summaryrefslogtreecommitdiff
path: root/tools/perf/scripts/python
diff options
context:
space:
mode:
authorAlexandre Hamamdjian <azkali.limited@gmail.com>2026-08-24 22:58:22 +0700
committerBjorn Andersson <andersson@kernel.org>2026-09-30 16:57:13 -0500
commit5df998721b264a401f343ceeec78191d6f1cc96d (patch)
tree8a7dd7db62563a6a6636193ff27c1580a3c911a3 /tools/perf/scripts/python
parentb1d387319cdfc4d745553ce4e1f735e3a3b730c0 (diff)
arm64: dts: qcom: qcs8550-ayaneo-pocket-ds: add the lower DSI panel
The Pocket DS drives two panels. The lower one is a 768x1024 IPS LCD on a Sitronix ST7703 DDIC hanging off mdss_dsi1, and all of the pieces it needs have now landed: the ayaneo,pocket-ds-lower-panel compatible in the ST7703 binding, the SGMicro SGM3804 dual-rail charge pump driver and the Silergy SY7758 backlight driver. The SGM3804 on i2c0 supplies the panel bias. Its two rails are enabled by separate GPIOs and both are programmed to 5.2 V; that figure is the vendor register configuration (0x00 = 0x0c, 0x01 = 0x0c), which lands on 5.2 V in the driver's 4.0 V + 100 mV linear range. tlmm 152 enables the positive rail and tlmm 153 the negative one. Only the positive rail has a consumer, because the ST7703 binding models a single vcc-supply, so the negative rail is marked regulator-always-on to keep AVEE up while the panel is on; a better modelling of the pair is welcome. The SY7758 backlight sits on the same bus. Its enable line was determined on the device: driving tlmm 42 low makes the chip stop acknowledging on i2c while the TCA6408 on the same segment keeps responding, so tlmm 42 gates the SY7758 alone rather than the bus. The TCA6408 expander supplies the rest: line 0 gates the panel IOVCC rail, modelled as a fixed regulator, and line 2 is the panel reset. sde_dsi1_active keeps its output-high because tlmm 137 has no GPIO consumer anywhere in the tree; pinctrl is the only thing asserting it. The upper panel is not added here. Its Chipone ICNA3512 binding makes disp-supply and blvdd-supply mandatory, and this board has no separately controllable rail for either; the vendor device tree drives the upper panel purely through a platform enable GPIO and points the panel node at the simulator supply list. It will follow once that wiring is confirmed. Signed-off-by: Alexandre Hamamdjian <azkali.limited@gmail.com> Link: https://patch.msgid.link/20260824-pocketds-v6-4-370161574ca2@gmail.com Signed-off-by: Bjorn Andersson <andersson@kernel.org>
Diffstat (limited to 'tools/perf/scripts/python')
0 files changed, 0 insertions, 0 deletions