diff options
Diffstat (limited to 'drivers/gpu/drm')
99 files changed, 1279 insertions, 695 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/Kconfig b/drivers/gpu/drm/amd/amdgpu/Kconfig index 53cf3971dfc3..61360e27715f 100644 --- a/drivers/gpu/drm/amd/amdgpu/Kconfig +++ b/drivers/gpu/drm/amd/amdgpu/Kconfig @@ -32,5 +32,4 @@ config DRM_AMDGPU_GART_DEBUGFS Selecting this option creates a debugfs file to inspect the mapped pages. Uses more memory for housekeeping, enable only for debugging. -source "drivers/gpu/drm/amd/powerplay/Kconfig" source "drivers/gpu/drm/amd/acp/Kconfig" diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 9ec262d4b8a2..248a05d02917 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -111,14 +111,10 @@ amdgpu-$(CONFIG_VGA_SWITCHEROO) += amdgpu_atpx_handler.o amdgpu-$(CONFIG_ACPI) += amdgpu_acpi.o amdgpu-$(CONFIG_MMU_NOTIFIER) += amdgpu_mn.o -ifneq ($(CONFIG_DRM_AMD_POWERPLAY),) - include $(FULL_AMD_PATH)/powerplay/Makefile amdgpu-y += $(AMD_POWERPLAY_FILES) -endif - obj-$(CONFIG_DRM_AMDGPU)+= amdgpu.o CFLAGS_amdgpu_trace_points.o := -I$(src) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 235f3902643a..039b57e4644c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1943,6 +1943,7 @@ struct amdgpu_ip_block_status { bool valid; bool sw; bool hw; + bool late_initialized; bool hang; }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index 22c11e7698c8..e3281d4e3e41 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -765,12 +765,20 @@ amdgpu_connector_lvds_detect(struct drm_connector *connector, bool force) return ret; } -static void amdgpu_connector_destroy(struct drm_connector *connector) +static void amdgpu_connector_unregister(struct drm_connector *connector) { struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector); - if (amdgpu_connector->ddc_bus->has_aux) + if (amdgpu_connector->ddc_bus->has_aux) { drm_dp_aux_unregister(&amdgpu_connector->ddc_bus->aux); + amdgpu_connector->ddc_bus->has_aux = false; + } +} + +static void amdgpu_connector_destroy(struct drm_connector *connector) +{ + struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector); + amdgpu_connector_free_edid(connector); kfree(amdgpu_connector->con_priv); drm_connector_unregister(connector); @@ -824,6 +832,7 @@ static const struct drm_connector_funcs amdgpu_connector_lvds_funcs = { .dpms = drm_helper_connector_dpms, .detect = amdgpu_connector_lvds_detect, .fill_modes = drm_helper_probe_single_connector_modes, + .early_unregister = amdgpu_connector_unregister, .destroy = amdgpu_connector_destroy, .set_property = amdgpu_connector_set_lcd_property, }; @@ -934,6 +943,7 @@ static const struct drm_connector_funcs amdgpu_connector_vga_funcs = { .dpms = drm_helper_connector_dpms, .detect = amdgpu_connector_vga_detect, .fill_modes = drm_helper_probe_single_connector_modes, + .early_unregister = amdgpu_connector_unregister, .destroy = amdgpu_connector_destroy, .set_property = amdgpu_connector_set_property, }; @@ -1201,6 +1211,7 @@ static const struct drm_connector_funcs amdgpu_connector_dvi_funcs = { .detect = amdgpu_connector_dvi_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = amdgpu_connector_set_property, + .early_unregister = amdgpu_connector_unregister, .destroy = amdgpu_connector_destroy, .force = amdgpu_connector_dvi_force, }; @@ -1491,6 +1502,7 @@ static const struct drm_connector_funcs amdgpu_connector_dp_funcs = { .detect = amdgpu_connector_dp_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = amdgpu_connector_set_property, + .early_unregister = amdgpu_connector_unregister, .destroy = amdgpu_connector_destroy, .force = amdgpu_connector_dvi_force, }; @@ -1500,6 +1512,7 @@ static const struct drm_connector_funcs amdgpu_connector_edp_funcs = { .detect = amdgpu_connector_dp_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = amdgpu_connector_set_lcd_property, + .early_unregister = amdgpu_connector_unregister, .destroy = amdgpu_connector_destroy, .force = amdgpu_connector_dvi_force, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index e203e5561107..a5e2fcbef0f0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -43,6 +43,9 @@ static int amdgpu_ctx_init(struct amdgpu_device *adev, struct amdgpu_ctx *ctx) ctx->rings[i].sequence = 1; ctx->rings[i].fences = &ctx->fences[amdgpu_sched_jobs * i]; } + + ctx->reset_counter = atomic_read(&adev->gpu_reset_counter); + /* create context entity for each ring */ for (i = 0; i < adev->num_rings; i++) { struct amdgpu_ring *ring = adev->rings[i]; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 99a15cad6789..b4f4a9239069 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -1408,22 +1408,25 @@ static int amdgpu_late_init(struct amdgpu_device *adev) for (i = 0; i < adev->num_ip_blocks; i++) { if (!adev->ip_block_status[i].valid) continue; - if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_UVD || - adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_VCE) - continue; - /* enable clockgating to save power */ - r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev, - AMD_CG_STATE_GATE); - if (r) { - DRM_ERROR("set_clockgating_state(gate) of IP block <%s> failed %d\n", adev->ip_blocks[i].funcs->name, r); - return r; - } if (adev->ip_blocks[i].funcs->late_init) { r = adev->ip_blocks[i].funcs->late_init((void *)adev); if (r) { DRM_ERROR("late_init of IP block <%s> failed %d\n", adev->ip_blocks[i].funcs->name, r); return r; } + adev->ip_block_status[i].late_initialized = true; + } + /* skip CG for VCE/UVD, it's handled specially */ + if (adev->ip_blocks[i].type != AMD_IP_BLOCK_TYPE_UVD && + adev->ip_blocks[i].type != AMD_IP_BLOCK_TYPE_VCE) { + /* enable clockgating to save power */ + r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev, + AMD_CG_STATE_GATE); + if (r) { + DRM_ERROR("set_clockgating_state(gate) of IP block <%s> failed %d\n", + adev->ip_blocks[i].funcs->name, r); + return r; + } } } @@ -1434,6 +1437,30 @@ static int amdgpu_fini(struct amdgpu_device *adev) { int i, r; + /* need to disable SMC first */ + for (i = 0; i < adev->num_ip_blocks; i++) { + if (!adev->ip_block_status[i].hw) + continue; + if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_SMC) { + /* ungate blocks before hw fini so that we can shutdown the blocks safely */ + r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev, + AMD_CG_STATE_UNGATE); + if (r) { + DRM_ERROR("set_clockgating_state(ungate) of IP block <%s> failed %d\n", + adev->ip_blocks[i].funcs->name, r); + return r; + } + r = adev->ip_blocks[i].funcs->hw_fini((void *)adev); + /* XXX handle errors */ + if (r) { + DRM_DEBUG("hw_fini of IP block <%s> failed %d\n", + adev->ip_blocks[i].funcs->name, r); + } + adev->ip_block_status[i].hw = false; + break; + } + } + for (i = adev->num_ip_blocks - 1; i >= 0; i--) { if (!adev->ip_block_status[i].hw) continue; @@ -1469,8 +1496,11 @@ static int amdgpu_fini(struct amdgpu_device *adev) } for (i = adev->num_ip_blocks - 1; i >= 0; i--) { + if (!adev->ip_block_status[i].late_initialized) + continue; if (adev->ip_blocks[i].funcs->late_fini) adev->ip_blocks[i].funcs->late_fini((void *)adev); + adev->ip_block_status[i].late_initialized = false; } return 0; @@ -1822,11 +1852,11 @@ void amdgpu_device_fini(struct amdgpu_device *adev) DRM_INFO("amdgpu: finishing device.\n"); adev->shutdown = true; + drm_crtc_force_disable_all(adev->ddev); /* evict vram memory */ amdgpu_bo_evict_vram(adev); amdgpu_ib_pool_fini(adev); amdgpu_fence_driver_fini(adev); - drm_crtc_force_disable_all(adev->ddev); amdgpu_fbdev_fini(adev); r = amdgpu_fini(adev); kfree(adev->ip_block_status); @@ -2069,7 +2099,8 @@ static bool amdgpu_check_soft_reset(struct amdgpu_device *adev) if (!adev->ip_block_status[i].valid) continue; if (adev->ip_blocks[i].funcs->check_soft_reset) - adev->ip_blocks[i].funcs->check_soft_reset(adev); + adev->ip_block_status[i].hang = + adev->ip_blocks[i].funcs->check_soft_reset(adev); if (adev->ip_block_status[i].hang) { DRM_INFO("IP block:%d is hang!\n", i); asic_hang = true; @@ -2098,12 +2129,20 @@ static int amdgpu_pre_soft_reset(struct amdgpu_device *adev) static bool amdgpu_need_full_reset(struct amdgpu_device *adev) { - if (adev->ip_block_status[AMD_IP_BLOCK_TYPE_GMC].hang || - adev->ip_block_status[AMD_IP_BLOCK_TYPE_SMC].hang || - adev->ip_block_status[AMD_IP_BLOCK_TYPE_ACP].hang || - adev->ip_block_status[AMD_IP_BLOCK_TYPE_DCE].hang) { - DRM_INFO("Some block need full reset!\n"); - return true; + int i; + + for (i = 0; i < adev->num_ip_blocks; i++) { + if (!adev->ip_block_status[i].valid) + continue; + if ((adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) || + (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_SMC) || + (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_ACP) || + (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_DCE)) { + if (adev->ip_block_status[i].hang) { + DRM_INFO("Some block need full reset!\n"); + return true; + } + } } return false; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c index fe36caf1b7d7..14f57d9915e3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.c @@ -113,24 +113,26 @@ void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev, printk("\n"); } + u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev) { struct drm_device *dev = adev->ddev; struct drm_crtc *crtc; struct amdgpu_crtc *amdgpu_crtc; - u32 line_time_us, vblank_lines; + u32 vblank_in_pixels; u32 vblank_time_us = 0xffffffff; /* if the displays are off, vblank time is max */ if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) { list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { amdgpu_crtc = to_amdgpu_crtc(crtc); if (crtc->enabled && amdgpu_crtc->enabled && amdgpu_crtc->hw_mode.clock) { - line_time_us = (amdgpu_crtc->hw_mode.crtc_htotal * 1000) / - amdgpu_crtc->hw_mode.clock; - vblank_lines = amdgpu_crtc->hw_mode.crtc_vblank_end - + vblank_in_pixels = + amdgpu_crtc->hw_mode.crtc_htotal * + (amdgpu_crtc->hw_mode.crtc_vblank_end - amdgpu_crtc->hw_mode.crtc_vdisplay + - (amdgpu_crtc->v_border * 2); - vblank_time_us = vblank_lines * line_time_us; + (amdgpu_crtc->v_border * 2)); + + vblank_time_us = vblank_in_pixels * 1000 / amdgpu_crtc->hw_mode.clock; break; } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index dbe89fb25694..71ed27eb3dde 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -174,7 +174,6 @@ module_param_named(sched_jobs, amdgpu_sched_jobs, int, 0444); MODULE_PARM_DESC(sched_hw_submission, "the max number of HW submissions (default 2)"); module_param_named(sched_hw_submission, amdgpu_sched_hw_submission, int, 0444); -#ifdef CONFIG_DRM_AMD_POWERPLAY MODULE_PARM_DESC(powerplay, "Powerplay component (1 = enable, 0 = disable, -1 = auto (default))"); module_param_named(powerplay, amdgpu_powerplay, int, 0444); @@ -183,7 +182,6 @@ module_param_named(powercontainment, amdgpu_powercontainment, int, 0444); MODULE_PARM_DESC(ppfeaturemask, "all power features enabled (default))"); module_param_named(ppfeaturemask, amdgpu_pp_feature_mask, int, 0444); -#endif MODULE_PARM_DESC(sclkdeepsleep, "SCLK Deep Sleep (1 = enable (default), 0 = disable)"); module_param_named(sclkdeepsleep, amdgpu_sclk_deep_sleep_en, int, 0444); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c index 34bab616588c..91d367399956 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c @@ -220,6 +220,7 @@ void amdgpu_i2c_destroy(struct amdgpu_i2c_chan *i2c) { if (!i2c) return; + WARN_ON(i2c->has_aux); i2c_del_adapter(&i2c->adapter); kfree(i2c); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c index 68ad24101a36..7532ff822aa7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c @@ -42,7 +42,6 @@ static int amdgpu_powerplay_init(struct amdgpu_device *adev) amd_pp = &(adev->powerplay); if (adev->pp_enabled) { -#ifdef CONFIG_DRM_AMD_POWERPLAY struct amd_pp_init *pp_init; pp_init = kzalloc(sizeof(struct amd_pp_init), GFP_KERNEL); @@ -55,7 +54,6 @@ static int amdgpu_powerplay_init(struct amdgpu_device *adev) pp_init->device = amdgpu_cgs_create_device(adev); ret = amd_powerplay_init(pp_init, amd_pp); kfree(pp_init); -#endif } else { amd_pp->pp_handle = (void *)adev; @@ -97,7 +95,6 @@ static int amdgpu_pp_early_init(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; int ret = 0; -#ifdef CONFIG_DRM_AMD_POWERPLAY switch (adev->asic_type) { case CHIP_POLARIS11: case CHIP_POLARIS10: @@ -120,9 +117,6 @@ static int amdgpu_pp_early_init(void *handle) adev->pp_enabled = false; break; } -#else - adev->pp_enabled = false; -#endif ret = amdgpu_powerplay_init(adev); if (ret) @@ -144,12 +138,11 @@ static int amdgpu_pp_late_init(void *handle) ret = adev->powerplay.ip_funcs->late_init( adev->powerplay.pp_handle); -#ifdef CONFIG_DRM_AMD_POWERPLAY if (adev->pp_enabled && adev->pm.dpm_enabled) { amdgpu_pm_sysfs_init(adev); amdgpu_dpm_dispatch_task(adev, AMD_PP_EVENT_COMPLETE_INIT, NULL, NULL); } -#endif + return ret; } @@ -162,10 +155,8 @@ static int amdgpu_pp_sw_init(void *handle) ret = adev->powerplay.ip_funcs->sw_init( adev->powerplay.pp_handle); -#ifdef CONFIG_DRM_AMD_POWERPLAY if (adev->pp_enabled) adev->pm.dpm_enabled = true; -#endif return ret; } @@ -216,7 +207,6 @@ static int amdgpu_pp_hw_fini(void *handle) static void amdgpu_pp_late_fini(void *handle) { -#ifdef CONFIG_DRM_AMD_POWERPLAY struct amdgpu_device *adev = (struct amdgpu_device *)handle; if (adev->pp_enabled) { @@ -227,7 +217,6 @@ static void amdgpu_pp_late_fini(void *handle) if (adev->powerplay.ip_funcs->late_fini) adev->powerplay.ip_funcs->late_fini( adev->powerplay.pp_handle); -#endif } static int amdgpu_pp_suspend(void *handle) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index e1fa8731d1e2..3cb5e903cd62 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -345,8 +345,8 @@ static int amdgpu_debugfs_ring_init(struct amdgpu_device *adev, ent = debugfs_create_file(name, S_IFREG | S_IRUGO, root, ring, &amdgpu_debugfs_ring_fops); - if (IS_ERR(ent)) - return PTR_ERR(ent); + if (!ent) + return -ENOMEM; i_size_write(ent->d_inode, ring->ring_size + 12); ring->ent = ent; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 887483b8b818..dcaf691f56b5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -555,10 +555,13 @@ struct amdgpu_ttm_tt { int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages) { struct amdgpu_ttm_tt *gtt = (void *)ttm; - int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY); + unsigned int flags = 0; unsigned pinned = 0; int r; + if (!(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY)) + flags |= FOLL_WRITE; + if (gtt->userflags & AMDGPU_GEM_USERPTR_ANONONLY) { /* check that we only use anonymous memory to prevent problems with writeback */ @@ -581,7 +584,7 @@ int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages) list_add(&guptask.list, >t->guptasks); spin_unlock(>t->guptasklock); - r = get_user_pages(userptr, num_pages, write, 0, p, NULL); + r = get_user_pages(userptr, num_pages, flags, p, NULL); spin_lock(>t->guptasklock); list_del(&guptask.list); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index 7a05f79818f1..cb3d252f3c78 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -273,7 +273,6 @@ int amdgpu_ucode_init_bo(struct amdgpu_device *adev) amdgpu_bo_unreserve(*bo); - fw_offset = 0; for (i = 0; i < AMDGPU_UCODE_ID_MAXIMUM; i++) { ucode = &adev->firmware.ucode[i]; if (ucode->fw) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index 3b03558ddb01..7fe8fd884f06 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -699,6 +699,20 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx) case 0x05000009: /* clock table */ break; + case 0x0500000c: /* hw config */ + switch (p->adev->asic_type) { +#ifdef CONFIG_DRM_AMDGPU_CIK + case CHIP_KAVERI: + case CHIP_MULLINS: +#endif + case CHIP_CARRIZO: + break; + default: + r = -EINVAL; + goto out; + } + break; + case 0x03000001: /* encode */ r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 10, idx + 9, *size, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index bc4b22c6fc08..06f24322e7c3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -878,13 +878,13 @@ static void amdgpu_vm_frag_ptes(struct amdgpu_pte_update_params *params, * allocation size to the fragment size. */ - const uint64_t frag_align = 1 << AMDGPU_LOG2_PAGES_PER_FRAG; + /* SI and newer are optimized for 64KB */ + uint64_t frag_flags = AMDGPU_PTE_FRAG(AMDGPU_LOG2_PAGES_PER_FRAG); + uint64_t frag_align = 1 << AMDGPU_LOG2_PAGES_PER_FRAG; uint64_t frag_start = ALIGN(start, frag_align); uint64_t frag_end = end & ~(frag_align - 1); - uint32_t frag; - /* system pages are non continuously */ if (params->src || !(flags & AMDGPU_PTE_VALID) || (frag_start >= frag_end)) { @@ -893,10 +893,6 @@ static void amdgpu_vm_frag_ptes(struct amdgpu_pte_update_params *params, return; } - /* use more than 64KB fragment size if possible */ - frag = lower_32_bits(frag_start | frag_end); - frag = likely(frag) ? __ffs(frag) : 31; - /* handle the 4K area at the beginning */ if (start != frag_start) { amdgpu_vm_update_ptes(params, vm, start, frag_start, @@ -906,7 +902,7 @@ static void amdgpu_vm_frag_ptes(struct amdgpu_pte_update_params *params, /* handle the area in the middle */ amdgpu_vm_update_ptes(params, vm, frag_start, frag_end, dst, - flags | AMDGPU_PTE_FRAG(frag)); + flags | frag_flags); /* handle the 4K area at the end */ if (frag_end != end) { diff --git a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c index f80a0834e889..3c082e143730 100644 --- a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c @@ -1514,14 +1514,16 @@ static int cz_dpm_set_powergating_state(void *handle, return 0; } -/* borrowed from KV, need future unify */ static int cz_dpm_get_temperature(struct amdgpu_device *adev) { int actual_temp = 0; - uint32_t temp = RREG32_SMC(0xC0300E0C); + uint32_t val = RREG32_SMC(ixTHM_TCON_CUR_TMP); + uint32_t temp = REG_GET_FIELD(val, THM_TCON_CUR_TMP, CUR_TEMP); - if (temp) + if (REG_GET_FIELD(val, THM_TCON_CUR_TMP, CUR_TEMP_RANGE_SEL)) actual_temp = 1000 * ((temp / 8) - 49); + else + actual_temp = 1000 * (temp / 8); return actual_temp; } diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 613ebb7ed50f..4108c686aa7c 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -3188,16 +3188,11 @@ static int dce_v10_0_wait_for_idle(void *handle) return 0; } -static int dce_v10_0_check_soft_reset(void *handle) +static bool dce_v10_0_check_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (dce_v10_0_is_display_hung(adev)) - adev->ip_block_status[AMD_IP_BLOCK_TYPE_DCE].hang = true; - else - adev->ip_block_status[AMD_IP_BLOCK_TYPE_DCE].hang = false; - - return 0; + return dce_v10_0_is_display_hung(adev); } static int dce_v10_0_soft_reset(void *handle) @@ -3205,9 +3200,6 @@ static int dce_v10_0_soft_reset(void *handle) u32 srbm_soft_reset = 0, tmp; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_DCE].hang) - return 0; - if (dce_v10_0_is_display_hung(adev)) srbm_soft_reset |= SRBM_SOFT_RESET__SOFT_RESET_DC_MASK; diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index 678f5eb6cbc2..f264b8f17ad1 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -3159,6 +3159,7 @@ static int dce_v11_0_sw_fini(void *handle) dce_v11_0_afmt_fini(adev); + drm_mode_config_cleanup(adev->ddev); adev->mode_info.mode_config_initialized = false; return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c index a754f2522ba2..c2bd9f045532 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_virtual.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_virtual.c @@ -99,15 +99,15 @@ static void dce_virtual_stop_mc_access(struct amdgpu_device *adev, struct amdgpu_mode_mc_save *save) { switch (adev->asic_type) { +#ifdef CONFIG_DRM_AMDGPU_CIK case CHIP_BONAIRE: case CHIP_HAWAII: case CHIP_KAVERI: case CHIP_KABINI: case CHIP_MULLINS: -#ifdef CONFIG_DRM_AMDGPU_CIK dce_v8_0_disable_dce(adev); -#endif break; +#endif case CHIP_FIJI: case CHIP_TONGA: dce_v10_0_disable_dce(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 6c6ff57b1c95..ee6a48a09214 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -4087,14 +4087,21 @@ static int gfx_v8_0_rlc_load_microcode(struct amdgpu_device *adev) static int gfx_v8_0_rlc_resume(struct amdgpu_device *adev) { int r; + u32 tmp; gfx_v8_0_rlc_stop(adev); /* disable CG */ - WREG32(mmRLC_CGCG_CGLS_CTRL, 0); + tmp = RREG32(mmRLC_CGCG_CGLS_CTRL); + tmp &= ~(RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK | + RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK); + WREG32(mmRLC_CGCG_CGLS_CTRL, tmp); if (adev->asic_type == CHIP_POLARIS11 || - adev->asic_type == CHIP_POLARIS10) - WREG32(mmRLC_CGCG_CGLS_CTRL_3D, 0); + adev->asic_type == CHIP_POLARIS10) { + tmp = RREG32(mmRLC_CGCG_CGLS_CTRL_3D); + tmp &= ~0x3; + WREG32(mmRLC_CGCG_CGLS_CTRL_3D, tmp); + } /* disable PG */ WREG32(mmRLC_PG_CNTL, 0); @@ -5137,7 +5144,7 @@ static int gfx_v8_0_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int gfx_v8_0_check_soft_reset(void *handle) +static bool gfx_v8_0_check_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 grbm_soft_reset = 0, srbm_soft_reset = 0; @@ -5189,16 +5196,14 @@ static int gfx_v8_0_check_soft_reset(void *handle) SRBM_SOFT_RESET, SOFT_RESET_SEM, 1); if (grbm_soft_reset || srbm_soft_reset) { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_GFX].hang = true; adev->gfx.grbm_soft_reset = grbm_soft_reset; adev->gfx.srbm_soft_reset = srbm_soft_reset; + return true; } else { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_GFX].hang = false; adev->gfx.grbm_soft_reset = 0; adev->gfx.srbm_soft_reset = 0; + return false; } - - return 0; } static void gfx_v8_0_inactive_hqd(struct amdgpu_device *adev, @@ -5226,7 +5231,8 @@ static int gfx_v8_0_pre_soft_reset(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 grbm_soft_reset = 0, srbm_soft_reset = 0; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_GFX].hang) + if ((!adev->gfx.grbm_soft_reset) && + (!adev->gfx.srbm_soft_reset)) return 0; grbm_soft_reset = adev->gfx.grbm_soft_reset; @@ -5264,7 +5270,8 @@ static int gfx_v8_0_soft_reset(void *handle) u32 grbm_soft_reset = 0, srbm_soft_reset = 0; u32 tmp; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_GFX].hang) + if ((!adev->gfx.grbm_soft_reset) && + (!adev->gfx.srbm_soft_reset)) return 0; grbm_soft_reset = adev->gfx.grbm_soft_reset; @@ -5334,7 +5341,8 @@ static int gfx_v8_0_post_soft_reset(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 grbm_soft_reset = 0, srbm_soft_reset = 0; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_GFX].hang) + if ((!adev->gfx.grbm_soft_reset) && + (!adev->gfx.srbm_soft_reset)) return 0; grbm_soft_reset = adev->gfx.grbm_soft_reset; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 1b319f5bc696..c22ef140a542 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -1099,7 +1099,7 @@ static int gmc_v8_0_wait_for_idle(void *handle) } -static int gmc_v8_0_check_soft_reset(void *handle) +static bool gmc_v8_0_check_soft_reset(void *handle) { u32 srbm_soft_reset = 0; struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -1116,20 +1116,19 @@ static int gmc_v8_0_check_soft_reset(void *handle) SRBM_SOFT_RESET, SOFT_RESET_MC, 1); } if (srbm_soft_reset) { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_GMC].hang = true; adev->mc.srbm_soft_reset = srbm_soft_reset; + return true; } else { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_GMC].hang = false; adev->mc.srbm_soft_reset = 0; + return false; } - return 0; } static int gmc_v8_0_pre_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_GMC].hang) + if (!adev->mc.srbm_soft_reset) return 0; gmc_v8_0_mc_stop(adev, &adev->mc.save); @@ -1145,7 +1144,7 @@ static int gmc_v8_0_soft_reset(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_GMC].hang) + if (!adev->mc.srbm_soft_reset) return 0; srbm_soft_reset = adev->mc.srbm_soft_reset; @@ -1175,7 +1174,7 @@ static int gmc_v8_0_post_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_GMC].hang) + if (!adev->mc.srbm_soft_reset) return 0; gmc_v8_0_mc_resume(adev, &adev->mc.save); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c index f325fd86430b..a9d10941fb53 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c @@ -1268,7 +1268,7 @@ static int sdma_v3_0_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int sdma_v3_0_check_soft_reset(void *handle) +static bool sdma_v3_0_check_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset = 0; @@ -1281,14 +1281,12 @@ static int sdma_v3_0_check_soft_reset(void *handle) } if (srbm_soft_reset) { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_SDMA].hang = true; adev->sdma.srbm_soft_reset = srbm_soft_reset; + return true; } else { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_SDMA].hang = false; adev->sdma.srbm_soft_reset = 0; + return false; } - - return 0; } static int sdma_v3_0_pre_soft_reset(void *handle) @@ -1296,7 +1294,7 @@ static int sdma_v3_0_pre_soft_reset(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset = 0; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_SDMA].hang) + if (!adev->sdma.srbm_soft_reset) return 0; srbm_soft_reset = adev->sdma.srbm_soft_reset; @@ -1315,7 +1313,7 @@ static int sdma_v3_0_post_soft_reset(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset = 0; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_SDMA].hang) + if (!adev->sdma.srbm_soft_reset) return 0; srbm_soft_reset = adev->sdma.srbm_soft_reset; @@ -1335,7 +1333,7 @@ static int sdma_v3_0_soft_reset(void *handle) u32 srbm_soft_reset = 0; u32 tmp; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_SDMA].hang) + if (!adev->sdma.srbm_soft_reset) return 0; srbm_soft_reset = adev->sdma.srbm_soft_reset; diff --git a/drivers/gpu/drm/amd/amdgpu/si_dpm.c b/drivers/gpu/drm/amd/amdgpu/si_dpm.c index 8bd08925b370..3de7bca5854b 100644 --- a/drivers/gpu/drm/amd/amdgpu/si_dpm.c +++ b/drivers/gpu/drm/amd/amdgpu/si_dpm.c @@ -3499,6 +3499,12 @@ static void si_apply_state_adjust_rules(struct amdgpu_device *adev, max_sclk = 75000; max_mclk = 80000; } + /* Limit clocks for some HD8600 parts */ + if (adev->pdev->device == 0x6660 && + adev->pdev->revision == 0x83) { + max_sclk = 75000; + max_mclk = 80000; + } if (rps->vce_active) { rps->evclk = adev->pm.dpm.vce_states[adev->pm.dpm.vce_level].evclk; diff --git a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c index d127d59f953a..b4ea229bb449 100644 --- a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c @@ -373,7 +373,7 @@ static int tonga_ih_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int tonga_ih_check_soft_reset(void *handle) +static bool tonga_ih_check_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset = 0; @@ -384,21 +384,19 @@ static int tonga_ih_check_soft_reset(void *handle) SOFT_RESET_IH, 1); if (srbm_soft_reset) { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_IH].hang = true; adev->irq.srbm_soft_reset = srbm_soft_reset; + return true; } else { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_IH].hang = false; adev->irq.srbm_soft_reset = 0; + return false; } - - return 0; } static int tonga_ih_pre_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_IH].hang) + if (!adev->irq.srbm_soft_reset) return 0; return tonga_ih_hw_fini(adev); @@ -408,7 +406,7 @@ static int tonga_ih_post_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_IH].hang) + if (!adev->irq.srbm_soft_reset) return 0; return tonga_ih_hw_init(adev); @@ -419,7 +417,7 @@ static int tonga_ih_soft_reset(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_IH].hang) + if (!adev->irq.srbm_soft_reset) return 0; srbm_soft_reset = adev->irq.srbm_soft_reset; diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c index e0fd9f21ed95..ab3df6d75656 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c @@ -770,7 +770,7 @@ static int uvd_v6_0_wait_for_idle(void *handle) } #define AMDGPU_UVD_STATUS_BUSY_MASK 0xfd -static int uvd_v6_0_check_soft_reset(void *handle) +static bool uvd_v6_0_check_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset = 0; @@ -782,19 +782,19 @@ static int uvd_v6_0_check_soft_reset(void *handle) srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_UVD, 1); if (srbm_soft_reset) { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_UVD].hang = true; adev->uvd.srbm_soft_reset = srbm_soft_reset; + return true; } else { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_UVD].hang = false; adev->uvd.srbm_soft_reset = 0; + return false; } - return 0; } + static int uvd_v6_0_pre_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_UVD].hang) + if (!adev->uvd.srbm_soft_reset) return 0; uvd_v6_0_stop(adev); @@ -806,7 +806,7 @@ static int uvd_v6_0_soft_reset(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_UVD].hang) + if (!adev->uvd.srbm_soft_reset) return 0; srbm_soft_reset = adev->uvd.srbm_soft_reset; @@ -836,7 +836,7 @@ static int uvd_v6_0_post_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_UVD].hang) + if (!adev->uvd.srbm_soft_reset) return 0; mdelay(5); diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c index 3f6db4ec0102..8533269ec160 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c @@ -561,7 +561,7 @@ static int vce_v3_0_wait_for_idle(void *handle) #define AMDGPU_VCE_STATUS_BUSY_MASK (VCE_STATUS_VCPU_REPORT_AUTO_BUSY_MASK | \ VCE_STATUS_VCPU_REPORT_RB0_BUSY_MASK) -static int vce_v3_0_check_soft_reset(void *handle) +static bool vce_v3_0_check_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset = 0; @@ -591,16 +591,15 @@ static int vce_v3_0_check_soft_reset(void *handle) srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_VCE1, 1); } WREG32_FIELD(GRBM_GFX_INDEX, INSTANCE_INDEX, 0); + mutex_unlock(&adev->grbm_idx_mutex); if (srbm_soft_reset) { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_VCE].hang = true; adev->vce.srbm_soft_reset = srbm_soft_reset; + return true; } else { - adev->ip_block_status[AMD_IP_BLOCK_TYPE_VCE].hang = false; adev->vce.srbm_soft_reset = 0; + return false; } - mutex_unlock(&adev->grbm_idx_mutex); - return 0; } static int vce_v3_0_soft_reset(void *handle) @@ -608,7 +607,7 @@ static int vce_v3_0_soft_reset(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; u32 srbm_soft_reset; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_VCE].hang) + if (!adev->vce.srbm_soft_reset) return 0; srbm_soft_reset = adev->vce.srbm_soft_reset; @@ -638,7 +637,7 @@ static int vce_v3_0_pre_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_VCE].hang) + if (!adev->vce.srbm_soft_reset) return 0; mdelay(5); @@ -651,7 +650,7 @@ static int vce_v3_0_post_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (!adev->ip_block_status[AMD_IP_BLOCK_TYPE_VCE].hang) + if (!adev->vce.srbm_soft_reset) return 0; mdelay(5); diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index c934b78c9e2f..bec8125bceb0 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -165,7 +165,7 @@ struct amd_ip_funcs { /* poll for idle */ int (*wait_for_idle)(void *handle); /* check soft reset the IP block */ - int (*check_soft_reset)(void *handle); + bool (*check_soft_reset)(void *handle); /* pre soft reset the IP block */ int (*pre_soft_reset)(void *handle); /* soft reset the IP block */ diff --git a/drivers/gpu/drm/amd/powerplay/Kconfig b/drivers/gpu/drm/amd/powerplay/Kconfig deleted file mode 100644 index af380335b425..000000000000 --- a/drivers/gpu/drm/amd/powerplay/Kconfig +++ /dev/null @@ -1,6 +0,0 @@ -config DRM_AMD_POWERPLAY - bool "Enable AMD powerplay component" - depends on DRM_AMDGPU - default n - help - select this option will enable AMD powerplay component. diff --git a/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c b/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c index 92b117843875..8cee4e0f9fde 100644 --- a/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c +++ b/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c @@ -49,6 +49,7 @@ static const pem_event_action * const uninitialize_event[] = { uninitialize_display_phy_access_tasks, disable_gfx_voltage_island_power_gating_tasks, disable_gfx_clock_gating_tasks, + uninitialize_thermal_controller_tasks, set_boot_state_tasks, adjust_power_state_tasks, disable_dynamic_state_management_tasks, diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c index 7e4fcbbbe086..960424913496 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c @@ -1785,6 +1785,21 @@ static int cz_get_max_high_clocks(struct pp_hwmgr *hwmgr, struct amd_pp_simple_c return 0; } +static int cz_thermal_get_temperature(struct pp_hwmgr *hwmgr) +{ + int actual_temp = 0; + uint32_t val = cgs_read_ind_register(hwmgr->device, + CGS_IND_REG__SMC, ixTHM_TCON_CUR_TMP); + uint32_t temp = PHM_GET_FIELD(val, THM_TCON_CUR_TMP, CUR_TEMP); + + if (PHM_GET_FIELD(val, THM_TCON_CUR_TMP, CUR_TEMP_RANGE_SEL)) + actual_temp = ((temp / 8) - 49) * PP_TEMPERATURE_UNITS_PER_CENTIGRADES; + else + actual_temp = (temp / 8) * PP_TEMPERATURE_UNITS_PER_CENTIGRADES; + + return actual_temp; +} + static int cz_read_sensor(struct pp_hwmgr *hwmgr, int idx, int32_t *value) { struct cz_hwmgr *cz_hwmgr = (struct cz_hwmgr *)(hwmgr->backend); @@ -1881,6 +1896,9 @@ static int cz_read_sensor(struct pp_hwmgr *hwmgr, int idx, int32_t *value) case AMDGPU_PP_SENSOR_VCE_POWER: *value = cz_hwmgr->vce_power_gated ? 0 : 1; return 0; + case AMDGPU_PP_SENSOR_GPU_TEMP: + *value = cz_thermal_get_temperature(hwmgr); + return 0; default: return -EINVAL; } diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c index 508245d49d33..609996c84ad5 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c @@ -1030,20 +1030,19 @@ static int smu7_disable_sclk_mclk_dpm(struct pp_hwmgr *hwmgr) struct smu7_hwmgr *data = (struct smu7_hwmgr *)(hwmgr->backend); /* disable SCLK dpm */ - if (!data->sclk_dpm_key_disabled) - PP_ASSERT_WITH_CODE( - (smum_send_msg_to_smc(hwmgr->smumgr, - PPSMC_MSG_DPM_Disable) == 0), - "Failed to disable SCLK DPM!", - return -EINVAL); + if (!data->sclk_dpm_key_disabled) { + PP_ASSERT_WITH_CODE(true == smum_is_dpm_running(hwmgr), + "Trying to disable SCLK DPM when DPM is disabled", + return 0); + smum_send_msg_to_smc(hwmgr->smumgr, PPSMC_MSG_DPM_Disable); + } /* disable MCLK dpm */ if (!data->mclk_dpm_key_disabled) { - PP_ASSERT_WITH_CODE( - (smum_send_msg_to_smc(hwmgr->smumgr, - PPSMC_MSG_MCLKDPM_Disable) == 0), - "Failed to disable MCLK DPM!", - return -EINVAL); + PP_ASSERT_WITH_CODE(true == smum_is_dpm_running(hwmgr), + "Trying to disable MCLK DPM when DPM is disabled", + return 0); + smum_send_msg_to_smc(hwmgr->smumgr, PPSMC_MSG_MCLKDPM_Disable); } return 0; @@ -1069,10 +1068,13 @@ static int smu7_stop_dpm(struct pp_hwmgr *hwmgr) return -EINVAL); } - if (smu7_disable_sclk_mclk_dpm(hwmgr)) { - printk(KERN_ERR "Failed to disable Sclk DPM and Mclk DPM!"); - return -EINVAL; - } + smu7_disable_sclk_mclk_dpm(hwmgr); + + PP_ASSERT_WITH_CODE(true == smum_is_dpm_running(hwmgr), + "Trying to disable voltage DPM when DPM is disabled", + return 0); + + smum_send_msg_to_smc(hwmgr->smumgr, PPSMC_MSG_Voltage_Cntl_Disable); return 0; } @@ -1226,7 +1228,7 @@ int smu7_enable_dpm_tasks(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE((0 == tmp_result), "Failed to enable VR hot GPIO interrupt!", result = tmp_result); - smum_send_msg_to_smc(hwmgr->smumgr, (PPSMC_Msg)PPSMC_HasDisplay); + smum_send_msg_to_smc(hwmgr->smumgr, (PPSMC_Msg)PPSMC_NoDisplay); tmp_result = smu7_enable_sclk_control(hwmgr); PP_ASSERT_WITH_CODE((0 == tmp_result), @@ -1306,6 +1308,12 @@ int smu7_disable_dpm_tasks(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE((tmp_result == 0), "Failed to disable thermal auto throttle!", result = tmp_result); + if (1 == PHM_READ_VFPF_INDIRECT_FIELD(hwmgr->device, CGS_IND_REG__SMC, FEATURE_STATUS, AVS_ON)) { + PP_ASSERT_WITH_CODE((0 == smum_send_msg_to_smc(hwmgr->smumgr, PPSMC_MSG_DisableAvfs)), + "Failed to disable AVFS!", + return -EINVAL); + } + tmp_result = smu7_stop_dpm(hwmgr); PP_ASSERT_WITH_CODE((tmp_result == 0), "Failed to stop DPM!", result = tmp_result); @@ -1452,8 +1460,10 @@ static int smu7_get_evv_voltages(struct pp_hwmgr *hwmgr) struct phm_ppt_v1_clock_voltage_dependency_table *sclk_table = NULL; - if (table_info != NULL) - sclk_table = table_info->vdd_dep_on_sclk; + if (table_info == NULL) + return -EINVAL; + + sclk_table = table_info->vdd_dep_on_sclk; for (i = 0; i < SMU7_MAX_LEAKAGE_COUNT; i++) { vv_id = ATOM_VIRTUAL_VOLTAGE_ID0 + i; @@ -3802,13 +3812,15 @@ static inline bool smu7_are_power_levels_equal(const struct smu7_performance_lev int smu7_check_states_equal(struct pp_hwmgr *hwmgr, const struct pp_hw_power_state *pstate1, const struct pp_hw_power_state *pstate2, bool *equal) { - const struct smu7_power_state *psa = cast_const_phw_smu7_power_state(pstate1); - const struct smu7_power_state *psb = cast_const_phw_smu7_power_state(pstate2); + const struct smu7_power_state *psa; + const struct smu7_power_state *psb; int i; if (pstate1 == NULL || pstate2 == NULL || equal == NULL) return -EINVAL; + psa = cast_const_phw_smu7_power_state(pstate1); + psb = cast_const_phw_smu7_power_state(pstate2); /* If the two states don't even have the same number of performance levels they cannot be the same state. */ if (psa->performance_level_count != psb->performance_level_count) { *equal = false; @@ -4324,6 +4336,7 @@ static const struct pp_hwmgr_func smu7_hwmgr_funcs = { .set_mclk_od = smu7_set_mclk_od, .get_clock_by_type = smu7_get_clock_by_type, .read_sensor = smu7_read_sensor, + .dynamic_state_management_disable = smu7_disable_dpm_tasks, }; uint8_t smu7_get_sleep_divider_id_from_clock(uint32_t clock, diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smc.c b/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smc.c index eda802bc63c8..8c889caba420 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smc.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/iceland_smc.c @@ -2458,7 +2458,7 @@ static int iceland_set_mc_special_registers(struct pp_hwmgr *hwmgr, PP_ASSERT_WITH_CODE((j <= SMU71_DISCRETE_MC_REGISTER_ARRAY_SIZE), "Invalid VramInfo table.", return -EINVAL); - if (!data->is_memory_gddr5) { + if (!data->is_memory_gddr5 && j < SMU71_DISCRETE_MC_REGISTER_ARRAY_SIZE) { table->mc_reg_address[j].s1 = mmMC_PMG_AUTO_CMD; table->mc_reg_address[j].s0 = mmMC_PMG_AUTO_CMD; for (k = 0; k < table->num_entries; k++) { diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c index 2f58e9e2a59c..a51f8cbcfe26 100644 --- a/drivers/gpu/drm/armada/armada_crtc.c +++ b/drivers/gpu/drm/armada/armada_crtc.c @@ -332,17 +332,19 @@ static void armada_drm_crtc_dpms(struct drm_crtc *crtc, int dpms) { struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc); - if (dcrtc->dpms != dpms) { - dcrtc->dpms = dpms; - if (!IS_ERR(dcrtc->clk) && !dpms_blanked(dpms)) - WARN_ON(clk_prepare_enable(dcrtc->clk)); - armada_drm_crtc_update(dcrtc); - if (!IS_ERR(dcrtc->clk) && dpms_blanked(dpms)) - clk_disable_unprepare(dcrtc->clk); + if (dpms_blanked(dcrtc->dpms) != dpms_blanked(dpms)) { if (dpms_blanked(dpms)) armada_drm_vblank_off(dcrtc); - else + else if (!IS_ERR(dcrtc->clk)) + WARN_ON(clk_prepare_enable(dcrtc->clk)); + dcrtc->dpms = dpms; + armada_drm_crtc_update(dcrtc); + if (!dpms_blanked(dpms)) drm_crtc_vblank_on(&dcrtc->crtc); + else if (!IS_ERR(dcrtc->clk)) + clk_disable_unprepare(dcrtc->clk); + } else if (dcrtc->dpms != dpms) { + dcrtc->dpms = dpms; } } diff --git a/drivers/gpu/drm/armada/armada_gem.c b/drivers/gpu/drm/armada/armada_gem.c index a5e428d27d2f..806791897304 100644 --- a/drivers/gpu/drm/armada/armada_gem.c +++ b/drivers/gpu/drm/armada/armada_gem.c @@ -387,7 +387,7 @@ int armada_gem_pwrite_ioctl(struct drm_device *dev, void *data, if (!access_ok(VERIFY_READ, ptr, args->size)) return -EFAULT; - ret = fault_in_multipages_readable(ptr, args->size); + ret = fault_in_pages_readable(ptr, args->size); if (ret) return ret; diff --git a/drivers/gpu/drm/drm_debugfs.c b/drivers/gpu/drm/drm_debugfs.c index 8ea72d821525..800055c39cdb 100644 --- a/drivers/gpu/drm/drm_debugfs.c +++ b/drivers/gpu/drm/drm_debugfs.c @@ -104,8 +104,8 @@ int drm_debugfs_create_files(const struct drm_info_list *files, int count, ent = debugfs_create_file(files[i].name, S_IFREG | S_IRUGO, root, tmp, &drm_debugfs_fops); if (!ent) { - DRM_ERROR("Cannot create /sys/kernel/debug/dri/%s/%s\n", - root->d_name.name, files[i].name); + DRM_ERROR("Cannot create /sys/kernel/debug/dri/%pd/%s\n", + root, files[i].name); kfree(tmp); ret = -1; goto fail; diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c index ac3924a877e0..3e6fe82c6d64 100644 --- a/drivers/gpu/drm/drm_dp_helper.c +++ b/drivers/gpu/drm/drm_dp_helper.c @@ -975,6 +975,12 @@ static void unlock_bus(struct i2c_adapter *i2c, unsigned int flags) mutex_unlock(&i2c_to_aux(i2c)->hw_mutex); } +static const struct i2c_lock_operations drm_dp_i2c_lock_ops = { + .lock_bus = lock_bus, + .trylock_bus = trylock_bus, + .unlock_bus = unlock_bus, +}; + /** * drm_dp_aux_init() - minimally initialise an aux channel * @aux: DisplayPort AUX channel @@ -992,9 +998,7 @@ void drm_dp_aux_init(struct drm_dp_aux *aux) aux->ddc.algo_data = aux; aux->ddc.retries = 3; - aux->ddc.lock_bus = lock_bus; - aux->ddc.trylock_bus = trylock_bus; - aux->ddc.unlock_bus = unlock_bus; + aux->ddc.lock_ops = &drm_dp_i2c_lock_ops; } EXPORT_SYMBOL(drm_dp_aux_init); diff --git a/drivers/gpu/drm/drm_info.c b/drivers/gpu/drm/drm_info.c index 1df2d33d0b40..ffb2ab389d1d 100644 --- a/drivers/gpu/drm/drm_info.c +++ b/drivers/gpu/drm/drm_info.c @@ -54,9 +54,6 @@ int drm_name_info(struct seq_file *m, void *data) mutex_lock(&dev->master_mutex); master = dev->master; - if (!master) - goto out_unlock; - seq_printf(m, "%s", dev->driver->name); if (dev->dev) seq_printf(m, " dev=%s", dev_name(dev->dev)); @@ -65,7 +62,6 @@ int drm_name_info(struct seq_file *m, void *data) if (dev->unique) seq_printf(m, " unique=%s", dev->unique); seq_printf(m, "\n"); -out_unlock: mutex_unlock(&dev->master_mutex); return 0; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c index cb86c7e5495c..d9230132dfbc 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c @@ -329,20 +329,34 @@ void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, unsigned int event, /* * Append a LINK to the submitted command buffer to return to * the ring buffer. return_target is the ring target address. - * We need three dwords: event, wait, link. + * We need at most 7 dwords in the return target: 2 cache flush + + * 2 semaphore stall + 1 event + 1 wait + 1 link. */ - return_dwords = 3; + return_dwords = 7; return_target = etnaviv_buffer_reserve(gpu, buffer, return_dwords); CMD_LINK(cmdbuf, return_dwords, return_target); /* - * Append event, wait and link pointing back to the wait - * command to the ring buffer. + * Append a cache flush, stall, event, wait and link pointing back to + * the wait command to the ring buffer. */ + if (gpu->exec_state == ETNA_PIPE_2D) { + CMD_LOAD_STATE(buffer, VIVS_GL_FLUSH_CACHE, + VIVS_GL_FLUSH_CACHE_PE2D); + } else { + CMD_LOAD_STATE(buffer, VIVS_GL_FLUSH_CACHE, + VIVS_GL_FLUSH_CACHE_DEPTH | + VIVS_GL_FLUSH_CACHE_COLOR); + CMD_LOAD_STATE(buffer, VIVS_TS_FLUSH_CACHE, + VIVS_TS_FLUSH_CACHE_FLUSH); + } + CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); + CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE); CMD_LOAD_STATE(buffer, VIVS_GL_EVENT, VIVS_GL_EVENT_EVENT_ID(event) | VIVS_GL_EVENT_FROM_PE); CMD_WAIT(buffer); - CMD_LINK(buffer, 2, return_target + 8); + CMD_LINK(buffer, 2, etnaviv_iommu_get_cmdbuf_va(gpu, buffer) + + buffer->user_size - 4); if (drm_debug & DRM_UT_DRIVER) pr_info("stream link to 0x%08x @ 0x%08x %p\n", diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c index 9ffca2478e02..3755ef935af4 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c @@ -744,19 +744,22 @@ static struct page **etnaviv_gem_userptr_do_get_pages( int ret = 0, pinned, npages = etnaviv_obj->base.size >> PAGE_SHIFT; struct page **pvec; uintptr_t ptr; + unsigned int flags = 0; pvec = drm_malloc_ab(npages, sizeof(struct page *)); if (!pvec) return ERR_PTR(-ENOMEM); + if (!etnaviv_obj->userptr.ro) + flags |= FOLL_WRITE; + pinned = 0; ptr = etnaviv_obj->userptr.ptr; down_read(&mm->mmap_sem); while (pinned < npages) { ret = get_user_pages_remote(task, mm, ptr, npages - pinned, - !etnaviv_obj->userptr.ro, 0, - pvec + pinned, NULL); + flags, pvec + pinned, NULL); if (ret < 0) break; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_mmu.c b/drivers/gpu/drm/etnaviv/etnaviv_mmu.c index d3796ed8d8c5..169ac96e8f08 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_mmu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_mmu.c @@ -330,7 +330,8 @@ u32 etnaviv_iommu_get_cmdbuf_va(struct etnaviv_gpu *gpu, return (u32)buf->vram_node.start; mutex_lock(&mmu->lock); - ret = etnaviv_iommu_find_iova(mmu, &buf->vram_node, buf->size); + ret = etnaviv_iommu_find_iova(mmu, &buf->vram_node, + buf->size + SZ_64K); if (ret < 0) { mutex_unlock(&mmu->lock); return 0; diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig index 83f61c513b7e..465d344f3391 100644 --- a/drivers/gpu/drm/exynos/Kconfig +++ b/drivers/gpu/drm/exynos/Kconfig @@ -38,7 +38,6 @@ config DRM_EXYNOS7_DECON config DRM_EXYNOS_MIXER bool "Mixer" - depends on !VIDEO_SAMSUNG_S5P_TV help Choose this option if you want to use Exynos Mixer for DRM. @@ -77,7 +76,7 @@ config DRM_EXYNOS_DP config DRM_EXYNOS_HDMI bool "HDMI" - depends on !VIDEO_SAMSUNG_S5P_TV && (DRM_EXYNOS_MIXER || DRM_EXYNOS5433_DECON) + depends on DRM_EXYNOS_MIXER || DRM_EXYNOS5433_DECON help Choose this option if you want to use Exynos HDMI for DRM. diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c index aa92decf4233..fbd13fabdf2d 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c @@ -488,7 +488,8 @@ static dma_addr_t *g2d_userptr_get_dma_addr(struct drm_device *drm_dev, goto err_free; } - ret = get_vaddr_frames(start, npages, true, true, g2d_userptr->vec); + ret = get_vaddr_frames(start, npages, FOLL_FORCE | FOLL_WRITE, + g2d_userptr->vec); if (ret != npages) { DRM_ERROR("failed to get user pages from userptr.\n"); if (ret < 0) diff --git a/drivers/gpu/drm/exynos/exynos_drm_iommu.h b/drivers/gpu/drm/exynos/exynos_drm_iommu.h index c8de4913fdbe..87f6b5672e11 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_iommu.h +++ b/drivers/gpu/drm/exynos/exynos_drm_iommu.h @@ -66,7 +66,7 @@ static inline int __exynos_iommu_create_mapping(struct exynos_drm_private *priv, if (ret) goto free_domain; - ret = iommu_dma_init_domain(domain, start, size); + ret = iommu_dma_init_domain(domain, start, size, NULL); if (ret) goto put_cookie; diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c index 3371635cd4d7..b2d5e188b1b8 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c @@ -51,6 +51,7 @@ static void fsl_dcu_drm_disable_crtc(struct drm_crtc *crtc) DCU_MODE_DCU_MODE(DCU_MODE_OFF)); regmap_write(fsl_dev->regmap, DCU_UPDATE_MODE, DCU_UPDATE_MODE_READREG); + clk_disable_unprepare(fsl_dev->pix_clk); } static void fsl_dcu_drm_crtc_enable(struct drm_crtc *crtc) @@ -58,6 +59,7 @@ static void fsl_dcu_drm_crtc_enable(struct drm_crtc *crtc) struct drm_device *dev = crtc->dev; struct fsl_dcu_drm_device *fsl_dev = dev->dev_private; + clk_prepare_enable(fsl_dev->pix_clk); regmap_update_bits(fsl_dev->regmap, DCU_DCU_MODE, DCU_MODE_DCU_MODE_MASK, DCU_MODE_DCU_MODE(DCU_MODE_NORMAL)); @@ -116,8 +118,6 @@ static void fsl_dcu_drm_crtc_mode_set_nofb(struct drm_crtc *crtc) DCU_THRESHOLD_LS_BF_VS(BF_VS_VAL) | DCU_THRESHOLD_OUT_BUF_HIGH(BUF_MAX_VAL) | DCU_THRESHOLD_OUT_BUF_LOW(BUF_MIN_VAL)); - regmap_write(fsl_dev->regmap, DCU_UPDATE_MODE, - DCU_UPDATE_MODE_READREG); return; } diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c index 0884c45aefe8..e04efbed1a54 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c @@ -267,12 +267,8 @@ static int fsl_dcu_drm_pm_resume(struct device *dev) return ret; } - ret = clk_prepare_enable(fsl_dev->pix_clk); - if (ret < 0) { - dev_err(dev, "failed to enable pix clk\n"); - goto disable_dcu_clk; - } - + if (fsl_dev->tcon) + fsl_tcon_bypass_enable(fsl_dev->tcon); fsl_dcu_drm_init_planes(fsl_dev->drm); drm_atomic_helper_resume(fsl_dev->drm, fsl_dev->state); @@ -284,10 +280,6 @@ static int fsl_dcu_drm_pm_resume(struct device *dev) enable_irq(fsl_dev->irq); return 0; - -disable_dcu_clk: - clk_disable_unprepare(fsl_dev->clk); - return ret; } #endif @@ -401,18 +393,12 @@ static int fsl_dcu_drm_probe(struct platform_device *pdev) goto disable_clk; } - ret = clk_prepare_enable(fsl_dev->pix_clk); - if (ret < 0) { - dev_err(dev, "failed to enable pix clk\n"); - goto unregister_pix_clk; - } - fsl_dev->tcon = fsl_tcon_init(dev); drm = drm_dev_alloc(driver, dev); if (IS_ERR(drm)) { ret = PTR_ERR(drm); - goto disable_pix_clk; + goto unregister_pix_clk; } fsl_dev->dev = dev; @@ -433,8 +419,6 @@ static int fsl_dcu_drm_probe(struct platform_device *pdev) unref: drm_dev_unref(drm); -disable_pix_clk: - clk_disable_unprepare(fsl_dev->pix_clk); unregister_pix_clk: clk_unregister(fsl_dev->pix_clk); disable_clk: @@ -447,7 +431,6 @@ static int fsl_dcu_drm_remove(struct platform_device *pdev) struct fsl_dcu_drm_device *fsl_dev = platform_get_drvdata(pdev); clk_disable_unprepare(fsl_dev->clk); - clk_disable_unprepare(fsl_dev->pix_clk); clk_unregister(fsl_dev->pix_clk); drm_put_dev(fsl_dev->drm); diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c index a7e5486bd1e9..9e6f7d8112b3 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c @@ -211,11 +211,6 @@ void fsl_dcu_drm_init_planes(struct drm_device *dev) for (j = 1; j <= fsl_dev->soc->layer_regs; j++) regmap_write(fsl_dev->regmap, DCU_CTRLDESCLN(i, j), 0); } - regmap_update_bits(fsl_dev->regmap, DCU_DCU_MODE, - DCU_MODE_DCU_MODE_MASK, - DCU_MODE_DCU_MODE(DCU_MODE_OFF)); - regmap_write(fsl_dev->regmap, DCU_UPDATE_MODE, - DCU_UPDATE_MODE_READREG); } struct drm_plane *fsl_dcu_drm_primary_create_plane(struct drm_device *dev) diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c index 26edcc899712..e1dd75b18118 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_rgb.c @@ -20,38 +20,6 @@ #include "fsl_dcu_drm_drv.h" #include "fsl_tcon.h" -static int -fsl_dcu_drm_encoder_atomic_check(struct drm_encoder *encoder, - struct drm_crtc_state *crtc_state, - struct drm_connector_state *conn_state) -{ - return 0; -} - -static void fsl_dcu_drm_encoder_disable(struct drm_encoder *encoder) -{ - struct drm_device *dev = encoder->dev; - struct fsl_dcu_drm_device *fsl_dev = dev->dev_private; - - if (fsl_dev->tcon) - fsl_tcon_bypass_disable(fsl_dev->tcon); -} - -static void fsl_dcu_drm_encoder_enable(struct drm_encoder *encoder) -{ - struct drm_device *dev = encoder->dev; - struct fsl_dcu_drm_device *fsl_dev = dev->dev_private; - - if (fsl_dev->tcon) - fsl_tcon_bypass_enable(fsl_dev->tcon); -} - -static const struct drm_encoder_helper_funcs encoder_helper_funcs = { - .atomic_check = fsl_dcu_drm_encoder_atomic_check, - .disable = fsl_dcu_drm_encoder_disable, - .enable = fsl_dcu_drm_encoder_enable, -}; - static void fsl_dcu_drm_encoder_destroy(struct drm_encoder *encoder) { drm_encoder_cleanup(encoder); @@ -68,13 +36,16 @@ int fsl_dcu_drm_encoder_create(struct fsl_dcu_drm_device *fsl_dev, int ret; encoder->possible_crtcs = 1; + + /* Use bypass mode for parallel RGB/LVDS encoder */ + if (fsl_dev->tcon) + fsl_tcon_bypass_enable(fsl_dev->tcon); + ret = drm_encoder_init(fsl_dev->drm, encoder, &encoder_funcs, DRM_MODE_ENCODER_LVDS, NULL); if (ret < 0) return ret; - drm_encoder_helper_add(encoder, &encoder_helper_funcs); - return 0; } diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 7f4e8adec8a8..bfb2efd8d4d4 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -1786,15 +1786,6 @@ void i915_reset(struct drm_i915_private *dev_priv) goto error; } - /* - * rps/rc6 re-init is necessary to restore state lost after the - * reset and the re-install of gt irqs. Skip for ironlake per - * previous concerns that it doesn't respond well to some forms - * of re-init after reset. - */ - intel_sanitize_gt_powersave(dev_priv); - intel_autoenable_gt_powersave(dev_priv); - wakeup: wake_up_bit(&error->flags, I915_RESET_IN_PROGRESS); return; @@ -1872,7 +1863,17 @@ static int i915_pm_resume(struct device *kdev) /* freeze: before creating the hibernation_image */ static int i915_pm_freeze(struct device *kdev) { - return i915_pm_suspend(kdev); + int ret; + + ret = i915_pm_suspend(kdev); + if (ret) + return ret; + + ret = i915_gem_freeze(kdev_to_i915(kdev)); + if (ret) + return ret; + + return 0; } static int i915_pm_freeze_late(struct device *kdev) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 4dd307ed4336..8b9ee4e390c0 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1984,11 +1984,11 @@ struct drm_i915_private { struct vlv_s0ix_state vlv_s0ix_state; enum { - I915_SKL_SAGV_UNKNOWN = 0, - I915_SKL_SAGV_DISABLED, - I915_SKL_SAGV_ENABLED, - I915_SKL_SAGV_NOT_CONTROLLED - } skl_sagv_status; + I915_SAGV_UNKNOWN = 0, + I915_SAGV_DISABLED, + I915_SAGV_ENABLED, + I915_SAGV_NOT_CONTROLLED + } sagv_status; struct { /* @@ -2276,21 +2276,19 @@ struct drm_i915_gem_object { /** Record of address bit 17 of each page at last unbind. */ unsigned long *bit_17; - union { - /** for phy allocated objects */ - struct drm_dma_handle *phys_handle; - - struct i915_gem_userptr { - uintptr_t ptr; - unsigned read_only :1; - unsigned workers :4; + struct i915_gem_userptr { + uintptr_t ptr; + unsigned read_only :1; + unsigned workers :4; #define I915_GEM_USERPTR_MAX_WORKERS 15 - struct i915_mm_struct *mm; - struct i915_mmu_object *mmu_object; - struct work_struct *work; - } userptr; - }; + struct i915_mm_struct *mm; + struct i915_mmu_object *mmu_object; + struct work_struct *work; + } userptr; + + /** for phys allocated objects */ + struct drm_dma_handle *phys_handle; }; static inline struct drm_i915_gem_object * @@ -3076,6 +3074,7 @@ int i915_gem_wait_ioctl(struct drm_device *dev, void *data, void i915_gem_load_init(struct drm_device *dev); void i915_gem_load_cleanup(struct drm_device *dev); void i915_gem_load_init_fences(struct drm_i915_private *dev_priv); +int i915_gem_freeze(struct drm_i915_private *dev_priv); int i915_gem_freeze_late(struct drm_i915_private *dev_priv); void *i915_gem_object_alloc(struct drm_device *dev); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 2c8106758922..947e82c2b175 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -860,7 +860,7 @@ i915_gem_gtt_pread(struct drm_device *dev, mutex_unlock(&dev->struct_mutex); if (likely(!i915.prefault_disable)) { - ret = fault_in_multipages_writeable(user_data, remain); + ret = fault_in_pages_writeable(user_data, remain); if (ret) { mutex_lock(&dev->struct_mutex); goto out_unpin; @@ -983,7 +983,7 @@ i915_gem_shmem_pread(struct drm_device *dev, mutex_unlock(&dev->struct_mutex); if (likely(!i915.prefault_disable) && !prefaulted) { - ret = fault_in_multipages_writeable(user_data, remain); + ret = fault_in_pages_writeable(user_data, remain); /* Userspace is tricking us, but we've already clobbered * its pages with the prefault and promised to write the * data up to the first fault. Hence ignore any errors @@ -1431,7 +1431,7 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data, return -EFAULT; if (likely(!i915.prefault_disable)) { - ret = fault_in_multipages_readable(u64_to_user_ptr(args->data_ptr), + ret = fault_in_pages_readable(u64_to_user_ptr(args->data_ptr), args->size); if (ret) return -EFAULT; @@ -2616,8 +2616,6 @@ static void i915_gem_reset_engine(struct intel_engine_cs *engine) list_for_each_entry_continue(request, &engine->request_list, link) if (request->ctx == incomplete_ctx) reset_request(request); - - engine->i915->gt.active_engines &= ~intel_engine_flag(engine); } void i915_gem_reset(struct drm_i915_private *dev_priv) @@ -2628,9 +2626,15 @@ void i915_gem_reset(struct drm_i915_private *dev_priv) for_each_engine(engine, dev_priv) i915_gem_reset_engine(engine); - mod_delayed_work(dev_priv->wq, &dev_priv->gt.idle_work, 0); i915_gem_restore_fences(&dev_priv->drm); + + if (dev_priv->gt.awake) { + intel_sanitize_gt_powersave(dev_priv); + intel_enable_gt_powersave(dev_priv); + if (INTEL_GEN(dev_priv) >= 6) + gen6_rps_busy(dev_priv); + } } static void nop_submit_request(struct drm_i915_gem_request *request) @@ -4589,6 +4593,19 @@ void i915_gem_load_cleanup(struct drm_device *dev) rcu_barrier(); } +int i915_gem_freeze(struct drm_i915_private *dev_priv) +{ + intel_runtime_pm_get(dev_priv); + + mutex_lock(&dev_priv->drm.struct_mutex); + i915_gem_shrink_all(dev_priv); + mutex_unlock(&dev_priv->drm.struct_mutex); + + intel_runtime_pm_put(dev_priv); + + return 0; +} + int i915_gem_freeze_late(struct drm_i915_private *dev_priv) { struct drm_i915_gem_object *obj; @@ -4612,7 +4629,8 @@ int i915_gem_freeze_late(struct drm_i915_private *dev_priv) * the objects as well. */ - i915_gem_shrink_all(dev_priv); + mutex_lock(&dev_priv->drm.struct_mutex); + i915_gem_shrink(dev_priv, -1UL, I915_SHRINK_UNBOUND); for (p = phases; *p; p++) { list_for_each_entry(obj, *p, global_list) { @@ -4620,6 +4638,7 @@ int i915_gem_freeze_late(struct drm_i915_private *dev_priv) obj->base.write_domain = I915_GEM_DOMAIN_CPU; } } + mutex_unlock(&dev_priv->drm.struct_mutex); return 0; } diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 33c85227643d..7adb4c77cc7f 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -451,8 +451,8 @@ static void *reloc_iomap(struct drm_i915_gem_object *obj, 0, ggtt->mappable_end, DRM_MM_SEARCH_DEFAULT, DRM_MM_CREATE_DEFAULT); - if (ret) - return ERR_PTR(ret); + if (ret) /* no inactive aperture space, use cpu reloc */ + return NULL; } else { ret = i915_vma_put_fence(vma); if (ret) { @@ -1253,7 +1253,7 @@ validate_exec_list(struct drm_device *dev, return -EFAULT; if (likely(!i915.prefault_disable)) { - if (fault_in_multipages_readable(ptr, length)) + if (fault_in_pages_readable(ptr, length)) return -EFAULT; } } diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c index 40978bc12ceb..8832f8ec1583 100644 --- a/drivers/gpu/drm/i915/i915_gem_request.c +++ b/drivers/gpu/drm/i915/i915_gem_request.c @@ -328,6 +328,7 @@ submit_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state) switch (state) { case FENCE_COMPLETE: + request->engine->last_submitted_seqno = request->fence.seqno; request->engine->submit_request(request); break; @@ -641,8 +642,8 @@ void __i915_add_request(struct drm_i915_gem_request *request, bool flush_caches) &request->submitq); request->emitted_jiffies = jiffies; - request->previous_seqno = engine->last_submitted_seqno; - engine->last_submitted_seqno = request->fence.seqno; + request->previous_seqno = engine->last_pending_seqno; + engine->last_pending_seqno = request->fence.seqno; i915_gem_active_set(&engine->last_request, request); list_add_tail(&request->link, &engine->request_list); list_add_tail(&request->ring_link, &ring->request_list); diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c index e537930c64b5..c6f780f5abc9 100644 --- a/drivers/gpu/drm/i915/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/i915_gem_userptr.c @@ -508,6 +508,10 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work) pvec = drm_malloc_gfp(npages, sizeof(struct page *), GFP_TEMPORARY); if (pvec != NULL) { struct mm_struct *mm = obj->userptr.mm->mm; + unsigned int flags = 0; + + if (!obj->userptr.read_only) + flags |= FOLL_WRITE; ret = -EFAULT; if (atomic_inc_not_zero(&mm->mm_users)) { @@ -517,7 +521,7 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work) (work->task, mm, obj->userptr.ptr + pinned * PAGE_SIZE, npages - pinned, - !obj->userptr.read_only, 0, + flags, pvec + pinned, NULL); if (ret < 0) break; diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c index 43358e18d34c..3106dcc06fe9 100644 --- a/drivers/gpu/drm/i915/i915_guc_submission.c +++ b/drivers/gpu/drm/i915/i915_guc_submission.c @@ -451,6 +451,18 @@ int i915_guc_wq_reserve(struct drm_i915_gem_request *request) return ret; } +void i915_guc_wq_unreserve(struct drm_i915_gem_request *request) +{ + const size_t wqi_size = sizeof(struct guc_wq_item); + struct i915_guc_client *gc = request->i915->guc.execbuf_client; + + GEM_BUG_ON(READ_ONCE(gc->wq_rsvd) < wqi_size); + + spin_lock(&gc->wq_lock); + gc->wq_rsvd -= wqi_size; + spin_unlock(&gc->wq_lock); +} + /* Construct a Work Item and append it to the GuC's Work Queue */ static void guc_wq_item_append(struct i915_guc_client *gc, struct drm_i915_gem_request *rq) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index c128fdbd24e4..3fc286cd1157 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -350,6 +350,9 @@ void gen6_reset_rps_interrupts(struct drm_i915_private *dev_priv) void gen6_enable_rps_interrupts(struct drm_i915_private *dev_priv) { + if (READ_ONCE(dev_priv->rps.interrupts_enabled)) + return; + spin_lock_irq(&dev_priv->irq_lock); WARN_ON_ONCE(dev_priv->rps.pm_iir); WARN_ON_ONCE(I915_READ(gen6_pm_iir(dev_priv)) & dev_priv->pm_rps_events); @@ -368,6 +371,9 @@ u32 gen6_sanitize_rps_pm_mask(struct drm_i915_private *dev_priv, u32 mask) void gen6_disable_rps_interrupts(struct drm_i915_private *dev_priv) { + if (!READ_ONCE(dev_priv->rps.interrupts_enabled)) + return; + spin_lock_irq(&dev_priv->irq_lock); dev_priv->rps.interrupts_enabled = false; @@ -2816,7 +2822,7 @@ semaphore_wait_to_signaller_ring(struct intel_engine_cs *engine, u32 ipehr, if (engine == signaller) continue; - if (offset == signaller->semaphore.signal_ggtt[engine->id]) + if (offset == signaller->semaphore.signal_ggtt[engine->hw_id]) return signaller; } } else { @@ -2826,13 +2832,13 @@ semaphore_wait_to_signaller_ring(struct intel_engine_cs *engine, u32 ipehr, if(engine == signaller) continue; - if (sync_bits == signaller->semaphore.mbox.wait[engine->id]) + if (sync_bits == signaller->semaphore.mbox.wait[engine->hw_id]) return signaller; } } - DRM_DEBUG_DRIVER("No signaller ring found for ring %i, ipehr 0x%08x, offset 0x%016llx\n", - engine->id, ipehr, offset); + DRM_DEBUG_DRIVER("No signaller ring found for %s, ipehr 0x%08x, offset 0x%016llx\n", + engine->name, ipehr, offset); return ERR_PTR(-ENODEV); } diff --git a/drivers/gpu/drm/i915/intel_breadcrumbs.c b/drivers/gpu/drm/i915/intel_breadcrumbs.c index 9bad14d22c95..495611b7068d 100644 --- a/drivers/gpu/drm/i915/intel_breadcrumbs.c +++ b/drivers/gpu/drm/i915/intel_breadcrumbs.c @@ -578,6 +578,36 @@ int intel_engine_init_breadcrumbs(struct intel_engine_cs *engine) return 0; } +static void cancel_fake_irq(struct intel_engine_cs *engine) +{ + struct intel_breadcrumbs *b = &engine->breadcrumbs; + + del_timer_sync(&b->hangcheck); + del_timer_sync(&b->fake_irq); + clear_bit(engine->id, &engine->i915->gpu_error.missed_irq_rings); +} + +void intel_engine_reset_breadcrumbs(struct intel_engine_cs *engine) +{ + struct intel_breadcrumbs *b = &engine->breadcrumbs; + + cancel_fake_irq(engine); + spin_lock(&b->lock); + + __intel_breadcrumbs_disable_irq(b); + if (intel_engine_has_waiter(engine)) { + b->timeout = wait_timeout(); + __intel_breadcrumbs_enable_irq(b); + if (READ_ONCE(b->irq_posted)) + wake_up_process(b->first_wait->tsk); + } else { + /* sanitize the IMR and unmask any auxiliary interrupts */ + irq_disable(engine); + } + + spin_unlock(&b->lock); +} + void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine) { struct intel_breadcrumbs *b = &engine->breadcrumbs; @@ -585,8 +615,7 @@ void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine) if (!IS_ERR_OR_NULL(b->signaler)) kthread_stop(b->signaler); - del_timer_sync(&b->hangcheck); - del_timer_sync(&b->fake_irq); + cancel_fake_irq(engine); } unsigned int intel_kick_waiters(struct drm_i915_private *i915) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 6c5c36eba6cb..4c21d2ec2c51 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -3408,6 +3408,8 @@ static void skylake_update_primary_plane(struct drm_plane *plane, dst_w--; dst_h--; + intel_crtc->dspaddr_offset = surf_addr; + intel_crtc->adjusted_x = src_x; intel_crtc->adjusted_y = src_y; @@ -3629,6 +3631,7 @@ void intel_finish_reset(struct drm_i915_private *dev_priv) intel_runtime_pm_disable_interrupts(dev_priv); intel_runtime_pm_enable_interrupts(dev_priv); + intel_pps_unlock_regs_wa(dev_priv); intel_modeset_init_hw(dev); spin_lock_irq(&dev_priv->irq_lock); @@ -9511,6 +9514,24 @@ static void ironlake_compute_dpll(struct intel_crtc *intel_crtc, if (intel_crtc_has_dp_encoder(crtc_state)) dpll |= DPLL_SDVO_HIGH_SPEED; + /* + * The high speed IO clock is only really required for + * SDVO/HDMI/DP, but we also enable it for CRT to make it + * possible to share the DPLL between CRT and HDMI. Enabling + * the clock needlessly does no real harm, except use up a + * bit of power potentially. + * + * We'll limit this to IVB with 3 pipes, since it has only two + * DPLLs and so DPLL sharing is the only way to get three pipes + * driving PCH ports at the same time. On SNB we could do this, + * and potentially avoid enabling the second DPLL, but it's not + * clear if it''s a win or loss power wise. No point in doing + * this on ILK at all since it has a fixed DPLL<->pipe mapping. + */ + if (INTEL_INFO(dev_priv)->num_pipes == 3 && + intel_crtc_has_type(crtc_state, INTEL_OUTPUT_ANALOG)) + dpll |= DPLL_SDVO_HIGH_SPEED; + /* compute bitmask from p1 value */ dpll |= (1 << (crtc_state->dpll.p1 - 1)) << DPLL_FPA01_P1_POST_DIV_SHIFT; /* also FPA1 */ @@ -14369,8 +14390,8 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state) * SKL workaround: bspec recommends we disable the SAGV when we * have more then one pipe enabled */ - if (IS_SKYLAKE(dev_priv) && !skl_can_enable_sagv(state)) - skl_disable_sagv(dev_priv); + if (!intel_can_enable_sagv(state)) + intel_disable_sagv(dev_priv); intel_modeset_verify_disabled(dev); } @@ -14427,9 +14448,8 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state) intel_modeset_verify_crtc(crtc, old_crtc_state, crtc->state); } - if (IS_SKYLAKE(dev_priv) && intel_state->modeset && - skl_can_enable_sagv(state)) - skl_enable_sagv(dev_priv); + if (intel_state->modeset && intel_can_enable_sagv(state)) + intel_enable_sagv(dev_priv); drm_atomic_helper_commit_hw_done(state); diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index acd0c51f74d5..14a3cf0b7213 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -4337,7 +4337,7 @@ intel_dp_unset_edid(struct intel_dp *intel_dp) intel_dp->has_audio = false; } -static void +static enum drm_connector_status intel_dp_long_pulse(struct intel_connector *intel_connector) { struct drm_connector *connector = &intel_connector->base; @@ -4361,7 +4361,7 @@ intel_dp_long_pulse(struct intel_connector *intel_connector) else status = connector_status_disconnected; - if (status != connector_status_connected) { + if (status == connector_status_disconnected) { intel_dp->compliance_test_active = 0; intel_dp->compliance_test_type = 0; intel_dp->compliance_test_data = 0; @@ -4423,8 +4423,8 @@ intel_dp_long_pulse(struct intel_connector *intel_connector) intel_dp->aux.i2c_defer_count = 0; intel_dp_set_edid(intel_dp); - - status = connector_status_connected; + if (is_edp(intel_dp) || intel_connector->detect_edid) + status = connector_status_connected; intel_dp->detect_done = true; /* Try to read the source of the interrupt */ @@ -4443,12 +4443,11 @@ intel_dp_long_pulse(struct intel_connector *intel_connector) } out: - if ((status != connector_status_connected) && - (intel_dp->is_mst == false)) + if (status != connector_status_connected && !intel_dp->is_mst) intel_dp_unset_edid(intel_dp); intel_display_power_put(to_i915(dev), power_domain); - return; + return status; } static enum drm_connector_status @@ -4457,7 +4456,7 @@ intel_dp_detect(struct drm_connector *connector, bool force) struct intel_dp *intel_dp = intel_attached_dp(connector); struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp); struct intel_encoder *intel_encoder = &intel_dig_port->base; - struct intel_connector *intel_connector = to_intel_connector(connector); + enum drm_connector_status status = connector->status; DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n", connector->base.id, connector->name); @@ -4472,14 +4471,11 @@ intel_dp_detect(struct drm_connector *connector, bool force) /* If full detect is not performed yet, do a full detect */ if (!intel_dp->detect_done) - intel_dp_long_pulse(intel_dp->attached_connector); + status = intel_dp_long_pulse(intel_dp->attached_connector); intel_dp->detect_done = false; - if (is_edp(intel_dp) || intel_connector->detect_edid) - return connector_status_connected; - else - return connector_status_disconnected; + return status; } static void @@ -4831,36 +4827,34 @@ intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd) port_name(intel_dig_port->port), long_hpd ? "long" : "short"); + if (long_hpd) { + intel_dp->detect_done = false; + return IRQ_NONE; + } + power_domain = intel_display_port_aux_power_domain(intel_encoder); intel_display_power_get(dev_priv, power_domain); - if (long_hpd) { - intel_dp_long_pulse(intel_dp->attached_connector); - if (intel_dp->is_mst) - ret = IRQ_HANDLED; - goto put_power; - - } else { - if (intel_dp->is_mst) { - if (intel_dp_check_mst_status(intel_dp) == -EINVAL) { - /* - * If we were in MST mode, and device is not - * there, get out of MST mode - */ - DRM_DEBUG_KMS("MST device may have disappeared %d vs %d\n", - intel_dp->is_mst, intel_dp->mst_mgr.mst_state); - intel_dp->is_mst = false; - drm_dp_mst_topology_mgr_set_mst(&intel_dp->mst_mgr, - intel_dp->is_mst); - goto put_power; - } + if (intel_dp->is_mst) { + if (intel_dp_check_mst_status(intel_dp) == -EINVAL) { + /* + * If we were in MST mode, and device is not + * there, get out of MST mode + */ + DRM_DEBUG_KMS("MST device may have disappeared %d vs %d\n", + intel_dp->is_mst, intel_dp->mst_mgr.mst_state); + intel_dp->is_mst = false; + drm_dp_mst_topology_mgr_set_mst(&intel_dp->mst_mgr, + intel_dp->is_mst); + intel_dp->detect_done = false; + goto put_power; } + } - if (!intel_dp->is_mst) { - if (!intel_dp_short_pulse(intel_dp)) { - intel_dp_long_pulse(intel_dp->attached_connector); - goto put_power; - } + if (!intel_dp->is_mst) { + if (!intel_dp_short_pulse(intel_dp)) { + intel_dp->detect_done = false; + goto put_power; } } diff --git a/drivers/gpu/drm/i915/intel_dpll_mgr.c b/drivers/gpu/drm/i915/intel_dpll_mgr.c index c26d18a574b6..1c59ca50c430 100644 --- a/drivers/gpu/drm/i915/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/intel_dpll_mgr.c @@ -1694,21 +1694,32 @@ bool bxt_ddi_dp_set_dpll_hw_state(int clock, return bxt_ddi_set_dpll_hw_state(clock, &clk_div, dpll_hw_state); } +static bool +bxt_ddi_hdmi_set_dpll_hw_state(struct intel_crtc *intel_crtc, + struct intel_crtc_state *crtc_state, int clock, + struct intel_dpll_hw_state *dpll_hw_state) +{ + struct bxt_clk_div clk_div = { }; + + bxt_ddi_hdmi_pll_dividers(intel_crtc, crtc_state, clock, &clk_div); + + return bxt_ddi_set_dpll_hw_state(clock, &clk_div, dpll_hw_state); +} + static struct intel_shared_dpll * bxt_get_dpll(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state, struct intel_encoder *encoder) { - struct bxt_clk_div clk_div = {0}; - struct intel_dpll_hw_state dpll_hw_state = {0}; + struct intel_dpll_hw_state dpll_hw_state = { }; struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); struct intel_digital_port *intel_dig_port; struct intel_shared_dpll *pll; int i, clock = crtc_state->port_clock; - if (encoder->type == INTEL_OUTPUT_HDMI - && !bxt_ddi_hdmi_pll_dividers(crtc, crtc_state, - clock, &clk_div)) + if (encoder->type == INTEL_OUTPUT_HDMI && + !bxt_ddi_hdmi_set_dpll_hw_state(crtc, crtc_state, clock, + &dpll_hw_state)) return NULL; if ((encoder->type == INTEL_OUTPUT_DP || diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 5145ff264c8e..3ffba2def09c 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -263,6 +263,7 @@ struct intel_panel { bool enabled; bool combination_mode; /* gen 2/4 only */ bool active_low_pwm; + bool alternate_pwm_increment; /* lpt+ */ /* PWM chip */ bool util_pin_active_low; /* bxt+ */ @@ -1732,9 +1733,9 @@ void ilk_wm_get_hw_state(struct drm_device *dev); void skl_wm_get_hw_state(struct drm_device *dev); void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, struct skl_ddb_allocation *ddb /* out */); -bool skl_can_enable_sagv(struct drm_atomic_state *state); -int skl_enable_sagv(struct drm_i915_private *dev_priv); -int skl_disable_sagv(struct drm_i915_private *dev_priv); +bool intel_can_enable_sagv(struct drm_atomic_state *state); +int intel_enable_sagv(struct drm_i915_private *dev_priv); +int intel_disable_sagv(struct drm_i915_private *dev_priv); bool skl_ddb_allocation_equals(const struct skl_ddb_allocation *old, const struct skl_ddb_allocation *new, enum pipe pipe); diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c index e405f1080296..025e232a4205 100644 --- a/drivers/gpu/drm/i915/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/intel_engine_cs.c @@ -210,9 +210,6 @@ void intel_engine_init_seqno(struct intel_engine_cs *engine, u32 seqno) void intel_engine_init_hangcheck(struct intel_engine_cs *engine) { memset(&engine->hangcheck, 0, sizeof(engine->hangcheck)); - clear_bit(engine->id, &engine->i915->gpu_error.missed_irq_rings); - if (intel_engine_has_waiter(engine)) - i915_queue_hangcheck(engine->i915); } static void intel_engine_init_requests(struct intel_engine_cs *engine) @@ -307,18 +304,6 @@ int intel_engine_init_common(struct intel_engine_cs *engine) return 0; } -void intel_engine_reset_irq(struct intel_engine_cs *engine) -{ - struct drm_i915_private *dev_priv = engine->i915; - - spin_lock_irq(&dev_priv->irq_lock); - if (intel_engine_has_waiter(engine)) - engine->irq_enable(engine); - else - engine->irq_disable(engine); - spin_unlock_irq(&dev_priv->irq_lock); -} - /** * intel_engines_cleanup_common - cleans up the engine state created by * the common initiailizers. diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h index b1ba86958811..5cdf7aa75be5 100644 --- a/drivers/gpu/drm/i915/intel_guc.h +++ b/drivers/gpu/drm/i915/intel_guc.h @@ -160,6 +160,7 @@ extern int intel_guc_resume(struct drm_device *dev); int i915_guc_submission_init(struct drm_i915_private *dev_priv); int i915_guc_submission_enable(struct drm_i915_private *dev_priv); int i915_guc_wq_reserve(struct drm_i915_gem_request *rq); +void i915_guc_wq_unreserve(struct drm_i915_gem_request *request); void i915_guc_submission_disable(struct drm_i915_private *dev_priv); void i915_guc_submission_fini(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 251143361f31..0adb879833ff 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -226,10 +226,16 @@ enum { /* Typical size of the average request (2 pipecontrols and a MI_BB) */ #define EXECLISTS_REQUEST_SIZE 64 /* bytes */ +#define WA_TAIL_DWORDS 2 + static int execlists_context_deferred_alloc(struct i915_gem_context *ctx, struct intel_engine_cs *engine); static int intel_lr_context_pin(struct i915_gem_context *ctx, struct intel_engine_cs *engine); +static void execlists_init_reg_state(u32 *reg_state, + struct i915_gem_context *ctx, + struct intel_engine_cs *engine, + struct intel_ring *ring); /** * intel_sanitize_enable_execlists() - sanitize i915.enable_execlists @@ -621,6 +627,10 @@ int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request request->ring = ce->ring; + ret = intel_lr_context_pin(request->ctx, engine); + if (ret) + return ret; + if (i915.enable_guc_submission) { /* * Check that the GuC has space for the request before @@ -629,21 +639,17 @@ int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request */ ret = i915_guc_wq_reserve(request); if (ret) - return ret; + goto err_unpin; } - ret = intel_lr_context_pin(request->ctx, engine); - if (ret) - return ret; - ret = intel_ring_begin(request, 0); if (ret) - goto err_unpin; + goto err_unreserve; if (!ce->initialised) { ret = engine->init_context(request); if (ret) - goto err_unpin; + goto err_unreserve; ce->initialised = true; } @@ -658,6 +664,9 @@ int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request request->reserved_space -= EXECLISTS_REQUEST_SIZE; return 0; +err_unreserve: + if (i915.enable_guc_submission) + i915_guc_wq_unreserve(request); err_unpin: intel_lr_context_unpin(request->ctx, engine); return ret; @@ -708,7 +717,6 @@ static int intel_lr_context_pin(struct i915_gem_context *ctx, { struct intel_context *ce = &ctx->engine[engine->id]; void *vaddr; - u32 *lrc_reg_state; int ret; lockdep_assert_held(&ctx->i915->drm.struct_mutex); @@ -727,17 +735,16 @@ static int intel_lr_context_pin(struct i915_gem_context *ctx, goto unpin_vma; } - lrc_reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE; - ret = intel_ring_pin(ce->ring); if (ret) goto unpin_map; intel_lr_context_descriptor_update(ctx, engine); - lrc_reg_state[CTX_RING_BUFFER_START+1] = + ce->lrc_reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE; + ce->lrc_reg_state[CTX_RING_BUFFER_START+1] = i915_ggtt_offset(ce->ring->vma); - ce->lrc_reg_state = lrc_reg_state; + ce->state->obj->dirty = true; /* Invalidate GuC TLB. */ @@ -1231,7 +1238,7 @@ static int gen8_init_common_ring(struct intel_engine_cs *engine) lrc_init_hws(engine); - intel_engine_reset_irq(engine); + intel_engine_reset_breadcrumbs(engine); I915_WRITE(RING_HWSTAM(engine->mmio_base), 0xffffffff); @@ -1289,8 +1296,21 @@ static void reset_common_ring(struct intel_engine_cs *engine, struct execlist_port *port = engine->execlist_port; struct intel_context *ce = &request->ctx->engine[engine->id]; + /* We want a simple context + ring to execute the breadcrumb update. + * We cannot rely on the context being intact across the GPU hang, + * so clear it and rebuild just what we need for the breadcrumb. + * All pending requests for this context will be zapped, and any + * future request will be after userspace has had the opportunity + * to recreate its own state. + */ + execlists_init_reg_state(ce->lrc_reg_state, + request->ctx, engine, ce->ring); + /* Move the RING_HEAD onto the breadcrumb, past the hanging batch */ + ce->lrc_reg_state[CTX_RING_BUFFER_START+1] = + i915_ggtt_offset(ce->ring->vma); ce->lrc_reg_state[CTX_RING_HEAD+1] = request->postfix; + request->ring->head = request->postfix; request->ring->last_retired_head = -1; intel_ring_update_space(request->ring); @@ -1310,6 +1330,9 @@ static void reset_common_ring(struct intel_engine_cs *engine, GEM_BUG_ON(request->ctx != port[0].request->ctx); port[0].count = 0; port[1].count = 0; + + /* Reset WaIdleLiteRestore:bdw,skl as well */ + request->tail = request->wa_tail - WA_TAIL_DWORDS * sizeof(u32); } static int intel_logical_ring_emit_pdps(struct drm_i915_gem_request *req) @@ -1547,7 +1570,6 @@ static void bxt_a_seqno_barrier(struct intel_engine_cs *engine) * used as a workaround for not being allowed to do lite * restore with HEAD==TAIL (WaIdleLiteRestore). */ -#define WA_TAIL_DWORDS 2 static int gen8_emit_request(struct drm_i915_gem_request *request) { @@ -1894,38 +1916,13 @@ static u32 intel_lr_indirect_ctx_offset(struct intel_engine_cs *engine) return indirect_ctx_offset; } -static int -populate_lr_context(struct i915_gem_context *ctx, - struct drm_i915_gem_object *ctx_obj, - struct intel_engine_cs *engine, - struct intel_ring *ring) +static void execlists_init_reg_state(u32 *reg_state, + struct i915_gem_context *ctx, + struct intel_engine_cs *engine, + struct intel_ring *ring) { - struct drm_i915_private *dev_priv = ctx->i915; - struct i915_hw_ppgtt *ppgtt = ctx->ppgtt; - void *vaddr; - u32 *reg_state; - int ret; - - if (!ppgtt) - ppgtt = dev_priv->mm.aliasing_ppgtt; - - ret = i915_gem_object_set_to_cpu_domain(ctx_obj, true); - if (ret) { - DRM_DEBUG_DRIVER("Could not set to CPU domain\n"); - return ret; - } - - vaddr = i915_gem_object_pin_map(ctx_obj, I915_MAP_WB); - if (IS_ERR(vaddr)) { - ret = PTR_ERR(vaddr); - DRM_DEBUG_DRIVER("Could not map object pages! (%d)\n", ret); - return ret; - } - ctx_obj->dirty = true; - - /* The second page of the context object contains some fields which must - * be set up prior to the first execution. */ - reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE; + struct drm_i915_private *dev_priv = engine->i915; + struct i915_hw_ppgtt *ppgtt = ctx->ppgtt ?: dev_priv->mm.aliasing_ppgtt; /* A context is actually a big batch buffer with several MI_LOAD_REGISTER_IMM * commands followed by (reg, value) pairs. The values we are setting here are @@ -1939,14 +1936,11 @@ populate_lr_context(struct i915_gem_context *ctx, _MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | (HAS_RESOURCE_STREAMER(dev_priv) ? - CTX_CTRL_RS_CTX_ENABLE : 0))); + CTX_CTRL_RS_CTX_ENABLE : 0))); ASSIGN_CTX_REG(reg_state, CTX_RING_HEAD, RING_HEAD(engine->mmio_base), 0); ASSIGN_CTX_REG(reg_state, CTX_RING_TAIL, RING_TAIL(engine->mmio_base), 0); - /* Ring buffer start address is not known until the buffer is pinned. - * It is written to the context image in execlists_update_context() - */ ASSIGN_CTX_REG(reg_state, CTX_RING_BUFFER_START, RING_START(engine->mmio_base), 0); ASSIGN_CTX_REG(reg_state, CTX_RING_BUFFER_CONTROL, @@ -2029,6 +2023,36 @@ populate_lr_context(struct i915_gem_context *ctx, ASSIGN_CTX_REG(reg_state, CTX_R_PWR_CLK_STATE, GEN8_R_PWR_CLK_STATE, make_rpcs(dev_priv)); } +} + +static int +populate_lr_context(struct i915_gem_context *ctx, + struct drm_i915_gem_object *ctx_obj, + struct intel_engine_cs *engine, + struct intel_ring *ring) +{ + void *vaddr; + int ret; + + ret = i915_gem_object_set_to_cpu_domain(ctx_obj, true); + if (ret) { + DRM_DEBUG_DRIVER("Could not set to CPU domain\n"); + return ret; + } + + vaddr = i915_gem_object_pin_map(ctx_obj, I915_MAP_WB); + if (IS_ERR(vaddr)) { + ret = PTR_ERR(vaddr); + DRM_DEBUG_DRIVER("Could not map object pages! (%d)\n", ret); + return ret; + } + ctx_obj->dirty = true; + + /* The second page of the context object contains some fields which must + * be set up prior to the first execution. */ + + execlists_init_reg_state(vaddr + LRC_STATE_PN * PAGE_SIZE, + ctx, engine, ring); i915_gem_object_unpin_map(ctx_obj); diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c index c10e9b0405e8..be4b4d546fd9 100644 --- a/drivers/gpu/drm/i915/intel_panel.c +++ b/drivers/gpu/drm/i915/intel_panel.c @@ -841,7 +841,7 @@ static void lpt_enable_backlight(struct intel_connector *connector) { struct drm_i915_private *dev_priv = to_i915(connector->base.dev); struct intel_panel *panel = &connector->panel; - u32 pch_ctl1, pch_ctl2; + u32 pch_ctl1, pch_ctl2, schicken; pch_ctl1 = I915_READ(BLC_PWM_PCH_CTL1); if (pch_ctl1 & BLM_PCH_PWM_ENABLE) { @@ -850,6 +850,22 @@ static void lpt_enable_backlight(struct intel_connector *connector) I915_WRITE(BLC_PWM_PCH_CTL1, pch_ctl1); } + if (HAS_PCH_LPT(dev_priv)) { + schicken = I915_READ(SOUTH_CHICKEN2); + if (panel->backlight.alternate_pwm_increment) + schicken |= LPT_PWM_GRANULARITY; + else + schicken &= ~LPT_PWM_GRANULARITY; + I915_WRITE(SOUTH_CHICKEN2, schicken); + } else { + schicken = I915_READ(SOUTH_CHICKEN1); + if (panel->backlight.alternate_pwm_increment) + schicken |= SPT_PWM_GRANULARITY; + else + schicken &= ~SPT_PWM_GRANULARITY; + I915_WRITE(SOUTH_CHICKEN1, schicken); + } + pch_ctl2 = panel->backlight.max << 16; I915_WRITE(BLC_PWM_PCH_CTL2, pch_ctl2); @@ -1242,10 +1258,10 @@ static u32 bxt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz) */ static u32 spt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz) { - struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct intel_panel *panel = &connector->panel; u32 mul; - if (I915_READ(SOUTH_CHICKEN1) & SPT_PWM_GRANULARITY) + if (panel->backlight.alternate_pwm_increment) mul = 128; else mul = 16; @@ -1261,9 +1277,10 @@ static u32 spt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz) static u32 lpt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz) { struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct intel_panel *panel = &connector->panel; u32 mul, clock; - if (I915_READ(SOUTH_CHICKEN2) & LPT_PWM_GRANULARITY) + if (panel->backlight.alternate_pwm_increment) mul = 16; else mul = 128; @@ -1414,6 +1431,13 @@ static int lpt_setup_backlight(struct intel_connector *connector, enum pipe unus struct drm_i915_private *dev_priv = to_i915(connector->base.dev); struct intel_panel *panel = &connector->panel; u32 pch_ctl1, pch_ctl2, val; + bool alt; + + if (HAS_PCH_LPT(dev_priv)) + alt = I915_READ(SOUTH_CHICKEN2) & LPT_PWM_GRANULARITY; + else + alt = I915_READ(SOUTH_CHICKEN1) & SPT_PWM_GRANULARITY; + panel->backlight.alternate_pwm_increment = alt; pch_ctl1 = I915_READ(BLC_PWM_PCH_CTL1); panel->backlight.active_low_pwm = pch_ctl1 & BLM_PCH_POLARITY; diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 1472400ddce3..e2f0a32279e7 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -2127,32 +2127,34 @@ static void intel_read_wm_latency(struct drm_device *dev, uint16_t wm[8]) GEN9_MEM_LATENCY_LEVEL_MASK; /* + * If a level n (n > 1) has a 0us latency, all levels m (m >= n) + * need to be disabled. We make sure to sanitize the values out + * of the punit to satisfy this requirement. + */ + for (level = 1; level <= max_level; level++) { + if (wm[level] == 0) { + for (i = level + 1; i <= max_level; i++) + wm[i] = 0; + break; + } + } + + /* * WaWmMemoryReadLatency:skl * * punit doesn't take into account the read latency so we need - * to add 2us to the various latency levels we retrieve from - * the punit. - * - W0 is a bit special in that it's the only level that - * can't be disabled if we want to have display working, so - * we always add 2us there. - * - For levels >=1, punit returns 0us latency when they are - * disabled, so we respect that and don't add 2us then - * - * Additionally, if a level n (n > 1) has a 0us latency, all - * levels m (m >= n) need to be disabled. We make sure to - * sanitize the values out of the punit to satisfy this - * requirement. + * to add 2us to the various latency levels we retrieve from the + * punit when level 0 response data us 0us. */ - wm[0] += 2; - for (level = 1; level <= max_level; level++) - if (wm[level] != 0) + if (wm[0] == 0) { + wm[0] += 2; + for (level = 1; level <= max_level; level++) { + if (wm[level] == 0) + break; wm[level] += 2; - else { - for (i = level + 1; i <= max_level; i++) - wm[i] = 0; - - break; } + } + } else if (IS_HASWELL(dev) || IS_BROADWELL(dev)) { uint64_t sskpd = I915_READ64(MCH_SSKPD); @@ -2877,6 +2879,19 @@ skl_wm_plane_id(const struct intel_plane *plane) } } +static bool +intel_has_sagv(struct drm_i915_private *dev_priv) +{ + if (IS_KABYLAKE(dev_priv)) + return true; + + if (IS_SKYLAKE(dev_priv) && + dev_priv->sagv_status != I915_SAGV_NOT_CONTROLLED) + return true; + + return false; +} + /* * SAGV dynamically adjusts the system agent voltage and clock frequencies * depending on power and performance requirements. The display engine access @@ -2889,12 +2904,14 @@ skl_wm_plane_id(const struct intel_plane *plane) * - We're not using an interlaced display configuration */ int -skl_enable_sagv(struct drm_i915_private *dev_priv) +intel_enable_sagv(struct drm_i915_private *dev_priv) { int ret; - if (dev_priv->skl_sagv_status == I915_SKL_SAGV_NOT_CONTROLLED || - dev_priv->skl_sagv_status == I915_SKL_SAGV_ENABLED) + if (!intel_has_sagv(dev_priv)) + return 0; + + if (dev_priv->sagv_status == I915_SAGV_ENABLED) return 0; DRM_DEBUG_KMS("Enabling the SAGV\n"); @@ -2910,21 +2927,21 @@ skl_enable_sagv(struct drm_i915_private *dev_priv) * Some skl systems, pre-release machines in particular, * don't actually have an SAGV. */ - if (ret == -ENXIO) { + if (IS_SKYLAKE(dev_priv) && ret == -ENXIO) { DRM_DEBUG_DRIVER("No SAGV found on system, ignoring\n"); - dev_priv->skl_sagv_status = I915_SKL_SAGV_NOT_CONTROLLED; + dev_priv->sagv_status = I915_SAGV_NOT_CONTROLLED; return 0; } else if (ret < 0) { DRM_ERROR("Failed to enable the SAGV\n"); return ret; } - dev_priv->skl_sagv_status = I915_SKL_SAGV_ENABLED; + dev_priv->sagv_status = I915_SAGV_ENABLED; return 0; } static int -skl_do_sagv_disable(struct drm_i915_private *dev_priv) +intel_do_sagv_disable(struct drm_i915_private *dev_priv) { int ret; uint32_t temp = GEN9_SAGV_DISABLE; @@ -2938,19 +2955,21 @@ skl_do_sagv_disable(struct drm_i915_private *dev_priv) } int -skl_disable_sagv(struct drm_i915_private *dev_priv) +intel_disable_sagv(struct drm_i915_private *dev_priv) { int ret, result; - if (dev_priv->skl_sagv_status == I915_SKL_SAGV_NOT_CONTROLLED || - dev_priv->skl_sagv_status == I915_SKL_SAGV_DISABLED) + if (!intel_has_sagv(dev_priv)) + return 0; + + if (dev_priv->sagv_status == I915_SAGV_DISABLED) return 0; DRM_DEBUG_KMS("Disabling the SAGV\n"); mutex_lock(&dev_priv->rps.hw_lock); /* bspec says to keep retrying for at least 1 ms */ - ret = wait_for(result = skl_do_sagv_disable(dev_priv), 1); + ret = wait_for(result = intel_do_sagv_disable(dev_priv), 1); mutex_unlock(&dev_priv->rps.hw_lock); if (ret == -ETIMEDOUT) { @@ -2962,20 +2981,20 @@ skl_disable_sagv(struct drm_i915_private *dev_priv) * Some skl systems, pre-release machines in particular, * don't actually have an SAGV. */ - if (result == -ENXIO) { + if (IS_SKYLAKE(dev_priv) && result == -ENXIO) { DRM_DEBUG_DRIVER("No SAGV found on system, ignoring\n"); - dev_priv->skl_sagv_status = I915_SKL_SAGV_NOT_CONTROLLED; + dev_priv->sagv_status = I915_SAGV_NOT_CONTROLLED; return 0; } else if (result < 0) { DRM_ERROR("Failed to disable the SAGV\n"); return result; } - dev_priv->skl_sagv_status = I915_SKL_SAGV_DISABLED; + dev_priv->sagv_status = I915_SAGV_DISABLED; return 0; } -bool skl_can_enable_sagv(struct drm_atomic_state *state) +bool intel_can_enable_sagv(struct drm_atomic_state *state) { struct drm_device *dev = state->dev; struct drm_i915_private *dev_priv = to_i915(dev); @@ -2984,6 +3003,9 @@ bool skl_can_enable_sagv(struct drm_atomic_state *state) enum pipe pipe; int level, plane; + if (!intel_has_sagv(dev_priv)) + return false; + /* * SKL workaround: bspec recommends we disable the SAGV when we have * more then one pipe enabled @@ -3472,29 +3494,14 @@ static uint32_t skl_wm_method1(uint32_t pixel_rate, uint8_t cpp, uint32_t latenc } static uint32_t skl_wm_method2(uint32_t pixel_rate, uint32_t pipe_htotal, - uint32_t horiz_pixels, uint8_t cpp, - uint64_t tiling, uint32_t latency) + uint32_t latency, uint32_t plane_blocks_per_line) { uint32_t ret; - uint32_t plane_bytes_per_line, plane_blocks_per_line; uint32_t wm_intermediate_val; if (latency == 0) return UINT_MAX; - plane_bytes_per_line = horiz_pixels * cpp; - - if (tiling == I915_FORMAT_MOD_Y_TILED || - tiling == I915_FORMAT_MOD_Yf_TILED) { - plane_bytes_per_line *= 4; - plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512); - plane_blocks_per_line /= 4; - } else if (tiling == DRM_FORMAT_MOD_NONE) { - plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512) + 1; - } else { - plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512); - } - wm_intermediate_val = latency * pixel_rate; ret = DIV_ROUND_UP(wm_intermediate_val, pipe_htotal * 1000) * plane_blocks_per_line; @@ -3545,6 +3552,7 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, uint8_t cpp; uint32_t width = 0, height = 0; uint32_t plane_pixel_rate; + uint32_t y_tile_minimum, y_min_scanlines; if (latency == 0 || !cstate->base.active || !intel_pstate->base.visible) { *enabled = false; @@ -3560,38 +3568,51 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, cpp = drm_format_plane_cpp(fb->pixel_format, 0); plane_pixel_rate = skl_adjusted_plane_pixel_rate(cstate, intel_pstate); + if (drm_rotation_90_or_270(pstate->rotation)) { + int cpp = (fb->pixel_format == DRM_FORMAT_NV12) ? + drm_format_plane_cpp(fb->pixel_format, 1) : + drm_format_plane_cpp(fb->pixel_format, 0); + + switch (cpp) { + case 1: + y_min_scanlines = 16; + break; + case 2: + y_min_scanlines = 8; + break; + default: + WARN(1, "Unsupported pixel depth for rotation"); + case 4: + y_min_scanlines = 4; + break; + } + } else { + y_min_scanlines = 4; + } + + plane_bytes_per_line = width * cpp; + if (fb->modifier[0] == I915_FORMAT_MOD_Y_TILED || + fb->modifier[0] == I915_FORMAT_MOD_Yf_TILED) { + plane_blocks_per_line = + DIV_ROUND_UP(plane_bytes_per_line * y_min_scanlines, 512); + plane_blocks_per_line /= y_min_scanlines; + } else if (fb->modifier[0] == DRM_FORMAT_MOD_NONE) { + plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512) + + 1; + } else { + plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512); + } + method1 = skl_wm_method1(plane_pixel_rate, cpp, latency); method2 = skl_wm_method2(plane_pixel_rate, cstate->base.adjusted_mode.crtc_htotal, - width, - cpp, - fb->modifier[0], - latency); + latency, + plane_blocks_per_line); - plane_bytes_per_line = width * cpp; - plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512); + y_tile_minimum = plane_blocks_per_line * y_min_scanlines; if (fb->modifier[0] == I915_FORMAT_MOD_Y_TILED || fb->modifier[0] == I915_FORMAT_MOD_Yf_TILED) { - uint32_t min_scanlines = 4; - uint32_t y_tile_minimum; - if (drm_rotation_90_or_270(pstate->rotation)) { - int cpp = (fb->pixel_format == DRM_FORMAT_NV12) ? - drm_format_plane_cpp(fb->pixel_format, 1) : - drm_format_plane_cpp(fb->pixel_format, 0); - - switch (cpp) { - case 1: - min_scanlines = 16; - break; - case 2: - min_scanlines = 8; - break; - case 8: - WARN(1, "Unsupported pixel depth for rotation"); - } - } - y_tile_minimum = plane_blocks_per_line * min_scanlines; selected_result = max(method2, y_tile_minimum); } else { if ((ddb_allocation / plane_blocks_per_line) >= 1) @@ -3605,10 +3626,12 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, if (level >= 1 && level <= 7) { if (fb->modifier[0] == I915_FORMAT_MOD_Y_TILED || - fb->modifier[0] == I915_FORMAT_MOD_Yf_TILED) - res_lines += 4; - else + fb->modifier[0] == I915_FORMAT_MOD_Yf_TILED) { + res_blocks += y_tile_minimum; + res_lines += y_min_scanlines; + } else { res_blocks++; + } } if (res_blocks >= ddb_allocation || res_lines > 31) { @@ -3939,6 +3962,41 @@ pipes_modified(struct drm_atomic_state *state) return ret; } +int +skl_ddb_add_affected_planes(struct intel_crtc_state *cstate) +{ + struct drm_atomic_state *state = cstate->base.state; + struct drm_device *dev = state->dev; + struct drm_crtc *crtc = cstate->base.crtc; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_atomic_state *intel_state = to_intel_atomic_state(state); + struct skl_ddb_allocation *new_ddb = &intel_state->wm_results.ddb; + struct skl_ddb_allocation *cur_ddb = &dev_priv->wm.skl_hw.ddb; + struct drm_plane_state *plane_state; + struct drm_plane *plane; + enum pipe pipe = intel_crtc->pipe; + int id; + + WARN_ON(!drm_atomic_get_existing_crtc_state(state, crtc)); + + drm_for_each_plane_mask(plane, dev, crtc->state->plane_mask) { + id = skl_wm_plane_id(to_intel_plane(plane)); + + if (skl_ddb_entry_equal(&cur_ddb->plane[pipe][id], + &new_ddb->plane[pipe][id]) && + skl_ddb_entry_equal(&cur_ddb->y_plane[pipe][id], + &new_ddb->y_plane[pipe][id])) + continue; + + plane_state = drm_atomic_get_plane_state(state, plane); + if (IS_ERR(plane_state)) + return PTR_ERR(plane_state); + } + + return 0; +} + static int skl_compute_ddb(struct drm_atomic_state *state) { @@ -4003,7 +4061,7 @@ skl_compute_ddb(struct drm_atomic_state *state) if (ret) return ret; - ret = drm_atomic_add_affected_planes(state, &intel_crtc->base); + ret = skl_ddb_add_affected_planes(cstate); if (ret) return ret; } diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 7a74750076c5..ed9955dce156 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -564,7 +564,7 @@ static int init_ring_common(struct intel_engine_cs *engine) else intel_ring_setup_status_page(engine); - intel_engine_reset_irq(engine); + intel_engine_reset_breadcrumbs(engine); /* Enforce ordering by reading HEAD register back */ I915_READ_HEAD(engine); diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 7f64d611159b..ec0b4a0c605d 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -328,6 +328,7 @@ struct intel_engine_cs { * inspecting request list. */ u32 last_submitted_seqno; + u32 last_pending_seqno; /* An RCU guarded pointer to the last request. No reference is * held to the request, users must carefully acquire a reference to @@ -492,7 +493,6 @@ int __intel_ring_space(int head, int tail, int size); void intel_ring_update_space(struct intel_ring *ring); void intel_engine_init_seqno(struct intel_engine_cs *engine, u32 seqno); -void intel_engine_reset_irq(struct intel_engine_cs *engine); void intel_engine_setup_common(struct intel_engine_cs *engine); int intel_engine_init_common(struct intel_engine_cs *engine); @@ -584,6 +584,7 @@ static inline bool intel_engine_wakeup(const struct intel_engine_cs *engine) return wakeup; } +void intel_engine_reset_breadcrumbs(struct intel_engine_cs *engine); void intel_engine_fini_breadcrumbs(struct intel_engine_cs *engine); unsigned int intel_kick_waiters(struct drm_i915_private *i915); unsigned int intel_kick_signalers(struct drm_i915_private *i915); diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index a9b6c936aadd..ee2306a79747 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -796,10 +796,9 @@ __unclaimed_reg_debug(struct drm_i915_private *dev_priv, const bool read, const bool before) { - if (WARN(check_for_unclaimed_mmio(dev_priv), - "Unclaimed register detected %s %s register 0x%x\n", - before ? "before" : "after", - read ? "reading" : "writing to", + if (WARN(check_for_unclaimed_mmio(dev_priv) && !before, + "Unclaimed %s register 0x%x\n", + read ? "read from" : "write to", i915_mmio_reg_offset(reg))) i915.mmio_debug--; /* Only report the first N failures */ } diff --git a/drivers/gpu/drm/msm/msm_perf.c b/drivers/gpu/drm/msm/msm_perf.c index 17fe4e53e0d1..1627294575cb 100644 --- a/drivers/gpu/drm/msm/msm_perf.c +++ b/drivers/gpu/drm/msm/msm_perf.c @@ -229,8 +229,8 @@ int msm_perf_debugfs_init(struct drm_minor *minor) perf->ent = debugfs_create_file("perf", S_IFREG | S_IRUGO, minor->debugfs_root, perf, &perf_debugfs_fops); if (!perf->ent) { - DRM_ERROR("Cannot create /sys/kernel/debug/dri/%s/perf\n", - minor->debugfs_root->d_name.name); + DRM_ERROR("Cannot create /sys/kernel/debug/dri/%pd/perf\n", + minor->debugfs_root); goto fail; } diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index 3a5fdfcd67ae..8487f461f05f 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -243,8 +243,8 @@ int msm_rd_debugfs_init(struct drm_minor *minor) rd->ent = debugfs_create_file("rd", S_IFREG | S_IRUGO, minor->debugfs_root, rd, &rd_debugfs_fops); if (!rd->ent) { - DRM_ERROR("Cannot create /sys/kernel/debug/dri/%s/rd\n", - minor->debugfs_root->d_name.name); + DRM_ERROR("Cannot create /sys/kernel/debug/dri/%pd/rd\n", + minor->debugfs_root); goto fail; } diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h index 7ea8aa7ca408..6bc712f32c8b 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h @@ -175,6 +175,7 @@ struct nvkm_device_func { void (*fini)(struct nvkm_device *, bool suspend); resource_size_t (*resource_addr)(struct nvkm_device *, unsigned bar); resource_size_t (*resource_size)(struct nvkm_device *, unsigned bar); + bool cpu_coherent; }; struct nvkm_device_quirk { diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index 66f31c3eb8ba..343b8659472c 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -209,7 +209,8 @@ nouveau_bo_new(struct drm_device *dev, int size, int align, nvbo->tile_flags = tile_flags; nvbo->bo.bdev = &drm->ttm.bdev; - nvbo->force_coherent = flags & TTM_PL_FLAG_UNCACHED; + if (!nvxx_device(&drm->device)->func->cpu_coherent) + nvbo->force_coherent = flags & TTM_PL_FLAG_UNCACHED; nvbo->page_shift = 12; if (drm->client.vm) { diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c index b1b693219db3..62ad0300cfa5 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c @@ -1614,6 +1614,7 @@ nvkm_device_pci_func = { .fini = nvkm_device_pci_fini, .resource_addr = nvkm_device_pci_resource_addr, .resource_size = nvkm_device_pci_resource_size, + .cpu_coherent = !IS_ENABLED(CONFIG_ARM), }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c index 939682f18788..9b638bd905ff 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c @@ -245,6 +245,7 @@ nvkm_device_tegra_func = { .fini = nvkm_device_tegra_fini, .resource_addr = nvkm_device_tegra_resource_addr, .resource_size = nvkm_device_tegra_resource_size, + .cpu_coherent = false, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/dmanv04.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/dmanv04.c index edec30fd3ecd..0a7b6ed5ed28 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/dmanv04.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/dmanv04.c @@ -37,7 +37,10 @@ nv04_fifo_dma_object_dtor(struct nvkm_fifo_chan *base, int cookie) { struct nv04_fifo_chan *chan = nv04_fifo_chan(base); struct nvkm_instmem *imem = chan->fifo->base.engine.subdev.device->imem; + + mutex_lock(&chan->fifo->base.engine.subdev.mutex); nvkm_ramht_remove(imem->ramht, cookie); + mutex_unlock(&chan->fifo->base.engine.subdev.mutex); } static int diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index a4e9f35da3a2..74f99bac08b1 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -1638,8 +1638,8 @@ static int avivo_crtc_do_set_base(struct drm_crtc *crtc, WREG32(AVIVO_D1MODE_VIEWPORT_SIZE + radeon_crtc->crtc_offset, (viewport_w << 16) | viewport_h); - /* set pageflip to happen anywhere in vblank interval */ - WREG32(AVIVO_D1MODE_MASTER_UPDATE_MODE + radeon_crtc->crtc_offset, 0); + /* set pageflip to happen only at start of vblank interval (front porch) */ + WREG32(AVIVO_D1MODE_MASTER_UPDATE_MODE + radeon_crtc->crtc_offset, 3); if (!atomic && fb && fb != crtc->primary->fb) { radeon_fb = to_radeon_framebuffer(fb); diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c index 6a4b020dd0b4..5a26eb4545aa 100644 --- a/drivers/gpu/drm/radeon/r600_dpm.c +++ b/drivers/gpu/drm/radeon/r600_dpm.c @@ -156,19 +156,20 @@ u32 r600_dpm_get_vblank_time(struct radeon_device *rdev) struct drm_device *dev = rdev->ddev; struct drm_crtc *crtc; struct radeon_crtc *radeon_crtc; - u32 line_time_us, vblank_lines; + u32 vblank_in_pixels; u32 vblank_time_us = 0xffffffff; /* if the displays are off, vblank time is max */ if (rdev->num_crtc && rdev->mode_info.mode_config_initialized) { list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { radeon_crtc = to_radeon_crtc(crtc); if (crtc->enabled && radeon_crtc->enabled && radeon_crtc->hw_mode.clock) { - line_time_us = (radeon_crtc->hw_mode.crtc_htotal * 1000) / - radeon_crtc->hw_mode.clock; - vblank_lines = radeon_crtc->hw_mode.crtc_vblank_end - - radeon_crtc->hw_mode.crtc_vdisplay + - (radeon_crtc->v_border * 2); - vblank_time_us = vblank_lines * line_time_us; + vblank_in_pixels = + radeon_crtc->hw_mode.crtc_htotal * + (radeon_crtc->hw_mode.crtc_vblank_end - + radeon_crtc->hw_mode.crtc_vdisplay + + (radeon_crtc->v_border * 2)); + + vblank_time_us = vblank_in_pixels * 1000 / radeon_crtc->hw_mode.clock; break; } } diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 50e96d2c593d..e18839d52e3e 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -927,6 +927,16 @@ radeon_lvds_detect(struct drm_connector *connector, bool force) return ret; } +static void radeon_connector_unregister(struct drm_connector *connector) +{ + struct radeon_connector *radeon_connector = to_radeon_connector(connector); + + if (radeon_connector->ddc_bus->has_aux) { + drm_dp_aux_unregister(&radeon_connector->ddc_bus->aux); + radeon_connector->ddc_bus->has_aux = false; + } +} + static void radeon_connector_destroy(struct drm_connector *connector) { struct radeon_connector *radeon_connector = to_radeon_connector(connector); @@ -984,6 +994,7 @@ static const struct drm_connector_funcs radeon_lvds_connector_funcs = { .dpms = drm_helper_connector_dpms, .detect = radeon_lvds_detect, .fill_modes = drm_helper_probe_single_connector_modes, + .early_unregister = radeon_connector_unregister, .destroy = radeon_connector_destroy, .set_property = radeon_lvds_set_property, }; @@ -1111,6 +1122,7 @@ static const struct drm_connector_funcs radeon_vga_connector_funcs = { .dpms = drm_helper_connector_dpms, .detect = radeon_vga_detect, .fill_modes = drm_helper_probe_single_connector_modes, + .early_unregister = radeon_connector_unregister, .destroy = radeon_connector_destroy, .set_property = radeon_connector_set_property, }; @@ -1188,6 +1200,7 @@ static const struct drm_connector_funcs radeon_tv_connector_funcs = { .dpms = drm_helper_connector_dpms, .detect = radeon_tv_detect, .fill_modes = drm_helper_probe_single_connector_modes, + .early_unregister = radeon_connector_unregister, .destroy = radeon_connector_destroy, .set_property = radeon_connector_set_property, }; @@ -1519,6 +1532,7 @@ static const struct drm_connector_funcs radeon_dvi_connector_funcs = { .detect = radeon_dvi_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = radeon_connector_set_property, + .early_unregister = radeon_connector_unregister, .destroy = radeon_connector_destroy, .force = radeon_dvi_force, }; @@ -1832,6 +1846,7 @@ static const struct drm_connector_funcs radeon_dp_connector_funcs = { .detect = radeon_dp_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = radeon_connector_set_property, + .early_unregister = radeon_connector_unregister, .destroy = radeon_connector_destroy, .force = radeon_dvi_force, }; @@ -1841,6 +1856,7 @@ static const struct drm_connector_funcs radeon_edp_connector_funcs = { .detect = radeon_dp_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = radeon_lvds_set_property, + .early_unregister = radeon_connector_unregister, .destroy = radeon_connector_destroy, .force = radeon_dvi_force, }; @@ -1850,6 +1866,7 @@ static const struct drm_connector_funcs radeon_lvds_bridge_connector_funcs = { .detect = radeon_dp_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = radeon_lvds_set_property, + .early_unregister = radeon_connector_unregister, .destroy = radeon_connector_destroy, .force = radeon_dvi_force, }; diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index 890171f08987..cdb8cb568c15 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -321,16 +321,30 @@ void radeon_crtc_handle_vblank(struct radeon_device *rdev, int crtc_id) update_pending = radeon_page_flip_pending(rdev, crtc_id); /* Has the pageflip already completed in crtc, or is it certain - * to complete in this vblank? + * to complete in this vblank? GET_DISTANCE_TO_VBLANKSTART provides + * distance to start of "fudged earlier" vblank in vpos, distance to + * start of real vblank in hpos. vpos >= 0 && hpos < 0 means we are in + * the last few scanlines before start of real vblank, where the vblank + * irq can fire, so we have sampled update_pending a bit too early and + * know the flip will complete at leading edge of the upcoming real + * vblank. On pre-AVIVO hardware, flips also complete inside the real + * vblank, not only at leading edge, so if update_pending for hpos >= 0 + * == inside real vblank, the flip will complete almost immediately. + * Note that this method of completion handling is still not 100% race + * free, as we could execute before the radeon_flip_work_func managed + * to run and set the RADEON_FLIP_SUBMITTED status, thereby we no-op, + * but the flip still gets programmed into hw and completed during + * vblank, leading to a delayed emission of the flip completion event. + * This applies at least to pre-AVIVO hardware, where flips are always + * completing inside vblank, not only at leading edge of vblank. */ if (update_pending && - (DRM_SCANOUTPOS_VALID & radeon_get_crtc_scanoutpos(rdev->ddev, - crtc_id, - USE_REAL_VBLANKSTART, - &vpos, &hpos, NULL, NULL, - &rdev->mode_info.crtcs[crtc_id]->base.hwmode)) && - ((vpos >= (99 * rdev->mode_info.crtcs[crtc_id]->base.hwmode.crtc_vdisplay)/100) || - (vpos < 0 && !ASIC_IS_AVIVO(rdev)))) { + (DRM_SCANOUTPOS_VALID & + radeon_get_crtc_scanoutpos(rdev->ddev, crtc_id, + GET_DISTANCE_TO_VBLANKSTART, + &vpos, &hpos, NULL, NULL, + &rdev->mode_info.crtcs[crtc_id]->base.hwmode)) && + ((vpos >= 0 && hpos < 0) || (hpos >= 0 && !ASIC_IS_AVIVO(rdev)))) { /* crtc didn't flip in this target vblank interval, * but flip is pending in crtc. Based on the current * scanout position we know that the current frame is @@ -438,16 +452,19 @@ static void radeon_flip_work_func(struct work_struct *__work) } /* Wait until we're out of the vertical blank period before the one - * targeted by the flip + * targeted by the flip. Always wait on pre DCE4 to avoid races with + * flip completion handling from vblank irq, as these old asics don't + * have reliable pageflip completion interrupts. */ while (radeon_crtc->enabled && - (radeon_get_crtc_scanoutpos(dev, work->crtc_id, 0, - &vpos, &hpos, NULL, NULL, - &crtc->hwmode) + (radeon_get_crtc_scanoutpos(dev, work->crtc_id, 0, + &vpos, &hpos, NULL, NULL, + &crtc->hwmode) & (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK)) == - (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK) && - (int)(work->target_vblank - - dev->driver->get_vblank_counter(dev, work->crtc_id)) > 0) + (DRM_SCANOUTPOS_VALID | DRM_SCANOUTPOS_IN_VBLANK) && + (!ASIC_IS_AVIVO(rdev) || + ((int) (work->target_vblank - + dev->driver->get_vblank_counter(dev, work->crtc_id)) > 0))) usleep_range(1000, 2000); /* We borrow the event spin lock for protecting flip_status */ @@ -1658,20 +1675,20 @@ int radeon_modeset_init(struct radeon_device *rdev) void radeon_modeset_fini(struct radeon_device *rdev) { - radeon_fbdev_fini(rdev); - kfree(rdev->mode_info.bios_hardcoded_edid); - - /* free i2c buses */ - radeon_i2c_fini(rdev); - if (rdev->mode_info.mode_config_initialized) { - radeon_afmt_fini(rdev); drm_kms_helper_poll_fini(rdev->ddev); radeon_hpd_fini(rdev); drm_crtc_force_disable_all(rdev->ddev); + radeon_fbdev_fini(rdev); + radeon_afmt_fini(rdev); drm_mode_config_cleanup(rdev->ddev); rdev->mode_info.mode_config_initialized = false; } + + kfree(rdev->mode_info.bios_hardcoded_edid); + + /* free i2c buses */ + radeon_i2c_fini(rdev); } static bool is_hdtv_mode(const struct drm_display_mode *mode) diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index 91c8f4339566..00ea0002b539 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -96,9 +96,10 @@ * 2.45.0 - Allow setting shader registers using DMA/COPY packet3 on SI * 2.46.0 - Add PFP_SYNC_ME support on evergreen * 2.47.0 - Add UVD_NO_OP register support + * 2.48.0 - TA_CS_BC_BASE_ADDR allowed on SI */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 47 +#define KMS_DRIVER_MINOR 48 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); diff --git a/drivers/gpu/drm/radeon/radeon_i2c.c b/drivers/gpu/drm/radeon/radeon_i2c.c index 021aa005623f..29f7817af821 100644 --- a/drivers/gpu/drm/radeon/radeon_i2c.c +++ b/drivers/gpu/drm/radeon/radeon_i2c.c @@ -982,9 +982,8 @@ void radeon_i2c_destroy(struct radeon_i2c_chan *i2c) { if (!i2c) return; + WARN_ON(i2c->has_aux); i2c_del_adapter(&i2c->adapter); - if (i2c->has_aux) - drm_dp_aux_unregister(&i2c->aux); kfree(i2c); } diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index 455268214b89..3de5e6e21662 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -566,7 +566,8 @@ static int radeon_ttm_tt_pin_userptr(struct ttm_tt *ttm) uint64_t userptr = gtt->userptr + pinned * PAGE_SIZE; struct page **pages = ttm->pages + pinned; - r = get_user_pages(userptr, num_pages, write, 0, pages, NULL); + r = get_user_pages(userptr, num_pages, write ? FOLL_WRITE : 0, + pages, NULL); if (r < 0) goto release_pages; diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c index 76c55c5d11ec..c55d653aaf5f 100644 --- a/drivers/gpu/drm/radeon/rv515.c +++ b/drivers/gpu/drm/radeon/rv515.c @@ -406,8 +406,9 @@ void rv515_mc_resume(struct radeon_device *rdev, struct rv515_mc_save *save) for (i = 0; i < rdev->num_crtc; i++) { if (save->crtc_enabled[i]) { tmp = RREG32(AVIVO_D1MODE_MASTER_UPDATE_MODE + crtc_offsets[i]); - if ((tmp & 0x7) != 0) { + if ((tmp & 0x7) != 3) { tmp &= ~0x7; + tmp |= 0x3; WREG32(AVIVO_D1MODE_MASTER_UPDATE_MODE + crtc_offsets[i], tmp); } tmp = RREG32(AVIVO_D1GRPH_UPDATE + crtc_offsets[i]); diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 7ee9aafbdf74..e402be8821c4 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -4431,6 +4431,7 @@ static bool si_vm_reg_valid(u32 reg) case SPI_CONFIG_CNTL: case SPI_CONFIG_CNTL_1: case TA_CNTL_AUX: + case TA_CS_BC_BASE_ADDR: return true; default: DRM_ERROR("Invalid register 0x%x in CS\n", reg); diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index ba7e3d19b38d..89bdf20344ae 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -3015,6 +3015,12 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev, if (rdev->pdev->device == 0x6811 && rdev->pdev->revision == 0x81) max_mclk = 120000; + /* limit sclk/mclk on Jet parts for stability */ + if (rdev->pdev->device == 0x6665 && + rdev->pdev->revision == 0xc3) { + max_sclk = 75000; + max_mclk = 80000; + } if (rps->vce_active) { rps->evclk = rdev->pm.dpm.vce_states[rdev->pm.dpm.vce_level].evclk; diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h index eb220eecba78..65a911ddd509 100644 --- a/drivers/gpu/drm/radeon/sid.h +++ b/drivers/gpu/drm/radeon/sid.h @@ -1145,6 +1145,7 @@ #define SPI_LB_CU_MASK 0x9354 #define TA_CNTL_AUX 0x9508 +#define TA_CS_BC_BASE_ADDR 0x950C #define CC_RB_BACKEND_DISABLE 0x98F4 #define BACKEND_DISABLE(x) ((x) << 16) diff --git a/drivers/gpu/drm/udl/udl_fb.c b/drivers/gpu/drm/udl/udl_fb.c index 9688bfa92ccd..611b6b9bb3cb 100644 --- a/drivers/gpu/drm/udl/udl_fb.c +++ b/drivers/gpu/drm/udl/udl_fb.c @@ -122,7 +122,7 @@ int udl_handle_damage(struct udl_framebuffer *fb, int x, int y, return 0; cmd = urb->transfer_buffer; - for (i = y; i < height ; i++) { + for (i = y; i < y + height ; i++) { const int line_offset = fb->base.pitches[0] * i; const int byte_offset = line_offset + (x * bpp); const int dev_byte_offset = (fb->base.width * bpp * i) + (x * bpp); diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c index 2682f07d8f1e..7f08d681a74b 100644 --- a/drivers/gpu/drm/vc4/vc4_crtc.c +++ b/drivers/gpu/drm/vc4/vc4_crtc.c @@ -229,7 +229,7 @@ int vc4_crtc_get_scanoutpos(struct drm_device *dev, unsigned int crtc_id, * and need to make things up in a approximative but consistent way. */ ret |= DRM_SCANOUTPOS_IN_VBLANK; - vblank_lines = mode->crtc_vtotal - mode->crtc_vdisplay; + vblank_lines = mode->vtotal - mode->vdisplay; if (flags & DRM_CALLED_FROM_VBLIRQ) { /* @@ -378,7 +378,7 @@ static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc) struct drm_crtc_state *state = crtc->state; struct drm_display_mode *mode = &state->adjusted_mode; bool interlace = mode->flags & DRM_MODE_FLAG_INTERLACE; - u32 vactive = (mode->vdisplay >> (interlace ? 1 : 0)); + u32 pixel_rep = (mode->flags & DRM_MODE_FLAG_DBLCLK) ? 2 : 1; u32 format = PV_CONTROL_FORMAT_24; bool debug_dump_regs = false; int clock_select = vc4_get_clock_select(crtc); @@ -394,47 +394,65 @@ static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc) CRTC_WRITE(PV_CONTROL, 0); CRTC_WRITE(PV_HORZA, - VC4_SET_FIELD(mode->htotal - mode->hsync_end, + VC4_SET_FIELD((mode->htotal - + mode->hsync_end) * pixel_rep, PV_HORZA_HBP) | - VC4_SET_FIELD(mode->hsync_end - mode->hsync_start, + VC4_SET_FIELD((mode->hsync_end - + mode->hsync_start) * pixel_rep, PV_HORZA_HSYNC)); CRTC_WRITE(PV_HORZB, - VC4_SET_FIELD(mode->hsync_start - mode->hdisplay, + VC4_SET_FIELD((mode->hsync_start - + mode->hdisplay) * pixel_rep, PV_HORZB_HFP) | - VC4_SET_FIELD(mode->hdisplay, PV_HORZB_HACTIVE)); + VC4_SET_FIELD(mode->hdisplay * pixel_rep, PV_HORZB_HACTIVE)); CRTC_WRITE(PV_VERTA, - VC4_SET_FIELD(mode->vtotal - mode->vsync_end, + VC4_SET_FIELD(mode->crtc_vtotal - mode->crtc_vsync_end, PV_VERTA_VBP) | - VC4_SET_FIELD(mode->vsync_end - mode->vsync_start, + VC4_SET_FIELD(mode->crtc_vsync_end - mode->crtc_vsync_start, PV_VERTA_VSYNC)); CRTC_WRITE(PV_VERTB, - VC4_SET_FIELD(mode->vsync_start - mode->vdisplay, + VC4_SET_FIELD(mode->crtc_vsync_start - mode->crtc_vdisplay, PV_VERTB_VFP) | - VC4_SET_FIELD(vactive, PV_VERTB_VACTIVE)); + VC4_SET_FIELD(mode->crtc_vdisplay, PV_VERTB_VACTIVE)); if (interlace) { CRTC_WRITE(PV_VERTA_EVEN, - VC4_SET_FIELD(mode->vtotal - mode->vsync_end - 1, + VC4_SET_FIELD(mode->crtc_vtotal - + mode->crtc_vsync_end - 1, PV_VERTA_VBP) | - VC4_SET_FIELD(mode->vsync_end - mode->vsync_start, + VC4_SET_FIELD(mode->crtc_vsync_end - + mode->crtc_vsync_start, PV_VERTA_VSYNC)); CRTC_WRITE(PV_VERTB_EVEN, - VC4_SET_FIELD(mode->vsync_start - mode->vdisplay, + VC4_SET_FIELD(mode->crtc_vsync_start - + mode->crtc_vdisplay, PV_VERTB_VFP) | - VC4_SET_FIELD(vactive, PV_VERTB_VACTIVE)); + VC4_SET_FIELD(mode->crtc_vdisplay, PV_VERTB_VACTIVE)); + + /* We set up first field even mode for HDMI. VEC's + * NTSC mode would want first field odd instead, once + * we support it (to do so, set ODD_FIRST and put the + * delay in VSYNCD_EVEN instead). + */ + CRTC_WRITE(PV_V_CONTROL, + PV_VCONTROL_CONTINUOUS | + PV_VCONTROL_INTERLACE | + VC4_SET_FIELD(mode->htotal * pixel_rep / 2, + PV_VCONTROL_ODD_DELAY)); + CRTC_WRITE(PV_VSYNCD_EVEN, 0); + } else { + CRTC_WRITE(PV_V_CONTROL, PV_VCONTROL_CONTINUOUS); } - CRTC_WRITE(PV_HACT_ACT, mode->hdisplay); + CRTC_WRITE(PV_HACT_ACT, mode->hdisplay * pixel_rep); - CRTC_WRITE(PV_V_CONTROL, - PV_VCONTROL_CONTINUOUS | - (interlace ? PV_VCONTROL_INTERLACE : 0)); CRTC_WRITE(PV_CONTROL, VC4_SET_FIELD(format, PV_CONTROL_FORMAT) | VC4_SET_FIELD(vc4_get_fifo_full_level(format), PV_CONTROL_FIFO_LEVEL) | + VC4_SET_FIELD(pixel_rep - 1, PV_CONTROL_PIXEL_REP) | PV_CONTROL_CLR_AT_START | PV_CONTROL_TRIGGER_UNDERFLOW | PV_CONTROL_WAIT_HSTART | @@ -544,16 +562,6 @@ static bool vc4_crtc_mode_fixup(struct drm_crtc *crtc, return false; } - /* - * Interlaced video modes got CRTC_INTERLACE_HALVE_V applied when - * coming from user space. We don't want this, as it screws up - * vblank timestamping, so fix it up. - */ - drm_mode_set_crtcinfo(adjusted_mode, 0); - - DRM_DEBUG_KMS("[CRTC:%d] adjusted_mode :\n", crtc->base.id); - drm_mode_debug_printmodeline(adjusted_mode); - return true; } diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 428e24919ef1..7c1e4d97486f 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -122,9 +122,16 @@ to_vc4_dev(struct drm_device *dev) struct vc4_bo { struct drm_gem_cma_object base; - /* seqno of the last job to render to this BO. */ + /* seqno of the last job to render using this BO. */ uint64_t seqno; + /* seqno of the last job to use the RCL to write to this BO. + * + * Note that this doesn't include binner overflow memory + * writes. + */ + uint64_t write_seqno; + /* List entry for the BO's position in either * vc4_exec_info->unref_list or vc4_dev->bo_cache.time_list */ @@ -216,6 +223,9 @@ struct vc4_exec_info { /* Sequence number for this bin/render job. */ uint64_t seqno; + /* Latest write_seqno of any BO that binning depends on. */ + uint64_t bin_dep_seqno; + /* Last current addresses the hardware was processing when the * hangcheck timer checked on us. */ @@ -230,6 +240,13 @@ struct vc4_exec_info { struct drm_gem_cma_object **bo; uint32_t bo_count; + /* List of BOs that are being written by the RCL. Other than + * the binner temporary storage, this is all the BOs written + * by the job. + */ + struct drm_gem_cma_object *rcl_write_bo[4]; + uint32_t rcl_write_bo_count; + /* Pointers for our position in vc4->job_list */ struct list_head head; @@ -307,18 +324,15 @@ struct vc4_exec_info { static inline struct vc4_exec_info * vc4_first_bin_job(struct vc4_dev *vc4) { - if (list_empty(&vc4->bin_job_list)) - return NULL; - return list_first_entry(&vc4->bin_job_list, struct vc4_exec_info, head); + return list_first_entry_or_null(&vc4->bin_job_list, + struct vc4_exec_info, head); } static inline struct vc4_exec_info * vc4_first_render_job(struct vc4_dev *vc4) { - if (list_empty(&vc4->render_job_list)) - return NULL; - return list_first_entry(&vc4->render_job_list, - struct vc4_exec_info, head); + return list_first_entry_or_null(&vc4->render_job_list, + struct vc4_exec_info, head); } static inline struct vc4_exec_info * diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c index 77daea6cb866..47a095f392f8 100644 --- a/drivers/gpu/drm/vc4/vc4_gem.c +++ b/drivers/gpu/drm/vc4/vc4_gem.c @@ -467,6 +467,11 @@ vc4_update_bo_seqnos(struct vc4_exec_info *exec, uint64_t seqno) list_for_each_entry(bo, &exec->unref_list, unref_head) { bo->seqno = seqno; } + + for (i = 0; i < exec->rcl_write_bo_count; i++) { + bo = to_vc4_bo(&exec->rcl_write_bo[i]->base); + bo->write_seqno = seqno; + } } /* Queues a struct vc4_exec_info for execution. If no job is @@ -669,6 +674,14 @@ vc4_get_bcl(struct drm_device *dev, struct vc4_exec_info *exec) goto fail; ret = vc4_validate_shader_recs(dev, exec); + if (ret) + goto fail; + + /* Block waiting on any previous rendering into the CS's VBO, + * IB, or textures, so that pixels are actually written by the + * time we try to read them. + */ + ret = vc4_wait_for_seqno(dev, exec->bin_dep_seqno, ~0ull, true); fail: drm_free_large(temp); diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index 68ad10634b29..c4cb2e26de32 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -62,6 +62,8 @@ struct vc4_hdmi { struct vc4_hdmi_encoder { struct vc4_encoder base; bool hdmi_monitor; + bool limited_rgb_range; + bool rgb_range_selectable; }; static inline struct vc4_hdmi_encoder * @@ -174,6 +176,9 @@ vc4_hdmi_connector_detect(struct drm_connector *connector, bool force) return connector_status_disconnected; } + if (drm_probe_ddc(vc4->hdmi->ddc)) + return connector_status_connected; + if (HDMI_READ(VC4_HDMI_HOTPLUG) & VC4_HDMI_HOTPLUG_CONNECTED) return connector_status_connected; else @@ -202,41 +207,22 @@ static int vc4_hdmi_connector_get_modes(struct drm_connector *connector) return -ENODEV; vc4_encoder->hdmi_monitor = drm_detect_hdmi_monitor(edid); + + if (edid && edid->input & DRM_EDID_INPUT_DIGITAL) { + vc4_encoder->rgb_range_selectable = + drm_rgb_quant_range_selectable(edid); + } + drm_mode_connector_update_edid_property(connector, edid); ret = drm_add_edid_modes(connector, edid); return ret; } -/* - * drm_helper_probe_single_connector_modes() applies drm_mode_set_crtcinfo to - * all modes with flag CRTC_INTERLACE_HALVE_V. We don't want this, as it - * screws up vblank timestamping for interlaced modes, so fix it up. - */ -static int vc4_hdmi_connector_probe_modes(struct drm_connector *connector, - uint32_t maxX, uint32_t maxY) -{ - struct drm_display_mode *mode; - int count; - - count = drm_helper_probe_single_connector_modes(connector, maxX, maxY); - if (count == 0) - return 0; - - DRM_DEBUG_KMS("[CONNECTOR:%d:%s] probed adapted modes :\n", - connector->base.id, connector->name); - list_for_each_entry(mode, &connector->modes, head) { - drm_mode_set_crtcinfo(mode, 0); - drm_mode_debug_printmodeline(mode); - } - - return count; -} - static const struct drm_connector_funcs vc4_hdmi_connector_funcs = { .dpms = drm_atomic_helper_connector_dpms, .detect = vc4_hdmi_connector_detect, - .fill_modes = vc4_hdmi_connector_probe_modes, + .fill_modes = drm_helper_probe_single_connector_modes, .destroy = vc4_hdmi_connector_destroy, .reset = drm_atomic_helper_connector_reset, .atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state, @@ -294,25 +280,143 @@ static const struct drm_encoder_funcs vc4_hdmi_encoder_funcs = { .destroy = vc4_hdmi_encoder_destroy, }; +static int vc4_hdmi_stop_packet(struct drm_encoder *encoder, + enum hdmi_infoframe_type type) +{ + struct drm_device *dev = encoder->dev; + struct vc4_dev *vc4 = to_vc4_dev(dev); + u32 packet_id = type - 0x80; + + HDMI_WRITE(VC4_HDMI_RAM_PACKET_CONFIG, + HDMI_READ(VC4_HDMI_RAM_PACKET_CONFIG) & ~BIT(packet_id)); + + return wait_for(!(HDMI_READ(VC4_HDMI_RAM_PACKET_STATUS) & + BIT(packet_id)), 100); +} + +static void vc4_hdmi_write_infoframe(struct drm_encoder *encoder, + union hdmi_infoframe *frame) +{ + struct drm_device *dev = encoder->dev; + struct vc4_dev *vc4 = to_vc4_dev(dev); + u32 packet_id = frame->any.type - 0x80; + u32 packet_reg = VC4_HDMI_GCP_0 + VC4_HDMI_PACKET_STRIDE * packet_id; + uint8_t buffer[VC4_HDMI_PACKET_STRIDE]; + ssize_t len, i; + int ret; + + WARN_ONCE(!(HDMI_READ(VC4_HDMI_RAM_PACKET_CONFIG) & + VC4_HDMI_RAM_PACKET_ENABLE), + "Packet RAM has to be on to store the packet."); + + len = hdmi_infoframe_pack(frame, buffer, sizeof(buffer)); + if (len < 0) + return; + + ret = vc4_hdmi_stop_packet(encoder, frame->any.type); + if (ret) { + DRM_ERROR("Failed to wait for infoframe to go idle: %d\n", ret); + return; + } + + for (i = 0; i < len; i += 7) { + HDMI_WRITE(packet_reg, + buffer[i + 0] << 0 | + buffer[i + 1] << 8 | + buffer[i + 2] << 16); + packet_reg += 4; + + HDMI_WRITE(packet_reg, + buffer[i + 3] << 0 | + buffer[i + 4] << 8 | + buffer[i + 5] << 16 | + buffer[i + 6] << 24); + packet_reg += 4; + } + + HDMI_WRITE(VC4_HDMI_RAM_PACKET_CONFIG, + HDMI_READ(VC4_HDMI_RAM_PACKET_CONFIG) | BIT(packet_id)); + ret = wait_for((HDMI_READ(VC4_HDMI_RAM_PACKET_STATUS) & + BIT(packet_id)), 100); + if (ret) + DRM_ERROR("Failed to wait for infoframe to start: %d\n", ret); +} + +static void vc4_hdmi_set_avi_infoframe(struct drm_encoder *encoder) +{ + struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder); + struct drm_crtc *crtc = encoder->crtc; + const struct drm_display_mode *mode = &crtc->state->adjusted_mode; + union hdmi_infoframe frame; + int ret; + + ret = drm_hdmi_avi_infoframe_from_display_mode(&frame.avi, mode); + if (ret < 0) { + DRM_ERROR("couldn't fill AVI infoframe\n"); + return; + } + + if (vc4_encoder->rgb_range_selectable) { + if (vc4_encoder->limited_rgb_range) { + frame.avi.quantization_range = + HDMI_QUANTIZATION_RANGE_LIMITED; + } else { + frame.avi.quantization_range = + HDMI_QUANTIZATION_RANGE_FULL; + } + } + + vc4_hdmi_write_infoframe(encoder, &frame); +} + +static void vc4_hdmi_set_spd_infoframe(struct drm_encoder *encoder) +{ + union hdmi_infoframe frame; + int ret; + + ret = hdmi_spd_infoframe_init(&frame.spd, "Broadcom", "Videocore"); + if (ret < 0) { + DRM_ERROR("couldn't fill SPD infoframe\n"); + return; + } + + frame.spd.sdi = HDMI_SPD_SDI_PC; + + vc4_hdmi_write_infoframe(encoder, &frame); +} + +static void vc4_hdmi_set_infoframes(struct drm_encoder *encoder) +{ + vc4_hdmi_set_avi_infoframe(encoder); + vc4_hdmi_set_spd_infoframe(encoder); +} + static void vc4_hdmi_encoder_mode_set(struct drm_encoder *encoder, struct drm_display_mode *unadjusted_mode, struct drm_display_mode *mode) { + struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder); struct drm_device *dev = encoder->dev; struct vc4_dev *vc4 = to_vc4_dev(dev); bool debug_dump_regs = false; bool hsync_pos = mode->flags & DRM_MODE_FLAG_PHSYNC; bool vsync_pos = mode->flags & DRM_MODE_FLAG_PVSYNC; - u32 vactive = (mode->vdisplay >> - ((mode->flags & DRM_MODE_FLAG_INTERLACE) ? 1 : 0)); - u32 verta = (VC4_SET_FIELD(mode->vsync_end - mode->vsync_start, + bool interlaced = mode->flags & DRM_MODE_FLAG_INTERLACE; + u32 pixel_rep = (mode->flags & DRM_MODE_FLAG_DBLCLK) ? 2 : 1; + u32 verta = (VC4_SET_FIELD(mode->crtc_vsync_end - mode->crtc_vsync_start, VC4_HDMI_VERTA_VSP) | - VC4_SET_FIELD(mode->vsync_start - mode->vdisplay, + VC4_SET_FIELD(mode->crtc_vsync_start - mode->crtc_vdisplay, VC4_HDMI_VERTA_VFP) | - VC4_SET_FIELD(vactive, VC4_HDMI_VERTA_VAL)); + VC4_SET_FIELD(mode->crtc_vdisplay, VC4_HDMI_VERTA_VAL)); u32 vertb = (VC4_SET_FIELD(0, VC4_HDMI_VERTB_VSPO) | - VC4_SET_FIELD(mode->vtotal - mode->vsync_end, + VC4_SET_FIELD(mode->crtc_vtotal - mode->crtc_vsync_end, VC4_HDMI_VERTB_VBP)); + u32 vertb_even = (VC4_SET_FIELD(0, VC4_HDMI_VERTB_VSPO) | + VC4_SET_FIELD(mode->crtc_vtotal - + mode->crtc_vsync_end - + interlaced, + VC4_HDMI_VERTB_VBP)); + u32 csc_ctl; if (debug_dump_regs) { DRM_INFO("HDMI regs before:\n"); @@ -321,7 +425,8 @@ static void vc4_hdmi_encoder_mode_set(struct drm_encoder *encoder, HD_WRITE(VC4_HD_VID_CTL, 0); - clk_set_rate(vc4->hdmi->pixel_clock, mode->clock * 1000); + clk_set_rate(vc4->hdmi->pixel_clock, mode->clock * 1000 * + ((mode->flags & DRM_MODE_FLAG_DBLCLK) ? 2 : 1)); HDMI_WRITE(VC4_HDMI_SCHEDULER_CONTROL, HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) | @@ -331,29 +436,62 @@ static void vc4_hdmi_encoder_mode_set(struct drm_encoder *encoder, HDMI_WRITE(VC4_HDMI_HORZA, (vsync_pos ? VC4_HDMI_HORZA_VPOS : 0) | (hsync_pos ? VC4_HDMI_HORZA_HPOS : 0) | - VC4_SET_FIELD(mode->hdisplay, VC4_HDMI_HORZA_HAP)); + VC4_SET_FIELD(mode->hdisplay * pixel_rep, + VC4_HDMI_HORZA_HAP)); HDMI_WRITE(VC4_HDMI_HORZB, - VC4_SET_FIELD(mode->htotal - mode->hsync_end, + VC4_SET_FIELD((mode->htotal - + mode->hsync_end) * pixel_rep, VC4_HDMI_HORZB_HBP) | - VC4_SET_FIELD(mode->hsync_end - mode->hsync_start, + VC4_SET_FIELD((mode->hsync_end - + mode->hsync_start) * pixel_rep, VC4_HDMI_HORZB_HSP) | - VC4_SET_FIELD(mode->hsync_start - mode->hdisplay, + VC4_SET_FIELD((mode->hsync_start - + mode->hdisplay) * pixel_rep, VC4_HDMI_HORZB_HFP)); HDMI_WRITE(VC4_HDMI_VERTA0, verta); HDMI_WRITE(VC4_HDMI_VERTA1, verta); - HDMI_WRITE(VC4_HDMI_VERTB0, vertb); + HDMI_WRITE(VC4_HDMI_VERTB0, vertb_even); HDMI_WRITE(VC4_HDMI_VERTB1, vertb); HD_WRITE(VC4_HD_VID_CTL, (vsync_pos ? 0 : VC4_HD_VID_CTL_VSYNC_LOW) | (hsync_pos ? 0 : VC4_HD_VID_CTL_HSYNC_LOW)); + csc_ctl = VC4_SET_FIELD(VC4_HD_CSC_CTL_ORDER_BGR, + VC4_HD_CSC_CTL_ORDER); + + if (vc4_encoder->hdmi_monitor && drm_match_cea_mode(mode) > 1) { + /* CEA VICs other than #1 requre limited range RGB + * output unless overridden by an AVI infoframe. + * Apply a colorspace conversion to squash 0-255 down + * to 16-235. The matrix here is: + * + * [ 0 0 0.8594 16] + * [ 0 0.8594 0 16] + * [ 0.8594 0 0 16] + * [ 0 0 0 1] + */ + csc_ctl |= VC4_HD_CSC_CTL_ENABLE; + csc_ctl |= VC4_HD_CSC_CTL_RGB2YCC; + csc_ctl |= VC4_SET_FIELD(VC4_HD_CSC_CTL_MODE_CUSTOM, + VC4_HD_CSC_CTL_MODE); + + HD_WRITE(VC4_HD_CSC_12_11, (0x000 << 16) | 0x000); + HD_WRITE(VC4_HD_CSC_14_13, (0x100 << 16) | 0x6e0); + HD_WRITE(VC4_HD_CSC_22_21, (0x6e0 << 16) | 0x000); + HD_WRITE(VC4_HD_CSC_24_23, (0x100 << 16) | 0x000); + HD_WRITE(VC4_HD_CSC_32_31, (0x000 << 16) | 0x6e0); + HD_WRITE(VC4_HD_CSC_34_33, (0x100 << 16) | 0x000); + vc4_encoder->limited_rgb_range = true; + } else { + vc4_encoder->limited_rgb_range = false; + } + /* The RGB order applies even when CSC is disabled. */ - HD_WRITE(VC4_HD_CSC_CTL, VC4_SET_FIELD(VC4_HD_CSC_CTL_ORDER_BGR, - VC4_HD_CSC_CTL_ORDER)); + HD_WRITE(VC4_HD_CSC_CTL, csc_ctl); HDMI_WRITE(VC4_HDMI_FIFO_CTL, VC4_HDMI_FIFO_CTL_MASTER_SLAVE_N); @@ -368,6 +506,8 @@ static void vc4_hdmi_encoder_disable(struct drm_encoder *encoder) struct drm_device *dev = encoder->dev; struct vc4_dev *vc4 = to_vc4_dev(dev); + HDMI_WRITE(VC4_HDMI_RAM_PACKET_CONFIG, 0); + HDMI_WRITE(VC4_HDMI_TX_PHY_RESET_CTL, 0xf << 16); HD_WRITE(VC4_HD_VID_CTL, HD_READ(VC4_HD_VID_CTL) & ~VC4_HD_VID_CTL_ENABLE); @@ -394,7 +534,7 @@ static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder) VC4_HDMI_SCHEDULER_CONTROL_MODE_HDMI); ret = wait_for(HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) & - VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE, 1); + VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE, 1000); WARN_ONCE(ret, "Timeout waiting for " "VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE\n"); } else { @@ -406,7 +546,7 @@ static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder) ~VC4_HDMI_SCHEDULER_CONTROL_MODE_HDMI); ret = wait_for(!(HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) & - VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE), 1); + VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE), 1000); WARN_ONCE(ret, "Timeout waiting for " "!VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE\n"); } @@ -420,9 +560,10 @@ static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder) HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) | VC4_HDMI_SCHEDULER_CONTROL_VERT_ALWAYS_KEEPOUT); - /* XXX: Set HDMI_RAM_PACKET_CONFIG (1 << 16) and set - * up the infoframe. - */ + HDMI_WRITE(VC4_HDMI_RAM_PACKET_CONFIG, + VC4_HDMI_RAM_PACKET_ENABLE); + + vc4_hdmi_set_infoframes(encoder); drift = HDMI_READ(VC4_HDMI_FIFO_CTL); drift &= VC4_HDMI_FIFO_VALID_WRITE_MASK; diff --git a/drivers/gpu/drm/vc4/vc4_regs.h b/drivers/gpu/drm/vc4/vc4_regs.h index 160942a9180e..1aa44c2db556 100644 --- a/drivers/gpu/drm/vc4/vc4_regs.h +++ b/drivers/gpu/drm/vc4/vc4_regs.h @@ -175,6 +175,8 @@ # define PV_CONTROL_CLR_AT_START BIT(14) # define PV_CONTROL_TRIGGER_UNDERFLOW BIT(13) # define PV_CONTROL_WAIT_HSTART BIT(12) +# define PV_CONTROL_PIXEL_REP_MASK VC4_MASK(5, 4) +# define PV_CONTROL_PIXEL_REP_SHIFT 4 # define PV_CONTROL_CLK_SELECT_DSI_VEC 0 # define PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI 1 # define PV_CONTROL_CLK_SELECT_MASK VC4_MASK(3, 2) @@ -183,6 +185,9 @@ # define PV_CONTROL_EN BIT(0) #define PV_V_CONTROL 0x04 +# define PV_VCONTROL_ODD_DELAY_MASK VC4_MASK(22, 6) +# define PV_VCONTROL_ODD_DELAY_SHIFT 6 +# define PV_VCONTROL_ODD_FIRST BIT(5) # define PV_VCONTROL_INTERLACE BIT(4) # define PV_VCONTROL_CONTINUOUS BIT(1) # define PV_VCONTROL_VIDEN BIT(0) @@ -438,6 +443,8 @@ #define VC4_HDMI_RAM_PACKET_CONFIG 0x0a0 # define VC4_HDMI_RAM_PACKET_ENABLE BIT(16) +#define VC4_HDMI_RAM_PACKET_STATUS 0x0a4 + #define VC4_HDMI_HORZA 0x0c4 # define VC4_HDMI_HORZA_VPOS BIT(14) # define VC4_HDMI_HORZA_HPOS BIT(13) @@ -499,6 +506,9 @@ #define VC4_HDMI_TX_PHY_RESET_CTL 0x2c0 +#define VC4_HDMI_GCP_0 0x400 +#define VC4_HDMI_PACKET_STRIDE 0x24 + #define VC4_HD_M_CTL 0x00c # define VC4_HD_M_REGISTER_FILE_STANDBY (3 << 6) # define VC4_HD_M_RAM_STANDBY (3 << 4) @@ -528,10 +538,17 @@ # define VC4_HD_CSC_CTL_MODE_SHIFT 2 # define VC4_HD_CSC_CTL_MODE_RGB_TO_SD_YPRPB 0 # define VC4_HD_CSC_CTL_MODE_RGB_TO_HD_YPRPB 1 -# define VC4_HD_CSC_CTL_MODE_CUSTOM 2 +# define VC4_HD_CSC_CTL_MODE_CUSTOM 3 # define VC4_HD_CSC_CTL_RGB2YCC BIT(1) # define VC4_HD_CSC_CTL_ENABLE BIT(0) +#define VC4_HD_CSC_12_11 0x044 +#define VC4_HD_CSC_14_13 0x048 +#define VC4_HD_CSC_22_21 0x04c +#define VC4_HD_CSC_24_23 0x050 +#define VC4_HD_CSC_32_31 0x054 +#define VC4_HD_CSC_34_33 0x058 + #define VC4_HD_FRAME_COUNT 0x068 /* HVS display list information. */ diff --git a/drivers/gpu/drm/vc4/vc4_render_cl.c b/drivers/gpu/drm/vc4/vc4_render_cl.c index 0f12418725e5..08886a309757 100644 --- a/drivers/gpu/drm/vc4/vc4_render_cl.c +++ b/drivers/gpu/drm/vc4/vc4_render_cl.c @@ -45,6 +45,8 @@ struct vc4_rcl_setup { struct drm_gem_cma_object *rcl; u32 next_offset; + + u32 next_write_bo_index; }; static inline void rcl_u8(struct vc4_rcl_setup *setup, u8 val) @@ -407,6 +409,8 @@ static int vc4_rcl_msaa_surface_setup(struct vc4_exec_info *exec, if (!*obj) return -EINVAL; + exec->rcl_write_bo[exec->rcl_write_bo_count++] = *obj; + if (surf->offset & 0xf) { DRM_ERROR("MSAA write must be 16b aligned.\n"); return -EINVAL; @@ -417,7 +421,8 @@ static int vc4_rcl_msaa_surface_setup(struct vc4_exec_info *exec, static int vc4_rcl_surface_setup(struct vc4_exec_info *exec, struct drm_gem_cma_object **obj, - struct drm_vc4_submit_rcl_surface *surf) + struct drm_vc4_submit_rcl_surface *surf, + bool is_write) { uint8_t tiling = VC4_GET_FIELD(surf->bits, VC4_LOADSTORE_TILE_BUFFER_TILING); @@ -440,6 +445,9 @@ static int vc4_rcl_surface_setup(struct vc4_exec_info *exec, if (!*obj) return -EINVAL; + if (is_write) + exec->rcl_write_bo[exec->rcl_write_bo_count++] = *obj; + if (surf->flags & VC4_SUBMIT_RCL_SURFACE_READ_IS_FULL_RES) { if (surf == &exec->args->zs_write) { DRM_ERROR("general zs write may not be a full-res.\n"); @@ -542,6 +550,8 @@ vc4_rcl_render_config_surface_setup(struct vc4_exec_info *exec, if (!*obj) return -EINVAL; + exec->rcl_write_bo[exec->rcl_write_bo_count++] = *obj; + if (tiling > VC4_TILING_FORMAT_LT) { DRM_ERROR("Bad tiling format\n"); return -EINVAL; @@ -599,15 +609,18 @@ int vc4_get_rcl(struct drm_device *dev, struct vc4_exec_info *exec) if (ret) return ret; - ret = vc4_rcl_surface_setup(exec, &setup.color_read, &args->color_read); + ret = vc4_rcl_surface_setup(exec, &setup.color_read, &args->color_read, + false); if (ret) return ret; - ret = vc4_rcl_surface_setup(exec, &setup.zs_read, &args->zs_read); + ret = vc4_rcl_surface_setup(exec, &setup.zs_read, &args->zs_read, + false); if (ret) return ret; - ret = vc4_rcl_surface_setup(exec, &setup.zs_write, &args->zs_write); + ret = vc4_rcl_surface_setup(exec, &setup.zs_write, &args->zs_write, + true); if (ret) return ret; diff --git a/drivers/gpu/drm/vc4/vc4_validate.c b/drivers/gpu/drm/vc4/vc4_validate.c index 9ce1d0adf882..26503e307438 100644 --- a/drivers/gpu/drm/vc4/vc4_validate.c +++ b/drivers/gpu/drm/vc4/vc4_validate.c @@ -267,6 +267,9 @@ validate_indexed_prim_list(VALIDATE_ARGS) if (!ib) return -EINVAL; + exec->bin_dep_seqno = max(exec->bin_dep_seqno, + to_vc4_bo(&ib->base)->write_seqno); + if (offset > ib->base.size || (ib->base.size - offset) / index_size < length) { DRM_ERROR("IB access overflow (%d + %d*%d > %zd)\n", @@ -555,8 +558,7 @@ static bool reloc_tex(struct vc4_exec_info *exec, void *uniform_data_u, struct vc4_texture_sample_info *sample, - uint32_t texture_handle_index) - + uint32_t texture_handle_index, bool is_cs) { struct drm_gem_cma_object *tex; uint32_t p0 = *(uint32_t *)(uniform_data_u + sample->p_offset[0]); @@ -714,6 +716,11 @@ reloc_tex(struct vc4_exec_info *exec, *validated_p0 = tex->paddr + p0; + if (is_cs) { + exec->bin_dep_seqno = max(exec->bin_dep_seqno, + to_vc4_bo(&tex->base)->write_seqno); + } + return true; fail: DRM_INFO("Texture p0 at %d: 0x%08x\n", sample->p_offset[0], p0); @@ -835,7 +842,8 @@ validate_gl_shader_rec(struct drm_device *dev, if (!reloc_tex(exec, uniform_data_u, &validated_shader->texture_samples[tex], - texture_handles_u[tex])) { + texture_handles_u[tex], + i == 2)) { return -EINVAL; } } @@ -867,6 +875,9 @@ validate_gl_shader_rec(struct drm_device *dev, uint32_t stride = *(uint8_t *)(pkt_u + o + 5); uint32_t max_index; + exec->bin_dep_seqno = max(exec->bin_dep_seqno, + to_vc4_bo(&vbo->base)->write_seqno); + if (state->addr & 0x8) stride |= (*(uint32_t *)(pkt_u + 100 + i * 4)) & ~0xff; diff --git a/drivers/gpu/drm/via/via_dmablit.c b/drivers/gpu/drm/via/via_dmablit.c index 7e2a12c4fed2..1a3ad769f8c8 100644 --- a/drivers/gpu/drm/via/via_dmablit.c +++ b/drivers/gpu/drm/via/via_dmablit.c @@ -241,8 +241,8 @@ via_lock_all_dma_pages(drm_via_sg_info_t *vsg, drm_via_dmablit_t *xfer) down_read(¤t->mm->mmap_sem); ret = get_user_pages((unsigned long)xfer->mem_addr, vsg->num_pages, - (vsg->direction == DMA_FROM_DEVICE), - 0, vsg->pages, NULL); + (vsg->direction == DMA_FROM_DEVICE) ? FOLL_WRITE : 0, + vsg->pages, NULL); up_read(¤t->mm->mmap_sem); if (ret != vsg->num_pages) { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index e8ae3dc476d1..18061a4bc2f2 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -241,15 +241,15 @@ static int vmwgfx_pm_notifier(struct notifier_block *nb, unsigned long val, void *ptr); MODULE_PARM_DESC(enable_fbdev, "Enable vmwgfx fbdev"); -module_param_named(enable_fbdev, enable_fbdev, int, 0600); +module_param_named(enable_fbdev, enable_fbdev, int, S_IRUSR | S_IWUSR); MODULE_PARM_DESC(force_dma_api, "Force using the DMA API for TTM pages"); -module_param_named(force_dma_api, vmw_force_iommu, int, 0600); +module_param_named(force_dma_api, vmw_force_iommu, int, S_IRUSR | S_IWUSR); MODULE_PARM_DESC(restrict_iommu, "Try to limit IOMMU usage for TTM pages"); -module_param_named(restrict_iommu, vmw_restrict_iommu, int, 0600); +module_param_named(restrict_iommu, vmw_restrict_iommu, int, S_IRUSR | S_IWUSR); MODULE_PARM_DESC(force_coherent, "Force coherent TTM pages"); -module_param_named(force_coherent, vmw_force_coherent, int, 0600); +module_param_named(force_coherent, vmw_force_coherent, int, S_IRUSR | S_IWUSR); MODULE_PARM_DESC(restrict_dma_mask, "Restrict DMA mask to 44 bits with IOMMU"); -module_param_named(restrict_dma_mask, vmw_restrict_dma_mask, int, 0600); +module_param_named(restrict_dma_mask, vmw_restrict_dma_mask, int, S_IRUSR | S_IWUSR); MODULE_PARM_DESC(assume_16bpp, "Assume 16-bpp when filtering modes"); module_param_named(assume_16bpp, vmw_assume_16bpp, int, 0600); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 070d750af16d..1e59a486bba8 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -43,7 +43,7 @@ #define VMWGFX_DRIVER_DATE "20160210" #define VMWGFX_DRIVER_MAJOR 2 -#define VMWGFX_DRIVER_MINOR 10 +#define VMWGFX_DRIVER_MINOR 11 #define VMWGFX_DRIVER_PATCHLEVEL 0 #define VMWGFX_FILE_PAGE_OFFSET 0x00100000 #define VMWGFX_FIFO_STATIC_SIZE (1024*1024) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index dc5beff2b4aa..c7b53d987f06 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -35,17 +35,37 @@ #define VMW_RES_HT_ORDER 12 /** + * enum vmw_resource_relocation_type - Relocation type for resources + * + * @vmw_res_rel_normal: Traditional relocation. The resource id in the + * command stream is replaced with the actual id after validation. + * @vmw_res_rel_nop: NOP relocation. The command is unconditionally replaced + * with a NOP. + * @vmw_res_rel_cond_nop: Conditional NOP relocation. If the resource id + * after validation is -1, the command is replaced with a NOP. Otherwise no + * action. + */ +enum vmw_resource_relocation_type { + vmw_res_rel_normal, + vmw_res_rel_nop, + vmw_res_rel_cond_nop, + vmw_res_rel_max +}; + +/** * struct vmw_resource_relocation - Relocation info for resources * * @head: List head for the software context's relocation list. * @res: Non-ref-counted pointer to the resource. - * @offset: Offset of 4 byte entries into the command buffer where the + * @offset: Offset of single byte entries into the command buffer where the * id that needs fixup is located. + * @rel_type: Type of relocation. */ struct vmw_resource_relocation { struct list_head head; const struct vmw_resource *res; - unsigned long offset; + u32 offset:29; + enum vmw_resource_relocation_type rel_type:3; }; /** @@ -109,7 +129,18 @@ static int vmw_bo_to_validate_list(struct vmw_sw_context *sw_context, struct vmw_dma_buffer *vbo, bool validate_as_mob, uint32_t *p_val_node); - +/** + * vmw_ptr_diff - Compute the offset from a to b in bytes + * + * @a: A starting pointer. + * @b: A pointer offset in the same address space. + * + * Returns: The offset in bytes between the two pointers. + */ +static size_t vmw_ptr_diff(void *a, void *b) +{ + return (unsigned long) b - (unsigned long) a; +} /** * vmw_resources_unreserve - unreserve resources previously reserved for @@ -409,11 +440,14 @@ static int vmw_resource_context_res_add(struct vmw_private *dev_priv, * @list: Pointer to head of relocation list. * @res: The resource. * @offset: Offset into the command buffer currently being parsed where the - * id that needs fixup is located. Granularity is 4 bytes. + * id that needs fixup is located. Granularity is one byte. + * @rel_type: Relocation type. */ static int vmw_resource_relocation_add(struct list_head *list, const struct vmw_resource *res, - unsigned long offset) + unsigned long offset, + enum vmw_resource_relocation_type + rel_type) { struct vmw_resource_relocation *rel; @@ -425,6 +459,7 @@ static int vmw_resource_relocation_add(struct list_head *list, rel->res = res; rel->offset = offset; + rel->rel_type = rel_type; list_add_tail(&rel->head, list); return 0; @@ -459,11 +494,24 @@ static void vmw_resource_relocations_apply(uint32_t *cb, { struct vmw_resource_relocation *rel; + /* Validate the struct vmw_resource_relocation member size */ + BUILD_BUG_ON(SVGA_CB_MAX_SIZE >= (1 << 29)); + BUILD_BUG_ON(vmw_res_rel_max >= (1 << 3)); + list_for_each_entry(rel, list, head) { - if (likely(rel->res != NULL)) - cb[rel->offset] = rel->res->id; - else - cb[rel->offset] = SVGA_3D_CMD_NOP; + u32 *addr = (u32 *)((unsigned long) cb + rel->offset); + switch (rel->rel_type) { + case vmw_res_rel_normal: + *addr = rel->res->id; + break; + case vmw_res_rel_nop: + *addr = SVGA_3D_CMD_NOP; + break; + default: + if (rel->res->id == -1) + *addr = SVGA_3D_CMD_NOP; + break; + } } } @@ -655,7 +703,9 @@ static int vmw_cmd_res_reloc_add(struct vmw_private *dev_priv, *p_val = NULL; ret = vmw_resource_relocation_add(&sw_context->res_relocations, res, - id_loc - sw_context->buf_start); + vmw_ptr_diff(sw_context->buf_start, + id_loc), + vmw_res_rel_normal); if (unlikely(ret != 0)) return ret; @@ -721,7 +771,8 @@ vmw_cmd_res_check(struct vmw_private *dev_priv, return vmw_resource_relocation_add (&sw_context->res_relocations, res, - id_loc - sw_context->buf_start); + vmw_ptr_diff(sw_context->buf_start, id_loc), + vmw_res_rel_normal); } ret = vmw_user_resource_lookup_handle(dev_priv, @@ -2143,10 +2194,10 @@ static int vmw_cmd_shader_define(struct vmw_private *dev_priv, return ret; return vmw_resource_relocation_add(&sw_context->res_relocations, - NULL, &cmd->header.id - - sw_context->buf_start); - - return 0; + NULL, + vmw_ptr_diff(sw_context->buf_start, + &cmd->header.id), + vmw_res_rel_nop); } /** @@ -2188,10 +2239,10 @@ static int vmw_cmd_shader_destroy(struct vmw_private *dev_priv, return ret; return vmw_resource_relocation_add(&sw_context->res_relocations, - NULL, &cmd->header.id - - sw_context->buf_start); - - return 0; + NULL, + vmw_ptr_diff(sw_context->buf_start, + &cmd->header.id), + vmw_res_rel_nop); } /** @@ -2848,8 +2899,7 @@ static int vmw_cmd_dx_cid_check(struct vmw_private *dev_priv, * @header: Pointer to the command header in the command stream. * * Check that the view exists, and if it was not created using this - * command batch, make sure it's validated (present in the device) so that - * the remove command will not confuse the device. + * command batch, conditionally make this command a NOP. */ static int vmw_cmd_dx_view_remove(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, @@ -2877,10 +2927,16 @@ static int vmw_cmd_dx_view_remove(struct vmw_private *dev_priv, return ret; /* - * Add view to the validate list iff it was not created using this - * command batch. + * If the view wasn't created during this command batch, it might + * have been removed due to a context swapout, so add a + * relocation to conditionally make this command a NOP to avoid + * device errors. */ - return vmw_view_res_val_add(sw_context, view); + return vmw_resource_relocation_add(&sw_context->res_relocations, + view, + vmw_ptr_diff(sw_context->buf_start, + &cmd->header.id), + vmw_res_rel_cond_nop); } /** @@ -3029,6 +3085,35 @@ static int vmw_cmd_dx_genmips(struct vmw_private *dev_priv, cmd->body.shaderResourceViewId); } +/** + * vmw_cmd_dx_transfer_from_buffer - + * Validate an SVGA_3D_CMD_DX_TRANSFER_FROM_BUFFER command + * + * @dev_priv: Pointer to a device private struct. + * @sw_context: The software context being used for this batch. + * @header: Pointer to the command header in the command stream. + */ +static int vmw_cmd_dx_transfer_from_buffer(struct vmw_private *dev_priv, + struct vmw_sw_context *sw_context, + SVGA3dCmdHeader *header) +{ + struct { + SVGA3dCmdHeader header; + SVGA3dCmdDXTransferFromBuffer body; + } *cmd = container_of(header, typeof(*cmd), header); + int ret; + + ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + user_surface_converter, + &cmd->body.srcSid, NULL); + if (ret != 0) + return ret; + + return vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface, + user_surface_converter, + &cmd->body.destSid, NULL); +} + static int vmw_cmd_check_not_3d(struct vmw_private *dev_priv, struct vmw_sw_context *sw_context, void *buf, uint32_t *size) @@ -3379,6 +3464,9 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = { &vmw_cmd_buffer_copy_check, true, false, true), VMW_CMD_DEF(SVGA_3D_CMD_DX_PRED_COPY_REGION, &vmw_cmd_pred_copy_check, true, false, true), + VMW_CMD_DEF(SVGA_3D_CMD_DX_TRANSFER_FROM_BUFFER, + &vmw_cmd_dx_transfer_from_buffer, + true, false, true), }; static int vmw_cmd_check(struct vmw_private *dev_priv, @@ -3848,14 +3936,14 @@ static void *vmw_execbuf_cmdbuf(struct vmw_private *dev_priv, int ret; *header = NULL; - if (!dev_priv->cman || kernel_commands) - return kernel_commands; - if (command_size > SVGA_CB_MAX_SIZE) { DRM_ERROR("Command buffer is too large.\n"); return ERR_PTR(-EINVAL); } + if (!dev_priv->cman || kernel_commands) + return kernel_commands; + /* If possible, add a little space for fencing. */ cmdbuf_size = command_size + 512; cmdbuf_size = min_t(size_t, cmdbuf_size, SVGA_CB_MAX_SIZE); @@ -4232,9 +4320,6 @@ void __vmw_execbuf_release_pinned_bo(struct vmw_private *dev_priv, ttm_bo_unref(&query_val.bo); ttm_bo_unref(&pinned_val.bo); vmw_dmabuf_unreference(&dev_priv->pinned_bo); - DRM_INFO("Dummy query bo pin count: %d\n", - dev_priv->dummy_query_bo->pin_count); - out_unlock: return; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c index c2a721a8cef9..b445ce9b9757 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c @@ -324,7 +324,7 @@ static void vmw_hw_surface_destroy(struct vmw_resource *res) if (res->id != -1) { cmd = vmw_fifo_reserve(dev_priv, vmw_surface_destroy_size()); - if (unlikely(cmd == NULL)) { + if (unlikely(!cmd)) { DRM_ERROR("Failed reserving FIFO space for surface " "destruction.\n"); return; @@ -397,7 +397,7 @@ static int vmw_legacy_srf_create(struct vmw_resource *res) submit_size = vmw_surface_define_size(srf); cmd = vmw_fifo_reserve(dev_priv, submit_size); - if (unlikely(cmd == NULL)) { + if (unlikely(!cmd)) { DRM_ERROR("Failed reserving FIFO space for surface " "creation.\n"); ret = -ENOMEM; @@ -446,11 +446,10 @@ static int vmw_legacy_srf_dma(struct vmw_resource *res, uint8_t *cmd; struct vmw_private *dev_priv = res->dev_priv; - BUG_ON(val_buf->bo == NULL); - + BUG_ON(!val_buf->bo); submit_size = vmw_surface_dma_size(srf); cmd = vmw_fifo_reserve(dev_priv, submit_size); - if (unlikely(cmd == NULL)) { + if (unlikely(!cmd)) { DRM_ERROR("Failed reserving FIFO space for surface " "DMA.\n"); return -ENOMEM; @@ -538,7 +537,7 @@ static int vmw_legacy_srf_destroy(struct vmw_resource *res) submit_size = vmw_surface_destroy_size(); cmd = vmw_fifo_reserve(dev_priv, submit_size); - if (unlikely(cmd == NULL)) { + if (unlikely(!cmd)) { DRM_ERROR("Failed reserving FIFO space for surface " "eviction.\n"); return -ENOMEM; @@ -578,7 +577,7 @@ static int vmw_surface_init(struct vmw_private *dev_priv, int ret; struct vmw_resource *res = &srf->res; - BUG_ON(res_free == NULL); + BUG_ON(!res_free); if (!dev_priv->has_mob) vmw_fifo_resource_inc(dev_priv); ret = vmw_resource_init(dev_priv, res, true, res_free, @@ -700,7 +699,6 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, struct drm_vmw_surface_create_req *req = &arg->req; struct drm_vmw_surface_arg *rep = &arg->rep; struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; - struct drm_vmw_size __user *user_sizes; int ret; int i, j; uint32_t cur_bo_offset; @@ -748,7 +746,7 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, } user_srf = kzalloc(sizeof(*user_srf), GFP_KERNEL); - if (unlikely(user_srf == NULL)) { + if (unlikely(!user_srf)) { ret = -ENOMEM; goto out_no_user_srf; } @@ -763,29 +761,21 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, memcpy(srf->mip_levels, req->mip_levels, sizeof(srf->mip_levels)); srf->num_sizes = num_sizes; user_srf->size = size; - - srf->sizes = kmalloc(srf->num_sizes * sizeof(*srf->sizes), GFP_KERNEL); - if (unlikely(srf->sizes == NULL)) { - ret = -ENOMEM; + srf->sizes = memdup_user((struct drm_vmw_size __user *)(unsigned long) + req->size_addr, + sizeof(*srf->sizes) * srf->num_sizes); + if (IS_ERR(srf->sizes)) { + ret = PTR_ERR(srf->sizes); goto out_no_sizes; } - srf->offsets = kmalloc(srf->num_sizes * sizeof(*srf->offsets), - GFP_KERNEL); - if (unlikely(srf->offsets == NULL)) { + srf->offsets = kmalloc_array(srf->num_sizes, + sizeof(*srf->offsets), + GFP_KERNEL); + if (unlikely(!srf->offsets)) { ret = -ENOMEM; goto out_no_offsets; } - user_sizes = (struct drm_vmw_size __user *)(unsigned long) - req->size_addr; - - ret = copy_from_user(srf->sizes, user_sizes, - srf->num_sizes * sizeof(*srf->sizes)); - if (unlikely(ret != 0)) { - ret = -EFAULT; - goto out_no_copy; - } - srf->base_size = *srf->sizes; srf->autogen_filter = SVGA3D_TEX_FILTER_NONE; srf->multisample_count = 0; @@ -923,7 +913,7 @@ vmw_surface_handle_reference(struct vmw_private *dev_priv, ret = -EINVAL; base = ttm_base_object_lookup_for_ref(dev_priv->tdev, handle); - if (unlikely(base == NULL)) { + if (unlikely(!base)) { DRM_ERROR("Could not find surface to reference.\n"); goto out_no_lookup; } @@ -1069,7 +1059,7 @@ static int vmw_gb_surface_create(struct vmw_resource *res) cmd = vmw_fifo_reserve(dev_priv, submit_len); cmd2 = (typeof(cmd2))cmd; - if (unlikely(cmd == NULL)) { + if (unlikely(!cmd)) { DRM_ERROR("Failed reserving FIFO space for surface " "creation.\n"); ret = -ENOMEM; @@ -1135,7 +1125,7 @@ static int vmw_gb_surface_bind(struct vmw_resource *res, submit_size = sizeof(*cmd1) + (res->backup_dirty ? sizeof(*cmd2) : 0); cmd1 = vmw_fifo_reserve(dev_priv, submit_size); - if (unlikely(cmd1 == NULL)) { + if (unlikely(!cmd1)) { DRM_ERROR("Failed reserving FIFO space for surface " "binding.\n"); return -ENOMEM; @@ -1185,7 +1175,7 @@ static int vmw_gb_surface_unbind(struct vmw_resource *res, submit_size = sizeof(*cmd3) + (readback ? sizeof(*cmd1) : sizeof(*cmd2)); cmd = vmw_fifo_reserve(dev_priv, submit_size); - if (unlikely(cmd == NULL)) { + if (unlikely(!cmd)) { DRM_ERROR("Failed reserving FIFO space for surface " "unbinding.\n"); return -ENOMEM; @@ -1244,7 +1234,7 @@ static int vmw_gb_surface_destroy(struct vmw_resource *res) vmw_binding_res_list_scrub(&res->binding_head); cmd = vmw_fifo_reserve(dev_priv, sizeof(*cmd)); - if (unlikely(cmd == NULL)) { + if (unlikely(!cmd)) { DRM_ERROR("Failed reserving FIFO space for surface " "destruction.\n"); mutex_unlock(&dev_priv->binding_mutex); @@ -1410,7 +1400,7 @@ int vmw_gb_surface_reference_ioctl(struct drm_device *dev, void *data, user_srf = container_of(base, struct vmw_user_surface, prime.base); srf = &user_srf->srf; - if (srf->res.backup == NULL) { + if (!srf->res.backup) { DRM_ERROR("Shared GB surface is missing a backup buffer.\n"); goto out_bad_resource; } @@ -1524,7 +1514,7 @@ int vmw_surface_gb_priv_define(struct drm_device *dev, } user_srf = kzalloc(sizeof(*user_srf), GFP_KERNEL); - if (unlikely(user_srf == NULL)) { + if (unlikely(!user_srf)) { ret = -ENOMEM; goto out_no_user_srf; } |