20 files changed, 677 insertions, 251 deletions

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index d09a6e02dc95..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);
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 15bfa9145d2b..a1103fc6597d 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -58,15 +58,15 @@ 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)");
 
 int i915_enable_fbc __read_mostly = -1;
 module_param_named(i915_enable_fbc, i915_enable_fbc, int, 0600);
@@ -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)
@@ -903,8 +925,9 @@ 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 4a9c1b979804..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;
@@ -221,6 +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 */
@@ -710,6 +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;
@@ -998,11 +1002,11 @@ 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 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 int i915_enable_rc6 __read_mostly;
 extern int i915_enable_fbc __read_mostly;
 extern bool i915_enable_hangcheck __read_mostly;
 
@@ -1308,6 +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);
@@ -1352,8 +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_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_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index b080cc824001..a26d5b0a3690 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -3303,10 +3303,10 @@
 /* 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)
@@ -3447,8 +3447,30 @@
 #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
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index e77a863a3833..d809b038ca88 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))
 
@@ -4670,6 +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.
@@ -4681,13 +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;
@@ -4758,6 +4761,11 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
 		}
 	}
 
+	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
@@ -5019,6 +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');
@@ -5480,7 +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;
@@ -7189,11 +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;
@@ -7212,10 +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;
@@ -7238,7 +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);
@@ -7247,6 +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;
@@ -7887,6 +7907,33 @@ 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);
@@ -7923,7 +7970,7 @@ 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;
 
@@ -8372,7 +8419,7 @@ 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);
@@ -8491,6 +8538,28 @@ 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 4d0358fad937..92b041b66e49 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -208,13 +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;
@@ -233,6 +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)
@@ -242,9 +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;
@@ -362,8 +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))
@@ -672,6 +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) {
@@ -689,7 +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;
@@ -817,10 +829,11 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
 	}
 
 	/*
-	 * There are three kinds of DP registers:
+	 * There are four kinds of DP registers:
 	 *
 	 * 	IBX PCH
-	 * 	CPU
+	 * 	SNB CPU
+	 *	IVB CPU
 	 * 	CPT PCH
 	 *
 	 * IBX PCH and CPU are the same for almost everything,
@@ -873,7 +886,25 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
 
 	/* Split out the IBX/CPU vs CPT settings */
 
-	if (!HAS_PCH_CPT(dev) || is_cpu_edp(intel_dp)) {
+	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;
+
+		/* 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)
@@ -1375,34 +1406,59 @@ 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
-#define I830_DP_VOLTAGE_MAX_CPT	    DP_TRAIN_VOLTAGE_SWING_1200
 
 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, uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
-	struct drm_device *dev = intel_dp->base.base.dev;
 	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);
-	int voltage_max;
+	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(adjust_request, lane);
@@ -1414,15 +1470,13 @@ intel_get_adjust_train(struct intel_dp *intel_dp, uint8_t link_status[DP_LINK_ST
 			p = this_p;
 	}
 
-	if (HAS_PCH_CPT(dev) && !is_cpu_edp(intel_dp))
-		voltage_max = I830_DP_VOLTAGE_MAX_CPT;
-	else
-		voltage_max = I830_DP_VOLTAGE_MAX;
+	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;
@@ -1494,6 +1548,37 @@ 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)
@@ -1599,7 +1684,8 @@ 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;
@@ -1613,7 +1699,11 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 		uint8_t	    link_status[DP_LINK_STATUS_SIZE];
 		uint32_t    signal_levels;
 
-		if (IS_GEN6(dev) && is_cpu_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 {
@@ -1622,7 +1712,7 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 			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;
@@ -1703,7 +1793,10 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 			break;
 		}
 
-		if (IS_GEN6(dev) && is_cpu_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 {
@@ -1711,7 +1804,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 			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;
@@ -1752,7 +1845,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 		++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;
@@ -1782,7 +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 {
@@ -1794,7 +1887,7 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 	msleep(17);
 
 	if (is_edp(intel_dp)) {
-		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_OFF_CPT;
 		else
 			DP |= DP_LINK_TRAIN_OFF;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index bd9a604b73da..a1b4343814e8 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 21f60b7d69a3..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);
 }
 
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/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/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 8cca91a93bde..dc279706ca70 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -390,6 +390,11 @@ extern int vmw_context_check(struct vmw_private *dev_priv,
 			     struct ttm_object_file *tfile,
 			     int id,
 			     struct vmw_resource **p_res);
+extern int vmw_user_lookup_handle(struct vmw_private *dev_priv,
+				  struct ttm_object_file *tfile,
+				  uint32_t handle,
+				  struct vmw_surface **out_surf,
+				  struct vmw_dma_buffer **out_buf);
 extern void vmw_surface_res_free(struct vmw_resource *res);
 extern int vmw_surface_init(struct vmw_private *dev_priv,
 			    struct vmw_surface *srf,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
index 03bbc2a6f9a7..a0c2f12b1e1b 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
@@ -33,6 +33,7 @@ bool vmw_fifo_have_3d(struct vmw_private *dev_priv)
 {
 	__le32 __iomem *fifo_mem = dev_priv->mmio_virt;
 	uint32_t fifo_min, hwversion;
+	const struct vmw_fifo_state *fifo = &dev_priv->fifo;
 
 	if (!(dev_priv->capabilities & SVGA_CAP_EXTENDED_FIFO))
 		return false;
@@ -41,7 +42,12 @@ bool vmw_fifo_have_3d(struct vmw_private *dev_priv)
 	if (fifo_min <= SVGA_FIFO_3D_HWVERSION * sizeof(unsigned int))
 		return false;
 
-	hwversion = ioread32(fifo_mem + SVGA_FIFO_3D_HWVERSION);
+	hwversion = ioread32(fifo_mem +
+			     ((fifo->capabilities &
+			       SVGA_FIFO_CAP_3D_HWVERSION_REVISED) ?
+			      SVGA_FIFO_3D_HWVERSION_REVISED :
+			      SVGA_FIFO_3D_HWVERSION));
+
 	if (hwversion == 0)
 		return false;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
index 3f6343502d1f..66917c6c3813 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
@@ -58,8 +58,14 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data,
 	case DRM_VMW_PARAM_FIFO_HW_VERSION:
 	{
 		__le32 __iomem *fifo_mem = dev_priv->mmio_virt;
-
-		param->value = ioread32(fifo_mem + SVGA_FIFO_3D_HWVERSION);
+		const struct vmw_fifo_state *fifo = &dev_priv->fifo;
+
+		param->value =
+			ioread32(fifo_mem +
+				 ((fifo->capabilities &
+				   SVGA_FIFO_CAP_3D_HWVERSION_REVISED) ?
+				  SVGA_FIFO_3D_HWVERSION_REVISED :
+				  SVGA_FIFO_3D_HWVERSION));
 		break;
 	}
 	default:
@@ -140,7 +146,7 @@ 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;
@@ -166,13 +172,7 @@ int vmw_present_ioctl(struct drm_device *dev, void *data,
 		ret = -EINVAL;
 		goto out_no_fb;
 	}
-
 	vfb = vmw_framebuffer_to_vfb(obj_to_fb(obj));
-	if (!vfb->dmabuf) {
-		DRM_ERROR("Framebuffer not dmabuf backed.\n");
-		ret = -EINVAL;
-		goto out_no_fb;
-	}
 
 	ret = ttm_read_lock(&vmaster->lock, true);
 	if (unlikely(ret != 0))
@@ -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 37d40545ed77..8aa1dbb45c67 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -31,6 +31,44 @@
 /* Might need a hrtimer here? */
 #define VMWGFX_PRESENT_RATE ((HZ / 60 > 0) ? HZ / 60 : 1)
 
+
+struct vmw_clip_rect {
+	int x1, x2, y1, y2;
+};
+
+/**
+ * Clip @num_rects number of @rects against @clip storing the
+ * results in @out_rects and the number of passed rects in @out_num.
+ */
+void vmw_clip_cliprects(struct drm_clip_rect *rects,
+			int num_rects,
+			struct vmw_clip_rect clip,
+			SVGASignedRect *out_rects,
+			int *out_num)
+{
+	int i, k;
+
+	for (i = 0, k = 0; i < num_rects; i++) {
+		int x1 = max_t(int, clip.x1, rects[i].x1);
+		int y1 = max_t(int, clip.y1, rects[i].y1);
+		int x2 = min_t(int, clip.x2, rects[i].x2);
+		int y2 = min_t(int, clip.y2, rects[i].y2);
+
+		if (x1 >= x2)
+			continue;
+		if (y1 >= y2)
+			continue;
+
+		out_rects[k].left   = x1;
+		out_rects[k].top    = y1;
+		out_rects[k].right  = x2;
+		out_rects[k].bottom = y2;
+		k++;
+	}
+
+	*out_num = k;
+}
+
 void vmw_display_unit_cleanup(struct vmw_display_unit *du)
 {
 	if (du->cursor_surface)
@@ -82,6 +120,43 @@ int vmw_cursor_update_image(struct vmw_private *dev_priv,
 	return 0;
 }
 
+int vmw_cursor_update_dmabuf(struct vmw_private *dev_priv,
+			     struct vmw_dma_buffer *dmabuf,
+			     u32 width, u32 height,
+			     u32 hotspotX, u32 hotspotY)
+{
+	struct ttm_bo_kmap_obj map;
+	unsigned long kmap_offset;
+	unsigned long kmap_num;
+	void *virtual;
+	bool dummy;
+	int ret;
+
+	kmap_offset = 0;
+	kmap_num = (width*height*4 + PAGE_SIZE - 1) >> PAGE_SHIFT;
+
+	ret = ttm_bo_reserve(&dmabuf->base, true, false, false, 0);
+	if (unlikely(ret != 0)) {
+		DRM_ERROR("reserve failed\n");
+		return -EINVAL;
+	}
+
+	ret = ttm_bo_kmap(&dmabuf->base, kmap_offset, kmap_num, &map);
+	if (unlikely(ret != 0))
+		goto err_unreserve;
+
+	virtual = ttm_kmap_obj_virtual(&map, &dummy);
+	ret = vmw_cursor_update_image(dev_priv, virtual, width, height,
+				      hotspotX, hotspotY);
+
+	ttm_bo_kunmap(&map);
+err_unreserve:
+	ttm_bo_unreserve(&dmabuf->base);
+
+	return ret;
+}
+
+
 void vmw_cursor_update_position(struct vmw_private *dev_priv,
 				bool show, int x, int y)
 {
@@ -110,24 +185,21 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
 		return -EINVAL;
 
 	if (handle) {
-		ret = vmw_user_surface_lookup_handle(dev_priv, tfile,
-						     handle, &surface);
-		if (!ret) {
-			if (!surface->snooper.image) {
-				DRM_ERROR("surface not suitable for cursor\n");
-				vmw_surface_unreference(&surface);
-				return -EINVAL;
-			}
-		} else {
-			ret = vmw_user_dmabuf_lookup(tfile,
-						     handle, &dmabuf);
-			if (ret) {
-				DRM_ERROR("failed to find surface or dmabuf: %i\n", ret);
-				return -EINVAL;
-			}
+		ret = vmw_user_lookup_handle(dev_priv, tfile,
+					     handle, &surface, &dmabuf);
+		if (ret) {
+			DRM_ERROR("failed to find surface or dmabuf: %i\n", ret);
+			return -EINVAL;
 		}
 	}
 
+	/* need to do this before taking down old image */
+	if (surface && !surface->snooper.image) {
+		DRM_ERROR("surface not suitable for cursor\n");
+		vmw_surface_unreference(&surface);
+		return -EINVAL;
+	}
+
 	/* takedown old cursor */
 	if (du->cursor_surface) {
 		du->cursor_surface->snooper.crtc = NULL;
@@ -146,36 +218,11 @@ int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
 		vmw_cursor_update_image(dev_priv, surface->snooper.image,
 					64, 64, du->hotspot_x, du->hotspot_y);
 	} else if (dmabuf) {
-		struct ttm_bo_kmap_obj map;
-		unsigned long kmap_offset;
-		unsigned long kmap_num;
-		void *virtual;
-		bool dummy;
-
 		/* vmw_user_surface_lookup takes one reference */
 		du->cursor_dmabuf = dmabuf;
 
-		kmap_offset = 0;
-		kmap_num = (64*64*4) >> PAGE_SHIFT;
-
-		ret = ttm_bo_reserve(&dmabuf->base, true, false, false, 0);
-		if (unlikely(ret != 0)) {
-			DRM_ERROR("reserve failed\n");
-			return -EINVAL;
-		}
-
-		ret = ttm_bo_kmap(&dmabuf->base, kmap_offset, kmap_num, &map);
-		if (unlikely(ret != 0))
-			goto err_unreserve;
-
-		virtual = ttm_kmap_obj_virtual(&map, &dummy);
-		vmw_cursor_update_image(dev_priv, virtual, 64, 64,
-					du->hotspot_x, du->hotspot_y);
-
-		ttm_bo_kunmap(&map);
-err_unreserve:
-		ttm_bo_unreserve(&dmabuf->base);
-
+		ret = vmw_cursor_update_dmabuf(dev_priv, dmabuf, width, height,
+					       du->hotspot_x, du->hotspot_y);
 	} else {
 		vmw_cursor_update_position(dev_priv, false, 0, 0);
 		return 0;
@@ -377,8 +424,9 @@ static int do_surface_dirty_sou(struct vmw_private *dev_priv,
 				struct drm_clip_rect *clips,
 				unsigned num_clips, int inc)
 {
-	struct drm_clip_rect *clips_ptr;
 	struct vmw_display_unit *units[VMWGFX_NUM_DISPLAY_UNITS];
+	struct drm_clip_rect *clips_ptr;
+	struct drm_clip_rect *tmp;
 	struct drm_crtc *crtc;
 	size_t fifo_size;
 	int i, num_units;
@@ -391,7 +439,6 @@ static int do_surface_dirty_sou(struct vmw_private *dev_priv,
 	} *cmd;
 	SVGASignedRect *blits;
 
-
 	num_units = 0;
 	list_for_each_entry(crtc, &dev_priv->dev->mode_config.crtc_list,
 			    head) {
@@ -402,13 +449,24 @@ static int do_surface_dirty_sou(struct vmw_private *dev_priv,
 
 	BUG_ON(!clips || !num_clips);
 
+	tmp = kzalloc(sizeof(*tmp) * num_clips, GFP_KERNEL);
+	if (unlikely(tmp == NULL)) {
+		DRM_ERROR("Temporary cliprect memory alloc failed.\n");
+		return -ENOMEM;
+	}
+
 	fifo_size = sizeof(*cmd) + sizeof(SVGASignedRect) * num_clips;
 	cmd = kzalloc(fifo_size, GFP_KERNEL);
 	if (unlikely(cmd == NULL)) {
 		DRM_ERROR("Temporary fifo memory alloc failed.\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto out_free_tmp;
 	}
 
+	/* setup blits pointer */
+	blits = (SVGASignedRect *)&cmd[1];
+
+	/* initial clip region */
 	left = clips->x1;
 	right = clips->x2;
 	top = clips->y1;
@@ -434,45 +492,60 @@ static int do_surface_dirty_sou(struct vmw_private *dev_priv,
 	cmd->body.srcRect.bottom = bottom;
 
 	clips_ptr = clips;
-	blits = (SVGASignedRect *)&cmd[1];
 	for (i = 0; i < num_clips; i++, clips_ptr += inc) {
-		blits[i].left   = clips_ptr->x1 - left;
-		blits[i].right  = clips_ptr->x2 - left;
-		blits[i].top    = clips_ptr->y1 - top;
-		blits[i].bottom = clips_ptr->y2 - top;
+		tmp[i].x1 = clips_ptr->x1 - left;
+		tmp[i].x2 = clips_ptr->x2 - left;
+		tmp[i].y1 = clips_ptr->y1 - top;
+		tmp[i].y2 = clips_ptr->y2 - top;
 	}
 
 	/* do per unit writing, reuse fifo for each */
 	for (i = 0; i < num_units; i++) {
 		struct vmw_display_unit *unit = units[i];
-		int clip_x1 = left - unit->crtc.x;
-		int clip_y1 = top - unit->crtc.y;
-		int clip_x2 = right - unit->crtc.x;
-		int clip_y2 = bottom - unit->crtc.y;
+		struct vmw_clip_rect clip;
+		int num;
+
+		clip.x1 = left - unit->crtc.x;
+		clip.y1 = top - unit->crtc.y;
+		clip.x2 = right - unit->crtc.x;
+		clip.y2 = bottom - unit->crtc.y;
 
 		/* skip any crtcs that misses the clip region */
-		if (clip_x1 >= unit->crtc.mode.hdisplay ||
-		    clip_y1 >= unit->crtc.mode.vdisplay ||
-		    clip_x2 <= 0 || clip_y2 <= 0)
+		if (clip.x1 >= unit->crtc.mode.hdisplay ||
+		    clip.y1 >= unit->crtc.mode.vdisplay ||
+		    clip.x2 <= 0 || clip.y2 <= 0)
 			continue;
 
+		/*
+		 * In order for the clip rects to be correctly scaled
+		 * the src and dest rects needs to be the same size.
+		 */
+		cmd->body.destRect.left = clip.x1;
+		cmd->body.destRect.right = clip.x2;
+		cmd->body.destRect.top = clip.y1;
+		cmd->body.destRect.bottom = clip.y2;
+
+		/* create a clip rect of the crtc in dest coords */
+		clip.x2 = unit->crtc.mode.hdisplay - clip.x1;
+		clip.y2 = unit->crtc.mode.vdisplay - clip.y1;
+		clip.x1 = 0 - clip.x1;
+		clip.y1 = 0 - clip.y1;
+
 		/* need to reset sid as it is changed by execbuf */
 		cmd->body.srcImage.sid = cpu_to_le32(framebuffer->user_handle);
-
 		cmd->body.destScreenId = unit->unit;
 
-		/*
-		 * The blit command is a lot more resilient then the
-		 * readback command when it comes to clip rects. So its
-		 * okay to go out of bounds.
-		 */
+		/* clip and write blits to cmd stream */
+		vmw_clip_cliprects(tmp, num_clips, clip, blits, &num);
 
-		cmd->body.destRect.left = clip_x1;
-		cmd->body.destRect.right = clip_x2;
-		cmd->body.destRect.top = clip_y1;
-		cmd->body.destRect.bottom = clip_y2;
+		/* if no cliprects hit skip this */
+		if (num == 0)
+			continue;
 
 
+		/* recalculate package length */
+		fifo_size = sizeof(*cmd) + sizeof(SVGASignedRect) * num;
+		cmd->header.size = cpu_to_le32(fifo_size - sizeof(cmd->header));
 		ret = vmw_execbuf_process(file_priv, dev_priv, NULL, cmd,
 					  fifo_size, 0, NULL);
 
@@ -480,7 +553,10 @@ static int do_surface_dirty_sou(struct vmw_private *dev_priv,
 			break;
 	}
 
+
 	kfree(cmd);
+out_free_tmp:
+	kfree(tmp);
 
 	return ret;
 }
@@ -556,6 +632,10 @@ static int vmw_kms_new_framebuffer_surface(struct vmw_private *dev_priv,
 	 * Sanity checks.
 	 */
 
+	/* Surface must be marked as a scanout. */
+	if (unlikely(!surface->scanout))
+		return -EINVAL;
+
 	if (unlikely(surface->mip_levels[0] != 1 ||
 		     surface->num_sizes != 1 ||
 		     surface->sizes[0].width < mode_cmd->width ||
@@ -782,6 +862,7 @@ static int do_dmabuf_dirty_sou(struct drm_file *file_priv,
 			int clip_y1 = clips_ptr->y1 - unit->crtc.y;
 			int clip_x2 = clips_ptr->x2 - unit->crtc.x;
 			int clip_y2 = clips_ptr->y2 - unit->crtc.y;
+			int move_x, move_y;
 
 			/* skip any crtcs that misses the clip region */
 			if (clip_x1 >= unit->crtc.mode.hdisplay ||
@@ -789,12 +870,21 @@ static int do_dmabuf_dirty_sou(struct drm_file *file_priv,
 			    clip_x2 <= 0 || clip_y2 <= 0)
 				continue;
 
+			/* clip size to crtc size */
+			clip_x2 = min_t(int, clip_x2, unit->crtc.mode.hdisplay);
+			clip_y2 = min_t(int, clip_y2, unit->crtc.mode.vdisplay);
+
+			/* translate both src and dest to bring clip into screen */
+			move_x = min_t(int, clip_x1, 0);
+			move_y = min_t(int, clip_y1, 0);
+
+			/* actual translate done here */
 			blits[hit_num].header = SVGA_CMD_BLIT_GMRFB_TO_SCREEN;
 			blits[hit_num].body.destScreenId = unit->unit;
-			blits[hit_num].body.srcOrigin.x = clips_ptr->x1;
-			blits[hit_num].body.srcOrigin.y = clips_ptr->y1;
-			blits[hit_num].body.destRect.left = clip_x1;
-			blits[hit_num].body.destRect.top = clip_y1;
+			blits[hit_num].body.srcOrigin.x = clips_ptr->x1 - move_x;
+			blits[hit_num].body.srcOrigin.y = clips_ptr->y1 - move_y;
+			blits[hit_num].body.destRect.left = clip_x1 - move_x;
+			blits[hit_num].body.destRect.top = clip_y1 - move_y;
 			blits[hit_num].body.destRect.right = clip_x2;
 			blits[hit_num].body.destRect.bottom = clip_y2;
 			hit_num++;
@@ -1033,46 +1123,29 @@ static struct drm_framebuffer *vmw_kms_fb_create(struct drm_device *dev,
 		return ERR_PTR(-ENOENT);
 	}
 
-	/**
-	 * End conditioned code.
-	 */
-
-	ret = vmw_user_surface_lookup_handle(dev_priv, tfile,
-					     mode_cmd->handle, &surface);
+	/* returns either a dmabuf or surface */
+	ret = vmw_user_lookup_handle(dev_priv, tfile,
+				     mode_cmd->handle,
+				     &surface, &bo);
 	if (ret)
-		goto try_dmabuf;
-
-	if (!surface->scanout)
-		goto err_not_scanout;
-
-	ret = vmw_kms_new_framebuffer_surface(dev_priv, file_priv, surface,
-					      &vfb, mode_cmd);
-
-	/* vmw_user_surface_lookup takes one ref so does new_fb */
-	vmw_surface_unreference(&surface);
-
-	if (ret) {
-		DRM_ERROR("failed to create vmw_framebuffer: %i\n", ret);
-		ttm_base_object_unref(&user_obj);
-		return ERR_PTR(ret);
-	} else
-		vfb->user_obj = user_obj;
-	return &vfb->base;
-
-try_dmabuf:
-	DRM_INFO("%s: trying buffer\n", __func__);
-
-	ret = vmw_user_dmabuf_lookup(tfile, mode_cmd->handle, &bo);
-	if (ret) {
-		DRM_ERROR("failed to find buffer: %i\n", ret);
-		return ERR_PTR(-ENOENT);
-	}
-
-	ret = vmw_kms_new_framebuffer_dmabuf(dev_priv, bo, &vfb,
-					     mode_cmd);
+		goto err_out;
+
+	/* Create the new framebuffer depending one what we got back */
+	if (bo)
+		ret = vmw_kms_new_framebuffer_dmabuf(dev_priv, bo, &vfb,
+						     mode_cmd);
+	else if (surface)
+		ret = vmw_kms_new_framebuffer_surface(dev_priv, file_priv,
+						      surface, &vfb, mode_cmd);
+	else
+		BUG();
 
-	/* vmw_user_dmabuf_lookup takes one ref so does new_fb */
-	vmw_dmabuf_unreference(&bo);
+err_out:
+	/* vmw_user_lookup_handle takes one ref so does new_fb */
+	if (bo)
+		vmw_dmabuf_unreference(&bo);
+	if (surface)
+		vmw_surface_unreference(&surface);
 
 	if (ret) {
 		DRM_ERROR("failed to create vmw_framebuffer: %i\n", ret);
@@ -1082,14 +1155,6 @@ try_dmabuf:
 		vfb->user_obj = user_obj;
 
 	return &vfb->base;
-
-err_not_scanout:
-	DRM_ERROR("surface not marked as scanout\n");
-	/* vmw_user_surface_lookup takes one ref */
-	vmw_surface_unreference(&surface);
-	ttm_base_object_unref(&user_obj);
-
-	return ERR_PTR(-EINVAL);
 }
 
 static struct drm_mode_config_funcs vmw_kms_funcs = {
@@ -1106,10 +1171,12 @@ int vmw_kms_present(struct vmw_private *dev_priv,
 		    uint32_t num_clips)
 {
 	struct vmw_display_unit *units[VMWGFX_NUM_DISPLAY_UNITS];
+	struct drm_clip_rect *tmp;
 	struct drm_crtc *crtc;
 	size_t fifo_size;
 	int i, k, num_units;
 	int ret = 0; /* silence warning */
+	int left, right, top, bottom;
 
 	struct {
 		SVGA3dCmdHeader header;
@@ -1127,60 +1194,95 @@ int vmw_kms_present(struct vmw_private *dev_priv,
 	BUG_ON(surface == NULL);
 	BUG_ON(!clips || !num_clips);
 
+	tmp = kzalloc(sizeof(*tmp) * num_clips, GFP_KERNEL);
+	if (unlikely(tmp == NULL)) {
+		DRM_ERROR("Temporary cliprect memory alloc failed.\n");
+		return -ENOMEM;
+	}
+
 	fifo_size = sizeof(*cmd) + sizeof(SVGASignedRect) * num_clips;
 	cmd = kmalloc(fifo_size, GFP_KERNEL);
 	if (unlikely(cmd == NULL)) {
 		DRM_ERROR("Failed to allocate temporary fifo memory.\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto out_free_tmp;
+	}
+
+	left = clips->x;
+	right = clips->x + clips->w;
+	top = clips->y;
+	bottom = clips->y + clips->h;
+
+	for (i = 1; i < num_clips; i++) {
+		left = min_t(int, left, (int)clips[i].x);
+		right = max_t(int, right, (int)clips[i].x + clips[i].w);
+		top = min_t(int, top, (int)clips[i].y);
+		bottom = max_t(int, bottom, (int)clips[i].y + clips[i].h);
 	}
 
 	/* only need to do this once */
 	memset(cmd, 0, fifo_size);
 	cmd->header.id = cpu_to_le32(SVGA_3D_CMD_BLIT_SURFACE_TO_SCREEN);
-	cmd->header.size = cpu_to_le32(fifo_size - sizeof(cmd->header));
-
-	cmd->body.srcRect.left = 0;
-	cmd->body.srcRect.right = surface->sizes[0].width;
-	cmd->body.srcRect.top = 0;
-	cmd->body.srcRect.bottom = surface->sizes[0].height;
 
 	blits = (SVGASignedRect *)&cmd[1];
+
+	cmd->body.srcRect.left = left;
+	cmd->body.srcRect.right = right;
+	cmd->body.srcRect.top = top;
+	cmd->body.srcRect.bottom = bottom;
+
 	for (i = 0; i < num_clips; i++) {
-		blits[i].left   = clips[i].x;
-		blits[i].right  = clips[i].x + clips[i].w;
-		blits[i].top    = clips[i].y;
-		blits[i].bottom = clips[i].y + clips[i].h;
+		tmp[i].x1 = clips[i].x - left;
+		tmp[i].x2 = clips[i].x + clips[i].w - left;
+		tmp[i].y1 = clips[i].y - top;
+		tmp[i].y2 = clips[i].y + clips[i].h - top;
 	}
 
 	for (k = 0; k < num_units; k++) {
 		struct vmw_display_unit *unit = units[k];
-		int clip_x1 = destX - unit->crtc.x;
-		int clip_y1 = destY - unit->crtc.y;
-		int clip_x2 = clip_x1 + surface->sizes[0].width;
-		int clip_y2 = clip_y1 + surface->sizes[0].height;
+		struct vmw_clip_rect clip;
+		int num;
+
+		clip.x1 = left + destX - unit->crtc.x;
+		clip.y1 = top + destY - unit->crtc.y;
+		clip.x2 = right + destX - unit->crtc.x;
+		clip.y2 = bottom + destY - unit->crtc.y;
 
 		/* skip any crtcs that misses the clip region */
-		if (clip_x1 >= unit->crtc.mode.hdisplay ||
-		    clip_y1 >= unit->crtc.mode.vdisplay ||
-		    clip_x2 <= 0 || clip_y2 <= 0)
+		if (clip.x1 >= unit->crtc.mode.hdisplay ||
+		    clip.y1 >= unit->crtc.mode.vdisplay ||
+		    clip.x2 <= 0 || clip.y2 <= 0)
 			continue;
 
+		/*
+		 * In order for the clip rects to be correctly scaled
+		 * the src and dest rects needs to be the same size.
+		 */
+		cmd->body.destRect.left = clip.x1;
+		cmd->body.destRect.right = clip.x2;
+		cmd->body.destRect.top = clip.y1;
+		cmd->body.destRect.bottom = clip.y2;
+
+		/* create a clip rect of the crtc in dest coords */
+		clip.x2 = unit->crtc.mode.hdisplay - clip.x1;
+		clip.y2 = unit->crtc.mode.vdisplay - clip.y1;
+		clip.x1 = 0 - clip.x1;
+		clip.y1 = 0 - clip.y1;
+
 		/* need to reset sid as it is changed by execbuf */
 		cmd->body.srcImage.sid = sid;
-
 		cmd->body.destScreenId = unit->unit;
 
-		/*
-		 * The blit command is a lot more resilient then the
-		 * readback command when it comes to clip rects. So its
-		 * okay to go out of bounds.
-		 */
+		/* clip and write blits to cmd stream */
+		vmw_clip_cliprects(tmp, num_clips, clip, blits, &num);
 
-		cmd->body.destRect.left = clip_x1;
-		cmd->body.destRect.right = clip_x2;
-		cmd->body.destRect.top = clip_y1;
-		cmd->body.destRect.bottom = clip_y2;
+		/* if no cliprects hit skip this */
+		if (num == 0)
+			continue;
 
+		/* recalculate package length */
+		fifo_size = sizeof(*cmd) + sizeof(SVGASignedRect) * num;
+		cmd->header.size = cpu_to_le32(fifo_size - sizeof(cmd->header));
 		ret = vmw_execbuf_process(file_priv, dev_priv, NULL, cmd,
 					  fifo_size, 0, NULL);
 
@@ -1189,6 +1291,8 @@ int vmw_kms_present(struct vmw_private *dev_priv,
 	}
 
 	kfree(cmd);
+out_free_tmp:
+	kfree(tmp);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
index af8e6e5bd964..e1cb8556355f 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
@@ -62,9 +62,14 @@ struct vmw_framebuffer {
 int vmw_cursor_update_image(struct vmw_private *dev_priv,
 			    u32 *image, u32 width, u32 height,
 			    u32 hotspotX, u32 hotspotY);
+int vmw_cursor_update_dmabuf(struct vmw_private *dev_priv,
+			     struct vmw_dma_buffer *dmabuf,
+			     u32 width, u32 height,
+			     u32 hotspotX, u32 hotspotY);
 void vmw_cursor_update_position(struct vmw_private *dev_priv,
 				bool show, int x, int y);
 
+
 /**
  * Base class display unit.
  *
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
index 90c5e3928491..8f8dbd43c33d 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
@@ -74,9 +74,10 @@ static int vmw_ldu_commit_list(struct vmw_private *dev_priv)
 {
 	struct vmw_legacy_display *lds = dev_priv->ldu_priv;
 	struct vmw_legacy_display_unit *entry;
+	struct vmw_display_unit *du = NULL;
 	struct drm_framebuffer *fb = NULL;
 	struct drm_crtc *crtc = NULL;
-	int i = 0;
+	int i = 0, ret;
 
 	/* If there is no display topology the host just assumes
 	 * that the guest will set the same layout as the host.
@@ -129,6 +130,25 @@ static int vmw_ldu_commit_list(struct vmw_private *dev_priv)
 
 	lds->last_num_active = lds->num_active;
 
+
+	/* Find the first du with a cursor. */
+	list_for_each_entry(entry, &lds->active, active) {
+		du = &entry->base;
+
+		if (!du->cursor_dmabuf)
+			continue;
+
+		ret = vmw_cursor_update_dmabuf(dev_priv,
+					       du->cursor_dmabuf,
+					       64, 64,
+					       du->hotspot_x,
+					       du->hotspot_y);
+		if (ret == 0)
+			break;
+
+		DRM_ERROR("Could not update cursor image\n");
+	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
index 86c5e4cceb31..1c7f09e26819 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c
@@ -1190,6 +1190,29 @@ void vmw_resource_unreserve(struct list_head *list)
 		write_unlock(lock);
 }
 
+/**
+ * Helper function that looks either a surface or dmabuf.
+ *
+ * The pointer this pointed at by out_surf and out_buf needs to be null.
+ */
+int vmw_user_lookup_handle(struct vmw_private *dev_priv,
+			   struct ttm_object_file *tfile,
+			   uint32_t handle,
+			   struct vmw_surface **out_surf,
+			   struct vmw_dma_buffer **out_buf)
+{
+	int ret;
+
+	BUG_ON(*out_surf || *out_buf);
+
+	ret = vmw_user_surface_lookup_handle(dev_priv, tfile, handle, out_surf);
+	if (!ret)
+		return 0;
+
+	ret = vmw_user_dmabuf_lookup(tfile, handle, out_buf);
+	return ret;
+}
+
 
 int vmw_user_surface_lookup_handle(struct vmw_private *dev_priv,
 				   struct ttm_object_file *tfile,