diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c
index e01f5eaaec82..38390f7c6ab6 100644
--- a/drivers/char/agp/intel-gtt.c
+++ b/drivers/char/agp/intel-gtt.c
@@ -667,7 +667,7 @@ static int intel_gtt_init(void)
 	gtt_map_size = intel_private.base.gtt_total_entries * 4;
 
 	intel_private.gtt = NULL;
-	if (INTEL_GTT_GEN < 6)
+	if (INTEL_GTT_GEN < 6 && INTEL_GTT_GEN > 2)
 		intel_private.gtt = ioremap_wc(intel_private.gtt_bus_addr,
 					       gtt_map_size);
 	if (intel_private.gtt == NULL)
diff --git a/drivers/gpu/drm/i915/dvo_ch7xxx.c b/drivers/gpu/drm/i915/dvo_ch7xxx.c
index 38f3a6cb8c7d..3edd981e0770 100644
--- a/drivers/gpu/drm/i915/dvo_ch7xxx.c
+++ b/drivers/gpu/drm/i915/dvo_ch7xxx.c
@@ -303,10 +303,10 @@ static bool ch7xxx_get_hw_state(struct intel_dvo_device *dvo)
 
 	ch7xxx_readb(dvo, CH7xxx_PM, &val);
 
-	if (val & CH7xxx_PM_FPD)
-		return false;
-	else
+	if (val & (CH7xxx_PM_DVIL | CH7xxx_PM_DVIP))
 		return true;
+	else
+		return false;
 }
 
 static void ch7xxx_dump_regs(struct intel_dvo_device *dvo)
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4f2831aa5fed..b84f7861e438 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1341,9 +1341,14 @@ int __must_check i915_gem_object_get_pages(struct drm_i915_gem_object *obj);
 static inline struct page *i915_gem_object_get_page(struct drm_i915_gem_object *obj, int n)
 {
 	struct scatterlist *sg = obj->pages->sgl;
-	while (n >= SG_MAX_SINGLE_ALLOC) {
+	int nents = obj->pages->nents;
+	while (nents > SG_MAX_SINGLE_ALLOC) {
+		if (n < SG_MAX_SINGLE_ALLOC - 1)
+			break;
+
 		sg = sg_chain_ptr(sg + SG_MAX_SINGLE_ALLOC - 1);
 		n -= SG_MAX_SINGLE_ALLOC - 1;
+		nents -= SG_MAX_SINGLE_ALLOC - 1;
 	}
 	return sg_page(sg+n);
 }
@@ -1427,7 +1432,7 @@ int __must_check i915_gpu_idle(struct drm_device *dev);
 int __must_check i915_gem_idle(struct drm_device *dev);
 int i915_add_request(struct intel_ring_buffer *ring,
 		     struct drm_file *file,
-		     struct drm_i915_gem_request *request);
+		     u32 *seqno);
 int __must_check i915_wait_seqno(struct intel_ring_buffer *ring,
 				 uint32_t seqno);
 int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 19dbdd7dd564..d33d02d13c96 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1955,11 +1955,12 @@ i915_gem_next_request_seqno(struct intel_ring_buffer *ring)
 int
 i915_add_request(struct intel_ring_buffer *ring,
 		 struct drm_file *file,
-		 struct drm_i915_gem_request *request)
+		 u32 *out_seqno)
 {
 	drm_i915_private_t *dev_priv = ring->dev->dev_private;
-	uint32_t seqno;
+	struct drm_i915_gem_request *request;
 	u32 request_ring_position;
+	u32 seqno;
 	int was_empty;
 	int ret;
 
@@ -1974,11 +1975,9 @@ i915_add_request(struct intel_ring_buffer *ring,
 	if (ret)
 		return ret;
 
-	if (request == NULL) {
-		request = kmalloc(sizeof(*request), GFP_KERNEL);
-		if (request == NULL)
-			return -ENOMEM;
-	}
+	request = kmalloc(sizeof(*request), GFP_KERNEL);
+	if (request == NULL)
+		return -ENOMEM;
 
 	seqno = i915_gem_next_request_seqno(ring);
 
@@ -2030,6 +2029,8 @@ i915_add_request(struct intel_ring_buffer *ring,
 		}
 	}
 
+	if (out_seqno)
+		*out_seqno = seqno;
 	return 0;
 }
 
@@ -3959,6 +3960,9 @@ i915_gem_init_hw(struct drm_device *dev)
 	if (!intel_enable_gtt())
 		return -EIO;
 
+	if (IS_HASWELL(dev) && (I915_READ(0x120010) == 1))
+		I915_WRITE(0x9008, I915_READ(0x9008) | 0xf0000);
+
 	i915_gem_l3_remap(dev);
 
 	i915_gem_init_swizzling(dev);
@@ -4098,7 +4102,6 @@ i915_gem_entervt_ioctl(struct drm_device *dev, void *data,
 	}
 
 	BUG_ON(!list_empty(&dev_priv->mm.active_list));
-	BUG_ON(!list_empty(&dev_priv->mm.inactive_list));
 	mutex_unlock(&dev->struct_mutex);
 
 	ret = drm_irq_install(dev);
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 64c1be0a9cfd..a4162ddff6c5 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -521,7 +521,7 @@
  */
 # define _3D_CHICKEN2_WM_READ_PIPELINED			(1 << 14)
 #define _3D_CHICKEN3	0x02090
-#define  _3D_CHICKEN_SF_DISABLE_FASTCLIP_CULL		(1 << 5)
+#define  _3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL		(1 << 5)
 
 #define MI_MODE		0x0209c
 # define VS_TIMER_DISPATCH				(1 << 6)
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 2b6ce9b2674a..682bd3729baf 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -3253,6 +3253,16 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 
 	if (HAS_PCH_CPT(dev))
 		intel_cpt_verify_modeset(dev, intel_crtc->pipe);
+
+	/*
+	 * There seems to be a race in PCH platform hw (at least on some
+	 * outputs) where an enabled pipe still completes any pageflip right
+	 * away (as if the pipe is off) instead of waiting for vblank. As soon
+	 * as the first vblank happend, everything works as expected. Hence just
+	 * wait for one vblank before returning to avoid strange things
+	 * happening.
+	 */
+	intel_wait_for_vblank(dev, intel_crtc->pipe);
 }
 
 static void ironlake_crtc_disable(struct drm_crtc *crtc)
@@ -7892,8 +7902,7 @@ static struct intel_quirk intel_quirks[] = {
 	/* ThinkPad T60 needs pipe A force quirk (bug #16494) */
 	{ 0x2782, 0x17aa, 0x201a, quirk_pipea_force },
 
-	/* 855 & before need to leave pipe A & dpll A up */
-	{ 0x3582, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force },
+	/* 830/845 need to leave pipe A & dpll A up */
 	{ 0x2562, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force },
 	{ 0x3577, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force },
 
@@ -8049,29 +8058,42 @@ static void intel_enable_pipe_a(struct drm_device *dev)
 
 }
 
+static bool
+intel_check_plane_mapping(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	u32 reg, val;
+
+	if (dev_priv->num_pipe == 1)
+		return true;
+
+	reg = DSPCNTR(!crtc->plane);
+	val = I915_READ(reg);
+
+	if ((val & DISPLAY_PLANE_ENABLE) &&
+	    (!!(val & DISPPLANE_SEL_PIPE_MASK) == crtc->pipe))
+		return false;
+
+	return true;
+}
+
 static void intel_sanitize_crtc(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 reg, val;
+	u32 reg;
 
 	/* Clear any frame start delays used for debugging left by the BIOS */
 	reg = PIPECONF(crtc->pipe);
 	I915_WRITE(reg, I915_READ(reg) & ~PIPECONF_FRAME_START_DELAY_MASK);
 
 	/* We need to sanitize the plane -> pipe mapping first because this will
-	 * disable the crtc (and hence change the state) if it is wrong. */
-	if (!HAS_PCH_SPLIT(dev)) {
+	 * disable the crtc (and hence change the state) if it is wrong. Note
+	 * that gen4+ has a fixed plane -> pipe mapping.  */
+	if (INTEL_INFO(dev)->gen < 4 && !intel_check_plane_mapping(crtc)) {
 		struct intel_connector *connector;
 		bool plane;
 
-		reg = DSPCNTR(crtc->plane);
-		val = I915_READ(reg);
-
-		if ((val & DISPLAY_PLANE_ENABLE) == 0 &&
-		    (!!(val & DISPPLANE_SEL_PIPE_MASK) == crtc->pipe))
-			goto ok;
-
 		DRM_DEBUG_KMS("[CRTC:%d] wrong plane connection detected!\n",
 			      crtc->base.base.id);
 
@@ -8095,7 +8117,6 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 		WARN_ON(crtc->active);
 		crtc->base.enabled = false;
 	}
-ok:
 
 	if (dev_priv->quirks & QUIRK_PIPEA_FORCE &&
 	    crtc->pipe == PIPE_A && !crtc->active) {
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index d1e8ddb2d6c0..1b727a5c9ee5 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -2369,8 +2369,9 @@ static void
 intel_dp_destroy(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
+	struct intel_dp *intel_dp = intel_attached_dp(connector);
 
-	if (intel_dpd_is_edp(dev))
+	if (is_edp(intel_dp))
 		intel_panel_destroy_backlight(dev);
 
 	drm_sysfs_connector_remove(connector);
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index ebff850a9ab6..495625914e4a 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -209,7 +209,6 @@ static void intel_overlay_unmap_regs(struct intel_overlay *overlay,
 }
 
 static int intel_overlay_do_wait_request(struct intel_overlay *overlay,
-					 struct drm_i915_gem_request *request,
 					 void (*tail)(struct intel_overlay *))
 {
 	struct drm_device *dev = overlay->dev;
@@ -218,12 +217,10 @@ static int intel_overlay_do_wait_request(struct intel_overlay *overlay,
 	int ret;
 
 	BUG_ON(overlay->last_flip_req);
-	ret = i915_add_request(ring, NULL, request);
-	if (ret) {
-	    kfree(request);
-	    return ret;
-	}
-	overlay->last_flip_req = request->seqno;
+	ret = i915_add_request(ring, NULL, &overlay->last_flip_req);
+	if (ret)
+		return ret;
+
 	overlay->flip_tail = tail;
 	ret = i915_wait_seqno(ring, overlay->last_flip_req);
 	if (ret)
@@ -240,7 +237,6 @@ static int intel_overlay_on(struct intel_overlay *overlay)
 	struct drm_device *dev = overlay->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
-	struct drm_i915_gem_request *request;
 	int ret;
 
 	BUG_ON(overlay->active);
@@ -248,17 +244,9 @@ static int intel_overlay_on(struct intel_overlay *overlay)
 
 	WARN_ON(IS_I830(dev) && !(dev_priv->quirks & QUIRK_PIPEA_FORCE));
 
-	request = kzalloc(sizeof(*request), GFP_KERNEL);
-	if (request == NULL) {
-		ret = -ENOMEM;
-		goto out;
-	}
-
 	ret = intel_ring_begin(ring, 4);
-	if (ret) {
-		kfree(request);
-		goto out;
-	}
+	if (ret)
+		return ret;
 
 	intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_ON);
 	intel_ring_emit(ring, overlay->flip_addr | OFC_UPDATE);
@@ -266,9 +254,7 @@ static int intel_overlay_on(struct intel_overlay *overlay)
 	intel_ring_emit(ring, MI_NOOP);
 	intel_ring_advance(ring);
 
-	ret = intel_overlay_do_wait_request(overlay, request, NULL);
-out:
-	return ret;
+	return intel_overlay_do_wait_request(overlay, NULL);
 }
 
 /* overlay needs to be enabled in OCMD reg */
@@ -278,17 +264,12 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 	struct drm_device *dev = overlay->dev;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
-	struct drm_i915_gem_request *request;
 	u32 flip_addr = overlay->flip_addr;
 	u32 tmp;
 	int ret;
 
 	BUG_ON(!overlay->active);
 
-	request = kzalloc(sizeof(*request), GFP_KERNEL);
-	if (request == NULL)
-		return -ENOMEM;
-
 	if (load_polyphase_filter)
 		flip_addr |= OFC_UPDATE;
 
@@ -298,22 +279,14 @@ static int intel_overlay_continue(struct intel_overlay *overlay,
 		DRM_DEBUG("overlay underrun, DOVSTA: %x\n", tmp);
 
 	ret = intel_ring_begin(ring, 2);
-	if (ret) {
-		kfree(request);
+	if (ret)
 		return ret;
-	}
+
 	intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_CONTINUE);
 	intel_ring_emit(ring, flip_addr);
 	intel_ring_advance(ring);
 
-	ret = i915_add_request(ring, NULL, request);
-	if (ret) {
-		kfree(request);
-		return ret;
-	}
-
-	overlay->last_flip_req = request->seqno;
-	return 0;
+	return i915_add_request(ring, NULL, &overlay->last_flip_req);
 }
 
 static void intel_overlay_release_old_vid_tail(struct intel_overlay *overlay)
@@ -349,15 +322,10 @@ static int intel_overlay_off(struct intel_overlay *overlay)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
 	u32 flip_addr = overlay->flip_addr;
-	struct drm_i915_gem_request *request;
 	int ret;
 
 	BUG_ON(!overlay->active);
 
-	request = kzalloc(sizeof(*request), GFP_KERNEL);
-	if (request == NULL)
-		return -ENOMEM;
-
 	/* According to intel docs the overlay hw may hang (when switching
 	 * off) without loading the filter coeffs. It is however unclear whether
 	 * this applies to the disabling of the overlay or to the switching off
@@ -365,10 +333,9 @@ static int intel_overlay_off(struct intel_overlay *overlay)
 	flip_addr |= OFC_UPDATE;
 
 	ret = intel_ring_begin(ring, 6);
-	if (ret) {
-		kfree(request);
+	if (ret)
 		return ret;
-	}
+
 	/* wait for overlay to go idle */
 	intel_ring_emit(ring, MI_OVERLAY_FLIP | MI_OVERLAY_CONTINUE);
 	intel_ring_emit(ring, flip_addr);
@@ -379,8 +346,7 @@ static int intel_overlay_off(struct intel_overlay *overlay)
 	intel_ring_emit(ring, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
 	intel_ring_advance(ring);
 
-	return intel_overlay_do_wait_request(overlay, request,
-					     intel_overlay_off_tail);
+	return intel_overlay_do_wait_request(overlay, intel_overlay_off_tail);
 }
 
 /* recover from an interruption due to a signal
@@ -425,24 +391,16 @@ static int intel_overlay_release_old_vid(struct intel_overlay *overlay)
 		return 0;
 
 	if (I915_READ(ISR) & I915_OVERLAY_PLANE_FLIP_PENDING_INTERRUPT) {
-		struct drm_i915_gem_request *request;
-
 		/* synchronous slowpath */
-		request = kzalloc(sizeof(*request), GFP_KERNEL);
-		if (request == NULL)
-			return -ENOMEM;
-
 		ret = intel_ring_begin(ring, 2);
-		if (ret) {
-			kfree(request);
+		if (ret)
 			return ret;
-		}
 
 		intel_ring_emit(ring, MI_WAIT_FOR_EVENT | MI_WAIT_FOR_OVERLAY_FLIP);
 		intel_ring_emit(ring, MI_NOOP);
 		intel_ring_advance(ring);
 
-		ret = intel_overlay_do_wait_request(overlay, request,
+		ret = intel_overlay_do_wait_request(overlay,
 						    intel_overlay_release_old_vid_tail);
 		if (ret)
 			return ret;
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index b3b4b6cea8b0..72f41aaa71ff 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -3442,8 +3442,8 @@ static void gen6_init_clock_gating(struct drm_device *dev)
 		   GEN6_RCCUNIT_CLOCK_GATE_DISABLE);
 
 	/* Bspec says we need to always set all mask bits. */
-	I915_WRITE(_3D_CHICKEN, (0xFFFF << 16) |
-		   _3D_CHICKEN_SF_DISABLE_FASTCLIP_CULL);
+	I915_WRITE(_3D_CHICKEN3, (0xFFFF << 16) |
+		   _3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL);
 
 	/*
 	 * According to the spec the following bits should be