Merge branch 'drm-next-4.10' of git://people.freedesktop.org/~agd5f/linux into drm-next

- lots of code cleanup
- lots of bug fixes
- expose rpm based fan info via hwmon
- lots of clock and powergating fixes
- SI register header cleanup and conversion to common format used by newer asics

* 'drm-next-4.10' of git://people.freedesktop.org/~agd5f/linux: (54 commits)
  drm/amdgpu: drop is_display_hung from display funcs
  drm/amdgpu/uvd: reduce IB parsing overhead on UVD5+ (v2)
  drm/amdgpu/uvd: consolidate code for fetching addr from ctx
  drm/amdgpu: Disable DPM in virtualization
  drm/amdgpu: use AMDGPU_GEM_CREATE_VRAM_CLEARED for VM PD/PTs (v2)
  drm/amdgpu: improve AMDGPU_GEM_CREATE_VRAM_CLEARED handling (v2)
  drm/amdgpu: fix error handling in amdgpu_bo_create_restricted
  drm/amdgpu: fix amdgpu_fill_buffer (v2)
  drm/amdgpu: remove amdgpu_irq_get_delayed
  amdgpu: Wrap dev_err() calls on vm faults with printk_ratelimit()
  amdgpu: Use dev_err() over vanilla printk() in vm_decode_fault()
  drm/amd/amdgpu: port of DCE v6 to new headers (v3)
  drm/amdgpu: cleanup unused iterator members for sdma v2.4
  drm/amdgpu: cleanup unused iterator members for sdma v3
  drm/amdgpu:impl vgt_flush for VI(V5)
  drm/amdgpu: enable uvd mgcg for Fiji.
  drm/amdgpu: refine cz uvd clock gate logic.
  drm/amdgpu: change log level to KERN_INFO in ci_dpm.c
  drm/amdgpu: always un-gate UVD REGS path.
  drm/amdgpu/sdma: fix typo in packet setup
  ...

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -92,13 +92,13 @@ extern int amdgpu_vm_debug;
 extern int amdgpu_sched_jobs;
 extern int amdgpu_sched_hw_submission;
 extern int amdgpu_powerplay;
-extern int amdgpu_powercontainment;
+extern int amdgpu_no_evict;
+extern int amdgpu_direct_gma_size;
 extern unsigned amdgpu_pcie_gen_cap;
 extern unsigned amdgpu_pcie_lane_cap;
 extern unsigned amdgpu_cg_mask;
 extern unsigned amdgpu_pg_mask;
 extern char *amdgpu_disable_cu;
-extern int amdgpu_sclk_deep_sleep_en;
 extern char *amdgpu_virtual_display;
 extern unsigned amdgpu_pp_feature_mask;
 extern int amdgpu_vram_page_split;
@@ -1633,7 +1633,6 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
 #define amdgpu_display_set_vga_render_state(adev, r) (adev)->mode_info.funcs->set_vga_render_state((adev), (r))
 #define amdgpu_display_vblank_get_counter(adev, crtc) (adev)->mode_info.funcs->vblank_get_counter((adev), (crtc))
 #define amdgpu_display_vblank_wait(adev, crtc) (adev)->mode_info.funcs->vblank_wait((adev), (crtc))
-#define amdgpu_display_is_display_hung(adev) (adev)->mode_info.funcs->is_display_hung((adev))
 #define amdgpu_display_backlight_set_level(adev, e, l) (adev)->mode_info.funcs->backlight_set_level((e), (l))
 #define amdgpu_display_backlight_get_level(adev, e) (adev)->mode_info.funcs->backlight_get_level((e))
 #define amdgpu_display_hpd_sense(adev, h) (adev)->mode_info.funcs->hpd_sense((adev), (h))

drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c

@@ -70,7 +70,7 @@ static bool igp_read_bios_from_vram(struct amdgpu_device *adev)
 		return false;
 	}
 	adev->bios = kmalloc(size, GFP_KERNEL);
-	if (adev->bios == NULL) {
+	if (!adev->bios) {
 		iounmap(bios);
 		return false;
 	}

drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c

@@ -841,16 +841,6 @@ static int amdgpu_cs_ib_vm_chunk(struct amdgpu_device *adev,
 	return amdgpu_cs_sync_rings(p);
 }
 
-static int amdgpu_cs_handle_lockup(struct amdgpu_device *adev, int r)
-{
-	if (r == -EDEADLK) {
-		r = amdgpu_gpu_reset(adev);
-		if (!r)
-			r = -EAGAIN;
-	}
-	return r;
-}
-
 static int amdgpu_cs_ib_fill(struct amdgpu_device *adev,
 			     struct amdgpu_cs_parser *parser)
 {
@@ -1054,29 +1044,29 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	r = amdgpu_cs_parser_init(&parser, data);
 	if (r) {
 		DRM_ERROR("Failed to initialize parser !\n");
-		amdgpu_cs_parser_fini(&parser, r, false);
-		r = amdgpu_cs_handle_lockup(adev, r);
-		return r;
-	}
-	r = amdgpu_cs_parser_bos(&parser, data);
-	if (r == -ENOMEM)
-		DRM_ERROR("Not enough memory for command submission!\n");
-	else if (r && r != -ERESTARTSYS)
-		DRM_ERROR("Failed to process the buffer list %d!\n", r);
-	else if (!r) {
-		reserved_buffers = true;
-		r = amdgpu_cs_ib_fill(adev, &parser);
+		goto out;
 	}
 
-	if (!r) {
-		r = amdgpu_cs_dependencies(adev, &parser);
-		if (r)
-			DRM_ERROR("Failed in the dependencies handling %d!\n", r);
+	r = amdgpu_cs_parser_bos(&parser, data);
+	if (r) {
+		if (r == -ENOMEM)
+			DRM_ERROR("Not enough memory for command submission!\n");
+		else if (r != -ERESTARTSYS)
+			DRM_ERROR("Failed to process the buffer list %d!\n", r);
+		goto out;
 	}
 
+	reserved_buffers = true;
+	r = amdgpu_cs_ib_fill(adev, &parser);
 	if (r)
 		goto out;
 
+	r = amdgpu_cs_dependencies(adev, &parser);
+	if (r) {
+		DRM_ERROR("Failed in the dependencies handling %d!\n", r);
+		goto out;
+	}
+
 	for (i = 0; i < parser.job->num_ibs; i++)
 		trace_amdgpu_cs(&parser, i);
 
@@ -1088,7 +1078,6 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 
 out:
 	amdgpu_cs_parser_fini(&parser, r, reserved_buffers);
-	r = amdgpu_cs_handle_lockup(adev, r);
 	return r;
 }
 

drivers/gpu/drm/amd/amdgpu/amdgpu_device.c

@@ -1017,8 +1017,8 @@ static void amdgpu_check_arguments(struct amdgpu_device *adev)
 		amdgpu_vm_block_size = 9;
 	}
 
-	if ((amdgpu_vram_page_split != -1 && amdgpu_vram_page_split < 16) ||
-	    !amdgpu_check_pot_argument(amdgpu_vram_page_split)) {
+	if (amdgpu_vram_page_split != -1 && (amdgpu_vram_page_split < 16 ||
+	    !amdgpu_check_pot_argument(amdgpu_vram_page_split))) {
 		dev_warn(adev->dev, "invalid VRAM page split (%d)\n",
 			 amdgpu_vram_page_split);
 		amdgpu_vram_page_split = 1024;

drivers/gpu/drm/amd/amdgpu/amdgpu_dpm.h

@@ -52,6 +52,8 @@ enum amdgpu_dpm_event_src {
 	AMDGPU_DPM_EVENT_SRC_DIGIAL_OR_EXTERNAL = 4
 };
 
+#define SCLK_DEEP_SLEEP_MASK 0x8
+
 struct amdgpu_ps {
 	u32 caps; /* vbios flags */
 	u32 class; /* vbios flags */
@@ -317,6 +319,11 @@ struct amdgpu_dpm_funcs {
 	      (adev)->powerplay.pp_funcs->get_fan_speed_percent((adev)->powerplay.pp_handle, (s)) : \
 	      (adev)->pm.funcs->get_fan_speed_percent((adev), (s)))
 
+#define amdgpu_dpm_get_fan_speed_rpm(adev, s) \
+	((adev)->pp_enabled ?						\
+	      (adev)->powerplay.pp_funcs->get_fan_speed_rpm((adev)->powerplay.pp_handle, (s)) : \
+	      -EINVAL)
+
 #define amdgpu_dpm_get_sclk(adev, l) \
 	((adev)->pp_enabled ?						\
 	      (adev)->powerplay.pp_funcs->get_sclk((adev)->powerplay.pp_handle, (l)) : \

drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c

@@ -91,8 +91,8 @@ int amdgpu_exp_hw_support = 0;
 int amdgpu_sched_jobs = 32;
 int amdgpu_sched_hw_submission = 2;
 int amdgpu_powerplay = -1;
-int amdgpu_powercontainment = 1;
-int amdgpu_sclk_deep_sleep_en = 1;
+int amdgpu_no_evict = 0;
+int amdgpu_direct_gma_size = 0;
 unsigned amdgpu_pcie_gen_cap = 0;
 unsigned amdgpu_pcie_lane_cap = 0;
 unsigned amdgpu_cg_mask = 0xffffffff;
@@ -182,14 +182,14 @@ module_param_named(sched_hw_submission, amdgpu_sched_hw_submission, int, 0444);
 MODULE_PARM_DESC(powerplay, "Powerplay component (1 = enable, 0 = disable, -1 = auto (default))");
 module_param_named(powerplay, amdgpu_powerplay, int, 0444);
 
-MODULE_PARM_DESC(powercontainment, "Power Containment (1 = enable (default), 0 = disable)");
-module_param_named(powercontainment, amdgpu_powercontainment, int, 0444);
-
 MODULE_PARM_DESC(ppfeaturemask, "all power features enabled (default))");
 module_param_named(ppfeaturemask, amdgpu_pp_feature_mask, int, 0444);
 
-MODULE_PARM_DESC(sclkdeepsleep, "SCLK Deep Sleep (1 = enable (default), 0 = disable)");
-module_param_named(sclkdeepsleep, amdgpu_sclk_deep_sleep_en, int, 0444);
+MODULE_PARM_DESC(no_evict, "Support pinning request from user space (1 = enable, 0 = disable (default))");
+module_param_named(no_evict, amdgpu_no_evict, int, 0444);
+
+MODULE_PARM_DESC(direct_gma_size, "Direct GMA size in megabytes (max 96MB)");
+module_param_named(direct_gma_size, amdgpu_direct_gma_size, int, 0444);
 
 MODULE_PARM_DESC(pcie_gen_cap, "PCIE Gen Caps (0: autodetect (default))");
 module_param_named(pcie_gen_cap, amdgpu_pcie_gen_cap, uint, 0444);

drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c

@@ -382,24 +382,27 @@ int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring,
 	if (!ring->fence_drv.fences)
 		return -ENOMEM;
 
-	timeout = msecs_to_jiffies(amdgpu_lockup_timeout);
-	if (timeout == 0) {
-		/*
-		 * FIXME:
-		 * Delayed workqueue cannot use it directly,
-		 * so the scheduler will not use delayed workqueue if
-		 * MAX_SCHEDULE_TIMEOUT is set.
-		 * Currently keep it simple and silly.
-		 */
-		timeout = MAX_SCHEDULE_TIMEOUT;
-	}
-	r = amd_sched_init(&ring->sched, &amdgpu_sched_ops,
-			   num_hw_submission,
-			   timeout, ring->name);
-	if (r) {
-		DRM_ERROR("Failed to create scheduler on ring %s.\n",
-			  ring->name);
-		return r;
+	/* No need to setup the GPU scheduler for KIQ ring */
+	if (ring->funcs->type != AMDGPU_RING_TYPE_KIQ) {
+		timeout = msecs_to_jiffies(amdgpu_lockup_timeout);
+		if (timeout == 0) {
+			/*
+			 * FIXME:
+			 * Delayed workqueue cannot use it directly,
+			 * so the scheduler will not use delayed workqueue if
+			 * MAX_SCHEDULE_TIMEOUT is set.
+			 * Currently keep it simple and silly.
+			 */
+			timeout = MAX_SCHEDULE_TIMEOUT;
+		}
+		r = amd_sched_init(&ring->sched, &amdgpu_sched_ops,
+				   num_hw_submission,
+				   timeout, ring->name);
+		if (r) {
+			DRM_ERROR("Failed to create scheduler on ring %s.\n",
+				  ring->name);
+			return r;
+		}
 	}
 
 	return 0;

drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c

@@ -424,15 +424,6 @@ int amdgpu_irq_get(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 	return 0;
 }
 
-bool amdgpu_irq_get_delayed(struct amdgpu_device *adev,
-			struct amdgpu_irq_src *src,
-			unsigned type)
-{
-	if ((type >= src->num_types) || !src->enabled_types)
-		return false;
-	return atomic_inc_return(&src->enabled_types[type]) == 1;
-}
-
 /**
  * amdgpu_irq_put - disable interrupt
  *

drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h

@@ -88,9 +88,6 @@ int amdgpu_irq_update(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 		      unsigned type);
 int amdgpu_irq_get(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 		   unsigned type);
-bool amdgpu_irq_get_delayed(struct amdgpu_device *adev,
-			    struct amdgpu_irq_src *src,
-			    unsigned type);
 int amdgpu_irq_put(struct amdgpu_device *adev, struct amdgpu_irq_src *src,
 		   unsigned type);
 bool amdgpu_irq_enabled(struct amdgpu_device *adev, struct amdgpu_irq_src *src,

drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h

@@ -271,8 +271,6 @@ struct amdgpu_display_funcs {
 	u32 (*vblank_get_counter)(struct amdgpu_device *adev, int crtc);
 	/* wait for vblank */
 	void (*vblank_wait)(struct amdgpu_device *adev, int crtc);
-	/* is dce hung */
-	bool (*is_display_hung)(struct amdgpu_device *adev);
 	/* set backlight level */
 	void (*backlight_set_level)(struct amdgpu_encoder *amdgpu_encoder,
 				    u8 level);

drivers/gpu/drm/amd/amdgpu/amdgpu_object.c

@@ -128,17 +128,6 @@ static void amdgpu_ttm_placement_init(struct amdgpu_device *adev,
 		if (flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS)
 			lpfn = adev->mc.real_vram_size >> PAGE_SHIFT;
 
-		if (flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS &&
-		    !(flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) &&
-		    adev->mc.visible_vram_size < adev->mc.real_vram_size) {
-			places[c].fpfn = visible_pfn;
-			places[c].lpfn = lpfn;
-			places[c].flags = TTM_PL_FLAG_WC |
-				TTM_PL_FLAG_UNCACHED | TTM_PL_FLAG_VRAM |
-				TTM_PL_FLAG_TOPDOWN;
-			c++;
-		}
-
 		places[c].fpfn = 0;
 		places[c].lpfn = lpfn;
 		places[c].flags = TTM_PL_FLAG_WC | TTM_PL_FLAG_UNCACHED |
@@ -382,39 +371,36 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev,
 
 	amdgpu_fill_placement_to_bo(bo, placement);
 	/* Kernel allocation are uninterruptible */
+
+	if (!resv) {
+		bool locked;
+
+		reservation_object_init(&bo->tbo.ttm_resv);
+		locked = ww_mutex_trylock(&bo->tbo.ttm_resv.lock);
+		WARN_ON(!locked);
+	}
 	r = ttm_bo_init(&adev->mman.bdev, &bo->tbo, size, type,
 			&bo->placement, page_align, !kernel, NULL,
-			acc_size, sg, resv, &amdgpu_ttm_bo_destroy);
-	if (unlikely(r != 0)) {
+			acc_size, sg, resv ? resv : &bo->tbo.ttm_resv,
+			&amdgpu_ttm_bo_destroy);
+	if (unlikely(r != 0))
 		return r;
-	}
 
 	if (flags & AMDGPU_GEM_CREATE_VRAM_CLEARED &&
 	    bo->tbo.mem.placement & TTM_PL_FLAG_VRAM) {
 		struct dma_fence *fence;
 
-		if (adev->mman.buffer_funcs_ring == NULL ||
-		   !adev->mman.buffer_funcs_ring->ready) {
-			r = -EBUSY;
-			goto fail_free;
-		}
-
-		r = amdgpu_bo_reserve(bo, false);
-		if (unlikely(r != 0))
-			goto fail_free;
-
-		amdgpu_ttm_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_VRAM);
-		r = ttm_bo_validate(&bo->tbo, &bo->placement, false, false);
-		if (unlikely(r != 0))
+		r = amdgpu_fill_buffer(bo, 0, bo->tbo.resv, &fence);
+		if (unlikely(r))
 			goto fail_unreserve;
 
-		amdgpu_fill_buffer(bo, 0, bo->tbo.resv, &fence);
 		amdgpu_bo_fence(bo, fence, false);
-		amdgpu_bo_unreserve(bo);
 		dma_fence_put(bo->tbo.moving);
 		bo->tbo.moving = dma_fence_get(fence);
 		dma_fence_put(fence);
 	}
+	if (!resv)
+		ww_mutex_unlock(&bo->tbo.resv->lock);
 	*bo_ptr = bo;
 
 	trace_amdgpu_bo_create(bo);
@@ -422,8 +408,7 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev,
 	return 0;
 
 fail_unreserve:
-	amdgpu_bo_unreserve(bo);
-fail_free:
+	ww_mutex_unlock(&bo->tbo.resv->lock);
 	amdgpu_bo_unref(&bo);
 	return r;
 }

drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c

@@ -737,6 +737,21 @@ static ssize_t amdgpu_hwmon_get_pwm1(struct device *dev,
 	return sprintf(buf, "%i\n", speed);
 }
 
+static ssize_t amdgpu_hwmon_get_fan1_input(struct device *dev,
+					   struct device_attribute *attr,
+					   char *buf)
+{
+	struct amdgpu_device *adev = dev_get_drvdata(dev);
+	int err;
+	u32 speed;
+
+	err = amdgpu_dpm_get_fan_speed_rpm(adev, &speed);
+	if (err)
+		return err;
+
+	return sprintf(buf, "%i\n", speed);
+}
+
 static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, amdgpu_hwmon_show_temp, NULL, 0);
 static SENSOR_DEVICE_ATTR(temp1_crit, S_IRUGO, amdgpu_hwmon_show_temp_thresh, NULL, 0);
 static SENSOR_DEVICE_ATTR(temp1_crit_hyst, S_IRUGO, amdgpu_hwmon_show_temp_thresh, NULL, 1);
@@ -744,6 +759,7 @@ static SENSOR_DEVICE_ATTR(pwm1, S_IRUGO | S_IWUSR, amdgpu_hwmon_get_pwm1, amdgpu
 static SENSOR_DEVICE_ATTR(pwm1_enable, S_IRUGO | S_IWUSR, amdgpu_hwmon_get_pwm1_enable, amdgpu_hwmon_set_pwm1_enable, 0);
 static SENSOR_DEVICE_ATTR(pwm1_min, S_IRUGO, amdgpu_hwmon_get_pwm1_min, NULL, 0);
 static SENSOR_DEVICE_ATTR(pwm1_max, S_IRUGO, amdgpu_hwmon_get_pwm1_max, NULL, 0);
+static SENSOR_DEVICE_ATTR(fan1_input, S_IRUGO, amdgpu_hwmon_get_fan1_input, NULL, 0);
 
 static struct attribute *hwmon_attributes[] = {
 	&sensor_dev_attr_temp1_input.dev_attr.attr,
@@ -753,6 +769,7 @@ static struct attribute *hwmon_attributes[] = {
 	&sensor_dev_attr_pwm1_enable.dev_attr.attr,
 	&sensor_dev_attr_pwm1_min.dev_attr.attr,
 	&sensor_dev_attr_pwm1_max.dev_attr.attr,
+	&sensor_dev_attr_fan1_input.dev_attr.attr,
 	NULL
 };
 
@@ -804,6 +821,10 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	     attr == &sensor_dev_attr_pwm1_min.dev_attr.attr))
 		return 0;
 
+	/* requires powerplay */
+	if (attr == &sensor_dev_attr_fan1_input.dev_attr.attr)
+		return 0;
+
 	return effective_mode;
 }
 

drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c

@@ -155,9 +155,6 @@ static int amdgpu_pp_sw_init(void *handle)
 		ret = adev->powerplay.ip_funcs->sw_init(
 					adev->powerplay.pp_handle);
 
-	if (adev->pp_enabled)
-		adev->pm.dpm_enabled = true;
-
 	return ret;
 }
 
@@ -187,6 +184,9 @@ static int amdgpu_pp_hw_init(void *handle)
 		ret = adev->powerplay.ip_funcs->hw_init(
 					adev->powerplay.pp_handle);
 
+	if ((amdgpu_dpm != 0) && !amdgpu_sriov_vf(adev))
+		adev->pm.dpm_enabled = true;
+
 	return ret;
 }
 

drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h

@@ -44,7 +44,8 @@ enum amdgpu_ring_type {
 	AMDGPU_RING_TYPE_COMPUTE,
 	AMDGPU_RING_TYPE_SDMA,
 	AMDGPU_RING_TYPE_UVD,
-	AMDGPU_RING_TYPE_VCE
+	AMDGPU_RING_TYPE_VCE,
+	AMDGPU_RING_TYPE_KIQ
 };
 
 struct amdgpu_device;

drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c

@@ -327,9 +327,8 @@ int amdgpu_sa_bo_new(struct amdgpu_sa_manager *sa_manager,
 		return -EINVAL;
 
 	*sa_bo = kmalloc(sizeof(struct amdgpu_sa_bo), GFP_KERNEL);
-	if ((*sa_bo) == NULL) {
+	if (!(*sa_bo))
 		return -ENOMEM;
-	}
 	(*sa_bo)->manager = sa_manager;
 	(*sa_bo)->fence = NULL;
 	INIT_LIST_HEAD(&(*sa_bo)->olist);

drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c

@@ -1382,28 +1382,40 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring,
 }
 
 int amdgpu_fill_buffer(struct amdgpu_bo *bo,
-		uint32_t src_data,
-		struct reservation_object *resv,
-		struct dma_fence **fence)
+		       uint32_t src_data,
+		       struct reservation_object *resv,
+		       struct dma_fence **fence)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
-	struct amdgpu_job *job;
+	uint32_t max_bytes = adev->mman.buffer_funcs->fill_max_bytes;
 	struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
 
-	uint32_t max_bytes, byte_count;
-	uint64_t dst_offset;
+	struct drm_mm_node *mm_node;
+	unsigned long num_pages;
 	unsigned int num_loops, num_dw;
-	unsigned int i;
+
+	struct amdgpu_job *job;
 	int r;
 
-	byte_count = bo->tbo.num_pages << PAGE_SHIFT;
-	max_bytes = adev->mman.buffer_funcs->fill_max_bytes;
-	num_loops = DIV_ROUND_UP(byte_count, max_bytes);
+	if (!ring->ready) {
+		DRM_ERROR("Trying to clear memory with ring turned off.\n");
+		return -EINVAL;
+	}
+
+	num_pages = bo->tbo.num_pages;
+	mm_node = bo->tbo.mem.mm_node;
+	num_loops = 0;
+	while (num_pages) {
+		uint32_t byte_count = mm_node->size << PAGE_SHIFT;
+
+		num_loops += DIV_ROUND_UP(byte_count, max_bytes);
+		num_pages -= mm_node->size;
+		++mm_node;
+	}
 	num_dw = num_loops * adev->mman.buffer_funcs->fill_num_dw;
 
 	/* for IB padding */
-	while (num_dw & 0x7)
-		num_dw++;
+	num_dw += 64;
 
 	r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, &job);
 	if (r)
@@ -1411,28 +1423,43 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 
 	if (resv) {
 		r = amdgpu_sync_resv(adev, &job->sync, resv,
-				AMDGPU_FENCE_OWNER_UNDEFINED);
+				     AMDGPU_FENCE_OWNER_UNDEFINED);
 		if (r) {
 			DRM_ERROR("sync failed (%d).\n", r);
 			goto error_free;
 		}
 	}
 
-	dst_offset = bo->tbo.mem.start << PAGE_SHIFT;
-	for (i = 0; i < num_loops; i++) {
-		uint32_t cur_size_in_bytes = min(byte_count, max_bytes);
+	num_pages = bo->tbo.num_pages;
+	mm_node = bo->tbo.mem.mm_node;
 
-		amdgpu_emit_fill_buffer(adev, &job->ibs[0], src_data,
-				dst_offset, cur_size_in_bytes);
+	while (num_pages) {
+		uint32_t byte_count = mm_node->size << PAGE_SHIFT;
+		uint64_t dst_addr;
 
-		dst_offset += cur_size_in_bytes;
-		byte_count -= cur_size_in_bytes;
+		r = amdgpu_mm_node_addr(&bo->tbo, mm_node,
+					&bo->tbo.mem, &dst_addr);
+		if (r)
+			return r;
+
+		while (byte_count) {
+			uint32_t cur_size_in_bytes = min(byte_count, max_bytes);
+
+			amdgpu_emit_fill_buffer(adev, &job->ibs[0], src_data,
+						dst_addr, cur_size_in_bytes);
+
+			dst_addr += cur_size_in_bytes;
+			byte_count -= cur_size_in_bytes;
+		}
+
+		num_pages -= mm_node->size;
+		++mm_node;
 	}
 
 	amdgpu_ring_pad_ib(ring, &job->ibs[0]);
 	WARN_ON(job->ibs[0].length_dw > num_dw);
 	r = amdgpu_job_submit(job, ring, &adev->mman.entity,
-			AMDGPU_FENCE_OWNER_UNDEFINED, fence);
+			      AMDGPU_FENCE_OWNER_UNDEFINED, fence);
 	if (r)
 		goto error_free;
 

drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c

@@ -360,6 +360,18 @@ static void amdgpu_uvd_force_into_uvd_segment(struct amdgpu_bo *abo)
 	}
 }
 
+static u64 amdgpu_uvd_get_addr_from_ctx(struct amdgpu_uvd_cs_ctx *ctx)
+{
+	uint32_t lo, hi;
+	uint64_t addr;
+
+	lo = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->data0);
+	hi = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->data1);
+	addr = ((uint64_t)lo) | (((uint64_t)hi) << 32);
+
+	return addr;
+}
+
 /**
  * amdgpu_uvd_cs_pass1 - first parsing round
  *
@@ -372,14 +384,10 @@ static int amdgpu_uvd_cs_pass1(struct amdgpu_uvd_cs_ctx *ctx)
 {
 	struct amdgpu_bo_va_mapping *mapping;
 	struct amdgpu_bo *bo;
-	uint32_t cmd, lo, hi;
-	uint64_t addr;
+	uint32_t cmd;
+	uint64_t addr = amdgpu_uvd_get_addr_from_ctx(ctx);
 	int r = 0;
 
-	lo = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->data0);
-	hi = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->data1);
-	addr = ((uint64_t)lo) | (((uint64_t)hi) << 32);
-
 	mapping = amdgpu_cs_find_mapping(ctx->parser, addr, &bo);
 	if (mapping == NULL) {
 		DRM_ERROR("Can't find BO for addr 0x%08Lx\n", addr);
@@ -698,18 +706,16 @@ static int amdgpu_uvd_cs_pass2(struct amdgpu_uvd_cs_ctx *ctx)
 {
 	struct amdgpu_bo_va_mapping *mapping;
 	struct amdgpu_bo *bo;
-	uint32_t cmd, lo, hi;
+	uint32_t cmd;
 	uint64_t start, end;
-	uint64_t addr;
+	uint64_t addr = amdgpu_uvd_get_addr_from_ctx(ctx);
 	int r;
 
-	lo = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->data0);
-	hi = amdgpu_get_ib_value(ctx->parser, ctx->ib_idx, ctx->data1);
-	addr = ((uint64_t)lo) | (((uint64_t)hi) << 32);
-
 	mapping = amdgpu_cs_find_mapping(ctx->parser, addr, &bo);
-	if (mapping == NULL)
+	if (mapping == NULL) {
+		DRM_ERROR("Can't find BO for addr 0x%08Lx\n", addr);
 		return -EINVAL;
+	}
 
 	start = amdgpu_bo_gpu_offset(bo);
 
@@ -893,10 +899,13 @@ int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, uint32_t ib_idx)
 	ctx.buf_sizes = buf_sizes;
 	ctx.ib_idx = ib_idx;
 
-	/* first round, make sure the buffers are actually in the UVD segment */
-	r = amdgpu_uvd_cs_packets(&ctx, amdgpu_uvd_cs_pass1);
-	if (r)
-		return r;
+	/* first round only required on chips without UVD 64 bit address support */
+	if (!parser->adev->uvd.address_64_bit) {
+		/* first round, make sure the buffers are actually in the UVD segment */
+		r = amdgpu_uvd_cs_packets(&ctx, amdgpu_uvd_cs_pass1);
+		if (r)
+			return r;
+	}
 
 	/* second round, patch buffer addresses into the command stream */
 	r = amdgpu_uvd_cs_packets(&ctx, amdgpu_uvd_cs_pass2);

drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c

@@ -529,70 +529,6 @@ static void amdgpu_vm_do_copy_ptes(struct amdgpu_pte_update_params *params,
 	amdgpu_vm_copy_pte(params->adev, params->ib, pe, src, count);
 }
 
-/**
- * amdgpu_vm_clear_bo - initially clear the page dir/table
- *
- * @adev: amdgpu_device pointer
- * @bo: bo to clear
- *
- * need to reserve bo first before calling it.
- */
-static int amdgpu_vm_clear_bo(struct amdgpu_device *adev,
-			      struct amdgpu_vm *vm,
-			      struct amdgpu_bo *bo)
-{
-	struct amdgpu_ring *ring;
-	struct dma_fence *fence = NULL;
-	struct amdgpu_job *job;
-	struct amdgpu_pte_update_params params;
-	unsigned entries;
-	uint64_t addr;
-	int r;
-
-	ring = container_of(vm->entity.sched, struct amdgpu_ring, sched);
-
-	r = reservation_object_reserve_shared(bo->tbo.resv);
-	if (r)
-		return r;
-
-	r = ttm_bo_validate(&bo->tbo, &bo->placement, true, false);
-	if (r)
-		goto error;
-
-	r = amdgpu_ttm_bind(&bo->tbo, &bo->tbo.mem);
-	if (r)
-		goto error;
-
-	addr = amdgpu_bo_gpu_offset(bo);
-	entries = amdgpu_bo_size(bo) / 8;
-
-	r = amdgpu_job_alloc_with_ib(adev, 64, &job);
-	if (r)
-		goto error;
-
-	memset(&params, 0, sizeof(params));
-	params.adev = adev;
-	params.ib = &job->ibs[0];
-	amdgpu_vm_do_set_ptes(&params, addr, 0, entries, 0, 0);
-	amdgpu_ring_pad_ib(ring, &job->ibs[0]);
-
-	WARN_ON(job->ibs[0].length_dw > 64);
-	r = amdgpu_job_submit(job, ring, &vm->entity,
-			      AMDGPU_FENCE_OWNER_VM, &fence);
-	if (r)
-		goto error_free;
-
-	amdgpu_bo_fence(bo, fence, true);
-	dma_fence_put(fence);
-	return 0;
-
-error_free:
-	amdgpu_job_free(job);
-
-error:
-	return r;
-}
-
 /**
  * amdgpu_vm_map_gart - Resolve gart mapping of addr
  *
@@ -1435,7 +1371,8 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 				     AMDGPU_GEM_DOMAIN_VRAM,
 				     AMDGPU_GEM_CREATE_NO_CPU_ACCESS |
 				     AMDGPU_GEM_CREATE_SHADOW |
-				     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
+				     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS |
+				     AMDGPU_GEM_CREATE_VRAM_CLEARED,
 				     NULL, resv, &pt);
 		if (r)
 			goto error_free;
@@ -1445,22 +1382,6 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 		 */
 		pt->parent = amdgpu_bo_ref(vm->page_directory);
 
-		r = amdgpu_vm_clear_bo(adev, vm, pt);
-		if (r) {
-			amdgpu_bo_unref(&pt->shadow);
-			amdgpu_bo_unref(&pt);
-			goto error_free;
-		}
-
-		if (pt->shadow) {
-			r = amdgpu_vm_clear_bo(adev, vm, pt->shadow);
-			if (r) {
-				amdgpu_bo_unref(&pt->shadow);
-				amdgpu_bo_unref(&pt);
-				goto error_free;
-			}
-		}
-
 		vm->page_tables[pt_idx].bo = pt;
 		vm->page_tables[pt_idx].addr = 0;
 	}
@@ -1642,7 +1563,8 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 			     AMDGPU_GEM_DOMAIN_VRAM,
 			     AMDGPU_GEM_CREATE_NO_CPU_ACCESS |
 			     AMDGPU_GEM_CREATE_SHADOW |
-			     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS,
+			     AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS |
+			     AMDGPU_GEM_CREATE_VRAM_CLEARED,
 			     NULL, NULL, &vm->page_directory);
 	if (r)
 		goto error_free_sched_entity;
@@ -1651,24 +1573,11 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	if (r)
 		goto error_free_page_directory;
 
-	r = amdgpu_vm_clear_bo(adev, vm, vm->page_directory);
-	if (r)
-		goto error_unreserve;
-
-	if (vm->page_directory->shadow) {
-		r = amdgpu_vm_clear_bo(adev, vm, vm->page_directory->shadow);
-		if (r)
-			goto error_unreserve;
-	}
-
 	vm->last_eviction_counter = atomic64_read(&adev->num_evictions);
 	amdgpu_bo_unreserve(vm->page_directory);
 
 	return 0;
 
-error_unreserve:
-	amdgpu_bo_unreserve(vm->page_directory);
-
 error_free_page_directory:
 	amdgpu_bo_unref(&vm->page_directory->shadow);
 	amdgpu_bo_unref(&vm->page_directory);

drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c

@@ -108,7 +108,7 @@ static int amdgpu_vram_mgr_new(struct ttm_mem_type_manager *man,
 		lpfn = man->size;
 
 	if (bo->flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS ||
-	    amdgpu_vram_page_split == -1) {
+	    place->lpfn || amdgpu_vram_page_split == -1) {
 		pages_per_node = ~0ul;
 		num_nodes = 1;
 	} else {

drivers/gpu/drm/amd/amdgpu/ci_dpm.c

@@ -4202,11 +4202,6 @@ static int ci_update_uvd_dpm(struct amdgpu_device *adev, bool gate)
 
 	if (!gate) {
 		/* turn the clocks on when decoding */
-		ret = amdgpu_set_clockgating_state(adev, AMD_IP_BLOCK_TYPE_UVD,
-						    AMD_CG_STATE_UNGATE);
-		if (ret)
-			return ret;
-
 		if (pi->caps_uvd_dpm ||
 		    (adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count <= 0))
 			pi->smc_state_table.UvdBootLevel = 0;
@@ -4223,9 +4218,6 @@ static int ci_update_uvd_dpm(struct amdgpu_device *adev, bool gate)
 		ret = ci_enable_uvd_dpm(adev, false);
 		if (ret)
 			return ret;
-
-		ret = amdgpu_set_clockgating_state(adev, AMD_IP_BLOCK_TYPE_UVD,
-						    AMD_CG_STATE_GATE);
 	}
 
 	return ret;
@@ -5896,7 +5888,7 @@ static int ci_dpm_init(struct amdgpu_device *adev)
 	pi->pcie_dpm_key_disabled = 0;
 	pi->thermal_sclk_dpm_enabled = 0;
 
-	if (amdgpu_sclk_deep_sleep_en)
+	if (amdgpu_pp_feature_mask & SCLK_DEEP_SLEEP_MASK)
 		pi->caps_sclk_ds = true;
 	else
 		pi->caps_sclk_ds = false;
@@ -5999,7 +5991,7 @@ static int ci_dpm_init(struct amdgpu_device *adev)
 			tmp |= CNB_PWRMGT_CNTL__DPM_ENABLED_MASK;
 			break;
 		default:
-			DRM_ERROR("Invalid PCC GPIO: %u!\n", gpio.shift);
+			DRM_INFO("Invalid PCC GPIO: %u!\n", gpio.shift);
 			break;
 		}
 		WREG32_SMC(ixCNB_PWRMGT_CNTL, tmp);