_init_start; + + ast_moutdwm(ast, 0x1E6E0120, param->reg_FREQ); + /* ECC Memory Initialization */ +#ifdef ECC + ast_moutdwm(ast, 0x1E6E007C, 0x00000000); + ast_moutdwm(ast, 0x1E6E0070, 0x221); + do { + data = ast_mindwm(ast, 0x1E6E0070); + } while (!(data & 0x00001000)); + ast_moutdwm(ast, 0x1E6E0070, 0x00000000); + ast_moutdwm(ast, 0x1E6E0050, 0x80000000); + ast_moutdwm(ast, 0x1E6E0050, 0x00000000); +#endif +} + +static void get_ddr2_info(struct ast_device *ast, struct ast2300_dram_param *param) +{ + u32 trap, trap_AC2, trap_MRS; + + ast_moutdwm(ast, 0x1E6E2000, 0x1688A8A8); + + /* Ger trap info */ + trap = (ast_mindwm(ast, 0x1E6E2070) >> 25) & 0x3; + trap_AC2 = (trap << 20) | (trap << 16); + trap_AC2 += 0x00110000; + trap_MRS = 0x00000040 | (trap << 4); + + param->reg_MADJ = 0x00034C4C; + param->reg_SADJ = 0x00001800; + param->reg_DRV = 0x000000F0; + param->reg_PERIOD = param->dram_freq; + param->rodt = 0; + + switch (param->dram_freq) { + case 264: + ast_moutdwm(ast, 0x1E6E2020, 0x0130); + param->wodt = 0; + param->reg_AC1 = 0x11101513; + param->reg_AC2 = 0x78117011; + param->reg_DQSIC = 0x00000092; + param->reg_MRS = 0x00000842; + param->reg_EMRS = 0x00000000; + param->reg_DRV = 0x000000F0; + param->reg_IOZ = 0x00000034; + param->reg_DQIDLY = 0x0000005A; + param->reg_FREQ = 0x00004AC0; + param->madj_max = 138; + param->dll2_finetune_step = 3; + break; + case 336: + ast_moutdwm(ast, 0x1E6E2020, 0x0190); + param->wodt = 1; + param->reg_AC1 = 0x22202613; + param->reg_AC2 = 0xAA009016 | trap_AC2; + param->reg_DQSIC = 0x000000BA; + param->reg_MRS = 0x00000A02 | trap_MRS; + param->reg_EMRS = 0x00000040; + param->reg_DRV = 0x000000FA; + param->reg_IOZ = 0x00000034; + param->reg_DQIDLY = 0x00000074; + param->reg_FREQ = 0x00004DC0; + param->madj_max = 96; + param->dll2_finetune_step = 3; + switch (param->dram_chipid) { + default: + case AST_DRAM_512Mx16: + param->reg_AC2 = 0xAA009012 | trap_AC2; + break; + case AST_DRAM_1Gx16: + param->reg_AC2 = 0xAA009016 | trap_AC2; + break; + case AST_DRAM_2Gx16: + param->reg_AC2 = 0xAA009023 | trap_AC2; + break; + case AST_DRAM_4Gx16: + param->reg_AC2 = 0xAA00903B | trap_AC2; + break; + } + break; + default: + case 396: + ast_moutdwm(ast, 0x1E6E2020, 0x03F1); + param->wodt = 1; + param->rodt = 0; + param->reg_AC1 = 0x33302714; + param->reg_AC2 = 0xCC00B01B | trap_AC2; + param->reg_DQSIC = 0x000000E2; + param->reg_MRS = 0x00000C02 | trap_MRS; + param->reg_EMRS = 0x00000040; + param->reg_DRV = 0x000000FA; + param->reg_IOZ = 0x00000034; + param->reg_DQIDLY = 0x00000089; + param->reg_FREQ = 0x00005040; + param->madj_max = 96; + param->dll2_finetune_step = 4; + + switch (param->dram_chipid) { + case AST_DRAM_512Mx16: + param->reg_AC2 = 0xCC00B016 | trap_AC2; + break; + default: + case AST_DRAM_1Gx16: + param->reg_AC2 = 0xCC00B01B | trap_AC2; + break; + case AST_DRAM_2Gx16: + param->reg_AC2 = 0xCC00B02B | trap_AC2; + break; + case AST_DRAM_4Gx16: + param->reg_AC2 = 0xCC00B03F | trap_AC2; + break; + } + + break; + + case 408: + ast_moutdwm(ast, 0x1E6E2020, 0x01F0); + param->wodt = 1; + param->rodt = 0; + param->reg_AC1 = 0x33302714; + param->reg_AC2 = 0xCC00B01B | trap_AC2; + param->reg_DQSIC = 0x000000E2; + param->reg_MRS = 0x00000C02 | trap_MRS; + param->reg_EMRS = 0x00000040; + param->reg_DRV = 0x000000FA; + param->reg_IOZ = 0x00000034; + param->reg_DQIDLY = 0x00000089; + param->reg_FREQ = 0x000050C0; + param->madj_max = 96; + param->dll2_finetune_step = 4; + + switch (param->dram_chipid) { + case AST_DRAM_512Mx16: + param->reg_AC2 = 0xCC00B016 | trap_AC2; + break; + default: + case AST_DRAM_1Gx16: + param->reg_AC2 = 0xCC00B01B | trap_AC2; + break; + case AST_DRAM_2Gx16: + param->reg_AC2 = 0xCC00B02B | trap_AC2; + break; + case AST_DRAM_4Gx16: + param->reg_AC2 = 0xCC00B03F | trap_AC2; + break; + } + + break; + case 456: + ast_moutdwm(ast, 0x1E6E2020, 0x0230); + param->wodt = 0; + param->reg_AC1 = 0x33302815; + param->reg_AC2 = 0xCD44B01E; + param->reg_DQSIC = 0x000000FC; + param->reg_MRS = 0x00000E72; + param->reg_EMRS = 0x00000000; + param->reg_DRV = 0x00000000; + param->reg_IOZ = 0x00000034; + param->reg_DQIDLY = 0x00000097; + param->reg_FREQ = 0x000052C0; + param->madj_max = 88; + param->dll2_finetune_step = 3; + break; + case 504: + ast_moutdwm(ast, 0x1E6E2020, 0x0261); + param->wodt = 1; + param->rodt = 1; + param->reg_AC1 = 0x33302815; + param->reg_AC2 = 0xDE44C022; + param->reg_DQSIC = 0x00000117; + param->reg_MRS = 0x00000E72; + param->reg_EMRS = 0x00000040; + param->reg_DRV = 0x0000000A; + param->reg_IOZ = 0x00000045; + param->reg_DQIDLY = 0x000000A0; + param->reg_FREQ = 0x000054C0; + param->madj_max = 79; + param->dll2_finetune_step = 3; + break; + case 528: + ast_moutdwm(ast, 0x1E6E2020, 0x0120); + param->wodt = 1; + param->rodt = 1; + param->reg_AC1 = 0x33302815; + param->reg_AC2 = 0xEF44D024; + param->reg_DQSIC = 0x00000125; + param->reg_MRS = 0x00000E72; + param->reg_EMRS = 0x00000004; + param->reg_DRV = 0x000000F9; + param->reg_IOZ = 0x00000045; + param->reg_DQIDLY = 0x000000A7; + param->reg_FREQ = 0x000055C0; + param->madj_max = 76; + param->dll2_finetune_step = 3; + break; + case 552: + ast_moutdwm(ast, 0x1E6E2020, 0x02A1); + param->wodt = 1; + param->rodt = 1; + param->reg_AC1 = 0x43402915; + param->reg_AC2 = 0xFF44E025; + param->reg_DQSIC = 0x00000132; + param->reg_MRS = 0x00000E72; + param->reg_EMRS = 0x00000040; + param->reg_DRV = 0x0000000A; + param->reg_IOZ = 0x00000045; + param->reg_DQIDLY = 0x000000AD; + param->reg_FREQ = 0x000056C0; + param->madj_max = 76; + param->dll2_finetune_step = 3; + break; + case 576: + ast_moutdwm(ast, 0x1E6E2020, 0x0140); + param->wodt = 1; + param->rodt = 1; + param->reg_AC1 = 0x43402915; + param->reg_AC2 = 0xFF44E027; + param->reg_DQSIC = 0x0000013F; + param->reg_MRS = 0x00000E72; + param->reg_EMRS = 0x00000004; + param->reg_DRV = 0x000000F5; + param->reg_IOZ = 0x00000045; + param->reg_DQIDLY = 0x000000B3; + param->reg_FREQ = 0x000057C0; + param->madj_max = 76; + param->dll2_finetune_step = 3; + break; + } + + switch (param->dram_chipid) { + case AST_DRAM_512Mx16: + param->dram_config = 0x100; + break; + default: + case AST_DRAM_1Gx16: + param->dram_config = 0x121; + break; + case AST_DRAM_2Gx16: + param->dram_config = 0x122; + break; + case AST_DRAM_4Gx16: + param->dram_config = 0x123; + break; + } /* switch size */ + + switch (param->vram_size) { + default: + case SZ_8M: + param->dram_config |= 0x00; + break; + case SZ_16M: + param->dram_config |= 0x04; + break; + case SZ_32M: + param->dram_config |= 0x08; + break; + case SZ_64M: + param->dram_config |= 0x0c; + break; + } +} + +static void ddr2_init(struct ast_device *ast, struct ast2300_dram_param *param) +{ + u32 data, data2, retry = 0; + +ddr2_init_start: + ast_moutdwm(ast, 0x1E6E0000, 0xFC600309); + ast_moutdwm(ast, 0x1E6E0018, 0x00000100); + ast_moutdwm(ast, 0x1E6E0024, 0x00000000); + ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ); + ast_moutdwm(ast, 0x1E6E0068, param->reg_SADJ); + udelay(10); + ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000); + udelay(10); + + ast_moutdwm(ast, 0x1E6E0004, param->dram_config); + ast_moutdwm(ast, 0x1E6E0008, 0x90040f); + ast_moutdwm(ast, 0x1E6E0010, param->reg_AC1); + ast_moutdwm(ast, 0x1E6E0014, param->reg_AC2); + ast_moutdwm(ast, 0x1E6E0020, param->reg_DQSIC); + ast_moutdwm(ast, 0x1E6E0080, 0x00000000); + ast_moutdwm(ast, 0x1E6E0084, 0x00000000); + ast_moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY); + ast_moutdwm(ast, 0x1E6E0018, 0x4000A130); + ast_moutdwm(ast, 0x1E6E0018, 0x00002330); + ast_moutdwm(ast, 0x1E6E0038, 0x00000000); + ast_moutdwm(ast, 0x1E6E0040, 0xFF808000); + ast_moutdwm(ast, 0x1E6E0044, 0x88848466); + ast_moutdwm(ast, 0x1E6E0048, 0x44440008); + ast_moutdwm(ast, 0x1E6E004C, 0x00000000); + ast_moutdwm(ast, 0x1E6E0050, 0x80000000); + ast_moutdwm(ast, 0x1E6E0050, 0x00000000); + ast_moutdwm(ast, 0x1E6E0054, 0); + ast_moutdwm(ast, 0x1E6E0060, param->reg_DRV); + ast_moutdwm(ast, 0x1E6E006C, param->reg_IOZ); + ast_moutdwm(ast, 0x1E6E0070, 0x00000000); + ast_moutdwm(ast, 0x1E6E0074, 0x00000000); + ast_moutdwm(ast, 0x1E6E0078, 0x00000000); + ast_moutdwm(ast, 0x1E6E007C, 0x00000000); + + /* Wait MCLK2X lock to MCLK */ + do { + data = ast_mindwm(ast, 0x1E6E001C); + } while (!(data & 0x08000000)); + data = ast_mindwm(ast, 0x1E6E001C); + data = (data >> 8) & 0xff; + while ((data & 0x08) || ((data & 0x7) < 2) || (data < 4)) { + data2 = (ast_mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4; + if ((data2 & 0xff) > param->madj_max) + break; + ast_moutdwm(ast, 0x1E6E0064, data2); + if (data2 & 0x00100000) + data2 = ((data2 & 0xff) >> 3) + 3; + else + data2 = ((data2 & 0xff) >> 2) + 5; + data = ast_mindwm(ast, 0x1E6E0068) & 0xffff00ff; + data2 += data & 0xff; + data = data | (data2 << 8); + ast_moutdwm(ast, 0x1E6E0068, data); + udelay(10); + ast_moutdwm(ast, 0x1E6E0064, ast_mindwm(ast, 0x1E6E0064) | 0xC0000); + udelay(10); + data = ast_mindwm(ast, 0x1E6E0018) & 0xfffff1ff; + ast_moutdwm(ast, 0x1E6E0018, data); + data = data | 0x200; + ast_moutdwm(ast, 0x1E6E0018, data); + do { + data = ast_mindwm(ast, 0x1E6E001C); + } while (!(data & 0x08000000)); + + data = ast_mindwm(ast, 0x1E6E001C); + data = (data >> 8) & 0xff; + } + ast_moutdwm(ast, 0x1E720058, ast_mindwm(ast, 0x1E6E0008) & 0xffff); + data = ast_mindwm(ast, 0x1E6E0018) | 0xC00; + ast_moutdwm(ast, 0x1E6E0018, data); + + ast_moutdwm(ast, 0x1E6E0034, 0x00000001); + ast_moutdwm(ast, 0x1E6E000C, 0x00000000); + udelay(50); + /* Mode Register Setting */ + ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100); + ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS); + ast_moutdwm(ast, 0x1E6E0028, 0x00000005); + ast_moutdwm(ast, 0x1E6E0028, 0x00000007); + ast_moutdwm(ast, 0x1E6E0028, 0x00000003); + ast_moutdwm(ast, 0x1E6E0028, 0x00000001); + + ast_moutdwm(ast, 0x1E6E000C, 0x00005C08); + ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS); + ast_moutdwm(ast, 0x1E6E0028, 0x00000001); + ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS | 0x380); + ast_moutdwm(ast, 0x1E6E0028, 0x00000003); + ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS); + ast_moutdwm(ast, 0x1E6E0028, 0x00000003); + + ast_moutdwm(ast, 0x1E6E000C, 0x7FFF5C01); + data = 0; + if (param->wodt) + data = 0x500; + if (param->rodt) + data = data | 0x3000 | ((param->reg_AC2 & 0x60000) >> 3); + ast_moutdwm(ast, 0x1E6E0034, data | 0x3); + ast_moutdwm(ast, 0x1E6E0120, param->reg_FREQ); + + /* Calibrate the DQSI delay */ + if ((cbr_dll2(ast, param) == false) && (retry++ < 10)) + goto ddr2_init_start; + + /* ECC Memory Initialization */ +#ifdef ECC + ast_moutdwm(ast, 0x1E6E007C, 0x00000000); + ast_moutdwm(ast, 0x1E6E0070, 0x221); + do { + data = ast_mindwm(ast, 0x1E6E0070); + } while (!(data & 0x00001000)); + ast_moutdwm(ast, 0x1E6E0070, 0x00000000); + ast_moutdwm(ast, 0x1E6E0050, 0x80000000); + ast_moutdwm(ast, 0x1E6E0050, 0x00000000); +#endif +} + +static void ast_post_chip_2300(struct ast_device *ast) +{ + struct ast2300_dram_param param; + u32 temp; + u8 reg; + + reg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); + if ((reg & 0x80) == 0) {/* vga only */ + ast_write32(ast, 0xf004, 0x1e6e0000); + ast_write32(ast, 0xf000, 0x1); + ast_write32(ast, 0x12000, 0x1688a8a8); + do { + ; + } while (ast_read32(ast, 0x12000) != 0x1); + + ast_write32(ast, 0x10000, 0xfc600309); + do { + ; + } while (ast_read32(ast, 0x10000) != 0x1); + + /* Slow down CPU/AHB CLK in VGA only mode */ + temp = ast_read32(ast, 0x12008); + temp |= 0x73; + ast_write32(ast, 0x12008, temp); + + param.dram_freq = 396; + param.dram_type = AST_DDR3; + temp = ast_mindwm(ast, 0x1e6e2070); + if (temp & 0x01000000) + param.dram_type = AST_DDR2; + switch (temp & 0x18000000) { + case 0: + param.dram_chipid = AST_DRAM_512Mx16; + break; + default: + case 0x08000000: + param.dram_chipid = AST_DRAM_1Gx16; + break; + case 0x10000000: + param.dram_chipid = AST_DRAM_2Gx16; + break; + case 0x18000000: + param.dram_chipid = AST_DRAM_4Gx16; + break; + } + switch (temp & 0x0c) { + default: + case 0x00: + param.vram_size = SZ_8M; + break; + case 0x04: + param.vram_size = SZ_16M; + break; + case 0x08: + param.vram_size = SZ_32M; + break; + case 0x0c: + param.vram_size = SZ_64M; + break; + } + + if (param.dram_type == AST_DDR3) { + get_ddr3_info(ast, ¶m); + ddr3_init(ast, ¶m); + } else { + get_ddr2_info(ast, ¶m); + ddr2_init(ast, ¶m); + } + + temp = ast_mindwm(ast, 0x1e6e2040); + ast_moutdwm(ast, 0x1e6e2040, temp | 0x40); + } + + /* wait ready */ + do { + reg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); + } while ((reg & 0x40) == 0); +} + +int ast_2300_post(struct ast_device *ast) +{ + ast_2300_set_def_ext_reg(ast); + + if (ast->config_mode == ast_use_p2a) { + ast_post_chip_2300(ast); + ast_init_3rdtx(ast); + } else { + if (ast->tx_chip == AST_TX_SIL164) { + /* Enable DVO */ + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); + } + } + + return 0; +} diff --git a/drivers/gpu/drm/ast/ast_2500.c b/drivers/gpu/drm/ast/ast_2500.c new file mode 100644 index 000000000000..1e541498ea67 --- /dev/null +++ b/drivers/gpu/drm/ast/ast_2500.c @@ -0,0 +1,569 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright 2012 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the + * "Software"), to deal in the Software without restriction, including + * without limitation the rights to use, copy, modify, merge, publish, + * distribute, sub license, and/or sell copies of the Software, and to + * permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, + * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR + * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE + * USE OR OTHER DEALINGS IN THE SOFTWARE. + * + * The above copyright notice and this permission notice (including the + * next paragraph) shall be included in all copies or substantial portions + * of the Software. + */ +/* + * Authors: Dave Airlie <airlied@redhat.com> + */ + +#include <linux/delay.h> + +#include <drm/drm_print.h> + +#include "ast_drv.h" +#include "ast_post.h" + +/* + * POST + */ + +/* + * AST2500 DRAM settings modules + */ + +#define REGTBL_NUM 17 +#define REGIDX_010 0 +#define REGIDX_014 1 +#define REGIDX_018 2 +#define REGIDX_020 3 +#define REGIDX_024 4 +#define REGIDX_02C 5 +#define REGIDX_030 6 +#define REGIDX_214 7 +#define REGIDX_2E0 8 +#define REGIDX_2E4 9 +#define REGIDX_2E8 10 +#define REGIDX_2EC 11 +#define REGIDX_2F0 12 +#define REGIDX_2F4 13 +#define REGIDX_2F8 14 +#define REGIDX_RFC 15 +#define REGIDX_PLL 16 + +static const u32 ast2500_ddr3_1600_timing_table[REGTBL_NUM] = { + 0x64604D38, /* 0x010 */ + 0x29690599, /* 0x014 */ + 0x00000300, /* 0x018 */ + 0x00000000, /* 0x020 */ + 0x00000000, /* 0x024 */ + 0x02181E70, /* 0x02C */ + 0x00000040, /* 0x030 */ + 0x00000024, /* 0x214 */ + 0x02001300, /* 0x2E0 */ + 0x0E0000A0, /* 0x2E4 */ + 0x000E001B, /* 0x2E8 */ + 0x35B8C105, /* 0x2EC */ + 0x08090408, /* 0x2F0 */ + 0x9B000800, /* 0x2F4 */ + 0x0E400A00, /* 0x2F8 */ + 0x9971452F, /* tRFC */ + 0x000071C1 /* PLL */ +}; + +static const u32 ast2500_ddr4_1600_timing_table[REGTBL_NUM] = { + 0x63604E37, /* 0x010 */ + 0xE97AFA99, /* 0x014 */ + 0x00019000, /* 0x018 */ + 0x08000000, /* 0x020 */ + 0x00000400, /* 0x024 */ + 0x00000410, /* 0x02C */ + 0x00000101, /* 0x030 */ + 0x00000024, /* 0x214 */ + 0x03002900, /* 0x2E0 */ + 0x0E0000A0, /* 0x2E4 */ + 0x000E001C, /* 0x2E8 */ + 0x35B8C106, /* 0x2EC */ + 0x08080607, /* 0x2F0 */ + 0x9B000900, /* 0x2F4 */ + 0x0E400A00, /* 0x2F8 */ + 0x99714545, /* tRFC */ + 0x000071C1 /* PLL */ +}; + +#define TIMEOUT 5000000 + +void ast_2500_patch_ahb(void __iomem *regs) +{ + u32 data; + + /* Clear bus lock condition */ + __ast_moutdwm(regs, 0x1e600000, 0xAEED1A03); + __ast_moutdwm(regs, 0x1e600084, 0x00010000); + __ast_moutdwm(regs, 0x1e600088, 0x00000000); + __ast_moutdwm(regs, 0x1e6e2000, 0x1688A8A8); + + data = __ast_mindwm(regs, 0x1e6e2070); + if (data & 0x08000000) { /* check fast reset */ + /* + * If "Fast restet" is enabled for ARM-ICE debugger, + * then WDT needs to enable, that + * WDT04 is WDT#1 Reload reg. + * WDT08 is WDT#1 counter restart reg to avoid system deadlock + * WDT0C is WDT#1 control reg + * [6:5]:= 01:Full chip + * [4]:= 1:1MHz clock source + * [1]:= 1:WDT will be cleeared and disabled after timeout occurs + * [0]:= 1:WDT enable + */ + __ast_moutdwm(regs, 0x1E785004, 0x00000010); + __ast_moutdwm(regs, 0x1E785008, 0x00004755); + __ast_moutdwm(regs, 0x1E78500c, 0x00000033); + udelay(1000); + } + + do { + __ast_moutdwm(regs, 0x1e6e2000, 0x1688A8A8); + data = __ast_mindwm(regs, 0x1e6e2000); + } while (data != 1); + + __ast_moutdwm(regs, 0x1e6e207c, 0x08000000); /* clear fast reset */ +} + +static bool mmc_test_single_2500(struct ast_device *ast, u32 datagen) +{ + return mmc_test(ast, datagen, 0x85); +} + +static bool cbr_test_2500(struct ast_device *ast) +{ + ast_moutdwm(ast, 0x1E6E0074, 0x0000FFFF); + ast_moutdwm(ast, 0x1E6E007C, 0xFF00FF00); + if (!mmc_test_burst(ast, 0)) + return false; + if (!mmc_test_single_2500(ast, 0)) + return false; + return true; +} + +static bool ddr_test_2500(struct ast_device *ast) +{ + ast_moutdwm(ast, 0x1E6E0074, 0x0000FFFF); + ast_moutdwm(ast, 0x1E6E007C, 0xFF00FF00); + if (!mmc_test_burst(ast, 0)) + return false; + if (!mmc_test_burst(ast, 1)) + return false; + if (!mmc_test_burst(ast, 2)) + return false; + if (!mmc_test_burst(ast, 3)) + return false; + if (!mmc_test_single_2500(ast, 0)) + return false; + return true; +} + +static void ddr_init_common_2500(struct ast_device *ast) +{ + ast_moutdwm(ast, 0x1E6E0034, 0x00020080); + ast_moutdwm(ast, 0x1E6E0008, 0x2003000F); + ast_moutdwm(ast, 0x1E6E0038, 0x00000FFF); + ast_moutdwm(ast, 0x1E6E0040, 0x88448844); + ast_moutdwm(ast, 0x1E6E0044, 0x24422288); + ast_moutdwm(ast, 0x1E6E0048, 0x22222222); + ast_moutdwm(ast, 0x1E6E004C, 0x22222222); + ast_moutdwm(ast, 0x1E6E0050, 0x80000000); + ast_moutdwm(ast, 0x1E6E0208, 0x00000000); + ast_moutdwm(ast, 0x1E6E0218, 0x00000000); + ast_moutdwm(ast, 0x1E6E0220, 0x00000000); + ast_moutdwm(ast, 0x1E6E0228, 0x00000000); + ast_moutdwm(ast, 0x1E6E0230, 0x00000000); + ast_moutdwm(ast, 0x1E6E02A8, 0x00000000); + ast_moutdwm(ast, 0x1E6E02B0, 0x00000000); + ast_moutdwm(ast, 0x1E6E0240, 0x86000000); + ast_moutdwm(ast, 0x1E6E0244, 0x00008600); + ast_moutdwm(ast, 0x1E6E0248, 0x80000000); + ast_moutdwm(ast, 0x1E6E024C, 0x80808080); +} + +static void ddr_phy_init_2500(struct ast_device *ast) +{ + u32 data, pass, timecnt; + + pass = 0; + ast_moutdwm(ast, 0x1E6E0060, 0x00000005); + while (!pass) { + for (timecnt = 0; timecnt < TIMEOUT; timecnt++) { + data = ast_mindwm(ast, 0x1E6E0060) & 0x1; + if (!data) + break; + } + if (timecnt != TIMEOUT) { + data = ast_mindwm(ast, 0x1E6E0300) & 0x000A0000; + if (!data) + pass = 1; + } + if (!pass) { + ast_moutdwm(ast, 0x1E6E0060, 0x00000000); + udelay(10); /* delay 10 us */ + ast_moutdwm(ast, 0x1E6E0060, 0x00000005); + } + } + + ast_moutdwm(ast, 0x1E6E0060, 0x00000006); +} + +/* + * Check DRAM Size + * 1Gb : 0x80000000 ~ 0x87FFFFFF + * 2Gb : 0x80000000 ~ 0x8FFFFFFF + * 4Gb : 0x80000000 ~ 0x9FFFFFFF + * 8Gb : 0x80000000 ~ 0xBFFFFFFF + */ +static void check_dram_size_2500(struct ast_device *ast, u32 tRFC) +{ + u32 reg_04, reg_14; + + reg_04 = ast_mindwm(ast, 0x1E6E0004) & 0xfffffffc; + reg_14 = ast_mindwm(ast, 0x1E6E0014) & 0xffffff00; + + ast_moutdwm(ast, 0xA0100000, 0x41424344); + ast_moutdwm(ast, 0x90100000, 0x35363738); + ast_moutdwm(ast, 0x88100000, 0x292A2B2C); + ast_moutdwm(ast, 0x80100000, 0x1D1E1F10); + + /* Check 8Gbit */ + if (ast_mindwm(ast, 0xA0100000) == 0x41424344) { + reg_04 |= 0x03; + reg_14 |= (tRFC >> 24) & 0xFF; + /* Check 4Gbit */ + } else if (ast_mindwm(ast, 0x90100000) == 0x35363738) { + reg_04 |= 0x02; + reg_14 |= (tRFC >> 16) & 0xFF; + /* Check 2Gbit */ + } else if (ast_mindwm(ast, 0x88100000) == 0x292A2B2C) { + reg_04 |= 0x01; + reg_14 |= (tRFC >> 8) & 0xFF; + } else { + reg_14 |= tRFC & 0xFF; + } + ast_moutdwm(ast, 0x1E6E0004, reg_04); + ast_moutdwm(ast, 0x1E6E0014, reg_14); +} + +static void enable_cache_2500(struct ast_device *ast) +{ + u32 reg_04, data; + + reg_04 = ast_mindwm(ast, 0x1E6E0004); + ast_moutdwm(ast, 0x1E6E0004, reg_04 | 0x1000); + + do + data = ast_mindwm(ast, 0x1E6E0004); + while (!(data & 0x80000)); + ast_moutdwm(ast, 0x1E6E0004, reg_04 | 0x400); +} + +static void set_mpll_2500(struct ast_device *ast) +{ + u32 addr, data, param; + + /* Reset MMC */ + ast_moutdwm(ast, 0x1E6E0000, 0xFC600309); + ast_moutdwm(ast, 0x1E6E0034, 0x00020080); + for (addr = 0x1e6e0004; addr < 0x1e6e0090;) { + ast_moutdwm(ast, addr, 0x0); + addr += 4; + } + ast_moutdwm(ast, 0x1E6E0034, 0x00020000); + + ast_moutdwm(ast, 0x1E6E2000, 0x1688A8A8); + data = ast_mindwm(ast, 0x1E6E2070) & 0x00800000; + if (data) { + /* CLKIN = 25MHz */ + param = 0x930023E0; + ast_moutdwm(ast, 0x1E6E2160, 0x00011320); + } else { + /* CLKIN = 24MHz */ + param = 0x93002400; + } + ast_moutdwm(ast, 0x1E6E2020, param); + udelay(100); +} + +static void reset_mmc_2500(struct ast_device *ast) +{ + ast_moutdwm(ast, 0x1E78505C, 0x00000004); + ast_moutdwm(ast, 0x1E785044, 0x00000001); + ast_moutdwm(ast, 0x1E785048, 0x00004755); + ast_moutdwm(ast, 0x1E78504C, 0x00000013); + mdelay(100); + ast_moutdwm(ast, 0x1E785054, 0x00000077); + ast_moutdwm(ast, 0x1E6E0000, 0xFC600309); +} + +static void ddr3_init_2500(struct ast_device *ast, const u32 *ddr_table) +{ + ast_moutdwm(ast, 0x1E6E0004, 0x00000303); + ast_moutdwm(ast, 0x1E6E0010, ddr_table[REGIDX_010]); + ast_moutdwm(ast, 0x1E6E0014, ddr_table[REGIDX_014]); + ast_moutdwm(ast, 0x1E6E0018, ddr_table[REGIDX_018]); + ast_moutdwm(ast, 0x1E6E0020, ddr_table[REGIDX_020]); /* MODEREG4/6 */ + ast_moutdwm(ast, 0x1E6E0024, ddr_table[REGIDX_024]); /* MODEREG5 */ + ast_moutdwm(ast, 0x1E6E002C, ddr_table[REGIDX_02C] | 0x100); /* MODEREG0/2 */ + ast_moutdwm(ast, 0x1E6E0030, ddr_table[REGIDX_030]); /* MODEREG1/3 */ + + /* DDR PHY Setting */ + ast_moutdwm(ast, 0x1E6E0200, 0x02492AAE); + ast_moutdwm(ast, 0x1E6E0204, 0x00001001); + ast_moutdwm(ast, 0x1E6E020C, 0x55E00B0B); + ast_moutdwm(ast, 0x1E6E0210, 0x20000000); + ast_moutdwm(ast, 0x1E6E0214, ddr_table[REGIDX_214]); + ast_moutdwm(ast, 0x1E6E02E0, ddr_table[REGIDX_2E0]); + ast_moutdwm(ast, 0x1E6E02E4, ddr_table[REGIDX_2E4]); + ast_moutdwm(ast, 0x1E6E02E8, ddr_table[REGIDX_2E8]); + ast_moutdwm(ast, 0x1E6E02EC, ddr_table[REGIDX_2EC]); + ast_moutdwm(ast, 0x1E6E02F0, ddr_table[REGIDX_2F0]); + ast_moutdwm(ast, 0x1E6E02F4, ddr_table[REGIDX_2F4]); + ast_moutdwm(ast, 0x1E6E02F8, ddr_table[REGIDX_2F8]); + ast_moutdwm(ast, 0x1E6E0290, 0x00100008); + ast_moutdwm(ast, 0x1E6E02C0, 0x00000006); + + /* Controller Setting */ + ast_moutdwm(ast, 0x1E6E0034, 0x00020091); + + /* Wait DDR PHY init done */ + ddr_phy_init_2500(ast); + + ast_moutdwm(ast, 0x1E6E0120, ddr_table[REGIDX_PLL]); + ast_moutdwm(ast, 0x1E6E000C, 0x42AA5C81); + ast_moutdwm(ast, 0x1E6E0034, 0x0001AF93); + + check_dram_size_2500(ast, ddr_table[REGIDX_RFC]); + enable_cache_2500(ast); + ast_moutdwm(ast, 0x1E6E001C, 0x00000008); + ast_moutdwm(ast, 0x1E6E0038, 0xFFFFFF00); +} + +static void ddr4_init_2500(struct ast_device *ast, const u32 *ddr_table) +{ + u32 data, data2, pass, retrycnt; + u32 ddr_vref, phy_vref; + u32 min_ddr_vref = 0, min_phy_vref = 0; + u32 max_ddr_vref = 0, max_phy_vref = 0; + + ast_moutdwm(ast, 0x1E6E0004, 0x00000313); + ast_moutdwm(ast, 0x1E6E0010, ddr_table[REGIDX_010]); + ast_moutdwm(ast, 0x1E6E0014, ddr_table[REGIDX_014]); + ast_moutdwm(ast, 0x1E6E0018, ddr_table[REGIDX_018]); + ast_moutdwm(ast, 0x1E6E0020, ddr_table[REGIDX_020]); /* MODEREG4/6 */ + ast_moutdwm(ast, 0x1E6E0024, ddr_table[REGIDX_024]); /* MODEREG5 */ + ast_moutdwm(ast, 0x1E6E002C, ddr_table[REGIDX_02C] | 0x100); /* MODEREG0/2 */ + ast_moutdwm(ast, 0x1E6E0030, ddr_table[REGIDX_030]); /* MODEREG1/3 */ + + /* DDR PHY Setting */ + ast_moutdwm(ast, 0x1E6E0200, 0x42492AAE); + ast_moutdwm(ast, 0x1E6E0204, 0x09002000); + ast_moutdwm(ast, 0x1E6E020C, 0x55E00B0B); + ast_moutdwm(ast, 0x1E6E0210, 0x20000000); + ast_moutdwm(ast, 0x1E6E0214, ddr_table[REGIDX_214]); + ast_moutdwm(ast, 0x1E6E02E0, ddr_table[REGIDX_2E0]); + ast_moutdwm(ast, 0x1E6E02E4, ddr_table[REGIDX_2E4]); + ast_moutdwm(ast, 0x1E6E02E8, ddr_table[REGIDX_2E8]); + ast_moutdwm(ast, 0x1E6E02EC, ddr_table[REGIDX_2EC]); + ast_moutdwm(ast, 0x1E6E02F0, ddr_table[REGIDX_2F0]); + ast_moutdwm(ast, 0x1E6E02F4, ddr_table[REGIDX_2F4]); + ast_moutdwm(ast, 0x1E6E02F8, ddr_table[REGIDX_2F8]); + ast_moutdwm(ast, 0x1E6E0290, 0x00100008); + ast_moutdwm(ast, 0x1E6E02C4, 0x3C183C3C); + ast_moutdwm(ast, 0x1E6E02C8, 0x00631E0E); + + /* Controller Setting */ + ast_moutdwm(ast, 0x1E6E0034, 0x0001A991); + + /* Train PHY Vref first */ + pass = 0; + + for (retrycnt = 0; retrycnt < 4 && pass == 0; retrycnt++) { + max_phy_vref = 0x0; + pass = 0; + ast_moutdwm(ast, 0x1E6E02C0, 0x00001C06); + for (phy_vref = 0x40; phy_vref < 0x80; phy_vref++) { + ast_moutdwm(ast, 0x1E6E000C, 0x00000000); + ast_moutdwm(ast, 0x1E6E0060, 0x00000000); + ast_moutdwm(ast, 0x1E6E02CC, phy_vref | (phy_vref << 8)); + /* Fire DFI Init */ + ddr_phy_init_2500(ast); + ast_moutdwm(ast, 0x1E6E000C, 0x00005C01); + if (cbr_test_2500(ast)) { + pass++; + data = ast_mindwm(ast, 0x1E6E03D0); + data2 = data >> 8; + data = data & 0xff; + if (data > data2) + data = data2; + if (max_phy_vref < data) { + max_phy_vref = data; + min_phy_vref = phy_vref; + } + } else if (pass > 0) { + break; + } + } + } + ast_moutdwm(ast, 0x1E6E02CC, min_phy_vref | (min_phy_vref << 8)); + + /* Train DDR Vref next */ + pass = 0; + + for (retrycnt = 0; retrycnt < 4 && pass == 0; retrycnt++) { + min_ddr_vref = 0xFF; + max_ddr_vref = 0x0; + pass = 0; + for (ddr_vref = 0x00; ddr_vref < 0x40; ddr_vref++) { + ast_moutdwm(ast, 0x1E6E000C, 0x00000000); + ast_moutdwm(ast, 0x1E6E0060, 0x00000000); + ast_moutdwm(ast, 0x1E6E02C0, 0x00000006 | (ddr_vref << 8)); + /* Fire DFI Init */ + ddr_phy_init_2500(ast); + ast_moutdwm(ast, 0x1E6E000C, 0x00005C01); + if (cbr_test_2500(ast)) { + pass++; + if (min_ddr_vref > ddr_vref) + min_ddr_vref = ddr_vref; + if (max_ddr_vref < ddr_vref) + max_ddr_vref = ddr_vref; + } else if (pass != 0) { + break; + } + } + } + + ast_moutdwm(ast, 0x1E6E000C, 0x00000000); + ast_moutdwm(ast, 0x1E6E0060, 0x00000000); + ddr_vref = (min_ddr_vref + max_ddr_vref + 1) >> 1; + ast_moutdwm(ast, 0x1E6E02C0, 0x00000006 | (ddr_vref << 8)); + + /* Wait DDR PHY init done */ + ddr_phy_init_2500(ast); + + ast_moutdwm(ast, 0x1E6E0120, ddr_table[REGIDX_PLL]); + ast_moutdwm(ast, 0x1E6E000C, 0x42AA5C81); + ast_moutdwm(ast, 0x1E6E0034, 0x0001AF93); + + check_dram_size_2500(ast, ddr_table[REGIDX_RFC]); + enable_cache_2500(ast); + ast_moutdwm(ast, 0x1E6E001C, 0x00000008); + ast_moutdwm(ast, 0x1E6E0038, 0xFFFFFF00); +} + +static bool ast_dram_init_2500(struct ast_device *ast) +{ + u32 data; + u32 max_tries = 5; + + do { + if (max_tries-- == 0) + return false; + set_mpll_2500(ast); + reset_mmc_2500(ast); + ddr_init_common_2500(ast); + + data = ast_mindwm(ast, 0x1E6E2070); + if (data & 0x01000000) + ddr4_init_2500(ast, ast2500_ddr4_1600_timing_table); + else + ddr3_init_2500(ast, ast2500_ddr3_1600_timing_table); + } while (!ddr_test_2500(ast)); + + ast_moutdwm(ast, 0x1E6E2040, ast_mindwm(ast, 0x1E6E2040) | 0x41); + + /* Patch code */ + data = ast_mindwm(ast, 0x1E6E200C) & 0xF9FFFFFF; + ast_moutdwm(ast, 0x1E6E200C, data | 0x10000000); + + return true; +} + +static void ast_post_chip_2500(struct ast_device *ast) +{ + struct drm_device *dev = &ast->base; + u32 temp; + u8 reg; + + reg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); + if ((reg & AST_IO_VGACRD0_VRAM_INIT_STATUS_MASK) == 0) {/* vga only */ + /* Clear bus lock condition */ + ast_2500_patch_ahb(ast->regs); + + /* Disable watchdog */ + ast_moutdwm(ast, 0x1E78502C, 0x00000000); + ast_moutdwm(ast, 0x1E78504C, 0x00000000); + + /* + * Reset USB port to patch USB unknown device issue + * SCU90 is Multi-function Pin Control #5 + * [29]:= 1:Enable USB2.0 Host port#1 (that the mutually shared USB2.0 Hub + * port). + * SCU94 is Multi-function Pin Control #6 + * [14:13]:= 1x:USB2.0 Host2 controller + * SCU70 is Hardware Strap reg + * [23]:= 1:CLKIN is 25MHz and USBCK1 = 24/48 MHz (determined by + * [18]: 0(24)/1(48) MHz) + * SCU7C is Write clear reg to SCU70 + * [23]:= write 1 and then SCU70[23] will be clear as 0b. + */ + ast_moutdwm(ast, 0x1E6E2090, 0x20000000); + ast_moutdwm(ast, 0x1E6E2094, 0x00004000); + if (ast_mindwm(ast, 0x1E6E2070) & 0x00800000) { + ast_moutdwm(ast, 0x1E6E207C, 0x00800000); + mdelay(100); + ast_moutdwm(ast, 0x1E6E2070, 0x00800000); + } + /* Modify eSPI reset pin */ + temp = ast_mindwm(ast, 0x1E6E2070); + if (temp & 0x02000000) + ast_moutdwm(ast, 0x1E6E207C, 0x00004000); + + /* Slow down CPU/AHB CLK in VGA only mode */ + temp = ast_read32(ast, 0x12008); + temp |= 0x73; + ast_write32(ast, 0x12008, temp); + + if (!ast_dram_init_2500(ast)) + drm_err(dev, "DRAM init failed !\n"); + + temp = ast_mindwm(ast, 0x1e6e2040); + ast_moutdwm(ast, 0x1e6e2040, temp | 0x40); + } + + /* wait ready */ + do { + reg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); + } while ((reg & 0x40) == 0); +} + +int ast_2500_post(struct ast_device *ast) +{ + ast_2300_set_def_ext_reg(ast); + + if (ast->config_mode == ast_use_p2a) { + ast_post_chip_2500(ast); + } else { + if (ast->tx_chip == AST_TX_SIL164) { + /* Enable DVO */ + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); + } + } + + return 0; +} diff --git a/drivers/gpu/drm/ast/ast_2600.c b/drivers/gpu/drm/ast/ast_2600.c new file mode 100644 index 000000000000..8d75a47444f5 --- /dev/null +++ b/drivers/gpu/drm/ast/ast_2600.c @@ -0,0 +1,44 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright 2012 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the + * "Software"), to deal in the Software without restriction, including + * without limitation the rights to use, copy, modify, merge, publish, + * distribute, sub license, and/or sell copies of the Software, and to + * permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, + * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR + * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE + * USE OR OTHER DEALINGS IN THE SOFTWARE. + * + * The above copyright notice and this permission notice (including the + * next paragraph) shall be included in all copies or substantial portions + * of the Software. + */ +/* + * Authors: Dave Airlie <airlied@redhat.com> + */ + +#include "ast_drv.h" +#include "ast_post.h" + +/* + * POST + */ + +int ast_2600_post(struct ast_device *ast) +{ + ast_2300_set_def_ext_reg(ast); + + if (ast->tx_chip == AST_TX_ASTDP) + return ast_dp_launch(ast); + + return 0; +} diff --git a/drivers/gpu/drm/ast/ast_dram_tables.h b/drivers/gpu/drm/ast/ast_dram_tables.h deleted file mode 100644 index 1e9ac9d6d26c..000000000000 --- a/drivers/gpu/drm/ast/ast_dram_tables.h +++ /dev/null @@ -1,207 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef AST_DRAM_TABLES_H -#define AST_DRAM_TABLES_H - -/* DRAM timing tables */ -struct ast_dramstruct { - u16 index; - u32 data; -}; - -static const struct ast_dramstruct ast2000_dram_table_data[] = { - { 0x0108, 0x00000000 }, - { 0x0120, 0x00004a21 }, - { 0xFF00, 0x00000043 }, - { 0x0000, 0xFFFFFFFF }, - { 0x0004, 0x00000089 }, - { 0x0008, 0x22331353 }, - { 0x000C, 0x0d07000b }, - { 0x0010, 0x11113333 }, - { 0x0020, 0x00110350 }, - { 0x0028, 0x1e0828f0 }, - { 0x0024, 0x00000001 }, - { 0x001C, 0x00000000 }, - { 0x0014, 0x00000003 }, - { 0xFF00, 0x00000043 }, - { 0x0018, 0x00000131 }, - { 0x0014, 0x00000001 }, - { 0xFF00, 0x00000043 }, - { 0x0018, 0x00000031 }, - { 0x0014, 0x00000001 }, - { 0xFF00, 0x00000043 }, - { 0x0028, 0x1e0828f1 }, - { 0x0024, 0x00000003 }, - { 0x002C, 0x1f0f28fb }, - { 0x0030, 0xFFFFFE01 }, - { 0xFFFF, 0xFFFFFFFF } -}; - -static const struct ast_dramstruct ast1100_dram_table_data[] = { - { 0x2000, 0x1688a8a8 }, - { 0x2020, 0x000041f0 }, - { 0xFF00, 0x00000043 }, - { 0x0000, 0xfc600309 }, - { 0x006C, 0x00909090 }, - { 0x0064, 0x00050000 }, - { 0x0004, 0x00000585 }, - { 0x0008, 0x0011030f }, - { 0x0010, 0x22201724 }, - { 0x0018, 0x1e29011a }, - { 0x0020, 0x00c82222 }, - { 0x0014, 0x01001523 }, - { 0x001C, 0x1024010d }, - { 0x0024, 0x00cb2522 }, - { 0x0038, 0xffffff82 }, - { 0x003C, 0x00000000 }, - { 0x0040, 0x00000000 }, - { 0x0044, 0x00000000 }, - { 0x0048, 0x00000000 }, - { 0x004C, 0x00000000 }, - { 0x0050, 0x00000000 }, - { 0x0054, 0x00000000 }, - { 0x0058, 0x00000000 }, - { 0x005C, 0x00000000 }, - { 0x0060, 0x032aa02a }, - { 0x0064, 0x002d3000 }, - { 0x0068, 0x00000000 }, - { 0x0070, 0x00000000 }, - { 0x0074, 0x00000000 }, - { 0x0078, 0x00000000 }, - { 0x007C, 0x00000000 }, - { 0x0034, 0x00000001 }, - { 0xFF00, 0x00000043 }, - { 0x002C, 0x00000732 }, - { 0x0030, 0x00000040 }, - { 0x0028, 0x00000005 }, - { 0x0028, 0x00000007 }, - { 0x0028, 0x00000003 }, - { 0x0028, 0x00000001 }, - { 0x000C, 0x00005a08 }, - { 0x002C, 0x00000632 }, - { 0x0028, 0x00000001 }, - { 0x0030, 0x000003c0 }, - { 0x0028, 0x00000003 }, - { 0x0030, 0x00000040 }, - { 0x0028, 0x00000003 }, - { 0x000C, 0x00005a21 }, - { 0x0034, 0x00007c03 }, - { 0x0120, 0x00004c41 }, - { 0xffff, 0xffffffff }, -}; - -static const struct ast_dramstruct ast2100_dram_table_data[] = { - { 0x2000, 0x1688a8a8 }, - { 0x2020, 0x00004120 }, - { 0xFF00, 0x00000043 }, - { 0x0000, 0xfc600309 }, - { 0x006C, 0x00909090 }, - { 0x0064, 0x00070000 }, - { 0x0004, 0x00000489 }, - { 0x0008, 0x0011030f }, - { 0x0010, 0x32302926 }, - { 0x0018, 0x274c0122 }, - { 0x0020, 0x00ce2222 }, - { 0x0014, 0x01001523 }, - { 0x001C, 0x1024010d }, - { 0x0024, 0x00cb2522 }, - { 0x0038, 0xffffff82 }, - { 0x003C, 0x00000000 }, - { 0x0040, 0x00000000 }, - { 0x0044, 0x00000000 }, - { 0x0048, 0x00000000 }, - { 0x004C, 0x00000000 }, - { 0x0050, 0x00000000 }, - { 0x0054, 0x00000000 }, - { 0x0058, 0x00000000 }, - { 0x005C, 0x00000000 }, - { 0x0060, 0x0f2aa02a }, - { 0x0064, 0x003f3005 }, - { 0x0068, 0x02020202 }, - { 0x0070, 0x00000000 }, - { 0x0074, 0x00000000 }, - { 0x0078, 0x00000000 }, - { 0x007C, 0x00000000 }, - { 0x0034, 0x00000001 }, - { 0xFF00, 0x00000043 }, - { 0x002C, 0x00000942 }, - { 0x0030, 0x00000040 }, - { 0x0028, 0x00000005 }, - { 0x0028, 0x00000007 }, - { 0x0028, 0x00000003 }, - { 0x0028, 0x00000001 }, - { 0x000C, 0x00005a08 }, - { 0x002C, 0x00000842 }, - { 0x0028, 0x00000001 }, - { 0x0030, 0x000003c0 }, - { 0x0028, 0x00000003 }, - { 0x0030, 0x00000040 }, - { 0x0028, 0x00000003 }, - { 0x000C, 0x00005a21 }, - { 0x0034, 0x00007c03 }, - { 0x0120, 0x00005061 }, - { 0xffff, 0xffffffff }, -}; - -/* - * AST2500 DRAM settings modules - */ -#define REGTBL_NUM 17 -#define REGIDX_010 0 -#define REGIDX_014 1 -#define REGIDX_018 2 -#define REGIDX_020 3 -#define REGIDX_024 4 -#define REGIDX_02C 5 -#define REGIDX_030 6 -#define REGIDX_214 7 -#define REGIDX_2E0 8 -#define REGIDX_2E4 9 -#define REGIDX_2E8 10 -#define REGIDX_2EC 11 -#define REGIDX_2F0 12 -#define REGIDX_2F4 13 -#define REGIDX_2F8 14 -#define REGIDX_RFC 15 -#define REGIDX_PLL 16 - -static const u32 ast2500_ddr3_1600_timing_table[REGTBL_NUM] = { - 0x64604D38, /* 0x010 */ - 0x29690599, /* 0x014 */ - 0x00000300, /* 0x018 */ - 0x00000000, /* 0x020 */ - 0x00000000, /* 0x024 */ - 0x02181E70, /* 0x02C */ - 0x00000040, /* 0x030 */ - 0x00000024, /* 0x214 */ - 0x02001300, /* 0x2E0 */ - 0x0E0000A0, /* 0x2E4 */ - 0x000E001B, /* 0x2E8 */ - 0x35B8C105, /* 0x2EC */ - 0x08090408, /* 0x2F0 */ - 0x9B000800, /* 0x2F4 */ - 0x0E400A00, /* 0x2F8 */ - 0x9971452F, /* tRFC */ - 0x000071C1 /* PLL */ -}; - -static const u32 ast2500_ddr4_1600_timing_table[REGTBL_NUM] = { - 0x63604E37, /* 0x010 */ - 0xE97AFA99, /* 0x014 */ - 0x00019000, /* 0x018 */ - 0x08000000, /* 0x020 */ - 0x00000400, /* 0x024 */ - 0x00000410, /* 0x02C */ - 0x00000101, /* 0x030 */ - 0x00000024, /* 0x214 */ - 0x03002900, /* 0x2E0 */ - 0x0E0000A0, /* 0x2E4 */ - 0x000E001C, /* 0x2E8 */ - 0x35B8C106, /* 0x2EC */ - 0x08080607, /* 0x2F0 */ - 0x9B000900, /* 0x2F4 */ - 0x0E400A00, /* 0x2F8 */ - 0x99714545, /* tRFC */ - 0x000071C1 /* PLL */ -}; - -#endif diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c index 054acda41909..473faa92d08c 100644 --- a/drivers/gpu/drm/ast/ast_drv.c +++ b/drivers/gpu/drm/ast/ast_drv.c @@ -64,7 +64,7 @@ static const struct drm_driver ast_driver = { .minor = DRIVER_MINOR, .patchlevel = DRIVER_PATCHLEVEL, - DRM_GEM_SHMEM_DRIVER_OPS_NO_MAP_SGT, + DRM_GEM_SHMEM_DRIVER_OPS, DRM_FBDEV_SHMEM_DRIVER_OPS, }; @@ -171,7 +171,7 @@ static int ast_detect_chip(struct pci_dev *pdev, /* Patch AST2500/AST2510 */ if ((pdev->revision & 0xf0) == 0x40) { if (!(vgacrd0 & AST_IO_VGACRD0_VRAM_INIT_STATUS_MASK)) - ast_patch_ahb_2500(regs); + ast_2500_patch_ahb(regs); } /* Double check that it's actually working */ diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 2ee402096cd9..e37a55295ed7 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -417,11 +417,26 @@ struct ast_crtc_state { int ast_mm_init(struct ast_device *ast); +/* ast_2000.c */ +int ast_2000_post(struct ast_device *ast); + +/* ast_2100.c */ +int ast_2100_post(struct ast_device *ast); + +/* ast_2300.c */ +int ast_2300_post(struct ast_device *ast); + +/* ast_2500.c */ +void ast_2500_patch_ahb(void __iomem *regs); +int ast_2500_post(struct ast_device *ast); + +/* ast_2600.c */ +int ast_2600_post(struct ast_device *ast); + /* ast post */ int ast_post_gpu(struct ast_device *ast); u32 ast_mindwm(struct ast_device *ast, u32 r); void ast_moutdwm(struct ast_device *ast, u32 r, u32 v); -void ast_patch_ahb_2500(void __iomem *regs); int ast_vga_output_init(struct ast_device *ast); int ast_sil164_output_init(struct ast_device *ast); diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c index 37568cf3822c..b72914dbed38 100644 --- a/drivers/gpu/drm/ast/ast_post.c +++ b/drivers/gpu/drm/ast/ast_post.c @@ -31,51 +31,10 @@ #include <drm/drm_print.h> -#include "ast_dram_tables.h" #include "ast_drv.h" +#include "ast_post.h" -static void ast_post_chip_2300(struct ast_device *ast); -static void ast_post_chip_2500(struct ast_device *ast); - -static const u8 extreginfo[] = { 0x0f, 0x04, 0x1c, 0xff }; -static const u8 extreginfo_ast2300[] = { 0x0f, 0x04, 0x1f, 0xff }; - -static void ast_set_def_ext_reg(struct ast_device *ast) -{ - u8 i, index, reg; - const u8 *ext_reg_info; - - /* reset scratch */ - for (i = 0x81; i <= 0x9f; i++) - ast_set_index_reg(ast, AST_IO_VGACRI, i, 0x00); - - if (IS_AST_GEN4(ast) || IS_AST_GEN5(ast) || IS_AST_GEN6(ast)) - ext_reg_info = extreginfo_ast2300; - else - ext_reg_info = extreginfo; - - index = 0xa0; - while (*ext_reg_info != 0xff) { - ast_set_index_reg_mask(ast, AST_IO_VGACRI, index, 0x00, *ext_reg_info); - index++; - ext_reg_info++; - } - - /* disable standard IO/MEM decode if secondary */ - /* ast_set_index_reg-mask(ast, AST_IO_VGACRI, 0xa1, 0xff, 0x3); */ - - /* Set Ext. Default */ - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0x8c, 0x00, 0x01); - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xb7, 0x00, 0x00); - - /* Enable RAMDAC for A1 */ - reg = 0x04; - if (IS_AST_GEN4(ast) || IS_AST_GEN5(ast) || IS_AST_GEN6(ast)) - reg |= 0x20; - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xb6, 0xff, reg); -} - -static u32 __ast_mindwm(void __iomem *regs, u32 r) +u32 __ast_mindwm(void __iomem *regs, u32 r) { u32 data; @@ -89,7 +48,7 @@ static u32 __ast_mindwm(void __iomem *regs, u32 r) return __ast_read32(regs, 0x10000 + (r & 0x0000ffff)); } -static void __ast_moutdwm(void __iomem *regs, u32 r, u32 v) +void __ast_moutdwm(void __iomem *regs, u32 r, u32 v) { u32 data; @@ -113,332 +72,38 @@ void ast_moutdwm(struct ast_device *ast, u32 r, u32 v) __ast_moutdwm(ast->regs, r, v); } -/* - * AST2100/2150 DLL CBR Setting - */ -#define CBR_SIZE_AST2150 ((16 << 10) - 1) -#define CBR_PASSNUM_AST2150 5 -#define CBR_THRESHOLD_AST2150 10 -#define CBR_THRESHOLD2_AST2150 10 -#define TIMEOUT_AST2150 5000000 - -#define CBR_PATNUM_AST2150 8 - -static const u32 pattern_AST2150[14] = { - 0xFF00FF00, - 0xCC33CC33, - 0xAA55AA55, - 0xFFFE0001, - 0x683501FE, - 0x0F1929B0, - 0x2D0B4346, - 0x60767F02, - 0x6FBE36A6, - 0x3A253035, - 0x3019686D, - 0x41C6167E, - 0x620152BF, - 0x20F050E0 -}; - -static u32 mmctestburst2_ast2150(struct ast_device *ast, u32 datagen) -{ - u32 data, timeout; - - ast_moutdwm(ast, 0x1e6e0070, 0x00000000); - ast_moutdwm(ast, 0x1e6e0070, 0x00000001 | (datagen << 3)); - timeout = 0; - do { - data = ast_mindwm(ast, 0x1e6e0070) & 0x40; - if (++timeout > TIMEOUT_AST2150) { - ast_moutdwm(ast, 0x1e6e0070, 0x00000000); - return 0xffffffff; - } - } while (!data); - ast_moutdwm(ast, 0x1e6e0070, 0x00000000); - ast_moutdwm(ast, 0x1e6e0070, 0x00000003 | (datagen << 3)); - timeout = 0; - do { - data = ast_mindwm(ast, 0x1e6e0070) & 0x40; - if (++timeout > TIMEOUT_AST2150) { - ast_moutdwm(ast, 0x1e6e0070, 0x00000000); - return 0xffffffff; - } - } while (!data); - data = (ast_mindwm(ast, 0x1e6e0070) & 0x80) >> 7; - ast_moutdwm(ast, 0x1e6e0070, 0x00000000); - return data; -} - -#if 0 /* unused in DDX driver - here for completeness */ -static u32 mmctestsingle2_ast2150(struct ast_device *ast, u32 datagen) -{ - u32 data, timeout; - - ast_moutdwm(ast, 0x1e6e0070, 0x00000000); - ast_moutdwm(ast, 0x1e6e0070, 0x00000005 | (datagen << 3)); - timeout = 0; - do { - data = ast_mindwm(ast, 0x1e6e0070) & 0x40; - if (++timeout > TIMEOUT_AST2150) { - ast_moutdwm(ast, 0x1e6e0070, 0x00000000); - return 0xffffffff; - } - } while (!data); - data = (ast_mindwm(ast, 0x1e6e0070) & 0x80) >> 7; - ast_moutdwm(ast, 0x1e6e0070, 0x00000000); - return data; -} -#endif - -static int cbrtest_ast2150(struct ast_device *ast) -{ - int i; - - for (i = 0; i < 8; i++) - if (mmctestburst2_ast2150(ast, i)) - return 0; - return 1; -} - -static int cbrscan_ast2150(struct ast_device *ast, int busw) -{ - u32 patcnt, loop; - - for (patcnt = 0; patcnt < CBR_PATNUM_AST2150; patcnt++) { - ast_moutdwm(ast, 0x1e6e007c, pattern_AST2150[patcnt]); - for (loop = 0; loop < CBR_PASSNUM_AST2150; loop++) { - if (cbrtest_ast2150(ast)) - break; - } - if (loop == CBR_PASSNUM_AST2150) - return 0; - } - return 1; -} - - -static void cbrdlli_ast2150(struct ast_device *ast, int busw) -{ - u32 dll_min[4], dll_max[4], dlli, data, passcnt; - -cbr_start: - dll_min[0] = dll_min[1] = dll_min[2] = dll_min[3] = 0xff; - dll_max[0] = dll_max[1] = dll_max[2] = dll_max[3] = 0x0; - passcnt = 0; - - for (dlli = 0; dlli < 100; dlli++) { - ast_moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24)); - data = cbrscan_ast2150(ast, busw); - if (data != 0) { - if (data & 0x1) { - if (dll_min[0] > dlli) - dll_min[0] = dlli; - if (dll_max[0] < dlli) - dll_max[0] = dlli; - } - passcnt++; - } else if (passcnt >= CBR_THRESHOLD_AST2150) - goto cbr_start; - } - if (dll_max[0] == 0 || (dll_max[0]-dll_min[0]) < CBR_THRESHOLD_AST2150) - goto cbr_start; - - dlli = dll_min[0] + (((dll_max[0] - dll_min[0]) * 7) >> 4); - ast_moutdwm(ast, 0x1e6e0068, dlli | (dlli << 8) | (dlli << 16) | (dlli << 24)); -} - - - -static void ast_init_dram_reg(struct ast_device *ast) -{ - u8 j; - u32 data, temp, i; - const struct ast_dramstruct *dram_reg_info; - - j = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); - - if ((j & 0x80) == 0) { /* VGA only */ - if (IS_AST_GEN1(ast)) { - dram_reg_info = ast2000_dram_table_data; - ast_write32(ast, 0xf004, 0x1e6e0000); - ast_write32(ast, 0xf000, 0x1); - ast_write32(ast, 0x10100, 0xa8); - - do { - ; - } while (ast_read32(ast, 0x10100) != 0xa8); - } else { /* GEN2/GEN3 */ - if (ast->chip == AST2100 || ast->chip == AST2200) - dram_reg_info = ast2100_dram_table_data; - else - dram_reg_info = ast1100_dram_table_data; - - ast_write32(ast, 0xf004, 0x1e6e0000); - ast_write32(ast, 0xf000, 0x1); - ast_write32(ast, 0x12000, 0x1688A8A8); - do { - ; - } while (ast_read32(ast, 0x12000) != 0x01); - - ast_write32(ast, 0x10000, 0xfc600309); - do { - ; - } while (ast_read32(ast, 0x10000) != 0x01); - } - - while (dram_reg_info->index != 0xffff) { - if (dram_reg_info->index == 0xff00) {/* delay fn */ - for (i = 0; i < 15; i++) - udelay(dram_reg_info->data); - } else if (dram_reg_info->index == 0x4 && !IS_AST_GEN1(ast)) { - data = dram_reg_info->data; - if (ast->dram_type == AST_DRAM_1Gx16) - data = 0x00000d89; - else if (ast->dram_type == AST_DRAM_1Gx32) - data = 0x00000c8d; - - temp = ast_read32(ast, 0x12070); - temp &= 0xc; - temp <<= 2; - ast_write32(ast, 0x10000 + dram_reg_info->index, data | temp); - } else - ast_write32(ast, 0x10000 + dram_reg_info->index, dram_reg_info->data); - dram_reg_info++; - } - - /* AST 2100/2150 DRAM calibration */ - data = ast_read32(ast, 0x10120); - if (data == 0x5061) { /* 266Mhz */ - data = ast_read32(ast, 0x10004); - if (data & 0x40) - cbrdlli_ast2150(ast, 16); /* 16 bits */ - else - cbrdlli_ast2150(ast, 32); /* 32 bits */ - } - - switch (AST_GEN(ast)) { - case 1: - temp = ast_read32(ast, 0x10140); - ast_write32(ast, 0x10140, temp | 0x40); - break; - case 2: - case 3: - temp = ast_read32(ast, 0x1200c); - ast_write32(ast, 0x1200c, temp & 0xfffffffd); - temp = ast_read32(ast, 0x12040); - ast_write32(ast, 0x12040, temp | 0x40); - break; - default: - break; - } - } - - /* wait ready */ - do { - j = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); - } while ((j & 0x40) == 0); -} - int ast_post_gpu(struct ast_device *ast) { int ret; - ast_set_def_ext_reg(ast); - if (AST_GEN(ast) >= 7) { - if (ast->tx_chip == AST_TX_ASTDP) { - ret = ast_dp_launch(ast); - if (ret) - return ret; - } + ret = ast_2600_post(ast); + if (ret) + return ret; } else if (AST_GEN(ast) >= 6) { - if (ast->config_mode == ast_use_p2a) { - ast_post_chip_2500(ast); - } else { - if (ast->tx_chip == AST_TX_SIL164) { - /* Enable DVO */ - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); - } - } + ret = ast_2500_post(ast); + if (ret) + return ret; } else if (AST_GEN(ast) >= 4) { - if (ast->config_mode == ast_use_p2a) { - ast_post_chip_2300(ast); - ast_init_3rdtx(ast); - } else { - if (ast->tx_chip == AST_TX_SIL164) { - /* Enable DVO */ - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); - } - } + ret = ast_2300_post(ast); + if (ret) + return ret; + } else if (AST_GEN(ast) >= 2) { + ret = ast_2100_post(ast); + if (ret) + return ret; } else { - if (ast->config_mode == ast_use_p2a) { - ast_init_dram_reg(ast); - } else { - if (ast->tx_chip == AST_TX_SIL164) { - /* Enable DVO */ - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); - } - } + ret = ast_2000_post(ast); + if (ret) + return ret; } return 0; } -/* AST 2300 DRAM settings */ -#define AST_DDR3 0 -#define AST_DDR2 1 - -struct ast2300_dram_param { - u32 dram_type; - u32 dram_chipid; - u32 dram_freq; - u32 vram_size; - u32 odt; - u32 wodt; - u32 rodt; - u32 dram_config; - u32 reg_PERIOD; - u32 reg_MADJ; - u32 reg_SADJ; - u32 reg_MRS; - u32 reg_EMRS; - u32 reg_AC1; - u32 reg_AC2; - u32 reg_DQSIC; - u32 reg_DRV; - u32 reg_IOZ; - u32 reg_DQIDLY; - u32 reg_FREQ; - u32 madj_max; - u32 dll2_finetune_step; -}; - -/* - * DQSI DLL CBR Setting - */ -#define CBR_SIZE0 ((1 << 10) - 1) -#define CBR_SIZE1 ((4 << 10) - 1) -#define CBR_SIZE2 ((64 << 10) - 1) -#define CBR_PASSNUM 5 -#define CBR_PASSNUM2 5 -#define CBR_THRESHOLD 10 -#define CBR_THRESHOLD2 10 #define TIMEOUT 5000000 -#define CBR_PATNUM 8 -static const u32 pattern[8] = { - 0xFF00FF00, - 0xCC33CC33, - 0xAA55AA55, - 0x88778877, - 0x92CC4D6E, - 0x543D3CDE, - 0xF1E843C7, - 0x7C61D253 -}; - -static bool mmc_test(struct ast_device *ast, u32 datagen, u8 test_ctl) +bool mmc_test(struct ast_device *ast, u32 datagen, u8 test_ctl) { u32 data, timeout; @@ -458,1657 +123,7 @@ static bool mmc_test(struct ast_device *ast, u32 datagen, u8 test_ctl) return true; } -static u32 mmc_test2(struct ast_device *ast, u32 datagen, u8 test_ctl) -{ - u32 data, timeout; - - ast_moutdwm(ast, 0x1e6e0070, 0x00000000); - ast_moutdwm(ast, 0x1e6e0070, (datagen << 3) | test_ctl); - timeout = 0; - do { - data = ast_mindwm(ast, 0x1e6e0070) & 0x1000; - if (++timeout > TIMEOUT) { - ast_moutdwm(ast, 0x1e6e0070, 0x0); - return 0xffffffff; - } - } while (!data); - data = ast_mindwm(ast, 0x1e6e0078); - data = (data | (data >> 16)) & 0xffff; - ast_moutdwm(ast, 0x1e6e0070, 0x00000000); - return data; -} - - -static bool mmc_test_burst(struct ast_device *ast, u32 datagen) +bool mmc_test_burst(struct ast_device *ast, u32 datagen) { return mmc_test(ast, datagen, 0xc1); } - -static u32 mmc_test_burst2(struct ast_device *ast, u32 datagen) -{ - return mmc_test2(ast, datagen, 0x41); -} - -static bool mmc_test_single(struct ast_device *ast, u32 datagen) -{ - return mmc_test(ast, datagen, 0xc5); -} - -static u32 mmc_test_single2(struct ast_device *ast, u32 datagen) -{ - return mmc_test2(ast, datagen, 0x05); -} - -static bool mmc_test_single_2500(struct ast_device *ast, u32 datagen) -{ - return mmc_test(ast, datagen, 0x85); -} - -static int cbr_test(struct ast_device *ast) -{ - u32 data; - int i; - data = mmc_test_single2(ast, 0); - if ((data & 0xff) && (data & 0xff00)) - return 0; - for (i = 0; i < 8; i++) { - data = mmc_test_burst2(ast, i); - if ((data & 0xff) && (data & 0xff00)) - return 0; - } - if (!data) - return 3; - else if (data & 0xff) - return 2; - return 1; -} - -static int cbr_scan(struct ast_device *ast) -{ - u32 data, data2, patcnt, loop; - - data2 = 3; - for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) { - ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]); - for (loop = 0; loop < CBR_PASSNUM2; loop++) { - if ((data = cbr_test(ast)) != 0) { - data2 &= data; - if (!data2) - return 0; - break; - } - } - if (loop == CBR_PASSNUM2) - return 0; - } - return data2; -} - -static u32 cbr_test2(struct ast_device *ast) -{ - u32 data; - - data = mmc_test_burst2(ast, 0); - if (data == 0xffff) - return 0; - data |= mmc_test_single2(ast, 0); - if (data == 0xffff) - return 0; - - return ~data & 0xffff; -} - -static u32 cbr_scan2(struct ast_device *ast) -{ - u32 data, data2, patcnt, loop; - - data2 = 0xffff; - for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) { - ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]); - for (loop = 0; loop < CBR_PASSNUM2; loop++) { - if ((data = cbr_test2(ast)) != 0) { - data2 &= data; - if (!data2) - return 0; - break; - } - } - if (loop == CBR_PASSNUM2) - return 0; - } - return data2; -} - -static bool cbr_test3(struct ast_device *ast) -{ - if (!mmc_test_burst(ast, 0)) - return false; - if (!mmc_test_single(ast, 0)) - return false; - return true; -} - -static bool cbr_scan3(struct ast_device *ast) -{ - u32 patcnt, loop; - - for (patcnt = 0; patcnt < CBR_PATNUM; patcnt++) { - ast_moutdwm(ast, 0x1e6e007c, pattern[patcnt]); - for (loop = 0; loop < 2; loop++) { - if (cbr_test3(ast)) - break; - } - if (loop == 2) - return false; - } - return true; -} - -static bool finetuneDQI_L(struct ast_device *ast, struct ast2300_dram_param *param) -{ - u32 gold_sadj[2], dllmin[16], dllmax[16], dlli, data, cnt, mask, passcnt, retry = 0; - bool status = false; -FINETUNE_START: - for (cnt = 0; cnt < 16; cnt++) { - dllmin[cnt] = 0xff; - dllmax[cnt] = 0x0; - } - passcnt = 0; - for (dlli = 0; dlli < 76; dlli++) { - ast_moutdwm(ast, 0x1E6E0068, 0x00001400 | (dlli << 16) | (dlli << 24)); - ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE1); - data = cbr_scan2(ast); - if (data != 0) { - mask = 0x00010001; - for (cnt = 0; cnt < 16; cnt++) { - if (data & mask) { - if (dllmin[cnt] > dlli) { - dllmin[cnt] = dlli; - } - if (dllmax[cnt] < dlli) { - dllmax[cnt] = dlli; - } - } - mask <<= 1; - } - passcnt++; - } else if (passcnt >= CBR_THRESHOLD2) { - break; - } - } - gold_sadj[0] = 0x0; - passcnt = 0; - for (cnt = 0; cnt < 16; cnt++) { - if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) { - gold_sadj[0] += dllmin[cnt]; - passcnt++; - } - } - if (retry++ > 10) - goto FINETUNE_DONE; - if (passcnt != 16) { - goto FINETUNE_START; - } - status = true; -FINETUNE_DONE: - gold_sadj[0] = gold_sadj[0] >> 4; - gold_sadj[1] = gold_sadj[0]; - - data = 0; - for (cnt = 0; cnt < 8; cnt++) { - data >>= 3; - if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) { - dlli = dllmin[cnt]; - if (gold_sadj[0] >= dlli) { - dlli = ((gold_sadj[0] - dlli) * 19) >> 5; - if (dlli > 3) { - dlli = 3; - } - } else { - dlli = ((dlli - gold_sadj[0]) * 19) >> 5; - if (dlli > 4) { - dlli = 4; - } - dlli = (8 - dlli) & 0x7; - } - data |= dlli << 21; - } - } - ast_moutdwm(ast, 0x1E6E0080, data); - - data = 0; - for (cnt = 8; cnt < 16; cnt++) { - data >>= 3; - if ((dllmax[cnt] > dllmin[cnt]) && ((dllmax[cnt] - dllmin[cnt]) >= CBR_THRESHOLD2)) { - dlli = dllmin[cnt]; - if (gold_sadj[1] >= dlli) { - dlli = ((gold_sadj[1] - dlli) * 19) >> 5; - if (dlli > 3) { - dlli = 3; - } else { - dlli = (dlli - 1) & 0x7; - } - } else { - dlli = ((dlli - gold_sadj[1]) * 19) >> 5; - dlli += 1; - if (dlli > 4) { - dlli = 4; - } - dlli = (8 - dlli) & 0x7; - } - data |= dlli << 21; - } - } - ast_moutdwm(ast, 0x1E6E0084, data); - return status; -} /* finetuneDQI_L */ - -static void finetuneDQSI(struct ast_device *ast) -{ - u32 dlli, dqsip, dqidly; - u32 reg_mcr18, reg_mcr0c, passcnt[2], diff; - u32 g_dqidly, g_dqsip, g_margin, g_side; - u16 pass[32][2][2]; - char tag[2][76]; - - /* Disable DQI CBR */ - reg_mcr0c = ast_mindwm(ast, 0x1E6E000C); - reg_mcr18 = ast_mindwm(ast, 0x1E6E0018); - reg_mcr18 &= 0x0000ffff; - ast_moutdwm(ast, 0x1E6E0018, reg_mcr18); - - for (dlli = 0; dlli < 76; dlli++) { - tag[0][dlli] = 0x0; - tag[1][dlli] = 0x0; - } - for (dqidly = 0; dqidly < 32; dqidly++) { - pass[dqidly][0][0] = 0xff; - pass[dqidly][0][1] = 0x0; - pass[dqidly][1][0] = 0xff; - pass[dqidly][1][1] = 0x0; - } - for (dqidly = 0; dqidly < 32; dqidly++) { - passcnt[0] = passcnt[1] = 0; - for (dqsip = 0; dqsip < 2; dqsip++) { - ast_moutdwm(ast, 0x1E6E000C, 0); - ast_moutdwm(ast, 0x1E6E0018, reg_mcr18 | (dqidly << 16) | (dqsip << 23)); - ast_moutdwm(ast, 0x1E6E000C, reg_mcr0c); - for (dlli = 0; dlli < 76; dlli++) { - ast_moutdwm(ast, 0x1E6E0068, 0x00001300 | (dlli << 16) | (dlli << 24)); - ast_moutdwm(ast, 0x1E6E0070, 0); - ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE0); - if (cbr_scan3(ast)) { - if (dlli == 0) - break; - passcnt[dqsip]++; - tag[dqsip][dlli] = 'P'; - if (dlli < pass[dqidly][dqsip][0]) - pass[dqidly][dqsip][0] = (u16) dlli; - if (dlli > pass[dqidly][dqsip][1]) - pass[dqidly][dqsip][1] = (u16) dlli; - } else if (passcnt[dqsip] >= 5) - break; - else { - pass[dqidly][dqsip][0] = 0xff; - pass[dqidly][dqsip][1] = 0x0; - } - } - } - if (passcnt[0] == 0 && passcnt[1] == 0) - dqidly++; - } - /* Search margin */ - g_dqidly = g_dqsip = g_margin = g_side = 0; - - for (dqidly = 0; dqidly < 32; dqidly++) { - for (dqsip = 0; dqsip < 2; dqsip++) { - if (pass[dqidly][dqsip][0] > pass[dqidly][dqsip][1]) - continue; - diff = pass[dqidly][dqsip][1] - pass[dqidly][dqsip][0]; - if ((diff+2) < g_margin) - continue; - passcnt[0] = passcnt[1] = 0; - for (dlli = pass[dqidly][dqsip][0]; dlli > 0 && tag[dqsip][dlli] != 0; dlli--, passcnt[0]++); - for (dlli = pass[dqidly][dqsip][1]; dlli < 76 && tag[dqsip][dlli] != 0; dlli++, passcnt[1]++); - if (passcnt[0] > passcnt[1]) - passcnt[0] = passcnt[1]; - passcnt[1] = 0; - if (passcnt[0] > g_side) - passcnt[1] = passcnt[0] - g_side; - if (diff > (g_margin+1) && (passcnt[1] > 0 || passcnt[0] > 8)) { - g_margin = diff; - g_dqidly = dqidly; - g_dqsip = dqsip; - g_side = passcnt[0]; - } else if (passcnt[1] > 1 && g_side < 8) { - if (diff > g_margin) - g_margin = diff; - g_dqidly = dqidly; - g_dqsip = dqsip; - g_side = passcnt[0]; - } - } - } - reg_mcr18 = reg_mcr18 | (g_dqidly << 16) | (g_dqsip << 23); - ast_moutdwm(ast, 0x1E6E0018, reg_mcr18); - -} -static bool cbr_dll2(struct ast_device *ast, struct ast2300_dram_param *param) -{ - u32 dllmin[2], dllmax[2], dlli, data, passcnt, retry = 0; - bool status = false; - - finetuneDQSI(ast); - if (finetuneDQI_L(ast, param) == false) - return status; - -CBR_START2: - dllmin[0] = dllmin[1] = 0xff; - dllmax[0] = dllmax[1] = 0x0; - passcnt = 0; - for (dlli = 0; dlli < 76; dlli++) { - ast_moutdwm(ast, 0x1E6E0068, 0x00001300 | (dlli << 16) | (dlli << 24)); - ast_moutdwm(ast, 0x1E6E0074, CBR_SIZE2); - data = cbr_scan(ast); - if (data != 0) { - if (data & 0x1) { - if (dllmin[0] > dlli) { - dllmin[0] = dlli; - } - if (dllmax[0] < dlli) { - dllmax[0] = dlli; - } - } - if (data & 0x2) { - if (dllmin[1] > dlli) { - dllmin[1] = dlli; - } - if (dllmax[1] < dlli) { - dllmax[1] = dlli; - } - } - passcnt++; - } else if (passcnt >= CBR_THRESHOLD) { - break; - } - } - if (retry++ > 10) - goto CBR_DONE2; - if (dllmax[0] == 0 || (dllmax[0]-dllmin[0]) < CBR_THRESHOLD) { - goto CBR_START2; - } - if (dllmax[1] == 0 || (dllmax[1]-dllmin[1]) < CBR_THRESHOLD) { - goto CBR_START2; - } - status = true; -CBR_DONE2: - dlli = (dllmin[1] + dllmax[1]) >> 1; - dlli <<= 8; - dlli += (dllmin[0] + dllmax[0]) >> 1; - ast_moutdwm(ast, 0x1E6E0068, ast_mindwm(ast, 0x1E720058) | (dlli << 16)); - return status; -} /* CBRDLL2 */ - -static void get_ddr3_info(struct ast_device *ast, struct ast2300_dram_param *param) -{ - u32 trap, trap_AC2, trap_MRS; - - ast_moutdwm(ast, 0x1E6E2000, 0x1688A8A8); - - /* Ger trap info */ - trap = (ast_mindwm(ast, 0x1E6E2070) >> 25) & 0x3; - trap_AC2 = 0x00020000 + (trap << 16); - trap_AC2 |= 0x00300000 + ((trap & 0x2) << 19); - trap_MRS = 0x00000010 + (trap << 4); - trap_MRS |= ((trap & 0x2) << 18); - - param->reg_MADJ = 0x00034C4C; - param->reg_SADJ = 0x00001800; - param->reg_DRV = 0x000000F0; - param->reg_PERIOD = param->dram_freq; - param->rodt = 0; - - switch (param->dram_freq) { - case 336: - ast_moutdwm(ast, 0x1E6E2020, 0x0190); - param->wodt = 0; - param->reg_AC1 = 0x22202725; - param->reg_AC2 = 0xAA007613 | trap_AC2; - param->reg_DQSIC = 0x000000BA; - param->reg_MRS = 0x04001400 | trap_MRS; - param->reg_EMRS = 0x00000000; - param->reg_IOZ = 0x00000023; - param->reg_DQIDLY = 0x00000074; - param->reg_FREQ = 0x00004DC0; - param->madj_max = 96; - param->dll2_finetune_step = 3; - switch (param->dram_chipid) { - default: - case AST_DRAM_512Mx16: - case AST_DRAM_1Gx16: - param->reg_AC2 = 0xAA007613 | trap_AC2; - break; - case AST_DRAM_2Gx16: - param->reg_AC2 = 0xAA00761C | trap_AC2; - break; - case AST_DRAM_4Gx16: - param->reg_AC2 = 0xAA007636 | trap_AC2; - break; - } - break; - default: - case 396: - ast_moutdwm(ast, 0x1E6E2020, 0x03F1); - param->wodt = 1; - param->reg_AC1 = 0x33302825; - param->reg_AC2 = 0xCC009617 | trap_AC2; - param->reg_DQSIC = 0x000000E2; - param->reg_MRS = 0x04001600 | trap_MRS; - param->reg_EMRS = 0x00000000; - param->reg_IOZ = 0x00000034; - param->reg_DRV = 0x000000FA; - param->reg_DQIDLY = 0x00000089; - param->reg_FREQ = 0x00005040; - param->madj_max = 96; - param->dll2_finetune_step = 4; - - switch (param->dram_chipid) { - default: - case AST_DRAM_512Mx16: - case AST_DRAM_1Gx16: - param->reg_AC2 = 0xCC009617 | trap_AC2; - break; - case AST_DRAM_2Gx16: - param->reg_AC2 = 0xCC009622 | trap_AC2; - break; - case AST_DRAM_4Gx16: - param->reg_AC2 = 0xCC00963F | trap_AC2; - break; - } - break; - - case 408: - ast_moutdwm(ast, 0x1E6E2020, 0x01F0); - param->wodt = 1; - param->reg_AC1 = 0x33302825; - param->reg_AC2 = 0xCC009617 | trap_AC2; - param->reg_DQSIC = 0x000000E2; - param->reg_MRS = 0x04001600 | trap_MRS; - param->reg_EMRS = 0x00000000; - param->reg_IOZ = 0x00000023; - param->reg_DRV = 0x000000FA; - param->reg_DQIDLY = 0x00000089; - param->reg_FREQ = 0x000050C0; - param->madj_max = 96; - param->dll2_finetune_step = 4; - - switch (param->dram_chipid) { - default: - case AST_DRAM_512Mx16: - case AST_DRAM_1Gx16: - param->reg_AC2 = 0xCC009617 | trap_AC2; - break; - case AST_DRAM_2Gx16: - param->reg_AC2 = 0xCC009622 | trap_AC2; - break; - case AST_DRAM_4Gx16: - param->reg_AC2 = 0xCC00963F | trap_AC2; - break; - } - - break; - case 456: - ast_moutdwm(ast, 0x1E6E2020, 0x0230); - param->wodt = 0; - param->reg_AC1 = 0x33302926; - param->reg_AC2 = 0xCD44961A; - param->reg_DQSIC = 0x000000FC; - param->reg_MRS = 0x00081830; - param->reg_EMRS = 0x00000000; - param->reg_IOZ = 0x00000045; - param->reg_DQIDLY = 0x00000097; - param->reg_FREQ = 0x000052C0; - param->madj_max = 88; - param->dll2_finetune_step = 4; - break; - case 504: - ast_moutdwm(ast, 0x1E6E2020, 0x0270); - param->wodt = 1; - param->reg_AC1 = 0x33302926; - param->reg_AC2 = 0xDE44A61D; - param->reg_DQSIC = 0x00000117; - param->reg_MRS = 0x00081A30; - param->reg_EMRS = 0x00000000; - param->reg_IOZ = 0x070000BB; - param->reg_DQIDLY = 0x000000A0; - param->reg_FREQ = 0x000054C0; - param->madj_max = 79; - param->dll2_finetune_step = 4; - break; - case 528: - ast_moutdwm(ast, 0x1E6E2020, 0x0290); - param->wodt = 1; - param->rodt = 1; - param->reg_AC1 = 0x33302926; - param->reg_AC2 = 0xEF44B61E; - param->reg_DQSIC = 0x00000125; - param->reg_MRS = 0x00081A30; - param->reg_EMRS = 0x00000040; - param->reg_DRV = 0x000000F5; - param->reg_IOZ = 0x00000023; - param->reg_DQIDLY = 0x00000088; - param->reg_FREQ = 0x000055C0; - param->madj_max = 76; - param->dll2_finetune_step = 3; - break; - case 576: - ast_moutdwm(ast, 0x1E6E2020, 0x0140); - param->reg_MADJ = 0x00136868; - param->reg_SADJ = 0x00004534; - param->wodt = 1; - param->rodt = 1; - param->reg_AC1 = 0x33302A37; - param->reg_AC2 = 0xEF56B61E; - param->reg_DQSIC = 0x0000013F; - param->reg_MRS = 0x00101A50; - param->reg_EMRS = 0x00000040; - param->reg_DRV = 0x000000FA; - param->reg_IOZ = 0x00000023; - param->reg_DQIDLY = 0x00000078; - param->reg_FREQ = 0x000057C0; - param->madj_max = 136; - param->dll2_finetune_step = 3; - break; - case 600: - ast_moutdwm(ast, 0x1E6E2020, 0x02E1); - param->reg_MADJ = 0x00136868; - param->reg_SADJ = 0x00004534; - param->wodt = 1; - param->rodt = 1; - param->reg_AC1 = 0x32302A37; - param->reg_AC2 = 0xDF56B61F; - param->reg_DQSIC = 0x0000014D; - param->reg_MRS = 0x00101A50; - param->reg_EMRS = 0x00000004; - param->reg_DRV = 0x000000F5; - param->reg_IOZ = 0x00000023; - param->reg_DQIDLY = 0x00000078; - param->reg_FREQ = 0x000058C0; - param->madj_max = 132; - param->dll2_finetune_step = 3; - break; - case 624: - ast_moutdwm(ast, 0x1E6E2020, 0x0160); - param->reg_MADJ = 0x00136868; - param->reg_SADJ = 0x00004534; - param->wodt = 1; - param->rodt = 1; - param->reg_AC1 = 0x32302A37; - param->reg_AC2 = 0xEF56B621; - param->reg_DQSIC = 0x0000015A; - param->reg_MRS = 0x02101A50; - param->reg_EMRS = 0x00000004; - param->reg_DRV = 0x000000F5; - param->reg_IOZ = 0x00000034; - param->reg_DQIDLY = 0x00000078; - param->reg_FREQ = 0x000059C0; - param->madj_max = 128; - param->dll2_finetune_step = 3; - break; - } /* switch freq */ - - switch (param->dram_chipid) { - case AST_DRAM_512Mx16: - param->dram_config = 0x130; - break; - default: - case AST_DRAM_1Gx16: - param->dram_config = 0x131; - break; - case AST_DRAM_2Gx16: - param->dram_config = 0x132; - break; - case AST_DRAM_4Gx16: - param->dram_config = 0x133; - break; - } /* switch size */ - - switch (param->vram_size) { - default: - case SZ_8M: - param->dram_config |= 0x00; - break; - case SZ_16M: - param->dram_config |= 0x04; - break; - case SZ_32M: - param->dram_config |= 0x08; - break; - case SZ_64M: - param->dram_config |= 0x0c; - break; - } - -} - -static void ddr3_init(struct ast_device *ast, struct ast2300_dram_param *param) -{ - u32 data, data2, retry = 0; - -ddr3_init_start: - ast_moutdwm(ast, 0x1E6E0000, 0xFC600309); - ast_moutdwm(ast, 0x1E6E0018, 0x00000100); - ast_moutdwm(ast, 0x1E6E0024, 0x00000000); - ast_moutdwm(ast, 0x1E6E0034, 0x00000000); - udelay(10); - ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ); - ast_moutdwm(ast, 0x1E6E0068, param->reg_SADJ); - udelay(10); - ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000); - udelay(10); - - ast_moutdwm(ast, 0x1E6E0004, param->dram_config); - ast_moutdwm(ast, 0x1E6E0008, 0x90040f); - ast_moutdwm(ast, 0x1E6E0010, param->reg_AC1); - ast_moutdwm(ast, 0x1E6E0014, param->reg_AC2); - ast_moutdwm(ast, 0x1E6E0020, param->reg_DQSIC); - ast_moutdwm(ast, 0x1E6E0080, 0x00000000); - ast_moutdwm(ast, 0x1E6E0084, 0x00000000); - ast_moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY); - ast_moutdwm(ast, 0x1E6E0018, 0x4000A170); - ast_moutdwm(ast, 0x1E6E0018, 0x00002370); - ast_moutdwm(ast, 0x1E6E0038, 0x00000000); - ast_moutdwm(ast, 0x1E6E0040, 0xFF444444); - ast_moutdwm(ast, 0x1E6E0044, 0x22222222); - ast_moutdwm(ast, 0x1E6E0048, 0x22222222); - ast_moutdwm(ast, 0x1E6E004C, 0x00000002); - ast_moutdwm(ast, 0x1E6E0050, 0x80000000); - ast_moutdwm(ast, 0x1E6E0050, 0x00000000); - ast_moutdwm(ast, 0x1E6E0054, 0); - ast_moutdwm(ast, 0x1E6E0060, param->reg_DRV); - ast_moutdwm(ast, 0x1E6E006C, param->reg_IOZ); - ast_moutdwm(ast, 0x1E6E0070, 0x00000000); - ast_moutdwm(ast, 0x1E6E0074, 0x00000000); - ast_moutdwm(ast, 0x1E6E0078, 0x00000000); - ast_moutdwm(ast, 0x1E6E007C, 0x00000000); - /* Wait MCLK2X lock to MCLK */ - do { - data = ast_mindwm(ast, 0x1E6E001C); - } while (!(data & 0x08000000)); - data = ast_mindwm(ast, 0x1E6E001C); - data = (data >> 8) & 0xff; - while ((data & 0x08) || ((data & 0x7) < 2) || (data < 4)) { - data2 = (ast_mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4; - if ((data2 & 0xff) > param->madj_max) { - break; - } - ast_moutdwm(ast, 0x1E6E0064, data2); - if (data2 & 0x00100000) { - data2 = ((data2 & 0xff) >> 3) + 3; - } else { - data2 = ((data2 & 0xff) >> 2) + 5; - } - data = ast_mindwm(ast, 0x1E6E0068) & 0xffff00ff; - data2 += data & 0xff; - data = data | (data2 << 8); - ast_moutdwm(ast, 0x1E6E0068, data); - udelay(10); - ast_moutdwm(ast, 0x1E6E0064, ast_mindwm(ast, 0x1E6E0064) | 0xC0000); - udelay(10); - data = ast_mindwm(ast, 0x1E6E0018) & 0xfffff1ff; - ast_moutdwm(ast, 0x1E6E0018, data); - data = data | 0x200; - ast_moutdwm(ast, 0x1E6E0018, data); - do { - data = ast_mindwm(ast, 0x1E6E001C); - } while (!(data & 0x08000000)); - - data = ast_mindwm(ast, 0x1E6E001C); - data = (data >> 8) & 0xff; - } - ast_moutdwm(ast, 0x1E720058, ast_mindwm(ast, 0x1E6E0068) & 0xffff); - data = ast_mindwm(ast, 0x1E6E0018) | 0xC00; - ast_moutdwm(ast, 0x1E6E0018, data); - - ast_moutdwm(ast, 0x1E6E0034, 0x00000001); - ast_moutdwm(ast, 0x1E6E000C, 0x00000040); - udelay(50); - /* Mode Register Setting */ - ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100); - ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS); - ast_moutdwm(ast, 0x1E6E0028, 0x00000005); - ast_moutdwm(ast, 0x1E6E0028, 0x00000007); - ast_moutdwm(ast, 0x1E6E0028, 0x00000003); - ast_moutdwm(ast, 0x1E6E0028, 0x00000001); - ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS); - ast_moutdwm(ast, 0x1E6E000C, 0x00005C08); - ast_moutdwm(ast, 0x1E6E0028, 0x00000001); - - ast_moutdwm(ast, 0x1E6E000C, 0x00005C01); - data = 0; - if (param->wodt) { - data = 0x300; - } - if (param->rodt) { - data = data | 0x3000 | ((param->reg_AC2 & 0x60000) >> 3); - } - ast_moutdwm(ast, 0x1E6E0034, data | 0x3); - - /* Calibrate the DQSI delay */ - if ((cbr_dll2(ast, param) == false) && (retry++ < 10)) - goto ddr3_init_start; - - ast_moutdwm(ast, 0x1E6E0120, param->reg_FREQ); - /* ECC Memory Initialization */ -#ifdef ECC - ast_moutdwm(ast, 0x1E6E007C, 0x00000000); - ast_moutdwm(ast, 0x1E6E0070, 0x221); - do { - data = ast_mindwm(ast, 0x1E6E0070); - } while (!(data & 0x00001000)); - ast_moutdwm(ast, 0x1E6E0070, 0x00000000); - ast_moutdwm(ast, 0x1E6E0050, 0x80000000); - ast_moutdwm(ast, 0x1E6E0050, 0x00000000); -#endif - - -} - -static void get_ddr2_info(struct ast_device *ast, struct ast2300_dram_param *param) -{ - u32 trap, trap_AC2, trap_MRS; - - ast_moutdwm(ast, 0x1E6E2000, 0x1688A8A8); - - /* Ger trap info */ - trap = (ast_mindwm(ast, 0x1E6E2070) >> 25) & 0x3; - trap_AC2 = (trap << 20) | (trap << 16); - trap_AC2 += 0x00110000; - trap_MRS = 0x00000040 | (trap << 4); - - - param->reg_MADJ = 0x00034C4C; - param->reg_SADJ = 0x00001800; - param->reg_DRV = 0x000000F0; - param->reg_PERIOD = param->dram_freq; - param->rodt = 0; - - switch (param->dram_freq) { - case 264: - ast_moutdwm(ast, 0x1E6E2020, 0x0130); - param->wodt = 0; - param->reg_AC1 = 0x11101513; - param->reg_AC2 = 0x78117011; - param->reg_DQSIC = 0x00000092; - param->reg_MRS = 0x00000842; - param->reg_EMRS = 0x00000000; - param->reg_DRV = 0x000000F0; - param->reg_IOZ = 0x00000034; - param->reg_DQIDLY = 0x0000005A; - param->reg_FREQ = 0x00004AC0; - param->madj_max = 138; - param->dll2_finetune_step = 3; - break; - case 336: - ast_moutdwm(ast, 0x1E6E2020, 0x0190); - param->wodt = 1; - param->reg_AC1 = 0x22202613; - param->reg_AC2 = 0xAA009016 | trap_AC2; - param->reg_DQSIC = 0x000000BA; - param->reg_MRS = 0x00000A02 | trap_MRS; - param->reg_EMRS = 0x00000040; - param->reg_DRV = 0x000000FA; - param->reg_IOZ = 0x00000034; - param->reg_DQIDLY = 0x00000074; - param->reg_FREQ = 0x00004DC0; - param->madj_max = 96; - param->dll2_finetune_step = 3; - switch (param->dram_chipid) { - default: - case AST_DRAM_512Mx16: - param->reg_AC2 = 0xAA009012 | trap_AC2; - break; - case AST_DRAM_1Gx16: - param->reg_AC2 = 0xAA009016 | trap_AC2; - break; - case AST_DRAM_2Gx16: - param->reg_AC2 = 0xAA009023 | trap_AC2; - break; - case AST_DRAM_4Gx16: - param->reg_AC2 = 0xAA00903B | trap_AC2; - break; - } - break; - default: - case 396: - ast_moutdwm(ast, 0x1E6E2020, 0x03F1); - param->wodt = 1; - param->rodt = 0; - param->reg_AC1 = 0x33302714; - param->reg_AC2 = 0xCC00B01B | trap_AC2; - param->reg_DQSIC = 0x000000E2; - param->reg_MRS = 0x00000C02 | trap_MRS; - param->reg_EMRS = 0x00000040; - param->reg_DRV = 0x000000FA; - param->reg_IOZ = 0x00000034; - param->reg_DQIDLY = 0x00000089; - param->reg_FREQ = 0x00005040; - param->madj_max = 96; - param->dll2_finetune_step = 4; - - switch (param->dram_chipid) { - case AST_DRAM_512Mx16: - param->reg_AC2 = 0xCC00B016 | trap_AC2; - break; - default: - case AST_DRAM_1Gx16: - param->reg_AC2 = 0xCC00B01B | trap_AC2; - break; - case AST_DRAM_2Gx16: - param->reg_AC2 = 0xCC00B02B | trap_AC2; - break; - case AST_DRAM_4Gx16: - param->reg_AC2 = 0xCC00B03F | trap_AC2; - break; - } - - break; - - case 408: - ast_moutdwm(ast, 0x1E6E2020, 0x01F0); - param->wodt = 1; - param->rodt = 0; - param->reg_AC1 = 0x33302714; - param->reg_AC2 = 0xCC00B01B | trap_AC2; - param->reg_DQSIC = 0x000000E2; - param->reg_MRS = 0x00000C02 | trap_MRS; - param->reg_EMRS = 0x00000040; - param->reg_DRV = 0x000000FA; - param->reg_IOZ = 0x00000034; - param->reg_DQIDLY = 0x00000089; - param->reg_FREQ = 0x000050C0; - param->madj_max = 96; - param->dll2_finetune_step = 4; - - switch (param->dram_chipid) { - case AST_DRAM_512Mx16: - param->reg_AC2 = 0xCC00B016 | trap_AC2; - break; - default: - case AST_DRAM_1Gx16: - param->reg_AC2 = 0xCC00B01B | trap_AC2; - break; - case AST_DRAM_2Gx16: - param->reg_AC2 = 0xCC00B02B | trap_AC2; - break; - case AST_DRAM_4Gx16: - param->reg_AC2 = 0xCC00B03F | trap_AC2; - break; - } - - break; - case 456: - ast_moutdwm(ast, 0x1E6E2020, 0x0230); - param->wodt = 0; - param->reg_AC1 = 0x33302815; - param->reg_AC2 = 0xCD44B01E; - param->reg_DQSIC = 0x000000FC; - param->reg_MRS = 0x00000E72; - param->reg_EMRS = 0x00000000; - param->reg_DRV = 0x00000000; - param->reg_IOZ = 0x00000034; - param->reg_DQIDLY = 0x00000097; - param->reg_FREQ = 0x000052C0; - param->madj_max = 88; - param->dll2_finetune_step = 3; - break; - case 504: - ast_moutdwm(ast, 0x1E6E2020, 0x0261); - param->wodt = 1; - param->rodt = 1; - param->reg_AC1 = 0x33302815; - param->reg_AC2 = 0xDE44C022; - param->reg_DQSIC = 0x00000117; - param->reg_MRS = 0x00000E72; - param->reg_EMRS = 0x00000040; - param->reg_DRV = 0x0000000A; - param->reg_IOZ = 0x00000045; - param->reg_DQIDLY = 0x000000A0; - param->reg_FREQ = 0x000054C0; - param->madj_max = 79; - param->dll2_finetune_step = 3; - break; - case 528: - ast_moutdwm(ast, 0x1E6E2020, 0x0120); - param->wodt = 1; - param->rodt = 1; - param->reg_AC1 = 0x33302815; - param->reg_AC2 = 0xEF44D024; - param->reg_DQSIC = 0x00000125; - param->reg_MRS = 0x00000E72; - param->reg_EMRS = 0x00000004; - param->reg_DRV = 0x000000F9; - param->reg_IOZ = 0x00000045; - param->reg_DQIDLY = 0x000000A7; - param->reg_FREQ = 0x000055C0; - param->madj_max = 76; - param->dll2_finetune_step = 3; - break; - case 552: - ast_moutdwm(ast, 0x1E6E2020, 0x02A1); - param->wodt = 1; - param->rodt = 1; - param->reg_AC1 = 0x43402915; - param->reg_AC2 = 0xFF44E025; - param->reg_DQSIC = 0x00000132; - param->reg_MRS = 0x00000E72; - param->reg_EMRS = 0x00000040; - param->reg_DRV = 0x0000000A; - param->reg_IOZ = 0x00000045; - param->reg_DQIDLY = 0x000000AD; - param->reg_FREQ = 0x000056C0; - param->madj_max = 76; - param->dll2_finetune_step = 3; - break; - case 576: - ast_moutdwm(ast, 0x1E6E2020, 0x0140); - param->wodt = 1; - param->rodt = 1; - param->reg_AC1 = 0x43402915; - param->reg_AC2 = 0xFF44E027; - param->reg_DQSIC = 0x0000013F; - param->reg_MRS = 0x00000E72; - param->reg_EMRS = 0x00000004; - param->reg_DRV = 0x000000F5; - param->reg_IOZ = 0x00000045; - param->reg_DQIDLY = 0x000000B3; - param->reg_FREQ = 0x000057C0; - param->madj_max = 76; - param->dll2_finetune_step = 3; - break; - } - - switch (param->dram_chipid) { - case AST_DRAM_512Mx16: - param->dram_config = 0x100; - break; - default: - case AST_DRAM_1Gx16: - param->dram_config = 0x121; - break; - case AST_DRAM_2Gx16: - param->dram_config = 0x122; - break; - case AST_DRAM_4Gx16: - param->dram_config = 0x123; - break; - } /* switch size */ - - switch (param->vram_size) { - default: - case SZ_8M: - param->dram_config |= 0x00; - break; - case SZ_16M: - param->dram_config |= 0x04; - break; - case SZ_32M: - param->dram_config |= 0x08; - break; - case SZ_64M: - param->dram_config |= 0x0c; - break; - } -} - -static void ddr2_init(struct ast_device *ast, struct ast2300_dram_param *param) -{ - u32 data, data2, retry = 0; - -ddr2_init_start: - ast_moutdwm(ast, 0x1E6E0000, 0xFC600309); - ast_moutdwm(ast, 0x1E6E0018, 0x00000100); - ast_moutdwm(ast, 0x1E6E0024, 0x00000000); - ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ); - ast_moutdwm(ast, 0x1E6E0068, param->reg_SADJ); - udelay(10); - ast_moutdwm(ast, 0x1E6E0064, param->reg_MADJ | 0xC0000); - udelay(10); - - ast_moutdwm(ast, 0x1E6E0004, param->dram_config); - ast_moutdwm(ast, 0x1E6E0008, 0x90040f); - ast_moutdwm(ast, 0x1E6E0010, param->reg_AC1); - ast_moutdwm(ast, 0x1E6E0014, param->reg_AC2); - ast_moutdwm(ast, 0x1E6E0020, param->reg_DQSIC); - ast_moutdwm(ast, 0x1E6E0080, 0x00000000); - ast_moutdwm(ast, 0x1E6E0084, 0x00000000); - ast_moutdwm(ast, 0x1E6E0088, param->reg_DQIDLY); - ast_moutdwm(ast, 0x1E6E0018, 0x4000A130); - ast_moutdwm(ast, 0x1E6E0018, 0x00002330); - ast_moutdwm(ast, 0x1E6E0038, 0x00000000); - ast_moutdwm(ast, 0x1E6E0040, 0xFF808000); - ast_moutdwm(ast, 0x1E6E0044, 0x88848466); - ast_moutdwm(ast, 0x1E6E0048, 0x44440008); - ast_moutdwm(ast, 0x1E6E004C, 0x00000000); - ast_moutdwm(ast, 0x1E6E0050, 0x80000000); - ast_moutdwm(ast, 0x1E6E0050, 0x00000000); - ast_moutdwm(ast, 0x1E6E0054, 0); - ast_moutdwm(ast, 0x1E6E0060, param->reg_DRV); - ast_moutdwm(ast, 0x1E6E006C, param->reg_IOZ); - ast_moutdwm(ast, 0x1E6E0070, 0x00000000); - ast_moutdwm(ast, 0x1E6E0074, 0x00000000); - ast_moutdwm(ast, 0x1E6E0078, 0x00000000); - ast_moutdwm(ast, 0x1E6E007C, 0x00000000); - - /* Wait MCLK2X lock to MCLK */ - do { - data = ast_mindwm(ast, 0x1E6E001C); - } while (!(data & 0x08000000)); - data = ast_mindwm(ast, 0x1E6E001C); - data = (data >> 8) & 0xff; - while ((data & 0x08) || ((data & 0x7) < 2) || (data < 4)) { - data2 = (ast_mindwm(ast, 0x1E6E0064) & 0xfff3ffff) + 4; - if ((data2 & 0xff) > param->madj_max) { - break; - } - ast_moutdwm(ast, 0x1E6E0064, data2); - if (data2 & 0x00100000) { - data2 = ((data2 & 0xff) >> 3) + 3; - } else { - data2 = ((data2 & 0xff) >> 2) + 5; - } - data = ast_mindwm(ast, 0x1E6E0068) & 0xffff00ff; - data2 += data & 0xff; - data = data | (data2 << 8); - ast_moutdwm(ast, 0x1E6E0068, data); - udelay(10); - ast_moutdwm(ast, 0x1E6E0064, ast_mindwm(ast, 0x1E6E0064) | 0xC0000); - udelay(10); - data = ast_mindwm(ast, 0x1E6E0018) & 0xfffff1ff; - ast_moutdwm(ast, 0x1E6E0018, data); - data = data | 0x200; - ast_moutdwm(ast, 0x1E6E0018, data); - do { - data = ast_mindwm(ast, 0x1E6E001C); - } while (!(data & 0x08000000)); - - data = ast_mindwm(ast, 0x1E6E001C); - data = (data >> 8) & 0xff; - } - ast_moutdwm(ast, 0x1E720058, ast_mindwm(ast, 0x1E6E0008) & 0xffff); - data = ast_mindwm(ast, 0x1E6E0018) | 0xC00; - ast_moutdwm(ast, 0x1E6E0018, data); - - ast_moutdwm(ast, 0x1E6E0034, 0x00000001); - ast_moutdwm(ast, 0x1E6E000C, 0x00000000); - udelay(50); - /* Mode Register Setting */ - ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS | 0x100); - ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS); - ast_moutdwm(ast, 0x1E6E0028, 0x00000005); - ast_moutdwm(ast, 0x1E6E0028, 0x00000007); - ast_moutdwm(ast, 0x1E6E0028, 0x00000003); - ast_moutdwm(ast, 0x1E6E0028, 0x00000001); - - ast_moutdwm(ast, 0x1E6E000C, 0x00005C08); - ast_moutdwm(ast, 0x1E6E002C, param->reg_MRS); - ast_moutdwm(ast, 0x1E6E0028, 0x00000001); - ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS | 0x380); - ast_moutdwm(ast, 0x1E6E0028, 0x00000003); - ast_moutdwm(ast, 0x1E6E0030, param->reg_EMRS); - ast_moutdwm(ast, 0x1E6E0028, 0x00000003); - - ast_moutdwm(ast, 0x1E6E000C, 0x7FFF5C01); - data = 0; - if (param->wodt) { - data = 0x500; - } - if (param->rodt) { - data = data | 0x3000 | ((param->reg_AC2 & 0x60000) >> 3); - } - ast_moutdwm(ast, 0x1E6E0034, data | 0x3); - ast_moutdwm(ast, 0x1E6E0120, param->reg_FREQ); - - /* Calibrate the DQSI delay */ - if ((cbr_dll2(ast, param) == false) && (retry++ < 10)) - goto ddr2_init_start; - - /* ECC Memory Initialization */ -#ifdef ECC - ast_moutdwm(ast, 0x1E6E007C, 0x00000000); - ast_moutdwm(ast, 0x1E6E0070, 0x221); - do { - data = ast_mindwm(ast, 0x1E6E0070); - } while (!(data & 0x00001000)); - ast_moutdwm(ast, 0x1E6E0070, 0x00000000); - ast_moutdwm(ast, 0x1E6E0050, 0x80000000); - ast_moutdwm(ast, 0x1E6E0050, 0x00000000); -#endif - -} - -static void ast_post_chip_2300(struct ast_device *ast) -{ - struct ast2300_dram_param param; - u32 temp; - u8 reg; - - reg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); - if ((reg & 0x80) == 0) {/* vga only */ - ast_write32(ast, 0xf004, 0x1e6e0000); - ast_write32(ast, 0xf000, 0x1); - ast_write32(ast, 0x12000, 0x1688a8a8); - do { - ; - } while (ast_read32(ast, 0x12000) != 0x1); - - ast_write32(ast, 0x10000, 0xfc600309); - do { - ; - } while (ast_read32(ast, 0x10000) != 0x1); - - /* Slow down CPU/AHB CLK in VGA only mode */ - temp = ast_read32(ast, 0x12008); - temp |= 0x73; - ast_write32(ast, 0x12008, temp); - - param.dram_freq = 396; - param.dram_type = AST_DDR3; - temp = ast_mindwm(ast, 0x1e6e2070); - if (temp & 0x01000000) - param.dram_type = AST_DDR2; - switch (temp & 0x18000000) { - case 0: - param.dram_chipid = AST_DRAM_512Mx16; - break; - default: - case 0x08000000: - param.dram_chipid = AST_DRAM_1Gx16; - break; - case 0x10000000: - param.dram_chipid = AST_DRAM_2Gx16; - break; - case 0x18000000: - param.dram_chipid = AST_DRAM_4Gx16; - break; - } - switch (temp & 0x0c) { - default: - case 0x00: - param.vram_size = SZ_8M; - break; - - case 0x04: - param.vram_size = SZ_16M; - break; - - case 0x08: - param.vram_size = SZ_32M; - break; - - case 0x0c: - param.vram_size = SZ_64M; - break; - } - - if (param.dram_type == AST_DDR3) { - get_ddr3_info(ast, ¶m); - ddr3_init(ast, ¶m); - } else { - get_ddr2_info(ast, ¶m); - ddr2_init(ast, ¶m); - } - - temp = ast_mindwm(ast, 0x1e6e2040); - ast_moutdwm(ast, 0x1e6e2040, temp | 0x40); - } - - /* wait ready */ - do { - reg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); - } while ((reg & 0x40) == 0); -} - -static bool cbr_test_2500(struct ast_device *ast) -{ - ast_moutdwm(ast, 0x1E6E0074, 0x0000FFFF); - ast_moutdwm(ast, 0x1E6E007C, 0xFF00FF00); - if (!mmc_test_burst(ast, 0)) - return false; - if (!mmc_test_single_2500(ast, 0)) - return false; - return true; -} - -static bool ddr_test_2500(struct ast_device *ast) -{ - ast_moutdwm(ast, 0x1E6E0074, 0x0000FFFF); - ast_moutdwm(ast, 0x1E6E007C, 0xFF00FF00); - if (!mmc_test_burst(ast, 0)) - return false; - if (!mmc_test_burst(ast, 1)) - return false; - if (!mmc_test_burst(ast, 2)) - return false; - if (!mmc_test_burst(ast, 3)) - return false; - if (!mmc_test_single_2500(ast, 0)) - return false; - return true; -} - -static void ddr_init_common_2500(struct ast_device *ast) -{ - ast_moutdwm(ast, 0x1E6E0034, 0x00020080); - ast_moutdwm(ast, 0x1E6E0008, 0x2003000F); - ast_moutdwm(ast, 0x1E6E0038, 0x00000FFF); - ast_moutdwm(ast, 0x1E6E0040, 0x88448844); - ast_moutdwm(ast, 0x1E6E0044, 0x24422288); - ast_moutdwm(ast, 0x1E6E0048, 0x22222222); - ast_moutdwm(ast, 0x1E6E004C, 0x22222222); - ast_moutdwm(ast, 0x1E6E0050, 0x80000000); - ast_moutdwm(ast, 0x1E6E0208, 0x00000000); - ast_moutdwm(ast, 0x1E6E0218, 0x00000000); - ast_moutdwm(ast, 0x1E6E0220, 0x00000000); - ast_moutdwm(ast, 0x1E6E0228, 0x00000000); - ast_moutdwm(ast, 0x1E6E0230, 0x00000000); - ast_moutdwm(ast, 0x1E6E02A8, 0x00000000); - ast_moutdwm(ast, 0x1E6E02B0, 0x00000000); - ast_moutdwm(ast, 0x1E6E0240, 0x86000000); - ast_moutdwm(ast, 0x1E6E0244, 0x00008600); - ast_moutdwm(ast, 0x1E6E0248, 0x80000000); - ast_moutdwm(ast, 0x1E6E024C, 0x80808080); -} - -static void ddr_phy_init_2500(struct ast_device *ast) -{ - u32 data, pass, timecnt; - - pass = 0; - ast_moutdwm(ast, 0x1E6E0060, 0x00000005); - while (!pass) { - for (timecnt = 0; timecnt < TIMEOUT; timecnt++) { - data = ast_mindwm(ast, 0x1E6E0060) & 0x1; - if (!data) - break; - } - if (timecnt != TIMEOUT) { - data = ast_mindwm(ast, 0x1E6E0300) & 0x000A0000; - if (!data) - pass = 1; - } - if (!pass) { - ast_moutdwm(ast, 0x1E6E0060, 0x00000000); - udelay(10); /* delay 10 us */ - ast_moutdwm(ast, 0x1E6E0060, 0x00000005); - } - } - - ast_moutdwm(ast, 0x1E6E0060, 0x00000006); -} - -/* - * Check DRAM Size - * 1Gb : 0x80000000 ~ 0x87FFFFFF - * 2Gb : 0x80000000 ~ 0x8FFFFFFF - * 4Gb : 0x80000000 ~ 0x9FFFFFFF - * 8Gb : 0x80000000 ~ 0xBFFFFFFF - */ -static void check_dram_size_2500(struct ast_device *ast, u32 tRFC) -{ - u32 reg_04, reg_14; - - reg_04 = ast_mindwm(ast, 0x1E6E0004) & 0xfffffffc; - reg_14 = ast_mindwm(ast, 0x1E6E0014) & 0xffffff00; - - ast_moutdwm(ast, 0xA0100000, 0x41424344); - ast_moutdwm(ast, 0x90100000, 0x35363738); - ast_moutdwm(ast, 0x88100000, 0x292A2B2C); - ast_moutdwm(ast, 0x80100000, 0x1D1E1F10); - - /* Check 8Gbit */ - if (ast_mindwm(ast, 0xA0100000) == 0x41424344) { - reg_04 |= 0x03; - reg_14 |= (tRFC >> 24) & 0xFF; - /* Check 4Gbit */ - } else if (ast_mindwm(ast, 0x90100000) == 0x35363738) { - reg_04 |= 0x02; - reg_14 |= (tRFC >> 16) & 0xFF; - /* Check 2Gbit */ - } else if (ast_mindwm(ast, 0x88100000) == 0x292A2B2C) { - reg_04 |= 0x01; - reg_14 |= (tRFC >> 8) & 0xFF; - } else { - reg_14 |= tRFC & 0xFF; - } - ast_moutdwm(ast, 0x1E6E0004, reg_04); - ast_moutdwm(ast, 0x1E6E0014, reg_14); -} - -static void enable_cache_2500(struct ast_device *ast) -{ - u32 reg_04, data; - - reg_04 = ast_mindwm(ast, 0x1E6E0004); - ast_moutdwm(ast, 0x1E6E0004, reg_04 | 0x1000); - - do - data = ast_mindwm(ast, 0x1E6E0004); - while (!(data & 0x80000)); - ast_moutdwm(ast, 0x1E6E0004, reg_04 | 0x400); -} - -static void set_mpll_2500(struct ast_device *ast) -{ - u32 addr, data, param; - - /* Reset MMC */ - ast_moutdwm(ast, 0x1E6E0000, 0xFC600309); - ast_moutdwm(ast, 0x1E6E0034, 0x00020080); - for (addr = 0x1e6e0004; addr < 0x1e6e0090;) { - ast_moutdwm(ast, addr, 0x0); - addr += 4; - } - ast_moutdwm(ast, 0x1E6E0034, 0x00020000); - - ast_moutdwm(ast, 0x1E6E2000, 0x1688A8A8); - data = ast_mindwm(ast, 0x1E6E2070) & 0x00800000; - if (data) { - /* CLKIN = 25MHz */ - param = 0x930023E0; - ast_moutdwm(ast, 0x1E6E2160, 0x00011320); - } else { - /* CLKIN = 24MHz */ - param = 0x93002400; - } - ast_moutdwm(ast, 0x1E6E2020, param); - udelay(100); -} - -static void reset_mmc_2500(struct ast_device *ast) -{ - ast_moutdwm(ast, 0x1E78505C, 0x00000004); - ast_moutdwm(ast, 0x1E785044, 0x00000001); - ast_moutdwm(ast, 0x1E785048, 0x00004755); - ast_moutdwm(ast, 0x1E78504C, 0x00000013); - mdelay(100); - ast_moutdwm(ast, 0x1E785054, 0x00000077); - ast_moutdwm(ast, 0x1E6E0000, 0xFC600309); -} - -static void ddr3_init_2500(struct ast_device *ast, const u32 *ddr_table) -{ - - ast_moutdwm(ast, 0x1E6E0004, 0x00000303); - ast_moutdwm(ast, 0x1E6E0010, ddr_table[REGIDX_010]); - ast_moutdwm(ast, 0x1E6E0014, ddr_table[REGIDX_014]); - ast_moutdwm(ast, 0x1E6E0018, ddr_table[REGIDX_018]); - ast_moutdwm(ast, 0x1E6E0020, ddr_table[REGIDX_020]); /* MODEREG4/6 */ - ast_moutdwm(ast, 0x1E6E0024, ddr_table[REGIDX_024]); /* MODEREG5 */ - ast_moutdwm(ast, 0x1E6E002C, ddr_table[REGIDX_02C] | 0x100); /* MODEREG0/2 */ - ast_moutdwm(ast, 0x1E6E0030, ddr_table[REGIDX_030]); /* MODEREG1/3 */ - - /* DDR PHY Setting */ - ast_moutdwm(ast, 0x1E6E0200, 0x02492AAE); - ast_moutdwm(ast, 0x1E6E0204, 0x00001001); - ast_moutdwm(ast, 0x1E6E020C, 0x55E00B0B); - ast_moutdwm(ast, 0x1E6E0210, 0x20000000); - ast_moutdwm(ast, 0x1E6E0214, ddr_table[REGIDX_214]); - ast_moutdwm(ast, 0x1E6E02E0, ddr_table[REGIDX_2E0]); - ast_moutdwm(ast, 0x1E6E02E4, ddr_table[REGIDX_2E4]); - ast_moutdwm(ast, 0x1E6E02E8, ddr_table[REGIDX_2E8]); - ast_moutdwm(ast, 0x1E6E02EC, ddr_table[REGIDX_2EC]); - ast_moutdwm(ast, 0x1E6E02F0, ddr_table[REGIDX_2F0]); - ast_moutdwm(ast, 0x1E6E02F4, ddr_table[REGIDX_2F4]); - ast_moutdwm(ast, 0x1E6E02F8, ddr_table[REGIDX_2F8]); - ast_moutdwm(ast, 0x1E6E0290, 0x00100008); - ast_moutdwm(ast, 0x1E6E02C0, 0x00000006); - - /* Controller Setting */ - ast_moutdwm(ast, 0x1E6E0034, 0x00020091); - - /* Wait DDR PHY init done */ - ddr_phy_init_2500(ast); - - ast_moutdwm(ast, 0x1E6E0120, ddr_table[REGIDX_PLL]); - ast_moutdwm(ast, 0x1E6E000C, 0x42AA5C81); - ast_moutdwm(ast, 0x1E6E0034, 0x0001AF93); - - check_dram_size_2500(ast, ddr_table[REGIDX_RFC]); - enable_cache_2500(ast); - ast_moutdwm(ast, 0x1E6E001C, 0x00000008); - ast_moutdwm(ast, 0x1E6E0038, 0xFFFFFF00); -} - -static void ddr4_init_2500(struct ast_device *ast, const u32 *ddr_table) -{ - u32 data, data2, pass, retrycnt; - u32 ddr_vref, phy_vref; - u32 min_ddr_vref = 0, min_phy_vref = 0; - u32 max_ddr_vref = 0, max_phy_vref = 0; - - ast_moutdwm(ast, 0x1E6E0004, 0x00000313); - ast_moutdwm(ast, 0x1E6E0010, ddr_table[REGIDX_010]); - ast_moutdwm(ast, 0x1E6E0014, ddr_table[REGIDX_014]); - ast_moutdwm(ast, 0x1E6E0018, ddr_table[REGIDX_018]); - ast_moutdwm(ast, 0x1E6E0020, ddr_table[REGIDX_020]); /* MODEREG4/6 */ - ast_moutdwm(ast, 0x1E6E0024, ddr_table[REGIDX_024]); /* MODEREG5 */ - ast_moutdwm(ast, 0x1E6E002C, ddr_table[REGIDX_02C] | 0x100); /* MODEREG0/2 */ - ast_moutdwm(ast, 0x1E6E0030, ddr_table[REGIDX_030]); /* MODEREG1/3 */ - - /* DDR PHY Setting */ - ast_moutdwm(ast, 0x1E6E0200, 0x42492AAE); - ast_moutdwm(ast, 0x1E6E0204, 0x09002000); - ast_moutdwm(ast, 0x1E6E020C, 0x55E00B0B); - ast_moutdwm(ast, 0x1E6E0210, 0x20000000); - ast_moutdwm(ast, 0x1E6E0214, ddr_table[REGIDX_214]); - ast_moutdwm(ast, 0x1E6E02E0, ddr_table[REGIDX_2E0]); - ast_moutdwm(ast, 0x1E6E02E4, ddr_table[REGIDX_2E4]); - ast_moutdwm(ast, 0x1E6E02E8, ddr_table[REGIDX_2E8]); - ast_moutdwm(ast, 0x1E6E02EC, ddr_table[REGIDX_2EC]); - ast_moutdwm(ast, 0x1E6E02F0, ddr_table[REGIDX_2F0]); - ast_moutdwm(ast, 0x1E6E02F4, ddr_table[REGIDX_2F4]); - ast_moutdwm(ast, 0x1E6E02F8, ddr_table[REGIDX_2F8]); - ast_moutdwm(ast, 0x1E6E0290, 0x00100008); - ast_moutdwm(ast, 0x1E6E02C4, 0x3C183C3C); - ast_moutdwm(ast, 0x1E6E02C8, 0x00631E0E); - - /* Controller Setting */ - ast_moutdwm(ast, 0x1E6E0034, 0x0001A991); - - /* Train PHY Vref first */ - pass = 0; - - for (retrycnt = 0; retrycnt < 4 && pass == 0; retrycnt++) { - max_phy_vref = 0x0; - pass = 0; - ast_moutdwm(ast, 0x1E6E02C0, 0x00001C06); - for (phy_vref = 0x40; phy_vref < 0x80; phy_vref++) { - ast_moutdwm(ast, 0x1E6E000C, 0x00000000); - ast_moutdwm(ast, 0x1E6E0060, 0x00000000); - ast_moutdwm(ast, 0x1E6E02CC, phy_vref | (phy_vref << 8)); - /* Fire DFI Init */ - ddr_phy_init_2500(ast); - ast_moutdwm(ast, 0x1E6E000C, 0x00005C01); - if (cbr_test_2500(ast)) { - pass++; - data = ast_mindwm(ast, 0x1E6E03D0); - data2 = data >> 8; - data = data & 0xff; - if (data > data2) - data = data2; - if (max_phy_vref < data) { - max_phy_vref = data; - min_phy_vref = phy_vref; - } - } else if (pass > 0) - break; - } - } - ast_moutdwm(ast, 0x1E6E02CC, min_phy_vref | (min_phy_vref << 8)); - - /* Train DDR Vref next */ - pass = 0; - - for (retrycnt = 0; retrycnt < 4 && pass == 0; retrycnt++) { - min_ddr_vref = 0xFF; - max_ddr_vref = 0x0; - pass = 0; - for (ddr_vref = 0x00; ddr_vref < 0x40; ddr_vref++) { - ast_moutdwm(ast, 0x1E6E000C, 0x00000000); - ast_moutdwm(ast, 0x1E6E0060, 0x00000000); - ast_moutdwm(ast, 0x1E6E02C0, 0x00000006 | (ddr_vref << 8)); - /* Fire DFI Init */ - ddr_phy_init_2500(ast); - ast_moutdwm(ast, 0x1E6E000C, 0x00005C01); - if (cbr_test_2500(ast)) { - pass++; - if (min_ddr_vref > ddr_vref) - min_ddr_vref = ddr_vref; - if (max_ddr_vref < ddr_vref) - max_ddr_vref = ddr_vref; - } else if (pass != 0) - break; - } - } - - ast_moutdwm(ast, 0x1E6E000C, 0x00000000); - ast_moutdwm(ast, 0x1E6E0060, 0x00000000); - ddr_vref = (min_ddr_vref + max_ddr_vref + 1) >> 1; - ast_moutdwm(ast, 0x1E6E02C0, 0x00000006 | (ddr_vref << 8)); - - /* Wait DDR PHY init done */ - ddr_phy_init_2500(ast); - - ast_moutdwm(ast, 0x1E6E0120, ddr_table[REGIDX_PLL]); - ast_moutdwm(ast, 0x1E6E000C, 0x42AA5C81); - ast_moutdwm(ast, 0x1E6E0034, 0x0001AF93); - - check_dram_size_2500(ast, ddr_table[REGIDX_RFC]); - enable_cache_2500(ast); - ast_moutdwm(ast, 0x1E6E001C, 0x00000008); - ast_moutdwm(ast, 0x1E6E0038, 0xFFFFFF00); -} - -static bool ast_dram_init_2500(struct ast_device *ast) -{ - u32 data; - u32 max_tries = 5; - - do { - if (max_tries-- == 0) - return false; - set_mpll_2500(ast); - reset_mmc_2500(ast); - ddr_init_common_2500(ast); - - data = ast_mindwm(ast, 0x1E6E2070); - if (data & 0x01000000) - ddr4_init_2500(ast, ast2500_ddr4_1600_timing_table); - else - ddr3_init_2500(ast, ast2500_ddr3_1600_timing_table); - } while (!ddr_test_2500(ast)); - - ast_moutdwm(ast, 0x1E6E2040, ast_mindwm(ast, 0x1E6E2040) | 0x41); - - /* Patch code */ - data = ast_mindwm(ast, 0x1E6E200C) & 0xF9FFFFFF; - ast_moutdwm(ast, 0x1E6E200C, data | 0x10000000); - - return true; -} - -void ast_patch_ahb_2500(void __iomem *regs) -{ - u32 data; - - /* Clear bus lock condition */ - __ast_moutdwm(regs, 0x1e600000, 0xAEED1A03); - __ast_moutdwm(regs, 0x1e600084, 0x00010000); - __ast_moutdwm(regs, 0x1e600088, 0x00000000); - __ast_moutdwm(regs, 0x1e6e2000, 0x1688A8A8); - - data = __ast_mindwm(regs, 0x1e6e2070); - if (data & 0x08000000) { /* check fast reset */ - /* - * If "Fast restet" is enabled for ARM-ICE debugger, - * then WDT needs to enable, that - * WDT04 is WDT#1 Reload reg. - * WDT08 is WDT#1 counter restart reg to avoid system deadlock - * WDT0C is WDT#1 control reg - * [6:5]:= 01:Full chip - * [4]:= 1:1MHz clock source - * [1]:= 1:WDT will be cleeared and disabled after timeout occurs - * [0]:= 1:WDT enable - */ - __ast_moutdwm(regs, 0x1E785004, 0x00000010); - __ast_moutdwm(regs, 0x1E785008, 0x00004755); - __ast_moutdwm(regs, 0x1E78500c, 0x00000033); - udelay(1000); - } - - do { - __ast_moutdwm(regs, 0x1e6e2000, 0x1688A8A8); - data = __ast_mindwm(regs, 0x1e6e2000); - } while (data != 1); - - __ast_moutdwm(regs, 0x1e6e207c, 0x08000000); /* clear fast reset */ -} - -void ast_post_chip_2500(struct ast_device *ast) -{ - struct drm_device *dev = &ast->base; - u32 temp; - u8 reg; - - reg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); - if ((reg & AST_IO_VGACRD0_VRAM_INIT_STATUS_MASK) == 0) {/* vga only */ - /* Clear bus lock condition */ - ast_patch_ahb_2500(ast->regs); - - /* Disable watchdog */ - ast_moutdwm(ast, 0x1E78502C, 0x00000000); - ast_moutdwm(ast, 0x1E78504C, 0x00000000); - - /* - * Reset USB port to patch USB unknown device issue - * SCU90 is Multi-function Pin Control #5 - * [29]:= 1:Enable USB2.0 Host port#1 (that the mutually shared USB2.0 Hub - * port). - * SCU94 is Multi-function Pin Control #6 - * [14:13]:= 1x:USB2.0 Host2 controller - * SCU70 is Hardware Strap reg - * [23]:= 1:CLKIN is 25MHz and USBCK1 = 24/48 MHz (determined by - * [18]: 0(24)/1(48) MHz) - * SCU7C is Write clear reg to SCU70 - * [23]:= write 1 and then SCU70[23] will be clear as 0b. - */ - ast_moutdwm(ast, 0x1E6E2090, 0x20000000); - ast_moutdwm(ast, 0x1E6E2094, 0x00004000); - if (ast_mindwm(ast, 0x1E6E2070) & 0x00800000) { - ast_moutdwm(ast, 0x1E6E207C, 0x00800000); - mdelay(100); - ast_moutdwm(ast, 0x1E6E2070, 0x00800000); - } - /* Modify eSPI reset pin */ - temp = ast_mindwm(ast, 0x1E6E2070); - if (temp & 0x02000000) - ast_moutdwm(ast, 0x1E6E207C, 0x00004000); - - /* Slow down CPU/AHB CLK in VGA only mode */ - temp = ast_read32(ast, 0x12008); - temp |= 0x73; - ast_write32(ast, 0x12008, temp); - - if (!ast_dram_init_2500(ast)) - drm_err(dev, "DRAM init failed !\n"); - - temp = ast_mindwm(ast, 0x1e6e2040); - ast_moutdwm(ast, 0x1e6e2040, temp | 0x40); - } - - /* wait ready */ - do { - reg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff); - } while ((reg & 0x40) == 0); -} diff --git a/drivers/gpu/drm/ast/ast_post.h b/drivers/gpu/drm/ast/ast_post.h new file mode 100644 index 000000000000..aa5d247bebe8 --- /dev/null +++ b/drivers/gpu/drm/ast/ast_post.h @@ -0,0 +1,50 @@ +/* SPDX-License-Identifier: MIT */ + +#ifndef AST_POST_H +#define AST_POST_H + +#include <linux/limits.h> +#include <linux/types.h> + +struct ast_device; + +/* DRAM timing tables */ +struct ast_dramstruct { + u16 index; + u32 data; +}; + +/* hardware fields */ +#define __AST_DRAMSTRUCT_DRAM_TYPE 0x0004 + +/* control commands */ +#define __AST_DRAMSTRUCT_UDELAY 0xff00 +#define __AST_DRAMSTRUCT_INVALID 0xffff + +#define __AST_DRAMSTRUCT_INDEX(_name) \ + (__AST_DRAMSTRUCT_ ## _name) + +#define AST_DRAMSTRUCT_INIT(_name, _value) \ + { __AST_DRAMSTRUCT_INDEX(_name), (_value) } + +#define AST_DRAMSTRUCT_UDELAY(_usecs) \ + AST_DRAMSTRUCT_INIT(UDELAY, _usecs) +#define AST_DRAMSTRUCT_INVALID \ + AST_DRAMSTRUCT_INIT(INVALID, U32_MAX) + +#define AST_DRAMSTRUCT_IS(_entry, _name) \ + ((_entry)->index == __AST_DRAMSTRUCT_INDEX(_name)) + +u32 __ast_mindwm(void __iomem *regs, u32 r); +void __ast_moutdwm(void __iomem *regs, u32 r, u32 v); + +bool mmc_test(struct ast_device *ast, u32 datagen, u8 test_ctl); +bool mmc_test_burst(struct ast_device *ast, u32 datagen); + +/* ast_2000.c */ +void ast_2000_set_def_ext_reg(struct ast_device *ast); + +/* ast_2300.c */ +void ast_2300_set_def_ext_reg(struct ast_device *ast); + +#endif diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c index 9df18a8f2e37..f59d19b4b81a 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c @@ -1262,9 +1262,7 @@ static int adv7511_probe(struct i2c_client *i2c) adv7511->bridge.ops = DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID | - DRM_BRIDGE_OP_HDMI | - DRM_BRIDGE_OP_HDMI_AUDIO | - DRM_BRIDGE_OP_HDMI_CEC_ADAPTER; + DRM_BRIDGE_OP_HDMI; if (adv7511->i2c_main->irq) adv7511->bridge.ops |= DRM_BRIDGE_OP_HPD; @@ -1272,6 +1270,7 @@ static int adv7511_probe(struct i2c_client *i2c) adv7511->bridge.product = adv7511->info->name; #ifdef CONFIG_DRM_I2C_ADV7511_AUDIO + adv7511->bridge.ops |= DRM_BRIDGE_OP_HDMI_AUDIO; adv7511->bridge.hdmi_audio_dev = dev; adv7511->bridge.hdmi_audio_max_i2s_playback_channels = 2; adv7511->bridge.hdmi_audio_i2s_formats = (SNDRV_PCM_FMTBIT_S16_LE | @@ -1284,6 +1283,7 @@ static int adv7511_probe(struct i2c_client *i2c) #endif #ifdef CONFIG_DRM_I2C_ADV7511_CEC + adv7511->bridge.ops |= DRM_BRIDGE_OP_HDMI_CEC_ADAPTER; adv7511->bridge.hdmi_cec_dev = dev; adv7511->bridge.hdmi_cec_adapter_name = dev_name(dev); adv7511->bridge.hdmi_cec_available_las = ADV7511_MAX_ADDRS; diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c index a1bc3e96dd35..ed35e567d117 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c +++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c @@ -1041,7 +1041,7 @@ static int analogix_dp_bridge_attach(struct drm_bridge *bridge, struct drm_encoder *encoder, enum drm_bridge_attach_flags flags) { - struct analogix_dp_device *dp = bridge->driver_private; + struct analogix_dp_device *dp = to_dp(bridge); struct drm_connector *connector = NULL; int ret = 0; @@ -1125,7 +1125,7 @@ struct drm_crtc *analogix_dp_get_new_crtc(struct analogix_dp_device *dp, static void analogix_dp_bridge_atomic_pre_enable(struct drm_bridge *bridge, struct drm_atomic_state *old_state) { - struct analogix_dp_device *dp = bridge->driver_private; + struct analogix_dp_device *dp = to_dp(bridge); struct drm_crtc *crtc; struct drm_crtc_state *old_crtc_state; @@ -1180,7 +1180,7 @@ out_dp_init: static void analogix_dp_bridge_atomic_enable(struct drm_bridge *bridge, struct drm_atomic_state *old_state) { - struct analogix_dp_device *dp = bridge->driver_private; + struct analogix_dp_device *dp = to_dp(bridge); struct drm_crtc *crtc; struct drm_crtc_state *old_crtc_state; int timeout_loop = 0; @@ -1217,7 +1217,7 @@ static void analogix_dp_bridge_atomic_enable(struct drm_bridge *bridge, static void analogix_dp_bridge_disable(struct drm_bridge *bridge) { - struct analogix_dp_device *dp = bridge->driver_private; + struct analogix_dp_device *dp = to_dp(bridge); if (dp->dpms_mode != DRM_MODE_DPMS_ON) return; @@ -1240,7 +1240,7 @@ static void analogix_dp_bridge_disable(struct drm_bridge *bridge) static void analogix_dp_bridge_atomic_disable(struct drm_bridge *bridge, struct drm_atomic_state *old_state) { - struct analogix_dp_device *dp = bridge->driver_private; + struct analogix_dp_device *dp = to_dp(bridge); struct drm_crtc *old_crtc, *new_crtc; struct drm_crtc_state *old_crtc_state = NULL; struct drm_crtc_state *new_crtc_state = NULL; @@ -1278,7 +1278,7 @@ out: static void analogix_dp_bridge_atomic_post_disable(struct drm_bridge *bridge, struct drm_atomic_state *old_state) { - struct analogix_dp_device *dp = bridge->driver_private; + struct analogix_dp_device *dp = to_dp(bridge); struct drm_crtc *crtc; struct drm_crtc_state *new_crtc_state; int ret; @@ -1300,7 +1300,7 @@ static void analogix_dp_bridge_mode_set(struct drm_bridge *bridge, const struct drm_display_mode *orig_mode, const struct drm_display_mode *mode) { - struct analogix_dp_device *dp = bridge->driver_private; + struct analogix_dp_device *dp = to_dp(bridge); struct drm_display_info *display_info = &dp->connector.display_info; struct video_info *video = &dp->video_info; struct device_node *dp_node = dp->dev->of_node; @@ -1385,25 +1385,6 @@ static const struct drm_bridge_funcs analogix_dp_bridge_funcs = { .attach = analogix_dp_bridge_attach, }; -static int analogix_dp_create_bridge(struct drm_device *drm_dev, - struct analogix_dp_device *dp) -{ - struct drm_bridge *bridge; - - bridge = devm_kzalloc(drm_dev->dev, sizeof(*bridge), GFP_KERNEL); - if (!bridge) { - DRM_ERROR("failed to allocate for drm bridge\n"); - return -ENOMEM; - } - - dp->bridge = bridge; - - bridge->driver_private = dp; - bridge->funcs = &analogix_dp_bridge_funcs; - - return drm_bridge_attach(dp->encoder, bridge, NULL, 0); -} - static int analogix_dp_dt_parse_pdata(struct analogix_dp_device *dp) { struct device_node *dp_node = dp->dev->of_node; @@ -1491,7 +1472,8 @@ analogix_dp_probe(struct device *dev, struct analogix_dp_plat_data *plat_data) return ERR_PTR(-EINVAL); } - dp = devm_kzalloc(dev, sizeof(struct analogix_dp_device), GFP_KERNEL); + dp = devm_drm_bridge_alloc(dev, struct analogix_dp_device, bridge, + &analogix_dp_bridge_funcs); if (!dp) return ERR_PTR(-ENOMEM); @@ -1643,7 +1625,7 @@ int analogix_dp_bind(struct analogix_dp_device *dp, struct drm_device *drm_dev) return ret; } - ret = analogix_dp_create_bridge(drm_dev, dp); + ret = drm_bridge_attach(dp->encoder, &dp->bridge, NULL, 0); if (ret) { DRM_ERROR("failed to create bridge (%d)\n", ret); goto err_unregister_aux; @@ -1660,7 +1642,7 @@ EXPORT_SYMBOL_GPL(analogix_dp_bind); void analogix_dp_unbind(struct analogix_dp_device *dp) { - analogix_dp_bridge_disable(dp->bridge); + analogix_dp_bridge_disable(&dp->bridge); dp->connector.funcs->destroy(&dp->connector); drm_panel_unprepare(dp->plat_data->panel); diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h index 2b54120ba4a3..b86e93f30ed6 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h +++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h @@ -11,6 +11,7 @@ #include <drm/display/drm_dp_helper.h> #include <drm/drm_crtc.h> +#include <drm/drm_bridge.h> #define DP_TIMEOUT_LOOP_COUNT 100 #define MAX_CR_LOOP 5 @@ -154,7 +155,7 @@ struct analogix_dp_device { struct device *dev; struct drm_device *drm_dev; struct drm_connector connector; - struct drm_bridge *bridge; + struct drm_bridge bridge; struct drm_dp_aux aux; struct clk *clock; unsigned int irq; diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c index f2f666b27d2d..c4997795db18 100644 --- a/drivers/gpu/drm/bridge/samsung-dsim.c +++ b/drivers/gpu/drm/bridge/samsung-dsim.c @@ -20,6 +20,7 @@ #include <linux/of.h> #include <linux/phy/phy.h> #include <linux/platform_device.h> +#include <linux/units.h> #include <video/mipi_display.h> @@ -558,10 +559,6 @@ static void samsung_dsim_reset(struct samsung_dsim *dsi) samsung_dsim_write(dsi, DSIM_SWRST_REG, reset_val); } -#ifndef MHZ -#define MHZ (1000 * 1000) -#endif - static unsigned long samsung_dsim_pll_find_pms(struct samsung_dsim *dsi, unsigned long fin, unsigned long fout, @@ -575,8 +572,8 @@ static unsigned long samsung_dsim_pll_find_pms(struct samsung_dsim *dsi, u16 _m, best_m; u8 _s, best_s; - p_min = DIV_ROUND_UP(fin, (driver_data->pll_fin_max * MHZ)); - p_max = fin / (driver_data->pll_fin_min * MHZ); + p_min = DIV_ROUND_UP(fin, (driver_data->pll_fin_max * HZ_PER_MHZ)); + p_max = fin / (driver_data->pll_fin_min * HZ_PER_MHZ); for (_p = p_min; _p <= p_max; ++_p) { for (_s = 0; _s <= 5; ++_s) { @@ -591,8 +588,8 @@ static unsigned long samsung_dsim_pll_find_pms(struct samsung_dsim *dsi, tmp = (u64)_m * fin; do_div(tmp, _p); - if (tmp < driver_data->min_freq * MHZ || - tmp > driver_data->max_freq * MHZ) + if (tmp < driver_data->min_freq * HZ_PER_MHZ || + tmp > driver_data->max_freq * HZ_PER_MHZ) continue; tmp = (u64)_m * fin; @@ -635,7 +632,7 @@ static unsigned long samsung_dsim_set_pll(struct samsung_dsim *dsi, * limit. */ fin = clk_get_rate(clk_get_parent(dsi->pll_clk)); - while (fin > driver_data->pll_fin_max * MHZ) + while (fin > driver_data->pll_fin_max * HZ_PER_MHZ) fin /= 2; clk_set_rate(dsi->pll_clk, fin); @@ -661,10 +658,11 @@ static unsigned long samsung_dsim_set_pll(struct samsung_dsim *dsi, if (driver_data->has_freqband) { static const unsigned long freq_bands[] = { - 100 * MHZ, 120 * MHZ, 160 * MHZ, 200 * MHZ, - 270 * MHZ, 320 * MHZ, 390 * MHZ, 450 * MHZ, - 510 * MHZ, 560 * MHZ, 640 * MHZ, 690 * MHZ, - 770 * MHZ, 870 * MHZ, 950 * MHZ, + 100 * HZ_PER_MHZ, 120 * HZ_PER_MHZ, 160 * HZ_PER_MHZ, + 200 * HZ_PER_MHZ, 270 * HZ_PER_MHZ, 320 * HZ_PER_MHZ, + 390 * HZ_PER_MHZ, 450 * HZ_PER_MHZ, 510 * HZ_PER_MHZ, + 560 * HZ_PER_MHZ, 640 * HZ_PER_MHZ, 690 * HZ_PER_MHZ, + 770 * HZ_PER_MHZ, 870 * HZ_PER_MHZ, 950 * HZ_PER_MHZ, }; int band; @@ -724,7 +722,7 @@ static int samsung_dsim_enable_clock(struct samsung_dsim *dsi) esc_div = DIV_ROUND_UP(byte_clk, dsi->esc_clk_rate); esc_clk = byte_clk / esc_div; - if (esc_clk > 20 * MHZ) { + if (esc_clk > 20 * HZ_PER_MHZ) { ++esc_div; esc_clk = byte_clk / esc_div; } @@ -899,8 +897,6 @@ static int samsung_dsim_init_link(struct samsung_dsim *dsi) * The user manual describes that following bits are ignored in * command mode. */ - if (!(dsi->mode_flags & MIPI_DSI_MODE_VSYNC_FLUSH)) - reg |= DSIM_MFLUSH_VS; if (dsi->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE) reg |= DSIM_SYNC_INFORM; if (dsi->mode_flags & MIPI_DSI_MODE_VIDEO_BURST) @@ -1236,43 +1232,34 @@ static void samsung_dsim_transfer_start(struct samsung_dsim *dsi) { unsigned long flags; struct samsung_dsim_transfer *xfer; - bool start = false; -again: spin_lock_irqsave(&dsi->transfer_lock, flags); - if (list_empty(&dsi->transfer_list)) { - spin_unlock_irqrestore(&dsi->transfer_lock, flags); - return; - } + while (!list_empty(&dsi->transfer_list)) { + xfer = list_first_entry(&dsi->transfer_list, + struct samsung_dsim_transfer, list); - xfer = list_first_entry(&dsi->transfer_list, - struct samsung_dsim_transfer, list); - - spin_unlock_irqrestore(&dsi->transfer_lock, flags); + spin_unlock_irqrestore(&dsi->transfer_lock, flags); - if (xfer->packet.payload_length && - xfer->tx_done == xfer->packet.payload_length) - /* waiting for RX */ - return; + if (xfer->packet.payload_length && + xfer->tx_done == xfer->packet.payload_length) + /* waiting for RX */ + return; - samsung_dsim_send_to_fifo(dsi, xfer); + samsung_dsim_send_to_fifo(dsi, xfer); - if (xfer->packet.payload_length || xfer->rx_len) - return; + if (xfer->packet.payload_length || xfer->rx_len) + return; - xfer->result = 0; - complete(&xfer->completed); + xfer->result = 0; + complete(&xfer->completed); - spin_lock_irqsave(&dsi->transfer_lock, flags); + spin_lock_irqsave(&dsi->transfer_lock, flags); - list_del_init(&xfer->list); - start = !list_empty(&dsi->transfer_list); + list_del_init(&xfer->list); + } spin_unlock_irqrestore(&dsi->transfer_lock, flags); - - if (start) - goto again; } static bool samsung_dsim_transfer_finish(struct samsung_dsim *dsi) diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c index 7e5449fb86a3..562fea47b3ec 100644 --- a/drivers/gpu/drm/bridge/tc358767.c +++ b/drivers/gpu/drm/bridge/tc358767.c @@ -344,6 +344,14 @@ #define COLOR_BAR_MODE_BARS 2 #define PLL_DBG 0x0a04 +enum tc_mode { + mode_dpi_to_edp = BIT(1) | BIT(2), + mode_dpi_to_dp = BIT(1), + mode_dsi_to_edp = BIT(0) | BIT(2), + mode_dsi_to_dp = BIT(0), + mode_dsi_to_dpi = BIT(0) | BIT(1), +}; + static bool tc_test_pattern; module_param_named(test, tc_test_pattern, bool, 0644); @@ -2327,7 +2335,6 @@ static int tc_probe_dpi_bridge_endpoint(struct tc_data *tc) if (bridge) { tc->panel_bridge = bridge; tc->bridge.type = DRM_MODE_CONNECTOR_DPI; - tc->bridge.funcs = &tc_dpi_bridge_funcs; return 0; } @@ -2360,7 +2367,6 @@ static int tc_probe_edp_bridge_endpoint(struct tc_data *tc) tc->bridge.type = DRM_MODE_CONNECTOR_DisplayPort; } - tc->bridge.funcs = &tc_edp_bridge_funcs; if (tc->hpd_pin >= 0) tc->bridge.ops |= DRM_BRIDGE_OP_DETECT; tc->bridge.ops |= DRM_BRIDGE_OP_EDID; @@ -2368,17 +2374,11 @@ static int tc_probe_edp_bridge_endpoint(struct tc_data *tc) return 0; } -static int tc_probe_bridge_endpoint(struct tc_data *tc) +static enum tc_mode tc_probe_get_mode(struct device *dev) { - struct device *dev = tc->dev; struct of_endpoint endpoint; struct device_node *node = NULL; - const u8 mode_dpi_to_edp = BIT(1) | BIT(2); - const u8 mode_dpi_to_dp = BIT(1); - const u8 mode_dsi_to_edp = BIT(0) | BIT(2); - const u8 mode_dsi_to_dp = BIT(0); - const u8 mode_dsi_to_dpi = BIT(0) | BIT(1); - u8 mode = 0; + enum tc_mode mode = 0; /* * Determine bridge configuration. @@ -2401,7 +2401,28 @@ static int tc_probe_bridge_endpoint(struct tc_data *tc) return -EINVAL; } mode |= BIT(endpoint.port); + } + + if (mode != mode_dpi_to_edp && + mode != mode_dpi_to_dp && + mode != mode_dsi_to_dpi && + mode != mode_dsi_to_edp && + mode != mode_dsi_to_dp) { + dev_warn(dev, "Invalid mode (0x%x) is not supported!\n", mode); + return -EINVAL; + } + + return mode; +} +static int tc_probe_bridge_endpoint(struct tc_data *tc, enum tc_mode mode) +{ + struct device *dev = tc->dev; + struct of_endpoint endpoint; + struct device_node *node = NULL; + + for_each_endpoint_of_node(dev->of_node, node) { + of_graph_parse_endpoint(node, &endpoint); if (endpoint.port == 2) { of_property_read_u8_array(node, "toshiba,pre-emphasis", tc->pre_emphasis, @@ -2427,24 +2448,28 @@ static int tc_probe_bridge_endpoint(struct tc_data *tc) return tc_probe_edp_bridge_endpoint(tc); } - dev_warn(dev, "Invalid mode (0x%x) is not supported!\n", mode); - + /* Should never happen, mode was validated by tc_probe_get_mode() */ return -EINVAL; } static int tc_probe(struct i2c_client *client) { struct device *dev = &client->dev; + const struct drm_bridge_funcs *funcs; struct tc_data *tc; + int mode; int ret; - tc = devm_kzalloc(dev, sizeof(*tc), GFP_KERNEL); - if (!tc) - return -ENOMEM; + mode = tc_probe_get_mode(dev); + funcs = (mode == mode_dsi_to_dpi) ? &tc_dpi_bridge_funcs : &tc_edp_bridge_funcs; + + tc = devm_drm_bridge_alloc(dev, struct tc_data, bridge, funcs); + if (IS_ERR(tc)) + return PTR_ERR(tc); tc->dev = dev; - ret = tc_probe_bridge_endpoint(tc); + ret = tc_probe_bridge_endpoint(tc, mode); if (ret) return ret; diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c index 78a50b947a08..3d0b4bc5129d 100644 --- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c +++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c @@ -1677,11 +1677,6 @@ static int ti_sn_bridge_gpio_set(struct gpio_chip *chip, unsigned int offset, { struct ti_sn65dsi86 *pdata = gpiochip_get_data(chip); - if (!test_bit(offset, pdata->gchip_output)) { - dev_err(pdata->dev, "Ignoring GPIO set while input\n"); - return -EPERM; - } - val &= 1; return regmap_update_bits(pdata->regmap, SN_GPIO_IO_REG, BIT(SN_GPIO_OUTPUT_SHIFT + offset), diff --git a/drivers/gpu/drm/ci/build-igt.sh b/drivers/gpu/drm/ci/build-igt.sh index caa2f4804ed5..eddb5f782a5e 100644 --- a/drivers/gpu/drm/ci/build-igt.sh +++ b/drivers/gpu/drm/ci/build-igt.sh @@ -71,4 +71,4 @@ tar -cf artifacts/igt.tar /igt # Pass needed files to the test stage S3_ARTIFACT_NAME="igt.tar.gz" gzip -c artifacts/igt.tar > ${S3_ARTIFACT_NAME} -s3_upload ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${KERNEL_ARCH}/ +ci-fairy s3cp --token-file "${S3_JWT_FILE}" ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${KERNEL_ARCH}/${S3_ARTIFACT_NAME} diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh index 6fb74c51abe2..ac5e7ed195cf 100644 --- a/drivers/gpu/drm/ci/build.sh +++ b/drivers/gpu/drm/ci/build.sh @@ -113,17 +113,6 @@ mkdir -p install/modules/ INSTALL_MOD_PATH=install/modules/ make modules_install if [[ ${DEBIAN_ARCH} = "arm64" ]]; then - make Image.lzma - mkimage \ - -f auto \ - -A arm \ - -O linux \ - -d arch/arm64/boot/Image.lzma \ - -C lzma\ - -b arch/arm64/boot/dts/qcom/sdm845-cheza-r3.dtb \ - /kernel/cheza-kernel - KERNEL_IMAGE_NAME+=" cheza-kernel" - # Make a gzipped copy of the Image for db410c. gzip -k /kernel/Image KERNEL_IMAGE_NAME+=" Image.gz" @@ -148,13 +137,13 @@ if [[ "$UPLOAD_TO_MINIO" = "1" ]]; then ls -l "${S3_JWT_FILE}" for f in $FILES_TO_UPLOAD; do - s3_upload /kernel/$f \ - https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/ + ci-fairy s3cp --token-file "${S3_JWT_FILE}" /kernel/$f \ + https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/$f done S3_ARTIFACT_NAME="kernel-files.tar.zst" tar --zstd -cf $S3_ARTIFACT_NAME install - s3_upload ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/ + ci-fairy s3cp --token-file "${S3_JWT_FILE}" ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/${S3_ARTIFACT_NAME} echo "Download vmlinux.xz from https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/vmlinux.xz" fi diff --git a/drivers/gpu/drm/ci/build.yml b/drivers/gpu/drm/ci/build.yml index 8eb56ebcf4aa..af27ff5de369 100644 --- a/drivers/gpu/drm/ci/build.yml +++ b/drivers/gpu/drm/ci/build.yml @@ -13,7 +13,7 @@ - .build - .use-debian/arm64_build tags: - - aarch64 + - $FDO_RUNNER_JOB_PRIORITY_TAG_AARCH64 variables: DEFCONFIG: "arch/arm/configs/multi_v7_defconfig" KERNEL_IMAGE_NAME: "zImage" @@ -24,7 +24,7 @@ - .build - .use-debian/arm64_build tags: - - aarch64 + - $FDO_RUNNER_JOB_PRIORITY_TAG_AARCH64 variables: DEFCONFIG: "arch/arm64/configs/defconfig" KERNEL_IMAGE_NAME: "Image" @@ -44,16 +44,22 @@ igt:arm32: extends: .build:arm32 + variables: + GIT_DEPTH: 10 script: - FDO_CI_CONCURRENT=${FDO_CI_CONCURRENT} bash drivers/gpu/drm/ci/build-igt.sh igt:arm64: extends: .build:arm64 + variables: + GIT_DEPTH: 10 script: - FDO_CI_CONCURRENT=${FDO_CI_CONCURRENT} bash drivers/gpu/drm/ci/build-igt.sh igt:x86_64: extends: .build:x86_64 + variables: + GIT_DEPTH: 10 script: - FDO_CI_CONCURRENT=${FDO_CI_CONCURRENT} bash drivers/gpu/drm/ci/build-igt.sh diff --git a/drivers/gpu/drm/ci/check-devicetrees.yml b/drivers/gpu/drm/ci/check-devicetrees.yml new file mode 100644 index 000000000000..727bd56018b8 --- /dev/null +++ b/drivers/gpu/drm/ci/check-devicetrees.yml @@ -0,0 +1,50 @@ +.dt-check-base: + stage: static-checks + timeout: "30m" + variables: + GIT_DEPTH: 1 + FF_USE_NEW_BASH_EVAL_STRATEGY: 'true' + SCHEMA: "display:gpu" + VENV_PATH: "/tmp/dtcheck-venv" + before_script: + - apt-get update -qq + # Minimum supported version of LLVM for building x86 kernels is 15.0.0. + # In mesa-ci containers, LLVM_VERSION is defined as a container-level property and is currently set to 19. + - apt-get install -y --no-install-recommends clang-${LLVM_VERSION} lld-${LLVM_VERSION} llvm-${LLVM_VERSION} python3-dev python3-venv python3-pip yamllint + - python3 -m venv "${VENV_PATH}" + - source "${VENV_PATH}/bin/activate" + - pip3 install dtschema + script: + - drivers/gpu/drm/ci/${SCRIPT_NAME} + artifacts: + when: on_failure + paths: + - ${ARTIFACT_FILE} + allow_failure: + exit_codes: + - 102 + +dtbs-check:arm32: + extends: + - .build:arm32 + - .dt-check-base + variables: + SCRIPT_NAME: "dtbs-check.sh" + ARTIFACT_FILE: "dtbs-check.log" + +dtbs-check:arm64: + extends: + - .build:arm64 + - .dt-check-base + variables: + SCRIPT_NAME: "dtbs-check.sh" + ARTIFACT_FILE: "dtbs-check.log" + +dt-binding-check: + extends: + - .build + - .use-debian/x86_64_build + - .dt-check-base + variables: + SCRIPT_NAME: "dt-binding-check.sh" + ARTIFACT_FILE: "dt-binding-check.log" diff --git a/drivers/gpu/drm/ci/container.yml b/drivers/gpu/drm/ci/container.yml index 56c95c2f91ae..5f90508578a3 100644 --- a/drivers/gpu/drm/ci/container.yml +++ b/drivers/gpu/drm/ci/container.yml @@ -20,31 +20,15 @@ debian/arm64_build: EXTRA_LOCAL_PACKAGES: "jq libasound2 libcairo2 libdw1 libglib2.0-0 libjson-c5" # Disable container jobs that we won't use -alpine/x86_64_build: - rules: - - when: never - -debian/arm32_test-base: - rules: - - when: never - -debian/arm32_test-gl: - rules: - - when: never - -debian/arm32_test-vk: - rules: - - when: never - -debian/arm64_test-gl: +debian/arm64_test-vk: rules: - when: never -debian/arm64_test-vk: +debian/baremetal_arm32_test-gl: rules: - when: never -debian/baremetal_arm32_test: +debian/baremetal_arm64_test-vk: rules: - when: never @@ -64,19 +48,19 @@ debian/x86_64_test-android: rules: - when: never -debian/x86_64_test-vk: +debian/x86_64_test-video: rules: - when: never -fedora/x86_64_build: +debian/x86_64_test-vk: rules: - when: never -debian/android_build: +fedora/x86_64_build: rules: - when: never -.debian/x86_64_test-android: +debian/android_build: rules: - when: never diff --git a/drivers/gpu/drm/ci/dt-binding-check.sh b/drivers/gpu/drm/ci/dt-binding-check.sh new file mode 100755 index 000000000000..99e1c0df84b7 --- /dev/null +++ b/drivers/gpu/drm/ci/dt-binding-check.sh @@ -0,0 +1,19 @@ +#!/bin/bash +# SPDX-License-Identifier: MIT + +set -euxo pipefail + +VENV_PATH="${VENV_PATH:-/tmp/dtschema-venv}" +source "${VENV_PATH}/bin/activate" + +if ! make -j"${FDO_CI_CONCURRENT:-4}" dt_binding_check \ + DT_SCHEMA_FILES="${SCHEMA:-}" 2>dt-binding-check.log; then + echo "ERROR: 'make dt_binding_check' failed. Please check dt-binding-check.log for details." + exit 1 +fi + +if [[ -s dt-binding-check.log ]]; then + echo "WARNING: dt_binding_check reported warnings. Please check dt-binding-check.log" \ + "for details." + exit 102 +fi diff --git a/drivers/gpu/drm/ci/dtbs-check.sh b/drivers/gpu/drm/ci/dtbs-check.sh new file mode 100755 index 000000000000..57842c452439 --- /dev/null +++ b/drivers/gpu/drm/ci/dtbs-check.sh @@ -0,0 +1,22 @@ +#!/bin/bash +# SPDX-License-Identifier: MIT + +set -euxo pipefail + +: "${KERNEL_ARCH:?ERROR: KERNEL_ARCH must be set}" +: "${LLVM_VERSION:?ERROR: LLVM_VERSION must be set}" + +./drivers/gpu/drm/ci/setup-llvm-links.sh + +make LLVM=1 ARCH="${KERNEL_ARCH}" defconfig + +if ! make -j"${FDO_CI_CONCURRENT:-4}" ARCH="${KERNEL_ARCH}" LLVM=1 dtbs_check \ + DT_SCHEMA_FILES="${SCHEMA:-}" 2>dtbs-check.log; then + echo "ERROR: 'make dtbs_check' failed. Please check dtbs-check.log for details." + exit 1 +fi + +if [[ -s dtbs-check.log ]]; then + echo "WARNING: dtbs_check reported warnings. Please check dtbs-check.log for details." + exit 102 +fi diff --git a/drivers/gpu/drm/ci/gitlab-ci.yml b/drivers/gpu/drm/ci/gitlab-ci.yml index ba75b3a7eca4..d502d146b177 100644 --- a/drivers/gpu/drm/ci/gitlab-ci.yml +++ b/drivers/gpu/drm/ci/gitlab-ci.yml @@ -1,17 +1,17 @@ variables: DRM_CI_PROJECT_PATH: &drm-ci-project-path mesa/mesa - DRM_CI_COMMIT_SHA: &drm-ci-commit-sha f73132f1215a37ce8ffc711a0136c90649aaf128 + DRM_CI_COMMIT_SHA: &drm-ci-commit-sha 02337aec715c25dae7ff2479d986f831c77fe536 UPSTREAM_REPO: https://gitlab.freedesktop.org/drm/kernel.git TARGET_BRANCH: drm-next - IGT_VERSION: 04bedb9238586b81d4d4ca62b02e584f6cfc77af + IGT_VERSION: 129d5b10baaadea1d6cd6377341c4cb42e7ee6fd DEQP_RUNNER_GIT_URL: https://gitlab.freedesktop.org/mesa/deqp-runner.git DEQP_RUNNER_GIT_TAG: v0.20.0 FDO_UPSTREAM_REPO: helen.fornazier/linux # The repo where the git-archive daily runs - MESA_TEMPLATES_COMMIT: &ci-templates-commit d5aa3941aa03c2f716595116354fb81eb8012acb + MESA_TEMPLATES_COMMIT: &ci-templates-commit c6aeb16f86e32525fa630fb99c66c4f3e62fc3cb DRM_CI_PROJECT_URL: https://gitlab.freedesktop.org/${DRM_CI_PROJECT_PATH} CI_PRE_CLONE_SCRIPT: |- set -o xtrace @@ -20,10 +20,8 @@ variables: rm download-git-cache.sh set +o xtrace S3_JWT_FILE: /s3_jwt - S3_JWT_HEADER_FILE: /s3_jwt_header S3_JWT_FILE_SCRIPT: |- echo -n '${S3_JWT}' > '${S3_JWT_FILE}' && - echo -n "Authorization: Bearer ${S3_JWT}" > '${S3_JWT_HEADER_FILE}' && unset CI_JOB_JWT S3_JWT # Unsetting vulnerable env variables S3_HOST: s3.freedesktop.org # This bucket is used to fetch the kernel image @@ -38,7 +36,11 @@ variables: JOB_ARTIFACTS_BASE: ${PIPELINE_ARTIFACTS_BASE}/${CI_JOB_ID} KERNEL_IMAGE_BASE: https://${S3_HOST}/${S3_KERNEL_BUCKET}/${KERNEL_REPO}/${KERNEL_TAG} LAVA_TAGS: subset-1-gfx - LAVA_JOB_PRIORITY: 30 + # Default priority for non-merge pipelines + FDO_RUNNER_JOB_PRIORITY_TAG_X86_64: "" # Empty tags are ignored by gitlab + FDO_RUNNER_JOB_PRIORITY_TAG_X86_64_KVM: kvm + FDO_RUNNER_JOB_PRIORITY_TAG_AARCH64: aarch64 + JOB_PRIORITY: 30 ARTIFACTS_BASE_URL: https://${CI_PROJECT_ROOT_NAMESPACE}.${CI_PAGES_DOMAIN}/-/${CI_PROJECT_NAME}/-/jobs/${CI_JOB_ID}/artifacts # Python scripts for structured logger PYTHONPATH: "$PYTHONPATH:$CI_PROJECT_DIR/install" @@ -73,15 +75,12 @@ default: include: - project: 'freedesktop/ci-templates' - ref: 16bc29078de5e0a067ff84a1a199a3760d3b3811 - file: - - '/templates/ci-fairy.yml' - - project: 'freedesktop/ci-templates' ref: *ci-templates-commit file: - '/templates/alpine.yml' - '/templates/debian.yml' - '/templates/fedora.yml' + - '/templates/ci-fairy.yml' - project: *drm-ci-project-path ref: *drm-ci-commit-sha file: @@ -105,20 +104,26 @@ include: - '/src/microsoft/ci/gitlab-ci-inc.yml' - '/src/nouveau/ci/gitlab-ci-inc.yml' - '/src/virtio/ci/gitlab-ci-inc.yml' + - 'docs/gitlab-ci.yml' - drivers/gpu/drm/ci/image-tags.yml - drivers/gpu/drm/ci/container.yml - drivers/gpu/drm/ci/static-checks.yml - drivers/gpu/drm/ci/build.yml - drivers/gpu/drm/ci/test.yml + - drivers/gpu/drm/ci/check-devicetrees.yml + - drivers/gpu/drm/ci/kunit.yml - 'https://gitlab.freedesktop.org/gfx-ci/lab-status/-/raw/main/lab-status.yml' stages: - sanity - container + - deploy - git-archive - build-for-tests - build-only + - static-checks + - kunit - code-validation - amdgpu - i915 @@ -232,16 +237,20 @@ stages: - _build/meson-logs/strace +python-artifacts: + variables: + GIT_DEPTH: 10 + + # Git archive -make git archive: +make-git-archive: extends: - .fdo.ci-fairy stage: git-archive rules: - !reference [.scheduled_pipeline-rules, rules] - # ensure we are running on packet tags: - - packet.net + - $FDO_RUNNER_JOB_PRIORITY_TAG_X86_64 script: # Remove drm-ci files we just added - rm -rf .gitlab-ci.* @@ -253,7 +262,7 @@ make git archive: - tar -cvzf ../$CI_PROJECT_NAME.tar.gz . # Use id_tokens for JWT auth - - s3_upload ../$CI_PROJECT_NAME.tar.gz https://$S3_HOST/${S3_GITCACHE_BUCKET}/$CI_PROJECT_NAMESPACE/$CI_PROJECT_NAME/ + - ci-fairy s3cp --token-file "${S3_JWT_FILE}" ../$CI_PROJECT_NAME.tar.gz https://$S3_HOST/${S3_GITCACHE_BUCKET}/$CI_PROJECT_NAMESPACE/$CI_PROJECT_NAME/$CI_PROJECT_NAME.tar.gz # Sanity checks of MR settings and commit logs @@ -261,6 +270,8 @@ sanity: extends: - .fdo.ci-fairy stage: sanity + tags: + - $FDO_RUNNER_JOB_PRIORITY_TAG_X86_64 rules: - if: *is-pre-merge when: on_success @@ -279,7 +290,6 @@ sanity: DEBIAN_BUILD_TAG DEBIAN_PYUTILS_TAG DEBIAN_TEST_GL_TAG - KERNEL_ROOTFS_TAG KERNEL_TAG PKG_REPO_REV ) @@ -295,14 +305,14 @@ sanity: when: on_failure reports: junit: check-*.xml - tags: - - placeholder-job mr-label-maker-test: extends: - .fdo.ci-fairy stage: sanity + tags: + - $FDO_RUNNER_JOB_PRIORITY_TAG_X86_64 rules: - !reference [.mr-label-maker-rules, rules] variables: @@ -325,3 +335,15 @@ mr-label-maker-test: optional: true - job: toml-lint optional: true + +deploy-docs: + rules: + - when: never + +linkcheck-docs: + rules: + - when: never + +test-docs: + rules: + - when: never diff --git a/drivers/gpu/drm/ci/igt_runner.sh b/drivers/gpu/drm/ci/igt_runner.sh index 2a0599f12c58..b24d4bc53cda 100755 --- a/drivers/gpu/drm/ci/igt_runner.sh +++ b/drivers/gpu/drm/ci/igt_runner.sh @@ -19,6 +19,7 @@ set +e cat /sys/kernel/debug/dri/*/state set -e +mkdir -p /lib/modules case "$DRIVER_NAME" in amdgpu|vkms) # Cannot use HWCI_KERNEL_MODULES as at that point we don't have the module in /lib diff --git a/drivers/gpu/drm/ci/image-tags.yml b/drivers/gpu/drm/ci/image-tags.yml index 53fe34b86578..7acc2e2a8eaa 100644 --- a/drivers/gpu/drm/ci/image-tags.yml +++ b/drivers/gpu/drm/ci/image-tags.yml @@ -1,24 +1,18 @@ variables: - CONTAINER_TAG: "20250328-mesa-uprev" - DEBIAN_X86_64_BUILD_BASE_IMAGE: "debian/x86_64_build-base" + CONTAINER_TAG: "20250502-mesa-uprev" DEBIAN_BASE_TAG: "${CONTAINER_TAG}" - - DEBIAN_X86_64_BUILD_IMAGE_PATH: "debian/x86_64_build" DEBIAN_BUILD_TAG: "${CONTAINER_TAG}" - KERNEL_ROOTFS_TAG: "${CONTAINER_TAG}" + DEBIAN_TEST_GL_TAG: "${CONTAINER_TAG}" # default kernel for rootfs before injecting the current kernel tree - KERNEL_TAG: "v6.13-rc4-mesa-5e77" + KERNEL_TAG: "v6.14-mesa-0bdd" KERNEL_REPO: "gfx-ci/linux" - PKG_REPO_REV: "bca9635d" - - DEBIAN_X86_64_TEST_BASE_IMAGE: "debian/x86_64_test-base" - DEBIAN_X86_64_TEST_IMAGE_GL_PATH: "debian/x86_64_test-gl" - DEBIAN_TEST_GL_TAG: "${CONTAINER_TAG}" + PKG_REPO_REV: "95bf62c" - DEBIAN_PYUTILS_IMAGE: "debian/x86_64_pyutils" DEBIAN_PYUTILS_TAG: "${CONTAINER_TAG}" + ALPINE_X86_64_BUILD_TAG: "${CONTAINER_TAG}" ALPINE_X86_64_LAVA_SSH_TAG: "${CONTAINER_TAG}" - CONDITIONAL_BUILD_ANGLE_TAG: fec96cc945650c5fe9f7188cabe80d8a + CONDITIONAL_BUILD_ANGLE_TAG: 384145a4023315dae658259bee07c43a + CONDITIONAL_BUILD_PIGLIT_TAG: a19e424b8a3f020dbf1b9dd29f220a4f diff --git a/drivers/gpu/drm/ci/kunit.sh b/drivers/gpu/drm/ci/kunit.sh new file mode 100755 index 000000000000..7a1052fd3f17 --- /dev/null +++ b/drivers/gpu/drm/ci/kunit.sh @@ -0,0 +1,16 @@ +#!/bin/bash +# SPDX-License-Identifier: MIT + +set -euxo pipefail + +: "${KERNEL_ARCH:?ERROR: KERNEL_ARCH must be set}" +: "${LLVM_VERSION:?ERROR: LLVM_VERSION must be set}" + +./drivers/gpu/drm/ci/setup-llvm-links.sh + +export PATH="/usr/bin:$PATH" + +./tools/testing/kunit/kunit.py run \ + --arch "${KERNEL_ARCH}" \ + --make_options LLVM=1 \ + --kunitconfig=drivers/gpu/drm/tests diff --git a/drivers/gpu/drm/ci/kunit.yml b/drivers/gpu/drm/ci/kunit.yml new file mode 100644 index 000000000000..0d5b2c4433d2 --- /dev/null +++ b/drivers/gpu/drm/ci/kunit.yml @@ -0,0 +1,37 @@ +.kunit-packages: &kunit-packages + - apt-get update -qq + # Minimum supported version of LLVM for building x86 kernels is 15.0.0. + # In mesa-ci containers, LLVM_VERSION is defined as a container-level property and is currently set to 19. + - apt-get install -y --no-install-recommends clang-${LLVM_VERSION} lld-${LLVM_VERSION} llvm-${LLVM_VERSION} + +.kunit-base: + stage: kunit + timeout: "30m" + variables: + GIT_DEPTH: 1 + script: + - drivers/gpu/drm/ci/kunit.sh + +kunit:arm32: + extends: + - .build:arm32 + - .kunit-base + before_script: + - *kunit-packages + - apt-get install -y --no-install-recommends qemu-system-arm + +kunit:arm64: + extends: + - .build:arm64 + - .kunit-base + before_script: + - *kunit-packages + - apt-get install -y --no-install-recommends qemu-system-aarch64 + +kunit:x86_64: + extends: + - .build:x86_64 + - .kunit-base + before_script: + - *kunit-packages + - apt-get install -y --no-install-recommends qemu-system-x86 diff --git a/drivers/gpu/drm/ci/lava-submit.sh b/drivers/gpu/drm/ci/lava-submit.sh index a1e8b34fb2d4..a295102c3468 100755 --- a/drivers/gpu/drm/ci/lava-submit.sh +++ b/drivers/gpu/drm/ci/lava-submit.sh @@ -41,7 +41,6 @@ section_start prepare_rootfs "Preparing root filesystem" set -ex -section_switch rootfs "Assembling root filesystem" ROOTFS_URL="$(get_path_to_artifact lava-rootfs.tar.zst)" [ $? != 1 ] || exit 1 @@ -54,7 +53,7 @@ cp artifacts/ci-common/init-*.sh results/job-rootfs-overlay/ cp "$SCRIPTS_DIR"/setup-test-env.sh results/job-rootfs-overlay/ tar zcf job-rootfs-overlay.tar.gz -C results/job-rootfs-overlay/ . -s3_upload job-rootfs-overlay.tar.gz "https://${JOB_ARTIFACTS_BASE}" +ci-fairy s3cp --token-file "${S3_JWT_FILE}" job-rootfs-overlay.tar.gz "https://${JOB_ROOTFS_OVERLAY_PATH}" # Prepare env vars for upload. section_switch variables "Environment variables passed through to device:" @@ -64,6 +63,9 @@ section_switch lava_submit "Submitting job for scheduling" touch results/lava.log tail -f results/lava.log & +# Ensure that we are printing the commands that are being executed, +# making it easier to debug the job in case it fails. +set -x PYTHONPATH=artifacts/ artifacts/lava/lava_job_submitter.py \ --farm "${FARM}" \ --device-type "${DEVICE_TYPE}" \ diff --git a/drivers/gpu/drm/ci/setup-llvm-links.sh b/drivers/gpu/drm/ci/setup-llvm-links.sh new file mode 100755 index 000000000000..ace33af82a3f --- /dev/null +++ b/drivers/gpu/drm/ci/setup-llvm-links.sh @@ -0,0 +1,13 @@ +#!/usr/bin/env bash +# SPDX-License-Identifier: MIT +set -euo pipefail + +ln -svf "$(which clang++-${LLVM_VERSION})" /usr/bin/clang++ +ln -svf "$(which clang-${LLVM_VERSION})" /usr/bin/clang +ln -svf "$(which ld.lld-${LLVM_VERSION})" /usr/bin/ld.lld +ln -svf "$(which lld-${LLVM_VERSION})" /usr/bin/lld +ln -svf "$(which llvm-ar-${LLVM_VERSION})" /usr/bin/llvm-ar +ln -svf "$(which llvm-nm-${LLVM_VERSION})" /usr/bin/llvm-nm +ln -svf "$(which llvm-objcopy-${LLVM_VERSION})" /usr/bin/llvm-objcopy +ln -svf "$(which llvm-readelf-${LLVM_VERSION})" /usr/bin/llvm-readelf +ln -svf "$(which llvm-strip-${LLVM_VERSION})" /usr/bin/llvm-strip diff --git a/drivers/gpu/drm/ci/test.yml b/drivers/gpu/drm/ci/test.yml index 84a25f0e783b..81147e86bfd0 100644 --- a/drivers/gpu/drm/ci/test.yml +++ b/drivers/gpu/drm/ci/test.yml @@ -24,7 +24,7 @@ .lava-igt:arm32: extends: - - .lava-test:arm32 + - .lava-arm32-test-gl variables: HWCI_TEST_SCRIPT: "/install/igt_runner.sh" DEBIAN_ARCH: "armhf" @@ -33,15 +33,14 @@ - testing:arm32 needs: - alpine/x86_64_lava_ssh_client - - kernel+rootfs_arm32 - - debian/x86_64_build + - debian/arm32_test-gl - python-artifacts - testing:arm32 - igt:arm32 .lava-igt:arm64: extends: - - .lava-test:arm64 + - .lava-arm64-test-gl variables: HWCI_TEST_SCRIPT: "/install/igt_runner.sh" DEBIAN_ARCH: "arm64" @@ -50,15 +49,14 @@ - testing:arm64 needs: - alpine/x86_64_lava_ssh_client - - kernel+rootfs_arm64 - - debian/x86_64_build + - debian/arm64_test-gl - python-artifacts - testing:arm64 - igt:arm64 .lava-igt:x86_64: extends: - - .lava-test:x86_64 + - .lava-x86_64-test-gl variables: HWCI_TEST_SCRIPT: "/install/igt_runner.sh" DEBIAN_ARCH: "amd64" @@ -67,16 +65,15 @@ - testing:x86_64 needs: - alpine/x86_64_lava_ssh_client - - kernel+rootfs_x86_64 - - debian/x86_64_build + - debian/x86_64_test-gl - python-artifacts - testing:x86_64 - igt:x86_64 .baremetal-igt-arm64: extends: - - .baremetal-test-arm64 - - .use-debian/baremetal_arm64_test + - .baremetal-test-arm64-gl + - .use-debian/baremetal_arm64_test-gl - .allow_failure_lockdep timeout: "1h30m" rules: @@ -91,7 +88,7 @@ BM_CMDLINE: "ip=dhcp console=ttyMSM0,115200n8 $BM_KERNEL_EXTRA_ARGS root=/dev/nfs rw nfsrootdebug nfsroot=,tcp,nfsvers=4.2 init=/init $BM_KERNELARGS" FARM: google needs: - - debian/baremetal_arm64_test + - debian/baremetal_arm64_test-gl - job: testing:arm64 artifacts: false - igt:arm64 @@ -101,19 +98,21 @@ .software-driver: stage: software-driver extends: + - .test-gl - .allow_failure_lockdep timeout: "1h30m" rules: - !reference [.scheduled_pipeline-rules, rules] - when: on_success - extends: - - .test-gl tags: - - kvm + - $FDO_RUNNER_JOB_PRIORITY_TAG_X86_64_KVM + before_script: + - !reference [default, before_script] + - rm -rf install + - tar -xf artifacts/install.tar script: - ln -sf $CI_PROJECT_DIR/install /install - mv install/bzImage /kernel/bzImage - - mkdir -p /lib/modules - install/crosvm-runner.sh install/igt_runner.sh needs: - debian/x86_64_test-gl @@ -180,20 +179,6 @@ msm:apq8096: script: - ./install/bare-metal/fastboot.sh || exit $? -msm:sdm845: - extends: - - .baremetal-igt-arm64 - stage: msm - parallel: 6 - variables: - DEVICE_TYPE: sdm845-cheza-r3 - DRIVER_NAME: msm - BM_KERNEL: https://${PIPELINE_ARTIFACTS_BASE}/arm64/cheza-kernel - GPU_VERSION: sdm845 - RUNNER_TAG: google-freedreno-cheza - script: - - ./install/bare-metal/cros-servo.sh || exit $? - msm:sm8350-hdk: extends: - .lava-igt:arm64 @@ -323,7 +308,7 @@ i915:cml: variables: DEVICE_TYPE: asus-C436FA-Flip-hatch GPU_VERSION: cml - RUNNER_TAG: mesa-ci-x86-64-lava-asus-C436FA-flip-hatch + RUNNER_TAG: mesa-ci-x86-64-lava-asus-C436FA-Flip-hatch i915:tgl: extends: diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt index d4b8ba3a54a9..154b047787b2 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt @@ -32,3 +32,8 @@ kms_display_modes@mst-extended-mode-negative # It causes other tests to fail, so skip it. kms_invalid_mode@overflow-vrefresh + +# sc7180 does not have APRIV, so memptrs is not protected. +# (Preemption is not supported on devices that do not have +# APRIV, so this is ok) +msm/msm_mapping@memptrs diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt index 022db559cc7d..a9bb3e1ad75c 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt @@ -23,3 +23,8 @@ core_hotunplug.* # https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 kms_display_modes@extended-mode-basic kms_display_modes@mst-extended-mode-negative + +# sc7180 does not have APRIV, so memptrs is not protected. +# (Preemption is not supported on devices that do not have +# APRIV, so this is ok) +msm/msm_mapping@memptrs diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt deleted file mode 100644 index 7a2ab58b706f..000000000000 --- a/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt +++ /dev/null @@ -1,29 +0,0 @@ -kms_color@ctm-0-25,Fail -kms_color@ctm-0-50,Fail -kms_color@ctm-0-75,Fail -kms_color@ctm-blue-to-red,Fail -kms_color@ctm-green-to-red,Fail -kms_color@ctm-negative,Fail -kms_color@ctm-red-to-blue,Fail -kms_color@ctm-signed,Fail -kms_cursor_legacy@basic-flip-before-cursor-atomic,Fail -kms_cursor_legacy@basic-flip-before-cursor-legacy,Fail -kms_cursor_legacy@cursor-vs-flip-atomic,Fail -kms_cursor_legacy@cursor-vs-flip-atomic-transitions,Fail -kms_cursor_legacy@cursor-vs-flip-atomic-transitions-varying-size,Fail -kms_cursor_legacy@cursor-vs-flip-legacy,Fail -kms_cursor_legacy@cursor-vs-flip-toggle,Fail -kms_cursor_legacy@cursor-vs-flip-varying-size,Fail -kms_cursor_legacy@flip-vs-cursor-atomic,Fail -kms_cursor_legacy@flip-vs-cursor-crc-atomic,Fail -kms_cursor_legacy@flip-vs-cursor-crc-legacy,Fail -kms_cursor_legacy@flip-vs-cursor-legacy,Fail -kms_flip@flip-vs-modeset-vs-hang,Fail -kms_flip@flip-vs-panning-vs-hang,Fail -kms_lease@lease-uevent,Fail -kms_pipe_crc_basic@compare-crc-sanitycheck-nv12,Fail -kms_plane_alpha_blend@alpha-7efc,Fail -kms_plane_alpha_blend@coverage-7efc,Fail -kms_plane_alpha_blend@coverage-vs-premult-vs-constant,Fail -kms_plane_cursor@overlay,Fail -kms_plane_cursor@viewport,Fail diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt deleted file mode 100644 index e32d73c6c98e..000000000000 --- a/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt +++ /dev/null @@ -1,139 +0,0 @@ -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@basic-flip-after-cursor-atomic - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@basic-flip-after-cursor-legacy - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@basic-flip-after-cursor-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@basic-flip-before-cursor-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@flip-vs-cursor-atomic-transitions - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@flip-vs-cursor-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@short-flip-after-cursor-atomic-transitions - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@short-flip-after-cursor-atomic-transitions-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@short-flip-after-cursor-toggle - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@short-flip-before-cursor-atomic-transitions - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@short-flip-before-cursor-atomic-transitions-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -msm/msm_shrink@copy-gpu-32 - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -msm/msm_shrink@copy-gpu-oom-32 - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-gf13702b8e -# Linux Version: 6.10.0-rc5 -kms_cursor_legacy@short-flip-before-cursor-toggle - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-gf13702b8e -# Linux Version: 6.10.0-rc5 -kms_cursor_legacy@flip-vs-cursor-toggle - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-gf13702b8e -# Linux Version: 6.10.0-rc5 -msm/msm_shrink@copy-mmap-oom-8 - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/64bc4bcf-de51-4e60-a9f7-1295a1e64c65@collabora.com/T/#t -# Failure Rate: 50 -# IGT Version: 1.28-ga73311079 -# Linux Version: 6.11.0-rc2 -kms_lease@page-flip-implicit-plane - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/64bc4bcf-de51-4e60-a9f7-1295a1e64c65@collabora.com/T/#t -# Failure Rate: 50 -# IGT Version: 1.28-ga73311079 -# Linux Version: 6.11.0-rc5 -kms_flip@flip-vs-expired-vblank - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://gitlab.freedesktop.org/drm/msm/-/issues/75 -# Failure Rate: 20 -# IGT Version: 1.30-g04bedb923 -# Linux Version: 6.14.0-rc4 -kms_flip@plain-flip-ts-check-interruptible diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt deleted file mode 100644 index 6c86d1953e11..000000000000 --- a/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt +++ /dev/null @@ -1,350 +0,0 @@ -# Hangs machine -kms_bw.* - -# Failing due to a bootloader/fw issue. The workaround in mesa CI involves these two patches -# https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/4b49f902ec6f2bb382cbbf489870573f4b43371e -# https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/38cdf4c5559771e2474ae0fecef8469f65147bc1 -msm/msm_mapping@* - -# Skip driver specific tests -^amdgpu.* -nouveau_.* -^panfrost.* -^v3d.* -^vc4.* -^vmwgfx* - -# Skip intel specific tests -gem_.* -i915_.* -tools_test.* -kms_dp_link_training.* - -# Currently fails and causes coverage loss for other tests -# since core_getversion also fails. -core_hotunplug.* - -# Whole machine hangs -kms_cursor_crc.* - -# IGT test crash -# IGT Version: 1.28-ga73311079 -# Linux Version: 6.11.0-rc2 -kms_content_protection@uevent - -# IGT issue. is_joiner_mode() should return false for non-Intel hardware. -# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 -kms_display_modes@extended-mode-basic -kms_display_modes@mst-extended-mode-negative - -# Kernel panic -msm/msm_recovery@hangcheck -# DEBUG - Begin test msm/msm_recovery@hangcheck -# Console: switching to colour dummy device 80x25 -# [ 489.526286] [IGT] msm_recovery: executing -# [ 489.531926] [IGT] msm_recovery: starting subtest hangcheck -# [ 492.808574] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 492.820358] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 492.831154] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 493.832570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 493.844177] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 493.854971] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 494.824633] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 494.836237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 494.847034] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 495.816570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 495.828170] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 495.838966] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 496.804643] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 496.816246] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 496.827041] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 497.832570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 497.844170] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 497.854963] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 498.820636] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 498.832232] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 498.843024] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 499.816568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 499.828163] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 499.838958] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 500.808570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 500.820165] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 500.830960] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 501.832570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 501.844175] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 501.854965] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 502.824568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 502.836171] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 502.846965] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 503.816570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 503.828176] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 503.838969] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 504.804640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 504.816237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 504.827033] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 505.828643] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 505.840247] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 505.851043] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 506.820637] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 506.832233] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 506.843026] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 507.816567] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 507.828171] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 507.838965] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 508.808568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 508.820173] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 508.830969] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 509.832568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 509.844173] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 509.854967] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 510.824568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 510.836162] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 510.846954] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 511.816569] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 511.828173] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 511.838968] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 512.804641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 512.816246] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 512.827040] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 513.828641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 513.840239] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 513.851035] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 514.824568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 514.836164] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 514.846959] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 515.812640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 515.824235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 515.835030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 515.912427] rcu: INFO: rcu_preempt self-detected stall on CPU -# [ 515.918398] rcu: 0-....: (6452 ticks this GP) idle=6afc/1/0x4000000000000000 softirq=12492/12697 fqs=3179 -# [ 515.929296] rcu: (t=6505 jiffies g=36205 q=58 ncpus=8) -# [ 515.934709] CPU: 0 UID: 0 PID: 126 Comm: sugov:0 Tainted: G W 6.14.0-rc4-gdddf15cff632 #1 -# [ 515.934727] Tainted: [W]=WARN -# [ 515.934732] Hardware name: Google Cheza (rev3+) (DT) -# [ 515.934739] pstate: 00400009 (nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) -# [ 515.934751] pc : rcu_core+0x59c/0xe68 -# [ 515.934769] lr : rcu_core+0x74/0xe68 -# [ 515.934781] sp : ffff800080003e50 -# [ 515.934785] x29: ffff800080003e50 x28: ffff225d038e9bc0 x27: 0000000000000002 -# [ 515.934805] x26: ffffc171a8ee6108 x25: ffffc171a85bc2c0 x24: ffff60ecd691e000 -# [ 515.934820] x23: ffffc171a85d15c0 x22: ffffc171a8f8d780 x21: ffff225e7eeef5c0 -# [ 515.934835] x20: ffffc171a8ef0e80 x19: ffffc171a85d15d1 x18: ffffc171a9461e70 -# [ 515.934850] x17: ffff60ecd691e000 x16: ffff800080000000 x15: 0000000000000000 -# [ 515.934866] x14: ffffc171a85d0780 x13: 0000000000000400 x12: 0000000000000000 -# [ 515.934880] x11: ffffc171a85ce900 x10: ffffc171a8ef5000 x9 : ffffc171a8ef0000 -# [ 515.934894] x8 : ffff800080003d88 x7 : ffffc171a8ee6100 x6 : ffff800080003de0 -# [ 515.934909] x5 : ffff800080003dc8 x4 : 0000000000000003 x3 : 0000000000000000 -# [ 515.934923] x2 : 0000000000000101 x1 : 0000000000000000 x0 : ffff225d038e9bc0 -# [ 515.934939] Call trace: -# [ 515.934945] rcu_core+0x59c/0xe68 (P) -# [ 515.934962] rcu_core_si+0x10/0x1c -# [ 515.934976] handle_softirqs+0x118/0x4b8 -# [ 515.934994] __do_softirq+0x14/0x20 -# [ 515.935007] ____do_softirq+0x10/0x1c -# [ 515.935021] call_on_irq_stack+0x24/0x4c -# [ 515.935034] do_softirq_own_stack+0x1c/0x28 -# [ 515.935048] __irq_exit_rcu+0x174/0x1b4 -# [ 515.935063] irq_exit_rcu+0x10/0x38 -# [ 515.935077] el1_interrupt+0x38/0x64 -# [ 515.935092] el1h_64_irq_handler+0x18/0x24 -# [ 515.935104] el1h_64_irq+0x6c/0x70 -# [ 515.935115] lock_acquire+0x1e0/0x338 (P) -# [ 515.935129] __mutex_lock+0xa8/0x4b8 -# [ 515.935144] mutex_lock_nested+0x24/0x30 -# [ 515.935159] _find_opp_table_unlocked+0x40/0xfc -# [ 515.935174] _find_key+0x64/0x16c -# [ 515.935184] dev_pm_opp_find_freq_exact+0x4c/0x74 -# [ 515.935197] qcom_cpufreq_hw_target_index+0xe8/0x128 -# [ 515.935211] __cpufreq_driver_target+0x144/0x29c -# [ 515.935227] sugov_work+0x58/0x74 -# [ 515.935239] kthread_worker_fn+0xf4/0x324 -# [ 515.935254] kthread+0x12c/0x208 -# [ 515.935266] ret_from_fork+0x10/0x20 -# [ 516.808569] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 516.820174] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 516.830968] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 517.828641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 517.840236] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 517.851032] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 518.820642] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 518.832237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 518.843030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 519.812636] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 519.824231] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 519.835026] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 520.808570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 520.820165] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 520.830959] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 521.828643] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 521.840238] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 521.851033] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 522.820636] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 522.832232] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 522.843027] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 523.812639] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 523.824239] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 523.835034] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 524.804640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 524.816235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 524.827026] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 525.828641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 525.840236] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 525.851031] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 526.820641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 526.832244] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 526.843041] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 527.812642] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 527.824242] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 527.835038] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 528.804639] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 528.816234] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 528.827027] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 529.832634] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 529.844231] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 529.855017] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 530.820646] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 530.832270] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 530.843065] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 531.812640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 531.824238] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 531.835030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 532.804640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 532.816237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 532.827031] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 533.828640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 533.840243] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 533.851037] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 534.820640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 534.832245] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 534.843038] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 535.812641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 535.824238] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 535.835033] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 536.804639] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 536.816235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 536.827030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 537.828640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 537.840234] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 537.851020] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 538.820640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 538.832235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 538.843027] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 539.812644] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 539.824247] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 539.835040] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 540.124426] watchdog: BUG: soft lockup - CPU#0 stuck for 49s! [sugov:0:126] -# [ 540.124439] Modules linked in: -# [ 540.124448] irq event stamp: 9912389 -# [ 540.124453] hardirqs last enabled at (9912388): [<ffffc171a767a24c>] exit_to_kernel_mode+0x38/0x130 -# [ 540.124473] hardirqs last disabled at (9912389): [<ffffc171a767a368>] el1_interrupt+0x24/0x64 -# [ 540.124486] softirqs last enabled at (9898068): [<ffffc171a62bc290>] handle_softirqs+0x4a0/0x4b8 -# [ 540.124505] softirqs last disabled at (9898071): [<ffffc171a62105b0>] __do_softirq+0x14/0x20 -# [ 540.124525] CPU: 0 UID: 0 PID: 126 Comm: sugov:0 Tainted: G W 6.14.0-rc4-gdddf15cff632 #1 -# [ 540.124540] Tainted: [W]=WARN -# [ 540.124544] Hardware name: Google Cheza (rev3+) (DT) -# [ 540.124549] pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) -# [ 540.124560] pc : xhci_urb_enqueue+0xbc/0x32c -# [ 540.124573] lr : xhci_urb_enqueue+0xb4/0x32c -# [ 540.124581] sp : ffff800080003c20 -# [ 540.124586] x29: ffff800080003c20 x28: 0000000000000000 x27: ffff225d00b1e6a0 -# [ 540.124602] x26: ffff225d01c3d800 x25: 0000000000000001 x24: 0000000000000006 -# [ 540.124617] x23: ffff225d044dc000 x22: ffff225d044dc000 x21: 0000000000000001 -# [ 540.124632] x20: ffff225d002d7280 x19: ffff225d0573a780 x18: ffff225e7eff0f50 -# [ 540.124647] x17: 000000000000cab0 x16: 0000000000000000 x15: ffff225d0353a000 -# [ 540.124661] x14: 0000000000000000 x13: 0000000000000820 x12: 0000000000000000 -# [ 540.124674] x11: ffff800080003a30 x10: 0000000000000001 x9 : 0000000000000000 -# [ 540.124689] x8 : ffff225d002d7300 x7 : 0000000000000000 x6 : 000000000000003f -# [ 540.124702] x5 : 00000000ffffffff x4 : 0000000000000920 x3 : 0000000000000080 -# [ 540.124716] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff225d002d7280 -# [ 540.124731] Call trace: -# [ 540.124736] xhci_urb_enqueue+0xbc/0x32c (P) -# [ 540.124751] usb_hcd_submit_urb+0x98/0x7fc -# [ 540.124766] usb_submit_urb+0x294/0x560 -# [ 540.124780] intr_callback+0x78/0x1fc -# [ 540.124798] __usb_hcd_giveback_urb+0x68/0x128 -# [ 540.124812] usb_giveback_urb_bh+0xa8/0x140 -# [ 540.124825] process_one_work+0x208/0x5e8 -# [ 540.124840] bh_worker+0x1a8/0x20c -# [ 540.124853] workqueue_softirq_action+0x78/0x88 -# [ 540.124868] tasklet_hi_action+0x14/0x3c -# [ 540.124883] handle_softirqs+0x118/0x4b8 -# [ 540.124897] __do_softirq+0x14/0x20 -# [ 540.124908] ____do_softirq+0x10/0x1c -# [ 540.124922] call_on_irq_stack+0x24/0x4c -# [ 540.124934] do_softirq_own_stack+0x1c/0x28 -# [ 540.124947] __irq_exit_rcu+0x174/0x1b4 -# [ 540.124961] irq_exit_rcu+0x10/0x38 -# [ 540.124976] el1_interrupt+0x38/0x64 -# [ 540.124987] el1h_64_irq_handler+0x18/0x24 -# [ 540.124998] el1h_64_irq+0x6c/0x70 -# [ 540.125009] lock_acquire+0x1e0/0x338 (P) -# [ 540.125023] __mutex_lock+0xa8/0x4b8 -# [ 540.125038] mutex_lock_nested+0x24/0x30 -# [ 540.125052] _find_opp_table_unlocked+0x40/0xfc -# [ 540.125067] _find_key+0x64/0x16c -# [ 540.125078] dev_pm_opp_find_freq_exact+0x4c/0x74 -# [ 540.125090] qcom_cpufreq_hw_target_index+0xe8/0x128 -# [ 540.125105] __cpufreq_driver_target+0x144/0x29c -# [ 540.125121] sugov_work+0x58/0x74 -# [ 540.125133] kthread_worker_fn+0xf4/0x324 -# [ 540.125148] kthread+0x12c/0x208 -# [ 540.125160] ret_from_fork+0x10/0x20 -# [ 540.125176] Kernel panic - not syncing: softlockup: hung tasks -# [ 540.423567] CPU: 0 UID: 0 PID: 126 Comm: sugov:0 Tainted: G W L 6.14.0-rc4-gdddf15cff632 #1 -# [ 540.433411] Tainted: [W]=WARN, [L]=SOFTLOCKUP -# [ 540.437901] Hardware name: Google Cheza (rev3+) (DT) -# [ 540.443022] Call trace: -# [ 540.445559] show_stack+0x18/0x24 (C) -# [ 540.449357] dump_stack_lvl+0x38/0xd0 -# [ 540.453157] dump_stack+0x18/0x24 -# [ 540.456599] panic+0x3bc/0x41c -# [ 540.459767] watchdog_timer_fn+0x254/0x2e4 -# [ 540.464005] __hrtimer_run_queues+0x3c4/0x440 -# [ 540.468508] hrtimer_interrupt+0xe4/0x244 -# [ 540.472662] arch_timer_handler_phys+0x2c/0x44 -# [ 540.477256] handle_percpu_devid_irq+0x90/0x1f0 -# [ 540.481943] handle_irq_desc+0x40/0x58 -# [ 540.485829] generic_handle_domain_irq+0x1c/0x28 -# [ 540.490604] gic_handle_irq+0x4c/0x11c -# [ 540.494483] do_interrupt_handler+0x50/0x84 -# [ 540.498811] el1_interrupt+0x34/0x64 -# [ 540.502518] el1h_64_irq_handler+0x18/0x24 -# [ 540.506758] el1h_64_irq+0x6c/0x70 -# [ 540.510279] xhci_urb_enqueue+0xbc/0x32c (P) -# [ 540.514693] usb_hcd_submit_urb+0x98/0x7fc -# [ 540.518932] usb_submit_urb+0x294/0x560 -# [ 540.522901] intr_callback+0x78/0x1fc -# [ 540.526700] __usb_hcd_giveback_urb+0x68/0x128 -# [ 540.531288] usb_giveback_urb_bh+0xa8/0x140 -# [ 540.535614] process_one_work+0x208/0x5e8 -# [ 540.539769] bh_worker+0x1a8/0x20c -# [ 540.543293] workqueue_softirq_action+0x78/0x88 -# [ 540.547980] tasklet_hi_action+0x14/0x3c -# [ 540.552038] handle_softirqs+0x118/0x4b8 -# [ 540.556096] __do_softirq+0x14/0x20 -# [ 540.559705] ____do_softirq+0x10/0x1c -# [ 540.563500] call_on_irq_stack+0x24/0x4c -# [ 540.567554] do_softirq_own_stack+0x1c/0x28 -# [ 540.571878] __irq_exit_rcu+0x174/0x1b4 -# [ 540.575849] irq_exit_rcu+0x10/0x38 -# [ 540.579462] el1_interrupt+0x38/0x64 -# [ 540.583158] el1h_64_irq_handler+0x18/0x24 -# [ 540.587397] el1h_64_irq+0x6c/0x70 -# [ 540.590918] lock_acquire+0x1e0/0x338 (P) -# [ 540.595060] __mutex_lock+0xa8/0x4b8 -# [ 540.598760] mutex_lock_nested+0x24/0x30 -# [ 540.602818] _find_opp_table_unlocked+0x40/0xfc -# [ 540.607503] _find_key+0x64/0x16c -# [ 540.610940] dev_pm_opp_find_freq_exact+0x4c/0x74 -# [ 540.615798] qcom_cpufreq_hw_target_index+0xe8/0x128 -# [ 540.620924] __cpufreq_driver_target+0x144/0x29c -# [ 540.625698] sugov_work+0x58/0x74 -# [ 540.629134] kthread_worker_fn+0xf4/0x324 -# [ 540.633278] kthread+0x12c/0x208 -# [ 540.636619] ret_from_fork+0x10/0x20 -# [ 540.640321] SMP: stopping secondary CPUs -# [ 540.644518] Kernel Offset: 0x417126200000 from 0xffff800080000000 -# [ 540.650848] PHYS_OFFSET: 0xfff0dda400000000 -# [ 540.655170] CPU features: 0x000,00000100,00901250,8200721b -# [ 540.660829] Memory Limit: none -# [ 540.663999] ---[ end Kernel panic - not syncing: softlockup: hung tasks ]--- diff --git a/drivers/gpu/drm/clients/drm_client_setup.c b/drivers/gpu/drm/clients/drm_client_setup.c index aec2fab6d2bf..72480db1f00d 100644 --- a/drivers/gpu/drm/clients/drm_client_setup.c +++ b/drivers/gpu/drm/clients/drm_client_setup.c @@ -4,6 +4,7 @@ #include <drm/clients/drm_client_setup.h> #include <drm/drm_device.h> +#include <drm/drm_drv.h> #include <drm/drm_fourcc.h> #include <drm/drm_print.h> @@ -33,6 +34,10 @@ MODULE_PARM_DESC(active, */ void drm_client_setup(struct drm_device *dev, const struct drm_format_info *format) { + if (!drm_core_check_feature(dev, DRIVER_MODESET)) { + drm_dbg(dev, "driver does not support mode-setting, skipping DRM clients\n"); + return; + } #ifdef CONFIG_DRM_FBDEV_EMULATION if (!strcmp(drm_client_default, "fbdev")) { diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c index 385a1bfdb272..1c3920297906 100644 --- a/drivers/gpu/drm/display/drm_dp_helper.c +++ b/drivers/gpu/drm/display/drm_dp_helper.c @@ -3957,23 +3957,31 @@ EXPORT_SYMBOL(drm_dp_pcon_convert_rgb_to_ycbcr); * Returns: %0 on success, negative error code on failure */ int drm_edp_backlight_set_level(struct drm_dp_aux *aux, const struct drm_edp_backlight_info *bl, - u16 level) + u32 level) { int ret; - u8 buf[2] = { 0 }; + unsigned int offset = DP_EDP_BACKLIGHT_BRIGHTNESS_MSB; + u8 buf[3] = { 0 }; /* The panel uses the PWM for controlling brightness levels */ - if (!bl->aux_set) + if (!(bl->aux_set || bl->luminance_set)) return 0; - if (bl->lsb_reg_used) { + if (bl->luminance_set) { + level = level * 1000; + level &= 0xffffff; + buf[0] = (level & 0x0000ff); + buf[1] = (level & 0x00ff00) >> 8; + buf[2] = (level & 0xff0000) >> 16; + offset = DP_EDP_PANEL_TARGET_LUMINANCE_VALUE; + } else if (bl->lsb_reg_used) { buf[0] = (level & 0xff00) >> 8; buf[1] = (level & 0x00ff); } else { buf[0] = level; } - ret = drm_dp_dpcd_write_data(aux, DP_EDP_BACKLIGHT_BRIGHTNESS_MSB, buf, sizeof(buf)); + ret = drm_dp_dpcd_write_data(aux, offset, buf, sizeof(buf)); if (ret < 0) { drm_err(aux->drm_dev, "%s: Failed to write aux backlight level: %d\n", @@ -4036,7 +4044,7 @@ drm_edp_backlight_set_enable(struct drm_dp_aux *aux, const struct drm_edp_backli * Returns: %0 on success, negative error code on failure. */ int drm_edp_backlight_enable(struct drm_dp_aux *aux, const struct drm_edp_backlight_info *bl, - const u16 level) + const u32 level) { int ret; u8 dpcd_buf; @@ -4046,6 +4054,9 @@ int drm_edp_backlight_enable(struct drm_dp_aux *aux, const struct drm_edp_backli else dpcd_buf = DP_EDP_BACKLIGHT_CONTROL_MODE_PWM; + if (bl->luminance_set) + dpcd_buf |= DP_EDP_PANEL_LUMINANCE_CONTROL_ENABLE; + if (bl->pwmgen_bit_count) { ret = drm_dp_dpcd_write_byte(aux, DP_EDP_PWMGEN_BIT_COUNT, bl->pwmgen_bit_count); if (ret < 0) @@ -4209,7 +4220,7 @@ drm_edp_backlight_probe_state(struct drm_dp_aux *aux, struct drm_edp_backlight_i u8 *current_mode) { int ret; - u8 buf[2]; + u8 buf[3]; u8 mode_reg; ret = drm_dp_dpcd_read_byte(aux, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, &mode_reg); @@ -4226,17 +4237,37 @@ drm_edp_backlight_probe_state(struct drm_dp_aux *aux, struct drm_edp_backlight_i if (*current_mode == DP_EDP_BACKLIGHT_CONTROL_MODE_DPCD) { int size = 1 + bl->lsb_reg_used; - ret = drm_dp_dpcd_read_data(aux, DP_EDP_BACKLIGHT_BRIGHTNESS_MSB, buf, size); - if (ret < 0) { - drm_dbg_kms(aux->drm_dev, "%s: Failed to read backlight level: %d\n", - aux->name, ret); - return ret; - } + if (bl->luminance_set) { + ret = drm_dp_dpcd_read_data(aux, DP_EDP_PANEL_TARGET_LUMINANCE_VALUE, + buf, sizeof(buf)); + if (ret < 0) { + drm_dbg_kms(aux->drm_dev, + "%s: Failed to read backlight level: %d\n", + aux->name, ret); + return ret; + } - if (bl->lsb_reg_used) - return (buf[0] << 8) | buf[1]; - else - return buf[0]; + /* + * Incase luminance is set we want to send the value back in nits but + * since DP_EDP_PANEL_TARGET_LUMINANCE stores values in millinits we + * need to divide by 1000. + */ + return (buf[0] | buf[1] << 8 | buf[2] << 16) / 1000; + } else { + ret = drm_dp_dpcd_read_data(aux, DP_EDP_BACKLIGHT_BRIGHTNESS_MSB, + buf, size); + if (ret < 0) { + drm_dbg_kms(aux->drm_dev, + "%s: Failed to read backlight level: %d\n", + aux->name, ret); + return ret; + } + + if (bl->lsb_reg_used) + return (buf[0] << 8) | buf[1]; + else + return buf[0]; + } } /* @@ -4251,10 +4282,12 @@ drm_edp_backlight_probe_state(struct drm_dp_aux *aux, struct drm_edp_backlight_i * interface. * @aux: The DP aux device to use for probing * @bl: The &drm_edp_backlight_info struct to fill out with information on the backlight + * @max_luminance: max luminance when need luminance is set as true * @driver_pwm_freq_hz: Optional PWM frequency from the driver in hz * @edp_dpcd: A cached copy of the eDP DPCD * @current_level: Where to store the probed brightness level, if any * @current_mode: Where to store the currently set backlight control mode + * @need_luminance: Tells us if a we want to manipulate backlight using luminance values * * Initializes a &drm_edp_backlight_info struct by probing @aux for it's backlight capabilities, * along with also probing the current and maximum supported brightness levels. @@ -4266,8 +4299,9 @@ drm_edp_backlight_probe_state(struct drm_dp_aux *aux, struct drm_edp_backlight_i */ int drm_edp_backlight_init(struct drm_dp_aux *aux, struct drm_edp_backlight_info *bl, + u32 max_luminance, u16 driver_pwm_freq_hz, const u8 edp_dpcd[EDP_DISPLAY_CTL_CAP_SIZE], - u16 *current_level, u8 *current_mode) + u32 *current_level, u8 *current_mode, bool need_luminance) { int ret; @@ -4277,18 +4311,26 @@ drm_edp_backlight_init(struct drm_dp_aux *aux, struct drm_edp_backlight_info *bl bl->aux_set = true; if (edp_dpcd[2] & DP_EDP_BACKLIGHT_BRIGHTNESS_BYTE_COUNT) bl->lsb_reg_used = true; + if ((edp_dpcd[0] & DP_EDP_15) && edp_dpcd[3] & + (DP_EDP_PANEL_LUMINANCE_CONTROL_CAPABLE) && need_luminance) + bl->luminance_set = true; /* Sanity check caps */ - if (!bl->aux_set && !(edp_dpcd[2] & DP_EDP_BACKLIGHT_BRIGHTNESS_PWM_PIN_CAP)) { + if (!bl->aux_set && !(edp_dpcd[2] & DP_EDP_BACKLIGHT_BRIGHTNESS_PWM_PIN_CAP) && + !bl->luminance_set) { drm_dbg_kms(aux->drm_dev, - "%s: Panel supports neither AUX or PWM brightness control? Aborting\n", + "%s: Panel does not support AUX, PWM or luminance-based brightness control. Aborting\n", aux->name); return -EINVAL; } - ret = drm_edp_backlight_probe_max(aux, bl, driver_pwm_freq_hz, edp_dpcd); - if (ret < 0) - return ret; + if (bl->luminance_set) { + bl->max = max_luminance; + } else { + ret = drm_edp_backlight_probe_max(aux, bl, driver_pwm_freq_hz, edp_dpcd); + if (ret < 0) + return ret; + } ret = drm_edp_backlight_probe_state(aux, bl, current_mode); if (ret < 0) @@ -4367,7 +4409,7 @@ int drm_panel_dp_aux_backlight(struct drm_panel *panel, struct drm_dp_aux *aux) { struct dp_aux_backlight *bl; struct backlight_properties props = { 0 }; - u16 current_level; + u32 current_level; u8 current_mode; u8 edp_dpcd[EDP_DISPLAY_CTL_CAP_SIZE]; int ret; @@ -4391,8 +4433,8 @@ int drm_panel_dp_aux_backlight(struct drm_panel *panel, struct drm_dp_aux *aux) bl->aux = aux; - ret = drm_edp_backlight_init(aux, &bl->info, 0, edp_dpcd, - ¤t_level, ¤t_mode); + ret = drm_edp_backlight_init(aux, &bl->info, 0, 0, edp_dpcd, + ¤t_level, ¤t_mode, false); if (ret < 0) return ret; diff --git a/drivers/gpu/drm/display/drm_hdmi_cec_helper.c b/drivers/gpu/drm/display/drm_hdmi_cec_helper.c index b4273c3522fa..3651ad0f76e0 100644 --- a/drivers/gpu/drm/display/drm_hdmi_cec_helper.c +++ b/drivers/gpu/drm/display/drm_hdmi_cec_helper.c @@ -69,7 +69,7 @@ static void drm_connector_hdmi_cec_adapter_unregister(struct drm_device *dev, vo struct drm_connector *connector = res; struct drm_connector_hdmi_cec_data *data = connector->cec.data; - cec_delete_adapter(data->adapter); + cec_unregister_adapter(data->adapter); if (data->funcs->uninit) data->funcs->uninit(connector); diff --git a/drivers/gpu/drm/display/drm_hdmi_helper.c b/drivers/gpu/drm/display/drm_hdmi_helper.c index 6063c155bdea..a237dc55805d 100644 --- a/drivers/gpu/drm/display/drm_hdmi_helper.c +++ b/drivers/gpu/drm/display/drm_hdmi_helper.c @@ -45,7 +45,7 @@ int drm_hdmi_infoframe_set_hdr_metadata(struct hdmi_drm_infoframe *frame, /* Sink EOTF is Bit map while infoframe is absolute values */ if (!is_eotf_supported(hdr_metadata->hdmi_metadata_type1.eotf, - connector->hdr_sink_metadata.hdmi_type1.eotf)) + connector->display_info.hdr_sink_metadata.hdmi_type1.eotf)) DRM_DEBUG_KMS("Unknown EOTF %d\n", hdr_metadata->hdmi_metadata_type1.eotf); err = hdmi_drm_infoframe_init(frame); diff --git a/drivers/gpu/drm/drm_bridge.c b/drivers/gpu/drm/drm_bridge.c index d6ce7b4c019f..0b450b334afd 100644 --- a/drivers/gpu/drm/drm_bridge.c +++ b/drivers/gpu/drm/drm_bridge.c @@ -295,6 +295,11 @@ EXPORT_SYMBOL(__devm_drm_bridge_alloc); */ void drm_bridge_add(struct drm_bridge *bridge) { + if (!bridge->container) + DRM_WARN("DRM bridge corrupted or not allocated by devm_drm_bridge_alloc()\n"); + + drm_bridge_get(bridge); + mutex_init(&bridge->hpd_mutex); if (bridge->ops & DRM_BRIDGE_OP_HDMI) @@ -342,6 +347,8 @@ void drm_bridge_remove(struct drm_bridge *bridge) mutex_unlock(&bridge_lock); mutex_destroy(&bridge->hpd_mutex); + + drm_bridge_put(bridge); } EXPORT_SYMBOL(drm_bridge_remove); @@ -407,11 +414,17 @@ int drm_bridge_attach(struct drm_encoder *encoder, struct drm_bridge *bridge, if (!encoder || !bridge) return -EINVAL; - if (previous && (!previous->dev || previous->encoder != encoder)) - return -EINVAL; + drm_bridge_get(bridge); - if (bridge->dev) - return -EBUSY; + if (previous && (!previous->dev || previous->encoder != encoder)) { + ret = -EINVAL; + goto err_put_bridge; + } + + if (bridge->dev) { + ret = -EBUSY; + goto err_put_bridge; + } bridge->dev = encoder->dev; bridge->encoder = encoder; @@ -460,6 +473,8 @@ err_reset_bridge: "failed to attach bridge %pOF to encoder %s\n", bridge->of_node, encoder->name); +err_put_bridge: + drm_bridge_put(bridge); return ret; } EXPORT_SYMBOL(drm_bridge_attach); @@ -480,6 +495,7 @@ void drm_bridge_detach(struct drm_bridge *bridge) list_del(&bridge->chain_node); bridge->dev = NULL; + drm_bridge_put(bridge); } /** diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c index 0167e0e0d425..272d6254ea47 100644 --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -1687,7 +1687,7 @@ EXPORT_SYMBOL(drm_hdmi_connector_get_output_format_name); * structure from userspace. This is received as blob and stored in * &drm_connector_state.hdr_output_metadata. It parses EDID and saves the * sink metadata in &struct hdr_sink_metadata, as - * &drm_connector.hdr_sink_metadata. Driver uses + * &drm_connector.display_info.hdr_sink_metadata. Driver uses * drm_hdmi_infoframe_set_hdr_metadata() helper to set the HDR metadata, * hdmi_drm_infoframe_pack() to pack the infoframe as per spec, in case of * HDMI encoder. diff --git a/drivers/gpu/drm/drm_debugfs.c b/drivers/gpu/drm/drm_debugfs.c index abceb28b23fc..365cf337529f 100644 --- a/drivers/gpu/drm/drm_debugfs.c +++ b/drivers/gpu/drm/drm_debugfs.c @@ -44,6 +44,9 @@ #include "drm_crtc_internal.h" #include "drm_internal.h" +static struct dentry *accel_debugfs_root; +static struct dentry *drm_debugfs_root; + /*************************************************** * Initialization, etc. **************************************************/ @@ -287,16 +290,120 @@ int drm_debugfs_remove_files(const struct drm_info_list *files, int count, } EXPORT_SYMBOL(drm_debugfs_remove_files); +void drm_debugfs_bridge_params(void) +{ + drm_bridge_debugfs_params(drm_debugfs_root); +} + +void drm_debugfs_init_root(void) +{ + drm_debugfs_root = debugfs_create_dir("dri", NULL); +#if IS_ENABLED(CONFIG_DRM_ACCEL) + accel_debugfs_root = debugfs_create_dir("accel", NULL); +#endif +} + +void drm_debugfs_remove_root(void) +{ +#if IS_ENABLED(CONFIG_DRM_ACCEL) + debugfs_remove(accel_debugfs_root); +#endif + debugfs_remove(drm_debugfs_root); +} + +static int drm_debugfs_proc_info_show(struct seq_file *m, void *unused) +{ + struct pid *pid; + struct task_struct *task; + struct drm_file *file = m->private; + + if (!file) + return -EINVAL; + + rcu_read_lock(); + pid = rcu_dereference(file->pid); + task = pid_task(pid, PIDTYPE_TGID); + + seq_printf(m, "pid: %d\n", task ? task->pid : 0); + seq_printf(m, "comm: %s\n", task ? task->comm : "Unset"); + rcu_read_unlock(); + return 0; +} + +static int drm_debufs_proc_info_open(struct inode *inode, struct file *file) +{ + return single_open(file, drm_debugfs_proc_info_show, inode->i_private); +} + +static const struct file_operations drm_debugfs_proc_info_fops = { + .owner = THIS_MODULE, + .open = drm_debufs_proc_info_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +/** + * drm_debugfs_clients_add - Add a per client debugfs directory + * @file: drm_file for a client + * + * Create the debugfs directory for each client. This will be used to populate + * driver specific data for each client. + * + * Also add the process information debugfs file for each client to tag + * which client belongs to which process. + */ +void drm_debugfs_clients_add(struct drm_file *file) +{ + char *client; + + client = kasprintf(GFP_KERNEL, "client-%llu", file->client_id); + if (!client) + return; + + /* Create a debugfs directory for the client in root on drm debugfs */ + file->debugfs_client = debugfs_create_dir(client, drm_debugfs_root); + kfree(client); + + debugfs_create_file("proc_info", 0444, file->debugfs_client, file, + &drm_debugfs_proc_info_fops); + + client = kasprintf(GFP_KERNEL, "../%s", file->minor->dev->unique); + if (!client) + return; + + /* Create a link from client_id to the drm device this client id belongs to */ + debugfs_create_symlink("device", file->debugfs_client, client); + kfree(client); +} + +/** + * drm_debugfs_clients_remove - removes all debugfs directories and files + * @file: drm_file for a client + * + * Removes the debugfs directories recursively from the client directory. + * + * There is also a possibility that debugfs files are open while the drm_file + * is released. + */ +void drm_debugfs_clients_remove(struct drm_file *file) +{ + debugfs_remove_recursive(file->debugfs_client); + file->debugfs_client = NULL; +} + /** * drm_debugfs_dev_init - create debugfs directory for the device * @dev: the device which we want to create the directory for - * @root: the parent directory depending on the device type * * Creates the debugfs directory for the device under the given root directory. */ -void drm_debugfs_dev_init(struct drm_device *dev, struct dentry *root) +void drm_debugfs_dev_init(struct drm_device *dev) { - dev->debugfs_root = debugfs_create_dir(dev->unique, root); + if (drm_core_check_feature(dev, DRIVER_COMPUTE_ACCEL)) + dev->debugfs_root = debugfs_create_dir(dev->unique, accel_debugfs_root); + else + dev->debugfs_root = debugfs_create_dir(dev->unique, drm_debugfs_root); } /** @@ -323,14 +430,13 @@ void drm_debugfs_dev_register(struct drm_device *dev) drm_atomic_debugfs_init(dev); } -int drm_debugfs_register(struct drm_minor *minor, int minor_id, - struct dentry *root) +int drm_debugfs_register(struct drm_minor *minor, int minor_id) { struct drm_device *dev = minor->dev; char name[64]; sprintf(name, "%d", minor_id); - minor->debugfs_symlink = debugfs_create_symlink(name, root, + minor->debugfs_symlink = debugfs_create_symlink(name, drm_debugfs_root, dev->unique); /* TODO: Only for compatibility with drivers */ diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c index 02556363e918..cdd591b11488 100644 --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -72,8 +72,6 @@ DEFINE_XARRAY_ALLOC(drm_minors_xa); */ static bool drm_core_init_complete; -static struct dentry *drm_debugfs_root; - DEFINE_STATIC_SRCU(drm_unplug_srcu); /* @@ -186,8 +184,7 @@ static int drm_minor_register(struct drm_device *dev, enum drm_minor_type type) return 0; if (minor->type != DRM_MINOR_ACCEL) { - ret = drm_debugfs_register(minor, minor->index, - drm_debugfs_root); + ret = drm_debugfs_register(minor, minor->index); if (ret) { DRM_ERROR("DRM: Failed to initialize /sys/kernel/debug/dri.\n"); goto err_debugfs; @@ -787,10 +784,7 @@ static int drm_dev_init(struct drm_device *dev, goto err; } - if (drm_core_check_feature(dev, DRIVER_COMPUTE_ACCEL)) - accel_debugfs_init(dev); - else - drm_debugfs_dev_init(dev, drm_debugfs_root); + drm_debugfs_dev_init(dev); return 0; @@ -1230,7 +1224,7 @@ static void drm_core_exit(void) drm_panic_exit(); accel_core_exit(); unregister_chrdev(DRM_MAJOR, "drm"); - debugfs_remove(drm_debugfs_root); + drm_debugfs_remove_root(); drm_sysfs_destroy(); WARN_ON(!xa_empty(&drm_minors_xa)); drm_connector_ida_destroy(); @@ -1249,8 +1243,8 @@ static int __init drm_core_init(void) goto error; } - drm_debugfs_root = debugfs_create_dir("dri", NULL); - drm_bridge_debugfs_params(drm_debugfs_root); + drm_debugfs_init_root(); + drm_debugfs_bridge_params(); ret = register_chrdev(DRM_MAJOR, "drm", &drm_stub_fops); if (ret < 0) diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index a876d7ef252b..e2e85345aa9a 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -5394,7 +5394,8 @@ static void fixup_detailed_cea_mode_clock(struct drm_connector *connector, static void drm_calculate_luminance_range(struct drm_connector *connector) { - struct hdr_static_metadata *hdr_metadata = &connector->hdr_sink_metadata.hdmi_type1; + const struct hdr_static_metadata *hdr_metadata = + &connector->display_info.hdr_sink_metadata.hdmi_type1; struct drm_luminance_range_info *luminance_range = &connector->display_info.luminance_range; static const u8 pre_computed_values[] = { @@ -5455,21 +5456,21 @@ static uint8_t hdr_metadata_type(const u8 *edid_ext) static void drm_parse_hdr_metadata_block(struct drm_connector *connector, const u8 *db) { + struct hdr_static_metadata *hdr_metadata = + &connector->display_info.hdr_sink_metadata.hdmi_type1; u16 len; len = cea_db_payload_len(db); - connector->hdr_sink_metadata.hdmi_type1.eotf = - eotf_supported(db); - connector->hdr_sink_metadata.hdmi_type1.metadata_type = - hdr_metadata_type(db); + hdr_metadata->eotf = eotf_supported(db); + hdr_metadata->metadata_type = hdr_metadata_type(db); if (len >= 4) - connector->hdr_sink_metadata.hdmi_type1.max_cll = db[4]; + hdr_metadata->max_cll = db[4]; if (len >= 5) - connector->hdr_sink_metadata.hdmi_type1.max_fall = db[5]; + hdr_metadata->max_fall = db[5]; if (len >= 6) { - connector->hdr_sink_metadata.hdmi_type1.min_cll = db[6]; + hdr_metadata->min_cll = db[6]; /* Calculate only when all values are available */ drm_calculate_luminance_range(connector); @@ -6617,7 +6618,7 @@ static void drm_reset_display_info(struct drm_connector *connector) info->has_hdmi_infoframe = false; info->rgb_quant_range_selectable = false; memset(&info->hdmi, 0, sizeof(info->hdmi)); - memset(&connector->hdr_sink_metadata, 0, sizeof(connector->hdr_sink_metadata)); + memset(&info->hdr_sink_metadata, 0, sizeof(info->hdr_sink_metadata)); info->edid_hdmi_rgb444_dc_modes = 0; info->edid_hdmi_ycbcr444_dc_modes = 0; diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c index 3952e27447ee..eebd1a05ee97 100644 --- a/drivers/gpu/drm/drm_file.c +++ b/drivers/gpu/drm/drm_file.c @@ -46,6 +46,7 @@ #include <drm/drm_file.h> #include <drm/drm_gem.h> #include <drm/drm_print.h> +#include <drm/drm_debugfs.h> #include "drm_crtc_internal.h" #include "drm_internal.h" @@ -168,6 +169,9 @@ struct drm_file *drm_file_alloc(struct drm_minor *minor) drm_prime_init_file_private(&file->prime); + if (!drm_core_check_feature(dev, DRIVER_COMPUTE_ACCEL)) + drm_debugfs_clients_add(file); + if (dev->driver->open) { ret = dev->driver->open(dev, file); if (ret < 0) @@ -182,6 +186,10 @@ out_prime_destroy: drm_syncobj_release(file); if (drm_core_check_feature(dev, DRIVER_GEM)) drm_gem_release(dev, file); + + if (!drm_core_check_feature(dev, DRIVER_COMPUTE_ACCEL)) + drm_debugfs_clients_remove(file); + put_pid(rcu_access_pointer(file->pid)); kfree(file); @@ -236,6 +244,9 @@ void drm_file_free(struct drm_file *file) (long)old_encode_dev(file->minor->kdev->devt), atomic_read(&dev->open_count)); + if (!drm_core_check_feature(dev, DRIVER_COMPUTE_ACCEL)) + drm_debugfs_clients_remove(file); + drm_events_release(file); if (drm_core_check_feature(dev, DRIVER_MODESET)) { diff --git a/drivers/gpu/drm/drm_format_helper.c b/drivers/gpu/drm/drm_format_helper.c index 99d9f7bbc261..8f3daf38ca63 100644 --- a/drivers/gpu/drm/drm_format_helper.c +++ b/drivers/gpu/drm/drm_format_helper.c @@ -559,18 +559,6 @@ static void drm_fb_xrgb8888_to_rgb565_line(void *dbuf, const void *sbuf, unsigne drm_fb_xfrm_line_32to16(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_rgb565); } -static __always_inline u32 drm_xrgb8888_to_rgb565_swab(u32 pix) -{ - return swab16(drm_pixel_xrgb8888_to_rgb565(pix)); -} - -/* TODO: implement this helper as conversion to RGB565|BIG_ENDIAN */ -static void drm_fb_xrgb8888_to_rgb565_swab_line(void *dbuf, const void *sbuf, - unsigned int pixels) -{ - drm_fb_xfrm_line_32to16(dbuf, sbuf, pixels, drm_xrgb8888_to_rgb565_swab); -} - /** * drm_fb_xrgb8888_to_rgb565 - Convert XRGB8888 to RGB565 clip buffer * @dst: Array of RGB565 destination buffers @@ -580,7 +568,6 @@ static void drm_fb_xrgb8888_to_rgb565_swab_line(void *dbuf, const void *sbuf, * @fb: DRM framebuffer * @clip: Clip rectangle area to copy * @state: Transform and conversion state - * @swab: Swap bytes * * This function copies parts of a framebuffer to display memory and converts the * color format during the process. Destination and framebuffer formats must match. The @@ -595,23 +582,56 @@ static void drm_fb_xrgb8888_to_rgb565_swab_line(void *dbuf, const void *sbuf, */ void drm_fb_xrgb8888_to_rgb565(struct iosys_map *dst, const unsigned int *dst_pitch, const struct iosys_map *src, const struct drm_framebuffer *fb, - const struct drm_rect *clip, struct drm_format_conv_state *state, - bool swab) + const struct drm_rect *clip, struct drm_format_conv_state *state) { static const u8 dst_pixsize[DRM_FORMAT_MAX_PLANES] = { 2, }; - void (*xfrm_line)(void *dbuf, const void *sbuf, unsigned int npixels); + drm_fb_xfrm(dst, dst_pitch, dst_pixsize, src, fb, clip, false, state, + drm_fb_xrgb8888_to_rgb565_line); +} +EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb565); + +static void drm_fb_xrgb8888_to_rgb565be_line(void *dbuf, const void *sbuf, + unsigned int pixels) +{ + drm_fb_xfrm_line_32to16(dbuf, sbuf, pixels, drm_pixel_xrgb8888_to_rgb565be); +} - if (swab) - xfrm_line = drm_fb_xrgb8888_to_rgb565_swab_line; - else - xfrm_line = drm_fb_xrgb8888_to_rgb565_line; +/** + * drm_fb_xrgb8888_to_rgb565be - Convert XRGB8888 to RGB565|DRM_FORMAT_BIG_ENDIAN clip buffer + * @dst: Array of RGB565BE destination buffers + * @dst_pitch: Array of numbers of bytes between the start of two consecutive scanlines + * within @dst; can be NULL if scanlines are stored next to each other. + * @src: Array of XRGB8888 source buffer + * @fb: DRM framebuffer + * @clip: Clip rectangle area to copy + * @state: Transform and conversion state + * + * This function copies parts of a framebuffer to display memory and converts the + * color format during the process. Destination and framebuffer formats must match. The + * parameters @dst, @dst_pitch and @src refer to arrays. Each array must have at + * least as many entries as there are planes in @fb's format. Each entry stores the + * value for the format's respective color plane at the same index. + * + * This function does not apply clipping on @dst (i.e. the destination is at the + * top-left corner). + * + * Drivers can use this function for RGB565BE devices that don't support XRGB8888 natively. + */ +void drm_fb_xrgb8888_to_rgb565be(struct iosys_map *dst, const unsigned int *dst_pitch, + const struct iosys_map *src, const struct drm_framebuffer *fb, + const struct drm_rect *clip, struct drm_format_conv_state *state) +{ + static const u8 dst_pixsize[DRM_FORMAT_MAX_PLANES] = { + 2, + }; - drm_fb_xfrm(dst, dst_pitch, dst_pixsize, src, fb, clip, false, state, xfrm_line); + drm_fb_xfrm(dst, dst_pitch, dst_pixsize, src, fb, clip, false, state, + drm_fb_xrgb8888_to_rgb565be_line); } -EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb565); +EXPORT_SYMBOL(drm_fb_xrgb8888_to_rgb565be); static void drm_fb_xrgb8888_to_xrgb1555_line(void *dbuf, const void *sbuf, unsigned int pixels) { @@ -1188,7 +1208,7 @@ int drm_fb_blit(struct iosys_map *dst, const unsigned int *dst_pitch, uint32_t d return 0; } else if (fb_format == DRM_FORMAT_XRGB8888) { if (dst_format == DRM_FORMAT_RGB565) { - drm_fb_xrgb8888_to_rgb565(dst, dst_pitch, src, fb, clip, state, false); + drm_fb_xrgb8888_to_rgb565(dst, dst_pitch, src, fb, clip, state); return 0; } else if (dst_format == DRM_FORMAT_XRGB1555) { drm_fb_xrgb8888_to_xrgb1555(dst, dst_pitch, src, fb, clip, state); diff --git a/drivers/gpu/drm/drm_format_internal.h b/drivers/gpu/drm/drm_format_internal.h index 9428d5cfebc5..ce29dd05bcc5 100644 --- a/drivers/gpu/drm/drm_format_internal.h +++ b/drivers/gpu/drm/drm_format_internal.h @@ -5,6 +5,7 @@ #include <linux/bits.h> #include <linux/types.h> +#include <linux/swab.h> /* * Each pixel-format conversion helper takes a raw pixel in a @@ -59,6 +60,11 @@ static inline u32 drm_pixel_xrgb8888_to_rgb565(u32 pix) ((pix & 0x000000f8) >> 3); } +static inline u32 drm_pixel_xrgb8888_to_rgb565be(u32 pix) +{ + return swab16(drm_pixel_xrgb8888_to_rgb565(pix)); +} + static inline u32 drm_pixel_xrgb8888_to_rgbx5551(u32 pix) { return ((pix & 0x00f80000) >> 8) | diff --git a/drivers/gpu/drm/drm_fourcc.c b/drivers/gpu/drm/drm_fourcc.c index 55ddd99fd7f6..2890e889dd15 100644 --- a/drivers/gpu/drm/drm_fourcc.c +++ b/drivers/gpu/drm/drm_fourcc.c @@ -238,6 +238,14 @@ const struct drm_format_info *__drm_format_info(u32 format) { .format = DRM_FORMAT_ABGR2101010, .depth = 30, .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true }, { .format = DRM_FORMAT_RGBA1010102, .depth = 30, .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true }, { .format = DRM_FORMAT_BGRA1010102, .depth = 30, .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true }, + { .format = DRM_FORMAT_RGB161616, .depth = 0, + .num_planes = 1, .char_per_block = { 6, 0, 0 }, + .block_w = { 1, 0, 0 }, .block_h = { 1, 0, 0 }, + .hsub = 1, .vsub = 1, .has_alpha = false }, + { .format = DRM_FORMAT_BGR161616, .depth = 0, + .num_planes = 1, .char_per_block = { 6, 0, 0 }, + .block_w = { 1, 0, 0 }, .block_h = { 1, 0, 0 }, + .hsub = 1, .vsub = 1, .has_alpha = false }, { .format = DRM_FORMAT_ARGB8888, .depth = 32, .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true }, { .format = DRM_FORMAT_ABGR8888, .depth = 32, .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true }, { .format = DRM_FORMAT_RGBA8888, .depth = 32, .num_planes = 1, .cpp = { 4, 0, 0 }, .hsub = 1, .vsub = 1, .has_alpha = true }, diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index 19d50d254fe6..0905ef6786e9 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -1429,12 +1429,14 @@ EXPORT_SYMBOL(drm_gem_lru_move_tail); * @nr_to_scan: The number of pages to try to reclaim * @remaining: The number of pages left to reclaim, should be initialized by caller * @shrink: Callback to try to shrink/reclaim the object. + * @ticket: Optional ww_acquire_ctx context to use for locking */ unsigned long drm_gem_lru_scan(struct drm_gem_lru *lru, unsigned int nr_to_scan, unsigned long *remaining, - bool (*shrink)(struct drm_gem_object *obj)) + bool (*shrink)(struct drm_gem_object *obj, struct ww_acquire_ctx *ticket), + struct ww_acquire_ctx *ticket) { struct drm_gem_lru still_in_lru; struct drm_gem_object *obj; @@ -1467,17 +1469,20 @@ drm_gem_lru_scan(struct drm_gem_lru *lru, */ mutex_unlock(lru->lock); + if (ticket) + ww_acquire_init(ticket, &reservation_ww_class); + /* * Note that this still needs to be trylock, since we can * hit shrinker in response to trying to get backing pages * for this obj (ie. while it's lock is already held) */ - if (!dma_resv_trylock(obj->resv)) { + if (!ww_mutex_trylock(&obj->resv->lock, ticket)) { *remaining += obj->size >> PAGE_SHIFT; goto tail; } - if (shrink(obj)) { + if (shrink(obj, ticket)) { freed += obj->size >> PAGE_SHIFT; /* @@ -1491,6 +1496,9 @@ drm_gem_lru_scan(struct drm_gem_lru *lru, dma_resv_unlock(obj->resv); + if (ticket) + ww_acquire_fini(ticket); + tail: drm_gem_object_put(obj); mutex_lock(lru->lock); diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c index ab198645d90f..5bb4c77db2c3 100644 --- a/drivers/gpu/drm/drm_gpusvm.c +++ b/drivers/gpu/drm/drm_gpusvm.c @@ -9,10 +9,9 @@ #include <linux/dma-mapping.h> #include <linux/export.h> #include <linux/hmm.h> +#include <linux/hugetlb_inline.h> #include <linux/memremap.h> -#include <linux/migrate.h> #include <linux/mm_types.h> -#include <linux/pagemap.h> #include <linux/slab.h> #include <drm/drm_device.h> @@ -109,21 +108,6 @@ */ /** - * DOC: Migration - * - * The migration support is quite simple, allowing migration between RAM and - * device memory at the range granularity. For example, GPU SVM currently does - * not support mixing RAM and device memory pages within a range. This means - * that upon GPU fault, the entire range can be migrated to device memory, and - * upon CPU fault, the entire range is migrated to RAM. Mixed RAM and device - * memory storage within a range could be added in the future if required. - * - * The reasoning for only supporting range granularity is as follows: it - * simplifies the implementation, and range sizes are driver-defined and should - * be relatively small. - */ - -/** * DOC: Partial Unmapping of Ranges * * Partial unmapping of ranges (e.g., 1M out of 2M is unmapped by CPU resulting @@ -192,12 +176,9 @@ * } * * if (driver_migration_policy(range)) { - * mmap_read_lock(mm); - * devmem = driver_alloc_devmem(); - * err = drm_gpusvm_migrate_to_devmem(gpusvm, range, - * devmem_allocation, - * &ctx); - * mmap_read_unlock(mm); + * err = drm_pagemap_populate_mm(driver_choose_drm_pagemap(), + * gpuva_start, gpuva_end, gpusvm->mm, + * ctx->timeslice_ms); * if (err) // CPU mappings may have changed * goto retry; * } @@ -290,97 +271,6 @@ npages_in_range(unsigned long start, unsigned long end) } /** - * struct drm_gpusvm_zdd - GPU SVM zone device data - * - * @refcount: Reference count for the zdd - * @devmem_allocation: device memory allocation - * @device_private_page_owner: Device private pages owner - * - * This structure serves as a generic wrapper installed in - * page->zone_device_data. It provides infrastructure for looking up a device - * memory allocation upon CPU page fault and asynchronously releasing device - * memory once the CPU has no page references. Asynchronous release is useful - * because CPU page references can be dropped in IRQ contexts, while releasing - * device memory likely requires sleeping locks. - */ -struct drm_gpusvm_zdd { - struct kref refcount; - struct drm_gpusvm_devmem *devmem_allocation; - void *device_private_page_owner; -}; - -/** - * drm_gpusvm_zdd_alloc() - Allocate a zdd structure. - * @device_private_page_owner: Device private pages owner - * - * This function allocates and initializes a new zdd structure. It sets up the - * reference count and initializes the destroy work. - * - * Return: Pointer to the allocated zdd on success, ERR_PTR() on failure. - */ -static struct drm_gpusvm_zdd * -drm_gpusvm_zdd_alloc(void *device_private_page_owner) -{ - struct drm_gpusvm_zdd *zdd; - - zdd = kmalloc(sizeof(*zdd), GFP_KERNEL); - if (!zdd) - return NULL; - - kref_init(&zdd->refcount); - zdd->devmem_allocation = NULL; - zdd->device_private_page_owner = device_private_page_owner; - - return zdd; -} - -/** - * drm_gpusvm_zdd_get() - Get a reference to a zdd structure. - * @zdd: Pointer to the zdd structure. - * - * This function increments the reference count of the provided zdd structure. - * - * Return: Pointer to the zdd structure. - */ -static struct drm_gpusvm_zdd *drm_gpusvm_zdd_get(struct drm_gpusvm_zdd *zdd) -{ - kref_get(&zdd->refcount); - return zdd; -} - -/** - * drm_gpusvm_zdd_destroy() - Destroy a zdd structure. - * @ref: Pointer to the reference count structure. - * - * This function queues the destroy_work of the zdd for asynchronous destruction. - */ -static void drm_gpusvm_zdd_destroy(struct kref *ref) -{ - struct drm_gpusvm_zdd *zdd = - container_of(ref, struct drm_gpusvm_zdd, refcount); - struct drm_gpusvm_devmem *devmem = zdd->devmem_allocation; - - if (devmem) { - complete_all(&devmem->detached); - if (devmem->ops->devmem_release) - devmem->ops->devmem_release(devmem); - } - kfree(zdd); -} - -/** - * drm_gpusvm_zdd_put() - Put a zdd reference. - * @zdd: Pointer to the zdd structure. - * - * This function decrements the reference count of the provided zdd structure - * and schedules its destruction if the count drops to zero. - */ -static void drm_gpusvm_zdd_put(struct drm_gpusvm_zdd *zdd) -{ - kref_put(&zdd->refcount, drm_gpusvm_zdd_destroy); -} - -/** * drm_gpusvm_range_find() - Find GPU SVM range from GPU SVM notifier * @notifier: Pointer to the GPU SVM notifier structure. * @start: Start address of the range @@ -946,7 +836,7 @@ retry: * process-many-malloc' fails. In the failure case, each process * mallocs 16k but the CPU VMA is ~128k which results in 64k SVM * ranges. When migrating the SVM ranges, some processes fail in - * drm_gpusvm_migrate_to_devmem with 'migrate.cpages != npages' + * drm_pagemap_migrate_to_devmem with 'migrate.cpages != npages' * and then upon drm_gpusvm_range_get_pages device pages from * other processes are collected + faulted in which creates all * sorts of problems. Unsure exactly how this happening, also @@ -1364,7 +1254,7 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm, .dev_private_owner = gpusvm->device_private_page_owner, }; struct mm_struct *mm = gpusvm->mm; - struct drm_gpusvm_zdd *zdd; + void *zdd; unsigned long timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT); unsigned long i, j; @@ -1447,6 +1337,7 @@ map_pages: } zdd = NULL; + pagemap = NULL; num_dma_mapped = 0; for (i = 0, j = 0; i < npages; ++j) { struct page *page = hmm_pfn_to_page(pfns[i]); @@ -1466,7 +1357,7 @@ map_pages: } pagemap = page_pgmap(page); - dpagemap = zdd->devmem_allocation->dpagemap; + dpagemap = drm_pagemap_page_to_dpagemap(page); if (drm_WARN_ON(gpusvm->drm, !dpagemap)) { /* * Raced. This is not supposed to happen @@ -1490,7 +1381,7 @@ map_pages: } else { dma_addr_t addr; - if (is_zone_device_page(page) || zdd) { + if (is_zone_device_page(page) || pagemap) { err = -EOPNOTSUPP; goto err_unmap; } @@ -1518,7 +1409,7 @@ map_pages: flags.has_dma_mapping = true; } - if (zdd) { + if (pagemap) { flags.has_devmem_pages = true; range->dpagemap = dpagemap; } @@ -1546,6 +1437,7 @@ EXPORT_SYMBOL_GPL(drm_gpusvm_range_get_pages); /** * drm_gpusvm_range_unmap_pages() - Unmap pages associated with a GPU SVM range + * drm_gpusvm_range_evict() - Evict GPU SVM range * @gpusvm: Pointer to the GPU SVM structure * @range: Pointer to the GPU SVM range structure * @ctx: GPU SVM context @@ -1576,562 +1468,11 @@ void drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm, EXPORT_SYMBOL_GPL(drm_gpusvm_range_unmap_pages); /** - * drm_gpusvm_migration_unlock_put_page() - Put a migration page - * @page: Pointer to the page to put - * - * This function unlocks and puts a page. - */ -static void drm_gpusvm_migration_unlock_put_page(struct page *page) -{ - unlock_page(page); - put_page(page); -} - -/** - * drm_gpusvm_migration_unlock_put_pages() - Put migration pages - * @npages: Number of pages - * @migrate_pfn: Array of migrate page frame numbers - * - * This function unlocks and puts an array of pages. - */ -static void drm_gpusvm_migration_unlock_put_pages(unsigned long npages, - unsigned long *migrate_pfn) -{ - unsigned long i; - - for (i = 0; i < npages; ++i) { - struct page *page; - - if (!migrate_pfn[i]) - continue; - - page = migrate_pfn_to_page(migrate_pfn[i]); - drm_gpusvm_migration_unlock_put_page(page); - migrate_pfn[i] = 0; - } -} - -/** - * drm_gpusvm_get_devmem_page() - Get a reference to a device memory page - * @page: Pointer to the page - * @zdd: Pointer to the GPU SVM zone device data - * - * This function associates the given page with the specified GPU SVM zone - * device data and initializes it for zone device usage. - */ -static void drm_gpusvm_get_devmem_page(struct page *page, - struct drm_gpusvm_zdd *zdd) -{ - page->zone_device_data = drm_gpusvm_zdd_get(zdd); - zone_device_page_init(page); -} - -/** - * drm_gpusvm_migrate_map_pages() - Map migration pages for GPU SVM migration - * @dev: The device for which the pages are being mapped - * @dma_addr: Array to store DMA addresses corresponding to mapped pages - * @migrate_pfn: Array of migrate page frame numbers to map - * @npages: Number of pages to map - * @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL) - * - * This function maps pages of memory for migration usage in GPU SVM. It - * iterates over each page frame number provided in @migrate_pfn, maps the - * corresponding page, and stores the DMA address in the provided @dma_addr - * array. - * - * Return: 0 on success, -EFAULT if an error occurs during mapping. - */ -static int drm_gpusvm_migrate_map_pages(struct device *dev, - dma_addr_t *dma_addr, - unsigned long *migrate_pfn, - unsigned long npages, - enum dma_data_direction dir) -{ - unsigned long i; - - for (i = 0; i < npages; ++i) { - struct page *page = migrate_pfn_to_page(migrate_pfn[i]); - - if (!page) - continue; - - if (WARN_ON_ONCE(is_zone_device_page(page))) - return -EFAULT; - - dma_addr[i] = dma_map_page(dev, page, 0, PAGE_SIZE, dir); - if (dma_mapping_error(dev, dma_addr[i])) - return -EFAULT; - } - - return 0; -} - -/** - * drm_gpusvm_migrate_unmap_pages() - Unmap pages previously mapped for GPU SVM migration - * @dev: The device for which the pages were mapped - * @dma_addr: Array of DMA addresses corresponding to mapped pages - * @npages: Number of pages to unmap - * @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL) - * - * This function unmaps previously mapped pages of memory for GPU Shared Virtual - * Memory (SVM). It iterates over each DMA address provided in @dma_addr, checks - * if it's valid and not already unmapped, and unmaps the corresponding page. - */ -static void drm_gpusvm_migrate_unmap_pages(struct device *dev, - dma_addr_t *dma_addr, - unsigned long npages, - enum dma_data_direction dir) -{ - unsigned long i; - - for (i = 0; i < npages; ++i) { - if (!dma_addr[i] || dma_mapping_error(dev, dma_addr[i])) - continue; - - dma_unmap_page(dev, dma_addr[i], PAGE_SIZE, dir); - } -} - -/** - * drm_gpusvm_migrate_to_devmem() - Migrate GPU SVM range to device memory + * drm_gpusvm_range_evict() - Evict GPU SVM range * @gpusvm: Pointer to the GPU SVM structure - * @range: Pointer to the GPU SVM range structure - * @devmem_allocation: Pointer to the device memory allocation. The caller - * should hold a reference to the device memory allocation, - * which should be dropped via ops->devmem_release or upon - * the failure of this function. - * @ctx: GPU SVM context - * - * This function migrates the specified GPU SVM range to device memory. It - * performs the necessary setup and invokes the driver-specific operations for - * migration to device memory. Upon successful return, @devmem_allocation can - * safely reference @range until ops->devmem_release is called which only upon - * successful return. Expected to be called while holding the mmap lock in read - * mode. - * - * Return: 0 on success, negative error code on failure. - */ -int drm_gpusvm_migrate_to_devmem(struct drm_gpusvm *gpusvm, - struct drm_gpusvm_range *range, - struct drm_gpusvm_devmem *devmem_allocation, - const struct drm_gpusvm_ctx *ctx) -{ - const struct drm_gpusvm_devmem_ops *ops = devmem_allocation->ops; - unsigned long start = drm_gpusvm_range_start(range), - end = drm_gpusvm_range_end(range); - struct migrate_vma migrate = { - .start = start, - .end = end, - .pgmap_owner = gpusvm->device_private_page_owner, - .flags = MIGRATE_VMA_SELECT_SYSTEM, - }; - struct mm_struct *mm = gpusvm->mm; - unsigned long i, npages = npages_in_range(start, end); - struct vm_area_struct *vas; - struct drm_gpusvm_zdd *zdd = NULL; - struct page **pages; - dma_addr_t *dma_addr; - void *buf; - int err; - - mmap_assert_locked(gpusvm->mm); - - if (!range->flags.migrate_devmem) - return -EINVAL; - - if (!ops->populate_devmem_pfn || !ops->copy_to_devmem || - !ops->copy_to_ram) - return -EOPNOTSUPP; - - vas = vma_lookup(mm, start); - if (!vas) { - err = -ENOENT; - goto err_out; - } - - if (end > vas->vm_end || start < vas->vm_start) { - err = -EINVAL; - goto err_out; - } - - if (!vma_is_anonymous(vas)) { - err = -EBUSY; - goto err_out; - } - - buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*dma_addr) + - sizeof(*pages), GFP_KERNEL); - if (!buf) { - err = -ENOMEM; - goto err_out; - } - dma_addr = buf + (2 * sizeof(*migrate.src) * npages); - pages = buf + (2 * sizeof(*migrate.src) + sizeof(*dma_addr)) * npages; - - zdd = drm_gpusvm_zdd_alloc(gpusvm->device_private_page_owner); - if (!zdd) { - err = -ENOMEM; - goto err_free; - } - - migrate.vma = vas; - migrate.src = buf; - migrate.dst = migrate.src + npages; - - err = migrate_vma_setup(&migrate); - if (err) - goto err_free; - - if (!migrate.cpages) { - err = -EFAULT; - goto err_free; - } - - if (migrate.cpages != npages) { - err = -EBUSY; - goto err_finalize; - } - - err = ops->populate_devmem_pfn(devmem_allocation, npages, migrate.dst); - if (err) - goto err_finalize; - - err = drm_gpusvm_migrate_map_pages(devmem_allocation->dev, dma_addr, - migrate.src, npages, DMA_TO_DEVICE); - if (err) - goto err_finalize; - - for (i = 0; i < npages; ++i) { - struct page *page = pfn_to_page(migrate.dst[i]); - - pages[i] = page; - migrate.dst[i] = migrate_pfn(migrate.dst[i]); - drm_gpusvm_get_devmem_page(page, zdd); - } - - err = ops->copy_to_devmem(pages, dma_addr, npages); - if (err) - goto err_finalize; - - /* Upon success bind devmem allocation to range and zdd */ - devmem_allocation->timeslice_expiration = get_jiffies_64() + - msecs_to_jiffies(ctx->timeslice_ms); - zdd->devmem_allocation = devmem_allocation; /* Owns ref */ - -err_finalize: - if (err) - drm_gpusvm_migration_unlock_put_pages(npages, migrate.dst); - migrate_vma_pages(&migrate); - migrate_vma_finalize(&migrate); - drm_gpusvm_migrate_unmap_pages(devmem_allocation->dev, dma_addr, npages, - DMA_TO_DEVICE); -err_free: - if (zdd) - drm_gpusvm_zdd_put(zdd); - kvfree(buf); -err_out: - return err; -} -EXPORT_SYMBOL_GPL(drm_gpusvm_migrate_to_devmem); - -/** - * drm_gpusvm_migrate_populate_ram_pfn() - Populate RAM PFNs for a VM area - * @vas: Pointer to the VM area structure, can be NULL - * @fault_page: Fault page - * @npages: Number of pages to populate - * @mpages: Number of pages to migrate - * @src_mpfn: Source array of migrate PFNs - * @mpfn: Array of migrate PFNs to populate - * @addr: Start address for PFN allocation - * - * This function populates the RAM migrate page frame numbers (PFNs) for the - * specified VM area structure. It allocates and locks pages in the VM area for - * RAM usage. If vas is non-NULL use alloc_page_vma for allocation, if NULL use - * alloc_page for allocation. - * - * Return: 0 on success, negative error code on failure. - */ -static int drm_gpusvm_migrate_populate_ram_pfn(struct vm_area_struct *vas, - struct page *fault_page, - unsigned long npages, - unsigned long *mpages, - unsigned long *src_mpfn, - unsigned long *mpfn, - unsigned long addr) -{ - unsigned long i; - - for (i = 0; i < npages; ++i, addr += PAGE_SIZE) { - struct page *page, *src_page; - - if (!(src_mpfn[i] & MIGRATE_PFN_MIGRATE)) - continue; - - src_page = migrate_pfn_to_page(src_mpfn[i]); - if (!src_page) - continue; - - if (fault_page) { - if (src_page->zone_device_data != - fault_page->zone_device_data) - continue; - } - - if (vas) - page = alloc_page_vma(GFP_HIGHUSER, vas, addr); - else - page = alloc_page(GFP_HIGHUSER); - - if (!page) - goto free_pages; - - mpfn[i] = migrate_pfn(page_to_pfn(page)); - } - - for (i = 0; i < npages; ++i) { - struct page *page = migrate_pfn_to_page(mpfn[i]); - - if (!page) - continue; - - WARN_ON_ONCE(!trylock_page(page)); - ++*mpages; - } - - return 0; - -free_pages: - for (i = 0; i < npages; ++i) { - struct page *page = migrate_pfn_to_page(mpfn[i]); - - if (!page) - continue; - - put_page(page); - mpfn[i] = 0; - } - return -ENOMEM; -} - -/** - * drm_gpusvm_evict_to_ram() - Evict GPU SVM range to RAM - * @devmem_allocation: Pointer to the device memory allocation - * - * Similar to __drm_gpusvm_migrate_to_ram but does not require mmap lock and - * migration done via migrate_device_* functions. - * - * Return: 0 on success, negative error code on failure. - */ -int drm_gpusvm_evict_to_ram(struct drm_gpusvm_devmem *devmem_allocation) -{ - const struct drm_gpusvm_devmem_ops *ops = devmem_allocation->ops; - unsigned long npages, mpages = 0; - struct page **pages; - unsigned long *src, *dst; - dma_addr_t *dma_addr; - void *buf; - int i, err = 0; - unsigned int retry_count = 2; - - npages = devmem_allocation->size >> PAGE_SHIFT; - -retry: - if (!mmget_not_zero(devmem_allocation->mm)) - return -EFAULT; - - buf = kvcalloc(npages, 2 * sizeof(*src) + sizeof(*dma_addr) + - sizeof(*pages), GFP_KERNEL); - if (!buf) { - err = -ENOMEM; - goto err_out; - } - src = buf; - dst = buf + (sizeof(*src) * npages); - dma_addr = buf + (2 * sizeof(*src) * npages); - pages = buf + (2 * sizeof(*src) + sizeof(*dma_addr)) * npages; - - err = ops->populate_devmem_pfn(devmem_allocation, npages, src); - if (err) - goto err_free; - - err = migrate_device_pfns(src, npages); - if (err) - goto err_free; - - err = drm_gpusvm_migrate_populate_ram_pfn(NULL, NULL, npages, &mpages, - src, dst, 0); - if (err || !mpages) - goto err_finalize; - - err = drm_gpusvm_migrate_map_pages(devmem_allocation->dev, dma_addr, - dst, npages, DMA_FROM_DEVICE); - if (err) - goto err_finalize; - - for (i = 0; i < npages; ++i) - pages[i] = migrate_pfn_to_page(src[i]); - - err = ops->copy_to_ram(pages, dma_addr, npages); - if (err) - goto err_finalize; - -err_finalize: - if (err) - drm_gpusvm_migration_unlock_put_pages(npages, dst); - migrate_device_pages(src, dst, npages); - migrate_device_finalize(src, dst, npages); - drm_gpusvm_migrate_unmap_pages(devmem_allocation->dev, dma_addr, npages, - DMA_FROM_DEVICE); -err_free: - kvfree(buf); -err_out: - mmput_async(devmem_allocation->mm); - - if (completion_done(&devmem_allocation->detached)) - return 0; - - if (retry_count--) { - cond_resched(); - goto retry; - } - - return err ?: -EBUSY; -} -EXPORT_SYMBOL_GPL(drm_gpusvm_evict_to_ram); - -/** - * __drm_gpusvm_migrate_to_ram() - Migrate GPU SVM range to RAM (internal) - * @vas: Pointer to the VM area structure - * @device_private_page_owner: Device private pages owner - * @page: Pointer to the page for fault handling (can be NULL) - * @fault_addr: Fault address - * @size: Size of migration - * - * This internal function performs the migration of the specified GPU SVM range - * to RAM. It sets up the migration, populates + dma maps RAM PFNs, and - * invokes the driver-specific operations for migration to RAM. - * - * Return: 0 on success, negative error code on failure. - */ -static int __drm_gpusvm_migrate_to_ram(struct vm_area_struct *vas, - void *device_private_page_owner, - struct page *page, - unsigned long fault_addr, - unsigned long size) -{ - struct migrate_vma migrate = { - .vma = vas, - .pgmap_owner = device_private_page_owner, - .flags = MIGRATE_VMA_SELECT_DEVICE_PRIVATE | - MIGRATE_VMA_SELECT_DEVICE_COHERENT, - .fault_page = page, - }; - struct drm_gpusvm_zdd *zdd; - const struct drm_gpusvm_devmem_ops *ops; - struct device *dev = NULL; - unsigned long npages, mpages = 0; - struct page **pages; - dma_addr_t *dma_addr; - unsigned long start, end; - void *buf; - int i, err = 0; - - if (page) { - zdd = page->zone_device_data; - if (time_before64(get_jiffies_64(), - zdd->devmem_allocation->timeslice_expiration)) - return 0; - } - - start = ALIGN_DOWN(fault_addr, size); - end = ALIGN(fault_addr + 1, size); - - /* Corner where VMA area struct has been partially unmapped */ - if (start < vas->vm_start) - start = vas->vm_start; - if (end > vas->vm_end) - end = vas->vm_end; - - migrate.start = start; - migrate.end = end; - npages = npages_in_range(start, end); - - buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*dma_addr) + - sizeof(*pages), GFP_KERNEL); - if (!buf) { - err = -ENOMEM; - goto err_out; - } - dma_addr = buf + (2 * sizeof(*migrate.src) * npages); - pages = buf + (2 * sizeof(*migrate.src) + sizeof(*dma_addr)) * npages; - - migrate.vma = vas; - migrate.src = buf; - migrate.dst = migrate.src + npages; - - err = migrate_vma_setup(&migrate); - if (err) - goto err_free; - - /* Raced with another CPU fault, nothing to do */ - if (!migrate.cpages) - goto err_free; - - if (!page) { - for (i = 0; i < npages; ++i) { - if (!(migrate.src[i] & MIGRATE_PFN_MIGRATE)) - continue; - - page = migrate_pfn_to_page(migrate.src[i]); - break; - } - - if (!page) - goto err_finalize; - } - zdd = page->zone_device_data; - ops = zdd->devmem_allocation->ops; - dev = zdd->devmem_allocation->dev; - - err = drm_gpusvm_migrate_populate_ram_pfn(vas, page, npages, &mpages, - migrate.src, migrate.dst, - start); - if (err) - goto err_finalize; - - err = drm_gpusvm_migrate_map_pages(dev, dma_addr, migrate.dst, npages, - DMA_FROM_DEVICE); - if (err) - goto err_finalize; - - for (i = 0; i < npages; ++i) - pages[i] = migrate_pfn_to_page(migrate.src[i]); - - err = ops->copy_to_ram(pages, dma_addr, npages); - if (err) - goto err_finalize; - -err_finalize: - if (err) - drm_gpusvm_migration_unlock_put_pages(npages, migrate.dst); - migrate_vma_pages(&migrate); - migrate_vma_finalize(&migrate); - if (dev) - drm_gpusvm_migrate_unmap_pages(dev, dma_addr, npages, - DMA_FROM_DEVICE); -err_free: - kvfree(buf); -err_out: - - return err; -} - -/** - * drm_gpusvm_range_evict - Evict GPU SVM range * @range: Pointer to the GPU SVM range to be removed * - * This function evicts the specified GPU SVM range. This function will not - * evict coherent pages. + * This function evicts the specified GPU SVM range. * * Return: 0 on success, a negative error code on failure. */ @@ -2184,60 +1525,6 @@ int drm_gpusvm_range_evict(struct drm_gpusvm *gpusvm, EXPORT_SYMBOL_GPL(drm_gpusvm_range_evict); /** - * drm_gpusvm_page_free() - Put GPU SVM zone device data associated with a page - * @page: Pointer to the page - * - * This function is a callback used to put the GPU SVM zone device data - * associated with a page when it is being released. - */ -static void drm_gpusvm_page_free(struct page *page) -{ - drm_gpusvm_zdd_put(page->zone_device_data); -} - -/** - * drm_gpusvm_migrate_to_ram() - Migrate GPU SVM range to RAM (page fault handler) - * @vmf: Pointer to the fault information structure - * - * This function is a page fault handler used to migrate a GPU SVM range to RAM. - * It retrieves the GPU SVM range information from the faulting page and invokes - * the internal migration function to migrate the range back to RAM. - * - * Return: VM_FAULT_SIGBUS on failure, 0 on success. - */ -static vm_fault_t drm_gpusvm_migrate_to_ram(struct vm_fault *vmf) -{ - struct drm_gpusvm_zdd *zdd = vmf->page->zone_device_data; - int err; - - err = __drm_gpusvm_migrate_to_ram(vmf->vma, - zdd->device_private_page_owner, - vmf->page, vmf->address, - zdd->devmem_allocation->size); - - return err ? VM_FAULT_SIGBUS : 0; -} - -/* - * drm_gpusvm_pagemap_ops - Device page map operations for GPU SVM - */ -static const struct dev_pagemap_ops drm_gpusvm_pagemap_ops = { - .page_free = drm_gpusvm_page_free, - .migrate_to_ram = drm_gpusvm_migrate_to_ram, -}; - -/** - * drm_gpusvm_pagemap_ops_get() - Retrieve GPU SVM device page map operations - * - * Return: Pointer to the GPU SVM device page map operations structure. - */ -const struct dev_pagemap_ops *drm_gpusvm_pagemap_ops_get(void) -{ - return &drm_gpusvm_pagemap_ops; -} -EXPORT_SYMBOL_GPL(drm_gpusvm_pagemap_ops_get); - -/** * drm_gpusvm_has_mapping() - Check if GPU SVM has mapping for the given address range * @gpusvm: Pointer to the GPU SVM structure. * @start: Start address @@ -2281,28 +1568,5 @@ void drm_gpusvm_range_set_unmapped(struct drm_gpusvm_range *range, } EXPORT_SYMBOL_GPL(drm_gpusvm_range_set_unmapped); -/** - * drm_gpusvm_devmem_init() - Initialize a GPU SVM device memory allocation - * - * @dev: Pointer to the device structure which device memory allocation belongs to - * @mm: Pointer to the mm_struct for the address space - * @ops: Pointer to the operations structure for GPU SVM device memory - * @dpagemap: The struct drm_pagemap we're allocating from. - * @size: Size of device memory allocation - */ -void drm_gpusvm_devmem_init(struct drm_gpusvm_devmem *devmem_allocation, - struct device *dev, struct mm_struct *mm, - const struct drm_gpusvm_devmem_ops *ops, - struct drm_pagemap *dpagemap, size_t size) -{ - init_completion(&devmem_allocation->detached); - devmem_allocation->dev = dev; - devmem_allocation->mm = mm; - devmem_allocation->ops = ops; - devmem_allocation->dpagemap = dpagemap; - devmem_allocation->size = size; -} -EXPORT_SYMBOL_GPL(drm_gpusvm_devmem_init); - MODULE_DESCRIPTION("DRM GPUSVM"); MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/drm_gpuvm.c b/drivers/gpu/drm/drm_gpuvm.c index e89b932e987c..bbc7fecb6f4a 100644 --- a/drivers/gpu/drm/drm_gpuvm.c +++ b/drivers/gpu/drm/drm_gpuvm.c @@ -2300,13 +2300,13 @@ __drm_gpuvm_sm_unmap(struct drm_gpuvm *gpuvm, } /** - * drm_gpuvm_sm_map() - creates the &drm_gpuva_op split/merge steps + * drm_gpuvm_sm_map() - calls the &drm_gpuva_op split/merge steps * @gpuvm: the &drm_gpuvm representing the GPU VA space + * @priv: pointer to a driver private data structure * @req_addr: the start address of the new mapping * @req_range: the range of the new mapping * @req_obj: the &drm_gem_object to map * @req_offset: the offset within the &drm_gem_object - * @priv: pointer to a driver private data structure * * This function iterates the given range of the GPU VA space. It utilizes the * &drm_gpuvm_ops to call back into the driver providing the split and merge @@ -2350,7 +2350,7 @@ drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm, void *priv, EXPORT_SYMBOL_GPL(drm_gpuvm_sm_map); /** - * drm_gpuvm_sm_unmap() - creates the &drm_gpuva_ops to split on unmap + * drm_gpuvm_sm_unmap() - calls the &drm_gpuva_ops to split on unmap * @gpuvm: the &drm_gpuvm representing the GPU VA space * @priv: pointer to a driver private data structure * @req_addr: the start address of the range to unmap @@ -2391,6 +2391,132 @@ drm_gpuvm_sm_unmap(struct drm_gpuvm *gpuvm, void *priv, } EXPORT_SYMBOL_GPL(drm_gpuvm_sm_unmap); +static int +drm_gpuva_sm_step_lock(struct drm_gpuva_op *op, void *priv) +{ + struct drm_exec *exec = priv; + + switch (op->op) { + case DRM_GPUVA_OP_REMAP: + if (op->remap.unmap->va->gem.obj) + return drm_exec_lock_obj(exec, op->remap.unmap->va->gem.obj); + return 0; + case DRM_GPUVA_OP_UNMAP: + if (op->unmap.va->gem.obj) + return drm_exec_lock_obj(exec, op->unmap.va->gem.obj); + return 0; + default: + return 0; + } +} + +static const struct drm_gpuvm_ops lock_ops = { + .sm_step_map = drm_gpuva_sm_step_lock, + .sm_step_remap = drm_gpuva_sm_step_lock, + .sm_step_unmap = drm_gpuva_sm_step_lock, +}; + +/** + * drm_gpuvm_sm_map_exec_lock() - locks the objects touched by a drm_gpuvm_sm_map() + * @gpuvm: the &drm_gpuvm representing the GPU VA space + * @exec: the &drm_exec locking context + * @num_fences: for newly mapped objects, the # of fences to reserve + * @req_addr: the start address of the range to unmap + * @req_range: the range of the mappings to unmap + * @req_obj: the &drm_gem_object to map + * @req_offset: the offset within the &drm_gem_object + * + * This function locks (drm_exec_lock_obj()) objects that will be unmapped/ + * remapped, and locks+prepares (drm_exec_prepare_object()) objects that + * will be newly mapped. + * + * The expected usage is: + * + * vm_bind { + * struct drm_exec exec; + * + * // IGNORE_DUPLICATES is required, INTERRUPTIBLE_WAIT is recommended: + * drm_exec_init(&exec, IGNORE_DUPLICATES | INTERRUPTIBLE_WAIT, 0); + * + * drm_exec_until_all_locked (&exec) { + * for_each_vm_bind_operation { + * switch (op->op) { + * case DRIVER_OP_UNMAP: + * ret = drm_gpuvm_sm_unmap_exec_lock(gpuvm, &exec, op->addr, op->range); + * break; + * case DRIVER_OP_MAP: + * ret = drm_gpuvm_sm_map_exec_lock(gpuvm, &exec, num_fences, + * op->addr, op->range, + * obj, op->obj_offset); + * break; + * } + * + * drm_exec_retry_on_contention(&exec); + * if (ret) + * return ret; + * } + * } + * } + * + * This enables all locking to be performed before the driver begins modifying + * the VM. This is safe to do in the case of overlapping DRIVER_VM_BIND_OPs, + * where an earlier op can alter the sequence of steps generated for a later + * op, because the later altered step will involve the same GEM object(s) + * already seen in the earlier locking step. For example: + * + * 1) An earlier driver DRIVER_OP_UNMAP op removes the need for a + * DRM_GPUVA_OP_REMAP/UNMAP step. This is safe because we've already + * locked the GEM object in the earlier DRIVER_OP_UNMAP op. + * + * 2) An earlier DRIVER_OP_MAP op overlaps with a later DRIVER_OP_MAP/UNMAP + * op, introducing a DRM_GPUVA_OP_REMAP/UNMAP that wouldn't have been + * required without the earlier DRIVER_OP_MAP. This is safe because we've + * already locked the GEM object in the earlier DRIVER_OP_MAP step. + * + * Returns: 0 on success or a negative error codec + */ +int +drm_gpuvm_sm_map_exec_lock(struct drm_gpuvm *gpuvm, + struct drm_exec *exec, unsigned int num_fences, + u64 req_addr, u64 req_range, + struct drm_gem_object *req_obj, u64 req_offset) +{ + if (req_obj) { + int ret = drm_exec_prepare_obj(exec, req_obj, num_fences); + if (ret) + return ret; + } + + return __drm_gpuvm_sm_map(gpuvm, &lock_ops, exec, + req_addr, req_range, + req_obj, req_offset); + +} +EXPORT_SYMBOL_GPL(drm_gpuvm_sm_map_exec_lock); + +/** + * drm_gpuvm_sm_unmap_exec_lock() - locks the objects touched by drm_gpuvm_sm_unmap() + * @gpuvm: the &drm_gpuvm representing the GPU VA space + * @exec: the &drm_exec locking context + * @req_addr: the start address of the range to unmap + * @req_range: the range of the mappings to unmap + * + * This function locks (drm_exec_lock_obj()) objects that will be unmapped/ + * remapped by drm_gpuvm_sm_unmap(). + * + * See drm_gpuvm_sm_map_exec_lock() for expected usage. + * + * Returns: 0 on success or a negative error code + */ +int +drm_gpuvm_sm_unmap_exec_lock(struct drm_gpuvm *gpuvm, struct drm_exec *exec, + u64 req_addr, u64 req_range) +{ + return __drm_gpuvm_sm_unmap(gpuvm, &lock_ops, exec, + req_addr, req_range); +} +EXPORT_SYMBOL_GPL(drm_gpuvm_sm_unmap_exec_lock); + static struct drm_gpuva_op * gpuva_op_alloc(struct drm_gpuvm *gpuvm) { diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h index 442eb31351dd..9078504e789c 100644 --- a/drivers/gpu/drm/drm_internal.h +++ b/drivers/gpu/drm/drm_internal.h @@ -182,8 +182,7 @@ void drm_gem_vunmap_locked(struct drm_gem_object *obj, struct iosys_map *map); #if defined(CONFIG_DEBUG_FS) void drm_debugfs_dev_fini(struct drm_device *dev); void drm_debugfs_dev_register(struct drm_device *dev); -int drm_debugfs_register(struct drm_minor *minor, int minor_id, - struct dentry *root); +int drm_debugfs_register(struct drm_minor *minor, int minor_id); void drm_debugfs_unregister(struct drm_minor *minor); void drm_debugfs_connector_add(struct drm_connector *connector); void drm_debugfs_connector_remove(struct drm_connector *connector); @@ -201,8 +200,7 @@ static inline void drm_debugfs_dev_register(struct drm_device *dev) { } -static inline int drm_debugfs_register(struct drm_minor *minor, int minor_id, - struct dentry *root) +static inline int drm_debugfs_register(struct drm_minor *minor, int minor_id) { return 0; } diff --git a/drivers/gpu/drm/drm_mipi_dbi.c b/drivers/gpu/drm/drm_mipi_dbi.c index ba4be6be5d28..e33c78fc8fbd 100644 --- a/drivers/gpu/drm/drm_mipi_dbi.c +++ b/drivers/gpu/drm/drm_mipi_dbi.c @@ -230,7 +230,13 @@ int mipi_dbi_buf_copy(void *dst, struct iosys_map *src, struct drm_framebuffer * case DRM_FORMAT_XRGB8888: switch (dbidev->pixel_format) { case DRM_FORMAT_RGB565: - drm_fb_xrgb8888_to_rgb565(&dst_map, NULL, src, fb, clip, fmtcnv_state, swap); + if (swap) { + drm_fb_xrgb8888_to_rgb565be(&dst_map, NULL, src, fb, clip, + fmtcnv_state); + } else { + drm_fb_xrgb8888_to_rgb565(&dst_map, NULL, src, fb, clip, + fmtcnv_state); + } break; case DRM_FORMAT_RGB888: drm_fb_xrgb8888_to_rgb888(&dst_map, NULL, src, fb, clip, fmtcnv_state); diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c new file mode 100644 index 000000000000..1da55322af12 --- /dev/null +++ b/drivers/gpu/drm/drm_pagemap.c @@ -0,0 +1,838 @@ +// SPDX-License-Identifier: GPL-2.0-only OR MIT +/* + * Copyright © 2024-2025 Intel Corporation + */ + +#include <linux/dma-mapping.h> +#include <linux/migrate.h> +#include <linux/pagemap.h> +#include <drm/drm_drv.h> +#include <drm/drm_pagemap.h> + +/** + * DOC: Overview + * + * The DRM pagemap layer is intended to augment the dev_pagemap functionality by + * providing a way to populate a struct mm_struct virtual range with device + * private pages and to provide helpers to abstract device memory allocations, + * to migrate memory back and forth between device memory and system RAM and + * to handle access (and in the future migration) between devices implementing + * a fast interconnect that is not necessarily visible to the rest of the + * system. + * + * Typically the DRM pagemap receives requests from one or more DRM GPU SVM + * instances to populate struct mm_struct virtual ranges with memory, and the + * migration is best effort only and may thus fail. The implementation should + * also handle device unbinding by blocking (return an -ENODEV) error for new + * population requests and after that migrate all device pages to system ram. + */ + +/** + * DOC: Migration + * + * Migration granularity typically follows the GPU SVM range requests, but + * if there are clashes, due to races or due to the fact that multiple GPU + * SVM instances have different views of the ranges used, and because of that + * parts of a requested range is already present in the requested device memory, + * the implementation has a variety of options. It can fail and it can choose + * to populate only the part of the range that isn't already in device memory, + * and it can evict the range to system before trying to migrate. Ideally an + * implementation would just try to migrate the missing part of the range and + * allocate just enough memory to do so. + * + * When migrating to system memory as a response to a cpu fault or a device + * memory eviction request, currently a full device memory allocation is + * migrated back to system. Moving forward this might need improvement for + * situations where a single page needs bouncing between system memory and + * device memory due to, for example, atomic operations. + * + * Key DRM pagemap components: + * + * - Device Memory Allocations: + * Embedded structure containing enough information for the drm_pagemap to + * migrate to / from device memory. + * + * - Device Memory Operations: + * Define the interface for driver-specific device memory operations + * release memory, populate pfns, and copy to / from device memory. + */ + +/** + * struct drm_pagemap_zdd - GPU SVM zone device data + * + * @refcount: Reference count for the zdd + * @devmem_allocation: device memory allocation + * @device_private_page_owner: Device private pages owner + * + * This structure serves as a generic wrapper installed in + * page->zone_device_data. It provides infrastructure for looking up a device + * memory allocation upon CPU page fault and asynchronously releasing device + * memory once the CPU has no page references. Asynchronous release is useful + * because CPU page references can be dropped in IRQ contexts, while releasing + * device memory likely requires sleeping locks. + */ +struct drm_pagemap_zdd { + struct kref refcount; + struct drm_pagemap_devmem *devmem_allocation; + void *device_private_page_owner; +}; + +/** + * drm_pagemap_zdd_alloc() - Allocate a zdd structure. + * @device_private_page_owner: Device private pages owner + * + * This function allocates and initializes a new zdd structure. It sets up the + * reference count and initializes the destroy work. + * + * Return: Pointer to the allocated zdd on success, ERR_PTR() on failure. + */ +static struct drm_pagemap_zdd * +drm_pagemap_zdd_alloc(void *device_private_page_owner) +{ + struct drm_pagemap_zdd *zdd; + + zdd = kmalloc(sizeof(*zdd), GFP_KERNEL); + if (!zdd) + return NULL; + + kref_init(&zdd->refcount); + zdd->devmem_allocation = NULL; + zdd->device_private_page_owner = device_private_page_owner; + + return zdd; +} + +/** + * drm_pagemap_zdd_get() - Get a reference to a zdd structure. + * @zdd: Pointer to the zdd structure. + * + * This function increments the reference count of the provided zdd structure. + * + * Return: Pointer to the zdd structure. + */ +static struct drm_pagemap_zdd *drm_pagemap_zdd_get(struct drm_pagemap_zdd *zdd) +{ + kref_get(&zdd->refcount); + return zdd; +} + +/** + * drm_pagemap_zdd_destroy() - Destroy a zdd structure. + * @ref: Pointer to the reference count structure. + * + * This function queues the destroy_work of the zdd for asynchronous destruction. + */ +static void drm_pagemap_zdd_destroy(struct kref *ref) +{ + struct drm_pagemap_zdd *zdd = + container_of(ref, struct drm_pagemap_zdd, refcount); + struct drm_pagemap_devmem *devmem = zdd->devmem_allocation; + + if (devmem) { + complete_all(&devmem->detached); + if (devmem->ops->devmem_release) + devmem->ops->devmem_release(devmem); + } + kfree(zdd); +} + +/** + * drm_pagemap_zdd_put() - Put a zdd reference. + * @zdd: Pointer to the zdd structure. + * + * This function decrements the reference count of the provided zdd structure + * and schedules its destruction if the count drops to zero. + */ +static void drm_pagemap_zdd_put(struct drm_pagemap_zdd *zdd) +{ + kref_put(&zdd->refcount, drm_pagemap_zdd_destroy); +} + +/** + * drm_pagemap_migration_unlock_put_page() - Put a migration page + * @page: Pointer to the page to put + * + * This function unlocks and puts a page. + */ +static void drm_pagemap_migration_unlock_put_page(struct page *page) +{ + unlock_page(page); + put_page(page); +} + +/** + * drm_pagemap_migration_unlock_put_pages() - Put migration pages + * @npages: Number of pages + * @migrate_pfn: Array of migrate page frame numbers + * + * This function unlocks and puts an array of pages. + */ +static void drm_pagemap_migration_unlock_put_pages(unsigned long npages, + unsigned long *migrate_pfn) +{ + unsigned long i; + + for (i = 0; i < npages; ++i) { + struct page *page; + + if (!migrate_pfn[i]) + continue; + + page = migrate_pfn_to_page(migrate_pfn[i]); + drm_pagemap_migration_unlock_put_page(page); + migrate_pfn[i] = 0; + } +} + +/** + * drm_pagemap_get_devmem_page() - Get a reference to a device memory page + * @page: Pointer to the page + * @zdd: Pointer to the GPU SVM zone device data + * + * This function associates the given page with the specified GPU SVM zone + * device data and initializes it for zone device usage. + */ +static void drm_pagemap_get_devmem_page(struct page *page, + struct drm_pagemap_zdd *zdd) +{ + page->zone_device_data = drm_pagemap_zdd_get(zdd); + zone_device_page_init(page); +} + +/** + * drm_pagemap_migrate_map_pages() - Map migration pages for GPU SVM migration + * @dev: The device for which the pages are being mapped + * @dma_addr: Array to store DMA addresses corresponding to mapped pages + * @migrate_pfn: Array of migrate page frame numbers to map + * @npages: Number of pages to map + * @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL) + * + * This function maps pages of memory for migration usage in GPU SVM. It + * iterates over each page frame number provided in @migrate_pfn, maps the + * corresponding page, and stores the DMA address in the provided @dma_addr + * array. + * + * Returns: 0 on success, -EFAULT if an error occurs during mapping. + */ +static int drm_pagemap_migrate_map_pages(struct device *dev, + dma_addr_t *dma_addr, + unsigned long *migrate_pfn, + unsigned long npages, + enum dma_data_direction dir) +{ + unsigned long i; + + for (i = 0; i < npages; ++i) { + struct page *page = migrate_pfn_to_page(migrate_pfn[i]); + + if (!page) + continue; + + if (WARN_ON_ONCE(is_zone_device_page(page))) + return -EFAULT; + + dma_addr[i] = dma_map_page(dev, page, 0, PAGE_SIZE, dir); + if (dma_mapping_error(dev, dma_addr[i])) + return -EFAULT; + } + + return 0; +} + +/** + * drm_pagemap_migrate_unmap_pages() - Unmap pages previously mapped for GPU SVM migration + * @dev: The device for which the pages were mapped + * @dma_addr: Array of DMA addresses corresponding to mapped pages + * @npages: Number of pages to unmap + * @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL) + * + * This function unmaps previously mapped pages of memory for GPU Shared Virtual + * Memory (SVM). It iterates over each DMA address provided in @dma_addr, checks + * if it's valid and not already unmapped, and unmaps the corresponding page. + */ +static void drm_pagemap_migrate_unmap_pages(struct device *dev, + dma_addr_t *dma_addr, + unsigned long npages, + enum dma_data_direction dir) +{ + unsigned long i; + + for (i = 0; i < npages; ++i) { + if (!dma_addr[i] || dma_mapping_error(dev, dma_addr[i])) + continue; + + dma_unmap_page(dev, dma_addr[i], PAGE_SIZE, dir); + } +} + +static unsigned long +npages_in_range(unsigned long start, unsigned long end) +{ + return (end - start) >> PAGE_SHIFT; +} + +/** + * drm_pagemap_migrate_to_devmem() - Migrate a struct mm_struct range to device memory + * @devmem_allocation: The device memory allocation to migrate to. + * The caller should hold a reference to the device memory allocation, + * and the reference is consumed by this function unless it returns with + * an error. + * @mm: Pointer to the struct mm_struct. + * @start: Start of the virtual address range to migrate. + * @end: End of the virtual address range to migrate. + * @timeslice_ms: The time requested for the migrated pagemap pages to + * be present in @mm before being allowed to be migrated back. + * @pgmap_owner: Not used currently, since only system memory is considered. + * + * This function migrates the specified virtual address range to device memory. + * It performs the necessary setup and invokes the driver-specific operations for + * migration to device memory. Expected to be called while holding the mmap lock in + * at least read mode. + * + * Note: The @timeslice_ms parameter can typically be used to force data to + * remain in pagemap pages long enough for a GPU to perform a task and to prevent + * a migration livelock. One alternative would be for the GPU driver to block + * in a mmu_notifier for the specified amount of time, but adding the + * functionality to the pagemap is likely nicer to the system as a whole. + * + * Return: %0 on success, negative error code on failure. + */ +int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, + struct mm_struct *mm, + unsigned long start, unsigned long end, + unsigned long timeslice_ms, + void *pgmap_owner) +{ + const struct drm_pagemap_devmem_ops *ops = devmem_allocation->ops; + struct migrate_vma migrate = { + .start = start, + .end = end, + .pgmap_owner = pgmap_owner, + .flags = MIGRATE_VMA_SELECT_SYSTEM, + }; + unsigned long i, npages = npages_in_range(start, end); + struct vm_area_struct *vas; + struct drm_pagemap_zdd *zdd = NULL; + struct page **pages; + dma_addr_t *dma_addr; + void *buf; + int err; + + mmap_assert_locked(mm); + + if (!ops->populate_devmem_pfn || !ops->copy_to_devmem || + !ops->copy_to_ram) + return -EOPNOTSUPP; + + vas = vma_lookup(mm, start); + if (!vas) { + err = -ENOENT; + goto err_out; + } + + if (end > vas->vm_end || start < vas->vm_start) { + err = -EINVAL; + goto err_out; + } + + if (!vma_is_anonymous(vas)) { + err = -EBUSY; + goto err_out; + } + + buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*dma_addr) + + sizeof(*pages), GFP_KERNEL); + if (!buf) { + err = -ENOMEM; + goto err_out; + } + dma_addr = buf + (2 * sizeof(*migrate.src) * npages); + pages = buf + (2 * sizeof(*migrate.src) + sizeof(*dma_addr)) * npages; + + zdd = drm_pagemap_zdd_alloc(pgmap_owner); + if (!zdd) { + err = -ENOMEM; + goto err_free; + } + + migrate.vma = vas; + migrate.src = buf; + migrate.dst = migrate.src + npages; + + err = migrate_vma_setup(&migrate); + if (err) + goto err_free; + + if (!migrate.cpages) { + err = -EFAULT; + goto err_free; + } + + if (migrate.cpages != npages) { + err = -EBUSY; + goto err_finalize; + } + + err = ops->populate_devmem_pfn(devmem_allocation, npages, migrate.dst); + if (err) + goto err_finalize; + + err = drm_pagemap_migrate_map_pages(devmem_allocation->dev, dma_addr, + migrate.src, npages, DMA_TO_DEVICE); + if (err) + goto err_finalize; + + for (i = 0; i < npages; ++i) { + struct page *page = pfn_to_page(migrate.dst[i]); + + pages[i] = page; + migrate.dst[i] = migrate_pfn(migrate.dst[i]); + drm_pagemap_get_devmem_page(page, zdd); + } + + err = ops->copy_to_devmem(pages, dma_addr, npages); + if (err) + goto err_finalize; + + /* Upon success bind devmem allocation to range and zdd */ + devmem_allocation->timeslice_expiration = get_jiffies_64() + + msecs_to_jiffies(timeslice_ms); + zdd->devmem_allocation = devmem_allocation; /* Owns ref */ + +err_finalize: + if (err) + drm_pagemap_migration_unlock_put_pages(npages, migrate.dst); + migrate_vma_pages(&migrate); + migrate_vma_finalize(&migrate); + drm_pagemap_migrate_unmap_pages(devmem_allocation->dev, dma_addr, npages, + DMA_TO_DEVICE); +err_free: + if (zdd) + drm_pagemap_zdd_put(zdd); + kvfree(buf); +err_out: + return err; +} +EXPORT_SYMBOL_GPL(drm_pagemap_migrate_to_devmem); + +/** + * drm_pagemap_migrate_populate_ram_pfn() - Populate RAM PFNs for a VM area + * @vas: Pointer to the VM area structure, can be NULL + * @fault_page: Fault page + * @npages: Number of pages to populate + * @mpages: Number of pages to migrate + * @src_mpfn: Source array of migrate PFNs + * @mpfn: Array of migrate PFNs to populate + * @addr: Start address for PFN allocation + * + * This function populates the RAM migrate page frame numbers (PFNs) for the + * specified VM area structure. It allocates and locks pages in the VM area for + * RAM usage. If vas is non-NULL use alloc_page_vma for allocation, if NULL use + * alloc_page for allocation. + * + * Return: 0 on success, negative error code on failure. + */ +static int drm_pagemap_migrate_populate_ram_pfn(struct vm_area_struct *vas, + struct page *fault_page, + unsigned long npages, + unsigned long *mpages, + unsigned long *src_mpfn, + unsigned long *mpfn, + unsigned long addr) +{ + unsigned long i; + + for (i = 0; i < npages; ++i, addr += PAGE_SIZE) { + struct page *page, *src_page; + + if (!(src_mpfn[i] & MIGRATE_PFN_MIGRATE)) + continue; + + src_page = migrate_pfn_to_page(src_mpfn[i]); + if (!src_page) + continue; + + if (fault_page) { + if (src_page->zone_device_data != + fault_page->zone_device_data) + continue; + } + + if (vas) + page = alloc_page_vma(GFP_HIGHUSER, vas, addr); + else + page = alloc_page(GFP_HIGHUSER); + + if (!page) + goto free_pages; + + mpfn[i] = migrate_pfn(page_to_pfn(page)); + } + + for (i = 0; i < npages; ++i) { + struct page *page = migrate_pfn_to_page(mpfn[i]); + + if (!page) + continue; + + WARN_ON_ONCE(!trylock_page(page)); + ++*mpages; + } + + return 0; + +free_pages: + for (i = 0; i < npages; ++i) { + struct page *page = migrate_pfn_to_page(mpfn[i]); + + if (!page) + continue; + + put_page(page); + mpfn[i] = 0; + } + return -ENOMEM; +} + +/** + * drm_pagemap_evict_to_ram() - Evict GPU SVM range to RAM + * @devmem_allocation: Pointer to the device memory allocation + * + * Similar to __drm_pagemap_migrate_to_ram but does not require mmap lock and + * migration done via migrate_device_* functions. + * + * Return: 0 on success, negative error code on failure. + */ +int drm_pagemap_evict_to_ram(struct drm_pagemap_devmem *devmem_allocation) +{ + const struct drm_pagemap_devmem_ops *ops = devmem_allocation->ops; + unsigned long npages, mpages = 0; + struct page **pages; + unsigned long *src, *dst; + dma_addr_t *dma_addr; + void *buf; + int i, err = 0; + unsigned int retry_count = 2; + + npages = devmem_allocation->size >> PAGE_SHIFT; + +retry: + if (!mmget_not_zero(devmem_allocation->mm)) + return -EFAULT; + + buf = kvcalloc(npages, 2 * sizeof(*src) + sizeof(*dma_addr) + + sizeof(*pages), GFP_KERNEL); + if (!buf) { + err = -ENOMEM; + goto err_out; + } + src = buf; + dst = buf + (sizeof(*src) * npages); + dma_addr = buf + (2 * sizeof(*src) * npages); + pages = buf + (2 * sizeof(*src) + sizeof(*dma_addr)) * npages; + + err = ops->populate_devmem_pfn(devmem_allocation, npages, src); + if (err) + goto err_free; + + err = migrate_device_pfns(src, npages); + if (err) + goto err_free; + + err = drm_pagemap_migrate_populate_ram_pfn(NULL, NULL, npages, &mpages, + src, dst, 0); + if (err || !mpages) + goto err_finalize; + + err = drm_pagemap_migrate_map_pages(devmem_allocation->dev, dma_addr, + dst, npages, DMA_FROM_DEVICE); + if (err) + goto err_finalize; + + for (i = 0; i < npages; ++i) + pages[i] = migrate_pfn_to_page(src[i]); + + err = ops->copy_to_ram(pages, dma_addr, npages); + if (err) + goto err_finalize; + +err_finalize: + if (err) + drm_pagemap_migration_unlock_put_pages(npages, dst); + migrate_device_pages(src, dst, npages); + migrate_device_finalize(src, dst, npages); + drm_pagemap_migrate_unmap_pages(devmem_allocation->dev, dma_addr, npages, + DMA_FROM_DEVICE); +err_free: + kvfree(buf); +err_out: + mmput_async(devmem_allocation->mm); + + if (completion_done(&devmem_allocation->detached)) + return 0; + + if (retry_count--) { + cond_resched(); + goto retry; + } + + return err ?: -EBUSY; +} +EXPORT_SYMBOL_GPL(drm_pagemap_evict_to_ram); + +/** + * __drm_pagemap_migrate_to_ram() - Migrate GPU SVM range to RAM (internal) + * @vas: Pointer to the VM area structure + * @device_private_page_owner: Device private pages owner + * @page: Pointer to the page for fault handling (can be NULL) + * @fault_addr: Fault address + * @size: Size of migration + * + * This internal function performs the migration of the specified GPU SVM range + * to RAM. It sets up the migration, populates + dma maps RAM PFNs, and + * invokes the driver-specific operations for migration to RAM. + * + * Return: 0 on success, negative error code on failure. + */ +static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas, + void *device_private_page_owner, + struct page *page, + unsigned long fault_addr, + unsigned long size) +{ + struct migrate_vma migrate = { + .vma = vas, + .pgmap_owner = device_private_page_owner, + .flags = MIGRATE_VMA_SELECT_DEVICE_PRIVATE | + MIGRATE_VMA_SELECT_DEVICE_COHERENT, + .fault_page = page, + }; + struct drm_pagemap_zdd *zdd; + const struct drm_pagemap_devmem_ops *ops; + struct device *dev = NULL; + unsigned long npages, mpages = 0; + struct page **pages; + dma_addr_t *dma_addr; + unsigned long start, end; + void *buf; + int i, err = 0; + + if (page) { + zdd = page->zone_device_data; + if (time_before64(get_jiffies_64(), + zdd->devmem_allocation->timeslice_expiration)) + return 0; + } + + start = ALIGN_DOWN(fault_addr, size); + end = ALIGN(fault_addr + 1, size); + + /* Corner where VMA area struct has been partially unmapped */ + if (start < vas->vm_start) + start = vas->vm_start; + if (end > vas->vm_end) + end = vas->vm_end; + + migrate.start = start; + migrate.end = end; + npages = npages_in_range(start, end); + + buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*dma_addr) + + sizeof(*pages), GFP_KERNEL); + if (!buf) { + err = -ENOMEM; + goto err_out; + } + dma_addr = buf + (2 * sizeof(*migrate.src) * npages); + pages = buf + (2 * sizeof(*migrate.src) + sizeof(*dma_addr)) * npages; + + migrate.vma = vas; + migrate.src = buf; + migrate.dst = migrate.src + npages; + + err = migrate_vma_setup(&migrate); + if (err) + goto err_free; + + /* Raced with another CPU fault, nothing to do */ + if (!migrate.cpages) + goto err_free; + + if (!page) { + for (i = 0; i < npages; ++i) { + if (!(migrate.src[i] & MIGRATE_PFN_MIGRATE)) + continue; + + page = migrate_pfn_to_page(migrate.src[i]); + break; + } + + if (!page) + goto err_finalize; + } + zdd = page->zone_device_data; + ops = zdd->devmem_allocation->ops; + dev = zdd->devmem_allocation->dev; + + err = drm_pagemap_migrate_populate_ram_pfn(vas, page, npages, &mpages, + migrate.src, migrate.dst, + start); + if (err) + goto err_finalize; + + err = drm_pagemap_migrate_map_pages(dev, dma_addr, migrate.dst, npages, + DMA_FROM_DEVICE); + if (err) + goto err_finalize; + + for (i = 0; i < npages; ++i) + pages[i] = migrate_pfn_to_page(migrate.src[i]); + + err = ops->copy_to_ram(pages, dma_addr, npages); + if (err) + goto err_finalize; + +err_finalize: + if (err) + drm_pagemap_migration_unlock_put_pages(npages, migrate.dst); + migrate_vma_pages(&migrate); + migrate_vma_finalize(&migrate); + if (dev) + drm_pagemap_migrate_unmap_pages(dev, dma_addr, npages, + DMA_FROM_DEVICE); +err_free: + kvfree(buf); +err_out: + + return err; +} + +/** + * drm_pagemap_page_free() - Put GPU SVM zone device data associated with a page + * @page: Pointer to the page + * + * This function is a callback used to put the GPU SVM zone device data + * associated with a page when it is being released. + */ +static void drm_pagemap_page_free(struct page *page) +{ + drm_pagemap_zdd_put(page->zone_device_data); +} + +/** + * drm_pagemap_migrate_to_ram() - Migrate a virtual range to RAM (page fault handler) + * @vmf: Pointer to the fault information structure + * + * This function is a page fault handler used to migrate a virtual range + * to ram. The device memory allocation in which the device page is found is + * migrated in its entirety. + * + * Returns: + * VM_FAULT_SIGBUS on failure, 0 on success. + */ +static vm_fault_t drm_pagemap_migrate_to_ram(struct vm_fault *vmf) +{ + struct drm_pagemap_zdd *zdd = vmf->page->zone_device_data; + int err; + + err = __drm_pagemap_migrate_to_ram(vmf->vma, + zdd->device_private_page_owner, + vmf->page, vmf->address, + zdd->devmem_allocation->size); + + return err ? VM_FAULT_SIGBUS : 0; +} + +static const struct dev_pagemap_ops drm_pagemap_pagemap_ops = { + .page_free = drm_pagemap_page_free, + .migrate_to_ram = drm_pagemap_migrate_to_ram, +}; + +/** + * drm_pagemap_pagemap_ops_get() - Retrieve GPU SVM device page map operations + * + * Returns: + * Pointer to the GPU SVM device page map operations structure. + */ +const struct dev_pagemap_ops *drm_pagemap_pagemap_ops_get(void) +{ + return &drm_pagemap_pagemap_ops; +} +EXPORT_SYMBOL_GPL(drm_pagemap_pagemap_ops_get); + +/** + * drm_pagemap_devmem_init() - Initialize a drm_pagemap device memory allocation + * + * @devmem_allocation: The struct drm_pagemap_devmem to initialize. + * @dev: Pointer to the device structure which device memory allocation belongs to + * @mm: Pointer to the mm_struct for the address space + * @ops: Pointer to the operations structure for GPU SVM device memory + * @dpagemap: The struct drm_pagemap we're allocating from. + * @size: Size of device memory allocation + */ +void drm_pagemap_devmem_init(struct drm_pagemap_devmem *devmem_allocation, + struct device *dev, struct mm_struct *mm, + const struct drm_pagemap_devmem_ops *ops, + struct drm_pagemap *dpagemap, size_t size) +{ + init_completion(&devmem_allocation->detached); + devmem_allocation->dev = dev; + devmem_allocation->mm = mm; + devmem_allocation->ops = ops; + devmem_allocation->dpagemap = dpagemap; + devmem_allocation->size = size; +} +EXPORT_SYMBOL_GPL(drm_pagemap_devmem_init); + +/** + * drm_pagemap_page_to_dpagemap() - Return a pointer the drm_pagemap of a page + * @page: The struct page. + * + * Return: A pointer to the struct drm_pagemap of a device private page that + * was populated from the struct drm_pagemap. If the page was *not* populated + * from a struct drm_pagemap, the result is undefined and the function call + * may result in dereferencing and invalid address. + */ +struct drm_pagemap *drm_pagemap_page_to_dpagemap(struct page *page) +{ + struct drm_pagemap_zdd *zdd = page->zone_device_data; + + return zdd->devmem_allocation->dpagemap; +} +EXPORT_SYMBOL_GPL(drm_pagemap_page_to_dpagemap); + +/** + * drm_pagemap_populate_mm() - Populate a virtual range with device memory pages + * @dpagemap: Pointer to the drm_pagemap managing the device memory + * @start: Start of the virtual range to populate. + * @end: End of the virtual range to populate. + * @mm: Pointer to the virtual address space. + * @timeslice_ms: The time requested for the migrated pagemap pages to + * be present in @mm before being allowed to be migrated back. + * + * Attempt to populate a virtual range with device memory pages, + * clearing them or migrating data from the existing pages if necessary. + * The function is best effort only, and implementations may vary + * in how hard they try to satisfy the request. + * + * Return: %0 on success, negative error code on error. If the hardware + * device was removed / unbound the function will return %-ENODEV. + */ +int drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, + unsigned long start, unsigned long end, + struct mm_struct *mm, + unsigned long timeslice_ms) +{ + int err; + + if (!mmget_not_zero(mm)) + return -EFAULT; + mmap_read_lock(mm); + err = dpagemap->ops->populate_mm(dpagemap, start, end, mm, + timeslice_ms); + mmap_read_unlock(mm); + mmput(mm); + + return err; +} +EXPORT_SYMBOL(drm_pagemap_populate_mm); diff --git a/drivers/gpu/drm/drm_panic_qr.rs b/drivers/gpu/drm/drm_panic_qr.rs index dd55b1cb764d..18492daae4b3 100644 --- a/drivers/gpu/drm/drm_panic_qr.rs +++ b/drivers/gpu/drm/drm_panic_qr.rs @@ -27,7 +27,7 @@ //! * <https://github.com/erwanvivien/fast_qr> //! * <https://github.com/bjguillot/qr> -use kernel::{prelude::*, str::CStr}; +use kernel::prelude::*; #[derive(Debug, Clone, Copy, PartialEq, Eq, Ord, PartialOrd)] struct Version(usize); diff --git a/drivers/gpu/drm/gud/gud_pipe.c b/drivers/gpu/drm/gud/gud_pipe.c index adadd526641d..8d548d08f127 100644 --- a/drivers/gpu/drm/gud/gud_pipe.c +++ b/drivers/gpu/drm/gud/gud_pipe.c @@ -188,8 +188,13 @@ retry: } else if (format->format == DRM_FORMAT_RGB332) { drm_fb_xrgb8888_to_rgb332(&dst, NULL, src, fb, rect, fmtcnv_state); } else if (format->format == DRM_FORMAT_RGB565) { - drm_fb_xrgb8888_to_rgb565(&dst, NULL, src, fb, rect, fmtcnv_state, - gud_is_big_endian()); + if (gud_is_big_endian()) { + drm_fb_xrgb8888_to_rgb565be(&dst, NULL, src, fb, rect, + fmtcnv_state); + } else { + drm_fb_xrgb8888_to_rgb565(&dst, NULL, src, fb, rect, + fmtcnv_state); + } } else if (format->format == DRM_FORMAT_RGB888) { drm_fb_xrgb8888_to_rgb888(&dst, NULL, src, fb, rect, fmtcnv_state); } else { diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c index 271b27c9de51..41228478b21c 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c +++ b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c @@ -145,7 +145,7 @@ intel_dp_aux_supports_hdr_backlight(struct intel_connector *connector) * ranges for such panels. */ if (display->params.enable_dpcd_backlight != INTEL_DP_AUX_BACKLIGHT_FORCE_INTEL && - !(connector->base.hdr_sink_metadata.hdmi_type1.metadata_type & + !(connector->base.display_info.hdr_sink_metadata.hdmi_type1.metadata_type & BIT(HDMI_STATIC_METADATA_TYPE1))) { drm_info(display->drm, "[CONNECTOR:%d:%s] Panel is missing HDR static metadata. Possible support for Intel HDR backlight interface is not used. If your backlight controls don't work try booting with i915.enable_dpcd_backlight=%d.\n", @@ -475,31 +475,6 @@ static u32 intel_dp_aux_vesa_get_backlight(struct intel_connector *connector, en return connector->panel.backlight.level; } -static int -intel_dp_aux_vesa_set_luminance(struct intel_connector *connector, u32 level) -{ - struct intel_dp *intel_dp = enc_to_intel_dp(connector->encoder); - u8 buf[3]; - int ret; - - level = level * 1000; - level &= 0xffffff; - buf[0] = (level & 0x0000ff); - buf[1] = (level & 0x00ff00) >> 8; - buf[2] = (level & 0xff0000) >> 16; - - ret = drm_dp_dpcd_write(&intel_dp->aux, DP_EDP_PANEL_TARGET_LUMINANCE_VALUE, - buf, sizeof(buf)); - if (ret != sizeof(buf)) { - drm_err(intel_dp->aux.drm_dev, - "%s: Failed to set VESA Aux Luminance: %d\n", - intel_dp->aux.name, ret); - return -EINVAL; - } else { - return 0; - } -} - static void intel_dp_aux_vesa_set_backlight(const struct drm_connector_state *conn_state, u32 level) { @@ -507,11 +482,6 @@ intel_dp_aux_vesa_set_backlight(const struct drm_connector_state *conn_state, u3 struct intel_panel *panel = &connector->panel; struct intel_dp *intel_dp = enc_to_intel_dp(connector->encoder); - if (panel->backlight.edp.vesa.luminance_control_support) { - if (!intel_dp_aux_vesa_set_luminance(connector, level)) - return; - } - if (!panel->backlight.edp.vesa.info.aux_set) { const u32 pwm_level = intel_backlight_level_to_pwm(connector, level); @@ -528,18 +498,6 @@ intel_dp_aux_vesa_enable_backlight(const struct intel_crtc_state *crtc_state, struct intel_connector *connector = to_intel_connector(conn_state->connector); struct intel_panel *panel = &connector->panel; struct intel_dp *intel_dp = enc_to_intel_dp(connector->encoder); - int ret; - - if (panel->backlight.edp.vesa.luminance_control_support) { - ret = drm_dp_dpcd_writeb(&intel_dp->aux, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, - DP_EDP_PANEL_LUMINANCE_CONTROL_ENABLE); - - if (ret == 1) - return; - - if (!intel_dp_aux_vesa_set_luminance(connector, level)) - return; - } if (!panel->backlight.edp.vesa.info.aux_enable) { u32 pwm_level; @@ -580,13 +538,41 @@ static int intel_dp_aux_vesa_setup_backlight(struct intel_connector *connector, &connector->base.display_info.luminance_range; struct intel_dp *intel_dp = intel_attached_dp(connector); struct intel_panel *panel = &connector->panel; - u16 current_level; + u32 current_level; u8 current_mode; int ret; - if (panel->backlight.edp.vesa.luminance_control_support) { + ret = drm_edp_backlight_init(&intel_dp->aux, &panel->backlight.edp.vesa.info, + luminance_range->max_luminance, + panel->vbt.backlight.pwm_freq_hz, + intel_dp->edp_dpcd, ¤t_level, ¤t_mode, + false); + if (ret < 0) + return ret; + + drm_dbg_kms(display->drm, + "[CONNECTOR:%d:%s] AUX VESA backlight enable is controlled through %s\n", + connector->base.base.id, connector->base.name, + dpcd_vs_pwm_str(panel->backlight.edp.vesa.info.aux_enable)); + drm_dbg_kms(display->drm, + "[CONNECTOR:%d:%s] AUX VESA backlight level is controlled through %s\n", + connector->base.base.id, connector->base.name, + dpcd_vs_pwm_str(panel->backlight.edp.vesa.info.aux_set)); + + if (!panel->backlight.edp.vesa.info.aux_set || + !panel->backlight.edp.vesa.info.aux_enable) { + ret = panel->backlight.pwm_funcs->setup(connector, pipe); + if (ret < 0) { + drm_err(display->drm, + "[CONNECTOR:%d:%s] Failed to setup PWM backlight controls for eDP backlight: %d\n", + connector->base.base.id, connector->base.name, ret); + return ret; + } + } + + if (panel->backlight.edp.vesa.info.luminance_set) { if (luminance_range->max_luminance) { - panel->backlight.max = luminance_range->max_luminance; + panel->backlight.max = panel->backlight.edp.vesa.info.max; panel->backlight.min = luminance_range->min_luminance; } else { panel->backlight.max = 512; @@ -597,54 +583,26 @@ static int intel_dp_aux_vesa_setup_backlight(struct intel_connector *connector, drm_dbg_kms(display->drm, "[CONNECTOR:%d:%s] AUX VESA Nits backlight level is controlled through DPCD\n", connector->base.base.id, connector->base.name); - } else { - ret = drm_edp_backlight_init(&intel_dp->aux, &panel->backlight.edp.vesa.info, - panel->vbt.backlight.pwm_freq_hz, intel_dp->edp_dpcd, - ¤t_level, ¤t_mode); - if (ret < 0) - return ret; - - drm_dbg_kms(display->drm, - "[CONNECTOR:%d:%s] AUX VESA backlight enable is controlled through %s\n", - connector->base.base.id, connector->base.name, - dpcd_vs_pwm_str(panel->backlight.edp.vesa.info.aux_enable)); - drm_dbg_kms(display->drm, - "[CONNECTOR:%d:%s] AUX VESA backlight level is controlled through %s\n", - connector->base.base.id, connector->base.name, - dpcd_vs_pwm_str(panel->backlight.edp.vesa.info.aux_set)); - - if (!panel->backlight.edp.vesa.info.aux_set || - !panel->backlight.edp.vesa.info.aux_enable) { - ret = panel->backlight.pwm_funcs->setup(connector, pipe); - if (ret < 0) { - drm_err(display->drm, - "[CONNECTOR:%d:%s] Failed to setup PWM backlight controls for eDP backlight: %d\n", - connector->base.base.id, connector->base.name, ret); - return ret; - } + } else if (panel->backlight.edp.vesa.info.aux_set) { + panel->backlight.max = panel->backlight.edp.vesa.info.max; + panel->backlight.min = 0; + if (current_mode == DP_EDP_BACKLIGHT_CONTROL_MODE_DPCD) { + panel->backlight.level = current_level; + panel->backlight.enabled = panel->backlight.level != 0; + } else { + panel->backlight.level = panel->backlight.max; + panel->backlight.enabled = false; } - - if (panel->backlight.edp.vesa.info.aux_set) { - panel->backlight.max = panel->backlight.edp.vesa.info.max; - panel->backlight.min = 0; - if (current_mode == DP_EDP_BACKLIGHT_CONTROL_MODE_DPCD) { - panel->backlight.level = current_level; - panel->backlight.enabled = panel->backlight.level != 0; - } else { - panel->backlight.level = panel->backlight.max; - panel->backlight.enabled = false; - } + } else { + panel->backlight.max = panel->backlight.pwm_level_max; + panel->backlight.min = panel->backlight.pwm_level_min; + if (current_mode == DP_EDP_BACKLIGHT_CONTROL_MODE_PWM) { + panel->backlight.level = + panel->backlight.pwm_funcs->get(connector, pipe); + panel->backlight.enabled = panel->backlight.pwm_enabled; } else { - panel->backlight.max = panel->backlight.pwm_level_max; - panel->backlight.min = panel->backlight.pwm_level_min; - if (current_mode == DP_EDP_BACKLIGHT_CONTROL_MODE_PWM) { - panel->backlight.level = - panel->backlight.pwm_funcs->get(connector, pipe); - panel->backlight.enabled = panel->backlight.pwm_enabled; - } else { - panel->backlight.level = panel->backlight.max; - panel->backlight.enabled = false; - } + panel->backlight.level = panel->backlight.max; + panel->backlight.enabled = false; } } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_wait.c b/drivers/gpu/drm/i915/gem/i915_gem_wait.c index 7127e90c1a8f..991666fd9f85 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_wait.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_wait.c @@ -106,11 +106,6 @@ static void fence_set_priority(struct dma_fence *fence, rcu_read_unlock(); } -static inline bool __dma_fence_is_chain(const struct dma_fence *fence) -{ - return fence->ops == &dma_fence_chain_ops; -} - void i915_gem_fence_wait_priority(struct dma_fence *fence, const struct i915_sched_attr *attr) { @@ -126,7 +121,7 @@ void i915_gem_fence_wait_priority(struct dma_fence *fence, for (i = 0; i < array->num_fences; i++) fence_set_priority(array->fences[i], attr); - } else if (__dma_fence_is_chain(fence)) { + } else if (dma_fence_is_chain(fence)) { struct dma_fence *iter; /* The chain is ordered; if we boost the last, we boost all */ diff --git a/drivers/gpu/drm/i915/gt/intel_engine.h b/drivers/gpu/drm/i915/gt/intel_engine.h index 325da0414d94..f6a98cf1e5a5 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine.h +++ b/drivers/gpu/drm/i915/gt/intel_engine.h @@ -79,6 +79,29 @@ struct lock_class_key; #define ENGINE_WRITE(...) __ENGINE_WRITE_OP(write, __VA_ARGS__) #define ENGINE_WRITE_FW(...) __ENGINE_WRITE_OP(write_fw, __VA_ARGS__) +#define __HAS_ENGINE(engine_mask, id) ((engine_mask) & BIT(id)) +#define HAS_ENGINE(gt, id) __HAS_ENGINE((gt)->info.engine_mask, id) + +#define __ENGINE_INSTANCES_MASK(mask, first, count) ({ \ + unsigned int first__ = (first); \ + unsigned int count__ = (count); \ + ((mask) & GENMASK(first__ + count__ - 1, first__)) >> first__; \ +}) + +#define ENGINE_INSTANCES_MASK(gt, first, count) \ + __ENGINE_INSTANCES_MASK((gt)->info.engine_mask, first, count) + +#define RCS_MASK(gt) \ + ENGINE_INSTANCES_MASK(gt, RCS0, I915_MAX_RCS) +#define BCS_MASK(gt) \ + ENGINE_INSTANCES_MASK(gt, BCS0, I915_MAX_BCS) +#define VDBOX_MASK(gt) \ + ENGINE_INSTANCES_MASK(gt, VCS0, I915_MAX_VCS) +#define VEBOX_MASK(gt) \ + ENGINE_INSTANCES_MASK(gt, VECS0, I915_MAX_VECS) +#define CCS_MASK(gt) \ + ENGINE_INSTANCES_MASK(gt, CCS0, I915_MAX_CCS) + #define GEN6_RING_FAULT_REG_READ(engine__) \ intel_uncore_read((engine__)->uncore, RING_FAULT_REG(engine__)) @@ -355,4 +378,12 @@ u64 intel_clamp_preempt_timeout_ms(struct intel_engine_cs *engine, u64 value); u64 intel_clamp_stop_timeout_ms(struct intel_engine_cs *engine, u64 value); u64 intel_clamp_timeslice_duration_ms(struct intel_engine_cs *engine, u64 value); +#define rb_to_uabi_engine(rb) \ + rb_entry_safe(rb, struct intel_engine_cs, uabi_node) + +#define for_each_uabi_engine(engine__, i915__) \ + for ((engine__) = rb_to_uabi_engine(rb_first(&(i915__)->uabi_engines));\ + (engine__); \ + (engine__) = rb_to_uabi_engine(rb_next(&(engine__)->uabi_node))) + #endif /* _INTEL_RINGBUFFER_H_ */ diff --git a/drivers/gpu/drm/i915/gt/intel_gsc.c b/drivers/gpu/drm/i915/gt/intel_gsc.c index 1e925c75fb08..c43febc862dc 100644 --- a/drivers/gpu/drm/i915/gt/intel_gsc.c +++ b/drivers/gpu/drm/i915/gt/intel_gsc.c @@ -284,7 +284,7 @@ static void gsc_irq_handler(struct intel_gt *gt, unsigned int intf_id) if (gt->gsc.intf[intf_id].irq < 0) return; - ret = generic_handle_irq(gt->gsc.intf[intf_id].irq); + ret = generic_handle_irq_safe(gt->gsc.intf[intf_id].irq); if (ret) gt_err_ratelimited(gt, "error handling GSC irq: %d\n", ret); } diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c index a876a34455f1..2a6d79abf25b 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c +++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c @@ -610,7 +610,6 @@ static int ring_context_alloc(struct intel_context *ce) /* One ringbuffer to rule them all */ GEM_BUG_ON(!engine->legacy.ring); ce->ring = engine->legacy.ring; - ce->timeline = intel_timeline_get(engine->legacy.timeline); GEM_BUG_ON(ce->state); if (engine->context_size) { @@ -623,6 +622,8 @@ static int ring_context_alloc(struct intel_context *ce) ce->state = vma; } + ce->timeline = intel_timeline_get(engine->legacy.timeline); + return 0; } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c index 9df80c325fc1..f360f020d8f1 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c @@ -313,8 +313,13 @@ static u32 guc_ctl_wa_flags(struct intel_guc *guc) * * The same WA bit is used for both and 22011391025 is applicable to * all DG2. + * + * Platforms post DG2 prevent this issue in hardware by stalling + * submissions. With this flag GuC will schedule as to avoid such + * stalls. */ - if (IS_DG2(gt->i915)) + if (IS_DG2(gt->i915) || + (CCS_MASK(gt) && GRAPHICS_VER_FULL(gt->i915) >= IP_VER(12, 70))) flags |= GUC_WA_DUAL_QUEUE; /* Wa_22011802037: graphics version 11/12 */ diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 5e4c49f0d5d4..4e4e89746aa6 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -66,8 +66,6 @@ struct intel_display; struct intel_pxp; struct vlv_s0ix_state; -#define GEM_QUIRK_PIN_SWIZZLED_PAGES BIT(0) - /* Data Stolen Memory (DSM) aka "i915 stolen memory" */ struct i915_dsm { /* @@ -354,14 +352,6 @@ static inline struct intel_gt *to_gt(const struct drm_i915_private *i915) return i915->gt[0]; } -#define rb_to_uabi_engine(rb) \ - rb_entry_safe(rb, struct intel_engine_cs, uabi_node) - -#define for_each_uabi_engine(engine__, i915__) \ - for ((engine__) = rb_to_uabi_engine(rb_first(&(i915__)->uabi_engines));\ - (engine__); \ - (engine__) = rb_to_uabi_engine(rb_next(&(engine__)->uabi_node))) - #define INTEL_INFO(i915) ((i915)->__info) #define RUNTIME_INFO(i915) (&(i915)->__runtime) #define DRIVER_CAPS(i915) (&(i915)->caps) @@ -570,29 +560,6 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define IS_GEN9_LP(i915) (IS_BROXTON(i915) || IS_GEMINILAKE(i915)) #define IS_GEN9_BC(i915) (GRAPHICS_VER(i915) == 9 && !IS_GEN9_LP(i915)) -#define __HAS_ENGINE(engine_mask, id) ((engine_mask) & BIT(id)) -#define HAS_ENGINE(gt, id) __HAS_ENGINE((gt)->info.engine_mask, id) - -#define __ENGINE_INSTANCES_MASK(mask, first, count) ({ \ - unsigned int first__ = (first); \ - unsigned int count__ = (count); \ - ((mask) & GENMASK(first__ + count__ - 1, first__)) >> first__; \ -}) - -#define ENGINE_INSTANCES_MASK(gt, first, count) \ - __ENGINE_INSTANCES_MASK((gt)->info.engine_mask, first, count) - -#define RCS_MASK(gt) \ - ENGINE_INSTANCES_MASK(gt, RCS0, I915_MAX_RCS) -#define BCS_MASK(gt) \ - ENGINE_INSTANCES_MASK(gt, BCS0, I915_MAX_BCS) -#define VDBOX_MASK(gt) \ - ENGINE_INSTANCES_MASK(gt, VCS0, I915_MAX_VCS) -#define VEBOX_MASK(gt) \ - ENGINE_INSTANCES_MASK(gt, VECS0, I915_MAX_VECS) -#define CCS_MASK(gt) \ - ENGINE_INSTANCES_MASK(gt, CCS0, I915_MAX_CCS) - #define HAS_MEDIA_RATIO_MODE(i915) (INTEL_INFO(i915)->has_media_ratio_mode) /* diff --git a/drivers/gpu/drm/i915/i915_gem.h b/drivers/gpu/drm/i915/i915_gem.h index 82e9d289398c..20b3cb29cfff 100644 --- a/drivers/gpu/drm/i915/i915_gem.h +++ b/drivers/gpu/drm/i915/i915_gem.h @@ -134,4 +134,6 @@ int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file); #define I915_GEM_IDLE_TIMEOUT (HZ / 5) +#define GEM_QUIRK_PIN_SWIZZLED_PAGES BIT(0) + #endif /* __I915_GEM_H__ */ diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 21006c7f615c..b2e311f4791a 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -663,7 +663,6 @@ static const struct intel_device_info dg1_info = { DGFX_FEATURES, .__runtime.graphics.ip.rel = 10, PLATFORM(INTEL_DG1), - .require_force_probe = 1, .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2), diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c index e5a188ce3185..5bc696bfbb0f 100644 --- a/drivers/gpu/drm/i915/i915_pmu.c +++ b/drivers/gpu/drm/i915/i915_pmu.c @@ -108,11 +108,11 @@ static unsigned int config_bit(const u64 config) return other_bit(config); } -static u32 config_mask(const u64 config) +static __always_inline u32 config_mask(const u64 config) { unsigned int bit = config_bit(config); - if (__builtin_constant_p(config)) + if (__builtin_constant_p(bit)) BUILD_BUG_ON(bit > BITS_PER_TYPE(typeof_member(struct i915_pmu, enable)) - 1); @@ -121,7 +121,7 @@ static u32 config_mask(const u64 config) BITS_PER_TYPE(typeof_member(struct i915_pmu, enable)) - 1); - return BIT(config_bit(config)); + return BIT(bit); } static bool is_engine_event(struct perf_event *event) diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index 632e316f8b05..25e97031d76e 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -1607,6 +1607,26 @@ err_rpm: return err; } +int i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags) +{ + struct i915_gem_ww_ctx ww; + int err; + + i915_gem_ww_ctx_init(&ww, true); +retry: + err = i915_gem_object_lock(vma->obj, &ww); + if (!err) + err = i915_vma_pin_ww(vma, &ww, size, alignment, flags); + if (err == -EDEADLK) { + err = i915_gem_ww_ctx_backoff(&ww); + if (!err) + goto retry; + } + i915_gem_ww_ctx_fini(&ww); + + return err; +} + static void flush_idle_contexts(struct intel_gt *gt) { struct intel_engine_cs *engine; diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h index 4ae610927fa7..0f9eee6d18d2 100644 --- a/drivers/gpu/drm/i915/i915_vma.h +++ b/drivers/gpu/drm/i915/i915_vma.h @@ -289,26 +289,8 @@ int __must_check i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww, u64 size, u64 alignment, u64 flags); -static inline int __must_check -i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags) -{ - struct i915_gem_ww_ctx ww; - int err; - - i915_gem_ww_ctx_init(&ww, true); -retry: - err = i915_gem_object_lock(vma->obj, &ww); - if (!err) - err = i915_vma_pin_ww(vma, &ww, size, alignment, flags); - if (err == -EDEADLK) { - err = i915_gem_ww_ctx_backoff(&ww); - if (!err) - goto retry; - } - i915_gem_ww_ctx_fini(&ww); - - return err; -} +int __must_check +i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags); int i915_ggtt_pin(struct i915_vma *vma, struct i915_gem_ww_ctx *ww, u32 align, unsigned int flags); diff --git a/drivers/gpu/drm/i915/selftests/intel_memory_region.c b/drivers/gpu/drm/i915/selftests/intel_memory_region.c index f08f6674911e..7b856b5090f9 100644 --- a/drivers/gpu/drm/i915/selftests/intel_memory_region.c +++ b/drivers/gpu/drm/i915/selftests/intel_memory_region.c @@ -413,15 +413,8 @@ static int igt_mock_splintered_region(void *arg) close_objects(mem, &objects); - /* - * While we should be able allocate everything without any flag - * restrictions, if we consider I915_BO_ALLOC_CONTIGUOUS then we are - * actually limited to the largest power-of-two for the region size i.e - * max_order, due to the inner workings of the buddy allocator. So make - * sure that does indeed hold true. - */ - - obj = igt_object_create(mem, &objects, size, I915_BO_ALLOC_CONTIGUOUS); + obj = igt_object_create(mem, &objects, roundup_pow_of_two(size), + I915_BO_ALLOC_CONTIGUOUS); if (!IS_ERR(obj)) { pr_err("%s too large contiguous allocation was not rejected\n", __func__); @@ -429,8 +422,7 @@ static int igt_mock_splintered_region(void *arg) goto out_close; } - obj = igt_object_create(mem, &objects, rounddown_pow_of_two(size), - I915_BO_ALLOC_CONTIGUOUS); + obj = igt_object_create(mem, &objects, size, I915_BO_ALLOC_CONTIGUOUS); if (IS_ERR(obj)) { pr_err("%s largest possible contiguous allocation failed\n", __func__); diff --git a/drivers/gpu/drm/imagination/pvr_power.c b/drivers/gpu/drm/imagination/pvr_power.c index 41f5d89e78b8..65642ded051d 100644 --- a/drivers/gpu/drm/imagination/pvr_power.c +++ b/drivers/gpu/drm/imagination/pvr_power.c @@ -340,6 +340,63 @@ pvr_power_device_idle(struct device *dev) return pvr_power_is_idle(pvr_dev) ? 0 : -EBUSY; } +static int +pvr_power_clear_error(struct pvr_device *pvr_dev) +{ + struct device *dev = from_pvr_device(pvr_dev)->dev; + int err; + + /* Ensure the device state is known and nothing is happening past this point */ + pm_runtime_disable(dev); + + /* Attempt to clear the runtime PM error by setting the current state again */ + if (pm_runtime_status_suspended(dev)) + err = pm_runtime_set_suspended(dev); + else + err = pm_runtime_set_active(dev); + + if (err) { + drm_err(from_pvr_device(pvr_dev), + "%s: Failed to clear runtime PM error (new error %d)\n", + __func__, err); + } + + pm_runtime_enable(dev); + + return err; +} + +/** + * pvr_power_get_clear() - Acquire a power reference, correcting any errors + * @pvr_dev: Device pointer + * + * Attempt to acquire a power reference on the device. If the runtime PM + * is in error state, attempt to clear the error and retry. + * + * Returns: + * * 0 on success, or + * * Any error code returned by pvr_power_get() or the runtime PM API. + */ +static int +pvr_power_get_clear(struct pvr_device *pvr_dev) +{ + int err; + + err = pvr_power_get(pvr_dev); + if (err == 0) + return err; + + drm_warn(from_pvr_device(pvr_dev), + "%s: pvr_power_get returned error %d, attempting recovery\n", + __func__, err); + + err = pvr_power_clear_error(pvr_dev); + if (err) + return err; + + return pvr_power_get(pvr_dev); +} + /** * pvr_power_reset() - Reset the GPU * @pvr_dev: Device pointer @@ -364,7 +421,7 @@ pvr_power_reset(struct pvr_device *pvr_dev, bool hard_reset) * Take a power reference during the reset. This should prevent any interference with the * power state during reset. */ - WARN_ON(pvr_power_get(pvr_dev)); + WARN_ON(pvr_power_get_clear(pvr_dev)); down_write(&pvr_dev->reset_sem); diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig index 7f127e2ae442..250246f81ea9 100644 --- a/drivers/gpu/drm/msm/Kconfig +++ b/drivers/gpu/drm/msm/Kconfig @@ -15,21 +15,13 @@ config DRM_MSM select IOMMU_IO_PGTABLE select QCOM_MDT_LOADER if ARCH_QCOM select REGULATOR - select DRM_CLIENT_SELECTION - select DRM_DISPLAY_DP_AUX_BUS - select DRM_DISPLAY_DP_HELPER - select DRM_DISPLAY_HELPER - select DRM_BRIDGE_CONNECTOR select DRM_EXEC - select DRM_KMS_HELPER - select DRM_PANEL - select DRM_BRIDGE - select DRM_PANEL_BRIDGE + select DRM_GPUVM select DRM_SCHED - select FB_SYSMEM_HELPERS if DRM_FBDEV_EMULATION select SHMEM select TMPFS select QCOM_SCM + select QCOM_UBWC_CONFIG select WANT_DEV_COREDUMP select SND_SOC_HDMI_CODEC if SND_SOC select SYNC_FILE @@ -66,6 +58,22 @@ config DRM_MSM_VALIDATE_XML Validate XML files with register definitions against rules-fd schema. This option is mostly targeting DRM MSM developers. If unsure, say N. +config DRM_MSM_KMS + def_bool n + depends on DRM_MSM + select DRM_BRIDGE + select DRM_BRIDGE_CONNECTOR + select DRM_CLIENT_SELECTION + select DRM_DISPLAY_HELPER + select DRM_KMS_HELPER + select DRM_PANEL + select DRM_PANEL_BRIDGE + +config DRM_MSM_KMS_FBDEV + def_bool DRM_FBDEV_EMULATION + depends on DRM_MSM_KMS + select FB_SYSMEM_HELPERS + config DRM_MSM_MDSS bool depends on DRM_MSM @@ -74,6 +82,7 @@ config DRM_MSM_MDSS config DRM_MSM_MDP4 bool "Enable MDP4 support in MSM DRM driver" depends on DRM_MSM + select DRM_MSM_KMS default y help Compile in support for the Mobile Display Processor v4 (MDP4) in @@ -84,6 +93,7 @@ config DRM_MSM_MDP5 bool "Enable MDP5 support in MSM DRM driver" depends on DRM_MSM select DRM_MSM_MDSS + select DRM_MSM_KMS default y help Compile in support for the Mobile Display Processor v5 (MDP5) in @@ -94,6 +104,7 @@ config DRM_MSM_DPU bool "Enable DPU support in MSM DRM driver" depends on DRM_MSM select DRM_MSM_MDSS + select DRM_MSM_KMS select DRM_DISPLAY_DSC_HELPER default y help @@ -104,8 +115,11 @@ config DRM_MSM_DPU config DRM_MSM_DP bool "Enable DisplayPort support in MSM DRM driver" depends on DRM_MSM + depends on DRM_MSM_KMS select DRM_DISPLAY_HDMI_AUDIO_HELPER select RATIONAL + select DRM_DISPLAY_DP_AUX_BUS + select DRM_DISPLAY_DP_HELPER default y help Compile in support for DP driver in MSM DRM driver. DP external @@ -115,6 +129,7 @@ config DRM_MSM_DP config DRM_MSM_DSI bool "Enable DSI support in MSM DRM driver" depends on DRM_MSM + depends on DRM_MSM_KMS select DRM_PANEL select DRM_MIPI_DSI select DRM_DISPLAY_DSC_HELPER @@ -170,6 +185,7 @@ config DRM_MSM_DSI_7NM_PHY config DRM_MSM_HDMI bool "Enable HDMI support in MSM DRM driver" depends on DRM_MSM + depends on DRM_MSM_KMS default y select DRM_DISPLAY_HDMI_HELPER select DRM_DISPLAY_HDMI_STATE_HELPER diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile index 7a2ada6e2d74..0c0dfb25f01b 100644 --- a/drivers/gpu/drm/msm/Makefile +++ b/drivers/gpu/drm/msm/Makefile @@ -100,18 +100,15 @@ msm-display-$(CONFIG_DRM_MSM_DPU) += \ msm-display-$(CONFIG_DRM_MSM_MDSS) += \ msm_mdss.o \ -msm-display-y += \ +msm-display-$(CONFIG_DRM_MSM_KMS) += \ disp/mdp_format.o \ disp/mdp_kms.o \ disp/msm_disp_snapshot.o \ disp/msm_disp_snapshot_util.o \ msm-y += \ - msm_atomic.o \ - msm_atomic_tracepoints.o \ msm_debugfs.o \ msm_drv.o \ - msm_fb.o \ msm_fence.o \ msm_gem.o \ msm_gem_prime.o \ @@ -122,21 +119,24 @@ msm-y += \ msm_gpu_devfreq.o \ msm_io_utils.o \ msm_iommu.o \ - msm_kms.o \ msm_perf.o \ msm_rd.o \ msm_ringbuffer.o \ msm_submitqueue.o \ + msm_syncobj.o \ msm_gpu_tracepoints.o \ -msm-$(CONFIG_DRM_FBDEV_EMULATION) += msm_fbdev.o +msm-$(CONFIG_DRM_MSM_KMS) += \ + msm_atomic.o \ + msm_atomic_tracepoints.o \ + msm_fb.o \ + msm_kms.o \ -msm-display-$(CONFIG_DEBUG_FS) += \ - dp/dp_debug.o +msm-$(CONFIG_DRM_MSM_KMS_FBDEV) += msm_fbdev.o msm-display-$(CONFIG_DRM_MSM_DP)+= dp/dp_aux.o \ - dp/dp_catalog.o \ dp/dp_ctrl.o \ + dp/dp_debug.o \ dp/dp_display.o \ dp/dp_drm.o \ dp/dp_link.o \ @@ -159,7 +159,8 @@ msm-display-$(CONFIG_DRM_MSM_DSI_14NM_PHY) += dsi/phy/dsi_phy_14nm.o msm-display-$(CONFIG_DRM_MSM_DSI_10NM_PHY) += dsi/phy/dsi_phy_10nm.o msm-display-$(CONFIG_DRM_MSM_DSI_7NM_PHY) += dsi/phy/dsi_phy_7nm.o -msm-y += $(adreno-y) $(msm-display-y) +msm-y += $(adreno-y) +msm-$(CONFIG_DRM_MSM_KMS) += $(msm-display-y) obj-$(CONFIG_DRM_MSM) += msm.o @@ -195,6 +196,11 @@ ADRENO_HEADERS = \ generated/a4xx.xml.h \ generated/a5xx.xml.h \ generated/a6xx.xml.h \ + generated/a6xx_descriptors.xml.h \ + generated/a6xx_enums.xml.h \ + generated/a6xx_perfcntrs.xml.h \ + generated/a7xx_enums.xml.h \ + generated/a7xx_perfcntrs.xml.h \ generated/a6xx_gmu.xml.h \ generated/adreno_common.xml.h \ generated/adreno_pm4.xml.h \ diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c index 379a3d346c30..ec38db45d8a3 100644 --- a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c @@ -113,7 +113,7 @@ static int a2xx_hw_init(struct msm_gpu *gpu) uint32_t *ptr, len; int i, ret; - a2xx_gpummu_params(gpu->aspace->mmu, &pt_base, &tran_error); + a2xx_gpummu_params(to_msm_vm(gpu->vm)->mmu, &pt_base, &tran_error); DBG("%s", gpu->name); @@ -466,19 +466,18 @@ static struct msm_gpu_state *a2xx_gpu_state_get(struct msm_gpu *gpu) return state; } -static struct msm_gem_address_space * -a2xx_create_address_space(struct msm_gpu *gpu, struct platform_device *pdev) +static struct drm_gpuvm * +a2xx_create_vm(struct msm_gpu *gpu, struct platform_device *pdev) { struct msm_mmu *mmu = a2xx_gpummu_new(&pdev->dev, gpu); - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; - aspace = msm_gem_address_space_create(mmu, "gpu", SZ_16M, - 0xfff * SZ_64K); + vm = msm_gem_vm_create(gpu->dev, mmu, "gpu", SZ_16M, 0xfff * SZ_64K, true); - if (IS_ERR(aspace) && !IS_ERR(mmu)) + if (IS_ERR(vm) && !IS_ERR(mmu)) mmu->funcs->destroy(mmu); - return aspace; + return vm; } static u32 a2xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) @@ -504,7 +503,7 @@ static const struct adreno_gpu_funcs funcs = { #endif .gpu_state_get = a2xx_gpu_state_get, .gpu_state_put = adreno_gpu_state_put, - .create_address_space = a2xx_create_address_space, + .create_vm = a2xx_create_vm, .get_rptr = a2xx_get_rptr, }, }; @@ -551,14 +550,6 @@ struct msm_gpu *a2xx_gpu_init(struct drm_device *dev) else adreno_gpu->registers = a220_registers; - if (!gpu->aspace) { - dev_err(dev->dev, "No memory protection without MMU\n"); - if (!allow_vram_carveout) { - ret = -ENXIO; - goto fail; - } - } - return gpu; fail: diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c index 39641551eeb6..0407c9bc8c1b 100644 --- a/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpummu.c @@ -29,13 +29,16 @@ static void a2xx_gpummu_detach(struct msm_mmu *mmu) } static int a2xx_gpummu_map(struct msm_mmu *mmu, uint64_t iova, - struct sg_table *sgt, size_t len, int prot) + struct sg_table *sgt, size_t off, size_t len, + int prot) { struct a2xx_gpummu *gpummu = to_a2xx_gpummu(mmu); unsigned idx = (iova - GPUMMU_VA_START) / GPUMMU_PAGE_SIZE; struct sg_dma_page_iter dma_iter; unsigned prot_bits = 0; + WARN_ON(off != 0); + if (prot & IOMMU_WRITE) prot_bits |= 1; if (prot & IOMMU_READ) @@ -71,10 +74,6 @@ static int a2xx_gpummu_unmap(struct msm_mmu *mmu, uint64_t iova, size_t len) return 0; } -static void a2xx_gpummu_resume_translation(struct msm_mmu *mmu) -{ -} - static void a2xx_gpummu_destroy(struct msm_mmu *mmu) { struct a2xx_gpummu *gpummu = to_a2xx_gpummu(mmu); @@ -90,7 +89,6 @@ static const struct msm_mmu_funcs funcs = { .map = a2xx_gpummu_map, .unmap = a2xx_gpummu_unmap, .destroy = a2xx_gpummu_destroy, - .resume_translation = a2xx_gpummu_resume_translation, }; struct msm_mmu *a2xx_gpummu_new(struct device *dev, struct msm_gpu *gpu) diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c index b6df115bb567..a956cd79195e 100644 --- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c @@ -526,7 +526,7 @@ static const struct adreno_gpu_funcs funcs = { .gpu_busy = a3xx_gpu_busy, .gpu_state_get = a3xx_gpu_state_get, .gpu_state_put = adreno_gpu_state_put, - .create_address_space = adreno_create_address_space, + .create_vm = adreno_create_vm, .get_rptr = a3xx_get_rptr, }, }; @@ -581,21 +581,6 @@ struct msm_gpu *a3xx_gpu_init(struct drm_device *dev) goto fail; } - if (!gpu->aspace) { - /* TODO we think it is possible to configure the GPU to - * restrict access to VRAM carveout. But the required - * registers are unknown. For now just bail out and - * limp along with just modesetting. If it turns out - * to not be possible to restrict access, then we must - * implement a cmdstream validator. - */ - DRM_DEV_ERROR(dev->dev, "No memory protection without IOMMU\n"); - if (!allow_vram_carveout) { - ret = -ENXIO; - goto fail; - } - } - icc_path = devm_of_icc_get(&pdev->dev, "gfx-mem"); if (IS_ERR(icc_path)) { ret = PTR_ERR(icc_path); diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c index f1b18a6663f7..83f6329accba 100644 --- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c @@ -645,7 +645,7 @@ static const struct adreno_gpu_funcs funcs = { .gpu_busy = a4xx_gpu_busy, .gpu_state_get = a4xx_gpu_state_get, .gpu_state_put = adreno_gpu_state_put, - .create_address_space = adreno_create_address_space, + .create_vm = adreno_create_vm, .get_rptr = a4xx_get_rptr, }, .get_timestamp = a4xx_get_timestamp, @@ -695,21 +695,6 @@ struct msm_gpu *a4xx_gpu_init(struct drm_device *dev) adreno_gpu->uche_trap_base = 0xffff0000ffff0000ull; - if (!gpu->aspace) { - /* TODO we think it is possible to configure the GPU to - * restrict access to VRAM carveout. But the required - * registers are unknown. For now just bail out and - * limp along with just modesetting. If it turns out - * to not be possible to restrict access, then we must - * implement a cmdstream validator. - */ - DRM_DEV_ERROR(dev->dev, "No memory protection without IOMMU\n"); - if (!allow_vram_carveout) { - ret = -ENXIO; - goto fail; - } - } - icc_path = devm_of_icc_get(&pdev->dev, "gfx-mem"); if (IS_ERR(icc_path)) { ret = PTR_ERR(icc_path); diff --git a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c index 169b8fe688f8..625a4e787d8f 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c @@ -116,13 +116,13 @@ reset_set(void *data, u64 val) adreno_gpu->fw[ADRENO_FW_PFP] = NULL; if (a5xx_gpu->pm4_bo) { - msm_gem_unpin_iova(a5xx_gpu->pm4_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->pm4_bo, gpu->vm); drm_gem_object_put(a5xx_gpu->pm4_bo); a5xx_gpu->pm4_bo = NULL; } if (a5xx_gpu->pfp_bo) { - msm_gem_unpin_iova(a5xx_gpu->pfp_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->pfp_bo, gpu->vm); drm_gem_object_put(a5xx_gpu->pfp_bo); a5xx_gpu->pfp_bo = NULL; } diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 650e5bac225f..4a04dc43a8e6 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -131,6 +131,8 @@ static void a5xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) struct msm_ringbuffer *ring = submit->ring; unsigned int i, ibs = 0; + adreno_check_and_reenable_stall(adreno_gpu); + if (IS_ENABLED(CONFIG_DRM_MSM_GPU_SUDO) && submit->in_rb) { ring->cur_ctx_seqno = 0; a5xx_submit_in_rb(gpu, submit); @@ -620,7 +622,7 @@ static int a5xx_ucode_load(struct msm_gpu *gpu) a5xx_gpu->shadow = msm_gem_kernel_new(gpu->dev, sizeof(u32) * gpu->nr_rings, MSM_BO_WC | MSM_BO_MAP_PRIV, - gpu->aspace, &a5xx_gpu->shadow_bo, + gpu->vm, &a5xx_gpu->shadow_bo, &a5xx_gpu->shadow_iova); if (IS_ERR(a5xx_gpu->shadow)) @@ -833,8 +835,8 @@ static int a5xx_hw_init(struct msm_gpu *gpu) gpu_write(gpu, REG_A5XX_RBBM_AHB_CNTL2, 0x0000003F); - BUG_ON(adreno_gpu->ubwc_config.highest_bank_bit < 13); - hbb = adreno_gpu->ubwc_config.highest_bank_bit - 13; + BUG_ON(adreno_gpu->ubwc_config->highest_bank_bit < 13); + hbb = adreno_gpu->ubwc_config->highest_bank_bit - 13; gpu_write(gpu, REG_A5XX_TPL1_MODE_CNTL, hbb << 7); gpu_write(gpu, REG_A5XX_RB_MODE_CNTL, hbb << 1); @@ -1040,22 +1042,22 @@ static void a5xx_destroy(struct msm_gpu *gpu) a5xx_preempt_fini(gpu); if (a5xx_gpu->pm4_bo) { - msm_gem_unpin_iova(a5xx_gpu->pm4_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->pm4_bo, gpu->vm); drm_gem_object_put(a5xx_gpu->pm4_bo); } if (a5xx_gpu->pfp_bo) { - msm_gem_unpin_iova(a5xx_gpu->pfp_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->pfp_bo, gpu->vm); drm_gem_object_put(a5xx_gpu->pfp_bo); } if (a5xx_gpu->gpmu_bo) { - msm_gem_unpin_iova(a5xx_gpu->gpmu_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->gpmu_bo, gpu->vm); drm_gem_object_put(a5xx_gpu->gpmu_bo); } if (a5xx_gpu->shadow_bo) { - msm_gem_unpin_iova(a5xx_gpu->shadow_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->shadow_bo, gpu->vm); drm_gem_object_put(a5xx_gpu->shadow_bo); } @@ -1455,7 +1457,7 @@ static int a5xx_crashdumper_init(struct msm_gpu *gpu, struct a5xx_crashdumper *dumper) { dumper->ptr = msm_gem_kernel_new(gpu->dev, - SZ_1M, MSM_BO_WC, gpu->aspace, + SZ_1M, MSM_BO_WC, gpu->vm, &dumper->bo, &dumper->iova); if (!IS_ERR(dumper->ptr)) @@ -1555,7 +1557,7 @@ static void a5xx_gpu_state_get_hlsq_regs(struct msm_gpu *gpu, if (a5xx_crashdumper_run(gpu, &dumper)) { kfree(a5xx_state->hlsqregs); - msm_gem_kernel_put(dumper.bo, gpu->aspace); + msm_gem_kernel_put(dumper.bo, gpu->vm); return; } @@ -1563,7 +1565,7 @@ static void a5xx_gpu_state_get_hlsq_regs(struct msm_gpu *gpu, memcpy(a5xx_state->hlsqregs, dumper.ptr + (256 * SZ_1K), count * sizeof(u32)); - msm_gem_kernel_put(dumper.bo, gpu->aspace); + msm_gem_kernel_put(dumper.bo, gpu->vm); } static struct msm_gpu_state *a5xx_gpu_state_get(struct msm_gpu *gpu) @@ -1711,7 +1713,7 @@ static const struct adreno_gpu_funcs funcs = { .gpu_busy = a5xx_gpu_busy, .gpu_state_get = a5xx_gpu_state_get, .gpu_state_put = a5xx_gpu_state_put, - .create_address_space = adreno_create_address_space, + .create_vm = adreno_create_vm, .get_rptr = a5xx_get_rptr, }, .get_timestamp = a5xx_get_timestamp, @@ -1754,6 +1756,7 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev) struct msm_drm_private *priv = dev->dev_private; struct platform_device *pdev = priv->gpu_pdev; struct adreno_platform_config *config = pdev->dev.platform_data; + const struct qcom_ubwc_cfg_data *common_cfg; struct a5xx_gpu *a5xx_gpu = NULL; struct adreno_gpu *adreno_gpu; struct msm_gpu *gpu; @@ -1784,21 +1787,20 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev) return ERR_PTR(ret); } - if (gpu->aspace) - msm_mmu_set_fault_handler(gpu->aspace->mmu, gpu, a5xx_fault_handler); + msm_mmu_set_fault_handler(to_msm_vm(gpu->vm)->mmu, gpu, + a5xx_fault_handler); /* Set up the preemption specific bits and pieces for each ringbuffer */ a5xx_preempt_init(gpu); - /* Set the highest bank bit */ - if (adreno_is_a540(adreno_gpu) || adreno_is_a530(adreno_gpu)) - adreno_gpu->ubwc_config.highest_bank_bit = 15; - else - adreno_gpu->ubwc_config.highest_bank_bit = 14; + /* Inherit the common config and make some necessary fixups */ + common_cfg = qcom_ubwc_config_get_data(); + if (IS_ERR(common_cfg)) + return ERR_CAST(common_cfg); - /* a5xx only supports UBWC 1.0, these are not configurable */ - adreno_gpu->ubwc_config.macrotile_mode = 0; - adreno_gpu->ubwc_config.ubwc_swizzle = 0x7; + /* Copy the data into the internal struct to drop the const qualifier (temporarily) */ + adreno_gpu->_ubwc_config = *common_cfg; + adreno_gpu->ubwc_config = &adreno_gpu->_ubwc_config; adreno_gpu->uche_trap_base = 0x0001ffffffff0000ull; diff --git a/drivers/gpu/drm/msm/adreno/a5xx_power.c b/drivers/gpu/drm/msm/adreno/a5xx_power.c index 6b91e0bd1514..d6da7351cfbb 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_power.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_power.c @@ -363,7 +363,7 @@ void a5xx_gpmu_ucode_init(struct msm_gpu *gpu) bosize = (cmds_size + (cmds_size / TYPE4_MAX_PAYLOAD) + 1) << 2; ptr = msm_gem_kernel_new(drm, bosize, - MSM_BO_WC | MSM_BO_GPU_READONLY, gpu->aspace, + MSM_BO_WC | MSM_BO_GPU_READONLY, gpu->vm, &a5xx_gpu->gpmu_bo, &a5xx_gpu->gpmu_iova); if (IS_ERR(ptr)) return; diff --git a/drivers/gpu/drm/msm/adreno/a5xx_preempt.c b/drivers/gpu/drm/msm/adreno/a5xx_preempt.c index b5f9d40687d5..e4924b5e1c48 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_preempt.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_preempt.c @@ -255,7 +255,7 @@ static int preempt_init_ring(struct a5xx_gpu *a5xx_gpu, ptr = msm_gem_kernel_new(gpu->dev, A5XX_PREEMPT_RECORD_SIZE + A5XX_PREEMPT_COUNTER_SIZE, - MSM_BO_WC | MSM_BO_MAP_PRIV, gpu->aspace, &bo, &iova); + MSM_BO_WC | MSM_BO_MAP_PRIV, gpu->vm, &bo, &iova); if (IS_ERR(ptr)) return PTR_ERR(ptr); @@ -263,9 +263,9 @@ static int preempt_init_ring(struct a5xx_gpu *a5xx_gpu, /* The buffer to store counters needs to be unprivileged */ counters = msm_gem_kernel_new(gpu->dev, A5XX_PREEMPT_COUNTER_SIZE, - MSM_BO_WC, gpu->aspace, &counters_bo, &counters_iova); + MSM_BO_WC, gpu->vm, &counters_bo, &counters_iova); if (IS_ERR(counters)) { - msm_gem_kernel_put(bo, gpu->aspace); + msm_gem_kernel_put(bo, gpu->vm); return PTR_ERR(counters); } @@ -296,8 +296,8 @@ void a5xx_preempt_fini(struct msm_gpu *gpu) int i; for (i = 0; i < gpu->nr_rings; i++) { - msm_gem_kernel_put(a5xx_gpu->preempt_bo[i], gpu->aspace); - msm_gem_kernel_put(a5xx_gpu->preempt_counters_bo[i], gpu->aspace); + msm_gem_kernel_put(a5xx_gpu->preempt_bo[i], gpu->vm); + msm_gem_kernel_put(a5xx_gpu->preempt_counters_bo[i], gpu->vm); } } diff --git a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c index 70f7ad806c34..00e1afd46b81 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c @@ -1335,7 +1335,7 @@ static const uint32_t a7xx_pwrup_reglist_regs[] = { REG_A6XX_RB_NC_MODE_CNTL, REG_A6XX_RB_CMP_DBG_ECO_CNTL, REG_A7XX_GRAS_NC_MODE_CNTL, - REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE, + REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ENABLE, REG_A6XX_UCHE_GBIF_GX_CONFIG, REG_A6XX_UCHE_CLIENT_PF, REG_A6XX_TPL1_DBG_ECO_CNTL1, @@ -1442,6 +1442,13 @@ static const struct adreno_info a7xx_gpus[] = { .gmu_cgc_mode = 0x00020202, }, .preempt_record_size = 4192 * SZ_1K, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 59, 1 }, + { 7, 2 }, + { 232, 3 }, + { 146, 4 }, + ), }, { .chip_ids = ADRENO_CHIP_IDS(0x43051401), /* "C520v2" */ .family = ADRENO_7XX_GEN3, @@ -1474,6 +1481,45 @@ static const struct adreno_info a7xx_gpus[] = { }, }, .preempt_record_size = 3572 * SZ_1K, + }, { + .chip_ids = ADRENO_CHIP_IDS(0x43030c00), + .family = ADRENO_7XX_GEN2, + .fw = { + [ADRENO_FW_SQE] = "gen71500_sqe.fw", + [ADRENO_FW_GMU] = "gen71500_gmu.bin", + }, + .gmem = SZ_1M + SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV | + ADRENO_QUIRK_PREEMPTION, + .init = a6xx_gpu_init, + .a6xx = &(const struct a6xx_info) { + .hwcg = a740_hwcg, + .protect = &a730_protect, + .pwrup_reglist = &a7xx_pwrup_reglist, + .gmu_chipid = 0x70f0000, + .gmu_cgc_mode = 0x00020222, + .bcms = (const struct a6xx_bcm[]) { + { .name = "SH0", .buswidth = 16 }, + { .name = "MC0", .buswidth = 4 }, + { + .name = "ACV", + .fixed = true, + .perfmode = BIT(3), + .perfmode_bw = 16500000, + }, + { /* sentinel */ }, + }, + }, + .preempt_record_size = 4192 * SZ_1K, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 294, 1 }, + { 263, 2 }, + { 233, 3 }, + { 141, 4 }, + ), } }; DECLARE_ADRENO_GPULIST(a7xx); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c index 38c0f8ef85c3..28e6705c6da6 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -1259,15 +1259,17 @@ int a6xx_gmu_stop(struct a6xx_gpu *a6xx_gpu) static void a6xx_gmu_memory_free(struct a6xx_gmu *gmu) { - msm_gem_kernel_put(gmu->hfi.obj, gmu->aspace); - msm_gem_kernel_put(gmu->debug.obj, gmu->aspace); - msm_gem_kernel_put(gmu->icache.obj, gmu->aspace); - msm_gem_kernel_put(gmu->dcache.obj, gmu->aspace); - msm_gem_kernel_put(gmu->dummy.obj, gmu->aspace); - msm_gem_kernel_put(gmu->log.obj, gmu->aspace); - - gmu->aspace->mmu->funcs->detach(gmu->aspace->mmu); - msm_gem_address_space_put(gmu->aspace); + struct msm_mmu *mmu = to_msm_vm(gmu->vm)->mmu; + + msm_gem_kernel_put(gmu->hfi.obj, gmu->vm); + msm_gem_kernel_put(gmu->debug.obj, gmu->vm); + msm_gem_kernel_put(gmu->icache.obj, gmu->vm); + msm_gem_kernel_put(gmu->dcache.obj, gmu->vm); + msm_gem_kernel_put(gmu->dummy.obj, gmu->vm); + msm_gem_kernel_put(gmu->log.obj, gmu->vm); + + mmu->funcs->detach(mmu); + drm_gpuvm_put(gmu->vm); } static int a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, struct a6xx_gmu_bo *bo, @@ -1296,7 +1298,7 @@ static int a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, struct a6xx_gmu_bo *bo, if (IS_ERR(bo->obj)) return PTR_ERR(bo->obj); - ret = msm_gem_get_and_pin_iova_range(bo->obj, gmu->aspace, &bo->iova, + ret = msm_gem_get_and_pin_iova_range(bo->obj, gmu->vm, &bo->iova, range_start, range_end); if (ret) { drm_gem_object_put(bo->obj); @@ -1311,7 +1313,7 @@ static int a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, struct a6xx_gmu_bo *bo, return 0; } -static int a6xx_gmu_memory_probe(struct a6xx_gmu *gmu) +static int a6xx_gmu_memory_probe(struct drm_device *drm, struct a6xx_gmu *gmu) { struct msm_mmu *mmu; @@ -1321,9 +1323,9 @@ static int a6xx_gmu_memory_probe(struct a6xx_gmu *gmu) if (IS_ERR(mmu)) return PTR_ERR(mmu); - gmu->aspace = msm_gem_address_space_create(mmu, "gmu", 0x0, 0x80000000); - if (IS_ERR(gmu->aspace)) - return PTR_ERR(gmu->aspace); + gmu->vm = msm_gem_vm_create(drm, mmu, "gmu", 0x0, 0x80000000, true); + if (IS_ERR(gmu->vm)) + return PTR_ERR(gmu->vm); return 0; } @@ -1940,7 +1942,7 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) if (ret) goto err_put_device; - ret = a6xx_gmu_memory_probe(gmu); + ret = a6xx_gmu_memory_probe(adreno_gpu->base.dev, gmu); if (ret) goto err_put_device; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h index b2d4489b4024..d1ce11131ba6 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h @@ -62,7 +62,7 @@ struct a6xx_gmu { /* For serializing communication with the GMU: */ struct mutex lock; - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; void __iomem *mmio; void __iomem *rscc; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index bf3758f010f4..45dd5fd1c2bf 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -111,7 +111,8 @@ static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu, struct msm_ringbuffer *ring, struct msm_gem_submit *submit) { bool sysprof = refcount_read(&a6xx_gpu->base.base.sysprof_active) > 1; - struct msm_file_private *ctx = submit->queue->ctx; + struct msm_context *ctx = submit->queue->ctx; + struct drm_gpuvm *vm = msm_context_vm(submit->dev, ctx); struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; phys_addr_t ttbr; u32 asid; @@ -120,7 +121,7 @@ static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu, if (ctx->seqno == ring->cur_ctx_seqno) return; - if (msm_iommu_pagetable_params(ctx->aspace->mmu, &ttbr, &asid)) + if (msm_iommu_pagetable_params(to_msm_vm(vm)->mmu, &ttbr, &asid)) return; if (adreno_gpu->info->family >= ADRENO_7XX_GEN1) { @@ -130,6 +131,20 @@ static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu, OUT_RING(ring, lower_32_bits(rbmemptr(ring, fence))); OUT_RING(ring, upper_32_bits(rbmemptr(ring, fence))); OUT_RING(ring, submit->seqno - 1); + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_SET_THREAD_BOTH); + + /* Reset state used to synchronize BR and BV */ + OUT_PKT7(ring, CP_RESET_CONTEXT_STATE, 1); + OUT_RING(ring, + CP_RESET_CONTEXT_STATE_0_CLEAR_ON_CHIP_TS | + CP_RESET_CONTEXT_STATE_0_CLEAR_RESOURCE_TABLE | + CP_RESET_CONTEXT_STATE_0_CLEAR_BV_BR_COUNTER | + CP_RESET_CONTEXT_STATE_0_RESET_GLOBAL_LOCAL_TS); + + OUT_PKT7(ring, CP_THREAD_CONTROL, 1); + OUT_RING(ring, CP_SET_THREAD_BR); } if (!sysprof) { @@ -212,6 +227,8 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) struct msm_ringbuffer *ring = submit->ring; unsigned int i, ibs = 0; + adreno_check_and_reenable_stall(adreno_gpu); + a6xx_set_pagetable(a6xx_gpu, ring, submit); get_stats_counter(ring, REG_A6XX_RBBM_PERFCTR_CP(0), @@ -335,6 +352,8 @@ static void a7xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) struct msm_ringbuffer *ring = submit->ring; unsigned int i, ibs = 0; + adreno_check_and_reenable_stall(adreno_gpu); + /* * Toggle concurrent binning for pagetable switch and set the thread to * BR since only it can execute the pagetable switch packets. @@ -585,117 +604,118 @@ static void a6xx_set_cp_protect(struct msm_gpu *gpu) gpu_write(gpu, REG_A6XX_CP_PROTECT(protect->count_max - 1), protect->regs[i]); } -static void a6xx_calc_ubwc_config(struct adreno_gpu *gpu) +static int a6xx_calc_ubwc_config(struct adreno_gpu *gpu) { - gpu->ubwc_config.rgb565_predicator = 0; - gpu->ubwc_config.uavflagprd_inv = 0; - gpu->ubwc_config.min_acc_len = 0; - gpu->ubwc_config.ubwc_swizzle = 0x6; - gpu->ubwc_config.macrotile_mode = 0; - gpu->ubwc_config.highest_bank_bit = 15; + const struct qcom_ubwc_cfg_data *common_cfg; + struct qcom_ubwc_cfg_data *cfg = &gpu->_ubwc_config; + + /* Inherit the common config and make some necessary fixups */ + common_cfg = qcom_ubwc_config_get_data(); + if (IS_ERR(common_cfg)) + return PTR_ERR(common_cfg); + + /* Copy the data into the internal struct to drop the const qualifier (temporarily) */ + *cfg = *common_cfg; + + cfg->ubwc_swizzle = 0x6; + cfg->highest_bank_bit = 15; if (adreno_is_a610(gpu)) { - gpu->ubwc_config.highest_bank_bit = 13; - gpu->ubwc_config.min_acc_len = 1; - gpu->ubwc_config.ubwc_swizzle = 0x7; + cfg->highest_bank_bit = 13; + cfg->ubwc_swizzle = 0x7; } if (adreno_is_a618(gpu)) - gpu->ubwc_config.highest_bank_bit = 14; + cfg->highest_bank_bit = 14; if (adreno_is_a619(gpu)) /* TODO: Should be 14 but causes corruption at e.g. 1920x1200 on DP */ - gpu->ubwc_config.highest_bank_bit = 13; + cfg->highest_bank_bit = 13; if (adreno_is_a619_holi(gpu)) - gpu->ubwc_config.highest_bank_bit = 13; - - if (adreno_is_a621(gpu)) { - gpu->ubwc_config.highest_bank_bit = 13; - gpu->ubwc_config.amsbc = 1; - gpu->ubwc_config.uavflagprd_inv = 2; - } - - if (adreno_is_a623(gpu)) { - gpu->ubwc_config.highest_bank_bit = 16; - gpu->ubwc_config.amsbc = 1; - gpu->ubwc_config.rgb565_predicator = 1; - gpu->ubwc_config.uavflagprd_inv = 2; - gpu->ubwc_config.macrotile_mode = 1; - } + cfg->highest_bank_bit = 13; - if (adreno_is_a640_family(gpu)) - gpu->ubwc_config.amsbc = 1; + if (adreno_is_a621(gpu)) + cfg->highest_bank_bit = 13; - if (adreno_is_a680(gpu)) - gpu->ubwc_config.macrotile_mode = 1; + if (adreno_is_a623(gpu)) + cfg->highest_bank_bit = 16; if (adreno_is_a650(gpu) || adreno_is_a660(gpu) || adreno_is_a690(gpu) || adreno_is_a730(gpu) || adreno_is_a740_family(gpu)) { - /* TODO: get ddr type from bootloader and use 2 for LPDDR4 */ - gpu->ubwc_config.highest_bank_bit = 16; - gpu->ubwc_config.amsbc = 1; - gpu->ubwc_config.rgb565_predicator = 1; - gpu->ubwc_config.uavflagprd_inv = 2; - gpu->ubwc_config.macrotile_mode = 1; + /* TODO: get ddr type from bootloader and use 15 for LPDDR4 */ + cfg->highest_bank_bit = 16; } if (adreno_is_a663(gpu)) { - gpu->ubwc_config.highest_bank_bit = 13; - gpu->ubwc_config.amsbc = 1; - gpu->ubwc_config.rgb565_predicator = 1; - gpu->ubwc_config.uavflagprd_inv = 2; - gpu->ubwc_config.macrotile_mode = 1; - gpu->ubwc_config.ubwc_swizzle = 0x4; + cfg->highest_bank_bit = 13; + cfg->ubwc_swizzle = 0x4; } - if (adreno_is_7c3(gpu)) { - gpu->ubwc_config.highest_bank_bit = 14; - gpu->ubwc_config.amsbc = 1; - gpu->ubwc_config.uavflagprd_inv = 2; - gpu->ubwc_config.macrotile_mode = 1; - } + if (adreno_is_7c3(gpu)) + cfg->highest_bank_bit = 14; - if (adreno_is_a702(gpu)) { - gpu->ubwc_config.highest_bank_bit = 14; - gpu->ubwc_config.min_acc_len = 1; - } + if (adreno_is_a702(gpu)) + cfg->highest_bank_bit = 14; + + if (cfg->highest_bank_bit != common_cfg->highest_bank_bit) + DRM_WARN_ONCE("Inconclusive highest_bank_bit value: %u (GPU) vs %u (UBWC_CFG)\n", + cfg->highest_bank_bit, common_cfg->highest_bank_bit); + + if (cfg->ubwc_swizzle != common_cfg->ubwc_swizzle) + DRM_WARN_ONCE("Inconclusive ubwc_swizzle value: %u (GPU) vs %u (UBWC_CFG)\n", + cfg->ubwc_swizzle, common_cfg->ubwc_swizzle); + + gpu->ubwc_config = &gpu->_ubwc_config; + + return 0; } static void a6xx_set_ubwc_config(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + const struct qcom_ubwc_cfg_data *cfg = adreno_gpu->ubwc_config; /* * We subtract 13 from the highest bank bit (13 is the minimum value * allowed by hw) and write the lowest two bits of the remaining value * as hbb_lo and the one above it as hbb_hi to the hardware. */ - BUG_ON(adreno_gpu->ubwc_config.highest_bank_bit < 13); - u32 hbb = adreno_gpu->ubwc_config.highest_bank_bit - 13; + BUG_ON(cfg->highest_bank_bit < 13); + u32 hbb = cfg->highest_bank_bit - 13; + bool rgb565_predicator = cfg->ubwc_enc_version >= UBWC_4_0; + u32 level2_swizzling_dis = !(cfg->ubwc_swizzle & UBWC_SWIZZLE_ENABLE_LVL2); + bool ubwc_mode = qcom_ubwc_get_ubwc_mode(cfg); + bool amsbc = cfg->ubwc_enc_version >= UBWC_3_0; + bool min_acc_len_64b = false; + u8 uavflagprd_inv = 0; u32 hbb_hi = hbb >> 2; u32 hbb_lo = hbb & 3; - u32 ubwc_mode = adreno_gpu->ubwc_config.ubwc_swizzle & 1; - u32 level2_swizzling_dis = !(adreno_gpu->ubwc_config.ubwc_swizzle & 2); + + if (adreno_is_a650_family(adreno_gpu) || adreno_is_a7xx(adreno_gpu)) + uavflagprd_inv = 2; + + if (adreno_is_a610(adreno_gpu) || adreno_is_a702(adreno_gpu)) + min_acc_len_64b = true; gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, level2_swizzling_dis << 12 | - adreno_gpu->ubwc_config.rgb565_predicator << 11 | - hbb_hi << 10 | adreno_gpu->ubwc_config.amsbc << 4 | - adreno_gpu->ubwc_config.min_acc_len << 3 | + rgb565_predicator << 11 | + hbb_hi << 10 | amsbc << 4 | + min_acc_len_64b << 3 | hbb_lo << 1 | ubwc_mode); gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, level2_swizzling_dis << 6 | hbb_hi << 4 | - adreno_gpu->ubwc_config.min_acc_len << 3 | + min_acc_len_64b << 3 | hbb_lo << 1 | ubwc_mode); gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, level2_swizzling_dis << 12 | hbb_hi << 10 | - adreno_gpu->ubwc_config.uavflagprd_inv << 4 | - adreno_gpu->ubwc_config.min_acc_len << 3 | + uavflagprd_inv << 4 | + min_acc_len_64b << 3 | hbb_lo << 1 | ubwc_mode); if (adreno_is_a7xx(adreno_gpu)) @@ -703,10 +723,10 @@ static void a6xx_set_ubwc_config(struct msm_gpu *gpu) FIELD_PREP(GENMASK(8, 5), hbb_lo)); gpu_write(gpu, REG_A6XX_UCHE_MODE_CNTL, - adreno_gpu->ubwc_config.min_acc_len << 23 | hbb_lo << 21); + min_acc_len_64b << 23 | hbb_lo << 21); gpu_write(gpu, REG_A6XX_RBBM_NC_MODE_CNTL, - adreno_gpu->ubwc_config.macrotile_mode); + cfg->macrotile_mode); } static void a7xx_patch_pwrup_reglist(struct msm_gpu *gpu) @@ -952,7 +972,7 @@ static int a6xx_ucode_load(struct msm_gpu *gpu) msm_gem_object_set_name(a6xx_gpu->sqe_bo, "sqefw"); if (!a6xx_ucode_check_version(a6xx_gpu, a6xx_gpu->sqe_bo)) { - msm_gem_unpin_iova(a6xx_gpu->sqe_bo, gpu->aspace); + msm_gem_unpin_iova(a6xx_gpu->sqe_bo, gpu->vm); drm_gem_object_put(a6xx_gpu->sqe_bo); a6xx_gpu->sqe_bo = NULL; @@ -969,7 +989,7 @@ static int a6xx_ucode_load(struct msm_gpu *gpu) a6xx_gpu->shadow = msm_gem_kernel_new(gpu->dev, sizeof(u32) * gpu->nr_rings, MSM_BO_WC | MSM_BO_MAP_PRIV, - gpu->aspace, &a6xx_gpu->shadow_bo, + gpu->vm, &a6xx_gpu->shadow_bo, &a6xx_gpu->shadow_iova); if (IS_ERR(a6xx_gpu->shadow)) @@ -980,7 +1000,7 @@ static int a6xx_ucode_load(struct msm_gpu *gpu) a6xx_gpu->pwrup_reglist_ptr = msm_gem_kernel_new(gpu->dev, PAGE_SIZE, MSM_BO_WC | MSM_BO_MAP_PRIV, - gpu->aspace, &a6xx_gpu->pwrup_reglist_bo, + gpu->vm, &a6xx_gpu->pwrup_reglist_bo, &a6xx_gpu->pwrup_reglist_iova); if (IS_ERR(a6xx_gpu->pwrup_reglist_ptr)) @@ -2193,12 +2213,12 @@ static void a6xx_destroy(struct msm_gpu *gpu) struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); if (a6xx_gpu->sqe_bo) { - msm_gem_unpin_iova(a6xx_gpu->sqe_bo, gpu->aspace); + msm_gem_unpin_iova(a6xx_gpu->sqe_bo, gpu->vm); drm_gem_object_put(a6xx_gpu->sqe_bo); } if (a6xx_gpu->shadow_bo) { - msm_gem_unpin_iova(a6xx_gpu->shadow_bo, gpu->aspace); + msm_gem_unpin_iova(a6xx_gpu->shadow_bo, gpu->vm); drm_gem_object_put(a6xx_gpu->shadow_bo); } @@ -2238,8 +2258,8 @@ static void a6xx_gpu_set_freq(struct msm_gpu *gpu, struct dev_pm_opp *opp, mutex_unlock(&a6xx_gpu->gmu.lock); } -static struct msm_gem_address_space * -a6xx_create_address_space(struct msm_gpu *gpu, struct platform_device *pdev) +static struct drm_gpuvm * +a6xx_create_vm(struct msm_gpu *gpu, struct platform_device *pdev) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); @@ -2253,22 +2273,21 @@ a6xx_create_address_space(struct msm_gpu *gpu, struct platform_device *pdev) !device_iommu_capable(&pdev->dev, IOMMU_CAP_CACHE_COHERENCY)) quirks |= IO_PGTABLE_QUIRK_ARM_OUTER_WBWA; - return adreno_iommu_create_address_space(gpu, pdev, quirks); + return adreno_iommu_create_vm(gpu, pdev, quirks); } -static struct msm_gem_address_space * -a6xx_create_private_address_space(struct msm_gpu *gpu) +static struct drm_gpuvm * +a6xx_create_private_vm(struct msm_gpu *gpu, bool kernel_managed) { struct msm_mmu *mmu; - mmu = msm_iommu_pagetable_create(gpu->aspace->mmu); + mmu = msm_iommu_pagetable_create(to_msm_vm(gpu->vm)->mmu, kernel_managed); if (IS_ERR(mmu)) return ERR_CAST(mmu); - return msm_gem_address_space_create(mmu, - "gpu", ADRENO_VM_START, - adreno_private_address_space_size(gpu)); + return msm_gem_vm_create(gpu->dev, mmu, "gpu", ADRENO_VM_START, + adreno_private_vm_size(gpu), kernel_managed); } static uint32_t a6xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring) @@ -2385,8 +2404,8 @@ static const struct adreno_gpu_funcs funcs = { .gpu_state_get = a6xx_gpu_state_get, .gpu_state_put = a6xx_gpu_state_put, #endif - .create_address_space = a6xx_create_address_space, - .create_private_address_space = a6xx_create_private_address_space, + .create_vm = a6xx_create_vm, + .create_private_vm = a6xx_create_private_vm, .get_rptr = a6xx_get_rptr, .progress = a6xx_progress, }, @@ -2414,8 +2433,8 @@ static const struct adreno_gpu_funcs funcs_gmuwrapper = { .gpu_state_get = a6xx_gpu_state_get, .gpu_state_put = a6xx_gpu_state_put, #endif - .create_address_space = a6xx_create_address_space, - .create_private_address_space = a6xx_create_private_address_space, + .create_vm = a6xx_create_vm, + .create_private_vm = a6xx_create_private_vm, .get_rptr = a6xx_get_rptr, .progress = a6xx_progress, }, @@ -2445,8 +2464,8 @@ static const struct adreno_gpu_funcs funcs_a7xx = { .gpu_state_get = a6xx_gpu_state_get, .gpu_state_put = a6xx_gpu_state_put, #endif - .create_address_space = a6xx_create_address_space, - .create_private_address_space = a6xx_create_private_address_space, + .create_vm = a6xx_create_vm, + .create_private_vm = a6xx_create_private_vm, .get_rptr = a6xx_get_rptr, .progress = a6xx_progress, }, @@ -2542,11 +2561,15 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev) adreno_gpu->uche_trap_base = 0x1fffffffff000ull; - if (gpu->aspace) - msm_mmu_set_fault_handler(gpu->aspace->mmu, gpu, - a6xx_fault_handler); + msm_mmu_set_fault_handler(to_msm_vm(gpu->vm)->mmu, gpu, + a6xx_fault_handler); + + ret = a6xx_calc_ubwc_config(adreno_gpu); + if (ret) { + a6xx_destroy(&(a6xx_gpu->base.base)); + return ERR_PTR(ret); + } - a6xx_calc_ubwc_config(adreno_gpu); /* Set up the preemption specific bits and pieces for each ringbuffer */ a6xx_preempt_init(gpu); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h index 9201a53dd341..6e71f617fc3d 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h @@ -6,6 +6,10 @@ #include "adreno_gpu.h" +#include "a6xx_enums.xml.h" +#include "a7xx_enums.xml.h" +#include "a6xx_perfcntrs.xml.h" +#include "a7xx_perfcntrs.xml.h" #include "a6xx.xml.h" #include "a6xx_gmu.h" diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c index 341a72a67401..faca2a0243ab 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c @@ -132,7 +132,7 @@ static int a6xx_crashdumper_init(struct msm_gpu *gpu, struct a6xx_crashdumper *dumper) { dumper->ptr = msm_gem_kernel_new(gpu->dev, - SZ_1M, MSM_BO_WC, gpu->aspace, + SZ_1M, MSM_BO_WC, gpu->vm, &dumper->bo, &dumper->iova); if (!IS_ERR(dumper->ptr)) @@ -158,7 +158,7 @@ static int a6xx_crashdumper_run(struct msm_gpu *gpu, /* Make sure all pending memory writes are posted */ wmb(); - gpu_write64(gpu, REG_A6XX_CP_CRASH_SCRIPT_BASE, dumper->iova); + gpu_write64(gpu, REG_A6XX_CP_CRASH_DUMP_SCRIPT_BASE, dumper->iova); gpu_write(gpu, REG_A6XX_CP_CRASH_DUMP_CNTL, 1); @@ -1619,7 +1619,7 @@ struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu) a7xx_get_clusters(gpu, a6xx_state, dumper); a7xx_get_dbgahb_clusters(gpu, a6xx_state, dumper); - msm_gem_kernel_put(dumper->bo, gpu->aspace); + msm_gem_kernel_put(dumper->bo, gpu->vm); } a7xx_get_post_crashdumper_registers(gpu, a6xx_state); @@ -1631,7 +1631,7 @@ struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu) a6xx_get_clusters(gpu, a6xx_state, dumper); a6xx_get_dbgahb_clusters(gpu, a6xx_state, dumper); - msm_gem_kernel_put(dumper->bo, gpu->aspace); + msm_gem_kernel_put(dumper->bo, gpu->vm); } } diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h index e545106c70be..95d93ac6812a 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h @@ -212,7 +212,7 @@ static const struct a6xx_shader_block { SHADER(A6XX_SP_LB_5_DATA, 0x200), SHADER(A6XX_SP_CB_BINDLESS_DATA, 0x800), SHADER(A6XX_SP_CB_LEGACY_DATA, 0x280), - SHADER(A6XX_SP_UAV_DATA, 0x80), + SHADER(A6XX_SP_GFX_UAV_BASE_DATA, 0x80), SHADER(A6XX_SP_INST_TAG, 0x80), SHADER(A6XX_SP_CB_BINDLESS_TAG, 0x80), SHADER(A6XX_SP_TMO_UMO_TAG, 0x80), diff --git a/drivers/gpu/drm/msm/adreno/a6xx_preempt.c b/drivers/gpu/drm/msm/adreno/a6xx_preempt.c index 3b17fd2dba89..6a12a35dabff 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_preempt.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_preempt.c @@ -210,7 +210,7 @@ void a6xx_preempt_hw_init(struct msm_gpu *gpu) gpu_write64(gpu, REG_A6XX_CP_CONTEXT_SWITCH_SMMU_INFO, 0); /* Enable the GMEM save/restore feature for preemption */ - gpu_write(gpu, REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE, 0x1); + gpu_write(gpu, REG_A6XX_RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ENABLE, 0x1); /* Reset the preemption state */ set_preempt_state(a6xx_gpu, PREEMPT_NONE); @@ -344,7 +344,7 @@ static int preempt_init_ring(struct a6xx_gpu *a6xx_gpu, ptr = msm_gem_kernel_new(gpu->dev, PREEMPT_RECORD_SIZE(adreno_gpu), - MSM_BO_WC | MSM_BO_MAP_PRIV, gpu->aspace, &bo, &iova); + MSM_BO_WC | MSM_BO_MAP_PRIV, gpu->vm, &bo, &iova); if (IS_ERR(ptr)) return PTR_ERR(ptr); @@ -362,7 +362,7 @@ static int preempt_init_ring(struct a6xx_gpu *a6xx_gpu, ptr = msm_gem_kernel_new(gpu->dev, PREEMPT_SMMU_INFO_SIZE, MSM_BO_WC | MSM_BO_MAP_PRIV | MSM_BO_GPU_READONLY, - gpu->aspace, &bo, &iova); + gpu->vm, &bo, &iova); if (IS_ERR(ptr)) return PTR_ERR(ptr); @@ -377,7 +377,7 @@ static int preempt_init_ring(struct a6xx_gpu *a6xx_gpu, struct a7xx_cp_smmu_info *smmu_info_ptr = ptr; - msm_iommu_pagetable_params(gpu->aspace->mmu, &ttbr, &asid); + msm_iommu_pagetable_params(to_msm_vm(gpu->vm)->mmu, &ttbr, &asid); smmu_info_ptr->magic = GEN7_CP_SMMU_INFO_MAGIC; smmu_info_ptr->ttbr0 = ttbr; @@ -405,7 +405,7 @@ void a6xx_preempt_fini(struct msm_gpu *gpu) int i; for (i = 0; i < gpu->nr_rings; i++) - msm_gem_kernel_put(a6xx_gpu->preempt_bo[i], gpu->aspace); + msm_gem_kernel_put(a6xx_gpu->preempt_bo[i], gpu->vm); } void a6xx_preempt_init(struct msm_gpu *gpu) @@ -431,7 +431,7 @@ void a6xx_preempt_init(struct msm_gpu *gpu) a6xx_gpu->preempt_postamble_ptr = msm_gem_kernel_new(gpu->dev, PAGE_SIZE, MSM_BO_WC | MSM_BO_MAP_PRIV | MSM_BO_GPU_READONLY, - gpu->aspace, &a6xx_gpu->preempt_postamble_bo, + gpu->vm, &a6xx_gpu->preempt_postamble_bo, &a6xx_gpu->preempt_postamble_iova); preempt_prepare_postamble(a6xx_gpu); diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c index f5e1490d07c1..50945bfe9b49 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_device.c +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c @@ -16,10 +16,6 @@ bool snapshot_debugbus = false; MODULE_PARM_DESC(snapshot_debugbus, "Include debugbus sections in GPU devcoredump (if not fused off)"); module_param_named(snapshot_debugbus, snapshot_debugbus, bool, 0600); -bool allow_vram_carveout = false; -MODULE_PARM_DESC(allow_vram_carveout, "Allow using VRAM Carveout, in place of IOMMU"); -module_param_named(allow_vram_carveout, allow_vram_carveout, bool, 0600); - int enable_preemption = -1; MODULE_PARM_DESC(enable_preemption, "Enable preemption (A7xx only) (1=on , 0=disable, -1=auto (default))"); module_param(enable_preemption, int, 0600); @@ -137,9 +133,8 @@ err_disable_rpm: return NULL; } -static int find_chipid(struct device *dev, uint32_t *chipid) +static int find_chipid(struct device_node *node, uint32_t *chipid) { - struct device_node *node = dev->of_node; const char *compat; int ret; @@ -173,15 +168,36 @@ static int find_chipid(struct device *dev, uint32_t *chipid) /* and if that fails, fall back to legacy "qcom,chipid" property: */ ret = of_property_read_u32(node, "qcom,chipid", chipid); if (ret) { - DRM_DEV_ERROR(dev, "could not parse qcom,chipid: %d\n", ret); + DRM_ERROR("%pOF: could not parse qcom,chipid: %d\n", + node, ret); return ret; } - dev_warn(dev, "Using legacy qcom,chipid binding!\n"); + pr_warn("%pOF: Using legacy qcom,chipid binding!\n", node); return 0; } +bool adreno_has_gpu(struct device_node *node) +{ + const struct adreno_info *info; + uint32_t chip_id; + int ret; + + ret = find_chipid(node, &chip_id); + if (ret) + return false; + + info = adreno_info(chip_id); + if (!info) { + pr_warn("%pOF: Unknown GPU revision: %"ADRENO_CHIPID_FMT"\n", + node, ADRENO_CHIPID_ARGS(chip_id)); + return false; + } + + return true; +} + static int adreno_bind(struct device *dev, struct device *master, void *data) { static struct adreno_platform_config config = {}; @@ -191,19 +207,18 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) struct msm_gpu *gpu; int ret; - ret = find_chipid(dev, &config.chip_id); - if (ret) + ret = find_chipid(dev->of_node, &config.chip_id); + /* We shouldn't have gotten this far if we can't parse the chip_id */ + if (WARN_ON(ret)) return ret; dev->platform_data = &config; priv->gpu_pdev = to_platform_device(dev); info = adreno_info(config.chip_id); - if (!info) { - dev_warn(drm->dev, "Unknown GPU revision: %"ADRENO_CHIPID_FMT"\n", - ADRENO_CHIPID_ARGS(config.chip_id)); + /* We shouldn't have gotten this far if we don't recognize the GPU: */ + if (WARN_ON(!info)) return -ENXIO; - } config.info = info; @@ -245,42 +260,23 @@ static const struct component_ops a3xx_ops = { .unbind = adreno_unbind, }; -static void adreno_device_register_headless(void) -{ - /* on imx5, we don't have a top-level mdp/dpu node - * this creates a dummy node for the driver for that case - */ - struct platform_device_info dummy_info = { - .parent = NULL, - .name = "msm", - .id = -1, - .res = NULL, - .num_res = 0, - .data = NULL, - .size_data = 0, - .dma_mask = ~0, - }; - platform_device_register_full(&dummy_info); -} - static int adreno_probe(struct platform_device *pdev) { + if (of_device_is_compatible(pdev->dev.of_node, "amd,imageon") || + msm_gpu_no_components()) + return msm_gpu_probe(pdev, &a3xx_ops); - int ret; - - ret = component_add(&pdev->dev, &a3xx_ops); - if (ret) - return ret; - - if (of_device_is_compatible(pdev->dev.of_node, "amd,imageon")) - adreno_device_register_headless(); - - return 0; + return component_add(&pdev->dev, &a3xx_ops); } static void adreno_remove(struct platform_device *pdev) { - component_del(&pdev->dev, &a3xx_ops); + struct msm_drm_private *priv = platform_get_drvdata(pdev); + + if (priv->kms_init) + component_del(&pdev->dev, &a3xx_ops); + else + msm_gpu_remove(pdev, &a3xx_ops); } static void adreno_shutdown(struct platform_device *pdev) diff --git a/drivers/gpu/drm/msm/adreno/adreno_gen7_9_0_snapshot.h b/drivers/gpu/drm/msm/adreno/adreno_gen7_9_0_snapshot.h index 9a327d543f27..e02cabb39f19 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gen7_9_0_snapshot.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gen7_9_0_snapshot.h @@ -1311,8 +1311,8 @@ static struct a6xx_indexed_registers gen7_9_0_cp_indexed_reg_list[] = { REG_A7XX_CP_BV_SQE_UCODE_DBG_DATA, 0x08000}, { "CP_BV_SQE_STAT_ADDR", REG_A7XX_CP_BV_SQE_STAT_ADDR, REG_A7XX_CP_BV_SQE_STAT_DATA, 0x00040}, - { "CP_RESOURCE_TBL", REG_A7XX_CP_RESOURCE_TBL_DBG_ADDR, - REG_A7XX_CP_RESOURCE_TBL_DBG_DATA, 0x04100}, + { "CP_RESOURCE_TBL", REG_A7XX_CP_RESOURCE_TABLE_DBG_ADDR, + REG_A7XX_CP_RESOURCE_TABLE_DBG_DATA, 0x04100}, { "CP_LPAC_DRAW_STATE_ADDR", REG_A7XX_CP_LPAC_DRAW_STATE_ADDR, REG_A7XX_CP_LPAC_DRAW_STATE_DATA, 0x00200}, { "CP_LPAC_ROQ", REG_A7XX_CP_LPAC_ROQ_DBG_ADDR, diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 2348ffb35f7e..f1230465bf0d 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -191,25 +191,27 @@ int adreno_zap_shader_load(struct msm_gpu *gpu, u32 pasid) return zap_shader_load_mdt(gpu, adreno_gpu->info->zapfw, pasid); } -struct msm_gem_address_space * -adreno_create_address_space(struct msm_gpu *gpu, - struct platform_device *pdev) +struct drm_gpuvm * +adreno_create_vm(struct msm_gpu *gpu, + struct platform_device *pdev) { - return adreno_iommu_create_address_space(gpu, pdev, 0); + return adreno_iommu_create_vm(gpu, pdev, 0); } -struct msm_gem_address_space * -adreno_iommu_create_address_space(struct msm_gpu *gpu, - struct platform_device *pdev, - unsigned long quirks) +struct drm_gpuvm * +adreno_iommu_create_vm(struct msm_gpu *gpu, + struct platform_device *pdev, + unsigned long quirks) { struct iommu_domain_geometry *geometry; struct msm_mmu *mmu; - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; u64 start, size; mmu = msm_iommu_gpu_new(&pdev->dev, gpu, quirks); - if (IS_ERR_OR_NULL(mmu)) + if (!mmu) + return ERR_PTR(-ENODEV); + else if (IS_ERR_OR_NULL(mmu)) return ERR_CAST(mmu); geometry = msm_iommu_get_geometry(mmu); @@ -224,16 +226,16 @@ adreno_iommu_create_address_space(struct msm_gpu *gpu, start = max_t(u64, SZ_16M, geometry->aperture_start); size = geometry->aperture_end - start + 1; - aspace = msm_gem_address_space_create(mmu, "gpu", - start & GENMASK_ULL(48, 0), size); + vm = msm_gem_vm_create(gpu->dev, mmu, "gpu", start & GENMASK_ULL(48, 0), + size, true); - if (IS_ERR(aspace) && !IS_ERR(mmu)) + if (IS_ERR(vm) && !IS_ERR(mmu)) mmu->funcs->destroy(mmu); - return aspace; + return vm; } -u64 adreno_private_address_space_size(struct msm_gpu *gpu) +u64 adreno_private_vm_size(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); struct adreno_smmu_priv *adreno_smmu = dev_get_drvdata(&gpu->pdev->dev); @@ -259,25 +261,59 @@ u64 adreno_private_address_space_size(struct msm_gpu *gpu) return BIT(ttbr1_cfg->ias) - ADRENO_VM_START; } +void adreno_check_and_reenable_stall(struct adreno_gpu *adreno_gpu) +{ + struct msm_gpu *gpu = &adreno_gpu->base; + struct msm_drm_private *priv = gpu->dev->dev_private; + unsigned long flags; + + /* + * Wait until the cooldown period has passed and we would actually + * collect a crashdump to re-enable stall-on-fault. + */ + spin_lock_irqsave(&priv->fault_stall_lock, flags); + if (!priv->stall_enabled && + ktime_after(ktime_get(), priv->stall_reenable_time) && + !READ_ONCE(gpu->crashstate)) { + struct msm_mmu *mmu = to_msm_vm(gpu->vm)->mmu; + + priv->stall_enabled = true; + + mmu->funcs->set_stall(mmu, true); + } + spin_unlock_irqrestore(&priv->fault_stall_lock, flags); +} + #define ARM_SMMU_FSR_TF BIT(1) #define ARM_SMMU_FSR_PF BIT(3) #define ARM_SMMU_FSR_EF BIT(4) +#define ARM_SMMU_FSR_SS BIT(30) int adreno_fault_handler(struct msm_gpu *gpu, unsigned long iova, int flags, struct adreno_smmu_fault_info *info, const char *block, u32 scratch[4]) { + struct msm_drm_private *priv = gpu->dev->dev_private; + struct msm_mmu *mmu = to_msm_vm(gpu->vm)->mmu; const char *type = "UNKNOWN"; - bool do_devcoredump = info && !READ_ONCE(gpu->crashstate); + bool do_devcoredump = info && (info->fsr & ARM_SMMU_FSR_SS) && + !READ_ONCE(gpu->crashstate); + unsigned long irq_flags; /* - * If we aren't going to be resuming later from fault_worker, then do - * it now. + * In case there is a subsequent storm of pagefaults, disable + * stall-on-fault for at least half a second. */ - if (!do_devcoredump) { - gpu->aspace->mmu->funcs->resume_translation(gpu->aspace->mmu); + spin_lock_irqsave(&priv->fault_stall_lock, irq_flags); + if (priv->stall_enabled) { + priv->stall_enabled = false; + + mmu->funcs->set_stall(mmu, false); } + priv->stall_reenable_time = ktime_add_ms(ktime_get(), 500); + spin_unlock_irqrestore(&priv->fault_stall_lock, irq_flags); + /* * Print a default message if we couldn't get the data from the * adreno-smmu-priv @@ -304,26 +340,37 @@ int adreno_fault_handler(struct msm_gpu *gpu, unsigned long iova, int flags, scratch[0], scratch[1], scratch[2], scratch[3]); if (do_devcoredump) { + struct msm_gpu_fault_info fault_info = {}; + /* Turn off the hangcheck timer to keep it from bothering us */ timer_delete(&gpu->hangcheck_timer); - gpu->fault_info.ttbr0 = info->ttbr0; - gpu->fault_info.iova = iova; - gpu->fault_info.flags = flags; - gpu->fault_info.type = type; - gpu->fault_info.block = block; + fault_info.ttbr0 = info->ttbr0; + fault_info.iova = iova; + fault_info.flags = flags; + fault_info.type = type; + fault_info.block = block; - kthread_queue_work(gpu->worker, &gpu->fault_work); + msm_gpu_fault_crashstate_capture(gpu, &fault_info); } return 0; } -int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, +static bool +adreno_smmu_has_prr(struct msm_gpu *gpu) +{ + struct adreno_smmu_priv *adreno_smmu = dev_get_drvdata(&gpu->pdev->dev); + return adreno_smmu && adreno_smmu->set_prr_addr; +} + +int adreno_get_param(struct msm_gpu *gpu, struct msm_context *ctx, uint32_t param, uint64_t *value, uint32_t *len) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); struct drm_device *drm = gpu->dev; + /* Note ctx can be NULL when called from rd_open(): */ + struct drm_gpuvm *vm = ctx ? msm_context_vm(drm, ctx) : NULL; /* No pointer params yet */ if (*len != 0) @@ -369,8 +416,8 @@ int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, *value = 0; return 0; case MSM_PARAM_FAULTS: - if (ctx->aspace) - *value = gpu->global_faults + ctx->aspace->faults; + if (vm) + *value = gpu->global_faults + to_msm_vm(vm)->faults; else *value = gpu->global_faults; return 0; @@ -378,36 +425,39 @@ int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, *value = gpu->suspend_count; return 0; case MSM_PARAM_VA_START: - if (ctx->aspace == gpu->aspace) + if (vm == gpu->vm) return UERR(EINVAL, drm, "requires per-process pgtables"); - *value = ctx->aspace->va_start; + *value = vm->mm_start; return 0; case MSM_PARAM_VA_SIZE: - if (ctx->aspace == gpu->aspace) + if (vm == gpu->vm) return UERR(EINVAL, drm, "requires per-process pgtables"); - *value = ctx->aspace->va_size; + *value = vm->mm_range; return 0; case MSM_PARAM_HIGHEST_BANK_BIT: - *value = adreno_gpu->ubwc_config.highest_bank_bit; + *value = adreno_gpu->ubwc_config->highest_bank_bit; return 0; case MSM_PARAM_RAYTRACING: *value = adreno_gpu->has_ray_tracing; return 0; case MSM_PARAM_UBWC_SWIZZLE: - *value = adreno_gpu->ubwc_config.ubwc_swizzle; + *value = adreno_gpu->ubwc_config->ubwc_swizzle; return 0; case MSM_PARAM_MACROTILE_MODE: - *value = adreno_gpu->ubwc_config.macrotile_mode; + *value = adreno_gpu->ubwc_config->macrotile_mode; return 0; case MSM_PARAM_UCHE_TRAP_BASE: *value = adreno_gpu->uche_trap_base; return 0; + case MSM_PARAM_HAS_PRR: + *value = adreno_smmu_has_prr(gpu); + return 0; default: return UERR(EINVAL, drm, "%s: invalid param: %u", gpu->name, param); } } -int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, +int adreno_set_param(struct msm_gpu *gpu, struct msm_context *ctx, uint32_t param, uint64_t value, uint32_t len) { struct drm_device *drm = gpu->dev; @@ -453,7 +503,22 @@ int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, case MSM_PARAM_SYSPROF: if (!capable(CAP_SYS_ADMIN)) return UERR(EPERM, drm, "invalid permissions"); - return msm_file_private_set_sysprof(ctx, gpu, value); + return msm_context_set_sysprof(ctx, gpu, value); + case MSM_PARAM_EN_VM_BIND: + /* We can only support VM_BIND with per-process pgtables: */ + if (ctx->vm == gpu->vm) + return UERR(EINVAL, drm, "requires per-process pgtables"); + + /* + * We can only swtich to VM_BIND mode if the VM has not yet + * been created: + */ + if (ctx->vm) + return UERR(EBUSY, drm, "VM already created"); + + ctx->userspace_managed_vm = value; + + return 0; default: return UERR(EINVAL, drm, "%s: invalid param: %u", gpu->name, param); } @@ -575,7 +640,7 @@ struct drm_gem_object *adreno_fw_create_bo(struct msm_gpu *gpu, void *ptr; ptr = msm_gem_kernel_new(gpu->dev, fw->size - 4, - MSM_BO_WC | MSM_BO_GPU_READONLY, gpu->aspace, &bo, iova); + MSM_BO_WC | MSM_BO_GPU_READONLY, gpu->vm, &bo, iova); if (IS_ERR(ptr)) return ERR_CAST(ptr); @@ -768,6 +833,7 @@ void adreno_gpu_state_destroy(struct msm_gpu_state *state) for (i = 0; state->bos && i < state->nr_bos; i++) kvfree(state->bos[i].data); + kfree(state->vm_logs); kfree(state->bos); kfree(state->comm); kfree(state->cmd); @@ -908,6 +974,16 @@ void adreno_show(struct msm_gpu *gpu, struct msm_gpu_state *state, info->ptes[0], info->ptes[1], info->ptes[2], info->ptes[3]); } + if (state->vm_logs) { + drm_puts(p, "vm-log:\n"); + for (i = 0; i < state->nr_vm_logs; i++) { + struct msm_gem_vm_log_entry *e = &state->vm_logs[i]; + drm_printf(p, " - %s:%d: 0x%016llx-0x%016llx\n", + e->op, e->queue_id, e->iova, + e->iova + e->range); + } + } + drm_printf(p, "rbbm-status: 0x%08x\n", state->rbbm_status); drm_puts(p, "ringbuffer:\n"); diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index a8f4bf416e64..9dc93c247196 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -12,13 +12,14 @@ #include <linux/firmware.h> #include <linux/iopoll.h> +#include <linux/soc/qcom/ubwc.h> + #include "msm_gpu.h" #include "adreno_common.xml.h" #include "adreno_pm4.xml.h" extern bool snapshot_debugbus; -extern bool allow_vram_carveout; enum { ADRENO_FW_PM4 = 0, @@ -205,44 +206,12 @@ struct adreno_gpu { /* firmware: */ const struct firmware *fw[ADRENO_FW_MAX]; - struct { - /** - * @rgb565_predicator: Unknown, introduced with A650 family, - * related to UBWC mode/ver 4 - */ - u32 rgb565_predicator; - /** @uavflagprd_inv: Unknown, introduced with A650 family */ - u32 uavflagprd_inv; - /** @min_acc_len: Whether the minimum access length is 64 bits */ - u32 min_acc_len; - /** - * @ubwc_swizzle: Whether to enable level 1, 2 & 3 bank swizzling. - * - * UBWC 1.0 always enables all three levels. - * UBWC 2.0 removes level 1 bank swizzling, leaving levels 2 & 3. - * UBWC 4.0 adds the optional ability to disable levels 2 & 3. - * - * This is a bitmask where BIT(0) enables level 1, BIT(1) - * controls level 2, and BIT(2) enables level 3. - */ - u32 ubwc_swizzle; - /** - * @highest_bank_bit: Highest Bank Bit - * - * The Highest Bank Bit value represents the bit of the highest - * DDR bank. This should ideally use DRAM type detection. - */ - u32 highest_bank_bit; - u32 amsbc; - /** - * @macrotile_mode: Macrotile Mode - * - * Whether to use 4-channel macrotiling mode or the newer - * 8-channel macrotiling mode introduced in UBWC 3.1. 0 is - * 4-channel and 1 is 8-channel. - */ - u32 macrotile_mode; - } ubwc_config; + /* + * The migration to the central UBWC config db is still in flight - keep + * a copy containing some local fixups until that's done. + */ + const struct qcom_ubwc_cfg_data *ubwc_config; + struct qcom_ubwc_cfg_data _ubwc_config; /* * Register offsets are different between some GPUs. @@ -580,10 +549,10 @@ static inline int adreno_is_a7xx(struct adreno_gpu *gpu) /* Put vm_start above 32b to catch issues with not setting xyz_BASE_HI */ #define ADRENO_VM_START 0x100000000ULL -u64 adreno_private_address_space_size(struct msm_gpu *gpu); -int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, +u64 adreno_private_vm_size(struct msm_gpu *gpu); +int adreno_get_param(struct msm_gpu *gpu, struct msm_context *ctx, uint32_t param, uint64_t *value, uint32_t *len); -int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, +int adreno_set_param(struct msm_gpu *gpu, struct msm_context *ctx, uint32_t param, uint64_t value, uint32_t len); const struct firmware *adreno_request_fw(struct adreno_gpu *adreno_gpu, const char *fwname); @@ -623,19 +592,21 @@ void adreno_show_object(struct drm_printer *p, void **ptr, int len, * Common helper function to initialize the default address space for arm-smmu * attached targets */ -struct msm_gem_address_space * -adreno_create_address_space(struct msm_gpu *gpu, - struct platform_device *pdev); +struct drm_gpuvm * +adreno_create_vm(struct msm_gpu *gpu, + struct platform_device *pdev); -struct msm_gem_address_space * -adreno_iommu_create_address_space(struct msm_gpu *gpu, - struct platform_device *pdev, - unsigned long quirks); +struct drm_gpuvm * +adreno_iommu_create_vm(struct msm_gpu *gpu, + struct platform_device *pdev, + unsigned long quirks); int adreno_fault_handler(struct msm_gpu *gpu, unsigned long iova, int flags, struct adreno_smmu_fault_info *info, const char *block, u32 scratch[4]); +void adreno_check_and_reenable_stall(struct adreno_gpu *gpu); + int adreno_read_speedbin(struct device *dev, u32 *speedbin); /* diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h index ffc4d4257ae5..56d3c38c8778 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_10_0_sm8650.h @@ -21,7 +21,6 @@ static const struct dpu_caps sm8650_dpu_caps = { static const struct dpu_mdp_cfg sm8650_mdp = { .name = "top_0", .base = 0, .len = 0x494, - .features = BIT(DPU_MDP_PERIPH_0_REMOVED), .clk_ctrls = { [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, }, @@ -31,32 +30,26 @@ static const struct dpu_ctl_cfg sm8650_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x1000, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x16000, .len = 0x1000, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x17000, .len = 0x1000, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x18000, .len = 0x1000, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x19000, .len = 0x1000, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a000, .len = 0x1000, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -139,7 +132,7 @@ static const struct dpu_lm_cfg sm8650_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -147,7 +140,7 @@ static const struct dpu_lm_cfg sm8650_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -155,7 +148,7 @@ static const struct dpu_lm_cfg sm8650_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -163,7 +156,7 @@ static const struct dpu_lm_cfg sm8650_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -171,14 +164,14 @@ static const struct dpu_lm_cfg sm8650_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -189,22 +182,18 @@ static const struct dpu_dspp_cfg sm8650_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -213,67 +202,57 @@ static const struct dpu_pingpong_cfg sm8650_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x69000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x6a000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x6b000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x6c000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x6d000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x6e000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), }, { .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, .base = 0x66000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, { .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, .base = 0x66400, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, { .name = "pingpong_cwb_2", .id = PINGPONG_CWB_2, .base = 0x7e000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_4, }, { .name = "pingpong_cwb_3", .id = PINGPONG_CWB_3, .base = 0x7e400, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_4, }, @@ -307,32 +286,30 @@ static const struct dpu_dsc_cfg sm8650_dsc[] = { { .name = "dce_0_0", .id = DSC_0, .base = 0x80000, .len = 0x6, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_0, }, { .name = "dce_0_1", .id = DSC_1, .base = 0x80000, .len = 0x6, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_1, }, { .name = "dce_1_0", .id = DSC_2, .base = 0x81000, .len = 0x6, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_0, }, { .name = "dce_1_1", .id = DSC_3, .base = 0x81000, .len = 0x6, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_1, }, { .name = "dce_2_0", .id = DSC_4, .base = 0x82000, .len = 0x6, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_0, }, { .name = "dce_2_1", .id = DSC_5, .base = 0x82000, .len = 0x6, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_1, }, }; @@ -341,7 +318,7 @@ static const struct dpu_wb_cfg sm8650_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .xin_id = 6, @@ -374,7 +351,6 @@ static const struct dpu_intf_cfg sm8650_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x34000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -383,7 +359,6 @@ static const struct dpu_intf_cfg sm8650_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x35000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -393,7 +368,6 @@ static const struct dpu_intf_cfg sm8650_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x36000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -403,7 +377,6 @@ static const struct dpu_intf_cfg sm8650_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x37000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_0_sm8750.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_0_sm8750.h new file mode 100644 index 000000000000..db8cc2d0112c --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_12_0_sm8750.h @@ -0,0 +1,494 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2025 Linaro Limited + * Copyright (c) 2022 Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_12_0_SM8750_H +#define _DPU_12_0_SM8750_H + +static const struct dpu_caps sm8750_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0xb, + .has_src_split = true, + .has_dim_layer = true, + .has_idle_pc = true, + .has_3d_merge = true, + .max_linewidth = 8192, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, +}; + +static const struct dpu_mdp_cfg sm8750_mdp = { + .name = "top_0", + .base = 0, .len = 0x494, + .clk_ctrls = { + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + }, +}; + +static const struct dpu_ctl_cfg sm8750_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x1000, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm8750_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x344, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &dpu_vig_sblk_qseed3_3_4, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x344, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_12", .id = SSPP_DMA4, + .base = 0x2c000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 14, + .type = SSPP_TYPE_DMA, + }, { + .name = "sspp_13", .id = SSPP_DMA5, + .base = 0x2e000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &dpu_dma_sblk, + .xin_id = 15, + .type = SSPP_TYPE_DMA, + }, +}; + +static const struct dpu_lm_cfg sm8750_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, { + .name = "lm_6", .id = LM_6, + .base = 0x4a000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_7, + .pingpong = PINGPONG_6, + }, { + .name = "lm_7", .id = LM_7, + .base = 0x4b000, .len = 0x400, + .features = MIXER_MSM8998_MASK, + .sblk = &sm8750_lm_sblk, + .lm_pair = LM_6, + .pingpong = PINGPONG_7, + }, +}; + +static const struct dpu_dspp_cfg sm8750_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .sblk = &sm8750_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm8750_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + }, { + .name = "pingpong_6", .id = PINGPONG_6, + .base = 0x6f000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 20), + }, { + .name = "pingpong_7", .id = PINGPONG_7, + .base = 0x70000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 21), + }, { + .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, + .base = 0x66000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_4, + }, { + .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, + .base = 0x66400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_4, + }, { + .name = "pingpong_cwb_2", .id = PINGPONG_CWB_2, + .base = 0x7e000, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_5, + }, { + .name = "pingpong_cwb_3", .id = PINGPONG_CWB_3, + .base = 0x7e400, .len = 0, + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_5, + }, +}; + +static const struct dpu_merge_3d_cfg sm8750_merge_3d[] = { + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x1c, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x1c, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x1c, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x51000, .len = 0x1c, + }, { + .name = "merge_3d_4", .id = MERGE_3D_4, + .base = 0x66700, .len = 0x1c, + }, { + .name = "merge_3d_5", .id = MERGE_3D_5, + .base = 0x7e700, .len = 0x1c, + }, +}; + +/* + * NOTE: Each display compression engine (DCE) contains dual hard + * slice DSC encoders so both share same base address but with + * its own different sub block address. + */ +static const struct dpu_dsc_cfg sm8750_dsc[] = { + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_2_0", .id = DSC_4, + .base = 0x82000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_2_1", .id = DSC_5, + .base = 0x82000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, { + .name = "dce_3_0", .id = DSC_6, + .base = 0x83000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_0, + }, { + .name = "dce_3_1", .id = DSC_7, + .base = 0x83000, .len = 0x8, + .features = BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &sm8750_dsc_sblk_1, + }, +}; + +static const struct dpu_wb_cfg sm8750_wb[] = { + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SDM845_MASK, + .format_list = wb2_formats_rgb_yuv, + .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), + .xin_id = 6, + .vbif_idx = VBIF_RT, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, +}; + +static const struct dpu_cwb_cfg sm8750_cwb[] = { + { + .name = "cwb_0", .id = CWB_0, + .base = 0x66200, .len = 0x20, + }, + { + .name = "cwb_1", .id = CWB_1, + .base = 0x66600, .len = 0x20, + }, + { + .name = "cwb_2", .id = CWB_2, + .base = 0x7e200, .len = 0x20, + }, + { + .name = "cwb_3", .id = CWB_3, + .base = 0x7e600, .len = 0x20, + }, +}; + +static const struct dpu_intf_cfg sm8750_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x4bc, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x4bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x4bc, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x4bc, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + }, +}; + +static const struct dpu_perf_cfg sm8750_perf_data = { + .max_bw_low = 18900000, + .max_bw_high = 28500000, + .min_core_ib = 2500000, + .min_llcc_ib = 0, + .min_dram_ib = 800000, + .min_prefill_lines = 35, + .danger_lut_tbl = {0x3ffff, 0x3ffff, 0x0}, + .safe_lut_tbl = {0xfe00, 0xfe00, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sc7180_qos_linear), + .entries = sc7180_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm8750_mdss_ver = { + .core_major_ver = 12, + .core_minor_ver = 0, +}; + +const struct dpu_mdss_cfg dpu_sm8750_cfg = { + .mdss_ver = &sm8750_mdss_ver, + .caps = &sm8750_dpu_caps, + .mdp = &sm8750_mdp, + .cdm = &dpu_cdm_5_x, + .ctl_count = ARRAY_SIZE(sm8750_ctl), + .ctl = sm8750_ctl, + .sspp_count = ARRAY_SIZE(sm8750_sspp), + .sspp = sm8750_sspp, + .mixer_count = ARRAY_SIZE(sm8750_lm), + .mixer = sm8750_lm, + .dspp_count = ARRAY_SIZE(sm8750_dspp), + .dspp = sm8750_dspp, + .pingpong_count = ARRAY_SIZE(sm8750_pp), + .pingpong = sm8750_pp, + .dsc_count = ARRAY_SIZE(sm8750_dsc), + .dsc = sm8750_dsc, + .merge_3d_count = ARRAY_SIZE(sm8750_merge_3d), + .merge_3d = sm8750_merge_3d, + .wb_count = ARRAY_SIZE(sm8750_wb), + .wb = sm8750_wb, + .cwb_count = ARRAY_SIZE(sm8750_cwb), + .cwb = sm8650_cwb, + .intf_count = ARRAY_SIZE(sm8750_intf), + .intf = sm8750_intf, + .vbif_count = ARRAY_SIZE(sm8650_vbif), + .vbif = sm8650_vbif, + .perf = &sm8750_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h index 39027a21c6fe..29e0eba91930 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_14_msm8937.h @@ -19,7 +19,6 @@ static const struct dpu_mdp_cfg msm8937_mdp[] = { { .name = "top_0", .base = 0x0, .len = 0x454, - .features = BIT(DPU_MDP_VSYNC_SEL), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_RGB0] = { .reg_off = 0x2ac, .bit_off = 4 }, @@ -116,7 +115,6 @@ static const struct dpu_dspp_cfg msm8937_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &msm8998_dspp_sblk, }, }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h index 8d1b43ea1663..cb1ee4b63f9f 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_15_msm8917.h @@ -19,7 +19,6 @@ static const struct dpu_mdp_cfg msm8917_mdp[] = { { .name = "top_0", .base = 0x0, .len = 0x454, - .features = BIT(DPU_MDP_VSYNC_SEL), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_RGB0] = { .reg_off = 0x2ac, .bit_off = 4 }, @@ -103,7 +102,6 @@ static const struct dpu_dspp_cfg msm8917_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &msm8998_dspp_sblk, }, }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h index 16c12499b24b..b44d02b48418 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_16_msm8953.h @@ -19,7 +19,6 @@ static const struct dpu_mdp_cfg msm8953_mdp[] = { { .name = "top_0", .base = 0x0, .len = 0x454, - .features = BIT(DPU_MDP_VSYNC_SEL), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_RGB0] = { .reg_off = 0x2ac, .bit_off = 4 }, @@ -116,7 +115,6 @@ static const struct dpu_dspp_cfg msm8953_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &msm8998_dspp_sblk, }, }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h index 91f514d28ac6..8af63db315b4 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_1_7_msm8996.h @@ -22,7 +22,6 @@ static const struct dpu_mdp_cfg msm8996_mdp[] = { { .name = "top_0", .base = 0x0, .len = 0x454, - .features = BIT(DPU_MDP_VSYNC_SEL), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -181,28 +180,24 @@ static const struct dpu_pingpong_cfg msm8996_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_MSM8996_MASK, .sblk = &msm8996_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_MSM8996_MASK, .sblk = &msm8996_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x71000, .len = 0xd4, - .features = PINGPONG_MSM8996_MASK, .sblk = &msm8996_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x71800, .len = 0xd4, - .features = PINGPONG_MSM8996_MASK, .sblk = &msm8996_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15), @@ -223,12 +218,10 @@ static const struct dpu_dspp_cfg msm8996_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &msm8998_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &msm8998_dspp_sblk, }, }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h index 413cd59dc0c4..f91220496082 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h @@ -23,7 +23,6 @@ static const struct dpu_caps msm8998_dpu_caps = { static const struct dpu_mdp_cfg msm8998_mdp = { .name = "top_0", .base = 0x0, .len = 0x458, - .features = BIT(DPU_MDP_VSYNC_SEL), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -170,28 +169,24 @@ static const struct dpu_pingpong_cfg msm8998_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x71000, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x71800, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15), @@ -212,12 +207,10 @@ static const struct dpu_dspp_cfg msm8998_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &msm8998_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &msm8998_dspp_sblk, }, }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h index b2eb7ca699e3..8f9a097147c0 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_2_sdm660.h @@ -22,7 +22,6 @@ static const struct dpu_caps sdm660_dpu_caps = { static const struct dpu_mdp_cfg sdm660_mdp = { .name = "top_0", .base = 0x0, .len = 0x458, - .features = BIT(DPU_MDP_VSYNC_SEL), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -141,28 +140,24 @@ static const struct dpu_pingpong_cfg sdm660_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x71000, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x71800, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15), @@ -183,12 +178,10 @@ static const struct dpu_dspp_cfg sdm660_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &msm8998_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &msm8998_dspp_sblk, }, }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h index 85e121ad84a0..0ad18bd273ff 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_3_sdm630.h @@ -22,7 +22,6 @@ static const struct dpu_caps sdm630_dpu_caps = { static const struct dpu_mdp_cfg sdm630_mdp = { .name = "top_0", .base = 0x0, .len = 0x458, - .features = BIT(DPU_MDP_VSYNC_SEL), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, @@ -115,14 +114,12 @@ static const struct dpu_pingpong_cfg sdm630_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x71000, .len = 0xd4, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), @@ -133,7 +130,6 @@ static const struct dpu_dspp_cfg sdm630_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &msm8998_dspp_sblk, }, }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h index 49363d7d5b93..5cc9f55d542b 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h @@ -23,7 +23,6 @@ static const struct dpu_caps sdm845_dpu_caps = { static const struct dpu_mdp_cfg sdm845_mdp = { .name = "top_0", .base = 0x0, .len = 0x45c, - .features = BIT(DPU_MDP_AUDIO_SELECT) | BIT(DPU_MDP_VSYNC_SEL), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -134,7 +133,7 @@ static const struct dpu_lm_cfg sdm845_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -142,7 +141,7 @@ static const struct dpu_lm_cfg sdm845_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -150,7 +149,7 @@ static const struct dpu_lm_cfg sdm845_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_2, @@ -158,7 +157,7 @@ static const struct dpu_lm_cfg sdm845_lm[] = { }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -170,22 +169,18 @@ static const struct dpu_dspp_cfg sdm845_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -194,28 +189,24 @@ static const struct dpu_pingpong_cfg sdm845_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x71000, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x71800, .len = 0xd4, - .features = PINGPONG_SDM845_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15), diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_1_sdm670.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_1_sdm670.h index c2fde980fb52..0f5e9babdeea 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_1_sdm670.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_1_sdm670.h @@ -11,7 +11,6 @@ static const struct dpu_mdp_cfg sdm670_mdp = { .name = "top_0", .base = 0x0, .len = 0x45c, - .features = BIT(DPU_MDP_AUDIO_SELECT), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -69,7 +68,7 @@ static const struct dpu_lm_cfg sdm670_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -77,7 +76,7 @@ static const struct dpu_lm_cfg sdm670_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -85,14 +84,14 @@ static const struct dpu_lm_cfg sdm670_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_2, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -103,12 +102,10 @@ static const struct dpu_dspp_cfg sdm670_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h index 08d38e1d420c..ae1b2ed96e9f 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h @@ -23,7 +23,6 @@ static const struct dpu_caps sm8150_dpu_caps = { static const struct dpu_mdp_cfg sm8150_mdp = { .name = "top_0", .base = 0x0, .len = 0x45c, - .features = BIT(DPU_MDP_AUDIO_SELECT), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -41,32 +40,26 @@ static const struct dpu_ctl_cfg sm8150_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x1200, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x1400, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x1600, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x1800, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a00, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -143,7 +136,7 @@ static const struct dpu_lm_cfg sm8150_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -151,7 +144,7 @@ static const struct dpu_lm_cfg sm8150_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -159,7 +152,7 @@ static const struct dpu_lm_cfg sm8150_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -167,7 +160,7 @@ static const struct dpu_lm_cfg sm8150_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -175,14 +168,14 @@ static const struct dpu_lm_cfg sm8150_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -193,22 +186,18 @@ static const struct dpu_dspp_cfg sm8150_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -217,42 +206,36 @@ static const struct dpu_pingpong_cfg sm8150_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x71000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x71800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x72000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x72800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), @@ -276,19 +259,15 @@ static const struct dpu_dsc_cfg sm8150_dsc[] = { { .name = "dsc_0", .id = DSC_0, .base = 0x80000, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_1", .id = DSC_1, .base = 0x80400, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_2", .id = DSC_2, .base = 0x80800, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_3", .id = DSC_3, .base = 0x80c00, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, }; @@ -296,7 +275,7 @@ static const struct dpu_wb_cfg sm8150_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, @@ -311,7 +290,6 @@ static const struct dpu_intf_cfg sm8150_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x6a000, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -320,7 +298,6 @@ static const struct dpu_intf_cfg sm8150_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2bc, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -330,7 +307,6 @@ static const struct dpu_intf_cfg sm8150_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x6b000, .len = 0x2bc, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -340,7 +316,6 @@ static const struct dpu_intf_cfg sm8150_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x6b800, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h index d6f8b1030c68..b572cfa7ed35 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h @@ -23,7 +23,6 @@ static const struct dpu_caps sc8180x_dpu_caps = { static const struct dpu_mdp_cfg sc8180x_mdp = { .name = "top_0", .base = 0x0, .len = 0x45c, - .features = BIT(DPU_MDP_AUDIO_SELECT), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -41,32 +40,26 @@ static const struct dpu_ctl_cfg sc8180x_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x1200, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x1400, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x1600, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x1800, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a00, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -143,7 +136,7 @@ static const struct dpu_lm_cfg sc8180x_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -151,7 +144,7 @@ static const struct dpu_lm_cfg sc8180x_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -159,7 +152,7 @@ static const struct dpu_lm_cfg sc8180x_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -167,7 +160,7 @@ static const struct dpu_lm_cfg sc8180x_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -175,14 +168,14 @@ static const struct dpu_lm_cfg sc8180x_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -193,22 +186,18 @@ static const struct dpu_dspp_cfg sc8180x_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -217,42 +206,36 @@ static const struct dpu_pingpong_cfg sc8180x_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x71000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x71800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x72000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x72800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), @@ -276,27 +259,21 @@ static const struct dpu_dsc_cfg sc8180x_dsc[] = { { .name = "dsc_0", .id = DSC_0, .base = 0x80000, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_1", .id = DSC_1, .base = 0x80400, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_2", .id = DSC_2, .base = 0x80800, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_3", .id = DSC_3, .base = 0x80c00, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_4", .id = DSC_4, .base = 0x81000, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_5", .id = DSC_5, .base = 0x81400, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, }; @@ -304,7 +281,7 @@ static const struct dpu_wb_cfg sc8180x_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, @@ -319,7 +296,6 @@ static const struct dpu_intf_cfg sc8180x_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x6a000, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -328,7 +304,6 @@ static const struct dpu_intf_cfg sc8180x_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2bc, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -338,7 +313,6 @@ static const struct dpu_intf_cfg sc8180x_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x6b000, .len = 0x2bc, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -350,7 +324,6 @@ static const struct dpu_intf_cfg sc8180x_intf[] = { { .name = "intf_3", .id = INTF_3, .base = 0x6b800, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = 999, .prog_fetch_lines_worst_case = 24, @@ -359,7 +332,6 @@ static const struct dpu_intf_cfg sc8180x_intf[] = { }, { .name = "intf_4", .id = INTF_4, .base = 0x6c000, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -368,7 +340,6 @@ static const struct dpu_intf_cfg sc8180x_intf[] = { }, { .name = "intf_5", .id = INTF_5, .base = 0x6c800, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_2, .prog_fetch_lines_worst_case = 24, @@ -383,6 +354,7 @@ static const struct dpu_perf_cfg sc8180x_perf_data = { .min_core_ib = 2400000, .min_llcc_ib = 800000, .min_dram_ib = 800000, + .min_prefill_lines = 24, .danger_lut_tbl = {0xf, 0xffff, 0x0}, .safe_lut_tbl = {0xfff0, 0xf000, 0xffff}, .qos_lut_tbl = { diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h index 71ba48b05656..a56c288ac10c 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_2_sm7150.h @@ -23,7 +23,6 @@ static const struct dpu_caps sm7150_dpu_caps = { static const struct dpu_mdp_cfg sm7150_mdp = { .name = "top_0", .base = 0x0, .len = 0x45c, - .features = BIT(DPU_MDP_AUDIO_SELECT), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -38,32 +37,26 @@ static const struct dpu_ctl_cfg sm7150_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x1200, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x1400, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x1600, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x1800, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a00, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -116,7 +109,7 @@ static const struct dpu_lm_cfg sm7150_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -124,7 +117,7 @@ static const struct dpu_lm_cfg sm7150_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -132,14 +125,14 @@ static const struct dpu_lm_cfg sm7150_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -150,12 +143,10 @@ static const struct dpu_dspp_cfg sm7150_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -164,28 +155,24 @@ static const struct dpu_pingpong_cfg sm7150_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x71000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x71800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), @@ -206,11 +193,9 @@ static const struct dpu_dsc_cfg sm7150_dsc[] = { { .name = "dsc_0", .id = DSC_0, .base = 0x80000, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_1", .id = DSC_1, .base = 0x80400, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, }; @@ -218,7 +203,6 @@ static const struct dpu_intf_cfg sm7150_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x6a000, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -227,7 +211,6 @@ static const struct dpu_intf_cfg sm7150_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2bc, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -237,7 +220,6 @@ static const struct dpu_intf_cfg sm7150_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x6b000, .len = 0x2bc, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -247,7 +229,6 @@ static const struct dpu_intf_cfg sm7150_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x6b800, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -260,7 +241,7 @@ static const struct dpu_wb_cfg sm7150_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_3_sm6150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_3_sm6150.h index da11830d4407..26883f6b66b3 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_3_sm6150.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_3_sm6150.h @@ -20,7 +20,6 @@ static const struct dpu_caps sm6150_dpu_caps = { static const struct dpu_mdp_cfg sm6150_mdp = { .name = "top_0", .base = 0x0, .len = 0x45c, - .features = 0, .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, @@ -35,32 +34,26 @@ static const struct dpu_ctl_cfg sm6150_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x1200, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x1400, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x1600, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x1800, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a00, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -113,20 +106,17 @@ static const struct dpu_lm_cfg sm6150_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_QCM2290_MASK, .sblk = &sdm845_lm_sblk, .pingpong = PINGPONG_0, .dspp = DSPP_0, }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_QCM2290_MASK, .sblk = &sdm845_lm_sblk, .pingpong = PINGPONG_1, }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_QCM2290_MASK, .sblk = &sdm845_lm_sblk, .pingpong = PINGPONG_2, }, @@ -136,7 +126,6 @@ static const struct dpu_dspp_cfg sm6150_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -145,19 +134,16 @@ static const struct dpu_pingpong_cfg sm6150_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x71000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, @@ -167,7 +153,7 @@ static const struct dpu_wb_cfg sm6150_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, @@ -182,7 +168,6 @@ static const struct dpu_intf_cfg sm6150_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x6a000, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -191,7 +176,6 @@ static const struct dpu_intf_cfg sm6150_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2c0, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -201,7 +185,6 @@ static const struct dpu_intf_cfg sm6150_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x6b800, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h index fcfb3774f7a1..fbf50f279e66 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h @@ -22,7 +22,6 @@ static const struct dpu_caps sm6125_dpu_caps = { static const struct dpu_mdp_cfg sm6125_mdp = { .name = "top_0", .base = 0x0, .len = 0x45c, - .features = 0, .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, @@ -35,32 +34,26 @@ static const struct dpu_ctl_cfg sm6125_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x1200, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x1400, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x1600, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x1800, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a00, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -97,7 +90,6 @@ static const struct dpu_lm_cfg sm6125_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_QCM2290_MASK, .sblk = &sdm845_lm_sblk, .pingpong = PINGPONG_0, .dspp = DSPP_0, @@ -105,7 +97,6 @@ static const struct dpu_lm_cfg sm6125_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_QCM2290_MASK, .sblk = &sdm845_lm_sblk, .pingpong = PINGPONG_1, .dspp = 0, @@ -117,7 +108,6 @@ static const struct dpu_dspp_cfg sm6125_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -126,14 +116,12 @@ static const struct dpu_pingpong_cfg sm6125_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .merge_3d = 0, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .merge_3d = 0, .sblk = &sdm845_pp_sblk, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), @@ -144,7 +132,7 @@ static const struct dpu_wb_cfg sm6125_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, @@ -159,7 +147,6 @@ static const struct dpu_intf_cfg sm6125_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x6a000, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -168,7 +155,6 @@ static const struct dpu_intf_cfg sm6125_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2c0, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = 0, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h index a86fdb33ebdd..7b8b7a1c2d76 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h @@ -39,32 +39,26 @@ static const struct dpu_ctl_cfg sm8250_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x1200, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x1400, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x1600, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x1800, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a00, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -141,7 +135,7 @@ static const struct dpu_lm_cfg sm8250_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -149,7 +143,7 @@ static const struct dpu_lm_cfg sm8250_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -157,7 +151,7 @@ static const struct dpu_lm_cfg sm8250_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -165,7 +159,7 @@ static const struct dpu_lm_cfg sm8250_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -173,14 +167,14 @@ static const struct dpu_lm_cfg sm8250_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -191,22 +185,18 @@ static const struct dpu_dspp_cfg sm8250_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -215,42 +205,36 @@ static const struct dpu_pingpong_cfg sm8250_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x71000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x71800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x72000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x72800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), @@ -274,19 +258,15 @@ static const struct dpu_dsc_cfg sm8250_dsc[] = { { .name = "dsc_0", .id = DSC_0, .base = 0x80000, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_1", .id = DSC_1, .base = 0x80400, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_2", .id = DSC_2, .base = 0x80800, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, { .name = "dsc_3", .id = DSC_3, .base = 0x80c00, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, }; @@ -294,7 +274,6 @@ static const struct dpu_intf_cfg sm8250_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x6a000, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -303,7 +282,6 @@ static const struct dpu_intf_cfg sm8250_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2c0, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -313,7 +291,6 @@ static const struct dpu_intf_cfg sm8250_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x6b000, .len = 0x2c0, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -323,7 +300,6 @@ static const struct dpu_intf_cfg sm8250_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x6b800, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -336,7 +312,7 @@ static const struct dpu_wb_cfg sm8250_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h index 842fcc5887fe..c990ba3b5db0 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h @@ -32,17 +32,14 @@ static const struct dpu_ctl_cfg sc7180_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x1200, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x1400, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, }; @@ -87,7 +84,7 @@ static const struct dpu_lm_cfg sc7180_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sc7180_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -95,7 +92,7 @@ static const struct dpu_lm_cfg sc7180_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sc7180_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -106,7 +103,6 @@ static const struct dpu_dspp_cfg sc7180_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -115,14 +111,12 @@ static const struct dpu_pingpong_cfg sc7180_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), @@ -133,7 +127,6 @@ static const struct dpu_intf_cfg sc7180_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x6a000, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -142,7 +135,6 @@ static const struct dpu_intf_cfg sc7180_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2c0, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -156,7 +148,7 @@ static const struct dpu_wb_cfg sc7180_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h index c5fd89dd7c89..343ff5482382 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h @@ -29,7 +29,6 @@ static const struct dpu_ctl_cfg sm6115_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, }; @@ -58,7 +57,6 @@ static const struct dpu_lm_cfg sm6115_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_QCM2290_MASK, .sblk = &qcm2290_lm_sblk, .pingpong = PINGPONG_0, .dspp = DSPP_0, @@ -69,7 +67,6 @@ static const struct dpu_dspp_cfg sm6115_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -78,7 +75,6 @@ static const struct dpu_pingpong_cfg sm6115_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), @@ -89,7 +85,6 @@ static const struct dpu_intf_cfg sm6115_intf[] = { { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2c0, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h index a234bb289d24..093d16bdc450 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h @@ -35,22 +35,18 @@ static const struct dpu_ctl_cfg sm6350_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x1200, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x1400, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x1600, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, }; @@ -95,7 +91,7 @@ static const struct dpu_lm_cfg sm6350_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sc7180_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -103,7 +99,7 @@ static const struct dpu_lm_cfg sm6350_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sc7180_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -115,7 +111,6 @@ static const struct dpu_dspp_cfg sm6350_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -124,14 +119,12 @@ static struct dpu_pingpong_cfg sm6350_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x70800, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), @@ -142,7 +135,6 @@ static const struct dpu_dsc_cfg sm6350_dsc[] = { { .name = "dsc_0", .id = DSC_0, .base = 0x80000, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, }; @@ -150,7 +142,7 @@ static const struct dpu_wb_cfg sm6350_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, @@ -165,7 +157,6 @@ static const struct dpu_intf_cfg sm6350_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x6a000, .len = 0x280, - .features = INTF_SC7180_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 35, @@ -174,7 +165,6 @@ static const struct dpu_intf_cfg sm6350_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2c0, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 35, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h index 53f3be28f6f6..47053bf9b0a2 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h @@ -29,7 +29,6 @@ static const struct dpu_ctl_cfg qcm2290_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, }; @@ -58,7 +57,6 @@ static const struct dpu_lm_cfg qcm2290_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_QCM2290_MASK, .sblk = &qcm2290_lm_sblk, .pingpong = PINGPONG_0, .dspp = DSPP_0, @@ -69,7 +67,6 @@ static const struct dpu_dspp_cfg qcm2290_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -78,7 +75,6 @@ static const struct dpu_pingpong_cfg qcm2290_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), @@ -89,7 +85,6 @@ static const struct dpu_intf_cfg qcm2290_intf[] = { { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2c0, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h index 3a3bc8e429be..98190ee7ec7a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h @@ -30,7 +30,6 @@ static const struct dpu_ctl_cfg sm6375_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x1000, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, }; @@ -59,7 +58,6 @@ static const struct dpu_lm_cfg sm6375_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_QCM2290_MASK, .sblk = &qcm2290_lm_sblk, .lm_pair = 0, .pingpong = PINGPONG_0, @@ -71,7 +69,6 @@ static const struct dpu_dspp_cfg sm6375_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -80,7 +77,6 @@ static const struct dpu_pingpong_cfg sm6375_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x70000, .len = 0xd4, - .features = PINGPONG_SM8150_MASK, .sblk = &sdm845_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), @@ -91,7 +87,6 @@ static const struct dpu_dsc_cfg sm6375_dsc[] = { { .name = "dsc_0", .id = DSC_0, .base = 0x80000, .len = 0x140, - .features = BIT(DPU_DSC_OUTPUT_CTRL), }, }; @@ -99,7 +94,6 @@ static const struct dpu_intf_cfg sm6375_intf[] = { { .name = "intf_1", .id = INTF_1, .base = 0x6a800, .len = 0x2c0, - .features = INTF_SC7180_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h index 90e86063a372..85aae40c210f 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h @@ -39,32 +39,26 @@ static const struct dpu_ctl_cfg sm8350_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x1e8, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x16000, .len = 0x1e8, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x17000, .len = 0x1e8, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x18000, .len = 0x1e8, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x19000, .len = 0x1e8, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a000, .len = 0x1e8, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -141,7 +135,7 @@ static const struct dpu_lm_cfg sm8350_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -149,7 +143,7 @@ static const struct dpu_lm_cfg sm8350_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -157,7 +151,7 @@ static const struct dpu_lm_cfg sm8350_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -165,7 +159,7 @@ static const struct dpu_lm_cfg sm8350_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -173,14 +167,14 @@ static const struct dpu_lm_cfg sm8350_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -191,22 +185,18 @@ static const struct dpu_dspp_cfg sm8350_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -215,42 +205,36 @@ static const struct dpu_pingpong_cfg sm8350_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x69000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x6a000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x6b000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x6c000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x6d000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x6e000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), @@ -279,22 +263,20 @@ static const struct dpu_dsc_cfg sm8350_dsc[] = { { .name = "dce_0_0", .id = DSC_0, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_0, }, { .name = "dce_0_1", .id = DSC_1, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_1, }, { .name = "dce_1_0", .id = DSC_2, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_0, }, { .name = "dce_1_1", .id = DSC_3, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_1, }, }; @@ -303,7 +285,7 @@ static const struct dpu_wb_cfg sm8350_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, @@ -318,7 +300,6 @@ static const struct dpu_intf_cfg sm8350_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x34000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -327,7 +308,6 @@ static const struct dpu_intf_cfg sm8350_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x35000, .len = 0x2c4, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -337,7 +317,6 @@ static const struct dpu_intf_cfg sm8350_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x36000, .len = 0x2c4, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -347,7 +326,6 @@ static const struct dpu_intf_cfg sm8350_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x37000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h index e9625c48c567..8f978b9c3452 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h @@ -32,22 +32,18 @@ static const struct dpu_ctl_cfg sc7280_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x1e8, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x16000, .len = 0x1e8, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x17000, .len = 0x1e8, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x18000, .len = 0x1e8, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, }; @@ -92,21 +88,21 @@ static const struct dpu_lm_cfg sc7280_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sc7180_lm_sblk, .pingpong = PINGPONG_0, .dspp = DSPP_0, }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sc7180_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sc7180_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -117,7 +113,6 @@ static const struct dpu_dspp_cfg sc7280_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -126,28 +121,24 @@ static const struct dpu_pingpong_cfg sc7280_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x69000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x6a000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x6b000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x6c000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = 0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), @@ -159,7 +150,7 @@ static const struct dpu_dsc_cfg sc7280_dsc[] = { { .name = "dce_0_0", .id = DSC_0, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_0, }, }; @@ -168,7 +159,7 @@ static const struct dpu_wb_cfg sc7280_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, @@ -183,7 +174,6 @@ static const struct dpu_intf_cfg sc7280_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x34000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -192,7 +182,6 @@ static const struct dpu_intf_cfg sc7280_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x35000, .len = 0x2c4, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -202,7 +191,6 @@ static const struct dpu_intf_cfg sc7280_intf[] = { }, { .name = "intf_5", .id = INTF_5, .base = 0x39000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h index 139f11321fea..303d33dc7783 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h @@ -21,7 +21,6 @@ static const struct dpu_caps sc8280xp_dpu_caps = { static const struct dpu_mdp_cfg sc8280xp_mdp = { .name = "top_0", .base = 0x0, .len = 0x494, - .features = BIT(DPU_MDP_PERIPH_0_REMOVED), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -39,32 +38,26 @@ static const struct dpu_ctl_cfg sc8280xp_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x16000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x17000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x18000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x19000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -141,7 +134,7 @@ static const struct dpu_lm_cfg sc8280xp_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -149,7 +142,7 @@ static const struct dpu_lm_cfg sc8280xp_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -157,7 +150,7 @@ static const struct dpu_lm_cfg sc8280xp_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -165,7 +158,7 @@ static const struct dpu_lm_cfg sc8280xp_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -173,14 +166,14 @@ static const struct dpu_lm_cfg sc8280xp_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -191,22 +184,18 @@ static const struct dpu_dspp_cfg sc8280xp_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -215,42 +204,36 @@ static const struct dpu_pingpong_cfg sc8280xp_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x69000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x6a000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x6b000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x6c000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x6d000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x6e000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), @@ -279,32 +262,28 @@ static const struct dpu_dsc_cfg sc8280xp_dsc[] = { { .name = "dce_0_0", .id = DSC_0, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_0, }, { .name = "dce_0_1", .id = DSC_1, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_1, }, { .name = "dce_1_0", .id = DSC_2, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_0, }, { .name = "dce_1_1", .id = DSC_3, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_1, }, { .name = "dce_2_0", .id = DSC_4, .base = 0x82000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_0, }, { .name = "dce_2_1", .id = DSC_5, .base = 0x82000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_1, }, }; @@ -314,7 +293,6 @@ static const struct dpu_intf_cfg sc8280xp_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x34000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -323,7 +301,6 @@ static const struct dpu_intf_cfg sc8280xp_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x35000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -333,7 +310,6 @@ static const struct dpu_intf_cfg sc8280xp_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x36000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -343,7 +319,6 @@ static const struct dpu_intf_cfg sc8280xp_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x37000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -352,7 +327,6 @@ static const struct dpu_intf_cfg sc8280xp_intf[] = { }, { .name = "intf_4", .id = INTF_4, .base = 0x38000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -361,7 +335,6 @@ static const struct dpu_intf_cfg sc8280xp_intf[] = { }, { .name = "intf_5", .id = INTF_5, .base = 0x39000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_3, .prog_fetch_lines_worst_case = 24, @@ -370,7 +343,6 @@ static const struct dpu_intf_cfg sc8280xp_intf[] = { }, { .name = "intf_6", .id = INTF_6, .base = 0x3a000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_2, .prog_fetch_lines_worst_case = 24, @@ -379,7 +351,6 @@ static const struct dpu_intf_cfg sc8280xp_intf[] = { }, { .name = "intf_7", .id = INTF_7, .base = 0x3b000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_2, .prog_fetch_lines_worst_case = 24, @@ -388,7 +359,6 @@ static const struct dpu_intf_cfg sc8280xp_intf[] = { }, { .name = "intf_8", .id = INTF_8, .base = 0x3c000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h index 461294143a90..b09a6af4c474 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h @@ -21,7 +21,6 @@ static const struct dpu_caps sm8450_dpu_caps = { static const struct dpu_mdp_cfg sm8450_mdp = { .name = "top_0", .base = 0x0, .len = 0x494, - .features = BIT(DPU_MDP_PERIPH_0_REMOVED), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -40,32 +39,26 @@ static const struct dpu_ctl_cfg sm8450_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x16000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x17000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x18000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x19000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -142,7 +135,7 @@ static const struct dpu_lm_cfg sm8450_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -150,7 +143,7 @@ static const struct dpu_lm_cfg sm8450_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -158,7 +151,7 @@ static const struct dpu_lm_cfg sm8450_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -166,7 +159,7 @@ static const struct dpu_lm_cfg sm8450_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -174,14 +167,14 @@ static const struct dpu_lm_cfg sm8450_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -192,22 +185,18 @@ static const struct dpu_dspp_cfg sm8450_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -216,55 +205,47 @@ static const struct dpu_pingpong_cfg sm8450_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x69000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x6a000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x6b000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x6c000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x6d000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x6e000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), }, { .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, .base = 0x65800, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, { .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, .base = 0x65c00, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, @@ -295,22 +276,20 @@ static const struct dpu_dsc_cfg sm8450_dsc[] = { { .name = "dce_0_0", .id = DSC_0, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_0, }, { .name = "dce_0_1", .id = DSC_1, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_1, }, { .name = "dce_1_0", .id = DSC_2, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_0, }, { .name = "dce_1_1", .id = DSC_3, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_1, }, }; @@ -319,7 +298,7 @@ static const struct dpu_wb_cfg sm8450_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, @@ -334,7 +313,6 @@ static const struct dpu_intf_cfg sm8450_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x34000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -343,7 +321,6 @@ static const struct dpu_intf_cfg sm8450_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x35000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -353,7 +330,6 @@ static const struct dpu_intf_cfg sm8450_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x36000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -363,7 +339,6 @@ static const struct dpu_intf_cfg sm8450_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x37000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h index c248b3b55c41..0f7b4a224e4c 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_4_sa8775p.h @@ -20,7 +20,6 @@ static const struct dpu_caps sa8775p_dpu_caps = { static const struct dpu_mdp_cfg sa8775p_mdp = { .name = "top_0", .base = 0x0, .len = 0x494, - .features = BIT(DPU_MDP_PERIPH_0_REMOVED), .clk_ctrls = { [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, @@ -39,32 +38,26 @@ static const struct dpu_ctl_cfg sa8775p_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x16000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x17000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x18000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x19000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a000, .len = 0x204, - .features = CTL_SC7280_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -141,7 +134,7 @@ static const struct dpu_lm_cfg sa8775p_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -149,7 +142,7 @@ static const struct dpu_lm_cfg sa8775p_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -157,7 +150,7 @@ static const struct dpu_lm_cfg sa8775p_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -165,7 +158,7 @@ static const struct dpu_lm_cfg sa8775p_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -173,14 +166,14 @@ static const struct dpu_lm_cfg sa8775p_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x400, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -191,22 +184,18 @@ static const struct dpu_dspp_cfg sa8775p_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -215,55 +204,47 @@ static const struct dpu_pingpong_cfg sa8775p_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x69000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x6a000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x6b000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x6c000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x6d000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x6e000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), }, { .name = "pingpong_6", .id = PINGPONG_CWB_0, .base = 0x65800, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, { .name = "pingpong_7", .id = PINGPONG_CWB_1, .base = 0x65c00, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, @@ -294,32 +275,28 @@ static const struct dpu_dsc_cfg sa8775p_dsc[] = { { .name = "dce_0_0", .id = DSC_0, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_0, }, { .name = "dce_0_1", .id = DSC_1, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_1, }, { .name = "dce_1_0", .id = DSC_2, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_0, }, { .name = "dce_1_1", .id = DSC_3, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_1, }, { .name = "dce_2_0", .id = DSC_4, .base = 0x82000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_0, }, { .name = "dce_2_1", .id = DSC_5, .base = 0x82000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_1, }, }; @@ -328,7 +305,7 @@ static const struct dpu_wb_cfg sa8775p_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .clk_ctrl = DPU_CLK_CTRL_WB2, @@ -344,7 +321,6 @@ static const struct dpu_intf_cfg sa8775p_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x34000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -353,7 +329,6 @@ static const struct dpu_intf_cfg sa8775p_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x35000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -363,7 +338,6 @@ static const struct dpu_intf_cfg sa8775p_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x36000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -373,7 +347,6 @@ static const struct dpu_intf_cfg sa8775p_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x37000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ .prog_fetch_lines_worst_case = 24, @@ -382,7 +355,6 @@ static const struct dpu_intf_cfg sa8775p_intf[] = { }, { .name = "intf_4", .id = INTF_4, .base = 0x38000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -391,7 +363,6 @@ static const struct dpu_intf_cfg sa8775p_intf[] = { }, { .name = "intf_6", .id = INTF_6, .base = 0x3A000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ .prog_fetch_lines_worst_case = 24, @@ -400,7 +371,6 @@ static const struct dpu_intf_cfg sa8775p_intf[] = { }, { .name = "intf_7", .id = INTF_7, .base = 0x3b000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ .prog_fetch_lines_worst_case = 24, @@ -409,7 +379,6 @@ static const struct dpu_intf_cfg sa8775p_intf[] = { }, { .name = "intf_8", .id = INTF_8, .base = 0x3c000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_1, /* pair with intf_4 for DP MST */ .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h index 59c7fdf28e89..465b6460f875 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h @@ -21,7 +21,6 @@ static const struct dpu_caps sm8550_dpu_caps = { static const struct dpu_mdp_cfg sm8550_mdp = { .name = "top_0", .base = 0, .len = 0x494, - .features = BIT(DPU_MDP_PERIPH_0_REMOVED), .clk_ctrls = { [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, }, @@ -31,32 +30,26 @@ static const struct dpu_ctl_cfg sm8550_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x16000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x17000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x18000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x19000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -139,7 +132,7 @@ static const struct dpu_lm_cfg sm8550_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -147,7 +140,7 @@ static const struct dpu_lm_cfg sm8550_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -155,7 +148,7 @@ static const struct dpu_lm_cfg sm8550_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -163,7 +156,7 @@ static const struct dpu_lm_cfg sm8550_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -171,14 +164,14 @@ static const struct dpu_lm_cfg sm8550_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -189,22 +182,18 @@ static const struct dpu_dspp_cfg sm8550_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -212,55 +201,47 @@ static const struct dpu_pingpong_cfg sm8550_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x69000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x6a000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x6b000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x6c000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x6d000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x6e000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), }, { .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, .base = 0x66000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, { .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, .base = 0x66400, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, @@ -291,22 +272,20 @@ static const struct dpu_dsc_cfg sm8550_dsc[] = { { .name = "dce_0_0", .id = DSC_0, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_0, }, { .name = "dce_0_1", .id = DSC_1, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_1, }, { .name = "dce_1_0", .id = DSC_2, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_0, }, { .name = "dce_1_1", .id = DSC_3, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_1, }, }; @@ -315,7 +294,7 @@ static const struct dpu_wb_cfg sm8550_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .xin_id = 6, @@ -329,7 +308,6 @@ static const struct dpu_intf_cfg sm8550_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x34000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -338,7 +316,6 @@ static const struct dpu_intf_cfg sm8550_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x35000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -348,7 +325,6 @@ static const struct dpu_intf_cfg sm8550_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x36000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -358,7 +334,6 @@ static const struct dpu_intf_cfg sm8550_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x37000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h index 5667d055fbd1..6caa7d40f368 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_1_sar2130p.h @@ -21,7 +21,6 @@ static const struct dpu_caps sar2130p_dpu_caps = { static const struct dpu_mdp_cfg sar2130p_mdp = { .name = "top_0", .base = 0, .len = 0x494, - .features = BIT(DPU_MDP_PERIPH_0_REMOVED), .clk_ctrls = { [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, }, @@ -31,32 +30,26 @@ static const struct dpu_ctl_cfg sar2130p_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x16000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x17000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x18000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x19000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -139,7 +132,7 @@ static const struct dpu_lm_cfg sar2130p_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -147,7 +140,7 @@ static const struct dpu_lm_cfg sar2130p_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -155,7 +148,7 @@ static const struct dpu_lm_cfg sar2130p_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -163,7 +156,7 @@ static const struct dpu_lm_cfg sar2130p_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -171,14 +164,14 @@ static const struct dpu_lm_cfg sar2130p_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -189,22 +182,18 @@ static const struct dpu_dspp_cfg sar2130p_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -212,55 +201,47 @@ static const struct dpu_pingpong_cfg sar2130p_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x69000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x6a000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x6b000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x6c000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x6d000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x6e000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), }, { .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, .base = 0x66000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, { .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, .base = 0x66400, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, @@ -291,22 +272,20 @@ static const struct dpu_dsc_cfg sar2130p_dsc[] = { { .name = "dce_0_0", .id = DSC_0, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_0, }, { .name = "dce_0_1", .id = DSC_1, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_1, }, { .name = "dce_1_0", .id = DSC_2, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_0, }, { .name = "dce_1_1", .id = DSC_3, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_1, }, }; @@ -315,7 +294,7 @@ static const struct dpu_wb_cfg sar2130p_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .xin_id = 6, @@ -329,7 +308,6 @@ static const struct dpu_intf_cfg sar2130p_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x34000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -338,7 +316,6 @@ static const struct dpu_intf_cfg sar2130p_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x35000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -348,7 +325,6 @@ static const struct dpu_intf_cfg sar2130p_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x36000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -358,7 +334,6 @@ static const struct dpu_intf_cfg sar2130p_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x37000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h index 52cc10aec1f9..7243eebb85f3 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_2_x1e80100.h @@ -20,7 +20,6 @@ static const struct dpu_caps x1e80100_dpu_caps = { static const struct dpu_mdp_cfg x1e80100_mdp = { .name = "top_0", .base = 0, .len = 0x494, - .features = BIT(DPU_MDP_PERIPH_0_REMOVED), .clk_ctrls = { [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, }, @@ -30,32 +29,26 @@ static const struct dpu_ctl_cfg x1e80100_ctl[] = { { .name = "ctl_0", .id = CTL_0, .base = 0x15000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, { .name = "ctl_1", .id = CTL_1, .base = 0x16000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), }, { .name = "ctl_2", .id = CTL_2, .base = 0x17000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, { .name = "ctl_3", .id = CTL_3, .base = 0x18000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, { .name = "ctl_4", .id = CTL_4, .base = 0x19000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, { .name = "ctl_5", .id = CTL_5, .base = 0x1a000, .len = 0x290, - .features = CTL_SM8550_MASK, .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; @@ -138,7 +131,7 @@ static const struct dpu_lm_cfg x1e80100_lm[] = { { .name = "lm_0", .id = LM_0, .base = 0x44000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_1, .pingpong = PINGPONG_0, @@ -146,7 +139,7 @@ static const struct dpu_lm_cfg x1e80100_lm[] = { }, { .name = "lm_1", .id = LM_1, .base = 0x45000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_0, .pingpong = PINGPONG_1, @@ -154,7 +147,7 @@ static const struct dpu_lm_cfg x1e80100_lm[] = { }, { .name = "lm_2", .id = LM_2, .base = 0x46000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_3, .pingpong = PINGPONG_2, @@ -162,7 +155,7 @@ static const struct dpu_lm_cfg x1e80100_lm[] = { }, { .name = "lm_3", .id = LM_3, .base = 0x47000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_2, .pingpong = PINGPONG_3, @@ -170,14 +163,14 @@ static const struct dpu_lm_cfg x1e80100_lm[] = { }, { .name = "lm_4", .id = LM_4, .base = 0x48000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_5, .pingpong = PINGPONG_4, }, { .name = "lm_5", .id = LM_5, .base = 0x49000, .len = 0x320, - .features = MIXER_SDM845_MASK, + .features = MIXER_MSM8998_MASK, .sblk = &sdm845_lm_sblk, .lm_pair = LM_4, .pingpong = PINGPONG_5, @@ -188,22 +181,18 @@ static const struct dpu_dspp_cfg x1e80100_dspp[] = { { .name = "dspp_0", .id = DSPP_0, .base = 0x54000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_1", .id = DSPP_1, .base = 0x56000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_2", .id = DSPP_2, .base = 0x58000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, { .name = "dspp_3", .id = DSPP_3, .base = 0x5a000, .len = 0x1800, - .features = DSPP_SC7180_MASK, .sblk = &sdm845_dspp_sblk, }, }; @@ -212,55 +201,47 @@ static const struct dpu_pingpong_cfg x1e80100_pp[] = { { .name = "pingpong_0", .id = PINGPONG_0, .base = 0x69000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), }, { .name = "pingpong_1", .id = PINGPONG_1, .base = 0x6a000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_0, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), }, { .name = "pingpong_2", .id = PINGPONG_2, .base = 0x6b000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), }, { .name = "pingpong_3", .id = PINGPONG_3, .base = 0x6c000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_1, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), }, { .name = "pingpong_4", .id = PINGPONG_4, .base = 0x6d000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), }, { .name = "pingpong_5", .id = PINGPONG_5, .base = 0x6e000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_2, .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), }, { .name = "pingpong_cwb_0", .id = PINGPONG_CWB_0, .base = 0x66000, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, { .name = "pingpong_cwb_1", .id = PINGPONG_CWB_1, .base = 0x66400, .len = 0, - .features = BIT(DPU_PINGPONG_DITHER), .sblk = &sc7280_pp_sblk, .merge_3d = MERGE_3D_3, }, @@ -291,22 +272,20 @@ static const struct dpu_dsc_cfg x1e80100_dsc[] = { { .name = "dce_0_0", .id = DSC_0, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_0, }, { .name = "dce_0_1", .id = DSC_1, .base = 0x80000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2), .sblk = &dsc_sblk_1, }, { .name = "dce_1_0", .id = DSC_2, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_0, }, { .name = "dce_1_1", .id = DSC_3, .base = 0x81000, .len = 0x4, - .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .features = BIT(DPU_DSC_NATIVE_42x_EN), .sblk = &dsc_sblk_1, }, }; @@ -315,7 +294,7 @@ static const struct dpu_wb_cfg x1e80100_wb[] = { { .name = "wb_2", .id = WB_2, .base = 0x65000, .len = 0x2c8, - .features = WB_SM8250_MASK, + .features = WB_SDM845_MASK, .format_list = wb2_formats_rgb_yuv, .num_formats = ARRAY_SIZE(wb2_formats_rgb_yuv), .xin_id = 6, @@ -330,7 +309,6 @@ static const struct dpu_intf_cfg x1e80100_intf[] = { { .name = "intf_0", .id = INTF_0, .base = 0x34000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -339,7 +317,6 @@ static const struct dpu_intf_cfg x1e80100_intf[] = { }, { .name = "intf_1", .id = INTF_1, .base = 0x35000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_0, .prog_fetch_lines_worst_case = 24, @@ -349,7 +326,6 @@ static const struct dpu_intf_cfg x1e80100_intf[] = { }, { .name = "intf_2", .id = INTF_2, .base = 0x36000, .len = 0x300, - .features = INTF_SC7280_MASK, .type = INTF_DSI, .controller_id = MSM_DSI_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -359,7 +335,6 @@ static const struct dpu_intf_cfg x1e80100_intf[] = { }, { .name = "intf_3", .id = INTF_3, .base = 0x37000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_0, /* pair with intf_0 for DP MST */ .prog_fetch_lines_worst_case = 24, @@ -368,7 +343,6 @@ static const struct dpu_intf_cfg x1e80100_intf[] = { }, { .name = "intf_4", .id = INTF_4, .base = 0x38000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_1, .prog_fetch_lines_worst_case = 24, @@ -377,7 +351,6 @@ static const struct dpu_intf_cfg x1e80100_intf[] = { }, { .name = "intf_5", .id = INTF_5, .base = 0x39000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_3, .prog_fetch_lines_worst_case = 24, @@ -386,7 +359,6 @@ static const struct dpu_intf_cfg x1e80100_intf[] = { }, { .name = "intf_6", .id = INTF_6, .base = 0x3A000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_DP, .controller_id = MSM_DP_CONTROLLER_2, .prog_fetch_lines_worst_case = 24, @@ -395,7 +367,6 @@ static const struct dpu_intf_cfg x1e80100_intf[] = { }, { .name = "intf_7", .id = INTF_7, .base = 0x3b000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_2, /* pair with intf_6 for DP MST */ .prog_fetch_lines_worst_case = 24, @@ -404,7 +375,6 @@ static const struct dpu_intf_cfg x1e80100_intf[] = { }, { .name = "intf_8", .id = INTF_8, .base = 0x3c000, .len = 0x280, - .features = INTF_SC7280_MASK, .type = INTF_NONE, .controller_id = MSM_DP_CONTROLLER_1, /* pair with intf_4 for DP MST */ .prog_fetch_lines_worst_case = 24, diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c index a4b0fe0d9899..d4b545448d74 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c @@ -320,14 +320,22 @@ static bool dpu_crtc_get_scanout_position(struct drm_crtc *crtc, } static void _dpu_crtc_setup_blend_cfg(struct dpu_crtc_mixer *mixer, - struct dpu_plane_state *pstate, const struct msm_format *format) + struct dpu_plane_state *pstate, + const struct msm_format *format, + const struct dpu_mdss_version *mdss_ver) { struct dpu_hw_mixer *lm = mixer->hw_lm; - uint32_t blend_op; - uint32_t fg_alpha, bg_alpha; + u32 blend_op; + u32 fg_alpha, bg_alpha, max_alpha; - fg_alpha = pstate->base.alpha >> 8; - bg_alpha = 0xff - fg_alpha; + if (mdss_ver->core_major_ver < 12) { + max_alpha = 0xff; + fg_alpha = pstate->base.alpha >> 8; + } else { + max_alpha = 0x3ff; + fg_alpha = pstate->base.alpha >> 6; + } + bg_alpha = max_alpha - fg_alpha; /* default to opaque blending */ if (pstate->base.pixel_blend_mode == DRM_MODE_BLEND_PIXEL_NONE || @@ -337,7 +345,7 @@ static void _dpu_crtc_setup_blend_cfg(struct dpu_crtc_mixer *mixer, } else if (pstate->base.pixel_blend_mode == DRM_MODE_BLEND_PREMULTI) { blend_op = DPU_BLEND_FG_ALPHA_FG_CONST | DPU_BLEND_BG_ALPHA_FG_PIXEL; - if (fg_alpha != 0xff) { + if (fg_alpha != max_alpha) { bg_alpha = fg_alpha; blend_op |= DPU_BLEND_BG_MOD_ALPHA | DPU_BLEND_BG_INV_MOD_ALPHA; @@ -348,7 +356,7 @@ static void _dpu_crtc_setup_blend_cfg(struct dpu_crtc_mixer *mixer, /* coverage blending */ blend_op = DPU_BLEND_FG_ALPHA_FG_PIXEL | DPU_BLEND_BG_ALPHA_FG_PIXEL; - if (fg_alpha != 0xff) { + if (fg_alpha != max_alpha) { bg_alpha = fg_alpha; blend_op |= DPU_BLEND_FG_MOD_ALPHA | DPU_BLEND_FG_INV_MOD_ALPHA | @@ -402,7 +410,7 @@ static void _dpu_crtc_blend_setup_pipe(struct drm_crtc *crtc, struct dpu_hw_stage_cfg *stage_cfg ) { - uint32_t lm_idx; + u32 lm_idx; enum dpu_sspp sspp_idx; struct drm_plane_state *state; @@ -442,12 +450,13 @@ static void _dpu_crtc_blend_setup_mixer(struct drm_crtc *crtc, struct dpu_plane_state *pstate = NULL; const struct msm_format *format; struct dpu_hw_ctl *ctl = mixer->lm_ctl; - - uint32_t lm_idx; + u32 lm_idx; bool bg_alpha_enable = false; DECLARE_BITMAP(active_fetch, SSPP_MAX); + DECLARE_BITMAP(active_pipes, SSPP_MAX); memset(active_fetch, 0, sizeof(active_fetch)); + memset(active_pipes, 0, sizeof(active_pipes)); drm_atomic_crtc_for_each_plane(plane, crtc) { state = plane->state; if (!state) @@ -465,6 +474,7 @@ static void _dpu_crtc_blend_setup_mixer(struct drm_crtc *crtc, bg_alpha_enable = true; set_bit(pstate->pipe.sspp->idx, active_fetch); + set_bit(pstate->pipe.sspp->idx, active_pipes); _dpu_crtc_blend_setup_pipe(crtc, plane, mixer, cstate->num_mixers, pstate->stage, @@ -473,6 +483,7 @@ static void _dpu_crtc_blend_setup_mixer(struct drm_crtc *crtc, if (pstate->r_pipe.sspp) { set_bit(pstate->r_pipe.sspp->idx, active_fetch); + set_bit(pstate->r_pipe.sspp->idx, active_pipes); _dpu_crtc_blend_setup_pipe(crtc, plane, mixer, cstate->num_mixers, pstate->stage, @@ -482,7 +493,8 @@ static void _dpu_crtc_blend_setup_mixer(struct drm_crtc *crtc, /* blend config update */ for (lm_idx = 0; lm_idx < cstate->num_mixers; lm_idx++) { - _dpu_crtc_setup_blend_cfg(mixer + lm_idx, pstate, format); + _dpu_crtc_setup_blend_cfg(mixer + lm_idx, pstate, format, + ctl->mdss_ver); if (bg_alpha_enable && !format->alpha_enable) mixer[lm_idx].mixer_op_mode = 0; @@ -495,6 +507,9 @@ static void _dpu_crtc_blend_setup_mixer(struct drm_crtc *crtc, if (ctl->ops.set_active_fetch_pipes) ctl->ops.set_active_fetch_pipes(ctl, active_fetch); + if (ctl->ops.set_active_pipes) + ctl->ops.set_active_pipes(ctl, active_pipes); + _dpu_crtc_program_lm_output_roi(crtc); } @@ -510,6 +525,7 @@ static void _dpu_crtc_blend_setup(struct drm_crtc *crtc) struct dpu_hw_ctl *ctl; struct dpu_hw_mixer *lm; struct dpu_hw_stage_cfg stage_cfg; + DECLARE_BITMAP(active_lms, LM_MAX); int i; DRM_DEBUG_ATOMIC("%s\n", dpu_crtc->name); @@ -521,10 +537,16 @@ static void _dpu_crtc_blend_setup(struct drm_crtc *crtc) mixer[i].lm_ctl); if (mixer[i].lm_ctl->ops.set_active_fetch_pipes) mixer[i].lm_ctl->ops.set_active_fetch_pipes(mixer[i].lm_ctl, NULL); + if (mixer[i].lm_ctl->ops.set_active_pipes) + mixer[i].lm_ctl->ops.set_active_pipes(mixer[i].lm_ctl, NULL); + + if (mixer[i].hw_lm->ops.clear_all_blendstages) + mixer[i].hw_lm->ops.clear_all_blendstages(mixer[i].hw_lm); } /* initialize stage cfg */ memset(&stage_cfg, 0, sizeof(struct dpu_hw_stage_cfg)); + memset(active_lms, 0, sizeof(active_lms)); _dpu_crtc_blend_setup_mixer(crtc, dpu_crtc, mixer, &stage_cfg); @@ -538,13 +560,22 @@ static void _dpu_crtc_blend_setup(struct drm_crtc *crtc) ctl->ops.update_pending_flush_mixer(ctl, mixer[i].hw_lm->idx); + set_bit(lm->idx, active_lms); + if (ctl->ops.set_active_lms) + ctl->ops.set_active_lms(ctl, active_lms); + DRM_DEBUG_ATOMIC("lm %d, op_mode 0x%X, ctl %d\n", mixer[i].hw_lm->idx - LM_0, mixer[i].mixer_op_mode, ctl->idx - CTL_0); - ctl->ops.setup_blendstage(ctl, mixer[i].hw_lm->idx, - &stage_cfg); + if (ctl->ops.setup_blendstage) + ctl->ops.setup_blendstage(ctl, mixer[i].hw_lm->idx, + &stage_cfg); + + if (lm->ops.setup_blendstage) + lm->ops.setup_blendstage(lm, mixer[i].hw_lm->idx, + &stage_cfg); } } @@ -711,7 +742,7 @@ void dpu_crtc_frame_event_cb(struct drm_crtc *crtc, u32 event) fevent->event = event; fevent->crtc = crtc; fevent->ts = ktime_get(); - kthread_queue_work(priv->event_thread[crtc_id].worker, &fevent->work); + kthread_queue_work(priv->kms->event_thread[crtc_id].worker, &fevent->work); } /** @@ -880,7 +911,7 @@ static void dpu_crtc_atomic_flush(struct drm_crtc *crtc, dev = crtc->dev; priv = dev->dev_private; - if (crtc->index >= ARRAY_SIZE(priv->event_thread)) { + if (crtc->index >= ARRAY_SIZE(priv->kms->event_thread)) { DPU_ERROR("invalid crtc index[%d]\n", crtc->index); return; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c index c0ed110a7d30..05e5f3463e30 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c @@ -264,7 +264,7 @@ bool dpu_encoder_needs_periph_flush(struct dpu_encoder_phys *phys_enc) mode = &phys_enc->cached_mode; return phys_enc->hw_intf->cap->type == INTF_DP && - msm_dp_needs_periph_flush(priv->dp[disp_info->h_tile_instance[0]], mode); + msm_dp_needs_periph_flush(priv->kms->dp[disp_info->h_tile_instance[0]], mode); } /** @@ -283,9 +283,9 @@ bool dpu_encoder_is_widebus_enabled(const struct drm_encoder *drm_enc) index = disp_info->h_tile_instance[0]; if (disp_info->intf_type == INTF_DP) - return msm_dp_wide_bus_available(priv->dp[index]); + return msm_dp_wide_bus_available(priv->kms->dp[index]); else if (disp_info->intf_type == INTF_DSI) - return msm_dsi_wide_bus_enabled(priv->dsi[index]); + return msm_dsi_wide_bus_enabled(priv->kms->dsi[index]); return false; } @@ -647,7 +647,7 @@ struct drm_dsc_config *dpu_encoder_get_dsc_config(struct drm_encoder *drm_enc) int index = dpu_enc->disp_info.h_tile_instance[0]; if (dpu_enc->disp_info.intf_type == INTF_DSI) - return msm_dsi_get_dsc_config(priv->dsi[index]); + return msm_dsi_get_dsc_config(priv->kms->dsi[index]); return NULL; } @@ -709,7 +709,8 @@ void dpu_encoder_update_topology(struct drm_encoder *drm_enc, if (fb && MSM_FORMAT_IS_YUV(msm_framebuffer_format(fb))) topology->num_cdm++; } else if (disp_info->intf_type == INTF_DP) { - if (msm_dp_is_yuv_420_enabled(priv->dp[disp_info->h_tile_instance[0]], adj_mode)) + if (msm_dp_is_yuv_420_enabled(priv->kms->dp[disp_info->h_tile_instance[0]], + adj_mode)) topology->num_cdm++; } } @@ -980,7 +981,7 @@ static int dpu_encoder_resource_control(struct drm_encoder *drm_enc, return 0; } - queue_delayed_work(priv->wq, &dpu_enc->delayed_off_work, + queue_delayed_work(priv->kms->wq, &dpu_enc->delayed_off_work, msecs_to_jiffies(dpu_enc->idle_timeout)); trace_dpu_enc_rc(DRMID(drm_enc), sw_event, @@ -2195,8 +2196,17 @@ static void dpu_encoder_helper_reset_mixers(struct dpu_encoder_phys *phys_enc) if (ctl->ops.setup_blendstage) ctl->ops.setup_blendstage(ctl, hw_mixer[i]->idx, NULL); + if (hw_mixer[i]->ops.clear_all_blendstages) + hw_mixer[i]->ops.clear_all_blendstages(hw_mixer[i]); + + if (ctl->ops.set_active_lms) + ctl->ops.set_active_lms(ctl, NULL); + if (ctl->ops.set_active_fetch_pipes) ctl->ops.set_active_fetch_pipes(ctl, NULL); + + if (ctl->ops.set_active_pipes) + ctl->ops.set_active_pipes(ctl, NULL); } } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c index a0ba55ab3c89..0ec6d67c7c70 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c @@ -69,7 +69,8 @@ static void _dpu_encoder_phys_cmd_update_intf_cfg( ctl->ops.setup_intf_cfg(ctl, &intf_cfg); /* setup which pp blk will connect to this intf */ - if (test_bit(DPU_CTL_ACTIVE_CFG, &ctl->caps->features) && phys_enc->hw_intf->ops.bind_pingpong_blk) + if (phys_enc->dpu_kms->catalog->mdss_ver->core_major_ver >= 5 && + phys_enc->hw_intf->ops.bind_pingpong_blk) phys_enc->hw_intf->ops.bind_pingpong_blk( phys_enc->hw_intf, phys_enc->hw_pp->idx); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c index 8a618841e3ea..0ba777bda253 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c @@ -94,17 +94,21 @@ static void drm_mode_to_intf_timing_params( timing->vsync_polarity = 0; } - /* for DP/EDP, Shift timings to align it to bottom right */ - if (phys_enc->hw_intf->cap->type == INTF_DP) { + timing->wide_bus_en = dpu_encoder_is_widebus_enabled(phys_enc->parent); + timing->compression_en = dpu_encoder_is_dsc_enabled(phys_enc->parent); + + /* + * For DP/EDP, Shift timings to align it to bottom right. + * wide_bus_en is set for everything excluding SDM845 & + * porch changes cause DisplayPort failure and HDMI tearing. + */ + if (phys_enc->hw_intf->cap->type == INTF_DP && timing->wide_bus_en) { timing->h_back_porch += timing->h_front_porch; timing->h_front_porch = 0; timing->v_back_porch += timing->v_front_porch; timing->v_front_porch = 0; } - timing->wide_bus_en = dpu_encoder_is_widebus_enabled(phys_enc->parent); - timing->compression_en = dpu_encoder_is_dsc_enabled(phys_enc->parent); - /* * for DP, divide the horizonal parameters by 2 when * widebus is enabled @@ -309,8 +313,7 @@ static void dpu_encoder_phys_vid_setup_timing_engine( spin_lock_irqsave(phys_enc->enc_spinlock, lock_flags); phys_enc->hw_intf->ops.setup_timing_gen(phys_enc->hw_intf, - &timing_params, fmt, - phys_enc->dpu_kms->catalog->mdss_ver); + &timing_params, fmt); phys_enc->hw_ctl->ops.setup_intf_cfg(phys_enc->hw_ctl, &intf_cfg); /* setup which pp blk will connect to this intf */ @@ -374,7 +377,7 @@ static void dpu_encoder_phys_vid_underrun_irq(void *arg) static bool dpu_encoder_phys_vid_needs_single_flush( struct dpu_encoder_phys *phys_enc) { - return !(phys_enc->hw_ctl->caps->features & BIT(DPU_CTL_ACTIVE_CFG)) && + return !(phys_enc->dpu_kms->catalog->mdss_ver->core_major_ver >= 5) && phys_enc->split_role != ENC_ROLE_SOLO; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c index 849fea580a4c..56a5b596554d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c @@ -218,7 +218,6 @@ static void dpu_encoder_phys_wb_setup_fb(struct dpu_encoder_phys *phys_enc, static void dpu_encoder_phys_wb_setup_ctl(struct dpu_encoder_phys *phys_enc) { struct dpu_hw_wb *hw_wb; - struct dpu_hw_ctl *ctl; struct dpu_hw_cdm *hw_cdm; if (!phys_enc) { @@ -227,10 +226,9 @@ static void dpu_encoder_phys_wb_setup_ctl(struct dpu_encoder_phys *phys_enc) } hw_wb = phys_enc->hw_wb; - ctl = phys_enc->hw_ctl; hw_cdm = phys_enc->hw_cdm; - if (test_bit(DPU_CTL_ACTIVE_CFG, &ctl->caps->features) && + if (phys_enc->dpu_kms->catalog->mdss_ver->core_major_ver >= 5 && (phys_enc->hw_ctl && phys_enc->hw_ctl->ops.setup_intf_cfg)) { struct dpu_hw_intf_cfg intf_cfg = {0}; @@ -534,7 +532,6 @@ static void dpu_encoder_phys_wb_enable(struct dpu_encoder_phys *phys_enc) static void dpu_encoder_phys_wb_disable(struct dpu_encoder_phys *phys_enc) { struct dpu_hw_wb *hw_wb = phys_enc->hw_wb; - struct dpu_hw_ctl *hw_ctl = phys_enc->hw_ctl; DPU_DEBUG("[wb:%d]\n", hw_wb->idx - WB_0); @@ -556,7 +553,7 @@ static void dpu_encoder_phys_wb_disable(struct dpu_encoder_phys *phys_enc) * WB support is added to those targets will need to add * the legacy teardown sequence as well. */ - if (hw_ctl->caps->features & BIT(DPU_CTL_ACTIVE_CFG)) + if (phys_enc->dpu_kms->catalog->mdss_ver->core_major_ver >= 5) dpu_encoder_helper_phys_cleanup(phys_enc); phys_enc->enable_state = DPU_ENC_DISABLED; @@ -566,7 +563,6 @@ static void dpu_encoder_phys_wb_prepare_wb_job(struct dpu_encoder_phys *phys_enc struct drm_writeback_job *job) { const struct msm_format *format; - struct msm_gem_address_space *aspace; struct dpu_hw_wb_cfg *wb_cfg; int ret; struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); @@ -576,13 +572,12 @@ static void dpu_encoder_phys_wb_prepare_wb_job(struct dpu_encoder_phys *phys_enc wb_enc->wb_job = job; wb_enc->wb_conn = job->connector; - aspace = phys_enc->dpu_kms->base.aspace; wb_cfg = &wb_enc->wb_cfg; memset(wb_cfg, 0, sizeof(struct dpu_hw_wb_cfg)); - ret = msm_framebuffer_prepare(job->fb, aspace, false); + ret = msm_framebuffer_prepare(job->fb, false); if (ret) { DPU_ERROR("prep fb failed, %d\n", ret); return; @@ -596,7 +591,7 @@ static void dpu_encoder_phys_wb_prepare_wb_job(struct dpu_encoder_phys *phys_enc return; } - dpu_format_populate_addrs(aspace, job->fb, &wb_cfg->dest); + dpu_format_populate_addrs(job->fb, &wb_cfg->dest); wb_cfg->dest.width = job->fb->width; wb_cfg->dest.height = job->fb->height; @@ -619,14 +614,11 @@ static void dpu_encoder_phys_wb_cleanup_wb_job(struct dpu_encoder_phys *phys_enc struct drm_writeback_job *job) { struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc); - struct msm_gem_address_space *aspace; if (!job->fb) return; - aspace = phys_enc->dpu_kms->base.aspace; - - msm_framebuffer_cleanup(job->fb, aspace, false); + msm_framebuffer_cleanup(job->fb, false); wb_enc->wb_job = NULL; wb_enc->wb_conn = NULL; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c index 59c9427da7dd..b0d585c5315c 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c @@ -274,15 +274,14 @@ int dpu_format_populate_plane_sizes( return _dpu_format_populate_plane_sizes_linear(fmt, fb, layout); } -static void _dpu_format_populate_addrs_ubwc(struct msm_gem_address_space *aspace, - struct drm_framebuffer *fb, +static void _dpu_format_populate_addrs_ubwc(struct drm_framebuffer *fb, struct dpu_hw_fmt_layout *layout) { const struct msm_format *fmt; uint32_t base_addr = 0; bool meta; - base_addr = msm_framebuffer_iova(fb, aspace, 0); + base_addr = msm_framebuffer_iova(fb, 0); fmt = msm_framebuffer_format(fb); meta = MSM_FORMAT_IS_UBWC(fmt); @@ -355,26 +354,23 @@ static void _dpu_format_populate_addrs_ubwc(struct msm_gem_address_space *aspace } } -static void _dpu_format_populate_addrs_linear(struct msm_gem_address_space *aspace, - struct drm_framebuffer *fb, +static void _dpu_format_populate_addrs_linear(struct drm_framebuffer *fb, struct dpu_hw_fmt_layout *layout) { unsigned int i; /* Populate addresses for simple formats here */ for (i = 0; i < layout->num_planes; ++i) - layout->plane_addr[i] = msm_framebuffer_iova(fb, aspace, i); -} + layout->plane_addr[i] = msm_framebuffer_iova(fb, i); + } /** * dpu_format_populate_addrs - populate buffer addresses based on * mmu, fb, and format found in the fb - * @aspace: address space pointer * @fb: framebuffer pointer * @layout: format layout structure to populate */ -void dpu_format_populate_addrs(struct msm_gem_address_space *aspace, - struct drm_framebuffer *fb, +void dpu_format_populate_addrs(struct drm_framebuffer *fb, struct dpu_hw_fmt_layout *layout) { const struct msm_format *fmt; @@ -384,7 +380,7 @@ void dpu_format_populate_addrs(struct msm_gem_address_space *aspace, /* Populate the addresses given the fb */ if (MSM_FORMAT_IS_UBWC(fmt) || MSM_FORMAT_IS_TILE(fmt)) - _dpu_format_populate_addrs_ubwc(aspace, fb, layout); + _dpu_format_populate_addrs_ubwc(fb, layout); else - _dpu_format_populate_addrs_linear(aspace, fb, layout); + _dpu_format_populate_addrs_linear(fb, layout); } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h index c6145d43aa3f..dc03f522e616 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.h @@ -31,8 +31,7 @@ static inline bool dpu_find_format(u32 format, const u32 *supported_formats, return false; } -void dpu_format_populate_addrs(struct msm_gem_address_space *aspace, - struct drm_framebuffer *fb, +void dpu_format_populate_addrs(struct drm_framebuffer *fb, struct dpu_hw_fmt_layout *layout); int dpu_format_populate_plane_sizes( diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c index c878fe196aeb..e824cd64fd3f 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c @@ -35,12 +35,12 @@ (VIG_MASK | BIT(DPU_SSPP_SCALER_QSEED3_COMPATIBLE)) #define VIG_SDM845_MASK_NO_SDMA \ - (VIG_MASK | BIT(DPU_SSPP_QOS_8LVL) | BIT(DPU_SSPP_SCALER_QSEED3_COMPATIBLE)) + (VIG_MASK | BIT(DPU_SSPP_SCALER_QSEED3_COMPATIBLE)) #define VIG_SDM845_MASK_SDMA \ (VIG_SDM845_MASK_NO_SDMA | BIT(DPU_SSPP_SMART_DMA_V2)) -#define VIG_QCM2290_MASK (VIG_BASE_MASK | BIT(DPU_SSPP_QOS_8LVL)) +#define VIG_QCM2290_MASK (VIG_BASE_MASK) #define DMA_MSM8953_MASK \ (BIT(DPU_SSPP_QOS)) @@ -60,7 +60,7 @@ (VIG_SC7280_MASK | BIT(DPU_SSPP_SMART_DMA_V2)) #define DMA_SDM845_MASK_NO_SDMA \ - (BIT(DPU_SSPP_QOS) | BIT(DPU_SSPP_QOS_8LVL) |\ + (BIT(DPU_SSPP_QOS) | \ BIT(DPU_SSPP_TS_PREFILL) | BIT(DPU_SSPP_TS_PREFILL_REC1) |\ BIT(DPU_SSPP_CDP) | BIT(DPU_SSPP_EXCL_RECT)) @@ -89,39 +89,6 @@ #define MIXER_MSM8998_MASK \ (BIT(DPU_MIXER_SOURCESPLIT)) -#define MIXER_SDM845_MASK \ - (BIT(DPU_MIXER_SOURCESPLIT) | BIT(DPU_DIM_LAYER) | BIT(DPU_MIXER_COMBINED_ALPHA)) - -#define MIXER_QCM2290_MASK \ - (BIT(DPU_DIM_LAYER) | BIT(DPU_MIXER_COMBINED_ALPHA)) - -#define PINGPONG_MSM8996_MASK \ - (BIT(DPU_PINGPONG_DSC)) - -#define PINGPONG_SDM845_MASK \ - (BIT(DPU_PINGPONG_DITHER) | BIT(DPU_PINGPONG_DSC)) - -#define PINGPONG_SM8150_MASK \ - (BIT(DPU_PINGPONG_DITHER) | BIT(DPU_PINGPONG_DSC)) - -#define CTL_SC7280_MASK \ - (BIT(DPU_CTL_ACTIVE_CFG) | \ - BIT(DPU_CTL_FETCH_ACTIVE) | \ - BIT(DPU_CTL_VM_CFG) | \ - BIT(DPU_CTL_DSPP_SUB_BLOCK_FLUSH)) - -#define CTL_SM8550_MASK \ - (CTL_SC7280_MASK | BIT(DPU_CTL_HAS_LAYER_EXT4)) - -#define DSPP_SC7180_MASK BIT(DPU_DSPP_PCC) - -#define INTF_SC7180_MASK \ - (BIT(DPU_INTF_INPUT_CTRL) | \ - BIT(DPU_INTF_STATUS_SUPPORTED) | \ - BIT(DPU_DATA_HCTL_EN)) - -#define INTF_SC7280_MASK (INTF_SC7180_MASK) - #define WB_SDM845_MASK (BIT(DPU_WB_LINE_MODE) | \ BIT(DPU_WB_UBWC) | \ BIT(DPU_WB_YUV_CONFIG) | \ @@ -131,9 +98,6 @@ BIT(DPU_WB_QOS_8LVL) | \ BIT(DPU_WB_CDP)) -#define WB_SM8250_MASK (WB_SDM845_MASK | \ - BIT(DPU_WB_INPUT_CTRL)) - #define DEFAULT_PIXEL_RAM_SIZE (50 * 1024) #define DEFAULT_DPU_LINE_WIDTH 2048 #define DEFAULT_DPU_OUTPUT_LINE_WIDTH 2560 @@ -362,6 +326,9 @@ static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_3_2 = static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_3_3 = _VIG_SBLK(SSPP_SCALER_VER(3, 3)); +static const struct dpu_sspp_sub_blks dpu_vig_sblk_qseed3_3_4 = + _VIG_SBLK(SSPP_SCALER_VER(3, 4)); + static const struct dpu_sspp_sub_blks dpu_rgb_sblk = _RGB_SBLK(); static const struct dpu_sspp_sub_blks dpu_dma_sblk = _DMA_SBLK(); @@ -396,6 +363,16 @@ static const struct dpu_lm_sub_blks sc7180_lm_sblk = { }, }; +static const struct dpu_lm_sub_blks sm8750_lm_sblk = { + .maxwidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .maxblendstages = 11, /* excluding base layer */ + .blendstage_base = { /* offsets relative to mixer base */ + /* 0x40 + n*0x30 */ + 0x40, 0x70, 0xa0, 0xd0, 0x100, 0x130, 0x160, 0x190, 0x1c0, + 0x1f0, 0x220 + }, +}; + static const struct dpu_lm_sub_blks qcm2290_lm_sblk = { .maxwidth = DEFAULT_DPU_LINE_WIDTH, .maxblendstages = 4, /* excluding base layer */ @@ -417,6 +394,11 @@ static const struct dpu_dspp_sub_blks sdm845_dspp_sblk = { .len = 0x90, .version = 0x40000}, }; +static const struct dpu_dspp_sub_blks sm8750_dspp_sblk = { + .pcc = {.name = "pcc", .base = 0x1700, + .len = 0x90, .version = 0x60000}, +}; + /************************************************************* * PINGPONG sub blocks config *************************************************************/ @@ -448,6 +430,16 @@ static const struct dpu_dsc_sub_blks dsc_sblk_1 = { .ctl = {.name = "ctl", .base = 0xF80, .len = 0x10}, }; +static const struct dpu_dsc_sub_blks sm8750_dsc_sblk_0 = { + .enc = {.name = "enc", .base = 0x100, .len = 0x100}, + .ctl = {.name = "ctl", .base = 0xF00, .len = 0x24}, +}; + +static const struct dpu_dsc_sub_blks sm8750_dsc_sblk_1 = { + .enc = {.name = "enc", .base = 0x200, .len = 0x100}, + .ctl = {.name = "ctl", .base = 0xF80, .len = 0x24}, +}; + /************************************************************* * CDM block config *************************************************************/ @@ -738,3 +730,4 @@ static const struct dpu_qos_lut_entry sc7180_qos_nrt[] = { #include "catalog/dpu_9_2_x1e80100.h" #include "catalog/dpu_10_0_sm8650.h" +#include "catalog/dpu_12_0_sm8750.h" diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h index 01dd6e65f777..a78bb2c334e3 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h @@ -29,26 +29,6 @@ #define MAX_XIN_COUNT 16 /** - * MDP TOP BLOCK features - * @DPU_MDP_PANIC_PER_PIPE Panic configuration needs to be done per pipe - * @DPU_MDP_10BIT_SUPPORT, Chipset supports 10 bit pixel formats - * @DPU_MDP_PERIPH_0_REMOVED Indicates that access to periph top0 block results - * in a failure - * @DPU_MDP_VSYNC_SEL Enables vsync source selection via MDP_VSYNC_SEL register - * (moved into INTF block since DPU 5.0.0) - * @DPU_MDP_MAX Maximum value - - */ -enum { - DPU_MDP_PANIC_PER_PIPE = 0x1, - DPU_MDP_10BIT_SUPPORT, - DPU_MDP_AUDIO_SELECT, - DPU_MDP_PERIPH_0_REMOVED, - DPU_MDP_VSYNC_SEL, - DPU_MDP_MAX -}; - -/** * SSPP sub-blocks/features * @DPU_SSPP_SCALER_QSEED2, QSEED2 algorithm support * @DPU_SSPP_SCALER_QSEED3_COMPATIBLE, QSEED3-compatible alogorithm support (includes QSEED3, QSEED3LITE and QSEED4) @@ -57,7 +37,6 @@ enum { * @DPU_SSPP_CSC_10BIT, Support of 10-bit Color space conversion * @DPU_SSPP_CURSOR, SSPP can be used as a cursor layer * @DPU_SSPP_QOS, SSPP support QoS control, danger/safe/creq - * @DPU_SSPP_QOS_8LVL, SSPP support 8-level QoS control * @DPU_SSPP_EXCL_RECT, SSPP supports exclusion rect * @DPU_SSPP_SMART_DMA_V1, SmartDMA 1.0 support * @DPU_SSPP_SMART_DMA_V2, SmartDMA 2.0 support @@ -75,7 +54,6 @@ enum { DPU_SSPP_CSC_10BIT, DPU_SSPP_CURSOR, DPU_SSPP_QOS, - DPU_SSPP_QOS_8LVL, DPU_SSPP_EXCL_RECT, DPU_SSPP_SMART_DMA_V1, DPU_SSPP_SMART_DMA_V2, @@ -88,20 +66,12 @@ enum { /* * MIXER sub-blocks/features - * @DPU_MIXER_LAYER Layer mixer layer blend configuration, * @DPU_MIXER_SOURCESPLIT Layer mixer supports source-split configuration - * @DPU_MIXER_GC Gamma correction block - * @DPU_DIM_LAYER Layer mixer supports dim layer - * @DPU_MIXER_COMBINED_ALPHA Layer mixer has combined alpha register * @DPU_MIXER_MAX maximum value */ enum { - DPU_MIXER_LAYER = 0x1, - DPU_MIXER_SOURCESPLIT, - DPU_MIXER_GC, - DPU_DIM_LAYER, - DPU_MIXER_COMBINED_ALPHA, - DPU_MIXER_MAX + DPU_MIXER_SOURCESPLIT = 0x1, + DPU_MIXER_MAX, }; /** @@ -114,57 +84,16 @@ enum { }; /** - * PINGPONG sub-blocks - * @DPU_PINGPONG_SPLIT PP block supports split fifo - * @DPU_PINGPONG_SLAVE PP block is a suitable slave for split fifo - * @DPU_PINGPONG_DITHER Dither blocks - * @DPU_PINGPONG_DSC PP block supports DSC - * @DPU_PINGPONG_MAX - */ -enum { - DPU_PINGPONG_SPLIT = 0x1, - DPU_PINGPONG_SLAVE, - DPU_PINGPONG_DITHER, - DPU_PINGPONG_DSC, - DPU_PINGPONG_MAX -}; - -/** * CTL sub-blocks * @DPU_CTL_SPLIT_DISPLAY: CTL supports video mode split display - * @DPU_CTL_FETCH_ACTIVE: Active CTL for fetch HW (SSPPs) - * @DPU_CTL_VM_CFG: CTL config to support multiple VMs - * @DPU_CTL_HAS_LAYER_EXT4: CTL has the CTL_LAYER_EXT4 register - * @DPU_CTL_DSPP_BLOCK_FLUSH: CTL config to support dspp sub-block flush * @DPU_CTL_MAX */ enum { DPU_CTL_SPLIT_DISPLAY = 0x1, - DPU_CTL_ACTIVE_CFG, - DPU_CTL_FETCH_ACTIVE, - DPU_CTL_VM_CFG, - DPU_CTL_HAS_LAYER_EXT4, - DPU_CTL_DSPP_SUB_BLOCK_FLUSH, DPU_CTL_MAX }; /** - * INTF sub-blocks - * @DPU_INTF_INPUT_CTRL Supports the setting of pp block from which - * pixel data arrives to this INTF - * @DPU_DATA_HCTL_EN Allows data to be transferred at different rate - * than video timing - * @DPU_INTF_STATUS_SUPPORTED INTF block has INTF_STATUS register - * @DPU_INTF_MAX - */ -enum { - DPU_INTF_INPUT_CTRL = 0x1, - DPU_DATA_HCTL_EN, - DPU_INTF_STATUS_SUPPORTED, - DPU_INTF_MAX -}; - -/** * WB sub-blocks and features * @DPU_WB_LINE_MODE Writeback module supports line/linear mode * @DPU_WB_BLOCK_MODE Writeback module supports block mode read @@ -180,8 +109,6 @@ enum { * @DPU_WB_QOS, Writeback supports QoS control, danger/safe/creq * @DPU_WB_QOS_8LVL, Writeback supports 8-level QoS control * @DPU_WB_CDP Writeback supports client driven prefetch - * @DPU_WB_INPUT_CTRL Writeback supports from which pp block input pixel - * data arrives. * @DPU_WB_CROP CWB supports cropping * @DPU_WB_MAX maximum value */ @@ -195,7 +122,6 @@ enum { DPU_WB_QOS, DPU_WB_QOS_8LVL, DPU_WB_CDP, - DPU_WB_INPUT_CTRL, DPU_WB_CROP, DPU_WB_MAX }; @@ -214,16 +140,11 @@ enum { /** * DSC sub-blocks/features - * @DPU_DSC_OUTPUT_CTRL Configure which PINGPONG block gets - * the pixel output from this DSC. - * @DPU_DSC_HW_REV_1_2 DSC block supports DSC 1.1 and 1.2 * @DPU_DSC_NATIVE_42x_EN Supports NATIVE_422_EN and NATIVE_420_EN encoding * @DPU_DSC_MAX */ enum { - DPU_DSC_OUTPUT_CTRL = 0x1, - DPU_DSC_HW_REV_1_2, - DPU_DSC_NATIVE_42x_EN, + DPU_DSC_NATIVE_42x_EN = 0x1, DPU_DSC_MAX }; @@ -233,14 +154,12 @@ enum { * @id: enum identifying this block * @base: register base offset to mdss * @len: length of hardware block - * @features bit mask identifying sub-blocks/features */ #define DPU_HW_BLK_INFO \ char name[DPU_HW_BLK_NAME_LEN]; \ u32 id; \ u32 base; \ - u32 len; \ - unsigned long features + u32 len /** * struct dpu_scaler_blk: Scaler information @@ -455,7 +374,6 @@ struct dpu_clk_ctrl_reg { /* struct dpu_mdp_cfg : MDP TOP-BLK instance info * @id: index identifying this block * @base: register base offset to mdss - * @features bit mask identifying sub-blocks/features * @clk_ctrls clock control register definition */ struct dpu_mdp_cfg { @@ -471,6 +389,7 @@ struct dpu_mdp_cfg { */ struct dpu_ctl_cfg { DPU_HW_BLK_INFO; + unsigned long features; unsigned int intr_start; }; @@ -486,6 +405,7 @@ struct dpu_ctl_cfg { */ struct dpu_sspp_cfg { DPU_HW_BLK_INFO; + unsigned long features; const struct dpu_sspp_sub_blks *sblk; u32 xin_id; enum dpu_clk_ctrl_type clk_ctrl; @@ -503,6 +423,7 @@ struct dpu_sspp_cfg { */ struct dpu_lm_cfg { DPU_HW_BLK_INFO; + unsigned long features; const struct dpu_lm_sub_blks *sblk; u32 pingpong; u32 dspp; @@ -513,7 +434,6 @@ struct dpu_lm_cfg { * struct dpu_dspp_cfg - information of DSPP blocks * @id enum identifying this block * @base register offset of this block - * @features bit mask identifying sub-blocks/features * supported by this block * @sblk sub-blocks information */ @@ -526,7 +446,6 @@ struct dpu_dspp_cfg { * struct dpu_pingpong_cfg - information of PING-PONG blocks * @id enum identifying this block * @base register offset of this block - * @features bit mask identifying sub-blocks/features * @intr_done: index for PINGPONG done interrupt * @intr_rdptr: index for PINGPONG readpointer done interrupt * @sblk sub-blocks information @@ -543,8 +462,6 @@ struct dpu_pingpong_cfg { * struct dpu_merge_3d_cfg - information of DSPP blocks * @id enum identifying this block * @base register offset of this block - * @features bit mask identifying sub-blocks/features - * supported by this block * @sblk sub-blocks information */ struct dpu_merge_3d_cfg { @@ -562,6 +479,7 @@ struct dpu_merge_3d_cfg { */ struct dpu_dsc_cfg { DPU_HW_BLK_INFO; + unsigned long features; const struct dpu_dsc_sub_blks *sblk; }; @@ -569,7 +487,6 @@ struct dpu_dsc_cfg { * struct dpu_intf_cfg - information of timing engine blocks * @id enum identifying this block * @base register offset of this block - * @features bit mask identifying sub-blocks/features * @type: Interface type(DSI, DP, HDMI) * @controller_id: Controller Instance ID in case of multiple of intf type * @prog_fetch_lines_worst_case Worst case latency num lines needed to prefetch @@ -600,6 +517,7 @@ struct dpu_intf_cfg { */ struct dpu_wb_cfg { DPU_HW_BLK_INFO; + unsigned long features; u8 vbif_idx; u32 maxlinewidth; u32 xin_id; @@ -668,6 +586,7 @@ struct dpu_vbif_qos_tbl { */ struct dpu_vbif_cfg { DPU_HW_BLK_INFO; + unsigned long features; u32 default_ot_rd_limit; u32 default_ot_wr_limit; u32 xin_halt_timeout; @@ -685,7 +604,6 @@ struct dpu_vbif_cfg { * @name string name for debug purposes * @id enum identifying this block * @base register offset of this block - * @features bit mask identifying sub-blocks/features */ struct dpu_cdm_cfg { DPU_HW_BLK_INFO; @@ -860,6 +778,7 @@ extern const struct dpu_mdss_cfg dpu_sm8450_cfg; extern const struct dpu_mdss_cfg dpu_sa8775p_cfg; extern const struct dpu_mdss_cfg dpu_sm8550_cfg; extern const struct dpu_mdss_cfg dpu_sm8650_cfg; +extern const struct dpu_mdss_cfg dpu_sm8750_cfg; extern const struct dpu_mdss_cfg dpu_x1e80100_cfg; #endif /* _DPU_HW_CATALOG_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c index 573e42b06ad0..ac834db2e4c1 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c @@ -42,6 +42,8 @@ #define CTL_INTF_FLUSH 0x110 #define CTL_CDM_FLUSH 0x114 #define CTL_PERIPH_FLUSH 0x128 +#define CTL_PIPE_ACTIVE 0x12c +#define CTL_LAYER_ACTIVE 0x130 #define CTL_INTF_MASTER 0x134 #define CTL_DSPP_n_FLUSH(n) ((0x13C) + ((n) * 4)) @@ -64,6 +66,8 @@ static const u32 fetch_tbl[SSPP_MAX] = {CTL_INVALID_BIT, 16, 17, 18, 19, CTL_INVALID_BIT, CTL_INVALID_BIT, CTL_INVALID_BIT, CTL_INVALID_BIT, 0, 1, 2, 3, 4, 5}; +static const u32 lm_tbl[LM_MAX] = {CTL_INVALID_BIT, 0, 1, 2, 3, 4, 5, 6, 7}; + static int _mixer_stages(const struct dpu_lm_cfg *mixer, int count, enum dpu_lm lm) { @@ -555,7 +559,7 @@ exit: DPU_REG_WRITE(c, CTL_LAYER_EXT(lm), mixercfg[1]); DPU_REG_WRITE(c, CTL_LAYER_EXT2(lm), mixercfg[2]); DPU_REG_WRITE(c, CTL_LAYER_EXT3(lm), mixercfg[3]); - if ((test_bit(DPU_CTL_HAS_LAYER_EXT4, &ctx->caps->features))) + if (ctx->mdss_ver->core_major_ver >= 9) DPU_REG_WRITE(c, CTL_LAYER_EXT4(lm), mixercfg[4]); } @@ -575,7 +579,7 @@ static void dpu_hw_ctl_intf_cfg_v1(struct dpu_hw_ctl *ctx, * per VM. Explicitly disable it until VM support is * added in SW. Power on reset value is not disable. */ - if ((test_bit(DPU_CTL_VM_CFG, &ctx->caps->features))) + if (ctx->mdss_ver->core_major_ver >= 7) mode_sel = CTL_DEFAULT_GROUP_ID << 28; if (cfg->intf_mode_sel == DPU_CTL_MODE_SEL_CMD) @@ -676,11 +680,18 @@ static void dpu_hw_ctl_reset_intf_cfg_v1(struct dpu_hw_ctl *ctx, merge3d_active); } - dpu_hw_ctl_clear_all_blendstages(ctx); + if (ctx->ops.clear_all_blendstages) + ctx->ops.clear_all_blendstages(ctx); + + if (ctx->ops.set_active_lms) + ctx->ops.set_active_lms(ctx, NULL); if (ctx->ops.set_active_fetch_pipes) ctx->ops.set_active_fetch_pipes(ctx, NULL); + if (ctx->ops.set_active_pipes) + ctx->ops.set_active_pipes(ctx, NULL); + if (cfg->intf) { intf_active = DPU_REG_READ(c, CTL_INTF_ACTIVE); intf_active &= ~BIT(cfg->intf - INTF_0); @@ -737,55 +748,39 @@ static void dpu_hw_ctl_set_active_fetch_pipes(struct dpu_hw_ctl *ctx, DPU_REG_WRITE(&ctx->hw, CTL_FETCH_PIPE_ACTIVE, val); } -static void _setup_ctl_ops(struct dpu_hw_ctl_ops *ops, - unsigned long cap) +static void dpu_hw_ctl_set_active_pipes(struct dpu_hw_ctl *ctx, + unsigned long *active_pipes) { - if (cap & BIT(DPU_CTL_ACTIVE_CFG)) { - ops->trigger_flush = dpu_hw_ctl_trigger_flush_v1; - ops->setup_intf_cfg = dpu_hw_ctl_intf_cfg_v1; - ops->reset_intf_cfg = dpu_hw_ctl_reset_intf_cfg_v1; - ops->update_pending_flush_intf = - dpu_hw_ctl_update_pending_flush_intf_v1; + int i; + u32 val = 0; - ops->update_pending_flush_periph = - dpu_hw_ctl_update_pending_flush_periph_v1; + if (active_pipes) { + for (i = 0; i < SSPP_MAX; i++) { + if (test_bit(i, active_pipes) && + fetch_tbl[i] != CTL_INVALID_BIT) + val |= BIT(fetch_tbl[i]); + } + } - ops->update_pending_flush_merge_3d = - dpu_hw_ctl_update_pending_flush_merge_3d_v1; - ops->update_pending_flush_wb = dpu_hw_ctl_update_pending_flush_wb_v1; - ops->update_pending_flush_cwb = dpu_hw_ctl_update_pending_flush_cwb_v1; - ops->update_pending_flush_dsc = - dpu_hw_ctl_update_pending_flush_dsc_v1; - ops->update_pending_flush_cdm = dpu_hw_ctl_update_pending_flush_cdm_v1; - } else { - ops->trigger_flush = dpu_hw_ctl_trigger_flush; - ops->setup_intf_cfg = dpu_hw_ctl_intf_cfg; - ops->update_pending_flush_intf = - dpu_hw_ctl_update_pending_flush_intf; - ops->update_pending_flush_wb = dpu_hw_ctl_update_pending_flush_wb; - ops->update_pending_flush_cdm = dpu_hw_ctl_update_pending_flush_cdm; + DPU_REG_WRITE(&ctx->hw, CTL_PIPE_ACTIVE, val); +} + +static void dpu_hw_ctl_set_active_lms(struct dpu_hw_ctl *ctx, + unsigned long *active_lms) +{ + int i; + u32 val = 0; + + if (active_lms) { + for (i = LM_0; i < LM_MAX; i++) { + if (test_bit(i, active_lms) && + lm_tbl[i] != CTL_INVALID_BIT) + val |= BIT(lm_tbl[i]); + } } - ops->clear_pending_flush = dpu_hw_ctl_clear_pending_flush; - ops->update_pending_flush = dpu_hw_ctl_update_pending_flush; - ops->get_pending_flush = dpu_hw_ctl_get_pending_flush; - ops->get_flush_register = dpu_hw_ctl_get_flush_register; - ops->trigger_start = dpu_hw_ctl_trigger_start; - ops->is_started = dpu_hw_ctl_is_started; - ops->trigger_pending = dpu_hw_ctl_trigger_pending; - ops->reset = dpu_hw_ctl_reset_control; - ops->wait_reset_status = dpu_hw_ctl_wait_reset_status; - ops->clear_all_blendstages = dpu_hw_ctl_clear_all_blendstages; - ops->setup_blendstage = dpu_hw_ctl_setup_blendstage; - ops->update_pending_flush_sspp = dpu_hw_ctl_update_pending_flush_sspp; - ops->update_pending_flush_mixer = dpu_hw_ctl_update_pending_flush_mixer; - if (cap & BIT(DPU_CTL_DSPP_SUB_BLOCK_FLUSH)) - ops->update_pending_flush_dspp = dpu_hw_ctl_update_pending_flush_dspp_sub_blocks; - else - ops->update_pending_flush_dspp = dpu_hw_ctl_update_pending_flush_dspp; - if (cap & BIT(DPU_CTL_FETCH_ACTIVE)) - ops->set_active_fetch_pipes = dpu_hw_ctl_set_active_fetch_pipes; -}; + DPU_REG_WRITE(&ctx->hw, CTL_LAYER_ACTIVE, val); +} /** * dpu_hw_ctl_init() - Initializes the ctl_path hw driver object. @@ -793,12 +788,14 @@ static void _setup_ctl_ops(struct dpu_hw_ctl_ops *ops, * @dev: Corresponding device for devres management * @cfg: ctl_path catalog entry for which driver object is required * @addr: mapped register io address of MDP + * @mdss_ver: dpu core's major and minor versions * @mixer_count: Number of mixers in @mixer * @mixer: Pointer to an array of Layer Mixers defined in the catalog */ struct dpu_hw_ctl *dpu_hw_ctl_init(struct drm_device *dev, const struct dpu_ctl_cfg *cfg, void __iomem *addr, + const struct dpu_mdss_version *mdss_ver, u32 mixer_count, const struct dpu_lm_cfg *mixer) { @@ -812,7 +809,59 @@ struct dpu_hw_ctl *dpu_hw_ctl_init(struct drm_device *dev, c->hw.log_mask = DPU_DBG_MASK_CTL; c->caps = cfg; - _setup_ctl_ops(&c->ops, c->caps->features); + c->mdss_ver = mdss_ver; + + if (mdss_ver->core_major_ver >= 5) { + c->ops.trigger_flush = dpu_hw_ctl_trigger_flush_v1; + c->ops.setup_intf_cfg = dpu_hw_ctl_intf_cfg_v1; + c->ops.reset_intf_cfg = dpu_hw_ctl_reset_intf_cfg_v1; + c->ops.update_pending_flush_intf = + dpu_hw_ctl_update_pending_flush_intf_v1; + + c->ops.update_pending_flush_periph = + dpu_hw_ctl_update_pending_flush_periph_v1; + + c->ops.update_pending_flush_merge_3d = + dpu_hw_ctl_update_pending_flush_merge_3d_v1; + c->ops.update_pending_flush_wb = dpu_hw_ctl_update_pending_flush_wb_v1; + c->ops.update_pending_flush_cwb = dpu_hw_ctl_update_pending_flush_cwb_v1; + c->ops.update_pending_flush_dsc = + dpu_hw_ctl_update_pending_flush_dsc_v1; + c->ops.update_pending_flush_cdm = dpu_hw_ctl_update_pending_flush_cdm_v1; + } else { + c->ops.trigger_flush = dpu_hw_ctl_trigger_flush; + c->ops.setup_intf_cfg = dpu_hw_ctl_intf_cfg; + c->ops.update_pending_flush_intf = + dpu_hw_ctl_update_pending_flush_intf; + c->ops.update_pending_flush_wb = dpu_hw_ctl_update_pending_flush_wb; + c->ops.update_pending_flush_cdm = dpu_hw_ctl_update_pending_flush_cdm; + } + c->ops.clear_pending_flush = dpu_hw_ctl_clear_pending_flush; + c->ops.update_pending_flush = dpu_hw_ctl_update_pending_flush; + c->ops.get_pending_flush = dpu_hw_ctl_get_pending_flush; + c->ops.get_flush_register = dpu_hw_ctl_get_flush_register; + c->ops.trigger_start = dpu_hw_ctl_trigger_start; + c->ops.is_started = dpu_hw_ctl_is_started; + c->ops.trigger_pending = dpu_hw_ctl_trigger_pending; + c->ops.reset = dpu_hw_ctl_reset_control; + c->ops.wait_reset_status = dpu_hw_ctl_wait_reset_status; + if (mdss_ver->core_major_ver < 12) { + c->ops.clear_all_blendstages = dpu_hw_ctl_clear_all_blendstages; + c->ops.setup_blendstage = dpu_hw_ctl_setup_blendstage; + } else { + c->ops.set_active_pipes = dpu_hw_ctl_set_active_pipes; + c->ops.set_active_lms = dpu_hw_ctl_set_active_lms; + } + c->ops.update_pending_flush_sspp = dpu_hw_ctl_update_pending_flush_sspp; + c->ops.update_pending_flush_mixer = dpu_hw_ctl_update_pending_flush_mixer; + if (mdss_ver->core_major_ver >= 7) + c->ops.update_pending_flush_dspp = dpu_hw_ctl_update_pending_flush_dspp_sub_blocks; + else + c->ops.update_pending_flush_dspp = dpu_hw_ctl_update_pending_flush_dspp; + + if (mdss_ver->core_major_ver >= 7) + c->ops.set_active_fetch_pipes = dpu_hw_ctl_set_active_fetch_pipes; + c->idx = cfg->id; c->mixer_count = mixer_count; c->mixer_hw_caps = mixer; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h index feb09590bc8f..15931b22ec94 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.h @@ -258,6 +258,23 @@ struct dpu_hw_ctl_ops { void (*set_active_fetch_pipes)(struct dpu_hw_ctl *ctx, unsigned long *fetch_active); + + /** + * Set active pipes attached to this CTL + * @ctx: ctl path ctx pointer + * @active_pipes: bitmap of enum dpu_sspp + */ + void (*set_active_pipes)(struct dpu_hw_ctl *ctx, + unsigned long *active_pipes); + + /** + * Set active layer mixers attached to this CTL + * @ctx: ctl path ctx pointer + * @active_lms: bitmap of enum dpu_lm + */ + void (*set_active_lms)(struct dpu_hw_ctl *ctx, + unsigned long *active_lms); + }; /** @@ -274,6 +291,7 @@ struct dpu_hw_ctl_ops { * @pending_cwb_flush_mask: pending CWB flush * @pending_dsc_flush_mask: pending DSC flush * @pending_cdm_flush_mask: pending CDM flush + * @mdss_ver: MDSS revision information * @ops: operation list */ struct dpu_hw_ctl { @@ -295,6 +313,8 @@ struct dpu_hw_ctl { u32 pending_dsc_flush_mask; u32 pending_cdm_flush_mask; + const struct dpu_mdss_version *mdss_ver; + /* ops */ struct dpu_hw_ctl_ops ops; }; @@ -312,6 +332,7 @@ static inline struct dpu_hw_ctl *to_dpu_hw_ctl(struct dpu_hw_blk *hw) struct dpu_hw_ctl *dpu_hw_ctl_init(struct drm_device *dev, const struct dpu_ctl_cfg *cfg, void __iomem *addr, + const struct dpu_mdss_version *mdss_ver, u32 mixer_count, const struct dpu_lm_cfg *mixer); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c index cec6d4e8baec..3a149caa7ff4 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.c @@ -181,26 +181,18 @@ static void dpu_hw_dsc_bind_pingpong_blk( DPU_REG_WRITE(c, dsc_ctl_offset, mux_cfg); } -static void _setup_dsc_ops(struct dpu_hw_dsc_ops *ops, - unsigned long cap) -{ - ops->dsc_disable = dpu_hw_dsc_disable; - ops->dsc_config = dpu_hw_dsc_config; - ops->dsc_config_thresh = dpu_hw_dsc_config_thresh; - if (cap & BIT(DPU_DSC_OUTPUT_CTRL)) - ops->dsc_bind_pingpong_blk = dpu_hw_dsc_bind_pingpong_blk; -}; - /** * dpu_hw_dsc_init() - Initializes the DSC hw driver object. * @dev: Corresponding device for devres management * @cfg: DSC catalog entry for which driver object is required * @addr: Mapped register io address of MDP + * @mdss_ver: dpu core's major and minor versions * Return: Error code or allocated dpu_hw_dsc context */ struct dpu_hw_dsc *dpu_hw_dsc_init(struct drm_device *dev, const struct dpu_dsc_cfg *cfg, - void __iomem *addr) + void __iomem *addr, + const struct dpu_mdss_version *mdss_ver) { struct dpu_hw_dsc *c; @@ -213,7 +205,12 @@ struct dpu_hw_dsc *dpu_hw_dsc_init(struct drm_device *dev, c->idx = cfg->id; c->caps = cfg; - _setup_dsc_ops(&c->ops, c->caps->features); + + c->ops.dsc_disable = dpu_hw_dsc_disable; + c->ops.dsc_config = dpu_hw_dsc_config; + c->ops.dsc_config_thresh = dpu_hw_dsc_config_thresh; + if (mdss_ver->core_major_ver >= 5) + c->ops.dsc_bind_pingpong_blk = dpu_hw_dsc_bind_pingpong_blk; return c; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h index fc171bdeca48..b7013c9822d2 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc.h @@ -64,7 +64,8 @@ struct dpu_hw_dsc { struct dpu_hw_dsc *dpu_hw_dsc_init(struct drm_device *dev, const struct dpu_dsc_cfg *cfg, - void __iomem *addr); + void __iomem *addr, + const struct dpu_mdss_version *mdss_ver); struct dpu_hw_dsc *dpu_hw_dsc_init_1_2(struct drm_device *dev, const struct dpu_dsc_cfg *cfg, diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c index b9c433567262..b3395e9c34a1 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dsc_1_2.c @@ -360,8 +360,7 @@ static void dpu_hw_dsc_bind_pingpong_blk_1_2(struct dpu_hw_dsc *hw_dsc, DPU_REG_WRITE(hw, sblk->ctl.base + DSC_CTL, mux_cfg); } -static void _setup_dcs_ops_1_2(struct dpu_hw_dsc_ops *ops, - const unsigned long features) +static void _setup_dcs_ops_1_2(struct dpu_hw_dsc_ops *ops) { ops->dsc_disable = dpu_hw_dsc_disable_1_2; ops->dsc_config = dpu_hw_dsc_config_1_2; @@ -391,7 +390,7 @@ struct dpu_hw_dsc *dpu_hw_dsc_init_1_2(struct drm_device *dev, c->idx = cfg->id; c->caps = cfg; - _setup_dcs_ops_1_2(&c->ops, c->caps->features); + _setup_dcs_ops_1_2(&c->ops); return c; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c index 829ca272873e..11fb1bc54fa9 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_dspp.c @@ -63,13 +63,6 @@ static void dpu_setup_dspp_pcc(struct dpu_hw_dspp *ctx, DPU_REG_WRITE(&ctx->hw, base, PCC_EN); } -static void _setup_dspp_ops(struct dpu_hw_dspp *c, - unsigned long features) -{ - if (test_bit(DPU_DSPP_PCC, &features)) - c->ops.setup_pcc = dpu_setup_dspp_pcc; -} - /** * dpu_hw_dspp_init() - Initializes the DSPP hw driver object. * should be called once before accessing every DSPP. @@ -97,7 +90,8 @@ struct dpu_hw_dspp *dpu_hw_dspp_init(struct drm_device *dev, /* Assign ops */ c->idx = cfg->id; c->cap = cfg; - _setup_dspp_ops(c, c->cap->features); + if (c->cap->sblk->pcc.base) + c->ops.setup_pcc = dpu_setup_dspp_pcc; return c; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c index fb1d25baa518..a80ac82a9625 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c @@ -98,8 +98,7 @@ static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *intf, const struct dpu_hw_intf_timing_params *p, - const struct msm_format *fmt, - const struct dpu_mdss_version *mdss_ver) + const struct msm_format *fmt) { struct dpu_hw_blk_reg_map *c = &intf->hw; u32 hsync_period, vsync_period; @@ -180,7 +179,7 @@ static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *intf, /* TODO: handle DSC+DP case, we only handle DSC+DSI case so far */ if (p->compression_en && !dp_intf && - mdss_ver->core_major_ver >= 7) + intf->mdss_ver->core_major_ver >= 7) intf_cfg2 |= INTF_CFG2_DCE_DATA_COMPRESS; hsync_data_start_x = hsync_start_x; @@ -238,7 +237,7 @@ static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *intf, DPU_REG_WRITE(c, INTF_FRAME_LINE_COUNT_EN, 0x3); DPU_REG_WRITE(c, INTF_CONFIG, intf_cfg); DPU_REG_WRITE(c, INTF_PANEL_FORMAT, panel_format); - if (intf->cap->features & BIT(DPU_DATA_HCTL_EN)) { + if (intf->mdss_ver->core_major_ver >= 5) { /* * DATA_HCTL_EN controls data timing which can be different from * video timing. It is recommended to enable it for all cases, except @@ -309,9 +308,8 @@ static void dpu_hw_intf_get_status( struct dpu_hw_intf_status *s) { struct dpu_hw_blk_reg_map *c = &intf->hw; - unsigned long cap = intf->cap->features; - if (cap & BIT(DPU_INTF_STATUS_SUPPORTED)) + if (intf->mdss_ver->core_major_ver >= 5) s->is_en = DPU_REG_READ(c, INTF_STATUS) & BIT(0); else s->is_en = DPU_REG_READ(c, INTF_TIMING_ENGINE_EN); @@ -580,6 +578,8 @@ struct dpu_hw_intf *dpu_hw_intf_init(struct drm_device *dev, c->idx = cfg->id; c->cap = cfg; + c->mdss_ver = mdss_rev; + c->ops.setup_timing_gen = dpu_hw_intf_setup_timing_engine; c->ops.setup_prg_fetch = dpu_hw_intf_setup_prg_fetch; c->ops.get_status = dpu_hw_intf_get_status; @@ -588,7 +588,7 @@ struct dpu_hw_intf *dpu_hw_intf_init(struct drm_device *dev, c->ops.setup_misr = dpu_hw_intf_setup_misr; c->ops.collect_misr = dpu_hw_intf_collect_misr; - if (cfg->features & BIT(DPU_INTF_INPUT_CTRL)) + if (mdss_rev->core_major_ver >= 5) c->ops.bind_pingpong_blk = dpu_hw_intf_bind_pingpong_blk; /* INTF TE is only for DSI interfaces */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h index 114be272ac0a..f31067a9aaf1 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h @@ -81,8 +81,7 @@ struct dpu_hw_intf_cmd_mode_cfg { struct dpu_hw_intf_ops { void (*setup_timing_gen)(struct dpu_hw_intf *intf, const struct dpu_hw_intf_timing_params *p, - const struct msm_format *fmt, - const struct dpu_mdss_version *mdss_ver); + const struct msm_format *fmt); void (*setup_prg_fetch)(struct dpu_hw_intf *intf, const struct dpu_hw_intf_prog_fetch *fetch); @@ -126,6 +125,8 @@ struct dpu_hw_intf { enum dpu_intf idx; const struct dpu_intf_cfg *cap; + const struct dpu_mdss_version *mdss_ver; + /* ops */ struct dpu_hw_intf_ops ops; }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c index 81b56f066519..e8a76d5192c2 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c @@ -19,12 +19,28 @@ /* These register are offset to mixer base + stage base */ #define LM_BLEND0_OP 0x00 + +/* <v12 DPU with offset to mixer base + stage base */ #define LM_BLEND0_CONST_ALPHA 0x04 #define LM_FG_COLOR_FILL_COLOR_0 0x08 #define LM_FG_COLOR_FILL_COLOR_1 0x0C #define LM_FG_COLOR_FILL_SIZE 0x10 #define LM_FG_COLOR_FILL_XY 0x14 +/* >= v12 DPU */ +#define LM_BG_SRC_SEL_V12 0x14 +#define LM_BG_SRC_SEL_V12_RESET_VALUE 0x0000c0c0 +#define LM_BORDER_COLOR_0_V12 0x1c +#define LM_BORDER_COLOR_1_V12 0x20 + +/* >= v12 DPU with offset to mixer base + stage base */ +#define LM_BLEND0_FG_SRC_SEL_V12 0x04 +#define LM_BLEND0_CONST_ALPHA_V12 0x08 +#define LM_FG_COLOR_FILL_COLOR_0_V12 0x0c +#define LM_FG_COLOR_FILL_COLOR_1_V12 0x10 +#define LM_FG_COLOR_FILL_SIZE_V12 0x14 +#define LM_FG_COLOR_FILL_XY_V12 0x18 + #define LM_BLEND0_FG_ALPHA 0x04 #define LM_BLEND0_BG_ALPHA 0x08 @@ -83,6 +99,22 @@ static void dpu_hw_lm_setup_border_color(struct dpu_hw_mixer *ctx, } } +static void dpu_hw_lm_setup_border_color_v12(struct dpu_hw_mixer *ctx, + struct dpu_mdss_color *color, + u8 border_en) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + + if (border_en) { + DPU_REG_WRITE(c, LM_BORDER_COLOR_0_V12, + (color->color_0 & 0x3ff) | + ((color->color_1 & 0x3ff) << 16)); + DPU_REG_WRITE(c, LM_BORDER_COLOR_1_V12, + (color->color_2 & 0x3ff) | + ((color->color_3 & 0x3ff) << 16)); + } +} + static void dpu_hw_lm_setup_misr(struct dpu_hw_mixer *ctx) { dpu_hw_setup_misr(&ctx->hw, LM_MISR_CTRL, 0x0); @@ -112,6 +144,27 @@ static void dpu_hw_lm_setup_blend_config_combined_alpha(struct dpu_hw_mixer *ctx DPU_REG_WRITE(c, LM_BLEND0_OP + stage_off, blend_op); } +static void +dpu_hw_lm_setup_blend_config_combined_alpha_v12(struct dpu_hw_mixer *ctx, + u32 stage, u32 fg_alpha, + u32 bg_alpha, u32 blend_op) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int stage_off; + u32 const_alpha; + + if (stage == DPU_STAGE_BASE) + return; + + stage_off = _stage_offset(ctx, stage); + if (WARN_ON(stage_off < 0)) + return; + + const_alpha = (bg_alpha & 0x3ff) | ((fg_alpha & 0x3ff) << 16); + DPU_REG_WRITE(c, LM_BLEND0_CONST_ALPHA_V12 + stage_off, const_alpha); + DPU_REG_WRITE(c, LM_BLEND0_OP + stage_off, blend_op); +} + static void dpu_hw_lm_setup_blend_config(struct dpu_hw_mixer *ctx, u32 stage, u32 fg_alpha, u32 bg_alpha, u32 blend_op) { @@ -144,18 +197,146 @@ static void dpu_hw_lm_setup_color3(struct dpu_hw_mixer *ctx, DPU_REG_WRITE(c, LM_OP_MODE, op_mode); } -static void _setup_mixer_ops(struct dpu_hw_lm_ops *ops, - unsigned long features) +static void dpu_hw_lm_setup_color3_v12(struct dpu_hw_mixer *ctx, + uint32_t mixer_op_mode) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int op_mode, stages, stage_off, i; + + stages = ctx->cap->sblk->maxblendstages; + if (stages <= 0) + return; + + for (i = DPU_STAGE_0; i <= stages; i++) { + stage_off = _stage_offset(ctx, i); + if (WARN_ON(stage_off < 0)) + return; + + /* set color_out3 bit in blend0_op when enabled in mixer_op_mode */ + op_mode = DPU_REG_READ(c, LM_BLEND0_OP + stage_off); + if (mixer_op_mode & BIT(i)) + op_mode |= BIT(30); + else + op_mode &= ~BIT(30); + + DPU_REG_WRITE(c, LM_BLEND0_OP + stage_off, op_mode); + } +} + +static int _set_staged_sspp(u32 stage, struct dpu_hw_stage_cfg *stage_cfg, + int pipes_per_stage, u32 *value) { - ops->setup_mixer_out = dpu_hw_lm_setup_out; - if (test_bit(DPU_MIXER_COMBINED_ALPHA, &features)) - ops->setup_blend_config = dpu_hw_lm_setup_blend_config_combined_alpha; + int i; + u32 pipe_type = 0, pipe_id = 0, rec_id = 0; + u32 src_sel[PIPES_PER_STAGE]; + + *value = LM_BG_SRC_SEL_V12_RESET_VALUE; + if (!stage_cfg || !pipes_per_stage) + return 0; + + for (i = 0; i < pipes_per_stage; i++) { + enum dpu_sspp pipe = stage_cfg->stage[stage][i]; + enum dpu_sspp_multirect_index rect_index = stage_cfg->multirect_index[stage][i]; + + src_sel[i] = LM_BG_SRC_SEL_V12_RESET_VALUE; + + if (!pipe) + continue; + + /* translate pipe data to SWI pipe_type, pipe_id */ + if (pipe >= SSPP_DMA0 && pipe <= SSPP_DMA5) { + pipe_type = 0; + pipe_id = pipe - SSPP_DMA0; + } else if (pipe >= SSPP_VIG0 && pipe <= SSPP_VIG3) { + pipe_type = 1; + pipe_id = pipe - SSPP_VIG0; + } else { + DPU_ERROR("invalid rec-%d pipe:%d\n", i, pipe); + return -EINVAL; + } + + /* translate rec data to SWI rec_id */ + if (rect_index == DPU_SSPP_RECT_SOLO || rect_index == DPU_SSPP_RECT_0) { + rec_id = 0; + } else if (rect_index == DPU_SSPP_RECT_1) { + rec_id = 1; + } else { + DPU_ERROR("invalid rec-%d rect_index:%d\n", i, rect_index); + rec_id = 0; + } + + /* calculate SWI value for rec-0 and rec-1 and store it temporary buffer */ + src_sel[i] = (((pipe_type & 0x3) << 6) | ((rec_id & 0x3) << 4) | (pipe_id & 0xf)); + } + + /* calculate final SWI register value for rec-0 and rec-1 */ + *value = 0; + for (i = 0; i < pipes_per_stage; i++) + *value |= src_sel[i] << (i * 8); + + return 0; +} + +static int dpu_hw_lm_setup_blendstage(struct dpu_hw_mixer *ctx, enum dpu_lm lm, + struct dpu_hw_stage_cfg *stage_cfg) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int i, ret, stages, stage_off, pipes_per_stage; + u32 value; + + stages = ctx->cap->sblk->maxblendstages; + if (stages <= 0) + return -EINVAL; + + if (test_bit(DPU_MIXER_SOURCESPLIT, &ctx->cap->features)) + pipes_per_stage = PIPES_PER_STAGE; else - ops->setup_blend_config = dpu_hw_lm_setup_blend_config; - ops->setup_alpha_out = dpu_hw_lm_setup_color3; - ops->setup_border_color = dpu_hw_lm_setup_border_color; - ops->setup_misr = dpu_hw_lm_setup_misr; - ops->collect_misr = dpu_hw_lm_collect_misr; + pipes_per_stage = 1; + + /* + * When stage configuration is empty, we can enable the + * border color by setting the corresponding LAYER_ACTIVE bit + * and un-staging all the pipes from the layer mixer. + */ + if (!stage_cfg) + DPU_REG_WRITE(c, LM_BG_SRC_SEL_V12, LM_BG_SRC_SEL_V12_RESET_VALUE); + + for (i = DPU_STAGE_0; i <= stages; i++) { + stage_off = _stage_offset(ctx, i); + if (stage_off < 0) + return stage_off; + + ret = _set_staged_sspp(i, stage_cfg, pipes_per_stage, &value); + if (ret) + return ret; + + DPU_REG_WRITE(c, LM_BLEND0_FG_SRC_SEL_V12 + stage_off, value); + } + + return 0; +} + +static int dpu_hw_lm_clear_all_blendstages(struct dpu_hw_mixer *ctx) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + int i, stages, stage_off; + + stages = ctx->cap->sblk->maxblendstages; + if (stages <= 0) + return -EINVAL; + + DPU_REG_WRITE(c, LM_BG_SRC_SEL_V12, LM_BG_SRC_SEL_V12_RESET_VALUE); + + for (i = DPU_STAGE_0; i <= stages; i++) { + stage_off = _stage_offset(ctx, i); + if (stage_off < 0) + return stage_off; + + DPU_REG_WRITE(c, LM_BLEND0_FG_SRC_SEL_V12 + stage_off, + LM_BG_SRC_SEL_V12_RESET_VALUE); + } + + return 0; } /** @@ -164,10 +345,12 @@ static void _setup_mixer_ops(struct dpu_hw_lm_ops *ops, * @dev: Corresponding device for devres management * @cfg: mixer catalog entry for which driver object is required * @addr: mapped register io address of MDP + * @mdss_ver: DPU core's major and minor versions */ struct dpu_hw_mixer *dpu_hw_lm_init(struct drm_device *dev, const struct dpu_lm_cfg *cfg, - void __iomem *addr) + void __iomem *addr, + const struct dpu_mdss_version *mdss_ver) { struct dpu_hw_mixer *c; @@ -186,7 +369,24 @@ struct dpu_hw_mixer *dpu_hw_lm_init(struct drm_device *dev, /* Assign ops */ c->idx = cfg->id; c->cap = cfg; - _setup_mixer_ops(&c->ops, c->cap->features); + c->ops.setup_mixer_out = dpu_hw_lm_setup_out; + if (mdss_ver->core_major_ver >= 12) + c->ops.setup_blend_config = dpu_hw_lm_setup_blend_config_combined_alpha_v12; + else if (mdss_ver->core_major_ver >= 4) + c->ops.setup_blend_config = dpu_hw_lm_setup_blend_config_combined_alpha; + else + c->ops.setup_blend_config = dpu_hw_lm_setup_blend_config; + if (mdss_ver->core_major_ver < 12) { + c->ops.setup_alpha_out = dpu_hw_lm_setup_color3; + c->ops.setup_border_color = dpu_hw_lm_setup_border_color; + } else { + c->ops.setup_alpha_out = dpu_hw_lm_setup_color3_v12; + c->ops.setup_blendstage = dpu_hw_lm_setup_blendstage; + c->ops.clear_all_blendstages = dpu_hw_lm_clear_all_blendstages; + c->ops.setup_border_color = dpu_hw_lm_setup_border_color_v12; + } + c->ops.setup_misr = dpu_hw_lm_setup_misr; + c->ops.collect_misr = dpu_hw_lm_collect_misr; return c; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h index 6f60fa9b3cd7..1b9ecd082d7f 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h @@ -11,6 +11,7 @@ #include "dpu_hw_util.h" struct dpu_hw_mixer; +struct dpu_hw_stage_cfg; struct dpu_hw_mixer_cfg { u32 out_width; @@ -49,6 +50,23 @@ struct dpu_hw_lm_ops { void (*setup_alpha_out)(struct dpu_hw_mixer *ctx, uint32_t mixer_op); /** + * Clear layer mixer to pipe configuration + * @ctx : mixer ctx pointer + * Returns: 0 on success or -error + */ + int (*clear_all_blendstages)(struct dpu_hw_mixer *ctx); + + /** + * Configure layer mixer to pipe configuration + * @ctx : mixer ctx pointer + * @lm : layer mixer enumeration + * @stage_cfg : blend stage configuration + * Returns: 0 on success or -error + */ + int (*setup_blendstage)(struct dpu_hw_mixer *ctx, enum dpu_lm lm, + struct dpu_hw_stage_cfg *stage_cfg); + + /** * setup_border_color : enable/disable border color */ void (*setup_border_color)(struct dpu_hw_mixer *ctx, @@ -95,6 +113,7 @@ static inline struct dpu_hw_mixer *to_dpu_hw_mixer(struct dpu_hw_blk *hw) struct dpu_hw_mixer *dpu_hw_lm_init(struct drm_device *dev, const struct dpu_lm_cfg *cfg, - void __iomem *addr); + void __iomem *addr, + const struct dpu_mdss_version *mdss_ver); #endif /*_DPU_HW_LM_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c index 0b3325f9c870..83b1dbecddd2 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_merge3d.c @@ -33,8 +33,7 @@ static void dpu_hw_merge_3d_setup_3d_mode(struct dpu_hw_merge_3d *merge_3d, } } -static void _setup_merge_3d_ops(struct dpu_hw_merge_3d *c, - unsigned long features) +static void _setup_merge_3d_ops(struct dpu_hw_merge_3d *c) { c->ops.setup_3d_mode = dpu_hw_merge_3d_setup_3d_mode; }; @@ -62,7 +61,7 @@ struct dpu_hw_merge_3d *dpu_hw_merge_3d_init(struct drm_device *dev, c->idx = cfg->id; c->caps = cfg; - _setup_merge_3d_ops(c, c->caps->features); + _setup_merge_3d_ops(c); return c; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c index 36c0ec775b92..138071be5649 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c @@ -319,13 +319,13 @@ struct dpu_hw_pingpong *dpu_hw_pingpong_init(struct drm_device *dev, c->ops.disable_autorefresh = dpu_hw_pp_disable_autorefresh; } - if (test_bit(DPU_PINGPONG_DSC, &cfg->features)) { + if (mdss_rev->core_major_ver < 7) { c->ops.setup_dsc = dpu_hw_pp_setup_dsc; c->ops.enable_dsc = dpu_hw_pp_dsc_enable; c->ops.disable_dsc = dpu_hw_pp_dsc_disable; } - if (test_bit(DPU_PINGPONG_DITHER, &cfg->features)) + if (mdss_rev->core_major_ver >= 3) c->ops.setup_dither = dpu_hw_pp_setup_dither; return c; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c index 32c7c8084553..6f1fc790ad6d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c @@ -10,11 +10,11 @@ #include "dpu_hw_sspp.h" #include "dpu_kms.h" -#include "msm_mdss.h" - #include <drm/drm_file.h> #include <drm/drm_managed.h> +#include <linux/soc/qcom/ubwc.h> + #define DPU_FETCH_CONFIG_RESET_VALUE 0x00000087 /* SSPP registers */ @@ -543,7 +543,7 @@ static void dpu_hw_sspp_setup_qos_lut(struct dpu_hw_sspp *ctx, return; _dpu_hw_setup_qos_lut(&ctx->hw, SSPP_DANGER_LUT, - test_bit(DPU_SSPP_QOS_8LVL, &ctx->cap->features), + ctx->mdss_ver->core_major_ver >= 4, cfg); } @@ -684,7 +684,7 @@ int _dpu_hw_sspp_init_debugfs(struct dpu_hw_sspp *hw_pipe, struct dpu_kms *kms, struct dpu_hw_sspp *dpu_hw_sspp_init(struct drm_device *dev, const struct dpu_sspp_cfg *cfg, void __iomem *addr, - const struct msm_mdss_data *mdss_data, + const struct qcom_ubwc_cfg_data *mdss_data, const struct dpu_mdss_version *mdss_rev) { struct dpu_hw_sspp *hw_pipe; @@ -703,6 +703,9 @@ struct dpu_hw_sspp *dpu_hw_sspp_init(struct drm_device *dev, hw_pipe->ubwc = mdss_data; hw_pipe->idx = cfg->id; hw_pipe->cap = cfg; + + hw_pipe->mdss_ver = mdss_rev; + _setup_layer_ops(hw_pipe, hw_pipe->cap->features, mdss_rev); return hw_pipe; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h index 56a0edf2a57c..bdac5c04bf79 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h @@ -308,12 +308,14 @@ struct dpu_hw_sspp_ops { struct dpu_hw_sspp { struct dpu_hw_blk base; struct dpu_hw_blk_reg_map hw; - const struct msm_mdss_data *ubwc; + const struct qcom_ubwc_cfg_data *ubwc; /* Pipe */ enum dpu_sspp idx; const struct dpu_sspp_cfg *cap; + const struct dpu_mdss_version *mdss_ver; + /* Ops */ struct dpu_hw_sspp_ops ops; }; @@ -323,7 +325,7 @@ struct dpu_kms; struct dpu_hw_sspp *dpu_hw_sspp_init(struct drm_device *dev, const struct dpu_sspp_cfg *cfg, void __iomem *addr, - const struct msm_mdss_data *mdss_data, + const struct qcom_ubwc_cfg_data *mdss_data, const struct dpu_mdss_version *mdss_rev); int _dpu_hw_sspp_init_debugfs(struct dpu_hw_sspp *hw_pipe, struct dpu_kms *kms, diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c index 562a3f4c5238..96dc10589bee 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c @@ -264,15 +264,15 @@ static void dpu_hw_dp_phy_intf_sel(struct dpu_hw_mdp *mdp, } static void _setup_mdp_ops(struct dpu_hw_mdp_ops *ops, - unsigned long cap, const struct dpu_mdss_version *mdss_rev) + const struct dpu_mdss_version *mdss_rev) { ops->setup_split_pipe = dpu_hw_setup_split_pipe; ops->setup_clk_force_ctrl = dpu_hw_setup_clk_force_ctrl; ops->get_danger_status = dpu_hw_get_danger_status; - if (cap & BIT(DPU_MDP_VSYNC_SEL)) + if (mdss_rev->core_major_ver < 5) ops->setup_vsync_source = dpu_hw_setup_vsync_sel; - else if (!(cap & BIT(DPU_MDP_PERIPH_0_REMOVED))) + else if (mdss_rev->core_major_ver < 8) ops->setup_vsync_source = dpu_hw_setup_wd_timer; ops->get_safe_status = dpu_hw_get_safe_status; @@ -280,7 +280,8 @@ static void _setup_mdp_ops(struct dpu_hw_mdp_ops *ops, if (mdss_rev->core_major_ver >= 5) ops->dp_phy_intf_sel = dpu_hw_dp_phy_intf_sel; - if (cap & BIT(DPU_MDP_AUDIO_SELECT)) + if (mdss_rev->core_major_ver == 4 || + mdss_rev->core_major_ver == 5) ops->intf_audio_select = dpu_hw_intf_audio_select; } @@ -312,7 +313,7 @@ struct dpu_hw_mdp *dpu_hw_mdptop_init(struct drm_device *dev, * Assign ops */ mdp->caps = cfg; - _setup_mdp_ops(&mdp->ops, mdp->caps->features, mdss_rev); + _setup_mdp_ops(&mdp->ops, mdss_rev); return mdp; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c index 4853e516c487..478a091aeccf 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c @@ -208,7 +208,7 @@ static void _setup_wb_ops(struct dpu_hw_wb_ops *ops, if (test_bit(DPU_WB_CDP, &features)) ops->setup_cdp = dpu_hw_wb_setup_cdp; - if (test_bit(DPU_WB_INPUT_CTRL, &features)) + if (mdss_rev->core_major_ver >= 5) ops->bind_pingpong_blk = dpu_hw_wb_bind_pingpong_blk; if (mdss_rev->core_major_ver >= 9) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c index 1fd82b6747e9..12dcb32b4724 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c @@ -20,9 +20,10 @@ #include <drm/drm_vblank.h> #include <drm/drm_writeback.h> +#include <linux/soc/qcom/ubwc.h> + #include "msm_drv.h" #include "msm_mmu.h" -#include "msm_mdss.h" #include "msm_gem.h" #include "disp/msm_disp_snapshot.h" @@ -582,7 +583,7 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev, struct msm_display_info info; int i, rc = 0; - if (!(priv->dsi[0] || priv->dsi[1])) + if (!(priv->kms->dsi[0] || priv->kms->dsi[1])) return rc; /* @@ -593,26 +594,26 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev, * * TODO: Support swapping DSI0 and DSI1 in the bonded setup. */ - for (i = 0; i < ARRAY_SIZE(priv->dsi); i++) { + for (i = 0; i < ARRAY_SIZE(priv->kms->dsi); i++) { int other = (i + 1) % 2; - if (!priv->dsi[i]) + if (!priv->kms->dsi[i]) continue; - if (msm_dsi_is_bonded_dsi(priv->dsi[i]) && - !msm_dsi_is_master_dsi(priv->dsi[i])) + if (msm_dsi_is_bonded_dsi(priv->kms->dsi[i]) && + !msm_dsi_is_master_dsi(priv->kms->dsi[i])) continue; memset(&info, 0, sizeof(info)); info.intf_type = INTF_DSI; info.h_tile_instance[info.num_of_h_tiles++] = i; - if (msm_dsi_is_bonded_dsi(priv->dsi[i])) + if (msm_dsi_is_bonded_dsi(priv->kms->dsi[i])) info.h_tile_instance[info.num_of_h_tiles++] = other; - info.is_cmd_mode = msm_dsi_is_cmd_mode(priv->dsi[i]); + info.is_cmd_mode = msm_dsi_is_cmd_mode(priv->kms->dsi[i]); - rc = dpu_kms_dsi_set_te_source(&info, priv->dsi[i]); + rc = dpu_kms_dsi_set_te_source(&info, priv->kms->dsi[i]); if (rc) { DPU_ERROR("failed to identify TE source for dsi display\n"); return rc; @@ -624,15 +625,15 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev, return PTR_ERR(encoder); } - rc = msm_dsi_modeset_init(priv->dsi[i], dev, encoder); + rc = msm_dsi_modeset_init(priv->kms->dsi[i], dev, encoder); if (rc) { DPU_ERROR("modeset_init failed for dsi[%d], rc = %d\n", i, rc); break; } - if (msm_dsi_is_bonded_dsi(priv->dsi[i]) && priv->dsi[other]) { - rc = msm_dsi_modeset_init(priv->dsi[other], dev, encoder); + if (msm_dsi_is_bonded_dsi(priv->kms->dsi[i]) && priv->kms->dsi[other]) { + rc = msm_dsi_modeset_init(priv->kms->dsi[other], dev, encoder); if (rc) { DPU_ERROR("modeset_init failed for dsi[%d], rc = %d\n", other, rc); @@ -654,8 +655,8 @@ static int _dpu_kms_initialize_displayport(struct drm_device *dev, int rc; int i; - for (i = 0; i < ARRAY_SIZE(priv->dp); i++) { - if (!priv->dp[i]) + for (i = 0; i < ARRAY_SIZE(priv->kms->dp); i++) { + if (!priv->kms->dp[i]) continue; memset(&info, 0, sizeof(info)); @@ -670,7 +671,7 @@ static int _dpu_kms_initialize_displayport(struct drm_device *dev, } yuv_supported = !!dpu_kms->catalog->cdm; - rc = msm_dp_modeset_init(priv->dp[i], dev, encoder, yuv_supported); + rc = msm_dp_modeset_init(priv->kms->dp[i], dev, encoder, yuv_supported); if (rc) { DPU_ERROR("modeset_init failed for DP, rc = %d\n", rc); return rc; @@ -688,7 +689,7 @@ static int _dpu_kms_initialize_hdmi(struct drm_device *dev, struct msm_display_info info; int rc; - if (!priv->hdmi) + if (!priv->kms->hdmi) return 0; memset(&info, 0, sizeof(info)); @@ -702,7 +703,7 @@ static int _dpu_kms_initialize_hdmi(struct drm_device *dev, return PTR_ERR(encoder); } - rc = msm_hdmi_modeset_init(priv->hdmi, dev, encoder); + rc = msm_hdmi_modeset_init(priv->kms->hdmi, dev, encoder); if (rc) { DPU_ERROR("modeset_init failed for DP, rc = %d\n", rc); return rc; @@ -874,12 +875,11 @@ static int _dpu_kms_drm_obj_init(struct dpu_kms *dpu_kms) ret = PTR_ERR(crtc); return ret; } - priv->num_crtcs++; } /* All CRTCs are compatible with all encoders */ drm_for_each_encoder(encoder, dev) - encoder->possible_crtcs = (1 << priv->num_crtcs) - 1; + encoder->possible_crtcs = (1 << dev->mode_config.num_crtc) - 1; return 0; } @@ -1022,7 +1022,7 @@ static void dpu_kms_mdp_snapshot(struct msm_disp_state *disp_state, struct msm_k dpu_kms->mmio + cat->wb[i].base, "%s", cat->wb[i].name); - if (cat->mdp[0].features & BIT(DPU_MDP_PERIPH_0_REMOVED)) { + if (dpu_kms->catalog->mdss_ver->core_major_ver >= 8) { msm_disp_snapshot_add_block(disp_state, MDP_PERIPH_TOP0, dpu_kms->mmio + cat->mdp[0].base, "top"); msm_disp_snapshot_add_block(disp_state, cat->mdp[0].len - MDP_PERIPH_TOP0_END, @@ -1043,7 +1043,7 @@ static void dpu_kms_mdp_snapshot(struct msm_disp_state *disp_state, struct msm_k msm_disp_snapshot_add_block(disp_state, cat->dsc[i].len, base, "%s", cat->dsc[i].name); - if (cat->dsc[i].features & BIT(DPU_DSC_HW_REV_1_2)) { + if (cat->mdss_ver->core_major_ver >= 7) { struct dpu_dsc_blk enc = cat->dsc[i].sblk->enc; struct dpu_dsc_blk ctl = cat->dsc[i].sblk->ctl; @@ -1095,26 +1095,26 @@ static void _dpu_kms_mmu_destroy(struct dpu_kms *dpu_kms) { struct msm_mmu *mmu; - if (!dpu_kms->base.aspace) + if (!dpu_kms->base.vm) return; - mmu = dpu_kms->base.aspace->mmu; + mmu = to_msm_vm(dpu_kms->base.vm)->mmu; mmu->funcs->detach(mmu); - msm_gem_address_space_put(dpu_kms->base.aspace); + drm_gpuvm_put(dpu_kms->base.vm); - dpu_kms->base.aspace = NULL; + dpu_kms->base.vm = NULL; } static int _dpu_kms_mmu_init(struct dpu_kms *dpu_kms) { - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; - aspace = msm_kms_init_aspace(dpu_kms->dev); - if (IS_ERR(aspace)) - return PTR_ERR(aspace); + vm = msm_kms_init_vm(dpu_kms->dev); + if (IS_ERR(vm)) + return PTR_ERR(vm); - dpu_kms->base.aspace = aspace; + dpu_kms->base.vm = vm; return 0; } @@ -1189,10 +1189,10 @@ static int dpu_kms_hw_init(struct msm_kms *kms) goto err_pm_put; } - dpu_kms->mdss = msm_mdss_get_mdss_data(dpu_kms->pdev->dev.parent); + dpu_kms->mdss = qcom_ubwc_config_get_data(); if (IS_ERR(dpu_kms->mdss)) { rc = PTR_ERR(dpu_kms->mdss); - DPU_ERROR("failed to get MDSS data: %d\n", rc); + DPU_ERROR("failed to get UBWC config data: %d\n", rc); goto err_pm_put; } @@ -1533,6 +1533,7 @@ static const struct of_device_id dpu_dt_match[] = { { .compatible = "qcom,sm8450-dpu", .data = &dpu_sm8450_cfg, }, { .compatible = "qcom,sm8550-dpu", .data = &dpu_sm8550_cfg, }, { .compatible = "qcom,sm8650-dpu", .data = &dpu_sm8650_cfg, }, + { .compatible = "qcom,sm8750-dpu", .data = &dpu_sm8750_cfg, }, { .compatible = "qcom,x1e80100-dpu", .data = &dpu_x1e80100_cfg, }, {} }; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h index a57ec2ec1060..993cf512f8c5 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h @@ -60,7 +60,7 @@ struct dpu_kms { struct msm_kms base; struct drm_device *dev; const struct dpu_mdss_cfg *catalog; - const struct msm_mdss_data *mdss; + const struct qcom_ubwc_cfg_data *mdss; /* io/register spaces: */ void __iomem *mmio, *vbif[VBIF_MAX]; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c index 421138bc3cb7..01171c535a27 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c @@ -17,8 +17,9 @@ #include <drm/drm_framebuffer.h> #include <drm/drm_gem_atomic_helper.h> +#include <linux/soc/qcom/ubwc.h> + #include "msm_drv.h" -#include "msm_mdss.h" #include "dpu_kms.h" #include "dpu_hw_sspp.h" #include "dpu_hw_util.h" @@ -71,7 +72,7 @@ static const uint32_t qcom_compressed_supported_formats[] = { /* * struct dpu_plane - local dpu plane structure - * @aspace: address space pointer + * @vm: address space pointer * @csc_ptr: Points to dpu_csc_cfg structure to use for current * @catalog: Points to dpu catalog structure * @revalidate: force revalidation of all the plane properties @@ -646,7 +647,6 @@ static int dpu_plane_prepare_fb(struct drm_plane *plane, struct drm_framebuffer *fb = new_state->fb; struct dpu_plane *pdpu = to_dpu_plane(plane); struct dpu_plane_state *pstate = to_dpu_plane_state(new_state); - struct dpu_kms *kms = _dpu_plane_get_kms(&pdpu->base); int ret; if (!new_state->fb) @@ -654,9 +654,6 @@ static int dpu_plane_prepare_fb(struct drm_plane *plane, DPU_DEBUG_PLANE(pdpu, "FB[%u]\n", fb->base.id); - /* cache aspace */ - pstate->aspace = kms->base.aspace; - /* * TODO: Need to sort out the msm_framebuffer_prepare() call below so * we can use msm_atomic_prepare_fb() instead of doing the @@ -664,13 +661,10 @@ static int dpu_plane_prepare_fb(struct drm_plane *plane, */ drm_gem_plane_helper_prepare_fb(plane, new_state); - if (pstate->aspace) { - ret = msm_framebuffer_prepare(new_state->fb, - pstate->aspace, pstate->needs_dirtyfb); - if (ret) { - DPU_ERROR("failed to prepare framebuffer\n"); - return ret; - } + ret = msm_framebuffer_prepare(new_state->fb, pstate->needs_dirtyfb); + if (ret) { + DPU_ERROR("failed to prepare framebuffer\n"); + return ret; } return 0; @@ -689,8 +683,7 @@ static void dpu_plane_cleanup_fb(struct drm_plane *plane, DPU_DEBUG_PLANE(pdpu, "FB[%u]\n", old_state->fb->base.id); - msm_framebuffer_cleanup(old_state->fb, old_pstate->aspace, - old_pstate->needs_dirtyfb); + msm_framebuffer_cleanup(old_state->fb, old_pstate->needs_dirtyfb); } static int dpu_plane_check_inline_rotation(struct dpu_plane *pdpu, @@ -1457,7 +1450,7 @@ static void dpu_plane_sspp_atomic_update(struct drm_plane *plane, pstate->needs_qos_remap |= (is_rt_pipe != pdpu->is_rt_pipe); pdpu->is_rt_pipe = is_rt_pipe; - dpu_format_populate_addrs(pstate->aspace, new_state->fb, &pstate->layout); + dpu_format_populate_addrs(new_state->fb, &pstate->layout); DPU_DEBUG_PLANE(pdpu, "FB[%u] " DRM_RECT_FP_FMT "->crtc%u " DRM_RECT_FMT ", %p4cc ubwc %d\n", fb->base.id, DRM_RECT_FP_ARG(&state->src), diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h index acd5725175cd..a3a6e9028333 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.h @@ -17,7 +17,6 @@ /** * struct dpu_plane_state: Define dpu extension of drm plane state object * @base: base drm plane state object - * @aspace: pointer to address space for input/output buffers * @pipe: software pipe description * @r_pipe: software pipe description of the second pipe * @pipe_cfg: software pipe configuration @@ -34,7 +33,6 @@ */ struct dpu_plane_state { struct drm_plane_state base; - struct msm_gem_address_space *aspace; struct dpu_sw_pipe pipe; struct dpu_sw_pipe r_pipe; struct dpu_sw_pipe_cfg pipe_cfg; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c index 2e296f79cba1..25382120cb1a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c @@ -40,7 +40,7 @@ static inline bool reserved_by_other(uint32_t *res_map, int idx, int dpu_rm_init(struct drm_device *dev, struct dpu_rm *rm, const struct dpu_mdss_cfg *cat, - const struct msm_mdss_data *mdss_data, + const struct qcom_ubwc_cfg_data *mdss_data, void __iomem *mmio) { int rc, i; @@ -60,7 +60,7 @@ int dpu_rm_init(struct drm_device *dev, struct dpu_hw_mixer *hw; const struct dpu_lm_cfg *lm = &cat->mixer[i]; - hw = dpu_hw_lm_init(dev, lm, mmio); + hw = dpu_hw_lm_init(dev, lm, mmio, cat->mdss_ver); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed lm object creation: err %d\n", rc); @@ -142,7 +142,7 @@ int dpu_rm_init(struct drm_device *dev, struct dpu_hw_ctl *hw; const struct dpu_ctl_cfg *ctl = &cat->ctl[i]; - hw = dpu_hw_ctl_init(dev, ctl, mmio, cat->mixer_count, cat->mixer); + hw = dpu_hw_ctl_init(dev, ctl, mmio, cat->mdss_ver, cat->mixer_count, cat->mixer); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed ctl object creation: err %d\n", rc); @@ -168,10 +168,10 @@ int dpu_rm_init(struct drm_device *dev, struct dpu_hw_dsc *hw; const struct dpu_dsc_cfg *dsc = &cat->dsc[i]; - if (test_bit(DPU_DSC_HW_REV_1_2, &dsc->features)) + if (cat->mdss_ver->core_major_ver >= 7) hw = dpu_hw_dsc_init_1_2(dev, dsc, mmio); else - hw = dpu_hw_dsc_init(dev, dsc, mmio); + hw = dpu_hw_dsc_init(dev, dsc, mmio, cat->mdss_ver); if (IS_ERR(hw)) { rc = PTR_ERR(hw); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h index aa62966056d4..ccd64404f12d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h @@ -69,7 +69,7 @@ struct msm_display_topology { int dpu_rm_init(struct drm_device *dev, struct dpu_rm *rm, const struct dpu_mdss_cfg *cat, - const struct msm_mdss_data *mdss_data, + const struct qcom_ubwc_cfg_data *mdss_data, void __iomem *mmio); int dpu_rm_reserve(struct dpu_rm *rm, diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c index b8610aa806ea..da53ca88251e 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c @@ -17,7 +17,6 @@ struct mdp4_crtc { struct drm_crtc base; char name[8]; - int id; int ovlp; enum mdp4_dma dma; bool enabled; @@ -120,7 +119,7 @@ static void unref_cursor_worker(struct drm_flip_work *work, void *val) struct mdp4_kms *mdp4_kms = get_kms(&mdp4_crtc->base); struct msm_kms *kms = &mdp4_kms->base.base; - msm_gem_unpin_iova(val, kms->aspace); + msm_gem_unpin_iova(val, kms->vm); drm_gem_object_put(val); } @@ -369,7 +368,7 @@ static void update_cursor(struct drm_crtc *crtc) if (next_bo) { /* take a obj ref + iova ref when we start scanning out: */ drm_gem_object_get(next_bo); - msm_gem_get_and_pin_iova(next_bo, kms->aspace, &iova); + msm_gem_get_and_pin_iova(next_bo, kms->vm, &iova); /* enable cursor: */ mdp4_write(mdp4_kms, REG_MDP4_DMA_CURSOR_SIZE(dma), @@ -427,7 +426,7 @@ static int mdp4_crtc_cursor_set(struct drm_crtc *crtc, } if (cursor_bo) { - ret = msm_gem_get_and_pin_iova(cursor_bo, kms->aspace, &iova); + ret = msm_gem_get_and_pin_iova(cursor_bo, kms->vm, &iova); if (ret) goto fail; } else { @@ -511,7 +510,7 @@ static void mdp4_crtc_vblank_irq(struct mdp_irq *irq, uint32_t irqstatus) if (pending & PENDING_CURSOR) { update_cursor(crtc); - drm_flip_work_commit(&mdp4_crtc->unref_cursor_work, priv->wq); + drm_flip_work_commit(&mdp4_crtc->unref_cursor_work, priv->kms->wq); } } @@ -539,7 +538,7 @@ static void mdp4_crtc_wait_for_flush_done(struct drm_crtc *crtc) mdp4_crtc->flushed_mask), msecs_to_jiffies(50)); if (ret <= 0) - dev_warn(dev->dev, "vblank time out, crtc=%d\n", mdp4_crtc->id); + dev_warn(dev->dev, "vblank time out, crtc=%s\n", mdp4_crtc->base.name); mdp4_crtc->flushed_mask = 0; @@ -624,7 +623,7 @@ static void mdp4_crtc_flip_cleanup(struct drm_device *dev, void *ptr) /* initialize crtc */ struct drm_crtc *mdp4_crtc_init(struct drm_device *dev, - struct drm_plane *plane, int id, int ovlp_id, + struct drm_plane *plane, int ovlp_id, enum mdp4_dma dma_id) { struct drm_crtc *crtc = NULL; @@ -639,8 +638,6 @@ struct drm_crtc *mdp4_crtc_init(struct drm_device *dev, crtc = &mdp4_crtc->base; - mdp4_crtc->id = id; - mdp4_crtc->ovlp = ovlp_id; mdp4_crtc->dma = dma_id; diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c index 7e942c1337b3..0952c7f18abd 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c @@ -122,15 +122,16 @@ static void mdp4_destroy(struct msm_kms *kms) { struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms)); struct device *dev = mdp4_kms->dev->dev; - struct msm_gem_address_space *aspace = kms->aspace; if (mdp4_kms->blank_cursor_iova) - msm_gem_unpin_iova(mdp4_kms->blank_cursor_bo, kms->aspace); + msm_gem_unpin_iova(mdp4_kms->blank_cursor_bo, kms->vm); drm_gem_object_put(mdp4_kms->blank_cursor_bo); - if (aspace) { - aspace->mmu->funcs->detach(aspace->mmu); - msm_gem_address_space_put(aspace); + if (kms->vm) { + struct msm_mmu *mmu = to_msm_vm(kms->vm)->mmu; + + mmu->funcs->detach(mmu); + drm_gpuvm_put(kms->vm); } if (mdp4_kms->rpm_enabled) @@ -249,9 +250,9 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, /* DTV can be hooked to DMA_E: */ encoder->possible_crtcs = 1 << 1; - if (priv->hdmi) { + if (priv->kms->hdmi) { /* Construct bridge/connector for HDMI: */ - ret = msm_hdmi_modeset_init(priv->hdmi, dev, encoder); + ret = msm_hdmi_modeset_init(priv->kms->hdmi, dev, encoder); if (ret) { DRM_DEV_ERROR(dev->dev, "failed to initialize HDMI: %d\n", ret); return ret; @@ -263,7 +264,7 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, /* only DSI1 supported for now */ dsi_id = 0; - if (!priv->dsi[dsi_id]) + if (!priv->kms->dsi[dsi_id]) break; encoder = mdp4_dsi_encoder_init(dev); @@ -277,7 +278,7 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, /* TODO: Add DMA_S later? */ encoder->possible_crtcs = 1 << DMA_P; - ret = msm_dsi_modeset_init(priv->dsi[dsi_id], dev, encoder); + ret = msm_dsi_modeset_init(priv->kms->dsi[dsi_id], dev, encoder); if (ret) { DRM_DEV_ERROR(dev->dev, "failed to initialize DSI: %d\n", ret); @@ -296,7 +297,6 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, static int modeset_init(struct mdp4_kms *mdp4_kms) { struct drm_device *dev = mdp4_kms->dev; - struct msm_drm_private *priv = dev->dev_private; struct drm_plane *plane; struct drm_crtc *crtc; int i, ret; @@ -338,7 +338,7 @@ static int modeset_init(struct mdp4_kms *mdp4_kms) goto fail; } - crtc = mdp4_crtc_init(dev, plane, priv->num_crtcs, i, + crtc = mdp4_crtc_init(dev, plane, i, mdp4_crtcs[i]); if (IS_ERR(crtc)) { DRM_DEV_ERROR(dev->dev, "failed to construct crtc for %s\n", @@ -346,8 +346,6 @@ static int modeset_init(struct mdp4_kms *mdp4_kms) ret = PTR_ERR(crtc); goto fail; } - - priv->num_crtcs++; } /* @@ -398,7 +396,7 @@ static int mdp4_kms_init(struct drm_device *dev) struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(priv->kms)); struct msm_kms *kms = NULL; struct msm_mmu *mmu; - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; int ret; u32 major, minor; unsigned long max_clk; @@ -467,19 +465,20 @@ static int mdp4_kms_init(struct drm_device *dev) } else if (!mmu) { DRM_DEV_INFO(dev->dev, "no iommu, fallback to phys " "contig buffers for scanout\n"); - aspace = NULL; + vm = NULL; } else { - aspace = msm_gem_address_space_create(mmu, - "mdp4", 0x1000, 0x100000000 - 0x1000); + vm = msm_gem_vm_create(dev, mmu, "mdp4", + 0x1000, 0x100000000 - 0x1000, + true); - if (IS_ERR(aspace)) { + if (IS_ERR(vm)) { if (!IS_ERR(mmu)) mmu->funcs->destroy(mmu); - ret = PTR_ERR(aspace); + ret = PTR_ERR(vm); goto fail; } - kms->aspace = aspace; + kms->vm = vm; } ret = modeset_init(mdp4_kms); @@ -496,7 +495,7 @@ static int mdp4_kms_init(struct drm_device *dev) goto fail; } - ret = msm_gem_get_and_pin_iova(mdp4_kms->blank_cursor_bo, kms->aspace, + ret = msm_gem_get_and_pin_iova(mdp4_kms->blank_cursor_bo, kms->vm, &mdp4_kms->blank_cursor_iova); if (ret) { DRM_DEV_ERROR(dev->dev, "could not pin blank-cursor bo: %d\n", ret); diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h index f9d988076337..fb348583dc84 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h @@ -185,7 +185,7 @@ void mdp4_crtc_set_config(struct drm_crtc *crtc, uint32_t config); void mdp4_crtc_set_intf(struct drm_crtc *crtc, enum mdp4_intf intf, int mixer); void mdp4_crtc_wait_for_commit_done(struct drm_crtc *crtc); struct drm_crtc *mdp4_crtc_init(struct drm_device *dev, - struct drm_plane *plane, int id, int ovlp_id, + struct drm_plane *plane, int ovlp_id, enum mdp4_dma dma_id); long mdp4_dtv_round_pixclk(struct drm_encoder *encoder, unsigned long rate); diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c index 3fefb2088008..098c3b5ff2b2 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c @@ -79,30 +79,25 @@ static const struct drm_plane_funcs mdp4_plane_funcs = { static int mdp4_plane_prepare_fb(struct drm_plane *plane, struct drm_plane_state *new_state) { - struct msm_drm_private *priv = plane->dev->dev_private; - struct msm_kms *kms = priv->kms; - if (!new_state->fb) return 0; drm_gem_plane_helper_prepare_fb(plane, new_state); - return msm_framebuffer_prepare(new_state->fb, kms->aspace, false); + return msm_framebuffer_prepare(new_state->fb, false); } static void mdp4_plane_cleanup_fb(struct drm_plane *plane, struct drm_plane_state *old_state) { struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane); - struct mdp4_kms *mdp4_kms = get_kms(plane); - struct msm_kms *kms = &mdp4_kms->base.base; struct drm_framebuffer *fb = old_state->fb; if (!fb) return; DBG("%s: cleanup: FB[%u]", mdp4_plane->name, fb->base.id); - msm_framebuffer_cleanup(fb, kms->aspace, false); + msm_framebuffer_cleanup(fb, false); } @@ -141,7 +136,6 @@ static void mdp4_plane_set_scanout(struct drm_plane *plane, { struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane); struct mdp4_kms *mdp4_kms = get_kms(plane); - struct msm_kms *kms = &mdp4_kms->base.base; enum mdp4_pipe pipe = mdp4_plane->pipe; mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRC_STRIDE_A(pipe), @@ -153,13 +147,13 @@ static void mdp4_plane_set_scanout(struct drm_plane *plane, MDP4_PIPE_SRC_STRIDE_B_P3(fb->pitches[3])); mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP0_BASE(pipe), - msm_framebuffer_iova(fb, kms->aspace, 0)); + msm_framebuffer_iova(fb, 0)); mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP1_BASE(pipe), - msm_framebuffer_iova(fb, kms->aspace, 1)); + msm_framebuffer_iova(fb, 1)); mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP2_BASE(pipe), - msm_framebuffer_iova(fb, kms->aspace, 2)); + msm_framebuffer_iova(fb, 2)); mdp4_write(mdp4_kms, REG_MDP4_PIPE_SRCP3_BASE(pipe), - msm_framebuffer_iova(fb, kms->aspace, 3)); + msm_framebuffer_iova(fb, 3)); } static void mdp4_write_csc_config(struct mdp4_kms *mdp4_kms, diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c index 0f653e62b4a0..4c4900a7beda 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c @@ -169,7 +169,7 @@ static void unref_cursor_worker(struct drm_flip_work *work, void *val) struct mdp5_kms *mdp5_kms = get_kms(&mdp5_crtc->base); struct msm_kms *kms = &mdp5_kms->base.base; - msm_gem_unpin_iova(val, kms->aspace); + msm_gem_unpin_iova(val, kms->vm); drm_gem_object_put(val); } @@ -993,7 +993,7 @@ static int mdp5_crtc_cursor_set(struct drm_crtc *crtc, if (!cursor_bo) return -ENOENT; - ret = msm_gem_get_and_pin_iova(cursor_bo, kms->aspace, + ret = msm_gem_get_and_pin_iova(cursor_bo, kms->vm, &mdp5_crtc->cursor.iova); if (ret) { drm_gem_object_put(cursor_bo); @@ -1196,7 +1196,7 @@ static void mdp5_crtc_vblank_irq(struct mdp_irq *irq, uint32_t irqstatus) } if (pending & PENDING_CURSOR) - drm_flip_work_commit(&mdp5_crtc->unref_cursor_work, priv->wq); + drm_flip_work_commit(&mdp5_crtc->unref_cursor_work, priv->kms->wq); } static void mdp5_crtc_err_irq(struct mdp_irq *irq, uint32_t irqstatus) diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c index 3fcca7a3d82e..5b6ca8dd929e 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c @@ -198,11 +198,12 @@ static void mdp5_destroy(struct mdp5_kms *mdp5_kms); static void mdp5_kms_destroy(struct msm_kms *kms) { struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); - struct msm_gem_address_space *aspace = kms->aspace; - if (aspace) { - aspace->mmu->funcs->detach(aspace->mmu); - msm_gem_address_space_put(aspace); + if (kms->vm) { + struct msm_mmu *mmu = to_msm_vm(kms->vm)->mmu; + + mmu->funcs->detach(mmu); + drm_gpuvm_put(kms->vm); } mdp_kms_destroy(&mdp5_kms->base); @@ -311,7 +312,7 @@ static int modeset_init_intf(struct mdp5_kms *mdp5_kms, DRM_DEV_INFO(dev->dev, "Skipping eDP interface %d\n", intf->num); break; case INTF_HDMI: - if (!priv->hdmi) + if (!priv->kms->hdmi) break; ctl = mdp5_ctlm_request(ctlm, intf->num); @@ -326,7 +327,7 @@ static int modeset_init_intf(struct mdp5_kms *mdp5_kms, break; } - ret = msm_hdmi_modeset_init(priv->hdmi, dev, encoder); + ret = msm_hdmi_modeset_init(priv->kms->hdmi, dev, encoder); break; case INTF_DSI: { @@ -334,14 +335,14 @@ static int modeset_init_intf(struct mdp5_kms *mdp5_kms, mdp5_cfg_get_hw_config(mdp5_kms->cfg); int dsi_id = get_dsi_id_from_intf(hw_cfg, intf->num); - if ((dsi_id >= ARRAY_SIZE(priv->dsi)) || (dsi_id < 0)) { + if ((dsi_id >= ARRAY_SIZE(priv->kms->dsi)) || (dsi_id < 0)) { DRM_DEV_ERROR(dev->dev, "failed to find dsi from intf %d\n", intf->num); ret = -EINVAL; break; } - if (!priv->dsi[dsi_id]) + if (!priv->kms->dsi[dsi_id]) break; ctl = mdp5_ctlm_request(ctlm, intf->num); @@ -356,9 +357,10 @@ static int modeset_init_intf(struct mdp5_kms *mdp5_kms, break; } - ret = msm_dsi_modeset_init(priv->dsi[dsi_id], dev, encoder); + ret = msm_dsi_modeset_init(priv->kms->dsi[dsi_id], dev, encoder); if (!ret) - mdp5_encoder_set_intf_mode(encoder, msm_dsi_is_cmd_mode(priv->dsi[dsi_id])); + mdp5_encoder_set_intf_mode(encoder, + msm_dsi_is_cmd_mode(priv->kms->dsi[dsi_id])); break; } @@ -374,7 +376,6 @@ static int modeset_init_intf(struct mdp5_kms *mdp5_kms, static int modeset_init(struct mdp5_kms *mdp5_kms) { struct drm_device *dev = mdp5_kms->dev; - struct msm_drm_private *priv = dev->dev_private; unsigned int num_crtcs; int i, ret, pi = 0, ci = 0; struct drm_plane *primary[MAX_BASES] = { NULL }; @@ -442,7 +443,6 @@ static int modeset_init(struct mdp5_kms *mdp5_kms) DRM_DEV_ERROR(dev->dev, "failed to construct crtc %d (%d)\n", i, ret); goto fail; } - priv->num_crtcs++; } /* @@ -450,7 +450,7 @@ static int modeset_init(struct mdp5_kms *mdp5_kms) * crtcs for the encoders */ drm_for_each_encoder(encoder, dev) - encoder->possible_crtcs = (1 << priv->num_crtcs) - 1; + encoder->possible_crtcs = (1 << dev->mode_config.num_crtc) - 1; return 0; @@ -500,7 +500,7 @@ static int mdp5_kms_init(struct drm_device *dev) struct mdp5_kms *mdp5_kms; struct mdp5_cfg *config; struct msm_kms *kms = priv->kms; - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; int i, ret; ret = mdp5_init(to_platform_device(dev->dev), dev); @@ -534,13 +534,13 @@ static int mdp5_kms_init(struct drm_device *dev) } mdelay(16); - aspace = msm_kms_init_aspace(mdp5_kms->dev); - if (IS_ERR(aspace)) { - ret = PTR_ERR(aspace); + vm = msm_kms_init_vm(mdp5_kms->dev); + if (IS_ERR(vm)) { + ret = PTR_ERR(vm); goto fail; } - kms->aspace = aspace; + kms->vm = vm; pm_runtime_put_sync(&pdev->dev); diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c index bb1601921938..7c790406d533 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c @@ -135,8 +135,6 @@ static const struct drm_plane_funcs mdp5_plane_funcs = { static int mdp5_plane_prepare_fb(struct drm_plane *plane, struct drm_plane_state *new_state) { - struct msm_drm_private *priv = plane->dev->dev_private; - struct msm_kms *kms = priv->kms; bool needs_dirtyfb = to_mdp5_plane_state(new_state)->needs_dirtyfb; if (!new_state->fb) @@ -144,14 +142,12 @@ static int mdp5_plane_prepare_fb(struct drm_plane *plane, drm_gem_plane_helper_prepare_fb(plane, new_state); - return msm_framebuffer_prepare(new_state->fb, kms->aspace, needs_dirtyfb); + return msm_framebuffer_prepare(new_state->fb, needs_dirtyfb); } static void mdp5_plane_cleanup_fb(struct drm_plane *plane, struct drm_plane_state *old_state) { - struct mdp5_kms *mdp5_kms = get_kms(plane); - struct msm_kms *kms = &mdp5_kms->base.base; struct drm_framebuffer *fb = old_state->fb; bool needed_dirtyfb = to_mdp5_plane_state(old_state)->needs_dirtyfb; @@ -159,7 +155,7 @@ static void mdp5_plane_cleanup_fb(struct drm_plane *plane, return; DBG("%s: cleanup: FB[%u]", plane->name, fb->base.id); - msm_framebuffer_cleanup(fb, kms->aspace, needed_dirtyfb); + msm_framebuffer_cleanup(fb, needed_dirtyfb); } static int mdp5_plane_atomic_check_with_state(struct drm_crtc_state *crtc_state, @@ -467,8 +463,6 @@ static void set_scanout_locked(struct mdp5_kms *mdp5_kms, enum mdp5_pipe pipe, struct drm_framebuffer *fb) { - struct msm_kms *kms = &mdp5_kms->base.base; - mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_STRIDE_A(pipe), MDP5_PIPE_SRC_STRIDE_A_P0(fb->pitches[0]) | MDP5_PIPE_SRC_STRIDE_A_P1(fb->pitches[1])); @@ -478,13 +472,13 @@ static void set_scanout_locked(struct mdp5_kms *mdp5_kms, MDP5_PIPE_SRC_STRIDE_B_P3(fb->pitches[3])); mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC0_ADDR(pipe), - msm_framebuffer_iova(fb, kms->aspace, 0)); + msm_framebuffer_iova(fb, 0)); mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC1_ADDR(pipe), - msm_framebuffer_iova(fb, kms->aspace, 1)); + msm_framebuffer_iova(fb, 1)); mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC2_ADDR(pipe), - msm_framebuffer_iova(fb, kms->aspace, 2)); + msm_framebuffer_iova(fb, 2)); mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC3_ADDR(pipe), - msm_framebuffer_iova(fb, kms->aspace, 3)); + msm_framebuffer_iova(fb, 3)); } /* Note: mdp5_plane->pipe_lock must be locked */ diff --git a/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c b/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c index 07a2c1e87219..071bcdea80f7 100644 --- a/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c +++ b/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c @@ -127,18 +127,18 @@ void msm_disp_snapshot_capture_state(struct msm_disp_state *disp_state) priv = drm_dev->dev_private; kms = priv->kms; - for (i = 0; i < ARRAY_SIZE(priv->dp); i++) { - if (!priv->dp[i]) + for (i = 0; i < ARRAY_SIZE(kms->dp); i++) { + if (!kms->dp[i]) continue; - msm_dp_snapshot(disp_state, priv->dp[i]); + msm_dp_snapshot(disp_state, kms->dp[i]); } - for (i = 0; i < ARRAY_SIZE(priv->dsi); i++) { - if (!priv->dsi[i]) + for (i = 0; i < ARRAY_SIZE(kms->dsi); i++) { + if (!kms->dsi[i]) continue; - msm_dsi_snapshot(disp_state, priv->dsi[i]); + msm_dsi_snapshot(disp_state, kms->dsi[i]); } if (kms->funcs->snapshot) diff --git a/drivers/gpu/drm/msm/dp/dp_audio.c b/drivers/gpu/drm/msm/dp/dp_audio.c index f8bfb908f9b4..959cf53be4b8 100644 --- a/drivers/gpu/drm/msm/dp/dp_audio.c +++ b/drivers/gpu/drm/msm/dp/dp_audio.c @@ -11,7 +11,6 @@ #include <drm/display/drm_dp_helper.h> #include <drm/drm_edid.h> -#include "dp_catalog.h" #include "dp_audio.h" #include "dp_drm.h" #include "dp_panel.h" @@ -22,13 +21,28 @@ struct msm_dp_audio_private { struct platform_device *pdev; struct drm_device *drm_dev; - struct msm_dp_catalog *catalog; + void __iomem *link_base; u32 channels; struct msm_dp_audio msm_dp_audio; }; +static inline u32 msm_dp_read_link(struct msm_dp_audio_private *audio, u32 offset) +{ + return readl_relaxed(audio->link_base + offset); +} + +static inline void msm_dp_write_link(struct msm_dp_audio_private *audio, + u32 offset, u32 data) +{ + /* + * To make sure link reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, audio->link_base + offset); +} + static void msm_dp_audio_stream_sdp(struct msm_dp_audio_private *audio) { struct dp_sdp_header sdp_hdr = { @@ -37,8 +51,12 @@ static void msm_dp_audio_stream_sdp(struct msm_dp_audio_private *audio) .HB2 = 0x00, .HB3 = audio->channels - 1, }; + u32 header[2]; - msm_dp_catalog_write_audio_stream(audio->catalog, &sdp_hdr); + msm_dp_utils_pack_sdp_header(&sdp_hdr, header); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_STREAM_0, header[0]); + msm_dp_write_link(audio, MMSS_DP_AUDIO_STREAM_1, header[1]); } static void msm_dp_audio_timestamp_sdp(struct msm_dp_audio_private *audio) @@ -49,8 +67,12 @@ static void msm_dp_audio_timestamp_sdp(struct msm_dp_audio_private *audio) .HB2 = 0x17, .HB3 = 0x0 | (0x11 << 2), }; + u32 header[2]; + + msm_dp_utils_pack_sdp_header(&sdp_hdr, header); - msm_dp_catalog_write_audio_timestamp(audio->catalog, &sdp_hdr); + msm_dp_write_link(audio, MMSS_DP_AUDIO_TIMESTAMP_0, header[0]); + msm_dp_write_link(audio, MMSS_DP_AUDIO_TIMESTAMP_1, header[1]); } static void msm_dp_audio_infoframe_sdp(struct msm_dp_audio_private *audio) @@ -61,8 +83,12 @@ static void msm_dp_audio_infoframe_sdp(struct msm_dp_audio_private *audio) .HB2 = 0x1b, .HB3 = 0x0 | (0x11 << 2), }; + u32 header[2]; - msm_dp_catalog_write_audio_infoframe(audio->catalog, &sdp_hdr); + msm_dp_utils_pack_sdp_header(&sdp_hdr, header); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_INFOFRAME_0, header[0]); + msm_dp_write_link(audio, MMSS_DP_AUDIO_INFOFRAME_1, header[1]); } static void msm_dp_audio_copy_management_sdp(struct msm_dp_audio_private *audio) @@ -73,8 +99,12 @@ static void msm_dp_audio_copy_management_sdp(struct msm_dp_audio_private *audio) .HB2 = 0x0f, .HB3 = 0x00, }; + u32 header[2]; + + msm_dp_utils_pack_sdp_header(&sdp_hdr, header); - msm_dp_catalog_write_audio_copy_mgmt(audio->catalog, &sdp_hdr); + msm_dp_write_link(audio, MMSS_DP_AUDIO_COPYMANAGEMENT_0, header[0]); + msm_dp_write_link(audio, MMSS_DP_AUDIO_COPYMANAGEMENT_1, header[1]); } static void msm_dp_audio_isrc_sdp(struct msm_dp_audio_private *audio) @@ -85,13 +115,53 @@ static void msm_dp_audio_isrc_sdp(struct msm_dp_audio_private *audio) .HB2 = 0x0f, .HB3 = 0x00, }; + u32 header[2]; + u32 reg; + + /* XXX: is it necessary to preserve this field? */ + reg = msm_dp_read_link(audio, MMSS_DP_AUDIO_ISRC_1); + sdp_hdr.HB3 = FIELD_GET(HEADER_3_MASK, reg); + + msm_dp_utils_pack_sdp_header(&sdp_hdr, header); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_ISRC_0, header[0]); + msm_dp_write_link(audio, MMSS_DP_AUDIO_ISRC_1, header[1]); +} + +static void msm_dp_audio_config_sdp(struct msm_dp_audio_private *audio) +{ + u32 sdp_cfg, sdp_cfg2; + + sdp_cfg = msm_dp_read_link(audio, MMSS_DP_SDP_CFG); + /* AUDIO_TIMESTAMP_SDP_EN */ + sdp_cfg |= BIT(1); + /* AUDIO_STREAM_SDP_EN */ + sdp_cfg |= BIT(2); + /* AUDIO_COPY_MANAGEMENT_SDP_EN */ + sdp_cfg |= BIT(5); + /* AUDIO_ISRC_SDP_EN */ + sdp_cfg |= BIT(6); + /* AUDIO_INFOFRAME_SDP_EN */ + sdp_cfg |= BIT(20); + + drm_dbg_dp(audio->drm_dev, "sdp_cfg = 0x%x\n", sdp_cfg); + + msm_dp_write_link(audio, MMSS_DP_SDP_CFG, sdp_cfg); + + sdp_cfg2 = msm_dp_read_link(audio, MMSS_DP_SDP_CFG2); + /* IFRM_REGSRC -> Do not use reg values */ + sdp_cfg2 &= ~BIT(0); + /* AUDIO_STREAM_HB3_REGSRC-> Do not use reg values */ + sdp_cfg2 &= ~BIT(1); + + drm_dbg_dp(audio->drm_dev, "sdp_cfg2 = 0x%x\n", sdp_cfg2); - msm_dp_catalog_write_audio_isrc(audio->catalog, &sdp_hdr); + msm_dp_write_link(audio, MMSS_DP_SDP_CFG2, sdp_cfg2); } static void msm_dp_audio_setup_sdp(struct msm_dp_audio_private *audio) { - msm_dp_catalog_audio_config_sdp(audio->catalog); + msm_dp_audio_config_sdp(audio); msm_dp_audio_stream_sdp(audio); msm_dp_audio_timestamp_sdp(audio); @@ -102,8 +172,7 @@ static void msm_dp_audio_setup_sdp(struct msm_dp_audio_private *audio) static void msm_dp_audio_setup_acr(struct msm_dp_audio_private *audio) { - u32 select = 0; - struct msm_dp_catalog *catalog = audio->catalog; + u32 select, acr_ctrl; switch (audio->msm_dp_audio.bw_code) { case DP_LINK_BW_1_62: @@ -124,13 +193,17 @@ static void msm_dp_audio_setup_acr(struct msm_dp_audio_private *audio) break; } - msm_dp_catalog_audio_config_acr(catalog, select); + acr_ctrl = select << 4 | BIT(31) | BIT(8) | BIT(14); + + drm_dbg_dp(audio->drm_dev, "select: %#x, acr_ctrl: %#x\n", + select, acr_ctrl); + + msm_dp_write_link(audio, MMSS_DP_AUDIO_ACR_CTRL, acr_ctrl); } static void msm_dp_audio_safe_to_exit_level(struct msm_dp_audio_private *audio) { - struct msm_dp_catalog *catalog = audio->catalog; - u32 safe_to_exit_level = 0; + u32 safe_to_exit_level, mainlink_levels; switch (audio->msm_dp_audio.lane_count) { case 1: @@ -150,14 +223,33 @@ static void msm_dp_audio_safe_to_exit_level(struct msm_dp_audio_private *audio) break; } - msm_dp_catalog_audio_sfe_level(catalog, safe_to_exit_level); + mainlink_levels = msm_dp_read_link(audio, REG_DP_MAINLINK_LEVELS); + mainlink_levels &= 0xFE0; + mainlink_levels |= safe_to_exit_level; + + drm_dbg_dp(audio->drm_dev, + "mainlink_level = 0x%x, safe_to_exit_level = 0x%x\n", + mainlink_levels, safe_to_exit_level); + + msm_dp_write_link(audio, REG_DP_MAINLINK_LEVELS, mainlink_levels); } static void msm_dp_audio_enable(struct msm_dp_audio_private *audio, bool enable) { - struct msm_dp_catalog *catalog = audio->catalog; + u32 audio_ctrl; + + audio_ctrl = msm_dp_read_link(audio, MMSS_DP_AUDIO_CFG); + + if (enable) + audio_ctrl |= BIT(0); + else + audio_ctrl &= ~BIT(0); + + drm_dbg_dp(audio->drm_dev, "dp_audio_cfg = 0x%x\n", audio_ctrl); - msm_dp_catalog_audio_enable(catalog, enable); + msm_dp_write_link(audio, MMSS_DP_AUDIO_CFG, audio_ctrl); + /* make sure audio engine is disabled */ + wmb(); } static struct msm_dp_audio_private *msm_dp_audio_get_data(struct msm_dp *msm_dp_display) @@ -246,13 +338,13 @@ void msm_dp_audio_shutdown(struct drm_connector *connector, } struct msm_dp_audio *msm_dp_audio_get(struct platform_device *pdev, - struct msm_dp_catalog *catalog) + void __iomem *link_base) { int rc = 0; struct msm_dp_audio_private *audio; struct msm_dp_audio *msm_dp_audio; - if (!pdev || !catalog) { + if (!pdev) { DRM_ERROR("invalid input\n"); rc = -EINVAL; goto error; @@ -265,7 +357,7 @@ struct msm_dp_audio *msm_dp_audio_get(struct platform_device *pdev, } audio->pdev = pdev; - audio->catalog = catalog; + audio->link_base = link_base; msm_dp_audio = &audio->msm_dp_audio; diff --git a/drivers/gpu/drm/msm/dp/dp_audio.h b/drivers/gpu/drm/msm/dp/dp_audio.h index 58fc14693e48..842278516c99 100644 --- a/drivers/gpu/drm/msm/dp/dp_audio.h +++ b/drivers/gpu/drm/msm/dp/dp_audio.h @@ -8,9 +8,10 @@ #include <linux/platform_device.h> -#include "dp_catalog.h" #include <sound/hdmi-codec.h> +struct drm_bridge; + /** * struct msm_dp_audio * @lane_count: number of lanes configured in current session @@ -27,13 +28,13 @@ struct msm_dp_audio { * Creates and instance of dp audio. * * @pdev: caller's platform device instance. - * @catalog: an instance of msm_dp_catalog module. + * @link_base: pointer to the msm_dp_link resource. * * Returns the error code in case of failure, otherwize * an instance of newly created msm_dp_module. */ struct msm_dp_audio *msm_dp_audio_get(struct platform_device *pdev, - struct msm_dp_catalog *catalog); + void __iomem *link_base); /** * msm_dp_audio_put() diff --git a/drivers/gpu/drm/msm/dp/dp_aux.c b/drivers/gpu/drm/msm/dp/dp_aux.c index bc8d46abfc61..3825a2fb48e2 100644 --- a/drivers/gpu/drm/msm/dp/dp_aux.c +++ b/drivers/gpu/drm/msm/dp/dp_aux.c @@ -4,6 +4,7 @@ */ #include <linux/delay.h> +#include <linux/iopoll.h> #include <linux/phy/phy.h> #include <drm/drm_print.h> @@ -22,7 +23,7 @@ enum msm_dp_aux_err { struct msm_dp_aux_private { struct device *dev; - struct msm_dp_catalog *catalog; + void __iomem *aux_base; struct phy *phy; @@ -45,6 +46,80 @@ struct msm_dp_aux_private { struct drm_dp_aux msm_dp_aux; }; +static inline u32 msm_dp_read_aux(struct msm_dp_aux_private *aux, u32 offset) +{ + return readl_relaxed(aux->aux_base + offset); +} + +static inline void msm_dp_write_aux(struct msm_dp_aux_private *aux, + u32 offset, u32 data) +{ + /* + * To make sure aux reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, aux->aux_base + offset); +} + +static void msm_dp_aux_clear_hw_interrupts(struct msm_dp_aux_private *aux) +{ + msm_dp_read_aux(aux, REG_DP_PHY_AUX_INTERRUPT_STATUS); + msm_dp_write_aux(aux, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0x1f); + msm_dp_write_aux(aux, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0x9f); + msm_dp_write_aux(aux, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0); +} + +/* + * NOTE: resetting AUX controller will also clear any pending HPD related interrupts + */ +static void msm_dp_aux_reset(struct msm_dp_aux_private *aux) +{ + u32 aux_ctrl; + + aux_ctrl = msm_dp_read_aux(aux, REG_DP_AUX_CTRL); + + aux_ctrl |= DP_AUX_CTRL_RESET; + msm_dp_write_aux(aux, REG_DP_AUX_CTRL, aux_ctrl); + usleep_range(1000, 1100); /* h/w recommended delay */ + + aux_ctrl &= ~DP_AUX_CTRL_RESET; + msm_dp_write_aux(aux, REG_DP_AUX_CTRL, aux_ctrl); +} + +static void msm_dp_aux_enable(struct msm_dp_aux_private *aux) +{ + u32 aux_ctrl; + + aux_ctrl = msm_dp_read_aux(aux, REG_DP_AUX_CTRL); + + msm_dp_write_aux(aux, REG_DP_TIMEOUT_COUNT, 0xffff); + msm_dp_write_aux(aux, REG_DP_AUX_LIMITS, 0xffff); + + aux_ctrl |= DP_AUX_CTRL_ENABLE; + msm_dp_write_aux(aux, REG_DP_AUX_CTRL, aux_ctrl); +} + +static void msm_dp_aux_disable(struct msm_dp_aux_private *aux) +{ + u32 aux_ctrl; + + aux_ctrl = msm_dp_read_aux(aux, REG_DP_AUX_CTRL); + aux_ctrl &= ~DP_AUX_CTRL_ENABLE; + msm_dp_write_aux(aux, REG_DP_AUX_CTRL, aux_ctrl); +} + +static int msm_dp_aux_wait_for_hpd_connect_state(struct msm_dp_aux_private *aux, + unsigned long wait_us) +{ + u32 state; + + /* poll for hpd connected status every 2ms and timeout after wait_us */ + return readl_poll_timeout(aux->aux_base + + REG_DP_DP_HPD_INT_STATUS, + state, state & DP_DP_HPD_STATE_STATUS_CONNECTED, + min(wait_us, 2000), wait_us); +} + #define MAX_AUX_RETRIES 5 static ssize_t msm_dp_aux_write(struct msm_dp_aux_private *aux, @@ -88,11 +163,11 @@ static ssize_t msm_dp_aux_write(struct msm_dp_aux_private *aux, /* index = 0, write */ if (i == 0) reg |= DP_AUX_DATA_INDEX_WRITE; - msm_dp_catalog_aux_write_data(aux->catalog, reg); + msm_dp_write_aux(aux, REG_DP_AUX_DATA, reg); } - msm_dp_catalog_aux_clear_trans(aux->catalog, false); - msm_dp_catalog_aux_clear_hw_interrupts(aux->catalog); + msm_dp_write_aux(aux, REG_DP_AUX_TRANS_CTRL, 0); + msm_dp_aux_clear_hw_interrupts(aux); reg = 0; /* Transaction number == 1 */ if (!aux->native) { /* i2c */ @@ -106,7 +181,7 @@ static ssize_t msm_dp_aux_write(struct msm_dp_aux_private *aux, } reg |= DP_AUX_TRANS_CTRL_GO; - msm_dp_catalog_aux_write_trans(aux->catalog, reg); + msm_dp_write_aux(aux, REG_DP_AUX_TRANS_CTRL, reg); return len; } @@ -139,20 +214,22 @@ static ssize_t msm_dp_aux_cmd_fifo_rx(struct msm_dp_aux_private *aux, u32 i, actual_i; u32 len = msg->size; - msm_dp_catalog_aux_clear_trans(aux->catalog, true); + data = msm_dp_read_aux(aux, REG_DP_AUX_TRANS_CTRL); + data &= ~DP_AUX_TRANS_CTRL_GO; + msm_dp_write_aux(aux, REG_DP_AUX_TRANS_CTRL, data); data = DP_AUX_DATA_INDEX_WRITE; /* INDEX_WRITE */ data |= DP_AUX_DATA_READ; /* read */ - msm_dp_catalog_aux_write_data(aux->catalog, data); + msm_dp_write_aux(aux, REG_DP_AUX_DATA, data); dp = msg->buffer; /* discard first byte */ - data = msm_dp_catalog_aux_read_data(aux->catalog); + data = msm_dp_read_aux(aux, REG_DP_AUX_DATA); for (i = 0; i < len; i++) { - data = msm_dp_catalog_aux_read_data(aux->catalog); + data = msm_dp_read_aux(aux, REG_DP_AUX_DATA); *dp++ = (u8)((data >> DP_AUX_DATA_OFFSET) & 0xff); actual_i = (data >> DP_AUX_DATA_INDEX_OFFSET) & 0xFF; @@ -335,8 +412,8 @@ static ssize_t msm_dp_aux_transfer(struct drm_dp_aux *msm_dp_aux, phy_calibrate(aux->phy); } /* reset aux if link is in connected state */ - if (msm_dp_catalog_link_is_connected(aux->catalog)) - msm_dp_catalog_aux_reset(aux->catalog); + if (msm_dp_aux_is_link_connected(msm_dp_aux)) + msm_dp_aux_reset(aux); } else { aux->retry_cnt = 0; switch (aux->aux_error_num) { @@ -369,9 +446,8 @@ exit: return ret; } -irqreturn_t msm_dp_aux_isr(struct drm_dp_aux *msm_dp_aux) +irqreturn_t msm_dp_aux_isr(struct drm_dp_aux *msm_dp_aux, u32 isr) { - u32 isr; struct msm_dp_aux_private *aux; if (!msm_dp_aux) { @@ -381,12 +457,6 @@ irqreturn_t msm_dp_aux_isr(struct drm_dp_aux *msm_dp_aux) aux = container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); - isr = msm_dp_catalog_aux_get_irq(aux->catalog); - - /* no interrupts pending, return immediately */ - if (!isr) - return IRQ_NONE; - if (!aux->cmd_busy) { DRM_ERROR("Unexpected DP AUX IRQ %#010x when not busy\n", isr); return IRQ_NONE; @@ -403,7 +473,7 @@ irqreturn_t msm_dp_aux_isr(struct drm_dp_aux *msm_dp_aux) if (isr & DP_INTR_AUX_ERROR) { aux->aux_error_num = DP_AUX_ERR_PHY; - msm_dp_catalog_aux_clear_hw_interrupts(aux->catalog); + msm_dp_aux_clear_hw_interrupts(aux); } else if (isr & DP_INTR_NACK_DEFER) { aux->aux_error_num = DP_AUX_ERR_NACK_DEFER; } else if (isr & DP_INTR_WRONG_ADDR) { @@ -444,7 +514,7 @@ void msm_dp_aux_reconfig(struct drm_dp_aux *msm_dp_aux) aux = container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); phy_calibrate(aux->phy); - msm_dp_catalog_aux_reset(aux->catalog); + msm_dp_aux_reset(aux); } void msm_dp_aux_init(struct drm_dp_aux *msm_dp_aux) @@ -460,7 +530,7 @@ void msm_dp_aux_init(struct drm_dp_aux *msm_dp_aux) mutex_lock(&aux->mutex); - msm_dp_catalog_aux_enable(aux->catalog, true); + msm_dp_aux_enable(aux); aux->retry_cnt = 0; aux->initted = true; @@ -476,7 +546,7 @@ void msm_dp_aux_deinit(struct drm_dp_aux *msm_dp_aux) mutex_lock(&aux->mutex); aux->initted = false; - msm_dp_catalog_aux_enable(aux->catalog, false); + msm_dp_aux_disable(aux); mutex_unlock(&aux->mutex); } @@ -517,23 +587,105 @@ static int msm_dp_wait_hpd_asserted(struct drm_dp_aux *msm_dp_aux, if (ret) return ret; - ret = msm_dp_catalog_aux_wait_for_hpd_connect_state(aux->catalog, wait_us); + ret = msm_dp_aux_wait_for_hpd_connect_state(aux, wait_us); pm_runtime_put_sync(aux->dev); return ret; } -struct drm_dp_aux *msm_dp_aux_get(struct device *dev, struct msm_dp_catalog *catalog, +void msm_dp_aux_hpd_enable(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + u32 reg; + + /* Configure REFTIMER and enable it */ + reg = msm_dp_read_aux(aux, REG_DP_DP_HPD_REFTIMER); + reg |= DP_DP_HPD_REFTIMER_ENABLE; + msm_dp_write_aux(aux, REG_DP_DP_HPD_REFTIMER, reg); + + /* Enable HPD */ + msm_dp_write_aux(aux, REG_DP_DP_HPD_CTRL, DP_DP_HPD_CTRL_HPD_EN); +} + +void msm_dp_aux_hpd_disable(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + u32 reg; + + reg = msm_dp_read_aux(aux, REG_DP_DP_HPD_REFTIMER); + reg &= ~DP_DP_HPD_REFTIMER_ENABLE; + msm_dp_write_aux(aux, REG_DP_DP_HPD_REFTIMER, reg); + + msm_dp_write_aux(aux, REG_DP_DP_HPD_CTRL, 0); +} + +void msm_dp_aux_hpd_intr_enable(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + u32 reg; + + reg = msm_dp_read_aux(aux, REG_DP_DP_HPD_INT_MASK); + reg |= DP_DP_HPD_INT_MASK; + msm_dp_write_aux(aux, REG_DP_DP_HPD_INT_MASK, + reg & DP_DP_HPD_INT_MASK); +} + +void msm_dp_aux_hpd_intr_disable(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + u32 reg; + + reg = msm_dp_read_aux(aux, REG_DP_DP_HPD_INT_MASK); + reg &= ~DP_DP_HPD_INT_MASK; + msm_dp_write_aux(aux, REG_DP_DP_HPD_INT_MASK, + reg & DP_DP_HPD_INT_MASK); +} + +u32 msm_dp_aux_get_hpd_intr_status(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + int isr, mask; + + isr = msm_dp_read_aux(aux, REG_DP_DP_HPD_INT_STATUS); + msm_dp_write_aux(aux, REG_DP_DP_HPD_INT_ACK, + (isr & DP_DP_HPD_INT_MASK)); + mask = msm_dp_read_aux(aux, REG_DP_DP_HPD_INT_MASK); + + /* + * We only want to return interrupts that are unmasked to the caller. + * However, the interrupt status field also contains other + * informational bits about the HPD state status, so we only mask + * out the part of the register that tells us about which interrupts + * are pending. + */ + return isr & (mask | ~DP_DP_HPD_INT_MASK); +} + +u32 msm_dp_aux_is_link_connected(struct drm_dp_aux *msm_dp_aux) +{ + struct msm_dp_aux_private *aux = + container_of(msm_dp_aux, struct msm_dp_aux_private, msm_dp_aux); + u32 status; + + status = msm_dp_read_aux(aux, REG_DP_DP_HPD_INT_STATUS); + status >>= DP_DP_HPD_STATE_STATUS_BITS_SHIFT; + status &= DP_DP_HPD_STATE_STATUS_BITS_MASK; + + return status; +} + +struct drm_dp_aux *msm_dp_aux_get(struct device *dev, struct phy *phy, - bool is_edp) + bool is_edp, + void __iomem *aux_base) { struct msm_dp_aux_private *aux; - if (!catalog) { - DRM_ERROR("invalid input\n"); - return ERR_PTR(-ENODEV); - } - aux = devm_kzalloc(dev, sizeof(*aux), GFP_KERNEL); if (!aux) return ERR_PTR(-ENOMEM); @@ -544,9 +696,9 @@ struct drm_dp_aux *msm_dp_aux_get(struct device *dev, struct msm_dp_catalog *cat mutex_init(&aux->mutex); aux->dev = dev; - aux->catalog = catalog; aux->phy = phy; aux->retry_cnt = 0; + aux->aux_base = aux_base; /* * Use the drm_dp_aux_init() to use the aux adapter diff --git a/drivers/gpu/drm/msm/dp/dp_aux.h b/drivers/gpu/drm/msm/dp/dp_aux.h index 39c5b4c8596a..4be02e8b4d0b 100644 --- a/drivers/gpu/drm/msm/dp/dp_aux.h +++ b/drivers/gpu/drm/msm/dp/dp_aux.h @@ -6,21 +6,28 @@ #ifndef _DP_AUX_H_ #define _DP_AUX_H_ -#include "dp_catalog.h" #include <drm/display/drm_dp_helper.h> int msm_dp_aux_register(struct drm_dp_aux *msm_dp_aux); void msm_dp_aux_unregister(struct drm_dp_aux *msm_dp_aux); -irqreturn_t msm_dp_aux_isr(struct drm_dp_aux *msm_dp_aux); +irqreturn_t msm_dp_aux_isr(struct drm_dp_aux *msm_dp_aux, u32 isr); void msm_dp_aux_enable_xfers(struct drm_dp_aux *msm_dp_aux, bool enabled); void msm_dp_aux_init(struct drm_dp_aux *msm_dp_aux); void msm_dp_aux_deinit(struct drm_dp_aux *msm_dp_aux); void msm_dp_aux_reconfig(struct drm_dp_aux *msm_dp_aux); +void msm_dp_aux_hpd_enable(struct drm_dp_aux *msm_dp_aux); +void msm_dp_aux_hpd_disable(struct drm_dp_aux *msm_dp_aux); +void msm_dp_aux_hpd_intr_enable(struct drm_dp_aux *msm_dp_aux); +void msm_dp_aux_hpd_intr_disable(struct drm_dp_aux *msm_dp_aux); +u32 msm_dp_aux_get_hpd_intr_status(struct drm_dp_aux *msm_dp_aux); +u32 msm_dp_aux_is_link_connected(struct drm_dp_aux *msm_dp_aux); + struct phy; -struct drm_dp_aux *msm_dp_aux_get(struct device *dev, struct msm_dp_catalog *catalog, +struct drm_dp_aux *msm_dp_aux_get(struct device *dev, struct phy *phy, - bool is_edp); + bool is_edp, + void __iomem *aux_base); void msm_dp_aux_put(struct drm_dp_aux *aux); #endif /*__DP_AUX_H_*/ diff --git a/drivers/gpu/drm/msm/dp/dp_catalog.c b/drivers/gpu/drm/msm/dp/dp_catalog.c deleted file mode 100644 index 7b7eadb2f83b..000000000000 --- a/drivers/gpu/drm/msm/dp/dp_catalog.c +++ /dev/null @@ -1,1298 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved. - */ - -#define pr_fmt(fmt) "[drm-dp] %s: " fmt, __func__ - -#include <linux/delay.h> -#include <linux/iopoll.h> -#include <linux/platform_device.h> -#include <linux/rational.h> -#include <drm/display/drm_dp_helper.h> -#include <drm/drm_print.h> - -#include "dp_catalog.h" -#include "dp_reg.h" - -#define POLLING_SLEEP_US 1000 -#define POLLING_TIMEOUT_US 10000 - -#define SCRAMBLER_RESET_COUNT_VALUE 0xFC - -#define DP_INTERRUPT_STATUS_ACK_SHIFT 1 -#define DP_INTERRUPT_STATUS_MASK_SHIFT 2 - -#define DP_INTF_CONFIG_DATABUS_WIDEN BIT(4) - -#define DP_INTERRUPT_STATUS1 \ - (DP_INTR_AUX_XFER_DONE| \ - DP_INTR_WRONG_ADDR | DP_INTR_TIMEOUT | \ - DP_INTR_NACK_DEFER | DP_INTR_WRONG_DATA_CNT | \ - DP_INTR_I2C_NACK | DP_INTR_I2C_DEFER | \ - DP_INTR_PLL_UNLOCKED | DP_INTR_AUX_ERROR) - -#define DP_INTERRUPT_STATUS1_ACK \ - (DP_INTERRUPT_STATUS1 << DP_INTERRUPT_STATUS_ACK_SHIFT) -#define DP_INTERRUPT_STATUS1_MASK \ - (DP_INTERRUPT_STATUS1 << DP_INTERRUPT_STATUS_MASK_SHIFT) - -#define DP_INTERRUPT_STATUS2 \ - (DP_INTR_READY_FOR_VIDEO | DP_INTR_IDLE_PATTERN_SENT | \ - DP_INTR_FRAME_END | DP_INTR_CRC_UPDATED) - -#define DP_INTERRUPT_STATUS2_ACK \ - (DP_INTERRUPT_STATUS2 << DP_INTERRUPT_STATUS_ACK_SHIFT) -#define DP_INTERRUPT_STATUS2_MASK \ - (DP_INTERRUPT_STATUS2 << DP_INTERRUPT_STATUS_MASK_SHIFT) - -#define DP_INTERRUPT_STATUS4 \ - (PSR_UPDATE_INT | PSR_CAPTURE_INT | PSR_EXIT_INT | \ - PSR_UPDATE_ERROR_INT | PSR_WAKE_ERROR_INT) - -#define DP_INTERRUPT_MASK4 \ - (PSR_UPDATE_MASK | PSR_CAPTURE_MASK | PSR_EXIT_MASK | \ - PSR_UPDATE_ERROR_MASK | PSR_WAKE_ERROR_MASK) - -#define DP_DEFAULT_AHB_OFFSET 0x0000 -#define DP_DEFAULT_AHB_SIZE 0x0200 -#define DP_DEFAULT_AUX_OFFSET 0x0200 -#define DP_DEFAULT_AUX_SIZE 0x0200 -#define DP_DEFAULT_LINK_OFFSET 0x0400 -#define DP_DEFAULT_LINK_SIZE 0x0C00 -#define DP_DEFAULT_P0_OFFSET 0x1000 -#define DP_DEFAULT_P0_SIZE 0x0400 - -struct dss_io_region { - size_t len; - void __iomem *base; -}; - -struct dss_io_data { - struct dss_io_region ahb; - struct dss_io_region aux; - struct dss_io_region link; - struct dss_io_region p0; -}; - -struct msm_dp_catalog_private { - struct device *dev; - struct drm_device *drm_dev; - struct dss_io_data io; - struct msm_dp_catalog msm_dp_catalog; -}; - -void msm_dp_catalog_snapshot(struct msm_dp_catalog *msm_dp_catalog, struct msm_disp_state *disp_state) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - struct dss_io_data *dss = &catalog->io; - - msm_disp_snapshot_add_block(disp_state, dss->ahb.len, dss->ahb.base, "dp_ahb"); - msm_disp_snapshot_add_block(disp_state, dss->aux.len, dss->aux.base, "dp_aux"); - msm_disp_snapshot_add_block(disp_state, dss->link.len, dss->link.base, "dp_link"); - msm_disp_snapshot_add_block(disp_state, dss->p0.len, dss->p0.base, "dp_p0"); -} - -static inline u32 msm_dp_read_aux(struct msm_dp_catalog_private *catalog, u32 offset) -{ - return readl_relaxed(catalog->io.aux.base + offset); -} - -static inline void msm_dp_write_aux(struct msm_dp_catalog_private *catalog, - u32 offset, u32 data) -{ - /* - * To make sure aux reg writes happens before any other operation, - * this function uses writel() instread of writel_relaxed() - */ - writel(data, catalog->io.aux.base + offset); -} - -static inline u32 msm_dp_read_ahb(const struct msm_dp_catalog_private *catalog, u32 offset) -{ - return readl_relaxed(catalog->io.ahb.base + offset); -} - -static inline void msm_dp_write_ahb(struct msm_dp_catalog_private *catalog, - u32 offset, u32 data) -{ - /* - * To make sure phy reg writes happens before any other operation, - * this function uses writel() instread of writel_relaxed() - */ - writel(data, catalog->io.ahb.base + offset); -} - -static inline void msm_dp_write_p0(struct msm_dp_catalog_private *catalog, - u32 offset, u32 data) -{ - /* - * To make sure interface reg writes happens before any other operation, - * this function uses writel() instread of writel_relaxed() - */ - writel(data, catalog->io.p0.base + offset); -} - -static inline u32 msm_dp_read_p0(struct msm_dp_catalog_private *catalog, - u32 offset) -{ - /* - * To make sure interface reg writes happens before any other operation, - * this function uses writel() instread of writel_relaxed() - */ - return readl_relaxed(catalog->io.p0.base + offset); -} - -static inline u32 msm_dp_read_link(struct msm_dp_catalog_private *catalog, u32 offset) -{ - return readl_relaxed(catalog->io.link.base + offset); -} - -static inline void msm_dp_write_link(struct msm_dp_catalog_private *catalog, - u32 offset, u32 data) -{ - /* - * To make sure link reg writes happens before any other operation, - * this function uses writel() instread of writel_relaxed() - */ - writel(data, catalog->io.link.base + offset); -} - -/* aux related catalog functions */ -u32 msm_dp_catalog_aux_read_data(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - return msm_dp_read_aux(catalog, REG_DP_AUX_DATA); -} - -int msm_dp_catalog_aux_write_data(struct msm_dp_catalog *msm_dp_catalog, u32 data) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - msm_dp_write_aux(catalog, REG_DP_AUX_DATA, data); - return 0; -} - -int msm_dp_catalog_aux_write_trans(struct msm_dp_catalog *msm_dp_catalog, u32 data) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - msm_dp_write_aux(catalog, REG_DP_AUX_TRANS_CTRL, data); - return 0; -} - -int msm_dp_catalog_aux_clear_trans(struct msm_dp_catalog *msm_dp_catalog, bool read) -{ - u32 data; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - if (read) { - data = msm_dp_read_aux(catalog, REG_DP_AUX_TRANS_CTRL); - data &= ~DP_AUX_TRANS_CTRL_GO; - msm_dp_write_aux(catalog, REG_DP_AUX_TRANS_CTRL, data); - } else { - msm_dp_write_aux(catalog, REG_DP_AUX_TRANS_CTRL, 0); - } - return 0; -} - -int msm_dp_catalog_aux_clear_hw_interrupts(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - msm_dp_read_aux(catalog, REG_DP_PHY_AUX_INTERRUPT_STATUS); - msm_dp_write_aux(catalog, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0x1f); - msm_dp_write_aux(catalog, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0x9f); - msm_dp_write_aux(catalog, REG_DP_PHY_AUX_INTERRUPT_CLEAR, 0); - return 0; -} - -/** - * msm_dp_catalog_aux_reset() - reset AUX controller - * - * @msm_dp_catalog: DP catalog structure - * - * return: void - * - * This function reset AUX controller - * - * NOTE: reset AUX controller will also clear any pending HPD related interrupts - * - */ -void msm_dp_catalog_aux_reset(struct msm_dp_catalog *msm_dp_catalog) -{ - u32 aux_ctrl; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - aux_ctrl = msm_dp_read_aux(catalog, REG_DP_AUX_CTRL); - - aux_ctrl |= DP_AUX_CTRL_RESET; - msm_dp_write_aux(catalog, REG_DP_AUX_CTRL, aux_ctrl); - usleep_range(1000, 1100); /* h/w recommended delay */ - - aux_ctrl &= ~DP_AUX_CTRL_RESET; - msm_dp_write_aux(catalog, REG_DP_AUX_CTRL, aux_ctrl); -} - -void msm_dp_catalog_aux_enable(struct msm_dp_catalog *msm_dp_catalog, bool enable) -{ - u32 aux_ctrl; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - aux_ctrl = msm_dp_read_aux(catalog, REG_DP_AUX_CTRL); - - if (enable) { - msm_dp_write_aux(catalog, REG_DP_TIMEOUT_COUNT, 0xffff); - msm_dp_write_aux(catalog, REG_DP_AUX_LIMITS, 0xffff); - aux_ctrl |= DP_AUX_CTRL_ENABLE; - } else { - aux_ctrl &= ~DP_AUX_CTRL_ENABLE; - } - - msm_dp_write_aux(catalog, REG_DP_AUX_CTRL, aux_ctrl); -} - -int msm_dp_catalog_aux_wait_for_hpd_connect_state(struct msm_dp_catalog *msm_dp_catalog, - unsigned long wait_us) -{ - u32 state; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - /* poll for hpd connected status every 2ms and timeout after wait_us */ - return readl_poll_timeout(catalog->io.aux.base + - REG_DP_DP_HPD_INT_STATUS, - state, state & DP_DP_HPD_STATE_STATUS_CONNECTED, - min(wait_us, 2000), wait_us); -} - -u32 msm_dp_catalog_aux_get_irq(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 intr, intr_ack; - - intr = msm_dp_read_ahb(catalog, REG_DP_INTR_STATUS); - intr &= ~DP_INTERRUPT_STATUS1_MASK; - intr_ack = (intr & DP_INTERRUPT_STATUS1) - << DP_INTERRUPT_STATUS_ACK_SHIFT; - msm_dp_write_ahb(catalog, REG_DP_INTR_STATUS, intr_ack | - DP_INTERRUPT_STATUS1_MASK); - - return intr; - -} - -/* controller related catalog functions */ -void msm_dp_catalog_ctrl_update_transfer_unit(struct msm_dp_catalog *msm_dp_catalog, - u32 msm_dp_tu, u32 valid_boundary, - u32 valid_boundary2) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - msm_dp_write_link(catalog, REG_DP_VALID_BOUNDARY, valid_boundary); - msm_dp_write_link(catalog, REG_DP_TU, msm_dp_tu); - msm_dp_write_link(catalog, REG_DP_VALID_BOUNDARY_2, valid_boundary2); -} - -void msm_dp_catalog_ctrl_state_ctrl(struct msm_dp_catalog *msm_dp_catalog, u32 state) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - msm_dp_write_link(catalog, REG_DP_STATE_CTRL, state); -} - -void msm_dp_catalog_ctrl_config_ctrl(struct msm_dp_catalog *msm_dp_catalog, u32 cfg) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - drm_dbg_dp(catalog->drm_dev, "DP_CONFIGURATION_CTRL=0x%x\n", cfg); - - msm_dp_write_link(catalog, REG_DP_CONFIGURATION_CTRL, cfg); -} - -void msm_dp_catalog_ctrl_lane_mapping(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 ln_0 = 0, ln_1 = 1, ln_2 = 2, ln_3 = 3; /* One-to-One mapping */ - u32 ln_mapping; - - ln_mapping = ln_0 << LANE0_MAPPING_SHIFT; - ln_mapping |= ln_1 << LANE1_MAPPING_SHIFT; - ln_mapping |= ln_2 << LANE2_MAPPING_SHIFT; - ln_mapping |= ln_3 << LANE3_MAPPING_SHIFT; - - msm_dp_write_link(catalog, REG_DP_LOGICAL2PHYSICAL_LANE_MAPPING, - ln_mapping); -} - -void msm_dp_catalog_ctrl_psr_mainlink_enable(struct msm_dp_catalog *msm_dp_catalog, - bool enable) -{ - u32 val; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - val = msm_dp_read_link(catalog, REG_DP_MAINLINK_CTRL); - - if (enable) - val |= DP_MAINLINK_CTRL_ENABLE; - else - val &= ~DP_MAINLINK_CTRL_ENABLE; - - msm_dp_write_link(catalog, REG_DP_MAINLINK_CTRL, val); -} - -void msm_dp_catalog_ctrl_mainlink_ctrl(struct msm_dp_catalog *msm_dp_catalog, - bool enable) -{ - u32 mainlink_ctrl; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - drm_dbg_dp(catalog->drm_dev, "enable=%d\n", enable); - if (enable) { - /* - * To make sure link reg writes happens before other operation, - * msm_dp_write_link() function uses writel() - */ - mainlink_ctrl = msm_dp_read_link(catalog, REG_DP_MAINLINK_CTRL); - - mainlink_ctrl &= ~(DP_MAINLINK_CTRL_RESET | - DP_MAINLINK_CTRL_ENABLE); - msm_dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl); - - mainlink_ctrl |= DP_MAINLINK_CTRL_RESET; - msm_dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl); - - mainlink_ctrl &= ~DP_MAINLINK_CTRL_RESET; - msm_dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl); - - mainlink_ctrl |= (DP_MAINLINK_CTRL_ENABLE | - DP_MAINLINK_FB_BOUNDARY_SEL); - msm_dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl); - } else { - mainlink_ctrl = msm_dp_read_link(catalog, REG_DP_MAINLINK_CTRL); - mainlink_ctrl &= ~DP_MAINLINK_CTRL_ENABLE; - msm_dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl); - } -} - -void msm_dp_catalog_ctrl_config_misc(struct msm_dp_catalog *msm_dp_catalog, - u32 colorimetry_cfg, - u32 test_bits_depth) -{ - u32 misc_val; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - misc_val = msm_dp_read_link(catalog, REG_DP_MISC1_MISC0); - - /* clear bpp bits */ - misc_val &= ~(0x07 << DP_MISC0_TEST_BITS_DEPTH_SHIFT); - misc_val |= colorimetry_cfg << DP_MISC0_COLORIMETRY_CFG_SHIFT; - misc_val |= test_bits_depth << DP_MISC0_TEST_BITS_DEPTH_SHIFT; - /* Configure clock to synchronous mode */ - misc_val |= DP_MISC0_SYNCHRONOUS_CLK; - - drm_dbg_dp(catalog->drm_dev, "misc settings = 0x%x\n", misc_val); - msm_dp_write_link(catalog, REG_DP_MISC1_MISC0, misc_val); -} - -void msm_dp_catalog_setup_peripheral_flush(struct msm_dp_catalog *msm_dp_catalog) -{ - u32 mainlink_ctrl, hw_revision; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - mainlink_ctrl = msm_dp_read_link(catalog, REG_DP_MAINLINK_CTRL); - - hw_revision = msm_dp_catalog_hw_revision(msm_dp_catalog); - if (hw_revision >= DP_HW_VERSION_1_2) - mainlink_ctrl |= DP_MAINLINK_FLUSH_MODE_SDE_PERIPH_UPDATE; - else - mainlink_ctrl |= DP_MAINLINK_FLUSH_MODE_UPDATE_SDP; - - msm_dp_write_link(catalog, REG_DP_MAINLINK_CTRL, mainlink_ctrl); -} - -void msm_dp_catalog_ctrl_config_msa(struct msm_dp_catalog *msm_dp_catalog, - u32 rate, u32 stream_rate_khz, - bool is_ycbcr_420) -{ - u32 pixel_m, pixel_n; - u32 mvid, nvid, pixel_div = 0, dispcc_input_rate; - u32 const nvid_fixed = DP_LINK_CONSTANT_N_VALUE; - u32 const link_rate_hbr2 = 540000; - u32 const link_rate_hbr3 = 810000; - unsigned long den, num; - - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - if (rate == link_rate_hbr3) - pixel_div = 6; - else if (rate == 162000 || rate == 270000) - pixel_div = 2; - else if (rate == link_rate_hbr2) - pixel_div = 4; - else - DRM_ERROR("Invalid pixel mux divider\n"); - - dispcc_input_rate = (rate * 10) / pixel_div; - - rational_best_approximation(dispcc_input_rate, stream_rate_khz, - (unsigned long)(1 << 16) - 1, - (unsigned long)(1 << 16) - 1, &den, &num); - - den = ~(den - num); - den = den & 0xFFFF; - pixel_m = num; - pixel_n = den; - - mvid = (pixel_m & 0xFFFF) * 5; - nvid = (0xFFFF & (~pixel_n)) + (pixel_m & 0xFFFF); - - if (nvid < nvid_fixed) { - u32 temp; - - temp = (nvid_fixed / nvid) * nvid; - mvid = (nvid_fixed / nvid) * mvid; - nvid = temp; - } - - if (is_ycbcr_420) - mvid /= 2; - - if (link_rate_hbr2 == rate) - nvid *= 2; - - if (link_rate_hbr3 == rate) - nvid *= 3; - - drm_dbg_dp(catalog->drm_dev, "mvid=0x%x, nvid=0x%x\n", mvid, nvid); - msm_dp_write_link(catalog, REG_DP_SOFTWARE_MVID, mvid); - msm_dp_write_link(catalog, REG_DP_SOFTWARE_NVID, nvid); - msm_dp_write_p0(catalog, MMSS_DP_DSC_DTO, 0x0); -} - -int msm_dp_catalog_ctrl_set_pattern_state_bit(struct msm_dp_catalog *msm_dp_catalog, - u32 state_bit) -{ - int bit, ret; - u32 data; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - bit = BIT(state_bit - 1); - drm_dbg_dp(catalog->drm_dev, "hw: bit=%d train=%d\n", bit, state_bit); - msm_dp_catalog_ctrl_state_ctrl(msm_dp_catalog, bit); - - bit = BIT(state_bit - 1) << DP_MAINLINK_READY_LINK_TRAINING_SHIFT; - - /* Poll for mainlink ready status */ - ret = readx_poll_timeout(readl, catalog->io.link.base + - REG_DP_MAINLINK_READY, - data, data & bit, - POLLING_SLEEP_US, POLLING_TIMEOUT_US); - if (ret < 0) { - DRM_ERROR("set state_bit for link_train=%d failed\n", state_bit); - return ret; - } - return 0; -} - -/** - * msm_dp_catalog_hw_revision() - retrieve DP hw revision - * - * @msm_dp_catalog: DP catalog structure - * - * Return: DP controller hw revision - * - */ -u32 msm_dp_catalog_hw_revision(const struct msm_dp_catalog *msm_dp_catalog) -{ - const struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - return msm_dp_read_ahb(catalog, REG_DP_HW_VERSION); -} - -/** - * msm_dp_catalog_ctrl_reset() - reset DP controller - * - * @msm_dp_catalog: DP catalog structure - * - * return: void - * - * This function reset the DP controller - * - * NOTE: reset DP controller will also clear any pending HPD related interrupts - * - */ -void msm_dp_catalog_ctrl_reset(struct msm_dp_catalog *msm_dp_catalog) -{ - u32 sw_reset; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - sw_reset = msm_dp_read_ahb(catalog, REG_DP_SW_RESET); - - sw_reset |= DP_SW_RESET; - msm_dp_write_ahb(catalog, REG_DP_SW_RESET, sw_reset); - usleep_range(1000, 1100); /* h/w recommended delay */ - - sw_reset &= ~DP_SW_RESET; - msm_dp_write_ahb(catalog, REG_DP_SW_RESET, sw_reset); -} - -bool msm_dp_catalog_ctrl_mainlink_ready(struct msm_dp_catalog *msm_dp_catalog) -{ - u32 data; - int ret; - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - /* Poll for mainlink ready status */ - ret = readl_poll_timeout(catalog->io.link.base + - REG_DP_MAINLINK_READY, - data, data & DP_MAINLINK_READY_FOR_VIDEO, - POLLING_SLEEP_US, POLLING_TIMEOUT_US); - if (ret < 0) { - DRM_ERROR("mainlink not ready\n"); - return false; - } - - return true; -} - -void msm_dp_catalog_ctrl_enable_irq(struct msm_dp_catalog *msm_dp_catalog, - bool enable) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - if (enable) { - msm_dp_write_ahb(catalog, REG_DP_INTR_STATUS, - DP_INTERRUPT_STATUS1_MASK); - msm_dp_write_ahb(catalog, REG_DP_INTR_STATUS2, - DP_INTERRUPT_STATUS2_MASK); - } else { - msm_dp_write_ahb(catalog, REG_DP_INTR_STATUS, 0x00); - msm_dp_write_ahb(catalog, REG_DP_INTR_STATUS2, 0x00); - } -} - -void msm_dp_catalog_hpd_config_intr(struct msm_dp_catalog *msm_dp_catalog, - u32 intr_mask, bool en) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - u32 config = msm_dp_read_aux(catalog, REG_DP_DP_HPD_INT_MASK); - - config = (en ? config | intr_mask : config & ~intr_mask); - - drm_dbg_dp(catalog->drm_dev, "intr_mask=%#x config=%#x\n", - intr_mask, config); - msm_dp_write_aux(catalog, REG_DP_DP_HPD_INT_MASK, - config & DP_DP_HPD_INT_MASK); -} - -void msm_dp_catalog_ctrl_hpd_enable(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - u32 reftimer = msm_dp_read_aux(catalog, REG_DP_DP_HPD_REFTIMER); - - /* Configure REFTIMER and enable it */ - reftimer |= DP_DP_HPD_REFTIMER_ENABLE; - msm_dp_write_aux(catalog, REG_DP_DP_HPD_REFTIMER, reftimer); - - /* Enable HPD */ - msm_dp_write_aux(catalog, REG_DP_DP_HPD_CTRL, DP_DP_HPD_CTRL_HPD_EN); -} - -void msm_dp_catalog_ctrl_hpd_disable(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - u32 reftimer = msm_dp_read_aux(catalog, REG_DP_DP_HPD_REFTIMER); - - reftimer &= ~DP_DP_HPD_REFTIMER_ENABLE; - msm_dp_write_aux(catalog, REG_DP_DP_HPD_REFTIMER, reftimer); - - msm_dp_write_aux(catalog, REG_DP_DP_HPD_CTRL, 0); -} - -static void msm_dp_catalog_enable_sdp(struct msm_dp_catalog_private *catalog) -{ - /* trigger sdp */ - msm_dp_write_link(catalog, MMSS_DP_SDP_CFG3, UPDATE_SDP); - msm_dp_write_link(catalog, MMSS_DP_SDP_CFG3, 0x0); -} - -void msm_dp_catalog_ctrl_config_psr(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 config; - - /* enable PSR1 function */ - config = msm_dp_read_link(catalog, REG_PSR_CONFIG); - config |= PSR1_SUPPORTED; - msm_dp_write_link(catalog, REG_PSR_CONFIG, config); - - msm_dp_write_ahb(catalog, REG_DP_INTR_MASK4, DP_INTERRUPT_MASK4); - msm_dp_catalog_enable_sdp(catalog); -} - -void msm_dp_catalog_ctrl_set_psr(struct msm_dp_catalog *msm_dp_catalog, bool enter) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 cmd; - - cmd = msm_dp_read_link(catalog, REG_PSR_CMD); - - cmd &= ~(PSR_ENTER | PSR_EXIT); - - if (enter) - cmd |= PSR_ENTER; - else - cmd |= PSR_EXIT; - - msm_dp_catalog_enable_sdp(catalog); - msm_dp_write_link(catalog, REG_PSR_CMD, cmd); -} - -u32 msm_dp_catalog_link_is_connected(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 status; - - status = msm_dp_read_aux(catalog, REG_DP_DP_HPD_INT_STATUS); - drm_dbg_dp(catalog->drm_dev, "aux status: %#x\n", status); - status >>= DP_DP_HPD_STATE_STATUS_BITS_SHIFT; - status &= DP_DP_HPD_STATE_STATUS_BITS_MASK; - - return status; -} - -u32 msm_dp_catalog_hpd_get_intr_status(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - int isr, mask; - - isr = msm_dp_read_aux(catalog, REG_DP_DP_HPD_INT_STATUS); - msm_dp_write_aux(catalog, REG_DP_DP_HPD_INT_ACK, - (isr & DP_DP_HPD_INT_MASK)); - mask = msm_dp_read_aux(catalog, REG_DP_DP_HPD_INT_MASK); - - /* - * We only want to return interrupts that are unmasked to the caller. - * However, the interrupt status field also contains other - * informational bits about the HPD state status, so we only mask - * out the part of the register that tells us about which interrupts - * are pending. - */ - return isr & (mask | ~DP_DP_HPD_INT_MASK); -} - -u32 msm_dp_catalog_ctrl_read_psr_interrupt_status(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 intr, intr_ack; - - intr = msm_dp_read_ahb(catalog, REG_DP_INTR_STATUS4); - intr_ack = (intr & DP_INTERRUPT_STATUS4) - << DP_INTERRUPT_STATUS_ACK_SHIFT; - msm_dp_write_ahb(catalog, REG_DP_INTR_STATUS4, intr_ack); - - return intr; -} - -int msm_dp_catalog_ctrl_get_interrupt(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 intr, intr_ack; - - intr = msm_dp_read_ahb(catalog, REG_DP_INTR_STATUS2); - intr &= ~DP_INTERRUPT_STATUS2_MASK; - intr_ack = (intr & DP_INTERRUPT_STATUS2) - << DP_INTERRUPT_STATUS_ACK_SHIFT; - msm_dp_write_ahb(catalog, REG_DP_INTR_STATUS2, - intr_ack | DP_INTERRUPT_STATUS2_MASK); - - return intr; -} - -void msm_dp_catalog_ctrl_phy_reset(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - msm_dp_write_ahb(catalog, REG_DP_PHY_CTRL, - DP_PHY_CTRL_SW_RESET | DP_PHY_CTRL_SW_RESET_PLL); - usleep_range(1000, 1100); /* h/w recommended delay */ - msm_dp_write_ahb(catalog, REG_DP_PHY_CTRL, 0x0); -} - -void msm_dp_catalog_ctrl_send_phy_pattern(struct msm_dp_catalog *msm_dp_catalog, - u32 pattern) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 value = 0x0; - - /* Make sure to clear the current pattern before starting a new one */ - msm_dp_write_link(catalog, REG_DP_STATE_CTRL, 0x0); - - drm_dbg_dp(catalog->drm_dev, "pattern: %#x\n", pattern); - switch (pattern) { - case DP_PHY_TEST_PATTERN_D10_2: - msm_dp_write_link(catalog, REG_DP_STATE_CTRL, - DP_STATE_CTRL_LINK_TRAINING_PATTERN1); - break; - case DP_PHY_TEST_PATTERN_ERROR_COUNT: - value &= ~(1 << 16); - msm_dp_write_link(catalog, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, - value); - value |= SCRAMBLER_RESET_COUNT_VALUE; - msm_dp_write_link(catalog, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, - value); - msm_dp_write_link(catalog, REG_DP_MAINLINK_LEVELS, - DP_MAINLINK_SAFE_TO_EXIT_LEVEL_2); - msm_dp_write_link(catalog, REG_DP_STATE_CTRL, - DP_STATE_CTRL_LINK_SYMBOL_ERR_MEASURE); - break; - case DP_PHY_TEST_PATTERN_PRBS7: - msm_dp_write_link(catalog, REG_DP_STATE_CTRL, - DP_STATE_CTRL_LINK_PRBS7); - break; - case DP_PHY_TEST_PATTERN_80BIT_CUSTOM: - msm_dp_write_link(catalog, REG_DP_STATE_CTRL, - DP_STATE_CTRL_LINK_TEST_CUSTOM_PATTERN); - /* 00111110000011111000001111100000 */ - msm_dp_write_link(catalog, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG0, - 0x3E0F83E0); - /* 00001111100000111110000011111000 */ - msm_dp_write_link(catalog, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG1, - 0x0F83E0F8); - /* 1111100000111110 */ - msm_dp_write_link(catalog, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG2, - 0x0000F83E); - break; - case DP_PHY_TEST_PATTERN_CP2520: - value = msm_dp_read_link(catalog, REG_DP_MAINLINK_CTRL); - value &= ~DP_MAINLINK_CTRL_SW_BYPASS_SCRAMBLER; - msm_dp_write_link(catalog, REG_DP_MAINLINK_CTRL, value); - - value = DP_HBR2_ERM_PATTERN; - msm_dp_write_link(catalog, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, - value); - value |= SCRAMBLER_RESET_COUNT_VALUE; - msm_dp_write_link(catalog, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, - value); - msm_dp_write_link(catalog, REG_DP_MAINLINK_LEVELS, - DP_MAINLINK_SAFE_TO_EXIT_LEVEL_2); - msm_dp_write_link(catalog, REG_DP_STATE_CTRL, - DP_STATE_CTRL_LINK_SYMBOL_ERR_MEASURE); - value = msm_dp_read_link(catalog, REG_DP_MAINLINK_CTRL); - value |= DP_MAINLINK_CTRL_ENABLE; - msm_dp_write_link(catalog, REG_DP_MAINLINK_CTRL, value); - break; - case DP_PHY_TEST_PATTERN_SEL_MASK: - msm_dp_write_link(catalog, REG_DP_MAINLINK_CTRL, - DP_MAINLINK_CTRL_ENABLE); - msm_dp_write_link(catalog, REG_DP_STATE_CTRL, - DP_STATE_CTRL_LINK_TRAINING_PATTERN4); - break; - default: - drm_dbg_dp(catalog->drm_dev, - "No valid test pattern requested: %#x\n", pattern); - break; - } -} - -u32 msm_dp_catalog_ctrl_read_phy_pattern(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - return msm_dp_read_link(catalog, REG_DP_MAINLINK_READY); -} - -/* panel related catalog functions */ -int msm_dp_catalog_panel_timing_cfg(struct msm_dp_catalog *msm_dp_catalog, u32 total, - u32 sync_start, u32 width_blanking, u32 msm_dp_active) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 reg; - - msm_dp_write_link(catalog, REG_DP_TOTAL_HOR_VER, total); - msm_dp_write_link(catalog, REG_DP_START_HOR_VER_FROM_SYNC, sync_start); - msm_dp_write_link(catalog, REG_DP_HSYNC_VSYNC_WIDTH_POLARITY, width_blanking); - msm_dp_write_link(catalog, REG_DP_ACTIVE_HOR_VER, msm_dp_active); - - reg = msm_dp_read_p0(catalog, MMSS_DP_INTF_CONFIG); - - if (msm_dp_catalog->wide_bus_en) - reg |= DP_INTF_CONFIG_DATABUS_WIDEN; - else - reg &= ~DP_INTF_CONFIG_DATABUS_WIDEN; - - - DRM_DEBUG_DP("wide_bus_en=%d reg=%#x\n", msm_dp_catalog->wide_bus_en, reg); - - msm_dp_write_p0(catalog, MMSS_DP_INTF_CONFIG, reg); - return 0; -} - -static void msm_dp_catalog_panel_send_vsc_sdp(struct msm_dp_catalog *msm_dp_catalog, struct dp_sdp *vsc_sdp) -{ - struct msm_dp_catalog_private *catalog; - u32 header[2]; - u32 val; - int i; - - catalog = container_of(msm_dp_catalog, struct msm_dp_catalog_private, msm_dp_catalog); - - msm_dp_utils_pack_sdp_header(&vsc_sdp->sdp_header, header); - - msm_dp_write_link(catalog, MMSS_DP_GENERIC0_0, header[0]); - msm_dp_write_link(catalog, MMSS_DP_GENERIC0_1, header[1]); - - for (i = 0; i < sizeof(vsc_sdp->db); i += 4) { - val = ((vsc_sdp->db[i]) | (vsc_sdp->db[i + 1] << 8) | (vsc_sdp->db[i + 2] << 16) | - (vsc_sdp->db[i + 3] << 24)); - msm_dp_write_link(catalog, MMSS_DP_GENERIC0_2 + i, val); - } -} - -static void msm_dp_catalog_panel_update_sdp(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog; - u32 hw_revision; - - catalog = container_of(msm_dp_catalog, struct msm_dp_catalog_private, msm_dp_catalog); - - hw_revision = msm_dp_catalog_hw_revision(msm_dp_catalog); - if (hw_revision < DP_HW_VERSION_1_2 && hw_revision >= DP_HW_VERSION_1_0) { - msm_dp_write_link(catalog, MMSS_DP_SDP_CFG3, 0x01); - msm_dp_write_link(catalog, MMSS_DP_SDP_CFG3, 0x00); - } -} - -void msm_dp_catalog_panel_enable_vsc_sdp(struct msm_dp_catalog *msm_dp_catalog, struct dp_sdp *vsc_sdp) -{ - struct msm_dp_catalog_private *catalog; - u32 cfg, cfg2, misc; - - catalog = container_of(msm_dp_catalog, struct msm_dp_catalog_private, msm_dp_catalog); - - cfg = msm_dp_read_link(catalog, MMSS_DP_SDP_CFG); - cfg2 = msm_dp_read_link(catalog, MMSS_DP_SDP_CFG2); - misc = msm_dp_read_link(catalog, REG_DP_MISC1_MISC0); - - cfg |= GEN0_SDP_EN; - msm_dp_write_link(catalog, MMSS_DP_SDP_CFG, cfg); - - cfg2 |= GENERIC0_SDPSIZE_VALID; - msm_dp_write_link(catalog, MMSS_DP_SDP_CFG2, cfg2); - - msm_dp_catalog_panel_send_vsc_sdp(msm_dp_catalog, vsc_sdp); - - /* indicates presence of VSC (BIT(6) of MISC1) */ - misc |= DP_MISC1_VSC_SDP; - - drm_dbg_dp(catalog->drm_dev, "vsc sdp enable=1\n"); - - pr_debug("misc settings = 0x%x\n", misc); - msm_dp_write_link(catalog, REG_DP_MISC1_MISC0, misc); - - msm_dp_catalog_panel_update_sdp(msm_dp_catalog); -} - -void msm_dp_catalog_panel_disable_vsc_sdp(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog; - u32 cfg, cfg2, misc; - - catalog = container_of(msm_dp_catalog, struct msm_dp_catalog_private, msm_dp_catalog); - - cfg = msm_dp_read_link(catalog, MMSS_DP_SDP_CFG); - cfg2 = msm_dp_read_link(catalog, MMSS_DP_SDP_CFG2); - misc = msm_dp_read_link(catalog, REG_DP_MISC1_MISC0); - - cfg &= ~GEN0_SDP_EN; - msm_dp_write_link(catalog, MMSS_DP_SDP_CFG, cfg); - - cfg2 &= ~GENERIC0_SDPSIZE_VALID; - msm_dp_write_link(catalog, MMSS_DP_SDP_CFG2, cfg2); - - /* switch back to MSA */ - misc &= ~DP_MISC1_VSC_SDP; - - drm_dbg_dp(catalog->drm_dev, "vsc sdp enable=0\n"); - - pr_debug("misc settings = 0x%x\n", misc); - msm_dp_write_link(catalog, REG_DP_MISC1_MISC0, misc); - - msm_dp_catalog_panel_update_sdp(msm_dp_catalog); -} - -void msm_dp_catalog_panel_tpg_enable(struct msm_dp_catalog *msm_dp_catalog, - struct drm_display_mode *drm_mode) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 hsync_period, vsync_period; - u32 display_v_start, display_v_end; - u32 hsync_start_x, hsync_end_x; - u32 v_sync_width; - u32 hsync_ctl; - u32 display_hctl; - - /* TPG config parameters*/ - hsync_period = drm_mode->htotal; - vsync_period = drm_mode->vtotal; - - display_v_start = ((drm_mode->vtotal - drm_mode->vsync_start) * - hsync_period); - display_v_end = ((vsync_period - (drm_mode->vsync_start - - drm_mode->vdisplay)) - * hsync_period) - 1; - - display_v_start += drm_mode->htotal - drm_mode->hsync_start; - display_v_end -= (drm_mode->hsync_start - drm_mode->hdisplay); - - hsync_start_x = drm_mode->htotal - drm_mode->hsync_start; - hsync_end_x = hsync_period - (drm_mode->hsync_start - - drm_mode->hdisplay) - 1; - - v_sync_width = drm_mode->vsync_end - drm_mode->vsync_start; - - hsync_ctl = (hsync_period << 16) | - (drm_mode->hsync_end - drm_mode->hsync_start); - display_hctl = (hsync_end_x << 16) | hsync_start_x; - - - msm_dp_write_p0(catalog, MMSS_DP_INTF_HSYNC_CTL, hsync_ctl); - msm_dp_write_p0(catalog, MMSS_DP_INTF_VSYNC_PERIOD_F0, vsync_period * - hsync_period); - msm_dp_write_p0(catalog, MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F0, v_sync_width * - hsync_period); - msm_dp_write_p0(catalog, MMSS_DP_INTF_VSYNC_PERIOD_F1, 0); - msm_dp_write_p0(catalog, MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F1, 0); - msm_dp_write_p0(catalog, MMSS_DP_INTF_DISPLAY_HCTL, display_hctl); - msm_dp_write_p0(catalog, MMSS_DP_INTF_ACTIVE_HCTL, 0); - msm_dp_write_p0(catalog, MMSS_INTF_DISPLAY_V_START_F0, display_v_start); - msm_dp_write_p0(catalog, MMSS_DP_INTF_DISPLAY_V_END_F0, display_v_end); - msm_dp_write_p0(catalog, MMSS_INTF_DISPLAY_V_START_F1, 0); - msm_dp_write_p0(catalog, MMSS_DP_INTF_DISPLAY_V_END_F1, 0); - msm_dp_write_p0(catalog, MMSS_DP_INTF_ACTIVE_V_START_F0, 0); - msm_dp_write_p0(catalog, MMSS_DP_INTF_ACTIVE_V_END_F0, 0); - msm_dp_write_p0(catalog, MMSS_DP_INTF_ACTIVE_V_START_F1, 0); - msm_dp_write_p0(catalog, MMSS_DP_INTF_ACTIVE_V_END_F1, 0); - msm_dp_write_p0(catalog, MMSS_DP_INTF_POLARITY_CTL, 0); - - msm_dp_write_p0(catalog, MMSS_DP_TPG_MAIN_CONTROL, - DP_TPG_CHECKERED_RECT_PATTERN); - msm_dp_write_p0(catalog, MMSS_DP_TPG_VIDEO_CONFIG, - DP_TPG_VIDEO_CONFIG_BPP_8BIT | - DP_TPG_VIDEO_CONFIG_RGB); - msm_dp_write_p0(catalog, MMSS_DP_BIST_ENABLE, - DP_BIST_ENABLE_DPBIST_EN); - msm_dp_write_p0(catalog, MMSS_DP_TIMING_ENGINE_EN, - DP_TIMING_ENGINE_EN_EN); - drm_dbg_dp(catalog->drm_dev, "%s: enabled tpg\n", __func__); -} - -void msm_dp_catalog_panel_tpg_disable(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - msm_dp_write_p0(catalog, MMSS_DP_TPG_MAIN_CONTROL, 0x0); - msm_dp_write_p0(catalog, MMSS_DP_BIST_ENABLE, 0x0); - msm_dp_write_p0(catalog, MMSS_DP_TIMING_ENGINE_EN, 0x0); -} - -static void __iomem *msm_dp_ioremap(struct platform_device *pdev, int idx, size_t *len) -{ - struct resource *res; - void __iomem *base; - - base = devm_platform_get_and_ioremap_resource(pdev, idx, &res); - if (!IS_ERR(base)) - *len = resource_size(res); - - return base; -} - -static int msm_dp_catalog_get_io(struct msm_dp_catalog_private *catalog) -{ - struct platform_device *pdev = to_platform_device(catalog->dev); - struct dss_io_data *dss = &catalog->io; - - dss->ahb.base = msm_dp_ioremap(pdev, 0, &dss->ahb.len); - if (IS_ERR(dss->ahb.base)) - return PTR_ERR(dss->ahb.base); - - dss->aux.base = msm_dp_ioremap(pdev, 1, &dss->aux.len); - if (IS_ERR(dss->aux.base)) { - /* - * The initial binding had a single reg, but in order to - * support variation in the sub-region sizes this was split. - * msm_dp_ioremap() will fail with -EINVAL here if only a single - * reg is specified, so fill in the sub-region offsets and - * lengths based on this single region. - */ - if (PTR_ERR(dss->aux.base) == -EINVAL) { - if (dss->ahb.len < DP_DEFAULT_P0_OFFSET + DP_DEFAULT_P0_SIZE) { - DRM_ERROR("legacy memory region not large enough\n"); - return -EINVAL; - } - - dss->ahb.len = DP_DEFAULT_AHB_SIZE; - dss->aux.base = dss->ahb.base + DP_DEFAULT_AUX_OFFSET; - dss->aux.len = DP_DEFAULT_AUX_SIZE; - dss->link.base = dss->ahb.base + DP_DEFAULT_LINK_OFFSET; - dss->link.len = DP_DEFAULT_LINK_SIZE; - dss->p0.base = dss->ahb.base + DP_DEFAULT_P0_OFFSET; - dss->p0.len = DP_DEFAULT_P0_SIZE; - } else { - DRM_ERROR("unable to remap aux region: %pe\n", dss->aux.base); - return PTR_ERR(dss->aux.base); - } - } else { - dss->link.base = msm_dp_ioremap(pdev, 2, &dss->link.len); - if (IS_ERR(dss->link.base)) { - DRM_ERROR("unable to remap link region: %pe\n", dss->link.base); - return PTR_ERR(dss->link.base); - } - - dss->p0.base = msm_dp_ioremap(pdev, 3, &dss->p0.len); - if (IS_ERR(dss->p0.base)) { - DRM_ERROR("unable to remap p0 region: %pe\n", dss->p0.base); - return PTR_ERR(dss->p0.base); - } - } - - return 0; -} - -struct msm_dp_catalog *msm_dp_catalog_get(struct device *dev) -{ - struct msm_dp_catalog_private *catalog; - int ret; - - catalog = devm_kzalloc(dev, sizeof(*catalog), GFP_KERNEL); - if (!catalog) - return ERR_PTR(-ENOMEM); - - catalog->dev = dev; - - ret = msm_dp_catalog_get_io(catalog); - if (ret) - return ERR_PTR(ret); - - return &catalog->msm_dp_catalog; -} - -void msm_dp_catalog_write_audio_stream(struct msm_dp_catalog *msm_dp_catalog, - struct dp_sdp_header *sdp_hdr) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 header[2]; - - msm_dp_utils_pack_sdp_header(sdp_hdr, header); - - msm_dp_write_link(catalog, MMSS_DP_AUDIO_STREAM_0, header[0]); - msm_dp_write_link(catalog, MMSS_DP_AUDIO_STREAM_1, header[1]); -} - -void msm_dp_catalog_write_audio_timestamp(struct msm_dp_catalog *msm_dp_catalog, - struct dp_sdp_header *sdp_hdr) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 header[2]; - - msm_dp_utils_pack_sdp_header(sdp_hdr, header); - - msm_dp_write_link(catalog, MMSS_DP_AUDIO_TIMESTAMP_0, header[0]); - msm_dp_write_link(catalog, MMSS_DP_AUDIO_TIMESTAMP_1, header[1]); -} - -void msm_dp_catalog_write_audio_infoframe(struct msm_dp_catalog *msm_dp_catalog, - struct dp_sdp_header *sdp_hdr) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 header[2]; - - msm_dp_utils_pack_sdp_header(sdp_hdr, header); - - msm_dp_write_link(catalog, MMSS_DP_AUDIO_INFOFRAME_0, header[0]); - msm_dp_write_link(catalog, MMSS_DP_AUDIO_INFOFRAME_1, header[1]); -} - -void msm_dp_catalog_write_audio_copy_mgmt(struct msm_dp_catalog *msm_dp_catalog, - struct dp_sdp_header *sdp_hdr) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - u32 header[2]; - - msm_dp_utils_pack_sdp_header(sdp_hdr, header); - - msm_dp_write_link(catalog, MMSS_DP_AUDIO_COPYMANAGEMENT_0, header[0]); - msm_dp_write_link(catalog, MMSS_DP_AUDIO_COPYMANAGEMENT_1, header[1]); -} - -void msm_dp_catalog_write_audio_isrc(struct msm_dp_catalog *msm_dp_catalog, - struct dp_sdp_header *sdp_hdr) -{ - struct msm_dp_catalog_private *catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - struct dp_sdp_header tmp = *sdp_hdr; - u32 header[2]; - u32 reg; - - /* XXX: is it necessary to preserve this field? */ - reg = msm_dp_read_link(catalog, MMSS_DP_AUDIO_ISRC_1); - tmp.HB3 = FIELD_GET(HEADER_3_MASK, reg); - - msm_dp_utils_pack_sdp_header(&tmp, header); - - msm_dp_write_link(catalog, MMSS_DP_AUDIO_ISRC_0, header[0]); - msm_dp_write_link(catalog, MMSS_DP_AUDIO_ISRC_1, header[1]); -} - -void msm_dp_catalog_audio_config_acr(struct msm_dp_catalog *msm_dp_catalog, u32 select) -{ - struct msm_dp_catalog_private *catalog; - u32 acr_ctrl; - - if (!msm_dp_catalog) - return; - - catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - acr_ctrl = select << 4 | BIT(31) | BIT(8) | BIT(14); - - drm_dbg_dp(catalog->drm_dev, "select: %#x, acr_ctrl: %#x\n", - select, acr_ctrl); - - msm_dp_write_link(catalog, MMSS_DP_AUDIO_ACR_CTRL, acr_ctrl); -} - -void msm_dp_catalog_audio_enable(struct msm_dp_catalog *msm_dp_catalog, bool enable) -{ - struct msm_dp_catalog_private *catalog; - u32 audio_ctrl; - - if (!msm_dp_catalog) - return; - - catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - audio_ctrl = msm_dp_read_link(catalog, MMSS_DP_AUDIO_CFG); - - if (enable) - audio_ctrl |= BIT(0); - else - audio_ctrl &= ~BIT(0); - - drm_dbg_dp(catalog->drm_dev, "dp_audio_cfg = 0x%x\n", audio_ctrl); - - msm_dp_write_link(catalog, MMSS_DP_AUDIO_CFG, audio_ctrl); - /* make sure audio engine is disabled */ - wmb(); -} - -void msm_dp_catalog_audio_config_sdp(struct msm_dp_catalog *msm_dp_catalog) -{ - struct msm_dp_catalog_private *catalog; - u32 sdp_cfg = 0; - u32 sdp_cfg2 = 0; - - if (!msm_dp_catalog) - return; - - catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - sdp_cfg = msm_dp_read_link(catalog, MMSS_DP_SDP_CFG); - /* AUDIO_TIMESTAMP_SDP_EN */ - sdp_cfg |= BIT(1); - /* AUDIO_STREAM_SDP_EN */ - sdp_cfg |= BIT(2); - /* AUDIO_COPY_MANAGEMENT_SDP_EN */ - sdp_cfg |= BIT(5); - /* AUDIO_ISRC_SDP_EN */ - sdp_cfg |= BIT(6); - /* AUDIO_INFOFRAME_SDP_EN */ - sdp_cfg |= BIT(20); - - drm_dbg_dp(catalog->drm_dev, "sdp_cfg = 0x%x\n", sdp_cfg); - - msm_dp_write_link(catalog, MMSS_DP_SDP_CFG, sdp_cfg); - - sdp_cfg2 = msm_dp_read_link(catalog, MMSS_DP_SDP_CFG2); - /* IFRM_REGSRC -> Do not use reg values */ - sdp_cfg2 &= ~BIT(0); - /* AUDIO_STREAM_HB3_REGSRC-> Do not use reg values */ - sdp_cfg2 &= ~BIT(1); - - drm_dbg_dp(catalog->drm_dev, "sdp_cfg2 = 0x%x\n", sdp_cfg2); - - msm_dp_write_link(catalog, MMSS_DP_SDP_CFG2, sdp_cfg2); -} - -void msm_dp_catalog_audio_sfe_level(struct msm_dp_catalog *msm_dp_catalog, u32 safe_to_exit_level) -{ - struct msm_dp_catalog_private *catalog; - u32 mainlink_levels; - - if (!msm_dp_catalog) - return; - - catalog = container_of(msm_dp_catalog, - struct msm_dp_catalog_private, msm_dp_catalog); - - mainlink_levels = msm_dp_read_link(catalog, REG_DP_MAINLINK_LEVELS); - mainlink_levels &= 0xFE0; - mainlink_levels |= safe_to_exit_level; - - drm_dbg_dp(catalog->drm_dev, - "mainlink_level = 0x%x, safe_to_exit_level = 0x%x\n", - mainlink_levels, safe_to_exit_level); - - msm_dp_write_link(catalog, REG_DP_MAINLINK_LEVELS, mainlink_levels); -} diff --git a/drivers/gpu/drm/msm/dp/dp_catalog.h b/drivers/gpu/drm/msm/dp/dp_catalog.h deleted file mode 100644 index 6678b0ac9a67..000000000000 --- a/drivers/gpu/drm/msm/dp/dp_catalog.h +++ /dev/null @@ -1,113 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Copyright (c) 2017-2020, The Linux Foundation. All rights reserved. - */ - -#ifndef _DP_CATALOG_H_ -#define _DP_CATALOG_H_ - -#include <drm/drm_modes.h> - -#include "dp_utils.h" -#include "disp/msm_disp_snapshot.h" - -/* interrupts */ -#define DP_INTR_HPD BIT(0) -#define DP_INTR_AUX_XFER_DONE BIT(3) -#define DP_INTR_WRONG_ADDR BIT(6) -#define DP_INTR_TIMEOUT BIT(9) -#define DP_INTR_NACK_DEFER BIT(12) -#define DP_INTR_WRONG_DATA_CNT BIT(15) -#define DP_INTR_I2C_NACK BIT(18) -#define DP_INTR_I2C_DEFER BIT(21) -#define DP_INTR_PLL_UNLOCKED BIT(24) -#define DP_INTR_AUX_ERROR BIT(27) - -#define DP_INTR_READY_FOR_VIDEO BIT(0) -#define DP_INTR_IDLE_PATTERN_SENT BIT(3) -#define DP_INTR_FRAME_END BIT(6) -#define DP_INTR_CRC_UPDATED BIT(9) - -#define DP_HW_VERSION_1_0 0x10000000 -#define DP_HW_VERSION_1_2 0x10020000 - -struct msm_dp_catalog { - bool wide_bus_en; -}; - -/* Debug module */ -void msm_dp_catalog_snapshot(struct msm_dp_catalog *msm_dp_catalog, struct msm_disp_state *disp_state); - -/* AUX APIs */ -u32 msm_dp_catalog_aux_read_data(struct msm_dp_catalog *msm_dp_catalog); -int msm_dp_catalog_aux_write_data(struct msm_dp_catalog *msm_dp_catalog, u32 data); -int msm_dp_catalog_aux_write_trans(struct msm_dp_catalog *msm_dp_catalog, u32 data); -int msm_dp_catalog_aux_clear_trans(struct msm_dp_catalog *msm_dp_catalog, bool read); -int msm_dp_catalog_aux_clear_hw_interrupts(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_aux_reset(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_aux_enable(struct msm_dp_catalog *msm_dp_catalog, bool enable); -int msm_dp_catalog_aux_wait_for_hpd_connect_state(struct msm_dp_catalog *msm_dp_catalog, - unsigned long wait_us); -u32 msm_dp_catalog_aux_get_irq(struct msm_dp_catalog *msm_dp_catalog); - -/* DP Controller APIs */ -void msm_dp_catalog_ctrl_state_ctrl(struct msm_dp_catalog *msm_dp_catalog, u32 state); -void msm_dp_catalog_ctrl_config_ctrl(struct msm_dp_catalog *msm_dp_catalog, u32 config); -void msm_dp_catalog_ctrl_lane_mapping(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_ctrl_mainlink_ctrl(struct msm_dp_catalog *msm_dp_catalog, bool enable); -void msm_dp_catalog_ctrl_psr_mainlink_enable(struct msm_dp_catalog *msm_dp_catalog, bool enable); -void msm_dp_catalog_setup_peripheral_flush(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_ctrl_config_misc(struct msm_dp_catalog *msm_dp_catalog, u32 cc, u32 tb); -void msm_dp_catalog_ctrl_config_msa(struct msm_dp_catalog *msm_dp_catalog, u32 rate, - u32 stream_rate_khz, bool is_ycbcr_420); -int msm_dp_catalog_ctrl_set_pattern_state_bit(struct msm_dp_catalog *msm_dp_catalog, u32 pattern); -u32 msm_dp_catalog_hw_revision(const struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_ctrl_reset(struct msm_dp_catalog *msm_dp_catalog); -bool msm_dp_catalog_ctrl_mainlink_ready(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_ctrl_enable_irq(struct msm_dp_catalog *msm_dp_catalog, bool enable); -void msm_dp_catalog_hpd_config_intr(struct msm_dp_catalog *msm_dp_catalog, - u32 intr_mask, bool en); -void msm_dp_catalog_ctrl_hpd_enable(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_ctrl_hpd_disable(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_ctrl_config_psr(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_ctrl_set_psr(struct msm_dp_catalog *msm_dp_catalog, bool enter); -u32 msm_dp_catalog_link_is_connected(struct msm_dp_catalog *msm_dp_catalog); -u32 msm_dp_catalog_hpd_get_intr_status(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_ctrl_phy_reset(struct msm_dp_catalog *msm_dp_catalog); -int msm_dp_catalog_ctrl_get_interrupt(struct msm_dp_catalog *msm_dp_catalog); -u32 msm_dp_catalog_ctrl_read_psr_interrupt_status(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_ctrl_update_transfer_unit(struct msm_dp_catalog *msm_dp_catalog, - u32 msm_dp_tu, u32 valid_boundary, - u32 valid_boundary2); -void msm_dp_catalog_ctrl_send_phy_pattern(struct msm_dp_catalog *msm_dp_catalog, - u32 pattern); -u32 msm_dp_catalog_ctrl_read_phy_pattern(struct msm_dp_catalog *msm_dp_catalog); - -/* DP Panel APIs */ -int msm_dp_catalog_panel_timing_cfg(struct msm_dp_catalog *msm_dp_catalog, u32 total, - u32 sync_start, u32 width_blanking, u32 msm_dp_active); -void msm_dp_catalog_panel_enable_vsc_sdp(struct msm_dp_catalog *msm_dp_catalog, struct dp_sdp *vsc_sdp); -void msm_dp_catalog_panel_disable_vsc_sdp(struct msm_dp_catalog *msm_dp_catalog); -void msm_dp_catalog_panel_tpg_enable(struct msm_dp_catalog *msm_dp_catalog, - struct drm_display_mode *drm_mode); -void msm_dp_catalog_panel_tpg_disable(struct msm_dp_catalog *msm_dp_catalog); - -struct msm_dp_catalog *msm_dp_catalog_get(struct device *dev); - -/* DP Audio APIs */ -void msm_dp_catalog_write_audio_stream(struct msm_dp_catalog *msm_dp_catalog, - struct dp_sdp_header *sdp_hdr); -void msm_dp_catalog_write_audio_timestamp(struct msm_dp_catalog *msm_dp_catalog, - struct dp_sdp_header *sdp_hdr); -void msm_dp_catalog_write_audio_infoframe(struct msm_dp_catalog *msm_dp_catalog, - struct dp_sdp_header *sdp_hdr); -void msm_dp_catalog_write_audio_copy_mgmt(struct msm_dp_catalog *msm_dp_catalog, - struct dp_sdp_header *sdp_hdr); -void msm_dp_catalog_write_audio_isrc(struct msm_dp_catalog *msm_dp_catalog, - struct dp_sdp_header *sdp_hdr); -void msm_dp_catalog_audio_config_acr(struct msm_dp_catalog *catalog, u32 select); -void msm_dp_catalog_audio_enable(struct msm_dp_catalog *catalog, bool enable); -void msm_dp_catalog_audio_config_sdp(struct msm_dp_catalog *catalog); -void msm_dp_catalog_audio_sfe_level(struct msm_dp_catalog *catalog, u32 safe_to_exit_level); - -#endif /* _DP_CATALOG_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.c b/drivers/gpu/drm/msm/dp/dp_ctrl.c index a50bfafbb4ea..c42fd2c17a32 100644 --- a/drivers/gpu/drm/msm/dp/dp_ctrl.c +++ b/drivers/gpu/drm/msm/dp/dp_ctrl.c @@ -6,14 +6,18 @@ #define pr_fmt(fmt) "[drm-dp] %s: " fmt, __func__ #include <linux/types.h> +#include <linux/clk.h> #include <linux/completion.h> #include <linux/delay.h> +#include <linux/iopoll.h> #include <linux/phy/phy.h> #include <linux/phy/phy-dp.h> #include <linux/pm_opp.h> +#include <linux/rational.h> #include <linux/string_choices.h> #include <drm/display/drm_dp_helper.h> +#include <drm/drm_device.h> #include <drm/drm_fixed.h> #include <drm/drm_print.h> @@ -21,11 +25,46 @@ #include "dp_ctrl.h" #include "dp_link.h" +#define POLLING_SLEEP_US 1000 +#define POLLING_TIMEOUT_US 10000 + #define DP_KHZ_TO_HZ 1000 #define IDLE_PATTERN_COMPLETION_TIMEOUT_JIFFIES (30 * HZ / 1000) /* 30 ms */ #define PSR_OPERATION_COMPLETION_TIMEOUT_JIFFIES (300 * HZ / 1000) /* 300 ms */ #define WAIT_FOR_VIDEO_READY_TIMEOUT_JIFFIES (HZ / 2) +#define DP_INTERRUPT_STATUS_ACK_SHIFT 1 +#define DP_INTERRUPT_STATUS_MASK_SHIFT 2 + +#define DP_INTERRUPT_STATUS1 \ + (DP_INTR_AUX_XFER_DONE| \ + DP_INTR_WRONG_ADDR | DP_INTR_TIMEOUT | \ + DP_INTR_NACK_DEFER | DP_INTR_WRONG_DATA_CNT | \ + DP_INTR_I2C_NACK | DP_INTR_I2C_DEFER | \ + DP_INTR_PLL_UNLOCKED | DP_INTR_AUX_ERROR) + +#define DP_INTERRUPT_STATUS1_ACK \ + (DP_INTERRUPT_STATUS1 << DP_INTERRUPT_STATUS_ACK_SHIFT) +#define DP_INTERRUPT_STATUS1_MASK \ + (DP_INTERRUPT_STATUS1 << DP_INTERRUPT_STATUS_MASK_SHIFT) + +#define DP_INTERRUPT_STATUS2 \ + (DP_INTR_READY_FOR_VIDEO | DP_INTR_IDLE_PATTERN_SENT | \ + DP_INTR_FRAME_END | DP_INTR_CRC_UPDATED) + +#define DP_INTERRUPT_STATUS2_ACK \ + (DP_INTERRUPT_STATUS2 << DP_INTERRUPT_STATUS_ACK_SHIFT) +#define DP_INTERRUPT_STATUS2_MASK \ + (DP_INTERRUPT_STATUS2 << DP_INTERRUPT_STATUS_MASK_SHIFT) + +#define DP_INTERRUPT_STATUS4 \ + (PSR_UPDATE_INT | PSR_CAPTURE_INT | PSR_EXIT_INT | \ + PSR_UPDATE_ERROR_INT | PSR_WAKE_ERROR_INT) + +#define DP_INTERRUPT_MASK4 \ + (PSR_UPDATE_MASK | PSR_CAPTURE_MASK | PSR_EXIT_MASK | \ + PSR_UPDATE_ERROR_MASK | PSR_WAKE_ERROR_MASK) + #define DP_CTRL_INTR_READY_FOR_VIDEO BIT(0) #define DP_CTRL_INTR_IDLE_PATTERN_SENT BIT(3) @@ -77,7 +116,8 @@ struct msm_dp_ctrl_private { struct drm_dp_aux *aux; struct msm_dp_panel *panel; struct msm_dp_link *link; - struct msm_dp_catalog *catalog; + void __iomem *ahb_base; + void __iomem *link_base; struct phy *phy; @@ -95,11 +135,43 @@ struct msm_dp_ctrl_private { struct completion psr_op_comp; struct completion video_comp; + u32 hw_revision; + bool core_clks_on; bool link_clks_on; bool stream_clks_on; }; +static inline u32 msm_dp_read_ahb(const struct msm_dp_ctrl_private *ctrl, u32 offset) +{ + return readl_relaxed(ctrl->ahb_base + offset); +} + +static inline void msm_dp_write_ahb(struct msm_dp_ctrl_private *ctrl, + u32 offset, u32 data) +{ + /* + * To make sure phy reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, ctrl->ahb_base + offset); +} + +static inline u32 msm_dp_read_link(struct msm_dp_ctrl_private *ctrl, u32 offset) +{ + return readl_relaxed(ctrl->link_base + offset); +} + +static inline void msm_dp_write_link(struct msm_dp_ctrl_private *ctrl, + u32 offset, u32 data) +{ + /* + * To make sure link reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, ctrl->link_base + offset); +} + static int msm_dp_aux_link_configure(struct drm_dp_aux *aux, struct msm_dp_link_info *link) { @@ -119,6 +191,179 @@ static int msm_dp_aux_link_configure(struct drm_dp_aux *aux, return 0; } +/* + * NOTE: resetting DP controller will also clear any pending HPD related interrupts + */ +void msm_dp_ctrl_reset(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl = + container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + u32 sw_reset; + + sw_reset = msm_dp_read_ahb(ctrl, REG_DP_SW_RESET); + + sw_reset |= DP_SW_RESET; + msm_dp_write_ahb(ctrl, REG_DP_SW_RESET, sw_reset); + usleep_range(1000, 1100); /* h/w recommended delay */ + + sw_reset &= ~DP_SW_RESET; + msm_dp_write_ahb(ctrl, REG_DP_SW_RESET, sw_reset); + + if (!ctrl->hw_revision) { + ctrl->hw_revision = msm_dp_read_ahb(ctrl, REG_DP_HW_VERSION); + ctrl->panel->hw_revision = ctrl->hw_revision; + } +} + +static u32 msm_dp_ctrl_get_aux_interrupt(struct msm_dp_ctrl_private *ctrl) +{ + u32 intr, intr_ack; + + intr = msm_dp_read_ahb(ctrl, REG_DP_INTR_STATUS); + intr &= ~DP_INTERRUPT_STATUS1_MASK; + intr_ack = (intr & DP_INTERRUPT_STATUS1) + << DP_INTERRUPT_STATUS_ACK_SHIFT; + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS, + intr_ack | DP_INTERRUPT_STATUS1_MASK); + + return intr; + +} + +static u32 msm_dp_ctrl_get_interrupt(struct msm_dp_ctrl_private *ctrl) +{ + u32 intr, intr_ack; + + intr = msm_dp_read_ahb(ctrl, REG_DP_INTR_STATUS2); + intr &= ~DP_INTERRUPT_STATUS2_MASK; + intr_ack = (intr & DP_INTERRUPT_STATUS2) + << DP_INTERRUPT_STATUS_ACK_SHIFT; + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS2, + intr_ack | DP_INTERRUPT_STATUS2_MASK); + + return intr; +} + +void msm_dp_ctrl_enable_irq(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl = + container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS, + DP_INTERRUPT_STATUS1_MASK); + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS2, + DP_INTERRUPT_STATUS2_MASK); +} + +void msm_dp_ctrl_disable_irq(struct msm_dp_ctrl *msm_dp_ctrl) +{ + struct msm_dp_ctrl_private *ctrl = + container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS, 0x00); + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS2, 0x00); +} + +static u32 msm_dp_ctrl_get_psr_interrupt(struct msm_dp_ctrl_private *ctrl) +{ + u32 intr, intr_ack; + + intr = msm_dp_read_ahb(ctrl, REG_DP_INTR_STATUS4); + intr_ack = (intr & DP_INTERRUPT_STATUS4) + << DP_INTERRUPT_STATUS_ACK_SHIFT; + msm_dp_write_ahb(ctrl, REG_DP_INTR_STATUS4, intr_ack); + + return intr; +} + +static void msm_dp_ctrl_config_psr_interrupt(struct msm_dp_ctrl_private *ctrl) +{ + msm_dp_write_ahb(ctrl, REG_DP_INTR_MASK4, DP_INTERRUPT_MASK4); +} + +static void msm_dp_ctrl_psr_mainlink_enable(struct msm_dp_ctrl_private *ctrl) +{ + u32 val; + + val = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + val |= DP_MAINLINK_CTRL_ENABLE; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, val); +} + +static void msm_dp_ctrl_psr_mainlink_disable(struct msm_dp_ctrl_private *ctrl) +{ + u32 val; + + val = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + val &= ~DP_MAINLINK_CTRL_ENABLE; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, val); +} + +static void msm_dp_ctrl_mainlink_enable(struct msm_dp_ctrl_private *ctrl) +{ + u32 mainlink_ctrl; + + drm_dbg_dp(ctrl->drm_dev, "enable\n"); + + mainlink_ctrl = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + + mainlink_ctrl &= ~(DP_MAINLINK_CTRL_RESET | + DP_MAINLINK_CTRL_ENABLE); + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); + + mainlink_ctrl |= DP_MAINLINK_CTRL_RESET; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); + + mainlink_ctrl &= ~DP_MAINLINK_CTRL_RESET; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); + + mainlink_ctrl |= (DP_MAINLINK_CTRL_ENABLE | + DP_MAINLINK_FB_BOUNDARY_SEL); + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); +} + +static void msm_dp_ctrl_mainlink_disable(struct msm_dp_ctrl_private *ctrl) +{ + u32 mainlink_ctrl; + + drm_dbg_dp(ctrl->drm_dev, "disable\n"); + + mainlink_ctrl = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + mainlink_ctrl &= ~DP_MAINLINK_CTRL_ENABLE; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); +} + +static void msm_dp_setup_peripheral_flush(struct msm_dp_ctrl_private *ctrl) +{ + u32 mainlink_ctrl; + + mainlink_ctrl = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + + if (ctrl->hw_revision >= DP_HW_VERSION_1_2) + mainlink_ctrl |= DP_MAINLINK_FLUSH_MODE_SDE_PERIPH_UPDATE; + else + mainlink_ctrl |= DP_MAINLINK_FLUSH_MODE_UPDATE_SDP; + + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, mainlink_ctrl); +} + +static bool msm_dp_ctrl_mainlink_ready(struct msm_dp_ctrl_private *ctrl) +{ + u32 data; + int ret; + + /* Poll for mainlink ready status */ + ret = readl_poll_timeout(ctrl->link_base + REG_DP_MAINLINK_READY, + data, data & DP_MAINLINK_READY_FOR_VIDEO, + POLLING_SLEEP_US, POLLING_TIMEOUT_US); + if (ret < 0) { + DRM_ERROR("mainlink not ready\n"); + return false; + } + + return true; +} + void msm_dp_ctrl_push_idle(struct msm_dp_ctrl *msm_dp_ctrl) { struct msm_dp_ctrl_private *ctrl; @@ -126,7 +371,7 @@ void msm_dp_ctrl_push_idle(struct msm_dp_ctrl *msm_dp_ctrl) ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); reinit_completion(&ctrl->idle_comp); - msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, DP_STATE_CTRL_PUSH_IDLE); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, DP_STATE_CTRL_PUSH_IDLE); if (!wait_for_completion_timeout(&ctrl->idle_comp, IDLE_PATTERN_COMPLETION_TIMEOUT_JIFFIES)) @@ -171,23 +416,50 @@ static void msm_dp_ctrl_config_ctrl(struct msm_dp_ctrl_private *ctrl) if (ctrl->panel->psr_cap.version) config |= DP_CONFIGURATION_CTRL_SEND_VSC; - msm_dp_catalog_ctrl_config_ctrl(ctrl->catalog, config); + drm_dbg_dp(ctrl->drm_dev, "DP_CONFIGURATION_CTRL=0x%x\n", config); + + msm_dp_write_link(ctrl, REG_DP_CONFIGURATION_CTRL, config); +} + +static void msm_dp_ctrl_lane_mapping(struct msm_dp_ctrl_private *ctrl) +{ + u32 ln_0 = 0, ln_1 = 1, ln_2 = 2, ln_3 = 3; /* One-to-One mapping */ + u32 ln_mapping; + + ln_mapping = ln_0 << LANE0_MAPPING_SHIFT; + ln_mapping |= ln_1 << LANE1_MAPPING_SHIFT; + ln_mapping |= ln_2 << LANE2_MAPPING_SHIFT; + ln_mapping |= ln_3 << LANE3_MAPPING_SHIFT; + + msm_dp_write_link(ctrl, REG_DP_LOGICAL2PHYSICAL_LANE_MAPPING, + ln_mapping); } static void msm_dp_ctrl_configure_source_params(struct msm_dp_ctrl_private *ctrl) { - u32 cc, tb; + u32 colorimetry_cfg, test_bits_depth, misc_val; - msm_dp_catalog_ctrl_lane_mapping(ctrl->catalog); - msm_dp_catalog_setup_peripheral_flush(ctrl->catalog); + msm_dp_ctrl_lane_mapping(ctrl); + msm_dp_setup_peripheral_flush(ctrl); msm_dp_ctrl_config_ctrl(ctrl); - tb = msm_dp_link_get_test_bits_depth(ctrl->link, - ctrl->panel->msm_dp_mode.bpp); - cc = msm_dp_link_get_colorimetry_config(ctrl->link); - msm_dp_catalog_ctrl_config_misc(ctrl->catalog, cc, tb); - msm_dp_panel_timing_cfg(ctrl->panel); + test_bits_depth = msm_dp_link_get_test_bits_depth(ctrl->link, ctrl->panel->msm_dp_mode.bpp); + colorimetry_cfg = msm_dp_link_get_colorimetry_config(ctrl->link); + + misc_val = msm_dp_read_link(ctrl, REG_DP_MISC1_MISC0); + + /* clear bpp bits */ + misc_val &= ~(0x07 << DP_MISC0_TEST_BITS_DEPTH_SHIFT); + misc_val |= colorimetry_cfg << DP_MISC0_COLORIMETRY_CFG_SHIFT; + misc_val |= test_bits_depth << DP_MISC0_TEST_BITS_DEPTH_SHIFT; + /* Configure clock to synchronous mode */ + misc_val |= DP_MISC0_SYNCHRONOUS_CLK; + + drm_dbg_dp(ctrl->drm_dev, "misc settings = 0x%x\n", misc_val); + msm_dp_write_link(ctrl, REG_DP_MISC1_MISC0, misc_val); + + msm_dp_panel_timing_cfg(ctrl->panel, ctrl->msm_dp_ctrl.wide_bus_en); } /* @@ -1003,8 +1275,9 @@ static void msm_dp_ctrl_setup_tr_unit(struct msm_dp_ctrl_private *ctrl) pr_debug("dp_tu=0x%x, valid_boundary=0x%x, valid_boundary2=0x%x\n", msm_dp_tu, valid_boundary, valid_boundary2); - msm_dp_catalog_ctrl_update_transfer_unit(ctrl->catalog, - msm_dp_tu, valid_boundary, valid_boundary2); + msm_dp_write_link(ctrl, REG_DP_VALID_BOUNDARY, valid_boundary); + msm_dp_write_link(ctrl, REG_DP_TU, msm_dp_tu); + msm_dp_write_link(ctrl, REG_DP_VALID_BOUNDARY_2, valid_boundary2); } static int msm_dp_ctrl_wait4video_ready(struct msm_dp_ctrl_private *ctrl) @@ -1113,6 +1386,30 @@ static bool msm_dp_ctrl_train_pattern_set(struct msm_dp_ctrl_private *ctrl, return ret == 1; } +static int msm_dp_ctrl_set_pattern_state_bit(struct msm_dp_ctrl_private *ctrl, + u32 state_bit) +{ + int bit, ret; + u32 data; + + bit = BIT(state_bit - 1); + drm_dbg_dp(ctrl->drm_dev, "hw: bit=%d train=%d\n", bit, state_bit); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, bit); + + bit = BIT(state_bit - 1) << DP_MAINLINK_READY_LINK_TRAINING_SHIFT; + + /* Poll for mainlink ready status */ + ret = readx_poll_timeout(readl, ctrl->link_base + REG_DP_MAINLINK_READY, + data, data & bit, + POLLING_SLEEP_US, POLLING_TIMEOUT_US); + if (ret < 0) { + DRM_ERROR("set state_bit for link_train=%d failed\n", state_bit); + return ret; + } + + return 0; +} + static int msm_dp_ctrl_link_train_1(struct msm_dp_ctrl_private *ctrl, int *training_step, enum drm_dp_phy dp_phy) { @@ -1124,11 +1421,11 @@ static int msm_dp_ctrl_link_train_1(struct msm_dp_ctrl_private *ctrl, delay_us = drm_dp_read_clock_recovery_delay(ctrl->aux, ctrl->panel->dpcd, dp_phy, false); - msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0); *training_step = DP_TRAINING_1; - ret = msm_dp_catalog_ctrl_set_pattern_state_bit(ctrl->catalog, 1); + ret = msm_dp_ctrl_set_pattern_state_bit(ctrl, 1); if (ret) return ret; msm_dp_ctrl_train_pattern_set(ctrl, DP_TRAINING_PATTERN_1 | @@ -1242,7 +1539,7 @@ static int msm_dp_ctrl_link_train_2(struct msm_dp_ctrl_private *ctrl, delay_us = drm_dp_read_channel_eq_delay(ctrl->aux, ctrl->panel->dpcd, dp_phy, false); - msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0); *training_step = DP_TRAINING_2; @@ -1257,7 +1554,7 @@ static int msm_dp_ctrl_link_train_2(struct msm_dp_ctrl_private *ctrl, state_ctrl_bit = 2; } - ret = msm_dp_catalog_ctrl_set_pattern_state_bit(ctrl->catalog, state_ctrl_bit); + ret = msm_dp_ctrl_set_pattern_state_bit(ctrl, state_ctrl_bit); if (ret) return ret; @@ -1359,7 +1656,7 @@ static int msm_dp_ctrl_link_train(struct msm_dp_ctrl_private *ctrl, } end: - msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0); return ret; } @@ -1369,7 +1666,7 @@ static int msm_dp_ctrl_setup_main_link(struct msm_dp_ctrl_private *ctrl, { int ret = 0; - msm_dp_catalog_ctrl_mainlink_ctrl(ctrl->catalog, true); + msm_dp_ctrl_mainlink_enable(ctrl); if (ctrl->link->sink_request & DP_TEST_LINK_PHY_TEST_PATTERN) return ret; @@ -1502,33 +1799,55 @@ static int msm_dp_ctrl_enable_mainlink_clocks(struct msm_dp_ctrl_private *ctrl) return ret; } -void msm_dp_ctrl_reset_irq_ctrl(struct msm_dp_ctrl *msm_dp_ctrl, bool enable) +static void msm_dp_ctrl_enable_sdp(struct msm_dp_ctrl_private *ctrl) { - struct msm_dp_ctrl_private *ctrl; + /* trigger sdp */ + msm_dp_write_link(ctrl, MMSS_DP_SDP_CFG3, UPDATE_SDP); + msm_dp_write_link(ctrl, MMSS_DP_SDP_CFG3, 0x0); +} - ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); +static void msm_dp_ctrl_psr_enter(struct msm_dp_ctrl_private *ctrl) +{ + u32 cmd; - msm_dp_catalog_ctrl_reset(ctrl->catalog); + cmd = msm_dp_read_link(ctrl, REG_PSR_CMD); - /* - * all dp controller programmable registers will not - * be reset to default value after DP_SW_RESET - * therefore interrupt mask bits have to be updated - * to enable/disable interrupts - */ - msm_dp_catalog_ctrl_enable_irq(ctrl->catalog, enable); + cmd &= ~(PSR_ENTER | PSR_EXIT); + cmd |= PSR_ENTER; + + msm_dp_ctrl_enable_sdp(ctrl); + msm_dp_write_link(ctrl, REG_PSR_CMD, cmd); +} + +static void msm_dp_ctrl_psr_exit(struct msm_dp_ctrl_private *ctrl) +{ + u32 cmd; + + cmd = msm_dp_read_link(ctrl, REG_PSR_CMD); + + cmd &= ~(PSR_ENTER | PSR_EXIT); + cmd |= PSR_EXIT; + + msm_dp_ctrl_enable_sdp(ctrl); + msm_dp_write_link(ctrl, REG_PSR_CMD, cmd); } void msm_dp_ctrl_config_psr(struct msm_dp_ctrl *msm_dp_ctrl) { - u8 cfg; struct msm_dp_ctrl_private *ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); + u32 cfg; if (!ctrl->panel->psr_cap.version) return; - msm_dp_catalog_ctrl_config_psr(ctrl->catalog); + /* enable PSR1 function */ + cfg = msm_dp_read_link(ctrl, REG_PSR_CONFIG); + cfg |= PSR1_SUPPORTED; + msm_dp_write_link(ctrl, REG_PSR_CONFIG, cfg); + + msm_dp_ctrl_config_psr_interrupt(ctrl); + msm_dp_ctrl_enable_sdp(ctrl); cfg = DP_PSR_ENABLE; drm_dp_dpcd_write(ctrl->aux, DP_PSR_EN_CFG, &cfg, 1); @@ -1554,29 +1873,37 @@ void msm_dp_ctrl_set_psr(struct msm_dp_ctrl *msm_dp_ctrl, bool enter) */ if (enter) { reinit_completion(&ctrl->psr_op_comp); - msm_dp_catalog_ctrl_set_psr(ctrl->catalog, true); + msm_dp_ctrl_psr_enter(ctrl); if (!wait_for_completion_timeout(&ctrl->psr_op_comp, PSR_OPERATION_COMPLETION_TIMEOUT_JIFFIES)) { DRM_ERROR("PSR_ENTRY timedout\n"); - msm_dp_catalog_ctrl_set_psr(ctrl->catalog, false); + msm_dp_ctrl_psr_exit(ctrl); return; } msm_dp_ctrl_push_idle(msm_dp_ctrl); - msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0); - msm_dp_catalog_ctrl_psr_mainlink_enable(ctrl->catalog, false); + msm_dp_ctrl_psr_mainlink_disable(ctrl); } else { - msm_dp_catalog_ctrl_psr_mainlink_enable(ctrl->catalog, true); + msm_dp_ctrl_psr_mainlink_enable(ctrl); - msm_dp_catalog_ctrl_set_psr(ctrl->catalog, false); - msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, DP_STATE_CTRL_SEND_VIDEO); + msm_dp_ctrl_psr_exit(ctrl); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, DP_STATE_CTRL_SEND_VIDEO); msm_dp_ctrl_wait4video_ready(ctrl); - msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, 0); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0); } } +static void msm_dp_ctrl_phy_reset(struct msm_dp_ctrl_private *ctrl) +{ + msm_dp_write_ahb(ctrl, REG_DP_PHY_CTRL, + DP_PHY_CTRL_SW_RESET | DP_PHY_CTRL_SW_RESET_PLL); + usleep_range(1000, 1100); /* h/w recommended delay */ + msm_dp_write_ahb(ctrl, REG_DP_PHY_CTRL, 0x0); +} + void msm_dp_ctrl_phy_init(struct msm_dp_ctrl *msm_dp_ctrl) { struct msm_dp_ctrl_private *ctrl; @@ -1585,7 +1912,7 @@ void msm_dp_ctrl_phy_init(struct msm_dp_ctrl *msm_dp_ctrl) ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); phy = ctrl->phy; - msm_dp_catalog_ctrl_phy_reset(ctrl->catalog); + msm_dp_ctrl_phy_reset(ctrl); phy_init(phy); drm_dbg_dp(ctrl->drm_dev, "phy=%p init=%d power_on=%d\n", @@ -1600,7 +1927,7 @@ void msm_dp_ctrl_phy_exit(struct msm_dp_ctrl *msm_dp_ctrl) ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); phy = ctrl->phy; - msm_dp_catalog_ctrl_phy_reset(ctrl->catalog); + msm_dp_ctrl_phy_reset(ctrl); phy_exit(phy); drm_dbg_dp(ctrl->drm_dev, "phy=%p init=%d power_on=%d\n", phy, phy->init_count, phy->power_count); @@ -1611,7 +1938,7 @@ static int msm_dp_ctrl_reinitialize_mainlink(struct msm_dp_ctrl_private *ctrl) struct phy *phy = ctrl->phy; int ret = 0; - msm_dp_catalog_ctrl_mainlink_ctrl(ctrl->catalog, false); + msm_dp_ctrl_mainlink_disable(ctrl); ctrl->phy_opts.dp.lanes = ctrl->link->link_params.num_lanes; phy_configure(phy, &ctrl->phy_opts); /* @@ -1642,9 +1969,9 @@ static int msm_dp_ctrl_deinitialize_mainlink(struct msm_dp_ctrl_private *ctrl) phy = ctrl->phy; - msm_dp_catalog_ctrl_mainlink_ctrl(ctrl->catalog, false); + msm_dp_ctrl_mainlink_disable(ctrl); - msm_dp_catalog_ctrl_reset(ctrl->catalog); + msm_dp_ctrl_reset(&ctrl->msm_dp_ctrl); dev_pm_opp_set_rate(ctrl->dev, 0); msm_dp_ctrl_link_clk_disable(&ctrl->msm_dp_ctrl); @@ -1676,13 +2003,96 @@ static int msm_dp_ctrl_link_maintenance(struct msm_dp_ctrl_private *ctrl) msm_dp_ctrl_clear_training_pattern(ctrl, DP_PHY_DPRX); - msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, DP_STATE_CTRL_SEND_VIDEO); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, DP_STATE_CTRL_SEND_VIDEO); ret = msm_dp_ctrl_wait4video_ready(ctrl); end: return ret; } +#define SCRAMBLER_RESET_COUNT_VALUE 0xFC + +static void msm_dp_ctrl_send_phy_pattern(struct msm_dp_ctrl_private *ctrl, + u32 pattern) +{ + u32 value = 0x0; + + /* Make sure to clear the current pattern before starting a new one */ + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, 0x0); + + drm_dbg_dp(ctrl->drm_dev, "pattern: %#x\n", pattern); + switch (pattern) { + case DP_PHY_TEST_PATTERN_D10_2: + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_TRAINING_PATTERN1); + break; + + case DP_PHY_TEST_PATTERN_ERROR_COUNT: + value &= ~(1 << 16); + msm_dp_write_link(ctrl, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, + value); + value |= SCRAMBLER_RESET_COUNT_VALUE; + msm_dp_write_link(ctrl, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, + value); + msm_dp_write_link(ctrl, REG_DP_MAINLINK_LEVELS, + DP_MAINLINK_SAFE_TO_EXIT_LEVEL_2); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_SYMBOL_ERR_MEASURE); + break; + + case DP_PHY_TEST_PATTERN_PRBS7: + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_PRBS7); + break; + + case DP_PHY_TEST_PATTERN_80BIT_CUSTOM: + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_TEST_CUSTOM_PATTERN); + /* 00111110000011111000001111100000 */ + msm_dp_write_link(ctrl, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG0, + 0x3E0F83E0); + /* 00001111100000111110000011111000 */ + msm_dp_write_link(ctrl, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG1, + 0x0F83E0F8); + /* 1111100000111110 */ + msm_dp_write_link(ctrl, REG_DP_TEST_80BIT_CUSTOM_PATTERN_REG2, + 0x0000F83E); + break; + + case DP_PHY_TEST_PATTERN_CP2520: + value = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + value &= ~DP_MAINLINK_CTRL_SW_BYPASS_SCRAMBLER; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, value); + + value = DP_HBR2_ERM_PATTERN; + msm_dp_write_link(ctrl, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, + value); + value |= SCRAMBLER_RESET_COUNT_VALUE; + msm_dp_write_link(ctrl, REG_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET, + value); + msm_dp_write_link(ctrl, REG_DP_MAINLINK_LEVELS, + DP_MAINLINK_SAFE_TO_EXIT_LEVEL_2); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_SYMBOL_ERR_MEASURE); + value = msm_dp_read_link(ctrl, REG_DP_MAINLINK_CTRL); + value |= DP_MAINLINK_CTRL_ENABLE; + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, value); + break; + + case DP_PHY_TEST_PATTERN_SEL_MASK: + msm_dp_write_link(ctrl, REG_DP_MAINLINK_CTRL, + DP_MAINLINK_CTRL_ENABLE); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, + DP_STATE_CTRL_LINK_TRAINING_PATTERN4); + break; + + default: + drm_dbg_dp(ctrl->drm_dev, + "No valid test pattern requested: %#x\n", pattern); + break; + } +} + static bool msm_dp_ctrl_send_phy_test_pattern(struct msm_dp_ctrl_private *ctrl) { bool success = false; @@ -1697,11 +2107,11 @@ static bool msm_dp_ctrl_send_phy_test_pattern(struct msm_dp_ctrl_private *ctrl) DRM_ERROR("Failed to set v/p levels\n"); return false; } - msm_dp_catalog_ctrl_send_phy_pattern(ctrl->catalog, pattern_requested); + msm_dp_ctrl_send_phy_pattern(ctrl, pattern_requested); msm_dp_ctrl_update_phy_vx_px(ctrl, DP_PHY_DPRX); msm_dp_link_send_test_response(ctrl->link); - pattern_sent = msm_dp_catalog_ctrl_read_phy_pattern(ctrl->catalog); + pattern_sent = msm_dp_read_link(ctrl, REG_DP_MAINLINK_READY); switch (pattern_sent) { case MR_LINK_TRAINING1: @@ -1898,7 +2308,7 @@ int msm_dp_ctrl_on_link(struct msm_dp_ctrl *msm_dp_ctrl) break; } else if (training_step == DP_TRAINING_1) { /* link train_1 failed */ - if (!msm_dp_catalog_link_is_connected(ctrl->catalog)) + if (!msm_dp_aux_is_link_connected(ctrl->aux)) break; drm_dp_dpcd_read_link_status(ctrl->aux, link_status); @@ -1923,7 +2333,7 @@ int msm_dp_ctrl_on_link(struct msm_dp_ctrl *msm_dp_ctrl) } } else if (training_step == DP_TRAINING_2) { /* link train_2 failed */ - if (!msm_dp_catalog_link_is_connected(ctrl->catalog)) + if (!msm_dp_aux_is_link_connected(ctrl->aux)) break; drm_dp_dpcd_read_link_status(ctrl->aux, link_status); @@ -1980,6 +2390,62 @@ static int msm_dp_ctrl_link_retrain(struct msm_dp_ctrl_private *ctrl) return msm_dp_ctrl_setup_main_link(ctrl, &training_step); } +static void msm_dp_ctrl_config_msa(struct msm_dp_ctrl_private *ctrl, + u32 rate, u32 stream_rate_khz, + bool is_ycbcr_420) +{ + u32 pixel_m, pixel_n; + u32 mvid, nvid, pixel_div = 0, dispcc_input_rate; + u32 const nvid_fixed = DP_LINK_CONSTANT_N_VALUE; + u32 const link_rate_hbr2 = 540000; + u32 const link_rate_hbr3 = 810000; + unsigned long den, num; + + if (rate == link_rate_hbr3) + pixel_div = 6; + else if (rate == 162000 || rate == 270000) + pixel_div = 2; + else if (rate == link_rate_hbr2) + pixel_div = 4; + else + DRM_ERROR("Invalid pixel mux divider\n"); + + dispcc_input_rate = (rate * 10) / pixel_div; + + rational_best_approximation(dispcc_input_rate, stream_rate_khz, + (unsigned long)(1 << 16) - 1, + (unsigned long)(1 << 16) - 1, &den, &num); + + den = ~(den - num); + den = den & 0xFFFF; + pixel_m = num; + pixel_n = den; + + mvid = (pixel_m & 0xFFFF) * 5; + nvid = (0xFFFF & (~pixel_n)) + (pixel_m & 0xFFFF); + + if (nvid < nvid_fixed) { + u32 temp; + + temp = (nvid_fixed / nvid) * nvid; + mvid = (nvid_fixed / nvid) * mvid; + nvid = temp; + } + + if (is_ycbcr_420) + mvid /= 2; + + if (link_rate_hbr2 == rate) + nvid *= 2; + + if (link_rate_hbr3 == rate) + nvid *= 3; + + drm_dbg_dp(ctrl->drm_dev, "mvid=0x%x, nvid=0x%x\n", mvid, nvid); + msm_dp_write_link(ctrl, REG_DP_SOFTWARE_MVID, mvid); + msm_dp_write_link(ctrl, REG_DP_SOFTWARE_NVID, nvid); +} + int msm_dp_ctrl_on_stream(struct msm_dp_ctrl *msm_dp_ctrl, bool force_link_train) { int ret = 0; @@ -2045,20 +2511,22 @@ int msm_dp_ctrl_on_stream(struct msm_dp_ctrl *msm_dp_ctrl, bool force_link_train msm_dp_ctrl_configure_source_params(ctrl); - msm_dp_catalog_ctrl_config_msa(ctrl->catalog, + msm_dp_ctrl_config_msa(ctrl, ctrl->link->link_params.rate, pixel_rate_orig, ctrl->panel->msm_dp_mode.out_fmt_is_yuv_420); + msm_dp_panel_clear_dsc_dto(ctrl->panel); + msm_dp_ctrl_setup_tr_unit(ctrl); - msm_dp_catalog_ctrl_state_ctrl(ctrl->catalog, DP_STATE_CTRL_SEND_VIDEO); + msm_dp_write_link(ctrl, REG_DP_STATE_CTRL, DP_STATE_CTRL_SEND_VIDEO); ret = msm_dp_ctrl_wait4video_ready(ctrl); if (ret) return ret; - mainlink_ready = msm_dp_catalog_ctrl_mainlink_ready(ctrl->catalog); + mainlink_ready = msm_dp_ctrl_mainlink_ready(ctrl); drm_dbg_dp(ctrl->drm_dev, "mainlink %s\n", mainlink_ready ? "READY" : "NOT READY"); @@ -2074,12 +2542,12 @@ void msm_dp_ctrl_off_link_stream(struct msm_dp_ctrl *msm_dp_ctrl) ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); phy = ctrl->phy; - msm_dp_catalog_panel_disable_vsc_sdp(ctrl->catalog); + msm_dp_panel_disable_vsc_sdp(ctrl->panel); /* set dongle to D3 (power off) mode */ msm_dp_link_psm_config(ctrl->link, &ctrl->panel->link_info, true); - msm_dp_catalog_ctrl_mainlink_ctrl(ctrl->catalog, false); + msm_dp_ctrl_mainlink_disable(ctrl); if (ctrl->stream_clks_on) { clk_disable_unprepare(ctrl->pixel_clk); @@ -2107,7 +2575,7 @@ void msm_dp_ctrl_off_link(struct msm_dp_ctrl *msm_dp_ctrl) ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); phy = ctrl->phy; - msm_dp_catalog_ctrl_mainlink_ctrl(ctrl->catalog, false); + msm_dp_ctrl_mainlink_disable(ctrl); dev_pm_opp_set_rate(ctrl->dev, 0); msm_dp_ctrl_link_clk_disable(&ctrl->msm_dp_ctrl); @@ -2129,11 +2597,11 @@ void msm_dp_ctrl_off(struct msm_dp_ctrl *msm_dp_ctrl) ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); phy = ctrl->phy; - msm_dp_catalog_panel_disable_vsc_sdp(ctrl->catalog); + msm_dp_panel_disable_vsc_sdp(ctrl->panel); - msm_dp_catalog_ctrl_mainlink_ctrl(ctrl->catalog, false); + msm_dp_ctrl_mainlink_disable(ctrl); - msm_dp_catalog_ctrl_reset(ctrl->catalog); + msm_dp_ctrl_reset(&ctrl->msm_dp_ctrl); if (ctrl->stream_clks_on) { clk_disable_unprepare(ctrl->pixel_clk); @@ -2160,7 +2628,7 @@ irqreturn_t msm_dp_ctrl_isr(struct msm_dp_ctrl *msm_dp_ctrl) ctrl = container_of(msm_dp_ctrl, struct msm_dp_ctrl_private, msm_dp_ctrl); if (ctrl->panel->psr_cap.version) { - isr = msm_dp_catalog_ctrl_read_psr_interrupt_status(ctrl->catalog); + isr = msm_dp_ctrl_get_psr_interrupt(ctrl); if (isr) complete(&ctrl->psr_op_comp); @@ -2175,8 +2643,7 @@ irqreturn_t msm_dp_ctrl_isr(struct msm_dp_ctrl *msm_dp_ctrl) drm_dbg_dp(ctrl->drm_dev, "PSR frame capture done\n"); } - isr = msm_dp_catalog_ctrl_get_interrupt(ctrl->catalog); - + isr = msm_dp_ctrl_get_interrupt(ctrl); if (isr & DP_CTRL_INTR_READY_FOR_VIDEO) { drm_dbg_dp(ctrl->drm_dev, "dp_video_ready\n"); @@ -2190,6 +2657,11 @@ irqreturn_t msm_dp_ctrl_isr(struct msm_dp_ctrl *msm_dp_ctrl) ret = IRQ_HANDLED; } + /* DP aux isr */ + isr = msm_dp_ctrl_get_aux_interrupt(ctrl); + if (isr) + ret |= msm_dp_aux_isr(ctrl->aux, isr); + return ret; } @@ -2245,14 +2717,14 @@ static int msm_dp_ctrl_clk_init(struct msm_dp_ctrl *msm_dp_ctrl) struct msm_dp_ctrl *msm_dp_ctrl_get(struct device *dev, struct msm_dp_link *link, struct msm_dp_panel *panel, struct drm_dp_aux *aux, - struct msm_dp_catalog *catalog, - struct phy *phy) + struct phy *phy, + void __iomem *ahb_base, + void __iomem *link_base) { struct msm_dp_ctrl_private *ctrl; int ret; - if (!dev || !panel || !aux || - !link || !catalog) { + if (!dev || !panel || !aux || !link) { DRM_ERROR("invalid input\n"); return ERR_PTR(-EINVAL); } @@ -2283,9 +2755,10 @@ struct msm_dp_ctrl *msm_dp_ctrl_get(struct device *dev, struct msm_dp_link *link ctrl->panel = panel; ctrl->aux = aux; ctrl->link = link; - ctrl->catalog = catalog; ctrl->dev = dev; ctrl->phy = phy; + ctrl->ahb_base = ahb_base; + ctrl->link_base = link_base; ret = msm_dp_ctrl_clk_init(&ctrl->msm_dp_ctrl); if (ret) { diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.h b/drivers/gpu/drm/msm/dp/dp_ctrl.h index b7abfedbf574..124b9b21bb7f 100644 --- a/drivers/gpu/drm/msm/dp/dp_ctrl.h +++ b/drivers/gpu/drm/msm/dp/dp_ctrl.h @@ -9,7 +9,6 @@ #include "dp_aux.h" #include "dp_panel.h" #include "dp_link.h" -#include "dp_catalog.h" struct msm_dp_ctrl { bool wide_bus_en; @@ -25,12 +24,15 @@ void msm_dp_ctrl_off(struct msm_dp_ctrl *msm_dp_ctrl); void msm_dp_ctrl_push_idle(struct msm_dp_ctrl *msm_dp_ctrl); irqreturn_t msm_dp_ctrl_isr(struct msm_dp_ctrl *msm_dp_ctrl); void msm_dp_ctrl_handle_sink_request(struct msm_dp_ctrl *msm_dp_ctrl); -struct msm_dp_ctrl *msm_dp_ctrl_get(struct device *dev, struct msm_dp_link *link, - struct msm_dp_panel *panel, struct drm_dp_aux *aux, - struct msm_dp_catalog *catalog, - struct phy *phy); - -void msm_dp_ctrl_reset_irq_ctrl(struct msm_dp_ctrl *msm_dp_ctrl, bool enable); +struct msm_dp_ctrl *msm_dp_ctrl_get(struct device *dev, + struct msm_dp_link *link, + struct msm_dp_panel *panel, + struct drm_dp_aux *aux, + struct phy *phy, + void __iomem *ahb_base, + void __iomem *link_base); + +void msm_dp_ctrl_reset(struct msm_dp_ctrl *msm_dp_ctrl); void msm_dp_ctrl_phy_init(struct msm_dp_ctrl *msm_dp_ctrl); void msm_dp_ctrl_phy_exit(struct msm_dp_ctrl *msm_dp_ctrl); void msm_dp_ctrl_irq_phy_exit(struct msm_dp_ctrl *msm_dp_ctrl); @@ -41,4 +43,7 @@ void msm_dp_ctrl_config_psr(struct msm_dp_ctrl *msm_dp_ctrl); int msm_dp_ctrl_core_clk_enable(struct msm_dp_ctrl *msm_dp_ctrl); void msm_dp_ctrl_core_clk_disable(struct msm_dp_ctrl *msm_dp_ctrl); +void msm_dp_ctrl_enable_irq(struct msm_dp_ctrl *msm_dp_ctrl); +void msm_dp_ctrl_disable_irq(struct msm_dp_ctrl *msm_dp_ctrl); + #endif /* _DP_CTRL_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_debug.c b/drivers/gpu/drm/msm/dp/dp_debug.c index 22fd946ee201..cf3838fcd154 100644 --- a/drivers/gpu/drm/msm/dp/dp_debug.c +++ b/drivers/gpu/drm/msm/dp/dp_debug.c @@ -5,11 +5,12 @@ #define pr_fmt(fmt)"[drm-dp] %s: " fmt, __func__ +#ifdef CONFIG_DEBUG_FS + #include <linux/debugfs.h> #include <drm/drm_connector.h> #include <drm/drm_file.h> -#include "dp_catalog.h" #include "dp_aux.h" #include "dp_ctrl.h" #include "dp_debug.h" @@ -235,3 +236,5 @@ int msm_dp_debug_init(struct device *dev, struct msm_dp_panel *panel, return 0; } + +#endif diff --git a/drivers/gpu/drm/msm/dp/dp_display.c b/drivers/gpu/drm/msm/dp/dp_display.c index 386c4669c831..d87d47cc7ec3 100644 --- a/drivers/gpu/drm/msm/dp/dp_display.c +++ b/drivers/gpu/drm/msm/dp/dp_display.c @@ -19,7 +19,6 @@ #include "msm_drv.h" #include "msm_kms.h" #include "dp_ctrl.h" -#include "dp_catalog.h" #include "dp_aux.h" #include "dp_reg.h" #include "dp_link.h" @@ -87,7 +86,6 @@ struct msm_dp_display_private { struct drm_device *drm_dev; - struct msm_dp_catalog *catalog; struct drm_dp_aux *aux; struct msm_dp_link *link; struct msm_dp_panel *panel; @@ -112,6 +110,18 @@ struct msm_dp_display_private { bool wide_bus_supported; struct msm_dp_audio *audio; + + void __iomem *ahb_base; + size_t ahb_len; + + void __iomem *aux_base; + size_t aux_len; + + void __iomem *link_base; + size_t link_len; + + void __iomem *p0_base; + size_t p0_len; }; struct msm_dp_desc { @@ -128,6 +138,11 @@ static const struct msm_dp_desc msm_dp_desc_sa8775p[] = { {} }; +static const struct msm_dp_desc msm_dp_desc_sdm845[] = { + { .io_start = 0x0ae90000, .id = MSM_DP_CONTROLLER_0 }, + {} +}; + static const struct msm_dp_desc msm_dp_desc_sc7180[] = { { .io_start = 0x0ae90000, .id = MSM_DP_CONTROLLER_0, .wide_bus_supported = true }, {} @@ -180,7 +195,7 @@ static const struct of_device_id msm_dp_dt_match[] = { { .compatible = "qcom,sc8180x-edp", .data = &msm_dp_desc_sc8180x }, { .compatible = "qcom,sc8280xp-dp", .data = &msm_dp_desc_sc8280xp }, { .compatible = "qcom,sc8280xp-edp", .data = &msm_dp_desc_sc8280xp }, - { .compatible = "qcom,sdm845-dp", .data = &msm_dp_desc_sc7180 }, + { .compatible = "qcom,sdm845-dp", .data = &msm_dp_desc_sdm845 }, { .compatible = "qcom,sm8350-dp", .data = &msm_dp_desc_sc7180 }, { .compatible = "qcom,sm8650-dp", .data = &msm_dp_desc_sm8650 }, { .compatible = "qcom,x1e80100-dp", .data = &msm_dp_desc_x1e80100 }, @@ -277,9 +292,7 @@ static int msm_dp_display_bind(struct device *dev, struct device *master, struct drm_device *drm = priv->dev; dp->msm_dp_display.drm_dev = drm; - priv->dp[dp->id] = &dp->msm_dp_display; - - + priv->kms->dp[dp->id] = &dp->msm_dp_display; dp->drm_dev = drm; dp->aux->drm_dev = drm; @@ -313,7 +326,7 @@ static void msm_dp_display_unbind(struct device *dev, struct device *master, msm_dp_aux_unregister(dp->aux); dp->drm_dev = NULL; dp->aux->drm_dev = NULL; - priv->dp[dp->id] = NULL; + priv->kms->dp[dp->id] = NULL; } static const struct component_ops msm_dp_display_comp_ops = { @@ -457,7 +470,8 @@ static void msm_dp_display_host_init(struct msm_dp_display_private *dp) dp->phy_initialized); msm_dp_ctrl_core_clk_enable(dp->ctrl); - msm_dp_ctrl_reset_irq_ctrl(dp->ctrl, true); + msm_dp_ctrl_reset(dp->ctrl); + msm_dp_ctrl_enable_irq(dp->ctrl); msm_dp_aux_init(dp->aux); dp->core_initialized = true; } @@ -468,7 +482,8 @@ static void msm_dp_display_host_deinit(struct msm_dp_display_private *dp) dp->msm_dp_display.connector_type, dp->core_initialized, dp->phy_initialized); - msm_dp_ctrl_reset_irq_ctrl(dp->ctrl, false); + msm_dp_ctrl_reset(dp->ctrl); + msm_dp_ctrl_disable_irq(dp->ctrl); msm_dp_aux_deinit(dp->aux); msm_dp_ctrl_core_clk_disable(dp->ctrl); dp->core_initialized = false; @@ -749,21 +764,10 @@ static int msm_dp_init_sub_modules(struct msm_dp_display_private *dp) dp->msm_dp_display.is_edp ? PHY_SUBMODE_EDP : PHY_SUBMODE_DP); if (rc) { DRM_ERROR("failed to set phy submode, rc = %d\n", rc); - dp->catalog = NULL; - goto error; - } - - dp->catalog = msm_dp_catalog_get(dev); - if (IS_ERR(dp->catalog)) { - rc = PTR_ERR(dp->catalog); - DRM_ERROR("failed to initialize catalog, rc = %d\n", rc); - dp->catalog = NULL; goto error; } - dp->aux = msm_dp_aux_get(dev, dp->catalog, - phy, - dp->msm_dp_display.is_edp); + dp->aux = msm_dp_aux_get(dev, phy, dp->msm_dp_display.is_edp, dp->aux_base); if (IS_ERR(dp->aux)) { rc = PTR_ERR(dp->aux); DRM_ERROR("failed to initialize aux, rc = %d\n", rc); @@ -779,7 +783,7 @@ static int msm_dp_init_sub_modules(struct msm_dp_display_private *dp) goto error_link; } - dp->panel = msm_dp_panel_get(dev, dp->aux, dp->link, dp->catalog); + dp->panel = msm_dp_panel_get(dev, dp->aux, dp->link, dp->link_base, dp->p0_base); if (IS_ERR(dp->panel)) { rc = PTR_ERR(dp->panel); DRM_ERROR("failed to initialize panel, rc = %d\n", rc); @@ -788,8 +792,7 @@ static int msm_dp_init_sub_modules(struct msm_dp_display_private *dp) } dp->ctrl = msm_dp_ctrl_get(dev, dp->link, dp->panel, dp->aux, - dp->catalog, - phy); + phy, dp->ahb_base, dp->link_base); if (IS_ERR(dp->ctrl)) { rc = PTR_ERR(dp->ctrl); DRM_ERROR("failed to initialize ctrl, rc = %d\n", rc); @@ -797,7 +800,7 @@ static int msm_dp_init_sub_modules(struct msm_dp_display_private *dp) goto error_ctrl; } - dp->audio = msm_dp_audio_get(dp->msm_dp_display.pdev, dp->catalog); + dp->audio = msm_dp_audio_get(dp->msm_dp_display.pdev, dp->link_base); if (IS_ERR(dp->audio)) { rc = PTR_ERR(dp->audio); pr_err("failed to initialize audio, rc = %d\n", rc); @@ -1020,7 +1023,14 @@ void msm_dp_snapshot(struct msm_disp_state *disp_state, struct msm_dp *dp) return; } - msm_dp_catalog_snapshot(msm_dp_display->catalog, disp_state); + msm_disp_snapshot_add_block(disp_state, msm_dp_display->ahb_len, + msm_dp_display->ahb_base, "dp_ahb"); + msm_disp_snapshot_add_block(disp_state, msm_dp_display->aux_len, + msm_dp_display->aux_base, "dp_aux"); + msm_disp_snapshot_add_block(disp_state, msm_dp_display->link_len, + msm_dp_display->link_base, "dp_link"); + msm_disp_snapshot_add_block(disp_state, msm_dp_display->p0_len, + msm_dp_display->p0_base, "dp_p0"); mutex_unlock(&msm_dp_display->event_mutex); } @@ -1143,7 +1153,7 @@ static irqreturn_t msm_dp_display_irq_handler(int irq, void *dev_id) return IRQ_NONE; } - hpd_isr_status = msm_dp_catalog_hpd_get_intr_status(dp->catalog); + hpd_isr_status = msm_dp_aux_get_hpd_intr_status(dp->aux); if (hpd_isr_status & 0x0F) { drm_dbg_dp(dp->drm_dev, "type=%d isr=0x%x\n", @@ -1170,9 +1180,6 @@ static irqreturn_t msm_dp_display_irq_handler(int irq, void *dev_id) /* DP controller isr */ ret |= msm_dp_ctrl_isr(dp->ctrl); - /* DP aux isr */ - ret |= msm_dp_aux_isr(dp->aux); - return ret; } @@ -1270,6 +1277,80 @@ static int msm_dp_display_get_connector_type(struct platform_device *pdev, return connector_type; } +static void __iomem *msm_dp_ioremap(struct platform_device *pdev, int idx, size_t *len) +{ + struct resource *res; + void __iomem *base; + + base = devm_platform_get_and_ioremap_resource(pdev, idx, &res); + if (!IS_ERR(base)) + *len = resource_size(res); + + return base; +} + +#define DP_DEFAULT_AHB_OFFSET 0x0000 +#define DP_DEFAULT_AHB_SIZE 0x0200 +#define DP_DEFAULT_AUX_OFFSET 0x0200 +#define DP_DEFAULT_AUX_SIZE 0x0200 +#define DP_DEFAULT_LINK_OFFSET 0x0400 +#define DP_DEFAULT_LINK_SIZE 0x0C00 +#define DP_DEFAULT_P0_OFFSET 0x1000 +#define DP_DEFAULT_P0_SIZE 0x0400 + +static int msm_dp_display_get_io(struct msm_dp_display_private *display) +{ + struct platform_device *pdev = display->msm_dp_display.pdev; + + display->ahb_base = msm_dp_ioremap(pdev, 0, &display->ahb_len); + if (IS_ERR(display->ahb_base)) + return PTR_ERR(display->ahb_base); + + display->aux_base = msm_dp_ioremap(pdev, 1, &display->aux_len); + if (IS_ERR(display->aux_base)) { + if (display->aux_base != ERR_PTR(-EINVAL)) { + DRM_ERROR("unable to remap aux region: %pe\n", display->aux_base); + return PTR_ERR(display->aux_base); + } + + /* + * The initial binding had a single reg, but in order to + * support variation in the sub-region sizes this was split. + * msm_dp_ioremap() will fail with -EINVAL here if only a single + * reg is specified, so fill in the sub-region offsets and + * lengths based on this single region. + */ + if (display->ahb_len < DP_DEFAULT_P0_OFFSET + DP_DEFAULT_P0_SIZE) { + DRM_ERROR("legacy memory region not large enough\n"); + return -EINVAL; + } + + display->ahb_len = DP_DEFAULT_AHB_SIZE; + display->aux_base = display->ahb_base + DP_DEFAULT_AUX_OFFSET; + display->aux_len = DP_DEFAULT_AUX_SIZE; + display->link_base = display->ahb_base + DP_DEFAULT_LINK_OFFSET; + display->link_len = DP_DEFAULT_LINK_SIZE; + display->p0_base = display->ahb_base + DP_DEFAULT_P0_OFFSET; + display->p0_len = DP_DEFAULT_P0_SIZE; + + return 0; + } + + display->link_base = msm_dp_ioremap(pdev, 2, &display->link_len); + if (IS_ERR(display->link_base)) { + DRM_ERROR("unable to remap link region: %pe\n", display->link_base); + return PTR_ERR(display->link_base); + } + + display->p0_base = msm_dp_ioremap(pdev, 3, &display->p0_len); + if (IS_ERR(display->p0_base)) { + DRM_ERROR("unable to remap p0 region: %pe\n", display->p0_base); + return PTR_ERR(display->p0_base); + } + + return 0; +} + static int msm_dp_display_probe(struct platform_device *pdev) { int rc = 0; @@ -1296,6 +1377,10 @@ static int msm_dp_display_probe(struct platform_device *pdev) dp->msm_dp_display.is_edp = (dp->msm_dp_display.connector_type == DRM_MODE_CONNECTOR_eDP); + rc = msm_dp_display_get_io(dp); + if (rc) + return rc; + rc = msm_dp_init_sub_modules(dp); if (rc) { DRM_ERROR("init sub module failed\n"); @@ -1358,7 +1443,7 @@ static int msm_dp_pm_runtime_suspend(struct device *dev) if (dp->msm_dp_display.is_edp) { msm_dp_display_host_phy_exit(dp); - msm_dp_catalog_ctrl_hpd_disable(dp->catalog); + msm_dp_aux_hpd_disable(dp->aux); } msm_dp_display_host_deinit(dp); @@ -1379,7 +1464,7 @@ static int msm_dp_pm_runtime_resume(struct device *dev) */ msm_dp_display_host_init(dp); if (dp->msm_dp_display.is_edp) { - msm_dp_catalog_ctrl_hpd_enable(dp->catalog); + msm_dp_aux_hpd_enable(dp->aux); msm_dp_display_host_phy_init(dp); } @@ -1641,8 +1726,6 @@ void msm_dp_bridge_mode_set(struct drm_bridge *drm_bridge, /* populate wide_bus_support to different layers */ msm_dp_display->ctrl->wide_bus_en = msm_dp_display->msm_dp_mode.out_fmt_is_yuv_420 ? false : msm_dp_display->wide_bus_supported; - msm_dp_display->catalog->wide_bus_en = - msm_dp_display->msm_dp_mode.out_fmt_is_yuv_420 ? false : msm_dp_display->wide_bus_supported; } void msm_dp_bridge_hpd_enable(struct drm_bridge *bridge) @@ -1666,10 +1749,8 @@ void msm_dp_bridge_hpd_enable(struct drm_bridge *bridge) return; } - msm_dp_catalog_ctrl_hpd_enable(dp->catalog); - - /* enable HDP interrupts */ - msm_dp_catalog_hpd_config_intr(dp->catalog, DP_DP_HPD_INT_MASK, true); + msm_dp_aux_hpd_enable(dp->aux); + msm_dp_aux_hpd_intr_enable(dp->aux); msm_dp_display->internal_hpd = true; mutex_unlock(&dp->event_mutex); @@ -1682,9 +1763,9 @@ void msm_dp_bridge_hpd_disable(struct drm_bridge *bridge) struct msm_dp_display_private *dp = container_of(msm_dp_display, struct msm_dp_display_private, msm_dp_display); mutex_lock(&dp->event_mutex); - /* disable HDP interrupts */ - msm_dp_catalog_hpd_config_intr(dp->catalog, DP_DP_HPD_INT_MASK, false); - msm_dp_catalog_ctrl_hpd_disable(dp->catalog); + + msm_dp_aux_hpd_intr_disable(dp->aux); + msm_dp_aux_hpd_disable(dp->aux); msm_dp_display->internal_hpd = false; diff --git a/drivers/gpu/drm/msm/dp/dp_link.c b/drivers/gpu/drm/msm/dp/dp_link.c index 92a9077959b3..66e1bbd80db3 100644 --- a/drivers/gpu/drm/msm/dp/dp_link.c +++ b/drivers/gpu/drm/msm/dp/dp_link.c @@ -5,6 +5,7 @@ #define pr_fmt(fmt) "[drm-dp] %s: " fmt, __func__ +#include <drm/drm_device.h> #include <drm/drm_print.h> #include "dp_reg.h" diff --git a/drivers/gpu/drm/msm/dp/dp_panel.c b/drivers/gpu/drm/msm/dp/dp_panel.c index 4e8ab75c771b..15b7f6c7146e 100644 --- a/drivers/gpu/drm/msm/dp/dp_panel.c +++ b/drivers/gpu/drm/msm/dp/dp_panel.c @@ -4,6 +4,7 @@ */ #include "dp_panel.h" +#include "dp_reg.h" #include "dp_utils.h" #include <drm/drm_connector.h> @@ -11,6 +12,10 @@ #include <drm/drm_of.h> #include <drm/drm_print.h> +#include <linux/io.h> + +#define DP_INTF_CONFIG_DATABUS_WIDEN BIT(4) + #define DP_MAX_NUM_DP_LANES 4 #define DP_LINK_RATE_HBR2 540000 /* kbytes */ @@ -20,10 +25,46 @@ struct msm_dp_panel_private { struct msm_dp_panel msm_dp_panel; struct drm_dp_aux *aux; struct msm_dp_link *link; - struct msm_dp_catalog *catalog; + void __iomem *link_base; + void __iomem *p0_base; bool panel_on; }; +static inline u32 msm_dp_read_link(struct msm_dp_panel_private *panel, u32 offset) +{ + return readl_relaxed(panel->link_base + offset); +} + +static inline void msm_dp_write_link(struct msm_dp_panel_private *panel, + u32 offset, u32 data) +{ + /* + * To make sure link reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, panel->link_base + offset); +} + +static inline void msm_dp_write_p0(struct msm_dp_panel_private *panel, + u32 offset, u32 data) +{ + /* + * To make sure interface reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + writel(data, panel->p0_base + offset); +} + +static inline u32 msm_dp_read_p0(struct msm_dp_panel_private *panel, + u32 offset) +{ + /* + * To make sure interface reg writes happens before any other operation, + * this function uses writel() instread of writel_relaxed() + */ + return readl_relaxed(panel->p0_base + offset); +} + static void msm_dp_panel_read_psr_cap(struct msm_dp_panel_private *panel) { ssize_t rlen; @@ -172,7 +213,7 @@ int msm_dp_panel_read_sink_caps(struct msm_dp_panel *msm_dp_panel, if (!msm_dp_panel->drm_edid) { DRM_ERROR("panel edid read failed\n"); /* check edid read fail is due to unplug */ - if (!msm_dp_catalog_link_is_connected(panel->catalog)) { + if (!msm_dp_aux_is_link_connected(panel->aux)) { rc = -ETIMEDOUT; goto end; } @@ -252,9 +293,85 @@ void msm_dp_panel_handle_sink_request(struct msm_dp_panel *msm_dp_panel) } } +static void msm_dp_panel_tpg_enable(struct msm_dp_panel *msm_dp_panel, + struct drm_display_mode *drm_mode) +{ + struct msm_dp_panel_private *panel = + container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + u32 hsync_period, vsync_period; + u32 display_v_start, display_v_end; + u32 hsync_start_x, hsync_end_x; + u32 v_sync_width; + u32 hsync_ctl; + u32 display_hctl; + + /* TPG config parameters*/ + hsync_period = drm_mode->htotal; + vsync_period = drm_mode->vtotal; + + display_v_start = ((drm_mode->vtotal - drm_mode->vsync_start) * + hsync_period); + display_v_end = ((vsync_period - (drm_mode->vsync_start - + drm_mode->vdisplay)) + * hsync_period) - 1; + + display_v_start += drm_mode->htotal - drm_mode->hsync_start; + display_v_end -= (drm_mode->hsync_start - drm_mode->hdisplay); + + hsync_start_x = drm_mode->htotal - drm_mode->hsync_start; + hsync_end_x = hsync_period - (drm_mode->hsync_start - + drm_mode->hdisplay) - 1; + + v_sync_width = drm_mode->vsync_end - drm_mode->vsync_start; + + hsync_ctl = (hsync_period << 16) | + (drm_mode->hsync_end - drm_mode->hsync_start); + display_hctl = (hsync_end_x << 16) | hsync_start_x; + + + msm_dp_write_p0(panel, MMSS_DP_INTF_HSYNC_CTL, hsync_ctl); + msm_dp_write_p0(panel, MMSS_DP_INTF_VSYNC_PERIOD_F0, vsync_period * + hsync_period); + msm_dp_write_p0(panel, MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F0, v_sync_width * + hsync_period); + msm_dp_write_p0(panel, MMSS_DP_INTF_VSYNC_PERIOD_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_VSYNC_PULSE_WIDTH_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_DISPLAY_HCTL, display_hctl); + msm_dp_write_p0(panel, MMSS_DP_INTF_ACTIVE_HCTL, 0); + msm_dp_write_p0(panel, MMSS_INTF_DISPLAY_V_START_F0, display_v_start); + msm_dp_write_p0(panel, MMSS_DP_INTF_DISPLAY_V_END_F0, display_v_end); + msm_dp_write_p0(panel, MMSS_INTF_DISPLAY_V_START_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_DISPLAY_V_END_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_ACTIVE_V_START_F0, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_ACTIVE_V_END_F0, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_ACTIVE_V_START_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_ACTIVE_V_END_F1, 0); + msm_dp_write_p0(panel, MMSS_DP_INTF_POLARITY_CTL, 0); + + msm_dp_write_p0(panel, MMSS_DP_TPG_MAIN_CONTROL, + DP_TPG_CHECKERED_RECT_PATTERN); + msm_dp_write_p0(panel, MMSS_DP_TPG_VIDEO_CONFIG, + DP_TPG_VIDEO_CONFIG_BPP_8BIT | + DP_TPG_VIDEO_CONFIG_RGB); + msm_dp_write_p0(panel, MMSS_DP_BIST_ENABLE, + DP_BIST_ENABLE_DPBIST_EN); + msm_dp_write_p0(panel, MMSS_DP_TIMING_ENGINE_EN, + DP_TIMING_ENGINE_EN_EN); + drm_dbg_dp(panel->drm_dev, "%s: enabled tpg\n", __func__); +} + +static void msm_dp_panel_tpg_disable(struct msm_dp_panel *msm_dp_panel) +{ + struct msm_dp_panel_private *panel = + container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + + msm_dp_write_p0(panel, MMSS_DP_TPG_MAIN_CONTROL, 0x0); + msm_dp_write_p0(panel, MMSS_DP_BIST_ENABLE, 0x0); + msm_dp_write_p0(panel, MMSS_DP_TIMING_ENGINE_EN, 0x0); +} + void msm_dp_panel_tpg_config(struct msm_dp_panel *msm_dp_panel, bool enable) { - struct msm_dp_catalog *catalog; struct msm_dp_panel_private *panel; if (!msm_dp_panel) { @@ -263,7 +380,6 @@ void msm_dp_panel_tpg_config(struct msm_dp_panel *msm_dp_panel, bool enable) } panel = container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); - catalog = panel->catalog; if (!panel->panel_on) { drm_dbg_dp(panel->drm_dev, @@ -272,18 +388,109 @@ void msm_dp_panel_tpg_config(struct msm_dp_panel *msm_dp_panel, bool enable) } if (!enable) { - msm_dp_catalog_panel_tpg_disable(catalog); + msm_dp_panel_tpg_disable(msm_dp_panel); return; } - drm_dbg_dp(panel->drm_dev, "calling catalog tpg_enable\n"); - msm_dp_catalog_panel_tpg_enable(catalog, &panel->msm_dp_panel.msm_dp_mode.drm_mode); + drm_dbg_dp(panel->drm_dev, "calling panel's tpg_enable\n"); + msm_dp_panel_tpg_enable(msm_dp_panel, &panel->msm_dp_panel.msm_dp_mode.drm_mode); +} + +void msm_dp_panel_clear_dsc_dto(struct msm_dp_panel *msm_dp_panel) +{ + struct msm_dp_panel_private *panel = + container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + + msm_dp_write_p0(panel, MMSS_DP_DSC_DTO, 0x0); +} + +static void msm_dp_panel_send_vsc_sdp(struct msm_dp_panel_private *panel, struct dp_sdp *vsc_sdp) +{ + u32 header[2]; + u32 val; + int i; + + msm_dp_utils_pack_sdp_header(&vsc_sdp->sdp_header, header); + + msm_dp_write_link(panel, MMSS_DP_GENERIC0_0, header[0]); + msm_dp_write_link(panel, MMSS_DP_GENERIC0_1, header[1]); + + for (i = 0; i < sizeof(vsc_sdp->db); i += 4) { + val = ((vsc_sdp->db[i]) | (vsc_sdp->db[i + 1] << 8) | (vsc_sdp->db[i + 2] << 16) | + (vsc_sdp->db[i + 3] << 24)); + msm_dp_write_link(panel, MMSS_DP_GENERIC0_2 + i, val); + } +} + +static void msm_dp_panel_update_sdp(struct msm_dp_panel_private *panel) +{ + u32 hw_revision = panel->msm_dp_panel.hw_revision; + + if (hw_revision >= DP_HW_VERSION_1_0 && + hw_revision < DP_HW_VERSION_1_2) { + msm_dp_write_link(panel, MMSS_DP_SDP_CFG3, UPDATE_SDP); + msm_dp_write_link(panel, MMSS_DP_SDP_CFG3, 0x0); + } +} + +void msm_dp_panel_enable_vsc_sdp(struct msm_dp_panel *msm_dp_panel, struct dp_sdp *vsc_sdp) +{ + struct msm_dp_panel_private *panel = + container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + u32 cfg, cfg2, misc; + + cfg = msm_dp_read_link(panel, MMSS_DP_SDP_CFG); + cfg2 = msm_dp_read_link(panel, MMSS_DP_SDP_CFG2); + misc = msm_dp_read_link(panel, REG_DP_MISC1_MISC0); + + cfg |= GEN0_SDP_EN; + msm_dp_write_link(panel, MMSS_DP_SDP_CFG, cfg); + + cfg2 |= GENERIC0_SDPSIZE_VALID; + msm_dp_write_link(panel, MMSS_DP_SDP_CFG2, cfg2); + + msm_dp_panel_send_vsc_sdp(panel, vsc_sdp); + + /* indicates presence of VSC (BIT(6) of MISC1) */ + misc |= DP_MISC1_VSC_SDP; + + drm_dbg_dp(panel->drm_dev, "vsc sdp enable=1\n"); + + pr_debug("misc settings = 0x%x\n", misc); + msm_dp_write_link(panel, REG_DP_MISC1_MISC0, misc); + + msm_dp_panel_update_sdp(panel); +} + +void msm_dp_panel_disable_vsc_sdp(struct msm_dp_panel *msm_dp_panel) +{ + struct msm_dp_panel_private *panel = + container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); + u32 cfg, cfg2, misc; + + cfg = msm_dp_read_link(panel, MMSS_DP_SDP_CFG); + cfg2 = msm_dp_read_link(panel, MMSS_DP_SDP_CFG2); + misc = msm_dp_read_link(panel, REG_DP_MISC1_MISC0); + + cfg &= ~GEN0_SDP_EN; + msm_dp_write_link(panel, MMSS_DP_SDP_CFG, cfg); + + cfg2 &= ~GENERIC0_SDPSIZE_VALID; + msm_dp_write_link(panel, MMSS_DP_SDP_CFG2, cfg2); + + /* switch back to MSA */ + misc &= ~DP_MISC1_VSC_SDP; + + drm_dbg_dp(panel->drm_dev, "vsc sdp enable=0\n"); + + pr_debug("misc settings = 0x%x\n", misc); + msm_dp_write_link(panel, REG_DP_MISC1_MISC0, misc); + + msm_dp_panel_update_sdp(panel); } static int msm_dp_panel_setup_vsc_sdp_yuv_420(struct msm_dp_panel *msm_dp_panel) { - struct msm_dp_catalog *catalog; - struct msm_dp_panel_private *panel; struct msm_dp_display_mode *msm_dp_mode; struct drm_dp_vsc_sdp vsc_sdp_data; struct dp_sdp vsc_sdp; @@ -294,8 +501,6 @@ static int msm_dp_panel_setup_vsc_sdp_yuv_420(struct msm_dp_panel *msm_dp_panel) return -EINVAL; } - panel = container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); - catalog = panel->catalog; msm_dp_mode = &msm_dp_panel->msm_dp_mode; memset(&vsc_sdp_data, 0, sizeof(vsc_sdp_data)); @@ -322,24 +527,23 @@ static int msm_dp_panel_setup_vsc_sdp_yuv_420(struct msm_dp_panel *msm_dp_panel) return len; } - msm_dp_catalog_panel_enable_vsc_sdp(catalog, &vsc_sdp); + msm_dp_panel_enable_vsc_sdp(msm_dp_panel, &vsc_sdp); return 0; } -int msm_dp_panel_timing_cfg(struct msm_dp_panel *msm_dp_panel) +int msm_dp_panel_timing_cfg(struct msm_dp_panel *msm_dp_panel, bool wide_bus_en) { u32 data, total_ver, total_hor; - struct msm_dp_catalog *catalog; struct msm_dp_panel_private *panel; struct drm_display_mode *drm_mode; u32 width_blanking; u32 sync_start; u32 msm_dp_active; u32 total; + u32 reg; panel = container_of(msm_dp_panel, struct msm_dp_panel_private, msm_dp_panel); - catalog = panel->catalog; drm_mode = &panel->msm_dp_panel.msm_dp_mode.drm_mode; drm_dbg_dp(panel->drm_dev, "width=%d hporch= %d %d %d\n", @@ -382,7 +586,20 @@ int msm_dp_panel_timing_cfg(struct msm_dp_panel *msm_dp_panel) msm_dp_active = data; - msm_dp_catalog_panel_timing_cfg(catalog, total, sync_start, width_blanking, msm_dp_active); + msm_dp_write_link(panel, REG_DP_TOTAL_HOR_VER, total); + msm_dp_write_link(panel, REG_DP_START_HOR_VER_FROM_SYNC, sync_start); + msm_dp_write_link(panel, REG_DP_HSYNC_VSYNC_WIDTH_POLARITY, width_blanking); + msm_dp_write_link(panel, REG_DP_ACTIVE_HOR_VER, msm_dp_active); + + reg = msm_dp_read_p0(panel, MMSS_DP_INTF_CONFIG); + if (wide_bus_en) + reg |= DP_INTF_CONFIG_DATABUS_WIDEN; + else + reg &= ~DP_INTF_CONFIG_DATABUS_WIDEN; + + drm_dbg_dp(panel->drm_dev, "wide_bus_en=%d reg=%#x\n", wide_bus_en, reg); + + msm_dp_write_p0(panel, MMSS_DP_INTF_CONFIG, reg); if (msm_dp_panel->msm_dp_mode.out_fmt_is_yuv_420) msm_dp_panel_setup_vsc_sdp_yuv_420(msm_dp_panel); @@ -486,13 +703,15 @@ static int msm_dp_panel_parse_dt(struct msm_dp_panel *msm_dp_panel) } struct msm_dp_panel *msm_dp_panel_get(struct device *dev, struct drm_dp_aux *aux, - struct msm_dp_link *link, struct msm_dp_catalog *catalog) + struct msm_dp_link *link, + void __iomem *link_base, + void __iomem *p0_base) { struct msm_dp_panel_private *panel; struct msm_dp_panel *msm_dp_panel; int ret; - if (!dev || !catalog || !aux || !link) { + if (!dev || !aux || !link) { DRM_ERROR("invalid input\n"); return ERR_PTR(-EINVAL); } @@ -503,8 +722,9 @@ struct msm_dp_panel *msm_dp_panel_get(struct device *dev, struct drm_dp_aux *aux panel->dev = dev; panel->aux = aux; - panel->catalog = catalog; panel->link = link; + panel->link_base = link_base; + panel->p0_base = p0_base; msm_dp_panel = &panel->msm_dp_panel; msm_dp_panel->max_bw_code = DP_LINK_BW_8_1; diff --git a/drivers/gpu/drm/msm/dp/dp_panel.h b/drivers/gpu/drm/msm/dp/dp_panel.h index 4906f4f09f24..d2cf401506dc 100644 --- a/drivers/gpu/drm/msm/dp/dp_panel.h +++ b/drivers/gpu/drm/msm/dp/dp_panel.h @@ -6,6 +6,7 @@ #ifndef _DP_PANEL_H_ #define _DP_PANEL_H_ +#include <drm/drm_modes.h> #include <drm/msm_drm.h> #include "dp_aux.h" @@ -38,6 +39,7 @@ struct msm_dp_panel { struct msm_dp_panel_psr psr_cap; bool video_test; bool vsc_sdp_supported; + u32 hw_revision; u32 max_dp_lanes; u32 max_dp_link_rate; @@ -47,7 +49,7 @@ struct msm_dp_panel { int msm_dp_panel_init_panel_info(struct msm_dp_panel *msm_dp_panel); int msm_dp_panel_deinit(struct msm_dp_panel *msm_dp_panel); -int msm_dp_panel_timing_cfg(struct msm_dp_panel *msm_dp_panel); +int msm_dp_panel_timing_cfg(struct msm_dp_panel *msm_dp_panel, bool wide_bus_en); int msm_dp_panel_read_sink_caps(struct msm_dp_panel *msm_dp_panel, struct drm_connector *connector); u32 msm_dp_panel_get_mode_bpp(struct msm_dp_panel *msm_dp_panel, u32 mode_max_bpp, @@ -57,6 +59,11 @@ int msm_dp_panel_get_modes(struct msm_dp_panel *msm_dp_panel, void msm_dp_panel_handle_sink_request(struct msm_dp_panel *msm_dp_panel); void msm_dp_panel_tpg_config(struct msm_dp_panel *msm_dp_panel, bool enable); +void msm_dp_panel_clear_dsc_dto(struct msm_dp_panel *msm_dp_panel); + +void msm_dp_panel_enable_vsc_sdp(struct msm_dp_panel *msm_dp_panel, struct dp_sdp *vsc_sdp); +void msm_dp_panel_disable_vsc_sdp(struct msm_dp_panel *msm_dp_panel); + /** * is_link_rate_valid() - validates the link rate * @lane_rate: link rate requested by the sink @@ -85,6 +92,8 @@ static inline bool is_lane_count_valid(u32 lane_count) } struct msm_dp_panel *msm_dp_panel_get(struct device *dev, struct drm_dp_aux *aux, - struct msm_dp_link *link, struct msm_dp_catalog *catalog); + struct msm_dp_link *link, + void __iomem *link_base, + void __iomem *p0_base); void msm_dp_panel_put(struct msm_dp_panel *msm_dp_panel); #endif /* _DP_PANEL_H_ */ diff --git a/drivers/gpu/drm/msm/dp/dp_reg.h b/drivers/gpu/drm/msm/dp/dp_reg.h index 3835c7f5cb98..7c44d4e2cf13 100644 --- a/drivers/gpu/drm/msm/dp/dp_reg.h +++ b/drivers/gpu/drm/msm/dp/dp_reg.h @@ -11,6 +11,8 @@ /* DP_TX Registers */ #define REG_DP_HW_VERSION (0x00000000) +#define DP_HW_VERSION_1_0 0x10000000 +#define DP_HW_VERSION_1_2 0x10020000 #define REG_DP_SW_RESET (0x00000010) #define DP_SW_RESET (0x00000001) @@ -21,8 +23,25 @@ #define REG_DP_CLK_CTRL (0x00000018) #define REG_DP_CLK_ACTIVE (0x0000001C) + #define REG_DP_INTR_STATUS (0x00000020) +#define DP_INTR_HPD BIT(0) +#define DP_INTR_AUX_XFER_DONE BIT(3) +#define DP_INTR_WRONG_ADDR BIT(6) +#define DP_INTR_TIMEOUT BIT(9) +#define DP_INTR_NACK_DEFER BIT(12) +#define DP_INTR_WRONG_DATA_CNT BIT(15) +#define DP_INTR_I2C_NACK BIT(18) +#define DP_INTR_I2C_DEFER BIT(21) +#define DP_INTR_PLL_UNLOCKED BIT(24) +#define DP_INTR_AUX_ERROR BIT(27) + #define REG_DP_INTR_STATUS2 (0x00000024) +#define DP_INTR_READY_FOR_VIDEO BIT(0) +#define DP_INTR_IDLE_PATTERN_SENT BIT(3) +#define DP_INTR_FRAME_END BIT(6) +#define DP_INTR_CRC_UPDATED BIT(9) + #define REG_DP_INTR_STATUS3 (0x00000028) #define REG_DP_INTR_STATUS4 (0x0000002C) diff --git a/drivers/gpu/drm/msm/dsi/dsi.c b/drivers/gpu/drm/msm/dsi/dsi.c index 296215877613..d8bb40ef820e 100644 --- a/drivers/gpu/drm/msm/dsi/dsi.c +++ b/drivers/gpu/drm/msm/dsi/dsi.c @@ -136,7 +136,7 @@ static int dsi_bind(struct device *dev, struct device *master, void *data) msm_dsi->next_bridge = ext_bridge; } - priv->dsi[msm_dsi->id] = msm_dsi; + priv->kms->dsi[msm_dsi->id] = msm_dsi; return 0; } @@ -148,7 +148,7 @@ static void dsi_unbind(struct device *dev, struct device *master, struct msm_dsi *msm_dsi = dev_get_drvdata(dev); msm_dsi_tx_buf_free(msm_dsi->host); - priv->dsi[msm_dsi->id] = NULL; + priv->kms->dsi[msm_dsi->id] = NULL; } static const struct component_ops dsi_ops = { diff --git a/drivers/gpu/drm/msm/dsi/dsi.h b/drivers/gpu/drm/msm/dsi/dsi.h index 87496db203d6..93c028a122f3 100644 --- a/drivers/gpu/drm/msm/dsi/dsi.h +++ b/drivers/gpu/drm/msm/dsi/dsi.h @@ -98,6 +98,7 @@ int msm_dsi_host_init(struct msm_dsi *msm_dsi); int msm_dsi_runtime_suspend(struct device *dev); int msm_dsi_runtime_resume(struct device *dev); int dsi_link_clk_set_rate_6g(struct msm_dsi_host *msm_host); +int dsi_link_clk_set_rate_6g_v2_9(struct msm_dsi_host *msm_host); int dsi_link_clk_set_rate_v2(struct msm_dsi_host *msm_host); int dsi_link_clk_enable_6g(struct msm_dsi_host *msm_host); int dsi_link_clk_enable_v2(struct msm_dsi_host *msm_host); @@ -115,6 +116,7 @@ int dsi_dma_base_get_6g(struct msm_dsi_host *msm_host, uint64_t *iova); int dsi_dma_base_get_v2(struct msm_dsi_host *msm_host, uint64_t *iova); int dsi_clk_init_v2(struct msm_dsi_host *msm_host); int dsi_clk_init_6g_v2(struct msm_dsi_host *msm_host); +int dsi_clk_init_6g_v2_9(struct msm_dsi_host *msm_host); int dsi_calc_clk_rate_v2(struct msm_dsi_host *msm_host, bool is_bonded_dsi); int dsi_calc_clk_rate_6g(struct msm_dsi_host *msm_host, bool is_bonded_dsi); void msm_dsi_host_snapshot(struct msm_disp_state *disp_state, struct mipi_dsi_host *host); diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c index 7675558ae2e5..fed8e9b67011 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c @@ -273,6 +273,18 @@ static const struct msm_dsi_host_cfg_ops msm_dsi_6g_v2_host_ops = { .calc_clk_rate = dsi_calc_clk_rate_6g, }; +static const struct msm_dsi_host_cfg_ops msm_dsi_6g_v2_9_host_ops = { + .link_clk_set_rate = dsi_link_clk_set_rate_6g_v2_9, + .link_clk_enable = dsi_link_clk_enable_6g, + .link_clk_disable = dsi_link_clk_disable_6g, + .clk_init_ver = dsi_clk_init_6g_v2_9, + .tx_buf_alloc = dsi_tx_buf_alloc_6g, + .tx_buf_get = dsi_tx_buf_get_6g, + .tx_buf_put = dsi_tx_buf_put_6g, + .dma_base_get = dsi_dma_base_get_6g, + .calc_clk_rate = dsi_calc_clk_rate_6g, +}; + static const struct msm_dsi_cfg_handler dsi_cfg_handlers[] = { {MSM_DSI_VER_MAJOR_V2, MSM_DSI_V2_VER_MINOR_8064, &apq8064_dsi_cfg, &msm_dsi_v2_host_ops}, @@ -318,6 +330,8 @@ static const struct msm_dsi_cfg_handler dsi_cfg_handlers[] = { &sm8550_dsi_cfg, &msm_dsi_6g_v2_host_ops}, {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_8_0, &sm8650_dsi_cfg, &msm_dsi_6g_v2_host_ops}, + {MSM_DSI_VER_MAJOR_6G, MSM_DSI_6G_VER_MINOR_V2_9_0, + &sm8650_dsi_cfg, &msm_dsi_6g_v2_9_host_ops}, }; const struct msm_dsi_cfg_handler *msm_dsi_cfg_get(u32 major, u32 minor) diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.h b/drivers/gpu/drm/msm/dsi/dsi_cfg.h index 65b0705fac0e..38f303f2ed04 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_cfg.h +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.h @@ -31,6 +31,7 @@ #define MSM_DSI_6G_VER_MINOR_V2_6_0 0x20060000 #define MSM_DSI_6G_VER_MINOR_V2_7_0 0x20070000 #define MSM_DSI_6G_VER_MINOR_V2_8_0 0x20080000 +#define MSM_DSI_6G_VER_MINOR_V2_9_0 0x20090000 #define MSM_DSI_V2_VER_MINOR_8064 0x0 diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c index 4d75529c0e85..e0de545d4077 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_host.c +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c @@ -119,6 +119,15 @@ struct msm_dsi_host { struct clk *pixel_clk; struct clk *byte_intf_clk; + /* + * Clocks which needs to be properly parented between DISPCC and DSI PHY + * PLL: + */ + struct clk *byte_src_clk; + struct clk *pixel_src_clk; + struct clk *dsi_pll_byte_clk; + struct clk *dsi_pll_pixel_clk; + unsigned long byte_clk_rate; unsigned long byte_intf_clk_rate; unsigned long pixel_clk_rate; @@ -143,7 +152,7 @@ struct msm_dsi_host { /* DSI 6G TX buffer*/ struct drm_gem_object *tx_gem_obj; - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; /* DSI v2 TX buffer */ void *tx_buf; @@ -269,6 +278,38 @@ int dsi_clk_init_6g_v2(struct msm_dsi_host *msm_host) return ret; } +int dsi_clk_init_6g_v2_9(struct msm_dsi_host *msm_host) +{ + struct device *dev = &msm_host->pdev->dev; + int ret; + + ret = dsi_clk_init_6g_v2(msm_host); + if (ret) + return ret; + + msm_host->byte_src_clk = devm_clk_get(dev, "byte_src"); + if (IS_ERR(msm_host->byte_src_clk)) + return dev_err_probe(dev, PTR_ERR(msm_host->byte_src_clk), + "can't get byte_src clock\n"); + + msm_host->dsi_pll_byte_clk = devm_clk_get(dev, "dsi_pll_byte"); + if (IS_ERR(msm_host->dsi_pll_byte_clk)) + return dev_err_probe(dev, PTR_ERR(msm_host->dsi_pll_byte_clk), + "can't get dsi_pll_byte clock\n"); + + msm_host->pixel_src_clk = devm_clk_get(dev, "pixel_src"); + if (IS_ERR(msm_host->pixel_src_clk)) + return dev_err_probe(dev, PTR_ERR(msm_host->pixel_src_clk), + "can't get pixel_src clock\n"); + + msm_host->dsi_pll_pixel_clk = devm_clk_get(dev, "dsi_pll_pixel"); + if (IS_ERR(msm_host->dsi_pll_pixel_clk)) + return dev_err_probe(dev, PTR_ERR(msm_host->dsi_pll_pixel_clk), + "can't get dsi_pll_pixel clock\n"); + + return 0; +} + static int dsi_clk_init(struct msm_dsi_host *msm_host) { struct platform_device *pdev = msm_host->pdev; @@ -370,6 +411,26 @@ int dsi_link_clk_set_rate_6g(struct msm_dsi_host *msm_host) return 0; } +int dsi_link_clk_set_rate_6g_v2_9(struct msm_dsi_host *msm_host) +{ + struct device *dev = &msm_host->pdev->dev; + int ret; + + /* + * DSI PHY PLLs have to be enabled to allow reparenting to them, so + * cannot use assigned-clock-parents. + */ + ret = clk_set_parent(msm_host->byte_src_clk, msm_host->dsi_pll_byte_clk); + if (ret) + dev_err(dev, "Failed to parent byte_src -> dsi_pll_byte: %d\n", ret); + + ret = clk_set_parent(msm_host->pixel_src_clk, msm_host->dsi_pll_pixel_clk); + if (ret) + dev_err(dev, "Failed to parent pixel_src -> dsi_pll_pixel: %d\n", ret); + + return dsi_link_clk_set_rate_6g(msm_host); +} + int dsi_link_clk_enable_6g(struct msm_dsi_host *msm_host) { int ret; @@ -1146,10 +1207,10 @@ int dsi_tx_buf_alloc_6g(struct msm_dsi_host *msm_host, int size) uint64_t iova; u8 *data; - msm_host->aspace = msm_gem_address_space_get(priv->kms->aspace); + msm_host->vm = drm_gpuvm_get(priv->kms->vm); data = msm_gem_kernel_new(dev, size, MSM_BO_WC, - msm_host->aspace, + msm_host->vm, &msm_host->tx_gem_obj, &iova); if (IS_ERR(data)) { @@ -1193,10 +1254,10 @@ void msm_dsi_tx_buf_free(struct mipi_dsi_host *host) return; if (msm_host->tx_gem_obj) { - msm_gem_kernel_put(msm_host->tx_gem_obj, msm_host->aspace); - msm_gem_address_space_put(msm_host->aspace); + msm_gem_kernel_put(msm_host->tx_gem_obj, msm_host->vm); + drm_gpuvm_put(msm_host->vm); msm_host->tx_gem_obj = NULL; - msm_host->aspace = NULL; + msm_host->vm = NULL; } if (msm_host->tx_buf) @@ -1327,7 +1388,7 @@ int dsi_dma_base_get_6g(struct msm_dsi_host *msm_host, uint64_t *dma_base) return -EINVAL; return msm_gem_get_and_pin_iova(msm_host->tx_gem_obj, - priv->kms->aspace, dma_base); + priv->kms->vm, dma_base); } int dsi_dma_base_get_v2(struct msm_dsi_host *msm_host, uint64_t *dma_base) diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c index 5973d7325699..221f12db5f8b 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c @@ -597,6 +597,8 @@ static const struct of_device_id dsi_phy_dt_match[] = { .data = &dsi_phy_4nm_8550_cfgs }, { .compatible = "qcom,sm8650-dsi-phy-4nm", .data = &dsi_phy_4nm_8650_cfgs }, + { .compatible = "qcom,sm8750-dsi-phy-3nm", + .data = &dsi_phy_3nm_8750_cfgs }, #endif {} }; diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h index 7ea608f620fe..c558f8df1684 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h @@ -63,6 +63,7 @@ extern const struct msm_dsi_phy_cfg dsi_phy_5nm_8775p_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_5nm_sar2130p_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_4nm_8550_cfgs; extern const struct msm_dsi_phy_cfg dsi_phy_4nm_8650_cfgs; +extern const struct msm_dsi_phy_cfg dsi_phy_3nm_8750_cfgs; struct msm_dsi_dphy_timing { u32 clk_zero; diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c index 9812b4d69197..af2e30f3f842 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c @@ -704,6 +704,13 @@ static int dsi_pll_10nm_init(struct msm_dsi_phy *phy) /* TODO: Remove this when we have proper display handover support */ msm_dsi_phy_pll_save_state(phy); + /* + * Store also proper vco_current_rate, because its value will be used in + * dsi_10nm_pll_restore_state(). + */ + if (!dsi_pll_10nm_vco_recalc_rate(&pll_10nm->clk_hw, VCO_REF_CLK_RATE)) + pll_10nm->vco_current_rate = pll_10nm->phy->cfg->min_pll_rate; + return 0; } diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c index c19890358b74..8c98f91a5930 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c @@ -51,6 +51,8 @@ #define DSI_PHY_7NM_QUIRK_V4_3 BIT(3) /* Hardware is V5.2 */ #define DSI_PHY_7NM_QUIRK_V5_2 BIT(4) +/* Hardware is V7.0 */ +#define DSI_PHY_7NM_QUIRK_V7_0 BIT(5) struct dsi_pll_config { bool enable_ssc; @@ -129,9 +131,30 @@ static void dsi_pll_calc_dec_frac(struct dsi_pll_7nm *pll, struct dsi_pll_config dec_multiple = div_u64(pll_freq * multiplier, divider); dec = div_u64_rem(dec_multiple, multiplier, &frac); - if (pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_PRE_V4_1) + if (pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_PRE_V4_1) { config->pll_clock_inverters = 0x28; - else if ((pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2)) { + } else if ((pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0)) { + if (pll_freq < 163000000ULL) + config->pll_clock_inverters = 0xa0; + else if (pll_freq < 175000000ULL) + config->pll_clock_inverters = 0x20; + else if (pll_freq < 325000000ULL) + config->pll_clock_inverters = 0xa0; + else if (pll_freq < 350000000ULL) + config->pll_clock_inverters = 0x20; + else if (pll_freq < 650000000ULL) + config->pll_clock_inverters = 0xa0; + else if (pll_freq < 700000000ULL) + config->pll_clock_inverters = 0x20; + else if (pll_freq < 1300000000ULL) + config->pll_clock_inverters = 0xa0; + else if (pll_freq < 2500000000ULL) + config->pll_clock_inverters = 0x20; + else if (pll_freq < 4000000000ULL) + config->pll_clock_inverters = 0x00; + else + config->pll_clock_inverters = 0x40; + } else if ((pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2)) { if (pll_freq <= 1300000000ULL) config->pll_clock_inverters = 0xa0; else if (pll_freq <= 2500000000ULL) @@ -250,7 +273,8 @@ static void dsi_pll_config_hzindep_reg(struct dsi_pll_7nm *pll) vco_config_1 = 0x01; } - if ((pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2)) { + if ((pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2) || + (pll->phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0)) { if (pll->vco_current_rate < 1557000000ULL) vco_config_1 = 0x08; else @@ -620,6 +644,7 @@ static int dsi_7nm_pll_restore_state(struct msm_dsi_phy *phy) static int dsi_7nm_set_usecase(struct msm_dsi_phy *phy) { struct dsi_pll_7nm *pll_7nm = to_pll_7nm(phy->vco_hw); + void __iomem *base = phy->base; u32 data = 0x0; /* internal PLL */ DBG("DSI PLL%d", pll_7nm->phy->id); @@ -629,6 +654,9 @@ static int dsi_7nm_set_usecase(struct msm_dsi_phy *phy) break; case MSM_DSI_PHY_MASTER: pll_7nm->slave = pll_7nm_list[(pll_7nm->phy->id + 1) % DSI_MAX]; + /* v7.0: Enable ATB_EN0 and alternate clock output to external phy */ + if (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0) + writel(0x07, base + REG_DSI_7nm_PHY_CMN_CTRL_5); break; case MSM_DSI_PHY_SLAVE: data = 0x1; /* external PLL */ @@ -907,7 +935,8 @@ static int dsi_7nm_phy_enable(struct msm_dsi_phy *phy, /* Request for REFGEN READY */ if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V4_3) || - (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2)) { + (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2) || + (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0)) { writel(0x1, phy->base + REG_DSI_7nm_PHY_CMN_GLBL_DIGTOP_SPARE10); udelay(500); } @@ -941,7 +970,20 @@ static int dsi_7nm_phy_enable(struct msm_dsi_phy *phy, lane_ctrl0 = 0x1f; } - if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2)) { + if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0)) { + if (phy->cphy_mode) { + /* TODO: different for second phy */ + vreg_ctrl_0 = 0x57; + vreg_ctrl_1 = 0x41; + glbl_rescode_top_ctrl = 0x3d; + glbl_rescode_bot_ctrl = 0x38; + } else { + vreg_ctrl_0 = 0x56; + vreg_ctrl_1 = 0x19; + glbl_rescode_top_ctrl = less_than_1500_mhz ? 0x3c : 0x03; + glbl_rescode_bot_ctrl = less_than_1500_mhz ? 0x38 : 0x3c; + } + } else if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2)) { if (phy->cphy_mode) { vreg_ctrl_0 = 0x45; vreg_ctrl_1 = 0x41; @@ -1003,6 +1045,7 @@ static int dsi_7nm_phy_enable(struct msm_dsi_phy *phy, /* program CMN_CTRL_4 for minor_ver 2 chipsets*/ if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2) || + (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0) || (readl(base + REG_DSI_7nm_PHY_CMN_REVISION_ID0) & (0xf0)) == 0x20) writel(0x04, base + REG_DSI_7nm_PHY_CMN_CTRL_4); @@ -1117,7 +1160,8 @@ static void dsi_7nm_phy_disable(struct msm_dsi_phy *phy) /* Turn off REFGEN Vote */ if ((phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V4_3) || - (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2)) { + (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V5_2) || + (phy->cfg->quirks & DSI_PHY_7NM_QUIRK_V7_0)) { writel(0x0, base + REG_DSI_7nm_PHY_CMN_GLBL_DIGTOP_SPARE10); wmb(); /* Delay to ensure HW removes vote before PHY shut down */ @@ -1384,3 +1428,26 @@ const struct msm_dsi_phy_cfg dsi_phy_4nm_8650_cfgs = { .num_dsi_phy = 2, .quirks = DSI_PHY_7NM_QUIRK_V5_2, }; + +const struct msm_dsi_phy_cfg dsi_phy_3nm_8750_cfgs = { + .has_phy_lane = true, + .regulator_data = dsi_phy_7nm_98000uA_regulators, + .num_regulators = ARRAY_SIZE(dsi_phy_7nm_98000uA_regulators), + .ops = { + .enable = dsi_7nm_phy_enable, + .disable = dsi_7nm_phy_disable, + .pll_init = dsi_pll_7nm_init, + .save_pll_state = dsi_7nm_pll_save_state, + .restore_pll_state = dsi_7nm_pll_restore_state, + .set_continuous_clock = dsi_7nm_set_continuous_clock, + }, + .min_pll_rate = 600000000UL, +#ifdef CONFIG_64BIT + .max_pll_rate = 5000000000UL, +#else + .max_pll_rate = ULONG_MAX, +#endif + .io_start = { 0xae95000, 0xae97000 }, + .num_dsi_phy = 2, + .quirks = DSI_PHY_7NM_QUIRK_V7_0, +}; diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c index 2fd388b892dc..5afac09c0d33 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi.c @@ -15,6 +15,7 @@ #include <drm/drm_of.h> #include <drm/display/drm_hdmi_state_helper.h> +#include "msm_kms.h" #include "hdmi.h" void msm_hdmi_set_mode(struct hdmi *hdmi, bool power_on) @@ -244,7 +245,7 @@ static int msm_hdmi_bind(struct device *dev, struct device *master, void *data) err = msm_hdmi_init(hdmi); if (err) return err; - priv->hdmi = hdmi; + priv->kms->hdmi = hdmi; return 0; } @@ -254,9 +255,9 @@ static void msm_hdmi_unbind(struct device *dev, struct device *master, { struct msm_drm_private *priv = dev_get_drvdata(master); - if (priv->hdmi) { - msm_hdmi_destroy(priv->hdmi); - priv->hdmi = NULL; + if (priv->kms->hdmi) { + msm_hdmi_destroy(priv->kms->hdmi); + priv->kms->hdmi = NULL; } } diff --git a/drivers/gpu/drm/msm/msm_debugfs.c b/drivers/gpu/drm/msm/msm_debugfs.c index 7ab607252d18..bbda865addae 100644 --- a/drivers/gpu/drm/msm/msm_debugfs.c +++ b/drivers/gpu/drm/msm/msm_debugfs.c @@ -117,6 +117,36 @@ static const struct file_operations msm_gpu_fops = { .release = msm_gpu_release, }; +#ifdef CONFIG_DRM_MSM_KMS +static int msm_fb_show(struct seq_file *m, void *arg) +{ + struct drm_info_node *node = m->private; + struct drm_device *dev = node->minor->dev; + struct drm_framebuffer *fb, *fbdev_fb = NULL; + + if (dev->fb_helper && dev->fb_helper->fb) { + seq_puts(m, "fbcon "); + fbdev_fb = dev->fb_helper->fb; + msm_framebuffer_describe(fbdev_fb, m); + } + + mutex_lock(&dev->mode_config.fb_lock); + list_for_each_entry(fb, &dev->mode_config.fb_list, head) { + if (fb == fbdev_fb) + continue; + + seq_puts(m, "user "); + msm_framebuffer_describe(fb, m); + } + mutex_unlock(&dev->mode_config.fb_lock); + + return 0; +} + +static struct drm_info_list msm_kms_debugfs_list[] = { + { "fb", msm_fb_show }, +}; + /* * Display Snapshot: */ @@ -180,6 +210,27 @@ static const struct file_operations msm_kms_fops = { .release = msm_kms_release, }; +static void msm_debugfs_kms_init(struct drm_minor *minor) +{ + struct drm_device *dev = minor->dev; + struct msm_drm_private *priv = dev->dev_private; + + drm_debugfs_create_files(msm_kms_debugfs_list, + ARRAY_SIZE(msm_kms_debugfs_list), + minor->debugfs_root, minor); + debugfs_create_file("kms", 0400, minor->debugfs_root, + dev, &msm_kms_fops); + + if (priv->kms->funcs->debugfs_init) + priv->kms->funcs->debugfs_init(priv->kms, minor); + +} +#else /* ! CONFIG_DRM_MSM_KMS */ +static void msm_debugfs_kms_init(struct drm_minor *minor) +{ +} +#endif + /* * Other debugfs: */ @@ -208,6 +259,35 @@ DEFINE_DEBUGFS_ATTRIBUTE(shrink_fops, shrink_get, shrink_set, "0x%08llx\n"); +/* + * Return the number of microseconds to wait until stall-on-fault is + * re-enabled. If 0 then it is already enabled or will be re-enabled on the + * next submit (unless there's a leftover devcoredump). This is useful for + * kernel tests that intentionally produce a fault and check the devcoredump to + * wait until the cooldown period is over. + */ + +static int +stall_reenable_time_get(void *data, u64 *val) +{ + struct msm_drm_private *priv = data; + unsigned long irq_flags; + + spin_lock_irqsave(&priv->fault_stall_lock, irq_flags); + + if (priv->stall_enabled) + *val = 0; + else + *val = max(ktime_us_delta(priv->stall_reenable_time, ktime_get()), 0); + + spin_unlock_irqrestore(&priv->fault_stall_lock, irq_flags); + + return 0; +} + +DEFINE_DEBUGFS_ATTRIBUTE(stall_reenable_time_fops, + stall_reenable_time_get, NULL, + "%lld\n"); static int msm_gem_show(struct seq_file *m, void *arg) { @@ -238,47 +318,23 @@ static int msm_mm_show(struct seq_file *m, void *arg) return 0; } -static int msm_fb_show(struct seq_file *m, void *arg) -{ - struct drm_info_node *node = m->private; - struct drm_device *dev = node->minor->dev; - struct drm_framebuffer *fb, *fbdev_fb = NULL; - - if (dev->fb_helper && dev->fb_helper->fb) { - seq_printf(m, "fbcon "); - fbdev_fb = dev->fb_helper->fb; - msm_framebuffer_describe(fbdev_fb, m); - } - - mutex_lock(&dev->mode_config.fb_lock); - list_for_each_entry(fb, &dev->mode_config.fb_list, head) { - if (fb == fbdev_fb) - continue; - - seq_printf(m, "user "); - msm_framebuffer_describe(fb, m); - } - mutex_unlock(&dev->mode_config.fb_lock); - - return 0; -} - static struct drm_info_list msm_debugfs_list[] = { {"gem", msm_gem_show}, { "mm", msm_mm_show }, }; -static struct drm_info_list msm_kms_debugfs_list[] = { - { "fb", msm_fb_show }, -}; - static int late_init_minor(struct drm_minor *minor) { + struct drm_device *dev = minor->dev; + struct msm_drm_private *priv = dev->dev_private; int ret; if (!minor) return 0; + if (!priv->gpu_pdev) + return 0; + ret = msm_rd_debugfs_init(minor); if (ret) { DRM_DEV_ERROR(minor->dev->dev, "could not install rd debugfs\n"); @@ -319,6 +375,9 @@ static void msm_debugfs_gpu_init(struct drm_minor *minor) debugfs_create_bool("disable_err_irq", 0600, minor->debugfs_root, &priv->disable_err_irq); + debugfs_create_file("stall_reenable_time_us", 0400, minor->debugfs_root, + priv, &stall_reenable_time_fops); + gpu_devfreq = debugfs_create_dir("devfreq", minor->debugfs_root); debugfs_create_bool("idle_clamp",0600, gpu_devfreq, @@ -343,20 +402,12 @@ void msm_debugfs_init(struct drm_minor *minor) if (priv->gpu_pdev) msm_debugfs_gpu_init(minor); - if (priv->kms) { - drm_debugfs_create_files(msm_kms_debugfs_list, - ARRAY_SIZE(msm_kms_debugfs_list), - minor->debugfs_root, minor); - debugfs_create_file("kms", S_IRUSR, minor->debugfs_root, - dev, &msm_kms_fops); - } + if (priv->kms) + msm_debugfs_kms_init(minor); debugfs_create_file("shrink", S_IRWXU, minor->debugfs_root, dev, &shrink_fops); - if (priv->kms && priv->kms->funcs->debugfs_init) - priv->kms->funcs->debugfs_init(priv->kms, minor); - fault_create_debugfs_attr("fail_gem_alloc", minor->debugfs_root, &fail_gem_alloc); fault_create_debugfs_attr("fail_gem_iova", minor->debugfs_root, diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index f316e6776f67..9dcc7a596a11 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -11,7 +11,6 @@ #include <linux/of_address.h> #include <linux/uaccess.h> -#include <drm/clients/drm_client_setup.h> #include <drm/drm_drv.h> #include <drm/drm_file.h> #include <drm/drm_ioctl.h> @@ -41,17 +40,12 @@ * - 1.10.0 - Add MSM_SUBMIT_BO_NO_IMPLICIT * - 1.11.0 - Add wait boost (MSM_WAIT_FENCE_BOOST, MSM_PREP_BOOST) * - 1.12.0 - Add MSM_INFO_SET_METADATA and MSM_INFO_GET_METADATA + * - 1.13.0 - Add VM_BIND */ #define MSM_VERSION_MAJOR 1 -#define MSM_VERSION_MINOR 12 +#define MSM_VERSION_MINOR 13 #define MSM_VERSION_PATCHLEVEL 0 -static void msm_deinit_vram(struct drm_device *ddev); - -static char *vram = "16m"; -MODULE_PARM_DESC(vram, "Configure VRAM size (for devices without IOMMU/GPUMMU)"); -module_param(vram, charp, 0); - bool dumpstate; MODULE_PARM_DESC(dumpstate, "Dump KMS state on errors"); module_param(dumpstate, bool, 0600); @@ -60,10 +54,19 @@ static bool modeset = true; MODULE_PARM_DESC(modeset, "Use kernel modesetting [KMS] (1=on (default), 0=disable)"); module_param(modeset, bool, 0600); +static bool separate_gpu_kms; +MODULE_PARM_DESC(separate_gpu_drm, "Use separate DRM device for the GPU (0=single DRM device for both GPU and display (default), 1=two DRM devices)"); +module_param(separate_gpu_kms, bool, 0400); + DECLARE_FAULT_ATTR(fail_gem_alloc); DECLARE_FAULT_ATTR(fail_gem_iova); -static int msm_drm_uninit(struct device *dev) +bool msm_gpu_no_components(void) +{ + return separate_gpu_kms; +} + +static int msm_drm_uninit(struct device *dev, const struct component_ops *gpu_ops) { struct platform_device *pdev = to_platform_device(dev); struct msm_drm_private *priv = platform_get_drvdata(pdev); @@ -79,16 +82,9 @@ static int msm_drm_uninit(struct device *dev) if (ddev->registered) { drm_dev_unregister(ddev); if (priv->kms) - drm_atomic_helper_shutdown(ddev); + msm_drm_kms_unregister(dev); } - /* We must cancel and cleanup any pending vblank enable/disable - * work before msm_irq_uninstall() to avoid work re-enabling an - * irq after uninstall has disabled it. - */ - - flush_workqueue(priv->wq); - msm_gem_shrinker_cleanup(ddev); msm_perf_debugfs_cleanup(priv); @@ -97,120 +93,19 @@ static int msm_drm_uninit(struct device *dev) if (priv->kms) msm_drm_kms_uninit(dev); - msm_deinit_vram(ddev); - - component_unbind_all(dev, ddev); + if (gpu_ops) + gpu_ops->unbind(dev, dev, NULL); + else + component_unbind_all(dev, ddev); ddev->dev_private = NULL; drm_dev_put(ddev); - destroy_workqueue(priv->wq); - return 0; } -bool msm_use_mmu(struct drm_device *dev) -{ - struct msm_drm_private *priv = dev->dev_private; - - /* - * a2xx comes with its own MMU - * On other platforms IOMMU can be declared specified either for the - * MDP/DPU device or for its parent, MDSS device. - */ - return priv->is_a2xx || - device_iommu_mapped(dev->dev) || - device_iommu_mapped(dev->dev->parent); -} - -static int msm_init_vram(struct drm_device *dev) -{ - struct msm_drm_private *priv = dev->dev_private; - struct device_node *node; - unsigned long size = 0; - int ret = 0; - - /* In the device-tree world, we could have a 'memory-region' - * phandle, which gives us a link to our "vram". Allocating - * is all nicely abstracted behind the dma api, but we need - * to know the entire size to allocate it all in one go. There - * are two cases: - * 1) device with no IOMMU, in which case we need exclusive - * access to a VRAM carveout big enough for all gpu - * buffers - * 2) device with IOMMU, but where the bootloader puts up - * a splash screen. In this case, the VRAM carveout - * need only be large enough for fbdev fb. But we need - * exclusive access to the buffer to avoid the kernel - * using those pages for other purposes (which appears - * as corruption on screen before we have a chance to - * load and do initial modeset) - */ - - node = of_parse_phandle(dev->dev->of_node, "memory-region", 0); - if (node) { - struct resource r; - ret = of_address_to_resource(node, 0, &r); - of_node_put(node); - if (ret) - return ret; - size = r.end - r.start + 1; - DRM_INFO("using VRAM carveout: %lx@%pa\n", size, &r.start); - - /* if we have no IOMMU, then we need to use carveout allocator. - * Grab the entire DMA chunk carved out in early startup in - * mach-msm: - */ - } else if (!msm_use_mmu(dev)) { - DRM_INFO("using %s VRAM carveout\n", vram); - size = memparse(vram, NULL); - } - - if (size) { - unsigned long attrs = 0; - void *p; - - priv->vram.size = size; - - drm_mm_init(&priv->vram.mm, 0, (size >> PAGE_SHIFT) - 1); - spin_lock_init(&priv->vram.lock); - - attrs |= DMA_ATTR_NO_KERNEL_MAPPING; - attrs |= DMA_ATTR_WRITE_COMBINE; - - /* note that for no-kernel-mapping, the vaddr returned - * is bogus, but non-null if allocation succeeded: - */ - p = dma_alloc_attrs(dev->dev, size, - &priv->vram.paddr, GFP_KERNEL, attrs); - if (!p) { - DRM_DEV_ERROR(dev->dev, "failed to allocate VRAM\n"); - priv->vram.paddr = 0; - return -ENOMEM; - } - - DRM_DEV_INFO(dev->dev, "VRAM: %08x->%08x\n", - (uint32_t)priv->vram.paddr, - (uint32_t)(priv->vram.paddr + size)); - } - - return ret; -} - -static void msm_deinit_vram(struct drm_device *ddev) -{ - struct msm_drm_private *priv = ddev->dev_private; - unsigned long attrs = DMA_ATTR_NO_KERNEL_MAPPING; - - if (!priv->vram.paddr) - return; - - drm_mm_takedown(&priv->vram.mm); - dma_free_attrs(ddev->dev, priv->vram.size, NULL, priv->vram.paddr, - attrs); -} - -static int msm_drm_init(struct device *dev, const struct drm_driver *drv) +static int msm_drm_init(struct device *dev, const struct drm_driver *drv, + const struct component_ops *gpu_ops) { struct msm_drm_private *priv = dev_get_drvdata(dev); struct drm_device *ddev; @@ -227,12 +122,6 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) ddev->dev_private = priv; priv->dev = ddev; - priv->wq = alloc_ordered_workqueue("msm", 0); - if (!priv->wq) { - ret = -ENOMEM; - goto err_put_dev; - } - INIT_LIST_HEAD(&priv->objects); mutex_init(&priv->obj_lock); @@ -245,6 +134,10 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) drm_gem_lru_init(&priv->lru.willneed, &priv->lru.lock); drm_gem_lru_init(&priv->lru.dontneed, &priv->lru.lock); + /* Initialize stall-on-fault */ + spin_lock_init(&priv->fault_stall_lock); + priv->stall_enabled = true; + /* Teach lockdep about lock ordering wrt. shrinker: */ fs_reclaim_acquire(GFP_KERNEL); might_lock(&priv->lru.lock); @@ -253,19 +146,18 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) if (priv->kms_init) { ret = drmm_mode_config_init(ddev); if (ret) - goto err_destroy_wq; + goto err_put_dev; } - ret = msm_init_vram(ddev); - if (ret) - goto err_destroy_wq; - dma_set_max_seg_size(dev, UINT_MAX); /* Bind all our sub-components: */ - ret = component_bind_all(dev, ddev); + if (gpu_ops) + ret = gpu_ops->bind(dev, dev, NULL); + else + ret = component_bind_all(dev, ddev); if (ret) - goto err_deinit_vram; + goto err_put_dev; ret = msm_gem_shrinker_init(ddev); if (ret) @@ -275,11 +167,6 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) ret = msm_drm_kms_init(dev, drv); if (ret) goto err_msm_uninit; - } else { - /* valid only for the dummy headless case, where of_node=NULL */ - WARN_ON(dev->of_node); - ddev->driver_features &= ~DRIVER_MODESET; - ddev->driver_features &= ~DRIVER_ATOMIC; } ret = drm_dev_register(ddev, 0); @@ -290,22 +177,16 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) if (ret) goto err_msm_uninit; - if (priv->kms_init) { - drm_kms_helper_poll_init(ddev); - drm_client_setup(ddev, NULL); - } + if (priv->kms_init) + msm_drm_kms_post_init(dev); return 0; err_msm_uninit: - msm_drm_uninit(dev); + msm_drm_uninit(dev, gpu_ops); return ret; -err_deinit_vram: - msm_deinit_vram(ddev); -err_destroy_wq: - destroy_workqueue(priv->wq); err_put_dev: drm_dev_put(ddev); @@ -329,11 +210,42 @@ static void load_gpu(struct drm_device *dev) mutex_unlock(&init_lock); } +/** + * msm_context_vm - lazily create the context's VM + * + * @dev: the drm device + * @ctx: the context + * + * The VM is lazily created, so that userspace has a chance to opt-in to having + * a userspace managed VM before the VM is created. + * + * Note that this does not return a reference to the VM. Once the VM is created, + * it exists for the lifetime of the context. + */ +struct drm_gpuvm *msm_context_vm(struct drm_device *dev, struct msm_context *ctx) +{ + static DEFINE_MUTEX(init_lock); + struct msm_drm_private *priv = dev->dev_private; + + /* Once ctx->vm is created it is valid for the lifetime of the context: */ + if (ctx->vm) + return ctx->vm; + + mutex_lock(&init_lock); + if (!ctx->vm) { + ctx->vm = msm_gpu_create_private_vm( + priv->gpu, current, !ctx->userspace_managed_vm); + + } + mutex_unlock(&init_lock); + + return ctx->vm; +} + static int context_init(struct drm_device *dev, struct drm_file *file) { static atomic_t ident = ATOMIC_INIT(0); - struct msm_drm_private *priv = dev->dev_private; - struct msm_file_private *ctx; + struct msm_context *ctx; ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); if (!ctx) @@ -345,7 +257,6 @@ static int context_init(struct drm_device *dev, struct drm_file *file) kref_init(&ctx->ref); msm_submitqueue_init(dev, ctx); - ctx->aspace = msm_gpu_create_private_address_space(priv->gpu, current); file->driver_priv = ctx; ctx->seqno = atomic_inc_return(&ident); @@ -363,23 +274,24 @@ static int msm_open(struct drm_device *dev, struct drm_file *file) return context_init(dev, file); } -static void context_close(struct msm_file_private *ctx) +static void context_close(struct msm_context *ctx) { + ctx->closed = true; msm_submitqueue_close(ctx); - msm_file_private_put(ctx); + msm_context_put(ctx); } static void msm_postclose(struct drm_device *dev, struct drm_file *file) { struct msm_drm_private *priv = dev->dev_private; - struct msm_file_private *ctx = file->driver_priv; + struct msm_context *ctx = file->driver_priv; /* * It is not possible to set sysprof param to non-zero if gpu * is not initialized: */ if (priv->gpu) - msm_file_private_set_sysprof(ctx, priv->gpu, 0); + msm_context_set_sysprof(ctx, priv->gpu, 0); context_close(ctx); } @@ -511,11 +423,14 @@ static int msm_ioctl_gem_info_iova(struct drm_device *dev, uint64_t *iova) { struct msm_drm_private *priv = dev->dev_private; - struct msm_file_private *ctx = file->driver_priv; + struct msm_context *ctx = file->driver_priv; if (!priv->gpu) return -EINVAL; + if (msm_context_is_vmbind(ctx)) + return UERR(EINVAL, dev, "VM_BIND is enabled"); + if (should_fail(&fail_gem_iova, obj->size)) return -ENOMEM; @@ -523,7 +438,7 @@ static int msm_ioctl_gem_info_iova(struct drm_device *dev, * Don't pin the memory here - just get an address so that userspace can * be productive */ - return msm_gem_get_iova(obj, ctx->aspace, iova); + return msm_gem_get_iova(obj, msm_context_vm(dev, ctx), iova); } static int msm_ioctl_gem_info_set_iova(struct drm_device *dev, @@ -531,19 +446,23 @@ static int msm_ioctl_gem_info_set_iova(struct drm_device *dev, uint64_t iova) { struct msm_drm_private *priv = dev->dev_private; - struct msm_file_private *ctx = file->driver_priv; + struct msm_context *ctx = file->driver_priv; + struct drm_gpuvm *vm = msm_context_vm(dev, ctx); if (!priv->gpu) return -EINVAL; + if (msm_context_is_vmbind(ctx)) + return UERR(EINVAL, dev, "VM_BIND is enabled"); + /* Only supported if per-process address space is supported: */ - if (priv->gpu->aspace == ctx->aspace) + if (priv->gpu->vm == vm) return UERR(EOPNOTSUPP, dev, "requires per-process pgtables"); if (should_fail(&fail_gem_iova, obj->size)) return -ENOMEM; - return msm_gem_set_iova(obj, ctx->aspace, iova); + return msm_gem_set_iova(obj, vm, iova); } static int msm_ioctl_gem_info_set_metadata(struct drm_gem_object *obj, @@ -551,6 +470,7 @@ static int msm_ioctl_gem_info_set_metadata(struct drm_gem_object *obj, u32 metadata_size) { struct msm_gem_object *msm_obj = to_msm_bo(obj); + void *new_metadata; void *buf; int ret; @@ -568,8 +488,14 @@ static int msm_ioctl_gem_info_set_metadata(struct drm_gem_object *obj, if (ret) goto out; - msm_obj->metadata = + new_metadata = krealloc(msm_obj->metadata, metadata_size, GFP_KERNEL); + if (!new_metadata) { + ret = -ENOMEM; + goto out; + } + + msm_obj->metadata = new_metadata; msm_obj->metadata_size = metadata_size; memcpy(msm_obj->metadata, buf, metadata_size); @@ -868,6 +794,7 @@ static const struct drm_ioctl_desc msm_ioctls[] = { DRM_IOCTL_DEF_DRV(MSM_SUBMITQUEUE_NEW, msm_ioctl_submitqueue_new, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_SUBMITQUEUE_CLOSE, msm_ioctl_submitqueue_close, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(MSM_SUBMITQUEUE_QUERY, msm_ioctl_submitqueue_query, DRM_RENDER_ALLOW), + DRM_IOCTL_DEF_DRV(MSM_VM_BIND, msm_ioctl_vm_bind, DRM_RENDER_ALLOW), }; static void msm_show_fdinfo(struct drm_printer *p, struct drm_file *file) @@ -889,17 +816,27 @@ static const struct file_operations fops = { .show_fdinfo = drm_show_fdinfo, }; +#define DRIVER_FEATURES_GPU ( \ + DRIVER_GEM | \ + DRIVER_GEM_GPUVA | \ + DRIVER_RENDER | \ + DRIVER_SYNCOBJ | \ + DRIVER_SYNCOBJ_TIMELINE | \ + 0 ) + +#define DRIVER_FEATURES_KMS ( \ + DRIVER_GEM | \ + DRIVER_ATOMIC | \ + DRIVER_MODESET | \ + 0 ) + static const struct drm_driver msm_driver = { - .driver_features = DRIVER_GEM | - DRIVER_RENDER | - DRIVER_ATOMIC | - DRIVER_MODESET | - DRIVER_SYNCOBJ_TIMELINE | - DRIVER_SYNCOBJ, + .driver_features = DRIVER_FEATURES_GPU | DRIVER_FEATURES_KMS, .open = msm_open, .postclose = msm_postclose, .dumb_create = msm_gem_dumb_create, .dumb_map_offset = msm_gem_dumb_map_offset, + .gem_prime_import = msm_gem_prime_import, .gem_prime_import_sg_table = msm_gem_prime_import_sg_table, #ifdef CONFIG_DEBUG_FS .debugfs_init = msm_debugfs_init, @@ -916,6 +853,45 @@ static const struct drm_driver msm_driver = { .patchlevel = MSM_VERSION_PATCHLEVEL, }; +static const struct drm_driver msm_kms_driver = { + .driver_features = DRIVER_FEATURES_KMS, + .open = msm_open, + .postclose = msm_postclose, + .dumb_create = msm_gem_dumb_create, + .dumb_map_offset = msm_gem_dumb_map_offset, + .gem_prime_import_sg_table = msm_gem_prime_import_sg_table, +#ifdef CONFIG_DEBUG_FS + .debugfs_init = msm_debugfs_init, +#endif + MSM_FBDEV_DRIVER_OPS, + .show_fdinfo = msm_show_fdinfo, + .fops = &fops, + .name = "msm-kms", + .desc = "MSM Snapdragon DRM", + .major = MSM_VERSION_MAJOR, + .minor = MSM_VERSION_MINOR, + .patchlevel = MSM_VERSION_PATCHLEVEL, +}; + +static const struct drm_driver msm_gpu_driver = { + .driver_features = DRIVER_FEATURES_GPU, + .open = msm_open, + .postclose = msm_postclose, + .gem_prime_import_sg_table = msm_gem_prime_import_sg_table, +#ifdef CONFIG_DEBUG_FS + .debugfs_init = msm_debugfs_init, +#endif + .show_fdinfo = msm_show_fdinfo, + .ioctls = msm_ioctls, + .num_ioctls = ARRAY_SIZE(msm_ioctls), + .fops = &fops, + .name = "msm", + .desc = "MSM Snapdragon DRM", + .major = MSM_VERSION_MAJOR, + .minor = MSM_VERSION_MINOR, + .patchlevel = MSM_VERSION_PATCHLEVEL, +}; + /* * Componentized driver support: */ @@ -926,7 +902,7 @@ static const struct drm_driver msm_driver = { * is no external component that we need to add since LVDS is within MDP4 * itself. */ -static int add_components_mdp(struct device *master_dev, +static int add_mdp_components(struct device *master_dev, struct component_match **matchptr) { struct device_node *np = master_dev->of_node; @@ -1030,7 +1006,7 @@ static int add_gpu_components(struct device *dev, if (!np) return 0; - if (of_device_is_available(np)) + if (of_device_is_available(np) && adreno_has_gpu(np)) drm_of_component_match_add(dev, matchptr, component_compare_of, np); of_node_put(np); @@ -1040,12 +1016,16 @@ static int add_gpu_components(struct device *dev, static int msm_drm_bind(struct device *dev) { - return msm_drm_init(dev, &msm_driver); + return msm_drm_init(dev, + msm_gpu_no_components() ? + &msm_kms_driver : + &msm_driver, + NULL); } static void msm_drm_unbind(struct device *dev) { - msm_drm_uninit(dev); + msm_drm_uninit(dev, NULL); } const struct component_master_ops msm_drm_ops = { @@ -1071,14 +1051,16 @@ int msm_drv_probe(struct device *master_dev, /* Add mdp components if we have KMS. */ if (kms_init) { - ret = add_components_mdp(master_dev, &match); + ret = add_mdp_components(master_dev, &match); if (ret) return ret; } - ret = add_gpu_components(master_dev, &match); - if (ret) - return ret; + if (!msm_gpu_no_components()) { + ret = add_gpu_components(master_dev, &match); + if (ret) + return ret; + } /* on all devices that I am aware of, iommu's which can map * any address the cpu can see are used: @@ -1094,29 +1076,34 @@ int msm_drv_probe(struct device *master_dev, return 0; } -/* - * Platform driver: - * Used only for headlesss GPU instances - */ - -static int msm_pdev_probe(struct platform_device *pdev) +int msm_gpu_probe(struct platform_device *pdev, + const struct component_ops *ops) { - return msm_drv_probe(&pdev->dev, NULL, NULL); + struct msm_drm_private *priv; + int ret; + + priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + platform_set_drvdata(pdev, priv); + + /* on all devices that I am aware of, iommu's which can map + * any address the cpu can see are used: + */ + ret = dma_set_mask_and_coherent(&pdev->dev, ~0); + if (ret) + return ret; + + return msm_drm_init(&pdev->dev, &msm_gpu_driver, ops); } -static void msm_pdev_remove(struct platform_device *pdev) +void msm_gpu_remove(struct platform_device *pdev, + const struct component_ops *ops) { - component_master_del(&pdev->dev, &msm_drm_ops); + msm_drm_uninit(&pdev->dev, ops); } -static struct platform_driver msm_platform_driver = { - .probe = msm_pdev_probe, - .remove = msm_pdev_remove, - .driver = { - .name = "msm", - }, -}; - static int __init msm_drm_register(void) { if (!modeset) @@ -1131,13 +1118,13 @@ static int __init msm_drm_register(void) adreno_register(); msm_mdp4_register(); msm_mdss_register(); - return platform_driver_register(&msm_platform_driver); + + return 0; } static void __exit msm_drm_unregister(void) { DBG("fini"); - platform_driver_unregister(&msm_platform_driver); msm_mdss_unregister(); msm_mdp4_unregister(); msm_dp_unregister(); diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index a65077855201..9875ca62e9ad 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -48,8 +48,6 @@ struct msm_rd_state; struct msm_perf_state; struct msm_gem_submit; struct msm_fence_context; -struct msm_gem_address_space; -struct msm_gem_vma; struct msm_disp_state; #define MAX_CRTCS 8 @@ -72,12 +70,6 @@ enum msm_dsi_controller { #define MSM_GPU_MAX_RINGS 4 -/* Commit/Event thread specific structure */ -struct msm_drm_thread { - struct drm_device *dev; - struct kthread_worker *worker; -}; - struct msm_drm_private { struct drm_device *dev; @@ -88,16 +80,6 @@ struct msm_drm_private { /* subordinate devices, if present: */ struct platform_device *gpu_pdev; - /* possibly this should be in the kms component, but it is - * shared by both mdp4 and mdp5.. - */ - struct hdmi *hdmi; - - /* DSI is shared by mdp4 and mdp5 */ - struct msm_dsi *dsi[MSM_DSI_CONTROLLER_COUNT]; - - struct msm_dp *dp[MSM_DP_CONTROLLER_COUNT]; - /* when we have more than one 'msm_gpu' these need to be an array: */ struct msm_gpu *gpu; @@ -177,23 +159,6 @@ struct msm_drm_private { struct mutex lock; } lru; - struct workqueue_struct *wq; - - unsigned int num_crtcs; - - struct msm_drm_thread event_thread[MAX_CRTCS]; - - /* VRAM carveout, used when no IOMMU: */ - struct { - unsigned long size; - dma_addr_t paddr; - /* NOTE: mm managed at the page level, size is in # of pages - * and position mm_node->start is in # of pages: - */ - struct drm_mm mm; - spinlock_t lock; /* Protects drm_mm node allocation/removal */ - } vram; - struct notifier_block vmap_notifier; struct shrinker *shrinker; @@ -222,6 +187,29 @@ struct msm_drm_private { * the sw hangcheck mechanism. */ bool disable_err_irq; + + /** + * @fault_stall_lock: + * + * Serialize changes to stall-on-fault state. + */ + spinlock_t fault_stall_lock; + + /** + * @fault_stall_reenable_time: + * + * If stall_enabled is false, when to reenable stall-on-fault. + * Protected by @fault_stall_lock. + */ + ktime_t stall_reenable_time; + + /** + * @stall_enabled: + * + * Whether stall-on-fault is currently enabled. Protected by + * @fault_stall_lock. + */ + bool stall_enabled; }; const struct msm_format *mdp_get_format(struct msm_kms *kms, uint32_t format, uint64_t modifier); @@ -241,11 +229,13 @@ void msm_crtc_disable_vblank(struct drm_crtc *crtc); int msm_register_mmu(struct drm_device *dev, struct msm_mmu *mmu); void msm_unregister_mmu(struct drm_device *dev, struct msm_mmu *mmu); -struct msm_gem_address_space *msm_kms_init_aspace(struct drm_device *dev); +struct drm_gpuvm *msm_kms_init_vm(struct drm_device *dev); bool msm_use_mmu(struct drm_device *dev); int msm_ioctl_gem_submit(struct drm_device *dev, void *data, - struct drm_file *file); + struct drm_file *file); +int msm_ioctl_vm_bind(struct drm_device *dev, void *data, + struct drm_file *file); #ifdef CONFIG_DEBUG_FS unsigned long msm_gem_shrinker_shrink(struct drm_device *dev, unsigned long nr_to_scan); @@ -257,17 +247,16 @@ void msm_gem_shrinker_cleanup(struct drm_device *dev); struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj); int msm_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map); void msm_gem_prime_vunmap(struct drm_gem_object *obj, struct iosys_map *map); +struct drm_gem_object *msm_gem_prime_import(struct drm_device *dev, struct dma_buf *buf); struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sg); +struct dma_buf *msm_gem_prime_export(struct drm_gem_object *obj, int flags); int msm_gem_prime_pin(struct drm_gem_object *obj); void msm_gem_prime_unpin(struct drm_gem_object *obj); -int msm_framebuffer_prepare(struct drm_framebuffer *fb, - struct msm_gem_address_space *aspace, bool needs_dirtyfb); -void msm_framebuffer_cleanup(struct drm_framebuffer *fb, - struct msm_gem_address_space *aspace, bool needed_dirtyfb); -uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb, - struct msm_gem_address_space *aspace, int plane); +int msm_framebuffer_prepare(struct drm_framebuffer *fb, bool needs_dirtyfb); +void msm_framebuffer_cleanup(struct drm_framebuffer *fb, bool needed_dirtyfb); +uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb, int plane); struct drm_gem_object *msm_framebuffer_bo(struct drm_framebuffer *fb, int plane); const struct msm_format *msm_framebuffer_format(struct drm_framebuffer *fb); struct drm_framebuffer *msm_framebuffer_create(struct drm_device *dev, @@ -275,7 +264,7 @@ struct drm_framebuffer *msm_framebuffer_create(struct drm_device *dev, struct drm_framebuffer * msm_alloc_stolen_fb(struct drm_device *dev, int w, int h, int p, uint32_t format); -#ifdef CONFIG_DRM_FBDEV_EMULATION +#ifdef CONFIG_DRM_MSM_KMS_FBDEV int msm_fbdev_driver_fbdev_probe(struct drm_fb_helper *helper, struct drm_fb_helper_surface_size *sizes); #define MSM_FBDEV_DRIVER_OPS \ @@ -360,6 +349,7 @@ static inline const char *msm_dsi_get_te_source(struct msm_dsi *msm_dsi) } #endif +struct msm_dp; #ifdef CONFIG_DRM_MSM_DP int __init msm_dp_register(void); void __exit msm_dp_unregister(void); @@ -553,6 +543,10 @@ extern const struct component_master_ops msm_drm_ops; int msm_kms_pm_prepare(struct device *dev); void msm_kms_pm_complete(struct device *dev); +int msm_gpu_probe(struct platform_device *pdev, + const struct component_ops *ops); +void msm_gpu_remove(struct platform_device *pdev, + const struct component_ops *ops); int msm_drv_probe(struct device *dev, int (*kms_init)(struct drm_device *dev), struct msm_kms *kms); @@ -560,4 +554,6 @@ void msm_kms_shutdown(struct platform_device *pdev); bool msm_disp_drv_should_bind(struct device *dev, bool dpu_driver); +bool msm_gpu_no_components(void); + #endif /* __MSM_DRV_H__ */ diff --git a/drivers/gpu/drm/msm/msm_fb.c b/drivers/gpu/drm/msm/msm_fb.c index 09268e416843..bc7c2bb8f01e 100644 --- a/drivers/gpu/drm/msm/msm_fb.c +++ b/drivers/gpu/drm/msm/msm_fb.c @@ -75,20 +75,22 @@ void msm_framebuffer_describe(struct drm_framebuffer *fb, struct seq_file *m) /* prepare/pin all the fb's bo's for scanout. */ -int msm_framebuffer_prepare(struct drm_framebuffer *fb, - struct msm_gem_address_space *aspace, - bool needs_dirtyfb) +int msm_framebuffer_prepare(struct drm_framebuffer *fb, bool needs_dirtyfb) { + struct msm_drm_private *priv = fb->dev->dev_private; + struct drm_gpuvm *vm = priv->kms->vm; struct msm_framebuffer *msm_fb = to_msm_framebuffer(fb); int ret, i, n = fb->format->num_planes; if (needs_dirtyfb) refcount_inc(&msm_fb->dirtyfb); - atomic_inc(&msm_fb->prepare_count); + if (atomic_inc_return(&msm_fb->prepare_count) > 1) + return 0; for (i = 0; i < n; i++) { - ret = msm_gem_get_and_pin_iova(fb->obj[i], aspace, &msm_fb->iova[i]); + msm_gem_vma_get(fb->obj[i]); + ret = msm_gem_get_and_pin_iova(fb->obj[i], vm, &msm_fb->iova[i]); drm_dbg_state(fb->dev, "FB[%u]: iova[%d]: %08llx (%d)\n", fb->base.id, i, msm_fb->iova[i], ret); if (ret) @@ -98,25 +100,28 @@ int msm_framebuffer_prepare(struct drm_framebuffer *fb, return 0; } -void msm_framebuffer_cleanup(struct drm_framebuffer *fb, - struct msm_gem_address_space *aspace, - bool needed_dirtyfb) +void msm_framebuffer_cleanup(struct drm_framebuffer *fb, bool needed_dirtyfb) { + struct msm_drm_private *priv = fb->dev->dev_private; + struct drm_gpuvm *vm = priv->kms->vm; struct msm_framebuffer *msm_fb = to_msm_framebuffer(fb); int i, n = fb->format->num_planes; if (needed_dirtyfb) refcount_dec(&msm_fb->dirtyfb); - for (i = 0; i < n; i++) - msm_gem_unpin_iova(fb->obj[i], aspace); + if (atomic_dec_return(&msm_fb->prepare_count)) + return; + + memset(msm_fb->iova, 0, sizeof(msm_fb->iova)); - if (!atomic_dec_return(&msm_fb->prepare_count)) - memset(msm_fb->iova, 0, sizeof(msm_fb->iova)); + for (i = 0; i < n; i++) { + msm_gem_unpin_iova(fb->obj[i], vm); + msm_gem_vma_put(fb->obj[i]); + } } -uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb, - struct msm_gem_address_space *aspace, int plane) +uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb, int plane) { struct msm_framebuffer *msm_fb = to_msm_framebuffer(fb); return msm_fb->iova[plane] + fb->offsets[plane]; diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c index c62249b1ab3d..b5969374d53f 100644 --- a/drivers/gpu/drm/msm/msm_fbdev.c +++ b/drivers/gpu/drm/msm/msm_fbdev.c @@ -122,7 +122,7 @@ int msm_fbdev_driver_fbdev_probe(struct drm_fb_helper *helper, * in panic (ie. lock-safe, etc) we could avoid pinning the * buffer now: */ - ret = msm_gem_get_and_pin_iova(bo, priv->kms->aspace, &paddr); + ret = msm_gem_get_and_pin_iova(bo, priv->kms->vm, &paddr); if (ret) { DRM_DEV_ERROR(dev->dev, "failed to get buffer obj iova: %d\n", ret); goto fail; diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index 2995e80fec3b..33d3354c6102 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -17,24 +17,9 @@ #include <trace/events/gpu_mem.h> #include "msm_drv.h" -#include "msm_fence.h" #include "msm_gem.h" #include "msm_gpu.h" -#include "msm_mmu.h" - -static dma_addr_t physaddr(struct drm_gem_object *obj) -{ - struct msm_gem_object *msm_obj = to_msm_bo(obj); - struct msm_drm_private *priv = obj->dev->dev_private; - return (((dma_addr_t)msm_obj->vram_node->start) << PAGE_SHIFT) + - priv->vram.paddr; -} - -static bool use_pages(struct drm_gem_object *obj) -{ - struct msm_gem_object *msm_obj = to_msm_bo(obj); - return !msm_obj->vram_node; -} +#include "msm_kms.h" static void update_device_mem(struct msm_drm_private *priv, ssize_t size) { @@ -44,7 +29,7 @@ static void update_device_mem(struct msm_drm_private *priv, ssize_t size) static void update_ctx_mem(struct drm_file *file, ssize_t size) { - struct msm_file_private *ctx = file->driver_priv; + struct msm_context *ctx = file->driver_priv; uint64_t ctx_mem = atomic64_add_return(size, &ctx->ctx_mem); rcu_read_lock(); /* Locks file->pid! */ @@ -55,13 +40,73 @@ static void update_ctx_mem(struct drm_file *file, ssize_t size) static int msm_gem_open(struct drm_gem_object *obj, struct drm_file *file) { + msm_gem_vma_get(obj); update_ctx_mem(file, obj->size); return 0; } +static void put_iova_spaces(struct drm_gem_object *obj, struct drm_gpuvm *vm, + bool close, const char *reason); + static void msm_gem_close(struct drm_gem_object *obj, struct drm_file *file) { + struct msm_context *ctx = file->driver_priv; + struct drm_exec exec; + update_ctx_mem(file, -obj->size); + msm_gem_vma_put(obj); + + /* + * If VM isn't created yet, nothing to cleanup. And in fact calling + * put_iova_spaces() with vm=NULL would be bad, in that it will tear- + * down the mappings of shared buffers in other contexts. + */ + if (!ctx->vm) + return; + + /* + * VM_BIND does not depend on implicit teardown of VMAs on handle + * close, but instead on implicit teardown of the VM when the device + * is closed (see msm_gem_vm_close()) + */ + if (msm_context_is_vmbind(ctx)) + return; + + /* + * TODO we might need to kick this to a queue to avoid blocking + * in CLOSE ioctl + */ + dma_resv_wait_timeout(obj->resv, DMA_RESV_USAGE_BOOKKEEP, false, + MAX_SCHEDULE_TIMEOUT); + + msm_gem_lock_vm_and_obj(&exec, obj, ctx->vm); + put_iova_spaces(obj, ctx->vm, true, "close"); + drm_exec_fini(&exec); /* drop locks */ +} + +/* + * Get/put for kms->vm VMA + */ + +void msm_gem_vma_get(struct drm_gem_object *obj) +{ + atomic_inc(&to_msm_bo(obj)->vma_ref); +} + +void msm_gem_vma_put(struct drm_gem_object *obj) +{ + struct msm_drm_private *priv = obj->dev->dev_private; + struct drm_exec exec; + + if (atomic_dec_return(&to_msm_bo(obj)->vma_ref)) + return; + + if (!priv->kms) + return; + + msm_gem_lock_vm_and_obj(&exec, obj, priv->kms->vm); + put_iova_spaces(obj, priv->kms->vm, true, "vma_put"); + drm_exec_fini(&exec); /* drop locks */ } /* @@ -135,36 +180,6 @@ static void update_lru(struct drm_gem_object *obj) mutex_unlock(&priv->lru.lock); } -/* allocate pages from VRAM carveout, used when no IOMMU: */ -static struct page **get_pages_vram(struct drm_gem_object *obj, int npages) -{ - struct msm_gem_object *msm_obj = to_msm_bo(obj); - struct msm_drm_private *priv = obj->dev->dev_private; - dma_addr_t paddr; - struct page **p; - int ret, i; - - p = kvmalloc_array(npages, sizeof(struct page *), GFP_KERNEL); - if (!p) - return ERR_PTR(-ENOMEM); - - spin_lock(&priv->vram.lock); - ret = drm_mm_insert_node(&priv->vram.mm, msm_obj->vram_node, npages); - spin_unlock(&priv->vram.lock); - if (ret) { - kvfree(p); - return ERR_PTR(ret); - } - - paddr = physaddr(obj); - for (i = 0; i < npages; i++) { - p[i] = pfn_to_page(__phys_to_pfn(paddr)); - paddr += PAGE_SIZE; - } - - return p; -} - static struct page **get_pages(struct drm_gem_object *obj) { struct msm_gem_object *msm_obj = to_msm_bo(obj); @@ -176,10 +191,7 @@ static struct page **get_pages(struct drm_gem_object *obj) struct page **p; int npages = obj->size >> PAGE_SHIFT; - if (use_pages(obj)) - p = drm_gem_get_pages(obj); - else - p = get_pages_vram(obj, npages); + p = drm_gem_get_pages(obj); if (IS_ERR(p)) { DRM_DEV_ERROR(dev->dev, "could not get pages: %ld\n", @@ -212,22 +224,17 @@ static struct page **get_pages(struct drm_gem_object *obj) return msm_obj->pages; } -static void put_pages_vram(struct drm_gem_object *obj) -{ - struct msm_gem_object *msm_obj = to_msm_bo(obj); - struct msm_drm_private *priv = obj->dev->dev_private; - - spin_lock(&priv->vram.lock); - drm_mm_remove_node(msm_obj->vram_node); - spin_unlock(&priv->vram.lock); - - kvfree(msm_obj->pages); -} - static void put_pages(struct drm_gem_object *obj) { struct msm_gem_object *msm_obj = to_msm_bo(obj); + /* + * Skip gpuvm in the object free path to avoid a WARN_ON() splat. + * See explaination in msm_gem_assert_locked() + */ + if (kref_read(&obj->refcount)) + drm_gpuvm_bo_gem_evict(obj, true); + if (msm_obj->pages) { if (msm_obj->sgt) { /* For non-cached buffers, ensure the new @@ -244,18 +251,14 @@ static void put_pages(struct drm_gem_object *obj) update_device_mem(obj->dev->dev_private, -obj->size); - if (use_pages(obj)) - drm_gem_put_pages(obj, msm_obj->pages, true, false); - else - put_pages_vram(obj); + drm_gem_put_pages(obj, msm_obj->pages, true, false); msm_obj->pages = NULL; update_lru(obj); } } -static struct page **msm_gem_get_pages_locked(struct drm_gem_object *obj, - unsigned madv) +struct page **msm_gem_get_pages_locked(struct drm_gem_object *obj, unsigned madv) { struct msm_gem_object *msm_obj = to_msm_bo(obj); @@ -397,48 +400,31 @@ uint64_t msm_gem_mmap_offset(struct drm_gem_object *obj) return offset; } -static struct msm_gem_vma *add_vma(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace) +static struct drm_gpuva *lookup_vma(struct drm_gem_object *obj, + struct drm_gpuvm *vm) { - struct msm_gem_object *msm_obj = to_msm_bo(obj); - struct msm_gem_vma *vma; + struct drm_gpuvm_bo *vm_bo; msm_gem_assert_locked(obj); - vma = msm_gem_vma_new(aspace); - if (!vma) - return ERR_PTR(-ENOMEM); - - list_add_tail(&vma->list, &msm_obj->vmas); - - return vma; -} - -static struct msm_gem_vma *lookup_vma(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace) -{ - struct msm_gem_object *msm_obj = to_msm_bo(obj); - struct msm_gem_vma *vma; + drm_gem_for_each_gpuvm_bo (vm_bo, obj) { + struct drm_gpuva *vma; - msm_gem_assert_locked(obj); + drm_gpuvm_bo_for_each_va (vma, vm_bo) { + if (vma->vm == vm) { + /* lookup_vma() should only be used in paths + * with at most one vma per vm + */ + GEM_WARN_ON(!list_is_singular(&vm_bo->list.gpuva)); - list_for_each_entry(vma, &msm_obj->vmas, list) { - if (vma->aspace == aspace) - return vma; + return vma; + } + } } return NULL; } -static void del_vma(struct msm_gem_vma *vma) -{ - if (!vma) - return; - - list_del(&vma->list); - kfree(vma); -} - /* * If close is true, this also closes the VMA (releasing the allocated * iova range) in addition to removing the iommu mapping. In the eviction @@ -446,71 +432,54 @@ static void del_vma(struct msm_gem_vma *vma) * mapping. */ static void -put_iova_spaces(struct drm_gem_object *obj, bool close) +put_iova_spaces(struct drm_gem_object *obj, struct drm_gpuvm *vm, + bool close, const char *reason) { - struct msm_gem_object *msm_obj = to_msm_bo(obj); - struct msm_gem_vma *vma; + struct drm_gpuvm_bo *vm_bo, *tmp; msm_gem_assert_locked(obj); - list_for_each_entry(vma, &msm_obj->vmas, list) { - if (vma->aspace) { - msm_gem_vma_purge(vma); + drm_gem_for_each_gpuvm_bo_safe (vm_bo, tmp, obj) { + struct drm_gpuva *vma, *vmatmp; + + if (vm && vm_bo->vm != vm) + continue; + + drm_gpuvm_bo_get(vm_bo); + + drm_gpuvm_bo_for_each_va_safe (vma, vmatmp, vm_bo) { + msm_gem_vma_unmap(vma, reason); if (close) msm_gem_vma_close(vma); } - } -} -/* Called with msm_obj locked */ -static void -put_iova_vmas(struct drm_gem_object *obj) -{ - struct msm_gem_object *msm_obj = to_msm_bo(obj); - struct msm_gem_vma *vma, *tmp; - - msm_gem_assert_locked(obj); - - list_for_each_entry_safe(vma, tmp, &msm_obj->vmas, list) { - del_vma(vma); + drm_gpuvm_bo_put(vm_bo); } } -static struct msm_gem_vma *get_vma_locked(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace, - u64 range_start, u64 range_end) +static struct drm_gpuva *get_vma_locked(struct drm_gem_object *obj, + struct drm_gpuvm *vm, u64 range_start, + u64 range_end) { - struct msm_gem_vma *vma; + struct drm_gpuva *vma; msm_gem_assert_locked(obj); - vma = lookup_vma(obj, aspace); + vma = lookup_vma(obj, vm); if (!vma) { - int ret; - - vma = add_vma(obj, aspace); - if (IS_ERR(vma)) - return vma; - - ret = msm_gem_vma_init(vma, obj->size, - range_start, range_end); - if (ret) { - del_vma(vma); - return ERR_PTR(ret); - } + vma = msm_gem_vma_new(vm, obj, 0, range_start, range_end); } else { - GEM_WARN_ON(vma->iova < range_start); - GEM_WARN_ON((vma->iova + obj->size) > range_end); + GEM_WARN_ON(vma->va.addr < range_start); + GEM_WARN_ON((vma->va.addr + obj->size) > range_end); } return vma; } -int msm_gem_pin_vma_locked(struct drm_gem_object *obj, struct msm_gem_vma *vma) +int msm_gem_prot(struct drm_gem_object *obj) { struct msm_gem_object *msm_obj = to_msm_bo(obj); - struct page **pages; int prot = IOMMU_READ; if (!(msm_obj->flags & MSM_BO_GPU_READONLY)) @@ -522,13 +491,22 @@ int msm_gem_pin_vma_locked(struct drm_gem_object *obj, struct msm_gem_vma *vma) if (msm_obj->flags & MSM_BO_CACHED_COHERENT) prot |= IOMMU_CACHE; + return prot; +} + +int msm_gem_pin_vma_locked(struct drm_gem_object *obj, struct drm_gpuva *vma) +{ + struct msm_gem_object *msm_obj = to_msm_bo(obj); + struct page **pages; + int prot = msm_gem_prot(obj); + msm_gem_assert_locked(obj); pages = msm_gem_get_pages_locked(obj, MSM_MADV_WILLNEED); if (IS_ERR(pages)) return PTR_ERR(pages); - return msm_gem_vma_map(vma, prot, msm_obj->sgt, obj->size); + return msm_gem_vma_map(vma, prot, msm_obj->sgt); } void msm_gem_unpin_locked(struct drm_gem_object *obj) @@ -560,28 +538,31 @@ void msm_gem_unpin_active(struct drm_gem_object *obj) update_lru_active(obj); } -struct msm_gem_vma *msm_gem_get_vma_locked(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace) +struct drm_gpuva *msm_gem_get_vma_locked(struct drm_gem_object *obj, + struct drm_gpuvm *vm) { - return get_vma_locked(obj, aspace, 0, U64_MAX); + return get_vma_locked(obj, vm, 0, U64_MAX); } static int get_and_pin_iova_range_locked(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace, uint64_t *iova, - u64 range_start, u64 range_end) + struct drm_gpuvm *vm, uint64_t *iova, + u64 range_start, u64 range_end) { - struct msm_gem_vma *vma; + struct drm_gpuva *vma; int ret; msm_gem_assert_locked(obj); - vma = get_vma_locked(obj, aspace, range_start, range_end); + if (to_msm_bo(obj)->flags & MSM_BO_NO_SHARE) + return -EINVAL; + + vma = get_vma_locked(obj, vm, range_start, range_end); if (IS_ERR(vma)) return PTR_ERR(vma); ret = msm_gem_pin_vma_locked(obj, vma); if (!ret) { - *iova = vma->iova; + *iova = vma->va.addr; pin_obj_locked(obj); } @@ -593,58 +574,59 @@ static int get_and_pin_iova_range_locked(struct drm_gem_object *obj, * limits iova to specified range (in pages) */ int msm_gem_get_and_pin_iova_range(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace, uint64_t *iova, - u64 range_start, u64 range_end) + struct drm_gpuvm *vm, uint64_t *iova, + u64 range_start, u64 range_end) { + struct drm_exec exec; int ret; - msm_gem_lock(obj); - ret = get_and_pin_iova_range_locked(obj, aspace, iova, range_start, range_end); - msm_gem_unlock(obj); + msm_gem_lock_vm_and_obj(&exec, obj, vm); + ret = get_and_pin_iova_range_locked(obj, vm, iova, range_start, range_end); + drm_exec_fini(&exec); /* drop locks */ return ret; } /* get iova and pin it. Should have a matching put */ -int msm_gem_get_and_pin_iova(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace, uint64_t *iova) +int msm_gem_get_and_pin_iova(struct drm_gem_object *obj, struct drm_gpuvm *vm, + uint64_t *iova) { - return msm_gem_get_and_pin_iova_range(obj, aspace, iova, 0, U64_MAX); + return msm_gem_get_and_pin_iova_range(obj, vm, iova, 0, U64_MAX); } /* * Get an iova but don't pin it. Doesn't need a put because iovas are currently * valid for the life of the object */ -int msm_gem_get_iova(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace, uint64_t *iova) +int msm_gem_get_iova(struct drm_gem_object *obj, struct drm_gpuvm *vm, + uint64_t *iova) { - struct msm_gem_vma *vma; + struct drm_gpuva *vma; + struct drm_exec exec; int ret = 0; - msm_gem_lock(obj); - vma = get_vma_locked(obj, aspace, 0, U64_MAX); + msm_gem_lock_vm_and_obj(&exec, obj, vm); + vma = get_vma_locked(obj, vm, 0, U64_MAX); if (IS_ERR(vma)) { ret = PTR_ERR(vma); } else { - *iova = vma->iova; + *iova = vma->va.addr; } - msm_gem_unlock(obj); + drm_exec_fini(&exec); /* drop locks */ return ret; } static int clear_iova(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace) + struct drm_gpuvm *vm) { - struct msm_gem_vma *vma = lookup_vma(obj, aspace); + struct drm_gpuva *vma = lookup_vma(obj, vm); if (!vma) return 0; - msm_gem_vma_purge(vma); + msm_gem_vma_unmap(vma, NULL); msm_gem_vma_close(vma); - del_vma(vma); return 0; } @@ -657,44 +639,54 @@ static int clear_iova(struct drm_gem_object *obj, * Setting an iova of zero will clear the vma. */ int msm_gem_set_iova(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace, uint64_t iova) + struct drm_gpuvm *vm, uint64_t iova) { + struct drm_exec exec; int ret = 0; - msm_gem_lock(obj); + msm_gem_lock_vm_and_obj(&exec, obj, vm); if (!iova) { - ret = clear_iova(obj, aspace); + ret = clear_iova(obj, vm); } else { - struct msm_gem_vma *vma; - vma = get_vma_locked(obj, aspace, iova, iova + obj->size); + struct drm_gpuva *vma; + vma = get_vma_locked(obj, vm, iova, iova + obj->size); if (IS_ERR(vma)) { ret = PTR_ERR(vma); - } else if (GEM_WARN_ON(vma->iova != iova)) { - clear_iova(obj, aspace); + } else if (GEM_WARN_ON(vma->va.addr != iova)) { + clear_iova(obj, vm); ret = -EBUSY; } } - msm_gem_unlock(obj); + drm_exec_fini(&exec); /* drop locks */ return ret; } +static bool is_kms_vm(struct drm_gpuvm *vm) +{ + struct msm_drm_private *priv = vm->drm->dev_private; + + return priv->kms && (priv->kms->vm == vm); +} + /* * Unpin a iova by updating the reference counts. The memory isn't actually * purged until something else (shrinker, mm_notifier, destroy, etc) decides * to get rid of it */ -void msm_gem_unpin_iova(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace) +void msm_gem_unpin_iova(struct drm_gem_object *obj, struct drm_gpuvm *vm) { - struct msm_gem_vma *vma; + struct drm_gpuva *vma; + struct drm_exec exec; - msm_gem_lock(obj); - vma = lookup_vma(obj, aspace); - if (!GEM_WARN_ON(!vma)) { + msm_gem_lock_vm_and_obj(&exec, obj, vm); + vma = lookup_vma(obj, vm); + if (vma) { msm_gem_unpin_locked(obj); } - msm_gem_unlock(obj); + if (!is_kms_vm(vm)) + put_iova_spaces(obj, vm, true, "close"); + drm_exec_fini(&exec); /* drop locks */ } int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev, @@ -853,7 +845,7 @@ void msm_gem_purge(struct drm_gem_object *obj) GEM_WARN_ON(!is_purgeable(msm_obj)); /* Get rid of any iommu mapping(s): */ - put_iova_spaces(obj, true); + put_iova_spaces(obj, NULL, false, "purge"); msm_gem_vunmap(obj); @@ -861,8 +853,6 @@ void msm_gem_purge(struct drm_gem_object *obj) put_pages(obj); - put_iova_vmas(obj); - mutex_lock(&priv->lru.lock); /* A one-way transition: */ msm_obj->madv = __MSM_MADV_PURGED; @@ -893,7 +883,7 @@ void msm_gem_evict(struct drm_gem_object *obj) GEM_WARN_ON(is_unevictable(msm_obj)); /* Get rid of any iommu mapping(s): */ - put_iova_spaces(obj, false); + put_iova_spaces(obj, NULL, false, "evict"); drm_vma_node_unmap(&obj->vma_node, dev->anon_inode->i_mapping); @@ -920,7 +910,7 @@ bool msm_gem_active(struct drm_gem_object *obj) if (to_msm_bo(obj)->pin_count) return true; - return !dma_resv_test_signaled(obj->resv, dma_resv_usage_rw(true)); + return !dma_resv_test_signaled(obj->resv, DMA_RESV_USAGE_BOOKKEEP); } int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout) @@ -959,11 +949,11 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m, { struct msm_gem_object *msm_obj = to_msm_bo(obj); struct dma_resv *robj = obj->resv; - struct msm_gem_vma *vma; uint64_t off = drm_vma_node_start(&obj->vma_node); const char *madv; - msm_gem_lock(obj); + if (!msm_gem_trylock(obj)) + return; stats->all.count++; stats->all.size += obj->size; @@ -1002,31 +992,33 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m, seq_printf(m, " %08zu %9s %-32s\n", obj->size, madv, msm_obj->name); - if (!list_empty(&msm_obj->vmas)) { + if (!list_empty(&obj->gpuva.list)) { + struct drm_gpuvm_bo *vm_bo; seq_puts(m, " vmas:"); - list_for_each_entry(vma, &msm_obj->vmas, list) { - const char *name, *comm; - if (vma->aspace) { - struct msm_gem_address_space *aspace = vma->aspace; + drm_gem_for_each_gpuvm_bo (vm_bo, obj) { + struct drm_gpuva *vma; + + drm_gpuvm_bo_for_each_va (vma, vm_bo) { + const char *name, *comm; + struct msm_gem_vm *vm = to_msm_vm(vma->vm); struct task_struct *task = - get_pid_task(aspace->pid, PIDTYPE_PID); + get_pid_task(vm->pid, PIDTYPE_PID); if (task) { comm = kstrdup(task->comm, GFP_KERNEL); put_task_struct(task); } else { comm = NULL; } - name = aspace->name; - } else { - name = comm = NULL; + name = vm->base.name; + + seq_printf(m, " [%s%s%s: vm=%p, %08llx, %smapped]", + name, comm ? ":" : "", comm ? comm : "", + vma->vm, vma->va.addr, + to_msm_vma(vma)->mapped ? "" : "un"); + kfree(comm); } - seq_printf(m, " [%s%s%s: aspace=%p, %08llx,%s]", - name, comm ? ":" : "", comm ? comm : "", - vma->aspace, vma->iova, - vma->mapped ? "mapped" : "unmapped"); - kfree(comm); } seq_puts(m, "\n"); @@ -1067,12 +1059,46 @@ static void msm_gem_free_object(struct drm_gem_object *obj) struct msm_gem_object *msm_obj = to_msm_bo(obj); struct drm_device *dev = obj->dev; struct msm_drm_private *priv = dev->dev_private; + struct drm_exec exec; mutex_lock(&priv->obj_lock); list_del(&msm_obj->node); mutex_unlock(&priv->obj_lock); - put_iova_spaces(obj, true); + /* + * We need to lock any VMs the object is still attached to, but not + * the object itself (see explaination in msm_gem_assert_locked()), + * so just open-code this special case. + * + * Note that we skip the dance if we aren't attached to any VM. This + * is load bearing. The driver needs to support two usage models: + * + * 1. Legacy kernel managed VM: Userspace expects the VMA's to be + * implicitly torn down when the object is freed, the VMA's do + * not hold a hard reference to the BO. + * + * 2. VM_BIND, userspace managed VM: The VMA holds a reference to the + * BO. This can be dropped when the VM is closed and it's associated + * VMAs are torn down. (See msm_gem_vm_close()). + * + * In the latter case the last reference to a BO can be dropped while + * we already have the VM locked. It would have already been removed + * from the gpuva list, but lockdep doesn't know that. Or understand + * the differences between the two usage models. + */ + if (!list_empty(&obj->gpuva.list)) { + drm_exec_init(&exec, 0, 0); + drm_exec_until_all_locked (&exec) { + struct drm_gpuvm_bo *vm_bo; + drm_gem_for_each_gpuvm_bo (vm_bo, obj) { + drm_exec_lock_obj(&exec, + drm_gpuvm_resv_obj(vm_bo->vm)); + drm_exec_retry_on_contention(&exec); + } + } + put_iova_spaces(obj, NULL, true, "free"); + drm_exec_fini(&exec); /* drop locks */ + } if (drm_gem_is_imported(obj)) { GEM_WARN_ON(msm_obj->vaddr); @@ -1082,13 +1108,18 @@ static void msm_gem_free_object(struct drm_gem_object *obj) */ kvfree(msm_obj->pages); - put_iova_vmas(obj); - drm_prime_gem_destroy(obj, msm_obj->sgt); } else { msm_gem_vunmap(obj); put_pages(obj); - put_iova_vmas(obj); + } + + if (msm_obj->flags & MSM_BO_NO_SHARE) { + struct drm_gem_object *r_obj = + container_of(obj->resv, struct drm_gem_object, _resv); + + /* Drop reference we hold to shared resv obj: */ + drm_gem_object_put(r_obj); } drm_gem_object_release(obj); @@ -1123,6 +1154,15 @@ int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file, if (name) msm_gem_object_set_name(obj, "%s", name); + if (flags & MSM_BO_NO_SHARE) { + struct msm_context *ctx = file->driver_priv; + struct drm_gem_object *r_obj = drm_gpuvm_resv_obj(ctx->vm); + + drm_gem_object_get(r_obj); + + obj->resv = r_obj->resv; + } + ret = drm_gem_handle_create(file, obj, handle); /* drop reference from allocate - handle holds it now */ @@ -1155,6 +1195,7 @@ static const struct drm_gem_object_funcs msm_gem_object_funcs = { .free = msm_gem_free_object, .open = msm_gem_open, .close = msm_gem_close, + .export = msm_gem_prime_export, .pin = msm_gem_prime_pin, .unpin = msm_gem_prime_unpin, .get_sg_table = msm_gem_prime_get_sg_table, @@ -1194,7 +1235,6 @@ static int msm_gem_new_impl(struct drm_device *dev, msm_obj->madv = MSM_MADV_WILLNEED; INIT_LIST_HEAD(&msm_obj->node); - INIT_LIST_HEAD(&msm_obj->vmas); *obj = &msm_obj->base; (*obj)->funcs = &msm_gem_object_funcs; @@ -1207,19 +1247,10 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev, uint32_t size, uint32 struct msm_drm_private *priv = dev->dev_private; struct msm_gem_object *msm_obj; struct drm_gem_object *obj = NULL; - bool use_vram = false; int ret; size = PAGE_ALIGN(size); - if (!msm_use_mmu(dev)) - use_vram = true; - else if ((flags & (MSM_BO_STOLEN | MSM_BO_SCANOUT)) && priv->vram.size) - use_vram = true; - - if (GEM_WARN_ON(use_vram && !priv->vram.size)) - return ERR_PTR(-EINVAL); - /* Disallow zero sized objects as they make the underlying * infrastructure grumpy */ @@ -1232,44 +1263,16 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev, uint32_t size, uint32 msm_obj = to_msm_bo(obj); - if (use_vram) { - struct msm_gem_vma *vma; - struct page **pages; - - drm_gem_private_object_init(dev, obj, size); - - msm_gem_lock(obj); - - vma = add_vma(obj, NULL); - msm_gem_unlock(obj); - if (IS_ERR(vma)) { - ret = PTR_ERR(vma); - goto fail; - } - - to_msm_bo(obj)->vram_node = &vma->node; - - msm_gem_lock(obj); - pages = get_pages(obj); - msm_gem_unlock(obj); - if (IS_ERR(pages)) { - ret = PTR_ERR(pages); - goto fail; - } - - vma->iova = physaddr(obj); - } else { - ret = drm_gem_object_init(dev, obj, size); - if (ret) - goto fail; - /* - * Our buffers are kept pinned, so allocating them from the - * MOVABLE zone is a really bad idea, and conflicts with CMA. - * See comments above new_inode() why this is required _and_ - * expected if you're going to pin these pages. - */ - mapping_set_gfp_mask(obj->filp->f_mapping, GFP_HIGHUSER); - } + ret = drm_gem_object_init(dev, obj, size); + if (ret) + goto fail; + /* + * Our buffers are kept pinned, so allocating them from the + * MOVABLE zone is a really bad idea, and conflicts with CMA. + * See comments above new_inode() why this is required _and_ + * expected if you're going to pin these pages. + */ + mapping_set_gfp_mask(obj->filp->f_mapping, GFP_HIGHUSER); drm_gem_lru_move_tail(&priv->lru.unbacked, obj); @@ -1297,12 +1300,6 @@ struct drm_gem_object *msm_gem_import(struct drm_device *dev, uint32_t size; int ret, npages; - /* if we don't have IOMMU, don't bother pretending we can import: */ - if (!msm_use_mmu(dev)) { - DRM_DEV_ERROR(dev->dev, "cannot import without IOMMU\n"); - return ERR_PTR(-EINVAL); - } - size = PAGE_ALIGN(dmabuf->size); ret = msm_gem_new_impl(dev, size, MSM_BO_WC, &obj); @@ -1348,9 +1345,9 @@ fail: return ERR_PTR(ret); } -void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size, - uint32_t flags, struct msm_gem_address_space *aspace, - struct drm_gem_object **bo, uint64_t *iova) +void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size, uint32_t flags, + struct drm_gpuvm *vm, struct drm_gem_object **bo, + uint64_t *iova) { void *vaddr; struct drm_gem_object *obj = msm_gem_new(dev, size, flags); @@ -1360,14 +1357,14 @@ void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size, return ERR_CAST(obj); if (iova) { - ret = msm_gem_get_and_pin_iova(obj, aspace, iova); + ret = msm_gem_get_and_pin_iova(obj, vm, iova); if (ret) goto err; } vaddr = msm_gem_get_vaddr(obj); if (IS_ERR(vaddr)) { - msm_gem_unpin_iova(obj, aspace); + msm_gem_unpin_iova(obj, vm); ret = PTR_ERR(vaddr); goto err; } @@ -1383,14 +1380,13 @@ err: } -void msm_gem_kernel_put(struct drm_gem_object *bo, - struct msm_gem_address_space *aspace) +void msm_gem_kernel_put(struct drm_gem_object *bo, struct drm_gpuvm *vm) { if (IS_ERR_OR_NULL(bo)) return; msm_gem_put_vaddr(bo); - msm_gem_unpin_iova(bo, aspace); + msm_gem_unpin_iova(bo, vm); drm_gem_object_put(bo); } diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h index ba5c4ff76292..88239da1cd72 100644 --- a/drivers/gpu/drm/msm/msm_gem.h +++ b/drivers/gpu/drm/msm/msm_gem.h @@ -7,9 +7,11 @@ #ifndef __MSM_GEM_H__ #define __MSM_GEM_H__ +#include "msm_mmu.h" #include <linux/kref.h> #include <linux/dma-resv.h> #include "drm/drm_exec.h" +#include "drm/drm_gpuvm.h" #include "drm/gpu_scheduler.h" #include "msm_drv.h" @@ -22,56 +24,173 @@ #define MSM_BO_STOLEN 0x10000000 /* try to use stolen/splash memory */ #define MSM_BO_MAP_PRIV 0x20000000 /* use IOMMU_PRIV when mapping */ -struct msm_gem_address_space { - const char *name; - /* NOTE: mm managed at the page level, size is in # of pages - * and position mm_node->start is in # of pages: +/** + * struct msm_gem_vm_log_entry - An entry in the VM log + * + * For userspace managed VMs, a log of recent VM updates is tracked and + * captured in GPU devcore dumps, to aid debugging issues caused by (for + * example) incorrectly synchronized VM updates + */ +struct msm_gem_vm_log_entry { + const char *op; + uint64_t iova; + uint64_t range; + int queue_id; +}; + +/** + * struct msm_gem_vm - VM object + * + * A VM object representing a GPU (or display or GMU or ...) virtual address + * space. + * + * In the case of GPU, if per-process address spaces are supported, the address + * space is split into two VMs, which map to TTBR0 and TTBR1 in the SMMU. TTBR0 + * is used for userspace objects, and is unique per msm_context/drm_file, while + * TTBR1 is the same for all processes. (The kernel controlled ringbuffer and + * a few other kernel controlled buffers live in TTBR1.) + * + * The GPU TTBR0 vm can be managed by userspace or by the kernel, depending on + * whether userspace supports VM_BIND. All other vm's are managed by the kernel. + * (Managed by kernel means the kernel is responsible for VA allocation.) + * + * Note that because VM_BIND allows a given BO to be mapped multiple times in + * a VM, and therefore have multiple VMA's in a VM, there is an extra object + * provided by drm_gpuvm infrastructure.. the drm_gpuvm_bo, which is not + * embedded in any larger driver structure. The GEM object holds a list of + * drm_gpuvm_bo, which in turn holds a list of msm_gem_vma. A linked vma + * holds a reference to the vm_bo, and drops it when the vma is unlinked. + * So we just need to call drm_gpuvm_bo_obtain() to return a ref to an + * existing vm_bo, or create a new one. Once the vma is linked, the ref + * to the vm_bo can be dropped (since the vma is holding one). + */ +struct msm_gem_vm { + /** @base: Inherit from drm_gpuvm. */ + struct drm_gpuvm base; + + /** + * @sched: Scheduler used for asynchronous VM_BIND request. + * + * Unused for kernel managed VMs (where all operations are synchronous). + */ + struct drm_gpu_scheduler sched; + + /** + * @prealloc_throttle: Used to throttle VM_BIND ops if too much pre- + * allocated memory is in flight. + * + * Because we have to pre-allocate pgtable pages for the worst case + * (ie. new mappings do not share any PTEs with existing mappings) + * we could end up consuming a lot of resources transiently. The + * prealloc_throttle puts an upper bound on that. + */ + struct { + /** @wait: Notified when preallocated resources are released */ + wait_queue_head_t wait; + + /** + * @in_flight: The # of preallocated pgtable pages in-flight + * for queued VM_BIND jobs. + */ + atomic_t in_flight; + } prealloc_throttle; + + /** + * @mm: Memory management for kernel managed VA allocations + * + * Only used for kernel managed VMs, unused for user managed VMs. + * + * Protected by @mm_lock. */ struct drm_mm mm; - spinlock_t lock; /* Protects drm_mm node allocation/removal */ + + /** @mmu: The mmu object which manages the pgtables */ struct msm_mmu *mmu; - struct kref kref; - /* For address spaces associated with a specific process, this + /** @mmu_lock: Protects access to the mmu */ + struct mutex mmu_lock; + + /** + * @pid: For address spaces associated with a specific process, this * will be non-NULL: */ struct pid *pid; - /* @faults: the number of GPU hangs associated with this address space */ + /** @last_fence: Fence for last pending work scheduled on the VM */ + struct dma_fence *last_fence; + + /** @log: A log of recent VM updates */ + struct msm_gem_vm_log_entry *log; + + /** @log_shift: length of @log is (1 << @log_shift) */ + uint32_t log_shift; + + /** @log_idx: index of next @log entry to write */ + uint32_t log_idx; + + /** @faults: the number of GPU hangs associated with this address space */ int faults; - /** @va_start: lowest possible address to allocate */ - uint64_t va_start; + /** @managed: is this a kernel managed VM? */ + bool managed; - /** @va_size: the size of the address space (in bytes) */ - uint64_t va_size; + /** + * @unusable: True if the VM has turned unusable because something + * bad happened during an asynchronous request. + * + * We don't try to recover from such failures, because this implies + * informing userspace about the specific operation that failed, and + * hoping the userspace driver can replay things from there. This all + * sounds very complicated for little gain. + * + * Instead, we should just flag the VM as unusable, and fail any + * further request targeting this VM. + * + * As an analogy, this would be mapped to a VK_ERROR_DEVICE_LOST + * situation, where the logical device needs to be re-created. + */ + bool unusable; }; +#define to_msm_vm(x) container_of(x, struct msm_gem_vm, base) -struct msm_gem_address_space * -msm_gem_address_space_get(struct msm_gem_address_space *aspace); - -void msm_gem_address_space_put(struct msm_gem_address_space *aspace); +struct drm_gpuvm * +msm_gem_vm_create(struct drm_device *drm, struct msm_mmu *mmu, const char *name, + u64 va_start, u64 va_size, bool managed); -struct msm_gem_address_space * -msm_gem_address_space_create(struct msm_mmu *mmu, const char *name, - u64 va_start, u64 size); +void msm_gem_vm_close(struct drm_gpuvm *gpuvm); +void msm_gem_vm_unusable(struct drm_gpuvm *gpuvm); struct msm_fence_context; +#define MSM_VMA_DUMP (DRM_GPUVA_USERBITS << 0) + +/** + * struct msm_gem_vma - a VMA mapping + * + * Represents a combination of a GEM object plus a VM. + */ struct msm_gem_vma { + /** @base: inherit from drm_gpuva */ + struct drm_gpuva base; + + /** + * @node: mm node for VA allocation + * + * Only used by kernel managed VMs + */ struct drm_mm_node node; - uint64_t iova; - struct msm_gem_address_space *aspace; - struct list_head list; /* node in msm_gem_object::vmas */ + + /** @mapped: Is this VMA mapped? */ bool mapped; }; +#define to_msm_vma(x) container_of(x, struct msm_gem_vma, base) -struct msm_gem_vma *msm_gem_vma_new(struct msm_gem_address_space *aspace); -int msm_gem_vma_init(struct msm_gem_vma *vma, int size, - u64 range_start, u64 range_end); -void msm_gem_vma_purge(struct msm_gem_vma *vma); -int msm_gem_vma_map(struct msm_gem_vma *vma, int prot, struct sg_table *sgt, int size); -void msm_gem_vma_close(struct msm_gem_vma *vma); +struct drm_gpuva * +msm_gem_vma_new(struct drm_gpuvm *vm, struct drm_gem_object *obj, + u64 offset, u64 range_start, u64 range_end); +void msm_gem_vma_unmap(struct drm_gpuva *vma, const char *reason); +int msm_gem_vma_map(struct drm_gpuva *vma, int prot, struct sg_table *sgt); +void msm_gem_vma_close(struct drm_gpuva *vma); struct msm_gem_object { struct drm_gem_object base; @@ -100,13 +219,6 @@ struct msm_gem_object { struct sg_table *sgt; void *vaddr; - struct list_head vmas; /* list of msm_gem_vma */ - - /* For physically contiguous buffers. Used when we don't have - * an IOMMU. Also used for stolen/splashscreen buffer. - */ - struct drm_mm_node *vram_node; - char name[32]; /* Identifier to print for the debugfs files */ /* userspace metadata backchannel */ @@ -119,27 +231,56 @@ struct msm_gem_object { * Protected by LRU lock. */ int pin_count; + + /** + * @vma_ref: Reference count of VMA users. + * + * With the vm_bo/vma holding a reference to the GEM object, we'd + * otherwise have to actively tear down a VMA when, for example, + * a buffer is unpinned for scanout, vs. the pre-drm_gpuvm approach + * where a VMA did not hold a reference to the BO, but instead was + * implicitly torn down when the BO was freed. + * + * To regain the lazy VMA teardown, we use the @vma_ref. It is + * incremented for any of the following: + * + * 1) the BO is exported as a dma_buf + * 2) the BO has open userspace handle + * + * All of those conditions will hold an reference to the BO, + * preventing it from being freed. So lazily keeping around the + * VMA will not prevent the BO from being freed. (Or rather, the + * reference loop is harmless in this case.) + * + * When the @vma_ref drops to zero, then kms->vm VMA will be + * torn down. + */ + atomic_t vma_ref; }; #define to_msm_bo(x) container_of(x, struct msm_gem_object, base) +void msm_gem_vma_get(struct drm_gem_object *obj); +void msm_gem_vma_put(struct drm_gem_object *obj); + uint64_t msm_gem_mmap_offset(struct drm_gem_object *obj); -int msm_gem_pin_vma_locked(struct drm_gem_object *obj, struct msm_gem_vma *vma); +int msm_gem_prot(struct drm_gem_object *obj); +int msm_gem_pin_vma_locked(struct drm_gem_object *obj, struct drm_gpuva *vma); void msm_gem_unpin_locked(struct drm_gem_object *obj); void msm_gem_unpin_active(struct drm_gem_object *obj); -struct msm_gem_vma *msm_gem_get_vma_locked(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace); -int msm_gem_get_iova(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace, uint64_t *iova); -int msm_gem_set_iova(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace, uint64_t iova); +struct drm_gpuva *msm_gem_get_vma_locked(struct drm_gem_object *obj, + struct drm_gpuvm *vm); +int msm_gem_get_iova(struct drm_gem_object *obj, struct drm_gpuvm *vm, + uint64_t *iova); +int msm_gem_set_iova(struct drm_gem_object *obj, struct drm_gpuvm *vm, + uint64_t iova); int msm_gem_get_and_pin_iova_range(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace, uint64_t *iova, - u64 range_start, u64 range_end); -int msm_gem_get_and_pin_iova(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace, uint64_t *iova); -void msm_gem_unpin_iova(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace); + struct drm_gpuvm *vm, uint64_t *iova, + u64 range_start, u64 range_end); +int msm_gem_get_and_pin_iova(struct drm_gem_object *obj, struct drm_gpuvm *vm, + uint64_t *iova); +void msm_gem_unpin_iova(struct drm_gem_object *obj, struct drm_gpuvm *vm); void msm_gem_pin_obj_locked(struct drm_gem_object *obj); +struct page **msm_gem_get_pages_locked(struct drm_gem_object *obj, unsigned madv); struct page **msm_gem_pin_pages_locked(struct drm_gem_object *obj); void msm_gem_unpin_pages_locked(struct drm_gem_object *obj); int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev, @@ -159,11 +300,10 @@ int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file, uint32_t size, uint32_t flags, uint32_t *handle, char *name); struct drm_gem_object *msm_gem_new(struct drm_device *dev, uint32_t size, uint32_t flags); -void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size, - uint32_t flags, struct msm_gem_address_space *aspace, - struct drm_gem_object **bo, uint64_t *iova); -void msm_gem_kernel_put(struct drm_gem_object *bo, - struct msm_gem_address_space *aspace); +void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size, uint32_t flags, + struct drm_gpuvm *vm, struct drm_gem_object **bo, + uint64_t *iova); +void msm_gem_kernel_put(struct drm_gem_object *bo, struct drm_gpuvm *vm); struct drm_gem_object *msm_gem_import(struct drm_device *dev, struct dma_buf *dmabuf, struct sg_table *sgt); __printf(2, 3) @@ -188,6 +328,12 @@ msm_gem_lock(struct drm_gem_object *obj) dma_resv_lock(obj->resv, NULL); } +static inline bool __must_check +msm_gem_trylock(struct drm_gem_object *obj) +{ + return dma_resv_trylock(obj->resv); +} + static inline int msm_gem_lock_interruptible(struct drm_gem_object *obj) { @@ -200,6 +346,37 @@ msm_gem_unlock(struct drm_gem_object *obj) dma_resv_unlock(obj->resv); } +/** + * msm_gem_lock_vm_and_obj() - Helper to lock an obj + VM + * @exec: the exec context helper which will be initalized + * @obj: the GEM object to lock + * @vm: the VM to lock + * + * Operations which modify a VM frequently need to lock both the VM and + * the object being mapped/unmapped/etc. This helper uses drm_exec to + * acquire both locks, dealing with potential deadlock/backoff scenarios + * which arise when multiple locks are involved. + */ +static inline int +msm_gem_lock_vm_and_obj(struct drm_exec *exec, + struct drm_gem_object *obj, + struct drm_gpuvm *vm) +{ + int ret = 0; + + drm_exec_init(exec, 0, 2); + drm_exec_until_all_locked (exec) { + ret = drm_exec_lock_obj(exec, drm_gpuvm_resv_obj(vm)); + if (!ret && (obj->resv != drm_gpuvm_resv(vm))) + ret = drm_exec_lock_obj(exec, obj); + drm_exec_retry_on_contention(exec); + if (GEM_WARN_ON(ret)) + break; + } + + return ret; +} + static inline void msm_gem_assert_locked(struct drm_gem_object *obj) { @@ -257,7 +434,7 @@ struct msm_gem_submit { struct kref ref; struct drm_device *dev; struct msm_gpu *gpu; - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; struct list_head node; /* node in ring submit list */ struct drm_exec exec; uint32_t seqno; /* Sequence number of the submit on the ring */ @@ -297,6 +474,7 @@ struct msm_gem_submit { struct drm_gem_object *obj; uint32_t handle; }; + struct drm_gpuvm_bo *vm_bo; uint64_t iova; } bos[]; }; @@ -320,14 +498,4 @@ static inline void msm_gem_submit_put(struct msm_gem_submit *submit) void msm_submit_retire(struct msm_gem_submit *submit); -/* helper to determine of a buffer in submit should be dumped, used for both - * devcoredump and debugfs cmdstream dumping: - */ -static inline bool -should_dump(struct msm_gem_submit *submit, int idx) -{ - extern bool rd_full; - return rd_full || (submit->bos[idx].flags & MSM_SUBMIT_BO_DUMP); -} - #endif /* __MSM_GEM_H__ */ diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c index 2e37913d5a6a..c0a33ac839cb 100644 --- a/drivers/gpu/drm/msm/msm_gem_prime.c +++ b/drivers/gpu/drm/msm/msm_gem_prime.c @@ -6,6 +6,7 @@ #include <linux/dma-buf.h> +#include <drm/drm_drv.h> #include <drm/drm_prime.h> #include "msm_drv.h" @@ -16,6 +17,9 @@ struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj) struct msm_gem_object *msm_obj = to_msm_bo(obj); int npages = obj->size >> PAGE_SHIFT; + if (msm_obj->flags & MSM_BO_NO_SHARE) + return ERR_PTR(-EINVAL); + if (WARN_ON(!msm_obj->pages)) /* should have already pinned! */ return ERR_PTR(-ENOMEM); @@ -39,12 +43,70 @@ void msm_gem_prime_vunmap(struct drm_gem_object *obj, struct iosys_map *map) msm_gem_put_vaddr_locked(obj); } +static void msm_gem_dmabuf_release(struct dma_buf *dma_buf) +{ + struct drm_gem_object *obj = dma_buf->priv; + + msm_gem_vma_put(obj); + drm_gem_dmabuf_release(dma_buf); +} + +static const struct dma_buf_ops msm_gem_prime_dmabuf_ops = { + .attach = drm_gem_map_attach, + .detach = drm_gem_map_detach, + .map_dma_buf = drm_gem_map_dma_buf, + .unmap_dma_buf = drm_gem_unmap_dma_buf, + .release = msm_gem_dmabuf_release, + .mmap = drm_gem_dmabuf_mmap, + .vmap = drm_gem_dmabuf_vmap, + .vunmap = drm_gem_dmabuf_vunmap, +}; + +struct drm_gem_object *msm_gem_prime_import(struct drm_device *dev, + struct dma_buf *buf) +{ + if (buf->ops == &msm_gem_prime_dmabuf_ops) { + struct drm_gem_object *obj = buf->priv; + if (obj->dev == dev) { + /* + * Importing dmabuf exported from our own gem increases + * refcount on gem itself instead of f_count of dmabuf. + */ + drm_gem_object_get(obj); + return obj; + } + } + + return drm_gem_prime_import(dev, buf); +} + struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sg) { return msm_gem_import(dev, attach->dmabuf, sg); } +struct dma_buf *msm_gem_prime_export(struct drm_gem_object *obj, int flags) +{ + if (to_msm_bo(obj)->flags & MSM_BO_NO_SHARE) + return ERR_PTR(-EPERM); + + msm_gem_vma_get(obj); + + struct drm_device *dev = obj->dev; + struct dma_buf_export_info exp_info = { + .exp_name = KBUILD_MODNAME, /* white lie for debug */ + .owner = dev->driver->fops->owner, + .ops = &msm_gem_prime_dmabuf_ops, + .size = obj->size, + .flags = flags, + .priv = obj, + .resv = obj->resv, + }; + + return drm_gem_dmabuf_export(dev, &exp_info); +} + int msm_gem_prime_pin(struct drm_gem_object *obj) { struct page **pages; @@ -53,6 +115,9 @@ int msm_gem_prime_pin(struct drm_gem_object *obj) if (drm_gem_is_imported(obj)) return 0; + if (to_msm_bo(obj)->flags & MSM_BO_NO_SHARE) + return -EINVAL; + pages = msm_gem_pin_pages_locked(obj); if (IS_ERR(pages)) ret = PTR_ERR(pages); diff --git a/drivers/gpu/drm/msm/msm_gem_shrinker.c b/drivers/gpu/drm/msm/msm_gem_shrinker.c index 07ca4ddfe4e3..1039e3c0a47b 100644 --- a/drivers/gpu/drm/msm/msm_gem_shrinker.c +++ b/drivers/gpu/drm/msm/msm_gem_shrinker.c @@ -44,7 +44,76 @@ msm_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc) } static bool -purge(struct drm_gem_object *obj) +with_vm_locks(struct ww_acquire_ctx *ticket, + void (*fn)(struct drm_gem_object *obj), + struct drm_gem_object *obj) +{ + /* + * Track last locked entry for for unwinding locks in error and + * success paths + */ + struct drm_gpuvm_bo *vm_bo, *last_locked = NULL; + int ret = 0; + + drm_gem_for_each_gpuvm_bo (vm_bo, obj) { + struct dma_resv *resv = drm_gpuvm_resv(vm_bo->vm); + + if (resv == obj->resv) + continue; + + ret = dma_resv_lock(resv, ticket); + + /* + * Since we already skip the case when the VM and obj + * share a resv (ie. _NO_SHARE objs), we don't expect + * to hit a double-locking scenario... which the lock + * unwinding cannot really cope with. + */ + WARN_ON(ret == -EALREADY); + + /* + * Don't bother with slow-lock / backoff / retry sequence, + * if we can't get the lock just give up and move on to + * the next object. + */ + if (ret) + goto out_unlock; + + /* + * Hold a ref to prevent the vm_bo from being freed + * and removed from the obj's gpuva list, as that would + * would result in missing the unlock below + */ + drm_gpuvm_bo_get(vm_bo); + + last_locked = vm_bo; + } + + fn(obj); + +out_unlock: + if (last_locked) { + drm_gem_for_each_gpuvm_bo (vm_bo, obj) { + struct dma_resv *resv = drm_gpuvm_resv(vm_bo->vm); + + if (resv == obj->resv) + continue; + + dma_resv_unlock(resv); + + /* Drop the ref taken while locking: */ + drm_gpuvm_bo_put(vm_bo); + + if (last_locked == vm_bo) + break; + } + } + + return ret == 0; +} + +static bool +purge(struct drm_gem_object *obj, struct ww_acquire_ctx *ticket) { if (!is_purgeable(to_msm_bo(obj))) return false; @@ -52,13 +121,11 @@ purge(struct drm_gem_object *obj) if (msm_gem_active(obj)) return false; - msm_gem_purge(obj); - - return true; + return with_vm_locks(ticket, msm_gem_purge, obj); } static bool -evict(struct drm_gem_object *obj) +evict(struct drm_gem_object *obj, struct ww_acquire_ctx *ticket) { if (is_unevictable(to_msm_bo(obj))) return false; @@ -66,43 +133,42 @@ evict(struct drm_gem_object *obj) if (msm_gem_active(obj)) return false; - msm_gem_evict(obj); - - return true; + return with_vm_locks(ticket, msm_gem_evict, obj); } static bool wait_for_idle(struct drm_gem_object *obj) { - enum dma_resv_usage usage = dma_resv_usage_rw(true); + enum dma_resv_usage usage = DMA_RESV_USAGE_BOOKKEEP; return dma_resv_wait_timeout(obj->resv, usage, false, 10) > 0; } static bool -active_purge(struct drm_gem_object *obj) +active_purge(struct drm_gem_object *obj, struct ww_acquire_ctx *ticket) { if (!wait_for_idle(obj)) return false; - return purge(obj); + return purge(obj, ticket); } static bool -active_evict(struct drm_gem_object *obj) +active_evict(struct drm_gem_object *obj, struct ww_acquire_ctx *ticket) { if (!wait_for_idle(obj)) return false; - return evict(obj); + return evict(obj, ticket); } static unsigned long msm_gem_shrinker_scan(struct shrinker *shrinker, struct shrink_control *sc) { struct msm_drm_private *priv = shrinker->private_data; + struct ww_acquire_ctx ticket; struct { struct drm_gem_lru *lru; - bool (*shrink)(struct drm_gem_object *obj); + bool (*shrink)(struct drm_gem_object *obj, struct ww_acquire_ctx *ticket); bool cond; unsigned long freed; unsigned long remaining; @@ -122,8 +188,9 @@ msm_gem_shrinker_scan(struct shrinker *shrinker, struct shrink_control *sc) continue; stages[i].freed = drm_gem_lru_scan(stages[i].lru, nr, - &stages[i].remaining, - stages[i].shrink); + &stages[i].remaining, + stages[i].shrink, + &ticket); nr -= stages[i].freed; freed += stages[i].freed; remaining += stages[i].remaining; @@ -164,7 +231,7 @@ msm_gem_shrinker_shrink(struct drm_device *dev, unsigned long nr_to_scan) static const int vmap_shrink_limit = 15; static bool -vmap_shrink(struct drm_gem_object *obj) +vmap_shrink(struct drm_gem_object *obj, struct ww_acquire_ctx *ticket) { if (!is_vunmapable(to_msm_bo(obj))) return false; @@ -192,7 +259,8 @@ msm_gem_shrinker_vmap(struct notifier_block *nb, unsigned long event, void *ptr) unmapped += drm_gem_lru_scan(lrus[idx], vmap_shrink_limit - unmapped, &remaining, - vmap_shrink); + vmap_shrink, + NULL); } *(unsigned long *)ptr += unmapped; diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index d9be0fe3d674..5f8e939a5906 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -4,6 +4,7 @@ * Author: Rob Clark <robdclark@gmail.com> */ +#include <linux/dma-fence-unwrap.h> #include <linux/file.h> #include <linux/sync_file.h> #include <linux/uaccess.h> @@ -16,6 +17,7 @@ #include "msm_gpu.h" #include "msm_gem.h" #include "msm_gpu_trace.h" +#include "msm_syncobj.h" /* For userspace errors, use DRM_UT_DRIVER.. so that userspace can enable * error msgs for debugging, but we don't spam dmesg by default @@ -64,7 +66,7 @@ static struct msm_gem_submit *submit_create(struct drm_device *dev, kref_init(&submit->ref); submit->dev = dev; - submit->aspace = queue->ctx->aspace; + submit->vm = msm_context_vm(dev, queue->ctx); submit->gpu = gpu; submit->cmd = (void *)&submit->bos[nr_bos]; submit->queue = queue; @@ -86,6 +88,15 @@ void __msm_gem_submit_destroy(struct kref *kref) container_of(kref, struct msm_gem_submit, ref); unsigned i; + /* + * In error paths, we could unref the submit without calling + * drm_sched_entity_push_job(), so msm_job_free() will never + * get called. Since drm_sched_job_cleanup() will NULL out + * s_fence, we can use that to detect this case. + */ + if (submit->base.s_fence) + drm_sched_job_cleanup(&submit->base); + if (submit->fence_id) { spin_lock(&submit->queue->idr_lock); idr_remove(&submit->queue->fence_idr, submit->fence_id); @@ -183,6 +194,7 @@ out: static int submit_lookup_cmds(struct msm_gem_submit *submit, struct drm_msm_gem_submit *args, struct drm_file *file) { + struct msm_context *ctx = file->driver_priv; unsigned i; size_t sz; int ret = 0; @@ -214,6 +226,20 @@ static int submit_lookup_cmds(struct msm_gem_submit *submit, goto out; } + if (msm_context_is_vmbind(ctx)) { + if (submit_cmd.nr_relocs) { + ret = SUBMIT_ERROR(EINVAL, submit, "nr_relocs must be zero"); + goto out; + } + + if (submit_cmd.submit_idx || submit_cmd.submit_offset) { + ret = SUBMIT_ERROR(EINVAL, submit, "submit_idx/offset must be zero"); + goto out; + } + + submit->cmd[i].iova = submit_cmd.iova; + } + submit->cmd[i].type = submit_cmd.type; submit->cmd[i].size = submit_cmd.size / 4; submit->cmd[i].offset = submit_cmd.submit_offset / 4; @@ -248,24 +274,48 @@ out: /* This is where we make sure all the bo's are reserved and pin'd: */ static int submit_lock_objects(struct msm_gem_submit *submit) { + unsigned flags = DRM_EXEC_INTERRUPTIBLE_WAIT; + struct drm_exec *exec = &submit->exec; int ret; - drm_exec_init(&submit->exec, DRM_EXEC_INTERRUPTIBLE_WAIT, submit->nr_bos); + if (msm_context_is_vmbind(submit->queue->ctx)) { + flags |= DRM_EXEC_IGNORE_DUPLICATES; + + drm_exec_init(&submit->exec, flags, submit->nr_bos); + + drm_exec_until_all_locked (&submit->exec) { + ret = drm_gpuvm_prepare_vm(submit->vm, exec, 1); + drm_exec_retry_on_contention(exec); + if (ret) + return ret; + + ret = drm_gpuvm_prepare_objects(submit->vm, exec, 1); + drm_exec_retry_on_contention(exec); + if (ret) + return ret; + } + + return 0; + } + + drm_exec_init(&submit->exec, flags, submit->nr_bos); drm_exec_until_all_locked (&submit->exec) { + ret = drm_exec_lock_obj(&submit->exec, + drm_gpuvm_resv_obj(submit->vm)); + drm_exec_retry_on_contention(&submit->exec); + if (ret) + return ret; for (unsigned i = 0; i < submit->nr_bos; i++) { struct drm_gem_object *obj = submit->bos[i].obj; ret = drm_exec_prepare_obj(&submit->exec, obj, 1); drm_exec_retry_on_contention(&submit->exec); if (ret) - goto error; + return ret; } } return 0; - -error: - return ret; } static int submit_fence_sync(struct msm_gem_submit *submit) @@ -300,10 +350,10 @@ static int submit_pin_objects(struct msm_gem_submit *submit) for (i = 0; i < submit->nr_bos; i++) { struct drm_gem_object *obj = submit->bos[i].obj; - struct msm_gem_vma *vma; + struct drm_gpuva *vma; /* if locking succeeded, pin bo: */ - vma = msm_gem_get_vma_locked(obj, submit->aspace); + vma = msm_gem_get_vma_locked(obj, submit->vm); if (IS_ERR(vma)) { ret = PTR_ERR(vma); break; @@ -313,7 +363,8 @@ static int submit_pin_objects(struct msm_gem_submit *submit) if (ret) break; - submit->bos[i].iova = vma->iova; + submit->bos[i].vm_bo = drm_gpuvm_bo_get(vma->vm_bo); + submit->bos[i].iova = vma->va.addr; } /* @@ -350,9 +401,18 @@ static void submit_unpin_objects(struct msm_gem_submit *submit) static void submit_attach_object_fences(struct msm_gem_submit *submit) { - int i; + struct msm_gem_vm *vm = to_msm_vm(submit->vm); + struct dma_fence *last_fence; + + if (msm_context_is_vmbind(submit->queue->ctx)) { + drm_gpuvm_resv_add_fence(submit->vm, &submit->exec, + submit->user_fence, + DMA_RESV_USAGE_BOOKKEEP, + DMA_RESV_USAGE_BOOKKEEP); + return; + } - for (i = 0; i < submit->nr_bos; i++) { + for (unsigned i = 0; i < submit->nr_bos; i++) { struct drm_gem_object *obj = submit->bos[i].obj; if (submit->bos[i].flags & MSM_SUBMIT_BO_WRITE) @@ -362,6 +422,10 @@ static void submit_attach_object_fences(struct msm_gem_submit *submit) dma_resv_add_fence(obj->resv, submit->user_fence, DMA_RESV_USAGE_READ); } + + last_fence = vm->last_fence; + vm->last_fence = dma_fence_unwrap_merge(submit->user_fence, last_fence); + dma_fence_put(last_fence); } static int submit_bo(struct msm_gem_submit *submit, uint32_t idx, @@ -450,14 +514,14 @@ out: */ static void submit_cleanup(struct msm_gem_submit *submit, bool error) { + if (submit->exec.objects) + drm_exec_fini(&submit->exec); + if (error) { submit_unpin_objects(submit); /* job wasn't enqueued to scheduler, so early retirement: */ msm_submit_retire(submit); } - - if (submit->exec.objects) - drm_exec_fini(&submit->exec); } void msm_submit_retire(struct msm_gem_submit *submit) @@ -466,190 +530,29 @@ void msm_submit_retire(struct msm_gem_submit *submit) for (i = 0; i < submit->nr_bos; i++) { struct drm_gem_object *obj = submit->bos[i].obj; + struct drm_gpuvm_bo *vm_bo = submit->bos[i].vm_bo; + msm_gem_lock(obj); + drm_gpuvm_bo_put(vm_bo); + msm_gem_unlock(obj); drm_gem_object_put(obj); } } -struct msm_submit_post_dep { - struct drm_syncobj *syncobj; - uint64_t point; - struct dma_fence_chain *chain; -}; - -static struct drm_syncobj **msm_parse_deps(struct msm_gem_submit *submit, - struct drm_file *file, - uint64_t in_syncobjs_addr, - uint32_t nr_in_syncobjs, - size_t syncobj_stride) -{ - struct drm_syncobj **syncobjs = NULL; - struct drm_msm_gem_submit_syncobj syncobj_desc = {0}; - int ret = 0; - uint32_t i, j; - - syncobjs = kcalloc(nr_in_syncobjs, sizeof(*syncobjs), - GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY); - if (!syncobjs) - return ERR_PTR(-ENOMEM); - - for (i = 0; i < nr_in_syncobjs; ++i) { - uint64_t address = in_syncobjs_addr + i * syncobj_stride; - - if (copy_from_user(&syncobj_desc, - u64_to_user_ptr(address), - min(syncobj_stride, sizeof(syncobj_desc)))) { - ret = -EFAULT; - break; - } - - if (syncobj_desc.point && - !drm_core_check_feature(submit->dev, DRIVER_SYNCOBJ_TIMELINE)) { - ret = SUBMIT_ERROR(EOPNOTSUPP, submit, "syncobj timeline unsupported"); - break; - } - - if (syncobj_desc.flags & ~MSM_SUBMIT_SYNCOBJ_FLAGS) { - ret = SUBMIT_ERROR(EINVAL, submit, "invalid syncobj flags: %x", syncobj_desc.flags); - break; - } - - ret = drm_sched_job_add_syncobj_dependency(&submit->base, file, - syncobj_desc.handle, syncobj_desc.point); - if (ret) - break; - - if (syncobj_desc.flags & MSM_SUBMIT_SYNCOBJ_RESET) { - syncobjs[i] = - drm_syncobj_find(file, syncobj_desc.handle); - if (!syncobjs[i]) { - ret = SUBMIT_ERROR(EINVAL, submit, "invalid syncobj handle: %u", i); - break; - } - } - } - - if (ret) { - for (j = 0; j <= i; ++j) { - if (syncobjs[j]) - drm_syncobj_put(syncobjs[j]); - } - kfree(syncobjs); - return ERR_PTR(ret); - } - return syncobjs; -} - -static void msm_reset_syncobjs(struct drm_syncobj **syncobjs, - uint32_t nr_syncobjs) -{ - uint32_t i; - - for (i = 0; syncobjs && i < nr_syncobjs; ++i) { - if (syncobjs[i]) - drm_syncobj_replace_fence(syncobjs[i], NULL); - } -} - -static struct msm_submit_post_dep *msm_parse_post_deps(struct drm_device *dev, - struct drm_file *file, - uint64_t syncobjs_addr, - uint32_t nr_syncobjs, - size_t syncobj_stride) -{ - struct msm_submit_post_dep *post_deps; - struct drm_msm_gem_submit_syncobj syncobj_desc = {0}; - int ret = 0; - uint32_t i, j; - - post_deps = kcalloc(nr_syncobjs, sizeof(*post_deps), - GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY); - if (!post_deps) - return ERR_PTR(-ENOMEM); - - for (i = 0; i < nr_syncobjs; ++i) { - uint64_t address = syncobjs_addr + i * syncobj_stride; - - if (copy_from_user(&syncobj_desc, - u64_to_user_ptr(address), - min(syncobj_stride, sizeof(syncobj_desc)))) { - ret = -EFAULT; - break; - } - - post_deps[i].point = syncobj_desc.point; - - if (syncobj_desc.flags) { - ret = UERR(EINVAL, dev, "invalid syncobj flags"); - break; - } - - if (syncobj_desc.point) { - if (!drm_core_check_feature(dev, - DRIVER_SYNCOBJ_TIMELINE)) { - ret = UERR(EOPNOTSUPP, dev, "syncobj timeline unsupported"); - break; - } - - post_deps[i].chain = dma_fence_chain_alloc(); - if (!post_deps[i].chain) { - ret = -ENOMEM; - break; - } - } - - post_deps[i].syncobj = - drm_syncobj_find(file, syncobj_desc.handle); - if (!post_deps[i].syncobj) { - ret = UERR(EINVAL, dev, "invalid syncobj handle"); - break; - } - } - - if (ret) { - for (j = 0; j <= i; ++j) { - dma_fence_chain_free(post_deps[j].chain); - if (post_deps[j].syncobj) - drm_syncobj_put(post_deps[j].syncobj); - } - - kfree(post_deps); - return ERR_PTR(ret); - } - - return post_deps; -} - -static void msm_process_post_deps(struct msm_submit_post_dep *post_deps, - uint32_t count, struct dma_fence *fence) -{ - uint32_t i; - - for (i = 0; post_deps && i < count; ++i) { - if (post_deps[i].chain) { - drm_syncobj_add_point(post_deps[i].syncobj, - post_deps[i].chain, - fence, post_deps[i].point); - post_deps[i].chain = NULL; - } else { - drm_syncobj_replace_fence(post_deps[i].syncobj, - fence); - } - } -} - int msm_ioctl_gem_submit(struct drm_device *dev, void *data, struct drm_file *file) { struct msm_drm_private *priv = dev->dev_private; struct drm_msm_gem_submit *args = data; - struct msm_file_private *ctx = file->driver_priv; + struct msm_context *ctx = file->driver_priv; struct msm_gem_submit *submit = NULL; struct msm_gpu *gpu = priv->gpu; struct msm_gpu_submitqueue *queue; struct msm_ringbuffer *ring; - struct msm_submit_post_dep *post_deps = NULL; + struct msm_syncobj_post_dep *post_deps = NULL; struct drm_syncobj **syncobjs_to_reset = NULL; + struct sync_file *sync_file = NULL; + unsigned cmds_to_parse; int out_fence_fd = -1; unsigned i; int ret; @@ -660,10 +563,8 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, if (args->pad) return -EINVAL; - if (unlikely(!ctx->aspace) && !capable(CAP_SYS_RAWIO)) { - DRM_ERROR_RATELIMITED("IOMMU support or CAP_SYS_RAWIO required!\n"); - return -EPERM; - } + if (to_msm_vm(ctx->vm)->unusable) + return UERR(EPIPE, dev, "context is unusable"); /* for now, we just have 3d pipe.. eventually this would need to * be more clever to dispatch to appropriate gpu module: @@ -684,6 +585,11 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, if (!queue) return -ENOENT; + if (queue->flags & MSM_SUBMITQUEUE_VM_BIND) { + ret = UERR(EINVAL, dev, "Invalid queue type"); + goto out_post_unlock; + } + ring = gpu->rb[queue->ring_nr]; if (args->flags & MSM_SUBMIT_FENCE_FD_OUT) { @@ -727,10 +633,10 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, } if (args->flags & MSM_SUBMIT_SYNCOBJ_IN) { - syncobjs_to_reset = msm_parse_deps(submit, file, - args->in_syncobjs, - args->nr_in_syncobjs, - args->syncobj_stride); + syncobjs_to_reset = msm_syncobj_parse_deps(dev, &submit->base, + file, args->in_syncobjs, + args->nr_in_syncobjs, + args->syncobj_stride); if (IS_ERR(syncobjs_to_reset)) { ret = PTR_ERR(syncobjs_to_reset); goto out_unlock; @@ -738,10 +644,10 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, } if (args->flags & MSM_SUBMIT_SYNCOBJ_OUT) { - post_deps = msm_parse_post_deps(dev, file, - args->out_syncobjs, - args->nr_out_syncobjs, - args->syncobj_stride); + post_deps = msm_syncobj_parse_post_deps(dev, file, + args->out_syncobjs, + args->nr_out_syncobjs, + args->syncobj_stride); if (IS_ERR(post_deps)) { ret = PTR_ERR(post_deps); goto out_unlock; @@ -771,7 +677,9 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, if (ret) goto out; - for (i = 0; i < args->nr_cmds; i++) { + cmds_to_parse = msm_context_is_vmbind(ctx) ? 0 : args->nr_cmds; + + for (i = 0; i < cmds_to_parse; i++) { struct drm_gem_object *obj; uint64_t iova; @@ -802,7 +710,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, goto out; } - submit->nr_cmds = i; + submit->nr_cmds = args->nr_cmds; idr_preload(GFP_KERNEL); @@ -860,7 +768,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, } if (ret == 0 && args->flags & MSM_SUBMIT_FENCE_FD_OUT) { - struct sync_file *sync_file = sync_file_create(submit->user_fence); + sync_file = sync_file_create(submit->user_fence); if (!sync_file) { ret = -ENOMEM; } else { @@ -874,6 +782,18 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, submit_attach_object_fences(submit); + if (msm_context_is_vmbind(ctx)) { + /* + * If we are not using VM_BIND, submit_pin_vmas() will validate + * just the BOs attached to the submit. In that case we don't + * need to validate the _entire_ vm, because userspace tracked + * what BOs are associated with the submit. + */ + ret = drm_gpuvm_validate(submit->vm, &submit->exec); + if (ret) + goto out; + } + /* The scheduler owns a ref now: */ msm_gem_submit_get(submit); @@ -884,18 +804,19 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, args->fence = submit->fence_id; queue->last_fence = submit->fence_id; - msm_reset_syncobjs(syncobjs_to_reset, args->nr_in_syncobjs); - msm_process_post_deps(post_deps, args->nr_out_syncobjs, - submit->user_fence); - + msm_syncobj_reset(syncobjs_to_reset, args->nr_in_syncobjs); + msm_syncobj_process_post_deps(post_deps, args->nr_out_syncobjs, submit->user_fence); out: submit_cleanup(submit, !!ret); out_unlock: mutex_unlock(&queue->lock); out_post_unlock: - if (ret && (out_fence_fd >= 0)) + if (ret && (out_fence_fd >= 0)) { put_unused_fd(out_fence_fd); + if (sync_file) + fput(sync_file->file); + } if (!IS_ERR_OR_NULL(submit)) { msm_gem_submit_put(submit); diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c index 11e842dda73c..3cd8562a5109 100644 --- a/drivers/gpu/drm/msm/msm_gem_vma.c +++ b/drivers/gpu/drm/msm/msm_gem_vma.c @@ -4,73 +4,319 @@ * Author: Rob Clark <robdclark@gmail.com> */ +#include "drm/drm_file.h" +#include "drm/msm_drm.h" +#include "linux/file.h" +#include "linux/sync_file.h" + #include "msm_drv.h" -#include "msm_fence.h" #include "msm_gem.h" +#include "msm_gpu.h" #include "msm_mmu.h" +#include "msm_syncobj.h" + +#define vm_dbg(fmt, ...) pr_debug("%s:%d: "fmt"\n", __func__, __LINE__, ##__VA_ARGS__) + +static uint vm_log_shift = 0; +MODULE_PARM_DESC(vm_log_shift, "Length of VM op log"); +module_param_named(vm_log_shift, vm_log_shift, uint, 0600); + +/** + * struct msm_vm_map_op - create new pgtable mapping + */ +struct msm_vm_map_op { + /** @iova: start address for mapping */ + uint64_t iova; + /** @range: size of the region to map */ + uint64_t range; + /** @offset: offset into @sgt to map */ + uint64_t offset; + /** @sgt: pages to map, or NULL for a PRR mapping */ + struct sg_table *sgt; + /** @prot: the mapping protection flags */ + int prot; + + /** + * @queue_id: The id of the submitqueue the operation is performed + * on, or zero for (in particular) UNMAP ops triggered outside of + * a submitqueue (ie. process cleanup) + */ + int queue_id; +}; + +/** + * struct msm_vm_unmap_op - unmap a range of pages from pgtable + */ +struct msm_vm_unmap_op { + /** @iova: start address for unmap */ + uint64_t iova; + /** @range: size of region to unmap */ + uint64_t range; + + /** @reason: The reason for the unmap */ + const char *reason; + + /** + * @queue_id: The id of the submitqueue the operation is performed + * on, or zero for (in particular) UNMAP ops triggered outside of + * a submitqueue (ie. process cleanup) + */ + int queue_id; +}; + +/** + * struct msm_vma_op - A MAP or UNMAP operation + */ +struct msm_vm_op { + /** @op: The operation type */ + enum { + MSM_VM_OP_MAP = 1, + MSM_VM_OP_UNMAP, + } op; + union { + /** @map: Parameters used if op == MSM_VMA_OP_MAP */ + struct msm_vm_map_op map; + /** @unmap: Parameters used if op == MSM_VMA_OP_UNMAP */ + struct msm_vm_unmap_op unmap; + }; + /** @node: list head in msm_vm_bind_job::vm_ops */ + struct list_head node; + + /** + * @obj: backing object for pages to be mapped/unmapped + * + * Async unmap ops, in particular, must hold a reference to the + * original GEM object backing the mapping that will be unmapped. + * But the same can be required in the map path, for example if + * there is not a corresponding unmap op, such as process exit. + * + * This ensures that the pages backing the mapping are not freed + * before the mapping is torn down. + */ + struct drm_gem_object *obj; +}; + +/** + * struct msm_vm_bind_job - Tracking for a VM_BIND ioctl + * + * A table of userspace requested VM updates (MSM_VM_BIND_OP_UNMAP/MAP/MAP_NULL) + * gets applied to the vm, generating a list of VM ops (MSM_VM_OP_MAP/UNMAP) + * which are applied to the pgtables asynchronously. For example a userspace + * requested MSM_VM_BIND_OP_MAP could end up generating both an MSM_VM_OP_UNMAP + * to unmap an existing mapping, and a MSM_VM_OP_MAP to apply the new mapping. + */ +struct msm_vm_bind_job { + /** @base: base class for drm_sched jobs */ + struct drm_sched_job base; + /** @vm: The VM being operated on */ + struct drm_gpuvm *vm; + /** @fence: The fence that is signaled when job completes */ + struct dma_fence *fence; + /** @queue: The queue that the job runs on */ + struct msm_gpu_submitqueue *queue; + /** @prealloc: Tracking for pre-allocated MMU pgtable pages */ + struct msm_mmu_prealloc prealloc; + /** @vm_ops: a list of struct msm_vm_op */ + struct list_head vm_ops; + /** @bos_pinned: are the GEM objects being bound pinned? */ + bool bos_pinned; + /** @nr_ops: the number of userspace requested ops */ + unsigned int nr_ops; + /** + * @ops: the userspace requested ops + * + * The userspace requested ops are copied/parsed and validated + * before we start applying the updates to try to do as much up- + * front error checking as possible, to avoid the VM being in an + * undefined state due to partially executed VM_BIND. + * + * This table also serves to hold a reference to the backing GEM + * objects. + */ + struct msm_vm_bind_op { + uint32_t op; + uint32_t flags; + union { + struct drm_gem_object *obj; + uint32_t handle; + }; + uint64_t obj_offset; + uint64_t iova; + uint64_t range; + } ops[]; +}; + +#define job_foreach_bo(obj, _job) \ + for (unsigned i = 0; i < (_job)->nr_ops; i++) \ + if ((obj = (_job)->ops[i].obj)) + +static inline struct msm_vm_bind_job *to_msm_vm_bind_job(struct drm_sched_job *job) +{ + return container_of(job, struct msm_vm_bind_job, base); +} static void -msm_gem_address_space_destroy(struct kref *kref) +msm_gem_vm_free(struct drm_gpuvm *gpuvm) { - struct msm_gem_address_space *aspace = container_of(kref, - struct msm_gem_address_space, kref); + struct msm_gem_vm *vm = container_of(gpuvm, struct msm_gem_vm, base); - drm_mm_takedown(&aspace->mm); - if (aspace->mmu) - aspace->mmu->funcs->destroy(aspace->mmu); - put_pid(aspace->pid); - kfree(aspace); + drm_mm_takedown(&vm->mm); + if (vm->mmu) + vm->mmu->funcs->destroy(vm->mmu); + dma_fence_put(vm->last_fence); + put_pid(vm->pid); + kfree(vm->log); + kfree(vm); } +/** + * msm_gem_vm_unusable() - Mark a VM as unusable + * @gpuvm: the VM to mark unusable + */ +void +msm_gem_vm_unusable(struct drm_gpuvm *gpuvm) +{ + struct msm_gem_vm *vm = to_msm_vm(gpuvm); + uint32_t vm_log_len = (1 << vm->log_shift); + uint32_t vm_log_mask = vm_log_len - 1; + uint32_t nr_vm_logs; + int first; + + vm->unusable = true; + + /* Bail if no log, or empty log: */ + if (!vm->log || !vm->log[0].op) + return; + + mutex_lock(&vm->mmu_lock); + + /* + * log_idx is the next entry to overwrite, meaning it is the oldest, or + * first, entry (other than the special case handled below where the + * log hasn't wrapped around yet) + */ + first = vm->log_idx; + + if (!vm->log[first].op) { + /* + * If the next log entry has not been written yet, then only + * entries 0 to idx-1 are valid (ie. we haven't wrapped around + * yet) + */ + nr_vm_logs = MAX(0, first - 1); + first = 0; + } else { + nr_vm_logs = vm_log_len; + } + + pr_err("vm-log:\n"); + for (int i = 0; i < nr_vm_logs; i++) { + int idx = (i + first) & vm_log_mask; + struct msm_gem_vm_log_entry *e = &vm->log[idx]; + pr_err(" - %s:%d: 0x%016llx-0x%016llx\n", + e->op, e->queue_id, e->iova, + e->iova + e->range); + } + + mutex_unlock(&vm->mmu_lock); +} -void msm_gem_address_space_put(struct msm_gem_address_space *aspace) +static void +vm_log(struct msm_gem_vm *vm, const char *op, uint64_t iova, uint64_t range, int queue_id) { - if (aspace) - kref_put(&aspace->kref, msm_gem_address_space_destroy); + int idx; + + if (!vm->managed) + lockdep_assert_held(&vm->mmu_lock); + + vm_dbg("%s:%p:%d: %016llx %016llx", op, vm, queue_id, iova, iova + range); + + if (!vm->log) + return; + + idx = vm->log_idx; + vm->log[idx].op = op; + vm->log[idx].iova = iova; + vm->log[idx].range = range; + vm->log[idx].queue_id = queue_id; + vm->log_idx = (vm->log_idx + 1) & ((1 << vm->log_shift) - 1); +} + +static void +vm_unmap_op(struct msm_gem_vm *vm, const struct msm_vm_unmap_op *op) +{ + const char *reason = op->reason; + + if (!reason) + reason = "unmap"; + + vm_log(vm, reason, op->iova, op->range, op->queue_id); + + vm->mmu->funcs->unmap(vm->mmu, op->iova, op->range); } -struct msm_gem_address_space * -msm_gem_address_space_get(struct msm_gem_address_space *aspace) +static int +vm_map_op(struct msm_gem_vm *vm, const struct msm_vm_map_op *op) { - if (!IS_ERR_OR_NULL(aspace)) - kref_get(&aspace->kref); + vm_log(vm, "map", op->iova, op->range, op->queue_id); - return aspace; + return vm->mmu->funcs->map(vm->mmu, op->iova, op->sgt, op->offset, + op->range, op->prot); } /* Actually unmap memory for the vma */ -void msm_gem_vma_purge(struct msm_gem_vma *vma) +void msm_gem_vma_unmap(struct drm_gpuva *vma, const char *reason) { - struct msm_gem_address_space *aspace = vma->aspace; - unsigned size = vma->node.size; + struct msm_gem_vm *vm = to_msm_vm(vma->vm); + struct msm_gem_vma *msm_vma = to_msm_vma(vma); /* Don't do anything if the memory isn't mapped */ - if (!vma->mapped) + if (!msm_vma->mapped) return; - aspace->mmu->funcs->unmap(aspace->mmu, vma->iova, size); + /* + * The mmu_lock is only needed when preallocation is used. But + * in that case we don't need to worry about recursion into + * shrinker + */ + if (!vm->managed) + mutex_lock(&vm->mmu_lock); - vma->mapped = false; + vm_unmap_op(vm, &(struct msm_vm_unmap_op){ + .iova = vma->va.addr, + .range = vma->va.range, + .reason = reason, + }); + + if (!vm->managed) + mutex_unlock(&vm->mmu_lock); + + msm_vma->mapped = false; } /* Map and pin vma: */ int -msm_gem_vma_map(struct msm_gem_vma *vma, int prot, - struct sg_table *sgt, int size) +msm_gem_vma_map(struct drm_gpuva *vma, int prot, struct sg_table *sgt) { - struct msm_gem_address_space *aspace = vma->aspace; + struct msm_gem_vm *vm = to_msm_vm(vma->vm); + struct msm_gem_vma *msm_vma = to_msm_vma(vma); int ret; - if (GEM_WARN_ON(!vma->iova)) + if (GEM_WARN_ON(!vma->va.addr)) return -EINVAL; - if (vma->mapped) + if (msm_vma->mapped) return 0; - vma->mapped = true; + msm_vma->mapped = true; - if (!aspace) - return 0; + /* + * The mmu_lock is only needed when preallocation is used. But + * in that case we don't need to worry about recursion into + * shrinker + */ + if (!vm->managed) + mutex_lock(&vm->mmu_lock); /* * NOTE: iommu/io-pgtable can allocate pages, so we cannot hold @@ -81,97 +327,1205 @@ msm_gem_vma_map(struct msm_gem_vma *vma, int prot, * Revisit this if we can come up with a scheme to pre-alloc pages * for the pgtable in map/unmap ops. */ - ret = aspace->mmu->funcs->map(aspace->mmu, vma->iova, sgt, size, prot); + ret = vm_map_op(vm, &(struct msm_vm_map_op){ + .iova = vma->va.addr, + .range = vma->va.range, + .offset = vma->gem.offset, + .sgt = sgt, + .prot = prot, + }); - if (ret) { - vma->mapped = false; - } + if (!vm->managed) + mutex_unlock(&vm->mmu_lock); + + if (ret) + msm_vma->mapped = false; return ret; } /* Close an iova. Warn if it is still in use */ -void msm_gem_vma_close(struct msm_gem_vma *vma) +void msm_gem_vma_close(struct drm_gpuva *vma) { - struct msm_gem_address_space *aspace = vma->aspace; + struct msm_gem_vm *vm = to_msm_vm(vma->vm); + struct msm_gem_vma *msm_vma = to_msm_vma(vma); - GEM_WARN_ON(vma->mapped); + GEM_WARN_ON(msm_vma->mapped); - spin_lock(&aspace->lock); - if (vma->iova) - drm_mm_remove_node(&vma->node); - spin_unlock(&aspace->lock); + drm_gpuvm_resv_assert_held(&vm->base); + + if (vma->gem.obj) + msm_gem_assert_locked(vma->gem.obj); - vma->iova = 0; + if (vma->va.addr && vm->managed) + drm_mm_remove_node(&msm_vma->node); - msm_gem_address_space_put(aspace); + drm_gpuva_remove(vma); + drm_gpuva_unlink(vma); + + kfree(vma); } -struct msm_gem_vma *msm_gem_vma_new(struct msm_gem_address_space *aspace) +/* Create a new vma and allocate an iova for it */ +struct drm_gpuva * +msm_gem_vma_new(struct drm_gpuvm *gpuvm, struct drm_gem_object *obj, + u64 offset, u64 range_start, u64 range_end) { + struct msm_gem_vm *vm = to_msm_vm(gpuvm); + struct drm_gpuvm_bo *vm_bo; struct msm_gem_vma *vma; + int ret; + + drm_gpuvm_resv_assert_held(&vm->base); vma = kzalloc(sizeof(*vma), GFP_KERNEL); if (!vma) - return NULL; + return ERR_PTR(-ENOMEM); + + if (vm->managed) { + BUG_ON(offset != 0); + BUG_ON(!obj); /* NULL mappings not valid for kernel managed VM */ + ret = drm_mm_insert_node_in_range(&vm->mm, &vma->node, + obj->size, PAGE_SIZE, 0, + range_start, range_end, 0); + + if (ret) + goto err_free_vma; + + range_start = vma->node.start; + range_end = range_start + obj->size; + } + + if (obj) + GEM_WARN_ON((range_end - range_start) > obj->size); + + drm_gpuva_init(&vma->base, range_start, range_end - range_start, obj, offset); + vma->mapped = false; + + ret = drm_gpuva_insert(&vm->base, &vma->base); + if (ret) + goto err_free_range; - vma->aspace = aspace; + if (!obj) + return &vma->base; - return vma; + vm_bo = drm_gpuvm_bo_obtain(&vm->base, obj); + if (IS_ERR(vm_bo)) { + ret = PTR_ERR(vm_bo); + goto err_va_remove; + } + + drm_gpuvm_bo_extobj_add(vm_bo); + drm_gpuva_link(&vma->base, vm_bo); + GEM_WARN_ON(drm_gpuvm_bo_put(vm_bo)); + + return &vma->base; + +err_va_remove: + drm_gpuva_remove(&vma->base); +err_free_range: + if (vm->managed) + drm_mm_remove_node(&vma->node); +err_free_vma: + kfree(vma); + return ERR_PTR(ret); } -/* Initialize a new vma and allocate an iova for it */ -int msm_gem_vma_init(struct msm_gem_vma *vma, int size, - u64 range_start, u64 range_end) +static int +msm_gem_vm_bo_validate(struct drm_gpuvm_bo *vm_bo, struct drm_exec *exec) { - struct msm_gem_address_space *aspace = vma->aspace; + struct drm_gem_object *obj = vm_bo->obj; + struct drm_gpuva *vma; int ret; - if (GEM_WARN_ON(!aspace)) - return -EINVAL; + vm_dbg("validate: %p", obj); + + msm_gem_assert_locked(obj); - if (GEM_WARN_ON(vma->iova)) - return -EBUSY; + drm_gpuvm_bo_for_each_va (vma, vm_bo) { + ret = msm_gem_pin_vma_locked(obj, vma); + if (ret) + return ret; + } - spin_lock(&aspace->lock); - ret = drm_mm_insert_node_in_range(&aspace->mm, &vma->node, - size, PAGE_SIZE, 0, - range_start, range_end, 0); - spin_unlock(&aspace->lock); + return 0; +} - if (ret) - return ret; +struct op_arg { + unsigned flags; + struct msm_vm_bind_job *job; +}; - vma->iova = vma->node.start; - vma->mapped = false; +static void +vm_op_enqueue(struct op_arg *arg, struct msm_vm_op _op) +{ + struct msm_vm_op *op = kmalloc(sizeof(*op), GFP_KERNEL); + *op = _op; + list_add_tail(&op->node, &arg->job->vm_ops); + + if (op->obj) + drm_gem_object_get(op->obj); +} + +static struct drm_gpuva * +vma_from_op(struct op_arg *arg, struct drm_gpuva_op_map *op) +{ + return msm_gem_vma_new(arg->job->vm, op->gem.obj, op->gem.offset, + op->va.addr, op->va.addr + op->va.range); +} + +static int +msm_gem_vm_sm_step_map(struct drm_gpuva_op *op, void *arg) +{ + struct msm_vm_bind_job *job = ((struct op_arg *)arg)->job; + struct drm_gem_object *obj = op->map.gem.obj; + struct drm_gpuva *vma; + struct sg_table *sgt; + unsigned prot; + + vma = vma_from_op(arg, &op->map); + if (WARN_ON(IS_ERR(vma))) + return PTR_ERR(vma); + + vm_dbg("%p:%p:%p: %016llx %016llx", vma->vm, vma, vma->gem.obj, + vma->va.addr, vma->va.range); + + vma->flags = ((struct op_arg *)arg)->flags; + + if (obj) { + sgt = to_msm_bo(obj)->sgt; + prot = msm_gem_prot(obj); + } else { + sgt = NULL; + prot = IOMMU_READ | IOMMU_WRITE; + } - kref_get(&aspace->kref); + vm_op_enqueue(arg, (struct msm_vm_op){ + .op = MSM_VM_OP_MAP, + .map = { + .sgt = sgt, + .iova = vma->va.addr, + .range = vma->va.range, + .offset = vma->gem.offset, + .prot = prot, + .queue_id = job->queue->id, + }, + .obj = vma->gem.obj, + }); + + to_msm_vma(vma)->mapped = true; + + return 0; +} + +static int +msm_gem_vm_sm_step_remap(struct drm_gpuva_op *op, void *arg) +{ + struct msm_vm_bind_job *job = ((struct op_arg *)arg)->job; + struct drm_gpuvm *vm = job->vm; + struct drm_gpuva *orig_vma = op->remap.unmap->va; + struct drm_gpuva *prev_vma = NULL, *next_vma = NULL; + struct drm_gpuvm_bo *vm_bo = orig_vma->vm_bo; + bool mapped = to_msm_vma(orig_vma)->mapped; + unsigned flags; + + vm_dbg("orig_vma: %p:%p:%p: %016llx %016llx", vm, orig_vma, + orig_vma->gem.obj, orig_vma->va.addr, orig_vma->va.range); + + if (mapped) { + uint64_t unmap_start, unmap_range; + + drm_gpuva_op_remap_to_unmap_range(&op->remap, &unmap_start, &unmap_range); + + vm_op_enqueue(arg, (struct msm_vm_op){ + .op = MSM_VM_OP_UNMAP, + .unmap = { + .iova = unmap_start, + .range = unmap_range, + .queue_id = job->queue->id, + }, + .obj = orig_vma->gem.obj, + }); + + /* + * Part of this GEM obj is still mapped, but we're going to kill the + * existing VMA and replace it with one or two new ones (ie. two if + * the unmapped range is in the middle of the existing (unmap) VMA). + * So just set the state to unmapped: + */ + to_msm_vma(orig_vma)->mapped = false; + } + + /* + * Hold a ref to the vm_bo between the msm_gem_vma_close() and the + * creation of the new prev/next vma's, in case the vm_bo is tracked + * in the VM's evict list: + */ + if (vm_bo) + drm_gpuvm_bo_get(vm_bo); + + /* + * The prev_vma and/or next_vma are replacing the unmapped vma, and + * therefore should preserve it's flags: + */ + flags = orig_vma->flags; + + msm_gem_vma_close(orig_vma); + + if (op->remap.prev) { + prev_vma = vma_from_op(arg, op->remap.prev); + if (WARN_ON(IS_ERR(prev_vma))) + return PTR_ERR(prev_vma); + + vm_dbg("prev_vma: %p:%p: %016llx %016llx", vm, prev_vma, prev_vma->va.addr, prev_vma->va.range); + to_msm_vma(prev_vma)->mapped = mapped; + prev_vma->flags = flags; + } + + if (op->remap.next) { + next_vma = vma_from_op(arg, op->remap.next); + if (WARN_ON(IS_ERR(next_vma))) + return PTR_ERR(next_vma); + + vm_dbg("next_vma: %p:%p: %016llx %016llx", vm, next_vma, next_vma->va.addr, next_vma->va.range); + to_msm_vma(next_vma)->mapped = mapped; + next_vma->flags = flags; + } + + if (!mapped) + drm_gpuvm_bo_evict(vm_bo, true); + + /* Drop the previous ref: */ + drm_gpuvm_bo_put(vm_bo); return 0; } -struct msm_gem_address_space * -msm_gem_address_space_create(struct msm_mmu *mmu, const char *name, - u64 va_start, u64 size) +static int +msm_gem_vm_sm_step_unmap(struct drm_gpuva_op *op, void *arg) +{ + struct msm_vm_bind_job *job = ((struct op_arg *)arg)->job; + struct drm_gpuva *vma = op->unmap.va; + struct msm_gem_vma *msm_vma = to_msm_vma(vma); + + vm_dbg("%p:%p:%p: %016llx %016llx", vma->vm, vma, vma->gem.obj, + vma->va.addr, vma->va.range); + + if (!msm_vma->mapped) + goto out_close; + + vm_op_enqueue(arg, (struct msm_vm_op){ + .op = MSM_VM_OP_UNMAP, + .unmap = { + .iova = vma->va.addr, + .range = vma->va.range, + .queue_id = job->queue->id, + }, + .obj = vma->gem.obj, + }); + + msm_vma->mapped = false; + +out_close: + msm_gem_vma_close(vma); + + return 0; +} + +static const struct drm_gpuvm_ops msm_gpuvm_ops = { + .vm_free = msm_gem_vm_free, + .vm_bo_validate = msm_gem_vm_bo_validate, + .sm_step_map = msm_gem_vm_sm_step_map, + .sm_step_remap = msm_gem_vm_sm_step_remap, + .sm_step_unmap = msm_gem_vm_sm_step_unmap, +}; + +static struct dma_fence * +msm_vma_job_run(struct drm_sched_job *_job) +{ + struct msm_vm_bind_job *job = to_msm_vm_bind_job(_job); + struct msm_gem_vm *vm = to_msm_vm(job->vm); + struct drm_gem_object *obj; + int ret = vm->unusable ? -EINVAL : 0; + + vm_dbg(""); + + mutex_lock(&vm->mmu_lock); + vm->mmu->prealloc = &job->prealloc; + + while (!list_empty(&job->vm_ops)) { + struct msm_vm_op *op = + list_first_entry(&job->vm_ops, struct msm_vm_op, node); + + switch (op->op) { + case MSM_VM_OP_MAP: + /* + * On error, stop trying to map new things.. but we + * still want to process the unmaps (or in particular, + * the drm_gem_object_put()s) + */ + if (!ret) + ret = vm_map_op(vm, &op->map); + break; + case MSM_VM_OP_UNMAP: + vm_unmap_op(vm, &op->unmap); + break; + } + drm_gem_object_put(op->obj); + list_del(&op->node); + kfree(op); + } + + vm->mmu->prealloc = NULL; + mutex_unlock(&vm->mmu_lock); + + /* + * We failed to perform at least _some_ of the pgtable updates, so + * now the VM is in an undefined state. Game over! + */ + if (ret) + msm_gem_vm_unusable(job->vm); + + job_foreach_bo (obj, job) { + msm_gem_lock(obj); + msm_gem_unpin_locked(obj); + msm_gem_unlock(obj); + } + + /* VM_BIND ops are synchronous, so no fence to wait on: */ + return NULL; +} + +static void +msm_vma_job_free(struct drm_sched_job *_job) { - struct msm_gem_address_space *aspace; + struct msm_vm_bind_job *job = to_msm_vm_bind_job(_job); + struct msm_gem_vm *vm = to_msm_vm(job->vm); + struct drm_gem_object *obj; + + vm->mmu->funcs->prealloc_cleanup(vm->mmu, &job->prealloc); + + atomic_sub(job->prealloc.count, &vm->prealloc_throttle.in_flight); + + drm_sched_job_cleanup(_job); + + job_foreach_bo (obj, job) + drm_gem_object_put(obj); + + msm_submitqueue_put(job->queue); + dma_fence_put(job->fence); + + /* In error paths, we could have unexecuted ops: */ + while (!list_empty(&job->vm_ops)) { + struct msm_vm_op *op = + list_first_entry(&job->vm_ops, struct msm_vm_op, node); + list_del(&op->node); + kfree(op); + } + + wake_up(&vm->prealloc_throttle.wait); + + kfree(job); +} + +static const struct drm_sched_backend_ops msm_vm_bind_ops = { + .run_job = msm_vma_job_run, + .free_job = msm_vma_job_free +}; + +/** + * msm_gem_vm_create() - Create and initialize a &msm_gem_vm + * @drm: the drm device + * @mmu: the backing MMU objects handling mapping/unmapping + * @name: the name of the VM + * @va_start: the start offset of the VA space + * @va_size: the size of the VA space + * @managed: is it a kernel managed VM? + * + * In a kernel managed VM, the kernel handles address allocation, and only + * synchronous operations are supported. In a user managed VM, userspace + * handles virtual address allocation, and both async and sync operations + * are supported. + */ +struct drm_gpuvm * +msm_gem_vm_create(struct drm_device *drm, struct msm_mmu *mmu, const char *name, + u64 va_start, u64 va_size, bool managed) +{ + /* + * We mostly want to use DRM_GPUVM_RESV_PROTECTED, except that + * makes drm_gpuvm_bo_evict() a no-op for extobjs (ie. we loose + * tracking that an extobj is evicted) :facepalm: + */ + enum drm_gpuvm_flags flags = 0; + struct msm_gem_vm *vm; + struct drm_gem_object *dummy_gem; + int ret = 0; if (IS_ERR(mmu)) return ERR_CAST(mmu); - aspace = kzalloc(sizeof(*aspace), GFP_KERNEL); - if (!aspace) + vm = kzalloc(sizeof(*vm), GFP_KERNEL); + if (!vm) return ERR_PTR(-ENOMEM); - spin_lock_init(&aspace->lock); - aspace->name = name; - aspace->mmu = mmu; - aspace->va_start = va_start; - aspace->va_size = size; + dummy_gem = drm_gpuvm_resv_object_alloc(drm); + if (!dummy_gem) { + ret = -ENOMEM; + goto err_free_vm; + } + + if (!managed) { + struct drm_sched_init_args args = { + .ops = &msm_vm_bind_ops, + .num_rqs = 1, + .credit_limit = 1, + .timeout = MAX_SCHEDULE_TIMEOUT, + .name = "msm-vm-bind", + .dev = drm->dev, + }; + + ret = drm_sched_init(&vm->sched, &args); + if (ret) + goto err_free_dummy; + + init_waitqueue_head(&vm->prealloc_throttle.wait); + } + + drm_gpuvm_init(&vm->base, name, flags, drm, dummy_gem, + va_start, va_size, 0, 0, &msm_gpuvm_ops); + drm_gem_object_put(dummy_gem); + + vm->mmu = mmu; + mutex_init(&vm->mmu_lock); + vm->managed = managed; + + drm_mm_init(&vm->mm, va_start, va_size); + + /* + * We don't really need vm log for kernel managed VMs, as the kernel + * is responsible for ensuring that GEM objs are mapped if they are + * used by a submit. Furthermore we piggyback on mmu_lock to serialize + * access to the log. + * + * Limit the max log_shift to 8 to prevent userspace from asking us + * for an unreasonable log size. + */ + if (!managed) + vm->log_shift = MIN(vm_log_shift, 8); + + if (vm->log_shift) { + vm->log = kmalloc_array(1 << vm->log_shift, sizeof(vm->log[0]), + GFP_KERNEL | __GFP_ZERO); + } + + return &vm->base; + +err_free_dummy: + drm_gem_object_put(dummy_gem); + +err_free_vm: + kfree(vm); + return ERR_PTR(ret); +} + +/** + * msm_gem_vm_close() - Close a VM + * @gpuvm: The VM to close + * + * Called when the drm device file is closed, to tear down VM related resources + * (which will drop refcounts to GEM objects that were still mapped into the + * VM at the time). + */ +void +msm_gem_vm_close(struct drm_gpuvm *gpuvm) +{ + struct msm_gem_vm *vm = to_msm_vm(gpuvm); + struct drm_gpuva *vma, *tmp; + struct drm_exec exec; + + /* + * For kernel managed VMs, the VMAs are torn down when the handle is + * closed, so nothing more to do. + */ + if (vm->managed) + return; - drm_mm_init(&aspace->mm, va_start, size); + if (vm->last_fence) + dma_fence_wait(vm->last_fence, false); + + /* Kill the scheduler now, so we aren't racing with it for cleanup: */ + drm_sched_stop(&vm->sched, NULL); + drm_sched_fini(&vm->sched); + + /* Tear down any remaining mappings: */ + drm_exec_init(&exec, 0, 2); + drm_exec_until_all_locked (&exec) { + drm_exec_lock_obj(&exec, drm_gpuvm_resv_obj(gpuvm)); + drm_exec_retry_on_contention(&exec); + + drm_gpuvm_for_each_va_safe (vma, tmp, gpuvm) { + struct drm_gem_object *obj = vma->gem.obj; + + /* + * MSM_BO_NO_SHARE objects share the same resv as the + * VM, in which case the obj is already locked: + */ + if (obj && (obj->resv == drm_gpuvm_resv(gpuvm))) + obj = NULL; + + if (obj) { + drm_exec_lock_obj(&exec, obj); + drm_exec_retry_on_contention(&exec); + } + + msm_gem_vma_unmap(vma, "close"); + msm_gem_vma_close(vma); + + if (obj) { + drm_exec_unlock_obj(&exec, obj); + } + } + } + drm_exec_fini(&exec); +} + + +static struct msm_vm_bind_job * +vm_bind_job_create(struct drm_device *dev, struct drm_file *file, + struct msm_gpu_submitqueue *queue, uint32_t nr_ops) +{ + struct msm_vm_bind_job *job; + uint64_t sz; + int ret; - kref_init(&aspace->kref); + sz = struct_size(job, ops, nr_ops); - return aspace; + if (sz > SIZE_MAX) + return ERR_PTR(-ENOMEM); + + job = kzalloc(sz, GFP_KERNEL | __GFP_NOWARN); + if (!job) + return ERR_PTR(-ENOMEM); + + ret = drm_sched_job_init(&job->base, queue->entity, 1, queue, + file->client_id); + if (ret) { + kfree(job); + return ERR_PTR(ret); + } + + job->vm = msm_context_vm(dev, queue->ctx); + job->queue = queue; + INIT_LIST_HEAD(&job->vm_ops); + + return job; +} + +static bool invalid_alignment(uint64_t addr) +{ + /* + * Technically this is about GPU alignment, not CPU alignment. But + * I've not seen any qcom SoC where the SMMU does not support the + * CPU's smallest page size. + */ + return !PAGE_ALIGNED(addr); +} + +static int +lookup_op(struct msm_vm_bind_job *job, const struct drm_msm_vm_bind_op *op) +{ + struct drm_device *dev = job->vm->drm; + int i = job->nr_ops++; + int ret = 0; + + job->ops[i].op = op->op; + job->ops[i].handle = op->handle; + job->ops[i].obj_offset = op->obj_offset; + job->ops[i].iova = op->iova; + job->ops[i].range = op->range; + job->ops[i].flags = op->flags; + + if (op->flags & ~MSM_VM_BIND_OP_FLAGS) + ret = UERR(EINVAL, dev, "invalid flags: %x\n", op->flags); + + if (invalid_alignment(op->iova)) + ret = UERR(EINVAL, dev, "invalid address: %016llx\n", op->iova); + + if (invalid_alignment(op->obj_offset)) + ret = UERR(EINVAL, dev, "invalid bo_offset: %016llx\n", op->obj_offset); + + if (invalid_alignment(op->range)) + ret = UERR(EINVAL, dev, "invalid range: %016llx\n", op->range); + + if (!drm_gpuvm_range_valid(job->vm, op->iova, op->range)) + ret = UERR(EINVAL, dev, "invalid range: %016llx, %016llx\n", op->iova, op->range); + + /* + * MAP must specify a valid handle. But the handle MBZ for + * UNMAP or MAP_NULL. + */ + if (op->op == MSM_VM_BIND_OP_MAP) { + if (!op->handle) + ret = UERR(EINVAL, dev, "invalid handle\n"); + } else if (op->handle) { + ret = UERR(EINVAL, dev, "handle must be zero\n"); + } + + switch (op->op) { + case MSM_VM_BIND_OP_MAP: + case MSM_VM_BIND_OP_MAP_NULL: + case MSM_VM_BIND_OP_UNMAP: + break; + default: + ret = UERR(EINVAL, dev, "invalid op: %u\n", op->op); + break; + } + + return ret; +} + +/* + * ioctl parsing, parameter validation, and GEM handle lookup + */ +static int +vm_bind_job_lookup_ops(struct msm_vm_bind_job *job, struct drm_msm_vm_bind *args, + struct drm_file *file, int *nr_bos) +{ + struct drm_device *dev = job->vm->drm; + int ret = 0; + int cnt = 0; + + if (args->nr_ops == 1) { + /* Single op case, the op is inlined: */ + ret = lookup_op(job, &args->op); + } else { + for (unsigned i = 0; i < args->nr_ops; i++) { + struct drm_msm_vm_bind_op op; + void __user *userptr = + u64_to_user_ptr(args->ops + (i * sizeof(op))); + + /* make sure we don't have garbage flags, in case we hit + * error path before flags is initialized: + */ + job->ops[i].flags = 0; + + if (copy_from_user(&op, userptr, sizeof(op))) { + ret = -EFAULT; + break; + } + + ret = lookup_op(job, &op); + if (ret) + break; + } + } + + if (ret) { + job->nr_ops = 0; + goto out; + } + + spin_lock(&file->table_lock); + + for (unsigned i = 0; i < args->nr_ops; i++) { + struct drm_gem_object *obj; + + if (!job->ops[i].handle) { + job->ops[i].obj = NULL; + continue; + } + + /* + * normally use drm_gem_object_lookup(), but for bulk lookup + * all under single table_lock just hit object_idr directly: + */ + obj = idr_find(&file->object_idr, job->ops[i].handle); + if (!obj) { + ret = UERR(EINVAL, dev, "invalid handle %u at index %u\n", job->ops[i].handle, i); + goto out_unlock; + } + + drm_gem_object_get(obj); + + job->ops[i].obj = obj; + cnt++; + } + + *nr_bos = cnt; + +out_unlock: + spin_unlock(&file->table_lock); + +out: + return ret; +} + +static void +prealloc_count(struct msm_vm_bind_job *job, + struct msm_vm_bind_op *first, + struct msm_vm_bind_op *last) +{ + struct msm_mmu *mmu = to_msm_vm(job->vm)->mmu; + + if (!first) + return; + + uint64_t start_iova = first->iova; + uint64_t end_iova = last->iova + last->range; + + mmu->funcs->prealloc_count(mmu, &job->prealloc, start_iova, end_iova - start_iova); +} + +static bool +ops_are_same_pte(struct msm_vm_bind_op *first, struct msm_vm_bind_op *next) +{ + /* + * Last level pte covers 2MB.. so we should merge two ops, from + * the PoV of figuring out how much pgtable pages to pre-allocate + * if they land in the same 2MB range: + */ + uint64_t pte_mask = ~(SZ_2M - 1); + return ((first->iova + first->range) & pte_mask) == (next->iova & pte_mask); +} + +/* + * Determine the amount of memory to prealloc for pgtables. For sparse images, + * in particular, userspace plays some tricks with the order of page mappings + * to get the desired swizzle pattern, resulting in a large # of tiny MAP ops. + * So detect when multiple MAP operations are physically contiguous, and count + * them as a single mapping. Otherwise the prealloc_count() will not realize + * they can share pagetable pages and vastly overcount. + */ +static int +vm_bind_prealloc_count(struct msm_vm_bind_job *job) +{ + struct msm_vm_bind_op *first = NULL, *last = NULL; + struct msm_gem_vm *vm = to_msm_vm(job->vm); + int ret; + + for (int i = 0; i < job->nr_ops; i++) { + struct msm_vm_bind_op *op = &job->ops[i]; + + /* We only care about MAP/MAP_NULL: */ + if (op->op == MSM_VM_BIND_OP_UNMAP) + continue; + + /* + * If op is contiguous with last in the current range, then + * it becomes the new last in the range and we continue + * looping: + */ + if (last && ops_are_same_pte(last, op)) { + last = op; + continue; + } + + /* + * If op is not contiguous with the current range, flush + * the current range and start anew: + */ + prealloc_count(job, first, last); + first = last = op; + } + + /* Flush the remaining range: */ + prealloc_count(job, first, last); + + /* + * Now that we know the needed amount to pre-alloc, throttle on pending + * VM_BIND jobs if we already have too much pre-alloc memory in flight + */ + ret = wait_event_interruptible( + vm->prealloc_throttle.wait, + atomic_read(&vm->prealloc_throttle.in_flight) <= 1024); + if (ret) + return ret; + + atomic_add(job->prealloc.count, &vm->prealloc_throttle.in_flight); + + return 0; +} + +/* + * Lock VM and GEM objects + */ +static int +vm_bind_job_lock_objects(struct msm_vm_bind_job *job, struct drm_exec *exec) +{ + int ret; + + /* Lock VM and objects: */ + drm_exec_until_all_locked (exec) { + ret = drm_exec_lock_obj(exec, drm_gpuvm_resv_obj(job->vm)); + drm_exec_retry_on_contention(exec); + if (ret) + return ret; + + for (unsigned i = 0; i < job->nr_ops; i++) { + const struct msm_vm_bind_op *op = &job->ops[i]; + + switch (op->op) { + case MSM_VM_BIND_OP_UNMAP: + ret = drm_gpuvm_sm_unmap_exec_lock(job->vm, exec, + op->iova, + op->obj_offset); + break; + case MSM_VM_BIND_OP_MAP: + case MSM_VM_BIND_OP_MAP_NULL: + ret = drm_gpuvm_sm_map_exec_lock(job->vm, exec, 1, + op->iova, op->range, + op->obj, op->obj_offset); + break; + default: + /* + * lookup_op() should have already thrown an error for + * invalid ops + */ + WARN_ON("unreachable"); + } + + drm_exec_retry_on_contention(exec); + if (ret) + return ret; + } + } + + return 0; +} + +/* + * Pin GEM objects, ensuring that we have backing pages. Pinning will move + * the object to the pinned LRU so that the shrinker knows to first consider + * other objects for evicting. + */ +static int +vm_bind_job_pin_objects(struct msm_vm_bind_job *job) +{ + struct drm_gem_object *obj; + + /* + * First loop, before holding the LRU lock, avoids holding the + * LRU lock while calling msm_gem_pin_vma_locked (which could + * trigger get_pages()) + */ + job_foreach_bo (obj, job) { + struct page **pages; + + pages = msm_gem_get_pages_locked(obj, MSM_MADV_WILLNEED); + if (IS_ERR(pages)) + return PTR_ERR(pages); + } + + struct msm_drm_private *priv = job->vm->drm->dev_private; + + /* + * A second loop while holding the LRU lock (a) avoids acquiring/dropping + * the LRU lock for each individual bo, while (b) avoiding holding the + * LRU lock while calling msm_gem_pin_vma_locked() (which could trigger + * get_pages() which could trigger reclaim.. and if we held the LRU lock + * could trigger deadlock with the shrinker). + */ + mutex_lock(&priv->lru.lock); + job_foreach_bo (obj, job) + msm_gem_pin_obj_locked(obj); + mutex_unlock(&priv->lru.lock); + + job->bos_pinned = true; + + return 0; +} + +/* + * Unpin GEM objects. Normally this is done after the bind job is run. + */ +static void +vm_bind_job_unpin_objects(struct msm_vm_bind_job *job) +{ + struct drm_gem_object *obj; + + if (!job->bos_pinned) + return; + + job_foreach_bo (obj, job) + msm_gem_unpin_locked(obj); + + job->bos_pinned = false; +} + +/* + * Pre-allocate pgtable memory, and translate the VM bind requests into a + * sequence of pgtable updates to be applied asynchronously. + */ +static int +vm_bind_job_prepare(struct msm_vm_bind_job *job) +{ + struct msm_gem_vm *vm = to_msm_vm(job->vm); + struct msm_mmu *mmu = vm->mmu; + int ret; + + ret = mmu->funcs->prealloc_allocate(mmu, &job->prealloc); + if (ret) + return ret; + + for (unsigned i = 0; i < job->nr_ops; i++) { + const struct msm_vm_bind_op *op = &job->ops[i]; + struct op_arg arg = { + .job = job, + }; + + switch (op->op) { + case MSM_VM_BIND_OP_UNMAP: + ret = drm_gpuvm_sm_unmap(job->vm, &arg, op->iova, + op->range); + break; + case MSM_VM_BIND_OP_MAP: + if (op->flags & MSM_VM_BIND_OP_DUMP) + arg.flags |= MSM_VMA_DUMP; + fallthrough; + case MSM_VM_BIND_OP_MAP_NULL: + ret = drm_gpuvm_sm_map(job->vm, &arg, op->iova, + op->range, op->obj, op->obj_offset); + break; + default: + /* + * lookup_op() should have already thrown an error for + * invalid ops + */ + BUG_ON("unreachable"); + } + + if (ret) { + /* + * If we've already started modifying the vm, we can't + * adequetly describe to userspace the intermediate + * state the vm is in. So throw up our hands! + */ + if (i > 0) + msm_gem_vm_unusable(job->vm); + return ret; + } + } + + return 0; +} + +/* + * Attach fences to the GEM objects being bound. This will signify to + * the shrinker that they are busy even after dropping the locks (ie. + * drm_exec_fini()) + */ +static void +vm_bind_job_attach_fences(struct msm_vm_bind_job *job) +{ + for (unsigned i = 0; i < job->nr_ops; i++) { + struct drm_gem_object *obj = job->ops[i].obj; + + if (!obj) + continue; + + dma_resv_add_fence(obj->resv, job->fence, + DMA_RESV_USAGE_KERNEL); + } +} + +int +msm_ioctl_vm_bind(struct drm_device *dev, void *data, struct drm_file *file) +{ + struct msm_drm_private *priv = dev->dev_private; + struct drm_msm_vm_bind *args = data; + struct msm_context *ctx = file->driver_priv; + struct msm_vm_bind_job *job = NULL; + struct msm_gpu *gpu = priv->gpu; + struct msm_gpu_submitqueue *queue; + struct msm_syncobj_post_dep *post_deps = NULL; + struct drm_syncobj **syncobjs_to_reset = NULL; + struct sync_file *sync_file = NULL; + struct dma_fence *fence; + int out_fence_fd = -1; + int ret, nr_bos = 0; + unsigned i; + + if (!gpu) + return -ENXIO; + + /* + * Maybe we could allow just UNMAP ops? OTOH userspace should just + * immediately close the device file and all will be torn down. + */ + if (to_msm_vm(ctx->vm)->unusable) + return UERR(EPIPE, dev, "context is unusable"); + + /* + * Technically, you cannot create a VM_BIND submitqueue in the first + * place, if you haven't opted in to VM_BIND context. But it is + * cleaner / less confusing, to check this case directly. + */ + if (!msm_context_is_vmbind(ctx)) + return UERR(EINVAL, dev, "context does not support vmbind"); + + if (args->flags & ~MSM_VM_BIND_FLAGS) + return UERR(EINVAL, dev, "invalid flags"); + + queue = msm_submitqueue_get(ctx, args->queue_id); + if (!queue) + return -ENOENT; + + if (!(queue->flags & MSM_SUBMITQUEUE_VM_BIND)) { + ret = UERR(EINVAL, dev, "Invalid queue type"); + goto out_post_unlock; + } + + if (args->flags & MSM_VM_BIND_FENCE_FD_OUT) { + out_fence_fd = get_unused_fd_flags(O_CLOEXEC); + if (out_fence_fd < 0) { + ret = out_fence_fd; + goto out_post_unlock; + } + } + + job = vm_bind_job_create(dev, file, queue, args->nr_ops); + if (IS_ERR(job)) { + ret = PTR_ERR(job); + goto out_post_unlock; + } + + ret = mutex_lock_interruptible(&queue->lock); + if (ret) + goto out_post_unlock; + + if (args->flags & MSM_VM_BIND_FENCE_FD_IN) { + struct dma_fence *in_fence; + + in_fence = sync_file_get_fence(args->fence_fd); + + if (!in_fence) { + ret = UERR(EINVAL, dev, "invalid in-fence"); + goto out_unlock; + } + + ret = drm_sched_job_add_dependency(&job->base, in_fence); + if (ret) + goto out_unlock; + } + + if (args->in_syncobjs > 0) { + syncobjs_to_reset = msm_syncobj_parse_deps(dev, &job->base, + file, args->in_syncobjs, + args->nr_in_syncobjs, + args->syncobj_stride); + if (IS_ERR(syncobjs_to_reset)) { + ret = PTR_ERR(syncobjs_to_reset); + goto out_unlock; + } + } + + if (args->out_syncobjs > 0) { + post_deps = msm_syncobj_parse_post_deps(dev, file, + args->out_syncobjs, + args->nr_out_syncobjs, + args->syncobj_stride); + if (IS_ERR(post_deps)) { + ret = PTR_ERR(post_deps); + goto out_unlock; + } + } + + ret = vm_bind_job_lookup_ops(job, args, file, &nr_bos); + if (ret) + goto out_unlock; + + ret = vm_bind_prealloc_count(job); + if (ret) + goto out_unlock; + + struct drm_exec exec; + unsigned flags = DRM_EXEC_IGNORE_DUPLICATES | DRM_EXEC_INTERRUPTIBLE_WAIT; + drm_exec_init(&exec, flags, nr_bos + 1); + + ret = vm_bind_job_lock_objects(job, &exec); + if (ret) + goto out; + + ret = vm_bind_job_pin_objects(job); + if (ret) + goto out; + + ret = vm_bind_job_prepare(job); + if (ret) + goto out; + + drm_sched_job_arm(&job->base); + + job->fence = dma_fence_get(&job->base.s_fence->finished); + + if (args->flags & MSM_VM_BIND_FENCE_FD_OUT) { + sync_file = sync_file_create(job->fence); + if (!sync_file) { + ret = -ENOMEM; + } else { + fd_install(out_fence_fd, sync_file->file); + args->fence_fd = out_fence_fd; + } + } + + if (ret) + goto out; + + vm_bind_job_attach_fences(job); + + /* + * The job can be free'd (and fence unref'd) at any point after + * drm_sched_entity_push_job(), so we need to hold our own ref + */ + fence = dma_fence_get(job->fence); + + drm_sched_entity_push_job(&job->base); + + msm_syncobj_reset(syncobjs_to_reset, args->nr_in_syncobjs); + msm_syncobj_process_post_deps(post_deps, args->nr_out_syncobjs, fence); + + dma_fence_put(fence); + +out: + if (ret) + vm_bind_job_unpin_objects(job); + + drm_exec_fini(&exec); +out_unlock: + mutex_unlock(&queue->lock); +out_post_unlock: + if (ret && (out_fence_fd >= 0)) { + put_unused_fd(out_fence_fd); + if (sync_file) + fput(sync_file->file); + } + + if (!IS_ERR_OR_NULL(job)) { + if (ret) + msm_vma_job_free(&job->base); + } else { + /* + * If the submit hasn't yet taken ownership of the queue + * then we need to drop the reference ourself: + */ + msm_submitqueue_put(queue); + } + + if (!IS_ERR_OR_NULL(post_deps)) { + for (i = 0; i < args->nr_out_syncobjs; ++i) { + kfree(post_deps[i].chain); + drm_syncobj_put(post_deps[i].syncobj); + } + kfree(post_deps); + } + + if (!IS_ERR_OR_NULL(syncobjs_to_reset)) { + for (i = 0; i < args->nr_in_syncobjs; ++i) { + if (syncobjs_to_reset[i]) + drm_syncobj_put(syncobjs_to_reset[i]); + } + kfree(syncobjs_to_reset); + } + + return ret; } diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 197871fdf508..c317b25a8162 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -148,7 +148,7 @@ int msm_gpu_pm_suspend(struct msm_gpu *gpu) return 0; } -void msm_gpu_show_fdinfo(struct msm_gpu *gpu, struct msm_file_private *ctx, +void msm_gpu_show_fdinfo(struct msm_gpu *gpu, struct msm_context *ctx, struct drm_printer *p) { drm_printf(p, "drm-engine-gpu:\t%llu ns\n", ctx->elapsed_ns); @@ -219,13 +219,14 @@ static void msm_gpu_devcoredump_free(void *data) } static void msm_gpu_crashstate_get_bo(struct msm_gpu_state *state, - struct drm_gem_object *obj, u64 iova, bool full) + struct drm_gem_object *obj, u64 iova, + bool full, size_t offset, size_t size) { struct msm_gpu_state_bo *state_bo = &state->bos[state->nr_bos]; struct msm_gem_object *msm_obj = to_msm_bo(obj); /* Don't record write only objects */ - state_bo->size = obj->size; + state_bo->size = size; state_bo->flags = msm_obj->flags; state_bo->iova = iova; @@ -236,28 +237,129 @@ static void msm_gpu_crashstate_get_bo(struct msm_gpu_state *state, if (full) { void *ptr; - state_bo->data = kvmalloc(obj->size, GFP_KERNEL); + state_bo->data = kvmalloc(size, GFP_KERNEL); if (!state_bo->data) goto out; - msm_gem_lock(obj); ptr = msm_gem_get_vaddr_active(obj); - msm_gem_unlock(obj); if (IS_ERR(ptr)) { kvfree(state_bo->data); state_bo->data = NULL; goto out; } - memcpy(state_bo->data, ptr, obj->size); - msm_gem_put_vaddr(obj); + memcpy(state_bo->data, ptr + offset, size); + msm_gem_put_vaddr_locked(obj); } out: state->nr_bos++; } +static void crashstate_get_bos(struct msm_gpu_state *state, struct msm_gem_submit *submit) +{ + extern bool rd_full; + + if (msm_context_is_vmbind(submit->queue->ctx)) { + struct drm_exec exec; + struct drm_gpuva *vma; + unsigned cnt = 0; + + drm_exec_init(&exec, DRM_EXEC_IGNORE_DUPLICATES, 0); + drm_exec_until_all_locked(&exec) { + cnt = 0; + + drm_exec_lock_obj(&exec, drm_gpuvm_resv_obj(submit->vm)); + drm_exec_retry_on_contention(&exec); + + drm_gpuvm_for_each_va (vma, submit->vm) { + if (!vma->gem.obj) + continue; + + cnt++; + drm_exec_lock_obj(&exec, vma->gem.obj); + drm_exec_retry_on_contention(&exec); + } + + } + + drm_gpuvm_for_each_va (vma, submit->vm) + cnt++; + + state->bos = kcalloc(cnt, sizeof(struct msm_gpu_state_bo), GFP_KERNEL); + + drm_gpuvm_for_each_va (vma, submit->vm) { + bool dump = rd_full || (vma->flags & MSM_VMA_DUMP); + + /* Skip MAP_NULL/PRR VMAs: */ + if (!vma->gem.obj) + continue; + + msm_gpu_crashstate_get_bo(state, vma->gem.obj, vma->va.addr, + dump, vma->gem.offset, vma->va.range); + } + + drm_exec_fini(&exec); + } else { + state->bos = kcalloc(submit->nr_bos, + sizeof(struct msm_gpu_state_bo), GFP_KERNEL); + + for (int i = 0; state->bos && i < submit->nr_bos; i++) { + struct drm_gem_object *obj = submit->bos[i].obj;; + bool dump = rd_full || (submit->bos[i].flags & MSM_SUBMIT_BO_DUMP); + + msm_gem_lock(obj); + msm_gpu_crashstate_get_bo(state, obj, submit->bos[i].iova, + dump, 0, obj->size); + msm_gem_unlock(obj); + } + } +} + +static void crashstate_get_vm_logs(struct msm_gpu_state *state, struct msm_gem_vm *vm) +{ + uint32_t vm_log_len = (1 << vm->log_shift); + uint32_t vm_log_mask = vm_log_len - 1; + int first; + + /* Bail if no log, or empty log: */ + if (!vm->log || !vm->log[0].op) + return; + + mutex_lock(&vm->mmu_lock); + + /* + * log_idx is the next entry to overwrite, meaning it is the oldest, or + * first, entry (other than the special case handled below where the + * log hasn't wrapped around yet) + */ + first = vm->log_idx; + + if (!vm->log[first].op) { + /* + * If the next log entry has not been written yet, then only + * entries 0 to idx-1 are valid (ie. we haven't wrapped around + * yet) + */ + state->nr_vm_logs = MAX(0, first - 1); + first = 0; + } else { + state->nr_vm_logs = vm_log_len; + } + + state->vm_logs = kmalloc_array( + state->nr_vm_logs, sizeof(vm->log[0]), GFP_KERNEL); + for (int i = 0; i < state->nr_vm_logs; i++) { + int idx = (i + first) & vm_log_mask; + + state->vm_logs[i] = vm->log[idx]; + } + + mutex_unlock(&vm->mmu_lock); +} + static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, - struct msm_gem_submit *submit, char *comm, char *cmd) + struct msm_gem_submit *submit, struct msm_gpu_fault_info *fault_info, + char *comm, char *cmd) { struct msm_gpu_state *state; @@ -276,28 +378,21 @@ static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, /* Fill in the additional crash state information */ state->comm = kstrdup(comm, GFP_KERNEL); state->cmd = kstrdup(cmd, GFP_KERNEL); - state->fault_info = gpu->fault_info; - - if (submit) { - int i; - - if (state->fault_info.ttbr0) { - struct msm_gpu_fault_info *info = &state->fault_info; - struct msm_mmu *mmu = submit->aspace->mmu; + if (fault_info) + state->fault_info = *fault_info; - msm_iommu_pagetable_params(mmu, &info->pgtbl_ttbr0, - &info->asid); - msm_iommu_pagetable_walk(mmu, info->iova, info->ptes); - } + if (submit && state->fault_info.ttbr0) { + struct msm_gpu_fault_info *info = &state->fault_info; + struct msm_mmu *mmu = to_msm_vm(submit->vm)->mmu; - state->bos = kcalloc(submit->nr_bos, - sizeof(struct msm_gpu_state_bo), GFP_KERNEL); + msm_iommu_pagetable_params(mmu, &info->pgtbl_ttbr0, + &info->asid); + msm_iommu_pagetable_walk(mmu, info->iova, info->ptes); + } - for (i = 0; state->bos && i < submit->nr_bos; i++) { - msm_gpu_crashstate_get_bo(state, submit->bos[i].obj, - submit->bos[i].iova, - should_dump(submit, i)); - } + if (submit) { + crashstate_get_vm_logs(state, to_msm_vm(submit->vm)); + crashstate_get_bos(state, submit); } /* Set the active crash state to be dumped on failure */ @@ -308,7 +403,8 @@ static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, } #else static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, - struct msm_gem_submit *submit, char *comm, char *cmd) + struct msm_gem_submit *submit, struct msm_gpu_fault_info *fault_info, + char *comm, char *cmd) { } #endif @@ -339,7 +435,7 @@ static void retire_submits(struct msm_gpu *gpu); static void get_comm_cmdline(struct msm_gem_submit *submit, char **comm, char **cmd) { - struct msm_file_private *ctx = submit->queue->ctx; + struct msm_context *ctx = submit->queue->ctx; struct task_struct *task; WARN_ON(!mutex_is_locked(&submit->gpu->lock)); @@ -386,8 +482,20 @@ static void recover_worker(struct kthread_work *work) /* Increment the fault counts */ submit->queue->faults++; - if (submit->aspace) - submit->aspace->faults++; + if (submit->vm) { + struct msm_gem_vm *vm = to_msm_vm(submit->vm); + + vm->faults++; + + /* + * If userspace has opted-in to VM_BIND (and therefore userspace + * management of the VM), faults mark the VM as unusuable. This + * matches vulkan expectations (vulkan is the main target for + * VM_BIND) + */ + if (!vm->managed) + msm_gem_vm_unusable(submit->vm); + } get_comm_cmdline(submit, &comm, &cmd); @@ -405,7 +513,7 @@ static void recover_worker(struct kthread_work *work) /* Record the crash state */ pm_runtime_get_sync(&gpu->pdev->dev); - msm_gpu_crashstate_capture(gpu, submit, comm, cmd); + msm_gpu_crashstate_capture(gpu, submit, NULL, comm, cmd); kfree(cmd); kfree(comm); @@ -459,9 +567,8 @@ out_unlock: msm_gpu_retire(gpu); } -static void fault_worker(struct kthread_work *work) +void msm_gpu_fault_crashstate_capture(struct msm_gpu *gpu, struct msm_gpu_fault_info *fault_info) { - struct msm_gpu *gpu = container_of(work, struct msm_gpu, fault_work); struct msm_gem_submit *submit; struct msm_ringbuffer *cur_ring = gpu->funcs->active_ring(gpu); char *comm = NULL, *cmd = NULL; @@ -484,16 +591,13 @@ static void fault_worker(struct kthread_work *work) /* Record the crash state */ pm_runtime_get_sync(&gpu->pdev->dev); - msm_gpu_crashstate_capture(gpu, submit, comm, cmd); + msm_gpu_crashstate_capture(gpu, submit, fault_info, comm, cmd); pm_runtime_put_sync(&gpu->pdev->dev); kfree(cmd); kfree(comm); resume_smmu: - memset(&gpu->fault_info, 0, sizeof(gpu->fault_info)); - gpu->aspace->mmu->funcs->resume_translation(gpu->aspace->mmu); - mutex_unlock(&gpu->lock); } @@ -829,10 +933,12 @@ static int get_clocks(struct platform_device *pdev, struct msm_gpu *gpu) } /* Return a new address space for a msm_drm_private instance */ -struct msm_gem_address_space * -msm_gpu_create_private_address_space(struct msm_gpu *gpu, struct task_struct *task) +struct drm_gpuvm * +msm_gpu_create_private_vm(struct msm_gpu *gpu, struct task_struct *task, + bool kernel_managed) { - struct msm_gem_address_space *aspace = NULL; + struct drm_gpuvm *vm = NULL; + if (!gpu) return NULL; @@ -840,16 +946,16 @@ msm_gpu_create_private_address_space(struct msm_gpu *gpu, struct task_struct *ta * If the target doesn't support private address spaces then return * the global one */ - if (gpu->funcs->create_private_address_space) { - aspace = gpu->funcs->create_private_address_space(gpu); - if (!IS_ERR(aspace)) - aspace->pid = get_pid(task_pid(task)); + if (gpu->funcs->create_private_vm) { + vm = gpu->funcs->create_private_vm(gpu, kernel_managed); + if (!IS_ERR(vm)) + to_msm_vm(vm)->pid = get_pid(task_pid(task)); } - if (IS_ERR_OR_NULL(aspace)) - aspace = msm_gem_address_space_get(gpu->aspace); + if (IS_ERR_OR_NULL(vm)) + vm = drm_gpuvm_get(gpu->vm); - return aspace; + return vm; } int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, @@ -882,7 +988,6 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, init_waitqueue_head(&gpu->retire_event); kthread_init_work(&gpu->retire_work, retire_worker); kthread_init_work(&gpu->recover_work, recover_worker); - kthread_init_work(&gpu->fault_work, fault_worker); priv->hangcheck_period = DRM_MSM_HANGCHECK_DEFAULT_PERIOD; @@ -944,19 +1049,15 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, msm_devfreq_init(gpu); - - gpu->aspace = gpu->funcs->create_address_space(gpu, pdev); - - if (gpu->aspace == NULL) - DRM_DEV_INFO(drm->dev, "%s: no IOMMU, fallback to VRAM carveout!\n", name); - else if (IS_ERR(gpu->aspace)) { - ret = PTR_ERR(gpu->aspace); + gpu->vm = gpu->funcs->create_vm(gpu, pdev); + if (IS_ERR(gpu->vm)) { + ret = PTR_ERR(gpu->vm); goto fail; } memptrs = msm_gem_kernel_new(drm, sizeof(struct msm_rbmemptrs) * nr_rings, - check_apriv(gpu, MSM_BO_WC), gpu->aspace, &gpu->memptrs_bo, + check_apriv(gpu, MSM_BO_WC), gpu->vm, &gpu->memptrs_bo, &memptrs_iova); if (IS_ERR(memptrs)) { @@ -1000,7 +1101,7 @@ fail: gpu->rb[i] = NULL; } - msm_gem_kernel_put(gpu->memptrs_bo, gpu->aspace); + msm_gem_kernel_put(gpu->memptrs_bo, gpu->vm); platform_set_drvdata(pdev, NULL); return ret; @@ -1017,11 +1118,12 @@ void msm_gpu_cleanup(struct msm_gpu *gpu) gpu->rb[i] = NULL; } - msm_gem_kernel_put(gpu->memptrs_bo, gpu->aspace); + msm_gem_kernel_put(gpu->memptrs_bo, gpu->vm); - if (!IS_ERR_OR_NULL(gpu->aspace)) { - gpu->aspace->mmu->funcs->detach(gpu->aspace->mmu); - msm_gem_address_space_put(gpu->aspace); + if (!IS_ERR_OR_NULL(gpu->vm)) { + struct msm_mmu *mmu = to_msm_vm(gpu->vm)->mmu; + mmu->funcs->detach(mmu); + drm_gpuvm_put(gpu->vm); } if (gpu->worker) { diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index e25009150579..b2a96544f92a 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -20,9 +20,10 @@ #include "msm_gem.h" struct msm_gem_submit; +struct msm_gem_vm_log_entry; struct msm_gpu_perfcntr; struct msm_gpu_state; -struct msm_file_private; +struct msm_context; struct msm_gpu_config { const char *ioname; @@ -44,9 +45,9 @@ struct msm_gpu_config { * + z180_gpu */ struct msm_gpu_funcs { - int (*get_param)(struct msm_gpu *gpu, struct msm_file_private *ctx, + int (*get_param)(struct msm_gpu *gpu, struct msm_context *ctx, uint32_t param, uint64_t *value, uint32_t *len); - int (*set_param)(struct msm_gpu *gpu, struct msm_file_private *ctx, + int (*set_param)(struct msm_gpu *gpu, struct msm_context *ctx, uint32_t param, uint64_t value, uint32_t len); int (*hw_init)(struct msm_gpu *gpu); @@ -78,10 +79,8 @@ struct msm_gpu_funcs { /* note: gpu_set_freq() can assume that we have been pm_resumed */ void (*gpu_set_freq)(struct msm_gpu *gpu, struct dev_pm_opp *opp, bool suspended); - struct msm_gem_address_space *(*create_address_space) - (struct msm_gpu *gpu, struct platform_device *pdev); - struct msm_gem_address_space *(*create_private_address_space) - (struct msm_gpu *gpu); + struct drm_gpuvm *(*create_vm)(struct msm_gpu *gpu, struct platform_device *pdev); + struct drm_gpuvm *(*create_private_vm)(struct msm_gpu *gpu, bool kernel_managed); uint32_t (*get_rptr)(struct msm_gpu *gpu, struct msm_ringbuffer *ring); /** @@ -236,7 +235,7 @@ struct msm_gpu { void __iomem *mmio; int irq; - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; /* Power Control: */ struct regulator *gpu_reg, *gpu_cx; @@ -253,12 +252,6 @@ struct msm_gpu { #define DRM_MSM_HANGCHECK_PROGRESS_RETRIES 3 struct timer_list hangcheck_timer; - /* Fault info for most recent iova fault: */ - struct msm_gpu_fault_info fault_info; - - /* work for handling GPU ioval faults: */ - struct kthread_work fault_work; - /* work for handling GPU recovery: */ struct kthread_work recover_work; @@ -347,26 +340,61 @@ struct msm_gpu_perfcntr { #define NR_SCHED_PRIORITIES (1 + DRM_SCHED_PRIORITY_LOW - DRM_SCHED_PRIORITY_HIGH) /** - * struct msm_file_private - per-drm_file context - * - * @queuelock: synchronizes access to submitqueues list - * @submitqueues: list of &msm_gpu_submitqueue created by userspace - * @queueid: counter incremented each time a submitqueue is created, - * used to assign &msm_gpu_submitqueue.id - * @aspace: the per-process GPU address-space - * @ref: reference count - * @seqno: unique per process seqno + * struct msm_context - per-drm_file context */ -struct msm_file_private { +struct msm_context { + /** @queuelock: synchronizes access to submitqueues list */ rwlock_t queuelock; + + /** @submitqueues: list of &msm_gpu_submitqueue created by userspace */ struct list_head submitqueues; + + /** + * @queueid: + * + * Counter incremented each time a submitqueue is created, used to + * assign &msm_gpu_submitqueue.id + */ int queueid; - struct msm_gem_address_space *aspace; + + /** + * @closed: The device file associated with this context has been closed. + * + * Once the device is closed, any submits that have not been written + * to the ring buffer are no-op'd. + */ + bool closed; + + /** + * @userspace_managed_vm: + * + * Has userspace opted-in to userspace managed VM (ie. VM_BIND) via + * MSM_PARAM_EN_VM_BIND? + */ + bool userspace_managed_vm; + + /** + * @vm: + * + * The per-process GPU address-space. Do not access directly, use + * msm_context_vm(). + */ + struct drm_gpuvm *vm; + + /** @kref: the reference count */ struct kref ref; + + /** + * @seqno: + * + * A unique per-process sequence number. Used to detect context + * switches, without relying on keeping a, potentially dangling, + * pointer to the previous context. + */ int seqno; /** - * sysprof: + * @sysprof: * * The value of MSM_PARAM_SYSPROF set by userspace. This is * intended to be used by system profiling tools like Mesa's @@ -384,21 +412,21 @@ struct msm_file_private { int sysprof; /** - * comm: Overridden task comm, see MSM_PARAM_COMM + * @comm: Overridden task comm, see MSM_PARAM_COMM * * Accessed under msm_gpu::lock */ char *comm; /** - * cmdline: Overridden task cmdline, see MSM_PARAM_CMDLINE + * @cmdline: Overridden task cmdline, see MSM_PARAM_CMDLINE * * Accessed under msm_gpu::lock */ char *cmdline; /** - * elapsed: + * @elapsed: * * The total (cumulative) elapsed time GPU was busy with rendering * from this context in ns. @@ -406,7 +434,7 @@ struct msm_file_private { uint64_t elapsed_ns; /** - * cycles: + * @cycles: * * The total (cumulative) GPU cycles elapsed attributed to this * context. @@ -414,7 +442,7 @@ struct msm_file_private { uint64_t cycles; /** - * entities: + * @entities: * * Table of per-priority-level sched entities used by submitqueues * associated with this &drm_file. Because some userspace apps @@ -427,7 +455,7 @@ struct msm_file_private { struct drm_sched_entity *entities[NR_SCHED_PRIORITIES * MSM_GPU_MAX_RINGS]; /** - * ctx_mem: + * @ctx_mem: * * Total amount of memory of GEM buffers with handles attached for * this context. @@ -435,6 +463,24 @@ struct msm_file_private { atomic64_t ctx_mem; }; +struct drm_gpuvm *msm_context_vm(struct drm_device *dev, struct msm_context *ctx); + +/** + * msm_context_is_vm_bind() - has userspace opted in to VM_BIND? + * + * @ctx: the drm_file context + * + * See MSM_PARAM_EN_VM_BIND. If userspace is managing the VM, it can + * do sparse binding including having multiple, potentially partial, + * mappings in the VM. Therefore certain legacy uabi (ie. GET_IOVA, + * SET_IOVA) are rejected because they don't have a sensible meaning. + */ +static inline bool +msm_context_is_vmbind(struct msm_context *ctx) +{ + return ctx->userspace_managed_vm; +} + /** * msm_gpu_convert_priority - Map userspace priority to ring # and sched priority * @@ -512,13 +558,16 @@ struct msm_gpu_submitqueue { u32 ring_nr; int faults; uint32_t last_fence; - struct msm_file_private *ctx; + struct msm_context *ctx; struct list_head node; struct idr fence_idr; struct spinlock idr_lock; struct mutex lock; struct kref ref; struct drm_sched_entity *entity; + + /** @_vm_bind_entity: used for @entity pointer for VM_BIND queues */ + struct drm_sched_entity _vm_bind_entity[0]; }; struct msm_gpu_state_bo { @@ -555,6 +604,9 @@ struct msm_gpu_state { struct msm_gpu_fault_info fault_info; + int nr_vm_logs; + struct msm_gem_vm_log_entry *vm_logs; + int nr_bos; struct msm_gpu_state_bo *bos; }; @@ -608,33 +660,32 @@ static inline void gpu_write64(struct msm_gpu *gpu, u32 reg, u64 val) int msm_gpu_pm_suspend(struct msm_gpu *gpu); int msm_gpu_pm_resume(struct msm_gpu *gpu); -void msm_gpu_show_fdinfo(struct msm_gpu *gpu, struct msm_file_private *ctx, +void msm_gpu_show_fdinfo(struct msm_gpu *gpu, struct msm_context *ctx, struct drm_printer *p); -int msm_submitqueue_init(struct drm_device *drm, struct msm_file_private *ctx); -struct msm_gpu_submitqueue *msm_submitqueue_get(struct msm_file_private *ctx, +int msm_submitqueue_init(struct drm_device *drm, struct msm_context *ctx); +struct msm_gpu_submitqueue *msm_submitqueue_get(struct msm_context *ctx, u32 id); int msm_submitqueue_create(struct drm_device *drm, - struct msm_file_private *ctx, + struct msm_context *ctx, u32 prio, u32 flags, u32 *id); -int msm_submitqueue_query(struct drm_device *drm, struct msm_file_private *ctx, +int msm_submitqueue_query(struct drm_device *drm, struct msm_context *ctx, struct drm_msm_submitqueue_query *args); -int msm_submitqueue_remove(struct msm_file_private *ctx, u32 id); -void msm_submitqueue_close(struct msm_file_private *ctx); +int msm_submitqueue_remove(struct msm_context *ctx, u32 id); +void msm_submitqueue_close(struct msm_context *ctx); void msm_submitqueue_destroy(struct kref *kref); -int msm_file_private_set_sysprof(struct msm_file_private *ctx, - struct msm_gpu *gpu, int sysprof); -void __msm_file_private_destroy(struct kref *kref); +int msm_context_set_sysprof(struct msm_context *ctx, struct msm_gpu *gpu, int sysprof); +void __msm_context_destroy(struct kref *kref); -static inline void msm_file_private_put(struct msm_file_private *ctx) +static inline void msm_context_put(struct msm_context *ctx) { - kref_put(&ctx->ref, __msm_file_private_destroy); + kref_put(&ctx->ref, __msm_context_destroy); } -static inline struct msm_file_private *msm_file_private_get( - struct msm_file_private *ctx) +static inline struct msm_context *msm_context_get( + struct msm_context *ctx) { kref_get(&ctx->ref); return ctx; @@ -662,12 +713,14 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, struct msm_gpu *gpu, const struct msm_gpu_funcs *funcs, const char *name, struct msm_gpu_config *config); -struct msm_gem_address_space * -msm_gpu_create_private_address_space(struct msm_gpu *gpu, struct task_struct *task); +struct drm_gpuvm * +msm_gpu_create_private_vm(struct msm_gpu *gpu, struct task_struct *task, + bool kernel_managed); void msm_gpu_cleanup(struct msm_gpu *gpu); struct msm_gpu *adreno_load_gpu(struct drm_device *dev); +bool adreno_has_gpu(struct device_node *node); void __init adreno_register(void); void __exit adreno_unregister(void); @@ -705,6 +758,8 @@ static inline void msm_gpu_crashstate_put(struct msm_gpu *gpu) mutex_unlock(&gpu->lock); } +void msm_gpu_fault_crashstate_capture(struct msm_gpu *gpu, struct msm_gpu_fault_info *fault_info); + /* * Simple macro to semi-cleanly add the MAP_PRIV flag for targets that can * support expanded privileges diff --git a/drivers/gpu/drm/msm/msm_gpu_trace.h b/drivers/gpu/drm/msm/msm_gpu_trace.h index 7f863282db0d..781bbe5540bd 100644 --- a/drivers/gpu/drm/msm/msm_gpu_trace.h +++ b/drivers/gpu/drm/msm/msm_gpu_trace.h @@ -205,6 +205,20 @@ TRACE_EVENT(msm_gpu_preemption_irq, TP_printk("preempted to %u", __entry->ring_id) ); +TRACE_EVENT(msm_mmu_prealloc_cleanup, + TP_PROTO(u32 count, u32 remaining), + TP_ARGS(count, remaining), + TP_STRUCT__entry( + __field(u32, count) + __field(u32, remaining) + ), + TP_fast_assign( + __entry->count = count; + __entry->remaining = remaining; + ), + TP_printk("count=%u, remaining=%u", __entry->count, __entry->remaining) +); + #endif #undef TRACE_INCLUDE_PATH diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c index fd73dcd3f30e..55c29f49b788 100644 --- a/drivers/gpu/drm/msm/msm_iommu.c +++ b/drivers/gpu/drm/msm/msm_iommu.c @@ -6,13 +6,18 @@ #include <linux/adreno-smmu-priv.h> #include <linux/io-pgtable.h> +#include <linux/kmemleak.h> #include "msm_drv.h" +#include "msm_gpu_trace.h" #include "msm_mmu.h" struct msm_iommu { struct msm_mmu base; struct iommu_domain *domain; atomic_t pagetables; + struct page *prr_page; + + struct kmem_cache *pt_cache; }; #define to_msm_iommu(x) container_of(x, struct msm_iommu, base) @@ -26,6 +31,9 @@ struct msm_iommu_pagetable { unsigned long pgsize_bitmap; /* Bitmap of page sizes in use */ phys_addr_t ttbr; u32 asid; + + /** @root_page_table: Stores the root page table pointer. */ + void *root_page_table; }; static struct msm_iommu_pagetable *to_pagetable(struct msm_mmu *mmu) { @@ -93,15 +101,24 @@ static int msm_iommu_pagetable_unmap(struct msm_mmu *mmu, u64 iova, { struct msm_iommu_pagetable *pagetable = to_pagetable(mmu); struct io_pgtable_ops *ops = pagetable->pgtbl_ops; + int ret = 0; while (size) { - size_t unmapped, pgsize, count; + size_t pgsize, count; + ssize_t unmapped; pgsize = calc_pgsize(pagetable, iova, iova, size, &count); unmapped = ops->unmap_pages(ops, iova, pgsize, count, NULL); - if (!unmapped) - break; + if (unmapped <= 0) { + ret = -EINVAL; + /* + * Continue attempting to unamp the remained of the + * range, so we don't end up with some dangling + * mapped pages + */ + unmapped = PAGE_SIZE; + } iova += unmapped; size -= unmapped; @@ -109,11 +126,42 @@ static int msm_iommu_pagetable_unmap(struct msm_mmu *mmu, u64 iova, iommu_flush_iotlb_all(to_msm_iommu(pagetable->parent)->domain); - return (size == 0) ? 0 : -EINVAL; + return ret; +} + +static int msm_iommu_pagetable_map_prr(struct msm_mmu *mmu, u64 iova, size_t len, int prot) +{ + struct msm_iommu_pagetable *pagetable = to_pagetable(mmu); + struct io_pgtable_ops *ops = pagetable->pgtbl_ops; + struct msm_iommu *iommu = to_msm_iommu(pagetable->parent); + phys_addr_t phys = page_to_phys(iommu->prr_page); + u64 addr = iova; + + while (len) { + size_t mapped = 0; + size_t size = PAGE_SIZE; + int ret; + + ret = ops->map_pages(ops, addr, phys, size, 1, prot, GFP_KERNEL, &mapped); + + /* map_pages could fail after mapping some of the pages, + * so update the counters before error handling. + */ + addr += mapped; + len -= mapped; + + if (ret) { + msm_iommu_pagetable_unmap(mmu, iova, addr - iova); + return -EINVAL; + } + } + + return 0; } static int msm_iommu_pagetable_map(struct msm_mmu *mmu, u64 iova, - struct sg_table *sgt, size_t len, int prot) + struct sg_table *sgt, size_t off, size_t len, + int prot) { struct msm_iommu_pagetable *pagetable = to_pagetable(mmu); struct io_pgtable_ops *ops = pagetable->pgtbl_ops; @@ -121,10 +169,26 @@ static int msm_iommu_pagetable_map(struct msm_mmu *mmu, u64 iova, u64 addr = iova; unsigned int i; + if (!sgt) + return msm_iommu_pagetable_map_prr(mmu, iova, len, prot); + for_each_sgtable_sg(sgt, sg, i) { size_t size = sg->length; phys_addr_t phys = sg_phys(sg); + if (!len) + break; + + if (size <= off) { + off -= size; + continue; + } + + phys += off; + size -= off; + size = min_t(size_t, size, len); + off = 0; + while (size) { size_t pgsize, count, mapped = 0; int ret; @@ -140,6 +204,7 @@ static int msm_iommu_pagetable_map(struct msm_mmu *mmu, u64 iova, phys += mapped; addr += mapped; size -= mapped; + len -= mapped; if (ret) { msm_iommu_pagetable_unmap(mmu, iova, addr - iova); @@ -162,9 +227,16 @@ static void msm_iommu_pagetable_destroy(struct msm_mmu *mmu) * If this is the last attached pagetable for the parent, * disable TTBR0 in the arm-smmu driver */ - if (atomic_dec_return(&iommu->pagetables) == 0) + if (atomic_dec_return(&iommu->pagetables) == 0) { adreno_smmu->set_ttbr0_cfg(adreno_smmu->cookie, NULL); + if (adreno_smmu->set_prr_bit) { + adreno_smmu->set_prr_bit(adreno_smmu->cookie, false); + __free_page(iommu->prr_page); + iommu->prr_page = NULL; + } + } + free_io_pgtable_ops(pagetable->pgtbl_ops); kfree(pagetable); } @@ -217,7 +289,148 @@ msm_iommu_pagetable_walk(struct msm_mmu *mmu, unsigned long iova, uint64_t ptes[ return 0; } +static void +msm_iommu_pagetable_prealloc_count(struct msm_mmu *mmu, struct msm_mmu_prealloc *p, + uint64_t iova, size_t len) +{ + u64 pt_count; + + /* + * L1, L2 and L3 page tables. + * + * We could optimize L3 allocation by iterating over the sgt and merging + * 2M contiguous blocks, but it's simpler to over-provision and return + * the pages if they're not used. + * + * The first level descriptor (v8 / v7-lpae page table format) encodes + * 30 bits of address. The second level encodes 29. For the 3rd it is + * 39. + * + * https://developer.arm.com/documentation/ddi0406/c/System-Level-Architecture/Virtual-Memory-System-Architecture--VMSA-/Long-descriptor-translation-table-format/Long-descriptor-translation-table-format-descriptors?lang=en#BEIHEFFB + */ + pt_count = ((ALIGN(iova + len, 1ull << 39) - ALIGN_DOWN(iova, 1ull << 39)) >> 39) + + ((ALIGN(iova + len, 1ull << 30) - ALIGN_DOWN(iova, 1ull << 30)) >> 30) + + ((ALIGN(iova + len, 1ull << 21) - ALIGN_DOWN(iova, 1ull << 21)) >> 21); + + p->count += pt_count; +} + +static struct kmem_cache * +get_pt_cache(struct msm_mmu *mmu) +{ + struct msm_iommu_pagetable *pagetable = to_pagetable(mmu); + return to_msm_iommu(pagetable->parent)->pt_cache; +} + +static int +msm_iommu_pagetable_prealloc_allocate(struct msm_mmu *mmu, struct msm_mmu_prealloc *p) +{ + struct kmem_cache *pt_cache = get_pt_cache(mmu); + int ret; + + p->pages = kvmalloc_array(p->count, sizeof(p->pages), GFP_KERNEL); + if (!p->pages) + return -ENOMEM; + + ret = kmem_cache_alloc_bulk(pt_cache, GFP_KERNEL, p->count, p->pages); + if (ret != p->count) { + p->count = ret; + return -ENOMEM; + } + + return 0; +} + +static void +msm_iommu_pagetable_prealloc_cleanup(struct msm_mmu *mmu, struct msm_mmu_prealloc *p) +{ + struct kmem_cache *pt_cache = get_pt_cache(mmu); + uint32_t remaining_pt_count = p->count - p->ptr; + + if (p->count > 0) + trace_msm_mmu_prealloc_cleanup(p->count, remaining_pt_count); + + kmem_cache_free_bulk(pt_cache, remaining_pt_count, &p->pages[p->ptr]); + kvfree(p->pages); +} + +/** + * alloc_pt() - Custom page table allocator + * @cookie: Cookie passed at page table allocation time. + * @size: Size of the page table. This size should be fixed, + * and determined at creation time based on the granule size. + * @gfp: GFP flags. + * + * We want a custom allocator so we can use a cache for page table + * allocations and amortize the cost of the over-reservation that's + * done to allow asynchronous VM operations. + * + * Return: non-NULL on success, NULL if the allocation failed for any + * reason. + */ +static void * +msm_iommu_pagetable_alloc_pt(void *cookie, size_t size, gfp_t gfp) +{ + struct msm_iommu_pagetable *pagetable = cookie; + struct msm_mmu_prealloc *p = pagetable->base.prealloc; + void *page; + + /* Allocation of the root page table happening during init. */ + if (unlikely(!pagetable->root_page_table)) { + struct page *p; + + p = alloc_pages_node(dev_to_node(pagetable->iommu_dev), + gfp | __GFP_ZERO, get_order(size)); + page = p ? page_address(p) : NULL; + pagetable->root_page_table = page; + return page; + } + + if (WARN_ON(!p) || WARN_ON(p->ptr >= p->count)) + return NULL; + + page = p->pages[p->ptr++]; + memset(page, 0, size); + + /* + * Page table entries don't use virtual addresses, which trips out + * kmemleak. kmemleak_alloc_phys() might work, but physical addresses + * are mixed with other fields, and I fear kmemleak won't detect that + * either. + * + * Let's just ignore memory passed to the page-table driver for now. + */ + kmemleak_ignore(page); + + return page; +} + + +/** + * free_pt() - Custom page table free function + * @cookie: Cookie passed at page table allocation time. + * @data: Page table to free. + * @size: Size of the page table. This size should be fixed, + * and determined at creation time based on the granule size. + */ +static void +msm_iommu_pagetable_free_pt(void *cookie, void *data, size_t size) +{ + struct msm_iommu_pagetable *pagetable = cookie; + + if (unlikely(pagetable->root_page_table == data)) { + free_pages((unsigned long)data, get_order(size)); + pagetable->root_page_table = NULL; + return; + } + + kmem_cache_free(get_pt_cache(&pagetable->base), data); +} + static const struct msm_mmu_funcs pagetable_funcs = { + .prealloc_count = msm_iommu_pagetable_prealloc_count, + .prealloc_allocate = msm_iommu_pagetable_prealloc_allocate, + .prealloc_cleanup = msm_iommu_pagetable_prealloc_cleanup, .map = msm_iommu_pagetable_map, .unmap = msm_iommu_pagetable_unmap, .destroy = msm_iommu_pagetable_destroy, @@ -268,7 +481,18 @@ static const struct iommu_flush_ops tlb_ops = { static int msm_gpu_fault_handler(struct iommu_domain *domain, struct device *dev, unsigned long iova, int flags, void *arg); -struct msm_mmu *msm_iommu_pagetable_create(struct msm_mmu *parent) +static size_t get_tblsz(const struct io_pgtable_cfg *cfg) +{ + int pg_shift, bits_per_level; + + pg_shift = __ffs(cfg->pgsize_bitmap); + /* arm_lpae_iopte is u64: */ + bits_per_level = pg_shift - ilog2(sizeof(u64)); + + return sizeof(u64) << bits_per_level; +} + +struct msm_mmu *msm_iommu_pagetable_create(struct msm_mmu *parent, bool kernel_managed) { struct adreno_smmu_priv *adreno_smmu = dev_get_drvdata(parent->dev); struct msm_iommu *iommu = to_msm_iommu(parent); @@ -302,6 +526,36 @@ struct msm_mmu *msm_iommu_pagetable_create(struct msm_mmu *parent) ttbr0_cfg.quirks &= ~IO_PGTABLE_QUIRK_ARM_TTBR1; ttbr0_cfg.tlb = &tlb_ops; + if (!kernel_managed) { + ttbr0_cfg.quirks |= IO_PGTABLE_QUIRK_NO_WARN; + + /* + * With userspace managed VM (aka VM_BIND), we need to pre- + * allocate pages ahead of time for map/unmap operations, + * handing them to io-pgtable via custom alloc/free ops as + * needed: + */ + ttbr0_cfg.alloc = msm_iommu_pagetable_alloc_pt; + ttbr0_cfg.free = msm_iommu_pagetable_free_pt; + + /* + * Restrict to single page granules. Otherwise we may run + * into a situation where userspace wants to unmap/remap + * only a part of a larger block mapping, which is not + * possible without unmapping the entire block. Which in + * turn could cause faults if the GPU is accessing other + * parts of the block mapping. + * + * Note that prior to commit 33729a5fc0ca ("iommu/io-pgtable-arm: + * Remove split on unmap behavior)" this was handled in + * io-pgtable-arm. But this apparently does not work + * correctly on SMMUv3. + */ + WARN_ON(!(ttbr0_cfg.pgsize_bitmap & PAGE_SIZE)); + ttbr0_cfg.pgsize_bitmap = PAGE_SIZE; + } + + pagetable->iommu_dev = ttbr1_cfg->iommu_dev; pagetable->pgtbl_ops = alloc_io_pgtable_ops(ARM_64_LPAE_S1, &ttbr0_cfg, pagetable); @@ -321,12 +575,30 @@ struct msm_mmu *msm_iommu_pagetable_create(struct msm_mmu *parent) kfree(pagetable); return ERR_PTR(ret); } + + BUG_ON(iommu->prr_page); + if (adreno_smmu->set_prr_bit) { + /* + * We need a zero'd page for two reasons: + * + * 1) Reserve a known physical address to use when + * mapping NULL / sparsely resident regions + * 2) Read back zero + * + * It appears the hw drops writes to the PRR region + * on the floor, but reads actually return whatever + * is in the PRR page. + */ + iommu->prr_page = alloc_page(GFP_KERNEL | __GFP_ZERO); + adreno_smmu->set_prr_addr(adreno_smmu->cookie, + page_to_phys(iommu->prr_page)); + adreno_smmu->set_prr_bit(adreno_smmu->cookie, true); + } } /* Needed later for TLB flush */ pagetable->parent = parent; pagetable->tlb = ttbr1_cfg->tlb; - pagetable->iommu_dev = ttbr1_cfg->iommu_dev; pagetable->pgsize_bitmap = ttbr0_cfg.pgsize_bitmap; pagetable->ttbr = ttbr0_cfg.arm_lpae_s1_cfg.ttbr; @@ -345,7 +617,6 @@ static int msm_gpu_fault_handler(struct iommu_domain *domain, struct device *dev unsigned long iova, int flags, void *arg) { struct msm_iommu *iommu = arg; - struct msm_mmu *mmu = &iommu->base; struct adreno_smmu_priv *adreno_smmu = dev_get_drvdata(iommu->base.dev); struct adreno_smmu_fault_info info, *ptr = NULL; @@ -359,9 +630,6 @@ static int msm_gpu_fault_handler(struct iommu_domain *domain, struct device *dev pr_warn_ratelimited("*** fault: iova=%16lx, flags=%d\n", iova, flags); - if (mmu->funcs->resume_translation) - mmu->funcs->resume_translation(mmu); - return 0; } @@ -376,12 +644,12 @@ static int msm_disp_fault_handler(struct iommu_domain *domain, struct device *de return -ENOSYS; } -static void msm_iommu_resume_translation(struct msm_mmu *mmu) +static void msm_iommu_set_stall(struct msm_mmu *mmu, bool enable) { struct adreno_smmu_priv *adreno_smmu = dev_get_drvdata(mmu->dev); - if (adreno_smmu->resume_translation) - adreno_smmu->resume_translation(adreno_smmu->cookie, true); + if (adreno_smmu->set_stall) + adreno_smmu->set_stall(adreno_smmu->cookie, enable); } static void msm_iommu_detach(struct msm_mmu *mmu) @@ -392,11 +660,14 @@ static void msm_iommu_detach(struct msm_mmu *mmu) } static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova, - struct sg_table *sgt, size_t len, int prot) + struct sg_table *sgt, size_t off, size_t len, + int prot) { struct msm_iommu *iommu = to_msm_iommu(mmu); size_t ret; + WARN_ON(off != 0); + /* The arm-smmu driver expects the addresses to be sign extended */ if (iova & BIT_ULL(48)) iova |= GENMASK_ULL(63, 49); @@ -423,6 +694,7 @@ static void msm_iommu_destroy(struct msm_mmu *mmu) { struct msm_iommu *iommu = to_msm_iommu(mmu); iommu_domain_free(iommu->domain); + kmem_cache_destroy(iommu->pt_cache); kfree(iommu); } @@ -431,7 +703,7 @@ static const struct msm_mmu_funcs funcs = { .map = msm_iommu_map, .unmap = msm_iommu_unmap, .destroy = msm_iommu_destroy, - .resume_translation = msm_iommu_resume_translation, + .set_stall = msm_iommu_set_stall, }; struct msm_mmu *msm_iommu_new(struct device *dev, unsigned long quirks) @@ -496,6 +768,14 @@ struct msm_mmu *msm_iommu_gpu_new(struct device *dev, struct msm_gpu *gpu, unsig return mmu; iommu = to_msm_iommu(mmu); + if (adreno_smmu && adreno_smmu->cookie) { + const struct io_pgtable_cfg *cfg = + adreno_smmu->get_ttbr1_cfg(adreno_smmu->cookie); + size_t tblsz = get_tblsz(cfg); + + iommu->pt_cache = + kmem_cache_create("msm-mmu-pt", tblsz, tblsz, 0, NULL); + } iommu_set_fault_handler(iommu->domain, msm_gpu_fault_handler, iommu); /* Enable stall on iommu fault: */ diff --git a/drivers/gpu/drm/msm/msm_kms.c b/drivers/gpu/drm/msm/msm_kms.c index 35d5397e73b4..6889f1c1e721 100644 --- a/drivers/gpu/drm/msm/msm_kms.c +++ b/drivers/gpu/drm/msm/msm_kms.c @@ -13,6 +13,7 @@ #include <drm/drm_drv.h> #include <drm/drm_mode_config.h> #include <drm/drm_vblank.h> +#include <drm/clients/drm_client_setup.h> #include "disp/msm_disp_snapshot.h" #include "msm_drv.h" @@ -137,7 +138,7 @@ static int vblank_ctrl_queue_work(struct msm_drm_private *priv, vbl_work->enable = enable; vbl_work->priv = priv; - queue_work(priv->wq, &vbl_work->work); + queue_work(priv->kms->wq, &vbl_work->work); return 0; } @@ -176,9 +177,9 @@ static int msm_kms_fault_handler(void *arg, unsigned long iova, int flags, void return -ENOSYS; } -struct msm_gem_address_space *msm_kms_init_aspace(struct drm_device *dev) +struct drm_gpuvm *msm_kms_init_vm(struct drm_device *dev) { - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; struct msm_mmu *mmu; struct device *mdp_dev = dev->dev; struct device *mdss_dev = mdp_dev->parent; @@ -204,17 +205,26 @@ struct msm_gem_address_space *msm_kms_init_aspace(struct drm_device *dev) return NULL; } - aspace = msm_gem_address_space_create(mmu, "mdp_kms", - 0x1000, 0x100000000 - 0x1000); - if (IS_ERR(aspace)) { - dev_err(mdp_dev, "aspace create, error %pe\n", aspace); + vm = msm_gem_vm_create(dev, mmu, "mdp_kms", + 0x1000, 0x100000000 - 0x1000, true); + if (IS_ERR(vm)) { + dev_err(mdp_dev, "vm create, error %pe\n", vm); mmu->funcs->destroy(mmu); - return aspace; + return vm; } - msm_mmu_set_fault_handler(aspace->mmu, kms, msm_kms_fault_handler); + msm_mmu_set_fault_handler(to_msm_vm(vm)->mmu, kms, msm_kms_fault_handler); - return aspace; + return vm; +} + +void msm_drm_kms_unregister(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct msm_drm_private *priv = platform_get_drvdata(pdev); + struct drm_device *ddev = priv->dev; + + drm_atomic_helper_shutdown(ddev); } void msm_drm_kms_uninit(struct device *dev) @@ -227,10 +237,17 @@ void msm_drm_kms_uninit(struct device *dev) BUG_ON(!kms); + /* We must cancel and cleanup any pending vblank enable/disable + * work before msm_irq_uninstall() to avoid work re-enabling an + * irq after uninstall has disabled it. + */ + + flush_workqueue(kms->wq); + /* clean up event worker threads */ - for (i = 0; i < priv->num_crtcs; i++) { - if (priv->event_thread[i].worker) - kthread_destroy_worker(priv->event_thread[i].worker); + for (i = 0; i < MAX_CRTCS; i++) { + if (kms->event_thread[i].worker) + kthread_destroy_worker(kms->event_thread[i].worker); } drm_kms_helper_poll_fini(ddev); @@ -261,7 +278,7 @@ int msm_drm_kms_init(struct device *dev, const struct drm_driver *drv) ret = priv->kms_init(ddev); if (ret) { DRM_DEV_ERROR(dev, "failed to load kms\n"); - return ret; + goto err_msm_uninit; } /* Enable normalization of plane zpos */ @@ -283,7 +300,7 @@ int msm_drm_kms_init(struct device *dev, const struct drm_driver *drv) struct msm_drm_thread *ev_thread; /* initialize event thread */ - ev_thread = &priv->event_thread[drm_crtc_index(crtc)]; + ev_thread = &kms->event_thread[drm_crtc_index(crtc)]; ev_thread->dev = ddev; ev_thread->worker = kthread_run_worker(0, "crtc_event:%d", crtc->base.id); if (IS_ERR(ev_thread->worker)) { @@ -296,7 +313,7 @@ int msm_drm_kms_init(struct device *dev, const struct drm_driver *drv) sched_set_fifo(ev_thread->worker->task); } - ret = drm_vblank_init(ddev, priv->num_crtcs); + ret = drm_vblank_init(ddev, ddev->mode_config.num_crtc); if (ret < 0) { DRM_DEV_ERROR(dev, "failed to initialize vblank\n"); goto err_msm_uninit; @@ -359,3 +376,13 @@ void msm_kms_shutdown(struct platform_device *pdev) if (drm && drm->registered && priv->kms) drm_atomic_helper_shutdown(drm); } + +void msm_drm_kms_post_init(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct msm_drm_private *priv = platform_get_drvdata(pdev); + struct drm_device *ddev = priv->dev; + + drm_kms_helper_poll_init(ddev); + drm_client_setup(ddev, NULL); +} diff --git a/drivers/gpu/drm/msm/msm_kms.h b/drivers/gpu/drm/msm/msm_kms.h index 43b58d052ee6..8a7be7b854de 100644 --- a/drivers/gpu/drm/msm/msm_kms.h +++ b/drivers/gpu/drm/msm/msm_kms.h @@ -13,6 +13,8 @@ #include "msm_drv.h" +#ifdef CONFIG_DRM_MSM_KMS + #define MAX_PLANE 4 /* As there are different display controller blocks depending on the @@ -127,10 +129,22 @@ struct msm_pending_timer { unsigned crtc_idx; }; +/* Commit/Event thread specific structure */ +struct msm_drm_thread { + struct drm_device *dev; + struct kthread_worker *worker; +}; + struct msm_kms { const struct msm_kms_funcs *funcs; struct drm_device *dev; + struct hdmi *hdmi; + + struct msm_dsi *dsi[MSM_DSI_CONTROLLER_COUNT]; + + struct msm_dp *dp[MSM_DP_CONTROLLER_COUNT]; + /* irq number to be passed on to msm_irq_install */ int irq; bool irq_requested; @@ -139,7 +153,7 @@ struct msm_kms { atomic_t fault_snapshot_capture; /* mapper-id used to request GEM buffer mapped for scanout: */ - struct msm_gem_address_space *aspace; + struct drm_gpuvm *vm; /* disp snapshot support */ struct kthread_worker *dump_worker; @@ -153,6 +167,9 @@ struct msm_kms { struct mutex commit_lock[MAX_CRTCS]; unsigned pending_crtc_mask; struct msm_pending_timer pending_timers[MAX_CRTCS]; + + struct workqueue_struct *wq; + struct msm_drm_thread event_thread[MAX_CRTCS]; }; static inline int msm_kms_init(struct msm_kms *kms, @@ -165,6 +182,10 @@ static inline int msm_kms_init(struct msm_kms *kms, kms->funcs = funcs; + kms->wq = alloc_ordered_workqueue("msm", 0); + if (!kms->wq) + return -ENOMEM; + for (i = 0; i < ARRAY_SIZE(kms->pending_timers); i++) { ret = msm_atomic_init_pending_timer(&kms->pending_timers[i], kms, i); if (ret) { @@ -181,6 +202,8 @@ static inline void msm_kms_destroy(struct msm_kms *kms) for (i = 0; i < ARRAY_SIZE(kms->pending_timers); i++) msm_atomic_destroy_pending_timer(&kms->pending_timers[i]); + + destroy_workqueue(kms->wq); } #define for_each_crtc_mask(dev, crtc, crtc_mask) \ @@ -192,6 +215,29 @@ static inline void msm_kms_destroy(struct msm_kms *kms) for_each_if (drm_crtc_mask(crtc) & (crtc_mask)) int msm_drm_kms_init(struct device *dev, const struct drm_driver *drv); +void msm_drm_kms_post_init(struct device *dev); +void msm_drm_kms_unregister(struct device *dev); void msm_drm_kms_uninit(struct device *dev); +#else /* ! CONFIG_DRM_MSM_KMS */ + +static inline int msm_drm_kms_init(struct device *dev, const struct drm_driver *drv) +{ + return -ENODEV; +} + +static inline void msm_drm_kms_post_init(struct device *dev) +{ +} + +static inline void msm_drm_kms_unregister(struct device *dev) +{ +} + +static inline void msm_drm_kms_uninit(struct device *dev) +{ +} + +#endif + #endif /* __MSM_KMS_H__ */ diff --git a/drivers/gpu/drm/msm/msm_mdss.c b/drivers/gpu/drm/msm/msm_mdss.c index 709979fcfab6..1f5fe7811e01 100644 --- a/drivers/gpu/drm/msm/msm_mdss.c +++ b/drivers/gpu/drm/msm/msm_mdss.c @@ -16,14 +16,17 @@ #include <linux/pm_runtime.h> #include <linux/reset.h> -#include "msm_mdss.h" +#include <linux/soc/qcom/ubwc.h> + #include "msm_kms.h" #include <generated/mdss.xml.h> #define MIN_IB_BW 400000000UL /* Min ib vote 400MB */ -#define DEFAULT_REG_BW 153600 /* Used in mdss fbdev driver */ +struct msm_mdss_data { + u32 reg_bus_bw; +}; struct msm_mdss { struct device *dev; @@ -36,7 +39,8 @@ struct msm_mdss { unsigned long enabled_mask; struct irq_domain *domain; } irq_controller; - const struct msm_mdss_data *mdss_data; + const struct qcom_ubwc_cfg_data *mdss_data; + u32 reg_bus_bw; struct icc_path *mdp_path[2]; u32 num_mdp_paths; struct icc_path *reg_bus_path; @@ -165,9 +169,9 @@ static int _msm_mdss_irq_domain_add(struct msm_mdss *msm_mdss) static void msm_mdss_setup_ubwc_dec_20(struct msm_mdss *msm_mdss) { - const struct msm_mdss_data *data = msm_mdss->mdss_data; + const struct qcom_ubwc_cfg_data *data = msm_mdss->mdss_data; u32 value = MDSS_UBWC_STATIC_UBWC_SWIZZLE(data->ubwc_swizzle) | - MDSS_UBWC_STATIC_HIGHEST_BANK_BIT(data->highest_bank_bit); + MDSS_UBWC_STATIC_HIGHEST_BANK_BIT(data->highest_bank_bit - 13); if (data->ubwc_bank_spread) value |= MDSS_UBWC_STATIC_UBWC_BANK_SPREAD; @@ -180,9 +184,9 @@ static void msm_mdss_setup_ubwc_dec_20(struct msm_mdss *msm_mdss) static void msm_mdss_setup_ubwc_dec_30(struct msm_mdss *msm_mdss) { - const struct msm_mdss_data *data = msm_mdss->mdss_data; + const struct qcom_ubwc_cfg_data *data = msm_mdss->mdss_data; u32 value = MDSS_UBWC_STATIC_UBWC_SWIZZLE(data->ubwc_swizzle & 0x1) | - MDSS_UBWC_STATIC_HIGHEST_BANK_BIT(data->highest_bank_bit); + MDSS_UBWC_STATIC_HIGHEST_BANK_BIT(data->highest_bank_bit - 13); if (data->macrotile_mode) value |= MDSS_UBWC_STATIC_MACROTILE_MODE; @@ -198,9 +202,9 @@ static void msm_mdss_setup_ubwc_dec_30(struct msm_mdss *msm_mdss) static void msm_mdss_setup_ubwc_dec_40(struct msm_mdss *msm_mdss) { - const struct msm_mdss_data *data = msm_mdss->mdss_data; + const struct qcom_ubwc_cfg_data *data = msm_mdss->mdss_data; u32 value = MDSS_UBWC_STATIC_UBWC_SWIZZLE(data->ubwc_swizzle) | - MDSS_UBWC_STATIC_HIGHEST_BANK_BIT(data->highest_bank_bit); + MDSS_UBWC_STATIC_HIGHEST_BANK_BIT(data->highest_bank_bit - 13); if (data->ubwc_bank_spread) value |= MDSS_UBWC_STATIC_UBWC_BANK_SPREAD; @@ -222,67 +226,22 @@ static void msm_mdss_setup_ubwc_dec_40(struct msm_mdss *msm_mdss) } } -#define MDSS_HW_MAJ_MIN \ - (MDSS_HW_VERSION_MAJOR__MASK | MDSS_HW_VERSION_MINOR__MASK) - -#define MDSS_HW_MSM8996 0x1007 -#define MDSS_HW_MSM8937 0x100e -#define MDSS_HW_MSM8953 0x1010 -#define MDSS_HW_MSM8998 0x3000 -#define MDSS_HW_SDM660 0x3002 -#define MDSS_HW_SDM630 0x3003 - -/* - * MDP5 platforms use generic qcom,mdp5 compat string, so we have to generate this data - */ -static const struct msm_mdss_data *msm_mdss_generate_mdp5_mdss_data(struct msm_mdss *mdss) +static void msm_mdss_setup_ubwc_dec_50(struct msm_mdss *msm_mdss) { - struct msm_mdss_data *data; - u32 hw_rev; - - data = devm_kzalloc(mdss->dev, sizeof(*data), GFP_KERNEL); - if (!data) - return NULL; - - hw_rev = readl_relaxed(mdss->mmio + REG_MDSS_HW_VERSION); - hw_rev = FIELD_GET(MDSS_HW_MAJ_MIN, hw_rev); - - if (hw_rev == MDSS_HW_MSM8996 || - hw_rev == MDSS_HW_MSM8937 || - hw_rev == MDSS_HW_MSM8953 || - hw_rev == MDSS_HW_MSM8998 || - hw_rev == MDSS_HW_SDM660 || - hw_rev == MDSS_HW_SDM630) { - data->ubwc_dec_version = UBWC_1_0; - data->ubwc_enc_version = UBWC_1_0; - } - - if (hw_rev == MDSS_HW_MSM8996 || - hw_rev == MDSS_HW_MSM8998) - data->highest_bank_bit = 2; - else - data->highest_bank_bit = 1; - - return data; -} - -const struct msm_mdss_data *msm_mdss_get_mdss_data(struct device *dev) -{ - struct msm_mdss *mdss; + const struct qcom_ubwc_cfg_data *data = msm_mdss->mdss_data; + u32 value = MDSS_UBWC_STATIC_UBWC_SWIZZLE(data->ubwc_swizzle) | + MDSS_UBWC_STATIC_HIGHEST_BANK_BIT(data->highest_bank_bit); - if (!dev) - return ERR_PTR(-EINVAL); + if (data->ubwc_bank_spread) + value |= MDSS_UBWC_STATIC_UBWC_BANK_SPREAD; - mdss = dev_get_drvdata(dev); + if (data->macrotile_mode) + value |= MDSS_UBWC_STATIC_MACROTILE_MODE; - /* - * We could not do it at the probe time, since hw revision register was - * not readable. Fill data structure now for the MDP5 platforms. - */ - if (!mdss->mdss_data && mdss->is_mdp5) - mdss->mdss_data = msm_mdss_generate_mdp5_mdss_data(mdss); + writel_relaxed(value, msm_mdss->mmio + REG_MDSS_UBWC_STATIC); - return mdss->mdss_data; + writel_relaxed(4, msm_mdss->mmio + REG_MDSS_UBWC_CTRL_2); + writel_relaxed(1, msm_mdss->mmio + REG_MDSS_UBWC_PREDICTION_MODE); } static int msm_mdss_enable(struct msm_mdss *msm_mdss) @@ -297,12 +256,8 @@ static int msm_mdss_enable(struct msm_mdss *msm_mdss) for (i = 0; i < msm_mdss->num_mdp_paths; i++) icc_set_bw(msm_mdss->mdp_path[i], 0, Bps_to_icc(MIN_IB_BW)); - if (msm_mdss->mdss_data && msm_mdss->mdss_data->reg_bus_bw) - icc_set_bw(msm_mdss->reg_bus_path, 0, - msm_mdss->mdss_data->reg_bus_bw); - else - icc_set_bw(msm_mdss->reg_bus_path, 0, - DEFAULT_REG_BW); + icc_set_bw(msm_mdss->reg_bus_path, 0, + msm_mdss->reg_bus_bw); ret = clk_bulk_prepare_enable(msm_mdss->num_clocks, msm_mdss->clocks); if (ret) { @@ -339,6 +294,9 @@ static int msm_mdss_enable(struct msm_mdss *msm_mdss) case UBWC_4_3: msm_mdss_setup_ubwc_dec_40(msm_mdss); break; + case UBWC_5_0: + msm_mdss_setup_ubwc_dec_50(msm_mdss); + break; default: dev_err(msm_mdss->dev, "Unsupported UBWC decoder version %x\n", msm_mdss->mdss_data->ubwc_dec_version); @@ -438,6 +396,7 @@ static int mdp5_mdss_parse_clock(struct platform_device *pdev, struct clk_bulk_d static struct msm_mdss *msm_mdss_init(struct platform_device *pdev, bool is_mdp5) { + const struct msm_mdss_data *mdss_data; struct msm_mdss *msm_mdss; int ret; int irq; @@ -450,7 +409,15 @@ static struct msm_mdss *msm_mdss_init(struct platform_device *pdev, bool is_mdp5 if (!msm_mdss) return ERR_PTR(-ENOMEM); - msm_mdss->mdss_data = of_device_get_match_data(&pdev->dev); + msm_mdss->mdss_data = qcom_ubwc_config_get_data(); + if (IS_ERR(msm_mdss->mdss_data)) + return ERR_CAST(msm_mdss->mdss_data); + + mdss_data = of_device_get_match_data(&pdev->dev); + if (!mdss_data) + return ERR_PTR(-EINVAL); + + msm_mdss->reg_bus_bw = mdss_data->reg_bus_bw; msm_mdss->mmio = devm_platform_ioremap_resource_byname(pdev, is_mdp5 ? "mdss_phys" : "mdss"); if (IS_ERR(msm_mdss->mmio)) @@ -569,205 +536,49 @@ static void mdss_remove(struct platform_device *pdev) msm_mdss_destroy(mdss); } -static const struct msm_mdss_data msm8998_data = { - .ubwc_enc_version = UBWC_1_0, - .ubwc_dec_version = UBWC_1_0, - .highest_bank_bit = 2, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data qcm2290_data = { - /* no UBWC */ - .highest_bank_bit = 0x2, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data sa8775p_data = { - .ubwc_enc_version = UBWC_4_0, - .ubwc_dec_version = UBWC_4_0, - .ubwc_swizzle = 4, - .ubwc_bank_spread = true, - .highest_bank_bit = 0, - .macrotile_mode = true, - .reg_bus_bw = 74000, -}; - -static const struct msm_mdss_data sar2130p_data = { - .ubwc_enc_version = UBWC_3_0, /* 4.0.2 in hw */ - .ubwc_dec_version = UBWC_4_3, - .ubwc_swizzle = 6, - .ubwc_bank_spread = true, - .highest_bank_bit = 0, - .macrotile_mode = 1, - .reg_bus_bw = 74000, -}; - -static const struct msm_mdss_data sc7180_data = { - .ubwc_enc_version = UBWC_2_0, - .ubwc_dec_version = UBWC_2_0, - .ubwc_swizzle = 6, - .ubwc_bank_spread = true, - .highest_bank_bit = 0x1, - .reg_bus_bw = 76800, +static const struct msm_mdss_data data_57k = { + .reg_bus_bw = 57000, }; -static const struct msm_mdss_data sc7280_data = { - .ubwc_enc_version = UBWC_3_0, - .ubwc_dec_version = UBWC_4_0, - .ubwc_swizzle = 6, - .ubwc_bank_spread = true, - .highest_bank_bit = 1, - .macrotile_mode = true, +static const struct msm_mdss_data data_74k = { .reg_bus_bw = 74000, }; -static const struct msm_mdss_data sc8180x_data = { - .ubwc_enc_version = UBWC_3_0, - .ubwc_dec_version = UBWC_3_0, - .highest_bank_bit = 3, - .macrotile_mode = true, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data sc8280xp_data = { - .ubwc_enc_version = UBWC_4_0, - .ubwc_dec_version = UBWC_4_0, - .ubwc_swizzle = 6, - .ubwc_bank_spread = true, - .highest_bank_bit = 3, - .macrotile_mode = true, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data sdm670_data = { - .ubwc_enc_version = UBWC_2_0, - .ubwc_dec_version = UBWC_2_0, - .highest_bank_bit = 1, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data sdm845_data = { - .ubwc_enc_version = UBWC_2_0, - .ubwc_dec_version = UBWC_2_0, - .highest_bank_bit = 2, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data sm6350_data = { - .ubwc_enc_version = UBWC_2_0, - .ubwc_dec_version = UBWC_2_0, - .ubwc_swizzle = 6, - .ubwc_bank_spread = true, - .highest_bank_bit = 1, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data sm7150_data = { - .ubwc_enc_version = UBWC_2_0, - .ubwc_dec_version = UBWC_2_0, - .highest_bank_bit = 1, +static const struct msm_mdss_data data_76k8 = { .reg_bus_bw = 76800, }; -static const struct msm_mdss_data sm8150_data = { - .ubwc_enc_version = UBWC_3_0, - .ubwc_dec_version = UBWC_3_0, - .highest_bank_bit = 2, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data sm6115_data = { - .ubwc_enc_version = UBWC_1_0, - .ubwc_dec_version = UBWC_2_0, - .ubwc_swizzle = 7, - .ubwc_bank_spread = true, - .highest_bank_bit = 0x1, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data sm6125_data = { - .ubwc_enc_version = UBWC_1_0, - .ubwc_dec_version = UBWC_3_0, - .ubwc_swizzle = 1, - .highest_bank_bit = 1, -}; - -static const struct msm_mdss_data sm6150_data = { - .ubwc_enc_version = UBWC_2_0, - .ubwc_dec_version = UBWC_2_0, - .highest_bank_bit = 1, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data sm8250_data = { - .ubwc_enc_version = UBWC_4_0, - .ubwc_dec_version = UBWC_4_0, - .ubwc_swizzle = 6, - .ubwc_bank_spread = true, - /* TODO: highest_bank_bit = 2 for LP_DDR4 */ - .highest_bank_bit = 3, - .macrotile_mode = true, - .reg_bus_bw = 76800, -}; - -static const struct msm_mdss_data sm8350_data = { - .ubwc_enc_version = UBWC_4_0, - .ubwc_dec_version = UBWC_4_0, - .ubwc_swizzle = 6, - .ubwc_bank_spread = true, - /* TODO: highest_bank_bit = 2 for LP_DDR4 */ - .highest_bank_bit = 3, - .macrotile_mode = true, - .reg_bus_bw = 74000, -}; - -static const struct msm_mdss_data sm8550_data = { - .ubwc_enc_version = UBWC_4_0, - .ubwc_dec_version = UBWC_4_3, - .ubwc_swizzle = 6, - .ubwc_bank_spread = true, - /* TODO: highest_bank_bit = 2 for LP_DDR4 */ - .highest_bank_bit = 3, - .macrotile_mode = true, - .reg_bus_bw = 57000, -}; - -static const struct msm_mdss_data x1e80100_data = { - .ubwc_enc_version = UBWC_4_0, - .ubwc_dec_version = UBWC_4_3, - .ubwc_swizzle = 6, - .ubwc_bank_spread = true, - /* TODO: highest_bank_bit = 2 for LP_DDR4 */ - .highest_bank_bit = 3, - .macrotile_mode = true, - /* TODO: Add reg_bus_bw with real value */ +static const struct msm_mdss_data data_153k6 = { + .reg_bus_bw = 153600, }; static const struct of_device_id mdss_dt_match[] = { - { .compatible = "qcom,mdss" }, - { .compatible = "qcom,msm8998-mdss", .data = &msm8998_data }, - { .compatible = "qcom,qcm2290-mdss", .data = &qcm2290_data }, - { .compatible = "qcom,sa8775p-mdss", .data = &sa8775p_data }, - { .compatible = "qcom,sar2130p-mdss", .data = &sar2130p_data }, - { .compatible = "qcom,sdm670-mdss", .data = &sdm670_data }, - { .compatible = "qcom,sdm845-mdss", .data = &sdm845_data }, - { .compatible = "qcom,sc7180-mdss", .data = &sc7180_data }, - { .compatible = "qcom,sc7280-mdss", .data = &sc7280_data }, - { .compatible = "qcom,sc8180x-mdss", .data = &sc8180x_data }, - { .compatible = "qcom,sc8280xp-mdss", .data = &sc8280xp_data }, - { .compatible = "qcom,sm6115-mdss", .data = &sm6115_data }, - { .compatible = "qcom,sm6125-mdss", .data = &sm6125_data }, - { .compatible = "qcom,sm6150-mdss", .data = &sm6150_data }, - { .compatible = "qcom,sm6350-mdss", .data = &sm6350_data }, - { .compatible = "qcom,sm6375-mdss", .data = &sm6350_data }, - { .compatible = "qcom,sm7150-mdss", .data = &sm7150_data }, - { .compatible = "qcom,sm8150-mdss", .data = &sm8150_data }, - { .compatible = "qcom,sm8250-mdss", .data = &sm8250_data }, - { .compatible = "qcom,sm8350-mdss", .data = &sm8350_data }, - { .compatible = "qcom,sm8450-mdss", .data = &sm8350_data }, - { .compatible = "qcom,sm8550-mdss", .data = &sm8550_data }, - { .compatible = "qcom,sm8650-mdss", .data = &sm8550_data}, - { .compatible = "qcom,x1e80100-mdss", .data = &x1e80100_data}, + { .compatible = "qcom,mdss", .data = &data_153k6 }, + { .compatible = "qcom,msm8998-mdss", .data = &data_76k8 }, + { .compatible = "qcom,qcm2290-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sa8775p-mdss", .data = &data_74k }, + { .compatible = "qcom,sar2130p-mdss", .data = &data_74k }, + { .compatible = "qcom,sdm670-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sdm845-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sc7180-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sc7280-mdss", .data = &data_74k }, + { .compatible = "qcom,sc8180x-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sc8280xp-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sm6115-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sm6125-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sm6150-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sm6350-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sm6375-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sm7150-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sm8150-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sm8250-mdss", .data = &data_76k8 }, + { .compatible = "qcom,sm8350-mdss", .data = &data_74k }, + { .compatible = "qcom,sm8450-mdss", .data = &data_74k }, + { .compatible = "qcom,sm8550-mdss", .data = &data_57k }, + { .compatible = "qcom,sm8650-mdss", .data = &data_57k }, + { .compatible = "qcom,sm8750-mdss", .data = &data_57k }, + /* TODO: x1e8: Add reg_bus_bw with real value */ + { .compatible = "qcom,x1e80100-mdss", .data = &data_153k6 }, {} }; MODULE_DEVICE_TABLE(of, mdss_dt_match); diff --git a/drivers/gpu/drm/msm/msm_mdss.h b/drivers/gpu/drm/msm/msm_mdss.h deleted file mode 100644 index 14dc53704314..000000000000 --- a/drivers/gpu/drm/msm/msm_mdss.h +++ /dev/null @@ -1,28 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Copyright (c) 2018, The Linux Foundation - */ - -#ifndef __MSM_MDSS_H__ -#define __MSM_MDSS_H__ - -struct msm_mdss_data { - u32 ubwc_enc_version; - /* can be read from register 0x58 */ - u32 ubwc_dec_version; - u32 ubwc_swizzle; - u32 highest_bank_bit; - bool ubwc_bank_spread; - bool macrotile_mode; - u32 reg_bus_bw; -}; - -#define UBWC_1_0 0x10000000 -#define UBWC_2_0 0x20000000 -#define UBWC_3_0 0x30000000 -#define UBWC_4_0 0x40000000 -#define UBWC_4_3 0x40030000 - -const struct msm_mdss_data *msm_mdss_get_mdss_data(struct device *dev); - -#endif /* __MSM_MDSS_H__ */ diff --git a/drivers/gpu/drm/msm/msm_mmu.h b/drivers/gpu/drm/msm/msm_mmu.h index daf91529e02b..8915662fbd4d 100644 --- a/drivers/gpu/drm/msm/msm_mmu.h +++ b/drivers/gpu/drm/msm/msm_mmu.h @@ -9,13 +9,21 @@ #include <linux/iommu.h> +struct msm_mmu_prealloc; +struct msm_mmu; +struct msm_gpu; + struct msm_mmu_funcs { void (*detach)(struct msm_mmu *mmu); + void (*prealloc_count)(struct msm_mmu *mmu, struct msm_mmu_prealloc *p, + uint64_t iova, size_t len); + int (*prealloc_allocate)(struct msm_mmu *mmu, struct msm_mmu_prealloc *p); + void (*prealloc_cleanup)(struct msm_mmu *mmu, struct msm_mmu_prealloc *p); int (*map)(struct msm_mmu *mmu, uint64_t iova, struct sg_table *sgt, - size_t len, int prot); + size_t off, size_t len, int prot); int (*unmap)(struct msm_mmu *mmu, uint64_t iova, size_t len); void (*destroy)(struct msm_mmu *mmu); - void (*resume_translation)(struct msm_mmu *mmu); + void (*set_stall)(struct msm_mmu *mmu, bool enable); }; enum msm_mmu_type { @@ -24,12 +32,38 @@ enum msm_mmu_type { MSM_MMU_IOMMU_PAGETABLE, }; +/** + * struct msm_mmu_prealloc - Tracking for pre-allocated pages for MMU updates. + */ +struct msm_mmu_prealloc { + /** @count: Number of pages reserved. */ + uint32_t count; + /** @ptr: Index of first unused page in @pages */ + uint32_t ptr; + /** + * @pages: Array of pages preallocated for MMU table updates. + * + * After a VM operation, there might be free pages remaining in this + * array (since the amount allocated is a worst-case). These are + * returned to the pt_cache at mmu->prealloc_cleanup(). + */ + void **pages; +}; + struct msm_mmu { const struct msm_mmu_funcs *funcs; struct device *dev; int (*handler)(void *arg, unsigned long iova, int flags, void *data); void *arg; enum msm_mmu_type type; + + /** + * @prealloc: pre-allocated pages for pgtable + * + * Set while a VM_BIND job is running, serialized under + * msm_gem_vm::mmu_lock. + */ + struct msm_mmu_prealloc *prealloc; }; static inline void msm_mmu_init(struct msm_mmu *mmu, struct device *dev, @@ -51,7 +85,7 @@ static inline void msm_mmu_set_fault_handler(struct msm_mmu *mmu, void *arg, mmu->handler = handler; } -struct msm_mmu *msm_iommu_pagetable_create(struct msm_mmu *parent); +struct msm_mmu *msm_iommu_pagetable_create(struct msm_mmu *parent, bool kernel_managed); int msm_iommu_pagetable_params(struct msm_mmu *mmu, phys_addr_t *ttbr, int *asid); diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index 39138e190cb9..54493a94dcb7 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -308,21 +308,11 @@ void msm_rd_debugfs_cleanup(struct msm_drm_private *priv) priv->hangrd = NULL; } -static void snapshot_buf(struct msm_rd_state *rd, - struct msm_gem_submit *submit, int idx, - uint64_t iova, uint32_t size, bool full) +static void snapshot_buf(struct msm_rd_state *rd, struct drm_gem_object *obj, + uint64_t iova, bool full, size_t offset, size_t size) { - struct drm_gem_object *obj = submit->bos[idx].obj; - unsigned offset = 0; const char *buf; - if (iova) { - offset = iova - submit->bos[idx].iova; - } else { - iova = submit->bos[idx].iova; - size = obj->size; - } - /* * Always write the GPUADDR header so can get a complete list of all the * buffers in the cmd @@ -333,10 +323,6 @@ static void snapshot_buf(struct msm_rd_state *rd, if (!full) return; - /* But only dump the contents of buffers marked READ */ - if (!(submit->bos[idx].flags & MSM_SUBMIT_BO_READ)) - return; - buf = msm_gem_get_vaddr_active(obj); if (IS_ERR(buf)) return; @@ -352,6 +338,7 @@ static void snapshot_buf(struct msm_rd_state *rd, void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit, const char *fmt, ...) { + extern bool rd_full; struct task_struct *task; char msg[256]; int i, n; @@ -385,16 +372,43 @@ void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit, rd_write_section(rd, RD_CMD, msg, ALIGN(n, 4)); - for (i = 0; i < submit->nr_bos; i++) - snapshot_buf(rd, submit, i, 0, 0, should_dump(submit, i)); + if (msm_context_is_vmbind(submit->queue->ctx)) { + struct drm_gpuva *vma; - for (i = 0; i < submit->nr_cmds; i++) { - uint32_t szd = submit->cmd[i].size; /* in dwords */ + drm_gpuvm_resv_assert_held(submit->vm); + + drm_gpuvm_for_each_va (vma, submit->vm) { + bool dump = rd_full || (vma->flags & MSM_VMA_DUMP); + + /* Skip MAP_NULL/PRR VMAs: */ + if (!vma->gem.obj) + continue; + + snapshot_buf(rd, vma->gem.obj, vma->va.addr, dump, + vma->gem.offset, vma->va.range); + } + + } else { + for (i = 0; i < submit->nr_bos; i++) { + struct drm_gem_object *obj = submit->bos[i].obj; + bool dump = rd_full || (submit->bos[i].flags & MSM_SUBMIT_BO_DUMP); + + snapshot_buf(rd, obj, submit->bos[i].iova, dump, 0, obj->size); + } + + for (i = 0; i < submit->nr_cmds; i++) { + uint32_t szd = submit->cmd[i].size; /* in dwords */ + int idx = submit->cmd[i].idx; + bool dump = rd_full || (submit->bos[idx].flags & MSM_SUBMIT_BO_DUMP); + + /* snapshot cmdstream bo's (if we haven't already): */ + if (!dump) { + struct drm_gem_object *obj = submit->bos[idx].obj; + size_t offset = submit->cmd[i].iova - submit->bos[idx].iova; - /* snapshot cmdstream bo's (if we haven't already): */ - if (!should_dump(submit, i)) { - snapshot_buf(rd, submit, submit->cmd[i].idx, - submit->cmd[i].iova, szd * 4, true); + snapshot_buf(rd, obj, submit->cmd[i].iova, true, + offset, szd * 4); + } } } diff --git a/drivers/gpu/drm/msm/msm_ringbuffer.c b/drivers/gpu/drm/msm/msm_ringbuffer.c index 89dce15eed3b..b2f612e5dc79 100644 --- a/drivers/gpu/drm/msm/msm_ringbuffer.c +++ b/drivers/gpu/drm/msm/msm_ringbuffer.c @@ -17,6 +17,7 @@ static struct dma_fence *msm_job_run(struct drm_sched_job *job) struct msm_fence_context *fctx = submit->ring->fctx; struct msm_gpu *gpu = submit->gpu; struct msm_drm_private *priv = gpu->dev->dev_private; + unsigned nr_cmds = submit->nr_cmds; int i; msm_fence_init(submit->hw_fence, fctx); @@ -36,8 +37,13 @@ static struct dma_fence *msm_job_run(struct drm_sched_job *job) /* TODO move submit path over to using a per-ring lock.. */ mutex_lock(&gpu->lock); + if (submit->queue->ctx->closed) + submit->nr_cmds = 0; + msm_gpu_submit(gpu, submit); + submit->nr_cmds = nr_cmds; + mutex_unlock(&gpu->lock); return dma_fence_get(submit->hw_fence); @@ -84,7 +90,7 @@ struct msm_ringbuffer *msm_ringbuffer_new(struct msm_gpu *gpu, int id, ring->start = msm_gem_kernel_new(gpu->dev, MSM_GPU_RINGBUFFER_SZ, check_apriv(gpu, MSM_BO_WC | MSM_BO_GPU_READONLY), - gpu->aspace, &ring->bo, &ring->iova); + gpu->vm, &ring->bo, &ring->iova); if (IS_ERR(ring->start)) { ret = PTR_ERR(ring->start); @@ -131,7 +137,7 @@ void msm_ringbuffer_destroy(struct msm_ringbuffer *ring) msm_fence_context_free(ring->fctx); - msm_gem_kernel_put(ring->bo, ring->gpu->aspace); + msm_gem_kernel_put(ring->bo, ring->gpu->vm); kfree(ring); } diff --git a/drivers/gpu/drm/msm/msm_submitqueue.c b/drivers/gpu/drm/msm/msm_submitqueue.c index 7fed1de63b5d..8617a82cd6b3 100644 --- a/drivers/gpu/drm/msm/msm_submitqueue.c +++ b/drivers/gpu/drm/msm/msm_submitqueue.c @@ -7,8 +7,7 @@ #include "msm_gpu.h" -int msm_file_private_set_sysprof(struct msm_file_private *ctx, - struct msm_gpu *gpu, int sysprof) +int msm_context_set_sysprof(struct msm_context *ctx, struct msm_gpu *gpu, int sysprof) { /* * Since pm_runtime and sysprof_active are both refcounts, we @@ -46,10 +45,10 @@ int msm_file_private_set_sysprof(struct msm_file_private *ctx, return 0; } -void __msm_file_private_destroy(struct kref *kref) +void __msm_context_destroy(struct kref *kref) { - struct msm_file_private *ctx = container_of(kref, - struct msm_file_private, ref); + struct msm_context *ctx = container_of(kref, + struct msm_context, ref); int i; for (i = 0; i < ARRAY_SIZE(ctx->entities); i++) { @@ -60,7 +59,7 @@ void __msm_file_private_destroy(struct kref *kref) kfree(ctx->entities[i]); } - msm_gem_address_space_put(ctx->aspace); + drm_gpuvm_put(ctx->vm); kfree(ctx->comm); kfree(ctx->cmdline); kfree(ctx); @@ -73,12 +72,15 @@ void msm_submitqueue_destroy(struct kref *kref) idr_destroy(&queue->fence_idr); - msm_file_private_put(queue->ctx); + if (queue->entity == &queue->_vm_bind_entity[0]) + drm_sched_entity_destroy(queue->entity); + + msm_context_put(queue->ctx); kfree(queue); } -struct msm_gpu_submitqueue *msm_submitqueue_get(struct msm_file_private *ctx, +struct msm_gpu_submitqueue *msm_submitqueue_get(struct msm_context *ctx, u32 id) { struct msm_gpu_submitqueue *entry; @@ -101,9 +103,9 @@ struct msm_gpu_submitqueue *msm_submitqueue_get(struct msm_file_private *ctx, return NULL; } -void msm_submitqueue_close(struct msm_file_private *ctx) +void msm_submitqueue_close(struct msm_context *ctx) { - struct msm_gpu_submitqueue *entry, *tmp; + struct msm_gpu_submitqueue *queue, *tmp; if (!ctx) return; @@ -112,14 +114,21 @@ void msm_submitqueue_close(struct msm_file_private *ctx) * No lock needed in close and there won't * be any more user ioctls coming our way */ - list_for_each_entry_safe(entry, tmp, &ctx->submitqueues, node) { - list_del(&entry->node); - msm_submitqueue_put(entry); + list_for_each_entry_safe(queue, tmp, &ctx->submitqueues, node) { + if (queue->entity == &queue->_vm_bind_entity[0]) + drm_sched_entity_flush(queue->entity, MAX_WAIT_SCHED_ENTITY_Q_EMPTY); + list_del(&queue->node); + msm_submitqueue_put(queue); } + + if (!ctx->vm) + return; + + msm_gem_vm_close(ctx->vm); } static struct drm_sched_entity * -get_sched_entity(struct msm_file_private *ctx, struct msm_ringbuffer *ring, +get_sched_entity(struct msm_context *ctx, struct msm_ringbuffer *ring, unsigned ring_nr, enum drm_sched_priority sched_prio) { static DEFINE_MUTEX(entity_lock); @@ -155,14 +164,12 @@ get_sched_entity(struct msm_file_private *ctx, struct msm_ringbuffer *ring, return ctx->entities[idx]; } -int msm_submitqueue_create(struct drm_device *drm, struct msm_file_private *ctx, +int msm_submitqueue_create(struct drm_device *drm, struct msm_context *ctx, u32 prio, u32 flags, u32 *id) { struct msm_drm_private *priv = drm->dev_private; struct msm_gpu_submitqueue *queue; enum drm_sched_priority sched_prio; - extern int enable_preemption; - bool preemption_supported; unsigned ring_nr; int ret; @@ -172,26 +179,53 @@ int msm_submitqueue_create(struct drm_device *drm, struct msm_file_private *ctx, if (!priv->gpu) return -ENODEV; - preemption_supported = priv->gpu->nr_rings == 1 && enable_preemption != 0; + if (flags & MSM_SUBMITQUEUE_VM_BIND) { + unsigned sz; - if (flags & MSM_SUBMITQUEUE_ALLOW_PREEMPT && preemption_supported) - return -EINVAL; + /* Not allowed for kernel managed VMs (ie. kernel allocs VA) */ + if (!msm_context_is_vmbind(ctx)) + return -EINVAL; - ret = msm_gpu_convert_priority(priv->gpu, prio, &ring_nr, &sched_prio); - if (ret) - return ret; + if (prio) + return -EINVAL; + + sz = struct_size(queue, _vm_bind_entity, 1); + queue = kzalloc(sz, GFP_KERNEL); + } else { + extern int enable_preemption; + bool preemption_supported = + priv->gpu->nr_rings == 1 && enable_preemption != 0; + + if (flags & MSM_SUBMITQUEUE_ALLOW_PREEMPT && preemption_supported) + return -EINVAL; - queue = kzalloc(sizeof(*queue), GFP_KERNEL); + ret = msm_gpu_convert_priority(priv->gpu, prio, &ring_nr, &sched_prio); + if (ret) + return ret; + + queue = kzalloc(sizeof(*queue), GFP_KERNEL); + } if (!queue) return -ENOMEM; kref_init(&queue->ref); queue->flags = flags; - queue->ring_nr = ring_nr; - queue->entity = get_sched_entity(ctx, priv->gpu->rb[ring_nr], - ring_nr, sched_prio); + if (flags & MSM_SUBMITQUEUE_VM_BIND) { + struct drm_gpu_scheduler *sched = &to_msm_vm(msm_context_vm(drm, ctx))->sched; + + queue->entity = &queue->_vm_bind_entity[0]; + + drm_sched_entity_init(queue->entity, DRM_SCHED_PRIORITY_KERNEL, + &sched, 1, NULL); + } else { + queue->ring_nr = ring_nr; + + queue->entity = get_sched_entity(ctx, priv->gpu->rb[ring_nr], + ring_nr, sched_prio); + } + if (IS_ERR(queue->entity)) { ret = PTR_ERR(queue->entity); kfree(queue); @@ -200,7 +234,7 @@ int msm_submitqueue_create(struct drm_device *drm, struct msm_file_private *ctx, write_lock(&ctx->queuelock); - queue->ctx = msm_file_private_get(ctx); + queue->ctx = msm_context_get(ctx); queue->id = ctx->queueid++; if (id) @@ -221,7 +255,7 @@ int msm_submitqueue_create(struct drm_device *drm, struct msm_file_private *ctx, * Create the default submit-queue (id==0), used for backwards compatibility * for userspace that pre-dates the introduction of submitqueues. */ -int msm_submitqueue_init(struct drm_device *drm, struct msm_file_private *ctx) +int msm_submitqueue_init(struct drm_device *drm, struct msm_context *ctx) { struct msm_drm_private *priv = drm->dev_private; int default_prio, max_priority; @@ -261,7 +295,7 @@ static int msm_submitqueue_query_faults(struct msm_gpu_submitqueue *queue, return ret ? -EFAULT : 0; } -int msm_submitqueue_query(struct drm_device *drm, struct msm_file_private *ctx, +int msm_submitqueue_query(struct drm_device *drm, struct msm_context *ctx, struct drm_msm_submitqueue_query *args) { struct msm_gpu_submitqueue *queue; @@ -282,7 +316,7 @@ int msm_submitqueue_query(struct drm_device *drm, struct msm_file_private *ctx, return ret; } -int msm_submitqueue_remove(struct msm_file_private *ctx, u32 id) +int msm_submitqueue_remove(struct msm_context *ctx, u32 id) { struct msm_gpu_submitqueue *entry; diff --git a/drivers/gpu/drm/msm/msm_syncobj.c b/drivers/gpu/drm/msm/msm_syncobj.c new file mode 100644 index 000000000000..4baa9f522c54 --- /dev/null +++ b/drivers/gpu/drm/msm/msm_syncobj.c @@ -0,0 +1,172 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (C) 2020 Google, Inc */ + +#include "drm/drm_drv.h" + +#include "msm_drv.h" +#include "msm_syncobj.h" + +struct drm_syncobj ** +msm_syncobj_parse_deps(struct drm_device *dev, + struct drm_sched_job *job, + struct drm_file *file, + uint64_t in_syncobjs_addr, + uint32_t nr_in_syncobjs, + size_t syncobj_stride) +{ + struct drm_syncobj **syncobjs = NULL; + struct drm_msm_syncobj syncobj_desc = {0}; + int ret = 0; + uint32_t i, j; + + syncobjs = kcalloc(nr_in_syncobjs, sizeof(*syncobjs), + GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY); + if (!syncobjs) + return ERR_PTR(-ENOMEM); + + for (i = 0; i < nr_in_syncobjs; ++i) { + uint64_t address = in_syncobjs_addr + i * syncobj_stride; + + if (copy_from_user(&syncobj_desc, + u64_to_user_ptr(address), + min(syncobj_stride, sizeof(syncobj_desc)))) { + ret = -EFAULT; + break; + } + + if (syncobj_desc.point && + !drm_core_check_feature(dev, DRIVER_SYNCOBJ_TIMELINE)) { + ret = UERR(EOPNOTSUPP, dev, "syncobj timeline unsupported"); + break; + } + + if (syncobj_desc.flags & ~MSM_SYNCOBJ_FLAGS) { + ret = UERR(EINVAL, dev, "invalid syncobj flags: %x", syncobj_desc.flags); + break; + } + + ret = drm_sched_job_add_syncobj_dependency(job, file, + syncobj_desc.handle, + syncobj_desc.point); + if (ret) + break; + + if (syncobj_desc.flags & MSM_SYNCOBJ_RESET) { + syncobjs[i] = drm_syncobj_find(file, syncobj_desc.handle); + if (!syncobjs[i]) { + ret = UERR(EINVAL, dev, "invalid syncobj handle: %u", i); + break; + } + } + } + + if (ret) { + for (j = 0; j <= i; ++j) { + if (syncobjs[j]) + drm_syncobj_put(syncobjs[j]); + } + kfree(syncobjs); + return ERR_PTR(ret); + } + return syncobjs; +} + +void +msm_syncobj_reset(struct drm_syncobj **syncobjs, uint32_t nr_syncobjs) +{ + uint32_t i; + + for (i = 0; syncobjs && i < nr_syncobjs; ++i) { + if (syncobjs[i]) + drm_syncobj_replace_fence(syncobjs[i], NULL); + } +} + +struct msm_syncobj_post_dep * +msm_syncobj_parse_post_deps(struct drm_device *dev, + struct drm_file *file, + uint64_t syncobjs_addr, + uint32_t nr_syncobjs, + size_t syncobj_stride) +{ + struct msm_syncobj_post_dep *post_deps; + struct drm_msm_syncobj syncobj_desc = {0}; + int ret = 0; + uint32_t i, j; + + post_deps = kcalloc(nr_syncobjs, sizeof(*post_deps), + GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY); + if (!post_deps) + return ERR_PTR(-ENOMEM); + + for (i = 0; i < nr_syncobjs; ++i) { + uint64_t address = syncobjs_addr + i * syncobj_stride; + + if (copy_from_user(&syncobj_desc, + u64_to_user_ptr(address), + min(syncobj_stride, sizeof(syncobj_desc)))) { + ret = -EFAULT; + break; + } + + post_deps[i].point = syncobj_desc.point; + + if (syncobj_desc.flags) { + ret = UERR(EINVAL, dev, "invalid syncobj flags"); + break; + } + + if (syncobj_desc.point) { + if (!drm_core_check_feature(dev, + DRIVER_SYNCOBJ_TIMELINE)) { + ret = UERR(EOPNOTSUPP, dev, "syncobj timeline unsupported"); + break; + } + + post_deps[i].chain = dma_fence_chain_alloc(); + if (!post_deps[i].chain) { + ret = -ENOMEM; + break; + } + } + + post_deps[i].syncobj = + drm_syncobj_find(file, syncobj_desc.handle); + if (!post_deps[i].syncobj) { + ret = UERR(EINVAL, dev, "invalid syncobj handle"); + break; + } + } + + if (ret) { + for (j = 0; j <= i; ++j) { + dma_fence_chain_free(post_deps[j].chain); + if (post_deps[j].syncobj) + drm_syncobj_put(post_deps[j].syncobj); + } + + kfree(post_deps); + return ERR_PTR(ret); + } + + return post_deps; +} + +void +msm_syncobj_process_post_deps(struct msm_syncobj_post_dep *post_deps, + uint32_t count, struct dma_fence *fence) +{ + uint32_t i; + + for (i = 0; post_deps && i < count; ++i) { + if (post_deps[i].chain) { + drm_syncobj_add_point(post_deps[i].syncobj, + post_deps[i].chain, + fence, post_deps[i].point); + post_deps[i].chain = NULL; + } else { + drm_syncobj_replace_fence(post_deps[i].syncobj, + fence); + } + } +} diff --git a/drivers/gpu/drm/msm/msm_syncobj.h b/drivers/gpu/drm/msm/msm_syncobj.h new file mode 100644 index 000000000000..bcaa15d01da0 --- /dev/null +++ b/drivers/gpu/drm/msm/msm_syncobj.h @@ -0,0 +1,37 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (C) 2020 Google, Inc */ + +#ifndef __MSM_GEM_SYNCOBJ_H__ +#define __MSM_GEM_SYNCOBJ_H__ + +#include "drm/drm_device.h" +#include "drm/drm_syncobj.h" +#include "drm/gpu_scheduler.h" + +struct msm_syncobj_post_dep { + struct drm_syncobj *syncobj; + uint64_t point; + struct dma_fence_chain *chain; +}; + +struct drm_syncobj ** +msm_syncobj_parse_deps(struct drm_device *dev, + struct drm_sched_job *job, + struct drm_file *file, + uint64_t in_syncobjs_addr, + uint32_t nr_in_syncobjs, + size_t syncobj_stride); + +void msm_syncobj_reset(struct drm_syncobj **syncobjs, uint32_t nr_syncobjs); + +struct msm_syncobj_post_dep * +msm_syncobj_parse_post_deps(struct drm_device *dev, + struct drm_file *file, + uint64_t syncobjs_addr, + uint32_t nr_syncobjs, + size_t syncobj_stride); + +void msm_syncobj_process_post_deps(struct msm_syncobj_post_dep *post_deps, + uint32_t count, struct dma_fence *fence); + +#endif /* __MSM_GEM_SYNCOBJ_H__ */ diff --git a/drivers/gpu/drm/msm/registers/adreno/a6xx.xml b/drivers/gpu/drm/msm/registers/adreno/a6xx.xml index 2db425abf0f3..d860fd94feae 100644 --- a/drivers/gpu/drm/msm/registers/adreno/a6xx.xml +++ b/drivers/gpu/drm/msm/registers/adreno/a6xx.xml @@ -5,6 +5,11 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <import file="freedreno_copyright.xml"/> <import file="adreno/adreno_common.xml"/> <import file="adreno/adreno_pm4.xml"/> +<import file="adreno/a6xx_enums.xml"/> +<import file="adreno/a7xx_enums.xml"/> +<import file="adreno/a6xx_perfcntrs.xml"/> +<import file="adreno/a7xx_perfcntrs.xml"/> +<import file="adreno/a6xx_descriptors.xml"/> <!-- Each register that is actually being used by driver should have "usage" defined, @@ -20,2205 +25,6 @@ is either overwritten by renderpass/blit (ib2) or not used if not overwritten by a particular renderpass/blit. --> -<!-- these might be same as a5xx --> -<enum name="a6xx_tile_mode"> - <value name="TILE6_LINEAR" value="0"/> - <value name="TILE6_2" value="2"/> - <value name="TILE6_3" value="3"/> -</enum> - -<enum name="a6xx_format"> - <value value="0x02" name="FMT6_A8_UNORM"/> - <value value="0x03" name="FMT6_8_UNORM"/> - <value value="0x04" name="FMT6_8_SNORM"/> - <value value="0x05" name="FMT6_8_UINT"/> - <value value="0x06" name="FMT6_8_SINT"/> - - <value value="0x08" name="FMT6_4_4_4_4_UNORM"/> - <value value="0x0a" name="FMT6_5_5_5_1_UNORM"/> - <value value="0x0c" name="FMT6_1_5_5_5_UNORM"/> <!-- read only --> - <value value="0x0e" name="FMT6_5_6_5_UNORM"/> - - <value value="0x0f" name="FMT6_8_8_UNORM"/> - <value value="0x10" name="FMT6_8_8_SNORM"/> - <value value="0x11" name="FMT6_8_8_UINT"/> - <value value="0x12" name="FMT6_8_8_SINT"/> - <value value="0x13" name="FMT6_L8_A8_UNORM"/> - - <value value="0x15" name="FMT6_16_UNORM"/> - <value value="0x16" name="FMT6_16_SNORM"/> - <value value="0x17" name="FMT6_16_FLOAT"/> - <value value="0x18" name="FMT6_16_UINT"/> - <value value="0x19" name="FMT6_16_SINT"/> - - <value value="0x21" name="FMT6_8_8_8_UNORM"/> - <value value="0x22" name="FMT6_8_8_8_SNORM"/> - <value value="0x23" name="FMT6_8_8_8_UINT"/> - <value value="0x24" name="FMT6_8_8_8_SINT"/> - - <value value="0x30" name="FMT6_8_8_8_8_UNORM"/> - <value value="0x31" name="FMT6_8_8_8_X8_UNORM"/> <!-- samples 1 for alpha --> - <value value="0x32" name="FMT6_8_8_8_8_SNORM"/> - <value value="0x33" name="FMT6_8_8_8_8_UINT"/> - <value value="0x34" name="FMT6_8_8_8_8_SINT"/> - - <value value="0x35" name="FMT6_9_9_9_E5_FLOAT"/> - - <value value="0x36" name="FMT6_10_10_10_2_UNORM"/> - <value value="0x37" name="FMT6_10_10_10_2_UNORM_DEST"/> - <value value="0x39" name="FMT6_10_10_10_2_SNORM"/> - <value value="0x3a" name="FMT6_10_10_10_2_UINT"/> - <value value="0x3b" name="FMT6_10_10_10_2_SINT"/> - - <value value="0x42" name="FMT6_11_11_10_FLOAT"/> - - <value value="0x43" name="FMT6_16_16_UNORM"/> - <value value="0x44" name="FMT6_16_16_SNORM"/> - <value value="0x45" name="FMT6_16_16_FLOAT"/> - <value value="0x46" name="FMT6_16_16_UINT"/> - <value value="0x47" name="FMT6_16_16_SINT"/> - - <value value="0x48" name="FMT6_32_UNORM"/> - <value value="0x49" name="FMT6_32_SNORM"/> - <value value="0x4a" name="FMT6_32_FLOAT"/> - <value value="0x4b" name="FMT6_32_UINT"/> - <value value="0x4c" name="FMT6_32_SINT"/> - <value value="0x4d" name="FMT6_32_FIXED"/> - - <value value="0x58" name="FMT6_16_16_16_UNORM"/> - <value value="0x59" name="FMT6_16_16_16_SNORM"/> - <value value="0x5a" name="FMT6_16_16_16_FLOAT"/> - <value value="0x5b" name="FMT6_16_16_16_UINT"/> - <value value="0x5c" name="FMT6_16_16_16_SINT"/> - - <value value="0x60" name="FMT6_16_16_16_16_UNORM"/> - <value value="0x61" name="FMT6_16_16_16_16_SNORM"/> - <value value="0x62" name="FMT6_16_16_16_16_FLOAT"/> - <value value="0x63" name="FMT6_16_16_16_16_UINT"/> - <value value="0x64" name="FMT6_16_16_16_16_SINT"/> - - <value value="0x65" name="FMT6_32_32_UNORM"/> - <value value="0x66" name="FMT6_32_32_SNORM"/> - <value value="0x67" name="FMT6_32_32_FLOAT"/> - <value value="0x68" name="FMT6_32_32_UINT"/> - <value value="0x69" name="FMT6_32_32_SINT"/> - <value value="0x6a" name="FMT6_32_32_FIXED"/> - - <value value="0x70" name="FMT6_32_32_32_UNORM"/> - <value value="0x71" name="FMT6_32_32_32_SNORM"/> - <value value="0x72" name="FMT6_32_32_32_UINT"/> - <value value="0x73" name="FMT6_32_32_32_SINT"/> - <value value="0x74" name="FMT6_32_32_32_FLOAT"/> - <value value="0x75" name="FMT6_32_32_32_FIXED"/> - - <value value="0x80" name="FMT6_32_32_32_32_UNORM"/> - <value value="0x81" name="FMT6_32_32_32_32_SNORM"/> - <value value="0x82" name="FMT6_32_32_32_32_FLOAT"/> - <value value="0x83" name="FMT6_32_32_32_32_UINT"/> - <value value="0x84" name="FMT6_32_32_32_32_SINT"/> - <value value="0x85" name="FMT6_32_32_32_32_FIXED"/> - - <value value="0x8c" name="FMT6_G8R8B8R8_422_UNORM"/> <!-- UYVY --> - <value value="0x8d" name="FMT6_R8G8R8B8_422_UNORM"/> <!-- YUYV --> - <value value="0x8e" name="FMT6_R8_G8B8_2PLANE_420_UNORM"/> <!-- NV12 --> - <value value="0x8f" name="FMT6_NV21"/> - <value value="0x90" name="FMT6_R8_G8_B8_3PLANE_420_UNORM"/> <!-- YV12 --> - - <value value="0x91" name="FMT6_Z24_UNORM_S8_UINT_AS_R8G8B8A8"/> - - <!-- Note: tiling/UBWC for these may be different from equivalent formats - For example FMT6_NV12_Y is not compatible with FMT6_8_UNORM - --> - <value value="0x94" name="FMT6_NV12_Y"/> - <value value="0x95" name="FMT6_NV12_UV"/> - <value value="0x96" name="FMT6_NV12_VU"/> - <value value="0x97" name="FMT6_NV12_4R"/> - <value value="0x98" name="FMT6_NV12_4R_Y"/> - <value value="0x99" name="FMT6_NV12_4R_UV"/> - <value value="0x9a" name="FMT6_P010"/> - <value value="0x9b" name="FMT6_P010_Y"/> - <value value="0x9c" name="FMT6_P010_UV"/> - <value value="0x9d" name="FMT6_TP10"/> - <value value="0x9e" name="FMT6_TP10_Y"/> - <value value="0x9f" name="FMT6_TP10_UV"/> - - <value value="0xa0" name="FMT6_Z24_UNORM_S8_UINT"/> - - <value value="0xab" name="FMT6_ETC2_RG11_UNORM"/> - <value value="0xac" name="FMT6_ETC2_RG11_SNORM"/> - <value value="0xad" name="FMT6_ETC2_R11_UNORM"/> - <value value="0xae" name="FMT6_ETC2_R11_SNORM"/> - <value value="0xaf" name="FMT6_ETC1"/> - <value value="0xb0" name="FMT6_ETC2_RGB8"/> - <value value="0xb1" name="FMT6_ETC2_RGBA8"/> - <value value="0xb2" name="FMT6_ETC2_RGB8A1"/> - <value value="0xb3" name="FMT6_DXT1"/> - <value value="0xb4" name="FMT6_DXT3"/> - <value value="0xb5" name="FMT6_DXT5"/> - <value value="0xb7" name="FMT6_RGTC1_UNORM"/> - <value value="0xb8" name="FMT6_RGTC1_SNORM"/> - <value value="0xbb" name="FMT6_RGTC2_UNORM"/> - <value value="0xbc" name="FMT6_RGTC2_SNORM"/> - <value value="0xbe" name="FMT6_BPTC_UFLOAT"/> - <value value="0xbf" name="FMT6_BPTC_FLOAT"/> - <value value="0xc0" name="FMT6_BPTC"/> - <value value="0xc1" name="FMT6_ASTC_4x4"/> - <value value="0xc2" name="FMT6_ASTC_5x4"/> - <value value="0xc3" name="FMT6_ASTC_5x5"/> - <value value="0xc4" name="FMT6_ASTC_6x5"/> - <value value="0xc5" name="FMT6_ASTC_6x6"/> - <value value="0xc6" name="FMT6_ASTC_8x5"/> - <value value="0xc7" name="FMT6_ASTC_8x6"/> - <value value="0xc8" name="FMT6_ASTC_8x8"/> - <value value="0xc9" name="FMT6_ASTC_10x5"/> - <value value="0xca" name="FMT6_ASTC_10x6"/> - <value value="0xcb" name="FMT6_ASTC_10x8"/> - <value value="0xcc" name="FMT6_ASTC_10x10"/> - <value value="0xcd" name="FMT6_ASTC_12x10"/> - <value value="0xce" name="FMT6_ASTC_12x12"/> - - <!-- for sampling stencil (integer, 2nd channel), not available on a630 --> - <value value="0xea" name="FMT6_Z24_UINT_S8_UINT"/> - - <!-- Not a hw enum, used internally in driver --> - <value value="0xff" name="FMT6_NONE"/> - -</enum> - -<!-- probably same as a5xx --> -<enum name="a6xx_polygon_mode"> - <value name="POLYMODE6_POINTS" value="1"/> - <value name="POLYMODE6_LINES" value="2"/> - <value name="POLYMODE6_TRIANGLES" value="3"/> -</enum> - -<enum name="a6xx_depth_format"> - <value name="DEPTH6_NONE" value="0"/> - <value name="DEPTH6_16" value="1"/> - <value name="DEPTH6_24_8" value="2"/> - <value name="DEPTH6_32" value="4"/> -</enum> - -<bitset name="a6x_cp_protect" inline="yes"> - <bitfield name="BASE_ADDR" low="0" high="17"/> - <bitfield name="MASK_LEN" low="18" high="30"/> - <bitfield name="READ" pos="31" type="boolean"/> -</bitset> - -<enum name="a6xx_shader_id"> - <value value="0x9" name="A6XX_TP0_TMO_DATA"/> - <value value="0xa" name="A6XX_TP0_SMO_DATA"/> - <value value="0xb" name="A6XX_TP0_MIPMAP_BASE_DATA"/> - <value value="0x19" name="A6XX_TP1_TMO_DATA"/> - <value value="0x1a" name="A6XX_TP1_SMO_DATA"/> - <value value="0x1b" name="A6XX_TP1_MIPMAP_BASE_DATA"/> - <value value="0x29" name="A6XX_SP_INST_DATA"/> - <value value="0x2a" name="A6XX_SP_LB_0_DATA"/> - <value value="0x2b" name="A6XX_SP_LB_1_DATA"/> - <value value="0x2c" name="A6XX_SP_LB_2_DATA"/> - <value value="0x2d" name="A6XX_SP_LB_3_DATA"/> - <value value="0x2e" name="A6XX_SP_LB_4_DATA"/> - <value value="0x2f" name="A6XX_SP_LB_5_DATA"/> - <value value="0x30" name="A6XX_SP_CB_BINDLESS_DATA"/> - <value value="0x31" name="A6XX_SP_CB_LEGACY_DATA"/> - <value value="0x32" name="A6XX_SP_UAV_DATA"/> - <value value="0x33" name="A6XX_SP_INST_TAG"/> - <value value="0x34" name="A6XX_SP_CB_BINDLESS_TAG"/> - <value value="0x35" name="A6XX_SP_TMO_UMO_TAG"/> - <value value="0x36" name="A6XX_SP_SMO_TAG"/> - <value value="0x37" name="A6XX_SP_STATE_DATA"/> - <value value="0x49" name="A6XX_HLSQ_CHUNK_CVS_RAM"/> - <value value="0x4a" name="A6XX_HLSQ_CHUNK_CPS_RAM"/> - <value value="0x4b" name="A6XX_HLSQ_CHUNK_CVS_RAM_TAG"/> - <value value="0x4c" name="A6XX_HLSQ_CHUNK_CPS_RAM_TAG"/> - <value value="0x4d" name="A6XX_HLSQ_ICB_CVS_CB_BASE_TAG"/> - <value value="0x4e" name="A6XX_HLSQ_ICB_CPS_CB_BASE_TAG"/> - <value value="0x50" name="A6XX_HLSQ_CVS_MISC_RAM"/> - <value value="0x51" name="A6XX_HLSQ_CPS_MISC_RAM"/> - <value value="0x52" name="A6XX_HLSQ_INST_RAM"/> - <value value="0x53" name="A6XX_HLSQ_GFX_CVS_CONST_RAM"/> - <value value="0x54" name="A6XX_HLSQ_GFX_CPS_CONST_RAM"/> - <value value="0x55" name="A6XX_HLSQ_CVS_MISC_RAM_TAG"/> - <value value="0x56" name="A6XX_HLSQ_CPS_MISC_RAM_TAG"/> - <value value="0x57" name="A6XX_HLSQ_INST_RAM_TAG"/> - <value value="0x58" name="A6XX_HLSQ_GFX_CVS_CONST_RAM_TAG"/> - <value value="0x59" name="A6XX_HLSQ_GFX_CPS_CONST_RAM_TAG"/> - <value value="0x5a" name="A6XX_HLSQ_PWR_REST_RAM"/> - <value value="0x5b" name="A6XX_HLSQ_PWR_REST_TAG"/> - <value value="0x60" name="A6XX_HLSQ_DATAPATH_META"/> - <value value="0x61" name="A6XX_HLSQ_FRONTEND_META"/> - <value value="0x62" name="A6XX_HLSQ_INDIRECT_META"/> - <value value="0x63" name="A6XX_HLSQ_BACKEND_META"/> - <value value="0x70" name="A6XX_SP_LB_6_DATA"/> - <value value="0x71" name="A6XX_SP_LB_7_DATA"/> - <value value="0x73" name="A6XX_HLSQ_INST_RAM_1"/> -</enum> - -<enum name="a7xx_statetype_id"> - <value value="0" name="A7XX_TP0_NCTX_REG"/> - <value value="1" name="A7XX_TP0_CTX0_3D_CVS_REG"/> - <value value="2" name="A7XX_TP0_CTX0_3D_CPS_REG"/> - <value value="3" name="A7XX_TP0_CTX1_3D_CVS_REG"/> - <value value="4" name="A7XX_TP0_CTX1_3D_CPS_REG"/> - <value value="5" name="A7XX_TP0_CTX2_3D_CPS_REG"/> - <value value="6" name="A7XX_TP0_CTX3_3D_CPS_REG"/> - <value value="9" name="A7XX_TP0_TMO_DATA"/> - <value value="10" name="A7XX_TP0_SMO_DATA"/> - <value value="11" name="A7XX_TP0_MIPMAP_BASE_DATA"/> - <value value="32" name="A7XX_SP_NCTX_REG"/> - <value value="33" name="A7XX_SP_CTX0_3D_CVS_REG"/> - <value value="34" name="A7XX_SP_CTX0_3D_CPS_REG"/> - <value value="35" name="A7XX_SP_CTX1_3D_CVS_REG"/> - <value value="36" name="A7XX_SP_CTX1_3D_CPS_REG"/> - <value value="37" name="A7XX_SP_CTX2_3D_CPS_REG"/> - <value value="38" name="A7XX_SP_CTX3_3D_CPS_REG"/> - <value value="39" name="A7XX_SP_INST_DATA"/> - <value value="40" name="A7XX_SP_INST_DATA_1"/> - <value value="41" name="A7XX_SP_LB_0_DATA"/> - <value value="42" name="A7XX_SP_LB_1_DATA"/> - <value value="43" name="A7XX_SP_LB_2_DATA"/> - <value value="44" name="A7XX_SP_LB_3_DATA"/> - <value value="45" name="A7XX_SP_LB_4_DATA"/> - <value value="46" name="A7XX_SP_LB_5_DATA"/> - <value value="47" name="A7XX_SP_LB_6_DATA"/> - <value value="48" name="A7XX_SP_LB_7_DATA"/> - <value value="49" name="A7XX_SP_CB_RAM"/> - <value value="50" name="A7XX_SP_LB_13_DATA"/> - <value value="51" name="A7XX_SP_LB_14_DATA"/> - <value value="52" name="A7XX_SP_INST_TAG"/> - <value value="53" name="A7XX_SP_INST_DATA_2"/> - <value value="54" name="A7XX_SP_TMO_TAG"/> - <value value="55" name="A7XX_SP_SMO_TAG"/> - <value value="56" name="A7XX_SP_STATE_DATA"/> - <value value="57" name="A7XX_SP_HWAVE_RAM"/> - <value value="58" name="A7XX_SP_L0_INST_BUF"/> - <value value="59" name="A7XX_SP_LB_8_DATA"/> - <value value="60" name="A7XX_SP_LB_9_DATA"/> - <value value="61" name="A7XX_SP_LB_10_DATA"/> - <value value="62" name="A7XX_SP_LB_11_DATA"/> - <value value="63" name="A7XX_SP_LB_12_DATA"/> - <value value="64" name="A7XX_HLSQ_DATAPATH_DSTR_META"/> - <value value="67" name="A7XX_HLSQ_L2STC_TAG_RAM"/> - <value value="68" name="A7XX_HLSQ_L2STC_INFO_CMD"/> - <value value="69" name="A7XX_HLSQ_CVS_BE_CTXT_BUF_RAM_TAG"/> - <value value="70" name="A7XX_HLSQ_CPS_BE_CTXT_BUF_RAM_TAG"/> - <value value="71" name="A7XX_HLSQ_GFX_CVS_BE_CTXT_BUF_RAM"/> - <value value="72" name="A7XX_HLSQ_GFX_CPS_BE_CTXT_BUF_RAM"/> - <value value="73" name="A7XX_HLSQ_CHUNK_CVS_RAM"/> - <value value="74" name="A7XX_HLSQ_CHUNK_CPS_RAM"/> - <value value="75" name="A7XX_HLSQ_CHUNK_CVS_RAM_TAG"/> - <value value="76" name="A7XX_HLSQ_CHUNK_CPS_RAM_TAG"/> - <value value="77" name="A7XX_HLSQ_ICB_CVS_CB_BASE_TAG"/> - <value value="78" name="A7XX_HLSQ_ICB_CPS_CB_BASE_TAG"/> - <value value="79" name="A7XX_HLSQ_CVS_MISC_RAM"/> - <value value="80" name="A7XX_HLSQ_CPS_MISC_RAM"/> - <value value="81" name="A7XX_HLSQ_CPS_MISC_RAM_1"/> - <value value="82" name="A7XX_HLSQ_INST_RAM"/> - <value value="83" name="A7XX_HLSQ_GFX_CVS_CONST_RAM"/> - <value value="84" name="A7XX_HLSQ_GFX_CPS_CONST_RAM"/> - <value value="85" name="A7XX_HLSQ_CVS_MISC_RAM_TAG"/> - <value value="86" name="A7XX_HLSQ_CPS_MISC_RAM_TAG"/> - <value value="87" name="A7XX_HLSQ_INST_RAM_TAG"/> - <value value="88" name="A7XX_HLSQ_GFX_CVS_CONST_RAM_TAG"/> - <value value="89" name="A7XX_HLSQ_GFX_CPS_CONST_RAM_TAG"/> - <value value="90" name="A7XX_HLSQ_GFX_LOCAL_MISC_RAM"/> - <value value="91" name="A7XX_HLSQ_GFX_LOCAL_MISC_RAM_TAG"/> - <value value="92" name="A7XX_HLSQ_INST_RAM_1"/> - <value value="93" name="A7XX_HLSQ_STPROC_META"/> - <value value="94" name="A7XX_HLSQ_BV_BE_META"/> - <value value="95" name="A7XX_HLSQ_INST_RAM_2"/> - <value value="96" name="A7XX_HLSQ_DATAPATH_META"/> - <value value="97" name="A7XX_HLSQ_FRONTEND_META"/> - <value value="98" name="A7XX_HLSQ_INDIRECT_META"/> - <value value="99" name="A7XX_HLSQ_BACKEND_META"/> -</enum> - -<enum name="a6xx_debugbus_id"> - <value value="0x1" name="A6XX_DBGBUS_CP"/> - <value value="0x2" name="A6XX_DBGBUS_RBBM"/> - <value value="0x3" name="A6XX_DBGBUS_VBIF"/> - <value value="0x4" name="A6XX_DBGBUS_HLSQ"/> - <value value="0x5" name="A6XX_DBGBUS_UCHE"/> - <value value="0x6" name="A6XX_DBGBUS_DPM"/> - <value value="0x7" name="A6XX_DBGBUS_TESS"/> - <value value="0x8" name="A6XX_DBGBUS_PC"/> - <value value="0x9" name="A6XX_DBGBUS_VFDP"/> - <value value="0xa" name="A6XX_DBGBUS_VPC"/> - <value value="0xb" name="A6XX_DBGBUS_TSE"/> - <value value="0xc" name="A6XX_DBGBUS_RAS"/> - <value value="0xd" name="A6XX_DBGBUS_VSC"/> - <value value="0xe" name="A6XX_DBGBUS_COM"/> - <value value="0x10" name="A6XX_DBGBUS_LRZ"/> - <value value="0x11" name="A6XX_DBGBUS_A2D"/> - <value value="0x12" name="A6XX_DBGBUS_CCUFCHE"/> - <value value="0x13" name="A6XX_DBGBUS_GMU_CX"/> - <value value="0x14" name="A6XX_DBGBUS_RBP"/> - <value value="0x15" name="A6XX_DBGBUS_DCS"/> - <value value="0x16" name="A6XX_DBGBUS_DBGC"/> - <value value="0x17" name="A6XX_DBGBUS_CX"/> - <value value="0x18" name="A6XX_DBGBUS_GMU_GX"/> - <value value="0x19" name="A6XX_DBGBUS_TPFCHE"/> - <value value="0x1a" name="A6XX_DBGBUS_GBIF_GX"/> - <value value="0x1d" name="A6XX_DBGBUS_GPC"/> - <value value="0x1e" name="A6XX_DBGBUS_LARC"/> - <value value="0x1f" name="A6XX_DBGBUS_HLSQ_SPTP"/> - <value value="0x20" name="A6XX_DBGBUS_RB_0"/> - <value value="0x21" name="A6XX_DBGBUS_RB_1"/> - <value value="0x22" name="A6XX_DBGBUS_RB_2"/> - <value value="0x24" name="A6XX_DBGBUS_UCHE_WRAPPER"/> - <value value="0x28" name="A6XX_DBGBUS_CCU_0"/> - <value value="0x29" name="A6XX_DBGBUS_CCU_1"/> - <value value="0x2a" name="A6XX_DBGBUS_CCU_2"/> - <value value="0x38" name="A6XX_DBGBUS_VFD_0"/> - <value value="0x39" name="A6XX_DBGBUS_VFD_1"/> - <value value="0x3a" name="A6XX_DBGBUS_VFD_2"/> - <value value="0x3b" name="A6XX_DBGBUS_VFD_3"/> - <value value="0x3c" name="A6XX_DBGBUS_VFD_4"/> - <value value="0x3d" name="A6XX_DBGBUS_VFD_5"/> - <value value="0x40" name="A6XX_DBGBUS_SP_0"/> - <value value="0x41" name="A6XX_DBGBUS_SP_1"/> - <value value="0x42" name="A6XX_DBGBUS_SP_2"/> - <value value="0x48" name="A6XX_DBGBUS_TPL1_0"/> - <value value="0x49" name="A6XX_DBGBUS_TPL1_1"/> - <value value="0x4a" name="A6XX_DBGBUS_TPL1_2"/> - <value value="0x4b" name="A6XX_DBGBUS_TPL1_3"/> - <value value="0x4c" name="A6XX_DBGBUS_TPL1_4"/> - <value value="0x4d" name="A6XX_DBGBUS_TPL1_5"/> - <value value="0x58" name="A6XX_DBGBUS_SPTP_0"/> - <value value="0x59" name="A6XX_DBGBUS_SPTP_1"/> - <value value="0x5a" name="A6XX_DBGBUS_SPTP_2"/> - <value value="0x5b" name="A6XX_DBGBUS_SPTP_3"/> - <value value="0x5c" name="A6XX_DBGBUS_SPTP_4"/> - <value value="0x5d" name="A6XX_DBGBUS_SPTP_5"/> -</enum> - -<enum name="a7xx_state_location"> - <value value="0" name="A7XX_HLSQ_STATE"/> - <value value="1" name="A7XX_HLSQ_DP"/> - <value value="2" name="A7XX_SP_TOP"/> - <value value="3" name="A7XX_USPTP"/> - <value value="4" name="A7XX_HLSQ_DP_STR"/> -</enum> - -<enum name="a7xx_pipe"> - <value value="0" name="A7XX_PIPE_NONE"/> - <value value="1" name="A7XX_PIPE_BR"/> - <value value="2" name="A7XX_PIPE_BV"/> - <value value="3" name="A7XX_PIPE_LPAC"/> -</enum> - -<enum name="a7xx_cluster"> - <value value="0" name="A7XX_CLUSTER_NONE"/> - <value value="1" name="A7XX_CLUSTER_FE"/> - <value value="2" name="A7XX_CLUSTER_SP_VS"/> - <value value="3" name="A7XX_CLUSTER_PC_VS"/> - <value value="4" name="A7XX_CLUSTER_GRAS"/> - <value value="5" name="A7XX_CLUSTER_SP_PS"/> - <value value="6" name="A7XX_CLUSTER_VPC_PS"/> - <value value="7" name="A7XX_CLUSTER_PS"/> -</enum> - -<enum name="a7xx_debugbus_id"> - <value value="1" name="A7XX_DBGBUS_CP_0_0"/> - <value value="2" name="A7XX_DBGBUS_CP_0_1"/> - <value value="3" name="A7XX_DBGBUS_RBBM"/> - <value value="5" name="A7XX_DBGBUS_GBIF_GX"/> - <value value="6" name="A7XX_DBGBUS_GBIF_CX"/> - <value value="7" name="A7XX_DBGBUS_HLSQ"/> - <value value="9" name="A7XX_DBGBUS_UCHE_0"/> - <value value="10" name="A7XX_DBGBUS_UCHE_1"/> - <value value="13" name="A7XX_DBGBUS_TESS_BR"/> - <value value="14" name="A7XX_DBGBUS_TESS_BV"/> - <value value="17" name="A7XX_DBGBUS_PC_BR"/> - <value value="18" name="A7XX_DBGBUS_PC_BV"/> - <value value="21" name="A7XX_DBGBUS_VFDP_BR"/> - <value value="22" name="A7XX_DBGBUS_VFDP_BV"/> - <value value="25" name="A7XX_DBGBUS_VPC_BR"/> - <value value="26" name="A7XX_DBGBUS_VPC_BV"/> - <value value="29" name="A7XX_DBGBUS_TSE_BR"/> - <value value="30" name="A7XX_DBGBUS_TSE_BV"/> - <value value="33" name="A7XX_DBGBUS_RAS_BR"/> - <value value="34" name="A7XX_DBGBUS_RAS_BV"/> - <value value="37" name="A7XX_DBGBUS_VSC"/> - <value value="39" name="A7XX_DBGBUS_COM_0"/> - <value value="43" name="A7XX_DBGBUS_LRZ_BR"/> - <value value="44" name="A7XX_DBGBUS_LRZ_BV"/> - <value value="47" name="A7XX_DBGBUS_UFC_0"/> - <value value="48" name="A7XX_DBGBUS_UFC_1"/> - <value value="55" name="A7XX_DBGBUS_GMU_GX"/> - <value value="59" name="A7XX_DBGBUS_DBGC"/> - <value value="60" name="A7XX_DBGBUS_CX"/> - <value value="61" name="A7XX_DBGBUS_GMU_CX"/> - <value value="62" name="A7XX_DBGBUS_GPC_BR"/> - <value value="63" name="A7XX_DBGBUS_GPC_BV"/> - <value value="66" name="A7XX_DBGBUS_LARC"/> - <value value="68" name="A7XX_DBGBUS_HLSQ_SPTP"/> - <value value="70" name="A7XX_DBGBUS_RB_0"/> - <value value="71" name="A7XX_DBGBUS_RB_1"/> - <value value="72" name="A7XX_DBGBUS_RB_2"/> - <value value="73" name="A7XX_DBGBUS_RB_3"/> - <value value="74" name="A7XX_DBGBUS_RB_4"/> - <value value="75" name="A7XX_DBGBUS_RB_5"/> - <value value="102" name="A7XX_DBGBUS_UCHE_WRAPPER"/> - <value value="106" name="A7XX_DBGBUS_CCU_0"/> - <value value="107" name="A7XX_DBGBUS_CCU_1"/> - <value value="108" name="A7XX_DBGBUS_CCU_2"/> - <value value="109" name="A7XX_DBGBUS_CCU_3"/> - <value value="110" name="A7XX_DBGBUS_CCU_4"/> - <value value="111" name="A7XX_DBGBUS_CCU_5"/> - <value value="138" name="A7XX_DBGBUS_VFD_BR_0"/> - <value value="139" name="A7XX_DBGBUS_VFD_BR_1"/> - <value value="140" name="A7XX_DBGBUS_VFD_BR_2"/> - <value value="141" name="A7XX_DBGBUS_VFD_BR_3"/> - <value value="142" name="A7XX_DBGBUS_VFD_BR_4"/> - <value value="143" name="A7XX_DBGBUS_VFD_BR_5"/> - <value value="144" name="A7XX_DBGBUS_VFD_BR_6"/> - <value value="145" name="A7XX_DBGBUS_VFD_BR_7"/> - <value value="202" name="A7XX_DBGBUS_VFD_BV_0"/> - <value value="203" name="A7XX_DBGBUS_VFD_BV_1"/> - <value value="204" name="A7XX_DBGBUS_VFD_BV_2"/> - <value value="205" name="A7XX_DBGBUS_VFD_BV_3"/> - <value value="234" name="A7XX_DBGBUS_USP_0"/> - <value value="235" name="A7XX_DBGBUS_USP_1"/> - <value value="236" name="A7XX_DBGBUS_USP_2"/> - <value value="237" name="A7XX_DBGBUS_USP_3"/> - <value value="238" name="A7XX_DBGBUS_USP_4"/> - <value value="239" name="A7XX_DBGBUS_USP_5"/> - <value value="266" name="A7XX_DBGBUS_TP_0"/> - <value value="267" name="A7XX_DBGBUS_TP_1"/> - <value value="268" name="A7XX_DBGBUS_TP_2"/> - <value value="269" name="A7XX_DBGBUS_TP_3"/> - <value value="270" name="A7XX_DBGBUS_TP_4"/> - <value value="271" name="A7XX_DBGBUS_TP_5"/> - <value value="272" name="A7XX_DBGBUS_TP_6"/> - <value value="273" name="A7XX_DBGBUS_TP_7"/> - <value value="274" name="A7XX_DBGBUS_TP_8"/> - <value value="275" name="A7XX_DBGBUS_TP_9"/> - <value value="276" name="A7XX_DBGBUS_TP_10"/> - <value value="277" name="A7XX_DBGBUS_TP_11"/> - <value value="330" name="A7XX_DBGBUS_USPTP_0"/> - <value value="331" name="A7XX_DBGBUS_USPTP_1"/> - <value value="332" name="A7XX_DBGBUS_USPTP_2"/> - <value value="333" name="A7XX_DBGBUS_USPTP_3"/> - <value value="334" name="A7XX_DBGBUS_USPTP_4"/> - <value value="335" name="A7XX_DBGBUS_USPTP_5"/> - <value value="336" name="A7XX_DBGBUS_USPTP_6"/> - <value value="337" name="A7XX_DBGBUS_USPTP_7"/> - <value value="338" name="A7XX_DBGBUS_USPTP_8"/> - <value value="339" name="A7XX_DBGBUS_USPTP_9"/> - <value value="340" name="A7XX_DBGBUS_USPTP_10"/> - <value value="341" name="A7XX_DBGBUS_USPTP_11"/> - <value value="396" name="A7XX_DBGBUS_CCHE_0"/> - <value value="397" name="A7XX_DBGBUS_CCHE_1"/> - <value value="398" name="A7XX_DBGBUS_CCHE_2"/> - <value value="408" name="A7XX_DBGBUS_VPC_DSTR_0"/> - <value value="409" name="A7XX_DBGBUS_VPC_DSTR_1"/> - <value value="410" name="A7XX_DBGBUS_VPC_DSTR_2"/> - <value value="411" name="A7XX_DBGBUS_HLSQ_DP_STR_0"/> - <value value="412" name="A7XX_DBGBUS_HLSQ_DP_STR_1"/> - <value value="413" name="A7XX_DBGBUS_HLSQ_DP_STR_2"/> - <value value="414" name="A7XX_DBGBUS_HLSQ_DP_STR_3"/> - <value value="415" name="A7XX_DBGBUS_HLSQ_DP_STR_4"/> - <value value="416" name="A7XX_DBGBUS_HLSQ_DP_STR_5"/> - <value value="443" name="A7XX_DBGBUS_UFC_DSTR_0"/> - <value value="444" name="A7XX_DBGBUS_UFC_DSTR_1"/> - <value value="445" name="A7XX_DBGBUS_UFC_DSTR_2"/> - <value value="446" name="A7XX_DBGBUS_CGC_SUBCORE"/> - <value value="447" name="A7XX_DBGBUS_CGC_CORE"/> -</enum> - -<enum name="a6xx_cp_perfcounter_select"> - <value value="0" name="PERF_CP_ALWAYS_COUNT"/> - <value value="1" name="PERF_CP_BUSY_GFX_CORE_IDLE"/> - <value value="2" name="PERF_CP_BUSY_CYCLES"/> - <value value="3" name="PERF_CP_NUM_PREEMPTIONS"/> - <value value="4" name="PERF_CP_PREEMPTION_REACTION_DELAY"/> - <value value="5" name="PERF_CP_PREEMPTION_SWITCH_OUT_TIME"/> - <value value="6" name="PERF_CP_PREEMPTION_SWITCH_IN_TIME"/> - <value value="7" name="PERF_CP_DEAD_DRAWS_IN_BIN_RENDER"/> - <value value="8" name="PERF_CP_PREDICATED_DRAWS_KILLED"/> - <value value="9" name="PERF_CP_MODE_SWITCH"/> - <value value="10" name="PERF_CP_ZPASS_DONE"/> - <value value="11" name="PERF_CP_CONTEXT_DONE"/> - <value value="12" name="PERF_CP_CACHE_FLUSH"/> - <value value="13" name="PERF_CP_LONG_PREEMPTIONS"/> - <value value="14" name="PERF_CP_SQE_I_CACHE_STARVE"/> - <value value="15" name="PERF_CP_SQE_IDLE"/> - <value value="16" name="PERF_CP_SQE_PM4_STARVE_RB_IB"/> - <value value="17" name="PERF_CP_SQE_PM4_STARVE_SDS"/> - <value value="18" name="PERF_CP_SQE_MRB_STARVE"/> - <value value="19" name="PERF_CP_SQE_RRB_STARVE"/> - <value value="20" name="PERF_CP_SQE_VSD_STARVE"/> - <value value="21" name="PERF_CP_VSD_DECODE_STARVE"/> - <value value="22" name="PERF_CP_SQE_PIPE_OUT_STALL"/> - <value value="23" name="PERF_CP_SQE_SYNC_STALL"/> - <value value="24" name="PERF_CP_SQE_PM4_WFI_STALL"/> - <value value="25" name="PERF_CP_SQE_SYS_WFI_STALL"/> - <value value="26" name="PERF_CP_SQE_T4_EXEC"/> - <value value="27" name="PERF_CP_SQE_LOAD_STATE_EXEC"/> - <value value="28" name="PERF_CP_SQE_SAVE_SDS_STATE"/> - <value value="29" name="PERF_CP_SQE_DRAW_EXEC"/> - <value value="30" name="PERF_CP_SQE_CTXT_REG_BUNCH_EXEC"/> - <value value="31" name="PERF_CP_SQE_EXEC_PROFILED"/> - <value value="32" name="PERF_CP_MEMORY_POOL_EMPTY"/> - <value value="33" name="PERF_CP_MEMORY_POOL_SYNC_STALL"/> - <value value="34" name="PERF_CP_MEMORY_POOL_ABOVE_THRESH"/> - <value value="35" name="PERF_CP_AHB_WR_STALL_PRE_DRAWS"/> - <value value="36" name="PERF_CP_AHB_STALL_SQE_GMU"/> - <value value="37" name="PERF_CP_AHB_STALL_SQE_WR_OTHER"/> - <value value="38" name="PERF_CP_AHB_STALL_SQE_RD_OTHER"/> - <value value="39" name="PERF_CP_CLUSTER0_EMPTY"/> - <value value="40" name="PERF_CP_CLUSTER1_EMPTY"/> - <value value="41" name="PERF_CP_CLUSTER2_EMPTY"/> - <value value="42" name="PERF_CP_CLUSTER3_EMPTY"/> - <value value="43" name="PERF_CP_CLUSTER4_EMPTY"/> - <value value="44" name="PERF_CP_CLUSTER5_EMPTY"/> - <value value="45" name="PERF_CP_PM4_DATA"/> - <value value="46" name="PERF_CP_PM4_HEADERS"/> - <value value="47" name="PERF_CP_VBIF_READ_BEATS"/> - <value value="48" name="PERF_CP_VBIF_WRITE_BEATS"/> - <value value="49" name="PERF_CP_SQE_INSTR_COUNTER"/> -</enum> - -<enum name="a6xx_rbbm_perfcounter_select"> - <value value="0" name="PERF_RBBM_ALWAYS_COUNT"/> - <value value="1" name="PERF_RBBM_ALWAYS_ON"/> - <value value="2" name="PERF_RBBM_TSE_BUSY"/> - <value value="3" name="PERF_RBBM_RAS_BUSY"/> - <value value="4" name="PERF_RBBM_PC_DCALL_BUSY"/> - <value value="5" name="PERF_RBBM_PC_VSD_BUSY"/> - <value value="6" name="PERF_RBBM_STATUS_MASKED"/> - <value value="7" name="PERF_RBBM_COM_BUSY"/> - <value value="8" name="PERF_RBBM_DCOM_BUSY"/> - <value value="9" name="PERF_RBBM_VBIF_BUSY"/> - <value value="10" name="PERF_RBBM_VSC_BUSY"/> - <value value="11" name="PERF_RBBM_TESS_BUSY"/> - <value value="12" name="PERF_RBBM_UCHE_BUSY"/> - <value value="13" name="PERF_RBBM_HLSQ_BUSY"/> -</enum> - -<enum name="a6xx_pc_perfcounter_select"> - <value value="0" name="PERF_PC_BUSY_CYCLES"/> - <value value="1" name="PERF_PC_WORKING_CYCLES"/> - <value value="2" name="PERF_PC_STALL_CYCLES_VFD"/> - <value value="3" name="PERF_PC_STALL_CYCLES_TSE"/> - <value value="4" name="PERF_PC_STALL_CYCLES_VPC"/> - <value value="5" name="PERF_PC_STALL_CYCLES_UCHE"/> - <value value="6" name="PERF_PC_STALL_CYCLES_TESS"/> - <value value="7" name="PERF_PC_STALL_CYCLES_TSE_ONLY"/> - <value value="8" name="PERF_PC_STALL_CYCLES_VPC_ONLY"/> - <value value="9" name="PERF_PC_PASS1_TF_STALL_CYCLES"/> - <value value="10" name="PERF_PC_STARVE_CYCLES_FOR_INDEX"/> - <value value="11" name="PERF_PC_STARVE_CYCLES_FOR_TESS_FACTOR"/> - <value value="12" name="PERF_PC_STARVE_CYCLES_FOR_VIZ_STREAM"/> - <value value="13" name="PERF_PC_STARVE_CYCLES_FOR_POSITION"/> - <value value="14" name="PERF_PC_STARVE_CYCLES_DI"/> - <value value="15" name="PERF_PC_VIS_STREAMS_LOADED"/> - <value value="16" name="PERF_PC_INSTANCES"/> - <value value="17" name="PERF_PC_VPC_PRIMITIVES"/> - <value value="18" name="PERF_PC_DEAD_PRIM"/> - <value value="19" name="PERF_PC_LIVE_PRIM"/> - <value value="20" name="PERF_PC_VERTEX_HITS"/> - <value value="21" name="PERF_PC_IA_VERTICES"/> - <value value="22" name="PERF_PC_IA_PRIMITIVES"/> - <value value="23" name="PERF_PC_GS_PRIMITIVES"/> - <value value="24" name="PERF_PC_HS_INVOCATIONS"/> - <value value="25" name="PERF_PC_DS_INVOCATIONS"/> - <value value="26" name="PERF_PC_VS_INVOCATIONS"/> - <value value="27" name="PERF_PC_GS_INVOCATIONS"/> - <value value="28" name="PERF_PC_DS_PRIMITIVES"/> - <value value="29" name="PERF_PC_VPC_POS_DATA_TRANSACTION"/> - <value value="30" name="PERF_PC_3D_DRAWCALLS"/> - <value value="31" name="PERF_PC_2D_DRAWCALLS"/> - <value value="32" name="PERF_PC_NON_DRAWCALL_GLOBAL_EVENTS"/> - <value value="33" name="PERF_TESS_BUSY_CYCLES"/> - <value value="34" name="PERF_TESS_WORKING_CYCLES"/> - <value value="35" name="PERF_TESS_STALL_CYCLES_PC"/> - <value value="36" name="PERF_TESS_STARVE_CYCLES_PC"/> - <value value="37" name="PERF_PC_TSE_TRANSACTION"/> - <value value="38" name="PERF_PC_TSE_VERTEX"/> - <value value="39" name="PERF_PC_TESS_PC_UV_TRANS"/> - <value value="40" name="PERF_PC_TESS_PC_UV_PATCHES"/> - <value value="41" name="PERF_PC_TESS_FACTOR_TRANS"/> -</enum> - -<enum name="a6xx_vfd_perfcounter_select"> - <value value="0" name="PERF_VFD_BUSY_CYCLES"/> - <value value="1" name="PERF_VFD_STALL_CYCLES_UCHE"/> - <value value="2" name="PERF_VFD_STALL_CYCLES_VPC_ALLOC"/> - <value value="3" name="PERF_VFD_STALL_CYCLES_SP_INFO"/> - <value value="4" name="PERF_VFD_STALL_CYCLES_SP_ATTR"/> - <value value="5" name="PERF_VFD_STARVE_CYCLES_UCHE"/> - <value value="6" name="PERF_VFD_RBUFFER_FULL"/> - <value value="7" name="PERF_VFD_ATTR_INFO_FIFO_FULL"/> - <value value="8" name="PERF_VFD_DECODED_ATTRIBUTE_BYTES"/> - <value value="9" name="PERF_VFD_NUM_ATTRIBUTES"/> - <value value="10" name="PERF_VFD_UPPER_SHADER_FIBERS"/> - <value value="11" name="PERF_VFD_LOWER_SHADER_FIBERS"/> - <value value="12" name="PERF_VFD_MODE_0_FIBERS"/> - <value value="13" name="PERF_VFD_MODE_1_FIBERS"/> - <value value="14" name="PERF_VFD_MODE_2_FIBERS"/> - <value value="15" name="PERF_VFD_MODE_3_FIBERS"/> - <value value="16" name="PERF_VFD_MODE_4_FIBERS"/> - <value value="17" name="PERF_VFD_TOTAL_VERTICES"/> - <value value="18" name="PERF_VFDP_STALL_CYCLES_VFD"/> - <value value="19" name="PERF_VFDP_STALL_CYCLES_VFD_INDEX"/> - <value value="20" name="PERF_VFDP_STALL_CYCLES_VFD_PROG"/> - <value value="21" name="PERF_VFDP_STARVE_CYCLES_PC"/> - <value value="22" name="PERF_VFDP_VS_STAGE_WAVES"/> -</enum> - -<enum name="a6xx_hlsq_perfcounter_select"> - <value value="0" name="PERF_HLSQ_BUSY_CYCLES"/> - <value value="1" name="PERF_HLSQ_STALL_CYCLES_UCHE"/> - <value value="2" name="PERF_HLSQ_STALL_CYCLES_SP_STATE"/> - <value value="3" name="PERF_HLSQ_STALL_CYCLES_SP_FS_STAGE"/> - <value value="4" name="PERF_HLSQ_UCHE_LATENCY_CYCLES"/> - <value value="5" name="PERF_HLSQ_UCHE_LATENCY_COUNT"/> - <value value="6" name="PERF_HLSQ_FS_STAGE_1X_WAVES"/> - <value value="7" name="PERF_HLSQ_FS_STAGE_2X_WAVES"/> - <value value="8" name="PERF_HLSQ_QUADS"/> - <value value="9" name="PERF_HLSQ_CS_INVOCATIONS"/> - <value value="10" name="PERF_HLSQ_COMPUTE_DRAWCALLS"/> - <value value="11" name="PERF_HLSQ_FS_DATA_WAIT_PROGRAMMING"/> - <value value="12" name="PERF_HLSQ_DUAL_FS_PROG_ACTIVE"/> - <value value="13" name="PERF_HLSQ_DUAL_VS_PROG_ACTIVE"/> - <value value="14" name="PERF_HLSQ_FS_BATCH_COUNT_ZERO"/> - <value value="15" name="PERF_HLSQ_VS_BATCH_COUNT_ZERO"/> - <value value="16" name="PERF_HLSQ_WAVE_PENDING_NO_QUAD"/> - <value value="17" name="PERF_HLSQ_WAVE_PENDING_NO_PRIM_BASE"/> - <value value="18" name="PERF_HLSQ_STALL_CYCLES_VPC"/> - <value value="19" name="PERF_HLSQ_PIXELS"/> - <value value="20" name="PERF_HLSQ_DRAW_MODE_SWITCH_VSFS_SYNC"/> -</enum> - -<enum name="a6xx_vpc_perfcounter_select"> - <value value="0" name="PERF_VPC_BUSY_CYCLES"/> - <value value="1" name="PERF_VPC_WORKING_CYCLES"/> - <value value="2" name="PERF_VPC_STALL_CYCLES_UCHE"/> - <value value="3" name="PERF_VPC_STALL_CYCLES_VFD_WACK"/> - <value value="4" name="PERF_VPC_STALL_CYCLES_HLSQ_PRIM_ALLOC"/> - <value value="5" name="PERF_VPC_STALL_CYCLES_PC"/> - <value value="6" name="PERF_VPC_STALL_CYCLES_SP_LM"/> - <value value="7" name="PERF_VPC_STARVE_CYCLES_SP"/> - <value value="8" name="PERF_VPC_STARVE_CYCLES_LRZ"/> - <value value="9" name="PERF_VPC_PC_PRIMITIVES"/> - <value value="10" name="PERF_VPC_SP_COMPONENTS"/> - <value value="11" name="PERF_VPC_STALL_CYCLES_VPCRAM_POS"/> - <value value="12" name="PERF_VPC_LRZ_ASSIGN_PRIMITIVES"/> - <value value="13" name="PERF_VPC_RB_VISIBLE_PRIMITIVES"/> - <value value="14" name="PERF_VPC_LM_TRANSACTION"/> - <value value="15" name="PERF_VPC_STREAMOUT_TRANSACTION"/> - <value value="16" name="PERF_VPC_VS_BUSY_CYCLES"/> - <value value="17" name="PERF_VPC_PS_BUSY_CYCLES"/> - <value value="18" name="PERF_VPC_VS_WORKING_CYCLES"/> - <value value="19" name="PERF_VPC_PS_WORKING_CYCLES"/> - <value value="20" name="PERF_VPC_STARVE_CYCLES_RB"/> - <value value="21" name="PERF_VPC_NUM_VPCRAM_READ_POS"/> - <value value="22" name="PERF_VPC_WIT_FULL_CYCLES"/> - <value value="23" name="PERF_VPC_VPCRAM_FULL_CYCLES"/> - <value value="24" name="PERF_VPC_LM_FULL_WAIT_FOR_INTP_END"/> - <value value="25" name="PERF_VPC_NUM_VPCRAM_WRITE"/> - <value value="26" name="PERF_VPC_NUM_VPCRAM_READ_SO"/> - <value value="27" name="PERF_VPC_NUM_ATTR_REQ_LM"/> -</enum> - -<enum name="a6xx_tse_perfcounter_select"> - <value value="0" name="PERF_TSE_BUSY_CYCLES"/> - <value value="1" name="PERF_TSE_CLIPPING_CYCLES"/> - <value value="2" name="PERF_TSE_STALL_CYCLES_RAS"/> - <value value="3" name="PERF_TSE_STALL_CYCLES_LRZ_BARYPLANE"/> - <value value="4" name="PERF_TSE_STALL_CYCLES_LRZ_ZPLANE"/> - <value value="5" name="PERF_TSE_STARVE_CYCLES_PC"/> - <value value="6" name="PERF_TSE_INPUT_PRIM"/> - <value value="7" name="PERF_TSE_INPUT_NULL_PRIM"/> - <value value="8" name="PERF_TSE_TRIVAL_REJ_PRIM"/> - <value value="9" name="PERF_TSE_CLIPPED_PRIM"/> - <value value="10" name="PERF_TSE_ZERO_AREA_PRIM"/> - <value value="11" name="PERF_TSE_FACENESS_CULLED_PRIM"/> - <value value="12" name="PERF_TSE_ZERO_PIXEL_PRIM"/> - <value value="13" name="PERF_TSE_OUTPUT_NULL_PRIM"/> - <value value="14" name="PERF_TSE_OUTPUT_VISIBLE_PRIM"/> - <value value="15" name="PERF_TSE_CINVOCATION"/> - <value value="16" name="PERF_TSE_CPRIMITIVES"/> - <value value="17" name="PERF_TSE_2D_INPUT_PRIM"/> - <value value="18" name="PERF_TSE_2D_ALIVE_CYCLES"/> - <value value="19" name="PERF_TSE_CLIP_PLANES"/> -</enum> - -<enum name="a6xx_ras_perfcounter_select"> - <value value="0" name="PERF_RAS_BUSY_CYCLES"/> - <value value="1" name="PERF_RAS_SUPERTILE_ACTIVE_CYCLES"/> - <value value="2" name="PERF_RAS_STALL_CYCLES_LRZ"/> - <value value="3" name="PERF_RAS_STARVE_CYCLES_TSE"/> - <value value="4" name="PERF_RAS_SUPER_TILES"/> - <value value="5" name="PERF_RAS_8X4_TILES"/> - <value value="6" name="PERF_RAS_MASKGEN_ACTIVE"/> - <value value="7" name="PERF_RAS_FULLY_COVERED_SUPER_TILES"/> - <value value="8" name="PERF_RAS_FULLY_COVERED_8X4_TILES"/> - <value value="9" name="PERF_RAS_PRIM_KILLED_INVISILBE"/> - <value value="10" name="PERF_RAS_SUPERTILE_GEN_ACTIVE_CYCLES"/> - <value value="11" name="PERF_RAS_LRZ_INTF_WORKING_CYCLES"/> - <value value="12" name="PERF_RAS_BLOCKS"/> -</enum> - -<enum name="a6xx_uche_perfcounter_select"> - <value value="0" name="PERF_UCHE_BUSY_CYCLES"/> - <value value="1" name="PERF_UCHE_STALL_CYCLES_ARBITER"/> - <value value="2" name="PERF_UCHE_VBIF_LATENCY_CYCLES"/> - <value value="3" name="PERF_UCHE_VBIF_LATENCY_SAMPLES"/> - <value value="4" name="PERF_UCHE_VBIF_READ_BEATS_TP"/> - <value value="5" name="PERF_UCHE_VBIF_READ_BEATS_VFD"/> - <value value="6" name="PERF_UCHE_VBIF_READ_BEATS_HLSQ"/> - <value value="7" name="PERF_UCHE_VBIF_READ_BEATS_LRZ"/> - <value value="8" name="PERF_UCHE_VBIF_READ_BEATS_SP"/> - <value value="9" name="PERF_UCHE_READ_REQUESTS_TP"/> - <value value="10" name="PERF_UCHE_READ_REQUESTS_VFD"/> - <value value="11" name="PERF_UCHE_READ_REQUESTS_HLSQ"/> - <value value="12" name="PERF_UCHE_READ_REQUESTS_LRZ"/> - <value value="13" name="PERF_UCHE_READ_REQUESTS_SP"/> - <value value="14" name="PERF_UCHE_WRITE_REQUESTS_LRZ"/> - <value value="15" name="PERF_UCHE_WRITE_REQUESTS_SP"/> - <value value="16" name="PERF_UCHE_WRITE_REQUESTS_VPC"/> - <value value="17" name="PERF_UCHE_WRITE_REQUESTS_VSC"/> - <value value="18" name="PERF_UCHE_EVICTS"/> - <value value="19" name="PERF_UCHE_BANK_REQ0"/> - <value value="20" name="PERF_UCHE_BANK_REQ1"/> - <value value="21" name="PERF_UCHE_BANK_REQ2"/> - <value value="22" name="PERF_UCHE_BANK_REQ3"/> - <value value="23" name="PERF_UCHE_BANK_REQ4"/> - <value value="24" name="PERF_UCHE_BANK_REQ5"/> - <value value="25" name="PERF_UCHE_BANK_REQ6"/> - <value value="26" name="PERF_UCHE_BANK_REQ7"/> - <value value="27" name="PERF_UCHE_VBIF_READ_BEATS_CH0"/> - <value value="28" name="PERF_UCHE_VBIF_READ_BEATS_CH1"/> - <value value="29" name="PERF_UCHE_GMEM_READ_BEATS"/> - <value value="30" name="PERF_UCHE_TPH_REF_FULL"/> - <value value="31" name="PERF_UCHE_TPH_VICTIM_FULL"/> - <value value="32" name="PERF_UCHE_TPH_EXT_FULL"/> - <value value="33" name="PERF_UCHE_VBIF_STALL_WRITE_DATA"/> - <value value="34" name="PERF_UCHE_DCMP_LATENCY_SAMPLES"/> - <value value="35" name="PERF_UCHE_DCMP_LATENCY_CYCLES"/> - <value value="36" name="PERF_UCHE_VBIF_READ_BEATS_PC"/> - <value value="37" name="PERF_UCHE_READ_REQUESTS_PC"/> - <value value="38" name="PERF_UCHE_RAM_READ_REQ"/> - <value value="39" name="PERF_UCHE_RAM_WRITE_REQ"/> -</enum> - -<enum name="a6xx_tp_perfcounter_select"> - <value value="0" name="PERF_TP_BUSY_CYCLES"/> - <value value="1" name="PERF_TP_STALL_CYCLES_UCHE"/> - <value value="2" name="PERF_TP_LATENCY_CYCLES"/> - <value value="3" name="PERF_TP_LATENCY_TRANS"/> - <value value="4" name="PERF_TP_FLAG_CACHE_REQUEST_SAMPLES"/> - <value value="5" name="PERF_TP_FLAG_CACHE_REQUEST_LATENCY"/> - <value value="6" name="PERF_TP_L1_CACHELINE_REQUESTS"/> - <value value="7" name="PERF_TP_L1_CACHELINE_MISSES"/> - <value value="8" name="PERF_TP_SP_TP_TRANS"/> - <value value="9" name="PERF_TP_TP_SP_TRANS"/> - <value value="10" name="PERF_TP_OUTPUT_PIXELS"/> - <value value="11" name="PERF_TP_FILTER_WORKLOAD_16BIT"/> - <value value="12" name="PERF_TP_FILTER_WORKLOAD_32BIT"/> - <value value="13" name="PERF_TP_QUADS_RECEIVED"/> - <value value="14" name="PERF_TP_QUADS_OFFSET"/> - <value value="15" name="PERF_TP_QUADS_SHADOW"/> - <value value="16" name="PERF_TP_QUADS_ARRAY"/> - <value value="17" name="PERF_TP_QUADS_GRADIENT"/> - <value value="18" name="PERF_TP_QUADS_1D"/> - <value value="19" name="PERF_TP_QUADS_2D"/> - <value value="20" name="PERF_TP_QUADS_BUFFER"/> - <value value="21" name="PERF_TP_QUADS_3D"/> - <value value="22" name="PERF_TP_QUADS_CUBE"/> - <value value="23" name="PERF_TP_DIVERGENT_QUADS_RECEIVED"/> - <value value="24" name="PERF_TP_PRT_NON_RESIDENT_EVENTS"/> - <value value="25" name="PERF_TP_OUTPUT_PIXELS_POINT"/> - <value value="26" name="PERF_TP_OUTPUT_PIXELS_BILINEAR"/> - <value value="27" name="PERF_TP_OUTPUT_PIXELS_MIP"/> - <value value="28" name="PERF_TP_OUTPUT_PIXELS_ANISO"/> - <value value="29" name="PERF_TP_OUTPUT_PIXELS_ZERO_LOD"/> - <value value="30" name="PERF_TP_FLAG_CACHE_REQUESTS"/> - <value value="31" name="PERF_TP_FLAG_CACHE_MISSES"/> - <value value="32" name="PERF_TP_L1_5_L2_REQUESTS"/> - <value value="33" name="PERF_TP_2D_OUTPUT_PIXELS"/> - <value value="34" name="PERF_TP_2D_OUTPUT_PIXELS_POINT"/> - <value value="35" name="PERF_TP_2D_OUTPUT_PIXELS_BILINEAR"/> - <value value="36" name="PERF_TP_2D_FILTER_WORKLOAD_16BIT"/> - <value value="37" name="PERF_TP_2D_FILTER_WORKLOAD_32BIT"/> - <value value="38" name="PERF_TP_TPA2TPC_TRANS"/> - <value value="39" name="PERF_TP_L1_MISSES_ASTC_1TILE"/> - <value value="40" name="PERF_TP_L1_MISSES_ASTC_2TILE"/> - <value value="41" name="PERF_TP_L1_MISSES_ASTC_4TILE"/> - <value value="42" name="PERF_TP_L1_5_L2_COMPRESS_REQS"/> - <value value="43" name="PERF_TP_L1_5_L2_COMPRESS_MISS"/> - <value value="44" name="PERF_TP_L1_BANK_CONFLICT"/> - <value value="45" name="PERF_TP_L1_5_MISS_LATENCY_CYCLES"/> - <value value="46" name="PERF_TP_L1_5_MISS_LATENCY_TRANS"/> - <value value="47" name="PERF_TP_QUADS_CONSTANT_MULTIPLIED"/> - <value value="48" name="PERF_TP_FRONTEND_WORKING_CYCLES"/> - <value value="49" name="PERF_TP_L1_TAG_WORKING_CYCLES"/> - <value value="50" name="PERF_TP_L1_DATA_WRITE_WORKING_CYCLES"/> - <value value="51" name="PERF_TP_PRE_L1_DECOM_WORKING_CYCLES"/> - <value value="52" name="PERF_TP_BACKEND_WORKING_CYCLES"/> - <value value="53" name="PERF_TP_FLAG_CACHE_WORKING_CYCLES"/> - <value value="54" name="PERF_TP_L1_5_CACHE_WORKING_CYCLES"/> - <value value="55" name="PERF_TP_STARVE_CYCLES_SP"/> - <value value="56" name="PERF_TP_STARVE_CYCLES_UCHE"/> -</enum> - -<enum name="a6xx_sp_perfcounter_select"> - <value value="0" name="PERF_SP_BUSY_CYCLES"/> - <value value="1" name="PERF_SP_ALU_WORKING_CYCLES"/> - <value value="2" name="PERF_SP_EFU_WORKING_CYCLES"/> - <value value="3" name="PERF_SP_STALL_CYCLES_VPC"/> - <value value="4" name="PERF_SP_STALL_CYCLES_TP"/> - <value value="5" name="PERF_SP_STALL_CYCLES_UCHE"/> - <value value="6" name="PERF_SP_STALL_CYCLES_RB"/> - <value value="7" name="PERF_SP_NON_EXECUTION_CYCLES"/> - <value value="8" name="PERF_SP_WAVE_CONTEXTS"/> - <value value="9" name="PERF_SP_WAVE_CONTEXT_CYCLES"/> - <value value="10" name="PERF_SP_FS_STAGE_WAVE_CYCLES"/> - <value value="11" name="PERF_SP_FS_STAGE_WAVE_SAMPLES"/> - <value value="12" name="PERF_SP_VS_STAGE_WAVE_CYCLES"/> - <value value="13" name="PERF_SP_VS_STAGE_WAVE_SAMPLES"/> - <value value="14" name="PERF_SP_FS_STAGE_DURATION_CYCLES"/> - <value value="15" name="PERF_SP_VS_STAGE_DURATION_CYCLES"/> - <value value="16" name="PERF_SP_WAVE_CTRL_CYCLES"/> - <value value="17" name="PERF_SP_WAVE_LOAD_CYCLES"/> - <value value="18" name="PERF_SP_WAVE_EMIT_CYCLES"/> - <value value="19" name="PERF_SP_WAVE_NOP_CYCLES"/> - <value value="20" name="PERF_SP_WAVE_WAIT_CYCLES"/> - <value value="21" name="PERF_SP_WAVE_FETCH_CYCLES"/> - <value value="22" name="PERF_SP_WAVE_IDLE_CYCLES"/> - <value value="23" name="PERF_SP_WAVE_END_CYCLES"/> - <value value="24" name="PERF_SP_WAVE_LONG_SYNC_CYCLES"/> - <value value="25" name="PERF_SP_WAVE_SHORT_SYNC_CYCLES"/> - <value value="26" name="PERF_SP_WAVE_JOIN_CYCLES"/> - <value value="27" name="PERF_SP_LM_LOAD_INSTRUCTIONS"/> - <value value="28" name="PERF_SP_LM_STORE_INSTRUCTIONS"/> - <value value="29" name="PERF_SP_LM_ATOMICS"/> - <value value="30" name="PERF_SP_GM_LOAD_INSTRUCTIONS"/> - <value value="31" name="PERF_SP_GM_STORE_INSTRUCTIONS"/> - <value value="32" name="PERF_SP_GM_ATOMICS"/> - <value value="33" name="PERF_SP_VS_STAGE_TEX_INSTRUCTIONS"/> - <value value="34" name="PERF_SP_VS_STAGE_EFU_INSTRUCTIONS"/> - <value value="35" name="PERF_SP_VS_STAGE_FULL_ALU_INSTRUCTIONS"/> - <value value="36" name="PERF_SP_VS_STAGE_HALF_ALU_INSTRUCTIONS"/> - <value value="37" name="PERF_SP_FS_STAGE_TEX_INSTRUCTIONS"/> - <value value="38" name="PERF_SP_FS_STAGE_CFLOW_INSTRUCTIONS"/> - <value value="39" name="PERF_SP_FS_STAGE_EFU_INSTRUCTIONS"/> - <value value="40" name="PERF_SP_FS_STAGE_FULL_ALU_INSTRUCTIONS"/> - <value value="41" name="PERF_SP_FS_STAGE_HALF_ALU_INSTRUCTIONS"/> - <value value="42" name="PERF_SP_FS_STAGE_BARY_INSTRUCTIONS"/> - <value value="43" name="PERF_SP_VS_INSTRUCTIONS"/> - <value value="44" name="PERF_SP_FS_INSTRUCTIONS"/> - <value value="45" name="PERF_SP_ADDR_LOCK_COUNT"/> - <value value="46" name="PERF_SP_UCHE_READ_TRANS"/> - <value value="47" name="PERF_SP_UCHE_WRITE_TRANS"/> - <value value="48" name="PERF_SP_EXPORT_VPC_TRANS"/> - <value value="49" name="PERF_SP_EXPORT_RB_TRANS"/> - <value value="50" name="PERF_SP_PIXELS_KILLED"/> - <value value="51" name="PERF_SP_ICL1_REQUESTS"/> - <value value="52" name="PERF_SP_ICL1_MISSES"/> - <value value="53" name="PERF_SP_HS_INSTRUCTIONS"/> - <value value="54" name="PERF_SP_DS_INSTRUCTIONS"/> - <value value="55" name="PERF_SP_GS_INSTRUCTIONS"/> - <value value="56" name="PERF_SP_CS_INSTRUCTIONS"/> - <value value="57" name="PERF_SP_GPR_READ"/> - <value value="58" name="PERF_SP_GPR_WRITE"/> - <value value="59" name="PERF_SP_FS_STAGE_HALF_EFU_INSTRUCTIONS"/> - <value value="60" name="PERF_SP_VS_STAGE_HALF_EFU_INSTRUCTIONS"/> - <value value="61" name="PERF_SP_LM_BANK_CONFLICTS"/> - <value value="62" name="PERF_SP_TEX_CONTROL_WORKING_CYCLES"/> - <value value="63" name="PERF_SP_LOAD_CONTROL_WORKING_CYCLES"/> - <value value="64" name="PERF_SP_FLOW_CONTROL_WORKING_CYCLES"/> - <value value="65" name="PERF_SP_LM_WORKING_CYCLES"/> - <value value="66" name="PERF_SP_DISPATCHER_WORKING_CYCLES"/> - <value value="67" name="PERF_SP_SEQUENCER_WORKING_CYCLES"/> - <value value="68" name="PERF_SP_LOW_EFFICIENCY_STARVED_BY_TP"/> - <value value="69" name="PERF_SP_STARVE_CYCLES_HLSQ"/> - <value value="70" name="PERF_SP_NON_EXECUTION_LS_CYCLES"/> - <value value="71" name="PERF_SP_WORKING_EU"/> - <value value="72" name="PERF_SP_ANY_EU_WORKING"/> - <value value="73" name="PERF_SP_WORKING_EU_FS_STAGE"/> - <value value="74" name="PERF_SP_ANY_EU_WORKING_FS_STAGE"/> - <value value="75" name="PERF_SP_WORKING_EU_VS_STAGE"/> - <value value="76" name="PERF_SP_ANY_EU_WORKING_VS_STAGE"/> - <value value="77" name="PERF_SP_WORKING_EU_CS_STAGE"/> - <value value="78" name="PERF_SP_ANY_EU_WORKING_CS_STAGE"/> - <value value="79" name="PERF_SP_GPR_READ_PREFETCH"/> - <value value="80" name="PERF_SP_GPR_READ_CONFLICT"/> - <value value="81" name="PERF_SP_GPR_WRITE_CONFLICT"/> - <value value="82" name="PERF_SP_GM_LOAD_LATENCY_CYCLES"/> - <value value="83" name="PERF_SP_GM_LOAD_LATENCY_SAMPLES"/> - <value value="84" name="PERF_SP_EXECUTABLE_WAVES"/> -</enum> - -<enum name="a6xx_rb_perfcounter_select"> - <value value="0" name="PERF_RB_BUSY_CYCLES"/> - <value value="1" name="PERF_RB_STALL_CYCLES_HLSQ"/> - <value value="2" name="PERF_RB_STALL_CYCLES_FIFO0_FULL"/> - <value value="3" name="PERF_RB_STALL_CYCLES_FIFO1_FULL"/> - <value value="4" name="PERF_RB_STALL_CYCLES_FIFO2_FULL"/> - <value value="5" name="PERF_RB_STARVE_CYCLES_SP"/> - <value value="6" name="PERF_RB_STARVE_CYCLES_LRZ_TILE"/> - <value value="7" name="PERF_RB_STARVE_CYCLES_CCU"/> - <value value="8" name="PERF_RB_STARVE_CYCLES_Z_PLANE"/> - <value value="9" name="PERF_RB_STARVE_CYCLES_BARY_PLANE"/> - <value value="10" name="PERF_RB_Z_WORKLOAD"/> - <value value="11" name="PERF_RB_HLSQ_ACTIVE"/> - <value value="12" name="PERF_RB_Z_READ"/> - <value value="13" name="PERF_RB_Z_WRITE"/> - <value value="14" name="PERF_RB_C_READ"/> - <value value="15" name="PERF_RB_C_WRITE"/> - <value value="16" name="PERF_RB_TOTAL_PASS"/> - <value value="17" name="PERF_RB_Z_PASS"/> - <value value="18" name="PERF_RB_Z_FAIL"/> - <value value="19" name="PERF_RB_S_FAIL"/> - <value value="20" name="PERF_RB_BLENDED_FXP_COMPONENTS"/> - <value value="21" name="PERF_RB_BLENDED_FP16_COMPONENTS"/> - <value value="22" name="PERF_RB_PS_INVOCATIONS"/> - <value value="23" name="PERF_RB_2D_ALIVE_CYCLES"/> - <value value="24" name="PERF_RB_2D_STALL_CYCLES_A2D"/> - <value value="25" name="PERF_RB_2D_STARVE_CYCLES_SRC"/> - <value value="26" name="PERF_RB_2D_STARVE_CYCLES_SP"/> - <value value="27" name="PERF_RB_2D_STARVE_CYCLES_DST"/> - <value value="28" name="PERF_RB_2D_VALID_PIXELS"/> - <value value="29" name="PERF_RB_3D_PIXELS"/> - <value value="30" name="PERF_RB_BLENDER_WORKING_CYCLES"/> - <value value="31" name="PERF_RB_ZPROC_WORKING_CYCLES"/> - <value value="32" name="PERF_RB_CPROC_WORKING_CYCLES"/> - <value value="33" name="PERF_RB_SAMPLER_WORKING_CYCLES"/> - <value value="34" name="PERF_RB_STALL_CYCLES_CCU_COLOR_READ"/> - <value value="35" name="PERF_RB_STALL_CYCLES_CCU_COLOR_WRITE"/> - <value value="36" name="PERF_RB_STALL_CYCLES_CCU_DEPTH_READ"/> - <value value="37" name="PERF_RB_STALL_CYCLES_CCU_DEPTH_WRITE"/> - <value value="38" name="PERF_RB_STALL_CYCLES_VPC"/> - <value value="39" name="PERF_RB_2D_INPUT_TRANS"/> - <value value="40" name="PERF_RB_2D_OUTPUT_RB_DST_TRANS"/> - <value value="41" name="PERF_RB_2D_OUTPUT_RB_SRC_TRANS"/> - <value value="42" name="PERF_RB_BLENDED_FP32_COMPONENTS"/> - <value value="43" name="PERF_RB_COLOR_PIX_TILES"/> - <value value="44" name="PERF_RB_STALL_CYCLES_CCU"/> - <value value="45" name="PERF_RB_EARLY_Z_ARB3_GRANT"/> - <value value="46" name="PERF_RB_LATE_Z_ARB3_GRANT"/> - <value value="47" name="PERF_RB_EARLY_Z_SKIP_GRANT"/> -</enum> - -<enum name="a6xx_vsc_perfcounter_select"> - <value value="0" name="PERF_VSC_BUSY_CYCLES"/> - <value value="1" name="PERF_VSC_WORKING_CYCLES"/> - <value value="2" name="PERF_VSC_STALL_CYCLES_UCHE"/> - <value value="3" name="PERF_VSC_EOT_NUM"/> - <value value="4" name="PERF_VSC_INPUT_TILES"/> -</enum> - -<enum name="a6xx_ccu_perfcounter_select"> - <value value="0" name="PERF_CCU_BUSY_CYCLES"/> - <value value="1" name="PERF_CCU_STALL_CYCLES_RB_DEPTH_RETURN"/> - <value value="2" name="PERF_CCU_STALL_CYCLES_RB_COLOR_RETURN"/> - <value value="3" name="PERF_CCU_STARVE_CYCLES_FLAG_RETURN"/> - <value value="4" name="PERF_CCU_DEPTH_BLOCKS"/> - <value value="5" name="PERF_CCU_COLOR_BLOCKS"/> - <value value="6" name="PERF_CCU_DEPTH_BLOCK_HIT"/> - <value value="7" name="PERF_CCU_COLOR_BLOCK_HIT"/> - <value value="8" name="PERF_CCU_PARTIAL_BLOCK_READ"/> - <value value="9" name="PERF_CCU_GMEM_READ"/> - <value value="10" name="PERF_CCU_GMEM_WRITE"/> - <value value="11" name="PERF_CCU_DEPTH_READ_FLAG0_COUNT"/> - <value value="12" name="PERF_CCU_DEPTH_READ_FLAG1_COUNT"/> - <value value="13" name="PERF_CCU_DEPTH_READ_FLAG2_COUNT"/> - <value value="14" name="PERF_CCU_DEPTH_READ_FLAG3_COUNT"/> - <value value="15" name="PERF_CCU_DEPTH_READ_FLAG4_COUNT"/> - <value value="16" name="PERF_CCU_DEPTH_READ_FLAG5_COUNT"/> - <value value="17" name="PERF_CCU_DEPTH_READ_FLAG6_COUNT"/> - <value value="18" name="PERF_CCU_DEPTH_READ_FLAG8_COUNT"/> - <value value="19" name="PERF_CCU_COLOR_READ_FLAG0_COUNT"/> - <value value="20" name="PERF_CCU_COLOR_READ_FLAG1_COUNT"/> - <value value="21" name="PERF_CCU_COLOR_READ_FLAG2_COUNT"/> - <value value="22" name="PERF_CCU_COLOR_READ_FLAG3_COUNT"/> - <value value="23" name="PERF_CCU_COLOR_READ_FLAG4_COUNT"/> - <value value="24" name="PERF_CCU_COLOR_READ_FLAG5_COUNT"/> - <value value="25" name="PERF_CCU_COLOR_READ_FLAG6_COUNT"/> - <value value="26" name="PERF_CCU_COLOR_READ_FLAG8_COUNT"/> - <value value="27" name="PERF_CCU_2D_RD_REQ"/> - <value value="28" name="PERF_CCU_2D_WR_REQ"/> -</enum> - -<enum name="a6xx_lrz_perfcounter_select"> - <value value="0" name="PERF_LRZ_BUSY_CYCLES"/> - <value value="1" name="PERF_LRZ_STARVE_CYCLES_RAS"/> - <value value="2" name="PERF_LRZ_STALL_CYCLES_RB"/> - <value value="3" name="PERF_LRZ_STALL_CYCLES_VSC"/> - <value value="4" name="PERF_LRZ_STALL_CYCLES_VPC"/> - <value value="5" name="PERF_LRZ_STALL_CYCLES_FLAG_PREFETCH"/> - <value value="6" name="PERF_LRZ_STALL_CYCLES_UCHE"/> - <value value="7" name="PERF_LRZ_LRZ_READ"/> - <value value="8" name="PERF_LRZ_LRZ_WRITE"/> - <value value="9" name="PERF_LRZ_READ_LATENCY"/> - <value value="10" name="PERF_LRZ_MERGE_CACHE_UPDATING"/> - <value value="11" name="PERF_LRZ_PRIM_KILLED_BY_MASKGEN"/> - <value value="12" name="PERF_LRZ_PRIM_KILLED_BY_LRZ"/> - <value value="13" name="PERF_LRZ_VISIBLE_PRIM_AFTER_LRZ"/> - <value value="14" name="PERF_LRZ_FULL_8X8_TILES"/> - <value value="15" name="PERF_LRZ_PARTIAL_8X8_TILES"/> - <value value="16" name="PERF_LRZ_TILE_KILLED"/> - <value value="17" name="PERF_LRZ_TOTAL_PIXEL"/> - <value value="18" name="PERF_LRZ_VISIBLE_PIXEL_AFTER_LRZ"/> - <value value="19" name="PERF_LRZ_FULLY_COVERED_TILES"/> - <value value="20" name="PERF_LRZ_PARTIAL_COVERED_TILES"/> - <value value="21" name="PERF_LRZ_FEEDBACK_ACCEPT"/> - <value value="22" name="PERF_LRZ_FEEDBACK_DISCARD"/> - <value value="23" name="PERF_LRZ_FEEDBACK_STALL"/> - <value value="24" name="PERF_LRZ_STALL_CYCLES_RB_ZPLANE"/> - <value value="25" name="PERF_LRZ_STALL_CYCLES_RB_BPLANE"/> - <value value="26" name="PERF_LRZ_STALL_CYCLES_VC"/> - <value value="27" name="PERF_LRZ_RAS_MASK_TRANS"/> -</enum> - -<enum name="a6xx_cmp_perfcounter_select"> - <value value="0" name="PERF_CMPDECMP_STALL_CYCLES_ARB"/> - <value value="1" name="PERF_CMPDECMP_VBIF_LATENCY_CYCLES"/> - <value value="2" name="PERF_CMPDECMP_VBIF_LATENCY_SAMPLES"/> - <value value="3" name="PERF_CMPDECMP_VBIF_READ_DATA_CCU"/> - <value value="4" name="PERF_CMPDECMP_VBIF_WRITE_DATA_CCU"/> - <value value="5" name="PERF_CMPDECMP_VBIF_READ_REQUEST"/> - <value value="6" name="PERF_CMPDECMP_VBIF_WRITE_REQUEST"/> - <value value="7" name="PERF_CMPDECMP_VBIF_READ_DATA"/> - <value value="8" name="PERF_CMPDECMP_VBIF_WRITE_DATA"/> - <value value="9" name="PERF_CMPDECMP_FLAG_FETCH_CYCLES"/> - <value value="10" name="PERF_CMPDECMP_FLAG_FETCH_SAMPLES"/> - <value value="11" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG1_COUNT"/> - <value value="12" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG2_COUNT"/> - <value value="13" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG3_COUNT"/> - <value value="14" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG4_COUNT"/> - <value value="15" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG5_COUNT"/> - <value value="16" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG6_COUNT"/> - <value value="17" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG8_COUNT"/> - <value value="18" name="PERF_CMPDECMP_COLOR_WRITE_FLAG1_COUNT"/> - <value value="19" name="PERF_CMPDECMP_COLOR_WRITE_FLAG2_COUNT"/> - <value value="20" name="PERF_CMPDECMP_COLOR_WRITE_FLAG3_COUNT"/> - <value value="21" name="PERF_CMPDECMP_COLOR_WRITE_FLAG4_COUNT"/> - <value value="22" name="PERF_CMPDECMP_COLOR_WRITE_FLAG5_COUNT"/> - <value value="23" name="PERF_CMPDECMP_COLOR_WRITE_FLAG6_COUNT"/> - <value value="24" name="PERF_CMPDECMP_COLOR_WRITE_FLAG8_COUNT"/> - <value value="25" name="PERF_CMPDECMP_2D_STALL_CYCLES_VBIF_REQ"/> - <value value="26" name="PERF_CMPDECMP_2D_STALL_CYCLES_VBIF_WR"/> - <value value="27" name="PERF_CMPDECMP_2D_STALL_CYCLES_VBIF_RETURN"/> - <value value="28" name="PERF_CMPDECMP_2D_RD_DATA"/> - <value value="29" name="PERF_CMPDECMP_2D_WR_DATA"/> - <value value="30" name="PERF_CMPDECMP_VBIF_READ_DATA_UCHE_CH0"/> - <value value="31" name="PERF_CMPDECMP_VBIF_READ_DATA_UCHE_CH1"/> - <value value="32" name="PERF_CMPDECMP_2D_OUTPUT_TRANS"/> - <value value="33" name="PERF_CMPDECMP_VBIF_WRITE_DATA_UCHE"/> - <value value="34" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG0_COUNT"/> - <value value="35" name="PERF_CMPDECMP_COLOR_WRITE_FLAG0_COUNT"/> - <value value="36" name="PERF_CMPDECMP_COLOR_WRITE_FLAGALPHA_COUNT"/> - <value value="37" name="PERF_CMPDECMP_2D_BUSY_CYCLES"/> - <value value="38" name="PERF_CMPDECMP_2D_REORDER_STARVE_CYCLES"/> - <value value="39" name="PERF_CMPDECMP_2D_PIXELS"/> -</enum> - -<!-- -Used in a6xx_2d_blit_cntl.. the value mostly seems to correlate to the -component type/size, so I think it relates to internal format used for -blending? The one exception is that 16b unorm and 32b float use the -same value... maybe 16b unorm is uncommon enough that it was just easier -to upconvert to 32b float internally? - - 8b unorm: 10 (sometimes 0, is the high bit part of something else?) -16b unorm: 4 - -32b int: 7 -16b int: 6 - 8b int: 5 - -32b float: 4 -16b float: 3 - --> -<enum name="a6xx_2d_ifmt"> - <value value="0x10" name="R2D_UNORM8"/> - <value value="0x7" name="R2D_INT32"/> - <value value="0x6" name="R2D_INT16"/> - <value value="0x5" name="R2D_INT8"/> - <value value="0x4" name="R2D_FLOAT32"/> - <value value="0x3" name="R2D_FLOAT16"/> - <value value="0x1" name="R2D_UNORM8_SRGB"/> - <value value="0x0" name="R2D_RAW"/> -</enum> - -<enum name="a6xx_ztest_mode"> - <doc>Allow early z-test and early-lrz (if applicable)</doc> - <value value="0x0" name="A6XX_EARLY_Z"/> - <doc>Disable early z-test and early-lrz test (if applicable)</doc> - <value value="0x1" name="A6XX_LATE_Z"/> - <doc> - A special mode that allows early-lrz test but disables - early-z test. Which might sound a bit funny, since - lrz-test happens before z-test. But as long as a couple - conditions are maintained this allows using lrz-test in - cases where fragment shader has kill/discard: - - 1) Disable lrz-write in cases where it is uncertain during - binning pass that a fragment will pass. Ie. if frag - shader has-kill, writes-z, or alpha/stencil test is - enabled. (For correctness, lrz-write must be disabled - when blend is enabled.) This is analogous to how a - z-prepass works. - - 2) Disable lrz-write and test if a depth-test direction - reversal is detected. Due to condition (1), the contents - of the lrz buffer are a conservative estimation of the - depth buffer during the draw pass. Meaning that geometry - that we know for certain will not be visible will not pass - lrz-test. But geometry which may be (or contributes to - blend) will pass the lrz-test. - - This allows us to keep early-lrz-test in cases where the frag - shader does not write-z (ie. we know the z-value before FS) - and does not have side-effects (image/ssbo writes, etc), but - does have kill/discard. Which turns out to be a common - enough case that it is useful to keep early-lrz test against - the conservative lrz buffer to discard fragments that we - know will definitely not be visible. - </doc> - <value value="0x2" name="A6XX_EARLY_LRZ_LATE_Z"/> - <doc>Not a real hw value, used internally by mesa</doc> - <value value="0x3" name="A6XX_INVALID_ZTEST"/> -</enum> - -<enum name="a6xx_tess_spacing"> - <value value="0x0" name="TESS_EQUAL"/> - <value value="0x2" name="TESS_FRACTIONAL_ODD"/> - <value value="0x3" name="TESS_FRACTIONAL_EVEN"/> -</enum> -<enum name="a6xx_tess_output"> - <value value="0x0" name="TESS_POINTS"/> - <value value="0x1" name="TESS_LINES"/> - <value value="0x2" name="TESS_CW_TRIS"/> - <value value="0x3" name="TESS_CCW_TRIS"/> -</enum> - -<enum name="a7xx_cp_perfcounter_select"> - <value value="0" name="A7XX_PERF_CP_ALWAYS_COUNT"/> - <value value="1" name="A7XX_PERF_CP_BUSY_GFX_CORE_IDLE"/> - <value value="2" name="A7XX_PERF_CP_BUSY_CYCLES"/> - <value value="3" name="A7XX_PERF_CP_NUM_PREEMPTIONS"/> - <value value="4" name="A7XX_PERF_CP_PREEMPTION_REACTION_DELAY"/> - <value value="5" name="A7XX_PERF_CP_PREEMPTION_SWITCH_OUT_TIME"/> - <value value="6" name="A7XX_PERF_CP_PREEMPTION_SWITCH_IN_TIME"/> - <value value="7" name="A7XX_PERF_CP_DEAD_DRAWS_IN_BIN_RENDER"/> - <value value="8" name="A7XX_PERF_CP_PREDICATED_DRAWS_KILLED"/> - <value value="9" name="A7XX_PERF_CP_MODE_SWITCH"/> - <value value="10" name="A7XX_PERF_CP_ZPASS_DONE"/> - <value value="11" name="A7XX_PERF_CP_CONTEXT_DONE"/> - <value value="12" name="A7XX_PERF_CP_CACHE_FLUSH"/> - <value value="13" name="A7XX_PERF_CP_LONG_PREEMPTIONS"/> - <value value="14" name="A7XX_PERF_CP_SQE_I_CACHE_STARVE"/> - <value value="15" name="A7XX_PERF_CP_SQE_IDLE"/> - <value value="16" name="A7XX_PERF_CP_SQE_PM4_STARVE_RB_IB"/> - <value value="17" name="A7XX_PERF_CP_SQE_PM4_STARVE_SDS"/> - <value value="18" name="A7XX_PERF_CP_SQE_MRB_STARVE"/> - <value value="19" name="A7XX_PERF_CP_SQE_RRB_STARVE"/> - <value value="20" name="A7XX_PERF_CP_SQE_VSD_STARVE"/> - <value value="21" name="A7XX_PERF_CP_VSD_DECODE_STARVE"/> - <value value="22" name="A7XX_PERF_CP_SQE_PIPE_OUT_STALL"/> - <value value="23" name="A7XX_PERF_CP_SQE_SYNC_STALL"/> - <value value="24" name="A7XX_PERF_CP_SQE_PM4_WFI_STALL"/> - <value value="25" name="A7XX_PERF_CP_SQE_SYS_WFI_STALL"/> - <value value="26" name="A7XX_PERF_CP_SQE_T4_EXEC"/> - <value value="27" name="A7XX_PERF_CP_SQE_LOAD_STATE_EXEC"/> - <value value="28" name="A7XX_PERF_CP_SQE_SAVE_SDS_STATE"/> - <value value="29" name="A7XX_PERF_CP_SQE_DRAW_EXEC"/> - <value value="30" name="A7XX_PERF_CP_SQE_CTXT_REG_BUNCH_EXEC"/> - <value value="31" name="A7XX_PERF_CP_SQE_EXEC_PROFILED"/> - <value value="32" name="A7XX_PERF_CP_MEMORY_POOL_EMPTY"/> - <value value="33" name="A7XX_PERF_CP_MEMORY_POOL_SYNC_STALL"/> - <value value="34" name="A7XX_PERF_CP_MEMORY_POOL_ABOVE_THRESH"/> - <value value="35" name="A7XX_PERF_CP_AHB_WR_STALL_PRE_DRAWS"/> - <value value="36" name="A7XX_PERF_CP_AHB_STALL_SQE_GMU"/> - <value value="37" name="A7XX_PERF_CP_AHB_STALL_SQE_WR_OTHER"/> - <value value="38" name="A7XX_PERF_CP_AHB_STALL_SQE_RD_OTHER"/> - <value value="39" name="A7XX_PERF_CP_CLUSTER0_EMPTY"/> - <value value="40" name="A7XX_PERF_CP_CLUSTER1_EMPTY"/> - <value value="41" name="A7XX_PERF_CP_CLUSTER2_EMPTY"/> - <value value="42" name="A7XX_PERF_CP_CLUSTER3_EMPTY"/> - <value value="43" name="A7XX_PERF_CP_CLUSTER4_EMPTY"/> - <value value="44" name="A7XX_PERF_CP_CLUSTER5_EMPTY"/> - <value value="45" name="A7XX_PERF_CP_PM4_DATA"/> - <value value="46" name="A7XX_PERF_CP_PM4_HEADERS"/> - <value value="47" name="A7XX_PERF_CP_VBIF_READ_BEATS"/> - <value value="48" name="A7XX_PERF_CP_VBIF_WRITE_BEATS"/> - <value value="49" name="A7XX_PERF_CP_SQE_INSTR_COUNTER"/> - <value value="50" name="A7XX_PERF_CP_RESERVED_50"/> - <value value="51" name="A7XX_PERF_CP_RESERVED_51"/> - <value value="52" name="A7XX_PERF_CP_RESERVED_52"/> - <value value="53" name="A7XX_PERF_CP_RESERVED_53"/> - <value value="54" name="A7XX_PERF_CP_RESERVED_54"/> - <value value="55" name="A7XX_PERF_CP_RESERVED_55"/> - <value value="56" name="A7XX_PERF_CP_RESERVED_56"/> - <value value="57" name="A7XX_PERF_CP_RESERVED_57"/> - <value value="58" name="A7XX_PERF_CP_RESERVED_58"/> - <value value="59" name="A7XX_PERF_CP_RESERVED_59"/> - <value value="60" name="A7XX_PERF_CP_CLUSTER0_FULL"/> - <value value="61" name="A7XX_PERF_CP_CLUSTER1_FULL"/> - <value value="62" name="A7XX_PERF_CP_CLUSTER2_FULL"/> - <value value="63" name="A7XX_PERF_CP_CLUSTER3_FULL"/> - <value value="64" name="A7XX_PERF_CP_CLUSTER4_FULL"/> - <value value="65" name="A7XX_PERF_CP_CLUSTER5_FULL"/> - <value value="66" name="A7XX_PERF_CP_CLUSTER6_FULL"/> - <value value="67" name="A7XX_PERF_CP_CLUSTER6_EMPTY"/> - <value value="68" name="A7XX_PERF_CP_ICACHE_MISSES"/> - <value value="69" name="A7XX_PERF_CP_ICACHE_HITS"/> - <value value="70" name="A7XX_PERF_CP_ICACHE_STALL"/> - <value value="71" name="A7XX_PERF_CP_DCACHE_MISSES"/> - <value value="72" name="A7XX_PERF_CP_DCACHE_HITS"/> - <value value="73" name="A7XX_PERF_CP_DCACHE_STALLS"/> - <value value="74" name="A7XX_PERF_CP_AQE_SQE_STALL"/> - <value value="75" name="A7XX_PERF_CP_SQE_AQE_STARVE"/> - <value value="76" name="A7XX_PERF_CP_PREEMPT_LATENCY"/> - <value value="77" name="A7XX_PERF_CP_SQE_MD8_STALL_CYCLES"/> - <value value="78" name="A7XX_PERF_CP_SQE_MESH_EXEC_CYCLES"/> - <value value="79" name="A7XX_PERF_CP_AQE_NUM_AS_CHUNKS"/> - <value value="80" name="A7XX_PERF_CP_AQE_NUM_MS_CHUNKS"/> -</enum> - -<enum name="a7xx_rbbm_perfcounter_select"> - <value value="0" name="A7XX_PERF_RBBM_ALWAYS_COUNT"/> - <value value="1" name="A7XX_PERF_RBBM_ALWAYS_ON"/> - <value value="2" name="A7XX_PERF_RBBM_TSE_BUSY"/> - <value value="3" name="A7XX_PERF_RBBM_RAS_BUSY"/> - <value value="4" name="A7XX_PERF_RBBM_PC_DCALL_BUSY"/> - <value value="5" name="A7XX_PERF_RBBM_PC_VSD_BUSY"/> - <value value="6" name="A7XX_PERF_RBBM_STATUS_MASKED"/> - <value value="7" name="A7XX_PERF_RBBM_COM_BUSY"/> - <value value="8" name="A7XX_PERF_RBBM_DCOM_BUSY"/> - <value value="9" name="A7XX_PERF_RBBM_VBIF_BUSY"/> - <value value="10" name="A7XX_PERF_RBBM_VSC_BUSY"/> - <value value="11" name="A7XX_PERF_RBBM_TESS_BUSY"/> - <value value="12" name="A7XX_PERF_RBBM_UCHE_BUSY"/> - <value value="13" name="A7XX_PERF_RBBM_HLSQ_BUSY"/> -</enum> - -<enum name="a7xx_pc_perfcounter_select"> - <value value="0" name="A7XX_PERF_PC_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_PC_WORKING_CYCLES"/> - <value value="2" name="A7XX_PERF_PC_STALL_CYCLES_VFD"/> - <value value="3" name="A7XX_PERF_PC_RESERVED"/> - <value value="4" name="A7XX_PERF_PC_STALL_CYCLES_VPC"/> - <value value="5" name="A7XX_PERF_PC_STALL_CYCLES_UCHE"/> - <value value="6" name="A7XX_PERF_PC_STALL_CYCLES_TESS"/> - <value value="7" name="A7XX_PERF_PC_STALL_CYCLES_VFD_ONLY"/> - <value value="8" name="A7XX_PERF_PC_STALL_CYCLES_VPC_ONLY"/> - <value value="9" name="A7XX_PERF_PC_PASS1_TF_STALL_CYCLES"/> - <value value="10" name="A7XX_PERF_PC_STARVE_CYCLES_FOR_INDEX"/> - <value value="11" name="A7XX_PERF_PC_STARVE_CYCLES_FOR_TESS_FACTOR"/> - <value value="12" name="A7XX_PERF_PC_STARVE_CYCLES_FOR_VIZ_STREAM"/> - <value value="13" name="A7XX_PERF_PC_STARVE_CYCLES_DI"/> - <value value="14" name="A7XX_PERF_PC_VIS_STREAMS_LOADED"/> - <value value="15" name="A7XX_PERF_PC_INSTANCES"/> - <value value="16" name="A7XX_PERF_PC_VPC_PRIMITIVES"/> - <value value="17" name="A7XX_PERF_PC_DEAD_PRIM"/> - <value value="18" name="A7XX_PERF_PC_LIVE_PRIM"/> - <value value="19" name="A7XX_PERF_PC_VERTEX_HITS"/> - <value value="20" name="A7XX_PERF_PC_IA_VERTICES"/> - <value value="21" name="A7XX_PERF_PC_IA_PRIMITIVES"/> - <value value="22" name="A7XX_PERF_PC_RESERVED_22"/> - <value value="23" name="A7XX_PERF_PC_HS_INVOCATIONS"/> - <value value="24" name="A7XX_PERF_PC_DS_INVOCATIONS"/> - <value value="25" name="A7XX_PERF_PC_VS_INVOCATIONS"/> - <value value="26" name="A7XX_PERF_PC_GS_INVOCATIONS"/> - <value value="27" name="A7XX_PERF_PC_DS_PRIMITIVES"/> - <value value="28" name="A7XX_PERF_PC_3D_DRAWCALLS"/> - <value value="29" name="A7XX_PERF_PC_2D_DRAWCALLS"/> - <value value="30" name="A7XX_PERF_PC_NON_DRAWCALL_GLOBAL_EVENTS"/> - <value value="31" name="A7XX_PERF_PC_TESS_BUSY_CYCLES"/> - <value value="32" name="A7XX_PERF_PC_TESS_WORKING_CYCLES"/> - <value value="33" name="A7XX_PERF_PC_TESS_STALL_CYCLES_PC"/> - <value value="34" name="A7XX_PERF_PC_TESS_STARVE_CYCLES_PC"/> - <value value="35" name="A7XX_PERF_PC_TESS_SINGLE_PRIM_CYCLES"/> - <value value="36" name="A7XX_PERF_PC_TESS_PC_UV_TRANS"/> - <value value="37" name="A7XX_PERF_PC_TESS_PC_UV_PATCHES"/> - <value value="38" name="A7XX_PERF_PC_TESS_FACTOR_TRANS"/> - <value value="39" name="A7XX_PERF_PC_TAG_CHECKED_VERTICES"/> - <value value="40" name="A7XX_PERF_PC_MESH_VS_WAVES"/> - <value value="41" name="A7XX_PERF_PC_MESH_DRAWS"/> - <value value="42" name="A7XX_PERF_PC_MESH_DEAD_DRAWS"/> - <value value="43" name="A7XX_PERF_PC_MESH_MVIS_EN_DRAWS"/> - <value value="44" name="A7XX_PERF_PC_MESH_DEAD_PRIM"/> - <value value="45" name="A7XX_PERF_PC_MESH_LIVE_PRIM"/> - <value value="46" name="A7XX_PERF_PC_MESH_PA_EN_PRIM"/> - <value value="47" name="A7XX_PERF_PC_STARVE_CYCLES_FOR_MVIS_STREAM"/> - <value value="48" name="A7XX_PERF_PC_STARVE_CYCLES_PREDRAW"/> - <value value="49" name="A7XX_PERF_PC_STALL_CYCLES_COMPUTE_GFX"/> - <value value="50" name="A7XX_PERF_PC_STALL_CYCLES_GFX_COMPUTE"/> - <value value="51" name="A7XX_PERF_PC_TESS_PC_MULTI_PATCH_TRANS"/> -</enum> - -<enum name="a7xx_vfd_perfcounter_select"> - <value value="0" name="A7XX_PERF_VFD_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_VFD_STALL_CYCLES_UCHE"/> - <value value="2" name="A7XX_PERF_VFD_STALL_CYCLES_VPC_ALLOC"/> - <value value="3" name="A7XX_PERF_VFD_STALL_CYCLES_SP_INFO"/> - <value value="4" name="A7XX_PERF_VFD_STALL_CYCLES_SP_ATTR"/> - <value value="5" name="A7XX_PERF_VFD_STARVE_CYCLES_UCHE"/> - <value value="6" name="A7XX_PERF_VFD_RBUFFER_FULL"/> - <value value="7" name="A7XX_PERF_VFD_ATTR_INFO_FIFO_FULL"/> - <value value="8" name="A7XX_PERF_VFD_DECODED_ATTRIBUTE_BYTES"/> - <value value="9" name="A7XX_PERF_VFD_NUM_ATTRIBUTES"/> - <value value="10" name="A7XX_PERF_VFD_UPPER_SHADER_FIBERS"/> - <value value="11" name="A7XX_PERF_VFD_LOWER_SHADER_FIBERS"/> - <value value="12" name="A7XX_PERF_VFD_MODE_0_FIBERS"/> - <value value="13" name="A7XX_PERF_VFD_MODE_1_FIBERS"/> - <value value="14" name="A7XX_PERF_VFD_MODE_2_FIBERS"/> - <value value="15" name="A7XX_PERF_VFD_MODE_3_FIBERS"/> - <value value="16" name="A7XX_PERF_VFD_MODE_4_FIBERS"/> - <value value="17" name="A7XX_PERF_VFD_TOTAL_VERTICES"/> - <value value="18" name="A7XX_PERF_VFDP_STALL_CYCLES_VFD"/> - <value value="19" name="A7XX_PERF_VFDP_STALL_CYCLES_VFD_INDEX"/> - <value value="20" name="A7XX_PERF_VFDP_STALL_CYCLES_VFD_PROG"/> - <value value="21" name="A7XX_PERF_VFDP_STARVE_CYCLES_PC"/> - <value value="22" name="A7XX_PERF_VFDP_VS_STAGE_WAVES"/> - <value value="23" name="A7XX_PERF_VFD_STALL_CYCLES_PRG_END_FE"/> - <value value="24" name="A7XX_PERF_VFD_STALL_CYCLES_CBSYNC"/> -</enum> - -<enum name="a7xx_hlsq_perfcounter_select"> - <value value="0" name="A7XX_PERF_HLSQ_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_HLSQ_STALL_CYCLES_UCHE"/> - <value value="2" name="A7XX_PERF_HLSQ_STALL_CYCLES_SP_STATE"/> - <value value="3" name="A7XX_PERF_HLSQ_STALL_CYCLES_SP_FS_STAGE"/> - <value value="4" name="A7XX_PERF_HLSQ_UCHE_LATENCY_CYCLES"/> - <value value="5" name="A7XX_PERF_HLSQ_UCHE_LATENCY_COUNT"/> - <value value="6" name="A7XX_PERF_HLSQ_RESERVED_6"/> - <value value="7" name="A7XX_PERF_HLSQ_RESERVED_7"/> - <value value="8" name="A7XX_PERF_HLSQ_RESERVED_8"/> - <value value="9" name="A7XX_PERF_HLSQ_RESERVED_9"/> - <value value="10" name="A7XX_PERF_HLSQ_COMPUTE_DRAWCALLS"/> - <value value="11" name="A7XX_PERF_HLSQ_FS_DATA_WAIT_PROGRAMMING"/> - <value value="12" name="A7XX_PERF_HLSQ_DUAL_FS_PROG_ACTIVE"/> - <value value="13" name="A7XX_PERF_HLSQ_DUAL_VS_PROG_ACTIVE"/> - <value value="14" name="A7XX_PERF_HLSQ_FS_BATCH_COUNT_ZERO"/> - <value value="15" name="A7XX_PERF_HLSQ_VS_BATCH_COUNT_ZERO"/> - <value value="16" name="A7XX_PERF_HLSQ_WAVE_PENDING_NO_QUAD"/> - <value value="17" name="A7XX_PERF_HLSQ_WAVE_PENDING_NO_PRIM_BASE"/> - <value value="18" name="A7XX_PERF_HLSQ_STALL_CYCLES_VPC"/> - <value value="19" name="A7XX_PERF_HLSQ_RESERVED_19"/> - <value value="20" name="A7XX_PERF_HLSQ_DRAW_MODE_SWITCH_VSFS_SYNC"/> - <value value="21" name="A7XX_PERF_HLSQ_VSBR_STALL_CYCLES"/> - <value value="22" name="A7XX_PERF_HLSQ_FS_STALL_CYCLES"/> - <value value="23" name="A7XX_PERF_HLSQ_LPAC_STALL_CYCLES"/> - <value value="24" name="A7XX_PERF_HLSQ_BV_STALL_CYCLES"/> - <value value="25" name="A7XX_PERF_HLSQ_VSBR_DEREF_CYCLES"/> - <value value="26" name="A7XX_PERF_HLSQ_FS_DEREF_CYCLES"/> - <value value="27" name="A7XX_PERF_HLSQ_LPAC_DEREF_CYCLES"/> - <value value="28" name="A7XX_PERF_HLSQ_BV_DEREF_CYCLES"/> - <value value="29" name="A7XX_PERF_HLSQ_VSBR_S2W_CYCLES"/> - <value value="30" name="A7XX_PERF_HLSQ_FS_S2W_CYCLES"/> - <value value="31" name="A7XX_PERF_HLSQ_LPAC_S2W_CYCLES"/> - <value value="32" name="A7XX_PERF_HLSQ_BV_S2W_CYCLES"/> - <value value="33" name="A7XX_PERF_HLSQ_VSBR_WAIT_FS_S2W"/> - <value value="34" name="A7XX_PERF_HLSQ_FS_WAIT_VS_S2W"/> - <value value="35" name="A7XX_PERF_HLSQ_LPAC_WAIT_VS_S2W"/> - <value value="36" name="A7XX_PERF_HLSQ_BV_WAIT_FS_S2W"/> - <value value="37" name="A7XX_PERF_HLSQ_VS_WAIT_CONST_RESOURCE"/> - <value value="38" name="A7XX_PERF_HLSQ_FS_WAIT_SAME_VS_S2W"/> - <value value="39" name="A7XX_PERF_HLSQ_FS_STARVING_SP"/> - <value value="40" name="A7XX_PERF_HLSQ_VS_DATA_WAIT_PROGRAMMING"/> - <value value="41" name="A7XX_PERF_HLSQ_BV_DATA_WAIT_PROGRAMMING"/> - <value value="42" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXTS_VS"/> - <value value="43" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXT_CYCLES_VS"/> - <value value="44" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXTS_FS"/> - <value value="45" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXT_CYCLES_FS"/> - <value value="46" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXTS_BV"/> - <value value="47" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXT_CYCLES_BV"/> - <value value="48" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXTS_LPAC"/> - <value value="49" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXT_CYCLES_LPAC"/> - <value value="50" name="A7XX_PERF_HLSQ_SPTROC_STCHE_WARMUP_INC_VS"/> - <value value="51" name="A7XX_PERF_HLSQ_SPTROC_STCHE_WARMUP_INC_FS"/> - <value value="52" name="A7XX_PERF_HLSQ_SPTROC_STCHE_WARMUP_INC_BV"/> - <value value="53" name="A7XX_PERF_HLSQ_SPTROC_STCHE_WARMUP_INC_LPAC"/> - <value value="54" name="A7XX_PERF_HLSQ_SPTROC_STCHE_MISS_INC_VS"/> - <value value="55" name="A7XX_PERF_HLSQ_SPTROC_STCHE_MISS_INC_FS"/> - <value value="56" name="A7XX_PERF_HLSQ_SPTROC_STCHE_MISS_INC_BV"/> - <value value="57" name="A7XX_PERF_HLSQ_SPTROC_STCHE_MISS_INC_LPAC"/> -</enum> - -<enum name="a7xx_vpc_perfcounter_select"> - <value value="0" name="A7XX_PERF_VPC_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_VPC_WORKING_CYCLES"/> - <value value="2" name="A7XX_PERF_VPC_STALL_CYCLES_UCHE"/> - <value value="3" name="A7XX_PERF_VPC_STALL_CYCLES_VFD_WACK"/> - <value value="4" name="A7XX_PERF_VPC_STALL_CYCLES_HLSQ_PRIM_ALLOC"/> - <value value="5" name="A7XX_PERF_VPC_RESERVED_5"/> - <value value="6" name="A7XX_PERF_VPC_STALL_CYCLES_SP_LM"/> - <value value="7" name="A7XX_PERF_VPC_STARVE_CYCLES_SP"/> - <value value="8" name="A7XX_PERF_VPC_STARVE_CYCLES_LRZ"/> - <value value="9" name="A7XX_PERF_VPC_PC_PRIMITIVES"/> - <value value="10" name="A7XX_PERF_VPC_SP_COMPONENTS"/> - <value value="11" name="A7XX_PERF_VPC_STALL_CYCLES_VPCRAM_POS"/> - <value value="12" name="A7XX_PERF_VPC_LRZ_ASSIGN_PRIMITIVES"/> - <value value="13" name="A7XX_PERF_VPC_RB_VISIBLE_PRIMITIVES"/> - <value value="14" name="A7XX_PERF_VPC_LM_TRANSACTION"/> - <value value="15" name="A7XX_PERF_VPC_STREAMOUT_TRANSACTION"/> - <value value="16" name="A7XX_PERF_VPC_VS_BUSY_CYCLES"/> - <value value="17" name="A7XX_PERF_VPC_PS_BUSY_CYCLES"/> - <value value="18" name="A7XX_PERF_VPC_VS_WORKING_CYCLES"/> - <value value="19" name="A7XX_PERF_VPC_PS_WORKING_CYCLES"/> - <value value="20" name="A7XX_PERF_VPC_STARVE_CYCLES_RB"/> - <value value="21" name="A7XX_PERF_VPC_NUM_VPCRAM_READ_POS"/> - <value value="22" name="A7XX_PERF_VPC_WIT_FULL_CYCLES"/> - <value value="23" name="A7XX_PERF_VPC_VPCRAM_FULL_CYCLES"/> - <value value="24" name="A7XX_PERF_VPC_LM_FULL_WAIT_FOR_INTP_END"/> - <value value="25" name="A7XX_PERF_VPC_NUM_VPCRAM_WRITE"/> - <value value="26" name="A7XX_PERF_VPC_NUM_VPCRAM_READ_SO"/> - <value value="27" name="A7XX_PERF_VPC_NUM_ATTR_REQ_LM"/> - <value value="28" name="A7XX_PERF_VPC_STALL_CYCLE_TSE"/> - <value value="29" name="A7XX_PERF_VPC_TSE_PRIMITIVES"/> - <value value="30" name="A7XX_PERF_VPC_GS_PRIMITIVES"/> - <value value="31" name="A7XX_PERF_VPC_TSE_TRANSACTIONS"/> - <value value="32" name="A7XX_PERF_VPC_STALL_CYCLES_CCU"/> - <value value="33" name="A7XX_PERF_VPC_NUM_WM_HIT"/> - <value value="34" name="A7XX_PERF_VPC_STALL_DQ_WACK"/> - <value value="35" name="A7XX_PERF_VPC_STALL_CYCLES_CCHE"/> - <value value="36" name="A7XX_PERF_VPC_STARVE_CYCLES_CCHE"/> - <value value="37" name="A7XX_PERF_VPC_NUM_PA_REQ"/> - <value value="38" name="A7XX_PERF_VPC_NUM_LM_REQ_HIT"/> - <value value="39" name="A7XX_PERF_VPC_CCHE_REQBUF_FULL"/> - <value value="40" name="A7XX_PERF_VPC_STALL_CYCLES_LM_ACK"/> - <value value="41" name="A7XX_PERF_VPC_STALL_CYCLES_PRG_END_FE"/> - <value value="42" name="A7XX_PERF_VPC_STALL_CYCLES_PRG_END_PCVS"/> - <value value="43" name="A7XX_PERF_VPC_STALL_CYCLES_PRG_END_VPCPS"/> -</enum> - -<enum name="a7xx_tse_perfcounter_select"> - <value value="0" name="A7XX_PERF_TSE_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_TSE_CLIPPING_CYCLES"/> - <value value="2" name="A7XX_PERF_TSE_STALL_CYCLES_RAS"/> - <value value="3" name="A7XX_PERF_TSE_STALL_CYCLES_LRZ_BARYPLANE"/> - <value value="4" name="A7XX_PERF_TSE_STALL_CYCLES_LRZ_ZPLANE"/> - <value value="5" name="A7XX_PERF_TSE_STARVE_CYCLES_PC"/> - <value value="6" name="A7XX_PERF_TSE_INPUT_PRIM"/> - <value value="7" name="A7XX_PERF_TSE_INPUT_NULL_PRIM"/> - <value value="8" name="A7XX_PERF_TSE_TRIVAL_REJ_PRIM"/> - <value value="9" name="A7XX_PERF_TSE_CLIPPED_PRIM"/> - <value value="10" name="A7XX_PERF_TSE_ZERO_AREA_PRIM"/> - <value value="11" name="A7XX_PERF_TSE_FACENESS_CULLED_PRIM"/> - <value value="12" name="A7XX_PERF_TSE_ZERO_PIXEL_PRIM"/> - <value value="13" name="A7XX_PERF_TSE_OUTPUT_NULL_PRIM"/> - <value value="14" name="A7XX_PERF_TSE_OUTPUT_VISIBLE_PRIM"/> - <value value="15" name="A7XX_PERF_TSE_CINVOCATION"/> - <value value="16" name="A7XX_PERF_TSE_CPRIMITIVES"/> - <value value="17" name="A7XX_PERF_TSE_2D_INPUT_PRIM"/> - <value value="18" name="A7XX_PERF_TSE_2D_ALIVE_CYCLES"/> - <value value="19" name="A7XX_PERF_TSE_CLIP_PLANES"/> -</enum> - -<enum name="a7xx_ras_perfcounter_select"> - <value value="0" name="A7XX_PERF_RAS_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_RAS_SUPERTILE_ACTIVE_CYCLES"/> - <value value="2" name="A7XX_PERF_RAS_STALL_CYCLES_LRZ"/> - <value value="3" name="A7XX_PERF_RAS_STARVE_CYCLES_TSE"/> - <value value="4" name="A7XX_PERF_RAS_SUPER_TILES"/> - <value value="5" name="A7XX_PERF_RAS_8X4_TILES"/> - <value value="6" name="A7XX_PERF_RAS_MASKGEN_ACTIVE"/> - <value value="7" name="A7XX_PERF_RAS_FULLY_COVERED_SUPER_TILES"/> - <value value="8" name="A7XX_PERF_RAS_FULLY_COVERED_8X4_TILES"/> - <value value="9" name="A7XX_PERF_RAS_PRIM_KILLED_INVISILBE"/> - <value value="10" name="A7XX_PERF_RAS_SUPERTILE_GEN_ACTIVE_CYCLES"/> - <value value="11" name="A7XX_PERF_RAS_LRZ_INTF_WORKING_CYCLES"/> - <value value="12" name="A7XX_PERF_RAS_BLOCKS"/> - <value value="13" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_0_WORKING_CC_l2"/> - <value value="14" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_1_WORKING_CC_l2"/> - <value value="15" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_2_WORKING_CC_l2"/> - <value value="16" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_3_WORKING_CC_l2"/> - <value value="17" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_4_WORKING_CC_l2"/> - <value value="18" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_5_WORKING_CC_l2"/> - <value value="19" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_6_WORKING_CC_l2"/> - <value value="20" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_7_WORKING_CC_l2"/> - <value value="21" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_8_WORKING_CC_l2"/> - <value value="22" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_9_WORKING_CC_l2"/> - <value value="23" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_10_WORKING_CC_l2"/> - <value value="24" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_11_WORKING_CC_l2"/> - <value value="25" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_12_WORKING_CC_l2"/> - <value value="26" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_13_WORKING_CC_l2"/> - <value value="27" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_14_WORKING_CC_l2"/> - <value value="28" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_15_WORKING_CC_l2"/> - <value value="29" name="A7XX_PERF_RAS_FALSE_PARTIAL_STILE"/> - -</enum> - -<enum name="a7xx_uche_perfcounter_select"> - <value value="0" name="A7XX_PERF_UCHE_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_UCHE_STALL_CYCLES_ARBITER"/> - <value value="2" name="A7XX_PERF_UCHE_VBIF_LATENCY_CYCLES"/> - <value value="3" name="A7XX_PERF_UCHE_VBIF_LATENCY_SAMPLES"/> - <value value="4" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_TP"/> - <value value="5" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_VFD"/> - <value value="6" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_HLSQ"/> - <value value="7" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_LRZ"/> - <value value="8" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_SP"/> - <value value="9" name="A7XX_PERF_UCHE_READ_REQUESTS_TP"/> - <value value="10" name="A7XX_PERF_UCHE_READ_REQUESTS_VFD"/> - <value value="11" name="A7XX_PERF_UCHE_READ_REQUESTS_HLSQ"/> - <value value="12" name="A7XX_PERF_UCHE_READ_REQUESTS_LRZ"/> - <value value="13" name="A7XX_PERF_UCHE_READ_REQUESTS_SP"/> - <value value="14" name="A7XX_PERF_UCHE_WRITE_REQUESTS_LRZ"/> - <value value="15" name="A7XX_PERF_UCHE_WRITE_REQUESTS_SP"/> - <value value="16" name="A7XX_PERF_UCHE_WRITE_REQUESTS_VPC"/> - <value value="17" name="A7XX_PERF_UCHE_WRITE_REQUESTS_VSC"/> - <value value="18" name="A7XX_PERF_UCHE_EVICTS"/> - <value value="19" name="A7XX_PERF_UCHE_BANK_REQ0"/> - <value value="20" name="A7XX_PERF_UCHE_BANK_REQ1"/> - <value value="21" name="A7XX_PERF_UCHE_BANK_REQ2"/> - <value value="22" name="A7XX_PERF_UCHE_BANK_REQ3"/> - <value value="23" name="A7XX_PERF_UCHE_BANK_REQ4"/> - <value value="24" name="A7XX_PERF_UCHE_BANK_REQ5"/> - <value value="25" name="A7XX_PERF_UCHE_BANK_REQ6"/> - <value value="26" name="A7XX_PERF_UCHE_BANK_REQ7"/> - <value value="27" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_CH0"/> - <value value="28" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_CH1"/> - <value value="29" name="A7XX_PERF_UCHE_GMEM_READ_BEATS"/> - <value value="30" name="A7XX_PERF_UCHE_TPH_REF_FULL"/> - <value value="31" name="A7XX_PERF_UCHE_TPH_VICTIM_FULL"/> - <value value="32" name="A7XX_PERF_UCHE_TPH_EXT_FULL"/> - <value value="33" name="A7XX_PERF_UCHE_VBIF_STALL_WRITE_DATA"/> - <value value="34" name="A7XX_PERF_UCHE_DCMP_LATENCY_SAMPLES"/> - <value value="35" name="A7XX_PERF_UCHE_DCMP_LATENCY_CYCLES"/> - <value value="36" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_PC"/> - <value value="37" name="A7XX_PERF_UCHE_READ_REQUESTS_PC"/> - <value value="38" name="A7XX_PERF_UCHE_RAM_READ_REQ"/> - <value value="39" name="A7XX_PERF_UCHE_RAM_WRITE_REQ"/> - <value value="40" name="A7XX_PERF_UCHE_STARVED_CYCLES_VBIF_DECMP"/> - <value value="41" name="A7XX_PERF_UCHE_STALL_CYCLES_DECMP"/> - <value value="42" name="A7XX_PERF_UCHE_ARBITER_STALL_CYCLES_VBIF"/> - <value value="43" name="A7XX_PERF_UCHE_READ_REQUESTS_TP_UBWC"/> - <value value="44" name="A7XX_PERF_UCHE_READ_REQUESTS_TP_NONUBWC"/> - <value value="45" name="A7XX_PERF_UCHE_READ_REQUESTS_TP_GMEM"/> - <value value="46" name="A7XX_PERF_UCHE_LONG_LINE_ALL_EVICTS_KAILUA"/> - <value value="47" name="A7XX_PERF_UCHE_LONG_LINE_PARTIAL_EVICTS_KAILUA"/> - <value value="48" name="A7XX_PERF_UCHE_TPH_CONFLICT_CL_CCHE"/> - <value value="49" name="A7XX_PERF_UCHE_TPH_CONFLICT_CL_OTHER_KAILUA"/> - <value value="50" name="A7XX_PERF_UCHE_DBANK_CONFLICT_CL_CCHE"/> - <value value="51" name="A7XX_PERF_UCHE_DBANK_CONFLICT_CL_OTHER_CLIENTS"/> - <value value="52" name="A7XX_PERF_UCHE_VBIF_WRITE_BEATS_CH0"/> - <value value="53" name="A7XX_PERF_UCHE_VBIF_WRITE_BEATS_CH1"/> - <value value="54" name="A7XX_PERF_UCHE_CCHE_TPH_QUEUE_FULL"/> - <value value="55" name="A7XX_PERF_UCHE_CCHE_DPH_QUEUE_FULL"/> - <value value="56" name="A7XX_PERF_UCHE_GMEM_WRITE_BEATS"/> - <value value="57" name="A7XX_PERF_UCHE_UBWC_READ_BEATS"/> - <value value="58" name="A7XX_PERF_UCHE_UBWC_WRITE_BEATS"/> -</enum> - -<enum name="a7xx_tp_perfcounter_select"> - <value value="0" name="A7XX_PERF_TP_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_TP_STALL_CYCLES_UCHE"/> - <value value="2" name="A7XX_PERF_TP_LATENCY_CYCLES"/> - <value value="3" name="A7XX_PERF_TP_LATENCY_TRANS"/> - <value value="4" name="A7XX_PERF_TP_FLAG_FIFO_DELAY_SAMPLES"/> - <value value="5" name="A7XX_PERF_TP_FLAG_FIFO_DELAY_CYCLES"/> - <value value="6" name="A7XX_PERF_TP_L1_CACHELINE_REQUESTS"/> - <value value="7" name="A7XX_PERF_TP_L1_CACHELINE_MISSES"/> - <value value="8" name="A7XX_PERF_TP_SP_TP_TRANS"/> - <value value="9" name="A7XX_PERF_TP_TP_SP_TRANS"/> - <value value="10" name="A7XX_PERF_TP_OUTPUT_PIXELS"/> - <value value="11" name="A7XX_PERF_TP_FILTER_WORKLOAD_16BIT"/> - <value value="12" name="A7XX_PERF_TP_FILTER_WORKLOAD_32BIT"/> - <value value="13" name="A7XX_PERF_TP_QUADS_RECEIVED"/> - <value value="14" name="A7XX_PERF_TP_QUADS_OFFSET"/> - <value value="15" name="A7XX_PERF_TP_QUADS_SHADOW"/> - <value value="16" name="A7XX_PERF_TP_QUADS_ARRAY"/> - <value value="17" name="A7XX_PERF_TP_QUADS_GRADIENT"/> - <value value="18" name="A7XX_PERF_TP_QUADS_1D"/> - <value value="19" name="A7XX_PERF_TP_QUADS_2D"/> - <value value="20" name="A7XX_PERF_TP_QUADS_BUFFER"/> - <value value="21" name="A7XX_PERF_TP_QUADS_3D"/> - <value value="22" name="A7XX_PERF_TP_QUADS_CUBE"/> - <value value="23" name="A7XX_PERF_TP_DIVERGENT_QUADS_RECEIVED"/> - <value value="24" name="A7XX_PERF_TP_PRT_NON_RESIDENT_EVENTS"/> - <value value="25" name="A7XX_PERF_TP_OUTPUT_PIXELS_POINT"/> - <value value="26" name="A7XX_PERF_TP_OUTPUT_PIXELS_BILINEAR"/> - <value value="27" name="A7XX_PERF_TP_OUTPUT_PIXELS_MIP"/> - <value value="28" name="A7XX_PERF_TP_OUTPUT_PIXELS_ANISO"/> - <value value="29" name="A7XX_PERF_TP_OUTPUT_PIXELS_ZERO_LOD"/> - <value value="30" name="A7XX_PERF_TP_FLAG_CACHE_REQUESTS"/> - <value value="31" name="A7XX_PERF_TP_FLAG_CACHE_MISSES"/> - <value value="32" name="A7XX_PERF_TP_L1_5_L2_REQUESTS"/> - <value value="33" name="A7XX_PERF_TP_2D_OUTPUT_PIXELS"/> - <value value="34" name="A7XX_PERF_TP_2D_OUTPUT_PIXELS_POINT"/> - <value value="35" name="A7XX_PERF_TP_2D_OUTPUT_PIXELS_BILINEAR"/> - <value value="36" name="A7XX_PERF_TP_2D_FILTER_WORKLOAD_16BIT"/> - <value value="37" name="A7XX_PERF_TP_2D_FILTER_WORKLOAD_32BIT"/> - <value value="38" name="A7XX_PERF_TP_TPA2TPC_TRANS"/> - <value value="39" name="A7XX_PERF_TP_L1_MISSES_ASTC_1TILE"/> - <value value="40" name="A7XX_PERF_TP_L1_MISSES_ASTC_2TILE"/> - <value value="41" name="A7XX_PERF_TP_L1_MISSES_ASTC_4TILE"/> - <value value="42" name="A7XX_PERF_TP_L1_5_COMPRESS_REQS"/> - <value value="43" name="A7XX_PERF_TP_L1_5_L2_COMPRESS_MISS"/> - <value value="44" name="A7XX_PERF_TP_L1_BANK_CONFLICT"/> - <value value="45" name="A7XX_PERF_TP_L1_5_MISS_LATENCY_CYCLES"/> - <value value="46" name="A7XX_PERF_TP_L1_5_MISS_LATENCY_TRANS"/> - <value value="47" name="A7XX_PERF_TP_QUADS_CONSTANT_MULTIPLIED"/> - <value value="48" name="A7XX_PERF_TP_FRONTEND_WORKING_CYCLES"/> - <value value="49" name="A7XX_PERF_TP_L1_TAG_WORKING_CYCLES"/> - <value value="50" name="A7XX_PERF_TP_L1_DATA_WRITE_WORKING_CYCLES"/> - <value value="51" name="A7XX_PERF_TP_PRE_L1_DECOM_WORKING_CYCLES"/> - <value value="52" name="A7XX_PERF_TP_BACKEND_WORKING_CYCLES"/> - <value value="53" name="A7XX_PERF_TP_L1_5_CACHE_WORKING_CYCLES"/> - <value value="54" name="A7XX_PERF_TP_STARVE_CYCLES_SP"/> - <value value="55" name="A7XX_PERF_TP_STARVE_CYCLES_UCHE"/> - <value value="56" name="A7XX_PERF_TP_STALL_CYCLES_UFC"/> - <value value="57" name="A7XX_PERF_TP_FORMAT_DECOMP"/> - <value value="58" name="A7XX_PERF_TP_FILTER_POINT_FP16"/> - <value value="59" name="A7XX_PERF_TP_FILTER_POINT_FP32"/> - <value value="60" name="A7XX_PERF_TP_LATENCY_FIFO_FULL"/> - <value value="61" name="A7XX_PERF_TP_RESERVED_61"/> - <value value="62" name="A7XX_PERF_TP_RESERVED_62"/> - <value value="63" name="A7XX_PERF_TP_RESERVED_63"/> - <value value="64" name="A7XX_PERF_TP_RESERVED_64"/> - <value value="65" name="A7XX_PERF_TP_RESERVED_65"/> - <value value="66" name="A7XX_PERF_TP_RESERVED_66"/> - <value value="67" name="A7XX_PERF_TP_RESERVED_67"/> - <value value="68" name="A7XX_PERF_TP_RESERVED_68"/> - <value value="69" name="A7XX_PERF_TP_RESERVED_69"/> - <value value="70" name="A7XX_PERF_TP_RESERVED_70"/> - <value value="71" name="A7XX_PERF_TP_RESERVED_71"/> - <value value="72" name="A7XX_PERF_TP_RESERVED_72"/> - <value value="73" name="A7XX_PERF_TP_RESERVED_73"/> - <value value="74" name="A7XX_PERF_TP_RESERVED_74"/> - <value value="75" name="A7XX_PERF_TP_RESERVED_75"/> - <value value="76" name="A7XX_PERF_TP_RESERVED_76"/> - <value value="77" name="A7XX_PERF_TP_RESERVED_77"/> - <value value="78" name="A7XX_PERF_TP_RESERVED_78"/> - <value value="79" name="A7XX_PERF_TP_RESERVED_79"/> - <value value="80" name="A7XX_PERF_TP_RESERVED_80"/> - <value value="81" name="A7XX_PERF_TP_RESERVED_81"/> - <value value="82" name="A7XX_PERF_TP_RESERVED_82"/> - <value value="83" name="A7XX_PERF_TP_RESERVED_83"/> - <value value="84" name="A7XX_PERF_TP_RESERVED_84"/> - <value value="85" name="A7XX_PERF_TP_RESERVED_85"/> - <value value="86" name="A7XX_PERF_TP_RESERVED_86"/> - <value value="87" name="A7XX_PERF_TP_RESERVED_87"/> - <value value="88" name="A7XX_PERF_TP_RESERVED_88"/> - <value value="89" name="A7XX_PERF_TP_RESERVED_89"/> - <value value="90" name="A7XX_PERF_TP_RESERVED_90"/> - <value value="91" name="A7XX_PERF_TP_RESERVED_91"/> - <value value="92" name="A7XX_PERF_TP_RESERVED_92"/> - <value value="93" name="A7XX_PERF_TP_RESERVED_93"/> - <value value="94" name="A7XX_PERF_TP_RESERVED_94"/> - <value value="95" name="A7XX_PERF_TP_RESERVED_95"/> - <value value="96" name="A7XX_PERF_TP_RESERVED_96"/> - <value value="97" name="A7XX_PERF_TP_RESERVED_97"/> - <value value="98" name="A7XX_PERF_TP_RESERVED_98"/> - <value value="99" name="A7XX_PERF_TP_RESERVED_99"/> - <value value="100" name="A7XX_PERF_TP_RESERVED_100"/> - <value value="101" name="A7XX_PERF_TP_RESERVED_101"/> - <value value="102" name="A7XX_PERF_TP_RESERVED_102"/> - <value value="103" name="A7XX_PERF_TP_RESERVED_103"/> - <value value="104" name="A7XX_PERF_TP_RESERVED_104"/> - <value value="105" name="A7XX_PERF_TP_RESERVED_105"/> - <value value="106" name="A7XX_PERF_TP_RESERVED_106"/> - <value value="107" name="A7XX_PERF_TP_RESERVED_107"/> - <value value="108" name="A7XX_PERF_TP_RESERVED_108"/> - <value value="109" name="A7XX_PERF_TP_RESERVED_109"/> - <value value="110" name="A7XX_PERF_TP_RESERVED_110"/> - <value value="111" name="A7XX_PERF_TP_RESERVED_111"/> - <value value="112" name="A7XX_PERF_TP_RESERVED_112"/> - <value value="113" name="A7XX_PERF_TP_RESERVED_113"/> - <value value="114" name="A7XX_PERF_TP_RESERVED_114"/> - <value value="115" name="A7XX_PERF_TP_RESERVED_115"/> - <value value="116" name="A7XX_PERF_TP_RESERVED_116"/> - <value value="117" name="A7XX_PERF_TP_RESERVED_117"/> - <value value="118" name="A7XX_PERF_TP_RESERVED_118"/> - <value value="119" name="A7XX_PERF_TP_RESERVED_119"/> - <value value="120" name="A7XX_PERF_TP_RESERVED_120"/> - <value value="121" name="A7XX_PERF_TP_RESERVED_121"/> - <value value="122" name="A7XX_PERF_TP_RESERVED_122"/> - <value value="123" name="A7XX_PERF_TP_RESERVED_123"/> - <value value="124" name="A7XX_PERF_TP_RESERVED_124"/> - <value value="125" name="A7XX_PERF_TP_RESERVED_125"/> - <value value="126" name="A7XX_PERF_TP_RESERVED_126"/> - <value value="127" name="A7XX_PERF_TP_RESERVED_127"/> - <value value="128" name="A7XX_PERF_TP_FORMAT_DECOMP_BILINEAR"/> - <value value="129" name="A7XX_PERF_TP_PACKED_POINT_BOTH_VALID_FP16"/> - <value value="130" name="A7XX_PERF_TP_PACKED_POINT_SINGLE_VALID_FP16"/> - <value value="131" name="A7XX_PERF_TP_PACKED_POINT_BOTH_VALID_FP32"/> - <value value="132" name="A7XX_PERF_TP_PACKED_POINT_SINGLE_VALID_FP32"/> -</enum> - -<enum name="a7xx_sp_perfcounter_select"> - <value value="0" name="A7XX_PERF_SP_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_SP_ALU_WORKING_CYCLES"/> - <value value="2" name="A7XX_PERF_SP_EFU_WORKING_CYCLES"/> - <value value="3" name="A7XX_PERF_SP_STALL_CYCLES_VPC"/> - <value value="4" name="A7XX_PERF_SP_STALL_CYCLES_TP"/> - <value value="5" name="A7XX_PERF_SP_STALL_CYCLES_UCHE"/> - <value value="6" name="A7XX_PERF_SP_STALL_CYCLES_RB"/> - <value value="7" name="A7XX_PERF_SP_NON_EXECUTION_CYCLES"/> - <value value="8" name="A7XX_PERF_SP_WAVE_CONTEXTS"/> - <value value="9" name="A7XX_PERF_SP_WAVE_CONTEXT_CYCLES"/> - <value value="10" name="A7XX_PERF_SP_STAGE_WAVE_CYCLES"/> - <value value="11" name="A7XX_PERF_SP_STAGE_WAVE_SAMPLES"/> - <value value="12" name="A7XX_PERF_SP_VS_STAGE_WAVE_CYCLES"/> - <value value="13" name="A7XX_PERF_SP_VS_STAGE_WAVE_SAMPLES"/> - <value value="14" name="A7XX_PERF_SP_FS_STAGE_DURATION_CYCLES"/> - <value value="15" name="A7XX_PERF_SP_VS_STAGE_DURATION_CYCLES"/> - <value value="16" name="A7XX_PERF_SP_WAVE_CTRL_CYCLES"/> - <value value="17" name="A7XX_PERF_SP_WAVE_LOAD_CYCLES"/> - <value value="18" name="A7XX_PERF_SP_WAVE_EMIT_CYCLES"/> - <value value="19" name="A7XX_PERF_SP_WAVE_NOP_CYCLES"/> - <value value="20" name="A7XX_PERF_SP_WAVE_WAIT_CYCLES"/> - <value value="21" name="A7XX_PERF_SP_WAVE_FETCH_CYCLES"/> - <value value="22" name="A7XX_PERF_SP_WAVE_IDLE_CYCLES"/> - <value value="23" name="A7XX_PERF_SP_WAVE_END_CYCLES"/> - <value value="24" name="A7XX_PERF_SP_WAVE_LONG_SYNC_CYCLES"/> - <value value="25" name="A7XX_PERF_SP_WAVE_SHORT_SYNC_CYCLES"/> - <value value="26" name="A7XX_PERF_SP_WAVE_JOIN_CYCLES"/> - <value value="27" name="A7XX_PERF_SP_LM_LOAD_INSTRUCTIONS"/> - <value value="28" name="A7XX_PERF_SP_LM_STORE_INSTRUCTIONS"/> - <value value="29" name="A7XX_PERF_SP_LM_ATOMICS"/> - <value value="30" name="A7XX_PERF_SP_GM_LOAD_INSTRUCTIONS"/> - <value value="31" name="A7XX_PERF_SP_GM_STORE_INSTRUCTIONS"/> - <value value="32" name="A7XX_PERF_SP_GM_ATOMICS"/> - <value value="33" name="A7XX_PERF_SP_VS_STAGE_TEX_INSTRUCTIONS"/> - <value value="34" name="A7XX_PERF_SP_VS_STAGE_EFU_INSTRUCTIONS"/> - <value value="35" name="A7XX_PERF_SP_VS_STAGE_FULL_ALU_INSTRUCTIONS"/> - <value value="36" name="A7XX_PERF_SP_VS_STAGE_HALF_ALU_INSTRUCTIONS"/> - <value value="37" name="A7XX_PERF_SP_FS_STAGE_TEX_INSTRUCTIONS"/> - <value value="38" name="A7XX_PERF_SP_FS_STAGE_CFLOW_INSTRUCTIONS"/> - <value value="39" name="A7XX_PERF_SP_FS_STAGE_EFU_INSTRUCTIONS"/> - <value value="40" name="A7XX_PERF_SP_FS_STAGE_FULL_ALU_INSTRUCTIONS"/> - <value value="41" name="A7XX_PERF_SP_FS_STAGE_HALF_ALU_INSTRUCTIONS"/> - <value value="42" name="A7XX_PERF_SP_FS_STAGE_BARY_INSTRUCTIONS"/> - <value value="43" name="A7XX_PERF_SP_VS_INSTRUCTIONS"/> - <value value="44" name="A7XX_PERF_SP_FS_INSTRUCTIONS"/> - <value value="45" name="A7XX_PERF_SP_ADDR_LOCK_COUNT"/> - <value value="46" name="A7XX_PERF_SP_UCHE_READ_TRANS"/> - <value value="47" name="A7XX_PERF_SP_UCHE_WRITE_TRANS"/> - <value value="48" name="A7XX_PERF_SP_EXPORT_VPC_TRANS"/> - <value value="49" name="A7XX_PERF_SP_EXPORT_RB_TRANS"/> - <value value="50" name="A7XX_PERF_SP_PIXELS_KILLED"/> - <value value="51" name="A7XX_PERF_SP_ICL1_REQUESTS"/> - <value value="52" name="A7XX_PERF_SP_ICL1_MISSES"/> - <value value="53" name="A7XX_PERF_SP_HS_INSTRUCTIONS"/> - <value value="54" name="A7XX_PERF_SP_DS_INSTRUCTIONS"/> - <value value="55" name="A7XX_PERF_SP_GS_INSTRUCTIONS"/> - <value value="56" name="A7XX_PERF_SP_CS_INSTRUCTIONS"/> - <value value="57" name="A7XX_PERF_SP_GPR_READ"/> - <value value="58" name="A7XX_PERF_SP_GPR_WRITE"/> - <value value="59" name="A7XX_PERF_SP_FS_STAGE_HALF_EFU_INSTRUCTIONS"/> - <value value="60" name="A7XX_PERF_SP_VS_STAGE_HALF_EFU_INSTRUCTIONS"/> - <value value="61" name="A7XX_PERF_SP_LM_BANK_CONFLICTS"/> - <value value="62" name="A7XX_PERF_SP_TEX_CONTROL_WORKING_CYCLES"/> - <value value="63" name="A7XX_PERF_SP_LOAD_CONTROL_WORKING_CYCLES"/> - <value value="64" name="A7XX_PERF_SP_FLOW_CONTROL_WORKING_CYCLES"/> - <value value="65" name="A7XX_PERF_SP_LM_WORKING_CYCLES"/> - <value value="66" name="A7XX_PERF_SP_DISPATCHER_WORKING_CYCLES"/> - <value value="67" name="A7XX_PERF_SP_SEQUENCER_WORKING_CYCLES"/> - <value value="68" name="A7XX_PERF_SP_LOW_EFFICIENCY_STARVED_BY_TP"/> - <value value="69" name="A7XX_PERF_SP_STARVE_CYCLES_HLSQ"/> - <value value="70" name="A7XX_PERF_SP_NON_EXECUTION_LS_CYCLES"/> - <value value="71" name="A7XX_PERF_SP_WORKING_EU"/> - <value value="72" name="A7XX_PERF_SP_ANY_EU_WORKING"/> - <value value="73" name="A7XX_PERF_SP_WORKING_EU_FS_STAGE"/> - <value value="74" name="A7XX_PERF_SP_ANY_EU_WORKING_FS_STAGE"/> - <value value="75" name="A7XX_PERF_SP_WORKING_EU_VS_STAGE"/> - <value value="76" name="A7XX_PERF_SP_ANY_EU_WORKING_VS_STAGE"/> - <value value="77" name="A7XX_PERF_SP_WORKING_EU_CS_STAGE"/> - <value value="78" name="A7XX_PERF_SP_ANY_EU_WORKING_CS_STAGE"/> - <value value="79" name="A7XX_PERF_SP_GPR_READ_PREFETCH"/> - <value value="80" name="A7XX_PERF_SP_GPR_READ_CONFLICT"/> - <value value="81" name="A7XX_PERF_SP_GPR_WRITE_CONFLICT"/> - <value value="82" name="A7XX_PERF_SP_GM_LOAD_LATENCY_CYCLES"/> - <value value="83" name="A7XX_PERF_SP_GM_LOAD_LATENCY_SAMPLES"/> - <value value="84" name="A7XX_PERF_SP_EXECUTABLE_WAVES"/> - <value value="85" name="A7XX_PERF_SP_ICL1_MISS_FETCH_CYCLES"/> - <value value="86" name="A7XX_PERF_SP_WORKING_EU_LPAC"/> - <value value="87" name="A7XX_PERF_SP_BYPASS_BUSY_CYCLES"/> - <value value="88" name="A7XX_PERF_SP_ANY_EU_WORKING_LPAC"/> - <value value="89" name="A7XX_PERF_SP_WAVE_ALU_CYCLES"/> - <value value="90" name="A7XX_PERF_SP_WAVE_EFU_CYCLES"/> - <value value="91" name="A7XX_PERF_SP_WAVE_INT_CYCLES"/> - <value value="92" name="A7XX_PERF_SP_WAVE_CSP_CYCLES"/> - <value value="93" name="A7XX_PERF_SP_EWAVE_CONTEXTS"/> - <value value="94" name="A7XX_PERF_SP_EWAVE_CONTEXT_CYCLES"/> - <value value="95" name="A7XX_PERF_SP_LPAC_BUSY_CYCLES"/> - <value value="96" name="A7XX_PERF_SP_LPAC_INSTRUCTIONS"/> - <value value="97" name="A7XX_PERF_SP_FS_STAGE_1X_WAVES"/> - <value value="98" name="A7XX_PERF_SP_FS_STAGE_2X_WAVES"/> - <value value="99" name="A7XX_PERF_SP_QUADS"/> - <value value="100" name="A7XX_PERF_SP_CS_INVOCATIONS"/> - <value value="101" name="A7XX_PERF_SP_PIXELS"/> - <value value="102" name="A7XX_PERF_SP_LPAC_DRAWCALLS"/> - <value value="103" name="A7XX_PERF_SP_PI_WORKING_CYCLES"/> - <value value="104" name="A7XX_PERF_SP_WAVE_INPUT_CYCLES"/> - <value value="105" name="A7XX_PERF_SP_WAVE_OUTPUT_CYCLES"/> - <value value="106" name="A7XX_PERF_SP_WAVE_HWAVE_WAIT_CYCLES"/> - <value value="107" name="A7XX_PERF_SP_WAVE_HWAVE_SYNC"/> - <value value="108" name="A7XX_PERF_SP_OUTPUT_3D_PIXELS"/> - <value value="109" name="A7XX_PERF_SP_FULL_ALU_MAD_INSTRUCTIONS"/> - <value value="110" name="A7XX_PERF_SP_HALF_ALU_MAD_INSTRUCTIONS"/> - <value value="111" name="A7XX_PERF_SP_FULL_ALU_MUL_INSTRUCTIONS"/> - <value value="112" name="A7XX_PERF_SP_HALF_ALU_MUL_INSTRUCTIONS"/> - <value value="113" name="A7XX_PERF_SP_FULL_ALU_ADD_INSTRUCTIONS"/> - <value value="114" name="A7XX_PERF_SP_HALF_ALU_ADD_INSTRUCTIONS"/> - <value value="115" name="A7XX_PERF_SP_BARY_FP32_INSTRUCTIONS"/> - <value value="116" name="A7XX_PERF_SP_ALU_GPR_READ_CYCLES"/> - <value value="117" name="A7XX_PERF_SP_ALU_DATA_FORWARDING_CYCLES"/> - <value value="118" name="A7XX_PERF_SP_LM_FULL_CYCLES"/> - <value value="119" name="A7XX_PERF_SP_TEXTURE_FETCH_LATENCY_CYCLES"/> - <value value="120" name="A7XX_PERF_SP_TEXTURE_FETCH_LATENCY_SAMPLES"/> - <value value="121" name="A7XX_PERF_SP_FS_STAGE_PI_TEX_INSTRUCTION"/> - <value value="122" name="A7XX_PERF_SP_RAY_QUERY_INSTRUCTIONS"/> - <value value="123" name="A7XX_PERF_SP_RBRT_KICKOFF_FIBERS"/> - <value value="124" name="A7XX_PERF_SP_RBRT_KICKOFF_DQUADS"/> - <value value="125" name="A7XX_PERF_SP_RTU_BUSY_CYCLES"/> - <value value="126" name="A7XX_PERF_SP_RTU_L0_HITS"/> - <value value="127" name="A7XX_PERF_SP_RTU_L0_MISSES"/> - <value value="128" name="A7XX_PERF_SP_RTU_L0_HIT_ON_MISS"/> - <value value="129" name="A7XX_PERF_SP_RTU_STALL_CYCLES_WAVE_QUEUE"/> - <value value="130" name="A7XX_PERF_SP_RTU_STALL_CYCLES_L0_HIT_QUEUE"/> - <value value="131" name="A7XX_PERF_SP_RTU_STALL_CYCLES_L0_MISS_QUEUE"/> - <value value="132" name="A7XX_PERF_SP_RTU_STALL_CYCLES_L0D_IDX_QUEUE"/> - <value value="133" name="A7XX_PERF_SP_RTU_STALL_CYCLES_L0DATA"/> - <value value="134" name="A7XX_PERF_SP_RTU_STALL_CYCLES_REPLACE_CNT"/> - <value value="135" name="A7XX_PERF_SP_RTU_STALL_CYCLES_MRG_CNT"/> - <value value="136" name="A7XX_PERF_SP_RTU_STALL_CYCLES_UCHE"/> - <value value="137" name="A7XX_PERF_SP_RTU_OPERAND_FETCH_STALL_CYCLES_L0"/> - <value value="138" name="A7XX_PERF_SP_RTU_OPERAND_FETCH_STALL_CYCLES_INS_FIFO"/> - <value value="139" name="A7XX_PERF_SP_RTU_BVH_FETCH_LATENCY_CYCLES"/> - <value value="140" name="A7XX_PERF_SP_RTU_BVH_FETCH_LATENCY_SAMPLES"/> - <value value="141" name="A7XX_PERF_SP_STCHE_MISS_INC_VS"/> - <value value="142" name="A7XX_PERF_SP_STCHE_MISS_INC_FS"/> - <value value="143" name="A7XX_PERF_SP_STCHE_MISS_INC_BV"/> - <value value="144" name="A7XX_PERF_SP_STCHE_MISS_INC_LPAC"/> - <value value="145" name="A7XX_PERF_SP_VGPR_ACTIVE_CONTEXTS"/> - <value value="146" name="A7XX_PERF_SP_PGPR_ALLOC_CONTEXTS"/> - <value value="147" name="A7XX_PERF_SP_VGPR_ALLOC_CONTEXTS"/> - <value value="148" name="A7XX_PERF_SP_RTU_RAY_BOX_INTERSECTIONS"/> - <value value="149" name="A7XX_PERF_SP_RTU_RAY_TRIANGLE_INTERSECTIONS"/> - <value value="150" name="A7XX_PERF_SP_SCH_STALL_CYCLES_RTU"/> -</enum> - -<enum name="a7xx_rb_perfcounter_select"> - <value value="0" name="A7XX_PERF_RB_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_RB_STALL_CYCLES_HLSQ"/> - <value value="2" name="A7XX_PERF_RB_STALL_CYCLES_FIFO0_FULL"/> - <value value="3" name="A7XX_PERF_RB_STALL_CYCLES_FIFO1_FULL"/> - <value value="4" name="A7XX_PERF_RB_STALL_CYCLES_FIFO2_FULL"/> - <value value="5" name="A7XX_PERF_RB_STARVE_CYCLES_SP"/> - <value value="6" name="A7XX_PERF_RB_STARVE_CYCLES_LRZ_TILE"/> - <value value="7" name="A7XX_PERF_RB_STARVE_CYCLES_CCU"/> - <value value="8" name="A7XX_PERF_RB_STARVE_CYCLES_Z_PLANE"/> - <value value="9" name="A7XX_PERF_RB_STARVE_CYCLES_BARY_PLANE"/> - <value value="10" name="A7XX_PERF_RB_Z_WORKLOAD"/> - <value value="11" name="A7XX_PERF_RB_HLSQ_ACTIVE"/> - <value value="12" name="A7XX_PERF_RB_Z_READ"/> - <value value="13" name="A7XX_PERF_RB_Z_WRITE"/> - <value value="14" name="A7XX_PERF_RB_C_READ"/> - <value value="15" name="A7XX_PERF_RB_C_WRITE"/> - <value value="16" name="A7XX_PERF_RB_TOTAL_PASS"/> - <value value="17" name="A7XX_PERF_RB_Z_PASS"/> - <value value="18" name="A7XX_PERF_RB_Z_FAIL"/> - <value value="19" name="A7XX_PERF_RB_S_FAIL"/> - <value value="20" name="A7XX_PERF_RB_BLENDED_FXP_COMPONENTS"/> - <value value="21" name="A7XX_PERF_RB_BLENDED_FP16_COMPONENTS"/> - <value value="22" name="A7XX_PERF_RB_PS_INVOCATIONS"/> - <value value="23" name="A7XX_PERF_RB_2D_ALIVE_CYCLES"/> - <value value="24" name="A7XX_PERF_RB_2D_STALL_CYCLES_A2D"/> - <value value="25" name="A7XX_PERF_RB_2D_STARVE_CYCLES_SRC"/> - <value value="26" name="A7XX_PERF_RB_2D_STARVE_CYCLES_SP"/> - <value value="27" name="A7XX_PERF_RB_2D_STARVE_CYCLES_DST"/> - <value value="28" name="A7XX_PERF_RB_2D_VALID_PIXELS"/> - <value value="29" name="A7XX_PERF_RB_3D_PIXELS"/> - <value value="30" name="A7XX_PERF_RB_BLENDER_WORKING_CYCLES"/> - <value value="31" name="A7XX_PERF_RB_ZPROC_WORKING_CYCLES"/> - <value value="32" name="A7XX_PERF_RB_CPROC_WORKING_CYCLES"/> - <value value="33" name="A7XX_PERF_RB_SAMPLER_WORKING_CYCLES"/> - <value value="34" name="A7XX_PERF_RB_STALL_CYCLES_CCU_COLOR_READ"/> - <value value="35" name="A7XX_PERF_RB_STALL_CYCLES_CCU_COLOR_WRITE"/> - <value value="36" name="A7XX_PERF_RB_STALL_CYCLES_CCU_DEPTH_READ"/> - <value value="37" name="A7XX_PERF_RB_STALL_CYCLES_CCU_DEPTH_WRITE"/> - <value value="38" name="A7XX_PERF_RB_STALL_CYCLES_VPC"/> - <value value="39" name="A7XX_PERF_RB_2D_INPUT_TRANS"/> - <value value="40" name="A7XX_PERF_RB_2D_OUTPUT_RB_DST_TRANS"/> - <value value="41" name="A7XX_PERF_RB_2D_OUTPUT_RB_SRC_TRANS"/> - <value value="42" name="A7XX_PERF_RB_BLENDED_FP32_COMPONENTS"/> - <value value="43" name="A7XX_PERF_RB_COLOR_PIX_TILES"/> - <value value="44" name="A7XX_PERF_RB_STALL_CYCLES_CCU"/> - <value value="45" name="A7XX_PERF_RB_EARLY_Z_ARB3_GRANT"/> - <value value="46" name="A7XX_PERF_RB_LATE_Z_ARB3_GRANT"/> - <value value="47" name="A7XX_PERF_RB_EARLY_Z_SKIP_GRANT"/> - <value value="48" name="A7XX_PERF_RB_VRS_1x1_QUADS"/> - <value value="49" name="A7XX_PERF_RB_VRS_2x1_QUADS"/> - <value value="50" name="A7XX_PERF_RB_VRS_1x2_QUADS"/> - <value value="51" name="A7XX_PERF_RB_VRS_2x2_QUADS"/> - <value value="52" name="A7XX_PERF_RB_VRS_4x2_QUADS"/> - <value value="53" name="A7XX_PERF_RB_VRS_4x4_QUADS"/> -</enum> - -<enum name="a7xx_vsc_perfcounter_select"> - <value value="0" name="A7XX_PERF_VSC_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_VSC_WORKING_CYCLES"/> - <value value="2" name="A7XX_PERF_VSC_STALL_CYCLES_UCHE"/> - <value value="3" name="A7XX_PERF_VSC_EOT_NUM"/> - <value value="4" name="A7XX_PERF_VSC_INPUT_TILES"/> -</enum> - -<enum name="a7xx_ccu_perfcounter_select"> - <value value="0" name="A7XX_PERF_CCU_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_CCU_STALL_CYCLES_RB_DEPTH_RETURN"/> - <value value="2" name="A7XX_PERF_CCU_STALL_CYCLES_RB_COLOR_RETURN"/> - <value value="3" name="A7XX_PERF_CCU_DEPTH_BLOCKS"/> - <value value="4" name="A7XX_PERF_CCU_COLOR_BLOCKS"/> - <value value="5" name="A7XX_PERF_CCU_DEPTH_BLOCK_HIT"/> - <value value="6" name="A7XX_PERF_CCU_COLOR_BLOCK_HIT"/> - <value value="7" name="A7XX_PERF_CCU_PARTIAL_BLOCK_READ"/> - <value value="8" name="A7XX_PERF_CCU_GMEM_READ"/> - <value value="9" name="A7XX_PERF_CCU_GMEM_WRITE"/> - <value value="10" name="A7XX_PERF_CCU_2D_RD_REQ"/> - <value value="11" name="A7XX_PERF_CCU_2D_WR_REQ"/> - <value value="12" name="A7XX_PERF_CCU_UBWC_COLOR_BLOCKS_CONCURRENT"/> - <value value="13" name="A7XX_PERF_CCU_UBWC_DEPTH_BLOCKS_CONCURRENT"/> - <value value="14" name="A7XX_PERF_CCU_COLOR_RESOLVE_DROPPED"/> - <value value="15" name="A7XX_PERF_CCU_DEPTH_RESOLVE_DROPPED"/> - <value value="16" name="A7XX_PERF_CCU_COLOR_RENDER_CONCURRENT"/> - <value value="17" name="A7XX_PERF_CCU_DEPTH_RENDER_CONCURRENT"/> - <value value="18" name="A7XX_PERF_CCU_COLOR_RESOLVE_AFTER_RENDER"/> - <value value="19" name="A7XX_PERF_CCU_DEPTH_RESOLVE_AFTER_RENDER"/> - <value value="20" name="A7XX_PERF_CCU_GMEM_EXTRA_DEPTH_READ"/> - <value value="21" name="A7XX_PERF_CCU_GMEM_COLOR_READ_4AA"/> - <value value="22" name="A7XX_PERF_CCU_GMEM_COLOR_READ_4AA_FULL"/> -</enum> - -<enum name="a7xx_lrz_perfcounter_select"> - <value value="0" name="A7XX_PERF_LRZ_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_LRZ_STARVE_CYCLES_RAS"/> - <value value="2" name="A7XX_PERF_LRZ_STALL_CYCLES_RB"/> - <value value="3" name="A7XX_PERF_LRZ_STALL_CYCLES_VSC"/> - <value value="4" name="A7XX_PERF_LRZ_STALL_CYCLES_VPC"/> - <value value="5" name="A7XX_PERF_LRZ_STALL_CYCLES_FLAG_PREFETCH"/> - <value value="6" name="A7XX_PERF_LRZ_STALL_CYCLES_UCHE"/> - <value value="7" name="A7XX_PERF_LRZ_LRZ_READ"/> - <value value="8" name="A7XX_PERF_LRZ_LRZ_WRITE"/> - <value value="9" name="A7XX_PERF_LRZ_READ_LATENCY"/> - <value value="10" name="A7XX_PERF_LRZ_MERGE_CACHE_UPDATING"/> - <value value="11" name="A7XX_PERF_LRZ_PRIM_KILLED_BY_MASKGEN"/> - <value value="12" name="A7XX_PERF_LRZ_PRIM_KILLED_BY_LRZ"/> - <value value="13" name="A7XX_PERF_LRZ_VISIBLE_PRIM_AFTER_LRZ"/> - <value value="14" name="A7XX_PERF_LRZ_FULL_8X8_TILES"/> - <value value="15" name="A7XX_PERF_LRZ_PARTIAL_8X8_TILES"/> - <value value="16" name="A7XX_PERF_LRZ_TILE_KILLED"/> - <value value="17" name="A7XX_PERF_LRZ_TOTAL_PIXEL"/> - <value value="18" name="A7XX_PERF_LRZ_VISIBLE_PIXEL_AFTER_LRZ"/> - <value value="19" name="A7XX_PERF_LRZ_FEEDBACK_ACCEPT"/> - <value value="20" name="A7XX_PERF_LRZ_FEEDBACK_DISCARD"/> - <value value="21" name="A7XX_PERF_LRZ_FEEDBACK_STALL"/> - <value value="22" name="A7XX_PERF_LRZ_STALL_CYCLES_RB_ZPLANE"/> - <value value="23" name="A7XX_PERF_LRZ_STALL_CYCLES_RB_BPLANE"/> - <value value="24" name="A7XX_PERF_LRZ_RAS_MASK_TRANS"/> - <value value="25" name="A7XX_PERF_LRZ_STALL_CYCLES_MVC"/> - <value value="26" name="A7XX_PERF_LRZ_TILE_KILLED_BY_IMAGE_VRS"/> - <value value="27" name="A7XX_PERF_LRZ_TILE_KILLED_BY_Z"/> -</enum> - -<enum name="a7xx_cmp_perfcounter_select"> - <value value="0" name="A7XX_PERF_CMPDECMP_STALL_CYCLES_ARB"/> - <value value="1" name="A7XX_PERF_CMPDECMP_VBIF_LATENCY_CYCLES"/> - <value value="2" name="A7XX_PERF_CMPDECMP_VBIF_LATENCY_SAMPLES"/> - <value value="3" name="A7XX_PERF_CMPDECMP_VBIF_READ_DATA_CCU"/> - <value value="4" name="A7XX_PERF_CMPDECMP_VBIF_WRITE_DATA_CCU"/> - <value value="5" name="A7XX_PERF_CMPDECMP_VBIF_READ_REQUEST"/> - <value value="6" name="A7XX_PERF_CMPDECMP_VBIF_WRITE_REQUEST"/> - <value value="7" name="A7XX_PERF_CMPDECMP_VBIF_READ_DATA"/> - <value value="8" name="A7XX_PERF_CMPDECMP_VBIF_WRITE_DATA"/> - <value value="9" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG1_COUNT"/> - <value value="10" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG2_COUNT"/> - <value value="11" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG3_COUNT"/> - <value value="12" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG4_COUNT"/> - <value value="13" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG5_COUNT"/> - <value value="14" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG6_COUNT"/> - <value value="15" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG8_COUNT"/> - <value value="16" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG1_COUNT"/> - <value value="17" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG2_COUNT"/> - <value value="18" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG3_COUNT"/> - <value value="19" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG4_COUNT"/> - <value value="20" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG5_COUNT"/> - <value value="21" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG6_COUNT"/> - <value value="22" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG8_COUNT"/> - <value value="23" name="A7XX_PERF_CMPDECMP_VBIF_READ_DATA_UCHE_CH0"/> - <value value="24" name="A7XX_PERF_CMPDECMP_VBIF_READ_DATA_UCHE_CH1"/> - <value value="25" name="A7XX_PERF_CMPDECMP_VBIF_WRITE_DATA_UCHE"/> - <value value="26" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG0_COUNT"/> - <value value="27" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG0_COUNT"/> - <value value="28" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAGALPHA_COUNT"/> - <value value="29" name="A7XX_PERF_CMPDECMP_RESOLVE_EVENTS"/> - <value value="30" name="A7XX_PERF_CMPDECMP_CONCURRENT_RESOLVE_EVENTS"/> - <value value="31" name="A7XX_PERF_CMPDECMP_DROPPED_CLEAR_EVENTS"/> - <value value="32" name="A7XX_PERF_CMPDECMP_ST_BLOCKS_CONCURRENT"/> - <value value="33" name="A7XX_PERF_CMPDECMP_LRZ_ST_BLOCKS_CONCURRENT"/> - <value value="34" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG0_COUNT"/> - <value value="35" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG1_COUNT"/> - <value value="36" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG2_COUNT"/> - <value value="37" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG3_COUNT"/> - <value value="38" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG4_COUNT"/> - <value value="39" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG5_COUNT"/> - <value value="40" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG6_COUNT"/> - <value value="41" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG8_COUNT"/> - <value value="42" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG0_COUNT"/> - <value value="43" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG1_COUNT"/> - <value value="44" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG2_COUNT"/> - <value value="45" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG3_COUNT"/> - <value value="46" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG4_COUNT"/> - <value value="47" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG5_COUNT"/> - <value value="48" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG6_COUNT"/> - <value value="49" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG8_COUNT"/> -</enum> - -<enum name="a7xx_gbif_perfcounter_select"> - <value value="0" name="A7XX_PERF_GBIF_RESERVED_0"/> - <value value="1" name="A7XX_PERF_GBIF_RESERVED_1"/> - <value value="2" name="A7XX_PERF_GBIF_RESERVED_2"/> - <value value="3" name="A7XX_PERF_GBIF_RESERVED_3"/> - <value value="4" name="A7XX_PERF_GBIF_RESERVED_4"/> - <value value="5" name="A7XX_PERF_GBIF_RESERVED_5"/> - <value value="6" name="A7XX_PERF_GBIF_RESERVED_6"/> - <value value="7" name="A7XX_PERF_GBIF_RESERVED_7"/> - <value value="8" name="A7XX_PERF_GBIF_RESERVED_8"/> - <value value="9" name="A7XX_PERF_GBIF_RESERVED_9"/> - <value value="10" name="A7XX_PERF_GBIF_AXI0_READ_REQUESTS_TOTAL"/> - <value value="11" name="A7XX_PERF_GBIF_AXI1_READ_REQUESTS_TOTAL"/> - <value value="12" name="A7XX_PERF_GBIF_RESERVED_12"/> - <value value="13" name="A7XX_PERF_GBIF_RESERVED_13"/> - <value value="14" name="A7XX_PERF_GBIF_RESERVED_14"/> - <value value="15" name="A7XX_PERF_GBIF_RESERVED_15"/> - <value value="16" name="A7XX_PERF_GBIF_RESERVED_16"/> - <value value="17" name="A7XX_PERF_GBIF_RESERVED_17"/> - <value value="18" name="A7XX_PERF_GBIF_RESERVED_18"/> - <value value="19" name="A7XX_PERF_GBIF_RESERVED_19"/> - <value value="20" name="A7XX_PERF_GBIF_RESERVED_20"/> - <value value="21" name="A7XX_PERF_GBIF_RESERVED_21"/> - <value value="22" name="A7XX_PERF_GBIF_AXI0_WRITE_REQUESTS_TOTAL"/> - <value value="23" name="A7XX_PERF_GBIF_AXI1_WRITE_REQUESTS_TOTAL"/> - <value value="24" name="A7XX_PERF_GBIF_RESERVED_24"/> - <value value="25" name="A7XX_PERF_GBIF_RESERVED_25"/> - <value value="26" name="A7XX_PERF_GBIF_RESERVED_26"/> - <value value="27" name="A7XX_PERF_GBIF_RESERVED_27"/> - <value value="28" name="A7XX_PERF_GBIF_RESERVED_28"/> - <value value="29" name="A7XX_PERF_GBIF_RESERVED_29"/> - <value value="30" name="A7XX_PERF_GBIF_RESERVED_30"/> - <value value="31" name="A7XX_PERF_GBIF_RESERVED_31"/> - <value value="32" name="A7XX_PERF_GBIF_RESERVED_32"/> - <value value="33" name="A7XX_PERF_GBIF_RESERVED_33"/> - <value value="34" name="A7XX_PERF_GBIF_AXI0_READ_DATA_BEATS_TOTAL"/> - <value value="35" name="A7XX_PERF_GBIF_AXI1_READ_DATA_BEATS_TOTAL"/> - <value value="36" name="A7XX_PERF_GBIF_RESERVED_36"/> - <value value="37" name="A7XX_PERF_GBIF_RESERVED_37"/> - <value value="38" name="A7XX_PERF_GBIF_RESERVED_38"/> - <value value="39" name="A7XX_PERF_GBIF_RESERVED_39"/> - <value value="40" name="A7XX_PERF_GBIF_RESERVED_40"/> - <value value="41" name="A7XX_PERF_GBIF_RESERVED_41"/> - <value value="42" name="A7XX_PERF_GBIF_RESERVED_42"/> - <value value="43" name="A7XX_PERF_GBIF_RESERVED_43"/> - <value value="44" name="A7XX_PERF_GBIF_RESERVED_44"/> - <value value="45" name="A7XX_PERF_GBIF_RESERVED_45"/> - <value value="46" name="A7XX_PERF_GBIF_AXI0_WRITE_DATA_BEATS_TOTAL"/> - <value value="47" name="A7XX_PERF_GBIF_AXI1_WRITE_DATA_BEATS_TOTAL"/> - <value value="48" name="A7XX_PERF_GBIF_RESERVED_48"/> - <value value="49" name="A7XX_PERF_GBIF_RESERVED_49"/> - <value value="50" name="A7XX_PERF_GBIF_RESERVED_50"/> - <value value="51" name="A7XX_PERF_GBIF_RESERVED_51"/> - <value value="52" name="A7XX_PERF_GBIF_RESERVED_52"/> - <value value="53" name="A7XX_PERF_GBIF_RESERVED_53"/> - <value value="54" name="A7XX_PERF_GBIF_RESERVED_54"/> - <value value="55" name="A7XX_PERF_GBIF_RESERVED_55"/> - <value value="56" name="A7XX_PERF_GBIF_RESERVED_56"/> - <value value="57" name="A7XX_PERF_GBIF_RESERVED_57"/> - <value value="58" name="A7XX_PERF_GBIF_RESERVED_58"/> - <value value="59" name="A7XX_PERF_GBIF_RESERVED_59"/> - <value value="60" name="A7XX_PERF_GBIF_RESERVED_60"/> - <value value="61" name="A7XX_PERF_GBIF_RESERVED_61"/> - <value value="62" name="A7XX_PERF_GBIF_RESERVED_62"/> - <value value="63" name="A7XX_PERF_GBIF_RESERVED_63"/> - <value value="64" name="A7XX_PERF_GBIF_RESERVED_64"/> - <value value="65" name="A7XX_PERF_GBIF_RESERVED_65"/> - <value value="66" name="A7XX_PERF_GBIF_RESERVED_66"/> - <value value="67" name="A7XX_PERF_GBIF_RESERVED_67"/> - <value value="68" name="A7XX_PERF_GBIF_CYCLES_CH0_HELD_OFF_RD_ALL"/> - <value value="69" name="A7XX_PERF_GBIF_CYCLES_CH1_HELD_OFF_RD_ALL"/> - <value value="70" name="A7XX_PERF_GBIF_CYCLES_CH0_HELD_OFF_WR_ALL"/> - <value value="71" name="A7XX_PERF_GBIF_CYCLES_CH1_HELD_OFF_WR_ALL"/> - <value value="72" name="A7XX_PERF_GBIF_AXI_CH0_REQUEST_HELD_OFF"/> - <value value="73" name="A7XX_PERF_GBIF_AXI_CH1_REQUEST_HELD_OFF"/> - <value value="74" name="A7XX_PERF_GBIF_AXI_REQUEST_HELD_OFF"/> - <value value="75" name="A7XX_PERF_GBIF_AXI_CH0_WRITE_DATA_HELD_OFF"/> - <value value="76" name="A7XX_PERF_GBIF_AXI_CH1_WRITE_DATA_HELD_OFF"/> - <value value="77" name="A7XX_PERF_GBIF_AXI_ALL_WRITE_DATA_HELD_OFF"/> - <value value="78" name="A7XX_PERF_GBIF_AXI_ALL_READ_BEATS"/> - <value value="79" name="A7XX_PERF_GBIF_AXI_ALL_WRITE_BEATS"/> - <value value="80" name="A7XX_PERF_GBIF_AXI_ALL_BEATS"/> -</enum> - -<enum name="a7xx_ufc_perfcounter_select"> - <value value="0" name="A7XX_PERF_UFC_BUSY_CYCLES"/> - <value value="1" name="A7XX_PERF_UFC_READ_DATA_VBIF"/> - <value value="2" name="A7XX_PERF_UFC_WRITE_DATA_VBIF"/> - <value value="3" name="A7XX_PERF_UFC_READ_REQUEST_VBIF"/> - <value value="4" name="A7XX_PERF_UFC_WRITE_REQUEST_VBIF"/> - <value value="5" name="A7XX_PERF_UFC_LRZ_FILTER_HIT"/> - <value value="6" name="A7XX_PERF_UFC_LRZ_FILTER_MISS"/> - <value value="7" name="A7XX_PERF_UFC_CRE_FILTER_HIT"/> - <value value="8" name="A7XX_PERF_UFC_CRE_FILTER_MISS"/> - <value value="9" name="A7XX_PERF_UFC_SP_FILTER_HIT"/> - <value value="10" name="A7XX_PERF_UFC_SP_FILTER_MISS"/> - <value value="11" name="A7XX_PERF_UFC_SP_REQUESTS"/> - <value value="12" name="A7XX_PERF_UFC_TP_FILTER_HIT"/> - <value value="13" name="A7XX_PERF_UFC_TP_FILTER_MISS"/> - <value value="14" name="A7XX_PERF_UFC_TP_REQUESTS"/> - <value value="15" name="A7XX_PERF_UFC_MAIN_HIT_LRZ_PREFETCH"/> - <value value="16" name="A7XX_PERF_UFC_MAIN_HIT_CRE_PREFETCH"/> - <value value="17" name="A7XX_PERF_UFC_MAIN_HIT_SP_PREFETCH"/> - <value value="18" name="A7XX_PERF_UFC_MAIN_HIT_TP_PREFETCH"/> - <value value="19" name="A7XX_PERF_UFC_MAIN_HIT_UBWC_READ"/> - <value value="20" name="A7XX_PERF_UFC_MAIN_HIT_UBWC_WRITE"/> - <value value="21" name="A7XX_PERF_UFC_MAIN_MISS_LRZ_PREFETCH"/> - <value value="22" name="A7XX_PERF_UFC_MAIN_MISS_CRE_PREFETCH"/> - <value value="23" name="A7XX_PERF_UFC_MAIN_MISS_SP_PREFETCH"/> - <value value="24" name="A7XX_PERF_UFC_MAIN_MISS_TP_PREFETCH"/> - <value value="25" name="A7XX_PERF_UFC_MAIN_MISS_UBWC_READ"/> - <value value="26" name="A7XX_PERF_UFC_MAIN_MISS_UBWC_WRITE"/> - <value value="27" name="A7XX_PERF_UFC_UBWC_READ_UFC_TRANS"/> - <value value="28" name="A7XX_PERF_UFC_UBWC_WRITE_UFC_TRANS"/> - <value value="29" name="A7XX_PERF_UFC_STALL_CYCLES_GBIF_CMD"/> - <value value="30" name="A7XX_PERF_UFC_STALL_CYCLES_GBIF_RDATA"/> - <value value="31" name="A7XX_PERF_UFC_STALL_CYCLES_GBIF_WDATA"/> - <value value="32" name="A7XX_PERF_UFC_STALL_CYCLES_UBWC_WR_FLAG"/> - <value value="33" name="A7XX_PERF_UFC_STALL_CYCLES_UBWC_FLAG_RTN"/> - <value value="34" name="A7XX_PERF_UFC_STALL_CYCLES_UBWC_EVENT"/> - <value value="35" name="A7XX_PERF_UFC_LRZ_PREFETCH_STALLED_CYCLES"/> - <value value="36" name="A7XX_PERF_UFC_CRE_PREFETCH_STALLED_CYCLES"/> - <value value="37" name="A7XX_PERF_UFC_SPTP_PREFETCH_STALLED_CYCLES"/> - <value value="38" name="A7XX_PERF_UFC_UBWC_RD_STALLED_CYCLES"/> - <value value="39" name="A7XX_PERF_UFC_UBWC_WR_STALLED_CYCLES"/> - <value value="40" name="A7XX_PERF_UFC_PREFETCH_STALLED_CYCLES"/> - <value value="41" name="A7XX_PERF_UFC_EVICTION_STALLED_CYCLES"/> - <value value="42" name="A7XX_PERF_UFC_LOCK_STALLED_CYCLES"/> - <value value="43" name="A7XX_PERF_UFC_MISS_LATENCY_CYCLES"/> - <value value="44" name="A7XX_PERF_UFC_MISS_LATENCY_SAMPLES"/> - <value value="45" name="A7XX_PERF_UFC_UBWC_REQ_STALLED_CYCLES"/> - <value value="46" name="A7XX_PERF_UFC_TP_HINT_TAG_MISS"/> - <value value="47" name="A7XX_PERF_UFC_TP_HINT_TAG_HIT_RDY"/> - <value value="48" name="A7XX_PERF_UFC_TP_HINT_TAG_HIT_NRDY"/> - <value value="49" name="A7XX_PERF_UFC_TP_HINT_IS_FCLEAR"/> - <value value="50" name="A7XX_PERF_UFC_TP_HINT_IS_ALPHA0"/> - <value value="51" name="A7XX_PERF_UFC_SP_L1_FILTER_HIT"/> - <value value="52" name="A7XX_PERF_UFC_SP_L1_FILTER_MISS"/> - <value value="53" name="A7XX_PERF_UFC_SP_L1_FILTER_REQUESTS"/> - <value value="54" name="A7XX_PERF_UFC_TP_L1_TAG_HIT_RDY"/> - <value value="55" name="A7XX_PERF_UFC_TP_L1_TAG_HIT_NRDY"/> - <value value="56" name="A7XX_PERF_UFC_TP_L1_TAG_MISS"/> - <value value="57" name="A7XX_PERF_UFC_TP_L1_FILTER_REQUESTS"/> -</enum> - <domain name="A6XX" width="32" prefix="variant" varset="chip"> <bitset name="A6XX_RBBM_INT_0_MASK" inline="no" varset="chip"> <bitfield name="RBBM_GPU_IDLE" pos="0" type="boolean"/> @@ -2371,7 +177,7 @@ to upconvert to 32b float internally? <reg32 offset="0x08ab" name="CP_CONTEXT_SWITCH_LEVEL_STATUS" variants="A7XX-"/> <array offset="0x08D0" name="CP_PERFCTR_CP_SEL" stride="1" length="14"/> <array offset="0x08e0" name="CP_BV_PERFCTR_CP_SEL" stride="1" length="7" variants="A7XX-"/> - <reg64 offset="0x0900" name="CP_CRASH_SCRIPT_BASE"/> + <reg64 offset="0x0900" name="CP_CRASH_DUMP_SCRIPT_BASE"/> <reg32 offset="0x0902" name="CP_CRASH_DUMP_CNTL"/> <reg32 offset="0x0903" name="CP_CRASH_DUMP_STATUS"/> <reg32 offset="0x0908" name="CP_SQE_STAT_ADDR"/> @@ -2400,22 +206,22 @@ to upconvert to 32b float internally? --> <reg64 offset="0x0934" name="CP_VSD_BASE"/> - <bitset name="a6xx_roq_stat" inline="yes"> + <bitset name="a6xx_roq_status" inline="yes"> <bitfield name="RPTR" low="0" high="9"/> <bitfield name="WPTR" low="16" high="25"/> </bitset> - <reg32 offset="0x0939" name="CP_ROQ_RB_STAT" type="a6xx_roq_stat"/> - <reg32 offset="0x093a" name="CP_ROQ_IB1_STAT" type="a6xx_roq_stat"/> - <reg32 offset="0x093b" name="CP_ROQ_IB2_STAT" type="a6xx_roq_stat"/> - <reg32 offset="0x093c" name="CP_ROQ_SDS_STAT" type="a6xx_roq_stat"/> - <reg32 offset="0x093d" name="CP_ROQ_MRB_STAT" type="a6xx_roq_stat"/> - <reg32 offset="0x093e" name="CP_ROQ_VSD_STAT" type="a6xx_roq_stat"/> - - <reg32 offset="0x0943" name="CP_IB1_DWORDS"/> - <reg32 offset="0x0944" name="CP_IB2_DWORDS"/> - <reg32 offset="0x0945" name="CP_SDS_DWORDS"/> - <reg32 offset="0x0946" name="CP_MRB_DWORDS"/> - <reg32 offset="0x0947" name="CP_VSD_DWORDS"/> + <reg32 offset="0x0939" name="CP_ROQ_RB_STATUS" type="a6xx_roq_status"/> + <reg32 offset="0x093a" name="CP_ROQ_IB1_STATUS" type="a6xx_roq_status"/> + <reg32 offset="0x093b" name="CP_ROQ_IB2_STATUS" type="a6xx_roq_status"/> + <reg32 offset="0x093c" name="CP_ROQ_SDS_STATUS" type="a6xx_roq_status"/> + <reg32 offset="0x093d" name="CP_ROQ_MRB_STATUS" type="a6xx_roq_status"/> + <reg32 offset="0x093e" name="CP_ROQ_VSD_STATUS" type="a6xx_roq_status"/> + + <reg32 offset="0x0943" name="CP_IB1_INIT_SIZE"/> + <reg32 offset="0x0944" name="CP_IB2_INIT_SIZE"/> + <reg32 offset="0x0945" name="CP_SDS_INIT_SIZE"/> + <reg32 offset="0x0946" name="CP_MRB_INIT_SIZE"/> + <reg32 offset="0x0947" name="CP_VSD_INIT_SIZE"/> <reg32 offset="0x0948" name="CP_ROQ_AVAIL_RB"> <doc>number of remaining dwords incl current dword being consumed?</doc> @@ -2451,6 +257,7 @@ to upconvert to 32b float internally? <reg32 offset="0x098D" name="CP_AHB_CNTL"/> <reg32 offset="0x0A00" name="CP_APERTURE_CNTL_HOST" variants="A6XX"/> <reg32 offset="0x0A00" name="CP_APERTURE_CNTL_HOST" type="a7xx_aperture_cntl" variants="A7XX-"/> + <reg32 offset="0x0A01" name="CP_APERTURE_CNTL_SQE" variants="A6XX"/> <reg32 offset="0x0A03" name="CP_APERTURE_CNTL_CD" variants="A6XX"/> <reg32 offset="0x0A03" name="CP_APERTURE_CNTL_CD" type="a7xx_aperture_cntl" variants="A7XX-"/> @@ -2468,8 +275,8 @@ to upconvert to 32b float internally? <reg32 offset="0x0a97" name="CP_BV_MEM_POOL_DBG_DATA" variants="A7XX-"/> <reg64 offset="0x0a98" name="CP_BV_RB_RPTR_ADDR" variants="A7XX-"/> - <reg32 offset="0x0a9a" name="CP_RESOURCE_TBL_DBG_ADDR" variants="A7XX-"/> - <reg32 offset="0x0a9b" name="CP_RESOURCE_TBL_DBG_DATA" variants="A7XX-"/> + <reg32 offset="0x0a9a" name="CP_RESOURCE_TABLE_DBG_ADDR" variants="A7XX-"/> + <reg32 offset="0x0a9b" name="CP_RESOURCE_TABLE_DBG_DATA" variants="A7XX-"/> <reg32 offset="0x0ad0" name="CP_BV_APRIV_CNTL" variants="A7XX-"/> <reg32 offset="0x0ada" name="CP_BV_CHICKEN_DBG" variants="A7XX-"/> @@ -2619,28 +426,17 @@ to upconvert to 32b float internally? vertices in, number of primnitives assembled etc. --> - <reg32 offset="0x0540" name="RBBM_PRIMCTR_0_LO"/> <!-- vs vertices in --> - <reg32 offset="0x0541" name="RBBM_PRIMCTR_0_HI"/> - <reg32 offset="0x0542" name="RBBM_PRIMCTR_1_LO"/> <!-- vs primitives out --> - <reg32 offset="0x0543" name="RBBM_PRIMCTR_1_HI"/> - <reg32 offset="0x0544" name="RBBM_PRIMCTR_2_LO"/> <!-- hs vertices in --> - <reg32 offset="0x0545" name="RBBM_PRIMCTR_2_HI"/> - <reg32 offset="0x0546" name="RBBM_PRIMCTR_3_LO"/> <!-- hs patches out --> - <reg32 offset="0x0547" name="RBBM_PRIMCTR_3_HI"/> - <reg32 offset="0x0548" name="RBBM_PRIMCTR_4_LO"/> <!-- dss vertices in --> - <reg32 offset="0x0549" name="RBBM_PRIMCTR_4_HI"/> - <reg32 offset="0x054a" name="RBBM_PRIMCTR_5_LO"/> <!-- ds primitives out --> - <reg32 offset="0x054b" name="RBBM_PRIMCTR_5_HI"/> - <reg32 offset="0x054c" name="RBBM_PRIMCTR_6_LO"/> <!-- gs primitives in --> - <reg32 offset="0x054d" name="RBBM_PRIMCTR_6_HI"/> - <reg32 offset="0x054e" name="RBBM_PRIMCTR_7_LO"/> <!-- gs primitives out --> - <reg32 offset="0x054f" name="RBBM_PRIMCTR_7_HI"/> - <reg32 offset="0x0550" name="RBBM_PRIMCTR_8_LO"/> <!-- gs primitives out --> - <reg32 offset="0x0551" name="RBBM_PRIMCTR_8_HI"/> - <reg32 offset="0x0552" name="RBBM_PRIMCTR_9_LO"/> <!-- raster primitives in --> - <reg32 offset="0x0553" name="RBBM_PRIMCTR_9_HI"/> - <reg32 offset="0x0554" name="RBBM_PRIMCTR_10_LO"/> - <reg32 offset="0x0555" name="RBBM_PRIMCTR_10_HI"/> + <reg64 offset="0x0540" name="RBBM_PIPESTAT_IAVERTICES"/> + <reg64 offset="0x0542" name="RBBM_PIPESTAT_IAPRIMITIVES"/> + <reg64 offset="0x0544" name="RBBM_PIPESTAT_VSINVOCATIONS"/> + <reg64 offset="0x0546" name="RBBM_PIPESTAT_HSINVOCATIONS"/> + <reg64 offset="0x0548" name="RBBM_PIPESTAT_DSINVOCATIONS"/> + <reg64 offset="0x054a" name="RBBM_PIPESTAT_GSINVOCATIONS"/> + <reg64 offset="0x054c" name="RBBM_PIPESTAT_GSPRIMITIVES"/> + <reg64 offset="0x054e" name="RBBM_PIPESTAT_CINVOCATIONS"/> + <reg64 offset="0x0550" name="RBBM_PIPESTAT_CPRIMITIVES"/> + <reg64 offset="0x0552" name="RBBM_PIPESTAT_PSINVOCATIONS"/> + <reg64 offset="0x0554" name="RBBM_PIPESTAT_CSINVOCATIONS"/> <reg32 offset="0xF400" name="RBBM_SECVID_TRUST_CNTL"/> <reg64 offset="0xF800" name="RBBM_SECVID_TSB_TRUSTED_BASE"/> @@ -2779,7 +575,7 @@ to upconvert to 32b float internally? <reg32 offset="0x0011f" name="RBBM_CGC_P2S_TRIG_CMD" variants="A7XX-"/> <reg32 offset="0x00120" name="RBBM_CLOCK_CNTL_TEX_FCHE"/> <reg32 offset="0x00121" name="RBBM_CLOCK_DELAY_TEX_FCHE"/> - <reg32 offset="0x00122" name="RBBM_CLOCK_HYST_TEX_FCHE"/> + <reg32 offset="0x00122" name="RBBM_CLOCK_HYST_TEX_FCHE" variants="A6XX"/> <reg32 offset="0x00122" name="RBBM_CGC_P2S_STATUS" variants="A7XX-"> <bitfield name="TXDONE" pos="0" type="boolean"/> </reg32> @@ -2840,7 +636,7 @@ to upconvert to 32b float internally? </reg32> <reg32 offset="0x062f" name="DBGC_CFG_DBGBUS_TRACE_BUF1"/> <reg32 offset="0x0630" name="DBGC_CFG_DBGBUS_TRACE_BUF2"/> - <array offset="0x0CD8" name="VSC_PERFCTR_VSC_SEL" stride="1" length="2"/> + <array offset="0x0CD8" name="VSC_PERFCTR_VSC_SEL" stride="1" length="2" variants="A6XX"/> <reg32 offset="0x0CD8" name="VSC_UNKNOWN_0CD8" variants="A7XX"> <doc> Set to true when binning, isn't changed afterwards @@ -2936,8 +732,8 @@ to upconvert to 32b float internally? <bitfield name="WIDTH" low="0" high="7" shr="5" type="uint"/> <bitfield name="HEIGHT" low="8" high="16" shr="4" type="uint"/> </reg32> - <reg64 offset="0x0c03" name="VSC_DRAW_STRM_SIZE_ADDRESS" type="waddress" usage="cmd"/> - <reg32 offset="0x0c06" name="VSC_BIN_COUNT" usage="rp_blit"> + <reg64 offset="0x0c03" name="VSC_SIZE_BASE" type="waddress" usage="cmd"/> + <reg32 offset="0x0c06" name="VSC_EXPANDED_BIN_CNTL" usage="rp_blit"> <bitfield name="NX" low="1" high="10" type="uint"/> <bitfield name="NY" low="11" high="20" type="uint"/> </reg32> @@ -2967,14 +763,14 @@ to upconvert to 32b float internally? LIMIT is set to PITCH - 64, to make room for a bit of overflow --> - <reg64 offset="0x0c30" name="VSC_PRIM_STRM_ADDRESS" type="waddress" usage="cmd"/> - <reg32 offset="0x0c32" name="VSC_PRIM_STRM_PITCH" usage="cmd"/> - <reg32 offset="0x0c33" name="VSC_PRIM_STRM_LIMIT" usage="cmd"/> - <reg64 offset="0x0c34" name="VSC_DRAW_STRM_ADDRESS" type="waddress" usage="cmd"/> - <reg32 offset="0x0c36" name="VSC_DRAW_STRM_PITCH" usage="cmd"/> - <reg32 offset="0x0c37" name="VSC_DRAW_STRM_LIMIT" usage="cmd"/> - - <array offset="0x0c38" name="VSC_STATE" stride="1" length="32" usage="rp_blit"> + <reg64 offset="0x0c30" name="VSC_PIPE_DATA_PRIM_BASE" type="waddress" usage="cmd"/> + <reg32 offset="0x0c32" name="VSC_PIPE_DATA_PRIM_STRIDE" usage="cmd"/> + <reg32 offset="0x0c33" name="VSC_PIPE_DATA_PRIM_LENGTH" usage="cmd"/> + <reg64 offset="0x0c34" name="VSC_PIPE_DATA_DRAW_BASE" type="waddress" usage="cmd"/> + <reg32 offset="0x0c36" name="VSC_PIPE_DATA_DRAW_STRIDE" usage="cmd"/> + <reg32 offset="0x0c37" name="VSC_PIPE_DATA_DRAW_LENGTH" usage="cmd"/> + + <array offset="0x0c38" name="VSC_CHANNEL_VISIBILITY" stride="1" length="32" usage="rp_blit"> <doc> Seems to be a bitmap of which tiles mapped to the VSC pipe contain geometry. @@ -2985,7 +781,7 @@ to upconvert to 32b float internally? <reg32 offset="0x0" name="REG"/> </array> - <array offset="0x0c58" name="VSC_PRIM_STRM_SIZE" stride="1" length="32" variants="A6XX" usage="rp_blit"> + <array offset="0x0c58" name="VSC_PIPE_DATA_PRIM_SIZE" stride="1" length="32" variants="A6XX" usage="rp_blit"> <doc> Has the size of data written to corresponding VSC_PRIM_STRM buffer. @@ -2993,10 +789,10 @@ to upconvert to 32b float internally? <reg32 offset="0x0" name="REG"/> </array> - <array offset="0x0c78" name="VSC_DRAW_STRM_SIZE" stride="1" length="32" variants="A6XX" usage="rp_blit"> + <array offset="0x0c78" name="VSC_PIPE_DATA_DRAW_SIZE" stride="1" length="32" variants="A6XX" usage="rp_blit"> <doc> Has the size of data written to corresponding VSC pipe, ie. - same thing that is written out to VSC_DRAW_STRM_SIZE_ADDRESS_LO/HI + same thing that is written out to VSC_SIZE_BASE </doc> <reg32 offset="0x0" name="REG"/> </array> @@ -3028,17 +824,17 @@ to upconvert to 32b float internally? <bitfield name="PERSP_DIVISION_DISABLE" pos="9" type="boolean"/> </reg32> - <bitset name="a6xx_gras_xs_cl_cntl" inline="yes"> + <bitset name="a6xx_gras_xs_clip_cull_distance" inline="yes"> <bitfield name="CLIP_MASK" low="0" high="7"/> <bitfield name="CULL_MASK" low="8" high="15"/> </bitset> - <reg32 offset="0x8001" name="GRAS_VS_CL_CNTL" type="a6xx_gras_xs_cl_cntl" usage="rp_blit"/> - <reg32 offset="0x8002" name="GRAS_DS_CL_CNTL" type="a6xx_gras_xs_cl_cntl" usage="rp_blit"/> - <reg32 offset="0x8003" name="GRAS_GS_CL_CNTL" type="a6xx_gras_xs_cl_cntl" usage="rp_blit"/> - <reg32 offset="0x8004" name="GRAS_MAX_LAYER_INDEX" low="0" high="10" type="uint" usage="rp_blit"/> + <reg32 offset="0x8001" name="GRAS_CL_VS_CLIP_CULL_DISTANCE" type="a6xx_gras_xs_clip_cull_distance" usage="rp_blit"/> + <reg32 offset="0x8002" name="GRAS_CL_DS_CLIP_CULL_DISTANCE" type="a6xx_gras_xs_clip_cull_distance" usage="rp_blit"/> + <reg32 offset="0x8003" name="GRAS_CL_GS_CLIP_CULL_DISTANCE" type="a6xx_gras_xs_clip_cull_distance" usage="rp_blit"/> + <reg32 offset="0x8004" name="GRAS_CL_ARRAY_SIZE" low="0" high="10" type="uint" usage="rp_blit"/> - <reg32 offset="0x8005" name="GRAS_CNTL" usage="rp_blit"> - <!-- see also RB_RENDER_CONTROL0 --> + <reg32 offset="0x8005" name="GRAS_CL_INTERP_CNTL" usage="rp_blit"> + <!-- see also RB_INTERP_CNTL --> <bitfield name="IJ_PERSP_PIXEL" pos="0" type="boolean"/> <bitfield name="IJ_PERSP_CENTROID" pos="1" type="boolean"/> <bitfield name="IJ_PERSP_SAMPLE" pos="2" type="boolean"/> @@ -3067,7 +863,7 @@ to upconvert to 32b float internally? <!-- <reg32 offset="0x80f0" name="GRAS_UNKNOWN_80F0" type="a6xx_reg_xy"/> --> <!-- 0x8006-0x800f invalid --> - <array offset="0x8010" name="GRAS_CL_VPORT" stride="6" length="16" usage="rp_blit"> + <array offset="0x8010" name="GRAS_CL_VIEWPORT" stride="6" length="16" usage="rp_blit"> <reg32 offset="0" name="XOFFSET" type="float"/> <reg32 offset="1" name="XSCALE" type="float"/> <reg32 offset="2" name="YOFFSET" type="float"/> @@ -3075,7 +871,7 @@ to upconvert to 32b float internally? <reg32 offset="4" name="ZOFFSET" type="float"/> <reg32 offset="5" name="ZSCALE" type="float"/> </array> - <array offset="0x8070" name="GRAS_CL_Z_CLAMP" stride="2" length="16" usage="rp_blit"> + <array offset="0x8070" name="GRAS_CL_VIEWPORT_ZCLAMP" stride="2" length="16" usage="rp_blit"> <reg32 offset="0" name="MIN" type="float"/> <reg32 offset="1" name="MAX" type="float"/> </array> @@ -3124,7 +920,12 @@ to upconvert to 32b float internally? <reg32 offset="0x8099" name="GRAS_SU_CONSERVATIVE_RAS_CNTL" usage="cmd"> <bitfield name="CONSERVATIVERASEN" pos="0" type="boolean"/> - <bitfield name="SHIFTAMOUNT" low="1" high="2"/> + <enum name="a6xx_shift_amount"> + <value value="0" name="NO_SHIFT"/> + <value value="1" name="HALF_PIXEL_SHIFT"/> + <value value="2" name="FULL_PIXEL_SHIFT"/> + </enum> + <bitfield name="SHIFTAMOUNT" low="1" high="2" type="a6xx_shift_amount"/> <bitfield name="INNERCONSERVATIVERASEN" pos="3" type="boolean"/> <bitfield name="UNK4" low="4" high="5"/> </reg32> @@ -3133,13 +934,13 @@ to upconvert to 32b float internally? <bitfield name="LINELENGTHEN" pos="1" type="boolean"/> </reg32> - <bitset name="a6xx_gras_layer_cntl" inline="yes"> + <bitset name="a6xx_gras_us_xs_siv_cntl" inline="yes"> <bitfield name="WRITES_LAYER" pos="0" type="boolean"/> <bitfield name="WRITES_VIEW" pos="1" type="boolean"/> </bitset> - <reg32 offset="0x809b" name="GRAS_VS_LAYER_CNTL" type="a6xx_gras_layer_cntl" usage="rp_blit"/> - <reg32 offset="0x809c" name="GRAS_GS_LAYER_CNTL" type="a6xx_gras_layer_cntl" usage="rp_blit"/> - <reg32 offset="0x809d" name="GRAS_DS_LAYER_CNTL" type="a6xx_gras_layer_cntl" usage="rp_blit"/> + <reg32 offset="0x809b" name="GRAS_SU_VS_SIV_CNTL" type="a6xx_gras_us_xs_siv_cntl" usage="rp_blit"/> + <reg32 offset="0x809c" name="GRAS_SU_GS_SIV_CNTL" type="a6xx_gras_us_xs_siv_cntl" usage="rp_blit"/> + <reg32 offset="0x809d" name="GRAS_SU_DS_SIV_CNTL" type="a6xx_gras_us_xs_siv_cntl" usage="rp_blit"/> <!-- 0x809e/0x809f invalid --> <enum name="a6xx_sequenced_thread_dist"> @@ -3213,13 +1014,13 @@ to upconvert to 32b float internally? <enum name="a6xx_lrz_feedback_mask"> <value value="0x0" name="LRZ_FEEDBACK_NONE"/> <value value="0x1" name="LRZ_FEEDBACK_EARLY_Z"/> - <value value="0x2" name="LRZ_FEEDBACK_EARLY_LRZ_LATE_Z"/> + <value value="0x2" name="LRZ_FEEDBACK_EARLY_Z_LATE_Z"/> <!-- We don't have a flag type and this flags combination is often used --> - <value value="0x3" name="LRZ_FEEDBACK_EARLY_Z_OR_EARLY_LRZ_LATE_Z"/> + <value value="0x3" name="LRZ_FEEDBACK_EARLY_Z_OR_EARLY_Z_LATE_Z"/> <value value="0x4" name="LRZ_FEEDBACK_LATE_Z"/> </enum> - <reg32 offset="0x80a1" name="GRAS_BIN_CONTROL" usage="rp_blit"> + <reg32 offset="0x80a1" name="GRAS_SC_BIN_CNTL" usage="rp_blit"> <bitfield name="BINW" low="0" high="5" shr="5" type="uint"/> <bitfield name="BINH" low="8" high="14" shr="4" type="uint"/> <bitfield name="RENDER_MODE" low="18" high="20" type="a6xx_render_mode"/> @@ -3235,22 +1036,22 @@ to upconvert to 32b float internally? <bitfield name="UNK27" pos="27"/> </reg32> - <reg32 offset="0x80a2" name="GRAS_RAS_MSAA_CNTL" usage="rp_blit"> + <reg32 offset="0x80a2" name="GRAS_SC_RAS_MSAA_CNTL" usage="rp_blit"> <bitfield name="SAMPLES" low="0" high="1" type="a3xx_msaa_samples"/> <bitfield name="UNK2" pos="2"/> <bitfield name="UNK3" pos="3"/> </reg32> - <reg32 offset="0x80a3" name="GRAS_DEST_MSAA_CNTL" usage="rp_blit"> + <reg32 offset="0x80a3" name="GRAS_SC_DEST_MSAA_CNTL" usage="rp_blit"> <bitfield name="SAMPLES" low="0" high="1" type="a3xx_msaa_samples"/> <bitfield name="MSAA_DISABLE" pos="2" type="boolean"/> </reg32> - <bitset name="a6xx_sample_config" inline="yes"> + <bitset name="a6xx_msaa_sample_pos_cntl" inline="yes"> <bitfield name="UNK0" pos="0"/> <bitfield name="LOCATION_ENABLE" pos="1" type="boolean"/> </bitset> - <bitset name="a6xx_sample_locations" inline="yes"> + <bitset name="a6xx_programmable_msaa_pos" inline="yes"> <bitfield name="SAMPLE_0_X" low="0" high="3" radix="4" type="fixed"/> <bitfield name="SAMPLE_0_Y" low="4" high="7" radix="4" type="fixed"/> <bitfield name="SAMPLE_1_X" low="8" high="11" radix="4" type="fixed"/> @@ -3261,9 +1062,9 @@ to upconvert to 32b float internally? <bitfield name="SAMPLE_3_Y" low="28" high="31" radix="4" type="fixed"/> </bitset> - <reg32 offset="0x80a4" name="GRAS_SAMPLE_CONFIG" type="a6xx_sample_config" usage="rp_blit"/> - <reg32 offset="0x80a5" name="GRAS_SAMPLE_LOCATION_0" type="a6xx_sample_locations" usage="rp_blit"/> - <reg32 offset="0x80a6" name="GRAS_SAMPLE_LOCATION_1" type="a6xx_sample_locations" usage="rp_blit"/> + <reg32 offset="0x80a4" name="GRAS_SC_MSAA_SAMPLE_POS_CNTL" type="a6xx_msaa_sample_pos_cntl" usage="rp_blit"/> + <reg32 offset="0x80a5" name="GRAS_SC_PROGRAMMABLE_MSAA_POS_0" type="a6xx_programmable_msaa_pos" usage="rp_blit"/> + <reg32 offset="0x80a6" name="GRAS_SC_PROGRAMMABLE_MSAA_POS_1" type="a6xx_programmable_msaa_pos" usage="rp_blit"/> <reg32 offset="0x80a7" name="GRAS_UNKNOWN_80A7" variants="A7XX-" usage="cmd"/> @@ -3286,13 +1087,36 @@ to upconvert to 32b float internally? <reg32 offset="0x80f0" name="GRAS_SC_WINDOW_SCISSOR_TL" type="a6xx_reg_xy" usage="rp_blit"/> <reg32 offset="0x80f1" name="GRAS_SC_WINDOW_SCISSOR_BR" type="a6xx_reg_xy" usage="rp_blit"/> - <!-- 0x80f4 - 0x80fa are used for VK_KHR_fragment_shading_rate --> - <reg64 offset="0x80f4" name="GRAS_UNKNOWN_80F4" variants="A7XX-" usage="cmd"/> - <reg64 offset="0x80f5" name="GRAS_UNKNOWN_80F5" variants="A7XX-" usage="cmd"/> - <reg64 offset="0x80f6" name="GRAS_UNKNOWN_80F6" variants="A7XX-" usage="cmd"/> - <reg64 offset="0x80f8" name="GRAS_UNKNOWN_80F8" variants="A7XX-" usage="cmd"/> - <reg64 offset="0x80f9" name="GRAS_UNKNOWN_80F9" variants="A7XX-" usage="cmd"/> - <reg64 offset="0x80fa" name="GRAS_UNKNOWN_80FA" variants="A7XX-" usage="cmd"/> + <enum name="a6xx_fsr_combiner"> + <value value="0" name="FSR_COMBINER_OP_KEEP"/> + <value value="1" name="FSR_COMBINER_OP_REPLACE"/> + <value value="2" name="FSR_COMBINER_OP_MIN"/> + <value value="3" name="FSR_COMBINER_OP_MAX"/> + <value value="4" name="FSR_COMBINER_OP_MUL"/> + </enum> + + <reg32 offset="0x80f4" name="GRAS_VRS_CONFIG" variants="A7XX-" usage="rp_blit"> + <bitfield name="PIPELINE_FSR_ENABLE" pos="0" type="boolean"/> + <bitfield name="FRAG_SIZE_X" low="1" high="2" type="uint"/> + <bitfield name="FRAG_SIZE_Y" low="3" high="4" type="uint"/> + <bitfield name="COMBINER_OP_1" low="5" high="7" type="a6xx_fsr_combiner"/> + <bitfield name="COMBINER_OP_2" low="8" high="10" type="a6xx_fsr_combiner"/> + <bitfield name="ATTACHMENT_FSR_ENABLE" pos="13" type="boolean"/> + <bitfield name="PRIMITIVE_FSR_ENABLE" pos="20" type="boolean"/> + </reg32> + <reg32 offset="0x80f5" name="GRAS_QUALITY_BUFFER_INFO" variants="A7XX-" usage="rp_blit"> + <bitfield name="LAYERED" pos="0" type="boolean"/> + <bitfield name="TILE_MODE" low="1" high="2" type="a6xx_tile_mode"/> + </reg32> + <reg32 offset="0x80f6" name="GRAS_QUALITY_BUFFER_DIMENSION" variants="A7XX-" usage="rp_blit"> + <bitfield name="WIDTH" low="0" high="15" type="uint"/> + <bitfield name="HEIGHT" low="16" high="31" type="uint"/> + </reg32> + <reg64 offset="0x80f8" name="GRAS_QUALITY_BUFFER_BASE" variants="A7XX-" type="waddress" usage="rp_blit"/> + <reg32 offset="0x80fa" name="GRAS_QUALITY_BUFFER_PITCH" variants="A7XX-" usage="rp_blit"> + <bitfield name="PITCH" shr="6" low="0" high="7" type="uint"/> + <bitfield name="ARRAY_PITCH" shr="6" low="10" high="28" type="uint"/> + </reg32> <enum name="a6xx_lrz_dir_status"> <value value="0x1" name="LRZ_DIR_LE"/> @@ -3313,7 +1137,7 @@ to upconvert to 32b float internally? </doc> <bitfield name="FC_ENABLE" pos="3" type="boolean" variants="A6XX"/> <!-- set when depth-test + depth-write enabled --> - <bitfield name="Z_TEST_ENABLE" pos="4" type="boolean"/> + <bitfield name="Z_WRITE_ENABLE" pos="4" type="boolean"/> <bitfield name="Z_BOUNDS_ENABLE" pos="5" type="boolean"/> <bitfield name="DIR" low="6" high="7" type="a6xx_lrz_dir_status"/> <doc> @@ -3339,14 +1163,13 @@ to upconvert to 32b float internally? <bitfield name="FRAGCOORDSAMPLEMODE" low="1" high="2" type="a6xx_fragcoord_sample_mode"/> </reg32> - <reg32 offset="0x8102" name="GRAS_LRZ_MRT_BUF_INFO_0" usage="rp_blit"> + <reg32 offset="0x8102" name="GRAS_LRZ_MRT_BUFFER_INFO_0" usage="rp_blit"> <bitfield name="COLOR_FORMAT" low="0" high="7" type="a6xx_format"/> </reg32> <reg64 offset="0x8103" name="GRAS_LRZ_BUFFER_BASE" align="256" type="waddress" usage="rp_blit"/> <reg32 offset="0x8105" name="GRAS_LRZ_BUFFER_PITCH" usage="rp_blit"> - <!-- TODO: fix the shr fields --> <bitfield name="PITCH" low="0" high="7" shr="5" type="uint"/> - <bitfield name="ARRAY_PITCH" low="10" high="28" shr="4" type="uint"/> + <bitfield name="ARRAY_PITCH" low="10" high="28" shr="8" type="uint"/> </reg32> <!-- @@ -3381,18 +1204,18 @@ to upconvert to 32b float internally? --> <reg64 offset="0x8106" name="GRAS_LRZ_FAST_CLEAR_BUFFER_BASE" align="64" type="waddress" usage="rp_blit"/> <!-- 0x8108 invalid --> - <reg32 offset="0x8109" name="GRAS_SAMPLE_CNTL" usage="rp_blit"> + <reg32 offset="0x8109" name="GRAS_LRZ_PS_SAMPLEFREQ_CNTL" usage="rp_blit"> <bitfield name="PER_SAMP_MODE" pos="0" type="boolean"/> </reg32> <!-- LRZ buffer represents a single array layer + mip level, and there is a single buffer per depth image. Thus to reuse LRZ between renderpasses it is necessary to track the depth view used in the past renderpass, which - GRAS_LRZ_DEPTH_VIEW is for. - GRAS_LRZ_CNTL checks if current value of GRAS_LRZ_DEPTH_VIEW is equal to + GRAS_LRZ_VIEW_INFO is for. + GRAS_LRZ_CNTL checks if current value of GRAS_LRZ_VIEW_INFO is equal to the value stored in the LRZ buffer, if not - LRZ is disabled. --> - <reg32 offset="0x810a" name="GRAS_LRZ_DEPTH_VIEW" usage="cmd"> + <reg32 offset="0x810a" name="GRAS_LRZ_VIEW_INFO" usage="cmd"> <bitfield name="BASE_LAYER" low="0" high="10" type="uint"/> <bitfield name="LAYER_COUNT" low="16" high="26" type="uint"/> <bitfield name="BASE_MIP_LEVEL" low="28" high="31" type="uint"/> @@ -3408,7 +1231,7 @@ to upconvert to 32b float internally? <reg32 offset="0x8110" name="GRAS_UNKNOWN_8110" low="0" high="1" usage="cmd"/> <!-- A bit tentative but it's a color and it is followed by LRZ_CLEAR --> - <reg32 offset="0x8111" name="GRAS_LRZ_CLEAR_DEPTH_F32" type="float" variants="A7XX-"/> + <reg32 offset="0x8111" name="GRAS_LRZ_DEPTH_CLEAR" type="float" variants="A7XX-"/> <reg32 offset="0x8113" name="GRAS_LRZ_DEPTH_BUFFER_INFO" variants="A7XX-" usage="rp_blit"> <bitfield name="DEPTH_FORMAT" low="0" high="2" type="a6xx_depth_format"/> @@ -3430,7 +1253,7 @@ to upconvert to 32b float internally? <value value="0x5" name="ROTATE_VFLIP"/> </enum> - <bitset name="a6xx_2d_blit_cntl" inline="yes"> + <bitset name="a6xx_a2d_bit_cntl" inline="yes"> <bitfield name="ROTATE" low="0" high="2" type="a6xx_rotation"/> <bitfield name="OVERWRITEEN" pos="3" type="boolean"/> <bitfield name="UNK4" low="4" high="6"/> @@ -3447,22 +1270,22 @@ to upconvert to 32b float internally? <bitfield name="UNK30" pos="30" type="boolean" variants="A7XX-"/> </bitset> - <reg32 offset="0x8400" name="GRAS_2D_BLIT_CNTL" type="a6xx_2d_blit_cntl" usage="rp_blit"/> + <reg32 offset="0x8400" name="GRAS_A2D_BLT_CNTL" type="a6xx_a2d_bit_cntl" usage="rp_blit"/> <!-- note: the low 8 bits for src coords are valid, probably fixed point it would be a bit weird though, since we subtract 1 from BR coords apparently signed, gallium driver uses negative coords and it works? --> - <reg32 offset="0x8401" name="GRAS_2D_SRC_TL_X" low="8" high="24" type="int" usage="rp_blit"/> - <reg32 offset="0x8402" name="GRAS_2D_SRC_BR_X" low="8" high="24" type="int" usage="rp_blit"/> - <reg32 offset="0x8403" name="GRAS_2D_SRC_TL_Y" low="8" high="24" type="int" usage="rp_blit"/> - <reg32 offset="0x8404" name="GRAS_2D_SRC_BR_Y" low="8" high="24" type="int" usage="rp_blit"/> - <reg32 offset="0x8405" name="GRAS_2D_DST_TL" type="a6xx_reg_xy" usage="rp_blit"/> - <reg32 offset="0x8406" name="GRAS_2D_DST_BR" type="a6xx_reg_xy" usage="rp_blit"/> + <reg32 offset="0x8401" name="GRAS_A2D_SRC_XMIN" low="8" high="24" type="int" usage="rp_blit"/> + <reg32 offset="0x8402" name="GRAS_A2D_SRC_XMAX" low="8" high="24" type="int" usage="rp_blit"/> + <reg32 offset="0x8403" name="GRAS_A2D_SRC_YMIN" low="8" high="24" type="int" usage="rp_blit"/> + <reg32 offset="0x8404" name="GRAS_A2D_SRC_YMAX" low="8" high="24" type="int" usage="rp_blit"/> + <reg32 offset="0x8405" name="GRAS_A2D_DEST_TL" type="a6xx_reg_xy" usage="rp_blit"/> + <reg32 offset="0x8406" name="GRAS_A2D_DEST_BR" type="a6xx_reg_xy" usage="rp_blit"/> <reg32 offset="0x8407" name="GRAS_2D_UNKNOWN_8407" low="0" high="31"/> <reg32 offset="0x8408" name="GRAS_2D_UNKNOWN_8408" low="0" high="31"/> <reg32 offset="0x8409" name="GRAS_2D_UNKNOWN_8409" low="0" high="31"/> - <reg32 offset="0x840a" name="GRAS_2D_RESOLVE_CNTL_1" type="a6xx_reg_xy" usage="rp_blit"/> - <reg32 offset="0x840b" name="GRAS_2D_RESOLVE_CNTL_2" type="a6xx_reg_xy" usage="rp_blit"/> + <reg32 offset="0x840a" name="GRAS_A2D_SCISSOR_TL" type="a6xx_reg_xy" usage="rp_blit"/> + <reg32 offset="0x840b" name="GRAS_A2D_SCISSOR_BR" type="a6xx_reg_xy" usage="rp_blit"/> <!-- 0x840c-0x85ff invalid --> <!-- always 0x880 ? (and 0 in a640/a650 traces?) --> @@ -3481,7 +1304,7 @@ to upconvert to 32b float internally? --> <!-- same as GRAS_BIN_CONTROL, but without bit 27: --> - <reg32 offset="0x8800" name="RB_BIN_CONTROL" variants="A6XX" usage="rp_blit"> + <reg32 offset="0x8800" name="RB_CNTL" variants="A6XX" usage="rp_blit"> <bitfield name="BINW" low="0" high="5" shr="5" type="uint"/> <bitfield name="BINH" low="8" high="14" shr="4" type="uint"/> <bitfield name="RENDER_MODE" low="18" high="20" type="a6xx_render_mode"/> @@ -3490,7 +1313,7 @@ to upconvert to 32b float internally? <bitfield name="LRZ_FEEDBACK_ZMODE_MASK" low="24" high="26" type="a6xx_lrz_feedback_mask"/> </reg32> - <reg32 offset="0x8800" name="RB_BIN_CONTROL" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0x8800" name="RB_CNTL" variants="A7XX-" usage="rp_blit"> <bitfield name="BINW" low="0" high="5" shr="5" type="uint"/> <bitfield name="BINH" low="8" high="14" shr="4" type="uint"/> <bitfield name="RENDER_MODE" low="18" high="20" type="a6xx_render_mode"/> @@ -3501,8 +1324,7 @@ to upconvert to 32b float internally? <reg32 offset="0x8801" name="RB_RENDER_CNTL" variants="A6XX" usage="rp_blit"> <bitfield name="CCUSINGLECACHELINESIZE" low="3" high="5"/> <bitfield name="EARLYVIZOUTEN" pos="6" type="boolean"/> - <!-- set during binning pass: --> - <bitfield name="BINNING" pos="7" type="boolean"/> + <bitfield name="FS_DISABLE" pos="7" type="boolean"/> <bitfield name="UNK8" low="8" high="10"/> <bitfield name="RASTER_MODE" pos="8" type="a6xx_raster_mode"/> <bitfield name="RASTER_DIRECTION" low="9" high="10" type="a6xx_raster_direction"/> @@ -3515,15 +1337,14 @@ to upconvert to 32b float internally? </reg32> <reg32 offset="0x8801" name="RB_RENDER_CNTL" variants="A7XX-" usage="rp_blit"> <bitfield name="EARLYVIZOUTEN" pos="6" type="boolean"/> - <!-- set during binning pass: --> - <bitfield name="BINNING" pos="7" type="boolean"/> + <bitfield name="FS_DISABLE" pos="7" type="boolean"/> <bitfield name="RASTER_MODE" pos="8" type="a6xx_raster_mode"/> <bitfield name="RASTER_DIRECTION" low="9" high="10" type="a6xx_raster_direction"/> <bitfield name="CONSERVATIVERASEN" pos="11" type="boolean"/> <bitfield name="INNERCONSERVATIVERASEN" pos="12" type="boolean"/> </reg32> <reg32 offset="0x8116" name="GRAS_SU_RENDER_CNTL" variants="A7XX-" usage="rp_blit"> - <bitfield name="BINNING" pos="7" type="boolean"/> + <bitfield name="FS_DISABLE" pos="7" type="boolean"/> </reg32> <reg32 offset="0x8802" name="RB_RAS_MSAA_CNTL" usage="rp_blit"> @@ -3536,16 +1357,16 @@ to upconvert to 32b float internally? <bitfield name="MSAA_DISABLE" pos="2" type="boolean"/> </reg32> - <reg32 offset="0x8804" name="RB_SAMPLE_CONFIG" type="a6xx_sample_config" usage="rp_blit"/> - <reg32 offset="0x8805" name="RB_SAMPLE_LOCATION_0" type="a6xx_sample_locations" usage="rp_blit"/> - <reg32 offset="0x8806" name="RB_SAMPLE_LOCATION_1" type="a6xx_sample_locations" usage="rp_blit"/> + <reg32 offset="0x8804" name="RB_MSAA_SAMPLE_POS_CNTL" type="a6xx_msaa_sample_pos_cntl" usage="rp_blit"/> + <reg32 offset="0x8805" name="RB_PROGRAMMABLE_MSAA_POS_0" type="a6xx_programmable_msaa_pos" usage="rp_blit"/> + <reg32 offset="0x8806" name="RB_PROGRAMMABLE_MSAA_POS_1" type="a6xx_programmable_msaa_pos" usage="rp_blit"/> <!-- 0x8807-0x8808 invalid --> <!-- note: maybe not actually called RB_RENDER_CONTROLn (since RB_RENDER_CNTL name comes from kernel and is probably right) --> - <reg32 offset="0x8809" name="RB_RENDER_CONTROL0" usage="rp_blit"> - <!-- see also GRAS_CNTL --> + <reg32 offset="0x8809" name="RB_INTERP_CNTL" usage="rp_blit"> + <!-- see also GRAS_CL_INTERP_CNTL --> <bitfield name="IJ_PERSP_PIXEL" pos="0" type="boolean"/> <bitfield name="IJ_PERSP_CENTROID" pos="1" type="boolean"/> <bitfield name="IJ_PERSP_SAMPLE" pos="2" type="boolean"/> @@ -3555,7 +1376,7 @@ to upconvert to 32b float internally? <bitfield name="COORD_MASK" low="6" high="9" type="hex"/> <bitfield name="UNK10" pos="10" type="boolean"/> </reg32> - <reg32 offset="0x880a" name="RB_RENDER_CONTROL1" usage="rp_blit"> + <reg32 offset="0x880a" name="RB_PS_INPUT_CNTL" usage="rp_blit"> <!-- enable bits for various FS sysvalue regs: --> <bitfield name="SAMPLEMASK" pos="0" type="boolean"/> <bitfield name="POSTDEPTHCOVERAGE" pos="1" type="boolean"/> @@ -3567,16 +1388,16 @@ to upconvert to 32b float internally? <bitfield name="FOVEATION" pos="8" type="boolean"/> </reg32> - <reg32 offset="0x880b" name="RB_FS_OUTPUT_CNTL0" usage="rp_blit"> + <reg32 offset="0x880b" name="RB_PS_OUTPUT_CNTL" usage="rp_blit"> <bitfield name="DUAL_COLOR_IN_ENABLE" pos="0" type="boolean"/> <bitfield name="FRAG_WRITES_Z" pos="1" type="boolean"/> <bitfield name="FRAG_WRITES_SAMPMASK" pos="2" type="boolean"/> <bitfield name="FRAG_WRITES_STENCILREF" pos="3" type="boolean"/> </reg32> - <reg32 offset="0x880c" name="RB_FS_OUTPUT_CNTL1" usage="rp_blit"> + <reg32 offset="0x880c" name="RB_PS_MRT_CNTL" usage="rp_blit"> <bitfield name="MRT" low="0" high="3" type="uint"/> </reg32> - <reg32 offset="0x880d" name="RB_RENDER_COMPONENTS" usage="rp_blit"> + <reg32 offset="0x880d" name="RB_PS_OUTPUT_MASK" usage="rp_blit"> <bitfield name="RT0" low="0" high="3"/> <bitfield name="RT1" low="4" high="7"/> <bitfield name="RT2" low="8" high="11"/> @@ -3608,7 +1429,7 @@ to upconvert to 32b float internally? <bitfield name="SRGB_MRT7" pos="7" type="boolean"/> </reg32> - <reg32 offset="0x8810" name="RB_SAMPLE_CNTL" usage="rp_blit"> + <reg32 offset="0x8810" name="RB_PS_SAMPLEFREQ_CNTL" usage="rp_blit"> <bitfield name="PER_SAMP_MODE" pos="0" type="boolean"/> </reg32> <reg32 offset="0x8811" name="RB_UNKNOWN_8811" low="4" high="6" usage="cmd"/> @@ -3672,18 +1493,18 @@ to upconvert to 32b float internally? <reg32 offset="0x7" name="BASE_GMEM" low="12" high="31" shr="12"/> </array> - <reg32 offset="0x8860" name="RB_BLEND_RED_F32" type="float" usage="rp_blit"/> - <reg32 offset="0x8861" name="RB_BLEND_GREEN_F32" type="float" usage="rp_blit"/> - <reg32 offset="0x8862" name="RB_BLEND_BLUE_F32" type="float" usage="rp_blit"/> - <reg32 offset="0x8863" name="RB_BLEND_ALPHA_F32" type="float" usage="rp_blit"/> - <reg32 offset="0x8864" name="RB_ALPHA_CONTROL" usage="cmd"> + <reg32 offset="0x8860" name="RB_BLEND_CONSTANT_RED_FP32" type="float" usage="rp_blit"/> + <reg32 offset="0x8861" name="RB_BLEND_CONSTANT_GREEN_FP32" type="float" usage="rp_blit"/> + <reg32 offset="0x8862" name="RB_BLEND_CONSTANT_BLUE_FP32" type="float" usage="rp_blit"/> + <reg32 offset="0x8863" name="RB_BLEND_CONSTANT_ALPHA_FP32" type="float" usage="rp_blit"/> + <reg32 offset="0x8864" name="RB_ALPHA_TEST_CNTL" usage="cmd"> <bitfield name="ALPHA_REF" low="0" high="7" type="hex"/> <bitfield name="ALPHA_TEST" pos="8" type="boolean"/> <bitfield name="ALPHA_TEST_FUNC" low="9" high="11" type="adreno_compare_func"/> </reg32> <reg32 offset="0x8865" name="RB_BLEND_CNTL" usage="rp_blit"> <!-- per-mrt enable bit --> - <bitfield name="ENABLE_BLEND" low="0" high="7"/> + <bitfield name="BLEND_READS_DEST" low="0" high="7"/> <bitfield name="INDEPENDENT_BLEND" pos="8" type="boolean"/> <bitfield name="DUAL_COLOR_IN_ENABLE" pos="9" type="boolean"/> <bitfield name="ALPHA_TO_COVERAGE" pos="10" type="boolean"/> @@ -3726,12 +1547,12 @@ to upconvert to 32b float internally? <reg32 offset="0x8873" name="RB_DEPTH_BUFFER_PITCH" low="0" high="13" shr="6" type="uint" usage="rp_blit"/> <reg32 offset="0x8874" name="RB_DEPTH_BUFFER_ARRAY_PITCH" low="0" high="27" shr="6" type="uint" usage="rp_blit"/> <reg64 offset="0x8875" name="RB_DEPTH_BUFFER_BASE" type="waddress" align="64" usage="rp_blit"/> - <reg32 offset="0x8877" name="RB_DEPTH_BUFFER_BASE_GMEM" low="12" high="31" shr="12" usage="rp_blit"/> + <reg32 offset="0x8877" name="RB_DEPTH_GMEM_BASE" low="12" high="31" shr="12" usage="rp_blit"/> - <reg32 offset="0x8878" name="RB_Z_BOUNDS_MIN" type="float" usage="rp_blit"/> - <reg32 offset="0x8879" name="RB_Z_BOUNDS_MAX" type="float" usage="rp_blit"/> + <reg32 offset="0x8878" name="RB_DEPTH_BOUND_MIN" type="float" usage="rp_blit"/> + <reg32 offset="0x8879" name="RB_DEPTH_BOUND_MAX" type="float" usage="rp_blit"/> <!-- 0x887a-0x887f invalid --> - <reg32 offset="0x8880" name="RB_STENCIL_CONTROL" usage="rp_blit"> + <reg32 offset="0x8880" name="RB_STENCIL_CNTL" usage="rp_blit"> <bitfield name="STENCIL_ENABLE" pos="0" type="boolean"/> <bitfield name="STENCIL_ENABLE_BF" pos="1" type="boolean"/> <!-- @@ -3753,11 +1574,11 @@ to upconvert to 32b float internally? <reg32 offset="0x8115" name="GRAS_SU_STENCIL_CNTL" usage="rp_blit"> <bitfield name="STENCIL_ENABLE" pos="0" type="boolean"/> </reg32> - <reg32 offset="0x8881" name="RB_STENCIL_INFO" variants="A6XX" usage="rp_blit"> + <reg32 offset="0x8881" name="RB_STENCIL_BUFFER_INFO" variants="A6XX" usage="rp_blit"> <bitfield name="SEPARATE_STENCIL" pos="0" type="boolean"/> <bitfield name="UNK1" pos="1" type="boolean"/> </reg32> - <reg32 offset="0x8881" name="RB_STENCIL_INFO" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0x8881" name="RB_STENCIL_BUFFER_INFO" variants="A7XX-" usage="rp_blit"> <bitfield name="SEPARATE_STENCIL" pos="0" type="boolean"/> <bitfield name="UNK1" pos="1" type="boolean"/> <bitfield name="TILEMODE" low="2" high="3" type="a6xx_tile_mode"/> @@ -3765,22 +1586,22 @@ to upconvert to 32b float internally? <reg32 offset="0x8882" name="RB_STENCIL_BUFFER_PITCH" low="0" high="11" shr="6" type="uint" usage="rp_blit"/> <reg32 offset="0x8883" name="RB_STENCIL_BUFFER_ARRAY_PITCH" low="0" high="23" shr="6" type="uint" usage="rp_blit"/> <reg64 offset="0x8884" name="RB_STENCIL_BUFFER_BASE" type="waddress" align="64" usage="rp_blit"/> - <reg32 offset="0x8886" name="RB_STENCIL_BUFFER_BASE_GMEM" low="12" high="31" shr="12" usage="rp_blit"/> - <reg32 offset="0x8887" name="RB_STENCILREF" usage="rp_blit"> + <reg32 offset="0x8886" name="RB_STENCIL_GMEM_BASE" low="12" high="31" shr="12" usage="rp_blit"/> + <reg32 offset="0x8887" name="RB_STENCIL_REF_CNTL" usage="rp_blit"> <bitfield name="REF" low="0" high="7"/> <bitfield name="BFREF" low="8" high="15"/> </reg32> - <reg32 offset="0x8888" name="RB_STENCILMASK" usage="rp_blit"> + <reg32 offset="0x8888" name="RB_STENCIL_MASK" usage="rp_blit"> <bitfield name="MASK" low="0" high="7"/> <bitfield name="BFMASK" low="8" high="15"/> </reg32> - <reg32 offset="0x8889" name="RB_STENCILWRMASK" usage="rp_blit"> + <reg32 offset="0x8889" name="RB_STENCIL_WRITE_MASK" usage="rp_blit"> <bitfield name="WRMASK" low="0" high="7"/> <bitfield name="BFWRMASK" low="8" high="15"/> </reg32> <!-- 0x888a-0x888f invalid --> <reg32 offset="0x8890" name="RB_WINDOW_OFFSET" type="a6xx_reg_xy" usage="rp_blit"/> - <reg32 offset="0x8891" name="RB_SAMPLE_COUNT_CONTROL" usage="cmd"> + <reg32 offset="0x8891" name="RB_SAMPLE_COUNTER_CNTL" usage="cmd"> <bitfield name="DISABLE" pos="0" type="boolean"/> <bitfield name="COPY" pos="1" type="boolean"/> </reg32> @@ -3791,27 +1612,27 @@ to upconvert to 32b float internally? <reg32 offset="0x8899" name="RB_UNKNOWN_8899" variants="A7XX-" usage="cmd"/> <!-- 0x8899-0x88bf invalid --> <!-- clamps depth value for depth test/write --> - <reg32 offset="0x88c0" name="RB_Z_CLAMP_MIN" type="float" usage="rp_blit"/> - <reg32 offset="0x88c1" name="RB_Z_CLAMP_MAX" type="float" usage="rp_blit"/> + <reg32 offset="0x88c0" name="RB_VIEWPORT_ZCLAMP_MIN" type="float" usage="rp_blit"/> + <reg32 offset="0x88c1" name="RB_VIEWPORT_ZCLAMP_MAX" type="float" usage="rp_blit"/> <!-- 0x88c2-0x88cf invalid--> - <reg32 offset="0x88d0" name="RB_UNKNOWN_88D0" usage="rp_blit"> + <reg32 offset="0x88d0" name="RB_RESOLVE_CNTL_0" usage="rp_blit"> <bitfield name="UNK0" low="0" high="12"/> <bitfield name="UNK16" low="16" high="26"/> </reg32> - <reg32 offset="0x88d1" name="RB_BLIT_SCISSOR_TL" type="a6xx_reg_xy" usage="rp_blit"/> - <reg32 offset="0x88d2" name="RB_BLIT_SCISSOR_BR" type="a6xx_reg_xy" usage="rp_blit"/> + <reg32 offset="0x88d1" name="RB_RESOLVE_CNTL_1" type="a6xx_reg_xy" usage="rp_blit"/> + <reg32 offset="0x88d2" name="RB_RESOLVE_CNTL_2" type="a6xx_reg_xy" usage="rp_blit"/> <!-- weird to duplicate other regs from same block?? --> - <reg32 offset="0x88d3" name="RB_BIN_CONTROL2" usage="rp_blit"> + <reg32 offset="0x88d3" name="RB_RESOLVE_CNTL_3" usage="rp_blit"> <bitfield name="BINW" low="0" high="5" shr="5" type="uint"/> <bitfield name="BINH" low="8" high="14" shr="4" type="uint"/> </reg32> - <reg32 offset="0x88d4" name="RB_WINDOW_OFFSET2" type="a6xx_reg_xy" usage="rp_blit"/> - <reg32 offset="0x88d5" name="RB_BLIT_GMEM_MSAA_CNTL" usage="rp_blit"> + <reg32 offset="0x88d4" name="RB_RESOLVE_WINDOW_OFFSET" type="a6xx_reg_xy" usage="rp_blit"/> + <reg32 offset="0x88d5" name="RB_RESOLVE_GMEM_BUFFER_INFO" usage="rp_blit"> <bitfield name="SAMPLES" low="3" high="4" type="a3xx_msaa_samples"/> </reg32> - <reg32 offset="0x88d6" name="RB_BLIT_BASE_GMEM" low="12" high="31" shr="12" usage="rp_blit"/> + <reg32 offset="0x88d6" name="RB_RESOLVE_GMEM_BUFFER_BASE" low="12" high="31" shr="12" usage="rp_blit"/> <!-- s/DST_FORMAT/DST_INFO/ probably: --> - <reg32 offset="0x88d7" name="RB_BLIT_DST_INFO" usage="rp_blit"> + <reg32 offset="0x88d7" name="RB_RESOLVE_SYSTEM_BUFFER_INFO" usage="rp_blit"> <bitfield name="TILE_MODE" low="0" high="1" type="a6xx_tile_mode"/> <bitfield name="FLAGS" pos="2" type="boolean"/> <bitfield name="SAMPLES" low="3" high="4" type="a3xx_msaa_samples"/> @@ -3820,25 +1641,31 @@ to upconvert to 32b float internally? <bitfield name="UNK15" pos="15" type="boolean"/> <bitfield name="MUTABLEEN" pos="16" type="boolean" variants="A7XX-"/> </reg32> - <reg64 offset="0x88d8" name="RB_BLIT_DST" type="waddress" align="64" usage="rp_blit"/> - <reg32 offset="0x88da" name="RB_BLIT_DST_PITCH" low="0" high="15" shr="6" type="uint" usage="rp_blit"/> + <reg64 offset="0x88d8" name="RB_RESOLVE_SYSTEM_BUFFER_BASE" type="waddress" align="64" usage="rp_blit"/> + <reg32 offset="0x88da" name="RB_RESOLVE_SYSTEM_BUFFER_PITCH" low="0" high="15" shr="6" type="uint" usage="rp_blit"/> <!-- array-pitch is size of layer --> - <reg32 offset="0x88db" name="RB_BLIT_DST_ARRAY_PITCH" low="0" high="28" shr="6" type="uint" usage="rp_blit"/> - <reg64 offset="0x88dc" name="RB_BLIT_FLAG_DST" type="waddress" align="64" usage="rp_blit"/> - <reg32 offset="0x88de" name="RB_BLIT_FLAG_DST_PITCH" usage="rp_blit"> + <reg32 offset="0x88db" name="RB_RESOLVE_SYSTEM_BUFFER_ARRAY_PITCH" low="0" high="28" shr="6" type="uint" usage="rp_blit"/> + <reg64 offset="0x88dc" name="RB_RESOLVE_SYSTEM_FLAG_BUFFER_BASE" type="waddress" align="64" usage="rp_blit"/> + <reg32 offset="0x88de" name="RB_RESOLVE_SYSTEM_FLAG_BUFFER_PITCH" usage="rp_blit"> <bitfield name="PITCH" low="0" high="10" shr="6" type="uint"/> <bitfield name="ARRAY_PITCH" low="11" high="27" shr="7" type="uint"/> </reg32> - <reg32 offset="0x88df" name="RB_BLIT_CLEAR_COLOR_DW0" usage="rp_blit"/> - <reg32 offset="0x88e0" name="RB_BLIT_CLEAR_COLOR_DW1" usage="rp_blit"/> - <reg32 offset="0x88e1" name="RB_BLIT_CLEAR_COLOR_DW2" usage="rp_blit"/> - <reg32 offset="0x88e2" name="RB_BLIT_CLEAR_COLOR_DW3" usage="rp_blit"/> + <reg32 offset="0x88df" name="RB_RESOLVE_CLEAR_COLOR_DW0" usage="rp_blit"/> + <reg32 offset="0x88e0" name="RB_RESOLVE_CLEAR_COLOR_DW1" usage="rp_blit"/> + <reg32 offset="0x88e1" name="RB_RESOLVE_CLEAR_COLOR_DW2" usage="rp_blit"/> + <reg32 offset="0x88e2" name="RB_RESOLVE_CLEAR_COLOR_DW3" usage="rp_blit"/> + + <enum name="a6xx_blit_event_type"> + <value value="0x0" name="BLIT_EVENT_STORE"/> + <value value="0x1" name="BLIT_EVENT_STORE_AND_CLEAR"/> + <value value="0x2" name="BLIT_EVENT_CLEAR"/> + <value value="0x3" name="BLIT_EVENT_LOAD"/> + </enum> <!-- seems somewhat similar to what we called RB_CLEAR_CNTL on a5xx: --> - <reg32 offset="0x88e3" name="RB_BLIT_INFO" usage="rp_blit"> - <bitfield name="UNK0" pos="0" type="boolean"/> <!-- s8 stencil restore/clear? But also color restore? --> - <bitfield name="GMEM" pos="1" type="boolean"/> <!-- set for restore and clear to gmem? --> + <reg32 offset="0x88e3" name="RB_RESOLVE_OPERATION" usage="rp_blit"> + <bitfield name="TYPE" low="0" high="1" type="a6xx_blit_event_type"/> <bitfield name="SAMPLE_0" pos="2" type="boolean"/> <!-- takes sample 0 instead of averaging --> <bitfield name="DEPTH" pos="3" type="boolean"/> <!-- z16/z32/z24s8/x24x8 clear or resolve? --> <doc> @@ -3853,16 +1680,20 @@ to upconvert to 32b float internally? <!-- set when this is the last resolve on a650+ --> <bitfield name="LAST" low="8" high="9"/> <!-- - a618 GLES: color render target number being resolved for RM6_RESOLVE, 0x8 for depth, 0x9 for separate stencil. - a618 VK: 0x8 for depth RM6_RESOLVE, 0x9 for separate stencil, 0 otherwise. - - We believe this is related to concurrent resolves + a618 GLES: color render target number being resolved for CCU_RESOLVE, 0x8 for depth, 0x9 for separate stencil. + a618 VK: 0x8 for depth CCU_RESOLVE, 0x9 for separate stencil, 0 otherwise. + a7xx VK: 0x8 for depth, 0x9 for separate stencil, 0x0 to 0x7 used for concurrent resolves of color render + targets inside a given resolve group. --> <bitfield name="BUFFER_ID" low="12" high="15"/> </reg32> - <reg32 offset="0x88e4" name="RB_UNKNOWN_88E4" variants="A7XX-" usage="rp_blit"> - <!-- Value conditioned based on predicate, changed before blits --> - <bitfield name="UNK0" pos="0" type="boolean"/> + + <enum name="a7xx_blit_clear_mode"> + <value value="0x0" name="CLEAR_MODE_SYSMEM"/> + <value value="0x1" name="CLEAR_MODE_GMEM"/> + </enum> + <reg32 offset="0x88e4" name="RB_CLEAR_TARGET" variants="A7XX-" usage="rp_blit"> + <bitfield name="CLEAR_MODE" pos="0" type="a7xx_blit_clear_mode"/> </reg32> <enum name="a6xx_ccu_cache_size"> @@ -3871,7 +1702,7 @@ to upconvert to 32b float internally? <value value="0x2" name="CCU_CACHE_SIZE_QUARTER"/> <value value="0x3" name="CCU_CACHE_SIZE_EIGHTH"/> </enum> - <reg32 offset="0x88e5" name="RB_CCU_CNTL2" variants="A7XX-" usage="cmd"> + <reg32 offset="0x88e5" name="RB_CCU_CACHE_CNTL" variants="A7XX-" usage="cmd"> <bitfield name="DEPTH_OFFSET_HI" pos="0" type="hex"/> <bitfield name="COLOR_OFFSET_HI" pos="2" type="hex"/> <bitfield name="DEPTH_CACHE_SIZE" low="10" high="11" type="a6xx_ccu_cache_size"/> @@ -3895,7 +1726,13 @@ to upconvert to 32b float internally? <bitfield name="PITCH" low="0" high="10" shr="6" type="uint"/> <bitfield name="ARRAY_PITCH" low="11" high="23" shr="7" type="uint"/> </reg32> - <reg32 offset="0x88f4" name="RB_UNKNOWN_88F4" low="0" high="2"/> + + <reg32 offset="0x88f4" name="RB_VRS_CONFIG" usage="rp_blit"> + <bitfield name="UNK2" pos="2" type="boolean"/> + <bitfield name="PIPELINE_FSR_ENABLE" pos="4" type="boolean"/> + <bitfield name="ATTACHMENT_FSR_ENABLE" pos="5" type="boolean"/> + <bitfield name="PRIMITIVE_FSR_ENABLE" pos="18" type="boolean"/> + </reg32> <!-- Connected to VK_EXT_fragment_density_map? --> <reg32 offset="0x88f5" name="RB_UNKNOWN_88F5" variants="A7XX-"/> <!-- 0x88f6-0x88ff invalid --> @@ -3906,7 +1743,7 @@ to upconvert to 32b float internally? <bitfield name="UNK8" low="8" high="10"/> <bitfield name="ARRAY_PITCH" low="11" high="27" shr="7" type="uint"/> </reg32> - <array offset="0x8903" name="RB_MRT_FLAG_BUFFER" stride="3" length="8" usage="rp_blit"> + <array offset="0x8903" name="RB_COLOR_FLAG_BUFFER" stride="3" length="8" usage="rp_blit"> <reg64 offset="0" name="ADDR" type="waddress" align="64"/> <reg32 offset="2" name="PITCH"> <bitfield name="PITCH" low="0" high="10" shr="6" type="uint"/> @@ -3915,10 +1752,10 @@ to upconvert to 32b float internally? </array> <!-- 0x891b-0x8926 invalid --> <doc> - RB_SAMPLE_COUNT_ADDR register is used up to (and including) a730. After that + RB_SAMPLE_COUNTER_BASE register is used up to (and including) a730. After that the address is specified through CP_EVENT_WRITE7::WRITE_SAMPLE_COUNT. </doc> - <reg64 offset="0x8927" name="RB_SAMPLE_COUNT_ADDR" type="waddress" align="16" usage="cmd"/> + <reg64 offset="0x8927" name="RB_SAMPLE_COUNTER_BASE" type="waddress" align="16" usage="cmd"/> <!-- 0x8929-0x89ff invalid --> <!-- TODO: there are some registers in the 0x8a00-0x8bff range --> @@ -3932,10 +1769,10 @@ to upconvert to 32b float internally? <reg32 offset="0x8a20" name="RB_UNKNOWN_8A20" variants="A6XX" usage="rp_blit"/> <reg32 offset="0x8a30" name="RB_UNKNOWN_8A30" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0x8c00" name="RB_2D_BLIT_CNTL" type="a6xx_2d_blit_cntl" usage="rp_blit"/> - <reg32 offset="0x8c01" name="RB_2D_UNKNOWN_8C01" low="0" high="31" usage="rp_blit"/> + <reg32 offset="0x8c00" name="RB_A2D_BLT_CNTL" type="a6xx_a2d_bit_cntl" usage="rp_blit"/> + <reg32 offset="0x8c01" name="RB_A2D_PIXEL_CNTL" low="0" high="31" usage="rp_blit"/> - <bitset name="a6xx_2d_src_surf_info" inline="yes"> + <bitset name="a6xx_a2d_src_texture_info" inline="yes"> <bitfield name="COLOR_FORMAT" low="0" high="7" type="a6xx_format"/> <bitfield name="TILE_MODE" low="8" high="9" type="a6xx_tile_mode"/> <bitfield name="COLOR_SWAP" low="10" high="11" type="a3xx_color_swap"/> @@ -3954,7 +1791,7 @@ to upconvert to 32b float internally? <bitfield name="MUTABLEEN" pos="29" type="boolean" variants="A7XX-"/> </bitset> - <bitset name="a6xx_2d_dst_surf_info" inline="yes"> + <bitset name="a6xx_a2d_dest_buffer_info" inline="yes"> <bitfield name="COLOR_FORMAT" low="0" high="7" type="a6xx_format"/> <bitfield name="TILE_MODE" low="8" high="9" type="a6xx_tile_mode"/> <bitfield name="COLOR_SWAP" low="10" high="11" type="a3xx_color_swap"/> @@ -3965,26 +1802,26 @@ to upconvert to 32b float internally? </bitset> <!-- 0x8c02-0x8c16 invalid --> - <reg32 offset="0x8c17" name="RB_2D_DST_INFO" type="a6xx_2d_dst_surf_info" usage="rp_blit"/> - <reg64 offset="0x8c18" name="RB_2D_DST" type="waddress" align="64" usage="rp_blit"/> - <reg32 offset="0x8c1a" name="RB_2D_DST_PITCH" low="0" high="15" shr="6" type="uint" usage="rp_blit"/> + <reg32 offset="0x8c17" name="RB_A2D_DEST_BUFFER_INFO" type="a6xx_a2d_dest_buffer_info" usage="rp_blit"/> + <reg64 offset="0x8c18" name="RB_A2D_DEST_BUFFER_BASE" type="waddress" align="64" usage="rp_blit"/> + <reg32 offset="0x8c1a" name="RB_A2D_DEST_BUFFER_PITCH" low="0" high="15" shr="6" type="uint" usage="rp_blit"/> <!-- this is a guess but seems likely (for NV12/IYUV): --> - <reg64 offset="0x8c1b" name="RB_2D_DST_PLANE1" type="waddress" align="64" usage="rp_blit"/> - <reg32 offset="0x8c1d" name="RB_2D_DST_PLANE_PITCH" low="0" high="15" shr="6" type="uint" usage="rp_blit"/> - <reg64 offset="0x8c1e" name="RB_2D_DST_PLANE2" type="waddress" align="64" usage="rp_blit"/> + <reg64 offset="0x8c1b" name="RB_A2D_DEST_BUFFER_BASE_1" type="waddress" align="64" usage="rp_blit"/> + <reg32 offset="0x8c1d" name="RB_A2D_DEST_BUFFER_PITCH_1" low="0" high="15" shr="6" type="uint" usage="rp_blit"/> + <reg64 offset="0x8c1e" name="RB_A2D_DEST_BUFFER_BASE_2" type="waddress" align="64" usage="rp_blit"/> - <reg64 offset="0x8c20" name="RB_2D_DST_FLAGS" type="waddress" align="64" usage="rp_blit"/> - <reg32 offset="0x8c22" name="RB_2D_DST_FLAGS_PITCH" low="0" high="7" shr="6" type="uint" usage="rp_blit"/> + <reg64 offset="0x8c20" name="RB_A2D_DEST_FLAG_BUFFER_BASE" type="waddress" align="64" usage="rp_blit"/> + <reg32 offset="0x8c22" name="RB_A2D_DEST_FLAG_BUFFER_PITCH" low="0" high="7" shr="6" type="uint" usage="rp_blit"/> <!-- this is a guess but seems likely (for NV12 with UBWC): --> - <reg64 offset="0x8c23" name="RB_2D_DST_FLAGS_PLANE" type="waddress" align="64" usage="rp_blit"/> - <reg32 offset="0x8c25" name="RB_2D_DST_FLAGS_PLANE_PITCH" low="0" high="7" shr="6" type="uint" usage="rp_blit"/> + <reg64 offset="0x8c23" name="RB_A2D_DEST_FLAG_BUFFER_BASE_1" type="waddress" align="64" usage="rp_blit"/> + <reg32 offset="0x8c25" name="RB_A2D_DEST_FLAG_BUFFER_PITCH_1" low="0" high="7" shr="6" type="uint" usage="rp_blit"/> <!-- TODO: 0x8c26-0x8c33 are all full 32-bit registers --> <!-- unlike a5xx, these are per channel values rather than packed --> - <reg32 offset="0x8c2c" name="RB_2D_SRC_SOLID_C0" usage="rp_blit"/> - <reg32 offset="0x8c2d" name="RB_2D_SRC_SOLID_C1" usage="rp_blit"/> - <reg32 offset="0x8c2e" name="RB_2D_SRC_SOLID_C2" usage="rp_blit"/> - <reg32 offset="0x8c2f" name="RB_2D_SRC_SOLID_C3" usage="rp_blit"/> + <reg32 offset="0x8c2c" name="RB_A2D_CLEAR_COLOR_DW0" usage="rp_blit"/> + <reg32 offset="0x8c2d" name="RB_A2D_CLEAR_COLOR_DW1" usage="rp_blit"/> + <reg32 offset="0x8c2e" name="RB_A2D_CLEAR_COLOR_DW2" usage="rp_blit"/> + <reg32 offset="0x8c2f" name="RB_A2D_CLEAR_COLOR_DW3" usage="rp_blit"/> <reg32 offset="0x8c34" name="RB_UNKNOWN_8C34" variants="A7XX-" usage="cmd"/> @@ -3996,7 +1833,7 @@ to upconvert to 32b float internally? <reg32 offset="0x8e04" name="RB_DBG_ECO_CNTL" usage="cmd"/> <!-- TODO: valid mask 0xfffffeff --> <reg32 offset="0x8e05" name="RB_ADDR_MODE_CNTL" pos="0" type="a5xx_address_mode"/> <!-- 0x02080000 in GMEM, zero otherwise? --> - <reg32 offset="0x8e06" name="RB_UNKNOWN_8E06" variants="A7XX-" usage="cmd"/> + <reg32 offset="0x8e06" name="RB_CCU_DBG_ECO_CNTL" variants="A7XX-" usage="cmd"/> <reg32 offset="0x8e07" name="RB_CCU_CNTL" usage="cmd" variants="A6XX"> <bitfield name="GMEM_FAST_CLEAR_DISABLE" pos="0" type="boolean"/> @@ -4017,10 +1854,21 @@ to upconvert to 32b float internally? <bitfield name="COLOR_OFFSET" low="23" high="31" shr="12" type="hex"/> <!--TODO: valid mask 0xfffffc1f --> </reg32> + <enum name="a7xx_concurrent_resolve_mode"> + <value value="0x0" name="CONCURRENT_RESOLVE_MODE_DISABLED"/> + <value value="0x1" name="CONCURRENT_RESOLVE_MODE_1"/> + <value value="0x2" name="CONCURRENT_RESOLVE_MODE_2"/> + </enum> + <enum name="a7xx_concurrent_unresolve_mode"> + <value value="0x0" name="CONCURRENT_UNRESOLVE_MODE_DISABLED"/> + <value value="0x1" name="CONCURRENT_UNRESOLVE_MODE_PARTIAL"/> + <value value="0x3" name="CONCURRENT_UNRESOLVE_MODE_FULL"/> + </enum> <reg32 offset="0x8e07" name="RB_CCU_CNTL" usage="cmd" variants="A7XX-"> <bitfield name="GMEM_FAST_CLEAR_DISABLE" pos="0" type="boolean"/> - <bitfield name="CONCURRENT_RESOLVE" pos="2" type="boolean"/> - <!-- rest of the bits were moved to RB_CCU_CNTL2 --> + <bitfield name="CONCURRENT_RESOLVE_MODE" low="2" high="3" type="a7xx_concurrent_resolve_mode"/> + <bitfield name="CONCURRENT_UNRESOLVE_MODE" low="5" high="6" type="a7xx_concurrent_unresolve_mode"/> + <!-- rest of the bits were moved to RB_CCU_CACHE_CNTL --> </reg32> <reg32 offset="0x8e08" name="RB_NC_MODE_CNTL"> <bitfield name="MODE" pos="0" type="boolean"/> @@ -4046,9 +1894,9 @@ to upconvert to 32b float internally? <reg32 offset="0x8e3d" name="RB_RB_SUB_BLOCK_SEL_CNTL_CD"/> <!-- 0x8e3e-0x8e4f invalid --> <!-- GMEM save/restore for preemption: --> - <reg32 offset="0x8e50" name="RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE" pos="0" type="boolean"/> + <reg32 offset="0x8e50" name="RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ENABLE" pos="0" type="boolean"/> <!-- address for GMEM save/restore? --> - <reg32 offset="0x8e51" name="RB_UNKNOWN_8E51" type="waddress" align="1"/> + <reg32 offset="0x8e51" name="RB_CONTEXT_SWITCH_GMEM_SAVE_RESTORE_ADDR" type="waddress" align="1"/> <!-- 0x8e53-0x8e7f invalid --> <reg32 offset="0x8e79" name="RB_UNKNOWN_8E79" variants="A7XX-" usage="cmd"/> <!-- 0x8e80-0x8e83 are valid --> @@ -4069,38 +1917,38 @@ to upconvert to 32b float internally? <bitfield name="CLIP_DIST_03_LOC" low="8" high="15" type="uint"/> <bitfield name="CLIP_DIST_47_LOC" low="16" high="23" type="uint"/> </bitset> - <reg32 offset="0x9101" name="VPC_VS_CLIP_CNTL" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> - <reg32 offset="0x9102" name="VPC_GS_CLIP_CNTL" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> - <reg32 offset="0x9103" name="VPC_DS_CLIP_CNTL" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> + <reg32 offset="0x9101" name="VPC_VS_CLIP_CULL_CNTL" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> + <reg32 offset="0x9102" name="VPC_GS_CLIP_CULL_CNTL" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> + <reg32 offset="0x9103" name="VPC_DS_CLIP_CULL_CNTL" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> - <reg32 offset="0x9311" name="VPC_VS_CLIP_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> - <reg32 offset="0x9312" name="VPC_GS_CLIP_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> - <reg32 offset="0x9313" name="VPC_DS_CLIP_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> + <reg32 offset="0x9311" name="VPC_VS_CLIP_CULL_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> + <reg32 offset="0x9312" name="VPC_GS_CLIP_CULL_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> + <reg32 offset="0x9313" name="VPC_DS_CLIP_CULL_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/> - <bitset name="a6xx_vpc_xs_layer_cntl" inline="yes"> + <bitset name="a6xx_vpc_xs_siv_cntl" inline="yes"> <bitfield name="LAYERLOC" low="0" high="7" type="uint"/> <bitfield name="VIEWLOC" low="8" high="15" type="uint"/> <bitfield name="SHADINGRATELOC" low="16" high="23" type="uint" variants="A7XX-"/> </bitset> - <reg32 offset="0x9104" name="VPC_VS_LAYER_CNTL" type="a6xx_vpc_xs_layer_cntl" usage="rp_blit"/> - <reg32 offset="0x9105" name="VPC_GS_LAYER_CNTL" type="a6xx_vpc_xs_layer_cntl" usage="rp_blit"/> - <reg32 offset="0x9106" name="VPC_DS_LAYER_CNTL" type="a6xx_vpc_xs_layer_cntl" usage="rp_blit"/> + <reg32 offset="0x9104" name="VPC_VS_SIV_CNTL" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/> + <reg32 offset="0x9105" name="VPC_GS_SIV_CNTL" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/> + <reg32 offset="0x9106" name="VPC_DS_SIV_CNTL" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/> - <reg32 offset="0x9314" name="VPC_VS_LAYER_CNTL_V2" type="a6xx_vpc_xs_layer_cntl" usage="rp_blit"/> - <reg32 offset="0x9315" name="VPC_GS_LAYER_CNTL_V2" type="a6xx_vpc_xs_layer_cntl" usage="rp_blit"/> - <reg32 offset="0x9316" name="VPC_DS_LAYER_CNTL_V2" type="a6xx_vpc_xs_layer_cntl" usage="rp_blit"/> + <reg32 offset="0x9314" name="VPC_VS_SIV_CNTL_V2" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/> + <reg32 offset="0x9315" name="VPC_GS_SIV_CNTL_V2" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/> + <reg32 offset="0x9316" name="VPC_DS_SIV_CNTL_V2" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/> <reg32 offset="0x9107" name="VPC_UNKNOWN_9107" variants="A6XX" usage="rp_blit"> - <!-- this mirrors PC_RASTER_CNTL::DISCARD, although it seems it's unused --> + <!-- this mirrors VPC_RAST_STREAM_CNTL::DISCARD, although it seems it's unused --> <bitfield name="RASTER_DISCARD" pos="0" type="boolean"/> <bitfield name="UNK2" pos="2" type="boolean"/> </reg32> - <reg32 offset="0x9108" name="VPC_POLYGON_MODE" usage="rp_blit"> + <reg32 offset="0x9108" name="VPC_RAST_CNTL" usage="rp_blit"> <bitfield name="MODE" low="0" high="1" type="a6xx_polygon_mode"/> </reg32> - <bitset name="a6xx_primitive_cntl_0" inline="yes"> + <bitset name="a6xx_pc_cntl" inline="yes"> <bitfield name="PRIMITIVE_RESTART" pos="0" type="boolean"/> <bitfield name="PROVOKING_VTX_LAST" pos="1" type="boolean"/> <bitfield name="D3D_VERTEX_ORDERING" pos="2" type="boolean"> @@ -4113,7 +1961,7 @@ to upconvert to 32b float internally? <bitfield name="UNK3" pos="3" type="boolean"/> </bitset> - <bitset name="a6xx_primitive_cntl_5" inline="yes"> + <bitset name="a6xx_gs_param_0" inline="yes"> <doc> geometry shader </doc> @@ -4125,7 +1973,7 @@ to upconvert to 32b float internally? <bitfield name="UNK18" pos="18"/> </bitset> - <bitset name="a6xx_multiview_cntl" inline="yes"> + <bitset name="a6xx_stereo_rendering_cntl" inline="yes"> <bitfield name="ENABLE" pos="0" type="boolean"/> <bitfield name="DISABLEMULTIPOS" pos="1" type="boolean"> <doc> @@ -4139,10 +1987,10 @@ to upconvert to 32b float internally? <bitfield name="VIEWS" low="2" high="6" type="uint"/> </bitset> - <reg32 offset="0x9109" name="VPC_PRIMITIVE_CNTL_0" type="a6xx_primitive_cntl_0" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0x910a" name="VPC_PRIMITIVE_CNTL_5" type="a6xx_primitive_cntl_5" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0x910b" name="VPC_MULTIVIEW_MASK" type="hex" low="0" high="15" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0x910c" name="VPC_MULTIVIEW_CNTL" type="a6xx_multiview_cntl" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0x9109" name="VPC_PC_CNTL" type="a6xx_pc_cntl" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0x910a" name="VPC_GS_PARAM_0" type="a6xx_gs_param_0" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0x910b" name="VPC_STEREO_RENDERING_VIEWMASK" type="hex" low="0" high="15" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0x910c" name="VPC_STEREO_RENDERING_CNTL" type="a6xx_stereo_rendering_cntl" variants="A7XX-" usage="rp_blit"/> <enum name="a6xx_varying_interp_mode"> <value value="0" name="INTERP_SMOOTH"/> @@ -4159,11 +2007,11 @@ to upconvert to 32b float internally? </enum> <!-- 0x9109-0x91ff invalid --> - <array offset="0x9200" name="VPC_VARYING_INTERP" stride="1" length="8" usage="rp_blit"> + <array offset="0x9200" name="VPC_VARYING_INTERP_MODE" stride="1" length="8" usage="rp_blit"> <doc>Packed array of a6xx_varying_interp_mode</doc> <reg32 offset="0x0" name="MODE"/> </array> - <array offset="0x9208" name="VPC_VARYING_PS_REPL" stride="1" length="8" usage="rp_blit"> + <array offset="0x9208" name="VPC_VARYING_REPLACE_MODE_0" stride="1" length="8" usage="rp_blit"> <doc>Packed array of a6xx_varying_ps_repl_mode</doc> <reg32 offset="0x0" name="MODE"/> </array> @@ -4172,12 +2020,12 @@ to upconvert to 32b float internally? <reg32 offset="0x9210" name="VPC_UNKNOWN_9210" low="0" high="31" variants="A6XX" usage="cmd"/> <reg32 offset="0x9211" name="VPC_UNKNOWN_9211" low="0" high="31" variants="A6XX" usage="cmd"/> - <array offset="0x9212" name="VPC_VAR" stride="1" length="4" usage="rp_blit"> + <array offset="0x9212" name="VPC_VARYING_LM_TRANSFER_CNTL_0" stride="1" length="4" usage="rp_blit"> <!-- one bit per varying component: --> <reg32 offset="0" name="DISABLE"/> </array> - <reg32 offset="0x9216" name="VPC_SO_CNTL" usage="rp_blit"> + <reg32 offset="0x9216" name="VPC_SO_MAPPING_WPTR" usage="rp_blit"> <!-- Choose which DWORD to write to. There is an array of (4 * 64) DWORD's, dumped in the devcoredump at @@ -4198,7 +2046,7 @@ to upconvert to 32b float internally? When EmitStreamVertex(N) happens, the HW goes to DWORD 64 * N and then "executes" the next 64 DWORD's. - This field is auto-incremented when VPC_SO_PROG is + This field is auto-incremented when VPC_SO_MAPPING_PORT is written to. --> <bitfield name="ADDR" low="0" high="7" type="hex"/> @@ -4206,7 +2054,7 @@ to upconvert to 32b float internally? <bitfield name="RESET" pos="16" type="boolean"/> </reg32> <!-- special register, write multiple times to load SO program (not readable) --> - <reg32 offset="0x9217" name="VPC_SO_PROG" usage="rp_blit"> + <reg32 offset="0x9217" name="VPC_SO_MAPPING_PORT" usage="rp_blit"> <bitfield name="A_BUF" low="0" high="1" type="uint"/> <bitfield name="A_OFF" low="2" high="10" shr="2" type="uint"/> <bitfield name="A_EN" pos="11" type="boolean"/> @@ -4215,7 +2063,7 @@ to upconvert to 32b float internally? <bitfield name="B_EN" pos="23" type="boolean"/> </reg32> - <reg64 offset="0x9218" name="VPC_SO_STREAM_COUNTS" type="waddress" align="32" usage="cmd"/> + <reg64 offset="0x9218" name="VPC_SO_QUERY_BASE" type="waddress" align="32" usage="cmd"/> <array offset="0x921a" name="VPC_SO" stride="7" length="4" usage="cmd"> <reg64 offset="0" name="BUFFER_BASE" type="waddress" align="32"/> @@ -4225,14 +2073,14 @@ to upconvert to 32b float internally? <reg64 offset="5" name="FLUSH_BASE" type="waddress" align="32"/> </array> - <reg32 offset="0x9236" name="VPC_POINT_COORD_INVERT" usage="cmd"> + <reg32 offset="0x9236" name="VPC_REPLACE_MODE_CNTL" usage="cmd"> <bitfield name="INVERT" pos="0" type="boolean"/> </reg32> <!-- 0x9237-0x92ff invalid --> <!-- always 0x0 ? --> <reg32 offset="0x9300" name="VPC_UNKNOWN_9300" low="0" high="2" usage="cmd"/> - <bitset name="a6xx_vpc_xs_pack" inline="yes"> + <bitset name="a6xx_vpc_xs_cntl" inline="yes"> <doc> num of varyings plus four for gl_Position (plus one if gl_PointSize) plus # of transform-feedback (streamout) varyings if using the @@ -4249,11 +2097,11 @@ to upconvert to 32b float internally? </doc> </bitfield> </bitset> - <reg32 offset="0x9301" name="VPC_VS_PACK" type="a6xx_vpc_xs_pack" usage="rp_blit"/> - <reg32 offset="0x9302" name="VPC_GS_PACK" type="a6xx_vpc_xs_pack" usage="rp_blit"/> - <reg32 offset="0x9303" name="VPC_DS_PACK" type="a6xx_vpc_xs_pack" usage="rp_blit"/> + <reg32 offset="0x9301" name="VPC_VS_CNTL" type="a6xx_vpc_xs_cntl" usage="rp_blit"/> + <reg32 offset="0x9302" name="VPC_GS_CNTL" type="a6xx_vpc_xs_cntl" usage="rp_blit"/> + <reg32 offset="0x9303" name="VPC_DS_CNTL" type="a6xx_vpc_xs_cntl" usage="rp_blit"/> - <reg32 offset="0x9304" name="VPC_CNTL_0" usage="rp_blit"> + <reg32 offset="0x9304" name="VPC_PS_CNTL" usage="rp_blit"> <bitfield name="NUMNONPOSVAR" low="0" high="7" type="uint"/> <!-- for fixed-function (i.e. no GS) gl_PrimitiveID in FS --> <bitfield name="PRIMIDLOC" low="8" high="15" type="uint"/> @@ -4272,7 +2120,7 @@ to upconvert to 32b float internally? </bitfield> </reg32> - <reg32 offset="0x9305" name="VPC_SO_STREAM_CNTL" usage="rp_blit"> + <reg32 offset="0x9305" name="VPC_SO_CNTL" usage="rp_blit"> <!-- It's offset by 1, and 0 means "disabled" --> @@ -4282,19 +2130,19 @@ to upconvert to 32b float internally? <bitfield name="BUF3_STREAM" low="9" high="11" type="uint"/> <bitfield name="STREAM_ENABLE" low="15" high="18" type="hex"/> </reg32> - <reg32 offset="0x9306" name="VPC_SO_DISABLE" usage="rp_blit"> + <reg32 offset="0x9306" name="VPC_SO_OVERRIDE" usage="rp_blit"> <bitfield name="DISABLE" pos="0" type="boolean"/> </reg32> - <reg32 offset="0x9307" name="VPC_POLYGON_MODE2" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0x9307" name="VPC_PS_RAST_CNTL" variants="A6XX-" usage="rp_blit"> <!-- A702 + A7xx --> <bitfield name="MODE" low="0" high="1" type="a6xx_polygon_mode"/> </reg32> - <reg32 offset="0x9308" name="VPC_ATTR_BUF_SIZE_GMEM" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0x9308" name="VPC_ATTR_BUF_GMEM_SIZE" variants="A7XX-" usage="rp_blit"> <bitfield name="SIZE_GMEM" low="0" high="31"/> </reg32> - <reg32 offset="0x9309" name="VPC_ATTR_BUF_BASE_GMEM" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0x9309" name="VPC_ATTR_BUF_GMEM_BASE" variants="A7XX-" usage="rp_blit"> <bitfield name="BASE_GMEM" low="0" high="31"/> </reg32> - <reg32 offset="0x9b09" name="PC_ATTR_BUF_SIZE_GMEM" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0x9b09" name="PC_ATTR_BUF_GMEM_SIZE" variants="A7XX-" usage="rp_blit"> <bitfield name="SIZE_GMEM" low="0" high="31"/> </reg32> @@ -4311,15 +2159,15 @@ to upconvert to 32b float internally? <!-- TODO: regs from 0x9624-0x963a --> <!-- 0x963b-0x97ff invalid --> - <reg32 offset="0x9800" name="PC_TESS_NUM_VERTEX" low="0" high="5" type="uint" usage="rp_blit"/> + <reg32 offset="0x9800" name="PC_HS_PARAM_0" low="0" high="5" type="uint" usage="rp_blit"/> <!-- always 0x0 ? --> - <reg32 offset="0x9801" name="PC_HS_INPUT_SIZE" usage="rp_blit"> + <reg32 offset="0x9801" name="PC_HS_PARAM_1" usage="rp_blit"> <bitfield name="SIZE" low="0" high="10" type="uint"/> <bitfield name="UNK13" pos="13"/> </reg32> - <reg32 offset="0x9802" name="PC_TESS_CNTL" usage="rp_blit"> + <reg32 offset="0x9802" name="PC_DS_PARAM" usage="rp_blit"> <bitfield name="SPACING" low="0" high="1" type="a6xx_tess_spacing"/> <bitfield name="OUTPUT" low="2" high="3" type="a6xx_tess_output"/> </reg32> @@ -4334,7 +2182,7 @@ to upconvert to 32b float internally? </reg32> <!-- New in a6xx gen3+ --> - <reg32 offset="0x9808" name="PC_SO_STREAM_CNTL" usage="rp_blit"> + <reg32 offset="0x9808" name="PC_DGEN_SO_CNTL" usage="rp_blit"> <bitfield name="STREAM_ENABLE" low="15" high="18" type="hex"/> </reg32> @@ -4344,15 +2192,15 @@ to upconvert to 32b float internally? <!-- 0x980b-0x983f invalid --> <!-- 0x9840 - 0x9842 are not readable --> - <reg32 offset="0x9840" name="PC_DRAW_CMD"> + <reg32 offset="0x9840" name="PC_DRAW_INITIATOR"> <bitfield name="STATE_ID" low="0" high="7"/> </reg32> - <reg32 offset="0x9841" name="PC_DISPATCH_CMD"> + <reg32 offset="0x9841" name="PC_KERNEL_INITIATOR"> <bitfield name="STATE_ID" low="0" high="7"/> </reg32> - <reg32 offset="0x9842" name="PC_EVENT_CMD"> + <reg32 offset="0x9842" name="PC_EVENT_INITIATOR"> <!-- I think only the low bit is actually used? --> <bitfield name="STATE_ID" low="16" high="23"/> <bitfield name="EVENT" low="0" high="6" type="vgt_event_type"/> @@ -4367,27 +2215,27 @@ to upconvert to 32b float internally? <!-- 0x9843-0x997f invalid --> - <reg32 offset="0x9981" name="PC_POLYGON_MODE" variants="A6XX" usage="rp_blit"> + <reg32 offset="0x9981" name="PC_DGEN_RAST_CNTL" variants="A6XX" usage="rp_blit"> <bitfield name="MODE" low="0" high="1" type="a6xx_polygon_mode"/> </reg32> - <reg32 offset="0x9809" name="PC_POLYGON_MODE" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0x9809" name="PC_DGEN_RAST_CNTL" variants="A7XX-" usage="rp_blit"> <bitfield name="MODE" low="0" high="1" type="a6xx_polygon_mode"/> </reg32> - <reg32 offset="0x9980" name="PC_RASTER_CNTL" variants="A6XX" usage="rp_blit"> + <reg32 offset="0x9980" name="VPC_RAST_STREAM_CNTL" variants="A6XX" usage="rp_blit"> <!-- which stream to send to GRAS --> <bitfield name="STREAM" low="0" high="1" type="uint"/> <!-- discard primitives before rasterization --> <bitfield name="DISCARD" pos="2" type="boolean"/> </reg32> - <!-- VPC_RASTER_CNTL --> - <reg32 offset="0x9107" name="PC_RASTER_CNTL" variants="A7XX-" usage="rp_blit"> + <!-- VPC_RAST_STREAM_CNTL --> + <reg32 offset="0x9107" name="VPC_RAST_STREAM_CNTL" variants="A7XX-" usage="rp_blit"> <!-- which stream to send to GRAS --> <bitfield name="STREAM" low="0" high="1" type="uint"/> <!-- discard primitives before rasterization --> <bitfield name="DISCARD" pos="2" type="boolean"/> </reg32> - <reg32 offset="0x9317" name="PC_RASTER_CNTL_V2" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0x9317" name="VPC_RAST_STREAM_CNTL_V2" variants="A7XX-" usage="rp_blit"> <!-- which stream to send to GRAS --> <bitfield name="STREAM" low="0" high="1" type="uint"/> <!-- discard primitives before rasterization --> @@ -4397,17 +2245,17 @@ to upconvert to 32b float internally? <!-- Both are a750+. Probably needed to correctly overlap execution of several draws. --> - <reg32 offset="0x9885" name="PC_TESS_PARAM_SIZE" variants="A7XX-" usage="cmd"/> + <reg32 offset="0x9885" name="PC_HS_BUFFER_SIZE" variants="A7XX-" usage="cmd"/> <!-- Blob adds a bit more space {0x10, 0x20, 0x30, 0x40} bytes, but the meaning of this additional space is not known. --> - <reg32 offset="0x9886" name="PC_TESS_FACTOR_SIZE" variants="A7XX-" usage="cmd"/> + <reg32 offset="0x9886" name="PC_TF_BUFFER_SIZE" variants="A7XX-" usage="cmd"/> <!-- 0x9982-0x9aff invalid --> - <reg32 offset="0x9b00" name="PC_PRIMITIVE_CNTL_0" type="a6xx_primitive_cntl_0" usage="rp_blit"/> + <reg32 offset="0x9b00" name="PC_CNTL" type="a6xx_pc_cntl" usage="rp_blit"/> - <bitset name="a6xx_xs_out_cntl" inline="yes"> + <bitset name="a6xx_pc_xs_cntl" inline="yes"> <doc> num of varyings plus four for gl_Position (plus one if gl_PointSize) plus # of transform-feedback (streamout) varyings if using the @@ -4417,19 +2265,19 @@ to upconvert to 32b float internally? <bitfield name="PSIZE" pos="8" type="boolean"/> <bitfield name="LAYER" pos="9" type="boolean"/> <bitfield name="VIEW" pos="10" type="boolean"/> - <!-- note: PC_VS_OUT_CNTL doesn't have the PRIMITIVE_ID bit --> + <!-- note: PC_VS_CNTL doesn't have the PRIMITIVE_ID bit --> <bitfield name="PRIMITIVE_ID" pos="11" type="boolean"/> <bitfield name="CLIP_MASK" low="16" high="23" type="uint"/> <bitfield name="SHADINGRATE" pos="24" type="boolean" variants="A7XX-"/> </bitset> - <reg32 offset="0x9b01" name="PC_VS_OUT_CNTL" type="a6xx_xs_out_cntl" usage="rp_blit"/> - <reg32 offset="0x9b02" name="PC_GS_OUT_CNTL" type="a6xx_xs_out_cntl" usage="rp_blit"/> + <reg32 offset="0x9b01" name="PC_VS_CNTL" type="a6xx_pc_xs_cntl" usage="rp_blit"/> + <reg32 offset="0x9b02" name="PC_GS_CNTL" type="a6xx_pc_xs_cntl" usage="rp_blit"/> <!-- since HS can't output anything, only PRIMITIVE_ID is valid --> - <reg32 offset="0x9b03" name="PC_HS_OUT_CNTL" type="a6xx_xs_out_cntl" usage="rp_blit"/> - <reg32 offset="0x9b04" name="PC_DS_OUT_CNTL" type="a6xx_xs_out_cntl" usage="rp_blit"/> + <reg32 offset="0x9b03" name="PC_HS_CNTL" type="a6xx_pc_xs_cntl" usage="rp_blit"/> + <reg32 offset="0x9b04" name="PC_DS_CNTL" type="a6xx_pc_xs_cntl" usage="rp_blit"/> - <reg32 offset="0x9b05" name="PC_PRIMITIVE_CNTL_5" type="a6xx_primitive_cntl_5" usage="rp_blit"/> + <reg32 offset="0x9b05" name="PC_GS_PARAM_0" type="a6xx_gs_param_0" usage="rp_blit"/> <reg32 offset="0x9b06" name="PC_PRIMITIVE_CNTL_6" variants="A6XX" usage="rp_blit"> <doc> @@ -4438,9 +2286,9 @@ to upconvert to 32b float internally? <bitfield name="STRIDE_IN_VPC" low="0" high="10" type="uint"/> </reg32> - <reg32 offset="0x9b07" name="PC_MULTIVIEW_CNTL" type="a6xx_multiview_cntl" usage="rp_blit"/> + <reg32 offset="0x9b07" name="PC_STEREO_RENDERING_CNTL" type="a6xx_stereo_rendering_cntl" usage="rp_blit"/> <!-- mask of enabled views, doesn't exist on A630 --> - <reg32 offset="0x9b08" name="PC_MULTIVIEW_MASK" type="hex" low="0" high="15" usage="rp_blit"/> + <reg32 offset="0x9b08" name="PC_STEREO_RENDERING_VIEWMASK" type="hex" low="0" high="15" usage="rp_blit"/> <!-- 0x9b09-0x9bff invalid --> <reg32 offset="0x9c00" name="PC_2D_EVENT_CMD"> <!-- special register (but note first 8 bits can be written/read) --> @@ -4451,31 +2299,31 @@ to upconvert to 32b float internally? <!-- TODO: 0x9e00-0xa000 range incomplete --> <reg32 offset="0x9e00" name="PC_DBG_ECO_CNTL"/> <reg32 offset="0x9e01" name="PC_ADDR_MODE_CNTL" type="a5xx_address_mode"/> - <reg64 offset="0x9e04" name="PC_DRAW_INDX_BASE"/> - <reg32 offset="0x9e06" name="PC_DRAW_FIRST_INDX" type="uint"/> - <reg32 offset="0x9e07" name="PC_DRAW_MAX_INDICES" type="uint"/> - <reg64 offset="0x9e08" name="PC_TESSFACTOR_ADDR" variants="A6XX" type="waddress" align="32" usage="cmd"/> - <reg64 offset="0x9810" name="PC_TESSFACTOR_ADDR" variants="A7XX-" type="waddress" align="32" usage="cmd"/> + <reg64 offset="0x9e04" name="PC_DMA_BASE"/> + <reg32 offset="0x9e06" name="PC_DMA_OFFSET" type="uint"/> + <reg32 offset="0x9e07" name="PC_DMA_SIZE" type="uint"/> + <reg64 offset="0x9e08" name="PC_TESS_BASE" variants="A6XX" type="waddress" align="32" usage="cmd"/> + <reg64 offset="0x9810" name="PC_TESS_BASE" variants="A7XX-" type="waddress" align="32" usage="cmd"/> - <reg32 offset="0x9e0b" name="PC_DRAW_INITIATOR" type="vgt_draw_initiator_a4xx"> + <reg32 offset="0x9e0b" name="PC_DRAWCALL_CNTL" type="vgt_draw_initiator_a4xx"> <doc> Possibly not really "initiating" the draw but the layout is similar to VGT_DRAW_INITIATOR on older gens </doc> </reg32> - <reg32 offset="0x9e0c" name="PC_DRAW_NUM_INSTANCES" type="uint"/> - <reg32 offset="0x9e0d" name="PC_DRAW_NUM_INDICES" type="uint"/> + <reg32 offset="0x9e0c" name="PC_DRAWCALL_INSTANCE_NUM" type="uint"/> + <reg32 offset="0x9e0d" name="PC_DRAWCALL_SIZE" type="uint"/> <!-- These match the contents of CP_SET_BIN_DATA (not written directly) --> - <reg32 offset="0x9e11" name="PC_VSTREAM_CONTROL"> + <reg32 offset="0x9e11" name="PC_VIS_STREAM_CNTL"> <bitfield name="UNK0" low="0" high="15"/> <bitfield name="VSC_SIZE" low="16" high="21" type="uint"/> <bitfield name="VSC_N" low="22" high="26" type="uint"/> </reg32> - <reg64 offset="0x9e12" name="PC_BIN_PRIM_STRM" type="waddress" align="32"/> - <reg64 offset="0x9e14" name="PC_BIN_DRAW_STRM" type="waddress" align="32"/> + <reg64 offset="0x9e12" name="PC_PVIS_STREAM_BIN_BASE" type="waddress" align="32"/> + <reg64 offset="0x9e14" name="PC_DVIS_STREAM_BIN_BASE" type="waddress" align="32"/> - <reg32 offset="0x9e1c" name="PC_VISIBILITY_OVERRIDE"> + <reg32 offset="0x9e1c" name="PC_DRAWCALL_CNTL_OVERRIDE"> <doc>Written by CP_SET_VISIBILITY_OVERRIDE handler</doc> <bitfield name="OVERRIDE" pos="0" type="boolean"/> </reg32> @@ -4488,18 +2336,18 @@ to upconvert to 32b float internally? <!-- always 0x0 --> <reg32 offset="0x9e72" name="PC_UNKNOWN_9E72" usage="cmd"/> - <reg32 offset="0xa000" name="VFD_CONTROL_0" usage="rp_blit"> + <reg32 offset="0xa000" name="VFD_CNTL_0" usage="rp_blit"> <bitfield name="FETCH_CNT" low="0" high="5" type="uint"/> <bitfield name="DECODE_CNT" low="8" high="13" type="uint"/> </reg32> - <reg32 offset="0xa001" name="VFD_CONTROL_1" usage="rp_blit"> + <reg32 offset="0xa001" name="VFD_CNTL_1" usage="rp_blit"> <bitfield name="REGID4VTX" low="0" high="7" type="a3xx_regid"/> <bitfield name="REGID4INST" low="8" high="15" type="a3xx_regid"/> <bitfield name="REGID4PRIMID" low="16" high="23" type="a3xx_regid"/> <!-- only used for VS in non-multi-position-output case --> <bitfield name="REGID4VIEWID" low="24" high="31" type="a3xx_regid"/> </reg32> - <reg32 offset="0xa002" name="VFD_CONTROL_2" usage="rp_blit"> + <reg32 offset="0xa002" name="VFD_CNTL_2" usage="rp_blit"> <bitfield name="REGID_HSRELPATCHID" low="0" high="7" type="a3xx_regid"> <doc> This is the ID of the current patch within the @@ -4512,32 +2360,32 @@ to upconvert to 32b float internally? </bitfield> <bitfield name="REGID_INVOCATIONID" low="8" high="15" type="a3xx_regid"/> </reg32> - <reg32 offset="0xa003" name="VFD_CONTROL_3" usage="rp_blit"> + <reg32 offset="0xa003" name="VFD_CNTL_3" usage="rp_blit"> <bitfield name="REGID_DSPRIMID" low="0" high="7" type="a3xx_regid"/> <bitfield name="REGID_DSRELPATCHID" low="8" high="15" type="a3xx_regid"/> <bitfield name="REGID_TESSX" low="16" high="23" type="a3xx_regid"/> <bitfield name="REGID_TESSY" low="24" high="31" type="a3xx_regid"/> </reg32> - <reg32 offset="0xa004" name="VFD_CONTROL_4" usage="rp_blit"> + <reg32 offset="0xa004" name="VFD_CNTL_4" usage="rp_blit"> <bitfield name="UNK0" low="0" high="7" type="a3xx_regid"/> </reg32> - <reg32 offset="0xa005" name="VFD_CONTROL_5" usage="rp_blit"> + <reg32 offset="0xa005" name="VFD_CNTL_5" usage="rp_blit"> <bitfield name="REGID_GSHEADER" low="0" high="7" type="a3xx_regid"/> <bitfield name="UNK8" low="8" high="15" type="a3xx_regid"/> </reg32> - <reg32 offset="0xa006" name="VFD_CONTROL_6" usage="rp_blit"> + <reg32 offset="0xa006" name="VFD_CNTL_6" usage="rp_blit"> <!-- True if gl_PrimitiveID is read via the FS --> <bitfield name="PRIMID4PSEN" pos="0" type="boolean"/> </reg32> - <reg32 offset="0xa007" name="VFD_MODE_CNTL" usage="cmd"> + <reg32 offset="0xa007" name="VFD_RENDER_MODE" usage="cmd"> <bitfield name="RENDER_MODE" low="0" high="2" type="a6xx_render_mode"/> </reg32> - <reg32 offset="0xa008" name="VFD_MULTIVIEW_CNTL" type="a6xx_multiview_cntl" usage="rp_blit"/> - <reg32 offset="0xa009" name="VFD_ADD_OFFSET" usage="cmd"> + <reg32 offset="0xa008" name="VFD_STEREO_RENDERING_CNTL" type="a6xx_stereo_rendering_cntl" usage="rp_blit"/> + <reg32 offset="0xa009" name="VFD_MODE_CNTL" usage="cmd"> <!-- add VFD_INDEX_OFFSET to REGID4VTX --> <bitfield name="VERTEX" pos="0" type="boolean"/> <!-- add VFD_INSTANCE_START_OFFSET to REGID4INST --> @@ -4546,14 +2394,14 @@ to upconvert to 32b float internally? <reg32 offset="0xa00e" name="VFD_INDEX_OFFSET" usage="rp_blit"/> <reg32 offset="0xa00f" name="VFD_INSTANCE_START_OFFSET" usage="rp_blit"/> - <array offset="0xa010" name="VFD_FETCH" stride="4" length="32" usage="rp_blit"> + <array offset="0xa010" name="VFD_VERTEX_BUFFER" stride="4" length="32" usage="rp_blit"> <reg64 offset="0x0" name="BASE" type="address" align="1"/> <reg32 offset="0x2" name="SIZE" type="uint"/> <reg32 offset="0x3" name="STRIDE" low="0" high="11" type="uint"/> </array> - <array offset="0xa090" name="VFD_DECODE" stride="2" length="32" usage="rp_blit"> + <array offset="0xa090" name="VFD_FETCH_INSTR" stride="2" length="32" usage="rp_blit"> <reg32 offset="0x0" name="INSTR"> - <!-- IDX and byte OFFSET into VFD_FETCH --> + <!-- IDX and byte OFFSET into VFD_VERTEX_BUFFER --> <bitfield name="IDX" low="0" high="4" type="uint"/> <bitfield name="OFFSET" low="5" high="16"/> <bitfield name="INSTANCED" pos="17" type="boolean"/> @@ -4573,7 +2421,7 @@ to upconvert to 32b float internally? <reg32 offset="0xa0f8" name="VFD_POWER_CNTL" low="0" high="2" usage="rp_blit"/> - <reg32 offset="0xa600" name="VFD_UNKNOWN_A600" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xa600" name="VFD_DBG_ECO_CNTL" variants="A7XX-" usage="cmd"/> <reg32 offset="0xa601" name="VFD_ADDR_MODE_CNTL" type="a5xx_address_mode"/> <array offset="0xa610" name="VFD_PERFCTR_VFD_SEL" stride="1" length="8" variants="A6XX"/> @@ -4588,7 +2436,7 @@ to upconvert to 32b float internally? <value value="1" name="THREAD128"/> </enum> - <bitset name="a6xx_sp_xs_ctrl_reg0" inline="yes"> + <bitset name="a6xx_sp_xs_cntl_0" inline="yes"> <!-- if set to SINGLE, only use 1 concurrent wave on each SP --> <bitfield name="THREADMODE" pos="0" type="a3xx_threadmode"/> <!-- @@ -4620,7 +2468,7 @@ to upconvert to 32b float internally? --> <bitfield name="BINDLESS_TEX" pos="0" type="boolean"/> <bitfield name="BINDLESS_SAMP" pos="1" type="boolean"/> - <bitfield name="BINDLESS_IBO" pos="2" type="boolean"/> + <bitfield name="BINDLESS_UAV" pos="2" type="boolean"/> <bitfield name="BINDLESS_UBO" pos="3" type="boolean"/> <bitfield name="ENABLED" pos="8" type="boolean"/> @@ -4630,17 +2478,17 @@ to upconvert to 32b float internally? --> <bitfield name="NTEX" low="9" high="16" type="uint"/> <bitfield name="NSAMP" low="17" high="21" type="uint"/> - <bitfield name="NIBO" low="22" high="28" type="uint"/> + <bitfield name="NUAV" low="22" high="28" type="uint"/> </bitset> - <bitset name="a6xx_sp_xs_prim_cntl" inline="yes"> + <bitset name="a6xx_sp_xs_output_cntl" inline="yes"> <!-- # of VS outputs including pos/psize --> <bitfield name="OUT" low="0" high="5" type="uint"/> <!-- FLAGS_REGID only for GS --> <bitfield name="FLAGS_REGID" low="6" high="13" type="a3xx_regid"/> </bitset> - <reg32 offset="0xa800" name="SP_VS_CTRL_REG0" type="a6xx_sp_xs_ctrl_reg0" usage="rp_blit"> + <reg32 offset="0xa800" name="SP_VS_CNTL_0" type="a6xx_sp_xs_cntl_0" usage="rp_blit"> <!-- This field actually controls all geometry stages. TCS, TES, and GS must have the same mergedregs setting as VS. @@ -4665,10 +2513,10 @@ to upconvert to 32b float internally? </reg32> <!-- bitmask of true/false conditions for VS brac.N instructions, bit N corresponds to brac.N --> - <reg32 offset="0xa801" name="SP_VS_BRANCH_COND" type="hex"/> + <reg32 offset="0xa801" name="SP_VS_BOOLEAN_CF_MASK" type="hex"/> <!-- # of VS outputs including pos/psize --> - <reg32 offset="0xa802" name="SP_VS_PRIMITIVE_CNTL" type="a6xx_sp_xs_prim_cntl" usage="rp_blit"/> - <array offset="0xa803" name="SP_VS_OUT" stride="1" length="16" usage="rp_blit"> + <reg32 offset="0xa802" name="SP_VS_OUTPUT_CNTL" type="a6xx_sp_xs_output_cntl" usage="rp_blit"/> + <array offset="0xa803" name="SP_VS_OUTPUT" stride="1" length="16" usage="rp_blit"> <reg32 offset="0x0" name="REG"> <bitfield name="A_REGID" low="0" high="7" type="a3xx_regid"/> <bitfield name="A_COMPMASK" low="8" high="11" type="hex"/> @@ -4678,12 +2526,12 @@ to upconvert to 32b float internally? </array> <!-- Starting with a5xx, position/psize outputs from shader end up in the - SP_VS_OUT map, with highest OUTLOCn position. (Generally they are + SP_VS_OUTPUT map, with highest OUTLOCn position. (Generally they are the last entries too, except when gl_PointCoord is used, blob inserts an extra varying after, but with a lower OUTLOC position. If present, psize is last, preceded by position. --> - <array offset="0xa813" name="SP_VS_VPC_DST" stride="1" length="8" usage="rp_blit"> + <array offset="0xa813" name="SP_VS_VPC_DEST" stride="1" length="8" usage="rp_blit"> <reg32 offset="0x0" name="REG"> <bitfield name="OUTLOC0" low="0" high="7" type="uint"/> <bitfield name="OUTLOC1" low="8" high="15" type="uint"/> @@ -4752,7 +2600,7 @@ to upconvert to 32b float internally? </bitfield> </bitset> - <bitset name="a6xx_sp_xs_pvt_mem_hw_stack_offset" inline="yes"> + <bitset name="a6xx_sp_xs_pvt_mem_stack_offset" inline="yes"> <doc> This seems to be be the equivalent of HWSTACKOFFSET in a3xx. The ldp/stp offset formula above isn't affected by @@ -4763,18 +2611,18 @@ to upconvert to 32b float internally? <bitfield name="OFFSET" low="0" high="18" shr="11"/> </bitset> - <reg32 offset="0xa81b" name="SP_VS_OBJ_FIRST_EXEC_OFFSET" type="uint" usage="rp_blit"/> - <reg64 offset="0xa81c" name="SP_VS_OBJ_START" type="address" align="32" usage="rp_blit"/> + <reg32 offset="0xa81b" name="SP_VS_PROGRAM_COUNTER_OFFSET" type="uint" usage="rp_blit"/> + <reg64 offset="0xa81c" name="SP_VS_BASE" type="address" align="32" usage="rp_blit"/> <reg32 offset="0xa81e" name="SP_VS_PVT_MEM_PARAM" type="a6xx_sp_xs_pvt_mem_param" usage="rp_blit"/> - <reg64 offset="0xa81f" name="SP_VS_PVT_MEM_ADDR" type="waddress" align="32" usage="rp_blit"/> + <reg64 offset="0xa81f" name="SP_VS_PVT_MEM_BASE" type="waddress" align="32" usage="rp_blit"/> <reg32 offset="0xa821" name="SP_VS_PVT_MEM_SIZE" type="a6xx_sp_xs_pvt_mem_size" usage="rp_blit"/> - <reg32 offset="0xa822" name="SP_VS_TEX_COUNT" low="0" high="7" type="uint" usage="rp_blit"/> + <reg32 offset="0xa822" name="SP_VS_TSIZE" low="0" high="7" type="uint" usage="rp_blit"/> <reg32 offset="0xa823" name="SP_VS_CONFIG" type="a6xx_sp_xs_config" usage="rp_blit"/> - <reg32 offset="0xa824" name="SP_VS_INSTRLEN" low="0" high="27" type="uint" usage="rp_blit"/> - <reg32 offset="0xa825" name="SP_VS_PVT_MEM_HW_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_hw_stack_offset" usage="rp_blit"/> - <reg32 offset="0xa82d" name="SP_VS_VGPR_CONFIG" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xa824" name="SP_VS_INSTR_SIZE" low="0" high="27" type="uint" usage="rp_blit"/> + <reg32 offset="0xa825" name="SP_VS_PVT_MEM_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_stack_offset" usage="rp_blit"/> + <reg32 offset="0xa82d" name="SP_VS_VGS_CNTL" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xa830" name="SP_HS_CTRL_REG0" type="a6xx_sp_xs_ctrl_reg0" usage="rp_blit"> + <reg32 offset="0xa830" name="SP_HS_CNTL_0" type="a6xx_sp_xs_cntl_0" usage="rp_blit"> <!-- There is no mergedregs bit, that comes from the VS. --> <bitfield name="EARLYPREAMBLE" pos="20" type="boolean"/> </reg32> @@ -4782,32 +2630,32 @@ to upconvert to 32b float internally? Total size of local storage in dwords divided by the wave size. The maximum value is 64. With the wave size being always 64 for HS, the maximum size of local storage should be: - 64 (wavesize) * 64 (SP_HS_WAVE_INPUT_SIZE) * 4 = 16k + 64 (wavesize) * 64 (SP_HS_CNTL_1) * 4 = 16k --> - <reg32 offset="0xa831" name="SP_HS_WAVE_INPUT_SIZE" low="0" high="7" type="uint" usage="rp_blit"/> - <reg32 offset="0xa832" name="SP_HS_BRANCH_COND" type="hex" usage="rp_blit"/> + <reg32 offset="0xa831" name="SP_HS_CNTL_1" low="0" high="7" type="uint" usage="rp_blit"/> + <reg32 offset="0xa832" name="SP_HS_BOOLEAN_CF_MASK" type="hex" usage="rp_blit"/> <!-- TODO: exact same layout as 0xa81b-0xa825 --> - <reg32 offset="0xa833" name="SP_HS_OBJ_FIRST_EXEC_OFFSET" type="uint" usage="rp_blit"/> - <reg64 offset="0xa834" name="SP_HS_OBJ_START" type="address" align="32" usage="rp_blit"/> + <reg32 offset="0xa833" name="SP_HS_PROGRAM_COUNTER_OFFSET" type="uint" usage="rp_blit"/> + <reg64 offset="0xa834" name="SP_HS_BASE" type="address" align="32" usage="rp_blit"/> <reg32 offset="0xa836" name="SP_HS_PVT_MEM_PARAM" type="a6xx_sp_xs_pvt_mem_param" usage="rp_blit"/> - <reg64 offset="0xa837" name="SP_HS_PVT_MEM_ADDR" type="waddress" align="32" usage="rp_blit"/> + <reg64 offset="0xa837" name="SP_HS_PVT_MEM_BASE" type="waddress" align="32" usage="rp_blit"/> <reg32 offset="0xa839" name="SP_HS_PVT_MEM_SIZE" type="a6xx_sp_xs_pvt_mem_size" usage="rp_blit"/> - <reg32 offset="0xa83a" name="SP_HS_TEX_COUNT" low="0" high="7" type="uint" usage="rp_blit"/> + <reg32 offset="0xa83a" name="SP_HS_TSIZE" low="0" high="7" type="uint" usage="rp_blit"/> <reg32 offset="0xa83b" name="SP_HS_CONFIG" type="a6xx_sp_xs_config" usage="rp_blit"/> - <reg32 offset="0xa83c" name="SP_HS_INSTRLEN" low="0" high="27" type="uint" usage="rp_blit"/> - <reg32 offset="0xa83d" name="SP_HS_PVT_MEM_HW_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_hw_stack_offset" usage="rp_blit"/> - <reg32 offset="0xa82f" name="SP_HS_VGPR_CONFIG" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xa83c" name="SP_HS_INSTR_SIZE" low="0" high="27" type="uint" usage="rp_blit"/> + <reg32 offset="0xa83d" name="SP_HS_PVT_MEM_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_stack_offset" usage="rp_blit"/> + <reg32 offset="0xa82f" name="SP_HS_VGS_CNTL" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xa840" name="SP_DS_CTRL_REG0" type="a6xx_sp_xs_ctrl_reg0" usage="rp_blit"> + <reg32 offset="0xa840" name="SP_DS_CNTL_0" type="a6xx_sp_xs_cntl_0" usage="rp_blit"> <!-- There is no mergedregs bit, that comes from the VS. --> <bitfield name="EARLYPREAMBLE" pos="20" type="boolean"/> </reg32> - <reg32 offset="0xa841" name="SP_DS_BRANCH_COND" type="hex"/> + <reg32 offset="0xa841" name="SP_DS_BOOLEAN_CF_MASK" type="hex"/> <!-- TODO: exact same layout as 0xa802-0xa81a --> - <reg32 offset="0xa842" name="SP_DS_PRIMITIVE_CNTL" type="a6xx_sp_xs_prim_cntl" usage="rp_blit"/> - <array offset="0xa843" name="SP_DS_OUT" stride="1" length="16" usage="rp_blit"> + <reg32 offset="0xa842" name="SP_DS_OUTPUT_CNTL" type="a6xx_sp_xs_output_cntl" usage="rp_blit"/> + <array offset="0xa843" name="SP_DS_OUTPUT" stride="1" length="16" usage="rp_blit"> <reg32 offset="0x0" name="REG"> <bitfield name="A_REGID" low="0" high="7" type="a3xx_regid"/> <bitfield name="A_COMPMASK" low="8" high="11" type="hex"/> @@ -4815,7 +2663,7 @@ to upconvert to 32b float internally? <bitfield name="B_COMPMASK" low="24" high="27" type="hex"/> </reg32> </array> - <array offset="0xa853" name="SP_DS_VPC_DST" stride="1" length="8" usage="rp_blit"> + <array offset="0xa853" name="SP_DS_VPC_DEST" stride="1" length="8" usage="rp_blit"> <reg32 offset="0x0" name="REG"> <bitfield name="OUTLOC0" low="0" high="7" type="uint"/> <bitfield name="OUTLOC1" low="8" high="15" type="uint"/> @@ -4825,22 +2673,22 @@ to upconvert to 32b float internally? </array> <!-- TODO: exact same layout as 0xa81b-0xa825 --> - <reg32 offset="0xa85b" name="SP_DS_OBJ_FIRST_EXEC_OFFSET" type="uint" usage="rp_blit"/> - <reg64 offset="0xa85c" name="SP_DS_OBJ_START" type="address" align="32" usage="rp_blit"/> + <reg32 offset="0xa85b" name="SP_DS_PROGRAM_COUNTER_OFFSET" type="uint" usage="rp_blit"/> + <reg64 offset="0xa85c" name="SP_DS_BASE" type="address" align="32" usage="rp_blit"/> <reg32 offset="0xa85e" name="SP_DS_PVT_MEM_PARAM" type="a6xx_sp_xs_pvt_mem_param" usage="rp_blit"/> - <reg64 offset="0xa85f" name="SP_DS_PVT_MEM_ADDR" type="waddress" align="32" usage="rp_blit"/> + <reg64 offset="0xa85f" name="SP_DS_PVT_MEM_BASE" type="waddress" align="32" usage="rp_blit"/> <reg32 offset="0xa861" name="SP_DS_PVT_MEM_SIZE" type="a6xx_sp_xs_pvt_mem_size" usage="rp_blit"/> - <reg32 offset="0xa862" name="SP_DS_TEX_COUNT" low="0" high="7" type="uint" usage="rp_blit"/> + <reg32 offset="0xa862" name="SP_DS_TSIZE" low="0" high="7" type="uint" usage="rp_blit"/> <reg32 offset="0xa863" name="SP_DS_CONFIG" type="a6xx_sp_xs_config" usage="rp_blit"/> - <reg32 offset="0xa864" name="SP_DS_INSTRLEN" low="0" high="27" type="uint" usage="rp_blit"/> - <reg32 offset="0xa865" name="SP_DS_PVT_MEM_HW_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_hw_stack_offset" usage="rp_blit"/> - <reg32 offset="0xa868" name="SP_DS_VGPR_CONFIG" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xa864" name="SP_DS_INSTR_SIZE" low="0" high="27" type="uint" usage="rp_blit"/> + <reg32 offset="0xa865" name="SP_DS_PVT_MEM_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_stack_offset" usage="rp_blit"/> + <reg32 offset="0xa868" name="SP_DS_VGS_CNTL" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xa870" name="SP_GS_CTRL_REG0" type="a6xx_sp_xs_ctrl_reg0" usage="rp_blit"> + <reg32 offset="0xa870" name="SP_GS_CNTL_0" type="a6xx_sp_xs_cntl_0" usage="rp_blit"> <!-- There is no mergedregs bit, that comes from the VS. --> <bitfield name="EARLYPREAMBLE" pos="20" type="boolean"/> </reg32> - <reg32 offset="0xa871" name="SP_GS_PRIM_SIZE" low="0" high="7" type="uint" usage="rp_blit"> + <reg32 offset="0xa871" name="SP_GS_CNTL_1" low="0" high="7" type="uint" usage="rp_blit"> <doc> Normally the size of the output of the last stage in dwords. It should be programmed as follows: @@ -4854,11 +2702,11 @@ to upconvert to 32b float internally? doesn't matter in practice. </doc> </reg32> - <reg32 offset="0xa872" name="SP_GS_BRANCH_COND" type="hex" usage="rp_blit"/> + <reg32 offset="0xa872" name="SP_GS_BOOLEAN_CF_MASK" type="hex" usage="rp_blit"/> <!-- TODO: exact same layout as 0xa802-0xa81a --> - <reg32 offset="0xa873" name="SP_GS_PRIMITIVE_CNTL" type="a6xx_sp_xs_prim_cntl" usage="rp_blit"/> - <array offset="0xa874" name="SP_GS_OUT" stride="1" length="16" usage="rp_blit"> + <reg32 offset="0xa873" name="SP_GS_OUTPUT_CNTL" type="a6xx_sp_xs_output_cntl" usage="rp_blit"/> + <array offset="0xa874" name="SP_GS_OUTPUT" stride="1" length="16" usage="rp_blit"> <reg32 offset="0x0" name="REG"> <bitfield name="A_REGID" low="0" high="7" type="a3xx_regid"/> <bitfield name="A_COMPMASK" low="8" high="11" type="hex"/> @@ -4867,7 +2715,7 @@ to upconvert to 32b float internally? </reg32> </array> - <array offset="0xa884" name="SP_GS_VPC_DST" stride="1" length="8" usage="rp_blit"> + <array offset="0xa884" name="SP_GS_VPC_DEST" stride="1" length="8" usage="rp_blit"> <reg32 offset="0x0" name="REG"> <bitfield name="OUTLOC0" low="0" high="7" type="uint"/> <bitfield name="OUTLOC1" low="8" high="15" type="uint"/> @@ -4877,29 +2725,29 @@ to upconvert to 32b float internally? </array> <!-- TODO: exact same layout as 0xa81b-0xa825 --> - <reg32 offset="0xa88c" name="SP_GS_OBJ_FIRST_EXEC_OFFSET" type="uint" usage="rp_blit"/> - <reg64 offset="0xa88d" name="SP_GS_OBJ_START" type="address" align="32" usage="rp_blit"/> + <reg32 offset="0xa88c" name="SP_GS_PROGRAM_COUNTER_OFFSET" type="uint" usage="rp_blit"/> + <reg64 offset="0xa88d" name="SP_GS_BASE" type="address" align="32" usage="rp_blit"/> <reg32 offset="0xa88f" name="SP_GS_PVT_MEM_PARAM" type="a6xx_sp_xs_pvt_mem_param" usage="rp_blit"/> - <reg64 offset="0xa890" name="SP_GS_PVT_MEM_ADDR" type="waddress" align="32" usage="rp_blit"/> + <reg64 offset="0xa890" name="SP_GS_PVT_MEM_BASE" type="waddress" align="32" usage="rp_blit"/> <reg32 offset="0xa892" name="SP_GS_PVT_MEM_SIZE" type="a6xx_sp_xs_pvt_mem_size" usage="rp_blit"/> - <reg32 offset="0xa893" name="SP_GS_TEX_COUNT" low="0" high="7" type="uint" usage="rp_blit"/> + <reg32 offset="0xa893" name="SP_GS_TSIZE" low="0" high="7" type="uint" usage="rp_blit"/> <reg32 offset="0xa894" name="SP_GS_CONFIG" type="a6xx_sp_xs_config" usage="rp_blit"/> - <reg32 offset="0xa895" name="SP_GS_INSTRLEN" low="0" high="27" type="uint" usage="rp_blit"/> - <reg32 offset="0xa896" name="SP_GS_PVT_MEM_HW_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_hw_stack_offset" usage="rp_blit"/> - <reg32 offset="0xa899" name="SP_GS_VGPR_CONFIG" variants="A7XX-" usage="cmd"/> - - <reg64 offset="0xa8a0" name="SP_VS_TEX_SAMP" type="address" align="16" usage="cmd"/> - <reg64 offset="0xa8a2" name="SP_HS_TEX_SAMP" type="address" align="16" usage="cmd"/> - <reg64 offset="0xa8a4" name="SP_DS_TEX_SAMP" type="address" align="16" usage="cmd"/> - <reg64 offset="0xa8a6" name="SP_GS_TEX_SAMP" type="address" align="16" usage="cmd"/> - <reg64 offset="0xa8a8" name="SP_VS_TEX_CONST" type="address" align="64" usage="cmd"/> - <reg64 offset="0xa8aa" name="SP_HS_TEX_CONST" type="address" align="64" usage="cmd"/> - <reg64 offset="0xa8ac" name="SP_DS_TEX_CONST" type="address" align="64" usage="cmd"/> - <reg64 offset="0xa8ae" name="SP_GS_TEX_CONST" type="address" align="64" usage="cmd"/> + <reg32 offset="0xa895" name="SP_GS_INSTR_SIZE" low="0" high="27" type="uint" usage="rp_blit"/> + <reg32 offset="0xa896" name="SP_GS_PVT_MEM_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_stack_offset" usage="rp_blit"/> + <reg32 offset="0xa899" name="SP_GS_VGS_CNTL" variants="A7XX-" usage="cmd"/> + + <reg64 offset="0xa8a0" name="SP_VS_SAMPLER_BASE" type="address" align="16" usage="cmd"/> + <reg64 offset="0xa8a2" name="SP_HS_SAMPLER_BASE" type="address" align="16" usage="cmd"/> + <reg64 offset="0xa8a4" name="SP_DS_SAMPLER_BASE" type="address" align="16" usage="cmd"/> + <reg64 offset="0xa8a6" name="SP_GS_SAMPLER_BASE" type="address" align="16" usage="cmd"/> + <reg64 offset="0xa8a8" name="SP_VS_TEXMEMOBJ_BASE" type="address" align="64" usage="cmd"/> + <reg64 offset="0xa8aa" name="SP_HS_TEXMEMOBJ_BASE" type="address" align="64" usage="cmd"/> + <reg64 offset="0xa8ac" name="SP_DS_TEXMEMOBJ_BASE" type="address" align="64" usage="cmd"/> + <reg64 offset="0xa8ae" name="SP_GS_TEXMEMOBJ_BASE" type="address" align="64" usage="cmd"/> <!-- TODO: 4 unknown bool registers 0xa8c0-0xa8c3 --> - <reg32 offset="0xa980" name="SP_FS_CTRL_REG0" type="a6xx_sp_xs_ctrl_reg0" usage="rp_blit"> + <reg32 offset="0xa980" name="SP_PS_CNTL_0" type="a6xx_sp_xs_cntl_0" usage="rp_blit"> <bitfield name="THREADSIZE" pos="20" type="a6xx_threadsize"/> <bitfield name="UNK21" pos="21" type="boolean"/> <bitfield name="VARYING" pos="22" type="boolean"/> @@ -4909,8 +2757,7 @@ to upconvert to 32b float internally? fine derivatives and quad subgroup ops. </doc> </bitfield> - <!-- note: vk blob uses bit24 --> - <bitfield name="UNK24" pos="24" type="boolean"/> + <bitfield name="INOUTREGOVERLAP" pos="24" type="boolean"/> <bitfield name="UNK25" pos="25" type="boolean"/> <bitfield name="PIXLODENABLE" pos="26" type="boolean"> <doc> @@ -4923,12 +2770,12 @@ to upconvert to 32b float internally? <bitfield name="EARLYPREAMBLE" pos="28" type="boolean"/> <bitfield name="MERGEDREGS" pos="31" type="boolean"/> </reg32> - <reg32 offset="0xa981" name="SP_FS_BRANCH_COND" type="hex"/> - <reg32 offset="0xa982" name="SP_FS_OBJ_FIRST_EXEC_OFFSET" type="uint" usage="rp_blit"/> - <reg64 offset="0xa983" name="SP_FS_OBJ_START" type="address" align="32" usage="rp_blit"/> - <reg32 offset="0xa985" name="SP_FS_PVT_MEM_PARAM" type="a6xx_sp_xs_pvt_mem_param" usage="rp_blit"/> - <reg64 offset="0xa986" name="SP_FS_PVT_MEM_ADDR" type="waddress" align="32" usage="rp_blit"/> - <reg32 offset="0xa988" name="SP_FS_PVT_MEM_SIZE" type="a6xx_sp_xs_pvt_mem_size" usage="rp_blit"/> + <reg32 offset="0xa981" name="SP_PS_BOOLEAN_CF_MASK" type="hex"/> + <reg32 offset="0xa982" name="SP_PS_PROGRAM_COUNTER_OFFSET" type="uint" usage="rp_blit"/> + <reg64 offset="0xa983" name="SP_PS_BASE" type="address" align="32" usage="rp_blit"/> + <reg32 offset="0xa985" name="SP_PS_PVT_MEM_PARAM" type="a6xx_sp_xs_pvt_mem_param" usage="rp_blit"/> + <reg64 offset="0xa986" name="SP_PS_PVT_MEM_BASE" type="waddress" align="32" usage="rp_blit"/> + <reg32 offset="0xa988" name="SP_PS_PVT_MEM_SIZE" type="a6xx_sp_xs_pvt_mem_size" usage="rp_blit"/> <reg32 offset="0xa989" name="SP_BLEND_CNTL" usage="rp_blit"> <!-- per-mrt enable bit --> @@ -4948,7 +2795,7 @@ to upconvert to 32b float internally? <bitfield name="SRGB_MRT6" pos="6" type="boolean"/> <bitfield name="SRGB_MRT7" pos="7" type="boolean"/> </reg32> - <reg32 offset="0xa98b" name="SP_FS_RENDER_COMPONENTS" usage="rp_blit"> + <reg32 offset="0xa98b" name="SP_PS_OUTPUT_MASK" usage="rp_blit"> <bitfield name="RT0" low="0" high="3"/> <bitfield name="RT1" low="4" high="7"/> <bitfield name="RT2" low="8" high="11"/> @@ -4958,17 +2805,17 @@ to upconvert to 32b float internally? <bitfield name="RT6" low="24" high="27"/> <bitfield name="RT7" low="28" high="31"/> </reg32> - <reg32 offset="0xa98c" name="SP_FS_OUTPUT_CNTL0" usage="rp_blit"> + <reg32 offset="0xa98c" name="SP_PS_OUTPUT_CNTL" usage="rp_blit"> <bitfield name="DUAL_COLOR_IN_ENABLE" pos="0" type="boolean"/> <bitfield name="DEPTH_REGID" low="8" high="15" type="a3xx_regid"/> <bitfield name="SAMPMASK_REGID" low="16" high="23" type="a3xx_regid"/> <bitfield name="STENCILREF_REGID" low="24" high="31" type="a3xx_regid"/> </reg32> - <reg32 offset="0xa98d" name="SP_FS_OUTPUT_CNTL1" usage="rp_blit"> + <reg32 offset="0xa98d" name="SP_PS_MRT_CNTL" usage="rp_blit"> <bitfield name="MRT" low="0" high="3" type="uint"/> </reg32> - <array offset="0xa98e" name="SP_FS_OUTPUT" stride="1" length="8" usage="rp_blit"> + <array offset="0xa98e" name="SP_PS_OUTPUT" stride="1" length="8" usage="rp_blit"> <doc>per MRT</doc> <reg32 offset="0x0" name="REG"> <bitfield name="REGID" low="0" high="7" type="a3xx_regid"/> @@ -4976,7 +2823,7 @@ to upconvert to 32b float internally? </reg32> </array> - <array offset="0xa996" name="SP_FS_MRT" stride="1" length="8" usage="rp_blit"> + <array offset="0xa996" name="SP_PS_MRT" stride="1" length="8" usage="rp_blit"> <reg32 offset="0" name="REG"> <bitfield name="COLOR_FORMAT" low="0" high="7" type="a6xx_format"/> <bitfield name="COLOR_SINT" pos="8" type="boolean"/> @@ -4985,7 +2832,7 @@ to upconvert to 32b float internally? </reg32> </array> - <reg32 offset="0xa99e" name="SP_FS_PREFETCH_CNTL" usage="rp_blit"> + <reg32 offset="0xa99e" name="SP_PS_INITIAL_TEX_LOAD_CNTL" usage="rp_blit"> <bitfield name="COUNT" low="0" high="2" type="uint"/> <bitfield name="IJ_WRITE_DISABLE" pos="3" type="boolean"/> <doc> @@ -5002,7 +2849,7 @@ to upconvert to 32b float internally? <!-- Blob never uses it --> <bitfield name="CONSTSLOTID4COORD" low="16" high="24" type="uint" variants="A7XX-"/> </reg32> - <array offset="0xa99f" name="SP_FS_PREFETCH" stride="1" length="4" variants="A6XX" usage="rp_blit"> + <array offset="0xa99f" name="SP_PS_INITIAL_TEX_LOAD" stride="1" length="4" variants="A6XX" usage="rp_blit"> <reg32 offset="0" name="CMD" variants="A6XX"> <bitfield name="SRC" low="0" high="6" type="uint"/> <bitfield name="SAMP_ID" low="7" high="10" type="uint"/> @@ -5016,7 +2863,7 @@ to upconvert to 32b float internally? <bitfield name="CMD" low="29" high="31" type="a6xx_tex_prefetch_cmd"/> </reg32> </array> - <array offset="0xa99f" name="SP_FS_PREFETCH" stride="1" length="4" variants="A7XX-" usage="rp_blit"> + <array offset="0xa99f" name="SP_PS_INITIAL_TEX_LOAD" stride="1" length="4" variants="A7XX-" usage="rp_blit"> <reg32 offset="0" name="CMD" variants="A7XX-"> <bitfield name="SRC" low="0" high="6" type="uint"/> <bitfield name="SAMP_ID" low="7" high="9" type="uint"/> @@ -5028,22 +2875,23 @@ to upconvert to 32b float internally? <bitfield name="CMD" low="26" high="29" type="a6xx_tex_prefetch_cmd"/> </reg32> </array> - <array offset="0xa9a3" name="SP_FS_BINDLESS_PREFETCH" stride="1" length="4" usage="rp_blit"> + <array offset="0xa9a3" name="SP_PS_INITIAL_TEX_INDEX" stride="1" length="4" usage="rp_blit"> <reg32 offset="0" name="CMD"> <bitfield name="SAMP_ID" low="0" high="15" type="uint"/> <bitfield name="TEX_ID" low="16" high="31" type="uint"/> </reg32> </array> - <reg32 offset="0xa9a7" name="SP_FS_TEX_COUNT" low="0" high="7" type="uint" usage="rp_blit"/> + <reg32 offset="0xa9a7" name="SP_PS_TSIZE" low="0" high="7" type="uint" usage="rp_blit"/> <reg32 offset="0xa9a8" name="SP_UNKNOWN_A9A8" low="0" high="16" usage="cmd"/> <!-- always 0x0 ? --> - <reg32 offset="0xa9a9" name="SP_FS_PVT_MEM_HW_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_hw_stack_offset" usage="rp_blit"/> + <reg32 offset="0xa9a9" name="SP_PS_PVT_MEM_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_stack_offset" usage="rp_blit"/> + <reg32 offset="0xa9ab" name="SP_PS_UNKNOWN_A9AB" variants="A7XX-" usage="cmd"/> <!-- TODO: unknown bool register at 0xa9aa, likely same as 0xa8c0-0xa8c3 but for FS --> - <reg32 offset="0xa9b0" name="SP_CS_CTRL_REG0" type="a6xx_sp_xs_ctrl_reg0" usage="cmd"> + <reg32 offset="0xa9b0" name="SP_CS_CNTL_0" type="a6xx_sp_xs_cntl_0" usage="cmd"> <bitfield name="THREADSIZE" pos="20" type="a6xx_threadsize"/> <!-- seems to make SP use less concurrent threads when possible? --> <bitfield name="UNK21" pos="21" type="boolean"/> @@ -5053,8 +2901,15 @@ to upconvert to 32b float internally? <bitfield name="MERGEDREGS" pos="31" type="boolean"/> </reg32> + <enum name="a6xx_const_ram_mode"> + <value value="0x0" name="CONSTLEN_128"/> + <value value="0x1" name="CONSTLEN_192"/> + <value value="0x2" name="CONSTLEN_256"/> + <value value="0x3" name="CONSTLEN_512"/> <!-- a7xx only --> + </enum> + <!-- set for compute shaders --> - <reg32 offset="0xa9b1" name="SP_CS_UNKNOWN_A9B1" usage="cmd"> + <reg32 offset="0xa9b1" name="SP_CS_CNTL_1" usage="cmd"> <bitfield name="SHARED_SIZE" low="0" high="4" type="uint"> <doc> If 0 - all 32k of shared storage is enabled, otherwise @@ -5065,32 +2920,36 @@ to upconvert to 32b float internally? always return 0) </doc> </bitfield> - <bitfield name="UNK5" pos="5" type="boolean"/> - <!-- always 1 ? --> - <bitfield name="UNK6" pos="6" type="boolean"/> + <bitfield name="CONSTANTRAMMODE" low="5" high="6" type="a6xx_const_ram_mode"> + <doc> + This defines the split between consts and local + memory in the Local Buffer. The programmed value + must be at least the actual CONSTLEN. + </doc> + </bitfield> </reg32> - <reg32 offset="0xa9b2" name="SP_CS_BRANCH_COND" type="hex" usage="cmd"/> - <reg32 offset="0xa9b3" name="SP_CS_OBJ_FIRST_EXEC_OFFSET" type="uint" usage="cmd"/> - <reg64 offset="0xa9b4" name="SP_CS_OBJ_START" type="address" align="32" usage="cmd"/> + <reg32 offset="0xa9b2" name="SP_CS_BOOLEAN_CF_MASK" type="hex" usage="cmd"/> + <reg32 offset="0xa9b3" name="SP_CS_PROGRAM_COUNTER_OFFSET" type="uint" usage="cmd"/> + <reg64 offset="0xa9b4" name="SP_CS_BASE" type="address" align="32" usage="cmd"/> <reg32 offset="0xa9b6" name="SP_CS_PVT_MEM_PARAM" type="a6xx_sp_xs_pvt_mem_param" usage="cmd"/> - <reg64 offset="0xa9b7" name="SP_CS_PVT_MEM_ADDR" align="32" usage="cmd"/> + <reg64 offset="0xa9b7" name="SP_CS_PVT_MEM_BASE" align="32" usage="cmd"/> <reg32 offset="0xa9b9" name="SP_CS_PVT_MEM_SIZE" type="a6xx_sp_xs_pvt_mem_size" usage="cmd"/> - <reg32 offset="0xa9ba" name="SP_CS_TEX_COUNT" low="0" high="7" type="uint" usage="cmd"/> + <reg32 offset="0xa9ba" name="SP_CS_TSIZE" low="0" high="7" type="uint" usage="cmd"/> <reg32 offset="0xa9bb" name="SP_CS_CONFIG" type="a6xx_sp_xs_config" usage="cmd"/> - <reg32 offset="0xa9bc" name="SP_CS_INSTRLEN" low="0" high="27" type="uint" usage="cmd"/> - <reg32 offset="0xa9bd" name="SP_CS_PVT_MEM_HW_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_hw_stack_offset" usage="cmd"/> + <reg32 offset="0xa9bc" name="SP_CS_INSTR_SIZE" low="0" high="27" type="uint" usage="cmd"/> + <reg32 offset="0xa9bd" name="SP_CS_PVT_MEM_STACK_OFFSET" type="a6xx_sp_xs_pvt_mem_stack_offset" usage="cmd"/> <reg32 offset="0xa9be" name="SP_CS_UNKNOWN_A9BE" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xa9c5" name="SP_CS_VGPR_CONFIG" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xa9c5" name="SP_CS_VGS_CNTL" variants="A7XX-" usage="cmd"/> - <!-- new in a6xx gen4, matches HLSQ_CS_CNTL_0 --> - <reg32 offset="0xa9c2" name="SP_CS_CNTL_0" usage="cmd"> + <!-- new in a6xx gen4, matches SP_CS_CONST_CONFIG_0 --> + <reg32 offset="0xa9c2" name="SP_CS_WIE_CNTL_0" usage="cmd"> <bitfield name="WGIDCONSTID" low="0" high="7" type="a3xx_regid"/> <bitfield name="WGSIZECONSTID" low="8" high="15" type="a3xx_regid"/> <bitfield name="WGOFFSETCONSTID" low="16" high="23" type="a3xx_regid"/> <bitfield name="LOCALIDREGID" low="24" high="31" type="a3xx_regid"/> </reg32> - <!-- new in a6xx gen4, matches HLSQ_CS_CNTL_1 --> - <reg32 offset="0xa9c3" name="SP_CS_CNTL_1" variants="A6XX" usage="cmd"> + <!-- new in a6xx gen4, matches SP_CS_WGE_CNTL --> + <reg32 offset="0xa9c3" name="SP_CS_WIE_CNTL_1" variants="A6XX" usage="cmd"> <!-- gl_LocalInvocationIndex --> <bitfield name="LINEARLOCALIDREGID" low="0" high="7" type="a3xx_regid"/> <!-- a650 has 6 "SP cores" (but 3 "SP"). this makes it use only @@ -5102,7 +2961,18 @@ to upconvert to 32b float internally? <bitfield name="THREADSIZE_SCALAR" pos="10" type="boolean"/> </reg32> - <reg32 offset="0xa9c3" name="SP_CS_CNTL_1" variants="A7XX-" usage="cmd"> + <enum name="a7xx_workitem_rast_order"> + <value value="0x0" name="WORKITEMRASTORDER_LINEAR"/> + <doc> + This is a fixed tiling, with 4x4 invocation outer tiles + containing 2x2 invocation inner tiles. The intent is to + improve cache locality with textures and images accessed + using gl_LocalInvocationID. + </doc> + <value value="0x1" name="WORKITEMRASTORDER_TILED"/> + </enum> + + <reg32 offset="0xa9c3" name="SP_CS_WIE_CNTL_1" variants="A7XX-" usage="cmd"> <!-- gl_LocalInvocationIndex --> <bitfield name="LINEARLOCALIDREGID" low="0" high="7" type="a3xx_regid"/> <!-- Must match SP_CS_CTRL --> @@ -5110,18 +2980,16 @@ to upconvert to 32b float internally? <!-- 1 thread per wave (would hang if THREAD128 is also set) --> <bitfield name="THREADSIZE_SCALAR" pos="9" type="boolean"/> - <!-- Affects getone. If enabled, getone sometimes executed 1? less times - than there are subgroups. - --> - <bitfield name="UNK15" pos="15" type="boolean"/> + <doc>How invocations/fibers within a workgroup are tiled.</doc> + <bitfield name="WORKITEMRASTORDER" pos="15" type="a7xx_workitem_rast_order"/> </reg32> <!-- TODO: two 64kb aligned addresses at a9d0/a9d2 --> - <reg64 offset="0xa9e0" name="SP_FS_TEX_SAMP" type="address" align="16" usage="rp_blit"/> - <reg64 offset="0xa9e2" name="SP_CS_TEX_SAMP" type="address" align="16" usage="cmd"/> - <reg64 offset="0xa9e4" name="SP_FS_TEX_CONST" type="address" align="64" usage="rp_blit"/> - <reg64 offset="0xa9e6" name="SP_CS_TEX_CONST" type="address" align="64" usage="cmd"/> + <reg64 offset="0xa9e0" name="SP_PS_SAMPLER_BASE" type="address" align="16" usage="rp_blit"/> + <reg64 offset="0xa9e2" name="SP_CS_SAMPLER_BASE" type="address" align="16" usage="cmd"/> + <reg64 offset="0xa9e4" name="SP_PS_TEXMEMOBJ_BASE" type="address" align="64" usage="rp_blit"/> + <reg64 offset="0xa9e6" name="SP_CS_TEXMEMOBJ_BASE" type="address" align="64" usage="cmd"/> <enum name="a6xx_bindless_descriptor_size"> <doc> @@ -5146,18 +3014,19 @@ to upconvert to 32b float internally? </array> <!-- - IBO state for compute shader: + UAV state for compute shader: --> - <reg64 offset="0xa9f2" name="SP_CS_IBO" type="address" align="16"/> - <reg32 offset="0xaa00" name="SP_CS_IBO_COUNT" low="0" high="6" type="uint"/> + <reg64 offset="0xa9f2" name="SP_CS_UAV_BASE" type="address" align="16" variants="A6XX"/> + <reg64 offset="0xa9f8" name="SP_CS_UAV_BASE" type="address" align="16" variants="A7XX"/> + <reg32 offset="0xaa00" name="SP_CS_USIZE" low="0" high="6" type="uint"/> <!-- Correlated with avgs/uvgs usage in FS --> - <reg32 offset="0xaa01" name="SP_FS_VGPR_CONFIG" type="uint" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xaa01" name="SP_PS_VGS_CNTL" type="uint" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xaa02" name="SP_PS_ALIASED_COMPONENTS_CONTROL" variants="A7XX-" usage="cmd"> + <reg32 offset="0xaa02" name="SP_PS_OUTPUT_CONST_CNTL" variants="A7XX-" usage="cmd"> <bitfield name="ENABLED" pos="0" type="boolean"/> </reg32> - <reg32 offset="0xaa03" name="SP_PS_ALIASED_COMPONENTS" variants="A7XX-" usage="cmd"> + <reg32 offset="0xaa03" name="SP_PS_OUTPUT_CONST_MASK" variants="A7XX-" usage="cmd"> <doc> Specify for which components the output color should be read from alias, e.g. for: @@ -5167,7 +3036,7 @@ to upconvert to 32b float internally? alias.1.b32.0 r1.x, c4.x alias.1.b32.0 r0.x, c0.x - the SP_PS_ALIASED_COMPONENTS would be 0x00001111 + the SP_PS_OUTPUT_CONST_MASK would be 0x00001111 </doc> <bitfield name="RT0" low="0" high="3"/> @@ -5193,7 +3062,7 @@ to upconvert to 32b float internally? <value value="0x2" name="ISAMMODE_GL"/> </enum> - <reg32 offset="0xab00" name="SP_MODE_CONTROL" usage="rp_blit"> + <reg32 offset="0xab00" name="SP_MODE_CNTL" usage="rp_blit"> <!-- When set, half register loads from the constant file will load a 32-bit value (so hc0.y loads the same value as c0.y) @@ -5210,16 +3079,16 @@ to upconvert to 32b float internally? <reg32 offset="0xab01" name="SP_UNKNOWN_AB01" variants="A7XX-" usage="cmd"/> <reg32 offset="0xab02" name="SP_UNKNOWN_AB02" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xab04" name="SP_FS_CONFIG" type="a6xx_sp_xs_config" usage="rp_blit"/> - <reg32 offset="0xab05" name="SP_FS_INSTRLEN" low="0" high="27" type="uint" usage="rp_blit"/> + <reg32 offset="0xab04" name="SP_PS_CONFIG" type="a6xx_sp_xs_config" usage="rp_blit"/> + <reg32 offset="0xab05" name="SP_PS_INSTR_SIZE" low="0" high="27" type="uint" usage="rp_blit"/> - <array offset="0xab10" name="SP_BINDLESS_BASE" stride="2" length="5" variants="A6XX" usage="rp_blit"> + <array offset="0xab10" name="SP_GFX_BINDLESS_BASE" stride="2" length="5" variants="A6XX" usage="rp_blit"> <reg64 offset="0" name="DESCRIPTOR" variants="A6XX"> <bitfield name="DESC_SIZE" low="0" high="1" type="a6xx_bindless_descriptor_size"/> <bitfield name="ADDR" low="2" high="63" shr="2" type="address"/> </reg64> </array> - <array offset="0xab0a" name="SP_BINDLESS_BASE" stride="2" length="8" variants="A7XX-" usage="rp_blit"> + <array offset="0xab0a" name="SP_GFX_BINDLESS_BASE" stride="2" length="8" variants="A7XX-" usage="rp_blit"> <reg64 offset="0" name="DESCRIPTOR" variants="A7XX-"> <bitfield name="DESC_SIZE" low="0" high="1" type="a6xx_bindless_descriptor_size"/> <bitfield name="ADDR" low="2" high="63" shr="2" type="address"/> @@ -5227,15 +3096,15 @@ to upconvert to 32b float internally? </array> <!-- - Combined IBO state for 3d pipe, used for Image and SSBO write/atomic - instructions VS/HS/DS/GS/FS. See SP_CS_IBO_* for compute shaders. + Combined UAV state for 3d pipe, used for Image and SSBO write/atomic + instructions VS/HS/DS/GS/FS. See SP_CS_UAV_BASE_* for compute shaders. --> - <reg64 offset="0xab1a" name="SP_IBO" type="address" align="16" usage="cmd"/> - <reg32 offset="0xab20" name="SP_IBO_COUNT" low="0" high="6" type="uint" usage="cmd"/> + <reg64 offset="0xab1a" name="SP_GFX_UAV_BASE" type="address" align="16" usage="cmd"/> + <reg32 offset="0xab20" name="SP_GFX_USIZE" low="0" high="6" type="uint" usage="cmd"/> <reg32 offset="0xab22" name="SP_UNKNOWN_AB22" variants="A7XX-" usage="cmd"/> - <bitset name="a6xx_sp_2d_dst_format" inline="yes"> + <bitset name="a6xx_sp_a2d_output_info" inline="yes"> <bitfield name="NORM" pos="0" type="boolean"/> <bitfield name="SINT" pos="1" type="boolean"/> <bitfield name="UINT" pos="2" type="boolean"/> @@ -5248,8 +3117,8 @@ to upconvert to 32b float internally? <bitfield name="MASK" low="12" high="15"/> </bitset> - <reg32 offset="0xacc0" name="SP_2D_DST_FORMAT" type="a6xx_sp_2d_dst_format" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xa9bf" name="SP_2D_DST_FORMAT" type="a6xx_sp_2d_dst_format" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xacc0" name="SP_A2D_OUTPUT_INFO" type="a6xx_sp_a2d_output_info" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xa9bf" name="SP_A2D_OUTPUT_INFO" type="a6xx_sp_a2d_output_info" variants="A7XX-" usage="rp_blit"/> <reg32 offset="0xae00" name="SP_DBG_ECO_CNTL" usage="cmd"/> <reg32 offset="0xae01" name="SP_ADDR_MODE_CNTL" pos="0" type="a5xx_address_mode"/> @@ -5257,16 +3126,16 @@ to upconvert to 32b float internally? <!-- TODO: valid bits 0x3c3f, see kernel --> </reg32> <reg32 offset="0xae03" name="SP_CHICKEN_BITS" usage="cmd"/> - <reg32 offset="0xae04" name="SP_FLOAT_CNTL" usage="cmd"> + <reg32 offset="0xae04" name="SP_NC_MODE_CNTL_2" usage="cmd"> <bitfield name="F16_NO_INF" pos="3" type="boolean"/> </reg32> <reg32 offset="0xae06" name="SP_UNKNOWN_AE06" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xae08" name="SP_UNKNOWN_AE08" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xae09" name="SP_UNKNOWN_AE09" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xae0a" name="SP_UNKNOWN_AE0A" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xae08" name="SP_CHICKEN_BITS_1" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xae09" name="SP_CHICKEN_BITS_2" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xae0a" name="SP_CHICKEN_BITS_3" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xae0f" name="SP_PERFCTR_ENABLE" usage="cmd"> + <reg32 offset="0xae0f" name="SP_PERFCTR_SHADER_MASK" usage="cmd"> <!-- some perfcntrs are affected by a per-stage enable bit (PERF_SP_ALU_WORKING_CYCLES for example) TODO: verify position of HS/DS/GS bits --> @@ -5281,7 +3150,7 @@ to upconvert to 32b float internally? <array offset="0xae60" name="SP_PERFCTR_HLSQ_SEL" stride="1" length="6" variants="A7XX-"/> <reg32 offset="0xae6a" name="SP_UNKNOWN_AE6A" variants="A7XX-" usage="cmd"/> <reg32 offset="0xae6b" name="SP_UNKNOWN_AE6B" variants="A7XX-" usage="cmd"/> - <reg32 offset="0xae6c" name="SP_UNKNOWN_AE6C" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xae6c" name="SP_HLSQ_DBG_ECO_CNTL" variants="A7XX-" usage="cmd"/> <reg32 offset="0xae6d" name="SP_READ_SEL" variants="A7XX-"> <bitfield name="LOCATION" low="18" high="19" type="a7xx_state_location"/> <bitfield name="PIPE" low="16" high="17" type="a7xx_pipe"/> @@ -5301,33 +3170,44 @@ to upconvert to 32b float internally? "a6xx_sp_ps_tp_cluster" but this actually specifies the border color base for compute shaders. --> - <reg64 offset="0xb180" name="SP_PS_TP_BORDER_COLOR_BASE_ADDR" type="address" align="128" usage="cmd"/> + <reg64 offset="0xb180" name="TPL1_CS_BORDER_COLOR_BASE" type="address" align="128" usage="cmd"/> <reg32 offset="0xb182" name="SP_UNKNOWN_B182" low="0" high="2" usage="cmd"/> <reg32 offset="0xb183" name="SP_UNKNOWN_B183" low="0" high="23" usage="cmd"/> <reg32 offset="0xb190" name="SP_UNKNOWN_B190"/> <reg32 offset="0xb191" name="SP_UNKNOWN_B191"/> - <!-- could be all the stuff below here is actually TPL1?? --> - - <reg32 offset="0xb300" name="SP_TP_RAS_MSAA_CNTL" usage="rp_blit"> + <reg32 offset="0xb300" name="TPL1_RAS_MSAA_CNTL" usage="rp_blit"> <bitfield name="SAMPLES" low="0" high="1" type="a3xx_msaa_samples"/> <bitfield name="UNK2" low="2" high="3"/> </reg32> - <reg32 offset="0xb301" name="SP_TP_DEST_MSAA_CNTL" usage="rp_blit"> + <reg32 offset="0xb301" name="TPL1_DEST_MSAA_CNTL" usage="rp_blit"> <bitfield name="SAMPLES" low="0" high="1" type="a3xx_msaa_samples"/> <bitfield name="MSAA_DISABLE" pos="2" type="boolean"/> </reg32> <!-- looks to work in the same way as a5xx: --> - <reg64 offset="0xb302" name="SP_TP_BORDER_COLOR_BASE_ADDR" type="address" align="128" usage="cmd"/> - <reg32 offset="0xb304" name="SP_TP_SAMPLE_CONFIG" type="a6xx_sample_config" usage="rp_blit"/> - <reg32 offset="0xb305" name="SP_TP_SAMPLE_LOCATION_0" type="a6xx_sample_locations" usage="rp_blit"/> - <reg32 offset="0xb306" name="SP_TP_SAMPLE_LOCATION_1" type="a6xx_sample_locations" usage="rp_blit"/> - <reg32 offset="0xb307" name="SP_TP_WINDOW_OFFSET" type="a6xx_reg_xy" usage="rp_blit"/> - <reg32 offset="0xb309" name="SP_TP_MODE_CNTL" usage="cmd"> + <reg64 offset="0xb302" name="TPL1_GFX_BORDER_COLOR_BASE" type="address" align="128" usage="cmd"/> + <reg32 offset="0xb304" name="TPL1_MSAA_SAMPLE_POS_CNTL" type="a6xx_msaa_sample_pos_cntl" usage="rp_blit"/> + <reg32 offset="0xb305" name="TPL1_PROGRAMMABLE_MSAA_POS_0" type="a6xx_programmable_msaa_pos" usage="rp_blit"/> + <reg32 offset="0xb306" name="TPL1_PROGRAMMABLE_MSAA_POS_1" type="a6xx_programmable_msaa_pos" usage="rp_blit"/> + <reg32 offset="0xb307" name="TPL1_WINDOW_OFFSET" type="a6xx_reg_xy" usage="rp_blit"/> + + <enum name="a6xx_coord_round"> + <value value="0" name="COORD_TRUNCATE"/> + <value value="1" name="COORD_ROUND_NEAREST_EVEN"/> + </enum> + + <enum name="a6xx_nearest_mode"> + <value value="0" name="ROUND_CLAMP_TRUNCATE"/> + <value value="1" name="CLAMP_ROUND_TRUNCATE"/> + </enum> + + <reg32 offset="0xb309" name="TPL1_MODE_CNTL" usage="cmd"> <bitfield name="ISAMMODE" low="0" high="1" type="a6xx_isam_mode"/> - <bitfield name="UNK3" low="2" high="7"/> + <bitfield name="TEXCOORDROUNDMODE" pos="2" type="a6xx_coord_round"/> + <bitfield name="NEARESTMIPSNAP" pos="5" type="a6xx_nearest_mode"/> + <bitfield name="DESTDATATYPEOVERRIDE" pos="7" type="boolean"/> </reg32> <reg32 offset="0xb310" name="SP_UNKNOWN_B310" variants="A7XX-" usage="cmd"/> @@ -5336,42 +3216,45 @@ to upconvert to 32b float internally? badly named or the functionality moved in a6xx. But downstream kernel calls this "a6xx_sp_ps_tp_2d_cluster" --> - <reg32 offset="0xb4c0" name="SP_PS_2D_SRC_INFO" type="a6xx_2d_src_surf_info" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb4c1" name="SP_PS_2D_SRC_SIZE" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb4c0" name="TPL1_A2D_SRC_TEXTURE_INFO" type="a6xx_a2d_src_texture_info" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb4c1" name="TPL1_A2D_SRC_TEXTURE_SIZE" variants="A6XX" usage="rp_blit"> <bitfield name="WIDTH" low="0" high="14" type="uint"/> <bitfield name="HEIGHT" low="15" high="29" type="uint"/> </reg32> - <reg64 offset="0xb4c2" name="SP_PS_2D_SRC" type="address" align="16" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb4c4" name="SP_PS_2D_SRC_PITCH" variants="A6XX" usage="rp_blit"> + <reg64 offset="0xb4c2" name="TPL1_A2D_SRC_TEXTURE_BASE" type="address" align="16" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb4c4" name="TPL1_A2D_SRC_TEXTURE_PITCH" variants="A6XX" usage="rp_blit"> <bitfield name="UNK0" low="0" high="8"/> <bitfield name="PITCH" low="9" high="23" shr="6" type="uint"/> </reg32> - <reg32 offset="0xb2c0" name="SP_PS_2D_SRC_INFO" type="a6xx_2d_src_surf_info" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xb2c1" name="SP_PS_2D_SRC_SIZE" variants="A7XX"> + <reg32 offset="0xb2c0" name="TPL1_A2D_SRC_TEXTURE_INFO" type="a6xx_a2d_src_texture_info" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xb2c1" name="TPL1_A2D_SRC_TEXTURE_SIZE" variants="A7XX"> <bitfield name="WIDTH" low="0" high="14" type="uint"/> <bitfield name="HEIGHT" low="15" high="29" type="uint"/> </reg32> - <reg64 offset="0xb2c2" name="SP_PS_2D_SRC" type="address" align="16" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xb2c4" name="SP_PS_2D_SRC_PITCH" variants="A7XX"> - <bitfield name="UNK0" low="0" high="8"/> - <bitfield name="PITCH" low="9" high="23" shr="6" type="uint"/> + <reg64 offset="0xb2c2" name="TPL1_A2D_SRC_TEXTURE_BASE" type="address" align="16" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xb2c4" name="TPL1_A2D_SRC_TEXTURE_PITCH" variants="A7XX"> + <!-- + Bits from 3..9 must be zero unless 'TPL1_A2D_BLT_CNTL::TYPE' + is A6XX_TEX_IMG_BUFFER, which allows for lower alignment. + --> + <bitfield name="PITCH" low="3" high="23" type="uint"/> </reg32> <!-- planes for NV12, etc. (TODO: not tested) --> - <reg64 offset="0xb4c5" name="SP_PS_2D_SRC_PLANE1" type="address" align="16" variants="A6XX"/> - <reg32 offset="0xb4c7" name="SP_PS_2D_SRC_PLANE_PITCH" low="0" high="11" shr="6" type="uint" variants="A6XX"/> - <reg64 offset="0xb4c8" name="SP_PS_2D_SRC_PLANE2" type="address" align="16" variants="A6XX"/> + <reg64 offset="0xb4c5" name="TPL1_A2D_SRC_TEXTURE_BASE_1" type="address" align="16" variants="A6XX"/> + <reg32 offset="0xb4c7" name="TPL1_A2D_SRC_TEXTURE_PITCH_1" low="0" high="11" shr="6" type="uint" variants="A6XX"/> + <reg64 offset="0xb4c8" name="TPL1_A2D_SRC_TEXTURE_BASE_2" type="address" align="16" variants="A6XX"/> - <reg64 offset="0xb2c5" name="SP_PS_2D_SRC_PLANE1" type="address" align="16" variants="A7XX-"/> - <reg32 offset="0xb2c7" name="SP_PS_2D_SRC_PLANE_PITCH" low="0" high="11" shr="6" type="uint" variants="A7XX-"/> - <reg64 offset="0xb2c8" name="SP_PS_2D_SRC_PLANE2" type="address" align="16" variants="A7XX-"/> + <reg64 offset="0xb2c5" name="TPL1_A2D_SRC_TEXTURE_BASE_1" type="address" align="16" variants="A7XX-"/> + <reg32 offset="0xb2c7" name="TPL1_A2D_SRC_TEXTURE_PITCH_1" low="0" high="11" shr="6" type="uint" variants="A7XX-"/> + <reg64 offset="0xb2c8" name="TPL1_A2D_SRC_TEXTURE_BASE_2" type="address" align="16" variants="A7XX-"/> - <reg64 offset="0xb4ca" name="SP_PS_2D_SRC_FLAGS" type="address" align="16" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb4cc" name="SP_PS_2D_SRC_FLAGS_PITCH" low="0" high="7" shr="6" type="uint" variants="A6XX" usage="rp_blit"/> + <reg64 offset="0xb4ca" name="TPL1_A2D_SRC_TEXTURE_FLAG_BASE" type="address" align="16" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb4cc" name="TPL1_A2D_SRC_TEXTURE_FLAG_PITCH" low="0" high="7" shr="6" type="uint" variants="A6XX" usage="rp_blit"/> - <reg64 offset="0xb2ca" name="SP_PS_2D_SRC_FLAGS" type="address" align="16" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xb2cc" name="SP_PS_2D_SRC_FLAGS_PITCH" low="0" high="7" shr="6" type="uint" variants="A7XX-" usage="rp_blit"/> + <reg64 offset="0xb2ca" name="TPL1_A2D_SRC_TEXTURE_FLAG_BASE" type="address" align="16" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xb2cc" name="TPL1_A2D_SRC_TEXTURE_FLAG_PITCH" low="0" high="7" shr="6" type="uint" variants="A7XX-" usage="rp_blit"/> <reg32 offset="0xb4cd" name="SP_PS_UNKNOWN_B4CD" low="6" high="31" variants="A6XX"/> <reg32 offset="0xb4ce" name="SP_PS_UNKNOWN_B4CE" low="0" high="31" variants="A6XX"/> @@ -5383,8 +3266,12 @@ to upconvert to 32b float internally? <reg32 offset="0xb2ce" name="SP_PS_UNKNOWN_B4CE" low="0" high="31" variants="A7XX"/> <reg32 offset="0xb2cf" name="SP_PS_UNKNOWN_B4CF" low="0" high="30" variants="A7XX"/> <reg32 offset="0xb2d0" name="SP_PS_UNKNOWN_B4D0" low="0" high="29" variants="A7XX"/> - <reg32 offset="0xb2d1" name="SP_PS_2D_WINDOW_OFFSET" type="a6xx_reg_xy" variants="A7XX"/> - <reg32 offset="0xb2d2" name="SP_PS_UNKNOWN_B2D2" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xb2d1" name="TPL1_A2D_WINDOW_OFFSET" type="a6xx_reg_xy" variants="A7XX"/> + <reg32 offset="0xb2d2" name="TPL1_A2D_BLT_CNTL" variants="A7XX-" usage="rp_blit"> + <bitfield name="RAW_COPY" pos="0" type="boolean"/> + <bitfield name="START_OFFSET_TEXELS" low="16" high="21"/> + <bitfield name="TYPE" low="29" high="31" type="a6xx_tex_type"/> + </reg32> <reg32 offset="0xab21" name="SP_WINDOW_OFFSET" type="a6xx_reg_xy" variants="A7XX-" usage="rp_blit"/> <!-- always 0x100000 or 0x1000000? --> @@ -5422,34 +3309,44 @@ to upconvert to 32b float internally? <!-- TODO: 4 more perfcntr sel at 0xb620 ? --> - <bitset name="a6xx_hlsq_xs_cntl" inline="yes"> + <bitset name="a6xx_xs_const_config" inline="yes"> <bitfield name="CONSTLEN" low="0" high="7" shr="2" type="uint"/> <bitfield name="ENABLED" pos="8" type="boolean"/> <bitfield name="READ_IMM_SHARED_CONSTS" pos="9" type="boolean" variants="A7XX-"/> </bitset> - <reg32 offset="0xb800" name="HLSQ_VS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb801" name="HLSQ_HS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb802" name="HLSQ_DS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb803" name="HLSQ_GS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb800" name="SP_VS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb801" name="SP_HS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb802" name="SP_DS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb803" name="SP_GS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xa827" name="HLSQ_VS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xa83f" name="HLSQ_HS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xa867" name="HLSQ_DS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xa898" name="HLSQ_GS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa827" name="SP_VS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa83f" name="SP_HS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa867" name="SP_DS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa898" name="SP_GS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xa9aa" name="HLSQ_FS_UNKNOWN_A9AA" variants="A7XX-" usage="rp_blit"> - <!-- Tentatively named, appears to disable consts being loaded via CP_LOAD_STATE6_FRAG --> - <bitfield name="CONSTS_LOAD_DISABLE" pos="0" type="boolean"/> + <reg32 offset="0xa9aa" name="SP_RENDER_CNTL" variants="A7XX-" usage="rp_blit"> + <bitfield name="FS_DISABLE" pos="0" type="boolean"/> </reg32> - <!-- Always 0 --> - <reg32 offset="0xa9ac" name="HLSQ_UNKNOWN_A9AC" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xa9ac" name="SP_DITHER_CNTL" variants="A7XX-" usage="cmd"> + <bitfield name="DITHER_MODE_MRT0" low="0" high="1" type="adreno_rb_dither_mode"/> + <bitfield name="DITHER_MODE_MRT1" low="2" high="3" type="adreno_rb_dither_mode"/> + <bitfield name="DITHER_MODE_MRT2" low="4" high="5" type="adreno_rb_dither_mode"/> + <bitfield name="DITHER_MODE_MRT3" low="6" high="7" type="adreno_rb_dither_mode"/> + <bitfield name="DITHER_MODE_MRT4" low="8" high="9" type="adreno_rb_dither_mode"/> + <bitfield name="DITHER_MODE_MRT5" low="10" high="11" type="adreno_rb_dither_mode"/> + <bitfield name="DITHER_MODE_MRT6" low="12" high="13" type="adreno_rb_dither_mode"/> + <bitfield name="DITHER_MODE_MRT7" low="14" high="15" type="adreno_rb_dither_mode"/> + </reg32> - <!-- Used in VK_KHR_fragment_shading_rate --> - <reg32 offset="0xa9ad" name="HLSQ_UNKNOWN_A9AD" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xa9ad" name="SP_VRS_CONFIG" variants="A7XX-" usage="rp_blit"> + <bitfield name="PIPELINE_FSR_ENABLE" pos="0" type="boolean"/> + <bitfield name="ATTACHMENT_FSR_ENABLE" pos="1" type="boolean"/> + <bitfield name="PRIMITIVE_FSR_ENABLE" pos="3" type="boolean"/> + </reg32> - <reg32 offset="0xa9ae" name="HLSQ_UNKNOWN_A9AE" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xa9ae" name="SP_PS_CNTL_1" variants="A7XX-" usage="rp_blit"> <bitfield name="SYSVAL_REGS_COUNT" low="0" high="7" type="uint"/> <!-- UNK8 is set on a730/a740 --> <bitfield name="UNK8" pos="8" type="boolean"/> @@ -5462,94 +3359,94 @@ to upconvert to 32b float internally? <reg32 offset="0xb823" name="HLSQ_LOAD_STATE_GEOM_DATA"/> - <bitset name="a6xx_hlsq_fs_cntl_0" inline="yes"> + <bitset name="a6xx_sp_ps_wave_cntl" inline="yes"> <!-- must match SP_FS_CTRL --> <bitfield name="THREADSIZE" pos="0" type="a6xx_threadsize"/> <bitfield name="VARYINGS" pos="1" type="boolean"/> <bitfield name="UNK2" low="2" high="11"/> </bitset> - <bitset name="a6xx_hlsq_control_3_reg" inline="yes"> + <bitset name="a6xx_sp_reg_prog_id_1" inline="yes"> <!-- register loaded with position (bary.f) --> <bitfield name="IJ_PERSP_PIXEL" low="0" high="7" type="a3xx_regid"/> <bitfield name="IJ_LINEAR_PIXEL" low="8" high="15" type="a3xx_regid"/> <bitfield name="IJ_PERSP_CENTROID" low="16" high="23" type="a3xx_regid"/> <bitfield name="IJ_LINEAR_CENTROID" low="24" high="31" type="a3xx_regid"/> </bitset> - <bitset name="a6xx_hlsq_control_4_reg" inline="yes"> + <bitset name="a6xx_sp_reg_prog_id_2" inline="yes"> <bitfield name="IJ_PERSP_SAMPLE" low="0" high="7" type="a3xx_regid"/> <bitfield name="IJ_LINEAR_SAMPLE" low="8" high="15" type="a3xx_regid"/> <bitfield name="XYCOORDREGID" low="16" high="23" type="a3xx_regid"/> <bitfield name="ZWCOORDREGID" low="24" high="31" type="a3xx_regid"/> </bitset> - <bitset name="a6xx_hlsq_control_5_reg" inline="yes"> + <bitset name="a6xx_sp_reg_prog_id_3" inline="yes"> <bitfield name="LINELENGTHREGID" low="0" high="7" type="a3xx_regid"/> <bitfield name="FOVEATIONQUALITYREGID" low="8" high="15" type="a3xx_regid"/> </bitset> - <reg32 offset="0xb980" type="a6xx_hlsq_fs_cntl_0" name="HLSQ_FS_CNTL_0" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb980" type="a6xx_sp_ps_wave_cntl" name="SP_PS_WAVE_CNTL" variants="A6XX" usage="rp_blit"/> <reg32 offset="0xb981" name="HLSQ_UNKNOWN_B981" pos="0" type="boolean" variants="A6XX"/> <!-- never used by blob --> - <reg32 offset="0xb982" name="HLSQ_CONTROL_1_REG" low="0" high="2" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb982" name="SP_LB_PARAM_LIMIT" low="0" high="2" variants="A6XX" usage="rp_blit"> <!-- Sets the maximum number of primitives allowed in one FS wave minus one, similarly to the A3xx field, except that it's not necessary to set it to anything but the maximum, since the hardware will simply emit smaller waves when it runs out of space. --> <bitfield name="PRIMALLOCTHRESHOLD" low="0" high="2" type="uint"/> </reg32> - <reg32 offset="0xb983" name="HLSQ_CONTROL_2_REG" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb983" name="SP_REG_PROG_ID_0" variants="A6XX" usage="rp_blit"> <bitfield name="FACEREGID" low="0" high="7" type="a3xx_regid"/> <!-- SAMPLEID is loaded into a half-precision register: --> <bitfield name="SAMPLEID" low="8" high="15" type="a3xx_regid"/> <bitfield name="SAMPLEMASK" low="16" high="23" type="a3xx_regid"/> <bitfield name="CENTERRHW" low="24" high="31" type="a3xx_regid"/> </reg32> - <reg32 offset="0xb984" type="a6xx_hlsq_control_3_reg" name="HLSQ_CONTROL_3_REG" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb985" type="a6xx_hlsq_control_4_reg" name="HLSQ_CONTROL_4_REG" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb986" type="a6xx_hlsq_control_5_reg" name="HLSQ_CONTROL_5_REG" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb987" name="HLSQ_CS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A6XX" usage="cmd"/> - <reg32 offset="0xa9c6" type="a6xx_hlsq_fs_cntl_0" name="HLSQ_FS_CNTL_0" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xa9c7" name="HLSQ_CONTROL_1_REG" low="0" high="2" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xb984" type="a6xx_sp_reg_prog_id_1" name="SP_REG_PROG_ID_1" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb985" type="a6xx_sp_reg_prog_id_2" name="SP_REG_PROG_ID_2" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb986" type="a6xx_sp_reg_prog_id_3" name="SP_REG_PROG_ID_3" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb987" name="SP_CS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A6XX" usage="cmd"/> + <reg32 offset="0xa9c6" type="a6xx_sp_ps_wave_cntl" name="SP_PS_WAVE_CNTL" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa9c7" name="SP_LB_PARAM_LIMIT" low="0" high="2" variants="A7XX-" usage="rp_blit"> <bitfield name="PRIMALLOCTHRESHOLD" low="0" high="2" type="uint"/> </reg32> - <reg32 offset="0xa9c8" name="HLSQ_CONTROL_2_REG" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xa9c8" name="SP_REG_PROG_ID_0" variants="A7XX-" usage="rp_blit"> <bitfield name="FACEREGID" low="0" high="7" type="a3xx_regid"/> <!-- SAMPLEID is loaded into a half-precision register: --> <bitfield name="SAMPLEID" low="8" high="15" type="a3xx_regid"/> <bitfield name="SAMPLEMASK" low="16" high="23" type="a3xx_regid"/> <bitfield name="CENTERRHW" low="24" high="31" type="a3xx_regid"/> </reg32> - <reg32 offset="0xa9c9" type="a6xx_hlsq_control_3_reg" name="HLSQ_CONTROL_3_REG" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xa9ca" type="a6xx_hlsq_control_4_reg" name="HLSQ_CONTROL_4_REG" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xa9cb" type="a6xx_hlsq_control_5_reg" name="HLSQ_CONTROL_5_REG" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xa9cd" name="HLSQ_CS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A7XX-" usage="cmd"/> + <reg32 offset="0xa9c9" type="a6xx_sp_reg_prog_id_1" name="SP_REG_PROG_ID_1" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa9ca" type="a6xx_sp_reg_prog_id_2" name="SP_REG_PROG_ID_2" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa9cb" type="a6xx_sp_reg_prog_id_3" name="SP_REG_PROG_ID_3" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa9cd" name="SP_CS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A7XX-" usage="cmd"/> <!-- TODO: what does KERNELDIM do exactly (blob sets it differently from turnip) --> - <reg32 offset="0xb990" name="HLSQ_CS_NDRANGE_0" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb990" name="SP_CS_NDRANGE_0" variants="A6XX" usage="rp_blit"> <bitfield name="KERNELDIM" low="0" high="1" type="uint"/> <!-- localsize is value minus one: --> <bitfield name="LOCALSIZEX" low="2" high="11" type="uint"/> <bitfield name="LOCALSIZEY" low="12" high="21" type="uint"/> <bitfield name="LOCALSIZEZ" low="22" high="31" type="uint"/> </reg32> - <reg32 offset="0xb991" name="HLSQ_CS_NDRANGE_1" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb991" name="SP_CS_NDRANGE_1" variants="A6XX" usage="rp_blit"> <bitfield name="GLOBALSIZE_X" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xb992" name="HLSQ_CS_NDRANGE_2" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb992" name="SP_CS_NDRANGE_2" variants="A6XX" usage="rp_blit"> <bitfield name="GLOBALOFF_X" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xb993" name="HLSQ_CS_NDRANGE_3" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb993" name="SP_CS_NDRANGE_3" variants="A6XX" usage="rp_blit"> <bitfield name="GLOBALSIZE_Y" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xb994" name="HLSQ_CS_NDRANGE_4" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb994" name="SP_CS_NDRANGE_4" variants="A6XX" usage="rp_blit"> <bitfield name="GLOBALOFF_Y" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xb995" name="HLSQ_CS_NDRANGE_5" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb995" name="SP_CS_NDRANGE_5" variants="A6XX" usage="rp_blit"> <bitfield name="GLOBALSIZE_Z" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xb996" name="HLSQ_CS_NDRANGE_6" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb996" name="SP_CS_NDRANGE_6" variants="A6XX" usage="rp_blit"> <bitfield name="GLOBALOFF_Z" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xb997" name="HLSQ_CS_CNTL_0" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb997" name="SP_CS_CONST_CONFIG_0" variants="A6XX" usage="rp_blit"> <!-- these are all vec3. first 3 need to be high regs - WGSIZECONSTID is the local size (from HLSQ_CS_NDRANGE_0) + WGSIZECONSTID is the local size (from SP_CS_NDRANGE_0) WGOFFSETCONSTID is WGIDCONSTID*WGSIZECONSTID --> <bitfield name="WGIDCONSTID" low="0" high="7" type="a3xx_regid"/> @@ -5557,7 +3454,7 @@ to upconvert to 32b float internally? <bitfield name="WGOFFSETCONSTID" low="16" high="23" type="a3xx_regid"/> <bitfield name="LOCALIDREGID" low="24" high="31" type="a3xx_regid"/> </reg32> - <reg32 offset="0xb998" name="HLSQ_CS_CNTL_1" variants="A6XX" usage="rp_blit"> + <reg32 offset="0xb998" name="SP_CS_WGE_CNTL" variants="A6XX" usage="rp_blit"> <!-- gl_LocalInvocationIndex --> <bitfield name="LINEARLOCALIDREGID" low="0" high="7" type="a3xx_regid"/> <!-- a650 has 6 "SP cores" (but 3 "SP"). this makes it use only @@ -5569,40 +3466,40 @@ to upconvert to 32b float internally? <bitfield name="THREADSIZE_SCALAR" pos="10" type="boolean"/> </reg32> <!--note: vulkan blob doesn't use these --> - <reg32 offset="0xb999" name="HLSQ_CS_KERNEL_GROUP_X" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb99a" name="HLSQ_CS_KERNEL_GROUP_Y" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xb99b" name="HLSQ_CS_KERNEL_GROUP_Z" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb999" name="SP_CS_KERNEL_GROUP_X" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb99a" name="SP_CS_KERNEL_GROUP_Y" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xb99b" name="SP_CS_KERNEL_GROUP_Z" variants="A6XX" usage="rp_blit"/> <!-- TODO: what does KERNELDIM do exactly (blob sets it differently from turnip) --> - <reg32 offset="0xa9d4" name="HLSQ_CS_NDRANGE_0" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xa9d4" name="SP_CS_NDRANGE_0" variants="A7XX-" usage="rp_blit"> <bitfield name="KERNELDIM" low="0" high="1" type="uint"/> <!-- localsize is value minus one: --> <bitfield name="LOCALSIZEX" low="2" high="11" type="uint"/> <bitfield name="LOCALSIZEY" low="12" high="21" type="uint"/> <bitfield name="LOCALSIZEZ" low="22" high="31" type="uint"/> </reg32> - <reg32 offset="0xa9d5" name="HLSQ_CS_NDRANGE_1" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xa9d5" name="SP_CS_NDRANGE_1" variants="A7XX-" usage="rp_blit"> <bitfield name="GLOBALSIZE_X" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xa9d6" name="HLSQ_CS_NDRANGE_2" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xa9d6" name="SP_CS_NDRANGE_2" variants="A7XX-" usage="rp_blit"> <bitfield name="GLOBALOFF_X" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xa9d7" name="HLSQ_CS_NDRANGE_3" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xa9d7" name="SP_CS_NDRANGE_3" variants="A7XX-" usage="rp_blit"> <bitfield name="GLOBALSIZE_Y" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xa9d8" name="HLSQ_CS_NDRANGE_4" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xa9d8" name="SP_CS_NDRANGE_4" variants="A7XX-" usage="rp_blit"> <bitfield name="GLOBALOFF_Y" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xa9d9" name="HLSQ_CS_NDRANGE_5" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xa9d9" name="SP_CS_NDRANGE_5" variants="A7XX-" usage="rp_blit"> <bitfield name="GLOBALSIZE_Z" low="0" high="31" type="uint"/> </reg32> - <reg32 offset="0xa9da" name="HLSQ_CS_NDRANGE_6" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xa9da" name="SP_CS_NDRANGE_6" variants="A7XX-" usage="rp_blit"> <bitfield name="GLOBALOFF_Z" low="0" high="31" type="uint"/> </reg32> <!--note: vulkan blob doesn't use these --> - <reg32 offset="0xa9dc" name="HLSQ_CS_KERNEL_GROUP_X" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xa9dd" name="HLSQ_CS_KERNEL_GROUP_Y" variants="A7XX-" usage="rp_blit"/> - <reg32 offset="0xa9de" name="HLSQ_CS_KERNEL_GROUP_Z" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa9dc" name="SP_CS_KERNEL_GROUP_X" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa9dd" name="SP_CS_KERNEL_GROUP_Y" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xa9de" name="SP_CS_KERNEL_GROUP_Z" variants="A7XX-" usage="rp_blit"/> <enum name="a7xx_cs_yalign"> <value name="CS_YALIGN_1" value="8"/> @@ -5611,19 +3508,29 @@ to upconvert to 32b float internally? <value name="CS_YALIGN_8" value="1"/> </enum> - <reg32 offset="0xa9db" name="HLSQ_CS_CNTL_1" variants="A7XX-" usage="rp_blit"> + <reg32 offset="0xa9db" name="SP_CS_WGE_CNTL" variants="A7XX-" usage="rp_blit"> <!-- gl_LocalInvocationIndex --> <bitfield name="LINEARLOCALIDREGID" low="0" high="7" type="a3xx_regid"/> <!-- Must match SP_CS_CTRL --> <bitfield name="THREADSIZE" pos="9" type="a6xx_threadsize"/> - <bitfield name="UNK11" pos="11" type="boolean"/> - <bitfield name="UNK22" pos="22" type="boolean"/> - <bitfield name="UNK26" pos="26" type="boolean"/> - <bitfield name="YALIGN" low="27" high="30" type="a7xx_cs_yalign"/> + <doc> + When this bit is enabled, the dispatch order interleaves + the z coordinate instead of launching all workgroups + with z=0, then all with z=1 and so on. + </doc> + <bitfield name="WORKGROUPRASTORDERZFIRSTEN" pos="11" type="boolean"/> + <doc> + When both fields are non-0 then the dispatcher uses + these tile sizes to launch workgroups in a tiled manner + when the x and y workgroup counts are + both more than 1. + </doc> + <bitfield name="WGTILEWIDTH" low="20" high="25"/> + <bitfield name="WGTILEHEIGHT" low="26" high="31"/> </reg32> - <reg32 offset="0xa9df" name="HLSQ_CS_LOCAL_SIZE" variants="A7XX-" usage="cmd"> - <!-- localsize is value minus one: --> + <reg32 offset="0xa9df" name="SP_CS_NDRANGE_7" variants="A7XX-" usage="cmd"> + <!-- The size of the last workgroup. localsize is value minus one: --> <bitfield name="LOCALSIZEX" low="2" high="11" type="uint"/> <bitfield name="LOCALSIZEY" low="12" high="21" type="uint"/> <bitfield name="LOCALSIZEZ" low="22" high="31" type="uint"/> @@ -5641,29 +3548,27 @@ to upconvert to 32b float internally? </reg64> </array> - <!-- new in a6xx gen4, mirror of SP_CS_UNKNOWN_A9B1? --> - <reg32 offset="0xb9d0" name="HLSQ_CS_UNKNOWN_B9D0" variants="A6XX" usage="cmd"> + <!-- new in a6xx gen4, mirror of SP_CS_CNTL_1? --> + <reg32 offset="0xb9d0" name="HLSQ_CS_CTRL_REG1" variants="A6XX" usage="cmd"> <bitfield name="SHARED_SIZE" low="0" high="4" type="uint"/> - <bitfield name="UNK5" pos="5" type="boolean"/> - <!-- always 1 ? --> - <bitfield name="UNK6" pos="6" type="boolean"/> + <bitfield name="CONSTANTRAMMODE" low="5" high="6" type="a6xx_const_ram_mode"/> </reg32> - <reg32 offset="0xbb00" name="HLSQ_DRAW_CMD" variants="A6XX"> + <reg32 offset="0xbb00" name="SP_DRAW_INITIATOR" variants="A6XX"> <bitfield name="STATE_ID" low="0" high="7"/> </reg32> - <reg32 offset="0xbb01" name="HLSQ_DISPATCH_CMD" variants="A6XX"> + <reg32 offset="0xbb01" name="SP_KERNEL_INITIATOR" variants="A6XX"> <bitfield name="STATE_ID" low="0" high="7"/> </reg32> - <reg32 offset="0xbb02" name="HLSQ_EVENT_CMD" variants="A6XX"> + <reg32 offset="0xbb02" name="SP_EVENT_INITIATOR" variants="A6XX"> <!-- I think only the low bit is actually used? --> <bitfield name="STATE_ID" low="16" high="23"/> <bitfield name="EVENT" low="0" high="6" type="vgt_event_type"/> </reg32> - <reg32 offset="0xbb08" name="HLSQ_INVALIDATE_CMD" variants="A6XX" usage="cmd"> + <reg32 offset="0xbb08" name="SP_UPDATE_CNTL" variants="A6XX" usage="cmd"> <doc> This register clears pending loads queued up by CP_LOAD_STATE6. Each bit resets a particular kind(s) of @@ -5678,8 +3583,8 @@ to upconvert to 32b float internally? <bitfield name="FS_STATE" pos="4" type="boolean"/> <bitfield name="CS_STATE" pos="5" type="boolean"/> - <bitfield name="CS_IBO" pos="6" type="boolean"/> - <bitfield name="GFX_IBO" pos="7" type="boolean"/> + <bitfield name="CS_UAV" pos="6" type="boolean"/> + <bitfield name="GFX_UAV" pos="7" type="boolean"/> <!-- Note: these only do something when HLSQ_SHARED_CONSTS is set to 1 --> <bitfield name="CS_SHARED_CONST" pos="19" type="boolean"/> @@ -5690,20 +3595,20 @@ to upconvert to 32b float internally? <bitfield name="GFX_BINDLESS" low="14" high="18" type="hex"/> </reg32> - <reg32 offset="0xab1c" name="HLSQ_DRAW_CMD" variants="A7XX-"> + <reg32 offset="0xab1c" name="SP_DRAW_INITIATOR" variants="A7XX-"> <bitfield name="STATE_ID" low="0" high="7"/> </reg32> - <reg32 offset="0xab1d" name="HLSQ_DISPATCH_CMD" variants="A7XX-"> + <reg32 offset="0xab1d" name="SP_KERNEL_INITIATOR" variants="A7XX-"> <bitfield name="STATE_ID" low="0" high="7"/> </reg32> - <reg32 offset="0xab1e" name="HLSQ_EVENT_CMD" variants="A7XX-"> + <reg32 offset="0xab1e" name="SP_EVENT_INITIATOR" variants="A7XX-"> <bitfield name="STATE_ID" low="16" high="23"/> <bitfield name="EVENT" low="0" high="6" type="vgt_event_type"/> </reg32> - <reg32 offset="0xab1f" name="HLSQ_INVALIDATE_CMD" variants="A7XX-" usage="cmd"> + <reg32 offset="0xab1f" name="SP_UPDATE_CNTL" variants="A7XX-" usage="cmd"> <doc> This register clears pending loads queued up by CP_LOAD_STATE6. Each bit resets a particular kind(s) of @@ -5718,18 +3623,18 @@ to upconvert to 32b float internally? <bitfield name="FS_STATE" pos="4" type="boolean"/> <bitfield name="CS_STATE" pos="5" type="boolean"/> - <bitfield name="CS_IBO" pos="6" type="boolean"/> - <bitfield name="GFX_IBO" pos="7" type="boolean"/> + <bitfield name="CS_UAV" pos="6" type="boolean"/> + <bitfield name="GFX_UAV" pos="7" type="boolean"/> <!-- SS6_BINDLESS: one bit per bindless base --> <bitfield name="CS_BINDLESS" low="9" high="16" type="hex"/> <bitfield name="GFX_BINDLESS" low="17" high="24" type="hex"/> </reg32> - <reg32 offset="0xbb10" name="HLSQ_FS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A6XX" usage="rp_blit"/> - <reg32 offset="0xab03" name="HLSQ_FS_CNTL" type="a6xx_hlsq_xs_cntl" variants="A7XX-" usage="rp_blit"/> + <reg32 offset="0xbb10" name="SP_PS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A6XX" usage="rp_blit"/> + <reg32 offset="0xab03" name="SP_PS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A7XX-" usage="rp_blit"/> - <array offset="0xab40" name="HLSQ_SHARED_CONSTS_IMM" stride="1" length="64" variants="A7XX-"/> + <array offset="0xab40" name="SP_SHARED_CONSTANT_GFX_0" stride="1" length="64" variants="A7XX-"/> <reg32 offset="0xbb11" name="HLSQ_SHARED_CONSTS" variants="A6XX" usage="cmd"> <doc> @@ -5738,7 +3643,7 @@ to upconvert to 32b float internally? const pool and 16 in the geometry const pool although only 8 are actually used (why?) and they are mapped to c504-c511 in each stage. Both VS and FS shared consts - are written using ST6_CONSTANTS/SB6_IBO, so that both + are written using ST6_CONSTANTS/SB6_UAV, so that both the geometry and FS shared consts can be written at once by using CP_LOAD_STATE6 rather than CP_LOAD_STATE6_FRAG/CP_LOAD_STATE6_GEOM. In addition @@ -5747,13 +3652,13 @@ to upconvert to 32b float internally? There is also a separate shared constant pool for CS, which is loaded through CP_LOAD_STATE6_FRAG with - ST6_UBO/ST6_IBO. However the only real difference for CS + ST6_UBO/ST6_UAV. However the only real difference for CS is the dword units. </doc> <bitfield name="ENABLE" pos="0" type="boolean"/> </reg32> - <!-- mirror of SP_BINDLESS_BASE --> + <!-- mirror of SP_GFX_BINDLESS_BASE --> <array offset="0xbb20" name="HLSQ_BINDLESS_BASE" stride="2" length="5" variants="A6XX" usage="cmd"> <reg64 offset="0" name="DESCRIPTOR"> <bitfield name="DESC_SIZE" low="0" high="1" type="a6xx_bindless_descriptor_size"/> @@ -5788,10 +3693,10 @@ to upconvert to 32b float internally? sequence. The sequence used internally for an event looks like: - write EVENT_CMD pipe register - write CP_EVENT_START - - write HLSQ_EVENT_CMD with event or HLSQ_DRAW_CMD - - write PC_EVENT_CMD with event or PC_DRAW_CMD - - write HLSQ_EVENT_CMD(CONTEXT_DONE) - - write PC_EVENT_CMD(CONTEXT_DONE) + - write SP_EVENT_INITIATOR with event or SP_DRAW_INITIATOR + - write PC_EVENT_INITIATOR with event or PC_DRAW_INITIATOR + - write SP_EVENT_INITIATOR(CONTEXT_DONE) + - write PC_EVENT_INITIATOR(CONTEXT_DONE) - write CP_EVENT_END Writing to CP_EVENT_END seems to actually trigger the context roll --> @@ -5809,193 +3714,6 @@ to upconvert to 32b float internally? </reg32> </domain> -<!-- Seems basically the same as a5xx, maybe move to common.xml.. --> -<domain name="A6XX_TEX_SAMP" width="32"> - <doc>Texture sampler dwords</doc> - <enum name="a6xx_tex_filter"> <!-- same as a4xx? --> - <value name="A6XX_TEX_NEAREST" value="0"/> - <value name="A6XX_TEX_LINEAR" value="1"/> - <value name="A6XX_TEX_ANISO" value="2"/> - <value name="A6XX_TEX_CUBIC" value="3"/> <!-- a650 only --> - </enum> - <enum name="a6xx_tex_clamp"> <!-- same as a4xx? --> - <value name="A6XX_TEX_REPEAT" value="0"/> - <value name="A6XX_TEX_CLAMP_TO_EDGE" value="1"/> - <value name="A6XX_TEX_MIRROR_REPEAT" value="2"/> - <value name="A6XX_TEX_CLAMP_TO_BORDER" value="3"/> - <value name="A6XX_TEX_MIRROR_CLAMP" value="4"/> - </enum> - <enum name="a6xx_tex_aniso"> <!-- same as a4xx? --> - <value name="A6XX_TEX_ANISO_1" value="0"/> - <value name="A6XX_TEX_ANISO_2" value="1"/> - <value name="A6XX_TEX_ANISO_4" value="2"/> - <value name="A6XX_TEX_ANISO_8" value="3"/> - <value name="A6XX_TEX_ANISO_16" value="4"/> - </enum> - <enum name="a6xx_reduction_mode"> - <value name="A6XX_REDUCTION_MODE_AVERAGE" value="0"/> - <value name="A6XX_REDUCTION_MODE_MIN" value="1"/> - <value name="A6XX_REDUCTION_MODE_MAX" value="2"/> - </enum> - - <reg32 offset="0" name="0"> - <bitfield name="MIPFILTER_LINEAR_NEAR" pos="0" type="boolean"/> - <bitfield name="XY_MAG" low="1" high="2" type="a6xx_tex_filter"/> - <bitfield name="XY_MIN" low="3" high="4" type="a6xx_tex_filter"/> - <bitfield name="WRAP_S" low="5" high="7" type="a6xx_tex_clamp"/> - <bitfield name="WRAP_T" low="8" high="10" type="a6xx_tex_clamp"/> - <bitfield name="WRAP_R" low="11" high="13" type="a6xx_tex_clamp"/> - <bitfield name="ANISO" low="14" high="16" type="a6xx_tex_aniso"/> - <bitfield name="LOD_BIAS" low="19" high="31" type="fixed" radix="8"/><!-- no idea how many bits for real --> - </reg32> - <reg32 offset="1" name="1"> - <bitfield name="CLAMPENABLE" pos="0" type="boolean"> - <doc> - clamp result to [0, 1] if the format is unorm or - [-1, 1] if the format is snorm, *after* - filtering. Has no effect for other formats. - </doc> - </bitfield> - <bitfield name="COMPARE_FUNC" low="1" high="3" type="adreno_compare_func"/> - <bitfield name="CUBEMAPSEAMLESSFILTOFF" pos="4" type="boolean"/> - <bitfield name="UNNORM_COORDS" pos="5" type="boolean"/> - <bitfield name="MIPFILTER_LINEAR_FAR" pos="6" type="boolean"/> - <bitfield name="MAX_LOD" low="8" high="19" type="ufixed" radix="8"/> - <bitfield name="MIN_LOD" low="20" high="31" type="ufixed" radix="8"/> - </reg32> - <reg32 offset="2" name="2"> - <bitfield name="REDUCTION_MODE" low="0" high="1" type="a6xx_reduction_mode"/> - <bitfield name="CHROMA_LINEAR" pos="5" type="boolean"/> - <bitfield name="BCOLOR" low="7" high="31"/> - </reg32> - <reg32 offset="3" name="3"/> -</domain> - -<domain name="A6XX_TEX_CONST" width="32" varset="chip"> - <doc>Texture constant dwords</doc> - <enum name="a6xx_tex_swiz"> <!-- same as a4xx? --> - <value name="A6XX_TEX_X" value="0"/> - <value name="A6XX_TEX_Y" value="1"/> - <value name="A6XX_TEX_Z" value="2"/> - <value name="A6XX_TEX_W" value="3"/> - <value name="A6XX_TEX_ZERO" value="4"/> - <value name="A6XX_TEX_ONE" value="5"/> - </enum> - <enum name="a6xx_tex_type"> <!-- same as a4xx? --> - <value name="A6XX_TEX_1D" value="0"/> - <value name="A6XX_TEX_2D" value="1"/> - <value name="A6XX_TEX_CUBE" value="2"/> - <value name="A6XX_TEX_3D" value="3"/> - <value name="A6XX_TEX_BUFFER" value="4"/> - </enum> - <reg32 offset="0" name="0"> - <bitfield name="TILE_MODE" low="0" high="1" type="a6xx_tile_mode"/> - <bitfield name="SRGB" pos="2" type="boolean"/> - <bitfield name="SWIZ_X" low="4" high="6" type="a6xx_tex_swiz"/> - <bitfield name="SWIZ_Y" low="7" high="9" type="a6xx_tex_swiz"/> - <bitfield name="SWIZ_Z" low="10" high="12" type="a6xx_tex_swiz"/> - <bitfield name="SWIZ_W" low="13" high="15" type="a6xx_tex_swiz"/> - <bitfield name="MIPLVLS" low="16" high="19" type="uint"/> - <!-- overlaps with MIPLVLS --> - <bitfield name="CHROMA_MIDPOINT_X" pos="16" type="boolean"/> - <bitfield name="CHROMA_MIDPOINT_Y" pos="18" type="boolean"/> - <bitfield name="SAMPLES" low="20" high="21" type="a3xx_msaa_samples"/> - <bitfield name="FMT" low="22" high="29" type="a6xx_format"/> - <!-- - Why is the swap needed in addition to SWIZ_*? The swap - is performed before border color replacement, while the - swizzle is applied after after it. - --> - <bitfield name="SWAP" low="30" high="31" type="a3xx_color_swap"/> - </reg32> - <reg32 offset="1" name="1"> - <bitfield name="WIDTH" low="0" high="14" type="uint"/> - <bitfield name="HEIGHT" low="15" high="29" type="uint"/> - <bitfield name="MUTABLEEN" pos="31" type="boolean" variants="A7XX-"/> - </reg32> - <reg32 offset="2" name="2"> - <!-- - These fields overlap PITCH, and are used instead of - PITCH/PITCHALIGN when TYPE is A6XX_TEX_BUFFER. - --> - <doc> probably for D3D structured UAVs, normally set to 1 </doc> - <bitfield name="STRUCTSIZETEXELS" low="4" high="15" type="uint"/> - <bitfield name="STARTOFFSETTEXELS" low="16" high="21" type="uint"/> - - <!-- minimum pitch (for mipmap levels): log2(pitchalign / 64) --> - <bitfield name="PITCHALIGN" low="0" high="3" type="uint"/> - <doc>Pitch in bytes (so actually stride)</doc> - <bitfield name="PITCH" low="7" high="28" type="uint"/> - <bitfield name="TYPE" low="29" high="31" type="a6xx_tex_type"/> - </reg32> - <reg32 offset="3" name="3"> - <!-- - ARRAY_PITCH is basically LAYERSZ for the first mipmap level, and - for 3d textures (laid out mipmap level first) MIN_LAYERSZ is the - layer size at the point that it stops being reduced moving to - higher (smaller) mipmap levels - --> - <bitfield name="ARRAY_PITCH" low="0" high="22" shr="12" type="uint"/> - <bitfield name="MIN_LAYERSZ" low="23" high="26" shr="12"/> - <!-- - by default levels with w < 16 are linear - TILE_ALL makes all levels have tiling - seems required when using UBWC, since all levels have UBWC (can possibly be disabled?) - --> - <bitfield name="TILE_ALL" pos="27" type="boolean"/> - <bitfield name="FLAG" pos="28" type="boolean"/> - </reg32> - <!-- for 2-3 plane format, BASE is flag buffer address (if enabled) - the address of the non-flag base buffer is determined automatically, - and must follow the flag buffer - --> - <reg32 offset="4" name="4"> - <bitfield name="BASE_LO" low="5" high="31" shr="5"/> - </reg32> - <reg32 offset="5" name="5"> - <bitfield name="BASE_HI" low="0" high="16"/> - <bitfield name="DEPTH" low="17" high="29" type="uint"/> - </reg32> - <reg32 offset="6" name="6"> - <!-- overlaps with PLANE_PITCH --> - <bitfield name="MIN_LOD_CLAMP" low="0" high="11" type="ufixed" radix="8"/> - <!-- pitch for plane 2 / plane 3 --> - <bitfield name="PLANE_PITCH" low="8" high="31" type="uint"/> - </reg32> - <!-- 7/8 is plane 2 address for planar formats --> - <reg32 offset="7" name="7"> - <bitfield name="FLAG_LO" low="5" high="31" shr="5"/> - </reg32> - <reg32 offset="8" name="8"> - <bitfield name="FLAG_HI" low="0" high="16"/> - </reg32> - <!-- 9/10 is plane 3 address for planar formats --> - <reg32 offset="9" name="9"> - <bitfield name="FLAG_BUFFER_ARRAY_PITCH" low="0" high="16" shr="4" type="uint"/> - </reg32> - <reg32 offset="10" name="10"> - <bitfield name="FLAG_BUFFER_PITCH" low="0" high="6" shr="6" type="uint"/> - <!-- log2 size of the first level, required for mipmapping --> - <bitfield name="FLAG_BUFFER_LOGW" low="8" high="11" type="uint"/> - <bitfield name="FLAG_BUFFER_LOGH" low="12" high="15" type="uint"/> - </reg32> - <reg32 offset="11" name="11"/> - <reg32 offset="12" name="12"/> - <reg32 offset="13" name="13"/> - <reg32 offset="14" name="14"/> - <reg32 offset="15" name="15"/> -</domain> - -<domain name="A6XX_UBO" width="32"> - <reg32 offset="0" name="0"> - <bitfield name="BASE_LO" low="0" high="31"/> - </reg32> - <reg32 offset="1" name="1"> - <bitfield name="BASE_HI" low="0" high="16"/> - <bitfield name="SIZE" low="17" high="31"/> <!-- size in vec4 (4xDWORD) units --> - </reg32> -</domain> - <domain name="A6XX_PDC" width="32"> <reg32 offset="0x1140" name="GPU_ENABLE_PDC"/> <reg32 offset="0x1148" name="GPU_SEQ_START_ADDR"/> diff --git a/drivers/gpu/drm/msm/registers/adreno/a6xx_descriptors.xml b/drivers/gpu/drm/msm/registers/adreno/a6xx_descriptors.xml new file mode 100644 index 000000000000..307d43dda8a2 --- /dev/null +++ b/drivers/gpu/drm/msm/registers/adreno/a6xx_descriptors.xml @@ -0,0 +1,198 @@ +<?xml version="1.0" encoding="UTF-8"?> +<database xmlns="http://nouveau.freedesktop.org/" +xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" +xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> +<import file="freedreno_copyright.xml"/> +<import file="adreno/adreno_common.xml"/> +<import file="adreno/adreno_pm4.xml"/> +<import file="adreno/a6xx_enums.xml"/> + +<domain name="A6XX_TEX_SAMP" width="32"> + <doc>Texture sampler dwords</doc> + <enum name="a6xx_tex_filter"> <!-- same as a4xx? --> + <value name="A6XX_TEX_NEAREST" value="0"/> + <value name="A6XX_TEX_LINEAR" value="1"/> + <value name="A6XX_TEX_ANISO" value="2"/> + <value name="A6XX_TEX_CUBIC" value="3"/> <!-- a650 only --> + </enum> + <enum name="a6xx_tex_clamp"> <!-- same as a4xx? --> + <value name="A6XX_TEX_REPEAT" value="0"/> + <value name="A6XX_TEX_CLAMP_TO_EDGE" value="1"/> + <value name="A6XX_TEX_MIRROR_REPEAT" value="2"/> + <value name="A6XX_TEX_CLAMP_TO_BORDER" value="3"/> + <value name="A6XX_TEX_MIRROR_CLAMP" value="4"/> + </enum> + <enum name="a6xx_tex_aniso"> <!-- same as a4xx? --> + <value name="A6XX_TEX_ANISO_1" value="0"/> + <value name="A6XX_TEX_ANISO_2" value="1"/> + <value name="A6XX_TEX_ANISO_4" value="2"/> + <value name="A6XX_TEX_ANISO_8" value="3"/> + <value name="A6XX_TEX_ANISO_16" value="4"/> + </enum> + <enum name="a6xx_reduction_mode"> + <value name="A6XX_REDUCTION_MODE_AVERAGE" value="0"/> + <value name="A6XX_REDUCTION_MODE_MIN" value="1"/> + <value name="A6XX_REDUCTION_MODE_MAX" value="2"/> + </enum> + <enum name="a6xx_fast_border_color"> + <!-- R B G A --> + <value name="A6XX_BORDER_COLOR_0_0_0_0" value="0"/> + <value name="A6XX_BORDER_COLOR_0_0_0_1" value="1"/> + <value name="A6XX_BORDER_COLOR_1_1_1_0" value="2"/> + <value name="A6XX_BORDER_COLOR_1_1_1_1" value="3"/> + </enum> + + <reg32 offset="0" name="0"> + <bitfield name="MIPFILTER_LINEAR_NEAR" pos="0" type="boolean"/> + <bitfield name="XY_MAG" low="1" high="2" type="a6xx_tex_filter"/> + <bitfield name="XY_MIN" low="3" high="4" type="a6xx_tex_filter"/> + <bitfield name="WRAP_S" low="5" high="7" type="a6xx_tex_clamp"/> + <bitfield name="WRAP_T" low="8" high="10" type="a6xx_tex_clamp"/> + <bitfield name="WRAP_R" low="11" high="13" type="a6xx_tex_clamp"/> + <bitfield name="ANISO" low="14" high="16" type="a6xx_tex_aniso"/> + <bitfield name="LOD_BIAS" low="19" high="31" type="fixed" radix="8"/><!-- no idea how many bits for real --> + </reg32> + <reg32 offset="1" name="1"> + <bitfield name="CLAMPENABLE" pos="0" type="boolean"> + <doc> + clamp result to [0, 1] if the format is unorm or + [-1, 1] if the format is snorm, *after* + filtering. Has no effect for other formats. + </doc> + </bitfield> + <bitfield name="COMPARE_FUNC" low="1" high="3" type="adreno_compare_func"/> + <bitfield name="CUBEMAPSEAMLESSFILTOFF" pos="4" type="boolean"/> + <bitfield name="UNNORM_COORDS" pos="5" type="boolean"/> + <bitfield name="MIPFILTER_LINEAR_FAR" pos="6" type="boolean"/> + <bitfield name="MAX_LOD" low="8" high="19" type="ufixed" radix="8"/> + <bitfield name="MIN_LOD" low="20" high="31" type="ufixed" radix="8"/> + </reg32> + <reg32 offset="2" name="2"> + <bitfield name="REDUCTION_MODE" low="0" high="1" type="a6xx_reduction_mode"/> + <bitfield name="FASTBORDERCOLOR" low="2" high="3" type="a6xx_fast_border_color"/> + <bitfield name="FASTBORDERCOLOREN" pos="4" type="boolean"/> + <bitfield name="CHROMA_LINEAR" pos="5" type="boolean"/> + <bitfield name="BCOLOR" low="7" high="31"/> + </reg32> + <reg32 offset="3" name="3"/> +</domain> + +<domain name="A6XX_TEX_CONST" width="32" varset="chip"> + <doc>Texture constant dwords</doc> + <enum name="a6xx_tex_swiz"> <!-- same as a4xx? --> + <value name="A6XX_TEX_X" value="0"/> + <value name="A6XX_TEX_Y" value="1"/> + <value name="A6XX_TEX_Z" value="2"/> + <value name="A6XX_TEX_W" value="3"/> + <value name="A6XX_TEX_ZERO" value="4"/> + <value name="A6XX_TEX_ONE" value="5"/> + </enum> + <reg32 offset="0" name="0"> + <bitfield name="TILE_MODE" low="0" high="1" type="a6xx_tile_mode"/> + <bitfield name="SRGB" pos="2" type="boolean"/> + <bitfield name="SWIZ_X" low="4" high="6" type="a6xx_tex_swiz"/> + <bitfield name="SWIZ_Y" low="7" high="9" type="a6xx_tex_swiz"/> + <bitfield name="SWIZ_Z" low="10" high="12" type="a6xx_tex_swiz"/> + <bitfield name="SWIZ_W" low="13" high="15" type="a6xx_tex_swiz"/> + <bitfield name="MIPLVLS" low="16" high="19" type="uint"/> + <!-- overlaps with MIPLVLS --> + <bitfield name="CHROMA_MIDPOINT_X" pos="16" type="boolean"/> + <bitfield name="CHROMA_MIDPOINT_Y" pos="18" type="boolean"/> + <bitfield name="SAMPLES" low="20" high="21" type="a3xx_msaa_samples"/> + <bitfield name="FMT" low="22" high="29" type="a6xx_format"/> + <!-- + Why is the swap needed in addition to SWIZ_*? The swap + is performed before border color replacement, while the + swizzle is applied after after it. + --> + <bitfield name="SWAP" low="30" high="31" type="a3xx_color_swap"/> + </reg32> + <reg32 offset="1" name="1"> + <bitfield name="WIDTH" low="0" high="14" type="uint"/> + <bitfield name="HEIGHT" low="15" high="29" type="uint"/> + <bitfield name="MUTABLEEN" pos="31" type="boolean" variants="A7XX-"/> + </reg32> + <reg32 offset="2" name="2"> + <!-- + These fields overlap PITCH, and are used instead of + PITCH/PITCHALIGN when TYPE is A6XX_TEX_BUFFER. + --> + <doc> probably for D3D structured UAVs, normally set to 1 </doc> + <bitfield name="STRUCTSIZETEXELS" low="4" high="15" type="uint"/> + <bitfield name="STARTOFFSETTEXELS" low="16" high="21" type="uint"/> + + <!-- minimum pitch (for mipmap levels): log2(pitchalign / 64) --> + <bitfield name="PITCHALIGN" low="0" high="3" type="uint"/> + <doc>Pitch in bytes (so actually stride)</doc> + <bitfield name="PITCH" low="7" high="28" type="uint"/> + <bitfield name="TYPE" low="29" high="31" type="a6xx_tex_type"/> + </reg32> + <reg32 offset="3" name="3"> + <!-- + ARRAY_PITCH is basically LAYERSZ for the first mipmap level, and + for 3d textures (laid out mipmap level first) MIN_LAYERSZ is the + layer size at the point that it stops being reduced moving to + higher (smaller) mipmap levels + --> + <bitfield name="ARRAY_PITCH" low="0" high="22" shr="12" type="uint"/> + <bitfield name="MIN_LAYERSZ" low="23" high="26" shr="12"/> + <!-- + by default levels with w < 16 are linear + TILE_ALL makes all levels have tiling + seems required when using UBWC, since all levels have UBWC (can possibly be disabled?) + --> + <bitfield name="TILE_ALL" pos="27" type="boolean"/> + <bitfield name="FLAG" pos="28" type="boolean"/> + </reg32> + <!-- for 2-3 plane format, BASE is flag buffer address (if enabled) + the address of the non-flag base buffer is determined automatically, + and must follow the flag buffer + --> + <reg32 offset="4" name="4"> + <bitfield name="BASE_LO" low="5" high="31" shr="5"/> + </reg32> + <reg32 offset="5" name="5"> + <bitfield name="BASE_HI" low="0" high="16"/> + <bitfield name="DEPTH" low="17" high="29" type="uint"/> + </reg32> + <reg32 offset="6" name="6"> + <!-- overlaps with PLANE_PITCH --> + <bitfield name="MIN_LOD_CLAMP" low="0" high="11" type="ufixed" radix="8"/> + <!-- pitch for plane 2 / plane 3 --> + <bitfield name="PLANE_PITCH" low="8" high="31" type="uint"/> + </reg32> + <!-- 7/8 is plane 2 address for planar formats --> + <reg32 offset="7" name="7"> + <bitfield name="FLAG_LO" low="5" high="31" shr="5"/> + </reg32> + <reg32 offset="8" name="8"> + <bitfield name="FLAG_HI" low="0" high="16"/> + </reg32> + <!-- 9/10 is plane 3 address for planar formats --> + <reg32 offset="9" name="9"> + <bitfield name="FLAG_BUFFER_ARRAY_PITCH" low="0" high="16" shr="4" type="uint"/> + </reg32> + <reg32 offset="10" name="10"> + <bitfield name="FLAG_BUFFER_PITCH" low="0" high="6" shr="6" type="uint"/> + <!-- log2 size of the first level, required for mipmapping --> + <bitfield name="FLAG_BUFFER_LOGW" low="8" high="11" type="uint"/> + <bitfield name="FLAG_BUFFER_LOGH" low="12" high="15" type="uint"/> + </reg32> + <reg32 offset="11" name="11"/> + <reg32 offset="12" name="12"/> + <reg32 offset="13" name="13"/> + <reg32 offset="14" name="14"/> + <reg32 offset="15" name="15"/> +</domain> + +<domain name="A6XX_UBO" width="32"> + <reg32 offset="0" name="0"> + <bitfield name="BASE_LO" low="0" high="31"/> + </reg32> + <reg32 offset="1" name="1"> + <bitfield name="BASE_HI" low="0" high="16"/> + <bitfield name="SIZE" low="17" high="31"/> <!-- size in vec4 (4xDWORD) units --> + </reg32> +</domain> + +</database> diff --git a/drivers/gpu/drm/msm/registers/adreno/a6xx_enums.xml b/drivers/gpu/drm/msm/registers/adreno/a6xx_enums.xml new file mode 100644 index 000000000000..665539b098c6 --- /dev/null +++ b/drivers/gpu/drm/msm/registers/adreno/a6xx_enums.xml @@ -0,0 +1,383 @@ +<?xml version="1.0" encoding="UTF-8"?> +<database xmlns="http://nouveau.freedesktop.org/" +xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" +xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> +<import file="freedreno_copyright.xml"/> +<import file="adreno/adreno_common.xml"/> +<import file="adreno/adreno_pm4.xml"/> + +<enum name="a6xx_tile_mode"> + <value name="TILE6_LINEAR" value="0"/> + <value name="TILE6_2" value="2"/> + <value name="TILE6_3" value="3"/> +</enum> + +<enum name="a6xx_format"> + <value value="0x02" name="FMT6_A8_UNORM"/> + <value value="0x03" name="FMT6_8_UNORM"/> + <value value="0x04" name="FMT6_8_SNORM"/> + <value value="0x05" name="FMT6_8_UINT"/> + <value value="0x06" name="FMT6_8_SINT"/> + + <value value="0x08" name="FMT6_4_4_4_4_UNORM"/> + <value value="0x0a" name="FMT6_5_5_5_1_UNORM"/> + <value value="0x0c" name="FMT6_1_5_5_5_UNORM"/> <!-- read only --> + <value value="0x0e" name="FMT6_5_6_5_UNORM"/> + + <value value="0x0f" name="FMT6_8_8_UNORM"/> + <value value="0x10" name="FMT6_8_8_SNORM"/> + <value value="0x11" name="FMT6_8_8_UINT"/> + <value value="0x12" name="FMT6_8_8_SINT"/> + <value value="0x13" name="FMT6_L8_A8_UNORM"/> + + <value value="0x15" name="FMT6_16_UNORM"/> + <value value="0x16" name="FMT6_16_SNORM"/> + <value value="0x17" name="FMT6_16_FLOAT"/> + <value value="0x18" name="FMT6_16_UINT"/> + <value value="0x19" name="FMT6_16_SINT"/> + + <value value="0x21" name="FMT6_8_8_8_UNORM"/> + <value value="0x22" name="FMT6_8_8_8_SNORM"/> + <value value="0x23" name="FMT6_8_8_8_UINT"/> + <value value="0x24" name="FMT6_8_8_8_SINT"/> + + <value value="0x30" name="FMT6_8_8_8_8_UNORM"/> + <value value="0x31" name="FMT6_8_8_8_X8_UNORM"/> <!-- samples 1 for alpha --> + <value value="0x32" name="FMT6_8_8_8_8_SNORM"/> + <value value="0x33" name="FMT6_8_8_8_8_UINT"/> + <value value="0x34" name="FMT6_8_8_8_8_SINT"/> + + <value value="0x35" name="FMT6_9_9_9_E5_FLOAT"/> + + <value value="0x36" name="FMT6_10_10_10_2_UNORM"/> + <value value="0x37" name="FMT6_10_10_10_2_UNORM_DEST"/> + <value value="0x39" name="FMT6_10_10_10_2_SNORM"/> + <value value="0x3a" name="FMT6_10_10_10_2_UINT"/> + <value value="0x3b" name="FMT6_10_10_10_2_SINT"/> + + <value value="0x42" name="FMT6_11_11_10_FLOAT"/> + + <value value="0x43" name="FMT6_16_16_UNORM"/> + <value value="0x44" name="FMT6_16_16_SNORM"/> + <value value="0x45" name="FMT6_16_16_FLOAT"/> + <value value="0x46" name="FMT6_16_16_UINT"/> + <value value="0x47" name="FMT6_16_16_SINT"/> + + <value value="0x48" name="FMT6_32_UNORM"/> + <value value="0x49" name="FMT6_32_SNORM"/> + <value value="0x4a" name="FMT6_32_FLOAT"/> + <value value="0x4b" name="FMT6_32_UINT"/> + <value value="0x4c" name="FMT6_32_SINT"/> + <value value="0x4d" name="FMT6_32_FIXED"/> + + <value value="0x58" name="FMT6_16_16_16_UNORM"/> + <value value="0x59" name="FMT6_16_16_16_SNORM"/> + <value value="0x5a" name="FMT6_16_16_16_FLOAT"/> + <value value="0x5b" name="FMT6_16_16_16_UINT"/> + <value value="0x5c" name="FMT6_16_16_16_SINT"/> + + <value value="0x60" name="FMT6_16_16_16_16_UNORM"/> + <value value="0x61" name="FMT6_16_16_16_16_SNORM"/> + <value value="0x62" name="FMT6_16_16_16_16_FLOAT"/> + <value value="0x63" name="FMT6_16_16_16_16_UINT"/> + <value value="0x64" name="FMT6_16_16_16_16_SINT"/> + + <value value="0x65" name="FMT6_32_32_UNORM"/> + <value value="0x66" name="FMT6_32_32_SNORM"/> + <value value="0x67" name="FMT6_32_32_FLOAT"/> + <value value="0x68" name="FMT6_32_32_UINT"/> + <value value="0x69" name="FMT6_32_32_SINT"/> + <value value="0x6a" name="FMT6_32_32_FIXED"/> + + <value value="0x70" name="FMT6_32_32_32_UNORM"/> + <value value="0x71" name="FMT6_32_32_32_SNORM"/> + <value value="0x72" name="FMT6_32_32_32_UINT"/> + <value value="0x73" name="FMT6_32_32_32_SINT"/> + <value value="0x74" name="FMT6_32_32_32_FLOAT"/> + <value value="0x75" name="FMT6_32_32_32_FIXED"/> + + <value value="0x80" name="FMT6_32_32_32_32_UNORM"/> + <value value="0x81" name="FMT6_32_32_32_32_SNORM"/> + <value value="0x82" name="FMT6_32_32_32_32_FLOAT"/> + <value value="0x83" name="FMT6_32_32_32_32_UINT"/> + <value value="0x84" name="FMT6_32_32_32_32_SINT"/> + <value value="0x85" name="FMT6_32_32_32_32_FIXED"/> + + <value value="0x8c" name="FMT6_G8R8B8R8_422_UNORM"/> <!-- UYVY --> + <value value="0x8d" name="FMT6_R8G8R8B8_422_UNORM"/> <!-- YUYV --> + <value value="0x8e" name="FMT6_R8_G8B8_2PLANE_420_UNORM"/> <!-- NV12 --> + <value value="0x8f" name="FMT6_NV21"/> + <value value="0x90" name="FMT6_R8_G8_B8_3PLANE_420_UNORM"/> <!-- YV12 --> + + <value value="0x91" name="FMT6_Z24_UNORM_S8_UINT_AS_R8G8B8A8"/> + + <!-- Note: tiling/UBWC for these may be different from equivalent formats + For example FMT6_NV12_Y is not compatible with FMT6_8_UNORM + --> + <value value="0x94" name="FMT6_NV12_Y"/> + <value value="0x95" name="FMT6_NV12_UV"/> + <value value="0x96" name="FMT6_NV12_VU"/> + <value value="0x97" name="FMT6_NV12_4R"/> + <value value="0x98" name="FMT6_NV12_4R_Y"/> + <value value="0x99" name="FMT6_NV12_4R_UV"/> + <value value="0x9a" name="FMT6_P010"/> + <value value="0x9b" name="FMT6_P010_Y"/> + <value value="0x9c" name="FMT6_P010_UV"/> + <value value="0x9d" name="FMT6_TP10"/> + <value value="0x9e" name="FMT6_TP10_Y"/> + <value value="0x9f" name="FMT6_TP10_UV"/> + + <value value="0xa0" name="FMT6_Z24_UNORM_S8_UINT"/> + + <value value="0xab" name="FMT6_ETC2_RG11_UNORM"/> + <value value="0xac" name="FMT6_ETC2_RG11_SNORM"/> + <value value="0xad" name="FMT6_ETC2_R11_UNORM"/> + <value value="0xae" name="FMT6_ETC2_R11_SNORM"/> + <value value="0xaf" name="FMT6_ETC1"/> + <value value="0xb0" name="FMT6_ETC2_RGB8"/> + <value value="0xb1" name="FMT6_ETC2_RGBA8"/> + <value value="0xb2" name="FMT6_ETC2_RGB8A1"/> + <value value="0xb3" name="FMT6_DXT1"/> + <value value="0xb4" name="FMT6_DXT3"/> + <value value="0xb5" name="FMT6_DXT5"/> + <value value="0xb6" name="FMT6_RGTC1_UNORM"/> + <value value="0xb7" name="FMT6_RGTC1_UNORM_FAST"/> + <value value="0xb8" name="FMT6_RGTC1_SNORM"/> + <value value="0xb9" name="FMT6_RGTC1_SNORM_FAST"/> + <value value="0xba" name="FMT6_RGTC2_UNORM"/> + <value value="0xbb" name="FMT6_RGTC2_UNORM_FAST"/> + <value value="0xbc" name="FMT6_RGTC2_SNORM"/> + <value value="0xbd" name="FMT6_RGTC2_SNORM_FAST"/> + <value value="0xbe" name="FMT6_BPTC_UFLOAT"/> + <value value="0xbf" name="FMT6_BPTC_FLOAT"/> + <value value="0xc0" name="FMT6_BPTC"/> + <value value="0xc1" name="FMT6_ASTC_4x4"/> + <value value="0xc2" name="FMT6_ASTC_5x4"/> + <value value="0xc3" name="FMT6_ASTC_5x5"/> + <value value="0xc4" name="FMT6_ASTC_6x5"/> + <value value="0xc5" name="FMT6_ASTC_6x6"/> + <value value="0xc6" name="FMT6_ASTC_8x5"/> + <value value="0xc7" name="FMT6_ASTC_8x6"/> + <value value="0xc8" name="FMT6_ASTC_8x8"/> + <value value="0xc9" name="FMT6_ASTC_10x5"/> + <value value="0xca" name="FMT6_ASTC_10x6"/> + <value value="0xcb" name="FMT6_ASTC_10x8"/> + <value value="0xcc" name="FMT6_ASTC_10x10"/> + <value value="0xcd" name="FMT6_ASTC_12x10"/> + <value value="0xce" name="FMT6_ASTC_12x12"/> + + <!-- for sampling stencil (integer, 2nd channel), not available on a630 --> + <value value="0xea" name="FMT6_Z24_UINT_S8_UINT"/> + + <!-- Not a hw enum, used internally in driver --> + <value value="0xff" name="FMT6_NONE"/> + +</enum> + +<!-- probably same as a5xx --> +<enum name="a6xx_polygon_mode"> + <value name="POLYMODE6_POINTS" value="1"/> + <value name="POLYMODE6_LINES" value="2"/> + <value name="POLYMODE6_TRIANGLES" value="3"/> +</enum> + +<enum name="a6xx_depth_format"> + <value name="DEPTH6_NONE" value="0"/> + <value name="DEPTH6_16" value="1"/> + <value name="DEPTH6_24_8" value="2"/> + <value name="DEPTH6_32" value="4"/> +</enum> + +<bitset name="a6x_cp_protect" inline="yes"> + <bitfield name="BASE_ADDR" low="0" high="17"/> + <bitfield name="MASK_LEN" low="18" high="30"/> + <bitfield name="READ" pos="31" type="boolean"/> +</bitset> + +<enum name="a6xx_shader_id"> + <value value="0x9" name="A6XX_TP0_TMO_DATA"/> + <value value="0xa" name="A6XX_TP0_SMO_DATA"/> + <value value="0xb" name="A6XX_TP0_MIPMAP_BASE_DATA"/> + <value value="0x19" name="A6XX_TP1_TMO_DATA"/> + <value value="0x1a" name="A6XX_TP1_SMO_DATA"/> + <value value="0x1b" name="A6XX_TP1_MIPMAP_BASE_DATA"/> + <value value="0x29" name="A6XX_SP_INST_DATA"/> + <value value="0x2a" name="A6XX_SP_LB_0_DATA"/> + <value value="0x2b" name="A6XX_SP_LB_1_DATA"/> + <value value="0x2c" name="A6XX_SP_LB_2_DATA"/> + <value value="0x2d" name="A6XX_SP_LB_3_DATA"/> + <value value="0x2e" name="A6XX_SP_LB_4_DATA"/> + <value value="0x2f" name="A6XX_SP_LB_5_DATA"/> + <value value="0x30" name="A6XX_SP_CB_BINDLESS_DATA"/> + <value value="0x31" name="A6XX_SP_CB_LEGACY_DATA"/> + <value value="0x32" name="A6XX_SP_GFX_UAV_BASE_DATA"/> + <value value="0x33" name="A6XX_SP_INST_TAG"/> + <value value="0x34" name="A6XX_SP_CB_BINDLESS_TAG"/> + <value value="0x35" name="A6XX_SP_TMO_UMO_TAG"/> + <value value="0x36" name="A6XX_SP_SMO_TAG"/> + <value value="0x37" name="A6XX_SP_STATE_DATA"/> + <value value="0x49" name="A6XX_HLSQ_CHUNK_CVS_RAM"/> + <value value="0x4a" name="A6XX_HLSQ_CHUNK_CPS_RAM"/> + <value value="0x4b" name="A6XX_HLSQ_CHUNK_CVS_RAM_TAG"/> + <value value="0x4c" name="A6XX_HLSQ_CHUNK_CPS_RAM_TAG"/> + <value value="0x4d" name="A6XX_HLSQ_ICB_CVS_CB_BASE_TAG"/> + <value value="0x4e" name="A6XX_HLSQ_ICB_CPS_CB_BASE_TAG"/> + <value value="0x50" name="A6XX_HLSQ_CVS_MISC_RAM"/> + <value value="0x51" name="A6XX_HLSQ_CPS_MISC_RAM"/> + <value value="0x52" name="A6XX_HLSQ_INST_RAM"/> + <value value="0x53" name="A6XX_HLSQ_GFX_CVS_CONST_RAM"/> + <value value="0x54" name="A6XX_HLSQ_GFX_CPS_CONST_RAM"/> + <value value="0x55" name="A6XX_HLSQ_CVS_MISC_RAM_TAG"/> + <value value="0x56" name="A6XX_HLSQ_CPS_MISC_RAM_TAG"/> + <value value="0x57" name="A6XX_HLSQ_INST_RAM_TAG"/> + <value value="0x58" name="A6XX_HLSQ_GFX_CVS_CONST_RAM_TAG"/> + <value value="0x59" name="A6XX_HLSQ_GFX_CPS_CONST_RAM_TAG"/> + <value value="0x5a" name="A6XX_HLSQ_PWR_REST_RAM"/> + <value value="0x5b" name="A6XX_HLSQ_PWR_REST_TAG"/> + <value value="0x60" name="A6XX_HLSQ_DATAPATH_META"/> + <value value="0x61" name="A6XX_HLSQ_FRONTEND_META"/> + <value value="0x62" name="A6XX_HLSQ_INDIRECT_META"/> + <value value="0x63" name="A6XX_HLSQ_BACKEND_META"/> + <value value="0x70" name="A6XX_SP_LB_6_DATA"/> + <value value="0x71" name="A6XX_SP_LB_7_DATA"/> + <value value="0x73" name="A6XX_HLSQ_INST_RAM_1"/> +</enum> + +<enum name="a6xx_debugbus_id"> + <value value="0x1" name="A6XX_DBGBUS_CP"/> + <value value="0x2" name="A6XX_DBGBUS_RBBM"/> + <value value="0x3" name="A6XX_DBGBUS_VBIF"/> + <value value="0x4" name="A6XX_DBGBUS_HLSQ"/> + <value value="0x5" name="A6XX_DBGBUS_UCHE"/> + <value value="0x6" name="A6XX_DBGBUS_DPM"/> + <value value="0x7" name="A6XX_DBGBUS_TESS"/> + <value value="0x8" name="A6XX_DBGBUS_PC"/> + <value value="0x9" name="A6XX_DBGBUS_VFDP"/> + <value value="0xa" name="A6XX_DBGBUS_VPC"/> + <value value="0xb" name="A6XX_DBGBUS_TSE"/> + <value value="0xc" name="A6XX_DBGBUS_RAS"/> + <value value="0xd" name="A6XX_DBGBUS_VSC"/> + <value value="0xe" name="A6XX_DBGBUS_COM"/> + <value value="0x10" name="A6XX_DBGBUS_LRZ"/> + <value value="0x11" name="A6XX_DBGBUS_A2D"/> + <value value="0x12" name="A6XX_DBGBUS_CCUFCHE"/> + <value value="0x13" name="A6XX_DBGBUS_GMU_CX"/> + <value value="0x14" name="A6XX_DBGBUS_RBP"/> + <value value="0x15" name="A6XX_DBGBUS_DCS"/> + <value value="0x16" name="A6XX_DBGBUS_DBGC"/> + <value value="0x17" name="A6XX_DBGBUS_CX"/> + <value value="0x18" name="A6XX_DBGBUS_GMU_GX"/> + <value value="0x19" name="A6XX_DBGBUS_TPFCHE"/> + <value value="0x1a" name="A6XX_DBGBUS_GBIF_GX"/> + <value value="0x1d" name="A6XX_DBGBUS_GPC"/> + <value value="0x1e" name="A6XX_DBGBUS_LARC"/> + <value value="0x1f" name="A6XX_DBGBUS_HLSQ_SPTP"/> + <value value="0x20" name="A6XX_DBGBUS_RB_0"/> + <value value="0x21" name="A6XX_DBGBUS_RB_1"/> + <value value="0x22" name="A6XX_DBGBUS_RB_2"/> + <value value="0x24" name="A6XX_DBGBUS_UCHE_WRAPPER"/> + <value value="0x28" name="A6XX_DBGBUS_CCU_0"/> + <value value="0x29" name="A6XX_DBGBUS_CCU_1"/> + <value value="0x2a" name="A6XX_DBGBUS_CCU_2"/> + <value value="0x38" name="A6XX_DBGBUS_VFD_0"/> + <value value="0x39" name="A6XX_DBGBUS_VFD_1"/> + <value value="0x3a" name="A6XX_DBGBUS_VFD_2"/> + <value value="0x3b" name="A6XX_DBGBUS_VFD_3"/> + <value value="0x3c" name="A6XX_DBGBUS_VFD_4"/> + <value value="0x3d" name="A6XX_DBGBUS_VFD_5"/> + <value value="0x40" name="A6XX_DBGBUS_SP_0"/> + <value value="0x41" name="A6XX_DBGBUS_SP_1"/> + <value value="0x42" name="A6XX_DBGBUS_SP_2"/> + <value value="0x48" name="A6XX_DBGBUS_TPL1_0"/> + <value value="0x49" name="A6XX_DBGBUS_TPL1_1"/> + <value value="0x4a" name="A6XX_DBGBUS_TPL1_2"/> + <value value="0x4b" name="A6XX_DBGBUS_TPL1_3"/> + <value value="0x4c" name="A6XX_DBGBUS_TPL1_4"/> + <value value="0x4d" name="A6XX_DBGBUS_TPL1_5"/> + <value value="0x58" name="A6XX_DBGBUS_SPTP_0"/> + <value value="0x59" name="A6XX_DBGBUS_SPTP_1"/> + <value value="0x5a" name="A6XX_DBGBUS_SPTP_2"/> + <value value="0x5b" name="A6XX_DBGBUS_SPTP_3"/> + <value value="0x5c" name="A6XX_DBGBUS_SPTP_4"/> + <value value="0x5d" name="A6XX_DBGBUS_SPTP_5"/> +</enum> + +<!-- +Used in a6xx_a2d_bit_cntl.. the value mostly seems to correlate to the +component type/size, so I think it relates to internal format used for +blending? The one exception is that 16b unorm and 32b float use the +same value... maybe 16b unorm is uncommon enough that it was just easier +to upconvert to 32b float internally? + + 8b unorm: 10 (sometimes 0, is the high bit part of something else?) +16b unorm: 4 + +32b int: 7 +16b int: 6 + 8b int: 5 + +32b float: 4 +16b float: 3 + --> +<enum name="a6xx_2d_ifmt"> + <value value="0x10" name="R2D_UNORM8"/> + <value value="0x7" name="R2D_INT32"/> + <value value="0x6" name="R2D_INT16"/> + <value value="0x5" name="R2D_INT8"/> + <value value="0x4" name="R2D_FLOAT32"/> + <value value="0x3" name="R2D_FLOAT16"/> + <value value="0x1" name="R2D_UNORM8_SRGB"/> + <value value="0x0" name="R2D_RAW"/> +</enum> + +<enum name="a6xx_tex_type"> + <value name="A6XX_TEX_1D" value="0"/> + <value name="A6XX_TEX_2D" value="1"/> + <value name="A6XX_TEX_CUBE" value="2"/> + <value name="A6XX_TEX_3D" value="3"/> + <value name="A6XX_TEX_BUFFER" value="4"/> + <doc> + A special buffer type for usage as the source for buffer + to image copies with lower alignment requirements than + A6XX_TEX_2D, available since A7XX. + </doc> + <value name="A6XX_TEX_IMG_BUFFER" value="5"/> +</enum> + +<enum name="a6xx_ztest_mode"> + <doc>Allow early z-test and early-lrz (if applicable)</doc> + <value value="0x0" name="A6XX_EARLY_Z"/> + <doc>Disable early z-test and early-lrz test (if applicable)</doc> + <value value="0x1" name="A6XX_LATE_Z"/> + <doc> + A special mode that allows early-lrz (if applicable) or early-z + tests, but also does late-z tests at which point it writes depth. + + This mode is used when fragment can be killed (via discard or + sample mask) after early-z tests and it writes depth. In such case + depth can be written only at late-z stage, but it's ok to use + early-z to discard fragments. + + However this mode is not compatible with: + - Lack of D/S attachment + - Stencil writes on stencil or depth test failures + - Per-sample shading + </doc> + <value value="0x2" name="A6XX_EARLY_Z_LATE_Z"/> + <doc>Not a real hw value, used internally by mesa</doc> + <value value="0x3" name="A6XX_INVALID_ZTEST"/> +</enum> + +<enum name="a6xx_tess_spacing"> + <value value="0x0" name="TESS_EQUAL"/> + <value value="0x2" name="TESS_FRACTIONAL_ODD"/> + <value value="0x3" name="TESS_FRACTIONAL_EVEN"/> +</enum> +<enum name="a6xx_tess_output"> + <value value="0x0" name="TESS_POINTS"/> + <value value="0x1" name="TESS_LINES"/> + <value value="0x2" name="TESS_CW_TRIS"/> + <value value="0x3" name="TESS_CCW_TRIS"/> +</enum> + +</database> diff --git a/drivers/gpu/drm/msm/registers/adreno/a6xx_perfcntrs.xml b/drivers/gpu/drm/msm/registers/adreno/a6xx_perfcntrs.xml new file mode 100644 index 000000000000..c446a2eb1120 --- /dev/null +++ b/drivers/gpu/drm/msm/registers/adreno/a6xx_perfcntrs.xml @@ -0,0 +1,600 @@ +<?xml version="1.0" encoding="UTF-8"?> +<database xmlns="http://nouveau.freedesktop.org/" +xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" +xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> +<import file="freedreno_copyright.xml"/> +<import file="adreno/adreno_common.xml"/> +<import file="adreno/adreno_pm4.xml"/> + +<enum name="a6xx_cp_perfcounter_select"> + <value value="0" name="PERF_CP_ALWAYS_COUNT"/> + <value value="1" name="PERF_CP_BUSY_GFX_CORE_IDLE"/> + <value value="2" name="PERF_CP_BUSY_CYCLES"/> + <value value="3" name="PERF_CP_NUM_PREEMPTIONS"/> + <value value="4" name="PERF_CP_PREEMPTION_REACTION_DELAY"/> + <value value="5" name="PERF_CP_PREEMPTION_SWITCH_OUT_TIME"/> + <value value="6" name="PERF_CP_PREEMPTION_SWITCH_IN_TIME"/> + <value value="7" name="PERF_CP_DEAD_DRAWS_IN_BIN_RENDER"/> + <value value="8" name="PERF_CP_PREDICATED_DRAWS_KILLED"/> + <value value="9" name="PERF_CP_MODE_SWITCH"/> + <value value="10" name="PERF_CP_ZPASS_DONE"/> + <value value="11" name="PERF_CP_CONTEXT_DONE"/> + <value value="12" name="PERF_CP_CACHE_FLUSH"/> + <value value="13" name="PERF_CP_LONG_PREEMPTIONS"/> + <value value="14" name="PERF_CP_SQE_I_CACHE_STARVE"/> + <value value="15" name="PERF_CP_SQE_IDLE"/> + <value value="16" name="PERF_CP_SQE_PM4_STARVE_RB_IB"/> + <value value="17" name="PERF_CP_SQE_PM4_STARVE_SDS"/> + <value value="18" name="PERF_CP_SQE_MRB_STARVE"/> + <value value="19" name="PERF_CP_SQE_RRB_STARVE"/> + <value value="20" name="PERF_CP_SQE_VSD_STARVE"/> + <value value="21" name="PERF_CP_VSD_DECODE_STARVE"/> + <value value="22" name="PERF_CP_SQE_PIPE_OUT_STALL"/> + <value value="23" name="PERF_CP_SQE_SYNC_STALL"/> + <value value="24" name="PERF_CP_SQE_PM4_WFI_STALL"/> + <value value="25" name="PERF_CP_SQE_SYS_WFI_STALL"/> + <value value="26" name="PERF_CP_SQE_T4_EXEC"/> + <value value="27" name="PERF_CP_SQE_LOAD_STATE_EXEC"/> + <value value="28" name="PERF_CP_SQE_SAVE_SDS_STATE"/> + <value value="29" name="PERF_CP_SQE_DRAW_EXEC"/> + <value value="30" name="PERF_CP_SQE_CTXT_REG_BUNCH_EXEC"/> + <value value="31" name="PERF_CP_SQE_EXEC_PROFILED"/> + <value value="32" name="PERF_CP_MEMORY_POOL_EMPTY"/> + <value value="33" name="PERF_CP_MEMORY_POOL_SYNC_STALL"/> + <value value="34" name="PERF_CP_MEMORY_POOL_ABOVE_THRESH"/> + <value value="35" name="PERF_CP_AHB_WR_STALL_PRE_DRAWS"/> + <value value="36" name="PERF_CP_AHB_STALL_SQE_GMU"/> + <value value="37" name="PERF_CP_AHB_STALL_SQE_WR_OTHER"/> + <value value="38" name="PERF_CP_AHB_STALL_SQE_RD_OTHER"/> + <value value="39" name="PERF_CP_CLUSTER0_EMPTY"/> + <value value="40" name="PERF_CP_CLUSTER1_EMPTY"/> + <value value="41" name="PERF_CP_CLUSTER2_EMPTY"/> + <value value="42" name="PERF_CP_CLUSTER3_EMPTY"/> + <value value="43" name="PERF_CP_CLUSTER4_EMPTY"/> + <value value="44" name="PERF_CP_CLUSTER5_EMPTY"/> + <value value="45" name="PERF_CP_PM4_DATA"/> + <value value="46" name="PERF_CP_PM4_HEADERS"/> + <value value="47" name="PERF_CP_VBIF_READ_BEATS"/> + <value value="48" name="PERF_CP_VBIF_WRITE_BEATS"/> + <value value="49" name="PERF_CP_SQE_INSTR_COUNTER"/> +</enum> + +<enum name="a6xx_rbbm_perfcounter_select"> + <value value="0" name="PERF_RBBM_ALWAYS_COUNT"/> + <value value="1" name="PERF_RBBM_ALWAYS_ON"/> + <value value="2" name="PERF_RBBM_TSE_BUSY"/> + <value value="3" name="PERF_RBBM_RAS_BUSY"/> + <value value="4" name="PERF_RBBM_PC_DCALL_BUSY"/> + <value value="5" name="PERF_RBBM_PC_VSD_BUSY"/> + <value value="6" name="PERF_RBBM_STATUS_MASKED"/> + <value value="7" name="PERF_RBBM_COM_BUSY"/> + <value value="8" name="PERF_RBBM_DCOM_BUSY"/> + <value value="9" name="PERF_RBBM_VBIF_BUSY"/> + <value value="10" name="PERF_RBBM_VSC_BUSY"/> + <value value="11" name="PERF_RBBM_TESS_BUSY"/> + <value value="12" name="PERF_RBBM_UCHE_BUSY"/> + <value value="13" name="PERF_RBBM_HLSQ_BUSY"/> +</enum> + +<enum name="a6xx_pc_perfcounter_select"> + <value value="0" name="PERF_PC_BUSY_CYCLES"/> + <value value="1" name="PERF_PC_WORKING_CYCLES"/> + <value value="2" name="PERF_PC_STALL_CYCLES_VFD"/> + <value value="3" name="PERF_PC_STALL_CYCLES_TSE"/> + <value value="4" name="PERF_PC_STALL_CYCLES_VPC"/> + <value value="5" name="PERF_PC_STALL_CYCLES_UCHE"/> + <value value="6" name="PERF_PC_STALL_CYCLES_TESS"/> + <value value="7" name="PERF_PC_STALL_CYCLES_TSE_ONLY"/> + <value value="8" name="PERF_PC_STALL_CYCLES_VPC_ONLY"/> + <value value="9" name="PERF_PC_PASS1_TF_STALL_CYCLES"/> + <value value="10" name="PERF_PC_STARVE_CYCLES_FOR_INDEX"/> + <value value="11" name="PERF_PC_STARVE_CYCLES_FOR_TESS_FACTOR"/> + <value value="12" name="PERF_PC_STARVE_CYCLES_FOR_VIZ_STREAM"/> + <value value="13" name="PERF_PC_STARVE_CYCLES_FOR_POSITION"/> + <value value="14" name="PERF_PC_STARVE_CYCLES_DI"/> + <value value="15" name="PERF_PC_VIS_STREAMS_LOADED"/> + <value value="16" name="PERF_PC_INSTANCES"/> + <value value="17" name="PERF_PC_VPC_PRIMITIVES"/> + <value value="18" name="PERF_PC_DEAD_PRIM"/> + <value value="19" name="PERF_PC_LIVE_PRIM"/> + <value value="20" name="PERF_PC_VERTEX_HITS"/> + <value value="21" name="PERF_PC_IA_VERTICES"/> + <value value="22" name="PERF_PC_IA_PRIMITIVES"/> + <value value="23" name="PERF_PC_GS_PRIMITIVES"/> + <value value="24" name="PERF_PC_HS_INVOCATIONS"/> + <value value="25" name="PERF_PC_DS_INVOCATIONS"/> + <value value="26" name="PERF_PC_VS_INVOCATIONS"/> + <value value="27" name="PERF_PC_GS_INVOCATIONS"/> + <value value="28" name="PERF_PC_DS_PRIMITIVES"/> + <value value="29" name="PERF_PC_VPC_POS_DATA_TRANSACTION"/> + <value value="30" name="PERF_PC_3D_DRAWCALLS"/> + <value value="31" name="PERF_PC_2D_DRAWCALLS"/> + <value value="32" name="PERF_PC_NON_DRAWCALL_GLOBAL_EVENTS"/> + <value value="33" name="PERF_TESS_BUSY_CYCLES"/> + <value value="34" name="PERF_TESS_WORKING_CYCLES"/> + <value value="35" name="PERF_TESS_STALL_CYCLES_PC"/> + <value value="36" name="PERF_TESS_STARVE_CYCLES_PC"/> + <value value="37" name="PERF_PC_TSE_TRANSACTION"/> + <value value="38" name="PERF_PC_TSE_VERTEX"/> + <value value="39" name="PERF_PC_TESS_PC_UV_TRANS"/> + <value value="40" name="PERF_PC_TESS_PC_UV_PATCHES"/> + <value value="41" name="PERF_PC_TESS_FACTOR_TRANS"/> +</enum> + +<enum name="a6xx_vfd_perfcounter_select"> + <value value="0" name="PERF_VFD_BUSY_CYCLES"/> + <value value="1" name="PERF_VFD_STALL_CYCLES_UCHE"/> + <value value="2" name="PERF_VFD_STALL_CYCLES_VPC_ALLOC"/> + <value value="3" name="PERF_VFD_STALL_CYCLES_SP_INFO"/> + <value value="4" name="PERF_VFD_STALL_CYCLES_SP_ATTR"/> + <value value="5" name="PERF_VFD_STARVE_CYCLES_UCHE"/> + <value value="6" name="PERF_VFD_RBUFFER_FULL"/> + <value value="7" name="PERF_VFD_ATTR_INFO_FIFO_FULL"/> + <value value="8" name="PERF_VFD_DECODED_ATTRIBUTE_BYTES"/> + <value value="9" name="PERF_VFD_NUM_ATTRIBUTES"/> + <value value="10" name="PERF_VFD_UPPER_SHADER_FIBERS"/> + <value value="11" name="PERF_VFD_LOWER_SHADER_FIBERS"/> + <value value="12" name="PERF_VFD_MODE_0_FIBERS"/> + <value value="13" name="PERF_VFD_MODE_1_FIBERS"/> + <value value="14" name="PERF_VFD_MODE_2_FIBERS"/> + <value value="15" name="PERF_VFD_MODE_3_FIBERS"/> + <value value="16" name="PERF_VFD_MODE_4_FIBERS"/> + <value value="17" name="PERF_VFD_TOTAL_VERTICES"/> + <value value="18" name="PERF_VFDP_STALL_CYCLES_VFD"/> + <value value="19" name="PERF_VFDP_STALL_CYCLES_VFD_INDEX"/> + <value value="20" name="PERF_VFDP_STALL_CYCLES_VFD_PROG"/> + <value value="21" name="PERF_VFDP_STARVE_CYCLES_PC"/> + <value value="22" name="PERF_VFDP_VS_STAGE_WAVES"/> +</enum> + +<enum name="a6xx_hlsq_perfcounter_select"> + <value value="0" name="PERF_HLSQ_BUSY_CYCLES"/> + <value value="1" name="PERF_HLSQ_STALL_CYCLES_UCHE"/> + <value value="2" name="PERF_HLSQ_STALL_CYCLES_SP_STATE"/> + <value value="3" name="PERF_HLSQ_STALL_CYCLES_SP_FS_STAGE"/> + <value value="4" name="PERF_HLSQ_UCHE_LATENCY_CYCLES"/> + <value value="5" name="PERF_HLSQ_UCHE_LATENCY_COUNT"/> + <value value="6" name="PERF_HLSQ_FS_STAGE_1X_WAVES"/> + <value value="7" name="PERF_HLSQ_FS_STAGE_2X_WAVES"/> + <value value="8" name="PERF_HLSQ_QUADS"/> + <value value="9" name="PERF_HLSQ_CS_INVOCATIONS"/> + <value value="10" name="PERF_HLSQ_COMPUTE_DRAWCALLS"/> + <value value="11" name="PERF_HLSQ_FS_DATA_WAIT_PROGRAMMING"/> + <value value="12" name="PERF_HLSQ_DUAL_FS_PROG_ACTIVE"/> + <value value="13" name="PERF_HLSQ_DUAL_VS_PROG_ACTIVE"/> + <value value="14" name="PERF_HLSQ_FS_BATCH_COUNT_ZERO"/> + <value value="15" name="PERF_HLSQ_VS_BATCH_COUNT_ZERO"/> + <value value="16" name="PERF_HLSQ_WAVE_PENDING_NO_QUAD"/> + <value value="17" name="PERF_HLSQ_WAVE_PENDING_NO_PRIM_BASE"/> + <value value="18" name="PERF_HLSQ_STALL_CYCLES_VPC"/> + <value value="19" name="PERF_HLSQ_PIXELS"/> + <value value="20" name="PERF_HLSQ_DRAW_MODE_SWITCH_VSFS_SYNC"/> +</enum> + +<enum name="a6xx_vpc_perfcounter_select"> + <value value="0" name="PERF_VPC_BUSY_CYCLES"/> + <value value="1" name="PERF_VPC_WORKING_CYCLES"/> + <value value="2" name="PERF_VPC_STALL_CYCLES_UCHE"/> + <value value="3" name="PERF_VPC_STALL_CYCLES_VFD_WACK"/> + <value value="4" name="PERF_VPC_STALL_CYCLES_HLSQ_PRIM_ALLOC"/> + <value value="5" name="PERF_VPC_STALL_CYCLES_PC"/> + <value value="6" name="PERF_VPC_STALL_CYCLES_SP_LM"/> + <value value="7" name="PERF_VPC_STARVE_CYCLES_SP"/> + <value value="8" name="PERF_VPC_STARVE_CYCLES_LRZ"/> + <value value="9" name="PERF_VPC_PC_PRIMITIVES"/> + <value value="10" name="PERF_VPC_SP_COMPONENTS"/> + <value value="11" name="PERF_VPC_STALL_CYCLES_VPCRAM_POS"/> + <value value="12" name="PERF_VPC_LRZ_ASSIGN_PRIMITIVES"/> + <value value="13" name="PERF_VPC_RB_VISIBLE_PRIMITIVES"/> + <value value="14" name="PERF_VPC_LM_TRANSACTION"/> + <value value="15" name="PERF_VPC_STREAMOUT_TRANSACTION"/> + <value value="16" name="PERF_VPC_VS_BUSY_CYCLES"/> + <value value="17" name="PERF_VPC_PS_BUSY_CYCLES"/> + <value value="18" name="PERF_VPC_VS_WORKING_CYCLES"/> + <value value="19" name="PERF_VPC_PS_WORKING_CYCLES"/> + <value value="20" name="PERF_VPC_STARVE_CYCLES_RB"/> + <value value="21" name="PERF_VPC_NUM_VPCRAM_READ_POS"/> + <value value="22" name="PERF_VPC_WIT_FULL_CYCLES"/> + <value value="23" name="PERF_VPC_VPCRAM_FULL_CYCLES"/> + <value value="24" name="PERF_VPC_LM_FULL_WAIT_FOR_INTP_END"/> + <value value="25" name="PERF_VPC_NUM_VPCRAM_WRITE"/> + <value value="26" name="PERF_VPC_NUM_VPCRAM_READ_SO"/> + <value value="27" name="PERF_VPC_NUM_ATTR_REQ_LM"/> +</enum> + +<enum name="a6xx_tse_perfcounter_select"> + <value value="0" name="PERF_TSE_BUSY_CYCLES"/> + <value value="1" name="PERF_TSE_CLIPPING_CYCLES"/> + <value value="2" name="PERF_TSE_STALL_CYCLES_RAS"/> + <value value="3" name="PERF_TSE_STALL_CYCLES_LRZ_BARYPLANE"/> + <value value="4" name="PERF_TSE_STALL_CYCLES_LRZ_ZPLANE"/> + <value value="5" name="PERF_TSE_STARVE_CYCLES_PC"/> + <value value="6" name="PERF_TSE_INPUT_PRIM"/> + <value value="7" name="PERF_TSE_INPUT_NULL_PRIM"/> + <value value="8" name="PERF_TSE_TRIVAL_REJ_PRIM"/> + <value value="9" name="PERF_TSE_CLIPPED_PRIM"/> + <value value="10" name="PERF_TSE_ZERO_AREA_PRIM"/> + <value value="11" name="PERF_TSE_FACENESS_CULLED_PRIM"/> + <value value="12" name="PERF_TSE_ZERO_PIXEL_PRIM"/> + <value value="13" name="PERF_TSE_OUTPUT_NULL_PRIM"/> + <value value="14" name="PERF_TSE_OUTPUT_VISIBLE_PRIM"/> + <value value="15" name="PERF_TSE_CINVOCATION"/> + <value value="16" name="PERF_TSE_CPRIMITIVES"/> + <value value="17" name="PERF_TSE_2D_INPUT_PRIM"/> + <value value="18" name="PERF_TSE_2D_ALIVE_CYCLES"/> + <value value="19" name="PERF_TSE_CLIP_PLANES"/> +</enum> + +<enum name="a6xx_ras_perfcounter_select"> + <value value="0" name="PERF_RAS_BUSY_CYCLES"/> + <value value="1" name="PERF_RAS_SUPERTILE_ACTIVE_CYCLES"/> + <value value="2" name="PERF_RAS_STALL_CYCLES_LRZ"/> + <value value="3" name="PERF_RAS_STARVE_CYCLES_TSE"/> + <value value="4" name="PERF_RAS_SUPER_TILES"/> + <value value="5" name="PERF_RAS_8X4_TILES"/> + <value value="6" name="PERF_RAS_MASKGEN_ACTIVE"/> + <value value="7" name="PERF_RAS_FULLY_COVERED_SUPER_TILES"/> + <value value="8" name="PERF_RAS_FULLY_COVERED_8X4_TILES"/> + <value value="9" name="PERF_RAS_PRIM_KILLED_INVISILBE"/> + <value value="10" name="PERF_RAS_SUPERTILE_GEN_ACTIVE_CYCLES"/> + <value value="11" name="PERF_RAS_LRZ_INTF_WORKING_CYCLES"/> + <value value="12" name="PERF_RAS_BLOCKS"/> +</enum> + +<enum name="a6xx_uche_perfcounter_select"> + <value value="0" name="PERF_UCHE_BUSY_CYCLES"/> + <value value="1" name="PERF_UCHE_STALL_CYCLES_ARBITER"/> + <value value="2" name="PERF_UCHE_VBIF_LATENCY_CYCLES"/> + <value value="3" name="PERF_UCHE_VBIF_LATENCY_SAMPLES"/> + <value value="4" name="PERF_UCHE_VBIF_READ_BEATS_TP"/> + <value value="5" name="PERF_UCHE_VBIF_READ_BEATS_VFD"/> + <value value="6" name="PERF_UCHE_VBIF_READ_BEATS_HLSQ"/> + <value value="7" name="PERF_UCHE_VBIF_READ_BEATS_LRZ"/> + <value value="8" name="PERF_UCHE_VBIF_READ_BEATS_SP"/> + <value value="9" name="PERF_UCHE_READ_REQUESTS_TP"/> + <value value="10" name="PERF_UCHE_READ_REQUESTS_VFD"/> + <value value="11" name="PERF_UCHE_READ_REQUESTS_HLSQ"/> + <value value="12" name="PERF_UCHE_READ_REQUESTS_LRZ"/> + <value value="13" name="PERF_UCHE_READ_REQUESTS_SP"/> + <value value="14" name="PERF_UCHE_WRITE_REQUESTS_LRZ"/> + <value value="15" name="PERF_UCHE_WRITE_REQUESTS_SP"/> + <value value="16" name="PERF_UCHE_WRITE_REQUESTS_VPC"/> + <value value="17" name="PERF_UCHE_WRITE_REQUESTS_VSC"/> + <value value="18" name="PERF_UCHE_EVICTS"/> + <value value="19" name="PERF_UCHE_BANK_REQ0"/> + <value value="20" name="PERF_UCHE_BANK_REQ1"/> + <value value="21" name="PERF_UCHE_BANK_REQ2"/> + <value value="22" name="PERF_UCHE_BANK_REQ3"/> + <value value="23" name="PERF_UCHE_BANK_REQ4"/> + <value value="24" name="PERF_UCHE_BANK_REQ5"/> + <value value="25" name="PERF_UCHE_BANK_REQ6"/> + <value value="26" name="PERF_UCHE_BANK_REQ7"/> + <value value="27" name="PERF_UCHE_VBIF_READ_BEATS_CH0"/> + <value value="28" name="PERF_UCHE_VBIF_READ_BEATS_CH1"/> + <value value="29" name="PERF_UCHE_GMEM_READ_BEATS"/> + <value value="30" name="PERF_UCHE_TPH_REF_FULL"/> + <value value="31" name="PERF_UCHE_TPH_VICTIM_FULL"/> + <value value="32" name="PERF_UCHE_TPH_EXT_FULL"/> + <value value="33" name="PERF_UCHE_VBIF_STALL_WRITE_DATA"/> + <value value="34" name="PERF_UCHE_DCMP_LATENCY_SAMPLES"/> + <value value="35" name="PERF_UCHE_DCMP_LATENCY_CYCLES"/> + <value value="36" name="PERF_UCHE_VBIF_READ_BEATS_PC"/> + <value value="37" name="PERF_UCHE_READ_REQUESTS_PC"/> + <value value="38" name="PERF_UCHE_RAM_READ_REQ"/> + <value value="39" name="PERF_UCHE_RAM_WRITE_REQ"/> +</enum> + +<enum name="a6xx_tp_perfcounter_select"> + <value value="0" name="PERF_TP_BUSY_CYCLES"/> + <value value="1" name="PERF_TP_STALL_CYCLES_UCHE"/> + <value value="2" name="PERF_TP_LATENCY_CYCLES"/> + <value value="3" name="PERF_TP_LATENCY_TRANS"/> + <value value="4" name="PERF_TP_FLAG_CACHE_REQUEST_SAMPLES"/> + <value value="5" name="PERF_TP_FLAG_CACHE_REQUEST_LATENCY"/> + <value value="6" name="PERF_TP_L1_CACHELINE_REQUESTS"/> + <value value="7" name="PERF_TP_L1_CACHELINE_MISSES"/> + <value value="8" name="PERF_TP_SP_TP_TRANS"/> + <value value="9" name="PERF_TP_TP_SP_TRANS"/> + <value value="10" name="PERF_TP_OUTPUT_PIXELS"/> + <value value="11" name="PERF_TP_FILTER_WORKLOAD_16BIT"/> + <value value="12" name="PERF_TP_FILTER_WORKLOAD_32BIT"/> + <value value="13" name="PERF_TP_QUADS_RECEIVED"/> + <value value="14" name="PERF_TP_QUADS_OFFSET"/> + <value value="15" name="PERF_TP_QUADS_SHADOW"/> + <value value="16" name="PERF_TP_QUADS_ARRAY"/> + <value value="17" name="PERF_TP_QUADS_GRADIENT"/> + <value value="18" name="PERF_TP_QUADS_1D"/> + <value value="19" name="PERF_TP_QUADS_2D"/> + <value value="20" name="PERF_TP_QUADS_BUFFER"/> + <value value="21" name="PERF_TP_QUADS_3D"/> + <value value="22" name="PERF_TP_QUADS_CUBE"/> + <value value="23" name="PERF_TP_DIVERGENT_QUADS_RECEIVED"/> + <value value="24" name="PERF_TP_PRT_NON_RESIDENT_EVENTS"/> + <value value="25" name="PERF_TP_OUTPUT_PIXELS_POINT"/> + <value value="26" name="PERF_TP_OUTPUT_PIXELS_BILINEAR"/> + <value value="27" name="PERF_TP_OUTPUT_PIXELS_MIP"/> + <value value="28" name="PERF_TP_OUTPUT_PIXELS_ANISO"/> + <value value="29" name="PERF_TP_OUTPUT_PIXELS_ZERO_LOD"/> + <value value="30" name="PERF_TP_FLAG_CACHE_REQUESTS"/> + <value value="31" name="PERF_TP_FLAG_CACHE_MISSES"/> + <value value="32" name="PERF_TP_L1_5_L2_REQUESTS"/> + <value value="33" name="PERF_TP_2D_OUTPUT_PIXELS"/> + <value value="34" name="PERF_TP_2D_OUTPUT_PIXELS_POINT"/> + <value value="35" name="PERF_TP_2D_OUTPUT_PIXELS_BILINEAR"/> + <value value="36" name="PERF_TP_2D_FILTER_WORKLOAD_16BIT"/> + <value value="37" name="PERF_TP_2D_FILTER_WORKLOAD_32BIT"/> + <value value="38" name="PERF_TP_TPA2TPC_TRANS"/> + <value value="39" name="PERF_TP_L1_MISSES_ASTC_1TILE"/> + <value value="40" name="PERF_TP_L1_MISSES_ASTC_2TILE"/> + <value value="41" name="PERF_TP_L1_MISSES_ASTC_4TILE"/> + <value value="42" name="PERF_TP_L1_5_L2_COMPRESS_REQS"/> + <value value="43" name="PERF_TP_L1_5_L2_COMPRESS_MISS"/> + <value value="44" name="PERF_TP_L1_BANK_CONFLICT"/> + <value value="45" name="PERF_TP_L1_5_MISS_LATENCY_CYCLES"/> + <value value="46" name="PERF_TP_L1_5_MISS_LATENCY_TRANS"/> + <value value="47" name="PERF_TP_QUADS_CONSTANT_MULTIPLIED"/> + <value value="48" name="PERF_TP_FRONTEND_WORKING_CYCLES"/> + <value value="49" name="PERF_TP_L1_TAG_WORKING_CYCLES"/> + <value value="50" name="PERF_TP_L1_DATA_WRITE_WORKING_CYCLES"/> + <value value="51" name="PERF_TP_PRE_L1_DECOM_WORKING_CYCLES"/> + <value value="52" name="PERF_TP_BACKEND_WORKING_CYCLES"/> + <value value="53" name="PERF_TP_FLAG_CACHE_WORKING_CYCLES"/> + <value value="54" name="PERF_TP_L1_5_CACHE_WORKING_CYCLES"/> + <value value="55" name="PERF_TP_STARVE_CYCLES_SP"/> + <value value="56" name="PERF_TP_STARVE_CYCLES_UCHE"/> +</enum> + +<enum name="a6xx_sp_perfcounter_select"> + <value value="0" name="PERF_SP_BUSY_CYCLES"/> + <value value="1" name="PERF_SP_ALU_WORKING_CYCLES"/> + <value value="2" name="PERF_SP_EFU_WORKING_CYCLES"/> + <value value="3" name="PERF_SP_STALL_CYCLES_VPC"/> + <value value="4" name="PERF_SP_STALL_CYCLES_TP"/> + <value value="5" name="PERF_SP_STALL_CYCLES_UCHE"/> + <value value="6" name="PERF_SP_STALL_CYCLES_RB"/> + <value value="7" name="PERF_SP_NON_EXECUTION_CYCLES"/> + <value value="8" name="PERF_SP_WAVE_CONTEXTS"/> + <value value="9" name="PERF_SP_WAVE_CONTEXT_CYCLES"/> + <value value="10" name="PERF_SP_FS_STAGE_WAVE_CYCLES"/> + <value value="11" name="PERF_SP_FS_STAGE_WAVE_SAMPLES"/> + <value value="12" name="PERF_SP_VS_STAGE_WAVE_CYCLES"/> + <value value="13" name="PERF_SP_VS_STAGE_WAVE_SAMPLES"/> + <value value="14" name="PERF_SP_FS_STAGE_DURATION_CYCLES"/> + <value value="15" name="PERF_SP_VS_STAGE_DURATION_CYCLES"/> + <value value="16" name="PERF_SP_WAVE_CTRL_CYCLES"/> + <value value="17" name="PERF_SP_WAVE_LOAD_CYCLES"/> + <value value="18" name="PERF_SP_WAVE_EMIT_CYCLES"/> + <value value="19" name="PERF_SP_WAVE_NOP_CYCLES"/> + <value value="20" name="PERF_SP_WAVE_WAIT_CYCLES"/> + <value value="21" name="PERF_SP_WAVE_FETCH_CYCLES"/> + <value value="22" name="PERF_SP_WAVE_IDLE_CYCLES"/> + <value value="23" name="PERF_SP_WAVE_END_CYCLES"/> + <value value="24" name="PERF_SP_WAVE_LONG_SYNC_CYCLES"/> + <value value="25" name="PERF_SP_WAVE_SHORT_SYNC_CYCLES"/> + <value value="26" name="PERF_SP_WAVE_JOIN_CYCLES"/> + <value value="27" name="PERF_SP_LM_LOAD_INSTRUCTIONS"/> + <value value="28" name="PERF_SP_LM_STORE_INSTRUCTIONS"/> + <value value="29" name="PERF_SP_LM_ATOMICS"/> + <value value="30" name="PERF_SP_GM_LOAD_INSTRUCTIONS"/> + <value value="31" name="PERF_SP_GM_STORE_INSTRUCTIONS"/> + <value value="32" name="PERF_SP_GM_ATOMICS"/> + <value value="33" name="PERF_SP_VS_STAGE_TEX_INSTRUCTIONS"/> + <value value="34" name="PERF_SP_VS_STAGE_EFU_INSTRUCTIONS"/> + <value value="35" name="PERF_SP_VS_STAGE_FULL_ALU_INSTRUCTIONS"/> + <value value="36" name="PERF_SP_VS_STAGE_HALF_ALU_INSTRUCTIONS"/> + <value value="37" name="PERF_SP_FS_STAGE_TEX_INSTRUCTIONS"/> + <value value="38" name="PERF_SP_FS_STAGE_CFLOW_INSTRUCTIONS"/> + <value value="39" name="PERF_SP_FS_STAGE_EFU_INSTRUCTIONS"/> + <value value="40" name="PERF_SP_FS_STAGE_FULL_ALU_INSTRUCTIONS"/> + <value value="41" name="PERF_SP_FS_STAGE_HALF_ALU_INSTRUCTIONS"/> + <value value="42" name="PERF_SP_FS_STAGE_BARY_INSTRUCTIONS"/> + <value value="43" name="PERF_SP_VS_INSTRUCTIONS"/> + <value value="44" name="PERF_SP_FS_INSTRUCTIONS"/> + <value value="45" name="PERF_SP_ADDR_LOCK_COUNT"/> + <value value="46" name="PERF_SP_UCHE_READ_TRANS"/> + <value value="47" name="PERF_SP_UCHE_WRITE_TRANS"/> + <value value="48" name="PERF_SP_EXPORT_VPC_TRANS"/> + <value value="49" name="PERF_SP_EXPORT_RB_TRANS"/> + <value value="50" name="PERF_SP_PIXELS_KILLED"/> + <value value="51" name="PERF_SP_ICL1_REQUESTS"/> + <value value="52" name="PERF_SP_ICL1_MISSES"/> + <value value="53" name="PERF_SP_HS_INSTRUCTIONS"/> + <value value="54" name="PERF_SP_DS_INSTRUCTIONS"/> + <value value="55" name="PERF_SP_GS_INSTRUCTIONS"/> + <value value="56" name="PERF_SP_CS_INSTRUCTIONS"/> + <value value="57" name="PERF_SP_GPR_READ"/> + <value value="58" name="PERF_SP_GPR_WRITE"/> + <value value="59" name="PERF_SP_FS_STAGE_HALF_EFU_INSTRUCTIONS"/> + <value value="60" name="PERF_SP_VS_STAGE_HALF_EFU_INSTRUCTIONS"/> + <value value="61" name="PERF_SP_LM_BANK_CONFLICTS"/> + <value value="62" name="PERF_SP_TEX_CONTROL_WORKING_CYCLES"/> + <value value="63" name="PERF_SP_LOAD_CONTROL_WORKING_CYCLES"/> + <value value="64" name="PERF_SP_FLOW_CONTROL_WORKING_CYCLES"/> + <value value="65" name="PERF_SP_LM_WORKING_CYCLES"/> + <value value="66" name="PERF_SP_DISPATCHER_WORKING_CYCLES"/> + <value value="67" name="PERF_SP_SEQUENCER_WORKING_CYCLES"/> + <value value="68" name="PERF_SP_LOW_EFFICIENCY_STARVED_BY_TP"/> + <value value="69" name="PERF_SP_STARVE_CYCLES_HLSQ"/> + <value value="70" name="PERF_SP_NON_EXECUTION_LS_CYCLES"/> + <value value="71" name="PERF_SP_WORKING_EU"/> + <value value="72" name="PERF_SP_ANY_EU_WORKING"/> + <value value="73" name="PERF_SP_WORKING_EU_FS_STAGE"/> + <value value="74" name="PERF_SP_ANY_EU_WORKING_FS_STAGE"/> + <value value="75" name="PERF_SP_WORKING_EU_VS_STAGE"/> + <value value="76" name="PERF_SP_ANY_EU_WORKING_VS_STAGE"/> + <value value="77" name="PERF_SP_WORKING_EU_CS_STAGE"/> + <value value="78" name="PERF_SP_ANY_EU_WORKING_CS_STAGE"/> + <value value="79" name="PERF_SP_GPR_READ_PREFETCH"/> + <value value="80" name="PERF_SP_GPR_READ_CONFLICT"/> + <value value="81" name="PERF_SP_GPR_WRITE_CONFLICT"/> + <value value="82" name="PERF_SP_GM_LOAD_LATENCY_CYCLES"/> + <value value="83" name="PERF_SP_GM_LOAD_LATENCY_SAMPLES"/> + <value value="84" name="PERF_SP_EXECUTABLE_WAVES"/> +</enum> + +<enum name="a6xx_rb_perfcounter_select"> + <value value="0" name="PERF_RB_BUSY_CYCLES"/> + <value value="1" name="PERF_RB_STALL_CYCLES_HLSQ"/> + <value value="2" name="PERF_RB_STALL_CYCLES_FIFO0_FULL"/> + <value value="3" name="PERF_RB_STALL_CYCLES_FIFO1_FULL"/> + <value value="4" name="PERF_RB_STALL_CYCLES_FIFO2_FULL"/> + <value value="5" name="PERF_RB_STARVE_CYCLES_SP"/> + <value value="6" name="PERF_RB_STARVE_CYCLES_LRZ_TILE"/> + <value value="7" name="PERF_RB_STARVE_CYCLES_CCU"/> + <value value="8" name="PERF_RB_STARVE_CYCLES_Z_PLANE"/> + <value value="9" name="PERF_RB_STARVE_CYCLES_BARY_PLANE"/> + <value value="10" name="PERF_RB_Z_WORKLOAD"/> + <value value="11" name="PERF_RB_HLSQ_ACTIVE"/> + <value value="12" name="PERF_RB_Z_READ"/> + <value value="13" name="PERF_RB_Z_WRITE"/> + <value value="14" name="PERF_RB_C_READ"/> + <value value="15" name="PERF_RB_C_WRITE"/> + <value value="16" name="PERF_RB_TOTAL_PASS"/> + <value value="17" name="PERF_RB_Z_PASS"/> + <value value="18" name="PERF_RB_Z_FAIL"/> + <value value="19" name="PERF_RB_S_FAIL"/> + <value value="20" name="PERF_RB_BLENDED_FXP_COMPONENTS"/> + <value value="21" name="PERF_RB_BLENDED_FP16_COMPONENTS"/> + <value value="22" name="PERF_RB_PS_INVOCATIONS"/> + <value value="23" name="PERF_RB_2D_ALIVE_CYCLES"/> + <value value="24" name="PERF_RB_2D_STALL_CYCLES_A2D"/> + <value value="25" name="PERF_RB_2D_STARVE_CYCLES_SRC"/> + <value value="26" name="PERF_RB_2D_STARVE_CYCLES_SP"/> + <value value="27" name="PERF_RB_2D_STARVE_CYCLES_DST"/> + <value value="28" name="PERF_RB_2D_VALID_PIXELS"/> + <value value="29" name="PERF_RB_3D_PIXELS"/> + <value value="30" name="PERF_RB_BLENDER_WORKING_CYCLES"/> + <value value="31" name="PERF_RB_ZPROC_WORKING_CYCLES"/> + <value value="32" name="PERF_RB_CPROC_WORKING_CYCLES"/> + <value value="33" name="PERF_RB_SAMPLER_WORKING_CYCLES"/> + <value value="34" name="PERF_RB_STALL_CYCLES_CCU_COLOR_READ"/> + <value value="35" name="PERF_RB_STALL_CYCLES_CCU_COLOR_WRITE"/> + <value value="36" name="PERF_RB_STALL_CYCLES_CCU_DEPTH_READ"/> + <value value="37" name="PERF_RB_STALL_CYCLES_CCU_DEPTH_WRITE"/> + <value value="38" name="PERF_RB_STALL_CYCLES_VPC"/> + <value value="39" name="PERF_RB_2D_INPUT_TRANS"/> + <value value="40" name="PERF_RB_2D_OUTPUT_RB_DST_TRANS"/> + <value value="41" name="PERF_RB_2D_OUTPUT_RB_SRC_TRANS"/> + <value value="42" name="PERF_RB_BLENDED_FP32_COMPONENTS"/> + <value value="43" name="PERF_RB_COLOR_PIX_TILES"/> + <value value="44" name="PERF_RB_STALL_CYCLES_CCU"/> + <value value="45" name="PERF_RB_EARLY_Z_ARB3_GRANT"/> + <value value="46" name="PERF_RB_LATE_Z_ARB3_GRANT"/> + <value value="47" name="PERF_RB_EARLY_Z_SKIP_GRANT"/> +</enum> + +<enum name="a6xx_vsc_perfcounter_select"> + <value value="0" name="PERF_VSC_BUSY_CYCLES"/> + <value value="1" name="PERF_VSC_WORKING_CYCLES"/> + <value value="2" name="PERF_VSC_STALL_CYCLES_UCHE"/> + <value value="3" name="PERF_VSC_EOT_NUM"/> + <value value="4" name="PERF_VSC_INPUT_TILES"/> +</enum> + +<enum name="a6xx_ccu_perfcounter_select"> + <value value="0" name="PERF_CCU_BUSY_CYCLES"/> + <value value="1" name="PERF_CCU_STALL_CYCLES_RB_DEPTH_RETURN"/> + <value value="2" name="PERF_CCU_STALL_CYCLES_RB_COLOR_RETURN"/> + <value value="3" name="PERF_CCU_STARVE_CYCLES_FLAG_RETURN"/> + <value value="4" name="PERF_CCU_DEPTH_BLOCKS"/> + <value value="5" name="PERF_CCU_COLOR_BLOCKS"/> + <value value="6" name="PERF_CCU_DEPTH_BLOCK_HIT"/> + <value value="7" name="PERF_CCU_COLOR_BLOCK_HIT"/> + <value value="8" name="PERF_CCU_PARTIAL_BLOCK_READ"/> + <value value="9" name="PERF_CCU_GMEM_READ"/> + <value value="10" name="PERF_CCU_GMEM_WRITE"/> + <value value="11" name="PERF_CCU_DEPTH_READ_FLAG0_COUNT"/> + <value value="12" name="PERF_CCU_DEPTH_READ_FLAG1_COUNT"/> + <value value="13" name="PERF_CCU_DEPTH_READ_FLAG2_COUNT"/> + <value value="14" name="PERF_CCU_DEPTH_READ_FLAG3_COUNT"/> + <value value="15" name="PERF_CCU_DEPTH_READ_FLAG4_COUNT"/> + <value value="16" name="PERF_CCU_DEPTH_READ_FLAG5_COUNT"/> + <value value="17" name="PERF_CCU_DEPTH_READ_FLAG6_COUNT"/> + <value value="18" name="PERF_CCU_DEPTH_READ_FLAG8_COUNT"/> + <value value="19" name="PERF_CCU_COLOR_READ_FLAG0_COUNT"/> + <value value="20" name="PERF_CCU_COLOR_READ_FLAG1_COUNT"/> + <value value="21" name="PERF_CCU_COLOR_READ_FLAG2_COUNT"/> + <value value="22" name="PERF_CCU_COLOR_READ_FLAG3_COUNT"/> + <value value="23" name="PERF_CCU_COLOR_READ_FLAG4_COUNT"/> + <value value="24" name="PERF_CCU_COLOR_READ_FLAG5_COUNT"/> + <value value="25" name="PERF_CCU_COLOR_READ_FLAG6_COUNT"/> + <value value="26" name="PERF_CCU_COLOR_READ_FLAG8_COUNT"/> + <value value="27" name="PERF_CCU_2D_RD_REQ"/> + <value value="28" name="PERF_CCU_2D_WR_REQ"/> +</enum> + +<enum name="a6xx_lrz_perfcounter_select"> + <value value="0" name="PERF_LRZ_BUSY_CYCLES"/> + <value value="1" name="PERF_LRZ_STARVE_CYCLES_RAS"/> + <value value="2" name="PERF_LRZ_STALL_CYCLES_RB"/> + <value value="3" name="PERF_LRZ_STALL_CYCLES_VSC"/> + <value value="4" name="PERF_LRZ_STALL_CYCLES_VPC"/> + <value value="5" name="PERF_LRZ_STALL_CYCLES_FLAG_PREFETCH"/> + <value value="6" name="PERF_LRZ_STALL_CYCLES_UCHE"/> + <value value="7" name="PERF_LRZ_LRZ_READ"/> + <value value="8" name="PERF_LRZ_LRZ_WRITE"/> + <value value="9" name="PERF_LRZ_READ_LATENCY"/> + <value value="10" name="PERF_LRZ_MERGE_CACHE_UPDATING"/> + <value value="11" name="PERF_LRZ_PRIM_KILLED_BY_MASKGEN"/> + <value value="12" name="PERF_LRZ_PRIM_KILLED_BY_LRZ"/> + <value value="13" name="PERF_LRZ_VISIBLE_PRIM_AFTER_LRZ"/> + <value value="14" name="PERF_LRZ_FULL_8X8_TILES"/> + <value value="15" name="PERF_LRZ_PARTIAL_8X8_TILES"/> + <value value="16" name="PERF_LRZ_TILE_KILLED"/> + <value value="17" name="PERF_LRZ_TOTAL_PIXEL"/> + <value value="18" name="PERF_LRZ_VISIBLE_PIXEL_AFTER_LRZ"/> + <value value="19" name="PERF_LRZ_FULLY_COVERED_TILES"/> + <value value="20" name="PERF_LRZ_PARTIAL_COVERED_TILES"/> + <value value="21" name="PERF_LRZ_FEEDBACK_ACCEPT"/> + <value value="22" name="PERF_LRZ_FEEDBACK_DISCARD"/> + <value value="23" name="PERF_LRZ_FEEDBACK_STALL"/> + <value value="24" name="PERF_LRZ_STALL_CYCLES_RB_ZPLANE"/> + <value value="25" name="PERF_LRZ_STALL_CYCLES_RB_BPLANE"/> + <value value="26" name="PERF_LRZ_STALL_CYCLES_VC"/> + <value value="27" name="PERF_LRZ_RAS_MASK_TRANS"/> +</enum> + +<enum name="a6xx_cmp_perfcounter_select"> + <value value="0" name="PERF_CMPDECMP_STALL_CYCLES_ARB"/> + <value value="1" name="PERF_CMPDECMP_VBIF_LATENCY_CYCLES"/> + <value value="2" name="PERF_CMPDECMP_VBIF_LATENCY_SAMPLES"/> + <value value="3" name="PERF_CMPDECMP_VBIF_READ_DATA_CCU"/> + <value value="4" name="PERF_CMPDECMP_VBIF_WRITE_DATA_CCU"/> + <value value="5" name="PERF_CMPDECMP_VBIF_READ_REQUEST"/> + <value value="6" name="PERF_CMPDECMP_VBIF_WRITE_REQUEST"/> + <value value="7" name="PERF_CMPDECMP_VBIF_READ_DATA"/> + <value value="8" name="PERF_CMPDECMP_VBIF_WRITE_DATA"/> + <value value="9" name="PERF_CMPDECMP_FLAG_FETCH_CYCLES"/> + <value value="10" name="PERF_CMPDECMP_FLAG_FETCH_SAMPLES"/> + <value value="11" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG1_COUNT"/> + <value value="12" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG2_COUNT"/> + <value value="13" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG3_COUNT"/> + <value value="14" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG4_COUNT"/> + <value value="15" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG5_COUNT"/> + <value value="16" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG6_COUNT"/> + <value value="17" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG8_COUNT"/> + <value value="18" name="PERF_CMPDECMP_COLOR_WRITE_FLAG1_COUNT"/> + <value value="19" name="PERF_CMPDECMP_COLOR_WRITE_FLAG2_COUNT"/> + <value value="20" name="PERF_CMPDECMP_COLOR_WRITE_FLAG3_COUNT"/> + <value value="21" name="PERF_CMPDECMP_COLOR_WRITE_FLAG4_COUNT"/> + <value value="22" name="PERF_CMPDECMP_COLOR_WRITE_FLAG5_COUNT"/> + <value value="23" name="PERF_CMPDECMP_COLOR_WRITE_FLAG6_COUNT"/> + <value value="24" name="PERF_CMPDECMP_COLOR_WRITE_FLAG8_COUNT"/> + <value value="25" name="PERF_CMPDECMP_2D_STALL_CYCLES_VBIF_REQ"/> + <value value="26" name="PERF_CMPDECMP_2D_STALL_CYCLES_VBIF_WR"/> + <value value="27" name="PERF_CMPDECMP_2D_STALL_CYCLES_VBIF_RETURN"/> + <value value="28" name="PERF_CMPDECMP_2D_RD_DATA"/> + <value value="29" name="PERF_CMPDECMP_2D_WR_DATA"/> + <value value="30" name="PERF_CMPDECMP_VBIF_READ_DATA_UCHE_CH0"/> + <value value="31" name="PERF_CMPDECMP_VBIF_READ_DATA_UCHE_CH1"/> + <value value="32" name="PERF_CMPDECMP_2D_OUTPUT_TRANS"/> + <value value="33" name="PERF_CMPDECMP_VBIF_WRITE_DATA_UCHE"/> + <value value="34" name="PERF_CMPDECMP_DEPTH_WRITE_FLAG0_COUNT"/> + <value value="35" name="PERF_CMPDECMP_COLOR_WRITE_FLAG0_COUNT"/> + <value value="36" name="PERF_CMPDECMP_COLOR_WRITE_FLAGALPHA_COUNT"/> + <value value="37" name="PERF_CMPDECMP_2D_BUSY_CYCLES"/> + <value value="38" name="PERF_CMPDECMP_2D_REORDER_STARVE_CYCLES"/> + <value value="39" name="PERF_CMPDECMP_2D_PIXELS"/> +</enum> + +</database> diff --git a/drivers/gpu/drm/msm/registers/adreno/a7xx_enums.xml b/drivers/gpu/drm/msm/registers/adreno/a7xx_enums.xml new file mode 100644 index 000000000000..661b0dd0f675 --- /dev/null +++ b/drivers/gpu/drm/msm/registers/adreno/a7xx_enums.xml @@ -0,0 +1,223 @@ +<?xml version="1.0" encoding="UTF-8"?> +<database xmlns="http://nouveau.freedesktop.org/" +xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" +xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> +<import file="freedreno_copyright.xml"/> +<import file="adreno/adreno_common.xml"/> +<import file="adreno/adreno_pm4.xml"/> + +<enum name="a7xx_statetype_id"> + <value value="0" name="A7XX_TP0_NCTX_REG"/> + <value value="1" name="A7XX_TP0_CTX0_3D_CVS_REG"/> + <value value="2" name="A7XX_TP0_CTX0_3D_CPS_REG"/> + <value value="3" name="A7XX_TP0_CTX1_3D_CVS_REG"/> + <value value="4" name="A7XX_TP0_CTX1_3D_CPS_REG"/> + <value value="5" name="A7XX_TP0_CTX2_3D_CPS_REG"/> + <value value="6" name="A7XX_TP0_CTX3_3D_CPS_REG"/> + <value value="9" name="A7XX_TP0_TMO_DATA"/> + <value value="10" name="A7XX_TP0_SMO_DATA"/> + <value value="11" name="A7XX_TP0_MIPMAP_BASE_DATA"/> + <value value="32" name="A7XX_SP_NCTX_REG"/> + <value value="33" name="A7XX_SP_CTX0_3D_CVS_REG"/> + <value value="34" name="A7XX_SP_CTX0_3D_CPS_REG"/> + <value value="35" name="A7XX_SP_CTX1_3D_CVS_REG"/> + <value value="36" name="A7XX_SP_CTX1_3D_CPS_REG"/> + <value value="37" name="A7XX_SP_CTX2_3D_CPS_REG"/> + <value value="38" name="A7XX_SP_CTX3_3D_CPS_REG"/> + <value value="39" name="A7XX_SP_INST_DATA"/> + <value value="40" name="A7XX_SP_INST_DATA_1"/> + <value value="41" name="A7XX_SP_LB_0_DATA"/> + <value value="42" name="A7XX_SP_LB_1_DATA"/> + <value value="43" name="A7XX_SP_LB_2_DATA"/> + <value value="44" name="A7XX_SP_LB_3_DATA"/> + <value value="45" name="A7XX_SP_LB_4_DATA"/> + <value value="46" name="A7XX_SP_LB_5_DATA"/> + <value value="47" name="A7XX_SP_LB_6_DATA"/> + <value value="48" name="A7XX_SP_LB_7_DATA"/> + <value value="49" name="A7XX_SP_CB_RAM"/> + <value value="50" name="A7XX_SP_LB_13_DATA"/> + <value value="51" name="A7XX_SP_LB_14_DATA"/> + <value value="52" name="A7XX_SP_INST_TAG"/> + <value value="53" name="A7XX_SP_INST_DATA_2"/> + <value value="54" name="A7XX_SP_TMO_TAG"/> + <value value="55" name="A7XX_SP_SMO_TAG"/> + <value value="56" name="A7XX_SP_STATE_DATA"/> + <value value="57" name="A7XX_SP_HWAVE_RAM"/> + <value value="58" name="A7XX_SP_L0_INST_BUF"/> + <value value="59" name="A7XX_SP_LB_8_DATA"/> + <value value="60" name="A7XX_SP_LB_9_DATA"/> + <value value="61" name="A7XX_SP_LB_10_DATA"/> + <value value="62" name="A7XX_SP_LB_11_DATA"/> + <value value="63" name="A7XX_SP_LB_12_DATA"/> + <value value="64" name="A7XX_HLSQ_DATAPATH_DSTR_META"/> + <value value="67" name="A7XX_HLSQ_L2STC_TAG_RAM"/> + <value value="68" name="A7XX_HLSQ_L2STC_INFO_CMD"/> + <value value="69" name="A7XX_HLSQ_CVS_BE_CTXT_BUF_RAM_TAG"/> + <value value="70" name="A7XX_HLSQ_CPS_BE_CTXT_BUF_RAM_TAG"/> + <value value="71" name="A7XX_HLSQ_GFX_CVS_BE_CTXT_BUF_RAM"/> + <value value="72" name="A7XX_HLSQ_GFX_CPS_BE_CTXT_BUF_RAM"/> + <value value="73" name="A7XX_HLSQ_CHUNK_CVS_RAM"/> + <value value="74" name="A7XX_HLSQ_CHUNK_CPS_RAM"/> + <value value="75" name="A7XX_HLSQ_CHUNK_CVS_RAM_TAG"/> + <value value="76" name="A7XX_HLSQ_CHUNK_CPS_RAM_TAG"/> + <value value="77" name="A7XX_HLSQ_ICB_CVS_CB_BASE_TAG"/> + <value value="78" name="A7XX_HLSQ_ICB_CPS_CB_BASE_TAG"/> + <value value="79" name="A7XX_HLSQ_CVS_MISC_RAM"/> + <value value="80" name="A7XX_HLSQ_CPS_MISC_RAM"/> + <value value="81" name="A7XX_HLSQ_CPS_MISC_RAM_1"/> + <value value="82" name="A7XX_HLSQ_INST_RAM"/> + <value value="83" name="A7XX_HLSQ_GFX_CVS_CONST_RAM"/> + <value value="84" name="A7XX_HLSQ_GFX_CPS_CONST_RAM"/> + <value value="85" name="A7XX_HLSQ_CVS_MISC_RAM_TAG"/> + <value value="86" name="A7XX_HLSQ_CPS_MISC_RAM_TAG"/> + <value value="87" name="A7XX_HLSQ_INST_RAM_TAG"/> + <value value="88" name="A7XX_HLSQ_GFX_CVS_CONST_RAM_TAG"/> + <value value="89" name="A7XX_HLSQ_GFX_CPS_CONST_RAM_TAG"/> + <value value="90" name="A7XX_HLSQ_GFX_LOCAL_MISC_RAM"/> + <value value="91" name="A7XX_HLSQ_GFX_LOCAL_MISC_RAM_TAG"/> + <value value="92" name="A7XX_HLSQ_INST_RAM_1"/> + <value value="93" name="A7XX_HLSQ_STPROC_META"/> + <value value="94" name="A7XX_HLSQ_BV_BE_META"/> + <value value="95" name="A7XX_HLSQ_INST_RAM_2"/> + <value value="96" name="A7XX_HLSQ_DATAPATH_META"/> + <value value="97" name="A7XX_HLSQ_FRONTEND_META"/> + <value value="98" name="A7XX_HLSQ_INDIRECT_META"/> + <value value="99" name="A7XX_HLSQ_BACKEND_META"/> +</enum> + +<enum name="a7xx_state_location"> + <value value="0" name="A7XX_HLSQ_STATE"/> + <value value="1" name="A7XX_HLSQ_DP"/> + <value value="2" name="A7XX_SP_TOP"/> + <value value="3" name="A7XX_USPTP"/> + <value value="4" name="A7XX_HLSQ_DP_STR"/> +</enum> + +<enum name="a7xx_pipe"> + <value value="0" name="A7XX_PIPE_NONE"/> + <value value="1" name="A7XX_PIPE_BR"/> + <value value="2" name="A7XX_PIPE_BV"/> + <value value="3" name="A7XX_PIPE_LPAC"/> +</enum> + +<enum name="a7xx_cluster"> + <value value="0" name="A7XX_CLUSTER_NONE"/> + <value value="1" name="A7XX_CLUSTER_FE"/> + <value value="2" name="A7XX_CLUSTER_SP_VS"/> + <value value="3" name="A7XX_CLUSTER_PC_VS"/> + <value value="4" name="A7XX_CLUSTER_GRAS"/> + <value value="5" name="A7XX_CLUSTER_SP_PS"/> + <value value="6" name="A7XX_CLUSTER_VPC_PS"/> + <value value="7" name="A7XX_CLUSTER_PS"/> +</enum> + +<enum name="a7xx_debugbus_id"> + <value value="1" name="A7XX_DBGBUS_CP_0_0"/> + <value value="2" name="A7XX_DBGBUS_CP_0_1"/> + <value value="3" name="A7XX_DBGBUS_RBBM"/> + <value value="5" name="A7XX_DBGBUS_GBIF_GX"/> + <value value="6" name="A7XX_DBGBUS_GBIF_CX"/> + <value value="7" name="A7XX_DBGBUS_HLSQ"/> + <value value="9" name="A7XX_DBGBUS_UCHE_0"/> + <value value="10" name="A7XX_DBGBUS_UCHE_1"/> + <value value="13" name="A7XX_DBGBUS_TESS_BR"/> + <value value="14" name="A7XX_DBGBUS_TESS_BV"/> + <value value="17" name="A7XX_DBGBUS_PC_BR"/> + <value value="18" name="A7XX_DBGBUS_PC_BV"/> + <value value="21" name="A7XX_DBGBUS_VFDP_BR"/> + <value value="22" name="A7XX_DBGBUS_VFDP_BV"/> + <value value="25" name="A7XX_DBGBUS_VPC_BR"/> + <value value="26" name="A7XX_DBGBUS_VPC_BV"/> + <value value="29" name="A7XX_DBGBUS_TSE_BR"/> + <value value="30" name="A7XX_DBGBUS_TSE_BV"/> + <value value="33" name="A7XX_DBGBUS_RAS_BR"/> + <value value="34" name="A7XX_DBGBUS_RAS_BV"/> + <value value="37" name="A7XX_DBGBUS_VSC"/> + <value value="39" name="A7XX_DBGBUS_COM_0"/> + <value value="43" name="A7XX_DBGBUS_LRZ_BR"/> + <value value="44" name="A7XX_DBGBUS_LRZ_BV"/> + <value value="47" name="A7XX_DBGBUS_UFC_0"/> + <value value="48" name="A7XX_DBGBUS_UFC_1"/> + <value value="55" name="A7XX_DBGBUS_GMU_GX"/> + <value value="59" name="A7XX_DBGBUS_DBGC"/> + <value value="60" name="A7XX_DBGBUS_CX"/> + <value value="61" name="A7XX_DBGBUS_GMU_CX"/> + <value value="62" name="A7XX_DBGBUS_GPC_BR"/> + <value value="63" name="A7XX_DBGBUS_GPC_BV"/> + <value value="66" name="A7XX_DBGBUS_LARC"/> + <value value="68" name="A7XX_DBGBUS_HLSQ_SPTP"/> + <value value="70" name="A7XX_DBGBUS_RB_0"/> + <value value="71" name="A7XX_DBGBUS_RB_1"/> + <value value="72" name="A7XX_DBGBUS_RB_2"/> + <value value="73" name="A7XX_DBGBUS_RB_3"/> + <value value="74" name="A7XX_DBGBUS_RB_4"/> + <value value="75" name="A7XX_DBGBUS_RB_5"/> + <value value="102" name="A7XX_DBGBUS_UCHE_WRAPPER"/> + <value value="106" name="A7XX_DBGBUS_CCU_0"/> + <value value="107" name="A7XX_DBGBUS_CCU_1"/> + <value value="108" name="A7XX_DBGBUS_CCU_2"/> + <value value="109" name="A7XX_DBGBUS_CCU_3"/> + <value value="110" name="A7XX_DBGBUS_CCU_4"/> + <value value="111" name="A7XX_DBGBUS_CCU_5"/> + <value value="138" name="A7XX_DBGBUS_VFD_BR_0"/> + <value value="139" name="A7XX_DBGBUS_VFD_BR_1"/> + <value value="140" name="A7XX_DBGBUS_VFD_BR_2"/> + <value value="141" name="A7XX_DBGBUS_VFD_BR_3"/> + <value value="142" name="A7XX_DBGBUS_VFD_BR_4"/> + <value value="143" name="A7XX_DBGBUS_VFD_BR_5"/> + <value value="144" name="A7XX_DBGBUS_VFD_BR_6"/> + <value value="145" name="A7XX_DBGBUS_VFD_BR_7"/> + <value value="202" name="A7XX_DBGBUS_VFD_BV_0"/> + <value value="203" name="A7XX_DBGBUS_VFD_BV_1"/> + <value value="204" name="A7XX_DBGBUS_VFD_BV_2"/> + <value value="205" name="A7XX_DBGBUS_VFD_BV_3"/> + <value value="234" name="A7XX_DBGBUS_USP_0"/> + <value value="235" name="A7XX_DBGBUS_USP_1"/> + <value value="236" name="A7XX_DBGBUS_USP_2"/> + <value value="237" name="A7XX_DBGBUS_USP_3"/> + <value value="238" name="A7XX_DBGBUS_USP_4"/> + <value value="239" name="A7XX_DBGBUS_USP_5"/> + <value value="266" name="A7XX_DBGBUS_TP_0"/> + <value value="267" name="A7XX_DBGBUS_TP_1"/> + <value value="268" name="A7XX_DBGBUS_TP_2"/> + <value value="269" name="A7XX_DBGBUS_TP_3"/> + <value value="270" name="A7XX_DBGBUS_TP_4"/> + <value value="271" name="A7XX_DBGBUS_TP_5"/> + <value value="272" name="A7XX_DBGBUS_TP_6"/> + <value value="273" name="A7XX_DBGBUS_TP_7"/> + <value value="274" name="A7XX_DBGBUS_TP_8"/> + <value value="275" name="A7XX_DBGBUS_TP_9"/> + <value value="276" name="A7XX_DBGBUS_TP_10"/> + <value value="277" name="A7XX_DBGBUS_TP_11"/> + <value value="330" name="A7XX_DBGBUS_USPTP_0"/> + <value value="331" name="A7XX_DBGBUS_USPTP_1"/> + <value value="332" name="A7XX_DBGBUS_USPTP_2"/> + <value value="333" name="A7XX_DBGBUS_USPTP_3"/> + <value value="334" name="A7XX_DBGBUS_USPTP_4"/> + <value value="335" name="A7XX_DBGBUS_USPTP_5"/> + <value value="336" name="A7XX_DBGBUS_USPTP_6"/> + <value value="337" name="A7XX_DBGBUS_USPTP_7"/> + <value value="338" name="A7XX_DBGBUS_USPTP_8"/> + <value value="339" name="A7XX_DBGBUS_USPTP_9"/> + <value value="340" name="A7XX_DBGBUS_USPTP_10"/> + <value value="341" name="A7XX_DBGBUS_USPTP_11"/> + <value value="396" name="A7XX_DBGBUS_CCHE_0"/> + <value value="397" name="A7XX_DBGBUS_CCHE_1"/> + <value value="398" name="A7XX_DBGBUS_CCHE_2"/> + <value value="408" name="A7XX_DBGBUS_VPC_DSTR_0"/> + <value value="409" name="A7XX_DBGBUS_VPC_DSTR_1"/> + <value value="410" name="A7XX_DBGBUS_VPC_DSTR_2"/> + <value value="411" name="A7XX_DBGBUS_HLSQ_DP_STR_0"/> + <value value="412" name="A7XX_DBGBUS_HLSQ_DP_STR_1"/> + <value value="413" name="A7XX_DBGBUS_HLSQ_DP_STR_2"/> + <value value="414" name="A7XX_DBGBUS_HLSQ_DP_STR_3"/> + <value value="415" name="A7XX_DBGBUS_HLSQ_DP_STR_4"/> + <value value="416" name="A7XX_DBGBUS_HLSQ_DP_STR_5"/> + <value value="443" name="A7XX_DBGBUS_UFC_DSTR_0"/> + <value value="444" name="A7XX_DBGBUS_UFC_DSTR_1"/> + <value value="445" name="A7XX_DBGBUS_UFC_DSTR_2"/> + <value value="446" name="A7XX_DBGBUS_CGC_SUBCORE"/> + <value value="447" name="A7XX_DBGBUS_CGC_CORE"/> +</enum> + +</database> diff --git a/drivers/gpu/drm/msm/registers/adreno/a7xx_perfcntrs.xml b/drivers/gpu/drm/msm/registers/adreno/a7xx_perfcntrs.xml new file mode 100644 index 000000000000..9bf78b0a854b --- /dev/null +++ b/drivers/gpu/drm/msm/registers/adreno/a7xx_perfcntrs.xml @@ -0,0 +1,1030 @@ +<?xml version="1.0" encoding="UTF-8"?> +<database xmlns="http://nouveau.freedesktop.org/" +xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" +xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> +<import file="freedreno_copyright.xml"/> +<import file="adreno/adreno_common.xml"/> +<import file="adreno/adreno_pm4.xml"/> + +<enum name="a7xx_cp_perfcounter_select"> + <value value="0" name="A7XX_PERF_CP_ALWAYS_COUNT"/> + <value value="1" name="A7XX_PERF_CP_BUSY_GFX_CORE_IDLE"/> + <value value="2" name="A7XX_PERF_CP_BUSY_CYCLES"/> + <value value="3" name="A7XX_PERF_CP_NUM_PREEMPTIONS"/> + <value value="4" name="A7XX_PERF_CP_PREEMPTION_REACTION_DELAY"/> + <value value="5" name="A7XX_PERF_CP_PREEMPTION_SWITCH_OUT_TIME"/> + <value value="6" name="A7XX_PERF_CP_PREEMPTION_SWITCH_IN_TIME"/> + <value value="7" name="A7XX_PERF_CP_DEAD_DRAWS_IN_BIN_RENDER"/> + <value value="8" name="A7XX_PERF_CP_PREDICATED_DRAWS_KILLED"/> + <value value="9" name="A7XX_PERF_CP_MODE_SWITCH"/> + <value value="10" name="A7XX_PERF_CP_ZPASS_DONE"/> + <value value="11" name="A7XX_PERF_CP_CONTEXT_DONE"/> + <value value="12" name="A7XX_PERF_CP_CACHE_FLUSH"/> + <value value="13" name="A7XX_PERF_CP_LONG_PREEMPTIONS"/> + <value value="14" name="A7XX_PERF_CP_SQE_I_CACHE_STARVE"/> + <value value="15" name="A7XX_PERF_CP_SQE_IDLE"/> + <value value="16" name="A7XX_PERF_CP_SQE_PM4_STARVE_RB_IB"/> + <value value="17" name="A7XX_PERF_CP_SQE_PM4_STARVE_SDS"/> + <value value="18" name="A7XX_PERF_CP_SQE_MRB_STARVE"/> + <value value="19" name="A7XX_PERF_CP_SQE_RRB_STARVE"/> + <value value="20" name="A7XX_PERF_CP_SQE_VSD_STARVE"/> + <value value="21" name="A7XX_PERF_CP_VSD_DECODE_STARVE"/> + <value value="22" name="A7XX_PERF_CP_SQE_PIPE_OUT_STALL"/> + <value value="23" name="A7XX_PERF_CP_SQE_SYNC_STALL"/> + <value value="24" name="A7XX_PERF_CP_SQE_PM4_WFI_STALL"/> + <value value="25" name="A7XX_PERF_CP_SQE_SYS_WFI_STALL"/> + <value value="26" name="A7XX_PERF_CP_SQE_T4_EXEC"/> + <value value="27" name="A7XX_PERF_CP_SQE_LOAD_STATE_EXEC"/> + <value value="28" name="A7XX_PERF_CP_SQE_SAVE_SDS_STATE"/> + <value value="29" name="A7XX_PERF_CP_SQE_DRAW_EXEC"/> + <value value="30" name="A7XX_PERF_CP_SQE_CTXT_REG_BUNCH_EXEC"/> + <value value="31" name="A7XX_PERF_CP_SQE_EXEC_PROFILED"/> + <value value="32" name="A7XX_PERF_CP_MEMORY_POOL_EMPTY"/> + <value value="33" name="A7XX_PERF_CP_MEMORY_POOL_SYNC_STALL"/> + <value value="34" name="A7XX_PERF_CP_MEMORY_POOL_ABOVE_THRESH"/> + <value value="35" name="A7XX_PERF_CP_AHB_WR_STALL_PRE_DRAWS"/> + <value value="36" name="A7XX_PERF_CP_AHB_STALL_SQE_GMU"/> + <value value="37" name="A7XX_PERF_CP_AHB_STALL_SQE_WR_OTHER"/> + <value value="38" name="A7XX_PERF_CP_AHB_STALL_SQE_RD_OTHER"/> + <value value="39" name="A7XX_PERF_CP_CLUSTER0_EMPTY"/> + <value value="40" name="A7XX_PERF_CP_CLUSTER1_EMPTY"/> + <value value="41" name="A7XX_PERF_CP_CLUSTER2_EMPTY"/> + <value value="42" name="A7XX_PERF_CP_CLUSTER3_EMPTY"/> + <value value="43" name="A7XX_PERF_CP_CLUSTER4_EMPTY"/> + <value value="44" name="A7XX_PERF_CP_CLUSTER5_EMPTY"/> + <value value="45" name="A7XX_PERF_CP_PM4_DATA"/> + <value value="46" name="A7XX_PERF_CP_PM4_HEADERS"/> + <value value="47" name="A7XX_PERF_CP_VBIF_READ_BEATS"/> + <value value="48" name="A7XX_PERF_CP_VBIF_WRITE_BEATS"/> + <value value="49" name="A7XX_PERF_CP_SQE_INSTR_COUNTER"/> + <value value="50" name="A7XX_PERF_CP_RESERVED_50"/> + <value value="51" name="A7XX_PERF_CP_RESERVED_51"/> + <value value="52" name="A7XX_PERF_CP_RESERVED_52"/> + <value value="53" name="A7XX_PERF_CP_RESERVED_53"/> + <value value="54" name="A7XX_PERF_CP_RESERVED_54"/> + <value value="55" name="A7XX_PERF_CP_RESERVED_55"/> + <value value="56" name="A7XX_PERF_CP_RESERVED_56"/> + <value value="57" name="A7XX_PERF_CP_RESERVED_57"/> + <value value="58" name="A7XX_PERF_CP_RESERVED_58"/> + <value value="59" name="A7XX_PERF_CP_RESERVED_59"/> + <value value="60" name="A7XX_PERF_CP_CLUSTER0_FULL"/> + <value value="61" name="A7XX_PERF_CP_CLUSTER1_FULL"/> + <value value="62" name="A7XX_PERF_CP_CLUSTER2_FULL"/> + <value value="63" name="A7XX_PERF_CP_CLUSTER3_FULL"/> + <value value="64" name="A7XX_PERF_CP_CLUSTER4_FULL"/> + <value value="65" name="A7XX_PERF_CP_CLUSTER5_FULL"/> + <value value="66" name="A7XX_PERF_CP_CLUSTER6_FULL"/> + <value value="67" name="A7XX_PERF_CP_CLUSTER6_EMPTY"/> + <value value="68" name="A7XX_PERF_CP_ICACHE_MISSES"/> + <value value="69" name="A7XX_PERF_CP_ICACHE_HITS"/> + <value value="70" name="A7XX_PERF_CP_ICACHE_STALL"/> + <value value="71" name="A7XX_PERF_CP_DCACHE_MISSES"/> + <value value="72" name="A7XX_PERF_CP_DCACHE_HITS"/> + <value value="73" name="A7XX_PERF_CP_DCACHE_STALLS"/> + <value value="74" name="A7XX_PERF_CP_AQE_SQE_STALL"/> + <value value="75" name="A7XX_PERF_CP_SQE_AQE_STARVE"/> + <value value="76" name="A7XX_PERF_CP_PREEMPT_LATENCY"/> + <value value="77" name="A7XX_PERF_CP_SQE_MD8_STALL_CYCLES"/> + <value value="78" name="A7XX_PERF_CP_SQE_MESH_EXEC_CYCLES"/> + <value value="79" name="A7XX_PERF_CP_AQE_NUM_AS_CHUNKS"/> + <value value="80" name="A7XX_PERF_CP_AQE_NUM_MS_CHUNKS"/> +</enum> + +<enum name="a7xx_rbbm_perfcounter_select"> + <value value="0" name="A7XX_PERF_RBBM_ALWAYS_COUNT"/> + <value value="1" name="A7XX_PERF_RBBM_ALWAYS_ON"/> + <value value="2" name="A7XX_PERF_RBBM_TSE_BUSY"/> + <value value="3" name="A7XX_PERF_RBBM_RAS_BUSY"/> + <value value="4" name="A7XX_PERF_RBBM_PC_DCALL_BUSY"/> + <value value="5" name="A7XX_PERF_RBBM_PC_VSD_BUSY"/> + <value value="6" name="A7XX_PERF_RBBM_STATUS_MASKED"/> + <value value="7" name="A7XX_PERF_RBBM_COM_BUSY"/> + <value value="8" name="A7XX_PERF_RBBM_DCOM_BUSY"/> + <value value="9" name="A7XX_PERF_RBBM_VBIF_BUSY"/> + <value value="10" name="A7XX_PERF_RBBM_VSC_BUSY"/> + <value value="11" name="A7XX_PERF_RBBM_TESS_BUSY"/> + <value value="12" name="A7XX_PERF_RBBM_UCHE_BUSY"/> + <value value="13" name="A7XX_PERF_RBBM_HLSQ_BUSY"/> +</enum> + +<enum name="a7xx_pc_perfcounter_select"> + <value value="0" name="A7XX_PERF_PC_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_PC_WORKING_CYCLES"/> + <value value="2" name="A7XX_PERF_PC_STALL_CYCLES_VFD"/> + <value value="3" name="A7XX_PERF_PC_RESERVED"/> + <value value="4" name="A7XX_PERF_PC_STALL_CYCLES_VPC"/> + <value value="5" name="A7XX_PERF_PC_STALL_CYCLES_UCHE"/> + <value value="6" name="A7XX_PERF_PC_STALL_CYCLES_TESS"/> + <value value="7" name="A7XX_PERF_PC_STALL_CYCLES_VFD_ONLY"/> + <value value="8" name="A7XX_PERF_PC_STALL_CYCLES_VPC_ONLY"/> + <value value="9" name="A7XX_PERF_PC_PASS1_TF_STALL_CYCLES"/> + <value value="10" name="A7XX_PERF_PC_STARVE_CYCLES_FOR_INDEX"/> + <value value="11" name="A7XX_PERF_PC_STARVE_CYCLES_FOR_TESS_FACTOR"/> + <value value="12" name="A7XX_PERF_PC_STARVE_CYCLES_FOR_VIZ_STREAM"/> + <value value="13" name="A7XX_PERF_PC_STARVE_CYCLES_DI"/> + <value value="14" name="A7XX_PERF_PC_VIS_STREAMS_LOADED"/> + <value value="15" name="A7XX_PERF_PC_INSTANCES"/> + <value value="16" name="A7XX_PERF_PC_VPC_PRIMITIVES"/> + <value value="17" name="A7XX_PERF_PC_DEAD_PRIM"/> + <value value="18" name="A7XX_PERF_PC_LIVE_PRIM"/> + <value value="19" name="A7XX_PERF_PC_VERTEX_HITS"/> + <value value="20" name="A7XX_PERF_PC_IA_VERTICES"/> + <value value="21" name="A7XX_PERF_PC_IA_PRIMITIVES"/> + <value value="22" name="A7XX_PERF_PC_RESERVED_22"/> + <value value="23" name="A7XX_PERF_PC_HS_INVOCATIONS"/> + <value value="24" name="A7XX_PERF_PC_DS_INVOCATIONS"/> + <value value="25" name="A7XX_PERF_PC_VS_INVOCATIONS"/> + <value value="26" name="A7XX_PERF_PC_GS_INVOCATIONS"/> + <value value="27" name="A7XX_PERF_PC_DS_PRIMITIVES"/> + <value value="28" name="A7XX_PERF_PC_3D_DRAWCALLS"/> + <value value="29" name="A7XX_PERF_PC_2D_DRAWCALLS"/> + <value value="30" name="A7XX_PERF_PC_NON_DRAWCALL_GLOBAL_EVENTS"/> + <value value="31" name="A7XX_PERF_PC_TESS_BUSY_CYCLES"/> + <value value="32" name="A7XX_PERF_PC_TESS_WORKING_CYCLES"/> + <value value="33" name="A7XX_PERF_PC_TESS_STALL_CYCLES_PC"/> + <value value="34" name="A7XX_PERF_PC_TESS_STARVE_CYCLES_PC"/> + <value value="35" name="A7XX_PERF_PC_TESS_SINGLE_PRIM_CYCLES"/> + <value value="36" name="A7XX_PERF_PC_TESS_PC_UV_TRANS"/> + <value value="37" name="A7XX_PERF_PC_TESS_PC_UV_PATCHES"/> + <value value="38" name="A7XX_PERF_PC_TESS_FACTOR_TRANS"/> + <value value="39" name="A7XX_PERF_PC_TAG_CHECKED_VERTICES"/> + <value value="40" name="A7XX_PERF_PC_MESH_VS_WAVES"/> + <value value="41" name="A7XX_PERF_PC_MESH_DRAWS"/> + <value value="42" name="A7XX_PERF_PC_MESH_DEAD_DRAWS"/> + <value value="43" name="A7XX_PERF_PC_MESH_MVIS_EN_DRAWS"/> + <value value="44" name="A7XX_PERF_PC_MESH_DEAD_PRIM"/> + <value value="45" name="A7XX_PERF_PC_MESH_LIVE_PRIM"/> + <value value="46" name="A7XX_PERF_PC_MESH_PA_EN_PRIM"/> + <value value="47" name="A7XX_PERF_PC_STARVE_CYCLES_FOR_MVIS_STREAM"/> + <value value="48" name="A7XX_PERF_PC_STARVE_CYCLES_PREDRAW"/> + <value value="49" name="A7XX_PERF_PC_STALL_CYCLES_COMPUTE_GFX"/> + <value value="50" name="A7XX_PERF_PC_STALL_CYCLES_GFX_COMPUTE"/> + <value value="51" name="A7XX_PERF_PC_TESS_PC_MULTI_PATCH_TRANS"/> +</enum> + +<enum name="a7xx_vfd_perfcounter_select"> + <value value="0" name="A7XX_PERF_VFD_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_VFD_STALL_CYCLES_UCHE"/> + <value value="2" name="A7XX_PERF_VFD_STALL_CYCLES_VPC_ALLOC"/> + <value value="3" name="A7XX_PERF_VFD_STALL_CYCLES_SP_INFO"/> + <value value="4" name="A7XX_PERF_VFD_STALL_CYCLES_SP_ATTR"/> + <value value="5" name="A7XX_PERF_VFD_STARVE_CYCLES_UCHE"/> + <value value="6" name="A7XX_PERF_VFD_RBUFFER_FULL"/> + <value value="7" name="A7XX_PERF_VFD_ATTR_INFO_FIFO_FULL"/> + <value value="8" name="A7XX_PERF_VFD_DECODED_ATTRIBUTE_BYTES"/> + <value value="9" name="A7XX_PERF_VFD_NUM_ATTRIBUTES"/> + <value value="10" name="A7XX_PERF_VFD_UPPER_SHADER_FIBERS"/> + <value value="11" name="A7XX_PERF_VFD_LOWER_SHADER_FIBERS"/> + <value value="12" name="A7XX_PERF_VFD_MODE_0_FIBERS"/> + <value value="13" name="A7XX_PERF_VFD_MODE_1_FIBERS"/> + <value value="14" name="A7XX_PERF_VFD_MODE_2_FIBERS"/> + <value value="15" name="A7XX_PERF_VFD_MODE_3_FIBERS"/> + <value value="16" name="A7XX_PERF_VFD_MODE_4_FIBERS"/> + <value value="17" name="A7XX_PERF_VFD_TOTAL_VERTICES"/> + <value value="18" name="A7XX_PERF_VFDP_STALL_CYCLES_VFD"/> + <value value="19" name="A7XX_PERF_VFDP_STALL_CYCLES_VFD_INDEX"/> + <value value="20" name="A7XX_PERF_VFDP_STALL_CYCLES_VFD_PROG"/> + <value value="21" name="A7XX_PERF_VFDP_STARVE_CYCLES_PC"/> + <value value="22" name="A7XX_PERF_VFDP_VS_STAGE_WAVES"/> + <value value="23" name="A7XX_PERF_VFD_STALL_CYCLES_PRG_END_FE"/> + <value value="24" name="A7XX_PERF_VFD_STALL_CYCLES_CBSYNC"/> +</enum> + +<enum name="a7xx_hlsq_perfcounter_select"> + <value value="0" name="A7XX_PERF_HLSQ_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_HLSQ_STALL_CYCLES_UCHE"/> + <value value="2" name="A7XX_PERF_HLSQ_STALL_CYCLES_SP_STATE"/> + <value value="3" name="A7XX_PERF_HLSQ_STALL_CYCLES_SP_FS_STAGE"/> + <value value="4" name="A7XX_PERF_HLSQ_UCHE_LATENCY_CYCLES"/> + <value value="5" name="A7XX_PERF_HLSQ_UCHE_LATENCY_COUNT"/> + <value value="6" name="A7XX_PERF_HLSQ_RESERVED_6"/> + <value value="7" name="A7XX_PERF_HLSQ_RESERVED_7"/> + <value value="8" name="A7XX_PERF_HLSQ_RESERVED_8"/> + <value value="9" name="A7XX_PERF_HLSQ_RESERVED_9"/> + <value value="10" name="A7XX_PERF_HLSQ_COMPUTE_DRAWCALLS"/> + <value value="11" name="A7XX_PERF_HLSQ_FS_DATA_WAIT_PROGRAMMING"/> + <value value="12" name="A7XX_PERF_HLSQ_DUAL_FS_PROG_ACTIVE"/> + <value value="13" name="A7XX_PERF_HLSQ_DUAL_VS_PROG_ACTIVE"/> + <value value="14" name="A7XX_PERF_HLSQ_FS_BATCH_COUNT_ZERO"/> + <value value="15" name="A7XX_PERF_HLSQ_VS_BATCH_COUNT_ZERO"/> + <value value="16" name="A7XX_PERF_HLSQ_WAVE_PENDING_NO_QUAD"/> + <value value="17" name="A7XX_PERF_HLSQ_WAVE_PENDING_NO_PRIM_BASE"/> + <value value="18" name="A7XX_PERF_HLSQ_STALL_CYCLES_VPC"/> + <value value="19" name="A7XX_PERF_HLSQ_RESERVED_19"/> + <value value="20" name="A7XX_PERF_HLSQ_DRAW_MODE_SWITCH_VSFS_SYNC"/> + <value value="21" name="A7XX_PERF_HLSQ_VSBR_STALL_CYCLES"/> + <value value="22" name="A7XX_PERF_HLSQ_FS_STALL_CYCLES"/> + <value value="23" name="A7XX_PERF_HLSQ_LPAC_STALL_CYCLES"/> + <value value="24" name="A7XX_PERF_HLSQ_BV_STALL_CYCLES"/> + <value value="25" name="A7XX_PERF_HLSQ_VSBR_DEREF_CYCLES"/> + <value value="26" name="A7XX_PERF_HLSQ_FS_DEREF_CYCLES"/> + <value value="27" name="A7XX_PERF_HLSQ_LPAC_DEREF_CYCLES"/> + <value value="28" name="A7XX_PERF_HLSQ_BV_DEREF_CYCLES"/> + <value value="29" name="A7XX_PERF_HLSQ_VSBR_S2W_CYCLES"/> + <value value="30" name="A7XX_PERF_HLSQ_FS_S2W_CYCLES"/> + <value value="31" name="A7XX_PERF_HLSQ_LPAC_S2W_CYCLES"/> + <value value="32" name="A7XX_PERF_HLSQ_BV_S2W_CYCLES"/> + <value value="33" name="A7XX_PERF_HLSQ_VSBR_WAIT_FS_S2W"/> + <value value="34" name="A7XX_PERF_HLSQ_FS_WAIT_VS_S2W"/> + <value value="35" name="A7XX_PERF_HLSQ_LPAC_WAIT_VS_S2W"/> + <value value="36" name="A7XX_PERF_HLSQ_BV_WAIT_FS_S2W"/> + <value value="37" name="A7XX_PERF_HLSQ_VS_WAIT_CONST_RESOURCE"/> + <value value="38" name="A7XX_PERF_HLSQ_FS_WAIT_SAME_VS_S2W"/> + <value value="39" name="A7XX_PERF_HLSQ_FS_STARVING_SP"/> + <value value="40" name="A7XX_PERF_HLSQ_VS_DATA_WAIT_PROGRAMMING"/> + <value value="41" name="A7XX_PERF_HLSQ_BV_DATA_WAIT_PROGRAMMING"/> + <value value="42" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXTS_VS"/> + <value value="43" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXT_CYCLES_VS"/> + <value value="44" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXTS_FS"/> + <value value="45" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXT_CYCLES_FS"/> + <value value="46" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXTS_BV"/> + <value value="47" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXT_CYCLES_BV"/> + <value value="48" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXTS_LPAC"/> + <value value="49" name="A7XX_PERF_HLSQ_STPROC_WAVE_CONTEXT_CYCLES_LPAC"/> + <value value="50" name="A7XX_PERF_HLSQ_SPTROC_STCHE_WARMUP_INC_VS"/> + <value value="51" name="A7XX_PERF_HLSQ_SPTROC_STCHE_WARMUP_INC_FS"/> + <value value="52" name="A7XX_PERF_HLSQ_SPTROC_STCHE_WARMUP_INC_BV"/> + <value value="53" name="A7XX_PERF_HLSQ_SPTROC_STCHE_WARMUP_INC_LPAC"/> + <value value="54" name="A7XX_PERF_HLSQ_SPTROC_STCHE_MISS_INC_VS"/> + <value value="55" name="A7XX_PERF_HLSQ_SPTROC_STCHE_MISS_INC_FS"/> + <value value="56" name="A7XX_PERF_HLSQ_SPTROC_STCHE_MISS_INC_BV"/> + <value value="57" name="A7XX_PERF_HLSQ_SPTROC_STCHE_MISS_INC_LPAC"/> +</enum> + +<enum name="a7xx_vpc_perfcounter_select"> + <value value="0" name="A7XX_PERF_VPC_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_VPC_WORKING_CYCLES"/> + <value value="2" name="A7XX_PERF_VPC_STALL_CYCLES_UCHE"/> + <value value="3" name="A7XX_PERF_VPC_STALL_CYCLES_VFD_WACK"/> + <value value="4" name="A7XX_PERF_VPC_STALL_CYCLES_HLSQ_PRIM_ALLOC"/> + <value value="5" name="A7XX_PERF_VPC_RESERVED_5"/> + <value value="6" name="A7XX_PERF_VPC_STALL_CYCLES_SP_LM"/> + <value value="7" name="A7XX_PERF_VPC_STARVE_CYCLES_SP"/> + <value value="8" name="A7XX_PERF_VPC_STARVE_CYCLES_LRZ"/> + <value value="9" name="A7XX_PERF_VPC_PC_PRIMITIVES"/> + <value value="10" name="A7XX_PERF_VPC_SP_COMPONENTS"/> + <value value="11" name="A7XX_PERF_VPC_STALL_CYCLES_VPCRAM_POS"/> + <value value="12" name="A7XX_PERF_VPC_LRZ_ASSIGN_PRIMITIVES"/> + <value value="13" name="A7XX_PERF_VPC_RB_VISIBLE_PRIMITIVES"/> + <value value="14" name="A7XX_PERF_VPC_LM_TRANSACTION"/> + <value value="15" name="A7XX_PERF_VPC_STREAMOUT_TRANSACTION"/> + <value value="16" name="A7XX_PERF_VPC_VS_BUSY_CYCLES"/> + <value value="17" name="A7XX_PERF_VPC_PS_BUSY_CYCLES"/> + <value value="18" name="A7XX_PERF_VPC_VS_WORKING_CYCLES"/> + <value value="19" name="A7XX_PERF_VPC_PS_WORKING_CYCLES"/> + <value value="20" name="A7XX_PERF_VPC_STARVE_CYCLES_RB"/> + <value value="21" name="A7XX_PERF_VPC_NUM_VPCRAM_READ_POS"/> + <value value="22" name="A7XX_PERF_VPC_WIT_FULL_CYCLES"/> + <value value="23" name="A7XX_PERF_VPC_VPCRAM_FULL_CYCLES"/> + <value value="24" name="A7XX_PERF_VPC_LM_FULL_WAIT_FOR_INTP_END"/> + <value value="25" name="A7XX_PERF_VPC_NUM_VPCRAM_WRITE"/> + <value value="26" name="A7XX_PERF_VPC_NUM_VPCRAM_READ_SO"/> + <value value="27" name="A7XX_PERF_VPC_NUM_ATTR_REQ_LM"/> + <value value="28" name="A7XX_PERF_VPC_STALL_CYCLE_TSE"/> + <value value="29" name="A7XX_PERF_VPC_TSE_PRIMITIVES"/> + <value value="30" name="A7XX_PERF_VPC_GS_PRIMITIVES"/> + <value value="31" name="A7XX_PERF_VPC_TSE_TRANSACTIONS"/> + <value value="32" name="A7XX_PERF_VPC_STALL_CYCLES_CCU"/> + <value value="33" name="A7XX_PERF_VPC_NUM_WM_HIT"/> + <value value="34" name="A7XX_PERF_VPC_STALL_DQ_WACK"/> + <value value="35" name="A7XX_PERF_VPC_STALL_CYCLES_CCHE"/> + <value value="36" name="A7XX_PERF_VPC_STARVE_CYCLES_CCHE"/> + <value value="37" name="A7XX_PERF_VPC_NUM_PA_REQ"/> + <value value="38" name="A7XX_PERF_VPC_NUM_LM_REQ_HIT"/> + <value value="39" name="A7XX_PERF_VPC_CCHE_REQBUF_FULL"/> + <value value="40" name="A7XX_PERF_VPC_STALL_CYCLES_LM_ACK"/> + <value value="41" name="A7XX_PERF_VPC_STALL_CYCLES_PRG_END_FE"/> + <value value="42" name="A7XX_PERF_VPC_STALL_CYCLES_PRG_END_PCVS"/> + <value value="43" name="A7XX_PERF_VPC_STALL_CYCLES_PRG_END_VPCPS"/> +</enum> + +<enum name="a7xx_tse_perfcounter_select"> + <value value="0" name="A7XX_PERF_TSE_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_TSE_CLIPPING_CYCLES"/> + <value value="2" name="A7XX_PERF_TSE_STALL_CYCLES_RAS"/> + <value value="3" name="A7XX_PERF_TSE_STALL_CYCLES_LRZ_BARYPLANE"/> + <value value="4" name="A7XX_PERF_TSE_STALL_CYCLES_LRZ_ZPLANE"/> + <value value="5" name="A7XX_PERF_TSE_STARVE_CYCLES_PC"/> + <value value="6" name="A7XX_PERF_TSE_INPUT_PRIM"/> + <value value="7" name="A7XX_PERF_TSE_INPUT_NULL_PRIM"/> + <value value="8" name="A7XX_PERF_TSE_TRIVAL_REJ_PRIM"/> + <value value="9" name="A7XX_PERF_TSE_CLIPPED_PRIM"/> + <value value="10" name="A7XX_PERF_TSE_ZERO_AREA_PRIM"/> + <value value="11" name="A7XX_PERF_TSE_FACENESS_CULLED_PRIM"/> + <value value="12" name="A7XX_PERF_TSE_ZERO_PIXEL_PRIM"/> + <value value="13" name="A7XX_PERF_TSE_OUTPUT_NULL_PRIM"/> + <value value="14" name="A7XX_PERF_TSE_OUTPUT_VISIBLE_PRIM"/> + <value value="15" name="A7XX_PERF_TSE_CINVOCATION"/> + <value value="16" name="A7XX_PERF_TSE_CPRIMITIVES"/> + <value value="17" name="A7XX_PERF_TSE_2D_INPUT_PRIM"/> + <value value="18" name="A7XX_PERF_TSE_2D_ALIVE_CYCLES"/> + <value value="19" name="A7XX_PERF_TSE_CLIP_PLANES"/> +</enum> + +<enum name="a7xx_ras_perfcounter_select"> + <value value="0" name="A7XX_PERF_RAS_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_RAS_SUPERTILE_ACTIVE_CYCLES"/> + <value value="2" name="A7XX_PERF_RAS_STALL_CYCLES_LRZ"/> + <value value="3" name="A7XX_PERF_RAS_STARVE_CYCLES_TSE"/> + <value value="4" name="A7XX_PERF_RAS_SUPER_TILES"/> + <value value="5" name="A7XX_PERF_RAS_8X4_TILES"/> + <value value="6" name="A7XX_PERF_RAS_MASKGEN_ACTIVE"/> + <value value="7" name="A7XX_PERF_RAS_FULLY_COVERED_SUPER_TILES"/> + <value value="8" name="A7XX_PERF_RAS_FULLY_COVERED_8X4_TILES"/> + <value value="9" name="A7XX_PERF_RAS_PRIM_KILLED_INVISILBE"/> + <value value="10" name="A7XX_PERF_RAS_SUPERTILE_GEN_ACTIVE_CYCLES"/> + <value value="11" name="A7XX_PERF_RAS_LRZ_INTF_WORKING_CYCLES"/> + <value value="12" name="A7XX_PERF_RAS_BLOCKS"/> + <value value="13" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_0_WORKING_CC_l2"/> + <value value="14" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_1_WORKING_CC_l2"/> + <value value="15" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_2_WORKING_CC_l2"/> + <value value="16" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_3_WORKING_CC_l2"/> + <value value="17" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_4_WORKING_CC_l2"/> + <value value="18" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_5_WORKING_CC_l2"/> + <value value="19" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_6_WORKING_CC_l2"/> + <value value="20" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_7_WORKING_CC_l2"/> + <value value="21" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_8_WORKING_CC_l2"/> + <value value="22" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_9_WORKING_CC_l2"/> + <value value="23" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_10_WORKING_CC_l2"/> + <value value="24" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_11_WORKING_CC_l2"/> + <value value="25" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_12_WORKING_CC_l2"/> + <value value="26" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_13_WORKING_CC_l2"/> + <value value="27" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_14_WORKING_CC_l2"/> + <value value="28" name="A7XX_PERF_RAS_SAMPLE_MASK_GEN_LANE_15_WORKING_CC_l2"/> + <value value="29" name="A7XX_PERF_RAS_FALSE_PARTIAL_STILE"/> + +</enum> + +<enum name="a7xx_uche_perfcounter_select"> + <value value="0" name="A7XX_PERF_UCHE_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_UCHE_STALL_CYCLES_ARBITER"/> + <value value="2" name="A7XX_PERF_UCHE_VBIF_LATENCY_CYCLES"/> + <value value="3" name="A7XX_PERF_UCHE_VBIF_LATENCY_SAMPLES"/> + <value value="4" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_TP"/> + <value value="5" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_VFD"/> + <value value="6" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_HLSQ"/> + <value value="7" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_LRZ"/> + <value value="8" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_SP"/> + <value value="9" name="A7XX_PERF_UCHE_READ_REQUESTS_TP"/> + <value value="10" name="A7XX_PERF_UCHE_READ_REQUESTS_VFD"/> + <value value="11" name="A7XX_PERF_UCHE_READ_REQUESTS_HLSQ"/> + <value value="12" name="A7XX_PERF_UCHE_READ_REQUESTS_LRZ"/> + <value value="13" name="A7XX_PERF_UCHE_READ_REQUESTS_SP"/> + <value value="14" name="A7XX_PERF_UCHE_WRITE_REQUESTS_LRZ"/> + <value value="15" name="A7XX_PERF_UCHE_WRITE_REQUESTS_SP"/> + <value value="16" name="A7XX_PERF_UCHE_WRITE_REQUESTS_VPC"/> + <value value="17" name="A7XX_PERF_UCHE_WRITE_REQUESTS_VSC"/> + <value value="18" name="A7XX_PERF_UCHE_EVICTS"/> + <value value="19" name="A7XX_PERF_UCHE_BANK_REQ0"/> + <value value="20" name="A7XX_PERF_UCHE_BANK_REQ1"/> + <value value="21" name="A7XX_PERF_UCHE_BANK_REQ2"/> + <value value="22" name="A7XX_PERF_UCHE_BANK_REQ3"/> + <value value="23" name="A7XX_PERF_UCHE_BANK_REQ4"/> + <value value="24" name="A7XX_PERF_UCHE_BANK_REQ5"/> + <value value="25" name="A7XX_PERF_UCHE_BANK_REQ6"/> + <value value="26" name="A7XX_PERF_UCHE_BANK_REQ7"/> + <value value="27" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_CH0"/> + <value value="28" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_CH1"/> + <value value="29" name="A7XX_PERF_UCHE_GMEM_READ_BEATS"/> + <value value="30" name="A7XX_PERF_UCHE_TPH_REF_FULL"/> + <value value="31" name="A7XX_PERF_UCHE_TPH_VICTIM_FULL"/> + <value value="32" name="A7XX_PERF_UCHE_TPH_EXT_FULL"/> + <value value="33" name="A7XX_PERF_UCHE_VBIF_STALL_WRITE_DATA"/> + <value value="34" name="A7XX_PERF_UCHE_DCMP_LATENCY_SAMPLES"/> + <value value="35" name="A7XX_PERF_UCHE_DCMP_LATENCY_CYCLES"/> + <value value="36" name="A7XX_PERF_UCHE_VBIF_READ_BEATS_PC"/> + <value value="37" name="A7XX_PERF_UCHE_READ_REQUESTS_PC"/> + <value value="38" name="A7XX_PERF_UCHE_RAM_READ_REQ"/> + <value value="39" name="A7XX_PERF_UCHE_RAM_WRITE_REQ"/> + <value value="40" name="A7XX_PERF_UCHE_STARVED_CYCLES_VBIF_DECMP"/> + <value value="41" name="A7XX_PERF_UCHE_STALL_CYCLES_DECMP"/> + <value value="42" name="A7XX_PERF_UCHE_ARBITER_STALL_CYCLES_VBIF"/> + <value value="43" name="A7XX_PERF_UCHE_READ_REQUESTS_TP_UBWC"/> + <value value="44" name="A7XX_PERF_UCHE_READ_REQUESTS_TP_NONUBWC"/> + <value value="45" name="A7XX_PERF_UCHE_READ_REQUESTS_TP_GMEM"/> + <value value="46" name="A7XX_PERF_UCHE_LONG_LINE_ALL_EVICTS_KAILUA"/> + <value value="47" name="A7XX_PERF_UCHE_LONG_LINE_PARTIAL_EVICTS_KAILUA"/> + <value value="48" name="A7XX_PERF_UCHE_TPH_CONFLICT_CL_CCHE"/> + <value value="49" name="A7XX_PERF_UCHE_TPH_CONFLICT_CL_OTHER_KAILUA"/> + <value value="50" name="A7XX_PERF_UCHE_DBANK_CONFLICT_CL_CCHE"/> + <value value="51" name="A7XX_PERF_UCHE_DBANK_CONFLICT_CL_OTHER_CLIENTS"/> + <value value="52" name="A7XX_PERF_UCHE_VBIF_WRITE_BEATS_CH0"/> + <value value="53" name="A7XX_PERF_UCHE_VBIF_WRITE_BEATS_CH1"/> + <value value="54" name="A7XX_PERF_UCHE_CCHE_TPH_QUEUE_FULL"/> + <value value="55" name="A7XX_PERF_UCHE_CCHE_DPH_QUEUE_FULL"/> + <value value="56" name="A7XX_PERF_UCHE_GMEM_WRITE_BEATS"/> + <value value="57" name="A7XX_PERF_UCHE_UBWC_READ_BEATS"/> + <value value="58" name="A7XX_PERF_UCHE_UBWC_WRITE_BEATS"/> +</enum> + +<enum name="a7xx_tp_perfcounter_select"> + <value value="0" name="A7XX_PERF_TP_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_TP_STALL_CYCLES_UCHE"/> + <value value="2" name="A7XX_PERF_TP_LATENCY_CYCLES"/> + <value value="3" name="A7XX_PERF_TP_LATENCY_TRANS"/> + <value value="4" name="A7XX_PERF_TP_FLAG_FIFO_DELAY_SAMPLES"/> + <value value="5" name="A7XX_PERF_TP_FLAG_FIFO_DELAY_CYCLES"/> + <value value="6" name="A7XX_PERF_TP_L1_CACHELINE_REQUESTS"/> + <value value="7" name="A7XX_PERF_TP_L1_CACHELINE_MISSES"/> + <value value="8" name="A7XX_PERF_TP_SP_TP_TRANS"/> + <value value="9" name="A7XX_PERF_TP_TP_SP_TRANS"/> + <value value="10" name="A7XX_PERF_TP_OUTPUT_PIXELS"/> + <value value="11" name="A7XX_PERF_TP_FILTER_WORKLOAD_16BIT"/> + <value value="12" name="A7XX_PERF_TP_FILTER_WORKLOAD_32BIT"/> + <value value="13" name="A7XX_PERF_TP_QUADS_RECEIVED"/> + <value value="14" name="A7XX_PERF_TP_QUADS_OFFSET"/> + <value value="15" name="A7XX_PERF_TP_QUADS_SHADOW"/> + <value value="16" name="A7XX_PERF_TP_QUADS_ARRAY"/> + <value value="17" name="A7XX_PERF_TP_QUADS_GRADIENT"/> + <value value="18" name="A7XX_PERF_TP_QUADS_1D"/> + <value value="19" name="A7XX_PERF_TP_QUADS_2D"/> + <value value="20" name="A7XX_PERF_TP_QUADS_BUFFER"/> + <value value="21" name="A7XX_PERF_TP_QUADS_3D"/> + <value value="22" name="A7XX_PERF_TP_QUADS_CUBE"/> + <value value="23" name="A7XX_PERF_TP_DIVERGENT_QUADS_RECEIVED"/> + <value value="24" name="A7XX_PERF_TP_PRT_NON_RESIDENT_EVENTS"/> + <value value="25" name="A7XX_PERF_TP_OUTPUT_PIXELS_POINT"/> + <value value="26" name="A7XX_PERF_TP_OUTPUT_PIXELS_BILINEAR"/> + <value value="27" name="A7XX_PERF_TP_OUTPUT_PIXELS_MIP"/> + <value value="28" name="A7XX_PERF_TP_OUTPUT_PIXELS_ANISO"/> + <value value="29" name="A7XX_PERF_TP_OUTPUT_PIXELS_ZERO_LOD"/> + <value value="30" name="A7XX_PERF_TP_FLAG_CACHE_REQUESTS"/> + <value value="31" name="A7XX_PERF_TP_FLAG_CACHE_MISSES"/> + <value value="32" name="A7XX_PERF_TP_L1_5_L2_REQUESTS"/> + <value value="33" name="A7XX_PERF_TP_2D_OUTPUT_PIXELS"/> + <value value="34" name="A7XX_PERF_TP_2D_OUTPUT_PIXELS_POINT"/> + <value value="35" name="A7XX_PERF_TP_2D_OUTPUT_PIXELS_BILINEAR"/> + <value value="36" name="A7XX_PERF_TP_2D_FILTER_WORKLOAD_16BIT"/> + <value value="37" name="A7XX_PERF_TP_2D_FILTER_WORKLOAD_32BIT"/> + <value value="38" name="A7XX_PERF_TP_TPA2TPC_TRANS"/> + <value value="39" name="A7XX_PERF_TP_L1_MISSES_ASTC_1TILE"/> + <value value="40" name="A7XX_PERF_TP_L1_MISSES_ASTC_2TILE"/> + <value value="41" name="A7XX_PERF_TP_L1_MISSES_ASTC_4TILE"/> + <value value="42" name="A7XX_PERF_TP_L1_5_COMPRESS_REQS"/> + <value value="43" name="A7XX_PERF_TP_L1_5_L2_COMPRESS_MISS"/> + <value value="44" name="A7XX_PERF_TP_L1_BANK_CONFLICT"/> + <value value="45" name="A7XX_PERF_TP_L1_5_MISS_LATENCY_CYCLES"/> + <value value="46" name="A7XX_PERF_TP_L1_5_MISS_LATENCY_TRANS"/> + <value value="47" name="A7XX_PERF_TP_QUADS_CONSTANT_MULTIPLIED"/> + <value value="48" name="A7XX_PERF_TP_FRONTEND_WORKING_CYCLES"/> + <value value="49" name="A7XX_PERF_TP_L1_TAG_WORKING_CYCLES"/> + <value value="50" name="A7XX_PERF_TP_L1_DATA_WRITE_WORKING_CYCLES"/> + <value value="51" name="A7XX_PERF_TP_PRE_L1_DECOM_WORKING_CYCLES"/> + <value value="52" name="A7XX_PERF_TP_BACKEND_WORKING_CYCLES"/> + <value value="53" name="A7XX_PERF_TP_L1_5_CACHE_WORKING_CYCLES"/> + <value value="54" name="A7XX_PERF_TP_STARVE_CYCLES_SP"/> + <value value="55" name="A7XX_PERF_TP_STARVE_CYCLES_UCHE"/> + <value value="56" name="A7XX_PERF_TP_STALL_CYCLES_UFC"/> + <value value="57" name="A7XX_PERF_TP_FORMAT_DECOMP"/> + <value value="58" name="A7XX_PERF_TP_FILTER_POINT_FP16"/> + <value value="59" name="A7XX_PERF_TP_FILTER_POINT_FP32"/> + <value value="60" name="A7XX_PERF_TP_LATENCY_FIFO_FULL"/> + <value value="61" name="A7XX_PERF_TP_RESERVED_61"/> + <value value="62" name="A7XX_PERF_TP_RESERVED_62"/> + <value value="63" name="A7XX_PERF_TP_RESERVED_63"/> + <value value="64" name="A7XX_PERF_TP_RESERVED_64"/> + <value value="65" name="A7XX_PERF_TP_RESERVED_65"/> + <value value="66" name="A7XX_PERF_TP_RESERVED_66"/> + <value value="67" name="A7XX_PERF_TP_RESERVED_67"/> + <value value="68" name="A7XX_PERF_TP_RESERVED_68"/> + <value value="69" name="A7XX_PERF_TP_RESERVED_69"/> + <value value="70" name="A7XX_PERF_TP_RESERVED_70"/> + <value value="71" name="A7XX_PERF_TP_RESERVED_71"/> + <value value="72" name="A7XX_PERF_TP_RESERVED_72"/> + <value value="73" name="A7XX_PERF_TP_RESERVED_73"/> + <value value="74" name="A7XX_PERF_TP_RESERVED_74"/> + <value value="75" name="A7XX_PERF_TP_RESERVED_75"/> + <value value="76" name="A7XX_PERF_TP_RESERVED_76"/> + <value value="77" name="A7XX_PERF_TP_RESERVED_77"/> + <value value="78" name="A7XX_PERF_TP_RESERVED_78"/> + <value value="79" name="A7XX_PERF_TP_RESERVED_79"/> + <value value="80" name="A7XX_PERF_TP_RESERVED_80"/> + <value value="81" name="A7XX_PERF_TP_RESERVED_81"/> + <value value="82" name="A7XX_PERF_TP_RESERVED_82"/> + <value value="83" name="A7XX_PERF_TP_RESERVED_83"/> + <value value="84" name="A7XX_PERF_TP_RESERVED_84"/> + <value value="85" name="A7XX_PERF_TP_RESERVED_85"/> + <value value="86" name="A7XX_PERF_TP_RESERVED_86"/> + <value value="87" name="A7XX_PERF_TP_RESERVED_87"/> + <value value="88" name="A7XX_PERF_TP_RESERVED_88"/> + <value value="89" name="A7XX_PERF_TP_RESERVED_89"/> + <value value="90" name="A7XX_PERF_TP_RESERVED_90"/> + <value value="91" name="A7XX_PERF_TP_RESERVED_91"/> + <value value="92" name="A7XX_PERF_TP_RESERVED_92"/> + <value value="93" name="A7XX_PERF_TP_RESERVED_93"/> + <value value="94" name="A7XX_PERF_TP_RESERVED_94"/> + <value value="95" name="A7XX_PERF_TP_RESERVED_95"/> + <value value="96" name="A7XX_PERF_TP_RESERVED_96"/> + <value value="97" name="A7XX_PERF_TP_RESERVED_97"/> + <value value="98" name="A7XX_PERF_TP_RESERVED_98"/> + <value value="99" name="A7XX_PERF_TP_RESERVED_99"/> + <value value="100" name="A7XX_PERF_TP_RESERVED_100"/> + <value value="101" name="A7XX_PERF_TP_RESERVED_101"/> + <value value="102" name="A7XX_PERF_TP_RESERVED_102"/> + <value value="103" name="A7XX_PERF_TP_RESERVED_103"/> + <value value="104" name="A7XX_PERF_TP_RESERVED_104"/> + <value value="105" name="A7XX_PERF_TP_RESERVED_105"/> + <value value="106" name="A7XX_PERF_TP_RESERVED_106"/> + <value value="107" name="A7XX_PERF_TP_RESERVED_107"/> + <value value="108" name="A7XX_PERF_TP_RESERVED_108"/> + <value value="109" name="A7XX_PERF_TP_RESERVED_109"/> + <value value="110" name="A7XX_PERF_TP_RESERVED_110"/> + <value value="111" name="A7XX_PERF_TP_RESERVED_111"/> + <value value="112" name="A7XX_PERF_TP_RESERVED_112"/> + <value value="113" name="A7XX_PERF_TP_RESERVED_113"/> + <value value="114" name="A7XX_PERF_TP_RESERVED_114"/> + <value value="115" name="A7XX_PERF_TP_RESERVED_115"/> + <value value="116" name="A7XX_PERF_TP_RESERVED_116"/> + <value value="117" name="A7XX_PERF_TP_RESERVED_117"/> + <value value="118" name="A7XX_PERF_TP_RESERVED_118"/> + <value value="119" name="A7XX_PERF_TP_RESERVED_119"/> + <value value="120" name="A7XX_PERF_TP_RESERVED_120"/> + <value value="121" name="A7XX_PERF_TP_RESERVED_121"/> + <value value="122" name="A7XX_PERF_TP_RESERVED_122"/> + <value value="123" name="A7XX_PERF_TP_RESERVED_123"/> + <value value="124" name="A7XX_PERF_TP_RESERVED_124"/> + <value value="125" name="A7XX_PERF_TP_RESERVED_125"/> + <value value="126" name="A7XX_PERF_TP_RESERVED_126"/> + <value value="127" name="A7XX_PERF_TP_RESERVED_127"/> + <value value="128" name="A7XX_PERF_TP_FORMAT_DECOMP_BILINEAR"/> + <value value="129" name="A7XX_PERF_TP_PACKED_POINT_BOTH_VALID_FP16"/> + <value value="130" name="A7XX_PERF_TP_PACKED_POINT_SINGLE_VALID_FP16"/> + <value value="131" name="A7XX_PERF_TP_PACKED_POINT_BOTH_VALID_FP32"/> + <value value="132" name="A7XX_PERF_TP_PACKED_POINT_SINGLE_VALID_FP32"/> +</enum> + +<enum name="a7xx_sp_perfcounter_select"> + <value value="0" name="A7XX_PERF_SP_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_SP_ALU_WORKING_CYCLES"/> + <value value="2" name="A7XX_PERF_SP_EFU_WORKING_CYCLES"/> + <value value="3" name="A7XX_PERF_SP_STALL_CYCLES_VPC"/> + <value value="4" name="A7XX_PERF_SP_STALL_CYCLES_TP"/> + <value value="5" name="A7XX_PERF_SP_STALL_CYCLES_UCHE"/> + <value value="6" name="A7XX_PERF_SP_STALL_CYCLES_RB"/> + <value value="7" name="A7XX_PERF_SP_NON_EXECUTION_CYCLES"/> + <value value="8" name="A7XX_PERF_SP_WAVE_CONTEXTS"/> + <value value="9" name="A7XX_PERF_SP_WAVE_CONTEXT_CYCLES"/> + <value value="10" name="A7XX_PERF_SP_STAGE_WAVE_CYCLES"/> + <value value="11" name="A7XX_PERF_SP_STAGE_WAVE_SAMPLES"/> + <value value="12" name="A7XX_PERF_SP_VS_STAGE_WAVE_CYCLES"/> + <value value="13" name="A7XX_PERF_SP_VS_STAGE_WAVE_SAMPLES"/> + <value value="14" name="A7XX_PERF_SP_FS_STAGE_DURATION_CYCLES"/> + <value value="15" name="A7XX_PERF_SP_VS_STAGE_DURATION_CYCLES"/> + <value value="16" name="A7XX_PERF_SP_WAVE_CTRL_CYCLES"/> + <value value="17" name="A7XX_PERF_SP_WAVE_LOAD_CYCLES"/> + <value value="18" name="A7XX_PERF_SP_WAVE_EMIT_CYCLES"/> + <value value="19" name="A7XX_PERF_SP_WAVE_NOP_CYCLES"/> + <value value="20" name="A7XX_PERF_SP_WAVE_WAIT_CYCLES"/> + <value value="21" name="A7XX_PERF_SP_WAVE_FETCH_CYCLES"/> + <value value="22" name="A7XX_PERF_SP_WAVE_IDLE_CYCLES"/> + <value value="23" name="A7XX_PERF_SP_WAVE_END_CYCLES"/> + <value value="24" name="A7XX_PERF_SP_WAVE_LONG_SYNC_CYCLES"/> + <value value="25" name="A7XX_PERF_SP_WAVE_SHORT_SYNC_CYCLES"/> + <value value="26" name="A7XX_PERF_SP_WAVE_JOIN_CYCLES"/> + <value value="27" name="A7XX_PERF_SP_LM_LOAD_INSTRUCTIONS"/> + <value value="28" name="A7XX_PERF_SP_LM_STORE_INSTRUCTIONS"/> + <value value="29" name="A7XX_PERF_SP_LM_ATOMICS"/> + <value value="30" name="A7XX_PERF_SP_GM_LOAD_INSTRUCTIONS"/> + <value value="31" name="A7XX_PERF_SP_GM_STORE_INSTRUCTIONS"/> + <value value="32" name="A7XX_PERF_SP_GM_ATOMICS"/> + <value value="33" name="A7XX_PERF_SP_VS_STAGE_TEX_INSTRUCTIONS"/> + <value value="34" name="A7XX_PERF_SP_VS_STAGE_EFU_INSTRUCTIONS"/> + <value value="35" name="A7XX_PERF_SP_VS_STAGE_FULL_ALU_INSTRUCTIONS"/> + <value value="36" name="A7XX_PERF_SP_VS_STAGE_HALF_ALU_INSTRUCTIONS"/> + <value value="37" name="A7XX_PERF_SP_FS_STAGE_TEX_INSTRUCTIONS"/> + <value value="38" name="A7XX_PERF_SP_FS_STAGE_CFLOW_INSTRUCTIONS"/> + <value value="39" name="A7XX_PERF_SP_FS_STAGE_EFU_INSTRUCTIONS"/> + <value value="40" name="A7XX_PERF_SP_FS_STAGE_FULL_ALU_INSTRUCTIONS"/> + <value value="41" name="A7XX_PERF_SP_FS_STAGE_HALF_ALU_INSTRUCTIONS"/> + <value value="42" name="A7XX_PERF_SP_FS_STAGE_BARY_INSTRUCTIONS"/> + <value value="43" name="A7XX_PERF_SP_VS_INSTRUCTIONS"/> + <value value="44" name="A7XX_PERF_SP_FS_INSTRUCTIONS"/> + <value value="45" name="A7XX_PERF_SP_ADDR_LOCK_COUNT"/> + <value value="46" name="A7XX_PERF_SP_UCHE_READ_TRANS"/> + <value value="47" name="A7XX_PERF_SP_UCHE_WRITE_TRANS"/> + <value value="48" name="A7XX_PERF_SP_EXPORT_VPC_TRANS"/> + <value value="49" name="A7XX_PERF_SP_EXPORT_RB_TRANS"/> + <value value="50" name="A7XX_PERF_SP_PIXELS_KILLED"/> + <value value="51" name="A7XX_PERF_SP_ICL1_REQUESTS"/> + <value value="52" name="A7XX_PERF_SP_ICL1_MISSES"/> + <value value="53" name="A7XX_PERF_SP_HS_INSTRUCTIONS"/> + <value value="54" name="A7XX_PERF_SP_DS_INSTRUCTIONS"/> + <value value="55" name="A7XX_PERF_SP_GS_INSTRUCTIONS"/> + <value value="56" name="A7XX_PERF_SP_CS_INSTRUCTIONS"/> + <value value="57" name="A7XX_PERF_SP_GPR_READ"/> + <value value="58" name="A7XX_PERF_SP_GPR_WRITE"/> + <value value="59" name="A7XX_PERF_SP_FS_STAGE_HALF_EFU_INSTRUCTIONS"/> + <value value="60" name="A7XX_PERF_SP_VS_STAGE_HALF_EFU_INSTRUCTIONS"/> + <value value="61" name="A7XX_PERF_SP_LM_BANK_CONFLICTS"/> + <value value="62" name="A7XX_PERF_SP_TEX_CONTROL_WORKING_CYCLES"/> + <value value="63" name="A7XX_PERF_SP_LOAD_CONTROL_WORKING_CYCLES"/> + <value value="64" name="A7XX_PERF_SP_FLOW_CONTROL_WORKING_CYCLES"/> + <value value="65" name="A7XX_PERF_SP_LM_WORKING_CYCLES"/> + <value value="66" name="A7XX_PERF_SP_DISPATCHER_WORKING_CYCLES"/> + <value value="67" name="A7XX_PERF_SP_SEQUENCER_WORKING_CYCLES"/> + <value value="68" name="A7XX_PERF_SP_LOW_EFFICIENCY_STARVED_BY_TP"/> + <value value="69" name="A7XX_PERF_SP_STARVE_CYCLES_HLSQ"/> + <value value="70" name="A7XX_PERF_SP_NON_EXECUTION_LS_CYCLES"/> + <value value="71" name="A7XX_PERF_SP_WORKING_EU"/> + <value value="72" name="A7XX_PERF_SP_ANY_EU_WORKING"/> + <value value="73" name="A7XX_PERF_SP_WORKING_EU_FS_STAGE"/> + <value value="74" name="A7XX_PERF_SP_ANY_EU_WORKING_FS_STAGE"/> + <value value="75" name="A7XX_PERF_SP_WORKING_EU_VS_STAGE"/> + <value value="76" name="A7XX_PERF_SP_ANY_EU_WORKING_VS_STAGE"/> + <value value="77" name="A7XX_PERF_SP_WORKING_EU_CS_STAGE"/> + <value value="78" name="A7XX_PERF_SP_ANY_EU_WORKING_CS_STAGE"/> + <value value="79" name="A7XX_PERF_SP_GPR_READ_PREFETCH"/> + <value value="80" name="A7XX_PERF_SP_GPR_READ_CONFLICT"/> + <value value="81" name="A7XX_PERF_SP_GPR_WRITE_CONFLICT"/> + <value value="82" name="A7XX_PERF_SP_GM_LOAD_LATENCY_CYCLES"/> + <value value="83" name="A7XX_PERF_SP_GM_LOAD_LATENCY_SAMPLES"/> + <value value="84" name="A7XX_PERF_SP_EXECUTABLE_WAVES"/> + <value value="85" name="A7XX_PERF_SP_ICL1_MISS_FETCH_CYCLES"/> + <value value="86" name="A7XX_PERF_SP_WORKING_EU_LPAC"/> + <value value="87" name="A7XX_PERF_SP_BYPASS_BUSY_CYCLES"/> + <value value="88" name="A7XX_PERF_SP_ANY_EU_WORKING_LPAC"/> + <value value="89" name="A7XX_PERF_SP_WAVE_ALU_CYCLES"/> + <value value="90" name="A7XX_PERF_SP_WAVE_EFU_CYCLES"/> + <value value="91" name="A7XX_PERF_SP_WAVE_INT_CYCLES"/> + <value value="92" name="A7XX_PERF_SP_WAVE_CSP_CYCLES"/> + <value value="93" name="A7XX_PERF_SP_EWAVE_CONTEXTS"/> + <value value="94" name="A7XX_PERF_SP_EWAVE_CONTEXT_CYCLES"/> + <value value="95" name="A7XX_PERF_SP_LPAC_BUSY_CYCLES"/> + <value value="96" name="A7XX_PERF_SP_LPAC_INSTRUCTIONS"/> + <value value="97" name="A7XX_PERF_SP_FS_STAGE_1X_WAVES"/> + <value value="98" name="A7XX_PERF_SP_FS_STAGE_2X_WAVES"/> + <value value="99" name="A7XX_PERF_SP_QUADS"/> + <value value="100" name="A7XX_PERF_SP_CS_INVOCATIONS"/> + <value value="101" name="A7XX_PERF_SP_PIXELS"/> + <value value="102" name="A7XX_PERF_SP_LPAC_DRAWCALLS"/> + <value value="103" name="A7XX_PERF_SP_PI_WORKING_CYCLES"/> + <value value="104" name="A7XX_PERF_SP_WAVE_INPUT_CYCLES"/> + <value value="105" name="A7XX_PERF_SP_WAVE_OUTPUT_CYCLES"/> + <value value="106" name="A7XX_PERF_SP_WAVE_HWAVE_WAIT_CYCLES"/> + <value value="107" name="A7XX_PERF_SP_WAVE_HWAVE_SYNC"/> + <value value="108" name="A7XX_PERF_SP_OUTPUT_3D_PIXELS"/> + <value value="109" name="A7XX_PERF_SP_FULL_ALU_MAD_INSTRUCTIONS"/> + <value value="110" name="A7XX_PERF_SP_HALF_ALU_MAD_INSTRUCTIONS"/> + <value value="111" name="A7XX_PERF_SP_FULL_ALU_MUL_INSTRUCTIONS"/> + <value value="112" name="A7XX_PERF_SP_HALF_ALU_MUL_INSTRUCTIONS"/> + <value value="113" name="A7XX_PERF_SP_FULL_ALU_ADD_INSTRUCTIONS"/> + <value value="114" name="A7XX_PERF_SP_HALF_ALU_ADD_INSTRUCTIONS"/> + <value value="115" name="A7XX_PERF_SP_BARY_FP32_INSTRUCTIONS"/> + <value value="116" name="A7XX_PERF_SP_ALU_GPR_READ_CYCLES"/> + <value value="117" name="A7XX_PERF_SP_ALU_DATA_FORWARDING_CYCLES"/> + <value value="118" name="A7XX_PERF_SP_LM_FULL_CYCLES"/> + <value value="119" name="A7XX_PERF_SP_TEXTURE_FETCH_LATENCY_CYCLES"/> + <value value="120" name="A7XX_PERF_SP_TEXTURE_FETCH_LATENCY_SAMPLES"/> + <value value="121" name="A7XX_PERF_SP_FS_STAGE_PI_TEX_INSTRUCTION"/> + <value value="122" name="A7XX_PERF_SP_RAY_QUERY_INSTRUCTIONS"/> + <value value="123" name="A7XX_PERF_SP_RBRT_KICKOFF_FIBERS"/> + <value value="124" name="A7XX_PERF_SP_RBRT_KICKOFF_DQUADS"/> + <value value="125" name="A7XX_PERF_SP_RTU_BUSY_CYCLES"/> + <value value="126" name="A7XX_PERF_SP_RTU_L0_HITS"/> + <value value="127" name="A7XX_PERF_SP_RTU_L0_MISSES"/> + <value value="128" name="A7XX_PERF_SP_RTU_L0_HIT_ON_MISS"/> + <value value="129" name="A7XX_PERF_SP_RTU_STALL_CYCLES_WAVE_QUEUE"/> + <value value="130" name="A7XX_PERF_SP_RTU_STALL_CYCLES_L0_HIT_QUEUE"/> + <value value="131" name="A7XX_PERF_SP_RTU_STALL_CYCLES_L0_MISS_QUEUE"/> + <value value="132" name="A7XX_PERF_SP_RTU_STALL_CYCLES_L0D_IDX_QUEUE"/> + <value value="133" name="A7XX_PERF_SP_RTU_STALL_CYCLES_L0DATA"/> + <value value="134" name="A7XX_PERF_SP_RTU_STALL_CYCLES_REPLACE_CNT"/> + <value value="135" name="A7XX_PERF_SP_RTU_STALL_CYCLES_MRG_CNT"/> + <value value="136" name="A7XX_PERF_SP_RTU_STALL_CYCLES_UCHE"/> + <value value="137" name="A7XX_PERF_SP_RTU_OPERAND_FETCH_STALL_CYCLES_L0"/> + <value value="138" name="A7XX_PERF_SP_RTU_OPERAND_FETCH_STALL_CYCLES_INS_FIFO"/> + <value value="139" name="A7XX_PERF_SP_RTU_BVH_FETCH_LATENCY_CYCLES"/> + <value value="140" name="A7XX_PERF_SP_RTU_BVH_FETCH_LATENCY_SAMPLES"/> + <value value="141" name="A7XX_PERF_SP_STCHE_MISS_INC_VS"/> + <value value="142" name="A7XX_PERF_SP_STCHE_MISS_INC_FS"/> + <value value="143" name="A7XX_PERF_SP_STCHE_MISS_INC_BV"/> + <value value="144" name="A7XX_PERF_SP_STCHE_MISS_INC_LPAC"/> + <value value="145" name="A7XX_PERF_SP_VGPR_ACTIVE_CONTEXTS"/> + <value value="146" name="A7XX_PERF_SP_PGPR_ALLOC_CONTEXTS"/> + <value value="147" name="A7XX_PERF_SP_VGPR_ALLOC_CONTEXTS"/> + <value value="148" name="A7XX_PERF_SP_RTU_RAY_BOX_INTERSECTIONS"/> + <value value="149" name="A7XX_PERF_SP_RTU_RAY_TRIANGLE_INTERSECTIONS"/> + <value value="150" name="A7XX_PERF_SP_SCH_STALL_CYCLES_RTU"/> +</enum> + +<enum name="a7xx_rb_perfcounter_select"> + <value value="0" name="A7XX_PERF_RB_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_RB_STALL_CYCLES_HLSQ"/> + <value value="2" name="A7XX_PERF_RB_STALL_CYCLES_FIFO0_FULL"/> + <value value="3" name="A7XX_PERF_RB_STALL_CYCLES_FIFO1_FULL"/> + <value value="4" name="A7XX_PERF_RB_STALL_CYCLES_FIFO2_FULL"/> + <value value="5" name="A7XX_PERF_RB_STARVE_CYCLES_SP"/> + <value value="6" name="A7XX_PERF_RB_STARVE_CYCLES_LRZ_TILE"/> + <value value="7" name="A7XX_PERF_RB_STARVE_CYCLES_CCU"/> + <value value="8" name="A7XX_PERF_RB_STARVE_CYCLES_Z_PLANE"/> + <value value="9" name="A7XX_PERF_RB_STARVE_CYCLES_BARY_PLANE"/> + <value value="10" name="A7XX_PERF_RB_Z_WORKLOAD"/> + <value value="11" name="A7XX_PERF_RB_HLSQ_ACTIVE"/> + <value value="12" name="A7XX_PERF_RB_Z_READ"/> + <value value="13" name="A7XX_PERF_RB_Z_WRITE"/> + <value value="14" name="A7XX_PERF_RB_C_READ"/> + <value value="15" name="A7XX_PERF_RB_C_WRITE"/> + <value value="16" name="A7XX_PERF_RB_TOTAL_PASS"/> + <value value="17" name="A7XX_PERF_RB_Z_PASS"/> + <value value="18" name="A7XX_PERF_RB_Z_FAIL"/> + <value value="19" name="A7XX_PERF_RB_S_FAIL"/> + <value value="20" name="A7XX_PERF_RB_BLENDED_FXP_COMPONENTS"/> + <value value="21" name="A7XX_PERF_RB_BLENDED_FP16_COMPONENTS"/> + <value value="22" name="A7XX_PERF_RB_PS_INVOCATIONS"/> + <value value="23" name="A7XX_PERF_RB_2D_ALIVE_CYCLES"/> + <value value="24" name="A7XX_PERF_RB_2D_STALL_CYCLES_A2D"/> + <value value="25" name="A7XX_PERF_RB_2D_STARVE_CYCLES_SRC"/> + <value value="26" name="A7XX_PERF_RB_2D_STARVE_CYCLES_SP"/> + <value value="27" name="A7XX_PERF_RB_2D_STARVE_CYCLES_DST"/> + <value value="28" name="A7XX_PERF_RB_2D_VALID_PIXELS"/> + <value value="29" name="A7XX_PERF_RB_3D_PIXELS"/> + <value value="30" name="A7XX_PERF_RB_BLENDER_WORKING_CYCLES"/> + <value value="31" name="A7XX_PERF_RB_ZPROC_WORKING_CYCLES"/> + <value value="32" name="A7XX_PERF_RB_CPROC_WORKING_CYCLES"/> + <value value="33" name="A7XX_PERF_RB_SAMPLER_WORKING_CYCLES"/> + <value value="34" name="A7XX_PERF_RB_STALL_CYCLES_CCU_COLOR_READ"/> + <value value="35" name="A7XX_PERF_RB_STALL_CYCLES_CCU_COLOR_WRITE"/> + <value value="36" name="A7XX_PERF_RB_STALL_CYCLES_CCU_DEPTH_READ"/> + <value value="37" name="A7XX_PERF_RB_STALL_CYCLES_CCU_DEPTH_WRITE"/> + <value value="38" name="A7XX_PERF_RB_STALL_CYCLES_VPC"/> + <value value="39" name="A7XX_PERF_RB_2D_INPUT_TRANS"/> + <value value="40" name="A7XX_PERF_RB_2D_OUTPUT_RB_DST_TRANS"/> + <value value="41" name="A7XX_PERF_RB_2D_OUTPUT_RB_SRC_TRANS"/> + <value value="42" name="A7XX_PERF_RB_BLENDED_FP32_COMPONENTS"/> + <value value="43" name="A7XX_PERF_RB_COLOR_PIX_TILES"/> + <value value="44" name="A7XX_PERF_RB_STALL_CYCLES_CCU"/> + <value value="45" name="A7XX_PERF_RB_EARLY_Z_ARB3_GRANT"/> + <value value="46" name="A7XX_PERF_RB_LATE_Z_ARB3_GRANT"/> + <value value="47" name="A7XX_PERF_RB_EARLY_Z_SKIP_GRANT"/> + <value value="48" name="A7XX_PERF_RB_VRS_1x1_QUADS"/> + <value value="49" name="A7XX_PERF_RB_VRS_2x1_QUADS"/> + <value value="50" name="A7XX_PERF_RB_VRS_1x2_QUADS"/> + <value value="51" name="A7XX_PERF_RB_VRS_2x2_QUADS"/> + <value value="52" name="A7XX_PERF_RB_VRS_4x2_QUADS"/> + <value value="53" name="A7XX_PERF_RB_VRS_4x4_QUADS"/> +</enum> + +<enum name="a7xx_vsc_perfcounter_select"> + <value value="0" name="A7XX_PERF_VSC_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_VSC_WORKING_CYCLES"/> + <value value="2" name="A7XX_PERF_VSC_STALL_CYCLES_UCHE"/> + <value value="3" name="A7XX_PERF_VSC_EOT_NUM"/> + <value value="4" name="A7XX_PERF_VSC_INPUT_TILES"/> +</enum> + +<enum name="a7xx_ccu_perfcounter_select"> + <value value="0" name="A7XX_PERF_CCU_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_CCU_STALL_CYCLES_RB_DEPTH_RETURN"/> + <value value="2" name="A7XX_PERF_CCU_STALL_CYCLES_RB_COLOR_RETURN"/> + <value value="3" name="A7XX_PERF_CCU_DEPTH_BLOCKS"/> + <value value="4" name="A7XX_PERF_CCU_COLOR_BLOCKS"/> + <value value="5" name="A7XX_PERF_CCU_DEPTH_BLOCK_HIT"/> + <value value="6" name="A7XX_PERF_CCU_COLOR_BLOCK_HIT"/> + <value value="7" name="A7XX_PERF_CCU_PARTIAL_BLOCK_READ"/> + <value value="8" name="A7XX_PERF_CCU_GMEM_READ"/> + <value value="9" name="A7XX_PERF_CCU_GMEM_WRITE"/> + <value value="10" name="A7XX_PERF_CCU_2D_RD_REQ"/> + <value value="11" name="A7XX_PERF_CCU_2D_WR_REQ"/> + <value value="12" name="A7XX_PERF_CCU_UBWC_COLOR_BLOCKS_CONCURRENT"/> + <value value="13" name="A7XX_PERF_CCU_UBWC_DEPTH_BLOCKS_CONCURRENT"/> + <value value="14" name="A7XX_PERF_CCU_COLOR_RESOLVE_DROPPED"/> + <value value="15" name="A7XX_PERF_CCU_DEPTH_RESOLVE_DROPPED"/> + <value value="16" name="A7XX_PERF_CCU_COLOR_RENDER_CONCURRENT"/> + <value value="17" name="A7XX_PERF_CCU_DEPTH_RENDER_CONCURRENT"/> + <value value="18" name="A7XX_PERF_CCU_COLOR_RESOLVE_AFTER_RENDER"/> + <value value="19" name="A7XX_PERF_CCU_DEPTH_RESOLVE_AFTER_RENDER"/> + <value value="20" name="A7XX_PERF_CCU_GMEM_EXTRA_DEPTH_READ"/> + <value value="21" name="A7XX_PERF_CCU_GMEM_COLOR_READ_4AA"/> + <value value="22" name="A7XX_PERF_CCU_GMEM_COLOR_READ_4AA_FULL"/> +</enum> + +<enum name="a7xx_lrz_perfcounter_select"> + <value value="0" name="A7XX_PERF_LRZ_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_LRZ_STARVE_CYCLES_RAS"/> + <value value="2" name="A7XX_PERF_LRZ_STALL_CYCLES_RB"/> + <value value="3" name="A7XX_PERF_LRZ_STALL_CYCLES_VSC"/> + <value value="4" name="A7XX_PERF_LRZ_STALL_CYCLES_VPC"/> + <value value="5" name="A7XX_PERF_LRZ_STALL_CYCLES_FLAG_PREFETCH"/> + <value value="6" name="A7XX_PERF_LRZ_STALL_CYCLES_UCHE"/> + <value value="7" name="A7XX_PERF_LRZ_LRZ_READ"/> + <value value="8" name="A7XX_PERF_LRZ_LRZ_WRITE"/> + <value value="9" name="A7XX_PERF_LRZ_READ_LATENCY"/> + <value value="10" name="A7XX_PERF_LRZ_MERGE_CACHE_UPDATING"/> + <value value="11" name="A7XX_PERF_LRZ_PRIM_KILLED_BY_MASKGEN"/> + <value value="12" name="A7XX_PERF_LRZ_PRIM_KILLED_BY_LRZ"/> + <value value="13" name="A7XX_PERF_LRZ_VISIBLE_PRIM_AFTER_LRZ"/> + <value value="14" name="A7XX_PERF_LRZ_FULL_8X8_TILES"/> + <value value="15" name="A7XX_PERF_LRZ_PARTIAL_8X8_TILES"/> + <value value="16" name="A7XX_PERF_LRZ_TILE_KILLED"/> + <value value="17" name="A7XX_PERF_LRZ_TOTAL_PIXEL"/> + <value value="18" name="A7XX_PERF_LRZ_VISIBLE_PIXEL_AFTER_LRZ"/> + <value value="19" name="A7XX_PERF_LRZ_FEEDBACK_ACCEPT"/> + <value value="20" name="A7XX_PERF_LRZ_FEEDBACK_DISCARD"/> + <value value="21" name="A7XX_PERF_LRZ_FEEDBACK_STALL"/> + <value value="22" name="A7XX_PERF_LRZ_STALL_CYCLES_RB_ZPLANE"/> + <value value="23" name="A7XX_PERF_LRZ_STALL_CYCLES_RB_BPLANE"/> + <value value="24" name="A7XX_PERF_LRZ_RAS_MASK_TRANS"/> + <value value="25" name="A7XX_PERF_LRZ_STALL_CYCLES_MVC"/> + <value value="26" name="A7XX_PERF_LRZ_TILE_KILLED_BY_IMAGE_VRS"/> + <value value="27" name="A7XX_PERF_LRZ_TILE_KILLED_BY_Z"/> +</enum> + +<enum name="a7xx_cmp_perfcounter_select"> + <value value="0" name="A7XX_PERF_CMPDECMP_STALL_CYCLES_ARB"/> + <value value="1" name="A7XX_PERF_CMPDECMP_VBIF_LATENCY_CYCLES"/> + <value value="2" name="A7XX_PERF_CMPDECMP_VBIF_LATENCY_SAMPLES"/> + <value value="3" name="A7XX_PERF_CMPDECMP_VBIF_READ_DATA_CCU"/> + <value value="4" name="A7XX_PERF_CMPDECMP_VBIF_WRITE_DATA_CCU"/> + <value value="5" name="A7XX_PERF_CMPDECMP_VBIF_READ_REQUEST"/> + <value value="6" name="A7XX_PERF_CMPDECMP_VBIF_WRITE_REQUEST"/> + <value value="7" name="A7XX_PERF_CMPDECMP_VBIF_READ_DATA"/> + <value value="8" name="A7XX_PERF_CMPDECMP_VBIF_WRITE_DATA"/> + <value value="9" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG1_COUNT"/> + <value value="10" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG2_COUNT"/> + <value value="11" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG3_COUNT"/> + <value value="12" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG4_COUNT"/> + <value value="13" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG5_COUNT"/> + <value value="14" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG6_COUNT"/> + <value value="15" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG8_COUNT"/> + <value value="16" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG1_COUNT"/> + <value value="17" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG2_COUNT"/> + <value value="18" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG3_COUNT"/> + <value value="19" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG4_COUNT"/> + <value value="20" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG5_COUNT"/> + <value value="21" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG6_COUNT"/> + <value value="22" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG8_COUNT"/> + <value value="23" name="A7XX_PERF_CMPDECMP_VBIF_READ_DATA_UCHE_CH0"/> + <value value="24" name="A7XX_PERF_CMPDECMP_VBIF_READ_DATA_UCHE_CH1"/> + <value value="25" name="A7XX_PERF_CMPDECMP_VBIF_WRITE_DATA_UCHE"/> + <value value="26" name="A7XX_PERF_CMPDECMP_DEPTH_WRITE_FLAG0_COUNT"/> + <value value="27" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAG0_COUNT"/> + <value value="28" name="A7XX_PERF_CMPDECMP_COLOR_WRITE_FLAGALPHA_COUNT"/> + <value value="29" name="A7XX_PERF_CMPDECMP_RESOLVE_EVENTS"/> + <value value="30" name="A7XX_PERF_CMPDECMP_CONCURRENT_RESOLVE_EVENTS"/> + <value value="31" name="A7XX_PERF_CMPDECMP_DROPPED_CLEAR_EVENTS"/> + <value value="32" name="A7XX_PERF_CMPDECMP_ST_BLOCKS_CONCURRENT"/> + <value value="33" name="A7XX_PERF_CMPDECMP_LRZ_ST_BLOCKS_CONCURRENT"/> + <value value="34" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG0_COUNT"/> + <value value="35" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG1_COUNT"/> + <value value="36" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG2_COUNT"/> + <value value="37" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG3_COUNT"/> + <value value="38" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG4_COUNT"/> + <value value="39" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG5_COUNT"/> + <value value="40" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG6_COUNT"/> + <value value="41" name="A7XX_PERF_CMPDECMP_DEPTH_READ_FLAG8_COUNT"/> + <value value="42" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG0_COUNT"/> + <value value="43" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG1_COUNT"/> + <value value="44" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG2_COUNT"/> + <value value="45" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG3_COUNT"/> + <value value="46" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG4_COUNT"/> + <value value="47" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG5_COUNT"/> + <value value="48" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG6_COUNT"/> + <value value="49" name="A7XX_PERF_CMPDECMP_COLOR_READ_FLAG8_COUNT"/> +</enum> + +<enum name="a7xx_gbif_perfcounter_select"> + <value value="0" name="A7XX_PERF_GBIF_RESERVED_0"/> + <value value="1" name="A7XX_PERF_GBIF_RESERVED_1"/> + <value value="2" name="A7XX_PERF_GBIF_RESERVED_2"/> + <value value="3" name="A7XX_PERF_GBIF_RESERVED_3"/> + <value value="4" name="A7XX_PERF_GBIF_RESERVED_4"/> + <value value="5" name="A7XX_PERF_GBIF_RESERVED_5"/> + <value value="6" name="A7XX_PERF_GBIF_RESERVED_6"/> + <value value="7" name="A7XX_PERF_GBIF_RESERVED_7"/> + <value value="8" name="A7XX_PERF_GBIF_RESERVED_8"/> + <value value="9" name="A7XX_PERF_GBIF_RESERVED_9"/> + <value value="10" name="A7XX_PERF_GBIF_AXI0_READ_REQUESTS_TOTAL"/> + <value value="11" name="A7XX_PERF_GBIF_AXI1_READ_REQUESTS_TOTAL"/> + <value value="12" name="A7XX_PERF_GBIF_RESERVED_12"/> + <value value="13" name="A7XX_PERF_GBIF_RESERVED_13"/> + <value value="14" name="A7XX_PERF_GBIF_RESERVED_14"/> + <value value="15" name="A7XX_PERF_GBIF_RESERVED_15"/> + <value value="16" name="A7XX_PERF_GBIF_RESERVED_16"/> + <value value="17" name="A7XX_PERF_GBIF_RESERVED_17"/> + <value value="18" name="A7XX_PERF_GBIF_RESERVED_18"/> + <value value="19" name="A7XX_PERF_GBIF_RESERVED_19"/> + <value value="20" name="A7XX_PERF_GBIF_RESERVED_20"/> + <value value="21" name="A7XX_PERF_GBIF_RESERVED_21"/> + <value value="22" name="A7XX_PERF_GBIF_AXI0_WRITE_REQUESTS_TOTAL"/> + <value value="23" name="A7XX_PERF_GBIF_AXI1_WRITE_REQUESTS_TOTAL"/> + <value value="24" name="A7XX_PERF_GBIF_RESERVED_24"/> + <value value="25" name="A7XX_PERF_GBIF_RESERVED_25"/> + <value value="26" name="A7XX_PERF_GBIF_RESERVED_26"/> + <value value="27" name="A7XX_PERF_GBIF_RESERVED_27"/> + <value value="28" name="A7XX_PERF_GBIF_RESERVED_28"/> + <value value="29" name="A7XX_PERF_GBIF_RESERVED_29"/> + <value value="30" name="A7XX_PERF_GBIF_RESERVED_30"/> + <value value="31" name="A7XX_PERF_GBIF_RESERVED_31"/> + <value value="32" name="A7XX_PERF_GBIF_RESERVED_32"/> + <value value="33" name="A7XX_PERF_GBIF_RESERVED_33"/> + <value value="34" name="A7XX_PERF_GBIF_AXI0_READ_DATA_BEATS_TOTAL"/> + <value value="35" name="A7XX_PERF_GBIF_AXI1_READ_DATA_BEATS_TOTAL"/> + <value value="36" name="A7XX_PERF_GBIF_RESERVED_36"/> + <value value="37" name="A7XX_PERF_GBIF_RESERVED_37"/> + <value value="38" name="A7XX_PERF_GBIF_RESERVED_38"/> + <value value="39" name="A7XX_PERF_GBIF_RESERVED_39"/> + <value value="40" name="A7XX_PERF_GBIF_RESERVED_40"/> + <value value="41" name="A7XX_PERF_GBIF_RESERVED_41"/> + <value value="42" name="A7XX_PERF_GBIF_RESERVED_42"/> + <value value="43" name="A7XX_PERF_GBIF_RESERVED_43"/> + <value value="44" name="A7XX_PERF_GBIF_RESERVED_44"/> + <value value="45" name="A7XX_PERF_GBIF_RESERVED_45"/> + <value value="46" name="A7XX_PERF_GBIF_AXI0_WRITE_DATA_BEATS_TOTAL"/> + <value value="47" name="A7XX_PERF_GBIF_AXI1_WRITE_DATA_BEATS_TOTAL"/> + <value value="48" name="A7XX_PERF_GBIF_RESERVED_48"/> + <value value="49" name="A7XX_PERF_GBIF_RESERVED_49"/> + <value value="50" name="A7XX_PERF_GBIF_RESERVED_50"/> + <value value="51" name="A7XX_PERF_GBIF_RESERVED_51"/> + <value value="52" name="A7XX_PERF_GBIF_RESERVED_52"/> + <value value="53" name="A7XX_PERF_GBIF_RESERVED_53"/> + <value value="54" name="A7XX_PERF_GBIF_RESERVED_54"/> + <value value="55" name="A7XX_PERF_GBIF_RESERVED_55"/> + <value value="56" name="A7XX_PERF_GBIF_RESERVED_56"/> + <value value="57" name="A7XX_PERF_GBIF_RESERVED_57"/> + <value value="58" name="A7XX_PERF_GBIF_RESERVED_58"/> + <value value="59" name="A7XX_PERF_GBIF_RESERVED_59"/> + <value value="60" name="A7XX_PERF_GBIF_RESERVED_60"/> + <value value="61" name="A7XX_PERF_GBIF_RESERVED_61"/> + <value value="62" name="A7XX_PERF_GBIF_RESERVED_62"/> + <value value="63" name="A7XX_PERF_GBIF_RESERVED_63"/> + <value value="64" name="A7XX_PERF_GBIF_RESERVED_64"/> + <value value="65" name="A7XX_PERF_GBIF_RESERVED_65"/> + <value value="66" name="A7XX_PERF_GBIF_RESERVED_66"/> + <value value="67" name="A7XX_PERF_GBIF_RESERVED_67"/> + <value value="68" name="A7XX_PERF_GBIF_CYCLES_CH0_HELD_OFF_RD_ALL"/> + <value value="69" name="A7XX_PERF_GBIF_CYCLES_CH1_HELD_OFF_RD_ALL"/> + <value value="70" name="A7XX_PERF_GBIF_CYCLES_CH0_HELD_OFF_WR_ALL"/> + <value value="71" name="A7XX_PERF_GBIF_CYCLES_CH1_HELD_OFF_WR_ALL"/> + <value value="72" name="A7XX_PERF_GBIF_AXI_CH0_REQUEST_HELD_OFF"/> + <value value="73" name="A7XX_PERF_GBIF_AXI_CH1_REQUEST_HELD_OFF"/> + <value value="74" name="A7XX_PERF_GBIF_AXI_REQUEST_HELD_OFF"/> + <value value="75" name="A7XX_PERF_GBIF_AXI_CH0_WRITE_DATA_HELD_OFF"/> + <value value="76" name="A7XX_PERF_GBIF_AXI_CH1_WRITE_DATA_HELD_OFF"/> + <value value="77" name="A7XX_PERF_GBIF_AXI_ALL_WRITE_DATA_HELD_OFF"/> + <value value="78" name="A7XX_PERF_GBIF_AXI_ALL_READ_BEATS"/> + <value value="79" name="A7XX_PERF_GBIF_AXI_ALL_WRITE_BEATS"/> + <value value="80" name="A7XX_PERF_GBIF_AXI_ALL_BEATS"/> +</enum> + +<enum name="a7xx_ufc_perfcounter_select"> + <value value="0" name="A7XX_PERF_UFC_BUSY_CYCLES"/> + <value value="1" name="A7XX_PERF_UFC_READ_DATA_VBIF"/> + <value value="2" name="A7XX_PERF_UFC_WRITE_DATA_VBIF"/> + <value value="3" name="A7XX_PERF_UFC_READ_REQUEST_VBIF"/> + <value value="4" name="A7XX_PERF_UFC_WRITE_REQUEST_VBIF"/> + <value value="5" name="A7XX_PERF_UFC_LRZ_FILTER_HIT"/> + <value value="6" name="A7XX_PERF_UFC_LRZ_FILTER_MISS"/> + <value value="7" name="A7XX_PERF_UFC_CRE_FILTER_HIT"/> + <value value="8" name="A7XX_PERF_UFC_CRE_FILTER_MISS"/> + <value value="9" name="A7XX_PERF_UFC_SP_FILTER_HIT"/> + <value value="10" name="A7XX_PERF_UFC_SP_FILTER_MISS"/> + <value value="11" name="A7XX_PERF_UFC_SP_REQUESTS"/> + <value value="12" name="A7XX_PERF_UFC_TP_FILTER_HIT"/> + <value value="13" name="A7XX_PERF_UFC_TP_FILTER_MISS"/> + <value value="14" name="A7XX_PERF_UFC_TP_REQUESTS"/> + <value value="15" name="A7XX_PERF_UFC_MAIN_HIT_LRZ_PREFETCH"/> + <value value="16" name="A7XX_PERF_UFC_MAIN_HIT_CRE_PREFETCH"/> + <value value="17" name="A7XX_PERF_UFC_MAIN_HIT_SP_PREFETCH"/> + <value value="18" name="A7XX_PERF_UFC_MAIN_HIT_TP_PREFETCH"/> + <value value="19" name="A7XX_PERF_UFC_MAIN_HIT_UBWC_READ"/> + <value value="20" name="A7XX_PERF_UFC_MAIN_HIT_UBWC_WRITE"/> + <value value="21" name="A7XX_PERF_UFC_MAIN_MISS_LRZ_PREFETCH"/> + <value value="22" name="A7XX_PERF_UFC_MAIN_MISS_CRE_PREFETCH"/> + <value value="23" name="A7XX_PERF_UFC_MAIN_MISS_SP_PREFETCH"/> + <value value="24" name="A7XX_PERF_UFC_MAIN_MISS_TP_PREFETCH"/> + <value value="25" name="A7XX_PERF_UFC_MAIN_MISS_UBWC_READ"/> + <value value="26" name="A7XX_PERF_UFC_MAIN_MISS_UBWC_WRITE"/> + <value value="27" name="A7XX_PERF_UFC_UBWC_READ_UFC_TRANS"/> + <value value="28" name="A7XX_PERF_UFC_UBWC_WRITE_UFC_TRANS"/> + <value value="29" name="A7XX_PERF_UFC_STALL_CYCLES_GBIF_CMD"/> + <value value="30" name="A7XX_PERF_UFC_STALL_CYCLES_GBIF_RDATA"/> + <value value="31" name="A7XX_PERF_UFC_STALL_CYCLES_GBIF_WDATA"/> + <value value="32" name="A7XX_PERF_UFC_STALL_CYCLES_UBWC_WR_FLAG"/> + <value value="33" name="A7XX_PERF_UFC_STALL_CYCLES_UBWC_FLAG_RTN"/> + <value value="34" name="A7XX_PERF_UFC_STALL_CYCLES_UBWC_EVENT"/> + <value value="35" name="A7XX_PERF_UFC_LRZ_PREFETCH_STALLED_CYCLES"/> + <value value="36" name="A7XX_PERF_UFC_CRE_PREFETCH_STALLED_CYCLES"/> + <value value="37" name="A7XX_PERF_UFC_SPTP_PREFETCH_STALLED_CYCLES"/> + <value value="38" name="A7XX_PERF_UFC_UBWC_RD_STALLED_CYCLES"/> + <value value="39" name="A7XX_PERF_UFC_UBWC_WR_STALLED_CYCLES"/> + <value value="40" name="A7XX_PERF_UFC_PREFETCH_STALLED_CYCLES"/> + <value value="41" name="A7XX_PERF_UFC_EVICTION_STALLED_CYCLES"/> + <value value="42" name="A7XX_PERF_UFC_LOCK_STALLED_CYCLES"/> + <value value="43" name="A7XX_PERF_UFC_MISS_LATENCY_CYCLES"/> + <value value="44" name="A7XX_PERF_UFC_MISS_LATENCY_SAMPLES"/> + <value value="45" name="A7XX_PERF_UFC_UBWC_REQ_STALLED_CYCLES"/> + <value value="46" name="A7XX_PERF_UFC_TP_HINT_TAG_MISS"/> + <value value="47" name="A7XX_PERF_UFC_TP_HINT_TAG_HIT_RDY"/> + <value value="48" name="A7XX_PERF_UFC_TP_HINT_TAG_HIT_NRDY"/> + <value value="49" name="A7XX_PERF_UFC_TP_HINT_IS_FCLEAR"/> + <value value="50" name="A7XX_PERF_UFC_TP_HINT_IS_ALPHA0"/> + <value value="51" name="A7XX_PERF_UFC_SP_L1_FILTER_HIT"/> + <value value="52" name="A7XX_PERF_UFC_SP_L1_FILTER_MISS"/> + <value value="53" name="A7XX_PERF_UFC_SP_L1_FILTER_REQUESTS"/> + <value value="54" name="A7XX_PERF_UFC_TP_L1_TAG_HIT_RDY"/> + <value value="55" name="A7XX_PERF_UFC_TP_L1_TAG_HIT_NRDY"/> + <value value="56" name="A7XX_PERF_UFC_TP_L1_TAG_MISS"/> + <value value="57" name="A7XX_PERF_UFC_TP_L1_FILTER_REQUESTS"/> +</enum> + +</database> diff --git a/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml b/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml index 5a6ae9fc3194..7abc08635495 100644 --- a/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml +++ b/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml @@ -21,9 +21,9 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <value name="HLSQ_FLUSH" value="7" variants="A3XX-A4XX"/> <value name="VIZQUERY_END" value="8" variants="A2XX"/> <value name="SC_WAIT_WC" value="9" variants="A2XX"/> - <value name="WRITE_PRIMITIVE_COUNTS" value="9" variants="A6XX"/> - <value name="START_PRIMITIVE_CTRS" value="11" variants="A6XX"/> - <value name="STOP_PRIMITIVE_CTRS" value="12" variants="A6XX"/> + <value name="WRITE_PRIMITIVE_COUNTS" value="9" variants="A6XX-"/> + <value name="START_PRIMITIVE_CTRS" value="11" variants="A6XX-"/> + <value name="STOP_PRIMITIVE_CTRS" value="12" variants="A6XX-"/> <!-- Not sure that these 4 events don't have the same meaning as on A5XX+ --> <value name="RST_PIX_CNT" value="13" variants="A2XX-A4XX"/> <value name="RST_VTX_CNT" value="14" variants="A2XX-A4XX"/> @@ -31,8 +31,8 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <value name="STAT_EVENT" value="16" variants="A2XX-A4XX"/> <value name="CACHE_FLUSH_AND_INV_TS_EVENT" value="20" variants="A2XX-A4XX"/> <doc> - If A6XX_RB_SAMPLE_COUNT_CONTROL.copy is true, writes OQ Z passed - sample counts to RB_SAMPLE_COUNT_ADDR. This writes to main + If A6XX_RB_SAMPLE_COUNTER_CNTL.copy is true, writes OQ Z passed + sample counts to RB_SAMPLE_COUNTER_BASE. This writes to main memory, skipping UCHE. </doc> <value name="ZPASS_DONE" value="21"/> @@ -98,6 +98,13 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <value name="BLIT" value="30" variants="A5XX-"/> <doc> + Flip between the primary and secondary LRZ buffers. This is used + for concurrent binning, so that BV can write to one buffer while + BR reads from the other. + </doc> + <value name="LRZ_FLIP_BUFFER" value="36" variants="A7XX"/> + + <doc> Clears based on GRAS_LRZ_CNTL configuration, could clear fast-clear buffer or LRZ direction. LRZ direction is stored at lrz_fc_offset + 0x200, has 1 byte which @@ -114,6 +121,7 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <value name="BLIT_OP_FILL_2D" value="39" variants="A5XX-"/> <value name="BLIT_OP_COPY_2D" value="40" variants="A5XX-A6XX"/> <value name="UNK_40" value="40" variants="A7XX"/> + <value name="LRZ_Q_CACHE_INVALIDATE" value="41" variants="A7XX"/> <value name="BLIT_OP_SCALE_2D" value="42" variants="A5XX-"/> <value name="CONTEXT_DONE_2D" value="43" variants="A5XX-"/> <value name="UNK_2C" value="44" variants="A5XX-"/> @@ -372,7 +380,7 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <value name="CP_LOAD_STATE" value="0x30" variants="A3XX"/> <value name="CP_LOAD_STATE4" value="0x30" variants="A4XX-A5XX"/> <doc>Conditionally load a IB based on a flag, prefetch enabled</doc> - <value name="CP_COND_INDIRECT_BUFFER_PFE" value="0x3a"/> + <value name="CP_COND_INDIRECT_BUFFER_PFE" value="0x3a" variants="A3XX-A5XX"/> <doc>Conditionally load a IB based on a flag, prefetch disabled</doc> <value name="CP_COND_INDIRECT_BUFFER_PFD" value="0x32" variants="A3XX"/> <doc>Load a buffer with pre-fetch enabled</doc> @@ -538,7 +546,7 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <value name="CP_LOAD_STATE6_GEOM" value="0x32" variants="A6XX-"/> <value name="CP_LOAD_STATE6_FRAG" value="0x34" variants="A6XX-"/> <!-- - Note: For IBO state (Image/SSBOs) which have shared state across + Note: For UAV state (Image/SSBOs) which have shared state across shader stages, for 3d pipeline CP_LOAD_STATE6 is used. But for compute shaders, CP_LOAD_STATE6_FRAG is used. Possibly they are interchangable. @@ -567,7 +575,7 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <value name="IN_PREEMPT" value="0x0f" variants="A6XX-"/> <!-- TODO do these exist on A5xx? --> - <value name="CP_SCRATCH_WRITE" value="0x4c" variants="A6XX"/> + <value name="CP_SCRATCH_WRITE" value="0x4c" variants="A6XX-"/> <value name="CP_REG_TO_MEM_OFFSET_MEM" value="0x74" variants="A6XX-"/> <value name="CP_REG_TO_MEM_OFFSET_REG" value="0x72" variants="A6XX-"/> <value name="CP_WAIT_MEM_GTE" value="0x14" variants="A6XX"/> @@ -650,6 +658,11 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <doc>Reset various on-chip state used for synchronization</doc> <value name="CP_RESET_CONTEXT_STATE" value="0x1f" variants="A7XX-"/> + + <doc>Invalidates the "CCHE" introduced on a740</doc> + <value name="CP_CCHE_INVALIDATE" value="0x3a" variants="A7XX-"/> + + <value name="CP_SCOPE_CNTL" value="0x6c" variants="A7XX-"/> </enum> @@ -792,14 +805,14 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) <value name="SB6_GS_SHADER" value="0xb"/> <value name="SB6_FS_SHADER" value="0xc"/> <value name="SB6_CS_SHADER" value="0xd"/> - <value name="SB6_IBO" value="0xe"/> - <value name="SB6_CS_IBO" value="0xf"/> + <value name="SB6_UAV" value="0xe"/> + <value name="SB6_CS_UAV" value="0xf"/> </enum> <enum name="a6xx_state_type"> <value name="ST6_SHADER" value="0"/> <value name="ST6_CONSTANTS" value="1"/> <value name="ST6_UBO" value="2"/> - <value name="ST6_IBO" value="3"/> + <value name="ST6_UAV" value="3"/> </enum> <enum name="a6xx_state_src"> <value name="SS6_DIRECT" value="0"/> @@ -1121,39 +1134,93 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) </reg32> </domain> +<enum name="a7xx_abs_mask_mode"> + <value name="ABS_MASK" value="0x1"/> + <value name="NO_ABS_MASK" value="0x0"/> +</enum> + <domain name="CP_SET_BIN_DATA5" width="32"> <reg32 offset="0" name="0"> + <bitfield name="VSC_MASK" low="0" high="15" type="hex"> + <doc> + A mask of bins, starting at VSC_N, whose + visibility is OR'd together. A value of 0 is + interpreted as 1 (i.e. just use VSC_N for + visbility) for backwards compatibility. Only + exists on a7xx. + </doc> + </bitfield> <!-- equiv to PC_VSTREAM_CONTROL.SIZE on a3xx/a4xx: --> <bitfield name="VSC_SIZE" low="16" high="21" type="uint"/> <!-- equiv to PC_VSTREAM_CONTROL.N on a3xx/a4xx: --> <bitfield name="VSC_N" low="22" high="26" type="uint"/> + <bitfield name="ABS_MASK" pos="28" type="a7xx_abs_mask_mode" addvariant="yes"> + <doc> + If this field is 1, VSC_MASK and VSC_N are + ignored and instead a new ordinal immediately + after specifies the full 32-bit mask of bins + to use. The mask is "absolute" instead of + relative to VSC_N. + </doc> + </bitfield> </reg32> - <!-- BIN_DATA_ADDR -> VSC_PIPE[p].DATA_ADDRESS --> - <reg32 offset="1" name="1"> - <bitfield name="BIN_DATA_ADDR_LO" low="0" high="31" type="hex"/> - </reg32> - <reg32 offset="2" name="2"> - <bitfield name="BIN_DATA_ADDR_HI" low="0" high="31" type="hex"/> - </reg32> - <!-- BIN_SIZE_ADDRESS -> VSC_SIZE_ADDRESS + (p * 4)--> - <reg32 offset="3" name="3"> - <bitfield name="BIN_SIZE_ADDRESS_LO" low="0" high="31"/> - </reg32> - <reg32 offset="4" name="4"> - <bitfield name="BIN_SIZE_ADDRESS_HI" low="0" high="31"/> - </reg32> - <!-- new on a6xx, where BIN_DATA_ADDR is the DRAW_STRM: --> - <reg32 offset="5" name="5"> - <bitfield name="BIN_PRIM_STRM_LO" low="0" high="31"/> - </reg32> - <reg32 offset="6" name="6"> - <bitfield name="BIN_PRIM_STRM_HI" low="0" high="31"/> - </reg32> - <!-- - a7xx adds a few more addresses to the end of the pkt - --> - <reg64 offset="7" name="7"/> - <reg64 offset="9" name="9"/> + <stripe varset="a7xx_abs_mask_mode" variants="NO_ABS_MASK"> + <!-- BIN_DATA_ADDR -> VSC_PIPE[p].DATA_ADDRESS --> + <reg32 offset="1" name="1"> + <bitfield name="BIN_DATA_ADDR_LO" low="0" high="31" type="hex"/> + </reg32> + <reg32 offset="2" name="2"> + <bitfield name="BIN_DATA_ADDR_HI" low="0" high="31" type="hex"/> + </reg32> + <!-- BIN_SIZE_ADDRESS -> VSC_SIZE_ADDRESS + (p * 4)--> + <reg32 offset="3" name="3"> + <bitfield name="BIN_SIZE_ADDRESS_LO" low="0" high="31"/> + </reg32> + <reg32 offset="4" name="4"> + <bitfield name="BIN_SIZE_ADDRESS_HI" low="0" high="31"/> + </reg32> + <!-- new on a6xx, where BIN_DATA_ADDR is the DRAW_STRM: --> + <reg32 offset="5" name="5"> + <bitfield name="BIN_PRIM_STRM_LO" low="0" high="31"/> + </reg32> + <reg32 offset="6" name="6"> + <bitfield name="BIN_PRIM_STRM_HI" low="0" high="31"/> + </reg32> + <!-- + a7xx adds a few more addresses to the end of the pkt + --> + <reg64 offset="7" name="7"/> + <reg64 offset="9" name="9"/> + </stripe> + <stripe varset="a7xx_abs_mask_mode" variants="ABS_MASK"> + <reg32 offset="1" name="ABS_MASK"/> + <!-- BIN_DATA_ADDR -> VSC_PIPE[p].DATA_ADDRESS --> + <reg32 offset="2" name="2"> + <bitfield name="BIN_DATA_ADDR_LO" low="0" high="31" type="hex"/> + </reg32> + <reg32 offset="3" name="3"> + <bitfield name="BIN_DATA_ADDR_HI" low="0" high="31" type="hex"/> + </reg32> + <!-- BIN_SIZE_ADDRESS -> VSC_SIZE_ADDRESS + (p * 4)--> + <reg32 offset="4" name="4"> + <bitfield name="BIN_SIZE_ADDRESS_LO" low="0" high="31"/> + </reg32> + <reg32 offset="5" name="5"> + <bitfield name="BIN_SIZE_ADDRESS_HI" low="0" high="31"/> + </reg32> + <!-- new on a6xx, where BIN_DATA_ADDR is the DRAW_STRM: --> + <reg32 offset="6" name="6"> + <bitfield name="BIN_PRIM_STRM_LO" low="0" high="31"/> + </reg32> + <reg32 offset="7" name="7"> + <bitfield name="BIN_PRIM_STRM_HI" low="0" high="31"/> + </reg32> + <!-- + a7xx adds a few more addresses to the end of the pkt + --> + <reg64 offset="8" name="8"/> + <reg64 offset="10" name="10"/> + </stripe> </domain> <domain name="CP_SET_BIN_DATA5_OFFSET" width="32"> @@ -1164,23 +1231,42 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) stream is recorded. </doc> <reg32 offset="0" name="0"> + <bitfield name="VSC_MASK" low="0" high="15" type="hex"/> <!-- equiv to PC_VSTREAM_CONTROL.SIZE on a3xx/a4xx: --> <bitfield name="VSC_SIZE" low="16" high="21" type="uint"/> <!-- equiv to PC_VSTREAM_CONTROL.N on a3xx/a4xx: --> <bitfield name="VSC_N" low="22" high="26" type="uint"/> + <bitfield name="ABS_MASK" pos="28" type="a7xx_abs_mask_mode" addvariant="yes"/> </reg32> - <!-- BIN_DATA_ADDR -> VSC_PIPE[p].DATA_ADDRESS --> - <reg32 offset="1" name="1"> - <bitfield name="BIN_DATA_OFFSET" low="0" high="31" type="uint"/> - </reg32> - <!-- BIN_SIZE_ADDRESS -> VSC_SIZE_ADDRESS + (p * 4)--> - <reg32 offset="2" name="2"> - <bitfield name="BIN_SIZE_OFFSET" low="0" high="31" type="uint"/> - </reg32> - <!-- BIN_DATA2_ADDR -> VSC_PIPE[p].DATA2_ADDRESS --> - <reg32 offset="3" name="3"> - <bitfield name="BIN_DATA2_OFFSET" low="0" high="31" type="uint"/> - </reg32> + <stripe varset="a7xx_abs_mask_mode" variants="NO_ABS_MASK"> + <!-- BIN_DATA_ADDR -> VSC_PIPE[p].DATA_ADDRESS --> + <reg32 offset="1" name="1"> + <bitfield name="BIN_DATA_OFFSET" low="0" high="31" type="uint"/> + </reg32> + <!-- BIN_SIZE_ADDRESS -> VSC_SIZE_ADDRESS + (p * 4)--> + <reg32 offset="2" name="2"> + <bitfield name="BIN_SIZE_OFFSET" low="0" high="31" type="uint"/> + </reg32> + <!-- BIN_DATA2_ADDR -> VSC_PIPE[p].DATA2_ADDRESS --> + <reg32 offset="3" name="3"> + <bitfield name="BIN_DATA2_OFFSET" low="0" high="31" type="uint"/> + </reg32> + </stripe> + <stripe varset="a7xx_abs_mask_mode" variants="ABS_MASK"> + <reg32 offset="1" name="ABS_MASK"/> + <!-- BIN_DATA_ADDR -> VSC_PIPE[p].DATA_ADDRESS --> + <reg32 offset="2" name="2"> + <bitfield name="BIN_DATA_OFFSET" low="0" high="31" type="uint"/> + </reg32> + <!-- BIN_SIZE_ADDRESS -> VSC_SIZE_ADDRESS + (p * 4)--> + <reg32 offset="3" name="3"> + <bitfield name="BIN_SIZE_OFFSET" low="0" high="31" type="uint"/> + </reg32> + <!-- BIN_DATA2_ADDR -> VSC_PIPE[p].DATA2_ADDRESS --> + <reg32 offset="4" name="4"> + <bitfield name="BIN_DATA2_OFFSET" low="0" high="31" type="uint"/> + </reg32> + </stripe> </domain> <domain name="CP_REG_RMW" width="32"> @@ -1198,6 +1284,9 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) </doc> <reg32 offset="0" name="0"> <bitfield name="DST_REG" low="0" high="17" type="hex"/> + <bitfield name="DST_SCRATCH" pos="19" type="boolean" varset="chip" variants="A7XX-"/> + <!-- skip implied CP_WAIT_FOR_IDLE + CP_WAIT_FOR_ME --> + <bitfield name="SKIP_WAIT_FOR_ME" pos="23" type="boolean" varset="chip" variants="A7XX-"/> <bitfield name="ROTATE" low="24" high="28" type="uint"/> <bitfield name="SRC1_ADD" pos="29" type="boolean"/> <bitfield name="SRC1_IS_REG" pos="30" type="boolean"/> @@ -1348,6 +1437,8 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) <bitfield name="SCRATCH" low="20" high="22" type="uint"/> <!-- number of registers/dwords copied is CNT + 1. --> <bitfield name="CNT" low="24" high="26" type="uint"/> + <!-- skip implied CP_WAIT_FOR_IDLE + CP_WAIT_FOR_ME --> + <bitfield name="SKIP_WAIT_FOR_ME" pos="27" type="boolean" varset="chip" variants="A7XX-"/> </reg32> </domain> @@ -1655,8 +1746,8 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) <bitfield name="WRITE_SAMPLE_COUNT" pos="12" type="boolean"/> <!-- Write sample count at (iova + 16) --> <bitfield name="SAMPLE_COUNT_END_OFFSET" pos="13" type="boolean"/> - <!-- *(iova + 8) = *(iova + 16) - *iova --> - <bitfield name="WRITE_SAMPLE_COUNT_DIFF" pos="14" type="boolean"/> + <!-- *(iova + 8) += *(iova + 16) - *iova --> + <bitfield name="WRITE_ACCUM_SAMPLE_COUNT_DIFF" pos="14" type="boolean"/> <!-- Next 4 flags are valid to set only when concurrent binning is enabled --> <!-- Increment 16b BV counter. Valid only in BV pipe --> @@ -1670,15 +1761,11 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) <bitfield name="WRITE_DST" pos="24" type="event_write_dst" addvariant="yes"/> <!-- Writes into WRITE_DST from WRITE_SRC. RB_DONE_TS requires WRITE_ENABLED. --> <bitfield name="WRITE_ENABLED" pos="27" type="boolean"/> + <bitfield name="IRQ" pos="31" type="boolean"/> </reg32> <stripe varset="event_write_dst" variants="EV_DST_RAM"> - <reg32 offset="1" name="1"> - <bitfield name="ADDR_0_LO" low="0" high="31"/> - </reg32> - <reg32 offset="2" name="2"> - <bitfield name="ADDR_0_HI" low="0" high="31"/> - </reg32> + <reg64 offset="1" name="1" type="waddress"/> <reg32 offset="3" name="3"> <bitfield name="PAYLOAD_0" low="0" high="31"/> </reg32> @@ -1773,13 +1860,23 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) <domain name="CP_SET_MARKER" width="32" varset="chip" prefix="chip" variants="A6XX-"> <doc>Tell CP the current operation mode, indicates save and restore procedure</doc> + <enum name="set_marker_mode"> + <value value="0" name="SET_RENDER_MODE"/> + <!-- IFPC - inter-frame power collapse --> + <value value="1" name="SET_IFPC_MODE"/> + </enum> + <enum name="a6xx_ifpc_mode"> + <value value="0" name="IFPC_ENABLE"/> + <value value="1" name="IFPC_DISABLE"/> + </enum> <enum name="a6xx_marker"> - <value value="1" name="RM6_BYPASS"/> - <value value="2" name="RM6_BINNING"/> - <value value="4" name="RM6_GMEM"/> - <value value="5" name="RM6_ENDVIS"/> - <value value="6" name="RM6_RESOLVE"/> - <value value="7" name="RM6_YIELD"/> + <value value="1" name="RM6_DIRECT_RENDER"/> + <value value="2" name="RM6_BIN_VISIBILITY"/> + <value value="3" name="RM6_BIN_DIRECT"/> + <value value="4" name="RM6_BIN_RENDER_START"/> + <value value="5" name="RM6_BIN_END_OF_DRAWS"/> + <value value="6" name="RM6_BIN_RESOLVE"/> + <value value="7" name="RM6_BIN_RENDER_END"/> <value value="8" name="RM6_COMPUTE"/> <value value="0xc" name="RM6_BLIT2DSCALE"/> <!-- no-op (at least on current sqe fw) --> @@ -1789,23 +1886,40 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) --> <value value="0xd" name="RM6_IB1LIST_START"/> <value value="0xe" name="RM6_IB1LIST_END"/> - <!-- IFPC - inter-frame power collapse --> - <value value="0x100" name="RM6_IFPC_ENABLE"/> - <value value="0x101" name="RM6_IFPC_DISABLE"/> </enum> <reg32 offset="0" name="0"> + <!-- if b8 is set, the low bits are interpreted differently (and b4 ignored) --> + <bitfield name="MARKER_MODE" pos="8" type="set_marker_mode" addvariant="yes"/> + + <bitfield name="MODE" low="0" high="3" type="a6xx_marker" varset="set_marker_mode" variants="SET_RENDER_MODE"/> + <!-- used by preemption to determine if GMEM needs to be saved or not --> + <bitfield name="USES_GMEM" pos="4" type="boolean" varset="set_marker_mode" variants="SET_RENDER_MODE"/> + + <bitfield name="IFPC_MODE" pos="0" type="a6xx_ifpc_mode" varset="set_marker_mode" variants="SET_IFPC_MODE"/> + <!-- - NOTE: blob driver and some versions of freedreno/turnip set - b4, which is unused (at least by current sqe fw), but interferes - with parsing if we extend the size of the bitfield to include - b8 (only sent by kernel mode driver). Really, the way the - parsing works in the firmware, only b0-b3 are considered, but - if b8 is set, the low bits are interpreted differently. To - model this, without getting confused by spurious b4, this is - described as two overlapping bitfields: - --> - <bitfield name="MODE" low="0" high="8" type="a6xx_marker"/> - <bitfield name="MARKER" low="0" high="3" type="a6xx_marker"/> + CP_SET_MARKER is used with these bits to create a + critical section around a workaround for ray tracing. + The workaround happens after BVH building, and appears + to invalidate the RTU's BVH node cache. It makes sure + that only one of BR/BV/LPAC is executing the + workaround at a time, and no draws using RT on BV/LPAC + are executing while the workaround is executed on BR (or + vice versa, that no draws on BV/BR using RT are executed + while the workaround executes on LPAC), by + hooking subsequent CP_EVENT_WRITE/CP_DRAW_*/CP_EXEC_CS. + The blob usage is: + + CP_SET_MARKER(RT_WA_START) + ... workaround here ... + CP_SET_MARKER(RT_WA_END) + ... + CP_SET_MARKER(SHADER_USES_RT) + CP_DRAW_INDX(...) or CP_EXEC_CS(...) + --> + <bitfield name="SHADER_USES_RT" pos="9" type="boolean" variants="A7XX-"/> + <bitfield name="RT_WA_START" pos="10" type="boolean" variants="A7XX-"/> + <bitfield name="RT_WA_END" pos="11" type="boolean" variants="A7XX-"/> </reg32> </domain> @@ -1832,9 +1946,9 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) If concurrent binning is disabled then BR also does binning so it will also write the "real" registers in BR. --> - <value value="8" name="DRAW_STRM_ADDRESS"/> - <value value="9" name="DRAW_STRM_SIZE_ADDRESS"/> - <value value="10" name="PRIM_STRM_ADDRESS"/> + <value value="8" name="VSC_PIPE_DATA_DRAW_BASE"/> + <value value="9" name="VSC_SIZE_BASE"/> + <value value="10" name="VSC_PIPE_DATA_PRIM_BASE"/> <value value="11" name="UNK_STRM_ADDRESS"/> <value value="12" name="UNK_STRM_SIZE_ADDRESS"/> @@ -1935,11 +2049,11 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) a bitmask of which modes pass the test. --> - <!-- RM6_BINNING --> + <!-- RM6_BIN_VISIBILITY --> <bitfield name="BINNING" pos="25" variants="RENDER_MODE" type="boolean"/> <!-- all others --> <bitfield name="GMEM" pos="26" variants="RENDER_MODE" type="boolean"/> - <!-- RM6_BYPASS --> + <!-- RM6_DIRECT_RENDER --> <bitfield name="SYSMEM" pos="27" variants="RENDER_MODE" type="boolean"/> <bitfield name="BV" pos="25" variants="THREAD_MODE" type="boolean"/> @@ -2014,10 +2128,10 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) <domain name="CP_SET_AMBLE" width="32"> <doc> - Used by the userspace and kernel drivers to set various IB's - which are executed during context save/restore for handling - state that isn't restored by the context switch routine itself. - </doc> + Used by the userspace and kernel drivers to set various IB's + which are executed during context save/restore for handling + state that isn't restored by the context switch routine itself. + </doc> <enum name="amble_type"> <value name="PREAMBLE_AMBLE_TYPE" value="0"> <doc>Executed unconditionally when switching back to the context.</doc> @@ -2087,12 +2201,12 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) <value name="UNK_EVENT_WRITE" value="0x4"/> <doc> Tracks GRAS_LRZ_CNTL::GREATER, GRAS_LRZ_CNTL::DIR, and - GRAS_LRZ_DEPTH_VIEW with previous values, and if one of + GRAS_LRZ_VIEW_INFO with previous values, and if one of the following is true: - GRAS_LRZ_CNTL::GREATER has changed - GRAS_LRZ_CNTL::DIR has changed, the old value is not CUR_DIR_GE, and the new value is not CUR_DIR_DISABLED - - GRAS_LRZ_DEPTH_VIEW has changed + - GRAS_LRZ_VIEW_INFO has changed then it does a LRZ_FLUSH with GRAS_LRZ_CNTL::ENABLE forced to 1. Only exists in a650_sqe.fw. @@ -2207,7 +2321,7 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) <domain name="CP_MEM_TO_SCRATCH_MEM" width="32"> <doc> - Best guess is that it is a faster way to fetch all the VSC_STATE registers + Best guess is that it is a faster way to fetch all the VSC_CHANNEL_VISIBILITY registers and keep them in a local scratch memory instead of fetching every time when skipping IBs. </doc> @@ -2255,7 +2369,18 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) <reg32 offset="0" name="0"> <bitfield name="CLEAR_ON_CHIP_TS" pos="0" type="boolean"/> <bitfield name="CLEAR_RESOURCE_TABLE" pos="1" type="boolean"/> - <bitfield name="CLEAR_GLOBAL_LOCAL_TS" pos="2" type="boolean"/> + <bitfield name="CLEAR_BV_BR_COUNTER" pos="2" type="boolean"/> + <bitfield name="RESET_GLOBAL_LOCAL_TS" pos="3" type="boolean"/> + </reg32> +</domain> + +<domain name="CP_SCOPE_CNTL" width="32"> + <enum name="cp_scope"> + <value value="0" name="INTERRUPTS"/> + </enum> + <reg32 offset="0" name="0"> + <bitfield name="DISABLE_PREEMPTION" pos="0" type="boolean"/> + <bitfield low="28" high="31" name="SCOPE" type="cp_scope"/> </reg32> </domain> diff --git a/drivers/gpu/drm/msm/registers/display/dsi_phy_7nm.xml b/drivers/gpu/drm/msm/registers/display/dsi_phy_7nm.xml index d2c8c46bb041..4e5ac0f25dea 100644 --- a/drivers/gpu/drm/msm/registers/display/dsi_phy_7nm.xml +++ b/drivers/gpu/drm/msm/registers/display/dsi_phy_7nm.xml @@ -26,6 +26,7 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <reg32 offset="0x00028" name="CTRL_1"/> <reg32 offset="0x0002c" name="CTRL_2"/> <reg32 offset="0x00030" name="CTRL_3"/> + <reg32 offset="0x001b0" name="CTRL_5"/> <reg32 offset="0x00034" name="LANE_CFG0"/> <reg32 offset="0x00038" name="LANE_CFG1"/> <reg32 offset="0x0003c" name="PLL_CNTRL"/> @@ -191,11 +192,24 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd"> <reg32 offset="0x01b0" name="COMMON_STATUS_ONE"/> <reg32 offset="0x01b4" name="COMMON_STATUS_TWO"/> <reg32 offset="0x01b8" name="BAND_SEL_CAL"/> + <!-- + Starting with SM8750, offset moved from 0x01bc to 0x01cc, however + we keep only one register map. That's not a problem, so far, + because this register is not used. The register map should be split + once it is going to be used. Comment out the code to prevent + any misuse due to the change in the offset. <reg32 offset="0x01bc" name="ICODE_ACCUM_STATUS_LOW"/> + <reg32 offset="0x01cc" name="ICODE_ACCUM_STATUS_LOW"/> + --> <reg32 offset="0x01c0" name="ICODE_ACCUM_STATUS_HIGH"/> <reg32 offset="0x01c4" name="FD_OUT_LOW"/> <reg32 offset="0x01c8" name="FD_OUT_HIGH"/> + <!-- + Starting with SM8750, offset moved from 0x01cc to 0x01bc, however + we keep only one register map. See above comment. <reg32 offset="0x01cc" name="ALOG_OBSV_BUS_STATUS_1"/> + <reg32 offset="0x01bc" name="ALOG_OBSV_BUS_STATUS_1"/> + --> <reg32 offset="0x01d0" name="PLL_MISC_CONFIG"/> <reg32 offset="0x01d4" name="FLL_CONFIG"/> <reg32 offset="0x01d8" name="FLL_FREQ_ACQ_TIME"/> diff --git a/drivers/gpu/drm/msm/registers/gen_header.py b/drivers/gpu/drm/msm/registers/gen_header.py index 3926485bb197..a409404627c7 100644 --- a/drivers/gpu/drm/msm/registers/gen_header.py +++ b/drivers/gpu/drm/msm/registers/gen_header.py @@ -11,6 +11,7 @@ import collections import argparse import time import datetime +import re class Error(Exception): def __init__(self, message): @@ -877,13 +878,14 @@ The rules-ng-ng source files this header was generated from are: """) maxlen = 0 for filepath in p.xml_files: - maxlen = max(maxlen, len(filepath)) + new_filepath = re.sub("^.+drivers","drivers",filepath) + maxlen = max(maxlen, len(new_filepath)) for filepath in p.xml_files: - pad = " " * (maxlen - len(filepath)) + pad = " " * (maxlen - len(new_filepath)) filesize = str(os.path.getsize(filepath)) filesize = " " * (7 - len(filesize)) + filesize filetime = time.ctime(os.path.getmtime(filepath)) - print("- " + filepath + pad + " (" + filesize + " bytes, from " + filetime + ")") + print("- " + new_filepath + pad + " (" + filesize + " bytes, from <stripped>)") if p.copyright_year: current_year = str(datetime.date.today().year) print() diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c index e5d37eee4301..e97e39abf3a2 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -1839,7 +1839,7 @@ nv50_sor_atomic_enable(struct drm_encoder *encoder, struct drm_atomic_state *sta backlight = nv_connector->backlight; if (backlight && backlight->uses_dpcd) drm_edp_backlight_enable(&nv_connector->aux, &backlight->edp_info, - (u16)backlight->dev->props.brightness); + backlight->dev->props.brightness); #endif break; diff --git a/drivers/gpu/drm/nouveau/nouveau_backlight.c b/drivers/gpu/drm/nouveau/nouveau_backlight.c index d47442125fa1..e006aaf57ff5 100644 --- a/drivers/gpu/drm/nouveau/nouveau_backlight.c +++ b/drivers/gpu/drm/nouveau/nouveau_backlight.c @@ -245,7 +245,7 @@ nv50_backlight_init(struct nouveau_backlight *bl, if (nv_conn->type == DCB_CONNECTOR_eDP) { int ret; - u16 current_level; + u32 current_level; u8 edp_dpcd[EDP_DISPLAY_CTL_CAP_SIZE]; u8 current_mode; @@ -261,8 +261,9 @@ nv50_backlight_init(struct nouveau_backlight *bl, NV_DEBUG(drm, "DPCD backlight controls supported on %s\n", nv_conn->base.name); - ret = drm_edp_backlight_init(&nv_conn->aux, &bl->edp_info, 0, edp_dpcd, - ¤t_level, ¤t_mode); + ret = drm_edp_backlight_init(&nv_conn->aux, &bl->edp_info, + 0, 0, edp_dpcd, + ¤t_level, ¤t_mode, false); if (ret < 0) return ret; diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c index add006fc8d81..c50ec347b30a 100644 --- a/drivers/gpu/drm/nouveau/nouveau_display.c +++ b/drivers/gpu/drm/nouveau/nouveau_display.c @@ -495,7 +495,7 @@ nouveau_display_hpd_work(struct work_struct *work) if (first_changed_connector) drm_connector_put(first_changed_connector); - pm_runtime_mark_last_busy(drm->dev->dev); + pm_runtime_mark_last_busy(dev->dev); noop: pm_runtime_put_autosuspend(dev->dev); } diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig index d5aa1c95c6a4..09b9f7ff9340 100644 --- a/drivers/gpu/drm/panel/Kconfig +++ b/drivers/gpu/drm/panel/Kconfig @@ -193,6 +193,16 @@ config DRM_PANEL_HIMAX_HX83112A Say Y here if you want to enable support for Himax HX83112A-based display panels, such as the one found in the Fairphone 4 smartphone. +config DRM_PANEL_HIMAX_HX83112B + tristate "Himax HX83112B-based DSI panel" + depends on OF + depends on DRM_MIPI_DSI + depends on BACKLIGHT_CLASS_DEVICE + select DRM_KMS_HELPER + help + Say Y here if you want to enable support for Himax HX83112B-based + display panels, such as the one found in the Fairphone 3 smartphone. + config DRM_PANEL_HIMAX_HX8394 tristate "HIMAX HX8394 MIPI-DSI LCD panels" depends on OF diff --git a/drivers/gpu/drm/panel/Makefile b/drivers/gpu/drm/panel/Makefile index 73a39bc72604..957555b49996 100644 --- a/drivers/gpu/drm/panel/Makefile +++ b/drivers/gpu/drm/panel/Makefile @@ -20,6 +20,7 @@ obj-$(CONFIG_DRM_PANEL_FEIYANG_FY07024DI26A30D) += panel-feiyang-fy07024di26a30d obj-$(CONFIG_DRM_PANEL_HIMAX_HX8279) += panel-himax-hx8279.o obj-$(CONFIG_DRM_PANEL_HIMAX_HX83102) += panel-himax-hx83102.o obj-$(CONFIG_DRM_PANEL_HIMAX_HX83112A) += panel-himax-hx83112a.o +obj-$(CONFIG_DRM_PANEL_HIMAX_HX83112B) += panel-himax-hx83112b.o obj-$(CONFIG_DRM_PANEL_HIMAX_HX8394) += panel-himax-hx8394.o obj-$(CONFIG_DRM_PANEL_ILITEK_IL9322) += panel-ilitek-ili9322.o obj-$(CONFIG_DRM_PANEL_ILITEK_ILI9341) += panel-ilitek-ili9341.o diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c index 6c45c9e879ec..3796c41629cc 100644 --- a/drivers/gpu/drm/panel/panel-edp.c +++ b/drivers/gpu/drm/panel/panel-edp.c @@ -1967,6 +1967,7 @@ static const struct edp_panel_entry edp_panels[] = { EDP_PANEL_ENTRY('C', 'M', 'N', 0x115e, &delay_200_500_e80_d50, "N116BCA-EA1"), EDP_PANEL_ENTRY('C', 'M', 'N', 0x1160, &delay_200_500_e80_d50, "N116BCJ-EAK"), EDP_PANEL_ENTRY('C', 'M', 'N', 0x1161, &delay_200_500_e80, "N116BCP-EA2"), + EDP_PANEL_ENTRY('C', 'M', 'N', 0x1163, &delay_200_500_e80_d50, "N116BCJ-EAK"), EDP_PANEL_ENTRY('C', 'M', 'N', 0x1247, &delay_200_500_e80_d50, "N120ACA-EA1"), EDP_PANEL_ENTRY('C', 'M', 'N', 0x142b, &delay_200_500_e80_d50, "N140HCA-EAC"), EDP_PANEL_ENTRY('C', 'M', 'N', 0x142e, &delay_200_500_e80_d50, "N140BGA-EA4"), diff --git a/drivers/gpu/drm/panel/panel-himax-hx83112b.c b/drivers/gpu/drm/panel/panel-himax-hx83112b.c new file mode 100644 index 000000000000..263f79a967de --- /dev/null +++ b/drivers/gpu/drm/panel/panel-himax-hx83112b.c @@ -0,0 +1,430 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Generated with linux-mdss-dsi-panel-driver-generator from vendor device tree. + * Copyright (c) 2025 Luca Weiss <luca@lucaweiss.eu> + */ + +#include <linux/backlight.h> +#include <linux/delay.h> +#include <linux/gpio/consumer.h> +#include <linux/mod_devicetable.h> +#include <linux/module.h> +#include <linux/regulator/consumer.h> + +#include <video/mipi_display.h> + +#include <drm/drm_mipi_dsi.h> +#include <drm/drm_modes.h> +#include <drm/drm_panel.h> +#include <drm/drm_probe_helper.h> + +/* Manufacturer specific DSI commands */ +#define HX83112B_SETPOWER1 0xb1 +#define HX83112B_SETDISP 0xb2 +#define HX83112B_SETDRV 0xb4 +#define HX83112B_SETEXTC 0xb9 +#define HX83112B_SETBANK 0xbd +#define HX83112B_SETDGCLUT 0xc1 +#define HX83112B_SETDISMO 0xc2 +#define HX83112B_UNKNOWN1 0xc6 +#define HX83112B_SETPANEL 0xcc +#define HX83112B_UNKNOWN2 0xd1 +#define HX83112B_SETPOWER2 0xd2 +#define HX83112B_SETGIP0 0xd3 +#define HX83112B_SETGIP1 0xd5 +#define HX83112B_SETGIP2 0xd6 +#define HX83112B_SETGIP3 0xd8 +#define HX83112B_SETIDLE 0xdd +#define HX83112B_UNKNOWN3 0xe7 +#define HX83112B_UNKNOWN4 0xe9 + +struct hx83112b_panel { + struct drm_panel panel; + struct mipi_dsi_device *dsi; + struct regulator_bulk_data *supplies; + struct gpio_desc *reset_gpio; +}; + +static const struct regulator_bulk_data hx83112b_supplies[] = { + { .supply = "iovcc" }, + { .supply = "vsn" }, + { .supply = "vsp" }, +}; + +static inline struct hx83112b_panel *to_hx83112b_panel(struct drm_panel *panel) +{ + return container_of(panel, struct hx83112b_panel, panel); +} + +static void hx83112b_reset(struct hx83112b_panel *ctx) +{ + gpiod_set_value_cansleep(ctx->reset_gpio, 0); + usleep_range(10000, 11000); + gpiod_set_value_cansleep(ctx->reset_gpio, 1); + usleep_range(10000, 11000); + gpiod_set_value_cansleep(ctx->reset_gpio, 0); + usleep_range(10000, 11000); +} + +static int hx83112b_on(struct hx83112b_panel *ctx) +{ + struct mipi_dsi_multi_context dsi_ctx = { .dsi = ctx->dsi }; + + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETEXTC, 0x83, 0x11, 0x2b); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDISMO, 0x08, 0x70); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDISP, 0x04, 0x38, 0x08, 0x70); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETPOWER1, + 0xf8, 0x27, 0x27, 0x00, 0x00, 0x0b, 0x0e, + 0x0b, 0x0e, 0x33); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETPOWER2, 0x2d, 0x2d); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDISP, + 0x80, 0x02, 0x18, 0x80, 0x70, 0x00, 0x08, + 0x1c, 0x08, 0x11, 0x05); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0xd1); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDISP, 0x00, 0x08); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x02); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDISP, 0xb5, 0x0a); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETIDLE, + 0x00, 0x00, 0x08, 0x1c, 0x08, 0x34, 0x34, + 0x88); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDRV, + 0x65, 0x6b, 0x00, 0x00, 0xd0, 0xd4, 0x36, + 0xcf, 0x06, 0xce, 0x00, 0xce, 0x00, 0x00, + 0x00, 0x07, 0x00, 0x2a, 0x07, 0x01, 0x07, + 0x00, 0x00, 0x2a); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0xc3); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDRV, 0x01, 0x67, 0x2a); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDGCLUT, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDGCLUT, + 0xff, 0xfb, 0xf9, 0xf6, 0xf4, 0xf1, 0xef, + 0xea, 0xe7, 0xe5, 0xe2, 0xdf, 0xdd, 0xda, + 0xd8, 0xd5, 0xd2, 0xcf, 0xcc, 0xc5, 0xbe, + 0xb7, 0xb0, 0xa8, 0xa0, 0x98, 0x8e, 0x85, + 0x7b, 0x72, 0x69, 0x5e, 0x53, 0x48, 0x3e, + 0x35, 0x2b, 0x22, 0x17, 0x0d, 0x09, 0x07, + 0x05, 0x01, 0x00, 0x26, 0xf0, 0x86, 0x25, + 0x6e, 0xb6, 0xdd, 0xf3, 0xd8, 0xcc, 0x9b, + 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x02); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDGCLUT, + 0xff, 0xfb, 0xf9, 0xf6, 0xf4, 0xf1, 0xef, + 0xea, 0xe7, 0xe5, 0xe2, 0xdf, 0xdd, 0xda, + 0xd8, 0xd5, 0xd2, 0xcf, 0xcc, 0xc5, 0xbe, + 0xb7, 0xb0, 0xa8, 0xa0, 0x98, 0x8e, 0x85, + 0x7b, 0x72, 0x69, 0x5e, 0x53, 0x48, 0x3e, + 0x35, 0x2b, 0x22, 0x17, 0x0d, 0x09, 0x07, + 0x05, 0x01, 0x00, 0x26, 0xf0, 0x86, 0x25, + 0x6e, 0xb6, 0xdd, 0xf3, 0xd8, 0xcc, 0x9b, + 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDGCLUT, + 0xff, 0xfb, 0xf9, 0xf6, 0xf4, 0xf1, 0xef, + 0xea, 0xe7, 0xe5, 0xe2, 0xdf, 0xdd, 0xda, + 0xd8, 0xd5, 0xd2, 0xcf, 0xcc, 0xc5, 0xbe, + 0xb7, 0xb0, 0xa8, 0xa0, 0x98, 0x8e, 0x85, + 0x7b, 0x72, 0x69, 0x5e, 0x53, 0x48, 0x3e, + 0x35, 0x2b, 0x22, 0x17, 0x0d, 0x09, 0x07, + 0x05, 0x01, 0x00, 0x26, 0xf0, 0x86, 0x25, + 0x6e, 0xb6, 0xdd, 0xf3, 0xd8, 0xcc, 0x9b, + 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETDISMO, 0xc8); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETPANEL, 0x08); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETGIP0, + 0x81, 0x00, 0x00, 0x00, 0x00, 0x01, 0x00, + 0x04, 0x00, 0x01, 0x13, 0x40, 0x04, 0x09, + 0x09, 0x0b, 0x0b, 0x32, 0x10, 0x08, 0x00, + 0x08, 0x32, 0x10, 0x08, 0x00, 0x08, 0x32, + 0x10, 0x08, 0x00, 0x08, 0x00, 0x00, 0x0a, + 0x08, 0x7b); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0xc5); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN1, 0xf7); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0xd4); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN1, 0x6e); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0xef); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETGIP0, 0x0c); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0xc8); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETGIP0, 0xa1); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETGIP1, + 0x18, 0x18, 0x19, 0x18, 0x18, 0x20, 0x18, + 0x18, 0x18, 0x10, 0x10, 0x18, 0x18, 0x00, + 0x00, 0x18, 0x18, 0x01, 0x01, 0x18, 0x18, + 0x28, 0x28, 0x18, 0x18, 0x18, 0x18, 0x18, + 0x2f, 0x2f, 0x30, 0x30, 0x31, 0x31, 0x35, + 0x35, 0x36, 0x36, 0x37, 0x37, 0x18, 0x18, + 0x18, 0x18, 0x18, 0x18, 0x18, 0x18, 0xfc, + 0xfc, 0x00, 0x00, 0xfc, 0xfc, 0x00, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETGIP2, + 0x18, 0x18, 0x19, 0x18, 0x18, 0x20, 0x19, + 0x18, 0x18, 0x10, 0x10, 0x18, 0x18, 0x00, + 0x00, 0x18, 0x18, 0x01, 0x01, 0x18, 0x18, + 0x28, 0x28, 0x18, 0x18, 0x18, 0x18, 0x18, + 0x2f, 0x2f, 0x30, 0x30, 0x31, 0x31, 0x35, + 0x35, 0x36, 0x36, 0x37, 0x37, 0x18, 0x18, + 0x18, 0x18, 0x18, 0x18, 0x18, 0x18); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETGIP3, + 0xaa, 0xaa, 0xaa, 0xaf, 0xea, 0xaa, 0xaa, + 0xaa, 0xaa, 0xaf, 0xea, 0xaa, 0xaa, 0xaa, + 0xab, 0xaf, 0xef, 0xaa, 0xaa, 0xaa, 0xaa, + 0xaf, 0xea, 0xaa); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETGIP3, + 0xaa, 0xaa, 0xab, 0xaf, 0xea, 0xaa, 0xaa, + 0xaa, 0xae, 0xaf, 0xea, 0xaa); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x02); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETGIP3, + 0xaa, 0xaa, 0xaa, 0xaf, 0xea, 0xaa, 0xaa, + 0xaa, 0xaa, 0xaf, 0xea, 0xaa); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETGIP3, + 0xba, 0xaa, 0xaa, 0xaf, 0xea, 0xaa, 0xaa, + 0xaa, 0xaa, 0xaf, 0xea, 0xaa, 0xba, 0xaa, + 0xaa, 0xaf, 0xea, 0xaa, 0xaa, 0xaa, 0xaa, + 0xaf, 0xea, 0xaa); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0xe4); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN3, 0x17, 0x69); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN3, + 0x09, 0x09, 0x00, 0x07, 0xe8, 0x00, 0x26, + 0x00, 0x07, 0x00, 0x00, 0xe8, 0x32, 0x00, + 0xe9, 0x0a, 0x0a, 0x00, 0x00, 0x00, 0x01, + 0x01, 0x00, 0x12, 0x04); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x01); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN3, + 0x02, 0x00, 0x01, 0x20, 0x01, 0x18, 0x08, + 0xa8, 0x09); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x02); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN3, 0x20, 0x20, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x03); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN3, + 0x00, 0xdc, 0x11, 0x70, 0x00, 0x20); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0xc9); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN3, + 0x2a, 0xce, 0x02, 0x70, 0x01, 0x04); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN4, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_SETBANK, 0x00); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, HX83112B_UNKNOWN2, 0x27); + mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 120); + mipi_dsi_dcs_set_display_on_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 20); + mipi_dsi_dcs_set_display_brightness_multi(&dsi_ctx, 0x0000); + mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_CONTROL_DISPLAY, + 0x24); + mipi_dsi_dcs_set_tear_on_multi(&dsi_ctx, MIPI_DSI_DCS_TEAR_MODE_VBLANK); + + return dsi_ctx.accum_err; +} + +static int hx83112b_off(struct hx83112b_panel *ctx) +{ + struct mipi_dsi_multi_context dsi_ctx = { .dsi = ctx->dsi }; + + mipi_dsi_dcs_set_display_off_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 20); + mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx); + mipi_dsi_msleep(&dsi_ctx, 120); + + return dsi_ctx.accum_err; +} + +static int hx83112b_prepare(struct drm_panel *panel) +{ + struct hx83112b_panel *ctx = to_hx83112b_panel(panel); + struct device *dev = &ctx->dsi->dev; + int ret; + + ret = regulator_bulk_enable(ARRAY_SIZE(hx83112b_supplies), ctx->supplies); + if (ret < 0) { + dev_err(dev, "Failed to enable regulators: %d\n", ret); + return ret; + } + + hx83112b_reset(ctx); + + ret = hx83112b_on(ctx); + if (ret < 0) { + dev_err(dev, "Failed to initialize panel: %d\n", ret); + gpiod_set_value_cansleep(ctx->reset_gpio, 1); + regulator_bulk_disable(ARRAY_SIZE(hx83112b_supplies), ctx->supplies); + return ret; + } + + return 0; +} + +static int hx83112b_unprepare(struct drm_panel *panel) +{ + struct hx83112b_panel *ctx = to_hx83112b_panel(panel); + struct device *dev = &ctx->dsi->dev; + int ret; + + ret = hx83112b_off(ctx); + if (ret < 0) + dev_err(dev, "Failed to un-initialize panel: %d\n", ret); + + gpiod_set_value_cansleep(ctx->reset_gpio, 1); + regulator_bulk_disable(ARRAY_SIZE(hx83112b_supplies), ctx->supplies); + + return 0; +} + +static const struct drm_display_mode hx83112b_mode = { + .clock = (1080 + 40 + 4 + 12) * (2160 + 32 + 2 + 2) * 60 / 1000, + .hdisplay = 1080, + .hsync_start = 1080 + 40, + .hsync_end = 1080 + 40 + 4, + .htotal = 1080 + 40 + 4 + 12, + .vdisplay = 2160, + .vsync_start = 2160 + 32, + .vsync_end = 2160 + 32 + 2, + .vtotal = 2160 + 32 + 2 + 2, + .width_mm = 65, + .height_mm = 128, + .type = DRM_MODE_TYPE_DRIVER, +}; + +static int hx83112b_get_modes(struct drm_panel *panel, + struct drm_connector *connector) +{ + return drm_connector_helper_get_modes_fixed(connector, &hx83112b_mode); +} + +static const struct drm_panel_funcs hx83112b_panel_funcs = { + .prepare = hx83112b_prepare, + .unprepare = hx83112b_unprepare, + .get_modes = hx83112b_get_modes, +}; + +static int hx83112b_bl_update_status(struct backlight_device *bl) +{ + struct mipi_dsi_device *dsi = bl_get_data(bl); + u16 brightness = backlight_get_brightness(bl); + int ret; + + dsi->mode_flags &= ~MIPI_DSI_MODE_LPM; + + ret = mipi_dsi_dcs_set_display_brightness_large(dsi, brightness); + if (ret < 0) + return ret; + + dsi->mode_flags |= MIPI_DSI_MODE_LPM; + + return 0; +} + +static const struct backlight_ops hx83112b_bl_ops = { + .update_status = hx83112b_bl_update_status, +}; + +static struct backlight_device * +hx83112b_create_backlight(struct mipi_dsi_device *dsi) +{ + struct device *dev = &dsi->dev; + const struct backlight_properties props = { + .type = BACKLIGHT_RAW, + .brightness = 4095, + .max_brightness = 4095, + }; + + return devm_backlight_device_register(dev, dev_name(dev), dev, dsi, + &hx83112b_bl_ops, &props); +} + +static int hx83112b_probe(struct mipi_dsi_device *dsi) +{ + struct device *dev = &dsi->dev; + struct hx83112b_panel *ctx; + int ret; + + ctx = devm_drm_panel_alloc(dev, struct hx83112b_panel, panel, + &hx83112b_panel_funcs, + DRM_MODE_CONNECTOR_DSI); + if (IS_ERR(ctx)) + return PTR_ERR(ctx); + + ret = devm_regulator_bulk_get_const(dev, + ARRAY_SIZE(hx83112b_supplies), + hx83112b_supplies, + &ctx->supplies); + if (ret < 0) + return ret; + + ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH); + if (IS_ERR(ctx->reset_gpio)) + return dev_err_probe(dev, PTR_ERR(ctx->reset_gpio), + "Failed to get reset-gpios\n"); + + ctx->dsi = dsi; + mipi_dsi_set_drvdata(dsi, ctx); + + dsi->lanes = 4; + dsi->format = MIPI_DSI_FMT_RGB888; + dsi->mode_flags = MIPI_DSI_MODE_VIDEO_BURST | + MIPI_DSI_CLOCK_NON_CONTINUOUS | + MIPI_DSI_MODE_VIDEO_NO_HSA | MIPI_DSI_MODE_LPM; + + ctx->panel.prepare_prev_first = true; + + ctx->panel.backlight = hx83112b_create_backlight(dsi); + if (IS_ERR(ctx->panel.backlight)) + return dev_err_probe(dev, PTR_ERR(ctx->panel.backlight), + "Failed to create backlight\n"); + + drm_panel_add(&ctx->panel); + + ret = mipi_dsi_attach(dsi); + if (ret < 0) { + drm_panel_remove(&ctx->panel); + return dev_err_probe(dev, ret, "Failed to attach to DSI host\n"); + } + + return 0; +} + +static void hx83112b_remove(struct mipi_dsi_device *dsi) +{ + struct hx83112b_panel *ctx = mipi_dsi_get_drvdata(dsi); + int ret; + + ret = mipi_dsi_detach(dsi); + if (ret < 0) + dev_err(&dsi->dev, "Failed to detach from DSI host: %d\n", ret); + + drm_panel_remove(&ctx->panel); +} + +static const struct of_device_id hx83112b_of_match[] = { + { .compatible = "djn,98-03057-6598b-i" }, + { /* sentinel */ } +}; +MODULE_DEVICE_TABLE(of, hx83112b_of_match); + +static struct mipi_dsi_driver hx83112b_driver = { + .probe = hx83112b_probe, + .remove = hx83112b_remove, + .driver = { + .name = "panel-himax-hx83112b", + .of_match_table = hx83112b_of_match, + }, +}; +module_mipi_dsi_driver(hx83112b_driver); + +MODULE_DESCRIPTION("DRM driver for hx83112b-equipped DSI panels"); +MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/panel/panel-raydium-rm67200.c b/drivers/gpu/drm/panel/panel-raydium-rm67200.c index 1da8225e6f7a..333faed62da7 100644 --- a/drivers/gpu/drm/panel/panel-raydium-rm67200.c +++ b/drivers/gpu/drm/panel/panel-raydium-rm67200.c @@ -36,12 +36,14 @@ static inline struct raydium_rm67200 *to_raydium_rm67200(struct drm_panel *panel static void raydium_rm67200_reset(struct raydium_rm67200 *ctx) { - gpiod_set_value_cansleep(ctx->reset_gpio, 0); - msleep(60); - gpiod_set_value_cansleep(ctx->reset_gpio, 1); - msleep(60); - gpiod_set_value_cansleep(ctx->reset_gpio, 0); - msleep(60); + if (ctx->reset_gpio) { + gpiod_set_value_cansleep(ctx->reset_gpio, 0); + msleep(60); + gpiod_set_value_cansleep(ctx->reset_gpio, 1); + msleep(60); + gpiod_set_value_cansleep(ctx->reset_gpio, 0); + msleep(60); + } } static void raydium_rm67200_write(struct mipi_dsi_multi_context *ctx, @@ -318,6 +320,7 @@ static void w552793baa_setup(struct mipi_dsi_multi_context *ctx) static int raydium_rm67200_prepare(struct drm_panel *panel) { struct raydium_rm67200 *ctx = to_raydium_rm67200(panel); + struct mipi_dsi_multi_context mctx = { .dsi = ctx->dsi }; int ret; ret = regulator_bulk_enable(ctx->num_supplies, ctx->supplies); @@ -328,6 +331,12 @@ static int raydium_rm67200_prepare(struct drm_panel *panel) msleep(60); + ctx->panel_info->panel_setup(&mctx); + mipi_dsi_dcs_exit_sleep_mode_multi(&mctx); + mipi_dsi_msleep(&mctx, 120); + mipi_dsi_dcs_set_display_on_multi(&mctx); + mipi_dsi_msleep(&mctx, 30); + return 0; } @@ -343,20 +352,6 @@ static int raydium_rm67200_unprepare(struct drm_panel *panel) return 0; } -static int raydium_rm67200_enable(struct drm_panel *panel) -{ - struct raydium_rm67200 *rm67200 = to_raydium_rm67200(panel); - struct mipi_dsi_multi_context ctx = { .dsi = rm67200->dsi }; - - rm67200->panel_info->panel_setup(&ctx); - mipi_dsi_dcs_exit_sleep_mode_multi(&ctx); - mipi_dsi_msleep(&ctx, 120); - mipi_dsi_dcs_set_display_on_multi(&ctx); - mipi_dsi_msleep(&ctx, 30); - - return ctx.accum_err; -} - static int raydium_rm67200_disable(struct drm_panel *panel) { struct raydium_rm67200 *rm67200 = to_raydium_rm67200(panel); @@ -381,7 +376,6 @@ static const struct drm_panel_funcs raydium_rm67200_funcs = { .prepare = raydium_rm67200_prepare, .unprepare = raydium_rm67200_unprepare, .get_modes = raydium_rm67200_get_modes, - .enable = raydium_rm67200_enable, .disable = raydium_rm67200_disable, }; @@ -409,7 +403,7 @@ static int raydium_rm67200_probe(struct mipi_dsi_device *dsi) if (ret < 0) return ret; - ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_LOW); + ctx->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW); if (IS_ERR(ctx->reset_gpio)) return dev_err_probe(dev, PTR_ERR(ctx->reset_gpio), "Failed to get reset-gpios\n"); @@ -470,6 +464,7 @@ static const struct raydium_rm67200_panel_info w552793baa_info = { .vtotal = 1952, .width_mm = 68, /* 68.04mm */ .height_mm = 121, /* 120.96mm */ + .flags = DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_NHSYNC, .type = DRM_MODE_TYPE_DRIVER, }, .regulators = w552793baa_regulators, diff --git a/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c b/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c index b5b9e80690f6..692020081524 100644 --- a/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c +++ b/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c @@ -244,7 +244,7 @@ static const struct s6d7aa0_panel_desc s6d7aa0_lsl080al02_desc = { .init_func = s6d7aa0_lsl080al02_init, .off_func = s6d7aa0_lsl080al02_off, .drm_mode = &s6d7aa0_lsl080al02_mode, - .mode_flags = MIPI_DSI_MODE_VSYNC_FLUSH | MIPI_DSI_MODE_VIDEO_NO_HFP, + .mode_flags = MIPI_DSI_MODE_VIDEO_NO_HFP, .bus_flags = 0, .has_backlight = false, diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e8aa0.c b/drivers/gpu/drm/panel/panel-samsung-s6e8aa0.c index 897df195f2f3..1b5c500d4f4e 100644 --- a/drivers/gpu/drm/panel/panel-samsung-s6e8aa0.c +++ b/drivers/gpu/drm/panel/panel-samsung-s6e8aa0.c @@ -992,7 +992,7 @@ static int s6e8aa0_probe(struct mipi_dsi_device *dsi) dsi->lanes = 4; dsi->format = MIPI_DSI_FMT_RGB888; dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST - | MIPI_DSI_MODE_VSYNC_FLUSH | MIPI_DSI_MODE_VIDEO_AUTO_VERT; + | MIPI_DSI_MODE_VIDEO_AUTO_VERT; ret = s6e8aa0_parse_dt(ctx); if (ret < 0) diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c index 842239b43cdf..1ea6c509a5d5 100644 --- a/drivers/gpu/drm/panfrost/panfrost_drv.c +++ b/drivers/gpu/drm/panfrost/panfrost_drv.c @@ -866,6 +866,8 @@ static const struct panfrost_compatible amlogic_data = { .vendor_quirk = panfrost_gpu_amlogic_quirk, }; +static const char * const mediatek_pm_domains[] = { "core0", "core1", "core2", + "core3", "core4" }; /* * The old data with two power supplies for MT8183 is here only to * keep retro-compatibility with older devicetrees, as DVFS will @@ -874,51 +876,53 @@ static const struct panfrost_compatible amlogic_data = { * On new devicetrees please use the _b variant with a single and * coupled regulators instead. */ -static const char * const mediatek_mt8183_supplies[] = { "mali", "sram", NULL }; -static const char * const mediatek_mt8183_pm_domains[] = { "core0", "core1", "core2" }; +static const char * const legacy_supplies[] = { "mali", "sram", NULL }; static const struct panfrost_compatible mediatek_mt8183_data = { - .num_supplies = ARRAY_SIZE(mediatek_mt8183_supplies) - 1, - .supply_names = mediatek_mt8183_supplies, - .num_pm_domains = ARRAY_SIZE(mediatek_mt8183_pm_domains), - .pm_domain_names = mediatek_mt8183_pm_domains, + .num_supplies = ARRAY_SIZE(legacy_supplies) - 1, + .supply_names = legacy_supplies, + .num_pm_domains = 3, + .pm_domain_names = mediatek_pm_domains, }; -static const char * const mediatek_mt8183_b_supplies[] = { "mali", NULL }; static const struct panfrost_compatible mediatek_mt8183_b_data = { - .num_supplies = ARRAY_SIZE(mediatek_mt8183_b_supplies) - 1, - .supply_names = mediatek_mt8183_b_supplies, - .num_pm_domains = ARRAY_SIZE(mediatek_mt8183_pm_domains), - .pm_domain_names = mediatek_mt8183_pm_domains, + .num_supplies = ARRAY_SIZE(default_supplies) - 1, + .supply_names = default_supplies, + .num_pm_domains = 3, + .pm_domain_names = mediatek_pm_domains, .pm_features = BIT(GPU_PM_CLK_DIS) | BIT(GPU_PM_VREG_OFF), }; -static const char * const mediatek_mt8186_pm_domains[] = { "core0", "core1" }; static const struct panfrost_compatible mediatek_mt8186_data = { - .num_supplies = ARRAY_SIZE(mediatek_mt8183_b_supplies) - 1, - .supply_names = mediatek_mt8183_b_supplies, - .num_pm_domains = ARRAY_SIZE(mediatek_mt8186_pm_domains), - .pm_domain_names = mediatek_mt8186_pm_domains, + .num_supplies = ARRAY_SIZE(default_supplies) - 1, + .supply_names = default_supplies, + .num_pm_domains = 2, + .pm_domain_names = mediatek_pm_domains, .pm_features = BIT(GPU_PM_CLK_DIS) | BIT(GPU_PM_VREG_OFF), }; -/* MT8188 uses the same power domains and power supplies as MT8183 */ static const struct panfrost_compatible mediatek_mt8188_data = { - .num_supplies = ARRAY_SIZE(mediatek_mt8183_b_supplies) - 1, - .supply_names = mediatek_mt8183_b_supplies, - .num_pm_domains = ARRAY_SIZE(mediatek_mt8183_pm_domains), - .pm_domain_names = mediatek_mt8183_pm_domains, + .num_supplies = ARRAY_SIZE(default_supplies) - 1, + .supply_names = default_supplies, + .num_pm_domains = 3, + .pm_domain_names = mediatek_pm_domains, .pm_features = BIT(GPU_PM_CLK_DIS) | BIT(GPU_PM_VREG_OFF), .gpu_quirks = BIT(GPU_QUIRK_FORCE_AARCH64_PGTABLE), }; -static const char * const mediatek_mt8192_supplies[] = { "mali", NULL }; -static const char * const mediatek_mt8192_pm_domains[] = { "core0", "core1", "core2", - "core3", "core4" }; static const struct panfrost_compatible mediatek_mt8192_data = { - .num_supplies = ARRAY_SIZE(mediatek_mt8192_supplies) - 1, - .supply_names = mediatek_mt8192_supplies, - .num_pm_domains = ARRAY_SIZE(mediatek_mt8192_pm_domains), - .pm_domain_names = mediatek_mt8192_pm_domains, + .num_supplies = ARRAY_SIZE(default_supplies) - 1, + .supply_names = default_supplies, + .num_pm_domains = 5, + .pm_domain_names = mediatek_pm_domains, + .pm_features = BIT(GPU_PM_CLK_DIS) | BIT(GPU_PM_VREG_OFF), + .gpu_quirks = BIT(GPU_QUIRK_FORCE_AARCH64_PGTABLE), +}; + +static const struct panfrost_compatible mediatek_mt8370_data = { + .num_supplies = ARRAY_SIZE(default_supplies) - 1, + .supply_names = default_supplies, + .num_pm_domains = 2, + .pm_domain_names = mediatek_pm_domains, .pm_features = BIT(GPU_PM_CLK_DIS) | BIT(GPU_PM_VREG_OFF), .gpu_quirks = BIT(GPU_QUIRK_FORCE_AARCH64_PGTABLE), }; @@ -945,6 +949,7 @@ static const struct of_device_id dt_match[] = { { .compatible = "mediatek,mt8186-mali", .data = &mediatek_mt8186_data }, { .compatible = "mediatek,mt8188-mali", .data = &mediatek_mt8188_data }, { .compatible = "mediatek,mt8192-mali", .data = &mediatek_mt8192_data }, + { .compatible = "mediatek,mt8370-mali", .data = &mediatek_mt8370_data }, { .compatible = "allwinner,sun50i-h616-mali", .data = &allwinner_h616_data }, {} }; diff --git a/drivers/gpu/drm/panthor/panthor_gem.c b/drivers/gpu/drm/panthor/panthor_gem.c index 7c00fd77758b..a123bc740ba1 100644 --- a/drivers/gpu/drm/panthor/panthor_gem.c +++ b/drivers/gpu/drm/panthor/panthor_gem.c @@ -16,10 +16,15 @@ #include "panthor_mmu.h" #ifdef CONFIG_DEBUG_FS -static void panthor_gem_debugfs_bo_add(struct panthor_device *ptdev, - struct panthor_gem_object *bo) +static void panthor_gem_debugfs_bo_init(struct panthor_gem_object *bo) { INIT_LIST_HEAD(&bo->debugfs.node); +} + +static void panthor_gem_debugfs_bo_add(struct panthor_gem_object *bo) +{ + struct panthor_device *ptdev = container_of(bo->base.base.dev, + struct panthor_device, base); bo->debugfs.creator.tgid = current->group_leader->pid; get_task_comm(bo->debugfs.creator.process_name, current->group_leader); @@ -44,14 +49,13 @@ static void panthor_gem_debugfs_bo_rm(struct panthor_gem_object *bo) static void panthor_gem_debugfs_set_usage_flags(struct panthor_gem_object *bo, u32 usage_flags) { - bo->debugfs.flags = usage_flags | PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED; + bo->debugfs.flags = usage_flags; + panthor_gem_debugfs_bo_add(bo); } #else -static void panthor_gem_debugfs_bo_add(struct panthor_device *ptdev, - struct panthor_gem_object *bo) -{} static void panthor_gem_debugfs_bo_rm(struct panthor_gem_object *bo) {} static void panthor_gem_debugfs_set_usage_flags(struct panthor_gem_object *bo, u32 usage_flags) {} +static void panthor_gem_debugfs_bo_init(struct panthor_gem_object *bo) {} #endif static void panthor_gem_free_object(struct drm_gem_object *obj) @@ -246,7 +250,7 @@ struct drm_gem_object *panthor_gem_create_object(struct drm_device *ddev, size_t drm_gem_gpuva_set_lock(&obj->base.base, &obj->gpuva_list_lock); mutex_init(&obj->label.lock); - panthor_gem_debugfs_bo_add(ptdev, obj); + panthor_gem_debugfs_bo_init(obj); return &obj->base.base; } @@ -285,6 +289,8 @@ panthor_gem_create_with_handle(struct drm_file *file, bo->base.base.resv = bo->exclusive_vm_root_gem->resv; } + panthor_gem_debugfs_set_usage_flags(bo, 0); + /* * Allocate an id of idr table where the obj is registered * and handle has the id what user can see. @@ -296,12 +302,6 @@ panthor_gem_create_with_handle(struct drm_file *file, /* drop reference from allocate - handle holds it now. */ drm_gem_object_put(&shmem->base); - /* - * No explicit flags are needed in the call below, since the - * function internally sets the INITIALIZED bit for us. - */ - panthor_gem_debugfs_set_usage_flags(bo, 0); - return ret; } @@ -387,7 +387,7 @@ static void panthor_gem_debugfs_bo_print(struct panthor_gem_object *bo, unsigned int refcount = kref_read(&bo->base.base.refcount); char creator_info[32] = {}; size_t resident_size; - u32 gem_usage_flags = bo->debugfs.flags & (u32)~PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED; + u32 gem_usage_flags = bo->debugfs.flags; u32 gem_state_flags = 0; /* Skip BOs being destroyed. */ @@ -436,8 +436,7 @@ void panthor_gem_debugfs_print_bos(struct panthor_device *ptdev, scoped_guard(mutex, &ptdev->gems.lock) { list_for_each_entry(bo, &ptdev->gems.node, debugfs.node) { - if (bo->debugfs.flags & PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED) - panthor_gem_debugfs_bo_print(bo, m, &totals); + panthor_gem_debugfs_bo_print(bo, m, &totals); } } diff --git a/drivers/gpu/drm/panthor/panthor_gem.h b/drivers/gpu/drm/panthor/panthor_gem.h index 4dd732dcd59f..8fc7215e9b90 100644 --- a/drivers/gpu/drm/panthor/panthor_gem.h +++ b/drivers/gpu/drm/panthor/panthor_gem.h @@ -35,9 +35,6 @@ enum panthor_debugfs_gem_usage_flags { /** @PANTHOR_DEBUGFS_GEM_USAGE_FLAG_FW_MAPPED: BO is mapped on the FW VM. */ PANTHOR_DEBUGFS_GEM_USAGE_FLAG_FW_MAPPED = BIT(PANTHOR_DEBUGFS_GEM_USAGE_FW_MAPPED_BIT), - - /** @PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED: BO is ready for DebugFS display. */ - PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED = BIT(31), }; /** diff --git a/drivers/gpu/drm/panthor/panthor_gpu.c b/drivers/gpu/drm/panthor/panthor_gpu.c index 534735518824..cb7a335e07d7 100644 --- a/drivers/gpu/drm/panthor/panthor_gpu.c +++ b/drivers/gpu/drm/panthor/panthor_gpu.c @@ -297,8 +297,9 @@ int panthor_gpu_block_power_on(struct panthor_device *ptdev, gpu_write64(ptdev, pwron_reg, mask); - ret = gpu_read64_relaxed_poll_timeout(ptdev, pwrtrans_reg, val, - !(mask & val), 100, timeout_us); + ret = gpu_read64_relaxed_poll_timeout(ptdev, rdy_reg, val, + (mask & val) == val, + 100, timeout_us); if (ret) { drm_err(&ptdev->base, "timeout waiting on %s:%llx readiness", blk_name, mask); diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c index a46613283393..266c57733136 100644 --- a/drivers/gpu/drm/radeon/evergreen_cs.c +++ b/drivers/gpu/drm/radeon/evergreen_cs.c @@ -211,7 +211,7 @@ static int evergreen_surface_check_linear_aligned(struct radeon_cs_parser *p, surf->base_align = track->group_size; surf->palign = palign; surf->halign = 1; - if (surf->nbx & (palign - 1)) { + if ((surf->nbx & (palign - 1)) && !(palign == 64 && surf->nbx == 32)) { if (prefix) { dev_warn(p->dev, "%s:%d %s pitch %d invalid must be aligned with %d\n", __func__, __LINE__, prefix, surf->nbx, palign); @@ -2661,6 +2661,95 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, } break; } + case PACKET3_COND_EXEC: + { + u64 offset; + + if (pkt->count != 2) { + DRM_ERROR("bad COND_EXEC (invalid count)\n"); + return -EINVAL; + } + r = radeon_cs_packet_next_reloc(p, &reloc, 0); + if (r) { + DRM_ERROR("bad COND_EXEC (missing reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx + 0); + offset += ((u64)(radeon_get_ib_value(p, idx + 1) & 0xff)) << 32UL; + if (offset & 0x7) { + DRM_ERROR("bad COND_EXEC (address not qwords aligned)\n"); + return -EINVAL; + } + if ((offset + 8) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad COND_EXEC bo too small: 0x%llx, 0x%lx\n", + offset + 8, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->gpu_offset; + ib[idx + 0] = offset; + ib[idx + 1] = upper_32_bits(offset) & 0xff; + break; + } + case PACKET3_COND_WRITE: + if (pkt->count != 7) { + DRM_ERROR("bad COND_WRITE (invalid count)\n"); + return -EINVAL; + } + if (idx_value & 0x10) { + u64 offset; + /* POLL is memory. */ + r = radeon_cs_packet_next_reloc(p, &reloc, 0); + if (r) { + DRM_ERROR("bad COND_WRITE (missing src reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx + 1); + offset += ((u64)(radeon_get_ib_value(p, idx + 2) & 0xff)) << 32; + if ((offset + 8) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad COND_WRITE src bo too small: 0x%llx, 0x%lx\n", + offset + 8, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->gpu_offset; + ib[idx + 1] = offset; + ib[idx + 2] = upper_32_bits(offset) & 0xff; + } else { + /* POLL is a reg. */ + reg = radeon_get_ib_value(p, idx + 1) << 2; + if (!evergreen_is_safe_reg(p, reg)) { + dev_warn(p->dev, "forbidden register 0x%08x at %d\n", + reg, idx + 1); + return -EINVAL; + } + } + if (idx_value & 0x100) { + u64 offset; + /* WRITE is memory. */ + r = radeon_cs_packet_next_reloc(p, &reloc, 0); + if (r) { + DRM_ERROR("bad COND_WRITE (missing dst reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx + 5); + offset += ((u64)(radeon_get_ib_value(p, idx + 6) & 0xff)) << 32; + if ((offset + 8) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad COND_WRITE dst bo too small: 0x%llx, 0x%lx\n", + offset + 8, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->gpu_offset; + ib[idx + 5] = offset; + ib[idx + 6] = upper_32_bits(offset) & 0xff; + } else { + /* WRITE is a reg. */ + reg = radeon_get_ib_value(p, idx + 5) << 2; + if (!evergreen_is_safe_reg(p, reg)) { + dev_warn(p->dev, "forbidden register 0x%08x at %d\n", + reg, idx + 5); + return -EINVAL; + } + } + break; case PACKET3_NOP: break; default: @@ -3406,7 +3495,12 @@ static int evergreen_vm_packet3_check(struct radeon_device *rdev, case CAYMAN_PACKET3_DEALLOC_STATE: break; case PACKET3_COND_WRITE: - if (idx_value & 0x100) { + if (!(idx_value & 0x10)) { + reg = ib[idx + 1] * 4; + if (!evergreen_vm_reg_valid(reg)) + return -EINVAL; + } + if (!(idx_value & 0x100)) { reg = ib[idx + 5] * 4; if (!evergreen_vm_reg_valid(reg)) return -EINVAL; diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index 267f082bc430..88e821d67af7 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -110,9 +110,10 @@ * 2.48.0 - TA_CS_BC_BASE_ADDR allowed on SI * 2.49.0 - DRM_RADEON_GEM_INFO ioctl returns correct vram_size/visible values * 2.50.0 - Allows unaligned shader loads on CIK. (needed by OpenGL) + * 2.51.0 - Add evergreen/cayman OpenGL 4.6 compatibility */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 50 +#define KMS_DRIVER_MINOR 51 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_no_wb; diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.c b/drivers/gpu/drm/rockchip/cdn-dp-core.c index 292c31de18f1..24f6b3879f4b 100644 --- a/drivers/gpu/drm/rockchip/cdn-dp-core.c +++ b/drivers/gpu/drm/rockchip/cdn-dp-core.c @@ -16,7 +16,9 @@ #include <sound/hdmi-codec.h> #include <drm/display/drm_dp_helper.h> +#include <drm/display/drm_hdmi_audio_helper.h> #include <drm/drm_atomic_helper.h> +#include <drm/drm_bridge_connector.h> #include <drm/drm_edid.h> #include <drm/drm_of.h> #include <drm/drm_probe_helper.h> @@ -25,9 +27,9 @@ #include "cdn-dp-core.h" #include "cdn-dp-reg.h" -static inline struct cdn_dp_device *connector_to_dp(struct drm_connector *connector) +static inline struct cdn_dp_device *bridge_to_dp(struct drm_bridge *bridge) { - return container_of(connector, struct cdn_dp_device, connector); + return container_of(bridge, struct cdn_dp_device, bridge); } static inline struct cdn_dp_device *encoder_to_dp(struct drm_encoder *encoder) @@ -231,9 +233,9 @@ static bool cdn_dp_check_sink_connection(struct cdn_dp_device *dp) } static enum drm_connector_status -cdn_dp_connector_detect(struct drm_connector *connector, bool force) +cdn_dp_bridge_detect(struct drm_bridge *bridge) { - struct cdn_dp_device *dp = connector_to_dp(connector); + struct cdn_dp_device *dp = bridge_to_dp(bridge); enum drm_connector_status status = connector_status_disconnected; mutex_lock(&dp->lock); @@ -244,41 +246,25 @@ cdn_dp_connector_detect(struct drm_connector *connector, bool force) return status; } -static void cdn_dp_connector_destroy(struct drm_connector *connector) +static const struct drm_edid * +cdn_dp_bridge_edid_read(struct drm_bridge *bridge, struct drm_connector *connector) { - drm_connector_unregister(connector); - drm_connector_cleanup(connector); -} - -static const struct drm_connector_funcs cdn_dp_atomic_connector_funcs = { - .detect = cdn_dp_connector_detect, - .destroy = cdn_dp_connector_destroy, - .fill_modes = drm_helper_probe_single_connector_modes, - .reset = drm_atomic_helper_connector_reset, - .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, - .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, -}; - -static int cdn_dp_connector_get_modes(struct drm_connector *connector) -{ - struct cdn_dp_device *dp = connector_to_dp(connector); - int ret = 0; + struct cdn_dp_device *dp = bridge_to_dp(bridge); + const struct drm_edid *drm_edid; mutex_lock(&dp->lock); - - ret = drm_edid_connector_add_modes(connector); - + drm_edid = drm_edid_read_custom(connector, cdn_dp_get_edid_block, dp); mutex_unlock(&dp->lock); - return ret; + return drm_edid; } static enum drm_mode_status -cdn_dp_connector_mode_valid(struct drm_connector *connector, - const struct drm_display_mode *mode) +cdn_dp_bridge_mode_valid(struct drm_bridge *bridge, + const struct drm_display_info *display_info, + const struct drm_display_mode *mode) { - struct cdn_dp_device *dp = connector_to_dp(connector); - struct drm_display_info *display_info = &dp->connector.display_info; + struct cdn_dp_device *dp = bridge_to_dp(bridge); u32 requested, actual, rate, sink_max, source_max = 0; u8 lanes, bpc; @@ -323,11 +309,6 @@ cdn_dp_connector_mode_valid(struct drm_connector *connector, return MODE_OK; } -static struct drm_connector_helper_funcs cdn_dp_connector_helper_funcs = { - .get_modes = cdn_dp_connector_get_modes, - .mode_valid = cdn_dp_connector_mode_valid, -}; - static int cdn_dp_firmware_init(struct cdn_dp_device *dp) { int ret; @@ -360,7 +341,6 @@ static int cdn_dp_firmware_init(struct cdn_dp_device *dp) static int cdn_dp_get_sink_capability(struct cdn_dp_device *dp) { - const struct drm_display_info *info = &dp->connector.display_info; int ret; if (!cdn_dp_check_sink_connection(dp)) @@ -373,17 +353,6 @@ static int cdn_dp_get_sink_capability(struct cdn_dp_device *dp) return ret; } - drm_edid_free(dp->drm_edid); - dp->drm_edid = drm_edid_read_custom(&dp->connector, - cdn_dp_get_edid_block, dp); - drm_edid_connector_update(&dp->connector, dp->drm_edid); - - dp->sink_has_audio = info->has_audio; - - if (dp->drm_edid) - DRM_DEV_DEBUG_KMS(dp->dev, "got edid: width[%d] x height[%d]\n", - info->width_mm / 10, info->height_mm / 10); - return 0; } @@ -488,10 +457,6 @@ static int cdn_dp_disable(struct cdn_dp_device *dp) dp->active = false; dp->max_lanes = 0; dp->max_rate = 0; - if (!dp->connected) { - drm_edid_free(dp->drm_edid); - dp->drm_edid = NULL; - } return 0; } @@ -546,26 +511,13 @@ err_clk_disable: return ret; } -static void cdn_dp_encoder_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode, - struct drm_display_mode *adjusted) +static void cdn_dp_bridge_mode_set(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + const struct drm_display_mode *adjusted) { - struct cdn_dp_device *dp = encoder_to_dp(encoder); - struct drm_display_info *display_info = &dp->connector.display_info; + struct cdn_dp_device *dp = bridge_to_dp(bridge); struct video_info *video = &dp->video_info; - switch (display_info->bpc) { - case 10: - video->color_depth = 10; - break; - case 6: - video->color_depth = 6; - break; - default: - video->color_depth = 8; - break; - } - video->color_fmt = PXL_RGB; video->v_sync_polarity = !!(mode->flags & DRM_MODE_FLAG_NVSYNC); video->h_sync_polarity = !!(mode->flags & DRM_MODE_FLAG_NHSYNC); @@ -592,19 +544,37 @@ static bool cdn_dp_check_link_status(struct cdn_dp_device *dp) return drm_dp_channel_eq_ok(link_status, min(port->lanes, sink_lanes)); } -static void cdn_dp_audio_handle_plugged_change(struct cdn_dp_device *dp, - bool plugged) +static void cdn_dp_display_info_update(struct cdn_dp_device *dp, + struct drm_display_info *display_info) { - if (dp->codec_dev) - dp->plugged_cb(dp->codec_dev, plugged); + struct video_info *video = &dp->video_info; + + switch (display_info->bpc) { + case 10: + video->color_depth = 10; + break; + case 6: + video->color_depth = 6; + break; + default: + video->color_depth = 8; + break; + } } -static void cdn_dp_encoder_enable(struct drm_encoder *encoder) +static void cdn_dp_bridge_atomic_enable(struct drm_bridge *bridge, struct drm_atomic_state *state) { - struct cdn_dp_device *dp = encoder_to_dp(encoder); + struct cdn_dp_device *dp = bridge_to_dp(bridge); + struct drm_connector *connector; int ret, val; - ret = drm_of_encoder_active_endpoint_id(dp->dev->of_node, encoder); + connector = drm_atomic_get_new_connector_for_encoder(state, bridge->encoder); + if (!connector) + return; + + cdn_dp_display_info_update(dp, &connector->display_info); + + ret = drm_of_encoder_active_endpoint_id(dp->dev->of_node, &dp->encoder.encoder); if (ret < 0) { DRM_DEV_ERROR(dp->dev, "Could not get vop id, %d", ret); return; @@ -625,7 +595,7 @@ static void cdn_dp_encoder_enable(struct drm_encoder *encoder) ret = cdn_dp_enable(dp); if (ret) { - DRM_DEV_ERROR(dp->dev, "Failed to enable encoder %d\n", + DRM_DEV_ERROR(dp->dev, "Failed to enable bridge %d\n", ret); goto out; } @@ -655,24 +625,21 @@ static void cdn_dp_encoder_enable(struct drm_encoder *encoder) goto out; } - cdn_dp_audio_handle_plugged_change(dp, true); - out: mutex_unlock(&dp->lock); } -static void cdn_dp_encoder_disable(struct drm_encoder *encoder) +static void cdn_dp_bridge_atomic_disable(struct drm_bridge *bridge, struct drm_atomic_state *state) { - struct cdn_dp_device *dp = encoder_to_dp(encoder); + struct cdn_dp_device *dp = bridge_to_dp(bridge); int ret; mutex_lock(&dp->lock); - cdn_dp_audio_handle_plugged_change(dp, false); if (dp->active) { ret = cdn_dp_disable(dp); if (ret) { - DRM_DEV_ERROR(dp->dev, "Failed to disable encoder %d\n", + DRM_DEV_ERROR(dp->dev, "Failed to disable bridge %d\n", ret); } } @@ -704,9 +671,6 @@ static int cdn_dp_encoder_atomic_check(struct drm_encoder *encoder, } static const struct drm_encoder_helper_funcs cdn_dp_encoder_helper_funcs = { - .mode_set = cdn_dp_encoder_mode_set, - .enable = cdn_dp_encoder_enable, - .disable = cdn_dp_encoder_disable, .atomic_check = cdn_dp_encoder_atomic_check, }; @@ -779,11 +743,12 @@ static int cdn_dp_parse_dt(struct cdn_dp_device *dp) return 0; } -static int cdn_dp_audio_hw_params(struct device *dev, void *data, - struct hdmi_codec_daifmt *daifmt, - struct hdmi_codec_params *params) +static int cdn_dp_audio_prepare(struct drm_connector *connector, + struct drm_bridge *bridge, + struct hdmi_codec_daifmt *daifmt, + struct hdmi_codec_params *params) { - struct cdn_dp_device *dp = dev_get_drvdata(dev); + struct cdn_dp_device *dp = bridge_to_dp(bridge); struct audio_info audio = { .sample_width = params->sample_width, .sample_rate = params->sample_rate, @@ -805,7 +770,7 @@ static int cdn_dp_audio_hw_params(struct device *dev, void *data, audio.format = AFMT_SPDIF; break; default: - DRM_DEV_ERROR(dev, "Invalid format %d\n", daifmt->fmt); + drm_err(bridge->dev, "Invalid format %d\n", daifmt->fmt); ret = -EINVAL; goto out; } @@ -819,9 +784,10 @@ out: return ret; } -static void cdn_dp_audio_shutdown(struct device *dev, void *data) +static void cdn_dp_audio_shutdown(struct drm_connector *connector, + struct drm_bridge *bridge) { - struct cdn_dp_device *dp = dev_get_drvdata(dev); + struct cdn_dp_device *dp = bridge_to_dp(bridge); int ret; mutex_lock(&dp->lock); @@ -835,10 +801,11 @@ out: mutex_unlock(&dp->lock); } -static int cdn_dp_audio_mute_stream(struct device *dev, void *data, +static int cdn_dp_audio_mute_stream(struct drm_connector *connector, + struct drm_bridge *bridge, bool enable, int direction) { - struct cdn_dp_device *dp = dev_get_drvdata(dev); + struct cdn_dp_device *dp = bridge_to_dp(bridge); int ret; mutex_lock(&dp->lock); @@ -854,57 +821,22 @@ out: return ret; } -static int cdn_dp_audio_get_eld(struct device *dev, void *data, - u8 *buf, size_t len) -{ - struct cdn_dp_device *dp = dev_get_drvdata(dev); - - memcpy(buf, dp->connector.eld, min(sizeof(dp->connector.eld), len)); - - return 0; -} - -static int cdn_dp_audio_hook_plugged_cb(struct device *dev, void *data, - hdmi_codec_plugged_cb fn, - struct device *codec_dev) -{ - struct cdn_dp_device *dp = dev_get_drvdata(dev); - - mutex_lock(&dp->lock); - dp->plugged_cb = fn; - dp->codec_dev = codec_dev; - cdn_dp_audio_handle_plugged_change(dp, dp->connected); - mutex_unlock(&dp->lock); - - return 0; -} - -static const struct hdmi_codec_ops audio_codec_ops = { - .hw_params = cdn_dp_audio_hw_params, - .audio_shutdown = cdn_dp_audio_shutdown, - .mute_stream = cdn_dp_audio_mute_stream, - .get_eld = cdn_dp_audio_get_eld, - .hook_plugged_cb = cdn_dp_audio_hook_plugged_cb, +static const struct drm_bridge_funcs cdn_dp_bridge_funcs = { + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, + .detect = cdn_dp_bridge_detect, + .edid_read = cdn_dp_bridge_edid_read, + .atomic_enable = cdn_dp_bridge_atomic_enable, + .atomic_disable = cdn_dp_bridge_atomic_disable, + .mode_valid = cdn_dp_bridge_mode_valid, + .mode_set = cdn_dp_bridge_mode_set, + + .dp_audio_prepare = cdn_dp_audio_prepare, + .dp_audio_mute_stream = cdn_dp_audio_mute_stream, + .dp_audio_shutdown = cdn_dp_audio_shutdown, }; -static int cdn_dp_audio_codec_init(struct cdn_dp_device *dp, - struct device *dev) -{ - struct hdmi_codec_pdata codec_data = { - .i2s = 1, - .spdif = 1, - .ops = &audio_codec_ops, - .max_i2s_channels = 8, - .no_capture_mute = 1, - }; - - dp->audio_pdev = platform_device_register_data( - dev, HDMI_CODEC_DRV_NAME, PLATFORM_DEVID_AUTO, - &codec_data, sizeof(codec_data)); - - return PTR_ERR_OR_ZERO(dp->audio_pdev); -} - static int cdn_dp_request_firmware(struct cdn_dp_device *dp) { int ret; @@ -1006,7 +938,9 @@ static void cdn_dp_pd_event_work(struct work_struct *work) out: mutex_unlock(&dp->lock); - drm_connector_helper_hpd_irq_event(&dp->connector); + drm_bridge_hpd_notify(&dp->bridge, + dp->connected ? connector_status_connected + : connector_status_disconnected); } static int cdn_dp_pd_event(struct notifier_block *nb, @@ -1062,26 +996,35 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data) drm_encoder_helper_add(encoder, &cdn_dp_encoder_helper_funcs); - connector = &dp->connector; - connector->polled = DRM_CONNECTOR_POLL_HPD; - connector->dpms = DRM_MODE_DPMS_OFF; - - ret = drm_connector_init(drm_dev, connector, - &cdn_dp_atomic_connector_funcs, - DRM_MODE_CONNECTOR_DisplayPort); - if (ret) { - DRM_ERROR("failed to initialize connector with drm\n"); - goto err_free_encoder; - } + dp->bridge.ops = + DRM_BRIDGE_OP_DETECT | + DRM_BRIDGE_OP_EDID | + DRM_BRIDGE_OP_HPD | + DRM_BRIDGE_OP_DP_AUDIO; + dp->bridge.of_node = dp->dev->of_node; + dp->bridge.type = DRM_MODE_CONNECTOR_DisplayPort; + dp->bridge.hdmi_audio_dev = dp->dev; + dp->bridge.hdmi_audio_max_i2s_playback_channels = 8; + dp->bridge.hdmi_audio_spdif_playback = 1; + dp->bridge.hdmi_audio_dai_port = -1; + + ret = devm_drm_bridge_add(dev, &dp->bridge); + if (ret) + return ret; - drm_connector_helper_add(connector, &cdn_dp_connector_helper_funcs); + ret = drm_bridge_attach(encoder, &dp->bridge, NULL, DRM_BRIDGE_ATTACH_NO_CONNECTOR); + if (ret) + return ret; - ret = drm_connector_attach_encoder(connector, encoder); - if (ret) { - DRM_ERROR("failed to attach connector and encoder\n"); - goto err_free_connector; + connector = drm_bridge_connector_init(drm_dev, encoder); + if (IS_ERR(connector)) { + ret = PTR_ERR(connector); + dev_err(dp->dev, "failed to init bridge connector: %d\n", ret); + return ret; } + drm_connector_attach_encoder(connector, encoder); + for (i = 0; i < dp->ports; i++) { port = dp->port[i]; @@ -1092,7 +1035,7 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data) if (ret) { DRM_DEV_ERROR(dev, "register EXTCON_DISP_DP notifier err\n"); - goto err_free_connector; + return ret; } } @@ -1101,30 +1044,19 @@ static int cdn_dp_bind(struct device *dev, struct device *master, void *data) schedule_work(&dp->event_work); return 0; - -err_free_connector: - drm_connector_cleanup(connector); -err_free_encoder: - drm_encoder_cleanup(encoder); - return ret; } static void cdn_dp_unbind(struct device *dev, struct device *master, void *data) { struct cdn_dp_device *dp = dev_get_drvdata(dev); struct drm_encoder *encoder = &dp->encoder.encoder; - struct drm_connector *connector = &dp->connector; cancel_work_sync(&dp->event_work); - cdn_dp_encoder_disable(encoder); encoder->funcs->destroy(encoder); - connector->funcs->destroy(connector); pm_runtime_disable(dev); if (dp->fw_loaded) release_firmware(dp->fw); - drm_edid_free(dp->drm_edid); - dp->drm_edid = NULL; } static const struct component_ops cdn_dp_component_ops = { @@ -1171,9 +1103,10 @@ static int cdn_dp_probe(struct platform_device *pdev) int ret; int i; - dp = devm_kzalloc(dev, sizeof(*dp), GFP_KERNEL); - if (!dp) - return -ENOMEM; + dp = devm_drm_bridge_alloc(dev, struct cdn_dp_device, bridge, + &cdn_dp_bridge_funcs); + if (IS_ERR(dp)) + return PTR_ERR(dp); dp->dev = dev; match = of_match_node(cdn_dp_dt_ids, pdev->dev.of_node); @@ -1209,19 +1142,11 @@ static int cdn_dp_probe(struct platform_device *pdev) mutex_init(&dp->lock); dev_set_drvdata(dev, dp); - ret = cdn_dp_audio_codec_init(dp, dev); - if (ret) - return ret; - ret = component_add(dev, &cdn_dp_component_ops); if (ret) - goto err_audio_deinit; + return ret; return 0; - -err_audio_deinit: - platform_device_unregister(dp->audio_pdev); - return ret; } static void cdn_dp_remove(struct platform_device *pdev) diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.h b/drivers/gpu/drm/rockchip/cdn-dp-core.h index 17498f576ce7..e9c30b9fd543 100644 --- a/drivers/gpu/drm/rockchip/cdn-dp-core.h +++ b/drivers/gpu/drm/rockchip/cdn-dp-core.h @@ -8,6 +8,7 @@ #define _CDN_DP_CORE_H #include <drm/display/drm_dp_helper.h> +#include <drm/drm_bridge.h> #include <drm/drm_panel.h> #include <drm/drm_probe_helper.h> #include <sound/hdmi-codec.h> @@ -65,12 +66,11 @@ struct cdn_dp_port { struct cdn_dp_device { struct device *dev; struct drm_device *drm_dev; - struct drm_connector connector; + struct drm_bridge bridge; struct rockchip_encoder encoder; struct drm_display_mode mode; struct platform_device *audio_pdev; struct work_struct event_work; - const struct drm_edid *drm_edid; struct mutex lock; bool connected; @@ -101,9 +101,5 @@ struct cdn_dp_device { int active_port; u8 dpcd[DP_RECEIVER_CAP_SIZE]; - bool sink_has_audio; - - hdmi_codec_plugged_cb plugged_cb; - struct device *codec_dev; }; #endif /* _CDN_DP_CORE_H */ diff --git a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c index f737e7d46e66..acb59b25d928 100644 --- a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c +++ b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c @@ -213,17 +213,13 @@ static int rockchip_hdmi_parse_dt(struct rockchip_hdmi *hdmi) if (IS_ERR(hdmi->ref_clk)) { ret = PTR_ERR(hdmi->ref_clk); - if (ret != -EPROBE_DEFER) - dev_err(hdmi->dev, "failed to get reference clock\n"); - return ret; + return dev_err_probe(hdmi->dev, ret, "failed to get reference clock\n"); } hdmi->grf_clk = devm_clk_get_optional(hdmi->dev, "grf"); if (IS_ERR(hdmi->grf_clk)) { ret = PTR_ERR(hdmi->grf_clk); - if (ret != -EPROBE_DEFER) - dev_err(hdmi->dev, "failed to get grf clock\n"); - return ret; + return dev_err_probe(hdmi->dev, ret, "failed to get grf clock\n"); } ret = devm_regulator_get_enable(hdmi->dev, "avdd-0v9"); @@ -573,17 +569,13 @@ static int dw_hdmi_rockchip_bind(struct device *dev, struct device *master, ret = rockchip_hdmi_parse_dt(hdmi); if (ret) { - if (ret != -EPROBE_DEFER) - dev_err(hdmi->dev, "Unable to parse OF data\n"); - return ret; + return dev_err_probe(hdmi->dev, ret, "Unable to parse OF data\n"); } hdmi->phy = devm_phy_optional_get(dev, "hdmi"); if (IS_ERR(hdmi->phy)) { ret = PTR_ERR(hdmi->phy); - if (ret != -EPROBE_DEFER) - dev_err(hdmi->dev, "failed to get phy\n"); - return ret; + return dev_err_probe(hdmi->dev, ret, "failed to get phy\n"); } if (hdmi->phy) { diff --git a/drivers/gpu/drm/rockchip/inno_hdmi.c b/drivers/gpu/drm/rockchip/inno_hdmi.c index db4b4038e51d..1ab3ad4bde9e 100644 --- a/drivers/gpu/drm/rockchip/inno_hdmi.c +++ b/drivers/gpu/drm/rockchip/inno_hdmi.c @@ -29,11 +29,360 @@ #include "rockchip_drm_drv.h" -#include "inno_hdmi.h" +#define INNO_HDMI_MIN_TMDS_CLOCK 25000000U -#define HIWORD_UPDATE(val, mask) ((val) | (mask) << 16) +#define DDC_SEGMENT_ADDR 0x30 + +#define HDMI_SCL_RATE (100 * 1000) + +#define DDC_BUS_FREQ_L 0x4b +#define DDC_BUS_FREQ_H 0x4c + +#define HDMI_SYS_CTRL 0x00 +#define m_RST_ANALOG BIT(6) +#define v_RST_ANALOG (0 << 6) +#define v_NOT_RST_ANALOG BIT(6) +#define m_RST_DIGITAL BIT(5) +#define v_RST_DIGITAL (0 << 5) +#define v_NOT_RST_DIGITAL BIT(5) +#define m_REG_CLK_INV BIT(4) +#define v_REG_CLK_NOT_INV (0 << 4) +#define v_REG_CLK_INV BIT(4) +#define m_VCLK_INV BIT(3) +#define v_VCLK_NOT_INV (0 << 3) +#define v_VCLK_INV BIT(3) +#define m_REG_CLK_SOURCE BIT(2) +#define v_REG_CLK_SOURCE_TMDS (0 << 2) +#define v_REG_CLK_SOURCE_SYS BIT(2) +#define m_POWER BIT(1) +#define v_PWR_ON (0 << 1) +#define v_PWR_OFF BIT(1) +#define m_INT_POL BIT(0) +#define v_INT_POL_HIGH 1 +#define v_INT_POL_LOW 0 + +#define HDMI_VIDEO_CONTRL1 0x01 +#define m_VIDEO_INPUT_FORMAT (7 << 1) +#define m_DE_SOURCE BIT(0) +#define v_VIDEO_INPUT_FORMAT(n) ((n) << 1) +#define v_DE_EXTERNAL 1 +#define v_DE_INTERNAL 0 +enum { + VIDEO_INPUT_SDR_RGB444 = 0, + VIDEO_INPUT_DDR_RGB444 = 5, + VIDEO_INPUT_DDR_YCBCR422 = 6 +}; -#define INNO_HDMI_MIN_TMDS_CLOCK 25000000U +#define HDMI_VIDEO_CONTRL2 0x02 +#define m_VIDEO_OUTPUT_COLOR (3 << 6) +#define m_VIDEO_INPUT_BITS (3 << 4) +#define m_VIDEO_INPUT_CSP BIT(0) +#define v_VIDEO_OUTPUT_COLOR(n) (((n) & 0x3) << 6) +#define v_VIDEO_INPUT_BITS(n) ((n) << 4) +#define v_VIDEO_INPUT_CSP(n) ((n) << 0) +enum { + VIDEO_INPUT_12BITS = 0, + VIDEO_INPUT_10BITS = 1, + VIDEO_INPUT_REVERT = 2, + VIDEO_INPUT_8BITS = 3, +}; + +#define HDMI_VIDEO_CONTRL 0x03 +#define m_VIDEO_AUTO_CSC BIT(7) +#define v_VIDEO_AUTO_CSC(n) ((n) << 7) +#define m_VIDEO_C0_C2_SWAP BIT(0) +#define v_VIDEO_C0_C2_SWAP(n) ((n) << 0) +enum { + C0_C2_CHANGE_ENABLE = 0, + C0_C2_CHANGE_DISABLE = 1, + AUTO_CSC_DISABLE = 0, + AUTO_CSC_ENABLE = 1, +}; + +#define HDMI_VIDEO_CONTRL3 0x04 +#define m_COLOR_DEPTH_NOT_INDICATED BIT(4) +#define m_SOF BIT(3) +#define m_COLOR_RANGE BIT(2) +#define m_CSC BIT(0) +#define v_COLOR_DEPTH_NOT_INDICATED(n) ((n) << 4) +#define v_SOF_ENABLE (0 << 3) +#define v_SOF_DISABLE BIT(3) +#define v_COLOR_RANGE_FULL BIT(2) +#define v_COLOR_RANGE_LIMITED (0 << 2) +#define v_CSC_ENABLE 1 +#define v_CSC_DISABLE 0 + +#define HDMI_AV_MUTE 0x05 +#define m_AVMUTE_CLEAR BIT(7) +#define m_AVMUTE_ENABLE BIT(6) +#define m_AUDIO_MUTE BIT(1) +#define m_VIDEO_BLACK BIT(0) +#define v_AVMUTE_CLEAR(n) ((n) << 7) +#define v_AVMUTE_ENABLE(n) ((n) << 6) +#define v_AUDIO_MUTE(n) ((n) << 1) +#define v_VIDEO_MUTE(n) ((n) << 0) + +#define HDMI_VIDEO_TIMING_CTL 0x08 +#define v_HSYNC_POLARITY(n) ((n) << 3) +#define v_VSYNC_POLARITY(n) ((n) << 2) +#define v_INETLACE(n) ((n) << 1) +#define v_EXTERANL_VIDEO(n) ((n) << 0) + +#define HDMI_VIDEO_EXT_HTOTAL_L 0x09 +#define HDMI_VIDEO_EXT_HTOTAL_H 0x0a +#define HDMI_VIDEO_EXT_HBLANK_L 0x0b +#define HDMI_VIDEO_EXT_HBLANK_H 0x0c +#define HDMI_VIDEO_EXT_HDELAY_L 0x0d +#define HDMI_VIDEO_EXT_HDELAY_H 0x0e +#define HDMI_VIDEO_EXT_HDURATION_L 0x0f +#define HDMI_VIDEO_EXT_HDURATION_H 0x10 +#define HDMI_VIDEO_EXT_VTOTAL_L 0x11 +#define HDMI_VIDEO_EXT_VTOTAL_H 0x12 +#define HDMI_VIDEO_EXT_VBLANK 0x13 +#define HDMI_VIDEO_EXT_VDELAY 0x14 +#define HDMI_VIDEO_EXT_VDURATION 0x15 + +#define HDMI_VIDEO_CSC_COEF 0x18 + +#define HDMI_AUDIO_CTRL1 0x35 +enum { + CTS_SOURCE_INTERNAL = 0, + CTS_SOURCE_EXTERNAL = 1, +}; + +#define v_CTS_SOURCE(n) ((n) << 7) + +enum { + DOWNSAMPLE_DISABLE = 0, + DOWNSAMPLE_1_2 = 1, + DOWNSAMPLE_1_4 = 2, +}; + +#define v_DOWN_SAMPLE(n) ((n) << 5) + +enum { + AUDIO_SOURCE_IIS = 0, + AUDIO_SOURCE_SPDIF = 1, +}; + +#define v_AUDIO_SOURCE(n) ((n) << 3) + +#define v_MCLK_ENABLE(n) ((n) << 2) + +enum { + MCLK_128FS = 0, + MCLK_256FS = 1, + MCLK_384FS = 2, + MCLK_512FS = 3, +}; + +#define v_MCLK_RATIO(n) (n) + +#define AUDIO_SAMPLE_RATE 0x37 + +enum { + AUDIO_32K = 0x3, + AUDIO_441K = 0x0, + AUDIO_48K = 0x2, + AUDIO_882K = 0x8, + AUDIO_96K = 0xa, + AUDIO_1764K = 0xc, + AUDIO_192K = 0xe, +}; + +#define AUDIO_I2S_MODE 0x38 + +enum { + I2S_CHANNEL_1_2 = 1, + I2S_CHANNEL_3_4 = 3, + I2S_CHANNEL_5_6 = 7, + I2S_CHANNEL_7_8 = 0xf +}; + +#define v_I2S_CHANNEL(n) ((n) << 2) + +enum { + I2S_STANDARD = 0, + I2S_LEFT_JUSTIFIED = 1, + I2S_RIGHT_JUSTIFIED = 2, +}; + +#define v_I2S_MODE(n) (n) + +#define AUDIO_I2S_MAP 0x39 +#define AUDIO_I2S_SWAPS_SPDIF 0x3a +#define v_SPIDF_FREQ(n) (n) + +#define N_32K 0x1000 +#define N_441K 0x1880 +#define N_882K 0x3100 +#define N_1764K 0x6200 +#define N_48K 0x1800 +#define N_96K 0x3000 +#define N_192K 0x6000 + +#define HDMI_AUDIO_CHANNEL_STATUS 0x3e +#define m_AUDIO_STATUS_NLPCM BIT(7) +#define m_AUDIO_STATUS_USE BIT(6) +#define m_AUDIO_STATUS_COPYRIGHT BIT(5) +#define m_AUDIO_STATUS_ADDITION (3 << 2) +#define m_AUDIO_STATUS_CLK_ACCURACY (2 << 0) +#define v_AUDIO_STATUS_NLPCM(n) (((n) & 1) << 7) +#define AUDIO_N_H 0x3f +#define AUDIO_N_M 0x40 +#define AUDIO_N_L 0x41 + +#define HDMI_AUDIO_CTS_H 0x45 +#define HDMI_AUDIO_CTS_M 0x46 +#define HDMI_AUDIO_CTS_L 0x47 + +#define HDMI_DDC_CLK_L 0x4b +#define HDMI_DDC_CLK_H 0x4c + +#define HDMI_EDID_SEGMENT_POINTER 0x4d +#define HDMI_EDID_WORD_ADDR 0x4e +#define HDMI_EDID_FIFO_OFFSET 0x4f +#define HDMI_EDID_FIFO_ADDR 0x50 + +#define HDMI_PACKET_SEND_MANUAL 0x9c +#define HDMI_PACKET_SEND_AUTO 0x9d +#define m_PACKET_GCP_EN BIT(7) +#define m_PACKET_MSI_EN BIT(6) +#define m_PACKET_SDI_EN BIT(5) +#define m_PACKET_VSI_EN BIT(4) +#define v_PACKET_GCP_EN(n) (((n) & 1) << 7) +#define v_PACKET_MSI_EN(n) (((n) & 1) << 6) +#define v_PACKET_SDI_EN(n) (((n) & 1) << 5) +#define v_PACKET_VSI_EN(n) (((n) & 1) << 4) + +#define HDMI_CONTROL_PACKET_BUF_INDEX 0x9f + +enum { + INFOFRAME_VSI = 0x05, + INFOFRAME_AVI = 0x06, + INFOFRAME_AAI = 0x08, +}; + +#define HDMI_CONTROL_PACKET_ADDR 0xa0 +#define HDMI_MAXIMUM_INFO_FRAME_SIZE 0x11 + +enum { + AVI_COLOR_MODE_RGB = 0, + AVI_COLOR_MODE_YCBCR422 = 1, + AVI_COLOR_MODE_YCBCR444 = 2, + AVI_COLORIMETRY_NO_DATA = 0, + + AVI_COLORIMETRY_SMPTE_170M = 1, + AVI_COLORIMETRY_ITU709 = 2, + AVI_COLORIMETRY_EXTENDED = 3, + + AVI_CODED_FRAME_ASPECT_NO_DATA = 0, + AVI_CODED_FRAME_ASPECT_4_3 = 1, + AVI_CODED_FRAME_ASPECT_16_9 = 2, + + ACTIVE_ASPECT_RATE_SAME_AS_CODED_FRAME = 0x08, + ACTIVE_ASPECT_RATE_4_3 = 0x09, + ACTIVE_ASPECT_RATE_16_9 = 0x0A, + ACTIVE_ASPECT_RATE_14_9 = 0x0B, +}; + +#define HDMI_HDCP_CTRL 0x52 +#define m_HDMI_DVI BIT(1) +#define v_HDMI_DVI(n) ((n) << 1) + +#define HDMI_INTERRUPT_MASK1 0xc0 +#define HDMI_INTERRUPT_STATUS1 0xc1 +#define m_INT_ACTIVE_VSYNC BIT(5) +#define m_INT_EDID_READY BIT(2) + +#define HDMI_INTERRUPT_MASK2 0xc2 +#define HDMI_INTERRUPT_STATUS2 0xc3 +#define m_INT_HDCP_ERR BIT(7) +#define m_INT_BKSV_FLAG BIT(6) +#define m_INT_HDCP_OK BIT(4) + +#define HDMI_STATUS 0xc8 +#define m_HOTPLUG BIT(7) +#define m_MASK_INT_HOTPLUG BIT(5) +#define m_INT_HOTPLUG BIT(1) +#define v_MASK_INT_HOTPLUG(n) (((n) & 0x1) << 5) + +#define HDMI_COLORBAR 0xc9 + +#define HDMI_PHY_SYNC 0xce +#define HDMI_PHY_SYS_CTL 0xe0 +#define m_TMDS_CLK_SOURCE BIT(5) +#define v_TMDS_FROM_PLL (0 << 5) +#define v_TMDS_FROM_GEN BIT(5) +#define m_PHASE_CLK BIT(4) +#define v_DEFAULT_PHASE (0 << 4) +#define v_SYNC_PHASE BIT(4) +#define m_TMDS_CURRENT_PWR BIT(3) +#define v_TURN_ON_CURRENT (0 << 3) +#define v_CAT_OFF_CURRENT BIT(3) +#define m_BANDGAP_PWR BIT(2) +#define v_BANDGAP_PWR_UP (0 << 2) +#define v_BANDGAP_PWR_DOWN BIT(2) +#define m_PLL_PWR BIT(1) +#define v_PLL_PWR_UP (0 << 1) +#define v_PLL_PWR_DOWN BIT(1) +#define m_TMDS_CHG_PWR BIT(0) +#define v_TMDS_CHG_PWR_UP (0 << 0) +#define v_TMDS_CHG_PWR_DOWN BIT(0) + +#define HDMI_PHY_CHG_PWR 0xe1 +#define v_CLK_CHG_PWR(n) (((n) & 1) << 3) +#define v_DATA_CHG_PWR(n) (((n) & 7) << 0) + +#define HDMI_PHY_DRIVER 0xe2 +#define v_CLK_MAIN_DRIVER(n) ((n) << 4) +#define v_DATA_MAIN_DRIVER(n) ((n) << 0) + +#define HDMI_PHY_PRE_EMPHASIS 0xe3 +#define v_PRE_EMPHASIS(n) (((n) & 7) << 4) +#define v_CLK_PRE_DRIVER(n) (((n) & 3) << 2) +#define v_DATA_PRE_DRIVER(n) (((n) & 3) << 0) + +#define HDMI_PHY_FEEDBACK_DIV_RATIO_LOW 0xe7 +#define v_FEEDBACK_DIV_LOW(n) ((n) & 0xff) +#define HDMI_PHY_FEEDBACK_DIV_RATIO_HIGH 0xe8 +#define v_FEEDBACK_DIV_HIGH(n) ((n) & 1) + +#define HDMI_PHY_PRE_DIV_RATIO 0xed +#define v_PRE_DIV_RATIO(n) ((n) & 0x1f) + +#define HDMI_CEC_CTRL 0xd0 +#define m_ADJUST_FOR_HISENSE BIT(6) +#define m_REJECT_RX_BROADCAST BIT(5) +#define m_BUSFREETIME_ENABLE BIT(2) +#define m_REJECT_RX BIT(1) +#define m_START_TX BIT(0) + +#define HDMI_CEC_DATA 0xd1 +#define HDMI_CEC_TX_OFFSET 0xd2 +#define HDMI_CEC_RX_OFFSET 0xd3 +#define HDMI_CEC_CLK_H 0xd4 +#define HDMI_CEC_CLK_L 0xd5 +#define HDMI_CEC_TX_LENGTH 0xd6 +#define HDMI_CEC_RX_LENGTH 0xd7 +#define HDMI_CEC_TX_INT_MASK 0xd8 +#define m_TX_DONE BIT(3) +#define m_TX_NOACK BIT(2) +#define m_TX_BROADCAST_REJ BIT(1) +#define m_TX_BUSNOTFREE BIT(0) + +#define HDMI_CEC_RX_INT_MASK 0xd9 +#define m_RX_LA_ERR BIT(4) +#define m_RX_GLITCH BIT(3) +#define m_RX_DONE BIT(0) + +#define HDMI_CEC_TX_INT 0xda +#define HDMI_CEC_RX_INT 0xdb +#define HDMI_CEC_BUSFREETIME_L 0xdc +#define HDMI_CEC_BUSFREETIME_H 0xdd +#define HDMI_CEC_LOGICADDR 0xde + +#define HIWORD_UPDATE(val, mask) ((val) | (mask) << 16) #define RK3036_GRF_SOC_CON2 0x148 #define RK3036_HDMI_PHSYNC BIT(4) @@ -255,22 +604,37 @@ static void inno_hdmi_power_up(struct inno_hdmi *hdmi, inno_hdmi_sys_power(hdmi, true); }; -static void inno_hdmi_reset(struct inno_hdmi *hdmi) +static void inno_hdmi_init_hw(struct inno_hdmi *hdmi) { u32 val; u32 msk; hdmi_modb(hdmi, HDMI_SYS_CTRL, m_RST_DIGITAL, v_NOT_RST_DIGITAL); - udelay(100); + usleep_range(100, 150); hdmi_modb(hdmi, HDMI_SYS_CTRL, m_RST_ANALOG, v_NOT_RST_ANALOG); - udelay(100); + usleep_range(100, 150); msk = m_REG_CLK_INV | m_REG_CLK_SOURCE | m_POWER | m_INT_POL; val = v_REG_CLK_INV | v_REG_CLK_SOURCE_SYS | v_PWR_ON | v_INT_POL_HIGH; hdmi_modb(hdmi, HDMI_SYS_CTRL, msk, val); inno_hdmi_standby(hdmi); + + /* + * When the controller isn't configured to an accurate + * video timing and there is no reference clock available, + * then the TMDS clock source would be switched to PCLK_HDMI, + * so we need to init the TMDS rate to PCLK rate, and + * reconfigure the DDC clock. + */ + if (hdmi->refclk) + inno_hdmi_i2c_init(hdmi, clk_get_rate(hdmi->refclk)); + else + inno_hdmi_i2c_init(hdmi, clk_get_rate(hdmi->pclk)); + + /* Unmute hotplug interrupt */ + hdmi_modb(hdmi, HDMI_STATUS, m_MASK_INT_HOTPLUG, v_MASK_INT_HOTPLUG(1)); } static int inno_hdmi_disable_frame(struct drm_connector *connector, @@ -775,8 +1139,7 @@ static int inno_hdmi_i2c_write(struct inno_hdmi *hdmi, struct i2c_msg *msgs) * we assume that each word write to this i2c adapter * should be the offset of EDID word address. */ - if ((msgs->len != 1) || - ((msgs->addr != DDC_ADDR) && (msgs->addr != DDC_SEGMENT_ADDR))) + if (msgs->len != 1 || (msgs->addr != DDC_ADDR && msgs->addr != DDC_SEGMENT_ADDR)) return -EINVAL; reinit_completion(&hdmi->i2c->cmp); @@ -867,10 +1230,9 @@ static struct i2c_adapter *inno_hdmi_i2c_adapter(struct inno_hdmi *hdmi) strscpy(adap->name, "Inno HDMI", sizeof(adap->name)); i2c_set_adapdata(adap, hdmi); - ret = i2c_add_adapter(adap); + ret = devm_i2c_add_adapter(hdmi->dev, adap); if (ret) { dev_warn(hdmi->dev, "cannot add %s I2C adapter\n", adap->name); - devm_kfree(hdmi->dev, i2c); return ERR_PTR(ret); } @@ -907,71 +1269,37 @@ static int inno_hdmi_bind(struct device *dev, struct device *master, if (IS_ERR(hdmi->regs)) return PTR_ERR(hdmi->regs); - hdmi->pclk = devm_clk_get(hdmi->dev, "pclk"); + hdmi->pclk = devm_clk_get_enabled(hdmi->dev, "pclk"); if (IS_ERR(hdmi->pclk)) return dev_err_probe(dev, PTR_ERR(hdmi->pclk), "Unable to get HDMI pclk\n"); - ret = clk_prepare_enable(hdmi->pclk); - if (ret) - return dev_err_probe(dev, ret, "Cannot enable HDMI pclk: %d\n", ret); - - hdmi->refclk = devm_clk_get_optional(hdmi->dev, "ref"); - if (IS_ERR(hdmi->refclk)) { - ret = dev_err_probe(dev, PTR_ERR(hdmi->refclk), "Unable to get HDMI refclk\n"); - goto err_disable_pclk; - } - - ret = clk_prepare_enable(hdmi->refclk); - if (ret) { - ret = dev_err_probe(dev, ret, "Cannot enable HDMI refclk: %d\n", ret); - goto err_disable_pclk; - } + hdmi->refclk = devm_clk_get_optional_enabled(hdmi->dev, "ref"); + if (IS_ERR(hdmi->refclk)) + return dev_err_probe(dev, PTR_ERR(hdmi->refclk), "Unable to get HDMI refclk\n"); if (hdmi->variant->dev_type == RK3036_HDMI) { hdmi->grf = syscon_regmap_lookup_by_phandle(dev->of_node, "rockchip,grf"); - if (IS_ERR(hdmi->grf)) { - ret = dev_err_probe(dev, PTR_ERR(hdmi->grf), - "Unable to get rockchip,grf\n"); - goto err_disable_clk; - } + if (IS_ERR(hdmi->grf)) + return dev_err_probe(dev, + PTR_ERR(hdmi->grf), "Unable to get rockchip,grf\n"); } irq = platform_get_irq(pdev, 0); - if (irq < 0) { - ret = irq; - goto err_disable_clk; - } + if (irq < 0) + return irq; - inno_hdmi_reset(hdmi); + inno_hdmi_init_hw(hdmi); hdmi->ddc = inno_hdmi_i2c_adapter(hdmi); - if (IS_ERR(hdmi->ddc)) { - ret = PTR_ERR(hdmi->ddc); - hdmi->ddc = NULL; - goto err_disable_clk; - } - - /* - * When the controller isn't configured to an accurate - * video timing and there is no reference clock available, - * then the TMDS clock source would be switched to PCLK_HDMI, - * so we need to init the TMDS rate to PCLK rate, and - * reconfigure the DDC clock. - */ - if (hdmi->refclk) - inno_hdmi_i2c_init(hdmi, clk_get_rate(hdmi->refclk)); - else - inno_hdmi_i2c_init(hdmi, clk_get_rate(hdmi->pclk)); + if (IS_ERR(hdmi->ddc)) + return PTR_ERR(hdmi->ddc); ret = inno_hdmi_register(drm, hdmi); if (ret) - goto err_put_adapter; + return ret; dev_set_drvdata(dev, hdmi); - /* Unmute hotplug interrupt */ - hdmi_modb(hdmi, HDMI_STATUS, m_MASK_INT_HOTPLUG, v_MASK_INT_HOTPLUG(1)); - ret = devm_request_threaded_irq(dev, irq, inno_hdmi_hardirq, inno_hdmi_irq, IRQF_SHARED, dev_name(dev), hdmi); @@ -982,12 +1310,6 @@ static int inno_hdmi_bind(struct device *dev, struct device *master, err_cleanup_hdmi: hdmi->connector.funcs->destroy(&hdmi->connector); hdmi->encoder.encoder.funcs->destroy(&hdmi->encoder.encoder); -err_put_adapter: - i2c_put_adapter(hdmi->ddc); -err_disable_clk: - clk_disable_unprepare(hdmi->refclk); -err_disable_pclk: - clk_disable_unprepare(hdmi->pclk); return ret; } @@ -998,10 +1320,6 @@ static void inno_hdmi_unbind(struct device *dev, struct device *master, hdmi->connector.funcs->destroy(&hdmi->connector); hdmi->encoder.encoder.funcs->destroy(&hdmi->encoder.encoder); - - i2c_put_adapter(hdmi->ddc); - clk_disable_unprepare(hdmi->refclk); - clk_disable_unprepare(hdmi->pclk); } static const struct component_ops inno_hdmi_ops = { diff --git a/drivers/gpu/drm/rockchip/inno_hdmi.h b/drivers/gpu/drm/rockchip/inno_hdmi.h deleted file mode 100644 index 8b7ef3fac485..000000000000 --- a/drivers/gpu/drm/rockchip/inno_hdmi.h +++ /dev/null @@ -1,349 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Copyright (C) Rockchip Electronics Co., Ltd. - * Zheng Yang <zhengyang@rock-chips.com> - * Yakir Yang <ykk@rock-chips.com> - */ - -#ifndef __INNO_HDMI_H__ -#define __INNO_HDMI_H__ - -#define DDC_SEGMENT_ADDR 0x30 - -#define HDMI_SCL_RATE (100*1000) -#define DDC_BUS_FREQ_L 0x4b -#define DDC_BUS_FREQ_H 0x4c - -#define HDMI_SYS_CTRL 0x00 -#define m_RST_ANALOG (1 << 6) -#define v_RST_ANALOG (0 << 6) -#define v_NOT_RST_ANALOG (1 << 6) -#define m_RST_DIGITAL (1 << 5) -#define v_RST_DIGITAL (0 << 5) -#define v_NOT_RST_DIGITAL (1 << 5) -#define m_REG_CLK_INV (1 << 4) -#define v_REG_CLK_NOT_INV (0 << 4) -#define v_REG_CLK_INV (1 << 4) -#define m_VCLK_INV (1 << 3) -#define v_VCLK_NOT_INV (0 << 3) -#define v_VCLK_INV (1 << 3) -#define m_REG_CLK_SOURCE (1 << 2) -#define v_REG_CLK_SOURCE_TMDS (0 << 2) -#define v_REG_CLK_SOURCE_SYS (1 << 2) -#define m_POWER (1 << 1) -#define v_PWR_ON (0 << 1) -#define v_PWR_OFF (1 << 1) -#define m_INT_POL (1 << 0) -#define v_INT_POL_HIGH 1 -#define v_INT_POL_LOW 0 - -#define HDMI_VIDEO_CONTRL1 0x01 -#define m_VIDEO_INPUT_FORMAT (7 << 1) -#define m_DE_SOURCE (1 << 0) -#define v_VIDEO_INPUT_FORMAT(n) (n << 1) -#define v_DE_EXTERNAL 1 -#define v_DE_INTERNAL 0 -enum { - VIDEO_INPUT_SDR_RGB444 = 0, - VIDEO_INPUT_DDR_RGB444 = 5, - VIDEO_INPUT_DDR_YCBCR422 = 6 -}; - -#define HDMI_VIDEO_CONTRL2 0x02 -#define m_VIDEO_OUTPUT_COLOR (3 << 6) -#define m_VIDEO_INPUT_BITS (3 << 4) -#define m_VIDEO_INPUT_CSP (1 << 0) -#define v_VIDEO_OUTPUT_COLOR(n) (((n) & 0x3) << 6) -#define v_VIDEO_INPUT_BITS(n) (n << 4) -#define v_VIDEO_INPUT_CSP(n) (n << 0) -enum { - VIDEO_INPUT_12BITS = 0, - VIDEO_INPUT_10BITS = 1, - VIDEO_INPUT_REVERT = 2, - VIDEO_INPUT_8BITS = 3, -}; - -#define HDMI_VIDEO_CONTRL 0x03 -#define m_VIDEO_AUTO_CSC (1 << 7) -#define v_VIDEO_AUTO_CSC(n) (n << 7) -#define m_VIDEO_C0_C2_SWAP (1 << 0) -#define v_VIDEO_C0_C2_SWAP(n) (n << 0) -enum { - C0_C2_CHANGE_ENABLE = 0, - C0_C2_CHANGE_DISABLE = 1, - AUTO_CSC_DISABLE = 0, - AUTO_CSC_ENABLE = 1, -}; - -#define HDMI_VIDEO_CONTRL3 0x04 -#define m_COLOR_DEPTH_NOT_INDICATED (1 << 4) -#define m_SOF (1 << 3) -#define m_COLOR_RANGE (1 << 2) -#define m_CSC (1 << 0) -#define v_COLOR_DEPTH_NOT_INDICATED(n) ((n) << 4) -#define v_SOF_ENABLE (0 << 3) -#define v_SOF_DISABLE (1 << 3) -#define v_COLOR_RANGE_FULL (1 << 2) -#define v_COLOR_RANGE_LIMITED (0 << 2) -#define v_CSC_ENABLE 1 -#define v_CSC_DISABLE 0 - -#define HDMI_AV_MUTE 0x05 -#define m_AVMUTE_CLEAR (1 << 7) -#define m_AVMUTE_ENABLE (1 << 6) -#define m_AUDIO_MUTE (1 << 1) -#define m_VIDEO_BLACK (1 << 0) -#define v_AVMUTE_CLEAR(n) (n << 7) -#define v_AVMUTE_ENABLE(n) (n << 6) -#define v_AUDIO_MUTE(n) (n << 1) -#define v_VIDEO_MUTE(n) (n << 0) - -#define HDMI_VIDEO_TIMING_CTL 0x08 -#define v_HSYNC_POLARITY(n) (n << 3) -#define v_VSYNC_POLARITY(n) (n << 2) -#define v_INETLACE(n) (n << 1) -#define v_EXTERANL_VIDEO(n) (n << 0) - -#define HDMI_VIDEO_EXT_HTOTAL_L 0x09 -#define HDMI_VIDEO_EXT_HTOTAL_H 0x0a -#define HDMI_VIDEO_EXT_HBLANK_L 0x0b -#define HDMI_VIDEO_EXT_HBLANK_H 0x0c -#define HDMI_VIDEO_EXT_HDELAY_L 0x0d -#define HDMI_VIDEO_EXT_HDELAY_H 0x0e -#define HDMI_VIDEO_EXT_HDURATION_L 0x0f -#define HDMI_VIDEO_EXT_HDURATION_H 0x10 -#define HDMI_VIDEO_EXT_VTOTAL_L 0x11 -#define HDMI_VIDEO_EXT_VTOTAL_H 0x12 -#define HDMI_VIDEO_EXT_VBLANK 0x13 -#define HDMI_VIDEO_EXT_VDELAY 0x14 -#define HDMI_VIDEO_EXT_VDURATION 0x15 - -#define HDMI_VIDEO_CSC_COEF 0x18 - -#define HDMI_AUDIO_CTRL1 0x35 -enum { - CTS_SOURCE_INTERNAL = 0, - CTS_SOURCE_EXTERNAL = 1, -}; -#define v_CTS_SOURCE(n) (n << 7) - -enum { - DOWNSAMPLE_DISABLE = 0, - DOWNSAMPLE_1_2 = 1, - DOWNSAMPLE_1_4 = 2, -}; -#define v_DOWN_SAMPLE(n) (n << 5) - -enum { - AUDIO_SOURCE_IIS = 0, - AUDIO_SOURCE_SPDIF = 1, -}; -#define v_AUDIO_SOURCE(n) (n << 3) - -#define v_MCLK_ENABLE(n) (n << 2) -enum { - MCLK_128FS = 0, - MCLK_256FS = 1, - MCLK_384FS = 2, - MCLK_512FS = 3, -}; -#define v_MCLK_RATIO(n) (n) - -#define AUDIO_SAMPLE_RATE 0x37 -enum { - AUDIO_32K = 0x3, - AUDIO_441K = 0x0, - AUDIO_48K = 0x2, - AUDIO_882K = 0x8, - AUDIO_96K = 0xa, - AUDIO_1764K = 0xc, - AUDIO_192K = 0xe, -}; - -#define AUDIO_I2S_MODE 0x38 -enum { - I2S_CHANNEL_1_2 = 1, - I2S_CHANNEL_3_4 = 3, - I2S_CHANNEL_5_6 = 7, - I2S_CHANNEL_7_8 = 0xf -}; -#define v_I2S_CHANNEL(n) ((n) << 2) -enum { - I2S_STANDARD = 0, - I2S_LEFT_JUSTIFIED = 1, - I2S_RIGHT_JUSTIFIED = 2, -}; -#define v_I2S_MODE(n) (n) - -#define AUDIO_I2S_MAP 0x39 -#define AUDIO_I2S_SWAPS_SPDIF 0x3a -#define v_SPIDF_FREQ(n) (n) - -#define N_32K 0x1000 -#define N_441K 0x1880 -#define N_882K 0x3100 -#define N_1764K 0x6200 -#define N_48K 0x1800 -#define N_96K 0x3000 -#define N_192K 0x6000 - -#define HDMI_AUDIO_CHANNEL_STATUS 0x3e -#define m_AUDIO_STATUS_NLPCM (1 << 7) -#define m_AUDIO_STATUS_USE (1 << 6) -#define m_AUDIO_STATUS_COPYRIGHT (1 << 5) -#define m_AUDIO_STATUS_ADDITION (3 << 2) -#define m_AUDIO_STATUS_CLK_ACCURACY (2 << 0) -#define v_AUDIO_STATUS_NLPCM(n) ((n & 1) << 7) -#define AUDIO_N_H 0x3f -#define AUDIO_N_M 0x40 -#define AUDIO_N_L 0x41 - -#define HDMI_AUDIO_CTS_H 0x45 -#define HDMI_AUDIO_CTS_M 0x46 -#define HDMI_AUDIO_CTS_L 0x47 - -#define HDMI_DDC_CLK_L 0x4b -#define HDMI_DDC_CLK_H 0x4c - -#define HDMI_EDID_SEGMENT_POINTER 0x4d -#define HDMI_EDID_WORD_ADDR 0x4e -#define HDMI_EDID_FIFO_OFFSET 0x4f -#define HDMI_EDID_FIFO_ADDR 0x50 - -#define HDMI_PACKET_SEND_MANUAL 0x9c -#define HDMI_PACKET_SEND_AUTO 0x9d -#define m_PACKET_GCP_EN (1 << 7) -#define m_PACKET_MSI_EN (1 << 6) -#define m_PACKET_SDI_EN (1 << 5) -#define m_PACKET_VSI_EN (1 << 4) -#define v_PACKET_GCP_EN(n) ((n & 1) << 7) -#define v_PACKET_MSI_EN(n) ((n & 1) << 6) -#define v_PACKET_SDI_EN(n) ((n & 1) << 5) -#define v_PACKET_VSI_EN(n) ((n & 1) << 4) - -#define HDMI_CONTROL_PACKET_BUF_INDEX 0x9f -enum { - INFOFRAME_VSI = 0x05, - INFOFRAME_AVI = 0x06, - INFOFRAME_AAI = 0x08, -}; - -#define HDMI_CONTROL_PACKET_ADDR 0xa0 -#define HDMI_MAXIMUM_INFO_FRAME_SIZE 0x11 -enum { - AVI_COLOR_MODE_RGB = 0, - AVI_COLOR_MODE_YCBCR422 = 1, - AVI_COLOR_MODE_YCBCR444 = 2, - AVI_COLORIMETRY_NO_DATA = 0, - - AVI_COLORIMETRY_SMPTE_170M = 1, - AVI_COLORIMETRY_ITU709 = 2, - AVI_COLORIMETRY_EXTENDED = 3, - - AVI_CODED_FRAME_ASPECT_NO_DATA = 0, - AVI_CODED_FRAME_ASPECT_4_3 = 1, - AVI_CODED_FRAME_ASPECT_16_9 = 2, - - ACTIVE_ASPECT_RATE_SAME_AS_CODED_FRAME = 0x08, - ACTIVE_ASPECT_RATE_4_3 = 0x09, - ACTIVE_ASPECT_RATE_16_9 = 0x0A, - ACTIVE_ASPECT_RATE_14_9 = 0x0B, -}; - -#define HDMI_HDCP_CTRL 0x52 -#define m_HDMI_DVI (1 << 1) -#define v_HDMI_DVI(n) (n << 1) - -#define HDMI_INTERRUPT_MASK1 0xc0 -#define HDMI_INTERRUPT_STATUS1 0xc1 -#define m_INT_ACTIVE_VSYNC (1 << 5) -#define m_INT_EDID_READY (1 << 2) - -#define HDMI_INTERRUPT_MASK2 0xc2 -#define HDMI_INTERRUPT_STATUS2 0xc3 -#define m_INT_HDCP_ERR (1 << 7) -#define m_INT_BKSV_FLAG (1 << 6) -#define m_INT_HDCP_OK (1 << 4) - -#define HDMI_STATUS 0xc8 -#define m_HOTPLUG (1 << 7) -#define m_MASK_INT_HOTPLUG (1 << 5) -#define m_INT_HOTPLUG (1 << 1) -#define v_MASK_INT_HOTPLUG(n) ((n & 0x1) << 5) - -#define HDMI_COLORBAR 0xc9 - -#define HDMI_PHY_SYNC 0xce -#define HDMI_PHY_SYS_CTL 0xe0 -#define m_TMDS_CLK_SOURCE (1 << 5) -#define v_TMDS_FROM_PLL (0 << 5) -#define v_TMDS_FROM_GEN (1 << 5) -#define m_PHASE_CLK (1 << 4) -#define v_DEFAULT_PHASE (0 << 4) -#define v_SYNC_PHASE (1 << 4) -#define m_TMDS_CURRENT_PWR (1 << 3) -#define v_TURN_ON_CURRENT (0 << 3) -#define v_CAT_OFF_CURRENT (1 << 3) -#define m_BANDGAP_PWR (1 << 2) -#define v_BANDGAP_PWR_UP (0 << 2) -#define v_BANDGAP_PWR_DOWN (1 << 2) -#define m_PLL_PWR (1 << 1) -#define v_PLL_PWR_UP (0 << 1) -#define v_PLL_PWR_DOWN (1 << 1) -#define m_TMDS_CHG_PWR (1 << 0) -#define v_TMDS_CHG_PWR_UP (0 << 0) -#define v_TMDS_CHG_PWR_DOWN (1 << 0) - -#define HDMI_PHY_CHG_PWR 0xe1 -#define v_CLK_CHG_PWR(n) ((n & 1) << 3) -#define v_DATA_CHG_PWR(n) ((n & 7) << 0) - -#define HDMI_PHY_DRIVER 0xe2 -#define v_CLK_MAIN_DRIVER(n) (n << 4) -#define v_DATA_MAIN_DRIVER(n) (n << 0) - -#define HDMI_PHY_PRE_EMPHASIS 0xe3 -#define v_PRE_EMPHASIS(n) ((n & 7) << 4) -#define v_CLK_PRE_DRIVER(n) ((n & 3) << 2) -#define v_DATA_PRE_DRIVER(n) ((n & 3) << 0) - -#define HDMI_PHY_FEEDBACK_DIV_RATIO_LOW 0xe7 -#define v_FEEDBACK_DIV_LOW(n) (n & 0xff) -#define HDMI_PHY_FEEDBACK_DIV_RATIO_HIGH 0xe8 -#define v_FEEDBACK_DIV_HIGH(n) (n & 1) - -#define HDMI_PHY_PRE_DIV_RATIO 0xed -#define v_PRE_DIV_RATIO(n) (n & 0x1f) - -#define HDMI_CEC_CTRL 0xd0 -#define m_ADJUST_FOR_HISENSE (1 << 6) -#define m_REJECT_RX_BROADCAST (1 << 5) -#define m_BUSFREETIME_ENABLE (1 << 2) -#define m_REJECT_RX (1 << 1) -#define m_START_TX (1 << 0) - -#define HDMI_CEC_DATA 0xd1 -#define HDMI_CEC_TX_OFFSET 0xd2 -#define HDMI_CEC_RX_OFFSET 0xd3 -#define HDMI_CEC_CLK_H 0xd4 -#define HDMI_CEC_CLK_L 0xd5 -#define HDMI_CEC_TX_LENGTH 0xd6 -#define HDMI_CEC_RX_LENGTH 0xd7 -#define HDMI_CEC_TX_INT_MASK 0xd8 -#define m_TX_DONE (1 << 3) -#define m_TX_NOACK (1 << 2) -#define m_TX_BROADCAST_REJ (1 << 1) -#define m_TX_BUSNOTFREE (1 << 0) - -#define HDMI_CEC_RX_INT_MASK 0xd9 -#define m_RX_LA_ERR (1 << 4) -#define m_RX_GLITCH (1 << 3) -#define m_RX_DONE (1 << 0) - -#define HDMI_CEC_TX_INT 0xda -#define HDMI_CEC_RX_INT 0xdb -#define HDMI_CEC_BUSFREETIME_L 0xdc -#define HDMI_CEC_BUSFREETIME_H 0xdd -#define HDMI_CEC_LOGICADDR 0xde - -#endif /* __INNO_HDMI_H__ */ diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c index d0f5fea15e21..186f6452a7d3 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c @@ -146,25 +146,6 @@ static void vop2_unlock(struct vop2 *vop2) mutex_unlock(&vop2->vop2_lock); } -/* - * Note: - * The write mask function is documented but missing on rk3566/8, writes - * to these bits have no effect. For newer soc(rk3588 and following) the - * write mask is needed for register writes. - * - * GLB_CFG_DONE_EN has no write mask bit. - * - */ -static void vop2_cfg_done(struct vop2_video_port *vp) -{ - struct vop2 *vop2 = vp->vop2; - u32 val = RK3568_REG_CFG_DONE__GLB_CFG_DONE_EN; - - val |= BIT(vp->id) | (BIT(vp->id) << 16); - - regmap_set_bits(vop2->map, RK3568_REG_CFG_DONE, val); -} - static void vop2_win_disable(struct vop2_win *win) { vop2_win_write(win, VOP2_WIN_ENABLE, 0); @@ -854,6 +835,11 @@ static void vop2_enable(struct vop2 *vop2) if (vop2->version == VOP_VERSION_RK3588) rk3588_vop2_power_domain_enable_all(vop2); + if (vop2->version <= VOP_VERSION_RK3588) { + vop2->old_layer_sel = vop2_readl(vop2, RK3568_OVL_LAYER_SEL); + vop2->old_port_sel = vop2_readl(vop2, RK3568_OVL_PORT_SEL); + } + vop2_writel(vop2, RK3568_REG_CFG_DONE, RK3568_REG_CFG_DONE__GLB_CFG_DONE_EN); /* @@ -2422,6 +2408,10 @@ static int vop2_create_crtcs(struct vop2 *vop2) break; } } + + if (!vp->primary_plane) + return dev_err_probe(drm->dev, -ENOENT, + "no primary plane for vp %d\n", i); } /* Register all unused window as overlay plane */ @@ -2724,6 +2714,7 @@ static int vop2_bind(struct device *dev, struct device *master, void *data) return dev_err_probe(drm->dev, vop2->irq, "cannot find irq for vop2\n"); mutex_init(&vop2->vop2_lock); + mutex_init(&vop2->ovl_lock); ret = devm_request_irq(dev, vop2->irq, vop2_isr, IRQF_SHARED, dev_name(dev), vop2); if (ret) diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.h b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.h index fc3ecb9fcd95..fa5c56f16047 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.h +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.h @@ -334,6 +334,19 @@ struct vop2 { /* optional internal rgb encoder */ struct rockchip_rgb *rgb; + /* + * Used to record layer selection configuration on rk356x/rk3588 + * as register RK3568_OVL_LAYER_SEL and RK3568_OVL_PORT_SEL are + * shared for all the Video Ports. + */ + u32 old_layer_sel; + u32 old_port_sel; + /* + * Ensure that the updates to these two registers(RKK3568_OVL_LAYER_SEL/RK3568_OVL_PORT_SEL) + * take effect in sequence. + */ + struct mutex ovl_lock; + /* must be put at the end of the struct */ struct vop2_win win[]; }; @@ -727,6 +740,7 @@ enum dst_factor_mode { #define RK3588_OVL_PORT_SEL__CLUSTER2 GENMASK(21, 20) #define RK3568_OVL_PORT_SEL__CLUSTER1 GENMASK(19, 18) #define RK3568_OVL_PORT_SEL__CLUSTER0 GENMASK(17, 16) +#define RK3588_OVL_PORT_SET__PORT3_MUX GENMASK(15, 12) #define RK3568_OVL_PORT_SET__PORT2_MUX GENMASK(11, 8) #define RK3568_OVL_PORT_SET__PORT1_MUX GENMASK(7, 4) #define RK3568_OVL_PORT_SET__PORT0_MUX GENMASK(3, 0) @@ -831,4 +845,23 @@ static inline struct vop2_win *to_vop2_win(struct drm_plane *p) return container_of(p, struct vop2_win, base); } +/* + * Note: + * The write mask function is documented but missing on rk3566/8, writes + * to these bits have no effect. For newer soc(rk3588 and following) the + * write mask is needed for register writes. + * + * GLB_CFG_DONE_EN has no write mask bit. + * + */ +static inline void vop2_cfg_done(struct vop2_video_port *vp) +{ + struct vop2 *vop2 = vp->vop2; + u32 val = RK3568_REG_CFG_DONE__GLB_CFG_DONE_EN; + + val |= BIT(vp->id) | (BIT(vp->id) << 16); + + regmap_set_bits(vop2->map, RK3568_REG_CFG_DONE, val); +} + #endif /* _ROCKCHIP_DRM_VOP2_H */ diff --git a/drivers/gpu/drm/rockchip/rockchip_lvds.c b/drivers/gpu/drm/rockchip/rockchip_lvds.c index a673779de3d2..2411260db51d 100644 --- a/drivers/gpu/drm/rockchip/rockchip_lvds.c +++ b/drivers/gpu/drm/rockchip/rockchip_lvds.c @@ -56,14 +56,13 @@ struct rockchip_lvds { struct drm_device *drm_dev; struct drm_panel *panel; struct drm_bridge *bridge; - struct drm_connector connector; struct rockchip_encoder encoder; struct dev_pin_info *pins; }; -static inline struct rockchip_lvds *connector_to_lvds(struct drm_connector *connector) +static inline struct rockchip_lvds *brige_to_lvds(struct drm_bridge *bridge) { - return container_of(connector, struct rockchip_lvds, connector); + return (struct rockchip_lvds *)bridge->driver_private; } static inline struct rockchip_lvds *encoder_to_lvds(struct drm_encoder *encoder) @@ -106,25 +105,21 @@ static inline int rockchip_lvds_name_to_output(const char *s) return -EINVAL; } -static const struct drm_connector_funcs rockchip_lvds_connector_funcs = { - .fill_modes = drm_helper_probe_single_connector_modes, - .destroy = drm_connector_cleanup, - .reset = drm_atomic_helper_connector_reset, - .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, - .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, -}; - -static int rockchip_lvds_connector_get_modes(struct drm_connector *connector) +static int +rockchip_lvds_bridge_get_modes(struct drm_bridge *bridge, struct drm_connector *connector) { - struct rockchip_lvds *lvds = connector_to_lvds(connector); + struct rockchip_lvds *lvds = brige_to_lvds(bridge); struct drm_panel *panel = lvds->panel; return drm_panel_get_modes(panel, connector); } static const -struct drm_connector_helper_funcs rockchip_lvds_connector_helper_funcs = { - .get_modes = rockchip_lvds_connector_get_modes, +struct drm_bridge_funcs rockchip_lvds_bridge_funcs = { + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, + .get_modes = rockchip_lvds_bridge_get_modes, }; static int @@ -606,26 +601,23 @@ static int rockchip_lvds_bind(struct device *dev, struct device *master, } drm_encoder_helper_add(encoder, lvds->soc_data->helper_funcs); - connector = &lvds->connector; if (lvds->panel) { - connector->dpms = DRM_MODE_DPMS_OFF; - ret = drm_connector_init(drm_dev, connector, - &rockchip_lvds_connector_funcs, - DRM_MODE_CONNECTOR_LVDS); - if (ret < 0) { - drm_err(drm_dev, - "failed to initialize connector: %d\n", ret); + lvds->bridge = drm_panel_bridge_add_typed(lvds->panel, DRM_MODE_CONNECTOR_LVDS); + if (IS_ERR(lvds->bridge)) { + ret = PTR_ERR(lvds->bridge); goto err_free_encoder; } + } - drm_connector_helper_add(connector, - &rockchip_lvds_connector_helper_funcs); - } else { - ret = drm_bridge_attach(encoder, lvds->bridge, NULL, - DRM_BRIDGE_ATTACH_NO_CONNECTOR); + if (lvds->bridge) { + lvds->bridge->driver_private = lvds; + lvds->bridge->ops = DRM_BRIDGE_OP_MODES; + lvds->bridge->funcs = &rockchip_lvds_bridge_funcs; + + ret = drm_bridge_attach(encoder, lvds->bridge, NULL, DRM_BRIDGE_ATTACH_NO_CONNECTOR); if (ret) - goto err_free_encoder; + goto err_free_bridge; connector = drm_bridge_connector_init(lvds->drm_dev, encoder); if (IS_ERR(connector)) { @@ -633,14 +625,14 @@ static int rockchip_lvds_bind(struct device *dev, struct device *master, "failed to initialize bridge connector: %pe\n", connector); ret = PTR_ERR(connector); - goto err_free_encoder; + goto err_free_bridge; } - } - ret = drm_connector_attach_encoder(connector, encoder); - if (ret < 0) { - drm_err(drm_dev, "failed to attach encoder: %d\n", ret); - goto err_free_connector; + ret = drm_connector_attach_encoder(connector, encoder); + if (ret < 0) { + drm_err(drm_dev, "failed to attach encoder: %d\n", ret); + goto err_free_bridge; + } } pm_runtime_enable(dev); @@ -649,8 +641,8 @@ static int rockchip_lvds_bind(struct device *dev, struct device *master, return 0; -err_free_connector: - drm_connector_cleanup(connector); +err_free_bridge: + drm_panel_bridge_remove(lvds->bridge); err_free_encoder: drm_encoder_cleanup(encoder); err_put_remote: @@ -670,8 +662,6 @@ static void rockchip_lvds_unbind(struct device *dev, struct device *master, encoder_funcs = lvds->soc_data->helper_funcs; encoder_funcs->disable(&lvds->encoder.encoder); pm_runtime_disable(dev); - drm_connector_cleanup(&lvds->connector); - drm_encoder_cleanup(&lvds->encoder.encoder); } static const struct component_ops rockchip_lvds_component_ops = { diff --git a/drivers/gpu/drm/rockchip/rockchip_vop2_reg.c b/drivers/gpu/drm/rockchip/rockchip_vop2_reg.c index 32c4ed685739..45c5e3987813 100644 --- a/drivers/gpu/drm/rockchip/rockchip_vop2_reg.c +++ b/drivers/gpu/drm/rockchip/rockchip_vop2_reg.c @@ -2052,12 +2052,55 @@ static void vop2_setup_alpha(struct vop2_video_port *vp) } } +static u32 rk3568_vop2_read_port_mux(struct vop2 *vop2) +{ + return vop2_readl(vop2, RK3568_OVL_PORT_SEL); +} + +static void rk3568_vop2_wait_for_port_mux_done(struct vop2 *vop2) +{ + u32 port_mux_sel; + int ret; + + /* + * Spin until the previous port_mux figuration is done. + */ + ret = readx_poll_timeout_atomic(rk3568_vop2_read_port_mux, vop2, port_mux_sel, + port_mux_sel == vop2->old_port_sel, 0, 50 * 1000); + if (ret) + DRM_DEV_ERROR(vop2->dev, "wait port_mux done timeout: 0x%x--0x%x\n", + port_mux_sel, vop2->old_port_sel); +} + +static u32 rk3568_vop2_read_layer_cfg(struct vop2 *vop2) +{ + return vop2_readl(vop2, RK3568_OVL_LAYER_SEL); +} + +static void rk3568_vop2_wait_for_layer_cfg_done(struct vop2 *vop2, u32 cfg) +{ + u32 atv_layer_cfg; + int ret; + + /* + * Spin until the previous layer configuration is done. + */ + ret = readx_poll_timeout_atomic(rk3568_vop2_read_layer_cfg, vop2, atv_layer_cfg, + atv_layer_cfg == cfg, 0, 50 * 1000); + if (ret) + DRM_DEV_ERROR(vop2->dev, "wait layer cfg done timeout: 0x%x--0x%x\n", + atv_layer_cfg, cfg); +} + static void rk3568_vop2_setup_layer_mixer(struct vop2_video_port *vp) { struct vop2 *vop2 = vp->vop2; struct drm_plane *plane; u32 layer_sel = 0; u32 port_sel; + u32 old_layer_sel = 0; + u32 atv_layer_sel = 0; + u32 old_port_sel = 0; u8 layer_id; u8 old_layer_id; u8 layer_sel_id; @@ -2069,19 +2112,18 @@ static void rk3568_vop2_setup_layer_mixer(struct vop2_video_port *vp) struct vop2_video_port *vp2 = &vop2->vps[2]; struct rockchip_crtc_state *vcstate = to_rockchip_crtc_state(vp->crtc.state); + mutex_lock(&vop2->ovl_lock); ovl_ctrl = vop2_readl(vop2, RK3568_OVL_CTRL); ovl_ctrl &= ~RK3568_OVL_CTRL__LAYERSEL_REGDONE_IMD; ovl_ctrl &= ~RK3568_OVL_CTRL__LAYERSEL_REGDONE_SEL; - ovl_ctrl |= FIELD_PREP(RK3568_OVL_CTRL__LAYERSEL_REGDONE_SEL, vp->id); if (vcstate->yuv_overlay) ovl_ctrl |= RK3568_OVL_CTRL__YUV_MODE(vp->id); else ovl_ctrl &= ~RK3568_OVL_CTRL__YUV_MODE(vp->id); - vop2_writel(vop2, RK3568_OVL_CTRL, ovl_ctrl); - - port_sel = vop2_readl(vop2, RK3568_OVL_PORT_SEL); + old_port_sel = vop2->old_port_sel; + port_sel = old_port_sel; port_sel &= RK3568_OVL_PORT_SEL__SEL_PORT; if (vp0->nlayers) @@ -2102,7 +2144,13 @@ static void rk3568_vop2_setup_layer_mixer(struct vop2_video_port *vp) else port_sel |= FIELD_PREP(RK3568_OVL_PORT_SET__PORT2_MUX, 8); - layer_sel = vop2_readl(vop2, RK3568_OVL_LAYER_SEL); + /* Fixed value for rk3588 */ + if (vop2->version == VOP_VERSION_RK3588) + port_sel |= FIELD_PREP(RK3588_OVL_PORT_SET__PORT3_MUX, 7); + + atv_layer_sel = vop2_readl(vop2, RK3568_OVL_LAYER_SEL); + old_layer_sel = vop2->old_layer_sel; + layer_sel = old_layer_sel; ofs = 0; for (i = 0; i < vp->id; i++) @@ -2186,8 +2234,37 @@ static void rk3568_vop2_setup_layer_mixer(struct vop2_video_port *vp) old_win->data->layer_sel_id[vp->id]); } + vop2->old_layer_sel = layer_sel; + vop2->old_port_sel = port_sel; + /* + * As the RK3568_OVL_LAYER_SEL and RK3568_OVL_PORT_SEL are shared by all Video Ports, + * and the configuration take effect by one Video Port's vsync. + * When performing layer migration or change the zpos of layers, there are two things + * to be observed and followed: + * 1. When a layer is migrated from one VP to another, the configuration of the layer + * can only take effect after the Port mux configuration is enabled. + * + * 2. When we change the zpos of layers, we must ensure that the change for the previous + * VP takes effect before we proceed to change the next VP. Otherwise, the new + * configuration might overwrite the previous one for the previous VP, or it could + * lead to the configuration of the previous VP being take effect along with the VSYNC + * of the new VP. + */ + if (layer_sel != old_layer_sel || port_sel != old_port_sel) + ovl_ctrl |= FIELD_PREP(RK3568_OVL_CTRL__LAYERSEL_REGDONE_SEL, vp->id); + vop2_writel(vop2, RK3568_OVL_CTRL, ovl_ctrl); + + if (port_sel != old_port_sel) { + vop2_writel(vop2, RK3568_OVL_PORT_SEL, port_sel); + vop2_cfg_done(vp); + rk3568_vop2_wait_for_port_mux_done(vop2); + } + + if (layer_sel != old_layer_sel && atv_layer_sel != old_layer_sel) + rk3568_vop2_wait_for_layer_cfg_done(vop2, vop2->old_layer_sel); + vop2_writel(vop2, RK3568_OVL_LAYER_SEL, layer_sel); - vop2_writel(vop2, RK3568_OVL_PORT_SEL, port_sel); + mutex_unlock(&vop2->ovl_lock); } static void rk3568_vop2_setup_dly_for_windows(struct vop2_video_port *vp) diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index c63543132f9d..81ad40d9582b 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -84,12 +84,6 @@ #define CREATE_TRACE_POINTS #include "gpu_scheduler_trace.h" -#ifdef CONFIG_LOCKDEP -static struct lockdep_map drm_sched_lockdep_map = { - .name = "drm_sched_lockdep_map" -}; -#endif - int drm_sched_policy = DRM_SCHED_POLICY_FIFO; /** @@ -269,38 +263,14 @@ drm_sched_rq_select_entity_rr(struct drm_gpu_scheduler *sched, entity = rq->current_entity; if (entity) { list_for_each_entry_continue(entity, &rq->entities, list) { - if (drm_sched_entity_is_ready(entity)) { - /* If we can't queue yet, preserve the current - * entity in terms of fairness. - */ - if (!drm_sched_can_queue(sched, entity)) { - spin_unlock(&rq->lock); - return ERR_PTR(-ENOSPC); - } - - rq->current_entity = entity; - reinit_completion(&entity->entity_idle); - spin_unlock(&rq->lock); - return entity; - } + if (drm_sched_entity_is_ready(entity)) + goto found; } } list_for_each_entry(entity, &rq->entities, list) { - if (drm_sched_entity_is_ready(entity)) { - /* If we can't queue yet, preserve the current entity in - * terms of fairness. - */ - if (!drm_sched_can_queue(sched, entity)) { - spin_unlock(&rq->lock); - return ERR_PTR(-ENOSPC); - } - - rq->current_entity = entity; - reinit_completion(&entity->entity_idle); - spin_unlock(&rq->lock); - return entity; - } + if (drm_sched_entity_is_ready(entity)) + goto found; if (entity == rq->current_entity) break; @@ -309,6 +279,22 @@ drm_sched_rq_select_entity_rr(struct drm_gpu_scheduler *sched, spin_unlock(&rq->lock); return NULL; + +found: + if (!drm_sched_can_queue(sched, entity)) { + /* + * If scheduler cannot take more jobs signal the caller to not + * consider lower priority queues. + */ + entity = ERR_PTR(-ENOSPC); + } else { + rq->current_entity = entity; + reinit_completion(&entity->entity_idle); + } + + spin_unlock(&rq->lock); + + return entity; } /** @@ -1261,6 +1247,25 @@ static void drm_sched_run_job_work(struct work_struct *w) drm_sched_run_job_queue(sched); } +static struct workqueue_struct *drm_sched_alloc_wq(const char *name) +{ +#if (IS_ENABLED(CONFIG_LOCKDEP)) + static struct lockdep_map map = { + .name = "drm_sched_lockdep_map" + }; + + /* + * Avoid leaking a lockdep map on each drm sched creation and + * destruction by using a single lockdep map for all drm sched + * allocated submit_wq. + */ + + return alloc_ordered_workqueue_lockdep_map(name, WQ_MEM_RECLAIM, &map); +#else + return alloc_ordered_workqueue(name, WQ_MEM_RECLAIM); +#endif +} + /** * drm_sched_init - Init a gpu scheduler instance * @@ -1301,13 +1306,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched, const struct drm_sched_init_ sched->submit_wq = args->submit_wq; sched->own_submit_wq = false; } else { -#ifdef CONFIG_LOCKDEP - sched->submit_wq = alloc_ordered_workqueue_lockdep_map(args->name, - WQ_MEM_RECLAIM, - &drm_sched_lockdep_map); -#else - sched->submit_wq = alloc_ordered_workqueue(args->name, WQ_MEM_RECLAIM); -#endif + sched->submit_wq = drm_sched_alloc_wq(args->name); if (!sched->submit_wq) return -ENOMEM; diff --git a/drivers/gpu/drm/scheduler/tests/mock_scheduler.c b/drivers/gpu/drm/scheduler/tests/mock_scheduler.c index 7f947ab9d322..49d067fecd67 100644 --- a/drivers/gpu/drm/scheduler/tests/mock_scheduler.c +++ b/drivers/gpu/drm/scheduler/tests/mock_scheduler.c @@ -200,12 +200,36 @@ static struct dma_fence *mock_sched_run_job(struct drm_sched_job *sched_job) return &job->hw_fence; } +/* + * Normally, drivers would take appropriate measures in this callback, such as + * killing the entity the faulty job is associated with, resetting the hardware + * and / or resubmitting non-faulty jobs. + * + * For the mock scheduler, there are no hardware rings to be resetted nor jobs + * to be resubmitted. Thus, this function merely ensures that + * a) timedout fences get signaled properly and removed from the pending list + * b) the mock scheduler framework gets informed about the timeout via a flag + * c) The drm_sched_job, not longer needed, gets freed + */ static enum drm_gpu_sched_stat mock_sched_timedout_job(struct drm_sched_job *sched_job) { + struct drm_mock_scheduler *sched = drm_sched_to_mock_sched(sched_job->sched); struct drm_mock_sched_job *job = drm_sched_job_to_mock_job(sched_job); + unsigned long flags; - job->flags |= DRM_MOCK_SCHED_JOB_TIMEDOUT; + spin_lock_irqsave(&sched->lock, flags); + if (!dma_fence_is_signaled_locked(&job->hw_fence)) { + list_del(&job->link); + job->flags |= DRM_MOCK_SCHED_JOB_TIMEDOUT; + dma_fence_set_error(&job->hw_fence, -ETIMEDOUT); + dma_fence_signal_locked(&job->hw_fence); + } + spin_unlock_irqrestore(&sched->lock, flags); + + dma_fence_put(&job->hw_fence); + drm_sched_job_cleanup(sched_job); + /* Mock job itself is freed by the kunit framework. */ return DRM_GPU_SCHED_STAT_NOMINAL; } diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c index d202b6c1eb8f..2c015f563de9 100644 --- a/drivers/gpu/drm/sti/sti_hda.c +++ b/drivers/gpu/drm/sti/sti_hda.c @@ -246,6 +246,7 @@ struct sti_hda { struct device dev; struct drm_device *drm_dev; struct drm_display_mode mode; + struct drm_bridge bridge; void __iomem *regs; void __iomem *video_dacs_ctrl; struct clk *clk_pix; @@ -262,6 +263,11 @@ struct sti_hda_connector { #define to_sti_hda_connector(x) \ container_of(x, struct sti_hda_connector, drm_connector) +static struct sti_hda *drm_bridge_to_sti_hda(struct drm_bridge *bridge) +{ + return container_of(bridge, struct sti_hda, bridge); +} + static u32 hda_read(struct sti_hda *hda, int offset) { return readl(hda->regs + offset); @@ -401,7 +407,7 @@ static void sti_hda_configure_awg(struct sti_hda *hda, u32 *awg_instr, int nb) static void sti_hda_disable(struct drm_bridge *bridge) { - struct sti_hda *hda = bridge->driver_private; + struct sti_hda *hda = drm_bridge_to_sti_hda(bridge); u32 val; if (!hda->enabled) @@ -426,7 +432,7 @@ static void sti_hda_disable(struct drm_bridge *bridge) static void sti_hda_pre_enable(struct drm_bridge *bridge) { - struct sti_hda *hda = bridge->driver_private; + struct sti_hda *hda = drm_bridge_to_sti_hda(bridge); u32 val, i, mode_idx; u32 src_filter_y, src_filter_c; u32 *coef_y, *coef_c; @@ -517,7 +523,7 @@ static void sti_hda_set_mode(struct drm_bridge *bridge, const struct drm_display_mode *mode, const struct drm_display_mode *adjusted_mode) { - struct sti_hda *hda = bridge->driver_private; + struct sti_hda *hda = drm_bridge_to_sti_hda(bridge); u32 mode_idx; int hddac_rate; int ret; @@ -677,7 +683,6 @@ static int sti_hda_bind(struct device *dev, struct device *master, void *data) struct drm_encoder *encoder; struct sti_hda_connector *connector; struct drm_connector *drm_connector; - struct drm_bridge *bridge; int err; /* Set the drm device handle */ @@ -693,13 +698,7 @@ static int sti_hda_bind(struct device *dev, struct device *master, void *data) connector->hda = hda; - bridge = devm_kzalloc(dev, sizeof(*bridge), GFP_KERNEL); - if (!bridge) - return -ENOMEM; - - bridge->driver_private = hda; - bridge->funcs = &sti_hda_bridge_funcs; - drm_bridge_attach(encoder, bridge, NULL, 0); + drm_bridge_attach(encoder, &hda->bridge, NULL, 0); connector->encoder = encoder; @@ -745,9 +744,9 @@ static int sti_hda_probe(struct platform_device *pdev) DRM_INFO("%s\n", __func__); - hda = devm_kzalloc(dev, sizeof(*hda), GFP_KERNEL); - if (!hda) - return -ENOMEM; + hda = devm_drm_bridge_alloc(dev, struct sti_hda, bridge, &sti_hda_bridge_funcs); + if (IS_ERR(hda)) + return PTR_ERR(hda); hda->dev = pdev->dev; hda->regs = devm_platform_ioremap_resource_byname(pdev, "hda-reg"); diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c index 37b8d619066e..4e7c3d78b2b9 100644 --- a/drivers/gpu/drm/sti/sti_hdmi.c +++ b/drivers/gpu/drm/sti/sti_hdmi.c @@ -168,6 +168,11 @@ struct sti_hdmi_connector { #define to_sti_hdmi_connector(x) \ container_of(x, struct sti_hdmi_connector, drm_connector) +static struct sti_hdmi *drm_bridge_to_sti_hdmi(struct drm_bridge *bridge) +{ + return container_of(bridge, struct sti_hdmi, bridge); +} + static const struct drm_prop_enum_list colorspace_mode_names[] = { { HDMI_COLORSPACE_RGB, "rgb" }, { HDMI_COLORSPACE_YUV422, "yuv422" }, @@ -749,7 +754,7 @@ static void hdmi_debugfs_init(struct sti_hdmi *hdmi, struct drm_minor *minor) static void sti_hdmi_disable(struct drm_bridge *bridge) { - struct sti_hdmi *hdmi = bridge->driver_private; + struct sti_hdmi *hdmi = drm_bridge_to_sti_hdmi(bridge); u32 val = hdmi_read(hdmi, HDMI_CFG); @@ -881,7 +886,7 @@ static int hdmi_audio_configure(struct sti_hdmi *hdmi) static void sti_hdmi_pre_enable(struct drm_bridge *bridge) { - struct sti_hdmi *hdmi = bridge->driver_private; + struct sti_hdmi *hdmi = drm_bridge_to_sti_hdmi(bridge); DRM_DEBUG_DRIVER("\n"); @@ -936,7 +941,7 @@ static void sti_hdmi_set_mode(struct drm_bridge *bridge, const struct drm_display_mode *mode, const struct drm_display_mode *adjusted_mode) { - struct sti_hdmi *hdmi = bridge->driver_private; + struct sti_hdmi *hdmi = drm_bridge_to_sti_hdmi(bridge); int ret; DRM_DEBUG_DRIVER("\n"); @@ -1273,7 +1278,6 @@ static int sti_hdmi_bind(struct device *dev, struct device *master, void *data) struct sti_hdmi_connector *connector; struct cec_connector_info conn_info; struct drm_connector *drm_connector; - struct drm_bridge *bridge; int err; /* Set the drm device handle */ @@ -1289,13 +1293,7 @@ static int sti_hdmi_bind(struct device *dev, struct device *master, void *data) connector->hdmi = hdmi; - bridge = devm_kzalloc(dev, sizeof(*bridge), GFP_KERNEL); - if (!bridge) - return -EINVAL; - - bridge->driver_private = hdmi; - bridge->funcs = &sti_hdmi_bridge_funcs; - drm_bridge_attach(encoder, bridge, NULL, 0); + drm_bridge_attach(encoder, &hdmi->bridge, NULL, 0); connector->encoder = encoder; @@ -1385,9 +1383,9 @@ static int sti_hdmi_probe(struct platform_device *pdev) DRM_INFO("%s\n", __func__); - hdmi = devm_kzalloc(dev, sizeof(*hdmi), GFP_KERNEL); - if (!hdmi) - return -ENOMEM; + hdmi = devm_drm_bridge_alloc(dev, struct sti_hdmi, bridge, &sti_hdmi_bridge_funcs); + if (IS_ERR(hdmi)) + return PTR_ERR(hdmi); ddc = of_parse_phandle(pdev->dev.of_node, "ddc", 0); if (ddc) { diff --git a/drivers/gpu/drm/sti/sti_hdmi.h b/drivers/gpu/drm/sti/sti_hdmi.h index 6d4c3f57bc46..91d43dd46f13 100644 --- a/drivers/gpu/drm/sti/sti_hdmi.h +++ b/drivers/gpu/drm/sti/sti_hdmi.h @@ -12,6 +12,7 @@ #include <media/cec-notifier.h> +#include <drm/drm_bridge.h> #include <drm/drm_modes.h> #include <drm/drm_property.h> @@ -86,6 +87,7 @@ struct sti_hdmi { struct hdmi_audio_params audio; struct drm_connector *drm_connector; struct cec_notifier *notifier; + struct drm_bridge bridge; }; u32 hdmi_read(struct sti_hdmi *hdmi, int offset); diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c index dbc1394f96b8..41a285ec889f 100644 --- a/drivers/gpu/drm/tegra/gem.c +++ b/drivers/gpu/drm/tegra/gem.c @@ -523,10 +523,10 @@ void tegra_bo_free_object(struct drm_gem_object *gem) if (tegra->domain) { tegra_bo_iommu_unmap(tegra, bo); - if (gem->import_attach) { + if (drm_gem_is_imported(gem)) { dma_buf_unmap_attachment_unlocked(gem->import_attach, bo->sgt, DMA_TO_DEVICE); - dma_buf_detach(gem->import_attach->dmabuf, gem->import_attach); + dma_buf_detach(gem->dma_buf, gem->import_attach); } } diff --git a/drivers/gpu/drm/tests/drm_format_helper_test.c b/drivers/gpu/drm/tests/drm_format_helper_test.c index ad06762db671..7299fa8971ce 100644 --- a/drivers/gpu/drm/tests/drm_format_helper_test.c +++ b/drivers/gpu/drm/tests/drm_format_helper_test.c @@ -735,13 +735,13 @@ static void drm_test_fb_xrgb8888_to_rgb565(struct kunit *test) NULL : &result->dst_pitch; drm_fb_xrgb8888_to_rgb565(&dst, dst_pitch, &src, &fb, ¶ms->clip, - &fmtcnv_state, false); + &fmtcnv_state); buf = le16buf_to_cpu(test, (__force const __le16 *)buf, dst_size / sizeof(__le16)); KUNIT_EXPECT_MEMEQ(test, buf, result->expected, dst_size); buf = dst.vaddr; /* restore original value of buf */ - drm_fb_xrgb8888_to_rgb565(&dst, &result->dst_pitch, &src, &fb, ¶ms->clip, - &fmtcnv_state, true); + drm_fb_xrgb8888_to_rgb565be(&dst, &result->dst_pitch, &src, &fb, ¶ms->clip, + &fmtcnv_state); buf = le16buf_to_cpu(test, (__force const __le16 *)buf, dst_size / sizeof(__le16)); KUNIT_EXPECT_MEMEQ(test, buf, result->expected_swab, dst_size); @@ -749,7 +749,7 @@ static void drm_test_fb_xrgb8888_to_rgb565(struct kunit *test) memset(buf, 0, dst_size); drm_fb_xrgb8888_to_rgb565(&dst, dst_pitch, &src, &fb, ¶ms->clip, - &fmtcnv_state, false); + &fmtcnv_state); buf = le16buf_to_cpu(test, (__force const __le16 *)buf, dst_size / sizeof(__le16)); KUNIT_EXPECT_MEMEQ(test, buf, result->expected, dst_size); } diff --git a/drivers/gpu/drm/tidss/Makefile b/drivers/gpu/drm/tidss/Makefile index 312645271014..b6d6becf1683 100644 --- a/drivers/gpu/drm/tidss/Makefile +++ b/drivers/gpu/drm/tidss/Makefile @@ -7,6 +7,7 @@ tidss-y := tidss_crtc.o \ tidss_irq.o \ tidss_plane.o \ tidss_scale_coefs.o \ - tidss_dispc.o + tidss_dispc.o \ + tidss_oldi.o obj-$(CONFIG_DRM_TIDSS) += tidss.o diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c index 21363ccbd763..c0277fa36425 100644 --- a/drivers/gpu/drm/tidss/tidss_dispc.c +++ b/drivers/gpu/drm/tidss/tidss_dispc.c @@ -146,7 +146,7 @@ static const u16 tidss_am65x_common_regs[DISPC_COMMON_REG_TABLE_LEN] = { const struct dispc_features dispc_am65x_feats = { .max_pclk_khz = { [DISPC_VP_DPI] = 165000, - [DISPC_VP_OLDI] = 165000, + [DISPC_VP_OLDI_AM65X] = 165000, }, .scaling = { @@ -176,7 +176,7 @@ const struct dispc_features dispc_am65x_feats = { .vp_name = { "vp1", "vp2" }, .ovr_name = { "ovr1", "ovr2" }, .vpclk_name = { "vp1", "vp2" }, - .vp_bus_type = { DISPC_VP_OLDI, DISPC_VP_DPI }, + .vp_bus_type = { DISPC_VP_OLDI_AM65X, DISPC_VP_DPI }, .vp_feat = { .color = { .has_ctm = true, @@ -491,7 +491,7 @@ struct dispc_device { void __iomem *base_ovr[TIDSS_MAX_PORTS]; void __iomem *base_vp[TIDSS_MAX_PORTS]; - struct regmap *oldi_io_ctrl; + struct regmap *am65x_oldi_io_ctrl; struct clk *vp_clk[TIDSS_MAX_PORTS]; @@ -566,6 +566,29 @@ static u32 dispc_vp_read(struct dispc_device *dispc, u32 hw_videoport, u16 reg) return ioread32(base + reg); } +int tidss_configure_oldi(struct tidss_device *tidss, u32 hw_videoport, + u32 oldi_cfg) +{ + u32 count = 0; + u32 oldi_reset_bit = BIT(5 + hw_videoport); + + dispc_vp_write(tidss->dispc, hw_videoport, DISPC_VP_DSS_OLDI_CFG, oldi_cfg); + + while (!(oldi_reset_bit & dispc_read(tidss->dispc, DSS_SYSSTATUS)) && + count < 10000) + count++; + + if (!(oldi_reset_bit & dispc_read(tidss->dispc, DSS_SYSSTATUS))) + return -ETIMEDOUT; + + return 0; +} + +void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport) +{ + dispc_vp_write(tidss->dispc, hw_videoport, DISPC_VP_DSS_OLDI_CFG, 0); +} + /* * TRM gives bitfields as start:end, where start is the higher bit * number. For example 7:0 @@ -1016,13 +1039,11 @@ void dispc_set_irqenable(struct dispc_device *dispc, dispc_irq_t mask) } } -enum dispc_oldi_mode_reg_val { SPWG_18 = 0, JEIDA_24 = 1, SPWG_24 = 2 }; - struct dispc_bus_format { u32 bus_fmt; u32 data_width; bool is_oldi_fmt; - enum dispc_oldi_mode_reg_val oldi_mode_reg_val; + enum oldi_mode_reg_val am65x_oldi_mode_reg_val; }; static const struct dispc_bus_format dispc_bus_formats[] = { @@ -1066,7 +1087,7 @@ int dispc_vp_bus_check(struct dispc_device *dispc, u32 hw_videoport, return -EINVAL; } - if (dispc->feat->vp_bus_type[hw_videoport] != DISPC_VP_OLDI && + if (dispc->feat->vp_bus_type[hw_videoport] != DISPC_VP_OLDI_AM65X && fmt->is_oldi_fmt) { dev_dbg(dispc->dev, "%s: %s is not OLDI-port\n", __func__, dispc->feat->vp_name[hw_videoport]); @@ -1076,23 +1097,23 @@ int dispc_vp_bus_check(struct dispc_device *dispc, u32 hw_videoport, return 0; } -static void dispc_oldi_tx_power(struct dispc_device *dispc, bool power) +static void dispc_am65x_oldi_tx_power(struct dispc_device *dispc, bool power) { - u32 val = power ? 0 : OLDI_PWRDN_TX; + u32 val = power ? 0 : AM65X_OLDI_PWRDN_TX; - if (WARN_ON(!dispc->oldi_io_ctrl)) + if (WARN_ON(!dispc->am65x_oldi_io_ctrl)) return; - regmap_update_bits(dispc->oldi_io_ctrl, OLDI_DAT0_IO_CTRL, - OLDI_PWRDN_TX, val); - regmap_update_bits(dispc->oldi_io_ctrl, OLDI_DAT1_IO_CTRL, - OLDI_PWRDN_TX, val); - regmap_update_bits(dispc->oldi_io_ctrl, OLDI_DAT2_IO_CTRL, - OLDI_PWRDN_TX, val); - regmap_update_bits(dispc->oldi_io_ctrl, OLDI_DAT3_IO_CTRL, - OLDI_PWRDN_TX, val); - regmap_update_bits(dispc->oldi_io_ctrl, OLDI_CLK_IO_CTRL, - OLDI_PWRDN_TX, val); + regmap_update_bits(dispc->am65x_oldi_io_ctrl, AM65X_OLDI_DAT0_IO_CTRL, + AM65X_OLDI_PWRDN_TX, val); + regmap_update_bits(dispc->am65x_oldi_io_ctrl, AM65X_OLDI_DAT1_IO_CTRL, + AM65X_OLDI_PWRDN_TX, val); + regmap_update_bits(dispc->am65x_oldi_io_ctrl, AM65X_OLDI_DAT2_IO_CTRL, + AM65X_OLDI_PWRDN_TX, val); + regmap_update_bits(dispc->am65x_oldi_io_ctrl, AM65X_OLDI_DAT3_IO_CTRL, + AM65X_OLDI_PWRDN_TX, val); + regmap_update_bits(dispc->am65x_oldi_io_ctrl, AM65X_OLDI_CLK_IO_CTRL, + AM65X_OLDI_PWRDN_TX, val); } static void dispc_set_num_datalines(struct dispc_device *dispc, @@ -1121,8 +1142,8 @@ static void dispc_set_num_datalines(struct dispc_device *dispc, VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, v, 10, 8); } -static void dispc_enable_oldi(struct dispc_device *dispc, u32 hw_videoport, - const struct dispc_bus_format *fmt) +static void dispc_enable_am65x_oldi(struct dispc_device *dispc, u32 hw_videoport, + const struct dispc_bus_format *fmt) { u32 oldi_cfg = 0; u32 oldi_reset_bit = BIT(5 + hw_videoport); @@ -1141,7 +1162,7 @@ static void dispc_enable_oldi(struct dispc_device *dispc, u32 hw_videoport, oldi_cfg |= BIT(7); /* DEPOL */ - oldi_cfg = FLD_MOD(oldi_cfg, fmt->oldi_mode_reg_val, 3, 1); + oldi_cfg = FLD_MOD(oldi_cfg, fmt->am65x_oldi_mode_reg_val, 3, 1); oldi_cfg |= BIT(12); /* SOFTRST */ @@ -1170,10 +1191,10 @@ void dispc_vp_prepare(struct dispc_device *dispc, u32 hw_videoport, if (WARN_ON(!fmt)) return; - if (dispc->feat->vp_bus_type[hw_videoport] == DISPC_VP_OLDI) { - dispc_oldi_tx_power(dispc, true); + if (dispc->feat->vp_bus_type[hw_videoport] == DISPC_VP_OLDI_AM65X) { + dispc_am65x_oldi_tx_power(dispc, true); - dispc_enable_oldi(dispc, hw_videoport, fmt); + dispc_enable_am65x_oldi(dispc, hw_videoport, fmt); } } @@ -1229,7 +1250,7 @@ void dispc_vp_enable(struct dispc_device *dispc, u32 hw_videoport, align = true; /* always use DE_HIGH for OLDI */ - if (dispc->feat->vp_bus_type[hw_videoport] == DISPC_VP_OLDI) + if (dispc->feat->vp_bus_type[hw_videoport] == DISPC_VP_OLDI_AM65X) ieo = false; dispc_vp_write(dispc, hw_videoport, DISPC_VP_POL_FREQ, @@ -1255,10 +1276,10 @@ void dispc_vp_disable(struct dispc_device *dispc, u32 hw_videoport) void dispc_vp_unprepare(struct dispc_device *dispc, u32 hw_videoport) { - if (dispc->feat->vp_bus_type[hw_videoport] == DISPC_VP_OLDI) { + if (dispc->feat->vp_bus_type[hw_videoport] == DISPC_VP_OLDI_AM65X) { dispc_vp_write(dispc, hw_videoport, DISPC_VP_DSS_OLDI_CFG, 0); - dispc_oldi_tx_power(dispc, false); + dispc_am65x_oldi_tx_power(dispc, false); } } @@ -1420,7 +1441,6 @@ void dispc_vp_disable_clk(struct dispc_device *dispc, u32 hw_videoport) * Calculate the percentage difference between the requested pixel clock rate * and the effective rate resulting from calculating the clock divider value. */ -static unsigned int dispc_pclk_diff(unsigned long rate, unsigned long real_rate) { int r = rate / 100, rr = real_rate / 100; @@ -2852,15 +2872,15 @@ static int dispc_iomap_resource(struct platform_device *pdev, const char *name, static int dispc_init_am65x_oldi_io_ctrl(struct device *dev, struct dispc_device *dispc) { - dispc->oldi_io_ctrl = + dispc->am65x_oldi_io_ctrl = syscon_regmap_lookup_by_phandle(dev->of_node, "ti,am65x-oldi-io-ctrl"); - if (PTR_ERR(dispc->oldi_io_ctrl) == -ENODEV) { - dispc->oldi_io_ctrl = NULL; - } else if (IS_ERR(dispc->oldi_io_ctrl)) { + if (PTR_ERR(dispc->am65x_oldi_io_ctrl) == -ENODEV) { + dispc->am65x_oldi_io_ctrl = NULL; + } else if (IS_ERR(dispc->am65x_oldi_io_ctrl)) { dev_err(dev, "%s: syscon_regmap_lookup_by_phandle failed %ld\n", - __func__, PTR_ERR(dispc->oldi_io_ctrl)); - return PTR_ERR(dispc->oldi_io_ctrl); + __func__, PTR_ERR(dispc->am65x_oldi_io_ctrl)); + return PTR_ERR(dispc->am65x_oldi_io_ctrl); } return 0; } diff --git a/drivers/gpu/drm/tidss/tidss_dispc.h b/drivers/gpu/drm/tidss/tidss_dispc.h index 28958514b8f5..b8614f62186c 100644 --- a/drivers/gpu/drm/tidss/tidss_dispc.h +++ b/drivers/gpu/drm/tidss/tidss_dispc.h @@ -58,7 +58,7 @@ struct dispc_errata { enum dispc_vp_bus_type { DISPC_VP_DPI, /* DPI output */ - DISPC_VP_OLDI, /* OLDI (LVDS) output */ + DISPC_VP_OLDI_AM65X, /* OLDI (LVDS) output for AM65x DSS */ DISPC_VP_INTERNAL, /* SoC internal routing */ DISPC_VP_TIED_OFF, /* Tied off / Unavailable */ DISPC_VP_MAX_BUS_TYPE, @@ -101,6 +101,11 @@ extern const struct dispc_features dispc_am62l_feats; extern const struct dispc_features dispc_am65x_feats; extern const struct dispc_features dispc_j721e_feats; +int tidss_configure_oldi(struct tidss_device *tidss, u32 hw_videoport, + u32 oldi_cfg); +void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport); +unsigned int dispc_pclk_diff(unsigned long rate, unsigned long real_rate); + void dispc_set_irqenable(struct dispc_device *dispc, dispc_irq_t mask); dispc_irq_t dispc_read_and_clear_irqstatus(struct dispc_device *dispc); diff --git a/drivers/gpu/drm/tidss/tidss_dispc_regs.h b/drivers/gpu/drm/tidss/tidss_dispc_regs.h index e88148e44937..50a3f28250ef 100644 --- a/drivers/gpu/drm/tidss/tidss_dispc_regs.h +++ b/drivers/gpu/drm/tidss/tidss_dispc_regs.h @@ -226,18 +226,35 @@ enum dispc_common_regs { #define DISPC_VP_DSS_DMA_THREADSIZE 0x170 /* J721E */ #define DISPC_VP_DSS_DMA_THREADSIZE_STATUS 0x174 /* J721E */ +/* OLDI Config Bits (DISPC_VP_DSS_OLDI_CFG) */ +#define OLDI_ENABLE BIT(0) +#define OLDI_MAP (BIT(1) | BIT(2) | BIT(3)) +#define OLDI_SRC BIT(4) +#define OLDI_CLONE_MODE BIT(5) +#define OLDI_MASTERSLAVE BIT(6) +#define OLDI_DEPOL BIT(7) +#define OLDI_MSB BIT(8) +#define OLDI_LBEN BIT(9) +#define OLDI_LBDATA BIT(10) +#define OLDI_DUALMODESYNC BIT(11) +#define OLDI_SOFTRST BIT(12) +#define OLDI_TPATCFG BIT(13) + +/* LVDS Format values for OLDI_MAP field in DISPC_VP_OLDI_CFG register */ +enum oldi_mode_reg_val { SPWG_18 = 0, JEIDA_24 = 1, SPWG_24 = 2 }; + /* * OLDI IO_CTRL register offsets. On AM654 the registers are found * from CTRL_MMR0, there the syscon regmap should map 0x14 bytes from * CTRLMMR0P1_OLDI_DAT0_IO_CTRL to CTRLMMR0P1_OLDI_CLK_IO_CTRL * register range. */ -#define OLDI_DAT0_IO_CTRL 0x00 -#define OLDI_DAT1_IO_CTRL 0x04 -#define OLDI_DAT2_IO_CTRL 0x08 -#define OLDI_DAT3_IO_CTRL 0x0C -#define OLDI_CLK_IO_CTRL 0x10 +#define AM65X_OLDI_DAT0_IO_CTRL 0x00 +#define AM65X_OLDI_DAT1_IO_CTRL 0x04 +#define AM65X_OLDI_DAT2_IO_CTRL 0x08 +#define AM65X_OLDI_DAT3_IO_CTRL 0x0C +#define AM65X_OLDI_CLK_IO_CTRL 0x10 -#define OLDI_PWRDN_TX BIT(8) +#define AM65X_OLDI_PWRDN_TX BIT(8) #endif /* __TIDSS_DISPC_REGS_H */ diff --git a/drivers/gpu/drm/tidss/tidss_drv.c b/drivers/gpu/drm/tidss/tidss_drv.c index f2a4f659f574..a1b12e52aca4 100644 --- a/drivers/gpu/drm/tidss/tidss_drv.c +++ b/drivers/gpu/drm/tidss/tidss_drv.c @@ -24,6 +24,7 @@ #include "tidss_drv.h" #include "tidss_kms.h" #include "tidss_irq.h" +#include "tidss_oldi.h" /* Power management */ @@ -147,6 +148,10 @@ static int tidss_probe(struct platform_device *pdev) return ret; } + ret = tidss_oldi_init(tidss); + if (ret) + return dev_err_probe(dev, ret, "failed to init OLDI\n"); + pm_runtime_enable(dev); pm_runtime_set_autosuspend_delay(dev, 1000); @@ -203,6 +208,8 @@ err_runtime_suspend: pm_runtime_dont_use_autosuspend(dev); pm_runtime_disable(dev); + tidss_oldi_deinit(tidss); + return ret; } @@ -227,6 +234,8 @@ static void tidss_remove(struct platform_device *pdev) pm_runtime_dont_use_autosuspend(dev); pm_runtime_disable(dev); + tidss_oldi_deinit(tidss); + /* devm allocated dispc goes away with the dev so mark it NULL */ dispc_remove(tidss); diff --git a/drivers/gpu/drm/tidss/tidss_drv.h b/drivers/gpu/drm/tidss/tidss_drv.h index 7f4f4282bc04..d14d5d28f0a3 100644 --- a/drivers/gpu/drm/tidss/tidss_drv.h +++ b/drivers/gpu/drm/tidss/tidss_drv.h @@ -11,8 +11,10 @@ #define TIDSS_MAX_PORTS 4 #define TIDSS_MAX_PLANES 4 +#define TIDSS_MAX_OLDI_TXES 2 typedef u32 dispc_irq_t; +struct tidss_oldi; struct tidss_device { struct drm_device ddev; /* DRM device for DSS */ @@ -27,6 +29,9 @@ struct tidss_device { unsigned int num_planes; struct drm_plane *planes[TIDSS_MAX_PLANES]; + unsigned int num_oldis; + struct tidss_oldi *oldis[TIDSS_MAX_OLDI_TXES]; + unsigned int irq; /* protects the irq masks field and irqenable/irqstatus registers */ diff --git a/drivers/gpu/drm/tidss/tidss_kms.c b/drivers/gpu/drm/tidss/tidss_kms.c index 19432c08ec6b..c34eb90cddbe 100644 --- a/drivers/gpu/drm/tidss/tidss_kms.c +++ b/drivers/gpu/drm/tidss/tidss_kms.c @@ -144,7 +144,7 @@ static int tidss_dispc_modeset_init(struct tidss_device *tidss) dev_dbg(dev, "Setting up panel for port %d\n", i); switch (feat->vp_bus_type[i]) { - case DISPC_VP_OLDI: + case DISPC_VP_OLDI_AM65X: enc_type = DRM_MODE_ENCODER_LVDS; conn_type = DRM_MODE_CONNECTOR_LVDS; break; diff --git a/drivers/gpu/drm/tidss/tidss_oldi.c b/drivers/gpu/drm/tidss/tidss_oldi.c new file mode 100644 index 000000000000..8223b8fec8ce --- /dev/null +++ b/drivers/gpu/drm/tidss/tidss_oldi.c @@ -0,0 +1,598 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * Copyright (C) 2025 - Texas Instruments Incorporated + * + * Aradhya Bhatia <a-bhatia1@ti.com> + */ + +#include <linux/clk.h> +#include <linux/of.h> +#include <linux/of_graph.h> +#include <linux/mfd/syscon.h> +#include <linux/media-bus-format.h> +#include <linux/regmap.h> + +#include <drm/drm_atomic_helper.h> +#include <drm/drm_bridge.h> +#include <drm/drm_of.h> + +#include "tidss_dispc.h" +#include "tidss_dispc_regs.h" +#include "tidss_oldi.h" + +struct tidss_oldi { + struct tidss_device *tidss; + struct device *dev; + + struct drm_bridge bridge; + struct drm_bridge *next_bridge; + + enum tidss_oldi_link_type link_type; + const struct oldi_bus_format *bus_format; + u32 oldi_instance; + int companion_instance; /* -1 when OLDI TX operates in Single-Link */ + u32 parent_vp; + + struct clk *serial; + struct regmap *io_ctrl; +}; + +struct oldi_bus_format { + u32 bus_fmt; + u32 data_width; + enum oldi_mode_reg_val oldi_mode_reg_val; + u32 input_bus_fmt; +}; + +static const struct oldi_bus_format oldi_bus_formats[] = { + { MEDIA_BUS_FMT_RGB666_1X7X3_SPWG, 18, SPWG_18, MEDIA_BUS_FMT_RGB666_1X18 }, + { MEDIA_BUS_FMT_RGB888_1X7X4_SPWG, 24, SPWG_24, MEDIA_BUS_FMT_RGB888_1X24 }, + { MEDIA_BUS_FMT_RGB888_1X7X4_JEIDA, 24, JEIDA_24, MEDIA_BUS_FMT_RGB888_1X24 }, +}; + +#define OLDI_IDLE_CLK_HZ 25000000 /*25 MHz */ + +static inline struct tidss_oldi * +drm_bridge_to_tidss_oldi(struct drm_bridge *bridge) +{ + return container_of(bridge, struct tidss_oldi, bridge); +} + +static int tidss_oldi_bridge_attach(struct drm_bridge *bridge, + struct drm_encoder *encoder, + enum drm_bridge_attach_flags flags) +{ + struct tidss_oldi *oldi = drm_bridge_to_tidss_oldi(bridge); + + if (!oldi->next_bridge) { + dev_err(oldi->dev, + "%s: OLDI%u Failure attach next bridge\n", + __func__, oldi->oldi_instance); + return -ENODEV; + } + + if (!(flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR)) { + dev_err(oldi->dev, + "%s: OLDI%u DRM_BRIDGE_ATTACH_NO_CONNECTOR is mandatory.\n", + __func__, oldi->oldi_instance); + return -EINVAL; + } + + return drm_bridge_attach(encoder, oldi->next_bridge, bridge, flags); +} + +static int +tidss_oldi_set_serial_clk(struct tidss_oldi *oldi, unsigned long rate) +{ + unsigned long new_rate; + int ret; + + ret = clk_set_rate(oldi->serial, rate); + if (ret) { + dev_err(oldi->dev, + "OLDI%u: failed to set serial clk rate to %lu Hz\n", + oldi->oldi_instance, rate); + return ret; + } + + new_rate = clk_get_rate(oldi->serial); + + if (dispc_pclk_diff(rate, new_rate) > 5) + dev_warn(oldi->dev, + "OLDI%u Clock rate %lu differs over 5%% from requested %lu\n", + oldi->oldi_instance, new_rate, rate); + + dev_dbg(oldi->dev, "OLDI%u: new rate %lu Hz (requested %lu Hz)\n", + oldi->oldi_instance, clk_get_rate(oldi->serial), rate); + + return 0; +} + +static void tidss_oldi_tx_power(struct tidss_oldi *oldi, bool enable) +{ + u32 mask; + + /* + * The power control bits are Active Low, and remain powered off by + * default. That is, the bits are set to 1. To power on the OLDI TXes, + * the bits must be cleared to 0. Since there are cases where not all + * OLDI TXes are being used, the power logic selectively powers them + * on. + * Setting the variable 'val' to particular bit masks, makes sure that + * the undesired OLDI TXes remain powered off. + */ + + if (enable) { + switch (oldi->link_type) { + case OLDI_MODE_SINGLE_LINK: + /* Power-on only the required OLDI TX's IO*/ + mask = OLDI_PWRDOWN_TX(oldi->oldi_instance) | OLDI_PWRDN_BG; + break; + case OLDI_MODE_CLONE_SINGLE_LINK: + case OLDI_MODE_DUAL_LINK: + /* Power-on both the OLDI TXes' IOs */ + mask = OLDI_PWRDOWN_TX(oldi->oldi_instance) | + OLDI_PWRDOWN_TX(oldi->companion_instance) | + OLDI_PWRDN_BG; + break; + default: + /* + * This code execution should never reach here as any + * OLDI with an unsupported OLDI mode would never get + * registered in the first place. + * However, power-off the OLDI in concern just in case. + */ + mask = OLDI_PWRDOWN_TX(oldi->oldi_instance); + enable = false; + break; + } + } else { + switch (oldi->link_type) { + case OLDI_MODE_CLONE_SINGLE_LINK: + case OLDI_MODE_DUAL_LINK: + mask = OLDI_PWRDOWN_TX(oldi->oldi_instance) | + OLDI_PWRDOWN_TX(oldi->companion_instance) | + OLDI_PWRDN_BG; + break; + case OLDI_MODE_SINGLE_LINK: + default: + mask = OLDI_PWRDOWN_TX(oldi->oldi_instance); + break; + } + } + + regmap_update_bits(oldi->io_ctrl, OLDI_PD_CTRL, mask, enable ? 0 : mask); +} + +static int tidss_oldi_config(struct tidss_oldi *oldi) +{ + const struct oldi_bus_format *bus_fmt = NULL; + u32 oldi_cfg = 0; + int ret; + + bus_fmt = oldi->bus_format; + + /* + * MASTERSLAVE and SRC bits of OLDI Config are always set to 0. + */ + + if (bus_fmt->data_width == 24) + oldi_cfg |= OLDI_MSB; + else if (bus_fmt->data_width != 18) + dev_warn(oldi->dev, + "OLDI%u: DSS port width %d not supported\n", + oldi->oldi_instance, bus_fmt->data_width); + + oldi_cfg |= OLDI_DEPOL; + + oldi_cfg = (oldi_cfg & (~OLDI_MAP)) | (bus_fmt->oldi_mode_reg_val << 1); + + oldi_cfg |= OLDI_SOFTRST; + + oldi_cfg |= OLDI_ENABLE; + + switch (oldi->link_type) { + case OLDI_MODE_SINGLE_LINK: + /* All configuration is done for this mode. */ + break; + + case OLDI_MODE_CLONE_SINGLE_LINK: + oldi_cfg |= OLDI_CLONE_MODE; + break; + + case OLDI_MODE_DUAL_LINK: + /* data-mapping field also indicates dual-link mode */ + oldi_cfg |= BIT(3); + oldi_cfg |= OLDI_DUALMODESYNC; + break; + + default: + dev_err(oldi->dev, "OLDI%u: Unsupported mode.\n", + oldi->oldi_instance); + return -EINVAL; + } + + ret = tidss_configure_oldi(oldi->tidss, oldi->parent_vp, oldi_cfg); + if (ret == -ETIMEDOUT) + dev_warn(oldi->dev, "OLDI%u: timeout waiting for OLDI reset done.\n", + oldi->oldi_instance); + + return ret; +} + +static void tidss_oldi_atomic_pre_enable(struct drm_bridge *bridge, + struct drm_atomic_state *state) +{ + struct tidss_oldi *oldi = drm_bridge_to_tidss_oldi(bridge); + struct drm_connector *connector; + struct drm_connector_state *conn_state; + struct drm_crtc_state *crtc_state; + struct drm_display_mode *mode; + + if (oldi->link_type == OLDI_MODE_SECONDARY_CLONE_SINGLE_LINK) + return; + + connector = drm_atomic_get_new_connector_for_encoder(state, + bridge->encoder); + if (WARN_ON(!connector)) + return; + + conn_state = drm_atomic_get_new_connector_state(state, connector); + if (WARN_ON(!conn_state)) + return; + + crtc_state = drm_atomic_get_new_crtc_state(state, conn_state->crtc); + if (WARN_ON(!crtc_state)) + return; + + mode = &crtc_state->adjusted_mode; + + /* Configure the OLDI params*/ + tidss_oldi_config(oldi); + + /* Set the OLDI serial clock (7 times the pixel clock) */ + tidss_oldi_set_serial_clk(oldi, mode->clock * 7 * 1000); + + /* Enable OLDI IO power */ + tidss_oldi_tx_power(oldi, true); +} + +static void tidss_oldi_atomic_post_disable(struct drm_bridge *bridge, + struct drm_atomic_state *state) +{ + struct tidss_oldi *oldi = drm_bridge_to_tidss_oldi(bridge); + + if (oldi->link_type == OLDI_MODE_SECONDARY_CLONE_SINGLE_LINK) + return; + + /* Disable OLDI IO power */ + tidss_oldi_tx_power(oldi, false); + + /* Set the OLDI serial clock to IDLE Frequency */ + tidss_oldi_set_serial_clk(oldi, OLDI_IDLE_CLK_HZ); + + /* Clear OLDI Config */ + tidss_disable_oldi(oldi->tidss, oldi->parent_vp); +} + +#define MAX_INPUT_SEL_FORMATS 1 + +static u32 *tidss_oldi_atomic_get_input_bus_fmts(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state, + u32 output_fmt, + unsigned int *num_input_fmts) +{ + struct tidss_oldi *oldi = drm_bridge_to_tidss_oldi(bridge); + u32 *input_fmts; + int i; + + *num_input_fmts = 0; + + for (i = 0; i < ARRAY_SIZE(oldi_bus_formats); i++) + if (oldi_bus_formats[i].bus_fmt == output_fmt) + break; + + if (i == ARRAY_SIZE(oldi_bus_formats)) + return NULL; + + input_fmts = kcalloc(MAX_INPUT_SEL_FORMATS, sizeof(*input_fmts), + GFP_KERNEL); + if (!input_fmts) + return NULL; + + *num_input_fmts = 1; + input_fmts[0] = oldi_bus_formats[i].input_bus_fmt; + oldi->bus_format = &oldi_bus_formats[i]; + + return input_fmts; +} + +static const struct drm_bridge_funcs tidss_oldi_bridge_funcs = { + .attach = tidss_oldi_bridge_attach, + .atomic_pre_enable = tidss_oldi_atomic_pre_enable, + .atomic_post_disable = tidss_oldi_atomic_post_disable, + .atomic_get_input_bus_fmts = tidss_oldi_atomic_get_input_bus_fmts, + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, +}; + +static int get_oldi_mode(struct device_node *oldi_tx, int *companion_instance) +{ + struct device_node *companion; + struct device_node *port0, *port1; + u32 companion_reg; + bool secondary_oldi = false; + int pixel_order; + + /* + * Find if the OLDI is paired with another OLDI for combined OLDI + * operation (dual-link or clone). + */ + companion = of_parse_phandle(oldi_tx, "ti,companion-oldi", 0); + if (!companion) + /* + * The OLDI TX does not have a companion, nor is it a + * secondary OLDI. It will operate independently. + */ + return OLDI_MODE_SINGLE_LINK; + + if (of_property_read_u32(companion, "reg", &companion_reg)) + return OLDI_MODE_UNSUPPORTED; + + if (companion_reg > (TIDSS_MAX_OLDI_TXES - 1)) + /* Invalid companion OLDI reg value. */ + return OLDI_MODE_UNSUPPORTED; + + *companion_instance = (int)companion_reg; + + if (of_property_read_bool(oldi_tx, "ti,secondary-oldi")) + secondary_oldi = true; + + /* + * We need to work out if the sink is expecting us to function in + * dual-link mode. We do this by looking at the DT port nodes, the + * OLDI TX ports are connected to. If they are marked as expecting + * even pixels and odd pixels, then we need to enable dual-link. + */ + port0 = of_graph_get_port_by_id(oldi_tx, 1); + port1 = of_graph_get_port_by_id(companion, 1); + pixel_order = drm_of_lvds_get_dual_link_pixel_order(port0, port1); + of_node_put(port0); + of_node_put(port1); + of_node_put(companion); + + switch (pixel_order) { + case -EINVAL: + /* + * The dual-link properties were not found in at least + * one of the sink nodes. Since 2 OLDI ports are present + * in the DT, it can be safely assumed that the required + * configuration is Clone Mode. + */ + return (secondary_oldi ? OLDI_MODE_SECONDARY_CLONE_SINGLE_LINK : + OLDI_MODE_CLONE_SINGLE_LINK); + + case DRM_LVDS_DUAL_LINK_ODD_EVEN_PIXELS: + /* + * Primary OLDI can only support "ODD" pixels. So, from its + * perspective, the pixel order has to be ODD-EVEN. + */ + return (secondary_oldi ? OLDI_MODE_UNSUPPORTED : + OLDI_MODE_DUAL_LINK); + + case DRM_LVDS_DUAL_LINK_EVEN_ODD_PIXELS: + /* + * Secondary OLDI can only support "EVEN" pixels. So, from its + * perspective, the pixel order has to be EVEN-ODD. + */ + return (secondary_oldi ? OLDI_MODE_SECONDARY_DUAL_LINK : + OLDI_MODE_UNSUPPORTED); + + default: + return OLDI_MODE_UNSUPPORTED; + } +} + +static int get_parent_dss_vp(struct device_node *oldi_tx, u32 *parent_vp) +{ + struct device_node *ep, *dss_port; + int ret; + + ep = of_graph_get_endpoint_by_regs(oldi_tx, OLDI_INPUT_PORT, -1); + if (ep) { + dss_port = of_graph_get_remote_port(ep); + if (!dss_port) { + ret = -ENODEV; + goto err_return_ep_port; + } + + ret = of_property_read_u32(dss_port, "reg", parent_vp); + + of_node_put(dss_port); +err_return_ep_port: + of_node_put(ep); + return ret; + } + + return -ENODEV; +} + +static const struct drm_bridge_timings default_tidss_oldi_timings = { + .input_bus_flags = DRM_BUS_FLAG_SYNC_SAMPLE_NEGEDGE + | DRM_BUS_FLAG_DE_HIGH, +}; + +void tidss_oldi_deinit(struct tidss_device *tidss) +{ + for (int i = 0; i < tidss->num_oldis; i++) { + if (tidss->oldis[i]) { + drm_bridge_remove(&tidss->oldis[i]->bridge); + tidss->oldis[i] = NULL; + } + } +} + +int tidss_oldi_init(struct tidss_device *tidss) +{ + struct tidss_oldi *oldi; + struct device_node *child; + struct drm_bridge *bridge; + u32 parent_vp, oldi_instance; + int companion_instance = -1; + enum tidss_oldi_link_type link_type = OLDI_MODE_UNSUPPORTED; + struct device_node *oldi_parent; + int ret = 0; + + tidss->num_oldis = 0; + + oldi_parent = of_get_child_by_name(tidss->dev->of_node, "oldi-transmitters"); + if (!oldi_parent) + /* Return gracefully */ + return 0; + + for_each_available_child_of_node(oldi_parent, child) { + ret = get_parent_dss_vp(child, &parent_vp); + if (ret) { + if (ret == -ENODEV) { + /* + * ENODEV means that this particular OLDI node + * is not connected with the DSS, which is not + * a harmful case. There could be another OLDI + * which may still be connected. + * Continue to search for that. + */ + ret = 0; + continue; + } + goto err_put_node; + } + + ret = of_property_read_u32(child, "reg", &oldi_instance); + if (ret) + goto err_put_node; + + /* + * Now that it's confirmed that OLDI is connected with DSS, + * let's continue getting the OLDI sinks ahead and other OLDI + * properties. + */ + bridge = devm_drm_of_get_bridge(tidss->dev, child, + OLDI_OUTPUT_PORT, 0); + if (IS_ERR(bridge)) { + /* + * Either there was no OLDI sink in the devicetree, or + * the OLDI sink has not been added yet. In any case, + * return. + * We don't want to have an OLDI node connected to DSS + * but not to any sink. + */ + ret = dev_err_probe(tidss->dev, PTR_ERR(bridge), + "no panel/bridge for OLDI%u.\n", + oldi_instance); + goto err_put_node; + } + + link_type = get_oldi_mode(child, &companion_instance); + if (link_type == OLDI_MODE_UNSUPPORTED) { + ret = dev_err_probe(tidss->dev, -EINVAL, + "OLDI%u: Unsupported OLDI connection.\n", + oldi_instance); + goto err_put_node; + } else if ((link_type == OLDI_MODE_SECONDARY_CLONE_SINGLE_LINK) || + (link_type == OLDI_MODE_CLONE_SINGLE_LINK)) { + /* + * The OLDI driver cannot support OLDI clone mode + * properly at present. + * The clone mode requires 2 working encoder-bridge + * pipelines, generating from the same crtc. The DRM + * framework does not support this at present. If + * there were to be, say, 2 OLDI sink bridges each + * connected to an OLDI TXes, they couldn't both be + * supported simultaneously. + * This driver still has some code pertaining to OLDI + * clone mode configuration in DSS hardware for future, + * when there is a better infrastructure in the DRM + * framework to support 2 encoder-bridge pipelines + * simultaneously. + * Till that time, this driver shall error out if it + * detects a clone mode configuration. + */ + ret = dev_err_probe(tidss->dev, -EOPNOTSUPP, + "The OLDI driver does not support Clone Mode at present.\n"); + goto err_put_node; + } else if (link_type == OLDI_MODE_SECONDARY_DUAL_LINK) { + /* + * This is the secondary OLDI node, which serves as a + * companion to the primary OLDI, when it is configured + * for the dual-link mode. Since the primary OLDI will + * be a part of bridge chain, no need to put this one + * too. Continue onto the next OLDI node. + */ + continue; + } + + oldi = devm_kzalloc(tidss->dev, sizeof(*oldi), GFP_KERNEL); + if (!oldi) { + ret = -ENOMEM; + goto err_put_node; + } + + oldi->parent_vp = parent_vp; + oldi->oldi_instance = oldi_instance; + oldi->companion_instance = companion_instance; + oldi->link_type = link_type; + oldi->dev = tidss->dev; + oldi->next_bridge = bridge; + + /* + * Only the primary OLDI needs to reference the io-ctrl system + * registers, and the serial clock. + * We don't require a check for secondary OLDI in dual-link mode + * because the driver will not create a drm_bridge instance. + * But the driver will need to create a drm_bridge instance, + * for secondary OLDI in clone mode (once it is supported). + */ + if (link_type != OLDI_MODE_SECONDARY_CLONE_SINGLE_LINK) { + oldi->io_ctrl = syscon_regmap_lookup_by_phandle(child, + "ti,oldi-io-ctrl"); + if (IS_ERR(oldi->io_ctrl)) { + ret = dev_err_probe(oldi->dev, PTR_ERR(oldi->io_ctrl), + "OLDI%u: syscon_regmap_lookup_by_phandle failed.\n", + oldi_instance); + goto err_put_node; + } + + oldi->serial = of_clk_get_by_name(child, "serial"); + if (IS_ERR(oldi->serial)) { + ret = dev_err_probe(oldi->dev, PTR_ERR(oldi->serial), + "OLDI%u: Failed to get serial clock.\n", + oldi_instance); + goto err_put_node; + } + } + + /* Register the bridge. */ + oldi->bridge.of_node = child; + oldi->bridge.driver_private = oldi; + oldi->bridge.funcs = &tidss_oldi_bridge_funcs; + oldi->bridge.timings = &default_tidss_oldi_timings; + + tidss->oldis[tidss->num_oldis++] = oldi; + oldi->tidss = tidss; + + drm_bridge_add(&oldi->bridge); + } + + of_node_put(child); + of_node_put(oldi_parent); + + return 0; + +err_put_node: + of_node_put(child); + of_node_put(oldi_parent); + return ret; +} diff --git a/drivers/gpu/drm/tidss/tidss_oldi.h b/drivers/gpu/drm/tidss/tidss_oldi.h new file mode 100644 index 000000000000..8cd535c5ee65 --- /dev/null +++ b/drivers/gpu/drm/tidss/tidss_oldi.h @@ -0,0 +1,43 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright (C) 2025 - Texas Instruments Incorporated + * + * Aradhya Bhatia <a-bhatia1@ti.com> + */ + +#ifndef __TIDSS_OLDI_H__ +#define __TIDSS_OLDI_H__ + +#include "tidss_drv.h" + +struct tidss_oldi; + +/* OLDI PORTS */ +#define OLDI_INPUT_PORT 0 +#define OLDI_OUTPUT_PORT 1 + +/* Control MMR Registers */ + +/* Register offsets */ +#define OLDI_PD_CTRL 0x100 +#define OLDI_LB_CTRL 0x104 + +/* Power control bits */ +#define OLDI_PWRDOWN_TX(n) BIT(n) + +/* LVDS Bandgap reference Enable/Disable */ +#define OLDI_PWRDN_BG BIT(8) + +enum tidss_oldi_link_type { + OLDI_MODE_UNSUPPORTED, + OLDI_MODE_SINGLE_LINK, + OLDI_MODE_CLONE_SINGLE_LINK, + OLDI_MODE_SECONDARY_CLONE_SINGLE_LINK, + OLDI_MODE_DUAL_LINK, + OLDI_MODE_SECONDARY_DUAL_LINK, +}; + +int tidss_oldi_init(struct tidss_device *tidss); +void tidss_oldi_deinit(struct tidss_device *tidss); + +#endif /* __TIDSS_OLDI_H__ */ diff --git a/drivers/gpu/drm/tiny/bochs.c b/drivers/gpu/drm/tiny/bochs.c index 8706763af8fb..8d3b7c4fa6a4 100644 --- a/drivers/gpu/drm/tiny/bochs.c +++ b/drivers/gpu/drm/tiny/bochs.c @@ -19,6 +19,7 @@ #include <drm/drm_gem_shmem_helper.h> #include <drm/drm_managed.h> #include <drm/drm_module.h> +#include <drm/drm_panic.h> #include <drm/drm_plane_helper.h> #include <drm/drm_probe_helper.h> @@ -469,10 +470,28 @@ static void bochs_primary_plane_helper_atomic_update(struct drm_plane *plane, bochs_hw_setformat(bochs, fb->format); } +static int bochs_primary_plane_helper_get_scanout_buffer(struct drm_plane *plane, + struct drm_scanout_buffer *sb) +{ + struct bochs_device *bochs = to_bochs_device(plane->dev); + struct iosys_map map = IOSYS_MAP_INIT_VADDR_IOMEM(bochs->fb_map); + + if (plane->state && plane->state->fb) { + sb->format = plane->state->fb->format; + sb->width = plane->state->fb->width; + sb->height = plane->state->fb->height; + sb->pitch[0] = plane->state->fb->pitches[0]; + sb->map[0] = map; + return 0; + } + return -ENODEV; +} + static const struct drm_plane_helper_funcs bochs_primary_plane_helper_funcs = { DRM_GEM_SHADOW_PLANE_HELPER_FUNCS, .atomic_check = bochs_primary_plane_helper_atomic_check, .atomic_update = bochs_primary_plane_helper_atomic_update, + .get_scanout_buffer = bochs_primary_plane_helper_get_scanout_buffer, }; static const struct drm_plane_funcs bochs_primary_plane_funcs = { diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index bb9c5c8e16b5..f4d9e68b21e7 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -526,11 +526,11 @@ static s64 ttm_bo_evict_cb(struct ttm_lru_walk *walk, struct ttm_buffer_object * return 0; if (bo->deleted) { - lret = ttm_bo_wait_ctx(bo, walk->ctx); + lret = ttm_bo_wait_ctx(bo, walk->arg.ctx); if (!lret) ttm_bo_cleanup_memtype_use(bo); } else { - lret = ttm_bo_evict(bo, walk->ctx); + lret = ttm_bo_evict(bo, walk->arg.ctx); } if (lret) @@ -566,8 +566,10 @@ static int ttm_bo_evict_alloc(struct ttm_device *bdev, struct ttm_bo_evict_walk evict_walk = { .walk = { .ops = &ttm_evict_walk_ops, - .ctx = ctx, - .ticket = ticket, + .arg = { + .ctx = ctx, + .ticket = ticket, + } }, .place = place, .evictor = evictor, @@ -576,7 +578,7 @@ static int ttm_bo_evict_alloc(struct ttm_device *bdev, }; s64 lret; - evict_walk.walk.trylock_only = true; + evict_walk.walk.arg.trylock_only = true; lret = ttm_lru_walk_for_evict(&evict_walk.walk, bdev, man, 1); /* One more attempt if we hit low limit? */ @@ -590,12 +592,12 @@ static int ttm_bo_evict_alloc(struct ttm_device *bdev, /* Reset low limit */ evict_walk.try_low = evict_walk.hit_low = false; /* If ticket-locking, repeat while making progress. */ - evict_walk.walk.trylock_only = false; + evict_walk.walk.arg.trylock_only = false; retry: do { /* The walk may clear the evict_walk.walk.ticket field */ - evict_walk.walk.ticket = ticket; + evict_walk.walk.arg.ticket = ticket; evict_walk.evicted = 0; lret = ttm_lru_walk_for_evict(&evict_walk.walk, bdev, man, 1); } while (!lret && evict_walk.evicted); @@ -1106,7 +1108,7 @@ ttm_bo_swapout_cb(struct ttm_lru_walk *walk, struct ttm_buffer_object *bo) struct ttm_place place = {.mem_type = bo->resource->mem_type}; struct ttm_bo_swapout_walk *swapout_walk = container_of(walk, typeof(*swapout_walk), walk); - struct ttm_operation_ctx *ctx = walk->ctx; + struct ttm_operation_ctx *ctx = walk->arg.ctx; s64 ret; /* @@ -1217,8 +1219,10 @@ s64 ttm_bo_swapout(struct ttm_device *bdev, struct ttm_operation_ctx *ctx, struct ttm_bo_swapout_walk swapout_walk = { .walk = { .ops = &ttm_swap_ops, - .ctx = ctx, - .trylock_only = true, + .arg = { + .ctx = ctx, + .trylock_only = true, + }, }, .gfp_flags = gfp_flags, }; diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 16cb4065214d..6502ced6169d 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -799,14 +799,15 @@ error_destroy_tt: return ret; } -static bool ttm_lru_walk_trylock(struct ttm_operation_ctx *ctx, - struct ttm_buffer_object *bo, - bool *needs_unlock) +static bool ttm_lru_walk_trylock(struct ttm_bo_lru_cursor *curs, + struct ttm_buffer_object *bo) { - *needs_unlock = false; + struct ttm_operation_ctx *ctx = curs->arg->ctx; + + curs->needs_unlock = false; if (dma_resv_trylock(bo->base.resv)) { - *needs_unlock = true; + curs->needs_unlock = true; return true; } @@ -818,27 +819,27 @@ static bool ttm_lru_walk_trylock(struct ttm_operation_ctx *ctx, return false; } -static int ttm_lru_walk_ticketlock(struct ttm_lru_walk *walk, - struct ttm_buffer_object *bo, - bool *needs_unlock) +static int ttm_lru_walk_ticketlock(struct ttm_bo_lru_cursor *curs, + struct ttm_buffer_object *bo) { + struct ttm_lru_walk_arg *arg = curs->arg; struct dma_resv *resv = bo->base.resv; int ret; - if (walk->ctx->interruptible) - ret = dma_resv_lock_interruptible(resv, walk->ticket); + if (arg->ctx->interruptible) + ret = dma_resv_lock_interruptible(resv, arg->ticket); else - ret = dma_resv_lock(resv, walk->ticket); + ret = dma_resv_lock(resv, arg->ticket); if (!ret) { - *needs_unlock = true; + curs->needs_unlock = true; /* * Only a single ticketlock per loop. Ticketlocks are prone * to return -EDEADLK causing the eviction to fail, so * after waiting for the ticketlock, revert back to * trylocking for this walk. */ - walk->ticket = NULL; + arg->ticket = NULL; } else if (ret == -EDEADLK) { /* Caller needs to exit the ww transaction. */ ret = -ENOSPC; @@ -847,12 +848,6 @@ static int ttm_lru_walk_ticketlock(struct ttm_lru_walk *walk, return ret; } -static void ttm_lru_walk_unlock(struct ttm_buffer_object *bo, bool locked) -{ - if (locked) - dma_resv_unlock(bo->base.resv); -} - /** * ttm_lru_walk_for_evict() - Perform a LRU list walk, with actions taken on * valid items. @@ -887,64 +882,21 @@ static void ttm_lru_walk_unlock(struct ttm_buffer_object *bo, bool locked) s64 ttm_lru_walk_for_evict(struct ttm_lru_walk *walk, struct ttm_device *bdev, struct ttm_resource_manager *man, s64 target) { - struct ttm_resource_cursor cursor; - struct ttm_resource *res; + struct ttm_bo_lru_cursor cursor; + struct ttm_buffer_object *bo; s64 progress = 0; s64 lret; - spin_lock(&bdev->lru_lock); - ttm_resource_cursor_init(&cursor, man); - ttm_resource_manager_for_each_res(&cursor, res) { - struct ttm_buffer_object *bo = res->bo; - bool bo_needs_unlock = false; - bool bo_locked = false; - int mem_type; - - /* - * Attempt a trylock before taking a reference on the bo, - * since if we do it the other way around, and the trylock fails, - * we need to drop the lru lock to put the bo. - */ - if (ttm_lru_walk_trylock(walk->ctx, bo, &bo_needs_unlock)) - bo_locked = true; - else if (!walk->ticket || walk->ctx->no_wait_gpu || - walk->trylock_only) - continue; - - if (!ttm_bo_get_unless_zero(bo)) { - ttm_lru_walk_unlock(bo, bo_needs_unlock); - continue; - } - - mem_type = res->mem_type; - spin_unlock(&bdev->lru_lock); - - lret = 0; - if (!bo_locked) - lret = ttm_lru_walk_ticketlock(walk, bo, &bo_needs_unlock); - - /* - * Note that in between the release of the lru lock and the - * ticketlock, the bo may have switched resource, - * and also memory type, since the resource may have been - * freed and allocated again with a different memory type. - * In that case, just skip it. - */ - if (!lret && bo->resource && bo->resource->mem_type == mem_type) - lret = walk->ops->process_bo(walk, bo); - - ttm_lru_walk_unlock(bo, bo_needs_unlock); - ttm_bo_put(bo); + ttm_bo_lru_for_each_reserved_guarded(&cursor, man, &walk->arg, bo) { + lret = walk->ops->process_bo(walk, bo); if (lret == -EBUSY || lret == -EALREADY) lret = 0; progress = (lret < 0) ? lret : progress + lret; - - spin_lock(&bdev->lru_lock); if (progress < 0 || progress >= target) break; } - ttm_resource_cursor_fini(&cursor); - spin_unlock(&bdev->lru_lock); + if (IS_ERR(bo)) + return PTR_ERR(bo); return progress; } @@ -982,44 +934,87 @@ EXPORT_SYMBOL(ttm_bo_lru_cursor_fini); * ttm_bo_lru_cursor_init() - Initialize a struct ttm_bo_lru_cursor * @curs: The ttm_bo_lru_cursor to initialize. * @man: The ttm resource_manager whose LRU lists to iterate over. - * @ctx: The ttm_operation_ctx to govern the locking. + * @arg: The ttm_lru_walk_arg to govern the walk. * - * Initialize a struct ttm_bo_lru_cursor. Currently only trylocking - * or prelocked buffer objects are available as detailed by - * @ctx::resv and @ctx::allow_res_evict. Ticketlocking is not - * supported. + * Initialize a struct ttm_bo_lru_cursor. * * Return: Pointer to @curs. The function does not fail. */ struct ttm_bo_lru_cursor * ttm_bo_lru_cursor_init(struct ttm_bo_lru_cursor *curs, struct ttm_resource_manager *man, - struct ttm_operation_ctx *ctx) + struct ttm_lru_walk_arg *arg) { memset(curs, 0, sizeof(*curs)); ttm_resource_cursor_init(&curs->res_curs, man); - curs->ctx = ctx; + curs->arg = arg; return curs; } EXPORT_SYMBOL(ttm_bo_lru_cursor_init); static struct ttm_buffer_object * -ttm_bo_from_res_reserved(struct ttm_resource *res, struct ttm_bo_lru_cursor *curs) +__ttm_bo_lru_cursor_next(struct ttm_bo_lru_cursor *curs) { - struct ttm_buffer_object *bo = res->bo; + spinlock_t *lru_lock = &curs->res_curs.man->bdev->lru_lock; + struct ttm_resource *res = NULL; + struct ttm_buffer_object *bo; + struct ttm_lru_walk_arg *arg = curs->arg; + bool first = !curs->bo; - if (!ttm_lru_walk_trylock(curs->ctx, bo, &curs->needs_unlock)) - return NULL; + ttm_bo_lru_cursor_cleanup_bo(curs); - if (!ttm_bo_get_unless_zero(bo)) { - if (curs->needs_unlock) - dma_resv_unlock(bo->base.resv); - return NULL; + spin_lock(lru_lock); + for (;;) { + int mem_type, ret = 0; + bool bo_locked = false; + + if (first) { + res = ttm_resource_manager_first(&curs->res_curs); + first = false; + } else { + res = ttm_resource_manager_next(&curs->res_curs); + } + if (!res) + break; + + bo = res->bo; + if (ttm_lru_walk_trylock(curs, bo)) + bo_locked = true; + else if (!arg->ticket || arg->ctx->no_wait_gpu || arg->trylock_only) + continue; + + if (!ttm_bo_get_unless_zero(bo)) { + if (curs->needs_unlock) + dma_resv_unlock(bo->base.resv); + continue; + } + + mem_type = res->mem_type; + spin_unlock(lru_lock); + if (!bo_locked) + ret = ttm_lru_walk_ticketlock(curs, bo); + + /* + * Note that in between the release of the lru lock and the + * ticketlock, the bo may have switched resource, + * and also memory type, since the resource may have been + * freed and allocated again with a different memory type. + * In that case, just skip it. + */ + curs->bo = bo; + if (!ret && bo->resource && bo->resource->mem_type == mem_type) + return bo; + + ttm_bo_lru_cursor_cleanup_bo(curs); + if (ret && ret != -EALREADY) + return ERR_PTR(ret); + + spin_lock(lru_lock); } - curs->bo = bo; - return bo; + spin_unlock(lru_lock); + return res ? bo : NULL; } /** @@ -1033,25 +1028,7 @@ ttm_bo_from_res_reserved(struct ttm_resource *res, struct ttm_bo_lru_cursor *cur */ struct ttm_buffer_object *ttm_bo_lru_cursor_next(struct ttm_bo_lru_cursor *curs) { - spinlock_t *lru_lock = &curs->res_curs.man->bdev->lru_lock; - struct ttm_resource *res = NULL; - struct ttm_buffer_object *bo; - - ttm_bo_lru_cursor_cleanup_bo(curs); - - spin_lock(lru_lock); - for (;;) { - res = ttm_resource_manager_next(&curs->res_curs); - if (!res) - break; - - bo = ttm_bo_from_res_reserved(res, curs); - if (bo) - break; - } - - spin_unlock(lru_lock); - return res ? bo : NULL; + return __ttm_bo_lru_cursor_next(curs); } EXPORT_SYMBOL(ttm_bo_lru_cursor_next); @@ -1065,21 +1042,8 @@ EXPORT_SYMBOL(ttm_bo_lru_cursor_next); */ struct ttm_buffer_object *ttm_bo_lru_cursor_first(struct ttm_bo_lru_cursor *curs) { - spinlock_t *lru_lock = &curs->res_curs.man->bdev->lru_lock; - struct ttm_buffer_object *bo; - struct ttm_resource *res; - - spin_lock(lru_lock); - res = ttm_resource_manager_first(&curs->res_curs); - if (!res) { - spin_unlock(lru_lock); - return NULL; - } - - bo = ttm_bo_from_res_reserved(res, curs); - spin_unlock(lru_lock); - - return bo ? bo : ttm_bo_lru_cursor_next(curs); + ttm_bo_lru_cursor_cleanup_bo(curs); + return __ttm_bo_lru_cursor_next(curs); } EXPORT_SYMBOL(ttm_bo_lru_cursor_first); diff --git a/drivers/gpu/drm/udl/udl_drv.c b/drivers/gpu/drm/udl/udl_drv.c index ce5ae7cacb90..1922988625eb 100644 --- a/drivers/gpu/drm/udl/udl_drv.c +++ b/drivers/gpu/drm/udl/udl_drv.c @@ -57,7 +57,7 @@ static const struct drm_driver driver = { /* GEM hooks */ .fops = &udl_driver_fops, - DRM_GEM_SHMEM_DRIVER_OPS_NO_MAP_SGT, + DRM_GEM_SHMEM_DRIVER_OPS, DRM_FBDEV_SHMEM_DRIVER_OPS, .name = DRIVER_NAME, diff --git a/drivers/gpu/drm/vgem/vgem_drv.c b/drivers/gpu/drm/vgem/vgem_drv.c index 2752ab4f1c97..260c64733972 100644 --- a/drivers/gpu/drm/vgem/vgem_drv.c +++ b/drivers/gpu/drm/vgem/vgem_drv.c @@ -32,7 +32,7 @@ #include <linux/dma-buf.h> #include <linux/module.h> -#include <linux/platform_device.h> +#include <linux/device/faux.h> #include <linux/shmem_fs.h> #include <linux/vmalloc.h> @@ -52,7 +52,7 @@ static struct vgem_device { struct drm_device drm; - struct platform_device *platform; + struct faux_device *faux_dev; } *vgem_device; static int vgem_open(struct drm_device *dev, struct drm_file *file) @@ -127,27 +127,27 @@ static const struct drm_driver vgem_driver = { static int __init vgem_init(void) { int ret; - struct platform_device *pdev; + struct faux_device *fdev; - pdev = platform_device_register_simple("vgem", -1, NULL, 0); - if (IS_ERR(pdev)) - return PTR_ERR(pdev); + fdev = faux_device_create("vgem", NULL, NULL); + if (!fdev) + return -ENODEV; - if (!devres_open_group(&pdev->dev, NULL, GFP_KERNEL)) { + if (!devres_open_group(&fdev->dev, NULL, GFP_KERNEL)) { ret = -ENOMEM; goto out_unregister; } - dma_coerce_mask_and_coherent(&pdev->dev, + dma_coerce_mask_and_coherent(&fdev->dev, DMA_BIT_MASK(64)); - vgem_device = devm_drm_dev_alloc(&pdev->dev, &vgem_driver, + vgem_device = devm_drm_dev_alloc(&fdev->dev, &vgem_driver, struct vgem_device, drm); if (IS_ERR(vgem_device)) { ret = PTR_ERR(vgem_device); goto out_devres; } - vgem_device->platform = pdev; + vgem_device->faux_dev = fdev; /* Final step: expose the device/driver to userspace */ ret = drm_dev_register(&vgem_device->drm, 0); @@ -157,19 +157,19 @@ static int __init vgem_init(void) return 0; out_devres: - devres_release_group(&pdev->dev, NULL); + devres_release_group(&fdev->dev, NULL); out_unregister: - platform_device_unregister(pdev); + faux_device_destroy(fdev); return ret; } static void __exit vgem_exit(void) { - struct platform_device *pdev = vgem_device->platform; + struct faux_device *fdev = vgem_device->faux_dev; drm_dev_unregister(&vgem_device->drm); - devres_release_group(&pdev->dev, NULL); - platform_device_unregister(pdev); + devres_release_group(&fdev->dev, NULL); + faux_device_destroy(fdev); } module_init(vgem_init); diff --git a/drivers/gpu/drm/vkms/vkms_crtc.c b/drivers/gpu/drm/vkms/vkms_crtc.c index 8c9898b9055d..e60573e0f3e9 100644 --- a/drivers/gpu/drm/vkms/vkms_crtc.c +++ b/drivers/gpu/drm/vkms/vkms_crtc.c @@ -302,8 +302,6 @@ struct vkms_output *vkms_crtc_init(struct drm_device *dev, struct drm_plane *pri vkms_out->composer_workq = drmm_alloc_ordered_workqueue(dev, "vkms_composer", 0); if (IS_ERR(vkms_out->composer_workq)) return ERR_CAST(vkms_out->composer_workq); - if (!vkms_out->composer_workq) - return ERR_PTR(-ENOMEM); return vkms_out; } diff --git a/drivers/gpu/drm/vkms/vkms_drv.c b/drivers/gpu/drm/vkms/vkms_drv.c index a24d1655f7b8..e8472d9b6e3b 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.c +++ b/drivers/gpu/drm/vkms/vkms_drv.c @@ -10,7 +10,7 @@ */ #include <linux/module.h> -#include <linux/platform_device.h> +#include <linux/device/faux.h> #include <linux/dma-mapping.h> #include <drm/clients/drm_client_setup.h> @@ -149,27 +149,27 @@ static int vkms_modeset_init(struct vkms_device *vkmsdev) static int vkms_create(struct vkms_config *config) { int ret; - struct platform_device *pdev; + struct faux_device *fdev; struct vkms_device *vkms_device; const char *dev_name; dev_name = vkms_config_get_device_name(config); - pdev = platform_device_register_simple(dev_name, -1, NULL, 0); - if (IS_ERR(pdev)) - return PTR_ERR(pdev); + fdev = faux_device_create(dev_name, NULL, NULL); + if (!fdev) + return -ENODEV; - if (!devres_open_group(&pdev->dev, NULL, GFP_KERNEL)) { + if (!devres_open_group(&fdev->dev, NULL, GFP_KERNEL)) { ret = -ENOMEM; goto out_unregister; } - vkms_device = devm_drm_dev_alloc(&pdev->dev, &vkms_driver, + vkms_device = devm_drm_dev_alloc(&fdev->dev, &vkms_driver, struct vkms_device, drm); if (IS_ERR(vkms_device)) { ret = PTR_ERR(vkms_device); goto out_devres; } - vkms_device->platform = pdev; + vkms_device->faux_dev = fdev; vkms_device->config = config; config->dev = vkms_device; @@ -203,9 +203,9 @@ static int vkms_create(struct vkms_config *config) return 0; out_devres: - devres_release_group(&pdev->dev, NULL); + devres_release_group(&fdev->dev, NULL); out_unregister: - platform_device_unregister(pdev); + faux_device_destroy(fdev); return ret; } @@ -231,19 +231,19 @@ static int __init vkms_init(void) static void vkms_destroy(struct vkms_config *config) { - struct platform_device *pdev; + struct faux_device *fdev; if (!config->dev) { DRM_INFO("vkms_device is NULL.\n"); return; } - pdev = config->dev->platform; + fdev = config->dev->faux_dev; drm_dev_unregister(&config->dev->drm); drm_atomic_helper_shutdown(&config->dev->drm); - devres_release_group(&pdev->dev, NULL); - platform_device_unregister(pdev); + devres_release_group(&fdev->dev, NULL); + faux_device_destroy(fdev); config->dev = NULL; } diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h index 68b5e3fce455..8013c31efe3b 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.h +++ b/drivers/gpu/drm/vkms/vkms_drv.h @@ -232,13 +232,13 @@ struct vkms_config; * struct vkms_device - Description of a VKMS device * * @drm - Base device in DRM - * @platform - Associated platform device + * @faux_dev - Associated faux device * @output - Configuration and sub-components of the VKMS device * @config: Configuration used in this VKMS device */ struct vkms_device { struct drm_device drm; - struct platform_device *platform; + struct faux_device *faux_dev; const struct vkms_config *config; }; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c b/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c index c55382167c1b..7057d852951b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_gem.c @@ -284,11 +284,10 @@ static void vmw_bo_print_info(int id, struct vmw_bo *bo, struct seq_file *m) seq_printf(m, "\t\t0x%08x: %12zu bytes %s, type = %s", id, bo->tbo.base.size, placement, type); - seq_printf(m, ", priority = %u, pin_count = %u, GEM refs = %d, TTM refs = %d", + seq_printf(m, ", priority = %u, pin_count = %u, GEM refs = %d", bo->tbo.priority, bo->tbo.pin_count, - kref_read(&bo->tbo.base.refcount), - kref_read(&bo->tbo.kref)); + kref_read(&bo->tbo.base.refcount)); seq_puts(m, "\n"); } diff --git a/drivers/gpu/drm/xe/Kconfig b/drivers/gpu/drm/xe/Kconfig index 30ed74ad29ab..2bb2bc052120 100644 --- a/drivers/gpu/drm/xe/Kconfig +++ b/drivers/gpu/drm/xe/Kconfig @@ -1,7 +1,8 @@ # SPDX-License-Identifier: GPL-2.0-only config DRM_XE tristate "Intel Xe2 Graphics" - depends on DRM && PCI && (m || (y && KUNIT=y)) + depends on DRM && PCI + depends on KUNIT || !KUNIT depends on INTEL_VSEC || !INTEL_VSEC depends on X86_PLATFORM_DEVICES || !(X86 && ACPI) select INTERVAL_TREE @@ -44,6 +45,7 @@ config DRM_XE select WANT_DEV_COREDUMP select AUXILIARY_BUS select HMM_MIRROR + select REGMAP if I2C help Driver for Intel Xe2 series GPUs and later. Experimental support for Xe series is also available. @@ -85,16 +87,18 @@ config DRM_XE_GPUSVM Enable this option if you want support for CPU to GPU address mirroring. - If in doubut say "Y". + If in doubt say "Y". -config DRM_XE_DEVMEM_MIRROR - bool "Enable device memory mirror" +config DRM_XE_PAGEMAP + bool "Enable device memory pool for SVM" depends on DRM_XE_GPUSVM select GET_FREE_REGION default y help - Disable this option only if you want to compile out without device - memory mirror. Will reduce KMD memory footprint when disabled. + Disable this option only if you don't want to expose local device + memory for SVM. Will reduce KMD memory footprint when disabled. + + If in doubt say "Y". config DRM_XE_FORCE_PROBE string "Force probe xe for selected Intel hardware IDs" diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile index c83d72de81dd..1201b8ab4702 100644 --- a/drivers/gpu/drm/xe/Makefile +++ b/drivers/gpu/drm/xe/Makefile @@ -80,6 +80,7 @@ xe-y += xe_bb.o \ xe_mmio.o \ xe_mocs.o \ xe_module.o \ + xe_nvm.o \ xe_oa.o \ xe_observation.o \ xe_pat.o \ @@ -124,6 +125,7 @@ xe-y += xe_bb.o \ xe_wait_user_fence.o \ xe_wopcm.o +xe-$(CONFIG_I2C) += xe_i2c.o xe-$(CONFIG_HMM_MIRROR) += xe_hmm.o xe-$(CONFIG_DRM_XE_GPUSVM) += xe_svm.o diff --git a/drivers/gpu/drm/xe/abi/guc_actions_abi.h b/drivers/gpu/drm/xe/abi/guc_actions_abi.h index ff4f412c28d8..81eb046aeebf 100644 --- a/drivers/gpu/drm/xe/abi/guc_actions_abi.h +++ b/drivers/gpu/drm/xe/abi/guc_actions_abi.h @@ -142,6 +142,7 @@ enum xe_guc_action { XE_GUC_ACTION_SET_ENG_UTIL_BUFF = 0x550A, XE_GUC_ACTION_SET_DEVICE_ENGINE_ACTIVITY_BUFFER = 0x550C, XE_GUC_ACTION_SET_FUNCTION_ENGINE_ACTIVITY_BUFFER = 0x550D, + XE_GUC_ACTION_OPT_IN_FEATURE_KLV = 0x550E, XE_GUC_ACTION_NOTIFY_MEMORY_CAT_ERROR = 0x6000, XE_GUC_ACTION_REPORT_PAGE_FAULT_REQ_DESC = 0x6002, XE_GUC_ACTION_PAGE_FAULT_RES_DESC = 0x6003, @@ -271,4 +272,7 @@ enum xe_guc_g2g_type { #define XE_G2G_DEREGISTER_TILE REG_GENMASK(15, 12) #define XE_G2G_DEREGISTER_TYPE REG_GENMASK(11, 8) +/* invalid type for XE_GUC_ACTION_NOTIFY_MEMORY_CAT_ERROR */ +#define XE_GUC_CAT_ERR_TYPE_INVALID 0xdeadbeef + #endif diff --git a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h index 7de8f827281f..0366a9da5977 100644 --- a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h +++ b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h @@ -16,6 +16,7 @@ * +===+=======+==============================================================+ * | 0 | 31:16 | **KEY** - KLV key identifier | * | | | - `GuC Self Config KLVs`_ | + * | | | - `GuC Opt In Feature KLVs`_ | * | | | - `GuC VGT Policy KLVs`_ | * | | | - `GuC VF Configuration KLVs`_ | * | | | | @@ -125,6 +126,33 @@ enum { }; /** + * DOC: GuC Opt In Feature KLVs + * + * `GuC KLV`_ keys available for use with OPT_IN_FEATURE_KLV + * + * _`GUC_KLV_OPT_IN_FEATURE_EXT_CAT_ERR_TYPE` : 0x4001 + * Adds an extra dword to the XE_GUC_ACTION_NOTIFY_MEMORY_CAT_ERROR G2H + * containing the type of the CAT error. On HW that does not support + * reporting the CAT error type, the extra dword is set to 0xdeadbeef. + * + * _`GUC_KLV_OPT_IN_FEATURE_DYNAMIC_INHIBIT_CONTEXT_SWITCH` : 0x4003 + * This KLV enables the Dynamic Inhibit Context Switch optimization, which + * consists in the GuC setting the CTX_CTRL_INHIBIT_SYN_CTX_SWITCH bit to + * zero in the CTX_CONTEXT_CONTROL register of LRCs that are submitted + * to an oversubscribed engine. This will cause those contexts to be + * switched out immediately if they hit an unsatisfied semaphore wait + * (instead of waiting the full timeslice duration). The bit is instead set + * to one if a single context is queued on the engine, to avoid it being + * switched out if there isn't another context that can run in its place. + */ + +#define GUC_KLV_OPT_IN_FEATURE_EXT_CAT_ERR_TYPE_KEY 0x4001 +#define GUC_KLV_OPT_IN_FEATURE_EXT_CAT_ERR_TYPE_LEN 0u + +#define GUC_KLV_OPT_IN_FEATURE_DYNAMIC_INHIBIT_CONTEXT_SWITCH_KEY 0x4003 +#define GUC_KLV_OPT_IN_FEATURE_DYNAMIC_INHIBIT_CONTEXT_SWITCH_LEN 0u + +/** * DOC: GuC VGT Policy KLVs * * `GuC KLV`_ keys available for use with PF2GUC_UPDATE_VGT_POLICY. diff --git a/drivers/gpu/drm/xe/display/xe_dsb_buffer.c b/drivers/gpu/drm/xe/display/xe_dsb_buffer.c index f95375451e2f..9f941fc2e36b 100644 --- a/drivers/gpu/drm/xe/display/xe_dsb_buffer.c +++ b/drivers/gpu/drm/xe/display/xe_dsb_buffer.c @@ -17,10 +17,7 @@ u32 intel_dsb_buffer_ggtt_offset(struct intel_dsb_buffer *dsb_buf) void intel_dsb_buffer_write(struct intel_dsb_buffer *dsb_buf, u32 idx, u32 val) { - struct xe_device *xe = dsb_buf->vma->bo->tile->xe; - iosys_map_wr(&dsb_buf->vma->bo->vmap, idx * 4, u32, val); - xe_device_l2_flush(xe); } u32 intel_dsb_buffer_read(struct intel_dsb_buffer *dsb_buf, u32 idx) @@ -30,12 +27,9 @@ u32 intel_dsb_buffer_read(struct intel_dsb_buffer *dsb_buf, u32 idx) void intel_dsb_buffer_memset(struct intel_dsb_buffer *dsb_buf, u32 idx, u32 val, size_t size) { - struct xe_device *xe = dsb_buf->vma->bo->tile->xe; - WARN_ON(idx > (dsb_buf->buf_size - size) / sizeof(*dsb_buf->cmd_buf)); iosys_map_memset(&dsb_buf->vma->bo->vmap, idx * 4, val, size); - xe_device_l2_flush(xe); } bool intel_dsb_buffer_create(struct intel_crtc *crtc, struct intel_dsb_buffer *dsb_buf, size_t size) @@ -74,9 +68,12 @@ void intel_dsb_buffer_cleanup(struct intel_dsb_buffer *dsb_buf) void intel_dsb_buffer_flush_map(struct intel_dsb_buffer *dsb_buf) { + struct xe_device *xe = dsb_buf->vma->bo->tile->xe; + /* * The memory barrier here is to ensure coherency of DSB vs MMIO, * both for weak ordering archs and discrete cards. */ - xe_device_wmb(dsb_buf->vma->bo->tile->xe); + xe_device_wmb(xe); + xe_device_l2_flush(xe); } diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c index 45122822b051..c38fba18effe 100644 --- a/drivers/gpu/drm/xe/display/xe_fb_pin.c +++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c @@ -163,6 +163,9 @@ static int __xe_pin_fb_vma_dpt(const struct intel_framebuffer *fb, vma->dpt = dpt; vma->node = dpt->ggtt_node[tile0->id]; + + /* Ensure DPT writes are flushed */ + xe_device_l2_flush(xe); return 0; } @@ -224,7 +227,7 @@ static int __xe_pin_fb_vma_ggtt(const struct intel_framebuffer *fb, goto out_unlock; } - ret = xe_ggtt_node_insert_locked(vma->node, bo->size, align, 0); + ret = xe_ggtt_node_insert_locked(vma->node, xe_bo_size(bo), align, 0); if (ret) { xe_ggtt_node_fini(vma->node); goto out_unlock; @@ -326,8 +329,6 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb, if (ret) goto err_unpin; - /* Ensure DPT writes are flushed */ - xe_device_l2_flush(xe); return vma; err_unpin: diff --git a/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c b/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c index b35a6f201d4a..30f1073141fc 100644 --- a/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c +++ b/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c @@ -85,7 +85,7 @@ static int intel_hdcp_gsc_initialize_message(struct xe_device *xe, cmd_in = xe_bo_ggtt_addr(bo); cmd_out = cmd_in + PAGE_SIZE; - xe_map_memset(xe, &bo->vmap, 0, 0, bo->size); + xe_map_memset(xe, &bo->vmap, 0, 0, xe_bo_size(bo)); gsc_context->hdcp_bo = bo; gsc_context->hdcp_cmd_in = cmd_in; diff --git a/drivers/gpu/drm/xe/regs/xe_gsc_regs.h b/drivers/gpu/drm/xe/regs/xe_gsc_regs.h index 7702364b65f1..9b66cc972a63 100644 --- a/drivers/gpu/drm/xe/regs/xe_gsc_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_gsc_regs.h @@ -16,6 +16,10 @@ #define MTL_GSC_HECI1_BASE 0x00116000 #define MTL_GSC_HECI2_BASE 0x00117000 +#define DG1_GSC_HECI2_BASE 0x00259000 +#define PVC_GSC_HECI2_BASE 0x00285000 +#define DG2_GSC_HECI2_BASE 0x00374000 + #define HECI_H_CSR(base) XE_REG((base) + 0x4) #define HECI_H_CSR_IE REG_BIT(0) #define HECI_H_CSR_IS REG_BIT(1) diff --git a/drivers/gpu/drm/xe/regs/xe_i2c_regs.h b/drivers/gpu/drm/xe/regs/xe_i2c_regs.h new file mode 100644 index 000000000000..af781c8e4a80 --- /dev/null +++ b/drivers/gpu/drm/xe/regs/xe_i2c_regs.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: MIT */ +#ifndef _XE_I2C_REGS_H_ +#define _XE_I2C_REGS_H_ + +#include <linux/pci_regs.h> + +#include "xe_reg_defs.h" +#include "xe_regs.h" + +#define I2C_BRIDGE_OFFSET (SOC_BASE + 0xd9000) +#define I2C_CONFIG_SPACE_OFFSET (SOC_BASE + 0xf6000) +#define I2C_MEM_SPACE_OFFSET (SOC_BASE + 0xf7400) + +#define REG_SG_REMAP_ADDR_PREFIX XE_REG(SOC_BASE + 0x0164) +#define REG_SG_REMAP_ADDR_POSTFIX XE_REG(SOC_BASE + 0x0168) + +#define I2C_CONFIG_CMD XE_REG(I2C_CONFIG_SPACE_OFFSET + PCI_COMMAND) +#define I2C_CONFIG_PMCSR XE_REG(I2C_CONFIG_SPACE_OFFSET + 0x84) + +#endif /* _XE_I2C_REGS_H_ */ diff --git a/drivers/gpu/drm/xe/regs/xe_irq_regs.h b/drivers/gpu/drm/xe/regs/xe_irq_regs.h index f0ecfcac4003..13635e4331d4 100644 --- a/drivers/gpu/drm/xe/regs/xe_irq_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_irq_regs.h @@ -19,6 +19,7 @@ #define MASTER_IRQ REG_BIT(31) #define GU_MISC_IRQ REG_BIT(29) #define DISPLAY_IRQ REG_BIT(16) +#define I2C_IRQ REG_BIT(12) #define GT_DW_IRQ(x) REG_BIT(x) /* diff --git a/drivers/gpu/drm/xe/regs/xe_pmt.h b/drivers/gpu/drm/xe/regs/xe_pmt.h index b0efd9b48d1e..2995d72c3f78 100644 --- a/drivers/gpu/drm/xe/regs/xe_pmt.h +++ b/drivers/gpu/drm/xe/regs/xe_pmt.h @@ -5,7 +5,7 @@ #ifndef _XE_PMT_H_ #define _XE_PMT_H_ -#define SOC_BASE 0x280000 +#include "xe_regs.h" #define BMG_PMT_BASE_OFFSET 0xDB000 #define BMG_DISCOVERY_OFFSET (SOC_BASE + BMG_PMT_BASE_OFFSET) diff --git a/drivers/gpu/drm/xe/regs/xe_regs.h b/drivers/gpu/drm/xe/regs/xe_regs.h index 3abb17d2ca33..1926b4044314 100644 --- a/drivers/gpu/drm/xe/regs/xe_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_regs.h @@ -7,6 +7,8 @@ #include "regs/xe_reg_defs.h" +#define SOC_BASE 0x280000 + #define GU_CNTL_PROTECTED XE_REG(0x10100C) #define DRIVERINT_FLR_DIS REG_BIT(31) diff --git a/drivers/gpu/drm/xe/tests/xe_bo.c b/drivers/gpu/drm/xe/tests/xe_bo.c index 77ca1ab527ec..bb469096d072 100644 --- a/drivers/gpu/drm/xe/tests/xe_bo.c +++ b/drivers/gpu/drm/xe/tests/xe_bo.c @@ -106,7 +106,7 @@ static int ccs_test_migrate(struct xe_tile *tile, struct xe_bo *bo, } /* Check last CCS value, or at least last value in page. */ - offset = xe_device_ccs_bytes(tile_to_xe(tile), bo->size); + offset = xe_device_ccs_bytes(tile_to_xe(tile), xe_bo_size(bo)); offset = min_t(u32, offset, PAGE_SIZE) / sizeof(u64) - 1; if (cpu_map[offset] != get_val) { KUNIT_FAIL(test, diff --git a/drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c b/drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c index 537766cdd882..d266882adc0e 100644 --- a/drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c +++ b/drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c @@ -32,7 +32,7 @@ static struct xe_bo *replacement_xe_managed_bo_create_pin_map(struct xe_device * bo->tile = tile; bo->ttm.bdev = &xe->ttm; - bo->size = size; + bo->ttm.base.size = size; iosys_map_set_vaddr(&bo->vmap, buf); if (flags & XE_BO_FLAG_GGTT) { @@ -43,7 +43,7 @@ static struct xe_bo *replacement_xe_managed_bo_create_pin_map(struct xe_device * KUNIT_ASSERT_EQ(test, 0, xe_ggtt_node_insert(bo->ggtt_node[tile->id], - bo->size, SZ_4K)); + xe_bo_size(bo), SZ_4K)); } return bo; diff --git a/drivers/gpu/drm/xe/tests/xe_migrate.c b/drivers/gpu/drm/xe/tests/xe_migrate.c index 4a65e3103f77..edd1e701aa1c 100644 --- a/drivers/gpu/drm/xe/tests/xe_migrate.c +++ b/drivers/gpu/drm/xe/tests/xe_migrate.c @@ -74,13 +74,13 @@ static void test_copy(struct xe_migrate *m, struct xe_bo *bo, { struct xe_device *xe = tile_to_xe(m->tile); u64 retval, expected = 0; - bool big = bo->size >= SZ_2M; + bool big = xe_bo_size(bo) >= SZ_2M; struct dma_fence *fence; const char *str = big ? "Copying big bo" : "Copying small bo"; int err; struct xe_bo *remote = xe_bo_create_locked(xe, m->tile, NULL, - bo->size, + xe_bo_size(bo), ttm_bo_type_kernel, region | XE_BO_FLAG_NEEDS_CPU_ACCESS | @@ -105,7 +105,7 @@ static void test_copy(struct xe_migrate *m, struct xe_bo *bo, goto out_unlock; } - xe_map_memset(xe, &remote->vmap, 0, 0xd0, remote->size); + xe_map_memset(xe, &remote->vmap, 0, 0xd0, xe_bo_size(remote)); fence = xe_migrate_clear(m, remote, remote->ttm.resource, XE_MIGRATE_CLEAR_FLAG_FULL); if (!sanity_fence_failed(xe, fence, big ? "Clearing remote big bo" : @@ -113,15 +113,15 @@ static void test_copy(struct xe_migrate *m, struct xe_bo *bo, retval = xe_map_rd(xe, &remote->vmap, 0, u64); check(retval, expected, "remote first offset should be cleared", test); - retval = xe_map_rd(xe, &remote->vmap, remote->size - 8, u64); + retval = xe_map_rd(xe, &remote->vmap, xe_bo_size(remote) - 8, u64); check(retval, expected, "remote last offset should be cleared", test); } dma_fence_put(fence); /* Try to copy 0xc0 from remote to vram with 2MB or 64KiB/4KiB pages */ - xe_map_memset(xe, &remote->vmap, 0, 0xc0, remote->size); - xe_map_memset(xe, &bo->vmap, 0, 0xd0, bo->size); + xe_map_memset(xe, &remote->vmap, 0, 0xc0, xe_bo_size(remote)); + xe_map_memset(xe, &bo->vmap, 0, 0xd0, xe_bo_size(bo)); expected = 0xc0c0c0c0c0c0c0c0; fence = xe_migrate_copy(m, remote, bo, remote->ttm.resource, @@ -131,15 +131,15 @@ static void test_copy(struct xe_migrate *m, struct xe_bo *bo, retval = xe_map_rd(xe, &bo->vmap, 0, u64); check(retval, expected, "remote -> vram bo first offset should be copied", test); - retval = xe_map_rd(xe, &bo->vmap, bo->size - 8, u64); + retval = xe_map_rd(xe, &bo->vmap, xe_bo_size(bo) - 8, u64); check(retval, expected, "remote -> vram bo offset should be copied", test); } dma_fence_put(fence); /* And other way around.. slightly hacky.. */ - xe_map_memset(xe, &remote->vmap, 0, 0xd0, remote->size); - xe_map_memset(xe, &bo->vmap, 0, 0xc0, bo->size); + xe_map_memset(xe, &remote->vmap, 0, 0xd0, xe_bo_size(remote)); + xe_map_memset(xe, &bo->vmap, 0, 0xc0, xe_bo_size(bo)); fence = xe_migrate_copy(m, bo, remote, bo->ttm.resource, remote->ttm.resource, false); @@ -148,7 +148,7 @@ static void test_copy(struct xe_migrate *m, struct xe_bo *bo, retval = xe_map_rd(xe, &remote->vmap, 0, u64); check(retval, expected, "vram -> remote bo first offset should be copied", test); - retval = xe_map_rd(xe, &remote->vmap, bo->size - 8, u64); + retval = xe_map_rd(xe, &remote->vmap, xe_bo_size(bo) - 8, u64); check(retval, expected, "vram -> remote bo last offset should be copied", test); } @@ -245,9 +245,9 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test) if (m->q->vm->flags & XE_VM_FLAG_64K) expected |= XE_PTE_PS64; if (xe_bo_is_vram(pt)) - xe_res_first(pt->ttm.resource, 0, pt->size, &src_it); + xe_res_first(pt->ttm.resource, 0, xe_bo_size(pt), &src_it); else - xe_res_first_sg(xe_bo_sg(pt), 0, pt->size, &src_it); + xe_res_first_sg(xe_bo_sg(pt), 0, xe_bo_size(pt), &src_it); emit_pte(m, bb, NUM_KERNEL_PDE - 1, xe_bo_is_vram(pt), false, &src_it, XE_PAGE_SIZE, pt->ttm.resource); @@ -276,7 +276,7 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test) /* Clear a small bo */ kunit_info(test, "Clearing small buffer object\n"); - xe_map_memset(xe, &tiny->vmap, 0, 0x22, tiny->size); + xe_map_memset(xe, &tiny->vmap, 0, 0x22, xe_bo_size(tiny)); expected = 0; fence = xe_migrate_clear(m, tiny, tiny->ttm.resource, XE_MIGRATE_CLEAR_FLAG_FULL); @@ -286,7 +286,7 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test) dma_fence_put(fence); retval = xe_map_rd(xe, &tiny->vmap, 0, u32); check(retval, expected, "Command clear small first value", test); - retval = xe_map_rd(xe, &tiny->vmap, tiny->size - 4, u32); + retval = xe_map_rd(xe, &tiny->vmap, xe_bo_size(tiny) - 4, u32); check(retval, expected, "Command clear small last value", test); kunit_info(test, "Copying small buffer object to system\n"); @@ -298,7 +298,7 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test) /* Clear a big bo */ kunit_info(test, "Clearing big buffer object\n"); - xe_map_memset(xe, &big->vmap, 0, 0x11, big->size); + xe_map_memset(xe, &big->vmap, 0, 0x11, xe_bo_size(big)); expected = 0; fence = xe_migrate_clear(m, big, big->ttm.resource, XE_MIGRATE_CLEAR_FLAG_FULL); @@ -308,7 +308,7 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test) dma_fence_put(fence); retval = xe_map_rd(xe, &big->vmap, 0, u32); check(retval, expected, "Command clear big first value", test); - retval = xe_map_rd(xe, &big->vmap, big->size - 4, u32); + retval = xe_map_rd(xe, &big->vmap, xe_bo_size(big) - 4, u32); check(retval, expected, "Command clear big last value", test); kunit_info(test, "Copying big buffer object to system\n"); @@ -370,7 +370,7 @@ static struct dma_fence *blt_copy(struct xe_tile *tile, struct xe_migrate *m = tile->migrate; struct xe_device *xe = gt_to_xe(gt); struct dma_fence *fence = NULL; - u64 size = src_bo->size; + u64 size = xe_bo_size(src_bo); struct xe_res_cursor src_it, dst_it; struct ttm_resource *src = src_bo->ttm.resource, *dst = dst_bo->ttm.resource; u64 src_L0_ofs, dst_L0_ofs; @@ -498,7 +498,7 @@ static void test_migrate(struct xe_device *xe, struct xe_tile *tile, long ret; expected = 0xd0d0d0d0d0d0d0d0; - xe_map_memset(xe, &sys_bo->vmap, 0, 0xd0, sys_bo->size); + xe_map_memset(xe, &sys_bo->vmap, 0, 0xd0, xe_bo_size(sys_bo)); fence = blt_copy(tile, sys_bo, vram_bo, false, "Blit copy from sysmem to vram", test); if (!sanity_fence_failed(xe, fence, "Blit copy from sysmem to vram", test)) { @@ -523,7 +523,7 @@ static void test_migrate(struct xe_device *xe, struct xe_tile *tile, retval = xe_map_rd(xe, &vram_bo->vmap, 0, u64); check(retval, expected, "Clear evicted vram data first value", test); - retval = xe_map_rd(xe, &vram_bo->vmap, vram_bo->size - 8, u64); + retval = xe_map_rd(xe, &vram_bo->vmap, xe_bo_size(vram_bo) - 8, u64); check(retval, expected, "Clear evicted vram data last value", test); fence = blt_copy(tile, vram_bo, ccs_bo, @@ -532,7 +532,7 @@ static void test_migrate(struct xe_device *xe, struct xe_tile *tile, retval = xe_map_rd(xe, &ccs_bo->vmap, 0, u64); check(retval, 0, "Clear ccs data first value", test); - retval = xe_map_rd(xe, &ccs_bo->vmap, ccs_bo->size - 8, u64); + retval = xe_map_rd(xe, &ccs_bo->vmap, xe_bo_size(ccs_bo) - 8, u64); check(retval, 0, "Clear ccs data last value", test); } dma_fence_put(fence); @@ -562,7 +562,7 @@ static void test_migrate(struct xe_device *xe, struct xe_tile *tile, retval = xe_map_rd(xe, &vram_bo->vmap, 0, u64); check(retval, expected, "Restored value must be equal to initial value", test); - retval = xe_map_rd(xe, &vram_bo->vmap, vram_bo->size - 8, u64); + retval = xe_map_rd(xe, &vram_bo->vmap, xe_bo_size(vram_bo) - 8, u64); check(retval, expected, "Restored value must be equal to initial value", test); fence = blt_copy(tile, vram_bo, ccs_bo, @@ -570,7 +570,7 @@ static void test_migrate(struct xe_device *xe, struct xe_tile *tile, if (!sanity_fence_failed(xe, fence, "Clear ccs buffer data", test)) { retval = xe_map_rd(xe, &ccs_bo->vmap, 0, u64); check(retval, 0, "Clear ccs data first value", test); - retval = xe_map_rd(xe, &ccs_bo->vmap, ccs_bo->size - 8, u64); + retval = xe_map_rd(xe, &ccs_bo->vmap, xe_bo_size(ccs_bo) - 8, u64); check(retval, 0, "Clear ccs data last value", test); } dma_fence_put(fence); @@ -583,7 +583,7 @@ static void test_clear(struct xe_device *xe, struct xe_tile *tile, u64 expected, retval; expected = 0xd0d0d0d0d0d0d0d0; - xe_map_memset(xe, &sys_bo->vmap, 0, 0xd0, sys_bo->size); + xe_map_memset(xe, &sys_bo->vmap, 0, 0xd0, xe_bo_size(sys_bo)); fence = blt_copy(tile, sys_bo, vram_bo, false, "Blit copy from sysmem to vram", test); if (!sanity_fence_failed(xe, fence, "Blit copy from sysmem to vram", test)) { @@ -597,7 +597,7 @@ static void test_clear(struct xe_device *xe, struct xe_tile *tile, if (!sanity_fence_failed(xe, fence, "Blit copy from vram to sysmem", test)) { retval = xe_map_rd(xe, &sys_bo->vmap, 0, u64); check(retval, expected, "Decompressed value must be equal to initial value", test); - retval = xe_map_rd(xe, &sys_bo->vmap, sys_bo->size - 8, u64); + retval = xe_map_rd(xe, &sys_bo->vmap, xe_bo_size(sys_bo) - 8, u64); check(retval, expected, "Decompressed value must be equal to initial value", test); } dma_fence_put(fence); @@ -615,7 +615,7 @@ static void test_clear(struct xe_device *xe, struct xe_tile *tile, if (!sanity_fence_failed(xe, fence, "Clear main buffer data", test)) { retval = xe_map_rd(xe, &sys_bo->vmap, 0, u64); check(retval, expected, "Clear main buffer first value", test); - retval = xe_map_rd(xe, &sys_bo->vmap, sys_bo->size - 8, u64); + retval = xe_map_rd(xe, &sys_bo->vmap, xe_bo_size(sys_bo) - 8, u64); check(retval, expected, "Clear main buffer last value", test); } dma_fence_put(fence); @@ -625,7 +625,7 @@ static void test_clear(struct xe_device *xe, struct xe_tile *tile, if (!sanity_fence_failed(xe, fence, "Clear ccs buffer data", test)) { retval = xe_map_rd(xe, &sys_bo->vmap, 0, u64); check(retval, expected, "Clear ccs data first value", test); - retval = xe_map_rd(xe, &sys_bo->vmap, sys_bo->size - 8, u64); + retval = xe_map_rd(xe, &sys_bo->vmap, xe_bo_size(sys_bo) - 8, u64); check(retval, expected, "Clear ccs data last value", test); } dma_fence_put(fence); diff --git a/drivers/gpu/drm/xe/tests/xe_pci.c b/drivers/gpu/drm/xe/tests/xe_pci.c index baccb657bd05..9c715e59f030 100644 --- a/drivers/gpu/drm/xe/tests/xe_pci.c +++ b/drivers/gpu/drm/xe/tests/xe_pci.c @@ -21,6 +21,18 @@ static void xe_ip_kunit_desc(const struct xe_ip *param, char *desc) KUNIT_ARRAY_PARAM(graphics_ip, graphics_ips, xe_ip_kunit_desc); KUNIT_ARRAY_PARAM(media_ip, media_ips, xe_ip_kunit_desc); +static void xe_pci_id_kunit_desc(const struct pci_device_id *param, char *desc) +{ + const struct xe_device_desc *dev_desc = + (const struct xe_device_desc *)param->driver_data; + + if (dev_desc) + snprintf(desc, KUNIT_PARAM_DESC_SIZE, "0x%X (%s)", + param->device, dev_desc->platform_name); +} + +KUNIT_ARRAY_PARAM(pci_id, pciidlist, xe_pci_id_kunit_desc); + /** * xe_pci_graphics_ip_gen_param - Generate graphics struct xe_ip parameters * @prev: the pointer to the previous parameter to iterate from or NULL @@ -55,6 +67,25 @@ const void *xe_pci_media_ip_gen_param(const void *prev, char *desc) } EXPORT_SYMBOL_IF_KUNIT(xe_pci_media_ip_gen_param); +/** + * xe_pci_id_gen_param - Generate struct pci_device_id parameters + * @prev: the pointer to the previous parameter to iterate from or NULL + * @desc: output buffer with minimum size of KUNIT_PARAM_DESC_SIZE + * + * This function prepares struct pci_device_id parameter. + * + * To be used only as a parameter generator function in &KUNIT_CASE_PARAM. + * + * Return: pointer to the next parameter or NULL if no more parameters + */ +const void *xe_pci_id_gen_param(const void *prev, char *desc) +{ + const struct pci_device_id *pci = pci_id_gen_params(prev, desc); + + return pci->driver_data ? pci : NULL; +} +EXPORT_SYMBOL_IF_KUNIT(xe_pci_id_gen_param); + static void fake_read_gmdid(struct xe_device *xe, enum xe_gmdid_type type, u32 *ver, u32 *revid) { diff --git a/drivers/gpu/drm/xe/tests/xe_pci_test.c b/drivers/gpu/drm/xe/tests/xe_pci_test.c index 95fed41f7ff2..37b344df2dc3 100644 --- a/drivers/gpu/drm/xe/tests/xe_pci_test.c +++ b/drivers/gpu/drm/xe/tests/xe_pci_test.c @@ -44,9 +44,21 @@ static void check_media_ip(struct kunit *test) KUNIT_ASSERT_EQ(test, mask, 0); } +static void check_platform_gt_count(struct kunit *test) +{ + const struct pci_device_id *pci = test->param_value; + const struct xe_device_desc *desc = + (const struct xe_device_desc *)pci->driver_data; + int max_gt = desc->max_gt_per_tile; + + KUNIT_ASSERT_GT(test, max_gt, 0); + KUNIT_ASSERT_LE(test, max_gt, XE_MAX_GT_PER_TILE); +} + static struct kunit_case xe_pci_tests[] = { KUNIT_CASE_PARAM(check_graphics_ip, xe_pci_graphics_ip_gen_param), KUNIT_CASE_PARAM(check_media_ip, xe_pci_media_ip_gen_param), + KUNIT_CASE_PARAM(check_platform_gt_count, xe_pci_id_gen_param), {} }; diff --git a/drivers/gpu/drm/xe/tests/xe_pci_test.h b/drivers/gpu/drm/xe/tests/xe_pci_test.h index 3a1df7a5e291..ce4d2b86b778 100644 --- a/drivers/gpu/drm/xe/tests/xe_pci_test.h +++ b/drivers/gpu/drm/xe/tests/xe_pci_test.h @@ -27,6 +27,7 @@ int xe_pci_fake_device_init(struct xe_device *xe); const void *xe_pci_graphics_ip_gen_param(const void *prev, char *desc); const void *xe_pci_media_ip_gen_param(const void *prev, char *desc); +const void *xe_pci_id_gen_param(const void *prev, char *desc); const void *xe_pci_live_device_gen_param(const void *prev, char *desc); #endif diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index 4e39188a021a..4e0355d0f406 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -19,6 +19,8 @@ #include <kunit/static_stub.h> +#include <trace/events/gpu_mem.h> + #include "xe_device.h" #include "xe_dma_buf.h" #include "xe_drm_client.h" @@ -418,6 +420,19 @@ static void xe_ttm_tt_account_subtract(struct xe_device *xe, struct ttm_tt *tt) xe_shrinker_mod_pages(xe->mem.shrinker, -(long)tt->num_pages, 0); } +static void update_global_total_pages(struct ttm_device *ttm_dev, + long num_pages) +{ +#if IS_ENABLED(CONFIG_TRACE_GPU_MEM) + struct xe_device *xe = ttm_to_xe_device(ttm_dev); + u64 global_total_pages = + atomic64_add_return(num_pages, &xe->global_total_pages); + + trace_gpu_mem_total(xe->drm.primary->index, 0, + global_total_pages << PAGE_SHIFT); +#endif +} + static struct ttm_tt *xe_ttm_tt_create(struct ttm_buffer_object *ttm_bo, u32 page_flags) { @@ -437,7 +452,7 @@ static struct ttm_tt *xe_ttm_tt_create(struct ttm_buffer_object *ttm_bo, extra_pages = 0; if (xe_bo_needs_ccs_pages(bo)) - extra_pages = DIV_ROUND_UP(xe_device_ccs_bytes(xe, bo->size), + extra_pages = DIV_ROUND_UP(xe_device_ccs_bytes(xe, xe_bo_size(bo)), PAGE_SIZE); /* @@ -525,6 +540,7 @@ static int xe_ttm_tt_populate(struct ttm_device *ttm_dev, struct ttm_tt *tt, xe_tt->purgeable = false; xe_ttm_tt_account_add(ttm_to_xe_device(ttm_dev), tt); + update_global_total_pages(ttm_dev, tt->num_pages); return 0; } @@ -541,6 +557,7 @@ static void xe_ttm_tt_unpopulate(struct ttm_device *ttm_dev, struct ttm_tt *tt) ttm_pool_free(&ttm_dev->pool, tt); xe_ttm_tt_account_subtract(xe, tt); + update_global_total_pages(ttm_dev, -(long)tt->num_pages); } static void xe_ttm_tt_destroy(struct ttm_device *ttm_dev, struct ttm_tt *tt) @@ -1122,7 +1139,7 @@ int xe_bo_notifier_prepare_pinned(struct xe_bo *bo) if (bo->flags & XE_BO_FLAG_PINNED_NORESTORE) goto out_unlock_bo; - backup = ___xe_bo_create_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, bo->size, + backup = ___xe_bo_create_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, xe_bo_size(bo), DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel, XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS | XE_BO_FLAG_PINNED); @@ -1200,7 +1217,8 @@ int xe_bo_evict_pinned(struct xe_bo *bo) goto out_unlock_bo; if (!backup) { - backup = ___xe_bo_create_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, bo->size, + backup = ___xe_bo_create_locked(xe, NULL, NULL, bo->ttm.base.resv, + NULL, xe_bo_size(bo), DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel, XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS | XE_BO_FLAG_PINNED); @@ -1254,7 +1272,7 @@ int xe_bo_evict_pinned(struct xe_bo *bo) } xe_map_memcpy_from(xe, backup->vmap.vaddr, &bo->vmap, 0, - bo->size); + xe_bo_size(bo)); } if (!bo->backup_obj) @@ -1347,7 +1365,7 @@ int xe_bo_restore_pinned(struct xe_bo *bo) } xe_map_memcpy_to(xe, &bo->vmap, 0, backup->vmap.vaddr, - bo->size); + xe_bo_size(bo)); } bo->backup_obj = NULL; @@ -1558,7 +1576,7 @@ static int xe_ttm_access_memory(struct ttm_buffer_object *ttm_bo, vram = res_to_mem_region(ttm_bo->resource); xe_res_first(ttm_bo->resource, offset & PAGE_MASK, - bo->size - (offset & PAGE_MASK), &cursor); + xe_bo_size(bo) - (offset & PAGE_MASK), &cursor); do { unsigned long page_offset = (offset & ~PAGE_MASK); @@ -1858,7 +1876,6 @@ struct xe_bo *___xe_bo_create_locked(struct xe_device *xe, struct xe_bo *bo, bo->ccs_cleared = false; bo->tile = tile; - bo->size = size; bo->flags = flags; bo->cpu_caching = cpu_caching; bo->ttm.base.funcs = &xe_gem_object_funcs; @@ -2036,7 +2053,7 @@ __xe_bo_create_locked(struct xe_device *xe, if (flags & XE_BO_FLAG_FIXED_PLACEMENT) { err = xe_ggtt_insert_bo_at(t->mem.ggtt, bo, - start + bo->size, U64_MAX); + start + xe_bo_size(bo), U64_MAX); } else { err = xe_ggtt_insert_bo(t->mem.ggtt, bo); } @@ -2234,7 +2251,7 @@ int xe_managed_bo_reinit_in_vram(struct xe_device *xe, struct xe_tile *tile, str xe_assert(xe, !(*src)->vmap.is_iomem); bo = xe_managed_bo_create_from_data(xe, tile, (*src)->vmap.vaddr, - (*src)->size, dst_flags); + xe_bo_size(*src), dst_flags); if (IS_ERR(bo)) return PTR_ERR(bo); @@ -2524,7 +2541,7 @@ int xe_bo_vmap(struct xe_bo *bo) * TODO: Fix up ttm_bo_vmap to do that, or fix up ttm_bo_kmap * to use struct iosys_map. */ - ret = ttm_bo_kmap(&bo->ttm, 0, bo->size >> PAGE_SHIFT, &bo->kmap); + ret = ttm_bo_kmap(&bo->ttm, 0, xe_bo_size(bo) >> PAGE_SHIFT, &bo->kmap); if (ret) return ret; diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h index 02ada1fb8a23..8559901e4088 100644 --- a/drivers/gpu/drm/xe/xe_bo.h +++ b/drivers/gpu/drm/xe/xe_bo.h @@ -238,6 +238,19 @@ xe_bo_main_addr(struct xe_bo *bo, size_t page_size) return xe_bo_addr(bo, 0, page_size); } +/** + * xe_bo_size() - Xe BO size + * @bo: The bo object. + * + * Simple helper to return Xe BO's size. + * + * Return: Xe BO's size + */ +static inline size_t xe_bo_size(struct xe_bo *bo) +{ + return bo->ttm.base.size; +} + static inline u32 __xe_bo_ggtt_addr(struct xe_bo *bo, u8 tile_id) { @@ -246,7 +259,7 @@ __xe_bo_ggtt_addr(struct xe_bo *bo, u8 tile_id) if (XE_WARN_ON(!ggtt_node)) return 0; - XE_WARN_ON(ggtt_node->base.size > bo->size); + XE_WARN_ON(ggtt_node->base.size > xe_bo_size(bo)); XE_WARN_ON(ggtt_node->base.start + ggtt_node->base.size > (1ull << 32)); return ggtt_node->base.start; } @@ -300,7 +313,7 @@ bool xe_bo_needs_ccs_pages(struct xe_bo *bo); static inline size_t xe_bo_ccs_pages_start(struct xe_bo *bo) { - return PAGE_ALIGN(bo->ttm.base.size); + return PAGE_ALIGN(xe_bo_size(bo)); } static inline bool xe_bo_has_pages(struct xe_bo *bo) diff --git a/drivers/gpu/drm/xe/xe_bo_types.h b/drivers/gpu/drm/xe/xe_bo_types.h index eb5e83c5f233..ff560d82496f 100644 --- a/drivers/gpu/drm/xe/xe_bo_types.h +++ b/drivers/gpu/drm/xe/xe_bo_types.h @@ -32,8 +32,6 @@ struct xe_bo { struct xe_bo *backup_obj; /** @parent_obj: Ref to parent bo if this a backup_obj */ struct xe_bo *parent_obj; - /** @size: Size of this buffer object */ - size_t size; /** @flags: flags for this buffer object */ u32 flags; /** @vm: VM this BO is attached to, for extobj this will be NULL */ @@ -86,7 +84,7 @@ struct xe_bo { u16 cpu_caching; /** @devmem_allocation: SVM device memory allocation */ - struct drm_gpusvm_devmem devmem_allocation; + struct drm_pagemap_devmem devmem_allocation; /** @vram_userfault_link: Link into @mem_access.vram_userfault.list */ struct list_head vram_userfault_link; diff --git a/drivers/gpu/drm/xe/xe_devcoredump.c b/drivers/gpu/drm/xe/xe_devcoredump.c index 7a8af2311318..203e3038cc81 100644 --- a/drivers/gpu/drm/xe/xe_devcoredump.c +++ b/drivers/gpu/drm/xe/xe_devcoredump.c @@ -171,14 +171,32 @@ static void xe_devcoredump_snapshot_free(struct xe_devcoredump_snapshot *ss) #define XE_DEVCOREDUMP_CHUNK_MAX (SZ_512M + SZ_1G) +/** + * xe_devcoredump_read() - Read data from the Xe device coredump snapshot + * @buffer: Destination buffer to copy the coredump data into + * @offset: Offset in the coredump data to start reading from + * @count: Number of bytes to read + * @data: Pointer to the xe_devcoredump structure + * @datalen: Length of the data (unused) + * + * Reads a chunk of the coredump snapshot data into the provided buffer. + * If the devcoredump is smaller than 1.5 GB (XE_DEVCOREDUMP_CHUNK_MAX), + * it is read directly from a pre-written buffer. For larger devcoredumps, + * the pre-written buffer must be periodically repopulated from the snapshot + * state due to kmalloc size limitations. + * + * Return: Number of bytes copied on success, or a negative error code on failure. + */ static ssize_t xe_devcoredump_read(char *buffer, loff_t offset, size_t count, void *data, size_t datalen) { struct xe_devcoredump *coredump = data; struct xe_devcoredump_snapshot *ss; - ssize_t byte_copied; + ssize_t byte_copied = 0; u32 chunk_offset; ssize_t new_chunk_position; + bool pm_needed = false; + int ret = 0; if (!coredump) return -ENODEV; @@ -188,20 +206,19 @@ static ssize_t xe_devcoredump_read(char *buffer, loff_t offset, /* Ensure delayed work is captured before continuing */ flush_work(&ss->work); - if (ss->read.size > XE_DEVCOREDUMP_CHUNK_MAX) + pm_needed = ss->read.size > XE_DEVCOREDUMP_CHUNK_MAX; + if (pm_needed) xe_pm_runtime_get(gt_to_xe(ss->gt)); mutex_lock(&coredump->lock); if (!ss->read.buffer) { - mutex_unlock(&coredump->lock); - return -ENODEV; + ret = -ENODEV; + goto unlock; } - if (offset >= ss->read.size) { - mutex_unlock(&coredump->lock); - return 0; - } + if (offset >= ss->read.size) + goto unlock; new_chunk_position = div_u64_rem(offset, XE_DEVCOREDUMP_CHUNK_MAX, @@ -221,12 +238,13 @@ static ssize_t xe_devcoredump_read(char *buffer, loff_t offset, ss->read.size - offset; memcpy(buffer, ss->read.buffer + chunk_offset, byte_copied); +unlock: mutex_unlock(&coredump->lock); - if (ss->read.size > XE_DEVCOREDUMP_CHUNK_MAX) + if (pm_needed) xe_pm_runtime_put(gt_to_xe(ss->gt)); - return byte_copied; + return byte_copied ? byte_copied : ret; } static void xe_devcoredump_free(void *data) @@ -313,13 +331,9 @@ static void devcoredump_snapshot(struct xe_devcoredump *coredump, { struct xe_devcoredump_snapshot *ss = &coredump->snapshot; struct xe_guc *guc = exec_queue_to_guc(q); - u32 adj_logical_mask = q->logical_mask; - u32 width_mask = (0x1 << q->width) - 1; const char *process_name = "no process"; - unsigned int fw_ref; bool cookie; - int i; ss->snapshot_time = ktime_get_real(); ss->boot_time = ktime_get_boottime(); @@ -335,14 +349,6 @@ static void devcoredump_snapshot(struct xe_devcoredump *coredump, INIT_WORK(&ss->work, xe_devcoredump_deferred_snap_work); cookie = dma_fence_begin_signalling(); - for (i = 0; q->width > 1 && i < XE_HW_ENGINE_MAX_INSTANCE;) { - if (adj_logical_mask & BIT(i)) { - adj_logical_mask |= width_mask << i; - i += q->width; - } else { - ++i; - } - } /* keep going if fw fails as we still want to save the memory and SW data */ fw_ref = xe_force_wake_get(gt_to_fw(q->gt), XE_FORCEWAKE_ALL); diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c index 8cfcfff250ca..6db09cfc8eb8 100644 --- a/drivers/gpu/drm/xe/xe_device.c +++ b/drivers/gpu/drm/xe/xe_device.c @@ -40,12 +40,14 @@ #include "xe_gt_printk.h" #include "xe_gt_sriov_vf.h" #include "xe_guc.h" +#include "xe_guc_pc.h" #include "xe_hw_engine_group.h" #include "xe_hwmon.h" +#include "xe_i2c.h" #include "xe_irq.h" -#include "xe_memirq.h" #include "xe_mmio.h" #include "xe_module.h" +#include "xe_nvm.h" #include "xe_oa.h" #include "xe_observation.h" #include "xe_pat.h" @@ -783,46 +785,16 @@ int xe_device_probe(struct xe_device *xe) if (err) return err; - err = xe_ttm_sys_mgr_init(xe); - if (err) - return err; - for_each_gt(gt, xe, id) { err = xe_gt_init_early(gt); if (err) return err; - - /* - * Only after this point can GT-specific MMIO operations - * (including things like communication with the GuC) - * be performed. - */ - xe_gt_mmio_init(gt); - - if (IS_SRIOV_VF(xe)) { - xe_guc_comm_init_early(>->uc.guc); - err = xe_gt_sriov_vf_bootstrap(gt); - if (err) - return err; - err = xe_gt_sriov_vf_query_config(gt); - if (err) - return err; - } } for_each_tile(tile, xe, id) { err = xe_ggtt_init_early(tile->mem.ggtt); if (err) return err; - err = xe_memirq_init(&tile->memirq); - if (err) - return err; - } - - for_each_gt(gt, xe, id) { - err = xe_gt_init_hwconfig(gt); - if (err) - return err; } err = xe_devcoredump_init(xe); @@ -850,6 +822,14 @@ int xe_device_probe(struct xe_device *xe) return err; } + /* + * Allow allocations only now to ensure xe_display_init_early() + * is the first to allocate, always. + */ + err = xe_ttm_sys_mgr_init(xe); + if (err) + return err; + /* Allocate and map stolen after potential VRAM resize */ err = xe_ttm_stolen_mgr_init(xe); if (err) @@ -881,6 +861,8 @@ int xe_device_probe(struct xe_device *xe) return err; } + xe_nvm_init(xe); + err = xe_heci_gsc_init(xe); if (err) return err; @@ -921,6 +903,10 @@ int xe_device_probe(struct xe_device *xe) if (err) goto err_unregister_display; + err = xe_i2c_probe(xe); + if (err) + goto err_unregister_display; + for_each_gt(gt, xe, id) xe_gt_sanitize_freq(gt); @@ -938,6 +924,8 @@ void xe_device_remove(struct xe_device *xe) { xe_display_unregister(xe); + xe_nvm_fini(xe); + drm_dev_unplug(&xe->drm); xe_bo_pci_dev_remove_all(xe); @@ -981,38 +969,15 @@ void xe_device_wmb(struct xe_device *xe) xe_mmio_write32(xe_root_tile_mmio(xe), VF_CAP_REG, 0); } -/** - * xe_device_td_flush() - Flush transient L3 cache entries - * @xe: The device - * - * Display engine has direct access to memory and is never coherent with L3/L4 - * caches (or CPU caches), however KMD is responsible for specifically flushing - * transient L3 GPU cache entries prior to the flip sequence to ensure scanout - * can happen from such a surface without seeing corruption. - * - * Display surfaces can be tagged as transient by mapping it using one of the - * various L3:XD PAT index modes on Xe2. - * - * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed - * at the end of each submission via PIPE_CONTROL for compute/render, since SA - * Media is not coherent with L3 and we want to support render-vs-media - * usescases. For other engines like copy/blt the HW internally forces uncached - * behaviour, hence why we can skip the TDF on such platforms. +/* + * Issue a TRANSIENT_FLUSH_REQUEST and wait for completion on each gt. */ -void xe_device_td_flush(struct xe_device *xe) +static void tdf_request_sync(struct xe_device *xe) { - struct xe_gt *gt; unsigned int fw_ref; + struct xe_gt *gt; u8 id; - if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20) - return; - - if (XE_WA(xe_root_mmio_gt(xe), 16023588340)) { - xe_device_l2_flush(xe); - return; - } - for_each_gt(gt, xe, id) { if (xe_gt_is_media_type(gt)) continue; @@ -1022,6 +987,7 @@ void xe_device_td_flush(struct xe_device *xe) return; xe_mmio_write32(>->mmio, XE2_TDF_CTRL, TRANSIENT_FLUSH_REQUEST); + /* * FIXME: We can likely do better here with our choice of * timeout. Currently we just assume the worst case, i.e. 150us, @@ -1052,15 +1018,52 @@ void xe_device_l2_flush(struct xe_device *xe) return; spin_lock(>->global_invl_lock); - xe_mmio_write32(>->mmio, XE2_GLOBAL_INVAL, 0x1); + xe_mmio_write32(>->mmio, XE2_GLOBAL_INVAL, 0x1); if (xe_mmio_wait32(>->mmio, XE2_GLOBAL_INVAL, 0x1, 0x0, 500, NULL, true)) xe_gt_err_once(gt, "Global invalidation timeout\n"); + spin_unlock(>->global_invl_lock); xe_force_wake_put(gt_to_fw(gt), fw_ref); } +/** + * xe_device_td_flush() - Flush transient L3 cache entries + * @xe: The device + * + * Display engine has direct access to memory and is never coherent with L3/L4 + * caches (or CPU caches), however KMD is responsible for specifically flushing + * transient L3 GPU cache entries prior to the flip sequence to ensure scanout + * can happen from such a surface without seeing corruption. + * + * Display surfaces can be tagged as transient by mapping it using one of the + * various L3:XD PAT index modes on Xe2. + * + * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed + * at the end of each submission via PIPE_CONTROL for compute/render, since SA + * Media is not coherent with L3 and we want to support render-vs-media + * usescases. For other engines like copy/blt the HW internally forces uncached + * behaviour, hence why we can skip the TDF on such platforms. + */ +void xe_device_td_flush(struct xe_device *xe) +{ + struct xe_gt *root_gt; + + if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20) + return; + + root_gt = xe_root_mmio_gt(xe); + if (XE_WA(root_gt, 16023588340)) { + /* A transient flush is not sufficient: flush the L2 */ + xe_device_l2_flush(xe); + } else { + xe_guc_pc_apply_flush_freq_limit(&root_gt->uc.guc.pc); + tdf_request_sync(xe); + xe_guc_pc_remove_flush_freq_limit(&root_gt->uc.guc.pc); + } +} + u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size) { return xe_device_has_flat_ccs(xe) ? diff --git a/drivers/gpu/drm/xe/xe_device.h b/drivers/gpu/drm/xe/xe_device.h index e4da797a984b..f0eb8150f185 100644 --- a/drivers/gpu/drm/xe/xe_device.h +++ b/drivers/gpu/drm/xe/xe_device.h @@ -60,35 +60,32 @@ static inline struct xe_tile *xe_device_get_root_tile(struct xe_device *xe) return &xe->tiles[0]; } +/* + * Highest GT/tile count for any platform. Used only for memory allocation + * sizing. Any logic looping over GTs or mapping userspace GT IDs into GT + * structures should use the per-platform xe->info.max_gt_per_tile instead. + */ #define XE_MAX_GT_PER_TILE 2 -static inline struct xe_gt *xe_tile_get_gt(struct xe_tile *tile, u8 gt_id) -{ - if (drm_WARN_ON(&tile_to_xe(tile)->drm, gt_id >= XE_MAX_GT_PER_TILE)) - gt_id = 0; - - return gt_id ? tile->media_gt : tile->primary_gt; -} - static inline struct xe_gt *xe_device_get_gt(struct xe_device *xe, u8 gt_id) { - struct xe_tile *root_tile = xe_device_get_root_tile(xe); + struct xe_tile *tile; struct xe_gt *gt; - /* - * FIXME: This only works for now because multi-tile and standalone - * media are mutually exclusive on the platforms we have today. - * - * id => GT mapping may change once we settle on how we want to handle - * our UAPI. - */ - if (MEDIA_VER(xe) >= 13) { - gt = xe_tile_get_gt(root_tile, gt_id); - } else { - if (drm_WARN_ON(&xe->drm, gt_id >= XE_MAX_TILES_PER_DEVICE)) - gt_id = 0; - - gt = xe->tiles[gt_id].primary_gt; + if (gt_id >= xe->info.tile_count * xe->info.max_gt_per_tile) + return NULL; + + tile = &xe->tiles[gt_id / xe->info.max_gt_per_tile]; + switch (gt_id % xe->info.max_gt_per_tile) { + default: + xe_assert(xe, false); + fallthrough; + case 0: + gt = tile->primary_gt; + break; + case 1: + gt = tile->media_gt; + break; } if (!gt) @@ -130,12 +127,8 @@ static inline bool xe_device_uc_enabled(struct xe_device *xe) for ((id__) = 1; (id__) < (xe__)->info.tile_count; (id__)++) \ for_each_if((tile__) = &(xe__)->tiles[(id__)]) -/* - * FIXME: This only works for now since multi-tile and standalone media - * happen to be mutually exclusive. Future platforms may change this... - */ #define for_each_gt(gt__, xe__, id__) \ - for ((id__) = 0; (id__) < (xe__)->info.gt_count; (id__)++) \ + for ((id__) = 0; (id__) < (xe__)->info.tile_count * (xe__)->info.max_gt_per_tile; (id__)++) \ for_each_if((gt__) = xe_device_get_gt((xe__), (id__))) static inline struct xe_force_wake *gt_to_fw(struct xe_gt *gt) diff --git a/drivers/gpu/drm/xe/xe_device_sysfs.c b/drivers/gpu/drm/xe/xe_device_sysfs.c index b9440f8c781e..e5fd0cd537bc 100644 --- a/drivers/gpu/drm/xe/xe_device_sysfs.c +++ b/drivers/gpu/drm/xe/xe_device_sysfs.c @@ -24,6 +24,12 @@ * * vram_d3cold_threshold - Report/change vram used threshold(in MB) below * which vram save/restore is permissible during runtime D3cold entry/exit. + * + * lb_fan_control_version - Fan control version provisioned by late binding. + * Exposed only if supported by the device. + * + * lb_voltage_regulator_version - Voltage regulator version provisioned by late + * binding. Exposed only if supported by the device. */ static ssize_t @@ -65,6 +71,135 @@ vram_d3cold_threshold_store(struct device *dev, struct device_attribute *attr, static DEVICE_ATTR_RW(vram_d3cold_threshold); +static ssize_t +lb_fan_control_version_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev)); + struct xe_tile *root = xe_device_get_root_tile(xe); + u32 cap, ver_low = FAN_TABLE, ver_high = FAN_TABLE; + u16 major = 0, minor = 0, hotfix = 0, build = 0; + int ret; + + xe_pm_runtime_get(xe); + + ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_CAPABILITY_STATUS, 0), + &cap, NULL); + if (ret) + goto out; + + if (REG_FIELD_GET(V1_FAN_PROVISIONED, cap)) { + ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_VERSION_LOW, 0), + &ver_low, NULL); + if (ret) + goto out; + + ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_VERSION_HIGH, 0), + &ver_high, NULL); + if (ret) + goto out; + + major = REG_FIELD_GET(MAJOR_VERSION_MASK, ver_low); + minor = REG_FIELD_GET(MINOR_VERSION_MASK, ver_low); + hotfix = REG_FIELD_GET(HOTFIX_VERSION_MASK, ver_high); + build = REG_FIELD_GET(BUILD_VERSION_MASK, ver_high); + } +out: + xe_pm_runtime_put(xe); + + return ret ?: sysfs_emit(buf, "%u.%u.%u.%u\n", major, minor, hotfix, build); +} +static DEVICE_ATTR_ADMIN_RO(lb_fan_control_version); + +static ssize_t +lb_voltage_regulator_version_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev)); + struct xe_tile *root = xe_device_get_root_tile(xe); + u32 cap, ver_low = VR_CONFIG, ver_high = VR_CONFIG; + u16 major = 0, minor = 0, hotfix = 0, build = 0; + int ret; + + xe_pm_runtime_get(xe); + + ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_CAPABILITY_STATUS, 0), + &cap, NULL); + if (ret) + goto out; + + if (REG_FIELD_GET(VR_PARAMS_PROVISIONED, cap)) { + ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_VERSION_LOW, 0), + &ver_low, NULL); + if (ret) + goto out; + + ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_VERSION_HIGH, 0), + &ver_high, NULL); + if (ret) + goto out; + + major = REG_FIELD_GET(MAJOR_VERSION_MASK, ver_low); + minor = REG_FIELD_GET(MINOR_VERSION_MASK, ver_low); + hotfix = REG_FIELD_GET(HOTFIX_VERSION_MASK, ver_high); + build = REG_FIELD_GET(BUILD_VERSION_MASK, ver_high); + } +out: + xe_pm_runtime_put(xe); + + return ret ?: sysfs_emit(buf, "%u.%u.%u.%u\n", major, minor, hotfix, build); +} +static DEVICE_ATTR_ADMIN_RO(lb_voltage_regulator_version); + +static int late_bind_create_files(struct device *dev) +{ + struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev)); + struct xe_tile *root = xe_device_get_root_tile(xe); + u32 cap; + int ret; + + xe_pm_runtime_get(xe); + + ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_CAPABILITY_STATUS, 0), + &cap, NULL); + if (ret) + goto out; + + if (REG_FIELD_GET(V1_FAN_SUPPORTED, cap)) { + ret = sysfs_create_file(&dev->kobj, &dev_attr_lb_fan_control_version.attr); + if (ret) + goto out; + } + + if (REG_FIELD_GET(VR_PARAMS_SUPPORTED, cap)) + ret = sysfs_create_file(&dev->kobj, &dev_attr_lb_voltage_regulator_version.attr); +out: + xe_pm_runtime_put(xe); + + return ret; +} + +static void late_bind_remove_files(struct device *dev) +{ + struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev)); + struct xe_tile *root = xe_device_get_root_tile(xe); + u32 cap; + int ret; + + xe_pm_runtime_get(xe); + + ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_CAPABILITY_STATUS, 0), + &cap, NULL); + if (ret) + goto out; + + if (REG_FIELD_GET(V1_FAN_SUPPORTED, cap)) + sysfs_remove_file(&dev->kobj, &dev_attr_lb_fan_control_version.attr); + + if (REG_FIELD_GET(VR_PARAMS_SUPPORTED, cap)) + sysfs_remove_file(&dev->kobj, &dev_attr_lb_voltage_regulator_version.attr); +out: + xe_pm_runtime_put(xe); +} + /** * DOC: PCIe Gen5 Limitations * @@ -151,8 +286,10 @@ static void xe_device_sysfs_fini(void *arg) if (xe->d3cold.capable) sysfs_remove_file(&xe->drm.dev->kobj, &dev_attr_vram_d3cold_threshold.attr); - if (xe->info.platform == XE_BATTLEMAGE) + if (xe->info.platform == XE_BATTLEMAGE) { sysfs_remove_files(&xe->drm.dev->kobj, auto_link_downgrade_attrs); + late_bind_remove_files(xe->drm.dev); + } } int xe_device_sysfs_init(struct xe_device *xe) @@ -170,6 +307,10 @@ int xe_device_sysfs_init(struct xe_device *xe) ret = sysfs_create_files(&dev->kobj, auto_link_downgrade_attrs); if (ret) return ret; + + ret = late_bind_create_files(dev); + if (ret) + return ret; } return devm_add_action_or_reset(dev, xe_device_sysfs_fini, xe); diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h index 003afb279a5e..d15b2793629e 100644 --- a/drivers/gpu/drm/xe/xe_device_types.h +++ b/drivers/gpu/drm/xe/xe_device_types.h @@ -32,7 +32,9 @@ struct dram_info; struct intel_display; +struct intel_dg_nvm_dev; struct xe_ggtt; +struct xe_i2c; struct xe_pat_ops; struct xe_pxp; @@ -105,7 +107,7 @@ struct xe_vram_region { void __iomem *mapping; /** @ttm: VRAM TTM manager */ struct xe_ttm_vram_mgr ttm; -#if IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR) +#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) /** @pagemap: Used to remap device memory as ZONE_DEVICE */ struct dev_pagemap pagemap; /** @@ -293,6 +295,8 @@ struct xe_device { u8 vram_flags; /** @info.tile_count: Number of tiles */ u8 tile_count; + /** @info.max_gt_per_tile: Number of GT IDs allocated to each tile */ + u8 max_gt_per_tile; /** @info.gt_count: Total number of GTs for entire device */ u8 gt_count; /** @info.vm_max_level: Max VM level */ @@ -316,6 +320,8 @@ struct xe_device { u8 has_fan_control:1; /** @info.has_flat_ccs: Whether flat CCS metadata is used */ u8 has_flat_ccs:1; + /** @info.has_gsc_nvm: Device has gsc non-volatile memory */ + u8 has_gsc_nvm:1; /** @info.has_heci_cscfi: device has heci cscfi */ u8 has_heci_cscfi:1; /** @info.has_heci_gscfi: device has heci gscfi */ @@ -549,6 +555,9 @@ struct xe_device { /** @heci_gsc: graphics security controller */ struct xe_heci_gsc heci_gsc; + /** @nvm: discrete graphics non-volatile memory */ + struct intel_dg_nvm_dev *nvm; + /** @oa: oa observation subsystem */ struct xe_oa oa; @@ -577,6 +586,9 @@ struct xe_device { /** @pmu: performance monitoring unit */ struct xe_pmu pmu; + /** @i2c: I2C host controller */ + struct xe_i2c *i2c; + /** @atomic_svm_timeslice_ms: Atomic SVM fault timeslice MS */ u32 atomic_svm_timeslice_ms; @@ -588,6 +600,14 @@ struct xe_device { u8 vm_inject_error_position; #endif +#if IS_ENABLED(CONFIG_TRACE_GPU_MEM) + /** + * @global_total_pages: global GPU page usage tracked for gpu_mem + * tracepoints + */ + atomic64_t global_total_pages; +#endif + /* private: */ #if IS_ENABLED(CONFIG_DRM_XE_DISPLAY) diff --git a/drivers/gpu/drm/xe/xe_drm_client.c b/drivers/gpu/drm/xe/xe_drm_client.c index 31f688e953d7..f931ff9b1ec0 100644 --- a/drivers/gpu/drm/xe/xe_drm_client.c +++ b/drivers/gpu/drm/xe/xe_drm_client.c @@ -167,7 +167,7 @@ void xe_drm_client_remove_bo(struct xe_bo *bo) static void bo_meminfo(struct xe_bo *bo, struct drm_memory_stats stats[TTM_NUM_MEM_TYPES]) { - u64 sz = bo->size; + u64 sz = xe_bo_size(bo); u32 mem_type = bo->ttm.resource->mem_type; xe_bo_assert_held(bo); diff --git a/drivers/gpu/drm/xe/xe_eu_stall.c b/drivers/gpu/drm/xe/xe_eu_stall.c index 96732613b4b7..af7916315ac6 100644 --- a/drivers/gpu/drm/xe/xe_eu_stall.c +++ b/drivers/gpu/drm/xe/xe_eu_stall.c @@ -258,11 +258,13 @@ static int set_prop_eu_stall_wait_num_reports(struct xe_device *xe, u64 value, static int set_prop_eu_stall_gt_id(struct xe_device *xe, u64 value, struct eu_stall_open_properties *props) { - if (value >= xe->info.gt_count) { + struct xe_gt *gt = xe_device_get_gt(xe, value); + + if (!gt) { drm_dbg(&xe->drm, "Invalid GT ID %llu for EU stall sampling\n", value); return -EINVAL; } - props->gt = xe_device_get_gt(xe, value); + props->gt = gt; return 0; } diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c index fee22358cc09..8991b4aed440 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.c +++ b/drivers/gpu/drm/xe/xe_exec_queue.c @@ -610,7 +610,7 @@ int xe_exec_queue_create_ioctl(struct drm_device *dev, void *data, if (XE_IOCTL_DBG(xe, err)) return -EFAULT; - if (XE_IOCTL_DBG(xe, eci[0].gt_id >= xe->info.gt_count)) + if (XE_IOCTL_DBG(xe, !xe_device_get_gt(xe, eci[0].gt_id))) return -EINVAL; if (args->flags & DRM_XE_EXEC_QUEUE_LOW_LATENCY_HINT) diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c index 7b11fa1356f0..29d4d3f51da1 100644 --- a/drivers/gpu/drm/xe/xe_ggtt.c +++ b/drivers/gpu/drm/xe/xe_ggtt.c @@ -238,6 +238,13 @@ int xe_ggtt_init_kunit(struct xe_ggtt *ggtt, u32 reserved, u32 size) } EXPORT_SYMBOL_IF_KUNIT(xe_ggtt_init_kunit); +static void dev_fini_ggtt(void *arg) +{ + struct xe_ggtt *ggtt = arg; + + drain_workqueue(ggtt->wq); +} + /** * xe_ggtt_init_early - Early GGTT initialization * @ggtt: the &xe_ggtt to be initialized @@ -290,6 +297,10 @@ int xe_ggtt_init_early(struct xe_ggtt *ggtt) if (err) return err; + err = devm_add_action_or_reset(xe->drm.dev, dev_fini_ggtt, ggtt); + if (err) + return err; + if (IS_SRIOV_VF(xe)) { err = xe_tile_sriov_vf_prepare_ggtt(ggtt->tile); if (err) @@ -410,7 +421,7 @@ int xe_ggtt_init(struct xe_ggtt *ggtt) goto err; } - xe_map_memset(xe, &ggtt->scratch->vmap, 0, 0, ggtt->scratch->size); + xe_map_memset(xe, &ggtt->scratch->vmap, 0, 0, xe_bo_size(ggtt->scratch)); xe_ggtt_initial_clear(ggtt); @@ -682,13 +693,13 @@ void xe_ggtt_map_bo(struct xe_ggtt *ggtt, struct xe_ggtt_node *node, return; start = node->base.start; - end = start + bo->size; + end = start + xe_bo_size(bo); pte = ggtt->pt_ops->pte_encode_flags(bo, pat_index); if (!xe_bo_is_vram(bo) && !xe_bo_is_stolen(bo)) { xe_assert(xe_bo_device(bo), bo->ttm.ttm); - for (xe_res_first_sg(xe_bo_sg(bo), 0, bo->size, &cur); + for (xe_res_first_sg(xe_bo_sg(bo), 0, xe_bo_size(bo), &cur); cur.remaining; xe_res_next(&cur, XE_PAGE_SIZE)) ggtt->pt_ops->ggtt_set_pte(ggtt, end - cur.remaining, pte | xe_res_dma(&cur)); @@ -696,7 +707,7 @@ void xe_ggtt_map_bo(struct xe_ggtt *ggtt, struct xe_ggtt_node *node, /* Prepend GPU offset */ pte |= vram_region_gpu_offset(bo->ttm.resource); - for (xe_res_first(bo->ttm.resource, 0, bo->size, &cur); + for (xe_res_first(bo->ttm.resource, 0, xe_bo_size(bo), &cur); cur.remaining; xe_res_next(&cur, XE_PAGE_SIZE)) ggtt->pt_ops->ggtt_set_pte(ggtt, end - cur.remaining, pte + cur.start); @@ -732,7 +743,7 @@ static int __xe_ggtt_insert_bo_at(struct xe_ggtt *ggtt, struct xe_bo *bo, if (XE_WARN_ON(bo->ggtt_node[tile_id])) { /* Someone's already inserted this BO in the GGTT */ - xe_tile_assert(ggtt->tile, bo->ggtt_node[tile_id]->base.size == bo->size); + xe_tile_assert(ggtt->tile, bo->ggtt_node[tile_id]->base.size == xe_bo_size(bo)); return 0; } @@ -751,7 +762,7 @@ static int __xe_ggtt_insert_bo_at(struct xe_ggtt *ggtt, struct xe_bo *bo, mutex_lock(&ggtt->lock); err = drm_mm_insert_node_in_range(&ggtt->mm, &bo->ggtt_node[tile_id]->base, - bo->size, alignment, 0, start, end, 0); + xe_bo_size(bo), alignment, 0, start, end, 0); if (err) { xe_ggtt_node_fini(bo->ggtt_node[tile_id]); bo->ggtt_node[tile_id] = NULL; @@ -812,7 +823,7 @@ void xe_ggtt_remove_bo(struct xe_ggtt *ggtt, struct xe_bo *bo) return; /* This BO is not currently in the GGTT */ - xe_tile_assert(ggtt->tile, bo->ggtt_node[tile_id]->base.size == bo->size); + xe_tile_assert(ggtt->tile, bo->ggtt_node[tile_id]->base.size == xe_bo_size(bo)); xe_ggtt_node_remove(bo->ggtt_node[tile_id], bo->flags & XE_BO_FLAG_GGTT_INVALIDATE); diff --git a/drivers/gpu/drm/xe/xe_gsc.c b/drivers/gpu/drm/xe/xe_gsc.c index 0bcf97063ff6..1d84bf2f2cef 100644 --- a/drivers/gpu/drm/xe/xe_gsc.c +++ b/drivers/gpu/drm/xe/xe_gsc.c @@ -59,7 +59,8 @@ static int memcpy_fw(struct xe_gsc *gsc) xe_map_memcpy_from(xe, storage, &gsc->fw.bo->vmap, 0, fw_size); xe_map_memcpy_to(xe, &gsc->private->vmap, 0, storage, fw_size); - xe_map_memset(xe, &gsc->private->vmap, fw_size, 0, gsc->private->size - fw_size); + xe_map_memset(xe, &gsc->private->vmap, fw_size, 0, + xe_bo_size(gsc->private) - fw_size); kfree(storage); @@ -82,7 +83,8 @@ static int emit_gsc_upload(struct xe_gsc *gsc) bb->cs[bb->len++] = GSC_FW_LOAD; bb->cs[bb->len++] = lower_32_bits(offset); bb->cs[bb->len++] = upper_32_bits(offset); - bb->cs[bb->len++] = (gsc->private->size / SZ_4K) | GSC_FW_LOAD_LIMIT_VALID; + bb->cs[bb->len++] = (xe_bo_size(gsc->private) / SZ_4K) | + GSC_FW_LOAD_LIMIT_VALID; job = xe_bb_create_job(gsc->q, bb); if (IS_ERR(job)) { diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c index 9752a38c0162..d397df056e4c 100644 --- a/drivers/gpu/drm/xe/xe_gt.c +++ b/drivers/gpu/drm/xe/xe_gt.c @@ -390,6 +390,7 @@ put_exec_queue: int xe_gt_init_early(struct xe_gt *gt) { + unsigned int fw_ref; int err; if (IS_SRIOV_PF(gt_to_xe(gt))) { @@ -419,6 +420,25 @@ int xe_gt_init_early(struct xe_gt *gt) xe_mocs_init_early(gt); + /* + * Only after this point can GT-specific MMIO operations + * (including things like communication with the GuC) + * be performed. + */ + xe_gt_mmio_init(gt); + + err = xe_uc_init_noalloc(>->uc); + if (err) + return err; + + fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); + if (!fw_ref) + return -ETIMEDOUT; + + xe_gt_mcr_init_early(gt); + xe_pat_init(gt); + xe_force_wake_put(gt_to_fw(gt), fw_ref); + return 0; } @@ -433,7 +453,7 @@ static void dump_pat_on_error(struct xe_gt *gt) xe_pat_dump(gt, &p); } -static int gt_fw_domain_init(struct xe_gt *gt) +static int gt_init_with_gt_forcewake(struct xe_gt *gt) { unsigned int fw_ref; int err; @@ -442,6 +462,14 @@ static int gt_fw_domain_init(struct xe_gt *gt) if (!fw_ref) return -ETIMEDOUT; + err = xe_uc_init(>->uc); + if (err) + goto err_force_wake; + + xe_gt_topology_init(gt); + xe_gt_mcr_init(gt); + xe_gt_enable_host_l2_vram(gt); + if (!xe_gt_is_media_type(gt)) { err = xe_ggtt_init(gt_to_tile(gt)->mem.ggtt); if (err) @@ -457,8 +485,10 @@ static int gt_fw_domain_init(struct xe_gt *gt) xe_gt_mcr_init(gt); err = xe_hw_engines_init_early(gt); - if (err) + if (err) { + dump_pat_on_error(gt); goto err_force_wake; + } err = xe_hw_engine_class_sysfs_init(gt); if (err) @@ -479,13 +509,12 @@ static int gt_fw_domain_init(struct xe_gt *gt) return 0; err_force_wake: - dump_pat_on_error(gt); xe_force_wake_put(gt_to_fw(gt), fw_ref); return err; } -static int all_fw_domain_init(struct xe_gt *gt) +static int gt_init_with_all_forcewake(struct xe_gt *gt) { unsigned int fw_ref; int err; @@ -544,7 +573,7 @@ static int all_fw_domain_init(struct xe_gt *gt) } } - err = xe_uc_init_hw(>->uc); + err = xe_uc_load_hw(>->uc); if (err) goto err_force_wake; @@ -572,39 +601,6 @@ err_force_wake: return err; } -/* - * Initialize enough GT to be able to load GuC in order to obtain hwconfig and - * enable CTB communication. - */ -int xe_gt_init_hwconfig(struct xe_gt *gt) -{ - unsigned int fw_ref; - int err; - - fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT); - if (!fw_ref) - return -ETIMEDOUT; - - xe_gt_mcr_init_early(gt); - xe_pat_init(gt); - - err = xe_uc_init(>->uc); - if (err) - goto out_fw; - - err = xe_uc_init_hwconfig(>->uc); - if (err) - goto out_fw; - - xe_gt_topology_init(gt); - xe_gt_mcr_init(gt); - xe_gt_enable_host_l2_vram(gt); - -out_fw: - xe_force_wake_put(gt_to_fw(gt), fw_ref); - return err; -} - static void xe_gt_fini(void *arg) { struct xe_gt *gt = arg; @@ -640,7 +636,7 @@ int xe_gt_init(struct xe_gt *gt) if (err) return err; - err = gt_fw_domain_init(gt); + err = gt_init_with_gt_forcewake(gt); if (err) return err; @@ -654,7 +650,7 @@ int xe_gt_init(struct xe_gt *gt) xe_force_wake_init_engines(gt, gt_to_fw(gt)); - err = all_fw_domain_init(gt); + err = gt_init_with_all_forcewake(gt); if (err) return err; @@ -742,7 +738,7 @@ static int vf_gt_restart(struct xe_gt *gt) if (err) return err; - err = xe_uc_init_hw(>->uc); + err = xe_uc_load_hw(>->uc); if (err) return err; @@ -780,7 +776,7 @@ static int do_gt_restart(struct xe_gt *gt) if (err) return err; - err = xe_uc_init_hw(>->uc); + err = xe_uc_load_hw(>->uc); if (err) return err; diff --git a/drivers/gpu/drm/xe/xe_gt.h b/drivers/gpu/drm/xe/xe_gt.h index 187fa6490eaf..8128ddfdd788 100644 --- a/drivers/gpu/drm/xe/xe_gt.h +++ b/drivers/gpu/drm/xe/xe_gt.h @@ -28,7 +28,6 @@ static inline bool xe_fault_inject_gt_reset(void) } struct xe_gt *xe_gt_alloc(struct xe_tile *tile); -int xe_gt_init_hwconfig(struct xe_gt *gt); int xe_gt_init_early(struct xe_gt *gt); int xe_gt_init(struct xe_gt *gt); void xe_gt_mmio_init(struct xe_gt *gt); diff --git a/drivers/gpu/drm/xe/xe_gt_mcr.c b/drivers/gpu/drm/xe/xe_gt_mcr.c index d4d9730f0d2c..64a2f0d6aaf9 100644 --- a/drivers/gpu/drm/xe/xe_gt_mcr.c +++ b/drivers/gpu/drm/xe/xe_gt_mcr.c @@ -420,12 +420,6 @@ static void init_steering_sqidi_psmi(struct xe_gt *gt) gt->steering[SQIDI_PSMI].instance_target = select & 0x1; } -static void init_steering_inst0(struct xe_gt *gt) -{ - gt->steering[INSTANCE0].group_target = 0; /* unused */ - gt->steering[INSTANCE0].instance_target = 0; /* unused */ -} - static const struct { const char *name; void (*init)(struct xe_gt *gt); @@ -436,7 +430,7 @@ static const struct { [DSS] = { "DSS", init_steering_dss }, [OADDRM] = { "OADDRM / GPMXMT", init_steering_oaddrm }, [SQIDI_PSMI] = { "SQIDI_PSMI", init_steering_sqidi_psmi }, - [INSTANCE0] = { "INSTANCE 0", init_steering_inst0 }, + [INSTANCE0] = { "INSTANCE 0", NULL }, [IMPLICIT_STEERING] = { "IMPLICIT", NULL }, }; @@ -446,25 +440,17 @@ static const struct { * * Perform early software only initialization of the MCR lock to allow * the synchronization on accessing the STEER_SEMAPHORE register and - * use the xe_gt_mcr_multicast_write() function. + * use the xe_gt_mcr_multicast_write() function, plus the minimum + * safe MCR registers required for VRAM/CCS probing. */ void xe_gt_mcr_init_early(struct xe_gt *gt) { + struct xe_device *xe = gt_to_xe(gt); + BUILD_BUG_ON(IMPLICIT_STEERING + 1 != NUM_STEERING_TYPES); BUILD_BUG_ON(ARRAY_SIZE(xe_steering_types) != NUM_STEERING_TYPES); spin_lock_init(>->mcr_lock); -} - -/** - * xe_gt_mcr_init - Normal initialization of the MCR support - * @gt: GT structure - * - * Perform normal initialization of the MCR for all usages. - */ -void xe_gt_mcr_init(struct xe_gt *gt) -{ - struct xe_device *xe = gt_to_xe(gt); if (IS_SRIOV_VF(xe)) return; @@ -505,10 +491,27 @@ void xe_gt_mcr_init(struct xe_gt *gt) } } + /* Mark instance 0 as initialized, we need this early for VRAM and CCS probe. */ + gt->steering[INSTANCE0].initialized = true; +} + +/** + * xe_gt_mcr_init - Normal initialization of the MCR support + * @gt: GT structure + * + * Perform normal initialization of the MCR for all usages. + */ +void xe_gt_mcr_init(struct xe_gt *gt) +{ + if (IS_SRIOV_VF(gt_to_xe(gt))) + return; + /* Select non-terminated steering target for each type */ - for (int i = 0; i < NUM_STEERING_TYPES; i++) + for (int i = 0; i < NUM_STEERING_TYPES; i++) { + gt->steering[i].initialized = true; if (gt->steering[i].ranges && xe_steering_types[i].init) xe_steering_types[i].init(gt); + } } /** @@ -570,6 +573,10 @@ bool xe_gt_mcr_get_nonterminated_steering(struct xe_gt *gt, for (int i = 0; gt->steering[type].ranges[i].end > 0; i++) { if (xe_mmio_in_range(>->mmio, >->steering[type].ranges[i], reg)) { + drm_WARN(>_to_xe(gt)->drm, !gt->steering[type].initialized, + "Uninitialized usage of MCR register %s/%#x\n", + xe_steering_types[type].name, reg.addr); + *group = gt->steering[type].group_target; *instance = gt->steering[type].instance_target; return true; diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c index 3522865c67c9..5a75d56d8558 100644 --- a/drivers/gpu/drm/xe/xe_gt_pagefault.c +++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c @@ -419,6 +419,7 @@ static int xe_alloc_pf_queue(struct xe_gt *gt, struct pf_queue *pf_queue) #define PF_MULTIPLIER 8 pf_queue->num_dw = (num_eus + XE_NUM_HW_ENGINES) * PF_MSG_LEN_DW * PF_MULTIPLIER; + pf_queue->num_dw = roundup_pow_of_two(pf_queue->num_dw); #undef PF_MULTIPLIER pf_queue->gt = gt; diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c index 3556c41c041b..d0cf1d80be07 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c @@ -104,13 +104,13 @@ static int pf_push_vf_buf_klvs(struct xe_gt *gt, unsigned int vfid, u32 num_klvs } if (IS_ENABLED(CONFIG_DRM_XE_DEBUG_SRIOV)) { - struct drm_printer p = xe_gt_info_printer(gt); + struct drm_printer p = xe_gt_dbg_printer(gt); void *klvs = xe_guc_buf_cpu_ptr(buf); char name[8]; - xe_gt_sriov_info(gt, "pushed %s config with %u KLV%s:\n", - xe_sriov_function_name(vfid, name, sizeof(name)), - num_klvs, str_plural(num_klvs)); + xe_gt_sriov_dbg(gt, "pushed %s config with %u KLV%s:\n", + xe_sriov_function_name(vfid, name, sizeof(name)), + num_klvs, str_plural(num_klvs)); xe_guc_klv_print(klvs, num_dwords, &p); } @@ -282,8 +282,8 @@ static u32 encode_config(u32 *cfg, const struct xe_gt_sriov_config *config, bool if (config->lmem_obj) { cfg[n++] = PREP_GUC_KLV_TAG(VF_CFG_LMEM_SIZE); - cfg[n++] = lower_32_bits(config->lmem_obj->size); - cfg[n++] = upper_32_bits(config->lmem_obj->size); + cfg[n++] = lower_32_bits(xe_bo_size(config->lmem_obj)); + cfg[n++] = upper_32_bits(xe_bo_size(config->lmem_obj)); } cfg[n++] = PREP_GUC_KLV_TAG(VF_CFG_EXEC_QUANTUM); @@ -1299,7 +1299,7 @@ static u64 pf_get_vf_config_lmem(struct xe_gt *gt, unsigned int vfid) struct xe_bo *bo; bo = config->lmem_obj; - return bo ? bo->size : 0; + return bo ? xe_bo_size(bo) : 0; } static int pf_distribute_config_lmem(struct xe_gt *gt, unsigned int vfid, u64 size) @@ -1388,7 +1388,7 @@ static int pf_update_vf_lmtt(struct xe_device *xe, unsigned int vfid) err = xe_lmtt_populate_pages(lmtt, vfid, bo, offset); if (err) goto fail; - offset += bo->size; + offset += xe_bo_size(bo); } } @@ -1469,12 +1469,12 @@ static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size) goto release; } - err = pf_push_vf_cfg_lmem(gt, vfid, bo->size); + err = pf_push_vf_cfg_lmem(gt, vfid, xe_bo_size(bo)); if (unlikely(err)) goto reset_lmtt; xe_gt_sriov_dbg_verbose(gt, "VF%u LMEM %zu (%zuM)\n", - vfid, bo->size, bo->size / SZ_1M); + vfid, xe_bo_size(bo), xe_bo_size(bo) / SZ_1M); return 0; reset_lmtt: @@ -2349,7 +2349,7 @@ int xe_gt_sriov_pf_config_restore(struct xe_gt *gt, unsigned int vfid, return -EINVAL; if (IS_ENABLED(CONFIG_DRM_XE_DEBUG_SRIOV)) { - struct drm_printer p = xe_gt_info_printer(gt); + struct drm_printer p = xe_gt_dbg_printer(gt); drm_printf(&p, "restoring VF%u config:\n", vfid); xe_guc_klv_print(buf, size / sizeof(u32), &p); @@ -2552,10 +2552,10 @@ int xe_gt_sriov_pf_config_print_lmem(struct xe_gt *gt, struct drm_printer *p) if (!config->lmem_obj) continue; - string_get_size(config->lmem_obj->size, 1, STRING_UNITS_2, + string_get_size(xe_bo_size(config->lmem_obj), 1, STRING_UNITS_2, buf, sizeof(buf)); drm_printf(p, "VF%u:\t%zu\t(%s)\n", - n, config->lmem_obj->size, buf); + n, xe_bo_size(config->lmem_obj), buf); } mutex_unlock(xe_gt_sriov_pf_master_mutex(gt)); diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c index 821cfcc34e6b..fa74b3e1a964 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_service.c @@ -266,7 +266,7 @@ static void pf_prepare_runtime_info(struct xe_gt *gt) read_many(gt, size, regs, values); if (IS_ENABLED(CONFIG_DRM_XE_DEBUG_SRIOV)) { - struct drm_printer p = xe_gt_info_printer(gt); + struct drm_printer p = xe_gt_dbg_printer(gt); xe_gt_sriov_pf_service_print_runtime(gt, &p); } diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c index 9b2fc9db55b8..93cd26dca070 100644 --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c @@ -966,7 +966,6 @@ u32 xe_gt_sriov_vf_read32(struct xe_gt *gt, struct xe_reg reg) struct vf_runtime_reg *rr; xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt))); - xe_gt_assert(gt, gt->sriov.vf.pf_version.major); xe_gt_assert(gt, !reg.vf); if (reg.addr == GMD_ID.addr) { diff --git a/drivers/gpu/drm/xe/xe_gt_types.h b/drivers/gpu/drm/xe/xe_gt_types.h index 7def0959da35..96344c604726 100644 --- a/drivers/gpu/drm/xe/xe_gt_types.h +++ b/drivers/gpu/drm/xe/xe_gt_types.h @@ -377,6 +377,8 @@ struct xe_gt { u16 group_target; /** @steering.instance_target: instance to steer accesses to */ u16 instance_target; + /** @steering.initialized: Whether this steering range is initialized */ + bool initialized; } steering[NUM_STEERING_TYPES]; /** diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c index 209e5d53c290..b1d1d6da3758 100644 --- a/drivers/gpu/drm/xe/xe_guc.c +++ b/drivers/gpu/drm/xe/xe_guc.c @@ -29,6 +29,7 @@ #include "xe_guc_db_mgr.h" #include "xe_guc_engine_activity.h" #include "xe_guc_hwconfig.h" +#include "xe_guc_klv_helpers.h" #include "xe_guc_log.h" #include "xe_guc_pc.h" #include "xe_guc_relay.h" @@ -59,7 +60,7 @@ static u32 guc_bo_ggtt_addr(struct xe_guc *guc, /* GuC addresses above GUC_GGTT_TOP don't map through the GTT */ xe_assert(xe, addr >= xe_wopcm_size(guc_to_xe(guc))); xe_assert(xe, addr < GUC_GGTT_TOP); - xe_assert(xe, bo->size <= GUC_GGTT_TOP - addr); + xe_assert(xe, xe_bo_size(bo) <= GUC_GGTT_TOP - addr); return addr; } @@ -420,7 +421,7 @@ static int guc_g2g_register(struct xe_guc *near_guc, struct xe_gt *far_gt, u32 t buf = base + G2G_DESC_AREA_SIZE + slot * G2G_BUFFER_SIZE; xe_assert(xe, (desc - base + G2G_DESC_SIZE) <= G2G_DESC_AREA_SIZE); - xe_assert(xe, (buf - base + G2G_BUFFER_SIZE) <= g2g_bo->size); + xe_assert(xe, (buf - base + G2G_BUFFER_SIZE) <= xe_bo_size(g2g_bo)); return guc_action_register_g2g_buffer(near_guc, type, far_tile, far_dev, desc, buf, G2G_BUFFER_SIZE); @@ -570,6 +571,86 @@ err_deregister: return err; } +static int __guc_opt_in_features_enable(struct xe_guc *guc, u64 addr, u32 num_dwords) +{ + u32 action[] = { + XE_GUC_ACTION_OPT_IN_FEATURE_KLV, + lower_32_bits(addr), + upper_32_bits(addr), + num_dwords + }; + + return xe_guc_ct_send_block(&guc->ct, action, ARRAY_SIZE(action)); +} + +static bool supports_dynamic_ics(struct xe_guc *guc) +{ + struct xe_device *xe = guc_to_xe(guc); + struct xe_gt *gt = guc_to_gt(guc); + + /* Dynamic ICS is available for PVC and Xe2 and newer platforms. */ + if (xe->info.platform != XE_PVC && GRAPHICS_VER(xe) < 20) + return false; + + /* + * The feature is currently not compatible with multi-lrc, so the GuC + * does not support it at all on the media engines (which are the main + * users of mlrc). On the primary GT side, to avoid it being used in + * conjunction with mlrc, we only enable it if we are in single CCS + * mode. + */ + if (xe_gt_is_media_type(gt) || gt->ccs_mode > 1) + return false; + + /* + * Dynamic ICS requires GuC v70.40.1, which maps to compatibility + * version v1.18.4. + */ + return GUC_SUBMIT_VER(guc) >= MAKE_GUC_VER(1, 18, 4); +} + +#define OPT_IN_MAX_DWORDS 16 +int xe_guc_opt_in_features_enable(struct xe_guc *guc) +{ + struct xe_device *xe = guc_to_xe(guc); + CLASS(xe_guc_buf, buf)(&guc->buf, OPT_IN_MAX_DWORDS); + u32 count = 0; + u32 *klvs; + int ret; + + if (!xe_guc_buf_is_valid(buf)) + return -ENOBUFS; + + klvs = xe_guc_buf_cpu_ptr(buf); + + /* + * The extra CAT error type opt-in was added in GuC v70.17.0, which maps + * to compatibility version v1.7.0. + * Note that the GuC allows enabling this KLV even on platforms that do + * not support the extra type; in such case the returned type variable + * will be set to a known invalid value which we can check against. + */ + if (GUC_SUBMIT_VER(guc) >= MAKE_GUC_VER(1, 7, 0)) + klvs[count++] = PREP_GUC_KLV_TAG(OPT_IN_FEATURE_EXT_CAT_ERR_TYPE); + + if (supports_dynamic_ics(guc)) + klvs[count++] = PREP_GUC_KLV_TAG(OPT_IN_FEATURE_DYNAMIC_INHIBIT_CONTEXT_SWITCH); + + if (count) { + xe_assert(xe, count <= OPT_IN_MAX_DWORDS); + + ret = __guc_opt_in_features_enable(guc, xe_guc_buf_flush(buf), count); + if (ret < 0) { + xe_gt_err(guc_to_gt(guc), + "failed to enable GuC opt-in features: %pe\n", + ERR_PTR(ret)); + return ret; + } + } + + return 0; +} + static void guc_fini_hw(void *arg) { struct xe_guc *guc = arg; @@ -577,7 +658,7 @@ static void guc_fini_hw(void *arg) unsigned int fw_ref; fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL); - xe_uc_fini_hw(&guc_to_gt(guc)->uc); + xe_uc_sanitize_reset(&guc_to_gt(guc)->uc); xe_force_wake_put(gt_to_fw(gt), fw_ref); guc_g2g_fini(guc); @@ -627,23 +708,51 @@ static int xe_guc_realloc_post_hwconfig(struct xe_guc *guc) return 0; } -static int vf_guc_init(struct xe_guc *guc) +static int vf_guc_init_noalloc(struct xe_guc *guc) { + struct xe_gt *gt = guc_to_gt(guc); int err; - xe_guc_comm_init_early(guc); - - err = xe_guc_ct_init(&guc->ct); + err = xe_gt_sriov_vf_bootstrap(gt); if (err) return err; - err = xe_guc_relay_init(&guc->relay); + err = xe_gt_sriov_vf_query_config(gt); if (err) return err; return 0; } +int xe_guc_init_noalloc(struct xe_guc *guc) +{ + struct xe_device *xe = guc_to_xe(guc); + struct xe_gt *gt = guc_to_gt(guc); + int ret; + + xe_guc_comm_init_early(guc); + + ret = xe_guc_ct_init_noalloc(&guc->ct); + if (ret) + goto out; + + ret = xe_guc_relay_init(&guc->relay); + if (ret) + goto out; + + if (IS_SRIOV_VF(xe)) { + ret = vf_guc_init_noalloc(guc); + if (ret) + goto out; + } + + return 0; + +out: + xe_gt_err(gt, "GuC init failed with %pe\n", ERR_PTR(ret)); + return ret; +} + int xe_guc_init(struct xe_guc *guc) { struct xe_device *xe = guc_to_xe(guc); @@ -653,13 +762,13 @@ int xe_guc_init(struct xe_guc *guc) guc->fw.type = XE_UC_FW_TYPE_GUC; ret = xe_uc_fw_init(&guc->fw); if (ret) - goto out; + return ret; if (!xe_uc_fw_is_enabled(&guc->fw)) return 0; if (IS_SRIOV_VF(xe)) { - ret = vf_guc_init(guc); + ret = xe_guc_ct_init(&guc->ct); if (ret) goto out; return 0; @@ -681,10 +790,6 @@ int xe_guc_init(struct xe_guc *guc) if (ret) goto out; - ret = xe_guc_relay_init(&guc->relay); - if (ret) - goto out; - xe_uc_fw_change_status(&guc->fw, XE_UC_FIRMWARE_LOADABLE); ret = devm_add_action_or_reset(xe->drm.dev, guc_fini_hw, guc); @@ -693,8 +798,6 @@ int xe_guc_init(struct xe_guc *guc) guc_init_params(guc); - xe_guc_comm_init_early(guc); - return 0; out: @@ -767,6 +870,10 @@ int xe_guc_post_load_init(struct xe_guc *guc) xe_guc_ads_populate_post_load(&guc->ads); + ret = xe_guc_opt_in_features_enable(guc); + if (ret) + return ret; + if (xe_guc_g2g_wanted(guc_to_xe(guc))) { ret = guc_g2g_start(guc); if (ret) @@ -1112,13 +1219,17 @@ static int vf_guc_min_load_for_hwconfig(struct xe_guc *guc) ret = xe_gt_sriov_vf_connect(gt); if (ret) - return ret; + goto err_out; ret = xe_gt_sriov_vf_query_runtime(gt); if (ret) - return ret; + goto err_out; return 0; + +err_out: + xe_guc_sanitize(guc); + return ret; } /** diff --git a/drivers/gpu/drm/xe/xe_guc.h b/drivers/gpu/drm/xe/xe_guc.h index 58338be44558..22cf019a11bf 100644 --- a/drivers/gpu/drm/xe/xe_guc.h +++ b/drivers/gpu/drm/xe/xe_guc.h @@ -26,6 +26,7 @@ struct drm_printer; void xe_guc_comm_init_early(struct xe_guc *guc); +int xe_guc_init_noalloc(struct xe_guc *guc); int xe_guc_init(struct xe_guc *guc); int xe_guc_init_post_hwconfig(struct xe_guc *guc); int xe_guc_post_load_init(struct xe_guc *guc); @@ -33,6 +34,7 @@ int xe_guc_reset(struct xe_guc *guc); int xe_guc_upload(struct xe_guc *guc); int xe_guc_min_load_for_hwconfig(struct xe_guc *guc); int xe_guc_enable_communication(struct xe_guc *guc); +int xe_guc_opt_in_features_enable(struct xe_guc *guc); int xe_guc_suspend(struct xe_guc *guc); void xe_guc_notify(struct xe_guc *guc); int xe_guc_auth_huc(struct xe_guc *guc, u32 rsa_addr); diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c index 07a027755627..131cfc56be00 100644 --- a/drivers/gpu/drm/xe/xe_guc_ads.c +++ b/drivers/gpu/drm/xe/xe_guc_ads.c @@ -890,7 +890,7 @@ void xe_guc_ads_populate_minimal(struct xe_guc_ads *ads) xe_gt_assert(gt, ads->bo); - xe_map_memset(ads_to_xe(ads), ads_to_map(ads), 0, 0, ads->bo->size); + xe_map_memset(ads_to_xe(ads), ads_to_map(ads), 0, 0, xe_bo_size(ads->bo)); guc_policies_init(ads); guc_golden_lrc_init(ads); guc_mapping_table_init_invalid(gt, &info_map); @@ -914,7 +914,7 @@ void xe_guc_ads_populate(struct xe_guc_ads *ads) xe_gt_assert(gt, ads->bo); - xe_map_memset(ads_to_xe(ads), ads_to_map(ads), 0, 0, ads->bo->size); + xe_map_memset(ads_to_xe(ads), ads_to_map(ads), 0, 0, xe_bo_size(ads->bo)); guc_policies_init(ads); fill_engine_enable_masks(gt, &info_map); guc_mmio_reg_state_init(ads); @@ -995,16 +995,6 @@ static int guc_ads_action_update_policies(struct xe_guc_ads *ads, u32 policy_off return xe_guc_ct_send(ct, action, ARRAY_SIZE(action), 0, 0); } -static int guc_ads_update_policies(struct xe_guc_ads *ads, const struct guc_policies *policies) -{ - CLASS(xe_guc_buf_from_data, buf)(&ads_to_guc(ads)->buf, policies, sizeof(*policies)); - - if (!xe_guc_buf_is_valid(buf)) - return -ENOBUFS; - - return guc_ads_action_update_policies(ads, xe_guc_buf_flush(buf)); -} - /** * xe_guc_ads_scheduler_policy_toggle_reset - Toggle reset policy * @ads: Additional data structures object @@ -1015,13 +1005,16 @@ static int guc_ads_update_policies(struct xe_guc_ads *ads, const struct guc_poli */ int xe_guc_ads_scheduler_policy_toggle_reset(struct xe_guc_ads *ads) { - struct xe_device *xe = ads_to_xe(ads); struct guc_policies *policies; - int ret; + struct xe_guc *guc = ads_to_guc(ads); + struct xe_device *xe = ads_to_xe(ads); + CLASS(xe_guc_buf, buf)(&guc->buf, sizeof(*policies)); - policies = kmalloc(sizeof(*policies), GFP_KERNEL); - if (!policies) - return -ENOMEM; + if (!xe_guc_buf_is_valid(buf)) + return -ENOBUFS; + + policies = xe_guc_buf_cpu_ptr(buf); + memset(policies, 0, sizeof(*policies)); policies->dpc_promote_time = ads_blob_read(ads, policies.dpc_promote_time); policies->max_num_work_items = ads_blob_read(ads, policies.max_num_work_items); @@ -1031,7 +1024,5 @@ int xe_guc_ads_scheduler_policy_toggle_reset(struct xe_guc_ads *ads) else policies->global_flags &= ~GLOBAL_POLICY_DISABLE_ENGINE_RESET; - ret = guc_ads_update_policies(ads, policies); - kfree(policies); - return ret; + return guc_ads_action_update_policies(ads, xe_guc_buf_flush(buf)); } diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c index 37509f619503..b6acccfcd351 100644 --- a/drivers/gpu/drm/xe/xe_guc_ct.c +++ b/drivers/gpu/drm/xe/xe_guc_ct.c @@ -35,6 +35,11 @@ #include "xe_pm.h" #include "xe_trace_guc.h" +static void receive_g2h(struct xe_guc_ct *ct); +static void g2h_worker_func(struct work_struct *w); +static void safe_mode_worker_func(struct work_struct *w); +static void ct_exit_safe_mode(struct xe_guc_ct *ct); + #if IS_ENABLED(CONFIG_DRM_XE_DEBUG) enum { /* Internal states, not error conditions */ @@ -80,6 +85,7 @@ struct g2h_fence { u16 error; u16 hint; u16 reason; + bool cancel; bool retry; bool fail; bool done; @@ -98,6 +104,13 @@ static void g2h_fence_init(struct g2h_fence *g2h_fence, u32 *response_buffer) g2h_fence->seqno = ~0x0; } +static void g2h_fence_cancel(struct g2h_fence *g2h_fence) +{ + g2h_fence->cancel = true; + g2h_fence->fail = true; + g2h_fence->done = true; +} + static bool g2h_fence_needs_alloc(struct g2h_fence *g2h_fence) { return g2h_fence->seqno == ~0x0; @@ -189,14 +202,11 @@ static void guc_ct_fini(struct drm_device *drm, void *arg) { struct xe_guc_ct *ct = arg; + ct_exit_safe_mode(ct); destroy_workqueue(ct->g2h_wq); xa_destroy(&ct->fence_lookup); } -static void receive_g2h(struct xe_guc_ct *ct); -static void g2h_worker_func(struct work_struct *w); -static void safe_mode_worker_func(struct work_struct *w); - static void primelockdep(struct xe_guc_ct *ct) { if (!IS_ENABLED(CONFIG_LOCKDEP)) @@ -207,12 +217,10 @@ static void primelockdep(struct xe_guc_ct *ct) fs_reclaim_release(GFP_KERNEL); } -int xe_guc_ct_init(struct xe_guc_ct *ct) +int xe_guc_ct_init_noalloc(struct xe_guc_ct *ct) { struct xe_device *xe = ct_to_xe(ct); struct xe_gt *gt = ct_to_gt(ct); - struct xe_tile *tile = gt_to_tile(gt); - struct xe_bo *bo; int err; xe_gt_assert(gt, !(guc_ct_size() % PAGE_SIZE)); @@ -238,6 +246,23 @@ int xe_guc_ct_init(struct xe_guc_ct *ct) primelockdep(ct); + err = drmm_add_action_or_reset(&xe->drm, guc_ct_fini, ct); + if (err) + return err; + + xe_gt_assert(gt, ct->state == XE_GUC_CT_STATE_NOT_INITIALIZED); + ct->state = XE_GUC_CT_STATE_DISABLED; + return 0; +} +ALLOW_ERROR_INJECTION(xe_guc_ct_init_noalloc, ERRNO); /* See xe_pci_probe() */ + +int xe_guc_ct_init(struct xe_guc_ct *ct) +{ + struct xe_device *xe = ct_to_xe(ct); + struct xe_gt *gt = ct_to_gt(ct); + struct xe_tile *tile = gt_to_tile(gt); + struct xe_bo *bo; + bo = xe_managed_bo_create_pin_map(xe, tile, guc_ct_size(), XE_BO_FLAG_SYSTEM | XE_BO_FLAG_GGTT | @@ -247,13 +272,6 @@ int xe_guc_ct_init(struct xe_guc_ct *ct) return PTR_ERR(bo); ct->bo = bo; - - err = drmm_add_action_or_reset(&xe->drm, guc_ct_fini, ct); - if (err) - return err; - - xe_gt_assert(gt, ct->state == XE_GUC_CT_STATE_NOT_INITIALIZED); - ct->state = XE_GUC_CT_STATE_DISABLED; return 0; } ALLOW_ERROR_INJECTION(xe_guc_ct_init, ERRNO); /* See xe_pci_probe() */ @@ -374,9 +392,13 @@ static int guc_ct_control_toggle(struct xe_guc_ct *ct, bool enable) return ret > 0 ? -EPROTO : ret; } -static void xe_guc_ct_set_state(struct xe_guc_ct *ct, +static void guc_ct_change_state(struct xe_guc_ct *ct, enum xe_guc_ct_state state) { + struct xe_gt *gt = ct_to_gt(ct); + struct g2h_fence *g2h_fence; + unsigned long idx; + mutex_lock(&ct->lock); /* Serialise dequeue_one_g2h() */ spin_lock_irq(&ct->fast_lock); /* Serialise CT fast-path */ @@ -388,8 +410,20 @@ static void xe_guc_ct_set_state(struct xe_guc_ct *ct, ct->g2h_outstanding = 0; ct->state = state; + xe_gt_dbg(gt, "GuC CT communication channel %s\n", + state == XE_GUC_CT_STATE_STOPPED ? "stopped" : + str_enabled_disabled(state == XE_GUC_CT_STATE_ENABLED)); + spin_unlock_irq(&ct->fast_lock); + /* cancel all in-flight send-recv requests */ + xa_for_each(&ct->fence_lookup, idx, g2h_fence) + g2h_fence_cancel(g2h_fence); + + /* make sure guc_ct_send_recv() will see g2h_fence changes */ + smp_mb(); + wake_up_all(&ct->g2h_fence_wq); + /* * Lockdep doesn't like this under the fast lock and he destroy only * needs to be serialized with the send path which ct lock provides. @@ -443,7 +477,7 @@ int xe_guc_ct_enable(struct xe_guc_ct *ct) xe_gt_assert(gt, !xe_guc_ct_enabled(ct)); - xe_map_memset(xe, &ct->bo->vmap, 0, 0, ct->bo->size); + xe_map_memset(xe, &ct->bo->vmap, 0, 0, xe_bo_size(ct->bo)); guc_ct_ctb_h2g_init(xe, &ct->ctbs.h2g, &ct->bo->vmap); guc_ct_ctb_g2h_init(xe, &ct->ctbs.g2h, &ct->bo->vmap); @@ -459,11 +493,10 @@ int xe_guc_ct_enable(struct xe_guc_ct *ct) if (err) goto err_out; - xe_guc_ct_set_state(ct, XE_GUC_CT_STATE_ENABLED); + guc_ct_change_state(ct, XE_GUC_CT_STATE_ENABLED); smp_mb(); wake_up_all(&ct->wq); - xe_gt_dbg(gt, "GuC CT communication channel enabled\n"); if (ct_needs_safe_mode(ct)) ct_enter_safe_mode(ct); @@ -504,7 +537,7 @@ static void stop_g2h_handler(struct xe_guc_ct *ct) */ void xe_guc_ct_disable(struct xe_guc_ct *ct) { - xe_guc_ct_set_state(ct, XE_GUC_CT_STATE_DISABLED); + guc_ct_change_state(ct, XE_GUC_CT_STATE_DISABLED); ct_exit_safe_mode(ct); stop_g2h_handler(ct); } @@ -520,7 +553,7 @@ void xe_guc_ct_stop(struct xe_guc_ct *ct) if (!xe_guc_ct_initialized(ct)) return; - xe_guc_ct_set_state(ct, XE_GUC_CT_STATE_STOPPED); + guc_ct_change_state(ct, XE_GUC_CT_STATE_STOPPED); stop_g2h_handler(ct); } @@ -1083,6 +1116,11 @@ retry_same_fence: goto retry; } if (g2h_fence.fail) { + if (g2h_fence.cancel) { + xe_gt_dbg(gt, "H2G request %#x canceled!\n", action[0]); + ret = -ECANCELED; + goto unlock; + } xe_gt_err(gt, "H2G request %#x failed: error %#x hint %#x\n", action[0], g2h_fence.error, g2h_fence.hint); ret = -EIO; @@ -1091,6 +1129,7 @@ retry_same_fence: if (ret > 0) ret = response_buffer ? g2h_fence.response_len : g2h_fence.response_data; +unlock: mutex_unlock(&ct->lock); return ret; @@ -1897,7 +1936,7 @@ static struct xe_guc_ct_snapshot *guc_ct_snapshot_alloc(struct xe_guc_ct *ct, bo return NULL; if (ct->bo && want_ctb) { - snapshot->ctb_size = ct->bo->size; + snapshot->ctb_size = xe_bo_size(ct->bo); snapshot->ctb = kmalloc(snapshot->ctb_size, atomic ? GFP_ATOMIC : GFP_KERNEL); } diff --git a/drivers/gpu/drm/xe/xe_guc_ct.h b/drivers/gpu/drm/xe/xe_guc_ct.h index 99c5dec446f2..18d4225e6502 100644 --- a/drivers/gpu/drm/xe/xe_guc_ct.h +++ b/drivers/gpu/drm/xe/xe_guc_ct.h @@ -11,6 +11,7 @@ struct drm_printer; struct xe_device; +int xe_guc_ct_init_noalloc(struct xe_guc_ct *ct); int xe_guc_ct_init(struct xe_guc_ct *ct); int xe_guc_ct_enable(struct xe_guc_ct *ct); void xe_guc_ct_disable(struct xe_guc_ct *ct); diff --git a/drivers/gpu/drm/xe/xe_guc_log.c b/drivers/gpu/drm/xe/xe_guc_log.c index 38039c411387..c01ccb35dc75 100644 --- a/drivers/gpu/drm/xe/xe_guc_log.c +++ b/drivers/gpu/drm/xe/xe_guc_log.c @@ -79,7 +79,7 @@ static struct xe_guc_log_snapshot *xe_guc_log_snapshot_alloc(struct xe_guc_log * * Also, can't use vmalloc as might be called from atomic context. So need * to break the buffer up into smaller chunks that can be allocated. */ - snapshot->size = log->bo->size; + snapshot->size = xe_bo_size(log->bo); snapshot->num_chunks = DIV_ROUND_UP(snapshot->size, GUC_LOG_CHUNK_SIZE); snapshot->copy = kcalloc(snapshot->num_chunks, sizeof(*snapshot->copy), diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c index 9fab5f5b10fa..68b192fe3b32 100644 --- a/drivers/gpu/drm/xe/xe_guc_pc.c +++ b/drivers/gpu/drm/xe/xe_guc_pc.c @@ -5,8 +5,11 @@ #include "xe_guc_pc.h" +#include <linux/cleanup.h> #include <linux/delay.h> +#include <linux/jiffies.h> #include <linux/ktime.h> +#include <linux/wait_bit.h> #include <drm/drm_managed.h> #include <drm/drm_print.h> @@ -52,9 +55,11 @@ #define LNL_MERT_FREQ_CAP 800 #define BMG_MERT_FREQ_CAP 2133 #define BMG_MIN_FREQ 1200 +#define BMG_MERT_FLUSH_FREQ_CAP 2600 #define SLPC_RESET_TIMEOUT_MS 5 /* roughly 5ms, but no need for precision */ #define SLPC_RESET_EXTENDED_TIMEOUT_MS 1000 /* To be used only at pc_start */ +#define SLPC_ACT_FREQ_TIMEOUT_MS 100 /** * DOC: GuC Power Conservation (PC) @@ -142,6 +147,36 @@ static int wait_for_pc_state(struct xe_guc_pc *pc, return -ETIMEDOUT; } +static int wait_for_flush_complete(struct xe_guc_pc *pc) +{ + const unsigned long timeout = msecs_to_jiffies(30); + + if (!wait_var_event_timeout(&pc->flush_freq_limit, + !atomic_read(&pc->flush_freq_limit), + timeout)) + return -ETIMEDOUT; + + return 0; +} + +static int wait_for_act_freq_limit(struct xe_guc_pc *pc, u32 freq) +{ + int timeout_us = SLPC_ACT_FREQ_TIMEOUT_MS * USEC_PER_MSEC; + int slept, wait = 10; + + for (slept = 0; slept < timeout_us;) { + if (xe_guc_pc_get_act_freq(pc) <= freq) + return 0; + + usleep_range(wait, wait << 1); + slept += wait; + wait <<= 1; + if (slept + wait > timeout_us) + wait = timeout_us - slept; + } + + return -ETIMEDOUT; +} static int pc_action_reset(struct xe_guc_pc *pc) { struct xe_guc_ct *ct = pc_to_ct(pc); @@ -554,6 +589,25 @@ u32 xe_guc_pc_get_rpn_freq(struct xe_guc_pc *pc) return pc->rpn_freq; } +static int xe_guc_pc_get_min_freq_locked(struct xe_guc_pc *pc, u32 *freq) +{ + int ret; + + lockdep_assert_held(&pc->freq_lock); + + /* Might be in the middle of a gt reset */ + if (!pc->freq_ready) + return -EAGAIN; + + ret = pc_action_query_task_state(pc); + if (ret) + return ret; + + *freq = pc_get_min_freq(pc); + + return 0; +} + /** * xe_guc_pc_get_min_freq - Get the min operational frequency * @pc: The GuC PC @@ -564,26 +618,28 @@ u32 xe_guc_pc_get_rpn_freq(struct xe_guc_pc *pc) */ int xe_guc_pc_get_min_freq(struct xe_guc_pc *pc, u32 *freq) { + guard(mutex)(&pc->freq_lock); + + return xe_guc_pc_get_min_freq_locked(pc, freq); +} + +static int xe_guc_pc_set_min_freq_locked(struct xe_guc_pc *pc, u32 freq) +{ int ret; - xe_device_assert_mem_access(pc_to_xe(pc)); + lockdep_assert_held(&pc->freq_lock); - mutex_lock(&pc->freq_lock); - if (!pc->freq_ready) { - /* Might be in the middle of a gt reset */ - ret = -EAGAIN; - goto out; - } + /* Might be in the middle of a gt reset */ + if (!pc->freq_ready) + return -EAGAIN; - ret = pc_action_query_task_state(pc); + ret = pc_set_min_freq(pc, freq); if (ret) - goto out; + return ret; - *freq = pc_get_min_freq(pc); + pc->user_requested_min = freq; -out: - mutex_unlock(&pc->freq_lock); - return ret; + return 0; } /** @@ -597,24 +653,28 @@ out: */ int xe_guc_pc_set_min_freq(struct xe_guc_pc *pc, u32 freq) { + guard(mutex)(&pc->freq_lock); + + return xe_guc_pc_set_min_freq_locked(pc, freq); +} + +static int xe_guc_pc_get_max_freq_locked(struct xe_guc_pc *pc, u32 *freq) +{ int ret; - mutex_lock(&pc->freq_lock); - if (!pc->freq_ready) { - /* Might be in the middle of a gt reset */ - ret = -EAGAIN; - goto out; - } + lockdep_assert_held(&pc->freq_lock); - ret = pc_set_min_freq(pc, freq); + /* Might be in the middle of a gt reset */ + if (!pc->freq_ready) + return -EAGAIN; + + ret = pc_action_query_task_state(pc); if (ret) - goto out; + return ret; - pc->user_requested_min = freq; + *freq = pc_get_max_freq(pc); -out: - mutex_unlock(&pc->freq_lock); - return ret; + return 0; } /** @@ -627,24 +687,28 @@ out: */ int xe_guc_pc_get_max_freq(struct xe_guc_pc *pc, u32 *freq) { + guard(mutex)(&pc->freq_lock); + + return xe_guc_pc_get_max_freq_locked(pc, freq); +} + +static int xe_guc_pc_set_max_freq_locked(struct xe_guc_pc *pc, u32 freq) +{ int ret; - mutex_lock(&pc->freq_lock); - if (!pc->freq_ready) { - /* Might be in the middle of a gt reset */ - ret = -EAGAIN; - goto out; - } + lockdep_assert_held(&pc->freq_lock); - ret = pc_action_query_task_state(pc); + /* Might be in the middle of a gt reset */ + if (!pc->freq_ready) + return -EAGAIN; + + ret = pc_set_max_freq(pc, freq); if (ret) - goto out; + return ret; - *freq = pc_get_max_freq(pc); + pc->user_requested_max = freq; -out: - mutex_unlock(&pc->freq_lock); - return ret; + return 0; } /** @@ -658,24 +722,14 @@ out: */ int xe_guc_pc_set_max_freq(struct xe_guc_pc *pc, u32 freq) { - int ret; - - mutex_lock(&pc->freq_lock); - if (!pc->freq_ready) { - /* Might be in the middle of a gt reset */ - ret = -EAGAIN; - goto out; + if (XE_WA(pc_to_gt(pc), 22019338487)) { + if (wait_for_flush_complete(pc) != 0) + return -EAGAIN; } - ret = pc_set_max_freq(pc, freq); - if (ret) - goto out; + guard(mutex)(&pc->freq_lock); - pc->user_requested_max = freq; - -out: - mutex_unlock(&pc->freq_lock); - return ret; + return xe_guc_pc_set_max_freq_locked(pc, freq); } /** @@ -873,30 +927,117 @@ static int pc_adjust_requested_freq(struct xe_guc_pc *pc) return ret; } -static int pc_set_mert_freq_cap(struct xe_guc_pc *pc) +static bool needs_flush_freq_limit(struct xe_guc_pc *pc) { - int ret = 0; + struct xe_gt *gt = pc_to_gt(pc); - if (XE_WA(pc_to_gt(pc), 22019338487)) { - /* - * Get updated min/max and stash them. - */ - ret = xe_guc_pc_get_min_freq(pc, &pc->stashed_min_freq); - if (!ret) - ret = xe_guc_pc_get_max_freq(pc, &pc->stashed_max_freq); - if (ret) - return ret; + return XE_WA(gt, 22019338487) && + pc->rp0_freq > BMG_MERT_FLUSH_FREQ_CAP; +} + +/** + * xe_guc_pc_apply_flush_freq_limit() - Limit max GT freq during L2 flush + * @pc: the xe_guc_pc object + * + * As per the WA, reduce max GT frequency during L2 cache flush + */ +void xe_guc_pc_apply_flush_freq_limit(struct xe_guc_pc *pc) +{ + struct xe_gt *gt = pc_to_gt(pc); + u32 max_freq; + int ret; + + if (!needs_flush_freq_limit(pc)) + return; + + guard(mutex)(&pc->freq_lock); + + ret = xe_guc_pc_get_max_freq_locked(pc, &max_freq); + if (!ret && max_freq > BMG_MERT_FLUSH_FREQ_CAP) { + ret = pc_set_max_freq(pc, BMG_MERT_FLUSH_FREQ_CAP); + if (ret) { + xe_gt_err_once(gt, "Failed to cap max freq on flush to %u, %pe\n", + BMG_MERT_FLUSH_FREQ_CAP, ERR_PTR(ret)); + return; + } + + atomic_set(&pc->flush_freq_limit, 1); /* - * Ensure min and max are bound by MERT_FREQ_CAP until driver loads. + * If user has previously changed max freq, stash that value to + * restore later, otherwise use the current max. New user + * requests wait on flush. */ - mutex_lock(&pc->freq_lock); - ret = pc_set_min_freq(pc, min(pc->rpe_freq, pc_max_freq_cap(pc))); - if (!ret) - ret = pc_set_max_freq(pc, min(pc->rp0_freq, pc_max_freq_cap(pc))); - mutex_unlock(&pc->freq_lock); + if (pc->user_requested_max != 0) + pc->stashed_max_freq = pc->user_requested_max; + else + pc->stashed_max_freq = max_freq; } + /* + * Wait for actual freq to go below the flush cap: even if the previous + * max was below cap, the current one might still be above it + */ + ret = wait_for_act_freq_limit(pc, BMG_MERT_FLUSH_FREQ_CAP); + if (ret) + xe_gt_err_once(gt, "Actual freq did not reduce to %u, %pe\n", + BMG_MERT_FLUSH_FREQ_CAP, ERR_PTR(ret)); +} + +/** + * xe_guc_pc_remove_flush_freq_limit() - Remove max GT freq limit after L2 flush completes. + * @pc: the xe_guc_pc object + * + * Retrieve the previous GT max frequency value. + */ +void xe_guc_pc_remove_flush_freq_limit(struct xe_guc_pc *pc) +{ + struct xe_gt *gt = pc_to_gt(pc); + int ret = 0; + + if (!needs_flush_freq_limit(pc)) + return; + + if (!atomic_read(&pc->flush_freq_limit)) + return; + + mutex_lock(&pc->freq_lock); + + ret = pc_set_max_freq(>->uc.guc.pc, pc->stashed_max_freq); + if (ret) + xe_gt_err_once(gt, "Failed to restore max freq %u:%d", + pc->stashed_max_freq, ret); + + atomic_set(&pc->flush_freq_limit, 0); + mutex_unlock(&pc->freq_lock); + wake_up_var(&pc->flush_freq_limit); +} + +static int pc_set_mert_freq_cap(struct xe_guc_pc *pc) +{ + int ret; + + if (!XE_WA(pc_to_gt(pc), 22019338487)) + return 0; + + guard(mutex)(&pc->freq_lock); + + /* + * Get updated min/max and stash them. + */ + ret = xe_guc_pc_get_min_freq_locked(pc, &pc->stashed_min_freq); + if (!ret) + ret = xe_guc_pc_get_max_freq_locked(pc, &pc->stashed_max_freq); + if (ret) + return ret; + + /* + * Ensure min and max are bound by MERT_FREQ_CAP until driver loads. + */ + ret = pc_set_min_freq(pc, min(pc->rpe_freq, pc_max_freq_cap(pc))); + if (!ret) + ret = pc_set_max_freq(pc, min(pc->rp0_freq, pc_max_freq_cap(pc))); + return ret; } diff --git a/drivers/gpu/drm/xe/xe_guc_pc.h b/drivers/gpu/drm/xe/xe_guc_pc.h index 0a2664d5c811..52ecdd5ddbff 100644 --- a/drivers/gpu/drm/xe/xe_guc_pc.h +++ b/drivers/gpu/drm/xe/xe_guc_pc.h @@ -38,5 +38,7 @@ u64 xe_guc_pc_mc6_residency(struct xe_guc_pc *pc); void xe_guc_pc_init_early(struct xe_guc_pc *pc); int xe_guc_pc_restore_stashed_freq(struct xe_guc_pc *pc); void xe_guc_pc_raise_unslice(struct xe_guc_pc *pc); +void xe_guc_pc_apply_flush_freq_limit(struct xe_guc_pc *pc); +void xe_guc_pc_remove_flush_freq_limit(struct xe_guc_pc *pc); #endif /* _XE_GUC_PC_H_ */ diff --git a/drivers/gpu/drm/xe/xe_guc_pc_types.h b/drivers/gpu/drm/xe/xe_guc_pc_types.h index 2978ac9a249b..c02053948a57 100644 --- a/drivers/gpu/drm/xe/xe_guc_pc_types.h +++ b/drivers/gpu/drm/xe/xe_guc_pc_types.h @@ -15,6 +15,8 @@ struct xe_guc_pc { /** @bo: GGTT buffer object that is shared with GuC PC */ struct xe_bo *bo; + /** @flush_freq_limit: 1 when max freq changes are limited by driver */ + atomic_t flush_freq_limit; /** @rp0_freq: HW RP0 frequency - The Maximum one */ u32 rp0_freq; /** @rpa_freq: HW RPa frequency - The Achievable one */ diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index df7a5a4eec74..5f91b9a6ab7d 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -908,12 +908,13 @@ static void xe_guc_exec_queue_lr_cleanup(struct work_struct *w) struct xe_exec_queue *q = ge->q; struct xe_guc *guc = exec_queue_to_guc(q); struct xe_gpu_scheduler *sched = &ge->sched; - bool wedged; + bool wedged = false; xe_gt_assert(guc_to_gt(guc), xe_exec_queue_is_lr(q)); trace_xe_exec_queue_lr_cleanup(q); - wedged = guc_submit_hint_wedged(exec_queue_to_guc(q)); + if (!exec_queue_killed(q)) + wedged = guc_submit_hint_wedged(exec_queue_to_guc(q)); /* Kill the run_job / process_msg entry points */ xe_sched_submission_stop(sched); @@ -1084,7 +1085,7 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job) int err = -ETIME; pid_t pid = -1; int i = 0; - bool wedged, skip_timeout_check; + bool wedged = false, skip_timeout_check; /* * TDR has fired before free job worker. Common if exec queue @@ -1130,7 +1131,8 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job) * doesn't work for SRIOV. For now assuming timeouts in wedged mode are * genuine timeouts. */ - wedged = guc_submit_hint_wedged(exec_queue_to_guc(q)); + if (!exec_queue_killed(q)) + wedged = guc_submit_hint_wedged(exec_queue_to_guc(q)); /* Engine state now stable, disable scheduling to check timestamp */ if (!wedged && exec_queue_registered(q)) { @@ -2090,12 +2092,16 @@ int xe_guc_exec_queue_memory_cat_error_handler(struct xe_guc *guc, u32 *msg, struct xe_gt *gt = guc_to_gt(guc); struct xe_exec_queue *q; u32 guc_id; + u32 type = XE_GUC_CAT_ERR_TYPE_INVALID; - if (unlikely(len < 1)) + if (unlikely(!len || len > 2)) return -EPROTO; guc_id = msg[0]; + if (len == 2) + type = msg[1]; + if (guc_id == GUC_ID_UNKNOWN) { /* * GuC uses GUC_ID_UNKNOWN if it can not map the CAT fault to any PF/VF @@ -2109,8 +2115,19 @@ int xe_guc_exec_queue_memory_cat_error_handler(struct xe_guc *guc, u32 *msg, if (unlikely(!q)) return -EPROTO; - xe_gt_dbg(gt, "Engine memory cat error: engine_class=%s, logical_mask: 0x%x, guc_id=%d", - xe_hw_engine_class_to_str(q->class), q->logical_mask, guc_id); + /* + * The type is HW-defined and changes based on platform, so we don't + * decode it in the kernel and only check if it is valid. + * See bspec 54047 and 72187 for details. + */ + if (type != XE_GUC_CAT_ERR_TYPE_INVALID) + xe_gt_dbg(gt, + "Engine memory CAT error [%u]: class=%s, logical_mask: 0x%x, guc_id=%d", + type, xe_hw_engine_class_to_str(q->class), q->logical_mask, guc_id); + else + xe_gt_dbg(gt, + "Engine memory CAT error: class=%s, logical_mask: 0x%x, guc_id=%d", + xe_hw_engine_class_to_str(q->class), q->logical_mask, guc_id); trace_xe_exec_queue_memory_cat_error(q); diff --git a/drivers/gpu/drm/xe/xe_heci_gsc.c b/drivers/gpu/drm/xe/xe_heci_gsc.c index 27d11e06a82b..6d7b62724126 100644 --- a/drivers/gpu/drm/xe/xe_heci_gsc.c +++ b/drivers/gpu/drm/xe/xe_heci_gsc.c @@ -11,15 +11,12 @@ #include "xe_device_types.h" #include "xe_drv.h" #include "xe_heci_gsc.h" +#include "regs/xe_gsc_regs.h" #include "xe_platform_types.h" #include "xe_survivability_mode.h" #define GSC_BAR_LENGTH 0x00000FFC -#define DG1_GSC_HECI2_BASE 0x259000 -#define PVC_GSC_HECI2_BASE 0x285000 -#define DG2_GSC_HECI2_BASE 0x374000 - static void heci_gsc_irq_mask(struct irq_data *d) { /* generic irq handling */ diff --git a/drivers/gpu/drm/xe/xe_huc.c b/drivers/gpu/drm/xe/xe_huc.c index 6a846e4cb221..7e43b2dd6a32 100644 --- a/drivers/gpu/drm/xe/xe_huc.c +++ b/drivers/gpu/drm/xe/xe_huc.c @@ -171,7 +171,7 @@ static int huc_auth_via_gsccs(struct xe_huc *huc) sizeof(struct pxp43_new_huc_auth_in)); wr_offset = huc_emit_pxp_auth_msg(xe, &pkt->vmap, wr_offset, xe_bo_ggtt_addr(huc->fw.bo), - huc->fw.bo->size); + xe_bo_size(huc->fw.bo)); do { err = xe_gsc_pkt_submit_kernel(>->uc.gsc, ggtt_offset, wr_offset, ggtt_offset + PXP43_HUC_AUTH_INOUT_SIZE, diff --git a/drivers/gpu/drm/xe/xe_hw_engine.c b/drivers/gpu/drm/xe/xe_hw_engine.c index 3439c8522d01..796ba8c34a16 100644 --- a/drivers/gpu/drm/xe/xe_hw_engine.c +++ b/drivers/gpu/drm/xe/xe_hw_engine.c @@ -1059,12 +1059,13 @@ struct xe_hw_engine * xe_hw_engine_lookup(struct xe_device *xe, struct drm_xe_engine_class_instance eci) { + struct xe_gt *gt = xe_device_get_gt(xe, eci.gt_id); unsigned int idx; if (eci.engine_class >= ARRAY_SIZE(user_to_xe_engine_class)) return NULL; - if (eci.gt_id >= xe->info.gt_count) + if (!gt) return NULL; idx = array_index_nospec(eci.engine_class, diff --git a/drivers/gpu/drm/xe/xe_hw_engine_group.c b/drivers/gpu/drm/xe/xe_hw_engine_group.c index 2d68c5b5262a..87a6dcb1b4b5 100644 --- a/drivers/gpu/drm/xe/xe_hw_engine_group.c +++ b/drivers/gpu/drm/xe/xe_hw_engine_group.c @@ -13,15 +13,6 @@ #include "xe_vm.h" static void -hw_engine_group_free(struct drm_device *drm, void *arg) -{ - struct xe_hw_engine_group *group = arg; - - destroy_workqueue(group->resume_wq); - kfree(group); -} - -static void hw_engine_group_resume_lr_jobs_func(struct work_struct *w) { struct xe_exec_queue *q; @@ -53,7 +44,7 @@ hw_engine_group_alloc(struct xe_device *xe) struct xe_hw_engine_group *group; int err; - group = kzalloc(sizeof(*group), GFP_KERNEL); + group = drmm_kzalloc(&xe->drm, sizeof(*group), GFP_KERNEL); if (!group) return ERR_PTR(-ENOMEM); @@ -61,14 +52,14 @@ hw_engine_group_alloc(struct xe_device *xe) if (!group->resume_wq) return ERR_PTR(-ENOMEM); + err = drmm_add_action_or_reset(&xe->drm, __drmm_workqueue_release, group->resume_wq); + if (err) + return ERR_PTR(err); + init_rwsem(&group->mode_sem); INIT_WORK(&group->resume_work, hw_engine_group_resume_lr_jobs_func); INIT_LIST_HEAD(&group->exec_queue_list); - err = drmm_add_action_or_reset(&xe->drm, hw_engine_group_free, group); - if (err) - return ERR_PTR(err); - return group; } diff --git a/drivers/gpu/drm/xe/xe_i2c.c b/drivers/gpu/drm/xe/xe_i2c.c new file mode 100644 index 000000000000..db9c0340be5c --- /dev/null +++ b/drivers/gpu/drm/xe/xe_i2c.c @@ -0,0 +1,329 @@ +// SPDX-License-Identifier: GPL-2.0 OR MIT +/* + * Intel Xe I2C attached Microcontroller Units (MCU) + * + * Copyright (C) 2025 Intel Corporation. + */ + +#include <linux/array_size.h> +#include <linux/container_of.h> +#include <linux/device.h> +#include <linux/err.h> +#include <linux/i2c.h> +#include <linux/ioport.h> +#include <linux/irq.h> +#include <linux/irqdomain.h> +#include <linux/notifier.h> +#include <linux/pci.h> +#include <linux/platform_device.h> +#include <linux/property.h> +#include <linux/regmap.h> +#include <linux/sprintf.h> +#include <linux/string.h> +#include <linux/types.h> +#include <linux/workqueue.h> + +#include "regs/xe_i2c_regs.h" +#include "regs/xe_irq_regs.h" + +#include "xe_device.h" +#include "xe_device_types.h" +#include "xe_i2c.h" +#include "xe_mmio.h" +#include "xe_platform_types.h" + +/** + * DOC: Xe I2C devices + * + * Register a platform device for the I2C host controller (Synpsys DesignWare + * I2C) if the registers of that controller are mapped to the MMIO, and also the + * I2C client device for the Add-In Management Controller (the MCU) attached to + * the host controller. + * + * See drivers/i2c/busses/i2c-designware-* for more information on the I2C host + * controller. + */ + +static const char adapter_name[] = "i2c_designware"; + +static const struct property_entry xe_i2c_adapter_properties[] = { + PROPERTY_ENTRY_STRING("compatible", "intel,xe-i2c"), + PROPERTY_ENTRY_U32("clock-frequency", I2C_MAX_FAST_MODE_PLUS_FREQ), + { } +}; + +static inline void xe_i2c_read_endpoint(struct xe_mmio *mmio, void *ep) +{ + u32 *val = ep; + + val[0] = xe_mmio_read32(mmio, REG_SG_REMAP_ADDR_PREFIX); + val[1] = xe_mmio_read32(mmio, REG_SG_REMAP_ADDR_POSTFIX); +} + +static void xe_i2c_client_work(struct work_struct *work) +{ + struct xe_i2c *i2c = container_of(work, struct xe_i2c, work); + struct i2c_board_info info = { + .type = "amc", + .flags = I2C_CLIENT_HOST_NOTIFY, + .addr = i2c->ep.addr[1], + }; + + i2c->client[0] = i2c_new_client_device(i2c->adapter, &info); +} + +static int xe_i2c_notifier(struct notifier_block *nb, unsigned long action, void *data) +{ + struct xe_i2c *i2c = container_of(nb, struct xe_i2c, bus_notifier); + struct i2c_adapter *adapter = i2c_verify_adapter(data); + struct device *dev = data; + + if (action == BUS_NOTIFY_ADD_DEVICE && + adapter && dev->parent == &i2c->pdev->dev) { + i2c->adapter = adapter; + schedule_work(&i2c->work); + return NOTIFY_OK; + } + + return NOTIFY_DONE; +} + +static int xe_i2c_register_adapter(struct xe_i2c *i2c) +{ + struct pci_dev *pci = to_pci_dev(i2c->drm_dev); + struct platform_device *pdev; + struct fwnode_handle *fwnode; + int ret; + + fwnode = fwnode_create_software_node(xe_i2c_adapter_properties, NULL); + if (!fwnode) + return -ENOMEM; + + /* + * Not using platform_device_register_full() here because we don't have + * a handle to the platform_device before it returns. xe_i2c_notifier() + * uses that handle, but it may be called before + * platform_device_register_full() is done. + */ + pdev = platform_device_alloc(adapter_name, pci_dev_id(pci)); + if (!pdev) { + ret = -ENOMEM; + goto err_fwnode_remove; + } + + if (i2c->adapter_irq) { + struct resource res; + + res = DEFINE_RES_IRQ_NAMED(i2c->adapter_irq, "xe_i2c"); + + ret = platform_device_add_resources(pdev, &res, 1); + if (ret) + goto err_pdev_put; + } + + pdev->dev.parent = i2c->drm_dev; + pdev->dev.fwnode = fwnode; + i2c->adapter_node = fwnode; + i2c->pdev = pdev; + + ret = platform_device_add(pdev); + if (ret) + goto err_pdev_put; + + return 0; + +err_pdev_put: + platform_device_put(pdev); +err_fwnode_remove: + fwnode_remove_software_node(fwnode); + + return ret; +} + +static void xe_i2c_unregister_adapter(struct xe_i2c *i2c) +{ + platform_device_unregister(i2c->pdev); + fwnode_remove_software_node(i2c->adapter_node); +} + +/** + * xe_i2c_irq_handler: Handler for I2C interrupts + * @xe: xe device instance + * @master_ctl: interrupt register + * + * Forward interrupts generated by the I2C host adapter to the I2C host adapter + * driver. + */ +void xe_i2c_irq_handler(struct xe_device *xe, u32 master_ctl) +{ + if (!xe->i2c || !xe->i2c->adapter_irq) + return; + + if (master_ctl & I2C_IRQ) + generic_handle_irq_safe(xe->i2c->adapter_irq); +} + +static int xe_i2c_irq_map(struct irq_domain *h, unsigned int virq, + irq_hw_number_t hw_irq_num) +{ + irq_set_chip_and_handler(virq, &dummy_irq_chip, handle_simple_irq); + return 0; +} + +static const struct irq_domain_ops xe_i2c_irq_ops = { + .map = xe_i2c_irq_map, +}; + +static int xe_i2c_create_irq(struct xe_i2c *i2c) +{ + struct irq_domain *domain; + + if (!(i2c->ep.capabilities & XE_I2C_EP_CAP_IRQ)) + return 0; + + domain = irq_domain_create_linear(dev_fwnode(i2c->drm_dev), 1, &xe_i2c_irq_ops, NULL); + if (!domain) + return -ENOMEM; + + i2c->adapter_irq = irq_create_mapping(domain, 0); + i2c->irqdomain = domain; + + return 0; +} + +static void xe_i2c_remove_irq(struct xe_i2c *i2c) +{ + if (!i2c->irqdomain) + return; + + irq_dispose_mapping(i2c->adapter_irq); + irq_domain_remove(i2c->irqdomain); +} + +static int xe_i2c_read(void *context, unsigned int reg, unsigned int *val) +{ + struct xe_i2c *i2c = context; + + *val = xe_mmio_read32(i2c->mmio, XE_REG(reg + I2C_MEM_SPACE_OFFSET)); + + return 0; +} + +static int xe_i2c_write(void *context, unsigned int reg, unsigned int val) +{ + struct xe_i2c *i2c = context; + + xe_mmio_write32(i2c->mmio, XE_REG(reg + I2C_MEM_SPACE_OFFSET), val); + + return 0; +} + +static const struct regmap_config i2c_regmap_config = { + .reg_bits = 32, + .val_bits = 32, + .reg_read = xe_i2c_read, + .reg_write = xe_i2c_write, + .fast_io = true, +}; + +void xe_i2c_pm_suspend(struct xe_device *xe) +{ + struct xe_mmio *mmio = xe_root_tile_mmio(xe); + + if (!xe->i2c || xe->i2c->ep.cookie != XE_I2C_EP_COOKIE_DEVICE) + return; + + xe_mmio_rmw32(mmio, I2C_CONFIG_PMCSR, PCI_PM_CTRL_STATE_MASK, (__force u32)PCI_D3hot); + drm_dbg(&xe->drm, "pmcsr: 0x%08x\n", xe_mmio_read32(mmio, I2C_CONFIG_PMCSR)); +} + +void xe_i2c_pm_resume(struct xe_device *xe, bool d3cold) +{ + struct xe_mmio *mmio = xe_root_tile_mmio(xe); + + if (!xe->i2c || xe->i2c->ep.cookie != XE_I2C_EP_COOKIE_DEVICE) + return; + + if (d3cold) + xe_mmio_rmw32(mmio, I2C_CONFIG_CMD, 0, PCI_COMMAND_MEMORY); + + xe_mmio_rmw32(mmio, I2C_CONFIG_PMCSR, PCI_PM_CTRL_STATE_MASK, (__force u32)PCI_D0); + drm_dbg(&xe->drm, "pmcsr: 0x%08x\n", xe_mmio_read32(mmio, I2C_CONFIG_PMCSR)); +} + +static void xe_i2c_remove(void *data) +{ + struct xe_i2c *i2c = data; + unsigned int i; + + for (i = 0; i < XE_I2C_MAX_CLIENTS; i++) + i2c_unregister_device(i2c->client[i]); + + bus_unregister_notifier(&i2c_bus_type, &i2c->bus_notifier); + xe_i2c_unregister_adapter(i2c); + xe_i2c_remove_irq(i2c); +} + +/** + * xe_i2c_probe: Probe the I2C host adapter and the I2C clients attached to it + * @xe: xe device instance + * + * Register all the I2C devices described in the I2C Endpoint data structure. + * + * Return: 0 on success, error code on failure + */ +int xe_i2c_probe(struct xe_device *xe) +{ + struct device *drm_dev = xe->drm.dev; + struct xe_i2c_endpoint ep; + struct regmap *regmap; + struct xe_i2c *i2c; + int ret; + + if (xe->info.platform != XE_BATTLEMAGE) + return 0; + + xe_i2c_read_endpoint(xe_root_tile_mmio(xe), &ep); + if (ep.cookie != XE_I2C_EP_COOKIE_DEVICE) + return 0; + + i2c = devm_kzalloc(drm_dev, sizeof(*i2c), GFP_KERNEL); + if (!i2c) + return -ENOMEM; + + INIT_WORK(&i2c->work, xe_i2c_client_work); + i2c->mmio = xe_root_tile_mmio(xe); + i2c->drm_dev = drm_dev; + i2c->ep = ep; + xe->i2c = i2c; + + /* PCI PM isn't aware of this device, bring it up and match it with SGUnit state. */ + xe_i2c_pm_resume(xe, true); + + regmap = devm_regmap_init(drm_dev, NULL, i2c, &i2c_regmap_config); + if (IS_ERR(regmap)) + return PTR_ERR(regmap); + + i2c->bus_notifier.notifier_call = xe_i2c_notifier; + ret = bus_register_notifier(&i2c_bus_type, &i2c->bus_notifier); + if (ret) + return ret; + + ret = xe_i2c_create_irq(i2c); + if (ret) + goto err_unregister_notifier; + + ret = xe_i2c_register_adapter(i2c); + if (ret) + goto err_remove_irq; + + return devm_add_action_or_reset(drm_dev, xe_i2c_remove, i2c); + +err_remove_irq: + xe_i2c_remove_irq(i2c); + +err_unregister_notifier: + bus_unregister_notifier(&i2c_bus_type, &i2c->bus_notifier); + + return ret; +} diff --git a/drivers/gpu/drm/xe/xe_i2c.h b/drivers/gpu/drm/xe/xe_i2c.h new file mode 100644 index 000000000000..b767ed8ce52b --- /dev/null +++ b/drivers/gpu/drm/xe/xe_i2c.h @@ -0,0 +1,62 @@ +/* SPDX-License-Identifier: MIT */ +#ifndef _XE_I2C_H_ +#define _XE_I2C_H_ + +#include <linux/bits.h> +#include <linux/notifier.h> +#include <linux/types.h> +#include <linux/workqueue.h> + +struct device; +struct fwnode_handle; +struct i2c_adapter; +struct i2c_client; +struct irq_domain; +struct platform_device; +struct xe_device; +struct xe_mmio; + +#define XE_I2C_MAX_CLIENTS 3 + +#define XE_I2C_EP_COOKIE_DEVICE 0xde + +/* Endpoint Capabilities */ +#define XE_I2C_EP_CAP_IRQ BIT(0) + +struct xe_i2c_endpoint { + u8 cookie; + u8 capabilities; + u16 addr[XE_I2C_MAX_CLIENTS]; +}; + +struct xe_i2c { + struct fwnode_handle *adapter_node; + struct platform_device *pdev; + struct i2c_adapter *adapter; + struct i2c_client *client[XE_I2C_MAX_CLIENTS]; + + struct notifier_block bus_notifier; + struct work_struct work; + + struct irq_domain *irqdomain; + int adapter_irq; + + struct xe_i2c_endpoint ep; + struct device *drm_dev; + + struct xe_mmio *mmio; +}; + +#if IS_ENABLED(CONFIG_I2C) +int xe_i2c_probe(struct xe_device *xe); +void xe_i2c_irq_handler(struct xe_device *xe, u32 master_ctl); +void xe_i2c_pm_suspend(struct xe_device *xe); +void xe_i2c_pm_resume(struct xe_device *xe, bool d3cold); +#else +static inline int xe_i2c_probe(struct xe_device *xe) { return 0; } +static inline void xe_i2c_irq_handler(struct xe_device *xe, u32 master_ctl) { } +static inline void xe_i2c_pm_suspend(struct xe_device *xe) { } +static inline void xe_i2c_pm_resume(struct xe_device *xe, bool d3cold) { } +#endif + +#endif diff --git a/drivers/gpu/drm/xe/xe_irq.c b/drivers/gpu/drm/xe/xe_irq.c index 5362d3174b06..c43e62dc692e 100644 --- a/drivers/gpu/drm/xe/xe_irq.c +++ b/drivers/gpu/drm/xe/xe_irq.c @@ -18,6 +18,7 @@ #include "xe_gt.h" #include "xe_guc.h" #include "xe_hw_engine.h" +#include "xe_i2c.h" #include "xe_memirq.h" #include "xe_mmio.h" #include "xe_pxp.h" @@ -476,6 +477,7 @@ static irqreturn_t dg1_irq_handler(int irq, void *arg) if (xe->info.has_heci_cscfi) xe_heci_csc_irq_handler(xe, master_ctl); xe_display_irq_handler(xe, master_ctl); + xe_i2c_irq_handler(xe, master_ctl); gu_misc_iir = gu_misc_irq_ack(xe, master_ctl); } } diff --git a/drivers/gpu/drm/xe/xe_lmtt.c b/drivers/gpu/drm/xe/xe_lmtt.c index 63db66df064b..381f576036d0 100644 --- a/drivers/gpu/drm/xe/xe_lmtt.c +++ b/drivers/gpu/drm/xe/xe_lmtt.c @@ -78,6 +78,9 @@ static struct xe_lmtt_pt *lmtt_pt_alloc(struct xe_lmtt *lmtt, unsigned int level } lmtt_assert(lmtt, xe_bo_is_vram(bo)); + lmtt_debug(lmtt, "level=%u addr=%#llx\n", level, (u64)xe_bo_main_addr(bo, XE_PAGE_SIZE)); + + xe_map_memset(lmtt_to_xe(lmtt), &bo->vmap, 0, 0, xe_bo_size(bo)); pt->level = level; pt->bo = bo; @@ -91,6 +94,9 @@ out: static void lmtt_pt_free(struct xe_lmtt_pt *pt) { + lmtt_debug(&pt->bo->tile->sriov.pf.lmtt, "level=%u addr=%llx\n", + pt->level, (u64)xe_bo_main_addr(pt->bo, XE_PAGE_SIZE)); + xe_bo_unpin_map_no_vm(pt->bo); kfree(pt); } @@ -226,9 +232,14 @@ static void lmtt_write_pte(struct xe_lmtt *lmtt, struct xe_lmtt_pt *pt, switch (lmtt->ops->lmtt_pte_size(level)) { case sizeof(u32): + lmtt_assert(lmtt, !overflows_type(pte, u32)); + lmtt_assert(lmtt, !pte || !iosys_map_rd(&pt->bo->vmap, idx * sizeof(u32), u32)); + xe_map_wr(lmtt_to_xe(lmtt), &pt->bo->vmap, idx * sizeof(u32), u32, pte); break; case sizeof(u64): + lmtt_assert(lmtt, !pte || !iosys_map_rd(&pt->bo->vmap, idx * sizeof(u64), u64)); + xe_map_wr(lmtt_to_xe(lmtt), &pt->bo->vmap, idx * sizeof(u64), u64, pte); break; default: @@ -386,11 +397,11 @@ static void lmtt_insert_bo(struct xe_lmtt *lmtt, unsigned int vfid, struct xe_bo u64 addr, vram_offset; lmtt_assert(lmtt, IS_ALIGNED(start, page_size)); - lmtt_assert(lmtt, IS_ALIGNED(bo->size, page_size)); + lmtt_assert(lmtt, IS_ALIGNED(xe_bo_size(bo), page_size)); lmtt_assert(lmtt, xe_bo_is_vram(bo)); vram_offset = vram_region_gpu_offset(bo->ttm.resource); - xe_res_first(bo->ttm.resource, 0, bo->size, &cur); + xe_res_first(bo->ttm.resource, 0, xe_bo_size(bo), &cur); while (cur.remaining) { addr = xe_res_dma(&cur); addr += vram_offset; /* XXX */ diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c index 37598588a54f..d2ad8fe737eb 100644 --- a/drivers/gpu/drm/xe/xe_lrc.c +++ b/drivers/gpu/drm/xe/xe_lrc.c @@ -717,8 +717,12 @@ static u32 __xe_lrc_ctx_timestamp_udw_offset(struct xe_lrc *lrc) static inline u32 __xe_lrc_indirect_ring_offset(struct xe_lrc *lrc) { - /* Indirect ring state page is at the very end of LRC */ - return lrc->size - LRC_INDIRECT_RING_STATE_SIZE; + return xe_bo_size(lrc->bo) - LRC_WA_BB_SIZE - LRC_INDIRECT_RING_STATE_SIZE; +} + +static inline u32 __xe_lrc_wa_bb_offset(struct xe_lrc *lrc) +{ + return xe_bo_size(lrc->bo) - LRC_WA_BB_SIZE; } #define DECL_MAP_ADDR_HELPERS(elem) \ @@ -973,11 +977,6 @@ struct wa_bb_setup { u32 *batch, size_t max_size); }; -static size_t wa_bb_offset(struct xe_lrc *lrc) -{ - return lrc->bo->size - LRC_WA_BB_SIZE; -} - static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe) { const size_t max_size = LRC_WA_BB_SIZE; @@ -993,7 +992,7 @@ static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe) return -ENOMEM; cmd = buf; } else { - cmd = lrc->bo->vmap.vaddr + wa_bb_offset(lrc); + cmd = lrc->bo->vmap.vaddr + __xe_lrc_wa_bb_offset(lrc); } remain = max_size / sizeof(*cmd); @@ -1017,13 +1016,13 @@ static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe) if (buf) { xe_map_memcpy_to(gt_to_xe(lrc->gt), &lrc->bo->vmap, - wa_bb_offset(lrc), buf, + __xe_lrc_wa_bb_offset(lrc), buf, (cmd - buf) * sizeof(*cmd)); kfree(buf); } xe_lrc_write_ctx_reg(lrc, CTX_BB_PER_CTX_PTR, xe_bo_ggtt_addr(lrc->bo) + - wa_bb_offset(lrc) + 1); + __xe_lrc_wa_bb_offset(lrc) + 1); return 0; @@ -1040,19 +1039,22 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, u32 init_flags) { struct xe_gt *gt = hwe->gt; + const u32 lrc_size = xe_gt_lrc_size(gt, hwe->class); + const u32 bo_size = ring_size + lrc_size + LRC_WA_BB_SIZE; struct xe_tile *tile = gt_to_tile(gt); struct xe_device *xe = gt_to_xe(gt); struct iosys_map map; void *init_data = NULL; u32 arb_enable; - u32 lrc_size; u32 bo_flags; int err; kref_init(&lrc->refcount); lrc->gt = gt; + lrc->size = lrc_size; lrc->flags = 0; - lrc_size = ring_size + xe_gt_lrc_size(gt, hwe->class); + lrc->ring.size = ring_size; + lrc->ring.tail = 0; if (xe_gt_has_indirect_ring_state(gt)) lrc->flags |= XE_LRC_FLAG_INDIRECT_RING_STATE; @@ -1065,17 +1067,12 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, * FIXME: Perma-pinning LRC as we don't yet support moving GGTT address * via VM bind calls. */ - lrc->bo = xe_bo_create_pin_map(xe, tile, NULL, - lrc_size + LRC_WA_BB_SIZE, + lrc->bo = xe_bo_create_pin_map(xe, tile, NULL, bo_size, ttm_bo_type_kernel, bo_flags); if (IS_ERR(lrc->bo)) return PTR_ERR(lrc->bo); - lrc->size = lrc_size; - lrc->ring.size = ring_size; - lrc->ring.tail = 0; - xe_hw_fence_ctx_init(&lrc->fence_ctx, hwe->gt, hwe->fence_irq, hwe->name); @@ -1096,10 +1093,9 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe, xe_map_memset(xe, &map, 0, 0, LRC_PPHWSP_SIZE); /* PPHWSP */ xe_map_memcpy_to(xe, &map, LRC_PPHWSP_SIZE, gt->default_lrc[hwe->class] + LRC_PPHWSP_SIZE, - xe_gt_lrc_size(gt, hwe->class) - LRC_PPHWSP_SIZE); + lrc_size - LRC_PPHWSP_SIZE); } else { - xe_map_memcpy_to(xe, &map, 0, init_data, - xe_gt_lrc_size(gt, hwe->class)); + xe_map_memcpy_to(xe, &map, 0, init_data, lrc_size); kfree(init_data); } @@ -1859,8 +1855,7 @@ struct xe_lrc_snapshot *xe_lrc_snapshot_capture(struct xe_lrc *lrc) snapshot->seqno = xe_lrc_seqno(lrc); snapshot->lrc_bo = xe_bo_get(lrc->bo); snapshot->lrc_offset = xe_lrc_pphwsp_offset(lrc); - snapshot->lrc_size = lrc->bo->size - snapshot->lrc_offset - - LRC_WA_BB_SIZE; + snapshot->lrc_size = lrc->size; snapshot->lrc_snapshot = NULL; snapshot->ctx_timestamp = lower_32_bits(xe_lrc_ctx_timestamp(lrc)); snapshot->ctx_job_timestamp = xe_lrc_ctx_job_timestamp(lrc); diff --git a/drivers/gpu/drm/xe/xe_lrc_types.h b/drivers/gpu/drm/xe/xe_lrc_types.h index 883e550a9423..2c7c81079801 100644 --- a/drivers/gpu/drm/xe/xe_lrc_types.h +++ b/drivers/gpu/drm/xe/xe_lrc_types.h @@ -22,7 +22,7 @@ struct xe_lrc { */ struct xe_bo *bo; - /** @size: size of lrc including any indirect ring state page */ + /** @size: size of the lrc and optional indirect ring state */ u32 size; /** @gt: gt which this LRC belongs to */ diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c index 8f8e9fdfb2a8..4e2bdf70eb70 100644 --- a/drivers/gpu/drm/xe/xe_migrate.c +++ b/drivers/gpu/drm/xe/xe_migrate.c @@ -82,7 +82,7 @@ struct xe_migrate { * of the instruction. Subtracting the instruction header (1 dword) and * address (2 dwords), that leaves 0x3FD dwords (0x1FE qwords) for PTE values. */ -#define MAX_PTE_PER_SDI 0x1FE +#define MAX_PTE_PER_SDI 0x1FEU /** * xe_tile_migrate_exec_queue() - Get this tile's migrate exec queue. @@ -203,7 +203,7 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, BUILD_BUG_ON(!(NUM_KERNEL_PDE & 1)); /* Need to be sure everything fits in the first PT, or create more */ - xe_tile_assert(tile, m->batch_base_ofs + batch->size < SZ_2M); + xe_tile_assert(tile, m->batch_base_ofs + xe_bo_size(batch) < SZ_2M); bo = xe_bo_create_pin_map(vm->xe, tile, vm, num_entries * XE_PAGE_SIZE, @@ -214,7 +214,7 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, return PTR_ERR(bo); /* PT30 & PT31 reserved for 2M identity map */ - pt29_ofs = bo->size - 3 * XE_PAGE_SIZE; + pt29_ofs = xe_bo_size(bo) - 3 * XE_PAGE_SIZE; entry = vm->pt_ops->pde_encode_bo(bo, pt29_ofs, pat_index); xe_pt_write(xe, &vm->pt_root[id]->bo->vmap, 0, entry); @@ -236,7 +236,7 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, if (!IS_DGFX(xe)) { /* Write out batch too */ m->batch_base_ofs = NUM_PT_SLOTS * XE_PAGE_SIZE; - for (i = 0; i < batch->size; + for (i = 0; i < xe_bo_size(batch); i += vm->flags & XE_VM_FLAG_64K ? XE_64K_PAGE_SIZE : XE_PAGE_SIZE) { entry = vm->pt_ops->pte_encode_bo(batch, i, @@ -247,13 +247,13 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, level++; } if (xe->info.has_usm) { - xe_tile_assert(tile, batch->size == SZ_1M); + xe_tile_assert(tile, xe_bo_size(batch) == SZ_1M); batch = tile->primary_gt->usm.bb_pool->bo; m->usm_batch_base_ofs = m->batch_base_ofs + SZ_1M; - xe_tile_assert(tile, batch->size == SZ_512K); + xe_tile_assert(tile, xe_bo_size(batch) == SZ_512K); - for (i = 0; i < batch->size; + for (i = 0; i < xe_bo_size(batch); i += vm->flags & XE_VM_FLAG_64K ? XE_64K_PAGE_SIZE : XE_PAGE_SIZE) { entry = vm->pt_ops->pte_encode_bo(batch, i, @@ -306,7 +306,7 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, /* Identity map the entire vram at 256GiB offset */ if (IS_DGFX(xe)) { - u64 pt30_ofs = bo->size - 2 * XE_PAGE_SIZE; + u64 pt30_ofs = xe_bo_size(bo) - 2 * XE_PAGE_SIZE; xe_migrate_program_identity(xe, vm, bo, map_ofs, IDENTITY_OFFSET, pat_index, pt30_ofs); @@ -321,7 +321,7 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, u16 comp_pat_index = xe->pat.idx[XE_CACHE_NONE_COMPRESSION]; u64 vram_offset = IDENTITY_OFFSET + DIV_ROUND_UP_ULL(xe->mem.vram.actual_physical_size, SZ_1G); - u64 pt31_ofs = bo->size - XE_PAGE_SIZE; + u64 pt31_ofs = xe_bo_size(bo) - XE_PAGE_SIZE; xe_assert(xe, xe->mem.vram.actual_physical_size <= (MAX_NUM_PTE - IDENTITY_OFFSET - IDENTITY_OFFSET / 2) * SZ_1G); @@ -768,7 +768,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m, struct xe_gt *gt = m->tile->primary_gt; struct xe_device *xe = gt_to_xe(gt); struct dma_fence *fence = NULL; - u64 size = src_bo->size; + u64 size = xe_bo_size(src_bo); struct xe_res_cursor src_it, dst_it, ccs_it; u64 src_L0_ofs, dst_L0_ofs; u32 src_L0_pt, dst_L0_pt; @@ -791,7 +791,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m, if (XE_WARN_ON(copy_ccs && src_bo != dst_bo)) return ERR_PTR(-EINVAL); - if (src_bo != dst_bo && XE_WARN_ON(src_bo->size != dst_bo->size)) + if (src_bo != dst_bo && XE_WARN_ON(xe_bo_size(src_bo) != xe_bo_size(dst_bo))) return ERR_PTR(-EINVAL); if (!src_is_vram) @@ -863,7 +863,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m, if (src_is_vram && xe_migrate_allow_identity(src_L0, &src_it)) xe_res_next(&src_it, src_L0); else - emit_pte(m, bb, src_L0_pt, src_is_vram, copy_system_ccs, + emit_pte(m, bb, src_L0_pt, src_is_vram, copy_system_ccs || use_comp_pat, &src_it, src_L0, src); if (dst_is_vram && xe_migrate_allow_identity(src_L0, &dst_it)) @@ -1064,7 +1064,7 @@ struct dma_fence *xe_migrate_clear(struct xe_migrate *m, struct xe_device *xe = gt_to_xe(gt); bool clear_only_system_ccs = false; struct dma_fence *fence = NULL; - u64 size = bo->size; + u64 size = xe_bo_size(bo); struct xe_res_cursor src_it; struct ttm_resource *src = dst; int err; @@ -1076,9 +1076,9 @@ struct dma_fence *xe_migrate_clear(struct xe_migrate *m, clear_only_system_ccs = true; if (!clear_vram) - xe_res_first_sg(xe_bo_sg(bo), 0, bo->size, &src_it); + xe_res_first_sg(xe_bo_sg(bo), 0, xe_bo_size(bo), &src_it); else - xe_res_first(src, 0, bo->size, &src_it); + xe_res_first(src, 0, xe_bo_size(bo), &src_it); while (size) { u64 clear_L0_ofs; @@ -1407,7 +1407,7 @@ __xe_migrate_update_pgtables(struct xe_migrate *m, if (idx == chunk) goto next_cmd; - xe_tile_assert(tile, pt_bo->size == SZ_4K); + xe_tile_assert(tile, xe_bo_size(pt_bo) == SZ_4K); /* Map a PT at most once */ if (pt_bo->update_index < 0) @@ -1553,15 +1553,17 @@ static u32 pte_update_cmd_size(u64 size) u64 entries = DIV_U64_ROUND_UP(size, XE_PAGE_SIZE); XE_WARN_ON(size > MAX_PREEMPTDISABLE_TRANSFER); + /* * MI_STORE_DATA_IMM command is used to update page table. Each - * instruction can update maximumly 0x1ff pte entries. To update - * n (n <= 0x1ff) pte entries, we need: - * 1 dword for the MI_STORE_DATA_IMM command header (opcode etc) - * 2 dword for the page table's physical location - * 2*n dword for value of pte to fill (each pte entry is 2 dwords) + * instruction can update maximumly MAX_PTE_PER_SDI pte entries. To + * update n (n <= MAX_PTE_PER_SDI) pte entries, we need: + * + * - 1 dword for the MI_STORE_DATA_IMM command header (opcode etc) + * - 2 dword for the page table's physical location + * - 2*n dword for value of pte to fill (each pte entry is 2 dwords) */ - num_dword = (1 + 2) * DIV_U64_ROUND_UP(entries, 0x1ff); + num_dword = (1 + 2) * DIV_U64_ROUND_UP(entries, MAX_PTE_PER_SDI); num_dword += entries * 2; return num_dword; @@ -1577,7 +1579,7 @@ static void build_pt_update_batch_sram(struct xe_migrate *m, ptes = DIV_ROUND_UP(size, XE_PAGE_SIZE); while (ptes) { - u32 chunk = min(0x1ffU, ptes); + u32 chunk = min(MAX_PTE_PER_SDI, ptes); bb->cs[bb->len++] = MI_STORE_DATA_IMM | MI_SDI_NUM_QW(chunk); bb->cs[bb->len++] = pt_offset; @@ -1866,7 +1868,7 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo, if (IS_ERR(dma_addr)) return PTR_ERR(dma_addr); - xe_res_first(bo->ttm.resource, offset, bo->size - offset, &cursor); + xe_res_first(bo->ttm.resource, offset, xe_bo_size(bo) - offset, &cursor); do { struct dma_fence *__fence; diff --git a/drivers/gpu/drm/xe/xe_mmio.c b/drivers/gpu/drm/xe/xe_mmio.c index 7357458bc0d2..751586d6806a 100644 --- a/drivers/gpu/drm/xe/xe_mmio.c +++ b/drivers/gpu/drm/xe/xe_mmio.c @@ -55,6 +55,7 @@ static void tiles_fini(void *arg) static void mmio_multi_tile_setup(struct xe_device *xe, size_t tile_mmio_size) { struct xe_tile *tile; + struct xe_gt *gt; u8 id; /* @@ -67,7 +68,7 @@ static void mmio_multi_tile_setup(struct xe_device *xe, size_t tile_mmio_size) /* Possibly override number of tile based on configuration register */ if (!xe->info.skip_mtcfg) { struct xe_mmio *mmio = xe_root_tile_mmio(xe); - u8 tile_count; + u8 tile_count, gt_count; u32 mtcfg; /* @@ -84,12 +85,15 @@ static void mmio_multi_tile_setup(struct xe_device *xe, size_t tile_mmio_size) xe->info.tile_count = tile_count; /* - * FIXME: Needs some work for standalone media, but - * should be impossible with multi-tile for now: - * multi-tile platform with standalone media doesn't - * exist + * We've already setup gt_count according to the full + * tile count. Re-calculate it to only include the GTs + * that belong to the remaining tile(s). */ - xe->info.gt_count = xe->info.tile_count; + gt_count = 0; + for_each_gt(gt, xe, id) + if (gt->info.id < tile_count * xe->info.max_gt_per_tile) + gt_count++; + xe->info.gt_count = gt_count; } } diff --git a/drivers/gpu/drm/xe/xe_nvm.c b/drivers/gpu/drm/xe/xe_nvm.c new file mode 100644 index 000000000000..61b0a1531a53 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_nvm.c @@ -0,0 +1,167 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright(c) 2019-2025, Intel Corporation. All rights reserved. + */ + +#include <linux/intel_dg_nvm_aux.h> +#include <linux/pci.h> + +#include "xe_device.h" +#include "xe_device_types.h" +#include "xe_mmio.h" +#include "xe_nvm.h" +#include "regs/xe_gsc_regs.h" +#include "xe_sriov.h" + +#define GEN12_GUNIT_NVM_BASE 0x00102040 +#define GEN12_DEBUG_NVM_BASE 0x00101018 + +#define GEN12_CNTL_PROTECTED_NVM_REG 0x0010100C + +#define GEN12_GUNIT_NVM_SIZE 0x80 +#define GEN12_DEBUG_NVM_SIZE 0x4 + +#define NVM_NON_POSTED_ERASE_CHICKEN_BIT BIT(13) + +#define HECI_FW_STATUS_2_NVM_ACCESS_MODE BIT(3) + +static const struct intel_dg_nvm_region regions[INTEL_DG_NVM_REGIONS] = { + [0] = { .name = "DESCRIPTOR", }, + [2] = { .name = "GSC", }, + [9] = { .name = "PADDING", }, + [11] = { .name = "OptionROM", }, + [12] = { .name = "DAM", }, +}; + +static void xe_nvm_release_dev(struct device *dev) +{ +} + +static bool xe_nvm_non_posted_erase(struct xe_device *xe) +{ + struct xe_gt *gt = xe_root_mmio_gt(xe); + + if (xe->info.platform != XE_BATTLEMAGE) + return false; + return !(xe_mmio_read32(>->mmio, XE_REG(GEN12_CNTL_PROTECTED_NVM_REG)) & + NVM_NON_POSTED_ERASE_CHICKEN_BIT); +} + +static bool xe_nvm_writable_override(struct xe_device *xe) +{ + struct xe_gt *gt = xe_root_mmio_gt(xe); + bool writable_override; + resource_size_t base; + + switch (xe->info.platform) { + case XE_BATTLEMAGE: + base = DG2_GSC_HECI2_BASE; + break; + case XE_PVC: + base = PVC_GSC_HECI2_BASE; + break; + case XE_DG2: + base = DG2_GSC_HECI2_BASE; + break; + case XE_DG1: + base = DG1_GSC_HECI2_BASE; + break; + default: + drm_err(&xe->drm, "Unknown platform\n"); + return true; + } + + writable_override = + !(xe_mmio_read32(>->mmio, HECI_FWSTS2(base)) & + HECI_FW_STATUS_2_NVM_ACCESS_MODE); + if (writable_override) + drm_info(&xe->drm, "NVM access overridden by jumper\n"); + return writable_override; +} + +int xe_nvm_init(struct xe_device *xe) +{ + struct pci_dev *pdev = to_pci_dev(xe->drm.dev); + struct auxiliary_device *aux_dev; + struct intel_dg_nvm_dev *nvm; + int ret; + + if (!xe->info.has_gsc_nvm) + return 0; + + /* No access to internal NVM from VFs */ + if (IS_SRIOV_VF(xe)) + return 0; + + /* Nvm pointer should be NULL here */ + if (WARN_ON(xe->nvm)) + return -EFAULT; + + xe->nvm = kzalloc(sizeof(*nvm), GFP_KERNEL); + if (!xe->nvm) + return -ENOMEM; + + nvm = xe->nvm; + + nvm->writable_override = xe_nvm_writable_override(xe); + nvm->non_posted_erase = xe_nvm_non_posted_erase(xe); + nvm->bar.parent = &pdev->resource[0]; + nvm->bar.start = GEN12_GUNIT_NVM_BASE + pdev->resource[0].start; + nvm->bar.end = nvm->bar.start + GEN12_GUNIT_NVM_SIZE - 1; + nvm->bar.flags = IORESOURCE_MEM; + nvm->bar.desc = IORES_DESC_NONE; + nvm->regions = regions; + + nvm->bar2.parent = &pdev->resource[0]; + nvm->bar2.start = GEN12_DEBUG_NVM_BASE + pdev->resource[0].start; + nvm->bar2.end = nvm->bar2.start + GEN12_DEBUG_NVM_SIZE - 1; + nvm->bar2.flags = IORESOURCE_MEM; + nvm->bar2.desc = IORES_DESC_NONE; + + aux_dev = &nvm->aux_dev; + + aux_dev->name = "nvm"; + aux_dev->id = (pci_domain_nr(pdev->bus) << 16) | pci_dev_id(pdev); + aux_dev->dev.parent = &pdev->dev; + aux_dev->dev.release = xe_nvm_release_dev; + + ret = auxiliary_device_init(aux_dev); + if (ret) { + drm_err(&xe->drm, "xe-nvm aux init failed %d\n", ret); + goto err; + } + + ret = auxiliary_device_add(aux_dev); + if (ret) { + drm_err(&xe->drm, "xe-nvm aux add failed %d\n", ret); + auxiliary_device_uninit(aux_dev); + goto err; + } + return 0; + +err: + kfree(nvm); + xe->nvm = NULL; + return ret; +} + +void xe_nvm_fini(struct xe_device *xe) +{ + struct intel_dg_nvm_dev *nvm = xe->nvm; + + if (!xe->info.has_gsc_nvm) + return; + + /* No access to internal NVM from VFs */ + if (IS_SRIOV_VF(xe)) + return; + + /* Nvm pointer should not be NULL here */ + if (WARN_ON(!nvm)) + return; + + auxiliary_device_delete(&nvm->aux_dev); + auxiliary_device_uninit(&nvm->aux_dev); + kfree(nvm); + xe->nvm = NULL; +} diff --git a/drivers/gpu/drm/xe/xe_nvm.h b/drivers/gpu/drm/xe/xe_nvm.h new file mode 100644 index 000000000000..7f3d5f57bed0 --- /dev/null +++ b/drivers/gpu/drm/xe/xe_nvm.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright(c) 2019-2025 Intel Corporation. All rights reserved. + */ + +#ifndef __XE_NVM_H__ +#define __XE_NVM_H__ + +struct xe_device; + +int xe_nvm_init(struct xe_device *xe); + +void xe_nvm_fini(struct xe_device *xe); + +#endif diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index 4829ed46a8b4..a3379d39f76d 100644 --- a/drivers/gpu/drm/xe/xe_oa.c +++ b/drivers/gpu/drm/xe/xe_oa.c @@ -403,7 +403,7 @@ static int xe_oa_append_reports(struct xe_oa_stream *stream, char __user *buf, static void xe_oa_init_oa_buffer(struct xe_oa_stream *stream) { u32 gtt_offset = xe_bo_ggtt_addr(stream->oa_buffer.bo); - int size_exponent = __ffs(stream->oa_buffer.bo->size); + int size_exponent = __ffs(xe_bo_size(stream->oa_buffer.bo)); u32 oa_buf = gtt_offset | OAG_OABUFFER_MEMORY_SELECT; struct xe_mmio *mmio = &stream->gt->mmio; unsigned long flags; @@ -435,7 +435,7 @@ static void xe_oa_init_oa_buffer(struct xe_oa_stream *stream) spin_unlock_irqrestore(&stream->oa_buffer.ptr_lock, flags); /* Zero out the OA buffer since we rely on zero report id and timestamp fields */ - memset(stream->oa_buffer.vaddr, 0, stream->oa_buffer.bo->size); + memset(stream->oa_buffer.vaddr, 0, xe_bo_size(stream->oa_buffer.bo)); } static u32 __format_to_oactrl(const struct xe_oa_format *format, int counter_sel_mask) @@ -1065,7 +1065,7 @@ static u32 oag_report_ctx_switches(const struct xe_oa_stream *stream) static u32 oag_buf_size_select(const struct xe_oa_stream *stream) { return _MASKED_FIELD(OAG_OA_DEBUG_BUF_SIZE_SELECT, - stream->oa_buffer.bo->size > SZ_16M ? + xe_bo_size(stream->oa_buffer.bo) > SZ_16M ? OAG_OA_DEBUG_BUF_SIZE_SELECT : 0); } @@ -1582,7 +1582,7 @@ static long xe_oa_status_locked(struct xe_oa_stream *stream, unsigned long arg) static long xe_oa_info_locked(struct xe_oa_stream *stream, unsigned long arg) { - struct drm_xe_oa_stream_info info = { .oa_buf_size = stream->oa_buffer.bo->size, }; + struct drm_xe_oa_stream_info info = { .oa_buf_size = xe_bo_size(stream->oa_buffer.bo), }; void __user *uaddr = (void __user *)arg; if (copy_to_user(uaddr, &info, sizeof(info))) @@ -1668,7 +1668,7 @@ static int xe_oa_mmap(struct file *file, struct vm_area_struct *vma) } /* Can mmap the entire OA buffer or nothing (no partial OA buffer mmaps) */ - if (vma->vm_end - vma->vm_start != stream->oa_buffer.bo->size) { + if (vma->vm_end - vma->vm_start != xe_bo_size(stream->oa_buffer.bo)) { drm_dbg(&stream->oa->xe->drm, "Wrong mmap size, must be OA buffer size\n"); return -EINVAL; } diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c index 89814b32e585..3614fcf3f088 100644 --- a/drivers/gpu/drm/xe/xe_pci.c +++ b/drivers/gpu/drm/xe/xe_pci.c @@ -38,43 +38,6 @@ enum toggle_d3cold { D3COLD_ENABLE, }; -struct xe_subplatform_desc { - enum xe_subplatform subplatform; - const char *name; - const u16 *pciidlist; -}; - -struct xe_device_desc { - /* Should only ever be set for platforms without GMD_ID */ - const struct xe_ip *pre_gmdid_graphics_ip; - /* Should only ever be set for platforms without GMD_ID */ - const struct xe_ip *pre_gmdid_media_ip; - - const char *platform_name; - const struct xe_subplatform_desc *subplatforms; - - enum xe_platform platform; - - u8 dma_mask_size; - u8 max_remote_tiles:2; - - u8 require_force_probe:1; - u8 is_dgfx:1; - - u8 has_display:1; - u8 has_fan_control:1; - u8 has_heci_gscfi:1; - u8 has_heci_cscfi:1; - u8 has_llc:1; - u8 has_mbx_power_limits:1; - u8 has_pxp:1; - u8 has_sriov:1; - u8 needs_scratch:1; - u8 skip_guc_pc:1; - u8 skip_mtcfg:1; - u8 skip_pcode:1; -}; - __diag_push(); __diag_ignore_all("-Woverride-init", "Allow field overrides in table"); @@ -140,7 +103,6 @@ static const struct xe_graphics_desc graphics_xelpg = { .has_asid = 1, \ .has_atomic_enable_pte_bit = 1, \ .has_flat_ccs = 1, \ - .has_indirect_ring_state = 1, \ .has_range_tlb_invalidation = 1, \ .has_usm = 1, \ .has_64bit_timestamp = 1, \ @@ -184,6 +146,7 @@ static const struct xe_ip graphics_ips[] = { { 2004, "Xe2_LPG", &graphics_xe2 }, { 3000, "Xe3_LPG", &graphics_xe2 }, { 3001, "Xe3_LPG", &graphics_xe2 }, + { 3003, "Xe3_LPG", &graphics_xe2 }, }; /* Pre-GMDID Media IPs */ @@ -196,6 +159,7 @@ static const struct xe_ip media_ips[] = { { 1301, "Xe2_HPM", &media_xelpmp }, { 2000, "Xe2_LPM", &media_xelpmp }, { 3000, "Xe3_LPM", &media_xelpmp }, + { 3002, "Xe3_LPM", &media_xelpmp }, }; static const struct xe_device_desc tgl_desc = { @@ -205,6 +169,7 @@ static const struct xe_device_desc tgl_desc = { .dma_mask_size = 39, .has_display = true, .has_llc = true, + .max_gt_per_tile = 1, .require_force_probe = true, }; @@ -215,6 +180,7 @@ static const struct xe_device_desc rkl_desc = { .dma_mask_size = 39, .has_display = true, .has_llc = true, + .max_gt_per_tile = 1, .require_force_probe = true, }; @@ -227,6 +193,7 @@ static const struct xe_device_desc adl_s_desc = { .dma_mask_size = 39, .has_display = true, .has_llc = true, + .max_gt_per_tile = 1, .require_force_probe = true, .subplatforms = (const struct xe_subplatform_desc[]) { { XE_SUBPLATFORM_ALDERLAKE_S_RPLS, "RPLS", adls_rpls_ids }, @@ -243,6 +210,7 @@ static const struct xe_device_desc adl_p_desc = { .dma_mask_size = 39, .has_display = true, .has_llc = true, + .max_gt_per_tile = 1, .require_force_probe = true, .subplatforms = (const struct xe_subplatform_desc[]) { { XE_SUBPLATFORM_ALDERLAKE_P_RPLU, "RPLU", adlp_rplu_ids }, @@ -257,6 +225,7 @@ static const struct xe_device_desc adl_n_desc = { .dma_mask_size = 39, .has_display = true, .has_llc = true, + .max_gt_per_tile = 1, .require_force_probe = true, }; @@ -270,7 +239,9 @@ static const struct xe_device_desc dg1_desc = { PLATFORM(DG1), .dma_mask_size = 39, .has_display = true, + .has_gsc_nvm = 1, .has_heci_gscfi = 1, + .max_gt_per_tile = 1, .require_force_probe = true, }; @@ -281,6 +252,7 @@ static const u16 dg2_g12_ids[] = { INTEL_DG2_G12_IDS(NOP), 0 }; #define DG2_FEATURES \ DGFX_FEATURES, \ PLATFORM(DG2), \ + .has_gsc_nvm = 1, \ .has_heci_gscfi = 1, \ .subplatforms = (const struct xe_subplatform_desc[]) { \ { XE_SUBPLATFORM_DG2_G10, "G10", dg2_g10_ids }, \ @@ -293,6 +265,7 @@ static const struct xe_device_desc ats_m_desc = { .pre_gmdid_graphics_ip = &graphics_ip_xehpg, .pre_gmdid_media_ip = &media_ip_xehpm, .dma_mask_size = 46, + .max_gt_per_tile = 1, .require_force_probe = true, DG2_FEATURES, @@ -303,6 +276,7 @@ static const struct xe_device_desc dg2_desc = { .pre_gmdid_graphics_ip = &graphics_ip_xehpg, .pre_gmdid_media_ip = &media_ip_xehpm, .dma_mask_size = 46, + .max_gt_per_tile = 1, .require_force_probe = true, DG2_FEATURES, @@ -317,7 +291,9 @@ static const __maybe_unused struct xe_device_desc pvc_desc = { PLATFORM(PVC), .dma_mask_size = 52, .has_display = false, + .has_gsc_nvm = 1, .has_heci_gscfi = 1, + .max_gt_per_tile = 1, .max_remote_tiles = 1, .require_force_probe = true, .has_mbx_power_limits = false, @@ -330,6 +306,7 @@ static const struct xe_device_desc mtl_desc = { .dma_mask_size = 46, .has_display = true, .has_pxp = true, + .max_gt_per_tile = 2, }; static const struct xe_device_desc lnl_desc = { @@ -337,6 +314,7 @@ static const struct xe_device_desc lnl_desc = { .dma_mask_size = 46, .has_display = true, .has_pxp = true, + .max_gt_per_tile = 2, .needs_scratch = true, }; @@ -347,7 +325,9 @@ static const struct xe_device_desc bmg_desc = { .has_display = true, .has_fan_control = true, .has_mbx_power_limits = true, + .has_gsc_nvm = 1, .has_heci_cscfi = 1, + .max_gt_per_tile = 2, .needs_scratch = true, }; @@ -356,7 +336,7 @@ static const struct xe_device_desc ptl_desc = { .dma_mask_size = 46, .has_display = true, .has_sriov = true, - .require_force_probe = true, + .max_gt_per_tile = 2, .needs_scratch = true, }; @@ -590,6 +570,7 @@ static int xe_info_init_early(struct xe_device *xe, xe->info.is_dgfx = desc->is_dgfx; xe->info.has_fan_control = desc->has_fan_control; xe->info.has_mbx_power_limits = desc->has_mbx_power_limits; + xe->info.has_gsc_nvm = desc->has_gsc_nvm; xe->info.has_heci_gscfi = desc->has_heci_gscfi; xe->info.has_heci_cscfi = desc->has_heci_cscfi; xe->info.has_llc = desc->has_llc; @@ -603,6 +584,10 @@ static int xe_info_init_early(struct xe_device *xe, xe->info.probe_display = IS_ENABLED(CONFIG_DRM_XE_DISPLAY) && xe_modparam.probe_display && desc->has_display; + + xe_assert(xe, desc->max_gt_per_tile > 0); + xe_assert(xe, desc->max_gt_per_tile <= XE_MAX_GT_PER_TILE); + xe->info.max_gt_per_tile = desc->max_gt_per_tile; xe->info.tile_count = 1 + desc->max_remote_tiles; err = xe_tile_init_early(xe_device_get_root_tile(xe), xe, 0); @@ -702,10 +687,11 @@ static int xe_info_init(struct xe_device *xe, */ for_each_tile(tile, xe, id) { gt = tile->primary_gt; - gt->info.id = xe->info.gt_count++; gt->info.type = XE_GT_TYPE_MAIN; + gt->info.id = tile->id * xe->info.max_gt_per_tile; gt->info.has_indirect_ring_state = graphics_desc->has_indirect_ring_state; gt->info.engine_mask = graphics_desc->hw_engine_mask; + xe->info.gt_count++; if (MEDIA_VER(xe) < 13 && media_desc) gt->info.engine_mask |= media_desc->hw_engine_mask; @@ -723,17 +709,10 @@ static int xe_info_init(struct xe_device *xe, gt = tile->media_gt; gt->info.type = XE_GT_TYPE_MEDIA; + gt->info.id = tile->id * xe->info.max_gt_per_tile + 1; gt->info.has_indirect_ring_state = media_desc->has_indirect_ring_state; gt->info.engine_mask = media_desc->hw_engine_mask; - - /* - * FIXME: At the moment multi-tile and standalone media are - * mutually exclusive on current platforms. We'll need to - * come up with a better way to number GTs if we ever wind - * up with platforms that support both together. - */ - drm_WARN_ON(&xe->drm, id != 0); - gt->info.id = xe->info.gt_count++; + xe->info.gt_count++; } return 0; diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h index ca6b10d35573..4de6f69ed975 100644 --- a/drivers/gpu/drm/xe/xe_pci_types.h +++ b/drivers/gpu/drm/xe/xe_pci_types.h @@ -8,6 +8,47 @@ #include <linux/types.h> +#include "xe_platform_types.h" + +struct xe_subplatform_desc { + enum xe_subplatform subplatform; + const char *name; + const u16 *pciidlist; +}; + +struct xe_device_desc { + /* Should only ever be set for platforms without GMD_ID */ + const struct xe_ip *pre_gmdid_graphics_ip; + /* Should only ever be set for platforms without GMD_ID */ + const struct xe_ip *pre_gmdid_media_ip; + + const char *platform_name; + const struct xe_subplatform_desc *subplatforms; + + enum xe_platform platform; + + u8 dma_mask_size; + u8 max_remote_tiles:2; + u8 max_gt_per_tile:2; + + u8 require_force_probe:1; + u8 is_dgfx:1; + + u8 has_display:1; + u8 has_fan_control:1; + u8 has_gsc_nvm:1; + u8 has_heci_gscfi:1; + u8 has_heci_cscfi:1; + u8 has_llc:1; + u8 has_mbx_power_limits:1; + u8 has_pxp:1; + u8 has_sriov:1; + u8 needs_scratch:1; + u8 skip_guc_pc:1; + u8 skip_mtcfg:1; + u8 skip_pcode:1; +}; + struct xe_graphics_desc { u8 va_bits; u8 vm_max_level; diff --git a/drivers/gpu/drm/xe/xe_pcode_api.h b/drivers/gpu/drm/xe/xe_pcode_api.h index 0befdea77db1..92bfcba51e19 100644 --- a/drivers/gpu/drm/xe/xe_pcode_api.h +++ b/drivers/gpu/drm/xe/xe_pcode_api.h @@ -50,6 +50,21 @@ #define READ_PL_FROM_FW 0x1 #define READ_PL_FROM_PCODE 0x0 +#define PCODE_LATE_BINDING 0x5C +#define GET_CAPABILITY_STATUS 0x0 +#define V1_FAN_SUPPORTED REG_BIT(0) +#define VR_PARAMS_SUPPORTED REG_BIT(3) +#define V1_FAN_PROVISIONED REG_BIT(16) +#define VR_PARAMS_PROVISIONED REG_BIT(19) +#define GET_VERSION_LOW 0x1 +#define GET_VERSION_HIGH 0x2 +#define MAJOR_VERSION_MASK REG_GENMASK(31, 16) +#define MINOR_VERSION_MASK REG_GENMASK(15, 0) +#define HOTFIX_VERSION_MASK REG_GENMASK(31, 16) +#define BUILD_VERSION_MASK REG_GENMASK(15, 0) +#define FAN_TABLE 1 +#define VR_CONFIG 2 + #define PCODE_FREQUENCY_CONFIG 0x6e /* Frequency Config Sub Commands (param1) */ #define PCODE_MBOX_FC_SC_READ_FUSED_P0 0x0 diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c index ff749edc005b..e279b47ba03b 100644 --- a/drivers/gpu/drm/xe/xe_pm.c +++ b/drivers/gpu/drm/xe/xe_pm.c @@ -19,6 +19,7 @@ #include "xe_ggtt.h" #include "xe_gt.h" #include "xe_guc.h" +#include "xe_i2c.h" #include "xe_irq.h" #include "xe_pcode.h" #include "xe_pxp.h" @@ -134,7 +135,7 @@ int xe_pm_suspend(struct xe_device *xe) /* FIXME: Super racey... */ err = xe_bo_evict_all(xe); if (err) - goto err_pxp; + goto err_display; for_each_gt(gt, xe, id) { err = xe_gt_suspend(gt); @@ -146,12 +147,13 @@ int xe_pm_suspend(struct xe_device *xe) xe_display_pm_suspend_late(xe); + xe_i2c_pm_suspend(xe); + drm_dbg(&xe->drm, "Device suspended\n"); return 0; err_display: xe_display_pm_resume(xe); -err_pxp: xe_pxp_pm_resume(xe->pxp); err: drm_dbg(&xe->drm, "Device suspend failed %d\n", err); @@ -191,6 +193,8 @@ int xe_pm_resume(struct xe_device *xe) if (err) goto err; + xe_i2c_pm_resume(xe, xe->d3cold.allowed); + xe_irq_resume(xe); for_each_gt(gt, xe, id) @@ -488,6 +492,8 @@ int xe_pm_runtime_suspend(struct xe_device *xe) xe_display_pm_runtime_suspend_late(xe); + xe_i2c_pm_suspend(xe); + xe_rpm_lockmap_release(xe); xe_pm_write_callback_task(xe, NULL); return 0; @@ -535,6 +541,8 @@ int xe_pm_runtime_resume(struct xe_device *xe) goto out; } + xe_i2c_pm_resume(xe, xe->d3cold.allowed); + xe_irq_resume(xe); for_each_gt(gt, xe, id) @@ -753,11 +761,13 @@ void xe_pm_assert_unbounded_bridge(struct xe_device *xe) } /** - * xe_pm_set_vram_threshold - Set a vram threshold for allowing/blocking D3Cold + * xe_pm_set_vram_threshold - Set a VRAM threshold for allowing/blocking D3Cold * @xe: xe device instance - * @threshold: VRAM size in bites for the D3cold threshold + * @threshold: VRAM size in MiB for the D3cold threshold * - * Returns 0 for success, negative error code otherwise. + * Return: + * * 0 - success + * * -EINVAL - invalid argument */ int xe_pm_set_vram_threshold(struct xe_device *xe, u32 threshold) { diff --git a/drivers/gpu/drm/xe/xe_pmu.c b/drivers/gpu/drm/xe/xe_pmu.c index 69df0e3520a5..cab51d826345 100644 --- a/drivers/gpu/drm/xe/xe_pmu.c +++ b/drivers/gpu/drm/xe/xe_pmu.c @@ -157,10 +157,13 @@ static bool event_gt_forcewake(struct perf_event *event) return true; } -static bool event_supported(struct xe_pmu *pmu, unsigned int gt, +static bool event_supported(struct xe_pmu *pmu, unsigned int gt_id, unsigned int id) { - if (gt >= XE_MAX_GT_PER_TILE) + struct xe_device *xe = container_of(pmu, typeof(*xe), pmu); + struct xe_gt *gt = xe_device_get_gt(xe, gt_id); + + if (!gt) return false; return id < sizeof(pmu->supported_events) * BITS_PER_BYTE && diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c index e8e1743dcb1e..d517ec9ddcbf 100644 --- a/drivers/gpu/drm/xe/xe_query.c +++ b/drivers/gpu/drm/xe/xe_query.c @@ -141,7 +141,7 @@ query_engine_cycles(struct xe_device *xe, return -EINVAL; eci = &resp.eci; - if (eci->gt_id >= XE_MAX_GT_PER_TILE) + if (eci->gt_id >= xe->info.max_gt_per_tile) return -EINVAL; gt = xe_device_get_gt(xe, eci->gt_id); @@ -368,6 +368,7 @@ static int query_gt_list(struct xe_device *xe, struct drm_xe_device_query *query struct drm_xe_query_gt_list __user *query_ptr = u64_to_user_ptr(query->data); struct drm_xe_query_gt_list *gt_list; + int iter = 0; u8 id; if (query->size == 0) { @@ -385,12 +386,12 @@ static int query_gt_list(struct xe_device *xe, struct drm_xe_device_query *query for_each_gt(gt, xe, id) { if (xe_gt_is_media_type(gt)) - gt_list->gt_list[id].type = DRM_XE_QUERY_GT_TYPE_MEDIA; + gt_list->gt_list[iter].type = DRM_XE_QUERY_GT_TYPE_MEDIA; else - gt_list->gt_list[id].type = DRM_XE_QUERY_GT_TYPE_MAIN; - gt_list->gt_list[id].tile_id = gt_to_tile(gt)->id; - gt_list->gt_list[id].gt_id = gt->info.id; - gt_list->gt_list[id].reference_clock = gt->info.reference_clock; + gt_list->gt_list[iter].type = DRM_XE_QUERY_GT_TYPE_MAIN; + gt_list->gt_list[iter].tile_id = gt_to_tile(gt)->id; + gt_list->gt_list[iter].gt_id = gt->info.id; + gt_list->gt_list[iter].reference_clock = gt->info.reference_clock; /* * The mem_regions indexes in the mask below need to * directly identify the struct @@ -406,19 +407,21 @@ static int query_gt_list(struct xe_device *xe, struct drm_xe_device_query *query * assumption. */ if (!IS_DGFX(xe)) - gt_list->gt_list[id].near_mem_regions = 0x1; + gt_list->gt_list[iter].near_mem_regions = 0x1; else - gt_list->gt_list[id].near_mem_regions = + gt_list->gt_list[iter].near_mem_regions = BIT(gt_to_tile(gt)->id) << 1; - gt_list->gt_list[id].far_mem_regions = xe->info.mem_region_mask ^ - gt_list->gt_list[id].near_mem_regions; + gt_list->gt_list[iter].far_mem_regions = xe->info.mem_region_mask ^ + gt_list->gt_list[iter].near_mem_regions; - gt_list->gt_list[id].ip_ver_major = + gt_list->gt_list[iter].ip_ver_major = REG_FIELD_GET(GMD_ID_ARCH_MASK, gt->info.gmdid); - gt_list->gt_list[id].ip_ver_minor = + gt_list->gt_list[iter].ip_ver_minor = REG_FIELD_GET(GMD_ID_RELEASE_MASK, gt->info.gmdid); - gt_list->gt_list[id].ip_ver_rev = + gt_list->gt_list[iter].ip_ver_rev = REG_FIELD_GET(GMD_ID_REVID, gt->info.gmdid); + + iter++; } if (copy_to_user(query_ptr, gt_list, size)) { diff --git a/drivers/gpu/drm/xe/xe_shrinker.c b/drivers/gpu/drm/xe/xe_shrinker.c index 125c836e0ee4..1c3c04d52f55 100644 --- a/drivers/gpu/drm/xe/xe_shrinker.c +++ b/drivers/gpu/drm/xe/xe_shrinker.c @@ -66,11 +66,15 @@ static s64 xe_shrinker_walk(struct xe_device *xe, struct ttm_resource_manager *man = ttm_manager_type(&xe->ttm, mem_type); struct ttm_bo_lru_cursor curs; struct ttm_buffer_object *ttm_bo; + struct ttm_lru_walk_arg arg = { + .ctx = ctx, + .trylock_only = true, + }; if (!man || !man->use_tt) continue; - ttm_bo_lru_for_each_reserved_guarded(&curs, man, ctx, ttm_bo) { + ttm_bo_lru_for_each_reserved_guarded(&curs, man, &arg, ttm_bo) { if (!ttm_bo_shrink_suitable(ttm_bo, ctx)) continue; @@ -82,6 +86,8 @@ static s64 xe_shrinker_walk(struct xe_device *xe, if (*scanned >= to_scan) break; } + /* Trylocks should never error, just fail. */ + xe_assert(xe, !IS_ERR(ttm_bo)); } return freed; diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c index 6526fe450e55..26e243c28994 100644 --- a/drivers/gpu/drm/xe/xe_sriov_vf.c +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c @@ -147,127 +147,113 @@ void xe_sriov_vf_init_early(struct xe_device *xe) xe_sriov_info(xe, "migration not supported by this module version\n"); } -/** - * vf_post_migration_requery_guc - Re-query GuC for current VF provisioning. +static bool gt_vf_post_migration_needed(struct xe_gt *gt) +{ + return test_bit(gt->info.id, >_to_xe(gt)->sriov.vf.migration.gt_flags); +} + +/* + * Notify GuCs marked in flags about resource fixups apply finished. * @xe: the &xe_device struct instance - * - * After migration, we need to re-query all VF configuration to make sure - * they match previous provisioning. Note that most of VF provisioning - * shall be the same, except GGTT range, since GGTT is not virtualized per-VF. - * - * Returns: 0 if the operation completed successfully, or a negative error - * code otherwise. + * @gt_flags: flags marking to which GTs the notification shall be sent */ -static int vf_post_migration_requery_guc(struct xe_device *xe) +static int vf_post_migration_notify_resfix_done(struct xe_device *xe, unsigned long gt_flags) { struct xe_gt *gt; unsigned int id; - int err, ret = 0; + int err = 0; for_each_gt(gt, xe, id) { - err = xe_gt_sriov_vf_query_config(gt); - ret = ret ?: err; + if (!test_bit(id, >_flags)) + continue; + /* skip asking GuC for RESFIX exit if new recovery request arrived */ + if (gt_vf_post_migration_needed(gt)) + continue; + err = xe_gt_sriov_vf_notify_resfix_done(gt); + if (err) + break; + clear_bit(id, >_flags); } - return ret; + if (gt_flags && !err) + drm_dbg(&xe->drm, "another recovery imminent, skipped some notifications\n"); + return err; } -static void vf_post_migration_fixup_ctb(struct xe_device *xe) +static int vf_get_next_migrated_gt_id(struct xe_device *xe) { struct xe_gt *gt; unsigned int id; - xe_assert(xe, IS_SRIOV_VF(xe)); - for_each_gt(gt, xe, id) { - s32 shift = xe_gt_sriov_vf_ggtt_shift(gt); - - xe_guc_ct_fixup_messages_with_ggtt(>->uc.guc.ct, shift); + if (test_and_clear_bit(id, &xe->sriov.vf.migration.gt_flags)) + return id; } + return -1; } -/* - * vf_post_migration_imminent - Check if post-restore recovery is coming. - * @xe: the &xe_device struct instance +/** + * Perform post-migration fixups on a single GT. * - * Return: True if migration recovery worker will soon be running. Any worker currently - * executing does not affect the result. + * After migration, GuC needs to be re-queried for VF configuration to check + * if it matches previous provisioning. Most of VF provisioning shall be the + * same, except GGTT range, since GGTT is not virtualized per-VF. If GGTT + * range has changed, we have to perform fixups - shift all GGTT references + * used anywhere within the driver. After the fixups in this function succeed, + * it is allowed to ask the GuC bound to this GT to continue normal operation. + * + * Returns: 0 if the operation completed successfully, or a negative error + * code otherwise. */ -static bool vf_post_migration_imminent(struct xe_device *xe) +static int gt_vf_post_migration_fixups(struct xe_gt *gt) { - return xe->sriov.vf.migration.gt_flags != 0 || - work_pending(&xe->sriov.vf.migration.worker); -} - -static bool vf_post_migration_fixup_ggtt_nodes(struct xe_device *xe) -{ - bool need_fixups = false; - struct xe_tile *tile; - unsigned int id; - - for_each_tile(tile, xe, id) { - struct xe_gt *gt = tile->primary_gt; - s64 shift; - - shift = xe_gt_sriov_vf_ggtt_shift(gt); - if (shift) { - need_fixups = true; - xe_tile_sriov_vf_fixup_ggtt_nodes(tile, shift); - } - } - return need_fixups; -} + s64 shift; + int err; -/* - * Notify all GuCs about resource fixups apply finished. - */ -static void vf_post_migration_notify_resfix_done(struct xe_device *xe) -{ - struct xe_gt *gt; - unsigned int id; + err = xe_gt_sriov_vf_query_config(gt); + if (err) + return err; - for_each_gt(gt, xe, id) { - if (vf_post_migration_imminent(xe)) - goto skip; - xe_gt_sriov_vf_notify_resfix_done(gt); + shift = xe_gt_sriov_vf_ggtt_shift(gt); + if (shift) { + xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift); + /* FIXME: add the recovery steps */ + xe_guc_ct_fixup_messages_with_ggtt(>->uc.guc.ct, shift); } - return; - -skip: - drm_dbg(&xe->drm, "another recovery imminent, skipping notifications\n"); + return 0; } static void vf_post_migration_recovery(struct xe_device *xe) { - bool need_fixups; - int err; + unsigned long fixed_gts = 0; + int id, err; drm_dbg(&xe->drm, "migration recovery in progress\n"); xe_pm_runtime_get(xe); - err = vf_post_migration_requery_guc(xe); - if (vf_post_migration_imminent(xe)) - goto defer; - if (unlikely(err)) - goto fail; + if (!vf_migration_supported(xe)) { xe_sriov_err(xe, "migration not supported by this module version\n"); err = -ENOTRECOVERABLE; goto fail; } - need_fixups = vf_post_migration_fixup_ggtt_nodes(xe); - /* FIXME: add the recovery steps */ - if (need_fixups) - vf_post_migration_fixup_ctb(xe); + while (id = vf_get_next_migrated_gt_id(xe), id >= 0) { + struct xe_gt *gt = xe_device_get_gt(xe, id); + + err = gt_vf_post_migration_fixups(gt); + if (err) + goto fail; + + set_bit(id, &fixed_gts); + } + + err = vf_post_migration_notify_resfix_done(xe, fixed_gts); + if (err) + goto fail; - vf_post_migration_notify_resfix_done(xe); xe_pm_runtime_put(xe); drm_notice(&xe->drm, "migration recovery ended\n"); return; -defer: - xe_pm_runtime_put(xe); - drm_dbg(&xe->drm, "migration recovery deferred\n"); - return; fail: xe_pm_runtime_put(xe); drm_err(&xe->drm, "migration recovery failed (%pe)\n", ERR_PTR(err)); @@ -282,18 +268,23 @@ static void migration_worker_func(struct work_struct *w) vf_post_migration_recovery(xe); } -static bool vf_ready_to_recovery_on_all_gts(struct xe_device *xe) +/* + * Check if post-restore recovery is coming on any of GTs. + * @xe: the &xe_device struct instance + * + * Return: True if migration recovery worker will soon be running. Any worker currently + * executing does not affect the result. + */ +static bool vf_ready_to_recovery_on_any_gts(struct xe_device *xe) { struct xe_gt *gt; unsigned int id; for_each_gt(gt, xe, id) { - if (!test_bit(id, &xe->sriov.vf.migration.gt_flags)) { - xe_gt_sriov_dbg_verbose(gt, "still not ready to recover\n"); - return false; - } + if (test_bit(id, &xe->sriov.vf.migration.gt_flags)) + return true; } - return true; + return false; } /** @@ -308,13 +299,9 @@ void xe_sriov_vf_start_migration_recovery(struct xe_device *xe) xe_assert(xe, IS_SRIOV_VF(xe)); - if (!vf_ready_to_recovery_on_all_gts(xe)) + if (!vf_ready_to_recovery_on_any_gts(xe)) return; - WRITE_ONCE(xe->sriov.vf.migration.gt_flags, 0); - /* Ensure other threads see that no flags are set now. */ - smp_mb(); - started = queue_work(xe->sriov.wq, &xe->sriov.vf.migration.worker); drm_info(&xe->drm, "VF migration recovery %s\n", started ? "scheduled" : "already in progress"); diff --git a/drivers/gpu/drm/xe/xe_step.c b/drivers/gpu/drm/xe/xe_step.c index c77b5c317fa0..10e88f2c9615 100644 --- a/drivers/gpu/drm/xe/xe_step.c +++ b/drivers/gpu/drm/xe/xe_step.c @@ -5,6 +5,7 @@ #include "xe_step.h" +#include <kunit/visibility.h> #include <linux/bitfield.h> #include "xe_device.h" @@ -255,3 +256,4 @@ const char *xe_step_name(enum xe_step step) return "**"; } } +EXPORT_SYMBOL_IF_KUNIT(xe_step_name); diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c index 1f710b3fc599..41705f5d52e3 100644 --- a/drivers/gpu/drm/xe/xe_survivability_mode.c +++ b/drivers/gpu/drm/xe/xe_survivability_mode.c @@ -14,6 +14,7 @@ #include "xe_device.h" #include "xe_gt.h" #include "xe_heci_gsc.h" +#include "xe_i2c.h" #include "xe_mmio.h" #include "xe_pcode_api.h" #include "xe_vsec.h" @@ -173,20 +174,22 @@ static int enable_survivability_mode(struct pci_dev *pdev) survivability->mode = true; ret = xe_heci_gsc_init(xe); - if (ret) { - /* - * But if it fails, device can't enter survivability - * so move it back for correct error handling - */ - survivability->mode = false; - return ret; - } + if (ret) + goto err; xe_vsec_init(xe); + ret = xe_i2c_probe(xe); + if (ret) + goto err; + dev_err(dev, "In Survivability Mode\n"); return 0; + +err: + survivability->mode = false; + return ret; } /** diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index 26418e9bdff0..a7ff5975873f 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -3,13 +3,17 @@ * Copyright © 2024 Intel Corporation */ +#include <drm/drm_drv.h> + #include "xe_bo.h" #include "xe_gt_stats.h" #include "xe_gt_tlb_invalidation.h" #include "xe_migrate.h" #include "xe_module.h" +#include "xe_pm.h" #include "xe_pt.h" #include "xe_svm.h" +#include "xe_tile.h" #include "xe_ttm_vram_mgr.h" #include "xe_vm.h" #include "xe_vm_types.h" @@ -295,7 +299,7 @@ static void xe_svm_garbage_collector_work_func(struct work_struct *w) up_write(&vm->lock); } -#if IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR) +#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) static struct xe_vram_region *page_to_vr(struct page *page) { @@ -483,16 +487,18 @@ static int xe_svm_copy_to_ram(struct page **pages, dma_addr_t *dma_addr, return xe_svm_copy(pages, dma_addr, npages, XE_SVM_COPY_TO_SRAM); } -static struct xe_bo *to_xe_bo(struct drm_gpusvm_devmem *devmem_allocation) +static struct xe_bo *to_xe_bo(struct drm_pagemap_devmem *devmem_allocation) { return container_of(devmem_allocation, struct xe_bo, devmem_allocation); } -static void xe_svm_devmem_release(struct drm_gpusvm_devmem *devmem_allocation) +static void xe_svm_devmem_release(struct drm_pagemap_devmem *devmem_allocation) { struct xe_bo *bo = to_xe_bo(devmem_allocation); + struct xe_device *xe = xe_bo_device(bo); xe_bo_put_async(bo); + xe_pm_runtime_put(xe); } static u64 block_offset_to_pfn(struct xe_vram_region *vr, u64 offset) @@ -505,7 +511,7 @@ static struct drm_buddy *tile_to_buddy(struct xe_tile *tile) return &tile->mem.vram.ttm.mm; } -static int xe_svm_populate_devmem_pfn(struct drm_gpusvm_devmem *devmem_allocation, +static int xe_svm_populate_devmem_pfn(struct drm_pagemap_devmem *devmem_allocation, unsigned long npages, unsigned long *pfn) { struct xe_bo *bo = to_xe_bo(devmem_allocation); @@ -528,7 +534,7 @@ static int xe_svm_populate_devmem_pfn(struct drm_gpusvm_devmem *devmem_allocatio return 0; } -static const struct drm_gpusvm_devmem_ops gpusvm_devmem_ops = { +static const struct drm_pagemap_devmem_ops dpagemap_devmem_ops = { .devmem_release = xe_svm_devmem_release, .populate_devmem_pfn = xe_svm_populate_devmem_pfn, .copy_to_devmem = xe_svm_copy_to_devmem, @@ -676,75 +682,69 @@ u64 xe_svm_find_vma_start(struct xe_vm *vm, u64 start, u64 end, struct xe_vma *v min(end, xe_vma_end(vma))); } -#if IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR) +#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) static struct xe_vram_region *tile_to_vr(struct xe_tile *tile) { return &tile->mem.vram; } -/** - * xe_svm_alloc_vram()- Allocate device memory pages for range, - * migrating existing data. - * @vm: The VM. - * @tile: tile to allocate vram from - * @range: SVM range - * @ctx: DRM GPU SVM context - * - * Return: 0 on success, error code on failure. - */ -int xe_svm_alloc_vram(struct xe_vm *vm, struct xe_tile *tile, - struct xe_svm_range *range, - const struct drm_gpusvm_ctx *ctx) +static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, + unsigned long start, unsigned long end, + struct mm_struct *mm, + unsigned long timeslice_ms) { - struct mm_struct *mm = vm->svm.gpusvm.mm; + struct xe_tile *tile = container_of(dpagemap, typeof(*tile), mem.vram.dpagemap); + struct xe_device *xe = tile_to_xe(tile); + struct device *dev = xe->drm.dev; struct xe_vram_region *vr = tile_to_vr(tile); struct drm_buddy_block *block; struct list_head *blocks; struct xe_bo *bo; - ktime_t end = 0; - int err; + ktime_t time_end = 0; + int err, idx; - range_debug(range, "ALLOCATE VRAM"); + if (!drm_dev_enter(&xe->drm, &idx)) + return -ENODEV; - if (!mmget_not_zero(mm)) - return -EFAULT; - mmap_read_lock(mm); + xe_pm_runtime_get(xe); -retry: - bo = xe_bo_create_locked(tile_to_xe(tile), NULL, NULL, - xe_svm_range_size(range), + retry: + bo = xe_bo_create_locked(tile_to_xe(tile), NULL, NULL, end - start, ttm_bo_type_device, XE_BO_FLAG_VRAM_IF_DGFX(tile) | XE_BO_FLAG_CPU_ADDR_MIRROR); if (IS_ERR(bo)) { err = PTR_ERR(bo); - if (xe_vm_validate_should_retry(NULL, err, &end)) + if (xe_vm_validate_should_retry(NULL, err, &time_end)) goto retry; - goto unlock; + goto out_pm_put; } - drm_gpusvm_devmem_init(&bo->devmem_allocation, - vm->xe->drm.dev, mm, - &gpusvm_devmem_ops, - &tile->mem.vram.dpagemap, - xe_svm_range_size(range)); + drm_pagemap_devmem_init(&bo->devmem_allocation, dev, mm, + &dpagemap_devmem_ops, + &tile->mem.vram.dpagemap, + end - start); blocks = &to_xe_ttm_vram_mgr_resource(bo->ttm.resource)->blocks; list_for_each_entry(block, blocks, link) block->private = vr; xe_bo_get(bo); - err = drm_gpusvm_migrate_to_devmem(&vm->svm.gpusvm, &range->base, - &bo->devmem_allocation, ctx); + + /* Ensure the device has a pm ref while there are device pages active. */ + xe_pm_runtime_get_noresume(xe); + err = drm_pagemap_migrate_to_devmem(&bo->devmem_allocation, mm, + start, end, timeslice_ms, + xe_svm_devm_owner(xe)); if (err) xe_svm_devmem_release(&bo->devmem_allocation); xe_bo_unlock(bo); xe_bo_put(bo); -unlock: - mmap_read_unlock(mm); - mmput(mm); +out_pm_put: + xe_pm_runtime_put(xe); + drm_dev_exit(idx); return err; } @@ -810,13 +810,13 @@ int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma, struct drm_gpusvm_ctx ctx = { .read_only = xe_vma_read_only(vma), .devmem_possible = IS_DGFX(vm->xe) && - IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR), + IS_ENABLED(CONFIG_DRM_XE_PAGEMAP), .check_pages_threshold = IS_DGFX(vm->xe) && - IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR) ? SZ_64K : 0, + IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) ? SZ_64K : 0, .devmem_only = atomic && IS_DGFX(vm->xe) && - IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR), + IS_ENABLED(CONFIG_DRM_XE_PAGEMAP), .timeslice_ms = atomic && IS_DGFX(vm->xe) && - IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR) ? + IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) ? vm->xe->atomic_svm_timeslice_ms : 0, }; struct xe_svm_range *range; @@ -852,7 +852,7 @@ retry: if (--migrate_try_count >= 0 && xe_svm_range_needs_migrate_to_vram(range, vma, IS_DGFX(vm->xe))) { - err = xe_svm_alloc_vram(vm, tile, range, &ctx); + err = xe_svm_alloc_vram(tile, range, &ctx); ctx.timeslice_ms <<= 1; /* Double timeslice if we have to retry */ if (err) { if (migrate_try_count || !ctx.devmem_only) { @@ -944,7 +944,7 @@ bool xe_svm_has_mapping(struct xe_vm *vm, u64 start, u64 end) */ int xe_svm_bo_evict(struct xe_bo *bo) { - return drm_gpusvm_evict_to_ram(&bo->devmem_allocation); + return drm_pagemap_evict_to_ram(&bo->devmem_allocation); } /** @@ -997,7 +997,31 @@ int xe_svm_range_get_pages(struct xe_vm *vm, struct xe_svm_range *range, return err; } -#if IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR) +#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) + +/** + * xe_svm_alloc_vram()- Allocate device memory pages for range, + * migrating existing data. + * @tile: tile to allocate vram from + * @range: SVM range + * @ctx: DRM GPU SVM context + * + * Return: 0 on success, error code on failure. + */ +int xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range *range, + const struct drm_gpusvm_ctx *ctx) +{ + struct drm_pagemap *dpagemap; + + xe_assert(tile_to_xe(tile), range->base.flags.migrate_devmem); + range_debug(range, "ALLOCATE VRAM"); + + dpagemap = xe_tile_local_pagemap(tile); + return drm_pagemap_populate_mm(dpagemap, xe_svm_range_start(range), + xe_svm_range_end(range), + range->base.gpusvm->mm, + ctx->timeslice_ms); +} static struct drm_pagemap_device_addr xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap, @@ -1023,6 +1047,7 @@ xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap, static const struct drm_pagemap_ops xe_drm_pagemap_ops = { .device_map = xe_drm_pagemap_device_map, + .populate_mm = xe_drm_pagemap_populate_mm, }; /** @@ -1054,7 +1079,7 @@ int xe_devm_add(struct xe_tile *tile, struct xe_vram_region *vr) vr->pagemap.range.start = res->start; vr->pagemap.range.end = res->end; vr->pagemap.nr_range = 1; - vr->pagemap.ops = drm_gpusvm_pagemap_ops_get(); + vr->pagemap.ops = drm_pagemap_pagemap_ops_get(); vr->pagemap.owner = xe_svm_devm_owner(xe); addr = devm_memremap_pages(dev, &vr->pagemap); @@ -1075,7 +1100,7 @@ int xe_devm_add(struct xe_tile *tile, struct xe_vram_region *vr) return 0; } #else -int xe_svm_alloc_vram(struct xe_vm *vm, struct xe_tile *tile, +int xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range *range, const struct drm_gpusvm_ctx *ctx) { diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h index 19ce4f2754a7..da9a69ea0bb1 100644 --- a/drivers/gpu/drm/xe/xe_svm.h +++ b/drivers/gpu/drm/xe/xe_svm.h @@ -70,8 +70,7 @@ int xe_svm_bo_evict(struct xe_bo *bo); void xe_svm_range_debug(struct xe_svm_range *range, const char *operation); -int xe_svm_alloc_vram(struct xe_vm *vm, struct xe_tile *tile, - struct xe_svm_range *range, +int xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range *range, const struct drm_gpusvm_ctx *ctx); struct xe_svm_range *xe_svm_range_find_or_insert(struct xe_vm *vm, u64 addr, @@ -237,10 +236,9 @@ void xe_svm_range_debug(struct xe_svm_range *range, const char *operation) { } -static inline -int xe_svm_alloc_vram(struct xe_vm *vm, struct xe_tile *tile, - struct xe_svm_range *range, - const struct drm_gpusvm_ctx *ctx) +static inline int +xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range *range, + const struct drm_gpusvm_ctx *ctx) { return -EOPNOTSUPP; } diff --git a/drivers/gpu/drm/xe/xe_tile.c b/drivers/gpu/drm/xe/xe_tile.c index 672faa0b67f1..86e9811e60ba 100644 --- a/drivers/gpu/drm/xe/xe_tile.c +++ b/drivers/gpu/drm/xe/xe_tile.c @@ -10,6 +10,7 @@ #include "xe_device.h" #include "xe_ggtt.h" #include "xe_gt.h" +#include "xe_memirq.h" #include "xe_migrate.h" #include "xe_pcode.h" #include "xe_sa.h" @@ -174,6 +175,12 @@ int xe_tile_init_noalloc(struct xe_tile *tile) int xe_tile_init(struct xe_tile *tile) { + int err; + + err = xe_memirq_init(&tile->memirq); + if (err) + return err; + tile->mem.kernel_bb_pool = xe_sa_bo_manager_init(tile, SZ_1M, 16); if (IS_ERR(tile->mem.kernel_bb_pool)) return PTR_ERR(tile->mem.kernel_bb_pool); diff --git a/drivers/gpu/drm/xe/xe_tile.h b/drivers/gpu/drm/xe/xe_tile.h index eb939316d55b..066a3d0cea79 100644 --- a/drivers/gpu/drm/xe/xe_tile.h +++ b/drivers/gpu/drm/xe/xe_tile.h @@ -16,4 +16,15 @@ int xe_tile_init(struct xe_tile *tile); void xe_tile_migrate_wait(struct xe_tile *tile); +#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) +static inline struct drm_pagemap *xe_tile_local_pagemap(struct xe_tile *tile) +{ + return &tile->mem.vram.dpagemap; +} +#else +static inline struct drm_pagemap *xe_tile_local_pagemap(struct xe_tile *tile) +{ + return NULL; +} +#endif #endif diff --git a/drivers/gpu/drm/xe/xe_trace_bo.h b/drivers/gpu/drm/xe/xe_trace_bo.h index ccebd5f0878e..86323cf3be2c 100644 --- a/drivers/gpu/drm/xe/xe_trace_bo.h +++ b/drivers/gpu/drm/xe/xe_trace_bo.h @@ -33,7 +33,7 @@ DECLARE_EVENT_CLASS(xe_bo, TP_fast_assign( __assign_str(dev); - __entry->size = bo->size; + __entry->size = xe_bo_size(bo); __entry->flags = bo->flags; __entry->vm = bo->vm; ), @@ -73,7 +73,7 @@ TRACE_EVENT(xe_bo_move, TP_fast_assign( __entry->bo = bo; - __entry->size = bo->size; + __entry->size = xe_bo_size(bo); __assign_str(new_placement_name); __assign_str(old_placement_name); __assign_str(device_id); diff --git a/drivers/gpu/drm/xe/xe_uc.c b/drivers/gpu/drm/xe/xe_uc.c index 3a8751a8b92d..3e0c3af235f2 100644 --- a/drivers/gpu/drm/xe/xe_uc.c +++ b/drivers/gpu/drm/xe/xe_uc.c @@ -33,6 +33,22 @@ uc_to_xe(struct xe_uc *uc) } /* Should be called once at driver load only */ +int xe_uc_init_noalloc(struct xe_uc *uc) +{ + int ret; + + ret = xe_guc_init_noalloc(&uc->guc); + if (ret) + goto err; + + /* HuC and GSC have no early dependencies and will be initialized during xe_uc_init(). */ + return 0; + +err: + xe_gt_err(uc_to_gt(uc), "Failed to early initialize uC (%pe)\n", ERR_PTR(ret)); + return ret; +} + int xe_uc_init(struct xe_uc *uc) { int ret; @@ -56,15 +72,17 @@ int xe_uc_init(struct xe_uc *uc) if (!xe_device_uc_enabled(uc_to_xe(uc))) return 0; - if (IS_SRIOV_VF(uc_to_xe(uc))) - return 0; + if (!IS_SRIOV_VF(uc_to_xe(uc))) { + ret = xe_wopcm_init(&uc->wopcm); + if (ret) + goto err; + } - ret = xe_wopcm_init(&uc->wopcm); + ret = xe_guc_min_load_for_hwconfig(&uc->guc); if (ret) goto err; return 0; - err: xe_gt_err(uc_to_gt(uc), "Failed to initialize uC (%pe)\n", ERR_PTR(ret)); return ret; @@ -126,28 +144,7 @@ int xe_uc_sanitize_reset(struct xe_uc *uc) return uc_reset(uc); } -/** - * xe_uc_init_hwconfig - minimally init Uc, read and parse hwconfig - * @uc: The UC object - * - * Return: 0 on success, negative error code on error. - */ -int xe_uc_init_hwconfig(struct xe_uc *uc) -{ - int ret; - - /* GuC submission not enabled, nothing to do */ - if (!xe_device_uc_enabled(uc_to_xe(uc))) - return 0; - - ret = xe_guc_min_load_for_hwconfig(&uc->guc); - if (ret) - return ret; - - return 0; -} - -static int vf_uc_init_hw(struct xe_uc *uc) +static int vf_uc_load_hw(struct xe_uc *uc) { int err; @@ -161,22 +158,30 @@ static int vf_uc_init_hw(struct xe_uc *uc) err = xe_gt_sriov_vf_connect(uc_to_gt(uc)); if (err) - return err; + goto err_out; uc->guc.submission_state.enabled = true; - err = xe_gt_record_default_lrcs(uc_to_gt(uc)); + err = xe_guc_opt_in_features_enable(&uc->guc); if (err) return err; + err = xe_gt_record_default_lrcs(uc_to_gt(uc)); + if (err) + goto err_out; + return 0; + +err_out: + xe_guc_sanitize(&uc->guc); + return err; } /* * Should be called during driver load, after every GT reset, and after every * suspend to reload / auth the firmwares. */ -int xe_uc_init_hw(struct xe_uc *uc) +int xe_uc_load_hw(struct xe_uc *uc) { int ret; @@ -185,7 +190,7 @@ int xe_uc_init_hw(struct xe_uc *uc) return 0; if (IS_SRIOV_VF(uc_to_xe(uc))) - return vf_uc_init_hw(uc); + return vf_uc_load_hw(uc); ret = xe_huc_upload(&uc->huc); if (ret) @@ -201,15 +206,15 @@ int xe_uc_init_hw(struct xe_uc *uc) ret = xe_gt_record_default_lrcs(uc_to_gt(uc)); if (ret) - return ret; + goto err_out; ret = xe_guc_post_load_init(&uc->guc); if (ret) - return ret; + goto err_out; ret = xe_guc_pc_start(&uc->guc.pc); if (ret) - return ret; + goto err_out; xe_guc_engine_activity_enable_stats(&uc->guc); @@ -221,11 +226,10 @@ int xe_uc_init_hw(struct xe_uc *uc) xe_gsc_load_start(&uc->gsc); return 0; -} -int xe_uc_fini_hw(struct xe_uc *uc) -{ - return xe_uc_sanitize_reset(uc); +err_out: + xe_guc_sanitize(&uc->guc); + return ret; } int xe_uc_reset_prepare(struct xe_uc *uc) diff --git a/drivers/gpu/drm/xe/xe_uc.h b/drivers/gpu/drm/xe/xe_uc.h index c23e6f5e2514..21c9306098cf 100644 --- a/drivers/gpu/drm/xe/xe_uc.h +++ b/drivers/gpu/drm/xe/xe_uc.h @@ -8,11 +8,10 @@ struct xe_uc; +int xe_uc_init_noalloc(struct xe_uc *uc); int xe_uc_init(struct xe_uc *uc); -int xe_uc_init_hwconfig(struct xe_uc *uc); int xe_uc_init_post_hwconfig(struct xe_uc *uc); -int xe_uc_init_hw(struct xe_uc *uc); -int xe_uc_fini_hw(struct xe_uc *uc); +int xe_uc_load_hw(struct xe_uc *uc); void xe_uc_gucrc_disable(struct xe_uc *uc); int xe_uc_reset_prepare(struct xe_uc *uc); void xe_uc_stop_prepare(struct xe_uc *uc); diff --git a/drivers/gpu/drm/xe/xe_uc_fw.c b/drivers/gpu/drm/xe/xe_uc_fw.c index 6d0869518652..9bbdde604923 100644 --- a/drivers/gpu/drm/xe/xe_uc_fw.c +++ b/drivers/gpu/drm/xe/xe_uc_fw.c @@ -115,10 +115,11 @@ struct fw_blobs_by_type { #define XE_GT_TYPE_ANY XE_GT_TYPE_UNINITIALIZED #define XE_GUC_FIRMWARE_DEFS(fw_def, mmp_ver, major_ver) \ - fw_def(BATTLEMAGE, GT_TYPE_ANY, major_ver(xe, guc, bmg, 70, 44, 1)) \ - fw_def(LUNARLAKE, GT_TYPE_ANY, major_ver(xe, guc, lnl, 70, 44, 1)) \ + fw_def(PANTHERLAKE, GT_TYPE_ANY, major_ver(xe, guc, ptl, 70, 47, 0)) \ + fw_def(BATTLEMAGE, GT_TYPE_ANY, major_ver(xe, guc, bmg, 70, 45, 2)) \ + fw_def(LUNARLAKE, GT_TYPE_ANY, major_ver(xe, guc, lnl, 70, 45, 2)) \ fw_def(METEORLAKE, GT_TYPE_ANY, major_ver(i915, guc, mtl, 70, 44, 1)) \ - fw_def(DG2, GT_TYPE_ANY, major_ver(i915, guc, dg2, 70, 44, 1)) \ + fw_def(DG2, GT_TYPE_ANY, major_ver(i915, guc, dg2, 70, 45, 2)) \ fw_def(DG1, GT_TYPE_ANY, major_ver(i915, guc, dg1, 70, 44, 1)) \ fw_def(ALDERLAKE_N, GT_TYPE_ANY, major_ver(i915, guc, tgl, 70, 44, 1)) \ fw_def(ALDERLAKE_P, GT_TYPE_ANY, major_ver(i915, guc, adlp, 70, 44, 1)) \ @@ -127,6 +128,7 @@ struct fw_blobs_by_type { fw_def(TIGERLAKE, GT_TYPE_ANY, major_ver(i915, guc, tgl, 70, 44, 1)) #define XE_HUC_FIRMWARE_DEFS(fw_def, mmp_ver, no_ver) \ + fw_def(PANTHERLAKE, GT_TYPE_ANY, no_ver(xe, huc, ptl)) \ fw_def(BATTLEMAGE, GT_TYPE_ANY, no_ver(xe, huc, bmg)) \ fw_def(LUNARLAKE, GT_TYPE_ANY, no_ver(xe, huc, lnl)) \ fw_def(METEORLAKE, GT_TYPE_ANY, no_ver(i915, huc_gsc, mtl)) \ diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index 04d1a43b81e3..e875ea4658a9 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -2913,7 +2913,7 @@ static int prefetch_ranges(struct xe_vm *vm, struct xe_vma_op *op) if (xe_svm_range_needs_migrate_to_vram(svm_range, vma, region)) { tile = &vm->xe->tiles[region_to_mem_type[region] - XE_PL_VRAM0]; - err = xe_svm_alloc_vram(vm, tile, svm_range, &ctx); + err = xe_svm_alloc_vram(tile, svm_range, &ctx); if (err) { drm_dbg(&vm->xe->drm, "VRAM allocation failed, retry from userspace, asid=%u, gpusvm=%p, errno=%pe\n", vm->usm.asid, &vm->svm.gpusvm, ERR_PTR(err)); @@ -3466,9 +3466,9 @@ static int xe_vm_bind_ioctl_validate_bo(struct xe_device *xe, struct xe_bo *bo, { u16 coh_mode; - if (XE_IOCTL_DBG(xe, range > bo->size) || + if (XE_IOCTL_DBG(xe, range > xe_bo_size(bo)) || XE_IOCTL_DBG(xe, obj_offset > - bo->size - range)) { + xe_bo_size(bo) - range)) { return -EINVAL; } @@ -3771,7 +3771,7 @@ struct dma_fence *xe_vm_bind_kernel_bo(struct xe_vm *vm, struct xe_bo *bo, xe_vma_ops_init(&vops, vm, q, NULL, 0); - ops = vm_bind_ioctl_ops_create(vm, &vops, bo, 0, addr, bo->size, + ops = vm_bind_ioctl_ops_create(vm, &vops, bo, 0, addr, xe_bo_size(bo), DRM_XE_VM_BIND_OP_MAP, 0, 0, vm->xe->pat.idx[cache_lvl]); if (IS_ERR(ops)) { diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c index 4a76de391abb..f51218a7a580 100644 --- a/drivers/gpu/drm/xe/xe_wa.c +++ b/drivers/gpu/drm/xe/xe_wa.c @@ -285,6 +285,18 @@ static const struct xe_rtp_entry_sr gt_was[] = { XE_RTP_ACTIONS(SET(VDBOX_CGCTL3F10(0), IECPUNIT_CLKGATE_DIS)), XE_RTP_ENTRY_FLAG(FOREACH_ENGINE), }, + { XE_RTP_NAME("16021865536"), + XE_RTP_RULES(MEDIA_VERSION(3002), + ENGINE_CLASS(VIDEO_DECODE)), + XE_RTP_ACTIONS(SET(VDBOX_CGCTL3F10(0), IECPUNIT_CLKGATE_DIS)), + XE_RTP_ENTRY_FLAG(FOREACH_ENGINE), + }, + { XE_RTP_NAME("16021867713"), + XE_RTP_RULES(MEDIA_VERSION(3002), + ENGINE_CLASS(VIDEO_DECODE)), + XE_RTP_ACTIONS(SET(VDBOX_CGCTL3F1C(0), MFXPIPE_CLKGATE_DIS)), + XE_RTP_ENTRY_FLAG(FOREACH_ENGINE), + }, { XE_RTP_NAME("14021486841"), XE_RTP_RULES(MEDIA_VERSION(3000), MEDIA_STEP(A0, B0), ENGINE_CLASS(VIDEO_DECODE)), @@ -644,6 +656,10 @@ static const struct xe_rtp_entry_sr engine_was[] = { XE_RTP_ACTIONS(SET(RING_PSMI_CTL(0), RC_SEMA_IDLE_MSG_DISABLE, XE_RTP_ACTION_FLAG(ENGINE_BASE))) }, + { XE_RTP_NAME("14021402888"), + XE_RTP_RULES(GRAPHICS_VERSION(3003), FUNC(xe_rtp_match_first_render_or_compute)), + XE_RTP_ACTIONS(SET(HALF_SLICE_CHICKEN7, CLEAR_OPTIMIZATION_DISABLE)) + }, }; static const struct xe_rtp_entry_sr lrc_was[] = { diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules index 0ee74a5b2407..e7ed5d583d68 100644 --- a/drivers/gpu/drm/xe/xe_wa_oob.rules +++ b/drivers/gpu/drm/xe/xe_wa_oob.rules @@ -34,14 +34,16 @@ 14022293748 GRAPHICS_VERSION_RANGE(2001, 2002) GRAPHICS_VERSION(2004) GRAPHICS_VERSION_RANGE(3000, 3001) + GRAPHICS_VERSION(3003) 22019794406 GRAPHICS_VERSION_RANGE(2001, 2002) GRAPHICS_VERSION(2004) GRAPHICS_VERSION_RANGE(3000, 3001) + GRAPHICS_VERSION(3003) 22019338487 MEDIA_VERSION(2000) - GRAPHICS_VERSION(2001) + GRAPHICS_VERSION(2001), FUNC(xe_rtp_match_not_sriov_vf) MEDIA_VERSION(3000), MEDIA_STEP(A0, B0), FUNC(xe_rtp_match_not_sriov_vf) 22019338487_display PLATFORM(LUNARLAKE) -16023588340 GRAPHICS_VERSION(2001) +16023588340 GRAPHICS_VERSION(2001), FUNC(xe_rtp_match_not_sriov_vf) 14019789679 GRAPHICS_VERSION(1255) GRAPHICS_VERSION_RANGE(1270, 2004) no_media_l3 MEDIA_VERSION(3000) @@ -58,8 +60,12 @@ no_media_l3 MEDIA_VERSION(3000) GRAPHICS_VERSION(1260), GRAPHICS_STEP(A0, B0) 16023105232 GRAPHICS_VERSION_RANGE(2001, 3001) MEDIA_VERSION_RANGE(1301, 3000) + MEDIA_VERSION(3002) + GRAPHICS_VERSION(3003) 16026508708 GRAPHICS_VERSION_RANGE(1200, 3001) MEDIA_VERSION_RANGE(1300, 3000) + MEDIA_VERSION(3002) + GRAPHICS_VERSION(3003) # SoC workaround - currently applies to all platforms with the following # primary GT GMDID diff --git a/drivers/gpu/trace/Kconfig b/drivers/gpu/trace/Kconfig index c24e9edd022e..cd3d19c4a201 100644 --- a/drivers/gpu/trace/Kconfig +++ b/drivers/gpu/trace/Kconfig @@ -1,4 +1,13 @@ # SPDX-License-Identifier: GPL-2.0-only config TRACE_GPU_MEM - bool + bool "Enable GPU memory usage tracepoints" + default n + help + Choose this option to enable tracepoints for tracking + global and per-process GPU memory usage. Intended for + performance profiling and required for Android. + + Tracepoint availability varies by GPU driver. + + If in doubt, say "N". diff --git a/drivers/i2c/busses/i2c-designware-platdrv.c b/drivers/i2c/busses/i2c-designware-platdrv.c index 879719e91df2..a35e4c64a1d4 100644 --- a/drivers/i2c/busses/i2c-designware-platdrv.c +++ b/drivers/i2c/busses/i2c-designware-platdrv.c @@ -101,7 +101,7 @@ static int bt1_i2c_request_regs(struct dw_i2c_dev *dev) } #endif -static int txgbe_i2c_request_regs(struct dw_i2c_dev *dev) +static int dw_i2c_get_parent_regmap(struct dw_i2c_dev *dev) { dev->map = dev_get_regmap(dev->dev->parent, NULL); if (!dev->map) @@ -123,12 +123,15 @@ static int dw_i2c_plat_request_regs(struct dw_i2c_dev *dev) struct platform_device *pdev = to_platform_device(dev->dev); int ret; + if (device_is_compatible(dev->dev, "intel,xe-i2c")) + return dw_i2c_get_parent_regmap(dev); + switch (dev->flags & MODEL_MASK) { case MODEL_BAIKAL_BT1: ret = bt1_i2c_request_regs(dev); break; case MODEL_WANGXUN_SP: - ret = txgbe_i2c_request_regs(dev); + ret = dw_i2c_get_parent_regmap(dev); break; default: dev->base = devm_platform_ioremap_resource(pdev, 0); @@ -205,25 +208,28 @@ static void i2c_dw_remove_lock_support(struct dw_i2c_dev *dev) static int dw_i2c_plat_probe(struct platform_device *pdev) { + u32 flags = (uintptr_t)device_get_match_data(&pdev->dev); struct device *device = &pdev->dev; struct i2c_adapter *adap; struct dw_i2c_dev *dev; int irq, ret; - irq = platform_get_irq(pdev, 0); - if (irq < 0) + irq = platform_get_irq_optional(pdev, 0); + if (irq == -ENXIO) + flags |= ACCESS_POLLING; + else if (irq < 0) return irq; dev = devm_kzalloc(device, sizeof(*dev), GFP_KERNEL); if (!dev) return -ENOMEM; - dev->flags = (uintptr_t)device_get_match_data(device); if (device_property_present(device, "wx,i2c-snps-model")) - dev->flags = MODEL_WANGXUN_SP | ACCESS_POLLING; + flags = MODEL_WANGXUN_SP | ACCESS_POLLING; dev->dev = device; dev->irq = irq; + dev->flags = flags; platform_set_drvdata(pdev, dev); ret = dw_i2c_plat_request_regs(dev); diff --git a/drivers/mtd/devices/Kconfig b/drivers/mtd/devices/Kconfig index aed653ce8fa2..46cebde79f34 100644 --- a/drivers/mtd/devices/Kconfig +++ b/drivers/mtd/devices/Kconfig @@ -183,6 +183,17 @@ config MTD_POWERNV_FLASH platforms from Linux. This device abstracts away the firmware interface for flash access. +config MTD_INTEL_DG + tristate "Intel Discrete Graphics non-volatile memory driver" + depends on AUXILIARY_BUS + depends on MTD + help + This provides an MTD device to access Intel Discrete Graphics + non-volatile memory. + + To compile this driver as a module, choose M here: the module + will be called mtd-intel-dg. + comment "Disk-On-Chip Device Drivers" config MTD_DOCG3 diff --git a/drivers/mtd/devices/Makefile b/drivers/mtd/devices/Makefile index d11eb2b8b6f8..9fe4ce9cffde 100644 --- a/drivers/mtd/devices/Makefile +++ b/drivers/mtd/devices/Makefile @@ -18,6 +18,7 @@ obj-$(CONFIG_MTD_SST25L) += sst25l.o obj-$(CONFIG_MTD_BCM47XXSFLASH) += bcm47xxsflash.o obj-$(CONFIG_MTD_ST_SPI_FSM) += st_spi_fsm.o obj-$(CONFIG_MTD_POWERNV_FLASH) += powernv_flash.o +obj-$(CONFIG_MTD_INTEL_DG) += mtd_intel_dg.o CFLAGS_docg3.o += -I$(src) diff --git a/drivers/mtd/devices/mtd_intel_dg.c b/drivers/mtd/devices/mtd_intel_dg.c new file mode 100644 index 000000000000..b438ee5aacc3 --- /dev/null +++ b/drivers/mtd/devices/mtd_intel_dg.c @@ -0,0 +1,830 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright(c) 2019-2025, Intel Corporation. All rights reserved. + */ + +#include <linux/bitfield.h> +#include <linux/bits.h> +#include <linux/cleanup.h> +#include <linux/delay.h> +#include <linux/device.h> +#include <linux/intel_dg_nvm_aux.h> +#include <linux/io.h> +#include <linux/io-64-nonatomic-lo-hi.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/mtd/mtd.h> +#include <linux/mtd/partitions.h> +#include <linux/string.h> +#include <linux/slab.h> +#include <linux/sizes.h> +#include <linux/types.h> + +struct intel_dg_nvm { + struct kref refcnt; + struct mtd_info mtd; + struct mutex lock; /* region access lock */ + void __iomem *base; + void __iomem *base2; + bool non_posted_erase; + + size_t size; + unsigned int nregions; + struct { + const char *name; + u8 id; + u64 offset; + u64 size; + unsigned int is_readable:1; + unsigned int is_writable:1; + } regions[] __counted_by(nregions); +}; + +#define NVM_TRIGGER_REG 0x00000000 +#define NVM_VALSIG_REG 0x00000010 +#define NVM_ADDRESS_REG 0x00000040 +#define NVM_REGION_ID_REG 0x00000044 +#define NVM_DEBUG_REG 0x00000000 +/* + * [15:0]-Erase size = 0x0010 4K 0x0080 32K 0x0100 64K + * [23:16]-Reserved + * [31:24]-Erase MEM RegionID + */ +#define NVM_ERASE_REG 0x00000048 +#define NVM_ACCESS_ERROR_REG 0x00000070 +#define NVM_ADDRESS_ERROR_REG 0x00000074 + +/* Flash Valid Signature */ +#define NVM_FLVALSIG 0x0FF0A55A + +#define NVM_MAP_ADDR_MASK GENMASK(7, 0) +#define NVM_MAP_ADDR_SHIFT 0x00000004 + +#define NVM_REGION_ID_DESCRIPTOR 0 +/* Flash Region Base Address */ +#define NVM_FRBA 0x40 +/* Flash Region __n - Flash Descriptor Record */ +#define NVM_FLREG(__n) (NVM_FRBA + ((__n) * 4)) +/* Flash Map 1 Register */ +#define NVM_FLMAP1_REG 0x18 +#define NVM_FLMSTR4_OFFSET 0x00C + +#define NVM_ACCESS_ERROR_PCIE_MASK 0x7 + +#define NVM_FREG_BASE_MASK GENMASK(15, 0) +#define NVM_FREG_ADDR_MASK GENMASK(31, 16) +#define NVM_FREG_ADDR_SHIFT 12 +#define NVM_FREG_MIN_REGION_SIZE 0xFFF + +#define NVM_NON_POSTED_ERASE_DONE BIT(23) +#define NVM_NON_POSTED_ERASE_DONE_ITER 3000 + +static inline void idg_nvm_set_region_id(struct intel_dg_nvm *nvm, u8 region) +{ + iowrite32((u32)region, nvm->base + NVM_REGION_ID_REG); +} + +static inline u32 idg_nvm_error(struct intel_dg_nvm *nvm) +{ + void __iomem *base = nvm->base; + + u32 reg = ioread32(base + NVM_ACCESS_ERROR_REG) & NVM_ACCESS_ERROR_PCIE_MASK; + + /* reset error bits */ + if (reg) + iowrite32(reg, base + NVM_ACCESS_ERROR_REG); + + return reg; +} + +static inline u32 idg_nvm_read32(struct intel_dg_nvm *nvm, u32 address) +{ + void __iomem *base = nvm->base; + + iowrite32(address, base + NVM_ADDRESS_REG); + + return ioread32(base + NVM_TRIGGER_REG); +} + +static inline u64 idg_nvm_read64(struct intel_dg_nvm *nvm, u32 address) +{ + void __iomem *base = nvm->base; + + iowrite32(address, base + NVM_ADDRESS_REG); + + return readq(base + NVM_TRIGGER_REG); +} + +static void idg_nvm_write32(struct intel_dg_nvm *nvm, u32 address, u32 data) +{ + void __iomem *base = nvm->base; + + iowrite32(address, base + NVM_ADDRESS_REG); + + iowrite32(data, base + NVM_TRIGGER_REG); +} + +static void idg_nvm_write64(struct intel_dg_nvm *nvm, u32 address, u64 data) +{ + void __iomem *base = nvm->base; + + iowrite32(address, base + NVM_ADDRESS_REG); + + writeq(data, base + NVM_TRIGGER_REG); +} + +static int idg_nvm_get_access_map(struct intel_dg_nvm *nvm, u32 *access_map) +{ + u32 fmstr4_addr; + u32 fmstr4; + u32 flmap1; + u32 fmba; + + idg_nvm_set_region_id(nvm, NVM_REGION_ID_DESCRIPTOR); + + flmap1 = idg_nvm_read32(nvm, NVM_FLMAP1_REG); + if (idg_nvm_error(nvm)) + return -EIO; + /* Get Flash Master Baser Address (FMBA) */ + fmba = (FIELD_GET(NVM_MAP_ADDR_MASK, flmap1) << NVM_MAP_ADDR_SHIFT); + fmstr4_addr = fmba + NVM_FLMSTR4_OFFSET; + + fmstr4 = idg_nvm_read32(nvm, fmstr4_addr); + if (idg_nvm_error(nvm)) + return -EIO; + + *access_map = fmstr4; + return 0; +} + +/* + * Region read/write access encoded in the access map + * in the following order from the lower bit: + * [3:0] regions 12-15 read state + * [7:4] regions 12-15 write state + * [19:8] regions 0-11 read state + * [31:20] regions 0-11 write state + */ +static bool idg_nvm_region_readable(u32 access_map, u8 region) +{ + if (region < 12) + return access_map & BIT(region + 8); /* [19:8] */ + else + return access_map & BIT(region - 12); /* [3:0] */ +} + +static bool idg_nvm_region_writable(u32 access_map, u8 region) +{ + if (region < 12) + return access_map & BIT(region + 20); /* [31:20] */ + else + return access_map & BIT(region - 8); /* [7:4] */ +} + +static int idg_nvm_is_valid(struct intel_dg_nvm *nvm) +{ + u32 is_valid; + + idg_nvm_set_region_id(nvm, NVM_REGION_ID_DESCRIPTOR); + + is_valid = idg_nvm_read32(nvm, NVM_VALSIG_REG); + if (idg_nvm_error(nvm)) + return -EIO; + + if (is_valid != NVM_FLVALSIG) + return -ENODEV; + + return 0; +} + +static unsigned int idg_nvm_get_region(const struct intel_dg_nvm *nvm, loff_t from) +{ + unsigned int i; + + for (i = 0; i < nvm->nregions; i++) { + if ((nvm->regions[i].offset + nvm->regions[i].size - 1) >= from && + nvm->regions[i].offset <= from && + nvm->regions[i].size != 0) + break; + } + + return i; +} + +static ssize_t idg_nvm_rewrite_partial(struct intel_dg_nvm *nvm, loff_t to, + loff_t offset, size_t len, const u32 *newdata) +{ + u32 data = idg_nvm_read32(nvm, to); + + if (idg_nvm_error(nvm)) + return -EIO; + + memcpy((u8 *)&data + offset, newdata, len); + + idg_nvm_write32(nvm, to, data); + if (idg_nvm_error(nvm)) + return -EIO; + + return len; +} + +static ssize_t idg_write(struct intel_dg_nvm *nvm, u8 region, + loff_t to, size_t len, const unsigned char *buf) +{ + size_t len_s = len; + size_t to_shift; + size_t len8; + size_t len4; + ssize_t ret; + size_t to4; + size_t i; + + idg_nvm_set_region_id(nvm, region); + + to4 = ALIGN_DOWN(to, sizeof(u32)); + to_shift = min(sizeof(u32) - ((size_t)to - to4), len); + if (to - to4) { + ret = idg_nvm_rewrite_partial(nvm, to4, to - to4, to_shift, (u32 *)&buf[0]); + if (ret < 0) + return ret; + + buf += to_shift; + to += to_shift; + len_s -= to_shift; + } + + if (!IS_ALIGNED(to, sizeof(u64)) && + ((to ^ (to + len_s)) & GENMASK(31, 10))) { + /* + * Workaround reads/writes across 1k-aligned addresses + * (start u32 before 1k, end u32 after) + * as this fails on hardware. + */ + u32 data; + + memcpy(&data, &buf[0], sizeof(u32)); + idg_nvm_write32(nvm, to, data); + if (idg_nvm_error(nvm)) + return -EIO; + buf += sizeof(u32); + to += sizeof(u32); + len_s -= sizeof(u32); + } + + len8 = ALIGN_DOWN(len_s, sizeof(u64)); + for (i = 0; i < len8; i += sizeof(u64)) { + u64 data; + + memcpy(&data, &buf[i], sizeof(u64)); + idg_nvm_write64(nvm, to + i, data); + if (idg_nvm_error(nvm)) + return -EIO; + } + + len4 = len_s - len8; + if (len4 >= sizeof(u32)) { + u32 data; + + memcpy(&data, &buf[i], sizeof(u32)); + idg_nvm_write32(nvm, to + i, data); + if (idg_nvm_error(nvm)) + return -EIO; + i += sizeof(u32); + len4 -= sizeof(u32); + } + + if (len4 > 0) { + ret = idg_nvm_rewrite_partial(nvm, to + i, 0, len4, (u32 *)&buf[i]); + if (ret < 0) + return ret; + } + + return len; +} + +static ssize_t idg_read(struct intel_dg_nvm *nvm, u8 region, + loff_t from, size_t len, unsigned char *buf) +{ + size_t len_s = len; + size_t from_shift; + size_t from4; + size_t len8; + size_t len4; + size_t i; + + idg_nvm_set_region_id(nvm, region); + + from4 = ALIGN_DOWN(from, sizeof(u32)); + from_shift = min(sizeof(u32) - ((size_t)from - from4), len); + + if (from - from4) { + u32 data = idg_nvm_read32(nvm, from4); + + if (idg_nvm_error(nvm)) + return -EIO; + memcpy(&buf[0], (u8 *)&data + (from - from4), from_shift); + len_s -= from_shift; + buf += from_shift; + from += from_shift; + } + + if (!IS_ALIGNED(from, sizeof(u64)) && + ((from ^ (from + len_s)) & GENMASK(31, 10))) { + /* + * Workaround reads/writes across 1k-aligned addresses + * (start u32 before 1k, end u32 after) + * as this fails on hardware. + */ + u32 data = idg_nvm_read32(nvm, from); + + if (idg_nvm_error(nvm)) + return -EIO; + memcpy(&buf[0], &data, sizeof(data)); + len_s -= sizeof(u32); + buf += sizeof(u32); + from += sizeof(u32); + } + + len8 = ALIGN_DOWN(len_s, sizeof(u64)); + for (i = 0; i < len8; i += sizeof(u64)) { + u64 data = idg_nvm_read64(nvm, from + i); + + if (idg_nvm_error(nvm)) + return -EIO; + + memcpy(&buf[i], &data, sizeof(data)); + } + + len4 = len_s - len8; + if (len4 >= sizeof(u32)) { + u32 data = idg_nvm_read32(nvm, from + i); + + if (idg_nvm_error(nvm)) + return -EIO; + memcpy(&buf[i], &data, sizeof(data)); + i += sizeof(u32); + len4 -= sizeof(u32); + } + + if (len4 > 0) { + u32 data = idg_nvm_read32(nvm, from + i); + + if (idg_nvm_error(nvm)) + return -EIO; + memcpy(&buf[i], &data, len4); + } + + return len; +} + +static ssize_t +idg_erase(struct intel_dg_nvm *nvm, u8 region, loff_t from, u64 len, u64 *fail_addr) +{ + void __iomem *base2 = nvm->base2; + void __iomem *base = nvm->base; + const u32 block = 0x10; + u32 iter = 0; + u32 reg; + u64 i; + + for (i = 0; i < len; i += SZ_4K) { + iowrite32(from + i, base + NVM_ADDRESS_REG); + iowrite32(region << 24 | block, base + NVM_ERASE_REG); + if (nvm->non_posted_erase) { + /* Wait for Erase Done */ + reg = ioread32(base2 + NVM_DEBUG_REG); + while (!(reg & NVM_NON_POSTED_ERASE_DONE) && + ++iter < NVM_NON_POSTED_ERASE_DONE_ITER) { + msleep(10); + reg = ioread32(base2 + NVM_DEBUG_REG); + } + if (reg & NVM_NON_POSTED_ERASE_DONE) { + /* Clear Erase Done */ + iowrite32(reg, base2 + NVM_DEBUG_REG); + } else { + *fail_addr = from + i; + return -ETIME; + } + } + /* Since the writes are via sgunit + * we cannot do back to back erases. + */ + msleep(50); + } + return len; +} + +static int intel_dg_nvm_init(struct intel_dg_nvm *nvm, struct device *device, + bool non_posted_erase) +{ + u32 access_map = 0; + unsigned int i, n; + int ret; + + /* clean error register, previous errors are ignored */ + idg_nvm_error(nvm); + + ret = idg_nvm_is_valid(nvm); + if (ret) { + dev_err(device, "The MEM is not valid %d\n", ret); + return ret; + } + + if (idg_nvm_get_access_map(nvm, &access_map)) + return -EIO; + + for (i = 0, n = 0; i < nvm->nregions; i++) { + u32 address, base, limit, region; + u8 id = nvm->regions[i].id; + + address = NVM_FLREG(id); + region = idg_nvm_read32(nvm, address); + + base = FIELD_GET(NVM_FREG_BASE_MASK, region) << NVM_FREG_ADDR_SHIFT; + limit = (FIELD_GET(NVM_FREG_ADDR_MASK, region) << NVM_FREG_ADDR_SHIFT) | + NVM_FREG_MIN_REGION_SIZE; + + dev_dbg(device, "[%d] %s: region: 0x%08X base: 0x%08x limit: 0x%08x\n", + id, nvm->regions[i].name, region, base, limit); + + if (base >= limit || (i > 0 && limit == 0)) { + dev_dbg(device, "[%d] %s: disabled\n", + id, nvm->regions[i].name); + nvm->regions[i].is_readable = 0; + continue; + } + + if (nvm->size < limit) + nvm->size = limit; + + nvm->regions[i].offset = base; + nvm->regions[i].size = limit - base + 1; + /* No write access to descriptor; mask it out*/ + nvm->regions[i].is_writable = idg_nvm_region_writable(access_map, id); + + nvm->regions[i].is_readable = idg_nvm_region_readable(access_map, id); + dev_dbg(device, "Registered, %s id=%d offset=%lld size=%lld rd=%d wr=%d\n", + nvm->regions[i].name, + nvm->regions[i].id, + nvm->regions[i].offset, + nvm->regions[i].size, + nvm->regions[i].is_readable, + nvm->regions[i].is_writable); + + if (nvm->regions[i].is_readable) + n++; + } + + nvm->non_posted_erase = non_posted_erase; + + dev_dbg(device, "Registered %d regions\n", n); + dev_dbg(device, "Non posted erase %d\n", nvm->non_posted_erase); + + /* Need to add 1 to the amount of memory + * so it is reported as an even block + */ + nvm->size += 1; + + return n; +} + +static int intel_dg_mtd_erase(struct mtd_info *mtd, struct erase_info *info) +{ + struct intel_dg_nvm *nvm = mtd->priv; + size_t total_len; + unsigned int idx; + ssize_t bytes; + loff_t from; + size_t len; + u8 region; + u64 addr; + + if (WARN_ON(!nvm)) + return -EINVAL; + + if (!IS_ALIGNED(info->addr, SZ_4K) || !IS_ALIGNED(info->len, SZ_4K)) { + dev_err(&mtd->dev, "unaligned erase %llx %llx\n", + info->addr, info->len); + info->fail_addr = MTD_FAIL_ADDR_UNKNOWN; + return -EINVAL; + } + + total_len = info->len; + addr = info->addr; + + guard(mutex)(&nvm->lock); + + while (total_len > 0) { + if (!IS_ALIGNED(addr, SZ_4K) || !IS_ALIGNED(total_len, SZ_4K)) { + dev_err(&mtd->dev, "unaligned erase %llx %zx\n", addr, total_len); + info->fail_addr = addr; + return -ERANGE; + } + + idx = idg_nvm_get_region(nvm, addr); + if (idx >= nvm->nregions) { + dev_err(&mtd->dev, "out of range"); + info->fail_addr = MTD_FAIL_ADDR_UNKNOWN; + return -ERANGE; + } + + from = addr - nvm->regions[idx].offset; + region = nvm->regions[idx].id; + len = total_len; + if (len > nvm->regions[idx].size - from) + len = nvm->regions[idx].size - from; + + dev_dbg(&mtd->dev, "erasing region[%d] %s from %llx len %zx\n", + region, nvm->regions[idx].name, from, len); + + bytes = idg_erase(nvm, region, from, len, &info->fail_addr); + if (bytes < 0) { + dev_dbg(&mtd->dev, "erase failed with %zd\n", bytes); + info->fail_addr += nvm->regions[idx].offset; + return bytes; + } + + addr += len; + total_len -= len; + } + + return 0; +} + +static int intel_dg_mtd_read(struct mtd_info *mtd, loff_t from, size_t len, + size_t *retlen, u_char *buf) +{ + struct intel_dg_nvm *nvm = mtd->priv; + unsigned int idx; + ssize_t ret; + u8 region; + + if (WARN_ON(!nvm)) + return -EINVAL; + + idx = idg_nvm_get_region(nvm, from); + + dev_dbg(&mtd->dev, "reading region[%d] %s from %lld len %zd\n", + nvm->regions[idx].id, nvm->regions[idx].name, from, len); + + if (idx >= nvm->nregions) { + dev_err(&mtd->dev, "out of range"); + return -ERANGE; + } + + from -= nvm->regions[idx].offset; + region = nvm->regions[idx].id; + if (len > nvm->regions[idx].size - from) + len = nvm->regions[idx].size - from; + + guard(mutex)(&nvm->lock); + + ret = idg_read(nvm, region, from, len, buf); + if (ret < 0) { + dev_dbg(&mtd->dev, "read failed with %zd\n", ret); + return ret; + } + + *retlen = ret; + + return 0; +} + +static int intel_dg_mtd_write(struct mtd_info *mtd, loff_t to, size_t len, + size_t *retlen, const u_char *buf) +{ + struct intel_dg_nvm *nvm = mtd->priv; + unsigned int idx; + ssize_t ret; + u8 region; + + if (WARN_ON(!nvm)) + return -EINVAL; + + idx = idg_nvm_get_region(nvm, to); + + dev_dbg(&mtd->dev, "writing region[%d] %s to %lld len %zd\n", + nvm->regions[idx].id, nvm->regions[idx].name, to, len); + + if (idx >= nvm->nregions) { + dev_err(&mtd->dev, "out of range"); + return -ERANGE; + } + + to -= nvm->regions[idx].offset; + region = nvm->regions[idx].id; + if (len > nvm->regions[idx].size - to) + len = nvm->regions[idx].size - to; + + guard(mutex)(&nvm->lock); + + ret = idg_write(nvm, region, to, len, buf); + if (ret < 0) { + dev_dbg(&mtd->dev, "write failed with %zd\n", ret); + return ret; + } + + *retlen = ret; + + return 0; +} + +static void intel_dg_nvm_release(struct kref *kref) +{ + struct intel_dg_nvm *nvm = container_of(kref, struct intel_dg_nvm, refcnt); + int i; + + pr_debug("freeing intel_dg nvm\n"); + for (i = 0; i < nvm->nregions; i++) + kfree(nvm->regions[i].name); + mutex_destroy(&nvm->lock); + kfree(nvm); +} + +static int intel_dg_mtd_get_device(struct mtd_info *mtd) +{ + struct mtd_info *master = mtd_get_master(mtd); + struct intel_dg_nvm *nvm = master->priv; + + if (WARN_ON(!nvm)) + return -EINVAL; + pr_debug("get mtd %s %d\n", mtd->name, kref_read(&nvm->refcnt)); + kref_get(&nvm->refcnt); + + return 0; +} + +static void intel_dg_mtd_put_device(struct mtd_info *mtd) +{ + struct mtd_info *master = mtd_get_master(mtd); + struct intel_dg_nvm *nvm = master->priv; + + if (WARN_ON(!nvm)) + return; + pr_debug("put mtd %s %d\n", mtd->name, kref_read(&nvm->refcnt)); + kref_put(&nvm->refcnt, intel_dg_nvm_release); +} + +static int intel_dg_nvm_init_mtd(struct intel_dg_nvm *nvm, struct device *device, + unsigned int nparts, bool writable_override) +{ + struct mtd_partition *parts = NULL; + unsigned int i, n; + int ret; + + dev_dbg(device, "registering with mtd\n"); + + nvm->mtd.owner = THIS_MODULE; + nvm->mtd.dev.parent = device; + nvm->mtd.flags = MTD_CAP_NORFLASH; + nvm->mtd.type = MTD_DATAFLASH; + nvm->mtd.priv = nvm; + nvm->mtd._write = intel_dg_mtd_write; + nvm->mtd._read = intel_dg_mtd_read; + nvm->mtd._erase = intel_dg_mtd_erase; + nvm->mtd._get_device = intel_dg_mtd_get_device; + nvm->mtd._put_device = intel_dg_mtd_put_device; + nvm->mtd.writesize = SZ_1; /* 1 byte granularity */ + nvm->mtd.erasesize = SZ_4K; /* 4K bytes granularity */ + nvm->mtd.size = nvm->size; + + parts = kcalloc(nvm->nregions, sizeof(*parts), GFP_KERNEL); + if (!parts) + return -ENOMEM; + + for (i = 0, n = 0; i < nvm->nregions && n < nparts; i++) { + if (!nvm->regions[i].is_readable) + continue; + parts[n].name = nvm->regions[i].name; + parts[n].offset = nvm->regions[i].offset; + parts[n].size = nvm->regions[i].size; + if (!nvm->regions[i].is_writable && !writable_override) + parts[n].mask_flags = MTD_WRITEABLE; + n++; + } + + ret = mtd_device_register(&nvm->mtd, parts, n); + + kfree(parts); + return ret; +} + +static int intel_dg_mtd_probe(struct auxiliary_device *aux_dev, + const struct auxiliary_device_id *aux_dev_id) +{ + struct intel_dg_nvm_dev *invm = auxiliary_dev_to_intel_dg_nvm_dev(aux_dev); + struct intel_dg_nvm *nvm; + struct device *device; + unsigned int nregions; + unsigned int i, n; + int ret; + + device = &aux_dev->dev; + + /* count available regions */ + for (nregions = 0, i = 0; i < INTEL_DG_NVM_REGIONS; i++) { + if (invm->regions[i].name) + nregions++; + } + + if (!nregions) { + dev_err(device, "no regions defined\n"); + return -ENODEV; + } + + nvm = kzalloc(struct_size(nvm, regions, nregions), GFP_KERNEL); + if (!nvm) + return -ENOMEM; + + kref_init(&nvm->refcnt); + mutex_init(&nvm->lock); + + for (n = 0, i = 0; i < INTEL_DG_NVM_REGIONS; i++) { + if (!invm->regions[i].name) + continue; + + char *name = kasprintf(GFP_KERNEL, "%s.%s", + dev_name(&aux_dev->dev), invm->regions[i].name); + if (!name) + continue; + nvm->regions[n].name = name; + nvm->regions[n].id = i; + n++; + } + nvm->nregions = n; /* in case where kasprintf fail */ + + nvm->base = devm_ioremap_resource(device, &invm->bar); + if (IS_ERR(nvm->base)) { + ret = PTR_ERR(nvm->base); + goto err; + } + + if (invm->non_posted_erase) { + nvm->base2 = devm_ioremap_resource(device, &invm->bar2); + if (IS_ERR(nvm->base2)) { + ret = PTR_ERR(nvm->base2); + goto err; + } + } + + ret = intel_dg_nvm_init(nvm, device, invm->non_posted_erase); + if (ret < 0) { + dev_err(device, "cannot initialize nvm %d\n", ret); + goto err; + } + + ret = intel_dg_nvm_init_mtd(nvm, device, ret, invm->writable_override); + if (ret) { + dev_err(device, "failed init mtd %d\n", ret); + goto err; + } + + dev_set_drvdata(&aux_dev->dev, nvm); + + return 0; + +err: + kref_put(&nvm->refcnt, intel_dg_nvm_release); + return ret; +} + +static void intel_dg_mtd_remove(struct auxiliary_device *aux_dev) +{ + struct intel_dg_nvm *nvm = dev_get_drvdata(&aux_dev->dev); + + if (!nvm) + return; + + mtd_device_unregister(&nvm->mtd); + + dev_set_drvdata(&aux_dev->dev, NULL); + + kref_put(&nvm->refcnt, intel_dg_nvm_release); +} + +static const struct auxiliary_device_id intel_dg_mtd_id_table[] = { + { + .name = "i915.nvm", + }, + { + .name = "xe.nvm", + }, + { + /* sentinel */ + } +}; +MODULE_DEVICE_TABLE(auxiliary, intel_dg_mtd_id_table); + +static struct auxiliary_driver intel_dg_mtd_driver = { + .probe = intel_dg_mtd_probe, + .remove = intel_dg_mtd_remove, + .driver = { + /* auxiliary_driver_register() sets .name to be the modname */ + }, + .id_table = intel_dg_mtd_id_table +}; +module_auxiliary_driver(intel_dg_mtd_driver); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Intel Corporation"); +MODULE_DESCRIPTION("Intel DGFX MTD driver"); diff --git a/drivers/soc/qcom/Kconfig b/drivers/soc/qcom/Kconfig index 58e63cf0036b..2caadbbcf830 100644 --- a/drivers/soc/qcom/Kconfig +++ b/drivers/soc/qcom/Kconfig @@ -296,3 +296,11 @@ config QCOM_PBS PBS trigger event to the PBS RAM. endmenu + +config QCOM_UBWC_CONFIG + tristate + help + Most Qualcomm SoCs feature a number of Universal Bandwidth Compression + (UBWC) engines across various IP blocks, which need to be initialized + with coherent configuration data. This module functions as a single + source of truth for that information. diff --git a/drivers/soc/qcom/Makefile b/drivers/soc/qcom/Makefile index acbca2ab5cc2..b7f1d2a57367 100644 --- a/drivers/soc/qcom/Makefile +++ b/drivers/soc/qcom/Makefile @@ -39,3 +39,4 @@ obj-$(CONFIG_QCOM_ICC_BWMON) += icc-bwmon.o qcom_ice-objs += ice.o obj-$(CONFIG_QCOM_INLINE_CRYPTO_ENGINE) += qcom_ice.o obj-$(CONFIG_QCOM_PBS) += qcom-pbs.o +obj-$(CONFIG_QCOM_UBWC_CONFIG) += ubwc_config.o diff --git a/drivers/soc/qcom/ubwc_config.c b/drivers/soc/qcom/ubwc_config.c new file mode 100644 index 000000000000..bd0a98aad9f3 --- /dev/null +++ b/drivers/soc/qcom/ubwc_config.c @@ -0,0 +1,282 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. + */ + +#include <linux/debugfs.h> +#include <linux/io.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/of_address.h> +#include <linux/platform_device.h> + +#include <linux/soc/qcom/ubwc.h> + +static const struct qcom_ubwc_cfg_data msm8937_data = { + .ubwc_enc_version = UBWC_1_0, + .ubwc_dec_version = UBWC_1_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL1 | + UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .highest_bank_bit = 14, +}; + +static const struct qcom_ubwc_cfg_data msm8998_data = { + .ubwc_enc_version = UBWC_1_0, + .ubwc_dec_version = UBWC_1_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL1 | + UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .highest_bank_bit = 15, +}; + +static const struct qcom_ubwc_cfg_data qcm2290_data = { + /* no UBWC */ + .highest_bank_bit = 15, +}; + +static const struct qcom_ubwc_cfg_data sa8775p_data = { + .ubwc_enc_version = UBWC_4_0, + .ubwc_dec_version = UBWC_4_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + .highest_bank_bit = 13, + .macrotile_mode = true, +}; + +static const struct qcom_ubwc_cfg_data sar2130p_data = { + .ubwc_enc_version = UBWC_3_0, /* 4.0.2 in hw */ + .ubwc_dec_version = UBWC_4_3, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + .highest_bank_bit = 13, + .macrotile_mode = true, +}; + +static const struct qcom_ubwc_cfg_data sc7180_data = { + .ubwc_enc_version = UBWC_2_0, + .ubwc_dec_version = UBWC_2_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + .highest_bank_bit = 14, +}; + +static const struct qcom_ubwc_cfg_data sc7280_data = { + .ubwc_enc_version = UBWC_3_0, + .ubwc_dec_version = UBWC_4_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + .highest_bank_bit = 14, + .macrotile_mode = true, +}; + +static const struct qcom_ubwc_cfg_data sc8180x_data = { + .ubwc_enc_version = UBWC_3_0, + .ubwc_dec_version = UBWC_3_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .highest_bank_bit = 16, + .macrotile_mode = true, +}; + +static const struct qcom_ubwc_cfg_data sc8280xp_data = { + .ubwc_enc_version = UBWC_4_0, + .ubwc_dec_version = UBWC_4_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + .highest_bank_bit = 16, + .macrotile_mode = true, +}; + +static const struct qcom_ubwc_cfg_data sdm670_data = { + .ubwc_enc_version = UBWC_2_0, + .ubwc_dec_version = UBWC_2_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .highest_bank_bit = 14, +}; + +static const struct qcom_ubwc_cfg_data sdm845_data = { + .ubwc_enc_version = UBWC_2_0, + .ubwc_dec_version = UBWC_2_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .highest_bank_bit = 15, +}; + +static const struct qcom_ubwc_cfg_data sm6115_data = { + .ubwc_enc_version = UBWC_1_0, + .ubwc_dec_version = UBWC_2_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL1 | + UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + .highest_bank_bit = 14, +}; + +static const struct qcom_ubwc_cfg_data sm6125_data = { + .ubwc_enc_version = UBWC_1_0, + .ubwc_dec_version = UBWC_3_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL1 | + UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .highest_bank_bit = 14, +}; + +static const struct qcom_ubwc_cfg_data sm6150_data = { + .ubwc_enc_version = UBWC_2_0, + .ubwc_dec_version = UBWC_2_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .highest_bank_bit = 14, +}; + +static const struct qcom_ubwc_cfg_data sm6350_data = { + .ubwc_enc_version = UBWC_2_0, + .ubwc_dec_version = UBWC_2_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + .highest_bank_bit = 14, +}; + +static const struct qcom_ubwc_cfg_data sm7150_data = { + .ubwc_enc_version = UBWC_2_0, + .ubwc_dec_version = UBWC_2_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .highest_bank_bit = 14, +}; + +static const struct qcom_ubwc_cfg_data sm8150_data = { + .ubwc_enc_version = UBWC_3_0, + .ubwc_dec_version = UBWC_3_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .highest_bank_bit = 15, +}; + +static const struct qcom_ubwc_cfg_data sm8250_data = { + .ubwc_enc_version = UBWC_4_0, + .ubwc_dec_version = UBWC_4_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + /* TODO: highest_bank_bit = 15 for LP_DDR4 */ + .highest_bank_bit = 16, + .macrotile_mode = true, +}; + +static const struct qcom_ubwc_cfg_data sm8350_data = { + .ubwc_enc_version = UBWC_4_0, + .ubwc_dec_version = UBWC_4_0, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + /* TODO: highest_bank_bit = 15 for LP_DDR4 */ + .highest_bank_bit = 16, + .macrotile_mode = true, +}; + +static const struct qcom_ubwc_cfg_data sm8550_data = { + .ubwc_enc_version = UBWC_4_0, + .ubwc_dec_version = UBWC_4_3, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + /* TODO: highest_bank_bit = 15 for LP_DDR4 */ + .highest_bank_bit = 16, + .macrotile_mode = true, +}; + +static const struct qcom_ubwc_cfg_data sm8750_data = { + .ubwc_enc_version = UBWC_5_0, + .ubwc_dec_version = UBWC_5_0, + .ubwc_swizzle = 6, + .ubwc_bank_spread = true, + /* TODO: highest_bank_bit = 15 for LP_DDR4 */ + .highest_bank_bit = 16, + .macrotile_mode = true, +}; + +static const struct qcom_ubwc_cfg_data x1e80100_data = { + .ubwc_enc_version = UBWC_4_0, + .ubwc_dec_version = UBWC_4_3, + .ubwc_swizzle = UBWC_SWIZZLE_ENABLE_LVL2 | + UBWC_SWIZZLE_ENABLE_LVL3, + .ubwc_bank_spread = true, + /* TODO: highest_bank_bit = 15 for LP_DDR4 */ + .highest_bank_bit = 16, + .macrotile_mode = true, +}; + +static const struct of_device_id qcom_ubwc_configs[] __maybe_unused = { + { .compatible = "qcom,apq8096", .data = &msm8998_data }, + { .compatible = "qcom,msm8917", .data = &msm8937_data }, + { .compatible = "qcom,msm8937", .data = &msm8937_data }, + { .compatible = "qcom,msm8953", .data = &msm8937_data }, + { .compatible = "qcom,msm8956", .data = &msm8937_data }, + { .compatible = "qcom,msm8976", .data = &msm8937_data }, + { .compatible = "qcom,msm8996", .data = &msm8998_data }, + { .compatible = "qcom,msm8998", .data = &msm8998_data }, + { .compatible = "qcom,qcm2290", .data = &qcm2290_data, }, + { .compatible = "qcom,qcm6490", .data = &sc7280_data, }, + { .compatible = "qcom,sa8155p", .data = &sm8150_data, }, + { .compatible = "qcom,sa8540p", .data = &sc8280xp_data, }, + { .compatible = "qcom,sa8775p", .data = &sa8775p_data, }, + { .compatible = "qcom,sar2130p", .data = &sar2130p_data }, + { .compatible = "qcom,sc7180", .data = &sc7180_data }, + { .compatible = "qcom,sc7280", .data = &sc7280_data, }, + { .compatible = "qcom,sc8180x", .data = &sc8180x_data, }, + { .compatible = "qcom,sc8280xp", .data = &sc8280xp_data, }, + { .compatible = "qcom,sdm630", .data = &msm8937_data }, + { .compatible = "qcom,sdm636", .data = &msm8937_data }, + { .compatible = "qcom,sdm660", .data = &msm8937_data }, + { .compatible = "qcom,sdm670", .data = &sdm670_data, }, + { .compatible = "qcom,sdm845", .data = &sdm845_data, }, + { .compatible = "qcom,sm4250", .data = &sm6115_data, }, + { .compatible = "qcom,sm6115", .data = &sm6115_data, }, + { .compatible = "qcom,sm6125", .data = &sm6125_data, }, + { .compatible = "qcom,sm6150", .data = &sm6150_data, }, + { .compatible = "qcom,sm6350", .data = &sm6350_data, }, + { .compatible = "qcom,sm6375", .data = &sm6350_data, }, + { .compatible = "qcom,sm7125", .data = &sc7180_data }, + { .compatible = "qcom,sm7150", .data = &sm7150_data, }, + { .compatible = "qcom,sm8150", .data = &sm8150_data, }, + { .compatible = "qcom,sm8250", .data = &sm8250_data, }, + { .compatible = "qcom,sm8350", .data = &sm8350_data, }, + { .compatible = "qcom,sm8450", .data = &sm8350_data, }, + { .compatible = "qcom,sm8550", .data = &sm8550_data, }, + { .compatible = "qcom,sm8650", .data = &sm8550_data, }, + { .compatible = "qcom,sm8750", .data = &sm8750_data, }, + { .compatible = "qcom,x1e80100", .data = &x1e80100_data, }, + { .compatible = "qcom,x1p42100", .data = &x1e80100_data, }, + { } +}; + +const struct qcom_ubwc_cfg_data *qcom_ubwc_config_get_data(void) +{ + const struct of_device_id *match; + struct device_node *root; + + root = of_find_node_by_path("/"); + if (!root) + return ERR_PTR(-ENODEV); + + match = of_match_node(qcom_ubwc_configs, root); + of_node_put(root); + if (!match) { + pr_err("Couldn't find UBWC config data for this platform!\n"); + return ERR_PTR(-EINVAL); + } + + return match->data; +} +EXPORT_SYMBOL_GPL(qcom_ubwc_config_get_data); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("UBWC config database for QTI SoCs"); diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig index c3da6c0bfca6..d51777df12d1 100644 --- a/drivers/video/Kconfig +++ b/drivers/video/Kconfig @@ -87,4 +87,6 @@ if FB_CORE || SGI_NEWPORT_CONSOLE endif +source "drivers/gpu/trace/Kconfig" + endmenu diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c index 25684f5d6523..d8eab4859fd4 100644 --- a/drivers/video/fbdev/core/fbcon.c +++ b/drivers/video/fbdev/core/fbcon.c @@ -953,13 +953,13 @@ static const char *fbcon_startup(void) int rows, cols; /* - * If num_registered_fb is zero, this is a call for the dummy part. + * If fbcon_num_registered_fb is zero, this is a call for the dummy part. * The frame buffer devices weren't initialized yet. */ if (!fbcon_num_registered_fb || info_idx == -1) return display_desc; /* - * Instead of blindly using registered_fb[0], we use info_idx, set by + * Instead of blindly using fbcon_registered_fb[0], we use info_idx, set by * fbcon_fb_registered(); */ info = fbcon_registered_fb[info_idx]; diff --git a/include/drm/display/drm_dp_helper.h b/include/drm/display/drm_dp_helper.h index 3e35a68b2b41..87caa4f1fdb8 100644 --- a/include/drm/display/drm_dp_helper.h +++ b/include/drm/display/drm_dp_helper.h @@ -843,6 +843,7 @@ drm_dp_has_quirk(const struct drm_dp_desc *desc, enum drm_dp_quirk quirk) * @lsb_reg_used: Do we also write values to the DP_EDP_BACKLIGHT_BRIGHTNESS_LSB register? * @aux_enable: Does the panel support the AUX enable cap? * @aux_set: Does the panel support setting the brightness through AUX? + * @luminance_set: Does the panel support setting the brightness through AUX using luminance values? * * This structure contains various data about an eDP backlight, which can be populated by using * drm_edp_backlight_init(). @@ -850,21 +851,23 @@ drm_dp_has_quirk(const struct drm_dp_desc *desc, enum drm_dp_quirk quirk) struct drm_edp_backlight_info { u8 pwmgen_bit_count; u8 pwm_freq_pre_divider; - u16 max; + u32 max; bool lsb_reg_used : 1; bool aux_enable : 1; bool aux_set : 1; + bool luminance_set : 1; }; int drm_edp_backlight_init(struct drm_dp_aux *aux, struct drm_edp_backlight_info *bl, + u32 max_luminance, u16 driver_pwm_freq_hz, const u8 edp_dpcd[EDP_DISPLAY_CTL_CAP_SIZE], - u16 *current_level, u8 *current_mode); + u32 *current_level, u8 *current_mode, bool need_luminance); int drm_edp_backlight_set_level(struct drm_dp_aux *aux, const struct drm_edp_backlight_info *bl, - u16 level); + u32 level); int drm_edp_backlight_enable(struct drm_dp_aux *aux, const struct drm_edp_backlight_info *bl, - u16 level); + u32 level); int drm_edp_backlight_disable(struct drm_dp_aux *aux, const struct drm_edp_backlight_info *bl); #if IS_ENABLED(CONFIG_DRM_KMS_HELPER) && (IS_BUILTIN(CONFIG_BACKLIGHT_CLASS_DEVICE) || \ diff --git a/include/drm/drm_accel.h b/include/drm/drm_accel.h index 038ccb02f9a3..20a665ec6f16 100644 --- a/include/drm/drm_accel.h +++ b/include/drm/drm_accel.h @@ -58,7 +58,6 @@ void accel_core_exit(void); int accel_core_init(void); void accel_set_device_instance_params(struct device *kdev, int index); int accel_open(struct inode *inode, struct file *filp); -void accel_debugfs_init(struct drm_device *dev); void accel_debugfs_register(struct drm_device *dev); #else @@ -77,10 +76,6 @@ static inline void accel_set_device_instance_params(struct device *kdev, int ind { } -static inline void accel_debugfs_init(struct drm_device *dev) -{ -} - static inline void accel_debugfs_register(struct drm_device *dev) { } diff --git a/include/drm/drm_bridge.h b/include/drm/drm_bridge.h index 7f66f9018c10..d2454ba83db3 100644 --- a/include/drm/drm_bridge.h +++ b/include/drm/drm_bridge.h @@ -1051,7 +1051,7 @@ enum drm_bridge_ops { */ DRM_BRIDGE_OP_HDMI_CEC_NOTIFIER = BIT(7), /** - * @DRM_BRIDGE_OP_HDMI_CEC_ADAPTER: The bridge requires CEC notifier + * @DRM_BRIDGE_OP_HDMI_CEC_ADAPTER: The bridge requires CEC adapter * to be present. */ DRM_BRIDGE_OP_HDMI_CEC_ADAPTER = BIT(8), diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h index 137773dd138e..8f34f4b8183d 100644 --- a/include/drm/drm_connector.h +++ b/include/drm/drm_connector.h @@ -800,6 +800,11 @@ struct drm_display_info { struct drm_hdmi_info hdmi; /** + * @hdr_sink_metadata: HDR Metadata Information read from sink + */ + struct hdr_sink_metadata hdr_sink_metadata; + + /** * @non_desktop: Non desktop display (HMD). */ bool non_desktop; @@ -2286,9 +2291,6 @@ struct drm_connector { */ struct llist_node free_node; - /** @hdr_sink_metadata: HDR Metadata Information read from sink */ - struct hdr_sink_metadata hdr_sink_metadata; - /** * @hdmi: HDMI-related variable and properties. */ diff --git a/include/drm/drm_debugfs.h b/include/drm/drm_debugfs.h index cf06cee4343f..ea8cba94208a 100644 --- a/include/drm/drm_debugfs.h +++ b/include/drm/drm_debugfs.h @@ -153,6 +153,9 @@ void drm_debugfs_add_files(struct drm_device *dev, int drm_debugfs_gpuva_info(struct seq_file *m, struct drm_gpuvm *gpuvm); + +void drm_debugfs_clients_add(struct drm_file *file); +void drm_debugfs_clients_remove(struct drm_file *file); #else static inline void drm_debugfs_create_files(const struct drm_info_list *files, int count, struct dentry *root, @@ -181,6 +184,14 @@ static inline int drm_debugfs_gpuva_info(struct seq_file *m, { return 0; } + +static inline void drm_debugfs_clients_add(struct drm_file *file) +{ +} + +static inline void drm_debugfs_clients_remove(struct drm_file *file) +{ +} #endif #endif /* _DRM_DEBUGFS_H_ */ diff --git a/include/drm/drm_drv.h b/include/drm/drm_drv.h index 3f76a32d6b84..42fc085f986d 100644 --- a/include/drm/drm_drv.h +++ b/include/drm/drm_drv.h @@ -572,9 +572,24 @@ static inline bool drm_firmware_drivers_only(void) } #if defined(CONFIG_DEBUG_FS) -void drm_debugfs_dev_init(struct drm_device *dev, struct dentry *root); +void drm_debugfs_dev_init(struct drm_device *dev); +void drm_debugfs_init_root(void); +void drm_debugfs_remove_root(void); +void drm_debugfs_bridge_params(void); #else -static inline void drm_debugfs_dev_init(struct drm_device *dev, struct dentry *root) +static inline void drm_debugfs_dev_init(struct drm_device *dev) +{ +} + +static inline void drm_debugfs_init_root(void) +{ +} + +static inline void drm_debugfs_remove_root(void) +{ +} + +static inline void drm_debugfs_bridge_params(void) { } #endif diff --git a/include/drm/drm_file.h b/include/drm/drm_file.h index 5c3b2aa3e69d..eab7546aad79 100644 --- a/include/drm/drm_file.h +++ b/include/drm/drm_file.h @@ -400,6 +400,13 @@ struct drm_file { * @client_name_lock: Protects @client_name. */ struct mutex client_name_lock; + + /** + * @debugfs_client: + * + * debugfs directory for each client under a drm node. + */ + struct dentry *debugfs_client; }; /** diff --git a/include/drm/drm_format_helper.h b/include/drm/drm_format_helper.h index 0d3ee2a1313f..562bc383ece4 100644 --- a/include/drm/drm_format_helper.h +++ b/include/drm/drm_format_helper.h @@ -82,8 +82,10 @@ void drm_fb_xrgb8888_to_rgb332(struct iosys_map *dst, const unsigned int *dst_pi const struct drm_rect *clip, struct drm_format_conv_state *state); void drm_fb_xrgb8888_to_rgb565(struct iosys_map *dst, const unsigned int *dst_pitch, const struct iosys_map *src, const struct drm_framebuffer *fb, - const struct drm_rect *clip, struct drm_format_conv_state *state, - bool swab); + const struct drm_rect *clip, struct drm_format_conv_state *state); +void drm_fb_xrgb8888_to_rgb565be(struct iosys_map *dst, const unsigned int *dst_pitch, + const struct iosys_map *src, const struct drm_framebuffer *fb, + const struct drm_rect *clip, struct drm_format_conv_state *state); void drm_fb_xrgb8888_to_xrgb1555(struct iosys_map *dst, const unsigned int *dst_pitch, const struct iosys_map *src, const struct drm_framebuffer *fb, const struct drm_rect *clip, struct drm_format_conv_state *state); diff --git a/include/drm/drm_gem.h b/include/drm/drm_gem.h index 1a79ec3fe45c..d3a7b43e2c63 100644 --- a/include/drm/drm_gem.h +++ b/include/drm/drm_gem.h @@ -560,10 +560,12 @@ void drm_gem_lru_init(struct drm_gem_lru *lru, struct mutex *lock); void drm_gem_lru_remove(struct drm_gem_object *obj); void drm_gem_lru_move_tail_locked(struct drm_gem_lru *lru, struct drm_gem_object *obj); void drm_gem_lru_move_tail(struct drm_gem_lru *lru, struct drm_gem_object *obj); -unsigned long drm_gem_lru_scan(struct drm_gem_lru *lru, - unsigned int nr_to_scan, - unsigned long *remaining, - bool (*shrink)(struct drm_gem_object *obj)); +unsigned long +drm_gem_lru_scan(struct drm_gem_lru *lru, + unsigned int nr_to_scan, + unsigned long *remaining, + bool (*shrink)(struct drm_gem_object *obj, struct ww_acquire_ctx *ticket), + struct ww_acquire_ctx *ticket); int drm_gem_evict_locked(struct drm_gem_object *obj); diff --git a/include/drm/drm_gem_shmem_helper.h b/include/drm/drm_gem_shmem_helper.h index 35f7466dca84..92f5db84b9c2 100644 --- a/include/drm/drm_gem_shmem_helper.h +++ b/include/drm/drm_gem_shmem_helper.h @@ -293,23 +293,11 @@ struct drm_gem_object *drm_gem_shmem_prime_import_no_map(struct drm_device *dev, /** * DRM_GEM_SHMEM_DRIVER_OPS - Default shmem GEM operations * - * This macro provides a shortcut for setting the shmem GEM operations in - * the &drm_driver structure. + * This macro provides a shortcut for setting the shmem GEM operations + * in the &drm_driver structure. Drivers that do not require an s/g table + * for imported buffers should use this. */ #define DRM_GEM_SHMEM_DRIVER_OPS \ - .gem_prime_import_sg_table = drm_gem_shmem_prime_import_sg_table, \ - .dumb_create = drm_gem_shmem_dumb_create - -/** - * DRM_GEM_SHMEM_DRIVER_OPS_NO_MAP_SGT - shmem GEM operations - * without mapping sg_table on - * imported buffer. - * - * This macro provides a shortcut for setting the shmem GEM operations in - * the &drm_driver structure for drivers that do not require a sg_table on - * imported buffers. - */ -#define DRM_GEM_SHMEM_DRIVER_OPS_NO_MAP_SGT \ .gem_prime_import = drm_gem_shmem_prime_import_no_map, \ .dumb_create = drm_gem_shmem_dumb_create diff --git a/include/drm/drm_gpusvm.h b/include/drm/drm_gpusvm.h index 6a5156476bf4..4aedc5423aff 100644 --- a/include/drm/drm_gpusvm.h +++ b/include/drm/drm_gpusvm.h @@ -16,92 +16,10 @@ struct drm_gpusvm; struct drm_gpusvm_notifier; struct drm_gpusvm_ops; struct drm_gpusvm_range; -struct drm_gpusvm_devmem; struct drm_pagemap; struct drm_pagemap_device_addr; /** - * struct drm_gpusvm_devmem_ops - Operations structure for GPU SVM device memory - * - * This structure defines the operations for GPU Shared Virtual Memory (SVM) - * device memory. These operations are provided by the GPU driver to manage device memory - * allocations and perform operations such as migration between device memory and system - * RAM. - */ -struct drm_gpusvm_devmem_ops { - /** - * @devmem_release: Release device memory allocation (optional) - * @devmem_allocation: device memory allocation - * - * Release device memory allocation and drop a reference to device - * memory allocation. - */ - void (*devmem_release)(struct drm_gpusvm_devmem *devmem_allocation); - - /** - * @populate_devmem_pfn: Populate device memory PFN (required for migration) - * @devmem_allocation: device memory allocation - * @npages: Number of pages to populate - * @pfn: Array of page frame numbers to populate - * - * Populate device memory page frame numbers (PFN). - * - * Return: 0 on success, a negative error code on failure. - */ - int (*populate_devmem_pfn)(struct drm_gpusvm_devmem *devmem_allocation, - unsigned long npages, unsigned long *pfn); - - /** - * @copy_to_devmem: Copy to device memory (required for migration) - * @pages: Pointer to array of device memory pages (destination) - * @dma_addr: Pointer to array of DMA addresses (source) - * @npages: Number of pages to copy - * - * Copy pages to device memory. - * - * Return: 0 on success, a negative error code on failure. - */ - int (*copy_to_devmem)(struct page **pages, - dma_addr_t *dma_addr, - unsigned long npages); - - /** - * @copy_to_ram: Copy to system RAM (required for migration) - * @pages: Pointer to array of device memory pages (source) - * @dma_addr: Pointer to array of DMA addresses (destination) - * @npages: Number of pages to copy - * - * Copy pages to system RAM. - * - * Return: 0 on success, a negative error code on failure. - */ - int (*copy_to_ram)(struct page **pages, - dma_addr_t *dma_addr, - unsigned long npages); -}; - -/** - * struct drm_gpusvm_devmem - Structure representing a GPU SVM device memory allocation - * - * @dev: Pointer to the device structure which device memory allocation belongs to - * @mm: Pointer to the mm_struct for the address space - * @detached: device memory allocations is detached from device pages - * @ops: Pointer to the operations structure for GPU SVM device memory - * @dpagemap: The struct drm_pagemap of the pages this allocation belongs to. - * @size: Size of device memory allocation - * @timeslice_expiration: Timeslice expiration in jiffies - */ -struct drm_gpusvm_devmem { - struct device *dev; - struct mm_struct *mm; - struct completion detached; - const struct drm_gpusvm_devmem_ops *ops; - struct drm_pagemap *dpagemap; - size_t size; - u64 timeslice_expiration; -}; - -/** * struct drm_gpusvm_ops - Operations structure for GPU SVM * * This structure defines the operations for GPU Shared Virtual Memory (SVM). @@ -361,15 +279,6 @@ void drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm, struct drm_gpusvm_range *range, const struct drm_gpusvm_ctx *ctx); -int drm_gpusvm_migrate_to_devmem(struct drm_gpusvm *gpusvm, - struct drm_gpusvm_range *range, - struct drm_gpusvm_devmem *devmem_allocation, - const struct drm_gpusvm_ctx *ctx); - -int drm_gpusvm_evict_to_ram(struct drm_gpusvm_devmem *devmem_allocation); - -const struct dev_pagemap_ops *drm_gpusvm_pagemap_ops_get(void); - bool drm_gpusvm_has_mapping(struct drm_gpusvm *gpusvm, unsigned long start, unsigned long end); @@ -380,11 +289,6 @@ drm_gpusvm_range_find(struct drm_gpusvm_notifier *notifier, unsigned long start, void drm_gpusvm_range_set_unmapped(struct drm_gpusvm_range *range, const struct mmu_notifier_range *mmu_range); -void drm_gpusvm_devmem_init(struct drm_gpusvm_devmem *devmem_allocation, - struct device *dev, struct mm_struct *mm, - const struct drm_gpusvm_devmem_ops *ops, - struct drm_pagemap *dpagemap, size_t size); - #ifdef CONFIG_LOCKDEP /** * drm_gpusvm_driver_set_lock() - Set the lock protecting accesses to GPU SVM diff --git a/include/drm/drm_gpuvm.h b/include/drm/drm_gpuvm.h index 2a9629377633..274532facfd6 100644 --- a/include/drm/drm_gpuvm.h +++ b/include/drm/drm_gpuvm.h @@ -1211,6 +1211,14 @@ int drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm, void *priv, int drm_gpuvm_sm_unmap(struct drm_gpuvm *gpuvm, void *priv, u64 addr, u64 range); +int drm_gpuvm_sm_map_exec_lock(struct drm_gpuvm *gpuvm, + struct drm_exec *exec, unsigned int num_fences, + u64 req_addr, u64 req_range, + struct drm_gem_object *obj, u64 offset); + +int drm_gpuvm_sm_unmap_exec_lock(struct drm_gpuvm *gpuvm, struct drm_exec *exec, + u64 req_addr, u64 req_range); + void drm_gpuva_map(struct drm_gpuvm *gpuvm, struct drm_gpuva *va, struct drm_gpuva_op_map *op); diff --git a/include/drm/drm_managed.h b/include/drm/drm_managed.h index 53017cc609ac..72bfac002c06 100644 --- a/include/drm/drm_managed.h +++ b/include/drm/drm_managed.h @@ -129,14 +129,25 @@ void __drmm_mutex_release(struct drm_device *dev, void *res); void __drmm_workqueue_release(struct drm_device *device, void *wq); +/** + * drmm_alloc_ordered_workqueue - &drm_device managed alloc_ordered_workqueue() + * @dev: DRM device + * @fmt: printf format for the name of the workqueue + * @flags: WQ_* flags (only WQ_FREEZABLE and WQ_MEM_RECLAIM are meaningful) + * @args: args for @fmt + * + * This is a &drm_device-managed version of alloc_ordered_workqueue(). The + * allocated workqueue is automatically destroyed on the final drm_dev_put(). + * + * Returns: workqueue on success, negative ERR_PTR otherwise. + */ #define drmm_alloc_ordered_workqueue(dev, fmt, flags, args...) \ ({ \ struct workqueue_struct *wq = alloc_ordered_workqueue(fmt, flags, ##args); \ wq ? ({ \ int ret = drmm_add_action_or_reset(dev, __drmm_workqueue_release, wq); \ ret ? ERR_PTR(ret) : wq; \ - }) : \ - wq; \ + }) : ERR_PTR(-ENOMEM); \ }) #endif diff --git a/include/drm/drm_mipi_dsi.h b/include/drm/drm_mipi_dsi.h index b37860f4a895..369b0d8830c3 100644 --- a/include/drm/drm_mipi_dsi.h +++ b/include/drm/drm_mipi_dsi.h @@ -130,8 +130,6 @@ struct mipi_dsi_host *of_find_mipi_dsi_host_by_node(struct device_node *node); #define MIPI_DSI_MODE_VIDEO_NO_HBP BIT(6) /* disable hsync-active area */ #define MIPI_DSI_MODE_VIDEO_NO_HSA BIT(7) -/* flush display FIFO on vsync pulse */ -#define MIPI_DSI_MODE_VSYNC_FLUSH BIT(8) /* disable EoT packets in HS mode */ #define MIPI_DSI_MODE_NO_EOT_PACKET BIT(9) /* device supports non-continuous clock behavior (DSI spec 5.6.1) */ diff --git a/include/drm/drm_pagemap.h b/include/drm/drm_pagemap.h index 202c157ff4d7..e5f20a1235be 100644 --- a/include/drm/drm_pagemap.h +++ b/include/drm/drm_pagemap.h @@ -7,6 +7,7 @@ #include <linux/types.h> struct drm_pagemap; +struct drm_pagemap_zdd; struct device; /** @@ -91,6 +92,35 @@ struct drm_pagemap_ops { struct device *dev, struct drm_pagemap_device_addr addr); + /** + * @populate_mm: Populate part of the mm with @dpagemap memory, + * migrating existing data. + * @dpagemap: The struct drm_pagemap managing the memory. + * @start: The virtual start address in @mm + * @end: The virtual end address in @mm + * @mm: Pointer to a live mm. The caller must have an mmget() + * reference. + * + * The caller will have the mm lock at least in read mode. + * Note that there is no guarantee that the memory is resident + * after the function returns, it's best effort only. + * When the mm is not using the memory anymore, + * it will be released. The struct drm_pagemap might have a + * mechanism in place to reclaim the memory and the data will + * then be migrated. Typically to system memory. + * The implementation should hold sufficient runtime power- + * references while pages are used in an address space and + * should ideally guard against hardware device unbind in + * a way such that device pages are migrated back to system + * followed by device page removal. The implementation should + * return -ENODEV after device removal. + * + * Return: 0 if successful. Negative error code on error. + */ + int (*populate_mm)(struct drm_pagemap *dpagemap, + unsigned long start, unsigned long end, + struct mm_struct *mm, + unsigned long timeslice_ms); }; /** @@ -104,4 +134,109 @@ struct drm_pagemap { struct device *dev; }; +struct drm_pagemap_devmem; + +/** + * struct drm_pagemap_devmem_ops - Operations structure for GPU SVM device memory + * + * This structure defines the operations for GPU Shared Virtual Memory (SVM) + * device memory. These operations are provided by the GPU driver to manage device memory + * allocations and perform operations such as migration between device memory and system + * RAM. + */ +struct drm_pagemap_devmem_ops { + /** + * @devmem_release: Release device memory allocation (optional) + * @devmem_allocation: device memory allocation + * + * Release device memory allocation and drop a reference to device + * memory allocation. + */ + void (*devmem_release)(struct drm_pagemap_devmem *devmem_allocation); + + /** + * @populate_devmem_pfn: Populate device memory PFN (required for migration) + * @devmem_allocation: device memory allocation + * @npages: Number of pages to populate + * @pfn: Array of page frame numbers to populate + * + * Populate device memory page frame numbers (PFN). + * + * Return: 0 on success, a negative error code on failure. + */ + int (*populate_devmem_pfn)(struct drm_pagemap_devmem *devmem_allocation, + unsigned long npages, unsigned long *pfn); + + /** + * @copy_to_devmem: Copy to device memory (required for migration) + * @pages: Pointer to array of device memory pages (destination) + * @dma_addr: Pointer to array of DMA addresses (source) + * @npages: Number of pages to copy + * + * Copy pages to device memory. + * + * Return: 0 on success, a negative error code on failure. + */ + int (*copy_to_devmem)(struct page **pages, + dma_addr_t *dma_addr, + unsigned long npages); + + /** + * @copy_to_ram: Copy to system RAM (required for migration) + * @pages: Pointer to array of device memory pages (source) + * @dma_addr: Pointer to array of DMA addresses (destination) + * @npages: Number of pages to copy + * + * Copy pages to system RAM. + * + * Return: 0 on success, a negative error code on failure. + */ + int (*copy_to_ram)(struct page **pages, + dma_addr_t *dma_addr, + unsigned long npages); +}; + +/** + * struct drm_pagemap_devmem - Structure representing a GPU SVM device memory allocation + * + * @dev: Pointer to the device structure which device memory allocation belongs to + * @mm: Pointer to the mm_struct for the address space + * @detached: device memory allocations is detached from device pages + * @ops: Pointer to the operations structure for GPU SVM device memory + * @dpagemap: The struct drm_pagemap of the pages this allocation belongs to. + * @size: Size of device memory allocation + * @timeslice_expiration: Timeslice expiration in jiffies + */ +struct drm_pagemap_devmem { + struct device *dev; + struct mm_struct *mm; + struct completion detached; + const struct drm_pagemap_devmem_ops *ops; + struct drm_pagemap *dpagemap; + size_t size; + u64 timeslice_expiration; +}; + +int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation, + struct mm_struct *mm, + unsigned long start, unsigned long end, + unsigned long timeslice_ms, + void *pgmap_owner); + +int drm_pagemap_evict_to_ram(struct drm_pagemap_devmem *devmem_allocation); + +const struct dev_pagemap_ops *drm_pagemap_pagemap_ops_get(void); + +struct drm_pagemap *drm_pagemap_page_to_dpagemap(struct page *page); + +void drm_pagemap_devmem_init(struct drm_pagemap_devmem *devmem_allocation, + struct device *dev, struct mm_struct *mm, + const struct drm_pagemap_devmem_ops *ops, + struct drm_pagemap *dpagemap, size_t size); + +int drm_pagemap_populate_mm(struct drm_pagemap *dpagemap, + unsigned long start, unsigned long end, + struct mm_struct *mm, + unsigned long timeslice_ms); + #endif diff --git a/include/drm/intel/pciids.h b/include/drm/intel/pciids.h index 129c3172aa5f..76f8d26f9cc9 100644 --- a/include/drm/intel/pciids.h +++ b/include/drm/intel/pciids.h @@ -846,6 +846,7 @@ /* BMG */ #define INTEL_BMG_IDS(MACRO__, ...) \ MACRO__(0xE202, ## __VA_ARGS__), \ + MACRO__(0xE209, ## __VA_ARGS__), \ MACRO__(0xE20B, ## __VA_ARGS__), \ MACRO__(0xE20C, ## __VA_ARGS__), \ MACRO__(0xE20D, ## __VA_ARGS__), \ @@ -871,6 +872,8 @@ MACRO__(0xB08F, ## __VA_ARGS__), \ MACRO__(0xB090, ## __VA_ARGS__), \ MACRO__(0xB0A0, ## __VA_ARGS__), \ - MACRO__(0xB0B0, ## __VA_ARGS__) + MACRO__(0xB0B0, ## __VA_ARGS__), \ + MACRO__(0xFD80, ## __VA_ARGS__), \ + MACRO__(0xFD81, ## __VA_ARGS__) #endif /* __PCIIDS_H__ */ diff --git a/include/drm/ttm/ttm_bo.h b/include/drm/ttm/ttm_bo.h index 5332ee74d95b..479b7ed075c0 100644 --- a/include/drm/ttm/ttm_bo.h +++ b/include/drm/ttm/ttm_bo.h @@ -207,11 +207,9 @@ struct ttm_lru_walk_ops { }; /** - * struct ttm_lru_walk - Structure describing a LRU walk. + * struct ttm_lru_walk_arg - Common part for the variants of BO LRU walk. */ -struct ttm_lru_walk { - /** @ops: Pointer to the ops structure. */ - const struct ttm_lru_walk_ops *ops; +struct ttm_lru_walk_arg { /** @ctx: Pointer to the struct ttm_operation_ctx. */ struct ttm_operation_ctx *ctx; /** @ticket: The struct ww_acquire_ctx if any. */ @@ -220,6 +218,16 @@ struct ttm_lru_walk { bool trylock_only; }; +/** + * struct ttm_lru_walk - Structure describing a LRU walk. + */ +struct ttm_lru_walk { + /** @ops: Pointer to the ops structure. */ + const struct ttm_lru_walk_ops *ops; + /** @arg: Common bo LRU walk arguments. */ + struct ttm_lru_walk_arg arg; +}; + s64 ttm_lru_walk_for_evict(struct ttm_lru_walk *walk, struct ttm_device *bdev, struct ttm_resource_manager *man, s64 target); @@ -468,11 +476,6 @@ struct ttm_bo_lru_cursor { /** @res_curs: Embedded struct ttm_resource_cursor. */ struct ttm_resource_cursor res_curs; /** - * @ctx: The struct ttm_operation_ctx used while looping. - * governs the locking mode. - */ - struct ttm_operation_ctx *ctx; - /** * @bo: Buffer object pointer if a buffer object is refcounted, * NULL otherwise. */ @@ -482,6 +485,8 @@ struct ttm_bo_lru_cursor { * unlock before the next iteration or after loop exit. */ bool needs_unlock; + /** @arg: Pointer to common BO LRU walk arguments. */ + struct ttm_lru_walk_arg *arg; }; void ttm_bo_lru_cursor_fini(struct ttm_bo_lru_cursor *curs); @@ -489,7 +494,7 @@ void ttm_bo_lru_cursor_fini(struct ttm_bo_lru_cursor *curs); struct ttm_bo_lru_cursor * ttm_bo_lru_cursor_init(struct ttm_bo_lru_cursor *curs, struct ttm_resource_manager *man, - struct ttm_operation_ctx *ctx); + struct ttm_lru_walk_arg *arg); struct ttm_buffer_object *ttm_bo_lru_cursor_first(struct ttm_bo_lru_cursor *curs); @@ -500,9 +505,9 @@ struct ttm_buffer_object *ttm_bo_lru_cursor_next(struct ttm_bo_lru_cursor *curs) */ DEFINE_CLASS(ttm_bo_lru_cursor, struct ttm_bo_lru_cursor *, if (_T) {ttm_bo_lru_cursor_fini(_T); }, - ttm_bo_lru_cursor_init(curs, man, ctx), + ttm_bo_lru_cursor_init(curs, man, arg), struct ttm_bo_lru_cursor *curs, struct ttm_resource_manager *man, - struct ttm_operation_ctx *ctx); + struct ttm_lru_walk_arg *arg); static inline void * class_ttm_bo_lru_cursor_lock_ptr(class_ttm_bo_lru_cursor_t *_T) { return *_T; } @@ -513,7 +518,7 @@ class_ttm_bo_lru_cursor_lock_ptr(class_ttm_bo_lru_cursor_t *_T) * resources on LRU lists. * @_cursor: struct ttm_bo_lru_cursor to use for the iteration. * @_man: The resource manager whose LRU lists to iterate over. - * @_ctx: The struct ttm_operation_context to govern the @_bo locking. + * @_arg: The struct ttm_lru_walk_arg to govern the LRU walk. * @_bo: The struct ttm_buffer_object pointer pointing to the buffer object * for the current iteration. * @@ -525,10 +530,15 @@ class_ttm_bo_lru_cursor_lock_ptr(class_ttm_bo_lru_cursor_t *_T) * up at looping termination, even if terminated prematurely by, for * example a return or break statement. Exiting the loop will also unlock * (if needed) and unreference @_bo. + * + * Return: If locking of a bo returns an error, then iteration is terminated + * and @_bo is set to a corresponding error pointer. It's illegal to + * dereference @_bo after loop exit. */ -#define ttm_bo_lru_for_each_reserved_guarded(_cursor, _man, _ctx, _bo) \ - scoped_guard(ttm_bo_lru_cursor, _cursor, _man, _ctx) \ - for ((_bo) = ttm_bo_lru_cursor_first(_cursor); (_bo); \ - (_bo) = ttm_bo_lru_cursor_next(_cursor)) +#define ttm_bo_lru_for_each_reserved_guarded(_cursor, _man, _arg, _bo) \ + scoped_guard(ttm_bo_lru_cursor, _cursor, _man, _arg) \ + for ((_bo) = ttm_bo_lru_cursor_first(_cursor); \ + !IS_ERR_OR_NULL(_bo); \ + (_bo) = ttm_bo_lru_cursor_next(_cursor)) #endif diff --git a/include/dt-bindings/power/qcom-rpmpd.h b/include/dt-bindings/power/qcom-rpmpd.h index d9b7bac30953..f15bcee7c928 100644 --- a/include/dt-bindings/power/qcom-rpmpd.h +++ b/include/dt-bindings/power/qcom-rpmpd.h @@ -240,6 +240,7 @@ #define RPMH_REGULATOR_LEVEL_TURBO_L2 432 #define RPMH_REGULATOR_LEVEL_TURBO_L3 448 #define RPMH_REGULATOR_LEVEL_TURBO_L4 452 +#define RPMH_REGULATOR_LEVEL_TURBO_L5 456 #define RPMH_REGULATOR_LEVEL_SUPER_TURBO 464 #define RPMH_REGULATOR_LEVEL_SUPER_TURBO_NO_CPR 480 diff --git a/include/linux/intel_dg_nvm_aux.h b/include/linux/intel_dg_nvm_aux.h new file mode 100644 index 000000000000..625d46a6b96e --- /dev/null +++ b/include/linux/intel_dg_nvm_aux.h @@ -0,0 +1,32 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright(c) 2019-2025, Intel Corporation. All rights reserved. + */ + +#ifndef __INTEL_DG_NVM_AUX_H__ +#define __INTEL_DG_NVM_AUX_H__ + +#include <linux/auxiliary_bus.h> +#include <linux/container_of.h> +#include <linux/ioport.h> +#include <linux/types.h> + +#define INTEL_DG_NVM_REGIONS 13 + +struct intel_dg_nvm_region { + const char *name; +}; + +struct intel_dg_nvm_dev { + struct auxiliary_device aux_dev; + bool writable_override; + bool non_posted_erase; + struct resource bar; + struct resource bar2; + const struct intel_dg_nvm_region *regions; +}; + +#define auxiliary_dev_to_intel_dg_nvm_dev(auxiliary_dev) \ + container_of(auxiliary_dev, struct intel_dg_nvm_dev, aux_dev) + +#endif /* __INTEL_DG_NVM_AUX_H__ */ diff --git a/include/linux/soc/qcom/ubwc.h b/include/linux/soc/qcom/ubwc.h new file mode 100644 index 000000000000..1ed8b1b16bc9 --- /dev/null +++ b/include/linux/soc/qcom/ubwc.h @@ -0,0 +1,75 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018, The Linux Foundation + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries. + */ + +#ifndef __QCOM_UBWC_H__ +#define __QCOM_UBWC_H__ + +#include <linux/bits.h> +#include <linux/types.h> + +struct qcom_ubwc_cfg_data { + u32 ubwc_enc_version; + /* Can be read from MDSS_BASE + 0x58 */ + u32 ubwc_dec_version; + + /** + * @ubwc_swizzle: Whether to enable level 1, 2 & 3 bank swizzling. + * + * UBWC 1.0 always enables all three levels. + * UBWC 2.0 removes level 1 bank swizzling, leaving levels 2 & 3. + * UBWC 4.0 adds the optional ability to disable levels 2 & 3. + */ + u32 ubwc_swizzle; +#define UBWC_SWIZZLE_ENABLE_LVL1 BIT(0) +#define UBWC_SWIZZLE_ENABLE_LVL2 BIT(1) +#define UBWC_SWIZZLE_ENABLE_LVL3 BIT(2) + + /** + * @highest_bank_bit: Highest Bank Bit + * + * The Highest Bank Bit value represents the bit of the highest + * DDR bank. This should ideally use DRAM type detection. + */ + int highest_bank_bit; + bool ubwc_bank_spread; + + /** + * @macrotile_mode: Macrotile Mode + * + * Whether to use 4-channel macrotiling mode or the newer + * 8-channel macrotiling mode introduced in UBWC 3.1. 0 is + * 4-channel and 1 is 8-channel. + */ + bool macrotile_mode; +}; + +#define UBWC_1_0 0x10000000 +#define UBWC_2_0 0x20000000 +#define UBWC_3_0 0x30000000 +#define UBWC_4_0 0x40000000 +#define UBWC_4_3 0x40030000 +#define UBWC_5_0 0x50000000 + +#if IS_ENABLED(CONFIG_QCOM_UBWC_CONFIG) +const struct qcom_ubwc_cfg_data *qcom_ubwc_config_get_data(void); +#else +static inline const struct qcom_ubwc_cfg_data *qcom_ubwc_config_get_data(void) +{ + return ERR_PTR(-EOPNOTSUPP); +} +#endif + +static inline bool qcom_ubwc_get_ubwc_mode(const struct qcom_ubwc_cfg_data *cfg) +{ + bool ret = cfg->ubwc_enc_version == UBWC_1_0; + + if (ret && !(cfg->ubwc_swizzle & UBWC_SWIZZLE_ENABLE_LVL1)) + pr_err("UBWC config discrepancy - level 1 swizzling disabled on UBWC 1.0\n"); + + return ret; +} + +#endif /* __QCOM_UBWC_H__ */ diff --git a/include/linux/sysfb.h b/include/linux/sysfb.h index 07cbab516942..b449665c686a 100644 --- a/include/linux/sysfb.h +++ b/include/linux/sysfb.h @@ -7,9 +7,13 @@ * Copyright (c) 2012-2013 David Herrmann <dh.herrmann@gmail.com> */ -#include <linux/kernel.h> +#include <linux/err.h> +#include <linux/types.h> + #include <linux/platform_data/simplefb.h> +struct device; +struct platform_device; struct screen_info; enum { diff --git a/include/uapi/drm/drm_fourcc.h b/include/uapi/drm/drm_fourcc.h index 6483f76a2165..ea91aa8afde9 100644 --- a/include/uapi/drm/drm_fourcc.h +++ b/include/uapi/drm/drm_fourcc.h @@ -210,6 +210,10 @@ extern "C" { #define DRM_FORMAT_RGBA1010102 fourcc_code('R', 'A', '3', '0') /* [31:0] R:G:B:A 10:10:10:2 little endian */ #define DRM_FORMAT_BGRA1010102 fourcc_code('B', 'A', '3', '0') /* [31:0] B:G:R:A 10:10:10:2 little endian */ +/* 48 bpp RGB */ +#define DRM_FORMAT_RGB161616 fourcc_code('R', 'G', '4', '8') /* [47:0] R:G:B 16:16:16 little endian */ +#define DRM_FORMAT_BGR161616 fourcc_code('B', 'G', '4', '8') /* [47:0] B:G:R 16:16:16 little endian */ + /* 64 bpp RGB */ #define DRM_FORMAT_XRGB16161616 fourcc_code('X', 'R', '4', '8') /* [63:0] x:R:G:B 16:16:16:16 little endian */ #define DRM_FORMAT_XBGR16161616 fourcc_code('X', 'B', '4', '8') /* [63:0] x:B:G:R 16:16:16:16 little endian */ @@ -218,7 +222,7 @@ extern "C" { #define DRM_FORMAT_ABGR16161616 fourcc_code('A', 'B', '4', '8') /* [63:0] A:B:G:R 16:16:16:16 little endian */ /* - * Floating point 64bpp RGB + * Half-Floating point - 16b/component * IEEE 754-2008 binary16 half-precision float * [15:0] sign:exponent:mantissa 1:5:10 */ @@ -228,6 +232,20 @@ extern "C" { #define DRM_FORMAT_ARGB16161616F fourcc_code('A', 'R', '4', 'H') /* [63:0] A:R:G:B 16:16:16:16 little endian */ #define DRM_FORMAT_ABGR16161616F fourcc_code('A', 'B', '4', 'H') /* [63:0] A:B:G:R 16:16:16:16 little endian */ +#define DRM_FORMAT_R16F fourcc_code('R', ' ', ' ', 'H') /* [15:0] R 16 little endian */ +#define DRM_FORMAT_GR1616F fourcc_code('G', 'R', ' ', 'H') /* [31:0] G:R 16:16 little endian */ +#define DRM_FORMAT_BGR161616F fourcc_code('B', 'G', 'R', 'H') /* [47:0] B:G:R 16:16:16 little endian */ + +/* + * Floating point - 32b/component + * IEEE 754-2008 binary32 float + * [31:0] sign:exponent:mantissa 1:8:23 + */ +#define DRM_FORMAT_R32F fourcc_code('R', ' ', ' ', 'F') /* [31:0] R 32 little endian */ +#define DRM_FORMAT_GR3232F fourcc_code('G', 'R', ' ', 'F') /* [63:0] R:G 32:32 little endian */ +#define DRM_FORMAT_BGR323232F fourcc_code('B', 'G', 'R', 'F') /* [95:0] R:G:B 32:32:32 little endian */ +#define DRM_FORMAT_ABGR32323232F fourcc_code('A', 'B', '8', 'F') /* [127:0] R:G:B:A 32:32:32:32 little endian */ + /* * RGBA format with 10-bit components packed in 64-bit per pixel, with 6 bits * of unused padding per component: diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h index 2342cb90857e..5c67294edc95 100644 --- a/include/uapi/drm/msm_drm.h +++ b/include/uapi/drm/msm_drm.h @@ -91,6 +91,32 @@ struct drm_msm_timespec { #define MSM_PARAM_UBWC_SWIZZLE 0x12 /* RO */ #define MSM_PARAM_MACROTILE_MODE 0x13 /* RO */ #define MSM_PARAM_UCHE_TRAP_BASE 0x14 /* RO */ +/* PRR (Partially Resident Region) is required for sparse residency: */ +#define MSM_PARAM_HAS_PRR 0x15 /* RO */ +/* MSM_PARAM_EN_VM_BIND is set to 1 to enable VM_BIND ops. + * + * With VM_BIND enabled, userspace is required to allocate iova and use the + * VM_BIND ops for map/unmap ioctls. MSM_INFO_SET_IOVA and MSM_INFO_GET_IOVA + * will be rejected. (The latter does not have a sensible meaning when a BO + * can have multiple and/or partial mappings.) + * + * With VM_BIND enabled, userspace does not include a submit_bo table in the + * SUBMIT ioctl (this will be rejected), the resident set is determined by + * the the VM_BIND ops. + * + * Enabling VM_BIND will fail on devices which do not have per-process pgtables. + * And it is not allowed to disable VM_BIND once it has been enabled. + * + * Enabling VM_BIND should be done (attempted) prior to allocating any BOs or + * submitqueues of type MSM_SUBMITQUEUE_VM_BIND. + * + * Relatedly, when VM_BIND mode is enabled, the kernel will not try to recover + * from GPU faults or failed async VM_BIND ops, in particular because it is + * difficult to communicate to userspace which op failed so that userspace + * could rewind and try again. When the VM is marked unusable, the SUBMIT + * ioctl will throw -EPIPE. + */ +#define MSM_PARAM_EN_VM_BIND 0x16 /* WO, once */ /* For backwards compat. The original support for preemption was based on * a single ring per priority level so # of priority levels equals the # @@ -114,6 +140,19 @@ struct drm_msm_param { #define MSM_BO_SCANOUT 0x00000001 /* scanout capable */ #define MSM_BO_GPU_READONLY 0x00000002 +/* Private buffers do not need to be explicitly listed in the SUBMIT + * ioctl, unless referenced by a drm_msm_gem_submit_cmd. Private + * buffers may NOT be imported/exported or used for scanout (or any + * other situation where buffers can be indefinitely pinned, but + * cases other than scanout are all kernel owned BOs which are not + * visible to userspace). + * + * In exchange for those constraints, all private BOs associated with + * a single context (drm_file) share a single dma_resv, and if there + * has been no eviction since the last submit, there are no per-BO + * bookeeping to do, significantly cutting the SUBMIT overhead. + */ +#define MSM_BO_NO_SHARE 0x00000004 #define MSM_BO_CACHE_MASK 0x000f0000 /* cache modes */ #define MSM_BO_CACHED 0x00010000 @@ -123,6 +162,7 @@ struct drm_msm_param { #define MSM_BO_FLAGS (MSM_BO_SCANOUT | \ MSM_BO_GPU_READONLY | \ + MSM_BO_NO_SHARE | \ MSM_BO_CACHE_MASK) struct drm_msm_gem_new { @@ -180,6 +220,17 @@ struct drm_msm_gem_cpu_fini { * Cmdstream Submission: */ +#define MSM_SYNCOBJ_RESET 0x00000001 /* Reset syncobj after wait. */ +#define MSM_SYNCOBJ_FLAGS ( \ + MSM_SYNCOBJ_RESET | \ + 0) + +struct drm_msm_syncobj { + __u32 handle; /* in, syncobj handle. */ + __u32 flags; /* in, from MSM_SUBMIT_SYNCOBJ_FLAGS */ + __u64 point; /* in, timepoint for timeline syncobjs. */ +}; + /* The value written into the cmdstream is logically: * * ((relocbuf->gpuaddr + reloc_offset) << shift) | or @@ -221,7 +272,10 @@ struct drm_msm_gem_submit_cmd { __u32 size; /* in, cmdstream size */ __u32 pad; __u32 nr_relocs; /* in, number of submit_reloc's */ - __u64 relocs; /* in, ptr to array of submit_reloc's */ + union { + __u64 relocs; /* in, ptr to array of submit_reloc's */ + __u64 iova; /* cmdstream address (for VM_BIND contexts) */ + }; }; /* Each buffer referenced elsewhere in the cmdstream submit (ie. the @@ -269,17 +323,6 @@ struct drm_msm_gem_submit_bo { MSM_SUBMIT_FENCE_SN_IN | \ 0) -#define MSM_SUBMIT_SYNCOBJ_RESET 0x00000001 /* Reset syncobj after wait. */ -#define MSM_SUBMIT_SYNCOBJ_FLAGS ( \ - MSM_SUBMIT_SYNCOBJ_RESET | \ - 0) - -struct drm_msm_gem_submit_syncobj { - __u32 handle; /* in, syncobj handle. */ - __u32 flags; /* in, from MSM_SUBMIT_SYNCOBJ_FLAGS */ - __u64 point; /* in, timepoint for timeline syncobjs. */ -}; - /* Each cmdstream submit consists of a table of buffers involved, and * one or more cmdstream buffers. This allows for conditional execution * (context-restore), and IB buffers needed for per tile/bin draw cmds. @@ -293,13 +336,80 @@ struct drm_msm_gem_submit { __u64 cmds; /* in, ptr to array of submit_cmd's */ __s32 fence_fd; /* in/out fence fd (see MSM_SUBMIT_FENCE_FD_IN/OUT) */ __u32 queueid; /* in, submitqueue id */ - __u64 in_syncobjs; /* in, ptr to array of drm_msm_gem_submit_syncobj */ - __u64 out_syncobjs; /* in, ptr to array of drm_msm_gem_submit_syncobj */ + __u64 in_syncobjs; /* in, ptr to array of drm_msm_syncobj */ + __u64 out_syncobjs; /* in, ptr to array of drm_msm_syncobj */ __u32 nr_in_syncobjs; /* in, number of entries in in_syncobj */ __u32 nr_out_syncobjs; /* in, number of entries in out_syncobj. */ __u32 syncobj_stride; /* in, stride of syncobj arrays. */ __u32 pad; /*in, reserved for future use, always 0. */ +}; +#define MSM_VM_BIND_OP_UNMAP 0 +#define MSM_VM_BIND_OP_MAP 1 +#define MSM_VM_BIND_OP_MAP_NULL 2 + +#define MSM_VM_BIND_OP_DUMP 1 +#define MSM_VM_BIND_OP_FLAGS ( \ + MSM_VM_BIND_OP_DUMP | \ + 0) + +/** + * struct drm_msm_vm_bind_op - bind/unbind op to run + */ +struct drm_msm_vm_bind_op { + /** @op: one of MSM_VM_BIND_OP_x */ + __u32 op; + /** @handle: GEM object handle, MBZ for UNMAP or MAP_NULL */ + __u32 handle; + /** @obj_offset: Offset into GEM object, MBZ for UNMAP or MAP_NULL */ + __u64 obj_offset; + /** @iova: Address to operate on */ + __u64 iova; + /** @range: Number of bites to to map/unmap */ + __u64 range; + /** @flags: Bitmask of MSM_VM_BIND_OP_FLAG_x */ + __u32 flags; + /** @pad: MBZ */ + __u32 pad; +}; + +#define MSM_VM_BIND_FENCE_FD_IN 0x00000001 +#define MSM_VM_BIND_FENCE_FD_OUT 0x00000002 +#define MSM_VM_BIND_FLAGS ( \ + MSM_VM_BIND_FENCE_FD_IN | \ + MSM_VM_BIND_FENCE_FD_OUT | \ + 0) + +/** + * struct drm_msm_vm_bind - Input of &DRM_IOCTL_MSM_VM_BIND + */ +struct drm_msm_vm_bind { + /** @flags: in, bitmask of MSM_VM_BIND_x */ + __u32 flags; + /** @nr_ops: the number of bind ops in this ioctl */ + __u32 nr_ops; + /** @fence_fd: in/out fence fd (see MSM_VM_BIND_FENCE_FD_IN/OUT) */ + __s32 fence_fd; + /** @queue_id: in, submitqueue id */ + __u32 queue_id; + /** @in_syncobjs: in, ptr to array of drm_msm_gem_syncobj */ + __u64 in_syncobjs; + /** @out_syncobjs: in, ptr to array of drm_msm_gem_syncobj */ + __u64 out_syncobjs; + /** @nr_in_syncobjs: in, number of entries in in_syncobj */ + __u32 nr_in_syncobjs; + /** @nr_out_syncobjs: in, number of entries in out_syncobj */ + __u32 nr_out_syncobjs; + /** @syncobj_stride: in, stride of syncobj arrays */ + __u32 syncobj_stride; + /** @op_stride: sizeof each struct drm_msm_vm_bind_op in @ops */ + __u32 op_stride; + union { + /** @op: used if num_ops == 1 */ + struct drm_msm_vm_bind_op op; + /** @ops: userptr to array of drm_msm_vm_bind_op if num_ops > 1 */ + __u64 ops; + }; }; #define MSM_WAIT_FENCE_BOOST 0x00000001 @@ -345,12 +455,19 @@ struct drm_msm_gem_madvise { /* * Draw queues allow the user to set specific submission parameter. Command * submissions specify a specific submitqueue to use. ID 0 is reserved for - * backwards compatibility as a "default" submitqueue + * backwards compatibility as a "default" submitqueue. + * + * Because VM_BIND async updates happen on the CPU, they must run on a + * virtual queue created with the flag MSM_SUBMITQUEUE_VM_BIND. If we had + * a way to do pgtable updates on the GPU, we could drop this restriction. */ #define MSM_SUBMITQUEUE_ALLOW_PREEMPT 0x00000001 +#define MSM_SUBMITQUEUE_VM_BIND 0x00000002 /* virtual queue for VM_BIND ops */ + #define MSM_SUBMITQUEUE_FLAGS ( \ MSM_SUBMITQUEUE_ALLOW_PREEMPT | \ + MSM_SUBMITQUEUE_VM_BIND | \ 0) /* @@ -388,6 +505,7 @@ struct drm_msm_submitqueue_query { #define DRM_MSM_SUBMITQUEUE_NEW 0x0A #define DRM_MSM_SUBMITQUEUE_CLOSE 0x0B #define DRM_MSM_SUBMITQUEUE_QUERY 0x0C +#define DRM_MSM_VM_BIND 0x0D #define DRM_IOCTL_MSM_GET_PARAM DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_GET_PARAM, struct drm_msm_param) #define DRM_IOCTL_MSM_SET_PARAM DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_SET_PARAM, struct drm_msm_param) @@ -401,6 +519,7 @@ struct drm_msm_submitqueue_query { #define DRM_IOCTL_MSM_SUBMITQUEUE_NEW DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_SUBMITQUEUE_NEW, struct drm_msm_submitqueue) #define DRM_IOCTL_MSM_SUBMITQUEUE_CLOSE DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_SUBMITQUEUE_CLOSE, __u32) #define DRM_IOCTL_MSM_SUBMITQUEUE_QUERY DRM_IOW (DRM_COMMAND_BASE + DRM_MSM_SUBMITQUEUE_QUERY, struct drm_msm_submitqueue_query) +#define DRM_IOCTL_MSM_VM_BIND DRM_IOWR(DRM_COMMAND_BASE + DRM_MSM_VM_BIND, struct drm_msm_vm_bind) #if defined(__cplusplus) } diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h index 8e8bbdec8c5c..e2426413488f 100644 --- a/include/uapi/drm/xe_drm.h +++ b/include/uapi/drm/xe_drm.h @@ -925,9 +925,9 @@ struct drm_xe_gem_mmap_offset { * - %DRM_XE_VM_CREATE_FLAG_LR_MODE - An LR, or Long Running VM accepts * exec submissions to its exec_queues that don't have an upper time * limit on the job execution time. But exec submissions to these - * don't allow any of the flags DRM_XE_SYNC_FLAG_SYNCOBJ, - * DRM_XE_SYNC_FLAG_TIMELINE_SYNCOBJ, DRM_XE_SYNC_FLAG_DMA_BUF, - * used as out-syncobjs, that is, together with DRM_XE_SYNC_FLAG_SIGNAL. + * don't allow any of the sync types DRM_XE_SYNC_TYPE_SYNCOBJ, + * DRM_XE_SYNC_TYPE_TIMELINE_SYNCOBJ, used as out-syncobjs, that is, + * together with sync flag DRM_XE_SYNC_FLAG_SIGNAL. * LR VMs can be created in recoverable page-fault mode using * DRM_XE_VM_CREATE_FLAG_FAULT_MODE, if the device supports it. * If that flag is omitted, the UMD can not rely on the slightly @@ -1394,7 +1394,7 @@ struct drm_xe_sync { /** * @timeline_value: Input for the timeline sync object. Needs to be - * different than 0 when used with %DRM_XE_SYNC_FLAG_TIMELINE_SYNCOBJ. + * different than 0 when used with %DRM_XE_SYNC_TYPE_TIMELINE_SYNCOBJ. */ __u64 timeline_value; diff --git a/rust/kernel/drm/driver.rs b/rust/kernel/drm/driver.rs index acb638086131..af93d46d03d3 100644 --- a/rust/kernel/drm/driver.rs +++ b/rust/kernel/drm/driver.rs @@ -10,7 +10,6 @@ use crate::{ drm, error::{to_result, Result}, prelude::*, - str::CStr, types::ARef, }; use macros::vtable; |