diff options
Diffstat (limited to 'drivers/gpu/drm')
56 files changed, 447 insertions, 181 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 6d83ccfa42ee..ef6e78224fdf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -353,7 +353,6 @@ enum amdgpu_kiq_irq { AMDGPU_CP_KIQ_IRQ_DRIVER0 = 0, AMDGPU_CP_KIQ_IRQ_LAST }; -#define SRIOV_USEC_TIMEOUT 1200000 /* wait 12 * 100ms for SRIOV */ #define MAX_KIQ_REG_WAIT 5000 /* in usecs, 5ms */ #define MAX_KIQ_REG_BAILOUT_INTERVAL 5 /* in msecs, 5ms */ #define MAX_KIQ_REG_TRY 1000 @@ -1124,6 +1123,7 @@ struct amdgpu_device { bool in_s3; bool in_s4; bool in_s0ix; + suspend_state_t last_suspend_state; enum pp_mp1_state mp1_state; struct amdgpu_doorbell_index doorbell_index; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index a30111d2c3ea..7f354cd532dc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -3510,6 +3510,7 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) amdgpu_device_mem_scratch_fini(adev); amdgpu_ib_pool_fini(adev); amdgpu_seq64_fini(adev); + amdgpu_doorbell_fini(adev); } if (adev->ip_blocks[i].version->funcs->sw_fini) { r = adev->ip_blocks[i].version->funcs->sw_fini(&adev->ip_blocks[i]); @@ -3643,6 +3644,13 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev) adev, adev->ip_blocks[i].version->type)) continue; + /* Since we skip suspend for S0i3, we need to cancel the delayed + * idle work here as the suspend callback never gets called. + */ + if (adev->in_s0ix && + adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GFX && + amdgpu_ip_version(adev, GC_HWIP, 0) >= IP_VERSION(10, 0, 0)) + cancel_delayed_work_sync(&adev->gfx.idle_work); /* skip suspend of gfx/mes and psp for S0ix * gfx is in gfxoff state, so on resume it will exit gfxoff just * like at runtime. PSP is also part of the always on hardware @@ -4851,7 +4859,6 @@ void amdgpu_device_fini_sw(struct amdgpu_device *adev) iounmap(adev->rmmio); adev->rmmio = NULL; - amdgpu_doorbell_fini(adev); drm_dev_exit(idx); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index dc2713ec95a5..9e738fae2b74 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -120,6 +120,8 @@ MODULE_FIRMWARE("amdgpu/vega20_ip_discovery.bin"); MODULE_FIRMWARE("amdgpu/raven_ip_discovery.bin"); MODULE_FIRMWARE("amdgpu/raven2_ip_discovery.bin"); MODULE_FIRMWARE("amdgpu/picasso_ip_discovery.bin"); +MODULE_FIRMWARE("amdgpu/arcturus_ip_discovery.bin"); +MODULE_FIRMWARE("amdgpu/aldebaran_ip_discovery.bin"); #define mmIP_DISCOVERY_VERSION 0x16A00 #define mmRCC_CONFIG_MEMSIZE 0xde3 diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c index 9f627caedc3f..667080cc9ae1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c @@ -75,11 +75,25 @@ static int amdgpu_dma_buf_attach(struct dma_buf *dmabuf, */ static int amdgpu_dma_buf_pin(struct dma_buf_attachment *attach) { - struct drm_gem_object *obj = attach->dmabuf->priv; - struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj); + struct dma_buf *dmabuf = attach->dmabuf; + struct amdgpu_bo *bo = gem_to_amdgpu_bo(dmabuf->priv); + u32 domains = bo->preferred_domains; - /* pin buffer into GTT */ - return amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT); + dma_resv_assert_held(dmabuf->resv); + + /* + * Try pinning into VRAM to allow P2P with RDMA NICs without ODP + * support if all attachments can do P2P. If any attachment can't do + * P2P just pin into GTT instead. + */ + list_for_each_entry(attach, &dmabuf->attachments, node) + if (!attach->peer2peer) + domains &= ~AMDGPU_GEM_DOMAIN_VRAM; + + if (domains & AMDGPU_GEM_DOMAIN_VRAM) + bo->flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED; + + return amdgpu_bo_pin(bo, domains); } /** @@ -134,9 +148,6 @@ static struct sg_table *amdgpu_dma_buf_map(struct dma_buf_attachment *attach, r = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx); if (r) return ERR_PTR(r); - - } else if (bo->tbo.resource->mem_type != TTM_PL_TT) { - return ERR_PTR(-EBUSY); } switch (bo->tbo.resource->mem_type) { @@ -184,7 +195,7 @@ static void amdgpu_dma_buf_unmap(struct dma_buf_attachment *attach, struct sg_table *sgt, enum dma_data_direction dir) { - if (sgt->sgl->page_link) { + if (sg_page(sgt->sgl)) { dma_unmap_sgtable(attach->dev, sgt, dir, 0); sg_free_table(sgt); kfree(sgt); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 26bf896f1444..24ee4710f807 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -2548,8 +2548,20 @@ static int amdgpu_pmops_suspend(struct device *dev) adev->in_s0ix = true; else if (amdgpu_acpi_is_s3_active(adev)) adev->in_s3 = true; - if (!adev->in_s0ix && !adev->in_s3) + if (!adev->in_s0ix && !adev->in_s3) { + /* don't allow going deep first time followed by s2idle the next time */ + if (adev->last_suspend_state != PM_SUSPEND_ON && + adev->last_suspend_state != pm_suspend_target_state) { + drm_err_once(drm_dev, "Unsupported suspend state %d\n", + pm_suspend_target_state); + return -EINVAL; + } return 0; + } + + /* cache the state last used for suspend */ + adev->last_suspend_state = pm_suspend_target_state; + return amdgpu_device_suspend(drm_dev, true); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 72af5e5a894a..cf2df7790077 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -1438,9 +1438,11 @@ static int amdgpu_gfx_run_cleaner_shader_job(struct amdgpu_ring *ring) struct amdgpu_device *adev = ring->adev; struct drm_gpu_scheduler *sched = &ring->sched; struct drm_sched_entity entity; + static atomic_t counter; struct dma_fence *f; struct amdgpu_job *job; struct amdgpu_ib *ib; + void *owner; int i, r; /* Initialize the scheduler entity */ @@ -1451,9 +1453,15 @@ static int amdgpu_gfx_run_cleaner_shader_job(struct amdgpu_ring *ring) goto err; } - r = amdgpu_job_alloc_with_ib(ring->adev, &entity, NULL, - 64, 0, - &job); + /* + * Use some unique dummy value as the owner to make sure we execute + * the cleaner shader on each submission. The value just need to change + * for each submission and is otherwise meaningless. + */ + owner = (void *)(unsigned long)atomic_inc_return(&counter); + + r = amdgpu_job_alloc_with_ib(ring->adev, &entity, owner, + 64, 0, &job); if (r) goto err; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 464625282872..ecb74ccf1d90 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -699,12 +699,10 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid, uint32_t flush_type, bool all_hub, uint32_t inst) { - u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT : - adev->usec_timeout; struct amdgpu_ring *ring = &adev->gfx.kiq[inst].ring; struct amdgpu_kiq *kiq = &adev->gfx.kiq[inst]; unsigned int ndw; - int r; + int r, cnt = 0; uint32_t seq; /* @@ -761,10 +759,21 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid, amdgpu_ring_commit(ring); spin_unlock(&adev->gfx.kiq[inst].ring_lock); - if (amdgpu_fence_wait_polling(ring, seq, usec_timeout) < 1) { + + r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT); + + might_sleep(); + while (r < 1 && cnt++ < MAX_KIQ_REG_TRY && + !amdgpu_reset_pending(adev->reset_domain)) { + msleep(MAX_KIQ_REG_BAILOUT_INTERVAL); + r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT); + } + + if (cnt > MAX_KIQ_REG_TRY) { dev_err(adev->dev, "timeout waiting for kiq fence\n"); r = -ETIME; - } + } else + r = 0; } error_unlock_reset: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 80cd6f5273db..0b9987781f76 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -163,8 +163,8 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain) * When GTT is just an alternative to VRAM make sure that we * only use it as fallback and still try to fill up VRAM first. */ - if (domain & abo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM && - !(adev->flags & AMD_IS_APU)) + if (abo->tbo.resource && !(adev->flags & AMD_IS_APU) && + domain & abo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) places[c].flags |= TTM_PL_FLAG_FALLBACK; c++; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c index 6da8994e0469..2d7f82e98df9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c @@ -24,6 +24,7 @@ #include <linux/dma-mapping.h> #include <drm/ttm/ttm_range_manager.h> +#include <drm/drm_drv.h> #include "amdgpu.h" #include "amdgpu_vm.h" @@ -907,6 +908,9 @@ int amdgpu_vram_mgr_init(struct amdgpu_device *adev) struct ttm_resource_manager *man = &mgr->manager; int err; + man->cg = drmm_cgroup_register_region(adev_to_drm(adev), "vram", adev->gmc.real_vram_size); + if (IS_ERR(man->cg)) + return PTR_ERR(man->cg); ttm_resource_manager_init(man, &adev->mman.bdev, adev->gmc.real_vram_size); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index a63ce747863f..23e6a05359c2 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -6114,7 +6114,7 @@ static int gfx_v10_0_cp_gfx_load_pfp_microcode(struct amdgpu_device *adev) } if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); tmp = RREG32_SOC15(GC, 0, mmCP_PFP_IC_BASE_CNTL); tmp = REG_SET_FIELD(tmp, CP_PFP_IC_BASE_CNTL, VMID, 0); @@ -6192,7 +6192,7 @@ static int gfx_v10_0_cp_gfx_load_ce_microcode(struct amdgpu_device *adev) } if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); tmp = RREG32_SOC15(GC, 0, mmCP_CE_IC_BASE_CNTL); tmp = REG_SET_FIELD(tmp, CP_CE_IC_BASE_CNTL, VMID, 0); @@ -6269,7 +6269,7 @@ static int gfx_v10_0_cp_gfx_load_me_microcode(struct amdgpu_device *adev) } if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); tmp = RREG32_SOC15(GC, 0, mmCP_ME_IC_BASE_CNTL); tmp = REG_SET_FIELD(tmp, CP_ME_IC_BASE_CNTL, VMID, 0); @@ -6644,7 +6644,7 @@ static int gfx_v10_0_cp_compute_load_microcode(struct amdgpu_device *adev) } if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); tmp = RREG32_SOC15(GC, 0, mmCP_CPC_IC_BASE_CNTL); tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, CACHE_POLICY, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index d57db42f9536..2a5c2a1ae3c7 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -2428,7 +2428,7 @@ static int gfx_v11_0_config_me_cache(struct amdgpu_device *adev, uint64_t addr) } if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); tmp = RREG32_SOC15(GC, 0, regCP_ME_IC_BASE_CNTL); tmp = REG_SET_FIELD(tmp, CP_ME_IC_BASE_CNTL, VMID, 0); @@ -2472,7 +2472,7 @@ static int gfx_v11_0_config_pfp_cache(struct amdgpu_device *adev, uint64_t addr) } if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); tmp = RREG32_SOC15(GC, 0, regCP_PFP_IC_BASE_CNTL); tmp = REG_SET_FIELD(tmp, CP_PFP_IC_BASE_CNTL, VMID, 0); @@ -2517,7 +2517,7 @@ static int gfx_v11_0_config_mec_cache(struct amdgpu_device *adev, uint64_t addr) } if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); tmp = RREG32_SOC15(GC, 0, regCP_CPC_IC_BASE_CNTL); tmp = REG_SET_FIELD(tmp, CP_CPC_IC_BASE_CNTL, CACHE_POLICY, 0); @@ -3153,7 +3153,7 @@ static int gfx_v11_0_cp_gfx_load_pfp_microcode_rs64(struct amdgpu_device *adev) amdgpu_bo_unreserve(adev->gfx.pfp.pfp_fw_data_obj); if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); WREG32_SOC15(GC, 0, regCP_PFP_IC_BASE_LO, lower_32_bits(adev->gfx.pfp.pfp_fw_gpu_addr)); @@ -3371,7 +3371,7 @@ static int gfx_v11_0_cp_gfx_load_me_microcode_rs64(struct amdgpu_device *adev) amdgpu_bo_unreserve(adev->gfx.me.me_fw_data_obj); if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); WREG32_SOC15(GC, 0, regCP_ME_IC_BASE_LO, lower_32_bits(adev->gfx.me.me_fw_gpu_addr)); @@ -4541,7 +4541,7 @@ static int gfx_v11_0_gfxhub_enable(struct amdgpu_device *adev) if (r) return r; - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ? false : true; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c index e7b58e470292..62a257a4a3e9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c @@ -2324,7 +2324,7 @@ static int gfx_v12_0_cp_gfx_load_pfp_microcode_rs64(struct amdgpu_device *adev) amdgpu_bo_unreserve(adev->gfx.pfp.pfp_fw_data_obj); if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); WREG32_SOC15(GC, 0, regCP_PFP_IC_BASE_LO, lower_32_bits(adev->gfx.pfp.pfp_fw_gpu_addr)); @@ -2468,7 +2468,7 @@ static int gfx_v12_0_cp_gfx_load_me_microcode_rs64(struct amdgpu_device *adev) amdgpu_bo_unreserve(adev->gfx.me.me_fw_data_obj); if (amdgpu_emu_mode == 1) - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); WREG32_SOC15(GC, 0, regCP_ME_IC_BASE_LO, lower_32_bits(adev->gfx.me.me_fw_gpu_addr)); @@ -3426,7 +3426,7 @@ static int gfx_v12_0_gfxhub_enable(struct amdgpu_device *adev) if (r) return r; - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ? false : true; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 95d894a231fc..809b3a882d0d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -268,7 +268,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng; /* flush hdp cache */ - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); /* This is necessary for SRIOV as well as for GFXOFF to function * properly under bare metal @@ -969,7 +969,7 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev) adev->hdp.funcs->init_registers(adev); /* Flush HDP after it is initialized */ - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ? false : true; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index ad099f136f84..e74e26b6a4f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -229,7 +229,7 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng; /* flush hdp cache */ - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); /* This is necessary for SRIOV as well as for GFXOFF to function * properly under bare metal @@ -899,7 +899,7 @@ static int gmc_v11_0_gart_enable(struct amdgpu_device *adev) return r; /* Flush HDP after it is initialized */ - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ? false : true; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c index 05c026d0b0d9..c6f290704d47 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c @@ -297,7 +297,7 @@ static void gmc_v12_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, return; /* flush hdp cache */ - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); /* This is necessary for SRIOV as well as for GFXOFF to function * properly under bare metal @@ -881,7 +881,7 @@ static int gmc_v12_0_gart_enable(struct amdgpu_device *adev) return r; /* Flush HDP after it is initialized */ - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ? false : true; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 783e0c3b86b4..5effe8327d29 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -2435,7 +2435,7 @@ static int gmc_v9_0_hw_init(struct amdgpu_ip_block *ip_block) adev->hdp.funcs->init_registers(adev); /* After HDP is initialized, flush HDP.*/ - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) value = false; diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index e65916ada23b..ef9538fbbf53 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -894,6 +894,10 @@ static void mes_v11_0_get_fw_version(struct amdgpu_device *adev) { int pipe; + /* return early if we have already fetched these */ + if (adev->mes.sched_version && adev->mes.kiq_version) + return; + /* get MES scheduler/KIQ versions */ mutex_lock(&adev->srbm_mutex); diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c index 183dd3346da5..e6ab617b9a40 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c @@ -1392,17 +1392,20 @@ static int mes_v12_0_queue_init(struct amdgpu_device *adev, mes_v12_0_queue_init_register(ring); } - /* get MES scheduler/KIQ versions */ - mutex_lock(&adev->srbm_mutex); - soc21_grbm_select(adev, 3, pipe, 0, 0); + if (((pipe == AMDGPU_MES_SCHED_PIPE) && !adev->mes.sched_version) || + ((pipe == AMDGPU_MES_KIQ_PIPE) && !adev->mes.kiq_version)) { + /* get MES scheduler/KIQ versions */ + mutex_lock(&adev->srbm_mutex); + soc21_grbm_select(adev, 3, pipe, 0, 0); - if (pipe == AMDGPU_MES_SCHED_PIPE) - adev->mes.sched_version = RREG32_SOC15(GC, 0, regCP_MES_GP3_LO); - else if (pipe == AMDGPU_MES_KIQ_PIPE && adev->enable_mes_kiq) - adev->mes.kiq_version = RREG32_SOC15(GC, 0, regCP_MES_GP3_LO); + if (pipe == AMDGPU_MES_SCHED_PIPE) + adev->mes.sched_version = RREG32_SOC15(GC, 0, regCP_MES_GP3_LO); + else if (pipe == AMDGPU_MES_KIQ_PIPE && adev->enable_mes_kiq) + adev->mes.kiq_version = RREG32_SOC15(GC, 0, regCP_MES_GP3_LO); - soc21_grbm_select(adev, 0, 0, 0, 0); - mutex_unlock(&adev->srbm_mutex); + soc21_grbm_select(adev, 0, 0, 0, 0); + mutex_unlock(&adev->srbm_mutex); + } return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index bb5dfc410a66..215543575f47 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -533,7 +533,7 @@ static int psp_v11_0_memory_training(struct psp_context *psp, uint32_t ops) } memcpy_toio(adev->mman.aper_base_kaddr, buf, sz); - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); vfree(buf); drm_dev_exit(idx); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c index cc621064610f..afdf8ce3b4c5 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c @@ -610,7 +610,7 @@ static int psp_v13_0_memory_training(struct psp_context *psp, uint32_t ops) } memcpy_toio(adev->mman.aper_base_kaddr, buf, sz); - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); vfree(buf); drm_dev_exit(idx); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c index 7c49c3f3c388..256288c6cd78 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c @@ -498,7 +498,7 @@ static int psp_v14_0_memory_training(struct psp_context *psp, uint32_t ops) } memcpy_toio(adev->mman.aper_base_kaddr, buf, sz); - adev->hdp.funcs->flush_hdp(adev, NULL); + amdgpu_device_flush_hdp(adev, NULL); vfree(buf); drm_dev_exit(idx); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index 86d8bc10d90a..9b3510e53112 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -239,6 +239,13 @@ static const struct amdgpu_video_codec_info cz_video_codecs_decode_array[] = .max_pixels_per_frame = 4096 * 4096, .max_level = 186, }, + { + .codec_type = AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_JPEG, + .max_width = 4096, + .max_height = 4096, + .max_pixels_per_frame = 4096 * 4096, + .max_level = 0, + }, }; static const struct amdgpu_video_codecs cz_video_codecs_decode = diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index e477d7509646..9bbee484d57c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -1983,9 +1983,6 @@ static void kfd_topology_set_capabilities(struct kfd_topology_device *dev) if (kfd_dbg_has_ttmps_always_setup(dev->gpu)) dev->node_props.debug_prop |= HSA_DBG_DISPATCH_INFO_ALWAYS_VALID; - if (dev->gpu->adev->sdma.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE) - dev->node_props.capability2 |= HSA_CAP2_PER_SDMA_QUEUE_RESET_SUPPORTED; - if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(10, 0, 0)) { if (KFD_GC_VERSION(dev->gpu) == IP_VERSION(9, 4, 3) || KFD_GC_VERSION(dev->gpu) == IP_VERSION(9, 4, 4)) @@ -2001,7 +1998,11 @@ static void kfd_topology_set_capabilities(struct kfd_topology_device *dev) dev->node_props.capability |= HSA_CAP_TRAP_DEBUG_PRECISE_MEMORY_OPERATIONS_SUPPORTED; - dev->node_props.capability |= HSA_CAP_PER_QUEUE_RESET_SUPPORTED; + 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/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index d0d8ad5368c3..9fed4471405f 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -1726,9 +1726,30 @@ static const struct dmi_system_id dmi_quirk_table[] = { .callback = edp0_on_dp1_callback, .matches = { DMI_MATCH(DMI_SYS_VENDOR, "HP"), + DMI_MATCH(DMI_PRODUCT_NAME, "HP EliteBook 645 14 inch G11 Notebook PC"), + }, + }, + { + .callback = edp0_on_dp1_callback, + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "HP"), DMI_MATCH(DMI_PRODUCT_NAME, "HP EliteBook 665 16 inch G11 Notebook PC"), }, }, + { + .callback = edp0_on_dp1_callback, + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "HP"), + DMI_MATCH(DMI_PRODUCT_NAME, "HP ProBook 445 14 inch G11 Notebook PC"), + }, + }, + { + .callback = edp0_on_dp1_callback, + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "HP"), + DMI_MATCH(DMI_PRODUCT_NAME, "HP ProBook 465 16 inch G11 Notebook PC"), + }, + }, {} /* TODO: refactor this from a fixed table to a dynamic option */ }; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c index 36a830a7440f..e8bdd7f0c460 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c @@ -113,6 +113,7 @@ bool amdgpu_dm_crtc_vrr_active(const struct dm_crtc_state *dm_state) * * Panel Replay and PSR SU * - Enable when: + * - VRR is disabled * - vblank counter is disabled * - entry is allowed: usermode demonstrates an adequate number of fast * commits) @@ -131,19 +132,20 @@ static void amdgpu_dm_crtc_set_panel_sr_feature( bool is_sr_active = (link->replay_settings.replay_allow_active || link->psr_settings.psr_allow_active); bool is_crc_window_active = false; + bool vrr_active = amdgpu_dm_crtc_vrr_active_irq(vblank_work->acrtc); #ifdef CONFIG_DRM_AMD_SECURE_DISPLAY is_crc_window_active = amdgpu_dm_crc_window_is_activated(&vblank_work->acrtc->base); #endif - if (link->replay_settings.replay_feature_enabled && + if (link->replay_settings.replay_feature_enabled && !vrr_active && allow_sr_entry && !is_sr_active && !is_crc_window_active) { amdgpu_dm_replay_enable(vblank_work->stream, true); } else if (vblank_enabled) { if (link->psr_settings.psr_version < DC_PSR_VERSION_SU_1 && is_sr_active) amdgpu_dm_psr_disable(vblank_work->stream, false); - } else if (link->psr_settings.psr_feature_enabled && + } else if (link->psr_settings.psr_feature_enabled && !vrr_active && allow_sr_entry && !is_sr_active && !is_crc_window_active) { struct amdgpu_dm_connector *aconn = @@ -244,6 +246,8 @@ static void amdgpu_dm_crtc_vblank_control_worker(struct work_struct *work) struct vblank_control_work *vblank_work = container_of(work, struct vblank_control_work, work); struct amdgpu_display_manager *dm = vblank_work->dm; + struct amdgpu_device *adev = drm_to_adev(dm->ddev); + int r; mutex_lock(&dm->dc_lock); @@ -271,8 +275,15 @@ static void amdgpu_dm_crtc_vblank_control_worker(struct work_struct *work) vblank_work->acrtc->dm_irq_params.allow_sr_entry); } - if (dm->active_vblank_irq_count == 0) + if (dm->active_vblank_irq_count == 0) { + r = amdgpu_dpm_pause_power_profile(adev, true); + if (r) + dev_warn(adev->dev, "failed to set default power profile mode\n"); dc_allow_idle_optimizations(dm->dc, true); + r = amdgpu_dpm_pause_power_profile(adev, false); + if (r) + dev_warn(adev->dev, "failed to restore the power profile mode\n"); + } mutex_unlock(&dm->dc_lock); 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 be54f0e696ce..5d16f36ec95c 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 @@ -2,6 +2,7 @@ // // Copyright 2024 Advanced Micro Devices, Inc. +#include <linux/vmalloc.h> #include "dml2_internal_types.h" #include "dml_top.h" @@ -13,11 +14,11 @@ static bool dml21_allocate_memory(struct dml2_context **dml_ctx) { - *dml_ctx = kzalloc(sizeof(struct dml2_context), GFP_KERNEL); + *dml_ctx = vzalloc(sizeof(struct dml2_context)); if (!(*dml_ctx)) return false; - (*dml_ctx)->v21.dml_init.dml2_instance = kzalloc(sizeof(struct dml2_instance), GFP_KERNEL); + (*dml_ctx)->v21.dml_init.dml2_instance = vzalloc(sizeof(struct dml2_instance)); if (!((*dml_ctx)->v21.dml_init.dml2_instance)) return false; @@ -27,7 +28,7 @@ static bool dml21_allocate_memory(struct dml2_context **dml_ctx) (*dml_ctx)->v21.mode_support.display_config = &(*dml_ctx)->v21.display_config; (*dml_ctx)->v21.mode_programming.display_config = (*dml_ctx)->v21.mode_support.display_config; - (*dml_ctx)->v21.mode_programming.programming = kzalloc(sizeof(struct dml2_display_cfg_programming), GFP_KERNEL); + (*dml_ctx)->v21.mode_programming.programming = vzalloc(sizeof(struct dml2_display_cfg_programming)); if (!((*dml_ctx)->v21.mode_programming.programming)) return false; @@ -86,6 +87,8 @@ static void dml21_init(const struct dc *in_dc, struct dml2_context **dml_ctx, co /* Store configuration options */ (*dml_ctx)->config = *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); @@ -96,6 +99,8 @@ static void dml21_init(const struct dc *in_dc, struct dml2_context **dml_ctx, co /*Initialize DML21 instance */ dml2_initialize_instance(&(*dml_ctx)->v21.dml_init); + + DC_FP_END(); } bool dml21_create(const struct dc *in_dc, struct dml2_context **dml_ctx, const struct dml2_configuration_options *config) @@ -111,8 +116,8 @@ bool dml21_create(const struct dc *in_dc, struct dml2_context **dml_ctx, const s void dml21_destroy(struct dml2_context *dml2) { - kfree(dml2->v21.dml_init.dml2_instance); - kfree(dml2->v21.mode_programming.programming); + vfree(dml2->v21.dml_init.dml2_instance); + vfree(dml2->v21.mode_programming.programming); } static void dml21_calculate_rq_and_dlg_params(const struct dc *dc, struct dc_state *context, struct resource_context *out_new_hw_state, @@ -283,11 +288,16 @@ bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml { bool out = false; + DC_FP_START(); + /* Use dml_validate_only for fast_validate path */ - if (fast_validate) { + if (fast_validate) out = dml21_check_mode_support(in_dc, context, dml_ctx); - } else + else out = dml21_mode_check_and_programming(in_dc, context, dml_ctx); + + DC_FP_END(); + return out; } @@ -426,8 +436,12 @@ void dml21_copy(struct dml2_context *dst_dml_ctx, dst_dml_ctx->v21.mode_programming.programming = dst_dml2_programming; + DC_FP_START(); + /* need to initialize copied instance for internal references to be correct */ dml2_initialize_instance(&dst_dml_ctx->v21.dml_init); + + DC_FP_END(); } bool dml21_create_copy(struct dml2_context **dst_dml_ctx, 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 939ee0708bd2..e89571874185 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c @@ -24,6 +24,8 @@ * */ +#include <linux/vmalloc.h> + #include "display_mode_core.h" #include "dml2_internal_types.h" #include "dml2_utils.h" @@ -732,17 +734,22 @@ bool dml2_validate(const struct dc *in_dc, struct dc_state *context, struct dml2 return out; } + DC_FP_START(); + /* Use dml_validate_only for fast_validate path */ if (fast_validate) out = dml2_validate_only(context); else out = dml2_validate_and_build_resource(in_dc, context); + + DC_FP_END(); + return out; } static inline struct dml2_context *dml2_allocate_memory(void) { - return (struct dml2_context *) kzalloc(sizeof(struct dml2_context), GFP_KERNEL); + return (struct dml2_context *) vzalloc(sizeof(struct dml2_context)); } static void dml2_init(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) @@ -779,11 +786,15 @@ static void dml2_init(const struct dc *in_dc, const struct dml2_configuration_op break; } + DC_FP_START(); + initialize_dml2_ip_params(*dml2, in_dc, &(*dml2)->v20.dml_core_ctx.ip); initialize_dml2_soc_bbox(*dml2, in_dc, &(*dml2)->v20.dml_core_ctx.soc); initialize_dml2_soc_states(*dml2, in_dc, &(*dml2)->v20.dml_core_ctx.soc, &(*dml2)->v20.dml_core_ctx.states); + + DC_FP_END(); } bool dml2_create(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) @@ -812,7 +823,7 @@ void dml2_destroy(struct dml2_context *dml2) if (dml2->architecture == dml2_architecture_21) dml21_destroy(dml2); - kfree(dml2); + vfree(dml2); } void dml2_extract_dram_and_fclk_change_support(struct dml2_context *dml2, diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index 2a9606118d89..21dc956b5f35 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -429,6 +429,7 @@ struct amd_pm_funcs { int (*set_pp_table)(void *handle, const char *buf, size_t size); void (*debugfs_print_current_performance_level)(void *handle, struct seq_file *m); int (*switch_power_profile)(void *handle, enum PP_SMC_POWER_PROFILE type, bool en); + int (*pause_power_profile)(void *handle, bool pause); /* export to amdgpu */ struct amd_vce_state *(*get_vce_clock_state)(void *handle, u32 idx); int (*dispatch_tasks)(void *handle, enum amd_pp_task task_id, diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c index 81e9b443ca0a..3533d43ed1e7 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c @@ -349,6 +349,25 @@ int amdgpu_dpm_switch_power_profile(struct amdgpu_device *adev, return ret; } +int amdgpu_dpm_pause_power_profile(struct amdgpu_device *adev, + bool pause) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (amdgpu_sriov_vf(adev)) + return 0; + + if (pp_funcs && pp_funcs->pause_power_profile) { + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->pause_power_profile( + adev->powerplay.pp_handle, pause); + mutex_unlock(&adev->pm.mutex); + } + + return ret; +} + int amdgpu_dpm_set_xgmi_pstate(struct amdgpu_device *adev, uint32_t pstate) { diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h index f93d287dbf13..4c0f7ad14816 100644 --- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h +++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h @@ -410,6 +410,8 @@ int amdgpu_dpm_set_xgmi_pstate(struct amdgpu_device *adev, int amdgpu_dpm_switch_power_profile(struct amdgpu_device *adev, enum PP_SMC_POWER_PROFILE type, bool en); +int amdgpu_dpm_pause_power_profile(struct amdgpu_device *adev, + bool pause); int amdgpu_dpm_baco_reset(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index 033c3229b555..46cce1d2aaf3 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -2398,7 +2398,11 @@ static int smu_switch_power_profile(void *handle, smu_power_profile_mode_get(smu, type); else smu_power_profile_mode_put(smu, type); - ret = smu_bump_power_profile_mode(smu, NULL, 0); + /* don't switch the active workload when paused */ + if (smu->pause_workload) + ret = 0; + else + ret = smu_bump_power_profile_mode(smu, NULL, 0); if (ret) { if (enable) smu_power_profile_mode_put(smu, type); @@ -2411,6 +2415,35 @@ static int smu_switch_power_profile(void *handle, return 0; } +static int smu_pause_power_profile(void *handle, + bool pause) +{ + struct smu_context *smu = handle; + struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); + u32 workload_mask = 1 << PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT; + int ret; + + if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) + return -EOPNOTSUPP; + + if (smu_dpm_ctx->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL && + smu_dpm_ctx->dpm_level != AMD_DPM_FORCED_LEVEL_PERF_DETERMINISM) { + smu->pause_workload = pause; + + /* force to bootup default profile */ + if (smu->pause_workload && smu->ppt_funcs->set_power_profile_mode) + ret = smu->ppt_funcs->set_power_profile_mode(smu, + workload_mask, + NULL, + 0); + else + ret = smu_bump_power_profile_mode(smu, NULL, 0); + return ret; + } + + return 0; +} + static enum amd_dpm_forced_level smu_get_performance_level(void *handle) { struct smu_context *smu = handle; @@ -3733,6 +3766,7 @@ static const struct amd_pm_funcs swsmu_pm_funcs = { .get_pp_table = smu_sys_get_pp_table, .set_pp_table = smu_sys_set_pp_table, .switch_power_profile = smu_switch_power_profile, + .pause_power_profile = smu_pause_power_profile, /* export to amdgpu */ .dispatch_tasks = smu_handle_dpm_task, .load_firmware = smu_load_microcode, 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 3ba169639f54..dd6d0e7aa242 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h @@ -558,6 +558,7 @@ struct smu_context { /* asic agnostic workload mask */ uint32_t workload_mask; + bool pause_workload; /* default/user workload preference */ uint32_t power_profile_mode; uint32_t workload_refcount[PP_SMC_POWER_PROFILE_COUNT]; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c index 78391d8f35a9..25fabf336a64 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c @@ -1204,7 +1204,7 @@ int smu_v11_0_set_fan_speed_rpm(struct smu_context *smu, uint32_t crystal_clock_freq = 2500; uint32_t tach_period; - if (speed == 0) + if (!speed || speed > UINT_MAX/8) return -EINVAL; /* * To prevent from possible overheat, some ASICs may have requirement diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c index 048be2872247..98b898a1de8f 100644 --- a/drivers/gpu/drm/i915/display/intel_bw.c +++ b/drivers/gpu/drm/i915/display/intel_bw.c @@ -244,6 +244,7 @@ static int icl_get_qgv_points(struct drm_i915_private *dev_priv, qi->deinterleave = 4; break; case INTEL_DRAM_GDDR: + case INTEL_DRAM_GDDR_ECC: qi->channel_width = 32; break; default: @@ -398,6 +399,12 @@ static const struct intel_sa_info xe2_hpd_sa_info = { /* Other values not used by simplified algorithm */ }; +static const struct intel_sa_info xe2_hpd_ecc_sa_info = { + .derating = 45, + .deprogbwlimit = 53, + /* Other values not used by simplified algorithm */ +}; + static int icl_get_bw_info(struct drm_i915_private *dev_priv, const struct intel_sa_info *sa) { struct intel_qgv_info qi = {}; @@ -740,10 +747,15 @@ static unsigned int icl_qgv_bw(struct drm_i915_private *i915, void intel_bw_init_hw(struct drm_i915_private *dev_priv) { + const struct dram_info *dram_info = &dev_priv->dram_info; + if (!HAS_DISPLAY(dev_priv)) return; - if (DISPLAY_VERx100(dev_priv) >= 1401 && IS_DGFX(dev_priv)) + if (DISPLAY_VERx100(dev_priv) >= 1401 && IS_DGFX(dev_priv) && + dram_info->type == INTEL_DRAM_GDDR_ECC) + xe2_hpd_get_bw_info(dev_priv, &xe2_hpd_ecc_sa_info); + else if (DISPLAY_VERx100(dev_priv) >= 1401 && IS_DGFX(dev_priv)) xe2_hpd_get_bw_info(dev_priv, &xe2_hpd_sa_info); else if (DISPLAY_VER(dev_priv) >= 14) tgl_get_bw_info(dev_priv, &mtl_sa_info); diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 3afb85fe8536..3b509c70fb58 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -968,7 +968,9 @@ static bool vrr_params_changed(const struct intel_crtc_state *old_crtc_state, old_crtc_state->vrr.vmin != new_crtc_state->vrr.vmin || old_crtc_state->vrr.vmax != new_crtc_state->vrr.vmax || old_crtc_state->vrr.guardband != new_crtc_state->vrr.guardband || - old_crtc_state->vrr.pipeline_full != new_crtc_state->vrr.pipeline_full; + old_crtc_state->vrr.pipeline_full != new_crtc_state->vrr.pipeline_full || + old_crtc_state->vrr.vsync_start != new_crtc_state->vrr.vsync_start || + old_crtc_state->vrr.vsync_end != new_crtc_state->vrr.vsync_end; } static bool cmrr_params_changed(const struct intel_crtc_state *old_crtc_state, diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h index 717286981687..7a3bb77c7af7 100644 --- a/drivers/gpu/drm/i915/display/intel_display_device.h +++ b/drivers/gpu/drm/i915/display/intel_display_device.h @@ -161,6 +161,7 @@ struct intel_display_platforms { #define HAS_DPT(__display) (DISPLAY_VER(__display) >= 13) #define HAS_DSB(__display) (DISPLAY_INFO(__display)->has_dsb) #define HAS_DSC(__display) (DISPLAY_RUNTIME_INFO(__display)->has_dsc) +#define HAS_DSC_3ENGINES(__display) (DISPLAY_VERx100(__display) == 1401 && HAS_DSC(__display)) #define HAS_DSC_MST(__display) (DISPLAY_VER(__display) >= 12 && HAS_DSC(__display)) #define HAS_FBC(__display) (DISPLAY_RUNTIME_INFO(__display)->fbc_mask != 0) #define HAS_FBC_DIRTY_RECT(__display) (DISPLAY_VER(__display) >= 30) diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index a236b5fc7a3d..392c3653d0d7 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -172,10 +172,28 @@ int intel_dp_link_symbol_clock(int rate) static int max_dprx_rate(struct intel_dp *intel_dp) { + struct intel_display *display = to_intel_display(intel_dp); + struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; + int max_rate; + if (intel_dp_tunnel_bw_alloc_is_enabled(intel_dp)) - return drm_dp_tunnel_max_dprx_rate(intel_dp->tunnel); + max_rate = drm_dp_tunnel_max_dprx_rate(intel_dp->tunnel); + else + max_rate = drm_dp_bw_code_to_link_rate(intel_dp->dpcd[DP_MAX_LINK_RATE]); - return drm_dp_bw_code_to_link_rate(intel_dp->dpcd[DP_MAX_LINK_RATE]); + /* + * Some broken eDP sinks illegally declare support for + * HBR3 without TPS4, and are unable to produce a stable + * output. Reject HBR3 when TPS4 is not available. + */ + if (max_rate >= 810000 && !drm_dp_tps4_supported(intel_dp->dpcd)) { + drm_dbg_kms(display->drm, + "[ENCODER:%d:%s] Rejecting HBR3 due to missing TPS4 support\n", + encoder->base.base.id, encoder->base.name); + max_rate = 540000; + } + + return max_rate; } static int max_dprx_lane_count(struct intel_dp *intel_dp) @@ -1032,10 +1050,11 @@ u8 intel_dp_dsc_get_slice_count(const struct intel_connector *connector, u8 test_slice_count = valid_dsc_slicecount[i] * num_joined_pipes; /* - * 3 DSC Slices per pipe need 3 DSC engines, - * which is supported only with Ultrajoiner. + * 3 DSC Slices per pipe need 3 DSC engines, which is supported only + * with Ultrajoiner only for some platforms. */ - if (valid_dsc_slicecount[i] == 3 && num_joined_pipes != 4) + if (valid_dsc_slicecount[i] == 3 && + (!HAS_DSC_3ENGINES(display) || num_joined_pipes != 4)) continue; if (test_slice_count > @@ -4170,6 +4189,9 @@ static void intel_edp_mso_init(struct intel_dp *intel_dp) static void intel_edp_set_sink_rates(struct intel_dp *intel_dp) { + struct intel_display *display = to_intel_display(intel_dp); + struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; + intel_dp->num_sink_rates = 0; if (intel_dp->edp_dpcd[0] >= DP_EDP_14) { @@ -4180,10 +4202,7 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp) sink_rates, sizeof(sink_rates)); for (i = 0; i < ARRAY_SIZE(sink_rates); i++) { - int val = le16_to_cpu(sink_rates[i]); - - if (val == 0) - break; + int rate; /* Value read multiplied by 200kHz gives the per-lane * link rate in kHz. The source rates are, however, @@ -4191,7 +4210,24 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp) * back to symbols is * (val * 200kHz)*(8/10 ch. encoding)*(1/8 bit to Byte) */ - intel_dp->sink_rates[i] = (val * 200) / 10; + rate = le16_to_cpu(sink_rates[i]) * 200 / 10; + + if (rate == 0) + break; + + /* + * Some broken eDP sinks illegally declare support for + * HBR3 without TPS4, and are unable to produce a stable + * output. Reject HBR3 when TPS4 is not available. + */ + if (rate >= 810000 && !drm_dp_tps4_supported(intel_dp->dpcd)) { + drm_dbg_kms(display->drm, + "[ENCODER:%d:%s] Rejecting HBR3 due to missing TPS4 support\n", + encoder->base.base.id, encoder->base.name); + break; + } + + intel_dp->sink_rates[i] = rate; } intel_dp->num_sink_rates = i; } diff --git a/drivers/gpu/drm/i915/display/intel_vblank.c b/drivers/gpu/drm/i915/display/intel_vblank.c index 4efd4f7d497a..7b240ce681a0 100644 --- a/drivers/gpu/drm/i915/display/intel_vblank.c +++ b/drivers/gpu/drm/i915/display/intel_vblank.c @@ -222,7 +222,9 @@ int intel_crtc_scanline_offset(const struct intel_crtc_state *crtc_state) * However if queried just before the start of vblank we'll get an * answer that's slightly in the future. */ - if (DISPLAY_VER(display) == 2) + if (DISPLAY_VER(display) >= 20 || display->platform.battlemage) + return 1; + else if (DISPLAY_VER(display) == 2) return -1; else if (HAS_DDI(display) && intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) return 2; diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c index 9378d5901c49..9ca42589da4d 100644 --- a/drivers/gpu/drm/i915/gt/intel_rc6.c +++ b/drivers/gpu/drm/i915/gt/intel_rc6.c @@ -117,21 +117,10 @@ static void gen11_rc6_enable(struct intel_rc6 *rc6) GEN6_RC_CTL_RC6_ENABLE | GEN6_RC_CTL_EI_MODE(1); - /* - * BSpec 52698 - Render powergating must be off. - * FIXME BSpec is outdated, disabling powergating for MTL is just - * temporary wa and should be removed after fixing real cause - * of forcewake timeouts. - */ - if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74))) - pg_enable = - GEN9_MEDIA_PG_ENABLE | - GEN11_MEDIA_SAMPLER_PG_ENABLE; - else - pg_enable = - GEN9_RENDER_PG_ENABLE | - GEN9_MEDIA_PG_ENABLE | - GEN11_MEDIA_SAMPLER_PG_ENABLE; + pg_enable = + GEN9_RENDER_PG_ENABLE | + GEN9_MEDIA_PG_ENABLE | + GEN11_MEDIA_SAMPLER_PG_ENABLE; if (GRAPHICS_VER(gt->i915) >= 12 && !IS_DG1(gt->i915)) { for (i = 0; i < I915_MAX_VCS; i++) diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c index d791f9baa11d..456d3372eef8 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c @@ -317,6 +317,11 @@ void intel_huc_init_early(struct intel_huc *huc) } } +void intel_huc_fini_late(struct intel_huc *huc) +{ + delayed_huc_load_fini(huc); +} + #define HUC_LOAD_MODE_STRING(x) (x ? "GSC" : "legacy") static int check_huc_loading_mode(struct intel_huc *huc) { @@ -414,12 +419,6 @@ out: void intel_huc_fini(struct intel_huc *huc) { - /* - * the fence is initialized in init_early, so we need to clean it up - * even if HuC loading is off. - */ - delayed_huc_load_fini(huc); - if (huc->heci_pkt) i915_vma_unpin_and_release(&huc->heci_pkt, 0); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.h b/drivers/gpu/drm/i915/gt/uc/intel_huc.h index d5e441b9e08d..921ad4b1687f 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.h @@ -55,6 +55,7 @@ struct intel_huc { int intel_huc_sanitize(struct intel_huc *huc); void intel_huc_init_early(struct intel_huc *huc); +void intel_huc_fini_late(struct intel_huc *huc); int intel_huc_init(struct intel_huc *huc); void intel_huc_fini(struct intel_huc *huc); int intel_huc_auth(struct intel_huc *huc, enum intel_huc_authentication_type type); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c index 90ba1b0b4c9d..4a3493e8d433 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c @@ -136,6 +136,7 @@ void intel_uc_init_late(struct intel_uc *uc) void intel_uc_driver_late_release(struct intel_uc *uc) { + intel_huc_fini_late(&uc->huc); } /** diff --git a/drivers/gpu/drm/i915/gvt/opregion.c b/drivers/gpu/drm/i915/gvt/opregion.c index 509f9ccae3a9..dbad4d853d3a 100644 --- a/drivers/gpu/drm/i915/gvt/opregion.c +++ b/drivers/gpu/drm/i915/gvt/opregion.c @@ -222,7 +222,6 @@ int intel_vgpu_init_opregion(struct intel_vgpu *vgpu) u8 *buf; struct opregion_header *header; struct vbt v; - const char opregion_signature[16] = OPREGION_SIGNATURE; gvt_dbg_core("init vgpu%d opregion\n", vgpu->id); vgpu_opregion(vgpu)->va = (void *)__get_free_pages(GFP_KERNEL | @@ -236,8 +235,10 @@ int intel_vgpu_init_opregion(struct intel_vgpu *vgpu) /* emulated opregion with VBT mailbox only */ buf = (u8 *)vgpu_opregion(vgpu)->va; header = (struct opregion_header *)buf; - memcpy(header->signature, opregion_signature, - sizeof(opregion_signature)); + + static_assert(sizeof(header->signature) == sizeof(OPREGION_SIGNATURE) - 1); + memcpy(header->signature, OPREGION_SIGNATURE, sizeof(header->signature)); + header->size = 0x8; header->opregion_ver = 0x02000000; header->mboxes = MBOX_VBT; diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index ffc346379cc2..54538b6f85df 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -305,6 +305,7 @@ struct drm_i915_private { INTEL_DRAM_DDR5, INTEL_DRAM_LPDDR5, INTEL_DRAM_GDDR, + INTEL_DRAM_GDDR_ECC, } type; u8 num_qgv_points; u8 num_psf_gv_points; diff --git a/drivers/gpu/drm/i915/selftests/i915_selftest.c b/drivers/gpu/drm/i915/selftests/i915_selftest.c index fee76c1d2f45..889281819c5b 100644 --- a/drivers/gpu/drm/i915/selftests/i915_selftest.c +++ b/drivers/gpu/drm/i915/selftests/i915_selftest.c @@ -23,7 +23,9 @@ #include <linux/random.h> +#include "gt/intel_gt.h" #include "gt/intel_gt_pm.h" +#include "gt/intel_gt_regs.h" #include "gt/uc/intel_gsc_fw.h" #include "i915_driver.h" @@ -253,11 +255,27 @@ int i915_mock_selftests(void) int i915_live_selftests(struct pci_dev *pdev) { struct drm_i915_private *i915 = pdev_to_i915(pdev); + struct intel_uncore *uncore = &i915->uncore; int err; + u32 pg_enable; + intel_wakeref_t wakeref; if (!i915_selftest.live) return 0; + /* + * FIXME Disable render powergating, this is temporary wa and should be removed + * after fixing real cause of forcewake timeouts. + */ + with_intel_runtime_pm(uncore->rpm, wakeref) { + if (IS_GFX_GT_IP_RANGE(to_gt(i915), IP_VER(12, 00), IP_VER(12, 74))) { + pg_enable = intel_uncore_read(uncore, GEN9_PG_ENABLE); + if (pg_enable & GEN9_RENDER_PG_ENABLE) + intel_uncore_write_fw(uncore, GEN9_PG_ENABLE, + pg_enable & ~GEN9_RENDER_PG_ENABLE); + } + } + __wait_gsc_proxy_completed(i915); __wait_gsc_huc_load_completed(i915); diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c index 9e310f4099f4..f60eedb0e92c 100644 --- a/drivers/gpu/drm/i915/soc/intel_dram.c +++ b/drivers/gpu/drm/i915/soc/intel_dram.c @@ -687,6 +687,10 @@ static int xelpdp_get_dram_info(struct drm_i915_private *i915) drm_WARN_ON(&i915->drm, !IS_DGFX(i915)); dram_info->type = INTEL_DRAM_GDDR; break; + case 9: + drm_WARN_ON(&i915->drm, !IS_DGFX(i915)); + dram_info->type = INTEL_DRAM_GDDR_ECC; + break; default: MISSING_CASE(val); return -EINVAL; diff --git a/drivers/gpu/drm/xe/instructions/xe_gpu_commands.h b/drivers/gpu/drm/xe/instructions/xe_gpu_commands.h index a255946b6f77..8cfcd3360896 100644 --- a/drivers/gpu/drm/xe/instructions/xe_gpu_commands.h +++ b/drivers/gpu/drm/xe/instructions/xe_gpu_commands.h @@ -41,6 +41,7 @@ #define GFX_OP_PIPE_CONTROL(len) ((0x3<<29)|(0x3<<27)|(0x2<<24)|((len)-2)) +#define PIPE_CONTROL0_L3_READ_ONLY_CACHE_INVALIDATE BIT(10) /* gen12 */ #define PIPE_CONTROL0_HDC_PIPELINE_FLUSH BIT(9) /* gen12 */ #define PIPE_CONTROL_COMMAND_CACHE_INVALIDATE (1<<29) diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h index 72ef0b6fc425..9f8667ebba85 100644 --- a/drivers/gpu/drm/xe/xe_device_types.h +++ b/drivers/gpu/drm/xe/xe_device_types.h @@ -585,6 +585,7 @@ struct xe_device { INTEL_DRAM_DDR5, INTEL_DRAM_LPDDR5, INTEL_DRAM_GDDR, + INTEL_DRAM_GDDR_ECC, } type; u8 num_qgv_points; u8 num_psf_gv_points; diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c index 03072e094991..084cbdeba8ea 100644 --- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c +++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c @@ -322,6 +322,13 @@ int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt) return 0; } +/* + * Ensure that roundup_pow_of_two(length) doesn't overflow. + * Note that roundup_pow_of_two() operates on unsigned long, + * not on u64. + */ +#define MAX_RANGE_TLB_INVALIDATION_LENGTH (rounddown_pow_of_two(ULONG_MAX)) + /** * xe_gt_tlb_invalidation_range - Issue a TLB invalidation on this GT for an * address range @@ -346,6 +353,7 @@ int xe_gt_tlb_invalidation_range(struct xe_gt *gt, struct xe_device *xe = gt_to_xe(gt); #define MAX_TLB_INVALIDATION_LEN 7 u32 action[MAX_TLB_INVALIDATION_LEN]; + u64 length = end - start; int len = 0; xe_gt_assert(gt, fence); @@ -358,11 +366,11 @@ int xe_gt_tlb_invalidation_range(struct xe_gt *gt, action[len++] = XE_GUC_ACTION_TLB_INVALIDATION; action[len++] = 0; /* seqno, replaced in send_tlb_invalidation */ - if (!xe->info.has_range_tlb_invalidation) { + if (!xe->info.has_range_tlb_invalidation || + length > MAX_RANGE_TLB_INVALIDATION_LENGTH) { action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL); } else { u64 orig_start = start; - u64 length = end - start; u64 align; if (length < SZ_4K) diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c index 85215313976c..43b1192ba61c 100644 --- a/drivers/gpu/drm/xe/xe_guc_pc.c +++ b/drivers/gpu/drm/xe/xe_guc_pc.c @@ -1070,6 +1070,7 @@ int xe_guc_pc_start(struct xe_guc_pc *pc) if (wait_for_pc_state(pc, SLPC_GLOBAL_STATE_RUNNING, SLPC_RESET_EXTENDED_TIMEOUT_MS)) { xe_gt_err(gt, "GuC PC Start failed: Dynamic GT frequency control and GT sleep states are now disabled.\n"); + ret = -EIO; goto out; } diff --git a/drivers/gpu/drm/xe/xe_hw_engine.c b/drivers/gpu/drm/xe/xe_hw_engine.c index 8c05fd30b7df..93241fd0a4ba 100644 --- a/drivers/gpu/drm/xe/xe_hw_engine.c +++ b/drivers/gpu/drm/xe/xe_hw_engine.c @@ -389,12 +389,6 @@ xe_hw_engine_setup_default_lrc_state(struct xe_hw_engine *hwe) blit_cctl_val, XE_RTP_ACTION_FLAG(ENGINE_BASE))) }, - /* Use Fixed slice CCS mode */ - { XE_RTP_NAME("RCU_MODE_FIXED_SLICE_CCS_MODE"), - XE_RTP_RULES(FUNC(xe_hw_engine_match_fixed_cslice_mode)), - XE_RTP_ACTIONS(FIELD_SET(RCU_MODE, RCU_MODE_FIXED_SLICE_CCS_MODE, - RCU_MODE_FIXED_SLICE_CCS_MODE)) - }, /* Disable WMTP if HW doesn't support it */ { XE_RTP_NAME("DISABLE_WMTP_ON_UNSUPPORTED_HW"), XE_RTP_RULES(FUNC(xe_rtp_cfeg_wmtp_disabled)), @@ -461,6 +455,12 @@ hw_engine_setup_default_state(struct xe_hw_engine *hwe) XE_RTP_ACTIONS(SET(CSFE_CHICKEN1(0), CS_PRIORITY_MEM_READ, XE_RTP_ACTION_FLAG(ENGINE_BASE))) }, + /* Use Fixed slice CCS mode */ + { XE_RTP_NAME("RCU_MODE_FIXED_SLICE_CCS_MODE"), + XE_RTP_RULES(FUNC(xe_hw_engine_match_fixed_cslice_mode)), + XE_RTP_ACTIONS(FIELD_SET(RCU_MODE, RCU_MODE_FIXED_SLICE_CCS_MODE, + RCU_MODE_FIXED_SLICE_CCS_MODE)) + }, }; xe_rtp_process_to_sr(&ctx, engine_entries, ARRAY_SIZE(engine_entries), &hwe->reg_sr); diff --git a/drivers/gpu/drm/xe/xe_hw_engine_class_sysfs.c b/drivers/gpu/drm/xe/xe_hw_engine_class_sysfs.c index b53e8d2accdb..a440442b4d72 100644 --- a/drivers/gpu/drm/xe/xe_hw_engine_class_sysfs.c +++ b/drivers/gpu/drm/xe/xe_hw_engine_class_sysfs.c @@ -32,14 +32,61 @@ bool xe_hw_engine_timeout_in_range(u64 timeout, u64 min, u64 max) return timeout >= min && timeout <= max; } -static void kobj_xe_hw_engine_release(struct kobject *kobj) +static void xe_hw_engine_sysfs_kobj_release(struct kobject *kobj) { kfree(kobj); } +static ssize_t xe_hw_engine_class_sysfs_attr_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct xe_device *xe = kobj_to_xe(kobj); + struct kobj_attribute *kattr; + ssize_t ret = -EIO; + + kattr = container_of(attr, struct kobj_attribute, attr); + if (kattr->show) { + xe_pm_runtime_get(xe); + ret = kattr->show(kobj, kattr, buf); + xe_pm_runtime_put(xe); + } + + return ret; +} + +static ssize_t xe_hw_engine_class_sysfs_attr_store(struct kobject *kobj, + struct attribute *attr, + const char *buf, + size_t count) +{ + struct xe_device *xe = kobj_to_xe(kobj); + struct kobj_attribute *kattr; + ssize_t ret = -EIO; + + kattr = container_of(attr, struct kobj_attribute, attr); + if (kattr->store) { + xe_pm_runtime_get(xe); + ret = kattr->store(kobj, kattr, buf, count); + xe_pm_runtime_put(xe); + } + + return ret; +} + +static const struct sysfs_ops xe_hw_engine_class_sysfs_ops = { + .show = xe_hw_engine_class_sysfs_attr_show, + .store = xe_hw_engine_class_sysfs_attr_store, +}; + static const struct kobj_type kobj_xe_hw_engine_type = { - .release = kobj_xe_hw_engine_release, - .sysfs_ops = &kobj_sysfs_ops + .release = xe_hw_engine_sysfs_kobj_release, + .sysfs_ops = &xe_hw_engine_class_sysfs_ops, +}; + +static const struct kobj_type kobj_xe_hw_engine_type_def = { + .release = xe_hw_engine_sysfs_kobj_release, + .sysfs_ops = &kobj_sysfs_ops, }; static ssize_t job_timeout_max_store(struct kobject *kobj, @@ -543,7 +590,7 @@ static int xe_add_hw_engine_class_defaults(struct xe_device *xe, if (!kobj) return -ENOMEM; - kobject_init(kobj, &kobj_xe_hw_engine_type); + kobject_init(kobj, &kobj_xe_hw_engine_type_def); err = kobject_add(kobj, parent, "%s", ".defaults"); if (err) goto err_object; @@ -559,57 +606,6 @@ err_object: return err; } -static void xe_hw_engine_sysfs_kobj_release(struct kobject *kobj) -{ - kfree(kobj); -} - -static ssize_t xe_hw_engine_class_sysfs_attr_show(struct kobject *kobj, - struct attribute *attr, - char *buf) -{ - struct xe_device *xe = kobj_to_xe(kobj); - struct kobj_attribute *kattr; - ssize_t ret = -EIO; - - kattr = container_of(attr, struct kobj_attribute, attr); - if (kattr->show) { - xe_pm_runtime_get(xe); - ret = kattr->show(kobj, kattr, buf); - xe_pm_runtime_put(xe); - } - - return ret; -} - -static ssize_t xe_hw_engine_class_sysfs_attr_store(struct kobject *kobj, - struct attribute *attr, - const char *buf, - size_t count) -{ - struct xe_device *xe = kobj_to_xe(kobj); - struct kobj_attribute *kattr; - ssize_t ret = -EIO; - - kattr = container_of(attr, struct kobj_attribute, attr); - if (kattr->store) { - xe_pm_runtime_get(xe); - ret = kattr->store(kobj, kattr, buf, count); - xe_pm_runtime_put(xe); - } - - return ret; -} - -static const struct sysfs_ops xe_hw_engine_class_sysfs_ops = { - .show = xe_hw_engine_class_sysfs_attr_show, - .store = xe_hw_engine_class_sysfs_attr_store, -}; - -static const struct kobj_type xe_hw_engine_sysfs_kobj_type = { - .release = xe_hw_engine_sysfs_kobj_release, - .sysfs_ops = &xe_hw_engine_class_sysfs_ops, -}; static void hw_engine_class_sysfs_fini(void *arg) { @@ -640,7 +636,7 @@ int xe_hw_engine_class_sysfs_init(struct xe_gt *gt) if (!kobj) return -ENOMEM; - kobject_init(kobj, &xe_hw_engine_sysfs_kobj_type); + kobject_init(kobj, &kobj_xe_hw_engine_type); err = kobject_add(kobj, gt->sysfs, "engines"); if (err) diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c index df4282c71bf0..5a3e89022c38 100644 --- a/drivers/gpu/drm/xe/xe_migrate.c +++ b/drivers/gpu/drm/xe/xe_migrate.c @@ -1177,7 +1177,7 @@ err: err_sync: /* Sync partial copies if any. FIXME: job_mutex? */ if (fence) { - dma_fence_wait(m->fence, false); + dma_fence_wait(fence, false); dma_fence_put(fence); } @@ -1547,7 +1547,7 @@ void xe_migrate_wait(struct xe_migrate *m) static u32 pte_update_cmd_size(u64 size) { u32 num_dword; - u64 entries = DIV_ROUND_UP(size, XE_PAGE_SIZE); + u64 entries = DIV_U64_ROUND_UP(size, XE_PAGE_SIZE); XE_WARN_ON(size > MAX_PREEMPTDISABLE_TRANSFER); /* @@ -1558,7 +1558,7 @@ static u32 pte_update_cmd_size(u64 size) * 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_ROUND_UP(entries, 0x1ff); + num_dword = (1 + 2) * DIV_U64_ROUND_UP(entries, 0x1ff); num_dword += entries * 2; return num_dword; diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c index 917fc16de866..a7582b097ae6 100644 --- a/drivers/gpu/drm/xe/xe_ring_ops.c +++ b/drivers/gpu/drm/xe/xe_ring_ops.c @@ -137,7 +137,8 @@ emit_pipe_control(u32 *dw, int i, u32 bit_group_0, u32 bit_group_1, u32 offset, static int emit_pipe_invalidate(u32 mask_flags, bool invalidate_tlb, u32 *dw, int i) { - u32 flags = PIPE_CONTROL_CS_STALL | + u32 flags0 = 0; + u32 flags1 = PIPE_CONTROL_CS_STALL | PIPE_CONTROL_COMMAND_CACHE_INVALIDATE | PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE | PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE | @@ -148,11 +149,15 @@ static int emit_pipe_invalidate(u32 mask_flags, bool invalidate_tlb, u32 *dw, PIPE_CONTROL_STORE_DATA_INDEX; if (invalidate_tlb) - flags |= PIPE_CONTROL_TLB_INVALIDATE; + flags1 |= PIPE_CONTROL_TLB_INVALIDATE; - flags &= ~mask_flags; + flags1 &= ~mask_flags; - return emit_pipe_control(dw, i, 0, flags, LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR, 0); + if (flags1 & PIPE_CONTROL_VF_CACHE_INVALIDATE) + flags0 |= PIPE_CONTROL0_L3_READ_ONLY_CACHE_INVALIDATE; + + return emit_pipe_control(dw, i, flags0, flags1, + LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR, 0); } static int emit_store_imm_ppgtt_posted(u64 addr, u64 value, diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index 3e829c87d7b4..f8c128524d9f 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -696,11 +696,14 @@ retry: 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); - xe_bo_unlock(bo); if (err) - xe_bo_put(bo); /* Creation ref */ + xe_svm_devmem_release(&bo->devmem_allocation); + + xe_bo_unlock(bo); + xe_bo_put(bo); unlock: mmap_read_unlock(mm); diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules index 0c738af24f7c..9b9e176992a8 100644 --- a/drivers/gpu/drm/xe/xe_wa_oob.rules +++ b/drivers/gpu/drm/xe/xe_wa_oob.rules @@ -32,8 +32,10 @@ GRAPHICS_VERSION(3001) 14022293748 GRAPHICS_VERSION(2001) GRAPHICS_VERSION(2004) + GRAPHICS_VERSION_RANGE(3000, 3001) 22019794406 GRAPHICS_VERSION(2001) GRAPHICS_VERSION(2004) + GRAPHICS_VERSION_RANGE(3000, 3001) 22019338487 MEDIA_VERSION(2000) GRAPHICS_VERSION(2001) MEDIA_VERSION(3000), MEDIA_STEP(A0, B0), FUNC(xe_rtp_match_not_sriov_vf) |