diff options
Diffstat (limited to 'drivers/gpu')
125 files changed, 1877 insertions, 858 deletions
diff --git a/drivers/gpu/drm/amd/acp/Kconfig b/drivers/gpu/drm/amd/acp/Kconfig index d968c2471412..0d12ebf66174 100644 --- a/drivers/gpu/drm/amd/acp/Kconfig +++ b/drivers/gpu/drm/amd/acp/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: MIT menu "ACP (Audio CoProcessor) Configuration" config DRM_AMD_ACP diff --git a/drivers/gpu/drm/amd/amdgpu/Kconfig b/drivers/gpu/drm/amd/amdgpu/Kconfig index 2e98c016cb47..9375e7f12420 100644 --- a/drivers/gpu/drm/amd/amdgpu/Kconfig +++ b/drivers/gpu/drm/amd/amdgpu/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: MIT config DRM_AMDGPU_SI bool "Enable amdgpu support for SI parts" depends on DRM_AMDGPU diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c index a97fb759e2f4..3e35a8f2c5e5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c @@ -613,7 +613,17 @@ static bool amdgpu_atpx_detect(void) bool d3_supported = false; struct pci_dev *parent_pdev; - while ((pdev = pci_get_class(PCI_BASE_CLASS_DISPLAY << 16, pdev)) != NULL) { + while ((pdev = pci_get_class(PCI_CLASS_DISPLAY_VGA << 8, pdev)) != NULL) { + vga_count++; + + has_atpx |= (amdgpu_atpx_pci_probe_handle(pdev) == true); + + parent_pdev = pci_upstream_bridge(pdev); + d3_supported |= parent_pdev && parent_pdev->bridge_d3; + amdgpu_atpx_get_quirks(pdev); + } + + while ((pdev = pci_get_class(PCI_CLASS_DISPLAY_OTHER << 8, pdev)) != NULL) { vga_count++; has_atpx |= (amdgpu_atpx_pci_probe_handle(pdev) == true); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 2cdaf3b2a721..6614d8a6f4c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -604,11 +604,8 @@ void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr) continue; } - for (i = 0; i < num_entities; i++) { - mutex_lock(&ctx->adev->lock_reset); + for (i = 0; i < num_entities; i++) drm_sched_entity_fini(&ctx->entities[0][i].entity); - mutex_unlock(&ctx->adev->lock_reset); - } } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 0ffc9447b573..30a1e3ac21d6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -142,7 +142,7 @@ int amdgpu_async_gfx_ring = 1; int amdgpu_mcbp = 0; int amdgpu_discovery = -1; int amdgpu_mes = 0; -int amdgpu_noretry = 1; +int amdgpu_noretry; int amdgpu_force_asic_type = -1; struct amdgpu_mgpu_info mgpu_info = { @@ -588,7 +588,7 @@ MODULE_PARM_DESC(mes, module_param_named(mes, amdgpu_mes, int, 0444); MODULE_PARM_DESC(noretry, - "Disable retry faults (0 = retry enabled, 1 = retry disabled (default))"); + "Disable retry faults (0 = retry enabled (default), 1 = retry disabled)"); module_param_named(noretry, amdgpu_noretry, int, 0644); /** @@ -1004,7 +1004,7 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x734F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI14}, /* Renoir */ - {0x1002, 0x1636, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RENOIR|AMD_IS_APU|AMD_EXP_HW_SUPPORT}, + {0x1002, 0x1636, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RENOIR|AMD_IS_APU}, /* Navi12 */ {0x1002, 0x7360, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI12|AMD_EXP_HW_SUPPORT}, @@ -1359,7 +1359,8 @@ static struct drm_driver kms_driver = { .driver_features = DRIVER_USE_AGP | DRIVER_ATOMIC | DRIVER_GEM | - DRIVER_RENDER | DRIVER_MODESET | DRIVER_SYNCOBJ, + DRIVER_RENDER | DRIVER_MODESET | DRIVER_SYNCOBJ | + DRIVER_SYNCOBJ_TIMELINE, .load = amdgpu_driver_load_kms, .open = amdgpu_driver_open_kms, .postclose = amdgpu_driver_postclose_kms, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 44be3a45b25e..e1b8d8daeafc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -1488,7 +1488,7 @@ out: /* Start rlc autoload after psp recieved all the gfx firmware */ if (psp->autoload_supported && ucode->ucode_id == (amdgpu_sriov_vf(adev) ? - AMDGPU_UCODE_ID_CP_MEC2 : AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM)) { + AMDGPU_UCODE_ID_CP_MEC2 : AMDGPU_UCODE_ID_RLC_G)) { ret = psp_rlc_autoload(psp); if (ret) { DRM_ERROR("Failed to start rlc autoload\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h index 410587b950f3..914acecda5cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h @@ -292,10 +292,10 @@ enum AMDGPU_UCODE_ID { AMDGPU_UCODE_ID_CP_MEC2_JT, AMDGPU_UCODE_ID_CP_MES, AMDGPU_UCODE_ID_CP_MES_DATA, - AMDGPU_UCODE_ID_RLC_G, AMDGPU_UCODE_ID_RLC_RESTORE_LIST_CNTL, AMDGPU_UCODE_ID_RLC_RESTORE_LIST_GPM_MEM, AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM, + AMDGPU_UCODE_ID_RLC_G, AMDGPU_UCODE_ID_STORAGE, AMDGPU_UCODE_ID_SMC, AMDGPU_UCODE_ID_UVD, diff --git a/drivers/gpu/drm/amd/amdgpu/df_v3_6.c b/drivers/gpu/drm/amd/amdgpu/df_v3_6.c index 16fbd2bc8ad1..4043ebcea5de 100644 --- a/drivers/gpu/drm/amd/amdgpu/df_v3_6.c +++ b/drivers/gpu/drm/amd/amdgpu/df_v3_6.c @@ -268,23 +268,29 @@ static void df_v3_6_update_medium_grain_clock_gating(struct amdgpu_device *adev, { u32 tmp; - /* Put DF on broadcast mode */ - adev->df_funcs->enable_broadcast_mode(adev, true); - - if (enable && (adev->cg_flags & AMD_CG_SUPPORT_DF_MGCG)) { - tmp = RREG32_SOC15(DF, 0, mmDF_PIE_AON0_DfGlobalClkGater); - tmp &= ~DF_PIE_AON0_DfGlobalClkGater__MGCGMode_MASK; - tmp |= DF_V3_6_MGCG_ENABLE_15_CYCLE_DELAY; - WREG32_SOC15(DF, 0, mmDF_PIE_AON0_DfGlobalClkGater, tmp); - } else { - tmp = RREG32_SOC15(DF, 0, mmDF_PIE_AON0_DfGlobalClkGater); - tmp &= ~DF_PIE_AON0_DfGlobalClkGater__MGCGMode_MASK; - tmp |= DF_V3_6_MGCG_DISABLE; - WREG32_SOC15(DF, 0, mmDF_PIE_AON0_DfGlobalClkGater, tmp); - } + if (adev->cg_flags & AMD_CG_SUPPORT_DF_MGCG) { + /* Put DF on broadcast mode */ + adev->df_funcs->enable_broadcast_mode(adev, true); + + if (enable) { + tmp = RREG32_SOC15(DF, 0, + mmDF_PIE_AON0_DfGlobalClkGater); + tmp &= ~DF_PIE_AON0_DfGlobalClkGater__MGCGMode_MASK; + tmp |= DF_V3_6_MGCG_ENABLE_15_CYCLE_DELAY; + WREG32_SOC15(DF, 0, + mmDF_PIE_AON0_DfGlobalClkGater, tmp); + } else { + tmp = RREG32_SOC15(DF, 0, + mmDF_PIE_AON0_DfGlobalClkGater); + tmp &= ~DF_PIE_AON0_DfGlobalClkGater__MGCGMode_MASK; + tmp |= DF_V3_6_MGCG_DISABLE; + WREG32_SOC15(DF, 0, + mmDF_PIE_AON0_DfGlobalClkGater, tmp); + } - /* Exit broadcast mode */ - adev->df_funcs->enable_broadcast_mode(adev, false); + /* Exit broadcast mode */ + adev->df_funcs->enable_broadcast_mode(adev, false); + } } static void df_v3_6_get_clockgating_state(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index f2c1b026397b..ba9e53a1abc3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -117,10 +117,13 @@ static const struct soc15_reg_golden golden_settings_gc_10_1[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CGTT_SCLK_CTRL, 0x10000000, 0x10000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CTRL2, 0xffffffff, 0x1402002f), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CTRL3, 0xffff9fff, 0x00001188), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_TIMEOUT_COUNTER, 0xffffffff, 0x00000800), SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE, 0x3fffffff, 0x08000009), SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_1, 0x00400000, 0x04440000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_2, 0x00000800, 0x00000820), SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmRMI_SPARE, 0xffffffff, 0xffff3101), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_CONFIG_CNTL, 0x001f0000, 0x00070104), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_ALU_CLK_CTRL, 0xffffffff, 0xffffffff), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_ARB_CONFIG, 0x00000100, 0x00000130), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_LDS_CLK_CTRL, 0xffffffff, 0xffffffff), @@ -162,10 +165,13 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_1[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CGTT_SCLK_CTRL, 0xffff0fff, 0x10000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CTRL2, 0xffffffff, 0x1402002f), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGL2C_CTRL3, 0xffffbfff, 0x00000188), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_BINNER_TIMEOUT_COUNTER, 0xffffffff, 0x00000800), SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE, 0x3fffffff, 0x08000009), SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_1, 0x00400000, 0x04440000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_ENHANCE_2, 0x00000800, 0x00000820), SOC15_REG_GOLDEN_VALUE(GC, 0, mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmRMI_SPARE, 0xffffffff, 0xffff3101), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmSPI_CONFIG_CNTL, 0x001f0000, 0x00070105), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_ALU_CLK_CTRL, 0xffffffff, 0xffffffff), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_ARB_CONFIG, 0x00000133, 0x00000130), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_LDS_CLK_CTRL, 0xffffffff, 0xffffffff), diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 983db77999e7..52a647d7022d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -6146,7 +6146,23 @@ static void gfx_v8_0_ring_emit_fence_gfx(struct amdgpu_ring *ring, u64 addr, bool write64bit = flags & AMDGPU_FENCE_FLAG_64BIT; bool int_sel = flags & AMDGPU_FENCE_FLAG_INT; - /* EVENT_WRITE_EOP - flush caches, send int */ + /* Workaround for cache flush problems. First send a dummy EOP + * event down the pipe with seq one below. + */ + amdgpu_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4)); + amdgpu_ring_write(ring, (EOP_TCL1_ACTION_EN | + EOP_TC_ACTION_EN | + EOP_TC_WB_ACTION_EN | + EVENT_TYPE(CACHE_FLUSH_AND_INV_TS_EVENT) | + EVENT_INDEX(5))); + amdgpu_ring_write(ring, addr & 0xfffffffc); + amdgpu_ring_write(ring, (upper_32_bits(addr) & 0xffff) | + DATA_SEL(1) | INT_SEL(0)); + amdgpu_ring_write(ring, lower_32_bits(seq - 1)); + amdgpu_ring_write(ring, upper_32_bits(seq - 1)); + + /* Then send the real EOP event down the pipe: + * EVENT_WRITE_EOP - flush caches, send int */ amdgpu_ring_write(ring, PACKET3(PACKET3_EVENT_WRITE_EOP, 4)); amdgpu_ring_write(ring, (EOP_TCL1_ACTION_EN | EOP_TC_ACTION_EN | @@ -6888,7 +6904,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = { 5 + /* COND_EXEC */ 7 + /* PIPELINE_SYNC */ VI_FLUSH_GPU_TLB_NUM_WREG * 5 + 9 + /* VM_FLUSH */ - 8 + /* FENCE for VM_FLUSH */ + 12 + /* FENCE for VM_FLUSH */ 20 + /* GDS switch */ 4 + /* double SWITCH_BUFFER, the first COND_EXEC jump to the place just @@ -6900,7 +6916,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = { 31 + /* DE_META */ 3 + /* CNTX_CTRL */ 5 + /* HDP_INVL */ - 8 + 8 + /* FENCE x2 */ + 12 + 12 + /* FENCE x2 */ 2, /* SWITCH_BUFFER */ .emit_ib_size = 4, /* gfx_v8_0_ring_emit_ib_gfx */ .emit_ib = gfx_v8_0_ring_emit_ib_gfx, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 66328ffa395a..97105a5bb246 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -1052,17 +1052,10 @@ static void gfx_v9_0_check_if_need_gfxoff(struct amdgpu_device *adev) case CHIP_VEGA20: break; case CHIP_RAVEN: - /* Disable GFXOFF on original raven. There are combinations - * of sbios and platforms that are not stable. - */ - if (!(adev->rev_id >= 0x8 || adev->pdev->device == 0x15d8)) - adev->pm.pp_feature &= ~PP_GFXOFF_MASK; - else if (!(adev->rev_id >= 0x8 || adev->pdev->device == 0x15d8) - &&((adev->gfx.rlc_fw_version != 106 && - adev->gfx.rlc_fw_version < 531) || - (adev->gfx.rlc_fw_version == 53815) || - (adev->gfx.rlc_feature_version < 1) || - !adev->gfx.rlc.is_rlc_v2_1)) + if (!(adev->rev_id >= 0x8 || + adev->pdev->device == 0x15d8) && + (adev->pm.fw_version < 0x41e2b || /* not raven1 fresh */ + !adev->gfx.rlc.is_rlc_v2_1)) /* without rlc save restore ucodes */ adev->pm.pp_feature &= ~PP_GFXOFF_MASK; if (adev->pm.pp_feature & PP_GFXOFF_MASK) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 232469507446..f5725336a5f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -219,6 +219,21 @@ static uint32_t gmc_v10_0_get_invalidate_req(unsigned int vmid, return req; } +/** + * gmc_v10_0_use_invalidate_semaphore - judge whether to use semaphore + * + * @adev: amdgpu_device pointer + * @vmhub: vmhub type + * + */ +static bool gmc_v10_0_use_invalidate_semaphore(struct amdgpu_device *adev, + uint32_t vmhub) +{ + return ((vmhub == AMDGPU_MMHUB_0 || + vmhub == AMDGPU_MMHUB_1) && + (!amdgpu_sriov_vf(adev))); +} + /* * GART * VMID 0 is the physical GPU addresses as used by the kernel. @@ -229,6 +244,7 @@ static uint32_t gmc_v10_0_get_invalidate_req(unsigned int vmid, static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, unsigned int vmhub, uint32_t flush_type) { + bool use_semaphore = gmc_v10_0_use_invalidate_semaphore(adev, vmhub); struct amdgpu_vmhub *hub = &adev->vmhub[vmhub]; u32 tmp = gmc_v10_0_get_invalidate_req(vmid, flush_type); /* Use register 17 for GART */ @@ -244,8 +260,7 @@ static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, */ /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ - if (vmhub == AMDGPU_MMHUB_0 || - vmhub == AMDGPU_MMHUB_1) { + if (use_semaphore) { for (i = 0; i < adev->usec_timeout; i++) { /* a read return value of 1 means semaphore acuqire */ tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_sem + eng); @@ -278,8 +293,7 @@ static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, } /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ - if (vmhub == AMDGPU_MMHUB_0 || - vmhub == AMDGPU_MMHUB_1) + if (use_semaphore) /* * add semaphore release after invalidation, * write with 0 means semaphore release @@ -369,6 +383,7 @@ error_alloc: static uint64_t gmc_v10_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, unsigned vmid, uint64_t pd_addr) { + bool use_semaphore = gmc_v10_0_use_invalidate_semaphore(ring->adev, ring->funcs->vmhub); struct amdgpu_vmhub *hub = &ring->adev->vmhub[ring->funcs->vmhub]; uint32_t req = gmc_v10_0_get_invalidate_req(vmid, 0); unsigned eng = ring->vm_inv_eng; @@ -381,8 +396,7 @@ static uint64_t gmc_v10_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, */ /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ - if (ring->funcs->vmhub == AMDGPU_MMHUB_0 || - ring->funcs->vmhub == AMDGPU_MMHUB_1) + if (use_semaphore) /* a read return value of 1 means semaphore acuqire */ amdgpu_ring_emit_reg_wait(ring, hub->vm_inv_eng0_sem + eng, 0x1, 0x1); @@ -398,8 +412,7 @@ static uint64_t gmc_v10_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, req, 1 << vmid); /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ - if (ring->funcs->vmhub == AMDGPU_MMHUB_0 || - ring->funcs->vmhub == AMDGPU_MMHUB_1) + if (use_semaphore) /* * add semaphore release after invalidation, * write with 0 means semaphore release diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 3c355fb5d2b4..a5b68b5e452f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -416,6 +416,24 @@ static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid, return req; } +/** + * gmc_v9_0_use_invalidate_semaphore - judge whether to use semaphore + * + * @adev: amdgpu_device pointer + * @vmhub: vmhub type + * + */ +static bool gmc_v9_0_use_invalidate_semaphore(struct amdgpu_device *adev, + uint32_t vmhub) +{ + return ((vmhub == AMDGPU_MMHUB_0 || + vmhub == AMDGPU_MMHUB_1) && + (!amdgpu_sriov_vf(adev)) && + (!(adev->asic_type == CHIP_RAVEN && + adev->rev_id < 0x8 && + adev->pdev->device == 0x15d8))); +} + /* * GART * VMID 0 is the physical GPU addresses as used by the kernel. @@ -435,6 +453,7 @@ static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid, static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, uint32_t vmhub, uint32_t flush_type) { + bool use_semaphore = gmc_v9_0_use_invalidate_semaphore(adev, vmhub); const unsigned eng = 17; u32 j, tmp; struct amdgpu_vmhub *hub; @@ -468,8 +487,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, */ /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ - if (vmhub == AMDGPU_MMHUB_0 || - vmhub == AMDGPU_MMHUB_1) { + if (use_semaphore) { for (j = 0; j < adev->usec_timeout; j++) { /* a read return value of 1 means semaphore acuqire */ tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_sem + eng); @@ -499,8 +517,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, } /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ - if (vmhub == AMDGPU_MMHUB_0 || - vmhub == AMDGPU_MMHUB_1) + if (use_semaphore) /* * add semaphore release after invalidation, * write with 0 means semaphore release @@ -518,6 +535,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, static uint64_t gmc_v9_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, unsigned vmid, uint64_t pd_addr) { + bool use_semaphore = gmc_v9_0_use_invalidate_semaphore(ring->adev, ring->funcs->vmhub); struct amdgpu_device *adev = ring->adev; struct amdgpu_vmhub *hub = &adev->vmhub[ring->funcs->vmhub]; uint32_t req = gmc_v9_0_get_invalidate_req(vmid, 0); @@ -531,8 +549,7 @@ static uint64_t gmc_v9_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, */ /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ - if (ring->funcs->vmhub == AMDGPU_MMHUB_0 || - ring->funcs->vmhub == AMDGPU_MMHUB_1) + if (use_semaphore) /* a read return value of 1 means semaphore acuqire */ amdgpu_ring_emit_reg_wait(ring, hub->vm_inv_eng0_sem + eng, 0x1, 0x1); @@ -548,8 +565,7 @@ static uint64_t gmc_v9_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, req, 1 << vmid); /* TODO: It needs to continue working on debugging with semaphore for GFXHUB as well. */ - if (ring->funcs->vmhub == AMDGPU_MMHUB_0 || - ring->funcs->vmhub == AMDGPU_MMHUB_1) + if (use_semaphore) /* * add semaphore release after invalidation, * write with 0 means semaphore release diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index 4ef4d31f5231..2f52b7f4d25c 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -254,7 +254,7 @@ static const struct soc15_reg_golden golden_settings_sdma_4_3[] = { SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC0_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_RLC1_RB_WPTR_POLL_CNTL, 0xfffffff7, 0x00403000), SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_PAGE, 0x000003ff, 0x000003c0), - SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_WATERMK, 0xfc000000, 0x00000000) + SOC15_REG_GOLDEN_VALUE(SDMA0, 0, mmSDMA0_UTCL1_WATERMK, 0xfc000000, 0x03fbe1fe) }; static u32 sdma_v4_0_get_reg_offset(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdkfd/Kconfig b/drivers/gpu/drm/amd/amdkfd/Kconfig index ba0e68057a89..b3672d10ea54 100644 --- a/drivers/gpu/drm/amd/amdkfd/Kconfig +++ b/drivers/gpu/drm/amd/amdkfd/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: MIT # # Heterogenous system architecture configuration # diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig index 313183b80032..ae161fe86ebb 100644 --- a/drivers/gpu/drm/amd/display/Kconfig +++ b/drivers/gpu/drm/amd/display/Kconfig @@ -1,4 +1,4 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: MIT menu "Display Engine Configuration" depends on DRM && DRM_AMDGPU 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 7aac9568d3be..803e59d97411 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -3356,27 +3356,21 @@ get_output_color_space(const struct dc_crtc_timing *dc_crtc_timing) return color_space; } -static void reduce_mode_colour_depth(struct dc_crtc_timing *timing_out) -{ - if (timing_out->display_color_depth <= COLOR_DEPTH_888) - return; - - timing_out->display_color_depth--; -} - -static void adjust_colour_depth_from_display_info(struct dc_crtc_timing *timing_out, - const struct drm_display_info *info) +static bool adjust_colour_depth_from_display_info( + struct dc_crtc_timing *timing_out, + const struct drm_display_info *info) { + enum dc_color_depth depth = timing_out->display_color_depth; int normalized_clk; - if (timing_out->display_color_depth <= COLOR_DEPTH_888) - return; do { normalized_clk = timing_out->pix_clk_100hz / 10; /* YCbCr 4:2:0 requires additional adjustment of 1/2 */ if (timing_out->pixel_encoding == PIXEL_ENCODING_YCBCR420) normalized_clk /= 2; /* Adjusting pix clock following on HDMI spec based on colour depth */ - switch (timing_out->display_color_depth) { + switch (depth) { + case COLOR_DEPTH_888: + break; case COLOR_DEPTH_101010: normalized_clk = (normalized_clk * 30) / 24; break; @@ -3387,14 +3381,15 @@ static void adjust_colour_depth_from_display_info(struct dc_crtc_timing *timing_ normalized_clk = (normalized_clk * 48) / 24; break; default: - return; + /* The above depths are the only ones valid for HDMI. */ + return false; } - if (normalized_clk <= info->max_tmds_clock) - return; - reduce_mode_colour_depth(timing_out); - - } while (timing_out->display_color_depth > COLOR_DEPTH_888); - + if (normalized_clk <= info->max_tmds_clock) { + timing_out->display_color_depth = depth; + return true; + } + } while (--depth > COLOR_DEPTH_666); + return false; } static void fill_stream_properties_from_drm_display_mode( @@ -3474,8 +3469,14 @@ static void fill_stream_properties_from_drm_display_mode( stream->out_transfer_func->type = TF_TYPE_PREDEFINED; stream->out_transfer_func->tf = TRANSFER_FUNCTION_SRGB; - if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) - adjust_colour_depth_from_display_info(timing_out, info); + if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) { + if (!adjust_colour_depth_from_display_info(timing_out, info) && + drm_mode_is_420_also(info, mode_in) && + timing_out->pixel_encoding != PIXEL_ENCODING_YCBCR420) { + timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR420; + adjust_colour_depth_from_display_info(timing_out, info); + } + } } static void fill_audio_info(struct audio_info *audio_info, diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index 7873abea4112..5c3fcaa47410 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -1625,6 +1625,7 @@ static enum bp_result construct_integrated_info( /* Don't need to check major revision as they are all 1 */ switch (revision.minor) { case 11: + case 12: result = get_integrated_info_v11(bp, info); break; default: diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c index 790a2d211bd6..35c55e54eac0 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c @@ -471,12 +471,28 @@ static void rn_notify_wm_ranges(struct clk_mgr *clk_mgr_base) } +static bool rn_are_clock_states_equal(struct dc_clocks *a, + struct dc_clocks *b) +{ + if (a->dispclk_khz != b->dispclk_khz) + return false; + else if (a->dppclk_khz != b->dppclk_khz) + return false; + else if (a->dcfclk_khz != b->dcfclk_khz) + return false; + else if (a->dcfclk_deep_sleep_khz != b->dcfclk_deep_sleep_khz) + return false; + + return true; +} + + static struct clk_mgr_funcs dcn21_funcs = { .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, .update_clocks = rn_update_clocks, .init_clocks = rn_init_clocks, .enable_pme_wa = rn_enable_pme_wa, - /* .dump_clk_registers = rn_dump_clk_registers, */ + .are_clock_states_equal = rn_are_clock_states_equal, .notify_wm_ranges = rn_notify_wm_ranges }; @@ -518,36 +534,83 @@ struct clk_bw_params rn_bw_params = { .num_entries = 4, }, - .wm_table = { - .entries = { - { - .wm_inst = WM_A, - .wm_type = WM_TYPE_PSTATE_CHG, - .pstate_latency_us = 23.84, - .valid = true, - }, - { - .wm_inst = WM_B, - .wm_type = WM_TYPE_PSTATE_CHG, - .pstate_latency_us = 23.84, - .valid = true, - }, - { - .wm_inst = WM_C, - .wm_type = WM_TYPE_PSTATE_CHG, - .pstate_latency_us = 23.84, - .valid = true, - }, - { - .wm_inst = WM_D, - .wm_type = WM_TYPE_PSTATE_CHG, - .pstate_latency_us = 23.84, - .valid = true, - }, +}; + +struct wm_table ddr4_wm_table = { + .entries = { + { + .wm_inst = WM_A, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.72, + .sr_exit_time_us = 6.09, + .sr_enter_plus_exit_time_us = 7.14, + .valid = true, + }, + { + .wm_inst = WM_B, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.72, + .sr_exit_time_us = 10.12, + .sr_enter_plus_exit_time_us = 11.48, + .valid = true, + }, + { + .wm_inst = WM_C, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.72, + .sr_exit_time_us = 10.12, + .sr_enter_plus_exit_time_us = 11.48, + .valid = true, + }, + { + .wm_inst = WM_D, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 11.72, + .sr_exit_time_us = 10.12, + .sr_enter_plus_exit_time_us = 11.48, + .valid = true, }, } }; +struct wm_table lpddr4_wm_table = { + .entries = { + { + .wm_inst = WM_A, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 23.84, + .sr_exit_time_us = 12.5, + .sr_enter_plus_exit_time_us = 17.0, + .valid = true, + }, + { + .wm_inst = WM_B, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 23.84, + .sr_exit_time_us = 12.5, + .sr_enter_plus_exit_time_us = 17.0, + .valid = true, + }, + { + .wm_inst = WM_C, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 23.84, + .sr_exit_time_us = 12.5, + .sr_enter_plus_exit_time_us = 17.0, + .valid = true, + }, + { + .wm_inst = WM_D, + .wm_type = WM_TYPE_PSTATE_CHG, + .pstate_latency_us = 23.84, + .sr_exit_time_us = 12.5, + .sr_enter_plus_exit_time_us = 17.0, + .valid = true, + }, + } +}; + + static unsigned int find_dcfclk_for_voltage(struct dpm_clocks *clock_table, unsigned int voltage) { int i; @@ -561,7 +624,7 @@ static unsigned int find_dcfclk_for_voltage(struct dpm_clocks *clock_table, unsi return 0; } -static void rn_clk_mgr_helper_populate_bw_params(struct clk_bw_params *bw_params, struct dpm_clocks *clock_table, struct hw_asic_id *asic_id) +static void rn_clk_mgr_helper_populate_bw_params(struct clk_bw_params *bw_params, struct dpm_clocks *clock_table, struct integrated_info *bios_info) { int i, j = 0; @@ -593,8 +656,8 @@ static void rn_clk_mgr_helper_populate_bw_params(struct clk_bw_params *bw_params bw_params->clk_table.entries[i].dcfclk_mhz = find_dcfclk_for_voltage(clock_table, clock_table->FClocks[j].Vol); } - bw_params->vram_type = asic_id->vram_type; - bw_params->num_channels = asic_id->vram_width / DDR4_DRAM_WIDTH; + bw_params->vram_type = bios_info->memory_type; + bw_params->num_channels = bios_info->ma_channel_number; for (i = 0; i < WM_SET_COUNT; i++) { bw_params->wm_table.entries[i].wm_inst = i; @@ -669,15 +732,24 @@ void rn_clk_mgr_construct( ASSERT(clk_mgr->base.dprefclk_khz == 600000); clk_mgr->base.dprefclk_khz = 600000; } + + if (ctx->dc_bios->integrated_info->memory_type == LpDdr4MemType) { + rn_bw_params.wm_table = lpddr4_wm_table; + } else { + rn_bw_params.wm_table = ddr4_wm_table; + } } dce_clock_read_ss_info(clk_mgr); + clk_mgr->base.bw_params = &rn_bw_params; if (pp_smu && pp_smu->rn_funcs.get_dpm_clock_table) { pp_smu->rn_funcs.get_dpm_clock_table(&pp_smu->rn_funcs.pp_smu, &clock_table); - rn_clk_mgr_helper_populate_bw_params(clk_mgr->base.bw_params, &clock_table, &ctx->asic_id); + if (ctx->dc_bios && ctx->dc_bios->integrated_info) { + rn_clk_mgr_helper_populate_bw_params (clk_mgr->base.bw_params, &clock_table, ctx->dc_bios->integrated_info); + } } if (!IS_FPGA_MAXIMUS_DC(ctx->dce_environment) && clk_mgr->smu_ver >= 0x00371500) { diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index 12ba6fdf89b7..4619f94f0ac7 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -372,7 +372,7 @@ bool dc_link_is_dp_sink_present(struct dc_link *link) if (GPIO_RESULT_OK != dal_ddc_open( ddc, GPIO_MODE_INPUT, GPIO_DDC_CONFIG_TYPE_MODE_I2C)) { - dal_gpio_destroy_ddc(&ddc); + dal_ddc_close(ddc); return present; } @@ -817,8 +817,8 @@ static bool dc_link_detect_helper(struct dc_link *link, } case SIGNAL_TYPE_EDP: { - read_current_link_settings_on_detect(link); detect_edp_sink_caps(link); + read_current_link_settings_on_detect(link); sink_caps.transaction_type = DDC_TRANSACTION_TYPE_I2C_OVER_AUX; sink_caps.signal = SIGNAL_TYPE_EDP; break; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c index 7f904d55c1bc..81789191d4ec 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_ddc.c @@ -586,7 +586,7 @@ bool dal_ddc_service_query_ddc_data( bool dal_ddc_submit_aux_command(struct ddc_service *ddc, struct aux_payload *payload) { - uint8_t retrieved = 0; + uint32_t retrieved = 0; bool ret = 0; if (!ddc) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index 0f59b68aa4c2..504055fc70e8 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -3522,7 +3522,14 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) if (link_enc->funcs->fec_set_enable && link->dpcd_caps.fec_cap.bits.FEC_CAPABLE) { if (link->fec_state == dc_link_fec_ready && enable) { - msleep(1); + /* Accord to DP spec, FEC enable sequence can first + * be transmitted anytime after 1000 LL codes have + * been transmitted on the link after link training + * completion. Using 1 lane RBR should have the maximum + * time for transmitting 1000 LL codes which is 6.173 us. + * So use 7 microseconds delay instead. + */ + udelay(7); link_enc->funcs->fec_set_enable(link_enc, true); link->fec_state = dc_link_fec_enabled; } else if (link->fec_state == dc_link_fec_enabled && !enable) { diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c index e472608faf33..793c0cec407f 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c @@ -583,6 +583,8 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc, uint8_t reply; bool payload_reply = true; enum aux_channel_operation_result operation_result; + bool retry_on_defer = false; + int aux_ack_retries = 0, aux_defer_retries = 0, aux_i2c_defer_retries = 0, @@ -613,8 +615,10 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc, break; case AUX_TRANSACTION_REPLY_AUX_DEFER: - case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_NACK: case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_DEFER: + retry_on_defer = true; + /* fall through */ + case AUX_TRANSACTION_REPLY_I2C_OVER_AUX_NACK: if (++aux_defer_retries >= AUX_MAX_DEFER_RETRIES) { goto fail; } else { @@ -647,15 +651,24 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc, break; case AUX_CHANNEL_OPERATION_FAILED_TIMEOUT: - if (++aux_timeout_retries >= AUX_MAX_TIMEOUT_RETRIES) - goto fail; - else { - /* - * DP 1.4, 2.8.2: AUX Transaction Response/Reply Timeouts - * According to the DP spec there should be 3 retries total - * with a 400us wait inbetween each. Hardware already waits - * for 550us therefore no wait is required here. - */ + // Check whether a DEFER had occurred before the timeout. + // If so, treat timeout as a DEFER. + if (retry_on_defer) { + if (++aux_defer_retries >= AUX_MAX_DEFER_RETRIES) + goto fail; + else if (payload->defer_delay > 0) + msleep(payload->defer_delay); + } else { + if (++aux_timeout_retries >= AUX_MAX_TIMEOUT_RETRIES) + goto fail; + else { + /* + * DP 1.4, 2.8.2: AUX Transaction Response/Reply Timeouts + * According to the DP spec there should be 3 retries total + * with a 400us wait inbetween each. Hardware already waits + * for 550us therefore no wait is required here. + */ + } } break; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile index 63f3bddba7da..10b47986526b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn20/Makefile @@ -1,3 +1,4 @@ +# SPDX-License-Identifier: MIT # # Makefile for DCN. diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index 09793336d84f..23ff2f1c75b5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -923,7 +923,9 @@ static const struct resource_caps res_cap_nv14 = { .num_dwb = 1, .num_ddc = 5, .num_vmid = 16, +#ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT .num_dsc = 5, +#endif }; static const struct dc_debug_options debug_defaults_drv = { @@ -1536,13 +1538,20 @@ enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state static void acquire_dsc(struct resource_context *res_ctx, const struct resource_pool *pool, - struct display_stream_compressor **dsc) + struct display_stream_compressor **dsc, + int pipe_idx) { int i; ASSERT(*dsc == NULL); *dsc = NULL; + if (pool->res_cap->num_dsc == pool->res_cap->num_opp) { + *dsc = pool->dscs[pipe_idx]; + res_ctx->is_dsc_acquired[pipe_idx] = true; + return; + } + /* Find first free DSC */ for (i = 0; i < pool->res_cap->num_dsc; i++) if (!res_ctx->is_dsc_acquired[i]) { @@ -1585,7 +1594,7 @@ static enum dc_status add_dsc_to_stream_resource(struct dc *dc, if (pipe_ctx->stream != dc_stream) continue; - acquire_dsc(&dc_ctx->res_ctx, pool, &pipe_ctx->stream_res.dsc); + acquire_dsc(&dc_ctx->res_ctx, pool, &pipe_ctx->stream_res.dsc, i); /* The number of DSCs can be less than the number of pipes */ if (!pipe_ctx->stream_res.dsc) { @@ -1785,7 +1794,7 @@ bool dcn20_split_stream_for_odm( next_odm_pipe->stream_res.opp = pool->opps[next_odm_pipe->pipe_idx]; #ifdef CONFIG_DRM_AMD_DC_DSC_SUPPORT if (next_odm_pipe->stream->timing.flags.DSC == 1) { - acquire_dsc(res_ctx, pool, &next_odm_pipe->stream_res.dsc); + acquire_dsc(res_ctx, pool, &next_odm_pipe->stream_res.dsc, next_odm_pipe->pipe_idx); ASSERT(next_odm_pipe->stream_res.dsc); if (next_odm_pipe->stream_res.dsc == NULL) return false; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c index 4b3401616434..fcb3877b4fcb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_stream_encoder.c @@ -492,15 +492,23 @@ void enc2_stream_encoder_dp_unblank( DP_VID_N_MUL, n_multiply); } - /* set DIG_START to 0x1 to reset FIFO */ + /* make sure stream is disabled before resetting steer fifo */ + REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_ENABLE, false); + REG_WAIT(DP_VID_STREAM_CNTL, DP_VID_STREAM_STATUS, 0, 10, 5000); + /* set DIG_START to 0x1 to reset FIFO */ REG_UPDATE(DIG_FE_CNTL, DIG_START, 1); + udelay(1); /* write 0 to take the FIFO out of reset */ REG_UPDATE(DIG_FE_CNTL, DIG_START, 0); - /* switch DP encoder to CRTC data */ + /* switch DP encoder to CRTC data, but reset it the fifo first. It may happen + * that it overflows during mode transition, and sometimes doesn't recover. + */ + REG_UPDATE(DP_STEER_FIFO, DP_STEER_FIFO_RESET, 1); + udelay(10); REG_UPDATE(DP_STEER_FIFO, DP_STEER_FIFO_RESET, 0); diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile index 14113ccf498d..5b8c17564bc1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn21/Makefile @@ -1,3 +1,4 @@ +# SPDX-License-Identifier: MIT # # Makefile for DCN21. diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c index 459bd9a5caed..b29b2c99a564 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c @@ -23,6 +23,8 @@ * */ +#include <linux/slab.h> + #include "dm_services.h" #include "dc.h" @@ -257,7 +259,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { .vmm_page_size_bytes = 4096, .dram_clock_change_latency_us = 23.84, .return_bus_width_bytes = 64, - .dispclk_dppclk_vco_speed_mhz = 3550, + .dispclk_dppclk_vco_speed_mhz = 3600, .xfc_bus_transport_time_us = 4, .xfc_xbuf_latency_tolerance_us = 4, .use_urgent_burst_bw = 1, @@ -1000,6 +1002,8 @@ static void calculate_wm_set_for_vlevel( pipes[0].clks_cfg.socclk_mhz = dml->soc.clock_limits[vlevel].socclk_mhz; dml->soc.dram_clock_change_latency_us = table_entry->pstate_latency_us; + dml->soc.sr_exit_time_us = table_entry->sr_exit_time_us; + dml->soc.sr_enter_plus_exit_time_us = table_entry->sr_enter_plus_exit_time_us; wm_set->urgent_ns = get_wm_urgent(dml, pipes, pipe_cnt) * 1000; wm_set->cstate_pstate.cstate_enter_plus_exit_ns = get_wm_stutter_enter_exit(dml, pipes, pipe_cnt) * 1000; @@ -1017,14 +1021,21 @@ static void calculate_wm_set_for_vlevel( static void patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb) { + int i; + kernel_fpu_begin(); if (dc->bb_overrides.sr_exit_time_ns) { - bb->sr_exit_time_us = dc->bb_overrides.sr_exit_time_ns / 1000.0; + for (i = 0; i < WM_SET_COUNT; i++) { + dc->clk_mgr->bw_params->wm_table.entries[i].sr_exit_time_us = + dc->bb_overrides.sr_exit_time_ns / 1000.0; + } } if (dc->bb_overrides.sr_enter_plus_exit_time_ns) { - bb->sr_enter_plus_exit_time_us = - dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; + for (i = 0; i < WM_SET_COUNT; i++) { + dc->clk_mgr->bw_params->wm_table.entries[i].sr_enter_plus_exit_time_us = + dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; + } } if (dc->bb_overrides.urgent_latency_ns) { @@ -1032,9 +1043,12 @@ static void patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_s } if (dc->bb_overrides.dram_clock_change_latency_ns) { - bb->dram_clock_change_latency_us = + for (i = 0; i < WM_SET_COUNT; i++) { + dc->clk_mgr->bw_params->wm_table.entries[i].pstate_latency_us = dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; + } } + kernel_fpu_end(); } diff --git a/drivers/gpu/drm/amd/display/dc/dsc/Makefile b/drivers/gpu/drm/amd/display/dc/dsc/Makefile index 970737217e53..641ffb7cfaed 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dsc/Makefile @@ -1,3 +1,4 @@ +# SPDX-License-Identifier: MIT # # Makefile for the 'dsc' sub-component of DAL. 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 4e18e77dcf42..026e6a2a2c44 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 @@ -69,6 +69,8 @@ struct wm_range_table_entry { unsigned int wm_inst; unsigned int wm_type; double pstate_latency_us; + double sr_exit_time_us; + double sr_enter_plus_exit_time_us; bool valid; }; diff --git a/drivers/gpu/drm/amd/display/include/i2caux_interface.h b/drivers/gpu/drm/amd/display/include/i2caux_interface.h index bb012cb1a9f5..c7fbb9c3ad6b 100644 --- a/drivers/gpu/drm/amd/display/include/i2caux_interface.h +++ b/drivers/gpu/drm/amd/display/include/i2caux_interface.h @@ -42,7 +42,7 @@ struct aux_payload { bool write; bool mot; uint32_t address; - uint8_t length; + uint32_t length; uint8_t *data; /* * used to return the reply type of the transaction diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index 16e69bbc69aa..5437b50e9f90 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -37,8 +37,8 @@ #define STATIC_SCREEN_RAMP_DELTA_REFRESH_RATE_PER_FRAME ((1000 / 60) * 65) /* Number of elements in the render times cache array */ #define RENDER_TIMES_MAX_COUNT 10 -/* Threshold to exit/exit BTR (to avoid frequent enter-exits at the lower limit) */ -#define BTR_MAX_MARGIN 2500 +/* Threshold to exit BTR (to avoid frequent enter-exits at the lower limit) */ +#define BTR_EXIT_MARGIN 2000 /* Threshold to change BTR multiplier (to avoid frequent changes) */ #define BTR_DRIFT_MARGIN 2000 /*Threshold to exit fixed refresh rate*/ @@ -254,22 +254,24 @@ static void apply_below_the_range(struct core_freesync *core_freesync, unsigned int delta_from_mid_point_in_us_1 = 0xFFFFFFFF; unsigned int delta_from_mid_point_in_us_2 = 0xFFFFFFFF; unsigned int frames_to_insert = 0; + unsigned int min_frame_duration_in_ns = 0; + unsigned int max_render_time_in_us = in_out_vrr->max_duration_in_us; unsigned int delta_from_mid_point_delta_in_us; - unsigned int max_render_time_in_us = - in_out_vrr->max_duration_in_us - in_out_vrr->btr.margin_in_us; + + min_frame_duration_in_ns = ((unsigned int) (div64_u64( + (1000000000ULL * 1000000), + in_out_vrr->max_refresh_in_uhz))); /* Program BTR */ - if ((last_render_time_in_us + in_out_vrr->btr.margin_in_us / 2) < max_render_time_in_us) { + if (last_render_time_in_us + BTR_EXIT_MARGIN < max_render_time_in_us) { /* Exit Below the Range */ if (in_out_vrr->btr.btr_active) { in_out_vrr->btr.frame_counter = 0; in_out_vrr->btr.btr_active = false; } - } else if (last_render_time_in_us > (max_render_time_in_us + in_out_vrr->btr.margin_in_us / 2)) { + } else if (last_render_time_in_us > max_render_time_in_us) { /* Enter Below the Range */ - if (!in_out_vrr->btr.btr_active) { - in_out_vrr->btr.btr_active = true; - } + in_out_vrr->btr.btr_active = true; } /* BTR set to "not active" so disengage */ @@ -325,9 +327,7 @@ static void apply_below_the_range(struct core_freesync *core_freesync, /* Choose number of frames to insert based on how close it * can get to the mid point of the variable range. */ - if ((frame_time_in_us / mid_point_frames_ceil) > in_out_vrr->min_duration_in_us && - (delta_from_mid_point_in_us_1 < delta_from_mid_point_in_us_2 || - mid_point_frames_floor < 2)) { + if (delta_from_mid_point_in_us_1 < delta_from_mid_point_in_us_2) { frames_to_insert = mid_point_frames_ceil; delta_from_mid_point_delta_in_us = delta_from_mid_point_in_us_2 - delta_from_mid_point_in_us_1; @@ -343,7 +343,7 @@ static void apply_below_the_range(struct core_freesync *core_freesync, if (in_out_vrr->btr.frames_to_insert != 0 && delta_from_mid_point_delta_in_us < BTR_DRIFT_MARGIN) { if (((last_render_time_in_us / in_out_vrr->btr.frames_to_insert) < - max_render_time_in_us) && + in_out_vrr->max_duration_in_us) && ((last_render_time_in_us / in_out_vrr->btr.frames_to_insert) > in_out_vrr->min_duration_in_us)) frames_to_insert = in_out_vrr->btr.frames_to_insert; @@ -796,11 +796,6 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, refresh_range = in_out_vrr->max_refresh_in_uhz - in_out_vrr->min_refresh_in_uhz; - in_out_vrr->btr.margin_in_us = in_out_vrr->max_duration_in_us - - 2 * in_out_vrr->min_duration_in_us; - if (in_out_vrr->btr.margin_in_us > BTR_MAX_MARGIN) - in_out_vrr->btr.margin_in_us = BTR_MAX_MARGIN; - in_out_vrr->supported = true; } @@ -816,7 +811,6 @@ void mod_freesync_build_vrr_params(struct mod_freesync *mod_freesync, in_out_vrr->btr.inserted_duration_in_us = 0; in_out_vrr->btr.frames_to_insert = 0; in_out_vrr->btr.frame_counter = 0; - in_out_vrr->btr.mid_point_in_us = (in_out_vrr->min_duration_in_us + in_out_vrr->max_duration_in_us) / 2; diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h index dbe7835aabcf..dc187844d10b 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_freesync.h @@ -92,7 +92,6 @@ struct mod_vrr_params_btr { uint32_t inserted_duration_in_us; uint32_t frames_to_insert; uint32_t frame_counter; - uint32_t margin_in_us; }; struct mod_vrr_params_fixed_refresh { diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c index 5ff7ccedfbed..a23729d3174b 100644 --- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c @@ -866,6 +866,7 @@ static int smu_sw_init(void *handle) smu->smu_baco.platform_support = false; mutex_init(&smu->sensor_lock); + mutex_init(&smu->metrics_lock); smu->watermarks_bitmap = 0; smu->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT; diff --git a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c index ce3566ca3e24..472e9fed411a 100644 --- a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c @@ -862,18 +862,21 @@ static int arcturus_get_metrics_table(struct smu_context *smu, struct smu_table_context *smu_table= &smu->smu_table; int ret = 0; + mutex_lock(&smu->metrics_lock); if (!smu_table->metrics_time || time_after(jiffies, smu_table->metrics_time + HZ / 1000)) { ret = smu_update_table(smu, SMU_TABLE_SMU_METRICS, 0, (void *)smu_table->metrics_table, false); if (ret) { pr_info("Failed to export SMU metrics table!\n"); + mutex_unlock(&smu->metrics_lock); return ret; } smu_table->metrics_time = jiffies; } memcpy(metrics_table, smu_table->metrics_table, sizeof(SmuMetrics_t)); + mutex_unlock(&smu->metrics_lock); return ret; } @@ -1313,12 +1316,17 @@ static int arcturus_get_power_profile_mode(struct smu_context *smu, "VR", "COMPUTE", "CUSTOM"}; + static const char *title[] = { + "PROFILE_INDEX(NAME)"}; uint32_t i, size = 0; int16_t workload_type = 0; if (!smu->pm_enabled || !buf) return -EINVAL; + size += sprintf(buf + size, "%16s\n", + title[0]); + for (i = 0; i <= PP_SMC_POWER_PROFILE_CUSTOM; i++) { /* * Conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h index ac9758305ab3..41fce75b263f 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h @@ -349,6 +349,7 @@ struct smu_context const struct pptable_funcs *ppt_funcs; struct mutex mutex; struct mutex sensor_lock; + struct mutex metrics_lock; uint64_t pool_size; struct smu_table_context smu_table; diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c index 4a14fd1f9fd5..ca62e92e5a4f 100644 --- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c @@ -562,17 +562,20 @@ static int navi10_get_metrics_table(struct smu_context *smu, struct smu_table_context *smu_table= &smu->smu_table; int ret = 0; + mutex_lock(&smu->metrics_lock); if (!smu_table->metrics_time || time_after(jiffies, smu_table->metrics_time + msecs_to_jiffies(100))) { ret = smu_update_table(smu, SMU_TABLE_SMU_METRICS, 0, (void *)smu_table->metrics_table, false); if (ret) { pr_info("Failed to export SMU metrics table!\n"); + mutex_unlock(&smu->metrics_lock); return ret; } smu_table->metrics_time = jiffies; } memcpy(metrics_table, smu_table->metrics_table, sizeof(SmuMetrics_t)); + mutex_unlock(&smu->metrics_lock); return ret; } diff --git a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c index 60b9ff097142..0d3a3b0a934e 100644 --- a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c +++ b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c @@ -1678,17 +1678,20 @@ static int vega20_get_metrics_table(struct smu_context *smu, struct smu_table_context *smu_table= &smu->smu_table; int ret = 0; + mutex_lock(&smu->metrics_lock); if (!smu_table->metrics_time || time_after(jiffies, smu_table->metrics_time + HZ / 1000)) { ret = smu_update_table(smu, SMU_TABLE_SMU_METRICS, 0, (void *)smu_table->metrics_table, false); if (ret) { pr_info("Failed to export SMU metrics table!\n"); + mutex_unlock(&smu->metrics_lock); return ret; } smu_table->metrics_time = jiffies; } memcpy(metrics_table, smu_table->metrics_table, sizeof(SmuMetrics_t)); + mutex_unlock(&smu->metrics_lock); return ret; } diff --git a/drivers/gpu/drm/arm/malidp_mw.c b/drivers/gpu/drm/arm/malidp_mw.c index 875a3a9eabfa..7d0e7b031e44 100644 --- a/drivers/gpu/drm/arm/malidp_mw.c +++ b/drivers/gpu/drm/arm/malidp_mw.c @@ -56,7 +56,7 @@ malidp_mw_connector_mode_valid(struct drm_connector *connector, return MODE_OK; } -const struct drm_connector_helper_funcs malidp_mw_connector_helper_funcs = { +static const struct drm_connector_helper_funcs malidp_mw_connector_helper_funcs = { .get_modes = malidp_mw_connector_get_modes, .mode_valid = malidp_mw_connector_mode_valid, }; diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index 273dd80fabf3..e6afe4faeca6 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -393,7 +393,7 @@ drm_dp_encode_sideband_req(const struct drm_dp_sideband_msg_req_body *req, memcpy(&buf[idx], req->u.i2c_read.transactions[i].bytes, req->u.i2c_read.transactions[i].num_bytes); idx += req->u.i2c_read.transactions[i].num_bytes; - buf[idx] = (req->u.i2c_read.transactions[i].no_stop_bit & 0x1) << 5; + buf[idx] = (req->u.i2c_read.transactions[i].no_stop_bit & 0x1) << 4; buf[idx] |= (req->u.i2c_read.transactions[i].i2c_transaction_delay & 0xf); idx++; } @@ -1190,6 +1190,8 @@ static int drm_dp_mst_wait_tx_reply(struct drm_dp_mst_branch *mstb, txmsg->state == DRM_DP_SIDEBAND_TX_SENT) { mstb->tx_slots[txmsg->seqno] = NULL; } + mgr->is_waiting_for_dwn_reply = false; + } out: if (unlikely(ret == -EIO) && drm_debug_enabled(DRM_UT_DP)) { @@ -1199,6 +1201,7 @@ out: } mutex_unlock(&mgr->qlock); + drm_dp_mst_kick_tx(mgr); return ret; } @@ -1913,73 +1916,90 @@ static u8 drm_dp_calculate_rad(struct drm_dp_mst_port *port, return parent_lct + 1; } -static int drm_dp_port_set_pdt(struct drm_dp_mst_port *port, u8 new_pdt) +static bool drm_dp_mst_is_dp_mst_end_device(u8 pdt, bool mcs) +{ + switch (pdt) { + case DP_PEER_DEVICE_DP_LEGACY_CONV: + case DP_PEER_DEVICE_SST_SINK: + return true; + case DP_PEER_DEVICE_MST_BRANCHING: + /* For sst branch device */ + if (!mcs) + return true; + + return false; + } + return true; +} + +static int +drm_dp_port_set_pdt(struct drm_dp_mst_port *port, u8 new_pdt, + bool new_mcs) { struct drm_dp_mst_topology_mgr *mgr = port->mgr; struct drm_dp_mst_branch *mstb; u8 rad[8], lct; int ret = 0; - if (port->pdt == new_pdt) + if (port->pdt == new_pdt && port->mcs == new_mcs) return 0; /* Teardown the old pdt, if there is one */ - switch (port->pdt) { - case DP_PEER_DEVICE_DP_LEGACY_CONV: - case DP_PEER_DEVICE_SST_SINK: - /* - * If the new PDT would also have an i2c bus, don't bother - * with reregistering it - */ - if (new_pdt == DP_PEER_DEVICE_DP_LEGACY_CONV || - new_pdt == DP_PEER_DEVICE_SST_SINK) { - port->pdt = new_pdt; - return 0; - } + if (port->pdt != DP_PEER_DEVICE_NONE) { + if (drm_dp_mst_is_dp_mst_end_device(port->pdt, port->mcs)) { + /* + * If the new PDT would also have an i2c bus, + * don't bother with reregistering it + */ + if (new_pdt != DP_PEER_DEVICE_NONE && + drm_dp_mst_is_dp_mst_end_device(new_pdt, new_mcs)) { + port->pdt = new_pdt; + port->mcs = new_mcs; + return 0; + } - /* remove i2c over sideband */ - drm_dp_mst_unregister_i2c_bus(&port->aux); - break; - case DP_PEER_DEVICE_MST_BRANCHING: - mutex_lock(&mgr->lock); - drm_dp_mst_topology_put_mstb(port->mstb); - port->mstb = NULL; - mutex_unlock(&mgr->lock); - break; + /* remove i2c over sideband */ + drm_dp_mst_unregister_i2c_bus(&port->aux); + } else { + mutex_lock(&mgr->lock); + drm_dp_mst_topology_put_mstb(port->mstb); + port->mstb = NULL; + mutex_unlock(&mgr->lock); + } } port->pdt = new_pdt; - switch (port->pdt) { - case DP_PEER_DEVICE_DP_LEGACY_CONV: - case DP_PEER_DEVICE_SST_SINK: - /* add i2c over sideband */ - ret = drm_dp_mst_register_i2c_bus(&port->aux); - break; + port->mcs = new_mcs; - case DP_PEER_DEVICE_MST_BRANCHING: - lct = drm_dp_calculate_rad(port, rad); - mstb = drm_dp_add_mst_branch_device(lct, rad); - if (!mstb) { - ret = -ENOMEM; - DRM_ERROR("Failed to create MSTB for port %p", port); - goto out; - } + if (port->pdt != DP_PEER_DEVICE_NONE) { + if (drm_dp_mst_is_dp_mst_end_device(port->pdt, port->mcs)) { + /* add i2c over sideband */ + ret = drm_dp_mst_register_i2c_bus(&port->aux); + } else { + lct = drm_dp_calculate_rad(port, rad); + mstb = drm_dp_add_mst_branch_device(lct, rad); + if (!mstb) { + ret = -ENOMEM; + DRM_ERROR("Failed to create MSTB for port %p", + port); + goto out; + } - mutex_lock(&mgr->lock); - port->mstb = mstb; - mstb->mgr = port->mgr; - mstb->port_parent = port; + mutex_lock(&mgr->lock); + port->mstb = mstb; + mstb->mgr = port->mgr; + mstb->port_parent = port; - /* - * Make sure this port's memory allocation stays - * around until its child MSTB releases it - */ - drm_dp_mst_get_port_malloc(port); - mutex_unlock(&mgr->lock); + /* + * Make sure this port's memory allocation stays + * around until its child MSTB releases it + */ + drm_dp_mst_get_port_malloc(port); + mutex_unlock(&mgr->lock); - /* And make sure we send a link address for this */ - ret = 1; - break; + /* And make sure we send a link address for this */ + ret = 1; + } } out: @@ -2132,9 +2152,8 @@ drm_dp_mst_port_add_connector(struct drm_dp_mst_branch *mstb, goto error; } - if ((port->pdt == DP_PEER_DEVICE_DP_LEGACY_CONV || - port->pdt == DP_PEER_DEVICE_SST_SINK) && - port->port_num >= DP_MST_LOGICAL_PORT_0) { + if (port->pdt != DP_PEER_DEVICE_NONE && + drm_dp_mst_is_dp_mst_end_device(port->pdt, port->mcs)) { port->cached_edid = drm_get_edid(port->connector, &port->aux.ddc); drm_connector_set_tile_property(port->connector); @@ -2198,6 +2217,7 @@ drm_dp_mst_handle_link_address_port(struct drm_dp_mst_branch *mstb, struct drm_dp_mst_port *port; int old_ddps = 0, ret; u8 new_pdt = DP_PEER_DEVICE_NONE; + bool new_mcs = 0; bool created = false, send_link_addr = false, changed = false; port = drm_dp_get_port(mstb, port_msg->port_number); @@ -2242,7 +2262,7 @@ drm_dp_mst_handle_link_address_port(struct drm_dp_mst_branch *mstb, port->input = port_msg->input_port; if (!port->input) new_pdt = port_msg->peer_device_type; - port->mcs = port_msg->mcs; + new_mcs = port_msg->mcs; port->ddps = port_msg->ddps; port->ldps = port_msg->legacy_device_plug_status; port->dpcd_rev = port_msg->dpcd_revision; @@ -2269,7 +2289,7 @@ drm_dp_mst_handle_link_address_port(struct drm_dp_mst_branch *mstb, } } - ret = drm_dp_port_set_pdt(port, new_pdt); + ret = drm_dp_port_set_pdt(port, new_pdt, new_mcs); if (ret == 1) { send_link_addr = true; } else if (ret < 0) { @@ -2283,7 +2303,8 @@ drm_dp_mst_handle_link_address_port(struct drm_dp_mst_branch *mstb, * we're coming out of suspend. In this case, always resend the link * address if there's an MSTB on this port */ - if (!created && port->pdt == DP_PEER_DEVICE_MST_BRANCHING) + if (!created && port->pdt == DP_PEER_DEVICE_MST_BRANCHING && + port->mcs) send_link_addr = true; if (port->connector) @@ -2318,8 +2339,9 @@ drm_dp_mst_handle_conn_stat(struct drm_dp_mst_branch *mstb, { struct drm_dp_mst_topology_mgr *mgr = mstb->mgr; struct drm_dp_mst_port *port; - int old_ddps, ret; + int old_ddps, old_input, ret, i; u8 new_pdt; + bool new_mcs; bool dowork = false, create_connector = false; port = drm_dp_get_port(mstb, conn_stat->port_number); @@ -2349,8 +2371,8 @@ drm_dp_mst_handle_conn_stat(struct drm_dp_mst_branch *mstb, } old_ddps = port->ddps; + old_input = port->input; port->input = conn_stat->input_port; - port->mcs = conn_stat->message_capability_status; port->ldps = conn_stat->legacy_device_plug_status; port->ddps = conn_stat->displayport_device_plug_status; @@ -2363,8 +2385,8 @@ drm_dp_mst_handle_conn_stat(struct drm_dp_mst_branch *mstb, } new_pdt = port->input ? DP_PEER_DEVICE_NONE : conn_stat->peer_device_type; - - ret = drm_dp_port_set_pdt(port, new_pdt); + new_mcs = conn_stat->message_capability_status; + ret = drm_dp_port_set_pdt(port, new_pdt, new_mcs); if (ret == 1) { dowork = true; } else if (ret < 0) { @@ -2373,6 +2395,28 @@ drm_dp_mst_handle_conn_stat(struct drm_dp_mst_branch *mstb, dowork = false; } + if (!old_input && old_ddps != port->ddps && !port->ddps) { + for (i = 0; i < mgr->max_payloads; i++) { + struct drm_dp_vcpi *vcpi = mgr->proposed_vcpis[i]; + struct drm_dp_mst_port *port_validated; + + if (!vcpi) + continue; + + port_validated = + container_of(vcpi, struct drm_dp_mst_port, vcpi); + port_validated = + drm_dp_mst_topology_get_port_validated(mgr, port_validated); + if (!port_validated) { + mutex_lock(&mgr->payload_lock); + vcpi->num_slots = 0; + mutex_unlock(&mgr->payload_lock); + } else { + drm_dp_mst_topology_put_port(port_validated); + } + } + } + if (port->connector) drm_modeset_unlock(&mgr->base.lock); else if (create_connector) @@ -2718,9 +2762,11 @@ static void process_single_down_tx_qlock(struct drm_dp_mst_topology_mgr *mgr) ret = process_single_tx_qlock(mgr, txmsg, false); if (ret == 1) { /* txmsg is sent it should be in the slots now */ + mgr->is_waiting_for_dwn_reply = true; list_del(&txmsg->next); } else if (ret) { DRM_DEBUG_KMS("failed to send msg in q %d\n", ret); + mgr->is_waiting_for_dwn_reply = false; list_del(&txmsg->next); if (txmsg->seqno != -1) txmsg->dst->tx_slots[txmsg->seqno] = NULL; @@ -2760,7 +2806,8 @@ static void drm_dp_queue_down_tx(struct drm_dp_mst_topology_mgr *mgr, drm_dp_mst_dump_sideband_msg_tx(&p, txmsg); } - if (list_is_singular(&mgr->tx_msg_downq)) + if (list_is_singular(&mgr->tx_msg_downq) && + !mgr->is_waiting_for_dwn_reply) process_single_down_tx_qlock(mgr); mutex_unlock(&mgr->qlock); } @@ -3678,6 +3725,7 @@ static int drm_dp_mst_handle_down_rep(struct drm_dp_mst_topology_mgr *mgr) mutex_lock(&mgr->qlock); txmsg->state = DRM_DP_SIDEBAND_TX_RX; mstb->tx_slots[slot] = NULL; + mgr->is_waiting_for_dwn_reply = false; mutex_unlock(&mgr->qlock); wake_up_all(&mgr->tx_waitq); @@ -3687,6 +3735,9 @@ static int drm_dp_mst_handle_down_rep(struct drm_dp_mst_topology_mgr *mgr) no_msg: drm_dp_mst_topology_put_mstb(mstb); clear_down_rep_recv: + mutex_lock(&mgr->qlock); + mgr->is_waiting_for_dwn_reply = false; + mutex_unlock(&mgr->qlock); memset(&mgr->down_rep_recv, 0, sizeof(struct drm_dp_sideband_msg_rx)); return 0; @@ -3896,6 +3947,8 @@ drm_dp_mst_detect_port(struct drm_connector *connector, switch (port->pdt) { case DP_PEER_DEVICE_NONE: case DP_PEER_DEVICE_MST_BRANCHING: + if (!port->mcs) + ret = connector_status_connected; break; case DP_PEER_DEVICE_SST_SINK: @@ -4497,7 +4550,7 @@ static void drm_dp_tx_work(struct work_struct *work) struct drm_dp_mst_topology_mgr *mgr = container_of(work, struct drm_dp_mst_topology_mgr, tx_work); mutex_lock(&mgr->qlock); - if (!list_empty(&mgr->tx_msg_downq)) + if (!list_empty(&mgr->tx_msg_downq) && !mgr->is_waiting_for_dwn_reply) process_single_down_tx_qlock(mgr); mutex_unlock(&mgr->qlock); } @@ -4508,7 +4561,7 @@ drm_dp_delayed_destroy_port(struct drm_dp_mst_port *port) if (port->connector) port->mgr->cbs->destroy_connector(port->mgr, port->connector); - drm_dp_port_set_pdt(port, DP_PEER_DEVICE_NONE); + drm_dp_port_set_pdt(port, DP_PEER_DEVICE_NONE, port->mcs); drm_dp_mst_put_port_malloc(port); } diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 8ebeccdeed23..d8e8f3960f4d 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -1283,7 +1283,7 @@ int drm_fb_helper_check_var(struct fb_var_screeninfo *var, * Changes struct fb_var_screeninfo are currently not pushed back * to KMS, hence fail if different settings are requested. */ - if (var->bits_per_pixel != fb->format->cpp[0] * 8 || + if (var->bits_per_pixel > fb->format->cpp[0] * 8 || var->xres > fb->width || var->yres > fb->height || var->xres_virtual > fb->width || var->yres_virtual > fb->height) { DRM_DEBUG("fb requested width/height/bpp can't fit in current fb " @@ -1309,6 +1309,11 @@ int drm_fb_helper_check_var(struct fb_var_screeninfo *var, } /* + * Likewise, bits_per_pixel should be rounded up to a supported value. + */ + var->bits_per_pixel = fb->format->cpp[0] * 8; + + /* * drm fbdev emulation doesn't support changing the pixel format at all, * so reject all pixel format changing requests. */ diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c index 7ae087b0504d..88b6fcaa20be 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c @@ -1313,6 +1313,7 @@ static int gsc_remove(struct platform_device *pdev) { struct device *dev = &pdev->dev; + component_del(dev, &gsc_component_ops); pm_runtime_dont_use_autosuspend(dev); pm_runtime_disable(dev); diff --git a/drivers/gpu/drm/gma500/gtt.c b/drivers/gpu/drm/gma500/gtt.c index afaf4bea21cf..9278bcfad1bf 100644 --- a/drivers/gpu/drm/gma500/gtt.c +++ b/drivers/gpu/drm/gma500/gtt.c @@ -503,7 +503,7 @@ int psb_gtt_init(struct drm_device *dev, int resume) * Map the GTT and the stolen memory area */ if (!resume) - dev_priv->gtt_map = ioremap_nocache(pg->gtt_phys_start, + dev_priv->gtt_map = ioremap(pg->gtt_phys_start, gtt_pages << PAGE_SHIFT); if (!dev_priv->gtt_map) { dev_err(dev->dev, "Failure to map gtt.\n"); diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c index 7005f8f69c68..0900052fc484 100644 --- a/drivers/gpu/drm/gma500/psb_drv.c +++ b/drivers/gpu/drm/gma500/psb_drv.c @@ -256,7 +256,7 @@ static int psb_driver_load(struct drm_device *dev, unsigned long flags) PSB_AUX_RESOURCE); resource_len = pci_resource_len(dev_priv->aux_pdev, PSB_AUX_RESOURCE); - dev_priv->aux_reg = ioremap_nocache(resource_start, + dev_priv->aux_reg = ioremap(resource_start, resource_len); if (!dev_priv->aux_reg) goto out_err; diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c index 2fd4ca91a62d..8dd5a43e5486 100644 --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c @@ -211,7 +211,7 @@ static int hibmc_hw_map(struct hibmc_drm_private *priv) ioaddr = pci_resource_start(pdev, 1); iosize = pci_resource_len(pdev, 1); - priv->mmio = devm_ioremap_nocache(dev->dev, ioaddr, iosize); + priv->mmio = devm_ioremap(dev->dev, ioaddr, iosize); if (!priv->mmio) { DRM_ERROR("Cannot map mmio region\n"); return -ENOMEM; diff --git a/drivers/gpu/drm/i915/display/intel_audio.c b/drivers/gpu/drm/i915/display/intel_audio.c index 85e6b2bbb34f..3a5ac13d5801 100644 --- a/drivers/gpu/drm/i915/display/intel_audio.c +++ b/drivers/gpu/drm/i915/display/intel_audio.c @@ -856,7 +856,7 @@ static unsigned long i915_audio_component_get_power(struct device *kdev) } /* Force CDCLK to 2*BCLK as long as we need audio powered. */ - if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) + if (IS_GEMINILAKE(dev_priv)) glk_force_audio_cdclk(dev_priv, true); if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) @@ -875,7 +875,7 @@ static void i915_audio_component_put_power(struct device *kdev, /* Stop forcing CDCLK to 2*BCLK if no need for audio to be powered. */ if (--dev_priv->audio_power_refcount == 0) - if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) + if (IS_GEMINILAKE(dev_priv)) glk_force_audio_cdclk(dev_priv, false); intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO, cookie); diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c index c7c2b349858d..2a27fb5d7dc6 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.c +++ b/drivers/gpu/drm/i915/display/intel_ddi.c @@ -3986,6 +3986,7 @@ static void intel_enable_ddi(struct intel_encoder *encoder, if (conn_state->content_protection == DRM_MODE_CONTENT_PROTECTION_DESIRED) intel_hdcp_enable(to_intel_connector(conn_state->connector), + crtc_state->cpu_transcoder, (u8)conn_state->hdcp_content_type); } @@ -4089,7 +4090,9 @@ static void intel_ddi_update_pipe(struct intel_encoder *encoder, if (conn_state->content_protection == DRM_MODE_CONTENT_PROTECTION_DESIRED || content_protection_type_changed) - intel_hdcp_enable(connector, (u8)conn_state->hdcp_content_type); + intel_hdcp_enable(connector, + crtc_state->cpu_transcoder, + (u8)conn_state->hdcp_content_type); } static void diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 6f5e3bd13ad1..301897791627 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -4515,8 +4515,6 @@ static void icl_disable_transcoder_port_sync(const struct intel_crtc_state *old_ { struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->base.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - i915_reg_t reg; - u32 trans_ddi_func_ctl2_val; if (old_crtc_state->master_transcoder == INVALID_TRANSCODER) return; @@ -4524,10 +4522,7 @@ static void icl_disable_transcoder_port_sync(const struct intel_crtc_state *old_ DRM_DEBUG_KMS("Disabling Transcoder Port Sync on Slave Transcoder %s\n", transcoder_name(old_crtc_state->cpu_transcoder)); - reg = TRANS_DDI_FUNC_CTL2(old_crtc_state->cpu_transcoder); - trans_ddi_func_ctl2_val = ~(PORT_SYNC_MODE_ENABLE | - PORT_SYNC_MODE_MASTER_SELECT_MASK); - I915_WRITE(reg, trans_ddi_func_ctl2_val); + I915_WRITE(TRANS_DDI_FUNC_CTL2(old_crtc_state->cpu_transcoder), 0); } static void intel_fdi_normal_train(struct intel_crtc *crtc) @@ -15112,7 +15107,7 @@ intel_prepare_plane_fb(struct drm_plane *plane, return ret; fb_obj_bump_render_priority(obj); - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_DIRTYFB); + i915_gem_object_flush_frontbuffer(obj, ORIGIN_DIRTYFB); if (!new_plane_state->base.fence) { /* implicit fencing */ struct dma_fence *fence; diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c index ce1b64f4dd44..12ba74788cce 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.c +++ b/drivers/gpu/drm/i915/display/intel_display_power.c @@ -3688,6 +3688,151 @@ static const struct i915_power_well_desc icl_power_wells[] = { }, }; +static const struct i915_power_well_desc ehl_power_wells[] = { + { + .name = "always-on", + .always_on = true, + .domains = POWER_DOMAIN_MASK, + .ops = &i9xx_always_on_power_well_ops, + .id = DISP_PW_ID_NONE, + }, + { + .name = "power well 1", + /* Handled by the DMC firmware */ + .always_on = true, + .domains = 0, + .ops = &hsw_power_well_ops, + .id = SKL_DISP_PW_1, + { + .hsw.regs = &hsw_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_PW_1, + .hsw.has_fuses = true, + }, + }, + { + .name = "DC off", + .domains = ICL_DISPLAY_DC_OFF_POWER_DOMAINS, + .ops = &gen9_dc_off_power_well_ops, + .id = SKL_DISP_DC_OFF, + }, + { + .name = "power well 2", + .domains = ICL_PW_2_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = SKL_DISP_PW_2, + { + .hsw.regs = &hsw_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_PW_2, + .hsw.has_fuses = true, + }, + }, + { + .name = "power well 3", + .domains = ICL_PW_3_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &hsw_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_PW_3, + .hsw.irq_pipe_mask = BIT(PIPE_B), + .hsw.has_vga = true, + .hsw.has_fuses = true, + }, + }, + { + .name = "DDI A IO", + .domains = ICL_DDI_IO_A_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_ddi_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_DDI_A, + }, + }, + { + .name = "DDI B IO", + .domains = ICL_DDI_IO_B_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_ddi_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_DDI_B, + }, + }, + { + .name = "DDI C IO", + .domains = ICL_DDI_IO_C_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_ddi_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_DDI_C, + }, + }, + { + .name = "DDI D IO", + .domains = ICL_DDI_IO_D_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_ddi_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_DDI_D, + }, + }, + { + .name = "AUX A", + .domains = ICL_AUX_A_IO_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_aux_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_AUX_A, + }, + }, + { + .name = "AUX B", + .domains = ICL_AUX_B_IO_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_aux_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_AUX_B, + }, + }, + { + .name = "AUX C", + .domains = ICL_AUX_C_TC1_IO_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_aux_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_AUX_C, + }, + }, + { + .name = "AUX D", + .domains = ICL_AUX_D_TC2_IO_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &icl_aux_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_AUX_D, + }, + }, + { + .name = "power well 4", + .domains = ICL_PW_4_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = DISP_PW_ID_NONE, + { + .hsw.regs = &hsw_power_well_regs, + .hsw.idx = ICL_PW_CTL_IDX_PW_4, + .hsw.has_fuses = true, + .hsw.irq_pipe_mask = BIT(PIPE_C), + }, + }, +}; + static const struct i915_power_well_desc tgl_power_wells[] = { { .name = "always-on", @@ -3832,7 +3977,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { { .name = "AUX A", .domains = TGL_AUX_A_IO_POWER_DOMAINS, - .ops = &icl_combo_phy_aux_power_well_ops, + .ops = &hsw_power_well_ops, .id = DISP_PW_ID_NONE, { .hsw.regs = &icl_aux_power_well_regs, @@ -3842,7 +3987,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { { .name = "AUX B", .domains = TGL_AUX_B_IO_POWER_DOMAINS, - .ops = &icl_combo_phy_aux_power_well_ops, + .ops = &hsw_power_well_ops, .id = DISP_PW_ID_NONE, { .hsw.regs = &icl_aux_power_well_regs, @@ -3852,7 +3997,7 @@ static const struct i915_power_well_desc tgl_power_wells[] = { { .name = "AUX C", .domains = TGL_AUX_C_IO_POWER_DOMAINS, - .ops = &icl_combo_phy_aux_power_well_ops, + .ops = &hsw_power_well_ops, .id = DISP_PW_ID_NONE, { .hsw.regs = &icl_aux_power_well_regs, @@ -4162,6 +4307,8 @@ int intel_power_domains_init(struct drm_i915_private *dev_priv) */ if (IS_GEN(dev_priv, 12)) { err = set_power_wells(power_domains, tgl_power_wells); + } else if (IS_ELKHARTLAKE(dev_priv)) { + err = set_power_wells(power_domains, ehl_power_wells); } else if (IS_GEN(dev_priv, 11)) { err = set_power_wells(power_domains, icl_power_wells); } else if (IS_CANNONLAKE(dev_priv)) { diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 050655a1a3d8..b05b2191b919 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -2414,9 +2414,6 @@ intel_dp_compute_config(struct intel_encoder *encoder, intel_psr_compute_config(intel_dp, pipe_config); - intel_hdcp_transcoder_config(intel_connector, - pipe_config->cpu_transcoder); - return 0; } diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c index 3111ecaeabd0..20616639b8ab 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.c +++ b/drivers/gpu/drm/i915/display/intel_fbc.c @@ -1284,7 +1284,7 @@ static int intel_sanitize_fbc_option(struct drm_i915_private *dev_priv) return 0; /* https://bugs.freedesktop.org/show_bug.cgi?id=108085 */ - if (IS_GEMINILAKE(dev_priv)) + if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) return 0; if (IS_BROADWELL(dev_priv) || INTEL_GEN(dev_priv) >= 9) diff --git a/drivers/gpu/drm/i915/display/intel_frontbuffer.c b/drivers/gpu/drm/i915/display/intel_frontbuffer.c index 84b164f31895..6cb02c912acc 100644 --- a/drivers/gpu/drm/i915/display/intel_frontbuffer.c +++ b/drivers/gpu/drm/i915/display/intel_frontbuffer.c @@ -229,11 +229,11 @@ static void frontbuffer_release(struct kref *ref) vma->display_alignment = I915_GTT_MIN_ALIGNMENT; spin_unlock(&obj->vma.lock); - obj->frontbuffer = NULL; + RCU_INIT_POINTER(obj->frontbuffer, NULL); spin_unlock(&to_i915(obj->base.dev)->fb_tracking.lock); i915_gem_object_put(obj); - kfree(front); + kfree_rcu(front, rcu); } struct intel_frontbuffer * @@ -242,11 +242,7 @@ intel_frontbuffer_get(struct drm_i915_gem_object *obj) struct drm_i915_private *i915 = to_i915(obj->base.dev); struct intel_frontbuffer *front; - spin_lock(&i915->fb_tracking.lock); - front = obj->frontbuffer; - if (front) - kref_get(&front->ref); - spin_unlock(&i915->fb_tracking.lock); + front = __intel_frontbuffer_get(obj); if (front) return front; @@ -262,13 +258,13 @@ intel_frontbuffer_get(struct drm_i915_gem_object *obj) i915_active_may_sleep(frontbuffer_retire)); spin_lock(&i915->fb_tracking.lock); - if (obj->frontbuffer) { + if (rcu_access_pointer(obj->frontbuffer)) { kfree(front); - front = obj->frontbuffer; + front = rcu_dereference_protected(obj->frontbuffer, true); kref_get(&front->ref); } else { i915_gem_object_get(obj); - obj->frontbuffer = front; + rcu_assign_pointer(obj->frontbuffer, front); } spin_unlock(&i915->fb_tracking.lock); diff --git a/drivers/gpu/drm/i915/display/intel_frontbuffer.h b/drivers/gpu/drm/i915/display/intel_frontbuffer.h index adc64d61a4a5..6d41f5394425 100644 --- a/drivers/gpu/drm/i915/display/intel_frontbuffer.h +++ b/drivers/gpu/drm/i915/display/intel_frontbuffer.h @@ -27,10 +27,10 @@ #include <linux/atomic.h> #include <linux/kref.h> +#include "gem/i915_gem_object_types.h" #include "i915_active.h" struct drm_i915_private; -struct drm_i915_gem_object; enum fb_op_origin { ORIGIN_GTT, @@ -45,6 +45,7 @@ struct intel_frontbuffer { atomic_t bits; struct i915_active write; struct drm_i915_gem_object *obj; + struct rcu_head rcu; }; void intel_frontbuffer_flip_prepare(struct drm_i915_private *i915, @@ -54,6 +55,35 @@ void intel_frontbuffer_flip_complete(struct drm_i915_private *i915, void intel_frontbuffer_flip(struct drm_i915_private *i915, unsigned frontbuffer_bits); +void intel_frontbuffer_put(struct intel_frontbuffer *front); + +static inline struct intel_frontbuffer * +__intel_frontbuffer_get(const struct drm_i915_gem_object *obj) +{ + struct intel_frontbuffer *front; + + if (likely(!rcu_access_pointer(obj->frontbuffer))) + return NULL; + + rcu_read_lock(); + do { + front = rcu_dereference(obj->frontbuffer); + if (!front) + break; + + if (unlikely(!kref_get_unless_zero(&front->ref))) + continue; + + if (likely(front == rcu_access_pointer(obj->frontbuffer))) + break; + + intel_frontbuffer_put(front); + } while (1); + rcu_read_unlock(); + + return front; +} + struct intel_frontbuffer * intel_frontbuffer_get(struct drm_i915_gem_object *obj); @@ -119,6 +149,4 @@ void intel_frontbuffer_track(struct intel_frontbuffer *old, struct intel_frontbuffer *new, unsigned int frontbuffer_bits); -void intel_frontbuffer_put(struct intel_frontbuffer *front); - #endif /* __INTEL_FRONTBUFFER_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index f1f41ca8402b..a448815d8fc2 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -1821,23 +1821,6 @@ enum mei_fw_tc intel_get_mei_fw_tc(enum transcoder cpu_transcoder) } } -void intel_hdcp_transcoder_config(struct intel_connector *connector, - enum transcoder cpu_transcoder) -{ - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); - struct intel_hdcp *hdcp = &connector->hdcp; - - if (!hdcp->shim) - return; - - if (INTEL_GEN(dev_priv) >= 12) { - mutex_lock(&hdcp->mutex); - hdcp->cpu_transcoder = cpu_transcoder; - hdcp->port_data.fw_tc = intel_get_mei_fw_tc(cpu_transcoder); - mutex_unlock(&hdcp->mutex); - } -} - static inline int initialize_hdcp_port_data(struct intel_connector *connector, const struct intel_hdcp_shim *shim) { @@ -1959,8 +1942,10 @@ int intel_hdcp_init(struct intel_connector *connector, return 0; } -int intel_hdcp_enable(struct intel_connector *connector, u8 content_type) +int intel_hdcp_enable(struct intel_connector *connector, + enum transcoder cpu_transcoder, u8 content_type) { + struct drm_i915_private *dev_priv = to_i915(connector->base.dev); struct intel_hdcp *hdcp = &connector->hdcp; unsigned long check_link_interval = DRM_HDCP_CHECK_PERIOD_MS; int ret = -EINVAL; @@ -1972,6 +1957,11 @@ int intel_hdcp_enable(struct intel_connector *connector, u8 content_type) WARN_ON(hdcp->value == DRM_MODE_CONTENT_PROTECTION_ENABLED); hdcp->content_type = content_type; + if (INTEL_GEN(dev_priv) >= 12) { + hdcp->cpu_transcoder = cpu_transcoder; + hdcp->port_data.fw_tc = intel_get_mei_fw_tc(cpu_transcoder); + } + /* * Considering that HDCP2.2 is more secure than HDCP1.4, If the setup * is capable of HDCP2.2, it is preferred to use HDCP2.2. diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.h b/drivers/gpu/drm/i915/display/intel_hdcp.h index 41c1053d9e38..f3c3272e712a 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.h +++ b/drivers/gpu/drm/i915/display/intel_hdcp.h @@ -21,11 +21,10 @@ enum transcoder; void intel_hdcp_atomic_check(struct drm_connector *connector, struct drm_connector_state *old_state, struct drm_connector_state *new_state); -void intel_hdcp_transcoder_config(struct intel_connector *connector, - enum transcoder cpu_transcoder); int intel_hdcp_init(struct intel_connector *connector, const struct intel_hdcp_shim *hdcp_shim); -int intel_hdcp_enable(struct intel_connector *connector, u8 content_type); +int intel_hdcp_enable(struct intel_connector *connector, + enum transcoder cpu_transcoder, u8 content_type); int intel_hdcp_disable(struct intel_connector *connector); bool is_hdcp_supported(struct drm_i915_private *dev_priv, enum port port); bool intel_hdcp_capable(struct intel_connector *connector); diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c index f6f5312205c4..f56fffc474fa 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.c +++ b/drivers/gpu/drm/i915/display/intel_hdmi.c @@ -2489,9 +2489,6 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder, return -EINVAL; } - intel_hdcp_transcoder_config(intel_hdmi->attached_connector, - pipe_config->cpu_transcoder); - return 0; } diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c index 848ce07a8ec2..8a98a1aa7adc 100644 --- a/drivers/gpu/drm/i915/display/intel_overlay.c +++ b/drivers/gpu/drm/i915/display/intel_overlay.c @@ -279,12 +279,21 @@ static void intel_overlay_flip_prepare(struct intel_overlay *overlay, struct i915_vma *vma) { enum pipe pipe = overlay->crtc->pipe; + struct intel_frontbuffer *from = NULL, *to = NULL; WARN_ON(overlay->old_vma); - intel_frontbuffer_track(overlay->vma ? overlay->vma->obj->frontbuffer : NULL, - vma ? vma->obj->frontbuffer : NULL, - INTEL_FRONTBUFFER_OVERLAY(pipe)); + if (overlay->vma) + from = intel_frontbuffer_get(overlay->vma->obj); + if (vma) + to = intel_frontbuffer_get(vma->obj); + + intel_frontbuffer_track(from, to, INTEL_FRONTBUFFER_OVERLAY(pipe)); + + if (to) + intel_frontbuffer_put(to); + if (from) + intel_frontbuffer_put(from); intel_frontbuffer_flip_prepare(overlay->i915, INTEL_FRONTBUFFER_OVERLAY(pipe)); @@ -766,7 +775,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay, ret = PTR_ERR(vma); goto out_pin_section; } - intel_frontbuffer_flush(new_bo->frontbuffer, ORIGIN_DIRTYFB); + i915_gem_object_flush_frontbuffer(new_bo, ORIGIN_DIRTYFB); if (!overlay->active) { u32 oconfig; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_busy.c b/drivers/gpu/drm/i915/gem/i915_gem_busy.c index 3d4f5775a4ba..25235ef630c1 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_busy.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_busy.c @@ -9,16 +9,16 @@ #include "i915_gem_ioctls.h" #include "i915_gem_object.h" -static __always_inline u32 __busy_read_flag(u8 id) +static __always_inline u32 __busy_read_flag(u16 id) { - if (id == (u8)I915_ENGINE_CLASS_INVALID) + if (id == (u16)I915_ENGINE_CLASS_INVALID) return 0xffff0000u; GEM_BUG_ON(id >= 16); return 0x10000u << id; } -static __always_inline u32 __busy_write_id(u8 id) +static __always_inline u32 __busy_write_id(u16 id) { /* * The uABI guarantees an active writer is also amongst the read @@ -29,14 +29,14 @@ static __always_inline u32 __busy_write_id(u8 id) * last_read - hence we always set both read and write busy for * last_write. */ - if (id == (u8)I915_ENGINE_CLASS_INVALID) + if (id == (u16)I915_ENGINE_CLASS_INVALID) return 0xffffffffu; return (id + 1) | __busy_read_flag(id); } static __always_inline unsigned int -__busy_set_if_active(const struct dma_fence *fence, u32 (*flag)(u8 id)) +__busy_set_if_active(const struct dma_fence *fence, u32 (*flag)(u16 id)) { const struct i915_request *rq; @@ -57,7 +57,7 @@ __busy_set_if_active(const struct dma_fence *fence, u32 (*flag)(u8 id)) return 0; /* Beware type-expansion follies! */ - BUILD_BUG_ON(!typecheck(u8, rq->engine->uabi_class)); + BUILD_BUG_ON(!typecheck(u16, rq->engine->uabi_class)); return flag(rq->engine->uabi_class); } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c index b9f504ba3b32..18ee708585a9 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_clflush.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_clflush.c @@ -20,7 +20,8 @@ static void __do_clflush(struct drm_i915_gem_object *obj) { GEM_BUG_ON(!i915_gem_object_has_pages(obj)); drm_clflush_sg(obj->mm.pages); - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_CPU); + + i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); } static int clflush_work(struct dma_fence_work *base) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c index 337ba17b1e0e..42385277c684 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_context.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c @@ -2167,8 +2167,7 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data, ext_data.fpriv = file->driver_priv; if (client_is_banned(ext_data.fpriv)) { DRM_DEBUG("client %s[%d] banned from creating ctx\n", - current->comm, - pid_nr(get_task_pid(current, PIDTYPE_PID))); + current->comm, task_pid_nr(current)); return -EIO; } diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c index 9937b4c341f1..f86400a191b0 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c @@ -664,7 +664,7 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data, i915_gem_object_unlock(obj); if (write_domain) - intel_frontbuffer_invalidate(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_invalidate_frontbuffer(obj, ORIGIN_CPU); out_unpin: i915_gem_object_unpin_pages(obj); @@ -784,7 +784,7 @@ int i915_gem_object_prepare_write(struct drm_i915_gem_object *obj, } out: - intel_frontbuffer_invalidate(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_invalidate_frontbuffer(obj, ORIGIN_CPU); obj->mm.dirty = true; /* return with the pages pinned */ return 0; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index f0998f1225af..bc3a67226163 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -2694,6 +2694,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, err = eb_submit(&eb); err_request: add_to_client(eb.request, file); + i915_request_get(eb.request); i915_request_add(eb.request); if (fences) @@ -2709,6 +2710,7 @@ err_request: fput(out_fence->file); } } + i915_request_put(eb.request); err_batch_unpin: if (eb.batch_flags & I915_DISPATCH_SECURE) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index a50296cce0d8..a596548c07bf 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -280,7 +280,7 @@ i915_gem_object_flush_write_domain(struct drm_i915_gem_object *obj, for_each_ggtt_vma(vma, obj) intel_gt_flush_ggtt_writes(vma->vm->gt); - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); for_each_ggtt_vma(vma, obj) { if (vma->iomap) @@ -308,6 +308,30 @@ i915_gem_object_flush_write_domain(struct drm_i915_gem_object *obj, obj->write_domain = 0; } +void __i915_gem_object_flush_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin) +{ + struct intel_frontbuffer *front; + + front = __intel_frontbuffer_get(obj); + if (front) { + intel_frontbuffer_flush(front, origin); + intel_frontbuffer_put(front); + } +} + +void __i915_gem_object_invalidate_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin) +{ + struct intel_frontbuffer *front; + + front = __intel_frontbuffer_get(obj); + if (front) { + intel_frontbuffer_invalidate(front, origin); + intel_frontbuffer_put(front); + } +} + void i915_gem_init__objects(struct drm_i915_private *i915) { INIT_WORK(&i915->mm.free_work, __i915_gem_free_work); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h index 458cd51331f1..4b93591fd5c7 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h @@ -13,8 +13,8 @@ #include <drm/i915_drm.h> +#include "display/intel_frontbuffer.h" #include "i915_gem_object_types.h" - #include "i915_gem_gtt.h" void i915_gem_init__objects(struct drm_i915_private *i915); @@ -463,4 +463,25 @@ int i915_gem_object_wait_priority(struct drm_i915_gem_object *obj, unsigned int flags, const struct i915_sched_attr *attr); +void __i915_gem_object_flush_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin); +void __i915_gem_object_invalidate_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin); + +static inline void +i915_gem_object_flush_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin) +{ + if (unlikely(rcu_access_pointer(obj->frontbuffer))) + __i915_gem_object_flush_frontbuffer(obj, origin); +} + +static inline void +i915_gem_object_invalidate_frontbuffer(struct drm_i915_gem_object *obj, + enum fb_op_origin origin) +{ + if (unlikely(rcu_access_pointer(obj->frontbuffer))) + __i915_gem_object_invalidate_frontbuffer(obj, origin); +} + #endif diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h index 96008374a412..e3f3944fbd90 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h @@ -150,7 +150,7 @@ struct drm_i915_gem_object { */ u16 write_domain; - struct intel_frontbuffer *frontbuffer; + struct intel_frontbuffer __rcu *frontbuffer; /** Current tiling stride for the object, if it's tiled. */ unsigned int tiling_and_stride; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c index 4c72d74d6576..0dbb44d30885 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c @@ -402,7 +402,7 @@ struct get_pages_work { static struct sg_table * __i915_gem_userptr_alloc_pages(struct drm_i915_gem_object *obj, - struct page **pvec, int num_pages) + struct page **pvec, unsigned long num_pages) { unsigned int max_segment = i915_sg_segment_size(); struct sg_table *st; @@ -448,9 +448,10 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work) { struct get_pages_work *work = container_of(_work, typeof(*work), work); struct drm_i915_gem_object *obj = work->obj; - const int npages = obj->base.size >> PAGE_SHIFT; + const unsigned long npages = obj->base.size >> PAGE_SHIFT; + unsigned long pinned; struct page **pvec; - int pinned, ret; + int ret; ret = -ENOMEM; pinned = 0; @@ -553,7 +554,7 @@ __i915_gem_userptr_get_pages_schedule(struct drm_i915_gem_object *obj) static int i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj) { - const int num_pages = obj->base.size >> PAGE_SHIFT; + const unsigned long num_pages = obj->base.size >> PAGE_SHIFT; struct mm_struct *mm = obj->userptr.mm->mm; struct page **pvec; struct sg_table *pages; diff --git a/drivers/gpu/drm/i915/gt/intel_context.c b/drivers/gpu/drm/i915/gt/intel_context.c index ef7bc41ffffa..5b7ff3ccfa8e 100644 --- a/drivers/gpu/drm/i915/gt/intel_context.c +++ b/drivers/gpu/drm/i915/gt/intel_context.c @@ -123,6 +123,10 @@ static int __context_pin_state(struct i915_vma *vma) if (err) return err; + err = i915_active_acquire(&vma->active); + if (err) + goto err_unpin; + /* * And mark it as a globally pinned object to let the shrinker know * it cannot reclaim the object until we release it. @@ -131,14 +135,44 @@ static int __context_pin_state(struct i915_vma *vma) vma->obj->mm.dirty = true; return 0; + +err_unpin: + i915_vma_unpin(vma); + return err; } static void __context_unpin_state(struct i915_vma *vma) { i915_vma_make_shrinkable(vma); + i915_active_release(&vma->active); __i915_vma_unpin(vma); } +static int __ring_active(struct intel_ring *ring) +{ + int err; + + err = i915_active_acquire(&ring->vma->active); + if (err) + return err; + + err = intel_ring_pin(ring); + if (err) + goto err_active; + + return 0; + +err_active: + i915_active_release(&ring->vma->active); + return err; +} + +static void __ring_retire(struct intel_ring *ring) +{ + intel_ring_unpin(ring); + i915_active_release(&ring->vma->active); +} + __i915_active_call static void __intel_context_retire(struct i915_active *active) { @@ -151,7 +185,7 @@ static void __intel_context_retire(struct i915_active *active) __context_unpin_state(ce->state); intel_timeline_unpin(ce->timeline); - intel_ring_unpin(ce->ring); + __ring_retire(ce->ring); intel_context_put(ce); } @@ -163,7 +197,7 @@ static int __intel_context_active(struct i915_active *active) intel_context_get(ce); - err = intel_ring_pin(ce->ring); + err = __ring_active(ce->ring); if (err) goto err_put; @@ -183,7 +217,7 @@ static int __intel_context_active(struct i915_active *active) err_timeline: intel_timeline_unpin(ce->timeline); err_ring: - intel_ring_unpin(ce->ring); + __ring_retire(ce->ring); err_put: intel_context_put(ce); return err; diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h index 17f1f1441efc..2b446474e010 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_types.h +++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h @@ -274,8 +274,8 @@ struct intel_engine_cs { u8 class; u8 instance; - u8 uabi_class; - u8 uabi_instance; + u16 uabi_class; + u16 uabi_instance; u32 uabi_capabilities; u32 context_size; diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm.c b/drivers/gpu/drm/i915/gt/intel_gt_pm.c index a459a42ad5c2..7e64b7d7d330 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_pm.c @@ -94,8 +94,9 @@ static int __gt_park(struct intel_wakeref *wf) intel_uncore_forcewake_put(&i915->uncore, FORCEWAKE_ALL); } + /* Defer dropping the display power well for 100ms, it's slow! */ GEM_BUG_ON(!wakeref); - intel_display_power_put(i915, POWER_DOMAIN_GT_IRQ, wakeref); + intel_display_power_put_async(i915, POWER_DOMAIN_GT_IRQ, wakeref); i915_globals_park(); diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c index 9fdefbdc3546..d925a1035c9d 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc.c +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c @@ -845,12 +845,6 @@ static const u8 *reg_offsets(const struct intel_engine_cs *engine) } } -static void unwind_wa_tail(struct i915_request *rq) -{ - rq->tail = intel_ring_wrap(rq->ring, rq->wa_tail - WA_TAIL_BYTES); - assert_ring_tail_valid(rq->ring, rq->tail); -} - static struct i915_request * __unwind_incomplete_requests(struct intel_engine_cs *engine) { @@ -863,12 +857,10 @@ __unwind_incomplete_requests(struct intel_engine_cs *engine) list_for_each_entry_safe_reverse(rq, rn, &engine->active.requests, sched.link) { - if (i915_request_completed(rq)) continue; /* XXX */ __i915_request_unsubmit(rq); - unwind_wa_tail(rq); /* * Push the request back into the queue for later resubmission. @@ -1161,13 +1153,29 @@ execlists_schedule_out(struct i915_request *rq) i915_request_put(rq); } -static u64 execlists_update_context(const struct i915_request *rq) +static u64 execlists_update_context(struct i915_request *rq) { struct intel_context *ce = rq->hw_context; - u64 desc; + u64 desc = ce->lrc_desc; + u32 tail; - ce->lrc_reg_state[CTX_RING_TAIL] = - intel_ring_set_tail(rq->ring, rq->tail); + /* + * WaIdleLiteRestore:bdw,skl + * + * We should never submit the context with the same RING_TAIL twice + * just in case we submit an empty ring, which confuses the HW. + * + * We append a couple of NOOPs (gen8_emit_wa_tail) after the end of + * the normal request to be able to always advance the RING_TAIL on + * subsequent resubmissions (for lite restore). Should that fail us, + * and we try and submit the same tail again, force the context + * reload. + */ + tail = intel_ring_set_tail(rq->ring, rq->tail); + if (unlikely(ce->lrc_reg_state[CTX_RING_TAIL] == tail)) + desc |= CTX_DESC_FORCE_RESTORE; + ce->lrc_reg_state[CTX_RING_TAIL] = tail; + rq->tail = rq->wa_tail; /* * Make sure the context image is complete before we submit it to HW. @@ -1186,13 +1194,11 @@ static u64 execlists_update_context(const struct i915_request *rq) */ mb(); - desc = ce->lrc_desc; - ce->lrc_desc &= ~CTX_DESC_FORCE_RESTORE; - /* Wa_1607138340:tgl */ if (IS_TGL_REVID(rq->i915, TGL_REVID_A0, TGL_REVID_A0)) desc |= CTX_DESC_FORCE_RESTORE; + ce->lrc_desc &= ~CTX_DESC_FORCE_RESTORE; return desc; } @@ -1703,16 +1709,6 @@ static void execlists_dequeue(struct intel_engine_cs *engine) return; } - - /* - * WaIdleLiteRestore:bdw,skl - * Apply the wa NOOPs to prevent - * ring:HEAD == rq:TAIL as we resubmit the - * request. See gen8_emit_fini_breadcrumb() for - * where we prepare the padding after the - * end of the request. - */ - last->tail = last->wa_tail; } } @@ -2668,6 +2664,14 @@ static u32 *gen9_init_indirectctx_bb(struct intel_engine_cs *engine, u32 *batch) /* WaFlushCoherentL3CacheLinesAtContextSwitch:skl,bxt,glk */ batch = gen8_emit_flush_coherentl3_wa(engine, batch); + /* WaClearSlmSpaceAtContextSwitch:skl,bxt,kbl,glk,cfl */ + batch = gen8_emit_pipe_control(batch, + PIPE_CONTROL_FLUSH_L3 | + PIPE_CONTROL_STORE_DATA_INDEX | + PIPE_CONTROL_CS_STALL | + PIPE_CONTROL_QW_WRITE, + LRC_PPHWSP_SCRATCH_ADDR); + batch = emit_lri(batch, lri, ARRAY_SIZE(lri)); /* WaMediaPoolStateCmdInWABB:bxt,glk */ @@ -4120,17 +4124,18 @@ static void virtual_context_destroy(struct kref *kref) for (n = 0; n < ve->num_siblings; n++) { struct intel_engine_cs *sibling = ve->siblings[n]; struct rb_node *node = &ve->nodes[sibling->id].rb; + unsigned long flags; if (RB_EMPTY_NODE(node)) continue; - spin_lock_irq(&sibling->active.lock); + spin_lock_irqsave(&sibling->active.lock, flags); /* Detachment is lazily performed in the execlists tasklet */ if (!RB_EMPTY_NODE(node)) rb_erase_cached(node, &sibling->execlists.virtual); - spin_unlock_irq(&sibling->active.lock); + spin_unlock_irqrestore(&sibling->active.lock, flags); } GEM_BUG_ON(__tasklet_is_scheduled(&ve->base.execlists.tasklet)); @@ -4419,9 +4424,11 @@ intel_execlists_create_virtual(struct i915_gem_context *ctx, ve->base.gt = siblings[0]->gt; ve->base.uncore = siblings[0]->uncore; ve->base.id = -1; + ve->base.class = OTHER_CLASS; ve->base.uabi_class = I915_ENGINE_CLASS_INVALID; ve->base.instance = I915_ENGINE_CLASS_INVALID_VIRTUAL; + ve->base.uabi_instance = I915_ENGINE_CLASS_INVALID_VIRTUAL; /* * The decision on whether to submit a request using semaphores diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c index a47d5a7c32c9..93026217c121 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c +++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c @@ -1413,14 +1413,6 @@ static inline int mi_set_context(struct i915_request *rq, u32 flags) int len; u32 *cs; - flags |= MI_MM_SPACE_GTT; - if (IS_HASWELL(i915)) - /* These flags are for resource streamer on HSW+ */ - flags |= HSW_MI_RS_SAVE_STATE_EN | HSW_MI_RS_RESTORE_STATE_EN; - else - /* We need to save the extended state for powersaving modes */ - flags |= MI_SAVE_EXT_STATE_EN | MI_RESTORE_EXT_STATE_EN; - len = 4; if (IS_GEN(i915, 7)) len += 2 + (num_engines ? 4 * num_engines + 6 : 0); @@ -1589,22 +1581,21 @@ static int switch_context(struct i915_request *rq) } if (ce->state) { - u32 hw_flags; + u32 flags; GEM_BUG_ON(rq->engine->id != RCS0); - /* - * The kernel context(s) is treated as pure scratch and is not - * expected to retain any state (as we sacrifice it during - * suspend and on resume it may be corrupted). This is ok, - * as nothing actually executes using the kernel context; it - * is purely used for flushing user contexts. - */ - hw_flags = 0; - if (i915_gem_context_is_kernel(rq->gem_context)) - hw_flags = MI_RESTORE_INHIBIT; + /* For resource streamer on HSW+ and power context elsewhere */ + BUILD_BUG_ON(HSW_MI_RS_SAVE_STATE_EN != MI_SAVE_EXT_STATE_EN); + BUILD_BUG_ON(HSW_MI_RS_RESTORE_STATE_EN != MI_RESTORE_EXT_STATE_EN); + + flags = MI_SAVE_EXT_STATE_EN | MI_MM_SPACE_GTT; + if (!i915_gem_context_is_kernel(rq->gem_context)) + flags |= MI_RESTORE_EXT_STATE_EN; + else + flags |= MI_RESTORE_INHIBIT; - ret = mi_set_context(rq, hw_flags); + ret = mi_set_context(rq, flags); if (ret) return ret; } diff --git a/drivers/gpu/drm/i915/gvt/dmabuf.c b/drivers/gpu/drm/i915/gvt/dmabuf.c index e451298d11c3..2477a1e5a166 100644 --- a/drivers/gpu/drm/i915/gvt/dmabuf.c +++ b/drivers/gpu/drm/i915/gvt/dmabuf.c @@ -36,13 +36,32 @@ #define GEN8_DECODE_PTE(pte) (pte & GENMASK_ULL(63, 12)) +static int vgpu_pin_dma_address(struct intel_vgpu *vgpu, + unsigned long size, + dma_addr_t dma_addr) +{ + int ret = 0; + + if (intel_gvt_hypervisor_dma_pin_guest_page(vgpu, dma_addr)) + ret = -EINVAL; + + return ret; +} + +static void vgpu_unpin_dma_address(struct intel_vgpu *vgpu, + dma_addr_t dma_addr) +{ + intel_gvt_hypervisor_dma_unmap_guest_page(vgpu, dma_addr); +} + static int vgpu_gem_get_pages( struct drm_i915_gem_object *obj) { struct drm_i915_private *dev_priv = to_i915(obj->base.dev); + struct intel_vgpu *vgpu; struct sg_table *st; struct scatterlist *sg; - int i, ret; + int i, j, ret; gen8_pte_t __iomem *gtt_entries; struct intel_vgpu_fb_info *fb_info; u32 page_num; @@ -51,6 +70,10 @@ static int vgpu_gem_get_pages( if (WARN_ON(!fb_info)) return -ENODEV; + vgpu = fb_info->obj->vgpu; + if (WARN_ON(!vgpu)) + return -ENODEV; + st = kmalloc(sizeof(*st), GFP_KERNEL); if (unlikely(!st)) return -ENOMEM; @@ -64,21 +87,53 @@ static int vgpu_gem_get_pages( gtt_entries = (gen8_pte_t __iomem *)dev_priv->ggtt.gsm + (fb_info->start >> PAGE_SHIFT); for_each_sg(st->sgl, sg, page_num, i) { + dma_addr_t dma_addr = + GEN8_DECODE_PTE(readq(>t_entries[i])); + if (vgpu_pin_dma_address(vgpu, PAGE_SIZE, dma_addr)) { + ret = -EINVAL; + goto out; + } + sg->offset = 0; sg->length = PAGE_SIZE; - sg_dma_address(sg) = - GEN8_DECODE_PTE(readq(>t_entries[i])); sg_dma_len(sg) = PAGE_SIZE; + sg_dma_address(sg) = dma_addr; } __i915_gem_object_set_pages(obj, st, PAGE_SIZE); +out: + if (ret) { + dma_addr_t dma_addr; + + for_each_sg(st->sgl, sg, i, j) { + dma_addr = sg_dma_address(sg); + if (dma_addr) + vgpu_unpin_dma_address(vgpu, dma_addr); + } + sg_free_table(st); + kfree(st); + } + + return ret; - return 0; } static void vgpu_gem_put_pages(struct drm_i915_gem_object *obj, struct sg_table *pages) { + struct scatterlist *sg; + + if (obj->base.dma_buf) { + struct intel_vgpu_fb_info *fb_info = obj->gvt_info; + struct intel_vgpu_dmabuf_obj *obj = fb_info->obj; + struct intel_vgpu *vgpu = obj->vgpu; + int i; + + for_each_sg(pages->sgl, sg, fb_info->size, i) + vgpu_unpin_dma_address(vgpu, + sg_dma_address(sg)); + } + sg_free_table(pages); kfree(pages); } @@ -163,6 +218,7 @@ static struct drm_i915_gem_object *vgpu_create_gem(struct drm_device *dev, drm_gem_private_object_init(dev, &obj->base, roundup(info->size, PAGE_SIZE)); i915_gem_object_init(obj, &intel_vgpu_gem_ops, &lock_class); + i915_gem_object_set_readonly(obj); obj->read_domains = I915_GEM_DOMAIN_GTT; obj->write_domain = 0; diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index bb9fe6bf5275..1043e6d564df 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -341,6 +341,10 @@ static int gdrst_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, gvt_dbg_mmio("vgpu%d: request VCS2 Reset\n", vgpu->id); engine_mask |= BIT(VCS1); } + if (data & GEN9_GRDOM_GUC) { + gvt_dbg_mmio("vgpu%d: request GUC Reset\n", vgpu->id); + vgpu_vreg_t(vgpu, GUC_STATUS) |= GS_MIA_IN_RESET; + } engine_mask &= INTEL_INFO(vgpu->gvt->dev_priv)->engine_mask; } @@ -1636,6 +1640,16 @@ static int edp_psr_imr_iir_write(struct intel_vgpu *vgpu, return 0; } +static int guc_status_read(struct intel_vgpu *vgpu, + unsigned int offset, void *p_data, + unsigned int bytes) +{ + /* keep MIA_IN_RESET before clearing */ + read_vreg(vgpu, offset, p_data, bytes); + vgpu_vreg(vgpu, offset) &= ~GS_MIA_IN_RESET; + return 0; +} + static int mmio_read_from_hw(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { @@ -2672,6 +2686,8 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_DH(EDP_PSR_IMR, D_BDW_PLUS, NULL, edp_psr_imr_iir_write); MMIO_DH(EDP_PSR_IIR, D_BDW_PLUS, NULL, edp_psr_imr_iir_write); + MMIO_DH(GUC_STATUS, D_ALL, guc_status_read, NULL); + return 0; } diff --git a/drivers/gpu/drm/i915/gvt/hypercall.h b/drivers/gpu/drm/i915/gvt/hypercall.h index 4862fb12778e..b19a3b1ea4c1 100644 --- a/drivers/gpu/drm/i915/gvt/hypercall.h +++ b/drivers/gpu/drm/i915/gvt/hypercall.h @@ -62,6 +62,8 @@ struct intel_gvt_mpt { unsigned long size, dma_addr_t *dma_addr); void (*dma_unmap_guest_page)(unsigned long handle, dma_addr_t dma_addr); + int (*dma_pin_guest_page)(unsigned long handle, dma_addr_t dma_addr); + int (*map_gfn_to_mfn)(unsigned long handle, unsigned long gfn, unsigned long mfn, unsigned int nr, bool map); int (*set_trap_area)(unsigned long handle, u64 start, u64 end, diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index 04a5a0d90823..3259a1fa69e1 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -1916,6 +1916,28 @@ err_unlock: return ret; } +static int kvmgt_dma_pin_guest_page(unsigned long handle, dma_addr_t dma_addr) +{ + struct kvmgt_guest_info *info; + struct gvt_dma *entry; + int ret = 0; + + if (!handle_valid(handle)) + return -ENODEV; + + info = (struct kvmgt_guest_info *)handle; + + mutex_lock(&info->vgpu->vdev.cache_lock); + entry = __gvt_cache_find_dma_addr(info->vgpu, dma_addr); + if (entry) + kref_get(&entry->ref); + else + ret = -ENOMEM; + mutex_unlock(&info->vgpu->vdev.cache_lock); + + return ret; +} + static void __gvt_dma_release(struct kref *ref) { struct gvt_dma *entry = container_of(ref, typeof(*entry), ref); @@ -2027,6 +2049,7 @@ static struct intel_gvt_mpt kvmgt_mpt = { .gfn_to_mfn = kvmgt_gfn_to_pfn, .dma_map_guest_page = kvmgt_dma_map_guest_page, .dma_unmap_guest_page = kvmgt_dma_unmap_guest_page, + .dma_pin_guest_page = kvmgt_dma_pin_guest_page, .set_opregion = kvmgt_set_opregion, .set_edid = kvmgt_set_edid, .get_vfio_device = kvmgt_get_vfio_device, diff --git a/drivers/gpu/drm/i915/gvt/mpt.h b/drivers/gpu/drm/i915/gvt/mpt.h index 0f9440128123..9ad224df9c68 100644 --- a/drivers/gpu/drm/i915/gvt/mpt.h +++ b/drivers/gpu/drm/i915/gvt/mpt.h @@ -255,6 +255,21 @@ static inline void intel_gvt_hypervisor_dma_unmap_guest_page( } /** + * intel_gvt_hypervisor_dma_pin_guest_page - pin guest dma buf + * @vgpu: a vGPU + * @dma_addr: guest dma addr + * + * Returns: + * 0 on success, negative error code if failed. + */ +static inline int +intel_gvt_hypervisor_dma_pin_guest_page(struct intel_vgpu *vgpu, + dma_addr_t dma_addr) +{ + return intel_gvt_host.mpt->dma_pin_guest_page(vgpu->handle, dma_addr); +} + +/** * intel_gvt_hypervisor_map_gfn_to_mfn - map a GFN region to MFN * @vgpu: a vGPU * @gfn: guest PFN diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c index d5a6e4e3d0fd..85bd9bf4f6ee 100644 --- a/drivers/gpu/drm/i915/gvt/vgpu.c +++ b/drivers/gpu/drm/i915/gvt/vgpu.c @@ -212,9 +212,9 @@ static void intel_gvt_update_vgpu_types(struct intel_gvt *gvt) */ void intel_gvt_activate_vgpu(struct intel_vgpu *vgpu) { - mutex_lock(&vgpu->gvt->lock); + mutex_lock(&vgpu->vgpu_lock); vgpu->active = true; - mutex_unlock(&vgpu->gvt->lock); + mutex_unlock(&vgpu->vgpu_lock); } /** diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index e29bc137e7ba..21aa08f55811 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1660,8 +1660,10 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, (IS_BROADWELL(dev_priv) || IS_GEN(dev_priv, 9)) /* WaRsDisableCoarsePowerGating:skl,cnl */ -#define NEEDS_WaRsDisableCoarsePowerGating(dev_priv) \ - (IS_CANNONLAKE(dev_priv) || IS_GEN(dev_priv, 9)) +#define NEEDS_WaRsDisableCoarsePowerGating(dev_priv) \ + (IS_CANNONLAKE(dev_priv) || \ + IS_SKL_GT3(dev_priv) || \ + IS_SKL_GT4(dev_priv)) #define HAS_GMBUS_IRQ(dev_priv) (INTEL_GEN(dev_priv) >= 4) #define HAS_GMBUS_BURST_READ(dev_priv) (INTEL_GEN(dev_priv) >= 10 || \ diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index b9eb6b3149b7..905890e3ac24 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -45,6 +45,7 @@ #include "gem/i915_gem_context.h" #include "gem/i915_gem_ioctls.h" #include "gem/i915_gem_pm.h" +#include "gt/intel_context.h" #include "gt/intel_engine_user.h" #include "gt/intel_gt.h" #include "gt/intel_gt_pm.h" @@ -160,7 +161,7 @@ i915_gem_phys_pwrite(struct drm_i915_gem_object *obj, * We manually control the domain here and pretend that it * remains coherent i.e. in the GTT domain, like shmem_pwrite. */ - intel_frontbuffer_invalidate(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_invalidate_frontbuffer(obj, ORIGIN_CPU); if (copy_from_user(vaddr, user_data, args->size)) return -EFAULT; @@ -168,7 +169,7 @@ i915_gem_phys_pwrite(struct drm_i915_gem_object *obj, drm_clflush_virt_range(vaddr, args->size); intel_gt_chipset_flush(&to_i915(obj->base.dev)->gt); - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); return 0; } @@ -588,7 +589,7 @@ i915_gem_gtt_pwrite_fast(struct drm_i915_gem_object *obj, goto out_unpin; } - intel_frontbuffer_invalidate(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_invalidate_frontbuffer(obj, ORIGIN_CPU); user_data = u64_to_user_ptr(args->data_ptr); offset = args->offset; @@ -630,7 +631,7 @@ i915_gem_gtt_pwrite_fast(struct drm_i915_gem_object *obj, user_data += page_length; offset += page_length; } - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); i915_gem_object_unlock_fence(obj, fence); out_unpin: @@ -720,7 +721,7 @@ i915_gem_shmem_pwrite(struct drm_i915_gem_object *obj, offset = 0; } - intel_frontbuffer_flush(obj->frontbuffer, ORIGIN_CPU); + i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); i915_gem_object_unlock_fence(obj, fence); return ret; @@ -1053,6 +1054,18 @@ out: return err; } +static int __intel_context_flush_retire(struct intel_context *ce) +{ + struct intel_timeline *tl; + + tl = intel_context_timeline_lock(ce); + if (IS_ERR(tl)) + return PTR_ERR(tl); + + intel_context_timeline_unlock(tl); + return 0; +} + static int __intel_engines_record_defaults(struct intel_gt *gt) { struct i915_request *requests[I915_NUM_ENGINES] = {}; @@ -1121,13 +1134,20 @@ err_rq: if (!rq) continue; - /* We want to be able to unbind the state from the GGTT */ - GEM_BUG_ON(intel_context_is_pinned(rq->hw_context)); - + GEM_BUG_ON(!test_bit(CONTEXT_ALLOC_BIT, + &rq->hw_context->flags)); state = rq->hw_context->state; if (!state) continue; + /* Serialise with retirement on another CPU */ + err = __intel_context_flush_retire(rq->hw_context); + if (err) + goto out; + + /* We want to be able to unbind the state from the GGTT */ + GEM_BUG_ON(intel_context_is_pinned(rq->hw_context)); + /* * As we will hold a reference to the logical state, it will * not be torn down with the context, and importantly the diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 6239a9adbf14..d6ce57d30958 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -1177,6 +1177,7 @@ gen8_ppgtt_insert_pte(struct i915_ppgtt *ppgtt, pd = i915_pd_entry(pdp, gen8_pd_index(idx, 2)); vaddr = kmap_atomic_px(i915_pt_entry(pd, gen8_pd_index(idx, 1))); do { + GEM_BUG_ON(iter->sg->length < I915_GTT_PAGE_SIZE); vaddr[gen8_pd_index(idx, 0)] = pte_encode | iter->dma; iter->dma += I915_GTT_PAGE_SIZE; @@ -1660,6 +1661,7 @@ static void gen6_ppgtt_insert_entries(struct i915_address_space *vm, vaddr = kmap_atomic_px(i915_pt_entry(pd, act_pt)); do { + GEM_BUG_ON(iter.sg->length < I915_GTT_PAGE_SIZE); vaddr[act_pte] = pte_encode | GEN6_PTE_ADDR_ENCODE(iter.dma); iter.dma += I915_GTT_PAGE_SIZE; @@ -2847,7 +2849,7 @@ static int ggtt_probe_common(struct i915_ggtt *ggtt, u64 size) * readback check when writing GTT PTE entries. */ if (IS_GEN9_LP(dev_priv) || INTEL_GEN(dev_priv) >= 10) - ggtt->gsm = ioremap_nocache(phys_addr, size); + ggtt->gsm = ioremap(phys_addr, size); else ggtt->gsm = ioremap_wc(phys_addr, size); if (!ggtt->gsm) { @@ -3304,7 +3306,7 @@ void i915_ggtt_disable_guc(struct i915_ggtt *ggtt) static void ggtt_restore_mappings(struct i915_ggtt *ggtt) { - struct i915_vma *vma, *vn; + struct i915_vma *vma; bool flush = false; int open; @@ -3319,15 +3321,12 @@ static void ggtt_restore_mappings(struct i915_ggtt *ggtt) open = atomic_xchg(&ggtt->vm.open, 0); /* clflush objects bound into the GGTT and rebind them. */ - list_for_each_entry_safe(vma, vn, &ggtt->vm.bound_list, vm_link) { + list_for_each_entry(vma, &ggtt->vm.bound_list, vm_link) { struct drm_i915_gem_object *obj = vma->obj; if (!i915_vma_is_bound(vma, I915_VMA_GLOBAL_BIND)) continue; - if (!__i915_vma_unbind(vma)) - continue; - clear_bit(I915_VMA_GLOBAL_BIND_BIT, __i915_vma_flags(vma)); WARN_ON(i915_vma_bind(vma, obj ? obj->cache_level : 0, diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index 65d7c2e599de..2ae14bc14931 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -2078,20 +2078,12 @@ gen8_update_reg_state_unlocked(const struct intel_context *ce, u32 *reg_state = ce->lrc_reg_state; int i; - if (IS_GEN(stream->perf->i915, 12)) { - u32 format = stream->oa_buffer.format; + reg_state[ctx_oactxctrl + 1] = + (stream->period_exponent << GEN8_OA_TIMER_PERIOD_SHIFT) | + (stream->periodic ? GEN8_OA_TIMER_ENABLE : 0) | + GEN8_OA_COUNTER_RESUME; - reg_state[ctx_oactxctrl + 1] = - (format << GEN12_OAR_OACONTROL_COUNTER_FORMAT_SHIFT) | - (stream->oa_config ? GEN12_OAR_OACONTROL_COUNTER_ENABLE : 0); - } else { - reg_state[ctx_oactxctrl + 1] = - (stream->period_exponent << GEN8_OA_TIMER_PERIOD_SHIFT) | - (stream->periodic ? GEN8_OA_TIMER_ENABLE : 0) | - GEN8_OA_COUNTER_RESUME; - } - - for (i = 0; !!ctx_flexeu0 && i < ARRAY_SIZE(flex_regs); i++) + for (i = 0; i < ARRAY_SIZE(flex_regs); i++) reg_state[ctx_flexeu0 + i * 2 + 1] = oa_config_flex_reg(stream->oa_config, flex_regs[i]); @@ -2224,34 +2216,51 @@ static int gen8_configure_context(struct i915_gem_context *ctx, return err; } -static int gen12_emit_oar_config(struct intel_context *ce, bool enable) +static int gen12_configure_oar_context(struct i915_perf_stream *stream, bool enable) { - struct i915_request *rq; - u32 *cs; - int err = 0; - - rq = i915_request_create(ce); - if (IS_ERR(rq)) - return PTR_ERR(rq); - - cs = intel_ring_begin(rq, 4); - if (IS_ERR(cs)) { - err = PTR_ERR(cs); - goto out; - } - - *cs++ = MI_LOAD_REGISTER_IMM(1); - *cs++ = i915_mmio_reg_offset(RING_CONTEXT_CONTROL(ce->engine->mmio_base)); - *cs++ = _MASKED_FIELD(GEN12_CTX_CTRL_OAR_CONTEXT_ENABLE, - enable ? GEN12_CTX_CTRL_OAR_CONTEXT_ENABLE : 0); - *cs++ = MI_NOOP; + int err; + struct intel_context *ce = stream->pinned_ctx; + u32 format = stream->oa_buffer.format; + struct flex regs_context[] = { + { + GEN8_OACTXCONTROL, + stream->perf->ctx_oactxctrl_offset + 1, + enable ? GEN8_OA_COUNTER_RESUME : 0, + }, + }; + /* Offsets in regs_lri are not used since this configuration is only + * applied using LRI. Initialize the correct offsets for posterity. + */ +#define GEN12_OAR_OACONTROL_OFFSET 0x5B0 + struct flex regs_lri[] = { + { + GEN12_OAR_OACONTROL, + GEN12_OAR_OACONTROL_OFFSET + 1, + (format << GEN12_OAR_OACONTROL_COUNTER_FORMAT_SHIFT) | + (enable ? GEN12_OAR_OACONTROL_COUNTER_ENABLE : 0) + }, + { + RING_CONTEXT_CONTROL(ce->engine->mmio_base), + CTX_CONTEXT_CONTROL, + _MASKED_FIELD(GEN12_CTX_CTRL_OAR_CONTEXT_ENABLE, + enable ? + GEN12_CTX_CTRL_OAR_CONTEXT_ENABLE : + 0) + }, + }; - intel_ring_advance(rq, cs); + /* Modify the context image of pinned context with regs_context*/ + err = intel_context_lock_pinned(ce); + if (err) + return err; -out: - i915_request_add(rq); + err = gen8_modify_context(ce, regs_context, ARRAY_SIZE(regs_context)); + intel_context_unlock_pinned(ce); + if (err) + return err; - return err; + /* Apply regs_lri using LRI with pinned context */ + return gen8_modify_self(ce, regs_lri, ARRAY_SIZE(regs_lri)); } /* @@ -2277,53 +2286,16 @@ out: * per-context OA state. * * Note: it's only the RCS/Render context that has any OA state. + * Note: the first flex register passed must always be R_PWR_CLK_STATE */ -static int lrc_configure_all_contexts(struct i915_perf_stream *stream, - const struct i915_oa_config *oa_config) +static int oa_configure_all_contexts(struct i915_perf_stream *stream, + struct flex *regs, + size_t num_regs) { struct drm_i915_private *i915 = stream->perf->i915; - /* The MMIO offsets for Flex EU registers aren't contiguous */ - const u32 ctx_flexeu0 = stream->perf->ctx_flexeu0_offset; -#define ctx_flexeuN(N) (ctx_flexeu0 + 2 * (N) + 1) - struct flex regs[] = { - { - GEN8_R_PWR_CLK_STATE, - CTX_R_PWR_CLK_STATE, - }, - { - IS_GEN(i915, 12) ? - GEN12_OAR_OACONTROL : GEN8_OACTXCONTROL, - stream->perf->ctx_oactxctrl_offset + 1, - }, - { EU_PERF_CNTL0, ctx_flexeuN(0) }, - { EU_PERF_CNTL1, ctx_flexeuN(1) }, - { EU_PERF_CNTL2, ctx_flexeuN(2) }, - { EU_PERF_CNTL3, ctx_flexeuN(3) }, - { EU_PERF_CNTL4, ctx_flexeuN(4) }, - { EU_PERF_CNTL5, ctx_flexeuN(5) }, - { EU_PERF_CNTL6, ctx_flexeuN(6) }, - }; -#undef ctx_flexeuN struct intel_engine_cs *engine; struct i915_gem_context *ctx, *cn; - size_t array_size = IS_GEN(i915, 12) ? 2 : ARRAY_SIZE(regs); - int i, err; - - if (IS_GEN(i915, 12)) { - u32 format = stream->oa_buffer.format; - - regs[1].value = - (format << GEN12_OAR_OACONTROL_COUNTER_FORMAT_SHIFT) | - (oa_config ? GEN12_OAR_OACONTROL_COUNTER_ENABLE : 0); - } else { - regs[1].value = - (stream->period_exponent << GEN8_OA_TIMER_PERIOD_SHIFT) | - (stream->periodic ? GEN8_OA_TIMER_ENABLE : 0) | - GEN8_OA_COUNTER_RESUME; - } - - for (i = 2; !!ctx_flexeu0 && i < array_size; i++) - regs[i].value = oa_config_flex_reg(oa_config, regs[i].reg); + int err; lockdep_assert_held(&stream->perf->lock); @@ -2353,7 +2325,7 @@ static int lrc_configure_all_contexts(struct i915_perf_stream *stream, spin_unlock(&i915->gem.contexts.lock); - err = gen8_configure_context(ctx, regs, array_size); + err = gen8_configure_context(ctx, regs, num_regs); if (err) { i915_gem_context_put(ctx); return err; @@ -2378,7 +2350,7 @@ static int lrc_configure_all_contexts(struct i915_perf_stream *stream, regs[0].value = intel_sseu_make_rpcs(i915, &ce->sseu); - err = gen8_modify_self(ce, regs, array_size); + err = gen8_modify_self(ce, regs, num_regs); if (err) return err; } @@ -2386,6 +2358,56 @@ static int lrc_configure_all_contexts(struct i915_perf_stream *stream, return 0; } +static int gen12_configure_all_contexts(struct i915_perf_stream *stream, + const struct i915_oa_config *oa_config) +{ + struct flex regs[] = { + { + GEN8_R_PWR_CLK_STATE, + CTX_R_PWR_CLK_STATE, + }, + }; + + return oa_configure_all_contexts(stream, regs, ARRAY_SIZE(regs)); +} + +static int lrc_configure_all_contexts(struct i915_perf_stream *stream, + const struct i915_oa_config *oa_config) +{ + /* The MMIO offsets for Flex EU registers aren't contiguous */ + const u32 ctx_flexeu0 = stream->perf->ctx_flexeu0_offset; +#define ctx_flexeuN(N) (ctx_flexeu0 + 2 * (N) + 1) + struct flex regs[] = { + { + GEN8_R_PWR_CLK_STATE, + CTX_R_PWR_CLK_STATE, + }, + { + GEN8_OACTXCONTROL, + stream->perf->ctx_oactxctrl_offset + 1, + }, + { EU_PERF_CNTL0, ctx_flexeuN(0) }, + { EU_PERF_CNTL1, ctx_flexeuN(1) }, + { EU_PERF_CNTL2, ctx_flexeuN(2) }, + { EU_PERF_CNTL3, ctx_flexeuN(3) }, + { EU_PERF_CNTL4, ctx_flexeuN(4) }, + { EU_PERF_CNTL5, ctx_flexeuN(5) }, + { EU_PERF_CNTL6, ctx_flexeuN(6) }, + }; +#undef ctx_flexeuN + int i; + + regs[1].value = + (stream->period_exponent << GEN8_OA_TIMER_PERIOD_SHIFT) | + (stream->periodic ? GEN8_OA_TIMER_ENABLE : 0) | + GEN8_OA_COUNTER_RESUME; + + for (i = 2; i < ARRAY_SIZE(regs); i++) + regs[i].value = oa_config_flex_reg(oa_config, regs[i].reg); + + return oa_configure_all_contexts(stream, regs, ARRAY_SIZE(regs)); +} + static int gen8_enable_metric_set(struct i915_perf_stream *stream) { struct intel_uncore *uncore = stream->uncore; @@ -2464,7 +2486,7 @@ static int gen12_enable_metric_set(struct i915_perf_stream *stream) * to make sure all slices/subslices are ON before writing to NOA * registers. */ - ret = lrc_configure_all_contexts(stream, oa_config); + ret = gen12_configure_all_contexts(stream, oa_config); if (ret) return ret; @@ -2474,8 +2496,7 @@ static int gen12_enable_metric_set(struct i915_perf_stream *stream) * requested this. */ if (stream->ctx) { - ret = gen12_emit_oar_config(stream->pinned_ctx, - oa_config != NULL); + ret = gen12_configure_oar_context(stream, true); if (ret) return ret; } @@ -2509,11 +2530,11 @@ static void gen12_disable_metric_set(struct i915_perf_stream *stream) struct intel_uncore *uncore = stream->uncore; /* Reset all contexts' slices/subslices configurations. */ - lrc_configure_all_contexts(stream, NULL); + gen12_configure_all_contexts(stream, NULL); /* disable the context save/restore or OAR counters */ if (stream->ctx) - gen12_emit_oar_config(stream->pinned_ctx, false); + gen12_configure_oar_context(stream, false); /* Make sure we disable noa to save power. */ intel_uncore_rmw(uncore, RPM_CONFIG1, GEN10_GT_NOA_ENABLE, 0); @@ -2713,7 +2734,8 @@ static int i915_oa_stream_init(struct i915_perf_stream *stream, return -EINVAL; } - if (!(props->sample_flags & SAMPLE_OA_REPORT)) { + if (!(props->sample_flags & SAMPLE_OA_REPORT) && + (INTEL_GEN(perf->i915) < 12 || !stream->ctx)) { DRM_DEBUG("Only OA report sampling supported\n"); return -EINVAL; } @@ -2745,7 +2767,7 @@ static int i915_oa_stream_init(struct i915_perf_stream *stream, format_size = perf->oa_formats[props->oa_format].size; - stream->sample_flags |= SAMPLE_OA_REPORT; + stream->sample_flags = props->sample_flags; stream->sample_size += format_size; stream->oa_buffer.format_size = format_size; @@ -2854,7 +2876,11 @@ void i915_oa_init_reg_state(const struct intel_context *ce, return; stream = engine->i915->perf.exclusive_stream; - if (stream) + /* + * For gen12, only CTX_R_PWR_CLK_STATE needs update, but the caller + * is already doing that, so nothing to be done for gen12 here. + */ + if (stream && INTEL_GEN(stream->perf->i915) < 12) gen8_update_reg_state_unlocked(ce, stream); } diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c index 2814218c5ba1..d6d2e6fb8674 100644 --- a/drivers/gpu/drm/i915/i915_pmu.c +++ b/drivers/gpu/drm/i915/i915_pmu.c @@ -144,61 +144,40 @@ static inline s64 ktime_since(const ktime_t kt) return ktime_to_ns(ktime_sub(ktime_get(), kt)); } -static u64 __pmu_estimate_rc6(struct i915_pmu *pmu) -{ - u64 val; - - /* - * We think we are runtime suspended. - * - * Report the delta from when the device was suspended to now, - * on top of the last known real value, as the approximated RC6 - * counter value. - */ - val = ktime_since(pmu->sleep_last); - val += pmu->sample[__I915_SAMPLE_RC6].cur; - - pmu->sample[__I915_SAMPLE_RC6_ESTIMATED].cur = val; - - return val; -} - -static u64 __pmu_update_rc6(struct i915_pmu *pmu, u64 val) -{ - /* - * If we are coming back from being runtime suspended we must - * be careful not to report a larger value than returned - * previously. - */ - if (val >= pmu->sample[__I915_SAMPLE_RC6_ESTIMATED].cur) { - pmu->sample[__I915_SAMPLE_RC6_ESTIMATED].cur = 0; - pmu->sample[__I915_SAMPLE_RC6].cur = val; - } else { - val = pmu->sample[__I915_SAMPLE_RC6_ESTIMATED].cur; - } - - return val; -} - static u64 get_rc6(struct intel_gt *gt) { struct drm_i915_private *i915 = gt->i915; struct i915_pmu *pmu = &i915->pmu; unsigned long flags; + bool awake = false; u64 val; - val = 0; if (intel_gt_pm_get_if_awake(gt)) { val = __get_rc6(gt); intel_gt_pm_put_async(gt); + awake = true; } spin_lock_irqsave(&pmu->lock, flags); - if (val) - val = __pmu_update_rc6(pmu, val); + if (awake) { + pmu->sample[__I915_SAMPLE_RC6].cur = val; + } else { + /* + * We think we are runtime suspended. + * + * Report the delta from when the device was suspended to now, + * on top of the last known real value, as the approximated RC6 + * counter value. + */ + val = ktime_since(pmu->sleep_last); + val += pmu->sample[__I915_SAMPLE_RC6].cur; + } + + if (val < pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur) + val = pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur; else - val = __pmu_estimate_rc6(pmu); + pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur = val; spin_unlock_irqrestore(&pmu->lock, flags); @@ -210,20 +189,11 @@ static void park_rc6(struct drm_i915_private *i915) struct i915_pmu *pmu = &i915->pmu; if (pmu->enable & config_enabled_mask(I915_PMU_RC6_RESIDENCY)) - __pmu_update_rc6(pmu, __get_rc6(&i915->gt)); + pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(&i915->gt); pmu->sleep_last = ktime_get(); } -static void unpark_rc6(struct drm_i915_private *i915) -{ - struct i915_pmu *pmu = &i915->pmu; - - /* Estimate how long we slept and accumulate that into rc6 counters */ - if (pmu->enable & config_enabled_mask(I915_PMU_RC6_RESIDENCY)) - __pmu_estimate_rc6(pmu); -} - #else static u64 get_rc6(struct intel_gt *gt) @@ -232,7 +202,6 @@ static u64 get_rc6(struct intel_gt *gt) } static void park_rc6(struct drm_i915_private *i915) {} -static void unpark_rc6(struct drm_i915_private *i915) {} #endif @@ -281,8 +250,6 @@ void i915_pmu_gt_unparked(struct drm_i915_private *i915) */ __i915_pmu_maybe_start_timer(pmu); - unpark_rc6(i915); - spin_unlock_irq(&pmu->lock); } @@ -1107,12 +1074,17 @@ void i915_pmu_register(struct drm_i915_private *i915) hrtimer_init(&pmu->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); pmu->timer.function = i915_sample; - if (!is_igp(i915)) + if (!is_igp(i915)) { pmu->name = kasprintf(GFP_KERNEL, - "i915-%s", + "i915_%s", dev_name(i915->drm.dev)); - else + if (pmu->name) { + /* tools/perf reserves colons as special. */ + strreplace((char *)pmu->name, ':', '_'); + } + } else { pmu->name = "i915"; + } if (!pmu->name) goto err; diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h index bf52e3983631..6c1647c5daf2 100644 --- a/drivers/gpu/drm/i915/i915_pmu.h +++ b/drivers/gpu/drm/i915/i915_pmu.h @@ -18,7 +18,7 @@ enum { __I915_SAMPLE_FREQ_ACT = 0, __I915_SAMPLE_FREQ_REQ, __I915_SAMPLE_RC6, - __I915_SAMPLE_RC6_ESTIMATED, + __I915_SAMPLE_RC6_LAST_REPORTED, __I915_NUM_PMU_SAMPLERS }; diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 73079b503724..094011b8f64d 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -4177,7 +4177,13 @@ enum { #define CPSSUNIT_CLKGATE_DIS REG_BIT(9) #define UNSLICE_UNIT_LEVEL_CLKGATE _MMIO(0x9434) -#define VFUNIT_CLKGATE_DIS (1 << 20) +#define VFUNIT_CLKGATE_DIS REG_BIT(20) +#define HSUNIT_CLKGATE_DIS REG_BIT(8) +#define VSUNIT_CLKGATE_DIS REG_BIT(3) + +#define UNSLICE_UNIT_LEVEL_CLKGATE2 _MMIO(0x94e4) +#define VSUNIT_CLKGATE_DIS_TGL REG_BIT(19) +#define PSDUNIT_CLKGATE_DIS REG_BIT(5) #define INF_UNIT_LEVEL_CLKGATE _MMIO(0x9560) #define CGPSF_CLKGATE_DIS (1 << 3) @@ -9405,11 +9411,9 @@ enum skl_power_gate { #define _ICL_AUX_REG_IDX(pw_idx) ((pw_idx) - ICL_PW_CTL_IDX_AUX_A) #define _ICL_AUX_ANAOVRD1_A 0x162398 #define _ICL_AUX_ANAOVRD1_B 0x6C398 -#define _TGL_AUX_ANAOVRD1_C 0x160398 #define ICL_AUX_ANAOVRD1(pw_idx) _MMIO(_PICK(_ICL_AUX_REG_IDX(pw_idx), \ _ICL_AUX_ANAOVRD1_A, \ - _ICL_AUX_ANAOVRD1_B, \ - _TGL_AUX_ANAOVRD1_C)) + _ICL_AUX_ANAOVRD1_B)) #define ICL_AUX_ANAOVRD1_LDO_BYPASS (1 << 7) #define ICL_AUX_ANAOVRD1_ENABLE (1 << 0) @@ -11994,7 +11998,7 @@ enum skl_power_gate { /* This register controls the Display State Buffer (DSB) engines. */ #define _DSBSL_INSTANCE_BASE 0x70B00 #define DSBSL_INSTANCE(pipe, id) (_DSBSL_INSTANCE_BASE + \ - (pipe) * 0x1000 + (id) * 100) + (pipe) * 0x1000 + (id) * 0x100) #define DSB_HEAD(pipe, id) _MMIO(DSBSL_INSTANCE(pipe, id) + 0x0) #define DSB_TAIL(pipe, id) _MMIO(DSBSL_INSTANCE(pipe, id) + 0x4) #define DSB_CTRL(pipe, id) _MMIO(DSBSL_INSTANCE(pipe, id) + 0x8) diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index bbd71af00a91..765bec89fc0d 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -300,11 +300,11 @@ void i915_request_retire_upto(struct i915_request *rq) } static int -__i915_request_await_execution(struct i915_request *rq, - struct i915_request *signal, - void (*hook)(struct i915_request *rq, - struct dma_fence *signal), - gfp_t gfp) +__await_execution(struct i915_request *rq, + struct i915_request *signal, + void (*hook)(struct i915_request *rq, + struct dma_fence *signal), + gfp_t gfp) { struct execute_cb *cb; @@ -341,6 +341,8 @@ __i915_request_await_execution(struct i915_request *rq, } spin_unlock_irq(&signal->lock); + /* Copy across semaphore status as we need the same behaviour */ + rq->sched.flags |= signal->sched.flags; return 0; } @@ -811,31 +813,21 @@ already_busywaiting(struct i915_request *rq) } static int -emit_semaphore_wait(struct i915_request *to, - struct i915_request *from, - gfp_t gfp) +__emit_semaphore_wait(struct i915_request *to, + struct i915_request *from, + u32 seqno) { const int has_token = INTEL_GEN(to->i915) >= 12; u32 hwsp_offset; - int len; + int len, err; u32 *cs; GEM_BUG_ON(INTEL_GEN(to->i915) < 8); - /* Just emit the first semaphore we see as request space is limited. */ - if (already_busywaiting(to) & from->engine->mask) - goto await_fence; - - if (i915_request_await_start(to, from) < 0) - goto await_fence; - - /* Only submit our spinner after the signaler is running! */ - if (__i915_request_await_execution(to, from, NULL, gfp)) - goto await_fence; - /* We need to pin the signaler's HWSP until we are finished reading. */ - if (intel_timeline_read_hwsp(from, to, &hwsp_offset)) - goto await_fence; + err = intel_timeline_read_hwsp(from, to, &hwsp_offset); + if (err) + return err; len = 4; if (has_token) @@ -858,7 +850,7 @@ emit_semaphore_wait(struct i915_request *to, MI_SEMAPHORE_POLL | MI_SEMAPHORE_SAD_GTE_SDD) + has_token; - *cs++ = from->fence.seqno; + *cs++ = seqno; *cs++ = hwsp_offset; *cs++ = 0; if (has_token) { @@ -867,6 +859,28 @@ emit_semaphore_wait(struct i915_request *to, } intel_ring_advance(to, cs); + return 0; +} + +static int +emit_semaphore_wait(struct i915_request *to, + struct i915_request *from, + gfp_t gfp) +{ + /* Just emit the first semaphore we see as request space is limited. */ + if (already_busywaiting(to) & from->engine->mask) + goto await_fence; + + if (i915_request_await_start(to, from) < 0) + goto await_fence; + + /* Only submit our spinner after the signaler is running! */ + if (__await_execution(to, from, NULL, gfp)) + goto await_fence; + + if (__emit_semaphore_wait(to, from, from->fence.seqno)) + goto await_fence; + to->sched.semaphores |= from->engine->mask; to->sched.flags |= I915_SCHED_HAS_SEMAPHORE_CHAIN; return 0; @@ -980,6 +994,57 @@ i915_request_await_dma_fence(struct i915_request *rq, struct dma_fence *fence) return 0; } +static bool intel_timeline_sync_has_start(struct intel_timeline *tl, + struct dma_fence *fence) +{ + return __intel_timeline_sync_is_later(tl, + fence->context, + fence->seqno - 1); +} + +static int intel_timeline_sync_set_start(struct intel_timeline *tl, + const struct dma_fence *fence) +{ + return __intel_timeline_sync_set(tl, fence->context, fence->seqno - 1); +} + +static int +__i915_request_await_execution(struct i915_request *to, + struct i915_request *from, + void (*hook)(struct i915_request *rq, + struct dma_fence *signal)) +{ + int err; + + /* Submit both requests at the same time */ + err = __await_execution(to, from, hook, I915_FENCE_GFP); + if (err) + return err; + + /* Squash repeated depenendices to the same timelines */ + if (intel_timeline_sync_has_start(i915_request_timeline(to), + &from->fence)) + return 0; + + /* Ensure both start together [after all semaphores in signal] */ + if (intel_engine_has_semaphores(to->engine)) + err = __emit_semaphore_wait(to, from, from->fence.seqno - 1); + else + err = i915_request_await_start(to, from); + if (err < 0) + return err; + + /* Couple the dependency tree for PI on this exposed to->fence */ + if (to->engine->schedule) { + err = i915_sched_node_add_dependency(&to->sched, &from->sched); + if (err < 0) + return err; + } + + return intel_timeline_sync_set_start(i915_request_timeline(to), + &from->fence); +} + int i915_request_await_execution(struct i915_request *rq, struct dma_fence *fence, @@ -1013,8 +1078,7 @@ i915_request_await_execution(struct i915_request *rq, if (dma_fence_is_i915(fence)) ret = __i915_request_await_execution(rq, to_request(fence), - hook, - I915_FENCE_GFP); + hook); else ret = i915_sw_fence_await_dma_fence(&rq->submit, fence, I915_FENCE_TIMEOUT, diff --git a/drivers/gpu/drm/i915/i915_scheduler.c b/drivers/gpu/drm/i915/i915_scheduler.c index 010d67f48ad9..247a9671bca5 100644 --- a/drivers/gpu/drm/i915/i915_scheduler.c +++ b/drivers/gpu/drm/i915/i915_scheduler.c @@ -474,7 +474,6 @@ void i915_sched_node_fini(struct i915_sched_node *node) * so we may be called out-of-order. */ list_for_each_entry_safe(dep, tmp, &node->signalers_list, signal_link) { - GEM_BUG_ON(!node_signaled(dep->signaler)); GEM_BUG_ON(!list_empty(&dep->dfs_link)); list_del(&dep->wait_link); diff --git a/drivers/gpu/drm/i915/i915_sw_fence_work.c b/drivers/gpu/drm/i915/i915_sw_fence_work.c index 07552cd544f2..8538ee7a521d 100644 --- a/drivers/gpu/drm/i915/i915_sw_fence_work.c +++ b/drivers/gpu/drm/i915/i915_sw_fence_work.c @@ -78,12 +78,11 @@ static const struct dma_fence_ops fence_ops = { void dma_fence_work_init(struct dma_fence_work *f, const struct dma_fence_work_ops *ops) { + f->ops = ops; spin_lock_init(&f->lock); dma_fence_init(&f->dma, &fence_ops, &f->lock, 0, 0); i915_sw_fence_init(&f->chain, fence_notify); INIT_WORK(&f->work, fence_work); - - f->ops = ops; } int dma_fence_work_chain(struct dma_fence_work *f, struct dma_fence *signal) diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index e5512f26e20a..01c822256b39 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -1104,8 +1104,14 @@ int i915_vma_move_to_active(struct i915_vma *vma, return err; if (flags & EXEC_OBJECT_WRITE) { - if (intel_frontbuffer_invalidate(obj->frontbuffer, ORIGIN_CS)) - i915_active_add_request(&obj->frontbuffer->write, rq); + struct intel_frontbuffer *front; + + front = __intel_frontbuffer_get(obj); + if (unlikely(front)) { + if (intel_frontbuffer_invalidate(front, ORIGIN_CS)) + i915_active_add_request(&front->write, rq); + intel_frontbuffer_put(front); + } dma_resv_add_excl_fence(vma->resv, &rq->fence); obj->write_domain = I915_GEM_DOMAIN_RENDER; diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 809bff955b5a..86379eddc908 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -4291,8 +4291,8 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *crtc_state, &crtc_state->wm.skl.optimal.planes[plane_id]; if (plane_id == PLANE_CURSOR) { - if (WARN_ON(wm->wm[level].min_ddb_alloc > - total[PLANE_CURSOR])) { + if (wm->wm[level].min_ddb_alloc > total[PLANE_CURSOR]) { + WARN_ON(wm->wm[level].min_ddb_alloc != U16_MAX); blocks = U32_MAX; break; } @@ -6565,6 +6565,17 @@ static void icl_init_clock_gating(struct drm_i915_private *dev_priv) /* WaEnable32PlaneMode:icl */ I915_WRITE(GEN9_CSFE_CHICKEN1_RCS, _MASKED_BIT_ENABLE(GEN11_ENABLE_32_PLANE_MODE)); + + /* + * Wa_1408615072:icl,ehl (vsunit) + * Wa_1407596294:icl,ehl (hsunit) + */ + intel_uncore_rmw(&dev_priv->uncore, UNSLICE_UNIT_LEVEL_CLKGATE, + 0, VSUNIT_CLKGATE_DIS | HSUNIT_CLKGATE_DIS); + + /* Wa_1407352427:icl,ehl */ + intel_uncore_rmw(&dev_priv->uncore, UNSLICE_UNIT_LEVEL_CLKGATE2, + 0, PSDUNIT_CLKGATE_DIS); } static void tgl_init_clock_gating(struct drm_i915_private *dev_priv) diff --git a/drivers/gpu/drm/i915/selftests/i915_random.h b/drivers/gpu/drm/i915/selftests/i915_random.h index 35cc69a3a1b9..05364eca20f7 100644 --- a/drivers/gpu/drm/i915/selftests/i915_random.h +++ b/drivers/gpu/drm/i915/selftests/i915_random.h @@ -25,6 +25,7 @@ #ifndef __I915_SELFTESTS_RANDOM_H__ #define __I915_SELFTESTS_RANDOM_H__ +#include <linux/math64.h> #include <linux/random.h> #include "../i915_selftest.h" diff --git a/drivers/gpu/drm/mcde/mcde_dsi.c b/drivers/gpu/drm/mcde/mcde_dsi.c index d6214d3c8b33..ef4c630afe3f 100644 --- a/drivers/gpu/drm/mcde/mcde_dsi.c +++ b/drivers/gpu/drm/mcde/mcde_dsi.c @@ -935,11 +935,13 @@ static int mcde_dsi_bind(struct device *dev, struct device *master, for_each_available_child_of_node(dev->of_node, child) { panel = of_drm_find_panel(child); if (IS_ERR(panel)) { - dev_err(dev, "failed to find panel try bridge (%lu)\n", + dev_err(dev, "failed to find panel try bridge (%ld)\n", PTR_ERR(panel)); + panel = NULL; + bridge = of_drm_find_bridge(child); if (IS_ERR(bridge)) { - dev_err(dev, "failed to find bridge (%lu)\n", + dev_err(dev, "failed to find bridge (%ld)\n", PTR_ERR(bridge)); return PTR_ERR(bridge); } diff --git a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c index f80a8ba75977..3305a94fc930 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c @@ -215,11 +215,12 @@ struct mtk_ddp_comp *mtk_drm_ddp_comp_for_plane(struct drm_crtc *crtc, struct mtk_drm_crtc *mtk_crtc = to_mtk_crtc(crtc); struct mtk_ddp_comp *comp; int i, count = 0; + unsigned int local_index = plane - mtk_crtc->planes; for (i = 0; i < mtk_crtc->ddp_comp_nr; i++) { comp = mtk_crtc->ddp_comp[i]; - if (plane->index < (count + mtk_ddp_comp_layer_nr(comp))) { - *local_layer = plane->index - count; + if (local_index < (count + mtk_ddp_comp_layer_nr(comp))) { + *local_layer = local_index - count; return comp; } count += mtk_ddp_comp_layer_nr(comp); @@ -310,7 +311,9 @@ static int mtk_crtc_ddp_hw_init(struct mtk_drm_crtc *mtk_crtc) plane_state = to_mtk_plane_state(plane->state); comp = mtk_drm_ddp_comp_for_plane(crtc, plane, &local_layer); - mtk_ddp_comp_layer_config(comp, local_layer, plane_state); + if (comp) + mtk_ddp_comp_layer_config(comp, local_layer, + plane_state); } return 0; @@ -386,8 +389,9 @@ static void mtk_crtc_ddp_config(struct drm_crtc *crtc) comp = mtk_drm_ddp_comp_for_plane(crtc, plane, &local_layer); - mtk_ddp_comp_layer_config(comp, local_layer, - plane_state); + if (comp) + mtk_ddp_comp_layer_config(comp, local_layer, + plane_state); plane_state->pending.config = false; } mtk_crtc->pending_planes = false; @@ -401,7 +405,9 @@ int mtk_drm_crtc_plane_check(struct drm_crtc *crtc, struct drm_plane *plane, struct mtk_ddp_comp *comp; comp = mtk_drm_ddp_comp_for_plane(crtc, plane, &local_layer); - return mtk_ddp_comp_layer_check(comp, local_layer, state); + if (comp) + return mtk_ddp_comp_layer_check(comp, local_layer, state); + return 0; } static void mtk_drm_crtc_atomic_enable(struct drm_crtc *crtc, diff --git a/drivers/gpu/drm/mediatek/mtk_dsi.c b/drivers/gpu/drm/mediatek/mtk_dsi.c index e9931bbbe846..d77c9f484ce3 100644 --- a/drivers/gpu/drm/mediatek/mtk_dsi.c +++ b/drivers/gpu/drm/mediatek/mtk_dsi.c @@ -230,28 +230,25 @@ static void mtk_dsi_mask(struct mtk_dsi *dsi, u32 offset, u32 mask, u32 data) static void mtk_dsi_phy_timconfig(struct mtk_dsi *dsi) { u32 timcon0, timcon1, timcon2, timcon3; - u32 ui, cycle_time; + u32 data_rate_mhz = DIV_ROUND_UP(dsi->data_rate, 1000000); struct mtk_phy_timing *timing = &dsi->phy_timing; - ui = DIV_ROUND_UP(1000000000, dsi->data_rate); - cycle_time = div_u64(8000000000ULL, dsi->data_rate); + timing->lpx = (60 * data_rate_mhz / (8 * 1000)) + 1; + timing->da_hs_prepare = (80 * data_rate_mhz + 4 * 1000) / 8000; + timing->da_hs_zero = (170 * data_rate_mhz + 10 * 1000) / 8000 + 1 - + timing->da_hs_prepare; + timing->da_hs_trail = timing->da_hs_prepare + 1; - timing->lpx = NS_TO_CYCLE(60, cycle_time); - timing->da_hs_prepare = NS_TO_CYCLE(50 + 5 * ui, cycle_time); - timing->da_hs_zero = NS_TO_CYCLE(110 + 6 * ui, cycle_time); - timing->da_hs_trail = NS_TO_CYCLE(77 + 4 * ui, cycle_time); + timing->ta_go = 4 * timing->lpx - 2; + timing->ta_sure = timing->lpx + 2; + timing->ta_get = 4 * timing->lpx; + timing->da_hs_exit = 2 * timing->lpx + 1; - timing->ta_go = 4 * timing->lpx; - timing->ta_sure = 3 * timing->lpx / 2; - timing->ta_get = 5 * timing->lpx; - timing->da_hs_exit = 2 * timing->lpx; - - timing->clk_hs_zero = NS_TO_CYCLE(336, cycle_time); - timing->clk_hs_trail = NS_TO_CYCLE(100, cycle_time) + 10; - - timing->clk_hs_prepare = NS_TO_CYCLE(64, cycle_time); - timing->clk_hs_post = NS_TO_CYCLE(80 + 52 * ui, cycle_time); - timing->clk_hs_exit = 2 * timing->lpx; + timing->clk_hs_prepare = 70 * data_rate_mhz / (8 * 1000); + timing->clk_hs_post = timing->clk_hs_prepare + 8; + timing->clk_hs_trail = timing->clk_hs_prepare; + timing->clk_hs_zero = timing->clk_hs_trail * 4; + timing->clk_hs_exit = 2 * timing->clk_hs_trail; timcon0 = timing->lpx | timing->da_hs_prepare << 8 | timing->da_hs_zero << 16 | timing->da_hs_trail << 24; @@ -482,27 +479,39 @@ static void mtk_dsi_config_vdo_timing(struct mtk_dsi *dsi) dsi_tmp_buf_bpp - 10); data_phy_cycles = timing->lpx + timing->da_hs_prepare + - timing->da_hs_zero + timing->da_hs_exit + 2; + timing->da_hs_zero + timing->da_hs_exit + 3; if (dsi->mode_flags & MIPI_DSI_MODE_VIDEO_BURST) { - if (vm->hfront_porch * dsi_tmp_buf_bpp > + if ((vm->hfront_porch + vm->hback_porch) * dsi_tmp_buf_bpp > data_phy_cycles * dsi->lanes + 18) { - horizontal_frontporch_byte = vm->hfront_porch * - dsi_tmp_buf_bpp - - data_phy_cycles * - dsi->lanes - 18; + horizontal_frontporch_byte = + vm->hfront_porch * dsi_tmp_buf_bpp - + (data_phy_cycles * dsi->lanes + 18) * + vm->hfront_porch / + (vm->hfront_porch + vm->hback_porch); + + horizontal_backporch_byte = + horizontal_backporch_byte - + (data_phy_cycles * dsi->lanes + 18) * + vm->hback_porch / + (vm->hfront_porch + vm->hback_porch); } else { DRM_WARN("HFP less than d-phy, FPS will under 60Hz\n"); horizontal_frontporch_byte = vm->hfront_porch * dsi_tmp_buf_bpp; } } else { - if (vm->hfront_porch * dsi_tmp_buf_bpp > + if ((vm->hfront_porch + vm->hback_porch) * dsi_tmp_buf_bpp > data_phy_cycles * dsi->lanes + 12) { - horizontal_frontporch_byte = vm->hfront_porch * - dsi_tmp_buf_bpp - - data_phy_cycles * - dsi->lanes - 12; + horizontal_frontporch_byte = + vm->hfront_porch * dsi_tmp_buf_bpp - + (data_phy_cycles * dsi->lanes + 12) * + vm->hfront_porch / + (vm->hfront_porch + vm->hback_porch); + horizontal_backporch_byte = horizontal_backporch_byte - + (data_phy_cycles * dsi->lanes + 12) * + vm->hback_porch / + (vm->hfront_porch + vm->hback_porch); } else { DRM_WARN("HFP less than d-phy, FPS will under 60Hz\n"); horizontal_frontporch_byte = vm->hfront_porch * diff --git a/drivers/gpu/drm/meson/meson_venc_cvbs.c b/drivers/gpu/drm/meson/meson_venc_cvbs.c index 9ab27aecfcf3..1bd6b6d15ffb 100644 --- a/drivers/gpu/drm/meson/meson_venc_cvbs.c +++ b/drivers/gpu/drm/meson/meson_venc_cvbs.c @@ -64,6 +64,25 @@ struct meson_cvbs_mode meson_cvbs_modes[MESON_CVBS_MODES_COUNT] = { }, }; +static const struct meson_cvbs_mode * +meson_cvbs_get_mode(const struct drm_display_mode *req_mode) +{ + int i; + + for (i = 0; i < MESON_CVBS_MODES_COUNT; ++i) { + struct meson_cvbs_mode *meson_mode = &meson_cvbs_modes[i]; + + if (drm_mode_match(req_mode, &meson_mode->mode, + DRM_MODE_MATCH_TIMINGS | + DRM_MODE_MATCH_CLOCK | + DRM_MODE_MATCH_FLAGS | + DRM_MODE_MATCH_3D_FLAGS)) + return meson_mode; + } + + return NULL; +} + /* Connector */ static void meson_cvbs_connector_destroy(struct drm_connector *connector) @@ -136,14 +155,8 @@ static int meson_venc_cvbs_encoder_atomic_check(struct drm_encoder *encoder, struct drm_crtc_state *crtc_state, struct drm_connector_state *conn_state) { - int i; - - for (i = 0; i < MESON_CVBS_MODES_COUNT; ++i) { - struct meson_cvbs_mode *meson_mode = &meson_cvbs_modes[i]; - - if (drm_mode_equal(&crtc_state->mode, &meson_mode->mode)) - return 0; - } + if (meson_cvbs_get_mode(&crtc_state->mode)) + return 0; return -EINVAL; } @@ -191,24 +204,17 @@ static void meson_venc_cvbs_encoder_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { + const struct meson_cvbs_mode *meson_mode = meson_cvbs_get_mode(mode); struct meson_venc_cvbs *meson_venc_cvbs = encoder_to_meson_venc_cvbs(encoder); struct meson_drm *priv = meson_venc_cvbs->priv; - int i; - for (i = 0; i < MESON_CVBS_MODES_COUNT; ++i) { - struct meson_cvbs_mode *meson_mode = &meson_cvbs_modes[i]; + if (meson_mode) { + meson_venci_cvbs_mode_set(priv, meson_mode->enci); - if (drm_mode_equal(mode, &meson_mode->mode)) { - meson_venci_cvbs_mode_set(priv, - meson_mode->enci); - - /* Setup 27MHz vclk2 for ENCI and VDAC */ - meson_vclk_setup(priv, MESON_VCLK_TARGET_CVBS, - MESON_VCLK_CVBS, MESON_VCLK_CVBS, - MESON_VCLK_CVBS, true); - break; - } + /* Setup 27MHz vclk2 for ENCI and VDAC */ + meson_vclk_setup(priv, MESON_VCLK_TARGET_CVBS, MESON_VCLK_CVBS, + MESON_VCLK_CVBS, MESON_VCLK_CVBS, true); } } diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.c b/drivers/gpu/drm/mgag200/mgag200_drv.c index d43951caeea0..b113876c2428 100644 --- a/drivers/gpu/drm/mgag200/mgag200_drv.c +++ b/drivers/gpu/drm/mgag200/mgag200_drv.c @@ -30,9 +30,8 @@ module_param_named(modeset, mgag200_modeset, int, 0400); static struct drm_driver driver; static const struct pci_device_id pciidlist[] = { - { PCI_VENDOR_ID_MATROX, 0x522, PCI_VENDOR_ID_SUN, 0x4852, 0, 0, + { PCI_VENDOR_ID_MATROX, 0x522, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_SE_A | MGAG200_FLAG_HW_BUG_NO_STARTADD}, - { PCI_VENDOR_ID_MATROX, 0x522, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_SE_A }, { PCI_VENDOR_ID_MATROX, 0x524, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_SE_B }, { PCI_VENDOR_ID_MATROX, 0x530, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_EV }, { PCI_VENDOR_ID_MATROX, 0x532, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_WB }, diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index c84f0a8b3f2c..ac678ace09a3 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -138,7 +138,7 @@ void __iomem *msm_ioremap(struct platform_device *pdev, const char *name, size = resource_size(res); - ptr = devm_ioremap_nocache(&pdev->dev, res->start, size); + ptr = devm_ioremap(&pdev->dev, res->start, size); if (!ptr) { DRM_DEV_ERROR(&pdev->dev, "failed to ioremap: %s\n", name); return ERR_PTR(-ENOMEM); diff --git a/drivers/gpu/drm/nouveau/dispnv50/atom.h b/drivers/gpu/drm/nouveau/dispnv50/atom.h index 43df86c38f58..24f7700768da 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/atom.h +++ b/drivers/gpu/drm/nouveau/dispnv50/atom.h @@ -114,6 +114,7 @@ struct nv50_head_atom { u8 nhsync:1; u8 nvsync:1; u8 depth:4; + u8 bpc; } or; /* Currently only used for MST */ diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c index 549486f1d937..63425e246018 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -326,9 +326,9 @@ nv50_outp_atomic_check_view(struct drm_encoder *encoder, * same size as the native one (e.g. different * refresh rate) */ - if (adjusted_mode->hdisplay == native_mode->hdisplay && - adjusted_mode->vdisplay == native_mode->vdisplay && - adjusted_mode->type & DRM_MODE_TYPE_DRIVER) + if (mode->hdisplay == native_mode->hdisplay && + mode->vdisplay == native_mode->vdisplay && + mode->type & DRM_MODE_TYPE_DRIVER) break; mode = native_mode; asyc->scaler.full = true; @@ -353,10 +353,20 @@ nv50_outp_atomic_check(struct drm_encoder *encoder, struct drm_crtc_state *crtc_state, struct drm_connector_state *conn_state) { - struct nouveau_connector *nv_connector = - nouveau_connector(conn_state->connector); - return nv50_outp_atomic_check_view(encoder, crtc_state, conn_state, - nv_connector->native_mode); + struct drm_connector *connector = conn_state->connector; + struct nouveau_connector *nv_connector = nouveau_connector(connector); + struct nv50_head_atom *asyh = nv50_head_atom(crtc_state); + int ret; + + ret = nv50_outp_atomic_check_view(encoder, crtc_state, conn_state, + nv_connector->native_mode); + if (ret) + return ret; + + if (crtc_state->mode_changed || crtc_state->connectors_changed) + asyh->or.bpc = connector->display_info.bpc; + + return 0; } /****************************************************************************** @@ -770,32 +780,54 @@ nv50_msto_atomic_check(struct drm_encoder *encoder, struct nv50_mstm *mstm = mstc->mstm; struct nv50_head_atom *asyh = nv50_head_atom(crtc_state); int slots; + int ret; + + ret = nv50_outp_atomic_check_view(encoder, crtc_state, conn_state, + mstc->native); + if (ret) + return ret; + + if (!crtc_state->mode_changed && !crtc_state->connectors_changed) + return 0; + + /* + * When restoring duplicated states, we need to make sure that the bw + * remains the same and avoid recalculating it, as the connector's bpc + * may have changed after the state was duplicated + */ + if (!state->duplicated) { + const int clock = crtc_state->adjusted_mode.clock; - if (crtc_state->mode_changed || crtc_state->connectors_changed) { /* - * When restoring duplicated states, we need to make sure that - * the bw remains the same and avoid recalculating it, as the - * connector's bpc may have changed after the state was - * duplicated + * XXX: Since we don't use HDR in userspace quite yet, limit + * the bpc to 8 to save bandwidth on the topology. In the + * future, we'll want to properly fix this by dynamically + * selecting the highest possible bpc that would fit in the + * topology */ - if (!state->duplicated) { - const int bpp = connector->display_info.bpc * 3; - const int clock = crtc_state->adjusted_mode.clock; + asyh->or.bpc = min(connector->display_info.bpc, 8U); + asyh->dp.pbn = drm_dp_calc_pbn_mode(clock, asyh->or.bpc * 3); + } - asyh->dp.pbn = drm_dp_calc_pbn_mode(clock, bpp); - } + slots = drm_dp_atomic_find_vcpi_slots(state, &mstm->mgr, mstc->port, + asyh->dp.pbn); + if (slots < 0) + return slots; - slots = drm_dp_atomic_find_vcpi_slots(state, &mstm->mgr, - mstc->port, - asyh->dp.pbn); - if (slots < 0) - return slots; + asyh->dp.tu = slots; - asyh->dp.tu = slots; - } + return 0; +} - return nv50_outp_atomic_check_view(encoder, crtc_state, conn_state, - mstc->native); +static u8 +nv50_dp_bpc_to_depth(unsigned int bpc) +{ + switch (bpc) { + case 6: return 0x2; + case 8: return 0x5; + case 10: /* fall-through */ + default: return 0x6; + } } static void @@ -808,7 +840,7 @@ nv50_msto_enable(struct drm_encoder *encoder) struct nv50_mstm *mstm = NULL; struct drm_connector *connector; struct drm_connector_list_iter conn_iter; - u8 proto, depth; + u8 proto; bool r; drm_connector_list_iter_begin(encoder->dev, &conn_iter); @@ -837,14 +869,8 @@ nv50_msto_enable(struct drm_encoder *encoder) else proto = 0x9; - switch (mstc->connector.display_info.bpc) { - case 6: depth = 0x2; break; - case 8: depth = 0x5; break; - case 10: - default: depth = 0x6; break; - } - - mstm->outp->update(mstm->outp, head->base.index, armh, proto, depth); + mstm->outp->update(mstm->outp, head->base.index, armh, proto, + nv50_dp_bpc_to_depth(armh->or.bpc)); msto->head = head; msto->mstc = mstc; @@ -1498,20 +1524,14 @@ nv50_sor_enable(struct drm_encoder *encoder) lvds.lvds.script |= 0x0200; } - if (nv_connector->base.display_info.bpc == 8) + if (asyh->or.bpc == 8) lvds.lvds.script |= 0x0200; } nvif_mthd(&disp->disp->object, 0, &lvds, sizeof(lvds)); break; case DCB_OUTPUT_DP: - if (nv_connector->base.display_info.bpc == 6) - depth = 0x2; - else - if (nv_connector->base.display_info.bpc == 8) - depth = 0x5; - else - depth = 0x6; + depth = nv50_dp_bpc_to_depth(asyh->or.bpc); if (nv_encoder->link & 1) proto = 0x8; @@ -1662,7 +1682,7 @@ nv50_pior_enable(struct drm_encoder *encoder) nv50_outp_acquire(nv_encoder); nv_connector = nouveau_encoder_connector_get(nv_encoder); - switch (nv_connector->base.display_info.bpc) { + switch (asyh->or.bpc) { case 10: asyh->or.depth = 0x6; break; case 8: asyh->or.depth = 0x5; break; case 6: asyh->or.depth = 0x2; break; diff --git a/drivers/gpu/drm/nouveau/dispnv50/head.c b/drivers/gpu/drm/nouveau/dispnv50/head.c index 71c23bf1fe25..c9692df2b76c 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/head.c +++ b/drivers/gpu/drm/nouveau/dispnv50/head.c @@ -81,18 +81,17 @@ nv50_head_atomic_check_dither(struct nv50_head_atom *armh, struct nv50_head_atom *asyh, struct nouveau_conn_atom *asyc) { - struct drm_connector *connector = asyc->state.connector; u32 mode = 0x00; if (asyc->dither.mode == DITHERING_MODE_AUTO) { - if (asyh->base.depth > connector->display_info.bpc * 3) + if (asyh->base.depth > asyh->or.bpc * 3) mode = DITHERING_MODE_DYNAMIC2X2; } else { mode = asyc->dither.mode; } if (asyc->dither.depth == DITHERING_DEPTH_AUTO) { - if (connector->display_info.bpc >= 8) + if (asyh->or.bpc >= 8) mode |= DITHERING_DEPTH_8BPC; } else { mode |= asyc->dither.depth; diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 5b413588b823..9a9a7f5003d3 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -245,14 +245,22 @@ nouveau_conn_atomic_duplicate_state(struct drm_connector *connector) void nouveau_conn_reset(struct drm_connector *connector) { + struct nouveau_connector *nv_connector = nouveau_connector(connector); struct nouveau_conn_atom *asyc; - if (WARN_ON(!(asyc = kzalloc(sizeof(*asyc), GFP_KERNEL)))) - return; + if (drm_drv_uses_atomic_modeset(connector->dev)) { + if (WARN_ON(!(asyc = kzalloc(sizeof(*asyc), GFP_KERNEL)))) + return; + + if (connector->state) + nouveau_conn_atomic_destroy_state(connector, + connector->state); + + __drm_atomic_helper_connector_reset(connector, &asyc->state); + } else { + asyc = &nv_connector->properties_state; + } - if (connector->state) - nouveau_conn_atomic_destroy_state(connector, connector->state); - __drm_atomic_helper_connector_reset(connector, &asyc->state); asyc->dither.mode = DITHERING_MODE_AUTO; asyc->dither.depth = DITHERING_DEPTH_AUTO; asyc->scaler.mode = DRM_MODE_SCALE_NONE; @@ -276,8 +284,14 @@ void nouveau_conn_attach_properties(struct drm_connector *connector) { struct drm_device *dev = connector->dev; - struct nouveau_conn_atom *armc = nouveau_conn_atom(connector->state); struct nouveau_display *disp = nouveau_display(dev); + struct nouveau_connector *nv_connector = nouveau_connector(connector); + struct nouveau_conn_atom *armc; + + if (drm_drv_uses_atomic_modeset(connector->dev)) + armc = nouveau_conn_atom(connector->state); + else + armc = &nv_connector->properties_state; /* Init DVI-I specific properties. */ if (connector->connector_type == DRM_MODE_CONNECTOR_DVII) @@ -748,9 +762,9 @@ static int nouveau_connector_set_property(struct drm_connector *connector, struct drm_property *property, uint64_t value) { - struct nouveau_conn_atom *asyc = nouveau_conn_atom(connector->state); struct nouveau_connector *nv_connector = nouveau_connector(connector); struct nouveau_encoder *nv_encoder = nv_connector->detected_encoder; + struct nouveau_conn_atom *asyc = &nv_connector->properties_state; struct drm_encoder *encoder = to_drm_encoder(nv_encoder); int ret; diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h index f43a8d63aef8..de84fb4708c7 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.h +++ b/drivers/gpu/drm/nouveau/nouveau_connector.h @@ -29,6 +29,7 @@ #include <nvif/notify.h> +#include <drm/drm_crtc.h> #include <drm/drm_edid.h> #include <drm/drm_encoder.h> #include <drm/drm_dp_helper.h> @@ -44,6 +45,60 @@ struct dcb_output; struct nouveau_backlight; #endif +#define nouveau_conn_atom(p) \ + container_of((p), struct nouveau_conn_atom, state) + +struct nouveau_conn_atom { + struct drm_connector_state state; + + struct { + /* The enum values specifically defined here match nv50/gf119 + * hw values, and the code relies on this. + */ + enum { + DITHERING_MODE_OFF = 0x00, + DITHERING_MODE_ON = 0x01, + DITHERING_MODE_DYNAMIC2X2 = 0x10 | DITHERING_MODE_ON, + DITHERING_MODE_STATIC2X2 = 0x18 | DITHERING_MODE_ON, + DITHERING_MODE_TEMPORAL = 0x20 | DITHERING_MODE_ON, + DITHERING_MODE_AUTO + } mode; + enum { + DITHERING_DEPTH_6BPC = 0x00, + DITHERING_DEPTH_8BPC = 0x02, + DITHERING_DEPTH_AUTO + } depth; + } dither; + + struct { + int mode; /* DRM_MODE_SCALE_* */ + struct { + enum { + UNDERSCAN_OFF, + UNDERSCAN_ON, + UNDERSCAN_AUTO, + } mode; + u32 hborder; + u32 vborder; + } underscan; + bool full; + } scaler; + + struct { + int color_vibrance; + int vibrant_hue; + } procamp; + + union { + struct { + bool dither:1; + bool scaler:1; + bool procamp:1; + }; + u8 mask; + } set; +}; + struct nouveau_connector { struct drm_connector base; enum dcb_connector_type type; @@ -63,6 +118,12 @@ struct nouveau_connector { #ifdef CONFIG_DRM_NOUVEAU_BACKLIGHT struct nouveau_backlight *backlight; #endif + /* + * Our connector property code expects a nouveau_conn_atom struct + * even on pre-nv50 where we do not support atomic. This embedded + * version gets used in the non atomic modeset case. + */ + struct nouveau_conn_atom properties_state; }; static inline struct nouveau_connector *nouveau_connector( @@ -121,61 +182,6 @@ extern int nouveau_ignorelid; extern int nouveau_duallink; extern int nouveau_hdmimhz; -#include <drm/drm_crtc.h> -#define nouveau_conn_atom(p) \ - container_of((p), struct nouveau_conn_atom, state) - -struct nouveau_conn_atom { - struct drm_connector_state state; - - struct { - /* The enum values specifically defined here match nv50/gf119 - * hw values, and the code relies on this. - */ - enum { - DITHERING_MODE_OFF = 0x00, - DITHERING_MODE_ON = 0x01, - DITHERING_MODE_DYNAMIC2X2 = 0x10 | DITHERING_MODE_ON, - DITHERING_MODE_STATIC2X2 = 0x18 | DITHERING_MODE_ON, - DITHERING_MODE_TEMPORAL = 0x20 | DITHERING_MODE_ON, - DITHERING_MODE_AUTO - } mode; - enum { - DITHERING_DEPTH_6BPC = 0x00, - DITHERING_DEPTH_8BPC = 0x02, - DITHERING_DEPTH_AUTO - } depth; - } dither; - - struct { - int mode; /* DRM_MODE_SCALE_* */ - struct { - enum { - UNDERSCAN_OFF, - UNDERSCAN_ON, - UNDERSCAN_AUTO, - } mode; - u32 hborder; - u32 vborder; - } underscan; - bool full; - } scaler; - - struct { - int color_vibrance; - int vibrant_hue; - } procamp; - - union { - struct { - bool dither:1; - bool scaler:1; - bool procamp:1; - }; - u8 mask; - } set; -}; - void nouveau_conn_attach_properties(struct drm_connector *); void nouveau_conn_reset(struct drm_connector *); struct drm_connector_state * diff --git a/drivers/gpu/drm/panfrost/panfrost_devfreq.c b/drivers/gpu/drm/panfrost/panfrost_devfreq.c index 4c4e8a30a1ac..536ba93b0f46 100644 --- a/drivers/gpu/drm/panfrost/panfrost_devfreq.c +++ b/drivers/gpu/drm/panfrost/panfrost_devfreq.c @@ -18,15 +18,18 @@ static void panfrost_devfreq_update_utilization(struct panfrost_device *pfdev); static int panfrost_devfreq_target(struct device *dev, unsigned long *freq, u32 flags) { - struct panfrost_device *pfdev = dev_get_drvdata(dev); + struct dev_pm_opp *opp; int err; + opp = devfreq_recommended_opp(dev, freq, flags); + if (IS_ERR(opp)) + return PTR_ERR(opp); + dev_pm_opp_put(opp); + err = dev_pm_opp_set_rate(dev, *freq); if (err) return err; - *freq = clk_get_rate(pfdev->clock); - return 0; } @@ -60,20 +63,10 @@ static int panfrost_devfreq_get_dev_status(struct device *dev, return 0; } -static int panfrost_devfreq_get_cur_freq(struct device *dev, unsigned long *freq) -{ - struct panfrost_device *pfdev = platform_get_drvdata(to_platform_device(dev)); - - *freq = clk_get_rate(pfdev->clock); - - return 0; -} - static struct devfreq_dev_profile panfrost_devfreq_profile = { .polling_ms = 50, /* ~3 frames */ .target = panfrost_devfreq_target, .get_dev_status = panfrost_devfreq_get_dev_status, - .get_cur_freq = panfrost_devfreq_get_cur_freq, }; int panfrost_devfreq_init(struct panfrost_device *pfdev) diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c index 9458dc6c750c..88b431a267af 100644 --- a/drivers/gpu/drm/panfrost/panfrost_drv.c +++ b/drivers/gpu/drm/panfrost/panfrost_drv.c @@ -78,8 +78,10 @@ static int panfrost_ioctl_get_param(struct drm_device *ddev, void *data, struct static int panfrost_ioctl_create_bo(struct drm_device *dev, void *data, struct drm_file *file) { + struct panfrost_file_priv *priv = file->driver_priv; struct panfrost_gem_object *bo; struct drm_panfrost_create_bo *args = data; + struct panfrost_gem_mapping *mapping; if (!args->size || args->pad || (args->flags & ~(PANFROST_BO_NOEXEC | PANFROST_BO_HEAP))) @@ -95,7 +97,14 @@ static int panfrost_ioctl_create_bo(struct drm_device *dev, void *data, if (IS_ERR(bo)) return PTR_ERR(bo); - args->offset = bo->node.start << PAGE_SHIFT; + mapping = panfrost_gem_mapping_get(bo, priv); + if (!mapping) { + drm_gem_object_put_unlocked(&bo->base.base); + return -EINVAL; + } + + args->offset = mapping->mmnode.start << PAGE_SHIFT; + panfrost_gem_mapping_put(mapping); return 0; } @@ -119,6 +128,11 @@ panfrost_lookup_bos(struct drm_device *dev, struct drm_panfrost_submit *args, struct panfrost_job *job) { + struct panfrost_file_priv *priv = file_priv->driver_priv; + struct panfrost_gem_object *bo; + unsigned int i; + int ret; + job->bo_count = args->bo_handle_count; if (!job->bo_count) @@ -130,9 +144,32 @@ panfrost_lookup_bos(struct drm_device *dev, if (!job->implicit_fences) return -ENOMEM; - return drm_gem_objects_lookup(file_priv, - (void __user *)(uintptr_t)args->bo_handles, - job->bo_count, &job->bos); + ret = drm_gem_objects_lookup(file_priv, + (void __user *)(uintptr_t)args->bo_handles, + job->bo_count, &job->bos); + if (ret) + return ret; + + job->mappings = kvmalloc_array(job->bo_count, + sizeof(struct panfrost_gem_mapping *), + GFP_KERNEL | __GFP_ZERO); + if (!job->mappings) + return -ENOMEM; + + for (i = 0; i < job->bo_count; i++) { + struct panfrost_gem_mapping *mapping; + + bo = to_panfrost_bo(job->bos[i]); + mapping = panfrost_gem_mapping_get(bo, priv); + if (!mapping) { + ret = -EINVAL; + break; + } + + job->mappings[i] = mapping; + } + + return ret; } /** @@ -303,21 +340,26 @@ static int panfrost_ioctl_mmap_bo(struct drm_device *dev, void *data, } /* Don't allow mmapping of heap objects as pages are not pinned. */ - if (to_panfrost_bo(gem_obj)->is_heap) - return -EINVAL; + if (to_panfrost_bo(gem_obj)->is_heap) { + ret = -EINVAL; + goto out; + } ret = drm_gem_create_mmap_offset(gem_obj); if (ret == 0) args->offset = drm_vma_node_offset_addr(&gem_obj->vma_node); - drm_gem_object_put_unlocked(gem_obj); +out: + drm_gem_object_put_unlocked(gem_obj); return ret; } static int panfrost_ioctl_get_bo_offset(struct drm_device *dev, void *data, struct drm_file *file_priv) { + struct panfrost_file_priv *priv = file_priv->driver_priv; struct drm_panfrost_get_bo_offset *args = data; + struct panfrost_gem_mapping *mapping; struct drm_gem_object *gem_obj; struct panfrost_gem_object *bo; @@ -328,18 +370,26 @@ static int panfrost_ioctl_get_bo_offset(struct drm_device *dev, void *data, } bo = to_panfrost_bo(gem_obj); - args->offset = bo->node.start << PAGE_SHIFT; - + mapping = panfrost_gem_mapping_get(bo, priv); drm_gem_object_put_unlocked(gem_obj); + + if (!mapping) + return -EINVAL; + + args->offset = mapping->mmnode.start << PAGE_SHIFT; + panfrost_gem_mapping_put(mapping); return 0; } static int panfrost_ioctl_madvise(struct drm_device *dev, void *data, struct drm_file *file_priv) { + struct panfrost_file_priv *priv = file_priv->driver_priv; struct drm_panfrost_madvise *args = data; struct panfrost_device *pfdev = dev->dev_private; struct drm_gem_object *gem_obj; + struct panfrost_gem_object *bo; + int ret = 0; gem_obj = drm_gem_object_lookup(file_priv, args->handle); if (!gem_obj) { @@ -347,23 +397,48 @@ static int panfrost_ioctl_madvise(struct drm_device *dev, void *data, return -ENOENT; } - args->retained = drm_gem_shmem_madvise(gem_obj, args->madv); + bo = to_panfrost_bo(gem_obj); - if (args->retained) { - struct panfrost_gem_object *bo = to_panfrost_bo(gem_obj); + mutex_lock(&pfdev->shrinker_lock); + mutex_lock(&bo->mappings.lock); + if (args->madv == PANFROST_MADV_DONTNEED) { + struct panfrost_gem_mapping *first; + + first = list_first_entry(&bo->mappings.list, + struct panfrost_gem_mapping, + node); + + /* + * If we want to mark the BO purgeable, there must be only one + * user: the caller FD. + * We could do something smarter and mark the BO purgeable only + * when all its users have marked it purgeable, but globally + * visible/shared BOs are likely to never be marked purgeable + * anyway, so let's not bother. + */ + if (!list_is_singular(&bo->mappings.list) || + WARN_ON_ONCE(first->mmu != &priv->mmu)) { + ret = -EINVAL; + goto out_unlock_mappings; + } + } - mutex_lock(&pfdev->shrinker_lock); + args->retained = drm_gem_shmem_madvise(gem_obj, args->madv); + if (args->retained) { if (args->madv == PANFROST_MADV_DONTNEED) - list_add_tail(&bo->base.madv_list, &pfdev->shrinker_list); + list_add_tail(&bo->base.madv_list, + &pfdev->shrinker_list); else if (args->madv == PANFROST_MADV_WILLNEED) list_del_init(&bo->base.madv_list); - - mutex_unlock(&pfdev->shrinker_lock); } +out_unlock_mappings: + mutex_unlock(&bo->mappings.lock); + mutex_unlock(&pfdev->shrinker_lock); + drm_gem_object_put_unlocked(gem_obj); - return 0; + return ret; } int panfrost_unstable_ioctl_check(void) @@ -443,7 +518,7 @@ panfrost_postclose(struct drm_device *dev, struct drm_file *file) { struct panfrost_file_priv *panfrost_priv = file->driver_priv; - panfrost_perfcnt_close(panfrost_priv); + panfrost_perfcnt_close(file); panfrost_job_close(panfrost_priv); panfrost_mmu_pgtable_free(panfrost_priv); diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.c b/drivers/gpu/drm/panfrost/panfrost_gem.c index deca0c30bbd4..17b654e1eb94 100644 --- a/drivers/gpu/drm/panfrost/panfrost_gem.c +++ b/drivers/gpu/drm/panfrost/panfrost_gem.c @@ -19,6 +19,22 @@ static void panfrost_gem_free_object(struct drm_gem_object *obj) struct panfrost_gem_object *bo = to_panfrost_bo(obj); struct panfrost_device *pfdev = obj->dev->dev_private; + /* + * Make sure the BO is no longer inserted in the shrinker list before + * taking care of the destruction itself. If we don't do that we have a + * race condition between this function and what's done in + * panfrost_gem_shrinker_scan(). + */ + mutex_lock(&pfdev->shrinker_lock); + list_del_init(&bo->base.madv_list); + mutex_unlock(&pfdev->shrinker_lock); + + /* + * If we still have mappings attached to the BO, there's a problem in + * our refcounting. + */ + WARN_ON_ONCE(!list_empty(&bo->mappings.list)); + if (bo->sgts) { int i; int n_sgt = bo->base.base.size / SZ_2M; @@ -33,15 +49,73 @@ static void panfrost_gem_free_object(struct drm_gem_object *obj) kfree(bo->sgts); } - mutex_lock(&pfdev->shrinker_lock); - if (!list_empty(&bo->base.madv_list)) - list_del(&bo->base.madv_list); - mutex_unlock(&pfdev->shrinker_lock); - drm_gem_shmem_free_object(obj); } -static int panfrost_gem_open(struct drm_gem_object *obj, struct drm_file *file_priv) +struct panfrost_gem_mapping * +panfrost_gem_mapping_get(struct panfrost_gem_object *bo, + struct panfrost_file_priv *priv) +{ + struct panfrost_gem_mapping *iter, *mapping = NULL; + + mutex_lock(&bo->mappings.lock); + list_for_each_entry(iter, &bo->mappings.list, node) { + if (iter->mmu == &priv->mmu) { + kref_get(&iter->refcount); + mapping = iter; + break; + } + } + mutex_unlock(&bo->mappings.lock); + + return mapping; +} + +static void +panfrost_gem_teardown_mapping(struct panfrost_gem_mapping *mapping) +{ + struct panfrost_file_priv *priv; + + if (mapping->active) + panfrost_mmu_unmap(mapping); + + priv = container_of(mapping->mmu, struct panfrost_file_priv, mmu); + spin_lock(&priv->mm_lock); + if (drm_mm_node_allocated(&mapping->mmnode)) + drm_mm_remove_node(&mapping->mmnode); + spin_unlock(&priv->mm_lock); +} + +static void panfrost_gem_mapping_release(struct kref *kref) +{ + struct panfrost_gem_mapping *mapping; + + mapping = container_of(kref, struct panfrost_gem_mapping, refcount); + + panfrost_gem_teardown_mapping(mapping); + drm_gem_object_put_unlocked(&mapping->obj->base.base); + kfree(mapping); +} + +void panfrost_gem_mapping_put(struct panfrost_gem_mapping *mapping) +{ + if (!mapping) + return; + + kref_put(&mapping->refcount, panfrost_gem_mapping_release); +} + +void panfrost_gem_teardown_mappings(struct panfrost_gem_object *bo) +{ + struct panfrost_gem_mapping *mapping; + + mutex_lock(&bo->mappings.lock); + list_for_each_entry(mapping, &bo->mappings.list, node) + panfrost_gem_teardown_mapping(mapping); + mutex_unlock(&bo->mappings.lock); +} + +int panfrost_gem_open(struct drm_gem_object *obj, struct drm_file *file_priv) { int ret; size_t size = obj->size; @@ -49,6 +123,16 @@ static int panfrost_gem_open(struct drm_gem_object *obj, struct drm_file *file_p struct panfrost_gem_object *bo = to_panfrost_bo(obj); unsigned long color = bo->noexec ? PANFROST_BO_NOEXEC : 0; struct panfrost_file_priv *priv = file_priv->driver_priv; + struct panfrost_gem_mapping *mapping; + + mapping = kzalloc(sizeof(*mapping), GFP_KERNEL); + if (!mapping) + return -ENOMEM; + + INIT_LIST_HEAD(&mapping->node); + kref_init(&mapping->refcount); + drm_gem_object_get(obj); + mapping->obj = bo; /* * Executable buffers cannot cross a 16MB boundary as the program @@ -61,37 +145,48 @@ static int panfrost_gem_open(struct drm_gem_object *obj, struct drm_file *file_p else align = size >= SZ_2M ? SZ_2M >> PAGE_SHIFT : 0; - bo->mmu = &priv->mmu; + mapping->mmu = &priv->mmu; spin_lock(&priv->mm_lock); - ret = drm_mm_insert_node_generic(&priv->mm, &bo->node, + ret = drm_mm_insert_node_generic(&priv->mm, &mapping->mmnode, size >> PAGE_SHIFT, align, color, 0); spin_unlock(&priv->mm_lock); if (ret) - return ret; + goto err; if (!bo->is_heap) { - ret = panfrost_mmu_map(bo); - if (ret) { - spin_lock(&priv->mm_lock); - drm_mm_remove_node(&bo->node); - spin_unlock(&priv->mm_lock); - } + ret = panfrost_mmu_map(mapping); + if (ret) + goto err; } + + mutex_lock(&bo->mappings.lock); + WARN_ON(bo->base.madv != PANFROST_MADV_WILLNEED); + list_add_tail(&mapping->node, &bo->mappings.list); + mutex_unlock(&bo->mappings.lock); + +err: + if (ret) + panfrost_gem_mapping_put(mapping); return ret; } -static void panfrost_gem_close(struct drm_gem_object *obj, struct drm_file *file_priv) +void panfrost_gem_close(struct drm_gem_object *obj, struct drm_file *file_priv) { - struct panfrost_gem_object *bo = to_panfrost_bo(obj); struct panfrost_file_priv *priv = file_priv->driver_priv; + struct panfrost_gem_object *bo = to_panfrost_bo(obj); + struct panfrost_gem_mapping *mapping = NULL, *iter; - if (bo->is_mapped) - panfrost_mmu_unmap(bo); + mutex_lock(&bo->mappings.lock); + list_for_each_entry(iter, &bo->mappings.list, node) { + if (iter->mmu == &priv->mmu) { + mapping = iter; + list_del(&iter->node); + break; + } + } + mutex_unlock(&bo->mappings.lock); - spin_lock(&priv->mm_lock); - if (drm_mm_node_allocated(&bo->node)) - drm_mm_remove_node(&bo->node); - spin_unlock(&priv->mm_lock); + panfrost_gem_mapping_put(mapping); } static int panfrost_gem_pin(struct drm_gem_object *obj) @@ -131,6 +226,8 @@ struct drm_gem_object *panfrost_gem_create_object(struct drm_device *dev, size_t if (!obj) return NULL; + INIT_LIST_HEAD(&obj->mappings.list); + mutex_init(&obj->mappings.lock); obj->base.base.funcs = &panfrost_gem_funcs; return &obj->base.base; diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.h b/drivers/gpu/drm/panfrost/panfrost_gem.h index 50920819cc16..ca1bc9019600 100644 --- a/drivers/gpu/drm/panfrost/panfrost_gem.h +++ b/drivers/gpu/drm/panfrost/panfrost_gem.h @@ -13,23 +13,46 @@ struct panfrost_gem_object { struct drm_gem_shmem_object base; struct sg_table *sgts; - struct panfrost_mmu *mmu; - struct drm_mm_node node; - bool is_mapped :1; + /* + * Use a list for now. If searching a mapping ever becomes the + * bottleneck, we should consider using an RB-tree, or even better, + * let the core store drm_gem_object_mapping entries (where we + * could place driver specific data) instead of drm_gem_object ones + * in its drm_file->object_idr table. + * + * struct drm_gem_object_mapping { + * struct drm_gem_object *obj; + * void *driver_priv; + * }; + */ + struct { + struct list_head list; + struct mutex lock; + } mappings; + bool noexec :1; bool is_heap :1; }; +struct panfrost_gem_mapping { + struct list_head node; + struct kref refcount; + struct panfrost_gem_object *obj; + struct drm_mm_node mmnode; + struct panfrost_mmu *mmu; + bool active :1; +}; + static inline struct panfrost_gem_object *to_panfrost_bo(struct drm_gem_object *obj) { return container_of(to_drm_gem_shmem_obj(obj), struct panfrost_gem_object, base); } -static inline -struct panfrost_gem_object *drm_mm_node_to_panfrost_bo(struct drm_mm_node *node) +static inline struct panfrost_gem_mapping * +drm_mm_node_to_panfrost_mapping(struct drm_mm_node *node) { - return container_of(node, struct panfrost_gem_object, node); + return container_of(node, struct panfrost_gem_mapping, mmnode); } struct drm_gem_object *panfrost_gem_create_object(struct drm_device *dev, size_t size); @@ -45,6 +68,16 @@ panfrost_gem_create_with_handle(struct drm_file *file_priv, u32 flags, uint32_t *handle); +int panfrost_gem_open(struct drm_gem_object *obj, struct drm_file *file_priv); +void panfrost_gem_close(struct drm_gem_object *obj, + struct drm_file *file_priv); + +struct panfrost_gem_mapping * +panfrost_gem_mapping_get(struct panfrost_gem_object *bo, + struct panfrost_file_priv *priv); +void panfrost_gem_mapping_put(struct panfrost_gem_mapping *mapping); +void panfrost_gem_teardown_mappings(struct panfrost_gem_object *bo); + void panfrost_gem_shrinker_init(struct drm_device *dev); void panfrost_gem_shrinker_cleanup(struct drm_device *dev); diff --git a/drivers/gpu/drm/panfrost/panfrost_gem_shrinker.c b/drivers/gpu/drm/panfrost/panfrost_gem_shrinker.c index 458f0fa68111..f5dd7b29bc95 100644 --- a/drivers/gpu/drm/panfrost/panfrost_gem_shrinker.c +++ b/drivers/gpu/drm/panfrost/panfrost_gem_shrinker.c @@ -39,11 +39,12 @@ panfrost_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc static bool panfrost_gem_purge(struct drm_gem_object *obj) { struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj); + struct panfrost_gem_object *bo = to_panfrost_bo(obj); if (!mutex_trylock(&shmem->pages_lock)) return false; - panfrost_mmu_unmap(to_panfrost_bo(obj)); + panfrost_gem_teardown_mappings(bo); drm_gem_shmem_purge_locked(obj); mutex_unlock(&shmem->pages_lock); diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c index d411eb6c8eb9..e364ee00f3d0 100644 --- a/drivers/gpu/drm/panfrost/panfrost_job.c +++ b/drivers/gpu/drm/panfrost/panfrost_job.c @@ -268,9 +268,20 @@ static void panfrost_job_cleanup(struct kref *ref) dma_fence_put(job->done_fence); dma_fence_put(job->render_done_fence); - if (job->bos) { + if (job->mappings) { for (i = 0; i < job->bo_count; i++) + panfrost_gem_mapping_put(job->mappings[i]); + kvfree(job->mappings); + } + + if (job->bos) { + struct panfrost_gem_object *bo; + + for (i = 0; i < job->bo_count; i++) { + bo = to_panfrost_bo(job->bos[i]); drm_gem_object_put_unlocked(job->bos[i]); + } + kvfree(job->bos); } diff --git a/drivers/gpu/drm/panfrost/panfrost_job.h b/drivers/gpu/drm/panfrost/panfrost_job.h index 62454128a792..bbd3ba97ff67 100644 --- a/drivers/gpu/drm/panfrost/panfrost_job.h +++ b/drivers/gpu/drm/panfrost/panfrost_job.h @@ -32,6 +32,7 @@ struct panfrost_job { /* Exclusive fences we have taken from the BOs to wait for */ struct dma_fence **implicit_fences; + struct panfrost_gem_mapping **mappings; struct drm_gem_object **bos; u32 bo_count; diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c index a3ed64a1f15e..763cfca886a7 100644 --- a/drivers/gpu/drm/panfrost/panfrost_mmu.c +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c @@ -269,14 +269,15 @@ static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu, return 0; } -int panfrost_mmu_map(struct panfrost_gem_object *bo) +int panfrost_mmu_map(struct panfrost_gem_mapping *mapping) { + struct panfrost_gem_object *bo = mapping->obj; struct drm_gem_object *obj = &bo->base.base; struct panfrost_device *pfdev = to_panfrost_device(obj->dev); struct sg_table *sgt; int prot = IOMMU_READ | IOMMU_WRITE; - if (WARN_ON(bo->is_mapped)) + if (WARN_ON(mapping->active)) return 0; if (bo->noexec) @@ -286,25 +287,28 @@ int panfrost_mmu_map(struct panfrost_gem_object *bo) if (WARN_ON(IS_ERR(sgt))) return PTR_ERR(sgt); - mmu_map_sg(pfdev, bo->mmu, bo->node.start << PAGE_SHIFT, prot, sgt); - bo->is_mapped = true; + mmu_map_sg(pfdev, mapping->mmu, mapping->mmnode.start << PAGE_SHIFT, + prot, sgt); + mapping->active = true; return 0; } -void panfrost_mmu_unmap(struct panfrost_gem_object *bo) +void panfrost_mmu_unmap(struct panfrost_gem_mapping *mapping) { + struct panfrost_gem_object *bo = mapping->obj; struct drm_gem_object *obj = &bo->base.base; struct panfrost_device *pfdev = to_panfrost_device(obj->dev); - struct io_pgtable_ops *ops = bo->mmu->pgtbl_ops; - u64 iova = bo->node.start << PAGE_SHIFT; - size_t len = bo->node.size << PAGE_SHIFT; + struct io_pgtable_ops *ops = mapping->mmu->pgtbl_ops; + u64 iova = mapping->mmnode.start << PAGE_SHIFT; + size_t len = mapping->mmnode.size << PAGE_SHIFT; size_t unmapped_len = 0; - if (WARN_ON(!bo->is_mapped)) + if (WARN_ON(!mapping->active)) return; - dev_dbg(pfdev->dev, "unmap: as=%d, iova=%llx, len=%zx", bo->mmu->as, iova, len); + dev_dbg(pfdev->dev, "unmap: as=%d, iova=%llx, len=%zx", + mapping->mmu->as, iova, len); while (unmapped_len < len) { size_t unmapped_page; @@ -318,8 +322,9 @@ void panfrost_mmu_unmap(struct panfrost_gem_object *bo) unmapped_len += pgsize; } - panfrost_mmu_flush_range(pfdev, bo->mmu, bo->node.start << PAGE_SHIFT, len); - bo->is_mapped = false; + panfrost_mmu_flush_range(pfdev, mapping->mmu, + mapping->mmnode.start << PAGE_SHIFT, len); + mapping->active = false; } static void mmu_tlb_inv_context_s1(void *cookie) @@ -394,10 +399,10 @@ void panfrost_mmu_pgtable_free(struct panfrost_file_priv *priv) free_io_pgtable_ops(mmu->pgtbl_ops); } -static struct panfrost_gem_object * -addr_to_drm_mm_node(struct panfrost_device *pfdev, int as, u64 addr) +static struct panfrost_gem_mapping * +addr_to_mapping(struct panfrost_device *pfdev, int as, u64 addr) { - struct panfrost_gem_object *bo = NULL; + struct panfrost_gem_mapping *mapping = NULL; struct panfrost_file_priv *priv; struct drm_mm_node *node; u64 offset = addr >> PAGE_SHIFT; @@ -418,8 +423,9 @@ found_mmu: drm_mm_for_each_node(node, &priv->mm) { if (offset >= node->start && offset < (node->start + node->size)) { - bo = drm_mm_node_to_panfrost_bo(node); - drm_gem_object_get(&bo->base.base); + mapping = drm_mm_node_to_panfrost_mapping(node); + + kref_get(&mapping->refcount); break; } } @@ -427,7 +433,7 @@ found_mmu: spin_unlock(&priv->mm_lock); out: spin_unlock(&pfdev->as_lock); - return bo; + return mapping; } #define NUM_FAULT_PAGES (SZ_2M / PAGE_SIZE) @@ -436,28 +442,30 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as, u64 addr) { int ret, i; + struct panfrost_gem_mapping *bomapping; struct panfrost_gem_object *bo; struct address_space *mapping; pgoff_t page_offset; struct sg_table *sgt; struct page **pages; - bo = addr_to_drm_mm_node(pfdev, as, addr); - if (!bo) + bomapping = addr_to_mapping(pfdev, as, addr); + if (!bomapping) return -ENOENT; + bo = bomapping->obj; if (!bo->is_heap) { dev_WARN(pfdev->dev, "matching BO is not heap type (GPU VA = %llx)", - bo->node.start << PAGE_SHIFT); + bomapping->mmnode.start << PAGE_SHIFT); ret = -EINVAL; goto err_bo; } - WARN_ON(bo->mmu->as != as); + WARN_ON(bomapping->mmu->as != as); /* Assume 2MB alignment and size multiple */ addr &= ~((u64)SZ_2M - 1); page_offset = addr >> PAGE_SHIFT; - page_offset -= bo->node.start; + page_offset -= bomapping->mmnode.start; mutex_lock(&bo->base.pages_lock); @@ -509,13 +517,14 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as, goto err_map; } - mmu_map_sg(pfdev, bo->mmu, addr, IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt); + mmu_map_sg(pfdev, bomapping->mmu, addr, + IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt); - bo->is_mapped = true; + bomapping->active = true; dev_dbg(pfdev->dev, "mapped page fault @ AS%d %llx", as, addr); - drm_gem_object_put_unlocked(&bo->base.base); + panfrost_gem_mapping_put(bomapping); return 0; diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.h b/drivers/gpu/drm/panfrost/panfrost_mmu.h index 7c5b6775ae23..44fc2edf63ce 100644 --- a/drivers/gpu/drm/panfrost/panfrost_mmu.h +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.h @@ -4,12 +4,12 @@ #ifndef __PANFROST_MMU_H__ #define __PANFROST_MMU_H__ -struct panfrost_gem_object; +struct panfrost_gem_mapping; struct panfrost_file_priv; struct panfrost_mmu; -int panfrost_mmu_map(struct panfrost_gem_object *bo); -void panfrost_mmu_unmap(struct panfrost_gem_object *bo); +int panfrost_mmu_map(struct panfrost_gem_mapping *mapping); +void panfrost_mmu_unmap(struct panfrost_gem_mapping *mapping); int panfrost_mmu_init(struct panfrost_device *pfdev); void panfrost_mmu_fini(struct panfrost_device *pfdev); diff --git a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c index 2dba192bf198..684820448be3 100644 --- a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c +++ b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c @@ -25,7 +25,7 @@ #define V4_SHADERS_PER_COREGROUP 4 struct panfrost_perfcnt { - struct panfrost_gem_object *bo; + struct panfrost_gem_mapping *mapping; size_t bosize; void *buf; struct panfrost_file_priv *user; @@ -49,7 +49,7 @@ static int panfrost_perfcnt_dump_locked(struct panfrost_device *pfdev) int ret; reinit_completion(&pfdev->perfcnt->dump_comp); - gpuva = pfdev->perfcnt->bo->node.start << PAGE_SHIFT; + gpuva = pfdev->perfcnt->mapping->mmnode.start << PAGE_SHIFT; gpu_write(pfdev, GPU_PERFCNT_BASE_LO, gpuva); gpu_write(pfdev, GPU_PERFCNT_BASE_HI, gpuva >> 32); gpu_write(pfdev, GPU_INT_CLEAR, @@ -67,9 +67,10 @@ static int panfrost_perfcnt_dump_locked(struct panfrost_device *pfdev) } static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev, - struct panfrost_file_priv *user, + struct drm_file *file_priv, unsigned int counterset) { + struct panfrost_file_priv *user = file_priv->driver_priv; struct panfrost_perfcnt *perfcnt = pfdev->perfcnt; struct drm_gem_shmem_object *bo; u32 cfg; @@ -88,17 +89,22 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev, if (IS_ERR(bo)) return PTR_ERR(bo); - perfcnt->bo = to_panfrost_bo(&bo->base); - /* Map the perfcnt buf in the address space attached to file_priv. */ - ret = panfrost_mmu_map(perfcnt->bo); + ret = panfrost_gem_open(&bo->base, file_priv); if (ret) goto err_put_bo; + perfcnt->mapping = panfrost_gem_mapping_get(to_panfrost_bo(&bo->base), + user); + if (!perfcnt->mapping) { + ret = -EINVAL; + goto err_close_bo; + } + perfcnt->buf = drm_gem_shmem_vmap(&bo->base); if (IS_ERR(perfcnt->buf)) { ret = PTR_ERR(perfcnt->buf); - goto err_put_bo; + goto err_put_mapping; } /* @@ -153,18 +159,26 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev, if (panfrost_has_hw_issue(pfdev, HW_ISSUE_8186)) gpu_write(pfdev, GPU_PRFCNT_TILER_EN, 0xffffffff); + /* The BO ref is retained by the mapping. */ + drm_gem_object_put_unlocked(&bo->base); + return 0; err_vunmap: - drm_gem_shmem_vunmap(&perfcnt->bo->base.base, perfcnt->buf); + drm_gem_shmem_vunmap(&bo->base, perfcnt->buf); +err_put_mapping: + panfrost_gem_mapping_put(perfcnt->mapping); +err_close_bo: + panfrost_gem_close(&bo->base, file_priv); err_put_bo: drm_gem_object_put_unlocked(&bo->base); return ret; } static int panfrost_perfcnt_disable_locked(struct panfrost_device *pfdev, - struct panfrost_file_priv *user) + struct drm_file *file_priv) { + struct panfrost_file_priv *user = file_priv->driver_priv; struct panfrost_perfcnt *perfcnt = pfdev->perfcnt; if (user != perfcnt->user) @@ -178,10 +192,11 @@ static int panfrost_perfcnt_disable_locked(struct panfrost_device *pfdev, GPU_PERFCNT_CFG_MODE(GPU_PERFCNT_CFG_MODE_OFF)); perfcnt->user = NULL; - drm_gem_shmem_vunmap(&perfcnt->bo->base.base, perfcnt->buf); + drm_gem_shmem_vunmap(&perfcnt->mapping->obj->base.base, perfcnt->buf); perfcnt->buf = NULL; - drm_gem_object_put_unlocked(&perfcnt->bo->base.base); - perfcnt->bo = NULL; + panfrost_gem_close(&perfcnt->mapping->obj->base.base, file_priv); + panfrost_gem_mapping_put(perfcnt->mapping); + perfcnt->mapping = NULL; pm_runtime_mark_last_busy(pfdev->dev); pm_runtime_put_autosuspend(pfdev->dev); @@ -191,7 +206,6 @@ static int panfrost_perfcnt_disable_locked(struct panfrost_device *pfdev, int panfrost_ioctl_perfcnt_enable(struct drm_device *dev, void *data, struct drm_file *file_priv) { - struct panfrost_file_priv *pfile = file_priv->driver_priv; struct panfrost_device *pfdev = dev->dev_private; struct panfrost_perfcnt *perfcnt = pfdev->perfcnt; struct drm_panfrost_perfcnt_enable *req = data; @@ -207,10 +221,10 @@ int panfrost_ioctl_perfcnt_enable(struct drm_device *dev, void *data, mutex_lock(&perfcnt->lock); if (req->enable) - ret = panfrost_perfcnt_enable_locked(pfdev, pfile, + ret = panfrost_perfcnt_enable_locked(pfdev, file_priv, req->counterset); else - ret = panfrost_perfcnt_disable_locked(pfdev, pfile); + ret = panfrost_perfcnt_disable_locked(pfdev, file_priv); mutex_unlock(&perfcnt->lock); return ret; @@ -248,15 +262,16 @@ out: return ret; } -void panfrost_perfcnt_close(struct panfrost_file_priv *pfile) +void panfrost_perfcnt_close(struct drm_file *file_priv) { + struct panfrost_file_priv *pfile = file_priv->driver_priv; struct panfrost_device *pfdev = pfile->pfdev; struct panfrost_perfcnt *perfcnt = pfdev->perfcnt; pm_runtime_get_sync(pfdev->dev); mutex_lock(&perfcnt->lock); if (perfcnt->user == pfile) - panfrost_perfcnt_disable_locked(pfdev, pfile); + panfrost_perfcnt_disable_locked(pfdev, file_priv); mutex_unlock(&perfcnt->lock); pm_runtime_mark_last_busy(pfdev->dev); pm_runtime_put_autosuspend(pfdev->dev); diff --git a/drivers/gpu/drm/panfrost/panfrost_perfcnt.h b/drivers/gpu/drm/panfrost/panfrost_perfcnt.h index 13b8fdaa1b43..8bbcf5f5fb33 100644 --- a/drivers/gpu/drm/panfrost/panfrost_perfcnt.h +++ b/drivers/gpu/drm/panfrost/panfrost_perfcnt.h @@ -9,7 +9,7 @@ void panfrost_perfcnt_sample_done(struct panfrost_device *pfdev); void panfrost_perfcnt_clean_cache_done(struct panfrost_device *pfdev); int panfrost_perfcnt_init(struct panfrost_device *pfdev); void panfrost_perfcnt_fini(struct panfrost_device *pfdev); -void panfrost_perfcnt_close(struct panfrost_file_priv *pfile); +void panfrost_perfcnt_close(struct drm_file *file_priv); int panfrost_ioctl_perfcnt_enable(struct drm_device *dev, void *data, struct drm_file *file_priv); int panfrost_ioctl_perfcnt_dump(struct drm_device *dev, void *data, diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index 098bc9f40b98..15bf8a207cb0 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -443,7 +443,7 @@ static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_ mem->bus.size); else mem->bus.addr = - ioremap_nocache(mem->bus.base + mem->bus.offset, + ioremap(mem->bus.base + mem->bus.offset, mem->bus.size); if (!mem->bus.addr) return -ENOMEM; diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.h b/drivers/gpu/drm/rockchip/cdn-dp-core.h index 83c4586665b4..81ac9b658a70 100644 --- a/drivers/gpu/drm/rockchip/cdn-dp-core.h +++ b/drivers/gpu/drm/rockchip/cdn-dp-core.h @@ -95,7 +95,7 @@ struct cdn_dp_device { struct cdn_dp_port *port[MAX_PHY]; u8 ports; u8 max_lanes; - u8 max_rate; + unsigned int max_rate; u8 lanes; int active_port; diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c index 68289b0b063a..68261c7f8c5f 100644 --- a/drivers/gpu/drm/sti/sti_dvo.c +++ b/drivers/gpu/drm/sti/sti_dvo.c @@ -534,7 +534,7 @@ static int sti_dvo_probe(struct platform_device *pdev) DRM_ERROR("Invalid dvo resource\n"); return -ENOMEM; } - dvo->regs = devm_ioremap_nocache(dev, res->start, + dvo->regs = devm_ioremap(dev, res->start, resource_size(res)); if (!dvo->regs) return -ENOMEM; diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c index 8f7bf33815fd..2bb32009d117 100644 --- a/drivers/gpu/drm/sti/sti_hda.c +++ b/drivers/gpu/drm/sti/sti_hda.c @@ -759,14 +759,14 @@ static int sti_hda_probe(struct platform_device *pdev) DRM_ERROR("Invalid hda resource\n"); return -ENOMEM; } - hda->regs = devm_ioremap_nocache(dev, res->start, resource_size(res)); + hda->regs = devm_ioremap(dev, res->start, resource_size(res)); if (!hda->regs) return -ENOMEM; res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "video-dacs-ctrl"); if (res) { - hda->video_dacs_ctrl = devm_ioremap_nocache(dev, res->start, + hda->video_dacs_ctrl = devm_ioremap(dev, res->start, resource_size(res)); if (!hda->video_dacs_ctrl) return -ENOMEM; diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c index 814560ead4e1..64ed102033c8 100644 --- a/drivers/gpu/drm/sti/sti_hdmi.c +++ b/drivers/gpu/drm/sti/sti_hdmi.c @@ -1393,7 +1393,7 @@ static int sti_hdmi_probe(struct platform_device *pdev) ret = -ENOMEM; goto release_adapter; } - hdmi->regs = devm_ioremap_nocache(dev, res->start, resource_size(res)); + hdmi->regs = devm_ioremap(dev, res->start, resource_size(res)); if (!hdmi->regs) { ret = -ENOMEM; goto release_adapter; diff --git a/drivers/gpu/drm/sti/sti_tvout.c b/drivers/gpu/drm/sti/sti_tvout.c index 5767e93dd1cd..c36a8da373cb 100644 --- a/drivers/gpu/drm/sti/sti_tvout.c +++ b/drivers/gpu/drm/sti/sti_tvout.c @@ -860,7 +860,7 @@ static int sti_tvout_probe(struct platform_device *pdev) DRM_ERROR("Invalid glue resource\n"); return -ENOMEM; } - tvout->regs = devm_ioremap_nocache(dev, res->start, resource_size(res)); + tvout->regs = devm_ioremap(dev, res->start, resource_size(res)); if (!tvout->regs) return -ENOMEM; diff --git a/drivers/gpu/drm/sti/sti_vtg.c b/drivers/gpu/drm/sti/sti_vtg.c index 0b17ac8a3faa..5e5f82b6a5d9 100644 --- a/drivers/gpu/drm/sti/sti_vtg.c +++ b/drivers/gpu/drm/sti/sti_vtg.c @@ -393,7 +393,7 @@ static int vtg_probe(struct platform_device *pdev) DRM_ERROR("Get memory resource failed\n"); return -ENOMEM; } - vtg->regs = devm_ioremap_nocache(dev, res->start, resource_size(res)); + vtg->regs = devm_ioremap(dev, res->start, resource_size(res)); if (!vtg->regs) { DRM_ERROR("failed to remap I/O memory\n"); return -ENOMEM; diff --git a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c index a7c4654445c7..68d4644ac2dc 100644 --- a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c +++ b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c @@ -685,8 +685,6 @@ static void sun4i_hdmi_unbind(struct device *dev, struct device *master, struct sun4i_hdmi *hdmi = dev_get_drvdata(dev); cec_unregister_adapter(hdmi->cec_adap); - drm_connector_cleanup(&hdmi->connector); - drm_encoder_cleanup(&hdmi->encoder); i2c_del_adapter(hdmi->i2c); i2c_put_adapter(hdmi->ddc_i2c); clk_disable_unprepare(hdmi->mod_clk); diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.c b/drivers/gpu/drm/sun4i/sun4i_tcon.c index 42651d737c55..c81cdce6ed55 100644 --- a/drivers/gpu/drm/sun4i/sun4i_tcon.c +++ b/drivers/gpu/drm/sun4i/sun4i_tcon.c @@ -489,7 +489,7 @@ static void sun4i_tcon0_mode_set_rgb(struct sun4i_tcon *tcon, WARN_ON(!tcon->quirks->has_channel_0); - tcon->dclk_min_div = 1; + tcon->dclk_min_div = tcon->quirks->dclk_min_div; tcon->dclk_max_div = 127; sun4i_tcon0_mode_set_common(tcon, mode); @@ -1426,12 +1426,14 @@ static int sun8i_r40_tcon_tv_set_mux(struct sun4i_tcon *tcon, static const struct sun4i_tcon_quirks sun4i_a10_quirks = { .has_channel_0 = true, .has_channel_1 = true, + .dclk_min_div = 4, .set_mux = sun4i_a10_tcon_set_mux, }; static const struct sun4i_tcon_quirks sun5i_a13_quirks = { .has_channel_0 = true, .has_channel_1 = true, + .dclk_min_div = 4, .set_mux = sun5i_a13_tcon_set_mux, }; @@ -1440,6 +1442,7 @@ static const struct sun4i_tcon_quirks sun6i_a31_quirks = { .has_channel_1 = true, .has_lvds_alt = true, .needs_de_be_mux = true, + .dclk_min_div = 1, .set_mux = sun6i_tcon_set_mux, }; @@ -1447,11 +1450,13 @@ static const struct sun4i_tcon_quirks sun6i_a31s_quirks = { .has_channel_0 = true, .has_channel_1 = true, .needs_de_be_mux = true, + .dclk_min_div = 1, }; static const struct sun4i_tcon_quirks sun7i_a20_quirks = { .has_channel_0 = true, .has_channel_1 = true, + .dclk_min_div = 4, /* Same display pipeline structure as A10 */ .set_mux = sun4i_a10_tcon_set_mux, }; @@ -1459,11 +1464,13 @@ static const struct sun4i_tcon_quirks sun7i_a20_quirks = { static const struct sun4i_tcon_quirks sun8i_a33_quirks = { .has_channel_0 = true, .has_lvds_alt = true, + .dclk_min_div = 1, }; static const struct sun4i_tcon_quirks sun8i_a83t_lcd_quirks = { .supports_lvds = true, .has_channel_0 = true, + .dclk_min_div = 1, }; static const struct sun4i_tcon_quirks sun8i_a83t_tv_quirks = { @@ -1477,11 +1484,13 @@ static const struct sun4i_tcon_quirks sun8i_r40_tv_quirks = { static const struct sun4i_tcon_quirks sun8i_v3s_quirks = { .has_channel_0 = true, + .dclk_min_div = 1, }; static const struct sun4i_tcon_quirks sun9i_a80_tcon_lcd_quirks = { - .has_channel_0 = true, - .needs_edp_reset = true, + .has_channel_0 = true, + .needs_edp_reset = true, + .dclk_min_div = 1, }; static const struct sun4i_tcon_quirks sun9i_a80_tcon_tv_quirks = { diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.h b/drivers/gpu/drm/sun4i/sun4i_tcon.h index f9f1fe80b206..a62ec826ae71 100644 --- a/drivers/gpu/drm/sun4i/sun4i_tcon.h +++ b/drivers/gpu/drm/sun4i/sun4i_tcon.h @@ -224,6 +224,7 @@ struct sun4i_tcon_quirks { bool needs_de_be_mux; /* sun6i needs mux to select backend */ bool needs_edp_reset; /* a80 edp reset needed for tcon0 access */ bool supports_lvds; /* Does the TCON support an LVDS output? */ + u8 dclk_min_div; /* minimum divider for TCON0 DCLK */ /* callback to handle tcon muxing options */ int (*set_mux)(struct sun4i_tcon *, const struct drm_encoder *); diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c index 2a9e67597375..a3612369750f 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c @@ -256,7 +256,7 @@ static int tilcdc_init(struct drm_driver *ddrv, struct device *dev) goto init_failed; } - priv->mmio = ioremap_nocache(res->start, resource_size(res)); + priv->mmio = ioremap(res->start, resource_size(res)); if (!priv->mmio) { dev_err(dev, "failed to ioremap\n"); ret = -ENOMEM; diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 6b0883a1776e..97fd1dafc3e8 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -218,7 +218,7 @@ static int ttm_mem_reg_ioremap(struct ttm_bo_device *bdev, struct ttm_mem_reg *m if (mem->placement & TTM_PL_FLAG_WC) addr = ioremap_wc(mem->bus.base + mem->bus.offset, mem->bus.size); else - addr = ioremap_nocache(mem->bus.base + mem->bus.offset, mem->bus.size); + addr = ioremap(mem->bus.base + mem->bus.offset, mem->bus.size); if (!addr) { (void) ttm_mem_io_lock(man, false); ttm_mem_io_free(bdev, mem); @@ -565,7 +565,7 @@ static int ttm_bo_ioremap(struct ttm_buffer_object *bo, map->virtual = ioremap_wc(bo->mem.bus.base + bo->mem.bus.offset + offset, size); else - map->virtual = ioremap_nocache(bo->mem.bus.base + bo->mem.bus.offset + offset, + map->virtual = ioremap(bo->mem.bus.base + bo->mem.bus.offset + offset, size); } return (!map->virtual) ? -ENOMEM : 0; diff --git a/drivers/gpu/drm/virtio/virtgpu_plane.c b/drivers/gpu/drm/virtio/virtgpu_plane.c index 390524143139..1635a9ff4794 100644 --- a/drivers/gpu/drm/virtio/virtgpu_plane.c +++ b/drivers/gpu/drm/virtio/virtgpu_plane.c @@ -232,6 +232,7 @@ static void virtio_gpu_cursor_plane_update(struct drm_plane *plane, if (!objs) return; virtio_gpu_array_add_obj(objs, vgfb->base.obj[0]); + virtio_gpu_array_lock_resv(objs); virtio_gpu_cmd_transfer_to_host_2d (vgdev, 0, plane->state->crtc_w, |