Merge tag 'drm-intel-fixes-2015-07-15' into drm-intel-next-queued

Backmerge fixes since it's getting out of hand again with the massive
split due to atomic between -next and 4.2-rc. All the bugfixes in
4.2-rc are addressed already (by converting more towards atomic
instead of minimal duct-tape) so just always pick the version in next
for the conflicts in modeset code.

All the other conflicts are just adjacent lines changed.

Conflicts:
	drivers/gpu/drm/i915/i915_drv.h
	drivers/gpu/drm/i915/i915_gem_gtt.c
	drivers/gpu/drm/i915/intel_display.c
	drivers/gpu/drm/i915/intel_drv.h
	drivers/gpu/drm/i915/intel_ringbuffer.h

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

84 files changed, 1733 insertions, 922 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 22866d1c3d69..01657830b470 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -425,6 +425,8 @@ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring,
 				   unsigned irq_type);
 int amdgpu_fence_emit(struct amdgpu_ring *ring, void *owner,
 		      struct amdgpu_fence **fence);
+int amdgpu_fence_recreate(struct amdgpu_ring *ring, void *owner,
+			  uint64_t seq, struct amdgpu_fence **fence);
 void amdgpu_fence_process(struct amdgpu_ring *ring);
 int amdgpu_fence_wait_next(struct amdgpu_ring *ring);
 int amdgpu_fence_wait_empty(struct amdgpu_ring *ring);
@@ -435,9 +437,6 @@ int amdgpu_fence_wait(struct amdgpu_fence *fence, bool interruptible);
 int amdgpu_fence_wait_any(struct amdgpu_device *adev,
 			  struct amdgpu_fence **fences,
 			  bool intr);
-long amdgpu_fence_wait_seq_timeout(struct amdgpu_device *adev,
-				   u64 *target_seq, bool intr,
-				   long timeout);
 struct amdgpu_fence *amdgpu_fence_ref(struct amdgpu_fence *fence);
 void amdgpu_fence_unref(struct amdgpu_fence **fence);
 
@@ -1622,6 +1621,7 @@ struct amdgpu_vce {
 	unsigned		fb_version;
 	atomic_t		handles[AMDGPU_MAX_VCE_HANDLES];
 	struct drm_file		*filp[AMDGPU_MAX_VCE_HANDLES];
+	uint32_t		img_size[AMDGPU_MAX_VCE_HANDLES];
 	struct delayed_work	idle_work;
 	const struct firmware	*fw;	/* VCE firmware */
 	struct amdgpu_ring	ring[AMDGPU_MAX_VCE_RINGS];
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
index 36d34e0afbc3..f82a2dd83874 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
@@ -30,6 +30,7 @@
 
 #include <drm/drmP.h>
 #include "amdgpu.h"
+#include "amdgpu_trace.h"
 
 static int amdgpu_bo_list_create(struct amdgpu_fpriv *fpriv,
 				 struct amdgpu_bo_list **result,
@@ -124,6 +125,8 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev,
 			gws_obj = entry->robj;
 		if (entry->prefered_domains == AMDGPU_GEM_DOMAIN_OA)
 			oa_obj = entry->robj;
+
+		trace_amdgpu_bo_list_set(list, entry->robj);
 	}
 
 	for (i = 0; i < list->num_entries; ++i)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index f09b2cba40ca..d63135bf29c0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -181,8 +181,6 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 		}
 		p->chunks[i].chunk_id = user_chunk.chunk_id;
 		p->chunks[i].length_dw = user_chunk.length_dw;
-		if (p->chunks[i].chunk_id == AMDGPU_CHUNK_ID_IB)
-			p->num_ibs++;
 
 		size = p->chunks[i].length_dw;
 		cdata = (void __user *)(unsigned long)user_chunk.chunk_data;
@@ -199,7 +197,12 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 			goto out;
 		}
 
-		if (p->chunks[i].chunk_id == AMDGPU_CHUNK_ID_FENCE) {
+		switch (p->chunks[i].chunk_id) {
+		case AMDGPU_CHUNK_ID_IB:
+			p->num_ibs++;
+			break;
+
+		case AMDGPU_CHUNK_ID_FENCE:
 			size = sizeof(struct drm_amdgpu_cs_chunk_fence);
 			if (p->chunks[i].length_dw * sizeof(uint32_t) >= size) {
 				uint32_t handle;
@@ -221,6 +224,14 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
 				r = -EINVAL;
 				goto out;
 			}
+			break;
+
+		case AMDGPU_CHUNK_ID_DEPENDENCIES:
+			break;
+
+		default:
+			r = -EINVAL;
+			goto out;
 		}
 	}
 
@@ -445,8 +456,9 @@ static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error, bo
 	for (i = 0; i < parser->nchunks; i++)
 		drm_free_large(parser->chunks[i].kdata);
 	kfree(parser->chunks);
-	for (i = 0; i < parser->num_ibs; i++)
-		amdgpu_ib_free(parser->adev, &parser->ibs[i]);
+	if (parser->ibs)
+		for (i = 0; i < parser->num_ibs; i++)
+			amdgpu_ib_free(parser->adev, &parser->ibs[i]);
 	kfree(parser->ibs);
 	if (parser->uf.bo)
 		drm_gem_object_unreference_unlocked(&parser->uf.bo->gem_base);
@@ -654,6 +666,55 @@ static int amdgpu_cs_ib_fill(struct amdgpu_device *adev,
 	return 0;
 }
 
+static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
+				  struct amdgpu_cs_parser *p)
+{
+	struct amdgpu_ib *ib;
+	int i, j, r;
+
+	if (!p->num_ibs)
+		return 0;
+
+	/* Add dependencies to first IB */
+	ib = &p->ibs[0];
+	for (i = 0; i < p->nchunks; ++i) {
+		struct drm_amdgpu_cs_chunk_dep *deps;
+		struct amdgpu_cs_chunk *chunk;
+		unsigned num_deps;
+
+		chunk = &p->chunks[i];
+
+		if (chunk->chunk_id != AMDGPU_CHUNK_ID_DEPENDENCIES)
+			continue;
+
+		deps = (struct drm_amdgpu_cs_chunk_dep *)chunk->kdata;
+		num_deps = chunk->length_dw * 4 /
+			sizeof(struct drm_amdgpu_cs_chunk_dep);
+
+		for (j = 0; j < num_deps; ++j) {
+			struct amdgpu_fence *fence;
+			struct amdgpu_ring *ring;
+
+			r = amdgpu_cs_get_ring(adev, deps[j].ip_type,
+					       deps[j].ip_instance,
+					       deps[j].ring, &ring);
+			if (r)
+				return r;
+
+			r = amdgpu_fence_recreate(ring, p->filp,
+						  deps[j].handle,
+						  &fence);
+			if (r)
+				return r;
+
+			amdgpu_sync_fence(&ib->sync, fence);
+			amdgpu_fence_unref(&fence);
+		}
+	}
+
+	return 0;
+}
+
 int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 {
 	struct amdgpu_device *adev = dev->dev_private;
@@ -688,11 +749,16 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 			else
 				DRM_ERROR("Failed to process the buffer list %d!\n", r);
 		}
-	} else {
+	}
+
+	if (!r) {
 		reserved_buffers = true;
 		r = amdgpu_cs_ib_fill(adev, &parser);
 	}
 
+	if (!r)
+		r = amdgpu_cs_dependencies(adev, &parser);
+
 	if (r) {
 		amdgpu_cs_parser_fini(&parser, r, reserved_buffers);
 		up_read(&adev->exclusive_lock);
@@ -730,9 +796,9 @@ int amdgpu_cs_wait_ioctl(struct drm_device *dev, void *data,
 {
 	union drm_amdgpu_wait_cs *wait = data;
 	struct amdgpu_device *adev = dev->dev_private;
-	uint64_t seq[AMDGPU_MAX_RINGS] = {0};
-	struct amdgpu_ring *ring = NULL;
 	unsigned long timeout = amdgpu_gem_timeout(wait->in.timeout);
+	struct amdgpu_fence *fence = NULL;
+	struct amdgpu_ring *ring = NULL;
 	struct amdgpu_ctx *ctx;
 	long r;
 
@@ -745,9 +811,12 @@ int amdgpu_cs_wait_ioctl(struct drm_device *dev, void *data,
 	if (r)
 		return r;
 
-	seq[ring->idx] = wait->in.handle;
+	r = amdgpu_fence_recreate(ring, filp, wait->in.handle, &fence);
+	if (r)
+		return r;
 
-	r = amdgpu_fence_wait_seq_timeout(adev, seq, true, timeout);
+	r = fence_wait_timeout(&fence->base, true, timeout);
+	amdgpu_fence_unref(&fence);
 	amdgpu_ctx_put(ctx);
 	if (r < 0)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index fec487d1c870..ba46be361c9b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1191,7 +1191,9 @@ static int amdgpu_early_init(struct amdgpu_device *adev)
 		return -EINVAL;
 	}
 
-
+	adev->ip_block_enabled = kcalloc(adev->num_ip_blocks, sizeof(bool), GFP_KERNEL);
+	if (adev->ip_block_enabled == NULL)
+		return -ENOMEM;
 
 	if (adev->ip_blocks == NULL) {
 		DRM_ERROR("No IP blocks found!\n");
@@ -1575,8 +1577,7 @@ void amdgpu_device_fini(struct amdgpu_device *adev)
 	amdgpu_fence_driver_fini(adev);
 	amdgpu_fbdev_fini(adev);
 	r = amdgpu_fini(adev);
-	if (adev->ip_block_enabled)
-		kfree(adev->ip_block_enabled);
+	kfree(adev->ip_block_enabled);
 	adev->ip_block_enabled = NULL;
 	adev->accel_working = false;
 	/* free i2c buses */
@@ -2000,4 +2001,10 @@ int amdgpu_debugfs_init(struct drm_minor *minor)
 void amdgpu_debugfs_cleanup(struct drm_minor *minor)
 {
 }
+#else
+static int amdgpu_debugfs_regs_init(struct amdgpu_device *adev)
+{
+	return 0;
+}
+static void amdgpu_debugfs_regs_cleanup(struct amdgpu_device *adev) { }
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 5c9918d01bf9..a7189a1fa6a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -136,6 +136,38 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, void *owner,
 }
 
 /**
+ * amdgpu_fence_recreate - recreate a fence from an user fence
+ *
+ * @ring: ring the fence is associated with
+ * @owner: creator of the fence
+ * @seq: user fence sequence number
+ * @fence: resulting amdgpu fence object
+ *
+ * Recreates a fence command from the user fence sequence number (all asics).
+ * Returns 0 on success, -ENOMEM on failure.
+ */
+int amdgpu_fence_recreate(struct amdgpu_ring *ring, void *owner,
+			  uint64_t seq, struct amdgpu_fence **fence)
+{
+	struct amdgpu_device *adev = ring->adev;
+
+	if (seq > ring->fence_drv.sync_seq[ring->idx])
+		return -EINVAL;
+
+	*fence = kmalloc(sizeof(struct amdgpu_fence), GFP_KERNEL);
+	if ((*fence) == NULL)
+		return -ENOMEM;
+
+	(*fence)->seq = seq;
+	(*fence)->ring = ring;
+	(*fence)->owner = owner;
+	fence_init(&(*fence)->base, &amdgpu_fence_ops,
+		&adev->fence_queue.lock, adev->fence_context + ring->idx,
+		(*fence)->seq);
+	return 0;
+}
+
+/**
  * amdgpu_fence_check_signaled - callback from fence_queue
  *
  * this function is called with fence_queue lock held, which is also used
@@ -517,12 +549,14 @@ static bool amdgpu_fence_any_seq_signaled(struct amdgpu_device *adev, u64 *seq)
  * the wait timeout, or an error for all other cases.
  * -EDEADLK is returned when a GPU lockup has been detected.
  */
-long amdgpu_fence_wait_seq_timeout(struct amdgpu_device *adev, u64 *target_seq,
-				   bool intr, long timeout)
+static long amdgpu_fence_wait_seq_timeout(struct amdgpu_device *adev,
+					  u64 *target_seq, bool intr,
+					  long timeout)
 {
 	uint64_t last_seq[AMDGPU_MAX_RINGS];
 	bool signaled;
-	int i, r;
+	int i;
+	long r;
 
 	if (timeout == 0) {
 		return amdgpu_fence_any_seq_signaled(adev, target_seq);
@@ -1023,7 +1057,7 @@ static int amdgpu_debugfs_fence_info(struct seq_file *m, void *data)
 
 		amdgpu_fence_process(ring);
 
-		seq_printf(m, "--- ring %d ---\n", i);
+		seq_printf(m, "--- ring %d (%s) ---\n", i, ring->name);
 		seq_printf(m, "Last signaled fence 0x%016llx\n",
 			   (unsigned long long)atomic64_read(&ring->fence_drv.last_seq));
 		seq_printf(m, "Last emitted        0x%016llx\n",
@@ -1031,7 +1065,8 @@ static int amdgpu_debugfs_fence_info(struct seq_file *m, void *data)
 
 		for (j = 0; j < AMDGPU_MAX_RINGS; ++j) {
 			struct amdgpu_ring *other = adev->rings[j];
-			if (i != j && other && other->fence_drv.initialized)
+			if (i != j && other && other->fence_drv.initialized &&
+			    ring->fence_drv.sync_seq[j])
 				seq_printf(m, "Last sync to ring %d 0x%016llx\n",
 					   j, ring->fence_drv.sync_seq[j]);
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 0ec222295fee..ae43b58c9733 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -352,7 +352,7 @@ unsigned long amdgpu_gem_timeout(uint64_t timeout_ns)
 	if (((int64_t)timeout_ns) < 0)
 		return MAX_SCHEDULE_TIMEOUT;
 
-	timeout = ktime_sub_ns(ktime_get(), timeout_ns);
+	timeout = ktime_sub(ns_to_ktime(timeout_ns), ktime_get());
 	if (ktime_to_ns(timeout) < 0)
 		return 0;
 
@@ -496,7 +496,7 @@ error_unreserve:
 error_free:
 	drm_free_large(vm_bos);
 
-	if (r)
+	if (r && r != -ERESTARTSYS)
 		DRM_ERROR("Couldn't update BO_VA (%d)\n", r);
 }
 
@@ -525,8 +525,8 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
-	invalid_flags = ~(AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_PAGE_WRITEABLE |
-			AMDGPU_VM_PAGE_EXECUTABLE);
+	invalid_flags = ~(AMDGPU_VM_DELAY_UPDATE | AMDGPU_VM_PAGE_READABLE |
+			AMDGPU_VM_PAGE_WRITEABLE | AMDGPU_VM_PAGE_EXECUTABLE);
 	if ((args->flags & invalid_flags)) {
 		dev_err(&dev->pdev->dev, "invalid flags 0x%08X vs 0x%08X\n",
 			args->flags, invalid_flags);
@@ -579,7 +579,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 		break;
 	}
 
-	if (!r)
+	if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE))
 		amdgpu_gem_va_update_vm(adev, bo_va);
 
 	drm_gem_object_unreference_unlocked(gobj);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
index b56dd64bd4ea..961d7265c286 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
@@ -30,19 +30,21 @@ TRACE_EVENT(amdgpu_cs,
 	    TP_PROTO(struct amdgpu_cs_parser *p, int i),
 	    TP_ARGS(p, i),
 	    TP_STRUCT__entry(
+			     __field(struct amdgpu_bo_list *, bo_list)
 			     __field(u32, ring)
 			     __field(u32, dw)
 			     __field(u32, fences)
 			     ),
 
 	    TP_fast_assign(
+			   __entry->bo_list = p->bo_list;
 			   __entry->ring = p->ibs[i].ring->idx;
 			   __entry->dw = p->ibs[i].length_dw;
 			   __entry->fences = amdgpu_fence_count_emitted(
 				p->ibs[i].ring);
 			   ),
-	    TP_printk("ring=%u, dw=%u, fences=%u",
-		      __entry->ring, __entry->dw,
+	    TP_printk("bo_list=%p, ring=%u, dw=%u, fences=%u",
+		      __entry->bo_list, __entry->ring, __entry->dw,
 		      __entry->fences)
 );
 
@@ -61,6 +63,54 @@ TRACE_EVENT(amdgpu_vm_grab_id,
 	    TP_printk("vmid=%u, ring=%u", __entry->vmid, __entry->ring)
 );
 
+TRACE_EVENT(amdgpu_vm_bo_map,
+	    TP_PROTO(struct amdgpu_bo_va *bo_va,
+		     struct amdgpu_bo_va_mapping *mapping),
+	    TP_ARGS(bo_va, mapping),
+	    TP_STRUCT__entry(
+			     __field(struct amdgpu_bo *, bo)
+			     __field(long, start)
+			     __field(long, last)
+			     __field(u64, offset)
+			     __field(u32, flags)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->bo = bo_va->bo;
+			   __entry->start = mapping->it.start;
+			   __entry->last = mapping->it.last;
+			   __entry->offset = mapping->offset;
+			   __entry->flags = mapping->flags;
+			   ),
+	    TP_printk("bo=%p, start=%lx, last=%lx, offset=%010llx, flags=%08x",
+		      __entry->bo, __entry->start, __entry->last,
+		      __entry->offset, __entry->flags)
+);
+
+TRACE_EVENT(amdgpu_vm_bo_unmap,
+	    TP_PROTO(struct amdgpu_bo_va *bo_va,
+		     struct amdgpu_bo_va_mapping *mapping),
+	    TP_ARGS(bo_va, mapping),
+	    TP_STRUCT__entry(
+			     __field(struct amdgpu_bo *, bo)
+			     __field(long, start)
+			     __field(long, last)
+			     __field(u64, offset)
+			     __field(u32, flags)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->bo = bo_va->bo;
+			   __entry->start = mapping->it.start;
+			   __entry->last = mapping->it.last;
+			   __entry->offset = mapping->offset;
+			   __entry->flags = mapping->flags;
+			   ),
+	    TP_printk("bo=%p, start=%lx, last=%lx, offset=%010llx, flags=%08x",
+		      __entry->bo, __entry->start, __entry->last,
+		      __entry->offset, __entry->flags)
+);
+
 TRACE_EVENT(amdgpu_vm_bo_update,
 	    TP_PROTO(struct amdgpu_bo_va_mapping *mapping),
 	    TP_ARGS(mapping),
@@ -121,6 +171,21 @@ TRACE_EVENT(amdgpu_vm_flush,
 		      __entry->pd_addr, __entry->ring, __entry->id)
 );
 
+TRACE_EVENT(amdgpu_bo_list_set,
+	    TP_PROTO(struct amdgpu_bo_list *list, struct amdgpu_bo *bo),
+	    TP_ARGS(list, bo),
+	    TP_STRUCT__entry(
+			     __field(struct amdgpu_bo_list *, list)
+			     __field(struct amdgpu_bo *, bo)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->list = list;
+			   __entry->bo = bo;
+			   ),
+	    TP_printk("list=%p, bo=%p", __entry->list, __entry->bo)
+);
+
 DECLARE_EVENT_CLASS(amdgpu_fence_request,
 
 	    TP_PROTO(struct drm_device *dev, int ring, u32 seqno),
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index d3706a498293..dd3415d2e45d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -674,7 +674,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm)
 		return 0;
 
 	if (gtt && gtt->userptr) {
-		ttm->sg = kcalloc(1, sizeof(struct sg_table), GFP_KERNEL);
+		ttm->sg = kzalloc(sizeof(struct sg_table), GFP_KERNEL);
 		if (!ttm->sg)
 			return -ENOMEM;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
index 1127a504f118..d3ca73090e39 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@@ -464,28 +464,42 @@ int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
  * @p: parser context
  * @lo: address of lower dword
  * @hi: address of higher dword
+ * @size: minimum size
  *
  * Patch relocation inside command stream with real buffer address
  */
-int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, uint32_t ib_idx, int lo, int hi)
+static int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, uint32_t ib_idx,
+			       int lo, int hi, unsigned size, uint32_t index)
 {
 	struct amdgpu_bo_va_mapping *mapping;
 	struct amdgpu_ib *ib = &p->ibs[ib_idx];
 	struct amdgpu_bo *bo;
 	uint64_t addr;
 
+	if (index == 0xffffffff)
+		index = 0;
+
 	addr = ((uint64_t)amdgpu_get_ib_value(p, ib_idx, lo)) |
 	       ((uint64_t)amdgpu_get_ib_value(p, ib_idx, hi)) << 32;
+	addr += ((uint64_t)size) * ((uint64_t)index);
 
 	mapping = amdgpu_cs_find_mapping(p, addr, &bo);
 	if (mapping == NULL) {
-		DRM_ERROR("Can't find BO for addr 0x%010Lx %d %d\n",
+		DRM_ERROR("Can't find BO for addr 0x%010Lx %d %d %d %d\n",
+			  addr, lo, hi, size, index);
+		return -EINVAL;
+	}
+
+	if ((addr + (uint64_t)size) >
+	    ((uint64_t)mapping->it.last + 1) * AMDGPU_GPU_PAGE_SIZE) {
+		DRM_ERROR("BO to small for addr 0x%010Lx %d %d\n",
 			  addr, lo, hi);
 		return -EINVAL;
 	}
 
 	addr -= ((uint64_t)mapping->it.start) * AMDGPU_GPU_PAGE_SIZE;
 	addr += amdgpu_bo_gpu_offset(bo);
+	addr -= ((uint64_t)size) * ((uint64_t)index);
 
 	ib->ptr[lo] = addr & 0xFFFFFFFF;
 	ib->ptr[hi] = addr >> 32;
@@ -494,6 +508,48 @@ int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, uint32_t ib_idx, int lo, int
 }
 
 /**
+ * amdgpu_vce_validate_handle - validate stream handle
+ *
+ * @p: parser context
+ * @handle: handle to validate
+ * @allocated: allocated a new handle?
+ *
+ * Validates the handle and return the found session index or -EINVAL
+ * we we don't have another free session index.
+ */
+static int amdgpu_vce_validate_handle(struct amdgpu_cs_parser *p,
+				      uint32_t handle, bool *allocated)
+{
+	unsigned i;
+
+	*allocated = false;
+
+	/* validate the handle */
+	for (i = 0; i < AMDGPU_MAX_VCE_HANDLES; ++i) {
+		if (atomic_read(&p->adev->vce.handles[i]) == handle) {
+			if (p->adev->vce.filp[i] != p->filp) {
+				DRM_ERROR("VCE handle collision detected!\n");
+				return -EINVAL;
+			}
+			return i;
+		}
+	}
+
+	/* handle not found try to alloc a new one */
+	for (i = 0; i < AMDGPU_MAX_VCE_HANDLES; ++i) {
+		if (!atomic_cmpxchg(&p->adev->vce.handles[i], 0, handle)) {
+			p->adev->vce.filp[i] = p->filp;
+			p->adev->vce.img_size[i] = 0;
+			*allocated = true;
+			return i;
+		}
+	}
+
+	DRM_ERROR("No more free VCE handles!\n");
+	return -EINVAL;
+}
+
+/**
  * amdgpu_vce_cs_parse - parse and validate the command stream
  *
  * @p: parser context
@@ -501,10 +557,15 @@ int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, uint32_t ib_idx, int lo, int
  */
 int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx)
 {
-	uint32_t handle = 0;
-	bool destroy = false;
-	int i, r, idx = 0;
 	struct amdgpu_ib *ib = &p->ibs[ib_idx];
+	unsigned fb_idx = 0, bs_idx = 0;
+	int session_idx = -1;
+	bool destroyed = false;
+	bool created = false;
+	bool allocated = false;
+	uint32_t tmp, handle = 0;
+	uint32_t *size = &tmp;
+	int i, r = 0, idx = 0;
 
 	amdgpu_vce_note_usage(p->adev);
 
@@ -514,16 +575,44 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx)
 
 		if ((len < 8) || (len & 3)) {
 			DRM_ERROR("invalid VCE command length (%d)!\n", len);
-			return -EINVAL;
+			r = -EINVAL;
+			goto out;
+		}
+
+		if (destroyed) {
+			DRM_ERROR("No other command allowed after destroy!\n");
+			r = -EINVAL;
+			goto out;
 		}
 
 		switch (cmd) {
 		case 0x00000001: // session
 			handle = amdgpu_get_ib_value(p, ib_idx, idx + 2);
+			session_idx = amdgpu_vce_validate_handle(p, handle,
+								 &allocated);
+			if (session_idx < 0)
+				return session_idx;
+			size = &p->adev->vce.img_size[session_idx];
 			break;
 
 		case 0x00000002: // task info
+			fb_idx = amdgpu_get_ib_value(p, ib_idx, idx + 6);
+			bs_idx = amdgpu_get_ib_value(p, ib_idx, idx + 7);
+			break;
+
 		case 0x01000001: // create
+			created = true;
+			if (!allocated) {
+				DRM_ERROR("Handle already in use!\n");
+				r = -EINVAL;
+				goto out;
+			}
+
+			*size = amdgpu_get_ib_value(p, ib_idx, idx + 8) *
+				amdgpu_get_ib_value(p, ib_idx, idx + 10) *
+				8 * 3 / 2;
+			break;
+
 		case 0x04000001: // config extension
 		case 0x04000002: // pic control
 		case 0x04000005: // rate control
@@ -534,60 +623,74 @@ int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx)
 			break;
 
 		case 0x03000001: // encode
-			r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 10, idx + 9);
+			r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 10, idx + 9,
+						*size, 0);
 			if (r)
-				return r;
+				goto out;
 
-			r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 12, idx + 11);
+			r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 12, idx + 11,
+						*size / 3, 0);
 			if (r)
-				return r;
+				goto out;
 			break;
 
 		case 0x02000001: // destroy
-			destroy = true;
+			destroyed = true;
 			break;
 
 		case 0x05000001: // context buffer
+			r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 3, idx + 2,
+						*size * 2, 0);
+			if (r)
+				goto out;
+			break;
+
 		case 0x05000004: // video bitstream buffer
+			tmp = amdgpu_get_ib_value(p, ib_idx, idx + 4);
+			r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 3, idx + 2,
+						tmp, bs_idx);
+			if (r)
+				goto out;
+			break;
+
 		case 0x05000005: // feedback buffer
-			r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 3, idx + 2);
+			r = amdgpu_vce_cs_reloc(p, ib_idx, idx + 3, idx + 2,
+						4096, fb_idx);
 			if (r)
-				return r;
+				goto out;
 			break;
 
 		default:
 			DRM_ERROR("invalid VCE command (0x%x)!\n", cmd);
-			return -EINVAL;
+			r = -EINVAL;
+			goto out;
 		}
 
-		idx += len / 4;
-	}
-
-	if (destroy) {
-		/* IB contains a destroy msg, free the handle */
-		for (i = 0; i < AMDGPU_MAX_VCE_HANDLES; ++i)
-			atomic_cmpxchg(&p->adev->vce.handles[i], handle, 0);
+		if (session_idx == -1) {
+			DRM_ERROR("no session command at start of IB\n");
+			r = -EINVAL;
+			goto out;
+		}
 
-		return 0;
+		idx += len / 4;
 	}
 
-	/* create or encode, validate the handle */
-	for (i = 0; i < AMDGPU_MAX_VCE_HANDLES; ++i) {
-		if (atomic_read(&p->adev->vce.handles[i]) == handle)
-			return 0;
+	if (allocated && !created) {
+		DRM_ERROR("New session without create command!\n");
+		r = -ENOENT;
 	}
 
-	/* handle not found try to alloc a new one */
-	for (i = 0; i < AMDGPU_MAX_VCE_HANDLES; ++i) {
-		if (!atomic_cmpxchg(&p->adev->vce.handles[i], 0, handle)) {
-			p->adev->vce.filp[i] = p->filp;
-			return 0;
-		}
+out:
+	if ((!r && destroyed) || (r && allocated)) {
+		/*
+		 * IB contains a destroy msg or we have allocated an
+		 * handle and got an error, anyway free the handle
+		 */
+		for (i = 0; i < AMDGPU_MAX_VCE_HANDLES; ++i)
+			atomic_cmpxchg(&p->adev->vce.handles[i], handle, 0);
 	}
 
-	DRM_ERROR("No more free VCE handles!\n");
-
-	return -EINVAL;
+	return r;
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
index b6a9d0956c60..7ccdb5927da5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
@@ -33,7 +33,6 @@ int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
 int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
 			       struct amdgpu_fence **fence);
 void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp);
-int amdgpu_vce_cs_reloc(struct amdgpu_cs_parser *p, uint32_t ib_idx, int lo, int hi);
 int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx);
 bool amdgpu_vce_ring_emit_semaphore(struct amdgpu_ring *ring,
 				    struct amdgpu_semaphore *semaphore,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 407882b233c7..9a4e3b63f1cb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -1001,6 +1001,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 
 	list_add(&mapping->list, &bo_va->mappings);
 	interval_tree_insert(&mapping->it, &vm->va);
+	trace_amdgpu_vm_bo_map(bo_va, mapping);
 
 	bo_va->addr = 0;
 
@@ -1058,6 +1059,7 @@ error_free:
 	mutex_lock(&vm->mutex);
 	list_del(&mapping->list);
 	interval_tree_remove(&mapping->it, &vm->va);
+	trace_amdgpu_vm_bo_unmap(bo_va, mapping);
 	kfree(mapping);
 
 error_unlock:
@@ -1099,6 +1101,7 @@ int amdgpu_vm_bo_unmap(struct amdgpu_device *adev,
 	mutex_lock(&vm->mutex);
 	list_del(&mapping->list);
 	interval_tree_remove(&mapping->it, &vm->va);
+	trace_amdgpu_vm_bo_unmap(bo_va, mapping);
 
 	if (bo_va->addr) {
 		/* clear the old address */
@@ -1139,6 +1142,7 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev,
 	list_for_each_entry_safe(mapping, next, &bo_va->mappings, list) {
 		list_del(&mapping->list);
 		interval_tree_remove(&mapping->it, &vm->va);
+		trace_amdgpu_vm_bo_unmap(bo_va, mapping);
 		if (bo_va->addr)
 			list_add(&mapping->list, &vm->freed);
 		else
diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c
index 5dab578d6462..341c56681841 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik.c
@@ -2256,10 +2256,6 @@ int cik_set_ip_blocks(struct amdgpu_device *adev)
 		return -EINVAL;
 	}
 
-	adev->ip_block_enabled = kcalloc(adev->num_ip_blocks, sizeof(bool), GFP_KERNEL);
-	if (adev->ip_block_enabled == NULL)
-		return -ENOMEM;
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/cikd.h b/drivers/gpu/drm/amd/amdgpu/cikd.h
index 220865a44814..d19085a97064 100644
--- a/drivers/gpu/drm/amd/amdgpu/cikd.h
+++ b/drivers/gpu/drm/amd/amdgpu/cikd.h
@@ -552,4 +552,10 @@
 #define VCE_CMD_IB_AUTO		0x00000005
 #define VCE_CMD_SEMAPHORE	0x00000006
 
+/* valid for both DEFAULT_MTYPE and APE1_MTYPE */
+enum {
+	MTYPE_CACHED = 0,
+	MTYPE_NONCACHED = 3
+};
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
index e4936a452bc6..f75a31df30bd 100644
--- a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
@@ -425,7 +425,7 @@ static int cz_dpm_init(struct amdgpu_device *adev)
 	pi->mgcg_cgtt_local1 = 0x0;
 	pi->clock_slow_down_step = 25000;
 	pi->skip_clock_slow_down = 1;
-	pi->enable_nb_ps_policy = 1;
+	pi->enable_nb_ps_policy = 0;
 	pi->caps_power_containment = true;
 	pi->caps_cac = true;
 	pi->didt_enabled = false;
diff --git a/drivers/gpu/drm/amd/amdgpu/cz_dpm.h b/drivers/gpu/drm/amd/amdgpu/cz_dpm.h
index 782a74107664..99e1afc89629 100644
--- a/drivers/gpu/drm/amd/amdgpu/cz_dpm.h
+++ b/drivers/gpu/drm/amd/amdgpu/cz_dpm.h
@@ -46,7 +46,7 @@
 
 /* Do not change the following, it is also defined in SMU8.h */
 #define SMU_EnabledFeatureScoreboard_AcpDpmOn		0x00000001
-#define SMU_EnabledFeatureScoreboard_SclkDpmOn		0x00100000
+#define SMU_EnabledFeatureScoreboard_SclkDpmOn		0x00200000
 #define SMU_EnabledFeatureScoreboard_UvdDpmOn		0x00800000
 #define SMU_EnabledFeatureScoreboard_VceDpmOn		0x01000000
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 5cde635978f9..6e77964f1b64 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -3403,19 +3403,25 @@ static int dce_v10_0_crtc_irq(struct amdgpu_device *adev,
 
 	switch (entry->src_data) {
 	case 0: /* vblank */
-		if (disp_int & interrupt_status_offsets[crtc].vblank) {
+		if (disp_int & interrupt_status_offsets[crtc].vblank)
 			dce_v10_0_crtc_vblank_int_ack(adev, crtc);
-			if (amdgpu_irq_enabled(adev, source, irq_type)) {
-				drm_handle_vblank(adev->ddev, crtc);
-			}
-			DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
+		else
+			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+		if (amdgpu_irq_enabled(adev, source, irq_type)) {
+			drm_handle_vblank(adev->ddev, crtc);
 		}
+		DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
+
 		break;
 	case 1: /* vline */
-		if (disp_int & interrupt_status_offsets[crtc].vline) {
+		if (disp_int & interrupt_status_offsets[crtc].vline)
 			dce_v10_0_crtc_vline_int_ack(adev, crtc);
-			DRM_DEBUG("IH: D%d vline\n", crtc + 1);
-		}
+		else
+			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+		DRM_DEBUG("IH: D%d vline\n", crtc + 1);
+
 		break;
 	default:
 		DRM_DEBUG("Unhandled interrupt: %d %d\n", entry->src_id, entry->src_data);
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 95efd98b202d..7f7abb0e0be5 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -3402,19 +3402,25 @@ static int dce_v11_0_crtc_irq(struct amdgpu_device *adev,
 
 	switch (entry->src_data) {
 	case 0: /* vblank */
-		if (disp_int & interrupt_status_offsets[crtc].vblank) {
+		if (disp_int & interrupt_status_offsets[crtc].vblank)
 			dce_v11_0_crtc_vblank_int_ack(adev, crtc);
-			if (amdgpu_irq_enabled(adev, source, irq_type)) {
-				drm_handle_vblank(adev->ddev, crtc);
-			}
-			DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
+		else
+			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+		if (amdgpu_irq_enabled(adev, source, irq_type)) {
+			drm_handle_vblank(adev->ddev, crtc);
 		}
+		DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
+
 		break;
 	case 1: /* vline */
-		if (disp_int & interrupt_status_offsets[crtc].vline) {
+		if (disp_int & interrupt_status_offsets[crtc].vline)
 			dce_v11_0_crtc_vline_int_ack(adev, crtc);
-			DRM_DEBUG("IH: D%d vline\n", crtc + 1);
-		}
+		else
+			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+		DRM_DEBUG("IH: D%d vline\n", crtc + 1);
+
 		break;
 	default:
 		DRM_DEBUG("Unhandled interrupt: %d %d\n", entry->src_id, entry->src_data);
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index 72c27ac915f2..08387dfd98a7 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -3237,19 +3237,25 @@ static int dce_v8_0_crtc_irq(struct amdgpu_device *adev,
 
 	switch (entry->src_data) {
 	case 0: /* vblank */
-		if (disp_int & interrupt_status_offsets[crtc].vblank) {
+		if (disp_int & interrupt_status_offsets[crtc].vblank)
 			WREG32(mmLB_VBLANK_STATUS + crtc_offsets[crtc], LB_VBLANK_STATUS__VBLANK_ACK_MASK);
-			if (amdgpu_irq_enabled(adev, source, irq_type)) {
-				drm_handle_vblank(adev->ddev, crtc);
-			}
-			DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
+		else
+			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+		if (amdgpu_irq_enabled(adev, source, irq_type)) {
+			drm_handle_vblank(adev->ddev, crtc);
 		}
+		DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
+
 		break;
 	case 1: /* vline */
-		if (disp_int & interrupt_status_offsets[crtc].vline) {
+		if (disp_int & interrupt_status_offsets[crtc].vline)
 			WREG32(mmLB_VLINE_STATUS + crtc_offsets[crtc], LB_VLINE_STATUS__VLINE_ACK_MASK);
-			DRM_DEBUG("IH: D%d vline\n", crtc + 1);
-		}
+		else
+			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+		DRM_DEBUG("IH: D%d vline\n", crtc + 1);
+
 		break;
 	default:
 		DRM_DEBUG("Unhandled interrupt: %d %d\n", entry->src_id, entry->src_data);
@@ -3379,7 +3385,7 @@ static int dce_v8_0_hpd_irq(struct amdgpu_device *adev,
 	uint32_t disp_int, mask, int_control, tmp;
 	unsigned hpd;
 
-	if (entry->src_data > 6) {
+	if (entry->src_data >= adev->mode_info.num_hpd) {
 		DRM_DEBUG("Unhandled interrupt: %d %d\n", entry->src_id, entry->src_data);
 		return 0;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index cb7907447b81..2c188fb9fd22 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -2010,6 +2010,46 @@ static void gfx_v7_0_setup_rb(struct amdgpu_device *adev,
 }
 
 /**
+ * gmc_v7_0_init_compute_vmid - gart enable
+ *
+ * @rdev: amdgpu_device pointer
+ *
+ * Initialize compute vmid sh_mem registers
+ *
+ */
+#define DEFAULT_SH_MEM_BASES	(0x6000)
+#define FIRST_COMPUTE_VMID	(8)
+#define LAST_COMPUTE_VMID	(16)
+static void gmc_v7_0_init_compute_vmid(struct amdgpu_device *adev)
+{
+	int i;
+	uint32_t sh_mem_config;
+	uint32_t sh_mem_bases;
+
+	/*
+	 * Configure apertures:
+	 * LDS:         0x60000000'00000000 - 0x60000001'00000000 (4GB)
+	 * Scratch:     0x60000001'00000000 - 0x60000002'00000000 (4GB)
+	 * GPUVM:       0x60010000'00000000 - 0x60020000'00000000 (1TB)
+	*/
+	sh_mem_bases = DEFAULT_SH_MEM_BASES | (DEFAULT_SH_MEM_BASES << 16);
+	sh_mem_config = SH_MEM_ALIGNMENT_MODE_UNALIGNED <<
+			SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT;
+	sh_mem_config |= MTYPE_NONCACHED << SH_MEM_CONFIG__DEFAULT_MTYPE__SHIFT;
+	mutex_lock(&adev->srbm_mutex);
+	for (i = FIRST_COMPUTE_VMID; i < LAST_COMPUTE_VMID; i++) {
+		cik_srbm_select(adev, 0, 0, 0, i);
+		/* CP and shaders */
+		WREG32(mmSH_MEM_CONFIG, sh_mem_config);
+		WREG32(mmSH_MEM_APE1_BASE, 1);
+		WREG32(mmSH_MEM_APE1_LIMIT, 0);
+		WREG32(mmSH_MEM_BASES, sh_mem_bases);
+	}
+	cik_srbm_select(adev, 0, 0, 0, 0);
+	mutex_unlock(&adev->srbm_mutex);
+}
+
+/**
  * gfx_v7_0_gpu_init - setup the 3D engine
  *
  * @adev: amdgpu_device pointer
@@ -2230,6 +2270,8 @@ static void gfx_v7_0_gpu_init(struct amdgpu_device *adev)
 	cik_srbm_select(adev, 0, 0, 0, 0);
 	mutex_unlock(&adev->srbm_mutex);
 
+	gmc_v7_0_init_compute_vmid(adev);
+
 	WREG32(mmSX_DEBUG_1, 0x20);
 
 	WREG32(mmTA_CNTL_AUX, 0x00010000);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 14242bd33363..7b683fb2173c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -1894,6 +1894,51 @@ static void gfx_v8_0_setup_rb(struct amdgpu_device *adev,
 	mutex_unlock(&adev->grbm_idx_mutex);
 }
 
+/**
+ * gmc_v8_0_init_compute_vmid - gart enable
+ *
+ * @rdev: amdgpu_device pointer
+ *
+ * Initialize compute vmid sh_mem registers
+ *
+ */
+#define DEFAULT_SH_MEM_BASES	(0x6000)
+#define FIRST_COMPUTE_VMID	(8)
+#define LAST_COMPUTE_VMID	(16)
+static void gmc_v8_0_init_compute_vmid(struct amdgpu_device *adev)
+{
+	int i;
+	uint32_t sh_mem_config;
+	uint32_t sh_mem_bases;
+
+	/*
+	 * Configure apertures:
+	 * LDS:         0x60000000'00000000 - 0x60000001'00000000 (4GB)
+	 * Scratch:     0x60000001'00000000 - 0x60000002'00000000 (4GB)
+	 * GPUVM:       0x60010000'00000000 - 0x60020000'00000000 (1TB)
+	 */
+	sh_mem_bases = DEFAULT_SH_MEM_BASES | (DEFAULT_SH_MEM_BASES << 16);
+
+	sh_mem_config = SH_MEM_ADDRESS_MODE_HSA64 <<
+			SH_MEM_CONFIG__ADDRESS_MODE__SHIFT |
+			SH_MEM_ALIGNMENT_MODE_UNALIGNED <<
+			SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT |
+			MTYPE_CC << SH_MEM_CONFIG__DEFAULT_MTYPE__SHIFT |
+			SH_MEM_CONFIG__PRIVATE_ATC_MASK;
+
+	mutex_lock(&adev->srbm_mutex);
+	for (i = FIRST_COMPUTE_VMID; i < LAST_COMPUTE_VMID; i++) {
+		vi_srbm_select(adev, 0, 0, 0, i);
+		/* CP and shaders */
+		WREG32(mmSH_MEM_CONFIG, sh_mem_config);
+		WREG32(mmSH_MEM_APE1_BASE, 1);
+		WREG32(mmSH_MEM_APE1_LIMIT, 0);
+		WREG32(mmSH_MEM_BASES, sh_mem_bases);
+	}
+	vi_srbm_select(adev, 0, 0, 0, 0);
+	mutex_unlock(&adev->srbm_mutex);
+}
+
 static void gfx_v8_0_gpu_init(struct amdgpu_device *adev)
 {
 	u32 gb_addr_config;
@@ -2113,6 +2158,8 @@ static void gfx_v8_0_gpu_init(struct amdgpu_device *adev)
 	vi_srbm_select(adev, 0, 0, 0, 0);
 	mutex_unlock(&adev->srbm_mutex);
 
+	gmc_v8_0_init_compute_vmid(adev);
+
 	mutex_lock(&adev->grbm_idx_mutex);
 	/*
 	 * making sure that the following register writes will be broadcasted
@@ -3081,7 +3128,7 @@ static int gfx_v8_0_cp_compute_resume(struct amdgpu_device *adev)
 				WREG32(mmCP_MEC_DOORBELL_RANGE_LOWER,
 				       AMDGPU_DOORBELL_KIQ << 2);
 				WREG32(mmCP_MEC_DOORBELL_RANGE_UPPER,
-				       AMDGPU_DOORBELL_MEC_RING7 << 2);
+						0x7FFFF << 2);
 			}
 			tmp = RREG32(mmCP_HQD_PQ_DOORBELL_CONTROL);
 			tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL,
@@ -3097,6 +3144,12 @@ static int gfx_v8_0_cp_compute_resume(struct amdgpu_device *adev)
 		WREG32(mmCP_HQD_PQ_DOORBELL_CONTROL,
 		       mqd->cp_hqd_pq_doorbell_control);
 
+		/* reset read and write pointers, similar to CP_RB0_WPTR/_RPTR */
+		ring->wptr = 0;
+		mqd->cp_hqd_pq_wptr = ring->wptr;
+		WREG32(mmCP_HQD_PQ_WPTR, mqd->cp_hqd_pq_wptr);
+		mqd->cp_hqd_pq_rptr = RREG32(mmCP_HQD_PQ_RPTR);
+
 		/* set the vmid for the queue */
 		mqd->cp_hqd_vmid = 0;
 		WREG32(mmCP_HQD_VMID, mqd->cp_hqd_vmid);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index e3c1fde75363..7bb37b93993f 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -439,6 +439,31 @@ static void sdma_v3_0_rlc_stop(struct amdgpu_device *adev)
 }
 
 /**
+ * sdma_v3_0_ctx_switch_enable - stop the async dma engines context switch
+ *
+ * @adev: amdgpu_device pointer
+ * @enable: enable/disable the DMA MEs context switch.
+ *
+ * Halt or unhalt the async dma engines context switch (VI).
+ */
+static void sdma_v3_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable)
+{
+	u32 f32_cntl;
+	int i;
+
+	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+		f32_cntl = RREG32(mmSDMA0_CNTL + sdma_offsets[i]);
+		if (enable)
+			f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL,
+					AUTO_CTXSW_ENABLE, 1);
+		else
+			f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL,
+					AUTO_CTXSW_ENABLE, 0);
+		WREG32(mmSDMA0_CNTL + sdma_offsets[i], f32_cntl);
+	}
+}
+
+/**
  * sdma_v3_0_enable - stop the async dma engines
  *
  * @adev: amdgpu_device pointer
@@ -648,6 +673,8 @@ static int sdma_v3_0_start(struct amdgpu_device *adev)
 
 	/* unhalt the MEs */
 	sdma_v3_0_enable(adev, true);
+	/* enable sdma ring preemption */
+	sdma_v3_0_ctx_switch_enable(adev, true);
 
 	/* start the gfx rings and rlc compute queues */
 	r = sdma_v3_0_gfx_resume(adev);
@@ -1079,6 +1106,7 @@ static int sdma_v3_0_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	sdma_v3_0_ctx_switch_enable(adev, false);
 	sdma_v3_0_enable(adev, false);
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 90fc93c2c1d0..fa5a4448531d 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -1189,10 +1189,6 @@ int vi_set_ip_blocks(struct amdgpu_device *adev)
 		return -EINVAL;
 	}
 
-	adev->ip_block_enabled = kcalloc(adev->num_ip_blocks, sizeof(bool), GFP_KERNEL);
-	if (adev->ip_block_enabled == NULL)
-		return -ENOMEM;
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 8a1f999daa24..9be007081b72 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -420,6 +420,12 @@ void kfd_unbind_process_from_device(struct kfd_dev *dev, unsigned int pasid)
 	pqm_uninit(&p->pqm);
 
 	pdd = kfd_get_process_device_data(dev, p);
+
+	if (!pdd) {
+		mutex_unlock(&p->mutex);
+		return;
+	}
+
 	if (pdd->reset_wavefronts) {
 		dbgdev_wave_reset_wavefronts(pdd->dev, p);
 		pdd->reset_wavefronts = false;
@@ -431,8 +437,7 @@ void kfd_unbind_process_from_device(struct kfd_dev *dev, unsigned int pasid)
 	 * We don't call amd_iommu_unbind_pasid() here
 	 * because the IOMMU called us.
 	 */
-	if (pdd)
-		pdd->bound = false;
+	pdd->bound = false;
 
 	mutex_unlock(&p->mutex);
 }
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index e469c4b2e8cc..c25728bc388a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -684,8 +684,6 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 			dev->node_props.cpu_core_id_base);
 	sysfs_show_32bit_prop(buffer, "simd_id_base",
 			dev->node_props.simd_id_base);
-	sysfs_show_32bit_prop(buffer, "capability",
-			dev->node_props.capability);
 	sysfs_show_32bit_prop(buffer, "max_waves_per_simd",
 			dev->node_props.max_waves_per_simd);
 	sysfs_show_32bit_prop(buffer, "lds_size_in_kb",
@@ -736,6 +734,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 			dev->gpu->kfd2kgd->get_fw_version(
 						dev->gpu->kgd,
 						KGD_ENGINE_MEC1));
+		sysfs_show_32bit_prop(buffer, "capability",
+				dev->node_props.capability);
 	}
 
 	return sysfs_show_32bit_prop(buffer, "max_engine_clk_ccompute",
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 79fe31e5851e..9c978d9de3b8 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -4730,7 +4730,7 @@ int drm_mode_connector_update_edid_property(struct drm_connector *connector,
 		return 0;
 
 	if (edid)
-		size = EDID_LENGTH + (1 + edid->extensions);
+		size = EDID_LENGTH * (1 + edid->extensions);
 
 	ret = drm_property_replace_global_blob(dev,
 					       &connector->edid_blob_ptr,
diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c
index f7213497ad6c..778bbb6425b8 100644
--- a/drivers/gpu/drm/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/drm_dp_mst_topology.c
@@ -867,8 +867,16 @@ static void drm_dp_destroy_port(struct kref *kref)
 		port->vcpi.num_slots = 0;
 
 		kfree(port->cached_edid);
-		if (port->connector)
-			(*port->mgr->cbs->destroy_connector)(mgr, port->connector);
+
+		/* we can't destroy the connector here, as
+		   we might be holding the mode_config.mutex
+		   from an EDID retrieval */
+		if (port->connector) {
+			mutex_lock(&mgr->destroy_connector_lock);
+			list_add(&port->connector->destroy_list, &mgr->destroy_connector_list);
+			mutex_unlock(&mgr->destroy_connector_lock);
+			schedule_work(&mgr->destroy_connector_work);
+		}
 		drm_dp_port_teardown_pdt(port, port->pdt);
 
 		if (!port->input && port->vcpi.vcpi > 0)
@@ -2649,6 +2657,30 @@ static void drm_dp_tx_work(struct work_struct *work)
 	mutex_unlock(&mgr->qlock);
 }
 
+static void drm_dp_destroy_connector_work(struct work_struct *work)
+{
+	struct drm_dp_mst_topology_mgr *mgr = container_of(work, struct drm_dp_mst_topology_mgr, destroy_connector_work);
+	struct drm_connector *connector;
+
+	/*
+	 * Not a regular list traverse as we have to drop the destroy
+	 * connector lock before destroying the connector, to avoid AB->BA
+	 * ordering between this lock and the config mutex.
+	 */
+	for (;;) {
+		mutex_lock(&mgr->destroy_connector_lock);
+		connector = list_first_entry_or_null(&mgr->destroy_connector_list, struct drm_connector, destroy_list);
+		if (!connector) {
+			mutex_unlock(&mgr->destroy_connector_lock);
+			break;
+		}
+		list_del(&connector->destroy_list);
+		mutex_unlock(&mgr->destroy_connector_lock);
+
+		mgr->cbs->destroy_connector(mgr, connector);
+	}
+}
+
 /**
  * drm_dp_mst_topology_mgr_init - initialise a topology manager
  * @mgr: manager struct to initialise
@@ -2668,10 +2700,13 @@ int drm_dp_mst_topology_mgr_init(struct drm_dp_mst_topology_mgr *mgr,
 	mutex_init(&mgr->lock);
 	mutex_init(&mgr->qlock);
 	mutex_init(&mgr->payload_lock);
+	mutex_init(&mgr->destroy_connector_lock);
 	INIT_LIST_HEAD(&mgr->tx_msg_upq);
 	INIT_LIST_HEAD(&mgr->tx_msg_downq);
+	INIT_LIST_HEAD(&mgr->destroy_connector_list);
 	INIT_WORK(&mgr->work, drm_dp_mst_link_probe_work);
 	INIT_WORK(&mgr->tx_work, drm_dp_tx_work);
+	INIT_WORK(&mgr->destroy_connector_work, drm_dp_destroy_connector_work);
 	init_waitqueue_head(&mgr->tx_waitq);
 	mgr->dev = dev;
 	mgr->aux = aux;
@@ -2696,6 +2731,7 @@ EXPORT_SYMBOL(drm_dp_mst_topology_mgr_init);
  */
 void drm_dp_mst_topology_mgr_destroy(struct drm_dp_mst_topology_mgr *mgr)
 {
+	flush_work(&mgr->destroy_connector_work);
 	mutex_lock(&mgr->payload_lock);
 	kfree(mgr->payloads);
 	mgr->payloads = NULL;
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 3b3c4f537e95..b7bf4ce8c012 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -46,13 +46,11 @@ MODULE_AUTHOR(CORE_AUTHOR);
 MODULE_DESCRIPTION(CORE_DESC);
 MODULE_LICENSE("GPL and additional rights");
 MODULE_PARM_DESC(debug, "Enable debug output");
-MODULE_PARM_DESC(atomic, "Enable experimental atomic KMS API");
 MODULE_PARM_DESC(vblankoffdelay, "Delay until vblank irq auto-disable [msecs] (0: never disable, <0: disable immediately)");
 MODULE_PARM_DESC(timestamp_precision_usec, "Max. error on timestamps [usecs]");
 MODULE_PARM_DESC(timestamp_monotonic, "Use monotonic timestamps");
 
 module_param_named(debug, drm_debug, int, 0600);
-module_param_named_unsafe(atomic, drm_atomic, bool, 0600);
 
 static DEFINE_SPINLOCK(drm_minor_lock);
 static struct idr drm_minors_idr;
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index 9bac1b7479af..b1d303fa2327 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -36,9 +36,6 @@
 
 #include <linux/pci.h>
 #include <linux/export.h>
-#ifdef CONFIG_X86
-#include <asm/mtrr.h>
-#endif
 
 static int drm_version(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv);
@@ -197,16 +194,7 @@ static int drm_getmap(struct drm_device *dev, void *data,
 	map->type = r_list->map->type;
 	map->flags = r_list->map->flags;
 	map->handle = (void *)(unsigned long) r_list->user_token;
-
-#ifdef CONFIG_X86
-	/*
-	 * There appears to be exactly one user of the mtrr index: dritest.
-	 * It's easy enough to keep it working on non-PAT systems.
-	 */
-	map->mtrr = phys_wc_to_mtrr_index(r_list->map->mtrr);
-#else
-	map->mtrr = -1;
-#endif
+	map->mtrr = arch_phys_wc_index(r_list->map->mtrr);
 
 	mutex_unlock(&dev->struct_mutex);
 
@@ -350,9 +338,6 @@ drm_setclientcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
 		file_priv->universal_planes = req->value;
 		break;
 	case DRM_CLIENT_CAP_ATOMIC:
-		/* for now, hide behind experimental drm.atomic moduleparam */
-		if (!drm_atomic)
-			return -EINVAL;
 		if (!drm_core_check_feature(dev, DRIVER_ATOMIC))
 			return -EINVAL;
 		if (req->value > 1)
diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index 487ddf5ffe51..0f6cd33b531f 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -217,7 +217,7 @@ static ssize_t status_store(struct device *device,
 
 	mutex_unlock(&dev->mode_config.mutex);
 
-	return ret;
+	return ret ? ret : count;
 }
 
 static ssize_t status_show(struct device *device,
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index f3f014df61a3..bc817da9fef7 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -1732,12 +1732,15 @@ static int i915_sr_status(struct seq_file *m, void *unused)
 
 	if (HAS_PCH_SPLIT(dev))
 		sr_enabled = I915_READ(WM1_LP_ILK) & WM1_LP_SR_EN;
-	else if (IS_CRESTLINE(dev) || IS_I945G(dev) || IS_I945GM(dev))
+	else if (IS_CRESTLINE(dev) || IS_G4X(dev) ||
+		 IS_I945G(dev) || IS_I945GM(dev))
 		sr_enabled = I915_READ(FW_BLC_SELF) & FW_BLC_SELF_EN;
 	else if (IS_I915GM(dev))
 		sr_enabled = I915_READ(INSTPM) & INSTPM_SELF_EN;
 	else if (IS_PINEVIEW(dev))
 		sr_enabled = I915_READ(DSPFW3) & PINEVIEW_SELF_REFRESH_EN;
+	else if (IS_VALLEYVIEW(dev))
+		sr_enabled = I915_READ(FW_BLC_SELF_VLV) & FW_CSPWRDWNEN;
 
 	intel_runtime_pm_put(dev_priv);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4af33d59d507..01fbdc57462a 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -866,6 +866,7 @@ struct intel_context {
 	struct kref ref;
 	int user_handle;
 	uint8_t remap_slice;
+	struct drm_i915_private *i915;
 	int flags;
 	struct drm_i915_file_private *file_priv;
 	struct i915_ctx_hang_stats hang_stats;
@@ -2073,8 +2074,6 @@ struct drm_i915_gem_object {
 	unsigned int cache_level:3;
 	unsigned int cache_dirty:1;
 
-	unsigned int has_dma_mapping:1;
-
 	unsigned int frontbuffer_bits:INTEL_FRONTBUFFER_BITS;
 
 	unsigned int pin_display;
@@ -3173,7 +3172,8 @@ void i915_debugfs_cleanup(struct drm_minor *minor);
 int i915_debugfs_connector_add(struct drm_connector *connector);
 void intel_display_crc_init(struct drm_device *dev);
 #else
-static inline int i915_debugfs_connector_add(struct drm_connector *connector) {}
+static inline int i915_debugfs_connector_add(struct drm_connector *connector)
+{ return 0; }
 static inline void intel_display_crc_init(struct drm_device *dev) {}
 #endif
 
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index aaabf3c259d5..d9f2701b4593 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -217,7 +217,6 @@ i915_gem_object_get_pages_phys(struct drm_i915_gem_object *obj)
 	sg_dma_len(sg) = obj->base.size;
 
 	obj->pages = st;
-	obj->has_dma_mapping = true;
 	return 0;
 }
 
@@ -269,8 +268,6 @@ i915_gem_object_put_pages_phys(struct drm_i915_gem_object *obj)
 
 	sg_free_table(obj->pages);
 	kfree(obj->pages);
-
-	obj->has_dma_mapping = false;
 }
 
 static void
@@ -2146,6 +2143,8 @@ i915_gem_object_put_pages_gtt(struct drm_i915_gem_object *obj)
 		obj->base.read_domains = obj->base.write_domain = I915_GEM_DOMAIN_CPU;
 	}
 
+	i915_gem_gtt_finish_object(obj);
+
 	if (i915_gem_object_needs_bit17_swizzle(obj))
 		i915_gem_object_save_bit_17_swizzle(obj);
 
@@ -2206,6 +2205,7 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 	struct sg_page_iter sg_iter;
 	struct page *page;
 	unsigned long last_pfn = 0;	/* suppress gcc warning */
+	int ret;
 	gfp_t gfp;
 
 	/* Assert that the object is not currently in any GPU domain. As it
@@ -2253,8 +2253,10 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 			 */
 			i915_gem_shrink_all(dev_priv);
 			page = shmem_read_mapping_page(mapping, i);
-			if (IS_ERR(page))
+			if (IS_ERR(page)) {
+				ret = PTR_ERR(page);
 				goto err_pages;
+			}
 		}
 #ifdef CONFIG_SWIOTLB
 		if (swiotlb_nr_tbl()) {
@@ -2283,6 +2285,10 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj)
 		sg_mark_end(sg);
 	obj->pages = st;
 
+	ret = i915_gem_gtt_prepare_object(obj);
+	if (ret)
+		goto err_pages;
+
 	if (i915_gem_object_needs_bit17_swizzle(obj))
 		i915_gem_object_do_bit_17_swizzle(obj);
 
@@ -2307,10 +2313,10 @@ err_pages:
 	 * space and so want to translate the error from shmemfs back to our
 	 * usual understanding of ENOMEM.
 	 */
-	if (PTR_ERR(page) == -ENOSPC)
-		return -ENOMEM;
-	else
-		return PTR_ERR(page);
+	if (ret == -ENOSPC)
+		ret = -ENOMEM;
+
+	return ret;
 }
 
 /* Ensure that the associated pages are gathered from the backing storage
@@ -2549,6 +2555,7 @@ void __i915_add_request(struct drm_i915_gem_request *request,
 	request->batch_obj = obj;
 
 	request->emitted_jiffies = jiffies;
+	ring->last_submitted_seqno = request->seqno;
 	list_add_tail(&request->list, &ring->request_list);
 
 	trace_i915_gem_request_add(request);
@@ -3279,8 +3286,8 @@ int i915_vma_unbind(struct i915_vma *vma)
 		} else if (vma->ggtt_view.pages) {
 			sg_free_table(vma->ggtt_view.pages);
 			kfree(vma->ggtt_view.pages);
-			vma->ggtt_view.pages = NULL;
 		}
+		vma->ggtt_view.pages = NULL;
 	}
 
 	drm_mm_remove_node(&vma->node);
@@ -3288,10 +3295,8 @@ int i915_vma_unbind(struct i915_vma *vma)
 
 	/* Since the unbound list is global, only move to that list if
 	 * no more VMAs exist. */
-	if (list_empty(&obj->vma_list)) {
-		i915_gem_gtt_finish_object(obj);
+	if (list_empty(&obj->vma_list))
 		list_move_tail(&obj->global_list, &dev_priv->mm.unbound_list);
-	}
 
 	/* And finally now the object is completely decoupled from this vma,
 	 * we can drop its hold on the backing storage and allow it to be
@@ -3819,22 +3824,16 @@ search_free:
 		goto err_remove_node;
 	}
 
-	ret = i915_gem_gtt_prepare_object(obj);
-	if (ret)
-		goto err_remove_node;
-
 	trace_i915_vma_bind(vma, flags);
 	ret = i915_vma_bind(vma, obj->cache_level, flags);
 	if (ret)
-		goto err_finish_gtt;
+		goto err_remove_node;
 
 	list_move_tail(&obj->global_list, &dev_priv->mm.bound_list);
 	list_add_tail(&vma->mm_list, &vm->inactive_list);
 
 	return vma;
 
-err_finish_gtt:
-	i915_gem_gtt_finish_object(obj);
 err_remove_node:
 	drm_mm_remove_node(&vma->node);
 err_free_vma:
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 4256b8e97e40..b77a8f78c35a 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -135,8 +135,7 @@ static int get_context_size(struct drm_device *dev)
 
 void i915_gem_context_free(struct kref *ctx_ref)
 {
-	struct intel_context *ctx = container_of(ctx_ref,
-						 typeof(*ctx), ref);
+	struct intel_context *ctx = container_of(ctx_ref, typeof(*ctx), ref);
 
 	trace_i915_context_free(ctx);
 
@@ -157,9 +156,7 @@ i915_gem_alloc_context_obj(struct drm_device *dev, size_t size)
 	struct drm_i915_gem_object *obj;
 	int ret;
 
-	obj = i915_gem_object_create_stolen(dev, size);
-	if (obj == NULL)
-		obj = i915_gem_alloc_object(dev, size);
+	obj = i915_gem_alloc_object(dev, size);
 	if (obj == NULL)
 		return ERR_PTR(-ENOMEM);
 
@@ -197,6 +194,7 @@ __create_hw_context(struct drm_device *dev,
 
 	kref_init(&ctx->ref);
 	list_add_tail(&ctx->link, &dev_priv->context_list);
+	ctx->i915 = dev_priv;
 
 	if (dev_priv->hw_context_size) {
 		struct drm_i915_gem_object *obj =
diff --git a/drivers/gpu/drm/i915/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
index 7998da27c500..e9c2bfd85b52 100644
--- a/drivers/gpu/drm/i915/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
@@ -256,7 +256,6 @@ static int i915_gem_object_get_pages_dmabuf(struct drm_i915_gem_object *obj)
 		return PTR_ERR(sg);
 
 	obj->pages = sg;
-	obj->has_dma_mapping = true;
 	return 0;
 }
 
@@ -264,7 +263,6 @@ static void i915_gem_object_put_pages_dmabuf(struct drm_i915_gem_object *obj)
 {
 	dma_buf_unmap_attachment(obj->base.import_attach,
 				 obj->pages, DMA_BIDIRECTIONAL);
-	obj->has_dma_mapping = false;
 }
 
 static const struct drm_i915_gem_object_ops i915_gem_object_dmabuf_ops = {
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 83577c615962..923a3c4bf0b7 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -32,6 +32,7 @@
 #include "i915_trace.h"
 #include "intel_drv.h"
 #include <linux/dma_remapping.h>
+#include <linux/uaccess.h>
 
 #define  __EXEC_OBJECT_HAS_PIN (1<<31)
 #define  __EXEC_OBJECT_HAS_FENCE (1<<30)
@@ -458,7 +459,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,
 	}
 
 	/* We can't wait for rendering with pagefaults disabled */
-	if (obj->active && in_atomic())
+	if (obj->active && pagefault_disabled())
 		return -EFAULT;
 
 	if (use_cpu_reloc(obj))
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index ed65f24867b4..c2a291e09bd9 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -585,17 +585,17 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 		struct i915_page_table *pt;
 
 		if (WARN_ON(!ppgtt->pdp.page_directory[pdpe]))
-			continue;
+			break;
 
 		pd = ppgtt->pdp.page_directory[pdpe];
 
 		if (WARN_ON(!pd->page_table[pde]))
-			continue;
+			break;
 
 		pt = pd->page_table[pde];
 
 		if (WARN_ON(!px_page(pt)))
-			continue;
+			break;
 
 		last_pte = pte + num_entries;
 		if (last_pte > GEN8_PTES)
@@ -1802,9 +1802,6 @@ void i915_gem_suspend_gtt_mappings(struct drm_device *dev)
 
 int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj)
 {
-	if (obj->has_dma_mapping)
-		return 0;
-
 	if (!dma_map_sg(&obj->base.dev->pdev->dev,
 			obj->pages->sgl, obj->pages->nents,
 			PCI_DMA_BIDIRECTIONAL))
@@ -2052,10 +2049,8 @@ void i915_gem_gtt_finish_object(struct drm_i915_gem_object *obj)
 
 	interruptible = do_idling(dev_priv);
 
-	if (!obj->has_dma_mapping)
-		dma_unmap_sg(&dev->pdev->dev,
-			     obj->pages->sgl, obj->pages->nents,
-			     PCI_DMA_BIDIRECTIONAL);
+	dma_unmap_sg(&dev->pdev->dev, obj->pages->sgl, obj->pages->nents,
+		     PCI_DMA_BIDIRECTIONAL);
 
 	undo_idling(dev_priv, interruptible);
 }
@@ -2593,6 +2588,8 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj;
 	struct i915_address_space *vm;
+	struct i915_vma *vma;
+	bool flush;
 
 	i915_check_and_clear_faults(dev);
 
@@ -2602,16 +2599,23 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 				       dev_priv->gtt.base.total,
 				       true);
 
+	/* Cache flush objects bound into GGTT and rebind them. */
+	vm = &dev_priv->gtt.base;
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		struct i915_vma *vma = i915_gem_obj_to_vma(obj,
-							   &dev_priv->gtt.base);
-		if (!vma)
-			continue;
+		flush = false;
+		list_for_each_entry(vma, &obj->vma_list, vma_link) {
+			if (vma->vm != vm)
+				continue;
 
-		i915_gem_clflush_object(obj, obj->pin_display);
-		WARN_ON(i915_vma_bind(vma, obj->cache_level, PIN_UPDATE));
-	}
+			WARN_ON(i915_vma_bind(vma, obj->cache_level,
+					      PIN_UPDATE));
 
+			flush = true;
+		}
+
+		if (flush)
+			i915_gem_clflush_object(obj, obj->pin_display);
+	}
 
 	if (INTEL_INFO(dev)->gen >= 8) {
 		if (IS_CHERRYVIEW(dev) || IS_BROXTON(dev))
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index de76d886cd1d..ed682a9a9cbb 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -316,7 +316,6 @@ _i915_gem_object_create_stolen(struct drm_device *dev,
 	if (obj->pages == NULL)
 		goto cleanup;
 
-	obj->has_dma_mapping = true;
 	i915_gem_object_pin_pages(obj);
 	obj->stolen = stolen;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 1f4e5a32a16e..8fd431bcdfd3 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -545,6 +545,26 @@ err:
 	return ret;
 }
 
+static int
+__i915_gem_userptr_set_pages(struct drm_i915_gem_object *obj,
+			     struct page **pvec, int num_pages)
+{
+	int ret;
+
+	ret = st_set_pages(&obj->pages, pvec, num_pages);
+	if (ret)
+		return ret;
+
+	ret = i915_gem_gtt_prepare_object(obj);
+	if (ret) {
+		sg_free_table(obj->pages);
+		kfree(obj->pages);
+		obj->pages = NULL;
+	}
+
+	return ret;
+}
+
 static void
 __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
 {
@@ -584,9 +604,12 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
 	if (obj->userptr.work != &work->work) {
 		ret = 0;
 	} else if (pinned == num_pages) {
-		ret = st_set_pages(&obj->pages, pvec, num_pages);
+		ret = __i915_gem_userptr_set_pages(obj, pvec, num_pages);
 		if (ret == 0) {
 			list_add_tail(&obj->global_list, &to_i915(dev)->mm.unbound_list);
+			obj->get_page.sg = obj->pages->sgl;
+			obj->get_page.last = 0;
+
 			pinned = 0;
 		}
 	}
@@ -693,7 +716,7 @@ i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 			}
 		}
 	} else {
-		ret = st_set_pages(&obj->pages, pvec, num_pages);
+		ret = __i915_gem_userptr_set_pages(obj, pvec, num_pages);
 		if (ret == 0) {
 			obj->userptr.work = NULL;
 			pinned = 0;
@@ -715,6 +738,8 @@ i915_gem_userptr_put_pages(struct drm_i915_gem_object *obj)
 	if (obj->madv != I915_MADV_WILLNEED)
 		obj->dirty = 0;
 
+	i915_gem_gtt_finish_object(obj);
+
 	for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0) {
 		struct page *page = sg_page_iter_page(&sg_iter);
 
diff --git a/drivers/gpu/drm/i915/i915_ioc32.c b/drivers/gpu/drm/i915/i915_ioc32.c
index a5a2d5b3f44b..97f3a5640289 100644
--- a/drivers/gpu/drm/i915/i915_ioc32.c
+++ b/drivers/gpu/drm/i915/i915_ioc32.c
@@ -84,7 +84,7 @@ long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	drm_ioctl_compat_t *fn = NULL;
 	int ret;
 
-	if (nr < DRM_COMMAND_BASE)
+	if (nr < DRM_COMMAND_BASE || nr >= DRM_COMMAND_END)
 		return drm_compat_ioctl(filp, cmd, arg);
 
 	if (nr < DRM_COMMAND_BASE + ARRAY_SIZE(i915_compat_ioctls))
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index a897f68485c2..d87f173a0179 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2486,18 +2486,11 @@ static void gen8_disable_vblank(struct drm_device *dev, int pipe)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
-static struct drm_i915_gem_request *
-ring_last_request(struct intel_engine_cs *ring)
-{
-	return list_entry(ring->request_list.prev,
-			  struct drm_i915_gem_request, list);
-}
-
 static bool
-ring_idle(struct intel_engine_cs *ring)
+ring_idle(struct intel_engine_cs *ring, u32 seqno)
 {
 	return (list_empty(&ring->request_list) ||
-		i915_gem_request_completed(ring_last_request(ring), false));
+		i915_seqno_passed(seqno, ring->last_submitted_seqno));
 }
 
 static bool
@@ -2719,7 +2712,7 @@ static void i915_hangcheck_elapsed(struct work_struct *work)
 		acthd = intel_ring_get_active_head(ring);
 
 		if (ring->hangcheck.seqno == seqno) {
-			if (ring_idle(ring)) {
+			if (ring_idle(ring, seqno)) {
 				ring->hangcheck.action = HANGCHECK_IDLE;
 
 				if (waitqueue_active(&ring->irq_queue)) {
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index ef5f69a1607f..e9a95df639f0 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -3532,6 +3532,7 @@ enum skl_disp_power_wells {
 #define   BLM_POLARITY_PNV			(1 << 0) /* pnv only */
 
 #define BLC_HIST_CTL	(dev_priv->info.display_mmio_offset + 0x61260)
+#define  BLM_HISTOGRAM_ENABLE			(1 << 31)
 
 /* New registers for PCH-split platforms. Safe where new bits show up, the
  * register layout machtes with gen4 BLC_PWM_CTL[12]. */
@@ -7031,6 +7032,9 @@ enum skl_disp_power_wells {
 #define   AUDIO_CP_READY(trans)		((1 << 1) << ((trans) * 4))
 #define   AUDIO_ELD_VALID(trans)	((1 << 0) << ((trans) * 4))
 
+#define HSW_AUD_CHICKENBIT			0x65f10
+#define   SKL_AUD_CODEC_WAKE_SIGNAL		(1 << 15)
+
 /* HSW Power Wells */
 #define HSW_PWR_WELL_BIOS			0x45400 /* CTL1 */
 #define HSW_PWR_WELL_DRIVER			0x45404 /* CTL2 */
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index 63328b6e8ea5..2f34c47bd4bf 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -727,7 +727,7 @@ DECLARE_EVENT_CLASS(i915_context,
 	TP_fast_assign(
 			__entry->ctx = ctx;
 			__entry->vm = ctx->ppgtt ? &ctx->ppgtt->base : NULL;
-			__entry->dev = ctx->file_priv->dev_priv->dev->primary->index;
+			__entry->dev = ctx->i915->dev->primary->index;
 	),
 
 	TP_printk("dev=%u, ctx=%p, ctx_vm=%p",
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index c4397c1e7e4f..dc32cf4585f8 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -471,6 +471,32 @@ static void i915_audio_component_put_power(struct device *dev)
 	intel_display_power_put(dev_to_i915(dev), POWER_DOMAIN_AUDIO);
 }
 
+static void i915_audio_component_codec_wake_override(struct device *dev,
+						     bool enable)
+{
+	struct drm_i915_private *dev_priv = dev_to_i915(dev);
+	u32 tmp;
+
+	if (!IS_SKYLAKE(dev_priv))
+		return;
+
+	/*
+	 * Enable/disable generating the codec wake signal, overriding the
+	 * internal logic to generate the codec wake to controller.
+	 */
+	tmp = I915_READ(HSW_AUD_CHICKENBIT);
+	tmp &= ~SKL_AUD_CODEC_WAKE_SIGNAL;
+	I915_WRITE(HSW_AUD_CHICKENBIT, tmp);
+	usleep_range(1000, 1500);
+
+	if (enable) {
+		tmp = I915_READ(HSW_AUD_CHICKENBIT);
+		tmp |= SKL_AUD_CODEC_WAKE_SIGNAL;
+		I915_WRITE(HSW_AUD_CHICKENBIT, tmp);
+		usleep_range(1000, 1500);
+	}
+}
+
 /* Get CDCLK in kHz  */
 static int i915_audio_component_get_cdclk_freq(struct device *dev)
 {
@@ -492,6 +518,7 @@ static const struct i915_audio_component_ops i915_audio_component_ops = {
 	.owner		= THIS_MODULE,
 	.get_power	= i915_audio_component_get_power,
 	.put_power	= i915_audio_component_put_power,
+	.codec_wake_override = i915_audio_component_codec_wake_override,
 	.get_cdclk_freq	= i915_audio_component_get_cdclk_freq,
 };
 
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index bd92e9ce4127..ede652867596 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -8096,7 +8096,7 @@ static void chv_crtc_clock_get(struct intel_crtc *crtc,
 	int pipe = pipe_config->cpu_transcoder;
 	enum dpio_channel port = vlv_pipe_to_channel(pipe);
 	intel_clock_t clock;
-	u32 cmn_dw13, pll_dw0, pll_dw1, pll_dw2;
+	u32 cmn_dw13, pll_dw0, pll_dw1, pll_dw2, pll_dw3;
 	int refclk = 100000;
 
 	mutex_lock(&dev_priv->sb_lock);
@@ -8104,10 +8104,13 @@ static void chv_crtc_clock_get(struct intel_crtc *crtc,
 	pll_dw0 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW0(port));
 	pll_dw1 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW1(port));
 	pll_dw2 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW2(port));
+	pll_dw3 = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW3(port));
 	mutex_unlock(&dev_priv->sb_lock);
 
 	clock.m1 = (pll_dw1 & 0x7) == DPIO_CHV_M1_DIV_BY_2 ? 2 : 0;
-	clock.m2 = ((pll_dw0 & 0xff) << 22) | (pll_dw2 & 0x3fffff);
+	clock.m2 = (pll_dw0 & 0xff) << 22;
+	if (pll_dw3 & DPIO_CHV_FRAC_DIV_EN)
+		clock.m2 |= pll_dw2 & 0x3fffff;
 	clock.n = (pll_dw1 >> DPIO_CHV_N_DIV_SHIFT) & 0xf;
 	clock.p1 = (cmn_dw13 >> DPIO_CHV_P1_DIV_SHIFT) & 0x7;
 	clock.p2 = (cmn_dw13 >> DPIO_CHV_P2_DIV_SHIFT) & 0x1f;
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 085d84156008..f1b9f939b435 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -900,10 +900,8 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 				continue;
 			}
 			if (status & DP_AUX_CH_CTL_DONE)
-				break;
+				goto done;
 		}
-		if (status & DP_AUX_CH_CTL_DONE)
-			break;
 	}
 
 	if ((status & DP_AUX_CH_CTL_DONE) == 0) {
@@ -912,6 +910,7 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 		goto out;
 	}
 
+done:
 	/* Check for timeout or receive error.
 	 * Timeouts occur when the sink is not connected
 	 */
@@ -1148,6 +1147,9 @@ skl_edp_set_pll_config(struct intel_crtc_state *pipe_config, int link_clock)
 static void
 hsw_dp_set_ddi_pll_sel(struct intel_crtc_state *pipe_config, int link_bw)
 {
+	memset(&pipe_config->dpll_hw_state, 0,
+	       sizeof(pipe_config->dpll_hw_state));
+
 	switch (link_bw) {
 	case DP_LINK_BW_1_62:
 		pipe_config->ddi_pll_sel = PORT_CLK_SEL_LCPLL_810;
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index 92072f56e418..a64f26c670af 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -486,7 +486,7 @@ gmbus_xfer(struct i2c_adapter *adapter,
 					       struct intel_gmbus,
 					       adapter);
 	struct drm_i915_private *dev_priv = bus->dev_priv;
-	int i, reg_offset;
+	int i = 0, inc, try = 0, reg_offset;
 	int ret = 0;
 
 	intel_aux_display_runtime_get(dev_priv);
@@ -499,12 +499,14 @@ gmbus_xfer(struct i2c_adapter *adapter,
 
 	reg_offset = dev_priv->gpio_mmio_base;
 
+retry:
 	I915_WRITE(GMBUS0 + reg_offset, bus->reg0);
 
-	for (i = 0; i < num; i++) {
+	for (; i < num; i += inc) {
+		inc = 1;
 		if (gmbus_is_index_read(msgs, i, num)) {
 			ret = gmbus_xfer_index_read(dev_priv, &msgs[i]);
-			i += 1;  /* set i to the index of the read xfer */
+			inc = 2; /* an index read is two msgs */
 		} else if (msgs[i].flags & I2C_M_RD) {
 			ret = gmbus_xfer_read(dev_priv, &msgs[i], 0);
 		} else {
@@ -576,6 +578,18 @@ clear_err:
 			 adapter->name, msgs[i].addr,
 			 (msgs[i].flags & I2C_M_RD) ? 'r' : 'w', msgs[i].len);
 
+	/*
+	 * Passive adapters sometimes NAK the first probe. Retry the first
+	 * message once on -ENXIO for GMBUS transfers; the bit banging algorithm
+	 * has retries internally. See also the retry loop in
+	 * drm_do_probe_ddc_edid, which bails out on the first -ENXIO.
+	 */
+	if (ret == -ENXIO && i == 0 && try++ == 0) {
+		DRM_DEBUG_KMS("GMBUS [%s] NAK on first message, retry\n",
+			      adapter->name);
+		goto retry;
+	}
+
 	goto out;
 
 timeout:
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 83ed0f7fcbe2..9faad82c42ec 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -1421,6 +1421,12 @@ static int gen8_init_common_ring(struct intel_engine_cs *ring)
 	I915_WRITE_IMR(ring, ~(ring->irq_enable_mask | ring->irq_keep_mask));
 	I915_WRITE(RING_HWSTAM(ring->mmio_base), 0xffffffff);
 
+	if (ring->status_page.obj) {
+		I915_WRITE(RING_HWS_PGA(ring->mmio_base),
+			   (u32)ring->status_page.gfx_addr);
+		POSTING_READ(RING_HWS_PGA(ring->mmio_base));
+	}
+
 	I915_WRITE(RING_MODE_GEN7(ring),
 		   _MASKED_BIT_DISABLE(GFX_REPLAY_MODE) |
 		   _MASKED_BIT_ENABLE(GFX_RUN_LIST_ENABLE));
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index 7df916e914a4..cb1c65739425 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -403,16 +403,6 @@ int intel_opregion_notify_adapter(struct drm_device *dev, pci_power_t state)
 	return -EINVAL;
 }
 
-/*
- * If the vendor backlight interface is not in use and ACPI backlight interface
- * is broken, do not bother processing backlight change requests from firmware.
- */
-static bool should_ignore_backlight_request(void)
-{
-	return acpi_video_backlight_support() &&
-	       !acpi_video_verify_backlight_support();
-}
-
 static u32 asle_set_backlight(struct drm_device *dev, u32 bclp)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -421,7 +411,7 @@ static u32 asle_set_backlight(struct drm_device *dev, u32 bclp)
 
 	DRM_DEBUG_DRIVER("bclp = 0x%08x\n", bclp);
 
-	if (should_ignore_backlight_request()) {
+	if (acpi_video_get_backlight_type() == acpi_backlight_native) {
 		DRM_DEBUG_KMS("opregion backlight request ignored\n");
 		return 0;
 	}
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 7d83527f95f7..55aad2322e10 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -907,6 +907,14 @@ static void i9xx_enable_backlight(struct intel_connector *connector)
 
 	/* XXX: combine this into above write? */
 	intel_panel_actually_set_backlight(connector, panel->backlight.level);
+
+	/*
+	 * Needed to enable backlight on some 855gm models. BLC_HIST_CTL is
+	 * 855gm only, but checking for gen2 is safe, as 855gm is the only gen2
+	 * that has backlight.
+	 */
+	if (IS_GEN2(dev))
+		I915_WRITE(BLC_HIST_CTL, BLM_HISTOGRAM_ENABLE);
 }
 
 static void i965_enable_backlight(struct intel_connector *connector)
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 1c14233d179f..177f7ed16cf0 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -1041,6 +1041,13 @@ static int skl_init_workarounds(struct intel_engine_cs *ring)
 		WA_SET_BIT_MASKED(HIZ_CHICKEN,
 				  BDW_HIZ_POWER_COMPILER_CLOCK_GATING_DISABLE);
 
+	if (INTEL_REVID(dev) == SKL_REVID_C0 ||
+	    INTEL_REVID(dev) == SKL_REVID_D0)
+		/* WaBarrierPerformanceFixDisable:skl */
+		WA_SET_BIT_MASKED(HDC_CHICKEN0,
+				  HDC_FENCE_DEST_SLM_DISABLE |
+				  HDC_BARRIER_PERFORMANCE_DISABLE);
+
 	if (INTEL_REVID(dev) <= SKL_REVID_D0) {
 		/*
 		 *Use Force Non-Coherent whenever executing a 3D context. This
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 0ea89ea30182..2e85fda94963 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -292,6 +292,13 @@ struct  intel_engine_cs {
 	 */
 	struct list_head request_list;
 
+	/**
+	 * Seqno of request most recently submitted to request_list.
+	 * Used exclusively by hang checker to avoid grabbing lock while
+	 * inspecting request list.
+	 */
+	u32 last_submitted_seqno;
+
 	bool gpu_caches_dirty;
 
 	wait_queue_head_t irq_queue;
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index d24ef75596a1..aa2fd751609c 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -2554,7 +2554,7 @@ intel_sdvo_analog_init(struct intel_sdvo *intel_sdvo, int device)
 
 	DRM_DEBUG_KMS("initialising analog device %d\n", device);
 
-	intel_sdvo_connector = kzalloc(sizeof(*intel_sdvo_connector), GFP_KERNEL);
+	intel_sdvo_connector = intel_sdvo_connector_alloc();
 	if (!intel_sdvo_connector)
 		return false;
 
diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c
index 6e84df9369a6..ad4b9010dfb0 100644
--- a/drivers/gpu/drm/mgag200/mgag200_mode.c
+++ b/drivers/gpu/drm/mgag200/mgag200_mode.c
@@ -1526,6 +1526,11 @@ static int mga_vga_mode_valid(struct drm_connector *connector,
 		return MODE_BANDWIDTH;
 	}
 
+	if ((mode->hdisplay % 8) != 0 || (mode->hsync_start % 8) != 0 ||
+	    (mode->hsync_end % 8) != 0 || (mode->htotal % 8) != 0) {
+		return MODE_H_ILLEGAL;
+	}
+
 	if (mode->crtc_hdisplay > 2048 || mode->crtc_hsync_start > 4096 ||
 	    mode->crtc_hsync_end > 4096 || mode->crtc_htotal > 4096 ||
 	    mode->crtc_vdisplay > 2048 || mode->crtc_vsync_start > 4096 ||
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 89049335b738..649024d4daf1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -23,6 +23,7 @@
  */
 
 #include <linux/console.h>
+#include <linux/delay.h>
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/pm_runtime.h>
@@ -666,6 +667,7 @@ nouveau_pmops_suspend(struct device *dev)
 	pci_save_state(pdev);
 	pci_disable_device(pdev);
 	pci_set_power_state(pdev, PCI_D3hot);
+	udelay(200);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 0e690bf19fc9..af1ee517f372 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -555,10 +555,7 @@ nouveau_gem_pushbuf_validate(struct nouveau_channel *chan,
 static inline void
 u_free(void *addr)
 {
-	if (!is_vmalloc_addr(addr))
-		kfree(addr);
-	else
-		vfree(addr);
+	kvfree(addr);
 }
 
 static inline void *
diff --git a/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c
index f2daad8c3d96..7841970de48d 100644
--- a/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c
+++ b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c
@@ -285,7 +285,7 @@ static int dmm_txn_commit(struct dmm_txn *txn, bool wait)
 
 	if (wait) {
 		if (!wait_for_completion_timeout(&engine->compl,
-				msecs_to_jiffies(1))) {
+				msecs_to_jiffies(100))) {
 			dev_err(dmm->dev, "timed out waiting for done\n");
 			ret = -ETIMEDOUT;
 		}
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.h b/drivers/gpu/drm/omapdrm/omap_drv.h
index ae2df41f216f..12081e61d45a 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.h
+++ b/drivers/gpu/drm/omapdrm/omap_drv.h
@@ -177,7 +177,7 @@ struct drm_framebuffer *omap_framebuffer_init(struct drm_device *dev,
 		struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_object **bos);
 struct drm_gem_object *omap_framebuffer_bo(struct drm_framebuffer *fb, int p);
 int omap_framebuffer_pin(struct drm_framebuffer *fb);
-int omap_framebuffer_unpin(struct drm_framebuffer *fb);
+void omap_framebuffer_unpin(struct drm_framebuffer *fb);
 void omap_framebuffer_update_scanout(struct drm_framebuffer *fb,
 		struct omap_drm_window *win, struct omap_overlay_info *info);
 struct drm_connector *omap_framebuffer_get_next_connector(
@@ -211,7 +211,7 @@ void omap_gem_dma_sync(struct drm_gem_object *obj,
 		enum dma_data_direction dir);
 int omap_gem_get_paddr(struct drm_gem_object *obj,
 		dma_addr_t *paddr, bool remap);
-int omap_gem_put_paddr(struct drm_gem_object *obj);
+void omap_gem_put_paddr(struct drm_gem_object *obj);
 int omap_gem_get_pages(struct drm_gem_object *obj, struct page ***pages,
 		bool remap);
 int omap_gem_put_pages(struct drm_gem_object *obj);
@@ -236,7 +236,7 @@ static inline int align_pitch(int pitch, int width, int bpp)
 	/* PVR needs alignment to 8 pixels.. right now that is the most
 	 * restrictive stride requirement..
 	 */
-	return ALIGN(pitch, 8 * bytespp);
+	return roundup(pitch, 8 * bytespp);
 }
 
 /* map crtc to vblank mask */
diff --git a/drivers/gpu/drm/omapdrm/omap_fb.c b/drivers/gpu/drm/omapdrm/omap_fb.c
index 0b967e76df1a..51b1219af87f 100644
--- a/drivers/gpu/drm/omapdrm/omap_fb.c
+++ b/drivers/gpu/drm/omapdrm/omap_fb.c
@@ -287,10 +287,10 @@ fail:
 }
 
 /* unpin, no longer being scanned out: */
-int omap_framebuffer_unpin(struct drm_framebuffer *fb)
+void omap_framebuffer_unpin(struct drm_framebuffer *fb)
 {
 	struct omap_framebuffer *omap_fb = to_omap_framebuffer(fb);
-	int ret, i, n = drm_format_num_planes(fb->pixel_format);
+	int i, n = drm_format_num_planes(fb->pixel_format);
 
 	mutex_lock(&omap_fb->lock);
 
@@ -298,24 +298,16 @@ int omap_framebuffer_unpin(struct drm_framebuffer *fb)
 
 	if (omap_fb->pin_count > 0) {
 		mutex_unlock(&omap_fb->lock);
-		return 0;
+		return;
 	}
 
 	for (i = 0; i < n; i++) {
 		struct plane *plane = &omap_fb->planes[i];
-		ret = omap_gem_put_paddr(plane->bo);
-		if (ret)
-			goto fail;
+		omap_gem_put_paddr(plane->bo);
 		plane->paddr = 0;
 	}
 
 	mutex_unlock(&omap_fb->lock);
-
-	return 0;
-
-fail:
-	mutex_unlock(&omap_fb->lock);
-	return ret;
 }
 
 struct drm_gem_object *omap_framebuffer_bo(struct drm_framebuffer *fb, int p)
diff --git a/drivers/gpu/drm/omapdrm/omap_fbdev.c b/drivers/gpu/drm/omapdrm/omap_fbdev.c
index 23b5a84389e3..720d16bce7e8 100644
--- a/drivers/gpu/drm/omapdrm/omap_fbdev.c
+++ b/drivers/gpu/drm/omapdrm/omap_fbdev.c
@@ -135,7 +135,7 @@ static int omap_fbdev_create(struct drm_fb_helper *helper,
 	fbdev->ywrap_enabled = priv->has_dmm && ywrap_enabled;
 	if (fbdev->ywrap_enabled) {
 		/* need to align pitch to page size if using DMM scrolling */
-		mode_cmd.pitches[0] = ALIGN(mode_cmd.pitches[0], PAGE_SIZE);
+		mode_cmd.pitches[0] = PAGE_ALIGN(mode_cmd.pitches[0]);
 	}
 
 	/* allocate backing bo */
diff --git a/drivers/gpu/drm/omapdrm/omap_gem.c b/drivers/gpu/drm/omapdrm/omap_gem.c
index 2ab77801cf5f..7ed08fdc4c42 100644
--- a/drivers/gpu/drm/omapdrm/omap_gem.c
+++ b/drivers/gpu/drm/omapdrm/omap_gem.c
@@ -808,10 +808,10 @@ fail:
 /* Release physical address, when DMA is no longer being performed.. this
  * could potentially unpin and unmap buffers from TILER
  */
-int omap_gem_put_paddr(struct drm_gem_object *obj)
+void omap_gem_put_paddr(struct drm_gem_object *obj)
 {
 	struct omap_gem_object *omap_obj = to_omap_bo(obj);
-	int ret = 0;
+	int ret;
 
 	mutex_lock(&obj->dev->struct_mutex);
 	if (omap_obj->paddr_cnt > 0) {
@@ -821,7 +821,6 @@ int omap_gem_put_paddr(struct drm_gem_object *obj)
 			if (ret) {
 				dev_err(obj->dev->dev,
 					"could not unpin pages: %d\n", ret);
-				goto fail;
 			}
 			ret = tiler_release(omap_obj->block);
 			if (ret) {
@@ -832,9 +831,8 @@ int omap_gem_put_paddr(struct drm_gem_object *obj)
 			omap_obj->block = NULL;
 		}
 	}
-fail:
+
 	mutex_unlock(&obj->dev->struct_mutex);
-	return ret;
 }
 
 /* Get rotated scanout address (only valid if already pinned), at the
@@ -1378,11 +1376,7 @@ struct drm_gem_object *omap_gem_new(struct drm_device *dev,
 
 	omap_obj = kzalloc(sizeof(*omap_obj), GFP_KERNEL);
 	if (!omap_obj)
-		goto fail;
-
-	spin_lock(&priv->list_lock);
-	list_add(&omap_obj->mm_list, &priv->obj_list);
-	spin_unlock(&priv->list_lock);
+		return NULL;
 
 	obj = &omap_obj->base;
 
@@ -1392,11 +1386,19 @@ struct drm_gem_object *omap_gem_new(struct drm_device *dev,
 		 */
 		omap_obj->vaddr =  dma_alloc_writecombine(dev->dev, size,
 				&omap_obj->paddr, GFP_KERNEL);
-		if (omap_obj->vaddr)
-			flags |= OMAP_BO_DMA;
+		if (!omap_obj->vaddr) {
+			kfree(omap_obj);
+
+			return NULL;
+		}
 
+		flags |= OMAP_BO_DMA;
 	}
 
+	spin_lock(&priv->list_lock);
+	list_add(&omap_obj->mm_list, &priv->obj_list);
+	spin_unlock(&priv->list_lock);
+
 	omap_obj->flags = flags;
 
 	if (flags & OMAP_BO_TILED) {
diff --git a/drivers/gpu/drm/omapdrm/omap_plane.c b/drivers/gpu/drm/omapdrm/omap_plane.c
index cfa8276c4deb..098904696a5c 100644
--- a/drivers/gpu/drm/omapdrm/omap_plane.c
+++ b/drivers/gpu/drm/omapdrm/omap_plane.c
@@ -17,6 +17,7 @@
  * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
+#include <drm/drm_atomic.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_plane_helper.h>
 
@@ -153,9 +154,34 @@ static void omap_plane_atomic_disable(struct drm_plane *plane,
 	dispc_ovl_enable(omap_plane->id, false);
 }
 
+static int omap_plane_atomic_check(struct drm_plane *plane,
+				   struct drm_plane_state *state)
+{
+	struct drm_crtc_state *crtc_state;
+
+	if (!state->crtc)
+		return 0;
+
+	crtc_state = drm_atomic_get_crtc_state(state->state, state->crtc);
+	if (IS_ERR(crtc_state))
+		return PTR_ERR(crtc_state);
+
+	if (state->crtc_x < 0 || state->crtc_y < 0)
+		return -EINVAL;
+
+	if (state->crtc_x + state->crtc_w > crtc_state->adjusted_mode.hdisplay)
+		return -EINVAL;
+
+	if (state->crtc_y + state->crtc_h > crtc_state->adjusted_mode.vdisplay)
+		return -EINVAL;
+
+	return 0;
+}
+
 static const struct drm_plane_helper_funcs omap_plane_helper_funcs = {
 	.prepare_fb = omap_plane_prepare_fb,
 	.cleanup_fb = omap_plane_cleanup_fb,
+	.atomic_check = omap_plane_atomic_check,
 	.atomic_update = omap_plane_atomic_update,
 	.atomic_disable = omap_plane_atomic_disable,
 };
diff --git a/drivers/gpu/drm/radeon/atombios.h b/drivers/gpu/drm/radeon/atombios.h
index 92be50c39ffd..ab89eed9ddd9 100644
--- a/drivers/gpu/drm/radeon/atombios.h
+++ b/drivers/gpu/drm/radeon/atombios.h
@@ -7944,8 +7944,8 @@ typedef struct {
 typedef struct {
   AMD_ACPI_DESCRIPTION_HEADER SHeader;
   UCHAR TableUUID[16];    //0x24
-  ULONG VBIOSImageOffset; //0x34. Offset to the first GOP_VBIOS_CONTENT block from the beginning of the stucture.
-  ULONG Lib1ImageOffset;  //0x38. Offset to the first GOP_LIB1_CONTENT block from the beginning of the stucture.
+  ULONG VBIOSImageOffset; //0x34. Offset to the first GOP_VBIOS_CONTENT block from the beginning of the structure.
+  ULONG Lib1ImageOffset;  //0x38. Offset to the first GOP_LIB1_CONTENT block from the beginning of the structure.
   ULONG Reserved[4];      //0x3C
 }UEFI_ACPI_VFCT;
 
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index e597ffc26563..dac78ad24b31 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -580,9 +580,6 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 		else
 			radeon_crtc->pll_flags |= RADEON_PLL_PREFER_LOW_REF_DIV;
 
-		/* if there is no audio, set MINM_OVER_MAXP  */
-		if (!drm_detect_monitor_audio(radeon_connector_edid(connector)))
-			radeon_crtc->pll_flags |= RADEON_PLL_PREFER_MINM_OVER_MAXP;
 		if (rdev->family < CHIP_RV770)
 			radeon_crtc->pll_flags |= RADEON_PLL_PREFER_MINM_OVER_MAXP;
 		/* use frac fb div on APUs */
@@ -1798,9 +1795,7 @@ static int radeon_get_shared_nondp_ppll(struct drm_crtc *crtc)
 			if ((crtc->mode.clock == test_crtc->mode.clock) &&
 			    (adjusted_clock == test_adjusted_clock) &&
 			    (radeon_crtc->ss_enabled == test_radeon_crtc->ss_enabled) &&
-			    (test_radeon_crtc->pll_id != ATOM_PPLL_INVALID) &&
-			    (drm_detect_monitor_audio(radeon_connector_edid(test_radeon_crtc->connector)) ==
-			     drm_detect_monitor_audio(radeon_connector_edid(radeon_crtc->connector))))
+			    (test_radeon_crtc->pll_id != ATOM_PPLL_INVALID))
 				return test_radeon_crtc->pll_id;
 		}
 	}
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index b0688b0c8908..248953d2fdb7 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -4604,6 +4604,31 @@ void cik_compute_set_wptr(struct radeon_device *rdev,
 	WDOORBELL32(ring->doorbell_index, ring->wptr);
 }
 
+static void cik_compute_stop(struct radeon_device *rdev,
+			     struct radeon_ring *ring)
+{
+	u32 j, tmp;
+
+	cik_srbm_select(rdev, ring->me, ring->pipe, ring->queue, 0);
+	/* Disable wptr polling. */
+	tmp = RREG32(CP_PQ_WPTR_POLL_CNTL);
+	tmp &= ~WPTR_POLL_EN;
+	WREG32(CP_PQ_WPTR_POLL_CNTL, tmp);
+	/* Disable HQD. */
+	if (RREG32(CP_HQD_ACTIVE) & 1) {
+		WREG32(CP_HQD_DEQUEUE_REQUEST, 1);
+		for (j = 0; j < rdev->usec_timeout; j++) {
+			if (!(RREG32(CP_HQD_ACTIVE) & 1))
+				break;
+			udelay(1);
+		}
+		WREG32(CP_HQD_DEQUEUE_REQUEST, 0);
+		WREG32(CP_HQD_PQ_RPTR, 0);
+		WREG32(CP_HQD_PQ_WPTR, 0);
+	}
+	cik_srbm_select(rdev, 0, 0, 0, 0);
+}
+
 /**
  * cik_cp_compute_enable - enable/disable the compute CP MEs
  *
@@ -4617,6 +4642,15 @@ static void cik_cp_compute_enable(struct radeon_device *rdev, bool enable)
 	if (enable)
 		WREG32(CP_MEC_CNTL, 0);
 	else {
+		/*
+		 * To make hibernation reliable we need to clear compute ring
+		 * configuration before halting the compute ring.
+		 */
+		mutex_lock(&rdev->srbm_mutex);
+		cik_compute_stop(rdev,&rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX]);
+		cik_compute_stop(rdev,&rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX]);
+		mutex_unlock(&rdev->srbm_mutex);
+
 		WREG32(CP_MEC_CNTL, (MEC_ME1_HALT | MEC_ME2_HALT));
 		rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX].ready = false;
 		rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX].ready = false;
@@ -7930,23 +7964,27 @@ restart_ih:
 		case 1: /* D1 vblank/vline */
 			switch (src_data) {
 			case 0: /* D1 vblank */
-				if (rdev->irq.stat_regs.cik.disp_int & LB_D1_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[0]) {
-						drm_handle_vblank(rdev->ddev, 0);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_vblank(rdev, 0);
-					rdev->irq.stat_regs.cik.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D1 vblank\n");
+				if (!(rdev->irq.stat_regs.cik.disp_int & LB_D1_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[0]) {
+					drm_handle_vblank(rdev->ddev, 0);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[0]))
+					radeon_crtc_handle_vblank(rdev, 0);
+				rdev->irq.stat_regs.cik.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D1 vblank\n");
+
 				break;
 			case 1: /* D1 vline */
-				if (rdev->irq.stat_regs.cik.disp_int & LB_D1_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int &= ~LB_D1_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D1 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int & LB_D1_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int &= ~LB_D1_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D1 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -7956,23 +7994,27 @@ restart_ih:
 		case 2: /* D2 vblank/vline */
 			switch (src_data) {
 			case 0: /* D2 vblank */
-				if (rdev->irq.stat_regs.cik.disp_int_cont & LB_D2_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[1]) {
-						drm_handle_vblank(rdev->ddev, 1);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_vblank(rdev, 1);
-					rdev->irq.stat_regs.cik.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D2 vblank\n");
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont & LB_D2_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[1]) {
+					drm_handle_vblank(rdev->ddev, 1);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[1]))
+					radeon_crtc_handle_vblank(rdev, 1);
+				rdev->irq.stat_regs.cik.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D2 vblank\n");
+
 				break;
 			case 1: /* D2 vline */
-				if (rdev->irq.stat_regs.cik.disp_int_cont & LB_D2_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont &= ~LB_D2_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D2 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont & LB_D2_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont &= ~LB_D2_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D2 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -7982,23 +8024,27 @@ restart_ih:
 		case 3: /* D3 vblank/vline */
 			switch (src_data) {
 			case 0: /* D3 vblank */
-				if (rdev->irq.stat_regs.cik.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[2]) {
-						drm_handle_vblank(rdev->ddev, 2);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[2]))
-						radeon_crtc_handle_vblank(rdev, 2);
-					rdev->irq.stat_regs.cik.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D3 vblank\n");
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[2]) {
+					drm_handle_vblank(rdev->ddev, 2);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[2]))
+					radeon_crtc_handle_vblank(rdev, 2);
+				rdev->irq.stat_regs.cik.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D3 vblank\n");
+
 				break;
 			case 1: /* D3 vline */
-				if (rdev->irq.stat_regs.cik.disp_int_cont2 & LB_D3_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont2 &= ~LB_D3_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D3 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont2 & LB_D3_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont2 &= ~LB_D3_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D3 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -8008,23 +8054,27 @@ restart_ih:
 		case 4: /* D4 vblank/vline */
 			switch (src_data) {
 			case 0: /* D4 vblank */
-				if (rdev->irq.stat_regs.cik.disp_int_cont3 & LB_D4_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[3]) {
-						drm_handle_vblank(rdev->ddev, 3);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[3]))
-						radeon_crtc_handle_vblank(rdev, 3);
-					rdev->irq.stat_regs.cik.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D4 vblank\n");
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont3 & LB_D4_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[3]) {
+					drm_handle_vblank(rdev->ddev, 3);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[3]))
+					radeon_crtc_handle_vblank(rdev, 3);
+				rdev->irq.stat_regs.cik.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D4 vblank\n");
+
 				break;
 			case 1: /* D4 vline */
-				if (rdev->irq.stat_regs.cik.disp_int_cont3 & LB_D4_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont3 &= ~LB_D4_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D4 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont3 & LB_D4_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont3 &= ~LB_D4_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D4 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -8034,23 +8084,27 @@ restart_ih:
 		case 5: /* D5 vblank/vline */
 			switch (src_data) {
 			case 0: /* D5 vblank */
-				if (rdev->irq.stat_regs.cik.disp_int_cont4 & LB_D5_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[4]) {
-						drm_handle_vblank(rdev->ddev, 4);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[4]))
-						radeon_crtc_handle_vblank(rdev, 4);
-					rdev->irq.stat_regs.cik.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D5 vblank\n");
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont4 & LB_D5_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[4]) {
+					drm_handle_vblank(rdev->ddev, 4);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[4]))
+					radeon_crtc_handle_vblank(rdev, 4);
+				rdev->irq.stat_regs.cik.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D5 vblank\n");
+
 				break;
 			case 1: /* D5 vline */
-				if (rdev->irq.stat_regs.cik.disp_int_cont4 & LB_D5_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont4 &= ~LB_D5_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D5 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont4 & LB_D5_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont4 &= ~LB_D5_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D5 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -8060,23 +8114,27 @@ restart_ih:
 		case 6: /* D6 vblank/vline */
 			switch (src_data) {
 			case 0: /* D6 vblank */
-				if (rdev->irq.stat_regs.cik.disp_int_cont5 & LB_D6_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[5]) {
-						drm_handle_vblank(rdev->ddev, 5);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[5]))
-						radeon_crtc_handle_vblank(rdev, 5);
-					rdev->irq.stat_regs.cik.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D6 vblank\n");
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont5 & LB_D6_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[5]) {
+					drm_handle_vblank(rdev->ddev, 5);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[5]))
+					radeon_crtc_handle_vblank(rdev, 5);
+				rdev->irq.stat_regs.cik.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D6 vblank\n");
+
 				break;
 			case 1: /* D6 vline */
-				if (rdev->irq.stat_regs.cik.disp_int_cont5 & LB_D6_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont5 &= ~LB_D6_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D6 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont5 & LB_D6_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont5 &= ~LB_D6_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D6 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -8096,88 +8154,112 @@ restart_ih:
 		case 42: /* HPD hotplug */
 			switch (src_data) {
 			case 0:
-				if (rdev->irq.stat_regs.cik.disp_int & DC_HPD1_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int &= ~DC_HPD1_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD1\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int & DC_HPD1_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int &= ~DC_HPD1_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD1\n");
+
 				break;
 			case 1:
-				if (rdev->irq.stat_regs.cik.disp_int_cont & DC_HPD2_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont &= ~DC_HPD2_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD2\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont & DC_HPD2_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont &= ~DC_HPD2_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD2\n");
+
 				break;
 			case 2:
-				if (rdev->irq.stat_regs.cik.disp_int_cont2 & DC_HPD3_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont2 &= ~DC_HPD3_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD3\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont2 & DC_HPD3_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont2 &= ~DC_HPD3_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD3\n");
+
 				break;
 			case 3:
-				if (rdev->irq.stat_regs.cik.disp_int_cont3 & DC_HPD4_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont3 &= ~DC_HPD4_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD4\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont3 & DC_HPD4_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont3 &= ~DC_HPD4_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD4\n");
+
 				break;
 			case 4:
-				if (rdev->irq.stat_regs.cik.disp_int_cont4 & DC_HPD5_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont4 &= ~DC_HPD5_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD5\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont4 & DC_HPD5_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont4 &= ~DC_HPD5_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD5\n");
+
 				break;
 			case 5:
-				if (rdev->irq.stat_regs.cik.disp_int_cont5 & DC_HPD6_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont5 &= ~DC_HPD6_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD6\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont5 & DC_HPD6_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont5 &= ~DC_HPD6_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD6\n");
+
 				break;
 			case 6:
-				if (rdev->irq.stat_regs.cik.disp_int & DC_HPD1_RX_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int &= ~DC_HPD1_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 1\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int & DC_HPD1_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int &= ~DC_HPD1_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 1\n");
+
 				break;
 			case 7:
-				if (rdev->irq.stat_regs.cik.disp_int_cont & DC_HPD2_RX_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont &= ~DC_HPD2_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 2\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont & DC_HPD2_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont &= ~DC_HPD2_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 2\n");
+
 				break;
 			case 8:
-				if (rdev->irq.stat_regs.cik.disp_int_cont2 & DC_HPD3_RX_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont2 &= ~DC_HPD3_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 3\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont2 & DC_HPD3_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont2 &= ~DC_HPD3_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 3\n");
+
 				break;
 			case 9:
-				if (rdev->irq.stat_regs.cik.disp_int_cont3 & DC_HPD4_RX_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont3 &= ~DC_HPD4_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 4\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont3 & DC_HPD4_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont3 &= ~DC_HPD4_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 4\n");
+
 				break;
 			case 10:
-				if (rdev->irq.stat_regs.cik.disp_int_cont4 & DC_HPD5_RX_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont4 &= ~DC_HPD5_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 5\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont4 & DC_HPD5_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont4 &= ~DC_HPD5_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 5\n");
+
 				break;
 			case 11:
-				if (rdev->irq.stat_regs.cik.disp_int_cont5 & DC_HPD6_RX_INTERRUPT) {
-					rdev->irq.stat_regs.cik.disp_int_cont5 &= ~DC_HPD6_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 6\n");
-				}
+				if (!(rdev->irq.stat_regs.cik.disp_int_cont5 & DC_HPD6_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.cik.disp_int_cont5 &= ~DC_HPD6_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 6\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index f86eb54e7763..d16f2eebd95e 100644
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -268,6 +268,17 @@ static void cik_sdma_gfx_stop(struct radeon_device *rdev)
 	}
 	rdev->ring[R600_RING_TYPE_DMA_INDEX].ready = false;
 	rdev->ring[CAYMAN_RING_TYPE_DMA1_INDEX].ready = false;
+
+	/* FIXME use something else than big hammer but after few days can not
+	 * seem to find good combination so reset SDMA blocks as it seems we
+	 * do not shut them down properly. This fix hibernation and does not
+	 * affect suspend to ram.
+	 */
+	WREG32(SRBM_SOFT_RESET, SOFT_RESET_SDMA | SOFT_RESET_SDMA1);
+	(void)RREG32(SRBM_SOFT_RESET);
+	udelay(50);
+	WREG32(SRBM_SOFT_RESET, 0);
+	(void)RREG32(SRBM_SOFT_RESET);
 }
 
 /**
diff --git a/drivers/gpu/drm/radeon/dce3_1_afmt.c b/drivers/gpu/drm/radeon/dce3_1_afmt.c
index f04205170b8a..cfa3a84a2af0 100644
--- a/drivers/gpu/drm/radeon/dce3_1_afmt.c
+++ b/drivers/gpu/drm/radeon/dce3_1_afmt.c
@@ -173,7 +173,7 @@ void dce3_2_hdmi_update_acr(struct drm_encoder *encoder, long offset,
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
 
-	WREG32(HDMI0_ACR_PACKET_CONTROL + offset,
+	WREG32(DCE3_HDMI0_ACR_PACKET_CONTROL + offset,
 		HDMI0_ACR_SOURCE |		/* select SW CTS value */
 		HDMI0_ACR_AUTO_SEND);	/* allow hw to sent ACR packets when required */
 
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 3a6d483a2c36..0acde1949c18 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -4924,7 +4924,7 @@ restart_ih:
 		return IRQ_NONE;
 
 	rptr = rdev->ih.rptr;
-	DRM_DEBUG("r600_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
+	DRM_DEBUG("evergreen_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
 
 	/* Order reading of wptr vs. reading of IH ring data */
 	rmb();
@@ -4942,23 +4942,27 @@ restart_ih:
 		case 1: /* D1 vblank/vline */
 			switch (src_data) {
 			case 0: /* D1 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[0]) {
-						drm_handle_vblank(rdev->ddev, 0);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_vblank(rdev, 0);
-					rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D1 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: D1 vblank - IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[0]) {
+					drm_handle_vblank(rdev->ddev, 0);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[0]))
+					radeon_crtc_handle_vblank(rdev, 0);
+				rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D1 vblank\n");
+
 				break;
 			case 1: /* D1 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D1 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: D1 vline - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D1 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -4968,23 +4972,27 @@ restart_ih:
 		case 2: /* D2 vblank/vline */
 			switch (src_data) {
 			case 0: /* D2 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[1]) {
-						drm_handle_vblank(rdev->ddev, 1);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_vblank(rdev, 1);
-					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D2 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: D2 vblank - IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[1]) {
+					drm_handle_vblank(rdev->ddev, 1);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[1]))
+					radeon_crtc_handle_vblank(rdev, 1);
+				rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D2 vblank\n");
+
 				break;
 			case 1: /* D2 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D2 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: D2 vline - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D2 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -4994,23 +5002,27 @@ restart_ih:
 		case 3: /* D3 vblank/vline */
 			switch (src_data) {
 			case 0: /* D3 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[2]) {
-						drm_handle_vblank(rdev->ddev, 2);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[2]))
-						radeon_crtc_handle_vblank(rdev, 2);
-					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D3 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: D3 vblank - IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[2]) {
+					drm_handle_vblank(rdev->ddev, 2);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[2]))
+					radeon_crtc_handle_vblank(rdev, 2);
+				rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D3 vblank\n");
+
 				break;
 			case 1: /* D3 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D3 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: D3 vline - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D3 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -5020,23 +5032,27 @@ restart_ih:
 		case 4: /* D4 vblank/vline */
 			switch (src_data) {
 			case 0: /* D4 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[3]) {
-						drm_handle_vblank(rdev->ddev, 3);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[3]))
-						radeon_crtc_handle_vblank(rdev, 3);
-					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D4 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: D4 vblank - IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[3]) {
+					drm_handle_vblank(rdev->ddev, 3);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[3]))
+					radeon_crtc_handle_vblank(rdev, 3);
+				rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D4 vblank\n");
+
 				break;
 			case 1: /* D4 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D4 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: D4 vline - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D4 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -5046,23 +5062,27 @@ restart_ih:
 		case 5: /* D5 vblank/vline */
 			switch (src_data) {
 			case 0: /* D5 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[4]) {
-						drm_handle_vblank(rdev->ddev, 4);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[4]))
-						radeon_crtc_handle_vblank(rdev, 4);
-					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D5 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: D5 vblank - IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[4]) {
+					drm_handle_vblank(rdev->ddev, 4);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[4]))
+					radeon_crtc_handle_vblank(rdev, 4);
+				rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D5 vblank\n");
+
 				break;
 			case 1: /* D5 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D5 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: D5 vline - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D5 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -5072,23 +5092,27 @@ restart_ih:
 		case 6: /* D6 vblank/vline */
 			switch (src_data) {
 			case 0: /* D6 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[5]) {
-						drm_handle_vblank(rdev->ddev, 5);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[5]))
-						radeon_crtc_handle_vblank(rdev, 5);
-					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D6 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: D6 vblank - IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[5]) {
+					drm_handle_vblank(rdev->ddev, 5);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[5]))
+					radeon_crtc_handle_vblank(rdev, 5);
+				rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D6 vblank\n");
+
 				break;
 			case 1: /* D6 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D6 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: D6 vline - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D6 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -5108,88 +5132,100 @@ restart_ih:
 		case 42: /* HPD hotplug */
 			switch (src_data) {
 			case 0:
-				if (rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int &= ~DC_HPD1_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD1\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int &= ~DC_HPD1_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD1\n");
 				break;
 			case 1:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont & DC_HPD2_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~DC_HPD2_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD2\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont & DC_HPD2_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont &= ~DC_HPD2_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD2\n");
 				break;
 			case 2:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & DC_HPD3_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~DC_HPD3_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD3\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont2 & DC_HPD3_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~DC_HPD3_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD3\n");
 				break;
 			case 3:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & DC_HPD4_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~DC_HPD4_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD4\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont3 & DC_HPD4_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~DC_HPD4_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD4\n");
 				break;
 			case 4:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & DC_HPD5_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~DC_HPD5_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD5\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont4 & DC_HPD5_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~DC_HPD5_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD5\n");
 				break;
 			case 5:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & DC_HPD6_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~DC_HPD6_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD6\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont5 & DC_HPD6_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~DC_HPD6_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD6\n");
 				break;
 			case 6:
-				if (rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int &= ~DC_HPD1_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 1\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int &= ~DC_HPD1_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 1\n");
 				break;
 			case 7:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont & DC_HPD2_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~DC_HPD2_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 2\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont & DC_HPD2_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont &= ~DC_HPD2_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 2\n");
 				break;
 			case 8:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & DC_HPD3_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~DC_HPD3_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 3\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont2 & DC_HPD3_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~DC_HPD3_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 3\n");
 				break;
 			case 9:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & DC_HPD4_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~DC_HPD4_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 4\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont3 & DC_HPD4_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~DC_HPD4_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 4\n");
 				break;
 			case 10:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & DC_HPD5_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~DC_HPD5_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 5\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont4 & DC_HPD5_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~DC_HPD5_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 5\n");
 				break;
 			case 11:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & DC_HPD6_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~DC_HPD6_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 6\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont5 & DC_HPD6_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~DC_HPD6_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 6\n");
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -5199,46 +5235,52 @@ restart_ih:
 		case 44: /* hdmi */
 			switch (src_data) {
 			case 0:
-				if (rdev->irq.stat_regs.evergreen.afmt_status1 & AFMT_AZ_FORMAT_WTRIG) {
-					rdev->irq.stat_regs.evergreen.afmt_status1 &= ~AFMT_AZ_FORMAT_WTRIG;
-					queue_hdmi = true;
-					DRM_DEBUG("IH: HDMI0\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.afmt_status1 & AFMT_AZ_FORMAT_WTRIG))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.afmt_status1 &= ~AFMT_AZ_FORMAT_WTRIG;
+				queue_hdmi = true;
+				DRM_DEBUG("IH: HDMI0\n");
 				break;
 			case 1:
-				if (rdev->irq.stat_regs.evergreen.afmt_status2 & AFMT_AZ_FORMAT_WTRIG) {
-					rdev->irq.stat_regs.evergreen.afmt_status2 &= ~AFMT_AZ_FORMAT_WTRIG;
-					queue_hdmi = true;
-					DRM_DEBUG("IH: HDMI1\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.afmt_status2 & AFMT_AZ_FORMAT_WTRIG))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.afmt_status2 &= ~AFMT_AZ_FORMAT_WTRIG;
+				queue_hdmi = true;
+				DRM_DEBUG("IH: HDMI1\n");
 				break;
 			case 2:
-				if (rdev->irq.stat_regs.evergreen.afmt_status3 & AFMT_AZ_FORMAT_WTRIG) {
-					rdev->irq.stat_regs.evergreen.afmt_status3 &= ~AFMT_AZ_FORMAT_WTRIG;
-					queue_hdmi = true;
-					DRM_DEBUG("IH: HDMI2\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.afmt_status3 & AFMT_AZ_FORMAT_WTRIG))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.afmt_status3 &= ~AFMT_AZ_FORMAT_WTRIG;
+				queue_hdmi = true;
+				DRM_DEBUG("IH: HDMI2\n");
 				break;
 			case 3:
-				if (rdev->irq.stat_regs.evergreen.afmt_status4 & AFMT_AZ_FORMAT_WTRIG) {
-					rdev->irq.stat_regs.evergreen.afmt_status4 &= ~AFMT_AZ_FORMAT_WTRIG;
-					queue_hdmi = true;
-					DRM_DEBUG("IH: HDMI3\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.afmt_status4 & AFMT_AZ_FORMAT_WTRIG))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.afmt_status4 &= ~AFMT_AZ_FORMAT_WTRIG;
+				queue_hdmi = true;
+				DRM_DEBUG("IH: HDMI3\n");
 				break;
 			case 4:
-				if (rdev->irq.stat_regs.evergreen.afmt_status5 & AFMT_AZ_FORMAT_WTRIG) {
-					rdev->irq.stat_regs.evergreen.afmt_status5 &= ~AFMT_AZ_FORMAT_WTRIG;
-					queue_hdmi = true;
-					DRM_DEBUG("IH: HDMI4\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.afmt_status5 & AFMT_AZ_FORMAT_WTRIG))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.afmt_status5 &= ~AFMT_AZ_FORMAT_WTRIG;
+				queue_hdmi = true;
+				DRM_DEBUG("IH: HDMI4\n");
 				break;
 			case 5:
-				if (rdev->irq.stat_regs.evergreen.afmt_status6 & AFMT_AZ_FORMAT_WTRIG) {
-					rdev->irq.stat_regs.evergreen.afmt_status6 &= ~AFMT_AZ_FORMAT_WTRIG;
-					queue_hdmi = true;
-					DRM_DEBUG("IH: HDMI5\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.afmt_status6 & AFMT_AZ_FORMAT_WTRIG))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.afmt_status6 &= ~AFMT_AZ_FORMAT_WTRIG;
+				queue_hdmi = true;
+				DRM_DEBUG("IH: HDMI5\n");
 				break;
 			default:
 				DRM_ERROR("Unhandled interrupt: %d %d\n", src_id, src_data);
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index 8e5aeeb058a5..158872eb78e4 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -2162,18 +2162,20 @@ static int cayman_startup(struct radeon_device *rdev)
 			DRM_ERROR("radeon: failed initializing UVD (%d).\n", r);
 	}
 
-	ring = &rdev->ring[TN_RING_TYPE_VCE1_INDEX];
-	if (ring->ring_size)
-		r = radeon_ring_init(rdev, ring, ring->ring_size, 0, 0x0);
+	if (rdev->family == CHIP_ARUBA) {
+		ring = &rdev->ring[TN_RING_TYPE_VCE1_INDEX];
+		if (ring->ring_size)
+			r = radeon_ring_init(rdev, ring, ring->ring_size, 0, 0x0);
 
-	ring = &rdev->ring[TN_RING_TYPE_VCE2_INDEX];
-	if (ring->ring_size)
-		r = radeon_ring_init(rdev, ring, ring->ring_size, 0, 0x0);
+		ring = &rdev->ring[TN_RING_TYPE_VCE2_INDEX];
+		if (ring->ring_size)
+			r = radeon_ring_init(rdev, ring, ring->ring_size, 0, 0x0);
 
-	if (!r)
-		r = vce_v1_0_init(rdev);
-	else if (r != -ENOENT)
-		DRM_ERROR("radeon: failed initializing VCE (%d).\n", r);
+		if (!r)
+			r = vce_v1_0_init(rdev);
+		if (r)
+			DRM_ERROR("radeon: failed initializing VCE (%d).\n", r);
+	}
 
 	r = radeon_ib_pool_init(rdev);
 	if (r) {
@@ -2396,7 +2398,8 @@ void cayman_fini(struct radeon_device *rdev)
 	radeon_irq_kms_fini(rdev);
 	uvd_v1_0_fini(rdev);
 	radeon_uvd_fini(rdev);
-	radeon_vce_fini(rdev);
+	if (rdev->family == CHIP_ARUBA)
+		radeon_vce_fini(rdev);
 	cayman_pcie_gart_fini(rdev);
 	r600_vram_scratch_fini(rdev);
 	radeon_gem_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index 35dafd77a639..4ea5b10ff5f4 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -4086,23 +4086,27 @@ restart_ih:
 		case 1: /* D1 vblank/vline */
 			switch (src_data) {
 			case 0: /* D1 vblank */
-				if (rdev->irq.stat_regs.r600.disp_int & LB_D1_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[0]) {
-						drm_handle_vblank(rdev->ddev, 0);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_vblank(rdev, 0);
-					rdev->irq.stat_regs.r600.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D1 vblank\n");
+				if (!(rdev->irq.stat_regs.r600.disp_int & LB_D1_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: D1 vblank - IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[0]) {
+					drm_handle_vblank(rdev->ddev, 0);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[0]))
+					radeon_crtc_handle_vblank(rdev, 0);
+				rdev->irq.stat_regs.r600.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D1 vblank\n");
+
 				break;
 			case 1: /* D1 vline */
-				if (rdev->irq.stat_regs.r600.disp_int & LB_D1_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.r600.disp_int &= ~LB_D1_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D1 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.r600.disp_int & LB_D1_VLINE_INTERRUPT))
+				    DRM_DEBUG("IH: D1 vline - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.r600.disp_int &= ~LB_D1_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D1 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -4112,23 +4116,27 @@ restart_ih:
 		case 5: /* D2 vblank/vline */
 			switch (src_data) {
 			case 0: /* D2 vblank */
-				if (rdev->irq.stat_regs.r600.disp_int & LB_D2_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[1]) {
-						drm_handle_vblank(rdev->ddev, 1);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_vblank(rdev, 1);
-					rdev->irq.stat_regs.r600.disp_int &= ~LB_D2_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D2 vblank\n");
+				if (!(rdev->irq.stat_regs.r600.disp_int & LB_D2_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: D2 vblank - IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[1]) {
+					drm_handle_vblank(rdev->ddev, 1);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[1]))
+					radeon_crtc_handle_vblank(rdev, 1);
+				rdev->irq.stat_regs.r600.disp_int &= ~LB_D2_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D2 vblank\n");
+
 				break;
 			case 1: /* D1 vline */
-				if (rdev->irq.stat_regs.r600.disp_int & LB_D2_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.r600.disp_int &= ~LB_D2_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D2 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.r600.disp_int & LB_D2_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: D2 vline - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.r600.disp_int &= ~LB_D2_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D2 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -4148,46 +4156,53 @@ restart_ih:
 		case 19: /* HPD/DAC hotplug */
 			switch (src_data) {
 			case 0:
-				if (rdev->irq.stat_regs.r600.disp_int & DC_HPD1_INTERRUPT) {
-					rdev->irq.stat_regs.r600.disp_int &= ~DC_HPD1_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD1\n");
-				}
+				if (!(rdev->irq.stat_regs.r600.disp_int & DC_HPD1_INTERRUPT))
+					DRM_DEBUG("IH: HPD1 - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.r600.disp_int &= ~DC_HPD1_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD1\n");
 				break;
 			case 1:
-				if (rdev->irq.stat_regs.r600.disp_int & DC_HPD2_INTERRUPT) {
-					rdev->irq.stat_regs.r600.disp_int &= ~DC_HPD2_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD2\n");
-				}
+				if (!(rdev->irq.stat_regs.r600.disp_int & DC_HPD2_INTERRUPT))
+					DRM_DEBUG("IH: HPD2 - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.r600.disp_int &= ~DC_HPD2_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD2\n");
 				break;
 			case 4:
-				if (rdev->irq.stat_regs.r600.disp_int_cont & DC_HPD3_INTERRUPT) {
-					rdev->irq.stat_regs.r600.disp_int_cont &= ~DC_HPD3_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD3\n");
-				}
+				if (!(rdev->irq.stat_regs.r600.disp_int_cont & DC_HPD3_INTERRUPT))
+					DRM_DEBUG("IH: HPD3 - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.r600.disp_int_cont &= ~DC_HPD3_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD3\n");
 				break;
 			case 5:
-				if (rdev->irq.stat_regs.r600.disp_int_cont & DC_HPD4_INTERRUPT) {
-					rdev->irq.stat_regs.r600.disp_int_cont &= ~DC_HPD4_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD4\n");
-				}
+				if (!(rdev->irq.stat_regs.r600.disp_int_cont & DC_HPD4_INTERRUPT))
+					DRM_DEBUG("IH: HPD4 - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.r600.disp_int_cont &= ~DC_HPD4_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD4\n");
 				break;
 			case 10:
-				if (rdev->irq.stat_regs.r600.disp_int_cont2 & DC_HPD5_INTERRUPT) {
-					rdev->irq.stat_regs.r600.disp_int_cont2 &= ~DC_HPD5_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD5\n");
-				}
+				if (!(rdev->irq.stat_regs.r600.disp_int_cont2 & DC_HPD5_INTERRUPT))
+					DRM_DEBUG("IH: HPD5 - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.r600.disp_int_cont2 &= ~DC_HPD5_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD5\n");
 				break;
 			case 12:
-				if (rdev->irq.stat_regs.r600.disp_int_cont2 & DC_HPD6_INTERRUPT) {
-					rdev->irq.stat_regs.r600.disp_int_cont2 &= ~DC_HPD6_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD6\n");
-				}
+				if (!(rdev->irq.stat_regs.r600.disp_int_cont2 & DC_HPD6_INTERRUPT))
+					DRM_DEBUG("IH: HPD6 - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.r600.disp_int_cont2 &= ~DC_HPD6_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD6\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -4197,18 +4212,22 @@ restart_ih:
 		case 21: /* hdmi */
 			switch (src_data) {
 			case 4:
-				if (rdev->irq.stat_regs.r600.hdmi0_status & HDMI0_AZ_FORMAT_WTRIG) {
-					rdev->irq.stat_regs.r600.hdmi0_status &= ~HDMI0_AZ_FORMAT_WTRIG;
-					queue_hdmi = true;
-					DRM_DEBUG("IH: HDMI0\n");
-				}
+				if (!(rdev->irq.stat_regs.r600.hdmi0_status & HDMI0_AZ_FORMAT_WTRIG))
+					DRM_DEBUG("IH: HDMI0 - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.r600.hdmi0_status &= ~HDMI0_AZ_FORMAT_WTRIG;
+				queue_hdmi = true;
+				DRM_DEBUG("IH: HDMI0\n");
+
 				break;
 			case 5:
-				if (rdev->irq.stat_regs.r600.hdmi1_status & HDMI0_AZ_FORMAT_WTRIG) {
-					rdev->irq.stat_regs.r600.hdmi1_status &= ~HDMI0_AZ_FORMAT_WTRIG;
-					queue_hdmi = true;
-					DRM_DEBUG("IH: HDMI1\n");
-				}
+				if (!(rdev->irq.stat_regs.r600.hdmi1_status & HDMI0_AZ_FORMAT_WTRIG))
+					DRM_DEBUG("IH: HDMI1 - IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.r600.hdmi1_status &= ~HDMI0_AZ_FORMAT_WTRIG;
+				queue_hdmi = true;
+				DRM_DEBUG("IH: HDMI1\n");
+
 				break;
 			default:
 				DRM_ERROR("Unhandled interrupt: %d %d\n", src_id, src_data);
diff --git a/drivers/gpu/drm/radeon/r600_cp.c b/drivers/gpu/drm/radeon/r600_cp.c
index 09e3f39925fa..98f9adaccc3d 100644
--- a/drivers/gpu/drm/radeon/r600_cp.c
+++ b/drivers/gpu/drm/radeon/r600_cp.c
@@ -2483,7 +2483,7 @@ int r600_cp_dispatch_texture(struct drm_device *dev,
 	struct drm_buf *buf;
 	u32 *buffer;
 	const u8 __user *data;
-	int size, pass_size;
+	unsigned int size, pass_size;
 	u64 src_offset, dst_offset;
 
 	if (!radeon_check_offset(dev_priv, tex->offset)) {
diff --git a/drivers/gpu/drm/radeon/radeon_audio.c b/drivers/gpu/drm/radeon/radeon_audio.c
index c89215275053..fa719c53449b 100644
--- a/drivers/gpu/drm/radeon/radeon_audio.c
+++ b/drivers/gpu/drm/radeon/radeon_audio.c
@@ -469,22 +469,22 @@ void radeon_audio_detect(struct drm_connector *connector,
 	dig = radeon_encoder->enc_priv;
 
 	if (status == connector_status_connected) {
-		struct radeon_connector *radeon_connector;
-		int sink_type;
-
 		if (!drm_detect_monitor_audio(radeon_connector_edid(connector))) {
 			radeon_encoder->audio = NULL;
 			return;
 		}
 
-		radeon_connector = to_radeon_connector(connector);
-		sink_type = radeon_dp_getsinktype(radeon_connector);
+		if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) {
+			struct radeon_connector *radeon_connector = to_radeon_connector(connector);
 
-		if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort &&
-			sink_type == CONNECTOR_OBJECT_ID_DISPLAYPORT)
-			radeon_encoder->audio = rdev->audio.dp_funcs;
-		else
+			if (radeon_dp_getsinktype(radeon_connector) ==
+			    CONNECTOR_OBJECT_ID_DISPLAYPORT)
+				radeon_encoder->audio = rdev->audio.dp_funcs;
+			else
+				radeon_encoder->audio = rdev->audio.hdmi_funcs;
+		} else {
 			radeon_encoder->audio = rdev->audio.hdmi_funcs;
+		}
 
 		dig->afmt->pin = radeon_audio_get_pin(connector->encoder);
 		radeon_audio_enable(rdev, dig->afmt->pin, 0xf);
diff --git a/drivers/gpu/drm/radeon/radeon_cursor.c b/drivers/gpu/drm/radeon/radeon_cursor.c
index 45e54060ee97..afaf346bd50e 100644
--- a/drivers/gpu/drm/radeon/radeon_cursor.c
+++ b/drivers/gpu/drm/radeon/radeon_cursor.c
@@ -91,15 +91,34 @@ static void radeon_show_cursor(struct drm_crtc *crtc)
 	struct radeon_device *rdev = crtc->dev->dev_private;
 
 	if (ASIC_IS_DCE4(rdev)) {
+		WREG32(EVERGREEN_CUR_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset,
+		       upper_32_bits(radeon_crtc->cursor_addr));
+		WREG32(EVERGREEN_CUR_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
+		       lower_32_bits(radeon_crtc->cursor_addr));
 		WREG32(RADEON_MM_INDEX, EVERGREEN_CUR_CONTROL + radeon_crtc->crtc_offset);
 		WREG32(RADEON_MM_DATA, EVERGREEN_CURSOR_EN |
 		       EVERGREEN_CURSOR_MODE(EVERGREEN_CURSOR_24_8_PRE_MULT) |
 		       EVERGREEN_CURSOR_URGENT_CONTROL(EVERGREEN_CURSOR_URGENT_1_2));
 	} else if (ASIC_IS_AVIVO(rdev)) {
+		if (rdev->family >= CHIP_RV770) {
+			if (radeon_crtc->crtc_id)
+				WREG32(R700_D2CUR_SURFACE_ADDRESS_HIGH,
+				       upper_32_bits(radeon_crtc->cursor_addr));
+			else
+				WREG32(R700_D1CUR_SURFACE_ADDRESS_HIGH,
+				       upper_32_bits(radeon_crtc->cursor_addr));
+		}
+
+		WREG32(AVIVO_D1CUR_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
+		       lower_32_bits(radeon_crtc->cursor_addr));
 		WREG32(RADEON_MM_INDEX, AVIVO_D1CUR_CONTROL + radeon_crtc->crtc_offset);
 		WREG32(RADEON_MM_DATA, AVIVO_D1CURSOR_EN |
 		       (AVIVO_D1CURSOR_MODE_24BPP << AVIVO_D1CURSOR_MODE_SHIFT));
 	} else {
+		/* offset is from DISP(2)_BASE_ADDRESS */
+		WREG32(RADEON_CUR_OFFSET + radeon_crtc->crtc_offset,
+		       radeon_crtc->cursor_addr - radeon_crtc->legacy_display_base_addr);
+
 		switch (radeon_crtc->crtc_id) {
 		case 0:
 			WREG32(RADEON_MM_INDEX, RADEON_CRTC_GEN_CNTL);
@@ -205,8 +224,9 @@ static int radeon_cursor_move_locked(struct drm_crtc *crtc, int x, int y)
 			| (x << 16)
 			| y));
 		/* offset is from DISP(2)_BASE_ADDRESS */
-		WREG32(RADEON_CUR_OFFSET + radeon_crtc->crtc_offset, (radeon_crtc->legacy_cursor_offset +
-								      (yorigin * 256)));
+		WREG32(RADEON_CUR_OFFSET + radeon_crtc->crtc_offset,
+		       radeon_crtc->cursor_addr - radeon_crtc->legacy_display_base_addr +
+		       yorigin * 256);
 	}
 
 	radeon_crtc->cursor_x = x;
@@ -227,53 +247,6 @@ int radeon_crtc_cursor_move(struct drm_crtc *crtc,
 	return ret;
 }
 
-static int radeon_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj)
-{
-	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
-	struct radeon_device *rdev = crtc->dev->dev_private;
-	struct radeon_bo *robj = gem_to_radeon_bo(obj);
-	uint64_t gpu_addr;
-	int ret;
-
-	ret = radeon_bo_reserve(robj, false);
-	if (unlikely(ret != 0))
-		goto fail;
-	/* Only 27 bit offset for legacy cursor */
-	ret = radeon_bo_pin_restricted(robj, RADEON_GEM_DOMAIN_VRAM,
-				       ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27,
-				       &gpu_addr);
-	radeon_bo_unreserve(robj);
-	if (ret)
-		goto fail;
-
-	if (ASIC_IS_DCE4(rdev)) {
-		WREG32(EVERGREEN_CUR_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset,
-		       upper_32_bits(gpu_addr));
-		WREG32(EVERGREEN_CUR_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
-		       gpu_addr & 0xffffffff);
-	} else if (ASIC_IS_AVIVO(rdev)) {
-		if (rdev->family >= CHIP_RV770) {
-			if (radeon_crtc->crtc_id)
-				WREG32(R700_D2CUR_SURFACE_ADDRESS_HIGH, upper_32_bits(gpu_addr));
-			else
-				WREG32(R700_D1CUR_SURFACE_ADDRESS_HIGH, upper_32_bits(gpu_addr));
-		}
-		WREG32(AVIVO_D1CUR_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
-		       gpu_addr & 0xffffffff);
-	} else {
-		radeon_crtc->legacy_cursor_offset = gpu_addr - radeon_crtc->legacy_display_base_addr;
-		/* offset is from DISP(2)_BASE_ADDRESS */
-		WREG32(RADEON_CUR_OFFSET + radeon_crtc->crtc_offset, radeon_crtc->legacy_cursor_offset);
-	}
-
-	return 0;
-
-fail:
-	drm_gem_object_unreference_unlocked(obj);
-
-	return ret;
-}
-
 int radeon_crtc_cursor_set2(struct drm_crtc *crtc,
 			    struct drm_file *file_priv,
 			    uint32_t handle,
@@ -283,7 +256,9 @@ int radeon_crtc_cursor_set2(struct drm_crtc *crtc,
 			    int32_t hot_y)
 {
 	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
+	struct radeon_device *rdev = crtc->dev->dev_private;
 	struct drm_gem_object *obj;
+	struct radeon_bo *robj;
 	int ret;
 
 	if (!handle) {
@@ -305,6 +280,23 @@ int radeon_crtc_cursor_set2(struct drm_crtc *crtc,
 		return -ENOENT;
 	}
 
+	robj = gem_to_radeon_bo(obj);
+	ret = radeon_bo_reserve(robj, false);
+	if (ret != 0) {
+		drm_gem_object_unreference_unlocked(obj);
+		return ret;
+	}
+	/* Only 27 bit offset for legacy cursor */
+	ret = radeon_bo_pin_restricted(robj, RADEON_GEM_DOMAIN_VRAM,
+				       ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27,
+				       &radeon_crtc->cursor_addr);
+	radeon_bo_unreserve(robj);
+	if (ret) {
+		DRM_ERROR("Failed to pin new cursor BO (%d)\n", ret);
+		drm_gem_object_unreference_unlocked(obj);
+		return ret;
+	}
+
 	radeon_crtc->cursor_width = width;
 	radeon_crtc->cursor_height = height;
 
@@ -323,13 +315,7 @@ int radeon_crtc_cursor_set2(struct drm_crtc *crtc,
 		radeon_crtc->cursor_hot_y = hot_y;
 	}
 
-	ret = radeon_set_cursor(crtc, obj);
-
-	if (ret)
-		DRM_ERROR("radeon_set_cursor returned %d, not changing cursor\n",
-			  ret);
-	else
-		radeon_show_cursor(crtc);
+	radeon_show_cursor(crtc);
 
 	radeon_lock_cursor(crtc, false);
 
@@ -341,8 +327,7 @@ unpin:
 			radeon_bo_unpin(robj);
 			radeon_bo_unreserve(robj);
 		}
-		if (radeon_crtc->cursor_bo != obj)
-			drm_gem_object_unreference_unlocked(radeon_crtc->cursor_bo);
+		drm_gem_object_unreference_unlocked(radeon_crtc->cursor_bo);
 	}
 
 	radeon_crtc->cursor_bo = obj;
@@ -360,7 +345,6 @@ unpin:
 void radeon_cursor_reset(struct drm_crtc *crtc)
 {
 	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
-	int ret;
 
 	if (radeon_crtc->cursor_bo) {
 		radeon_lock_cursor(crtc, true);
@@ -368,12 +352,7 @@ void radeon_cursor_reset(struct drm_crtc *crtc)
 		radeon_cursor_move_locked(crtc, radeon_crtc->cursor_x,
 					  radeon_crtc->cursor_y);
 
-		ret = radeon_set_cursor(crtc, radeon_crtc->cursor_bo);
-		if (ret)
-			DRM_ERROR("radeon_set_cursor returned %d, not showing "
-				  "cursor\n", ret);
-		else
-			radeon_show_cursor(crtc);
+		radeon_show_cursor(crtc);
 
 		radeon_lock_cursor(crtc, false);
 	}
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 13e207e0dff0..d8319dae8358 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -1080,6 +1080,22 @@ static bool radeon_check_pot_argument(int arg)
 }
 
 /**
+ * Determine a sensible default GART size according to ASIC family.
+ *
+ * @family ASIC family name
+ */
+static int radeon_gart_size_auto(enum radeon_family family)
+{
+	/* default to a larger gart size on newer asics */
+	if (family >= CHIP_TAHITI)
+		return 2048;
+	else if (family >= CHIP_RV770)
+		return 1024;
+	else
+		return 512;
+}
+
+/**
  * radeon_check_arguments - validate module params
  *
  * @rdev: radeon_device pointer
@@ -1097,27 +1113,17 @@ static void radeon_check_arguments(struct radeon_device *rdev)
 	}
 
 	if (radeon_gart_size == -1) {
-		/* default to a larger gart size on newer asics */
-		if (rdev->family >= CHIP_RV770)
-			radeon_gart_size = 1024;
-		else
-			radeon_gart_size = 512;
+		radeon_gart_size = radeon_gart_size_auto(rdev->family);
 	}
 	/* gtt size must be power of two and greater or equal to 32M */
 	if (radeon_gart_size < 32) {
 		dev_warn(rdev->dev, "gart size (%d) too small\n",
 				radeon_gart_size);
-		if (rdev->family >= CHIP_RV770)
-			radeon_gart_size = 1024;
-		else
-			radeon_gart_size = 512;
+		radeon_gart_size = radeon_gart_size_auto(rdev->family);
 	} else if (!radeon_check_pot_argument(radeon_gart_size)) {
 		dev_warn(rdev->dev, "gart size (%d) must be a power of 2\n",
 				radeon_gart_size);
-		if (rdev->family >= CHIP_RV770)
-			radeon_gart_size = 1024;
-		else
-			radeon_gart_size = 512;
+		radeon_gart_size = radeon_gart_size_auto(rdev->family);
 	}
 	rdev->mc.gtt_size = (uint64_t)radeon_gart_size << 20;
 
@@ -1463,6 +1469,21 @@ int radeon_device_init(struct radeon_device *rdev,
 	if (r)
 		DRM_ERROR("ib ring test failed (%d).\n", r);
 
+	/*
+	 * Turks/Thames GPU will freeze whole laptop if DPM is not restarted
+	 * after the CP ring have chew one packet at least. Hence here we stop
+	 * and restart DPM after the radeon_ib_ring_tests().
+	 */
+	if (rdev->pm.dpm_enabled &&
+	    (rdev->pm.pm_method == PM_METHOD_DPM) &&
+	    (rdev->family == CHIP_TURKS) &&
+	    (rdev->flags & RADEON_IS_MOBILITY)) {
+		mutex_lock(&rdev->pm.mutex);
+		radeon_dpm_disable(rdev);
+		radeon_dpm_enable(rdev);
+		mutex_unlock(&rdev->pm.mutex);
+	}
+
 	if ((radeon_testing & 1)) {
 		if (rdev->accel_working)
 			radeon_test_moves(rdev);
@@ -1557,11 +1578,21 @@ int radeon_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon)
 		drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
 	}
 
-	/* unpin the front buffers */
+	/* unpin the front buffers and cursors */
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
 		struct radeon_framebuffer *rfb = to_radeon_framebuffer(crtc->primary->fb);
 		struct radeon_bo *robj;
 
+		if (radeon_crtc->cursor_bo) {
+			struct radeon_bo *robj = gem_to_radeon_bo(radeon_crtc->cursor_bo);
+			r = radeon_bo_reserve(robj, false);
+			if (r == 0) {
+				radeon_bo_unpin(robj);
+				radeon_bo_unreserve(robj);
+			}
+		}
+
 		if (rfb == NULL || rfb->obj == NULL) {
 			continue;
 		}
@@ -1624,6 +1655,7 @@ int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon)
 {
 	struct drm_connector *connector;
 	struct radeon_device *rdev = dev->dev_private;
+	struct drm_crtc *crtc;
 	int r;
 
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
@@ -1663,6 +1695,27 @@ int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon)
 
 	radeon_restore_bios_scratch_regs(rdev);
 
+	/* pin cursors */
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
+
+		if (radeon_crtc->cursor_bo) {
+			struct radeon_bo *robj = gem_to_radeon_bo(radeon_crtc->cursor_bo);
+			r = radeon_bo_reserve(robj, false);
+			if (r == 0) {
+				/* Only 27 bit offset for legacy cursor */
+				r = radeon_bo_pin_restricted(robj,
+							     RADEON_GEM_DOMAIN_VRAM,
+							     ASIC_IS_AVIVO(rdev) ?
+							     0 : 1 << 27,
+							     &radeon_crtc->cursor_addr);
+				if (r != 0)
+					DRM_ERROR("Failed to pin cursor BO (%d)\n", r);
+				radeon_bo_unreserve(robj);
+			}
+		}
+	}
+
 	/* init dig PHYs, disp eng pll */
 	if (rdev->is_atom_bios) {
 		radeon_atom_encoder_init(rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_dp_mst.c b/drivers/gpu/drm/radeon/radeon_dp_mst.c
index 2b98ed3e684d..257b10be5cda 100644
--- a/drivers/gpu/drm/radeon/radeon_dp_mst.c
+++ b/drivers/gpu/drm/radeon/radeon_dp_mst.c
@@ -663,12 +663,17 @@ int
 radeon_dp_mst_probe(struct radeon_connector *radeon_connector)
 {
 	struct radeon_connector_atom_dig *dig_connector = radeon_connector->con_priv;
+	struct drm_device *dev = radeon_connector->base.dev;
+	struct radeon_device *rdev = dev->dev_private;
 	int ret;
 	u8 msg[1];
 
 	if (!radeon_mst)
 		return 0;
 
+	if (!ASIC_IS_DCE5(rdev))
+		return 0;
+
 	if (dig_connector->dpcd[DP_DPCD_REV] < 0x12)
 		return 0;
 
diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c
index ac3c1310b953..013ec7106e55 100644
--- a/drivers/gpu/drm/radeon/radeon_gem.c
+++ b/drivers/gpu/drm/radeon/radeon_gem.c
@@ -428,7 +428,6 @@ int radeon_gem_mmap_ioctl(struct drm_device *dev, void *data,
 int radeon_gem_busy_ioctl(struct drm_device *dev, void *data,
 			  struct drm_file *filp)
 {
-	struct radeon_device *rdev = dev->dev_private;
 	struct drm_radeon_gem_busy *args = data;
 	struct drm_gem_object *gobj;
 	struct radeon_bo *robj;
@@ -440,10 +439,16 @@ int radeon_gem_busy_ioctl(struct drm_device *dev, void *data,
 		return -ENOENT;
 	}
 	robj = gem_to_radeon_bo(gobj);
-	r = radeon_bo_wait(robj, &cur_placement, true);
+
+	r = reservation_object_test_signaled_rcu(robj->tbo.resv, true);
+	if (r == 0)
+		r = -EBUSY;
+	else
+		r = 0;
+
+	cur_placement = ACCESS_ONCE(robj->tbo.mem.mem_type);
 	args->domain = radeon_mem_type_to_domain(cur_placement);
 	drm_gem_object_unreference_unlocked(gobj);
-	r = radeon_gem_handle_lockup(rdev, r);
 	return r;
 }
 
@@ -471,6 +476,7 @@ int radeon_gem_wait_idle_ioctl(struct drm_device *dev, void *data,
 		r = ret;
 
 	/* Flush HDP cache via MMIO if necessary */
+	cur_placement = ACCESS_ONCE(robj->tbo.mem.mem_type);
 	if (rdev->asic->mmio_hdp_flush &&
 	    radeon_mem_type_to_domain(cur_placement) == RADEON_GEM_DOMAIN_VRAM)
 		robj->rdev->asic->mmio_hdp_flush(rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 9632e886ddc3..4a119c255ba9 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -576,6 +576,9 @@ static int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		if (radeon_get_allowed_info_register(rdev, *value, value))
 			return -EINVAL;
 		break;
+	case RADEON_INFO_VA_UNMAP_WORKING:
+		*value = true;
+		break;
 	case RADEON_INFO_GPU_RESET_COUNTER:
 		*value = atomic_read(&rdev->gpu_reset_counter);
 		break;
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 6de5459316b5..07909d817381 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -343,7 +343,6 @@ struct radeon_crtc {
 	int max_cursor_width;
 	int max_cursor_height;
 	uint32_t legacy_display_base_addr;
-	uint32_t legacy_cursor_offset;
 	enum radeon_rmx_type rmx_type;
 	u8 h_border;
 	u8 v_border;
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index edafd3c2b170..06ac59fe332a 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -719,7 +719,7 @@ static int radeon_ttm_tt_populate(struct ttm_tt *ttm)
 		return 0;
 
 	if (gtt && gtt->userptr) {
-		ttm->sg = kcalloc(1, sizeof(struct sg_table), GFP_KERNEL);
+		ttm->sg = kzalloc(sizeof(struct sg_table), GFP_KERNEL);
 		if (!ttm->sg)
 			return -ENOMEM;
 
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index 9739ded91b7a..48d97c040f49 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -457,14 +457,16 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 		/* make sure object fit at this offset */
 		eoffset = soffset + size;
 		if (soffset >= eoffset) {
-			return -EINVAL;
+			r = -EINVAL;
+			goto error_unreserve;
 		}
 
 		last_pfn = eoffset / RADEON_GPU_PAGE_SIZE;
 		if (last_pfn > rdev->vm_manager.max_pfn) {
 			dev_err(rdev->dev, "va above limit (0x%08X > 0x%08X)\n",
 				last_pfn, rdev->vm_manager.max_pfn);
-			return -EINVAL;
+			r = -EINVAL;
+			goto error_unreserve;
 		}
 
 	} else {
@@ -485,42 +487,41 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 				"(bo %p 0x%010lx 0x%010lx)\n", bo_va->bo,
 				soffset, tmp->bo, tmp->it.start, tmp->it.last);
 			mutex_unlock(&vm->mutex);
-			return -EINVAL;
+			r = -EINVAL;
+			goto error_unreserve;
 		}
 	}
 
 	if (bo_va->it.start || bo_va->it.last) {
-		spin_lock(&vm->status_lock);
-		if (list_empty(&bo_va->vm_status)) {
-			/* add a clone of the bo_va to clear the old address */
-			struct radeon_bo_va *tmp;
-			spin_unlock(&vm->status_lock);
-			tmp = kzalloc(sizeof(struct radeon_bo_va), GFP_KERNEL);
-			if (!tmp) {
-				mutex_unlock(&vm->mutex);
-				return -ENOMEM;
-			}
-			tmp->it.start = bo_va->it.start;
-			tmp->it.last = bo_va->it.last;
-			tmp->vm = vm;
-			tmp->bo = radeon_bo_ref(bo_va->bo);
-			spin_lock(&vm->status_lock);
-			list_add(&tmp->vm_status, &vm->freed);
+		/* add a clone of the bo_va to clear the old address */
+		struct radeon_bo_va *tmp;
+		tmp = kzalloc(sizeof(struct radeon_bo_va), GFP_KERNEL);
+		if (!tmp) {
+			mutex_unlock(&vm->mutex);
+			r = -ENOMEM;
+			goto error_unreserve;
 		}
-		spin_unlock(&vm->status_lock);
+		tmp->it.start = bo_va->it.start;
+		tmp->it.last = bo_va->it.last;
+		tmp->vm = vm;
+		tmp->bo = radeon_bo_ref(bo_va->bo);
 
 		interval_tree_remove(&bo_va->it, &vm->va);
+		spin_lock(&vm->status_lock);
 		bo_va->it.start = 0;
 		bo_va->it.last = 0;
+		list_del_init(&bo_va->vm_status);
+		list_add(&tmp->vm_status, &vm->freed);
+		spin_unlock(&vm->status_lock);
 	}
 
 	if (soffset || eoffset) {
+		spin_lock(&vm->status_lock);
 		bo_va->it.start = soffset;
 		bo_va->it.last = eoffset - 1;
-		interval_tree_insert(&bo_va->it, &vm->va);
-		spin_lock(&vm->status_lock);
 		list_add(&bo_va->vm_status, &vm->cleared);
 		spin_unlock(&vm->status_lock);
+		interval_tree_insert(&bo_va->it, &vm->va);
 	}
 
 	bo_va->flags = flags;
@@ -555,7 +556,6 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 		r = radeon_vm_clear_bo(rdev, pt);
 		if (r) {
 			radeon_bo_unref(&pt);
-			radeon_bo_reserve(bo_va->bo, false);
 			return r;
 		}
 
@@ -575,6 +575,10 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev,
 
 	mutex_unlock(&vm->mutex);
 	return 0;
+
+error_unreserve:
+	radeon_bo_unreserve(bo_va->bo);
+	return r;
 }
 
 /**
@@ -1122,12 +1126,12 @@ void radeon_vm_bo_rmv(struct radeon_device *rdev,
 		interval_tree_remove(&bo_va->it, &vm->va);
 
 	spin_lock(&vm->status_lock);
-	if (list_empty(&bo_va->vm_status)) {
+	list_del(&bo_va->vm_status);
+	if (bo_va->it.start || bo_va->it.last) {
 		bo_va->bo = radeon_bo_ref(bo_va->bo);
 		list_add(&bo_va->vm_status, &vm->freed);
 	} else {
 		radeon_fence_unref(&bo_va->last_pt_update);
-		list_del(&bo_va->vm_status);
 		kfree(bo_va);
 	}
 	spin_unlock(&vm->status_lock);
@@ -1151,7 +1155,8 @@ void radeon_vm_bo_invalidate(struct radeon_device *rdev,
 
 	list_for_each_entry(bo_va, &bo->va, bo_list) {
 		spin_lock(&bo_va->vm->status_lock);
-		if (list_empty(&bo_va->vm_status))
+		if (list_empty(&bo_va->vm_status) &&
+		    (bo_va->it.start || bo_va->it.last))
 			list_add(&bo_va->vm_status, &bo_va->vm->invalidated);
 		spin_unlock(&bo_va->vm->status_lock);
 	}
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 26388b5dd6ed..07037e32dea3 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -6466,23 +6466,27 @@ restart_ih:
 		case 1: /* D1 vblank/vline */
 			switch (src_data) {
 			case 0: /* D1 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[0]) {
-						drm_handle_vblank(rdev->ddev, 0);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[0]))
-						radeon_crtc_handle_vblank(rdev, 0);
-					rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D1 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[0]) {
+					drm_handle_vblank(rdev->ddev, 0);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[0]))
+					radeon_crtc_handle_vblank(rdev, 0);
+				rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D1 vblank\n");
+
 				break;
 			case 1: /* D1 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D1 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int & LB_D1_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D1 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -6492,23 +6496,27 @@ restart_ih:
 		case 2: /* D2 vblank/vline */
 			switch (src_data) {
 			case 0: /* D2 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[1]) {
-						drm_handle_vblank(rdev->ddev, 1);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[1]))
-						radeon_crtc_handle_vblank(rdev, 1);
-					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D2 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[1]) {
+					drm_handle_vblank(rdev->ddev, 1);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[1]))
+					radeon_crtc_handle_vblank(rdev, 1);
+				rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D2 vblank\n");
+
 				break;
 			case 1: /* D2 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D2 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont & LB_D2_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D2 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -6518,23 +6526,27 @@ restart_ih:
 		case 3: /* D3 vblank/vline */
 			switch (src_data) {
 			case 0: /* D3 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[2]) {
-						drm_handle_vblank(rdev->ddev, 2);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[2]))
-						radeon_crtc_handle_vblank(rdev, 2);
-					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D3 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[2]) {
+					drm_handle_vblank(rdev->ddev, 2);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[2]))
+					radeon_crtc_handle_vblank(rdev, 2);
+				rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D3 vblank\n");
+
 				break;
 			case 1: /* D3 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D3 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont2 & LB_D3_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D3 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -6544,23 +6556,27 @@ restart_ih:
 		case 4: /* D4 vblank/vline */
 			switch (src_data) {
 			case 0: /* D4 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[3]) {
-						drm_handle_vblank(rdev->ddev, 3);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[3]))
-						radeon_crtc_handle_vblank(rdev, 3);
-					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D4 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[3]) {
+					drm_handle_vblank(rdev->ddev, 3);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[3]))
+					radeon_crtc_handle_vblank(rdev, 3);
+				rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D4 vblank\n");
+
 				break;
 			case 1: /* D4 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D4 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont3 & LB_D4_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D4 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -6570,23 +6586,27 @@ restart_ih:
 		case 5: /* D5 vblank/vline */
 			switch (src_data) {
 			case 0: /* D5 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[4]) {
-						drm_handle_vblank(rdev->ddev, 4);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[4]))
-						radeon_crtc_handle_vblank(rdev, 4);
-					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D5 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[4]) {
+					drm_handle_vblank(rdev->ddev, 4);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[4]))
+					radeon_crtc_handle_vblank(rdev, 4);
+				rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D5 vblank\n");
+
 				break;
 			case 1: /* D5 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D5 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont4 & LB_D5_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D5 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -6596,23 +6616,27 @@ restart_ih:
 		case 6: /* D6 vblank/vline */
 			switch (src_data) {
 			case 0: /* D6 vblank */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VBLANK_INTERRUPT) {
-					if (rdev->irq.crtc_vblank_int[5]) {
-						drm_handle_vblank(rdev->ddev, 5);
-						rdev->pm.vblank_sync = true;
-						wake_up(&rdev->irq.vblank_queue);
-					}
-					if (atomic_read(&rdev->irq.pflip[5]))
-						radeon_crtc_handle_vblank(rdev, 5);
-					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
-					DRM_DEBUG("IH: D6 vblank\n");
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VBLANK_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				if (rdev->irq.crtc_vblank_int[5]) {
+					drm_handle_vblank(rdev->ddev, 5);
+					rdev->pm.vblank_sync = true;
+					wake_up(&rdev->irq.vblank_queue);
 				}
+				if (atomic_read(&rdev->irq.pflip[5]))
+					radeon_crtc_handle_vblank(rdev, 5);
+				rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
+				DRM_DEBUG("IH: D6 vblank\n");
+
 				break;
 			case 1: /* D6 vline */
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VLINE_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VLINE_INTERRUPT;
-					DRM_DEBUG("IH: D6 vline\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont5 & LB_D6_VLINE_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VLINE_INTERRUPT;
+				DRM_DEBUG("IH: D6 vline\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
@@ -6632,88 +6656,112 @@ restart_ih:
 		case 42: /* HPD hotplug */
 			switch (src_data) {
 			case 0:
-				if (rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int &= ~DC_HPD1_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD1\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int &= ~DC_HPD1_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD1\n");
+
 				break;
 			case 1:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont & DC_HPD2_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~DC_HPD2_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD2\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont & DC_HPD2_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont &= ~DC_HPD2_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD2\n");
+
 				break;
 			case 2:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & DC_HPD3_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~DC_HPD3_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD3\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont2 & DC_HPD3_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~DC_HPD3_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD3\n");
+
 				break;
 			case 3:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & DC_HPD4_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~DC_HPD4_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD4\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont3 & DC_HPD4_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~DC_HPD4_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD4\n");
+
 				break;
 			case 4:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & DC_HPD5_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~DC_HPD5_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD5\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont4 & DC_HPD5_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~DC_HPD5_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD5\n");
+
 				break;
 			case 5:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & DC_HPD6_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~DC_HPD6_INTERRUPT;
-					queue_hotplug = true;
-					DRM_DEBUG("IH: HPD6\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont5 & DC_HPD6_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~DC_HPD6_INTERRUPT;
+				queue_hotplug = true;
+				DRM_DEBUG("IH: HPD6\n");
+
 				break;
 			case 6:
-				if (rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int &= ~DC_HPD1_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 1\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int & DC_HPD1_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int &= ~DC_HPD1_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 1\n");
+
 				break;
 			case 7:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont & DC_HPD2_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont &= ~DC_HPD2_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 2\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont & DC_HPD2_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont &= ~DC_HPD2_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 2\n");
+
 				break;
 			case 8:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont2 & DC_HPD3_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~DC_HPD3_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 3\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont2 & DC_HPD3_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~DC_HPD3_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 3\n");
+
 				break;
 			case 9:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont3 & DC_HPD4_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~DC_HPD4_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 4\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont3 & DC_HPD4_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~DC_HPD4_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 4\n");
+
 				break;
 			case 10:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont4 & DC_HPD5_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~DC_HPD5_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 5\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont4 & DC_HPD5_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~DC_HPD5_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 5\n");
+
 				break;
 			case 11:
-				if (rdev->irq.stat_regs.evergreen.disp_int_cont5 & DC_HPD6_RX_INTERRUPT) {
-					rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~DC_HPD6_RX_INTERRUPT;
-					queue_dp = true;
-					DRM_DEBUG("IH: HPD_RX 6\n");
-				}
+				if (!(rdev->irq.stat_regs.evergreen.disp_int_cont5 & DC_HPD6_RX_INTERRUPT))
+					DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
+
+				rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~DC_HPD6_RX_INTERRUPT;
+				queue_dp = true;
+				DRM_DEBUG("IH: HPD_RX 6\n");
+
 				break;
 			default:
 				DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data);
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
index 3962176ee713..01b558fe3695 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
@@ -21,6 +21,7 @@
 #include <drm/drm_fb_helper.h>
 #include <linux/dma-mapping.h>
 #include <linux/pm_runtime.h>
+#include <linux/module.h>
 #include <linux/of_graph.h>
 #include <linux/component.h>
 
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
index 4557f335a8a5..dc65161d7cad 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
@@ -19,6 +19,7 @@
 #include <drm/drm_plane_helper.h>
 
 #include <linux/kernel.h>
+#include <linux/module.h>
 #include <linux/platform_device.h>
 #include <linux/clk.h>
 #include <linux/of.h>
diff --git a/drivers/gpu/drm/vgem/vgem_drv.c b/drivers/gpu/drm/vgem/vgem_drv.c
index 7a207ca547be..6394547cf67a 100644
--- a/drivers/gpu/drm/vgem/vgem_drv.c
+++ b/drivers/gpu/drm/vgem/vgem_drv.c
@@ -328,6 +328,8 @@ static int __init vgem_init(void)
 		goto out;
 	}
 
+	drm_dev_set_unique(vgem_device, "vgem");
+
 	ret  = drm_dev_register(vgem_device, 0);
 
 	if (ret)
diff --git a/drivers/gpu/ipu-v3/ipu-common.c b/drivers/gpu/ipu-v3/ipu-common.c
index 67bab5c36056..6d2f39d36e44 100644
--- a/drivers/gpu/ipu-v3/ipu-common.c
+++ b/drivers/gpu/ipu-v3/ipu-common.c
@@ -1119,10 +1119,9 @@ static int ipu_irq_init(struct ipu_soc *ipu)
 		ct->regs.mask = IPU_INT_CTRL(i / 32);
 	}
 
-	irq_set_chained_handler(ipu->irq_sync, ipu_irq_handler);
-	irq_set_handler_data(ipu->irq_sync, ipu);
-	irq_set_chained_handler(ipu->irq_err, ipu_err_irq_handler);
-	irq_set_handler_data(ipu->irq_err, ipu);
+	irq_set_chained_handler_and_data(ipu->irq_sync, ipu_irq_handler, ipu);
+	irq_set_chained_handler_and_data(ipu->irq_err, ipu_err_irq_handler,
+					 ipu);
 
 	return 0;
 }
@@ -1131,10 +1130,8 @@ static void ipu_irq_exit(struct ipu_soc *ipu)
 {
 	int i, irq;
 
-	irq_set_chained_handler(ipu->irq_err, NULL);
-	irq_set_handler_data(ipu->irq_err, NULL);
-	irq_set_chained_handler(ipu->irq_sync, NULL);
-	irq_set_handler_data(ipu->irq_sync, NULL);
+	irq_set_chained_handler_and_data(ipu->irq_err, NULL, NULL);
+	irq_set_chained_handler_and_data(ipu->irq_sync, NULL, NULL);
 
 	/* TODO: remove irq_domain_generic_chips */