Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net

Conflicts:
	drivers/net/ethernet/emulex/benet/be.h
	drivers/net/usb/qmi_wwan.c
	drivers/net/wireless/brcm80211/brcmfmac/dhd_bus.h
	include/net/netfilter/nf_conntrack_synproxy.h
	include/net/secure_seq.h

The conflicts are of two varieties:

1) Conflicts with Joe Perches's 'extern' removal from header file
   function declarations.  Usually it's an argument signature change
   or a function being added/removed.  The resolutions are trivial.

2) Some overlapping changes in qmi_wwan.c and be.h, one commit adds
   a new value, another changes an existing value.  That sort of
   thing.

Signed-off-by: David S. Miller <davem@davemloft.net>

44 files changed, 538 insertions, 272 deletions

diff --git a/drivers/gpu/drm/drm_context.c b/drivers/gpu/drm/drm_context.c
index b4fb86d89850..224ff965bcf7 100644
--- a/drivers/gpu/drm/drm_context.c
+++ b/drivers/gpu/drm/drm_context.c
@@ -42,6 +42,10 @@
 
 #include <drm/drmP.h>
 
+/******************************************************************/
+/** \name Context bitmap support */
+/*@{*/
+
 /**
  * Free a handle from the context bitmap.
  *
@@ -52,48 +56,13 @@
  * in drm_device::ctx_idr, while holding the drm_device::struct_mutex
  * lock.
  */
-static void drm_ctxbitmap_free(struct drm_device * dev, int ctx_handle)
+void drm_ctxbitmap_free(struct drm_device * dev, int ctx_handle)
 {
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return;
-
 	mutex_lock(&dev->struct_mutex);
 	idr_remove(&dev->ctx_idr, ctx_handle);
 	mutex_unlock(&dev->struct_mutex);
 }
 
-/******************************************************************/
-/** \name Context bitmap support */
-/*@{*/
-
-void drm_legacy_ctxbitmap_release(struct drm_device *dev,
-				  struct drm_file *file_priv)
-{
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return;
-
-	mutex_lock(&dev->ctxlist_mutex);
-	if (!list_empty(&dev->ctxlist)) {
-		struct drm_ctx_list *pos, *n;
-
-		list_for_each_entry_safe(pos, n, &dev->ctxlist, head) {
-			if (pos->tag == file_priv &&
-			    pos->handle != DRM_KERNEL_CONTEXT) {
-				if (dev->driver->context_dtor)
-					dev->driver->context_dtor(dev,
-								  pos->handle);
-
-				drm_ctxbitmap_free(dev, pos->handle);
-
-				list_del(&pos->head);
-				kfree(pos);
-				--dev->ctx_count;
-			}
-		}
-	}
-	mutex_unlock(&dev->ctxlist_mutex);
-}
-
 /**
  * Context bitmap allocation.
  *
@@ -121,12 +90,10 @@ static int drm_ctxbitmap_next(struct drm_device * dev)
  *
  * Initialise the drm_device::ctx_idr
  */
-void drm_legacy_ctxbitmap_init(struct drm_device * dev)
+int drm_ctxbitmap_init(struct drm_device * dev)
 {
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return;
-
 	idr_init(&dev->ctx_idr);
+	return 0;
 }
 
 /**
@@ -137,7 +104,7 @@ void drm_legacy_ctxbitmap_init(struct drm_device * dev)
  * Free all idr members using drm_ctx_sarea_free helper function
  * while holding the drm_device::struct_mutex lock.
  */
-void drm_legacy_ctxbitmap_cleanup(struct drm_device * dev)
+void drm_ctxbitmap_cleanup(struct drm_device * dev)
 {
 	mutex_lock(&dev->struct_mutex);
 	idr_destroy(&dev->ctx_idr);
@@ -169,9 +136,6 @@ int drm_getsareactx(struct drm_device *dev, void *data,
 	struct drm_local_map *map;
 	struct drm_map_list *_entry;
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return -EINVAL;
-
 	mutex_lock(&dev->struct_mutex);
 
 	map = idr_find(&dev->ctx_idr, request->ctx_id);
@@ -216,9 +180,6 @@ int drm_setsareactx(struct drm_device *dev, void *data,
 	struct drm_local_map *map = NULL;
 	struct drm_map_list *r_list = NULL;
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return -EINVAL;
-
 	mutex_lock(&dev->struct_mutex);
 	list_for_each_entry(r_list, &dev->maplist, head) {
 		if (r_list->map
@@ -319,9 +280,6 @@ int drm_resctx(struct drm_device *dev, void *data,
 	struct drm_ctx ctx;
 	int i;
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return -EINVAL;
-
 	if (res->count >= DRM_RESERVED_CONTEXTS) {
 		memset(&ctx, 0, sizeof(ctx));
 		for (i = 0; i < DRM_RESERVED_CONTEXTS; i++) {
@@ -352,9 +310,6 @@ int drm_addctx(struct drm_device *dev, void *data,
 	struct drm_ctx_list *ctx_entry;
 	struct drm_ctx *ctx = data;
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return -EINVAL;
-
 	ctx->handle = drm_ctxbitmap_next(dev);
 	if (ctx->handle == DRM_KERNEL_CONTEXT) {
 		/* Skip kernel's context and get a new one. */
@@ -398,9 +353,6 @@ int drm_getctx(struct drm_device *dev, void *data, struct drm_file *file_priv)
 {
 	struct drm_ctx *ctx = data;
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return -EINVAL;
-
 	/* This is 0, because we don't handle any context flags */
 	ctx->flags = 0;
 
@@ -423,9 +375,6 @@ int drm_switchctx(struct drm_device *dev, void *data,
 {
 	struct drm_ctx *ctx = data;
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return -EINVAL;
-
 	DRM_DEBUG("%d\n", ctx->handle);
 	return drm_context_switch(dev, dev->last_context, ctx->handle);
 }
@@ -446,9 +395,6 @@ int drm_newctx(struct drm_device *dev, void *data,
 {
 	struct drm_ctx *ctx = data;
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return -EINVAL;
-
 	DRM_DEBUG("%d\n", ctx->handle);
 	drm_context_switch_complete(dev, file_priv, ctx->handle);
 
@@ -471,9 +417,6 @@ int drm_rmctx(struct drm_device *dev, void *data,
 {
 	struct drm_ctx *ctx = data;
 
-	if (drm_core_check_feature(dev, DRIVER_MODESET))
-		return -EINVAL;
-
 	DRM_DEBUG("%d\n", ctx->handle);
 	if (ctx->handle != DRM_KERNEL_CONTEXT) {
 		if (dev->driver->context_dtor)
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index 4be8e09a32ef..3f84277d7036 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -439,7 +439,26 @@ int drm_release(struct inode *inode, struct file *filp)
 	if (dev->driver->driver_features & DRIVER_GEM)
 		drm_gem_release(dev, file_priv);
 
-	drm_legacy_ctxbitmap_release(dev, file_priv);
+	mutex_lock(&dev->ctxlist_mutex);
+	if (!list_empty(&dev->ctxlist)) {
+		struct drm_ctx_list *pos, *n;
+
+		list_for_each_entry_safe(pos, n, &dev->ctxlist, head) {
+			if (pos->tag == file_priv &&
+			    pos->handle != DRM_KERNEL_CONTEXT) {
+				if (dev->driver->context_dtor)
+					dev->driver->context_dtor(dev,
+								  pos->handle);
+
+				drm_ctxbitmap_free(dev, pos->handle);
+
+				list_del(&pos->head);
+				kfree(pos);
+				--dev->ctx_count;
+			}
+		}
+	}
+	mutex_unlock(&dev->ctxlist_mutex);
 
 	mutex_lock(&dev->struct_mutex);
 
diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c
index e7eb0276f7f1..39d864576be4 100644
--- a/drivers/gpu/drm/drm_stub.c
+++ b/drivers/gpu/drm/drm_stub.c
@@ -292,7 +292,13 @@ int drm_fill_in_dev(struct drm_device *dev,
 			goto error_out_unreg;
 	}
 
-	drm_legacy_ctxbitmap_init(dev);
+
+
+	retcode = drm_ctxbitmap_init(dev);
+	if (retcode) {
+		DRM_ERROR("Cannot allocate memory for context bitmap.\n");
+		goto error_out_unreg;
+	}
 
 	if (driver->driver_features & DRIVER_GEM) {
 		retcode = drm_gem_init(dev);
@@ -446,7 +452,7 @@ void drm_put_dev(struct drm_device *dev)
 		drm_rmmap(dev, r_list->map);
 	drm_ht_remove(&dev->map_hash);
 
-	drm_legacy_ctxbitmap_cleanup(dev);
+	drm_ctxbitmap_cleanup(dev);
 
 	if (drm_core_check_feature(dev, DRIVER_MODESET))
 		drm_put_minor(&dev->control);
diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index 4752f223e5b2..45b6ef595965 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -56,7 +56,7 @@ config DRM_EXYNOS_IPP
 
 config DRM_EXYNOS_FIMC
 	bool "Exynos DRM FIMC"
-	depends on DRM_EXYNOS_IPP && MFD_SYSCON && OF
+	depends on DRM_EXYNOS_IPP && MFD_SYSCON
 	help
 	  Choose this option if you want to use Exynos FIMC for DRM.
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_buf.c b/drivers/gpu/drm/exynos/exynos_drm_buf.c
index 3445a0f3a6b2..9c8088462c26 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_buf.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_buf.c
@@ -63,7 +63,8 @@ static int lowlevel_buffer_allocate(struct drm_device *dev,
 			return -ENOMEM;
 		}
 
-		buf->kvaddr = dma_alloc_attrs(dev->dev, buf->size,
+		buf->kvaddr = (void __iomem *)dma_alloc_attrs(dev->dev,
+					buf->size,
 					&buf->dma_addr, GFP_KERNEL,
 					&buf->dma_attrs);
 		if (!buf->kvaddr) {
@@ -90,9 +91,9 @@ static int lowlevel_buffer_allocate(struct drm_device *dev,
 	}
 
 	buf->sgt = drm_prime_pages_to_sg(buf->pages, nr_pages);
-	if (!buf->sgt) {
+	if (IS_ERR(buf->sgt)) {
 		DRM_ERROR("failed to get sg table.\n");
-		ret = -ENOMEM;
+		ret = PTR_ERR(buf->sgt);
 		goto err_free_attrs;
 	}
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index 78e868bcf1ec..e7c2f2d07f19 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -99,12 +99,13 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
 		if (is_drm_iommu_supported(dev)) {
 			unsigned int nr_pages = buffer->size >> PAGE_SHIFT;
 
-			buffer->kvaddr = vmap(buffer->pages, nr_pages, VM_MAP,
+			buffer->kvaddr = (void __iomem *) vmap(buffer->pages,
+					nr_pages, VM_MAP,
 					pgprot_writecombine(PAGE_KERNEL));
 		} else {
 			phys_addr_t dma_addr = buffer->dma_addr;
 			if (dma_addr)
-				buffer->kvaddr = phys_to_virt(dma_addr);
+				buffer->kvaddr = (void __iomem *)phys_to_virt(dma_addr);
 			else
 				buffer->kvaddr = (void __iomem *)NULL;
 		}
diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index b1f8fc69023f..60e84043aa34 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -707,8 +707,7 @@ tda998x_encoder_dpms(struct drm_encoder *encoder, int mode)
 		reg_write(encoder, REG_VIP_CNTRL_2, priv->vip_cntrl_2);
 		break;
 	case DRM_MODE_DPMS_OFF:
-		/* disable audio and video ports */
-		reg_write(encoder, REG_ENA_AP, 0x00);
+		/* disable video ports */
 		reg_write(encoder, REG_ENA_VP_0, 0x00);
 		reg_write(encoder, REG_ENA_VP_1, 0x00);
 		reg_write(encoder, REG_ENA_VP_2, 0x00);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 8507c6d1e642..cdfb9da0e4ce 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1392,14 +1392,11 @@ out:
 		if (i915_terminally_wedged(&dev_priv->gpu_error))
 			return VM_FAULT_SIGBUS;
 	case -EAGAIN:
-		/* Give the error handler a chance to run and move the
-		 * objects off the GPU active list. Next time we service the
-		 * fault, we should be able to transition the page into the
-		 * GTT without touching the GPU (and so avoid further
-		 * EIO/EGAIN). If the GPU is wedged, then there is no issue
-		 * with coherency, just lost writes.
+		/*
+		 * EAGAIN means the gpu is hung and we'll wait for the error
+		 * handler to reset everything when re-faulting in
+		 * i915_mutex_lock_interruptible.
 		 */
-		set_need_resched();
 	case 0:
 	case -ERESTARTSYS:
 	case -EINTR:
@@ -4803,10 +4800,10 @@ i915_gem_inactive_count(struct shrinker *shrinker, struct shrink_control *sc)
 
 	if (!mutex_trylock(&dev->struct_mutex)) {
 		if (!mutex_is_locked_by(&dev->struct_mutex, current))
-			return SHRINK_STOP;
+			return 0;
 
 		if (dev_priv->mm.shrinker_no_lock_stealing)
-			return SHRINK_STOP;
+			return 0;
 
 		unlock = false;
 	}
@@ -4904,10 +4901,10 @@ i915_gem_inactive_scan(struct shrinker *shrinker, struct shrink_control *sc)
 
 	if (!mutex_trylock(&dev->struct_mutex)) {
 		if (!mutex_is_locked_by(&dev->struct_mutex, current))
-			return 0;
+			return SHRINK_STOP;
 
 		if (dev_priv->mm.shrinker_no_lock_stealing)
-			return 0;
+			return SHRINK_STOP;
 
 		unlock = false;
 	}
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index aba9d7498996..dae364f0028c 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -143,8 +143,10 @@ static void i915_error_vprintf(struct drm_i915_error_state_buf *e,
 
 	/* Seek the first printf which is hits start position */
 	if (e->pos < e->start) {
-		len = vsnprintf(NULL, 0, f, args);
-		if (!__i915_error_seek(e, len))
+		va_list tmp;
+
+		va_copy(tmp, args);
+		if (!__i915_error_seek(e, vsnprintf(NULL, 0, f, tmp)))
 			return;
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 83cce0cdb769..4b91228fd9bd 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1469,6 +1469,34 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 	return ret;
 }
 
+static void i915_error_wake_up(struct drm_i915_private *dev_priv,
+			       bool reset_completed)
+{
+	struct intel_ring_buffer *ring;
+	int i;
+
+	/*
+	 * Notify all waiters for GPU completion events that reset state has
+	 * been changed, and that they need to restart their wait after
+	 * checking for potential errors (and bail out to drop locks if there is
+	 * a gpu reset pending so that i915_error_work_func can acquire them).
+	 */
+
+	/* Wake up __wait_seqno, potentially holding dev->struct_mutex. */
+	for_each_ring(ring, dev_priv, i)
+		wake_up_all(&ring->irq_queue);
+
+	/* Wake up intel_crtc_wait_for_pending_flips, holding crtc->mutex. */
+	wake_up_all(&dev_priv->pending_flip_queue);
+
+	/*
+	 * Signal tasks blocked in i915_gem_wait_for_error that the pending
+	 * reset state is cleared.
+	 */
+	if (reset_completed)
+		wake_up_all(&dev_priv->gpu_error.reset_queue);
+}
+
 /**
  * i915_error_work_func - do process context error handling work
  * @work: work struct
@@ -1483,11 +1511,10 @@ static void i915_error_work_func(struct work_struct *work)
 	drm_i915_private_t *dev_priv = container_of(error, drm_i915_private_t,
 						    gpu_error);
 	struct drm_device *dev = dev_priv->dev;
-	struct intel_ring_buffer *ring;
 	char *error_event[] = { I915_ERROR_UEVENT "=1", NULL };
 	char *reset_event[] = { I915_RESET_UEVENT "=1", NULL };
 	char *reset_done_event[] = { I915_ERROR_UEVENT "=0", NULL };
-	int i, ret;
+	int ret;
 
 	kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE, error_event);
 
@@ -1506,8 +1533,16 @@ static void i915_error_work_func(struct work_struct *work)
 		kobject_uevent_env(&dev->primary->kdev.kobj, KOBJ_CHANGE,
 				   reset_event);
 
+		/*
+		 * All state reset _must_ be completed before we update the
+		 * reset counter, for otherwise waiters might miss the reset
+		 * pending state and not properly drop locks, resulting in
+		 * deadlocks with the reset work.
+		 */
 		ret = i915_reset(dev);
 
+		intel_display_handle_reset(dev);
+
 		if (ret == 0) {
 			/*
 			 * After all the gem state is reset, increment the reset
@@ -1528,12 +1563,11 @@ static void i915_error_work_func(struct work_struct *work)
 			atomic_set(&error->reset_counter, I915_WEDGED);
 		}
 
-		for_each_ring(ring, dev_priv, i)
-			wake_up_all(&ring->irq_queue);
-
-		intel_display_handle_reset(dev);
-
-		wake_up_all(&dev_priv->gpu_error.reset_queue);
+		/*
+		 * Note: The wake_up also serves as a memory barrier so that
+		 * waiters see the update value of the reset counter atomic_t.
+		 */
+		i915_error_wake_up(dev_priv, true);
 	}
 }
 
@@ -1642,8 +1676,6 @@ static void i915_report_and_clear_eir(struct drm_device *dev)
 void i915_handle_error(struct drm_device *dev, bool wedged)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_ring_buffer *ring;
-	int i;
 
 	i915_capture_error_state(dev);
 	i915_report_and_clear_eir(dev);
@@ -1653,11 +1685,19 @@ void i915_handle_error(struct drm_device *dev, bool wedged)
 				&dev_priv->gpu_error.reset_counter);
 
 		/*
-		 * Wakeup waiting processes so that the reset work item
-		 * doesn't deadlock trying to grab various locks.
+		 * Wakeup waiting processes so that the reset work function
+		 * i915_error_work_func doesn't deadlock trying to grab various
+		 * locks. By bumping the reset counter first, the woken
+		 * processes will see a reset in progress and back off,
+		 * releasing their locks and then wait for the reset completion.
+		 * We must do this for _all_ gpu waiters that might hold locks
+		 * that the reset work needs to acquire.
+		 *
+		 * Note: The wake_up serves as the required memory barrier to
+		 * ensure that the waiters see the updated value of the reset
+		 * counter atomic_t.
 		 */
-		for_each_ring(ring, dev_priv, i)
-			wake_up_all(&ring->irq_queue);
+		i915_error_wake_up(dev_priv, false);
 	}
 
 	/*
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 63aca49d11a8..63de2701b974 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -778,7 +778,7 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc)
 			/* Can only use the always-on power well for eDP when
 			 * not using the panel fitter, and when not using motion
 			  * blur mitigation (which we don't support). */
-			if (intel_crtc->config.pch_pfit.size)
+			if (intel_crtc->config.pch_pfit.enabled)
 				temp |= TRANS_DDI_EDP_INPUT_A_ONOFF;
 			else
 				temp |= TRANS_DDI_EDP_INPUT_A_ON;
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 2489d0b4c7d2..e5822e79f912 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2249,7 +2249,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 		I915_WRITE(PIPESRC(intel_crtc->pipe),
 			   ((crtc->mode.hdisplay - 1) << 16) |
 			   (crtc->mode.vdisplay - 1));
-		if (!intel_crtc->config.pch_pfit.size &&
+		if (!intel_crtc->config.pch_pfit.enabled &&
 		    (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) ||
 		     intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))) {
 			I915_WRITE(PF_CTL(intel_crtc->pipe), 0);
@@ -3203,7 +3203,7 @@ static void ironlake_pfit_enable(struct intel_crtc *crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe = crtc->pipe;
 
-	if (crtc->config.pch_pfit.size) {
+	if (crtc->config.pch_pfit.enabled) {
 		/* Force use of hard-coded filter coefficients
 		 * as some pre-programmed values are broken,
 		 * e.g. x201.
@@ -3428,7 +3428,7 @@ static void ironlake_pfit_disable(struct intel_crtc *crtc)
 
 	/* To avoid upsetting the power well on haswell only disable the pfit if
 	 * it's in use. The hw state code will make sure we get this right. */
-	if (crtc->config.pch_pfit.size) {
+	if (crtc->config.pch_pfit.enabled) {
 		I915_WRITE(PF_CTL(pipe), 0);
 		I915_WRITE(PF_WIN_POS(pipe), 0);
 		I915_WRITE(PF_WIN_SZ(pipe), 0);
@@ -4775,6 +4775,10 @@ static void i9xx_set_pipeconf(struct intel_crtc *intel_crtc)
 
 	pipeconf = 0;
 
+	if (dev_priv->quirks & QUIRK_PIPEA_FORCE &&
+	    I915_READ(PIPECONF(intel_crtc->pipe)) & PIPECONF_ENABLE)
+		pipeconf |= PIPECONF_ENABLE;
+
 	if (intel_crtc->pipe == 0 && INTEL_INFO(dev)->gen < 4) {
 		/* Enable pixel doubling when the dot clock is > 90% of the (display)
 		 * core speed.
@@ -4877,9 +4881,6 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 		return -EINVAL;
 	}
 
-	/* Ensure that the cursor is valid for the new mode before changing... */
-	intel_crtc_update_cursor(crtc, true);
-
 	if (is_lvds && dev_priv->lvds_downclock_avail) {
 		/*
 		 * Ensure we match the reduced clock's P to the target clock.
@@ -5768,9 +5769,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 		intel_crtc->config.dpll.p2 = clock.p2;
 	}
 
-	/* Ensure that the cursor is valid for the new mode before changing... */
-	intel_crtc_update_cursor(crtc, true);
-
 	/* CPU eDP is the only output that doesn't need a PCH PLL of its own. */
 	if (intel_crtc->config.has_pch_encoder) {
 		fp = i9xx_dpll_compute_fp(&intel_crtc->config.dpll);
@@ -5859,6 +5857,7 @@ static void ironlake_get_pfit_config(struct intel_crtc *crtc,
 	tmp = I915_READ(PF_CTL(crtc->pipe));
 
 	if (tmp & PF_ENABLE) {
+		pipe_config->pch_pfit.enabled = true;
 		pipe_config->pch_pfit.pos = I915_READ(PF_WIN_POS(crtc->pipe));
 		pipe_config->pch_pfit.size = I915_READ(PF_WIN_SZ(crtc->pipe));
 
@@ -6236,7 +6235,7 @@ static void haswell_modeset_global_resources(struct drm_device *dev)
 		if (!crtc->base.enabled)
 			continue;
 
-		if (crtc->pipe != PIPE_A || crtc->config.pch_pfit.size ||
+		if (crtc->pipe != PIPE_A || crtc->config.pch_pfit.enabled ||
 		    crtc->config.cpu_transcoder != TRANSCODER_EDP)
 			enable = true;
 	}
@@ -6259,9 +6258,6 @@ static int haswell_crtc_mode_set(struct drm_crtc *crtc,
 	if (!intel_ddi_pll_mode_set(crtc))
 		return -EINVAL;
 
-	/* Ensure that the cursor is valid for the new mode before changing... */
-	intel_crtc_update_cursor(crtc, true);
-
 	if (intel_crtc->config.has_dp_encoder)
 		intel_dp_set_m_n(intel_crtc);
 
@@ -6494,15 +6490,15 @@ static void haswell_write_eld(struct drm_connector *connector,
 
 	/* Set ELD valid state */
 	tmp = I915_READ(aud_cntrl_st2);
-	DRM_DEBUG_DRIVER("HDMI audio: pin eld vld status=0x%8x\n", tmp);
+	DRM_DEBUG_DRIVER("HDMI audio: pin eld vld status=0x%08x\n", tmp);
 	tmp |= (AUDIO_ELD_VALID_A << (pipe * 4));
 	I915_WRITE(aud_cntrl_st2, tmp);
 	tmp = I915_READ(aud_cntrl_st2);
-	DRM_DEBUG_DRIVER("HDMI audio: eld vld status=0x%8x\n", tmp);
+	DRM_DEBUG_DRIVER("HDMI audio: eld vld status=0x%08x\n", tmp);
 
 	/* Enable HDMI mode */
 	tmp = I915_READ(aud_config);
-	DRM_DEBUG_DRIVER("HDMI audio: audio conf: 0x%8x\n", tmp);
+	DRM_DEBUG_DRIVER("HDMI audio: audio conf: 0x%08x\n", tmp);
 	/* clear N_programing_enable and N_value_index */
 	tmp &= ~(AUD_CONFIG_N_VALUE_INDEX | AUD_CONFIG_N_PROG_ENABLE);
 	I915_WRITE(aud_config, tmp);
@@ -6937,7 +6933,8 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc,
 	intel_crtc->cursor_width = width;
 	intel_crtc->cursor_height = height;
 
-	intel_crtc_update_cursor(crtc, intel_crtc->cursor_bo != NULL);
+	if (intel_crtc->active)
+		intel_crtc_update_cursor(crtc, intel_crtc->cursor_bo != NULL);
 
 	return 0;
 fail_unpin:
@@ -6956,7 +6953,8 @@ static int intel_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 	intel_crtc->cursor_x = x;
 	intel_crtc->cursor_y = y;
 
-	intel_crtc_update_cursor(crtc, intel_crtc->cursor_bo != NULL);
+	if (intel_crtc->active)
+		intel_crtc_update_cursor(crtc, intel_crtc->cursor_bo != NULL);
 
 	return 0;
 }
@@ -8205,9 +8203,10 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 		      pipe_config->gmch_pfit.control,
 		      pipe_config->gmch_pfit.pgm_ratios,
 		      pipe_config->gmch_pfit.lvds_border_bits);
-	DRM_DEBUG_KMS("pch pfit: pos: 0x%08x, size: 0x%08x\n",
+	DRM_DEBUG_KMS("pch pfit: pos: 0x%08x, size: 0x%08x, %s\n",
 		      pipe_config->pch_pfit.pos,
-		      pipe_config->pch_pfit.size);
+		      pipe_config->pch_pfit.size,
+		      pipe_config->pch_pfit.enabled ? "enabled" : "disabled");
 	DRM_DEBUG_KMS("ips: %i\n", pipe_config->ips_enabled);
 }
 
@@ -8603,8 +8602,11 @@ intel_pipe_config_compare(struct drm_device *dev,
 	if (INTEL_INFO(dev)->gen < 4)
 		PIPE_CONF_CHECK_I(gmch_pfit.pgm_ratios);
 	PIPE_CONF_CHECK_I(gmch_pfit.lvds_border_bits);
-	PIPE_CONF_CHECK_I(pch_pfit.pos);
-	PIPE_CONF_CHECK_I(pch_pfit.size);
+	PIPE_CONF_CHECK_I(pch_pfit.enabled);
+	if (current_config->pch_pfit.enabled) {
+		PIPE_CONF_CHECK_I(pch_pfit.pos);
+		PIPE_CONF_CHECK_I(pch_pfit.size);
+	}
 
 	PIPE_CONF_CHECK_I(ips_enabled);
 
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 2151d13772b8..79c14e298ba6 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -588,7 +588,18 @@ intel_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode,
 			DRM_DEBUG_KMS("aux_ch native nack\n");
 			return -EREMOTEIO;
 		case AUX_NATIVE_REPLY_DEFER:
-			udelay(100);
+			/*
+			 * For now, just give more slack to branch devices. We
+			 * could check the DPCD for I2C bit rate capabilities,
+			 * and if available, adjust the interval. We could also
+			 * be more careful with DP-to-Legacy adapters where a
+			 * long legacy cable may force very low I2C bit rates.
+			 */
+			if (intel_dp->dpcd[DP_DOWNSTREAMPORT_PRESENT] &
+			    DP_DWN_STRM_PORT_PRESENT)
+				usleep_range(500, 600);
+			else
+				usleep_range(300, 400);
 			continue;
 		default:
 			DRM_ERROR("aux_ch invalid native reply 0x%02x\n",
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index a47799e832c6..28cae80495e2 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -280,6 +280,7 @@ struct intel_crtc_config {
 	struct {
 		u32 pos;
 		u32 size;
+		bool enabled;
 	} pch_pfit;
 
 	/* FDI configuration, only valid if has_pch_encoder is set. */
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 406303b509c1..7fa7df546c1e 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -263,6 +263,8 @@ static bool intel_dvo_compute_config(struct intel_encoder *encoder,
 		C(vtotal);
 		C(clock);
 #undef C
+
+		drm_mode_set_crtcinfo(adjusted_mode, 0);
 	}
 
 	if (intel_dvo->dev.dev_ops->mode_fixup)
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 42114ecbae0e..293564a2896a 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -112,6 +112,7 @@ intel_pch_panel_fitting(struct intel_crtc *intel_crtc,
 done:
 	pipe_config->pch_pfit.pos = (x << 16) | y;
 	pipe_config->pch_pfit.size = (width << 16) | height;
+	pipe_config->pch_pfit.enabled = pipe_config->pch_pfit.size != 0;
 }
 
 static void
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 0c115cc4899f..dd176b7296c1 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -2096,16 +2096,16 @@ static uint32_t ilk_pipe_pixel_rate(struct drm_device *dev,
 				    struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	uint32_t pixel_rate, pfit_size;
+	uint32_t pixel_rate;
 
 	pixel_rate = intel_crtc->config.adjusted_mode.clock;
 
 	/* We only use IF-ID interlacing. If we ever use PF-ID we'll need to
 	 * adjust the pixel_rate here. */
 
-	pfit_size = intel_crtc->config.pch_pfit.size;
-	if (pfit_size) {
+	if (intel_crtc->config.pch_pfit.enabled) {
 		uint64_t pipe_w, pipe_h, pfit_w, pfit_h;
+		uint32_t pfit_size = intel_crtc->config.pch_pfit.size;
 
 		pipe_w = intel_crtc->config.requested_mode.hdisplay;
 		pipe_h = intel_crtc->config.requested_mode.vdisplay;
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 85037b9d4934..49482fd5b76c 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -788,6 +788,8 @@ static void intel_sdvo_get_dtd_from_mode(struct intel_sdvo_dtd *dtd,
 	uint16_t h_sync_offset, v_sync_offset;
 	int mode_clock;
 
+	memset(dtd, 0, sizeof(*dtd));
+
 	width = mode->hdisplay;
 	height = mode->vdisplay;
 
@@ -830,44 +832,51 @@ static void intel_sdvo_get_dtd_from_mode(struct intel_sdvo_dtd *dtd,
 	if (mode->flags & DRM_MODE_FLAG_PVSYNC)
 		dtd->part2.dtd_flags |= DTD_FLAG_VSYNC_POSITIVE;
 
-	dtd->part2.sdvo_flags = 0;
 	dtd->part2.v_sync_off_high = v_sync_offset & 0xc0;
-	dtd->part2.reserved = 0;
 }
 
-static void intel_sdvo_get_mode_from_dtd(struct drm_display_mode * mode,
+static void intel_sdvo_get_mode_from_dtd(struct drm_display_mode *pmode,
 					 const struct intel_sdvo_dtd *dtd)
 {
-	mode->hdisplay = dtd->part1.h_active;
-	mode->hdisplay += ((dtd->part1.h_high >> 4) & 0x0f) << 8;
-	mode->hsync_start = mode->hdisplay + dtd->part2.h_sync_off;
-	mode->hsync_start += (dtd->part2.sync_off_width_high & 0xc0) << 2;
-	mode->hsync_end = mode->hsync_start + dtd->part2.h_sync_width;
-	mode->hsync_end += (dtd->part2.sync_off_width_high & 0x30) << 4;
-	mode->htotal = mode->hdisplay + dtd->part1.h_blank;
-	mode->htotal += (dtd->part1.h_high & 0xf) << 8;
-
-	mode->vdisplay = dtd->part1.v_active;
-	mode->vdisplay += ((dtd->part1.v_high >> 4) & 0x0f) << 8;
-	mode->vsync_start = mode->vdisplay;
-	mode->vsync_start += (dtd->part2.v_sync_off_width >> 4) & 0xf;
-	mode->vsync_start += (dtd->part2.sync_off_width_high & 0x0c) << 2;
-	mode->vsync_start += dtd->part2.v_sync_off_high & 0xc0;
-	mode->vsync_end = mode->vsync_start +
+	struct drm_display_mode mode = {};
+
+	mode.hdisplay = dtd->part1.h_active;
+	mode.hdisplay += ((dtd->part1.h_high >> 4) & 0x0f) << 8;
+	mode.hsync_start = mode.hdisplay + dtd->part2.h_sync_off;
+	mode.hsync_start += (dtd->part2.sync_off_width_high & 0xc0) << 2;
+	mode.hsync_end = mode.hsync_start + dtd->part2.h_sync_width;
+	mode.hsync_end += (dtd->part2.sync_off_width_high & 0x30) << 4;
+	mode.htotal = mode.hdisplay + dtd->part1.h_blank;
+	mode.htotal += (dtd->part1.h_high & 0xf) << 8;
+
+	mode.vdisplay = dtd->part1.v_active;
+	mode.vdisplay += ((dtd->part1.v_high >> 4) & 0x0f) << 8;
+	mode.vsync_start = mode.vdisplay;
+	mode.vsync_start += (dtd->part2.v_sync_off_width >> 4) & 0xf;
+	mode.vsync_start += (dtd->part2.sync_off_width_high & 0x0c) << 2;
+	mode.vsync_start += dtd->part2.v_sync_off_high & 0xc0;
+	mode.vsync_end = mode.vsync_start +
 		(dtd->part2.v_sync_off_width & 0xf);
-	mode->vsync_end += (dtd->part2.sync_off_width_high & 0x3) << 4;
-	mode->vtotal = mode->vdisplay + dtd->part1.v_blank;
-	mode->vtotal += (dtd->part1.v_high & 0xf) << 8;
+	mode.vsync_end += (dtd->part2.sync_off_width_high & 0x3) << 4;
+	mode.vtotal = mode.vdisplay + dtd->part1.v_blank;
+	mode.vtotal += (dtd->part1.v_high & 0xf) << 8;
 
-	mode->clock = dtd->part1.clock * 10;
+	mode.clock = dtd->part1.clock * 10;
 
-	mode->flags &= ~(DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC);
 	if (dtd->part2.dtd_flags & DTD_FLAG_INTERLACE)
-		mode->flags |= DRM_MODE_FLAG_INTERLACE;
+		mode.flags |= DRM_MODE_FLAG_INTERLACE;
 	if (dtd->part2.dtd_flags & DTD_FLAG_HSYNC_POSITIVE)
-		mode->flags |= DRM_MODE_FLAG_PHSYNC;
+		mode.flags |= DRM_MODE_FLAG_PHSYNC;
+	else
+		mode.flags |= DRM_MODE_FLAG_NHSYNC;
 	if (dtd->part2.dtd_flags & DTD_FLAG_VSYNC_POSITIVE)
-		mode->flags |= DRM_MODE_FLAG_PVSYNC;
+		mode.flags |= DRM_MODE_FLAG_PVSYNC;
+	else
+		mode.flags |= DRM_MODE_FLAG_NVSYNC;
+
+	drm_mode_set_crtcinfo(&mode, 0);
+
+	drm_mode_copy(pmode, &mode);
 }
 
 static bool intel_sdvo_check_supp_encode(struct intel_sdvo *intel_sdvo)
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index f2c6d7909ae2..dd6f84bf6c22 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -916,6 +916,14 @@ intel_tv_compute_config(struct intel_encoder *encoder,
 	DRM_DEBUG_KMS("forcing bpc to 8 for TV\n");
 	pipe_config->pipe_bpp = 8*3;
 
+	/* TV has it's own notion of sync and other mode flags, so clear them. */
+	pipe_config->adjusted_mode.flags = 0;
+
+	/*
+	 * FIXME: We don't check whether the input mode is actually what we want
+	 * or whether userspace is doing something stupid.
+	 */
+
 	return true;
 }
 
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
index a60584763b61..a0b9d8a95b16 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
@@ -124,6 +124,8 @@ void adreno_recover(struct msm_gpu *gpu)
 
 	/* reset completed fence seqno, just discard anything pending: */
 	adreno_gpu->memptrs->fence = gpu->submitted_fence;
+	adreno_gpu->memptrs->rptr  = 0;
+	adreno_gpu->memptrs->wptr  = 0;
 
 	gpu->funcs->pm_resume(gpu);
 	ret = gpu->funcs->hw_init(gpu);
@@ -229,7 +231,7 @@ void adreno_idle(struct msm_gpu *gpu)
 			return;
 	} while(time_before(jiffies, t));
 
-	DRM_ERROR("timeout waiting for %s to drain ringbuffer!\n", gpu->name);
+	DRM_ERROR("%s: timeout waiting to drain ringbuffer!\n", gpu->name);
 
 	/* TODO maybe we need to reset GPU here to recover from hang? */
 }
@@ -256,11 +258,17 @@ void adreno_wait_ring(struct msm_gpu *gpu, uint32_t ndwords)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	uint32_t freedwords;
+	unsigned long t = jiffies + ADRENO_IDLE_TIMEOUT;
 	do {
 		uint32_t size = gpu->rb->size / 4;
 		uint32_t wptr = get_wptr(gpu->rb);
 		uint32_t rptr = adreno_gpu->memptrs->rptr;
 		freedwords = (rptr + (size - 1) - wptr) % size;
+
+		if (time_after(jiffies, t)) {
+			DRM_ERROR("%s: timeout waiting for ringbuffer space\n", gpu->name);
+			break;
+		}
 	} while(freedwords < ndwords);
 }
 
diff --git a/drivers/gpu/drm/msm/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/mdp4/mdp4_kms.c
index 5db5bbaedae2..bc7fd11ad8be 100644
--- a/drivers/gpu/drm/msm/mdp4/mdp4_kms.c
+++ b/drivers/gpu/drm/msm/mdp4/mdp4_kms.c
@@ -19,8 +19,6 @@
 #include "msm_drv.h"
 #include "mdp4_kms.h"
 
-#include <mach/iommu.h>
-
 static struct mdp4_platform_config *mdp4_get_config(struct platform_device *dev);
 
 static int mdp4_hw_init(struct msm_kms *kms)
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index 864c9773636b..b3a2f1629041 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -18,8 +18,6 @@
 #include "msm_drv.h"
 #include "msm_gpu.h"
 
-#include <mach/iommu.h>
-
 static void msm_fb_output_poll_changed(struct drm_device *dev)
 {
 	struct msm_drm_private *priv = dev->dev_private;
@@ -62,6 +60,8 @@ int msm_iommu_attach(struct drm_device *dev, struct iommu_domain *iommu,
 	int i, ret;
 
 	for (i = 0; i < cnt; i++) {
+		/* TODO maybe some day msm iommu won't require this hack: */
+		struct device *msm_iommu_get_ctx(const char *ctx_name);
 		struct device *ctx = msm_iommu_get_ctx(names[i]);
 		if (!ctx)
 			continue;
@@ -199,7 +199,7 @@ static int msm_load(struct drm_device *dev, unsigned long flags)
 		 * imx drm driver on iMX5
 		 */
 		dev_err(dev->dev, "failed to load kms\n");
-		ret = PTR_ERR(priv->kms);
+		ret = PTR_ERR(kms);
 		goto fail;
 	}
 
@@ -499,25 +499,41 @@ int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence,
 		struct timespec *timeout)
 {
 	struct msm_drm_private *priv = dev->dev_private;
-	unsigned long timeout_jiffies = timespec_to_jiffies(timeout);
-	unsigned long start_jiffies = jiffies;
-	unsigned long remaining_jiffies;
 	int ret;
 
-	if (time_after(start_jiffies, timeout_jiffies))
-		remaining_jiffies = 0;
-	else
-		remaining_jiffies = timeout_jiffies - start_jiffies;
-
-	ret = wait_event_interruptible_timeout(priv->fence_event,
-			priv->completed_fence >= fence,
-			remaining_jiffies);
-	if (ret == 0) {
-		DBG("timeout waiting for fence: %u (completed: %u)",
-				fence, priv->completed_fence);
-		ret = -ETIMEDOUT;
-	} else if (ret != -ERESTARTSYS) {
-		ret = 0;
+	if (!priv->gpu)
+		return 0;
+
+	if (fence > priv->gpu->submitted_fence) {
+		DRM_ERROR("waiting on invalid fence: %u (of %u)\n",
+				fence, priv->gpu->submitted_fence);
+		return -EINVAL;
+	}
+
+	if (!timeout) {
+		/* no-wait: */
+		ret = fence_completed(dev, fence) ? 0 : -EBUSY;
+	} else {
+		unsigned long timeout_jiffies = timespec_to_jiffies(timeout);
+		unsigned long start_jiffies = jiffies;
+		unsigned long remaining_jiffies;
+
+		if (time_after(start_jiffies, timeout_jiffies))
+			remaining_jiffies = 0;
+		else
+			remaining_jiffies = timeout_jiffies - start_jiffies;
+
+		ret = wait_event_interruptible_timeout(priv->fence_event,
+				fence_completed(dev, fence),
+				remaining_jiffies);
+
+		if (ret == 0) {
+			DBG("timeout waiting for fence: %u (completed: %u)",
+					fence, priv->completed_fence);
+			ret = -ETIMEDOUT;
+		} else if (ret != -ERESTARTSYS) {
+			ret = 0;
+		}
 	}
 
 	return ret;
@@ -681,7 +697,7 @@ static struct drm_driver msm_driver = {
 	.gem_vm_ops         = &vm_ops,
 	.dumb_create        = msm_gem_dumb_create,
 	.dumb_map_offset    = msm_gem_dumb_map_offset,
-	.dumb_destroy       = msm_gem_dumb_destroy,
+	.dumb_destroy       = drm_gem_dumb_destroy,
 #ifdef CONFIG_DEBUG_FS
 	.debugfs_init       = msm_debugfs_init,
 	.debugfs_cleanup    = msm_debugfs_cleanup,
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index 80d75094bf0a..df8f1d084bc1 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -153,7 +153,7 @@ void *msm_gem_vaddr(struct drm_gem_object *obj);
 int msm_gem_queue_inactive_work(struct drm_gem_object *obj,
 		struct work_struct *work);
 void msm_gem_move_to_active(struct drm_gem_object *obj,
-		struct msm_gpu *gpu, uint32_t fence);
+		struct msm_gpu *gpu, bool write, uint32_t fence);
 void msm_gem_move_to_inactive(struct drm_gem_object *obj);
 int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op,
 		struct timespec *timeout);
@@ -191,6 +191,12 @@ u32 msm_readl(const void __iomem *addr);
 #define DBG(fmt, ...) DRM_DEBUG(fmt"\n", ##__VA_ARGS__)
 #define VERB(fmt, ...) if (0) DRM_DEBUG(fmt"\n", ##__VA_ARGS__)
 
+static inline bool fence_completed(struct drm_device *dev, uint32_t fence)
+{
+	struct msm_drm_private *priv = dev->dev_private;
+	return priv->completed_fence >= fence;
+}
+
 static inline int align_pitch(int width, int bpp)
 {
 	int bytespp = (bpp + 7) / 8;
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index 6b5a6c8c7658..2bae46c66a30 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -40,9 +40,9 @@ static struct page **get_pages(struct drm_gem_object *obj)
 		}
 
 		msm_obj->sgt = drm_prime_pages_to_sg(p, npages);
-		if (!msm_obj->sgt) {
+		if (IS_ERR(msm_obj->sgt)) {
 			dev_err(dev->dev, "failed to allocate sgt\n");
-			return ERR_PTR(-ENOMEM);
+			return ERR_CAST(msm_obj->sgt);
 		}
 
 		msm_obj->pages = p;
@@ -159,7 +159,6 @@ out_unlock:
 out:
 	switch (ret) {
 	case -EAGAIN:
-		set_need_resched();
 	case 0:
 	case -ERESTARTSYS:
 	case -EINTR:
@@ -320,13 +319,6 @@ int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev,
 			MSM_BO_SCANOUT | MSM_BO_WC, &args->handle);
 }
 
-int msm_gem_dumb_destroy(struct drm_file *file, struct drm_device *dev,
-		uint32_t handle)
-{
-	/* No special work needed, drop the reference and see what falls out */
-	return drm_gem_handle_delete(file, handle);
-}
-
 int msm_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev,
 		uint32_t handle, uint64_t *offset)
 {
@@ -393,11 +385,14 @@ int msm_gem_queue_inactive_work(struct drm_gem_object *obj,
 }
 
 void msm_gem_move_to_active(struct drm_gem_object *obj,
-		struct msm_gpu *gpu, uint32_t fence)
+		struct msm_gpu *gpu, bool write, uint32_t fence)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	msm_obj->gpu = gpu;
-	msm_obj->fence = fence;
+	if (write)
+		msm_obj->write_fence = fence;
+	else
+		msm_obj->read_fence = fence;
 	list_del_init(&msm_obj->mm_list);
 	list_add_tail(&msm_obj->mm_list, &gpu->active_list);
 }
@@ -411,7 +406,8 @@ void msm_gem_move_to_inactive(struct drm_gem_object *obj)
 	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
 
 	msm_obj->gpu = NULL;
-	msm_obj->fence = 0;
+	msm_obj->read_fence = 0;
+	msm_obj->write_fence = 0;
 	list_del_init(&msm_obj->mm_list);
 	list_add_tail(&msm_obj->mm_list, &priv->inactive_list);
 
@@ -433,8 +429,18 @@ int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op,
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
 	int ret = 0;
 
-	if (is_active(msm_obj) && !(op & MSM_PREP_NOSYNC))
-		ret = msm_wait_fence_interruptable(dev, msm_obj->fence, timeout);
+	if (is_active(msm_obj)) {
+		uint32_t fence = 0;
+
+		if (op & MSM_PREP_READ)
+			fence = msm_obj->write_fence;
+		if (op & MSM_PREP_WRITE)
+			fence = max(fence, msm_obj->read_fence);
+		if (op & MSM_PREP_NOSYNC)
+			timeout = NULL;
+
+		ret = msm_wait_fence_interruptable(dev, fence, timeout);
+	}
 
 	/* TODO cache maintenance */
 
@@ -455,9 +461,10 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m)
 	uint64_t off = drm_vma_node_start(&obj->vma_node);
 
 	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
-	seq_printf(m, "%08x: %c(%d) %2d (%2d) %08llx %p %d\n",
+	seq_printf(m, "%08x: %c(r=%u,w=%u) %2d (%2d) %08llx %p %d\n",
 			msm_obj->flags, is_active(msm_obj) ? 'A' : 'I',
-			msm_obj->fence, obj->name, obj->refcount.refcount.counter,
+			msm_obj->read_fence, msm_obj->write_fence,
+			obj->name, obj->refcount.refcount.counter,
 			off, msm_obj->vaddr, obj->size);
 }
 
diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h
index d746f13d283c..0676f32e2c6a 100644
--- a/drivers/gpu/drm/msm/msm_gem.h
+++ b/drivers/gpu/drm/msm/msm_gem.h
@@ -36,7 +36,7 @@ struct msm_gem_object {
 	 */
 	struct list_head mm_list;
 	struct msm_gpu *gpu;     /* non-null if active */
-	uint32_t fence;
+	uint32_t read_fence, write_fence;
 
 	/* Transiently in the process of submit ioctl, objects associated
 	 * with the submit are on submit->bo_list.. this only lasts for
diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 3e1ef3a00f60..5281d4bc37f7 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -78,7 +78,7 @@ static int submit_lookup_objects(struct msm_gem_submit *submit,
 		}
 
 		if (submit_bo.flags & BO_INVALID_FLAGS) {
-			DBG("invalid flags: %x", submit_bo.flags);
+			DRM_ERROR("invalid flags: %x\n", submit_bo.flags);
 			ret = -EINVAL;
 			goto out_unlock;
 		}
@@ -92,7 +92,7 @@ static int submit_lookup_objects(struct msm_gem_submit *submit,
 		 */
 		obj = idr_find(&file->object_idr, submit_bo.handle);
 		if (!obj) {
-			DBG("invalid handle %u at index %u", submit_bo.handle, i);
+			DRM_ERROR("invalid handle %u at index %u\n", submit_bo.handle, i);
 			ret = -EINVAL;
 			goto out_unlock;
 		}
@@ -100,7 +100,7 @@ static int submit_lookup_objects(struct msm_gem_submit *submit,
 		msm_obj = to_msm_bo(obj);
 
 		if (!list_empty(&msm_obj->submit_entry)) {
-			DBG("handle %u at index %u already on submit list",
+			DRM_ERROR("handle %u at index %u already on submit list\n",
 					submit_bo.handle, i);
 			ret = -EINVAL;
 			goto out_unlock;
@@ -216,8 +216,9 @@ static int submit_bo(struct msm_gem_submit *submit, uint32_t idx,
 		struct msm_gem_object **obj, uint32_t *iova, bool *valid)
 {
 	if (idx >= submit->nr_bos) {
-		DBG("invalid buffer index: %u (out of %u)", idx, submit->nr_bos);
-		return EINVAL;
+		DRM_ERROR("invalid buffer index: %u (out of %u)\n",
+				idx, submit->nr_bos);
+		return -EINVAL;
 	}
 
 	if (obj)
@@ -239,7 +240,7 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob
 	int ret;
 
 	if (offset % 4) {
-		DBG("non-aligned cmdstream buffer: %u", offset);
+		DRM_ERROR("non-aligned cmdstream buffer: %u\n", offset);
 		return -EINVAL;
 	}
 
@@ -266,7 +267,7 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob
 			return -EFAULT;
 
 		if (submit_reloc.submit_offset % 4) {
-			DBG("non-aligned reloc offset: %u",
+			DRM_ERROR("non-aligned reloc offset: %u\n",
 					submit_reloc.submit_offset);
 			return -EINVAL;
 		}
@@ -276,7 +277,7 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob
 
 		if ((off >= (obj->base.size / 4)) ||
 				(off < last_offset)) {
-			DBG("invalid offset %u at reloc %u", off, i);
+			DRM_ERROR("invalid offset %u at reloc %u\n", off, i);
 			return -EINVAL;
 		}
 
@@ -374,14 +375,15 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 			goto out;
 
 		if (submit_cmd.size % 4) {
-			DBG("non-aligned cmdstream buffer size: %u",
+			DRM_ERROR("non-aligned cmdstream buffer size: %u\n",
 					submit_cmd.size);
 			ret = -EINVAL;
 			goto out;
 		}
 
-		if (submit_cmd.size >= msm_obj->base.size) {
-			DBG("invalid cmdstream size: %u", submit_cmd.size);
+		if ((submit_cmd.size + submit_cmd.submit_offset) >=
+				msm_obj->base.size) {
+			DRM_ERROR("invalid cmdstream size: %u\n", submit_cmd.size);
 			ret = -EINVAL;
 			goto out;
 		}
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index e1e1ec9321ff..3bab937965d1 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -29,13 +29,14 @@
 static void bs_init(struct msm_gpu *gpu, struct platform_device *pdev)
 {
 	struct drm_device *dev = gpu->dev;
-	struct kgsl_device_platform_data *pdata = pdev->dev.platform_data;
+	struct kgsl_device_platform_data *pdata;
 
 	if (!pdev) {
 		dev_err(dev->dev, "could not find dtv pdata\n");
 		return;
 	}
 
+	pdata = pdev->dev.platform_data;
 	if (pdata->bus_scale_table) {
 		gpu->bsc = msm_bus_scale_register_client(pdata->bus_scale_table);
 		DBG("bus scale client: %08x", gpu->bsc);
@@ -230,6 +231,8 @@ static void hangcheck_timer_reset(struct msm_gpu *gpu)
 static void hangcheck_handler(unsigned long data)
 {
 	struct msm_gpu *gpu = (struct msm_gpu *)data;
+	struct drm_device *dev = gpu->dev;
+	struct msm_drm_private *priv = dev->dev_private;
 	uint32_t fence = gpu->funcs->last_fence(gpu);
 
 	if (fence != gpu->hangcheck_fence) {
@@ -237,14 +240,22 @@ static void hangcheck_handler(unsigned long data)
 		gpu->hangcheck_fence = fence;
 	} else if (fence < gpu->submitted_fence) {
 		/* no progress and not done.. hung! */
-		struct msm_drm_private *priv = gpu->dev->dev_private;
 		gpu->hangcheck_fence = fence;
+		dev_err(dev->dev, "%s: hangcheck detected gpu lockup!\n",
+				gpu->name);
+		dev_err(dev->dev, "%s:     completed fence: %u\n",
+				gpu->name, fence);
+		dev_err(dev->dev, "%s:     submitted fence: %u\n",
+				gpu->name, gpu->submitted_fence);
 		queue_work(priv->wq, &gpu->recover_work);
 	}
 
 	/* if still more pending work, reset the hangcheck timer: */
 	if (gpu->submitted_fence > gpu->hangcheck_fence)
 		hangcheck_timer_reset(gpu);
+
+	/* workaround for missing irq: */
+	queue_work(priv->wq, &gpu->retire_work);
 }
 
 /*
@@ -265,7 +276,8 @@ static void retire_worker(struct work_struct *work)
 		obj = list_first_entry(&gpu->active_list,
 				struct msm_gem_object, mm_list);
 
-		if (obj->fence <= fence) {
+		if ((obj->read_fence <= fence) &&
+				(obj->write_fence <= fence)) {
 			/* move to inactive: */
 			msm_gem_move_to_inactive(&obj->base);
 			msm_gem_put_iova(&obj->base, gpu->id);
@@ -321,7 +333,11 @@ int msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 					submit->gpu->id, &iova);
 		}
 
-		msm_gem_move_to_active(&msm_obj->base, gpu, submit->fence);
+		if (submit->bos[i].flags & MSM_SUBMIT_BO_READ)
+			msm_gem_move_to_active(&msm_obj->base, gpu, false, submit->fence);
+
+		if (submit->bos[i].flags & MSM_SUBMIT_BO_WRITE)
+			msm_gem_move_to_active(&msm_obj->base, gpu, true, submit->fence);
 	}
 	hangcheck_timer_reset(gpu);
 	mutex_unlock(&dev->struct_mutex);
diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c
index 05ff315e8e9e..b162e98a2953 100644
--- a/drivers/gpu/drm/radeon/btc_dpm.c
+++ b/drivers/gpu/drm/radeon/btc_dpm.c
@@ -1168,6 +1168,23 @@ static const struct radeon_blacklist_clocks btc_blacklist_clocks[] =
         { 25000, 30000, RADEON_SCLK_UP }
 };
 
+void btc_get_max_clock_from_voltage_dependency_table(struct radeon_clock_voltage_dependency_table *table,
+						     u32 *max_clock)
+{
+	u32 i, clock = 0;
+
+	if ((table == NULL) || (table->count == 0)) {
+		*max_clock = clock;
+		return;
+	}
+
+	for (i = 0; i < table->count; i++) {
+		if (clock < table->entries[i].clk)
+			clock = table->entries[i].clk;
+	}
+	*max_clock = clock;
+}
+
 void btc_apply_voltage_dependency_rules(struct radeon_clock_voltage_dependency_table *table,
 					u32 clock, u16 max_voltage, u16 *voltage)
 {
@@ -2080,6 +2097,7 @@ static void btc_apply_state_adjust_rules(struct radeon_device *rdev,
 	bool disable_mclk_switching;
 	u32 mclk, sclk;
 	u16 vddc, vddci;
+	u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc;
 
 	if ((rdev->pm.dpm.new_active_crtc_count > 1) ||
 	    btc_dpm_vblank_too_short(rdev))
@@ -2121,6 +2139,39 @@ static void btc_apply_state_adjust_rules(struct radeon_device *rdev,
 			ps->low.vddci = max_limits->vddci;
 	}
 
+	/* limit clocks to max supported clocks based on voltage dependency tables */
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk,
+							&max_sclk_vddc);
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk,
+							&max_mclk_vddci);
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk,
+							&max_mclk_vddc);
+
+	if (max_sclk_vddc) {
+		if (ps->low.sclk > max_sclk_vddc)
+			ps->low.sclk = max_sclk_vddc;
+		if (ps->medium.sclk > max_sclk_vddc)
+			ps->medium.sclk = max_sclk_vddc;
+		if (ps->high.sclk > max_sclk_vddc)
+			ps->high.sclk = max_sclk_vddc;
+	}
+	if (max_mclk_vddci) {
+		if (ps->low.mclk > max_mclk_vddci)
+			ps->low.mclk = max_mclk_vddci;
+		if (ps->medium.mclk > max_mclk_vddci)
+			ps->medium.mclk = max_mclk_vddci;
+		if (ps->high.mclk > max_mclk_vddci)
+			ps->high.mclk = max_mclk_vddci;
+	}
+	if (max_mclk_vddc) {
+		if (ps->low.mclk > max_mclk_vddc)
+			ps->low.mclk = max_mclk_vddc;
+		if (ps->medium.mclk > max_mclk_vddc)
+			ps->medium.mclk = max_mclk_vddc;
+		if (ps->high.mclk > max_mclk_vddc)
+			ps->high.mclk = max_mclk_vddc;
+	}
+
 	/* XXX validate the min clocks required for display */
 
 	if (disable_mclk_switching) {
diff --git a/drivers/gpu/drm/radeon/btc_dpm.h b/drivers/gpu/drm/radeon/btc_dpm.h
index 1a15e0e41950..3b6f12b7760b 100644
--- a/drivers/gpu/drm/radeon/btc_dpm.h
+++ b/drivers/gpu/drm/radeon/btc_dpm.h
@@ -46,6 +46,8 @@ void btc_adjust_clock_combinations(struct radeon_device *rdev,
 				   struct rv7xx_pl *pl);
 void btc_apply_voltage_dependency_rules(struct radeon_clock_voltage_dependency_table *table,
 					u32 clock, u16 max_voltage, u16 *voltage);
+void btc_get_max_clock_from_voltage_dependency_table(struct radeon_clock_voltage_dependency_table *table,
+						     u32 *max_clock);
 void btc_apply_voltage_delta_rules(struct radeon_device *rdev,
 				   u16 max_vddc, u16 max_vddci,
 				   u16 *vddc, u16 *vddci);
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 899627443030..51e947a97edf 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -146,6 +146,8 @@ static const struct ci_pt_config_reg didt_config_ci[] =
 };
 
 extern u8 rv770_get_memory_module_index(struct radeon_device *rdev);
+extern void btc_get_max_clock_from_voltage_dependency_table(struct radeon_clock_voltage_dependency_table *table,
+							    u32 *max_clock);
 extern int ni_copy_and_switch_arb_sets(struct radeon_device *rdev,
 				       u32 arb_freq_src, u32 arb_freq_dest);
 extern u8 si_get_ddr3_mclk_frequency_ratio(u32 memory_clock);
@@ -712,6 +714,7 @@ static void ci_apply_state_adjust_rules(struct radeon_device *rdev,
 	struct radeon_clock_and_voltage_limits *max_limits;
 	bool disable_mclk_switching;
 	u32 sclk, mclk;
+	u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc;
 	int i;
 
 	if ((rdev->pm.dpm.new_active_crtc_count > 1) ||
@@ -739,6 +742,29 @@ static void ci_apply_state_adjust_rules(struct radeon_device *rdev,
 		}
 	}
 
+	/* limit clocks to max supported clocks based on voltage dependency tables */
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk,
+							&max_sclk_vddc);
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk,
+							&max_mclk_vddci);
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk,
+							&max_mclk_vddc);
+
+	for (i = 0; i < ps->performance_level_count; i++) {
+		if (max_sclk_vddc) {
+			if (ps->performance_levels[i].sclk > max_sclk_vddc)
+				ps->performance_levels[i].sclk = max_sclk_vddc;
+		}
+		if (max_mclk_vddci) {
+			if (ps->performance_levels[i].mclk > max_mclk_vddci)
+				ps->performance_levels[i].mclk = max_mclk_vddci;
+		}
+		if (max_mclk_vddc) {
+			if (ps->performance_levels[i].mclk > max_mclk_vddc)
+				ps->performance_levels[i].mclk = max_mclk_vddc;
+		}
+	}
+
 	/* XXX validate the min clocks required for display */
 
 	if (disable_mclk_switching) {
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index adbdb6503b05..d02fd1c045d5 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -2845,10 +2845,8 @@ static void cik_gpu_init(struct radeon_device *rdev)
 		rdev->config.cik.tile_config |= (3 << 0);
 		break;
 	}
-	if ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT)
-		rdev->config.cik.tile_config |= 1 << 4;
-	else
-		rdev->config.cik.tile_config |= 0 << 4;
+	rdev->config.cik.tile_config |=
+		((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) << 4;
 	rdev->config.cik.tile_config |=
 		((gb_addr_config & PIPE_INTERLEAVE_SIZE_MASK) >> PIPE_INTERLEAVE_SIZE_SHIFT) << 8;
 	rdev->config.cik.tile_config |=
@@ -4456,8 +4454,8 @@ static int cik_mc_init(struct radeon_device *rdev)
 	rdev->mc.aper_base = pci_resource_start(rdev->pdev, 0);
 	rdev->mc.aper_size = pci_resource_len(rdev->pdev, 0);
 	/* size in MB on si */
-	rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024;
-	rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE) * 1024 * 1024;
+	rdev->mc.mc_vram_size = RREG32(CONFIG_MEMSIZE) * 1024ULL * 1024ULL;
+	rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE) * 1024ULL * 1024ULL;
 	rdev->mc.visible_vram_size = rdev->mc.aper_size;
 	si_vram_gtt_location(rdev, &rdev->mc);
 	radeon_update_bandwidth_info(rdev);
@@ -4735,12 +4733,13 @@ static void cik_vm_decode_fault(struct radeon_device *rdev,
 	u32 mc_id = (status & MEMORY_CLIENT_ID_MASK) >> MEMORY_CLIENT_ID_SHIFT;
 	u32 vmid = (status & FAULT_VMID_MASK) >> FAULT_VMID_SHIFT;
 	u32 protections = (status & PROTECTIONS_MASK) >> PROTECTIONS_SHIFT;
-	char *block = (char *)&mc_client;
+	char block[5] = { mc_client >> 24, (mc_client >> 16) & 0xff,
+		(mc_client >> 8) & 0xff, mc_client & 0xff, 0 };
 
-	printk("VM fault (0x%02x, vmid %d) at page %u, %s from %s (%d)\n",
+	printk("VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n",
 	       protections, vmid, addr,
 	       (status & MEMORY_CLIENT_RW_MASK) ? "write" : "read",
-	       block, mc_id);
+	       block, mc_client, mc_id);
 }
 
 /**
diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c
index 6c398a456d78..f26339028154 100644
--- a/drivers/gpu/drm/radeon/ni_dpm.c
+++ b/drivers/gpu/drm/radeon/ni_dpm.c
@@ -787,6 +787,7 @@ static void ni_apply_state_adjust_rules(struct radeon_device *rdev,
 	bool disable_mclk_switching;
 	u32 mclk, sclk;
 	u16 vddc, vddci;
+	u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc;
 	int i;
 
 	if ((rdev->pm.dpm.new_active_crtc_count > 1) ||
@@ -813,6 +814,29 @@ static void ni_apply_state_adjust_rules(struct radeon_device *rdev,
 		}
 	}
 
+	/* limit clocks to max supported clocks based on voltage dependency tables */
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk,
+							&max_sclk_vddc);
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk,
+							&max_mclk_vddci);
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk,
+							&max_mclk_vddc);
+
+	for (i = 0; i < ps->performance_level_count; i++) {
+		if (max_sclk_vddc) {
+			if (ps->performance_levels[i].sclk > max_sclk_vddc)
+				ps->performance_levels[i].sclk = max_sclk_vddc;
+		}
+		if (max_mclk_vddci) {
+			if (ps->performance_levels[i].mclk > max_mclk_vddci)
+				ps->performance_levels[i].mclk = max_mclk_vddci;
+		}
+		if (max_mclk_vddc) {
+			if (ps->performance_levels[i].mclk > max_mclk_vddc)
+				ps->performance_levels[i].mclk = max_mclk_vddc;
+		}
+	}
+
 	/* XXX validate the min clocks required for display */
 
 	if (disable_mclk_switching) {
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 24175717307b..d71333033b2b 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -2933,9 +2933,11 @@ static int r100_debugfs_cp_ring_info(struct seq_file *m, void *data)
 	seq_printf(m, "CP_RB_RPTR 0x%08x\n", rdp);
 	seq_printf(m, "%u free dwords in ring\n", ring->ring_free_dw);
 	seq_printf(m, "%u dwords in ring\n", count);
-	for (j = 0; j <= count; j++) {
-		i = (rdp + j) & ring->ptr_mask;
-		seq_printf(m, "r[%04d]=0x%08x\n", i, ring->ring[i]);
+	if (ring->ready) {
+		for (j = 0; j <= count; j++) {
+			i = (rdp + j) & ring->ptr_mask;
+			seq_printf(m, "r[%04d]=0x%08x\n", i, ring->ring[i]);
+		}
 	}
 	return 0;
 }
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index e65f211a7be0..5513d8f06252 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -1084,7 +1084,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
 				rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].dclk =
 					le16_to_cpu(uvd_clk->usDClkLow) | (uvd_clk->ucDClkHigh << 16);
 				rdev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].v =
-					le16_to_cpu(limits->entries[i].usVoltage);
+					le16_to_cpu(entry->usVoltage);
 				entry = (ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record *)
 					((u8 *)entry + sizeof(ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record));
 			}
diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c
index f443010ce90b..b0fa6002af3e 100644
--- a/drivers/gpu/drm/radeon/r600_hdmi.c
+++ b/drivers/gpu/drm/radeon/r600_hdmi.c
@@ -257,10 +257,7 @@ void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 	 * number (coefficient of two integer numbers.  DCCG_AUDIO_DTOx_PHASE
 	 * is the numerator, DCCG_AUDIO_DTOx_MODULE is the denominator
 	 */
-	if (ASIC_IS_DCE3(rdev)) {
-		/* according to the reg specs, this should DCE3.2 only, but in
-		 * practice it seems to cover DCE3.0 as well.
-		 */
+	if (ASIC_IS_DCE32(rdev)) {
 		if (dig->dig_encoder == 0) {
 			dto_cntl = RREG32(DCCG_AUDIO_DTO0_CNTL) & ~DCCG_AUDIO_DTO_WALLCLOCK_RATIO_MASK;
 			dto_cntl |= DCCG_AUDIO_DTO_WALLCLOCK_RATIO(wallclock_ratio);
@@ -276,8 +273,21 @@ void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 			WREG32(DCCG_AUDIO_DTO1_MODULE, dto_modulo);
 			WREG32(DCCG_AUDIO_DTO_SELECT, 1); /* select DTO1 */
 		}
+	} else if (ASIC_IS_DCE3(rdev)) {
+		/* according to the reg specs, this should DCE3.2 only, but in
+		 * practice it seems to cover DCE3.0/3.1 as well.
+		 */
+		if (dig->dig_encoder == 0) {
+			WREG32(DCCG_AUDIO_DTO0_PHASE, base_rate * 100);
+			WREG32(DCCG_AUDIO_DTO0_MODULE, clock * 100);
+			WREG32(DCCG_AUDIO_DTO_SELECT, 0); /* select DTO0 */
+		} else {
+			WREG32(DCCG_AUDIO_DTO1_PHASE, base_rate * 100);
+			WREG32(DCCG_AUDIO_DTO1_MODULE, clock * 100);
+			WREG32(DCCG_AUDIO_DTO_SELECT, 1); /* select DTO1 */
+		}
 	} else {
-		/* according to the reg specs, this should be DCE2.0 and DCE3.0 */
+		/* according to the reg specs, this should be DCE2.0 and DCE3.0/3.1 */
 		WREG32(AUDIO_DTO, AUDIO_DTO_PHASE(base_rate / 10) |
 		       AUDIO_DTO_MODULE(clock / 10));
 	}
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index 5003385a7512..8f7e04538fd6 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -1004,6 +1004,8 @@ static struct radeon_asic rv6xx_asic = {
 		.wait_for_vblank = &avivo_wait_for_vblank,
 		.set_backlight_level = &atombios_set_backlight_level,
 		.get_backlight_level = &atombios_get_backlight_level,
+		.hdmi_enable = &r600_hdmi_enable,
+		.hdmi_setmode = &r600_hdmi_setmode,
 	},
 	.copy = {
 		.blit = &r600_copy_cpdma,
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 404e25d285ba..f79ee184ffd5 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -1367,6 +1367,7 @@ bool radeon_atombios_get_ppll_ss_info(struct radeon_device *rdev,
 	int index = GetIndexIntoMasterTable(DATA, PPLL_SS_Info);
 	uint16_t data_offset, size;
 	struct _ATOM_SPREAD_SPECTRUM_INFO *ss_info;
+	struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT *ss_assign;
 	uint8_t frev, crev;
 	int i, num_indices;
 
@@ -1378,18 +1379,21 @@ bool radeon_atombios_get_ppll_ss_info(struct radeon_device *rdev,
 
 		num_indices = (size - sizeof(ATOM_COMMON_TABLE_HEADER)) /
 			sizeof(ATOM_SPREAD_SPECTRUM_ASSIGNMENT);
-
+		ss_assign = (struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT*)
+			((u8 *)&ss_info->asSS_Info[0]);
 		for (i = 0; i < num_indices; i++) {
-			if (ss_info->asSS_Info[i].ucSS_Id == id) {
+			if (ss_assign->ucSS_Id == id) {
 				ss->percentage =
-					le16_to_cpu(ss_info->asSS_Info[i].usSpreadSpectrumPercentage);
-				ss->type = ss_info->asSS_Info[i].ucSpreadSpectrumType;
-				ss->step = ss_info->asSS_Info[i].ucSS_Step;
-				ss->delay = ss_info->asSS_Info[i].ucSS_Delay;
-				ss->range = ss_info->asSS_Info[i].ucSS_Range;
-				ss->refdiv = ss_info->asSS_Info[i].ucRecommendedRef_Div;
+					le16_to_cpu(ss_assign->usSpreadSpectrumPercentage);
+				ss->type = ss_assign->ucSpreadSpectrumType;
+				ss->step = ss_assign->ucSS_Step;
+				ss->delay = ss_assign->ucSS_Delay;
+				ss->range = ss_assign->ucSS_Range;
+				ss->refdiv = ss_assign->ucRecommendedRef_Div;
 				return true;
 			}
+			ss_assign = (struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT*)
+				((u8 *)ss_assign + sizeof(struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT));
 		}
 	}
 	return false;
@@ -1477,6 +1481,12 @@ union asic_ss_info {
 	struct _ATOM_ASIC_INTERNAL_SS_INFO_V3 info_3;
 };
 
+union asic_ss_assignment {
+	struct _ATOM_ASIC_SS_ASSIGNMENT v1;
+	struct _ATOM_ASIC_SS_ASSIGNMENT_V2 v2;
+	struct _ATOM_ASIC_SS_ASSIGNMENT_V3 v3;
+};
+
 bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev,
 				      struct radeon_atom_ss *ss,
 				      int id, u32 clock)
@@ -1485,6 +1495,7 @@ bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev,
 	int index = GetIndexIntoMasterTable(DATA, ASIC_InternalSS_Info);
 	uint16_t data_offset, size;
 	union asic_ss_info *ss_info;
+	union asic_ss_assignment *ss_assign;
 	uint8_t frev, crev;
 	int i, num_indices;
 
@@ -1509,45 +1520,52 @@ bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev,
 			num_indices = (size - sizeof(ATOM_COMMON_TABLE_HEADER)) /
 				sizeof(ATOM_ASIC_SS_ASSIGNMENT);
 
+			ss_assign = (union asic_ss_assignment *)((u8 *)&ss_info->info.asSpreadSpectrum[0]);
 			for (i = 0; i < num_indices; i++) {
-				if ((ss_info->info.asSpreadSpectrum[i].ucClockIndication == id) &&
-				    (clock <= le32_to_cpu(ss_info->info.asSpreadSpectrum[i].ulTargetClockRange))) {
+				if ((ss_assign->v1.ucClockIndication == id) &&
+				    (clock <= le32_to_cpu(ss_assign->v1.ulTargetClockRange))) {
 					ss->percentage =
-						le16_to_cpu(ss_info->info.asSpreadSpectrum[i].usSpreadSpectrumPercentage);
-					ss->type = ss_info->info.asSpreadSpectrum[i].ucSpreadSpectrumMode;
-					ss->rate = le16_to_cpu(ss_info->info.asSpreadSpectrum[i].usSpreadRateInKhz);
+						le16_to_cpu(ss_assign->v1.usSpreadSpectrumPercentage);
+					ss->type = ss_assign->v1.ucSpreadSpectrumMode;
+					ss->rate = le16_to_cpu(ss_assign->v1.usSpreadRateInKhz);
 					return true;
 				}
+				ss_assign = (union asic_ss_assignment *)
+					((u8 *)ss_assign + sizeof(ATOM_ASIC_SS_ASSIGNMENT));
 			}
 			break;
 		case 2:
 			num_indices = (size - sizeof(ATOM_COMMON_TABLE_HEADER)) /
 				sizeof(ATOM_ASIC_SS_ASSIGNMENT_V2);
+			ss_assign = (union asic_ss_assignment *)((u8 *)&ss_info->info_2.asSpreadSpectrum[0]);
 			for (i = 0; i < num_indices; i++) {
-				if ((ss_info->info_2.asSpreadSpectrum[i].ucClockIndication == id) &&
-				    (clock <= le32_to_cpu(ss_info->info_2.asSpreadSpectrum[i].ulTargetClockRange))) {
+				if ((ss_assign->v2.ucClockIndication == id) &&
+				    (clock <= le32_to_cpu(ss_assign->v2.ulTargetClockRange))) {
 					ss->percentage =
-						le16_to_cpu(ss_info->info_2.asSpreadSpectrum[i].usSpreadSpectrumPercentage);
-					ss->type = ss_info->info_2.asSpreadSpectrum[i].ucSpreadSpectrumMode;
-					ss->rate = le16_to_cpu(ss_info->info_2.asSpreadSpectrum[i].usSpreadRateIn10Hz);
+						le16_to_cpu(ss_assign->v2.usSpreadSpectrumPercentage);
+					ss->type = ss_assign->v2.ucSpreadSpectrumMode;
+					ss->rate = le16_to_cpu(ss_assign->v2.usSpreadRateIn10Hz);
 					if ((crev == 2) &&
 					    ((id == ASIC_INTERNAL_ENGINE_SS) ||
 					     (id == ASIC_INTERNAL_MEMORY_SS)))
 						ss->rate /= 100;
 					return true;
 				}
+				ss_assign = (union asic_ss_assignment *)
+					((u8 *)ss_assign + sizeof(ATOM_ASIC_SS_ASSIGNMENT_V2));
 			}
 			break;
 		case 3:
 			num_indices = (size - sizeof(ATOM_COMMON_TABLE_HEADER)) /
 				sizeof(ATOM_ASIC_SS_ASSIGNMENT_V3);
+			ss_assign = (union asic_ss_assignment *)((u8 *)&ss_info->info_3.asSpreadSpectrum[0]);
 			for (i = 0; i < num_indices; i++) {
-				if ((ss_info->info_3.asSpreadSpectrum[i].ucClockIndication == id) &&
-				    (clock <= le32_to_cpu(ss_info->info_3.asSpreadSpectrum[i].ulTargetClockRange))) {
+				if ((ss_assign->v3.ucClockIndication == id) &&
+				    (clock <= le32_to_cpu(ss_assign->v3.ulTargetClockRange))) {
 					ss->percentage =
-						le16_to_cpu(ss_info->info_3.asSpreadSpectrum[i].usSpreadSpectrumPercentage);
-					ss->type = ss_info->info_3.asSpreadSpectrum[i].ucSpreadSpectrumMode;
-					ss->rate = le16_to_cpu(ss_info->info_3.asSpreadSpectrum[i].usSpreadRateIn10Hz);
+						le16_to_cpu(ss_assign->v3.usSpreadSpectrumPercentage);
+					ss->type = ss_assign->v3.ucSpreadSpectrumMode;
+					ss->rate = le16_to_cpu(ss_assign->v3.usSpreadRateIn10Hz);
 					if ((id == ASIC_INTERNAL_ENGINE_SS) ||
 					    (id == ASIC_INTERNAL_MEMORY_SS))
 						ss->rate /= 100;
@@ -1555,6 +1573,8 @@ bool radeon_atombios_get_asic_ss_info(struct radeon_device *rdev,
 						radeon_atombios_get_igp_ss_overrides(rdev, ss, id);
 					return true;
 				}
+				ss_assign = (union asic_ss_assignment *)
+					((u8 *)ss_assign + sizeof(ATOM_ASIC_SS_ASSIGNMENT_V3));
 			}
 			break;
 		default:
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index ac6ece61a476..66c222836631 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -85,8 +85,9 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
 		   VRAM, also but everything into VRAM on AGP cards to avoid
 		   image corruptions */
 		if (p->ring == R600_RING_TYPE_UVD_INDEX &&
-		    (i == 0 || p->rdev->flags & RADEON_IS_AGP)) {
-			/* TODO: is this still needed for NI+ ? */
+		    p->rdev->family < CHIP_PALM &&
+		    (i == 0 || drm_pci_device_is_agp(p->rdev->ddev))) {
+
 			p->relocs[i].lobj.domain =
 				RADEON_GEM_DOMAIN_VRAM;
 
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index e29faa73b574..841d0e09be3e 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -1320,13 +1320,22 @@ int radeon_device_init(struct radeon_device *rdev,
 			return r;
 	}
 	if ((radeon_testing & 1)) {
-		radeon_test_moves(rdev);
+		if (rdev->accel_working)
+			radeon_test_moves(rdev);
+		else
+			DRM_INFO("radeon: acceleration disabled, skipping move tests\n");
 	}
 	if ((radeon_testing & 2)) {
-		radeon_test_syncing(rdev);
+		if (rdev->accel_working)
+			radeon_test_syncing(rdev);
+		else
+			DRM_INFO("radeon: acceleration disabled, skipping sync tests\n");
 	}
 	if (radeon_benchmarking) {
-		radeon_benchmark(rdev, radeon_benchmarking);
+		if (rdev->accel_working)
+			radeon_benchmark(rdev, radeon_benchmarking);
+		else
+			DRM_INFO("radeon: acceleration disabled, skipping benchmarks\n");
 	}
 	return 0;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 87e1d69e8fdb..ac07ad1d4f8c 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -1002,7 +1002,7 @@ static void radeon_pm_resume_old(struct radeon_device *rdev)
 {
 	/* set up the default clocks if the MC ucode is loaded */
 	if ((rdev->family >= CHIP_BARTS) &&
-	    (rdev->family <= CHIP_HAINAN) &&
+	    (rdev->family <= CHIP_CAYMAN) &&
 	    rdev->mc_fw) {
 		if (rdev->pm.default_vddc)
 			radeon_atom_set_voltage(rdev, rdev->pm.default_vddc,
@@ -1046,7 +1046,7 @@ static void radeon_pm_resume_dpm(struct radeon_device *rdev)
 	if (ret) {
 		DRM_ERROR("radeon: dpm resume failed\n");
 		if ((rdev->family >= CHIP_BARTS) &&
-		    (rdev->family <= CHIP_HAINAN) &&
+		    (rdev->family <= CHIP_CAYMAN) &&
 		    rdev->mc_fw) {
 			if (rdev->pm.default_vddc)
 				radeon_atom_set_voltage(rdev, rdev->pm.default_vddc,
@@ -1097,7 +1097,7 @@ static int radeon_pm_init_old(struct radeon_device *rdev)
 		radeon_pm_init_profile(rdev);
 		/* set up the default clocks if the MC ucode is loaded */
 		if ((rdev->family >= CHIP_BARTS) &&
-		    (rdev->family <= CHIP_HAINAN) &&
+		    (rdev->family <= CHIP_CAYMAN) &&
 		    rdev->mc_fw) {
 			if (rdev->pm.default_vddc)
 				radeon_atom_set_voltage(rdev, rdev->pm.default_vddc,
@@ -1183,7 +1183,7 @@ static int radeon_pm_init_dpm(struct radeon_device *rdev)
 	if (ret) {
 		rdev->pm.dpm_enabled = false;
 		if ((rdev->family >= CHIP_BARTS) &&
-		    (rdev->family <= CHIP_HAINAN) &&
+		    (rdev->family <= CHIP_CAYMAN) &&
 		    rdev->mc_fw) {
 			if (rdev->pm.default_vddc)
 				radeon_atom_set_voltage(rdev, rdev->pm.default_vddc,
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index 46a25f037b84..18254e1c3e71 100644
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -839,9 +839,11 @@ static int radeon_debugfs_ring_info(struct seq_file *m, void *data)
 	 * packet that is the root issue
 	 */
 	i = (ring->rptr + ring->ptr_mask + 1 - 32) & ring->ptr_mask;
-	for (j = 0; j <= (count + 32); j++) {
-		seq_printf(m, "r[%5d]=0x%08x\n", i, ring->ring[i]);
-		i = (i + 1) & ring->ptr_mask;
+	if (ring->ready) {
+		for (j = 0; j <= (count + 32); j++) {
+			seq_printf(m, "r[%5d]=0x%08x\n", i, ring->ring[i]);
+			i = (i + 1) & ring->ptr_mask;
+		}
 	}
 	return 0;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c
index 1a01bbff9bfa..a0f11856ddde 100644
--- a/drivers/gpu/drm/radeon/radeon_uvd.c
+++ b/drivers/gpu/drm/radeon/radeon_uvd.c
@@ -476,8 +476,7 @@ static int radeon_uvd_cs_reloc(struct radeon_cs_parser *p,
 		return -EINVAL;
 	}
 
-	/* TODO: is this still necessary on NI+ ? */
-	if ((cmd == 0 || cmd == 0x3) &&
+	if (p->rdev->family < CHIP_PALM && (cmd == 0 || cmd == 0x3) &&
 	    (start >> 28) != (p->rdev->uvd.gpu_addr >> 28)) {
 		DRM_ERROR("msg/fb buffer %LX-%LX out of 256MB segment!\n",
 			  start, end);
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index cfe5d4d28915..9ace28702c76 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -2910,6 +2910,7 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
 	bool disable_sclk_switching = false;
 	u32 mclk, sclk;
 	u16 vddc, vddci;
+	u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc;
 	int i;
 
 	if ((rdev->pm.dpm.new_active_crtc_count > 1) ||
@@ -2943,6 +2944,29 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
 		}
 	}
 
+	/* limit clocks to max supported clocks based on voltage dependency tables */
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk,
+							&max_sclk_vddc);
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddci_dependency_on_mclk,
+							&max_mclk_vddci);
+	btc_get_max_clock_from_voltage_dependency_table(&rdev->pm.dpm.dyn_state.vddc_dependency_on_mclk,
+							&max_mclk_vddc);
+
+	for (i = 0; i < ps->performance_level_count; i++) {
+		if (max_sclk_vddc) {
+			if (ps->performance_levels[i].sclk > max_sclk_vddc)
+				ps->performance_levels[i].sclk = max_sclk_vddc;
+		}
+		if (max_mclk_vddci) {
+			if (ps->performance_levels[i].mclk > max_mclk_vddci)
+				ps->performance_levels[i].mclk = max_mclk_vddci;
+		}
+		if (max_mclk_vddc) {
+			if (ps->performance_levels[i].mclk > max_mclk_vddc)
+				ps->performance_levels[i].mclk = max_mclk_vddc;
+		}
+	}
+
 	/* XXX validate the min clocks required for display */
 
 	if (disable_mclk_switching) {
diff --git a/drivers/gpu/drm/radeon/uvd_v1_0.c b/drivers/gpu/drm/radeon/uvd_v1_0.c
index 7266805d9786..3100fa9cb52f 100644
--- a/drivers/gpu/drm/radeon/uvd_v1_0.c
+++ b/drivers/gpu/drm/radeon/uvd_v1_0.c
@@ -212,8 +212,8 @@ int uvd_v1_0_start(struct radeon_device *rdev)
 	/* enable VCPU clock */
 	WREG32(UVD_VCPU_CNTL,  1 << 9);
 
-	/* enable UMC */
-	WREG32_P(UVD_LMI_CTRL2, 0, ~(1 << 8));
+	/* enable UMC and NC0 */
+	WREG32_P(UVD_LMI_CTRL2, 1 << 13, ~((1 << 8) | (1 << 13)));
 
 	/* boot up the VCPU */
 	WREG32(UVD_SOFT_RESET, 0);