diff options
Diffstat (limited to 'drivers/gpu/drm')
140 files changed, 1857 insertions, 1048 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index d0102cfc8efb..b0fc116296cb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -151,6 +151,7 @@ extern int amdgpu_compute_multipipe; extern int amdgpu_gpu_recovery; extern int amdgpu_emu_mode; extern uint amdgpu_smu_memory_pool_size; +extern uint amdgpu_dc_feature_mask; extern struct amdgpu_mgpu_info mgpu_info; #ifdef CONFIG_DRM_AMDGPU_SI @@ -232,7 +233,7 @@ enum amdgpu_kiq_irq { #define MAX_KIQ_REG_WAIT 5000 /* in usecs, 5ms */ #define MAX_KIQ_REG_BAILOUT_INTERVAL 5 /* in msecs, 5ms */ -#define MAX_KIQ_REG_TRY 20 +#define MAX_KIQ_REG_TRY 80 /* 20 -> 80 */ int amdgpu_device_ip_set_clockgating_state(void *dev, enum amd_ip_block_type block_type, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index c31a8849e9f8..1580ec60b89f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -501,8 +501,11 @@ void amdgpu_amdkfd_set_compute_idle(struct kgd_dev *kgd, bool idle) { struct amdgpu_device *adev = (struct amdgpu_device *)kgd; - amdgpu_dpm_switch_power_profile(adev, - PP_SMC_POWER_PROFILE_COMPUTE, !idle); + if (adev->powerplay.pp_funcs && + adev->powerplay.pp_funcs->switch_power_profile) + amdgpu_dpm_switch_power_profile(adev, + PP_SMC_POWER_PROFILE_COMPUTE, + !idle); } bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c index 8816c697b205..387f1cf1dc20 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c @@ -330,7 +330,9 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device, case CHIP_TOPAZ: if (((adev->pdev->device == 0x6900) && (adev->pdev->revision == 0x81)) || ((adev->pdev->device == 0x6900) && (adev->pdev->revision == 0x83)) || - ((adev->pdev->device == 0x6907) && (adev->pdev->revision == 0x87))) { + ((adev->pdev->device == 0x6907) && (adev->pdev->revision == 0x87)) || + ((adev->pdev->device == 0x6900) && (adev->pdev->revision == 0xD1)) || + ((adev->pdev->device == 0x6900) && (adev->pdev->revision == 0xD3))) { info->is_kicker = true; strcpy(fw_name, "amdgpu/topaz_k_smc.bin"); } else @@ -351,7 +353,6 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device, if (type == CGS_UCODE_ID_SMU) { if (((adev->pdev->device == 0x67ef) && ((adev->pdev->revision == 0xe0) || - (adev->pdev->revision == 0xe2) || (adev->pdev->revision == 0xe5))) || ((adev->pdev->device == 0x67ff) && ((adev->pdev->revision == 0xcf) || @@ -359,8 +360,13 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device, (adev->pdev->revision == 0xff)))) { info->is_kicker = true; strcpy(fw_name, "amdgpu/polaris11_k_smc.bin"); - } else + } else if ((adev->pdev->device == 0x67ef) && + (adev->pdev->revision == 0xe2)) { + info->is_kicker = true; + strcpy(fw_name, "amdgpu/polaris11_k2_smc.bin"); + } else { strcpy(fw_name, "amdgpu/polaris11_smc.bin"); + } } else if (type == CGS_UCODE_ID_SMU_SK) { strcpy(fw_name, "amdgpu/polaris11_smc_sk.bin"); } @@ -375,17 +381,35 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device, (adev->pdev->revision == 0xe7) || (adev->pdev->revision == 0xef))) || ((adev->pdev->device == 0x6fdf) && - (adev->pdev->revision == 0xef))) { + ((adev->pdev->revision == 0xef) || + (adev->pdev->revision == 0xff)))) { info->is_kicker = true; strcpy(fw_name, "amdgpu/polaris10_k_smc.bin"); - } else + } else if ((adev->pdev->device == 0x67df) && + ((adev->pdev->revision == 0xe1) || + (adev->pdev->revision == 0xf7))) { + info->is_kicker = true; + strcpy(fw_name, "amdgpu/polaris10_k2_smc.bin"); + } else { strcpy(fw_name, "amdgpu/polaris10_smc.bin"); + } } else if (type == CGS_UCODE_ID_SMU_SK) { strcpy(fw_name, "amdgpu/polaris10_smc_sk.bin"); } break; case CHIP_POLARIS12: - strcpy(fw_name, "amdgpu/polaris12_smc.bin"); + if (((adev->pdev->device == 0x6987) && + ((adev->pdev->revision == 0xc0) || + (adev->pdev->revision == 0xc3))) || + ((adev->pdev->device == 0x6981) && + ((adev->pdev->revision == 0x00) || + (adev->pdev->revision == 0x01) || + (adev->pdev->revision == 0x10)))) { + info->is_kicker = true; + strcpy(fw_name, "amdgpu/polaris12_k_smc.bin"); + } else { + strcpy(fw_name, "amdgpu/polaris12_smc.bin"); + } break; case CHIP_VEGAM: strcpy(fw_name, "amdgpu/vegam_smc.bin"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 663043c8f0f5..0acc8dee2cb8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -124,14 +124,14 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, union drm_amdgpu_cs goto free_chunk; } + mutex_lock(&p->ctx->lock); + /* skip guilty context job */ if (atomic_read(&p->ctx->guilty) == 1) { ret = -ECANCELED; goto free_chunk; } - mutex_lock(&p->ctx->lock); - /* get chunks */ chunk_array_user = u64_to_user_ptr(cs->in.chunks); if (copy_from_user(chunk_array, chunk_array_user, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index f9b54236102d..95f4c4139fc6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -39,6 +39,7 @@ const unsigned int amdgpu_ctx_num_entities[AMDGPU_HW_IP_NUM] = { [AMDGPU_HW_IP_UVD_ENC] = 1, [AMDGPU_HW_IP_VCN_DEC] = 1, [AMDGPU_HW_IP_VCN_ENC] = 1, + [AMDGPU_HW_IP_VCN_JPEG] = 1, }; static int amdgput_ctx_total_num_entities(void) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index 6748cd7fc129..686a26de50f9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -626,6 +626,13 @@ int amdgpu_display_modeset_create_props(struct amdgpu_device *adev) "dither", amdgpu_dither_enum_list, sz); + if (amdgpu_device_has_dc_support(adev)) { + adev->mode_info.max_bpc_property = + drm_property_create_range(adev->ddev, 0, "max bpc", 8, 16); + if (!adev->mode_info.max_bpc_property) + return -ENOMEM; + } + return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 943dbf3c5da1..74b611e8a1b1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -127,6 +127,9 @@ int amdgpu_compute_multipipe = -1; int amdgpu_gpu_recovery = -1; /* auto */ int amdgpu_emu_mode = 0; uint amdgpu_smu_memory_pool_size = 0; +/* FBC (bit 0) disabled by default*/ +uint amdgpu_dc_feature_mask = 0; + struct amdgpu_mgpu_info mgpu_info = { .mutex = __MUTEX_INITIALIZER(mgpu_info.mutex), }; @@ -631,6 +634,14 @@ module_param(halt_if_hws_hang, int, 0644); MODULE_PARM_DESC(halt_if_hws_hang, "Halt if HWS hang is detected (0 = off (default), 1 = on)"); #endif +/** + * DOC: dcfeaturemask (uint) + * Override display features enabled. See enum DC_FEATURE_MASK in drivers/gpu/drm/amd/include/amd_shared.h. + * The default is the current set of stable display features. + */ +MODULE_PARM_DESC(dcfeaturemask, "all stable DC features enabled (default))"); +module_param_named(dcfeaturemask, amdgpu_dc_feature_mask, uint, 0444); + static const struct pci_device_id pciidlist[] = { #ifdef CONFIG_DRM_AMDGPU_SI {0x1002, 0x6780, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI}, @@ -861,7 +872,13 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x6864, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, {0x1002, 0x6867, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, {0x1002, 0x6868, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x6869, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x686a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x686b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, {0x1002, 0x686c, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x686d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x686e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x686f, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, {0x1002, 0x687f, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, /* Vega 12 */ {0x1002, 0x69A0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA12}, @@ -874,6 +891,7 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x66A1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, {0x1002, 0x66A2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, {0x1002, 0x66A3, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, + {0x1002, 0x66A4, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, {0x1002, 0x66A7, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, {0x1002, 0x66AF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, /* Raven */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 81732a84c2ab..8f3d44e5e787 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -467,9 +467,6 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file if (!info->return_size || !info->return_pointer) return -EINVAL; - /* Ensure IB tests are run on ring */ - flush_delayed_work(&adev->late_init_work); - switch (info->query) { case AMDGPU_INFO_ACCEL_WORKING: ui32 = adev->accel_working; @@ -950,6 +947,9 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) struct amdgpu_fpriv *fpriv; int r, pasid; + /* Ensure IB tests are run on ring */ + flush_delayed_work(&adev->late_init_work); + file_priv->driver_priv = NULL; r = pm_runtime_get_sync(dev->dev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h index b9e9e8b02fb7..d1b4d9b6aae0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h @@ -339,6 +339,8 @@ struct amdgpu_mode_info { struct drm_property *audio_property; /* FMT dithering */ struct drm_property *dither_property; + /* maximum number of bits per channel for monitor color */ + struct drm_property *max_bpc_property; /* hardcoded DFP edid from BIOS */ struct edid *bios_hardcoded_edid; int bios_hardcoded_edid_size; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 352b30409060..0877ff9a9594 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -181,7 +181,7 @@ static unsigned amdgpu_vm_num_entries(struct amdgpu_device *adev, if (level == adev->vm_manager.root_level) /* For the root directory */ - return round_up(adev->vm_manager.max_pfn, 1 << shift) >> shift; + return round_up(adev->vm_manager.max_pfn, 1ULL << shift) >> shift; else if (level != AMDGPU_VM_PTB) /* Everything in between */ return 512; @@ -1632,13 +1632,6 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, continue; } - /* First check if the entry is already handled */ - if (cursor.pfn < frag_start) { - cursor.entry->huge = true; - amdgpu_vm_pt_next(adev, &cursor); - continue; - } - /* If it isn't already handled it can't be a huge page */ if (cursor.entry->huge) { /* Add the entry to the relocated list to update it. */ @@ -1663,9 +1656,11 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, if (!amdgpu_vm_pt_descendant(adev, &cursor)) return -ENOENT; continue; - } else if (frag >= parent_shift) { + } else if (frag >= parent_shift && + cursor.level - 1 != adev->vm_manager.root_level) { /* If the fragment size is even larger than the parent - * shift we should go up one level and check it again. + * shift we should go up one level and check it again + * unless one level up is the root level. */ if (!amdgpu_vm_pt_ancestor(&cursor)) return -ENOENT; @@ -1673,10 +1668,10 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, } /* Looks good so far, calculate parameters for the update */ - incr = AMDGPU_GPU_PAGE_SIZE << shift; + incr = (uint64_t)AMDGPU_GPU_PAGE_SIZE << shift; mask = amdgpu_vm_entries_mask(adev, cursor.level); pe_start = ((cursor.pfn >> shift) & mask) * 8; - entry_end = (mask + 1) << shift; + entry_end = (uint64_t)(mask + 1) << shift; entry_end += cursor.pfn & ~(entry_end - 1); entry_end = min(entry_end, end); @@ -1689,7 +1684,7 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, flags | AMDGPU_PTE_FRAG(frag)); pe_start += nptes * 8; - dst += nptes * AMDGPU_GPU_PAGE_SIZE << shift; + dst += (uint64_t)nptes * AMDGPU_GPU_PAGE_SIZE << shift; frag_start = upd_end; if (frag_start >= frag_end) { @@ -1701,8 +1696,17 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, } } while (frag_start < entry_end); - if (frag >= shift) + if (amdgpu_vm_pt_descendant(adev, &cursor)) { + /* Mark all child entries as huge */ + while (cursor.pfn < frag_start) { + cursor.entry->huge = true; + amdgpu_vm_pt_next(adev, &cursor); + } + + } else if (frag >= shift) { + /* or just move on to the next on the same level. */ amdgpu_vm_pt_next(adev, &cursor); + } } return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 6d7baf59d6e1..21363b2b2ee5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -2440,12 +2440,13 @@ static void gfx_v9_0_rlc_start(struct amdgpu_device *adev) #endif WREG32_FIELD15(GC, 0, RLC_CNTL, RLC_ENABLE_F32, 1); + udelay(50); /* carrizo do enable cp interrupt after cp inited */ - if (!(adev->flags & AMD_IS_APU)) + if (!(adev->flags & AMD_IS_APU)) { gfx_v9_0_enable_gui_idle_interrupt(adev, true); - - udelay(50); + udelay(50); + } #ifdef AMDGPU_RLC_DEBUG_RETRY /* RLC_GPM_GENERAL_6 : RLC Ucode version */ diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c index ceb7847b504f..bfa317ad20a9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c @@ -72,7 +72,7 @@ static void gfxhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev) /* Program the system aperture low logical page number. */ WREG32_SOC15(GC, 0, mmMC_VM_SYSTEM_APERTURE_LOW_ADDR, - min(adev->gmc.vram_start, adev->gmc.agp_start) >> 18); + min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); if (adev->asic_type == CHIP_RAVEN && adev->rev_id >= 0x8) /* @@ -82,11 +82,11 @@ static void gfxhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev) * to get rid of the VM fault and hardware hang. */ WREG32_SOC15(GC, 0, mmMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - max((adev->gmc.vram_end >> 18) + 0x1, + max((adev->gmc.fb_end >> 18) + 0x1, adev->gmc.agp_end >> 18)); else WREG32_SOC15(GC, 0, mmMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - max(adev->gmc.vram_end, adev->gmc.agp_end) >> 18); + max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18); /* Set default page address. */ value = adev->vram_scratch.gpu_addr - adev->gmc.vram_start diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index e1c2b4e9c7b2..73ad02aea2b2 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -46,6 +46,7 @@ MODULE_FIRMWARE("amdgpu/tahiti_mc.bin"); MODULE_FIRMWARE("amdgpu/pitcairn_mc.bin"); MODULE_FIRMWARE("amdgpu/verde_mc.bin"); MODULE_FIRMWARE("amdgpu/oland_mc.bin"); +MODULE_FIRMWARE("amdgpu/hainan_mc.bin"); MODULE_FIRMWARE("amdgpu/si58_mc.bin"); #define MC_SEQ_MISC0__MT__MASK 0xf0000000 diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 1d3265c97b70..747c068379dc 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -56,6 +56,9 @@ MODULE_FIRMWARE("amdgpu/tonga_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris11_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris10_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris12_mc.bin"); +MODULE_FIRMWARE("amdgpu/polaris11_k_mc.bin"); +MODULE_FIRMWARE("amdgpu/polaris10_k_mc.bin"); +MODULE_FIRMWARE("amdgpu/polaris12_k_mc.bin"); static const u32 golden_settings_tonga_a11[] = { @@ -224,13 +227,39 @@ static int gmc_v8_0_init_microcode(struct amdgpu_device *adev) chip_name = "tonga"; break; case CHIP_POLARIS11: - chip_name = "polaris11"; + if (((adev->pdev->device == 0x67ef) && + ((adev->pdev->revision == 0xe0) || + (adev->pdev->revision == 0xe5))) || + ((adev->pdev->device == 0x67ff) && + ((adev->pdev->revision == 0xcf) || + (adev->pdev->revision == 0xef) || + (adev->pdev->revision == 0xff)))) + chip_name = "polaris11_k"; + else if ((adev->pdev->device == 0x67ef) && + (adev->pdev->revision == 0xe2)) + chip_name = "polaris11_k"; + else + chip_name = "polaris11"; break; case CHIP_POLARIS10: - chip_name = "polaris10"; + if ((adev->pdev->device == 0x67df) && + ((adev->pdev->revision == 0xe1) || + (adev->pdev->revision == 0xf7))) + chip_name = "polaris10_k"; + else + chip_name = "polaris10"; break; case CHIP_POLARIS12: - chip_name = "polaris12"; + if (((adev->pdev->device == 0x6987) && + ((adev->pdev->revision == 0xc0) || + (adev->pdev->revision == 0xc3))) || + ((adev->pdev->device == 0x6981) && + ((adev->pdev->revision == 0x00) || + (adev->pdev->revision == 0x01) || + (adev->pdev->revision == 0x10)))) + chip_name = "polaris12_k"; + else + chip_name = "polaris12"; break; case CHIP_FIJI: case CHIP_CARRIZO: @@ -337,7 +366,7 @@ static int gmc_v8_0_polaris_mc_load_microcode(struct amdgpu_device *adev) const struct mc_firmware_header_v1_0 *hdr; const __le32 *fw_data = NULL; const __le32 *io_mc_regs = NULL; - u32 data, vbios_version; + u32 data; int i, ucode_size, regs_size; /* Skip MC ucode loading on SR-IOV capable boards. @@ -348,13 +377,6 @@ static int gmc_v8_0_polaris_mc_load_microcode(struct amdgpu_device *adev) if (amdgpu_sriov_bios(adev)) return 0; - WREG32(mmMC_SEQ_IO_DEBUG_INDEX, 0x9F); - data = RREG32(mmMC_SEQ_IO_DEBUG_DATA); - vbios_version = data & 0xf; - - if (vbios_version == 0) - return 0; - if (!adev->gmc.fw) return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index fd23ba1226a5..a0db67adc34c 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -90,7 +90,7 @@ static void mmhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev) /* Program the system aperture low logical page number. */ WREG32_SOC15(MMHUB, 0, mmMC_VM_SYSTEM_APERTURE_LOW_ADDR, - min(adev->gmc.vram_start, adev->gmc.agp_start) >> 18); + min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); if (adev->asic_type == CHIP_RAVEN && adev->rev_id >= 0x8) /* @@ -100,11 +100,11 @@ static void mmhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev) * to get rid of the VM fault and hardware hang. */ WREG32_SOC15(MMHUB, 0, mmMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - max((adev->gmc.vram_end >> 18) + 0x1, + max((adev->gmc.fb_end >> 18) + 0x1, adev->gmc.agp_end >> 18)); else WREG32_SOC15(MMHUB, 0, mmMC_VM_SYSTEM_APERTURE_HIGH_ADDR, - max(adev->gmc.vram_end, adev->gmc.agp_end) >> 18); + max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18); /* Set default page address. */ value = adev->vram_scratch.gpu_addr - adev->gmc.vram_start + diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index bf5e6a413dee..4cc0dcb1a187 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -65,6 +65,13 @@ #define mmMP0_MISC_LIGHT_SLEEP_CTRL 0x01ba #define mmMP0_MISC_LIGHT_SLEEP_CTRL_BASE_IDX 0 +/* for Vega20 register name change */ +#define mmHDP_MEM_POWER_CTRL 0x00d4 +#define HDP_MEM_POWER_CTRL__IPH_MEM_POWER_CTRL_EN_MASK 0x00000001L +#define HDP_MEM_POWER_CTRL__IPH_MEM_POWER_LS_EN_MASK 0x00000002L +#define HDP_MEM_POWER_CTRL__RC_MEM_POWER_CTRL_EN_MASK 0x00010000L +#define HDP_MEM_POWER_CTRL__RC_MEM_POWER_LS_EN_MASK 0x00020000L +#define mmHDP_MEM_POWER_CTRL_BASE_IDX 0 /* * Indirect registers accessor */ @@ -870,15 +877,33 @@ static void soc15_update_hdp_light_sleep(struct amdgpu_device *adev, bool enable { uint32_t def, data; - def = data = RREG32(SOC15_REG_OFFSET(HDP, 0, mmHDP_MEM_POWER_LS)); + if (adev->asic_type == CHIP_VEGA20) { + def = data = RREG32(SOC15_REG_OFFSET(HDP, 0, mmHDP_MEM_POWER_CTRL)); - if (enable && (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS)) - data |= HDP_MEM_POWER_LS__LS_ENABLE_MASK; - else - data &= ~HDP_MEM_POWER_LS__LS_ENABLE_MASK; + if (enable && (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS)) + data |= HDP_MEM_POWER_CTRL__IPH_MEM_POWER_CTRL_EN_MASK | + HDP_MEM_POWER_CTRL__IPH_MEM_POWER_LS_EN_MASK | + HDP_MEM_POWER_CTRL__RC_MEM_POWER_CTRL_EN_MASK | + HDP_MEM_POWER_CTRL__RC_MEM_POWER_LS_EN_MASK; + else + data &= ~(HDP_MEM_POWER_CTRL__IPH_MEM_POWER_CTRL_EN_MASK | + HDP_MEM_POWER_CTRL__IPH_MEM_POWER_LS_EN_MASK | + HDP_MEM_POWER_CTRL__RC_MEM_POWER_CTRL_EN_MASK | + HDP_MEM_POWER_CTRL__RC_MEM_POWER_LS_EN_MASK); - if (def != data) - WREG32(SOC15_REG_OFFSET(HDP, 0, mmHDP_MEM_POWER_LS), data); + if (def != data) + WREG32(SOC15_REG_OFFSET(HDP, 0, mmHDP_MEM_POWER_CTRL), data); + } else { + def = data = RREG32(SOC15_REG_OFFSET(HDP, 0, mmHDP_MEM_POWER_LS)); + + if (enable && (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS)) + data |= HDP_MEM_POWER_LS__LS_ENABLE_MASK; + else + data &= ~HDP_MEM_POWER_LS__LS_ENABLE_MASK; + + if (def != data) + WREG32(SOC15_REG_OFFSET(HDP, 0, mmHDP_MEM_POWER_LS), data); + } } static void soc15_update_drm_clock_gating(struct amdgpu_device *adev, bool enable) diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index eae90922fdbe..322e09b5b448 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -48,6 +48,7 @@ static void vcn_v1_0_set_enc_ring_funcs(struct amdgpu_device *adev); static void vcn_v1_0_set_jpeg_ring_funcs(struct amdgpu_device *adev); static void vcn_v1_0_set_irq_funcs(struct amdgpu_device *adev); static void vcn_v1_0_jpeg_ring_set_patch_ring(struct amdgpu_ring *ring, uint32_t ptr); +static int vcn_v1_0_set_powergating_state(void *handle, enum amd_powergating_state state); /** * vcn_v1_0_early_init - set function pointers @@ -222,7 +223,7 @@ static int vcn_v1_0_hw_fini(void *handle) struct amdgpu_ring *ring = &adev->vcn.ring_dec; if (RREG32_SOC15(VCN, 0, mmUVD_STATUS)) - vcn_v1_0_stop(adev); + vcn_v1_0_set_powergating_state(adev, AMD_PG_STATE_GATE); ring->ready = false; diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c index a99f71797aa3..a0fda6f9252a 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c @@ -129,7 +129,7 @@ static int vega10_ih_irq_init(struct amdgpu_device *adev) else wptr_off = adev->wb.gpu_addr + (adev->irq.ih.wptr_offs * 4); WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_LO, lower_32_bits(wptr_off)); - WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_HI, upper_32_bits(wptr_off) & 0xFF); + WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_HI, upper_32_bits(wptr_off) & 0xFFFF); /* set rptr, wptr to 0 */ WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c index 2d4473557b0d..d13fc4fcb517 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c +++ b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c @@ -49,6 +49,7 @@ int vega20_reg_base_init(struct amdgpu_device *adev) adev->reg_offset[SMUIO_HWIP][i] = (uint32_t *)(&(SMUIO_BASE.instance[i])); adev->reg_offset[NBIF_HWIP][i] = (uint32_t *)(&(NBIO_BASE.instance[i])); adev->reg_offset[THM_HWIP][i] = (uint32_t *)(&(THM_BASE.instance[i])); + adev->reg_offset[CLK_HWIP][i] = (uint32_t *)(&(CLK_BASE.instance[i])); } return 0; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index a9f18ea7e354..e4ded890b1cb 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -337,12 +337,19 @@ static const struct kfd_deviceid supported_devices[] = { { 0x6864, &vega10_device_info }, /* Vega10 */ { 0x6867, &vega10_device_info }, /* Vega10 */ { 0x6868, &vega10_device_info }, /* Vega10 */ + { 0x6869, &vega10_device_info }, /* Vega10 */ + { 0x686A, &vega10_device_info }, /* Vega10 */ + { 0x686B, &vega10_device_info }, /* Vega10 */ { 0x686C, &vega10_vf_device_info }, /* Vega10 vf*/ + { 0x686D, &vega10_device_info }, /* Vega10 */ + { 0x686E, &vega10_device_info }, /* Vega10 */ + { 0x686F, &vega10_device_info }, /* Vega10 */ { 0x687F, &vega10_device_info }, /* Vega10 */ { 0x66a0, &vega20_device_info }, /* Vega20 */ { 0x66a1, &vega20_device_info }, /* Vega20 */ { 0x66a2, &vega20_device_info }, /* Vega20 */ { 0x66a3, &vega20_device_info }, /* Vega20 */ + { 0x66a4, &vega20_device_info }, /* Vega20 */ { 0x66a7, &vega20_device_info }, /* Vega20 */ { 0x66af, &vega20_device_info } /* Vega20 */ }; 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 b0df6dc9a775..5a6edf65c9ea 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -429,6 +429,9 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) adev->asic_type < CHIP_RAVEN) init_data.flags.gpu_vm_support = true; + if (amdgpu_dc_feature_mask & DC_FBC_MASK) + init_data.flags.fbc_support = true; + /* Display Core create. */ adev->dm.dc = dc_create(&init_data); @@ -1524,13 +1527,6 @@ static int amdgpu_dm_backlight_update_status(struct backlight_device *bd) { struct amdgpu_display_manager *dm = bl_get_data(bd); - /* - * PWM interperts 0 as 100% rather than 0% because of HW - * limitation for level 0.So limiting minimum brightness level - * to 1. - */ - if (bd->props.brightness < 1) - return 1; if (dc_link_set_backlight_level(dm->backlight_link, bd->props.brightness, 0, 0)) return 0; @@ -2362,8 +2358,15 @@ static void update_stream_scaling_settings(const struct drm_display_mode *mode, static enum dc_color_depth convert_color_depth_from_display_info(const struct drm_connector *connector) { + struct dm_connector_state *dm_conn_state = + to_dm_connector_state(connector->state); uint32_t bpc = connector->display_info.bpc; + /* TODO: Remove this when there's support for max_bpc in drm */ + if (dm_conn_state && bpc > dm_conn_state->max_bpc) + /* Round down to nearest even number. */ + bpc = dm_conn_state->max_bpc - (dm_conn_state->max_bpc & 1); + switch (bpc) { case 0: /* @@ -2551,9 +2554,9 @@ static void fill_audio_info(struct audio_info *audio_info, cea_revision = drm_connector->display_info.cea_rev; - strncpy(audio_info->display_name, + strscpy(audio_info->display_name, edid_caps->display_name, - AUDIO_INFO_DISPLAY_NAME_SIZE_IN_CHARS - 1); + AUDIO_INFO_DISPLAY_NAME_SIZE_IN_CHARS); if (cea_revision >= 3) { audio_info->mode_count = edid_caps->audio_mode_count; @@ -2707,18 +2710,11 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, drm_connector = &aconnector->base; if (!aconnector->dc_sink) { - /* - * Create dc_sink when necessary to MST - * Don't apply fake_sink to MST - */ - if (aconnector->mst_port) { - dm_dp_mst_dc_sink_create(drm_connector); - return stream; + if (!aconnector->mst_port) { + sink = create_fake_sink(aconnector); + if (!sink) + return stream; } - - sink = create_fake_sink(aconnector); - if (!sink) - return stream; } else { sink = aconnector->dc_sink; } @@ -2954,6 +2950,9 @@ int amdgpu_dm_connector_atomic_set_property(struct drm_connector *connector, } else if (property == adev->mode_info.underscan_property) { dm_new_state->underscan_enable = val; ret = 0; + } else if (property == adev->mode_info.max_bpc_property) { + dm_new_state->max_bpc = val; + ret = 0; } return ret; @@ -2996,6 +2995,9 @@ int amdgpu_dm_connector_atomic_get_property(struct drm_connector *connector, } else if (property == adev->mode_info.underscan_property) { *val = dm_state->underscan_enable; ret = 0; + } else if (property == adev->mode_info.max_bpc_property) { + *val = dm_state->max_bpc; + ret = 0; } return ret; } @@ -3040,6 +3042,7 @@ void amdgpu_dm_connector_funcs_reset(struct drm_connector *connector) state->underscan_enable = false; state->underscan_hborder = 0; state->underscan_vborder = 0; + state->max_bpc = 8; __drm_atomic_helper_connector_reset(connector, &state->base); } @@ -3061,6 +3064,7 @@ amdgpu_dm_connector_atomic_duplicate_state(struct drm_connector *connector) new_state->freesync_capable = state->freesync_capable; new_state->freesync_enable = state->freesync_enable; + new_state->max_bpc = state->max_bpc; return &new_state->base; } @@ -3308,7 +3312,7 @@ void dm_drm_plane_destroy_state(struct drm_plane *plane, static const struct drm_plane_funcs dm_plane_funcs = { .update_plane = drm_atomic_helper_update_plane, .disable_plane = drm_atomic_helper_disable_plane, - .destroy = drm_plane_cleanup, + .destroy = drm_primary_helper_destroy, .reset = dm_drm_plane_reset, .atomic_duplicate_state = dm_drm_plane_duplicate_state, .atomic_destroy_state = dm_drm_plane_destroy_state, @@ -3648,7 +3652,7 @@ amdgpu_dm_create_common_mode(struct drm_encoder *encoder, mode->hdisplay = hdisplay; mode->vdisplay = vdisplay; mode->type &= ~DRM_MODE_TYPE_PREFERRED; - strncpy(mode->name, name, DRM_DISPLAY_MODE_LEN); + strscpy(mode->name, name, DRM_DISPLAY_MODE_LEN); return mode; @@ -3806,6 +3810,9 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, drm_object_attach_property(&aconnector->base.base, adev->mode_info.underscan_vborder_property, 0); + drm_object_attach_property(&aconnector->base.base, + adev->mode_info.max_bpc_property, + 0); } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 978b34a5011c..6e069d777ab2 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -160,8 +160,6 @@ struct amdgpu_dm_connector { struct mutex hpd_lock; bool fake_enable; - - bool mst_connected; }; #define to_amdgpu_dm_connector(x) container_of(x, struct amdgpu_dm_connector, base) @@ -206,6 +204,7 @@ struct dm_connector_state { enum amdgpu_rmx_type scaling; uint8_t underscan_vborder; uint8_t underscan_hborder; + uint8_t max_bpc; bool underscan_enable; bool freesync_enable; bool freesync_capable; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 03601d717fed..1b0d209d8367 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -205,40 +205,6 @@ static const struct drm_connector_funcs dm_dp_mst_connector_funcs = { .atomic_get_property = amdgpu_dm_connector_atomic_get_property }; -void dm_dp_mst_dc_sink_create(struct drm_connector *connector) -{ - struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); - struct dc_sink *dc_sink; - struct dc_sink_init_data init_params = { - .link = aconnector->dc_link, - .sink_signal = SIGNAL_TYPE_DISPLAY_PORT_MST }; - - /* FIXME none of this is safe. we shouldn't touch aconnector here in - * atomic_check - */ - - /* - * TODO: Need to further figure out why ddc.algo is NULL while MST port exists - */ - if (!aconnector->port || !aconnector->port->aux.ddc.algo) - return; - - ASSERT(aconnector->edid); - - dc_sink = dc_link_add_remote_sink( - aconnector->dc_link, - (uint8_t *)aconnector->edid, - (aconnector->edid->extensions + 1) * EDID_LENGTH, - &init_params); - - dc_sink->priv = aconnector; - aconnector->dc_sink = dc_sink; - - if (aconnector->dc_sink) - amdgpu_dm_update_freesync_caps( - connector, aconnector->edid); -} - static int dm_dp_mst_get_modes(struct drm_connector *connector) { struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); @@ -319,12 +285,7 @@ dm_dp_create_fake_mst_encoder(struct amdgpu_dm_connector *connector) struct amdgpu_device *adev = dev->dev_private; struct amdgpu_encoder *amdgpu_encoder; struct drm_encoder *encoder; - const struct drm_connector_helper_funcs *connector_funcs = - connector->base.helper_private; - struct drm_encoder *enc_master = - connector_funcs->best_encoder(&connector->base); - DRM_DEBUG_KMS("enc master is %p\n", enc_master); amdgpu_encoder = kzalloc(sizeof(*amdgpu_encoder), GFP_KERNEL); if (!amdgpu_encoder) return NULL; @@ -354,25 +315,6 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr, struct amdgpu_device *adev = dev->dev_private; struct amdgpu_dm_connector *aconnector; struct drm_connector *connector; - struct drm_connector_list_iter conn_iter; - - drm_connector_list_iter_begin(dev, &conn_iter); - drm_for_each_connector_iter(connector, &conn_iter) { - aconnector = to_amdgpu_dm_connector(connector); - if (aconnector->mst_port == master - && !aconnector->port) { - DRM_INFO("DM_MST: reusing connector: %p [id: %d] [master: %p]\n", - aconnector, connector->base.id, aconnector->mst_port); - - aconnector->port = port; - drm_connector_set_path_property(connector, pathprop); - - drm_connector_list_iter_end(&conn_iter); - aconnector->mst_connected = true; - return &aconnector->base; - } - } - drm_connector_list_iter_end(&conn_iter); aconnector = kzalloc(sizeof(*aconnector), GFP_KERNEL); if (!aconnector) @@ -400,10 +342,9 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr, master->connector_id); aconnector->mst_encoder = dm_dp_create_fake_mst_encoder(master); + drm_connector_attach_encoder(&aconnector->base, + &aconnector->mst_encoder->base); - /* - * TODO: understand why this one is needed - */ drm_object_attach_property( &connector->base, dev->mode_config.path_property, @@ -421,8 +362,6 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr, */ amdgpu_dm_connector_funcs_reset(connector); - aconnector->mst_connected = true; - DRM_INFO("DM_MST: added connector: %p [id: %d] [master: %p]\n", aconnector, connector->base.id, aconnector->mst_port); @@ -434,6 +373,9 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr, static void dm_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, struct drm_connector *connector) { + struct amdgpu_dm_connector *master = container_of(mgr, struct amdgpu_dm_connector, mst_mgr); + struct drm_device *dev = master->base.dev; + struct amdgpu_device *adev = dev->dev_private; struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); DRM_INFO("DM_MST: Disabling connector: %p [id: %d] [master: %p]\n", @@ -447,7 +389,10 @@ static void dm_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, aconnector->dc_sink = NULL; } - aconnector->mst_connected = false; + drm_connector_unregister(connector); + if (adev->mode_info.rfbdev) + drm_fb_helper_remove_one_connector(&adev->mode_info.rfbdev->helper, connector); + drm_connector_put(connector); } static void dm_dp_mst_hotplug(struct drm_dp_mst_topology_mgr *mgr) @@ -458,18 +403,10 @@ static void dm_dp_mst_hotplug(struct drm_dp_mst_topology_mgr *mgr) drm_kms_helper_hotplug_event(dev); } -static void dm_dp_mst_link_status_reset(struct drm_connector *connector) -{ - mutex_lock(&connector->dev->mode_config.mutex); - drm_connector_set_link_status_property(connector, DRM_MODE_LINK_STATUS_BAD); - mutex_unlock(&connector->dev->mode_config.mutex); -} - static void dm_dp_mst_register_connector(struct drm_connector *connector) { struct drm_device *dev = connector->dev; struct amdgpu_device *adev = dev->dev_private; - struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); if (adev->mode_info.rfbdev) drm_fb_helper_add_one_connector(&adev->mode_info.rfbdev->helper, connector); @@ -477,9 +414,6 @@ static void dm_dp_mst_register_connector(struct drm_connector *connector) DRM_ERROR("adev->mode_info.rfbdev is NULL\n"); drm_connector_register(connector); - - if (aconnector->mst_connected) - dm_dp_mst_link_status_reset(connector); } static const struct drm_dp_mst_topology_cbs dm_mst_cbs = { diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h index 8cf51da26657..2da851b40042 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h @@ -31,6 +31,5 @@ struct amdgpu_dm_connector; void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm, struct amdgpu_dm_connector *aconnector); -void dm_dp_mst_dc_sink_create(struct drm_connector *connector); #endif 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 fb04a4ad141f..5da2186b3615 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -1722,7 +1722,7 @@ static void write_i2c_retimer_setting( i2c_success = i2c_write(pipe_ctx, slave_address, buffer, sizeof(buffer)); RETIMER_REDRIVER_INFO("retimer write to slave_address = 0x%x,\ - offset = 0x%d, reg_val = 0x%d, i2c_success = %d\n", + offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) /* Write failure */ @@ -1734,7 +1734,7 @@ static void write_i2c_retimer_setting( i2c_success = i2c_write(pipe_ctx, slave_address, buffer, sizeof(buffer)); RETIMER_REDRIVER_INFO("retimer write to slave_address = 0x%x,\ - offset = 0x%d, reg_val = 0x%d, i2c_success = %d\n", + offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n", slave_address, buffer[0], buffer[1], i2c_success?1:0); if (!i2c_success) /* Write failure */ diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 199527171100..b57fa61b3034 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -169,6 +169,7 @@ struct link_training_settings; struct dc_config { bool gpu_vm_support; bool disable_disp_pll_sharing; + bool fbc_support; }; enum visual_confirm { diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index b75ede5f84f7..a6bcb90e8419 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -1736,7 +1736,12 @@ static void set_static_screen_control(struct pipe_ctx **pipe_ctx, if (events->force_trigger) value |= 0x1; - value |= 0x84; + if (num_pipes) { + struct dc *dc = pipe_ctx[0]->stream->ctx->dc; + + if (dc->fbc_compressor) + value |= 0x84; + } for (i = 0; i < num_pipes; i++) pipe_ctx[i]->stream_res.tg->funcs-> @@ -2507,6 +2512,8 @@ static void pplib_apply_display_requirements( dc, context->bw.dce.sclk_khz); + pp_display_cfg->min_dcfclock_khz = pp_display_cfg->min_engine_clock_khz; + pp_display_cfg->min_engine_clock_deep_sleep_khz = context->bw.dce.sclk_deep_sleep_khz; diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c index e3624ca24574..7c9fd9052ee2 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c @@ -1362,7 +1362,8 @@ static bool construct( pool->base.sw_i2cs[i] = NULL; } - dc->fbc_compressor = dce110_compressor_create(ctx); + if (dc->config.fbc_support) + dc->fbc_compressor = dce110_compressor_create(ctx); if (!underlay_create(ctx, &pool->base)) goto res_create_fail; diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index 2083c308007c..470d7b89071a 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -133,6 +133,10 @@ enum PP_FEATURE_MASK { PP_AVFS_MASK = 0x40000, }; +enum DC_FEATURE_MASK { + DC_FBC_MASK = 0x1, +}; + /** * struct amd_ip_funcs - general hooks for managing amdgpu IP Blocks */ diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h index d2e7c0fa96c2..8eb0bb241210 100644 --- a/drivers/gpu/drm/amd/include/atomfirmware.h +++ b/drivers/gpu/drm/amd/include/atomfirmware.h @@ -1325,7 +1325,7 @@ struct atom_smu_info_v3_3 { struct atom_common_table_header table_header; uint8_t smuip_min_ver; uint8_t smuip_max_ver; - uint8_t smu_rsd1; + uint8_t waflclk_ss_mode; uint8_t gpuclk_ss_mode; uint16_t sclk_ss_percentage; uint16_t sclk_ss_rate_10hz; @@ -1355,7 +1355,10 @@ struct atom_smu_info_v3_3 { uint32_t syspll3_1_vco_freq_10khz; uint32_t bootup_fclk_10khz; uint32_t bootup_waflclk_10khz; - uint32_t reserved[3]; + uint32_t smu_info_caps; + uint16_t waflclk_ss_percentage; // in unit of 0.001% + uint16_t smuinitoffset; + uint32_t reserved; }; /* diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c index 85119c2bdcc8..a2a7e0e94aa6 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c @@ -80,7 +80,9 @@ int phm_enable_dynamic_state_management(struct pp_hwmgr *hwmgr) PHM_FUNC_CHECK(hwmgr); adev = hwmgr->adev; - if (smum_is_dpm_running(hwmgr) && !amdgpu_passthrough(adev)) { + /* Skip for suspend/resume case */ + if (smum_is_dpm_running(hwmgr) && !amdgpu_passthrough(adev) + && adev->in_suspend) { pr_info("dpm has been enabled\n"); return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c index 47ac92369739..0173d0480024 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c @@ -352,6 +352,9 @@ int hwmgr_handle_task(struct pp_hwmgr *hwmgr, enum amd_pp_task task_id, switch (task_id) { case AMD_PP_TASK_DISPLAY_CONFIG_CHANGE: + ret = phm_pre_display_configuration_changed(hwmgr); + if (ret) + return ret; ret = phm_set_cpu_power_state(hwmgr); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c index 91ffb7bc4ee7..56437866d120 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c @@ -265,8 +265,6 @@ int psm_adjust_power_state_dynamic(struct pp_hwmgr *hwmgr, bool skip, if (skip) return 0; - phm_pre_display_configuration_changed(hwmgr); - phm_display_configuration_changed(hwmgr); if (hwmgr->ps) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c index ed35ec0341e6..b61a01f55284 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c @@ -3589,8 +3589,10 @@ static int smu7_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, cons } if (i >= sclk_table->count) { - data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; - sclk_table->dpm_levels[i-1].value = sclk; + if (sclk > sclk_table->dpm_levels[i-1].value) { + data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; + sclk_table->dpm_levels[i-1].value = sclk; + } } else { /* TODO: Check SCLK in DAL's minimum clocks * in case DeepSleep divider update is required. @@ -3607,8 +3609,10 @@ static int smu7_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, cons } if (i >= mclk_table->count) { - data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; - mclk_table->dpm_levels[i-1].value = mclk; + if (mclk > mclk_table->dpm_levels[i-1].value) { + data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; + mclk_table->dpm_levels[i-1].value = mclk; + } } if (data->display_timing.num_existing_displays != hwmgr->display_config->num_display) @@ -4525,12 +4529,12 @@ static int smu7_get_sclk_od(struct pp_hwmgr *hwmgr) struct smu7_single_dpm_table *sclk_table = &(data->dpm_table.sclk_table); struct smu7_single_dpm_table *golden_sclk_table = &(data->golden_dpm_table.sclk_table); - int value; + int value = sclk_table->dpm_levels[sclk_table->count - 1].value; + int golden_value = golden_sclk_table->dpm_levels + [golden_sclk_table->count - 1].value; - value = (sclk_table->dpm_levels[sclk_table->count - 1].value - - golden_sclk_table->dpm_levels[golden_sclk_table->count - 1].value) * - 100 / - golden_sclk_table->dpm_levels[golden_sclk_table->count - 1].value; + value -= golden_value; + value = DIV_ROUND_UP(value * 100, golden_value); return value; } @@ -4567,12 +4571,12 @@ static int smu7_get_mclk_od(struct pp_hwmgr *hwmgr) struct smu7_single_dpm_table *mclk_table = &(data->dpm_table.mclk_table); struct smu7_single_dpm_table *golden_mclk_table = &(data->golden_dpm_table.mclk_table); - int value; + int value = mclk_table->dpm_levels[mclk_table->count - 1].value; + int golden_value = golden_mclk_table->dpm_levels + [golden_mclk_table->count - 1].value; - value = (mclk_table->dpm_levels[mclk_table->count - 1].value - - golden_mclk_table->dpm_levels[golden_mclk_table->count - 1].value) * - 100 / - golden_mclk_table->dpm_levels[golden_mclk_table->count - 1].value; + value -= golden_value; + value = DIV_ROUND_UP(value * 100, golden_value); return value; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c index 99a33c33a32c..101c09b212ad 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c @@ -713,20 +713,20 @@ int smu_set_watermarks_for_clocks_ranges(void *wt_table, for (i = 0; i < wm_with_clock_ranges->num_wm_dmif_sets; i++) { table->WatermarkRow[1][i].MinClock = cpu_to_le16((uint16_t) - (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_min_dcfclk_clk_in_khz) / - 1000); + (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_min_dcfclk_clk_in_khz / + 1000)); table->WatermarkRow[1][i].MaxClock = cpu_to_le16((uint16_t) - (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_max_dcfclk_clk_in_khz) / - 1000); + (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_max_dcfclk_clk_in_khz / + 1000)); table->WatermarkRow[1][i].MinUclk = cpu_to_le16((uint16_t) - (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_min_mem_clk_in_khz) / - 1000); + (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_min_mem_clk_in_khz / + 1000)); table->WatermarkRow[1][i].MaxUclk = cpu_to_le16((uint16_t) - (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_max_mem_clk_in_khz) / - 1000); + (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_max_mem_clk_in_khz / + 1000)); table->WatermarkRow[1][i].WmSetting = (uint8_t) wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_set_id; } @@ -734,20 +734,20 @@ int smu_set_watermarks_for_clocks_ranges(void *wt_table, for (i = 0; i < wm_with_clock_ranges->num_wm_mcif_sets; i++) { table->WatermarkRow[0][i].MinClock = cpu_to_le16((uint16_t) - (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_min_socclk_clk_in_khz) / - 1000); + (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_min_socclk_clk_in_khz / + 1000)); table->WatermarkRow[0][i].MaxClock = cpu_to_le16((uint16_t) - (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_max_socclk_clk_in_khz) / - 1000); + (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_max_socclk_clk_in_khz / + 1000)); table->WatermarkRow[0][i].MinUclk = cpu_to_le16((uint16_t) - (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_min_mem_clk_in_khz) / - 1000); + (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_min_mem_clk_in_khz / + 1000)); table->WatermarkRow[0][i].MaxUclk = cpu_to_le16((uint16_t) - (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_max_mem_clk_in_khz) / - 1000); + (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_max_mem_clk_in_khz / + 1000)); table->WatermarkRow[0][i].WmSetting = (uint8_t) wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c index 8c4db86bb4b7..79c86247d0ac 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c @@ -3266,8 +3266,10 @@ static int vega10_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, co } if (i >= sclk_table->count) { - data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; - sclk_table->dpm_levels[i-1].value = sclk; + if (sclk > sclk_table->dpm_levels[i-1].value) { + data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; + sclk_table->dpm_levels[i-1].value = sclk; + } } for (i = 0; i < mclk_table->count; i++) { @@ -3276,8 +3278,10 @@ static int vega10_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, co } if (i >= mclk_table->count) { - data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; - mclk_table->dpm_levels[i-1].value = mclk; + if (mclk > mclk_table->dpm_levels[i-1].value) { + data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; + mclk_table->dpm_levels[i-1].value = mclk; + } } if (data->display_timing.num_existing_displays != hwmgr->display_config->num_display) @@ -4522,15 +4526,13 @@ static int vega10_get_sclk_od(struct pp_hwmgr *hwmgr) struct vega10_single_dpm_table *sclk_table = &(data->dpm_table.gfx_table); struct vega10_single_dpm_table *golden_sclk_table = &(data->golden_dpm_table.gfx_table); - int value; - - value = (sclk_table->dpm_levels[sclk_table->count - 1].value - - golden_sclk_table->dpm_levels - [golden_sclk_table->count - 1].value) * - 100 / - golden_sclk_table->dpm_levels + int value = sclk_table->dpm_levels[sclk_table->count - 1].value; + int golden_value = golden_sclk_table->dpm_levels [golden_sclk_table->count - 1].value; + value -= golden_value; + value = DIV_ROUND_UP(value * 100, golden_value); + return value; } @@ -4575,16 +4577,13 @@ static int vega10_get_mclk_od(struct pp_hwmgr *hwmgr) struct vega10_single_dpm_table *mclk_table = &(data->dpm_table.mem_table); struct vega10_single_dpm_table *golden_mclk_table = &(data->golden_dpm_table.mem_table); - int value; - - value = (mclk_table->dpm_levels - [mclk_table->count - 1].value - - golden_mclk_table->dpm_levels - [golden_mclk_table->count - 1].value) * - 100 / - golden_mclk_table->dpm_levels + int value = mclk_table->dpm_levels[mclk_table->count - 1].value; + int golden_value = golden_mclk_table->dpm_levels [golden_mclk_table->count - 1].value; + value -= golden_value; + value = DIV_ROUND_UP(value * 100, golden_value); + return value; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c index 74bc37308dc0..54364444ecd1 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega12_hwmgr.c @@ -2243,12 +2243,12 @@ static int vega12_get_sclk_od(struct pp_hwmgr *hwmgr) struct vega12_single_dpm_table *sclk_table = &(data->dpm_table.gfx_table); struct vega12_single_dpm_table *golden_sclk_table = &(data->golden_dpm_table.gfx_table); - int value; + int value = sclk_table->dpm_levels[sclk_table->count - 1].value; + int golden_value = golden_sclk_table->dpm_levels + [golden_sclk_table->count - 1].value; - value = (sclk_table->dpm_levels[sclk_table->count - 1].value - - golden_sclk_table->dpm_levels[golden_sclk_table->count - 1].value) * - 100 / - golden_sclk_table->dpm_levels[golden_sclk_table->count - 1].value; + value -= golden_value; + value = DIV_ROUND_UP(value * 100, golden_value); return value; } @@ -2264,16 +2264,13 @@ static int vega12_get_mclk_od(struct pp_hwmgr *hwmgr) struct vega12_single_dpm_table *mclk_table = &(data->dpm_table.mem_table); struct vega12_single_dpm_table *golden_mclk_table = &(data->golden_dpm_table.mem_table); - int value; - - value = (mclk_table->dpm_levels - [mclk_table->count - 1].value - - golden_mclk_table->dpm_levels - [golden_mclk_table->count - 1].value) * - 100 / - golden_mclk_table->dpm_levels + int value = mclk_table->dpm_levels[mclk_table->count - 1].value; + int golden_value = golden_mclk_table->dpm_levels [golden_mclk_table->count - 1].value; + value -= golden_value; + value = DIV_ROUND_UP(value * 100, golden_value); + return value; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c index 57143d51e3ee..3b7fce5d7258 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c @@ -75,7 +75,17 @@ static void vega20_set_default_registry_data(struct pp_hwmgr *hwmgr) data->phy_clk_quad_eqn_b = PPREGKEY_VEGA20QUADRATICEQUATION_DFLT; data->phy_clk_quad_eqn_c = PPREGKEY_VEGA20QUADRATICEQUATION_DFLT; - data->registry_data.disallowed_features = 0x0; + /* + * Disable the following features for now: + * GFXCLK DS + * SOCLK DS + * LCLK DS + * DCEFCLK DS + * FCLK DS + * MP1CLK DS + * MP0CLK DS + */ + data->registry_data.disallowed_features = 0xE0041C00; data->registry_data.od_state_in_dc_support = 0; data->registry_data.thermal_support = 1; data->registry_data.skip_baco_hardware = 0; @@ -120,6 +130,7 @@ static void vega20_set_default_registry_data(struct pp_hwmgr *hwmgr) data->registry_data.disable_auto_wattman = 1; data->registry_data.auto_wattman_debug = 0; data->registry_data.auto_wattman_sample_period = 100; + data->registry_data.fclk_gfxclk_ratio = 0; data->registry_data.auto_wattman_threshold = 50; data->registry_data.gfxoff_controlled_by_driver = 1; data->gfxoff_allowed = false; @@ -829,6 +840,28 @@ static int vega20_enable_all_smu_features(struct pp_hwmgr *hwmgr) return 0; } +static int vega20_notify_smc_display_change(struct pp_hwmgr *hwmgr) +{ + struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend); + + if (data->smu_features[GNLD_DPM_UCLK].enabled) + return smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_SetUclkFastSwitch, + 1); + + return 0; +} + +static int vega20_send_clock_ratio(struct pp_hwmgr *hwmgr) +{ + struct vega20_hwmgr *data = + (struct vega20_hwmgr *)(hwmgr->backend); + + return smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_SetFclkGfxClkRatio, + data->registry_data.fclk_gfxclk_ratio); +} + static int vega20_disable_all_smu_features(struct pp_hwmgr *hwmgr) { struct vega20_hwmgr *data = @@ -1290,12 +1323,13 @@ static int vega20_get_sclk_od( &(data->dpm_table.gfx_table); struct vega20_single_dpm_table *golden_sclk_table = &(data->golden_dpm_table.gfx_table); - int value; + int value = sclk_table->dpm_levels[sclk_table->count - 1].value; + int golden_value = golden_sclk_table->dpm_levels + [golden_sclk_table->count - 1].value; /* od percentage */ - value = DIV_ROUND_UP((sclk_table->dpm_levels[sclk_table->count - 1].value - - golden_sclk_table->dpm_levels[golden_sclk_table->count - 1].value) * 100, - golden_sclk_table->dpm_levels[golden_sclk_table->count - 1].value); + value -= golden_value; + value = DIV_ROUND_UP(value * 100, golden_value); return value; } @@ -1335,12 +1369,13 @@ static int vega20_get_mclk_od( &(data->dpm_table.mem_table); struct vega20_single_dpm_table *golden_mclk_table = &(data->golden_dpm_table.mem_table); - int value; + int value = mclk_table->dpm_levels[mclk_table->count - 1].value; + int golden_value = golden_mclk_table->dpm_levels + [golden_mclk_table->count - 1].value; /* od percentage */ - value = DIV_ROUND_UP((mclk_table->dpm_levels[mclk_table->count - 1].value - - golden_mclk_table->dpm_levels[golden_mclk_table->count - 1].value) * 100, - golden_mclk_table->dpm_levels[golden_mclk_table->count - 1].value); + value -= golden_value; + value = DIV_ROUND_UP(value * 100, golden_value); return value; } @@ -1532,6 +1567,16 @@ static int vega20_enable_dpm_tasks(struct pp_hwmgr *hwmgr) "[EnableDPMTasks] Failed to enable all smu features!", return result); + result = vega20_notify_smc_display_change(hwmgr); + PP_ASSERT_WITH_CODE(!result, + "[EnableDPMTasks] Failed to notify smc display change!", + return result); + + result = vega20_send_clock_ratio(hwmgr); + PP_ASSERT_WITH_CODE(!result, + "[EnableDPMTasks] Failed to send clock ratio!", + return result); + /* Initialize UVD/VCE powergating state */ vega20_init_powergate_state(hwmgr); @@ -1615,14 +1660,15 @@ static uint32_t vega20_find_highest_dpm_level( return i; } -static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) +static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_mask) { struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend); uint32_t min_freq; int ret = 0; - if (data->smu_features[GNLD_DPM_GFXCLK].enabled) { + if (data->smu_features[GNLD_DPM_GFXCLK].enabled && + (feature_mask & FEATURE_DPM_GFXCLK_MASK)) { min_freq = data->dpm_table.gfx_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, @@ -1631,7 +1677,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UCLK].enabled) { + if (data->smu_features[GNLD_DPM_UCLK].enabled && + (feature_mask & FEATURE_DPM_UCLK_MASK)) { min_freq = data->dpm_table.mem_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, @@ -1647,7 +1694,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UVD].enabled) { + if (data->smu_features[GNLD_DPM_UVD].enabled && + (feature_mask & FEATURE_DPM_UVD_MASK)) { min_freq = data->dpm_table.vclk_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1665,7 +1713,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_VCE].enabled) { + if (data->smu_features[GNLD_DPM_VCE].enabled && + (feature_mask & FEATURE_DPM_VCE_MASK)) { min_freq = data->dpm_table.eclk_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1675,7 +1724,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_SOCCLK].enabled) { + if (data->smu_features[GNLD_DPM_SOCCLK].enabled && + (feature_mask & FEATURE_DPM_SOCCLK_MASK)) { min_freq = data->dpm_table.soc_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1688,14 +1738,15 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret; } -static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) +static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr, uint32_t feature_mask) { struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend); uint32_t max_freq; int ret = 0; - if (data->smu_features[GNLD_DPM_GFXCLK].enabled) { + if (data->smu_features[GNLD_DPM_GFXCLK].enabled && + (feature_mask & FEATURE_DPM_GFXCLK_MASK)) { max_freq = data->dpm_table.gfx_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1705,7 +1756,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UCLK].enabled) { + if (data->smu_features[GNLD_DPM_UCLK].enabled && + (feature_mask & FEATURE_DPM_UCLK_MASK)) { max_freq = data->dpm_table.mem_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1715,7 +1767,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UVD].enabled) { + if (data->smu_features[GNLD_DPM_UVD].enabled && + (feature_mask & FEATURE_DPM_UVD_MASK)) { max_freq = data->dpm_table.vclk_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1732,7 +1785,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_VCE].enabled) { + if (data->smu_features[GNLD_DPM_VCE].enabled && + (feature_mask & FEATURE_DPM_VCE_MASK)) { max_freq = data->dpm_table.eclk_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1742,7 +1796,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_SOCCLK].enabled) { + if (data->smu_features[GNLD_DPM_SOCCLK].enabled && + (feature_mask & FEATURE_DPM_SOCCLK_MASK)) { max_freq = data->dpm_table.soc_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1972,19 +2027,6 @@ static int vega20_read_sensor(struct pp_hwmgr *hwmgr, int idx, return ret; } -static int vega20_notify_smc_display_change(struct pp_hwmgr *hwmgr, - bool has_disp) -{ - struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend); - - if (data->smu_features[GNLD_DPM_UCLK].enabled) - return smum_send_msg_to_smc_with_parameter(hwmgr, - PPSMC_MSG_SetUclkFastSwitch, - has_disp ? 1 : 0); - - return 0; -} - int vega20_display_clock_voltage_request(struct pp_hwmgr *hwmgr, struct pp_display_clock_request *clock_req) { @@ -2044,13 +2086,6 @@ static int vega20_notify_smc_display_config_after_ps_adjustment( struct pp_display_clock_request clock_req; int ret = 0; - if ((hwmgr->display_config->num_display > 1) && - !hwmgr->display_config->multi_monitor_in_sync && - !hwmgr->display_config->nb_pstate_switch_disable) - vega20_notify_smc_display_change(hwmgr, false); - else - vega20_notify_smc_display_change(hwmgr, true); - min_clocks.dcefClock = hwmgr->display_config->min_dcef_set_clk; min_clocks.dcefClockInSR = hwmgr->display_config->min_dcef_deep_sleep_set_clk; min_clocks.memoryClock = hwmgr->display_config->min_mem_set_clock; @@ -2101,12 +2136,12 @@ static int vega20_force_dpm_highest(struct pp_hwmgr *hwmgr) data->dpm_table.mem_table.dpm_state.soft_max_level = data->dpm_table.mem_table.dpm_levels[soft_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to highest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); @@ -2133,12 +2168,12 @@ static int vega20_force_dpm_lowest(struct pp_hwmgr *hwmgr) data->dpm_table.mem_table.dpm_state.soft_max_level = data->dpm_table.mem_table.dpm_levels[soft_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to highest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); @@ -2151,12 +2186,12 @@ static int vega20_unforce_dpm_levels(struct pp_hwmgr *hwmgr) { int ret = 0; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload DPM Bootup Levels!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload DPM Max Levels!", return ret); @@ -2214,12 +2249,12 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr, data->dpm_table.gfx_table.dpm_state.soft_max_level = data->dpm_table.gfx_table.dpm_levels[soft_max_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, FEATURE_DPM_GFXCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to lowest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, FEATURE_DPM_GFXCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); @@ -2234,12 +2269,12 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr, data->dpm_table.mem_table.dpm_state.soft_max_level = data->dpm_table.mem_table.dpm_levels[soft_max_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, FEATURE_DPM_UCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to lowest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, FEATURE_DPM_UCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h index 56fe6a0d42e8..25faaa5c5b10 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h @@ -328,6 +328,7 @@ struct vega20_registry_data { uint8_t disable_auto_wattman; uint32_t auto_wattman_debug; uint32_t auto_wattman_sample_period; + uint32_t fclk_gfxclk_ratio; uint8_t auto_wattman_threshold; uint8_t log_avfs_param; uint8_t enable_enginess; diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu7_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/smu7_ppsmc.h index 62f36ba2435b..c1a99dfe4913 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu7_ppsmc.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu7_ppsmc.h @@ -386,6 +386,8 @@ typedef uint16_t PPSMC_Result; #define PPSMC_MSG_AgmResetPsm ((uint16_t) 0x403) #define PPSMC_MSG_ReadVftCell ((uint16_t) 0x404) +#define PPSMC_MSG_ApplyAvfsCksOffVoltage ((uint16_t) 0x415) + #define PPSMC_MSG_GFX_CU_PG_ENABLE ((uint16_t) 0x280) #define PPSMC_MSG_GFX_CU_PG_DISABLE ((uint16_t) 0x281) #define PPSMC_MSG_GetCurrPkgPwr ((uint16_t) 0x282) diff --git a/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h index 45d64a81e945..4f63a736ea0e 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h +++ b/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h @@ -105,7 +105,8 @@ #define PPSMC_MSG_SetSystemVirtualDramAddrHigh 0x4B #define PPSMC_MSG_SetSystemVirtualDramAddrLow 0x4C #define PPSMC_MSG_WaflTest 0x4D -// Unused ID 0x4E to 0x50 +#define PPSMC_MSG_SetFclkGfxClkRatio 0x4E +// Unused ID 0x4F to 0x50 #define PPSMC_MSG_AllowGfxOff 0x51 #define PPSMC_MSG_DisallowGfxOff 0x52 #define PPSMC_MSG_GetPptLimit 0x53 diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c index 872d3824337b..a1e0ac9ae248 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c @@ -1985,6 +1985,12 @@ int polaris10_thermal_avfs_enable(struct pp_hwmgr *hwmgr) smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableAvfs); + /* Apply avfs cks-off voltages to avoid the overshoot + * when switching to the highest sclk frequency + */ + if (data->apply_avfs_cks_off_voltage) + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ApplyAvfsCksOffVoltage); + return 0; } diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c index 99d5e4f98f49..a6edd5df33b0 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c @@ -37,10 +37,13 @@ MODULE_FIRMWARE("amdgpu/fiji_smc.bin"); MODULE_FIRMWARE("amdgpu/polaris10_smc.bin"); MODULE_FIRMWARE("amdgpu/polaris10_smc_sk.bin"); MODULE_FIRMWARE("amdgpu/polaris10_k_smc.bin"); +MODULE_FIRMWARE("amdgpu/polaris10_k2_smc.bin"); MODULE_FIRMWARE("amdgpu/polaris11_smc.bin"); MODULE_FIRMWARE("amdgpu/polaris11_smc_sk.bin"); MODULE_FIRMWARE("amdgpu/polaris11_k_smc.bin"); +MODULE_FIRMWARE("amdgpu/polaris11_k2_smc.bin"); MODULE_FIRMWARE("amdgpu/polaris12_smc.bin"); +MODULE_FIRMWARE("amdgpu/polaris12_k_smc.bin"); MODULE_FIRMWARE("amdgpu/vegam_smc.bin"); MODULE_FIRMWARE("amdgpu/vega10_smc.bin"); MODULE_FIRMWARE("amdgpu/vega10_acg_smc.bin"); diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c index 69dab82a3771..bf589c53b908 100644 --- a/drivers/gpu/drm/ast/ast_drv.c +++ b/drivers/gpu/drm/ast/ast_drv.c @@ -60,8 +60,29 @@ static const struct pci_device_id pciidlist[] = { MODULE_DEVICE_TABLE(pci, pciidlist); +static void ast_kick_out_firmware_fb(struct pci_dev *pdev) +{ + struct apertures_struct *ap; + bool primary = false; + + ap = alloc_apertures(1); + if (!ap) + return; + + ap->ranges[0].base = pci_resource_start(pdev, 0); + ap->ranges[0].size = pci_resource_len(pdev, 0); + +#ifdef CONFIG_X86 + primary = pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW; +#endif + drm_fb_helper_remove_conflicting_framebuffers(ap, "astdrmfb", primary); + kfree(ap); +} + static int ast_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { + ast_kick_out_firmware_fb(pdev); + return drm_get_pci_dev(pdev, ent, &driver); } diff --git a/drivers/gpu/drm/ast/ast_fb.c b/drivers/gpu/drm/ast/ast_fb.c index 0cd827e11fa2..de26df0c6044 100644 --- a/drivers/gpu/drm/ast/ast_fb.c +++ b/drivers/gpu/drm/ast/ast_fb.c @@ -263,6 +263,7 @@ static void ast_fbdev_destroy(struct drm_device *dev, { struct ast_framebuffer *afb = &afbdev->afb; + drm_crtc_force_disable_all(dev); drm_fb_helper_unregister_fbi(&afbdev->helper); if (afb->obj) { diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index dac355812adc..373700c05a00 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -583,7 +583,8 @@ void ast_driver_unload(struct drm_device *dev) drm_mode_config_cleanup(dev); ast_mm_fini(ast); - pci_iounmap(dev->pdev, ast->ioregs); + if (ast->ioregs != ast->regs + AST_IO_MM_OFFSET) + pci_iounmap(dev->pdev, ast->ioregs); pci_iounmap(dev->pdev, ast->regs); kfree(ast); } diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 5e77d456d9bb..8bb355d5d43d 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -568,6 +568,7 @@ static int ast_crtc_do_set_base(struct drm_crtc *crtc, } ast_bo_unreserve(bo); + ast_set_offset_reg(crtc); ast_set_start_address_crt1(crtc, (u32)gpu_addr); return 0; @@ -972,9 +973,21 @@ static int get_clock(void *i2c_priv) { struct ast_i2c_chan *i2c = i2c_priv; struct ast_private *ast = i2c->dev->dev_private; - uint32_t val; + uint32_t val, val2, count, pass; + + count = 0; + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4) & 0x01; + do { + val2 = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4) & 0x01; + if (val == val2) { + pass++; + } else { + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4) & 0x01; + } + } while ((pass < 5) && (count++ < 0x10000)); - val = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4; return val & 1 ? 1 : 0; } @@ -982,9 +995,21 @@ static int get_data(void *i2c_priv) { struct ast_i2c_chan *i2c = i2c_priv; struct ast_private *ast = i2c->dev->dev_private; - uint32_t val; + uint32_t val, val2, count, pass; + + count = 0; + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5) & 0x01; + do { + val2 = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5) & 0x01; + if (val == val2) { + pass++; + } else { + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5) & 0x01; + } + } while ((pass < 5) && (count++ < 0x10000)); - val = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5; return val & 1 ? 1 : 0; } @@ -997,7 +1022,7 @@ static void set_clock(void *i2c_priv, int clock) for (i = 0; i < 0x10000; i++) { ujcrb7 = ((clock & 0x01) ? 0 : 1); - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xfe, ujcrb7); + ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xf4, ujcrb7); jtemp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x01); if (ujcrb7 == jtemp) break; @@ -1013,7 +1038,7 @@ static void set_data(void *i2c_priv, int data) for (i = 0; i < 0x10000; i++) { ujcrb7 = ((data & 0x01) ? 0 : 1) << 2; - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xfb, ujcrb7); + ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xf1, ujcrb7); jtemp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x04); if (ujcrb7 == jtemp) break; @@ -1254,7 +1279,7 @@ static int ast_cursor_move(struct drm_crtc *crtc, ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xc7, ((y >> 8) & 0x07)); /* dummy write to fire HWC */ - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xCB, 0xFF, 0x00); + ast_show_cursor(crtc); return 0; } diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c index 680566d97adc..10243965ee7c 100644 --- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c +++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c @@ -54,7 +54,7 @@ #define SN_AUX_ADDR_7_0_REG 0x76 #define SN_AUX_LENGTH_REG 0x77 #define SN_AUX_CMD_REG 0x78 -#define AUX_CMD_SEND BIT(1) +#define AUX_CMD_SEND BIT(0) #define AUX_CMD_REQ(x) ((x) << 4) #define SN_AUX_RDATA_REG(x) (0x79 + (x)) #define SN_SSC_CONFIG_REG 0x93 diff --git a/drivers/gpu/drm/drm_auth.c b/drivers/gpu/drm/drm_auth.c index d9c0f7573905..1669c42c40ed 100644 --- a/drivers/gpu/drm/drm_auth.c +++ b/drivers/gpu/drm/drm_auth.c @@ -142,6 +142,7 @@ static int drm_new_set_master(struct drm_device *dev, struct drm_file *fpriv) lockdep_assert_held_once(&dev->master_mutex); + WARN_ON(fpriv->is_master); old_master = fpriv->master; fpriv->master = drm_master_create(dev); if (!fpriv->master) { @@ -170,6 +171,7 @@ out_err: /* drop references and restore old master on failure */ drm_master_put(&fpriv->master); fpriv->master = old_master; + fpriv->is_master = 0; return ret; } diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index 5ff1d79b86c4..0e0df398222d 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -1275,6 +1275,9 @@ static struct drm_dp_mst_branch *drm_dp_get_mst_branch_device(struct drm_dp_mst_ mutex_lock(&mgr->lock); mstb = mgr->mst_primary; + if (!mstb) + goto out; + for (i = 0; i < lct - 1; i++) { int shift = (i % 2) ? 0 : 4; int port_num = (rad[i / 2] >> shift) & 0xf; diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index a502f3e519fd..9d64f874f965 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -71,7 +71,7 @@ MODULE_PARM_DESC(drm_fbdev_overalloc, #if IS_ENABLED(CONFIG_DRM_FBDEV_LEAK_PHYS_SMEM) static bool drm_leak_fbdev_smem = false; module_param_unsafe(drm_leak_fbdev_smem, bool, 0600); -MODULE_PARM_DESC(fbdev_emulation, +MODULE_PARM_DESC(drm_leak_fbdev_smem, "Allow unsafe leaking fbdev physical smem address [default=false]"); #endif @@ -219,6 +219,9 @@ int drm_fb_helper_single_add_all_connectors(struct drm_fb_helper *fb_helper) mutex_lock(&fb_helper->lock); drm_connector_list_iter_begin(dev, &conn_iter); drm_for_each_connector_iter(connector, &conn_iter) { + if (connector->connector_type == DRM_MODE_CONNECTOR_WRITEBACK) + continue; + ret = __drm_fb_helper_add_one_connector(fb_helper, connector); if (ret) goto fail; diff --git a/drivers/gpu/drm/drm_fourcc.c b/drivers/gpu/drm/drm_fourcc.c index 90a1c846fc25..8aaa5e86a979 100644 --- a/drivers/gpu/drm/drm_fourcc.c +++ b/drivers/gpu/drm/drm_fourcc.c @@ -97,9 +97,9 @@ EXPORT_SYMBOL(drm_mode_legacy_fb_format); /** * drm_driver_legacy_fb_format - compute drm fourcc code from legacy description + * @dev: DRM device * @bpp: bits per pixels * @depth: bit depth per pixel - * @native: use host native byte order * * Computes a drm fourcc pixel format code for the given @bpp/@depth values. * Unlike drm_mode_legacy_fb_format() this looks at the drivers mode_config, diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h index 0c4eb4a9ab31..51e06defc8d8 100644 --- a/drivers/gpu/drm/drm_internal.h +++ b/drivers/gpu/drm/drm_internal.h @@ -104,6 +104,8 @@ struct device *drm_sysfs_minor_alloc(struct drm_minor *minor); int drm_sysfs_connector_add(struct drm_connector *connector); void drm_sysfs_connector_remove(struct drm_connector *connector); +void drm_sysfs_lease_event(struct drm_device *dev); + /* drm_gem.c */ int drm_gem_init(struct drm_device *dev); void drm_gem_destroy(struct drm_device *dev); diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c index 94bd872d56c4..7e6746b2d704 100644 --- a/drivers/gpu/drm/drm_ioctl.c +++ b/drivers/gpu/drm/drm_ioctl.c @@ -37,6 +37,7 @@ #include <linux/pci.h> #include <linux/export.h> +#include <linux/nospec.h> /** * DOC: getunique and setversion story @@ -800,13 +801,17 @@ long drm_ioctl(struct file *filp, if (is_driver_ioctl) { /* driver ioctl */ - if (nr - DRM_COMMAND_BASE >= dev->driver->num_ioctls) + unsigned int index = nr - DRM_COMMAND_BASE; + + if (index >= dev->driver->num_ioctls) goto err_i1; - ioctl = &dev->driver->ioctls[nr - DRM_COMMAND_BASE]; + index = array_index_nospec(index, dev->driver->num_ioctls); + ioctl = &dev->driver->ioctls[index]; } else { /* core ioctl */ if (nr >= DRM_CORE_IOCTL_COUNT) goto err_i1; + nr = array_index_nospec(nr, DRM_CORE_IOCTL_COUNT); ioctl = &drm_ioctls[nr]; } @@ -888,6 +893,7 @@ bool drm_ioctl_flags(unsigned int nr, unsigned int *flags) if (nr >= DRM_CORE_IOCTL_COUNT) return false; + nr = array_index_nospec(nr, DRM_CORE_IOCTL_COUNT); *flags = drm_ioctls[nr].flags; return true; diff --git a/drivers/gpu/drm/drm_lease.c b/drivers/gpu/drm/drm_lease.c index 24a177ea5417..c61680ad962d 100644 --- a/drivers/gpu/drm/drm_lease.c +++ b/drivers/gpu/drm/drm_lease.c @@ -296,7 +296,7 @@ void drm_lease_destroy(struct drm_master *master) if (master->lessor) { /* Tell the master to check the lessee list */ - drm_sysfs_hotplug_event(dev); + drm_sysfs_lease_event(dev); drm_master_put(&master->lessor); } diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c index b3c1daad1169..ecb7b33002bb 100644 --- a/drivers/gpu/drm/drm_sysfs.c +++ b/drivers/gpu/drm/drm_sysfs.c @@ -301,6 +301,16 @@ void drm_sysfs_connector_remove(struct drm_connector *connector) connector->kdev = NULL; } +void drm_sysfs_lease_event(struct drm_device *dev) +{ + char *event_string = "LEASE=1"; + char *envp[] = { event_string, NULL }; + + DRM_DEBUG("generating lease event\n"); + + kobject_uevent_env(&dev->primary->kdev->kobj, KOBJ_CHANGE, envp); +} + /** * drm_sysfs_hotplug_event - generate a DRM uevent * @dev: DRM device diff --git a/drivers/gpu/drm/etnaviv/etnaviv_sched.c b/drivers/gpu/drm/etnaviv/etnaviv_sched.c index e7c3ed6c9a2e..9b476368aa31 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_sched.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_sched.c @@ -93,7 +93,7 @@ static void etnaviv_sched_timedout_job(struct drm_sched_job *sched_job) * If the GPU managed to complete this jobs fence, the timout is * spurious. Bail out. */ - if (fence_completed(gpu, submit->out_fence->seqno)) + if (dma_fence_is_signaled(submit->out_fence)) return; /* diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c index 94529aa82339..aef487dd8731 100644 --- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c @@ -164,13 +164,6 @@ static u32 decon_get_frame_count(struct decon_context *ctx, bool end) return frm; } -static u32 decon_get_vblank_counter(struct exynos_drm_crtc *crtc) -{ - struct decon_context *ctx = crtc->ctx; - - return decon_get_frame_count(ctx, false); -} - static void decon_setup_trigger(struct decon_context *ctx) { if (!ctx->crtc->i80_mode && !(ctx->out_type & I80_HW_TRG)) @@ -536,7 +529,6 @@ static const struct exynos_drm_crtc_ops decon_crtc_ops = { .disable = decon_disable, .enable_vblank = decon_enable_vblank, .disable_vblank = decon_disable_vblank, - .get_vblank_counter = decon_get_vblank_counter, .atomic_begin = decon_atomic_begin, .update_plane = decon_update_plane, .disable_plane = decon_disable_plane, @@ -554,7 +546,6 @@ static int decon_bind(struct device *dev, struct device *master, void *data) int ret; ctx->drm_dev = drm_dev; - drm_dev->max_vblank_count = 0xffffffff; for (win = ctx->first_win; win < WINDOWS_NR; win++) { ctx->configs[win].pixel_formats = decon_formats; diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c index eea90251808f..2696289ecc78 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c @@ -162,16 +162,6 @@ static void exynos_drm_crtc_disable_vblank(struct drm_crtc *crtc) exynos_crtc->ops->disable_vblank(exynos_crtc); } -static u32 exynos_drm_crtc_get_vblank_counter(struct drm_crtc *crtc) -{ - struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc); - - if (exynos_crtc->ops->get_vblank_counter) - return exynos_crtc->ops->get_vblank_counter(exynos_crtc); - - return 0; -} - static const struct drm_crtc_funcs exynos_crtc_funcs = { .set_config = drm_atomic_helper_set_config, .page_flip = drm_atomic_helper_page_flip, @@ -181,7 +171,6 @@ static const struct drm_crtc_funcs exynos_crtc_funcs = { .atomic_destroy_state = drm_atomic_helper_crtc_destroy_state, .enable_vblank = exynos_drm_crtc_enable_vblank, .disable_vblank = exynos_drm_crtc_disable_vblank, - .get_vblank_counter = exynos_drm_crtc_get_vblank_counter, }; struct exynos_drm_crtc *exynos_drm_crtc_create(struct drm_device *drm_dev, diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index ec9604f1272b..5e61e707f955 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -135,7 +135,6 @@ struct exynos_drm_crtc_ops { void (*disable)(struct exynos_drm_crtc *crtc); int (*enable_vblank)(struct exynos_drm_crtc *crtc); void (*disable_vblank)(struct exynos_drm_crtc *crtc); - u32 (*get_vblank_counter)(struct exynos_drm_crtc *crtc); enum drm_mode_status (*mode_valid)(struct exynos_drm_crtc *crtc, const struct drm_display_mode *mode); bool (*mode_fixup)(struct exynos_drm_crtc *crtc, diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c index 07af7758066d..d81e62ae286a 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c @@ -14,6 +14,7 @@ #include <drm/drmP.h> #include <drm/drm_crtc_helper.h> +#include <drm/drm_fb_helper.h> #include <drm/drm_mipi_dsi.h> #include <drm/drm_panel.h> #include <drm/drm_atomic_helper.h> @@ -1474,12 +1475,12 @@ static int exynos_dsi_create_connector(struct drm_encoder *encoder) { struct exynos_dsi *dsi = encoder_to_dsi(encoder); struct drm_connector *connector = &dsi->connector; + struct drm_device *drm = encoder->dev; int ret; connector->polled = DRM_CONNECTOR_POLL_HPD; - ret = drm_connector_init(encoder->dev, connector, - &exynos_dsi_connector_funcs, + ret = drm_connector_init(drm, connector, &exynos_dsi_connector_funcs, DRM_MODE_CONNECTOR_DSI); if (ret) { DRM_ERROR("Failed to initialize connector with drm\n"); @@ -1489,7 +1490,12 @@ static int exynos_dsi_create_connector(struct drm_encoder *encoder) connector->status = connector_status_disconnected; drm_connector_helper_add(connector, &exynos_dsi_connector_helper_funcs); drm_connector_attach_encoder(connector, encoder); + if (!drm->registered) + return 0; + connector->funcs->reset(connector); + drm_fb_helper_add_one_connector(drm->fb_helper, connector); + drm_connector_register(connector); return 0; } @@ -1527,7 +1533,9 @@ static int exynos_dsi_host_attach(struct mipi_dsi_host *host, } dsi->panel = of_drm_find_panel(device->dev.of_node); - if (dsi->panel) { + if (IS_ERR(dsi->panel)) { + dsi->panel = NULL; + } else { drm_panel_attach(dsi->panel, &dsi->connector); dsi->connector.status = connector_status_connected; } diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c index 918dd2c82209..01d182289efa 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c @@ -192,7 +192,7 @@ int exynos_drm_fbdev_init(struct drm_device *dev) struct drm_fb_helper *helper; int ret; - if (!dev->mode_config.num_crtc || !dev->mode_config.num_connector) + if (!dev->mode_config.num_crtc) return 0; fbdev = kzalloc(sizeof(*fbdev), GFP_KERNEL); diff --git a/drivers/gpu/drm/i915/gvt/aperture_gm.c b/drivers/gpu/drm/i915/gvt/aperture_gm.c index fe754022e356..359d37d5c958 100644 --- a/drivers/gpu/drm/i915/gvt/aperture_gm.c +++ b/drivers/gpu/drm/i915/gvt/aperture_gm.c @@ -61,10 +61,12 @@ static int alloc_gm(struct intel_vgpu *vgpu, bool high_gm) } mutex_lock(&dev_priv->drm.struct_mutex); + mmio_hw_access_pre(dev_priv); ret = i915_gem_gtt_insert(&dev_priv->ggtt.vm, node, size, I915_GTT_PAGE_SIZE, I915_COLOR_UNEVICTABLE, start, end, flags); + mmio_hw_access_post(dev_priv); mutex_unlock(&dev_priv->drm.struct_mutex); if (ret) gvt_err("fail to alloc %s gm space from host\n", diff --git a/drivers/gpu/drm/i915/gvt/fb_decoder.c b/drivers/gpu/drm/i915/gvt/fb_decoder.c index 481896fb712a..85e6736f0a32 100644 --- a/drivers/gpu/drm/i915/gvt/fb_decoder.c +++ b/drivers/gpu/drm/i915/gvt/fb_decoder.c @@ -235,7 +235,7 @@ int intel_vgpu_decode_primary_plane(struct intel_vgpu *vgpu, plane->bpp = skl_pixel_formats[fmt].bpp; plane->drm_format = skl_pixel_formats[fmt].drm_format; } else { - plane->tiled = !!(val & DISPPLANE_TILED); + plane->tiled = val & DISPPLANE_TILED; fmt = bdw_format_to_drm(val & DISPPLANE_PIXFORMAT_MASK); plane->bpp = bdw_pixel_formats[fmt].bpp; plane->drm_format = bdw_pixel_formats[fmt].drm_format; diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c index 2402395a068d..c7103dd2d8d5 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.c +++ b/drivers/gpu/drm/i915/gvt/gtt.c @@ -1905,7 +1905,6 @@ static struct intel_vgpu_mm *intel_vgpu_create_ggtt_mm(struct intel_vgpu *vgpu) vgpu_free_mm(mm); return ERR_PTR(-ENOMEM); } - mm->ggtt_mm.last_partial_off = -1UL; return mm; } @@ -1930,7 +1929,6 @@ void _intel_vgpu_mm_release(struct kref *mm_ref) invalidate_ppgtt_mm(mm); } else { vfree(mm->ggtt_mm.virtual_ggtt); - mm->ggtt_mm.last_partial_off = -1UL; } vgpu_free_mm(mm); @@ -2168,6 +2166,8 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off, struct intel_gvt_gtt_entry e, m; dma_addr_t dma_addr; int ret; + struct intel_gvt_partial_pte *partial_pte, *pos, *n; + bool partial_update = false; if (bytes != 4 && bytes != 8) return -EINVAL; @@ -2178,68 +2178,57 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off, if (!vgpu_gmadr_is_valid(vgpu, gma)) return 0; - ggtt_get_guest_entry(ggtt_mm, &e, g_gtt_index); - + e.type = GTT_TYPE_GGTT_PTE; memcpy((void *)&e.val64 + (off & (info->gtt_entry_size - 1)), p_data, bytes); /* If ggtt entry size is 8 bytes, and it's split into two 4 bytes - * write, we assume the two 4 bytes writes are consecutive. - * Otherwise, we abort and report error + * write, save the first 4 bytes in a list and update virtual + * PTE. Only update shadow PTE when the second 4 bytes comes. */ if (bytes < info->gtt_entry_size) { - if (ggtt_mm->ggtt_mm.last_partial_off == -1UL) { - /* the first partial part*/ - ggtt_mm->ggtt_mm.last_partial_off = off; - ggtt_mm->ggtt_mm.last_partial_data = e.val64; - return 0; - } else if ((g_gtt_index == - (ggtt_mm->ggtt_mm.last_partial_off >> - info->gtt_entry_size_shift)) && - (off != ggtt_mm->ggtt_mm.last_partial_off)) { - /* the second partial part */ - - int last_off = ggtt_mm->ggtt_mm.last_partial_off & - (info->gtt_entry_size - 1); - - memcpy((void *)&e.val64 + last_off, - (void *)&ggtt_mm->ggtt_mm.last_partial_data + - last_off, bytes); - - ggtt_mm->ggtt_mm.last_partial_off = -1UL; - } else { - int last_offset; - - gvt_vgpu_err("failed to populate guest ggtt entry: abnormal ggtt entry write sequence, last_partial_off=%lx, offset=%x, bytes=%d, ggtt entry size=%d\n", - ggtt_mm->ggtt_mm.last_partial_off, off, - bytes, info->gtt_entry_size); - - /* set host ggtt entry to scratch page and clear - * virtual ggtt entry as not present for last - * partially write offset - */ - last_offset = ggtt_mm->ggtt_mm.last_partial_off & - (~(info->gtt_entry_size - 1)); - - ggtt_get_host_entry(ggtt_mm, &m, last_offset); - ggtt_invalidate_pte(vgpu, &m); - ops->set_pfn(&m, gvt->gtt.scratch_mfn); - ops->clear_present(&m); - ggtt_set_host_entry(ggtt_mm, &m, last_offset); - ggtt_invalidate(gvt->dev_priv); - - ggtt_get_guest_entry(ggtt_mm, &e, last_offset); - ops->clear_present(&e); - ggtt_set_guest_entry(ggtt_mm, &e, last_offset); - - ggtt_mm->ggtt_mm.last_partial_off = off; - ggtt_mm->ggtt_mm.last_partial_data = e.val64; + bool found = false; + + list_for_each_entry_safe(pos, n, + &ggtt_mm->ggtt_mm.partial_pte_list, list) { + if (g_gtt_index == pos->offset >> + info->gtt_entry_size_shift) { + if (off != pos->offset) { + /* the second partial part*/ + int last_off = pos->offset & + (info->gtt_entry_size - 1); + + memcpy((void *)&e.val64 + last_off, + (void *)&pos->data + last_off, + bytes); + + list_del(&pos->list); + kfree(pos); + found = true; + break; + } + + /* update of the first partial part */ + pos->data = e.val64; + ggtt_set_guest_entry(ggtt_mm, &e, g_gtt_index); + return 0; + } + } - return 0; + if (!found) { + /* the first partial part */ + partial_pte = kzalloc(sizeof(*partial_pte), GFP_KERNEL); + if (!partial_pte) + return -ENOMEM; + partial_pte->offset = off; + partial_pte->data = e.val64; + list_add_tail(&partial_pte->list, + &ggtt_mm->ggtt_mm.partial_pte_list); + partial_update = true; } } - if (ops->test_present(&e)) { + if (!partial_update && (ops->test_present(&e))) { gfn = ops->get_pfn(&e); m = e; @@ -2263,16 +2252,18 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off, } else ops->set_pfn(&m, dma_addr >> PAGE_SHIFT); } else { - ggtt_get_host_entry(ggtt_mm, &m, g_gtt_index); - ggtt_invalidate_pte(vgpu, &m); ops->set_pfn(&m, gvt->gtt.scratch_mfn); ops->clear_present(&m); } out: + ggtt_set_guest_entry(ggtt_mm, &e, g_gtt_index); + + ggtt_get_host_entry(ggtt_mm, &e, g_gtt_index); + ggtt_invalidate_pte(vgpu, &e); + ggtt_set_host_entry(ggtt_mm, &m, g_gtt_index); ggtt_invalidate(gvt->dev_priv); - ggtt_set_guest_entry(ggtt_mm, &e, g_gtt_index); return 0; } @@ -2430,6 +2421,8 @@ int intel_vgpu_init_gtt(struct intel_vgpu *vgpu) intel_vgpu_reset_ggtt(vgpu, false); + INIT_LIST_HEAD(>t->ggtt_mm->ggtt_mm.partial_pte_list); + return create_scratch_page_tree(vgpu); } @@ -2454,6 +2447,15 @@ static void intel_vgpu_destroy_all_ppgtt_mm(struct intel_vgpu *vgpu) static void intel_vgpu_destroy_ggtt_mm(struct intel_vgpu *vgpu) { + struct intel_gvt_partial_pte *pos, *next; + + list_for_each_entry_safe(pos, next, + &vgpu->gtt.ggtt_mm->ggtt_mm.partial_pte_list, + list) { + gvt_dbg_mm("partial PTE update on hold 0x%lx : 0x%llx\n", + pos->offset, pos->data); + kfree(pos); + } intel_vgpu_destroy_mm(vgpu->gtt.ggtt_mm); vgpu->gtt.ggtt_mm = NULL; } diff --git a/drivers/gpu/drm/i915/gvt/gtt.h b/drivers/gpu/drm/i915/gvt/gtt.h index 7a9b36176efb..d8cb04cc946d 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.h +++ b/drivers/gpu/drm/i915/gvt/gtt.h @@ -35,7 +35,6 @@ #define _GVT_GTT_H_ #define I915_GTT_PAGE_SHIFT 12 -#define I915_GTT_PAGE_MASK (~(I915_GTT_PAGE_SIZE - 1)) struct intel_vgpu_mm; @@ -133,6 +132,12 @@ enum intel_gvt_mm_type { #define GVT_RING_CTX_NR_PDPS GEN8_3LVL_PDPES +struct intel_gvt_partial_pte { + unsigned long offset; + u64 data; + struct list_head list; +}; + struct intel_vgpu_mm { enum intel_gvt_mm_type type; struct intel_vgpu *vgpu; @@ -157,8 +162,7 @@ struct intel_vgpu_mm { } ppgtt_mm; struct { void *virtual_ggtt; - unsigned long last_partial_off; - u64 last_partial_data; + struct list_head partial_pte_list; } ggtt_mm; }; }; diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index 90f50f67909a..aa280bb07125 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -1609,7 +1609,7 @@ static int bxt_gt_disp_pwron_write(struct intel_vgpu *vgpu, return 0; } -static int bxt_edp_psr_imr_iir_write(struct intel_vgpu *vgpu, +static int edp_psr_imr_iir_write(struct intel_vgpu *vgpu, unsigned int offset, void *p_data, unsigned int bytes) { vgpu_vreg(vgpu, offset) = 0; @@ -2607,6 +2607,9 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_DFH(_MMIO(0x1a178), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL); MMIO_DFH(_MMIO(0x1a17c), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL); MMIO_DFH(_MMIO(0x2217c), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL); + + 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); return 0; } @@ -3205,9 +3208,6 @@ static int init_bxt_mmio_info(struct intel_gvt *gvt) MMIO_D(HSW_TVIDEO_DIP_GCP(TRANSCODER_B), D_BXT); MMIO_D(HSW_TVIDEO_DIP_GCP(TRANSCODER_C), D_BXT); - MMIO_DH(EDP_PSR_IMR, D_BXT, NULL, bxt_edp_psr_imr_iir_write); - MMIO_DH(EDP_PSR_IIR, D_BXT, NULL, bxt_edp_psr_imr_iir_write); - MMIO_D(RC6_CTX_BASE, D_BXT); MMIO_D(GEN8_PUSHBUS_CONTROL, D_BXT); diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c index 10e63eea5492..d6e02c15ef97 100644 --- a/drivers/gpu/drm/i915/gvt/mmio_context.c +++ b/drivers/gpu/drm/i915/gvt/mmio_context.c @@ -131,7 +131,7 @@ static struct engine_mmio gen9_engine_mmio_list[] __cacheline_aligned = { {RCS, GAMT_CHKN_BIT_REG, 0x0, false}, /* 0x4ab8 */ {RCS, GEN9_GAMT_ECO_REG_RW_IA, 0x0, false}, /* 0x4ab0 */ - {RCS, GEN9_CSFE_CHICKEN1_RCS, 0x0, false}, /* 0x20d4 */ + {RCS, GEN9_CSFE_CHICKEN1_RCS, 0xffff, false}, /* 0x20d4 */ {RCS, GEN8_GARBCNTL, 0x0, false}, /* 0xb004 */ {RCS, GEN7_FF_THREAD_MODE, 0x0, false}, /* 0x20a0 */ @@ -158,6 +158,8 @@ static void load_render_mocs(struct drm_i915_private *dev_priv) int ring_id, i; for (ring_id = 0; ring_id < ARRAY_SIZE(regs); ring_id++) { + if (!HAS_ENGINE(dev_priv, ring_id)) + continue; offset.reg = regs[ring_id]; for (i = 0; i < GEN9_MOCS_SIZE; i++) { gen9_render_mocs.control_table[ring_id][i] = diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 44e2c0f5ec50..47062ee979cf 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -1175,8 +1175,6 @@ skl_dram_get_channels_info(struct drm_i915_private *dev_priv) return -EINVAL; } - dram_info->valid_dimm = true; - /* * If any of the channel is single rank channel, worst case output * will be same as if single rank memory, so consider single rank @@ -1193,8 +1191,7 @@ skl_dram_get_channels_info(struct drm_i915_private *dev_priv) return -EINVAL; } - if (ch0.is_16gb_dimm || ch1.is_16gb_dimm) - dram_info->is_16gb_dimm = true; + dram_info->is_16gb_dimm = ch0.is_16gb_dimm || ch1.is_16gb_dimm; dev_priv->dram_info.symmetric_memory = intel_is_dram_symmetric(val_ch0, val_ch1, @@ -1314,7 +1311,6 @@ bxt_get_dram_info(struct drm_i915_private *dev_priv) return -EINVAL; } - dram_info->valid_dimm = true; dram_info->valid = true; return 0; } @@ -1327,12 +1323,17 @@ intel_get_dram_info(struct drm_i915_private *dev_priv) int ret; dram_info->valid = false; - dram_info->valid_dimm = false; - dram_info->is_16gb_dimm = false; dram_info->rank = I915_DRAM_RANK_INVALID; dram_info->bandwidth_kbps = 0; dram_info->num_channels = 0; + /* + * Assume 16Gb DIMMs are present until proven otherwise. + * This is only used for the level 0 watermark latency + * w/a which does not apply to bxt/glk. + */ + dram_info->is_16gb_dimm = !IS_GEN9_LP(dev_priv); + if (INTEL_GEN(dev_priv) < 9 || IS_GEMINILAKE(dev_priv)) return; @@ -1443,6 +1444,7 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv) intel_uncore_sanitize(dev_priv); + intel_gt_init_workarounds(dev_priv); i915_gem_load_init_fences(dev_priv); /* On the 945G/GM, the chipset reports the MSI capability on the diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 8624b4bdc242..872a2e159a5f 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -67,6 +67,7 @@ #include "intel_ringbuffer.h" #include "intel_uncore.h" #include "intel_wopcm.h" +#include "intel_workarounds.h" #include "intel_uc.h" #include "i915_gem.h" @@ -1805,6 +1806,7 @@ struct drm_i915_private { int dpio_phy_iosf_port[I915_NUM_PHYS_VLV]; struct i915_workarounds workarounds; + struct i915_wa_list gt_wa_list; struct i915_frontbuffer_tracking fb_tracking; @@ -1948,7 +1950,6 @@ struct drm_i915_private { struct dram_info { bool valid; - bool valid_dimm; bool is_16gb_dimm; u8 num_channels; enum dram_rank { @@ -2149,6 +2150,8 @@ struct drm_i915_private { struct delayed_work idle_work; ktime_t last_init_time; + + struct i915_vma *scratch; } gt; /* perform PHY state sanity checks? */ @@ -3871,4 +3874,9 @@ static inline int intel_hws_csb_write_index(struct drm_i915_private *i915) return I915_HWS_CSB_WRITE_INDEX; } +static inline u32 i915_scratch_offset(const struct drm_i915_private *i915) +{ + return i915_ggtt_offset(i915->gt.scratch); +} + #endif diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 0c8aa57ce83b..6ae9a6080cc8 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -5305,7 +5305,7 @@ int i915_gem_init_hw(struct drm_i915_private *dev_priv) } } - intel_gt_workarounds_apply(dev_priv); + intel_gt_apply_workarounds(dev_priv); i915_gem_init_swizzling(dev_priv); @@ -5500,6 +5500,44 @@ err_active: goto out_ctx; } +static int +i915_gem_init_scratch(struct drm_i915_private *i915, unsigned int size) +{ + struct drm_i915_gem_object *obj; + struct i915_vma *vma; + int ret; + + obj = i915_gem_object_create_stolen(i915, size); + if (!obj) + obj = i915_gem_object_create_internal(i915, size); + if (IS_ERR(obj)) { + DRM_ERROR("Failed to allocate scratch page\n"); + return PTR_ERR(obj); + } + + vma = i915_vma_instance(obj, &i915->ggtt.vm, NULL); + if (IS_ERR(vma)) { + ret = PTR_ERR(vma); + goto err_unref; + } + + ret = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH); + if (ret) + goto err_unref; + + i915->gt.scratch = vma; + return 0; + +err_unref: + i915_gem_object_put(obj); + return ret; +} + +static void i915_gem_fini_scratch(struct drm_i915_private *i915) +{ + i915_vma_unpin_and_release(&i915->gt.scratch, 0); +} + int i915_gem_init(struct drm_i915_private *dev_priv) { int ret; @@ -5546,12 +5584,19 @@ int i915_gem_init(struct drm_i915_private *dev_priv) goto err_unlock; } - ret = i915_gem_contexts_init(dev_priv); + ret = i915_gem_init_scratch(dev_priv, + IS_GEN2(dev_priv) ? SZ_256K : PAGE_SIZE); if (ret) { GEM_BUG_ON(ret == -EIO); goto err_ggtt; } + ret = i915_gem_contexts_init(dev_priv); + if (ret) { + GEM_BUG_ON(ret == -EIO); + goto err_scratch; + } + ret = intel_engines_init(dev_priv); if (ret) { GEM_BUG_ON(ret == -EIO); @@ -5624,6 +5669,8 @@ err_pm: err_context: if (ret != -EIO) i915_gem_contexts_fini(dev_priv); +err_scratch: + i915_gem_fini_scratch(dev_priv); err_ggtt: err_unlock: intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL); @@ -5675,8 +5722,11 @@ void i915_gem_fini(struct drm_i915_private *dev_priv) intel_uc_fini(dev_priv); i915_gem_cleanup_engines(dev_priv); i915_gem_contexts_fini(dev_priv); + i915_gem_fini_scratch(dev_priv); mutex_unlock(&dev_priv->drm.struct_mutex); + intel_wa_list_free(&dev_priv->gt_wa_list); + intel_cleanup_gt_powersave(dev_priv); intel_uc_fini_misc(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 09187286d346..1aaccbe7e1de 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -460,7 +460,7 @@ eb_validate_vma(struct i915_execbuffer *eb, * any non-page-aligned or non-canonical addresses. */ if (unlikely(entry->flags & EXEC_OBJECT_PINNED && - entry->offset != gen8_canonical_addr(entry->offset & PAGE_MASK))) + entry->offset != gen8_canonical_addr(entry->offset & I915_GTT_PAGE_MASK))) return -EINVAL; /* pad_to_size was once a reserved field, so sanitize it */ diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 56c7f8637311..07999fe09ad2 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -1757,7 +1757,7 @@ static void gen6_dump_ppgtt(struct i915_hw_ppgtt *base, struct seq_file *m) if (i == 4) continue; - seq_printf(m, "\t\t(%03d, %04d) %08lx: ", + seq_printf(m, "\t\t(%03d, %04d) %08llx: ", pde, pte, (pde * GEN6_PTES + pte) * I915_GTT_PAGE_SIZE); for (i = 0; i < 4; i++) { @@ -3413,6 +3413,11 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt) ggtt->vm.insert_page = bxt_vtd_ggtt_insert_page__BKL; if (ggtt->vm.clear_range != nop_clear_range) ggtt->vm.clear_range = bxt_vtd_ggtt_clear_range__BKL; + + /* Prevent recursively calling stop_machine() and deadlocks. */ + dev_info(dev_priv->drm.dev, + "Disabling error capture for VT-d workaround\n"); + i915_disable_error_state(dev_priv, -ENODEV); } ggtt->invalidate = gen6_ggtt_invalidate; diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h index 7e2af5f4f39b..28039290655c 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.h +++ b/drivers/gpu/drm/i915/i915_gem_gtt.h @@ -42,13 +42,15 @@ #include "i915_selftest.h" #include "i915_timeline.h" -#define I915_GTT_PAGE_SIZE_4K BIT(12) -#define I915_GTT_PAGE_SIZE_64K BIT(16) -#define I915_GTT_PAGE_SIZE_2M BIT(21) +#define I915_GTT_PAGE_SIZE_4K BIT_ULL(12) +#define I915_GTT_PAGE_SIZE_64K BIT_ULL(16) +#define I915_GTT_PAGE_SIZE_2M BIT_ULL(21) #define I915_GTT_PAGE_SIZE I915_GTT_PAGE_SIZE_4K #define I915_GTT_MAX_PAGE_SIZE I915_GTT_PAGE_SIZE_2M +#define I915_GTT_PAGE_MASK -I915_GTT_PAGE_SIZE + #define I915_GTT_MIN_ALIGNMENT I915_GTT_PAGE_SIZE #define I915_FENCE_REG_NONE -1 @@ -659,20 +661,20 @@ int i915_gem_gtt_insert(struct i915_address_space *vm, u64 start, u64 end, unsigned int flags); /* Flags used by pin/bind&friends. */ -#define PIN_NONBLOCK BIT(0) -#define PIN_MAPPABLE BIT(1) -#define PIN_ZONE_4G BIT(2) -#define PIN_NONFAULT BIT(3) -#define PIN_NOEVICT BIT(4) - -#define PIN_MBZ BIT(5) /* I915_VMA_PIN_OVERFLOW */ -#define PIN_GLOBAL BIT(6) /* I915_VMA_GLOBAL_BIND */ -#define PIN_USER BIT(7) /* I915_VMA_LOCAL_BIND */ -#define PIN_UPDATE BIT(8) - -#define PIN_HIGH BIT(9) -#define PIN_OFFSET_BIAS BIT(10) -#define PIN_OFFSET_FIXED BIT(11) +#define PIN_NONBLOCK BIT_ULL(0) +#define PIN_MAPPABLE BIT_ULL(1) +#define PIN_ZONE_4G BIT_ULL(2) +#define PIN_NONFAULT BIT_ULL(3) +#define PIN_NOEVICT BIT_ULL(4) + +#define PIN_MBZ BIT_ULL(5) /* I915_VMA_PIN_OVERFLOW */ +#define PIN_GLOBAL BIT_ULL(6) /* I915_VMA_GLOBAL_BIND */ +#define PIN_USER BIT_ULL(7) /* I915_VMA_LOCAL_BIND */ +#define PIN_UPDATE BIT_ULL(8) + +#define PIN_HIGH BIT_ULL(9) +#define PIN_OFFSET_BIAS BIT_ULL(10) +#define PIN_OFFSET_FIXED BIT_ULL(11) #define PIN_OFFSET_MASK (-I915_GTT_PAGE_SIZE) #endif diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 8762d17b6659..db4128d6c09b 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -648,6 +648,9 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, return 0; } + if (IS_ERR(error)) + return PTR_ERR(error); + if (*error->error_msg) err_printf(m, "%s\n", error->error_msg); err_printf(m, "Kernel: " UTS_RELEASE "\n"); @@ -1492,7 +1495,7 @@ static void gem_record_rings(struct i915_gpu_state *error) if (HAS_BROKEN_CS_TLB(i915)) ee->wa_batchbuffer = i915_error_object_create(i915, - engine->scratch); + i915->gt.scratch); request_record_user_bo(request, ee); ee->ctx = @@ -1859,6 +1862,7 @@ void i915_capture_error_state(struct drm_i915_private *i915, error = i915_capture_gpu_state(i915); if (!error) { DRM_DEBUG_DRIVER("out of memory, not capturing error state\n"); + i915_disable_error_state(i915, -ENOMEM); return; } @@ -1914,5 +1918,14 @@ void i915_reset_error_state(struct drm_i915_private *i915) i915->gpu_error.first_error = NULL; spin_unlock_irq(&i915->gpu_error.lock); - i915_gpu_state_put(error); + if (!IS_ERR(error)) + i915_gpu_state_put(error); +} + +void i915_disable_error_state(struct drm_i915_private *i915, int err) +{ + spin_lock_irq(&i915->gpu_error.lock); + if (!i915->gpu_error.first_error) + i915->gpu_error.first_error = ERR_PTR(err); + spin_unlock_irq(&i915->gpu_error.lock); } diff --git a/drivers/gpu/drm/i915/i915_gpu_error.h b/drivers/gpu/drm/i915/i915_gpu_error.h index 8710fb18ed74..3ec89a504de5 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.h +++ b/drivers/gpu/drm/i915/i915_gpu_error.h @@ -343,6 +343,7 @@ static inline void i915_gpu_state_put(struct i915_gpu_state *gpu) struct i915_gpu_state *i915_first_error_state(struct drm_i915_private *i915); void i915_reset_error_state(struct drm_i915_private *i915); +void i915_disable_error_state(struct drm_i915_private *i915, int err); #else @@ -355,13 +356,18 @@ static inline void i915_capture_error_state(struct drm_i915_private *dev_priv, static inline struct i915_gpu_state * i915_first_error_state(struct drm_i915_private *i915) { - return NULL; + return ERR_PTR(-ENODEV); } static inline void i915_reset_error_state(struct drm_i915_private *i915) { } +static inline void i915_disable_error_state(struct drm_i915_private *i915, + int err) +{ +} + #endif /* IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR) */ #endif /* _I915_GPU_ERROR_H_ */ diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 7c491ea3d052..e31c27e45734 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -2095,8 +2095,12 @@ enum i915_power_well_id { /* ICL PHY DFLEX registers */ #define PORT_TX_DFLEXDPMLE1 _MMIO(0x1638C0) -#define DFLEXDPMLE1_DPMLETC_MASK(n) (0xf << (4 * (n))) -#define DFLEXDPMLE1_DPMLETC(n, x) ((x) << (4 * (n))) +#define DFLEXDPMLE1_DPMLETC_MASK(tc_port) (0xf << (4 * (tc_port))) +#define DFLEXDPMLE1_DPMLETC_ML0(tc_port) (1 << (4 * (tc_port))) +#define DFLEXDPMLE1_DPMLETC_ML1_0(tc_port) (3 << (4 * (tc_port))) +#define DFLEXDPMLE1_DPMLETC_ML3(tc_port) (8 << (4 * (tc_port))) +#define DFLEXDPMLE1_DPMLETC_ML3_2(tc_port) (12 << (4 * (tc_port))) +#define DFLEXDPMLE1_DPMLETC_ML3_0(tc_port) (15 << (4 * (tc_port))) /* BXT PHY Ref registers */ #define _PORT_REF_DW3_A 0x16218C @@ -4593,12 +4597,12 @@ enum { #define DRM_DIP_ENABLE (1 << 28) #define PSR_VSC_BIT_7_SET (1 << 27) -#define VSC_SELECT_MASK (0x3 << 26) -#define VSC_SELECT_SHIFT 26 -#define VSC_DIP_HW_HEA_DATA (0 << 26) -#define VSC_DIP_HW_HEA_SW_DATA (1 << 26) -#define VSC_DIP_HW_DATA_SW_HEA (2 << 26) -#define VSC_DIP_SW_HEA_DATA (3 << 26) +#define VSC_SELECT_MASK (0x3 << 25) +#define VSC_SELECT_SHIFT 25 +#define VSC_DIP_HW_HEA_DATA (0 << 25) +#define VSC_DIP_HW_HEA_SW_DATA (1 << 25) +#define VSC_DIP_HW_DATA_SW_HEA (2 << 25) +#define VSC_DIP_SW_HEA_DATA (3 << 25) #define VDIP_ENABLE_PPS (1 << 24) /* Panel power sequencing */ diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c index 769f3f586661..ee3ca2de983b 100644 --- a/drivers/gpu/drm/i915/intel_audio.c +++ b/drivers/gpu/drm/i915/intel_audio.c @@ -144,6 +144,9 @@ static const struct { /* HDMI N/CTS table */ #define TMDS_297M 297000 #define TMDS_296M 296703 +#define TMDS_594M 594000 +#define TMDS_593M 593407 + static const struct { int sample_rate; int clock; @@ -164,6 +167,20 @@ static const struct { { 176400, TMDS_297M, 18816, 247500 }, { 192000, TMDS_296M, 23296, 281250 }, { 192000, TMDS_297M, 20480, 247500 }, + { 44100, TMDS_593M, 8918, 937500 }, + { 44100, TMDS_594M, 9408, 990000 }, + { 48000, TMDS_593M, 5824, 562500 }, + { 48000, TMDS_594M, 6144, 594000 }, + { 32000, TMDS_593M, 5824, 843750 }, + { 32000, TMDS_594M, 3072, 445500 }, + { 88200, TMDS_593M, 17836, 937500 }, + { 88200, TMDS_594M, 18816, 990000 }, + { 96000, TMDS_593M, 11648, 562500 }, + { 96000, TMDS_594M, 12288, 594000 }, + { 176400, TMDS_593M, 35672, 937500 }, + { 176400, TMDS_594M, 37632, 990000 }, + { 192000, TMDS_593M, 23296, 562500 }, + { 192000, TMDS_594M, 24576, 594000 }, }; /* get AUD_CONFIG_PIXEL_CLOCK_HDMI_* value for mode */ diff --git a/drivers/gpu/drm/i915/intel_cdclk.c b/drivers/gpu/drm/i915/intel_cdclk.c index 29075c763428..8d74276029e6 100644 --- a/drivers/gpu/drm/i915/intel_cdclk.c +++ b/drivers/gpu/drm/i915/intel_cdclk.c @@ -2138,16 +2138,8 @@ void intel_set_cdclk(struct drm_i915_private *dev_priv, static int intel_pixel_rate_to_cdclk(struct drm_i915_private *dev_priv, int pixel_rate) { - if (INTEL_GEN(dev_priv) >= 10) + if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) return DIV_ROUND_UP(pixel_rate, 2); - else if (IS_GEMINILAKE(dev_priv)) - /* - * FIXME: Avoid using a pixel clock that is more than 99% of the cdclk - * as a temporary workaround. Use a higher cdclk instead. (Note that - * intel_compute_max_dotclk() limits the max pixel clock to 99% of max - * cdclk.) - */ - return DIV_ROUND_UP(pixel_rate * 100, 2 * 99); else if (IS_GEN9(dev_priv) || IS_BROADWELL(dev_priv) || IS_HASWELL(dev_priv)) return pixel_rate; @@ -2543,14 +2535,8 @@ static int intel_compute_max_dotclk(struct drm_i915_private *dev_priv) { int max_cdclk_freq = dev_priv->max_cdclk_freq; - if (INTEL_GEN(dev_priv) >= 10) + if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) return 2 * max_cdclk_freq; - else if (IS_GEMINILAKE(dev_priv)) - /* - * FIXME: Limiting to 99% as a temporary workaround. See - * intel_min_cdclk() for details. - */ - return 2 * max_cdclk_freq * 99 / 100; else if (IS_GEN9(dev_priv) || IS_BROADWELL(dev_priv) || IS_HASWELL(dev_priv)) return max_cdclk_freq; diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c index 0ef0c6448d53..01fa98299bae 100644 --- a/drivers/gpu/drm/i915/intel_device_info.c +++ b/drivers/gpu/drm/i915/intel_device_info.c @@ -474,7 +474,7 @@ static void broadwell_sseu_info_init(struct drm_i915_private *dev_priv) u8 eu_disabled_mask; u32 n_disabled; - if (!(sseu->subslice_mask[ss] & BIT(ss))) + if (!(sseu->subslice_mask[s] & BIT(ss))) /* skip disabled subslice */ continue; diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 9741cc419e1b..c9878dd1f7cd 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2890,6 +2890,7 @@ intel_find_initial_plane_obj(struct intel_crtc *intel_crtc, return; valid_fb: + intel_state->base.rotation = plane_config->rotation; intel_fill_fb_ggtt_view(&intel_state->view, fb, intel_state->base.rotation); intel_state->color_plane[0].stride = @@ -4850,8 +4851,31 @@ static void cpt_verify_modeset(struct drm_device *dev, int pipe) * chroma samples for both of the luma samples, and thus we don't * actually get the expected MPEG2 chroma siting convention :( * The same behaviour is observed on pre-SKL platforms as well. + * + * Theory behind the formula (note that we ignore sub-pixel + * source coordinates): + * s = source sample position + * d = destination sample position + * + * Downscaling 4:1: + * -0.5 + * | 0.0 + * | | 1.5 (initial phase) + * | | | + * v v v + * | s | s | s | s | + * | d | + * + * Upscaling 1:4: + * -0.5 + * | -0.375 (initial phase) + * | | 0.0 + * | | | + * v v v + * | s | + * | d | d | d | d | */ -u16 skl_scaler_calc_phase(int sub, bool chroma_cosited) +u16 skl_scaler_calc_phase(int sub, int scale, bool chroma_cosited) { int phase = -0x8000; u16 trip = 0; @@ -4859,6 +4883,15 @@ u16 skl_scaler_calc_phase(int sub, bool chroma_cosited) if (chroma_cosited) phase += (sub - 1) * 0x8000 / sub; + phase += scale / (2 * sub); + + /* + * Hardware initial phase limited to [-0.5:1.5]. + * Since the max hardware scale factor is 3.0, we + * should never actually excdeed 1.0 here. + */ + WARN_ON(phase < -0x8000 || phase > 0x18000); + if (phase < 0) phase = 0x10000 + phase; else @@ -5067,13 +5100,20 @@ static void skylake_pfit_enable(struct intel_crtc *crtc) if (crtc->config->pch_pfit.enabled) { u16 uv_rgb_hphase, uv_rgb_vphase; + int pfit_w, pfit_h, hscale, vscale; int id; if (WARN_ON(crtc->config->scaler_state.scaler_id < 0)) return; - uv_rgb_hphase = skl_scaler_calc_phase(1, false); - uv_rgb_vphase = skl_scaler_calc_phase(1, false); + pfit_w = (crtc->config->pch_pfit.size >> 16) & 0xFFFF; + pfit_h = crtc->config->pch_pfit.size & 0xFFFF; + + hscale = (crtc->config->pipe_src_w << 16) / pfit_w; + vscale = (crtc->config->pipe_src_h << 16) / pfit_h; + + uv_rgb_hphase = skl_scaler_calc_phase(1, hscale, false); + uv_rgb_vphase = skl_scaler_calc_phase(1, vscale, false); id = scaler_state->scaler_id; I915_WRITE(SKL_PS_CTRL(pipe, id), PS_SCALER_EN | @@ -7843,8 +7883,15 @@ i9xx_get_initial_plane_config(struct intel_crtc *crtc, plane_config->tiling = I915_TILING_X; fb->modifier = I915_FORMAT_MOD_X_TILED; } + + if (val & DISPPLANE_ROTATE_180) + plane_config->rotation = DRM_MODE_ROTATE_180; } + if (IS_CHERRYVIEW(dev_priv) && pipe == PIPE_B && + val & DISPPLANE_MIRROR) + plane_config->rotation |= DRM_MODE_REFLECT_X; + pixel_format = val & DISPPLANE_PIXFORMAT_MASK; fourcc = i9xx_format_to_fourcc(pixel_format); fb->format = drm_format_info(fourcc); @@ -8913,6 +8960,29 @@ skylake_get_initial_plane_config(struct intel_crtc *crtc, goto error; } + /* + * DRM_MODE_ROTATE_ is counter clockwise to stay compatible with Xrandr + * while i915 HW rotation is clockwise, thats why this swapping. + */ + switch (val & PLANE_CTL_ROTATE_MASK) { + case PLANE_CTL_ROTATE_0: + plane_config->rotation = DRM_MODE_ROTATE_0; + break; + case PLANE_CTL_ROTATE_90: + plane_config->rotation = DRM_MODE_ROTATE_270; + break; + case PLANE_CTL_ROTATE_180: + plane_config->rotation = DRM_MODE_ROTATE_180; + break; + case PLANE_CTL_ROTATE_270: + plane_config->rotation = DRM_MODE_ROTATE_90; + break; + } + + if (INTEL_GEN(dev_priv) >= 10 && + val & PLANE_CTL_FLIP_HORIZONTAL) + plane_config->rotation |= DRM_MODE_REFLECT_X; + base = I915_READ(PLANE_SURF(pipe, plane_id)) & 0xfffff000; plane_config->base = base; @@ -12768,17 +12838,12 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state) intel_check_cpu_fifo_underruns(dev_priv); intel_check_pch_fifo_underruns(dev_priv); - if (!new_crtc_state->active) { - /* - * Make sure we don't call initial_watermarks - * for ILK-style watermark updates. - * - * No clue what this is supposed to achieve. - */ - if (INTEL_GEN(dev_priv) >= 9) - dev_priv->display.initial_watermarks(intel_state, - to_intel_crtc_state(new_crtc_state)); - } + /* FIXME unify this for all platforms */ + if (!new_crtc_state->active && + !HAS_GMCH_DISPLAY(dev_priv) && + dev_priv->display.initial_watermarks) + dev_priv->display.initial_watermarks(intel_state, + to_intel_crtc_state(new_crtc_state)); } } @@ -14646,7 +14711,7 @@ static int intel_framebuffer_init(struct intel_framebuffer *intel_fb, fb->height < SKL_MIN_YUV_420_SRC_H || (fb->width % 4) != 0 || (fb->height % 4) != 0)) { DRM_DEBUG_KMS("src dimensions not correct for NV12\n"); - return -EINVAL; + goto err; } for (i = 0; i < fb->format->num_planes; i++) { @@ -15233,6 +15298,14 @@ retry: ret = drm_atomic_add_affected_planes(state, crtc); if (ret) goto out; + + /* + * FIXME hack to force a LUT update to avoid the + * plane update forcing the pipe gamma on without + * having a proper LUT loaded. Remove once we + * have readout for pipe gamma enable. + */ + crtc_state->color_mgmt_changed = true; } } diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c index 1b00f8ea145b..a911691dbd0f 100644 --- a/drivers/gpu/drm/i915/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/intel_dp_mst.c @@ -452,6 +452,10 @@ static struct drm_connector *intel_dp_add_mst_connector(struct drm_dp_mst_topolo if (!intel_connector) return NULL; + intel_connector->get_hw_state = intel_dp_mst_get_hw_state; + intel_connector->mst_port = intel_dp; + intel_connector->port = port; + connector = &intel_connector->base; ret = drm_connector_init(dev, connector, &intel_dp_mst_connector_funcs, DRM_MODE_CONNECTOR_DisplayPort); @@ -462,10 +466,6 @@ static struct drm_connector *intel_dp_add_mst_connector(struct drm_dp_mst_topolo drm_connector_helper_add(connector, &intel_dp_mst_connector_helper_funcs); - intel_connector->get_hw_state = intel_dp_mst_get_hw_state; - intel_connector->mst_port = intel_dp; - intel_connector->port = port; - for_each_pipe(dev_priv, pipe) { struct drm_encoder *enc = &intel_dp->mst_encoders[pipe]->base.base; diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index f8dc84b2d2d3..db6fa1d0cbda 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -547,6 +547,7 @@ struct intel_initial_plane_config { unsigned int tiling; int size; u32 base; + u8 rotation; }; #define SKL_MIN_SRC_W 8 @@ -1646,7 +1647,7 @@ void intel_mode_from_pipe_config(struct drm_display_mode *mode, void intel_crtc_arm_fifo_underrun(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state); -u16 skl_scaler_calc_phase(int sub, bool chroma_center); +u16 skl_scaler_calc_phase(int sub, int scale, bool chroma_center); int skl_update_scaler_crtc(struct intel_crtc_state *crtc_state); int skl_max_scale(const struct intel_crtc_state *crtc_state, u32 pixel_format); diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c index 217ed3ee1cab..76b5f94ea6cb 100644 --- a/drivers/gpu/drm/i915/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/intel_engine_cs.c @@ -490,46 +490,6 @@ void intel_engine_setup_common(struct intel_engine_cs *engine) intel_engine_init_cmd_parser(engine); } -int intel_engine_create_scratch(struct intel_engine_cs *engine, - unsigned int size) -{ - struct drm_i915_gem_object *obj; - struct i915_vma *vma; - int ret; - - WARN_ON(engine->scratch); - - obj = i915_gem_object_create_stolen(engine->i915, size); - if (!obj) - obj = i915_gem_object_create_internal(engine->i915, size); - if (IS_ERR(obj)) { - DRM_ERROR("Failed to allocate scratch page\n"); - return PTR_ERR(obj); - } - - vma = i915_vma_instance(obj, &engine->i915->ggtt.vm, NULL); - if (IS_ERR(vma)) { - ret = PTR_ERR(vma); - goto err_unref; - } - - ret = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH); - if (ret) - goto err_unref; - - engine->scratch = vma; - return 0; - -err_unref: - i915_gem_object_put(obj); - return ret; -} - -void intel_engine_cleanup_scratch(struct intel_engine_cs *engine) -{ - i915_vma_unpin_and_release(&engine->scratch, 0); -} - static void cleanup_status_page(struct intel_engine_cs *engine) { if (HWS_NEEDS_PHYSICAL(engine->i915)) { @@ -704,8 +664,6 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine) { struct drm_i915_private *i915 = engine->i915; - intel_engine_cleanup_scratch(engine); - cleanup_status_page(engine); intel_engine_fini_breadcrumbs(engine); @@ -720,6 +678,8 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine) __intel_context_unpin(i915->kernel_context, engine); i915_timeline_fini(&engine->timeline); + + intel_wa_list_free(&engine->wa_list); } u64 intel_engine_get_active_head(const struct intel_engine_cs *engine) diff --git a/drivers/gpu/drm/i915/intel_hotplug.c b/drivers/gpu/drm/i915/intel_hotplug.c index 648a13c6043c..9a8018130237 100644 --- a/drivers/gpu/drm/i915/intel_hotplug.c +++ b/drivers/gpu/drm/i915/intel_hotplug.c @@ -228,7 +228,9 @@ static void intel_hpd_irq_storm_reenable_work(struct work_struct *work) drm_for_each_connector_iter(connector, &conn_iter) { struct intel_connector *intel_connector = to_intel_connector(connector); - if (intel_connector->encoder->hpd_pin == pin) { + /* Don't check MST ports, they don't have pins */ + if (!intel_connector->mst_port && + intel_connector->encoder->hpd_pin == pin) { if (connector->polled != intel_connector->polled) DRM_DEBUG_DRIVER("Reenabling HPD on connector %s\n", connector->name); @@ -395,37 +397,54 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv, struct intel_encoder *encoder; bool storm_detected = false; bool queue_dig = false, queue_hp = false; + u32 long_hpd_pulse_mask = 0; + u32 short_hpd_pulse_mask = 0; + enum hpd_pin pin; if (!pin_mask) return; spin_lock(&dev_priv->irq_lock); + + /* + * Determine whether ->hpd_pulse() exists for each pin, and + * whether we have a short or a long pulse. This is needed + * as each pin may have up to two encoders (HDMI and DP) and + * only the one of them (DP) will have ->hpd_pulse(). + */ for_each_intel_encoder(&dev_priv->drm, encoder) { - enum hpd_pin pin = encoder->hpd_pin; bool has_hpd_pulse = intel_encoder_has_hpd_pulse(encoder); + enum port port = encoder->port; + bool long_hpd; + pin = encoder->hpd_pin; if (!(BIT(pin) & pin_mask)) continue; - if (has_hpd_pulse) { - bool long_hpd = long_mask & BIT(pin); - enum port port = encoder->port; + if (!has_hpd_pulse) + continue; - DRM_DEBUG_DRIVER("digital hpd port %c - %s\n", port_name(port), - long_hpd ? "long" : "short"); - /* - * For long HPD pulses we want to have the digital queue happen, - * but we still want HPD storm detection to function. - */ - queue_dig = true; - if (long_hpd) { - dev_priv->hotplug.long_port_mask |= (1 << port); - } else { - /* for short HPD just trigger the digital queue */ - dev_priv->hotplug.short_port_mask |= (1 << port); - continue; - } + long_hpd = long_mask & BIT(pin); + + DRM_DEBUG_DRIVER("digital hpd port %c - %s\n", port_name(port), + long_hpd ? "long" : "short"); + queue_dig = true; + + if (long_hpd) { + long_hpd_pulse_mask |= BIT(pin); + dev_priv->hotplug.long_port_mask |= BIT(port); + } else { + short_hpd_pulse_mask |= BIT(pin); + dev_priv->hotplug.short_port_mask |= BIT(port); } + } + + /* Now process each pin just once */ + for_each_hpd_pin(pin) { + bool long_hpd; + + if (!(BIT(pin) & pin_mask)) + continue; if (dev_priv->hotplug.stats[pin].state == HPD_DISABLED) { /* @@ -442,11 +461,22 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv, if (dev_priv->hotplug.stats[pin].state != HPD_ENABLED) continue; - if (!has_hpd_pulse) { + /* + * Delegate to ->hpd_pulse() if one of the encoders for this + * pin has it, otherwise let the hotplug_work deal with this + * pin directly. + */ + if (((short_hpd_pulse_mask | long_hpd_pulse_mask) & BIT(pin))) { + long_hpd = long_hpd_pulse_mask & BIT(pin); + } else { dev_priv->hotplug.event_bits |= BIT(pin); + long_hpd = true; queue_hp = true; } + if (!long_hpd) + continue; + if (intel_hpd_irq_storm_detect(dev_priv, pin)) { dev_priv->hotplug.event_bits &= ~BIT(pin); storm_detected = true; diff --git a/drivers/gpu/drm/i915/intel_lpe_audio.c b/drivers/gpu/drm/i915/intel_lpe_audio.c index cdf19553ffac..5d5336fbe7b0 100644 --- a/drivers/gpu/drm/i915/intel_lpe_audio.c +++ b/drivers/gpu/drm/i915/intel_lpe_audio.c @@ -297,8 +297,10 @@ void intel_lpe_audio_teardown(struct drm_i915_private *dev_priv) lpe_audio_platdev_destroy(dev_priv); irq_free_desc(dev_priv->lpe_audio.irq); -} + dev_priv->lpe_audio.irq = -1; + dev_priv->lpe_audio.platdev = NULL; +} /** * intel_lpe_audio_notify() - notify lpe audio event diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 43957bb37a42..58d1d3d47dd3 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -424,7 +424,8 @@ static u64 execlists_update_context(struct i915_request *rq) reg_state[CTX_RING_TAIL+1] = intel_ring_set_tail(rq->ring, rq->tail); - /* True 32b PPGTT with dynamic page allocation: update PDP + /* + * True 32b PPGTT with dynamic page allocation: update PDP * registers and point the unallocated PDPs to scratch page. * PML4 is allocated during ppgtt init, so this is not needed * in 48-bit mode. @@ -432,6 +433,22 @@ static u64 execlists_update_context(struct i915_request *rq) if (ppgtt && !i915_vm_is_48bit(&ppgtt->vm)) execlists_update_context_pdps(ppgtt, reg_state); + /* + * Make sure the context image is complete before we submit it to HW. + * + * Ostensibly, writes (including the WCB) should be flushed prior to + * an uncached write such as our mmio register access, the empirical + * evidence (esp. on Braswell) suggests that the WC write into memory + * may not be visible to the HW prior to the completion of the UC + * register write and that we may begin execution from the context + * before its image is complete leading to invalid PD chasing. + * + * Furthermore, Braswell, at least, wants a full mb to be sure that + * the writes are coherent in memory (visible to the GPU) prior to + * execution, and not just visible to other CPUs (as is the result of + * wmb). + */ + mb(); return ce->lrc_desc; } @@ -1431,9 +1448,10 @@ static int execlists_request_alloc(struct i915_request *request) static u32 * gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *engine, u32 *batch) { + /* NB no one else is allowed to scribble over scratch + 256! */ *batch++ = MI_STORE_REGISTER_MEM_GEN8 | MI_SRM_LRM_GLOBAL_GTT; *batch++ = i915_mmio_reg_offset(GEN8_L3SQCREG4); - *batch++ = i915_ggtt_offset(engine->scratch) + 256; + *batch++ = i915_scratch_offset(engine->i915) + 256; *batch++ = 0; *batch++ = MI_LOAD_REGISTER_IMM(1); @@ -1447,7 +1465,7 @@ gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *engine, u32 *batch) *batch++ = MI_LOAD_REGISTER_MEM_GEN8 | MI_SRM_LRM_GLOBAL_GTT; *batch++ = i915_mmio_reg_offset(GEN8_L3SQCREG4); - *batch++ = i915_ggtt_offset(engine->scratch) + 256; + *batch++ = i915_scratch_offset(engine->i915) + 256; *batch++ = 0; return batch; @@ -1484,7 +1502,7 @@ static u32 *gen8_init_indirectctx_bb(struct intel_engine_cs *engine, u32 *batch) PIPE_CONTROL_GLOBAL_GTT_IVB | PIPE_CONTROL_CS_STALL | PIPE_CONTROL_QW_WRITE, - i915_ggtt_offset(engine->scratch) + + i915_scratch_offset(engine->i915) + 2 * CACHELINE_BYTES); *batch++ = MI_ARB_ON_OFF | MI_ARB_ENABLE; @@ -1561,7 +1579,7 @@ static u32 *gen9_init_indirectctx_bb(struct intel_engine_cs *engine, u32 *batch) PIPE_CONTROL_GLOBAL_GTT_IVB | PIPE_CONTROL_CS_STALL | PIPE_CONTROL_QW_WRITE, - i915_ggtt_offset(engine->scratch) + i915_scratch_offset(engine->i915) + 2 * CACHELINE_BYTES); } @@ -1781,6 +1799,8 @@ static bool unexpected_starting_state(struct intel_engine_cs *engine) static int gen8_init_common_ring(struct intel_engine_cs *engine) { + intel_engine_apply_workarounds(engine); + intel_mocs_init_engine(engine); intel_engine_reset_breadcrumbs(engine); @@ -2127,7 +2147,7 @@ static int gen8_emit_flush_render(struct i915_request *request, { struct intel_engine_cs *engine = request->engine; u32 scratch_addr = - i915_ggtt_offset(engine->scratch) + 2 * CACHELINE_BYTES; + i915_scratch_offset(engine->i915) + 2 * CACHELINE_BYTES; bool vf_flush_wa = false, dc_flush_wa = false; u32 *cs, flags = 0; int len; @@ -2464,10 +2484,6 @@ int logical_render_ring_init(struct intel_engine_cs *engine) if (ret) return ret; - ret = intel_engine_create_scratch(engine, PAGE_SIZE); - if (ret) - goto err_cleanup_common; - ret = intel_init_workaround_bb(engine); if (ret) { /* @@ -2479,11 +2495,9 @@ int logical_render_ring_init(struct intel_engine_cs *engine) ret); } - return 0; + intel_engine_init_workarounds(engine); -err_cleanup_common: - intel_engine_cleanup_common(engine); - return ret; + return 0; } int logical_xcs_ring_init(struct intel_engine_cs *engine) diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 1db9b8328275..3fe358db1276 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -2493,6 +2493,9 @@ static uint32_t ilk_compute_pri_wm(const struct intel_crtc_state *cstate, uint32_t method1, method2; int cpp; + if (mem_value == 0) + return U32_MAX; + if (!intel_wm_plane_visible(cstate, pstate)) return 0; @@ -2522,6 +2525,9 @@ static uint32_t ilk_compute_spr_wm(const struct intel_crtc_state *cstate, uint32_t method1, method2; int cpp; + if (mem_value == 0) + return U32_MAX; + if (!intel_wm_plane_visible(cstate, pstate)) return 0; @@ -2545,6 +2551,9 @@ static uint32_t ilk_compute_cur_wm(const struct intel_crtc_state *cstate, { int cpp; + if (mem_value == 0) + return U32_MAX; + if (!intel_wm_plane_visible(cstate, pstate)) return 0; @@ -2881,8 +2890,7 @@ static void intel_read_wm_latency(struct drm_i915_private *dev_priv, * any underrun. If not able to get Dimm info assume 16GB dimm * to avoid any underrun. */ - if (!dev_priv->dram_info.valid_dimm || - dev_priv->dram_info.is_16gb_dimm) + if (dev_priv->dram_info.is_16gb_dimm) wm[0] += 1; } else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) { @@ -3009,6 +3017,34 @@ static void snb_wm_latency_quirk(struct drm_i915_private *dev_priv) intel_print_wm_latency(dev_priv, "Cursor", dev_priv->wm.cur_latency); } +static void snb_wm_lp3_irq_quirk(struct drm_i915_private *dev_priv) +{ + /* + * On some SNB machines (Thinkpad X220 Tablet at least) + * LP3 usage can cause vblank interrupts to be lost. + * The DEIIR bit will go high but it looks like the CPU + * never gets interrupted. + * + * It's not clear whether other interrupt source could + * be affected or if this is somehow limited to vblank + * interrupts only. To play it safe we disable LP3 + * watermarks entirely. + */ + if (dev_priv->wm.pri_latency[3] == 0 && + dev_priv->wm.spr_latency[3] == 0 && + dev_priv->wm.cur_latency[3] == 0) + return; + + dev_priv->wm.pri_latency[3] = 0; + dev_priv->wm.spr_latency[3] = 0; + dev_priv->wm.cur_latency[3] = 0; + + DRM_DEBUG_KMS("LP3 watermarks disabled due to potential for lost interrupts\n"); + intel_print_wm_latency(dev_priv, "Primary", dev_priv->wm.pri_latency); + intel_print_wm_latency(dev_priv, "Sprite", dev_priv->wm.spr_latency); + intel_print_wm_latency(dev_priv, "Cursor", dev_priv->wm.cur_latency); +} + static void ilk_setup_wm_latency(struct drm_i915_private *dev_priv) { intel_read_wm_latency(dev_priv, dev_priv->wm.pri_latency); @@ -3025,8 +3061,10 @@ static void ilk_setup_wm_latency(struct drm_i915_private *dev_priv) intel_print_wm_latency(dev_priv, "Sprite", dev_priv->wm.spr_latency); intel_print_wm_latency(dev_priv, "Cursor", dev_priv->wm.cur_latency); - if (IS_GEN6(dev_priv)) + if (IS_GEN6(dev_priv)) { snb_wm_latency_quirk(dev_priv); + snb_wm_lp3_irq_quirk(dev_priv); + } } static void skl_setup_wm_latency(struct drm_i915_private *dev_priv) diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index d0ef50bf930a..1f8d2a66c791 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -69,19 +69,28 @@ unsigned int intel_ring_update_space(struct intel_ring *ring) static int gen2_render_ring_flush(struct i915_request *rq, u32 mode) { + unsigned int num_store_dw; u32 cmd, *cs; cmd = MI_FLUSH; - + num_store_dw = 0; if (mode & EMIT_INVALIDATE) cmd |= MI_READ_FLUSH; + if (mode & EMIT_FLUSH) + num_store_dw = 4; - cs = intel_ring_begin(rq, 2); + cs = intel_ring_begin(rq, 2 + 3 * num_store_dw); if (IS_ERR(cs)) return PTR_ERR(cs); *cs++ = cmd; - *cs++ = MI_NOOP; + while (num_store_dw--) { + *cs++ = MI_STORE_DWORD_IMM | MI_MEM_VIRTUAL; + *cs++ = i915_scratch_offset(rq->i915); + *cs++ = 0; + } + *cs++ = MI_FLUSH | MI_NO_WRITE_FLUSH; + intel_ring_advance(rq, cs); return 0; @@ -91,6 +100,7 @@ static int gen4_render_ring_flush(struct i915_request *rq, u32 mode) { u32 cmd, *cs; + int i; /* * read/write caches: @@ -127,12 +137,43 @@ gen4_render_ring_flush(struct i915_request *rq, u32 mode) cmd |= MI_INVALIDATE_ISP; } - cs = intel_ring_begin(rq, 2); + i = 2; + if (mode & EMIT_INVALIDATE) + i += 20; + + cs = intel_ring_begin(rq, i); if (IS_ERR(cs)) return PTR_ERR(cs); *cs++ = cmd; - *cs++ = MI_NOOP; + + /* + * A random delay to let the CS invalidate take effect? Without this + * delay, the GPU relocation path fails as the CS does not see + * the updated contents. Just as important, if we apply the flushes + * to the EMIT_FLUSH branch (i.e. immediately after the relocation + * write and before the invalidate on the next batch), the relocations + * still fail. This implies that is a delay following invalidation + * that is required to reset the caches as opposed to a delay to + * ensure the memory is written. + */ + if (mode & EMIT_INVALIDATE) { + *cs++ = GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE; + *cs++ = i915_scratch_offset(rq->i915) | PIPE_CONTROL_GLOBAL_GTT; + *cs++ = 0; + *cs++ = 0; + + for (i = 0; i < 12; i++) + *cs++ = MI_FLUSH; + + *cs++ = GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE; + *cs++ = i915_scratch_offset(rq->i915) | PIPE_CONTROL_GLOBAL_GTT; + *cs++ = 0; + *cs++ = 0; + } + + *cs++ = cmd; + intel_ring_advance(rq, cs); return 0; @@ -178,8 +219,7 @@ gen4_render_ring_flush(struct i915_request *rq, u32 mode) static int intel_emit_post_sync_nonzero_flush(struct i915_request *rq) { - u32 scratch_addr = - i915_ggtt_offset(rq->engine->scratch) + 2 * CACHELINE_BYTES; + u32 scratch_addr = i915_scratch_offset(rq->i915) + 2 * CACHELINE_BYTES; u32 *cs; cs = intel_ring_begin(rq, 6); @@ -212,8 +252,7 @@ intel_emit_post_sync_nonzero_flush(struct i915_request *rq) static int gen6_render_ring_flush(struct i915_request *rq, u32 mode) { - u32 scratch_addr = - i915_ggtt_offset(rq->engine->scratch) + 2 * CACHELINE_BYTES; + u32 scratch_addr = i915_scratch_offset(rq->i915) + 2 * CACHELINE_BYTES; u32 *cs, flags = 0; int ret; @@ -282,8 +321,7 @@ gen7_render_ring_cs_stall_wa(struct i915_request *rq) static int gen7_render_ring_flush(struct i915_request *rq, u32 mode) { - u32 scratch_addr = - i915_ggtt_offset(rq->engine->scratch) + 2 * CACHELINE_BYTES; + u32 scratch_addr = i915_scratch_offset(rq->i915) + 2 * CACHELINE_BYTES; u32 *cs, flags = 0; /* @@ -937,7 +975,7 @@ i965_emit_bb_start(struct i915_request *rq, } /* Just userspace ABI convention to limit the wa batch bo to a resonable size */ -#define I830_BATCH_LIMIT (256*1024) +#define I830_BATCH_LIMIT SZ_256K #define I830_TLB_ENTRIES (2) #define I830_WA_SIZE max(I830_TLB_ENTRIES*4096, I830_BATCH_LIMIT) static int @@ -945,7 +983,9 @@ i830_emit_bb_start(struct i915_request *rq, u64 offset, u32 len, unsigned int dispatch_flags) { - u32 *cs, cs_offset = i915_ggtt_offset(rq->engine->scratch); + u32 *cs, cs_offset = i915_scratch_offset(rq->i915); + + GEM_BUG_ON(rq->i915->gt.scratch->size < I830_WA_SIZE); cs = intel_ring_begin(rq, 6); if (IS_ERR(cs)) @@ -1403,7 +1443,6 @@ static int intel_init_ring_buffer(struct intel_engine_cs *engine) { struct i915_timeline *timeline; struct intel_ring *ring; - unsigned int size; int err; intel_engine_setup_common(engine); @@ -1428,21 +1467,12 @@ static int intel_init_ring_buffer(struct intel_engine_cs *engine) GEM_BUG_ON(engine->buffer); engine->buffer = ring; - size = PAGE_SIZE; - if (HAS_BROKEN_CS_TLB(engine->i915)) - size = I830_WA_SIZE; - err = intel_engine_create_scratch(engine, size); - if (err) - goto err_unpin; - err = intel_engine_init_common(engine); if (err) - goto err_scratch; + goto err_unpin; return 0; -err_scratch: - intel_engine_cleanup_scratch(engine); err_unpin: intel_ring_unpin(ring); err_ring: @@ -1516,7 +1546,7 @@ static int flush_pd_dir(struct i915_request *rq) /* Stall until the page table load is complete */ *cs++ = MI_STORE_REGISTER_MEM | MI_SRM_LRM_GLOBAL_GTT; *cs++ = i915_mmio_reg_offset(RING_PP_DIR_BASE(engine)); - *cs++ = i915_ggtt_offset(engine->scratch); + *cs++ = i915_scratch_offset(rq->i915); *cs++ = MI_NOOP; intel_ring_advance(rq, cs); @@ -1625,7 +1655,7 @@ static inline int mi_set_context(struct i915_request *rq, u32 flags) /* Insert a delay before the next switch! */ *cs++ = MI_STORE_REGISTER_MEM | MI_SRM_LRM_GLOBAL_GTT; *cs++ = i915_mmio_reg_offset(last_reg); - *cs++ = i915_ggtt_offset(engine->scratch); + *cs++ = i915_scratch_offset(rq->i915); *cs++ = MI_NOOP; } *cs++ = MI_ARB_ON_OFF | MI_ARB_ENABLE; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 2dfa585712c2..767a7192c969 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -15,6 +15,7 @@ #include "i915_selftest.h" #include "i915_timeline.h" #include "intel_gpu_commands.h" +#include "intel_workarounds.h" struct drm_printer; struct i915_sched_attr; @@ -440,7 +441,7 @@ struct intel_engine_cs { struct intel_hw_status_page status_page; struct i915_ctx_workarounds wa_ctx; - struct i915_vma *scratch; + struct i915_wa_list wa_list; u32 irq_keep_mask; /* always keep these interrupts */ u32 irq_enable_mask; /* bitmask to enable ring interrupt */ @@ -898,10 +899,6 @@ void intel_engine_setup_common(struct intel_engine_cs *engine); int intel_engine_init_common(struct intel_engine_cs *engine); void intel_engine_cleanup_common(struct intel_engine_cs *engine); -int intel_engine_create_scratch(struct intel_engine_cs *engine, - unsigned int size); -void intel_engine_cleanup_scratch(struct intel_engine_cs *engine); - int intel_init_render_ring_buffer(struct intel_engine_cs *engine); int intel_init_bsd_ring_buffer(struct intel_engine_cs *engine); int intel_init_blt_ring_buffer(struct intel_engine_cs *engine); diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index 0fdabce647ab..44e4491a4918 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -2749,6 +2749,12 @@ static const struct i915_power_well_desc icl_power_wells[] = { }, }, { + .name = "DC off", + .domains = ICL_DISPLAY_DC_OFF_POWER_DOMAINS, + .ops = &gen9_dc_off_power_well_ops, + .id = DISP_PW_ID_NONE, + }, + { .name = "power well 2", .domains = ICL_PW_2_POWER_DOMAINS, .ops = &hsw_power_well_ops, @@ -2760,12 +2766,6 @@ static const struct i915_power_well_desc icl_power_wells[] = { }, }, { - .name = "DC off", - .domains = ICL_DISPLAY_DC_OFF_POWER_DOMAINS, - .ops = &gen9_dc_off_power_well_ops, - .id = DISP_PW_ID_NONE, - }, - { .name = "power well 3", .domains = ICL_PW_3_POWER_DOMAINS, .ops = &hsw_power_well_ops, @@ -3176,8 +3176,7 @@ static u8 intel_dbuf_max_slices(struct drm_i915_private *dev_priv) void icl_dbuf_slices_update(struct drm_i915_private *dev_priv, u8 req_slices) { - u8 hw_enabled_slices = dev_priv->wm.skl_hw.ddb.enabled_slices; - u32 val; + const u8 hw_enabled_slices = dev_priv->wm.skl_hw.ddb.enabled_slices; bool ret; if (req_slices > intel_dbuf_max_slices(dev_priv)) { @@ -3188,7 +3187,6 @@ void icl_dbuf_slices_update(struct drm_i915_private *dev_priv, if (req_slices == hw_enabled_slices || req_slices == 0) return; - val = I915_READ(DBUF_CTL_S2); if (req_slices > hw_enabled_slices) ret = intel_dbuf_slice_set(dev_priv, DBUF_CTL_S2, true); else diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c index 5fd2f7bf3927..d3090a7537bb 100644 --- a/drivers/gpu/drm/i915/intel_sprite.c +++ b/drivers/gpu/drm/i915/intel_sprite.c @@ -302,13 +302,65 @@ skl_plane_max_stride(struct intel_plane *plane, return min(8192 * cpp, 32768); } +static void +skl_program_scaler(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state) +{ + struct drm_i915_private *dev_priv = to_i915(plane->base.dev); + enum pipe pipe = plane->pipe; + int scaler_id = plane_state->scaler_id; + const struct intel_scaler *scaler = + &crtc_state->scaler_state.scalers[scaler_id]; + int crtc_x = plane_state->base.dst.x1; + int crtc_y = plane_state->base.dst.y1; + uint32_t crtc_w = drm_rect_width(&plane_state->base.dst); + uint32_t crtc_h = drm_rect_height(&plane_state->base.dst); + u16 y_hphase, uv_rgb_hphase; + u16 y_vphase, uv_rgb_vphase; + int hscale, vscale; + + hscale = drm_rect_calc_hscale(&plane_state->base.src, + &plane_state->base.dst, + 0, INT_MAX); + vscale = drm_rect_calc_vscale(&plane_state->base.src, + &plane_state->base.dst, + 0, INT_MAX); + + /* TODO: handle sub-pixel coordinates */ + if (plane_state->base.fb->format->format == DRM_FORMAT_NV12) { + y_hphase = skl_scaler_calc_phase(1, hscale, false); + y_vphase = skl_scaler_calc_phase(1, vscale, false); + + /* MPEG2 chroma siting convention */ + uv_rgb_hphase = skl_scaler_calc_phase(2, hscale, true); + uv_rgb_vphase = skl_scaler_calc_phase(2, vscale, false); + } else { + /* not used */ + y_hphase = 0; + y_vphase = 0; + + uv_rgb_hphase = skl_scaler_calc_phase(1, hscale, false); + uv_rgb_vphase = skl_scaler_calc_phase(1, vscale, false); + } + + I915_WRITE_FW(SKL_PS_CTRL(pipe, scaler_id), + PS_SCALER_EN | PS_PLANE_SEL(plane->id) | scaler->mode); + I915_WRITE_FW(SKL_PS_PWR_GATE(pipe, scaler_id), 0); + I915_WRITE_FW(SKL_PS_VPHASE(pipe, scaler_id), + PS_Y_PHASE(y_vphase) | PS_UV_RGB_PHASE(uv_rgb_vphase)); + I915_WRITE_FW(SKL_PS_HPHASE(pipe, scaler_id), + PS_Y_PHASE(y_hphase) | PS_UV_RGB_PHASE(uv_rgb_hphase)); + I915_WRITE_FW(SKL_PS_WIN_POS(pipe, scaler_id), (crtc_x << 16) | crtc_y); + I915_WRITE_FW(SKL_PS_WIN_SZ(pipe, scaler_id), (crtc_w << 16) | crtc_h); +} + void skl_update_plane(struct intel_plane *plane, const struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state) { struct drm_i915_private *dev_priv = to_i915(plane->base.dev); - const struct drm_framebuffer *fb = plane_state->base.fb; enum plane_id plane_id = plane->id; enum pipe pipe = plane->pipe; u32 plane_ctl = plane_state->ctl; @@ -318,8 +370,6 @@ skl_update_plane(struct intel_plane *plane, u32 aux_stride = skl_plane_stride(plane_state, 1); int crtc_x = plane_state->base.dst.x1; int crtc_y = plane_state->base.dst.y1; - uint32_t crtc_w = drm_rect_width(&plane_state->base.dst); - uint32_t crtc_h = drm_rect_height(&plane_state->base.dst); uint32_t x = plane_state->color_plane[0].x; uint32_t y = plane_state->color_plane[0].y; uint32_t src_w = drm_rect_width(&plane_state->base.src) >> 16; @@ -329,8 +379,6 @@ skl_update_plane(struct intel_plane *plane, /* Sizes are 0 based */ src_w--; src_h--; - crtc_w--; - crtc_h--; spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); @@ -353,41 +401,8 @@ skl_update_plane(struct intel_plane *plane, (plane_state->color_plane[1].y << 16) | plane_state->color_plane[1].x); - /* program plane scaler */ if (plane_state->scaler_id >= 0) { - int scaler_id = plane_state->scaler_id; - const struct intel_scaler *scaler = - &crtc_state->scaler_state.scalers[scaler_id]; - u16 y_hphase, uv_rgb_hphase; - u16 y_vphase, uv_rgb_vphase; - - /* TODO: handle sub-pixel coordinates */ - if (fb->format->format == DRM_FORMAT_NV12) { - y_hphase = skl_scaler_calc_phase(1, false); - y_vphase = skl_scaler_calc_phase(1, false); - - /* MPEG2 chroma siting convention */ - uv_rgb_hphase = skl_scaler_calc_phase(2, true); - uv_rgb_vphase = skl_scaler_calc_phase(2, false); - } else { - /* not used */ - y_hphase = 0; - y_vphase = 0; - - uv_rgb_hphase = skl_scaler_calc_phase(1, false); - uv_rgb_vphase = skl_scaler_calc_phase(1, false); - } - - I915_WRITE_FW(SKL_PS_CTRL(pipe, scaler_id), - PS_SCALER_EN | PS_PLANE_SEL(plane_id) | scaler->mode); - I915_WRITE_FW(SKL_PS_PWR_GATE(pipe, scaler_id), 0); - I915_WRITE_FW(SKL_PS_VPHASE(pipe, scaler_id), - PS_Y_PHASE(y_vphase) | PS_UV_RGB_PHASE(uv_rgb_vphase)); - I915_WRITE_FW(SKL_PS_HPHASE(pipe, scaler_id), - PS_Y_PHASE(y_hphase) | PS_UV_RGB_PHASE(uv_rgb_hphase)); - I915_WRITE_FW(SKL_PS_WIN_POS(pipe, scaler_id), (crtc_x << 16) | crtc_y); - I915_WRITE_FW(SKL_PS_WIN_SZ(pipe, scaler_id), - ((crtc_w + 1) << 16)|(crtc_h + 1)); + skl_program_scaler(plane, crtc_state, plane_state); I915_WRITE_FW(PLANE_POS(pipe, plane_id), 0); } else { diff --git a/drivers/gpu/drm/i915/intel_workarounds.c b/drivers/gpu/drm/i915/intel_workarounds.c index 4bcdeaf8d98f..6e580891db96 100644 --- a/drivers/gpu/drm/i915/intel_workarounds.c +++ b/drivers/gpu/drm/i915/intel_workarounds.c @@ -48,6 +48,20 @@ * - Public functions to init or apply the given workaround type. */ +static void wa_init_start(struct i915_wa_list *wal, const char *name) +{ + wal->name = name; +} + +static void wa_init_finish(struct i915_wa_list *wal) +{ + if (!wal->count) + return; + + DRM_DEBUG_DRIVER("Initialized %u %s workarounds\n", + wal->count, wal->name); +} + static void wa_add(struct drm_i915_private *i915, i915_reg_t reg, const u32 mask, const u32 val) { @@ -580,160 +594,175 @@ int intel_ctx_workarounds_emit(struct i915_request *rq) return 0; } -static void bdw_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void +wal_add(struct i915_wa_list *wal, const struct i915_wa *wa) +{ + const unsigned int grow = 1 << 4; + + GEM_BUG_ON(!is_power_of_2(grow)); + + if (IS_ALIGNED(wal->count, grow)) { /* Either uninitialized or full. */ + struct i915_wa *list; + + list = kmalloc_array(ALIGN(wal->count + 1, grow), sizeof(*wa), + GFP_KERNEL); + if (!list) { + DRM_ERROR("No space for workaround init!\n"); + return; + } + + if (wal->list) + memcpy(list, wal->list, sizeof(*wa) * wal->count); + + wal->list = list; + } + + wal->list[wal->count++] = *wa; +} + +static void +wa_masked_en(struct i915_wa_list *wal, i915_reg_t reg, u32 val) +{ + struct i915_wa wa = { + .reg = reg, + .mask = val, + .val = _MASKED_BIT_ENABLE(val) + }; + + wal_add(wal, &wa); +} + +static void +wa_write_masked_or(struct i915_wa_list *wal, i915_reg_t reg, u32 mask, + u32 val) { + struct i915_wa wa = { + .reg = reg, + .mask = mask, + .val = val + }; + + wal_add(wal, &wa); } -static void chv_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void +wa_write(struct i915_wa_list *wal, i915_reg_t reg, u32 val) { + wa_write_masked_or(wal, reg, ~0, val); } -static void gen9_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void +wa_write_or(struct i915_wa_list *wal, i915_reg_t reg, u32 val) { - /* WaContextSwitchWithConcurrentTLBInvalidate:skl,bxt,kbl,glk,cfl */ - I915_WRITE(GEN9_CSFE_CHICKEN1_RCS, - _MASKED_BIT_ENABLE(GEN9_PREEMPT_GPGPU_SYNC_SWITCH_DISABLE)); + wa_write_masked_or(wal, reg, val, val); +} - /* WaEnableLbsSlaRetryTimerDecrement:skl,bxt,kbl,glk,cfl */ - I915_WRITE(BDW_SCRATCH1, I915_READ(BDW_SCRATCH1) | - GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE); +static void gen9_gt_workarounds_init(struct drm_i915_private *i915) +{ + struct i915_wa_list *wal = &i915->gt_wa_list; /* WaDisableKillLogic:bxt,skl,kbl */ - if (!IS_COFFEELAKE(dev_priv)) - I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | - ECOCHK_DIS_TLB); + if (!IS_COFFEELAKE(i915)) + wa_write_or(wal, + GAM_ECOCHK, + ECOCHK_DIS_TLB); - if (HAS_LLC(dev_priv)) { + if (HAS_LLC(i915)) { /* WaCompressedResourceSamplerPbeMediaNewHashMode:skl,kbl * * Must match Display Engine. See * WaCompressedResourceDisplayNewHashMode. */ - I915_WRITE(MMCD_MISC_CTRL, - I915_READ(MMCD_MISC_CTRL) | - MMCD_PCLA | - MMCD_HOTSPOT_EN); + wa_write_or(wal, + MMCD_MISC_CTRL, + MMCD_PCLA | MMCD_HOTSPOT_EN); } /* WaDisableHDCInvalidation:skl,bxt,kbl,cfl */ - I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | - BDW_DISABLE_HDC_INVALIDATION); - - /* WaProgramL3SqcReg1DefaultForPerf:bxt,glk */ - if (IS_GEN9_LP(dev_priv)) { - u32 val = I915_READ(GEN8_L3SQCREG1); - - val &= ~L3_PRIO_CREDITS_MASK; - val |= L3_GENERAL_PRIO_CREDITS(62) | L3_HIGH_PRIO_CREDITS(2); - I915_WRITE(GEN8_L3SQCREG1, val); - } - - /* WaOCLCoherentLineFlush:skl,bxt,kbl,cfl */ - I915_WRITE(GEN8_L3SQCREG4, - I915_READ(GEN8_L3SQCREG4) | GEN8_LQSC_FLUSH_COHERENT_LINES); - - /* WaEnablePreemptionGranularityControlByUMD:skl,bxt,kbl,cfl,[cnl] */ - I915_WRITE(GEN7_FF_SLICE_CS_CHICKEN1, - _MASKED_BIT_ENABLE(GEN9_FFSC_PERCTX_PREEMPT_CTRL)); + wa_write_or(wal, + GAM_ECOCHK, + BDW_DISABLE_HDC_INVALIDATION); } -static void skl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void skl_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaEnableGapsTsvCreditFix:skl */ - I915_WRITE(GEN8_GARBCNTL, - I915_READ(GEN8_GARBCNTL) | GEN9_GAPS_TSV_CREDIT_DISABLE); + gen9_gt_workarounds_init(i915); /* WaDisableGafsUnitClkGating:skl */ - I915_WRITE(GEN7_UCGCTL4, - I915_READ(GEN7_UCGCTL4) | GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); + wa_write_or(wal, + GEN7_UCGCTL4, + GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); /* WaInPlaceDecompressionHang:skl */ - if (IS_SKL_REVID(dev_priv, SKL_REVID_H0, REVID_FOREVER)) - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); + if (IS_SKL_REVID(i915, SKL_REVID_H0, REVID_FOREVER)) + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void bxt_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void bxt_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaDisablePooledEuLoadBalancingFix:bxt */ - I915_WRITE(FF_SLICE_CS_CHICKEN2, - _MASKED_BIT_ENABLE(GEN9_POOLED_EU_LOAD_BALANCING_FIX_DISABLE)); + gen9_gt_workarounds_init(i915); /* WaInPlaceDecompressionHang:bxt */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void kbl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void kbl_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaEnableGapsTsvCreditFix:kbl */ - I915_WRITE(GEN8_GARBCNTL, - I915_READ(GEN8_GARBCNTL) | GEN9_GAPS_TSV_CREDIT_DISABLE); + gen9_gt_workarounds_init(i915); /* WaDisableDynamicCreditSharing:kbl */ - if (IS_KBL_REVID(dev_priv, 0, KBL_REVID_B0)) - I915_WRITE(GAMT_CHKN_BIT_REG, - I915_READ(GAMT_CHKN_BIT_REG) | - GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING); + if (IS_KBL_REVID(i915, 0, KBL_REVID_B0)) + wa_write_or(wal, + GAMT_CHKN_BIT_REG, + GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING); /* WaDisableGafsUnitClkGating:kbl */ - I915_WRITE(GEN7_UCGCTL4, - I915_READ(GEN7_UCGCTL4) | GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); + wa_write_or(wal, + GEN7_UCGCTL4, + GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); /* WaInPlaceDecompressionHang:kbl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); - - /* WaKBLVECSSemaphoreWaitPoll:kbl */ - if (IS_KBL_REVID(dev_priv, KBL_REVID_A0, KBL_REVID_E0)) { - struct intel_engine_cs *engine; - unsigned int tmp; - - for_each_engine(engine, dev_priv, tmp) { - if (engine->id == RCS) - continue; - - I915_WRITE(RING_SEMA_WAIT_POLL(engine->mmio_base), 1); - } - } + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void glk_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void glk_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + gen9_gt_workarounds_init(i915); } -static void cfl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void cfl_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaEnableGapsTsvCreditFix:cfl */ - I915_WRITE(GEN8_GARBCNTL, - I915_READ(GEN8_GARBCNTL) | GEN9_GAPS_TSV_CREDIT_DISABLE); + gen9_gt_workarounds_init(i915); /* WaDisableGafsUnitClkGating:cfl */ - I915_WRITE(GEN7_UCGCTL4, - I915_READ(GEN7_UCGCTL4) | GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); + wa_write_or(wal, + GEN7_UCGCTL4, + GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); /* WaInPlaceDecompressionHang:cfl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } static void wa_init_mcr(struct drm_i915_private *dev_priv) { const struct sseu_dev_info *sseu = &(INTEL_INFO(dev_priv)->sseu); - u32 mcr; + struct i915_wa_list *wal = &dev_priv->gt_wa_list; u32 mcr_slice_subslice_mask; /* @@ -770,8 +799,6 @@ static void wa_init_mcr(struct drm_i915_private *dev_priv) WARN_ON((enabled_mask & disabled_mask) != enabled_mask); } - mcr = I915_READ(GEN8_MCR_SELECTOR); - if (INTEL_GEN(dev_priv) >= 11) mcr_slice_subslice_mask = GEN11_MCR_SLICE_MASK | GEN11_MCR_SUBSLICE_MASK; @@ -789,148 +816,170 @@ static void wa_init_mcr(struct drm_i915_private *dev_priv) * occasions, such as INSTDONE, where this value is dependent * on s/ss combo, the read should be done with read_subslice_reg. */ - mcr &= ~mcr_slice_subslice_mask; - mcr |= intel_calculate_mcr_s_ss_select(dev_priv); - I915_WRITE(GEN8_MCR_SELECTOR, mcr); + wa_write_masked_or(wal, + GEN8_MCR_SELECTOR, + mcr_slice_subslice_mask, + intel_calculate_mcr_s_ss_select(dev_priv)); } -static void cnl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void cnl_gt_workarounds_init(struct drm_i915_private *i915) { - wa_init_mcr(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; + + wa_init_mcr(i915); /* WaDisableI2mCycleOnWRPort:cnl (pre-prod) */ - if (IS_CNL_REVID(dev_priv, CNL_REVID_B0, CNL_REVID_B0)) - I915_WRITE(GAMT_CHKN_BIT_REG, - I915_READ(GAMT_CHKN_BIT_REG) | - GAMT_CHKN_DISABLE_I2M_CYCLE_ON_WR_PORT); + if (IS_CNL_REVID(i915, CNL_REVID_B0, CNL_REVID_B0)) + wa_write_or(wal, + GAMT_CHKN_BIT_REG, + GAMT_CHKN_DISABLE_I2M_CYCLE_ON_WR_PORT); /* WaInPlaceDecompressionHang:cnl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); - - /* WaEnablePreemptionGranularityControlByUMD:cnl */ - I915_WRITE(GEN7_FF_SLICE_CS_CHICKEN1, - _MASKED_BIT_ENABLE(GEN9_FFSC_PERCTX_PREEMPT_CTRL)); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void icl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void icl_gt_workarounds_init(struct drm_i915_private *i915) { - wa_init_mcr(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* This is not an Wa. Enable for better image quality */ - I915_WRITE(_3D_CHICKEN3, - _MASKED_BIT_ENABLE(_3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE)); + wa_init_mcr(i915); /* WaInPlaceDecompressionHang:icl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); - - /* WaPipelineFlushCoherentLines:icl */ - I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) | - GEN8_LQSC_FLUSH_COHERENT_LINES); - - /* Wa_1405543622:icl - * Formerly known as WaGAPZPriorityScheme - */ - I915_WRITE(GEN8_GARBCNTL, I915_READ(GEN8_GARBCNTL) | - GEN11_ARBITRATION_PRIO_ORDER_MASK); - - /* Wa_1604223664:icl - * Formerly known as WaL3BankAddressHashing - */ - I915_WRITE(GEN8_GARBCNTL, - (I915_READ(GEN8_GARBCNTL) & ~GEN11_HASH_CTRL_EXCL_MASK) | - GEN11_HASH_CTRL_EXCL_BIT0); - I915_WRITE(GEN11_GLBLINVL, - (I915_READ(GEN11_GLBLINVL) & ~GEN11_BANK_HASH_ADDR_EXCL_MASK) | - GEN11_BANK_HASH_ADDR_EXCL_BIT0); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); /* WaModifyGamTlbPartitioning:icl */ - I915_WRITE(GEN11_GACB_PERF_CTRL, - (I915_READ(GEN11_GACB_PERF_CTRL) & ~GEN11_HASH_CTRL_MASK) | - GEN11_HASH_CTRL_BIT0 | GEN11_HASH_CTRL_BIT4); - - /* Wa_1405733216:icl - * Formerly known as WaDisableCleanEvicts - */ - I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) | - GEN11_LQSC_CLEAN_EVICT_DISABLE); + wa_write_masked_or(wal, + GEN11_GACB_PERF_CTRL, + GEN11_HASH_CTRL_MASK, + GEN11_HASH_CTRL_BIT0 | GEN11_HASH_CTRL_BIT4); /* Wa_1405766107:icl * Formerly known as WaCL2SFHalfMaxAlloc */ - I915_WRITE(GEN11_LSN_UNSLCVC, I915_READ(GEN11_LSN_UNSLCVC) | - GEN11_LSN_UNSLCVC_GAFS_HALF_SF_MAXALLOC | - GEN11_LSN_UNSLCVC_GAFS_HALF_CL2_MAXALLOC); + wa_write_or(wal, + GEN11_LSN_UNSLCVC, + GEN11_LSN_UNSLCVC_GAFS_HALF_SF_MAXALLOC | + GEN11_LSN_UNSLCVC_GAFS_HALF_CL2_MAXALLOC); /* Wa_220166154:icl * Formerly known as WaDisCtxReload */ - I915_WRITE(GAMW_ECO_DEV_RW_IA_REG, I915_READ(GAMW_ECO_DEV_RW_IA_REG) | - GAMW_ECO_DEV_CTX_RELOAD_DISABLE); + wa_write_or(wal, + GEN8_GAMW_ECO_DEV_RW_IA, + GAMW_ECO_DEV_CTX_RELOAD_DISABLE); /* Wa_1405779004:icl (pre-prod) */ - if (IS_ICL_REVID(dev_priv, ICL_REVID_A0, ICL_REVID_A0)) - I915_WRITE(SLICE_UNIT_LEVEL_CLKGATE, - I915_READ(SLICE_UNIT_LEVEL_CLKGATE) | - MSCUNIT_CLKGATE_DIS); + if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_A0)) + wa_write_or(wal, + SLICE_UNIT_LEVEL_CLKGATE, + MSCUNIT_CLKGATE_DIS); /* Wa_1406680159:icl */ - I915_WRITE(SUBSLICE_UNIT_LEVEL_CLKGATE, - I915_READ(SUBSLICE_UNIT_LEVEL_CLKGATE) | - GWUNIT_CLKGATE_DIS); - - /* Wa_1604302699:icl */ - I915_WRITE(GEN10_L3_CHICKEN_MODE_REGISTER, - I915_READ(GEN10_L3_CHICKEN_MODE_REGISTER) | - GEN11_I2M_WRITE_DISABLE); + wa_write_or(wal, + SUBSLICE_UNIT_LEVEL_CLKGATE, + GWUNIT_CLKGATE_DIS); /* Wa_1406838659:icl (pre-prod) */ - if (IS_ICL_REVID(dev_priv, ICL_REVID_A0, ICL_REVID_B0)) - I915_WRITE(INF_UNIT_LEVEL_CLKGATE, - I915_READ(INF_UNIT_LEVEL_CLKGATE) | - CGPSF_CLKGATE_DIS); - - /* WaForwardProgressSoftReset:icl */ - I915_WRITE(GEN10_SCRATCH_LNCF2, - I915_READ(GEN10_SCRATCH_LNCF2) | - PMFLUSHDONE_LNICRSDROP | - PMFLUSH_GAPL3UNBLOCK | - PMFLUSHDONE_LNEBLK); + if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_B0)) + wa_write_or(wal, + INF_UNIT_LEVEL_CLKGATE, + CGPSF_CLKGATE_DIS); /* Wa_1406463099:icl * Formerly known as WaGamTlbPendError */ - I915_WRITE(GAMT_CHKN_BIT_REG, - I915_READ(GAMT_CHKN_BIT_REG) | - GAMT_CHKN_DISABLE_L3_COH_PIPE); + wa_write_or(wal, + GAMT_CHKN_BIT_REG, + GAMT_CHKN_DISABLE_L3_COH_PIPE); } -void intel_gt_workarounds_apply(struct drm_i915_private *dev_priv) +void intel_gt_init_workarounds(struct drm_i915_private *i915) { - if (INTEL_GEN(dev_priv) < 8) + struct i915_wa_list *wal = &i915->gt_wa_list; + + wa_init_start(wal, "GT"); + + if (INTEL_GEN(i915) < 8) return; - else if (IS_BROADWELL(dev_priv)) - bdw_gt_workarounds_apply(dev_priv); - else if (IS_CHERRYVIEW(dev_priv)) - chv_gt_workarounds_apply(dev_priv); - else if (IS_SKYLAKE(dev_priv)) - skl_gt_workarounds_apply(dev_priv); - else if (IS_BROXTON(dev_priv)) - bxt_gt_workarounds_apply(dev_priv); - else if (IS_KABYLAKE(dev_priv)) - kbl_gt_workarounds_apply(dev_priv); - else if (IS_GEMINILAKE(dev_priv)) - glk_gt_workarounds_apply(dev_priv); - else if (IS_COFFEELAKE(dev_priv)) - cfl_gt_workarounds_apply(dev_priv); - else if (IS_CANNONLAKE(dev_priv)) - cnl_gt_workarounds_apply(dev_priv); - else if (IS_ICELAKE(dev_priv)) - icl_gt_workarounds_apply(dev_priv); + else if (IS_BROADWELL(i915)) + return; + else if (IS_CHERRYVIEW(i915)) + return; + else if (IS_SKYLAKE(i915)) + skl_gt_workarounds_init(i915); + else if (IS_BROXTON(i915)) + bxt_gt_workarounds_init(i915); + else if (IS_KABYLAKE(i915)) + kbl_gt_workarounds_init(i915); + else if (IS_GEMINILAKE(i915)) + glk_gt_workarounds_init(i915); + else if (IS_COFFEELAKE(i915)) + cfl_gt_workarounds_init(i915); + else if (IS_CANNONLAKE(i915)) + cnl_gt_workarounds_init(i915); + else if (IS_ICELAKE(i915)) + icl_gt_workarounds_init(i915); else - MISSING_CASE(INTEL_GEN(dev_priv)); + MISSING_CASE(INTEL_GEN(i915)); + + wa_init_finish(wal); +} + +static enum forcewake_domains +wal_get_fw_for_rmw(struct drm_i915_private *dev_priv, + const struct i915_wa_list *wal) +{ + enum forcewake_domains fw = 0; + struct i915_wa *wa; + unsigned int i; + + for (i = 0, wa = wal->list; i < wal->count; i++, wa++) + fw |= intel_uncore_forcewake_for_reg(dev_priv, + wa->reg, + FW_REG_READ | + FW_REG_WRITE); + + return fw; +} + +static void +wa_list_apply(struct drm_i915_private *dev_priv, const struct i915_wa_list *wal) +{ + enum forcewake_domains fw; + unsigned long flags; + struct i915_wa *wa; + unsigned int i; + + if (!wal->count) + return; + + fw = wal_get_fw_for_rmw(dev_priv, wal); + + spin_lock_irqsave(&dev_priv->uncore.lock, flags); + intel_uncore_forcewake_get__locked(dev_priv, fw); + + for (i = 0, wa = wal->list; i < wal->count; i++, wa++) { + u32 val = I915_READ_FW(wa->reg); + + val &= ~wa->mask; + val |= wa->val; + + I915_WRITE_FW(wa->reg, val); + } + + intel_uncore_forcewake_put__locked(dev_priv, fw); + spin_unlock_irqrestore(&dev_priv->uncore.lock, flags); + + DRM_DEBUG_DRIVER("Applied %u %s workarounds\n", wal->count, wal->name); +} + +void intel_gt_apply_workarounds(struct drm_i915_private *dev_priv) +{ + wa_list_apply(dev_priv, &dev_priv->gt_wa_list); } struct whitelist { @@ -1077,6 +1126,146 @@ void intel_whitelist_workarounds_apply(struct intel_engine_cs *engine) whitelist_apply(engine, whitelist_build(engine, &w)); } +static void rcs_engine_wa_init(struct intel_engine_cs *engine) +{ + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->wa_list; + + if (IS_ICELAKE(i915)) { + /* This is not an Wa. Enable for better image quality */ + wa_masked_en(wal, + _3D_CHICKEN3, + _3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE); + + /* WaPipelineFlushCoherentLines:icl */ + wa_write_or(wal, + GEN8_L3SQCREG4, + GEN8_LQSC_FLUSH_COHERENT_LINES); + + /* + * Wa_1405543622:icl + * Formerly known as WaGAPZPriorityScheme + */ + wa_write_or(wal, + GEN8_GARBCNTL, + GEN11_ARBITRATION_PRIO_ORDER_MASK); + + /* + * Wa_1604223664:icl + * Formerly known as WaL3BankAddressHashing + */ + wa_write_masked_or(wal, + GEN8_GARBCNTL, + GEN11_HASH_CTRL_EXCL_MASK, + GEN11_HASH_CTRL_EXCL_BIT0); + wa_write_masked_or(wal, + GEN11_GLBLINVL, + GEN11_BANK_HASH_ADDR_EXCL_MASK, + GEN11_BANK_HASH_ADDR_EXCL_BIT0); + + /* + * Wa_1405733216:icl + * Formerly known as WaDisableCleanEvicts + */ + wa_write_or(wal, + GEN8_L3SQCREG4, + GEN11_LQSC_CLEAN_EVICT_DISABLE); + + /* Wa_1604302699:icl */ + wa_write_or(wal, + GEN10_L3_CHICKEN_MODE_REGISTER, + GEN11_I2M_WRITE_DISABLE); + + /* WaForwardProgressSoftReset:icl */ + wa_write_or(wal, + GEN10_SCRATCH_LNCF2, + PMFLUSHDONE_LNICRSDROP | + PMFLUSH_GAPL3UNBLOCK | + PMFLUSHDONE_LNEBLK); + } + + if (IS_GEN9(i915) || IS_CANNONLAKE(i915)) { + /* WaEnablePreemptionGranularityControlByUMD:skl,bxt,kbl,cfl,cnl */ + wa_masked_en(wal, + GEN7_FF_SLICE_CS_CHICKEN1, + GEN9_FFSC_PERCTX_PREEMPT_CTRL); + } + + if (IS_SKYLAKE(i915) || IS_KABYLAKE(i915) || IS_COFFEELAKE(i915)) { + /* WaEnableGapsTsvCreditFix:skl,kbl,cfl */ + wa_write_or(wal, + GEN8_GARBCNTL, + GEN9_GAPS_TSV_CREDIT_DISABLE); + } + + if (IS_BROXTON(i915)) { + /* WaDisablePooledEuLoadBalancingFix:bxt */ + wa_masked_en(wal, + FF_SLICE_CS_CHICKEN2, + GEN9_POOLED_EU_LOAD_BALANCING_FIX_DISABLE); + } + + if (IS_GEN9(i915)) { + /* WaContextSwitchWithConcurrentTLBInvalidate:skl,bxt,kbl,glk,cfl */ + wa_masked_en(wal, + GEN9_CSFE_CHICKEN1_RCS, + GEN9_PREEMPT_GPGPU_SYNC_SWITCH_DISABLE); + + /* WaEnableLbsSlaRetryTimerDecrement:skl,bxt,kbl,glk,cfl */ + wa_write_or(wal, + BDW_SCRATCH1, + GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE); + + /* WaProgramL3SqcReg1DefaultForPerf:bxt,glk */ + if (IS_GEN9_LP(i915)) + wa_write_masked_or(wal, + GEN8_L3SQCREG1, + L3_PRIO_CREDITS_MASK, + L3_GENERAL_PRIO_CREDITS(62) | + L3_HIGH_PRIO_CREDITS(2)); + + /* WaOCLCoherentLineFlush:skl,bxt,kbl,cfl */ + wa_write_or(wal, + GEN8_L3SQCREG4, + GEN8_LQSC_FLUSH_COHERENT_LINES); + } +} + +static void xcs_engine_wa_init(struct intel_engine_cs *engine) +{ + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->wa_list; + + /* WaKBLVECSSemaphoreWaitPoll:kbl */ + if (IS_KBL_REVID(i915, KBL_REVID_A0, KBL_REVID_E0)) { + wa_write(wal, + RING_SEMA_WAIT_POLL(engine->mmio_base), + 1); + } +} + +void intel_engine_init_workarounds(struct intel_engine_cs *engine) +{ + struct i915_wa_list *wal = &engine->wa_list; + + if (GEM_WARN_ON(INTEL_GEN(engine->i915) < 8)) + return; + + wa_init_start(wal, engine->name); + + if (engine->id == RCS) + rcs_engine_wa_init(engine); + else + xcs_engine_wa_init(engine); + + wa_init_finish(wal); +} + +void intel_engine_apply_workarounds(struct intel_engine_cs *engine) +{ + wa_list_apply(engine->i915, &engine->wa_list); +} + #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) #include "selftests/intel_workarounds.c" #endif diff --git a/drivers/gpu/drm/i915/intel_workarounds.h b/drivers/gpu/drm/i915/intel_workarounds.h index b11d0623e626..979695a53964 100644 --- a/drivers/gpu/drm/i915/intel_workarounds.h +++ b/drivers/gpu/drm/i915/intel_workarounds.h @@ -7,11 +7,35 @@ #ifndef _I915_WORKAROUNDS_H_ #define _I915_WORKAROUNDS_H_ +#include <linux/slab.h> + +struct i915_wa { + i915_reg_t reg; + u32 mask; + u32 val; +}; + +struct i915_wa_list { + const char *name; + struct i915_wa *list; + unsigned int count; +}; + +static inline void intel_wa_list_free(struct i915_wa_list *wal) +{ + kfree(wal->list); + memset(wal, 0, sizeof(*wal)); +} + int intel_ctx_workarounds_init(struct drm_i915_private *dev_priv); int intel_ctx_workarounds_emit(struct i915_request *rq); -void intel_gt_workarounds_apply(struct drm_i915_private *dev_priv); +void intel_gt_init_workarounds(struct drm_i915_private *dev_priv); +void intel_gt_apply_workarounds(struct drm_i915_private *dev_priv); void intel_whitelist_workarounds_apply(struct intel_engine_cs *engine); +void intel_engine_init_workarounds(struct intel_engine_cs *engine); +void intel_engine_apply_workarounds(struct intel_engine_cs *engine); + #endif diff --git a/drivers/gpu/drm/i915/selftests/huge_pages.c b/drivers/gpu/drm/i915/selftests/huge_pages.c index 8d03f64eabd7..5c22f2c8d4cf 100644 --- a/drivers/gpu/drm/i915/selftests/huge_pages.c +++ b/drivers/gpu/drm/i915/selftests/huge_pages.c @@ -551,7 +551,7 @@ static int igt_mock_ppgtt_misaligned_dma(void *arg) err = igt_check_page_sizes(vma); if (vma->page_sizes.gtt != I915_GTT_PAGE_SIZE_4K) { - pr_err("page_sizes.gtt=%u, expected %lu\n", + pr_err("page_sizes.gtt=%u, expected %llu\n", vma->page_sizes.gtt, I915_GTT_PAGE_SIZE_4K); err = -EINVAL; } diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c index 8e2e269db97e..127d81513671 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c @@ -1337,7 +1337,7 @@ static int igt_gtt_reserve(void *arg) GEM_BUG_ON(!drm_mm_node_allocated(&vma->node)); if (vma->node.start != total || vma->node.size != 2*I915_GTT_PAGE_SIZE) { - pr_err("i915_gem_gtt_reserve (pass 1) placement failed, found (%llx + %llx), expected (%llx + %lx)\n", + pr_err("i915_gem_gtt_reserve (pass 1) placement failed, found (%llx + %llx), expected (%llx + %llx)\n", vma->node.start, vma->node.size, total, 2*I915_GTT_PAGE_SIZE); err = -EINVAL; @@ -1386,7 +1386,7 @@ static int igt_gtt_reserve(void *arg) GEM_BUG_ON(!drm_mm_node_allocated(&vma->node)); if (vma->node.start != total || vma->node.size != 2*I915_GTT_PAGE_SIZE) { - pr_err("i915_gem_gtt_reserve (pass 2) placement failed, found (%llx + %llx), expected (%llx + %lx)\n", + pr_err("i915_gem_gtt_reserve (pass 2) placement failed, found (%llx + %llx), expected (%llx + %llx)\n", vma->node.start, vma->node.size, total, 2*I915_GTT_PAGE_SIZE); err = -EINVAL; @@ -1430,7 +1430,7 @@ static int igt_gtt_reserve(void *arg) GEM_BUG_ON(!drm_mm_node_allocated(&vma->node)); if (vma->node.start != offset || vma->node.size != 2*I915_GTT_PAGE_SIZE) { - pr_err("i915_gem_gtt_reserve (pass 3) placement failed, found (%llx + %llx), expected (%llx + %lx)\n", + pr_err("i915_gem_gtt_reserve (pass 3) placement failed, found (%llx + %llx), expected (%llx + %llx)\n", vma->node.start, vma->node.size, offset, 2*I915_GTT_PAGE_SIZE); err = -EINVAL; diff --git a/drivers/gpu/drm/mediatek/mtk_dsi.c b/drivers/gpu/drm/mediatek/mtk_dsi.c index 66df1b177959..27b507eb4a99 100644 --- a/drivers/gpu/drm/mediatek/mtk_dsi.c +++ b/drivers/gpu/drm/mediatek/mtk_dsi.c @@ -818,10 +818,13 @@ static int mtk_dsi_create_conn_enc(struct drm_device *drm, struct mtk_dsi *dsi) dsi->encoder.possible_crtcs = 1; /* If there's a bridge, attach to it and let it create the connector */ - ret = drm_bridge_attach(&dsi->encoder, dsi->bridge, NULL); - if (ret) { - DRM_ERROR("Failed to attach bridge to drm\n"); - + if (dsi->bridge) { + ret = drm_bridge_attach(&dsi->encoder, dsi->bridge, NULL); + if (ret) { + DRM_ERROR("Failed to attach bridge to drm\n"); + goto err_encoder_cleanup; + } + } else { /* Otherwise create our own connector and attach to a panel */ ret = mtk_dsi_create_connector(drm, dsi); if (ret) diff --git a/drivers/gpu/drm/meson/meson_crtc.c b/drivers/gpu/drm/meson/meson_crtc.c index 05520202c967..191b314f9e9e 100644 --- a/drivers/gpu/drm/meson/meson_crtc.c +++ b/drivers/gpu/drm/meson/meson_crtc.c @@ -45,6 +45,7 @@ struct meson_crtc { struct drm_crtc base; struct drm_pending_vblank_event *event; struct meson_drm *priv; + bool enabled; }; #define to_meson_crtc(x) container_of(x, struct meson_crtc, base) @@ -80,8 +81,7 @@ static const struct drm_crtc_funcs meson_crtc_funcs = { }; -static void meson_crtc_atomic_enable(struct drm_crtc *crtc, - struct drm_crtc_state *old_state) +static void meson_crtc_enable(struct drm_crtc *crtc) { struct meson_crtc *meson_crtc = to_meson_crtc(crtc); struct drm_crtc_state *crtc_state = crtc->state; @@ -101,6 +101,22 @@ static void meson_crtc_atomic_enable(struct drm_crtc *crtc, writel_bits_relaxed(VPP_POSTBLEND_ENABLE, VPP_POSTBLEND_ENABLE, priv->io_base + _REG(VPP_MISC)); + drm_crtc_vblank_on(crtc); + + meson_crtc->enabled = true; +} + +static void meson_crtc_atomic_enable(struct drm_crtc *crtc, + struct drm_crtc_state *old_state) +{ + struct meson_crtc *meson_crtc = to_meson_crtc(crtc); + struct meson_drm *priv = meson_crtc->priv; + + DRM_DEBUG_DRIVER("\n"); + + if (!meson_crtc->enabled) + meson_crtc_enable(crtc); + priv->viu.osd1_enabled = true; } @@ -110,6 +126,8 @@ static void meson_crtc_atomic_disable(struct drm_crtc *crtc, struct meson_crtc *meson_crtc = to_meson_crtc(crtc); struct meson_drm *priv = meson_crtc->priv; + drm_crtc_vblank_off(crtc); + priv->viu.osd1_enabled = false; priv->viu.osd1_commit = false; @@ -124,6 +142,8 @@ static void meson_crtc_atomic_disable(struct drm_crtc *crtc, crtc->state->event = NULL; } + + meson_crtc->enabled = false; } static void meson_crtc_atomic_begin(struct drm_crtc *crtc, @@ -132,6 +152,9 @@ static void meson_crtc_atomic_begin(struct drm_crtc *crtc, struct meson_crtc *meson_crtc = to_meson_crtc(crtc); unsigned long flags; + if (crtc->state->enable && !meson_crtc->enabled) + meson_crtc_enable(crtc); + if (crtc->state->event) { WARN_ON(drm_crtc_vblank_get(crtc) != 0); diff --git a/drivers/gpu/drm/meson/meson_dw_hdmi.c b/drivers/gpu/drm/meson/meson_dw_hdmi.c index df7247cd93f9..2cb2ad26d716 100644 --- a/drivers/gpu/drm/meson/meson_dw_hdmi.c +++ b/drivers/gpu/drm/meson/meson_dw_hdmi.c @@ -706,6 +706,7 @@ static const struct regmap_config meson_dw_hdmi_regmap_config = { .reg_read = meson_dw_hdmi_reg_read, .reg_write = meson_dw_hdmi_reg_write, .max_register = 0x10000, + .fast_io = true, }; static bool meson_hdmi_connector_is_available(struct device *dev) diff --git a/drivers/gpu/drm/meson/meson_venc.c b/drivers/gpu/drm/meson/meson_venc.c index 514245e69b38..be76f3d64bf2 100644 --- a/drivers/gpu/drm/meson/meson_venc.c +++ b/drivers/gpu/drm/meson/meson_venc.c @@ -71,6 +71,7 @@ */ /* HHI Registers */ +#define HHI_GCLK_MPEG2 0x148 /* 0x52 offset in data sheet */ #define HHI_VDAC_CNTL0 0x2F4 /* 0xbd offset in data sheet */ #define HHI_VDAC_CNTL1 0x2F8 /* 0xbe offset in data sheet */ #define HHI_HDMI_PHY_CNTL0 0x3a0 /* 0xe8 offset in data sheet */ @@ -714,6 +715,7 @@ struct meson_hdmi_venc_vic_mode { { 5, &meson_hdmi_encp_mode_1080i60 }, { 20, &meson_hdmi_encp_mode_1080i50 }, { 32, &meson_hdmi_encp_mode_1080p24 }, + { 33, &meson_hdmi_encp_mode_1080p50 }, { 34, &meson_hdmi_encp_mode_1080p30 }, { 31, &meson_hdmi_encp_mode_1080p50 }, { 16, &meson_hdmi_encp_mode_1080p60 }, @@ -854,6 +856,13 @@ void meson_venc_hdmi_mode_set(struct meson_drm *priv, int vic, unsigned int sof_lines; unsigned int vsync_lines; + /* Use VENCI for 480i and 576i and double HDMI pixels */ + if (mode->flags & DRM_MODE_FLAG_DBLCLK) { + hdmi_repeat = true; + use_enci = true; + venc_hdmi_latency = 1; + } + if (meson_venc_hdmi_supported_vic(vic)) { vmode = meson_venc_hdmi_get_vic_vmode(vic); if (!vmode) { @@ -865,13 +874,7 @@ void meson_venc_hdmi_mode_set(struct meson_drm *priv, int vic, } else { meson_venc_hdmi_get_dmt_vmode(mode, &vmode_dmt); vmode = &vmode_dmt; - } - - /* Use VENCI for 480i and 576i and double HDMI pixels */ - if (mode->flags & DRM_MODE_FLAG_DBLCLK) { - hdmi_repeat = true; - use_enci = true; - venc_hdmi_latency = 1; + use_enci = false; } /* Repeat VENC pixels for 480/576i/p, 720p50/60 and 1080p50/60 */ @@ -1529,10 +1532,12 @@ unsigned int meson_venci_get_field(struct meson_drm *priv) void meson_venc_enable_vsync(struct meson_drm *priv) { writel_relaxed(2, priv->io_base + _REG(VENC_INTCTRL)); + regmap_update_bits(priv->hhi, HHI_GCLK_MPEG2, BIT(25), BIT(25)); } void meson_venc_disable_vsync(struct meson_drm *priv) { + regmap_update_bits(priv->hhi, HHI_GCLK_MPEG2, BIT(25), 0); writel_relaxed(0, priv->io_base + _REG(VENC_INTCTRL)); } diff --git a/drivers/gpu/drm/meson/meson_viu.c b/drivers/gpu/drm/meson/meson_viu.c index 6bcfa527c180..26a0857878bf 100644 --- a/drivers/gpu/drm/meson/meson_viu.c +++ b/drivers/gpu/drm/meson/meson_viu.c @@ -184,18 +184,18 @@ void meson_viu_set_osd_lut(struct meson_drm *priv, enum viu_lut_sel_e lut_sel, if (lut_sel == VIU_LUT_OSD_OETF) { writel(0, priv->io_base + _REG(addr_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_OETF_LUT_SIZE / 2); i++) writel(r_map[i * 2] | (r_map[i * 2 + 1] << 16), priv->io_base + _REG(data_port)); writel(r_map[OSD_OETF_LUT_SIZE - 1] | (g_map[0] << 16), priv->io_base + _REG(data_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_OETF_LUT_SIZE / 2); i++) writel(g_map[i * 2 + 1] | (g_map[i * 2 + 2] << 16), priv->io_base + _REG(data_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_OETF_LUT_SIZE / 2); i++) writel(b_map[i * 2] | (b_map[i * 2 + 1] << 16), priv->io_base + _REG(data_port)); @@ -211,18 +211,18 @@ void meson_viu_set_osd_lut(struct meson_drm *priv, enum viu_lut_sel_e lut_sel, } else if (lut_sel == VIU_LUT_OSD_EOTF) { writel(0, priv->io_base + _REG(addr_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_EOTF_LUT_SIZE / 2); i++) writel(r_map[i * 2] | (r_map[i * 2 + 1] << 16), priv->io_base + _REG(data_port)); writel(r_map[OSD_EOTF_LUT_SIZE - 1] | (g_map[0] << 16), priv->io_base + _REG(data_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_EOTF_LUT_SIZE / 2); i++) writel(g_map[i * 2 + 1] | (g_map[i * 2 + 2] << 16), priv->io_base + _REG(data_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_EOTF_LUT_SIZE / 2); i++) writel(b_map[i * 2] | (b_map[i * 2 + 1] << 16), priv->io_base + _REG(data_port)); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c index d4530d60767b..ca169f013a14 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c @@ -1594,7 +1594,6 @@ struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane, NULL); drm_crtc_helper_add(crtc, &dpu_crtc_helper_funcs); - plane->crtc = crtc; /* save user friendly CRTC name for later */ snprintf(dpu_crtc->name, DPU_CRTC_NAME_SIZE, "crtc%u", crtc->base.id); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c index 96cdf06e7da2..d31d8281424e 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c @@ -488,8 +488,6 @@ static void dpu_encoder_destroy(struct drm_encoder *drm_enc) drm_encoder_cleanup(drm_enc); mutex_destroy(&dpu_enc->enc_lock); - - kfree(dpu_enc); } void dpu_encoder_helper_split_config( diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c index bfcd165e96df..d743e7ca6a3c 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c @@ -216,7 +216,7 @@ static const struct dpu_format dpu_format_map[] = { INTERLEAVED_RGB_FMT(XBGR8888, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, - true, 4, 0, + false, 4, 0, DPU_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBA8888, diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c index 4c03f0b7343e..41bec570c518 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c @@ -39,6 +39,8 @@ #define DSI_PIXEL_PLL_CLK 1 #define NUM_PROVIDED_CLKS 2 +#define VCO_REF_CLK_RATE 19200000 + struct dsi_pll_regs { u32 pll_prop_gain_rate; u32 pll_lockdet_rate; @@ -316,7 +318,7 @@ static int dsi_pll_10nm_vco_set_rate(struct clk_hw *hw, unsigned long rate, parent_rate); pll_10nm->vco_current_rate = rate; - pll_10nm->vco_ref_clk_rate = parent_rate; + pll_10nm->vco_ref_clk_rate = VCO_REF_CLK_RATE; dsi_pll_setup_config(pll_10nm); diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c index c79659ca5706..adbdce3aeda0 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi.c @@ -332,6 +332,12 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi, goto fail; } + ret = msm_hdmi_hpd_enable(hdmi->connector); + if (ret < 0) { + DRM_DEV_ERROR(&hdmi->pdev->dev, "failed to enable HPD: %d\n", ret); + goto fail; + } + encoder->bridge = hdmi->bridge; priv->bridges[priv->num_bridges++] = hdmi->bridge; @@ -571,7 +577,7 @@ static int msm_hdmi_bind(struct device *dev, struct device *master, void *data) { struct drm_device *drm = dev_get_drvdata(master); struct msm_drm_private *priv = drm->dev_private; - static struct hdmi_platform_config *hdmi_cfg; + struct hdmi_platform_config *hdmi_cfg; struct hdmi *hdmi; struct device_node *of_node = dev->of_node; int i, err; diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.h b/drivers/gpu/drm/msm/hdmi/hdmi.h index accc9a61611d..5c5df6ab2a57 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.h +++ b/drivers/gpu/drm/msm/hdmi/hdmi.h @@ -245,6 +245,7 @@ void msm_hdmi_bridge_destroy(struct drm_bridge *bridge); void msm_hdmi_connector_irq(struct drm_connector *connector); struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi); +int msm_hdmi_hpd_enable(struct drm_connector *connector); /* * i2c adapter for ddc: diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c index e9c9a0af508e..30e908dfded7 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c @@ -167,8 +167,9 @@ static void enable_hpd_clocks(struct hdmi *hdmi, bool enable) } } -static int hpd_enable(struct hdmi_connector *hdmi_connector) +int msm_hdmi_hpd_enable(struct drm_connector *connector) { + struct hdmi_connector *hdmi_connector = to_hdmi_connector(connector); struct hdmi *hdmi = hdmi_connector->hdmi; const struct hdmi_platform_config *config = hdmi->config; struct device *dev = &hdmi->pdev->dev; @@ -450,7 +451,6 @@ struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi) { struct drm_connector *connector = NULL; struct hdmi_connector *hdmi_connector; - int ret; hdmi_connector = kzalloc(sizeof(*hdmi_connector), GFP_KERNEL); if (!hdmi_connector) @@ -471,12 +471,6 @@ struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi) connector->interlace_allowed = 0; connector->doublescan_allowed = 0; - ret = hpd_enable(hdmi_connector); - if (ret) { - dev_err(&hdmi->pdev->dev, "failed to enable HPD: %d\n", ret); - return ERR_PTR(ret); - } - drm_connector_attach_encoder(connector, hdmi->encoder); return connector; diff --git a/drivers/gpu/drm/msm/msm_atomic.c b/drivers/gpu/drm/msm/msm_atomic.c index 4bcdeca7479d..2088a20eb270 100644 --- a/drivers/gpu/drm/msm/msm_atomic.c +++ b/drivers/gpu/drm/msm/msm_atomic.c @@ -34,7 +34,12 @@ static void msm_atomic_wait_for_commit_done(struct drm_device *dev, if (!new_crtc_state->active) continue; + if (drm_crtc_vblank_get(crtc)) + continue; + kms->funcs->wait_for_crtc_commit_done(kms, crtc); + + drm_crtc_vblank_put(crtc); } } diff --git a/drivers/gpu/drm/msm/msm_debugfs.c b/drivers/gpu/drm/msm/msm_debugfs.c index f0da0d3c8a80..d756436c1fcd 100644 --- a/drivers/gpu/drm/msm/msm_debugfs.c +++ b/drivers/gpu/drm/msm/msm_debugfs.c @@ -84,7 +84,7 @@ static int msm_gpu_open(struct inode *inode, struct file *file) ret = mutex_lock_interruptible(&dev->struct_mutex); if (ret) - return ret; + goto free_priv; pm_runtime_get_sync(&gpu->pdev->dev); show_priv->state = gpu->funcs->gpu_state_get(gpu); @@ -94,13 +94,20 @@ static int msm_gpu_open(struct inode *inode, struct file *file) if (IS_ERR(show_priv->state)) { ret = PTR_ERR(show_priv->state); - kfree(show_priv); - return ret; + goto free_priv; } show_priv->dev = dev; - return single_open(file, msm_gpu_show, show_priv); + ret = single_open(file, msm_gpu_show, show_priv); + if (ret) + goto free_priv; + + return 0; + +free_priv: + kfree(show_priv); + return ret; } static const struct file_operations msm_gpu_fops = { diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 4904d0d41409..dcff812c63d0 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -553,17 +553,18 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) kthread_run(kthread_worker_fn, &priv->disp_thread[i].worker, "crtc_commit:%d", priv->disp_thread[i].crtc_id); - ret = sched_setscheduler(priv->disp_thread[i].thread, - SCHED_FIFO, ¶m); - if (ret) - pr_warn("display thread priority update failed: %d\n", - ret); - if (IS_ERR(priv->disp_thread[i].thread)) { dev_err(dev, "failed to create crtc_commit kthread\n"); priv->disp_thread[i].thread = NULL; + goto err_msm_uninit; } + ret = sched_setscheduler(priv->disp_thread[i].thread, + SCHED_FIFO, ¶m); + if (ret) + dev_warn(dev, "disp_thread set priority failed: %d\n", + ret); + /* initialize event thread */ priv->event_thread[i].crtc_id = priv->crtcs[i]->base.id; kthread_init_worker(&priv->event_thread[i].worker); @@ -572,6 +573,12 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) kthread_run(kthread_worker_fn, &priv->event_thread[i].worker, "crtc_event:%d", priv->event_thread[i].crtc_id); + if (IS_ERR(priv->event_thread[i].thread)) { + dev_err(dev, "failed to create crtc_event kthread\n"); + priv->event_thread[i].thread = NULL; + goto err_msm_uninit; + } + /** * event thread should also run at same priority as disp_thread * because it is handling frame_done events. A lower priority @@ -580,34 +587,10 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) * failure at crtc commit level. */ ret = sched_setscheduler(priv->event_thread[i].thread, - SCHED_FIFO, ¶m); + SCHED_FIFO, ¶m); if (ret) - pr_warn("display event thread priority update failed: %d\n", - ret); - - if (IS_ERR(priv->event_thread[i].thread)) { - dev_err(dev, "failed to create crtc_event kthread\n"); - priv->event_thread[i].thread = NULL; - } - - if ((!priv->disp_thread[i].thread) || - !priv->event_thread[i].thread) { - /* clean up previously created threads if any */ - for ( ; i >= 0; i--) { - if (priv->disp_thread[i].thread) { - kthread_stop( - priv->disp_thread[i].thread); - priv->disp_thread[i].thread = NULL; - } - - if (priv->event_thread[i].thread) { - kthread_stop( - priv->event_thread[i].thread); - priv->event_thread[i].thread = NULL; - } - } - goto err_msm_uninit; - } + dev_warn(dev, "event_thread set priority failed:%d\n", + ret); } ret = drm_vblank_init(ddev, priv->num_crtcs); diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index 7a7923e6220d..6942604ad9a8 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -317,6 +317,9 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob uint32_t *ptr; int ret = 0; + if (!nr_relocs) + return 0; + if (offset % 4) { DRM_ERROR("non-aligned cmdstream buffer: %u\n", offset); return -EINVAL; @@ -410,7 +413,6 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, struct msm_file_private *ctx = file->driver_priv; struct msm_gem_submit *submit; struct msm_gpu *gpu = priv->gpu; - struct dma_fence *in_fence = NULL; struct sync_file *sync_file = NULL; struct msm_gpu_submitqueue *queue; struct msm_ringbuffer *ring; @@ -443,6 +445,8 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, ring = gpu->rb[queue->prio]; if (args->flags & MSM_SUBMIT_FENCE_FD_IN) { + struct dma_fence *in_fence; + in_fence = sync_file_get_fence(args->fence_fd); if (!in_fence) @@ -452,11 +456,13 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, * Wait if the fence is from a foreign context, or if the fence * array contains any fence from a foreign context. */ - if (!dma_fence_match_context(in_fence, ring->fctx->context)) { + ret = 0; + if (!dma_fence_match_context(in_fence, ring->fctx->context)) ret = dma_fence_wait(in_fence, true); - if (ret) - return ret; - } + + dma_fence_put(in_fence); + if (ret) + return ret; } ret = mutex_lock_interruptible(&dev->struct_mutex); @@ -582,8 +588,6 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, } out: - if (in_fence) - dma_fence_put(in_fence); submit_cleanup(submit); if (ret) msm_gem_submit_free(submit); diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 11aac8337066..2b7c8946adba 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -345,6 +345,10 @@ static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, { struct msm_gpu_state *state; + /* Check if the target supports capturing crash state */ + if (!gpu->funcs->gpu_state_get) + return; + /* Only save one crash state at a time */ if (gpu->crashstate) return; @@ -434,10 +438,9 @@ static void recover_worker(struct work_struct *work) if (submit) { struct task_struct *task; - rcu_read_lock(); - task = pid_task(submit->pid, PIDTYPE_PID); + task = get_pid_task(submit->pid, PIDTYPE_PID); if (task) { - comm = kstrdup(task->comm, GFP_ATOMIC); + comm = kstrdup(task->comm, GFP_KERNEL); /* * So slightly annoying, in other paths like @@ -450,10 +453,10 @@ static void recover_worker(struct work_struct *work) * about the submit going away. */ mutex_unlock(&dev->struct_mutex); - cmd = kstrdup_quotable_cmdline(task, GFP_ATOMIC); + cmd = kstrdup_quotable_cmdline(task, GFP_KERNEL); + put_task_struct(task); mutex_lock(&dev->struct_mutex); } - rcu_read_unlock(); if (comm && cmd) { dev_err(dev->dev, "%s: offending task: %s (%s)\n", diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c index b23d33622f37..2a90aa4caec0 100644 --- a/drivers/gpu/drm/msm/msm_iommu.c +++ b/drivers/gpu/drm/msm/msm_iommu.c @@ -66,7 +66,7 @@ static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova, // pm_runtime_get_sync(mmu->dev); ret = iommu_map_sg(iommu->domain, iova, sgt->sgl, sgt->nents, prot); // pm_runtime_put_sync(mmu->dev); - WARN_ON(ret < 0); + WARN_ON(!ret); return (ret == len) ? 0 : -EINVAL; } diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index cca933458439..0c2c8d2c631f 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -316,10 +316,11 @@ static void snapshot_buf(struct msm_rd_state *rd, uint64_t iova, uint32_t size) { struct msm_gem_object *obj = submit->bos[idx].obj; + unsigned offset = 0; const char *buf; if (iova) { - buf += iova - submit->bos[idx].iova; + offset = iova - submit->bos[idx].iova; } else { iova = submit->bos[idx].iova; size = obj->base.size; @@ -340,6 +341,8 @@ static void snapshot_buf(struct msm_rd_state *rd, if (IS_ERR(buf)) return; + buf += offset; + rd_write_section(rd, RD_BUFFER_CONTENTS, buf, size); msm_gem_put_vaddr(&obj->base); diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c index 6cbbae3f438b..db1bf7f88c1f 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -198,6 +198,22 @@ nv50_dmac_create(struct nvif_device *device, struct nvif_object *disp, /****************************************************************************** * EVO channel helpers *****************************************************************************/ +static void +evo_flush(struct nv50_dmac *dmac) +{ + /* Push buffer fetches are not coherent with BAR1, we need to ensure + * writes have been flushed right through to VRAM before writing PUT. + */ + if (dmac->push.type & NVIF_MEM_VRAM) { + struct nvif_device *device = dmac->base.device; + nvif_wr32(&device->object, 0x070000, 0x00000001); + nvif_msec(device, 2000, + if (!(nvif_rd32(&device->object, 0x070000) & 0x00000002)) + break; + ); + } +} + u32 * evo_wait(struct nv50_dmac *evoc, int nr) { @@ -208,6 +224,7 @@ evo_wait(struct nv50_dmac *evoc, int nr) mutex_lock(&dmac->lock); if (put + nr >= (PAGE_SIZE / 4) - 8) { dmac->ptr[put] = 0x20000000; + evo_flush(dmac); nvif_wr32(&dmac->base.user, 0x0000, 0x00000000); if (nvif_msec(device, 2000, @@ -230,17 +247,7 @@ evo_kick(u32 *push, struct nv50_dmac *evoc) { struct nv50_dmac *dmac = evoc; - /* Push buffer fetches are not coherent with BAR1, we need to ensure - * writes have been flushed right through to VRAM before writing PUT. - */ - if (dmac->push.type & NVIF_MEM_VRAM) { - struct nvif_device *device = dmac->base.device; - nvif_wr32(&device->object, 0x070000, 0x00000001); - nvif_msec(device, 2000, - if (!(nvif_rd32(&device->object, 0x070000) & 0x00000002)) - break; - ); - } + evo_flush(dmac); nvif_wr32(&dmac->base.user, 0x0000, (push - dmac->ptr) << 2); mutex_unlock(&dmac->lock); @@ -1264,6 +1271,7 @@ nv50_mstm_del(struct nv50_mstm **pmstm) { struct nv50_mstm *mstm = *pmstm; if (mstm) { + drm_dp_mst_topology_mgr_destroy(&mstm->mgr); kfree(*pmstm); *pmstm = NULL; } diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 2b2baf6e0e0d..d2928d43f29a 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -1171,10 +1171,16 @@ nouveau_platform_device_create(const struct nvkm_device_tegra_func *func, goto err_free; } + err = nouveau_drm_device_init(drm); + if (err) + goto err_put; + platform_set_drvdata(pdev, drm); return drm; +err_put: + drm_dev_put(drm); err_free: nvkm_device_del(pdevice); diff --git a/drivers/gpu/drm/omapdrm/displays/panel-dpi.c b/drivers/gpu/drm/omapdrm/displays/panel-dpi.c index 1f8161b041be..465120809eb3 100644 --- a/drivers/gpu/drm/omapdrm/displays/panel-dpi.c +++ b/drivers/gpu/drm/omapdrm/displays/panel-dpi.c @@ -177,6 +177,7 @@ static int panel_dpi_probe(struct platform_device *pdev) dssdev->type = OMAP_DISPLAY_TYPE_DPI; dssdev->owner = THIS_MODULE; dssdev->of_ports = BIT(0); + drm_bus_flags_from_videomode(&ddata->vm, &dssdev->bus_flags); omapdss_display_init(dssdev); omapdss_device_register(dssdev); diff --git a/drivers/gpu/drm/omapdrm/dss/dsi.c b/drivers/gpu/drm/omapdrm/dss/dsi.c index 394c129cfb3b..00a9c2ab9e6c 100644 --- a/drivers/gpu/drm/omapdrm/dss/dsi.c +++ b/drivers/gpu/drm/omapdrm/dss/dsi.c @@ -5409,15 +5409,24 @@ static int dsi_probe(struct platform_device *pdev) /* DSI on OMAP3 doesn't have register DSI_GNQ, set number * of data to 3 by default */ - if (dsi->data->quirks & DSI_QUIRK_GNQ) + if (dsi->data->quirks & DSI_QUIRK_GNQ) { + dsi_runtime_get(dsi); /* NB_DATA_LANES */ dsi->num_lanes_supported = 1 + REG_GET(dsi, DSI_GNQ, 11, 9); - else + dsi_runtime_put(dsi); + } else { dsi->num_lanes_supported = 3; + } + + r = of_platform_populate(dev->of_node, NULL, NULL, dev); + if (r) { + DSSERR("Failed to populate DSI child devices: %d\n", r); + goto err_pm_disable; + } r = dsi_init_output(dsi); if (r) - goto err_pm_disable; + goto err_of_depopulate; r = dsi_probe_of(dsi); if (r) { @@ -5425,10 +5434,6 @@ static int dsi_probe(struct platform_device *pdev) goto err_uninit_output; } - r = of_platform_populate(dev->of_node, NULL, NULL, dev); - if (r) - DSSERR("Failed to populate DSI child devices: %d\n", r); - r = component_add(&pdev->dev, &dsi_component_ops); if (r) goto err_uninit_output; @@ -5437,6 +5442,8 @@ static int dsi_probe(struct platform_device *pdev) err_uninit_output: dsi_uninit_output(dsi); +err_of_depopulate: + of_platform_depopulate(dev); err_pm_disable: pm_runtime_disable(dev); return r; @@ -5470,19 +5477,12 @@ static int dsi_runtime_suspend(struct device *dev) /* wait for current handler to finish before turning the DSI off */ synchronize_irq(dsi->irq); - dispc_runtime_put(dsi->dss->dispc); - return 0; } static int dsi_runtime_resume(struct device *dev) { struct dsi_data *dsi = dev_get_drvdata(dev); - int r; - - r = dispc_runtime_get(dsi->dss->dispc); - if (r) - return r; dsi->is_enabled = true; /* ensure the irq handler sees the is_enabled value */ diff --git a/drivers/gpu/drm/omapdrm/dss/dss.c b/drivers/gpu/drm/omapdrm/dss/dss.c index 1aaf260aa9b8..7553c7fc1c45 100644 --- a/drivers/gpu/drm/omapdrm/dss/dss.c +++ b/drivers/gpu/drm/omapdrm/dss/dss.c @@ -1484,16 +1484,23 @@ static int dss_probe(struct platform_device *pdev) dss); /* Add all the child devices as components. */ + r = of_platform_populate(pdev->dev.of_node, NULL, NULL, &pdev->dev); + if (r) + goto err_uninit_debugfs; + omapdss_gather_components(&pdev->dev); device_for_each_child(&pdev->dev, &match, dss_add_child_component); r = component_master_add_with_match(&pdev->dev, &dss_component_ops, match); if (r) - goto err_uninit_debugfs; + goto err_of_depopulate; return 0; +err_of_depopulate: + of_platform_depopulate(&pdev->dev); + err_uninit_debugfs: dss_debugfs_remove_file(dss->debugfs.clk); dss_debugfs_remove_file(dss->debugfs.dss); @@ -1522,6 +1529,8 @@ static int dss_remove(struct platform_device *pdev) { struct dss_device *dss = platform_get_drvdata(pdev); + of_platform_depopulate(&pdev->dev); + component_master_del(&pdev->dev, &dss_component_ops); dss_debugfs_remove_file(dss->debugfs.clk); diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4.c b/drivers/gpu/drm/omapdrm/dss/hdmi4.c index cf6230eac31a..aabdda394c9c 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi4.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi4.c @@ -635,10 +635,14 @@ static int hdmi4_bind(struct device *dev, struct device *master, void *data) hdmi->dss = dss; - r = hdmi_pll_init(dss, hdmi->pdev, &hdmi->pll, &hdmi->wp); + r = hdmi_runtime_get(hdmi); if (r) return r; + r = hdmi_pll_init(dss, hdmi->pdev, &hdmi->pll, &hdmi->wp); + if (r) + goto err_runtime_put; + r = hdmi4_cec_init(hdmi->pdev, &hdmi->core, &hdmi->wp); if (r) goto err_pll_uninit; @@ -652,12 +656,16 @@ static int hdmi4_bind(struct device *dev, struct device *master, void *data) hdmi->debugfs = dss_debugfs_create_file(dss, "hdmi", hdmi_dump_regs, hdmi); + hdmi_runtime_put(hdmi); + return 0; err_cec_uninit: hdmi4_cec_uninit(&hdmi->core); err_pll_uninit: hdmi_pll_uninit(&hdmi->pll); +err_runtime_put: + hdmi_runtime_put(hdmi); return r; } @@ -833,32 +841,6 @@ static int hdmi4_remove(struct platform_device *pdev) return 0; } -static int hdmi_runtime_suspend(struct device *dev) -{ - struct omap_hdmi *hdmi = dev_get_drvdata(dev); - - dispc_runtime_put(hdmi->dss->dispc); - - return 0; -} - -static int hdmi_runtime_resume(struct device *dev) -{ - struct omap_hdmi *hdmi = dev_get_drvdata(dev); - int r; - - r = dispc_runtime_get(hdmi->dss->dispc); - if (r < 0) - return r; - - return 0; -} - -static const struct dev_pm_ops hdmi_pm_ops = { - .runtime_suspend = hdmi_runtime_suspend, - .runtime_resume = hdmi_runtime_resume, -}; - static const struct of_device_id hdmi_of_match[] = { { .compatible = "ti,omap4-hdmi", }, {}, @@ -869,7 +851,6 @@ struct platform_driver omapdss_hdmi4hw_driver = { .remove = hdmi4_remove, .driver = { .name = "omapdss_hdmi", - .pm = &hdmi_pm_ops, .of_match_table = hdmi_of_match, .suppress_bind_attrs = true, }, diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5.c b/drivers/gpu/drm/omapdrm/dss/hdmi5.c index b0e4a7463f8c..9e8556f67a29 100644 --- a/drivers/gpu/drm/omapdrm/dss/hdmi5.c +++ b/drivers/gpu/drm/omapdrm/dss/hdmi5.c @@ -825,32 +825,6 @@ static int hdmi5_remove(struct platform_device *pdev) return 0; } -static int hdmi_runtime_suspend(struct device *dev) -{ - struct omap_hdmi *hdmi = dev_get_drvdata(dev); - - dispc_runtime_put(hdmi->dss->dispc); - - return 0; -} - -static int hdmi_runtime_resume(struct device *dev) -{ - struct omap_hdmi *hdmi = dev_get_drvdata(dev); - int r; - - r = dispc_runtime_get(hdmi->dss->dispc); - if (r < 0) - return r; - - return 0; -} - -static const struct dev_pm_ops hdmi_pm_ops = { - .runtime_suspend = hdmi_runtime_suspend, - .runtime_resume = hdmi_runtime_resume, -}; - static const struct of_device_id hdmi_of_match[] = { { .compatible = "ti,omap5-hdmi", }, { .compatible = "ti,dra7-hdmi", }, @@ -862,7 +836,6 @@ struct platform_driver omapdss_hdmi5hw_driver = { .remove = hdmi5_remove, .driver = { .name = "omapdss_hdmi5", - .pm = &hdmi_pm_ops, .of_match_table = hdmi_of_match, .suppress_bind_attrs = true, }, diff --git a/drivers/gpu/drm/omapdrm/dss/omapdss.h b/drivers/gpu/drm/omapdrm/dss/omapdss.h index 1f698a95a94a..33e15cb77efa 100644 --- a/drivers/gpu/drm/omapdrm/dss/omapdss.h +++ b/drivers/gpu/drm/omapdrm/dss/omapdss.h @@ -432,7 +432,7 @@ struct omap_dss_device { const struct omap_dss_driver *driver; const struct omap_dss_device_ops *ops; unsigned long ops_flags; - unsigned long bus_flags; + u32 bus_flags; /* helper variable for driver suspend/resume */ bool activate_after_resume; diff --git a/drivers/gpu/drm/omapdrm/dss/venc.c b/drivers/gpu/drm/omapdrm/dss/venc.c index ff0b18c8e4ac..b5f52727f8b1 100644 --- a/drivers/gpu/drm/omapdrm/dss/venc.c +++ b/drivers/gpu/drm/omapdrm/dss/venc.c @@ -946,19 +946,12 @@ static int venc_runtime_suspend(struct device *dev) if (venc->tv_dac_clk) clk_disable_unprepare(venc->tv_dac_clk); - dispc_runtime_put(venc->dss->dispc); - return 0; } static int venc_runtime_resume(struct device *dev) { struct venc_device *venc = dev_get_drvdata(dev); - int r; - - r = dispc_runtime_get(venc->dss->dispc); - if (r < 0) - return r; if (venc->tv_dac_clk) clk_prepare_enable(venc->tv_dac_clk); diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c index 62928ec0e7db..caffc547ef97 100644 --- a/drivers/gpu/drm/omapdrm/omap_crtc.c +++ b/drivers/gpu/drm/omapdrm/omap_crtc.c @@ -350,11 +350,14 @@ static void omap_crtc_arm_event(struct drm_crtc *crtc) static void omap_crtc_atomic_enable(struct drm_crtc *crtc, struct drm_crtc_state *old_state) { + struct omap_drm_private *priv = crtc->dev->dev_private; struct omap_crtc *omap_crtc = to_omap_crtc(crtc); int ret; DBG("%s", omap_crtc->name); + priv->dispc_ops->runtime_get(priv->dispc); + spin_lock_irq(&crtc->dev->event_lock); drm_crtc_vblank_on(crtc); ret = drm_crtc_vblank_get(crtc); @@ -367,6 +370,7 @@ static void omap_crtc_atomic_enable(struct drm_crtc *crtc, static void omap_crtc_atomic_disable(struct drm_crtc *crtc, struct drm_crtc_state *old_state) { + struct omap_drm_private *priv = crtc->dev->dev_private; struct omap_crtc *omap_crtc = to_omap_crtc(crtc); DBG("%s", omap_crtc->name); @@ -379,6 +383,8 @@ static void omap_crtc_atomic_disable(struct drm_crtc *crtc, spin_unlock_irq(&crtc->dev->event_lock); drm_crtc_vblank_off(crtc); + + priv->dispc_ops->runtime_put(priv->dispc); } static enum drm_mode_status omap_crtc_mode_valid(struct drm_crtc *crtc, diff --git a/drivers/gpu/drm/omapdrm/omap_encoder.c b/drivers/gpu/drm/omapdrm/omap_encoder.c index 452e625f6ce3..933ebc9f9faa 100644 --- a/drivers/gpu/drm/omapdrm/omap_encoder.c +++ b/drivers/gpu/drm/omapdrm/omap_encoder.c @@ -52,17 +52,44 @@ static const struct drm_encoder_funcs omap_encoder_funcs = { .destroy = omap_encoder_destroy, }; +static void omap_encoder_hdmi_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *adjusted_mode) +{ + struct drm_device *dev = encoder->dev; + struct omap_encoder *omap_encoder = to_omap_encoder(encoder); + struct omap_dss_device *dssdev = omap_encoder->output; + struct drm_connector *connector; + bool hdmi_mode; + + hdmi_mode = false; + list_for_each_entry(connector, &dev->mode_config.connector_list, head) { + if (connector->encoder == encoder) { + hdmi_mode = omap_connector_get_hdmi_mode(connector); + break; + } + } + + if (dssdev->ops->hdmi.set_hdmi_mode) + dssdev->ops->hdmi.set_hdmi_mode(dssdev, hdmi_mode); + + if (hdmi_mode && dssdev->ops->hdmi.set_infoframe) { + struct hdmi_avi_infoframe avi; + int r; + + r = drm_hdmi_avi_infoframe_from_display_mode(&avi, adjusted_mode, + false); + if (r == 0) + dssdev->ops->hdmi.set_infoframe(dssdev, &avi); + } +} + static void omap_encoder_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { - struct drm_device *dev = encoder->dev; struct omap_encoder *omap_encoder = to_omap_encoder(encoder); - struct drm_connector *connector; struct omap_dss_device *dssdev; struct videomode vm = { 0 }; - bool hdmi_mode; - int r; drm_display_mode_to_videomode(adjusted_mode, &vm); @@ -112,27 +139,8 @@ static void omap_encoder_mode_set(struct drm_encoder *encoder, } /* Set the HDMI mode and HDMI infoframe if applicable. */ - hdmi_mode = false; - list_for_each_entry(connector, &dev->mode_config.connector_list, head) { - if (connector->encoder == encoder) { - hdmi_mode = omap_connector_get_hdmi_mode(connector); - break; - } - } - - dssdev = omap_encoder->output; - - if (dssdev->ops->hdmi.set_hdmi_mode) - dssdev->ops->hdmi.set_hdmi_mode(dssdev, hdmi_mode); - - if (hdmi_mode && dssdev->ops->hdmi.set_infoframe) { - struct hdmi_avi_infoframe avi; - - r = drm_hdmi_avi_infoframe_from_display_mode(&avi, adjusted_mode, - false); - if (r == 0) - dssdev->ops->hdmi.set_infoframe(dssdev, &avi); - } + if (omap_encoder->output->output_type == OMAP_DISPLAY_TYPE_HDMI) + omap_encoder_hdmi_mode_set(encoder, adjusted_mode); } static void omap_encoder_disable(struct drm_encoder *encoder) diff --git a/drivers/gpu/drm/rcar-du/rcar_du_group.c b/drivers/gpu/drm/rcar-du/rcar_du_group.c index d85f0a1c1581..cebf313c6e1f 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_group.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_group.c @@ -202,10 +202,25 @@ void rcar_du_group_put(struct rcar_du_group *rgrp) static void __rcar_du_group_start_stop(struct rcar_du_group *rgrp, bool start) { - struct rcar_du_crtc *rcrtc = &rgrp->dev->crtcs[rgrp->index * 2]; + struct rcar_du_device *rcdu = rgrp->dev; + + /* + * Group start/stop is controlled by the DRES and DEN bits of DSYSR0 + * for the first group and DSYSR2 for the second group. On most DU + * instances, this maps to the first CRTC of the group, and we can just + * use rcar_du_crtc_dsysr_clr_set() to access the correct DSYSR. On + * M3-N, however, DU2 doesn't exist, but DSYSR2 does. We thus need to + * access the register directly using group read/write. + */ + if (rcdu->info->channels_mask & BIT(rgrp->index * 2)) { + struct rcar_du_crtc *rcrtc = &rgrp->dev->crtcs[rgrp->index * 2]; - rcar_du_crtc_dsysr_clr_set(rcrtc, DSYSR_DRES | DSYSR_DEN, - start ? DSYSR_DEN : DSYSR_DRES); + rcar_du_crtc_dsysr_clr_set(rcrtc, DSYSR_DRES | DSYSR_DEN, + start ? DSYSR_DEN : DSYSR_DRES); + } else { + rcar_du_group_write(rgrp, DSYSR, + start ? DSYSR_DEN : DSYSR_DRES); + } } void rcar_du_group_start_stop(struct rcar_du_group *rgrp, bool start) diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c index 941f35233b1f..5864cb452c5c 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c @@ -448,11 +448,6 @@ static int rockchip_drm_platform_remove(struct platform_device *pdev) return 0; } -static void rockchip_drm_platform_shutdown(struct platform_device *pdev) -{ - rockchip_drm_platform_remove(pdev); -} - static const struct of_device_id rockchip_drm_dt_ids[] = { { .compatible = "rockchip,display-subsystem", }, { /* sentinel */ }, @@ -462,7 +457,6 @@ MODULE_DEVICE_TABLE(of, rockchip_drm_dt_ids); static struct platform_driver rockchip_drm_platform_driver = { .probe = rockchip_drm_platform_probe, .remove = rockchip_drm_platform_remove, - .shutdown = rockchip_drm_platform_shutdown, .driver = { .name = "rockchip-drm", .of_match_table = rockchip_drm_dt_ids, diff --git a/drivers/gpu/drm/sun4i/sun4i_lvds.c b/drivers/gpu/drm/sun4i/sun4i_lvds.c index af7dcb6da351..e7eb0d1e17be 100644 --- a/drivers/gpu/drm/sun4i/sun4i_lvds.c +++ b/drivers/gpu/drm/sun4i/sun4i_lvds.c @@ -75,7 +75,7 @@ static void sun4i_lvds_encoder_enable(struct drm_encoder *encoder) DRM_DEBUG_DRIVER("Enabling LVDS output\n"); - if (!IS_ERR(tcon->panel)) { + if (tcon->panel) { drm_panel_prepare(tcon->panel); drm_panel_enable(tcon->panel); } @@ -88,7 +88,7 @@ static void sun4i_lvds_encoder_disable(struct drm_encoder *encoder) DRM_DEBUG_DRIVER("Disabling LVDS output\n"); - if (!IS_ERR(tcon->panel)) { + if (tcon->panel) { drm_panel_disable(tcon->panel); drm_panel_unprepare(tcon->panel); } diff --git a/drivers/gpu/drm/sun4i/sun4i_rgb.c b/drivers/gpu/drm/sun4i/sun4i_rgb.c index bf068da6b12e..f4a22689eb54 100644 --- a/drivers/gpu/drm/sun4i/sun4i_rgb.c +++ b/drivers/gpu/drm/sun4i/sun4i_rgb.c @@ -135,7 +135,7 @@ static void sun4i_rgb_encoder_enable(struct drm_encoder *encoder) DRM_DEBUG_DRIVER("Enabling RGB output\n"); - if (!IS_ERR(tcon->panel)) { + if (tcon->panel) { drm_panel_prepare(tcon->panel); drm_panel_enable(tcon->panel); } @@ -148,7 +148,7 @@ static void sun4i_rgb_encoder_disable(struct drm_encoder *encoder) DRM_DEBUG_DRIVER("Disabling RGB output\n"); - if (!IS_ERR(tcon->panel)) { + if (tcon->panel) { drm_panel_disable(tcon->panel); drm_panel_unprepare(tcon->panel); } diff --git a/drivers/gpu/drm/sun4i/sun4i_tcon.c b/drivers/gpu/drm/sun4i/sun4i_tcon.c index c78cd35a1294..f949287d926c 100644 --- a/drivers/gpu/drm/sun4i/sun4i_tcon.c +++ b/drivers/gpu/drm/sun4i/sun4i_tcon.c @@ -491,7 +491,8 @@ static void sun4i_tcon0_mode_set_rgb(struct sun4i_tcon *tcon, sun4i_tcon0_mode_set_common(tcon, mode); /* Set dithering if needed */ - sun4i_tcon0_mode_set_dithering(tcon, tcon->panel->connector); + if (tcon->panel) + sun4i_tcon0_mode_set_dithering(tcon, tcon->panel->connector); /* Adjust clock delay */ clk_delay = sun4i_tcon_get_clk_delay(mode, 0); @@ -555,7 +556,7 @@ static void sun4i_tcon0_mode_set_rgb(struct sun4i_tcon *tcon, * Following code is a way to avoid quirks all around TCON * and DOTCLOCK drivers. */ - if (!IS_ERR(tcon->panel)) { + if (tcon->panel) { struct drm_panel *panel = tcon->panel; struct drm_connector *connector = panel->connector; struct drm_display_info display_info = connector->display_info; diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index ba80150d1052..895d77d799e4 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -492,8 +492,10 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, if (!fbo) return -ENOMEM; - ttm_bo_get(bo); fbo->base = *bo; + fbo->base.mem.placement |= TTM_PL_FLAG_NO_EVICT; + + ttm_bo_get(bo); fbo->bo = bo; /** diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c index 127468785f74..1f94b9affe4b 100644 --- a/drivers/gpu/drm/vc4/vc4_kms.c +++ b/drivers/gpu/drm/vc4/vc4_kms.c @@ -214,6 +214,12 @@ static int vc4_atomic_commit(struct drm_device *dev, return 0; } + /* We know for sure we don't want an async update here. Set + * state->legacy_cursor_update to false to prevent + * drm_atomic_helper_setup_commit() from auto-completing + * commit->flip_done. + */ + state->legacy_cursor_update = false; ret = drm_atomic_helper_setup_commit(state, nonblock); if (ret) return ret; diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c index 9dc3fcbd290b..c6635f23918a 100644 --- a/drivers/gpu/drm/vc4/vc4_plane.c +++ b/drivers/gpu/drm/vc4/vc4_plane.c @@ -807,7 +807,7 @@ void vc4_plane_async_set_fb(struct drm_plane *plane, struct drm_framebuffer *fb) static void vc4_plane_atomic_async_update(struct drm_plane *plane, struct drm_plane_state *state) { - struct vc4_plane_state *vc4_state = to_vc4_plane_state(plane->state); + struct vc4_plane_state *vc4_state, *new_vc4_state; if (plane->state->fb != state->fb) { vc4_plane_async_set_fb(plane, state->fb); @@ -828,7 +828,18 @@ static void vc4_plane_atomic_async_update(struct drm_plane *plane, plane->state->src_y = state->src_y; /* Update the display list based on the new crtc_x/y. */ - vc4_plane_atomic_check(plane, plane->state); + vc4_plane_atomic_check(plane, state); + + new_vc4_state = to_vc4_plane_state(state); + vc4_state = to_vc4_plane_state(plane->state); + + /* Update the current vc4_state pos0, pos2 and ptr0 dlist entries. */ + vc4_state->dlist[vc4_state->pos0_offset] = + new_vc4_state->dlist[vc4_state->pos0_offset]; + vc4_state->dlist[vc4_state->pos2_offset] = + new_vc4_state->dlist[vc4_state->pos2_offset]; + vc4_state->dlist[vc4_state->ptr0_offset] = + new_vc4_state->dlist[vc4_state->ptr0_offset]; /* Note that we can't just call vc4_plane_write_dlist() * because that would smash the context data that the HVS is diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index 61a84b958d67..d7a2dfb8ee9b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -49,6 +49,8 @@ #define VMWGFX_REPO "In Tree" +#define VMWGFX_VALIDATION_MEM_GRAN (16*PAGE_SIZE) + /** * Fully encoded drm commands. Might move to vmw_drm.h @@ -918,7 +920,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) spin_unlock(&dev_priv->cap_lock); } - + vmw_validation_mem_init_ttm(dev_priv, VMWGFX_VALIDATION_MEM_GRAN); ret = vmw_kms_init(dev_priv); if (unlikely(ret != 0)) goto out_no_kms; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 59f614225bcd..aca974b14b55 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -606,6 +606,9 @@ struct vmw_private { struct vmw_cmdbuf_man *cman; DECLARE_BITMAP(irqthread_pending, VMW_IRQTHREAD_MAX); + + /* Validation memory reservation */ + struct vmw_validation_mem vvm; }; static inline struct vmw_surface *vmw_res_to_srf(struct vmw_resource *res) @@ -846,6 +849,8 @@ extern int vmw_ttm_global_init(struct vmw_private *dev_priv); extern void vmw_ttm_global_release(struct vmw_private *dev_priv); extern int vmw_mmap(struct file *filp, struct vm_area_struct *vma); +extern void vmw_validation_mem_init_ttm(struct vmw_private *dev_priv, + size_t gran); /** * TTM buffer object driver - vmwgfx_ttm_buffer.c */ diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 5a6b70ba137a..f2d13a72c05d 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -1738,7 +1738,6 @@ static int vmw_cmd_check_define_gmrfb(struct vmw_private *dev_priv, void *buf) { struct vmw_buffer_object *vmw_bo; - int ret; struct { uint32_t header; @@ -1748,7 +1747,6 @@ static int vmw_cmd_check_define_gmrfb(struct vmw_private *dev_priv, return vmw_translate_guest_ptr(dev_priv, sw_context, &cmd->body.ptr, &vmw_bo); - return ret; } @@ -3837,6 +3835,8 @@ int vmw_execbuf_process(struct drm_file *file_priv, struct sync_file *sync_file = NULL; DECLARE_VAL_CONTEXT(val_ctx, &sw_context->res_ht, 1); + vmw_validation_set_val_mem(&val_ctx, &dev_priv->vvm); + if (flags & DRM_VMW_EXECBUF_FLAG_EXPORT_FENCE_FD) { out_fence_fd = get_unused_fd_flags(O_CLOEXEC); if (out_fence_fd < 0) { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c index 7b1e5a5cbd2c..f88247046721 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c @@ -96,3 +96,39 @@ void vmw_ttm_global_release(struct vmw_private *dev_priv) drm_global_item_unref(&dev_priv->bo_global_ref.ref); drm_global_item_unref(&dev_priv->mem_global_ref); } + +/* struct vmw_validation_mem callback */ +static int vmw_vmt_reserve(struct vmw_validation_mem *m, size_t size) +{ + static struct ttm_operation_ctx ctx = {.interruptible = false, + .no_wait_gpu = false}; + struct vmw_private *dev_priv = container_of(m, struct vmw_private, vvm); + + return ttm_mem_global_alloc(vmw_mem_glob(dev_priv), size, &ctx); +} + +/* struct vmw_validation_mem callback */ +static void vmw_vmt_unreserve(struct vmw_validation_mem *m, size_t size) +{ + struct vmw_private *dev_priv = container_of(m, struct vmw_private, vvm); + + return ttm_mem_global_free(vmw_mem_glob(dev_priv), size); +} + +/** + * vmw_validation_mem_init_ttm - Interface the validation memory tracker + * to ttm. + * @dev_priv: Pointer to struct vmw_private. The reason we choose a vmw private + * rather than a struct vmw_validation_mem is to make sure assumption in the + * callbacks that struct vmw_private derives from struct vmw_validation_mem + * holds true. + * @gran: The recommended allocation granularity + */ +void vmw_validation_mem_init_ttm(struct vmw_private *dev_priv, size_t gran) +{ + struct vmw_validation_mem *vvm = &dev_priv->vvm; + + vvm->reserve_mem = vmw_vmt_reserve; + vvm->unreserve_mem = vmw_vmt_unreserve; + vvm->gran = gran; +} diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c index 184025fa938e..f116f092e00b 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c @@ -104,11 +104,25 @@ void *vmw_validation_mem_alloc(struct vmw_validation_context *ctx, return NULL; if (ctx->mem_size_left < size) { - struct page *page = alloc_page(GFP_KERNEL | __GFP_ZERO); + struct page *page; + if (ctx->vm && ctx->vm_size_left < PAGE_SIZE) { + int ret = ctx->vm->reserve_mem(ctx->vm, ctx->vm->gran); + + if (ret) + return NULL; + + ctx->vm_size_left += ctx->vm->gran; + ctx->total_mem += ctx->vm->gran; + } + + page = alloc_page(GFP_KERNEL | __GFP_ZERO); if (!page) return NULL; + if (ctx->vm) + ctx->vm_size_left -= PAGE_SIZE; + list_add_tail(&page->lru, &ctx->page_list); ctx->page_address = page_address(page); ctx->mem_size_left = PAGE_SIZE; @@ -138,6 +152,11 @@ static void vmw_validation_mem_free(struct vmw_validation_context *ctx) } ctx->mem_size_left = 0; + if (ctx->vm && ctx->total_mem) { + ctx->vm->unreserve_mem(ctx->vm, ctx->total_mem); + ctx->total_mem = 0; + ctx->vm_size_left = 0; + } } /** diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h index b57e3292c386..3b396fea40d7 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h @@ -34,6 +34,21 @@ #include <drm/ttm/ttm_execbuf_util.h> /** + * struct vmw_validation_mem - Custom interface to provide memory reservations + * for the validation code. + * @reserve_mem: Callback to reserve memory + * @unreserve_mem: Callback to unreserve memory + * @gran: Reservation granularity. Contains a hint how much memory should + * be reserved in each call to @reserve_mem(). A slow implementation may want + * reservation to be done in large batches. + */ +struct vmw_validation_mem { + int (*reserve_mem)(struct vmw_validation_mem *m, size_t size); + void (*unreserve_mem)(struct vmw_validation_mem *m, size_t size); + size_t gran; +}; + +/** * struct vmw_validation_context - Per command submission validation context * @ht: Hash table used to find resource- or buffer object duplicates * @resource_list: List head for resource validation metadata @@ -47,6 +62,10 @@ * buffer objects * @mem_size_left: Free memory left in the last page in @page_list * @page_address: Kernel virtual address of the last page in @page_list + * @vm: A pointer to the memory reservation interface or NULL if no + * memory reservation is needed. + * @vm_size_left: Amount of reserved memory that so far has not been allocated. + * @total_mem: Amount of reserved memory. */ struct vmw_validation_context { struct drm_open_hash *ht; @@ -59,6 +78,9 @@ struct vmw_validation_context { unsigned int merge_dups; unsigned int mem_size_left; u8 *page_address; + struct vmw_validation_mem *vm; + size_t vm_size_left; + size_t total_mem; }; struct vmw_buffer_object; @@ -102,6 +124,21 @@ vmw_validation_has_bos(struct vmw_validation_context *ctx) } /** + * vmw_validation_set_val_mem - Register a validation mem object for + * validation memory reservation + * @ctx: The validation context + * @vm: Pointer to a struct vmw_validation_mem + * + * Must be set before the first attempt to allocate validation memory. + */ +static inline void +vmw_validation_set_val_mem(struct vmw_validation_context *ctx, + struct vmw_validation_mem *vm) +{ + ctx->vm = vm; +} + +/** * vmw_validation_set_ht - Register a hash table for duplicate finding * @ctx: The validation context * @ht: Pointer to a hash table to use for duplicate finding |