Merge 3.13-rc5 into staging-next

We want these fixes here to handle some merge issues.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

70 files changed, 1012 insertions, 321 deletions

diff --git a/drivers/gpu/drm/armada/armada_drm.h b/drivers/gpu/drm/armada/armada_drm.h
index eef09ec9a5ff..a72cae03b99b 100644
--- a/drivers/gpu/drm/armada/armada_drm.h
+++ b/drivers/gpu/drm/armada/armada_drm.h
@@ -103,6 +103,7 @@ void armada_drm_queue_unref_work(struct drm_device *,
 extern const struct drm_mode_config_funcs armada_drm_mode_config_funcs;
 
 int armada_fbdev_init(struct drm_device *);
+void armada_fbdev_lastclose(struct drm_device *);
 void armada_fbdev_fini(struct drm_device *);
 
 int armada_overlay_plane_create(struct drm_device *, unsigned long);
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index 4f2b28354915..62d0ff3efddf 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -321,6 +321,11 @@ static struct drm_ioctl_desc armada_ioctls[] = {
 		DRM_UNLOCKED),
 };
 
+static void armada_drm_lastclose(struct drm_device *dev)
+{
+	armada_fbdev_lastclose(dev);
+}
+
 static const struct file_operations armada_drm_fops = {
 	.owner			= THIS_MODULE,
 	.llseek			= no_llseek,
@@ -337,7 +342,7 @@ static struct drm_driver armada_drm_driver = {
 	.open			= NULL,
 	.preclose		= NULL,
 	.postclose		= NULL,
-	.lastclose		= NULL,
+	.lastclose		= armada_drm_lastclose,
 	.unload			= armada_drm_unload,
 	.get_vblank_counter	= drm_vblank_count,
 	.enable_vblank		= armada_drm_enable_vblank,
diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c
index dd5ea77dac96..948cb14c561e 100644
--- a/drivers/gpu/drm/armada/armada_fbdev.c
+++ b/drivers/gpu/drm/armada/armada_fbdev.c
@@ -105,9 +105,9 @@ static int armada_fb_create(struct drm_fb_helper *fbh,
 	drm_fb_helper_fill_fix(info, dfb->fb.pitches[0], dfb->fb.depth);
 	drm_fb_helper_fill_var(info, fbh, sizes->fb_width, sizes->fb_height);
 
-	DRM_DEBUG_KMS("allocated %dx%d %dbpp fb: 0x%08x\n",
-		dfb->fb.width, dfb->fb.height,
-		dfb->fb.bits_per_pixel, obj->phys_addr);
+	DRM_DEBUG_KMS("allocated %dx%d %dbpp fb: 0x%08llx\n",
+		dfb->fb.width, dfb->fb.height, dfb->fb.bits_per_pixel,
+		(unsigned long long)obj->phys_addr);
 
 	return 0;
 
@@ -177,6 +177,16 @@ int armada_fbdev_init(struct drm_device *dev)
 	return ret;
 }
 
+void armada_fbdev_lastclose(struct drm_device *dev)
+{
+	struct armada_private *priv = dev->dev_private;
+
+	drm_modeset_lock_all(dev);
+	if (priv->fbdev)
+		drm_fb_helper_restore_fbdev_mode(priv->fbdev);
+	drm_modeset_unlock_all(dev);
+}
+
 void armada_fbdev_fini(struct drm_device *dev)
 {
 	struct armada_private *priv = dev->dev_private;
@@ -192,11 +202,11 @@ void armada_fbdev_fini(struct drm_device *dev)
 			framebuffer_release(info);
 		}
 
+		drm_fb_helper_fini(fbh);
+
 		if (fbh->fb)
 			fbh->fb->funcs->destroy(fbh->fb);
 
-		drm_fb_helper_fini(fbh);
-
 		priv->fbdev = NULL;
 	}
 }
diff --git a/drivers/gpu/drm/armada/armada_gem.c b/drivers/gpu/drm/armada/armada_gem.c
index 9f2356bae7fd..887816f43476 100644
--- a/drivers/gpu/drm/armada/armada_gem.c
+++ b/drivers/gpu/drm/armada/armada_gem.c
@@ -172,8 +172,9 @@ armada_gem_linear_back(struct drm_device *dev, struct armada_gem_object *obj)
 		obj->dev_addr = obj->linear->start;
 	}
 
-	DRM_DEBUG_DRIVER("obj %p phys %#x dev %#x\n",
-			 obj, obj->phys_addr, obj->dev_addr);
+	DRM_DEBUG_DRIVER("obj %p phys %#llx dev %#llx\n", obj,
+			 (unsigned long long)obj->phys_addr,
+			 (unsigned long long)obj->dev_addr);
 
 	return 0;
 }
@@ -557,7 +558,6 @@ armada_gem_prime_import(struct drm_device *dev, struct dma_buf *buf)
 			 * refcount on the gem object itself.
 			 */
 			drm_gem_object_reference(obj);
-			dma_buf_put(buf);
 			return obj;
 		}
 	}
@@ -573,6 +573,7 @@ armada_gem_prime_import(struct drm_device *dev, struct dma_buf *buf)
 	}
 
 	dobj->obj.import_attach = attach;
+	get_dma_buf(buf);
 
 	/*
 	 * Don't call dma_buf_map_attachment() here - it maps the
diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index fb7cf0e796f6..8835dcddfac3 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -68,6 +68,8 @@
 #define EDID_QUIRK_DETAILED_SYNC_PP		(1 << 6)
 /* Force reduced-blanking timings for detailed modes */
 #define EDID_QUIRK_FORCE_REDUCED_BLANKING	(1 << 7)
+/* Force 8bpc */
+#define EDID_QUIRK_FORCE_8BPC			(1 << 8)
 
 struct detailed_mode_closure {
 	struct drm_connector *connector;
@@ -128,6 +130,9 @@ static struct edid_quirk {
 
 	/* Medion MD 30217 PG */
 	{ "MED", 0x7b8, EDID_QUIRK_PREFER_LARGE_75 },
+
+	/* Panel in Samsung NP700G7A-S01PL notebook reports 6bpc */
+	{ "SEC", 0xd033, EDID_QUIRK_FORCE_8BPC },
 };
 
 /*
@@ -2674,7 +2679,7 @@ static int add_3d_struct_modes(struct drm_connector *connector, u16 structure,
 	int modes = 0;
 	u8 cea_mode;
 
-	if (video_db == NULL || video_index > video_len)
+	if (video_db == NULL || video_index >= video_len)
 		return 0;
 
 	/* CEA modes are numbered 1..127 */
@@ -2701,7 +2706,7 @@ static int add_3d_struct_modes(struct drm_connector *connector, u16 structure,
 	if (structure & (1 << 8)) {
 		newmode = drm_mode_duplicate(dev, &edid_cea_modes[cea_mode]);
 		if (newmode) {
-			newmode->flags = DRM_MODE_FLAG_3D_SIDE_BY_SIDE_HALF;
+			newmode->flags |= DRM_MODE_FLAG_3D_SIDE_BY_SIDE_HALF;
 			drm_mode_probed_add(connector, newmode);
 			modes++;
 		}
@@ -3435,6 +3440,9 @@ int drm_add_edid_modes(struct drm_connector *connector, struct edid *edid)
 
 	drm_add_display_info(edid, &connector->display_info);
 
+	if (quirks & EDID_QUIRK_FORCE_8BPC)
+		connector->display_info.bpc = 8;
+
 	return num_modes;
 }
 EXPORT_SYMBOL(drm_add_edid_modes);
diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c
index f53d5246979c..66dd3a001cf1 100644
--- a/drivers/gpu/drm/drm_stub.c
+++ b/drivers/gpu/drm/drm_stub.c
@@ -566,11 +566,11 @@ err_unload:
 	if (dev->driver->unload)
 		dev->driver->unload(dev);
 err_primary_node:
-	drm_put_minor(dev->primary);
+	drm_unplug_minor(dev->primary);
 err_render_node:
-	drm_put_minor(dev->render);
+	drm_unplug_minor(dev->render);
 err_control_node:
-	drm_put_minor(dev->control);
+	drm_unplug_minor(dev->control);
 err_agp:
 	if (dev->driver->bus->agp_destroy)
 		dev->driver->bus->agp_destroy(dev);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index b676006a95a0..22b8f5eced80 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -173,28 +173,37 @@ static int exynos_drm_open(struct drm_device *dev, struct drm_file *file)
 static void exynos_drm_preclose(struct drm_device *dev,
 					struct drm_file *file)
 {
+	exynos_drm_subdrv_close(dev, file);
+}
+
+static void exynos_drm_postclose(struct drm_device *dev, struct drm_file *file)
+{
 	struct exynos_drm_private *private = dev->dev_private;
-	struct drm_pending_vblank_event *e, *t;
+	struct drm_pending_vblank_event *v, *vt;
+	struct drm_pending_event *e, *et;
 	unsigned long flags;
 
-	/* release events of current file */
+	if (!file->driver_priv)
+		return;
+
+	/* Release all events not unhandled by page flip handler. */
 	spin_lock_irqsave(&dev->event_lock, flags);
-	list_for_each_entry_safe(e, t, &private->pageflip_event_list,
+	list_for_each_entry_safe(v, vt, &private->pageflip_event_list,
 			base.link) {
-		if (e->base.file_priv == file) {
-			list_del(&e->base.link);
-			e->base.destroy(&e->base);
+		if (v->base.file_priv == file) {
+			list_del(&v->base.link);
+			drm_vblank_put(dev, v->pipe);
+			v->base.destroy(&v->base);
 		}
 	}
-	spin_unlock_irqrestore(&dev->event_lock, flags);
 
-	exynos_drm_subdrv_close(dev, file);
-}
+	/* Release all events handled by page flip handler but not freed. */
+	list_for_each_entry_safe(e, et, &file->event_list, link) {
+		list_del(&e->link);
+		e->destroy(e);
+	}
+	spin_unlock_irqrestore(&dev->event_lock, flags);
 
-static void exynos_drm_postclose(struct drm_device *dev, struct drm_file *file)
-{
-	if (!file->driver_priv)
-		return;
 
 	kfree(file->driver_priv);
 	file->driver_priv = NULL;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 23da72b5eae9..a61878bf5dcd 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -31,7 +31,7 @@
 #include "exynos_drm_iommu.h"
 
 /*
- * FIMD is stand for Fully Interactive Mobile Display and
+ * FIMD stands for Fully Interactive Mobile Display and
  * as a display controller, it transfers contents drawn on memory
  * to a LCD Panel through Display Interfaces such as RGB or
  * CPU Interface.
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 0cab2d045135..5c648425c1e0 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -83,6 +83,14 @@ void i915_update_dri1_breadcrumb(struct drm_device *dev)
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct drm_i915_master_private *master_priv;
 
+	/*
+	 * The dri breadcrumb update races against the drm master disappearing.
+	 * Instead of trying to fix this (this is by far not the only ums issue)
+	 * just don't do the update in kms mode.
+	 */
+	if (drm_core_check_feature(dev, DRIVER_MODESET))
+		return;
+
 	if (dev->primary->master) {
 		master_priv = dev->primary->master->driver_priv;
 		if (master_priv->sarea_priv)
@@ -1490,16 +1498,9 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	spin_lock_init(&dev_priv->uncore.lock);
 	spin_lock_init(&dev_priv->mm.object_stat_lock);
 	mutex_init(&dev_priv->dpio_lock);
-	mutex_init(&dev_priv->rps.hw_lock);
 	mutex_init(&dev_priv->modeset_restore_lock);
 
-	mutex_init(&dev_priv->pc8.lock);
-	dev_priv->pc8.requirements_met = false;
-	dev_priv->pc8.gpu_idle = false;
-	dev_priv->pc8.irqs_disabled = false;
-	dev_priv->pc8.enabled = false;
-	dev_priv->pc8.disable_count = 2; /* requirements_met + gpu_idle */
-	INIT_DELAYED_WORK(&dev_priv->pc8.enable_work, hsw_enable_pc8_work);
+	intel_pm_setup(dev);
 
 	intel_display_crc_init(dev);
 
@@ -1603,7 +1604,6 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	}
 
 	intel_irq_init(dev);
-	intel_pm_init(dev);
 	intel_uncore_sanitize(dev);
 
 	/* Try to make sure MCHBAR is enabled before poking at it */
@@ -1848,8 +1848,10 @@ void i915_driver_lastclose(struct drm_device * dev)
 
 void i915_driver_preclose(struct drm_device * dev, struct drm_file *file_priv)
 {
+	mutex_lock(&dev->struct_mutex);
 	i915_gem_context_close(dev, file_priv);
 	i915_gem_release(dev, file_priv);
+	mutex_unlock(&dev->struct_mutex);
 }
 
 void i915_driver_postclose(struct drm_device *dev, struct drm_file *file)
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 989be12cdd6e..5b7b7e06cb3a 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -534,8 +534,10 @@ static int i915_drm_freeze(struct drm_device *dev)
 		 * Disable CRTCs directly since we want to preserve sw state
 		 * for _thaw.
 		 */
+		mutex_lock(&dev->mode_config.mutex);
 		list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
 			dev_priv->display.crtc_disable(crtc);
+		mutex_unlock(&dev->mode_config.mutex);
 
 		intel_modeset_suspend_hw(dev);
 	}
@@ -649,6 +651,7 @@ static int __i915_drm_thaw(struct drm_device *dev, bool restore_gtt_mappings)
 		intel_modeset_init_hw(dev);
 
 		drm_modeset_lock_all(dev);
+		drm_mode_config_reset(dev);
 		intel_modeset_setup_hw_state(dev, true);
 		drm_modeset_unlock_all(dev);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ccdbecca070d..90fcccba17b0 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1755,8 +1755,13 @@ struct drm_i915_file_private {
 #define IS_MOBILE(dev)		(INTEL_INFO(dev)->is_mobile)
 #define IS_HSW_EARLY_SDV(dev)	(IS_HASWELL(dev) && \
 				 ((dev)->pdev->device & 0xFF00) == 0x0C00)
-#define IS_ULT(dev)		(IS_HASWELL(dev) && \
+#define IS_BDW_ULT(dev)		(IS_BROADWELL(dev) && \
+				 (((dev)->pdev->device & 0xf) == 0x2  || \
+				 ((dev)->pdev->device & 0xf) == 0x6 || \
+				 ((dev)->pdev->device & 0xf) == 0xe))
+#define IS_HSW_ULT(dev)		(IS_HASWELL(dev) && \
 				 ((dev)->pdev->device & 0xFF00) == 0x0A00)
+#define IS_ULT(dev)		(IS_HSW_ULT(dev) || IS_BDW_ULT(dev))
 #define IS_HSW_GT3(dev)		(IS_HASWELL(dev) && \
 				 ((dev)->pdev->device & 0x00F0) == 0x0020)
 #define IS_PRELIMINARY_HW(intel_info) ((intel_info)->is_preliminary)
@@ -1901,9 +1906,7 @@ void i915_queue_hangcheck(struct drm_device *dev);
 void i915_handle_error(struct drm_device *dev, bool wedged);
 
 extern void intel_irq_init(struct drm_device *dev);
-extern void intel_pm_init(struct drm_device *dev);
 extern void intel_hpd_init(struct drm_device *dev);
-extern void intel_pm_init(struct drm_device *dev);
 
 extern void intel_uncore_sanitize(struct drm_device *dev);
 extern void intel_uncore_early_sanitize(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 12bbd5eac70d..621c7c67a643 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -4442,10 +4442,9 @@ i915_gem_init_hw(struct drm_device *dev)
 	if (dev_priv->ellc_size)
 		I915_WRITE(HSW_IDICR, I915_READ(HSW_IDICR) | IDIHASHMSK(0xf));
 
-	if (IS_HSW_GT3(dev))
-		I915_WRITE(MI_PREDICATE_RESULT_2, LOWER_SLICE_ENABLED);
-	else
-		I915_WRITE(MI_PREDICATE_RESULT_2, LOWER_SLICE_DISABLED);
+	if (IS_HASWELL(dev))
+		I915_WRITE(MI_PREDICATE_RESULT_2, IS_HSW_GT3(dev) ?
+			   LOWER_SLICE_ENABLED : LOWER_SLICE_DISABLED);
 
 	if (HAS_PCH_NOP(dev)) {
 		u32 temp = I915_READ(GEN7_MSG_CTL);
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 72a3df32292f..b0f42b9ca037 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -347,10 +347,8 @@ void i915_gem_context_close(struct drm_device *dev, struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
 
-	mutex_lock(&dev->struct_mutex);
 	idr_for_each(&file_priv->context_idr, context_idr_cleanup, NULL);
 	idr_destroy(&file_priv->context_idr);
-	mutex_unlock(&dev->struct_mutex);
 }
 
 static struct i915_hw_context *
@@ -423,11 +421,21 @@ static int do_switch(struct i915_hw_context *to)
 	if (ret)
 		return ret;
 
-	/* Clear this page out of any CPU caches for coherent swap-in/out. Note
+	/*
+	 * Pin can switch back to the default context if we end up calling into
+	 * evict_everything - as a last ditch gtt defrag effort that also
+	 * switches to the default context. Hence we need to reload from here.
+	 */
+	from = ring->last_context;
+
+	/*
+	 * Clear this page out of any CPU caches for coherent swap-in/out. Note
 	 * that thanks to write = false in this call and us not setting any gpu
 	 * write domains when putting a context object onto the active list
 	 * (when switching away from it), this won't block.
-	 * XXX: We need a real interface to do this instead of trickery. */
+	 *
+	 * XXX: We need a real interface to do this instead of trickery.
+	 */
 	ret = i915_gem_object_set_to_gtt_domain(to->obj, false);
 	if (ret) {
 		i915_gem_object_unpin(to->obj);
diff --git a/drivers/gpu/drm/i915/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
index 7d5752fda5f1..9bb533e0d762 100644
--- a/drivers/gpu/drm/i915/i915_gem_dmabuf.c
+++ b/drivers/gpu/drm/i915/i915_gem_dmabuf.c
@@ -125,13 +125,15 @@ static void *i915_gem_dmabuf_vmap(struct dma_buf *dma_buf)
 
 	ret = i915_gem_object_get_pages(obj);
 	if (ret)
-		goto error;
+		goto err;
+
+	i915_gem_object_pin_pages(obj);
 
 	ret = -ENOMEM;
 
 	pages = drm_malloc_ab(obj->base.size >> PAGE_SHIFT, sizeof(*pages));
 	if (pages == NULL)
-		goto error;
+		goto err_unpin;
 
 	i = 0;
 	for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0)
@@ -141,15 +143,16 @@ static void *i915_gem_dmabuf_vmap(struct dma_buf *dma_buf)
 	drm_free_large(pages);
 
 	if (!obj->dma_buf_vmapping)
-		goto error;
+		goto err_unpin;
 
 	obj->vmapping_count = 1;
-	i915_gem_object_pin_pages(obj);
 out_unlock:
 	mutex_unlock(&dev->struct_mutex);
 	return obj->dma_buf_vmapping;
 
-error:
+err_unpin:
+	i915_gem_object_unpin_pages(obj);
+err:
 	mutex_unlock(&dev->struct_mutex);
 	return ERR_PTR(ret);
 }
diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c
index b7376533633d..8f3adc7d0dc8 100644
--- a/drivers/gpu/drm/i915/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/i915_gem_evict.c
@@ -88,6 +88,7 @@ i915_gem_evict_something(struct drm_device *dev, struct i915_address_space *vm,
 	} else
 		drm_mm_init_scan(&vm->mm, min_size, alignment, cache_level);
 
+search_again:
 	/* First see if there is a large enough contiguous idle region... */
 	list_for_each_entry(vma, &vm->inactive_list, mm_list) {
 		if (mark_free(vma, &unwind_list))
@@ -115,10 +116,17 @@ none:
 		list_del_init(&vma->exec_list);
 	}
 
-	/* We expect the caller to unpin, evict all and try again, or give up.
-	 * So calling i915_gem_evict_vm() is unnecessary.
+	/* Can we unpin some objects such as idle hw contents,
+	 * or pending flips?
 	 */
-	return -ENOSPC;
+	ret = nonblocking ? -ENOSPC : i915_gpu_idle(dev);
+	if (ret)
+		return ret;
+
+	/* Only idle the GPU and repeat the search once */
+	i915_gem_retire_requests(dev);
+	nonblocking = true;
+	goto search_again;
 
 found:
 	/* drm_mm doesn't allow any other other operations while
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 885d595e0e02..b7e787fb4649 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -33,6 +33,9 @@
 #include "intel_drv.h"
 #include <linux/dma_remapping.h>
 
+#define  __EXEC_OBJECT_HAS_PIN (1<<31)
+#define  __EXEC_OBJECT_HAS_FENCE (1<<30)
+
 struct eb_vmas {
 	struct list_head vmas;
 	int and;
@@ -187,7 +190,28 @@ static struct i915_vma *eb_get_vma(struct eb_vmas *eb, unsigned long handle)
 	}
 }
 
-static void eb_destroy(struct eb_vmas *eb) {
+static void
+i915_gem_execbuffer_unreserve_vma(struct i915_vma *vma)
+{
+	struct drm_i915_gem_exec_object2 *entry;
+	struct drm_i915_gem_object *obj = vma->obj;
+
+	if (!drm_mm_node_allocated(&vma->node))
+		return;
+
+	entry = vma->exec_entry;
+
+	if (entry->flags & __EXEC_OBJECT_HAS_FENCE)
+		i915_gem_object_unpin_fence(obj);
+
+	if (entry->flags & __EXEC_OBJECT_HAS_PIN)
+		i915_gem_object_unpin(obj);
+
+	entry->flags &= ~(__EXEC_OBJECT_HAS_FENCE | __EXEC_OBJECT_HAS_PIN);
+}
+
+static void eb_destroy(struct eb_vmas *eb)
+{
 	while (!list_empty(&eb->vmas)) {
 		struct i915_vma *vma;
 
@@ -195,6 +219,7 @@ static void eb_destroy(struct eb_vmas *eb) {
 				       struct i915_vma,
 				       exec_list);
 		list_del_init(&vma->exec_list);
+		i915_gem_execbuffer_unreserve_vma(vma);
 		drm_gem_object_unreference(&vma->obj->base);
 	}
 	kfree(eb);
@@ -478,9 +503,6 @@ i915_gem_execbuffer_relocate(struct eb_vmas *eb,
 	return ret;
 }
 
-#define  __EXEC_OBJECT_HAS_PIN (1<<31)
-#define  __EXEC_OBJECT_HAS_FENCE (1<<30)
-
 static int
 need_reloc_mappable(struct i915_vma *vma)
 {
@@ -552,26 +574,6 @@ i915_gem_execbuffer_reserve_vma(struct i915_vma *vma,
 	return 0;
 }
 
-static void
-i915_gem_execbuffer_unreserve_vma(struct i915_vma *vma)
-{
-	struct drm_i915_gem_exec_object2 *entry;
-	struct drm_i915_gem_object *obj = vma->obj;
-
-	if (!drm_mm_node_allocated(&vma->node))
-		return;
-
-	entry = vma->exec_entry;
-
-	if (entry->flags & __EXEC_OBJECT_HAS_FENCE)
-		i915_gem_object_unpin_fence(obj);
-
-	if (entry->flags & __EXEC_OBJECT_HAS_PIN)
-		i915_gem_object_unpin(obj);
-
-	entry->flags &= ~(__EXEC_OBJECT_HAS_FENCE | __EXEC_OBJECT_HAS_PIN);
-}
-
 static int
 i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
 			    struct list_head *vmas,
@@ -670,13 +672,14 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,
 				goto err;
 		}
 
-err:		/* Decrement pin count for bound objects */
-		list_for_each_entry(vma, vmas, exec_list)
-			i915_gem_execbuffer_unreserve_vma(vma);
-
+err:
 		if (ret != -ENOSPC || retry++)
 			return ret;
 
+		/* Decrement pin count for bound objects */
+		list_for_each_entry(vma, vmas, exec_list)
+			i915_gem_execbuffer_unreserve_vma(vma);
+
 		ret = i915_gem_evict_vm(vm, true);
 		if (ret)
 			return ret;
@@ -708,6 +711,7 @@ i915_gem_execbuffer_relocate_slow(struct drm_device *dev,
 	while (!list_empty(&eb->vmas)) {
 		vma = list_first_entry(&eb->vmas, struct i915_vma, exec_list);
 		list_del_init(&vma->exec_list);
+		i915_gem_execbuffer_unreserve_vma(vma);
 		drm_gem_object_unreference(&vma->obj->base);
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 3620a1b0a73c..c79dd2b1f70e 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -57,7 +57,9 @@ typedef gen8_gtt_pte_t gen8_ppgtt_pde_t;
 #define HSW_WB_LLC_AGE3			HSW_CACHEABILITY_CONTROL(0x2)
 #define HSW_WB_LLC_AGE0			HSW_CACHEABILITY_CONTROL(0x3)
 #define HSW_WB_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0xb)
+#define HSW_WB_ELLC_LLC_AGE3		HSW_CACHEABILITY_CONTROL(0x8)
 #define HSW_WT_ELLC_LLC_AGE0		HSW_CACHEABILITY_CONTROL(0x6)
+#define HSW_WT_ELLC_LLC_AGE3		HSW_CACHEABILITY_CONTROL(0x7)
 
 #define GEN8_PTES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_gtt_pte_t))
 #define GEN8_PDES_PER_PAGE		(PAGE_SIZE / sizeof(gen8_ppgtt_pde_t))
@@ -185,10 +187,10 @@ static gen6_gtt_pte_t iris_pte_encode(dma_addr_t addr,
 	case I915_CACHE_NONE:
 		break;
 	case I915_CACHE_WT:
-		pte |= HSW_WT_ELLC_LLC_AGE0;
+		pte |= HSW_WT_ELLC_LLC_AGE3;
 		break;
 	default:
-		pte |= HSW_WB_ELLC_LLC_AGE0;
+		pte |= HSW_WB_ELLC_LLC_AGE3;
 		break;
 	}
 
@@ -335,8 +337,8 @@ static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
 		kfree(ppgtt->gen8_pt_dma_addr[i]);
 	}
 
-	__free_pages(ppgtt->gen8_pt_pages, ppgtt->num_pt_pages << PAGE_SHIFT);
-	__free_pages(ppgtt->pd_pages, ppgtt->num_pd_pages << PAGE_SHIFT);
+	__free_pages(ppgtt->gen8_pt_pages, get_order(ppgtt->num_pt_pages << PAGE_SHIFT));
+	__free_pages(ppgtt->pd_pages, get_order(ppgtt->num_pd_pages << PAGE_SHIFT));
 }
 
 /**
@@ -1239,6 +1241,11 @@ static inline unsigned int gen8_get_total_gtt_size(u16 bdw_gmch_ctl)
 	bdw_gmch_ctl &= BDW_GMCH_GGMS_MASK;
 	if (bdw_gmch_ctl)
 		bdw_gmch_ctl = 1 << bdw_gmch_ctl;
+	if (bdw_gmch_ctl > 4) {
+		WARN_ON(!i915_preliminary_hw_support);
+		return 4<<20;
+	}
+
 	return bdw_gmch_ctl << 20;
 }
 
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index f9eafb6ed523..ee2742122a02 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -235,6 +235,7 @@
  */
 #define MI_LOAD_REGISTER_IMM(x)	MI_INSTR(0x22, 2*x-1)
 #define MI_STORE_REGISTER_MEM(x) MI_INSTR(0x24, 2*x-1)
+#define  MI_SRM_LRM_GLOBAL_GTT		(1<<22)
 #define MI_FLUSH_DW		MI_INSTR(0x26, 1) /* for GEN6 */
 #define   MI_FLUSH_DW_STORE_INDEX	(1<<21)
 #define   MI_INVALIDATE_TLB		(1<<18)
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 330077bcd0bd..526c8ded16b0 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -173,7 +173,7 @@ static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port)
 		ddi_translations = ddi_translations_dp;
 		break;
 	case PORT_D:
-		if (intel_dpd_is_edp(dev))
+		if (intel_dp_is_edp(dev, PORT_D))
 			ddi_translations = ddi_translations_edp;
 		else
 			ddi_translations = ddi_translations_dp;
@@ -1158,9 +1158,10 @@ static void intel_ddi_post_disable(struct intel_encoder *intel_encoder)
 	if (wait)
 		intel_wait_ddi_buf_idle(dev_priv, port);
 
-	if (type == INTEL_OUTPUT_EDP) {
+	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 		ironlake_edp_panel_vdd_on(intel_dp);
+		intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_OFF);
 		ironlake_edp_panel_off(intel_dp);
 	}
 
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 7ec8b488bb1d..8b8bde7dce53 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -5815,7 +5815,7 @@ static void intel_set_pipe_csc(struct drm_crtc *crtc)
 		uint16_t postoff = 0;
 
 		if (intel_crtc->config.limited_color_range)
-			postoff = (16 * (1 << 13) / 255) & 0x1fff;
+			postoff = (16 * (1 << 12) / 255) & 0x1fff;
 
 		I915_WRITE(PIPE_CSC_POSTOFF_HI(pipe), postoff);
 		I915_WRITE(PIPE_CSC_POSTOFF_ME(pipe), postoff);
@@ -6402,7 +6402,7 @@ static void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 
 	/* Make sure we're not on PC8 state before disabling PC8, otherwise
 	 * we'll hang the machine! */
-	dev_priv->uncore.funcs.force_wake_get(dev_priv);
+	gen6_gt_force_wake_get(dev_priv);
 
 	if (val & LCPLL_POWER_DOWN_ALLOW) {
 		val &= ~LCPLL_POWER_DOWN_ALLOW;
@@ -6436,7 +6436,7 @@ static void hsw_restore_lcpll(struct drm_i915_private *dev_priv)
 			DRM_ERROR("Switching back to LCPLL failed\n");
 	}
 
-	dev_priv->uncore.funcs.force_wake_put(dev_priv);
+	gen6_gt_force_wake_put(dev_priv);
 }
 
 void hsw_enable_pc8_work(struct work_struct *__work)
@@ -8354,7 +8354,8 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 		intel_ring_emit(ring, ~(DERRMR_PIPEA_PRI_FLIP_DONE |
 					DERRMR_PIPEB_PRI_FLIP_DONE |
 					DERRMR_PIPEC_PRI_FLIP_DONE));
-		intel_ring_emit(ring, MI_STORE_REGISTER_MEM(1));
+		intel_ring_emit(ring, MI_STORE_REGISTER_MEM(1) |
+				MI_SRM_LRM_GLOBAL_GTT);
 		intel_ring_emit(ring, DERRMR);
 		intel_ring_emit(ring, ring->scratch.gtt_offset + 256);
 	}
@@ -9134,7 +9135,7 @@ intel_pipe_config_compare(struct drm_device *dev,
 	if (IS_G4X(dev) || INTEL_INFO(dev)->gen >= 5)
 		PIPE_CONF_CHECK_I(pipe_bpp);
 
-	if (!IS_HASWELL(dev)) {
+	if (!HAS_DDI(dev)) {
 		PIPE_CONF_CHECK_CLOCK_FUZZY(adjusted_mode.crtc_clock);
 		PIPE_CONF_CHECK_CLOCK_FUZZY(port_clock);
 	}
@@ -10049,7 +10050,7 @@ static void intel_setup_outputs(struct drm_device *dev)
 			intel_ddi_init(dev, PORT_D);
 	} else if (HAS_PCH_SPLIT(dev)) {
 		int found;
-		dpd_is_edp = intel_dpd_is_edp(dev);
+		dpd_is_edp = intel_dp_is_edp(dev, PORT_D);
 
 		if (has_edp_a(dev))
 			intel_dp_init(dev, DP_A, PORT_A);
@@ -10086,8 +10087,7 @@ static void intel_setup_outputs(struct drm_device *dev)
 			intel_hdmi_init(dev, VLV_DISPLAY_BASE + GEN4_HDMIC,
 					PORT_C);
 			if (I915_READ(VLV_DISPLAY_BASE + DP_C) & DP_DETECTED)
-				intel_dp_init(dev, VLV_DISPLAY_BASE + DP_C,
-					      PORT_C);
+				intel_dp_init(dev, VLV_DISPLAY_BASE + DP_C, PORT_C);
 		}
 
 		intel_dsi_init(dev);
@@ -11036,8 +11036,6 @@ void intel_modeset_setup_hw_state(struct drm_device *dev,
 	}
 
 	intel_modeset_check_state(dev);
-
-	drm_mode_config_reset(dev);
 }
 
 void intel_modeset_gem_init(struct drm_device *dev)
@@ -11046,7 +11044,10 @@ void intel_modeset_gem_init(struct drm_device *dev)
 
 	intel_setup_overlay(dev);
 
+	drm_modeset_lock_all(dev);
+	drm_mode_config_reset(dev);
 	intel_modeset_setup_hw_state(dev, false);
+	drm_modeset_unlock_all(dev);
 }
 
 void intel_modeset_cleanup(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 0b2e842fef01..30c627c7b7ba 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -3326,11 +3326,19 @@ intel_trans_dp_port_sel(struct drm_crtc *crtc)
 }
 
 /* check the VBT to see whether the eDP is on DP-D port */
-bool intel_dpd_is_edp(struct drm_device *dev)
+bool intel_dp_is_edp(struct drm_device *dev, enum port port)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	union child_device_config *p_child;
 	int i;
+	static const short port_mapping[] = {
+		[PORT_B] = PORT_IDPB,
+		[PORT_C] = PORT_IDPC,
+		[PORT_D] = PORT_IDPD,
+	};
+
+	if (port == PORT_A)
+		return true;
 
 	if (!dev_priv->vbt.child_dev_num)
 		return false;
@@ -3338,7 +3346,7 @@ bool intel_dpd_is_edp(struct drm_device *dev)
 	for (i = 0; i < dev_priv->vbt.child_dev_num; i++) {
 		p_child = dev_priv->vbt.child_dev + i;
 
-		if (p_child->common.dvo_port == PORT_IDPD &&
+		if (p_child->common.dvo_port == port_mapping[port] &&
 		    (p_child->common.device_type & DEVICE_TYPE_eDP_BITS) ==
 		    (DEVICE_TYPE_eDP & DEVICE_TYPE_eDP_BITS))
 			return true;
@@ -3616,26 +3624,10 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 	intel_dp->DP = I915_READ(intel_dp->output_reg);
 	intel_dp->attached_connector = intel_connector;
 
-	type = DRM_MODE_CONNECTOR_DisplayPort;
-	/*
-	 * FIXME : We need to initialize built-in panels before external panels.
-	 * For X0, DP_C is fixed as eDP. Revisit this as part of VLV eDP cleanup
-	 */
-	switch (port) {
-	case PORT_A:
+	if (intel_dp_is_edp(dev, port))
 		type = DRM_MODE_CONNECTOR_eDP;
-		break;
-	case PORT_C:
-		if (IS_VALLEYVIEW(dev))
-			type = DRM_MODE_CONNECTOR_eDP;
-		break;
-	case PORT_D:
-		if (HAS_PCH_SPLIT(dev) && intel_dpd_is_edp(dev))
-			type = DRM_MODE_CONNECTOR_eDP;
-		break;
-	default:	/* silence GCC warning */
-		break;
-	}
+	else
+		type = DRM_MODE_CONNECTOR_DisplayPort;
 
 	/*
 	 * For eDP we always set the encoder type to INTEL_OUTPUT_EDP, but
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 1e49aa8f5377..79f91f26e288 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -708,7 +708,7 @@ void intel_dp_encoder_destroy(struct drm_encoder *encoder);
 void intel_dp_check_link_status(struct intel_dp *intel_dp);
 bool intel_dp_compute_config(struct intel_encoder *encoder,
 			     struct intel_crtc_config *pipe_config);
-bool intel_dpd_is_edp(struct drm_device *dev);
+bool intel_dp_is_edp(struct drm_device *dev, enum port port);
 void ironlake_edp_backlight_on(struct intel_dp *intel_dp);
 void ironlake_edp_backlight_off(struct intel_dp *intel_dp);
 void ironlake_edp_panel_on(struct intel_dp *intel_dp);
@@ -821,6 +821,7 @@ void intel_update_sprite_watermarks(struct drm_plane *plane,
 				    uint32_t sprite_width, int pixel_size,
 				    bool enabled, bool scaled);
 void intel_init_pm(struct drm_device *dev);
+void intel_pm_setup(struct drm_device *dev);
 bool intel_fbc_enabled(struct drm_device *dev);
 void intel_update_fbc(struct drm_device *dev);
 void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index f161ac02c4f6..e6f782d1c669 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -451,7 +451,9 @@ static u32 intel_panel_get_backlight(struct drm_device *dev,
 
 	spin_lock_irqsave(&dev_priv->backlight.lock, flags);
 
-	if (HAS_PCH_SPLIT(dev)) {
+	if (IS_BROADWELL(dev)) {
+		val = I915_READ(BLC_PWM_PCH_CTL2) & BACKLIGHT_DUTY_CYCLE_MASK;
+	} else if (HAS_PCH_SPLIT(dev)) {
 		val = I915_READ(BLC_PWM_CPU_CTL) & BACKLIGHT_DUTY_CYCLE_MASK;
 	} else {
 		if (IS_VALLEYVIEW(dev))
@@ -479,6 +481,13 @@ static u32 intel_panel_get_backlight(struct drm_device *dev,
 	return val;
 }
 
+static void intel_bdw_panel_set_backlight(struct drm_device *dev, u32 level)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 val = I915_READ(BLC_PWM_PCH_CTL2) & ~BACKLIGHT_DUTY_CYCLE_MASK;
+	I915_WRITE(BLC_PWM_PCH_CTL2, val | level);
+}
+
 static void intel_pch_panel_set_backlight(struct drm_device *dev, u32 level)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -496,7 +505,9 @@ static void intel_panel_actually_set_backlight(struct drm_device *dev,
 	DRM_DEBUG_DRIVER("set backlight PWM = %d\n", level);
 	level = intel_panel_compute_brightness(dev, pipe, level);
 
-	if (HAS_PCH_SPLIT(dev))
+	if (IS_BROADWELL(dev))
+		return intel_bdw_panel_set_backlight(dev, level);
+	else if (HAS_PCH_SPLIT(dev))
 		return intel_pch_panel_set_backlight(dev, level);
 
 	if (is_backlight_combination_mode(dev)) {
@@ -666,7 +677,16 @@ void intel_panel_enable_backlight(struct intel_connector *connector)
 		POSTING_READ(reg);
 		I915_WRITE(reg, tmp | BLM_PWM_ENABLE);
 
-		if (HAS_PCH_SPLIT(dev) &&
+		if (IS_BROADWELL(dev)) {
+			/*
+			 * Broadwell requires PCH override to drive the PCH
+			 * backlight pin. The above will configure the CPU
+			 * backlight pin, which we don't plan to use.
+			 */
+			tmp = I915_READ(BLC_PWM_PCH_CTL1);
+			tmp |= BLM_PCH_OVERRIDE_ENABLE | BLM_PCH_PWM_ENABLE;
+			I915_WRITE(BLC_PWM_PCH_CTL1, tmp);
+		} else if (HAS_PCH_SPLIT(dev) &&
 		    !(dev_priv->quirks & QUIRK_NO_PCH_PWM_ENABLE)) {
 			tmp = I915_READ(BLC_PWM_PCH_CTL1);
 			tmp |= BLM_PCH_PWM_ENABLE;
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index caf2ee4e5441..3657ab43c8fd 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1180,7 +1180,7 @@ static bool g4x_compute_wm0(struct drm_device *dev,
 
 	adjusted_mode = &to_intel_crtc(crtc)->config.adjusted_mode;
 	clock = adjusted_mode->crtc_clock;
-	htotal = adjusted_mode->htotal;
+	htotal = adjusted_mode->crtc_htotal;
 	hdisplay = to_intel_crtc(crtc)->config.pipe_src_w;
 	pixel_size = crtc->fb->bits_per_pixel / 8;
 
@@ -1267,7 +1267,7 @@ static bool g4x_compute_srwm(struct drm_device *dev,
 	crtc = intel_get_crtc_for_plane(dev, plane);
 	adjusted_mode = &to_intel_crtc(crtc)->config.adjusted_mode;
 	clock = adjusted_mode->crtc_clock;
-	htotal = adjusted_mode->htotal;
+	htotal = adjusted_mode->crtc_htotal;
 	hdisplay = to_intel_crtc(crtc)->config.pipe_src_w;
 	pixel_size = crtc->fb->bits_per_pixel / 8;
 
@@ -1498,7 +1498,7 @@ static void i965_update_wm(struct drm_crtc *unused_crtc)
 		const struct drm_display_mode *adjusted_mode =
 			&to_intel_crtc(crtc)->config.adjusted_mode;
 		int clock = adjusted_mode->crtc_clock;
-		int htotal = adjusted_mode->htotal;
+		int htotal = adjusted_mode->crtc_htotal;
 		int hdisplay = to_intel_crtc(crtc)->config.pipe_src_w;
 		int pixel_size = crtc->fb->bits_per_pixel / 8;
 		unsigned long line_time_us;
@@ -1624,7 +1624,7 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 		const struct drm_display_mode *adjusted_mode =
 			&to_intel_crtc(enabled)->config.adjusted_mode;
 		int clock = adjusted_mode->crtc_clock;
-		int htotal = adjusted_mode->htotal;
+		int htotal = adjusted_mode->crtc_htotal;
 		int hdisplay = to_intel_crtc(enabled)->config.pipe_src_w;
 		int pixel_size = enabled->fb->bits_per_pixel / 8;
 		unsigned long line_time_us;
@@ -1776,7 +1776,7 @@ static bool ironlake_compute_srwm(struct drm_device *dev, int level, int plane,
 	crtc = intel_get_crtc_for_plane(dev, plane);
 	adjusted_mode = &to_intel_crtc(crtc)->config.adjusted_mode;
 	clock = adjusted_mode->crtc_clock;
-	htotal = adjusted_mode->htotal;
+	htotal = adjusted_mode->crtc_htotal;
 	hdisplay = to_intel_crtc(crtc)->config.pipe_src_w;
 	pixel_size = crtc->fb->bits_per_pixel / 8;
 
@@ -2469,8 +2469,9 @@ hsw_compute_linetime_wm(struct drm_device *dev, struct drm_crtc *crtc)
 	/* The WM are computed with base on how long it takes to fill a single
 	 * row at the given clock rate, multiplied by 8.
 	 * */
-	linetime = DIV_ROUND_CLOSEST(mode->htotal * 1000 * 8, mode->clock);
-	ips_linetime = DIV_ROUND_CLOSEST(mode->htotal * 1000 * 8,
+	linetime = DIV_ROUND_CLOSEST(mode->crtc_htotal * 1000 * 8,
+				     mode->crtc_clock);
+	ips_linetime = DIV_ROUND_CLOSEST(mode->crtc_htotal * 1000 * 8,
 					 intel_ddi_get_cdclk_freq(dev_priv));
 
 	return PIPE_WM_LINETIME_IPS_LINETIME(ips_linetime) |
@@ -5684,6 +5685,7 @@ static void __intel_set_power_well(struct drm_device *dev, bool enable)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	bool is_enabled, enable_requested;
+	unsigned long irqflags;
 	uint32_t tmp;
 
 	tmp = I915_READ(HSW_PWR_WELL_DRIVER);
@@ -5701,9 +5703,24 @@ static void __intel_set_power_well(struct drm_device *dev, bool enable)
 				      HSW_PWR_WELL_STATE_ENABLED), 20))
 				DRM_ERROR("Timeout enabling power well\n");
 		}
+
+		if (IS_BROADWELL(dev)) {
+			spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
+			I915_WRITE(GEN8_DE_PIPE_IMR(PIPE_B),
+				   dev_priv->de_irq_mask[PIPE_B]);
+			I915_WRITE(GEN8_DE_PIPE_IER(PIPE_B),
+				   ~dev_priv->de_irq_mask[PIPE_B] |
+				   GEN8_PIPE_VBLANK);
+			I915_WRITE(GEN8_DE_PIPE_IMR(PIPE_C),
+				   dev_priv->de_irq_mask[PIPE_C]);
+			I915_WRITE(GEN8_DE_PIPE_IER(PIPE_C),
+				   ~dev_priv->de_irq_mask[PIPE_C] |
+				   GEN8_PIPE_VBLANK);
+			POSTING_READ(GEN8_DE_PIPE_IER(PIPE_C));
+			spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
+		}
 	} else {
 		if (enable_requested) {
-			unsigned long irqflags;
 			enum pipe p;
 
 			I915_WRITE(HSW_PWR_WELL_DRIVER, 0);
@@ -6129,10 +6146,19 @@ int vlv_freq_opcode(int ddr_freq, int val)
 	return val;
 }
 
-void intel_pm_init(struct drm_device *dev)
+void intel_pm_setup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
+	mutex_init(&dev_priv->rps.hw_lock);
+
+	mutex_init(&dev_priv->pc8.lock);
+	dev_priv->pc8.requirements_met = false;
+	dev_priv->pc8.gpu_idle = false;
+	dev_priv->pc8.irqs_disabled = false;
+	dev_priv->pc8.enabled = false;
+	dev_priv->pc8.disable_count = 2; /* requirements_met + gpu_idle */
+	INIT_DELAYED_WORK(&dev_priv->pc8.enable_work, hsw_enable_pc8_work);
 	INIT_DELAYED_WORK(&dev_priv->rps.delayed_resume_work,
 			  intel_gen6_powersave_work);
 }
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index b620337e6d67..c2f09d456300 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -965,6 +965,7 @@ void intel_ring_setup_status_page(struct intel_ring_buffer *ring)
 	} else if (IS_GEN6(ring->dev)) {
 		mmio = RING_HWS_PGA_GEN6(ring->mmio_base);
 	} else {
+		/* XXX: gen8 returns to sanity */
 		mmio = RING_HWS_PGA(ring->mmio_base);
 	}
 
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 0b02078a0b84..25cbe073c388 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -784,6 +784,7 @@ static int gen6_do_reset(struct drm_device *dev)
 int intel_gpu_reset(struct drm_device *dev)
 {
 	switch (INTEL_INFO(dev)->gen) {
+	case 8:
 	case 7:
 	case 6: return gen6_do_reset(dev);
 	case 5: return ironlake_do_reset(dev);
diff --git a/drivers/gpu/drm/nouveau/Makefile b/drivers/gpu/drm/nouveau/Makefile
index edcf801613e6..b3fa1ba191b7 100644
--- a/drivers/gpu/drm/nouveau/Makefile
+++ b/drivers/gpu/drm/nouveau/Makefile
@@ -59,6 +59,7 @@ nouveau-y += core/subdev/clock/nv40.o
 nouveau-y += core/subdev/clock/nv50.o
 nouveau-y += core/subdev/clock/nv84.o
 nouveau-y += core/subdev/clock/nva3.o
+nouveau-y += core/subdev/clock/nvaa.o
 nouveau-y += core/subdev/clock/nvc0.o
 nouveau-y += core/subdev/clock/nve0.o
 nouveau-y += core/subdev/clock/pllnv04.o
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv50.c b/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
index db139827047c..db3fc7be856a 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv50.c
@@ -283,7 +283,7 @@ nv50_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nvaa_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
@@ -311,7 +311,7 @@ nv50_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nouveau_bios_oclass;
 		device->oclass[NVDEV_SUBDEV_GPIO   ] = &nv50_gpio_oclass;
 		device->oclass[NVDEV_SUBDEV_I2C    ] = &nv94_i2c_oclass;
-		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nv84_clock_oclass;
+		device->oclass[NVDEV_SUBDEV_CLOCK  ] =  nvaa_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv84_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] = &nv50_devinit_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c
index 5f555788121c..e6352bd5b4ff 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv50.c
@@ -33,6 +33,7 @@
 #include <engine/dmaobj.h>
 #include <engine/fifo.h>
 
+#include "nv04.h"
 #include "nv50.h"
 
 /*******************************************************************************
@@ -460,6 +461,8 @@ nv50_fifo_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	nv_subdev(priv)->intr = nv04_fifo_intr;
 	nv_engine(priv)->cclass = &nv50_fifo_cclass;
 	nv_engine(priv)->sclass = nv50_fifo_sclass;
+	priv->base.pause = nv04_fifo_pause;
+	priv->base.start = nv04_fifo_start;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
index 0908dc834c84..fe0f41e65d9b 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv84.c
@@ -35,6 +35,7 @@
 #include <engine/dmaobj.h>
 #include <engine/fifo.h>
 
+#include "nv04.h"
 #include "nv50.h"
 
 /*******************************************************************************
@@ -432,6 +433,8 @@ nv84_fifo_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	nv_subdev(priv)->intr = nv04_fifo_intr;
 	nv_engine(priv)->cclass = &nv84_fifo_cclass;
 	nv_engine(priv)->sclass = nv84_fifo_sclass;
+	priv->base.pause = nv04_fifo_pause;
+	priv->base.start = nv04_fifo_start;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/core/engine/software/nv50.c b/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
index b574dd4bb828..5ce686ee729e 100644
--- a/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/software/nv50.c
@@ -176,7 +176,7 @@ nv50_software_context_ctor(struct nouveau_object *parent,
 	if (ret)
 		return ret;
 
-	chan->vblank.nr_event = pdisp->vblank->index_nr;
+	chan->vblank.nr_event = pdisp ? pdisp->vblank->index_nr : 0;
 	chan->vblank.event = kzalloc(chan->vblank.nr_event *
 				     sizeof(*chan->vblank.event), GFP_KERNEL);
 	if (!chan->vblank.event)
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/clock.h b/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
index e2675bc0edba..8f4ced75444a 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/clock.h
@@ -14,6 +14,9 @@ enum nv_clk_src {
 	nv_clk_src_hclk,
 	nv_clk_src_hclkm3,
 	nv_clk_src_hclkm3d2,
+	nv_clk_src_hclkm2d3, /* NVAA */
+	nv_clk_src_hclkm4, /* NVAA */
+	nv_clk_src_cclk, /* NVAA */
 
 	nv_clk_src_host,
 
@@ -127,6 +130,7 @@ extern struct nouveau_oclass nv04_clock_oclass;
 extern struct nouveau_oclass nv40_clock_oclass;
 extern struct nouveau_oclass *nv50_clock_oclass;
 extern struct nouveau_oclass *nv84_clock_oclass;
+extern struct nouveau_oclass *nvaa_clock_oclass;
 extern struct nouveau_oclass nva3_clock_oclass;
 extern struct nouveau_oclass nvc0_clock_oclass;
 extern struct nouveau_oclass nve0_clock_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
index da50c1b12928..30c1f3a4158e 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nv04.c
@@ -69,6 +69,11 @@ nv04_clock_pll_prog(struct nouveau_clock *clk, u32 reg1,
 	return 0;
 }
 
+static struct nouveau_clocks
+nv04_domain[] = {
+	{ nv_clk_src_max }
+};
+
 static int
 nv04_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 		struct nouveau_oclass *oclass, void *data, u32 size,
@@ -77,7 +82,7 @@ nv04_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
 	struct nv04_clock_priv *priv;
 	int ret;
 
-	ret = nouveau_clock_create(parent, engine, oclass, NULL, &priv);
+	ret = nouveau_clock_create(parent, engine, oclass, nv04_domain, &priv);
 	*pobject = nv_object(priv);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c b/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c
new file mode 100644
index 000000000000..7a723b4f564d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/clock/nvaa.c
@@ -0,0 +1,445 @@
+/*
+ * Copyright 2012 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+
+#include <engine/fifo.h>
+#include <subdev/bios.h>
+#include <subdev/bios/pll.h>
+#include <subdev/timer.h>
+#include <subdev/clock.h>
+
+#include "pll.h"
+
+struct nvaa_clock_priv {
+	struct nouveau_clock base;
+	enum nv_clk_src csrc, ssrc, vsrc;
+	u32 cctrl, sctrl;
+	u32 ccoef, scoef;
+	u32 cpost, spost;
+	u32 vdiv;
+};
+
+static u32
+read_div(struct nouveau_clock *clk)
+{
+	return nv_rd32(clk, 0x004600);
+}
+
+static u32
+read_pll(struct nouveau_clock *clk, u32 base)
+{
+	u32 ctrl = nv_rd32(clk, base + 0);
+	u32 coef = nv_rd32(clk, base + 4);
+	u32 ref = clk->read(clk, nv_clk_src_href);
+	u32 post_div = 0;
+	u32 clock = 0;
+	int N1, M1;
+
+	switch (base){
+	case 0x4020:
+		post_div = 1 << ((nv_rd32(clk, 0x4070) & 0x000f0000) >> 16);
+		break;
+	case 0x4028:
+		post_div = (nv_rd32(clk, 0x4040) & 0x000f0000) >> 16;
+		break;
+	default:
+		break;
+	}
+
+	N1 = (coef & 0x0000ff00) >> 8;
+	M1 = (coef & 0x000000ff);
+	if ((ctrl & 0x80000000) && M1) {
+		clock = ref * N1 / M1;
+		clock = clock / post_div;
+	}
+
+	return clock;
+}
+
+static int
+nvaa_clock_read(struct nouveau_clock *clk, enum nv_clk_src src)
+{
+	struct nvaa_clock_priv *priv = (void *)clk;
+	u32 mast = nv_rd32(clk, 0x00c054);
+	u32 P = 0;
+
+	switch (src) {
+	case nv_clk_src_crystal:
+		return nv_device(priv)->crystal;
+	case nv_clk_src_href:
+		return 100000; /* PCIE reference clock */
+	case nv_clk_src_hclkm4:
+		return clk->read(clk, nv_clk_src_href) * 4;
+	case nv_clk_src_hclkm2d3:
+		return clk->read(clk, nv_clk_src_href) * 2 / 3;
+	case nv_clk_src_host:
+		switch (mast & 0x000c0000) {
+		case 0x00000000: return clk->read(clk, nv_clk_src_hclkm2d3);
+		case 0x00040000: break;
+		case 0x00080000: return clk->read(clk, nv_clk_src_hclkm4);
+		case 0x000c0000: return clk->read(clk, nv_clk_src_cclk);
+		}
+		break;
+	case nv_clk_src_core:
+		P = (nv_rd32(clk, 0x004028) & 0x00070000) >> 16;
+
+		switch (mast & 0x00000003) {
+		case 0x00000000: return clk->read(clk, nv_clk_src_crystal) >> P;
+		case 0x00000001: return 0;
+		case 0x00000002: return clk->read(clk, nv_clk_src_hclkm4) >> P;
+		case 0x00000003: return read_pll(clk, 0x004028) >> P;
+		}
+		break;
+	case nv_clk_src_cclk:
+		if ((mast & 0x03000000) != 0x03000000)
+			return clk->read(clk, nv_clk_src_core);
+
+		if ((mast & 0x00000200) == 0x00000000)
+			return clk->read(clk, nv_clk_src_core);
+
+		switch (mast & 0x00000c00) {
+		case 0x00000000: return clk->read(clk, nv_clk_src_href);
+		case 0x00000400: return clk->read(clk, nv_clk_src_hclkm4);
+		case 0x00000800: return clk->read(clk, nv_clk_src_hclkm2d3);
+		default: return 0;
+		}
+	case nv_clk_src_shader:
+		P = (nv_rd32(clk, 0x004020) & 0x00070000) >> 16;
+		switch (mast & 0x00000030) {
+		case 0x00000000:
+			if (mast & 0x00000040)
+				return clk->read(clk, nv_clk_src_href) >> P;
+			return clk->read(clk, nv_clk_src_crystal) >> P;
+		case 0x00000010: break;
+		case 0x00000020: return read_pll(clk, 0x004028) >> P;
+		case 0x00000030: return read_pll(clk, 0x004020) >> P;
+		}
+		break;
+	case nv_clk_src_mem:
+		return 0;
+		break;
+	case nv_clk_src_vdec:
+		P = (read_div(clk) & 0x00000700) >> 8;
+
+		switch (mast & 0x00400000) {
+		case 0x00400000:
+			return clk->read(clk, nv_clk_src_core) >> P;
+			break;
+		default:
+			return 500000 >> P;
+			break;
+		}
+		break;
+	default:
+		break;
+	}
+
+	nv_debug(priv, "unknown clock source %d 0x%08x\n", src, mast);
+	return 0;
+}
+
+static u32
+calc_pll(struct nvaa_clock_priv *priv, u32 reg,
+	 u32 clock, int *N, int *M, int *P)
+{
+	struct nouveau_bios *bios = nouveau_bios(priv);
+	struct nvbios_pll pll;
+	struct nouveau_clock *clk = &priv->base;
+	int ret;
+
+	ret = nvbios_pll_parse(bios, reg, &pll);
+	if (ret)
+		return 0;
+
+	pll.vco2.max_freq = 0;
+	pll.refclk = clk->read(clk, nv_clk_src_href);
+	if (!pll.refclk)
+		return 0;
+
+	return nv04_pll_calc(nv_subdev(priv), &pll, clock, N, M, NULL, NULL, P);
+}
+
+static inline u32
+calc_P(u32 src, u32 target, int *div)
+{
+	u32 clk0 = src, clk1 = src;
+	for (*div = 0; *div <= 7; (*div)++) {
+		if (clk0 <= target) {
+			clk1 = clk0 << (*div ? 1 : 0);
+			break;
+		}
+		clk0 >>= 1;
+	}
+
+	if (target - clk0 <= clk1 - target)
+		return clk0;
+	(*div)--;
+	return clk1;
+}
+
+static int
+nvaa_clock_calc(struct nouveau_clock *clk, struct nouveau_cstate *cstate)
+{
+	struct nvaa_clock_priv *priv = (void *)clk;
+	const int shader = cstate->domain[nv_clk_src_shader];
+	const int core = cstate->domain[nv_clk_src_core];
+	const int vdec = cstate->domain[nv_clk_src_vdec];
+	u32 out = 0, clock = 0;
+	int N, M, P1, P2 = 0;
+	int divs = 0;
+
+	/* cclk: find suitable source, disable PLL if we can */
+	if (core < clk->read(clk, nv_clk_src_hclkm4))
+		out = calc_P(clk->read(clk, nv_clk_src_hclkm4), core, &divs);
+
+	/* Calculate clock * 2, so shader clock can use it too */
+	clock = calc_pll(priv, 0x4028, (core << 1), &N, &M, &P1);
+
+	if (abs(core - out) <=
+	    abs(core - (clock >> 1))) {
+		priv->csrc = nv_clk_src_hclkm4;
+		priv->cctrl = divs << 16;
+	} else {
+		/* NVCTRL is actually used _after_ NVPOST, and after what we
+		 * call NVPLL. To make matters worse, NVPOST is an integer
+		 * divider instead of a right-shift number. */
+		if(P1 > 2) {
+			P2 = P1 - 2;
+			P1 = 2;
+		}
+
+		priv->csrc = nv_clk_src_core;
+		priv->ccoef = (N << 8) | M;
+
+		priv->cctrl = (P2 + 1) << 16;
+		priv->cpost = (1 << P1) << 16;
+	}
+
+	/* sclk: nvpll + divisor, href or spll */
+	out = 0;
+	if (shader == clk->read(clk, nv_clk_src_href)) {
+		priv->ssrc = nv_clk_src_href;
+	} else {
+		clock = calc_pll(priv, 0x4020, shader, &N, &M, &P1);
+		if (priv->csrc == nv_clk_src_core) {
+			out = calc_P((core << 1), shader, &divs);
+		}
+
+		if (abs(shader - out) <=
+		    abs(shader - clock) &&
+		   (divs + P2) <= 7) {
+			priv->ssrc = nv_clk_src_core;
+			priv->sctrl = (divs + P2) << 16;
+		} else {
+			priv->ssrc = nv_clk_src_shader;
+			priv->scoef = (N << 8) | M;
+			priv->sctrl = P1 << 16;
+		}
+	}
+
+	/* vclk */
+	out = calc_P(core, vdec, &divs);
+	clock = calc_P(500000, vdec, &P1);
+	if(abs(vdec - out) <=
+	   abs(vdec - clock)) {
+		priv->vsrc = nv_clk_src_cclk;
+		priv->vdiv = divs << 16;
+	} else {
+		priv->vsrc = nv_clk_src_vdec;
+		priv->vdiv = P1 << 16;
+	}
+
+	/* Print strategy! */
+	nv_debug(priv, "nvpll: %08x %08x %08x\n",
+			priv->ccoef, priv->cpost, priv->cctrl);
+	nv_debug(priv, " spll: %08x %08x %08x\n",
+			priv->scoef, priv->spost, priv->sctrl);
+	nv_debug(priv, " vdiv: %08x\n", priv->vdiv);
+	if (priv->csrc == nv_clk_src_hclkm4)
+		nv_debug(priv, "core: hrefm4\n");
+	else
+		nv_debug(priv, "core: nvpll\n");
+
+	if (priv->ssrc == nv_clk_src_hclkm4)
+		nv_debug(priv, "shader: hrefm4\n");
+	else if (priv->ssrc == nv_clk_src_core)
+		nv_debug(priv, "shader: nvpll\n");
+	else
+		nv_debug(priv, "shader: spll\n");
+
+	if (priv->vsrc == nv_clk_src_hclkm4)
+		nv_debug(priv, "vdec: 500MHz\n");
+	else
+		nv_debug(priv, "vdec: core\n");
+
+	return 0;
+}
+
+static int
+nvaa_clock_prog(struct nouveau_clock *clk)
+{
+	struct nvaa_clock_priv *priv = (void *)clk;
+	struct nouveau_fifo *pfifo = nouveau_fifo(clk);
+	unsigned long flags;
+	u32 pllmask = 0, mast, ptherm_gate;
+	int ret = -EBUSY;
+
+	/* halt and idle execution engines */
+	ptherm_gate = nv_mask(clk, 0x020060, 0x00070000, 0x00000000);
+	nv_mask(clk, 0x002504, 0x00000001, 0x00000001);
+	/* Wait until the interrupt handler is finished */
+	if (!nv_wait(clk, 0x000100, 0xffffffff, 0x00000000))
+		goto resume;
+
+	if (pfifo)
+		pfifo->pause(pfifo, &flags);
+
+	if (!nv_wait(clk, 0x002504, 0x00000010, 0x00000010))
+		goto resume;
+	if (!nv_wait(clk, 0x00251c, 0x0000003f, 0x0000003f))
+		goto resume;
+
+	/* First switch to safe clocks: href */
+	mast = nv_mask(clk, 0xc054, 0x03400e70, 0x03400640);
+	mast &= ~0x00400e73;
+	mast |= 0x03000000;
+
+	switch (priv->csrc) {
+	case nv_clk_src_hclkm4:
+		nv_mask(clk, 0x4028, 0x00070000, priv->cctrl);
+		mast |= 0x00000002;
+		break;
+	case nv_clk_src_core:
+		nv_wr32(clk, 0x402c, priv->ccoef);
+		nv_wr32(clk, 0x4028, 0x80000000 | priv->cctrl);
+		nv_wr32(clk, 0x4040, priv->cpost);
+		pllmask |= (0x3 << 8);
+		mast |= 0x00000003;
+		break;
+	default:
+		nv_warn(priv,"Reclocking failed: unknown core clock\n");
+		goto resume;
+	}
+
+	switch (priv->ssrc) {
+	case nv_clk_src_href:
+		nv_mask(clk, 0x4020, 0x00070000, 0x00000000);
+		/* mast |= 0x00000000; */
+		break;
+	case nv_clk_src_core:
+		nv_mask(clk, 0x4020, 0x00070000, priv->sctrl);
+		mast |= 0x00000020;
+		break;
+	case nv_clk_src_shader:
+		nv_wr32(clk, 0x4024, priv->scoef);
+		nv_wr32(clk, 0x4020, 0x80000000 | priv->sctrl);
+		nv_wr32(clk, 0x4070, priv->spost);
+		pllmask |= (0x3 << 12);
+		mast |= 0x00000030;
+		break;
+	default:
+		nv_warn(priv,"Reclocking failed: unknown sclk clock\n");
+		goto resume;
+	}
+
+	if (!nv_wait(clk, 0x004080, pllmask, pllmask)) {
+		nv_warn(priv,"Reclocking failed: unstable PLLs\n");
+		goto resume;
+	}
+
+	switch (priv->vsrc) {
+	case nv_clk_src_cclk:
+		mast |= 0x00400000;
+	default:
+		nv_wr32(clk, 0x4600, priv->vdiv);
+	}
+
+	nv_wr32(clk, 0xc054, mast);
+	ret = 0;
+
+resume:
+	if (pfifo)
+		pfifo->start(pfifo, &flags);
+
+	nv_mask(clk, 0x002504, 0x00000001, 0x00000000);
+	nv_wr32(clk, 0x020060, ptherm_gate);
+
+	/* Disable some PLLs and dividers when unused */
+	if (priv->csrc != nv_clk_src_core) {
+		nv_wr32(clk, 0x4040, 0x00000000);
+		nv_mask(clk, 0x4028, 0x80000000, 0x00000000);
+	}
+
+	if (priv->ssrc != nv_clk_src_shader) {
+		nv_wr32(clk, 0x4070, 0x00000000);
+		nv_mask(clk, 0x4020, 0x80000000, 0x00000000);
+	}
+
+	return ret;
+}
+
+static void
+nvaa_clock_tidy(struct nouveau_clock *clk)
+{
+}
+
+static struct nouveau_clocks
+nvaa_domains[] = {
+	{ nv_clk_src_crystal, 0xff },
+	{ nv_clk_src_href   , 0xff },
+	{ nv_clk_src_core   , 0xff, 0, "core", 1000 },
+	{ nv_clk_src_shader , 0xff, 0, "shader", 1000 },
+	{ nv_clk_src_vdec   , 0xff, 0, "vdec", 1000 },
+	{ nv_clk_src_max }
+};
+
+static int
+nvaa_clock_ctor(struct nouveau_object *parent, struct nouveau_object *engine,
+		struct nouveau_oclass *oclass, void *data, u32 size,
+		struct nouveau_object **pobject)
+{
+	struct nvaa_clock_priv *priv;
+	int ret;
+
+	ret = nouveau_clock_create(parent, engine, oclass, nvaa_domains, &priv);
+	*pobject = nv_object(priv);
+	if (ret)
+		return ret;
+
+	priv->base.read = nvaa_clock_read;
+	priv->base.calc = nvaa_clock_calc;
+	priv->base.prog = nvaa_clock_prog;
+	priv->base.tidy = nvaa_clock_tidy;
+	return 0;
+}
+
+struct nouveau_oclass *
+nvaa_clock_oclass = &(struct nouveau_oclass) {
+	.handle = NV_SUBDEV(CLOCK, 0xaa),
+	.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nvaa_clock_ctor,
+		.dtor = _nouveau_clock_dtor,
+		.init = _nouveau_clock_init,
+		.fini = _nouveau_clock_fini,
+	},
+};
diff --git a/drivers/gpu/drm/nouveau/dispnv04/overlay.c b/drivers/gpu/drm/nouveau/dispnv04/overlay.c
index 3618ac6b6316..32e7064b819b 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/overlay.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/overlay.c
@@ -58,8 +58,8 @@ struct nouveau_plane {
 };
 
 static uint32_t formats[] = {
-	DRM_FORMAT_NV12,
 	DRM_FORMAT_UYVY,
+	DRM_FORMAT_NV12,
 };
 
 /* Sine can be approximated with
@@ -99,13 +99,28 @@ nv10_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
 	struct nouveau_bo *cur = nv_plane->cur;
 	bool flip = nv_plane->flip;
-	int format = ALIGN(src_w * 4, 0x100);
 	int soff = NV_PCRTC0_SIZE * nv_crtc->index;
 	int soff2 = NV_PCRTC0_SIZE * !nv_crtc->index;
-	int ret;
+	int format, ret;
+
+	/* Source parameters given in 16.16 fixed point, ignore fractional. */
+	src_x >>= 16;
+	src_y >>= 16;
+	src_w >>= 16;
+	src_h >>= 16;
+
+	format = ALIGN(src_w * 4, 0x100);
 
 	if (format > 0xffff)
-		return -EINVAL;
+		return -ERANGE;
+
+	if (dev->chipset >= 0x30) {
+		if (crtc_w < (src_w >> 1) || crtc_h < (src_h >> 1))
+			return -ERANGE;
+	} else {
+		if (crtc_w < (src_w >> 3) || crtc_h < (src_h >> 3))
+			return -ERANGE;
+	}
 
 	ret = nouveau_bo_pin(nv_fb->nvbo, TTM_PL_FLAG_VRAM);
 	if (ret)
@@ -113,12 +128,6 @@ nv10_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 
 	nv_plane->cur = nv_fb->nvbo;
 
-	/* Source parameters given in 16.16 fixed point, ignore fractional. */
-	src_x = src_x >> 16;
-	src_y = src_y >> 16;
-	src_w = src_w >> 16;
-	src_h = src_h >> 16;
-
 	nv_mask(dev, NV_PCRTC_ENGINE_CTRL + soff, NV_CRTC_FSEL_OVERLAY, NV_CRTC_FSEL_OVERLAY);
 	nv_mask(dev, NV_PCRTC_ENGINE_CTRL + soff2, NV_CRTC_FSEL_OVERLAY, 0);
 
@@ -245,14 +254,25 @@ nv10_overlay_init(struct drm_device *device)
 {
 	struct nouveau_device *dev = nouveau_dev(device);
 	struct nouveau_plane *plane = kzalloc(sizeof(struct nouveau_plane), GFP_KERNEL);
+	int num_formats = ARRAY_SIZE(formats);
 	int ret;
 
 	if (!plane)
 		return;
 
+	switch (dev->chipset) {
+	case 0x10:
+	case 0x11:
+	case 0x15:
+	case 0x1a:
+	case 0x20:
+		num_formats = 1;
+		break;
+	}
+
 	ret = drm_plane_init(device, &plane->base, 3 /* both crtc's */,
 			     &nv10_plane_funcs,
-			     formats, ARRAY_SIZE(formats), false);
+			     formats, num_formats, false);
 	if (ret)
 		goto err;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 7809d92183c4..29c3efdfc7dd 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -608,6 +608,7 @@ nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 	fence = nouveau_fence_ref(new_bo->bo.sync_obj);
 	spin_unlock(&new_bo->bo.bdev->fence_lock);
 	ret = nouveau_fence_sync(fence, chan);
+	nouveau_fence_unref(&fence);
 	if (ret)
 		return ret;
 
@@ -701,7 +702,7 @@ nouveau_finish_page_flip(struct nouveau_channel *chan,
 
 	s = list_first_entry(&fctx->flip, struct nouveau_page_flip_state, head);
 	if (s->event)
-		drm_send_vblank_event(dev, -1, s->event);
+		drm_send_vblank_event(dev, s->crtc, s->event);
 
 	list_del(&s->head);
 	if (ps)
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 7a3759f1c41a..98a22e6e27a1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -858,6 +858,12 @@ static int nouveau_pmops_runtime_suspend(struct device *dev)
 	if (nouveau_runtime_pm == 0)
 		return -EINVAL;
 
+	/* are we optimus enabled? */
+	if (nouveau_runtime_pm == -1 && !nouveau_is_optimus() && !nouveau_is_v1_dsm()) {
+		DRM_DEBUG_DRIVER("failing to power off - not optimus\n");
+		return -EINVAL;
+	}
+
 	nv_debug_level(SILENT);
 	drm_kms_helper_poll_disable(drm_dev);
 	vga_switcheroo_set_dynamic_switch(pdev, VGA_SWITCHEROO_OFF);
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index f8e66c08b11a..4e384a2f99c3 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -1265,7 +1265,7 @@ nv50_crtc_gamma_set(struct drm_crtc *crtc, u16 *r, u16 *g, u16 *b,
 		    uint32_t start, uint32_t size)
 {
 	struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-	u32 end = max(start + size, (u32)256);
+	u32 end = min_t(u32, start + size, 256);
 	u32 i;
 
 	for (i = start; i < end; i++) {
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index 80a20120e625..b1970596a782 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -1196,7 +1196,9 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
 	} else if ((rdev->family == CHIP_TAHITI) ||
 		   (rdev->family == CHIP_PITCAIRN))
 		fb_format |= SI_GRPH_PIPE_CONFIG(SI_ADDR_SURF_P8_32x32_8x16);
-	else if (rdev->family == CHIP_VERDE)
+	else if ((rdev->family == CHIP_VERDE) ||
+		 (rdev->family == CHIP_OLAND) ||
+		 (rdev->family == CHIP_HAINAN)) /* for completeness.  HAINAN has no display hw */
 		fb_format |= SI_GRPH_PIPE_CONFIG(SI_ADDR_SURF_P4_8x16);
 
 	switch (radeon_crtc->crtc_id) {
diff --git a/drivers/gpu/drm/radeon/atombios_i2c.c b/drivers/gpu/drm/radeon/atombios_i2c.c
index 0652ee0a2098..f685035dbe39 100644
--- a/drivers/gpu/drm/radeon/atombios_i2c.c
+++ b/drivers/gpu/drm/radeon/atombios_i2c.c
@@ -44,7 +44,7 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
 	PROCESS_I2C_CHANNEL_TRANSACTION_PS_ALLOCATION args;
 	int index = GetIndexIntoMasterTable(COMMAND, ProcessI2cChannelTransaction);
 	unsigned char *base;
-	u16 out;
+	u16 out = cpu_to_le16(0);
 
 	memset(&args, 0, sizeof(args));
 
@@ -55,11 +55,14 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan,
 			DRM_ERROR("hw i2c: tried to write too many bytes (%d vs 3)\n", num);
 			return -EINVAL;
 		}
-		args.ucRegIndex = buf[0];
-		if (num > 1) {
+		if (buf == NULL)
+			args.ucRegIndex = 0;
+		else
+			args.ucRegIndex = buf[0];
+		if (num)
 			num--;
+		if (num)
 			memcpy(&out, &buf[1], num);
-		}
 		args.lpI2CDataOut = cpu_to_le16(out);
 	} else {
 		if (num > ATOM_MAX_HW_I2C_READ) {
@@ -96,14 +99,14 @@ int radeon_atom_hw_i2c_xfer(struct i2c_adapter *i2c_adap,
 	struct radeon_i2c_chan *i2c = i2c_get_adapdata(i2c_adap);
 	struct i2c_msg *p;
 	int i, remaining, current_count, buffer_offset, max_bytes, ret;
-	u8 buf = 0, flags;
+	u8 flags;
 
 	/* check for bus probe */
 	p = &msgs[0];
 	if ((num == 1) && (p->len == 0)) {
 		ret = radeon_process_i2c_ch(i2c,
 					    p->addr, HW_I2C_WRITE,
-					    &buf, 1);
+					    NULL, 0);
 		if (ret)
 			return ret;
 		else
diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index 0300727a4f70..d08b83c6267b 100644
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -458,7 +458,7 @@ int cik_copy_dma(struct radeon_device *rdev,
 		radeon_ring_write(ring, 0); /* src/dst endian swap */
 		radeon_ring_write(ring, src_offset & 0xffffffff);
 		radeon_ring_write(ring, upper_32_bits(src_offset) & 0xffffffff);
-		radeon_ring_write(ring, dst_offset & 0xfffffffc);
+		radeon_ring_write(ring, dst_offset & 0xffffffff);
 		radeon_ring_write(ring, upper_32_bits(dst_offset) & 0xffffffff);
 		src_offset += cur_size_in_bytes;
 		dst_offset += cur_size_in_bytes;
diff --git a/drivers/gpu/drm/radeon/dce6_afmt.c b/drivers/gpu/drm/radeon/dce6_afmt.c
index 009f46e0ce72..de86493cbc44 100644
--- a/drivers/gpu/drm/radeon/dce6_afmt.c
+++ b/drivers/gpu/drm/radeon/dce6_afmt.c
@@ -93,11 +93,13 @@ void dce6_afmt_select_pin(struct drm_encoder *encoder)
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	u32 offset = dig->afmt->offset;
+	u32 offset;
 
-	if (!dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->afmt->pin)
 		return;
 
+	offset = dig->afmt->offset;
+
 	WREG32(AFMT_AUDIO_SRC_CONTROL + offset,
 	       AFMT_AUDIO_SRC_SELECT(dig->afmt->pin->id));
 }
@@ -112,7 +114,7 @@ void dce6_afmt_write_latency_fields(struct drm_encoder *encoder,
 	struct radeon_connector *radeon_connector = NULL;
 	u32 tmp = 0, offset;
 
-	if (!dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->afmt->pin)
 		return;
 
 	offset = dig->afmt->pin->offset;
@@ -156,7 +158,7 @@ void dce6_afmt_write_speaker_allocation(struct drm_encoder *encoder)
 	u8 *sadb;
 	int sad_count;
 
-	if (!dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->afmt->pin)
 		return;
 
 	offset = dig->afmt->pin->offset;
@@ -217,7 +219,7 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder)
 		{ AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO },
 	};
 
-	if (!dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->afmt->pin)
 		return;
 
 	offset = dig->afmt->pin->offset;
diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c
index cdc003085a76..49c4d48f54d6 100644
--- a/drivers/gpu/drm/radeon/ni_dpm.c
+++ b/drivers/gpu/drm/radeon/ni_dpm.c
@@ -785,8 +785,8 @@ static void ni_apply_state_adjust_rules(struct radeon_device *rdev,
 	struct ni_ps *ps = ni_get_ps(rps);
 	struct radeon_clock_and_voltage_limits *max_limits;
 	bool disable_mclk_switching;
-	u32 mclk, sclk;
-	u16 vddc, vddci;
+	u32 mclk;
+	u16 vddci;
 	u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc;
 	int i;
 
@@ -839,24 +839,14 @@ static void ni_apply_state_adjust_rules(struct radeon_device *rdev,
 
 	/* XXX validate the min clocks required for display */
 
+	/* adjust low state */
 	if (disable_mclk_switching) {
-		mclk  = ps->performance_levels[ps->performance_level_count - 1].mclk;
-		sclk = ps->performance_levels[0].sclk;
-		vddc = ps->performance_levels[0].vddc;
-		vddci = ps->performance_levels[ps->performance_level_count - 1].vddci;
-	} else {
-		sclk = ps->performance_levels[0].sclk;
-		mclk = ps->performance_levels[0].mclk;
-		vddc = ps->performance_levels[0].vddc;
-		vddci = ps->performance_levels[0].vddci;
+		ps->performance_levels[0].mclk =
+			ps->performance_levels[ps->performance_level_count - 1].mclk;
+		ps->performance_levels[0].vddci =
+			ps->performance_levels[ps->performance_level_count - 1].vddci;
 	}
 
-	/* adjusted low state */
-	ps->performance_levels[0].sclk = sclk;
-	ps->performance_levels[0].mclk = mclk;
-	ps->performance_levels[0].vddc = vddc;
-	ps->performance_levels[0].vddci = vddci;
-
 	btc_skip_blacklist_clocks(rdev, max_limits->sclk, max_limits->mclk,
 				  &ps->performance_levels[0].sclk,
 				  &ps->performance_levels[0].mclk);
@@ -868,11 +858,15 @@ static void ni_apply_state_adjust_rules(struct radeon_device *rdev,
 			ps->performance_levels[i].vddc = ps->performance_levels[i - 1].vddc;
 	}
 
+	/* adjust remaining states */
 	if (disable_mclk_switching) {
 		mclk = ps->performance_levels[0].mclk;
+		vddci = ps->performance_levels[0].vddci;
 		for (i = 1; i < ps->performance_level_count; i++) {
 			if (mclk < ps->performance_levels[i].mclk)
 				mclk = ps->performance_levels[i].mclk;
+			if (vddci < ps->performance_levels[i].vddci)
+				vddci = ps->performance_levels[i].vddci;
 		}
 		for (i = 0; i < ps->performance_level_count; i++) {
 			ps->performance_levels[i].mclk = mclk;
diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c
index 4b89262f3f0e..b7d3ecba43e3 100644
--- a/drivers/gpu/drm/radeon/r600_hdmi.c
+++ b/drivers/gpu/drm/radeon/r600_hdmi.c
@@ -304,9 +304,9 @@ 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)) {
+	} else {
 		/* according to the reg specs, this should DCE3.2 only, but in
-		 * practice it seems to cover DCE3.0/3.1 as well.
+		 * practice it seems to cover DCE2.0/3.0/3.1 as well.
 		 */
 		if (dig->dig_encoder == 0) {
 			WREG32(DCCG_AUDIO_DTO0_PHASE, base_rate * 100);
@@ -317,10 +317,6 @@ void r600_audio_set_dto(struct drm_encoder *encoder, u32 clock)
 			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 DCE3.0/3.1 */
-		WREG32(AUDIO_DTO, AUDIO_DTO_PHASE(base_rate / 10) |
-		       AUDIO_DTO_MODULE(clock / 10));
 	}
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index ecf2a3960c07..b1f990d0eaa1 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -2710,10 +2710,10 @@ void radeon_vm_fence(struct radeon_device *rdev,
 		     struct radeon_vm *vm,
 		     struct radeon_fence *fence);
 uint64_t radeon_vm_map_gart(struct radeon_device *rdev, uint64_t addr);
-int radeon_vm_bo_update_pte(struct radeon_device *rdev,
-			    struct radeon_vm *vm,
-			    struct radeon_bo *bo,
-			    struct ttm_mem_reg *mem);
+int radeon_vm_bo_update(struct radeon_device *rdev,
+			struct radeon_vm *vm,
+			struct radeon_bo *bo,
+			struct ttm_mem_reg *mem);
 void radeon_vm_bo_invalidate(struct radeon_device *rdev,
 			     struct radeon_bo *bo);
 struct radeon_bo_va *radeon_vm_bo_find(struct radeon_vm *vm,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index e354ce94cdd1..c0425bb6223a 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -2021,7 +2021,7 @@ static struct radeon_asic ci_asic = {
 		.hdmi_setmode = &evergreen_hdmi_setmode,
 	},
 	.copy = {
-		.blit = NULL,
+		.blit = &cik_copy_cpdma,
 		.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
 		.dma = &cik_copy_dma,
 		.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
@@ -2122,7 +2122,7 @@ static struct radeon_asic kv_asic = {
 		.hdmi_setmode = &evergreen_hdmi_setmode,
 	},
 	.copy = {
-		.blit = NULL,
+		.blit = &cik_copy_cpdma,
 		.blit_ring_index = RADEON_RING_TYPE_GFX_INDEX,
 		.dma = &cik_copy_dma,
 		.dma_ring_index = R600_RING_TYPE_DMA_INDEX,
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index f79ee184ffd5..5c39bf7c3d88 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -2918,7 +2918,7 @@ int radeon_atom_get_memory_pll_dividers(struct radeon_device *rdev,
 			mpll_param->dll_speed = args.ucDllSpeed;
 			mpll_param->bwcntl = args.ucBWCntl;
 			mpll_param->vco_mode =
-				(args.ucPllCntlFlag & MPLL_CNTL_FLAG_VCO_MODE_MASK) ? 1 : 0;
+				(args.ucPllCntlFlag & MPLL_CNTL_FLAG_VCO_MODE_MASK);
 			mpll_param->yclk_sel =
 				(args.ucPllCntlFlag & MPLL_CNTL_FLAG_BYPASS_DQ_PLL) ? 1 : 0;
 			mpll_param->qdr =
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index f41594b2eeac..0b366169d64d 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -360,13 +360,13 @@ static int radeon_bo_vm_update_pte(struct radeon_cs_parser *parser,
 	struct radeon_bo *bo;
 	int r;
 
-	r = radeon_vm_bo_update_pte(rdev, vm, rdev->ring_tmp_bo.bo, &rdev->ring_tmp_bo.bo->tbo.mem);
+	r = radeon_vm_bo_update(rdev, vm, rdev->ring_tmp_bo.bo, &rdev->ring_tmp_bo.bo->tbo.mem);
 	if (r) {
 		return r;
 	}
 	list_for_each_entry(lobj, &parser->validated, tv.head) {
 		bo = lobj->bo;
-		r = radeon_vm_bo_update_pte(parser->rdev, vm, bo, &bo->tbo.mem);
+		r = radeon_vm_bo_update(parser->rdev, vm, bo, &bo->tbo.mem);
 		if (r) {
 			return r;
 		}
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 9f5ff28864f6..1958b36ad0e5 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -508,15 +508,6 @@ static const struct file_operations radeon_driver_kms_fops = {
 #endif
 };
 
-
-static void
-radeon_pci_shutdown(struct pci_dev *pdev)
-{
-	struct drm_device *dev = pci_get_drvdata(pdev);
-
-	radeon_driver_unload_kms(dev);
-}
-
 static struct drm_driver kms_driver = {
 	.driver_features =
 	    DRIVER_USE_AGP |
@@ -586,7 +577,6 @@ static struct pci_driver radeon_kms_pci_driver = {
 	.probe = radeon_pci_probe,
 	.remove = radeon_pci_remove,
 	.driver.pm = &radeon_pm_ops,
-	.shutdown = radeon_pci_shutdown,
 };
 
 static int __init radeon_init(void)
diff --git a/drivers/gpu/drm/radeon/radeon_drv.h b/drivers/gpu/drm/radeon/radeon_drv.h
index 543dcfae7e6f..00e0d449021c 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.h
+++ b/drivers/gpu/drm/radeon/radeon_drv.h
@@ -108,9 +108,10 @@
  * 1.31- Add support for num Z pipes from GET_PARAM
  * 1.32- fixes for rv740 setup
  * 1.33- Add r6xx/r7xx const buffer support
+ * 1.34- fix evergreen/cayman GS register
  */
 #define DRIVER_MAJOR		1
-#define DRIVER_MINOR		33
+#define DRIVER_MINOR		34
 #define DRIVER_PATCHLEVEL	0
 
 long radeon_drm_ioctl(struct file *filp,
diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c
index 3044e504f4ec..96e440061bdb 100644
--- a/drivers/gpu/drm/radeon/radeon_gart.c
+++ b/drivers/gpu/drm/radeon/radeon_gart.c
@@ -29,6 +29,7 @@
 #include <drm/radeon_drm.h>
 #include "radeon.h"
 #include "radeon_reg.h"
+#include "radeon_trace.h"
 
 /*
  * GART
@@ -737,6 +738,7 @@ struct radeon_fence *radeon_vm_grab_id(struct radeon_device *rdev,
 	for (i = 0; i < 2; ++i) {
 		if (choices[i]) {
 			vm->id = choices[i];
+			trace_radeon_vm_grab_id(vm->id, ring);
 			return rdev->vm_manager.active[choices[i]];
 		}
 	}
@@ -1116,7 +1118,7 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
 }
 
 /**
- * radeon_vm_bo_update_pte - map a bo into the vm page table
+ * radeon_vm_bo_update - map a bo into the vm page table
  *
  * @rdev: radeon_device pointer
  * @vm: requested vm
@@ -1128,10 +1130,10 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev,
  *
  * Object have to be reserved & global and local mutex must be locked!
  */
-int radeon_vm_bo_update_pte(struct radeon_device *rdev,
-			    struct radeon_vm *vm,
-			    struct radeon_bo *bo,
-			    struct ttm_mem_reg *mem)
+int radeon_vm_bo_update(struct radeon_device *rdev,
+			struct radeon_vm *vm,
+			struct radeon_bo *bo,
+			struct ttm_mem_reg *mem)
 {
 	struct radeon_ib ib;
 	struct radeon_bo_va *bo_va;
@@ -1176,6 +1178,8 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev,
 		bo_va->valid = false;
 	}
 
+	trace_radeon_vm_bo_update(bo_va);
+
 	nptes = radeon_bo_ngpu_pages(bo);
 
 	/* assume two extra pdes in case the mapping overlaps the borders */
@@ -1257,7 +1261,7 @@ int radeon_vm_bo_rmv(struct radeon_device *rdev,
 	mutex_lock(&rdev->vm_manager.lock);
 	mutex_lock(&bo_va->vm->mutex);
 	if (bo_va->soffset) {
-		r = radeon_vm_bo_update_pte(rdev, bo_va->vm, bo_va->bo, NULL);
+		r = radeon_vm_bo_update(rdev, bo_va->vm, bo_va->bo, NULL);
 	}
 	mutex_unlock(&rdev->vm_manager.lock);
 	list_del(&bo_va->vm_list);
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index d1385ccc672c..984097b907ef 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -537,8 +537,7 @@ static ssize_t radeon_hwmon_show_temp(struct device *dev,
 				      struct device_attribute *attr,
 				      char *buf)
 {
-	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct radeon_device *rdev = ddev->dev_private;
+	struct radeon_device *rdev = dev_get_drvdata(dev);
 	int temp;
 
 	if (rdev->asic->pm.get_temperature)
@@ -553,8 +552,7 @@ static ssize_t radeon_hwmon_show_temp_thresh(struct device *dev,
 					     struct device_attribute *attr,
 					     char *buf)
 {
-	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct radeon_device *rdev = ddev->dev_private;
+	struct radeon_device *rdev = dev_get_drvdata(dev);
 	int hyst = to_sensor_dev_attr(attr)->index;
 	int temp;
 
@@ -566,23 +564,14 @@ static ssize_t radeon_hwmon_show_temp_thresh(struct device *dev,
 	return snprintf(buf, PAGE_SIZE, "%d\n", temp);
 }
 
-static ssize_t radeon_hwmon_show_name(struct device *dev,
-				      struct device_attribute *attr,
-				      char *buf)
-{
-	return sprintf(buf, "radeon\n");
-}
-
 static SENSOR_DEVICE_ATTR(temp1_input, S_IRUGO, radeon_hwmon_show_temp, NULL, 0);
 static SENSOR_DEVICE_ATTR(temp1_crit, S_IRUGO, radeon_hwmon_show_temp_thresh, NULL, 0);
 static SENSOR_DEVICE_ATTR(temp1_crit_hyst, S_IRUGO, radeon_hwmon_show_temp_thresh, NULL, 1);
-static SENSOR_DEVICE_ATTR(name, S_IRUGO, radeon_hwmon_show_name, NULL, 0);
 
 static struct attribute *hwmon_attributes[] = {
 	&sensor_dev_attr_temp1_input.dev_attr.attr,
 	&sensor_dev_attr_temp1_crit.dev_attr.attr,
 	&sensor_dev_attr_temp1_crit_hyst.dev_attr.attr,
-	&sensor_dev_attr_name.dev_attr.attr,
 	NULL
 };
 
@@ -590,8 +579,7 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 					struct attribute *attr, int index)
 {
 	struct device *dev = container_of(kobj, struct device, kobj);
-	struct drm_device *ddev = dev_get_drvdata(dev);
-	struct radeon_device *rdev = ddev->dev_private;
+	struct radeon_device *rdev = dev_get_drvdata(dev);
 
 	/* Skip limit attributes if DPM is not enabled */
 	if (rdev->pm.pm_method != PM_METHOD_DPM &&
@@ -607,11 +595,15 @@ static const struct attribute_group hwmon_attrgroup = {
 	.is_visible = hwmon_attributes_visible,
 };
 
+static const struct attribute_group *hwmon_groups[] = {
+	&hwmon_attrgroup,
+	NULL
+};
+
 static int radeon_hwmon_init(struct radeon_device *rdev)
 {
 	int err = 0;
-
-	rdev->pm.int_hwmon_dev = NULL;
+	struct device *hwmon_dev;
 
 	switch (rdev->pm.int_thermal_type) {
 	case THERMAL_TYPE_RV6XX:
@@ -624,20 +616,13 @@ static int radeon_hwmon_init(struct radeon_device *rdev)
 	case THERMAL_TYPE_KV:
 		if (rdev->asic->pm.get_temperature == NULL)
 			return err;
-		rdev->pm.int_hwmon_dev = hwmon_device_register(rdev->dev);
-		if (IS_ERR(rdev->pm.int_hwmon_dev)) {
-			err = PTR_ERR(rdev->pm.int_hwmon_dev);
+		hwmon_dev = hwmon_device_register_with_groups(rdev->dev,
+							      "radeon", rdev,
+							      hwmon_groups);
+		if (IS_ERR(hwmon_dev)) {
+			err = PTR_ERR(hwmon_dev);
 			dev_err(rdev->dev,
 				"Unable to register hwmon device: %d\n", err);
-			break;
-		}
-		dev_set_drvdata(rdev->pm.int_hwmon_dev, rdev->ddev);
-		err = sysfs_create_group(&rdev->pm.int_hwmon_dev->kobj,
-					 &hwmon_attrgroup);
-		if (err) {
-			dev_err(rdev->dev,
-				"Unable to create hwmon sysfs file: %d\n", err);
-			hwmon_device_unregister(rdev->dev);
 		}
 		break;
 	default:
@@ -647,14 +632,6 @@ static int radeon_hwmon_init(struct radeon_device *rdev)
 	return err;
 }
 
-static void radeon_hwmon_fini(struct radeon_device *rdev)
-{
-	if (rdev->pm.int_hwmon_dev) {
-		sysfs_remove_group(&rdev->pm.int_hwmon_dev->kobj, &hwmon_attrgroup);
-		hwmon_device_unregister(rdev->pm.int_hwmon_dev);
-	}
-}
-
 static void radeon_dpm_thermal_work_handler(struct work_struct *work)
 {
 	struct radeon_device *rdev =
@@ -1337,8 +1314,6 @@ static void radeon_pm_fini_old(struct radeon_device *rdev)
 
 	if (rdev->pm.power_state)
 		kfree(rdev->pm.power_state);
-
-	radeon_hwmon_fini(rdev);
 }
 
 static void radeon_pm_fini_dpm(struct radeon_device *rdev)
@@ -1358,8 +1333,6 @@ static void radeon_pm_fini_dpm(struct radeon_device *rdev)
 
 	if (rdev->pm.power_state)
 		kfree(rdev->pm.power_state);
-
-	radeon_hwmon_fini(rdev);
 }
 
 void radeon_pm_fini(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/radeon_trace.h b/drivers/gpu/drm/radeon/radeon_trace.h
index 9f0e18172b6e..0473257d4078 100644
--- a/drivers/gpu/drm/radeon/radeon_trace.h
+++ b/drivers/gpu/drm/radeon/radeon_trace.h
@@ -47,6 +47,39 @@ TRACE_EVENT(radeon_cs,
 		      __entry->fences)
 );
 
+TRACE_EVENT(radeon_vm_grab_id,
+	    TP_PROTO(unsigned vmid, int ring),
+	    TP_ARGS(vmid, ring),
+	    TP_STRUCT__entry(
+			     __field(u32, vmid)
+			     __field(u32, ring)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->vmid = vmid;
+			   __entry->ring = ring;
+			   ),
+	    TP_printk("vmid=%u, ring=%u", __entry->vmid, __entry->ring)
+);
+
+TRACE_EVENT(radeon_vm_bo_update,
+	    TP_PROTO(struct radeon_bo_va *bo_va),
+	    TP_ARGS(bo_va),
+	    TP_STRUCT__entry(
+			     __field(u64, soffset)
+			     __field(u64, eoffset)
+			     __field(u32, flags)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->soffset = bo_va->soffset;
+			   __entry->eoffset = bo_va->eoffset;
+			   __entry->flags = bo_va->flags;
+			   ),
+	    TP_printk("soffs=%010llx, eoffs=%010llx, flags=%08x",
+		      __entry->soffset, __entry->eoffset, __entry->flags)
+);
+
 TRACE_EVENT(radeon_vm_set_page,
 	    TP_PROTO(uint64_t pe, uint64_t addr, unsigned count,
 		     uint32_t incr, uint32_t flags),
diff --git a/drivers/gpu/drm/radeon/reg_srcs/cayman b/drivers/gpu/drm/radeon/reg_srcs/cayman
index a072fa8c46b0..d46b58d078aa 100644
--- a/drivers/gpu/drm/radeon/reg_srcs/cayman
+++ b/drivers/gpu/drm/radeon/reg_srcs/cayman
@@ -21,7 +21,7 @@ cayman 0x9400
 0x000089AC VGT_COMPUTE_THREAD_GOURP_SIZE
 0x000089B0 VGT_HS_OFFCHIP_PARAM
 0x00008A14 PA_CL_ENHANCE
-0x00008A60 PA_SC_LINE_STIPPLE_VALUE
+0x00008A60 PA_SU_LINE_STIPPLE_VALUE
 0x00008B10 PA_SC_LINE_STIPPLE_STATE
 0x00008BF0 PA_SC_ENHANCE
 0x00008D8C SQ_DYN_GPR_CNTL_PS_FLUSH_REQ
@@ -532,7 +532,7 @@ cayman 0x9400
 0x00028B84 PA_SU_POLY_OFFSET_FRONT_OFFSET
 0x00028B88 PA_SU_POLY_OFFSET_BACK_SCALE
 0x00028B8C PA_SU_POLY_OFFSET_BACK_OFFSET
-0x00028B74 VGT_GS_INSTANCE_CNT
+0x00028B90 VGT_GS_INSTANCE_CNT
 0x00028BD4 PA_SC_CENTROID_PRIORITY_0
 0x00028BD8 PA_SC_CENTROID_PRIORITY_1
 0x00028BDC PA_SC_LINE_CNTL
diff --git a/drivers/gpu/drm/radeon/reg_srcs/evergreen b/drivers/gpu/drm/radeon/reg_srcs/evergreen
index b912a37689bf..57745c8761c8 100644
--- a/drivers/gpu/drm/radeon/reg_srcs/evergreen
+++ b/drivers/gpu/drm/radeon/reg_srcs/evergreen
@@ -22,7 +22,7 @@ evergreen 0x9400
 0x000089A4 VGT_COMPUTE_START_Z
 0x000089AC VGT_COMPUTE_THREAD_GOURP_SIZE
 0x00008A14 PA_CL_ENHANCE
-0x00008A60 PA_SC_LINE_STIPPLE_VALUE
+0x00008A60 PA_SU_LINE_STIPPLE_VALUE
 0x00008B10 PA_SC_LINE_STIPPLE_STATE
 0x00008BF0 PA_SC_ENHANCE
 0x00008D8C SQ_DYN_GPR_CNTL_PS_FLUSH_REQ
@@ -545,7 +545,7 @@ evergreen 0x9400
 0x00028B84 PA_SU_POLY_OFFSET_FRONT_OFFSET
 0x00028B88 PA_SU_POLY_OFFSET_BACK_SCALE
 0x00028B8C PA_SU_POLY_OFFSET_BACK_OFFSET
-0x00028B74 VGT_GS_INSTANCE_CNT
+0x00028B90 VGT_GS_INSTANCE_CNT
 0x00028C00 PA_SC_LINE_CNTL
 0x00028C08 PA_SU_VTX_CNTL
 0x00028C0C PA_CL_GB_VERT_CLIP_ADJ
diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c
index 1c560629575a..e7dab069cccf 100644
--- a/drivers/gpu/drm/radeon/rs690.c
+++ b/drivers/gpu/drm/radeon/rs690.c
@@ -162,6 +162,16 @@ static void rs690_mc_init(struct radeon_device *rdev)
 	base = RREG32_MC(R_000100_MCCFG_FB_LOCATION);
 	base = G_000100_MC_FB_START(base) << 16;
 	rdev->mc.igp_sideport_enabled = radeon_atombios_sideport_present(rdev);
+	/* Some boards seem to be configured for 128MB of sideport memory,
+	 * but really only have 64MB.  Just skip the sideport and use
+	 * UMA memory.
+	 */
+	if (rdev->mc.igp_sideport_enabled &&
+	    (rdev->mc.real_vram_size == (384 * 1024 * 1024))) {
+		base += 128 * 1024 * 1024;
+		rdev->mc.real_vram_size -= 128 * 1024 * 1024;
+		rdev->mc.mc_vram_size = rdev->mc.real_vram_size;
+	}
 
 	/* Use K8 direct mapping for fast fb access. */ 
 	rdev->fastfb_working = false;
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 6a64ccaa0695..a36736dab5e0 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -3882,8 +3882,15 @@ static int si_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) * 1024ULL * 1024ULL;
-	rdev->mc.real_vram_size = RREG32(CONFIG_MEMSIZE) * 1024ULL * 1024ULL;
+	tmp = RREG32(CONFIG_MEMSIZE);
+	/* some boards may have garbage in the upper 16 bits */
+	if (tmp & 0xffff0000) {
+		DRM_INFO("Probable bad vram size: 0x%08x\n", tmp);
+		if (tmp & 0xffff)
+			tmp &= 0xffff;
+	}
+	rdev->mc.mc_vram_size = tmp * 1024ULL * 1024ULL;
+	rdev->mc.real_vram_size = rdev->mc.mc_vram_size;
 	rdev->mc.visible_vram_size = rdev->mc.aper_size;
 	si_vram_gtt_location(rdev, &rdev->mc);
 	radeon_update_bandwidth_info(rdev);
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 28e178137718..07eba596d458 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -135,11 +135,11 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 	unsigned int num_relocs = args->num_relocs;
 	unsigned int num_waitchks = args->num_waitchks;
 	struct drm_tegra_cmdbuf __user *cmdbufs =
-		(void * __user)(uintptr_t)args->cmdbufs;
+		(void __user *)(uintptr_t)args->cmdbufs;
 	struct drm_tegra_reloc __user *relocs =
-		(void * __user)(uintptr_t)args->relocs;
+		(void __user *)(uintptr_t)args->relocs;
 	struct drm_tegra_waitchk __user *waitchks =
-		(void * __user)(uintptr_t)args->waitchks;
+		(void __user *)(uintptr_t)args->waitchks;
 	struct drm_tegra_syncpt syncpt;
 	struct host1x_job *job;
 	int err;
@@ -163,9 +163,10 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 		struct drm_tegra_cmdbuf cmdbuf;
 		struct host1x_bo *bo;
 
-		err = copy_from_user(&cmdbuf, cmdbufs, sizeof(cmdbuf));
-		if (err)
+		if (copy_from_user(&cmdbuf, cmdbufs, sizeof(cmdbuf))) {
+			err = -EFAULT;
 			goto fail;
+		}
 
 		bo = host1x_bo_lookup(drm, file, cmdbuf.handle);
 		if (!bo) {
@@ -178,10 +179,11 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 		cmdbufs++;
 	}
 
-	err = copy_from_user(job->relocarray, relocs,
-			     sizeof(*relocs) * num_relocs);
-	if (err)
+	if (copy_from_user(job->relocarray, relocs,
+			   sizeof(*relocs) * num_relocs)) {
+		err = -EFAULT;
 		goto fail;
+	}
 
 	while (num_relocs--) {
 		struct host1x_reloc *reloc = &job->relocarray[num_relocs];
@@ -199,15 +201,17 @@ int tegra_drm_submit(struct tegra_drm_context *context,
 		}
 	}
 
-	err = copy_from_user(job->waitchk, waitchks,
-			     sizeof(*waitchks) * num_waitchks);
-	if (err)
+	if (copy_from_user(job->waitchk, waitchks,
+			   sizeof(*waitchks) * num_waitchks)) {
+		err = -EFAULT;
 		goto fail;
+	}
 
-	err = copy_from_user(&syncpt, (void * __user)(uintptr_t)args->syncpts,
-			     sizeof(syncpt));
-	if (err)
+	if (copy_from_user(&syncpt, (void __user *)(uintptr_t)args->syncpts,
+			   sizeof(syncpt))) {
+		err = -EFAULT;
 		goto fail;
+	}
 
 	job->is_addr_reg = context->client->ops->is_addr_reg;
 	job->syncpt_incrs = syncpt.incrs;
@@ -573,7 +577,7 @@ static void tegra_debugfs_cleanup(struct drm_minor *minor)
 }
 #endif
 
-struct drm_driver tegra_drm_driver = {
+static struct drm_driver tegra_drm_driver = {
 	.driver_features = DRIVER_MODESET | DRIVER_GEM,
 	.load = tegra_drm_load,
 	.unload = tegra_drm_unload,
diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h
index fdfe259ed7f8..7da0b923131f 100644
--- a/drivers/gpu/drm/tegra/drm.h
+++ b/drivers/gpu/drm/tegra/drm.h
@@ -116,7 +116,7 @@ host1x_client_to_dc(struct host1x_client *client)
 
 static inline struct tegra_dc *to_tegra_dc(struct drm_crtc *crtc)
 {
-	return container_of(crtc, struct tegra_dc, base);
+	return crtc ? container_of(crtc, struct tegra_dc, base) : NULL;
 }
 
 static inline void tegra_dc_writel(struct tegra_dc *dc, unsigned long value,
diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c
index 490f7719e317..a3835e7de184 100644
--- a/drivers/gpu/drm/tegra/fb.c
+++ b/drivers/gpu/drm/tegra/fb.c
@@ -247,7 +247,7 @@ static int tegra_fbdev_probe(struct drm_fb_helper *helper,
 		 info->var.yoffset * fb->pitches[0];
 
 	drm->mode_config.fb_base = (resource_size_t)bo->paddr;
-	info->screen_base = bo->vaddr + offset;
+	info->screen_base = (void __iomem *)bo->vaddr + offset;
 	info->screen_size = size;
 	info->fix.smem_start = (unsigned long)(bo->paddr + offset);
 	info->fix.smem_len = size;
diff --git a/drivers/gpu/drm/tegra/rgb.c b/drivers/gpu/drm/tegra/rgb.c
index ba47ca4fb880..3b29018913a5 100644
--- a/drivers/gpu/drm/tegra/rgb.c
+++ b/drivers/gpu/drm/tegra/rgb.c
@@ -14,6 +14,8 @@
 
 struct tegra_rgb {
 	struct tegra_output output;
+	struct tegra_dc *dc;
+
 	struct clk *clk_parent;
 	struct clk *clk;
 };
@@ -84,18 +86,18 @@ static void tegra_dc_write_regs(struct tegra_dc *dc,
 
 static int tegra_output_rgb_enable(struct tegra_output *output)
 {
-	struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc);
+	struct tegra_rgb *rgb = to_rgb(output);
 
-	tegra_dc_write_regs(dc, rgb_enable, ARRAY_SIZE(rgb_enable));
+	tegra_dc_write_regs(rgb->dc, rgb_enable, ARRAY_SIZE(rgb_enable));
 
 	return 0;
 }
 
 static int tegra_output_rgb_disable(struct tegra_output *output)
 {
-	struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc);
+	struct tegra_rgb *rgb = to_rgb(output);
 
-	tegra_dc_write_regs(dc, rgb_disable, ARRAY_SIZE(rgb_disable));
+	tegra_dc_write_regs(rgb->dc, rgb_disable, ARRAY_SIZE(rgb_disable));
 
 	return 0;
 }
@@ -146,6 +148,7 @@ int tegra_dc_rgb_probe(struct tegra_dc *dc)
 
 	rgb->output.dev = dc->dev;
 	rgb->output.of_node = np;
+	rgb->dc = dc;
 
 	err = tegra_output_probe(&rgb->output);
 	if (err < 0)
diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
index b249ab9b1eb2..6440eeac22d2 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
@@ -169,9 +169,9 @@ static int ttm_bo_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	}
 
 	page_offset = ((address - vma->vm_start) >> PAGE_SHIFT) +
-	    drm_vma_node_start(&bo->vma_node) - vma->vm_pgoff;
-	page_last = vma_pages(vma) +
-	    drm_vma_node_start(&bo->vma_node) - vma->vm_pgoff;
+		vma->vm_pgoff - drm_vma_node_start(&bo->vma_node);
+	page_last = vma_pages(vma) + vma->vm_pgoff -
+		drm_vma_node_start(&bo->vma_node);
 
 	if (unlikely(page_offset >= bo->num_pages)) {
 		retval = VM_FAULT_SIGBUS;
diff --git a/drivers/gpu/drm/udl/udl_gem.c b/drivers/gpu/drm/udl/udl_gem.c
index 24ffbe990736..8d67b943ac05 100644
--- a/drivers/gpu/drm/udl/udl_gem.c
+++ b/drivers/gpu/drm/udl/udl_gem.c
@@ -125,6 +125,12 @@ static int udl_gem_get_pages(struct udl_gem_object *obj, gfp_t gfpmask)
 
 static void udl_gem_put_pages(struct udl_gem_object *obj)
 {
+	if (obj->base.import_attach) {
+		drm_free_large(obj->pages);
+		obj->pages = NULL;
+		return;
+	}
+
 	drm_gem_put_pages(&obj->base, obj->pages, false, false);
 	obj->pages = NULL;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c b/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c
index 7776e6f0aef6..0489c6152482 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_buffer.c
@@ -150,6 +150,8 @@ struct vmw_ttm_tt {
 	bool mapped;
 };
 
+const size_t vmw_tt_size = sizeof(struct vmw_ttm_tt);
+
 /**
  * Helper functions to advance a struct vmw_piter iterator.
  *
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index db85985c7086..20890ad8408b 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -615,6 +615,7 @@ extern int vmw_mmap(struct file *filp, struct vm_area_struct *vma);
  * TTM buffer object driver - vmwgfx_buffer.c
  */
 
+extern const size_t vmw_tt_size;
 extern struct ttm_placement vmw_vram_placement;
 extern struct ttm_placement vmw_vram_ne_placement;
 extern struct ttm_placement vmw_vram_sys_placement;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
index a51f48e3e917..45d5b5ab6ca9 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
@@ -68,6 +68,9 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data,
 				  SVGA_FIFO_3D_HWVERSION));
 		break;
 	}
+	case DRM_VMW_PARAM_MAX_SURF_MEMORY:
+		param->value = dev_priv->memory_size;
+		break;
 	default:
 		DRM_ERROR("Illegal vmwgfx get param request: %d\n",
 			  param->param);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index ecb3d867b426..03f1c2038631 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -75,6 +75,7 @@ void vmw_display_unit_cleanup(struct vmw_display_unit *du)
 		vmw_surface_unreference(&du->cursor_surface);
 	if (du->cursor_dmabuf)
 		vmw_dmabuf_unreference(&du->cursor_dmabuf);
+	drm_sysfs_connector_remove(&du->connector);
 	drm_crtc_cleanup(&du->crtc);
 	drm_encoder_cleanup(&du->encoder);
 	drm_connector_cleanup(&du->connector);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
index 79f7e8e60529..a055a26819c2 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
@@ -260,6 +260,7 @@ static int vmw_ldu_crtc_set_config(struct drm_mode_set *set)
 		connector->encoder = NULL;
 		encoder->crtc = NULL;
 		crtc->fb = NULL;
+		crtc->enabled = false;
 
 		vmw_ldu_del_active(dev_priv, ldu);
 
@@ -285,6 +286,7 @@ static int vmw_ldu_crtc_set_config(struct drm_mode_set *set)
 	crtc->x = set->x;
 	crtc->y = set->y;
 	crtc->mode = *mode;
+	crtc->enabled = true;
 
 	vmw_ldu_add_active(dev_priv, ldu, vfb);
 
@@ -369,6 +371,8 @@ static int vmw_ldu_init(struct vmw_private *dev_priv, unsigned unit)
 	encoder->possible_crtcs = (1 << unit);
 	encoder->possible_clones = 0;
 
+	(void) drm_sysfs_connector_add(connector);
+
 	drm_crtc_init(dev, crtc, &vmw_legacy_crtc_funcs);
 
 	drm_mode_crtc_set_gamma_size(crtc, 256);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
index efe2b74c5eb1..9b5ea2ac7ddf 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
@@ -352,6 +352,38 @@ int vmw_user_lookup_handle(struct vmw_private *dev_priv,
 /**
  * Buffer management.
  */
+
+/**
+ * vmw_dmabuf_acc_size - Calculate the pinned memory usage of buffers
+ *
+ * @dev_priv: Pointer to a struct vmw_private identifying the device.
+ * @size: The requested buffer size.
+ * @user: Whether this is an ordinary dma buffer or a user dma buffer.
+ */
+static size_t vmw_dmabuf_acc_size(struct vmw_private *dev_priv, size_t size,
+				  bool user)
+{
+	static size_t struct_size, user_struct_size;
+	size_t num_pages = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	size_t page_array_size = ttm_round_pot(num_pages * sizeof(void *));
+
+	if (unlikely(struct_size == 0)) {
+		size_t backend_size = ttm_round_pot(vmw_tt_size);
+
+		struct_size = backend_size +
+			ttm_round_pot(sizeof(struct vmw_dma_buffer));
+		user_struct_size = backend_size +
+			ttm_round_pot(sizeof(struct vmw_user_dma_buffer));
+	}
+
+	if (dev_priv->map_mode == vmw_dma_alloc_coherent)
+		page_array_size +=
+			ttm_round_pot(num_pages * sizeof(dma_addr_t));
+
+	return ((user) ? user_struct_size : struct_size) +
+		page_array_size;
+}
+
 void vmw_dmabuf_bo_free(struct ttm_buffer_object *bo)
 {
 	struct vmw_dma_buffer *vmw_bo = vmw_dma_buffer(bo);
@@ -359,6 +391,13 @@ void vmw_dmabuf_bo_free(struct ttm_buffer_object *bo)
 	kfree(vmw_bo);
 }
 
+static void vmw_user_dmabuf_destroy(struct ttm_buffer_object *bo)
+{
+	struct vmw_user_dma_buffer *vmw_user_bo = vmw_user_dma_buffer(bo);
+
+	ttm_prime_object_kfree(vmw_user_bo, prime);
+}
+
 int vmw_dmabuf_init(struct vmw_private *dev_priv,
 		    struct vmw_dma_buffer *vmw_bo,
 		    size_t size, struct ttm_placement *placement,
@@ -368,28 +407,23 @@ int vmw_dmabuf_init(struct vmw_private *dev_priv,
 	struct ttm_bo_device *bdev = &dev_priv->bdev;
 	size_t acc_size;
 	int ret;
+	bool user = (bo_free == &vmw_user_dmabuf_destroy);
 
-	BUG_ON(!bo_free);
+	BUG_ON(!bo_free && (!user && (bo_free != vmw_dmabuf_bo_free)));
 
-	acc_size = ttm_bo_acc_size(bdev, size, sizeof(struct vmw_dma_buffer));
+	acc_size = vmw_dmabuf_acc_size(dev_priv, size, user);
 	memset(vmw_bo, 0, sizeof(*vmw_bo));
 
 	INIT_LIST_HEAD(&vmw_bo->res_list);
 
 	ret = ttm_bo_init(bdev, &vmw_bo->base, size,
-			  ttm_bo_type_device, placement,
+			  (user) ? ttm_bo_type_device :
+			  ttm_bo_type_kernel, placement,
 			  0, interruptible,
 			  NULL, acc_size, NULL, bo_free);
 	return ret;
 }
 
-static void vmw_user_dmabuf_destroy(struct ttm_buffer_object *bo)
-{
-	struct vmw_user_dma_buffer *vmw_user_bo = vmw_user_dma_buffer(bo);
-
-	ttm_prime_object_kfree(vmw_user_bo, prime);
-}
-
 static void vmw_user_dmabuf_release(struct ttm_base_object **p_base)
 {
 	struct vmw_user_dma_buffer *vmw_user_bo;
@@ -781,54 +815,55 @@ err_ref:
 }
 
 
+/**
+ * vmw_dumb_create - Create a dumb kms buffer
+ *
+ * @file_priv: Pointer to a struct drm_file identifying the caller.
+ * @dev: Pointer to the drm device.
+ * @args: Pointer to a struct drm_mode_create_dumb structure
+ *
+ * This is a driver callback for the core drm create_dumb functionality.
+ * Note that this is very similar to the vmw_dmabuf_alloc ioctl, except
+ * that the arguments have a different format.
+ */
 int vmw_dumb_create(struct drm_file *file_priv,
 		    struct drm_device *dev,
 		    struct drm_mode_create_dumb *args)
 {
 	struct vmw_private *dev_priv = vmw_priv(dev);
 	struct vmw_master *vmaster = vmw_master(file_priv->master);
-	struct vmw_user_dma_buffer *vmw_user_bo;
-	struct ttm_buffer_object *tmp;
+	struct vmw_dma_buffer *dma_buf;
 	int ret;
 
 	args->pitch = args->width * ((args->bpp + 7) / 8);
 	args->size = args->pitch * args->height;
 
-	vmw_user_bo = kzalloc(sizeof(*vmw_user_bo), GFP_KERNEL);
-	if (vmw_user_bo == NULL)
-		return -ENOMEM;
-
 	ret = ttm_read_lock(&vmaster->lock, true);
-	if (ret != 0) {
-		kfree(vmw_user_bo);
+	if (unlikely(ret != 0))
 		return ret;
-	}
 
-	ret = vmw_dmabuf_init(dev_priv, &vmw_user_bo->dma, args->size,
-			      &vmw_vram_sys_placement, true,
-			      &vmw_user_dmabuf_destroy);
-	if (ret != 0)
-		goto out_no_dmabuf;
-
-	tmp = ttm_bo_reference(&vmw_user_bo->dma.base);
-	ret = ttm_prime_object_init(vmw_fpriv(file_priv)->tfile,
-				    args->size,
-				    &vmw_user_bo->prime,
-				    false,
-				    ttm_buffer_type,
-				    &vmw_user_dmabuf_release, NULL);
+	ret = vmw_user_dmabuf_alloc(dev_priv, vmw_fpriv(file_priv)->tfile,
+				    args->size, false, &args->handle,
+				    &dma_buf);
 	if (unlikely(ret != 0))
-		goto out_no_base_object;
-
-	args->handle = vmw_user_bo->prime.base.hash.key;
+		goto out_no_dmabuf;
 
-out_no_base_object:
-	ttm_bo_unref(&tmp);
+	vmw_dmabuf_unreference(&dma_buf);
 out_no_dmabuf:
 	ttm_read_unlock(&vmaster->lock);
 	return ret;
 }
 
+/**
+ * vmw_dumb_map_offset - Return the address space offset of a dumb buffer
+ *
+ * @file_priv: Pointer to a struct drm_file identifying the caller.
+ * @dev: Pointer to the drm device.
+ * @handle: Handle identifying the dumb buffer.
+ * @offset: The address space offset returned.
+ *
+ * This is a driver callback for the core drm dumb_map_offset functionality.
+ */
 int vmw_dumb_map_offset(struct drm_file *file_priv,
 			struct drm_device *dev, uint32_t handle,
 			uint64_t *offset)
@@ -846,6 +881,15 @@ int vmw_dumb_map_offset(struct drm_file *file_priv,
 	return 0;
 }
 
+/**
+ * vmw_dumb_destroy - Destroy a dumb boffer
+ *
+ * @file_priv: Pointer to a struct drm_file identifying the caller.
+ * @dev: Pointer to the drm device.
+ * @handle: Handle identifying the dumb buffer.
+ *
+ * This is a driver callback for the core drm dumb_destroy functionality.
+ */
 int vmw_dumb_destroy(struct drm_file *file_priv,
 		     struct drm_device *dev,
 		     uint32_t handle)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
index 26387c3d5a21..22406c8651ea 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
@@ -310,6 +310,7 @@ static int vmw_sou_crtc_set_config(struct drm_mode_set *set)
 		crtc->fb = NULL;
 		crtc->x = 0;
 		crtc->y = 0;
+		crtc->enabled = false;
 
 		vmw_sou_del_active(dev_priv, sou);
 
@@ -370,6 +371,7 @@ static int vmw_sou_crtc_set_config(struct drm_mode_set *set)
 		crtc->fb = NULL;
 		crtc->x = 0;
 		crtc->y = 0;
+		crtc->enabled = false;
 
 		return ret;
 	}
@@ -382,6 +384,7 @@ static int vmw_sou_crtc_set_config(struct drm_mode_set *set)
 	crtc->fb = fb;
 	crtc->x = set->x;
 	crtc->y = set->y;
+	crtc->enabled = true;
 
 	return 0;
 }
@@ -464,6 +467,8 @@ static int vmw_sou_init(struct vmw_private *dev_priv, unsigned unit)
 	encoder->possible_crtcs = (1 << unit);
 	encoder->possible_clones = 0;
 
+	(void) drm_sysfs_connector_add(connector);
+
 	drm_crtc_init(dev, crtc, &vmw_screen_object_crtc_funcs);
 
 	drm_mode_crtc_set_gamma_size(crtc, 256);