diff options
Diffstat (limited to 'drivers/gpu/drm')
99 files changed, 1261 insertions, 652 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 0165783..f7b49d5c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1130,6 +1130,9 @@ struct amdgpu_gfx { uint32_t me_feature_version; uint32_t ce_feature_version; uint32_t pfp_feature_version; + uint32_t rlc_feature_version; + uint32_t mec_feature_version; + uint32_t mec2_feature_version; struct amdgpu_ring gfx_ring[AMDGPU_MAX_GFX_RINGS]; unsigned num_gfx_rings; struct amdgpu_ring compute_ring[AMDGPU_MAX_COMPUTE_RINGS]; @@ -1614,6 +1617,9 @@ struct amdgpu_uvd { #define AMDGPU_MAX_VCE_HANDLES 16 #define AMDGPU_VCE_FIRMWARE_OFFSET 256 +#define AMDGPU_VCE_HARVEST_VCE0 (1 << 0) +#define AMDGPU_VCE_HARVEST_VCE1 (1 << 1) + struct amdgpu_vce { struct amdgpu_bo *vcpu_bo; uint64_t gpu_addr; @@ -1626,6 +1632,7 @@ struct amdgpu_vce { const struct firmware *fw; /* VCE firmware */ struct amdgpu_ring ring[AMDGPU_MAX_VCE_RINGS]; struct amdgpu_irq_src irq; + unsigned harvest_config; }; /* @@ -1635,6 +1642,7 @@ struct amdgpu_sdma { /* SDMA firmware */ const struct firmware *fw; uint32_t fw_version; + uint32_t feature_version; struct amdgpu_ring ring; }; @@ -1862,6 +1870,12 @@ typedef void (*amdgpu_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t); typedef uint32_t (*amdgpu_block_rreg_t)(struct amdgpu_device*, uint32_t, uint32_t); typedef void (*amdgpu_block_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t, uint32_t); +struct amdgpu_ip_block_status { + bool valid; + bool sw; + bool hw; +}; + struct amdgpu_device { struct device *dev; struct drm_device *ddev; @@ -2004,7 +2018,7 @@ struct amdgpu_device { const struct amdgpu_ip_block_version *ip_blocks; int num_ip_blocks; - bool *ip_block_enabled; + struct amdgpu_ip_block_status *ip_block_status; struct mutex mn_lock; DECLARE_HASHTABLE(mn_hash, 7); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index d63135b..1f040d8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -669,6 +669,7 @@ static int amdgpu_cs_ib_fill(struct amdgpu_device *adev, static int amdgpu_cs_dependencies(struct amdgpu_device *adev, struct amdgpu_cs_parser *p) { + struct amdgpu_fpriv *fpriv = p->filp->driver_priv; struct amdgpu_ib *ib; int i, j, r; @@ -694,6 +695,7 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev, for (j = 0; j < num_deps; ++j) { struct amdgpu_fence *fence; struct amdgpu_ring *ring; + struct amdgpu_ctx *ctx; r = amdgpu_cs_get_ring(adev, deps[j].ip_type, deps[j].ip_instance, @@ -701,14 +703,21 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev, if (r) return r; + ctx = amdgpu_ctx_get(fpriv, deps[j].ctx_id); + if (ctx == NULL) + return -EINVAL; + r = amdgpu_fence_recreate(ring, p->filp, deps[j].handle, &fence); - if (r) + if (r) { + amdgpu_ctx_put(ctx); return r; + } amdgpu_sync_fence(&ib->sync, fence); amdgpu_fence_unref(&fence); + amdgpu_ctx_put(ctx); } } @@ -808,12 +817,16 @@ int amdgpu_cs_wait_ioctl(struct drm_device *dev, void *data, r = amdgpu_cs_get_ring(adev, wait->in.ip_type, wait->in.ip_instance, wait->in.ring, &ring); - if (r) + if (r) { + amdgpu_ctx_put(ctx); return r; + } r = amdgpu_fence_recreate(ring, filp, wait->in.handle, &fence); - if (r) + if (r) { + amdgpu_ctx_put(ctx); return r; + } r = fence_wait_timeout(&fence->base, true, timeout); amdgpu_fence_unref(&fence); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index ba46be3..99f158e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1191,8 +1191,9 @@ static int amdgpu_early_init(struct amdgpu_device *adev) return -EINVAL; } - adev->ip_block_enabled = kcalloc(adev->num_ip_blocks, sizeof(bool), GFP_KERNEL); - if (adev->ip_block_enabled == NULL) + adev->ip_block_status = kcalloc(adev->num_ip_blocks, + sizeof(struct amdgpu_ip_block_status), GFP_KERNEL); + if (adev->ip_block_status == NULL) return -ENOMEM; if (adev->ip_blocks == NULL) { @@ -1203,14 +1204,19 @@ static int amdgpu_early_init(struct amdgpu_device *adev) for (i = 0; i < adev->num_ip_blocks; i++) { if ((amdgpu_ip_block_mask & (1 << i)) == 0) { DRM_ERROR("disabled ip block: %d\n", i); - adev->ip_block_enabled[i] = false; + adev->ip_block_status[i].valid = false; } else { if (adev->ip_blocks[i].funcs->early_init) { r = adev->ip_blocks[i].funcs->early_init((void *)adev); - if (r) + if (r == -ENOENT) + adev->ip_block_status[i].valid = false; + else if (r) return r; + else + adev->ip_block_status[i].valid = true; + } else { + adev->ip_block_status[i].valid = true; } - adev->ip_block_enabled[i] = true; } } @@ -1222,11 +1228,12 @@ static int amdgpu_init(struct amdgpu_device *adev) int i, r; for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].valid) continue; r = adev->ip_blocks[i].funcs->sw_init((void *)adev); if (r) return r; + adev->ip_block_status[i].sw = true; /* need to do gmc hw init early so we can allocate gpu mem */ if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) { r = amdgpu_vram_scratch_init(adev); @@ -1238,11 +1245,12 @@ static int amdgpu_init(struct amdgpu_device *adev) r = amdgpu_wb_init(adev); if (r) return r; + adev->ip_block_status[i].hw = true; } } for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].sw) continue; /* gmc hw init is done early */ if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) @@ -1250,6 +1258,7 @@ static int amdgpu_init(struct amdgpu_device *adev) r = adev->ip_blocks[i].funcs->hw_init((void *)adev); if (r) return r; + adev->ip_block_status[i].hw = true; } return 0; @@ -1260,7 +1269,7 @@ static int amdgpu_late_init(struct amdgpu_device *adev) int i = 0, r; for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].valid) continue; /* enable clockgating to save power */ r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev, @@ -1282,7 +1291,7 @@ static int amdgpu_fini(struct amdgpu_device *adev) int i, r; for (i = adev->num_ip_blocks - 1; i >= 0; i--) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].hw) continue; if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) { amdgpu_wb_fini(adev); @@ -1295,14 +1304,16 @@ static int amdgpu_fini(struct amdgpu_device *adev) return r; r = adev->ip_blocks[i].funcs->hw_fini((void *)adev); /* XXX handle errors */ + adev->ip_block_status[i].hw = false; } for (i = adev->num_ip_blocks - 1; i >= 0; i--) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].sw) continue; r = adev->ip_blocks[i].funcs->sw_fini((void *)adev); /* XXX handle errors */ - adev->ip_block_enabled[i] = false; + adev->ip_block_status[i].sw = false; + adev->ip_block_status[i].valid = false; } return 0; @@ -1313,7 +1324,7 @@ static int amdgpu_suspend(struct amdgpu_device *adev) int i, r; for (i = adev->num_ip_blocks - 1; i >= 0; i--) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].valid) continue; /* ungate blocks so that suspend can properly shut them down */ r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev, @@ -1331,7 +1342,7 @@ static int amdgpu_resume(struct amdgpu_device *adev) int i, r; for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_block_enabled[i]) + if (!adev->ip_block_status[i].valid) continue; r = adev->ip_blocks[i].funcs->resume(adev); if (r) @@ -1577,8 +1588,8 @@ void amdgpu_device_fini(struct amdgpu_device *adev) amdgpu_fence_driver_fini(adev); amdgpu_fbdev_fini(adev); r = amdgpu_fini(adev); - kfree(adev->ip_block_enabled); - adev->ip_block_enabled = NULL; + kfree(adev->ip_block_status); + adev->ip_block_status = NULL; adev->accel_working = false; /* free i2c buses */ amdgpu_i2c_fini(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index ae43b58..4afc507 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -449,7 +449,7 @@ out: * vital here, so they are not reported back to userspace. */ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev, - struct amdgpu_bo_va *bo_va) + struct amdgpu_bo_va *bo_va, uint32_t operation) { struct ttm_validate_buffer tv, *entry; struct amdgpu_bo_list_entry *vm_bos; @@ -485,7 +485,9 @@ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev, if (r) goto error_unlock; - r = amdgpu_vm_bo_update(adev, bo_va, &bo_va->bo->tbo.mem); + + if (operation == AMDGPU_VA_OP_MAP) + r = amdgpu_vm_bo_update(adev, bo_va, &bo_va->bo->tbo.mem); error_unlock: mutex_unlock(&bo_va->vm->mutex); @@ -580,7 +582,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, } if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE)) - amdgpu_gem_va_update_vm(adev, bo_va); + amdgpu_gem_va_update_vm(adev, bo_va, args->operation); drm_gem_object_unreference_unlocked(gobj); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index 52dff75..bc0fac6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -180,16 +180,16 @@ int amdgpu_ib_schedule(struct amdgpu_device *adev, unsigned num_ibs, if (vm) { /* do context switch */ amdgpu_vm_flush(ring, vm, ib->sync.last_vm_update); - } - if (vm && ring->funcs->emit_gds_switch) - amdgpu_ring_emit_gds_switch(ring, ib->vm->ids[ring->idx].id, - ib->gds_base, ib->gds_size, - ib->gws_base, ib->gws_size, - ib->oa_base, ib->oa_size); + if (ring->funcs->emit_gds_switch) + amdgpu_ring_emit_gds_switch(ring, ib->vm->ids[ring->idx].id, + ib->gds_base, ib->gds_size, + ib->gws_base, ib->gws_size, + ib->oa_base, ib->oa_size); - if (ring->funcs->emit_hdp_flush) - amdgpu_ring_emit_hdp_flush(ring); + if (ring->funcs->emit_hdp_flush) + amdgpu_ring_emit_hdp_flush(ring); + } old_ctx = ring->current_ctx; for (i = 0; i < num_ibs; ++i) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 5533434..3bfe67d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -235,7 +235,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file for (i = 0; i < adev->num_ip_blocks; i++) { if (adev->ip_blocks[i].type == type && - adev->ip_block_enabled[i]) { + adev->ip_block_status[i].valid) { ip.hw_ip_version_major = adev->ip_blocks[i].major; ip.hw_ip_version_minor = adev->ip_blocks[i].minor; ip.capabilities_flags = 0; @@ -274,7 +274,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file for (i = 0; i < adev->num_ip_blocks; i++) if (adev->ip_blocks[i].type == type && - adev->ip_block_enabled[i] && + adev->ip_block_status[i].valid && count < AMDGPU_HW_IP_INSTANCE_MAX_COUNT) count++; @@ -317,16 +317,17 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file break; case AMDGPU_INFO_FW_GFX_RLC: fw_info.ver = adev->gfx.rlc_fw_version; - fw_info.feature = 0; + fw_info.feature = adev->gfx.rlc_feature_version; break; case AMDGPU_INFO_FW_GFX_MEC: - if (info->query_fw.index == 0) + if (info->query_fw.index == 0) { fw_info.ver = adev->gfx.mec_fw_version; - else if (info->query_fw.index == 1) + fw_info.feature = adev->gfx.mec_feature_version; + } else if (info->query_fw.index == 1) { fw_info.ver = adev->gfx.mec2_fw_version; - else + fw_info.feature = adev->gfx.mec2_feature_version; + } else return -EINVAL; - fw_info.feature = 0; break; case AMDGPU_INFO_FW_SMC: fw_info.ver = adev->pm.fw_version; @@ -336,7 +337,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file if (info->query_fw.index >= 2) return -EINVAL; fw_info.ver = adev->sdma[info->query_fw.index].fw_version; - fw_info.feature = 0; + fw_info.feature = adev->sdma[info->query_fw.index].feature_version; break; default: return -EINVAL; @@ -416,7 +417,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file return n ? -EFAULT : 0; } case AMDGPU_INFO_DEV_INFO: { - struct drm_amdgpu_info_device dev_info; + struct drm_amdgpu_info_device dev_info = {}; struct amdgpu_cu_info cu_info; dev_info.device_id = dev->pdev->device; @@ -459,6 +460,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file memcpy(&dev_info.cu_bitmap[0], &cu_info.bitmap[0], sizeof(cu_info.bitmap)); dev_info.vram_type = adev->mc.vram_type; dev_info.vram_bit_width = adev->mc.vram_width; + dev_info.vce_harvest_config = adev->vce.harvest_config; return copy_to_user(out, &dev_info, min((size_t)size, sizeof(dev_info))) ? -EFAULT : 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 2f7a5ef..f5c2255 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -374,7 +374,7 @@ static int amdgpu_uvd_cs_msg_decode(uint32_t *msg, unsigned buf_sizes[]) unsigned height_in_mb = ALIGN(height / 16, 2); unsigned fs_in_mb = width_in_mb * height_in_mb; - unsigned image_size, tmp, min_dpb_size, num_dpb_buffer; + unsigned image_size, tmp, min_dpb_size, num_dpb_buffer, min_ctx_size; image_size = width * height; image_size += image_size / 2; @@ -466,6 +466,8 @@ static int amdgpu_uvd_cs_msg_decode(uint32_t *msg, unsigned buf_sizes[]) num_dpb_buffer = (le32_to_cpu(msg[59]) & 0xff) + 2; min_dpb_size = image_size * num_dpb_buffer; + min_ctx_size = ((width + 255) / 16) * ((height + 255) / 16) + * 16 * num_dpb_buffer + 52 * 1024; break; default: @@ -486,6 +488,7 @@ static int amdgpu_uvd_cs_msg_decode(uint32_t *msg, unsigned buf_sizes[]) buf_sizes[0x1] = dpb_size; buf_sizes[0x2] = image_size; + buf_sizes[0x4] = min_ctx_size; return 0; } @@ -628,6 +631,13 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx) return -EINVAL; } + } else if (cmd == 0x206) { + if ((end - start) < ctx->buf_sizes[4]) { + DRM_ERROR("buffer (%d) to small (%d / %d)!\n", cmd, + (unsigned)(end - start), + ctx->buf_sizes[4]); + return -EINVAL; + } } else if ((cmd != 0x100) && (cmd != 0x204)) { DRM_ERROR("invalid UVD command %X!\n", cmd); return -EINVAL; @@ -755,9 +765,10 @@ int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, uint32_t ib_idx) struct amdgpu_uvd_cs_ctx ctx = {}; unsigned buf_sizes[] = { [0x00000000] = 2048, - [0x00000001] = 32 * 1024 * 1024, - [0x00000002] = 2048 * 1152 * 3, + [0x00000001] = 0xFFFFFFFF, + [0x00000002] = 0xFFFFFFFF, [0x00000003] = 2048, + [0x00000004] = 0xFFFFFFFF, }; struct amdgpu_ib *ib = &parser->ibs[ib_idx]; int r; diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c index ab83cc1..15df46c 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c @@ -500,6 +500,7 @@ static int cik_sdma_load_microcode(struct amdgpu_device *adev) amdgpu_ucode_print_sdma_hdr(&hdr->header); fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4; adev->sdma[i].fw_version = le32_to_cpu(hdr->header.ucode_version); + adev->sdma[i].feature_version = le32_to_cpu(hdr->ucode_feature_version); fw_data = (const __le32 *) (adev->sdma[i].fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes)); WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], 0); diff --git a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c index f75a31d..ace870a 100644 --- a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c @@ -494,29 +494,67 @@ static void cz_dpm_fini(struct amdgpu_device *adev) amdgpu_free_extended_power_table(adev); } +#define ixSMUSVI_NB_CURRENTVID 0xD8230044 +#define CURRENT_NB_VID_MASK 0xff000000 +#define CURRENT_NB_VID__SHIFT 24 +#define ixSMUSVI_GFX_CURRENTVID 0xD8230048 +#define CURRENT_GFX_VID_MASK 0xff000000 +#define CURRENT_GFX_VID__SHIFT 24 + static void cz_dpm_debugfs_print_current_performance_level(struct amdgpu_device *adev, struct seq_file *m) { + struct cz_power_info *pi = cz_get_pi(adev); struct amdgpu_clock_voltage_dependency_table *table = &adev->pm.dpm.dyn_state.vddc_dependency_on_sclk; - u32 current_index = - (RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX) & - TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX_MASK) >> - TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX__SHIFT; - u32 sclk, tmp; - u16 vddc; - - if (current_index >= NUM_SCLK_LEVELS) { - seq_printf(m, "invalid dpm profile %d\n", current_index); + struct amdgpu_uvd_clock_voltage_dependency_table *uvd_table = + &adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table; + struct amdgpu_vce_clock_voltage_dependency_table *vce_table = + &adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table; + u32 sclk_index = REG_GET_FIELD(RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX), + TARGET_AND_CURRENT_PROFILE_INDEX, CURR_SCLK_INDEX); + u32 uvd_index = REG_GET_FIELD(RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX_2), + TARGET_AND_CURRENT_PROFILE_INDEX_2, CURR_UVD_INDEX); + u32 vce_index = REG_GET_FIELD(RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX_2), + TARGET_AND_CURRENT_PROFILE_INDEX_2, CURR_VCE_INDEX); + u32 sclk, vclk, dclk, ecclk, tmp; + u16 vddnb, vddgfx; + + if (sclk_index >= NUM_SCLK_LEVELS) { + seq_printf(m, "invalid sclk dpm profile %d\n", sclk_index); } else { - sclk = table->entries[current_index].clk; - tmp = (RREG32_SMC(ixSMU_VOLTAGE_STATUS) & - SMU_VOLTAGE_STATUS__SMU_VOLTAGE_CURRENT_LEVEL_MASK) >> - SMU_VOLTAGE_STATUS__SMU_VOLTAGE_CURRENT_LEVEL__SHIFT; - vddc = cz_convert_8bit_index_to_voltage(adev, (u16)tmp); - seq_printf(m, "power level %d sclk: %u vddc: %u\n", - current_index, sclk, vddc); + sclk = table->entries[sclk_index].clk; + seq_printf(m, "%u sclk: %u\n", sclk_index, sclk); + } + + tmp = (RREG32_SMC(ixSMUSVI_NB_CURRENTVID) & + CURRENT_NB_VID_MASK) >> CURRENT_NB_VID__SHIFT; + vddnb = cz_convert_8bit_index_to_voltage(adev, (u16)tmp); + tmp = (RREG32_SMC(ixSMUSVI_GFX_CURRENTVID) & + CURRENT_GFX_VID_MASK) >> CURRENT_GFX_VID__SHIFT; + vddgfx = cz_convert_8bit_index_to_voltage(adev, (u16)tmp); + seq_printf(m, "vddnb: %u vddgfx: %u\n", vddnb, vddgfx); + + seq_printf(m, "uvd %sabled\n", pi->uvd_power_gated ? "dis" : "en"); + if (!pi->uvd_power_gated) { + if (uvd_index >= CZ_MAX_HARDWARE_POWERLEVELS) { + seq_printf(m, "invalid uvd dpm level %d\n", uvd_index); + } else { + vclk = uvd_table->entries[uvd_index].vclk; + dclk = uvd_table->entries[uvd_index].dclk; + seq_printf(m, "%u uvd vclk: %u dclk: %u\n", uvd_index, vclk, dclk); + } + } + + seq_printf(m, "vce %sabled\n", pi->vce_power_gated ? "dis" : "en"); + if (!pi->vce_power_gated) { + if (vce_index >= CZ_MAX_HARDWARE_POWERLEVELS) { + seq_printf(m, "invalid vce dpm level %d\n", vce_index); + } else { + ecclk = vce_table->entries[vce_index].ecclk; + seq_printf(m, "%u vce ecclk: %u\n", vce_index, ecclk); + } } } @@ -1679,25 +1717,31 @@ static int cz_dpm_unforce_dpm_levels(struct amdgpu_device *adev) if (ret) return ret; - DRM_INFO("DPM unforce state min=%d, max=%d.\n", - pi->sclk_dpm.soft_min_clk, - pi->sclk_dpm.soft_max_clk); + DRM_DEBUG("DPM unforce state min=%d, max=%d.\n", + pi->sclk_dpm.soft_min_clk, + pi->sclk_dpm.soft_max_clk); return 0; } static int cz_dpm_force_dpm_level(struct amdgpu_device *adev, - enum amdgpu_dpm_forced_level level) + enum amdgpu_dpm_forced_level level) { int ret = 0; switch (level) { case AMDGPU_DPM_FORCED_LEVEL_HIGH: + ret = cz_dpm_unforce_dpm_levels(adev); + if (ret) + return ret; ret = cz_dpm_force_highest(adev); if (ret) return ret; break; case AMDGPU_DPM_FORCED_LEVEL_LOW: + ret = cz_dpm_unforce_dpm_levels(adev); + if (ret) + return ret; ret = cz_dpm_force_lowest(adev); if (ret) return ret; @@ -1711,6 +1755,8 @@ static int cz_dpm_force_dpm_level(struct amdgpu_device *adev, break; } + adev->pm.dpm.forced_level = level; + return ret; } diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 6e77964..e70a26f 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -2632,6 +2632,7 @@ static void dce_v10_0_crtc_dpms(struct drm_crtc *crtc, int mode) struct drm_device *dev = crtc->dev; struct amdgpu_device *adev = dev->dev_private; struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); + unsigned type; switch (mode) { case DRM_MODE_DPMS_ON: @@ -2640,6 +2641,9 @@ static void dce_v10_0_crtc_dpms(struct drm_crtc *crtc, int mode) dce_v10_0_vga_enable(crtc, true); amdgpu_atombios_crtc_blank(crtc, ATOM_DISABLE); dce_v10_0_vga_enable(crtc, false); + /* Make sure VBLANK interrupt is still enabled */ + type = amdgpu_crtc_idx_to_irq_type(adev, amdgpu_crtc->crtc_id); + amdgpu_irq_update(adev, &adev->crtc_irq, type); drm_vblank_post_modeset(dev, amdgpu_crtc->crtc_id); dce_v10_0_crtc_load_lut(crtc); break; diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index 7f7abb0..dcb402e 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -2631,6 +2631,7 @@ static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode) struct drm_device *dev = crtc->dev; struct amdgpu_device *adev = dev->dev_private; struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); + unsigned type; switch (mode) { case DRM_MODE_DPMS_ON: @@ -2639,6 +2640,9 @@ static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode) dce_v11_0_vga_enable(crtc, true); amdgpu_atombios_crtc_blank(crtc, ATOM_DISABLE); dce_v11_0_vga_enable(crtc, false); + /* Make sure VBLANK interrupt is still enabled */ + type = amdgpu_crtc_idx_to_irq_type(adev, amdgpu_crtc->crtc_id); + amdgpu_irq_update(adev, &adev->crtc_irq, type); drm_vblank_post_modeset(dev, amdgpu_crtc->crtc_id); dce_v11_0_crtc_load_lut(crtc); break; diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index 08387df..cc050a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -2566,6 +2566,7 @@ static void dce_v8_0_crtc_dpms(struct drm_crtc *crtc, int mode) struct drm_device *dev = crtc->dev; struct amdgpu_device *adev = dev->dev_private; struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); + unsigned type; switch (mode) { case DRM_MODE_DPMS_ON: @@ -2574,6 +2575,9 @@ static void dce_v8_0_crtc_dpms(struct drm_crtc *crtc, int mode) dce_v8_0_vga_enable(crtc, true); amdgpu_atombios_crtc_blank(crtc, ATOM_DISABLE); dce_v8_0_vga_enable(crtc, false); + /* Make sure VBLANK interrupt is still enabled */ + type = amdgpu_crtc_idx_to_irq_type(adev, amdgpu_crtc->crtc_id); + amdgpu_irq_update(adev, &adev->crtc_irq, type); drm_vblank_post_modeset(dev, amdgpu_crtc->crtc_id); dce_v8_0_crtc_load_lut(crtc); break; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 2c188fb..0d8bf2c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -2561,7 +2561,7 @@ static bool gfx_v7_0_ring_emit_semaphore(struct amdgpu_ring *ring, * sheduling on the ring. This function schedules the IB * on the gfx ring for execution by the GPU. */ -static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring, +static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, struct amdgpu_ib *ib) { bool need_ctx_switch = ring->current_ctx != ib->ctx; @@ -2569,15 +2569,10 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring, u32 next_rptr = ring->wptr + 5; /* drop the CE preamble IB for the same context */ - if ((ring->type == AMDGPU_RING_TYPE_GFX) && - (ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && - !need_ctx_switch) + if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && !need_ctx_switch) return; - if (ring->type == AMDGPU_RING_TYPE_COMPUTE) - control |= INDIRECT_BUFFER_VALID; - - if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) + if (need_ctx_switch) next_rptr += 2; next_rptr += 4; @@ -2588,7 +2583,7 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring, amdgpu_ring_write(ring, next_rptr); /* insert SWITCH_BUFFER packet before first IB in the ring frame */ - if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) { + if (need_ctx_switch) { amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); amdgpu_ring_write(ring, 0); } @@ -2611,6 +2606,35 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring, amdgpu_ring_write(ring, control); } +static void gfx_v7_0_ring_emit_ib_compute(struct amdgpu_ring *ring, + struct amdgpu_ib *ib) +{ + u32 header, control = 0; + u32 next_rptr = ring->wptr + 5; + + control |= INDIRECT_BUFFER_VALID; + next_rptr += 4; + amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + amdgpu_ring_write(ring, WRITE_DATA_DST_SEL(5) | WR_CONFIRM); + amdgpu_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc); + amdgpu_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff); + amdgpu_ring_write(ring, next_rptr); + + header = PACKET3(PACKET3_INDIRECT_BUFFER, 2); + + control |= ib->length_dw | + (ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0); + + amdgpu_ring_write(ring, header); + amdgpu_ring_write(ring, +#ifdef __BIG_ENDIAN + (2 << 0) | +#endif + (ib->gpu_addr & 0xFFFFFFFC)); + amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF); + amdgpu_ring_write(ring, control); +} + /** * gfx_v7_0_ring_test_ib - basic ring IB test * @@ -3056,6 +3080,8 @@ static int gfx_v7_0_cp_compute_load_microcode(struct amdgpu_device *adev) mec_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data; amdgpu_ucode_print_gfx_hdr(&mec_hdr->header); adev->gfx.mec_fw_version = le32_to_cpu(mec_hdr->header.ucode_version); + adev->gfx.mec_feature_version = le32_to_cpu( + mec_hdr->ucode_feature_version); gfx_v7_0_cp_compute_enable(adev, false); @@ -3078,6 +3104,8 @@ static int gfx_v7_0_cp_compute_load_microcode(struct amdgpu_device *adev) mec2_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec2_fw->data; amdgpu_ucode_print_gfx_hdr(&mec2_hdr->header); adev->gfx.mec2_fw_version = le32_to_cpu(mec2_hdr->header.ucode_version); + adev->gfx.mec2_feature_version = le32_to_cpu( + mec2_hdr->ucode_feature_version); /* MEC2 */ fw_data = (const __le32 *) @@ -4042,6 +4070,8 @@ static int gfx_v7_0_rlc_resume(struct amdgpu_device *adev) hdr = (const struct rlc_firmware_header_v1_0 *)adev->gfx.rlc_fw->data; amdgpu_ucode_print_rlc_hdr(&hdr->header); adev->gfx.rlc_fw_version = le32_to_cpu(hdr->header.ucode_version); + adev->gfx.rlc_feature_version = le32_to_cpu( + hdr->ucode_feature_version); gfx_v7_0_rlc_stop(adev); @@ -5098,7 +5128,7 @@ static void gfx_v7_0_print_status(void *handle) dev_info(adev->dev, " CP_HPD_EOP_CONTROL=0x%08X\n", RREG32(mmCP_HPD_EOP_CONTROL)); - for (queue = 0; queue < 8; i++) { + for (queue = 0; queue < 8; queue++) { cik_srbm_select(adev, me, pipe, queue, 0); dev_info(adev->dev, " queue: %d\n", queue); dev_info(adev->dev, " CP_PQ_WPTR_POLL_CNTL=0x%08X\n", @@ -5555,7 +5585,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = { .get_wptr = gfx_v7_0_ring_get_wptr_gfx, .set_wptr = gfx_v7_0_ring_set_wptr_gfx, .parse_cs = NULL, - .emit_ib = gfx_v7_0_ring_emit_ib, + .emit_ib = gfx_v7_0_ring_emit_ib_gfx, .emit_fence = gfx_v7_0_ring_emit_fence_gfx, .emit_semaphore = gfx_v7_0_ring_emit_semaphore, .emit_vm_flush = gfx_v7_0_ring_emit_vm_flush, @@ -5571,7 +5601,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = { .get_wptr = gfx_v7_0_ring_get_wptr_compute, .set_wptr = gfx_v7_0_ring_set_wptr_compute, .parse_cs = NULL, - .emit_ib = gfx_v7_0_ring_emit_ib, + .emit_ib = gfx_v7_0_ring_emit_ib_compute, .emit_fence = gfx_v7_0_ring_emit_fence_compute, .emit_semaphore = gfx_v7_0_ring_emit_semaphore, .emit_vm_flush = gfx_v7_0_ring_emit_vm_flush, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 7b683fb..20e2cfd 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -587,6 +587,7 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) int err; struct amdgpu_firmware_info *info = NULL; const struct common_firmware_header *header = NULL; + const struct gfx_firmware_header_v1_0 *cp_hdr; DRM_DEBUG("\n"); @@ -611,6 +612,9 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) err = amdgpu_ucode_validate(adev->gfx.pfp_fw); if (err) goto out; + cp_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.pfp_fw->data; + adev->gfx.pfp_fw_version = le32_to_cpu(cp_hdr->header.ucode_version); + adev->gfx.pfp_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_me.bin", chip_name); err = request_firmware(&adev->gfx.me_fw, fw_name, adev->dev); @@ -619,6 +623,9 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) err = amdgpu_ucode_validate(adev->gfx.me_fw); if (err) goto out; + cp_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.me_fw->data; + adev->gfx.me_fw_version = le32_to_cpu(cp_hdr->header.ucode_version); + adev->gfx.me_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ce.bin", chip_name); err = request_firmware(&adev->gfx.ce_fw, fw_name, adev->dev); @@ -627,12 +634,18 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) err = amdgpu_ucode_validate(adev->gfx.ce_fw); if (err) goto out; + cp_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.ce_fw->data; + adev->gfx.ce_fw_version = le32_to_cpu(cp_hdr->header.ucode_version); + adev->gfx.ce_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", chip_name); err = request_firmware(&adev->gfx.rlc_fw, fw_name, adev->dev); if (err) goto out; err = amdgpu_ucode_validate(adev->gfx.rlc_fw); + cp_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.rlc_fw->data; + adev->gfx.rlc_fw_version = le32_to_cpu(cp_hdr->header.ucode_version); + adev->gfx.rlc_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec.bin", chip_name); err = request_firmware(&adev->gfx.mec_fw, fw_name, adev->dev); @@ -641,6 +654,9 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) err = amdgpu_ucode_validate(adev->gfx.mec_fw); if (err) goto out; + cp_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data; + adev->gfx.mec_fw_version = le32_to_cpu(cp_hdr->header.ucode_version); + adev->gfx.mec_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec2.bin", chip_name); err = request_firmware(&adev->gfx.mec2_fw, fw_name, adev->dev); @@ -648,6 +664,12 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev) err = amdgpu_ucode_validate(adev->gfx.mec2_fw); if (err) goto out; + cp_hdr = (const struct gfx_firmware_header_v1_0 *) + adev->gfx.mec2_fw->data; + adev->gfx.mec2_fw_version = le32_to_cpu( + cp_hdr->header.ucode_version); + adev->gfx.mec2_feature_version = le32_to_cpu( + cp_hdr->ucode_feature_version); } else { err = 0; adev->gfx.mec2_fw = NULL; @@ -1813,10 +1835,7 @@ static u32 gfx_v8_0_get_rb_disabled(struct amdgpu_device *adev, u32 data, mask; data = RREG32(mmCC_RB_BACKEND_DISABLE); - if (data & 1) - data &= CC_RB_BACKEND_DISABLE__BACKEND_DISABLE_MASK; - else - data = 0; + data &= CC_RB_BACKEND_DISABLE__BACKEND_DISABLE_MASK; data |= RREG32(mmGC_USER_RB_BACKEND_DISABLE); @@ -1986,6 +2005,7 @@ static void gfx_v8_0_gpu_init(struct amdgpu_device *adev) adev->gfx.config.max_shader_engines = 1; adev->gfx.config.max_tile_pipes = 2; adev->gfx.config.max_sh_per_se = 1; + adev->gfx.config.max_backends_per_se = 2; switch (adev->pdev->revision) { case 0xc4: @@ -1994,7 +2014,6 @@ static void gfx_v8_0_gpu_init(struct amdgpu_device *adev) case 0xcc: /* B10 */ adev->gfx.config.max_cu_per_sh = 8; - adev->gfx.config.max_backends_per_se = 2; break; case 0xc5: case 0x81: @@ -2003,14 +2022,12 @@ static void gfx_v8_0_gpu_init(struct amdgpu_device *adev) case 0xcd: /* B8 */ adev->gfx.config.max_cu_per_sh = 6; - adev->gfx.config.max_backends_per_se = 2; break; case 0xc6: case 0xca: case 0xce: /* B6 */ adev->gfx.config.max_cu_per_sh = 6; - adev->gfx.config.max_backends_per_se = 2; break; case 0xc7: case 0x87: @@ -2018,7 +2035,6 @@ static void gfx_v8_0_gpu_init(struct amdgpu_device *adev) default: /* B4 */ adev->gfx.config.max_cu_per_sh = 4; - adev->gfx.config.max_backends_per_se = 1; break; } @@ -2278,7 +2294,6 @@ static int gfx_v8_0_rlc_load_microcode(struct amdgpu_device *adev) hdr = (const struct rlc_firmware_header_v2_0 *)adev->gfx.rlc_fw->data; amdgpu_ucode_print_rlc_hdr(&hdr->header); - adev->gfx.rlc_fw_version = le32_to_cpu(hdr->header.ucode_version); fw_data = (const __le32 *)(adev->gfx.rlc_fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes)); @@ -2364,12 +2379,6 @@ static int gfx_v8_0_cp_gfx_load_microcode(struct amdgpu_device *adev) amdgpu_ucode_print_gfx_hdr(&pfp_hdr->header); amdgpu_ucode_print_gfx_hdr(&ce_hdr->header); amdgpu_ucode_print_gfx_hdr(&me_hdr->header); - adev->gfx.pfp_fw_version = le32_to_cpu(pfp_hdr->header.ucode_version); - adev->gfx.ce_fw_version = le32_to_cpu(ce_hdr->header.ucode_version); - adev->gfx.me_fw_version = le32_to_cpu(me_hdr->header.ucode_version); - adev->gfx.me_feature_version = le32_to_cpu(me_hdr->ucode_feature_version); - adev->gfx.ce_feature_version = le32_to_cpu(ce_hdr->ucode_feature_version); - adev->gfx.pfp_feature_version = le32_to_cpu(pfp_hdr->ucode_feature_version); gfx_v8_0_cp_gfx_enable(adev, false); @@ -2625,7 +2634,6 @@ static int gfx_v8_0_cp_compute_load_microcode(struct amdgpu_device *adev) mec_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data; amdgpu_ucode_print_gfx_hdr(&mec_hdr->header); - adev->gfx.mec_fw_version = le32_to_cpu(mec_hdr->header.ucode_version); fw_data = (const __le32 *) (adev->gfx.mec_fw->data + @@ -2644,7 +2652,6 @@ static int gfx_v8_0_cp_compute_load_microcode(struct amdgpu_device *adev) mec2_hdr = (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec2_fw->data; amdgpu_ucode_print_gfx_hdr(&mec2_hdr->header); - adev->gfx.mec2_fw_version = le32_to_cpu(mec2_hdr->header.ucode_version); fw_data = (const __le32 *) (adev->gfx.mec2_fw->data + @@ -3128,7 +3135,7 @@ static int gfx_v8_0_cp_compute_resume(struct amdgpu_device *adev) WREG32(mmCP_MEC_DOORBELL_RANGE_LOWER, AMDGPU_DOORBELL_KIQ << 2); WREG32(mmCP_MEC_DOORBELL_RANGE_UPPER, - 0x7FFFF << 2); + AMDGPU_DOORBELL_MEC_RING7 << 2); } tmp = RREG32(mmCP_HQD_PQ_DOORBELL_CONTROL); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, @@ -3756,7 +3763,7 @@ static void gfx_v8_0_ring_emit_hdp_flush(struct amdgpu_ring *ring) amdgpu_ring_write(ring, 0x20); /* poll interval */ } -static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring, +static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, struct amdgpu_ib *ib) { bool need_ctx_switch = ring->current_ctx != ib->ctx; @@ -3764,15 +3771,10 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring, u32 next_rptr = ring->wptr + 5; /* drop the CE preamble IB for the same context */ - if ((ring->type == AMDGPU_RING_TYPE_GFX) && - (ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && - !need_ctx_switch) + if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && !need_ctx_switch) return; - if (ring->type == AMDGPU_RING_TYPE_COMPUTE) - control |= INDIRECT_BUFFER_VALID; - - if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) + if (need_ctx_switch) next_rptr += 2; next_rptr += 4; @@ -3783,7 +3785,7 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring, amdgpu_ring_write(ring, next_rptr); /* insert SWITCH_BUFFER packet before first IB in the ring frame */ - if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) { + if (need_ctx_switch) { amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); amdgpu_ring_write(ring, 0); } @@ -3806,6 +3808,36 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring, amdgpu_ring_write(ring, control); } +static void gfx_v8_0_ring_emit_ib_compute(struct amdgpu_ring *ring, + struct amdgpu_ib *ib) +{ + u32 header, control = 0; + u32 next_rptr = ring->wptr + 5; + + control |= INDIRECT_BUFFER_VALID; + + next_rptr += 4; + amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + amdgpu_ring_write(ring, WRITE_DATA_DST_SEL(5) | WR_CONFIRM); + amdgpu_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc); + amdgpu_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff); + amdgpu_ring_write(ring, next_rptr); + + header = PACKET3(PACKET3_INDIRECT_BUFFER, 2); + + control |= ib->length_dw | + (ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0); + + amdgpu_ring_write(ring, header); + amdgpu_ring_write(ring, +#ifdef __BIG_ENDIAN + (2 << 0) | +#endif + (ib->gpu_addr & 0xFFFFFFFC)); + amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF); + amdgpu_ring_write(ring, control); +} + static void gfx_v8_0_ring_emit_fence_gfx(struct amdgpu_ring *ring, u64 addr, u64 seq, unsigned flags) { @@ -4227,7 +4259,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = { .get_wptr = gfx_v8_0_ring_get_wptr_gfx, .set_wptr = gfx_v8_0_ring_set_wptr_gfx, .parse_cs = NULL, - .emit_ib = gfx_v8_0_ring_emit_ib, + .emit_ib = gfx_v8_0_ring_emit_ib_gfx, .emit_fence = gfx_v8_0_ring_emit_fence_gfx, .emit_semaphore = gfx_v8_0_ring_emit_semaphore, .emit_vm_flush = gfx_v8_0_ring_emit_vm_flush, @@ -4243,7 +4275,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = { .get_wptr = gfx_v8_0_ring_get_wptr_compute, .set_wptr = gfx_v8_0_ring_set_wptr_compute, .parse_cs = NULL, - .emit_ib = gfx_v8_0_ring_emit_ib, + .emit_ib = gfx_v8_0_ring_emit_ib_compute, .emit_fence = gfx_v8_0_ring_emit_fence_compute, .emit_semaphore = gfx_v8_0_ring_emit_semaphore, .emit_vm_flush = gfx_v8_0_ring_emit_vm_flush, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c index d789588..a988dfb 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c @@ -121,6 +121,7 @@ static int sdma_v2_4_init_microcode(struct amdgpu_device *adev) int err, i; struct amdgpu_firmware_info *info = NULL; const struct common_firmware_header *header = NULL; + const struct sdma_firmware_header_v1_0 *hdr; DRM_DEBUG("\n"); @@ -142,6 +143,9 @@ static int sdma_v2_4_init_microcode(struct amdgpu_device *adev) err = amdgpu_ucode_validate(adev->sdma[i].fw); if (err) goto out; + hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma[i].fw->data; + adev->sdma[i].fw_version = le32_to_cpu(hdr->header.ucode_version); + adev->sdma[i].feature_version = le32_to_cpu(hdr->ucode_feature_version); if (adev->firmware.smu_load) { info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SDMA0 + i]; @@ -541,8 +545,6 @@ static int sdma_v2_4_load_microcode(struct amdgpu_device *adev) hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma[i].fw->data; amdgpu_ucode_print_sdma_hdr(&hdr->header); fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4; - adev->sdma[i].fw_version = le32_to_cpu(hdr->header.ucode_version); - fw_data = (const __le32 *) (adev->sdma[i].fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes)); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c index 7bb37b9..2b86569 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c @@ -159,6 +159,7 @@ static int sdma_v3_0_init_microcode(struct amdgpu_device *adev) int err, i; struct amdgpu_firmware_info *info = NULL; const struct common_firmware_header *header = NULL; + const struct sdma_firmware_header_v1_0 *hdr; DRM_DEBUG("\n"); @@ -183,6 +184,9 @@ static int sdma_v3_0_init_microcode(struct amdgpu_device *adev) err = amdgpu_ucode_validate(adev->sdma[i].fw); if (err) goto out; + hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma[i].fw->data; + adev->sdma[i].fw_version = le32_to_cpu(hdr->header.ucode_version); + adev->sdma[i].feature_version = le32_to_cpu(hdr->ucode_feature_version); if (adev->firmware.smu_load) { info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SDMA0 + i]; @@ -630,8 +634,6 @@ static int sdma_v3_0_load_microcode(struct amdgpu_device *adev) hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma[i].fw->data; amdgpu_ucode_print_sdma_hdr(&hdr->header); fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4; - adev->sdma[i].fw_version = le32_to_cpu(hdr->header.ucode_version); - fw_data = (const __le32 *) (adev->sdma[i].fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes)); diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c index d62c4002..d1064ca 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c @@ -35,6 +35,8 @@ #include "oss/oss_2_0_d.h" #include "oss/oss_2_0_sh_mask.h" #include "gca/gfx_8_0_d.h" +#include "smu/smu_7_1_2_d.h" +#include "smu/smu_7_1_2_sh_mask.h" #define GRBM_GFX_INDEX__VCE_INSTANCE__SHIFT 0x04 #define GRBM_GFX_INDEX__VCE_INSTANCE_MASK 0x10 @@ -112,6 +114,10 @@ static int vce_v3_0_start(struct amdgpu_device *adev) mutex_lock(&adev->grbm_idx_mutex); for (idx = 0; idx < 2; ++idx) { + + if (adev->vce.harvest_config & (1 << idx)) + continue; + if(idx == 0) WREG32_P(mmGRBM_GFX_INDEX, 0, ~GRBM_GFX_INDEX__VCE_INSTANCE_MASK); @@ -190,10 +196,52 @@ static int vce_v3_0_start(struct amdgpu_device *adev) return 0; } +#define ixVCE_HARVEST_FUSE_MACRO__ADDRESS 0xC0014074 +#define VCE_HARVEST_FUSE_MACRO__SHIFT 27 +#define VCE_HARVEST_FUSE_MACRO__MASK 0x18000000 + +static unsigned vce_v3_0_get_harvest_config(struct amdgpu_device *adev) +{ + u32 tmp; + unsigned ret; + + if (adev->flags & AMDGPU_IS_APU) + tmp = (RREG32_SMC(ixVCE_HARVEST_FUSE_MACRO__ADDRESS) & + VCE_HARVEST_FUSE_MACRO__MASK) >> + VCE_HARVEST_FUSE_MACRO__SHIFT; + else + tmp = (RREG32_SMC(ixCC_HARVEST_FUSES) & + CC_HARVEST_FUSES__VCE_DISABLE_MASK) >> + CC_HARVEST_FUSES__VCE_DISABLE__SHIFT; + + switch (tmp) { + case 1: + ret = AMDGPU_VCE_HARVEST_VCE0; + break; + case 2: + ret = AMDGPU_VCE_HARVEST_VCE1; + break; + case 3: + ret = AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1; + break; + default: + ret = 0; + } + + return ret; +} + static int vce_v3_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->vce.harvest_config = vce_v3_0_get_harvest_config(adev); + + if ((adev->vce.harvest_config & + (AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1)) == + (AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1)) + return -ENOENT; + vce_v3_0_set_ring_funcs(adev); vce_v3_0_set_irq_funcs(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index fa5a444..68552da 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -122,6 +122,32 @@ static void vi_smc_wreg(struct amdgpu_device *adev, u32 reg, u32 v) spin_unlock_irqrestore(&adev->smc_idx_lock, flags); } +/* smu_8_0_d.h */ +#define mmMP0PUB_IND_INDEX 0x180 +#define mmMP0PUB_IND_DATA 0x181 + +static u32 cz_smc_rreg(struct amdgpu_device *adev, u32 reg) +{ + unsigned long flags; + u32 r; + + spin_lock_irqsave(&adev->smc_idx_lock, flags); + WREG32(mmMP0PUB_IND_INDEX, (reg)); + r = RREG32(mmMP0PUB_IND_DATA); + spin_unlock_irqrestore(&adev->smc_idx_lock, flags); + return r; +} + +static void cz_smc_wreg(struct amdgpu_device *adev, u32 reg, u32 v) +{ + unsigned long flags; + + spin_lock_irqsave(&adev->smc_idx_lock, flags); + WREG32(mmMP0PUB_IND_INDEX, (reg)); + WREG32(mmMP0PUB_IND_DATA, (v)); + spin_unlock_irqrestore(&adev->smc_idx_lock, flags); +} + static u32 vi_uvd_ctx_rreg(struct amdgpu_device *adev, u32 reg) { unsigned long flags; @@ -1222,8 +1248,13 @@ static int vi_common_early_init(void *handle) bool smc_enabled = false; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - adev->smc_rreg = &vi_smc_rreg; - adev->smc_wreg = &vi_smc_wreg; + if (adev->flags & AMDGPU_IS_APU) { + adev->smc_rreg = &cz_smc_rreg; + adev->smc_wreg = &cz_smc_wreg; + } else { + adev->smc_rreg = &vi_smc_rreg; + adev->smc_wreg = &vi_smc_wreg; + } adev->pcie_rreg = &vi_pcie_rreg; adev->pcie_wreg = &vi_pcie_wreg; adev->uvd_ctx_rreg = &vi_uvd_ctx_rreg; diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c index 42d2ffa..01ffe9b 100644 --- a/drivers/gpu/drm/armada/armada_crtc.c +++ b/drivers/gpu/drm/armada/armada_crtc.c @@ -531,8 +531,6 @@ static int armada_drm_crtc_mode_set(struct drm_crtc *crtc, drm_crtc_vblank_off(crtc); - crtc->mode = *adj; - val = dcrtc->dumb_ctrl & ~CFG_DUMB_ENA; if (val != dcrtc->dumb_ctrl) { dcrtc->dumb_ctrl = val; diff --git a/drivers/gpu/drm/armada/armada_gem.c b/drivers/gpu/drm/armada/armada_gem.c index 580e10a..60a688e 100644 --- a/drivers/gpu/drm/armada/armada_gem.c +++ b/drivers/gpu/drm/armada/armada_gem.c @@ -69,8 +69,9 @@ void armada_gem_free_object(struct drm_gem_object *obj) if (dobj->obj.import_attach) { /* We only ever display imported data */ - dma_buf_unmap_attachment(dobj->obj.import_attach, dobj->sgt, - DMA_TO_DEVICE); + if (dobj->sgt) + dma_buf_unmap_attachment(dobj->obj.import_attach, + dobj->sgt, DMA_TO_DEVICE); drm_prime_gem_destroy(&dobj->obj, NULL); } diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c index c5b06fd..e939fab 100644 --- a/drivers/gpu/drm/armada/armada_overlay.c +++ b/drivers/gpu/drm/armada/armada_overlay.c @@ -7,6 +7,7 @@ * published by the Free Software Foundation. */ #include <drm/drmP.h> +#include <drm/drm_plane_helper.h> #include "armada_crtc.h" #include "armada_drm.h" #include "armada_fb.h" @@ -85,16 +86,8 @@ static void armada_plane_vbl(struct armada_crtc *dcrtc, void *data) if (fb) armada_drm_queue_unref_work(dcrtc->crtc.dev, fb); -} -static unsigned armada_limit(int start, unsigned size, unsigned max) -{ - int end = start + size; - if (end < 0) - return 0; - if (start < 0) - start = 0; - return (unsigned)end > max ? max - start : end - start; + wake_up(&dplane->vbl.wait); } static int @@ -105,26 +98,39 @@ armada_plane_update(struct drm_plane *plane, struct drm_crtc *crtc, { struct armada_plane *dplane = drm_to_armada_plane(plane); struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc); + struct drm_rect src = { + .x1 = src_x, + .y1 = src_y, + .x2 = src_x + src_w, + .y2 = src_y + src_h, + }; + struct drm_rect dest = { + .x1 = crtc_x, + .y1 = crtc_y, + .x2 = crtc_x + crtc_w, + .y2 = crtc_y + crtc_h, + }; + const struct drm_rect clip = { + .x2 = crtc->mode.hdisplay, + .y2 = crtc->mode.vdisplay, + }; uint32_t val, ctrl0; unsigned idx = 0; + bool visible; int ret; - crtc_w = armada_limit(crtc_x, crtc_w, dcrtc->crtc.mode.hdisplay); - crtc_h = armada_limit(crtc_y, crtc_h, dcrtc->crtc.mode.vdisplay); + ret = drm_plane_helper_check_update(plane, crtc, fb, &src, &dest, &clip, + 0, INT_MAX, true, false, &visible); + if (ret) + return ret; + ctrl0 = CFG_DMA_FMT(drm_fb_to_armada_fb(fb)->fmt) | CFG_DMA_MOD(drm_fb_to_armada_fb(fb)->mod) | CFG_CBSH_ENA | CFG_DMA_HSMOOTH | CFG_DMA_ENA; /* Does the position/size result in nothing to display? */ - if (crtc_w == 0 || crtc_h == 0) { + if (!visible) ctrl0 &= ~CFG_DMA_ENA; - } - - /* - * FIXME: if the starting point is off screen, we need to - * adjust src_x, src_y, src_w, src_h appropriately, and - * according to the scale. - */ if (!dcrtc->plane) { dcrtc->plane = plane; @@ -134,15 +140,19 @@ armada_plane_update(struct drm_plane *plane, struct drm_crtc *crtc, /* FIXME: overlay on an interlaced display */ /* Just updating the position/size? */ if (plane->fb == fb && dplane->ctrl0 == ctrl0) { - val = (src_h & 0xffff0000) | src_w >> 16; + val = (drm_rect_height(&src) & 0xffff0000) | + drm_rect_width(&src) >> 16; dplane->src_hw = val; writel_relaxed(val, dcrtc->base + LCD_SPU_DMA_HPXL_VLN); - val = crtc_h << 16 | crtc_w; + + val = drm_rect_height(&dest) << 16 | drm_rect_width(&dest); dplane->dst_hw = val; writel_relaxed(val, dcrtc->base + LCD_SPU_DZM_HPXL_VLN); - val = crtc_y << 16 | crtc_x; + + val = dest.y1 << 16 | dest.x1; dplane->dst_yx = val; writel_relaxed(val, dcrtc->base + LCD_SPU_DMA_OVSA_HPXL_VLN); + return 0; } else if (~dplane->ctrl0 & ctrl0 & CFG_DMA_ENA) { /* Power up the Y/U/V FIFOs on ENA 0->1 transitions */ @@ -150,15 +160,14 @@ armada_plane_update(struct drm_plane *plane, struct drm_crtc *crtc, dcrtc->base + LCD_SPU_SRAM_PARA1); } - ret = wait_event_timeout(dplane->vbl.wait, - list_empty(&dplane->vbl.update.node), - HZ/25); - if (ret < 0) - return ret; + wait_event_timeout(dplane->vbl.wait, + list_empty(&dplane->vbl.update.node), + HZ/25); if (plane->fb != fb) { struct armada_gem_object *obj = drm_fb_obj(fb); - uint32_t sy, su, sv; + uint32_t addr[3], pixel_format; + int i, num_planes, hsub; /* * Take a reference on the new framebuffer - we want to @@ -178,26 +187,39 @@ armada_plane_update(struct drm_plane *plane, struct drm_crtc *crtc, older_fb); } - src_y >>= 16; - src_x >>= 16; - sy = obj->dev_addr + fb->offsets[0] + src_y * fb->pitches[0] + - src_x * fb->bits_per_pixel / 8; - su = obj->dev_addr + fb->offsets[1] + src_y * fb->pitches[1] + - src_x; - sv = obj->dev_addr + fb->offsets[2] + src_y * fb->pitches[2] + - src_x; + src_y = src.y1 >> 16; + src_x = src.x1 >> 16; - armada_reg_queue_set(dplane->vbl.regs, idx, sy, + pixel_format = fb->pixel_format; + hsub = drm_format_horz_chroma_subsampling(pixel_format); + num_planes = drm_format_num_planes(pixel_format); + + /* + * Annoyingly, shifting a YUYV-format image by one pixel + * causes the U/V planes to toggle. Toggle the UV swap. + * (Unfortunately, this causes momentary colour flickering.) + */ + if (src_x & (hsub - 1) && num_planes == 1) + ctrl0 ^= CFG_DMA_MOD(CFG_SWAPUV); + + for (i = 0; i < num_planes; i++) + addr[i] = obj->dev_addr + fb->offsets[i] + + src_y * fb->pitches[i] + + src_x * drm_format_plane_cpp(pixel_format, i); + for (; i < ARRAY_SIZE(addr); i++) + addr[i] = 0; + + armada_reg_queue_set(dplane->vbl.regs, idx, addr[0], LCD_SPU_DMA_START_ADDR_Y0); - armada_reg_queue_set(dplane->vbl.regs, idx, su, + armada_reg_queue_set(dplane->vbl.regs, idx, addr[1], LCD_SPU_DMA_START_ADDR_U0); - armada_reg_queue_set(dplane->vbl.regs, idx, sv, + armada_reg_queue_set(dplane->vbl.regs, idx, addr[2], LCD_SPU_DMA_START_ADDR_V0); - armada_reg_queue_set(dplane->vbl.regs, idx, sy, + armada_reg_queue_set(dplane->vbl.regs, idx, addr[0], LCD_SPU_DMA_START_ADDR_Y1); - armada_reg_queue_set(dplane->vbl.regs, idx, su, + armada_reg_queue_set(dplane->vbl.regs, idx, addr[1], LCD_SPU_DMA_START_ADDR_U1); - armada_reg_queue_set(dplane->vbl.regs, idx, sv, + armada_reg_queue_set(dplane->vbl.regs, idx, addr[2], LCD_SPU_DMA_START_ADDR_V1); val = fb->pitches[0] << 16 | fb->pitches[0]; @@ -208,24 +230,27 @@ armada_plane_update(struct drm_plane *plane, struct drm_crtc *crtc, LCD_SPU_DMA_PITCH_UV); } - val = (src_h & 0xffff0000) | src_w >> 16; + val = (drm_rect_height(&src) & 0xffff0000) | drm_rect_width(&src) >> 16; if (dplane->src_hw != val) { dplane->src_hw = val; armada_reg_queue_set(dplane->vbl.regs, idx, val, LCD_SPU_DMA_HPXL_VLN); } - val = crtc_h << 16 | crtc_w; + + val = drm_rect_height(&dest) << 16 | drm_rect_width(&dest); if (dplane->dst_hw != val) { dplane->dst_hw = val; armada_reg_queue_set(dplane->vbl.regs, idx, val, LCD_SPU_DZM_HPXL_VLN); } - val = crtc_y << 16 | crtc_x; + + val = dest.y1 << 16 | dest.x1; if (dplane->dst_yx != val) { dplane->dst_yx = val; armada_reg_queue_set(dplane->vbl.regs, idx, val, LCD_SPU_DMA_OVSA_HPXL_VLN); } + if (dplane->ctrl0 != ctrl0) { dplane->ctrl0 = ctrl0; armada_reg_queue_mod(dplane->vbl.regs, idx, ctrl0, @@ -279,7 +304,11 @@ static int armada_plane_disable(struct drm_plane *plane) static void armada_plane_destroy(struct drm_plane *plane) { - kfree(plane); + struct armada_plane *dplane = drm_to_armada_plane(plane); + + drm_plane_cleanup(plane); + + kfree(dplane); } static int armada_plane_set_property(struct drm_plane *plane, diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c index f69b925..5ae5c69 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c @@ -355,6 +355,7 @@ int atmel_hlcdc_crtc_create(struct drm_device *dev) planes->overlays[i]->base.possible_crtcs = 1 << crtc->id; drm_crtc_helper_add(&crtc->base, &lcdc_crtc_helper_funcs); + drm_crtc_vblank_reset(&crtc->base); dc->crtc = &crtc->base; diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c index 60b0c13..ef6182b 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c @@ -313,20 +313,20 @@ static int atmel_hlcdc_dc_load(struct drm_device *dev) pm_runtime_enable(dev->dev); - ret = atmel_hlcdc_dc_modeset_init(dev); + ret = drm_vblank_init(dev, 1); if (ret < 0) { - dev_err(dev->dev, "failed to initialize mode setting\n"); + dev_err(dev->dev, "failed to initialize vblank\n"); goto err_periph_clk_disable; } - drm_mode_config_reset(dev); - - ret = drm_vblank_init(dev, 1); + ret = atmel_hlcdc_dc_modeset_init(dev); if (ret < 0) { - dev_err(dev->dev, "failed to initialize vblank\n"); + dev_err(dev->dev, "failed to initialize mode setting\n"); goto err_periph_clk_disable; } + drm_mode_config_reset(dev); + pm_runtime_get_sync(dev->dev); ret = drm_irq_install(dev, dc->hlcdc->irq); pm_runtime_put_sync(dev->dev); @@ -559,7 +559,7 @@ static int atmel_hlcdc_dc_drm_remove(struct platform_device *pdev) return 0; } -#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP static int atmel_hlcdc_dc_drm_suspend(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 5b59d5ad..9dcc728 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -196,7 +196,12 @@ update_connector_routing(struct drm_atomic_state *state, int conn_idx) } funcs = connector->helper_private; - new_encoder = funcs->best_encoder(connector); + + if (funcs->atomic_best_encoder) + new_encoder = funcs->atomic_best_encoder(connector, + connector_state); + else + new_encoder = funcs->best_encoder(connector); if (!new_encoder) { DRM_DEBUG_ATOMIC("No suitable encoder found for [CONNECTOR:%d:%s]\n", @@ -229,6 +234,9 @@ update_connector_routing(struct drm_atomic_state *state, int conn_idx) } } + if (WARN_ON(!connector_state->crtc)) + return -EINVAL; + connector_state->best_encoder = new_encoder; idx = drm_crtc_index(connector_state->crtc); diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index b9ba061..fed7483 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -2706,8 +2706,11 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data, if (!drm_core_check_feature(dev, DRIVER_MODESET)) return -EINVAL; - /* For some reason crtc x/y offsets are signed internally. */ - if (crtc_req->x > INT_MAX || crtc_req->y > INT_MAX) + /* + * Universal plane src offsets are only 16.16, prevent havoc for + * drivers using universal plane code internally. + */ + if (crtc_req->x & 0xffff0000 || crtc_req->y & 0xffff0000) return -ERANGE; drm_modeset_lock_all(dev); @@ -5395,12 +5398,9 @@ void drm_mode_config_reset(struct drm_device *dev) if (encoder->funcs->reset) encoder->funcs->reset(encoder); - list_for_each_entry(connector, &dev->mode_config.connector_list, head) { - connector->status = connector_status_unknown; - + list_for_each_entry(connector, &dev->mode_config.connector_list, head) if (connector->funcs->reset) connector->funcs->reset(connector); - } } EXPORT_SYMBOL(drm_mode_config_reset); diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index 778bbb6..eb603f1de 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -873,9 +873,10 @@ static void drm_dp_destroy_port(struct kref *kref) from an EDID retrieval */ if (port->connector) { mutex_lock(&mgr->destroy_connector_lock); - list_add(&port->connector->destroy_list, &mgr->destroy_connector_list); + list_add(&port->next, &mgr->destroy_connector_list); mutex_unlock(&mgr->destroy_connector_lock); schedule_work(&mgr->destroy_connector_work); + return; } drm_dp_port_teardown_pdt(port, port->pdt); @@ -1294,7 +1295,6 @@ retry: goto retry; } DRM_DEBUG_KMS("failed to dpcd write %d %d\n", tosend, ret); - WARN(1, "fail\n"); return -EIO; } @@ -2660,7 +2660,7 @@ static void drm_dp_tx_work(struct work_struct *work) static void drm_dp_destroy_connector_work(struct work_struct *work) { struct drm_dp_mst_topology_mgr *mgr = container_of(work, struct drm_dp_mst_topology_mgr, destroy_connector_work); - struct drm_connector *connector; + struct drm_dp_mst_port *port; /* * Not a regular list traverse as we have to drop the destroy @@ -2669,15 +2669,21 @@ static void drm_dp_destroy_connector_work(struct work_struct *work) */ for (;;) { mutex_lock(&mgr->destroy_connector_lock); - connector = list_first_entry_or_null(&mgr->destroy_connector_list, struct drm_connector, destroy_list); - if (!connector) { + port = list_first_entry_or_null(&mgr->destroy_connector_list, struct drm_dp_mst_port, next); + if (!port) { mutex_unlock(&mgr->destroy_connector_lock); break; } - list_del(&connector->destroy_list); + list_del(&port->next); mutex_unlock(&mgr->destroy_connector_lock); - mgr->cbs->destroy_connector(mgr, connector); + mgr->cbs->destroy_connector(mgr, port->connector); + + drm_dp_port_teardown_pdt(port, port->pdt); + + if (!port->input && port->vcpi.vcpi > 0) + drm_dp_mst_put_payload_id(mgr, port->vcpi.vcpi); + kfree(port); } } diff --git a/drivers/gpu/drm/drm_ioc32.c b/drivers/gpu/drm/drm_ioc32.c index aa8bbb4..9cfcd0a 100644 --- a/drivers/gpu/drm/drm_ioc32.c +++ b/drivers/gpu/drm/drm_ioc32.c @@ -70,6 +70,8 @@ #define DRM_IOCTL_WAIT_VBLANK32 DRM_IOWR(0x3a, drm_wait_vblank32_t) +#define DRM_IOCTL_MODE_ADDFB232 DRM_IOWR(0xb8, drm_mode_fb_cmd232_t) + typedef struct drm_version_32 { int version_major; /**< Major version */ int version_minor; /**< Minor version */ @@ -1016,6 +1018,63 @@ static int compat_drm_wait_vblank(struct file *file, unsigned int cmd, return 0; } +typedef struct drm_mode_fb_cmd232 { + u32 fb_id; + u32 width; + u32 height; + u32 pixel_format; + u32 flags; + u32 handles[4]; + u32 pitches[4]; + u32 offsets[4]; + u64 modifier[4]; +} __attribute__((packed)) drm_mode_fb_cmd232_t; + +static int compat_drm_mode_addfb2(struct file *file, unsigned int cmd, + unsigned long arg) +{ + struct drm_mode_fb_cmd232 __user *argp = (void __user *)arg; + struct drm_mode_fb_cmd232 req32; + struct drm_mode_fb_cmd2 __user *req64; + int i; + int err; + + if (copy_from_user(&req32, argp, sizeof(req32))) + return -EFAULT; + + req64 = compat_alloc_user_space(sizeof(*req64)); + + if (!access_ok(VERIFY_WRITE, req64, sizeof(*req64)) + || __put_user(req32.width, &req64->width) + || __put_user(req32.height, &req64->height) + || __put_user(req32.pixel_format, &req64->pixel_format) + || __put_user(req32.flags, &req64->flags)) + return -EFAULT; + + for (i = 0; i < 4; i++) { + if (__put_user(req32.handles[i], &req64->handles[i])) + return -EFAULT; + if (__put_user(req32.pitches[i], &req64->pitches[i])) + return -EFAULT; + if (__put_user(req32.offsets[i], &req64->offsets[i])) + return -EFAULT; + if (__put_user(req32.modifier[i], &req64->modifier[i])) + return -EFAULT; + } + + err = drm_ioctl(file, DRM_IOCTL_MODE_ADDFB2, (unsigned long)req64); + if (err) + return err; + + if (__get_user(req32.fb_id, &req64->fb_id)) + return -EFAULT; + + if (copy_to_user(argp, &req32, sizeof(req32))) + return -EFAULT; + + return 0; +} + static drm_ioctl_compat_t *drm_compat_ioctls[] = { [DRM_IOCTL_NR(DRM_IOCTL_VERSION32)] = compat_drm_version, [DRM_IOCTL_NR(DRM_IOCTL_GET_UNIQUE32)] = compat_drm_getunique, @@ -1048,6 +1107,7 @@ static drm_ioctl_compat_t *drm_compat_ioctls[] = { [DRM_IOCTL_NR(DRM_IOCTL_UPDATE_DRAW32)] = compat_drm_update_draw, #endif [DRM_IOCTL_NR(DRM_IOCTL_WAIT_VBLANK32)] = compat_drm_wait_vblank, + [DRM_IOCTL_NR(DRM_IOCTL_MODE_ADDFB232)] = compat_drm_mode_addfb2, }; /** diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c index f9cc68f..b50fa0a 100644 --- a/drivers/gpu/drm/drm_irq.c +++ b/drivers/gpu/drm/drm_irq.c @@ -75,7 +75,7 @@ module_param_named(timestamp_precision_usec, drm_timestamp_precision, int, 0600) module_param_named(timestamp_monotonic, drm_timestamp_monotonic, int, 0600); static void store_vblank(struct drm_device *dev, int crtc, - unsigned vblank_count_inc, + u32 vblank_count_inc, struct timeval *t_vblank) { struct drm_vblank_crtc *vblank = &dev->vblank[crtc]; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index 842d6b8..2a65235 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -1745,7 +1745,6 @@ static int fimc_probe(struct platform_device *pdev) spin_lock_init(&ctx->lock); platform_set_drvdata(pdev, ctx); - pm_runtime_set_active(dev); pm_runtime_enable(dev); ret = exynos_drm_ippdrv_register(ippdrv); diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c index 8040ed2..f1c6b76 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c @@ -593,8 +593,7 @@ static int gsc_src_set_transf(struct device *dev, gsc_write(cfg, GSC_IN_CON); - ctx->rotation = cfg & - (GSC_IN_ROT_90 | GSC_IN_ROT_270) ? 1 : 0; + ctx->rotation = (cfg & GSC_IN_ROT_90) ? 1 : 0; *swap = ctx->rotation; return 0; @@ -857,8 +856,7 @@ static int gsc_dst_set_transf(struct device *dev, gsc_write(cfg, GSC_IN_CON); - ctx->rotation = cfg & - (GSC_IN_ROT_90 | GSC_IN_ROT_270) ? 1 : 0; + ctx->rotation = (cfg & GSC_IN_ROT_90) ? 1 : 0; *swap = ctx->rotation; return 0; diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c index 99e2864..4a00990 100644 --- a/drivers/gpu/drm/exynos/exynos_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_hdmi.c @@ -1064,6 +1064,7 @@ static int hdmi_get_modes(struct drm_connector *connector) { struct hdmi_context *hdata = ctx_from_connector(connector); struct edid *edid; + int ret; if (!hdata->ddc_adpt) return -ENODEV; @@ -1079,7 +1080,11 @@ static int hdmi_get_modes(struct drm_connector *connector) drm_mode_connector_update_edid_property(connector, edid); - return drm_add_edid_modes(connector, edid); + ret = drm_add_edid_modes(connector, edid); + + kfree(edid); + + return ret; } static int hdmi_find_phy_conf(struct hdmi_context *hdata, u32 pixel_clock) diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c index cae98db..4706b56 100644 --- a/drivers/gpu/drm/exynos/exynos_mixer.c +++ b/drivers/gpu/drm/exynos/exynos_mixer.c @@ -718,6 +718,10 @@ static irqreturn_t mixer_irq_handler(int irq, void *arg) /* handling VSYNC */ if (val & MXR_INT_STATUS_VSYNC) { + /* vsync interrupt use different bit for read and clear */ + val |= MXR_INT_CLEAR_VSYNC; + val &= ~MXR_INT_STATUS_VSYNC; + /* interlace scan need to check shadow register */ if (ctx->interlace) { base = mixer_reg_read(res, MXR_GRAPHIC_BASE(0)); @@ -743,11 +747,6 @@ static irqreturn_t mixer_irq_handler(int irq, void *arg) out: /* clear interrupts */ - if (~val & MXR_INT_EN_VSYNC) { - /* vsync interrupt use different bit for read and clear */ - val &= ~MXR_INT_EN_VSYNC; - val |= MXR_INT_CLEAR_VSYNC; - } mixer_reg_write(res, MXR_INT_STATUS, val); spin_unlock(&res->reg_slock); @@ -907,8 +906,8 @@ static int mixer_enable_vblank(struct exynos_drm_crtc *crtc) } /* enable vsync interrupt */ - mixer_reg_writemask(res, MXR_INT_EN, MXR_INT_EN_VSYNC, - MXR_INT_EN_VSYNC); + mixer_reg_writemask(res, MXR_INT_STATUS, ~0, MXR_INT_CLEAR_VSYNC); + mixer_reg_writemask(res, MXR_INT_EN, ~0, MXR_INT_EN_VSYNC); return 0; } @@ -918,7 +917,13 @@ static void mixer_disable_vblank(struct exynos_drm_crtc *crtc) struct mixer_context *mixer_ctx = crtc->ctx; struct mixer_resources *res = &mixer_ctx->mixer_res; + if (!mixer_ctx->powered) { + mixer_ctx->int_en &= MXR_INT_EN_VSYNC; + return; + } + /* disable vsync interrupt */ + mixer_reg_writemask(res, MXR_INT_STATUS, ~0, MXR_INT_CLEAR_VSYNC); mixer_reg_writemask(res, MXR_INT_EN, 0, MXR_INT_EN_VSYNC); } @@ -1047,6 +1052,8 @@ static void mixer_enable(struct exynos_drm_crtc *crtc) mixer_reg_writemask(res, MXR_STATUS, ~0, MXR_STATUS_SOFT_RESET); + if (ctx->int_en & MXR_INT_EN_VSYNC) + mixer_reg_writemask(res, MXR_INT_STATUS, ~0, MXR_INT_CLEAR_VSYNC); mixer_reg_write(res, MXR_INT_EN, ctx->int_en); mixer_win_reset(ctx); } diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c index fe1599d..424228b 100644 --- a/drivers/gpu/drm/i2c/tda998x_drv.c +++ b/drivers/gpu/drm/i2c/tda998x_drv.c @@ -606,8 +606,6 @@ static void tda998x_write_if(struct tda998x_priv *priv, uint8_t bit, uint16_t addr, uint8_t *buf, size_t size) { - buf[PB(0)] = tda998x_cksum(buf, size); - reg_clear(priv, REG_DIP_IF_FLAGS, bit); reg_write_range(priv, addr, buf, size); reg_set(priv, REG_DIP_IF_FLAGS, bit); @@ -627,6 +625,8 @@ tda998x_write_aif(struct tda998x_priv *priv, struct tda998x_encoder_params *p) buf[PB(4)] = p->audio_frame[4]; buf[PB(5)] = p->audio_frame[5] & 0xf8; /* DM_INH + LSV */ + buf[PB(0)] = tda998x_cksum(buf, sizeof(buf)); + tda998x_write_if(priv, DIP_IF_FLAGS_IF4, REG_IF4_HB0, buf, sizeof(buf)); } diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 542fac6..fd1de45 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -826,6 +826,7 @@ struct intel_context { struct kref ref; int user_handle; uint8_t remap_slice; + struct drm_i915_private *i915; struct drm_i915_file_private *file_priv; struct i915_ctx_hang_stats hang_stats; struct i915_hw_ppgtt *ppgtt; @@ -2036,8 +2037,6 @@ struct drm_i915_gem_object { unsigned int cache_level:3; unsigned int cache_dirty:1; - unsigned int has_dma_mapping:1; - unsigned int frontbuffer_bits:INTEL_FRONTBUFFER_BITS; unsigned int pin_display; @@ -3116,7 +3115,8 @@ void i915_debugfs_cleanup(struct drm_minor *minor); int i915_debugfs_connector_add(struct drm_connector *connector); void intel_display_crc_init(struct drm_device *dev); #else -static inline int i915_debugfs_connector_add(struct drm_connector *connector) {} +static inline int i915_debugfs_connector_add(struct drm_connector *connector) +{ return 0; } static inline void intel_display_crc_init(struct drm_device *dev) {} #endif @@ -3303,15 +3303,14 @@ int intel_freq_opcode(struct drm_i915_private *dev_priv, int val); #define I915_READ64(reg) dev_priv->uncore.funcs.mmio_readq(dev_priv, (reg), true) #define I915_READ64_2x32(lower_reg, upper_reg) ({ \ - u32 upper = I915_READ(upper_reg); \ - u32 lower = I915_READ(lower_reg); \ - u32 tmp = I915_READ(upper_reg); \ - if (upper != tmp) { \ - upper = tmp; \ - lower = I915_READ(lower_reg); \ - WARN_ON(I915_READ(upper_reg) != upper); \ - } \ - (u64)upper << 32 | lower; }) + u32 upper, lower, tmp; \ + tmp = I915_READ(upper_reg); \ + do { \ + upper = tmp; \ + lower = I915_READ(lower_reg); \ + tmp = I915_READ(upper_reg); \ + } while (upper != tmp); \ + (u64)upper << 32 | lower; }) #define POSTING_READ(reg) (void)I915_READ_NOTRACE(reg) #define POSTING_READ16(reg) (void)I915_READ16_NOTRACE(reg) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 248fd1a..52b446b 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -213,7 +213,6 @@ i915_gem_object_get_pages_phys(struct drm_i915_gem_object *obj) sg_dma_len(sg) = obj->base.size; obj->pages = st; - obj->has_dma_mapping = true; return 0; } @@ -265,8 +264,6 @@ i915_gem_object_put_pages_phys(struct drm_i915_gem_object *obj) sg_free_table(obj->pages); kfree(obj->pages); - - obj->has_dma_mapping = false; } static void @@ -2139,6 +2136,8 @@ i915_gem_object_put_pages_gtt(struct drm_i915_gem_object *obj) obj->base.read_domains = obj->base.write_domain = I915_GEM_DOMAIN_CPU; } + i915_gem_gtt_finish_object(obj); + if (i915_gem_object_needs_bit17_swizzle(obj)) i915_gem_object_save_bit_17_swizzle(obj); @@ -2199,6 +2198,7 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj) struct sg_page_iter sg_iter; struct page *page; unsigned long last_pfn = 0; /* suppress gcc warning */ + int ret; gfp_t gfp; /* Assert that the object is not currently in any GPU domain. As it @@ -2246,8 +2246,10 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj) */ i915_gem_shrink_all(dev_priv); page = shmem_read_mapping_page(mapping, i); - if (IS_ERR(page)) + if (IS_ERR(page)) { + ret = PTR_ERR(page); goto err_pages; + } } #ifdef CONFIG_SWIOTLB if (swiotlb_nr_tbl()) { @@ -2276,6 +2278,10 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj) sg_mark_end(sg); obj->pages = st; + ret = i915_gem_gtt_prepare_object(obj); + if (ret) + goto err_pages; + if (i915_gem_object_needs_bit17_swizzle(obj)) i915_gem_object_do_bit_17_swizzle(obj); @@ -2300,10 +2306,10 @@ err_pages: * space and so want to translate the error from shmemfs back to our * usual understanding of ENOMEM. */ - if (PTR_ERR(page) == -ENOSPC) - return -ENOMEM; - else - return PTR_ERR(page); + if (ret == -ENOSPC) + ret = -ENOMEM; + + return ret; } /* Ensure that the associated pages are gathered from the backing storage @@ -2542,6 +2548,7 @@ int __i915_add_request(struct intel_engine_cs *ring, } request->emitted_jiffies = jiffies; + ring->last_submitted_seqno = request->seqno; list_add_tail(&request->list, &ring->request_list); request->file_priv = NULL; @@ -3247,10 +3254,8 @@ int i915_vma_unbind(struct i915_vma *vma) /* Since the unbound list is global, only move to that list if * no more VMAs exist. */ - if (list_empty(&obj->vma_list)) { - i915_gem_gtt_finish_object(obj); + if (list_empty(&obj->vma_list)) list_move_tail(&obj->global_list, &dev_priv->mm.unbound_list); - } /* And finally now the object is completely decoupled from this vma, * we can drop its hold on the backing storage and allow it to be @@ -3768,22 +3773,16 @@ search_free: goto err_remove_node; } - ret = i915_gem_gtt_prepare_object(obj); - if (ret) - goto err_remove_node; - trace_i915_vma_bind(vma, flags); ret = i915_vma_bind(vma, obj->cache_level, flags); if (ret) - goto err_finish_gtt; + goto err_remove_node; list_move_tail(&obj->global_list, &dev_priv->mm.bound_list); list_add_tail(&vma->mm_list, &vm->inactive_list); return vma; -err_finish_gtt: - i915_gem_gtt_finish_object(obj); err_remove_node: drm_mm_remove_node(&vma->node); err_free_vma: diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index d65cbe6..48afa77 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -135,8 +135,7 @@ static int get_context_size(struct drm_device *dev) void i915_gem_context_free(struct kref *ctx_ref) { - struct intel_context *ctx = container_of(ctx_ref, - typeof(*ctx), ref); + struct intel_context *ctx = container_of(ctx_ref, typeof(*ctx), ref); trace_i915_context_free(ctx); @@ -195,6 +194,7 @@ __create_hw_context(struct drm_device *dev, kref_init(&ctx->ref); list_add_tail(&ctx->link, &dev_priv->context_list); + ctx->i915 = dev_priv; if (dev_priv->hw_context_size) { struct drm_i915_gem_object *obj = diff --git a/drivers/gpu/drm/i915/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/i915_gem_dmabuf.c index 7998da2..e9c2bfd 100644 --- a/drivers/gpu/drm/i915/i915_gem_dmabuf.c +++ b/drivers/gpu/drm/i915/i915_gem_dmabuf.c @@ -256,7 +256,6 @@ static int i915_gem_object_get_pages_dmabuf(struct drm_i915_gem_object *obj) return PTR_ERR(sg); obj->pages = sg; - obj->has_dma_mapping = true; return 0; } @@ -264,7 +263,6 @@ static void i915_gem_object_put_pages_dmabuf(struct drm_i915_gem_object *obj) { dma_buf_unmap_attachment(obj->base.import_attach, obj->pages, DMA_BIDIRECTIONAL); - obj->has_dma_mapping = false; } static const struct drm_i915_gem_object_ops i915_gem_object_dmabuf_ops = { diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index dcc6a88..31e8269 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -1723,9 +1723,6 @@ void i915_gem_suspend_gtt_mappings(struct drm_device *dev) int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj) { - if (obj->has_dma_mapping) - return 0; - if (!dma_map_sg(&obj->base.dev->pdev->dev, obj->pages->sgl, obj->pages->nents, PCI_DMA_BIDIRECTIONAL)) @@ -1926,6 +1923,17 @@ static int ggtt_bind_vma(struct i915_vma *vma, vma->vm->insert_entries(vma->vm, pages, vma->node.start, cache_level, pte_flags); + + /* Note the inconsistency here is due to absence of the + * aliasing ppgtt on gen4 and earlier. Though we always + * request PIN_USER for execbuffer (translated to LOCAL_BIND), + * without the appgtt, we cannot honour that request and so + * must substitute it with a global binding. Since we do this + * behind the upper layers back, we need to explicitly set + * the bound flag ourselves. + */ + vma->bound |= GLOBAL_BIND; + } if (dev_priv->mm.aliasing_ppgtt && flags & LOCAL_BIND) { @@ -1972,10 +1980,8 @@ void i915_gem_gtt_finish_object(struct drm_i915_gem_object *obj) interruptible = do_idling(dev_priv); - if (!obj->has_dma_mapping) - dma_unmap_sg(&dev->pdev->dev, - obj->pages->sgl, obj->pages->nents, - PCI_DMA_BIDIRECTIONAL); + dma_unmap_sg(&dev->pdev->dev, obj->pages->sgl, obj->pages->nents, + PCI_DMA_BIDIRECTIONAL); undo_idling(dev_priv, interruptible); } diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c index 348ed5a..8b5b784 100644 --- a/drivers/gpu/drm/i915/i915_gem_stolen.c +++ b/drivers/gpu/drm/i915/i915_gem_stolen.c @@ -416,7 +416,6 @@ _i915_gem_object_create_stolen(struct drm_device *dev, if (obj->pages == NULL) goto cleanup; - obj->has_dma_mapping = true; i915_gem_object_pin_pages(obj); obj->stolen = stolen; diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c index d61e74a..d19c9db 100644 --- a/drivers/gpu/drm/i915/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c @@ -183,18 +183,8 @@ i915_gem_detect_bit_6_swizzle(struct drm_device *dev) if (IS_GEN4(dev)) { uint32_t ddc2 = I915_READ(DCC2); - if (!(ddc2 & DCC2_MODIFIED_ENHANCED_DISABLE)) { - /* Since the swizzling may vary within an - * object, we have no idea what the swizzling - * is for any page in particular. Thus we - * cannot migrate tiled pages using the GPU, - * nor can we tell userspace what the exact - * swizzling is for any object. - */ + if (!(ddc2 & DCC2_MODIFIED_ENHANCED_DISABLE)) dev_priv->quirks |= QUIRK_PIN_SWIZZLED_PAGES; - swizzle_x = I915_BIT_6_SWIZZLE_UNKNOWN; - swizzle_y = I915_BIT_6_SWIZZLE_UNKNOWN; - } } if (dcc == 0xffffffff) { @@ -474,7 +464,10 @@ i915_gem_get_tiling(struct drm_device *dev, void *data, } /* Hide bit 17 from the user -- see comment in i915_gem_set_tiling */ - args->phys_swizzle_mode = args->swizzle_mode; + if (dev_priv->quirks & QUIRK_PIN_SWIZZLED_PAGES) + args->phys_swizzle_mode = I915_BIT_6_SWIZZLE_UNKNOWN; + else + args->phys_swizzle_mode = args->swizzle_mode; if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_17) args->swizzle_mode = I915_BIT_6_SWIZZLE_9; if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_10_17) diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c index 1f4e5a3..8fd431b 100644 --- a/drivers/gpu/drm/i915/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/i915_gem_userptr.c @@ -545,6 +545,26 @@ err: return ret; } +static int +__i915_gem_userptr_set_pages(struct drm_i915_gem_object *obj, + struct page **pvec, int num_pages) +{ + int ret; + + ret = st_set_pages(&obj->pages, pvec, num_pages); + if (ret) + return ret; + + ret = i915_gem_gtt_prepare_object(obj); + if (ret) { + sg_free_table(obj->pages); + kfree(obj->pages); + obj->pages = NULL; + } + + return ret; +} + static void __i915_gem_userptr_get_pages_worker(struct work_struct *_work) { @@ -584,9 +604,12 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work) if (obj->userptr.work != &work->work) { ret = 0; } else if (pinned == num_pages) { - ret = st_set_pages(&obj->pages, pvec, num_pages); + ret = __i915_gem_userptr_set_pages(obj, pvec, num_pages); if (ret == 0) { list_add_tail(&obj->global_list, &to_i915(dev)->mm.unbound_list); + obj->get_page.sg = obj->pages->sgl; + obj->get_page.last = 0; + pinned = 0; } } @@ -693,7 +716,7 @@ i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj) } } } else { - ret = st_set_pages(&obj->pages, pvec, num_pages); + ret = __i915_gem_userptr_set_pages(obj, pvec, num_pages); if (ret == 0) { obj->userptr.work = NULL; pinned = 0; @@ -715,6 +738,8 @@ i915_gem_userptr_put_pages(struct drm_i915_gem_object *obj) if (obj->madv != I915_MADV_WILLNEED) obj->dirty = 0; + i915_gem_gtt_finish_object(obj); + for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0) { struct page *page = sg_page_iter_page(&sg_iter); diff --git a/drivers/gpu/drm/i915/i915_ioc32.c b/drivers/gpu/drm/i915/i915_ioc32.c index 176de63..23aa04c 100644 --- a/drivers/gpu/drm/i915/i915_ioc32.c +++ b/drivers/gpu/drm/i915/i915_ioc32.c @@ -204,7 +204,7 @@ long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) drm_ioctl_compat_t *fn = NULL; int ret; - if (nr < DRM_COMMAND_BASE) + if (nr < DRM_COMMAND_BASE || nr >= DRM_COMMAND_END) return drm_compat_ioctl(filp, cmd, arg); if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(i915_compat_ioctls)) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index e6bb72d..984e2fe 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -2706,18 +2706,11 @@ static void gen8_disable_vblank(struct drm_device *dev, int pipe) spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags); } -static struct drm_i915_gem_request * -ring_last_request(struct intel_engine_cs *ring) -{ - return list_entry(ring->request_list.prev, - struct drm_i915_gem_request, list); -} - static bool -ring_idle(struct intel_engine_cs *ring) +ring_idle(struct intel_engine_cs *ring, u32 seqno) { return (list_empty(&ring->request_list) || - i915_gem_request_completed(ring_last_request(ring), false)); + i915_seqno_passed(seqno, ring->last_submitted_seqno)); } static bool @@ -2939,7 +2932,7 @@ static void i915_hangcheck_elapsed(struct work_struct *work) acthd = intel_ring_get_active_head(ring); if (ring->hangcheck.seqno == seqno) { - if (ring_idle(ring)) { + if (ring_idle(ring, seqno)) { ring->hangcheck.action = HANGCHECK_IDLE; if (waitqueue_active(&ring->irq_queue)) { diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h index 497cba5..849a259 100644 --- a/drivers/gpu/drm/i915/i915_trace.h +++ b/drivers/gpu/drm/i915/i915_trace.h @@ -727,7 +727,7 @@ DECLARE_EVENT_CLASS(i915_context, TP_fast_assign( __entry->ctx = ctx; __entry->vm = ctx->ppgtt ? &ctx->ppgtt->base : NULL; - __entry->dev = ctx->file_priv->dev_priv->dev->primary->index; + __entry->dev = ctx->i915->dev->primary->index; ), TP_printk("dev=%u, ctx=%p, ctx_vm=%p", diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c index 7ed8033..8e35e0d 100644 --- a/drivers/gpu/drm/i915/intel_atomic.c +++ b/drivers/gpu/drm/i915/intel_atomic.c @@ -129,8 +129,9 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *state, bool async) { - int ret; - int i; + struct drm_crtc_state *crtc_state; + struct drm_crtc *crtc; + int ret, i; if (async) { DRM_DEBUG_KMS("i915 does not yet support async commit\n"); @@ -142,48 +143,18 @@ int intel_atomic_commit(struct drm_device *dev, return ret; /* Point of no return */ - - /* - * FIXME: The proper sequence here will eventually be: - * - * drm_atomic_helper_swap_state(dev, state) - * drm_atomic_helper_commit_modeset_disables(dev, state); - * drm_atomic_helper_commit_planes(dev, state); - * drm_atomic_helper_commit_modeset_enables(dev, state); - * drm_atomic_helper_wait_for_vblanks(dev, state); - * drm_atomic_helper_cleanup_planes(dev, state); - * drm_atomic_state_free(state); - * - * once we have full atomic modeset. For now, just manually update - * plane states to avoid clobbering good states with dummy states - * while nuclear pageflipping. - */ - for (i = 0; i < dev->mode_config.num_total_plane; i++) { - struct drm_plane *plane = state->planes[i]; - - if (!plane) - continue; - - plane->state->state = state; - swap(state->plane_states[i], plane->state); - plane->state->state = NULL; - } + drm_atomic_helper_swap_state(dev, state); /* swap crtc_scaler_state */ - for (i = 0; i < dev->mode_config.num_crtc; i++) { - struct drm_crtc *crtc = state->crtcs[i]; - if (!crtc) { - continue; - } - - to_intel_crtc(crtc)->config->scaler_state = - to_intel_crtc_state(state->crtc_states[i])->scaler_state; + for_each_crtc_in_state(state, crtc, crtc_state, i) { + to_intel_crtc(crtc)->config = to_intel_crtc_state(crtc->state); if (INTEL_INFO(dev)->gen >= 9) skl_detach_scalers(to_intel_crtc(crtc)); + + drm_atomic_helper_commit_planes_on_crtc(crtc_state); } - drm_atomic_helper_commit_planes(dev, state); drm_atomic_helper_wait_for_vblanks(dev, state); drm_atomic_helper_cleanup_planes(dev, state); drm_atomic_state_free(state); diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 647b140..87476ff 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -6315,9 +6315,6 @@ static void intel_crtc_disable(struct drm_crtc *crtc) struct drm_connector *connector; struct drm_i915_private *dev_priv = dev->dev_private; - /* crtc should still be enabled when we disable it. */ - WARN_ON(!crtc->state->enable); - intel_crtc_disable_planes(crtc); dev_priv->display.crtc_disable(crtc); dev_priv->display.off(crtc); @@ -11829,7 +11826,9 @@ encoder_retry: goto encoder_retry; } - pipe_config->dither = pipe_config->pipe_bpp != base_bpp; + /* Dithering seems to not pass-through bits correctly when it should, so + * only enable it on 6bpc panels. */ + pipe_config->dither = pipe_config->pipe_bpp == 6*3; DRM_DEBUG_KMS("plane bpp: %i, pipe bpp: %i, dithering: %i\n", base_bpp, pipe_config->pipe_bpp, pipe_config->dither); @@ -12591,7 +12590,8 @@ static int __intel_set_mode(struct drm_crtc *modeset_crtc, continue; if (!crtc_state->enable) { - intel_crtc_disable(crtc); + if (crtc->state->enable) + intel_crtc_disable(crtc); } else if (crtc->state->enable) { intel_crtc_disable_planes(crtc); dev_priv->display.crtc_disable(crtc); @@ -12626,17 +12626,17 @@ static int __intel_set_mode(struct drm_crtc *modeset_crtc, modeset_update_crtc_power_domains(state); - drm_atomic_helper_commit_planes(dev, state); - /* Now enable the clocks, plane, pipe, and connectors that we set up. */ for_each_crtc_in_state(state, crtc, crtc_state, i) { - if (!needs_modeset(crtc->state) || !crtc->state->enable) + if (!needs_modeset(crtc->state) || !crtc->state->enable) { + drm_atomic_helper_commit_planes_on_crtc(crtc_state); continue; + } update_scanline_offset(to_intel_crtc(crtc)); dev_priv->display.crtc_enable(crtc); - intel_crtc_enable_planes(crtc); + drm_atomic_helper_commit_planes_on_crtc(crtc_state); } /* FIXME: add subpixel order */ @@ -12893,20 +12893,11 @@ intel_modeset_stage_output_state(struct drm_device *dev, return 0; } -static bool primary_plane_visible(struct drm_crtc *crtc) -{ - struct intel_plane_state *plane_state = - to_intel_plane_state(crtc->primary->state); - - return plane_state->visible; -} - static int intel_crtc_set_config(struct drm_mode_set *set) { struct drm_device *dev; struct drm_atomic_state *state = NULL; struct intel_crtc_state *pipe_config; - bool primary_plane_was_visible; int ret; BUG_ON(!set); @@ -12945,38 +12936,8 @@ static int intel_crtc_set_config(struct drm_mode_set *set) intel_update_pipe_size(to_intel_crtc(set->crtc)); - primary_plane_was_visible = primary_plane_visible(set->crtc); - ret = intel_set_mode_with_config(set->crtc, pipe_config, true); - if (ret == 0 && - pipe_config->base.enable && - pipe_config->base.planes_changed && - !needs_modeset(&pipe_config->base)) { - struct intel_crtc *intel_crtc = to_intel_crtc(set->crtc); - - /* - * We need to make sure the primary plane is re-enabled if it - * has previously been turned off. - */ - if (ret == 0 && !primary_plane_was_visible && - primary_plane_visible(set->crtc)) { - WARN_ON(!intel_crtc->active); - intel_post_enable_primary(set->crtc); - } - - /* - * In the fastboot case this may be our only check of the - * state after boot. It would be better to only do it on - * the first update, but we don't have a nice way of doing that - * (and really, set_config isn't used much for high freq page - * flipping, so increasing its cost here shouldn't be a big - * deal). - */ - if (i915.fastboot && ret == 0) - intel_modeset_check_state(set->crtc->dev); - } - if (ret) { DRM_DEBUG_KMS("failed to set mode on [CRTC:%d], err = %d\n", set->crtc->base.id, ret); @@ -13276,7 +13237,7 @@ intel_check_primary_plane(struct drm_plane *plane, if (ret) return ret; - if (intel_crtc->active) { + if (crtc_state ? crtc_state->base.active : intel_crtc->active) { struct intel_plane_state *old_state = to_intel_plane_state(plane->state); @@ -13307,6 +13268,9 @@ intel_check_primary_plane(struct drm_plane *plane, */ if (IS_BROADWELL(dev)) intel_crtc->atomic.wait_vblank = true; + + if (crtc_state) + intel_crtc->atomic.post_enable_primary = true; } /* @@ -13319,6 +13283,10 @@ intel_check_primary_plane(struct drm_plane *plane, if (!state->visible || !fb) intel_crtc->atomic.disable_ips = true; + if (!state->visible && old_state->visible && + crtc_state && !needs_modeset(&crtc_state->base)) + intel_crtc->atomic.pre_disable_primary = true; + intel_crtc->atomic.fb_bits |= INTEL_FRONTBUFFER_PRIMARY(intel_crtc->pipe); @@ -15036,6 +15004,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev) struct intel_plane_state *plane_state; memset(crtc->config, 0, sizeof(*crtc->config)); + crtc->config->base.crtc = &crtc->base; crtc->config->quirks |= PIPE_CONFIG_QUIRK_INHERITED_MODE; diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 6e8faa2..1df0e1f 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -93,9 +93,6 @@ static const struct dp_link_dpll chv_dpll[] = { static const int skl_rates[] = { 162000, 216000, 270000, 324000, 432000, 540000 }; -static const int chv_rates[] = { 162000, 202500, 210000, 216000, - 243000, 270000, 324000, 405000, - 420000, 432000, 540000 }; static const int default_rates[] = { 162000, 270000, 540000 }; /** @@ -1169,24 +1166,31 @@ intel_dp_sink_rates(struct intel_dp *intel_dp, const int **sink_rates) return (intel_dp_max_link_bw(intel_dp) >> 3) + 1; } +static bool intel_dp_source_supports_hbr2(struct drm_device *dev) +{ + /* WaDisableHBR2:skl */ + if (IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_B0) + return false; + + if ((IS_HASWELL(dev) && !IS_HSW_ULX(dev)) || IS_BROADWELL(dev) || + (INTEL_INFO(dev)->gen >= 9)) + return true; + else + return false; +} + static int intel_dp_source_rates(struct drm_device *dev, const int **source_rates) { if (IS_SKYLAKE(dev)) { *source_rates = skl_rates; return ARRAY_SIZE(skl_rates); - } else if (IS_CHERRYVIEW(dev)) { - *source_rates = chv_rates; - return ARRAY_SIZE(chv_rates); } *source_rates = default_rates; - if (IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_B0) - /* WaDisableHBR2:skl */ - return (DP_LINK_BW_2_7 >> 3) + 1; - else if (INTEL_INFO(dev)->gen >= 8 || - (IS_HASWELL(dev) && !IS_HSW_ULX(dev))) + /* This depends on the fact that 5.4 is last value in the array */ + if (intel_dp_source_supports_hbr2(dev)) return (DP_LINK_BW_5_4 >> 3) + 1; else return (DP_LINK_BW_2_7 >> 3) + 1; @@ -3941,10 +3945,15 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp) } } - /* Training Pattern 3 support, both source and sink */ + /* Training Pattern 3 support, Intel platforms that support HBR2 alone + * have support for TP3 hence that check is used along with dpcd check + * to ensure TP3 can be enabled. + * SKL < B0: due it's WaDisableHBR2 is the only exception where TP3 is + * supported but still not enabled. + */ if (intel_dp->dpcd[DP_DPCD_REV] >= 0x12 && intel_dp->dpcd[DP_MAX_LANE_COUNT] & DP_TPS3_SUPPORTED && - (IS_HASWELL(dev_priv) || INTEL_INFO(dev_priv)->gen >= 8)) { + intel_dp_source_supports_hbr2(dev)) { intel_dp->use_tps3 = true; DRM_DEBUG_KMS("Displayport TPS3 supported\n"); } else diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c index 6e4cc53..600afdb 100644 --- a/drivers/gpu/drm/i915/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/intel_dp_mst.c @@ -357,6 +357,16 @@ intel_dp_mst_mode_valid(struct drm_connector *connector, return MODE_OK; } +static struct drm_encoder *intel_mst_atomic_best_encoder(struct drm_connector *connector, + struct drm_connector_state *state) +{ + struct intel_connector *intel_connector = to_intel_connector(connector); + struct intel_dp *intel_dp = intel_connector->mst_port; + struct intel_crtc *crtc = to_intel_crtc(state->crtc); + + return &intel_dp->mst_encoders[crtc->pipe]->base.base; +} + static struct drm_encoder *intel_mst_best_encoder(struct drm_connector *connector) { struct intel_connector *intel_connector = to_intel_connector(connector); @@ -367,6 +377,7 @@ static struct drm_encoder *intel_mst_best_encoder(struct drm_connector *connecto static const struct drm_connector_helper_funcs intel_dp_mst_connector_helper_funcs = { .get_modes = intel_dp_mst_get_modes, .mode_valid = intel_dp_mst_mode_valid, + .atomic_best_encoder = intel_mst_atomic_best_encoder, .best_encoder = intel_mst_best_encoder, }; diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 9b74ffa..7f2161a 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -1012,6 +1012,8 @@ static int intel_lr_context_pin(struct intel_engine_cs *ring, ret = intel_pin_and_map_ringbuffer_obj(ring->dev, ringbuf); if (ret) goto unpin_ctx_obj; + + ctx_obj->dirty = true; } return ret; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index e539314..4be66f6 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -275,6 +275,13 @@ struct intel_engine_cs { * Do we have some not yet emitted requests outstanding? */ struct drm_i915_gem_request *outstanding_lazy_request; + /** + * Seqno of request most recently submitted to request_list. + * Used exclusively by hang checker to avoid grabbing lock while + * inspecting request list. + */ + u32 last_submitted_seqno; + bool gpu_caches_dirty; wait_queue_head_t irq_queue; diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index a6d8a3e..260389a 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -1274,10 +1274,12 @@ int i915_reg_read_ioctl(struct drm_device *dev, struct drm_i915_private *dev_priv = dev->dev_private; struct drm_i915_reg_read *reg = data; struct register_whitelist const *entry = whitelist; + unsigned size; + u64 offset; int i, ret = 0; for (i = 0; i < ARRAY_SIZE(whitelist); i++, entry++) { - if (entry->offset == reg->offset && + if (entry->offset == (reg->offset & -entry->size) && (1 << INTEL_INFO(dev)->gen & entry->gen_bitmask)) break; } @@ -1285,23 +1287,33 @@ int i915_reg_read_ioctl(struct drm_device *dev, if (i == ARRAY_SIZE(whitelist)) return -EINVAL; + /* We use the low bits to encode extra flags as the register should + * be naturally aligned (and those that are not so aligned merely + * limit the available flags for that register). + */ + offset = entry->offset; + size = entry->size; + size |= reg->offset ^ offset; + intel_runtime_pm_get(dev_priv); - switch (entry->size) { + switch (size) { + case 8 | 1: + reg->val = I915_READ64_2x32(offset, offset+4); + break; case 8: - reg->val = I915_READ64(reg->offset); + reg->val = I915_READ64(offset); break; case 4: - reg->val = I915_READ(reg->offset); + reg->val = I915_READ(offset); break; case 2: - reg->val = I915_READ16(reg->offset); + reg->val = I915_READ16(offset); break; case 1: - reg->val = I915_READ8(reg->offset); + reg->val = I915_READ8(offset); break; default: - MISSING_CASE(entry->size); ret = -EINVAL; goto out; } diff --git a/drivers/gpu/drm/imx/imx-tve.c b/drivers/gpu/drm/imx/imx-tve.c index 214ecee..e671ad3 100644 --- a/drivers/gpu/drm/imx/imx-tve.c +++ b/drivers/gpu/drm/imx/imx-tve.c @@ -301,7 +301,7 @@ static void imx_tve_encoder_prepare(struct drm_encoder *encoder) switch (tve->mode) { case TVE_MODE_VGA: - imx_drm_set_bus_format_pins(encoder, MEDIA_BUS_FMT_YUV8_1X24, + imx_drm_set_bus_format_pins(encoder, MEDIA_BUS_FMT_GBR888_1X24, tve->hsync_pin, tve->vsync_pin); break; case TVE_MODE_TVOUT: diff --git a/drivers/gpu/drm/imx/parallel-display.c b/drivers/gpu/drm/imx/parallel-display.c index 74a9ce4..b4deb9c 100644 --- a/drivers/gpu/drm/imx/parallel-display.c +++ b/drivers/gpu/drm/imx/parallel-display.c @@ -21,6 +21,7 @@ #include <drm/drm_panel.h> #include <linux/videodev2.h> #include <video/of_display_timing.h> +#include <linux/of_graph.h> #include "imx-drm.h" @@ -208,7 +209,7 @@ static int imx_pd_bind(struct device *dev, struct device *master, void *data) { struct drm_device *drm = data; struct device_node *np = dev->of_node; - struct device_node *panel_node; + struct device_node *port; const u8 *edidp; struct imx_parallel_display *imxpd; int ret; @@ -234,11 +235,19 @@ static int imx_pd_bind(struct device *dev, struct device *master, void *data) imxpd->bus_format = MEDIA_BUS_FMT_RGB666_1X24_CPADHI; } - panel_node = of_parse_phandle(np, "fsl,panel", 0); - if (panel_node) { - imxpd->panel = of_drm_find_panel(panel_node); - if (!imxpd->panel) - return -EPROBE_DEFER; + /* port@1 is the output port */ + port = of_graph_get_port_by_id(np, 1); + if (port) { + struct device_node *endpoint, *remote; + + endpoint = of_get_child_by_name(port, "endpoint"); + if (endpoint) { + remote = of_graph_get_remote_port_parent(endpoint); + if (remote) + imxpd->panel = of_drm_find_panel(remote); + if (!imxpd->panel) + return -EPROBE_DEFER; + } } imxpd->dev = dev; diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c index 0d1dbb7..247a424 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c @@ -220,13 +220,15 @@ static int mdp4_plane_mode_set(struct drm_plane *plane, uint32_t op_mode = 0; uint32_t phasex_step = MDP4_VG_PHASE_STEP_DEFAULT; uint32_t phasey_step = MDP4_VG_PHASE_STEP_DEFAULT; - enum mdp4_frame_format frame_type = mdp4_get_frame_format(fb); + enum mdp4_frame_format frame_type; if (!(crtc && fb)) { DBG("%s: disabled!", mdp4_plane->name); return 0; } + frame_type = mdp4_get_frame_format(fb); + /* src values are in Q16 fixed point, convert to integer: */ src_x = src_x >> 16; src_y = src_y >> 16; diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c index 206f758..e253db5 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c @@ -76,7 +76,20 @@ static void mdp5_prepare_commit(struct msm_kms *kms, struct drm_atomic_state *st static void mdp5_complete_commit(struct msm_kms *kms, struct drm_atomic_state *state) { + int i; struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms)); + int nplanes = mdp5_kms->dev->mode_config.num_total_plane; + + for (i = 0; i < nplanes; i++) { + struct drm_plane *plane = state->planes[i]; + struct drm_plane_state *plane_state = state->plane_states[i]; + + if (!plane) + continue; + + mdp5_plane_complete_commit(plane, plane_state); + } + mdp5_disable(mdp5_kms); } diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h index e0eb245..e79ac09 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h @@ -227,6 +227,8 @@ void mdp5_plane_install_properties(struct drm_plane *plane, struct drm_mode_object *obj); uint32_t mdp5_plane_get_flush(struct drm_plane *plane); void mdp5_plane_complete_flip(struct drm_plane *plane); +void mdp5_plane_complete_commit(struct drm_plane *plane, + struct drm_plane_state *state); enum mdp5_pipe mdp5_plane_pipe(struct drm_plane *plane); struct drm_plane *mdp5_plane_init(struct drm_device *dev, enum mdp5_pipe pipe, bool private_plane, uint32_t reg_offset); diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c index 57b8f56..2227556 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c @@ -31,8 +31,6 @@ struct mdp5_plane { uint32_t nformats; uint32_t formats[32]; - - bool enabled; }; #define to_mdp5_plane(x) container_of(x, struct mdp5_plane, base) @@ -56,22 +54,6 @@ static bool plane_enabled(struct drm_plane_state *state) return state->fb && state->crtc; } -static int mdp5_plane_disable(struct drm_plane *plane) -{ - struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane); - struct mdp5_kms *mdp5_kms = get_kms(plane); - enum mdp5_pipe pipe = mdp5_plane->pipe; - - DBG("%s: disable", mdp5_plane->name); - - if (mdp5_kms) { - /* Release the memory we requested earlier from the SMP: */ - mdp5_smp_release(mdp5_kms->smp, pipe); - } - - return 0; -} - static void mdp5_plane_destroy(struct drm_plane *plane) { struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane); @@ -224,7 +206,6 @@ static void mdp5_plane_atomic_update(struct drm_plane *plane, if (!plane_enabled(state)) { to_mdp5_plane_state(state)->pending = true; - mdp5_plane_disable(plane); } else if (to_mdp5_plane_state(state)->mode_changed) { int ret; to_mdp5_plane_state(state)->pending = true; @@ -602,6 +583,20 @@ uint32_t mdp5_plane_get_flush(struct drm_plane *plane) return mdp5_plane->flush_mask; } +/* called after vsync in thread context */ +void mdp5_plane_complete_commit(struct drm_plane *plane, + struct drm_plane_state *state) +{ + struct mdp5_kms *mdp5_kms = get_kms(plane); + struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane); + enum mdp5_pipe pipe = mdp5_plane->pipe; + + if (!plane_enabled(plane->state)) { + DBG("%s: free SMP", mdp5_plane->name); + mdp5_smp_release(mdp5_kms->smp, pipe); + } +} + /* initialize plane */ struct drm_plane *mdp5_plane_init(struct drm_device *dev, enum mdp5_pipe pipe, bool private_plane, uint32_t reg_offset) diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c index 16702ae..64a27d8 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c @@ -34,22 +34,44 @@ * and CANNOT be re-allocated (eg: MMB0 and MMB1 both tied to RGB0). * * For each block that can be dynamically allocated, it can be either - * free, or pending/in-use by a client. The updates happen in three steps: + * free: + * The block is free. + * + * pending: + * The block is allocated to some client and not free. + * + * configured: + * The block is allocated to some client, and assigned to that + * client in MDP5_MDP_SMP_ALLOC registers. + * + * inuse: + * The block is being actively used by a client. + * + * The updates happen in the following steps: * * 1) mdp5_smp_request(): * When plane scanout is setup, calculate required number of - * blocks needed per client, and request. Blocks not inuse or - * pending by any other client are added to client's pending - * set. + * blocks needed per client, and request. Blocks neither inuse nor + * configured nor pending by any other client are added to client's + * pending set. + * For shrinking, blocks in pending but not in configured can be freed + * directly, but those already in configured will be freed later by + * mdp5_smp_commit. * * 2) mdp5_smp_configure(): * As hw is programmed, before FLUSH, MDP5_MDP_SMP_ALLOC registers * are configured for the union(pending, inuse) + * Current pending is copied to configured. + * It is assumed that mdp5_smp_request and mdp5_smp_configure not run + * concurrently for the same pipe. * * 3) mdp5_smp_commit(): - * After next vblank, copy pending -> inuse. Optionally update + * After next vblank, copy configured -> inuse. Optionally update * MDP5_SMP_ALLOC registers if there are newly unused blocks * + * 4) mdp5_smp_release(): + * Must be called after the pipe is disabled and no longer uses any SMB + * * On the next vblank after changes have been committed to hw, the * client's pending blocks become it's in-use blocks (and no-longer * in-use blocks become available to other clients). @@ -77,6 +99,9 @@ struct mdp5_smp { struct mdp5_client_smp_state client_state[MAX_CLIENTS]; }; +static void update_smp_state(struct mdp5_smp *smp, + u32 cid, mdp5_smp_state_t *assigned); + static inline struct mdp5_kms *get_kms(struct mdp5_smp *smp) { @@ -149,7 +174,12 @@ static int smp_request_block(struct mdp5_smp *smp, for (i = cur_nblks; i > nblks; i--) { int blk = find_first_bit(ps->pending, cnt); clear_bit(blk, ps->pending); - /* don't clear in global smp_state until _commit() */ + + /* clear in global smp_state if not in configured + * otherwise until _commit() + */ + if (!test_bit(blk, ps->configured)) + clear_bit(blk, smp->state); } } @@ -223,10 +253,33 @@ int mdp5_smp_request(struct mdp5_smp *smp, enum mdp5_pipe pipe, u32 fmt, u32 wid /* Release SMP blocks for all clients of the pipe */ void mdp5_smp_release(struct mdp5_smp *smp, enum mdp5_pipe pipe) { - int i, nblks; + int i; + unsigned long flags; + int cnt = smp->blk_cnt; + + for (i = 0; i < pipe2nclients(pipe); i++) { + mdp5_smp_state_t assigned; + u32 cid = pipe2client(pipe, i); + struct mdp5_client_smp_state *ps = &smp->client_state[cid]; + + spin_lock_irqsave(&smp->state_lock, flags); + + /* clear hw assignment */ + bitmap_or(assigned, ps->inuse, ps->configured, cnt); + update_smp_state(smp, CID_UNUSED, &assigned); + + /* free to global pool */ + bitmap_andnot(smp->state, smp->state, ps->pending, cnt); + bitmap_andnot(smp->state, smp->state, assigned, cnt); + + /* clear client's infor */ + bitmap_zero(ps->pending, cnt); + bitmap_zero(ps->configured, cnt); + bitmap_zero(ps->inuse, cnt); + + spin_unlock_irqrestore(&smp->state_lock, flags); + } - for (i = 0, nblks = 0; i < pipe2nclients(pipe); i++) - smp_request_block(smp, pipe2client(pipe, i), 0); set_fifo_thresholds(smp, pipe, 0); } @@ -274,12 +327,20 @@ void mdp5_smp_configure(struct mdp5_smp *smp, enum mdp5_pipe pipe) u32 cid = pipe2client(pipe, i); struct mdp5_client_smp_state *ps = &smp->client_state[cid]; - bitmap_or(assigned, ps->inuse, ps->pending, cnt); + /* + * if vblank has not happened since last smp_configure + * skip the configure for now + */ + if (!bitmap_equal(ps->inuse, ps->configured, cnt)) + continue; + + bitmap_copy(ps->configured, ps->pending, cnt); + bitmap_or(assigned, ps->inuse, ps->configured, cnt); update_smp_state(smp, cid, &assigned); } } -/* step #3: after vblank, copy pending -> inuse: */ +/* step #3: after vblank, copy configured -> inuse: */ void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe) { int cnt = smp->blk_cnt; @@ -295,7 +356,7 @@ void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe) * using, which can be released and made available to other * clients: */ - if (bitmap_andnot(released, ps->inuse, ps->pending, cnt)) { + if (bitmap_andnot(released, ps->inuse, ps->configured, cnt)) { unsigned long flags; spin_lock_irqsave(&smp->state_lock, flags); @@ -306,7 +367,7 @@ void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe) update_smp_state(smp, CID_UNUSED, &released); } - bitmap_copy(ps->inuse, ps->pending, cnt); + bitmap_copy(ps->inuse, ps->configured, cnt); } } diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h index e47179f..5b6c236 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h @@ -23,6 +23,7 @@ struct mdp5_client_smp_state { mdp5_smp_state_t inuse; + mdp5_smp_state_t configured; mdp5_smp_state_t pending; }; diff --git a/drivers/gpu/drm/msm/msm_atomic.c b/drivers/gpu/drm/msm/msm_atomic.c index 1b22d8b..1ceb4f2 100644 --- a/drivers/gpu/drm/msm/msm_atomic.c +++ b/drivers/gpu/drm/msm/msm_atomic.c @@ -283,12 +283,8 @@ int msm_atomic_commit(struct drm_device *dev, timeout = ktime_add_ms(ktime_get(), 1000); - ret = msm_wait_fence_interruptable(dev, c->fence, &timeout); - if (ret) { - WARN_ON(ret); // TODO unswap state back? or?? - commit_destroy(c); - return ret; - } + /* uninterruptible wait */ + msm_wait_fence(dev, c->fence, &timeout, false); complete_commit(c); diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index b7ef56e..d3467b1 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -637,8 +637,8 @@ static void msm_debugfs_cleanup(struct drm_minor *minor) * Fences: */ -int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence, - ktime_t *timeout) +int msm_wait_fence(struct drm_device *dev, uint32_t fence, + ktime_t *timeout , bool interruptible) { struct msm_drm_private *priv = dev->dev_private; int ret; @@ -667,7 +667,12 @@ int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence, remaining_jiffies = timespec_to_jiffies(&ts); } - ret = wait_event_interruptible_timeout(priv->fence_event, + if (interruptible) + ret = wait_event_interruptible_timeout(priv->fence_event, + fence_completed(dev, fence), + remaining_jiffies); + else + ret = wait_event_timeout(priv->fence_event, fence_completed(dev, fence), remaining_jiffies); @@ -853,7 +858,7 @@ static int msm_ioctl_wait_fence(struct drm_device *dev, void *data, return -EINVAL; } - return msm_wait_fence_interruptable(dev, args->fence, &timeout); + return msm_wait_fence(dev, args->fence, &timeout, true); } static const struct drm_ioctl_desc msm_ioctls[] = { diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index e7c5ea1..4ff0ec9 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -164,8 +164,8 @@ int msm_atomic_commit(struct drm_device *dev, int msm_register_mmu(struct drm_device *dev, struct msm_mmu *mmu); -int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence, - ktime_t *timeout); +int msm_wait_fence(struct drm_device *dev, uint32_t fence, + ktime_t *timeout, bool interruptible); int msm_queue_fence_cb(struct drm_device *dev, struct msm_fence_cb *cb, uint32_t fence); void msm_update_fence(struct drm_device *dev, uint32_t fence); diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index f211b80..c76cc85 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -460,7 +460,7 @@ int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout) if (op & MSM_PREP_NOSYNC) timeout = NULL; - ret = msm_wait_fence_interruptable(dev, fence, timeout); + ret = msm_wait_fence(dev, fence, timeout, true); } /* TODO cache maintenance */ diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c index dd7a7ab..831461b 100644 --- a/drivers/gpu/drm/msm/msm_gem_prime.c +++ b/drivers/gpu/drm/msm/msm_gem_prime.c @@ -23,8 +23,12 @@ struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj) { struct msm_gem_object *msm_obj = to_msm_bo(obj); - BUG_ON(!msm_obj->sgt); /* should have already pinned! */ - return msm_obj->sgt; + int npages = obj->size >> PAGE_SHIFT; + + if (WARN_ON(!msm_obj->pages)) /* should have already pinned! */ + return NULL; + + return drm_prime_pages_to_sg(msm_obj->pages, npages); } void *msm_gem_prime_vmap(struct drm_gem_object *obj) diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 649024d..477cbb1 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -128,6 +128,7 @@ nouveau_cli_destroy(struct nouveau_cli *cli) nvkm_vm_ref(NULL, &nvxx_client(&cli->base)->vm, NULL); nvif_client_fini(&cli->base); usif_client_fini(cli); + kfree(cli); } static void @@ -865,8 +866,10 @@ nouveau_drm_preclose(struct drm_device *dev, struct drm_file *fpriv) pm_runtime_get_sync(dev->dev); + mutex_lock(&cli->mutex); if (cli->abi16) nouveau_abi16_fini(cli->abi16); + mutex_unlock(&cli->mutex); mutex_lock(&drm->client.mutex); list_del(&cli->head); diff --git a/drivers/gpu/drm/nouveau/nouveau_platform.c b/drivers/gpu/drm/nouveau/nouveau_platform.c index 775277f..dcfbbfa 100644 --- a/drivers/gpu/drm/nouveau/nouveau_platform.c +++ b/drivers/gpu/drm/nouveau/nouveau_platform.c @@ -92,6 +92,8 @@ static int nouveau_platform_power_down(struct nouveau_platform_gpu *gpu) return 0; } +#if IS_ENABLED(CONFIG_IOMMU_API) + static void nouveau_platform_probe_iommu(struct device *dev, struct nouveau_platform_gpu *gpu) { @@ -158,6 +160,20 @@ static void nouveau_platform_remove_iommu(struct device *dev, } } +#else + +static void nouveau_platform_probe_iommu(struct device *dev, + struct nouveau_platform_gpu *gpu) +{ +} + +static void nouveau_platform_remove_iommu(struct device *dev, + struct nouveau_platform_gpu *gpu) +{ +} + +#endif + static int nouveau_platform_probe(struct platform_device *pdev) { struct nouveau_platform_gpu *gpu; diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c index 18f4497..7464aef3 100644 --- a/drivers/gpu/drm/nouveau/nouveau_ttm.c +++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c @@ -175,15 +175,24 @@ nouveau_gart_manager_new(struct ttm_mem_type_manager *man, node->page_shift = 12; switch (drm->device.info.family) { + case NV_DEVICE_INFO_V0_TNT: + case NV_DEVICE_INFO_V0_CELSIUS: + case NV_DEVICE_INFO_V0_KELVIN: + case NV_DEVICE_INFO_V0_RANKINE: + case NV_DEVICE_INFO_V0_CURIE: + break; case NV_DEVICE_INFO_V0_TESLA: if (drm->device.info.chipset != 0x50) node->memtype = (nvbo->tile_flags & 0x7f00) >> 8; break; case NV_DEVICE_INFO_V0_FERMI: case NV_DEVICE_INFO_V0_KEPLER: + case NV_DEVICE_INFO_V0_MAXWELL: node->memtype = (nvbo->tile_flags & 0xff00) >> 8; break; default: + NV_WARN(drm, "%s: unhandled family type %x\n", __func__, + drm->device.info.family); break; } diff --git a/drivers/gpu/drm/nouveau/nv04_fbcon.c b/drivers/gpu/drm/nouveau/nv04_fbcon.c index 4ef602c..495c576 100644 --- a/drivers/gpu/drm/nouveau/nv04_fbcon.c +++ b/drivers/gpu/drm/nouveau/nv04_fbcon.c @@ -203,7 +203,7 @@ nv04_fbcon_accel_init(struct fb_info *info) if (ret) return ret; - if (RING_SPACE(chan, 49)) { + if (RING_SPACE(chan, 49 + (device->info.chipset >= 0x11 ? 4 : 0))) { nouveau_fbcon_gpu_lockup(info); return 0; } diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c index 7da7958..981342d 100644 --- a/drivers/gpu/drm/nouveau/nv50_display.c +++ b/drivers/gpu/drm/nouveau/nv50_display.c @@ -979,7 +979,7 @@ nv50_crtc_cursor_show_hide(struct nouveau_crtc *nv_crtc, bool show, bool update) { struct nv50_mast *mast = nv50_mast(nv_crtc->base.dev); - if (show && nv_crtc->cursor.nvbo) + if (show && nv_crtc->cursor.nvbo && nv_crtc->base.enabled) nv50_crtc_cursor_show(nv_crtc); else nv50_crtc_cursor_hide(nv_crtc); diff --git a/drivers/gpu/drm/nouveau/nv50_fbcon.c b/drivers/gpu/drm/nouveau/nv50_fbcon.c index 394c89a..901130b 100644 --- a/drivers/gpu/drm/nouveau/nv50_fbcon.c +++ b/drivers/gpu/drm/nouveau/nv50_fbcon.c @@ -188,7 +188,7 @@ nv50_fbcon_accel_init(struct fb_info *info) if (ret) return ret; - ret = RING_SPACE(chan, 59); + ret = RING_SPACE(chan, 58); if (ret) { nouveau_fbcon_gpu_lockup(info); return ret; @@ -252,6 +252,7 @@ nv50_fbcon_accel_init(struct fb_info *info) OUT_RING(chan, info->var.yres_virtual); OUT_RING(chan, upper_32_bits(fb->vma.offset)); OUT_RING(chan, lower_32_bits(fb->vma.offset)); + FIRE_RING(chan); return 0; } diff --git a/drivers/gpu/drm/nouveau/nvc0_fbcon.c b/drivers/gpu/drm/nouveau/nvc0_fbcon.c index 6124667..fcd2e5f 100644 --- a/drivers/gpu/drm/nouveau/nvc0_fbcon.c +++ b/drivers/gpu/drm/nouveau/nvc0_fbcon.c @@ -188,7 +188,7 @@ nvc0_fbcon_accel_init(struct fb_info *info) return -EINVAL; } - ret = RING_SPACE(chan, 60); + ret = RING_SPACE(chan, 58); if (ret) { WARN_ON(1); nouveau_fbcon_gpu_lockup(info); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c index 9ef6728..7f2f05f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c @@ -809,7 +809,7 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int or, u32 ctrl, case 0x00000900: type = DCB_OUTPUT_DP; mask = 2; break; default: nv_error(priv, "unknown SOR mc 0x%08x\n", ctrl); - return 0x0000; + return NULL; } } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c index 5606c25..ca11ddb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c @@ -663,6 +663,37 @@ gf100_gr_zbc_init(struct gf100_gr_priv *priv) gf100_gr_zbc_clear_depth(priv, index); } +/** + * Wait until GR goes idle. GR is considered idle if it is disabled by the + * MC (0x200) register, or GR is not busy and a context switch is not in + * progress. + */ +int +gf100_gr_wait_idle(struct gf100_gr_priv *priv) +{ + unsigned long end_jiffies = jiffies + msecs_to_jiffies(2000); + bool gr_enabled, ctxsw_active, gr_busy; + + do { + /* + * required to make sure FIFO_ENGINE_STATUS (0x2640) is + * up-to-date + */ + nv_rd32(priv, 0x400700); + + gr_enabled = nv_rd32(priv, 0x200) & 0x1000; + ctxsw_active = nv_rd32(priv, 0x2640) & 0x8000; + gr_busy = nv_rd32(priv, 0x40060c) & 0x1; + + if (!gr_enabled || (!gr_busy && !ctxsw_active)) + return 0; + } while (time_before(jiffies, end_jiffies)); + + nv_error(priv, "wait for idle timeout (en: %d, ctxsw: %d, busy: %d)\n", + gr_enabled, ctxsw_active, gr_busy); + return -EAGAIN; +} + void gf100_gr_mmio(struct gf100_gr_priv *priv, const struct gf100_gr_pack *p) { @@ -699,7 +730,13 @@ gf100_gr_icmd(struct gf100_gr_priv *priv, const struct gf100_gr_pack *p) while (addr < next) { nv_wr32(priv, 0x400200, addr); - nv_wait(priv, 0x400700, 0x00000002, 0x00000000); + /** + * Wait for GR to go idle after submitting a + * GO_IDLE bundle + */ + if ((addr & 0xffff) == 0xe100) + gf100_gr_wait_idle(priv); + nv_wait(priv, 0x400700, 0x00000004, 0x00000000); addr += init->pitch; } } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h index 8af1a89..c9533fd 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h @@ -181,6 +181,7 @@ struct gf100_gr_oclass { int ppc_nr; }; +int gf100_gr_wait_idle(struct gf100_gr_priv *); void gf100_gr_mmio(struct gf100_gr_priv *, const struct gf100_gr_pack *); void gf100_gr_icmd(struct gf100_gr_priv *, const struct gf100_gr_pack *); void gf100_gr_mthd(struct gf100_gr_priv *, const struct gf100_gr_pack *); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c index 2006c44..4cf36a3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c @@ -332,9 +332,12 @@ static void nvkm_perfctx_dtor(struct nvkm_object *object) { struct nvkm_pm *ppm = (void *)object->engine; + struct nvkm_perfctx *ctx = (void *)object; + mutex_lock(&nv_subdev(ppm)->mutex); - nvkm_engctx_destroy(&ppm->context->base); - ppm->context = NULL; + nvkm_engctx_destroy(&ctx->base); + if (ppm->context == ctx) + ppm->context = NULL; mutex_unlock(&nv_subdev(ppm)->mutex); } @@ -355,12 +358,11 @@ nvkm_perfctx_ctor(struct nvkm_object *parent, struct nvkm_object *engine, mutex_lock(&nv_subdev(ppm)->mutex); if (ppm->context == NULL) ppm->context = ctx; - mutex_unlock(&nv_subdev(ppm)->mutex); - if (ctx != ppm->context) - return -EBUSY; + ret = -EBUSY; + mutex_unlock(&nv_subdev(ppm)->mutex); - return 0; + return ret; } struct nvkm_oclass diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c index f67cdae..f4611e3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c @@ -1285,6 +1285,44 @@ init_zm_reg_sequence(struct nvbios_init *init) } /** + * INIT_PLL_INDIRECT - opcode 0x59 + * + */ +static void +init_pll_indirect(struct nvbios_init *init) +{ + struct nvkm_bios *bios = init->bios; + u32 reg = nv_ro32(bios, init->offset + 1); + u16 addr = nv_ro16(bios, init->offset + 5); + u32 freq = (u32)nv_ro16(bios, addr) * 1000; + + trace("PLL_INDIRECT\tR[0x%06x] =PLL= VBIOS[%04x] = %dkHz\n", + reg, addr, freq); + init->offset += 7; + + init_prog_pll(init, reg, freq); +} + +/** + * INIT_ZM_REG_INDIRECT - opcode 0x5a + * + */ +static void +init_zm_reg_indirect(struct nvbios_init *init) +{ + struct nvkm_bios *bios = init->bios; + u32 reg = nv_ro32(bios, init->offset + 1); + u16 addr = nv_ro16(bios, init->offset + 5); + u32 data = nv_ro32(bios, addr); + + trace("ZM_REG_INDIRECT\tR[0x%06x] = VBIOS[0x%04x] = 0x%08x\n", + reg, addr, data); + init->offset += 7; + + init_wr32(init, addr, data); +} + +/** * INIT_SUB_DIRECT - opcode 0x5b * */ @@ -2145,6 +2183,8 @@ static struct nvbios_init_opcode { [0x56] = { init_condition_time }, [0x57] = { init_ltime }, [0x58] = { init_zm_reg_sequence }, + [0x59] = { init_pll_indirect }, + [0x5a] = { init_zm_reg_indirect }, [0x5b] = { init_sub_direct }, [0x5c] = { init_jump }, [0x5e] = { init_i2c_if }, diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c index 822d32a2..065e9f5 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c @@ -180,7 +180,8 @@ gt215_clk_info(struct nvkm_clk *clock, int clk, u32 khz, struct gt215_clk_info *info) { struct gt215_clk_priv *priv = (void *)clock; - u32 oclk, sclk, sdiv, diff; + u32 oclk, sclk, sdiv; + s32 diff; info->clk = 0; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c index c0fdb89..24dcdfb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c @@ -38,6 +38,14 @@ gk20a_ibus_init_priv_ring(struct gk20a_ibus_priv *priv) nv_wr32(priv, 0x12004c, 0x4); nv_wr32(priv, 0x122204, 0x2); nv_rd32(priv, 0x122204); + + /* + * Bug: increase clock timeout to avoid operation failure at high + * gpcclk rate. + */ + nv_wr32(priv, 0x122354, 0x800); + nv_wr32(priv, 0x128328, 0x800); + nv_wr32(priv, 0x124320, 0x800); } static void diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c index 80614f1..282143f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c @@ -50,7 +50,12 @@ nv04_instobj_dtor(struct nvkm_object *object) { struct nv04_instmem_priv *priv = (void *)nvkm_instmem(object); struct nv04_instobj_priv *node = (void *)object; + struct nvkm_subdev *subdev = (void *)priv; + + mutex_lock(&subdev->mutex); nvkm_mm_free(&priv->heap, &node->mem); + mutex_unlock(&subdev->mutex); + nvkm_instobj_destroy(&node->base); } @@ -62,6 +67,7 @@ nv04_instobj_ctor(struct nvkm_object *parent, struct nvkm_object *engine, struct nv04_instmem_priv *priv = (void *)nvkm_instmem(parent); struct nv04_instobj_priv *node; struct nvkm_instobj_args *args = data; + struct nvkm_subdev *subdev = (void *)priv; int ret; if (!args->align) @@ -72,8 +78,10 @@ nv04_instobj_ctor(struct nvkm_object *parent, struct nvkm_object *engine, if (ret) return ret; + mutex_lock(&subdev->mutex); ret = nvkm_mm_head(&priv->heap, 0, 1, args->size, args->size, args->align, &node->mem); + mutex_unlock(&subdev->mutex); if (ret) return ret; diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index dd39f43..c387259 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -2299,8 +2299,7 @@ radeon_atom_encoder_mode_set(struct drm_encoder *encoder, encoder_mode = atombios_get_encoder_mode(encoder); if (connector && (radeon_audio != 0) && ((encoder_mode == ATOM_ENCODER_MODE_HDMI) || - (ENCODER_MODE_IS_DP(encoder_mode) && - drm_detect_monitor_audio(radeon_connector_edid(connector))))) + ENCODER_MODE_IS_DP(encoder_mode))) radeon_audio_mode_set(encoder, adjusted_mode); } diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c index 8730562..4a09947 100644 --- a/drivers/gpu/drm/radeon/ci_dpm.c +++ b/drivers/gpu/drm/radeon/ci_dpm.c @@ -5818,7 +5818,7 @@ int ci_dpm_init(struct radeon_device *rdev) tmp |= DPM_ENABLED; break; default: - DRM_ERROR("Invalid PCC GPIO: %u!\n", gpio.shift); + DRM_DEBUG("Invalid PCC GPIO: %u!\n", gpio.shift); break; } WREG32_SMC(CNB_PWRMGT_CNTL, tmp); diff --git a/drivers/gpu/drm/radeon/dce6_afmt.c b/drivers/gpu/drm/radeon/dce6_afmt.c index 68fd9fc..44480c1 100644 --- a/drivers/gpu/drm/radeon/dce6_afmt.c +++ b/drivers/gpu/drm/radeon/dce6_afmt.c @@ -93,30 +93,26 @@ void dce6_afmt_select_pin(struct drm_encoder *encoder) struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - u32 offset; - if (!dig || !dig->afmt || !dig->afmt->pin) + if (!dig || !dig->afmt || !dig->pin) return; - offset = dig->afmt->offset; - - WREG32(AFMT_AUDIO_SRC_CONTROL + offset, - AFMT_AUDIO_SRC_SELECT(dig->afmt->pin->id)); + WREG32(AFMT_AUDIO_SRC_CONTROL + dig->afmt->offset, + AFMT_AUDIO_SRC_SELECT(dig->pin->id)); } void dce6_afmt_write_latency_fields(struct drm_encoder *encoder, - struct drm_connector *connector, struct drm_display_mode *mode) + struct drm_connector *connector, + struct drm_display_mode *mode) { struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - u32 tmp = 0, offset; + u32 tmp = 0; - if (!dig || !dig->afmt || !dig->afmt->pin) + if (!dig || !dig->afmt || !dig->pin) return; - offset = dig->afmt->pin->offset; - if (mode->flags & DRM_MODE_FLAG_INTERLACE) { if (connector->latency_present[1]) tmp = VIDEO_LIPSYNC(connector->video_latency[1]) | @@ -130,24 +126,24 @@ void dce6_afmt_write_latency_fields(struct drm_encoder *encoder, else tmp = VIDEO_LIPSYNC(0) | AUDIO_LIPSYNC(0); } - WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC, tmp); + WREG32_ENDPOINT(dig->pin->offset, + AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC, tmp); } void dce6_afmt_hdmi_write_speaker_allocation(struct drm_encoder *encoder, - u8 *sadb, int sad_count) + u8 *sadb, int sad_count) { struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - u32 offset, tmp; + u32 tmp; - if (!dig || !dig->afmt || !dig->afmt->pin) + if (!dig || !dig->afmt || !dig->pin) return; - offset = dig->afmt->pin->offset; - /* program the speaker allocation */ - tmp = RREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER); + tmp = RREG32_ENDPOINT(dig->pin->offset, + AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER); tmp &= ~(DP_CONNECTION | SPEAKER_ALLOCATION_MASK); /* set HDMI mode */ tmp |= HDMI_CONNECTION; @@ -155,24 +151,24 @@ void dce6_afmt_hdmi_write_speaker_allocation(struct drm_encoder *encoder, tmp |= SPEAKER_ALLOCATION(sadb[0]); else tmp |= SPEAKER_ALLOCATION(5); /* stereo */ - WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp); + WREG32_ENDPOINT(dig->pin->offset, + AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp); } void dce6_afmt_dp_write_speaker_allocation(struct drm_encoder *encoder, - u8 *sadb, int sad_count) + u8 *sadb, int sad_count) { struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - u32 offset, tmp; + u32 tmp; - if (!dig || !dig->afmt || !dig->afmt->pin) + if (!dig || !dig->afmt || !dig->pin) return; - offset = dig->afmt->pin->offset; - /* program the speaker allocation */ - tmp = RREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER); + tmp = RREG32_ENDPOINT(dig->pin->offset, + AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER); tmp &= ~(HDMI_CONNECTION | SPEAKER_ALLOCATION_MASK); /* set DP mode */ tmp |= DP_CONNECTION; @@ -180,13 +176,13 @@ void dce6_afmt_dp_write_speaker_allocation(struct drm_encoder *encoder, tmp |= SPEAKER_ALLOCATION(sadb[0]); else tmp |= SPEAKER_ALLOCATION(5); /* stereo */ - WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp); + WREG32_ENDPOINT(dig->pin->offset, + AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp); } void dce6_afmt_write_sad_regs(struct drm_encoder *encoder, - struct cea_sad *sads, int sad_count) + struct cea_sad *sads, int sad_count) { - u32 offset; int i; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; @@ -206,11 +202,9 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder, { AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO }, }; - if (!dig || !dig->afmt || !dig->afmt->pin) + if (!dig || !dig->afmt || !dig->pin) return; - offset = dig->afmt->pin->offset; - for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) { u32 value = 0; u8 stereo_freqs = 0; @@ -237,7 +231,7 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder, value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs); - WREG32_ENDPOINT(offset, eld_reg_to_type[i][0], value); + WREG32_ENDPOINT(dig->pin->offset, eld_reg_to_type[i][0], value); } } @@ -253,7 +247,7 @@ void dce6_audio_enable(struct radeon_device *rdev, } void dce6_hdmi_audio_set_dto(struct radeon_device *rdev, - struct radeon_crtc *crtc, unsigned int clock) + struct radeon_crtc *crtc, unsigned int clock) { /* Two dtos; generally use dto0 for HDMI */ u32 value = 0; @@ -272,7 +266,7 @@ void dce6_hdmi_audio_set_dto(struct radeon_device *rdev, } void dce6_dp_audio_set_dto(struct radeon_device *rdev, - struct radeon_crtc *crtc, unsigned int clock) + struct radeon_crtc *crtc, unsigned int clock) { /* Two dtos; generally use dto1 for DP */ u32 value = 0; diff --git a/drivers/gpu/drm/radeon/radeon_audio.c b/drivers/gpu/drm/radeon/radeon_audio.c index fa719c5..fbc8d88 100644 --- a/drivers/gpu/drm/radeon/radeon_audio.c +++ b/drivers/gpu/drm/radeon/radeon_audio.c @@ -245,6 +245,28 @@ static struct radeon_audio_funcs dce6_dp_funcs = { static void radeon_audio_enable(struct radeon_device *rdev, struct r600_audio_pin *pin, u8 enable_mask) { + struct drm_encoder *encoder; + struct radeon_encoder *radeon_encoder; + struct radeon_encoder_atom_dig *dig; + int pin_count = 0; + + if (!pin) + return; + + if (rdev->mode_info.mode_config_initialized) { + list_for_each_entry(encoder, &rdev->ddev->mode_config.encoder_list, head) { + if (radeon_encoder_is_digital(encoder)) { + radeon_encoder = to_radeon_encoder(encoder); + dig = radeon_encoder->enc_priv; + if (dig->pin == pin) + pin_count++; + } + } + + if ((pin_count > 1) && (enable_mask == 0)) + return; + } + if (rdev->audio.funcs->enable) rdev->audio.funcs->enable(rdev, pin, enable_mask); } @@ -336,24 +358,13 @@ void radeon_audio_endpoint_wreg(struct radeon_device *rdev, u32 offset, static void radeon_audio_write_sad_regs(struct drm_encoder *encoder) { - struct radeon_encoder *radeon_encoder; - struct drm_connector *connector; - struct radeon_connector *radeon_connector = NULL; + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); + struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct cea_sad *sads; int sad_count; - list_for_each_entry(connector, - &encoder->dev->mode_config.connector_list, head) { - if (connector->encoder == encoder) { - radeon_connector = to_radeon_connector(connector); - break; - } - } - - if (!radeon_connector) { - DRM_ERROR("Couldn't find encoder's connector\n"); + if (!connector) return; - } sad_count = drm_edid_to_sad(radeon_connector_edid(connector), &sads); if (sad_count <= 0) { @@ -362,8 +373,6 @@ static void radeon_audio_write_sad_regs(struct drm_encoder *encoder) } BUG_ON(!sads); - radeon_encoder = to_radeon_encoder(encoder); - if (radeon_encoder->audio && radeon_encoder->audio->write_sad_regs) radeon_encoder->audio->write_sad_regs(encoder, sads, sad_count); @@ -372,27 +381,16 @@ static void radeon_audio_write_sad_regs(struct drm_encoder *encoder) static void radeon_audio_write_speaker_allocation(struct drm_encoder *encoder) { + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); - struct drm_connector *connector; - struct radeon_connector *radeon_connector = NULL; u8 *sadb = NULL; int sad_count; - list_for_each_entry(connector, - &encoder->dev->mode_config.connector_list, head) { - if (connector->encoder == encoder) { - radeon_connector = to_radeon_connector(connector); - break; - } - } - - if (!radeon_connector) { - DRM_ERROR("Couldn't find encoder's connector\n"); + if (!connector) return; - } - sad_count = drm_edid_to_speaker_allocation( - radeon_connector_edid(connector), &sadb); + sad_count = drm_edid_to_speaker_allocation(radeon_connector_edid(connector), + &sadb); if (sad_count < 0) { DRM_DEBUG("Couldn't read Speaker Allocation Data Block: %d\n", sad_count); @@ -406,26 +404,13 @@ static void radeon_audio_write_speaker_allocation(struct drm_encoder *encoder) } static void radeon_audio_write_latency_fields(struct drm_encoder *encoder, - struct drm_display_mode *mode) + struct drm_display_mode *mode) { - struct radeon_encoder *radeon_encoder; - struct drm_connector *connector; - struct radeon_connector *radeon_connector = 0; - - list_for_each_entry(connector, - &encoder->dev->mode_config.connector_list, head) { - if (connector->encoder == encoder) { - radeon_connector = to_radeon_connector(connector); - break; - } - } + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); + struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); - if (!radeon_connector) { - DRM_ERROR("Couldn't find encoder's connector\n"); + if (!connector) return; - } - - radeon_encoder = to_radeon_encoder(encoder); if (radeon_encoder->audio && radeon_encoder->audio->write_latency_fields) radeon_encoder->audio->write_latency_fields(encoder, connector, mode); @@ -451,29 +436,23 @@ static void radeon_audio_select_pin(struct drm_encoder *encoder) } void radeon_audio_detect(struct drm_connector *connector, + struct drm_encoder *encoder, enum drm_connector_status status) { - struct radeon_device *rdev; - struct radeon_encoder *radeon_encoder; + struct drm_device *dev = connector->dev; + struct radeon_device *rdev = dev->dev_private; + struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig; - if (!connector || !connector->encoder) + if (!radeon_audio_chipset_supported(rdev)) return; - rdev = connector->encoder->dev->dev_private; - - if (!radeon_audio_chipset_supported(rdev)) + if (!radeon_encoder_is_digital(encoder)) return; - radeon_encoder = to_radeon_encoder(connector->encoder); dig = radeon_encoder->enc_priv; if (status == connector_status_connected) { - if (!drm_detect_monitor_audio(radeon_connector_edid(connector))) { - radeon_encoder->audio = NULL; - return; - } - if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) { struct radeon_connector *radeon_connector = to_radeon_connector(connector); @@ -486,11 +465,17 @@ void radeon_audio_detect(struct drm_connector *connector, radeon_encoder->audio = rdev->audio.hdmi_funcs; } - dig->afmt->pin = radeon_audio_get_pin(connector->encoder); - radeon_audio_enable(rdev, dig->afmt->pin, 0xf); + if (drm_detect_monitor_audio(radeon_connector_edid(connector))) { + if (!dig->pin) + dig->pin = radeon_audio_get_pin(encoder); + radeon_audio_enable(rdev, dig->pin, 0xf); + } else { + radeon_audio_enable(rdev, dig->pin, 0); + dig->pin = NULL; + } } else { - radeon_audio_enable(rdev, dig->afmt->pin, 0); - dig->afmt->pin = NULL; + radeon_audio_enable(rdev, dig->pin, 0); + dig->pin = NULL; } } @@ -518,29 +503,18 @@ static void radeon_audio_set_dto(struct drm_encoder *encoder, unsigned int clock } static int radeon_audio_set_avi_packet(struct drm_encoder *encoder, - struct drm_display_mode *mode) + struct drm_display_mode *mode) { struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; - struct drm_connector *connector; - struct radeon_connector *radeon_connector = NULL; + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE]; struct hdmi_avi_infoframe frame; int err; - list_for_each_entry(connector, - &encoder->dev->mode_config.connector_list, head) { - if (connector->encoder == encoder) { - radeon_connector = to_radeon_connector(connector); - break; - } - } - - if (!radeon_connector) { - DRM_ERROR("Couldn't find encoder's connector\n"); - return -ENOENT; - } + if (!connector) + return -EINVAL; err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode); if (err < 0) { @@ -563,8 +537,8 @@ static int radeon_audio_set_avi_packet(struct drm_encoder *encoder, return err; } - if (dig && dig->afmt && - radeon_encoder->audio && radeon_encoder->audio->set_avi_packet) + if (dig && dig->afmt && radeon_encoder->audio && + radeon_encoder->audio->set_avi_packet) radeon_encoder->audio->set_avi_packet(rdev, dig->afmt->offset, buffer, sizeof(buffer)); @@ -722,30 +696,41 @@ static void radeon_audio_hdmi_mode_set(struct drm_encoder *encoder, { struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); if (!dig || !dig->afmt) return; - radeon_audio_set_mute(encoder, true); + if (!connector) + return; - radeon_audio_write_speaker_allocation(encoder); - radeon_audio_write_sad_regs(encoder); - radeon_audio_write_latency_fields(encoder, mode); - radeon_audio_set_dto(encoder, mode->clock); - radeon_audio_set_vbi_packet(encoder); - radeon_hdmi_set_color_depth(encoder); - radeon_audio_update_acr(encoder, mode->clock); - radeon_audio_set_audio_packet(encoder); - radeon_audio_select_pin(encoder); + if (drm_detect_monitor_audio(radeon_connector_edid(connector))) { + radeon_audio_set_mute(encoder, true); - if (radeon_audio_set_avi_packet(encoder, mode) < 0) - return; + radeon_audio_write_speaker_allocation(encoder); + radeon_audio_write_sad_regs(encoder); + radeon_audio_write_latency_fields(encoder, mode); + radeon_audio_set_dto(encoder, mode->clock); + radeon_audio_set_vbi_packet(encoder); + radeon_hdmi_set_color_depth(encoder); + radeon_audio_update_acr(encoder, mode->clock); + radeon_audio_set_audio_packet(encoder); + radeon_audio_select_pin(encoder); + + if (radeon_audio_set_avi_packet(encoder, mode) < 0) + return; - radeon_audio_set_mute(encoder, false); + radeon_audio_set_mute(encoder, false); + } else { + radeon_hdmi_set_color_depth(encoder); + + if (radeon_audio_set_avi_packet(encoder, mode) < 0) + return; + } } static void radeon_audio_dp_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode) + struct drm_display_mode *mode) { struct drm_device *dev = encoder->dev; struct radeon_device *rdev = dev->dev_private; @@ -759,22 +744,27 @@ static void radeon_audio_dp_mode_set(struct drm_encoder *encoder, if (!dig || !dig->afmt) return; - radeon_audio_write_speaker_allocation(encoder); - radeon_audio_write_sad_regs(encoder); - radeon_audio_write_latency_fields(encoder, mode); - if (rdev->clock.dp_extclk || ASIC_IS_DCE5(rdev)) - radeon_audio_set_dto(encoder, rdev->clock.default_dispclk * 10); - else - radeon_audio_set_dto(encoder, dig_connector->dp_clock); - radeon_audio_set_audio_packet(encoder); - radeon_audio_select_pin(encoder); - - if (radeon_audio_set_avi_packet(encoder, mode) < 0) + if (!connector) return; + + if (drm_detect_monitor_audio(radeon_connector_edid(connector))) { + radeon_audio_write_speaker_allocation(encoder); + radeon_audio_write_sad_regs(encoder); + radeon_audio_write_latency_fields(encoder, mode); + if (rdev->clock.dp_extclk || ASIC_IS_DCE5(rdev)) + radeon_audio_set_dto(encoder, rdev->clock.default_dispclk * 10); + else + radeon_audio_set_dto(encoder, dig_connector->dp_clock); + radeon_audio_set_audio_packet(encoder); + radeon_audio_select_pin(encoder); + + if (radeon_audio_set_avi_packet(encoder, mode) < 0) + return; + } } void radeon_audio_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode) + struct drm_display_mode *mode) { struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); diff --git a/drivers/gpu/drm/radeon/radeon_audio.h b/drivers/gpu/drm/radeon/radeon_audio.h index 8438304..059cc30 100644 --- a/drivers/gpu/drm/radeon/radeon_audio.h +++ b/drivers/gpu/drm/radeon/radeon_audio.h @@ -68,7 +68,8 @@ struct radeon_audio_funcs int radeon_audio_init(struct radeon_device *rdev); void radeon_audio_detect(struct drm_connector *connector, - enum drm_connector_status status); + struct drm_encoder *encoder, + enum drm_connector_status status); u32 radeon_audio_endpoint_rreg(struct radeon_device *rdev, u32 offset, u32 reg); void radeon_audio_endpoint_wreg(struct radeon_device *rdev, diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c index 3e5f6b7..c097d3a 100644 --- a/drivers/gpu/drm/radeon/radeon_combios.c +++ b/drivers/gpu/drm/radeon/radeon_combios.c @@ -1255,10 +1255,15 @@ struct radeon_encoder_lvds *radeon_combios_get_lvds_info(struct radeon_encoder if ((RBIOS16(tmp) == lvds->native_mode.hdisplay) && (RBIOS16(tmp + 2) == lvds->native_mode.vdisplay)) { + u32 hss = (RBIOS16(tmp + 21) - RBIOS16(tmp + 19) - 1) * 8; + + if (hss > lvds->native_mode.hdisplay) + hss = (10 - 1) * 8; + lvds->native_mode.htotal = lvds->native_mode.hdisplay + (RBIOS16(tmp + 17) - RBIOS16(tmp + 19)) * 8; lvds->native_mode.hsync_start = lvds->native_mode.hdisplay + - (RBIOS16(tmp + 21) - RBIOS16(tmp + 19) - 1) * 8; + hss; lvds->native_mode.hsync_end = lvds->native_mode.hsync_start + (RBIOS8(tmp + 23) * 8); diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index cebb65e..94b21ae 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -1379,8 +1379,16 @@ out: /* updated in get modes as well since we need to know if it's analog or digital */ radeon_connector_update_scratch_regs(connector, ret); - if (radeon_audio != 0) - radeon_audio_detect(connector, ret); + if ((radeon_audio != 0) && radeon_connector->use_digital) { + const struct drm_connector_helper_funcs *connector_funcs = + connector->helper_private; + + encoder = connector_funcs->best_encoder(connector); + if (encoder && (encoder->encoder_type == DRM_MODE_ENCODER_TMDS)) { + radeon_connector_get_edid(connector); + radeon_audio_detect(connector, encoder, ret); + } + } exit: pm_runtime_mark_last_busy(connector->dev->dev); @@ -1717,8 +1725,10 @@ radeon_dp_detect(struct drm_connector *connector, bool force) radeon_connector_update_scratch_regs(connector, ret); - if (radeon_audio != 0) - radeon_audio_detect(connector, ret); + if ((radeon_audio != 0) && encoder) { + radeon_connector_get_edid(connector); + radeon_audio_detect(connector, encoder, ret); + } out: pm_runtime_mark_last_busy(connector->dev->dev); diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c index 5450fa9..c4777c8 100644 --- a/drivers/gpu/drm/radeon/radeon_gart.c +++ b/drivers/gpu/drm/radeon/radeon_gart.c @@ -260,8 +260,10 @@ void radeon_gart_unbind(struct radeon_device *rdev, unsigned offset, } } } - mb(); - radeon_gart_tlb_flush(rdev); + if (rdev->gart.ptr) { + mb(); + radeon_gart_tlb_flush(rdev); + } } /** @@ -306,8 +308,10 @@ int radeon_gart_bind(struct radeon_device *rdev, unsigned offset, page_base += RADEON_GPU_PAGE_SIZE; } } - mb(); - radeon_gart_tlb_flush(rdev); + if (rdev->gart.ptr) { + mb(); + radeon_gart_tlb_flush(rdev); + } return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index 013ec71..3dcc573 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -36,6 +36,7 @@ void radeon_gem_object_free(struct drm_gem_object *gobj) if (robj) { if (robj->gem_base.import_attach) drm_prime_gem_destroy(&robj->gem_base, robj->tbo.sg); + radeon_mn_unregister(robj); radeon_bo_unref(&robj); } } diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c index 1162bfa..171d3e4 100644 --- a/drivers/gpu/drm/radeon/radeon_irq_kms.c +++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c @@ -79,6 +79,11 @@ static void radeon_hotplug_work_func(struct work_struct *work) struct drm_mode_config *mode_config = &dev->mode_config; struct drm_connector *connector; + /* we can race here at startup, some boards seem to trigger + * hotplug irqs when they shouldn't. */ + if (!rdev->mode_info.mode_config_initialized) + return; + mutex_lock(&mode_config->mutex); if (mode_config->num_connector) { list_for_each_entry(connector, &mode_config->connector_list, head) diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index 07909d8..aecc3e3 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -237,7 +237,6 @@ struct radeon_afmt { int offset; bool last_buffer_filled_status; int id; - struct r600_audio_pin *pin; }; struct radeon_mode_info { @@ -439,6 +438,7 @@ struct radeon_encoder_atom_dig { uint8_t backlight_level; int panel_mode; struct radeon_afmt *afmt; + struct r600_audio_pin *pin; int active_mst_links; }; diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c index 318165d..6763627 100644 --- a/drivers/gpu/drm/radeon/radeon_object.c +++ b/drivers/gpu/drm/radeon/radeon_object.c @@ -75,7 +75,6 @@ static void radeon_ttm_bo_destroy(struct ttm_buffer_object *tbo) bo = container_of(tbo, struct radeon_bo, tbo); radeon_update_memory_usage(bo, bo->tbo.mem.mem_type, -1); - radeon_mn_unregister(bo); mutex_lock(&bo->rdev->gem.mutex); list_del_init(&bo->list); diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index 1dbdf32..787cd8f 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -2926,6 +2926,7 @@ static struct si_dpm_quirk si_dpm_quirk_list[] = { /* PITCAIRN - https://bugs.freedesktop.org/show_bug.cgi?id=76490 */ { PCI_VENDOR_ID_ATI, 0x6810, 0x1462, 0x3036, 0, 120000 }, { PCI_VENDOR_ID_ATI, 0x6811, 0x174b, 0xe271, 0, 120000 }, + { PCI_VENDOR_ID_ATI, 0x6810, 0x174b, 0xe271, 85000, 90000 }, { 0, 0, 0, 0 }, }; diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c index 01b558f..9a0c291 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c @@ -555,7 +555,6 @@ static struct platform_driver rockchip_drm_platform_driver = { .probe = rockchip_drm_platform_probe, .remove = rockchip_drm_platform_remove, .driver = { - .owner = THIS_MODULE, .name = "rockchip-drm", .of_match_table = rockchip_drm_dt_ids, .pm = &rockchip_drm_pm_ops, diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fb.c b/drivers/gpu/drm/rockchip/rockchip_drm_fb.c index 77d5289..002645b 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_fb.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_fb.c @@ -162,7 +162,8 @@ static void rockchip_drm_output_poll_changed(struct drm_device *dev) struct rockchip_drm_private *private = dev->dev_private; struct drm_fb_helper *fb_helper = &private->fbdev_helper; - drm_fb_helper_hotplug_event(fb_helper); + if (fb_helper) + drm_fb_helper_hotplug_event(fb_helper); } static const struct drm_mode_config_funcs rockchip_drm_mode_config_funcs = { diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c index eb2282c..eba5f8a 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c @@ -54,55 +54,56 @@ static void rockchip_gem_free_buf(struct rockchip_gem_object *rk_obj) &rk_obj->dma_attrs); } -int rockchip_gem_mmap_buf(struct drm_gem_object *obj, - struct vm_area_struct *vma) +static int rockchip_drm_gem_object_mmap(struct drm_gem_object *obj, + struct vm_area_struct *vma) + { + int ret; struct rockchip_gem_object *rk_obj = to_rockchip_obj(obj); struct drm_device *drm = obj->dev; - unsigned long vm_size; - vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP; - vm_size = vma->vm_end - vma->vm_start; - - if (vm_size > obj->size) - return -EINVAL; + /* + * dma_alloc_attrs() allocated a struct page table for rk_obj, so clear + * VM_PFNMAP flag that was set by drm_gem_mmap_obj()/drm_gem_mmap(). + */ + vma->vm_flags &= ~VM_PFNMAP; - return dma_mmap_attrs(drm->dev, vma, rk_obj->kvaddr, rk_obj->dma_addr, + ret = dma_mmap_attrs(drm->dev, vma, rk_obj->kvaddr, rk_obj->dma_addr, obj->size, &rk_obj->dma_attrs); + if (ret) + drm_gem_vm_close(vma); + + return ret; } -/* drm driver mmap file operations */ -int rockchip_gem_mmap(struct file *filp, struct vm_area_struct *vma) +int rockchip_gem_mmap_buf(struct drm_gem_object *obj, + struct vm_area_struct *vma) { - struct drm_file *priv = filp->private_data; - struct drm_device *dev = priv->minor->dev; - struct drm_gem_object *obj; - struct drm_vma_offset_node *node; + struct drm_device *drm = obj->dev; int ret; - if (drm_device_is_unplugged(dev)) - return -ENODEV; + mutex_lock(&drm->struct_mutex); + ret = drm_gem_mmap_obj(obj, obj->size, vma); + mutex_unlock(&drm->struct_mutex); + if (ret) + return ret; - mutex_lock(&dev->struct_mutex); + return rockchip_drm_gem_object_mmap(obj, vma); +} - node = drm_vma_offset_exact_lookup(dev->vma_offset_manager, - vma->vm_pgoff, - vma_pages(vma)); - if (!node) { - mutex_unlock(&dev->struct_mutex); - DRM_ERROR("failed to find vma node.\n"); - return -EINVAL; - } else if (!drm_vma_node_is_allowed(node, filp)) { - mutex_unlock(&dev->struct_mutex); - return -EACCES; - } +/* drm driver mmap file operations */ +int rockchip_gem_mmap(struct file *filp, struct vm_area_struct *vma) +{ + struct drm_gem_object *obj; + int ret; - obj = container_of(node, struct drm_gem_object, vma_node); - ret = rockchip_gem_mmap_buf(obj, vma); + ret = drm_gem_mmap(filp, vma); + if (ret) + return ret; - mutex_unlock(&dev->struct_mutex); + obj = vma->vm_private_data; - return ret; + return rockchip_drm_gem_object_mmap(obj, vma); } struct rockchip_gem_object * diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c index dc65161..34b78e7 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c @@ -170,6 +170,7 @@ struct vop_win_phy { struct vop_reg enable; struct vop_reg format; + struct vop_reg rb_swap; struct vop_reg act_info; struct vop_reg dsp_info; struct vop_reg dsp_st; @@ -199,8 +200,12 @@ struct vop_data { static const uint32_t formats_01[] = { DRM_FORMAT_XRGB8888, DRM_FORMAT_ARGB8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_ABGR8888, DRM_FORMAT_RGB888, + DRM_FORMAT_BGR888, DRM_FORMAT_RGB565, + DRM_FORMAT_BGR565, DRM_FORMAT_NV12, DRM_FORMAT_NV16, DRM_FORMAT_NV24, @@ -209,8 +214,12 @@ static const uint32_t formats_01[] = { static const uint32_t formats_234[] = { DRM_FORMAT_XRGB8888, DRM_FORMAT_ARGB8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_ABGR8888, DRM_FORMAT_RGB888, + DRM_FORMAT_BGR888, DRM_FORMAT_RGB565, + DRM_FORMAT_BGR565, }; static const struct vop_win_phy win01_data = { @@ -218,6 +227,7 @@ static const struct vop_win_phy win01_data = { .nformats = ARRAY_SIZE(formats_01), .enable = VOP_REG(WIN0_CTRL0, 0x1, 0), .format = VOP_REG(WIN0_CTRL0, 0x7, 1), + .rb_swap = VOP_REG(WIN0_CTRL0, 0x1, 12), .act_info = VOP_REG(WIN0_ACT_INFO, 0x1fff1fff, 0), .dsp_info = VOP_REG(WIN0_DSP_INFO, 0x0fff0fff, 0), .dsp_st = VOP_REG(WIN0_DSP_ST, 0x1fff1fff, 0), @@ -234,6 +244,7 @@ static const struct vop_win_phy win23_data = { .nformats = ARRAY_SIZE(formats_234), .enable = VOP_REG(WIN2_CTRL0, 0x1, 0), .format = VOP_REG(WIN2_CTRL0, 0x7, 1), + .rb_swap = VOP_REG(WIN2_CTRL0, 0x1, 12), .dsp_info = VOP_REG(WIN2_DSP_INFO0, 0x0fff0fff, 0), .dsp_st = VOP_REG(WIN2_DSP_ST0, 0x1fff1fff, 0), .yrgb_mst = VOP_REG(WIN2_MST0, 0xffffffff, 0), @@ -242,15 +253,6 @@ static const struct vop_win_phy win23_data = { .dst_alpha_ctl = VOP_REG(WIN2_DST_ALPHA_CTRL, 0xff, 0), }; -static const struct vop_win_phy cursor_data = { - .data_formats = formats_234, - .nformats = ARRAY_SIZE(formats_234), - .enable = VOP_REG(HWC_CTRL0, 0x1, 0), - .format = VOP_REG(HWC_CTRL0, 0x7, 1), - .dsp_st = VOP_REG(HWC_DSP_ST, 0x1fff1fff, 0), - .yrgb_mst = VOP_REG(HWC_MST, 0xffffffff, 0), -}; - static const struct vop_ctrl ctrl_data = { .standby = VOP_REG(SYS_CTRL, 0x1, 22), .gate_en = VOP_REG(SYS_CTRL, 0x1, 23), @@ -282,14 +284,14 @@ static const struct vop_reg_data vop_init_reg_table[] = { /* * Note: rk3288 has a dedicated 'cursor' window, however, that window requires * special support to get alpha blending working. For now, just use overlay - * window 1 for the drm cursor. + * window 3 for the drm cursor. + * */ static const struct vop_win_data rk3288_vop_win_data[] = { { .base = 0x00, .phy = &win01_data, .type = DRM_PLANE_TYPE_PRIMARY }, - { .base = 0x40, .phy = &win01_data, .type = DRM_PLANE_TYPE_CURSOR }, + { .base = 0x40, .phy = &win01_data, .type = DRM_PLANE_TYPE_OVERLAY }, { .base = 0x00, .phy = &win23_data, .type = DRM_PLANE_TYPE_OVERLAY }, - { .base = 0x50, .phy = &win23_data, .type = DRM_PLANE_TYPE_OVERLAY }, - { .base = 0x00, .phy = &cursor_data, .type = DRM_PLANE_TYPE_OVERLAY }, + { .base = 0x50, .phy = &win23_data, .type = DRM_PLANE_TYPE_CURSOR }, }; static const struct vop_data rk3288_vop = { @@ -352,15 +354,32 @@ static inline void vop_mask_write_relaxed(struct vop *vop, uint32_t offset, } } +static bool has_rb_swapped(uint32_t format) +{ + switch (format) { + case DRM_FORMAT_XBGR8888: + case DRM_FORMAT_ABGR8888: + case DRM_FORMAT_BGR888: + case DRM_FORMAT_BGR565: + return true; + default: + return false; + } +} + static enum vop_data_format vop_convert_format(uint32_t format) { switch (format) { case DRM_FORMAT_XRGB8888: case DRM_FORMAT_ARGB8888: + case DRM_FORMAT_XBGR8888: + case DRM_FORMAT_ABGR8888: return VOP_FMT_ARGB8888; case DRM_FORMAT_RGB888: + case DRM_FORMAT_BGR888: return VOP_FMT_RGB888; case DRM_FORMAT_RGB565: + case DRM_FORMAT_BGR565: return VOP_FMT_RGB565; case DRM_FORMAT_NV12: return VOP_FMT_YUV420SP; @@ -378,6 +397,7 @@ static bool is_alpha_support(uint32_t format) { switch (format) { case DRM_FORMAT_ARGB8888: + case DRM_FORMAT_ABGR8888: return true; default: return false; @@ -588,6 +608,7 @@ static int vop_update_plane_event(struct drm_plane *plane, enum vop_data_format format; uint32_t val; bool is_alpha; + bool rb_swap; bool visible; int ret; struct drm_rect dest = { @@ -621,6 +642,7 @@ static int vop_update_plane_event(struct drm_plane *plane, return 0; is_alpha = is_alpha_support(fb->pixel_format); + rb_swap = has_rb_swapped(fb->pixel_format); format = vop_convert_format(fb->pixel_format); if (format < 0) return format; @@ -689,6 +711,7 @@ static int vop_update_plane_event(struct drm_plane *plane, val = (dsp_sty - 1) << 16; val |= (dsp_stx - 1) & 0xffff; VOP_WIN_SET(vop, win, dsp_st, val); + VOP_WIN_SET(vop, win, rb_swap, rb_swap); if (is_alpha) { VOP_WIN_SET(vop, win, dst_alpha_ctl, diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 882cccd..ac6fe40 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -490,7 +490,8 @@ pgprot_t ttm_io_prot(uint32_t caching_flags, pgprot_t tmp) else if (boot_cpu_data.x86 > 3) tmp = pgprot_noncached(tmp); #endif -#if defined(__ia64__) || defined(__arm__) || defined(__powerpc__) +#if defined(__ia64__) || defined(__arm__) || defined(__aarch64__) || \ + defined(__powerpc__) if (caching_flags & TTM_PL_FLAG_WC) tmp = pgprot_writecombine(tmp); else diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c index 3077f15..624d941 100644 --- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c +++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c @@ -963,14 +963,13 @@ void ttm_dma_unpopulate(struct ttm_dma_tt *ttm_dma, struct device *dev) } else { pool->npages_free += count; list_splice(&ttm_dma->pages_list, &pool->free_list); - npages = count; - if (pool->npages_free > _manager->options.max_size) { + /* + * Wait to have at at least NUM_PAGES_TO_ALLOC number of pages + * to free in order to minimize calls to set_memory_wb(). + */ + if (pool->npages_free >= (_manager->options.max_size + + NUM_PAGES_TO_ALLOC)) npages = pool->npages_free - _manager->options.max_size; - /* free at least NUM_PAGES_TO_ALLOC number of pages - * to reduce calls to set_memory_wb */ - if (npages < NUM_PAGES_TO_ALLOC) - npages = NUM_PAGES_TO_ALLOC; - } } spin_unlock_irqrestore(&pool->lock, irq_flags); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 654c8da..97ad3bc 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -2492,7 +2492,7 @@ int vmw_execbuf_process(struct drm_file *file_priv, ret = ttm_eu_reserve_buffers(&ticket, &sw_context->validate_nodes, true, NULL); if (unlikely(ret != 0)) - goto out_err; + goto out_err_nores; ret = vmw_validate_buffers(dev_priv, sw_context); if (unlikely(ret != 0)) @@ -2536,6 +2536,7 @@ int vmw_execbuf_process(struct drm_file *file_priv, vmw_resource_relocations_free(&sw_context->res_relocations); vmw_fifo_commit(dev_priv, command_size); + mutex_unlock(&dev_priv->binding_mutex); vmw_query_bo_switch_commit(dev_priv, sw_context); ret = vmw_execbuf_fence_commands(file_priv, dev_priv, @@ -2551,7 +2552,6 @@ int vmw_execbuf_process(struct drm_file *file_priv, DRM_ERROR("Fence submission error. Syncing.\n"); vmw_resource_list_unreserve(&sw_context->resource_list, false); - mutex_unlock(&dev_priv->binding_mutex); ttm_eu_fence_buffer_objects(&ticket, &sw_context->validate_nodes, (void *) fence); |