Merge tag 'v3.2-rc6' of /home/airlied/devel/kernel/linux-2.6 into drm-core-next

Merge in the upstream tree to bring in the mainline fixes.

Conflicts:
	drivers/gpu/drm/exynos/exynos_drm_fbdev.c
	drivers/gpu/drm/nouveau/nouveau_sgdma.c

85 files changed, 2112 insertions, 994 deletions

diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 45079046b6d5..2418429a9836 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -9,7 +9,6 @@ menuconfig DRM
 	depends on (AGP || AGP=n) && !EMULATED_CMPXCHG && MMU
 	select I2C
 	select I2C_ALGOBIT
-	select SLOW_WORK
 	help
 	  Kernel-level support for the Direct Rendering Infrastructure (DRI)
 	  introduced in XFree86 4.0. If you say Y here, you need to select
@@ -96,6 +95,7 @@ config DRM_I915
 	select FB_CFB_IMAGEBLIT
 	# i915 depends on ACPI_VIDEO when ACPI is enabled
 	# but for select to work, need to select ACPI_VIDEO's dependencies, ick
+	select BACKLIGHT_LCD_SUPPORT if ACPI
 	select BACKLIGHT_CLASS_DEVICE if ACPI
 	select VIDEO_OUTPUT_CONTROL if ACPI
 	select INPUT if ACPI
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 0d1faa72e1ff..f259a2563204 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -2340,6 +2340,10 @@ int drm_mode_dirtyfb_ioctl(struct drm_device *dev,
 	}
 
 	if (num_clips && clips_ptr) {
+		if (num_clips < 0 || num_clips > DRM_MODE_FB_DIRTY_MAX_CLIPS) {
+			ret = -EINVAL;
+			goto out_err1;
+		}
 		clips = kzalloc(num_clips * sizeof(*clips), GFP_KERNEL);
 		if (!clips) {
 			ret = -ENOMEM;
@@ -2585,8 +2589,10 @@ struct drm_property *drm_property_create(struct drm_device *dev, int flags,
 	property->num_values = num_values;
 	INIT_LIST_HEAD(&property->enum_blob_list);
 
-	if (name)
+	if (name) {
 		strncpy(property->name, name, DRM_PROP_NAME_LEN);
+		property->name[DRM_PROP_NAME_LEN-1] = '\0';
+	}
 
 	list_add_tail(&property->head, &dev->mode_config.property_list);
 	return property;
diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index ccbdc0b5854c..42f86e71479a 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -457,6 +457,30 @@ done:
 EXPORT_SYMBOL(drm_crtc_helper_set_mode);
 
 
+static int
+drm_crtc_helper_disable(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_connector *connector;
+	struct drm_encoder *encoder;
+
+	/* Decouple all encoders and their attached connectors from this crtc */
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
+		if (encoder->crtc != crtc)
+			continue;
+
+		list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+			if (connector->encoder != encoder)
+				continue;
+
+			connector->encoder = NULL;
+		}
+	}
+
+	drm_helper_disable_unused_functions(dev);
+	return 0;
+}
+
 /**
  * drm_crtc_helper_set_config - set a new config from userspace
  * @crtc: CRTC to setup
@@ -485,6 +509,7 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 	struct drm_connector *save_connectors, *connector;
 	int count = 0, ro, fail = 0;
 	struct drm_crtc_helper_funcs *crtc_funcs;
+	struct drm_mode_set save_set;
 	int ret = 0;
 	int i;
 
@@ -510,8 +535,7 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 				(int)set->num_connectors, set->x, set->y);
 	} else {
 		DRM_DEBUG_KMS("[CRTC:%d] [NOFB]\n", set->crtc->base.id);
-		set->mode = NULL;
-		set->num_connectors = 0;
+		return drm_crtc_helper_disable(set->crtc);
 	}
 
 	dev = set->crtc->dev;
@@ -557,6 +581,12 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 		save_connectors[count++] = *connector;
 	}
 
+	save_set.crtc = set->crtc;
+	save_set.mode = &set->crtc->mode;
+	save_set.x = set->crtc->x;
+	save_set.y = set->crtc->y;
+	save_set.fb = set->crtc->fb;
+
 	/* We should be able to check here if the fb has the same properties
 	 * and then just flip_or_move it */
 	if (set->crtc->fb != set->fb) {
@@ -722,6 +752,12 @@ fail:
 		*connector = save_connectors[count++];
 	}
 
+	/* Try to restore the config */
+	if (mode_changed &&
+	    !drm_crtc_helper_set_mode(save_set.crtc, save_set.mode, save_set.x,
+				      save_set.y, save_set.fb))
+		DRM_ERROR("failed to restore config after modeset failure\n");
+
 	kfree(save_connectors);
 	kfree(save_encoders);
 	kfree(save_crtcs);
diff --git a/drivers/gpu/drm/drm_debugfs.c b/drivers/gpu/drm/drm_debugfs.c
index d067c12ba940..1c7a1c0d3edd 100644
--- a/drivers/gpu/drm/drm_debugfs.c
+++ b/drivers/gpu/drm/drm_debugfs.c
@@ -118,7 +118,10 @@ int drm_debugfs_create_files(struct drm_info_list *files, int count,
 		tmp->minor = minor;
 		tmp->dent = ent;
 		tmp->info_ent = &files[i];
-		list_add(&(tmp->list), &(minor->debugfs_nodes.list));
+
+		mutex_lock(&minor->debugfs_lock);
+		list_add(&tmp->list, &minor->debugfs_list);
+		mutex_unlock(&minor->debugfs_lock);
 	}
 	return 0;
 
@@ -146,7 +149,8 @@ int drm_debugfs_init(struct drm_minor *minor, int minor_id,
 	char name[64];
 	int ret;
 
-	INIT_LIST_HEAD(&minor->debugfs_nodes.list);
+	INIT_LIST_HEAD(&minor->debugfs_list);
+	mutex_init(&minor->debugfs_lock);
 	sprintf(name, "%d", minor_id);
 	minor->debugfs_root = debugfs_create_dir(name, root);
 	if (!minor->debugfs_root) {
@@ -192,8 +196,9 @@ int drm_debugfs_remove_files(struct drm_info_list *files, int count,
 	struct drm_info_node *tmp;
 	int i;
 
+	mutex_lock(&minor->debugfs_lock);
 	for (i = 0; i < count; i++) {
-		list_for_each_safe(pos, q, &minor->debugfs_nodes.list) {
+		list_for_each_safe(pos, q, &minor->debugfs_list) {
 			tmp = list_entry(pos, struct drm_info_node, list);
 			if (tmp->info_ent == &files[i]) {
 				debugfs_remove(tmp->dent);
@@ -202,6 +207,7 @@ int drm_debugfs_remove_files(struct drm_info_list *files, int count,
 			}
 		}
 	}
+	mutex_unlock(&minor->debugfs_lock);
 	return 0;
 }
 EXPORT_SYMBOL(drm_debugfs_remove_files);
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index eaf25ffd9a46..bc5febe45762 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -125,7 +125,7 @@ static struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_SG_ALLOC, drm_sg_alloc_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF(DRM_IOCTL_SG_FREE, drm_sg_free, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 
-	DRM_IOCTL_DEF(DRM_IOCTL_WAIT_VBLANK, drm_wait_vblank, 0),
+	DRM_IOCTL_DEF(DRM_IOCTL_WAIT_VBLANK, drm_wait_vblank, DRM_UNLOCKED),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_MODESET_CTL, drm_modeset_ctl, 0),
 
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index cb3794a00f98..44a5d0ad8b7c 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -110,10 +110,7 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc)
 	/* Prevent vblank irq processing while disabling vblank irqs,
 	 * so no updates of timestamps or count can happen after we've
 	 * disabled. Needed to prevent races in case of delayed irq's.
-	 * Disable preemption, so vblank_time_lock is held as short as
-	 * possible, even under a kernel with PREEMPT_RT patches.
 	 */
-	preempt_disable();
 	spin_lock_irqsave(&dev->vblank_time_lock, irqflags);
 
 	dev->driver->disable_vblank(dev, crtc);
@@ -164,7 +161,6 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc)
 	clear_vblank_timestamps(dev, crtc);
 
 	spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
-	preempt_enable();
 }
 
 static void vblank_disable_fn(unsigned long arg)
@@ -407,13 +403,16 @@ int drm_irq_uninstall(struct drm_device *dev)
 	/*
 	 * Wake up any waiters so they don't hang.
 	 */
-	spin_lock_irqsave(&dev->vbl_lock, irqflags);
-	for (i = 0; i < dev->num_crtcs; i++) {
-		DRM_WAKEUP(&dev->vbl_queue[i]);
-		dev->vblank_enabled[i] = 0;
-		dev->last_vblank[i] = dev->driver->get_vblank_counter(dev, i);
+	if (dev->num_crtcs) {
+		spin_lock_irqsave(&dev->vbl_lock, irqflags);
+		for (i = 0; i < dev->num_crtcs; i++) {
+			DRM_WAKEUP(&dev->vbl_queue[i]);
+			dev->vblank_enabled[i] = 0;
+			dev->last_vblank[i] =
+				dev->driver->get_vblank_counter(dev, i);
+		}
+		spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 	}
-	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 
 	if (!irq_enabled)
 		return -EINVAL;
@@ -886,10 +885,6 @@ int drm_vblank_get(struct drm_device *dev, int crtc)
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
 	/* Going from 0->1 means we have to enable interrupts again */
 	if (atomic_add_return(1, &dev->vblank_refcount[crtc]) == 1) {
-		/* Disable preemption while holding vblank_time_lock. Do
-		 * it explicitely to guard against PREEMPT_RT kernel.
-		 */
-		preempt_disable();
 		spin_lock_irqsave(&dev->vblank_time_lock, irqflags2);
 		if (!dev->vblank_enabled[crtc]) {
 			/* Enable vblank irqs under vblank_time_lock protection.
@@ -909,7 +904,6 @@ int drm_vblank_get(struct drm_device *dev, int crtc)
 			}
 		}
 		spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags2);
-		preempt_enable();
 	} else {
 		if (!dev->vblank_enabled[crtc]) {
 			atomic_dec(&dev->vblank_refcount[crtc]);
@@ -1125,6 +1119,7 @@ static int drm_queue_vblank_event(struct drm_device *dev, int pipe,
 		trace_drm_vblank_event_delivered(current->pid, pipe,
 						 vblwait->request.sequence);
 	} else {
+		/* drm_handle_vblank_events will call drm_vblank_put */
 		list_add_tail(&e->base.link, &dev->vblank_event_list);
 		vblwait->reply.sequence = vblwait->request.sequence;
 	}
@@ -1205,8 +1200,12 @@ int drm_wait_vblank(struct drm_device *dev, void *data,
 		goto done;
 	}
 
-	if (flags & _DRM_VBLANK_EVENT)
+	if (flags & _DRM_VBLANK_EVENT) {
+		/* must hold on to the vblank ref until the event fires
+		 * drm_vblank_put will be called asynchronously
+		 */
 		return drm_queue_vblank_event(dev, crtc, vblwait, file_priv);
+	}
 
 	if ((flags & _DRM_VBLANK_NEXTONMISS) &&
 	    (seq - vblwait->request.sequence) <= (1<<23)) {
diff --git a/drivers/gpu/drm/exynos/exynos_drm_buf.c b/drivers/gpu/drm/exynos/exynos_drm_buf.c
index 6f8afea94fc9..2bb07bca511a 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_buf.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_buf.c
@@ -27,82 +27,84 @@
 #include "drm.h"
 
 #include "exynos_drm_drv.h"
+#include "exynos_drm_gem.h"
 #include "exynos_drm_buf.h"
 
-static DEFINE_MUTEX(exynos_drm_buf_lock);
-
 static int lowlevel_buffer_allocate(struct drm_device *dev,
-		struct exynos_drm_buf_entry *entry)
+		struct exynos_drm_gem_buf *buffer)
 {
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	entry->vaddr = dma_alloc_writecombine(dev->dev, entry->size,
-			(dma_addr_t *)&entry->paddr, GFP_KERNEL);
-	if (!entry->paddr) {
+	buffer->kvaddr = dma_alloc_writecombine(dev->dev, buffer->size,
+			&buffer->dma_addr, GFP_KERNEL);
+	if (!buffer->kvaddr) {
 		DRM_ERROR("failed to allocate buffer.\n");
 		return -ENOMEM;
 	}
 
-	DRM_DEBUG_KMS("allocated : vaddr(0x%x), paddr(0x%x), size(0x%x)\n",
-			(unsigned int)entry->vaddr, entry->paddr, entry->size);
+	DRM_DEBUG_KMS("vaddr(0x%lx), dma_addr(0x%lx), size(0x%lx)\n",
+			(unsigned long)buffer->kvaddr,
+			(unsigned long)buffer->dma_addr,
+			buffer->size);
 
 	return 0;
 }
 
 static void lowlevel_buffer_deallocate(struct drm_device *dev,
-		struct exynos_drm_buf_entry *entry)
+		struct exynos_drm_gem_buf *buffer)
 {
 	DRM_DEBUG_KMS("%s.\n", __FILE__);
 
-	if (entry->paddr && entry->vaddr && entry->size)
-		dma_free_writecombine(dev->dev, entry->size, entry->vaddr,
-				entry->paddr);
+	if (buffer->dma_addr && buffer->size)
+		dma_free_writecombine(dev->dev, buffer->size, buffer->kvaddr,
+				(dma_addr_t)buffer->dma_addr);
 	else
-		DRM_DEBUG_KMS("entry data is null.\n");
+		DRM_DEBUG_KMS("buffer data are invalid.\n");
 }
 
-struct exynos_drm_buf_entry *exynos_drm_buf_create(struct drm_device *dev,
+struct exynos_drm_gem_buf *exynos_drm_buf_create(struct drm_device *dev,
 		unsigned int size)
 {
-	struct exynos_drm_buf_entry *entry;
+	struct exynos_drm_gem_buf *buffer;
 
 	DRM_DEBUG_KMS("%s.\n", __FILE__);
+	DRM_DEBUG_KMS("desired size = 0x%x\n", size);
 
-	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
-	if (!entry) {
-		DRM_ERROR("failed to allocate exynos_drm_buf_entry.\n");
+	buffer = kzalloc(sizeof(*buffer), GFP_KERNEL);
+	if (!buffer) {
+		DRM_ERROR("failed to allocate exynos_drm_gem_buf.\n");
 		return ERR_PTR(-ENOMEM);
 	}
 
-	entry->size = size;
+	buffer->size = size;
 
 	/*
 	 * allocate memory region with size and set the memory information
-	 * to vaddr and paddr of a entry object.
+	 * to vaddr and dma_addr of a buffer object.
 	 */
-	if (lowlevel_buffer_allocate(dev, entry) < 0) {
-		kfree(entry);
-		entry = NULL;
+	if (lowlevel_buffer_allocate(dev, buffer) < 0) {
+		kfree(buffer);
+		buffer = NULL;
 		return ERR_PTR(-ENOMEM);
 	}
 
-	return entry;
+	return buffer;
 }
 
 void exynos_drm_buf_destroy(struct drm_device *dev,
-		struct exynos_drm_buf_entry *entry)
+		struct exynos_drm_gem_buf *buffer)
 {
 	DRM_DEBUG_KMS("%s.\n", __FILE__);
 
-	if (!entry) {
-		DRM_DEBUG_KMS("entry is null.\n");
+	if (!buffer) {
+		DRM_DEBUG_KMS("buffer is null.\n");
 		return;
 	}
 
-	lowlevel_buffer_deallocate(dev, entry);
+	lowlevel_buffer_deallocate(dev, buffer);
 
-	kfree(entry);
-	entry = NULL;
+	kfree(buffer);
+	buffer = NULL;
 }
 
 MODULE_AUTHOR("Inki Dae <inki.dae@samsung.com>");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_buf.h b/drivers/gpu/drm/exynos/exynos_drm_buf.h
index 045d59eab01a..6e91f9caa5db 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_buf.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_buf.h
@@ -26,28 +26,15 @@
 #ifndef _EXYNOS_DRM_BUF_H_
 #define _EXYNOS_DRM_BUF_H_
 
-/*
- * exynos drm buffer entry structure.
- *
- * @paddr: physical address of allocated memory.
- * @vaddr: kernel virtual address of allocated memory.
- * @size: size of allocated memory.
- */
-struct exynos_drm_buf_entry {
-	dma_addr_t paddr;
-	void __iomem *vaddr;
-	unsigned int size;
-};
-
 /* allocate physical memory. */
-struct exynos_drm_buf_entry *exynos_drm_buf_create(struct drm_device *dev,
+struct exynos_drm_gem_buf *exynos_drm_buf_create(struct drm_device *dev,
 		unsigned int size);
 
-/* get physical memory information of a drm framebuffer. */
-struct exynos_drm_buf_entry *exynos_drm_fb_get_buf(struct drm_framebuffer *fb);
+/* get memory information of a drm framebuffer. */
+struct exynos_drm_gem_buf *exynos_drm_fb_get_buf(struct drm_framebuffer *fb);
 
 /* remove allocated physical memory. */
 void exynos_drm_buf_destroy(struct drm_device *dev,
-		struct exynos_drm_buf_entry *entry);
+		struct exynos_drm_gem_buf *buffer);
 
 #endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_connector.c b/drivers/gpu/drm/exynos/exynos_drm_connector.c
index 985d9e768728..d620b0784257 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_connector.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_connector.c
@@ -37,6 +37,8 @@
 
 struct exynos_drm_connector {
 	struct drm_connector	drm_connector;
+	uint32_t		encoder_id;
+	struct exynos_drm_manager *manager;
 };
 
 /* convert exynos_video_timings to drm_display_mode */
@@ -47,6 +49,7 @@ convert_to_display_mode(struct drm_display_mode *mode,
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
 	mode->clock = timing->pixclock / 1000;
+	mode->vrefresh = timing->refresh;
 
 	mode->hdisplay = timing->xres;
 	mode->hsync_start = mode->hdisplay + timing->left_margin;
@@ -57,6 +60,12 @@ convert_to_display_mode(struct drm_display_mode *mode,
 	mode->vsync_start = mode->vdisplay + timing->upper_margin;
 	mode->vsync_end = mode->vsync_start + timing->vsync_len;
 	mode->vtotal = mode->vsync_end + timing->lower_margin;
+
+	if (timing->vmode & FB_VMODE_INTERLACED)
+		mode->flags |= DRM_MODE_FLAG_INTERLACE;
+
+	if (timing->vmode & FB_VMODE_DOUBLE)
+		mode->flags |= DRM_MODE_FLAG_DBLSCAN;
 }
 
 /* convert drm_display_mode to exynos_video_timings */
@@ -69,7 +78,7 @@ convert_to_video_timing(struct fb_videomode *timing,
 	memset(timing, 0, sizeof(*timing));
 
 	timing->pixclock = mode->clock * 1000;
-	timing->refresh = mode->vrefresh;
+	timing->refresh = drm_mode_vrefresh(mode);
 
 	timing->xres = mode->hdisplay;
 	timing->left_margin = mode->hsync_start - mode->hdisplay;
@@ -92,15 +101,16 @@ convert_to_video_timing(struct fb_videomode *timing,
 
 static int exynos_drm_connector_get_modes(struct drm_connector *connector)
 {
-	struct exynos_drm_manager *manager =
-				exynos_drm_get_manager(connector->encoder);
-	struct exynos_drm_display *display = manager->display;
+	struct exynos_drm_connector *exynos_connector =
+					to_exynos_connector(connector);
+	struct exynos_drm_manager *manager = exynos_connector->manager;
+	struct exynos_drm_display_ops *display_ops = manager->display_ops;
 	unsigned int count;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	if (!display) {
-		DRM_DEBUG_KMS("display is null.\n");
+	if (!display_ops) {
+		DRM_DEBUG_KMS("display_ops is null.\n");
 		return 0;
 	}
 
@@ -112,7 +122,7 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector)
 	 * P.S. in case of lcd panel, count is always 1 if success
 	 * because lcd panel has only one mode.
 	 */
-	if (display->get_edid) {
+	if (display_ops->get_edid) {
 		int ret;
 		void *edid;
 
@@ -122,7 +132,7 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector)
 			return 0;
 		}
 
-		ret = display->get_edid(manager->dev, connector,
+		ret = display_ops->get_edid(manager->dev, connector,
 						edid, MAX_EDID);
 		if (ret < 0) {
 			DRM_ERROR("failed to get edid data.\n");
@@ -140,8 +150,8 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector)
 		struct drm_display_mode *mode = drm_mode_create(connector->dev);
 		struct fb_videomode *timing;
 
-		if (display->get_timing)
-			timing = display->get_timing(manager->dev);
+		if (display_ops->get_timing)
+			timing = display_ops->get_timing(manager->dev);
 		else {
 			drm_mode_destroy(connector->dev, mode);
 			return 0;
@@ -162,9 +172,10 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector)
 static int exynos_drm_connector_mode_valid(struct drm_connector *connector,
 					    struct drm_display_mode *mode)
 {
-	struct exynos_drm_manager *manager =
-				exynos_drm_get_manager(connector->encoder);
-	struct exynos_drm_display *display = manager->display;
+	struct exynos_drm_connector *exynos_connector =
+					to_exynos_connector(connector);
+	struct exynos_drm_manager *manager = exynos_connector->manager;
+	struct exynos_drm_display_ops *display_ops = manager->display_ops;
 	struct fb_videomode timing;
 	int ret = MODE_BAD;
 
@@ -172,8 +183,8 @@ static int exynos_drm_connector_mode_valid(struct drm_connector *connector,
 
 	convert_to_video_timing(&timing, mode);
 
-	if (display && display->check_timing)
-		if (!display->check_timing(manager->dev, (void *)&timing))
+	if (display_ops && display_ops->check_timing)
+		if (!display_ops->check_timing(manager->dev, (void *)&timing))
 			ret = MODE_OK;
 
 	return ret;
@@ -181,9 +192,25 @@ static int exynos_drm_connector_mode_valid(struct drm_connector *connector,
 
 struct drm_encoder *exynos_drm_best_encoder(struct drm_connector *connector)
 {
+	struct drm_device *dev = connector->dev;
+	struct exynos_drm_connector *exynos_connector =
+					to_exynos_connector(connector);
+	struct drm_mode_object *obj;
+	struct drm_encoder *encoder;
+
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	return connector->encoder;
+	obj = drm_mode_object_find(dev, exynos_connector->encoder_id,
+				   DRM_MODE_OBJECT_ENCODER);
+	if (!obj) {
+		DRM_DEBUG_KMS("Unknown ENCODER ID %d\n",
+				exynos_connector->encoder_id);
+		return NULL;
+	}
+
+	encoder = obj_to_encoder(obj);
+
+	return encoder;
 }
 
 static struct drm_connector_helper_funcs exynos_connector_helper_funcs = {
@@ -196,15 +223,17 @@ static struct drm_connector_helper_funcs exynos_connector_helper_funcs = {
 static enum drm_connector_status
 exynos_drm_connector_detect(struct drm_connector *connector, bool force)
 {
-	struct exynos_drm_manager *manager =
-				exynos_drm_get_manager(connector->encoder);
-	struct exynos_drm_display *display = manager->display;
+	struct exynos_drm_connector *exynos_connector =
+					to_exynos_connector(connector);
+	struct exynos_drm_manager *manager = exynos_connector->manager;
+	struct exynos_drm_display_ops *display_ops =
+					manager->display_ops;
 	enum drm_connector_status status = connector_status_disconnected;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	if (display && display->is_connected) {
-		if (display->is_connected(manager->dev))
+	if (display_ops && display_ops->is_connected) {
+		if (display_ops->is_connected(manager->dev))
 			status = connector_status_connected;
 		else
 			status = connector_status_disconnected;
@@ -251,9 +280,11 @@ struct drm_connector *exynos_drm_connector_create(struct drm_device *dev,
 
 	connector = &exynos_connector->drm_connector;
 
-	switch (manager->display->type) {
+	switch (manager->display_ops->type) {
 	case EXYNOS_DISPLAY_TYPE_HDMI:
 		type = DRM_MODE_CONNECTOR_HDMIA;
+		connector->interlace_allowed = true;
+		connector->polled = DRM_CONNECTOR_POLL_HPD;
 		break;
 	default:
 		type = DRM_MODE_CONNECTOR_Unknown;
@@ -267,7 +298,10 @@ struct drm_connector *exynos_drm_connector_create(struct drm_device *dev,
 	if (err)
 		goto err_connector;
 
+	exynos_connector->encoder_id = encoder->base.id;
+	exynos_connector->manager = manager;
 	connector->encoder = encoder;
+
 	err = drm_mode_connector_attach_encoder(connector, encoder);
 	if (err) {
 		DRM_ERROR("failed to attach a connector to a encoder\n");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index 73893e5068a4..7777d41d1cda 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -29,36 +29,17 @@
 #include "drmP.h"
 #include "drm_crtc_helper.h"
 
+#include "exynos_drm_crtc.h"
 #include "exynos_drm_drv.h"
 #include "exynos_drm_fb.h"
 #include "exynos_drm_encoder.h"
+#include "exynos_drm_gem.h"
 #include "exynos_drm_buf.h"
 
 #define to_exynos_crtc(x)	container_of(x, struct exynos_drm_crtc,\
 				drm_crtc)
 
 /*
- * Exynos specific crtc postion structure.
- *
- * @fb_x: offset x on a framebuffer to be displyed
- *	- the unit is screen coordinates.
- * @fb_y: offset y on a framebuffer to be displayed
- *	- the unit is screen coordinates.
- * @crtc_x: offset x on hardware screen.
- * @crtc_y: offset y on hardware screen.
- * @crtc_w: width of hardware screen.
- * @crtc_h: height of hardware screen.
- */
-struct exynos_drm_crtc_pos {
-	unsigned int fb_x;
-	unsigned int fb_y;
-	unsigned int crtc_x;
-	unsigned int crtc_y;
-	unsigned int crtc_w;
-	unsigned int crtc_h;
-};
-
-/*
  * Exynos specific crtc structure.
  *
  * @drm_crtc: crtc object.
@@ -85,30 +66,31 @@ static void exynos_drm_crtc_apply(struct drm_crtc *crtc)
 
 	exynos_drm_fn_encoder(crtc, overlay,
 			exynos_drm_encoder_crtc_mode_set);
-	exynos_drm_fn_encoder(crtc, NULL, exynos_drm_encoder_crtc_commit);
+	exynos_drm_fn_encoder(crtc, &exynos_crtc->pipe,
+			exynos_drm_encoder_crtc_commit);
 }
 
-static int exynos_drm_overlay_update(struct exynos_drm_overlay *overlay,
-				       struct drm_framebuffer *fb,
-				       struct drm_display_mode *mode,
-				       struct exynos_drm_crtc_pos *pos)
+int exynos_drm_overlay_update(struct exynos_drm_overlay *overlay,
+			      struct drm_framebuffer *fb,
+			      struct drm_display_mode *mode,
+			      struct exynos_drm_crtc_pos *pos)
 {
-	struct exynos_drm_buf_entry *entry;
+	struct exynos_drm_gem_buf *buffer;
 	unsigned int actual_w;
 	unsigned int actual_h;
 
-	entry = exynos_drm_fb_get_buf(fb);
-	if (!entry) {
-		DRM_LOG_KMS("entry is null.\n");
+	buffer = exynos_drm_fb_get_buf(fb);
+	if (!buffer) {
+		DRM_LOG_KMS("buffer is null.\n");
 		return -EFAULT;
 	}
 
-	overlay->paddr = entry->paddr;
-	overlay->vaddr = entry->vaddr;
+	overlay->dma_addr = buffer->dma_addr;
+	overlay->vaddr = buffer->kvaddr;
 
-	DRM_DEBUG_KMS("vaddr = 0x%lx, paddr = 0x%lx\n",
+	DRM_DEBUG_KMS("vaddr = 0x%lx, dma_addr = 0x%lx\n",
 			(unsigned long)overlay->vaddr,
-			(unsigned long)overlay->paddr);
+			(unsigned long)overlay->dma_addr);
 
 	actual_w = min((mode->hdisplay - pos->crtc_x), pos->crtc_w);
 	actual_h = min((mode->vdisplay - pos->crtc_y), pos->crtc_h);
@@ -171,9 +153,26 @@ static int exynos_drm_crtc_update(struct drm_crtc *crtc)
 
 static void exynos_drm_crtc_dpms(struct drm_crtc *crtc, int mode)
 {
-	DRM_DEBUG_KMS("%s\n", __FILE__);
+	struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc);
 
-	/* TODO */
+	DRM_DEBUG_KMS("crtc[%d] mode[%d]\n", crtc->base.id, mode);
+
+	switch (mode) {
+	case DRM_MODE_DPMS_ON:
+		exynos_drm_fn_encoder(crtc, &exynos_crtc->pipe,
+				exynos_drm_encoder_crtc_commit);
+		break;
+	case DRM_MODE_DPMS_STANDBY:
+	case DRM_MODE_DPMS_SUSPEND:
+	case DRM_MODE_DPMS_OFF:
+		/* TODO */
+		exynos_drm_fn_encoder(crtc, NULL,
+				exynos_drm_encoder_crtc_disable);
+		break;
+	default:
+		DRM_DEBUG_KMS("unspecified mode %d\n", mode);
+		break;
+	}
 }
 
 static void exynos_drm_crtc_prepare(struct drm_crtc *crtc)
@@ -185,9 +184,12 @@ static void exynos_drm_crtc_prepare(struct drm_crtc *crtc)
 
 static void exynos_drm_crtc_commit(struct drm_crtc *crtc)
 {
+	struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc);
+
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	/* drm framework doesn't check NULL. */
+	exynos_drm_fn_encoder(crtc, &exynos_crtc->pipe,
+			exynos_drm_encoder_crtc_commit);
 }
 
 static bool
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.h b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
index c584042d6d2c..25f72a62cb88 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
@@ -35,4 +35,29 @@ int exynos_drm_crtc_create(struct drm_device *dev, unsigned int nr);
 int exynos_drm_crtc_enable_vblank(struct drm_device *dev, int crtc);
 void exynos_drm_crtc_disable_vblank(struct drm_device *dev, int crtc);
 
+/*
+ * Exynos specific crtc postion structure.
+ *
+ * @fb_x: offset x on a framebuffer to be displyed
+ *	- the unit is screen coordinates.
+ * @fb_y: offset y on a framebuffer to be displayed
+ *	- the unit is screen coordinates.
+ * @crtc_x: offset x on hardware screen.
+ * @crtc_y: offset y on hardware screen.
+ * @crtc_w: width of hardware screen.
+ * @crtc_h: height of hardware screen.
+ */
+struct exynos_drm_crtc_pos {
+	unsigned int fb_x;
+	unsigned int fb_y;
+	unsigned int crtc_x;
+	unsigned int crtc_y;
+	unsigned int crtc_w;
+	unsigned int crtc_h;
+};
+
+int exynos_drm_overlay_update(struct exynos_drm_overlay *overlay,
+			      struct drm_framebuffer *fb,
+			      struct drm_display_mode *mode,
+			      struct exynos_drm_crtc_pos *pos);
 #endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 83810cbe3c17..53e2216de61d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -27,6 +27,7 @@
 
 #include "drmP.h"
 #include "drm.h"
+#include "drm_crtc_helper.h"
 
 #include <drm/exynos_drm.h>
 
@@ -61,6 +62,9 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 
 	drm_mode_config_init(dev);
 
+	/* init kms poll for handling hpd */
+	drm_kms_helper_poll_init(dev);
+
 	exynos_drm_mode_config_init(dev);
 
 	/*
@@ -116,6 +120,7 @@ static int exynos_drm_unload(struct drm_device *dev)
 	exynos_drm_fbdev_fini(dev);
 	exynos_drm_device_unregister(dev);
 	drm_vblank_cleanup(dev);
+	drm_kms_helper_poll_fini(dev);
 	drm_mode_config_cleanup(dev);
 	kfree(dev->dev_private);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h
index c03683f2ae72..5e02e6ecc2e0 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h
@@ -29,6 +29,7 @@
 #ifndef _EXYNOS_DRM_DRV_H_
 #define _EXYNOS_DRM_DRV_H_
 
+#include <linux/module.h>
 #include "drm.h"
 
 #define MAX_CRTC	2
@@ -79,8 +80,8 @@ struct exynos_drm_overlay_ops {
  * @scan_flag: interlace or progressive way.
  *	(it could be DRM_MODE_FLAG_*)
  * @bpp: pixel size.(in bit)
- * @paddr: bus(accessed by dma) physical memory address to this overlay
- *		and this is physically continuous.
+ * @dma_addr: bus(accessed by dma) address to the memory region allocated
+ *	for a overlay.
  * @vaddr: virtual memory addresss to this overlay.
  * @default_win: a window to be enabled.
  * @color_key: color key on or off.
@@ -108,7 +109,7 @@ struct exynos_drm_overlay {
 	unsigned int scan_flag;
 	unsigned int bpp;
 	unsigned int pitch;
-	dma_addr_t paddr;
+	dma_addr_t dma_addr;
 	void __iomem *vaddr;
 
 	bool default_win;
@@ -130,7 +131,7 @@ struct exynos_drm_overlay {
  * @check_timing: check if timing is valid or not.
  * @power_on: display device on or off.
  */
-struct exynos_drm_display {
+struct exynos_drm_display_ops {
 	enum exynos_drm_output_type type;
 	bool (*is_connected)(struct device *dev);
 	int (*get_edid)(struct device *dev, struct drm_connector *connector,
@@ -146,12 +147,14 @@ struct exynos_drm_display {
  * @mode_set: convert drm_display_mode to hw specific display mode and
  *	      would be called by encoder->mode_set().
  * @commit: set current hw specific display mode to hw.
+ * @disable: disable hardware specific display mode.
  * @enable_vblank: specific driver callback for enabling vblank interrupt.
  * @disable_vblank: specific driver callback for disabling vblank interrupt.
  */
 struct exynos_drm_manager_ops {
 	void (*mode_set)(struct device *subdrv_dev, void *mode);
 	void (*commit)(struct device *subdrv_dev);
+	void (*disable)(struct device *subdrv_dev);
 	int (*enable_vblank)(struct device *subdrv_dev);
 	void (*disable_vblank)(struct device *subdrv_dev);
 };
@@ -178,7 +181,7 @@ struct exynos_drm_manager {
 	int pipe;
 	struct exynos_drm_manager_ops *ops;
 	struct exynos_drm_overlay_ops *overlay_ops;
-	struct exynos_drm_display *display;
+	struct exynos_drm_display_ops *display_ops;
 };
 
 /*
diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.c b/drivers/gpu/drm/exynos/exynos_drm_encoder.c
index 7cf6fa86a67e..153061415baf 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_encoder.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.c
@@ -53,15 +53,36 @@ static void exynos_drm_encoder_dpms(struct drm_encoder *encoder, int mode)
 	struct drm_device *dev = encoder->dev;
 	struct drm_connector *connector;
 	struct exynos_drm_manager *manager = exynos_drm_get_manager(encoder);
+	struct exynos_drm_manager_ops *manager_ops = manager->ops;
 
 	DRM_DEBUG_KMS("%s, encoder dpms: %d\n", __FILE__, mode);
 
+	switch (mode) {
+	case DRM_MODE_DPMS_ON:
+		if (manager_ops && manager_ops->commit)
+			manager_ops->commit(manager->dev);
+		break;
+	case DRM_MODE_DPMS_STANDBY:
+	case DRM_MODE_DPMS_SUSPEND:
+	case DRM_MODE_DPMS_OFF:
+		/* TODO */
+		if (manager_ops && manager_ops->disable)
+			manager_ops->disable(manager->dev);
+		break;
+	default:
+		DRM_ERROR("unspecified mode %d\n", mode);
+		break;
+	}
+
 	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
 		if (connector->encoder == encoder) {
-			struct exynos_drm_display *display = manager->display;
+			struct exynos_drm_display_ops *display_ops =
+							manager->display_ops;
 
-			if (display && display->power_on)
-				display->power_on(manager->dev, mode);
+			DRM_DEBUG_KMS("connector[%d] dpms[%d]\n",
+					connector->base.id, mode);
+			if (display_ops && display_ops->power_on)
+				display_ops->power_on(manager->dev, mode);
 		}
 	}
 }
@@ -116,15 +137,11 @@ static void exynos_drm_encoder_commit(struct drm_encoder *encoder)
 {
 	struct exynos_drm_manager *manager = exynos_drm_get_manager(encoder);
 	struct exynos_drm_manager_ops *manager_ops = manager->ops;
-	struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
 	if (manager_ops && manager_ops->commit)
 		manager_ops->commit(manager->dev);
-
-	if (overlay_ops && overlay_ops->commit)
-		overlay_ops->commit(manager->dev);
 }
 
 static struct drm_crtc *
@@ -208,10 +225,23 @@ void exynos_drm_fn_encoder(struct drm_crtc *crtc, void *data,
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_encoder *encoder;
+	struct exynos_drm_private *private = dev->dev_private;
+	struct exynos_drm_manager *manager;
 
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
-		if (encoder->crtc != crtc)
-			continue;
+		/*
+		 * if crtc is detached from encoder, check pipe,
+		 * otherwise check crtc attached to encoder
+		 */
+		if (!encoder->crtc) {
+			manager = to_exynos_encoder(encoder)->manager;
+			if (manager->pipe < 0 ||
+					private->crtc[manager->pipe] != crtc)
+				continue;
+		} else {
+			if (encoder->crtc != crtc)
+				continue;
+		}
 
 		fn(encoder, data);
 	}
@@ -250,8 +280,18 @@ void exynos_drm_encoder_crtc_commit(struct drm_encoder *encoder, void *data)
 	struct exynos_drm_manager *manager =
 		to_exynos_encoder(encoder)->manager;
 	struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops;
+	int crtc = *(int *)data;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	/*
+	 * when crtc is detached from encoder, this pipe is used
+	 * to select manager operation
+	 */
+	manager->pipe = crtc;
 
-	overlay_ops->commit(manager->dev);
+	if (overlay_ops && overlay_ops->commit)
+		overlay_ops->commit(manager->dev);
 }
 
 void exynos_drm_encoder_crtc_mode_set(struct drm_encoder *encoder, void *data)
@@ -261,7 +301,28 @@ void exynos_drm_encoder_crtc_mode_set(struct drm_encoder *encoder, void *data)
 	struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops;
 	struct exynos_drm_overlay *overlay = data;
 
-	overlay_ops->mode_set(manager->dev, overlay);
+	if (overlay_ops && overlay_ops->mode_set)
+		overlay_ops->mode_set(manager->dev, overlay);
+}
+
+void exynos_drm_encoder_crtc_disable(struct drm_encoder *encoder, void *data)
+{
+	struct exynos_drm_manager *manager =
+		to_exynos_encoder(encoder)->manager;
+	struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops;
+
+	DRM_DEBUG_KMS("\n");
+
+	if (overlay_ops && overlay_ops->disable)
+		overlay_ops->disable(manager->dev);
+
+	/*
+	 * crtc is already detached from encoder and last
+	 * function for detaching is properly done, so
+	 * clear pipe from manager to prevent repeated call
+	 */
+	if (!encoder->crtc)
+		manager->pipe = -1;
 }
 
 MODULE_AUTHOR("Inki Dae <inki.dae@samsung.com>");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.h b/drivers/gpu/drm/exynos/exynos_drm_encoder.h
index 5ecd645d06a9..a22acfbf0e4e 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_encoder.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.h
@@ -41,5 +41,6 @@ void exynos_drm_enable_vblank(struct drm_encoder *encoder, void *data);
 void exynos_drm_disable_vblank(struct drm_encoder *encoder, void *data);
 void exynos_drm_encoder_crtc_commit(struct drm_encoder *encoder, void *data);
 void exynos_drm_encoder_crtc_mode_set(struct drm_encoder *encoder, void *data);
+void exynos_drm_encoder_crtc_disable(struct drm_encoder *encoder, void *data);
 
 #endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c
index 48d29cfd5240..5bf4a1ac7f82 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fb.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c
@@ -29,7 +29,9 @@
 #include "drmP.h"
 #include "drm_crtc.h"
 #include "drm_crtc_helper.h"
+#include "drm_fb_helper.h"
 
+#include "exynos_drm_drv.h"
 #include "exynos_drm_fb.h"
 #include "exynos_drm_buf.h"
 #include "exynos_drm_gem.h"
@@ -41,14 +43,14 @@
  *
  * @fb: drm framebuffer obejct.
  * @exynos_gem_obj: exynos specific gem object containing a gem object.
- * @entry: pointer to exynos drm buffer entry object.
- *	- containing only the information to physically continuous memory
- *	region allocated at default framebuffer creation.
+ * @buffer: pointer to exynos_drm_gem_buffer object.
+ *	- contain the memory information to memory region allocated
+ *	at default framebuffer creation.
  */
 struct exynos_drm_fb {
 	struct drm_framebuffer		fb;
 	struct exynos_drm_gem_obj	*exynos_gem_obj;
-	struct exynos_drm_buf_entry	*entry;
+	struct exynos_drm_gem_buf	*buffer;
 };
 
 static void exynos_drm_fb_destroy(struct drm_framebuffer *fb)
@@ -63,8 +65,8 @@ static void exynos_drm_fb_destroy(struct drm_framebuffer *fb)
 	 * default framebuffer has no gem object so
 	 * a buffer of the default framebuffer should be released at here.
 	 */
-	if (!exynos_fb->exynos_gem_obj && exynos_fb->entry)
-		exynos_drm_buf_destroy(fb->dev, exynos_fb->entry);
+	if (!exynos_fb->exynos_gem_obj && exynos_fb->buffer)
+		exynos_drm_buf_destroy(fb->dev, exynos_fb->buffer);
 
 	kfree(exynos_fb);
 	exynos_fb = NULL;
@@ -143,29 +145,29 @@ exynos_drm_fb_init(struct drm_file *file_priv, struct drm_device *dev,
 	 */
 	if (!mode_cmd->handle) {
 		if (!file_priv) {
-			struct exynos_drm_buf_entry *entry;
+			struct exynos_drm_gem_buf *buffer;
 
 			/*
 			 * in case that file_priv is NULL, it allocates
 			 * only buffer and this buffer would be used
 			 * for default framebuffer.
 			 */
-			entry = exynos_drm_buf_create(dev, size);
-			if (IS_ERR(entry)) {
-				ret = PTR_ERR(entry);
+			buffer = exynos_drm_buf_create(dev, size);
+			if (IS_ERR(buffer)) {
+				ret = PTR_ERR(buffer);
 				goto err_buffer;
 			}
 
-			exynos_fb->entry = entry;
+			exynos_fb->buffer = buffer;
 
-			DRM_LOG_KMS("default fb: paddr = 0x%lx, size = 0x%x\n",
-					(unsigned long)entry->paddr, size);
+			DRM_LOG_KMS("default: dma_addr = 0x%lx, size = 0x%x\n",
+					(unsigned long)buffer->dma_addr, size);
 
 			goto out;
 		} else {
-			exynos_gem_obj = exynos_drm_gem_create(file_priv, dev,
-							size,
-							&mode_cmd->handle);
+			exynos_gem_obj = exynos_drm_gem_create(dev, file_priv,
+							&mode_cmd->handle,
+							size);
 			if (IS_ERR(exynos_gem_obj)) {
 				ret = PTR_ERR(exynos_gem_obj);
 				goto err_buffer;
@@ -189,10 +191,10 @@ exynos_drm_fb_init(struct drm_file *file_priv, struct drm_device *dev,
 	 * so that default framebuffer has no its own gem object,
 	 * only its own buffer object.
 	 */
-	exynos_fb->entry = exynos_gem_obj->entry;
+	exynos_fb->buffer = exynos_gem_obj->buffer;
 
-	DRM_LOG_KMS("paddr = 0x%lx, size = 0x%x, gem object = 0x%x\n",
-			(unsigned long)exynos_fb->entry->paddr, size,
+	DRM_LOG_KMS("dma_addr = 0x%lx, size = 0x%x, gem object = 0x%x\n",
+			(unsigned long)exynos_fb->buffer->dma_addr, size,
 			(unsigned int)&exynos_gem_obj->base);
 
 out:
@@ -220,26 +222,36 @@ struct drm_framebuffer *exynos_drm_fb_create(struct drm_device *dev,
 	return exynos_drm_fb_init(file_priv, dev, mode_cmd);
 }
 
-struct exynos_drm_buf_entry *exynos_drm_fb_get_buf(struct drm_framebuffer *fb)
+struct exynos_drm_gem_buf *exynos_drm_fb_get_buf(struct drm_framebuffer *fb)
 {
 	struct exynos_drm_fb *exynos_fb = to_exynos_fb(fb);
-	struct exynos_drm_buf_entry *entry;
+	struct exynos_drm_gem_buf *buffer;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	entry = exynos_fb->entry;
-	if (!entry)
+	buffer = exynos_fb->buffer;
+	if (!buffer)
 		return NULL;
 
-	DRM_DEBUG_KMS("vaddr = 0x%lx, paddr = 0x%lx\n",
-			(unsigned long)entry->vaddr,
-			(unsigned long)entry->paddr);
+	DRM_DEBUG_KMS("vaddr = 0x%lx, dma_addr = 0x%lx\n",
+			(unsigned long)buffer->kvaddr,
+			(unsigned long)buffer->dma_addr);
 
-	return entry;
+	return buffer;
+}
+
+static void exynos_drm_output_poll_changed(struct drm_device *dev)
+{
+	struct exynos_drm_private *private = dev->dev_private;
+	struct drm_fb_helper *fb_helper = private->fb_helper;
+
+	if (fb_helper)
+		drm_fb_helper_hotplug_event(fb_helper);
 }
 
 static struct drm_mode_config_funcs exynos_drm_mode_config_funcs = {
 	.fb_create = exynos_drm_fb_create,
+	.output_poll_changed = exynos_drm_output_poll_changed,
 };
 
 void exynos_drm_mode_config_init(struct drm_device *dev)
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index 81fba29b696d..f79f768a56ca 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -33,6 +33,7 @@
 
 #include "exynos_drm_drv.h"
 #include "exynos_drm_fb.h"
+#include "exynos_drm_gem.h"
 #include "exynos_drm_buf.h"
 
 #define MAX_CONNECTOR		4
@@ -85,15 +86,13 @@ static struct fb_ops exynos_drm_fb_ops = {
 };
 
 static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
-				     struct drm_framebuffer *fb,
-				     unsigned int fb_width,
-				     unsigned int fb_height)
+				     struct drm_framebuffer *fb)
 {
 	struct fb_info *fbi = helper->fbdev;
 	struct drm_device *dev = helper->dev;
 	struct exynos_drm_fbdev *exynos_fb = to_exynos_fbdev(helper);
-	struct exynos_drm_buf_entry *entry;
-	unsigned int size = fb_width * fb_height * (fb->bits_per_pixel >> 3);
+	struct exynos_drm_gem_buf *buffer;
+	unsigned int size = fb->width * fb->height * (fb->bits_per_pixel >> 3);
 	unsigned long offset;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
@@ -101,20 +100,20 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
 	exynos_fb->fb = fb;
 
 	drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->depth);
-	drm_fb_helper_fill_var(fbi, helper, fb_width, fb_height);
+	drm_fb_helper_fill_var(fbi, helper, fb->width, fb->height);
 
-	entry = exynos_drm_fb_get_buf(fb);
-	if (!entry) {
-		DRM_LOG_KMS("entry is null.\n");
+	buffer = exynos_drm_fb_get_buf(fb);
+	if (!buffer) {
+		DRM_LOG_KMS("buffer is null.\n");
 		return -EFAULT;
 	}
 
 	offset = fbi->var.xoffset * (fb->bits_per_pixel >> 3);
 	offset += fbi->var.yoffset * fb->pitches[0];
 
-	dev->mode_config.fb_base = entry->paddr;
-	fbi->screen_base = entry->vaddr + offset;
-	fbi->fix.smem_start = entry->paddr + offset;
+	dev->mode_config.fb_base = (resource_size_t)buffer->dma_addr;
+	fbi->screen_base = buffer->kvaddr + offset;
+	fbi->fix.smem_start = (unsigned long)(buffer->dma_addr + offset);
 	fbi->screen_size = size;
 	fbi->fix.smem_len = size;
 
@@ -171,8 +170,7 @@ static int exynos_drm_fbdev_create(struct drm_fb_helper *helper,
 		goto out;
 	}
 
-	ret = exynos_drm_fbdev_update(helper, helper->fb, sizes->fb_width,
-			sizes->fb_height);
+	ret = exynos_drm_fbdev_update(helper, helper->fb);
 	if (ret < 0)
 		fb_dealloc_cmap(&fbi->cmap);
 
@@ -235,8 +233,7 @@ static int exynos_drm_fbdev_recreate(struct drm_fb_helper *helper,
 	}
 
 	helper->fb = exynos_fbdev->fb;
-	return exynos_drm_fbdev_update(helper, helper->fb, sizes->fb_width,
-			sizes->fb_height);
+	return exynos_drm_fbdev_update(helper, helper->fb);
 }
 
 static int exynos_drm_fbdev_probe(struct drm_fb_helper *helper,
@@ -405,6 +402,18 @@ int exynos_drm_fbdev_reinit(struct drm_device *dev)
 	fb_helper = private->fb_helper;
 
 	if (fb_helper) {
+		struct list_head temp_list;
+
+		INIT_LIST_HEAD(&temp_list);
+
+		/*
+		 * fb_helper is reintialized but kernel fb is reused
+		 * so kernel_fb_list need to be backuped and restored
+		 */
+		if (!list_empty(&fb_helper->kernel_fb_list))
+			list_replace_init(&fb_helper->kernel_fb_list,
+					&temp_list);
+
 		drm_fb_helper_fini(fb_helper);
 
 		ret = drm_fb_helper_init(dev, fb_helper,
@@ -414,6 +423,9 @@ int exynos_drm_fbdev_reinit(struct drm_device *dev)
 			return ret;
 		}
 
+		if (!list_empty(&temp_list))
+			list_replace(&temp_list, &fb_helper->kernel_fb_list);
+
 		ret = drm_fb_helper_single_add_all_connectors(fb_helper);
 		if (ret < 0) {
 			DRM_ERROR("failed to add fb helper to connectors\n");
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 4659c88cdd9b..db3b3d9e731d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -64,7 +64,7 @@ struct fimd_win_data {
 	unsigned int		fb_width;
 	unsigned int		fb_height;
 	unsigned int		bpp;
-	dma_addr_t		paddr;
+	dma_addr_t		dma_addr;
 	void __iomem		*vaddr;
 	unsigned int		buf_offsize;
 	unsigned int		line_size;	/* bytes */
@@ -124,7 +124,7 @@ static int fimd_display_power_on(struct device *dev, int mode)
 	return 0;
 }
 
-static struct exynos_drm_display fimd_display = {
+static struct exynos_drm_display_ops fimd_display_ops = {
 	.type = EXYNOS_DISPLAY_TYPE_LCD,
 	.is_connected = fimd_display_is_connected,
 	.get_timing = fimd_get_timing,
@@ -177,6 +177,40 @@ static void fimd_commit(struct device *dev)
 	writel(val, ctx->regs + VIDCON0);
 }
 
+static void fimd_disable(struct device *dev)
+{
+	struct fimd_context *ctx = get_fimd_context(dev);
+	struct exynos_drm_subdrv *subdrv = &ctx->subdrv;
+	struct drm_device *drm_dev = subdrv->drm_dev;
+	struct exynos_drm_manager *manager = &subdrv->manager;
+	u32 val;
+
+	DRM_DEBUG_KMS("%s\n", __FILE__);
+
+	/* fimd dma off */
+	val = readl(ctx->regs + VIDCON0);
+	val &= ~(VIDCON0_ENVID | VIDCON0_ENVID_F);
+	writel(val, ctx->regs + VIDCON0);
+
+	/*
+	 * if vblank is enabled status with dma off then
+	 * it disables vsync interrupt.
+	 */
+	if (drm_dev->vblank_enabled[manager->pipe] &&
+		atomic_read(&drm_dev->vblank_refcount[manager->pipe])) {
+		drm_vblank_put(drm_dev, manager->pipe);
+
+		/*
+		 * if vblank_disable_allowed is 0 then disable
+		 * vsync interrupt right now else the vsync interrupt
+		 * would be disabled by drm timer once a current process
+		 * gives up ownershop of vblank event.
+		 */
+		if (!drm_dev->vblank_disable_allowed)
+			drm_vblank_off(drm_dev, manager->pipe);
+	}
+}
+
 static int fimd_enable_vblank(struct device *dev)
 {
 	struct fimd_context *ctx = get_fimd_context(dev);
@@ -220,6 +254,7 @@ static void fimd_disable_vblank(struct device *dev)
 
 static struct exynos_drm_manager_ops fimd_manager_ops = {
 	.commit = fimd_commit,
+	.disable = fimd_disable,
 	.enable_vblank = fimd_enable_vblank,
 	.disable_vblank = fimd_disable_vblank,
 };
@@ -251,7 +286,7 @@ static void fimd_win_mode_set(struct device *dev,
 	win_data->ovl_height = overlay->crtc_height;
 	win_data->fb_width = overlay->fb_width;
 	win_data->fb_height = overlay->fb_height;
-	win_data->paddr = overlay->paddr + offset;
+	win_data->dma_addr = overlay->dma_addr + offset;
 	win_data->vaddr = overlay->vaddr + offset;
 	win_data->bpp = overlay->bpp;
 	win_data->buf_offsize = (overlay->fb_width - overlay->crtc_width) *
@@ -263,7 +298,7 @@ static void fimd_win_mode_set(struct device *dev,
 	DRM_DEBUG_KMS("ovl_width = %d, ovl_height = %d\n",
 			win_data->ovl_width, win_data->ovl_height);
 	DRM_DEBUG_KMS("paddr = 0x%lx, vaddr = 0x%lx\n",
-			(unsigned long)win_data->paddr,
+			(unsigned long)win_data->dma_addr,
 			(unsigned long)win_data->vaddr);
 	DRM_DEBUG_KMS("fb_width = %d, crtc_width = %d\n",
 			overlay->fb_width, overlay->crtc_width);
@@ -376,16 +411,16 @@ static void fimd_win_commit(struct device *dev)
 	writel(val, ctx->regs + SHADOWCON);
 
 	/* buffer start address */
-	val = win_data->paddr;
+	val = (unsigned long)win_data->dma_addr;
 	writel(val, ctx->regs + VIDWx_BUF_START(win, 0));
 
 	/* buffer end address */
 	size = win_data->fb_width * win_data->ovl_height * (win_data->bpp >> 3);
-	val = win_data->paddr + size;
+	val = (unsigned long)(win_data->dma_addr + size);
 	writel(val, ctx->regs + VIDWx_BUF_END(win, 0));
 
 	DRM_DEBUG_KMS("start addr = 0x%lx, end addr = 0x%lx, size = 0x%lx\n",
-			(unsigned long)win_data->paddr, val, size);
+			(unsigned long)win_data->dma_addr, val, size);
 	DRM_DEBUG_KMS("ovl_width = %d, ovl_height = %d\n",
 			win_data->ovl_width, win_data->ovl_height);
 
@@ -447,7 +482,6 @@ static void fimd_win_commit(struct device *dev)
 static void fimd_win_disable(struct device *dev)
 {
 	struct fimd_context *ctx = get_fimd_context(dev);
-	struct fimd_win_data *win_data;
 	int win = ctx->default_win;
 	u32 val;
 
@@ -456,8 +490,6 @@ static void fimd_win_disable(struct device *dev)
 	if (win < 0 || win > WINDOWS_NR)
 		return;
 
-	win_data = &ctx->win_data[win];
-
 	/* protect windows */
 	val = readl(ctx->regs + SHADOWCON);
 	val |= SHADOWCON_WINx_PROTECT(win);
@@ -528,6 +560,16 @@ static irqreturn_t fimd_irq_handler(int irq, void *dev_id)
 		/* VSYNC interrupt */
 		writel(VIDINTCON1_INT_FRAME, ctx->regs + VIDINTCON1);
 
+	/*
+	 * in case that vblank_disable_allowed is 1, it could induce
+	 * the problem that manager->pipe could be -1 because with
+	 * disable callback, vsync interrupt isn't disabled and at this moment,
+	 * vsync interrupt could occur. the vsync interrupt would be disabled
+	 * by timer handler later.
+	 */
+	if (manager->pipe == -1)
+		return IRQ_HANDLED;
+
 	drm_handle_vblank(drm_dev, manager->pipe);
 	fimd_finish_pageflip(drm_dev, manager->pipe);
 
@@ -548,13 +590,6 @@ static int fimd_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
 	 */
 	drm_dev->irq_enabled = 1;
 
-	/*
-	 * with vblank_disable_allowed = 1, vblank interrupt will be disabled
-	 * by drm timer once a current process gives up ownership of
-	 * vblank event.(drm_vblank_put function was called)
-	 */
-	drm_dev->vblank_disable_allowed = 1;
-
 	return 0;
 }
 
@@ -731,7 +766,7 @@ static int __devinit fimd_probe(struct platform_device *pdev)
 	subdrv->manager.pipe = -1;
 	subdrv->manager.ops = &fimd_manager_ops;
 	subdrv->manager.overlay_ops = &fimd_overlay_ops;
-	subdrv->manager.display = &fimd_display;
+	subdrv->manager.display_ops = &fimd_display_ops;
 	subdrv->manager.dev = dev;
 
 	platform_set_drvdata(pdev, ctx);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index a8e7a88906ed..aba0fe47f7ea 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -62,40 +62,28 @@ static unsigned int get_gem_mmap_offset(struct drm_gem_object *obj)
 	return (unsigned int)obj->map_list.hash.key << PAGE_SHIFT;
 }
 
-struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_file *file_priv,
-		struct drm_device *dev, unsigned int size,
-		unsigned int *handle)
+static struct exynos_drm_gem_obj
+		*exynos_drm_gem_init(struct drm_device *drm_dev,
+			struct drm_file *file_priv, unsigned int *handle,
+			unsigned int size)
 {
 	struct exynos_drm_gem_obj *exynos_gem_obj;
-	struct exynos_drm_buf_entry *entry;
 	struct drm_gem_object *obj;
 	int ret;
 
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
-	size = roundup(size, PAGE_SIZE);
-
 	exynos_gem_obj = kzalloc(sizeof(*exynos_gem_obj), GFP_KERNEL);
 	if (!exynos_gem_obj) {
 		DRM_ERROR("failed to allocate exynos gem object.\n");
 		return ERR_PTR(-ENOMEM);
 	}
 
-	/* allocate the new buffer object and memory region. */
-	entry = exynos_drm_buf_create(dev, size);
-	if (!entry) {
-		kfree(exynos_gem_obj);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	exynos_gem_obj->entry = entry;
-
 	obj = &exynos_gem_obj->base;
 
-	ret = drm_gem_object_init(dev, obj, size);
+	ret = drm_gem_object_init(drm_dev, obj, size);
 	if (ret < 0) {
-		DRM_ERROR("failed to initailize gem object.\n");
-		goto err_obj_init;
+		DRM_ERROR("failed to initialize gem object.\n");
+		ret = -EINVAL;
+		goto err_object_init;
 	}
 
 	DRM_DEBUG_KMS("created file object = 0x%x\n", (unsigned int)obj->filp);
@@ -127,24 +115,50 @@ err_handle_create:
 err_create_mmap_offset:
 	drm_gem_object_release(obj);
 
-err_obj_init:
-	exynos_drm_buf_destroy(dev, exynos_gem_obj->entry);
-
+err_object_init:
 	kfree(exynos_gem_obj);
 
 	return ERR_PTR(ret);
 }
 
+struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev,
+				struct drm_file *file_priv,
+				unsigned int *handle, unsigned long size)
+{
+
+	struct exynos_drm_gem_obj *exynos_gem_obj = NULL;
+	struct exynos_drm_gem_buf *buffer;
+
+	size = roundup(size, PAGE_SIZE);
+
+	DRM_DEBUG_KMS("%s: size = 0x%lx\n", __FILE__, size);
+
+	buffer = exynos_drm_buf_create(dev, size);
+	if (IS_ERR(buffer)) {
+		return ERR_CAST(buffer);
+	}
+
+	exynos_gem_obj = exynos_drm_gem_init(dev, file_priv, handle, size);
+	if (IS_ERR(exynos_gem_obj)) {
+		exynos_drm_buf_destroy(dev, buffer);
+		return exynos_gem_obj;
+	}
+
+	exynos_gem_obj->buffer = buffer;
+
+	return exynos_gem_obj;
+}
+
 int exynos_drm_gem_create_ioctl(struct drm_device *dev, void *data,
-		struct drm_file *file_priv)
+					struct drm_file *file_priv)
 {
 	struct drm_exynos_gem_create *args = data;
-	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct exynos_drm_gem_obj *exynos_gem_obj = NULL;
 
-	DRM_DEBUG_KMS("%s : size = 0x%x\n", __FILE__, args->size);
+	DRM_DEBUG_KMS("%s\n", __FILE__);
 
-	exynos_gem_obj = exynos_drm_gem_create(file_priv, dev, args->size,
-			&args->handle);
+	exynos_gem_obj = exynos_drm_gem_create(dev, file_priv,
+						&args->handle, args->size);
 	if (IS_ERR(exynos_gem_obj))
 		return PTR_ERR(exynos_gem_obj);
 
@@ -175,7 +189,7 @@ static int exynos_drm_gem_mmap_buffer(struct file *filp,
 {
 	struct drm_gem_object *obj = filp->private_data;
 	struct exynos_drm_gem_obj *exynos_gem_obj = to_exynos_gem_obj(obj);
-	struct exynos_drm_buf_entry *entry;
+	struct exynos_drm_gem_buf *buffer;
 	unsigned long pfn, vm_size;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
@@ -187,20 +201,20 @@ static int exynos_drm_gem_mmap_buffer(struct file *filp,
 
 	vm_size = vma->vm_end - vma->vm_start;
 	/*
-	 * a entry contains information to physically continuous memory
+	 * a buffer contains information to physically continuous memory
 	 * allocated by user request or at framebuffer creation.
 	 */
-	entry = exynos_gem_obj->entry;
+	buffer = exynos_gem_obj->buffer;
 
 	/* check if user-requested size is valid. */
-	if (vm_size > entry->size)
+	if (vm_size > buffer->size)
 		return -EINVAL;
 
 	/*
 	 * get page frame number to physical memory to be mapped
 	 * to user space.
 	 */
-	pfn = exynos_gem_obj->entry->paddr >> PAGE_SHIFT;
+	pfn = ((unsigned long)exynos_gem_obj->buffer->dma_addr) >> PAGE_SHIFT;
 
 	DRM_DEBUG_KMS("pfn = 0x%lx\n", pfn);
 
@@ -281,7 +295,7 @@ void exynos_drm_gem_free_object(struct drm_gem_object *gem_obj)
 
 	exynos_gem_obj = to_exynos_gem_obj(gem_obj);
 
-	exynos_drm_buf_destroy(gem_obj->dev, exynos_gem_obj->entry);
+	exynos_drm_buf_destroy(gem_obj->dev, exynos_gem_obj->buffer);
 
 	kfree(exynos_gem_obj);
 }
@@ -302,8 +316,8 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv,
 	args->pitch = args->width * args->bpp >> 3;
 	args->size = args->pitch * args->height;
 
-	exynos_gem_obj = exynos_drm_gem_create(file_priv, dev, args->size,
-							&args->handle);
+	exynos_gem_obj = exynos_drm_gem_create(dev, file_priv, &args->handle,
+							args->size);
 	if (IS_ERR(exynos_gem_obj))
 		return PTR_ERR(exynos_gem_obj);
 
@@ -360,7 +374,8 @@ int exynos_drm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 	mutex_lock(&dev->struct_mutex);
 
-	pfn = (exynos_gem_obj->entry->paddr >> PAGE_SHIFT) + page_offset;
+	pfn = (((unsigned long)exynos_gem_obj->buffer->dma_addr) >>
+			PAGE_SHIFT) + page_offset;
 
 	ret = vm_insert_mixed(vma, (unsigned long)vmf->virtual_address, pfn);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h
index e5fc0148277b..ef8797334e6d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h
@@ -30,13 +30,29 @@
 			struct exynos_drm_gem_obj, base)
 
 /*
+ * exynos drm gem buffer structure.
+ *
+ * @kvaddr: kernel virtual address to allocated memory region.
+ * @dma_addr: bus address(accessed by dma) to allocated memory region.
+ *	- this address could be physical address without IOMMU and
+ *	device address with IOMMU.
+ * @size: size of allocated memory region.
+ */
+struct exynos_drm_gem_buf {
+	void __iomem		*kvaddr;
+	dma_addr_t		dma_addr;
+	unsigned long		size;
+};
+
+/*
  * exynos drm buffer structure.
  *
  * @base: a gem object.
  *	- a new handle to this gem object would be created
  *	by drm_gem_handle_create().
- * @entry: pointer to exynos drm buffer entry object.
- *	- containing the information to physically
+ * @buffer: a pointer to exynos_drm_gem_buffer object.
+ *	- contain the information to memory region allocated
+ *	by user request or at framebuffer creation.
  *	continuous memory region allocated by user request
  *	or at framebuffer creation.
  *
@@ -45,13 +61,13 @@
  */
 struct exynos_drm_gem_obj {
 	struct drm_gem_object base;
-	struct exynos_drm_buf_entry *entry;
+	struct exynos_drm_gem_buf *buffer;
 };
 
 /* create a new buffer and get a new gem handle. */
-struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_file *file_priv,
-		struct drm_device *dev, unsigned int size,
-		unsigned int *handle);
+struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev,
+		struct drm_file *file_priv,
+		unsigned int *handle, unsigned long size);
 
 /*
  * request gem object creation and buffer allocation as the size
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index d14b44e13f51..004b048c5192 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -62,6 +62,7 @@ static int i915_capabilities(struct seq_file *m, void *data)
 	const struct intel_device_info *info = INTEL_INFO(dev);
 
 	seq_printf(m, "gen: %d\n", info->gen);
+	seq_printf(m, "pch: %d\n", INTEL_PCH_TYPE(dev));
 #define B(x) seq_printf(m, #x ": %s\n", yesno(info->x))
 	B(is_mobile);
 	B(is_i85x);
@@ -636,11 +637,16 @@ static int i915_ringbuffer_info(struct seq_file *m, void *data)
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring;
+	int ret;
 
 	ring = &dev_priv->ring[(uintptr_t)node->info_ent->data];
 	if (ring->size == 0)
 		return 0;
 
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
 	seq_printf(m, "Ring %s:\n", ring->name);
 	seq_printf(m, "  Head :    %08x\n", I915_READ_HEAD(ring) & HEAD_ADDR);
 	seq_printf(m, "  Tail :    %08x\n", I915_READ_TAIL(ring) & TAIL_ADDR);
@@ -654,6 +660,8 @@ static int i915_ringbuffer_info(struct seq_file *m, void *data)
 	seq_printf(m, "  Control : %08x\n", I915_READ_CTL(ring));
 	seq_printf(m, "  Start :   %08x\n", I915_READ_START(ring));
 
+	mutex_unlock(&dev->struct_mutex);
+
 	return 0;
 }
 
@@ -842,7 +850,16 @@ static int i915_rstdby_delays(struct seq_file *m, void *unused)
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	u16 crstanddelay = I915_READ16(CRSTANDVID);
+	u16 crstanddelay;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	crstanddelay = I915_READ16(CRSTANDVID);
+
+	mutex_unlock(&dev->struct_mutex);
 
 	seq_printf(m, "w/ctx: %d, w/o ctx: %d\n", (crstanddelay >> 8) & 0x3f, (crstanddelay & 0x3f));
 
@@ -940,7 +957,11 @@ static int i915_delayfreq_table(struct seq_file *m, void *unused)
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	u32 delayfreq;
-	int i;
+	int ret, i;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
 
 	for (i = 0; i < 16; i++) {
 		delayfreq = I915_READ(PXVFREQ_BASE + i * 4);
@@ -948,6 +969,8 @@ static int i915_delayfreq_table(struct seq_file *m, void *unused)
 			   (delayfreq & PXVFREQ_PX_MASK) >> PXVFREQ_PX_SHIFT);
 	}
 
+	mutex_unlock(&dev->struct_mutex);
+
 	return 0;
 }
 
@@ -962,13 +985,19 @@ static int i915_inttoext_table(struct seq_file *m, void *unused)
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	u32 inttoext;
-	int i;
+	int ret, i;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
 
 	for (i = 1; i <= 32; i++) {
 		inttoext = I915_READ(INTTOEXT_BASE_ILK + i * 4);
 		seq_printf(m, "INTTOEXT%02d: 0x%08x\n", i, inttoext);
 	}
 
+	mutex_unlock(&dev->struct_mutex);
+
 	return 0;
 }
 
@@ -977,9 +1006,19 @@ static int i915_drpc_info(struct seq_file *m, void *unused)
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
-	u32 rgvmodectl = I915_READ(MEMMODECTL);
-	u32 rstdbyctl = I915_READ(RSTDBYCTL);
-	u16 crstandvid = I915_READ16(CRSTANDVID);
+	u32 rgvmodectl, rstdbyctl;
+	u16 crstandvid;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	rgvmodectl = I915_READ(MEMMODECTL);
+	rstdbyctl = I915_READ(RSTDBYCTL);
+	crstandvid = I915_READ16(CRSTANDVID);
+
+	mutex_unlock(&dev->struct_mutex);
 
 	seq_printf(m, "HD boost: %s\n", (rgvmodectl & MEMMODE_BOOST_EN) ?
 		   "yes" : "no");
@@ -1167,9 +1206,16 @@ static int i915_gfxec(struct seq_file *m, void *unused)
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
 
 	seq_printf(m, "GFXEC: %ld\n", (unsigned long)I915_READ(0x112f4));
 
+	mutex_unlock(&dev->struct_mutex);
+
 	return 0;
 }
 
@@ -1506,7 +1552,10 @@ drm_add_fake_info_node(struct drm_minor *minor,
 	node->minor = minor;
 	node->dent = ent;
 	node->info_ent = (void *) key;
-	list_add(&node->list, &minor->debugfs_nodes.list);
+
+	mutex_lock(&minor->debugfs_lock);
+	list_add(&node->list, &minor->debugfs_list);
+	mutex_unlock(&minor->debugfs_lock);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index a9533c54c93c..a9ae374861e7 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1454,6 +1454,14 @@ unsigned long i915_chipset_val(struct drm_i915_private *dev_priv)
 
 	diff1 = now - dev_priv->last_time1;
 
+	/* Prevent division-by-zero if we are asking too fast.
+	 * Also, we don't get interesting results if we are polling
+	 * faster than once in 10ms, so just return the saved value
+	 * in such cases.
+	 */
+	if (diff1 <= 10)
+		return dev_priv->chipset_power;
+
 	count1 = I915_READ(DMIEC);
 	count2 = I915_READ(DDREC);
 	count3 = I915_READ(CSIEC);
@@ -1484,6 +1492,8 @@ unsigned long i915_chipset_val(struct drm_i915_private *dev_priv)
 	dev_priv->last_count1 = total_count;
 	dev_priv->last_time1 = now;
 
+	dev_priv->chipset_power = ret;
+
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 9f592703c369..22c8ab70db2c 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -58,17 +58,17 @@ module_param_named(powersave, i915_powersave, int, 0600);
 MODULE_PARM_DESC(powersave,
 		"Enable powersavings, fbc, downclocking, etc. (default: true)");
 
-unsigned int i915_semaphores __read_mostly = 0;
+int i915_semaphores __read_mostly = -1;
 module_param_named(semaphores, i915_semaphores, int, 0600);
 MODULE_PARM_DESC(semaphores,
-		"Use semaphores for inter-ring sync (default: false)");
+		"Use semaphores for inter-ring sync (default: -1 (use per-chip defaults))");
 
-unsigned int i915_enable_rc6 __read_mostly = 0;
+int i915_enable_rc6 __read_mostly = -1;
 module_param_named(i915_enable_rc6, i915_enable_rc6, int, 0600);
 MODULE_PARM_DESC(i915_enable_rc6,
-		"Enable power-saving render C-state 6 (default: true)");
+		"Enable power-saving render C-state 6 (default: -1 (use per-chip default)");
 
-unsigned int i915_enable_fbc __read_mostly = -1;
+int i915_enable_fbc __read_mostly = -1;
 module_param_named(i915_enable_fbc, i915_enable_fbc, int, 0600);
 MODULE_PARM_DESC(i915_enable_fbc,
 		"Enable frame buffer compression for power savings "
@@ -80,7 +80,7 @@ MODULE_PARM_DESC(lvds_downclock,
 		"Use panel (LVDS/eDP) downclocking for power savings "
 		"(default: false)");
 
-unsigned int i915_panel_use_ssc __read_mostly = -1;
+int i915_panel_use_ssc __read_mostly = -1;
 module_param_named(lvds_use_ssc, i915_panel_use_ssc, int, 0600);
 MODULE_PARM_DESC(lvds_use_ssc,
 		"Use Spread Spectrum Clock with panels [LVDS/eDP] "
@@ -107,7 +107,7 @@ static struct drm_driver driver;
 extern int intel_agp_enabled;
 
 #define INTEL_VGA_DEVICE(id, info) {		\
-	.class = PCI_CLASS_DISPLAY_VGA << 8,	\
+	.class = PCI_BASE_CLASS_DISPLAY << 16,	\
 	.class_mask = 0xff0000,			\
 	.vendor = 0x8086,			\
 	.device = id,				\
@@ -328,7 +328,7 @@ void intel_detect_pch(struct drm_device *dev)
 	}
 }
 
-static void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
+void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
 {
 	int count;
 
@@ -344,6 +344,22 @@ static void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
 		udelay(10);
 }
 
+void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv)
+{
+	int count;
+
+	count = 0;
+	while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_MT_ACK) & 1))
+		udelay(10);
+
+	I915_WRITE_NOTRACE(FORCEWAKE_MT, (1<<16) | 1);
+	POSTING_READ(FORCEWAKE_MT);
+
+	count = 0;
+	while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_MT_ACK) & 1) == 0)
+		udelay(10);
+}
+
 /*
  * Generally this is called implicitly by the register read function. However,
  * if some sequence requires the GT to not power down then this function should
@@ -356,15 +372,21 @@ void gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
 
 	/* Forcewake is atomic in case we get in here without the lock */
 	if (atomic_add_return(1, &dev_priv->forcewake_count) == 1)
-		__gen6_gt_force_wake_get(dev_priv);
+		dev_priv->display.force_wake_get(dev_priv);
 }
 
-static void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
+void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
 {
 	I915_WRITE_NOTRACE(FORCEWAKE, 0);
 	POSTING_READ(FORCEWAKE);
 }
 
+void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv)
+{
+	I915_WRITE_NOTRACE(FORCEWAKE_MT, (1<<16) | 0);
+	POSTING_READ(FORCEWAKE_MT);
+}
+
 /*
  * see gen6_gt_force_wake_get()
  */
@@ -373,7 +395,7 @@ void gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
 	WARN_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex));
 
 	if (atomic_dec_and_test(&dev_priv->forcewake_count))
-		__gen6_gt_force_wake_put(dev_priv);
+		dev_priv->display.force_wake_put(dev_priv);
 }
 
 void __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
@@ -804,8 +826,8 @@ static const struct file_operations i915_driver_fops = {
 };
 
 static struct drm_driver driver = {
-	/* don't use mtrr's here, the Xserver or user space app should
-	 * deal with them for intel hardware.
+	/* Don't use MTRRs here; the Xserver or userspace app should
+	 * deal with them for Intel hardware.
 	 */
 	.driver_features =
 	    DRIVER_USE_AGP | DRIVER_REQUIRE_AGP | /* DRIVER_USE_MTRR |*/
@@ -904,8 +926,9 @@ MODULE_LICENSE("GPL and additional rights");
 /* We give fast paths for the really cool registers */
 #define NEEDS_FORCE_WAKE(dev_priv, reg) \
 	(((dev_priv)->info->gen >= 6) && \
-	((reg) < 0x40000) && \
-	((reg) != FORCEWAKE))
+	 ((reg) < 0x40000) &&		 \
+	 ((reg) != FORCEWAKE) &&	 \
+	 ((reg) != ECOBUS))
 
 #define __i915_read(x, y) \
 u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg) { \
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 06a37f4fd74b..554bef7a3b9c 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -107,6 +107,7 @@ struct opregion_header;
 struct opregion_acpi;
 struct opregion_swsci;
 struct opregion_asle;
+struct drm_i915_private;
 
 struct intel_opregion {
 	struct opregion_header *header;
@@ -126,6 +127,9 @@ struct drm_i915_master_private {
 	struct _drm_i915_sarea *sarea_priv;
 };
 #define I915_FENCE_REG_NONE -1
+#define I915_MAX_NUM_FENCES 16
+/* 16 fences + sign bit for FENCE_REG_NONE */
+#define I915_MAX_NUM_FENCE_BITS 5
 
 struct drm_i915_fence_reg {
 	struct list_head lru_list;
@@ -168,7 +172,7 @@ struct drm_i915_error_state {
 	u32 instdone1;
 	u32 seqno;
 	u64 bbaddr;
-	u64 fence[16];
+	u64 fence[I915_MAX_NUM_FENCES];
 	struct timeval time;
 	struct drm_i915_error_object {
 		int page_count;
@@ -182,7 +186,7 @@ struct drm_i915_error_state {
 		u32 gtt_offset;
 		u32 read_domains;
 		u32 write_domain;
-		s32 fence_reg:5;
+		s32 fence_reg:I915_MAX_NUM_FENCE_BITS;
 		s32 pinned:2;
 		u32 tiling:2;
 		u32 dirty:1;
@@ -218,6 +222,8 @@ struct drm_i915_display_funcs {
 			  struct drm_i915_gem_object *obj);
 	int (*update_plane)(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 			    int x, int y);
+	void (*force_wake_get)(struct drm_i915_private *dev_priv);
+	void (*force_wake_put)(struct drm_i915_private *dev_priv);
 	/* clock updates for mode set */
 	/* cursor updates */
 	/* render clock increase/decrease */
@@ -375,7 +381,7 @@ typedef struct drm_i915_private {
 	struct notifier_block lid_notifier;
 
 	int crt_ddc_pin;
-	struct drm_i915_fence_reg fence_regs[16]; /* assume 965 */
+	struct drm_i915_fence_reg fence_regs[I915_MAX_NUM_FENCES]; /* assume 965 */
 	int fence_reg_start; /* 4 if userland hasn't ioctl'd us yet */
 	int num_fence_regs; /* 8 on pre-965, 16 otherwise */
 
@@ -506,7 +512,7 @@ typedef struct drm_i915_private {
 	u8 saveAR[21];
 	u8 saveDACMASK;
 	u8 saveCR[37];
-	uint64_t saveFENCE[16];
+	uint64_t saveFENCE[I915_MAX_NUM_FENCES];
 	u32 saveCURACNTR;
 	u32 saveCURAPOS;
 	u32 saveCURABASE;
@@ -707,6 +713,7 @@ typedef struct drm_i915_private {
 
 	u64 last_count1;
 	unsigned long last_time1;
+	unsigned long chipset_power;
 	u64 last_count2;
 	struct timespec last_time2;
 	unsigned long gfx_power;
@@ -777,10 +784,8 @@ struct drm_i915_gem_object {
 	 * Fence register bits (if any) for this object.  Will be set
 	 * as needed when mapped into the GTT.
 	 * Protected by dev->struct_mutex.
-	 *
-	 * Size: 4 bits for 16 fences + sign (for FENCE_REG_NONE)
 	 */
-	signed int fence_reg:5;
+	signed int fence_reg:I915_MAX_NUM_FENCE_BITS;
 
 	/**
 	 * Advice: are the backing pages purgeable?
@@ -997,12 +1002,12 @@ extern int i915_max_ioctl;
 extern unsigned int i915_fbpercrtc __always_unused;
 extern int i915_panel_ignore_lid __read_mostly;
 extern unsigned int i915_powersave __read_mostly;
-extern unsigned int i915_semaphores __read_mostly;
+extern int i915_semaphores __read_mostly;
 extern unsigned int i915_lvds_downclock __read_mostly;
-extern unsigned int i915_panel_use_ssc __read_mostly;
+extern int i915_panel_use_ssc __read_mostly;
 extern int i915_vbt_sdvo_panel_type __read_mostly;
-extern unsigned int i915_enable_rc6 __read_mostly;
-extern unsigned int i915_enable_fbc __read_mostly;
+extern int i915_enable_rc6 __read_mostly;
+extern int i915_enable_fbc __read_mostly;
 extern bool i915_enable_hangcheck __read_mostly;
 
 extern int i915_suspend(struct drm_device *dev, pm_message_t state);
@@ -1307,6 +1312,11 @@ extern void gen6_set_rps(struct drm_device *dev, u8 val);
 extern void intel_detect_pch(struct drm_device *dev);
 extern int intel_trans_dp_port_sel(struct drm_crtc *crtc);
 
+extern void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv);
+extern void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv);
+extern void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv);
+extern void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv);
+
 /* overlay */
 #ifdef CONFIG_DEBUG_FS
 extern struct intel_overlay_error_state *intel_overlay_capture_error_state(struct drm_device *dev);
@@ -1351,8 +1361,9 @@ void __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv);
 /* We give fast paths for the really cool registers */
 #define NEEDS_FORCE_WAKE(dev_priv, reg) \
 	(((dev_priv)->info->gen >= 6) && \
-	((reg) < 0x40000) && \
-	((reg) != FORCEWAKE))
+	 ((reg) < 0x40000) &&		 \
+	 ((reg) != FORCEWAKE) &&	 \
+	 ((reg) != ECOBUS))
 
 #define __i915_read(x, y) \
 	u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 6651c36b6e8a..8359dc777041 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1396,7 +1396,7 @@ i915_gem_mmap_gtt(struct drm_file *file,
 
 	if (obj->base.size > dev_priv->mm.gtt_mappable_end) {
 		ret = -E2BIG;
-		goto unlock;
+		goto out;
 	}
 
 	if (obj->madv != I915_MADV_WILLNEED) {
@@ -1745,7 +1745,7 @@ static void i915_gem_reset_fences(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i;
 
-	for (i = 0; i < 16; i++) {
+	for (i = 0; i < dev_priv->num_fence_regs; i++) {
 		struct drm_i915_fence_reg *reg = &dev_priv->fence_regs[i];
 		struct drm_i915_gem_object *obj = reg->obj;
 
@@ -3512,9 +3512,11 @@ i915_gem_busy_ioctl(struct drm_device *dev, void *data,
 			 * so emit a request to do so.
 			 */
 			request = kzalloc(sizeof(*request), GFP_KERNEL);
-			if (request)
+			if (request) {
 				ret = i915_add_request(obj->ring, NULL, request);
-			else
+				if (ret)
+					kfree(request);
+			} else
 				ret = -ENOMEM;
 		}
 
@@ -3613,7 +3615,7 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
 	obj->base.write_domain = I915_GEM_DOMAIN_CPU;
 	obj->base.read_domains = I915_GEM_DOMAIN_CPU;
 
-	if (IS_GEN6(dev)) {
+	if (IS_GEN6(dev) || IS_GEN7(dev)) {
 		/* On Gen6, we can have the GPU use the LLC (the CPU
 		 * cache) for about a 10% performance improvement
 		 * compared to uncached.  Graphics requests other than
@@ -3877,7 +3879,7 @@ i915_gem_load(struct drm_device *dev)
 	INIT_LIST_HEAD(&dev_priv->mm.gtt_list);
 	for (i = 0; i < I915_NUM_RINGS; i++)
 		init_ring_lists(&dev_priv->ring[i]);
-	for (i = 0; i < 16; i++)
+	for (i = 0; i < I915_MAX_NUM_FENCES; i++)
 		INIT_LIST_HEAD(&dev_priv->fence_regs[i].lru_list);
 	INIT_DELAYED_WORK(&dev_priv->mm.retire_work,
 			  i915_gem_retire_work_handler);
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 3693e83a97f3..c681dc149d2a 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -32,6 +32,7 @@
 #include "i915_drv.h"
 #include "i915_trace.h"
 #include "intel_drv.h"
+#include <linux/dma_remapping.h>
 
 struct change_domains {
 	uint32_t invalidate_domains;
@@ -746,6 +747,22 @@ i915_gem_execbuffer_flush(struct drm_device *dev,
 	return 0;
 }
 
+static bool
+intel_enable_semaphores(struct drm_device *dev)
+{
+	if (INTEL_INFO(dev)->gen < 6)
+		return 0;
+
+	if (i915_semaphores >= 0)
+		return i915_semaphores;
+
+	/* Enable semaphores on SNB when IO remapping is off */
+	if (INTEL_INFO(dev)->gen == 6)
+		return !intel_iommu_enabled;
+
+	return 1;
+}
+
 static int
 i915_gem_execbuffer_sync_rings(struct drm_i915_gem_object *obj,
 			       struct intel_ring_buffer *to)
@@ -758,7 +775,7 @@ i915_gem_execbuffer_sync_rings(struct drm_i915_gem_object *obj,
 		return 0;
 
 	/* XXX gpu semaphores are implicated in various hard hangs on SNB */
-	if (INTEL_INFO(obj->base.dev)->gen < 6 || !i915_semaphores)
+	if (!intel_enable_semaphores(obj->base.dev))
 		return i915_gem_object_wait_rendering(obj);
 
 	idx = intel_ring_sync_index(from, to);
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 96643ee240da..3700df47ad93 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -824,6 +824,7 @@ static void i915_gem_record_fences(struct drm_device *dev,
 
 	/* Fences */
 	switch (INTEL_INFO(dev)->gen) {
+	case 7:
 	case 6:
 		for (i = 0; i < 16; i++)
 			error->fence[i] = I915_READ64(FENCE_REG_SANDYBRIDGE_0 + (i * 8));
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 517bf0cda3e5..853f2f0acaa2 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1553,12 +1553,21 @@
  */
 #define   PP_READY		(1 << 30)
 #define   PP_SEQUENCE_NONE	(0 << 28)
-#define   PP_SEQUENCE_ON	(1 << 28)
-#define   PP_SEQUENCE_OFF	(2 << 28)
-#define   PP_SEQUENCE_MASK	0x30000000
+#define   PP_SEQUENCE_POWER_UP	(1 << 28)
+#define   PP_SEQUENCE_POWER_DOWN (2 << 28)
+#define   PP_SEQUENCE_MASK	(3 << 28)
+#define   PP_SEQUENCE_SHIFT	28
 #define   PP_CYCLE_DELAY_ACTIVE	(1 << 27)
-#define   PP_SEQUENCE_STATE_ON_IDLE (1 << 3)
 #define   PP_SEQUENCE_STATE_MASK 0x0000000f
+#define   PP_SEQUENCE_STATE_OFF_IDLE	(0x0 << 0)
+#define   PP_SEQUENCE_STATE_OFF_S0_1	(0x1 << 0)
+#define   PP_SEQUENCE_STATE_OFF_S0_2	(0x2 << 0)
+#define   PP_SEQUENCE_STATE_OFF_S0_3	(0x3 << 0)
+#define   PP_SEQUENCE_STATE_ON_IDLE	(0x8 << 0)
+#define   PP_SEQUENCE_STATE_ON_S1_0	(0x9 << 0)
+#define   PP_SEQUENCE_STATE_ON_S1_2	(0xa << 0)
+#define   PP_SEQUENCE_STATE_ON_S1_3	(0xb << 0)
+#define   PP_SEQUENCE_STATE_RESET	(0xf << 0)
 #define PP_CONTROL	0x61204
 #define   POWER_TARGET_ON	(1 << 0)
 #define PP_ON_DELAYS	0x61208
@@ -3295,10 +3304,10 @@
 /* or SDVOB */
 #define HDMIB   0xe1140
 #define  PORT_ENABLE    (1 << 31)
-#define  TRANSCODER_A   (0)
-#define  TRANSCODER_B   (1 << 30)
-#define  TRANSCODER(pipe)	((pipe) << 30)
-#define  TRANSCODER_MASK   (1 << 30)
+#define  TRANSCODER(pipe)       ((pipe) << 30)
+#define  TRANSCODER_CPT(pipe)   ((pipe) << 29)
+#define  TRANSCODER_MASK        (1 << 30)
+#define  TRANSCODER_MASK_CPT    (3 << 29)
 #define  COLOR_FORMAT_8bpc      (0)
 #define  COLOR_FORMAT_12bpc     (3 << 26)
 #define  SDVOB_HOTPLUG_ENABLE   (1 << 23)
@@ -3439,12 +3448,38 @@
 #define  EDP_LINK_TRAIN_800_1200MV_0DB_SNB_B	(0x38<<22)
 #define  EDP_LINK_TRAIN_VOL_EMP_MASK_SNB	(0x3f<<22)
 
+/* IVB */
+#define EDP_LINK_TRAIN_400MV_0DB_IVB		(0x24 <<22)
+#define EDP_LINK_TRAIN_400MV_3_5DB_IVB		(0x2a <<22)
+#define EDP_LINK_TRAIN_400MV_6DB_IVB		(0x2f <<22)
+#define EDP_LINK_TRAIN_600MV_0DB_IVB		(0x30 <<22)
+#define EDP_LINK_TRAIN_600MV_3_5DB_IVB		(0x36 <<22)
+#define EDP_LINK_TRAIN_800MV_0DB_IVB		(0x38 <<22)
+#define EDP_LINK_TRAIN_800MV_3_5DB_IVB		(0x33 <<22)
+
+/* legacy values */
+#define EDP_LINK_TRAIN_500MV_0DB_IVB		(0x00 <<22)
+#define EDP_LINK_TRAIN_1000MV_0DB_IVB		(0x20 <<22)
+#define EDP_LINK_TRAIN_500MV_3_5DB_IVB		(0x02 <<22)
+#define EDP_LINK_TRAIN_1000MV_3_5DB_IVB		(0x22 <<22)
+#define EDP_LINK_TRAIN_1000MV_6DB_IVB		(0x23 <<22)
+
+#define  EDP_LINK_TRAIN_VOL_EMP_MASK_IVB	(0x3f<<22)
+
 #define  FORCEWAKE				0xA18C
 #define  FORCEWAKE_ACK				0x130090
+#define  FORCEWAKE_MT				0xa188 /* multi-threaded */
+#define  FORCEWAKE_MT_ACK			0x130040
+#define  ECOBUS					0xa180
+#define    FORCEWAKE_MT_ENABLE			(1<<5)
 
 #define  GT_FIFO_FREE_ENTRIES			0x120008
 #define    GT_FIFO_NUM_RESERVED_ENTRIES		20
 
+#define GEN6_UCGCTL2				0x9404
+# define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE		(1 << 12)
+# define GEN6_RCCUNIT_CLOCK_GATE_DISABLE		(1 << 11)
+
 #define GEN6_RPNSWREQ				0xA008
 #define   GEN6_TURBO_DISABLE			(1<<31)
 #define   GEN6_FREQUENCY(x)			((x)<<25)
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index f8f602d76650..7886e4fb60e3 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -370,6 +370,7 @@ static void i915_save_modeset_reg(struct drm_device *dev)
 
 	/* Fences */
 	switch (INTEL_INFO(dev)->gen) {
+	case 7:
 	case 6:
 		for (i = 0; i < 16; i++)
 			dev_priv->saveFENCE[i] = I915_READ64(FENCE_REG_SANDYBRIDGE_0 + (i * 8));
@@ -404,6 +405,7 @@ static void i915_restore_modeset_reg(struct drm_device *dev)
 
 	/* Fences */
 	switch (INTEL_INFO(dev)->gen) {
+	case 7:
 	case 6:
 		for (i = 0; i < 16; i++)
 			I915_WRITE64(FENCE_REG_SANDYBRIDGE_0 + (i * 8), dev_priv->saveFENCE[i]);
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 8ecbc2f11633..5a3e7853003f 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -38,8 +38,8 @@
 #include "i915_drv.h"
 #include "i915_trace.h"
 #include "drm_dp_helper.h"
-
 #include "drm_crtc_helper.h"
+#include <linux/dma_remapping.h>
 
 #define HAS_eDP (intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))
 
@@ -2933,7 +2933,8 @@ static void ironlake_pch_enable(struct drm_crtc *crtc)
 
 	/* For PCH DP, enable TRANS_DP_CTL */
 	if (HAS_PCH_CPT(dev) &&
-	    intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT)) {
+	    (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT) ||
+	     intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))) {
 		u32 bpc = (I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK) >> 5;
 		reg = TRANS_DP_CTL(pipe);
 		temp = I915_READ(reg);
@@ -4669,6 +4670,7 @@ static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv)
 /**
  * intel_choose_pipe_bpp_dither - figure out what color depth the pipe should send
  * @crtc: CRTC structure
+ * @mode: requested mode
  *
  * A pipe may be connected to one or more outputs.  Based on the depth of the
  * attached framebuffer, choose a good color depth to use on the pipe.
@@ -4680,13 +4682,15 @@ static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv)
  *    HDMI supports only 8bpc or 12bpc, so clamp to 8bpc with dither for 10bpc
  *    Displays may support a restricted set as well, check EDID and clamp as
  *      appropriate.
+ *    DP may want to dither down to 6bpc to fit larger modes
  *
  * RETURNS:
  * Dithering requirement (i.e. false if display bpc and pipe bpc match,
  * true if they don't match).
  */
 static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
-					 unsigned int *pipe_bpp)
+					 unsigned int *pipe_bpp,
+					 struct drm_display_mode *mode)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -4711,7 +4715,7 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
 				lvds_bpc = 6;
 
 			if (lvds_bpc < display_bpc) {
-				DRM_DEBUG_DRIVER("clamping display bpc (was %d) to LVDS (%d)\n", display_bpc, lvds_bpc);
+				DRM_DEBUG_KMS("clamping display bpc (was %d) to LVDS (%d)\n", display_bpc, lvds_bpc);
 				display_bpc = lvds_bpc;
 			}
 			continue;
@@ -4722,7 +4726,7 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
 			unsigned int edp_bpc = dev_priv->edp.bpp / 3;
 
 			if (edp_bpc < display_bpc) {
-				DRM_DEBUG_DRIVER("clamping display bpc (was %d) to eDP (%d)\n", display_bpc, edp_bpc);
+				DRM_DEBUG_KMS("clamping display bpc (was %d) to eDP (%d)\n", display_bpc, edp_bpc);
 				display_bpc = edp_bpc;
 			}
 			continue;
@@ -4737,7 +4741,7 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
 			/* Don't use an invalid EDID bpc value */
 			if (connector->display_info.bpc &&
 			    connector->display_info.bpc < display_bpc) {
-				DRM_DEBUG_DRIVER("clamping display bpc (was %d) to EDID reported max of %d\n", display_bpc, connector->display_info.bpc);
+				DRM_DEBUG_KMS("clamping display bpc (was %d) to EDID reported max of %d\n", display_bpc, connector->display_info.bpc);
 				display_bpc = connector->display_info.bpc;
 			}
 		}
@@ -4748,15 +4752,20 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
 		 */
 		if (intel_encoder->type == INTEL_OUTPUT_HDMI) {
 			if (display_bpc > 8 && display_bpc < 12) {
-				DRM_DEBUG_DRIVER("forcing bpc to 12 for HDMI\n");
+				DRM_DEBUG_KMS("forcing bpc to 12 for HDMI\n");
 				display_bpc = 12;
 			} else {
-				DRM_DEBUG_DRIVER("forcing bpc to 8 for HDMI\n");
+				DRM_DEBUG_KMS("forcing bpc to 8 for HDMI\n");
 				display_bpc = 8;
 			}
 		}
 	}
 
+	if (mode->private_flags & INTEL_MODE_DP_FORCE_6BPC) {
+		DRM_DEBUG_KMS("Dithering DP to 6bpc\n");
+		display_bpc = 6;
+	}
+
 	/*
 	 * We could just drive the pipe at the highest bpc all the time and
 	 * enable dithering as needed, but that costs bandwidth.  So choose
@@ -4789,8 +4798,8 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
 
 	display_bpc = min(display_bpc, bpc);
 
-	DRM_DEBUG_DRIVER("setting pipe bpc to %d (max display bpc %d)\n",
-			 bpc, display_bpc);
+	DRM_DEBUG_KMS("setting pipe bpc to %d (max display bpc %d)\n",
+		      bpc, display_bpc);
 
 	*pipe_bpp = display_bpc * 3;
 
@@ -5018,6 +5027,16 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 			pipeconf &= ~PIPECONF_DOUBLE_WIDE;
 	}
 
+	/* default to 8bpc */
+	pipeconf &= ~(PIPECONF_BPP_MASK | PIPECONF_DITHER_EN);
+	if (is_dp) {
+		if (mode->private_flags & INTEL_MODE_DP_FORCE_6BPC) {
+			pipeconf |= PIPECONF_BPP_6 |
+				    PIPECONF_DITHER_EN |
+				    PIPECONF_DITHER_TYPE_SP;
+		}
+	}
+
 	dpll |= DPLL_VCO_ENABLE;
 
 	DRM_DEBUG_KMS("Mode for pipe %c:\n", pipe == 0 ? 'A' : 'B');
@@ -5479,7 +5498,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 	/* determine panel color depth */
 	temp = I915_READ(PIPECONF(pipe));
 	temp &= ~PIPE_BPC_MASK;
-	dither = intel_choose_pipe_bpp_dither(crtc, &pipe_bpp);
+	dither = intel_choose_pipe_bpp_dither(crtc, &pipe_bpp, mode);
 	switch (pipe_bpp) {
 	case 18:
 		temp |= PIPE_6BPC;
@@ -5671,7 +5690,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 	pipeconf &= ~PIPECONF_DITHER_TYPE_MASK;
 	if ((is_lvds && dev_priv->lvds_dither) || dither) {
 		pipeconf |= PIPECONF_DITHER_EN;
-		pipeconf |= PIPECONF_DITHER_TYPE_ST1;
+		pipeconf |= PIPECONF_DITHER_TYPE_SP;
 	}
 	if (is_dp || intel_encoder_is_pch_edp(&has_edp_encoder->base)) {
 		intel_dp_set_m_n(crtc, mode, adjusted_mode);
@@ -7188,11 +7207,16 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	work->old_fb_obj = intel_fb->obj;
 	INIT_WORK(&work->work, intel_unpin_work_fn);
 
+	ret = drm_vblank_get(dev, intel_crtc->pipe);
+	if (ret)
+		goto free_work;
+
 	/* We borrow the event spin lock for protecting unpin_work */
 	spin_lock_irqsave(&dev->event_lock, flags);
 	if (intel_crtc->unpin_work) {
 		spin_unlock_irqrestore(&dev->event_lock, flags);
 		kfree(work);
+		drm_vblank_put(dev, intel_crtc->pipe);
 
 		DRM_DEBUG_DRIVER("flip queue: crtc already busy\n");
 		return -EBUSY;
@@ -7211,10 +7235,6 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	crtc->fb = fb;
 
-	ret = drm_vblank_get(dev, intel_crtc->pipe);
-	if (ret)
-		goto cleanup_objs;
-
 	work->pending_flip_obj = obj;
 
 	work->enable_stall_check = true;
@@ -7237,7 +7257,6 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 cleanup_pending:
 	atomic_sub(1 << intel_crtc->plane, &work->old_fb_obj->pending_flip);
-cleanup_objs:
 	drm_gem_object_unreference(&work->old_fb_obj->base);
 	drm_gem_object_unreference(&obj->base);
 	mutex_unlock(&dev->struct_mutex);
@@ -7246,6 +7265,8 @@ cleanup_objs:
 	intel_crtc->unpin_work = NULL;
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 
+	drm_vblank_put(dev, intel_crtc->pipe);
+free_work:
 	kfree(work);
 
 	return ret;
@@ -7891,6 +7912,33 @@ void intel_init_emon(struct drm_device *dev)
 	dev_priv->corr = (lcfuse & LCFUSE_HIV_MASK);
 }
 
+static bool intel_enable_rc6(struct drm_device *dev)
+{
+	/*
+	 * Respect the kernel parameter if it is set
+	 */
+	if (i915_enable_rc6 >= 0)
+		return i915_enable_rc6;
+
+	/*
+	 * Disable RC6 on Ironlake
+	 */
+	if (INTEL_INFO(dev)->gen == 5)
+		return 0;
+
+	/*
+	 * Enable rc6 on Sandybridge if DMA remapping is disabled
+	 */
+	if (INTEL_INFO(dev)->gen == 6) {
+		DRM_DEBUG_DRIVER("Sandybridge: intel_iommu_enabled %s -- RC6 %sabled\n",
+				 intel_iommu_enabled ? "true" : "false",
+				 !intel_iommu_enabled ? "en" : "dis");
+		return !intel_iommu_enabled;
+	}
+	DRM_DEBUG_DRIVER("RC6 enabled\n");
+	return 1;
+}
+
 void gen6_enable_rps(struct drm_i915_private *dev_priv)
 {
 	u32 rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
@@ -7927,7 +7975,7 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
 	I915_WRITE(GEN6_RC6p_THRESHOLD, 100000);
 	I915_WRITE(GEN6_RC6pp_THRESHOLD, 64000); /* unused */
 
-	if (i915_enable_rc6)
+	if (intel_enable_rc6(dev_priv->dev))
 		rc6_mask = GEN6_RC_CTL_RC6p_ENABLE |
 			GEN6_RC_CTL_RC6_ENABLE;
 
@@ -8153,6 +8201,20 @@ static void gen6_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(WM2_LP_ILK, 0);
 	I915_WRITE(WM1_LP_ILK, 0);
 
+	/* According to the BSpec vol1g, bit 12 (RCPBUNIT) clock
+	 * gating disable must be set.  Failure to set it results in
+	 * flickering pixels due to Z write ordering failures after
+	 * some amount of runtime in the Mesa "fire" demo, and Unigine
+	 * Sanctuary and Tropics, and apparently anything else with
+	 * alpha test or pixel discard.
+	 *
+	 * According to the spec, bit 11 (RCCUNIT) must also be set,
+	 * but we didn't debug actual testcases to find it out.
+	 */
+	I915_WRITE(GEN6_UCGCTL2,
+		   GEN6_RCPBUNIT_CLOCK_GATE_DISABLE |
+		   GEN6_RCCUNIT_CLOCK_GATE_DISABLE);
+
 	/*
 	 * According to the spec the following bits should be
 	 * set in order to enable memory self-refresh and fbc:
@@ -8362,7 +8424,7 @@ void ironlake_enable_rc6(struct drm_device *dev)
 	/* rc6 disabled by default due to repeated reports of hanging during
 	 * boot and resume.
 	 */
-	if (!i915_enable_rc6)
+	if (!intel_enable_rc6(dev))
 		return;
 
 	mutex_lock(&dev->struct_mutex);
@@ -8481,6 +8543,28 @@ static void intel_init_display(struct drm_device *dev)
 
 	/* For FIFO watermark updates */
 	if (HAS_PCH_SPLIT(dev)) {
+		dev_priv->display.force_wake_get = __gen6_gt_force_wake_get;
+		dev_priv->display.force_wake_put = __gen6_gt_force_wake_put;
+
+		/* IVB configs may use multi-threaded forcewake */
+		if (IS_IVYBRIDGE(dev)) {
+			u32	ecobus;
+
+			mutex_lock(&dev->struct_mutex);
+			__gen6_gt_force_wake_mt_get(dev_priv);
+			ecobus = I915_READ(ECOBUS);
+			__gen6_gt_force_wake_mt_put(dev_priv);
+			mutex_unlock(&dev->struct_mutex);
+
+			if (ecobus & FORCEWAKE_MT_ENABLE) {
+				DRM_DEBUG_KMS("Using MT version of forcewake\n");
+				dev_priv->display.force_wake_get =
+					__gen6_gt_force_wake_mt_get;
+				dev_priv->display.force_wake_put =
+					__gen6_gt_force_wake_mt_put;
+			}
+		}
+
 		if (HAS_PCH_IBX(dev))
 			dev_priv->display.init_pch_clock_gating = ibx_init_clock_gating;
 		else if (HAS_PCH_CPT(dev))
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 09b318b0227f..92b041b66e49 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -59,7 +59,6 @@ struct intel_dp {
 	struct i2c_algo_dp_aux_data algo;
 	bool is_pch_edp;
 	uint8_t	train_set[4];
-	uint8_t link_status[DP_LINK_STATUS_SIZE];
 	int panel_power_up_delay;
 	int panel_power_down_delay;
 	int panel_power_cycle_delay;
@@ -68,7 +67,6 @@ struct intel_dp {
 	struct drm_display_mode *panel_fixed_mode;  /* for eDP */
 	struct delayed_work panel_vdd_work;
 	bool want_panel_vdd;
-	unsigned long panel_off_jiffies;
 };
 
 /**
@@ -157,16 +155,12 @@ intel_edp_link_config(struct intel_encoder *intel_encoder,
 static int
 intel_dp_max_lane_count(struct intel_dp *intel_dp)
 {
-	int max_lane_count = 4;
-
-	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11) {
-		max_lane_count = intel_dp->dpcd[DP_MAX_LANE_COUNT] & 0x1f;
-		switch (max_lane_count) {
-		case 1: case 2: case 4:
-			break;
-		default:
-			max_lane_count = 4;
-		}
+	int max_lane_count = intel_dp->dpcd[DP_MAX_LANE_COUNT] & 0x1f;
+	switch (max_lane_count) {
+	case 1: case 2: case 4:
+		break;
+	default:
+		max_lane_count = 4;
 	}
 	return max_lane_count;
 }
@@ -214,13 +208,15 @@ intel_dp_link_clock(uint8_t link_bw)
  */
 
 static int
-intel_dp_link_required(struct intel_dp *intel_dp, int pixel_clock)
+intel_dp_link_required(struct intel_dp *intel_dp, int pixel_clock, int check_bpp)
 {
 	struct drm_crtc *crtc = intel_dp->base.base.crtc;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int bpp = 24;
 
-	if (intel_crtc)
+	if (check_bpp)
+		bpp = check_bpp;
+	else if (intel_crtc)
 		bpp = intel_crtc->bpp;
 
 	return (pixel_clock * bpp + 9) / 10;
@@ -239,6 +235,7 @@ intel_dp_mode_valid(struct drm_connector *connector,
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
 	int max_link_clock = intel_dp_link_clock(intel_dp_max_link_bw(intel_dp));
 	int max_lanes = intel_dp_max_lane_count(intel_dp);
+	int max_rate, mode_rate;
 
 	if (is_edp(intel_dp) && intel_dp->panel_fixed_mode) {
 		if (mode->hdisplay > intel_dp->panel_fixed_mode->hdisplay)
@@ -248,9 +245,17 @@ intel_dp_mode_valid(struct drm_connector *connector,
 			return MODE_PANEL;
 	}
 
-	if (intel_dp_link_required(intel_dp, mode->clock)
-	    > intel_dp_max_data_rate(max_link_clock, max_lanes))
-		return MODE_CLOCK_HIGH;
+	mode_rate = intel_dp_link_required(intel_dp, mode->clock, 0);
+	max_rate = intel_dp_max_data_rate(max_link_clock, max_lanes);
+
+	if (mode_rate > max_rate) {
+			mode_rate = intel_dp_link_required(intel_dp,
+							   mode->clock, 18);
+			if (mode_rate > max_rate)
+				return MODE_CLOCK_HIGH;
+			else
+				mode->private_flags |= INTEL_MODE_DP_FORCE_6BPC;
+	}
 
 	if (mode->clock < 10000)
 		return MODE_CLOCK_LOW;
@@ -368,8 +373,8 @@ intel_dp_aux_ch(struct intel_dp *intel_dp,
 	 * clock divider.
 	 */
 	if (is_cpu_edp(intel_dp)) {
-		if (IS_GEN6(dev))
-			aux_clock_divider = 200; /* SNB eDP input clock at 400Mhz */
+		if (IS_GEN6(dev) || IS_GEN7(dev))
+			aux_clock_divider = 200; /* SNB & IVB eDP input clock at 400Mhz */
 		else
 			aux_clock_divider = 225; /* eDP input clock at 450Mhz */
 	} else if (HAS_PCH_SPLIT(dev))
@@ -678,6 +683,7 @@ intel_dp_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode,
 	int lane_count, clock;
 	int max_lane_count = intel_dp_max_lane_count(intel_dp);
 	int max_clock = intel_dp_max_link_bw(intel_dp) == DP_LINK_BW_2_7 ? 1 : 0;
+	int bpp = mode->private_flags & INTEL_MODE_DP_FORCE_6BPC ? 18 : 0;
 	static int bws[2] = { DP_LINK_BW_1_62, DP_LINK_BW_2_7 };
 
 	if (is_edp(intel_dp) && intel_dp->panel_fixed_mode) {
@@ -695,7 +701,7 @@ intel_dp_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode,
 		for (clock = 0; clock <= max_clock; clock++) {
 			int link_avail = intel_dp_max_data_rate(intel_dp_link_clock(bws[clock]), lane_count);
 
-			if (intel_dp_link_required(intel_dp, mode->clock)
+			if (intel_dp_link_required(intel_dp, mode->clock, bpp)
 					<= link_avail) {
 				intel_dp->link_bw = bws[clock];
 				intel_dp->lane_count = lane_count;
@@ -768,12 +774,11 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
 			continue;
 
 		intel_dp = enc_to_intel_dp(encoder);
-		if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT) {
+		if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT ||
+		    intel_dp->base.type == INTEL_OUTPUT_EDP)
+		{
 			lane_count = intel_dp->lane_count;
 			break;
-		} else if (is_edp(intel_dp)) {
-			lane_count = dev_priv->edp.lanes;
-			break;
 		}
 	}
 
@@ -810,6 +815,7 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
 		  struct drm_display_mode *adjusted_mode)
 {
 	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	struct drm_crtc *crtc = intel_dp->base.base.crtc;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -822,18 +828,32 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
 			ironlake_edp_pll_off(encoder);
 	}
 
-	intel_dp->DP = DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0;
-	intel_dp->DP |= intel_dp->color_range;
+	/*
+	 * There are four kinds of DP registers:
+	 *
+	 * 	IBX PCH
+	 * 	SNB CPU
+	 *	IVB CPU
+	 * 	CPT PCH
+	 *
+	 * IBX PCH and CPU are the same for almost everything,
+	 * except that the CPU DP PLL is configured in this
+	 * register
+	 *
+	 * CPT PCH is quite different, having many bits moved
+	 * to the TRANS_DP_CTL register instead. That
+	 * configuration happens (oddly) in ironlake_pch_enable
+	 */
+
+	/* Preserve the BIOS-computed detected bit. This is
+	 * supposed to be read-only.
+	 */
+	intel_dp->DP = I915_READ(intel_dp->output_reg) & DP_DETECTED;
+	intel_dp->DP |=  DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0;
 
-	if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
-		intel_dp->DP |= DP_SYNC_HS_HIGH;
-	if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
-		intel_dp->DP |= DP_SYNC_VS_HIGH;
+	/* Handle DP bits in common between all three register formats */
 
-	if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
-		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
-	else
-		intel_dp->DP |= DP_LINK_TRAIN_OFF;
+	intel_dp->DP |= DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0;
 
 	switch (intel_dp->lane_count) {
 	case 1:
@@ -852,59 +872,124 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
 		intel_dp->DP |= DP_AUDIO_OUTPUT_ENABLE;
 		intel_write_eld(encoder, adjusted_mode);
 	}
-
 	memset(intel_dp->link_configuration, 0, DP_LINK_CONFIGURATION_SIZE);
 	intel_dp->link_configuration[0] = intel_dp->link_bw;
 	intel_dp->link_configuration[1] = intel_dp->lane_count;
 	intel_dp->link_configuration[8] = DP_SET_ANSI_8B10B;
-
 	/*
 	 * Check for DPCD version > 1.1 and enhanced framing support
 	 */
 	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 &&
 	    (intel_dp->dpcd[DP_MAX_LANE_COUNT] & DP_ENHANCED_FRAME_CAP)) {
 		intel_dp->link_configuration[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
-		intel_dp->DP |= DP_ENHANCED_FRAMING;
 	}
 
-	/* CPT DP's pipe select is decided in TRANS_DP_CTL */
-	if (intel_crtc->pipe == 1 && !HAS_PCH_CPT(dev))
-		intel_dp->DP |= DP_PIPEB_SELECT;
+	/* Split out the IBX/CPU vs CPT settings */
+
+	if (is_cpu_edp(intel_dp) && IS_GEN7(dev)) {
+		if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
+			intel_dp->DP |= DP_SYNC_HS_HIGH;
+		if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
+			intel_dp->DP |= DP_SYNC_VS_HIGH;
+		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
+
+		if (intel_dp->link_configuration[1] & DP_LANE_COUNT_ENHANCED_FRAME_EN)
+			intel_dp->DP |= DP_ENHANCED_FRAMING;
+
+		intel_dp->DP |= intel_crtc->pipe << 29;
 
-	if (is_cpu_edp(intel_dp)) {
 		/* don't miss out required setting for eDP */
 		intel_dp->DP |= DP_PLL_ENABLE;
 		if (adjusted_mode->clock < 200000)
 			intel_dp->DP |= DP_PLL_FREQ_160MHZ;
 		else
 			intel_dp->DP |= DP_PLL_FREQ_270MHZ;
+	} else if (!HAS_PCH_CPT(dev) || is_cpu_edp(intel_dp)) {
+		intel_dp->DP |= intel_dp->color_range;
+
+		if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
+			intel_dp->DP |= DP_SYNC_HS_HIGH;
+		if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
+			intel_dp->DP |= DP_SYNC_VS_HIGH;
+		intel_dp->DP |= DP_LINK_TRAIN_OFF;
+
+		if (intel_dp->link_configuration[1] & DP_LANE_COUNT_ENHANCED_FRAME_EN)
+			intel_dp->DP |= DP_ENHANCED_FRAMING;
+
+		if (intel_crtc->pipe == 1)
+			intel_dp->DP |= DP_PIPEB_SELECT;
+
+		if (is_cpu_edp(intel_dp)) {
+			/* don't miss out required setting for eDP */
+			intel_dp->DP |= DP_PLL_ENABLE;
+			if (adjusted_mode->clock < 200000)
+				intel_dp->DP |= DP_PLL_FREQ_160MHZ;
+			else
+				intel_dp->DP |= DP_PLL_FREQ_270MHZ;
+		}
+	} else {
+		intel_dp->DP |= DP_LINK_TRAIN_OFF_CPT;
 	}
 }
 
-static void ironlake_wait_panel_off(struct intel_dp *intel_dp)
+#define IDLE_ON_MASK		(PP_ON | 0 	  | PP_SEQUENCE_MASK | 0                     | PP_SEQUENCE_STATE_MASK)
+#define IDLE_ON_VALUE   	(PP_ON | 0 	  | PP_SEQUENCE_NONE | 0                     | PP_SEQUENCE_STATE_ON_IDLE)
+
+#define IDLE_OFF_MASK		(PP_ON | 0        | PP_SEQUENCE_MASK | 0                     | PP_SEQUENCE_STATE_MASK)
+#define IDLE_OFF_VALUE		(0     | 0        | PP_SEQUENCE_NONE | 0                     | PP_SEQUENCE_STATE_OFF_IDLE)
+
+#define IDLE_CYCLE_MASK		(PP_ON | 0        | PP_SEQUENCE_MASK | PP_CYCLE_DELAY_ACTIVE | PP_SEQUENCE_STATE_MASK)
+#define IDLE_CYCLE_VALUE	(0     | 0        | PP_SEQUENCE_NONE | 0                     | PP_SEQUENCE_STATE_OFF_IDLE)
+
+static void ironlake_wait_panel_status(struct intel_dp *intel_dp,
+				       u32 mask,
+				       u32 value)
 {
-	unsigned long	off_time;
-	unsigned long	delay;
+	struct drm_device *dev = intel_dp->base.base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	DRM_DEBUG_KMS("Wait for panel power off time\n");
+	DRM_DEBUG_KMS("mask %08x value %08x status %08x control %08x\n",
+		      mask, value,
+		      I915_READ(PCH_PP_STATUS),
+		      I915_READ(PCH_PP_CONTROL));
 
-	if (ironlake_edp_have_panel_power(intel_dp) ||
-	    ironlake_edp_have_panel_vdd(intel_dp))
-	{
-		DRM_DEBUG_KMS("Panel still on, no delay needed\n");
-		return;
+	if (_wait_for((I915_READ(PCH_PP_STATUS) & mask) == value, 5000, 10)) {
+		DRM_ERROR("Panel status timeout: status %08x control %08x\n",
+			  I915_READ(PCH_PP_STATUS),
+			  I915_READ(PCH_PP_CONTROL));
 	}
+}
 
-	off_time = intel_dp->panel_off_jiffies + msecs_to_jiffies(intel_dp->panel_power_down_delay);
-	if (time_after(jiffies, off_time)) {
-		DRM_DEBUG_KMS("Time already passed");
-		return;
-	}
-	delay = jiffies_to_msecs(off_time - jiffies);
-	if (delay > intel_dp->panel_power_down_delay)
-		delay = intel_dp->panel_power_down_delay;
-	DRM_DEBUG_KMS("Waiting an additional %ld ms\n", delay);
-	msleep(delay);
+static void ironlake_wait_panel_on(struct intel_dp *intel_dp)
+{
+	DRM_DEBUG_KMS("Wait for panel power on\n");
+	ironlake_wait_panel_status(intel_dp, IDLE_ON_MASK, IDLE_ON_VALUE);
+}
+
+static void ironlake_wait_panel_off(struct intel_dp *intel_dp)
+{
+	DRM_DEBUG_KMS("Wait for panel power off time\n");
+	ironlake_wait_panel_status(intel_dp, IDLE_OFF_MASK, IDLE_OFF_VALUE);
+}
+
+static void ironlake_wait_panel_power_cycle(struct intel_dp *intel_dp)
+{
+	DRM_DEBUG_KMS("Wait for panel power cycle\n");
+	ironlake_wait_panel_status(intel_dp, IDLE_CYCLE_MASK, IDLE_CYCLE_VALUE);
+}
+
+
+/* Read the current pp_control value, unlocking the register if it
+ * is locked
+ */
+
+static  u32 ironlake_get_pp_control(struct drm_i915_private *dev_priv)
+{
+	u32	control = I915_READ(PCH_PP_CONTROL);
+
+	control &= ~PANEL_UNLOCK_MASK;
+	control |= PANEL_UNLOCK_REGS;
+	return control;
 }
 
 static void ironlake_edp_panel_vdd_on(struct intel_dp *intel_dp)
@@ -921,15 +1006,16 @@ static void ironlake_edp_panel_vdd_on(struct intel_dp *intel_dp)
 	     "eDP VDD already requested on\n");
 
 	intel_dp->want_panel_vdd = true;
+
 	if (ironlake_edp_have_panel_vdd(intel_dp)) {
 		DRM_DEBUG_KMS("eDP VDD already on\n");
 		return;
 	}
 
-	ironlake_wait_panel_off(intel_dp);
-	pp = I915_READ(PCH_PP_CONTROL);
-	pp &= ~PANEL_UNLOCK_MASK;
-	pp |= PANEL_UNLOCK_REGS;
+	if (!ironlake_edp_have_panel_power(intel_dp))
+		ironlake_wait_panel_power_cycle(intel_dp);
+
+	pp = ironlake_get_pp_control(dev_priv);
 	pp |= EDP_FORCE_VDD;
 	I915_WRITE(PCH_PP_CONTROL, pp);
 	POSTING_READ(PCH_PP_CONTROL);
@@ -952,9 +1038,7 @@ static void ironlake_panel_vdd_off_sync(struct intel_dp *intel_dp)
 	u32 pp;
 
 	if (!intel_dp->want_panel_vdd && ironlake_edp_have_panel_vdd(intel_dp)) {
-		pp = I915_READ(PCH_PP_CONTROL);
-		pp &= ~PANEL_UNLOCK_MASK;
-		pp |= PANEL_UNLOCK_REGS;
+		pp = ironlake_get_pp_control(dev_priv);
 		pp &= ~EDP_FORCE_VDD;
 		I915_WRITE(PCH_PP_CONTROL, pp);
 		POSTING_READ(PCH_PP_CONTROL);
@@ -962,7 +1046,8 @@ static void ironlake_panel_vdd_off_sync(struct intel_dp *intel_dp)
 		/* Make sure sequencer is idle before allowing subsequent activity */
 		DRM_DEBUG_KMS("PCH_PP_STATUS: 0x%08x PCH_PP_CONTROL: 0x%08x\n",
 			      I915_READ(PCH_PP_STATUS), I915_READ(PCH_PP_CONTROL));
-		intel_dp->panel_off_jiffies = jiffies;
+
+		msleep(intel_dp->panel_power_down_delay);
 	}
 }
 
@@ -972,9 +1057,9 @@ static void ironlake_panel_vdd_work(struct work_struct *__work)
 						 struct intel_dp, panel_vdd_work);
 	struct drm_device *dev = intel_dp->base.base.dev;
 
-	mutex_lock(&dev->struct_mutex);
+	mutex_lock(&dev->mode_config.mutex);
 	ironlake_panel_vdd_off_sync(intel_dp);
-	mutex_unlock(&dev->struct_mutex);
+	mutex_unlock(&dev->mode_config.mutex);
 }
 
 static void ironlake_edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync)
@@ -984,7 +1069,7 @@ static void ironlake_edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync)
 
 	DRM_DEBUG_KMS("Turn eDP VDD off %d\n", intel_dp->want_panel_vdd);
 	WARN(!intel_dp->want_panel_vdd, "eDP VDD not forced on");
-	
+
 	intel_dp->want_panel_vdd = false;
 
 	if (sync) {
@@ -1000,23 +1085,25 @@ static void ironlake_edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync)
 	}
 }
 
-/* Returns true if the panel was already on when called */
 static void ironlake_edp_panel_on(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 pp, idle_on_mask = PP_ON | PP_SEQUENCE_STATE_ON_IDLE;
+	u32 pp;
 
 	if (!is_edp(intel_dp))
 		return;
-	if (ironlake_edp_have_panel_power(intel_dp))
+
+	DRM_DEBUG_KMS("Turn eDP power on\n");
+
+	if (ironlake_edp_have_panel_power(intel_dp)) {
+		DRM_DEBUG_KMS("eDP power already on\n");
 		return;
+	}
 
-	ironlake_wait_panel_off(intel_dp);
-	pp = I915_READ(PCH_PP_CONTROL);
-	pp &= ~PANEL_UNLOCK_MASK;
-	pp |= PANEL_UNLOCK_REGS;
+	ironlake_wait_panel_power_cycle(intel_dp);
 
+	pp = ironlake_get_pp_control(dev_priv);
 	if (IS_GEN5(dev)) {
 		/* ILK workaround: disable reset around power sequence */
 		pp &= ~PANEL_POWER_RESET;
@@ -1025,13 +1112,13 @@ static void ironlake_edp_panel_on(struct intel_dp *intel_dp)
 	}
 
 	pp |= POWER_TARGET_ON;
+	if (!IS_GEN5(dev))
+		pp |= PANEL_POWER_RESET;
+
 	I915_WRITE(PCH_PP_CONTROL, pp);
 	POSTING_READ(PCH_PP_CONTROL);
 
-	if (wait_for((I915_READ(PCH_PP_STATUS) & idle_on_mask) == idle_on_mask,
-		     5000))
-		DRM_ERROR("panel on wait timed out: 0x%08x\n",
-			  I915_READ(PCH_PP_STATUS));
+	ironlake_wait_panel_on(intel_dp);
 
 	if (IS_GEN5(dev)) {
 		pp |= PANEL_POWER_RESET; /* restore panel reset bit */
@@ -1040,46 +1127,25 @@ static void ironlake_edp_panel_on(struct intel_dp *intel_dp)
 	}
 }
 
-static void ironlake_edp_panel_off(struct drm_encoder *encoder)
+static void ironlake_edp_panel_off(struct intel_dp *intel_dp)
 {
-	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
-	struct drm_device *dev = encoder->dev;
+	struct drm_device *dev = intel_dp->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 pp, idle_off_mask = PP_ON | PP_SEQUENCE_MASK |
-		PP_CYCLE_DELAY_ACTIVE | PP_SEQUENCE_STATE_MASK;
+	u32 pp;
 
 	if (!is_edp(intel_dp))
 		return;
-	pp = I915_READ(PCH_PP_CONTROL);
-	pp &= ~PANEL_UNLOCK_MASK;
-	pp |= PANEL_UNLOCK_REGS;
 
-	if (IS_GEN5(dev)) {
-		/* ILK workaround: disable reset around power sequence */
-		pp &= ~PANEL_POWER_RESET;
-		I915_WRITE(PCH_PP_CONTROL, pp);
-		POSTING_READ(PCH_PP_CONTROL);
-	}
+	DRM_DEBUG_KMS("Turn eDP power off\n");
 
-	intel_dp->panel_off_jiffies = jiffies;
+	WARN(intel_dp->want_panel_vdd, "Cannot turn power off while VDD is on\n");
 
-	if (IS_GEN5(dev)) {
-		pp &= ~POWER_TARGET_ON;
-		I915_WRITE(PCH_PP_CONTROL, pp);
-		POSTING_READ(PCH_PP_CONTROL);
-		pp &= ~POWER_TARGET_ON;
-		I915_WRITE(PCH_PP_CONTROL, pp);
-		POSTING_READ(PCH_PP_CONTROL);
-		msleep(intel_dp->panel_power_cycle_delay);
-
-		if (wait_for((I915_READ(PCH_PP_STATUS) & idle_off_mask) == 0, 5000))
-			DRM_ERROR("panel off wait timed out: 0x%08x\n",
-				  I915_READ(PCH_PP_STATUS));
+	pp = ironlake_get_pp_control(dev_priv);
+	pp &= ~(POWER_TARGET_ON | EDP_FORCE_VDD | PANEL_POWER_RESET | EDP_BLC_ENABLE);
+	I915_WRITE(PCH_PP_CONTROL, pp);
+	POSTING_READ(PCH_PP_CONTROL);
 
-		pp |= PANEL_POWER_RESET; /* restore panel reset bit */
-		I915_WRITE(PCH_PP_CONTROL, pp);
-		POSTING_READ(PCH_PP_CONTROL);
-	}
+	ironlake_wait_panel_off(intel_dp);
 }
 
 static void ironlake_edp_backlight_on(struct intel_dp *intel_dp)
@@ -1099,9 +1165,7 @@ static void ironlake_edp_backlight_on(struct intel_dp *intel_dp)
 	 * allowing it to appear.
 	 */
 	msleep(intel_dp->backlight_on_delay);
-	pp = I915_READ(PCH_PP_CONTROL);
-	pp &= ~PANEL_UNLOCK_MASK;
-	pp |= PANEL_UNLOCK_REGS;
+	pp = ironlake_get_pp_control(dev_priv);
 	pp |= EDP_BLC_ENABLE;
 	I915_WRITE(PCH_PP_CONTROL, pp);
 	POSTING_READ(PCH_PP_CONTROL);
@@ -1117,9 +1181,7 @@ static void ironlake_edp_backlight_off(struct intel_dp *intel_dp)
 		return;
 
 	DRM_DEBUG_KMS("\n");
-	pp = I915_READ(PCH_PP_CONTROL);
-	pp &= ~PANEL_UNLOCK_MASK;
-	pp |= PANEL_UNLOCK_REGS;
+	pp = ironlake_get_pp_control(dev_priv);
 	pp &= ~EDP_BLC_ENABLE;
 	I915_WRITE(PCH_PP_CONTROL, pp);
 	POSTING_READ(PCH_PP_CONTROL);
@@ -1187,17 +1249,18 @@ static void intel_dp_prepare(struct drm_encoder *encoder)
 {
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 
+	ironlake_edp_backlight_off(intel_dp);
+	ironlake_edp_panel_off(intel_dp);
+
 	/* Wake up the sink first */
 	ironlake_edp_panel_vdd_on(intel_dp);
 	intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
+	intel_dp_link_down(intel_dp);
 	ironlake_edp_panel_vdd_off(intel_dp, false);
 
 	/* Make sure the panel is off before trying to
 	 * change the mode
 	 */
-	ironlake_edp_backlight_off(intel_dp);
-	intel_dp_link_down(intel_dp);
-	ironlake_edp_panel_off(encoder);
 }
 
 static void intel_dp_commit(struct drm_encoder *encoder)
@@ -1211,7 +1274,6 @@ static void intel_dp_commit(struct drm_encoder *encoder)
 	intel_dp_start_link_train(intel_dp);
 	ironlake_edp_panel_on(intel_dp);
 	ironlake_edp_panel_vdd_off(intel_dp, true);
-
 	intel_dp_complete_link_train(intel_dp);
 	ironlake_edp_backlight_on(intel_dp);
 
@@ -1230,16 +1292,20 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 	uint32_t dp_reg = I915_READ(intel_dp->output_reg);
 
 	if (mode != DRM_MODE_DPMS_ON) {
+		ironlake_edp_backlight_off(intel_dp);
+		ironlake_edp_panel_off(intel_dp);
+
 		ironlake_edp_panel_vdd_on(intel_dp);
-		if (is_edp(intel_dp))
-			ironlake_edp_backlight_off(intel_dp);
 		intel_dp_sink_dpms(intel_dp, mode);
 		intel_dp_link_down(intel_dp);
-		ironlake_edp_panel_off(encoder);
-		if (is_edp(intel_dp) && !is_pch_edp(intel_dp))
-			ironlake_edp_pll_off(encoder);
 		ironlake_edp_panel_vdd_off(intel_dp, false);
+
+		if (is_cpu_edp(intel_dp))
+			ironlake_edp_pll_off(encoder);
 	} else {
+		if (is_cpu_edp(intel_dp))
+			ironlake_edp_pll_on(encoder);
+
 		ironlake_edp_panel_vdd_on(intel_dp);
 		intel_dp_sink_dpms(intel_dp, mode);
 		if (!(dp_reg & DP_PORT_EN)) {
@@ -1247,7 +1313,6 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 			ironlake_edp_panel_on(intel_dp);
 			ironlake_edp_panel_vdd_off(intel_dp, true);
 			intel_dp_complete_link_train(intel_dp);
-			ironlake_edp_backlight_on(intel_dp);
 		} else
 			ironlake_edp_panel_vdd_off(intel_dp, false);
 		ironlake_edp_backlight_on(intel_dp);
@@ -1285,11 +1350,11 @@ intel_dp_aux_native_read_retry(struct intel_dp *intel_dp, uint16_t address,
  * link status information
  */
 static bool
-intel_dp_get_link_status(struct intel_dp *intel_dp)
+intel_dp_get_link_status(struct intel_dp *intel_dp, uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
 	return intel_dp_aux_native_read_retry(intel_dp,
 					      DP_LANE0_1_STATUS,
-					      intel_dp->link_status,
+					      link_status,
 					      DP_LINK_STATUS_SIZE);
 }
 
@@ -1301,27 +1366,25 @@ intel_dp_link_status(uint8_t link_status[DP_LINK_STATUS_SIZE],
 }
 
 static uint8_t
-intel_get_adjust_request_voltage(uint8_t link_status[DP_LINK_STATUS_SIZE],
+intel_get_adjust_request_voltage(uint8_t adjust_request[2],
 				 int lane)
 {
-	int	    i = DP_ADJUST_REQUEST_LANE0_1 + (lane >> 1);
 	int	    s = ((lane & 1) ?
 			 DP_ADJUST_VOLTAGE_SWING_LANE1_SHIFT :
 			 DP_ADJUST_VOLTAGE_SWING_LANE0_SHIFT);
-	uint8_t l = intel_dp_link_status(link_status, i);
+	uint8_t l = adjust_request[lane>>1];
 
 	return ((l >> s) & 3) << DP_TRAIN_VOLTAGE_SWING_SHIFT;
 }
 
 static uint8_t
-intel_get_adjust_request_pre_emphasis(uint8_t link_status[DP_LINK_STATUS_SIZE],
+intel_get_adjust_request_pre_emphasis(uint8_t adjust_request[2],
 				      int lane)
 {
-	int	    i = DP_ADJUST_REQUEST_LANE0_1 + (lane >> 1);
 	int	    s = ((lane & 1) ?
 			 DP_ADJUST_PRE_EMPHASIS_LANE1_SHIFT :
 			 DP_ADJUST_PRE_EMPHASIS_LANE0_SHIFT);
-	uint8_t l = intel_dp_link_status(link_status, i);
+	uint8_t l = adjust_request[lane>>1];
 
 	return ((l >> s) & 3) << DP_TRAIN_PRE_EMPHASIS_SHIFT;
 }
@@ -1343,34 +1406,63 @@ static char	*link_train_names[] = {
  * These are source-specific values; current Intel hardware supports
  * a maximum voltage of 800mV and a maximum pre-emphasis of 6dB
  */
-#define I830_DP_VOLTAGE_MAX	    DP_TRAIN_VOLTAGE_SWING_800
 
 static uint8_t
-intel_dp_pre_emphasis_max(uint8_t voltage_swing)
+intel_dp_voltage_max(struct intel_dp *intel_dp)
 {
-	switch (voltage_swing & DP_TRAIN_VOLTAGE_SWING_MASK) {
-	case DP_TRAIN_VOLTAGE_SWING_400:
-		return DP_TRAIN_PRE_EMPHASIS_6;
-	case DP_TRAIN_VOLTAGE_SWING_600:
-		return DP_TRAIN_PRE_EMPHASIS_6;
-	case DP_TRAIN_VOLTAGE_SWING_800:
-		return DP_TRAIN_PRE_EMPHASIS_3_5;
-	case DP_TRAIN_VOLTAGE_SWING_1200:
-	default:
-		return DP_TRAIN_PRE_EMPHASIS_0;
+	struct drm_device *dev = intel_dp->base.base.dev;
+
+	if (IS_GEN7(dev) && is_cpu_edp(intel_dp))
+		return DP_TRAIN_VOLTAGE_SWING_800;
+	else if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
+		return DP_TRAIN_VOLTAGE_SWING_1200;
+	else
+		return DP_TRAIN_VOLTAGE_SWING_800;
+}
+
+static uint8_t
+intel_dp_pre_emphasis_max(struct intel_dp *intel_dp, uint8_t voltage_swing)
+{
+	struct drm_device *dev = intel_dp->base.base.dev;
+
+	if (IS_GEN7(dev) && is_cpu_edp(intel_dp)) {
+		switch (voltage_swing & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			return DP_TRAIN_PRE_EMPHASIS_6;
+		case DP_TRAIN_VOLTAGE_SWING_600:
+		case DP_TRAIN_VOLTAGE_SWING_800:
+			return DP_TRAIN_PRE_EMPHASIS_3_5;
+		default:
+			return DP_TRAIN_PRE_EMPHASIS_0;
+		}
+	} else {
+		switch (voltage_swing & DP_TRAIN_VOLTAGE_SWING_MASK) {
+		case DP_TRAIN_VOLTAGE_SWING_400:
+			return DP_TRAIN_PRE_EMPHASIS_6;
+		case DP_TRAIN_VOLTAGE_SWING_600:
+			return DP_TRAIN_PRE_EMPHASIS_6;
+		case DP_TRAIN_VOLTAGE_SWING_800:
+			return DP_TRAIN_PRE_EMPHASIS_3_5;
+		case DP_TRAIN_VOLTAGE_SWING_1200:
+		default:
+			return DP_TRAIN_PRE_EMPHASIS_0;
+		}
 	}
 }
 
 static void
-intel_get_adjust_train(struct intel_dp *intel_dp)
+intel_get_adjust_train(struct intel_dp *intel_dp, uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
 	uint8_t v = 0;
 	uint8_t p = 0;
 	int lane;
+	uint8_t	*adjust_request = link_status + (DP_ADJUST_REQUEST_LANE0_1 - DP_LANE0_1_STATUS);
+	uint8_t voltage_max;
+	uint8_t preemph_max;
 
 	for (lane = 0; lane < intel_dp->lane_count; lane++) {
-		uint8_t this_v = intel_get_adjust_request_voltage(intel_dp->link_status, lane);
-		uint8_t this_p = intel_get_adjust_request_pre_emphasis(intel_dp->link_status, lane);
+		uint8_t this_v = intel_get_adjust_request_voltage(adjust_request, lane);
+		uint8_t this_p = intel_get_adjust_request_pre_emphasis(adjust_request, lane);
 
 		if (this_v > v)
 			v = this_v;
@@ -1378,18 +1470,20 @@ intel_get_adjust_train(struct intel_dp *intel_dp)
 			p = this_p;
 	}
 
-	if (v >= I830_DP_VOLTAGE_MAX)
-		v = I830_DP_VOLTAGE_MAX | DP_TRAIN_MAX_SWING_REACHED;
+	voltage_max = intel_dp_voltage_max(intel_dp);
+	if (v >= voltage_max)
+		v = voltage_max | DP_TRAIN_MAX_SWING_REACHED;
 
-	if (p >= intel_dp_pre_emphasis_max(v))
-		p = intel_dp_pre_emphasis_max(v) | DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
+	preemph_max = intel_dp_pre_emphasis_max(intel_dp, v);
+	if (p >= preemph_max)
+		p = preemph_max | DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
 
 	for (lane = 0; lane < 4; lane++)
 		intel_dp->train_set[lane] = v | p;
 }
 
 static uint32_t
-intel_dp_signal_levels(uint8_t train_set, int lane_count)
+intel_dp_signal_levels(uint8_t train_set)
 {
 	uint32_t	signal_levels = 0;
 
@@ -1454,13 +1548,43 @@ intel_gen6_edp_signal_levels(uint8_t train_set)
 	}
 }
 
+/* Gen7's DP voltage swing and pre-emphasis control */
+static uint32_t
+intel_gen7_edp_signal_levels(uint8_t train_set)
+{
+	int signal_levels = train_set & (DP_TRAIN_VOLTAGE_SWING_MASK |
+					 DP_TRAIN_PRE_EMPHASIS_MASK);
+	switch (signal_levels) {
+	case DP_TRAIN_VOLTAGE_SWING_400 | DP_TRAIN_PRE_EMPHASIS_0:
+		return EDP_LINK_TRAIN_400MV_0DB_IVB;
+	case DP_TRAIN_VOLTAGE_SWING_400 | DP_TRAIN_PRE_EMPHASIS_3_5:
+		return EDP_LINK_TRAIN_400MV_3_5DB_IVB;
+	case DP_TRAIN_VOLTAGE_SWING_400 | DP_TRAIN_PRE_EMPHASIS_6:
+		return EDP_LINK_TRAIN_400MV_6DB_IVB;
+
+	case DP_TRAIN_VOLTAGE_SWING_600 | DP_TRAIN_PRE_EMPHASIS_0:
+		return EDP_LINK_TRAIN_600MV_0DB_IVB;
+	case DP_TRAIN_VOLTAGE_SWING_600 | DP_TRAIN_PRE_EMPHASIS_3_5:
+		return EDP_LINK_TRAIN_600MV_3_5DB_IVB;
+
+	case DP_TRAIN_VOLTAGE_SWING_800 | DP_TRAIN_PRE_EMPHASIS_0:
+		return EDP_LINK_TRAIN_800MV_0DB_IVB;
+	case DP_TRAIN_VOLTAGE_SWING_800 | DP_TRAIN_PRE_EMPHASIS_3_5:
+		return EDP_LINK_TRAIN_800MV_3_5DB_IVB;
+
+	default:
+		DRM_DEBUG_KMS("Unsupported voltage swing/pre-emphasis level:"
+			      "0x%x\n", signal_levels);
+		return EDP_LINK_TRAIN_500MV_0DB_IVB;
+	}
+}
+
 static uint8_t
 intel_get_lane_status(uint8_t link_status[DP_LINK_STATUS_SIZE],
 		      int lane)
 {
-	int i = DP_LANE0_1_STATUS + (lane >> 1);
 	int s = (lane & 1) * 4;
-	uint8_t l = intel_dp_link_status(link_status, i);
+	uint8_t l = link_status[lane>>1];
 
 	return (l >> s) & 0xf;
 }
@@ -1485,18 +1609,18 @@ intel_clock_recovery_ok(uint8_t link_status[DP_LINK_STATUS_SIZE], int lane_count
 			 DP_LANE_CHANNEL_EQ_DONE|\
 			 DP_LANE_SYMBOL_LOCKED)
 static bool
-intel_channel_eq_ok(struct intel_dp *intel_dp)
+intel_channel_eq_ok(struct intel_dp *intel_dp, uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
 	uint8_t lane_align;
 	uint8_t lane_status;
 	int lane;
 
-	lane_align = intel_dp_link_status(intel_dp->link_status,
+	lane_align = intel_dp_link_status(link_status,
 					  DP_LANE_ALIGN_STATUS_UPDATED);
 	if ((lane_align & DP_INTERLANE_ALIGN_DONE) == 0)
 		return false;
 	for (lane = 0; lane < intel_dp->lane_count; lane++) {
-		lane_status = intel_get_lane_status(intel_dp->link_status, lane);
+		lane_status = intel_get_lane_status(link_status, lane);
 		if ((lane_status & CHANNEL_EQ_BITS) != CHANNEL_EQ_BITS)
 			return false;
 	}
@@ -1521,8 +1645,9 @@ intel_dp_set_link_train(struct intel_dp *intel_dp,
 
 	ret = intel_dp_aux_native_write(intel_dp,
 					DP_TRAINING_LANE0_SET,
-					intel_dp->train_set, 4);
-	if (ret != 4)
+					intel_dp->train_set,
+					intel_dp->lane_count);
+	if (ret != intel_dp->lane_count)
 		return false;
 
 	return true;
@@ -1538,7 +1663,7 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 	int i;
 	uint8_t voltage;
 	bool clock_recovery = false;
-	int tries;
+	int voltage_tries, loop_tries;
 	u32 reg;
 	uint32_t DP = intel_dp->DP;
 
@@ -1559,26 +1684,35 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 				  DP_LINK_CONFIGURATION_SIZE);
 
 	DP |= DP_PORT_EN;
-	if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
+
+	if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
 		DP &= ~DP_LINK_TRAIN_MASK_CPT;
 	else
 		DP &= ~DP_LINK_TRAIN_MASK;
 	memset(intel_dp->train_set, 0, 4);
 	voltage = 0xff;
-	tries = 0;
+	voltage_tries = 0;
+	loop_tries = 0;
 	clock_recovery = false;
 	for (;;) {
 		/* Use intel_dp->train_set[0] to set the voltage and pre emphasis values */
+		uint8_t	    link_status[DP_LINK_STATUS_SIZE];
 		uint32_t    signal_levels;
-		if (IS_GEN6(dev) && is_edp(intel_dp)) {
+
+
+		if (IS_GEN7(dev) && is_cpu_edp(intel_dp)) {
+			signal_levels = intel_gen7_edp_signal_levels(intel_dp->train_set[0]);
+			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_IVB) | signal_levels;
+		} else if (IS_GEN6(dev) && is_cpu_edp(intel_dp)) {
 			signal_levels = intel_gen6_edp_signal_levels(intel_dp->train_set[0]);
 			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_SNB) | signal_levels;
 		} else {
-			signal_levels = intel_dp_signal_levels(intel_dp->train_set[0], intel_dp->lane_count);
+			signal_levels = intel_dp_signal_levels(intel_dp->train_set[0]);
+			DRM_DEBUG_KMS("training pattern 1 signal levels %08x\n", signal_levels);
 			DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
 		}
 
-		if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
+		if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
 			reg = DP | DP_LINK_TRAIN_PAT_1_CPT;
 		else
 			reg = DP | DP_LINK_TRAIN_PAT_1;
@@ -1590,10 +1724,13 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 		/* Set training pattern 1 */
 
 		udelay(100);
-		if (!intel_dp_get_link_status(intel_dp))
+		if (!intel_dp_get_link_status(intel_dp, link_status)) {
+			DRM_ERROR("failed to get link status\n");
 			break;
+		}
 
-		if (intel_clock_recovery_ok(intel_dp->link_status, intel_dp->lane_count)) {
+		if (intel_clock_recovery_ok(link_status, intel_dp->lane_count)) {
+			DRM_DEBUG_KMS("clock recovery OK\n");
 			clock_recovery = true;
 			break;
 		}
@@ -1602,20 +1739,30 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 		for (i = 0; i < intel_dp->lane_count; i++)
 			if ((intel_dp->train_set[i] & DP_TRAIN_MAX_SWING_REACHED) == 0)
 				break;
-		if (i == intel_dp->lane_count)
-			break;
+		if (i == intel_dp->lane_count) {
+			++loop_tries;
+			if (loop_tries == 5) {
+				DRM_DEBUG_KMS("too many full retries, give up\n");
+				break;
+			}
+			memset(intel_dp->train_set, 0, 4);
+			voltage_tries = 0;
+			continue;
+		}
 
 		/* Check to see if we've tried the same voltage 5 times */
 		if ((intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK) == voltage) {
-			++tries;
-			if (tries == 5)
+			++voltage_tries;
+			if (voltage_tries == 5) {
+				DRM_DEBUG_KMS("too many voltage retries, give up\n");
 				break;
+			}
 		} else
-			tries = 0;
+			voltage_tries = 0;
 		voltage = intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK;
 
 		/* Compute new intel_dp->train_set as requested by target */
-		intel_get_adjust_train(intel_dp);
+		intel_get_adjust_train(intel_dp, link_status);
 	}
 
 	intel_dp->DP = DP;
@@ -1638,6 +1785,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 	for (;;) {
 		/* Use intel_dp->train_set[0] to set the voltage and pre emphasis values */
 		uint32_t    signal_levels;
+		uint8_t	    link_status[DP_LINK_STATUS_SIZE];
 
 		if (cr_tries > 5) {
 			DRM_ERROR("failed to train DP, aborting\n");
@@ -1645,15 +1793,18 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 			break;
 		}
 
-		if (IS_GEN6(dev) && is_edp(intel_dp)) {
+		if (IS_GEN7(dev) && is_cpu_edp(intel_dp)) {
+			signal_levels = intel_gen7_edp_signal_levels(intel_dp->train_set[0]);
+			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_IVB) | signal_levels;
+		} else if (IS_GEN6(dev) && is_cpu_edp(intel_dp)) {
 			signal_levels = intel_gen6_edp_signal_levels(intel_dp->train_set[0]);
 			DP = (DP & ~EDP_LINK_TRAIN_VOL_EMP_MASK_SNB) | signal_levels;
 		} else {
-			signal_levels = intel_dp_signal_levels(intel_dp->train_set[0], intel_dp->lane_count);
+			signal_levels = intel_dp_signal_levels(intel_dp->train_set[0]);
 			DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
 		}
 
-		if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
+		if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
 			reg = DP | DP_LINK_TRAIN_PAT_2_CPT;
 		else
 			reg = DP | DP_LINK_TRAIN_PAT_2;
@@ -1665,17 +1816,17 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 			break;
 
 		udelay(400);
-		if (!intel_dp_get_link_status(intel_dp))
+		if (!intel_dp_get_link_status(intel_dp, link_status))
 			break;
 
 		/* Make sure clock is still ok */
-		if (!intel_clock_recovery_ok(intel_dp->link_status, intel_dp->lane_count)) {
+		if (!intel_clock_recovery_ok(link_status, intel_dp->lane_count)) {
 			intel_dp_start_link_train(intel_dp);
 			cr_tries++;
 			continue;
 		}
 
-		if (intel_channel_eq_ok(intel_dp)) {
+		if (intel_channel_eq_ok(intel_dp, link_status)) {
 			channel_eq = true;
 			break;
 		}
@@ -1690,11 +1841,11 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 		}
 
 		/* Compute new intel_dp->train_set as requested by target */
-		intel_get_adjust_train(intel_dp);
+		intel_get_adjust_train(intel_dp, link_status);
 		++tries;
 	}
 
-	if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
+	if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
 		reg = DP | DP_LINK_TRAIN_OFF_CPT;
 	else
 		reg = DP | DP_LINK_TRAIN_OFF;
@@ -1724,7 +1875,7 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 		udelay(100);
 	}
 
-	if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp)) {
+	if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp))) {
 		DP &= ~DP_LINK_TRAIN_MASK_CPT;
 		I915_WRITE(intel_dp->output_reg, DP | DP_LINK_TRAIN_PAT_IDLE_CPT);
 	} else {
@@ -1735,8 +1886,12 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 
 	msleep(17);
 
-	if (is_edp(intel_dp))
-		DP |= DP_LINK_TRAIN_OFF;
+	if (is_edp(intel_dp)) {
+		if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
+			DP |= DP_LINK_TRAIN_OFF_CPT;
+		else
+			DP |= DP_LINK_TRAIN_OFF;
+	}
 
 	if (!HAS_PCH_CPT(dev) &&
 	    I915_READ(intel_dp->output_reg) & DP_PIPEB_SELECT) {
@@ -1822,6 +1977,7 @@ static void
 intel_dp_check_link_status(struct intel_dp *intel_dp)
 {
 	u8 sink_irq_vector;
+	u8 link_status[DP_LINK_STATUS_SIZE];
 
 	if (intel_dp->dpms_mode != DRM_MODE_DPMS_ON)
 		return;
@@ -1830,7 +1986,7 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 		return;
 
 	/* Try to read receiver status if the link appears to be up */
-	if (!intel_dp_get_link_status(intel_dp)) {
+	if (!intel_dp_get_link_status(intel_dp, link_status)) {
 		intel_dp_link_down(intel_dp);
 		return;
 	}
@@ -1855,7 +2011,7 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 			DRM_DEBUG_DRIVER("CP or sink specific irq unhandled\n");
 	}
 
-	if (!intel_channel_eq_ok(intel_dp)) {
+	if (!intel_channel_eq_ok(intel_dp, link_status)) {
 		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
 			      drm_get_encoder_name(&intel_dp->base.base));
 		intel_dp_start_link_train(intel_dp);
@@ -2179,7 +2335,8 @@ intel_trans_dp_port_sel(struct drm_crtc *crtc)
 			continue;
 
 		intel_dp = enc_to_intel_dp(encoder);
-		if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT)
+		if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT ||
+		    intel_dp->base.type == INTEL_OUTPUT_EDP)
 			return intel_dp->output_reg;
 	}
 
@@ -2321,7 +2478,7 @@ intel_dp_init(struct drm_device *dev, int output_reg)
 
 		cur.t8 = (pp_on & PANEL_LIGHT_ON_DELAY_MASK) >>
 			PANEL_LIGHT_ON_DELAY_SHIFT;
-		
+
 		cur.t9 = (pp_off & PANEL_LIGHT_OFF_DELAY_MASK) >>
 			PANEL_LIGHT_OFF_DELAY_SHIFT;
 
@@ -2354,11 +2511,10 @@ intel_dp_init(struct drm_device *dev, int output_reg)
 		DRM_DEBUG_KMS("backlight on delay %d, off delay %d\n",
 			      intel_dp->backlight_on_delay, intel_dp->backlight_off_delay);
 
-		intel_dp->panel_off_jiffies = jiffies - intel_dp->panel_power_down_delay;
-
 		ironlake_edp_panel_vdd_on(intel_dp);
 		ret = intel_dp_get_dpcd(intel_dp);
 		ironlake_edp_panel_vdd_off(intel_dp, false);
+
 		if (ret) {
 			if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11)
 				dev_priv->no_aux_handshake =
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 23c56221fe8f..82a459bfccbc 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -110,6 +110,7 @@
 /* drm_display_mode->private_flags */
 #define INTEL_MODE_PIXEL_MULTIPLIER_SHIFT (0x0)
 #define INTEL_MODE_PIXEL_MULTIPLIER_MASK (0xf << INTEL_MODE_PIXEL_MULTIPLIER_SHIFT)
+#define INTEL_MODE_DP_FORCE_6BPC (0x10)
 
 static inline void
 intel_mode_set_pixel_multiplier(struct drm_display_mode *mode,
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 42f165a520de..e44191132ac4 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -715,6 +715,14 @@ static const struct dmi_system_id intel_no_lvds[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "EB1007"),
 		},
 	},
+	{
+		.callback = intel_no_lvds_dmi_callback,
+		.ident = "Asus AT5NM10T-I",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "AT5NM10T-I"),
+		},
+	},
 
 	{ }	/* terminating entry */
 };
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 499d4c0dbeeb..04d79fd1dc9d 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -178,13 +178,10 @@ u32 intel_panel_get_max_backlight(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev)) {
 		max >>= 16;
 	} else {
-		if (IS_PINEVIEW(dev)) {
+		if (INTEL_INFO(dev)->gen < 4)
 			max >>= 17;
-		} else {
+		else
 			max >>= 16;
-			if (INTEL_INFO(dev)->gen < 4)
-				max &= ~1;
-		}
 
 		if (is_backlight_combination_mode(dev))
 			max *= 0xff;
@@ -203,13 +200,12 @@ u32 intel_panel_get_backlight(struct drm_device *dev)
 		val = I915_READ(BLC_PWM_CPU_CTL) & BACKLIGHT_DUTY_CYCLE_MASK;
 	} else {
 		val = I915_READ(BLC_PWM_CTL) & BACKLIGHT_DUTY_CYCLE_MASK;
-		if (IS_PINEVIEW(dev))
+		if (INTEL_INFO(dev)->gen < 4)
 			val >>= 1;
 
 		if (is_backlight_combination_mode(dev)) {
 			u8 lbpc;
 
-			val &= ~1;
 			pci_read_config_byte(dev->pdev, PCI_LBPC, &lbpc);
 			val *= lbpc;
 		}
@@ -246,11 +242,9 @@ static void intel_panel_actually_set_backlight(struct drm_device *dev, u32 level
 	}
 
 	tmp = I915_READ(BLC_PWM_CTL);
-	if (IS_PINEVIEW(dev)) {
-		tmp &= ~(BACKLIGHT_DUTY_CYCLE_MASK - 1);
+	if (INTEL_INFO(dev)->gen < 4) 
 		level <<= 1;
-	} else
-		tmp &= ~BACKLIGHT_DUTY_CYCLE_MASK;
+	tmp &= ~BACKLIGHT_DUTY_CYCLE_MASK;
 	I915_WRITE(BLC_PWM_CTL, tmp | level);
 }
 
@@ -326,7 +320,8 @@ static int intel_panel_update_status(struct backlight_device *bd)
 static int intel_panel_get_brightness(struct backlight_device *bd)
 {
 	struct drm_device *dev = bl_get_data(bd);
-	return intel_panel_get_backlight(dev);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	return dev_priv->backlight_level;
 }
 
 static const struct backlight_ops intel_panel_bl_ops = {
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 3003fb25aefd..f7b9268df266 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -50,6 +50,7 @@
 #define IS_TMDS(c)	(c->output_flag & SDVO_TMDS_MASK)
 #define IS_LVDS(c)	(c->output_flag & SDVO_LVDS_MASK)
 #define IS_TV_OR_LVDS(c) (c->output_flag & (SDVO_TV_MASK | SDVO_LVDS_MASK))
+#define IS_DIGITAL(c) (c->output_flag & (SDVO_TMDS_MASK | SDVO_LVDS_MASK))
 
 
 static const char *tv_format_names[] = {
@@ -1086,8 +1087,12 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder,
 		}
 		sdvox |= (9 << 19) | SDVO_BORDER_ENABLE;
 	}
-	if (intel_crtc->pipe == 1)
-		sdvox |= SDVO_PIPE_B_SELECT;
+
+	if (INTEL_PCH_TYPE(dev) >= PCH_CPT)
+		sdvox |= TRANSCODER_CPT(intel_crtc->pipe);
+	else
+		sdvox |= TRANSCODER(intel_crtc->pipe);
+
 	if (intel_sdvo->has_hdmi_audio)
 		sdvox |= SDVO_AUDIO_ENABLE;
 
@@ -1314,6 +1319,18 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector)
 	return status;
 }
 
+static bool
+intel_sdvo_connector_matches_edid(struct intel_sdvo_connector *sdvo,
+				  struct edid *edid)
+{
+	bool monitor_is_digital = !!(edid->input & DRM_EDID_INPUT_DIGITAL);
+	bool connector_is_digital = !!IS_DIGITAL(sdvo);
+
+	DRM_DEBUG_KMS("connector_is_digital? %d, monitor_is_digital? %d\n",
+		      connector_is_digital, monitor_is_digital);
+	return connector_is_digital == monitor_is_digital;
+}
+
 static enum drm_connector_status
 intel_sdvo_detect(struct drm_connector *connector, bool force)
 {
@@ -1358,10 +1375,12 @@ intel_sdvo_detect(struct drm_connector *connector, bool force)
 		if (edid == NULL)
 			edid = intel_sdvo_get_analog_edid(connector);
 		if (edid != NULL) {
-			if (edid->input & DRM_EDID_INPUT_DIGITAL)
-				ret = connector_status_disconnected;
-			else
+			if (intel_sdvo_connector_matches_edid(intel_sdvo_connector,
+							      edid))
 				ret = connector_status_connected;
+			else
+				ret = connector_status_disconnected;
+
 			connector->display_info.raw_edid = NULL;
 			kfree(edid);
 		} else
@@ -1402,11 +1421,8 @@ static void intel_sdvo_get_ddc_modes(struct drm_connector *connector)
 		edid = intel_sdvo_get_analog_edid(connector);
 
 	if (edid != NULL) {
-		struct intel_sdvo_connector *intel_sdvo_connector = to_intel_sdvo_connector(connector);
-		bool monitor_is_digital = !!(edid->input & DRM_EDID_INPUT_DIGITAL);
-		bool connector_is_digital = !!IS_TMDS(intel_sdvo_connector);
-
-		if (connector_is_digital == monitor_is_digital) {
+		if (intel_sdvo_connector_matches_edid(to_intel_sdvo_connector(connector),
+						      edid)) {
 			drm_mode_connector_update_edid_property(connector, edid);
 			drm_add_edid_modes(connector, edid);
 		}
diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.c b/drivers/gpu/drm/nouveau/nouveau_bios.c
index 032a82098136..5fc201b49d30 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bios.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bios.c
@@ -640,10 +640,9 @@ static int
 nv50_pll_set(struct drm_device *dev, uint32_t reg, uint32_t clk)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	uint32_t reg0 = nv_rd32(dev, reg + 0);
-	uint32_t reg1 = nv_rd32(dev, reg + 4);
 	struct nouveau_pll_vals pll;
 	struct pll_lims pll_limits;
+	u32 ctrl, mask, coef;
 	int ret;
 
 	ret = get_pll_limits(dev, reg, &pll_limits);
@@ -654,15 +653,20 @@ nv50_pll_set(struct drm_device *dev, uint32_t reg, uint32_t clk)
 	if (!clk)
 		return -ERANGE;
 
-	reg0 = (reg0 & 0xfff8ffff) | (pll.log2P << 16);
-	reg1 = (reg1 & 0xffff0000) | (pll.N1 << 8) | pll.M1;
-
-	if (dev_priv->vbios.execute) {
-		still_alive();
-		nv_wr32(dev, reg + 4, reg1);
-		nv_wr32(dev, reg + 0, reg0);
+	coef = pll.N1 << 8 | pll.M1;
+	ctrl = pll.log2P << 16;
+	mask = 0x00070000;
+	if (reg == 0x004008) {
+		mask |= 0x01f80000;
+		ctrl |= (pll_limits.log2p_bias << 19);
+		ctrl |= (pll.log2P << 22);
 	}
 
+	if (!dev_priv->vbios.execute)
+		return 0;
+
+	nv_mask(dev, reg + 0, mask, ctrl);
+	nv_wr32(dev, reg + 4, coef);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index b1b33a108b31..f12dd0f39211 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -153,7 +153,7 @@ set_placement_range(struct nouveau_bo *nvbo, uint32_t type)
 
 	if (dev_priv->card_type == NV_10 &&
 	    nvbo->tile_mode && (type & TTM_PL_FLAG_VRAM) &&
-	    nvbo->bo.mem.num_pages < vram_pages / 2) {
+	    nvbo->bo.mem.num_pages < vram_pages / 4) {
 		/*
 		 * Make sure that the color and depth buffers are handled
 		 * by independent memory controller units. Up to a 9x
diff --git a/drivers/gpu/drm/nouveau/nouveau_channel.c b/drivers/gpu/drm/nouveau/nouveau_channel.c
index a319d5646ea9..bb6ec9ef8676 100644
--- a/drivers/gpu/drm/nouveau/nouveau_channel.c
+++ b/drivers/gpu/drm/nouveau/nouveau_channel.c
@@ -158,6 +158,7 @@ nouveau_channel_alloc(struct drm_device *dev, struct nouveau_channel **chan_ret,
 	INIT_LIST_HEAD(&chan->nvsw.vbl_wait);
 	INIT_LIST_HEAD(&chan->nvsw.flip);
 	INIT_LIST_HEAD(&chan->fence.pending);
+	spin_lock_init(&chan->fence.lock);
 
 	/* setup channel's memory and vm */
 	ret = nouveau_gpuobj_channel_init(chan, vram_handle, gart_handle);
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index e0d275e1c96c..cea6696b1906 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -710,7 +710,7 @@ nouveau_connector_mode_valid(struct drm_connector *connector,
 	case OUTPUT_DP:
 		max_clock  = nv_encoder->dp.link_nr;
 		max_clock *= nv_encoder->dp.link_bw;
-		clock = clock * nouveau_connector_bpp(connector) / 8;
+		clock = clock * nouveau_connector_bpp(connector) / 10;
 		break;
 	default:
 		BUG_ON(1);
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 2531ef54c3e9..7e88cd7f2b99 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -369,3 +369,48 @@ nouveau_finish_page_flip(struct nouveau_channel *chan,
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 	return 0;
 }
+
+int
+nouveau_display_dumb_create(struct drm_file *file_priv, struct drm_device *dev,
+			    struct drm_mode_create_dumb *args)
+{
+	struct nouveau_bo *bo;
+	int ret;
+
+	args->pitch = roundup(args->width * (args->bpp / 8), 256);
+	args->size = args->pitch * args->height;
+	args->size = roundup(args->size, PAGE_SIZE);
+
+	ret = nouveau_gem_new(dev, args->size, 0, TTM_PL_FLAG_VRAM, 0, 0, &bo);
+	if (ret)
+		return ret;
+
+	ret = drm_gem_handle_create(file_priv, bo->gem, &args->handle);
+	drm_gem_object_unreference_unlocked(bo->gem);
+	return ret;
+}
+
+int
+nouveau_display_dumb_destroy(struct drm_file *file_priv, struct drm_device *dev,
+			     uint32_t handle)
+{
+	return drm_gem_handle_delete(file_priv, handle);
+}
+
+int
+nouveau_display_dumb_map_offset(struct drm_file *file_priv,
+				struct drm_device *dev,
+				uint32_t handle, uint64_t *poffset)
+{
+	struct drm_gem_object *gem;
+
+	gem = drm_gem_object_lookup(dev, file_priv, handle);
+	if (gem) {
+		struct nouveau_bo *bo = gem->driver_private;
+		*poffset = bo->bo.addr_space_offset;
+		drm_gem_object_unreference_unlocked(gem);
+		return 0;
+	}
+
+	return -ENOENT;
+}
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.c b/drivers/gpu/drm/nouveau/nouveau_drv.c
index d661bc5e3945..f0a60afac446 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.c
@@ -434,6 +434,10 @@ static struct drm_driver driver = {
 	.gem_open_object = nouveau_gem_object_open,
 	.gem_close_object = nouveau_gem_object_close,
 
+	.dumb_create = nouveau_display_dumb_create,
+	.dumb_map_offset = nouveau_display_dumb_map_offset,
+	.dumb_destroy = nouveau_display_dumb_destroy,
+
 	.name = DRIVER_NAME,
 	.desc = DRIVER_DESC,
 #ifdef GIT_REVISION
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index 0c53e39fc6c9..dfddb7e078a1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -1421,6 +1421,12 @@ int nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 			   struct drm_pending_vblank_event *event);
 int nouveau_finish_page_flip(struct nouveau_channel *,
 			     struct nouveau_page_flip_state *);
+int nouveau_display_dumb_create(struct drm_file *, struct drm_device *,
+				struct drm_mode_create_dumb *args);
+int nouveau_display_dumb_map_offset(struct drm_file *, struct drm_device *,
+				    uint32_t handle, uint64_t *offset);
+int nouveau_display_dumb_destroy(struct drm_file *, struct drm_device *,
+				 uint32_t handle);
 
 /* nv10_gpio.c */
 int nv10_gpio_get(struct drm_device *dev, enum dcb_gpio_tag tag);
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index defffd140781..dbb151834121 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -488,6 +488,7 @@ int nouveau_fbcon_init(struct drm_device *dev)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nouveau_fbdev *nfbdev;
+	int preferred_bpp;
 	int ret;
 
 	nfbdev = kzalloc(sizeof(struct nouveau_fbdev), GFP_KERNEL);
@@ -506,7 +507,15 @@ int nouveau_fbcon_init(struct drm_device *dev)
 	}
 
 	drm_fb_helper_single_add_all_connectors(&nfbdev->helper);
-	drm_fb_helper_initial_config(&nfbdev->helper, 32);
+
+	if (dev_priv->vram_size <= 32 * 1024 * 1024)
+		preferred_bpp = 8;
+	else if (dev_priv->vram_size <= 64 * 1024 * 1024)
+		preferred_bpp = 16;
+	else
+		preferred_bpp = 32;
+
+	drm_fb_helper_initial_config(&nfbdev->helper, preferred_bpp);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c
index 81116cfea275..2f6daae68b9d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.c
@@ -539,8 +539,6 @@ nouveau_fence_channel_init(struct nouveau_channel *chan)
 			return ret;
 	}
 
-	INIT_LIST_HEAD(&chan->fence.pending);
-	spin_lock_init(&chan->fence.lock);
 	atomic_set(&chan->fence.last_sequence_irq, 0);
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_i2c.c b/drivers/gpu/drm/nouveau/nouveau_i2c.c
index c6143df48b9f..d39b2202b197 100644
--- a/drivers/gpu/drm/nouveau/nouveau_i2c.c
+++ b/drivers/gpu/drm/nouveau/nouveau_i2c.c
@@ -333,7 +333,7 @@ nouveau_i2c_identify(struct drm_device *dev, const char *what,
 
 	NV_DEBUG(dev, "Probing %ss on I2C bus: %d\n", what, index);
 
-	for (i = 0; info[i].addr; i++) {
+	for (i = 0; i2c && info[i].addr; i++) {
 		if (nouveau_probe_i2c_addr(i2c, info[i].addr) &&
 		    (!match || match(i2c, &info[i]))) {
 			NV_INFO(dev, "Detected %s: %s\n", what, info[i].type);
diff --git a/drivers/gpu/drm/nouveau/nouveau_object.c b/drivers/gpu/drm/nouveau/nouveau_object.c
index 02222c540aee..960c0ae0c0c3 100644
--- a/drivers/gpu/drm/nouveau/nouveau_object.c
+++ b/drivers/gpu/drm/nouveau/nouveau_object.c
@@ -680,7 +680,7 @@ nouveau_gpuobj_channel_init_pramin(struct nouveau_channel *chan)
 		return ret;
 	}
 
-	ret = drm_mm_init(&chan->ramin_heap, base, size);
+	ret = drm_mm_init(&chan->ramin_heap, base, size - base);
 	if (ret) {
 		NV_ERROR(dev, "Error creating PRAMIN heap: %d\n", ret);
 		nouveau_gpuobj_ref(NULL, &chan->ramin);
diff --git a/drivers/gpu/drm/nouveau/nouveau_perf.c b/drivers/gpu/drm/nouveau/nouveau_perf.c
index 9f178aa94162..33d03fbf00df 100644
--- a/drivers/gpu/drm/nouveau/nouveau_perf.c
+++ b/drivers/gpu/drm/nouveau/nouveau_perf.c
@@ -239,7 +239,7 @@ nouveau_perf_init(struct drm_device *dev)
 	if(version == 0x15) {
 		memtimings->timing =
 				kcalloc(entries, sizeof(*memtimings->timing), GFP_KERNEL);
-		if(!memtimings) {
+		if (!memtimings->timing) {
 			NV_WARN(dev,"Could not allocate memtiming table\n");
 			return;
 		}
diff --git a/drivers/gpu/drm/nouveau/nouveau_state.c b/drivers/gpu/drm/nouveau/nouveau_state.c
index 82478e0998e5..d8831ab42bb9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_state.c
+++ b/drivers/gpu/drm/nouveau/nouveau_state.c
@@ -579,6 +579,14 @@ nouveau_card_init(struct drm_device *dev)
 	if (ret)
 		goto out_display_early;
 
+	/* workaround an odd issue on nvc1 by disabling the device's
+	 * nosnoop capability.  hopefully won't cause issues until a
+	 * better fix is found - assuming there is one...
+	 */
+	if (dev_priv->chipset == 0xc1) {
+		nv_mask(dev, 0x00088080, 0x00000800, 0x00000000);
+	}
+
 	nouveau_pm_init(dev);
 
 	ret = engine->vram.init(dev);
@@ -1102,12 +1110,13 @@ int nouveau_load(struct drm_device *dev, unsigned long flags)
 	dev_priv->noaccel = !!nouveau_noaccel;
 	if (nouveau_noaccel == -1) {
 		switch (dev_priv->chipset) {
-		case 0xc1: /* known broken */
-		case 0xc8: /* never tested */
+#if 0
+		case 0xXX: /* known broken */
 			NV_INFO(dev, "acceleration disabled by default, pass "
 				     "noaccel=0 to force enable\n");
 			dev_priv->noaccel = true;
 			break;
+#endif
 		default:
 			dev_priv->noaccel = false;
 			break;
diff --git a/drivers/gpu/drm/nouveau/nv40_pm.c b/drivers/gpu/drm/nouveau/nv40_pm.c
index bbc0b9c7e1f7..e676b0d53478 100644
--- a/drivers/gpu/drm/nouveau/nv40_pm.c
+++ b/drivers/gpu/drm/nouveau/nv40_pm.c
@@ -57,12 +57,14 @@ read_pll_2(struct drm_device *dev, u32 reg)
 	int P = (ctrl & 0x00070000) >> 16;
 	u32 ref = 27000, clk = 0;
 
-	if (ctrl & 0x80000000)
+	if ((ctrl & 0x80000000) && M1) {
 		clk = ref * N1 / M1;
-
-	if (!(ctrl & 0x00000100)) {
-		if (ctrl & 0x40000000)
-			clk = clk * N2 / M2;
+		if ((ctrl & 0x40000100) == 0x40000000) {
+			if (M2)
+				clk = clk * N2 / M2;
+			else
+				clk = 0;
+		}
 	}
 
 	return clk >> P;
@@ -177,6 +179,11 @@ nv40_pm_clocks_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl)
 	}
 
 	/* memory clock */
+	if (!perflvl->memory) {
+		info->mpll_ctrl = 0x00000000;
+		goto out;
+	}
+
 	ret = nv40_calc_pll(dev, 0x004020, &pll, perflvl->memory,
 			    &N1, &M1, &N2, &M2, &log2P);
 	if (ret < 0)
@@ -264,6 +271,9 @@ nv40_pm_clocks_set(struct drm_device *dev, void *pre_state)
 	mdelay(5);
 	nv_mask(dev, 0x00c040, 0x00000333, info->ctrl);
 
+	if (!info->mpll_ctrl)
+		goto resume;
+
 	/* wait for vblank start on active crtcs, disable memory access */
 	for (i = 0; i < 2; i++) {
 		if (!(crtc_mask & (1 << i)))
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index d23ca00e7d62..06de250fe617 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -616,7 +616,7 @@ nv50_display_unk10_handler(struct drm_device *dev)
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nv50_display *disp = nv50_display(dev);
 	u32 unk30 = nv_rd32(dev, 0x610030), mc;
-	int i, crtc, or, type = OUTPUT_ANY;
+	int i, crtc, or = 0, type = OUTPUT_ANY;
 
 	NV_DEBUG_KMS(dev, "0x610030: 0x%08x\n", unk30);
 	disp->irq.dcb = NULL;
@@ -708,7 +708,7 @@ nv50_display_unk20_handler(struct drm_device *dev)
 	struct nv50_display *disp = nv50_display(dev);
 	u32 unk30 = nv_rd32(dev, 0x610030), tmp, pclk, script, mc = 0;
 	struct dcb_entry *dcb;
-	int i, crtc, or, type = OUTPUT_ANY;
+	int i, crtc, or = 0, type = OUTPUT_ANY;
 
 	NV_DEBUG_KMS(dev, "0x610030: 0x%08x\n", unk30);
 	dcb = disp->irq.dcb;
diff --git a/drivers/gpu/drm/nouveau/nv50_graph.c b/drivers/gpu/drm/nouveau/nv50_graph.c
index 8c979b31ff61..ac601f7c4e1a 100644
--- a/drivers/gpu/drm/nouveau/nv50_graph.c
+++ b/drivers/gpu/drm/nouveau/nv50_graph.c
@@ -131,8 +131,8 @@ nv50_graph_init(struct drm_device *dev, int engine)
 	NV_DEBUG(dev, "\n");
 
 	/* master reset */
-	nv_mask(dev, 0x000200, 0x00200100, 0x00000000);
-	nv_mask(dev, 0x000200, 0x00200100, 0x00200100);
+	nv_mask(dev, 0x000200, 0x00201000, 0x00000000);
+	nv_mask(dev, 0x000200, 0x00201000, 0x00201000);
 	nv_wr32(dev, 0x40008c, 0x00000004); /* HW_CTX_SWITCH_ENABLED */
 
 	/* reset/enable traps and interrupts */
diff --git a/drivers/gpu/drm/nouveau/nv50_grctx.c b/drivers/gpu/drm/nouveau/nv50_grctx.c
index d05c2c3b2444..4b46d6968566 100644
--- a/drivers/gpu/drm/nouveau/nv50_grctx.c
+++ b/drivers/gpu/drm/nouveau/nv50_grctx.c
@@ -601,7 +601,7 @@ nv50_graph_construct_mmio(struct nouveau_grctx *ctx)
 					gr_def(ctx, offset + 0x1c, 0x00880000);
 					break;
 				case 0x86:
-					gr_def(ctx, offset + 0x1c, 0x008c0000);
+					gr_def(ctx, offset + 0x1c, 0x018c0000);
 					break;
 				case 0x92:
 				case 0x96:
diff --git a/drivers/gpu/drm/nouveau/nv50_vram.c b/drivers/gpu/drm/nouveau/nv50_vram.c
index 9da23838e63e..2e45e57fd869 100644
--- a/drivers/gpu/drm/nouveau/nv50_vram.c
+++ b/drivers/gpu/drm/nouveau/nv50_vram.c
@@ -160,7 +160,7 @@ nv50_vram_rblock(struct drm_device *dev)
 	colbits  =  (r4 & 0x0000f000) >> 12;
 	rowbitsa = ((r4 & 0x000f0000) >> 16) + 8;
 	rowbitsb = ((r4 & 0x00f00000) >> 20) + 8;
-	banks    = ((r4 & 0x01000000) ? 8 : 4);
+	banks    = 1 << (((r4 & 0x03000000) >> 24) + 2);
 
 	rowsize = parts * banks * (1 << colbits) * 8;
 	predicted = rowsize << rowbitsa;
diff --git a/drivers/gpu/drm/nouveau/nvc0_graph.c b/drivers/gpu/drm/nouveau/nvc0_graph.c
index bbdbc51830c8..ecfafd70cf0e 100644
--- a/drivers/gpu/drm/nouveau/nvc0_graph.c
+++ b/drivers/gpu/drm/nouveau/nvc0_graph.c
@@ -157,8 +157,8 @@ nvc0_graph_create_context_mmio_list(struct nouveau_channel *chan)
 	struct nvc0_graph_priv *priv = nv_engine(chan->dev, NVOBJ_ENGINE_GR);
 	struct nvc0_graph_chan *grch = chan->engctx[NVOBJ_ENGINE_GR];
 	struct drm_device *dev = chan->dev;
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	int i = 0, gpc, tp, ret;
-	u32 magic;
 
 	ret = nouveau_gpuobj_new(dev, chan, 0x2000, 256, NVOBJ_FLAG_VM,
 				 &grch->unk408004);
@@ -207,14 +207,37 @@ nvc0_graph_create_context_mmio_list(struct nouveau_channel *chan)
 	nv_wo32(grch->mmio, i++ * 4, 0x0041880c);
 	nv_wo32(grch->mmio, i++ * 4, 0x80000018);
 
-	magic = 0x02180000;
-	nv_wo32(grch->mmio, i++ * 4, 0x00405830);
-	nv_wo32(grch->mmio, i++ * 4, magic);
-	for (gpc = 0; gpc < priv->gpc_nr; gpc++) {
-		for (tp = 0; tp < priv->tp_nr[gpc]; tp++, magic += 0x0324) {
-			u32 reg = 0x504520 + (gpc * 0x8000) + (tp * 0x0800);
-			nv_wo32(grch->mmio, i++ * 4, reg);
-			nv_wo32(grch->mmio, i++ * 4, magic);
+	if (dev_priv->chipset != 0xc1) {
+		u32 magic = 0x02180000;
+		nv_wo32(grch->mmio, i++ * 4, 0x00405830);
+		nv_wo32(grch->mmio, i++ * 4, magic);
+		for (gpc = 0; gpc < priv->gpc_nr; gpc++) {
+			for (tp = 0; tp < priv->tp_nr[gpc]; tp++) {
+				u32 reg = TP_UNIT(gpc, tp, 0x520);
+				nv_wo32(grch->mmio, i++ * 4, reg);
+				nv_wo32(grch->mmio, i++ * 4, magic);
+				magic += 0x0324;
+			}
+		}
+	} else {
+		u32 magic = 0x02180000;
+		nv_wo32(grch->mmio, i++ * 4, 0x00405830);
+		nv_wo32(grch->mmio, i++ * 4, magic | 0x0000218);
+		nv_wo32(grch->mmio, i++ * 4, 0x004064c4);
+		nv_wo32(grch->mmio, i++ * 4, 0x0086ffff);
+		for (gpc = 0; gpc < priv->gpc_nr; gpc++) {
+			for (tp = 0; tp < priv->tp_nr[gpc]; tp++) {
+				u32 reg = TP_UNIT(gpc, tp, 0x520);
+				nv_wo32(grch->mmio, i++ * 4, reg);
+				nv_wo32(grch->mmio, i++ * 4, (1 << 28) | magic);
+				magic += 0x0324;
+			}
+			for (tp = 0; tp < priv->tp_nr[gpc]; tp++) {
+				u32 reg = TP_UNIT(gpc, tp, 0x544);
+				nv_wo32(grch->mmio, i++ * 4, reg);
+				nv_wo32(grch->mmio, i++ * 4, magic);
+				magic += 0x0324;
+			}
 		}
 	}
 
@@ -358,6 +381,8 @@ nvc0_graph_init_gpc_0(struct drm_device *dev)
 	u8  tpnr[GPC_MAX];
 	int i, gpc, tpc;
 
+	nv_wr32(dev, TP_UNIT(0, 0, 0x5c), 1); /* affects TFB offset queries */
+
 	/*
 	 *      TP      ROP UNKVAL(magic_not_rop_nr)
 	 * 450: 4/0/0/0 2        3
diff --git a/drivers/gpu/drm/nouveau/nvc0_grctx.c b/drivers/gpu/drm/nouveau/nvc0_grctx.c
index dd0e6a736b3b..96b0b93d94ca 100644
--- a/drivers/gpu/drm/nouveau/nvc0_grctx.c
+++ b/drivers/gpu/drm/nouveau/nvc0_grctx.c
@@ -1812,6 +1812,7 @@ nvc0_grctx_generate(struct nouveau_channel *chan)
 		/* calculate first set of magics */
 		memcpy(tpnr, priv->tp_nr, sizeof(priv->tp_nr));
 
+		gpc = -1;
 		for (tp = 0; tp < priv->tp_total; tp++) {
 			do {
 				gpc = (gpc + 1) % priv->gpc_nr;
@@ -1861,30 +1862,26 @@ nvc0_grctx_generate(struct nouveau_channel *chan)
 
 	if (1) {
 		u32 tp_mask = 0, tp_set = 0;
-		u8  tpnr[GPC_MAX];
+		u8  tpnr[GPC_MAX], a, b;
 
 		memcpy(tpnr, priv->tp_nr, sizeof(priv->tp_nr));
 		for (gpc = 0; gpc < priv->gpc_nr; gpc++)
 			tp_mask |= ((1 << priv->tp_nr[gpc]) - 1) << (gpc * 8);
 
-		gpc = -1;
-		for (i = 0, gpc = -1; i < 32; i++) {
-			int ltp = i * (priv->tp_total - 1) / 32;
-
-			do {
-				gpc = (gpc + 1) % priv->gpc_nr;
-			} while (!tpnr[gpc]);
-			tp = priv->tp_nr[gpc] - tpnr[gpc]--;
+		for (i = 0, gpc = -1, b = -1; i < 32; i++) {
+			a = (i * (priv->tp_total - 1)) / 32;
+			if (a != b) {
+				b = a;
+				do {
+					gpc = (gpc + 1) % priv->gpc_nr;
+				} while (!tpnr[gpc]);
+				tp = priv->tp_nr[gpc] - tpnr[gpc]--;
 
-			tp_set |= 1 << ((gpc * 8) + tp);
+				tp_set |= 1 << ((gpc * 8) + tp);
+			}
 
-			do {
-				nv_wr32(dev, 0x406800 + (i * 0x20), tp_set);
-				tp_set ^= tp_mask;
-				nv_wr32(dev, 0x406c00 + (i * 0x20), tp_set);
-				tp_set ^= tp_mask;
-			} while (ltp == (++i * (priv->tp_total - 1) / 32));
-			i--;
+			nv_wr32(dev, 0x406800 + (i * 0x20), tp_set);
+			nv_wr32(dev, 0x406c00 + (i * 0x20), tp_set ^ tp_mask);
 		}
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nvc0_vram.c b/drivers/gpu/drm/nouveau/nvc0_vram.c
index edbfe9360ae2..ce984d573a51 100644
--- a/drivers/gpu/drm/nouveau/nvc0_vram.c
+++ b/drivers/gpu/drm/nouveau/nvc0_vram.c
@@ -43,7 +43,7 @@ static const u8 types[256] = {
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 	0, 0, 0, 3, 3, 3, 3, 1, 1, 1, 1, 0, 0, 0, 0, 0,
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3,
-	3, 3, 3, 3, 0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3,
+	3, 3, 3, 1, 0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3,
 	3, 3, 0, 0, 0, 0, 0, 0, 3, 0, 0, 3, 0, 3, 0, 3,
 	3, 0, 3, 3, 3, 3, 3, 0, 0, 3, 0, 3, 0, 3, 3, 0,
 	3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 0, 1, 1, 0
@@ -110,22 +110,26 @@ nvc0_vram_init(struct drm_device *dev)
 	u32 bsize = nv_rd32(dev, 0x10f20c);
 	u32 offset, length;
 	bool uniform = true;
-	int ret, i;
+	int ret, part;
 
 	NV_DEBUG(dev, "0x100800: 0x%08x\n", nv_rd32(dev, 0x100800));
 	NV_DEBUG(dev, "parts 0x%08x bcast_mem_amount 0x%08x\n", parts, bsize);
 
 	/* read amount of vram attached to each memory controller */
-	for (i = 0; i < parts; i++) {
-		u32 psize = nv_rd32(dev, 0x11020c + (i * 0x1000));
+	part = 0;
+	while (parts) {
+		u32 psize = nv_rd32(dev, 0x11020c + (part++ * 0x1000));
+		if (psize == 0)
+			continue;
+		parts--;
+
 		if (psize != bsize) {
 			if (psize < bsize)
 				bsize = psize;
 			uniform = false;
 		}
 
-		NV_DEBUG(dev, "%d: mem_amount 0x%08x\n", i, psize);
-
+		NV_DEBUG(dev, "%d: mem_amount 0x%08x\n", part, psize);
 		dev_priv->vram_size += (u64)psize << 20;
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nvd0_display.c b/drivers/gpu/drm/nouveau/nvd0_display.c
index 23d63b4b3d77..cb006a718e70 100644
--- a/drivers/gpu/drm/nouveau/nvd0_display.c
+++ b/drivers/gpu/drm/nouveau/nvd0_display.c
@@ -780,7 +780,7 @@ nvd0_sor_dpms(struct drm_encoder *encoder, int mode)
 			continue;
 
 		if (nv_partner != nv_encoder &&
-		    nv_partner->dcb->or == nv_encoder->or) {
+		    nv_partner->dcb->or == nv_encoder->dcb->or) {
 			if (nv_partner->last_dpms == DRM_MODE_DPMS_ON)
 				return;
 			break;
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index 7567ff2510e0..457bbad3cbf9 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -1107,9 +1107,40 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
 		return -EINVAL;
 	}
 
-	if (tiling_flags & RADEON_TILING_MACRO)
+	if (tiling_flags & RADEON_TILING_MACRO) {
+		if (rdev->family >= CHIP_CAYMAN)
+			tmp = rdev->config.cayman.tile_config;
+		else
+			tmp = rdev->config.evergreen.tile_config;
+
+		switch ((tmp & 0xf0) >> 4) {
+		case 0: /* 4 banks */
+			fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_4_BANK);
+			break;
+		case 1: /* 8 banks */
+		default:
+			fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_8_BANK);
+			break;
+		case 2: /* 16 banks */
+			fb_format |= EVERGREEN_GRPH_NUM_BANKS(EVERGREEN_ADDR_SURF_16_BANK);
+			break;
+		}
+
+		switch ((tmp & 0xf000) >> 12) {
+		case 0: /* 1KB rows */
+		default:
+			fb_format |= EVERGREEN_GRPH_TILE_SPLIT(EVERGREEN_ADDR_SURF_TILE_SPLIT_1KB);
+			break;
+		case 1: /* 2KB rows */
+			fb_format |= EVERGREEN_GRPH_TILE_SPLIT(EVERGREEN_ADDR_SURF_TILE_SPLIT_2KB);
+			break;
+		case 2: /* 4KB rows */
+			fb_format |= EVERGREEN_GRPH_TILE_SPLIT(EVERGREEN_ADDR_SURF_TILE_SPLIT_4KB);
+			break;
+		}
+
 		fb_format |= EVERGREEN_GRPH_ARRAY_MODE(EVERGREEN_GRPH_ARRAY_2D_TILED_THIN1);
-	else if (tiling_flags & RADEON_TILING_MICRO)
+	} else if (tiling_flags & RADEON_TILING_MICRO)
 		fb_format |= EVERGREEN_GRPH_ARRAY_MODE(EVERGREEN_GRPH_ARRAY_1D_TILED_THIN1);
 
 	switch (radeon_crtc->crtc_id) {
@@ -1522,12 +1553,6 @@ static bool atombios_crtc_mode_fixup(struct drm_crtc *crtc,
 				     struct drm_display_mode *mode,
 				     struct drm_display_mode *adjusted_mode)
 {
-	struct drm_device *dev = crtc->dev;
-	struct radeon_device *rdev = dev->dev_private;
-
-	/* adjust pm to upcoming mode change */
-	radeon_pm_compute_clocks(rdev);
-
 	if (!radeon_crtc_scaling_mode_fixup(crtc, mode, adjusted_mode))
 		return false;
 	return true;
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index a0de48542f71..6fb335a4fdda 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -283,7 +283,7 @@ int radeon_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode,
 		}
 	}
 
-	DRM_ERROR("aux i2c too many retries, giving up\n");
+	DRM_DEBUG_KMS("aux i2c too many retries, giving up\n");
 	return -EREMOTEIO;
 }
 
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index e4c384b9511c..5e00d1670aa9 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -82,6 +82,7 @@ u32 evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset);
+	int i;
 
 	/* Lock the graphics update lock */
 	tmp |= EVERGREEN_GRPH_UPDATE_LOCK;
@@ -99,7 +100,11 @@ u32 evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	       (u32)crtc_base);
 
 	/* Wait for update_pending to go high. */
-	while (!(RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) & EVERGREEN_GRPH_SURFACE_UPDATE_PENDING));
+	for (i = 0; i < rdev->usec_timeout; i++) {
+		if (RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) & EVERGREEN_GRPH_SURFACE_UPDATE_PENDING)
+			break;
+		udelay(1);
+	}
 	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
 
 	/* Unlock the lock, so double-buffering can take place inside vblank */
@@ -157,6 +162,57 @@ int sumo_get_temp(struct radeon_device *rdev)
 	return actual_temp * 1000;
 }
 
+void sumo_pm_init_profile(struct radeon_device *rdev)
+{
+	int idx;
+
+	/* default */
+	rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_off_ps_idx = rdev->pm.default_power_state_index;
+	rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_on_ps_idx = rdev->pm.default_power_state_index;
+	rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_on_cm_idx = 0;
+
+	/* low,mid sh/mh */
+	if (rdev->flags & RADEON_IS_MOBILITY)
+		idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
+	else
+		idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
+
+	rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_cm_idx = 0;
+
+	rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_cm_idx = 0;
+
+	rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_cm_idx = 0;
+
+	rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_cm_idx = 0;
+
+	/* high sh/mh */
+	idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
+	rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_cm_idx =
+		rdev->pm.power_state[idx].num_clock_modes - 1;
+
+	rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_ps_idx = idx;
+	rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_cm_idx = 0;
+	rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_cm_idx =
+		rdev->pm.power_state[idx].num_clock_modes - 1;
+}
+
 void evergreen_pm_misc(struct radeon_device *rdev)
 {
 	int req_ps_idx = rdev->pm.requested_power_state_index;
@@ -1219,7 +1275,7 @@ void evergreen_mc_program(struct radeon_device *rdev)
 		WREG32(MC_VM_SYSTEM_APERTURE_HIGH_ADDR,
 			rdev->mc.vram_end >> 12);
 	}
-	WREG32(MC_VM_SYSTEM_APERTURE_DEFAULT_ADDR, 0);
+	WREG32(MC_VM_SYSTEM_APERTURE_DEFAULT_ADDR, rdev->vram_scratch.gpu_addr >> 12);
 	if (rdev->flags & RADEON_IS_IGP) {
 		tmp = RREG32(MC_FUS_VM_FB_OFFSET) & 0x000FFFFF;
 		tmp |= ((rdev->mc.vram_end >> 20) & 0xF) << 24;
diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index 7fdfa8ea7570..cd4590aae154 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -38,6 +38,7 @@ struct evergreen_cs_track {
 	u32			group_size;
 	u32			nbanks;
 	u32			npipes;
+	u32			row_size;
 	/* value we track */
 	u32			nsamples;
 	u32			cb_color_base_last[12];
@@ -77,6 +78,44 @@ struct evergreen_cs_track {
 	struct radeon_bo	*db_s_write_bo;
 };
 
+static u32 evergreen_cs_get_aray_mode(u32 tiling_flags)
+{
+	if (tiling_flags & RADEON_TILING_MACRO)
+		return ARRAY_2D_TILED_THIN1;
+	else if (tiling_flags & RADEON_TILING_MICRO)
+		return ARRAY_1D_TILED_THIN1;
+	else
+		return ARRAY_LINEAR_GENERAL;
+}
+
+static u32 evergreen_cs_get_num_banks(u32 nbanks)
+{
+	switch (nbanks) {
+	case 2:
+		return ADDR_SURF_2_BANK;
+	case 4:
+		return ADDR_SURF_4_BANK;
+	case 8:
+	default:
+		return ADDR_SURF_8_BANK;
+	case 16:
+		return ADDR_SURF_16_BANK;
+	}
+}
+
+static u32 evergreen_cs_get_tile_split(u32 row_size)
+{
+	switch (row_size) {
+	case 1:
+	default:
+		return ADDR_SURF_TILE_SPLIT_1KB;
+	case 2:
+		return ADDR_SURF_TILE_SPLIT_2KB;
+	case 4:
+		return ADDR_SURF_TILE_SPLIT_4KB;
+	}
+}
+
 static void evergreen_cs_track_init(struct evergreen_cs_track *track)
 {
 	int i;
@@ -480,21 +519,22 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		}
 		break;
 	case DB_Z_INFO:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
-		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
-			return -EINVAL;
-		}
 		track->db_z_info = radeon_get_ib_value(p, idx);
-		ib[idx] &= ~Z_ARRAY_MODE(0xf);
-		track->db_z_info &= ~Z_ARRAY_MODE(0xf);
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
-			ib[idx] |= Z_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-			track->db_z_info |= Z_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-		} else {
-			ib[idx] |= Z_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
-			track->db_z_info |= Z_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
+		if (!p->keep_tiling_flags) {
+			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			if (r) {
+				dev_warn(p->dev, "bad SET_CONTEXT_REG "
+						"0x%04X\n", reg);
+				return -EINVAL;
+			}
+			ib[idx] &= ~Z_ARRAY_MODE(0xf);
+			track->db_z_info &= ~Z_ARRAY_MODE(0xf);
+			ib[idx] |= Z_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
+			track->db_z_info |= Z_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
+				ib[idx] |= DB_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks));
+				ib[idx] |= DB_TILE_SPLIT(evergreen_cs_get_tile_split(track->row_size));
+			}
 		}
 		break;
 	case DB_STENCIL_INFO:
@@ -607,40 +647,34 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR5_INFO:
 	case CB_COLOR6_INFO:
 	case CB_COLOR7_INFO:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
-		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
-			return -EINVAL;
-		}
 		tmp = (reg - CB_COLOR0_INFO) / 0x3c;
 		track->cb_color_info[tmp] = radeon_get_ib_value(p, idx);
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
-			ib[idx] |= CB_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-			track->cb_color_info[tmp] |= CB_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-		} else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) {
-			ib[idx] |= CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
-			track->cb_color_info[tmp] |= CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
+		if (!p->keep_tiling_flags) {
+			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			if (r) {
+				dev_warn(p->dev, "bad SET_CONTEXT_REG "
+						"0x%04X\n", reg);
+				return -EINVAL;
+			}
+			ib[idx] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
+			track->cb_color_info[tmp] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
 		}
 		break;
 	case CB_COLOR8_INFO:
 	case CB_COLOR9_INFO:
 	case CB_COLOR10_INFO:
 	case CB_COLOR11_INFO:
-		r = evergreen_cs_packet_next_reloc(p, &reloc);
-		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
-			return -EINVAL;
-		}
 		tmp = ((reg - CB_COLOR8_INFO) / 0x1c) + 8;
 		track->cb_color_info[tmp] = radeon_get_ib_value(p, idx);
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
-			ib[idx] |= CB_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-			track->cb_color_info[tmp] |= CB_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-		} else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) {
-			ib[idx] |= CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
-			track->cb_color_info[tmp] |= CB_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
+		if (!p->keep_tiling_flags) {
+			r = evergreen_cs_packet_next_reloc(p, &reloc);
+			if (r) {
+				dev_warn(p->dev, "bad SET_CONTEXT_REG "
+						"0x%04X\n", reg);
+				return -EINVAL;
+			}
+			ib[idx] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
+			track->cb_color_info[tmp] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
 		}
 		break;
 	case CB_COLOR0_PITCH:
@@ -695,6 +729,16 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR9_ATTRIB:
 	case CB_COLOR10_ATTRIB:
 	case CB_COLOR11_ATTRIB:
+		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		if (r) {
+			dev_warn(p->dev, "bad SET_CONTEXT_REG "
+					"0x%04X\n", reg);
+			return -EINVAL;
+		}
+		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
+			ib[idx] |= CB_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks));
+			ib[idx] |= CB_TILE_SPLIT(evergreen_cs_get_tile_split(track->row_size));
+		}
 		break;
 	case CB_COLOR0_DIM:
 	case CB_COLOR1_DIM:
@@ -1311,10 +1355,16 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 					return -EINVAL;
 				}
 				ib[idx+1+(i*8)+2] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
-				if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-					ib[idx+1+(i*8)+1] |= TEX_ARRAY_MODE(ARRAY_2D_TILED_THIN1);
-				else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
-					ib[idx+1+(i*8)+1] |= TEX_ARRAY_MODE(ARRAY_1D_TILED_THIN1);
+				if (!p->keep_tiling_flags) {
+					ib[idx+1+(i*8)+1] |=
+						TEX_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags));
+					if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
+						ib[idx+1+(i*8)+6] |=
+							TEX_TILE_SPLIT(evergreen_cs_get_tile_split(track->row_size));
+						ib[idx+1+(i*8)+7] |=
+							TEX_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks));
+					}
+				}
 				texture = reloc->robj;
 				/* tex mip base */
 				r = evergreen_cs_packet_next_reloc(p, &reloc);
@@ -1414,6 +1464,7 @@ int evergreen_cs_parse(struct radeon_cs_parser *p)
 {
 	struct radeon_cs_packet pkt;
 	struct evergreen_cs_track *track;
+	u32 tmp;
 	int r;
 
 	if (p->track == NULL) {
@@ -1422,9 +1473,63 @@ int evergreen_cs_parse(struct radeon_cs_parser *p)
 		if (track == NULL)
 			return -ENOMEM;
 		evergreen_cs_track_init(track);
-		track->npipes = p->rdev->config.evergreen.tiling_npipes;
-		track->nbanks = p->rdev->config.evergreen.tiling_nbanks;
-		track->group_size = p->rdev->config.evergreen.tiling_group_size;
+		if (p->rdev->family >= CHIP_CAYMAN)
+			tmp = p->rdev->config.cayman.tile_config;
+		else
+			tmp = p->rdev->config.evergreen.tile_config;
+
+		switch (tmp & 0xf) {
+		case 0:
+			track->npipes = 1;
+			break;
+		case 1:
+		default:
+			track->npipes = 2;
+			break;
+		case 2:
+			track->npipes = 4;
+			break;
+		case 3:
+			track->npipes = 8;
+			break;
+		}
+
+		switch ((tmp & 0xf0) >> 4) {
+		case 0:
+			track->nbanks = 4;
+			break;
+		case 1:
+		default:
+			track->nbanks = 8;
+			break;
+		case 2:
+			track->nbanks = 16;
+			break;
+		}
+
+		switch ((tmp & 0xf00) >> 8) {
+		case 0:
+			track->group_size = 256;
+			break;
+		case 1:
+		default:
+			track->group_size = 512;
+			break;
+		}
+
+		switch ((tmp & 0xf000) >> 12) {
+		case 0:
+			track->row_size = 1;
+			break;
+		case 1:
+		default:
+			track->row_size = 2;
+			break;
+		case 2:
+			track->row_size = 4;
+			break;
+		}
+
 		p->track = track;
 	}
 	do {
diff --git a/drivers/gpu/drm/radeon/evergreen_reg.h b/drivers/gpu/drm/radeon/evergreen_reg.h
index c781c92c3451..7d7f2155e34c 100644
--- a/drivers/gpu/drm/radeon/evergreen_reg.h
+++ b/drivers/gpu/drm/radeon/evergreen_reg.h
@@ -42,6 +42,17 @@
 #       define EVERGREEN_GRPH_DEPTH_8BPP                0
 #       define EVERGREEN_GRPH_DEPTH_16BPP               1
 #       define EVERGREEN_GRPH_DEPTH_32BPP               2
+#       define EVERGREEN_GRPH_NUM_BANKS(x)              (((x) & 0x3) << 2)
+#       define EVERGREEN_ADDR_SURF_2_BANK               0
+#       define EVERGREEN_ADDR_SURF_4_BANK               1
+#       define EVERGREEN_ADDR_SURF_8_BANK               2
+#       define EVERGREEN_ADDR_SURF_16_BANK              3
+#       define EVERGREEN_GRPH_Z(x)                      (((x) & 0x3) << 4)
+#       define EVERGREEN_GRPH_BANK_WIDTH(x)             (((x) & 0x3) << 6)
+#       define EVERGREEN_ADDR_SURF_BANK_WIDTH_1         0
+#       define EVERGREEN_ADDR_SURF_BANK_WIDTH_2         1
+#       define EVERGREEN_ADDR_SURF_BANK_WIDTH_4         2
+#       define EVERGREEN_ADDR_SURF_BANK_WIDTH_8         3
 #       define EVERGREEN_GRPH_FORMAT(x)                 (((x) & 0x7) << 8)
 /* 8 BPP */
 #       define EVERGREEN_GRPH_FORMAT_INDEXED            0
@@ -61,6 +72,24 @@
 #       define EVERGREEN_GRPH_FORMAT_8B_BGRA1010102     5
 #       define EVERGREEN_GRPH_FORMAT_RGB111110          6
 #       define EVERGREEN_GRPH_FORMAT_BGR101111          7
+#       define EVERGREEN_GRPH_BANK_HEIGHT(x)            (((x) & 0x3) << 11)
+#       define EVERGREEN_ADDR_SURF_BANK_HEIGHT_1        0
+#       define EVERGREEN_ADDR_SURF_BANK_HEIGHT_2        1
+#       define EVERGREEN_ADDR_SURF_BANK_HEIGHT_4        2
+#       define EVERGREEN_ADDR_SURF_BANK_HEIGHT_8        3
+#       define EVERGREEN_GRPH_TILE_SPLIT(x)             (((x) & 0x7) << 13)
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_64B       0
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_128B      1
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_256B      2
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_512B      3
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_1KB       4
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_2KB       5
+#       define EVERGREEN_ADDR_SURF_TILE_SPLIT_4KB       6
+#       define EVERGREEN_GRPH_MACRO_TILE_ASPECT(x)      (((x) & 0x3) << 18)
+#       define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_1  0
+#       define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_2  1
+#       define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_4  2
+#       define EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_8  3
 #       define EVERGREEN_GRPH_ARRAY_MODE(x)             (((x) & 0x7) << 20)
 #       define EVERGREEN_GRPH_ARRAY_LINEAR_GENERAL      0
 #       define EVERGREEN_GRPH_ARRAY_LINEAR_ALIGNED      1
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index b937c49054d9..e00039e59a75 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -899,6 +899,10 @@
 #define DB_HTILE_DATA_BASE				0x28014
 #define DB_Z_INFO					0x28040
 #       define Z_ARRAY_MODE(x)                          ((x) << 4)
+#       define DB_TILE_SPLIT(x)                         (((x) & 0x7) << 8)
+#       define DB_NUM_BANKS(x)                          (((x) & 0x3) << 12)
+#       define DB_BANK_WIDTH(x)                         (((x) & 0x3) << 16)
+#       define DB_BANK_HEIGHT(x)                        (((x) & 0x3) << 20)
 #define DB_STENCIL_INFO					0x28044
 #define DB_Z_READ_BASE					0x28048
 #define DB_STENCIL_READ_BASE				0x2804c
@@ -951,6 +955,29 @@
 #	define CB_SF_EXPORT_FULL			0
 #	define CB_SF_EXPORT_NORM			1
 #define	CB_COLOR0_ATTRIB				0x28c74
+#       define CB_TILE_SPLIT(x)                         (((x) & 0x7) << 5)
+#       define ADDR_SURF_TILE_SPLIT_64B                 0
+#       define ADDR_SURF_TILE_SPLIT_128B                1
+#       define ADDR_SURF_TILE_SPLIT_256B                2
+#       define ADDR_SURF_TILE_SPLIT_512B                3
+#       define ADDR_SURF_TILE_SPLIT_1KB                 4
+#       define ADDR_SURF_TILE_SPLIT_2KB                 5
+#       define ADDR_SURF_TILE_SPLIT_4KB                 6
+#       define CB_NUM_BANKS(x)                          (((x) & 0x3) << 10)
+#       define ADDR_SURF_2_BANK                         0
+#       define ADDR_SURF_4_BANK                         1
+#       define ADDR_SURF_8_BANK                         2
+#       define ADDR_SURF_16_BANK                        3
+#       define CB_BANK_WIDTH(x)                         (((x) & 0x3) << 13)
+#       define ADDR_SURF_BANK_WIDTH_1                   0
+#       define ADDR_SURF_BANK_WIDTH_2                   1
+#       define ADDR_SURF_BANK_WIDTH_4                   2
+#       define ADDR_SURF_BANK_WIDTH_8                   3
+#       define CB_BANK_HEIGHT(x)                        (((x) & 0x3) << 16)
+#       define ADDR_SURF_BANK_HEIGHT_1                  0
+#       define ADDR_SURF_BANK_HEIGHT_2                  1
+#       define ADDR_SURF_BANK_HEIGHT_4                  2
+#       define ADDR_SURF_BANK_HEIGHT_8                  3
 #define	CB_COLOR0_DIM					0x28c78
 /* only CB0-7 blocks have these regs */
 #define	CB_COLOR0_CMASK					0x28c7c
@@ -1137,7 +1164,11 @@
 #	define SQ_SEL_1					5
 #define SQ_TEX_RESOURCE_WORD5_0                         0x30014
 #define SQ_TEX_RESOURCE_WORD6_0                         0x30018
+#       define TEX_TILE_SPLIT(x)                        (((x) & 0x7) << 29)
 #define SQ_TEX_RESOURCE_WORD7_0                         0x3001c
+#       define TEX_BANK_WIDTH(x)                        (((x) & 0x3) << 8)
+#       define TEX_BANK_HEIGHT(x)                       (((x) & 0x3) << 10)
+#       define TEX_NUM_BANKS(x)                         (((x) & 0x3) << 16)
 
 #define SQ_VTX_CONSTANT_WORD0_0				0x30000
 #define SQ_VTX_CONSTANT_WORD1_0				0x30004
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index ad158ea49901..bfc08f6320f8 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -187,13 +187,18 @@ u32 r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = ((u32)crtc_base) | RADEON_CRTC_OFFSET__OFFSET_LOCK;
+	int i;
 
 	/* Lock the graphics update lock */
 	/* update the scanout addresses */
 	WREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset, tmp);
 
 	/* Wait for update_pending to go high. */
-	while (!(RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) & RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET));
+	for (i = 0; i < rdev->usec_timeout; i++) {
+		if (RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) & RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET)
+			break;
+		udelay(1);
+	}
 	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
 
 	/* Unlock the lock, so double-buffering can take place inside vblank */
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 400b26df652a..c93bc64707e1 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -701,16 +701,21 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 			return r;
 		}
 
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-			tile_flags |= R300_TXO_MACRO_TILE;
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
-			tile_flags |= R300_TXO_MICRO_TILE;
-		else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
-			tile_flags |= R300_TXO_MICRO_TILE_SQUARE;
-
-		tmp = idx_value + ((u32)reloc->lobj.gpu_offset);
-		tmp |= tile_flags;
-		ib[idx] = tmp;
+		if (p->keep_tiling_flags) {
+			ib[idx] = (idx_value & 31) | /* keep the 1st 5 bits */
+				  ((idx_value & ~31) + (u32)reloc->lobj.gpu_offset);
+		} else {
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
+				tile_flags |= R300_TXO_MACRO_TILE;
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
+				tile_flags |= R300_TXO_MICRO_TILE;
+			else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
+				tile_flags |= R300_TXO_MICRO_TILE_SQUARE;
+
+			tmp = idx_value + ((u32)reloc->lobj.gpu_offset);
+			tmp |= tile_flags;
+			ib[idx] = tmp;
+		}
 		track->textures[i].robj = reloc->robj;
 		track->tex_dirty = true;
 		break;
@@ -760,24 +765,26 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		/* RB3D_COLORPITCH1 */
 		/* RB3D_COLORPITCH2 */
 		/* RB3D_COLORPITCH3 */
-		r = r100_cs_packet_next_reloc(p, &reloc);
-		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
-			return r;
-		}
+		if (!p->keep_tiling_flags) {
+			r = r100_cs_packet_next_reloc(p, &reloc);
+			if (r) {
+				DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
+					  idx, reg);
+				r100_cs_dump_packet(p, pkt);
+				return r;
+			}
 
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-			tile_flags |= R300_COLOR_TILE_ENABLE;
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
-			tile_flags |= R300_COLOR_MICROTILE_ENABLE;
-		else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
-			tile_flags |= R300_COLOR_MICROTILE_SQUARE_ENABLE;
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
+				tile_flags |= R300_COLOR_TILE_ENABLE;
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
+				tile_flags |= R300_COLOR_MICROTILE_ENABLE;
+			else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
+				tile_flags |= R300_COLOR_MICROTILE_SQUARE_ENABLE;
 
-		tmp = idx_value & ~(0x7 << 16);
-		tmp |= tile_flags;
-		ib[idx] = tmp;
+			tmp = idx_value & ~(0x7 << 16);
+			tmp |= tile_flags;
+			ib[idx] = tmp;
+		}
 		i = (reg - 0x4E38) >> 2;
 		track->cb[i].pitch = idx_value & 0x3FFE;
 		switch (((idx_value >> 21) & 0xF)) {
@@ -843,25 +850,26 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		break;
 	case 0x4F24:
 		/* ZB_DEPTHPITCH */
-		r = r100_cs_packet_next_reloc(p, &reloc);
-		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
-			r100_cs_dump_packet(p, pkt);
-			return r;
-		}
-
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-			tile_flags |= R300_DEPTHMACROTILE_ENABLE;
-		if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
-			tile_flags |= R300_DEPTHMICROTILE_TILED;
-		else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
-			tile_flags |= R300_DEPTHMICROTILE_TILED_SQUARE;
+		if (!p->keep_tiling_flags) {
+			r = r100_cs_packet_next_reloc(p, &reloc);
+			if (r) {
+				DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
+					  idx, reg);
+				r100_cs_dump_packet(p, pkt);
+				return r;
+			}
 
-		tmp = idx_value & ~(0x7 << 16);
-		tmp |= tile_flags;
-		ib[idx] = tmp;
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
+				tile_flags |= R300_DEPTHMACROTILE_ENABLE;
+			if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
+				tile_flags |= R300_DEPTHMICROTILE_TILED;
+			else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO_SQUARE)
+				tile_flags |= R300_DEPTHMICROTILE_TILED_SQUARE;
 
+			tmp = idx_value & ~(0x7 << 16);
+			tmp |= tile_flags;
+			ib[idx] = tmp;
+		}
 		track->zb.pitch = idx_value & 0x3FFC;
 		track->zb_dirty = true;
 		break;
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index 19afc43ad173..9cdda0b3b081 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -288,24 +288,6 @@ void r600_pm_get_dynpm_state(struct radeon_device *rdev)
 		  pcie_lanes);
 }
 
-static int r600_pm_get_type_index(struct radeon_device *rdev,
-				  enum radeon_pm_state_type ps_type,
-				  int instance)
-{
-	int i;
-	int found_instance = -1;
-
-	for (i = 0; i < rdev->pm.num_power_states; i++) {
-		if (rdev->pm.power_state[i].type == ps_type) {
-			found_instance++;
-			if (found_instance == instance)
-				return i;
-		}
-	}
-	/* return default if no match */
-	return rdev->pm.default_power_state_index;
-}
-
 void rs780_pm_init_profile(struct radeon_device *rdev)
 {
 	if (rdev->pm.num_power_states == 2) {
@@ -421,6 +403,8 @@ void rs780_pm_init_profile(struct radeon_device *rdev)
 
 void r600_pm_init_profile(struct radeon_device *rdev)
 {
+	int idx;
+
 	if (rdev->family == CHIP_R600) {
 		/* XXX */
 		/* default */
@@ -502,81 +486,43 @@ void r600_pm_init_profile(struct radeon_device *rdev)
 			rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_off_cm_idx = 0;
 			rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_on_cm_idx = 2;
 			/* low sh */
-			if (rdev->flags & RADEON_IS_MOBILITY) {
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_cm_idx = 0;
-			} else {
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_cm_idx = 0;
-			}
+			if (rdev->flags & RADEON_IS_MOBILITY)
+				idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
+			else
+				idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
+			rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_cm_idx = 0;
+			rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_cm_idx = 0;
 			/* mid sh */
-			if (rdev->flags & RADEON_IS_MOBILITY) {
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0);
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_cm_idx = 1;
-			} else {
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_cm_idx = 1;
-			}
+			rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_cm_idx = 0;
+			rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_cm_idx = 1;
 			/* high sh */
-			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_ps_idx =
-				r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
-			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_ps_idx =
-				r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
+			idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0);
+			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_ps_idx = idx;
 			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_cm_idx = 0;
 			rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_cm_idx = 2;
 			/* low mh */
-			if (rdev->flags & RADEON_IS_MOBILITY) {
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 1);
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 1);
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_cm_idx = 0;
-			} else {
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_cm_idx = 0;
-			}
+			if (rdev->flags & RADEON_IS_MOBILITY)
+				idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 1);
+			else
+				idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
+			rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_cm_idx = 0;
+			rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_cm_idx = 0;
 			/* mid mh */
-			if (rdev->flags & RADEON_IS_MOBILITY) {
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 1);
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 1);
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_cm_idx = 1;
-			} else {
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_ps_idx =
-					r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_cm_idx = 0;
-				rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_cm_idx = 1;
-			}
+			rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_cm_idx = 0;
+			rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_cm_idx = 1;
 			/* high mh */
-			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_ps_idx =
-				r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
-			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_ps_idx =
-				r600_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
+			idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 1);
+			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_ps_idx = idx;
+			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_ps_idx = idx;
 			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_cm_idx = 0;
 			rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_cm_idx = 2;
 		}
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 0a2e023c1557..cb1acffd2430 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -941,7 +941,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->db_depth_control = radeon_get_ib_value(p, idx);
 		break;
 	case R_028010_DB_DEPTH_INFO:
-		if (r600_cs_packet_next_is_pkt3_nop(p)) {
+		if (!p->keep_tiling_flags &&
+		    r600_cs_packet_next_is_pkt3_nop(p)) {
 			r = r600_cs_packet_next_reloc(p, &reloc);
 			if (r) {
 				dev_warn(p->dev, "bad SET_CONTEXT_REG "
@@ -992,7 +993,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case R_0280B4_CB_COLOR5_INFO:
 	case R_0280B8_CB_COLOR6_INFO:
 	case R_0280BC_CB_COLOR7_INFO:
-		if (r600_cs_packet_next_is_pkt3_nop(p)) {
+		if (!p->keep_tiling_flags &&
+		     r600_cs_packet_next_is_pkt3_nop(p)) {
 			r = r600_cs_packet_next_reloc(p, &reloc);
 			if (r) {
 				dev_err(p->dev, "bad SET_CONTEXT_REG 0x%04X\n", reg);
@@ -1291,10 +1293,12 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p,  u32 idx,
 	mip_offset <<= 8;
 
 	word0 = radeon_get_ib_value(p, idx + 0);
-	if (tiling_flags & RADEON_TILING_MACRO)
-		word0 |= S_038000_TILE_MODE(V_038000_ARRAY_2D_TILED_THIN1);
-	else if (tiling_flags & RADEON_TILING_MICRO)
-		word0 |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
+	if (!p->keep_tiling_flags) {
+		if (tiling_flags & RADEON_TILING_MACRO)
+			word0 |= S_038000_TILE_MODE(V_038000_ARRAY_2D_TILED_THIN1);
+		else if (tiling_flags & RADEON_TILING_MICRO)
+			word0 |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
+	}
 	word1 = radeon_get_ib_value(p, idx + 1);
 	w0 = G_038000_TEX_WIDTH(word0) + 1;
 	h0 = G_038004_TEX_HEIGHT(word1) + 1;
@@ -1621,10 +1625,12 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 					return -EINVAL;
 				}
 				base_offset = (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
-				if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-					ib[idx+1+(i*7)+0] |= S_038000_TILE_MODE(V_038000_ARRAY_2D_TILED_THIN1);
-				else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
-					ib[idx+1+(i*7)+0] |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
+				if (!p->keep_tiling_flags) {
+					if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
+						ib[idx+1+(i*7)+0] |= S_038000_TILE_MODE(V_038000_ARRAY_2D_TILED_THIN1);
+					else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
+						ib[idx+1+(i*7)+0] |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
+				}
 				texture = reloc->robj;
 				/* tex mip base */
 				r = r600_cs_packet_next_reloc(p, &reloc);
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 83b2e016a4a1..c8f4dbd2d17c 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -610,7 +610,8 @@ struct radeon_cs_parser {
 	struct radeon_ib	*ib;
 	void			*track;
 	unsigned		family;
-	int parser_error;
+	int			parser_error;
+	bool			keep_tiling_flags;
 };
 
 extern int radeon_cs_update_pages(struct radeon_cs_parser *p, int pg_idx);
@@ -783,8 +784,7 @@ struct radeon_pm_clock_info {
 
 struct radeon_power_state {
 	enum radeon_pm_state_type type;
-	/* XXX: use a define for num clock modes */
-	struct radeon_pm_clock_info clock_info[8];
+	struct radeon_pm_clock_info *clock_info;
 	/* number of valid clock modes in this power state */
 	int num_clock_modes;
 	struct radeon_pm_clock_info *default_clock_mode;
@@ -854,6 +854,9 @@ struct radeon_pm {
 	struct device	        *int_hwmon_dev;
 };
 
+int radeon_pm_get_type_index(struct radeon_device *rdev,
+			     enum radeon_pm_state_type ps_type,
+			     int instance);
 
 /*
  * Benchmarking
@@ -1141,6 +1144,48 @@ struct r600_vram_scratch {
 	u64				gpu_addr;
 };
 
+
+/*
+ * Mutex which allows recursive locking from the same process.
+ */
+struct radeon_mutex {
+	struct mutex		mutex;
+	struct task_struct	*owner;
+	int			level;
+};
+
+static inline void radeon_mutex_init(struct radeon_mutex *mutex)
+{
+	mutex_init(&mutex->mutex);
+	mutex->owner = NULL;
+	mutex->level = 0;
+}
+
+static inline void radeon_mutex_lock(struct radeon_mutex *mutex)
+{
+	if (mutex_trylock(&mutex->mutex)) {
+		/* The mutex was unlocked before, so it's ours now */
+		mutex->owner = current;
+	} else if (mutex->owner != current) {
+		/* Another process locked the mutex, take it */
+		mutex_lock(&mutex->mutex);
+		mutex->owner = current;
+	}
+	/* Otherwise the mutex was already locked by this process */
+
+	mutex->level++;
+}
+
+static inline void radeon_mutex_unlock(struct radeon_mutex *mutex)
+{
+	if (--mutex->level > 0)
+		return;
+
+	mutex->owner = NULL;
+	mutex_unlock(&mutex->mutex);
+}
+
+
 /*
  * Core structure, functions and helpers.
  */
@@ -1196,7 +1241,7 @@ struct radeon_device {
 	struct radeon_gem		gem;
 	struct radeon_pm		pm;
 	uint32_t			bios_scratch[RADEON_BIOS_NUM_SCRATCH];
-	struct mutex			cs_mutex;
+	struct radeon_mutex		cs_mutex;
 	struct radeon_wb		wb;
 	struct radeon_dummy_page	dummy_page;
 	bool				gpu_lockup;
diff --git a/drivers/gpu/drm/radeon/radeon_acpi.c b/drivers/gpu/drm/radeon/radeon_acpi.c
index 3f6636bb2d7f..3516a6081dcf 100644
--- a/drivers/gpu/drm/radeon/radeon_acpi.c
+++ b/drivers/gpu/drm/radeon/radeon_acpi.c
@@ -35,7 +35,8 @@ static int radeon_atif_call(acpi_handle handle)
 
 	/* Fail only if calling the method fails and ATIF is supported */
 	if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) {
-		printk(KERN_DEBUG "failed to evaluate ATIF got %s\n", acpi_format_exception(status));
+		DRM_DEBUG_DRIVER("failed to evaluate ATIF got %s\n",
+				 acpi_format_exception(status));
 		kfree(buffer.pointer);
 		return 1;
 	}
@@ -50,13 +51,13 @@ int radeon_acpi_init(struct radeon_device *rdev)
 	acpi_handle handle;
 	int ret;
 
-	/* No need to proceed if we're sure that ATIF is not supported */
-	if (!ASIC_IS_AVIVO(rdev) || !rdev->bios)
-		return 0;
-
 	/* Get the device handle */
 	handle = DEVICE_ACPI_HANDLE(&rdev->pdev->dev);
 
+	/* No need to proceed if we're sure that ATIF is not supported */
+	if (!ASIC_IS_AVIVO(rdev) || !rdev->bios || !handle)
+		return 0;
+
 	/* Call the ATIF method */
 	ret = radeon_atif_call(handle);
 	if (ret)
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index e2944566ffea..a2e1eae114ef 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -834,7 +834,7 @@ static struct radeon_asic sumo_asic = {
 	.pm_misc = &evergreen_pm_misc,
 	.pm_prepare = &evergreen_pm_prepare,
 	.pm_finish = &evergreen_pm_finish,
-	.pm_init_profile = &rs780_pm_init_profile,
+	.pm_init_profile = &sumo_pm_init_profile,
 	.pm_get_dynpm_state = &r600_pm_get_dynpm_state,
 	.pre_page_flip = &evergreen_pre_page_flip,
 	.page_flip = &evergreen_page_flip,
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 85f14f0337e4..59914842a729 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -413,6 +413,7 @@ extern int evergreen_cs_parse(struct radeon_cs_parser *p);
 extern void evergreen_pm_misc(struct radeon_device *rdev);
 extern void evergreen_pm_prepare(struct radeon_device *rdev);
 extern void evergreen_pm_finish(struct radeon_device *rdev);
+extern void sumo_pm_init_profile(struct radeon_device *rdev);
 extern void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc);
 extern u32 evergreen_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base);
 extern void evergreen_post_page_flip(struct radeon_device *rdev, int crtc);
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index 08d0b94332e6..d24baf30efcb 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -62,6 +62,87 @@ union atom_supported_devices {
 	struct _ATOM_SUPPORTED_DEVICES_INFO_2d1 info_2d1;
 };
 
+static void radeon_lookup_i2c_gpio_quirks(struct radeon_device *rdev,
+					  ATOM_GPIO_I2C_ASSIGMENT *gpio,
+					  u8 index)
+{
+	/* r4xx mask is technically not used by the hw, so patch in the legacy mask bits */
+	if ((rdev->family == CHIP_R420) ||
+	    (rdev->family == CHIP_R423) ||
+	    (rdev->family == CHIP_RV410)) {
+		if ((le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x0018) ||
+		    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x0019) ||
+		    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x001a)) {
+			gpio->ucClkMaskShift = 0x19;
+			gpio->ucDataMaskShift = 0x18;
+		}
+	}
+
+	/* some evergreen boards have bad data for this entry */
+	if (ASIC_IS_DCE4(rdev)) {
+		if ((index == 7) &&
+		    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1936) &&
+		    (gpio->sucI2cId.ucAccess == 0)) {
+			gpio->sucI2cId.ucAccess = 0x97;
+			gpio->ucDataMaskShift = 8;
+			gpio->ucDataEnShift = 8;
+			gpio->ucDataY_Shift = 8;
+			gpio->ucDataA_Shift = 8;
+		}
+	}
+
+	/* some DCE3 boards have bad data for this entry */
+	if (ASIC_IS_DCE3(rdev)) {
+		if ((index == 4) &&
+		    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1fda) &&
+		    (gpio->sucI2cId.ucAccess == 0x94))
+			gpio->sucI2cId.ucAccess = 0x14;
+	}
+}
+
+static struct radeon_i2c_bus_rec radeon_get_bus_rec_for_i2c_gpio(ATOM_GPIO_I2C_ASSIGMENT *gpio)
+{
+	struct radeon_i2c_bus_rec i2c;
+
+	memset(&i2c, 0, sizeof(struct radeon_i2c_bus_rec));
+
+	i2c.mask_clk_reg = le16_to_cpu(gpio->usClkMaskRegisterIndex) * 4;
+	i2c.mask_data_reg = le16_to_cpu(gpio->usDataMaskRegisterIndex) * 4;
+	i2c.en_clk_reg = le16_to_cpu(gpio->usClkEnRegisterIndex) * 4;
+	i2c.en_data_reg = le16_to_cpu(gpio->usDataEnRegisterIndex) * 4;
+	i2c.y_clk_reg = le16_to_cpu(gpio->usClkY_RegisterIndex) * 4;
+	i2c.y_data_reg = le16_to_cpu(gpio->usDataY_RegisterIndex) * 4;
+	i2c.a_clk_reg = le16_to_cpu(gpio->usClkA_RegisterIndex) * 4;
+	i2c.a_data_reg = le16_to_cpu(gpio->usDataA_RegisterIndex) * 4;
+	i2c.mask_clk_mask = (1 << gpio->ucClkMaskShift);
+	i2c.mask_data_mask = (1 << gpio->ucDataMaskShift);
+	i2c.en_clk_mask = (1 << gpio->ucClkEnShift);
+	i2c.en_data_mask = (1 << gpio->ucDataEnShift);
+	i2c.y_clk_mask = (1 << gpio->ucClkY_Shift);
+	i2c.y_data_mask = (1 << gpio->ucDataY_Shift);
+	i2c.a_clk_mask = (1 << gpio->ucClkA_Shift);
+	i2c.a_data_mask = (1 << gpio->ucDataA_Shift);
+
+	if (gpio->sucI2cId.sbfAccess.bfHW_Capable)
+		i2c.hw_capable = true;
+	else
+		i2c.hw_capable = false;
+
+	if (gpio->sucI2cId.ucAccess == 0xa0)
+		i2c.mm_i2c = true;
+	else
+		i2c.mm_i2c = false;
+
+	i2c.i2c_id = gpio->sucI2cId.ucAccess;
+
+	if (i2c.mask_clk_reg)
+		i2c.valid = true;
+	else
+		i2c.valid = false;
+
+	return i2c;
+}
+
 static struct radeon_i2c_bus_rec radeon_lookup_i2c_gpio(struct radeon_device *rdev,
 							       uint8_t id)
 {
@@ -85,59 +166,10 @@ static struct radeon_i2c_bus_rec radeon_lookup_i2c_gpio(struct radeon_device *rd
 		for (i = 0; i < num_indices; i++) {
 			gpio = &i2c_info->asGPIO_Info[i];
 
-			/* some evergreen boards have bad data for this entry */
-			if (ASIC_IS_DCE4(rdev)) {
-				if ((i == 7) &&
-				    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1936) &&
-				    (gpio->sucI2cId.ucAccess == 0)) {
-					gpio->sucI2cId.ucAccess = 0x97;
-					gpio->ucDataMaskShift = 8;
-					gpio->ucDataEnShift = 8;
-					gpio->ucDataY_Shift = 8;
-					gpio->ucDataA_Shift = 8;
-				}
-			}
-
-			/* some DCE3 boards have bad data for this entry */
-			if (ASIC_IS_DCE3(rdev)) {
-				if ((i == 4) &&
-				    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1fda) &&
-				    (gpio->sucI2cId.ucAccess == 0x94))
-					gpio->sucI2cId.ucAccess = 0x14;
-			}
+			radeon_lookup_i2c_gpio_quirks(rdev, gpio, i);
 
 			if (gpio->sucI2cId.ucAccess == id) {
-				i2c.mask_clk_reg = le16_to_cpu(gpio->usClkMaskRegisterIndex) * 4;
-				i2c.mask_data_reg = le16_to_cpu(gpio->usDataMaskRegisterIndex) * 4;
-				i2c.en_clk_reg = le16_to_cpu(gpio->usClkEnRegisterIndex) * 4;
-				i2c.en_data_reg = le16_to_cpu(gpio->usDataEnRegisterIndex) * 4;
-				i2c.y_clk_reg = le16_to_cpu(gpio->usClkY_RegisterIndex) * 4;
-				i2c.y_data_reg = le16_to_cpu(gpio->usDataY_RegisterIndex) * 4;
-				i2c.a_clk_reg = le16_to_cpu(gpio->usClkA_RegisterIndex) * 4;
-				i2c.a_data_reg = le16_to_cpu(gpio->usDataA_RegisterIndex) * 4;
-				i2c.mask_clk_mask = (1 << gpio->ucClkMaskShift);
-				i2c.mask_data_mask = (1 << gpio->ucDataMaskShift);
-				i2c.en_clk_mask = (1 << gpio->ucClkEnShift);
-				i2c.en_data_mask = (1 << gpio->ucDataEnShift);
-				i2c.y_clk_mask = (1 << gpio->ucClkY_Shift);
-				i2c.y_data_mask = (1 << gpio->ucDataY_Shift);
-				i2c.a_clk_mask = (1 << gpio->ucClkA_Shift);
-				i2c.a_data_mask = (1 << gpio->ucDataA_Shift);
-
-				if (gpio->sucI2cId.sbfAccess.bfHW_Capable)
-					i2c.hw_capable = true;
-				else
-					i2c.hw_capable = false;
-
-				if (gpio->sucI2cId.ucAccess == 0xa0)
-					i2c.mm_i2c = true;
-				else
-					i2c.mm_i2c = false;
-
-				i2c.i2c_id = gpio->sucI2cId.ucAccess;
-
-				if (i2c.mask_clk_reg)
-					i2c.valid = true;
+				i2c = radeon_get_bus_rec_for_i2c_gpio(gpio);
 				break;
 			}
 		}
@@ -157,8 +189,6 @@ void radeon_atombios_i2c_init(struct radeon_device *rdev)
 	int i, num_indices;
 	char stmp[32];
 
-	memset(&i2c, 0, sizeof(struct radeon_i2c_bus_rec));
-
 	if (atom_parse_data_header(ctx, index, &size, NULL, NULL, &data_offset)) {
 		i2c_info = (struct _ATOM_GPIO_I2C_INFO *)(ctx->bios + data_offset);
 
@@ -167,60 +197,12 @@ void radeon_atombios_i2c_init(struct radeon_device *rdev)
 
 		for (i = 0; i < num_indices; i++) {
 			gpio = &i2c_info->asGPIO_Info[i];
-			i2c.valid = false;
-
-			/* some evergreen boards have bad data for this entry */
-			if (ASIC_IS_DCE4(rdev)) {
-				if ((i == 7) &&
-				    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1936) &&
-				    (gpio->sucI2cId.ucAccess == 0)) {
-					gpio->sucI2cId.ucAccess = 0x97;
-					gpio->ucDataMaskShift = 8;
-					gpio->ucDataEnShift = 8;
-					gpio->ucDataY_Shift = 8;
-					gpio->ucDataA_Shift = 8;
-				}
-			}
 
-			/* some DCE3 boards have bad data for this entry */
-			if (ASIC_IS_DCE3(rdev)) {
-				if ((i == 4) &&
-				    (le16_to_cpu(gpio->usClkMaskRegisterIndex) == 0x1fda) &&
-				    (gpio->sucI2cId.ucAccess == 0x94))
-					gpio->sucI2cId.ucAccess = 0x14;
-			}
+			radeon_lookup_i2c_gpio_quirks(rdev, gpio, i);
 
-			i2c.mask_clk_reg = le16_to_cpu(gpio->usClkMaskRegisterIndex) * 4;
-			i2c.mask_data_reg = le16_to_cpu(gpio->usDataMaskRegisterIndex) * 4;
-			i2c.en_clk_reg = le16_to_cpu(gpio->usClkEnRegisterIndex) * 4;
-			i2c.en_data_reg = le16_to_cpu(gpio->usDataEnRegisterIndex) * 4;
-			i2c.y_clk_reg = le16_to_cpu(gpio->usClkY_RegisterIndex) * 4;
-			i2c.y_data_reg = le16_to_cpu(gpio->usDataY_RegisterIndex) * 4;
-			i2c.a_clk_reg = le16_to_cpu(gpio->usClkA_RegisterIndex) * 4;
-			i2c.a_data_reg = le16_to_cpu(gpio->usDataA_RegisterIndex) * 4;
-			i2c.mask_clk_mask = (1 << gpio->ucClkMaskShift);
-			i2c.mask_data_mask = (1 << gpio->ucDataMaskShift);
-			i2c.en_clk_mask = (1 << gpio->ucClkEnShift);
-			i2c.en_data_mask = (1 << gpio->ucDataEnShift);
-			i2c.y_clk_mask = (1 << gpio->ucClkY_Shift);
-			i2c.y_data_mask = (1 << gpio->ucDataY_Shift);
-			i2c.a_clk_mask = (1 << gpio->ucClkA_Shift);
-			i2c.a_data_mask = (1 << gpio->ucDataA_Shift);
-
-			if (gpio->sucI2cId.sbfAccess.bfHW_Capable)
-				i2c.hw_capable = true;
-			else
-				i2c.hw_capable = false;
-
-			if (gpio->sucI2cId.ucAccess == 0xa0)
-				i2c.mm_i2c = true;
-			else
-				i2c.mm_i2c = false;
+			i2c = radeon_get_bus_rec_for_i2c_gpio(gpio);
 
-			i2c.i2c_id = gpio->sucI2cId.ucAccess;
-
-			if (i2c.mask_clk_reg) {
-				i2c.valid = true;
+			if (i2c.valid) {
 				sprintf(stmp, "0x%x", i2c.i2c_id);
 				rdev->i2c_bus[i] = radeon_i2c_create(rdev->ddev, &i2c, stmp);
 			}
@@ -1996,10 +1978,14 @@ static int radeon_atombios_parse_power_table_1_3(struct radeon_device *rdev)
 		return state_index;
 	/* last mode is usually default, array is low to high */
 	for (i = 0; i < num_modes; i++) {
+		rdev->pm.power_state[state_index].clock_info =
+			kzalloc(sizeof(struct radeon_pm_clock_info) * 1, GFP_KERNEL);
+		if (!rdev->pm.power_state[state_index].clock_info)
+			return state_index;
+		rdev->pm.power_state[state_index].num_clock_modes = 1;
 		rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
 		switch (frev) {
 		case 1:
-			rdev->pm.power_state[state_index].num_clock_modes = 1;
 			rdev->pm.power_state[state_index].clock_info[0].mclk =
 				le16_to_cpu(power_info->info.asPowerPlayInfo[i].usMemoryClock);
 			rdev->pm.power_state[state_index].clock_info[0].sclk =
@@ -2035,7 +2021,6 @@ static int radeon_atombios_parse_power_table_1_3(struct radeon_device *rdev)
 			state_index++;
 			break;
 		case 2:
-			rdev->pm.power_state[state_index].num_clock_modes = 1;
 			rdev->pm.power_state[state_index].clock_info[0].mclk =
 				le32_to_cpu(power_info->info_2.asPowerPlayInfo[i].ulMemoryClock);
 			rdev->pm.power_state[state_index].clock_info[0].sclk =
@@ -2072,7 +2057,6 @@ static int radeon_atombios_parse_power_table_1_3(struct radeon_device *rdev)
 			state_index++;
 			break;
 		case 3:
-			rdev->pm.power_state[state_index].num_clock_modes = 1;
 			rdev->pm.power_state[state_index].clock_info[0].mclk =
 				le32_to_cpu(power_info->info_3.asPowerPlayInfo[i].ulMemoryClock);
 			rdev->pm.power_state[state_index].clock_info[0].sclk =
@@ -2257,7 +2241,7 @@ static void radeon_atombios_parse_pplib_non_clock_info(struct radeon_device *rde
 		rdev->pm.default_power_state_index = state_index;
 		rdev->pm.power_state[state_index].default_clock_mode =
 			&rdev->pm.power_state[state_index].clock_info[mode_index - 1];
-		if (ASIC_IS_DCE5(rdev)) {
+		if (ASIC_IS_DCE5(rdev) && !(rdev->flags & RADEON_IS_IGP)) {
 			/* NI chips post without MC ucode, so default clocks are strobe mode only */
 			rdev->pm.default_sclk = rdev->pm.power_state[state_index].clock_info[0].sclk;
 			rdev->pm.default_mclk = rdev->pm.power_state[state_index].clock_info[0].mclk;
@@ -2377,17 +2361,31 @@ static int radeon_atombios_parse_power_table_4_5(struct radeon_device *rdev)
 			 le16_to_cpu(power_info->pplib.usNonClockInfoArrayOffset) +
 			 (power_state->v1.ucNonClockStateIndex *
 			  power_info->pplib.ucNonClockSize));
-		for (j = 0; j < (power_info->pplib.ucStateEntrySize - 1); j++) {
-			clock_info = (union pplib_clock_info *)
-				(mode_info->atom_context->bios + data_offset +
-				 le16_to_cpu(power_info->pplib.usClockInfoArrayOffset) +
-				 (power_state->v1.ucClockStateIndices[j] *
-				  power_info->pplib.ucClockInfoSize));
-			valid = radeon_atombios_parse_pplib_clock_info(rdev,
-								       state_index, mode_index,
-								       clock_info);
-			if (valid)
-				mode_index++;
+		rdev->pm.power_state[i].clock_info = kzalloc(sizeof(struct radeon_pm_clock_info) *
+							     ((power_info->pplib.ucStateEntrySize - 1) ?
+							      (power_info->pplib.ucStateEntrySize - 1) : 1),
+							     GFP_KERNEL);
+		if (!rdev->pm.power_state[i].clock_info)
+			return state_index;
+		if (power_info->pplib.ucStateEntrySize - 1) {
+			for (j = 0; j < (power_info->pplib.ucStateEntrySize - 1); j++) {
+				clock_info = (union pplib_clock_info *)
+					(mode_info->atom_context->bios + data_offset +
+					 le16_to_cpu(power_info->pplib.usClockInfoArrayOffset) +
+					 (power_state->v1.ucClockStateIndices[j] *
+					  power_info->pplib.ucClockInfoSize));
+				valid = radeon_atombios_parse_pplib_clock_info(rdev,
+									       state_index, mode_index,
+									       clock_info);
+				if (valid)
+					mode_index++;
+			}
+		} else {
+			rdev->pm.power_state[state_index].clock_info[0].mclk =
+				rdev->clock.default_mclk;
+			rdev->pm.power_state[state_index].clock_info[0].sclk =
+				rdev->clock.default_sclk;
+			mode_index++;
 		}
 		rdev->pm.power_state[state_index].num_clock_modes = mode_index;
 		if (mode_index) {
@@ -2456,18 +2454,32 @@ static int radeon_atombios_parse_power_table_6(struct radeon_device *rdev)
 		non_clock_array_index = i; /* power_state->v2.nonClockInfoIndex */
 		non_clock_info = (struct _ATOM_PPLIB_NONCLOCK_INFO *)
 			&non_clock_info_array->nonClockInfo[non_clock_array_index];
-		for (j = 0; j < power_state->v2.ucNumDPMLevels; j++) {
-			clock_array_index = power_state->v2.clockInfoIndex[j];
-			/* XXX this might be an inagua bug... */
-			if (clock_array_index >= clock_info_array->ucNumEntries)
-				continue;
-			clock_info = (union pplib_clock_info *)
-				&clock_info_array->clockInfo[clock_array_index];
-			valid = radeon_atombios_parse_pplib_clock_info(rdev,
-								       state_index, mode_index,
-								       clock_info);
-			if (valid)
-				mode_index++;
+		rdev->pm.power_state[i].clock_info = kzalloc(sizeof(struct radeon_pm_clock_info) *
+							     (power_state->v2.ucNumDPMLevels ?
+							      power_state->v2.ucNumDPMLevels : 1),
+							     GFP_KERNEL);
+		if (!rdev->pm.power_state[i].clock_info)
+			return state_index;
+		if (power_state->v2.ucNumDPMLevels) {
+			for (j = 0; j < power_state->v2.ucNumDPMLevels; j++) {
+				clock_array_index = power_state->v2.clockInfoIndex[j];
+				/* XXX this might be an inagua bug... */
+				if (clock_array_index >= clock_info_array->ucNumEntries)
+					continue;
+				clock_info = (union pplib_clock_info *)
+					&clock_info_array->clockInfo[clock_array_index];
+				valid = radeon_atombios_parse_pplib_clock_info(rdev,
+									       state_index, mode_index,
+									       clock_info);
+				if (valid)
+					mode_index++;
+			}
+		} else {
+			rdev->pm.power_state[state_index].clock_info[0].mclk =
+				rdev->clock.default_mclk;
+			rdev->pm.power_state[state_index].clock_info[0].sclk =
+				rdev->clock.default_sclk;
+			mode_index++;
 		}
 		rdev->pm.power_state[state_index].num_clock_modes = mode_index;
 		if (mode_index) {
@@ -2524,19 +2536,23 @@ void radeon_atombios_get_power_modes(struct radeon_device *rdev)
 	} else {
 		rdev->pm.power_state = kzalloc(sizeof(struct radeon_power_state), GFP_KERNEL);
 		if (rdev->pm.power_state) {
-			/* add the default mode */
-			rdev->pm.power_state[state_index].type =
-				POWER_STATE_TYPE_DEFAULT;
-			rdev->pm.power_state[state_index].num_clock_modes = 1;
-			rdev->pm.power_state[state_index].clock_info[0].mclk = rdev->clock.default_mclk;
-			rdev->pm.power_state[state_index].clock_info[0].sclk = rdev->clock.default_sclk;
-			rdev->pm.power_state[state_index].default_clock_mode =
-				&rdev->pm.power_state[state_index].clock_info[0];
-			rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
-			rdev->pm.power_state[state_index].pcie_lanes = 16;
-			rdev->pm.default_power_state_index = state_index;
-			rdev->pm.power_state[state_index].flags = 0;
-			state_index++;
+			rdev->pm.power_state[0].clock_info =
+				kzalloc(sizeof(struct radeon_pm_clock_info) * 1, GFP_KERNEL);
+			if (rdev->pm.power_state[0].clock_info) {
+				/* add the default mode */
+				rdev->pm.power_state[state_index].type =
+					POWER_STATE_TYPE_DEFAULT;
+				rdev->pm.power_state[state_index].num_clock_modes = 1;
+				rdev->pm.power_state[state_index].clock_info[0].mclk = rdev->clock.default_mclk;
+				rdev->pm.power_state[state_index].clock_info[0].sclk = rdev->clock.default_sclk;
+				rdev->pm.power_state[state_index].default_clock_mode =
+					&rdev->pm.power_state[state_index].clock_info[0];
+				rdev->pm.power_state[state_index].clock_info[0].voltage.type = VOLTAGE_NONE;
+				rdev->pm.power_state[state_index].pcie_lanes = 16;
+				rdev->pm.default_power_state_index = state_index;
+				rdev->pm.power_state[state_index].flags = 0;
+				state_index++;
+			}
 		}
 	}
 
diff --git a/drivers/gpu/drm/radeon/radeon_benchmark.c b/drivers/gpu/drm/radeon/radeon_benchmark.c
index 5cafc90de7f8..17e1a9b2d8fb 100644
--- a/drivers/gpu/drm/radeon/radeon_benchmark.c
+++ b/drivers/gpu/drm/radeon/radeon_benchmark.c
@@ -98,7 +98,7 @@ static void radeon_benchmark_move(struct radeon_device *rdev, unsigned size,
 	struct radeon_bo *sobj = NULL;
 	uint64_t saddr, daddr;
 	int r, n;
-	unsigned int time;
+	int time;
 
 	n = RADEON_BENCHMARK_ITERATIONS;
 	r = radeon_bo_create(rdev, size, PAGE_SIZE, true, sdomain, &sobj);
diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c
index 8bf83c4b4147..81fc100be7e1 100644
--- a/drivers/gpu/drm/radeon/radeon_combios.c
+++ b/drivers/gpu/drm/radeon/radeon_combios.c
@@ -2563,14 +2563,17 @@ void radeon_combios_get_power_modes(struct radeon_device *rdev)
 
 	/* allocate 2 power states */
 	rdev->pm.power_state = kzalloc(sizeof(struct radeon_power_state) * 2, GFP_KERNEL);
-	if (!rdev->pm.power_state) {
-		rdev->pm.default_power_state_index = state_index;
-		rdev->pm.num_power_states = 0;
-
-		rdev->pm.current_power_state_index = rdev->pm.default_power_state_index;
-		rdev->pm.current_clock_mode_index = 0;
-		return;
-	}
+	if (rdev->pm.power_state) {
+		/* allocate 1 clock mode per state */
+		rdev->pm.power_state[0].clock_info =
+			kzalloc(sizeof(struct radeon_pm_clock_info) * 1, GFP_KERNEL);
+		rdev->pm.power_state[1].clock_info =
+			kzalloc(sizeof(struct radeon_pm_clock_info) * 1, GFP_KERNEL);
+		if (!rdev->pm.power_state[0].clock_info ||
+		    !rdev->pm.power_state[1].clock_info)
+			goto pm_failed;
+	} else
+		goto pm_failed;
 
 	/* check for a thermal chip */
 	offset = combios_get_table_offset(dev, COMBIOS_OVERDRIVE_INFO_TABLE);
@@ -2735,6 +2738,14 @@ default_mode:
 
 	rdev->pm.current_power_state_index = rdev->pm.default_power_state_index;
 	rdev->pm.current_clock_mode_index = 0;
+	return;
+
+pm_failed:
+	rdev->pm.default_power_state_index = state_index;
+	rdev->pm.num_power_states = 0;
+
+	rdev->pm.current_power_state_index = rdev->pm.default_power_state_index;
+	rdev->pm.current_clock_mode_index = 0;
 }
 
 void radeon_external_tmds_setup(struct drm_encoder *encoder)
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index fae00c0d75aa..29afd71e0840 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -93,7 +93,7 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 {
 	struct drm_radeon_cs *cs = data;
 	uint64_t *chunk_array_ptr;
-	unsigned size, i;
+	unsigned size, i, flags = 0;
 
 	if (!cs->num_chunks) {
 		return 0;
@@ -140,6 +140,10 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 			if (p->chunks[i].length_dw == 0)
 				return -EINVAL;
 		}
+		if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS &&
+		    !p->chunks[i].length_dw) {
+			return -EINVAL;
+		}
 
 		p->chunks[i].length_dw = user_chunk.length_dw;
 		p->chunks[i].user_ptr = (void __user *)(unsigned long)user_chunk.chunk_data;
@@ -155,6 +159,9 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 					       p->chunks[i].user_ptr, size)) {
 				return -EFAULT;
 			}
+			if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS) {
+				flags = p->chunks[i].kdata[0];
+			}
 		} else {
 			p->chunks[i].kpage[0] = kmalloc(PAGE_SIZE, GFP_KERNEL);
 			p->chunks[i].kpage[1] = kmalloc(PAGE_SIZE, GFP_KERNEL);
@@ -174,6 +181,8 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 			  p->chunks[p->chunk_ib_idx].length_dw);
 		return -EINVAL;
 	}
+
+	p->keep_tiling_flags = (flags & RADEON_CS_KEEP_TILING_FLAGS) != 0;
 	return 0;
 }
 
@@ -222,7 +231,7 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	struct radeon_cs_chunk *ib_chunk;
 	int r;
 
-	mutex_lock(&rdev->cs_mutex);
+	radeon_mutex_lock(&rdev->cs_mutex);
 	/* initialize parser */
 	memset(&parser, 0, sizeof(struct radeon_cs_parser));
 	parser.filp = filp;
@@ -233,14 +242,14 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	if (r) {
 		DRM_ERROR("Failed to initialize parser !\n");
 		radeon_cs_parser_fini(&parser, r);
-		mutex_unlock(&rdev->cs_mutex);
+		radeon_mutex_unlock(&rdev->cs_mutex);
 		return r;
 	}
 	r =  radeon_ib_get(rdev, &parser.ib);
 	if (r) {
 		DRM_ERROR("Failed to get ib !\n");
 		radeon_cs_parser_fini(&parser, r);
-		mutex_unlock(&rdev->cs_mutex);
+		radeon_mutex_unlock(&rdev->cs_mutex);
 		return r;
 	}
 	r = radeon_cs_parser_relocs(&parser);
@@ -248,7 +257,7 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		if (r != -ERESTARTSYS)
 			DRM_ERROR("Failed to parse relocation %d!\n", r);
 		radeon_cs_parser_fini(&parser, r);
-		mutex_unlock(&rdev->cs_mutex);
+		radeon_mutex_unlock(&rdev->cs_mutex);
 		return r;
 	}
 	/* Copy the packet into the IB, the parser will read from the
@@ -260,14 +269,14 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	if (r || parser.parser_error) {
 		DRM_ERROR("Invalid command stream !\n");
 		radeon_cs_parser_fini(&parser, r);
-		mutex_unlock(&rdev->cs_mutex);
+		radeon_mutex_unlock(&rdev->cs_mutex);
 		return r;
 	}
 	r = radeon_cs_finish_pages(&parser);
 	if (r) {
 		DRM_ERROR("Invalid command stream !\n");
 		radeon_cs_parser_fini(&parser, r);
-		mutex_unlock(&rdev->cs_mutex);
+		radeon_mutex_unlock(&rdev->cs_mutex);
 		return r;
 	}
 	r = radeon_ib_schedule(rdev, parser.ib);
@@ -275,7 +284,7 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		DRM_ERROR("Failed to schedule IB !\n");
 	}
 	radeon_cs_parser_fini(&parser, r);
-	mutex_unlock(&rdev->cs_mutex);
+	radeon_mutex_unlock(&rdev->cs_mutex);
 	return r;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 7c31321df45b..fb347a80486f 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -716,7 +716,7 @@ int radeon_device_init(struct radeon_device *rdev,
 
 	/* mutex initialization are all done here so we
 	 * can recall function without having locking issues */
-	mutex_init(&rdev->cs_mutex);
+	radeon_mutex_init(&rdev->cs_mutex);
 	mutex_init(&rdev->ib_pool.mutex);
 	mutex_init(&rdev->cp.mutex);
 	mutex_init(&rdev->dc_hw_i2c_mutex);
@@ -961,6 +961,9 @@ int radeon_gpu_reset(struct radeon_device *rdev)
 	int r;
 	int resched;
 
+	/* Prevent CS ioctl from interfering */
+	radeon_mutex_lock(&rdev->cs_mutex);
+
 	radeon_save_bios_scratch_regs(rdev);
 	/* block TTM */
 	resched = ttm_bo_lock_delayed_workqueue(&rdev->mman.bdev);
@@ -973,10 +976,15 @@ int radeon_gpu_reset(struct radeon_device *rdev)
 		radeon_restore_bios_scratch_regs(rdev);
 		drm_helper_resume_force_mode(rdev->ddev);
 		ttm_bo_unlock_delayed_workqueue(&rdev->mman.bdev, resched);
-		return 0;
 	}
-	/* bad news, how to tell it to userspace ? */
-	dev_info(rdev->dev, "GPU reset failed\n");
+
+	radeon_mutex_unlock(&rdev->cs_mutex);
+
+	if (r) {
+		/* bad news, how to tell it to userspace ? */
+		dev_info(rdev->dev, "GPU reset failed\n");
+	}
+
 	return r;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index e42c34b98c7b..c3ef1d266f88 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -53,9 +53,10 @@
  *   2.9.0 - r600 tiling (s3tc,rgtc) working, SET_PREDICATION packet 3 on r600 + eg, backend query
  *   2.10.0 - fusion 2D tiling
  *   2.11.0 - backend map, initial compute support for the CS checker
+ *   2.12.0 - RADEON_CS_KEEP_TILING_FLAGS
  */
 #define KMS_DRIVER_MAJOR	2
-#define KMS_DRIVER_MINOR	11
+#define KMS_DRIVER_MINOR	12
 #define KMS_DRIVER_PATCHLEVEL	0
 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags);
 int radeon_driver_unload_kms(struct drm_device *dev);
diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c
index 06e413e6a920..4b27efa4405b 100644
--- a/drivers/gpu/drm/radeon/radeon_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_encoders.c
@@ -233,13 +233,12 @@ u16 radeon_encoder_get_dp_bridge_encoder_id(struct drm_encoder *encoder)
 		switch (radeon_encoder->encoder_id) {
 		case ENCODER_OBJECT_ID_TRAVIS:
 		case ENCODER_OBJECT_ID_NUTMEG:
-			return true;
+			return radeon_encoder->encoder_id;
 		default:
-			return false;
+			return ENCODER_OBJECT_ID_NONE;
 		}
 	}
-
-	return false;
+	return ENCODER_OBJECT_ID_NONE;
 }
 
 void radeon_panel_mode_fixup(struct drm_encoder *encoder,
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
index 95b93604b679..25a19c483075 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
@@ -991,12 +991,6 @@ static bool radeon_crtc_mode_fixup(struct drm_crtc *crtc,
 				   struct drm_display_mode *mode,
 				   struct drm_display_mode *adjusted_mode)
 {
-	struct drm_device *dev = crtc->dev;
-	struct radeon_device *rdev = dev->dev_private;
-
-	/* adjust pm to upcoming mode change */
-	radeon_pm_compute_clocks(rdev);
-
 	if (!radeon_crtc_scaling_mode_fixup(crtc, mode, adjusted_mode))
 		return false;
 	return true;
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 6fabe89fa6a1..78a665bd9519 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -53,6 +53,24 @@ static void radeon_pm_set_clocks(struct radeon_device *rdev);
 
 #define ACPI_AC_CLASS           "ac_adapter"
 
+int radeon_pm_get_type_index(struct radeon_device *rdev,
+			     enum radeon_pm_state_type ps_type,
+			     int instance)
+{
+	int i;
+	int found_instance = -1;
+
+	for (i = 0; i < rdev->pm.num_power_states; i++) {
+		if (rdev->pm.power_state[i].type == ps_type) {
+			found_instance++;
+			if (found_instance == instance)
+				return i;
+		}
+	}
+	/* return default if no match */
+	return rdev->pm.default_power_state_index;
+}
+
 #ifdef CONFIG_ACPI
 static int radeon_acpi_event(struct notifier_block *nb,
 			     unsigned long val,
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index 481b99e89f65..b1053d640423 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -62,6 +62,7 @@ u32 rs600_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset);
+	int i;
 
 	/* Lock the graphics update lock */
 	tmp |= AVIVO_D1GRPH_UPDATE_LOCK;
@@ -74,7 +75,11 @@ u32 rs600_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	       (u32)crtc_base);
 
 	/* Wait for update_pending to go high. */
-	while (!(RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING));
+	for (i = 0; i < rdev->usec_timeout; i++) {
+		if (RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING)
+			break;
+		udelay(1);
+	}
 	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
 
 	/* Unlock the lock, so double-buffering can take place inside vblank */
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index a983f410ab89..23ae1c60ab3d 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -47,6 +47,7 @@ u32 rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
 	u32 tmp = RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset);
+	int i;
 
 	/* Lock the graphics update lock */
 	tmp |= AVIVO_D1GRPH_UPDATE_LOCK;
@@ -66,7 +67,11 @@ u32 rv770_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 	       (u32)crtc_base);
 
 	/* Wait for update_pending to go high. */
-	while (!(RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING));
+	for (i = 0; i < rdev->usec_timeout; i++) {
+		if (RREG32(AVIVO_D1GRPH_UPDATE + radeon_crtc->crtc_offset) & AVIVO_D1GRPH_SURFACE_UPDATE_PENDING)
+			break;
+		udelay(1);
+	}
 	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
 
 	/* Unlock the lock, so double-buffering can take place inside vblank */
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 119b6e3ff906..2f0eab66ece6 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -562,10 +562,16 @@ retry:
 		return ret;
 
 	spin_lock(&glob->lru_lock);
+
+	if (unlikely(list_empty(&bo->ddestroy))) {
+		spin_unlock(&glob->lru_lock);
+		return 0;
+	}
+
 	ret = ttm_bo_reserve_locked(bo, interruptible,
 				    no_wait_reserve, false, 0);
 
-	if (unlikely(ret != 0) || list_empty(&bo->ddestroy)) {
+	if (unlikely(ret != 0)) {
 		spin_unlock(&glob->lru_lock);
 		return ret;
 	}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
index 3f6343502d1f..5ff561d4e0b4 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
@@ -140,7 +140,7 @@ int vmw_present_ioctl(struct drm_device *dev, void *data,
 		goto out_clips;
 	}
 
-	clips = kzalloc(num_clips * sizeof(*clips), GFP_KERNEL);
+	clips = kcalloc(num_clips, sizeof(*clips), GFP_KERNEL);
 	if (clips == NULL) {
 		DRM_ERROR("Failed to allocate clip rect list.\n");
 		ret = -ENOMEM;
@@ -232,7 +232,7 @@ int vmw_present_readback_ioctl(struct drm_device *dev, void *data,
 		goto out_clips;
 	}
 
-	clips = kzalloc(num_clips * sizeof(*clips), GFP_KERNEL);
+	clips = kcalloc(num_clips, sizeof(*clips), GFP_KERNEL);
 	if (clips == NULL) {
 		DRM_ERROR("Failed to allocate clip rect list.\n");
 		ret = -ENOMEM;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index 0585987f2945..1748a7142aca 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -105,6 +105,10 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
 	struct vmw_dma_buffer *dmabuf = NULL;
 	int ret;
 
+	/* A lot of the code assumes this */
+	if (handle && (width != 64 || height != 64))
+		return -EINVAL;
+
 	if (handle) {
 		ret = vmw_user_surface_lookup_handle(dev_priv, tfile,
 						     handle, &surface);
@@ -410,8 +414,9 @@ static int do_surface_dirty_sou(struct vmw_private *dev_priv,
 	top = clips->y1;
 	bottom = clips->y2;
 
-	clips_ptr = clips;
-	for (i = 1; i < num_clips; i++, clips_ptr += inc) {
+	/* skip the first clip rect */
+	for (i = 1, clips_ptr = clips + inc;
+	     i < num_clips; i++, clips_ptr += inc) {
 		left = min_t(int, left, (int)clips_ptr->x1);
 		right = max_t(int, right, (int)clips_ptr->x2);
 		top = min_t(int, top, (int)clips_ptr->y1);
@@ -1331,7 +1336,10 @@ int vmw_kms_close(struct vmw_private *dev_priv)
 	 * drm_encoder_cleanup which takes the lock we deadlock.
 	 */
 	drm_mode_config_cleanup(dev_priv->dev);
-	vmw_kms_close_legacy_display_system(dev_priv);
+	if (dev_priv->sou_priv)
+		vmw_kms_close_screen_object_display(dev_priv);
+	else
+		vmw_kms_close_legacy_display_system(dev_priv);
 	return 0;
 }
 
@@ -1809,7 +1817,8 @@ int vmw_kms_update_layout_ioctl(struct drm_device *dev, void *data,
 	}
 
 	rects_size = arg->num_outputs * sizeof(struct drm_vmw_rect);
-	rects = kzalloc(rects_size, GFP_KERNEL);
+	rects = kcalloc(arg->num_outputs, sizeof(struct drm_vmw_rect),
+			GFP_KERNEL);
 	if (unlikely(!rects)) {
 		ret = -ENOMEM;
 		goto out_unlock;
@@ -1824,10 +1833,10 @@ int vmw_kms_update_layout_ioctl(struct drm_device *dev, void *data,
 	}
 
 	for (i = 0; i < arg->num_outputs; ++i) {
-		if (rects->x < 0 ||
-		    rects->y < 0 ||
-		    rects->x + rects->w > mode_config->max_width ||
-		    rects->y + rects->h > mode_config->max_height) {
+		if (rects[i].x < 0 ||
+		    rects[i].y < 0 ||
+		    rects[i].x + rects[i].w > mode_config->max_width ||
+		    rects[i].y + rects[i].h > mode_config->max_height) {
 			DRM_ERROR("Invalid GUI layout.\n");
 			ret = -EINVAL;
 			goto out_free;