diff options
| author | Dave Airlie <airlied@redhat.com> | 2017-07-20 04:31:43 +0300 |
|---|---|---|
| committer | Dave Airlie <airlied@redhat.com> | 2017-07-20 04:31:43 +0300 |
| commit | 2d62c799f8ffac4f7ffba6a4e7f148827dfc24c7 (patch) | |
| tree | dbd05cdff844362a80d31e21a825b38b16967fb2 /drivers | |
| parent | 5771a8c08880cdca3bfb4a3fc6d309d6bba20877 (diff) | |
| parent | 58947144af34a08eee3388c1f039f199e80c0c6f (diff) | |
| download | linux-2d62c799f8ffac4f7ffba6a4e7f148827dfc24c7.tar.xz | |
Merge tag 'drm-intel-next-2017-07-17' of git://anongit.freedesktop.org/git/drm-intel into drm-next
2nd round of 4.14 features:
- prep for deferred fbdev setup
- refactor fixed 16.16 computations and skl+ wm code (Mahesh Kumar)
- more cnl paches (Rodrigo, Imre et al)
- tighten context cleanup and handling (Chris Wilson)
- fix interlaced handling on skl+ (Mahesh Kumar)
- small bits as usual
* tag 'drm-intel-next-2017-07-17' of git://anongit.freedesktop.org/git/drm-intel: (84 commits)
drm/i915: Update DRIVER_DATE to 20170717
drm/i915: Protect against deferred fbdev setup
drm/i915/fbdev: Always forward hotplug events
drm/i915/skl+: unify cpp value in WM calculation
drm/i915/skl+: WM calculation don't require height
drm/i915: Addition wrapper for fixed16.16 operation
drm/i915: cleanup fixed-point wrappers naming
drm/i915: Always perform internal fixed16 division in 64 bits
drm/i915: take-out common clamping code of fixed16 wrappers
drm/i915/cnl: Add missing type case.
drm/i915/cnl: Add max allowed Cannonlake DC.
drm/i915: Make DP-MST connector info work
drm/i915/cnl: Get DDI clock based on PLLs.
drm/i915/cnl: Inherit RPS stuff from previous platforms.
drm/i915/cnl: Gen10 render context size.
drm/i915/cnl: Don't trust VBT's alternate pin for port D for now.
drm/i915: Fix the kernel panic when using aliasing ppgtt
drm/i915/cnl: Cannonlake color init.
drm/i915/cnl: Add force wake for gen10+.
x86/gpu: CNL uses the same GMS values as SKL
...
Diffstat (limited to 'drivers')
48 files changed, 1519 insertions, 637 deletions
diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig index a5cd5dacf055..e9e64e8e9765 100644 --- a/drivers/gpu/drm/i915/Kconfig +++ b/drivers/gpu/drm/i915/Kconfig @@ -21,6 +21,7 @@ config DRM_I915 select ACPI_BUTTON if ACPI select SYNC_FILE select IOSF_MBI + select CRC32 help Choose this option if you have a system that has "Intel Graphics Media Accelerator" or "HD Graphics" integrated graphics, diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index 4f7057d62d88..0e2e36ad6196 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -616,7 +616,7 @@ err: void intel_vgpu_clean_gvt_context(struct intel_vgpu *vgpu) { - i915_gem_context_put_unlocked(vgpu->shadow_ctx); + i915_gem_context_put(vgpu->shadow_ctx); } int intel_vgpu_init_gvt_context(struct intel_vgpu *vgpu) diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 00d8967c8512..2ef75c1a6119 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -1159,7 +1159,7 @@ static int i915_frequency_info(struct seq_file *m, void *unused) intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL); reqf = I915_READ(GEN6_RPNSWREQ); - if (IS_GEN9(dev_priv)) + if (INTEL_GEN(dev_priv) >= 9) reqf >>= 23; else { reqf &= ~GEN6_TURBO_DISABLE; @@ -1181,7 +1181,7 @@ static int i915_frequency_info(struct seq_file *m, void *unused) rpdownei = I915_READ(GEN6_RP_CUR_DOWN_EI) & GEN6_CURIAVG_MASK; rpcurdown = I915_READ(GEN6_RP_CUR_DOWN) & GEN6_CURBSYTAVG_MASK; rpprevdown = I915_READ(GEN6_RP_PREV_DOWN) & GEN6_CURBSYTAVG_MASK; - if (IS_GEN9(dev_priv)) + if (INTEL_GEN(dev_priv) >= 9) cagf = (rpstat & GEN9_CAGF_MASK) >> GEN9_CAGF_SHIFT; else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) cagf = (rpstat & HSW_CAGF_MASK) >> HSW_CAGF_SHIFT; @@ -1210,7 +1210,7 @@ static int i915_frequency_info(struct seq_file *m, void *unused) dev_priv->rps.pm_intrmsk_mbz); seq_printf(m, "GT_PERF_STATUS: 0x%08x\n", gt_perf_status); seq_printf(m, "Render p-state ratio: %d\n", - (gt_perf_status & (IS_GEN9(dev_priv) ? 0x1ff00 : 0xff00)) >> 8); + (gt_perf_status & (INTEL_GEN(dev_priv) >= 9 ? 0x1ff00 : 0xff00)) >> 8); seq_printf(m, "Render p-state VID: %d\n", gt_perf_status & 0xff); seq_printf(m, "Render p-state limit: %d\n", @@ -1241,18 +1241,21 @@ static int i915_frequency_info(struct seq_file *m, void *unused) max_freq = (IS_GEN9_LP(dev_priv) ? rp_state_cap >> 0 : rp_state_cap >> 16) & 0xff; - max_freq *= (IS_GEN9_BC(dev_priv) ? GEN9_FREQ_SCALER : 1); + max_freq *= (IS_GEN9_BC(dev_priv) || + IS_CANNONLAKE(dev_priv) ? GEN9_FREQ_SCALER : 1); seq_printf(m, "Lowest (RPN) frequency: %dMHz\n", intel_gpu_freq(dev_priv, max_freq)); max_freq = (rp_state_cap & 0xff00) >> 8; - max_freq *= (IS_GEN9_BC(dev_priv) ? GEN9_FREQ_SCALER : 1); + max_freq *= (IS_GEN9_BC(dev_priv) || + IS_CANNONLAKE(dev_priv) ? GEN9_FREQ_SCALER : 1); seq_printf(m, "Nominal (RP1) frequency: %dMHz\n", intel_gpu_freq(dev_priv, max_freq)); max_freq = (IS_GEN9_LP(dev_priv) ? rp_state_cap >> 16 : rp_state_cap >> 0) & 0xff; - max_freq *= (IS_GEN9_BC(dev_priv) ? GEN9_FREQ_SCALER : 1); + max_freq *= (IS_GEN9_BC(dev_priv) || + IS_CANNONLAKE(dev_priv) ? GEN9_FREQ_SCALER : 1); seq_printf(m, "Max non-overclocked (RP0) frequency: %dMHz\n", intel_gpu_freq(dev_priv, max_freq)); seq_printf(m, "Max overclocked frequency: %dMHz\n", @@ -1407,6 +1410,23 @@ static int i915_hangcheck_info(struct seq_file *m, void *unused) return 0; } +static int i915_reset_info(struct seq_file *m, void *unused) +{ + struct drm_i915_private *dev_priv = node_to_i915(m->private); + struct i915_gpu_error *error = &dev_priv->gpu_error; + struct intel_engine_cs *engine; + enum intel_engine_id id; + + seq_printf(m, "full gpu reset = %u\n", i915_reset_count(error)); + + for_each_engine(engine, dev_priv, id) { + seq_printf(m, "%s = %u\n", engine->name, + i915_reset_engine_count(error, engine)); + } + + return 0; +} + static int ironlake_drpc_info(struct seq_file *m) { struct drm_i915_private *dev_priv = node_to_i915(m->private); @@ -1838,7 +1858,7 @@ static int i915_ring_freq_table(struct seq_file *m, void *unused) if (ret) goto out; - if (IS_GEN9_BC(dev_priv)) { + if (IS_GEN9_BC(dev_priv) || IS_CANNONLAKE(dev_priv)) { /* Convert GT frequency to 50 HZ units */ min_gpu_freq = dev_priv->rps.min_freq_softlimit / GEN9_FREQ_SCALER; @@ -1858,7 +1878,8 @@ static int i915_ring_freq_table(struct seq_file *m, void *unused) &ia_freq); seq_printf(m, "%d\t\t%d\t\t\t\t%d\n", intel_gpu_freq(dev_priv, (gpu_freq * - (IS_GEN9_BC(dev_priv) ? + (IS_GEN9_BC(dev_priv) || + IS_CANNONLAKE(dev_priv) ? GEN9_FREQ_SCALER : 1))), ((ia_freq >> 0) & 0xff) * 100, ((ia_freq >> 8) & 0xff) * 100); @@ -1914,7 +1935,7 @@ static int i915_gem_framebuffer_info(struct seq_file *m, void *data) return ret; #ifdef CONFIG_DRM_FBDEV_EMULATION - if (dev_priv->fbdev) { + if (dev_priv->fbdev && dev_priv->fbdev->helper.fb) { fbdev_fb = to_intel_framebuffer(dev_priv->fbdev->helper.fb); seq_printf(m, "fbcon size: %d x %d, depth %d, %d bpp, modifier 0x%llx, refcount %d, obj ", @@ -1970,7 +1991,7 @@ static int i915_context_status(struct seq_file *m, void *unused) if (ret) return ret; - list_for_each_entry(ctx, &dev_priv->context_list, link) { + list_for_each_entry(ctx, &dev_priv->contexts.list, link) { seq_printf(m, "HW context %u ", ctx->hw_id); if (ctx->pid) { struct task_struct *task; @@ -2076,7 +2097,7 @@ static int i915_dump_lrc(struct seq_file *m, void *unused) if (ret) return ret; - list_for_each_entry(ctx, &dev_priv->context_list, link) + list_for_each_entry(ctx, &dev_priv->contexts.list, link) for_each_engine(engine, dev_priv, id) i915_dump_lrc_obj(m, ctx, engine); @@ -2310,6 +2331,8 @@ static int i915_rps_boost_info(struct seq_file *m, void *data) seq_printf(m, "GPU busy? %s [%d requests]\n", yesno(dev_priv->gt.awake), dev_priv->gt.active_requests); seq_printf(m, "CPU waiting? %d\n", count_irq_waiters(dev_priv)); + seq_printf(m, "Boosts outstanding? %d\n", + atomic_read(&dev_priv->rps.num_waiters)); seq_printf(m, "Frequency requested %d\n", intel_gpu_freq(dev_priv, dev_priv->rps.cur_freq)); seq_printf(m, " min hard:%d, soft:%d; max soft:%d, hard:%d\n", @@ -2323,22 +2346,20 @@ static int i915_rps_boost_info(struct seq_file *m, void *data) intel_gpu_freq(dev_priv, dev_priv->rps.boost_freq)); mutex_lock(&dev->filelist_mutex); - spin_lock(&dev_priv->rps.client_lock); list_for_each_entry_reverse(file, &dev->filelist, lhead) { struct drm_i915_file_private *file_priv = file->driver_priv; struct task_struct *task; rcu_read_lock(); task = pid_task(file->pid, PIDTYPE_PID); - seq_printf(m, "%s [%d]: %d boosts%s\n", + seq_printf(m, "%s [%d]: %d boosts\n", task ? task->comm : "<unknown>", task ? task->pid : -1, - file_priv->rps.boosts, - list_empty(&file_priv->rps.link) ? "" : ", active"); + atomic_read(&file_priv->rps.boosts)); rcu_read_unlock(); } - seq_printf(m, "Kernel (anonymous) boosts: %d\n", dev_priv->rps.boosts); - spin_unlock(&dev_priv->rps.client_lock); + seq_printf(m, "Kernel (anonymous) boosts: %d\n", + atomic_read(&dev_priv->rps.boosts)); mutex_unlock(&dev->filelist_mutex); if (INTEL_GEN(dev_priv) >= 6 && @@ -3289,6 +3310,7 @@ static int i915_display_info(struct seq_file *m, void *unused) static int i915_engine_info(struct seq_file *m, void *unused) { struct drm_i915_private *dev_priv = node_to_i915(m->private); + struct i915_gpu_error *error = &dev_priv->gpu_error; struct intel_engine_cs *engine; enum intel_engine_id id; @@ -3312,6 +3334,8 @@ static int i915_engine_info(struct seq_file *m, void *unused) engine->hangcheck.seqno, jiffies_to_msecs(jiffies - engine->hangcheck.action_timestamp), engine->timeline->inflight_seqnos); + seq_printf(m, "\tReset count: %d\n", + i915_reset_engine_count(error, engine)); rcu_read_lock(); @@ -3758,13 +3782,18 @@ static ssize_t i915_displayport_test_active_write(struct file *file, drm_connector_list_iter_begin(dev, &conn_iter); drm_for_each_connector_iter(connector, &conn_iter) { + struct intel_encoder *encoder; + if (connector->connector_type != DRM_MODE_CONNECTOR_DisplayPort) continue; - if (connector->status == connector_status_connected && - connector->encoder != NULL) { - intel_dp = enc_to_intel_dp(connector->encoder); + encoder = to_intel_encoder(connector->encoder); + if (encoder && encoder->type == INTEL_OUTPUT_DP_MST) + continue; + + if (encoder && connector->status == connector_status_connected) { + intel_dp = enc_to_intel_dp(&encoder->base); status = kstrtoint(input_buffer, 10, &val); if (status < 0) break; @@ -3796,13 +3825,18 @@ static int i915_displayport_test_active_show(struct seq_file *m, void *data) drm_connector_list_iter_begin(dev, &conn_iter); drm_for_each_connector_iter(connector, &conn_iter) { + struct intel_encoder *encoder; + if (connector->connector_type != DRM_MODE_CONNECTOR_DisplayPort) continue; - if (connector->status == connector_status_connected && - connector->encoder != NULL) { - intel_dp = enc_to_intel_dp(connector->encoder); + encoder = to_intel_encoder(connector->encoder); + if (encoder && encoder->type == INTEL_OUTPUT_DP_MST) + continue; + + if (encoder && connector->status == connector_status_connected) { + intel_dp = enc_to_intel_dp(&encoder->base); if (intel_dp->compliance.test_active) seq_puts(m, "1"); else @@ -3842,13 +3876,18 @@ static int i915_displayport_test_data_show(struct seq_file *m, void *data) drm_connector_list_iter_begin(dev, &conn_iter); drm_for_each_connector_iter(connector, &conn_iter) { + struct intel_encoder *encoder; + if (connector->connector_type != DRM_MODE_CONNECTOR_DisplayPort) continue; - if (connector->status == connector_status_connected && - connector->encoder != NULL) { - intel_dp = enc_to_intel_dp(connector->encoder); + encoder = to_intel_encoder(connector->encoder); + if (encoder && encoder->type == INTEL_OUTPUT_DP_MST) + continue; + + if (encoder && connector->status == connector_status_connected) { + intel_dp = enc_to_intel_dp(&encoder->base); if (intel_dp->compliance.test_type == DP_TEST_LINK_EDID_READ) seq_printf(m, "%lx", @@ -3895,13 +3934,18 @@ static int i915_displayport_test_type_show(struct seq_file *m, void *data) drm_connector_list_iter_begin(dev, &conn_iter); drm_for_each_connector_iter(connector, &conn_iter) { + struct intel_encoder *encoder; + if (connector->connector_type != DRM_MODE_CONNECTOR_DisplayPort) continue; - if (connector->status == connector_status_connected && - connector->encoder != NULL) { - intel_dp = enc_to_intel_dp(connector->encoder); + encoder = to_intel_encoder(connector->encoder); + if (encoder && encoder->type == INTEL_OUTPUT_DP_MST) + continue; + + if (encoder && connector->status == connector_status_connected) { + intel_dp = enc_to_intel_dp(&encoder->base); seq_printf(m, "%02lx", intel_dp->compliance.test_type); } else seq_puts(m, "0"); @@ -4824,6 +4868,7 @@ static const struct drm_info_list i915_debugfs_list[] = { {"i915_huc_load_status", i915_huc_load_status_info, 0}, {"i915_frequency_info", i915_frequency_info, 0}, {"i915_hangcheck_info", i915_hangcheck_info, 0}, + {"i915_reset_info", i915_reset_info, 0}, {"i915_drpc_info", i915_drpc_info, 0}, {"i915_emon_status", i915_emon_status, 0}, {"i915_ring_freq_table", i915_ring_freq_table, 0}, diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index fc307e03943c..6f750efe9c3d 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -132,9 +132,13 @@ static enum intel_pch intel_virt_detect_pch(struct drm_i915_private *dev_priv) DRM_DEBUG_KMS("Assuming Ibex Peak PCH\n"); } else if (IS_GEN6(dev_priv) || IS_IVYBRIDGE(dev_priv)) { ret = PCH_CPT; - DRM_DEBUG_KMS("Assuming CouarPoint PCH\n"); + DRM_DEBUG_KMS("Assuming CougarPoint PCH\n"); } else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) { ret = PCH_LPT; + if (IS_HSW_ULT(dev_priv) || IS_BDW_ULT(dev_priv)) + dev_priv->pch_id = INTEL_PCH_LPT_LP_DEVICE_ID_TYPE; + else + dev_priv->pch_id = INTEL_PCH_LPT_DEVICE_ID_TYPE; DRM_DEBUG_KMS("Assuming LynxPoint PCH\n"); } else if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv)) { ret = PCH_SPT; @@ -173,29 +177,25 @@ static void intel_detect_pch(struct drm_i915_private *dev_priv) while ((pch = pci_get_class(PCI_CLASS_BRIDGE_ISA << 8, pch))) { if (pch->vendor == PCI_VENDOR_ID_INTEL) { unsigned short id = pch->device & INTEL_PCH_DEVICE_ID_MASK; - unsigned short id_ext = pch->device & - INTEL_PCH_DEVICE_ID_MASK_EXT; + + dev_priv->pch_id = id; if (id == INTEL_PCH_IBX_DEVICE_ID_TYPE) { - dev_priv->pch_id = id; dev_priv->pch_type = PCH_IBX; DRM_DEBUG_KMS("Found Ibex Peak PCH\n"); WARN_ON(!IS_GEN5(dev_priv)); } else if (id == INTEL_PCH_CPT_DEVICE_ID_TYPE) { - dev_priv->pch_id = id; dev_priv->pch_type = PCH_CPT; DRM_DEBUG_KMS("Found CougarPoint PCH\n"); - WARN_ON(!(IS_GEN6(dev_priv) || - IS_IVYBRIDGE(dev_priv))); + WARN_ON(!IS_GEN6(dev_priv) && + !IS_IVYBRIDGE(dev_priv)); } else if (id == INTEL_PCH_PPT_DEVICE_ID_TYPE) { /* PantherPoint is CPT compatible */ - dev_priv->pch_id = id; dev_priv->pch_type = PCH_CPT; DRM_DEBUG_KMS("Found PantherPoint PCH\n"); - WARN_ON(!(IS_GEN6(dev_priv) || - IS_IVYBRIDGE(dev_priv))); + WARN_ON(!IS_GEN6(dev_priv) && + !IS_IVYBRIDGE(dev_priv)); } else if (id == INTEL_PCH_LPT_DEVICE_ID_TYPE) { - dev_priv->pch_id = id; dev_priv->pch_type = PCH_LPT; DRM_DEBUG_KMS("Found LynxPoint PCH\n"); WARN_ON(!IS_HASWELL(dev_priv) && @@ -203,51 +203,60 @@ static void intel_detect_pch(struct drm_i915_private *dev_priv) WARN_ON(IS_HSW_ULT(dev_priv) || IS_BDW_ULT(dev_priv)); } else if (id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) { - dev_priv->pch_id = id; dev_priv->pch_type = PCH_LPT; DRM_DEBUG_KMS("Found LynxPoint LP PCH\n"); WARN_ON(!IS_HASWELL(dev_priv) && !IS_BROADWELL(dev_priv)); WARN_ON(!IS_HSW_ULT(dev_priv) && !IS_BDW_ULT(dev_priv)); + } else if (id == INTEL_PCH_WPT_DEVICE_ID_TYPE) { + /* WildcatPoint is LPT compatible */ + dev_priv->pch_type = PCH_LPT; + DRM_DEBUG_KMS("Found WildcatPoint PCH\n"); + WARN_ON(!IS_HASWELL(dev_priv) && + !IS_BROADWELL(dev_priv)); + WARN_ON(IS_HSW_ULT(dev_priv) || + IS_BDW_ULT(dev_priv)); + } else if (id == INTEL_PCH_WPT_LP_DEVICE_ID_TYPE) { + /* WildcatPoint is LPT compatible */ + dev_priv->pch_type = PCH_LPT; + DRM_DEBUG_KMS("Found WildcatPoint LP PCH\n"); + WARN_ON(!IS_HASWELL(dev_priv) && + !IS_BROADWELL(dev_priv)); + WARN_ON(!IS_HSW_ULT(dev_priv) && + !IS_BDW_ULT(dev_priv)); } else if (id == INTEL_PCH_SPT_DEVICE_ID_TYPE) { - dev_priv->pch_id = id; dev_priv->pch_type = PCH_SPT; DRM_DEBUG_KMS("Found SunrisePoint PCH\n"); WARN_ON(!IS_SKYLAKE(dev_priv) && !IS_KABYLAKE(dev_priv)); - } else if (id_ext == INTEL_PCH_SPT_LP_DEVICE_ID_TYPE) { - dev_priv->pch_id = id_ext; + } else if (id == INTEL_PCH_SPT_LP_DEVICE_ID_TYPE) { dev_priv->pch_type = PCH_SPT; DRM_DEBUG_KMS("Found SunrisePoint LP PCH\n"); WARN_ON(!IS_SKYLAKE(dev_priv) && !IS_KABYLAKE(dev_priv)); } else if (id == INTEL_PCH_KBP_DEVICE_ID_TYPE) { - dev_priv->pch_id = id; dev_priv->pch_type = PCH_KBP; DRM_DEBUG_KMS("Found KabyPoint PCH\n"); WARN_ON(!IS_SKYLAKE(dev_priv) && !IS_KABYLAKE(dev_priv)); } else if (id == INTEL_PCH_CNP_DEVICE_ID_TYPE) { - dev_priv->pch_id = id; dev_priv->pch_type = PCH_CNP; DRM_DEBUG_KMS("Found CannonPoint PCH\n"); WARN_ON(!IS_CANNONLAKE(dev_priv) && !IS_COFFEELAKE(dev_priv)); - } else if (id_ext == INTEL_PCH_CNP_LP_DEVICE_ID_TYPE) { - dev_priv->pch_id = id_ext; + } else if (id == INTEL_PCH_CNP_LP_DEVICE_ID_TYPE) { dev_priv->pch_type = PCH_CNP; DRM_DEBUG_KMS("Found CannonPoint LP PCH\n"); WARN_ON(!IS_CANNONLAKE(dev_priv) && !IS_COFFEELAKE(dev_priv)); - } else if ((id == INTEL_PCH_P2X_DEVICE_ID_TYPE) || - (id == INTEL_PCH_P3X_DEVICE_ID_TYPE) || - ((id == INTEL_PCH_QEMU_DEVICE_ID_TYPE) && + } else if (id == INTEL_PCH_P2X_DEVICE_ID_TYPE || + id == INTEL_PCH_P3X_DEVICE_ID_TYPE || + (id == INTEL_PCH_QEMU_DEVICE_ID_TYPE && pch->subsystem_vendor == PCI_SUBVENDOR_ID_REDHAT_QUMRANET && pch->subsystem_device == PCI_SUBDEVICE_ID_QEMU)) { - dev_priv->pch_id = id; dev_priv->pch_type = intel_virt_detect_pch(dev_priv); } else @@ -331,6 +340,8 @@ static int i915_getparam(struct drm_device *dev, void *data, break; case I915_PARAM_HAS_GPU_RESET: value = i915.enable_hangcheck && intel_has_gpu_reset(dev_priv); + if (value && intel_has_reset_engine(dev_priv)) + value = 2; break; case I915_PARAM_HAS_RESOURCE_STREAMER: value = HAS_RESOURCE_STREAMER(dev_priv); @@ -585,16 +596,18 @@ static const struct vga_switcheroo_client_ops i915_switcheroo_ops = { static void i915_gem_fini(struct drm_i915_private *dev_priv) { + flush_workqueue(dev_priv->wq); + mutex_lock(&dev_priv->drm.struct_mutex); intel_uc_fini_hw(dev_priv); i915_gem_cleanup_engines(dev_priv); - i915_gem_context_fini(dev_priv); + i915_gem_contexts_fini(dev_priv); i915_gem_cleanup_userptr(dev_priv); mutex_unlock(&dev_priv->drm.struct_mutex); i915_gem_drain_freed_objects(dev_priv); - WARN_ON(!list_empty(&dev_priv->context_list)); + WARN_ON(!list_empty(&dev_priv->contexts.list)); } static int i915_load_modeset_init(struct drm_device *dev) @@ -1427,9 +1440,10 @@ static void i915_driver_release(struct drm_device *dev) static int i915_driver_open(struct drm_device *dev, struct drm_file *file) { + struct drm_i915_private *i915 = to_i915(dev); int ret; - ret = i915_gem_open(dev, file); + ret = i915_gem_open(i915, file); if (ret) return ret; @@ -1459,7 +1473,7 @@ static void i915_driver_postclose(struct drm_device *dev, struct drm_file *file) struct drm_i915_file_private *file_priv = file->driver_priv; mutex_lock(&dev->struct_mutex); - i915_gem_context_close(dev, file); + i915_gem_context_close(file); i915_gem_release(dev, file); mutex_unlock(&dev->struct_mutex); @@ -1911,9 +1925,72 @@ wakeup: error: i915_gem_set_wedged(dev_priv); + i915_gem_retire_requests(dev_priv); goto finish; } +/** + * i915_reset_engine - reset GPU engine to recover from a hang + * @engine: engine to reset + * + * Reset a specific GPU engine. Useful if a hang is detected. + * Returns zero on successful reset or otherwise an error code. + * + * Procedure is: + * - identifies the request that caused the hang and it is dropped + * - reset engine (which will force the engine to idle) + * - re-init/configure engine + */ +int i915_reset_engine(struct intel_engine_cs *engine) +{ + struct i915_gpu_error *error = &engine->i915->gpu_error; + struct drm_i915_gem_request *active_request; + int ret; + + GEM_BUG_ON(!test_bit(I915_RESET_ENGINE + engine->id, &error->flags)); + + DRM_DEBUG_DRIVER("resetting %s\n", engine->name); + + active_request = i915_gem_reset_prepare_engine(engine); + if (IS_ERR(active_request)) { + DRM_DEBUG_DRIVER("Previous reset failed, promote to full reset\n"); + ret = PTR_ERR(active_request); + goto out; + } + + /* + * The request that caused the hang is stuck on elsp, we know the + * active request and can drop it, adjust head to skip the offending + * request to resume executing remaining requests in the queue. + */ + i915_gem_reset_engine(engine, active_request); + + /* Finally, reset just this engine. */ + ret = intel_gpu_reset(engine->i915, intel_engine_flag(engine)); + + i915_gem_reset_finish_engine(engine); + + if (ret) { + /* If we fail here, we expect to fallback to a global reset */ + DRM_DEBUG_DRIVER("Failed to reset %s, ret=%d\n", + engine->name, ret); + goto out; + } + + /* + * The engine and its registers (and workarounds in case of render) + * have been reset to their default values. Follow the init_ring + * process to program RING_MODE, HWSP and re-enable submission. + */ + ret = engine->init_hw(engine); + if (ret) + goto out; + + error->reset_engine_count[engine->id]++; +out: + return ret; +} + static int i915_pm_suspend(struct device *kdev) { struct pci_dev *pdev = to_pci_dev(kdev); diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index e1f7c97a338a..559fdc7bb393 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -80,8 +80,8 @@ #define DRIVER_NAME "i915" #define DRIVER_DESC "Intel Graphics" -#define DRIVER_DATE "20170619" -#define DRIVER_TIMESTAMP 1497857498 +#define DRIVER_DATE "20170717" +#define DRIVER_TIMESTAMP 1500275179 /* Use I915_STATE_WARN(x) and I915_STATE_WARN_ON() (rather than WARN() and * WARN_ON()) for hw state sanity checks to check for unexpected conditions @@ -122,7 +122,7 @@ static inline bool is_fixed16_zero(uint_fixed_16_16_t val) return false; } -static inline uint_fixed_16_16_t u32_to_fixed_16_16(uint32_t val) +static inline uint_fixed_16_16_t u32_to_fixed16(uint32_t val) { uint_fixed_16_16_t fp; @@ -132,17 +132,17 @@ static inline uint_fixed_16_16_t u32_to_fixed_16_16(uint32_t val) return fp; } -static inline uint32_t fixed_16_16_to_u32_round_up(uint_fixed_16_16_t fp) +static inline uint32_t fixed16_to_u32_round_up(uint_fixed_16_16_t fp) { return DIV_ROUND_UP(fp.val, 1 << 16); } -static inline uint32_t fixed_16_16_to_u32(uint_fixed_16_16_t fp) +static inline uint32_t fixed16_to_u32(uint_fixed_16_16_t fp) { return fp.val >> 16; } -static inline uint_fixed_16_16_t min_fixed_16_16(uint_fixed_16_16_t min1, +static inline uint_fixed_16_16_t min_fixed16(uint_fixed_16_16_t min1, uint_fixed_16_16_t min2) { uint_fixed_16_16_t min; @@ -151,7 +151,7 @@ static inline uint_fixed_16_16_t min_fixed_16_16(uint_fixed_16_16_t min1, return min; } -static inline uint_fixed_16_16_t max_fixed_16_16(uint_fixed_16_16_t max1, +static inline uint_fixed_16_16_t max_fixed16(uint_fixed_16_16_t max1, uint_fixed_16_16_t max2) { uint_fixed_16_16_t max; @@ -160,6 +160,14 @@ static inline uint_fixed_16_16_t max_fixed_16_16(uint_fixed_16_16_t max1, return max; } +static inline uint_fixed_16_16_t clamp_u64_to_fixed16(uint64_t val) +{ + uint_fixed_16_16_t fp; + WARN_ON(val >> 32); + fp.val = clamp_t(uint32_t, val, 0, ~0); + return fp; +} + static inline uint32_t div_round_up_fixed16(uint_fixed_16_16_t val, uint_fixed_16_16_t d) { @@ -170,48 +178,30 @@ static inline uint32_t mul_round_up_u32_fixed16(uint32_t val, uint_fixed_16_16_t mul) { uint64_t intermediate_val; - uint32_t result; intermediate_val = (uint64_t) val * mul.val; intermediate_val = DIV_ROUND_UP_ULL(intermediate_val, 1 << 16); WARN_ON(intermediate_val >> 32); - result = clamp_t(uint32_t, intermediate_val, 0, ~0); - return result; + return clamp_t(uint32_t, intermediate_val, 0, ~0); } static inline uint_fixed_16_16_t mul_fixed16(uint_fixed_16_16_t val, uint_fixed_16_16_t mul) { uint64_t intermediate_val; - uint_fixed_16_16_t fp; intermediate_val = (uint64_t) val.val * mul.val; intermediate_val = intermediate_val >> 16; - WARN_ON(intermediate_val >> 32); - fp.val = clamp_t(uint32_t, intermediate_val, 0, ~0); - return fp; + return clamp_u64_to_fixed16(intermediate_val); } -static inline uint_fixed_16_16_t fixed_16_16_div(uint32_t val, uint32_t d) +static inline uint_fixed_16_16_t div_fixed16(uint32_t val, uint32_t d) { - uint_fixed_16_16_t fp, res; - - fp = u32_to_fixed_16_16(val); - res.val = DIV_ROUND_UP(fp.val, d); - return res; -} - -static inline uint_fixed_16_16_t fixed_16_16_div_u64(uint32_t val, uint32_t d) -{ - uint_fixed_16_16_t res; uint64_t interm_val; interm_val = (uint64_t)val << 16; interm_val = DIV_ROUND_UP_ULL(interm_val, d); - WARN_ON(interm_val >> 32); - res.val = (uint32_t) interm_val; - - return res; + return clamp_u64_to_fixed16(interm_val); } static inline uint32_t div_round_up_u32_fixed16(uint32_t val, @@ -225,16 +215,32 @@ static inline uint32_t div_round_up_u32_fixed16(uint32_t val, return clamp_t(uint32_t, interm_val, 0, ~0); } -static inline uint_fixed_16_16_t mul_u32_fixed_16_16(uint32_t val, +static inline uint_fixed_16_16_t mul_u32_fixed16(uint32_t val, uint_fixed_16_16_t mul) { uint64_t intermediate_val; - uint_fixed_16_16_t fp; intermediate_val = (uint64_t) val * mul.val; - WARN_ON(intermediate_val >> 32); - fp.val = (uint32_t) intermediate_val; - return fp; + return clamp_u64_to_fixed16(intermediate_val); +} + +static inline uint_fixed_16_16_t add_fixed16(uint_fixed_16_16_t add1, + uint_fixed_16_16_t add2) +{ + uint64_t interm_sum; + + interm_sum = (uint64_t) add1.val + add2.val; + return clamp_u64_to_fixed16(interm_sum); +} + +static inline uint_fixed_16_16_t add_fixed16_u32(uint_fixed_16_16_t add1, + uint32_t add2) +{ + uint64_t interm_sum; + uint_fixed_16_16_t interm_add2 = u32_to_fixed16(add2); + + interm_sum = (uint64_t) add1.val + interm_add2.val; + return clamp_u64_to_fixed16(interm_sum); } static inline const char *yesno(bool v) @@ -584,8 +590,7 @@ struct drm_i915_file_private { struct idr context_idr; struct intel_rps_client { - struct list_head link; - unsigned boosts; + atomic_t boosts; } rps; unsigned int bsd_engine; @@ -753,6 +758,7 @@ struct intel_csr { func(has_csr); \ func(has_ddi); \ func(has_dp_mst); \ + func(has_reset_engine); \ func(has_fbc); \ func(has_fpga_dbg); \ func(has_full_ppgtt); \ @@ -917,6 +923,7 @@ struct i915_gpu_state { enum intel_engine_hangcheck_action hangcheck_action; struct i915_address_space *vm; int num_requests; + u32 reset_count; /* position of active request inside the ring */ u32 rq_head, rq_post, rq_tail; @@ -1149,8 +1156,8 @@ struct i915_psr { enum intel_pch { PCH_NONE = 0, /* No PCH present */ PCH_IBX, /* Ibexpeak PCH */ - PCH_CPT, /* Cougarpoint PCH */ - PCH_LPT, /* Lynxpoint PCH */ + PCH_CPT, /* Cougarpoint/Pantherpoint PCH */ + PCH_LPT, /* Lynxpoint/Wildcatpoint PCH */ PCH_SPT, /* Sunrisepoint PCH */ PCH_KBP, /* Kabypoint PCH */ PCH_CNP, /* Cannonpoint PCH */ @@ -1166,6 +1173,7 @@ enum intel_sbi_destination { #define QUIRK_INVERT_BRIGHTNESS (1<<2) #define QUIRK_BACKLIGHT_PRESENT (1<<3) #define QUIRK_PIN_SWIZZLED_PAGES (1<<5) +#define QUIRK_INCREASE_T12_DELAY (1<<6) struct intel_fbdev; struct intel_fbc_work; @@ -1301,13 +1309,10 @@ struct intel_gen6_power_mgmt { int last_adj; enum { LOW_POWER, BETWEEN, HIGH_POWER } power; - spinlock_t client_lock; - struct list_head clients; - bool client_boost; - bool enabled; struct delayed_work autoenable_work; - unsigned boosts; + atomic_t num_waiters; + atomic_t boosts; /* manual wa residency calculations */ struct intel_rps_ei ei; @@ -1550,6 +1555,12 @@ struct i915_gpu_error { * inspect the bit and do the reset directly, otherwise the worker * waits for the struct_mutex. * + * #I915_RESET_ENGINE[num_engines] - Since the driver doesn't need to + * acquire the struct_mutex to reset an engine, we need an explicit + * flag to prevent two concurrent reset attempts in the same engine. + * As the number of engines continues to grow, allocate the flags from + * the most significant bits. + * * #I915_WEDGED - If reset fails and we can no longer use the GPU, * we set the #I915_WEDGED bit. Prior to command submission, e.g. * i915_gem_request_alloc(), this bit is checked and the sequence @@ -1559,6 +1570,10 @@ struct i915_gpu_error { #define I915_RESET_BACKOFF 0 #define I915_RESET_HANDOFF 1 #define I915_WEDGED (BITS_PER_LONG - 1) +#define I915_RESET_ENGINE (I915_WEDGED - I915_NUM_ENGINES) + + /** Number of times an engine has been reset */ + u32 reset_engine_count[I915_NUM_ENGINES]; /** * Waitqueue to signal when a hang is detected. Used to for waiters @@ -2236,13 +2251,6 @@ struct drm_i915_private { DECLARE_HASHTABLE(mm_structs, 7); struct mutex mm_lock; - /* The hw wants to have a stable context identifier for the lifetime - * of the context (for OA, PASID, faults, etc). This is limited - * in execlists to 21 bits. - */ - struct ida context_hw_ida; -#define MAX_CONTEXT_HW_ID (1<<21) /* exclusive */ - /* Kernel Modesetting */ struct intel_crtc *plane_to_crtc_mapping[I915_MAX_PIPES]; @@ -2321,7 +2329,18 @@ struct drm_i915_private { */ struct mutex av_mutex; - struct list_head context_list; + struct { + struct list_head list; + struct llist_head free_list; + struct work_struct free_work; + + /* The hw wants to have a stable context identifier for the + * lifetime of the context (for OA, PASID, faults, etc). + * This is limited in execlists to 21 bits. + */ + struct ida hw_ida; +#define MAX_CONTEXT_HW_ID (1<<21) /* exclusive */ + } contexts; u32 fdi_rx_config; @@ -2996,16 +3015,17 @@ intel_info(const struct drm_i915_private *dev_priv) #define HAS_POOLED_EU(dev_priv) ((dev_priv)->info.has_pooled_eu) -#define INTEL_PCH_DEVICE_ID_MASK 0xff00 -#define INTEL_PCH_DEVICE_ID_MASK_EXT 0xff80 +#define INTEL_PCH_DEVICE_ID_MASK 0xff80 #define INTEL_PCH_IBX_DEVICE_ID_TYPE 0x3b00 #define INTEL_PCH_CPT_DEVICE_ID_TYPE 0x1c00 #define INTEL_PCH_PPT_DEVICE_ID_TYPE 0x1e00 #define INTEL_PCH_LPT_DEVICE_ID_TYPE 0x8c00 #define INTEL_PCH_LPT_LP_DEVICE_ID_TYPE 0x9c00 +#define INTEL_PCH_WPT_DEVICE_ID_TYPE 0x8c80 +#define INTEL_PCH_WPT_LP_DEVICE_ID_TYPE 0x9c80 #define INTEL_PCH_SPT_DEVICE_ID_TYPE 0xA100 #define INTEL_PCH_SPT_LP_DEVICE_ID_TYPE 0x9D00 -#define INTEL_PCH_KBP_DEVICE_ID_TYPE 0xA200 +#define INTEL_PCH_KBP_DEVICE_ID_TYPE 0xA280 #define INTEL_PCH_CNP_DEVICE_ID_TYPE 0xA300 #define INTEL_PCH_CNP_LP_DEVICE_ID_TYPE 0x9D80 #define INTEL_PCH_P2X_DEVICE_ID_TYPE 0x7100 @@ -3020,9 +3040,11 @@ intel_info(const struct drm_i915_private *dev_priv) #define HAS_PCH_SPT(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_SPT) #define HAS_PCH_LPT(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_LPT) #define HAS_PCH_LPT_LP(dev_priv) \ - ((dev_priv)->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) + ((dev_priv)->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE || \ + (dev_priv)->pch_id == INTEL_PCH_WPT_LP_DEVICE_ID_TYPE) #define HAS_PCH_LPT_H(dev_priv) \ - ((dev_priv)->pch_id == INTEL_PCH_LPT_DEVICE_ID_TYPE) + ((dev_priv)->pch_id == INTEL_PCH_LPT_DEVICE_ID_TYPE || \ + (dev_priv)->pch_id == INTEL_PCH_WPT_DEVICE_ID_TYPE) #define HAS_PCH_CPT(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_CPT) #define HAS_PCH_IBX(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_IBX) #define HAS_PCH_NOP(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_NOP) @@ -3089,6 +3111,8 @@ extern void i915_driver_unload(struct drm_device *dev); extern int intel_gpu_reset(struct drm_i915_private *dev_priv, u32 engine_mask); extern bool intel_has_gpu_reset(struct drm_i915_private *dev_priv); extern void i915_reset(struct drm_i915_private *dev_priv); +extern int i915_reset_engine(struct intel_engine_cs *engine); +extern bool intel_has_reset_engine(struct drm_i915_private *dev_priv); extern int intel_guc_reset(struct drm_i915_private *dev_priv); extern void intel_engine_init_hangcheck(struct intel_engine_cs *engine); extern void intel_hangcheck_init(struct drm_i915_private *dev_priv); @@ -3461,11 +3485,22 @@ static inline u32 i915_reset_count(struct i915_gpu_error *error) return READ_ONCE(error->reset_count); } +static inline u32 i915_reset_engine_count(struct i915_gpu_error *error, + struct intel_engine_cs *engine) +{ + return READ_ONCE(error->reset_engine_count[engine->id]); +} + +struct drm_i915_gem_request * +i915_gem_reset_prepare_engine(struct intel_engine_cs *engine); int i915_gem_reset_prepare(struct drm_i915_private *dev_priv); void i915_gem_reset(struct drm_i915_private *dev_priv); +void i915_gem_reset_finish_engine(struct intel_engine_cs *engine); void i915_gem_reset_finish(struct drm_i915_private *dev_priv); void i915_gem_set_wedged(struct drm_i915_private *dev_priv); bool i915_gem_unset_wedged(struct drm_i915_private *dev_priv); +void i915_gem_reset_engine(struct intel_engine_cs *engine, + struct drm_i915_gem_request *request); void i915_gem_init_mmio(struct drm_i915_private *i915); int __must_check i915_gem_init(struct drm_i915_private *dev_priv); @@ -3499,7 +3534,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, void i915_gem_object_unpin_from_display_plane(struct i915_vma *vma); int i915_gem_object_attach_phys(struct drm_i915_gem_object *obj, int align); -int i915_gem_open(struct drm_device *dev, struct drm_file *file); +int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file); void i915_gem_release(struct drm_device *dev, struct drm_file *file); int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, @@ -3531,38 +3566,23 @@ void i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj, struct sg_table *pages); static inline struct i915_gem_context * -i915_gem_context_lookup(struct drm_i915_file_private *file_priv, u32 id) +__i915_gem_context_lookup_rcu(struct drm_i915_file_private *file_priv, u32 id) { - struct i915_gem_context *ctx; - - lockdep_assert_held(&file_priv->dev_priv->drm.struct_mutex); - - ctx = idr_find(&file_priv->context_idr, id); - if (!ctx) - return ERR_PTR(-ENOENT); - - return ctx; + return idr_find(&file_priv->context_idr, id); } static inline struct i915_gem_context * -i915_gem_context_get(struct i915_gem_context *ctx) -{ - kref_get(&ctx->ref); - return ctx; -} - -static inline void i915_gem_context_put(struct i915_gem_context *ctx) +i915_gem_context_lookup(struct drm_i915_file_private *file_priv, u32 id) { - lockdep_assert_held(&ctx->i915->drm.struct_mutex); - kref_put(&ctx->ref, i915_gem_context_free); -} + struct i915_gem_context *ctx; -static inline void i915_gem_context_put_unlocked(struct i915_gem_context *ctx) -{ - struct mutex *lock = &ctx->i915->drm.struct_mutex; + rcu_read_lock(); + ctx = __i915_gem_context_lookup_rcu(file_priv, id); + if (ctx && !kref_get_unless_zero(&ctx->ref)) + ctx = NULL; + rcu_read_unlock(); - if (kref_put_mutex(&ctx->ref, i915_gem_context_free, lock)) - mutex_unlock(lock); + return ctx; } static inline struct intel_timeline * diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 969bac8404f1..d6f9b4cb6e9b 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -388,7 +388,7 @@ i915_gem_object_wait_fence(struct dma_fence *fence, */ if (rps) { if (INTEL_GEN(rq->i915) >= 6) - gen6_rps_boost(rq->i915, rps, rq->emitted_jiffies); + gen6_rps_boost(rq, rps); else rps = NULL; } @@ -399,22 +399,6 @@ out: if (flags & I915_WAIT_LOCKED && i915_gem_request_completed(rq)) i915_gem_request_retire_upto(rq); - if (rps && i915_gem_request_global_seqno(rq) == intel_engine_last_submit(rq->engine)) { - /* The GPU is now idle and this client has stalled. - * Since no other client has submitted a request in the - * meantime, assume that this client is the only one - * supplying work to the GPU but is unable to keep that - * work supplied because it is waiting. Since the GPU is - * then never kept fully busy, RPS autoclocking will - * keep the clocks relatively low, causing further delays. - * Compensate by giving the synchronous client credit for - * a waitboost next time. - */ - spin_lock(&rq->i915->rps.client_lock); - list_del_init(&rps->link); - spin_unlock(&rq->i915->rps.client_lock); - } - return timeout; } @@ -2832,46 +2816,64 @@ static bool engine_stalled(struct intel_engine_cs *engine) return true; } +/* + * Ensure irq handler finishes, and not run again. + * Also return the active request so that we only search for it once. + */ +struct drm_i915_gem_request * +i915_gem_reset_prepare_engine(struct intel_engine_cs *engine) +{ + struct drm_i915_gem_request *request = NULL; + + /* Prevent the signaler thread from updating the request + * state (by calling dma_fence_signal) as we are processing + * the reset. The write from the GPU of the seqno is + * asynchronous and the signaler thread may see a different + * value to us and declare the request complete, even though + * the reset routine have picked that request as the active + * (incomplete) request. This conflict is not handled + * gracefully! + */ + kthread_park(engine->breadcrumbs.signaler); + + /* Prevent request submission to the hardware until we have + * completed the reset in i915_gem_reset_finish(). If a request + * is completed by one engine, it may then queue a request + * to a second via its engine->irq_tasklet *just* as we are + * calling engine->init_hw() and also writing the ELSP. + * Turning off the engine->irq_tasklet until the reset is over + * prevents the race. + */ + tasklet_kill(&engine->irq_tasklet); + tasklet_disable(&engine->irq_tasklet); + + if (engine->irq_seqno_barrier) + engine->irq_seqno_barrier(engine); + + if (engine_stalled(engine)) { + request = i915_gem_find_active_request(engine); + if (request && request->fence.error == -EIO) + request = ERR_PTR(-EIO); /* Previous reset failed! */ + } + + return request; +} + int i915_gem_reset_prepare(struct drm_i915_private *dev_priv) { struct intel_engine_cs *engine; + struct drm_i915_gem_request *request; enum intel_engine_id id; int err = 0; - /* Ensure irq handler finishes, and not run again. */ for_each_engine(engine, dev_priv, id) { - struct drm_i915_gem_request *request; - - /* Prevent the signaler thread from updating the request - * state (by calling dma_fence_signal) as we are processing - * the reset. The write from the GPU of the seqno is - * asynchronous and the signaler thread may see a different - * value to us and declare the request complete, even though - * the reset routine have picked that request as the active - * (incomplete) request. This conflict is not handled - * gracefully! - */ - kthread_park(engine->breadcrumbs.signaler); - - /* Prevent request submission to the hardware until we have - * completed the reset in i915_gem_reset_finish(). If a request - * is completed by one engine, it may then queue a request - * to a second via its engine->irq_tasklet *just* as we are - * calling engine->init_hw() and also writing the ELSP. - * Turning off the engine->irq_tasklet until the reset is over - * prevents the race. - */ - tasklet_kill(&engine->irq_tasklet); - tasklet_disable(&engine->irq_tasklet); - - if (engine->irq_seqno_barrier) - engine->irq_seqno_barrier(engine); - - if (engine_stalled(engine)) { - request = i915_gem_find_active_request(engine); - if (request && request->fence.error == -EIO) - err = -EIO; /* Previous reset failed! */ + request = i915_gem_reset_prepare_engine(engine); + if (IS_ERR(request)) { + err = PTR_ERR(request); + continue; } + + engine->hangcheck.active_request = request; } i915_gem_revoke_fences(dev_priv); @@ -2925,7 +2927,7 @@ static void engine_skip_context(struct drm_i915_gem_request *request) static bool i915_gem_reset_request(struct drm_i915_gem_request *request) { /* Read once and return the resolution */ - const bool guilty = engine_stalled(request->engine); + const bool guilty = !i915_gem_request_completed(request); /* The guilty request will get skipped on a hung engine. * @@ -2959,11 +2961,9 @@ static bool i915_gem_reset_request(struct drm_i915_gem_request *request) return guilty; } -static void i915_gem_reset_engine(struct intel_engine_cs *engine) +void i915_gem_reset_engine(struct intel_engine_cs *engine, + struct drm_i915_gem_request *request) { - struct drm_i915_gem_request *request; - - request = i915_gem_find_active_request(engine); if (request && i915_gem_reset_request(request)) { DRM_DEBUG_DRIVER("resetting %s to restart from tail of request 0x%x\n", engine->name, request->global_seqno); @@ -2989,7 +2989,7 @@ void i915_gem_reset(struct drm_i915_private *dev_priv) for_each_engine(engine, dev_priv, id) { struct i915_gem_context *ctx; - i915_gem_reset_engine(engine); + i915_gem_reset_engine(engine, engine->hangcheck.active_request); ctx = fetch_and_zero(&engine->last_retired_context); if (ctx) engine->context_unpin(engine, ctx); @@ -3005,6 +3005,12 @@ void i915_gem_reset(struct drm_i915_private *dev_priv) } } +void i915_gem_reset_finish_engine(struct intel_engine_cs *engine) +{ + tasklet_enable(&engine->irq_tasklet); + kthread_unpark(engine->breadcrumbs.signaler); +} + void i915_gem_reset_finish(struct drm_i915_private *dev_priv) { struct intel_engine_cs *engine; @@ -3013,8 +3019,8 @@ void i915_gem_reset_finish(struct drm_i915_private *dev_priv) lockdep_assert_held(&dev_priv->drm.struct_mutex); for_each_engine(engine, dev_priv, id) { - tasklet_enable(&engine->irq_tasklet); - kthread_unpark(engine->breadcrumbs.signaler); + engine->hangcheck.active_request = NULL; + i915_gem_reset_finish_engine(engine); } } @@ -3041,7 +3047,8 @@ static void engine_set_wedged(struct intel_engine_cs *engine) /* Mark all executing requests as skipped */ spin_lock_irqsave(&engine->timeline->lock, flags); list_for_each_entry(request, &engine->timeline->requests, link) - dma_fence_set_error(&request->fence, -EIO); + if (!i915_gem_request_completed(request)) + dma_fence_set_error(&request->fence, -EIO); spin_unlock_irqrestore(&engine->timeline->lock, flags); /* Mark all pending requests as complete so that any concurrent @@ -3071,6 +3078,13 @@ static void engine_set_wedged(struct intel_engine_cs *engine) engine->execlist_first = NULL; spin_unlock_irqrestore(&engine->timeline->lock, flags); + + /* The port is checked prior to scheduling a tasklet, but + * just in case we have suspended the tasklet to do the + * wedging make sure that when it wakes, it decides there + * is no work to do by clearing the irq_posted bit. + */ + clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted); } } @@ -3080,6 +3094,7 @@ static int __i915_gem_set_wedged_BKL(void *data) struct intel_engine_cs *engine; enum intel_engine_id id; + set_bit(I915_WEDGED, &i915->gpu_error.flags); for_each_engine(engine, i915, id) engine_set_wedged(engine); @@ -3088,20 +3103,7 @@ static int __i915_gem_set_wedged_BKL(void *data) void i915_gem_set_wedged(struct drm_i915_private *dev_priv) { - lockdep_assert_held(&dev_priv->drm.struct_mutex); - set_bit(I915_WEDGED, &dev_priv->gpu_error.flags); - - /* Retire completed requests first so the list of inflight/incomplete - * requests is accurate and we don't try and mark successful requests - * as in error during __i915_gem_set_wedged_BKL(). - */ - i915_gem_retire_requests(dev_priv); - stop_machine(__i915_gem_set_wedged_BKL, dev_priv, NULL); - - i915_gem_context_lost(dev_priv); - - mod_delayed_work(dev_priv->wq, &dev_priv->gt.idle_work, 0); } bool i915_gem_unset_wedged(struct drm_i915_private *i915) @@ -3156,6 +3158,7 @@ bool i915_gem_unset_wedged(struct drm_i915_private *i915) * context and do not require stop_machine(). */ intel_engines_reset_default_submission(i915); + i915_gem_contexts_lost(i915); smp_mb__before_atomic(); /* complete takeover before enabling execbuf */ clear_bit(I915_WEDGED, &i915->gpu_error.flags); @@ -4565,7 +4568,7 @@ int i915_gem_suspend(struct drm_i915_private *dev_priv) goto err_unlock; assert_kernel_context_is_current(dev_priv); - i915_gem_context_lost(dev_priv); + i915_gem_contexts_lost(dev_priv); mutex_unlock(&dev->struct_mutex); intel_guc_suspend(dev_priv); @@ -4579,8 +4582,6 @@ int i915_gem_suspend(struct drm_i915_private *dev_priv) while (flush_delayed_work(&dev_priv->gt.idle_work)) ; - i915_gem_drain_freed_objects(dev_priv); - /* Assert that we sucessfully flushed all the work and * reset the GPU back to its idle, low power state. */ @@ -4812,7 +4813,7 @@ int i915_gem_init(struct drm_i915_private *dev_priv) if (ret) goto out_unlock; - ret = i915_gem_context_init(dev_priv); + ret = i915_gem_contexts_init(dev_priv); if (ret) goto out_unlock; @@ -4922,7 +4923,6 @@ i915_gem_load_init(struct drm_i915_private *dev_priv) if (err) goto err_priorities; - INIT_LIST_HEAD(&dev_priv->context_list); INIT_WORK(&dev_priv->mm.free_work, __i915_gem_free_work); init_llist_head(&dev_priv->mm.free_list); INIT_LIST_HEAD(&dev_priv->mm.unbound_list); @@ -5038,15 +5038,9 @@ void i915_gem_release(struct drm_device *dev, struct drm_file *file) list_for_each_entry(request, &file_priv->mm.request_list, client_link) request->file_priv = NULL; spin_unlock(&file_priv->mm.lock); - - if (!list_empty(&file_priv->rps.link)) { - spin_lock(&to_i915(dev)->rps.client_lock); - list_del(&file_priv->rps.link); - spin_unlock(&to_i915(dev)->rps.client_lock); - } } -int i915_gem_open(struct drm_device *dev, struct drm_file *file) +int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file) { struct drm_i915_file_private *file_priv; int ret; @@ -5058,16 +5052,15 @@ int i915_gem_open(struct drm_device *dev, struct drm_file *file) return -ENOMEM; file->driver_priv = file_priv; - file_priv->dev_priv = to_i915(dev); + file_priv->dev_priv = i915; file_priv->file = file; - INIT_LIST_HEAD(&file_priv->rps.link); spin_lock_init(&file_priv->mm.lock); INIT_LIST_HEAD(&file_priv->mm.request_list); file_priv->bsd_engine = -1; - ret = i915_gem_context_open(dev, file); + ret = i915_gem_context_open(i915, file); if (ret) kfree(file_priv); diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index 39ed58a21fc1..1a87d04e7937 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -158,13 +158,11 @@ static void vma_lut_free(struct i915_gem_context *ctx) kvfree(lut->ht); } -void i915_gem_context_free(struct kref *ctx_ref) +static void i915_gem_context_free(struct i915_gem_context *ctx) { - struct i915_gem_context *ctx = container_of(ctx_ref, typeof(*ctx), ref); int i; lockdep_assert_held(&ctx->i915->drm.struct_mutex); - trace_i915_context_free(ctx); GEM_BUG_ON(!i915_gem_context_is_closed(ctx)); vma_lut_free(ctx); @@ -188,8 +186,54 @@ void i915_gem_context_free(struct kref *ctx_ref) list_del(&ctx->link); - ida_simple_remove(&ctx->i915->context_hw_ida, ctx->hw_id); - kfree(ctx); + ida_simple_remove(&ctx->i915->contexts.hw_ida, ctx->hw_id); + kfree_rcu(ctx, rcu); +} + +static void contexts_free(struct drm_i915_private *i915) +{ + struct llist_node *freed = llist_del_all(&i915->contexts.free_list); + struct i915_gem_context *ctx, *cn; + + lockdep_assert_held(&i915->drm.struct_mutex); + + llist_for_each_entry_safe(ctx, cn, freed, free_link) + i915_gem_context_free(ctx); +} + +static void contexts_free_first(struct drm_i915_private *i915) +{ + struct i915_gem_context *ctx; + struct llist_node *freed; + + lockdep_assert_held(&i915->drm.struct_mutex); + + freed = llist_del_first(&i915->contexts.free_list); + if (!freed) + return; + + ctx = container_of(freed, typeof(*ctx), free_link); + i915_gem_context_free(ctx); +} + +static void contexts_free_worker(struct work_struct *work) +{ + struct drm_i915_private *i915 = + container_of(work, typeof(*i915), contexts.free_work); + + mutex_lock(&i915->drm.struct_mutex); + contexts_free(i915); + mutex_unlock(&i915->drm.struct_mutex); +} + +void i915_gem_context_release(struct kref *ref) +{ + struct i915_gem_context *ctx = container_of(ref, typeof(*ctx), ref); + struct drm_i915_private *i915 = ctx->i915; + + trace_i915_context_free(ctx); + if (llist_add(&ctx->free_link, &i915->contexts.free_list)) + queue_work(i915->wq, &i915->contexts.free_work); } static void context_close(struct i915_gem_context *ctx) @@ -205,7 +249,7 @@ static int assign_hw_id(struct drm_i915_private *dev_priv, unsigned *out) { int ret; - ret = ida_simple_get(&dev_priv->context_hw_ida, + ret = ida_simple_get(&dev_priv->contexts.hw_ida, 0, MAX_CONTEXT_HW_ID, GFP_KERNEL); if (ret < 0) { /* Contexts are only released when no longer active. @@ -213,7 +257,7 @@ static int assign_hw_id(struct drm_i915_private *dev_priv, unsigned *out) * stale contexts and try again. */ i915_gem_retire_requests(dev_priv); - ret = ida_simple_get(&dev_priv->context_hw_ida, + ret = ida_simple_get(&dev_priv->contexts.hw_ida, 0, MAX_CONTEXT_HW_ID, GFP_KERNEL); if (ret < 0) return ret; @@ -265,7 +309,7 @@ __create_hw_context(struct drm_i915_private *dev_priv, } kref_init(&ctx->ref); - list_add_tail(&ctx->link, &dev_priv->context_list); + list_add_tail(&ctx->link, &dev_priv->contexts.list); ctx->i915 = dev_priv; ctx->priority = I915_PRIORITY_NORMAL; @@ -354,6 +398,9 @@ i915_gem_create_context(struct drm_i915_private *dev_priv, lockdep_assert_held(&dev_priv->drm.struct_mutex); + /* Reap the most stale context */ + contexts_free_first(dev_priv); + ctx = __create_hw_context(dev_priv, file_priv); if (IS_ERR(ctx)) return ctx; @@ -418,7 +465,7 @@ out: return ctx; } -int i915_gem_context_init(struct drm_i915_private *dev_priv) +int i915_gem_contexts_init(struct drm_i915_private *dev_priv) { struct i915_gem_context *ctx; @@ -427,6 +474,10 @@ int i915_gem_context_init(struct drm_i915_private *dev_priv) if (WARN_ON(dev_priv->kernel_context)) return 0; + INIT_LIST_HEAD(&dev_priv->contexts.list); + INIT_WORK(&dev_priv->contexts.free_work, contexts_free_worker); + init_llist_head(&dev_priv->contexts.free_list); + if (intel_vgpu_active(dev_priv) && HAS_LOGICAL_RING_CONTEXTS(dev_priv)) { if (!i915.enable_execlists) { @@ -437,7 +488,7 @@ int i915_gem_context_init(struct drm_i915_private *dev_priv) /* Using the simple ida interface, the max is limited by sizeof(int) */ BUILD_BUG_ON(MAX_CONTEXT_HW_ID > INT_MAX); - ida_init(&dev_priv->context_hw_ida); + ida_init(&dev_priv->contexts.hw_ida); ctx = i915_gem_create_context(dev_priv, NULL); if (IS_ERR(ctx)) { @@ -463,7 +514,7 @@ int i915_gem_context_init(struct drm_i915_private *dev_priv) return 0; } -void i915_gem_context_lost(struct drm_i915_private *dev_priv) +void i915_gem_contexts_lost(struct drm_i915_private *dev_priv) { struct intel_engine_cs *engine; enum intel_engine_id id; @@ -484,7 +535,7 @@ void i915_gem_context_lost(struct drm_i915_private *dev_priv) if (!i915.enable_execlists) { struct i915_gem_context *ctx; - list_for_each_entry(ctx, &dev_priv->context_list, link) { + list_for_each_entry(ctx, &dev_priv->contexts.list, link) { if (!i915_gem_context_is_default(ctx)) continue; @@ -503,18 +554,20 @@ void i915_gem_context_lost(struct drm_i915_private *dev_priv) } } -void i915_gem_context_fini(struct drm_i915_private *dev_priv) +void i915_gem_contexts_fini(struct drm_i915_private *i915) { - struct i915_gem_context *dctx = dev_priv->kernel_context; - - lockdep_assert_held(&dev_priv->drm.struct_mutex); + struct i915_gem_context *ctx; - GEM_BUG_ON(!i915_gem_context_is_kernel(dctx)); + lockdep_assert_held(&i915->drm.struct_mutex); - context_close(dctx); - dev_priv->kernel_context = NULL; + /* Keep the context so that we can free it immediately ourselves */ + ctx = i915_gem_context_get(fetch_and_zero(&i915->kernel_context)); + GEM_BUG_ON(!i915_gem_context_is_kernel(ctx)); + context_close(ctx); + i915_gem_context_free(ctx); - ida_destroy(&dev_priv->context_hw_ida); + /* Must free all deferred contexts (via flush_workqueue) first */ + ida_destroy(&i915->contexts.hw_ida); } static int context_idr_cleanup(int id, void *p, void *data) @@ -525,32 +578,32 @@ static int context_idr_cleanup(int id, void *p, void *data) return 0; } -int i915_gem_context_open(struct drm_device *dev, struct drm_file *file) +int i915_gem_context_open(struct drm_i915_private *i915, + struct drm_file *file) { struct drm_i915_file_private *file_priv = file->driver_priv; struct i915_gem_context *ctx; idr_init(&file_priv->context_idr); - mutex_lock(&dev->struct_mutex); - ctx = i915_gem_create_context(to_i915(dev), file_priv); - mutex_unlock(&dev->struct_mutex); - - GEM_BUG_ON(i915_gem_context_is_kernel(ctx)); - + mutex_lock(&i915->drm.struct_mutex); + ctx = i915_gem_create_context(i915, file_priv); + mutex_unlock(&i915->drm.struct_mutex); if (IS_ERR(ctx)) { idr_destroy(&file_priv->context_idr); return PTR_ERR(ctx); } + GEM_BUG_ON(i915_gem_context_is_kernel(ctx)); + return 0; } -void i915_gem_context_close(struct drm_device *dev, struct drm_file *file) +void i915_gem_context_close(struct drm_file *file) { struct drm_i915_file_private *file_priv = file->driver_priv; - lockdep_assert_held(&dev->struct_mutex); + lockdep_assert_held(&file_priv->dev_priv->drm.struct_mutex); idr_for_each(&file_priv->context_idr, context_idr_cleanup, NULL); idr_destroy(&file_priv->context_idr); @@ -981,20 +1034,19 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data, if (args->ctx_id == DEFAULT_CONTEXT_HANDLE) return -ENOENT; - ret = i915_mutex_lock_interruptible(dev); - if (ret) - return ret; - ctx = i915_gem_context_lookup(file_priv, args->ctx_id); - if (IS_ERR(ctx)) { - mutex_unlock(&dev->struct_mutex); - return PTR_ERR(ctx); - } + if (!ctx) + return -ENOENT; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + goto out; __destroy_hw_context(ctx, file_priv); mutex_unlock(&dev->struct_mutex); - DRM_DEBUG("HW context %d destroyed\n", args->ctx_id); +out: + i915_gem_context_put(ctx); return 0; } @@ -1004,17 +1056,11 @@ int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data, struct drm_i915_file_private *file_priv = file->driver_priv; struct drm_i915_gem_context_param *args = data; struct i915_gem_context *ctx; - int ret; - - ret = i915_mutex_lock_interruptible(dev); - if (ret) - return ret; + int ret = 0; ctx = i915_gem_context_lookup(file_priv, args->ctx_id); - if (IS_ERR(ctx)) { - mutex_unlock(&dev->struct_mutex); - return PTR_ERR(ctx); - } + if (!ctx) + return -ENOENT; args->size = 0; switch (args->param) { @@ -1042,8 +1088,8 @@ int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data, ret = -EINVAL; break; } - mutex_unlock(&dev->struct_mutex); + i915_gem_context_put(ctx); return ret; } @@ -1055,15 +1101,13 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data, struct i915_gem_context *ctx; int ret; + ctx = i915_gem_context_lookup(file_priv, args->ctx_id); + if (!ctx) + return -ENOENT; + ret = i915_mutex_lock_interruptible(dev); if (ret) - return ret; - - ctx = i915_gem_context_lookup(file_priv, args->ctx_id); - if (IS_ERR(ctx)) { - mutex_unlock(&dev->struct_mutex); - return PTR_ERR(ctx); - } + goto out; switch (args->param) { case I915_CONTEXT_PARAM_BAN_PERIOD: @@ -1101,6 +1145,8 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data, } mutex_unlock(&dev->struct_mutex); +out: + i915_gem_context_put(ctx); return ret; } @@ -1115,27 +1161,31 @@ int i915_gem_context_reset_stats_ioctl(struct drm_device *dev, if (args->flags || args->pad) return -EINVAL; - ret = i915_mutex_lock_interruptible(dev); - if (ret) - return ret; + ret = -ENOENT; + rcu_read_lock(); + ctx = __i915_gem_context_lookup_rcu(file->driver_priv, args->ctx_id); + if (!ctx) + goto out; - ctx = i915_gem_context_lookup(file->driver_priv, args->ctx_id); - if (IS_ERR(ctx)) { - mutex_unlock(&dev->struct_mutex); - return PTR_ERR(ctx); - } + /* + * We opt for unserialised reads here. This may result in tearing + * in the extremely unlikely event of a GPU hang on this context + * as we are querying them. If we need that extra layer of protection, + * we should wrap the hangstats with a seqlock. + */ if (capable(CAP_SYS_ADMIN)) args->reset_count = i915_reset_count(&dev_priv->gpu_error); else args->reset_count = 0; - args->batch_active = ctx->guilty_count; - args->batch_pending = ctx->active_count; + args->batch_active = READ_ONCE(ctx->guilty_count); + args->batch_pending = READ_ONCE(ctx->active_count); - mutex_unlock(&dev->struct_mutex); - - return 0; + ret = 0; +out: + rcu_read_unlock(); + return ret; } #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) diff --git a/drivers/gpu/drm/i915/i915_gem_context.h b/drivers/gpu/drm/i915/i915_gem_context.h index 82c99ba92ad3..04320f80f9f4 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.h +++ b/drivers/gpu/drm/i915/i915_gem_context.h @@ -86,6 +86,7 @@ struct i915_gem_context { /** link: place with &drm_i915_private.context_list */ struct list_head link; + struct llist_node free_link; /** * @ref: reference count @@ -99,6 +100,11 @@ struct i915_gem_context { struct kref ref; /** + * @rcu: rcu_head for deferred freeing. + */ + struct rcu_head rcu; + + /** * @flags: small set of booleans */ unsigned long flags; @@ -273,14 +279,18 @@ static inline bool i915_gem_context_is_kernel(struct i915_gem_context *ctx) } /* i915_gem_context.c */ -int __must_check i915_gem_context_init(struct drm_i915_private *dev_priv); -void i915_gem_context_lost(struct drm_i915_private *dev_priv); -void i915_gem_context_fini(struct drm_i915_private *dev_priv); -int i915_gem_context_open(struct drm_device *dev, struct drm_file *file); -void i915_gem_context_close(struct drm_device *dev, struct drm_file *file); +int __must_check i915_gem_contexts_init(struct drm_i915_private *dev_priv); +void i915_gem_contexts_lost(struct drm_i915_private *dev_priv); +void i915_gem_contexts_fini(struct drm_i915_private *dev_priv); + +int i915_gem_context_open(struct drm_i915_private *i915, + struct drm_file *file); +void i915_gem_context_close(struct drm_file *file); + int i915_switch_context(struct drm_i915_gem_request *req); int i915_gem_switch_to_kernel_context(struct drm_i915_private *dev_priv); -void i915_gem_context_free(struct kref *ctx_ref); + +void i915_gem_context_release(struct kref *ctx_ref); struct i915_gem_context * i915_gem_context_create_gvt(struct drm_device *dev); @@ -295,4 +305,16 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data, int i915_gem_context_reset_stats_ioctl(struct drm_device *dev, void *data, struct drm_file *file); +static inline struct i915_gem_context * +i915_gem_context_get(struct i915_gem_context *ctx) +{ + kref_get(&ctx->ref); + return ctx; +} + +static inline void i915_gem_context_put(struct i915_gem_context *ctx) +{ + kref_put(&ctx->ref, i915_gem_context_release); +} + #endif /* !__I915_GEM_CONTEXT_H__ */ diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 054b2e54cdaf..929f275e67aa 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -675,16 +675,17 @@ static int eb_select_context(struct i915_execbuffer *eb) struct i915_gem_context *ctx; ctx = i915_gem_context_lookup(eb->file->driver_priv, eb->args->rsvd1); - if (unlikely(IS_ERR(ctx))) - return PTR_ERR(ctx); + if (unlikely(!ctx)) + return -ENOENT; if (unlikely(i915_gem_context_is_banned(ctx))) { DRM_DEBUG("Context %u tried to submit while banned\n", ctx->user_handle); + i915_gem_context_put(ctx); return -EIO; } - eb->ctx = i915_gem_context_get(ctx); + eb->ctx = ctx; eb->vm = ctx->ppgtt ? &ctx->ppgtt->base : &eb->i915->ggtt.base; eb->context_flags = 0; @@ -2134,7 +2135,6 @@ i915_gem_do_execbuffer(struct drm_device *dev, if (DBG_FORCE_RELOC || !(args->flags & I915_EXEC_NO_RELOC)) args->flags |= __EXEC_HAS_RELOC; eb.exec = exec; - eb.ctx = NULL; eb.invalid_flags = __EXEC_OBJECT_UNKNOWN_FLAGS; if (USES_FULL_PPGTT(eb.i915)) eb.invalid_flags |= EXEC_OBJECT_NEEDS_GTT; @@ -2192,6 +2192,10 @@ i915_gem_do_execbuffer(struct drm_device *dev, GEM_BUG_ON(!eb.lut_size); + err = eb_select_context(&eb); + if (unlikely(err)) + goto err_destroy; + /* * Take a local wakeref for preparing to dispatch the execbuf as * we expect to access the hardware fairly frequently in the @@ -2200,14 +2204,11 @@ i915_gem_do_execbuffer(struct drm_device *dev, * 100ms. */ intel_runtime_pm_get(eb.i915); + err = i915_mutex_lock_interruptible(dev); if (err) goto err_rpm; - err = eb_select_context(&eb); - if (unlikely(err)) - goto err_unlock; - err = eb_relocate(&eb); if (err) /* @@ -2343,11 +2344,11 @@ err_batch_unpin: err_vma: if (eb.exec) eb_release_vmas(&eb); - i915_gem_context_put(eb.ctx); -err_unlock: mutex_unlock(&dev->struct_mutex); err_rpm: intel_runtime_pm_put(eb.i915); + i915_gem_context_put(eb.ctx); +err_destroy: eb_destroy(&eb); err_out_fence: if (out_fence_fd != -1) diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 61fc7e90a7da..10aa7762d9a6 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -207,8 +207,7 @@ static int ppgtt_bind_vma(struct i915_vma *vma, if (vma->obj->gt_ro) pte_flags |= PTE_READ_ONLY; - vma->vm->insert_entries(vma->vm, vma->pages, vma->node.start, - cache_level, pte_flags); + vma->vm->insert_entries(vma->vm, vma, cache_level, pte_flags); return 0; } @@ -907,37 +906,35 @@ gen8_ppgtt_insert_pte_entries(struct i915_hw_ppgtt *ppgtt, } static void gen8_ppgtt_insert_3lvl(struct i915_address_space *vm, - struct sg_table *pages, - u64 start, + struct i915_vma *vma, enum i915_cache_level cache_level, u32 unused) { struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm); struct sgt_dma iter = { - .sg = pages->sgl, + .sg = vma->pages->sgl, .dma = sg_dma_address(iter.sg), .max = iter.dma + iter.sg->length, }; - struct gen8_insert_pte idx = gen8_insert_pte(start); + struct gen8_insert_pte idx = gen8_insert_pte(vma->node.start); gen8_ppgtt_insert_pte_entries(ppgtt, &ppgtt->pdp, &iter, &idx, cache_level); } static void gen8_ppgtt_insert_4lvl(struct i915_address_space *vm, - struct sg_table *pages, - u64 start, + struct i915_vma *vma, enum i915_cache_level cache_level, u32 unused) { struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm); struct sgt_dma iter = { - .sg = pages->sgl, + .sg = vma->pages->sgl, .dma = sg_dma_address(iter.sg), .max = iter.dma + iter.sg->length, }; struct i915_page_directory_pointer **pdps = ppgtt->pml4.pdps; - struct gen8_insert_pte idx = gen8_insert_pte(start); + struct gen8_insert_pte idx = gen8_insert_pte(vma->node.start); while (gen8_ppgtt_insert_pte_entries(ppgtt, pdps[idx.pml4e++], &iter, &idx, cache_level)) @@ -1621,13 +1618,12 @@ static void gen6_ppgtt_clear_range(struct i915_address_space *vm, } static void gen6_ppgtt_insert_entries(struct i915_address_space *vm, - struct sg_table *pages, - u64 start, + struct i915_vma *vma, enum i915_cache_level cache_level, u32 flags) { struct i915_hw_ppgtt *ppgtt = i915_vm_to_ppgtt(vm); - unsigned first_entry = start >> PAGE_SHIFT; + unsigned first_entry = vma->node.start >> PAGE_SHIFT; unsigned act_pt = first_entry / GEN6_PTES; unsigned act_pte = first_entry % GEN6_PTES; const u32 pte_encode = vm->pte_encode(0, cache_level, flags); @@ -1635,7 +1631,7 @@ static void gen6_ppgtt_insert_entries(struct i915_address_space *vm, gen6_pte_t *vaddr; vaddr = kmap_atomic_px(ppgtt->pd.page_table[act_pt]); - iter.sg = pages->sgl; + iter.sg = vma->pages->sgl; iter.dma = sg_dma_address(iter.sg); iter.max = iter.dma + iter.sg->length; do { @@ -2090,8 +2086,7 @@ static void gen8_ggtt_insert_page(struct i915_address_space *vm, } static void gen8_ggtt_insert_entries(struct i915_address_space *vm, - struct sg_table *st, - u64 start, + struct i915_vma *vma, enum i915_cache_level level, u32 unused) { @@ -2102,8 +2097,8 @@ static void gen8_ggtt_insert_entries(struct i915_address_space *vm, dma_addr_t addr; gtt_entries = (gen8_pte_t __iomem *)ggtt->gsm; - gtt_entries += start >> PAGE_SHIFT; - for_each_sgt_dma(addr, sgt_iter, st) + gtt_entries += vma->node.start >> PAGE_SHIFT; + for_each_sgt_dma(addr, sgt_iter, vma->pages) gen8_set_pte(gtt_entries++, pte_encode | addr); wmb(); @@ -2137,17 +2132,16 @@ static void gen6_ggtt_insert_page(struct i915_address_space *vm, * mapped BAR (dev_priv->mm.gtt->gtt). */ static void gen6_ggtt_insert_entries(struct i915_address_space *vm, - struct sg_table *st, - u64 start, + struct i915_vma *vma, enum i915_cache_level level, u32 flags) { struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm); gen6_pte_t __iomem *entries = (gen6_pte_t __iomem *)ggtt->gsm; - unsigned int i = start >> PAGE_SHIFT; + unsigned int i = vma->node.start >> PAGE_SHIFT; struct sgt_iter iter; dma_addr_t addr; - for_each_sgt_dma(addr, iter, st) + for_each_sgt_dma(addr, iter, vma->pages) iowrite32(vm->pte_encode(addr, level, flags), &entries[i++]); wmb(); @@ -2229,8 +2223,7 @@ static void bxt_vtd_ggtt_insert_page__BKL(struct i915_address_space *vm, struct insert_entries { struct i915_address_space *vm; - struct sg_table *st; - u64 start; + struct i915_vma *vma; enum i915_cache_level level; }; @@ -2238,19 +2231,18 @@ static int bxt_vtd_ggtt_insert_entries__cb(void *_arg) { struct insert_entries *arg = _arg; - gen8_ggtt_insert_entries(arg->vm, arg->st, arg->start, arg->level, 0); + gen8_ggtt_insert_entries(arg->vm, arg->vma, arg->level, 0); bxt_vtd_ggtt_wa(arg->vm); return 0; } static void bxt_vtd_ggtt_insert_entries__BKL(struct i915_address_space *vm, - struct sg_table *st, - u64 start, + struct i915_vma *vma, enum i915_cache_level level, u32 unused) { - struct insert_entries arg = { vm, st, start, level }; + struct insert_entries arg = { vm, vma, level }; stop_machine(bxt_vtd_ggtt_insert_entries__cb, &arg, NULL); } @@ -2316,15 +2308,15 @@ static void i915_ggtt_insert_page(struct i915_address_space *vm, } static void i915_ggtt_insert_entries(struct i915_address_space *vm, - struct sg_table *pages, - u64 start, + struct i915_vma *vma, enum i915_cache_level cache_level, u32 unused) { unsigned int flags = (cache_level == I915_CACHE_NONE) ? AGP_USER_MEMORY : AGP_USER_CACHED_MEMORY; - intel_gtt_insert_sg_entries(pages, start >> PAGE_SHIFT, flags); + intel_gtt_insert_sg_entries(vma->pages, vma->node.start >> PAGE_SHIFT, + flags); } static void i915_ggtt_clear_range(struct i915_address_space *vm, @@ -2353,8 +2345,7 @@ static int ggtt_bind_vma(struct i915_vma *vma, pte_flags |= PTE_READ_ONLY; intel_runtime_pm_get(i915); - vma->vm->insert_entries(vma->vm, vma->pages, vma->node.start, - cache_level, pte_flags); + vma->vm->insert_entries(vma->vm, vma, cache_level, pte_flags); intel_runtime_pm_put(i915); /* @@ -2407,16 +2398,13 @@ static int aliasing_gtt_bind_vma(struct i915_vma *vma, goto err_pages; } - appgtt->base.insert_entries(&appgtt->base, - vma->pages, vma->node.start, - cache_level, pte_flags); + appgtt->base.insert_entries(&appgtt->base, vma, cache_level, + pte_flags); } if (flags & I915_VMA_GLOBAL_BIND) { intel_runtime_pm_get(i915); - vma->vm->insert_entries(vma->vm, - vma->pages, vma->node.start, - cache_level, pte_flags); + vma->vm->insert_entries(vma->vm, vma, cache_level, pte_flags); intel_runtime_pm_put(i915); } diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h index 1b2a56c3e5d3..b4e3aa7c0ce1 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.h +++ b/drivers/gpu/drm/i915/i915_gem_gtt.h @@ -313,8 +313,7 @@ struct i915_address_space { enum i915_cache_level cache_level, u32 flags); void (*insert_entries)(struct i915_address_space *vm, - struct sg_table *st, - u64 start, + struct i915_vma *vma, enum i915_cache_level cache_level, u32 flags); void (*cleanup)(struct i915_address_space *vm); diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c index 8c59c79cbd8b..483af8921060 100644 --- a/drivers/gpu/drm/i915/i915_gem_request.c +++ b/drivers/gpu/drm/i915/i915_gem_request.c @@ -384,7 +384,11 @@ static void i915_gem_request_retire(struct drm_i915_gem_request *request) engine->context_unpin(engine, engine->last_retired_context); engine->last_retired_context = request->ctx; - dma_fence_signal(&request->fence); + spin_lock_irq(&request->lock); + if (request->waitboost) + atomic_dec(&request->i915->rps.num_waiters); + dma_fence_signal_locked(&request->fence); + spin_unlock_irq(&request->lock); i915_priotree_fini(request->i915, &request->priotree); i915_gem_request_put(request); @@ -639,6 +643,7 @@ i915_gem_request_alloc(struct intel_engine_cs *engine, req->file_priv = NULL; req->batch = NULL; req->capture_list = NULL; + req->waitboost = false; /* * Reserve space in the ring buffer for all the commands required to diff --git a/drivers/gpu/drm/i915/i915_gem_request.h b/drivers/gpu/drm/i915/i915_gem_request.h index 7579b9702c22..49a4c8994ff0 100644 --- a/drivers/gpu/drm/i915/i915_gem_request.h +++ b/drivers/gpu/drm/i915/i915_gem_request.h @@ -184,6 +184,8 @@ struct drm_i915_gem_request { /** Time at which this request was emitted, in jiffies. */ unsigned long emitted_jiffies; + bool waitboost; + /** engine->request_list entry for this request */ struct list_head link; diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index e18f350bc364..ae70283470a6 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -463,6 +463,7 @@ static void error_print_engine(struct drm_i915_error_state_buf *m, err_printf(m, " hangcheck action timestamp: %lu, %u ms ago\n", ee->hangcheck_timestamp, jiffies_to_msecs(jiffies - ee->hangcheck_timestamp)); + err_printf(m, " engine reset count: %u\n", ee->reset_count); error_print_request(m, " ELSP[0]: ", &ee->execlist[0]); error_print_request(m, " ELSP[1]: ", &ee->execlist[1]); @@ -1236,6 +1237,8 @@ static void error_record_engine_registers(struct i915_gpu_state *error, ee->hangcheck_timestamp = engine->hangcheck.action_timestamp; ee->hangcheck_action = engine->hangcheck.action; ee->hangcheck_stalled = engine->hangcheck.stalled; + ee->reset_count = i915_reset_engine_count(&dev_priv->gpu_error, + engine); if (USES_PPGTT(dev_priv)) { int i; diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 4cd9ee1ba332..1d33cea01a1b 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -1091,18 +1091,6 @@ static u32 vlv_wa_c0_ei(struct drm_i915_private *dev_priv, u32 pm_iir) return events; } -static bool any_waiters(struct drm_i915_private *dev_priv) -{ - struct intel_engine_cs *engine; - enum intel_engine_id id; - - for_each_engine(engine, dev_priv, id) - if (intel_engine_has_waiter(engine)) - return true; - - return false; -} - static void gen6_pm_rps_work(struct work_struct *work) { struct drm_i915_private *dev_priv = @@ -1114,7 +1102,7 @@ static void gen6_pm_rps_work(struct work_struct *work) spin_lock_irq(&dev_priv->irq_lock); if (dev_priv->rps.interrupts_enabled) { pm_iir = fetch_and_zero(&dev_priv->rps.pm_iir); - client_boost = fetch_and_zero(&dev_priv->rps.client_boost); + client_boost = atomic_read(&dev_priv->rps.num_waiters); } spin_unlock_irq(&dev_priv->irq_lock); @@ -1131,7 +1119,7 @@ static void gen6_pm_rps_work(struct work_struct *work) new_delay = dev_priv->rps.cur_freq; min = dev_priv->rps.min_freq_softlimit; max = dev_priv->rps.max_freq_softlimit; - if (client_boost || any_waiters(dev_priv)) + if (client_boost) max = dev_priv->rps.max_freq; if (client_boost && new_delay < dev_priv->rps.boost_freq) { new_delay = dev_priv->rps.boost_freq; @@ -1144,7 +1132,7 @@ static void gen6_pm_rps_work(struct work_struct *work) if (new_delay >= dev_priv->rps.max_freq_softlimit) adj = 0; - } else if (client_boost || any_waiters(dev_priv)) { + } else if (client_boost) { adj = 0; } else if (pm_iir & GEN6_PM_RP_DOWN_TIMEOUT) { if (dev_priv->rps.cur_freq > dev_priv->rps.efficient_freq) @@ -2599,60 +2587,93 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg) return ret; } +struct wedge_me { + struct delayed_work work; + struct drm_i915_private *i915; + const char *name; +}; + +static void wedge_me(struct work_struct *work) +{ + struct wedge_me *w = container_of(work, typeof(*w), work.work); + + dev_err(w->i915->drm.dev, + "%s timed out, cancelling all in-flight rendering.\n", + w->name); + i915_gem_set_wedged(w->i915); +} + +static void __init_wedge(struct wedge_me *w, + struct drm_i915_private *i915, + long timeout, + const char *name) +{ + w->i915 = i915; + w->name = name; + + INIT_DELAYED_WORK_ONSTACK(&w->work, wedge_me); + schedule_delayed_work(&w->work, timeout); +} + +static void __fini_wedge(struct wedge_me *w) +{ + cancel_delayed_work_sync(&w->work); + destroy_delayed_work_on_stack(&w->work); + w->i915 = NULL; +} + +#define i915_wedge_on_timeout(W, DEV, TIMEOUT) \ + for (__init_wedge((W), (DEV), (TIMEOUT), __func__); \ + (W)->i915; \ + __fini_wedge((W))) + /** - * i915_reset_and_wakeup - do process context error handling work + * i915_reset_device - do process context error handling work * @dev_priv: i915 device private * * Fire an error uevent so userspace can see that a hang or error * was detected. */ -static void i915_reset_and_wakeup(struct drm_i915_private *dev_priv) +static void i915_reset_device(struct drm_i915_private *dev_priv) { struct kobject *kobj = &dev_priv->drm.primary->kdev->kobj; char *error_event[] = { I915_ERROR_UEVENT "=1", NULL }; char *reset_event[] = { I915_RESET_UEVENT "=1", NULL }; char *reset_done_event[] = { I915_ERROR_UEVENT "=0", NULL }; + struct wedge_me w; kobject_uevent_env(kobj, KOBJ_CHANGE, error_event); DRM_DEBUG_DRIVER("resetting chip\n"); kobject_uevent_env(kobj, KOBJ_CHANGE, reset_event); - intel_prepare_reset(dev_priv); + /* Use a watchdog to ensure that our reset completes */ + i915_wedge_on_timeout(&w, dev_priv, 5*HZ) { + intel_prepare_reset(dev_priv); - set_bit(I915_RESET_HANDOFF, &dev_priv->gpu_error.flags); - wake_up_all(&dev_priv->gpu_error.wait_queue); + /* Signal that locked waiters should reset the GPU */ + set_bit(I915_RESET_HANDOFF, &dev_priv->gpu_error.flags); + wake_up_all(&dev_priv->gpu_error.wait_queue); - do { - /* - * All state reset _must_ be completed before we update the - * reset counter, for otherwise waiters might miss the reset - * pending state and not properly drop locks, resulting in - * deadlocks with the reset work. + /* Wait for anyone holding the lock to wakeup, without + * blocking indefinitely on struct_mutex. */ - if (mutex_trylock(&dev_priv->drm.struct_mutex)) { - i915_reset(dev_priv); - mutex_unlock(&dev_priv->drm.struct_mutex); - } - - /* We need to wait for anyone holding the lock to wakeup */ - } while (wait_on_bit_timeout(&dev_priv->gpu_error.flags, - I915_RESET_HANDOFF, - TASK_UNINTERRUPTIBLE, - HZ)); + do { + if (mutex_trylock(&dev_priv->drm.struct_mutex)) { + i915_reset(dev_priv); + mutex_unlock(&dev_priv->drm.struct_mutex); + } + } while (wait_on_bit_timeout(&dev_priv->gpu_error.flags, + I915_RESET_HANDOFF, + TASK_UNINTERRUPTIBLE, + 1)); - intel_finish_reset(dev_priv); + intel_finish_reset(dev_priv); + } if (!test_bit(I915_WEDGED, &dev_priv->gpu_error.flags)) kobject_uevent_env(kobj, KOBJ_CHANGE, reset_done_event); - - /* - * Note: The wake_up also serves as a memory barrier so that - * waiters see the updated value of the dev_priv->gpu_error. - */ - clear_bit(I915_RESET_BACKOFF, &dev_priv->gpu_error.flags); - wake_up_all(&dev_priv->gpu_error.reset_queue); } static inline void @@ -2722,6 +2743,8 @@ void i915_handle_error(struct drm_i915_private *dev_priv, u32 engine_mask, const char *fmt, ...) { + struct intel_engine_cs *engine; + unsigned int tmp; va_list args; char error_msg[80]; @@ -2741,14 +2764,56 @@ void i915_handle_error(struct drm_i915_private *dev_priv, i915_capture_error_state(dev_priv, engine_mask, error_msg); i915_clear_error_registers(dev_priv); + /* + * Try engine reset when available. We fall back to full reset if + * single reset fails. + */ + if (intel_has_reset_engine(dev_priv)) { + for_each_engine_masked(engine, dev_priv, engine_mask, tmp) { + BUILD_BUG_ON(I915_RESET_HANDOFF >= I915_RESET_ENGINE); + if (test_and_set_bit(I915_RESET_ENGINE + engine->id, + &dev_priv->gpu_error.flags)) + continue; + + if (i915_reset_engine(engine) == 0) + engine_mask &= ~intel_engine_flag(engine); + + clear_bit(I915_RESET_ENGINE + engine->id, + &dev_priv->gpu_error.flags); + wake_up_bit(&dev_priv->gpu_error.flags, + I915_RESET_ENGINE + engine->id); + } + } + if (!engine_mask) goto out; - if (test_and_set_bit(I915_RESET_BACKOFF, - &dev_priv->gpu_error.flags)) + /* Full reset needs the mutex, stop any other user trying to do so. */ + if (test_and_set_bit(I915_RESET_BACKOFF, &dev_priv->gpu_error.flags)) { + wait_event(dev_priv->gpu_error.reset_queue, + !test_bit(I915_RESET_BACKOFF, + &dev_priv->gpu_error.flags)); goto out; + } + + /* Prevent any other reset-engine attempt. */ + for_each_engine(engine, dev_priv, tmp) { + while (test_and_set_bit(I915_RESET_ENGINE + engine->id, + &dev_priv->gpu_error.flags)) + wait_on_bit(&dev_priv->gpu_error.flags, + I915_RESET_ENGINE + engine->id, + TASK_UNINTERRUPTIBLE); + } - i915_reset_and_wakeup(dev_priv); + i915_reset_device(dev_priv); + + for_each_engine(engine, dev_priv, tmp) { + clear_bit(I915_RESET_ENGINE + engine->id, + &dev_priv->gpu_error.flags); + } + + clear_bit(I915_RESET_BACKOFF, &dev_priv->gpu_error.flags); + wake_up_all(&dev_priv->gpu_error.reset_queue); out: intel_runtime_pm_put(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c index b6a7e363d076..88b9d3e6713a 100644 --- a/drivers/gpu/drm/i915/i915_params.c +++ b/drivers/gpu/drm/i915/i915_params.c @@ -46,7 +46,7 @@ struct i915_params i915 __read_mostly = { .prefault_disable = 0, .load_detect_test = 0, .force_reset_modeset_test = 0, - .reset = true, + .reset = 2, .error_capture = true, .invert_brightness = 0, .disable_display = 0, @@ -63,8 +63,9 @@ struct i915_params i915 __read_mostly = { .huc_firmware_path = NULL, .enable_dp_mst = true, .inject_load_failure = 0, - .enable_dpcd_backlight = false, + .enable_dpcd_backlight = -1, .enable_gvt = false, + .enable_dbc = true, }; module_param_named(modeset, i915.modeset, int, 0400); @@ -115,8 +116,8 @@ MODULE_PARM_DESC(vbt_sdvo_panel_type, "Override/Ignore selection of SDVO panel mode in the VBT " "(-2=ignore, -1=auto [default], index in VBT BIOS table)"); -module_param_named_unsafe(reset, i915.reset, bool, 0600); -MODULE_PARM_DESC(reset, "Attempt GPU resets (default: true)"); +module_param_named_unsafe(reset, i915.reset, int, 0600); +MODULE_PARM_DESC(reset, "Attempt GPU resets (0=disabled, 1=full gpu reset, 2=engine reset [default])"); #if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR) module_param_named(error_capture, i915.error_capture, bool, 0600); @@ -246,10 +247,15 @@ MODULE_PARM_DESC(enable_dp_mst, module_param_named_unsafe(inject_load_failure, i915.inject_load_failure, uint, 0400); MODULE_PARM_DESC(inject_load_failure, "Force an error after a number of failure check points (0:disabled (default), N:force failure at the Nth failure check point)"); -module_param_named(enable_dpcd_backlight, i915.enable_dpcd_backlight, bool, 0600); +module_param_named_unsafe(enable_dpcd_backlight, i915.enable_dpcd_backlight, int, 0600); MODULE_PARM_DESC(enable_dpcd_backlight, - "Enable support for DPCD backlight control (default:false)"); + "Enable support for DPCD backlight control " + "(-1:auto (default), 0:force disable, 1:force enabled if supported"); module_param_named(enable_gvt, i915.enable_gvt, bool, 0400); MODULE_PARM_DESC(enable_gvt, "Enable support for Intel GVT-g graphics virtualization host support(default:false)"); + +module_param_named_unsafe(enable_dbc, i915.enable_dbc, bool, 0600); +MODULE_PARM_DESC(enable_dbc, + "Enable support for dynamic backlight control (default:true)"); diff --git a/drivers/gpu/drm/i915/i915_params.h b/drivers/gpu/drm/i915/i915_params.h index 34148cc8637c..057e203e6bda 100644 --- a/drivers/gpu/drm/i915/i915_params.h +++ b/drivers/gpu/drm/i915/i915_params.h @@ -51,7 +51,9 @@ func(int, use_mmio_flip); \ func(int, mmio_debug); \ func(int, edp_vswing); \ + func(int, reset); \ func(unsigned int, inject_load_failure); \ + func(int, enable_dpcd_backlight); \ /* leave bools at the end to not create holes */ \ func(bool, alpha_support); \ func(bool, enable_cmd_parser); \ @@ -60,14 +62,13 @@ func(bool, prefault_disable); \ func(bool, load_detect_test); \ func(bool, force_reset_modeset_test); \ - func(bool, reset); \ func(bool, error_capture); \ func(bool, disable_display); \ func(bool, verbose_state_checks); \ func(bool, nuclear_pageflip); \ func(bool, enable_dp_mst); \ - func(bool, enable_dpcd_backlight); \ - func(bool, enable_gvt) + func(bool, enable_gvt); \ + func(bool, enable_dbc) #define MEMBER(T, member) T member struct i915_params { diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 506ec32b9e53..a1e6b696bcfa 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -310,7 +310,8 @@ static const struct intel_device_info intel_haswell_info = { BDW_COLORS, \ .has_logical_ring_contexts = 1, \ .has_full_48bit_ppgtt = 1, \ - .has_64bit_reloc = 1 + .has_64bit_reloc = 1, \ + .has_reset_engine = 1 #define BDW_PLATFORM \ BDW_FEATURES, \ @@ -342,6 +343,7 @@ static const struct intel_device_info intel_cherryview_info = { .has_gmch_display = 1, .has_aliasing_ppgtt = 1, .has_full_ppgtt = 1, + .has_reset_engine = 1, .display_mmio_offset = VLV_DISPLAY_BASE, GEN_CHV_PIPEOFFSETS, CURSOR_OFFSETS, @@ -387,6 +389,7 @@ static const struct intel_device_info intel_skylake_gt3_info = { .has_aliasing_ppgtt = 1, \ .has_full_ppgtt = 1, \ .has_full_48bit_ppgtt = 1, \ + .has_reset_engine = 1, \ GEN_DEFAULT_PIPEOFFSETS, \ IVB_CURSOR_OFFSETS, \ BDW_COLORS @@ -446,6 +449,7 @@ static const struct intel_device_info intel_cannonlake_info = { .gen = 10, .ddb_size = 1024, .has_csr = 1, + .color = { .degamma_lut_size = 0, .gamma_lut_size = 1024 } }; /* diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index 9cd22f83b0cf..d9f77a4d85db 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -1746,7 +1746,7 @@ static int gen8_configure_all_contexts(struct drm_i915_private *dev_priv, goto out; /* Update all contexts now that we've stalled the submission. */ - list_for_each_entry(ctx, &dev_priv->context_list, link) { + list_for_each_entry(ctx, &dev_priv->contexts.list, link) { struct intel_context *ce = &ctx->engine[RCS]; u32 *regs; @@ -2444,7 +2444,7 @@ static void i915_perf_destroy_locked(struct i915_perf_stream *stream) list_del(&stream->link); if (stream->ctx) - i915_gem_context_put_unlocked(stream->ctx); + i915_gem_context_put(stream->ctx); kfree(stream); } @@ -2633,7 +2633,7 @@ err_alloc: kfree(stream); err_ctx: if (specific_ctx) - i915_gem_context_put_unlocked(specific_ctx); + i915_gem_context_put(specific_ctx); err: return ret; } diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 64cc674b652a..c712d01f92ab 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -3522,7 +3522,7 @@ enum skl_disp_power_wells { #define INTERVAL_1_28_US(us) roundup(((us) * 100) >> 7, 25) #define INTERVAL_1_33_US(us) (((us) * 3) >> 2) #define INTERVAL_0_833_US(us) (((us) * 6) / 5) -#define GT_INTERVAL_FROM_US(dev_priv, us) (IS_GEN9(dev_priv) ? \ +#define GT_INTERVAL_FROM_US(dev_priv, us) (INTEL_GEN(dev_priv) >= 9 ? \ (IS_GEN9_LP(dev_priv) ? \ INTERVAL_0_833_US(us) : \ INTERVAL_1_33_US(us)) : \ @@ -3531,7 +3531,7 @@ enum skl_disp_power_wells { #define INTERVAL_1_28_TO_US(interval) (((interval) << 7) / 100) #define INTERVAL_1_33_TO_US(interval) (((interval) << 2) / 3) #define INTERVAL_0_833_TO_US(interval) (((interval) * 5) / 6) -#define GT_PM_INTERVAL_TO_US(dev_priv, interval) (IS_GEN9(dev_priv) ? \ +#define GT_PM_INTERVAL_TO_US(dev_priv, interval) (INTEL_GEN(dev_priv) >= 9 ? \ (IS_GEN9_LP(dev_priv) ? \ INTERVAL_0_833_TO_US(interval) : \ INTERVAL_1_33_TO_US(interval)) : \ @@ -8343,6 +8343,7 @@ enum { #define DPLL_CFGCR0_LINK_RATE_3240 (6 << 25) #define DPLL_CFGCR0_LINK_RATE_4050 (7 << 25) #define DPLL_CFGCR0_DCO_FRACTION_MASK (0x7fff << 10) +#define DPLL_CFGCR0_DCO_FRAC_SHIFT (10) #define DPLL_CFGCR0_DCO_FRACTION(x) ((x) << 10) #define DPLL_CFGCR0_DCO_INTEGER_MASK (0x3ff) #define CNL_DPLL_CFGCR0(pll) _MMIO_PLL(pll, _CNL_DPLL0_CFGCR0, _CNL_DPLL1_CFGCR0) @@ -8350,6 +8351,7 @@ enum { #define _CNL_DPLL0_CFGCR1 0x6C004 #define _CNL_DPLL1_CFGCR1 0x6C084 #define DPLL_CFGCR1_QDIV_RATIO_MASK (0xff << 10) +#define DPLL_CFGCR1_QDIV_RATIO_SHIFT (10) #define DPLL_CFGCR1_QDIV_RATIO(x) ((x) << 10) #define DPLL_CFGCR1_QDIV_MODE(x) ((x) << 9) #define DPLL_CFGCR1_KDIV_MASK (7 << 6) diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c index 1eef3fae4db3..7fcf00622c4c 100644 --- a/drivers/gpu/drm/i915/i915_sysfs.c +++ b/drivers/gpu/drm/i915/i915_sysfs.c @@ -96,7 +96,7 @@ static struct attribute *rc6_attrs[] = { NULL }; -static struct attribute_group rc6_attr_group = { +static const struct attribute_group rc6_attr_group = { .name = power_group_name, .attrs = rc6_attrs }; @@ -107,7 +107,7 @@ static struct attribute *rc6p_attrs[] = { NULL }; -static struct attribute_group rc6p_attr_group = { +static const struct attribute_group rc6p_attr_group = { .name = power_group_name, .attrs = rc6p_attrs }; @@ -117,7 +117,7 @@ static struct attribute *media_rc6_attrs[] = { NULL }; -static struct attribute_group media_rc6_attr_group = { +static const struct attribute_group media_rc6_attr_group = { .name = power_group_name, .attrs = media_rc6_attrs }; @@ -209,7 +209,7 @@ i915_l3_write(struct file *filp, struct kobject *kobj, memcpy(*remap_info + (offset/4), buf, count); /* NB: We defer the remapping until we switch to the context */ - list_for_each_entry(ctx, &dev_priv->context_list, link) + list_for_each_entry(ctx, &dev_priv->contexts.list, link) ctx->remap_slice |= (1<<slice); ret = count; @@ -253,7 +253,7 @@ static ssize_t gt_act_freq_mhz_show(struct device *kdev, ret = intel_gpu_freq(dev_priv, (freq >> 8) & 0xff); } else { u32 rpstat = I915_READ(GEN6_RPSTAT1); - if (IS_GEN9(dev_priv)) + if (INTEL_GEN(dev_priv) >= 9) ret = (rpstat & GEN9_CAGF_MASK) >> GEN9_CAGF_SHIFT; else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) ret = (rpstat & HSW_CAGF_MASK) >> HSW_CAGF_SHIFT; diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index 1cfe137cdc32..958be0a95960 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -579,11 +579,17 @@ err_unpin: static void i915_vma_destroy(struct i915_vma *vma) { + int i; + GEM_BUG_ON(vma->node.allocated); GEM_BUG_ON(i915_vma_is_active(vma)); GEM_BUG_ON(!i915_vma_is_closed(vma)); GEM_BUG_ON(vma->fence); + for (i = 0; i < ARRAY_SIZE(vma->last_read); i++) + GEM_BUG_ON(i915_gem_active_isset(&vma->last_read[i])); + GEM_BUG_ON(i915_gem_active_isset(&vma->last_fence)); + list_del(&vma->vm_link); if (!i915_vma_is_ggtt(vma)) i915_ppgtt_put(i915_vm_to_ppgtt(vma->vm)); @@ -680,9 +686,8 @@ int i915_vma_unbind(struct i915_vma *vma) __i915_vma_unpin(vma); if (ret) return ret; - - GEM_BUG_ON(i915_vma_is_active(vma)); } + GEM_BUG_ON(i915_vma_is_active(vma)); if (i915_vma_is_pinned(vma)) return -EBUSY; diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.c b/drivers/gpu/drm/i915/intel_atomic_plane.c index 4325cb0a04f5..ee76fab7bb6f 100644 --- a/drivers/gpu/drm/i915/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/intel_atomic_plane.c @@ -114,6 +114,8 @@ int intel_plane_atomic_check_with_state(struct intel_crtc_state *crtc_state, struct drm_i915_private *dev_priv = to_i915(plane->dev); struct drm_plane_state *state = &intel_state->base; struct intel_plane *intel_plane = to_intel_plane(plane); + const struct drm_display_mode *adjusted_mode = + &crtc_state->base.adjusted_mode; int ret; /* @@ -173,6 +175,19 @@ int intel_plane_atomic_check_with_state(struct intel_crtc_state *crtc_state, if (ret) return ret; + /* + * Y-tiling is not supported in IF-ID Interlace mode in + * GEN9 and above. + */ + if (state->fb && INTEL_GEN(dev_priv) >= 9 && crtc_state->base.enable && + adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) { + if (state->fb->modifier == I915_FORMAT_MOD_Y_TILED || + state->fb->modifier == I915_FORMAT_MOD_Yf_TILED) { + DRM_DEBUG_KMS("Y/Yf tiling not supported in IF-ID mode\n"); + return -EINVAL; + } + } + /* FIXME pre-g4x don't work like this */ if (intel_state->base.visible) crtc_state->active_planes |= BIT(intel_plane->id); diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c index 639d45c1dd2e..82b144cdfa1d 100644 --- a/drivers/gpu/drm/i915/intel_bios.c +++ b/drivers/gpu/drm/i915/intel_bios.c @@ -1187,6 +1187,15 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port, if (is_dvi) { info->alternate_ddc_pin = ddc_pin; + /* + * All VBTs that we got so far for B Stepping has this + * information wrong for Port D. So, let's just ignore for now. + */ + if (IS_CNL_REVID(dev_priv, CNL_REVID_B0, CNL_REVID_B0) && + port == PORT_D) { + info->alternate_ddc_pin = 0; + } + sanitize_ddc_pin(dev_priv, port); } diff --git a/drivers/gpu/drm/i915/intel_color.c b/drivers/gpu/drm/i915/intel_color.c index 306c6b06b330..f85d57555957 100644 --- a/drivers/gpu/drm/i915/intel_color.c +++ b/drivers/gpu/drm/i915/intel_color.c @@ -615,7 +615,7 @@ void intel_color_init(struct drm_crtc *crtc) IS_BROXTON(dev_priv)) { dev_priv->display.load_csc_matrix = i9xx_load_csc_matrix; dev_priv->display.load_luts = broadwell_load_luts; - } else if (IS_GEMINILAKE(dev_priv)) { + } else if (IS_GEMINILAKE(dev_priv) || IS_CANNONLAKE(dev_priv)) { dev_priv->display.load_csc_matrix = i9xx_load_csc_matrix; dev_priv->display.load_luts = glk_load_luts; } else { diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c index 80e96f1f49d2..efb13582dc73 100644 --- a/drivers/gpu/drm/i915/intel_ddi.c +++ b/drivers/gpu/drm/i915/intel_ddi.c @@ -1103,6 +1103,62 @@ static int skl_calc_wrpll_link(struct drm_i915_private *dev_priv, return dco_freq / (p0 * p1 * p2 * 5); } +static int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv, + uint32_t pll_id) +{ + uint32_t cfgcr0, cfgcr1; + uint32_t p0, p1, p2, dco_freq, ref_clock; + + cfgcr0 = I915_READ(CNL_DPLL_CFGCR0(pll_id)); + cfgcr1 = I915_READ(CNL_DPLL_CFGCR1(pll_id)); + + p0 = cfgcr1 & DPLL_CFGCR1_PDIV_MASK; + p2 = cfgcr1 & DPLL_CFGCR1_KDIV_MASK; + + if (cfgcr1 & DPLL_CFGCR1_QDIV_MODE(1)) + p1 = (cfgcr1 & DPLL_CFGCR1_QDIV_RATIO_MASK) >> + DPLL_CFGCR1_QDIV_RATIO_SHIFT; + else + p1 = 1; + + + switch (p0) { + case DPLL_CFGCR1_PDIV_2: + p0 = 2; + break; + case DPLL_CFGCR1_PDIV_3: + p0 = 3; + break; + case DPLL_CFGCR1_PDIV_5: + p0 = 5; + break; + case DPLL_CFGCR1_PDIV_7: + p0 = 7; + break; + } + + switch (p2) { + case DPLL_CFGCR1_KDIV_1: + p2 = 1; + break; + case DPLL_CFGCR1_KDIV_2: + p2 = 2; + break; + case DPLL_CFGCR1_KDIV_4: + p2 = 4; + break; + } + + ref_clock = dev_priv->cdclk.hw.ref; + + dco_freq = (cfgcr0 & DPLL_CFGCR0_DCO_INTEGER_MASK) * ref_clock; + + dco_freq += (((cfgcr0 & DPLL_CFGCR0_DCO_FRACTION_MASK) >> + DPLL_CFGCR0_DCO_FRAC_SHIFT) * ref_clock) / 0x8000; + + return dco_freq / (p0 * p1 * p2 * 5); +} + static void ddi_dotclock_get(struct intel_crtc_state *pipe_config) { int dotclock; @@ -1124,6 +1180,59 @@ static void ddi_dotclock_get(struct intel_crtc_state *pipe_config) pipe_config->base.adjusted_mode.crtc_clock = dotclock; } +static void cnl_ddi_clock_get(struct intel_encoder *encoder, + struct intel_crtc_state *pipe_config) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + int link_clock = 0; + uint32_t cfgcr0, pll_id; + + pll_id = intel_get_shared_dpll_id(dev_priv, pipe_config->shared_dpll); + + cfgcr0 = I915_READ(CNL_DPLL_CFGCR0(pll_id)); + + if (cfgcr0 & DPLL_CFGCR0_HDMI_MODE) { + link_clock = cnl_calc_wrpll_link(dev_priv, pll_id); + } else { + link_clock = cfgcr0 & DPLL_CFGCR0_LINK_RATE_MASK; + + switch (link_clock) { + case DPLL_CFGCR0_LINK_RATE_810: + link_clock = 81000; + break; + case DPLL_CFGCR0_LINK_RATE_1080: + link_clock = 108000; + break; + case DPLL_CFGCR0_LINK_RATE_1350: + link_clock = 135000; + break; + case DPLL_CFGCR0_LINK_RATE_1620: + link_clock = 162000; + break; + case DPLL_CFGCR0_LINK_RATE_2160: + link_clock = 216000; + break; + case DPLL_CFGCR0_LINK_RATE_2700: + link_clock = 270000; + break; + case DPLL_CFGCR0_LINK_RATE_3240: + link_clock = 324000; + break; + case DPLL_CFGCR0_LINK_RATE_4050: + link_clock = 405000; + break; + default: + WARN(1, "Unsupported link rate\n"); + break; + } + link_clock *= 2; + } + + pipe_config->port_clock = link_clock; + + ddi_dotclock_get(pipe_config); +} + static void skl_ddi_clock_get(struct intel_encoder *encoder, struct intel_crtc_state *pipe_config) { @@ -1267,6 +1376,8 @@ void intel_ddi_clock_get(struct intel_encoder *encoder, skl_ddi_clock_get(encoder, pipe_config); else if (IS_GEN9_LP(dev_priv)) bxt_ddi_clock_get(encoder, pipe_config); + else if (IS_CANNONLAKE(dev_priv)) + cnl_ddi_clock_get(encoder, pipe_config); } void intel_ddi_set_pipe_settings(const struct intel_crtc_state *crtc_state) @@ -1868,9 +1979,12 @@ static void cnl_ddi_vswing_sequence(struct intel_encoder *encoder, u32 level) if ((intel_dp) && (type == INTEL_OUTPUT_EDP || type == INTEL_OUTPUT_DP)) { width = intel_dp->lane_count; rate = intel_dp->link_rate; - } else { + } else if (type == INTEL_OUTPUT_HDMI) { width = 4; /* Rate is always < than 6GHz for HDMI */ + } else { + MISSING_CASE(type); + return; } /* diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c index 77d3214e1a77..5f91ddc78c7a 100644 --- a/drivers/gpu/drm/i915/intel_device_info.c +++ b/drivers/gpu/drm/i915/intel_device_info.c @@ -363,7 +363,7 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv) */ if (fuse_strap & ILK_INTERNAL_DISPLAY_DISABLE || sfuse_strap & SFUSE_STRAP_DISPLAY_DISABLED || - (dev_priv->pch_type == PCH_CPT && + (HAS_PCH_CPT(dev_priv) && !(sfuse_strap & SFUSE_STRAP_FUSE_LOCK))) { DRM_INFO("Display fused off, disabling\n"); info->num_pipes = 0; diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index dec9e58545a1..bb9c9c3c391f 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -3311,7 +3311,7 @@ u32 skl_plane_ctl(const struct intel_crtc_state *crtc_state, plane_ctl = PLANE_CTL_ENABLE; - if (!IS_GEMINILAKE(dev_priv)) { + if (!IS_GEMINILAKE(dev_priv) && !IS_CANNONLAKE(dev_priv)) { plane_ctl |= PLANE_CTL_PIPE_GAMMA_ENABLE | PLANE_CTL_PIPE_CSC_ENABLE | @@ -3367,7 +3367,7 @@ static void skylake_update_primary_plane(struct intel_plane *plane, spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); - if (IS_GEMINILAKE(dev_priv)) { + if (IS_GEMINILAKE(dev_priv) || IS_CANNONLAKE(dev_priv)) { I915_WRITE_FW(PLANE_COLOR_CTL(pipe, plane_id), PLANE_COLOR_PIPE_GAMMA_ENABLE | PLANE_COLOR_PIPE_CSC_ENABLE | @@ -4612,6 +4612,9 @@ skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach, &crtc_state->scaler_state; struct intel_crtc *intel_crtc = to_intel_crtc(crtc_state->base.crtc); + struct drm_i915_private *dev_priv = to_i915(intel_crtc->base.dev); + const struct drm_display_mode *adjusted_mode = + &crtc_state->base.adjusted_mode; int need_scaling; /* @@ -4622,6 +4625,18 @@ skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach, need_scaling = src_w != dst_w || src_h != dst_h; /* + * Scaling/fitting not supported in IF-ID mode in GEN9+ + * TODO: Interlace fetch mode doesn't support YUV420 planar formats. + * Once NV12 is enabled, handle it here while allocating scaler + * for NV12. + */ + if (INTEL_GEN(dev_priv) >= 9 && crtc_state->base.enable && + need_scaling && adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) { + DRM_DEBUG_KMS("Pipe/Plane scaling not supported with IF-ID mode\n"); + return -EINVAL; + } + + /* * if plane is being disabled or scaler is no more required or force detach * - free scaler binded to this plane/crtc * - in order to do this, update crtc->scaler_usage @@ -14765,6 +14780,17 @@ static void quirk_backlight_present(struct drm_device *dev) DRM_INFO("applying backlight present quirk\n"); } +/* Toshiba Satellite P50-C-18C requires T12 delay to be min 800ms + * which is 300 ms greater than eDP spec T12 min. + */ +static void quirk_increase_t12_delay(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = to_i915(dev); + + dev_priv->quirks |= QUIRK_INCREASE_T12_DELAY; + DRM_INFO("Applying T12 delay quirk\n"); +} + struct intel_quirk { int device; int subsystem_vendor; @@ -14848,6 +14874,9 @@ static struct intel_quirk intel_quirks[] = { /* Dell Chromebook 11 (2015 version) */ { 0x0a16, 0x1028, 0x0a35, quirk_backlight_present }, + + /* Toshiba Satellite P50-C-18C */ + { 0x191B, 0x1179, 0xF840, quirk_increase_t12_delay }, }; static void intel_init_quirks(struct drm_device *dev) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 64fa774c855b..2d42d09428c9 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -4418,8 +4418,6 @@ static bool ibx_digital_port_connected(struct drm_i915_private *dev_priv, u32 bit; switch (port->port) { - case PORT_A: - return true; case PORT_B: bit = SDE_PORTB_HOTPLUG; break; @@ -4443,8 +4441,6 @@ static bool cpt_digital_port_connected(struct drm_i915_private *dev_priv, u32 bit; switch (port->port) { - case PORT_A: - return true; case PORT_B: bit = SDE_PORTB_HOTPLUG_CPT; break; @@ -4454,12 +4450,28 @@ static bool cpt_digital_port_connected(struct drm_i915_private *dev_priv, case PORT_D: bit = SDE_PORTD_HOTPLUG_CPT; break; + default: + MISSING_CASE(port->port); + return false; + } + + return I915_READ(SDEISR) & bit; +} + +static bool spt_digital_port_connected(struct drm_i915_private *dev_priv, + struct intel_digital_port *port) +{ + u32 bit; + + switch (port->port) { + case PORT_A: + bit = SDE_PORTA_HOTPLUG_SPT; + break; case PORT_E: bit = SDE_PORTE_HOTPLUG_SPT; break; default: - MISSING_CASE(port->port); - return false; + return cpt_digital_port_connected(dev_priv, port); } return I915_READ(SDEISR) & bit; @@ -4511,6 +4523,42 @@ static bool gm45_digital_port_connected(struct drm_i915_private *dev_priv, return I915_READ(PORT_HOTPLUG_STAT) & bit; } +static bool ilk_digital_port_connected(struct drm_i915_private *dev_priv, + struct intel_digital_port *port) +{ + if (port->port == PORT_A) + return I915_READ(DEISR) & DE_DP_A_HOTPLUG; + else + return ibx_digital_port_connected(dev_priv, port); +} + +static bool snb_digital_port_connected(struct drm_i915_private *dev_priv, + struct intel_digital_port *port) +{ + if (port->port == PORT_A) + return I915_READ(DEISR) & DE_DP_A_HOTPLUG; + else + return cpt_digital_port_connected(dev_priv, port); +} + +static bool ivb_digital_port_connected(struct drm_i915_private *dev_priv, + struct intel_digital_port *port) +{ + if (port->port == PORT_A) + return I915_READ(DEISR) & DE_DP_A_HOTPLUG_IVB; + else + return cpt_digital_port_connected(dev_priv, port); +} + +static bool bdw_digital_port_connected(struct drm_i915_private *dev_priv, + struct intel_digital_port *port) +{ + if (port->port == PORT_A) + return I915_READ(GEN8_DE_PORT_ISR) & GEN8_PORT_DP_A_HOTPLUG; + else + return cpt_digital_port_connected(dev_priv, port); +} + static bool bxt_digital_port_connected(struct drm_i915_private *dev_priv, struct intel_digital_port *intel_dig_port) { @@ -4547,16 +4595,25 @@ static bool bxt_digital_port_connected(struct drm_i915_private *dev_priv, bool intel_digital_port_connected(struct drm_i915_private *dev_priv, struct intel_digital_port *port) { - if (HAS_PCH_IBX(dev_priv)) - return ibx_digital_port_connected(dev_priv, port); - else if (HAS_PCH_SPLIT(dev_priv)) - return cpt_digital_port_connected(dev_priv, port); + if (HAS_GMCH_DISPLAY(dev_priv)) { + if (IS_GM45(dev_priv)) + return gm45_digital_port_connected(dev_priv, port); + else + return g4x_digital_port_connected(dev_priv, port); + } + + if (IS_GEN5(dev_priv)) + return ilk_digital_port_connected(dev_priv, port); + else if (IS_GEN6(dev_priv)) + return snb_digital_port_connected(dev_priv, port); + else if (IS_GEN7(dev_priv)) + return ivb_digital_port_connected(dev_priv, port); + else if (IS_GEN8(dev_priv)) + return bdw_digital_port_connected(dev_priv, port); else if (IS_GEN9_LP(dev_priv)) return bxt_digital_port_connected(dev_priv, port); - else if (IS_GM45(dev_priv)) - return gm45_digital_port_connected(dev_priv, port); else - return g4x_digital_port_connected(dev_priv, port); + return spt_digital_port_connected(dev_priv, port); } static struct edid * @@ -5121,12 +5178,8 @@ intel_pps_readout_hw_state(struct drm_i915_private *dev_priv, PANEL_POWER_DOWN_DELAY_SHIFT; if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv)) { - u16 tmp = (pp_ctl & BXT_POWER_CYCLE_DELAY_MASK) >> - BXT_POWER_CYCLE_DELAY_SHIFT; - if (tmp > 0) - seq->t11_t12 = (tmp - 1) * 1000; - else - seq->t11_t12 = 0; + seq->t11_t12 = ((pp_ctl & BXT_POWER_CYCLE_DELAY_MASK) >> + BXT_POWER_CYCLE_DELAY_SHIFT) * 1000; } else { seq->t11_t12 = ((pp_div & PANEL_POWER_CYCLE_DELAY_MASK) >> PANEL_POWER_CYCLE_DELAY_SHIFT) * 1000; @@ -5177,6 +5230,21 @@ intel_dp_init_panel_power_sequencer(struct drm_device *dev, intel_pps_dump_state("cur", &cur); vbt = dev_priv->vbt.edp.pps; + /* On Toshiba Satellite P50-C-18C system the VBT T12 delay + * of 500ms appears to be too short. Ocassionally the panel + * just fails to power back on. Increasing the delay to 800ms + * seems sufficient to avoid this problem. + */ + if (dev_priv->quirks & QUIRK_INCREASE_T12_DELAY) { + vbt.t11_t12 = max_t(u16, vbt.t11_t12, 800 * 10); + DRM_DEBUG_KMS("Increasing T12 panel delay as per the quirk to %d\n", + vbt.t11_t12); + } + /* T11_T12 delay is special and actually in units of 100ms, but zero + * based in the hw (so we need to add 100 ms). But the sw vbt + * table multiplies it with 1000 to make it in units of 100usec, + * too. */ + vbt.t11_t12 += 100 * 10; /* Upper limits from eDP 1.3 spec. Note that we use the clunky units of * our hw here, which are all in 100usec. */ @@ -5280,7 +5348,7 @@ intel_dp_init_panel_power_sequencer_registers(struct drm_device *dev, if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv)) { pp_div = I915_READ(regs.pp_ctrl); pp_div &= ~BXT_POWER_CYCLE_DELAY_MASK; - pp_div |= (DIV_ROUND_UP((seq->t11_t12 + 1), 1000) + pp_div |= (DIV_ROUND_UP(seq->t11_t12, 1000) << BXT_POWER_CYCLE_DELAY_SHIFT); } else { pp_div = ((100 * div)/2 - 1) << PP_REFERENCE_DIVIDER_SHIFT; diff --git a/drivers/gpu/drm/i915/intel_dp_aux_backlight.c b/drivers/gpu/drm/i915/intel_dp_aux_backlight.c index 228ca06d9f0b..b25cd88fc1c5 100644 --- a/drivers/gpu/drm/i915/intel_dp_aux_backlight.c +++ b/drivers/gpu/drm/i915/intel_dp_aux_backlight.c @@ -98,13 +98,105 @@ intel_dp_aux_set_backlight(const struct drm_connector_state *conn_state, u32 lev } } +/* + * Set PWM Frequency divider to match desired frequency in vbt. + * The PWM Frequency is calculated as 27Mhz / (F x P). + * - Where F = PWM Frequency Pre-Divider value programmed by field 7:0 of the + * EDP_BACKLIGHT_FREQ_SET register (DPCD Address 00728h) + * - Where P = 2^Pn, where Pn is the value programmed by field 4:0 of the + * EDP_PWMGEN_BIT_COUNT register (DPCD Address 00724h) + */ +static bool intel_dp_aux_set_pwm_freq(struct intel_connector *connector) +{ + struct drm_i915_private *dev_priv = to_i915(connector->base.dev); + struct intel_dp *intel_dp = enc_to_intel_dp(&connector->encoder->base); + int freq, fxp, fxp_min, fxp_max, fxp_actual, f = 1; + u8 pn, pn_min, pn_max; + + /* Find desired value of (F x P) + * Note that, if F x P is out of supported range, the maximum value or + * minimum value will applied automatically. So no need to check that. + */ + freq = dev_priv->vbt.backlight.pwm_freq_hz; + DRM_DEBUG_KMS("VBT defined backlight frequency %u Hz\n", freq); + if (!freq) { + DRM_DEBUG_KMS("Use panel default backlight frequency\n"); + return false; + } + + fxp = DIV_ROUND_CLOSEST(KHz(DP_EDP_BACKLIGHT_FREQ_BASE_KHZ), freq); + + /* Use highest possible value of Pn for more granularity of brightness + * adjustment while satifying the conditions below. + * - Pn is in the range of Pn_min and Pn_max + * - F is in the range of 1 and 255 + * - FxP is within 25% of desired value. + * Note: 25% is arbitrary value and may need some tweak. + */ + if (drm_dp_dpcd_readb(&intel_dp->aux, + DP_EDP_PWMGEN_BIT_COUNT_CAP_MIN, &pn_min) != 1) { + DRM_DEBUG_KMS("Failed to read pwmgen bit count cap min\n"); + return false; + } + if (drm_dp_dpcd_readb(&intel_dp->aux, + DP_EDP_PWMGEN_BIT_COUNT_CAP_MAX, &pn_max) != 1) { + DRM_DEBUG_KMS("Failed to read pwmgen bit count cap max\n"); + return false; + } + pn_min &= DP_EDP_PWMGEN_BIT_COUNT_MASK; + pn_max &= DP_EDP_PWMGEN_BIT_COUNT_MASK; + + fxp_min = DIV_ROUND_CLOSEST(fxp * 3, 4); + fxp_max = DIV_ROUND_CLOSEST(fxp * 5, 4); + if (fxp_min < (1 << pn_min) || (255 << pn_max) < fxp_max) { + DRM_DEBUG_KMS("VBT defined backlight frequency out of range\n"); + return false; + } + + for (pn = pn_max; pn >= pn_min; pn--) { + f = clamp(DIV_ROUND_CLOSEST(fxp, 1 << pn), 1, 255); + fxp_actual = f << pn; + if (fxp_min <= fxp_actual && fxp_actual <= fxp_max) + break; + } + + if (drm_dp_dpcd_writeb(&intel_dp->aux, + DP_EDP_PWMGEN_BIT_COUNT, pn) < 0) { + DRM_DEBUG_KMS("Failed to write aux pwmgen bit count\n"); + return false; + } + if (drm_dp_dpcd_writeb(&intel_dp->aux, + DP_EDP_BACKLIGHT_FREQ_SET, (u8) f) < 0) { + DRM_DEBUG_KMS("Failed to write aux backlight freq\n"); + return false; + } + return true; +} + +/* +* Set minimum / maximum dynamic brightness percentage. This value is expressed +* as the percentage of normal brightness in 5% increments. +*/ +static bool +intel_dp_aux_set_dynamic_backlight_percent(struct intel_dp *intel_dp, + u32 min, u32 max) +{ + u8 dbc[] = { DIV_ROUND_CLOSEST(min, 5), DIV_ROUND_CLOSEST(max, 5) }; + + if (drm_dp_dpcd_write(&intel_dp->aux, DP_EDP_DBC_MINIMUM_BRIGHTNESS_SET, + dbc, sizeof(dbc)) < 0) { + DRM_DEBUG_KMS("Failed to write aux DBC brightness level\n"); + return false; + } + return true; +} + static void intel_dp_aux_enable_backlight(const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state) { struct intel_connector *connector = to_intel_connector(conn_state->connector); struct intel_dp *intel_dp = enc_to_intel_dp(&connector->encoder->base); - uint8_t dpcd_buf = 0; - uint8_t edp_backlight_mode = 0; + uint8_t dpcd_buf, new_dpcd_buf, edp_backlight_mode; if (drm_dp_dpcd_readb(&intel_dp->aux, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, &dpcd_buf) != 1) { @@ -113,18 +205,15 @@ static void intel_dp_aux_enable_backlight(const struct intel_crtc_state *crtc_st return; } + new_dpcd_buf = dpcd_buf; edp_backlight_mode = dpcd_buf & DP_EDP_BACKLIGHT_CONTROL_MODE_MASK; switch (edp_backlight_mode) { case DP_EDP_BACKLIGHT_CONTROL_MODE_PWM: case DP_EDP_BACKLIGHT_CONTROL_MODE_PRESET: case DP_EDP_BACKLIGHT_CONTROL_MODE_PRODUCT: - dpcd_buf &= ~DP_EDP_BACKLIGHT_CONTROL_MODE_MASK; - dpcd_buf |= DP_EDP_BACKLIGHT_CONTROL_MODE_DPCD; - if (drm_dp_dpcd_writeb(&intel_dp->aux, - DP_EDP_BACKLIGHT_MODE_SET_REGISTER, dpcd_buf) < 0) { - DRM_DEBUG_KMS("Failed to write aux backlight mode\n"); - } + new_dpcd_buf &= ~DP_EDP_BACKLIGHT_CONTROL_MODE_MASK; + new_dpcd_buf |= DP_EDP_BACKLIGHT_CONTROL_MODE_DPCD; break; /* Do nothing when it is already DPCD mode */ @@ -133,6 +222,25 @@ static void intel_dp_aux_enable_backlight(const struct intel_crtc_state *crtc_st break; } + if (intel_dp->edp_dpcd[2] & DP_EDP_BACKLIGHT_FREQ_AUX_SET_CAP) + if (intel_dp_aux_set_pwm_freq(connector)) + new_dpcd_buf |= DP_EDP_BACKLIGHT_FREQ_AUX_SET_ENABLE; + + if (i915.enable_dbc && + (intel_dp->edp_dpcd[2] & DP_EDP_DYNAMIC_BACKLIGHT_CAP)) { + if(intel_dp_aux_set_dynamic_backlight_percent(intel_dp, 0, 100)) { + new_dpcd_buf |= DP_EDP_DYNAMIC_BACKLIGHT_ENABLE; + DRM_DEBUG_KMS("Enable dynamic brightness.\n"); + } + } + + if (new_dpcd_buf != dpcd_buf) { + if (drm_dp_dpcd_writeb(&intel_dp->aux, + DP_EDP_BACKLIGHT_MODE_SET_REGISTER, new_dpcd_buf) < 0) { + DRM_DEBUG_KMS("Failed to write aux backlight mode\n"); + } + } + set_aux_backlight_enable(intel_dp, true); intel_dp_aux_set_backlight(conn_state, connector->panel.backlight.level); } @@ -169,15 +277,66 @@ intel_dp_aux_display_control_capable(struct intel_connector *connector) /* Check the eDP Display control capabilities registers to determine if * the panel can support backlight control over the aux channel */ - if (intel_dp->edp_dpcd[1] & DP_EDP_TCON_BACKLIGHT_ADJUSTMENT_CAP && - (intel_dp->edp_dpcd[2] & DP_EDP_BACKLIGHT_BRIGHTNESS_AUX_SET_CAP) && - !(intel_dp->edp_dpcd[2] & DP_EDP_BACKLIGHT_BRIGHTNESS_PWM_PIN_CAP)) { + if ((intel_dp->edp_dpcd[1] & DP_EDP_TCON_BACKLIGHT_ADJUSTMENT_CAP) && + (intel_dp->edp_dpcd[2] & DP_EDP_BACKLIGHT_BRIGHTNESS_AUX_SET_CAP)) { DRM_DEBUG_KMS("AUX Backlight Control Supported!\n"); return true; } return false; } +/* + * Heuristic function whether we should use AUX for backlight adjustment or not. + * + * We should use AUX for backlight brightness adjustment if panel doesn't this + * via PWM pin or using AUX is better than using PWM pin. + * + * The heuristic to determine that using AUX pin is better than using PWM pin is + * that the panel support any of the feature list here. + * - Regional backlight brightness adjustment + * - Backlight PWM frequency set + * - More than 8 bits resolution of brightness level + * - Backlight enablement via AUX and not by BL_ENABLE pin + * + * If all above are not true, assume that using PWM pin is better. + */ +static bool +intel_dp_aux_display_control_heuristic(struct intel_connector *connector) +{ + struct intel_dp *intel_dp = enc_to_intel_dp(&connector->encoder->base); + uint8_t reg_val; + + /* Panel doesn't support adjusting backlight brightness via PWN pin */ + if (!(intel_dp->edp_dpcd[2] & DP_EDP_BACKLIGHT_BRIGHTNESS_PWM_PIN_CAP)) + return true; + + /* Panel supports regional backlight brightness adjustment */ + if (drm_dp_dpcd_readb(&intel_dp->aux, DP_EDP_GENERAL_CAP_3, + ®_val) != 1) { + DRM_DEBUG_KMS("Failed to read DPCD register 0x%x\n", + DP_EDP_GENERAL_CAP_3); + return false; + } + if (reg_val > 0) + return true; + + /* Panel supports backlight PWM frequency set */ + if (intel_dp->edp_dpcd[2] & DP_EDP_BACKLIGHT_FREQ_AUX_SET_CAP) + return true; + + /* Panel supports more than 8 bits resolution of brightness level */ + if (intel_dp->edp_dpcd[2] & DP_EDP_BACKLIGHT_BRIGHTNESS_BYTE_COUNT) + return true; + + /* Panel supports enabling backlight via AUX but not by BL_ENABLE pin */ + if ((intel_dp->edp_dpcd[1] & DP_EDP_BACKLIGHT_AUX_ENABLE_CAP) && + !(intel_dp->edp_dpcd[1] & DP_EDP_BACKLIGHT_PIN_ENABLE_CAP)) + return true; + + return false; + +} + int intel_dp_aux_init_backlight_funcs(struct intel_connector *intel_connector) { struct intel_panel *panel = &intel_connector->panel; @@ -188,6 +347,10 @@ int intel_dp_aux_init_backlight_funcs(struct intel_connector *intel_connector) if (!intel_dp_aux_display_control_capable(intel_connector)) return -ENODEV; + if (i915.enable_dpcd_backlight == -1 && + !intel_dp_aux_display_control_heuristic(intel_connector)) + return -ENODEV; + panel->backlight.setup = intel_dp_aux_setup_backlight; panel->backlight.enable = intel_dp_aux_enable_backlight; panel->backlight.disable = intel_dp_aux_disable_backlight; diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index d93efb49a2e2..d17a32437f07 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -1858,9 +1858,8 @@ void intel_suspend_gt_powersave(struct drm_i915_private *dev_priv); void gen6_rps_busy(struct drm_i915_private *dev_priv); void gen6_rps_reset_ei(struct drm_i915_private *dev_priv); void gen6_rps_idle(struct drm_i915_private *dev_priv); -void gen6_rps_boost(struct drm_i915_private *dev_priv, - struct intel_rps_client *rps, - unsigned long submitted); +void gen6_rps_boost(struct drm_i915_gem_request *rq, + struct intel_rps_client *rps); void intel_queue_rps_boost_for_request(struct drm_i915_gem_request *req); void g4x_wm_get_hw_state(struct drm_device *dev); void vlv_wm_get_hw_state(struct drm_device *dev); diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c index 5b4de719bec3..24db316e0fd1 100644 --- a/drivers/gpu/drm/i915/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/intel_engine_cs.c @@ -149,6 +149,7 @@ __intel_engine_context_size(struct drm_i915_private *dev_priv, u8 class) switch (INTEL_GEN(dev_priv)) { default: MISSING_CASE(INTEL_GEN(dev_priv)); + case 10: case 9: return GEN9_LR_CONTEXT_RENDER_SIZE; case 8: @@ -291,11 +292,9 @@ cleanup: */ int intel_engines_init(struct drm_i915_private *dev_priv) { - struct intel_device_info *device_info = mkwrite_device_info(dev_priv); struct intel_engine_cs *engine; enum intel_engine_id id, err_id; - unsigned int mask = 0; - int err = 0; + int err; for_each_engine(engine, dev_priv, id) { const struct engine_class_info *class_info = @@ -306,40 +305,30 @@ int intel_engines_init(struct drm_i915_private *dev_priv) init = class_info->init_execlists; else init = class_info->init_legacy; - if (!init) { - kfree(engine); - dev_priv->engine[id] = NULL; - continue; - } + + err = -EINVAL; + err_id = id; + + if (GEM_WARN_ON(!init)) + goto cleanup; err = init(engine); - if (err) { - err_id = id; + if (err) goto cleanup; - } GEM_BUG_ON(!engine->submit_request); - mask |= ENGINE_MASK(id); } - /* - * Catch failures to update intel_engines table when the new engines - * are added to the driver by a warning and disabling the forgotten - * engines. - */ - if (WARN_ON(mask != INTEL_INFO(dev_priv)->ring_mask)) - device_info->ring_mask = mask; - - device_info->num_rings = hweight32(mask); - return 0; cleanup: for_each_engine(engine, dev_priv, id) { - if (id >= err_id) + if (id >= err_id) { kfree(engine); - else + dev_priv->engine[id] = NULL; + } else { dev_priv->gt.cleanup_engine(engine); + } } return err; } @@ -1340,6 +1329,7 @@ void intel_engines_mark_idle(struct drm_i915_private *i915) for_each_engine(engine, i915, id) { intel_engine_disarm_breadcrumbs(engine); i915_gem_batch_pool_fini(&engine->batch_pool); + tasklet_kill(&engine->irq_tasklet); engine->no_priolist = false; } } diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c index 0c4cde6b2e6f..ee1a5b937590 100644 --- a/drivers/gpu/drm/i915/intel_fbdev.c +++ b/drivers/gpu/drm/i915/intel_fbdev.c @@ -813,7 +813,7 @@ void intel_fbdev_output_poll_changed(struct drm_device *dev) { struct intel_fbdev *ifbdev = to_i915(dev)->fbdev; - if (ifbdev && ifbdev->vma) + if (ifbdev) drm_fb_helper_hotplug_event(&ifbdev->helper); } diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 7404cf2aac28..699868d81de8 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -2071,7 +2071,7 @@ void intel_lr_context_resume(struct drm_i915_private *dev_priv) * So to avoid that we reset the context images upon resume. For * simplicity, we just zero everything out. */ - list_for_each_entry(ctx, &dev_priv->context_list, link) { + list_for_each_entry(ctx, &dev_priv->contexts.list, link) { for_each_engine(engine, dev_priv, id) { struct intel_context *ce = &ctx->engine[engine->id]; u32 *reg; diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 48ea0fca1f72..ee2a349cfe68 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -3837,7 +3837,7 @@ skl_plane_downscale_amount(const struct intel_crtc_state *cstate, uint_fixed_16_16_t downscale_h, downscale_w; if (WARN_ON(!intel_wm_plane_visible(cstate, pstate))) - return u32_to_fixed_16_16(0); + return u32_to_fixed16(0); /* n.b., src is 16.16 fixed point, dst is whole integer */ if (plane->id == PLANE_CURSOR) { @@ -3861,10 +3861,10 @@ skl_plane_downscale_amount(const struct intel_crtc_state *cstate, dst_h = drm_rect_height(&pstate->base.dst); } - fp_w_ratio = fixed_16_16_div(src_w, dst_w); - fp_h_ratio = fixed_16_16_div(src_h, dst_h); - downscale_w = max_fixed_16_16(fp_w_ratio, u32_to_fixed_16_16(1)); - downscale_h = max_fixed_16_16(fp_h_ratio, u32_to_fixed_16_16(1)); + fp_w_ratio = div_fixed16(src_w, dst_w); + fp_h_ratio = div_fixed16(src_h, dst_h); + downscale_w = max_fixed16(fp_w_ratio, u32_to_fixed16(1)); + downscale_h = max_fixed16(fp_h_ratio, u32_to_fixed16(1)); return mul_fixed16(downscale_w, downscale_h); } @@ -3872,7 +3872,7 @@ skl_plane_downscale_amount(const struct intel_crtc_state *cstate, static uint_fixed_16_16_t skl_pipe_downscale_amount(const struct intel_crtc_state *crtc_state) { - uint_fixed_16_16_t pipe_downscale = u32_to_fixed_16_16(1); + uint_fixed_16_16_t pipe_downscale = u32_to_fixed16(1); if (!crtc_state->base.enable) return pipe_downscale; @@ -3891,10 +3891,10 @@ skl_pipe_downscale_amount(const struct intel_crtc_state *crtc_state) if (!dst_w || !dst_h) return pipe_downscale; - fp_w_ratio = fixed_16_16_div(src_w, dst_w); - fp_h_ratio = fixed_16_16_div(src_h, dst_h); - downscale_w = max_fixed_16_16(fp_w_ratio, u32_to_fixed_16_16(1)); - downscale_h = max_fixed_16_16(fp_h_ratio, u32_to_fixed_16_16(1)); + fp_w_ratio = div_fixed16(src_w, dst_w); + fp_h_ratio = div_fixed16(src_h, dst_h); + downscale_w = max_fixed16(fp_w_ratio, u32_to_fixed16(1)); + downscale_h = max_fixed16(fp_h_ratio, u32_to_fixed16(1)); pipe_downscale = mul_fixed16(downscale_w, downscale_h); } @@ -3913,14 +3913,14 @@ int skl_check_pipe_max_pixel_rate(struct intel_crtc *intel_crtc, int crtc_clock, dotclk; uint32_t pipe_max_pixel_rate; uint_fixed_16_16_t pipe_downscale; - uint_fixed_16_16_t max_downscale = u32_to_fixed_16_16(1); + uint_fixed_16_16_t max_downscale = u32_to_fixed16(1); if (!cstate->base.enable) return 0; drm_atomic_crtc_state_for_each_plane_state(plane, pstate, crtc_state) { uint_fixed_16_16_t plane_downscale; - uint_fixed_16_16_t fp_9_div_8 = fixed_16_16_div(9, 8); + uint_fixed_16_16_t fp_9_div_8 = div_fixed16(9, 8); int bpp; if (!intel_wm_plane_visible(cstate, @@ -3938,7 +3938,7 @@ int skl_check_pipe_max_pixel_rate(struct intel_crtc *intel_crtc, plane_downscale = mul_fixed16(plane_downscale, fp_9_div_8); - max_downscale = max_fixed_16_16(plane_downscale, max_downscale); + max_downscale = max_fixed16(plane_downscale, max_downscale); } pipe_downscale = skl_pipe_downscale_amount(cstate); @@ -4276,7 +4276,7 @@ static uint_fixed_16_16_t skl_wm_method1(uint32_t pixel_rate, uint8_t cpp, return FP_16_16_MAX; wm_intermediate_val = latency * pixel_rate * cpp; - ret = fixed_16_16_div_u64(wm_intermediate_val, 1000 * 512); + ret = div_fixed16(wm_intermediate_val, 1000 * 512); return ret; } @@ -4294,7 +4294,7 @@ static uint_fixed_16_16_t skl_wm_method2(uint32_t pixel_rate, wm_intermediate_val = latency * pixel_rate; wm_intermediate_val = DIV_ROUND_UP(wm_intermediate_val, pipe_htotal * 1000); - ret = mul_u32_fixed_16_16(wm_intermediate_val, plane_blocks_per_line); + ret = mul_u32_fixed16(wm_intermediate_val, plane_blocks_per_line); return ret; } @@ -4306,15 +4306,15 @@ intel_get_linetime_us(struct intel_crtc_state *cstate) uint_fixed_16_16_t linetime_us; if (!cstate->base.active) - return u32_to_fixed_16_16(0); + return u32_to_fixed16(0); pixel_rate = cstate->pixel_rate; if (WARN_ON(pixel_rate == 0)) - return u32_to_fixed_16_16(0); + return u32_to_fixed16(0); crtc_htotal = cstate->base.adjusted_mode.crtc_htotal; - linetime_us = fixed_16_16_div_u64(crtc_htotal * 1000, pixel_rate); + linetime_us = div_fixed16(crtc_htotal * 1000, pixel_rate); return linetime_us; } @@ -4361,7 +4361,7 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, uint32_t plane_bytes_per_line; uint32_t res_blocks, res_lines; uint8_t cpp; - uint32_t width = 0, height = 0; + uint32_t width = 0; uint32_t plane_pixel_rate; uint_fixed_16_16_t y_tile_minimum; uint32_t y_min_scanlines; @@ -4390,7 +4390,6 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, if (plane->id == PLANE_CURSOR) { width = intel_pstate->base.crtc_w; - height = intel_pstate->base.crtc_h; } else { /* * Src coordinates are already rotated by 270 degrees for @@ -4398,16 +4397,13 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, * GTT mapping), hence no need to account for rotation here. */ width = drm_rect_width(&intel_pstate->base.src) >> 16; - height = drm_rect_height(&intel_pstate->base.src) >> 16; } - cpp = fb->format->cpp[0]; + cpp = (fb->format->format == DRM_FORMAT_NV12) ? fb->format->cpp[1] : + fb->format->cpp[0]; plane_pixel_rate = skl_adjusted_plane_pixel_rate(cstate, intel_pstate); if (drm_rotation_90_or_270(pstate->rotation)) { - int cpp = (fb->format->format == DRM_FORMAT_NV12) ? - fb->format->cpp[1] : - fb->format->cpp[0]; switch (cpp) { case 1: @@ -4434,14 +4430,14 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, if (y_tiled) { interm_pbpl = DIV_ROUND_UP(plane_bytes_per_line * y_min_scanlines, 512); - plane_blocks_per_line = fixed_16_16_div(interm_pbpl, + plane_blocks_per_line = div_fixed16(interm_pbpl, y_min_scanlines); } else if (x_tiled) { interm_pbpl = DIV_ROUND_UP(plane_bytes_per_line, 512); - plane_blocks_per_line = u32_to_fixed_16_16(interm_pbpl); + plane_blocks_per_line = u32_to_fixed16(interm_pbpl); } else { interm_pbpl = DIV_ROUND_UP(plane_bytes_per_line, 512) + 1; - plane_blocks_per_line = u32_to_fixed_16_16(interm_pbpl); + plane_blocks_per_line = u32_to_fixed16(interm_pbpl); } method1 = skl_wm_method1(plane_pixel_rate, cpp, latency); @@ -4450,35 +4446,35 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, latency, plane_blocks_per_line); - y_tile_minimum = mul_u32_fixed_16_16(y_min_scanlines, - plane_blocks_per_line); + y_tile_minimum = mul_u32_fixed16(y_min_scanlines, + plane_blocks_per_line); if (y_tiled) { - selected_result = max_fixed_16_16(method2, y_tile_minimum); + selected_result = max_fixed16(method2, y_tile_minimum); } else { uint32_t linetime_us; - linetime_us = fixed_16_16_to_u32_round_up( + linetime_us = fixed16_to_u32_round_up( intel_get_linetime_us(cstate)); if ((cpp * cstate->base.adjusted_mode.crtc_htotal / 512 < 1) && (plane_bytes_per_line / 512 < 1)) selected_result = method2; else if ((ddb_allocation && ddb_allocation / - fixed_16_16_to_u32_round_up(plane_blocks_per_line)) >= 1) - selected_result = min_fixed_16_16(method1, method2); + fixed16_to_u32_round_up(plane_blocks_per_line)) >= 1) + selected_result = min_fixed16(method1, method2); else if (latency >= linetime_us) - selected_result = min_fixed_16_16(method1, method2); + selected_result = min_fixed16(method1, method2); else selected_result = method1; } - res_blocks = fixed_16_16_to_u32_round_up(selected_result) + 1; + res_blocks = fixed16_to_u32_round_up(selected_result) + 1; res_lines = div_round_up_fixed16(selected_result, plane_blocks_per_line); if (level >= 1 && level <= 7) { if (y_tiled) { - res_blocks += fixed_16_16_to_u32_round_up(y_tile_minimum); + res_blocks += fixed16_to_u32_round_up(y_tile_minimum); res_lines += y_min_scanlines; } else { res_blocks++; @@ -4563,8 +4559,7 @@ skl_compute_linetime_wm(struct intel_crtc_state *cstate) if (is_fixed16_zero(linetime_us)) return 0; - linetime_wm = fixed_16_16_to_u32_round_up(mul_u32_fixed_16_16(8, - linetime_us)); + linetime_wm = fixed16_to_u32_round_up(mul_u32_fixed16(8, linetime_us)); /* Display WA #1135: bxt. */ if (IS_BROXTON(dev_priv) && dev_priv->ipc_enabled) @@ -5852,7 +5847,7 @@ static u32 intel_rps_limits(struct drm_i915_private *dev_priv, u8 val) * the hw runs at the minimal clock before selecting the desired * frequency, if the down threshold expires in that window we will not * receive a down interrupt. */ - if (IS_GEN9(dev_priv)) { + if (INTEL_GEN(dev_priv) >= 9) { limits = (dev_priv->rps.max_freq_softlimit) << 23; if (val <= dev_priv->rps.min_freq_softlimit) limits |= (dev_priv->rps.min_freq_softlimit) << 14; @@ -5994,7 +5989,7 @@ static int gen6_set_rps(struct drm_i915_private *dev_priv, u8 val) if (val != dev_priv->rps.cur_freq) { gen6_set_rps_thresholds(dev_priv, val); - if (IS_GEN9(dev_priv)) + if (INTEL_GEN(dev_priv) >= 9) I915_WRITE(GEN6_RPNSWREQ, GEN9_FREQUENCY(val)); else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) @@ -6126,47 +6121,35 @@ void gen6_rps_idle(struct drm_i915_private *dev_priv) gen6_sanitize_rps_pm_mask(dev_priv, ~0)); } mutex_unlock(&dev_priv->rps.hw_lock); - - spin_lock(&dev_priv->rps.client_lock); - while (!list_empty(&dev_priv->rps.clients)) - list_del_init(dev_priv->rps.clients.next); - spin_unlock(&dev_priv->rps.client_lock); } -void gen6_rps_boost(struct drm_i915_private *dev_priv, - struct intel_rps_client *rps, - unsigned long submitted) +void gen6_rps_boost(struct drm_i915_gem_request *rq, + struct intel_rps_client *rps) { + struct drm_i915_private *i915 = rq->i915; + bool boost; + /* This is intentionally racy! We peek at the state here, then * validate inside the RPS worker. */ - if (!(dev_priv->gt.awake && - dev_priv->rps.enabled && - dev_priv->rps.cur_freq < dev_priv->rps.boost_freq)) + if (!i915->rps.enabled) return; - /* Force a RPS boost (and don't count it against the client) if - * the GPU is severely congested. - */ - if (rps && time_after(jiffies, submitted + DRM_I915_THROTTLE_JIFFIES)) - rps = NULL; - - spin_lock(&dev_priv->rps.client_lock); - if (rps == NULL || list_empty(&rps->link)) { - spin_lock_irq(&dev_priv->irq_lock); - if (dev_priv->rps.interrupts_enabled) { - dev_priv->rps.client_boost = true; - schedule_work(&dev_priv->rps.work); - } - spin_unlock_irq(&dev_priv->irq_lock); - - if (rps != NULL) { - list_add(&rps->link, &dev_priv->rps.clients); - rps->boosts++; - } else - dev_priv->rps.boosts++; + boost = false; + spin_lock_irq(&rq->lock); + if (!rq->waitboost && !i915_gem_request_completed(rq)) { + atomic_inc(&i915->rps.num_waiters); + rq->waitboost = true; + boost = true; } - spin_unlock(&dev_priv->rps.client_lock); + spin_unlock_irq(&rq->lock); + if (!boost) + return; + + if (READ_ONCE(i915->rps.cur_freq) < i915->rps.boost_freq) + schedule_work(&i915->rps.work); + + atomic_inc(rps ? &rps->boosts : &i915->rps.boosts); } int intel_set_rps(struct drm_i915_private *dev_priv, u8 val) @@ -6365,7 +6348,7 @@ static void gen6_init_rps_frequencies(struct drm_i915_private *dev_priv) dev_priv->rps.efficient_freq = dev_priv->rps.rp1_freq; if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv) || - IS_GEN9_BC(dev_priv)) { + IS_GEN9_BC(dev_priv) || IS_CANNONLAKE(dev_priv)) { u32 ddcc_status = 0; if (sandybridge_pcode_read(dev_priv, @@ -6378,7 +6361,7 @@ static void gen6_init_rps_frequencies(struct drm_i915_private *dev_priv) dev_priv->rps.max_freq); } - if (IS_GEN9_BC(dev_priv)) { + if (IS_GEN9_BC(dev_priv) || IS_CANNONLAKE(dev_priv)) { /* Store the frequency values in 16.66 MHZ units, which is * the natural hardware unit for SKL */ @@ -6684,7 +6667,7 @@ static void gen6_update_ring_freq(struct drm_i915_private *dev_priv) /* convert DDR frequency from units of 266.6MHz to bandwidth */ min_ring_freq = mult_frac(min_ring_freq, 8, 3); - if (IS_GEN9_BC(dev_priv)) { + if (IS_GEN9_BC(dev_priv) || IS_CANNONLAKE(dev_priv)) { /* Convert GT frequency to 50 HZ units */ min_gpu_freq = dev_priv->rps.min_freq / GEN9_FREQ_SCALER; max_gpu_freq = dev_priv->rps.max_freq / GEN9_FREQ_SCALER; @@ -6702,7 +6685,7 @@ static void gen6_update_ring_freq(struct drm_i915_private *dev_priv) int diff = max_gpu_freq - gpu_freq; unsigned int ia_freq = 0, ring_freq = 0; - if (IS_GEN9_BC(dev_priv)) { + if (IS_GEN9_BC(dev_priv) || IS_CANNONLAKE(dev_priv)) { /* * ring_freq = 2 * GT. ring_freq is in 100MHz units * No floor required for ring frequency on SKL. @@ -7833,7 +7816,7 @@ void intel_enable_gt_powersave(struct drm_i915_private *dev_priv) } else if (INTEL_GEN(dev_priv) >= 9) { gen9_enable_rc6(dev_priv); gen9_enable_rps(dev_priv); - if (IS_GEN9_BC(dev_priv)) + if (IS_GEN9_BC(dev_priv) || IS_CANNONLAKE(dev_priv)) gen6_update_ring_freq(dev_priv); } else if (IS_BROADWELL(dev_priv)) { gen8_enable_rps(dev_priv); @@ -9078,7 +9061,7 @@ static int chv_freq_opcode(struct drm_i915_private *dev_priv, int val) int intel_gpu_freq(struct drm_i915_private *dev_priv, int val) { - if (IS_GEN9(dev_priv)) + if (INTEL_GEN(dev_priv) >= 9) return DIV_ROUND_CLOSEST(val * GT_FREQUENCY_MULTIPLIER, GEN9_FREQ_SCALER); else if (IS_CHERRYVIEW(dev_priv)) @@ -9091,7 +9074,7 @@ int intel_gpu_freq(struct drm_i915_private *dev_priv, int val) int intel_freq_opcode(struct drm_i915_private *dev_priv, int val) { - if (IS_GEN9(dev_priv)) + if (INTEL_GEN(dev_priv) >= 9) return DIV_ROUND_CLOSEST(val * GEN9_FREQ_SCALER, GT_FREQUENCY_MULTIPLIER); else if (IS_CHERRYVIEW(dev_priv)) @@ -9113,7 +9096,7 @@ static void __intel_rps_boost_work(struct work_struct *work) struct drm_i915_gem_request *req = boost->req; if (!i915_gem_request_completed(req)) - gen6_rps_boost(req->i915, NULL, req->emitted_jiffies); + gen6_rps_boost(req, NULL); i915_gem_request_put(req); kfree(boost); @@ -9142,11 +9125,10 @@ void intel_queue_rps_boost_for_request(struct drm_i915_gem_request *req) void intel_pm_setup(struct drm_i915_private *dev_priv) { mutex_init(&dev_priv->rps.hw_lock); - spin_lock_init(&dev_priv->rps.client_lock); INIT_DELAYED_WORK(&dev_priv->rps.autoenable_work, __intel_autoenable_gt_powersave); - INIT_LIST_HEAD(&dev_priv->rps.clients); + atomic_set(&dev_priv->rps.num_waiters, 0); dev_priv->pm.suspended = false; atomic_set(&dev_priv->pm.wakeref_count, 0); diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index acd1da9b62a3..5224b7abb8a3 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -2140,7 +2140,7 @@ static void intel_ring_default_vfuncs(struct drm_i915_private *dev_priv, engine->emit_breadcrumb = gen6_sema_emit_breadcrumb; - num_rings = hweight32(INTEL_INFO(dev_priv)->ring_mask) - 1; + num_rings = INTEL_INFO(dev_priv)->num_rings - 1; if (INTEL_GEN(dev_priv) >= 8) { engine->emit_breadcrumb_sz += num_rings * 6; } else { @@ -2184,8 +2184,7 @@ int intel_init_render_ring_buffer(struct intel_engine_cs *engine) engine->semaphore.signal = gen8_rcs_signal; - num_rings = - hweight32(INTEL_INFO(dev_priv)->ring_mask) - 1; + num_rings = INTEL_INFO(dev_priv)->num_rings - 1; engine->emit_breadcrumb_sz += num_rings * 8; } } else if (INTEL_GEN(dev_priv) >= 6) { diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 6aa20ac8cde3..d33c93444c0d 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -121,6 +121,7 @@ struct intel_engine_hangcheck { unsigned long action_timestamp; int deadlock; struct intel_instdone instdone; + struct drm_i915_gem_request *active_request; bool stalled; }; diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index efe80ed5fd4d..f630d632a976 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -341,6 +341,59 @@ static void skl_power_well_pre_disable(struct drm_i915_private *dev_priv, 1 << PIPE_C | 1 << PIPE_B); } +static void gen9_wait_for_power_well_enable(struct drm_i915_private *dev_priv, + struct i915_power_well *power_well) +{ + int id = power_well->id; + + /* Timeout for PW1:10 us, AUX:not specified, other PWs:20 us. */ + WARN_ON(intel_wait_for_register(dev_priv, + HSW_PWR_WELL_DRIVER, + SKL_POWER_WELL_STATE(id), + SKL_POWER_WELL_STATE(id), + 1)); +} + +static u32 gen9_power_well_requesters(struct drm_i915_private *dev_priv, int id) +{ + u32 req_mask = SKL_POWER_WELL_REQ(id); + u32 ret; + + ret = I915_READ(HSW_PWR_WELL_BIOS) & req_mask ? 1 : 0; + ret |= I915_READ(HSW_PWR_WELL_DRIVER) & req_mask ? 2 : 0; + ret |= I915_READ(HSW_PWR_WELL_KVMR) & req_mask ? 4 : 0; + ret |= I915_READ(HSW_PWR_WELL_DEBUG) & req_mask ? 8 : 0; + + return ret; +} + +static void gen9_wait_for_power_well_disable(struct drm_i915_private *dev_priv, + struct i915_power_well *power_well) +{ + int id = power_well->id; + bool disabled; + u32 reqs; + + /* + * Bspec doesn't require waiting for PWs to get disabled, but still do + * this for paranoia. The known cases where a PW will be forced on: + * - a KVMR request on any power well via the KVMR request register + * - a DMC request on PW1 and MISC_IO power wells via the BIOS and + * DEBUG request registers + * Skip the wait in case any of the request bits are set and print a + * diagnostic message. + */ + wait_for((disabled = !(I915_READ(HSW_PWR_WELL_DRIVER) & + SKL_POWER_WELL_STATE(id))) || + (reqs = gen9_power_well_requesters(dev_priv, id)), 1); + if (disabled) + return; + + DRM_DEBUG_KMS("%s forced on (bios:%d driver:%d kvmr:%d debug:%d)\n", + power_well->name, + !!(reqs & 1), !!(reqs & 2), !!(reqs & 4), !!(reqs & 8)); +} + static void hsw_set_power_well(struct drm_i915_private *dev_priv, struct i915_power_well *power_well, bool enable) { @@ -549,7 +602,9 @@ static void assert_can_enable_dc9(struct drm_i915_private *dev_priv) "DC9 already programmed to be enabled.\n"); WARN_ONCE(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5, "DC5 still not disabled to enable DC9.\n"); - WARN_ONCE(I915_READ(HSW_PWR_WELL_DRIVER), "Power well on.\n"); + WARN_ONCE(I915_READ(HSW_PWR_WELL_DRIVER) & + SKL_POWER_WELL_REQ(SKL_DISP_PW_2), + "Power well 2 on.\n"); WARN_ONCE(intel_irqs_enabled(dev_priv), "Interrupts not disabled yet.\n"); @@ -744,45 +799,6 @@ void skl_disable_dc6(struct drm_i915_private *dev_priv) gen9_set_dc_state(dev_priv, DC_STATE_DISABLE); } -static void -gen9_sanitize_power_well_requests(struct drm_i915_private *dev_priv, - struct i915_power_well *power_well) -{ - enum skl_disp_power_wells power_well_id = power_well->id; - u32 val; - u32 mask; - - mask = SKL_POWER_WELL_REQ(power_well_id); - - val = I915_READ(HSW_PWR_WELL_KVMR); - if (WARN_ONCE(val & mask, "Clearing unexpected KVMR request for %s\n", - power_well->name)) - I915_WRITE(HSW_PWR_WELL_KVMR, val & ~mask); - - val = I915_READ(HSW_PWR_WELL_BIOS); - val |= I915_READ(HSW_PWR_WELL_DEBUG); - - if (!(val & mask)) - return; - - /* - * DMC is known to force on the request bits for power well 1 on SKL - * and BXT and the misc IO power well on SKL but we don't expect any - * other request bits to be set, so WARN for those. - */ - if (power_well_id == SKL_DISP_PW_1 || - (IS_GEN9_BC(dev_priv) && - power_well_id == SKL_DISP_PW_MISC_IO)) - DRM_DEBUG_DRIVER("Clearing auxiliary requests for %s forced on " - "by DMC\n", power_well->name); - else - WARN_ONCE(1, "Clearing unexpected auxiliary requests for %s\n", - power_well->name); - - I915_WRITE(HSW_PWR_WELL_BIOS, val & ~mask); - I915_WRITE(HSW_PWR_WELL_DEBUG, val & ~mask); -} - static void skl_set_power_well(struct drm_i915_private *dev_priv, struct i915_power_well *power_well, bool enable) { @@ -846,6 +862,8 @@ static void skl_set_power_well(struct drm_i915_private *dev_priv, DRM_DEBUG_KMS("Enabling %s\n", power_well->name); check_fuse_status = true; } + + gen9_wait_for_power_well_enable(dev_priv, power_well); } else { if (enable_requested) { I915_WRITE(HSW_PWR_WELL_DRIVER, tmp & ~req_mask); @@ -853,14 +871,9 @@ static void skl_set_power_well(struct drm_i915_private *dev_priv, DRM_DEBUG_KMS("Disabling %s\n", power_well->name); } - gen9_sanitize_power_well_requests(dev_priv, power_well); + gen9_wait_for_power_well_disable(dev_priv, power_well); } - if (wait_for(!!(I915_READ(HSW_PWR_WELL_DRIVER) & state_mask) == enable, - 1)) - DRM_ERROR("%s %s timeout\n", - power_well->name, enable ? "enable" : "disable"); - if (check_fuse_status) { if (power_well->id == SKL_DISP_PW_1) { if (intel_wait_for_register(dev_priv, @@ -2479,7 +2492,7 @@ static uint32_t get_allowed_dc_mask(const struct drm_i915_private *dev_priv, int requested_dc; int max_dc; - if (IS_GEN9_BC(dev_priv)) { + if (IS_GEN9_BC(dev_priv) || IS_CANNONLAKE(dev_priv)) { max_dc = 2; mask = 0; } else if (IS_GEN9_LP(dev_priv)) { @@ -2694,13 +2707,18 @@ static void skl_display_core_uninit(struct drm_i915_private *dev_priv) mutex_lock(&power_domains->lock); - well = lookup_power_well(dev_priv, SKL_DISP_PW_MISC_IO); - intel_power_well_disable(dev_priv, well); - + /* + * BSpec says to keep the MISC IO power well enabled here, only + * remove our request for power well 1. + * Note that even though the driver's request is removed power well 1 + * may stay enabled after this due to DMC's own request on it. + */ well = lookup_power_well(dev_priv, SKL_DISP_PW_1); intel_power_well_disable(dev_priv, well); mutex_unlock(&power_domains->lock); + + usleep_range(10, 30); /* 10 us delay per Bspec */ } void bxt_display_core_init(struct drm_i915_private *dev_priv, @@ -2751,13 +2769,19 @@ void bxt_display_core_uninit(struct drm_i915_private *dev_priv) /* The spec doesn't call for removing the reset handshake flag */ - /* Disable PG1 */ + /* + * Disable PW1 (PG1). + * Note that even though the driver's request is removed power well 1 + * may stay enabled after this due to DMC's own request on it. + */ mutex_lock(&power_domains->lock); well = lookup_power_well(dev_priv, SKL_DISP_PW_1); intel_power_well_disable(dev_priv, well); mutex_unlock(&power_domains->lock); + + usleep_range(10, 30); /* 10 us delay per Bspec */ } #define CNL_PROCMON_IDX(val) \ @@ -2821,7 +2845,10 @@ static void cnl_display_core_init(struct drm_i915_private *dev_priv, bool resume val |= CL_POWER_DOWN_ENABLE; I915_WRITE(CNL_PORT_CL1CM_DW5, val); - /* 4. Enable Power Well 1 (PG1) and Aux IO Power */ + /* + * 4. Enable Power Well 1 (PG1). + * The AUX IO power wells will be enabled on demand. + */ mutex_lock(&power_domains->lock); well = lookup_power_well(dev_priv, SKL_DISP_PW_1); intel_power_well_enable(dev_priv, well); @@ -2853,12 +2880,18 @@ static void cnl_display_core_uninit(struct drm_i915_private *dev_priv) /* 3. Disable CD clock */ cnl_uninit_cdclk(dev_priv); - /* 4. Disable Power Well 1 (PG1) and Aux IO Power */ + /* + * 4. Disable Power Well 1 (PG1). + * The AUX IO power wells are toggled on demand, so they are already + * disabled at this point. + */ mutex_lock(&power_domains->lock); well = lookup_power_well(dev_priv, SKL_DISP_PW_1); intel_power_well_disable(dev_priv, well); mutex_unlock(&power_domains->lock); + usleep_range(10, 30); /* 10 us delay per Bspec */ + /* 5. Disable Comp */ val = I915_READ(CHICKEN_MISC_2); val |= COMP_PWR_DOWN; diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 3f8f30b412cd..f902922d4ae6 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -1343,7 +1343,7 @@ static void intel_sdvo_pre_enable(struct intel_encoder *intel_encoder, sdvox |= (9 << 19) | SDVO_BORDER_ENABLE; } - if (INTEL_PCH_TYPE(dev_priv) >= PCH_CPT) + if (HAS_PCH_CPT(dev_priv)) sdvox |= SDVO_PIPE_SEL_CPT(crtc->pipe); else sdvox |= SDVO_PIPE_SEL(crtc->pipe); diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c index 0c650c2cbca8..94f9a1332dbf 100644 --- a/drivers/gpu/drm/i915/intel_sprite.c +++ b/drivers/gpu/drm/i915/intel_sprite.c @@ -262,7 +262,7 @@ skl_update_plane(struct intel_plane *plane, spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); - if (IS_GEMINILAKE(dev_priv)) { + if (IS_GEMINILAKE(dev_priv) || IS_CANNONLAKE(dev_priv)) { I915_WRITE_FW(PLANE_COLOR_CTL(pipe, plane_id), PLANE_COLOR_PIPE_GAMMA_ENABLE | PLANE_COLOR_PIPE_CSC_ENABLE | diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 9882724bc2b6..deb4430541cf 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -643,7 +643,7 @@ find_fw_domain(struct drm_i915_private *dev_priv, u32 offset) { .start = (s), .end = (e), .domains = (d) } #define HAS_FWTABLE(dev_priv) \ - (IS_GEN9(dev_priv) || \ + (INTEL_GEN(dev_priv) >= 9 || \ IS_CHERRYVIEW(dev_priv) || \ IS_VALLEYVIEW(dev_priv)) @@ -1072,7 +1072,7 @@ static void intel_uncore_fw_domains_init(struct drm_i915_private *dev_priv) dev_priv->uncore.fw_clear = _MASKED_BIT_DISABLE(FORCEWAKE_KERNEL); } - if (IS_GEN9(dev_priv)) { + if (INTEL_GEN(dev_priv) >= 9) { dev_priv->uncore.funcs.force_wake_get = fw_domains_get; dev_priv->uncore.funcs.force_wake_put = fw_domains_put; fw_domain_init(dev_priv, FW_DOMAIN_ID_RENDER, @@ -1719,6 +1719,17 @@ bool intel_has_gpu_reset(struct drm_i915_private *dev_priv) return intel_get_gpu_reset(dev_priv) != NULL; } +/* + * When GuC submission is enabled, GuC manages ELSP and can initiate the + * engine reset too. For now, fall back to full GPU reset if it is enabled. + */ +bool intel_has_reset_engine(struct drm_i915_private *dev_priv) +{ + return (dev_priv->info.has_reset_engine && + !dev_priv->guc.execbuf_client && + i915.reset >= 2); +} + int intel_guc_reset(struct drm_i915_private *dev_priv) { int ret; diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c index 50710e3f1caa..6b132caffa18 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c @@ -197,6 +197,9 @@ static int lowlevel_hole(struct drm_i915_private *i915, { I915_RND_STATE(seed_prng); unsigned int size; + struct i915_vma mock_vma; + + memset(&mock_vma, 0, sizeof(struct i915_vma)); /* Keep creating larger objects until one cannot fit into the hole */ for (size = 12; (hole_end - hole_start) >> size; size++) { @@ -255,8 +258,11 @@ static int lowlevel_hole(struct drm_i915_private *i915, vm->allocate_va_range(vm, addr, BIT_ULL(size))) break; - vm->insert_entries(vm, obj->mm.pages, addr, - I915_CACHE_NONE, 0); + mock_vma.pages = obj->mm.pages; + mock_vma.node.size = BIT_ULL(size); + mock_vma.node.start = addr; + + vm->insert_entries(vm, &mock_vma, I915_CACHE_NONE, 0); } count = n; diff --git a/drivers/gpu/drm/i915/selftests/i915_vma.c b/drivers/gpu/drm/i915/selftests/i915_vma.c index fb9072d5877f..2e86ec136b35 100644 --- a/drivers/gpu/drm/i915/selftests/i915_vma.c +++ b/drivers/gpu/drm/i915/selftests/i915_vma.c @@ -186,16 +186,20 @@ static int igt_vma_create(void *arg) goto end; } - list_for_each_entry_safe(ctx, cn, &contexts, link) + list_for_each_entry_safe(ctx, cn, &contexts, link) { + list_del_init(&ctx->link); mock_context_close(ctx); + } } end: /* Final pass to lookup all created contexts */ err = create_vmas(i915, &objects, &contexts); out: - list_for_each_entry_safe(ctx, cn, &contexts, link) + list_for_each_entry_safe(ctx, cn, &contexts, link) { + list_del_init(&ctx->link); mock_context_close(ctx); + } list_for_each_entry_safe(obj, on, &objects, st_link) i915_gem_object_put(obj); diff --git a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c index aa31d6c0cdfb..7096c3911cd3 100644 --- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c +++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c @@ -316,6 +316,56 @@ static int igt_global_reset(void *arg) GEM_BUG_ON(test_bit(I915_RESET_HANDOFF, &i915->gpu_error.flags)); clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + wake_up_all(&i915->gpu_error.reset_queue); + + if (i915_terminally_wedged(&i915->gpu_error)) + err = -EIO; + + return err; +} + +static int igt_reset_engine(void *arg) +{ + struct drm_i915_private *i915 = arg; + struct intel_engine_cs *engine; + enum intel_engine_id id; + unsigned int reset_count, reset_engine_count; + int err = 0; + + /* Check that we can issue a global GPU and engine reset */ + + if (!intel_has_reset_engine(i915)) + return 0; + + for_each_engine(engine, i915, id) { + set_bit(I915_RESET_ENGINE + engine->id, &i915->gpu_error.flags); + reset_count = i915_reset_count(&i915->gpu_error); + reset_engine_count = i915_reset_engine_count(&i915->gpu_error, + engine); + + err = i915_reset_engine(engine); + if (err) { + pr_err("i915_reset_engine failed\n"); + break; + } + + if (i915_reset_count(&i915->gpu_error) != reset_count) { + pr_err("Full GPU reset recorded! (engine reset expected)\n"); + err = -EINVAL; + break; + } + + if (i915_reset_engine_count(&i915->gpu_error, engine) == + reset_engine_count) { + pr_err("No %s engine reset recorded!\n", engine->name); + err = -EINVAL; + break; + } + + clear_bit(I915_RESET_ENGINE + engine->id, + &i915->gpu_error.flags); + } + if (i915_terminally_wedged(&i915->gpu_error)) err = -EIO; @@ -404,6 +454,7 @@ fini: unlock: mutex_unlock(&i915->drm.struct_mutex); clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + wake_up_all(&i915->gpu_error.reset_queue); if (i915_terminally_wedged(&i915->gpu_error)) return -EIO; @@ -519,11 +570,117 @@ fini: unlock: mutex_unlock(&i915->drm.struct_mutex); clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + wake_up_all(&i915->gpu_error.reset_queue); + + if (i915_terminally_wedged(&i915->gpu_error)) + return -EIO; + + return err; +} + +static int igt_render_engine_reset_fallback(void *arg) +{ + struct drm_i915_private *i915 = arg; + struct intel_engine_cs *engine = i915->engine[RCS]; + struct hang h; + struct drm_i915_gem_request *rq; + unsigned int reset_count, reset_engine_count; + int err = 0; + + /* Check that we can issue a global GPU and engine reset */ + + if (!intel_has_reset_engine(i915)) + return 0; + + set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + mutex_lock(&i915->drm.struct_mutex); + + err = hang_init(&h, i915); + if (err) + goto err_unlock; + + rq = hang_create_request(&h, engine, i915->kernel_context); + if (IS_ERR(rq)) { + err = PTR_ERR(rq); + goto err_fini; + } + + i915_gem_request_get(rq); + __i915_add_request(rq, true); + + /* make reset engine fail */ + rq->fence.error = -EIO; + + if (!wait_for_hang(&h, rq)) { + pr_err("Failed to start request %x\n", rq->fence.seqno); + err = -EIO; + goto err_request; + } + + reset_engine_count = i915_reset_engine_count(&i915->gpu_error, engine); + reset_count = fake_hangcheck(rq); + + /* unlock since we'll call handle_error */ + mutex_unlock(&i915->drm.struct_mutex); + clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + wake_up_all(&i915->gpu_error.reset_queue); + + i915_handle_error(i915, intel_engine_flag(engine), "live test"); + + if (i915_reset_engine_count(&i915->gpu_error, engine) != + reset_engine_count) { + pr_err("render engine reset recorded! (full reset expected)\n"); + err = -EINVAL; + goto out_rq; + } + + if (i915_reset_count(&i915->gpu_error) == reset_count) { + pr_err("No full GPU reset recorded!\n"); + err = -EINVAL; + goto out_rq; + } + + /* + * by using fence.error = -EIO, full reset sets the wedged flag, do one + * more full reset to re-enable the hw. + */ + if (i915_terminally_wedged(&i915->gpu_error)) { + set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + rq->fence.error = 0; + + mutex_lock(&i915->drm.struct_mutex); + set_bit(I915_RESET_HANDOFF, &i915->gpu_error.flags); + i915_reset(i915); + GEM_BUG_ON(test_bit(I915_RESET_HANDOFF, + &i915->gpu_error.flags)); + mutex_unlock(&i915->drm.struct_mutex); + + if (i915_reset_count(&i915->gpu_error) == reset_count) { + pr_err("No full GPU reset recorded!\n"); + err = -EINVAL; + goto out_rq; + } + } + +out_rq: + i915_gem_request_put(rq); + hang_fini(&h); +out_backoff: + clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + wake_up_all(&i915->gpu_error.reset_queue); if (i915_terminally_wedged(&i915->gpu_error)) return -EIO; return err; + +err_request: + i915_gem_request_put(rq); +err_fini: + hang_fini(&h); +err_unlock: + mutex_unlock(&i915->drm.struct_mutex); + goto out_backoff; } int intel_hangcheck_live_selftests(struct drm_i915_private *i915) @@ -531,8 +688,10 @@ int intel_hangcheck_live_selftests(struct drm_i915_private *i915) static const struct i915_subtest tests[] = { SUBTEST(igt_hang_sanitycheck), SUBTEST(igt_global_reset), + SUBTEST(igt_reset_engine), SUBTEST(igt_wait_reset), SUBTEST(igt_reset_queue), + SUBTEST(igt_render_engine_reset_fallback), }; if (!intel_has_gpu_reset(i915)) diff --git a/drivers/gpu/drm/i915/selftests/mock_context.c b/drivers/gpu/drm/i915/selftests/mock_context.c index f8b9cc212b02..9c7c68181f82 100644 --- a/drivers/gpu/drm/i915/selftests/mock_context.c +++ b/drivers/gpu/drm/i915/selftests/mock_context.c @@ -48,7 +48,7 @@ mock_context(struct drm_i915_private *i915, if (!ctx->vma_lut.ht) goto err_free; - ret = ida_simple_get(&i915->context_hw_ida, + ret = ida_simple_get(&i915->contexts.hw_ida, 0, MAX_CONTEXT_HW_ID, GFP_KERNEL); if (ret < 0) goto err_vma_ht; @@ -86,3 +86,12 @@ void mock_context_close(struct i915_gem_context *ctx) i915_gem_context_put(ctx); } + +void mock_init_contexts(struct drm_i915_private *i915) +{ + INIT_LIST_HEAD(&i915->contexts.list); + ida_init(&i915->contexts.hw_ida); + + INIT_WORK(&i915->contexts.free_work, contexts_free_worker); + init_llist_head(&i915->contexts.free_list); +} diff --git a/drivers/gpu/drm/i915/selftests/mock_context.h b/drivers/gpu/drm/i915/selftests/mock_context.h index 2427e5c0916a..383941a61124 100644 --- a/drivers/gpu/drm/i915/selftests/mock_context.h +++ b/drivers/gpu/drm/i915/selftests/mock_context.h @@ -25,6 +25,8 @@ #ifndef __MOCK_CONTEXT_H #define __MOCK_CONTEXT_H +void mock_init_contexts(struct drm_i915_private *i915); + struct i915_gem_context * mock_context(struct drm_i915_private *i915, const char *name); diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c index 627e2aa09766..47613d20bba8 100644 --- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c +++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c @@ -57,11 +57,12 @@ static void mock_device_release(struct drm_device *dev) cancel_delayed_work_sync(&i915->gt.retire_work); cancel_delayed_work_sync(&i915->gt.idle_work); + flush_workqueue(i915->wq); mutex_lock(&i915->drm.struct_mutex); for_each_engine(engine, i915, id) mock_engine_free(engine); - i915_gem_context_fini(i915); + i915_gem_contexts_fini(i915); mutex_unlock(&i915->drm.struct_mutex); drain_workqueue(i915->wq); @@ -160,7 +161,7 @@ struct drm_i915_private *mock_gem_device(void) INIT_LIST_HEAD(&i915->mm.unbound_list); INIT_LIST_HEAD(&i915->mm.bound_list); - ida_init(&i915->context_hw_ida); + mock_init_contexts(i915); INIT_DELAYED_WORK(&i915->gt.retire_work, mock_retire_work_handler); INIT_DELAYED_WORK(&i915->gt.idle_work, mock_idle_work_handler); diff --git a/drivers/gpu/drm/i915/selftests/mock_gtt.c b/drivers/gpu/drm/i915/selftests/mock_gtt.c index a61309c7cb3e..f2118cf535a0 100644 --- a/drivers/gpu/drm/i915/selftests/mock_gtt.c +++ b/drivers/gpu/drm/i915/selftests/mock_gtt.c @@ -33,8 +33,7 @@ static void mock_insert_page(struct i915_address_space *vm, } static void mock_insert_entries(struct i915_address_space *vm, - struct sg_table *st, - u64 start, + struct i915_vma *vma, enum i915_cache_level level, u32 flags) { } |