Merge branch 'drm-intel-next' of ssh://master.kernel.org/pub/scm/linux/kernel/git/keithp/linux-2.6 into drm-core-next

* 'drm-intel-next' of ssh://master.kernel.org/pub/scm/linux/kernel/git/keithp/linux-2.6: (52 commits)
  drm/i915: provide module parameter description
  drm/i915: add module parameter compiler hints
  drm/i915/bios: Avoid temporary allocation whilst searching for downclock
  drm/i915: Cache GT fifo count for SandyBridge
  i915: Fix opregion notifications
  drm/i915: TVDAC_STATE_CHG does not indicate successful load-detect
  drm/i915: Select correct pipe during TV detect
  drm/i915/ringbuffer: Idling requires waiting for the ring to be empty
  Revert "drm/i915: enable rc6 by default"
  drm/i915: Clean up i915_driver_load failure path
  drm/i915: Enable i915 frame buffer compression by default
  drm/i915: Share the common work of disabling active FBC before updating
  drm/i915: Perform intel_enable_fbc() from a delayed task
  drm/i915: Disable FBC across page-flipping
  drm/i915: Set persistent-mode for ILK/SNB framebuffer compression
  drm/i915: Use of a CPU fence is mandatory to update FBC regions upon CPU writes
  drm/i915: Remove vestigial pitch from post-gen2 FBC control routines
  drm/i915: Replace direct calls to vfunc.disable_fbc with intel_disable_fbc()
  drm/i915: Only export the generic intel_disable_fbc() interface
  drm/i915: Enable GPU reset on Ivybridge.
  ...

20 files changed, 1101 insertions, 404 deletions

diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index 92369655dca..f88a9b2c977 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -560,6 +560,11 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 			mode_changed = true;
 		} else if (set->fb == NULL) {
 			mode_changed = true;
+		} else if (set->fb->depth != set->crtc->fb->depth) {
+			mode_changed = true;
+		} else if (set->fb->bits_per_pixel !=
+			   set->crtc->fb->bits_per_pixel) {
+			mode_changed = true;
 		} else
 			fb_changed = true;
 	}
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 0a893f7400f..e2662497d50 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -865,7 +865,7 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused)
 			   MEMSTAT_VID_SHIFT);
 		seq_printf(m, "Current P-state: %d\n",
 			   (rgvstat & MEMSTAT_PSTATE_MASK) >> MEMSTAT_PSTATE_SHIFT);
-	} else if (IS_GEN6(dev)) {
+	} else if (IS_GEN6(dev) || IS_GEN7(dev)) {
 		u32 gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS);
 		u32 rp_state_limits = I915_READ(GEN6_RP_STATE_LIMITS);
 		u32 rp_state_cap = I915_READ(GEN6_RP_STATE_CAP);
@@ -1123,6 +1123,44 @@ static int i915_emon_status(struct seq_file *m, void *unused)
 	return 0;
 }
 
+static int i915_ring_freq_table(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;
+	int gpu_freq, ia_freq;
+
+	if (!(IS_GEN6(dev) || IS_GEN7(dev))) {
+		seq_printf(m, "unsupported on this chipset\n");
+		return 0;
+	}
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	seq_printf(m, "GPU freq (MHz)\tEffective CPU freq (MHz)\n");
+
+	for (gpu_freq = dev_priv->min_delay; gpu_freq <= dev_priv->max_delay;
+	     gpu_freq++) {
+		I915_WRITE(GEN6_PCODE_DATA, gpu_freq);
+		I915_WRITE(GEN6_PCODE_MAILBOX, GEN6_PCODE_READY |
+			   GEN6_PCODE_READ_MIN_FREQ_TABLE);
+		if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) &
+			      GEN6_PCODE_READY) == 0, 10)) {
+			DRM_ERROR("pcode read of freq table timed out\n");
+			continue;
+		}
+		ia_freq = I915_READ(GEN6_PCODE_DATA);
+		seq_printf(m, "%d\t\t%d\n", gpu_freq * 50, ia_freq * 100);
+	}
+
+	mutex_unlock(&dev->struct_mutex);
+
+	return 0;
+}
+
 static int i915_gfxec(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
@@ -1430,6 +1468,7 @@ static struct drm_info_list i915_debugfs_list[] = {
 	{"i915_inttoext_table", i915_inttoext_table, 0},
 	{"i915_drpc_info", i915_drpc_info, 0},
 	{"i915_emon_status", i915_emon_status, 0},
+	{"i915_ring_freq_table", i915_ring_freq_table, 0},
 	{"i915_gfxec", i915_gfxec, 0},
 	{"i915_fbc_status", i915_fbc_status, 0},
 	{"i915_sr_status", i915_sr_status, 0},
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index e1787022d6c..12712824a6d 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1073,6 +1073,9 @@ static void i915_setup_compression(struct drm_device *dev, int size)
 	unsigned long cfb_base;
 	unsigned long ll_base = 0;
 
+	/* Just in case the BIOS is doing something questionable. */
+	intel_disable_fbc(dev);
+
 	compressed_fb = drm_mm_search_free(&dev_priv->mm.stolen, size, 4096, 0);
 	if (compressed_fb)
 		compressed_fb = drm_mm_get_block(compressed_fb, size, 4096);
@@ -1099,7 +1102,6 @@ static void i915_setup_compression(struct drm_device *dev, int size)
 
 	dev_priv->cfb_size = size;
 
-	intel_disable_fbc(dev);
 	dev_priv->compressed_fb = compressed_fb;
 	if (HAS_PCH_SPLIT(dev))
 		I915_WRITE(ILK_DPFC_CB_BASE, compressed_fb->start);
@@ -1943,7 +1945,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	if (!dev_priv->mm.gtt) {
 		DRM_ERROR("Failed to initialize GTT\n");
 		ret = -ENODEV;
-		goto out_iomapfree;
+		goto out_rmmap;
 	}
 
 	agp_size = dev_priv->mm.gtt->gtt_mappable_entries << PAGE_SHIFT;
@@ -1987,7 +1989,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	if (dev_priv->wq == NULL) {
 		DRM_ERROR("Failed to create our workqueue.\n");
 		ret = -ENOMEM;
-		goto out_iomapfree;
+		goto out_mtrrfree;
 	}
 
 	/* enable GEM by default */
@@ -2074,13 +2076,21 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	return 0;
 
 out_gem_unload:
+	if (dev_priv->mm.inactive_shrinker.shrink)
+		unregister_shrinker(&dev_priv->mm.inactive_shrinker);
+
 	if (dev->pdev->msi_enabled)
 		pci_disable_msi(dev->pdev);
 
 	intel_teardown_gmbus(dev);
 	intel_teardown_mchbar(dev);
 	destroy_workqueue(dev_priv->wq);
-out_iomapfree:
+out_mtrrfree:
+	if (dev_priv->mm.gtt_mtrr >= 0) {
+		mtrr_del(dev_priv->mm.gtt_mtrr, dev->agp->base,
+			 dev->agp->agp_info.aper_size * 1024 * 1024);
+		dev_priv->mm.gtt_mtrr = -1;
+	}
 	io_mapping_free(dev_priv->mm.gtt_mapping);
 out_rmmap:
 	pci_iounmap(dev->pdev, dev_priv->regs);
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 013d304455b..ce045a8cf82 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -37,38 +37,70 @@
 #include <linux/console.h>
 #include "drm_crtc_helper.h"
 
-static int i915_modeset = -1;
+static int i915_modeset __read_mostly = -1;
 module_param_named(modeset, i915_modeset, int, 0400);
+MODULE_PARM_DESC(modeset,
+		"Use kernel modesetting [KMS] (0=DRM_I915_KMS from .config, "
+		"1=on, -1=force vga console preference [default])");
 
-unsigned int i915_fbpercrtc = 0;
+unsigned int i915_fbpercrtc __always_unused = 0;
 module_param_named(fbpercrtc, i915_fbpercrtc, int, 0400);
 
-int i915_panel_ignore_lid = 0;
+int i915_panel_ignore_lid __read_mostly = 0;
 module_param_named(panel_ignore_lid, i915_panel_ignore_lid, int, 0600);
+MODULE_PARM_DESC(panel_ignore_lid,
+		"Override lid status (0=autodetect [default], 1=lid open, "
+		"-1=lid closed)");
 
-unsigned int i915_powersave = 1;
+unsigned int i915_powersave __read_mostly = 1;
 module_param_named(powersave, i915_powersave, int, 0600);
+MODULE_PARM_DESC(powersave,
+		"Enable powersavings, fbc, downclocking, etc. (default: true)");
 
-unsigned int i915_semaphores = 0;
+unsigned int i915_semaphores __read_mostly = 0;
 module_param_named(semaphores, i915_semaphores, int, 0600);
+MODULE_PARM_DESC(semaphores,
+		"Use semaphores for inter-ring sync (default: false)");
 
-unsigned int i915_enable_rc6 = 1;
+unsigned int i915_enable_rc6 __read_mostly = 0;
 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)");
 
-unsigned int i915_enable_fbc = 0;
+unsigned 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 "
+		"(default: false)");
 
-unsigned int i915_lvds_downclock = 0;
+unsigned int i915_lvds_downclock __read_mostly = 0;
 module_param_named(lvds_downclock, i915_lvds_downclock, int, 0400);
+MODULE_PARM_DESC(lvds_downclock,
+		"Use panel (LVDS/eDP) downclocking for power savings "
+		"(default: false)");
 
-unsigned int i915_panel_use_ssc = 1;
+unsigned 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] "
+		"(default: true)");
 
-int i915_vbt_sdvo_panel_type = -1;
+int i915_vbt_sdvo_panel_type __read_mostly = -1;
 module_param_named(vbt_sdvo_panel_type, i915_vbt_sdvo_panel_type, int, 0600);
+MODULE_PARM_DESC(vbt_sdvo_panel_type,
+		"Override selection of SDVO panel mode in the VBT "
+		"(default: auto)");
 
-static bool i915_try_reset = true;
+static bool i915_try_reset __read_mostly = true;
 module_param_named(reset, i915_try_reset, bool, 0600);
+MODULE_PARM_DESC(reset, "Attempt GPU resets (default: true)");
+
+bool i915_enable_hangcheck __read_mostly = true;
+module_param_named(enable_hangcheck, i915_enable_hangcheck, bool, 0644);
+MODULE_PARM_DESC(enable_hangcheck,
+		"Periodically check GPU activity for detecting hangs. "
+		"WARNING: Disabling this can cause system wide hangs. "
+		"(default: true)");
 
 static struct drm_driver driver;
 extern int intel_agp_enabled;
@@ -345,12 +377,17 @@ void gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
 
 void __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
 {
-	int loop = 500;
-	u32 fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
-	while (fifo < 20 && loop--) {
-		udelay(10);
-		fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
+	if (dev_priv->gt_fifo_count < GT_FIFO_NUM_RESERVED_ENTRIES ) {
+		int loop = 500;
+		u32 fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
+		while (fifo <= GT_FIFO_NUM_RESERVED_ENTRIES && loop--) {
+			udelay(10);
+			fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
+		}
+		WARN_ON(loop < 0 && fifo <= GT_FIFO_NUM_RESERVED_ENTRIES);
+		dev_priv->gt_fifo_count = fifo;
 	}
+	dev_priv->gt_fifo_count--;
 }
 
 static int i915_drm_freeze(struct drm_device *dev)
@@ -577,6 +614,7 @@ int i915_reset(struct drm_device *dev, u8 flags)
 	if (get_seconds() - dev_priv->last_gpu_reset < 5) {
 		DRM_ERROR("GPU hanging too fast, declaring wedged!\n");
 	} else switch (INTEL_INFO(dev)->gen) {
+	case 7:
 	case 6:
 		ret = gen6_do_reset(dev, flags);
 		/* If reset with a user forcewake, try to restore */
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index f245c588ae9..78cdd158287 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -214,6 +214,8 @@ struct drm_i915_display_funcs {
 	int (*queue_flip)(struct drm_device *dev, struct drm_crtc *crtc,
 			  struct drm_framebuffer *fb,
 			  struct drm_i915_gem_object *obj);
+	int (*update_plane)(struct drm_crtc *crtc, struct drm_framebuffer *fb,
+			    int x, int y);
 	/* clock updates for mode set */
 	/* cursor updates */
 	/* render clock increase/decrease */
@@ -264,6 +266,7 @@ enum intel_pch {
 #define QUIRK_PIPEA_FORCE (1<<0)
 
 struct intel_fbdev;
+struct intel_fbc_work;
 
 typedef struct drm_i915_private {
 	struct drm_device *dev;
@@ -274,6 +277,7 @@ typedef struct drm_i915_private {
 	int relative_constants_mode;
 
 	void __iomem *regs;
+	u32 gt_fifo_count;
 
 	struct intel_gmbus {
 		struct i2c_adapter adapter;
@@ -328,11 +332,10 @@ typedef struct drm_i915_private {
 	uint32_t last_instdone1;
 
 	unsigned long cfb_size;
-	unsigned long cfb_pitch;
-	unsigned long cfb_offset;
-	int cfb_fence;
-	int cfb_plane;
+	unsigned int cfb_fb;
+	enum plane cfb_plane;
 	int cfb_y;
+	struct intel_fbc_work *fbc_work;
 
 	struct intel_opregion opregion;
 
@@ -985,15 +988,16 @@ struct drm_i915_file_private {
 
 extern struct drm_ioctl_desc i915_ioctls[];
 extern int i915_max_ioctl;
-extern unsigned int i915_fbpercrtc;
-extern int i915_panel_ignore_lid;
-extern unsigned int i915_powersave;
-extern unsigned int i915_semaphores;
-extern unsigned int i915_lvds_downclock;
-extern unsigned int i915_panel_use_ssc;
-extern int i915_vbt_sdvo_panel_type;
-extern unsigned int i915_enable_rc6;
-extern unsigned int i915_enable_fbc;
+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 unsigned int i915_lvds_downclock __read_mostly;
+extern unsigned 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 bool i915_enable_hangcheck __read_mostly;
 
 extern int i915_suspend(struct drm_device *dev, pm_message_t state);
 extern int i915_resume(struct drm_device *dev);
@@ -1163,7 +1167,7 @@ void i915_gem_clflush_object(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_object_set_domain(struct drm_i915_gem_object *obj,
 					    uint32_t read_domains,
 					    uint32_t write_domain);
-int __must_check i915_gem_object_flush_gpu(struct drm_i915_gem_object *obj);
+int __must_check i915_gem_object_finish_gpu(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_init_ringbuffer(struct drm_device *dev);
 void i915_gem_cleanup_ringbuffer(struct drm_device *dev);
 void i915_gem_do_init(struct drm_device *dev,
@@ -1182,7 +1186,8 @@ int __must_check
 i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj,
 				  bool write);
 int __must_check
-i915_gem_object_set_to_display_plane(struct drm_i915_gem_object *obj,
+i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
+				     u32 alignment,
 				     struct intel_ring_buffer *pipelined);
 int i915_gem_attach_phys_object(struct drm_device *dev,
 				struct drm_i915_gem_object *obj,
@@ -1196,9 +1201,14 @@ void i915_gem_release(struct drm_device *dev, struct drm_file *file);
 uint32_t
 i915_gem_get_unfenced_gtt_alignment(struct drm_i915_gem_object *obj);
 
+int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
+				    enum i915_cache_level cache_level);
+
 /* i915_gem_gtt.c */
 void i915_gem_restore_gtt_mappings(struct drm_device *dev);
 int __must_check i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj);
+void i915_gem_gtt_rebind_object(struct drm_i915_gem_object *obj,
+				enum i915_cache_level cache_level);
 void i915_gem_gtt_unbind_object(struct drm_i915_gem_object *obj);
 
 /* i915_gem_evict.c */
@@ -1280,12 +1290,8 @@ extern void intel_modeset_init(struct drm_device *dev);
 extern void intel_modeset_gem_init(struct drm_device *dev);
 extern void intel_modeset_cleanup(struct drm_device *dev);
 extern int intel_modeset_vga_set_state(struct drm_device *dev, bool state);
-extern void i8xx_disable_fbc(struct drm_device *dev);
-extern void g4x_disable_fbc(struct drm_device *dev);
-extern void ironlake_disable_fbc(struct drm_device *dev);
-extern void intel_disable_fbc(struct drm_device *dev);
-extern void intel_enable_fbc(struct drm_crtc *crtc, unsigned long interval);
 extern bool intel_fbc_enabled(struct drm_device *dev);
+extern void intel_disable_fbc(struct drm_device *dev);
 extern bool ironlake_set_drps(struct drm_device *dev, u8 val);
 extern void ironlake_enable_rc6(struct drm_device *dev);
 extern void gen6_set_rps(struct drm_device *dev, u8 val);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 5c0d1247f45..e9d1d5c3a69 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1771,8 +1771,11 @@ i915_add_request(struct intel_ring_buffer *ring,
 	ring->outstanding_lazy_request = false;
 
 	if (!dev_priv->mm.suspended) {
-		mod_timer(&dev_priv->hangcheck_timer,
-			  jiffies + msecs_to_jiffies(DRM_I915_HANGCHECK_PERIOD));
+		if (i915_enable_hangcheck) {
+			mod_timer(&dev_priv->hangcheck_timer,
+				  jiffies +
+				  msecs_to_jiffies(DRM_I915_HANGCHECK_PERIOD));
+		}
 		if (was_empty)
 			queue_delayed_work(dev_priv->wq,
 					   &dev_priv->mm.retire_work, HZ);
@@ -2143,6 +2146,30 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj)
 	return 0;
 }
 
+static void i915_gem_object_finish_gtt(struct drm_i915_gem_object *obj)
+{
+	u32 old_write_domain, old_read_domains;
+
+	/* Act a barrier for all accesses through the GTT */
+	mb();
+
+	/* Force a pagefault for domain tracking on next user access */
+	i915_gem_release_mmap(obj);
+
+	if ((obj->base.read_domains & I915_GEM_DOMAIN_GTT) == 0)
+		return;
+
+	old_read_domains = obj->base.read_domains;
+	old_write_domain = obj->base.write_domain;
+
+	obj->base.read_domains &= ~I915_GEM_DOMAIN_GTT;
+	obj->base.write_domain &= ~I915_GEM_DOMAIN_GTT;
+
+	trace_i915_gem_object_change_domain(obj,
+					    old_read_domains,
+					    old_write_domain);
+}
+
 /**
  * Unbinds an object from the GTT aperture.
  */
@@ -2159,23 +2186,28 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj)
 		return -EINVAL;
 	}
 
-	/* blow away mappings if mapped through GTT */
-	i915_gem_release_mmap(obj);
-
-	/* Move the object to the CPU domain to ensure that
-	 * any possible CPU writes while it's not in the GTT
-	 * are flushed when we go to remap it. This will
-	 * also ensure that all pending GPU writes are finished
-	 * before we unbind.
-	 */
-	ret = i915_gem_object_set_to_cpu_domain(obj, 1);
+	ret = i915_gem_object_finish_gpu(obj);
 	if (ret == -ERESTARTSYS)
 		return ret;
 	/* Continue on if we fail due to EIO, the GPU is hung so we
 	 * should be safe and we need to cleanup or else we might
 	 * cause memory corruption through use-after-free.
 	 */
+
+	i915_gem_object_finish_gtt(obj);
+
+	/* Move the object to the CPU domain to ensure that
+	 * any possible CPU writes while it's not in the GTT
+	 * are flushed when we go to remap it.
+	 */
+	if (ret == 0)
+		ret = i915_gem_object_set_to_cpu_domain(obj, 1);
+	if (ret == -ERESTARTSYS)
+		return ret;
 	if (ret) {
+		/* In the event of a disaster, abandon all caches and
+		 * hope for the best.
+		 */
 		i915_gem_clflush_object(obj);
 		obj->base.read_domains = obj->base.write_domain = I915_GEM_DOMAIN_CPU;
 	}
@@ -2997,51 +3029,139 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 	return 0;
 }
 
+int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
+				    enum i915_cache_level cache_level)
+{
+	int ret;
+
+	if (obj->cache_level == cache_level)
+		return 0;
+
+	if (obj->pin_count) {
+		DRM_DEBUG("can not change the cache level of pinned objects\n");
+		return -EBUSY;
+	}
+
+	if (obj->gtt_space) {
+		ret = i915_gem_object_finish_gpu(obj);
+		if (ret)
+			return ret;
+
+		i915_gem_object_finish_gtt(obj);
+
+		/* Before SandyBridge, you could not use tiling or fence
+		 * registers with snooped memory, so relinquish any fences
+		 * currently pointing to our region in the aperture.
+		 */
+		if (INTEL_INFO(obj->base.dev)->gen < 6) {
+			ret = i915_gem_object_put_fence(obj);
+			if (ret)
+				return ret;
+		}
+
+		i915_gem_gtt_rebind_object(obj, cache_level);
+	}
+
+	if (cache_level == I915_CACHE_NONE) {
+		u32 old_read_domains, old_write_domain;
+
+		/* If we're coming from LLC cached, then we haven't
+		 * actually been tracking whether the data is in the
+		 * CPU cache or not, since we only allow one bit set
+		 * in obj->write_domain and have been skipping the clflushes.
+		 * Just set it to the CPU cache for now.
+		 */
+		WARN_ON(obj->base.write_domain & ~I915_GEM_DOMAIN_CPU);
+		WARN_ON(obj->base.read_domains & ~I915_GEM_DOMAIN_CPU);
+
+		old_read_domains = obj->base.read_domains;
+		old_write_domain = obj->base.write_domain;
+
+		obj->base.read_domains = I915_GEM_DOMAIN_CPU;
+		obj->base.write_domain = I915_GEM_DOMAIN_CPU;
+
+		trace_i915_gem_object_change_domain(obj,
+						    old_read_domains,
+						    old_write_domain);
+	}
+
+	obj->cache_level = cache_level;
+	return 0;
+}
+
 /*
- * Prepare buffer for display plane. Use uninterruptible for possible flush
- * wait, as in modesetting process we're not supposed to be interrupted.
+ * Prepare buffer for display plane (scanout, cursors, etc).
+ * Can be called from an uninterruptible phase (modesetting) and allows
+ * any flushes to be pipelined (for pageflips).
+ *
+ * For the display plane, we want to be in the GTT but out of any write
+ * domains. So in many ways this looks like set_to_gtt_domain() apart from the
+ * ability to pipeline the waits, pinning and any additional subtleties
+ * that may differentiate the display plane from ordinary buffers.
  */
 int
-i915_gem_object_set_to_display_plane(struct drm_i915_gem_object *obj,
+i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj,
+				     u32 alignment,
 				     struct intel_ring_buffer *pipelined)
 {
-	uint32_t old_read_domains;
+	u32 old_read_domains, old_write_domain;
 	int ret;
 
-	/* Not valid to be called on unbound objects. */
-	if (obj->gtt_space == NULL)
-		return -EINVAL;
-
 	ret = i915_gem_object_flush_gpu_write_domain(obj);
 	if (ret)
 		return ret;
 
-
-	/* Currently, we are always called from an non-interruptible context. */
 	if (pipelined != obj->ring) {
 		ret = i915_gem_object_wait_rendering(obj);
 		if (ret)
 			return ret;
 	}
 
+	/* The display engine is not coherent with the LLC cache on gen6.  As
+	 * a result, we make sure that the pinning that is about to occur is
+	 * done with uncached PTEs. This is lowest common denominator for all
+	 * chipsets.
+	 *
+	 * However for gen6+, we could do better by using the GFDT bit instead
+	 * of uncaching, which would allow us to flush all the LLC-cached data
+	 * with that bit in the PTE to main memory with just one PIPE_CONTROL.
+	 */
+	ret = i915_gem_object_set_cache_level(obj, I915_CACHE_NONE);
+	if (ret)
+		return ret;
+
+	/* As the user may map the buffer once pinned in the display plane
+	 * (e.g. libkms for the bootup splash), we have to ensure that we
+	 * always use map_and_fenceable for all scanout buffers.
+	 */
+	ret = i915_gem_object_pin(obj, alignment, true);
+	if (ret)
+		return ret;
+
 	i915_gem_object_flush_cpu_write_domain(obj);
 
+	old_write_domain = obj->base.write_domain;
 	old_read_domains = obj->base.read_domains;
+
+	/* It should now be out of any other write domains, and we can update
+	 * the domain values for our changes.
+	 */
+	BUG_ON((obj->base.write_domain & ~I915_GEM_DOMAIN_GTT) != 0);
 	obj->base.read_domains |= I915_GEM_DOMAIN_GTT;
 
 	trace_i915_gem_object_change_domain(obj,
 					    old_read_domains,
-					    obj->base.write_domain);
+					    old_write_domain);
 
 	return 0;
 }
 
 int
-i915_gem_object_flush_gpu(struct drm_i915_gem_object *obj)
+i915_gem_object_finish_gpu(struct drm_i915_gem_object *obj)
 {
 	int ret;
 
-	if (!obj->active)
+	if ((obj->base.read_domains & I915_GEM_GPU_DOMAINS) == 0)
 		return 0;
 
 	if (obj->base.write_domain & I915_GEM_GPU_DOMAINS) {
@@ -3050,6 +3170,9 @@ i915_gem_object_flush_gpu(struct drm_i915_gem_object *obj)
 			return ret;
 	}
 
+	/* Ensure that we invalidate the GPU's caches and TLBs. */
+	obj->base.read_domains &= ~I915_GEM_GPU_DOMAINS;
+
 	return i915_gem_object_wait_rendering(obj);
 }
 
@@ -3576,7 +3699,23 @@ 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;
 
-	obj->cache_level = I915_CACHE_NONE;
+	if (IS_GEN6(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
+		 * display scanout are coherent with the CPU in
+		 * accessing this cache.  This means in this mode we
+		 * don't need to clflush on the CPU side, and on the
+		 * GPU side we only need to flush internal caches to
+		 * get data visible to the CPU.
+		 *
+		 * However, we maintain the display planes as UC, and so
+		 * need to rebind when first used as such.
+		 */
+		obj->cache_level = I915_CACHE_LLC;
+	} else
+		obj->cache_level = I915_CACHE_NONE;
+
 	obj->base.driver_private = NULL;
 	obj->fence_reg = I915_FENCE_REG_NONE;
 	INIT_LIST_HEAD(&obj->mm_list);
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index e46b645773c..7a709cd8d54 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -59,24 +59,8 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 			      (dev_priv->mm.gtt_end - dev_priv->mm.gtt_start) / PAGE_SIZE);
 
 	list_for_each_entry(obj, &dev_priv->mm.gtt_list, gtt_list) {
-		unsigned int agp_type =
-			cache_level_to_agp_type(dev, obj->cache_level);
-
 		i915_gem_clflush_object(obj);
-
-		if (dev_priv->mm.gtt->needs_dmar) {
-			BUG_ON(!obj->sg_list);
-
-			intel_gtt_insert_sg_entries(obj->sg_list,
-						    obj->num_sg,
-						    obj->gtt_space->start >> PAGE_SHIFT,
-						    agp_type);
-		} else
-			intel_gtt_insert_pages(obj->gtt_space->start
-						   >> PAGE_SHIFT,
-					       obj->base.size >> PAGE_SHIFT,
-					       obj->pages,
-					       agp_type);
+		i915_gem_gtt_rebind_object(obj, obj->cache_level);
 	}
 
 	intel_gtt_chipset_flush();
@@ -110,6 +94,27 @@ int i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj)
 	return 0;
 }
 
+void i915_gem_gtt_rebind_object(struct drm_i915_gem_object *obj,
+				enum i915_cache_level cache_level)
+{
+	struct drm_device *dev = obj->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned int agp_type = cache_level_to_agp_type(dev, cache_level);
+
+	if (dev_priv->mm.gtt->needs_dmar) {
+		BUG_ON(!obj->sg_list);
+
+		intel_gtt_insert_sg_entries(obj->sg_list,
+					    obj->num_sg,
+					    obj->gtt_space->start >> PAGE_SHIFT,
+					    agp_type);
+	} else
+		intel_gtt_insert_pages(obj->gtt_space->start >> PAGE_SHIFT,
+				       obj->base.size >> PAGE_SHIFT,
+				       obj->pages,
+				       agp_type);
+}
+
 void i915_gem_gtt_unbind_object(struct drm_i915_gem_object *obj)
 {
 	intel_gtt_clear_range(obj->gtt_space->start >> PAGE_SHIFT,
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 3b03f85ea62..23d1ae67d27 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -361,10 +361,12 @@ static void notify_ring(struct drm_device *dev,
 
 	ring->irq_seqno = seqno;
 	wake_up_all(&ring->irq_queue);
-
-	dev_priv->hangcheck_count = 0;
-	mod_timer(&dev_priv->hangcheck_timer,
-		  jiffies + msecs_to_jiffies(DRM_I915_HANGCHECK_PERIOD));
+	if (i915_enable_hangcheck) {
+		dev_priv->hangcheck_count = 0;
+		mod_timer(&dev_priv->hangcheck_timer,
+			  jiffies +
+			  msecs_to_jiffies(DRM_I915_HANGCHECK_PERIOD));
+	}
 }
 
 static void gen6_pm_rps_work(struct work_struct *work)
@@ -1664,6 +1666,9 @@ void i915_hangcheck_elapsed(unsigned long data)
 	uint32_t acthd, instdone, instdone1;
 	bool err = false;
 
+	if (!i915_enable_hangcheck)
+		return;
+
 	/* If all work is done then ACTHD clearly hasn't advanced. */
 	if (i915_hangcheck_ring_idle(&dev_priv->ring[RCS], &err) &&
 	    i915_hangcheck_ring_idle(&dev_priv->ring[VCS], &err) &&
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 5d5def756c9..02db299f621 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -579,6 +579,7 @@
 #define   DPFC_CTL_PLANEA	(0<<30)
 #define   DPFC_CTL_PLANEB	(1<<30)
 #define   DPFC_CTL_FENCE_EN	(1<<29)
+#define   DPFC_CTL_PERSISTENT_MODE	(1<<25)
 #define   DPFC_SR_EN		(1<<10)
 #define   DPFC_CTL_LIMIT_1X	(0<<6)
 #define   DPFC_CTL_LIMIT_2X	(1<<6)
@@ -3360,6 +3361,7 @@
 #define  FORCEWAKE_ACK				0x130090
 
 #define  GT_FIFO_FREE_ENTRIES			0x120008
+#define    GT_FIFO_NUM_RESERVED_ENTRIES		20
 
 #define GEN6_RPNSWREQ				0xA008
 #define   GEN6_TURBO_DISABLE			(1<<31)
@@ -3434,7 +3436,9 @@
 #define GEN6_PCODE_MAILBOX			0x138124
 #define   GEN6_PCODE_READY			(1<<31)
 #define   GEN6_READ_OC_PARAMS			0xc
-#define   GEN6_PCODE_WRITE_MIN_FREQ_TABLE	0x9
+#define   GEN6_PCODE_WRITE_MIN_FREQ_TABLE	0x8
+#define   GEN6_PCODE_READ_MIN_FREQ_TABLE	0x9
 #define GEN6_PCODE_DATA				0x138128
+#define   GEN6_PCODE_FREQ_IA_RATIO_SHIFT	8
 
 #endif /* _I915_REG_H_ */
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index 5257cfc34c3..285758603ac 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -760,15 +760,13 @@ static void i915_restore_display(struct drm_device *dev)
 	/* FIXME: restore TV & SDVO state */
 
 	/* only restore FBC info on the platform that supports FBC*/
+	intel_disable_fbc(dev);
 	if (I915_HAS_FBC(dev)) {
 		if (HAS_PCH_SPLIT(dev)) {
-			ironlake_disable_fbc(dev);
 			I915_WRITE(ILK_DPFC_CB_BASE, dev_priv->saveDPFC_CB_BASE);
 		} else if (IS_GM45(dev)) {
-			g4x_disable_fbc(dev);
 			I915_WRITE(DPFC_CB_BASE, dev_priv->saveDPFC_CB_BASE);
 		} else {
-			i8xx_disable_fbc(dev);
 			I915_WRITE(FBC_CFB_BASE, dev_priv->saveFBC_CFB_BASE);
 			I915_WRITE(FBC_LL_BASE, dev_priv->saveFBC_LL_BASE);
 			I915_WRITE(FBC_CONTROL2, dev_priv->saveFBC_CONTROL2);
@@ -878,8 +876,10 @@ int i915_restore_state(struct drm_device *dev)
 		intel_init_emon(dev);
 	}
 
-	if (IS_GEN6(dev))
+	if (IS_GEN6(dev)) {
 		gen6_enable_rps(dev_priv);
+		gen6_update_ring_freq(dev_priv);
+	}
 
 	mutex_lock(&dev->struct_mutex);
 
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index 927442a1192..61abef8a811 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -74,7 +74,7 @@ get_blocksize(void *p)
 
 static void
 fill_detail_timing_data(struct drm_display_mode *panel_fixed_mode,
-			struct lvds_dvo_timing *dvo_timing)
+			const struct lvds_dvo_timing *dvo_timing)
 {
 	panel_fixed_mode->hdisplay = (dvo_timing->hactive_hi << 8) |
 		dvo_timing->hactive_lo;
@@ -115,20 +115,75 @@ fill_detail_timing_data(struct drm_display_mode *panel_fixed_mode,
 	drm_mode_set_name(panel_fixed_mode);
 }
 
+static bool
+lvds_dvo_timing_equal_size(const struct lvds_dvo_timing *a,
+			   const struct lvds_dvo_timing *b)
+{
+	if (a->hactive_hi != b->hactive_hi ||
+	    a->hactive_lo != b->hactive_lo)
+		return false;
+
+	if (a->hsync_off_hi != b->hsync_off_hi ||
+	    a->hsync_off_lo != b->hsync_off_lo)
+		return false;
+
+	if (a->hsync_pulse_width != b->hsync_pulse_width)
+		return false;
+
+	if (a->hblank_hi != b->hblank_hi ||
+	    a->hblank_lo != b->hblank_lo)
+		return false;
+
+	if (a->vactive_hi != b->vactive_hi ||
+	    a->vactive_lo != b->vactive_lo)
+		return false;
+
+	if (a->vsync_off != b->vsync_off)
+		return false;
+
+	if (a->vsync_pulse_width != b->vsync_pulse_width)
+		return false;
+
+	if (a->vblank_hi != b->vblank_hi ||
+	    a->vblank_lo != b->vblank_lo)
+		return false;
+
+	return true;
+}
+
+static const struct lvds_dvo_timing *
+get_lvds_dvo_timing(const struct bdb_lvds_lfp_data *lvds_lfp_data,
+		    const struct bdb_lvds_lfp_data_ptrs *lvds_lfp_data_ptrs,
+		    int index)
+{
+	/*
+	 * the size of fp_timing varies on the different platform.
+	 * So calculate the DVO timing relative offset in LVDS data
+	 * entry to get the DVO timing entry
+	 */
+
+	int lfp_data_size =
+		lvds_lfp_data_ptrs->ptr[1].dvo_timing_offset -
+		lvds_lfp_data_ptrs->ptr[0].dvo_timing_offset;
+	int dvo_timing_offset =
+		lvds_lfp_data_ptrs->ptr[0].dvo_timing_offset -
+		lvds_lfp_data_ptrs->ptr[0].fp_timing_offset;
+	char *entry = (char *)lvds_lfp_data->data + lfp_data_size * index;
+
+	return (struct lvds_dvo_timing *)(entry + dvo_timing_offset);
+}
+
 /* Try to find integrated panel data */
 static void
 parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 			    struct bdb_header *bdb)
 {
-	struct bdb_lvds_options *lvds_options;
-	struct bdb_lvds_lfp_data *lvds_lfp_data;
-	struct bdb_lvds_lfp_data_ptrs *lvds_lfp_data_ptrs;
-	struct bdb_lvds_lfp_data_entry *entry;
-	struct lvds_dvo_timing *dvo_timing;
+	const struct bdb_lvds_options *lvds_options;
+	const struct bdb_lvds_lfp_data *lvds_lfp_data;
+	const struct bdb_lvds_lfp_data_ptrs *lvds_lfp_data_ptrs;
+	const struct lvds_dvo_timing *panel_dvo_timing;
 	struct drm_display_mode *panel_fixed_mode;
-	int lfp_data_size, dvo_timing_offset;
-	int i, temp_downclock;
-	struct drm_display_mode *temp_mode;
+	int i, downclock;
 
 	lvds_options = find_section(bdb, BDB_LVDS_OPTIONS);
 	if (!lvds_options)
@@ -150,75 +205,44 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 
 	dev_priv->lvds_vbt = 1;
 
-	lfp_data_size = lvds_lfp_data_ptrs->ptr[1].dvo_timing_offset -
-		lvds_lfp_data_ptrs->ptr[0].dvo_timing_offset;
-	entry = (struct bdb_lvds_lfp_data_entry *)
-		((uint8_t *)lvds_lfp_data->data + (lfp_data_size *
-						   lvds_options->panel_type));
-	dvo_timing_offset = lvds_lfp_data_ptrs->ptr[0].dvo_timing_offset -
-		lvds_lfp_data_ptrs->ptr[0].fp_timing_offset;
-
-	/*
-	 * the size of fp_timing varies on the different platform.
-	 * So calculate the DVO timing relative offset in LVDS data
-	 * entry to get the DVO timing entry
-	 */
-	dvo_timing = (struct lvds_dvo_timing *)
-			((unsigned char *)entry + dvo_timing_offset);
+	panel_dvo_timing = get_lvds_dvo_timing(lvds_lfp_data,
+					       lvds_lfp_data_ptrs,
+					       lvds_options->panel_type);
 
 	panel_fixed_mode = kzalloc(sizeof(*panel_fixed_mode), GFP_KERNEL);
 	if (!panel_fixed_mode)
 		return;
 
-	fill_detail_timing_data(panel_fixed_mode, dvo_timing);
+	fill_detail_timing_data(panel_fixed_mode, panel_dvo_timing);
 
 	dev_priv->lfp_lvds_vbt_mode = panel_fixed_mode;
 
 	DRM_DEBUG_KMS("Found panel mode in BIOS VBT tables:\n");
 	drm_mode_debug_printmodeline(panel_fixed_mode);
 
-	temp_mode = kzalloc(sizeof(*temp_mode), GFP_KERNEL);
-	temp_downclock = panel_fixed_mode->clock;
 	/*
-	 * enumerate the LVDS panel timing info entry in VBT to check whether
-	 * the LVDS downclock is found.
+	 * Iterate over the LVDS panel timing info to find the lowest clock
+	 * for the native resolution.
 	 */
+	downclock = panel_dvo_timing->clock;
 	for (i = 0; i < 16; i++) {
-		entry = (struct bdb_lvds_lfp_data_entry *)
-			((uint8_t *)lvds_lfp_data->data + (lfp_data_size * i));
-		dvo_timing = (struct lvds_dvo_timing *)
-			((unsigned char *)entry + dvo_timing_offset);
-
-		fill_detail_timing_data(temp_mode, dvo_timing);
-
-		if (temp_mode->hdisplay == panel_fixed_mode->hdisplay &&
-		temp_mode->hsync_start == panel_fixed_mode->hsync_start &&
-		temp_mode->hsync_end == panel_fixed_mode->hsync_end &&
-		temp_mode->htotal == panel_fixed_mode->htotal &&
-		temp_mode->vdisplay == panel_fixed_mode->vdisplay &&
-		temp_mode->vsync_start == panel_fixed_mode->vsync_start &&
-		temp_mode->vsync_end == panel_fixed_mode->vsync_end &&
-		temp_mode->vtotal == panel_fixed_mode->vtotal &&
-		temp_mode->clock < temp_downclock) {
-			/*
-			 * downclock is already found. But we expect
-			 * to find the lower downclock.
-			 */
-			temp_downclock = temp_mode->clock;
-		}
-		/* clear it to zero */
-		memset(temp_mode, 0, sizeof(*temp_mode));
+		const struct lvds_dvo_timing *dvo_timing;
+
+		dvo_timing = get_lvds_dvo_timing(lvds_lfp_data,
+						 lvds_lfp_data_ptrs,
+						 i);
+		if (lvds_dvo_timing_equal_size(dvo_timing, panel_dvo_timing) &&
+		    dvo_timing->clock < downclock)
+			downclock = dvo_timing->clock;
 	}
-	kfree(temp_mode);
-	if (temp_downclock < panel_fixed_mode->clock &&
-	    i915_lvds_downclock) {
+
+	if (downclock < panel_dvo_timing->clock && i915_lvds_downclock) {
 		dev_priv->lvds_downclock_avail = 1;
-		dev_priv->lvds_downclock = temp_downclock;
+		dev_priv->lvds_downclock = downclock * 10;
 		DRM_DEBUG_KMS("LVDS downclock is found in VBT. "
 			      "Normal Clock %dKHz, downclock %dKHz\n",
-			      temp_downclock, panel_fixed_mode->clock);
+			      panel_fixed_mode->clock, 10*downclock);
 	}
-	return;
 }
 
 /* Try to find sdvo panel data */
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 21b6f93fe91..b5b15bda71d 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -24,6 +24,7 @@
  *	Eric Anholt <eric@anholt.net>
  */
 
+#include <linux/cpufreq.h>
 #include <linux/module.h>
 #include <linux/input.h>
 #include <linux/i2c.h>
@@ -1157,12 +1158,15 @@ static void intel_enable_transcoder(struct drm_i915_private *dev_priv,
 
 	reg = TRANSCONF(pipe);
 	val = I915_READ(reg);
-	/*
-	 * make the BPC in transcoder be consistent with
-	 * that in pipeconf reg.
-	 */
-	val &= ~PIPE_BPC_MASK;
-	val |= I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK;
+
+	if (HAS_PCH_IBX(dev_priv->dev)) {
+		/*
+		 * make the BPC in transcoder be consistent with
+		 * that in pipeconf reg.
+		 */
+		val &= ~PIPE_BPC_MASK;
+		val |= I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK;
+	}
 	I915_WRITE(reg, val | TRANS_ENABLE);
 	if (wait_for(I915_READ(reg) & TRANS_STATE_ENABLE, 100))
 		DRM_ERROR("failed to enable transcoder %d\n", pipe);
@@ -1380,6 +1384,28 @@ static void intel_disable_pch_ports(struct drm_i915_private *dev_priv,
 	disable_pch_hdmi(dev_priv, pipe, HDMID);
 }
 
+static void i8xx_disable_fbc(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 fbc_ctl;
+
+	/* Disable compression */
+	fbc_ctl = I915_READ(FBC_CONTROL);
+	if ((fbc_ctl & FBC_CTL_EN) == 0)
+		return;
+
+	fbc_ctl &= ~FBC_CTL_EN;
+	I915_WRITE(FBC_CONTROL, fbc_ctl);
+
+	/* Wait for compressing bit to clear */
+	if (wait_for((I915_READ(FBC_STATUS) & FBC_STAT_COMPRESSING) == 0, 10)) {
+		DRM_DEBUG_KMS("FBC idle timed out\n");
+		return;
+	}
+
+	DRM_DEBUG_KMS("disabled FBC\n");
+}
+
 static void i8xx_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
 {
 	struct drm_device *dev = crtc->dev;
@@ -1388,36 +1414,25 @@ static void i8xx_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
 	struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb);
 	struct drm_i915_gem_object *obj = intel_fb->obj;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int cfb_pitch;
 	int plane, i;
 	u32 fbc_ctl, fbc_ctl2;
 
-	if (fb->pitch == dev_priv->cfb_pitch &&
-	    obj->fence_reg == dev_priv->cfb_fence &&
-	    intel_crtc->plane == dev_priv->cfb_plane &&
-	    I915_READ(FBC_CONTROL) & FBC_CTL_EN)
-		return;
-
-	i8xx_disable_fbc(dev);
-
-	dev_priv->cfb_pitch = dev_priv->cfb_size / FBC_LL_SIZE;
-
-	if (fb->pitch < dev_priv->cfb_pitch)
-		dev_priv->cfb_pitch = fb->pitch;
+	cfb_pitch = dev_priv->cfb_size / FBC_LL_SIZE;
+	if (fb->pitch < cfb_pitch)
+		cfb_pitch = fb->pitch;
 
 	/* FBC_CTL wants 64B units */
-	dev_priv->cfb_pitch = (dev_priv->cfb_pitch / 64) - 1;
-	dev_priv->cfb_fence = obj->fence_reg;
-	dev_priv->cfb_plane = intel_crtc->plane;
-	plane = dev_priv->cfb_plane == 0 ? FBC_CTL_PLANEA : FBC_CTL_PLANEB;
+	cfb_pitch = (cfb_pitch / 64) - 1;
+	plane = intel_crtc->plane == 0 ? FBC_CTL_PLANEA : FBC_CTL_PLANEB;
 
 	/* Clear old tags */
 	for (i = 0; i < (FBC_LL_SIZE / 32) + 1; i++)
 		I915_WRITE(FBC_TAG + (i * 4), 0);
 
 	/* Set it up... */
-	fbc_ctl2 = FBC_CTL_FENCE_DBL | FBC_CTL_IDLE_IMM | plane;
-	if (obj->tiling_mode != I915_TILING_NONE)
-		fbc_ctl2 |= FBC_CTL_CPU_FENCE;
+	fbc_ctl2 = FBC_CTL_FENCE_DBL | FBC_CTL_IDLE_IMM | FBC_CTL_CPU_FENCE;
+	fbc_ctl2 |= plane;
 	I915_WRITE(FBC_CONTROL2, fbc_ctl2);
 	I915_WRITE(FBC_FENCE_OFF, crtc->y);
 
@@ -1425,36 +1440,13 @@ static void i8xx_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
 	fbc_ctl = FBC_CTL_EN | FBC_CTL_PERIODIC;
 	if (IS_I945GM(dev))
 		fbc_ctl |= FBC_CTL_C3_IDLE; /* 945 needs special SR handling */
-	fbc_ctl |= (dev_priv->cfb_pitch & 0xff) << FBC_CTL_STRIDE_SHIFT;
+	fbc_ctl |= (cfb_pitch & 0xff) << FBC_CTL_STRIDE_SHIFT;
 	fbc_ctl |= (interval & 0x2fff) << FBC_CTL_INTERVAL_SHIFT;
-	if (obj->tiling_mode != I915_TILING_NONE)
-		fbc_ctl |= dev_priv->cfb_fence;
-	I915_WRITE(FBC_CONTROL, fbc_ctl);
-
-	DRM_DEBUG_KMS("enabled FBC, pitch %ld, yoff %d, plane %d, ",
-		      dev_priv->cfb_pitch, crtc->y, dev_priv->cfb_plane);
-}
-
-void i8xx_disable_fbc(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 fbc_ctl;
-
-	/* Disable compression */
-	fbc_ctl = I915_READ(FBC_CONTROL);
-	if ((fbc_ctl & FBC_CTL_EN) == 0)
-		return;
-
-	fbc_ctl &= ~FBC_CTL_EN;
+	fbc_ctl |= obj->fence_reg;
 	I915_WRITE(FBC_CONTROL, fbc_ctl);
 
-	/* Wait for compressing bit to clear */
-	if (wait_for((I915_READ(FBC_STATUS) & FBC_STAT_COMPRESSING) == 0, 10)) {
-		DRM_DEBUG_KMS("FBC idle timed out\n");
-		return;
-	}
-
-	DRM_DEBUG_KMS("disabled FBC\n");
+	DRM_DEBUG_KMS("enabled FBC, pitch %d, yoff %d, plane %d, ",
+		      cfb_pitch, crtc->y, intel_crtc->plane);
 }
 
 static bool i8xx_fbc_enabled(struct drm_device *dev)
@@ -1476,30 +1468,9 @@ static void g4x_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
 	unsigned long stall_watermark = 200;
 	u32 dpfc_ctl;
 
-	dpfc_ctl = I915_READ(DPFC_CONTROL);
-	if (dpfc_ctl & DPFC_CTL_EN) {
-		if (dev_priv->cfb_pitch == dev_priv->cfb_pitch / 64 - 1 &&
-		    dev_priv->cfb_fence == obj->fence_reg &&
-		    dev_priv->cfb_plane == intel_crtc->plane &&
-		    dev_priv->cfb_y == crtc->y)
-			return;
-
-		I915_WRITE(DPFC_CONTROL, dpfc_ctl & ~DPFC_CTL_EN);
-		intel_wait_for_vblank(dev, intel_crtc->pipe);
-	}
-
-	dev_priv->cfb_pitch = (dev_priv->cfb_pitch / 64) - 1;
-	dev_priv->cfb_fence = obj->fence_reg;
-	dev_priv->cfb_plane = intel_crtc->plane;
-	dev_priv->cfb_y = crtc->y;
-
 	dpfc_ctl = plane | DPFC_SR_EN | DPFC_CTL_LIMIT_1X;
-	if (obj->tiling_mode != I915_TILING_NONE) {
-		dpfc_ctl |= DPFC_CTL_FENCE_EN | dev_priv->cfb_fence;
-		I915_WRITE(DPFC_CHICKEN, DPFC_HT_MODIFY);
-	} else {
-		I915_WRITE(DPFC_CHICKEN, ~DPFC_HT_MODIFY);
-	}
+	dpfc_ctl |= DPFC_CTL_FENCE_EN | obj->fence_reg;
+	I915_WRITE(DPFC_CHICKEN, DPFC_HT_MODIFY);
 
 	I915_WRITE(DPFC_RECOMP_CTL, DPFC_RECOMP_STALL_EN |
 		   (stall_watermark << DPFC_RECOMP_STALL_WM_SHIFT) |
@@ -1512,7 +1483,7 @@ static void g4x_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
 	DRM_DEBUG_KMS("enabled fbc on plane %d\n", intel_crtc->plane);
 }
 
-void g4x_disable_fbc(struct drm_device *dev)
+static void g4x_disable_fbc(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 dpfc_ctl;
@@ -1567,32 +1538,12 @@ static void ironlake_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
 	u32 dpfc_ctl;
 
 	dpfc_ctl = I915_READ(ILK_DPFC_CONTROL);
-	if (dpfc_ctl & DPFC_CTL_EN) {
-		if (dev_priv->cfb_pitch == dev_priv->cfb_pitch / 64 - 1 &&
-		    dev_priv->cfb_fence == obj->fence_reg &&
-		    dev_priv->cfb_plane == intel_crtc->plane &&
-		    dev_priv->cfb_offset == obj->gtt_offset &&
-		    dev_priv->cfb_y == crtc->y)
-			return;
-
-		I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl & ~DPFC_CTL_EN);
-		intel_wait_for_vblank(dev, intel_crtc->pipe);
-	}
-
-	dev_priv->cfb_pitch = (dev_priv->cfb_pitch / 64) - 1;
-	dev_priv->cfb_fence = obj->fence_reg;
-	dev_priv->cfb_plane = intel_crtc->plane;
-	dev_priv->cfb_offset = obj->gtt_offset;
-	dev_priv->cfb_y = crtc->y;
-
 	dpfc_ctl &= DPFC_RESERVED;
 	dpfc_ctl |= (plane | DPFC_CTL_LIMIT_1X);
-	if (obj->tiling_mode != I915_TILING_NONE) {
-		dpfc_ctl |= (DPFC_CTL_FENCE_EN | dev_priv->cfb_fence);
-		I915_WRITE(ILK_DPFC_CHICKEN, DPFC_HT_MODIFY);
-	} else {
-		I915_WRITE(ILK_DPFC_CHICKEN, ~DPFC_HT_MODIFY);
-	}
+	/* Set persistent mode for front-buffer rendering, ala X. */
+	dpfc_ctl |= DPFC_CTL_PERSISTENT_MODE;
+	dpfc_ctl |= (DPFC_CTL_FENCE_EN | obj->fence_reg);
+	I915_WRITE(ILK_DPFC_CHICKEN, DPFC_HT_MODIFY);
 
 	I915_WRITE(ILK_DPFC_RECOMP_CTL, DPFC_RECOMP_STALL_EN |
 		   (stall_watermark << DPFC_RECOMP_STALL_WM_SHIFT) |
@@ -1604,7 +1555,7 @@ static void ironlake_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
 
 	if (IS_GEN6(dev)) {
 		I915_WRITE(SNB_DPFC_CTL_SA,
-			   SNB_CPU_FENCE_ENABLE | dev_priv->cfb_fence);
+			   SNB_CPU_FENCE_ENABLE | obj->fence_reg);
 		I915_WRITE(DPFC_CPU_FENCE_OFFSET, crtc->y);
 		sandybridge_blit_fbc_update(dev);
 	}
@@ -1612,7 +1563,7 @@ static void ironlake_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
 	DRM_DEBUG_KMS("enabled fbc on plane %d\n", intel_crtc->plane);
 }
 
-void ironlake_disable_fbc(struct drm_device *dev)
+static void ironlake_disable_fbc(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 dpfc_ctl;
@@ -1644,24 +1595,109 @@ bool intel_fbc_enabled(struct drm_device *dev)
 	return dev_priv->display.fbc_enabled(dev);
 }
 
-void intel_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
+static void intel_fbc_work_fn(struct work_struct *__work)
 {
-	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
+	struct intel_fbc_work *work =
+		container_of(to_delayed_work(__work),
+			     struct intel_fbc_work, work);
+	struct drm_device *dev = work->crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	mutex_lock(&dev->struct_mutex);
+	if (work == dev_priv->fbc_work) {
+		/* Double check that we haven't switched fb without cancelling
+		 * the prior work.
+		 */
+		if (work->crtc->fb == work->fb) {
+			dev_priv->display.enable_fbc(work->crtc,
+						     work->interval);
+
+			dev_priv->cfb_plane = to_intel_crtc(work->crtc)->plane;
+			dev_priv->cfb_fb = work->crtc->fb->base.id;
+			dev_priv->cfb_y = work->crtc->y;
+		}
+
+		dev_priv->fbc_work = NULL;
+	}
+	mutex_unlock(&dev->struct_mutex);
+
+	kfree(work);
+}
+
+static void intel_cancel_fbc_work(struct drm_i915_private *dev_priv)
+{
+	if (dev_priv->fbc_work == NULL)
+		return;
+
+	DRM_DEBUG_KMS("cancelling pending FBC enable\n");
+
+	/* Synchronisation is provided by struct_mutex and checking of
+	 * dev_priv->fbc_work, so we can perform the cancellation
+	 * entirely asynchronously.
+	 */
+	if (cancel_delayed_work(&dev_priv->fbc_work->work))
+		/* tasklet was killed before being run, clean up */
+		kfree(dev_priv->fbc_work);
+
+	/* Mark the work as no longer wanted so that if it does
+	 * wake-up (because the work was already running and waiting
+	 * for our mutex), it will discover that is no longer
+	 * necessary to run.
+	 */
+	dev_priv->fbc_work = NULL;
+}
+
+static void intel_enable_fbc(struct drm_crtc *crtc, unsigned long interval)
+{
+	struct intel_fbc_work *work;
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (!dev_priv->display.enable_fbc)
 		return;
 
-	dev_priv->display.enable_fbc(crtc, interval);
+	intel_cancel_fbc_work(dev_priv);
+
+	work = kzalloc(sizeof *work, GFP_KERNEL);
+	if (work == NULL) {
+		dev_priv->display.enable_fbc(crtc, interval);
+		return;
+	}
+
+	work->crtc = crtc;
+	work->fb = crtc->fb;
+	work->interval = interval;
+	INIT_DELAYED_WORK(&work->work, intel_fbc_work_fn);
+
+	dev_priv->fbc_work = work;
+
+	DRM_DEBUG_KMS("scheduling delayed FBC enable\n");
+
+	/* Delay the actual enabling to let pageflipping cease and the
+	 * display to settle before starting the compression. Note that
+	 * this delay also serves a second purpose: it allows for a
+	 * vblank to pass after disabling the FBC before we attempt
+	 * to modify the control registers.
+	 *
+	 * A more complicated solution would involve tracking vblanks
+	 * following the termination of the page-flipping sequence
+	 * and indeed performing the enable as a co-routine and not
+	 * waiting synchronously upon the vblank.
+	 */
+	schedule_delayed_work(&work->work, msecs_to_jiffies(50));
 }
 
 void intel_disable_fbc(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
+	intel_cancel_fbc_work(dev_priv);
+
 	if (!dev_priv->display.disable_fbc)
 		return;
 
 	dev_priv->display.disable_fbc(dev);
+	dev_priv->cfb_plane = -1;
 }
 
 /**
@@ -1760,8 +1796,13 @@ static void intel_update_fbc(struct drm_device *dev)
 		dev_priv->no_fbc_reason = FBC_BAD_PLANE;
 		goto out_disable;
 	}
-	if (obj->tiling_mode != I915_TILING_X) {
-		DRM_DEBUG_KMS("framebuffer not tiled, disabling compression\n");
+
+	/* The use of a CPU fence is mandatory in order to detect writes
+	 * by the CPU to the scanout and trigger updates to the FBC.
+	 */
+	if (obj->tiling_mode != I915_TILING_X ||
+	    obj->fence_reg == I915_FENCE_REG_NONE) {
+		DRM_DEBUG_KMS("framebuffer not tiled or fenced, disabling compression\n");
 		dev_priv->no_fbc_reason = FBC_NOT_TILED;
 		goto out_disable;
 	}
@@ -1770,6 +1811,44 @@ static void intel_update_fbc(struct drm_device *dev)
 	if (in_dbg_master())
 		goto out_disable;
 
+	/* If the scanout has not changed, don't modify the FBC settings.
+	 * Note that we make the fundamental assumption that the fb->obj
+	 * cannot be unpinned (and have its GTT offset and fence revoked)
+	 * without first being decoupled from the scanout and FBC disabled.
+	 */
+	if (dev_priv->cfb_plane == intel_crtc->plane &&
+	    dev_priv->cfb_fb == fb->base.id &&
+	    dev_priv->cfb_y == crtc->y)
+		return;
+
+	if (intel_fbc_enabled(dev)) {
+		/* We update FBC along two paths, after changing fb/crtc
+		 * configuration (modeswitching) and after page-flipping
+		 * finishes. For the latter, we know that not only did
+		 * we disable the FBC at the start of the page-flip
+		 * sequence, but also more than one vblank has passed.
+		 *
+		 * For the former case of modeswitching, it is possible
+		 * to switch between two FBC valid configurations
+		 * instantaneously so we do need to disable the FBC
+		 * before we can modify its control registers. We also
+		 * have to wait for the next vblank for that to take
+		 * effect. However, since we delay enabling FBC we can
+		 * assume that a vblank has passed since disabling and
+		 * that we can safely alter the registers in the deferred
+		 * callback.
+		 *
+		 * In the scenario that we go from a valid to invalid
+		 * and then back to valid FBC configuration we have
+		 * no strict enforcement that a vblank occurred since
+		 * disabling the FBC. However, along all current pipe
+		 * disabling paths we do need to wait for a vblank at
+		 * some point. And we wait before enabling FBC anyway.
+		 */
+		DRM_DEBUG_KMS("disabling active FBC for update\n");
+		intel_disable_fbc(dev);
+	}
+
 	intel_enable_fbc(crtc, 500);
 	return;
 
@@ -1812,14 +1891,10 @@ intel_pin_and_fence_fb_obj(struct drm_device *dev,
 	}
 
 	dev_priv->mm.interruptible = false;
-	ret = i915_gem_object_pin(obj, alignment, true);
+	ret = i915_gem_object_pin_to_display_plane(obj, alignment, pipelined);
 	if (ret)
 		goto err_interruptible;
 
-	ret = i915_gem_object_set_to_display_plane(obj, pipelined);
-	if (ret)
-		goto err_unpin;
-
 	/* Install a fence for tiled scan-out. Pre-i965 always needs a
 	 * fence, whereas 965+ only requires a fence if using
 	 * framebuffer compression.  For simplicity, we always install
@@ -1841,10 +1916,8 @@ err_interruptible:
 	return ret;
 }
 
-/* Assume fb object is pinned & idle & fenced and just update base pointers */
-static int
-intel_pipe_set_base_atomic(struct drm_crtc *crtc, struct drm_framebuffer *fb,
-			   int x, int y, enum mode_set_atomic state)
+static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb,
+			     int x, int y)
 {
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1887,7 +1960,7 @@ intel_pipe_set_base_atomic(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 		dspcntr |= DISPPLANE_32BPP_NO_ALPHA;
 		break;
 	default:
-		DRM_ERROR("Unknown color depth\n");
+		DRM_ERROR("Unknown color depth %d\n", fb->bits_per_pixel);
 		return -EINVAL;
 	}
 	if (INTEL_INFO(dev)->gen >= 4) {
@@ -1897,10 +1970,6 @@ intel_pipe_set_base_atomic(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 			dspcntr &= ~DISPPLANE_TILED;
 	}
 
-	if (HAS_PCH_SPLIT(dev))
-		/* must disable */
-		dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE;
-
 	I915_WRITE(reg, dspcntr);
 
 	Start = obj->gtt_offset;
@@ -1917,6 +1986,99 @@ intel_pipe_set_base_atomic(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 		I915_WRITE(DSPADDR(plane), Start + Offset);
 	POSTING_READ(reg);
 
+	return 0;
+}
+
+static int ironlake_update_plane(struct drm_crtc *crtc,
+				 struct drm_framebuffer *fb, int x, int y)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_framebuffer *intel_fb;
+	struct drm_i915_gem_object *obj;
+	int plane = intel_crtc->plane;
+	unsigned long Start, Offset;
+	u32 dspcntr;
+	u32 reg;
+
+	switch (plane) {
+	case 0:
+	case 1:
+		break;
+	default:
+		DRM_ERROR("Can't update plane %d in SAREA\n", plane);
+		return -EINVAL;
+	}
+
+	intel_fb = to_intel_framebuffer(fb);
+	obj = intel_fb->obj;
+
+	reg = DSPCNTR(plane);
+	dspcntr = I915_READ(reg);
+	/* Mask out pixel format bits in case we change it */
+	dspcntr &= ~DISPPLANE_PIXFORMAT_MASK;
+	switch (fb->bits_per_pixel) {
+	case 8:
+		dspcntr |= DISPPLANE_8BPP;
+		break;
+	case 16:
+		if (fb->depth != 16)
+			return -EINVAL;
+
+		dspcntr |= DISPPLANE_16BPP;
+		break;
+	case 24:
+	case 32:
+		if (fb->depth == 24)
+			dspcntr |= DISPPLANE_32BPP_NO_ALPHA;
+		else if (fb->depth == 30)
+			dspcntr |= DISPPLANE_32BPP_30BIT_NO_ALPHA;
+		else
+			return -EINVAL;
+		break;
+	default:
+		DRM_ERROR("Unknown color depth %d\n", fb->bits_per_pixel);
+		return -EINVAL;
+	}
+
+	if (obj->tiling_mode != I915_TILING_NONE)
+		dspcntr |= DISPPLANE_TILED;
+	else
+		dspcntr &= ~DISPPLANE_TILED;
+
+	/* must disable */
+	dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE;
+
+	I915_WRITE(reg, dspcntr);
+
+	Start = obj->gtt_offset;
+	Offset = y * fb->pitch + x * (fb->bits_per_pixel / 8);
+
+	DRM_DEBUG_KMS("Writing base %08lX %08lX %d %d %d\n",
+		      Start, Offset, x, y, fb->pitch);
+	I915_WRITE(DSPSTRIDE(plane), fb->pitch);
+	I915_WRITE(DSPSURF(plane), Start);
+	I915_WRITE(DSPTILEOFF(plane), (y << 16) | x);
+	I915_WRITE(DSPADDR(plane), Offset);
+	POSTING_READ(reg);
+
+	return 0;
+}
+
+/* Assume fb object is pinned & idle & fenced and just update base pointers */
+static int
+intel_pipe_set_base_atomic(struct drm_crtc *crtc, struct drm_framebuffer *fb,
+			   int x, int y, enum mode_set_atomic state)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int ret;
+
+	ret = dev_priv->display.update_plane(crtc, fb, x, y);
+	if (ret)
+		return ret;
+
 	intel_update_fbc(dev);
 	intel_increase_pllclock(crtc);
 
@@ -1971,7 +2133,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 		 * This should only fail upon a hung GPU, in which case we
 		 * can safely continue.
 		 */
-		ret = i915_gem_object_flush_gpu(obj);
+		ret = i915_gem_object_finish_gpu(obj);
 		(void) ret;
 	}
 
@@ -2622,6 +2784,7 @@ 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)) {
+		u32 bpc = (I915_READ(PIPECONF(pipe)) & PIPE_BPC_MASK) >> 5;
 		reg = TRANS_DP_CTL(pipe);
 		temp = I915_READ(reg);
 		temp &= ~(TRANS_DP_PORT_SEL_MASK |
@@ -2629,7 +2792,7 @@ static void ironlake_pch_enable(struct drm_crtc *crtc)
 			  TRANS_DP_BPC_MASK);
 		temp |= (TRANS_DP_OUTPUT_ENABLE |
 			 TRANS_DP_ENH_FRAMING);
-		temp |= TRANS_DP_8BPC;
+		temp |= bpc << 9; /* same format but at 11:9 */
 
 		if (crtc->mode.flags & DRM_MODE_FLAG_PHSYNC)
 			temp |= TRANS_DP_HSYNC_ACTIVE_HIGH;
@@ -2732,9 +2895,8 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 
 	intel_disable_plane(dev_priv, plane, pipe);
 
-	if (dev_priv->cfb_plane == plane &&
-	    dev_priv->display.disable_fbc)
-		dev_priv->display.disable_fbc(dev);
+	if (dev_priv->cfb_plane == plane)
+		intel_disable_fbc(dev);
 
 	intel_disable_pipe(dev_priv, pipe);
 
@@ -2898,9 +3060,8 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 	intel_crtc_dpms_overlay(intel_crtc, false);
 	intel_crtc_update_cursor(crtc, false);
 
-	if (dev_priv->cfb_plane == plane &&
-	    dev_priv->display.disable_fbc)
-		dev_priv->display.disable_fbc(dev);
+	if (dev_priv->cfb_plane == plane)
+		intel_disable_fbc(dev);
 
 	intel_disable_plane(dev_priv, plane, pipe);
 	intel_disable_pipe(dev_priv, pipe);
@@ -4308,6 +4469,133 @@ static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv)
 	return dev_priv->lvds_use_ssc && i915_panel_use_ssc;
 }
 
+/**
+ * intel_choose_pipe_bpp_dither - figure out what color depth the pipe should send
+ * @crtc: CRTC structure
+ *
+ * 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.
+ *
+ * If possible, match the pipe depth to the fb depth.  In some cases, this
+ * isn't ideal, because the connected output supports a lesser or restricted
+ * set of depths.  Resolve that here:
+ *    LVDS typically supports only 6bpc, so clamp down in that case
+ *    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.
+ *
+ * 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)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_encoder *encoder;
+	struct drm_connector *connector;
+	unsigned int display_bpc = UINT_MAX, bpc;
+
+	/* Walk the encoders & connectors on this crtc, get min bpc */
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
+		struct intel_encoder *intel_encoder = to_intel_encoder(encoder);
+
+		if (encoder->crtc != crtc)
+			continue;
+
+		if (intel_encoder->type == INTEL_OUTPUT_LVDS) {
+			unsigned int lvds_bpc;
+
+			if ((I915_READ(PCH_LVDS) & LVDS_A3_POWER_MASK) ==
+			    LVDS_A3_POWER_UP)
+				lvds_bpc = 8;
+			else
+				lvds_bpc = 6;
+
+			if (lvds_bpc < display_bpc) {
+				DRM_DEBUG_DRIVER("clamping display bpc (was %d) to LVDS (%d)\n", display_bpc, lvds_bpc);
+				display_bpc = lvds_bpc;
+			}
+			continue;
+		}
+
+		if (intel_encoder->type == INTEL_OUTPUT_EDP) {
+			/* Use VBT settings if we have an eDP panel */
+			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);
+				display_bpc = edp_bpc;
+			}
+			continue;
+		}
+
+		/* Not one of the known troublemakers, check the EDID */
+		list_for_each_entry(connector, &dev->mode_config.connector_list,
+				    head) {
+			if (connector->encoder != encoder)
+				continue;
+
+			if (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);
+				display_bpc = connector->display_info.bpc;
+			}
+		}
+
+		/*
+		 * HDMI is either 12 or 8, so if the display lets 10bpc sneak
+		 * through, clamp it down.  (Note: >12bpc will be caught below.)
+		 */
+		if (intel_encoder->type == INTEL_OUTPUT_HDMI) {
+			if (display_bpc > 8 && display_bpc < 12) {
+				DRM_DEBUG_DRIVER("forcing bpc to 12 for HDMI\n");
+				display_bpc = 12;
+			} else {
+				DRM_DEBUG_DRIVER("forcing bpc to 8 for HDMI\n");
+				display_bpc = 8;
+			}
+		}
+	}
+
+	/*
+	 * We could just drive the pipe at the highest bpc all the time and
+	 * enable dithering as needed, but that costs bandwidth.  So choose
+	 * the minimum value that expresses the full color range of the fb but
+	 * also stays within the max display bpc discovered above.
+	 */
+
+	switch (crtc->fb->depth) {
+	case 8:
+		bpc = 8; /* since we go through a colormap */
+		break;
+	case 15:
+	case 16:
+		bpc = 6; /* min is 18bpp */
+		break;
+	case 24:
+		bpc = min((unsigned int)8, display_bpc);
+		break;
+	case 30:
+		bpc = min((unsigned int)10, display_bpc);
+		break;
+	case 48:
+		bpc = min((unsigned int)12, display_bpc);
+		break;
+	default:
+		DRM_DEBUG("unsupported depth, assuming 24 bits\n");
+		bpc = min((unsigned int)8, display_bpc);
+		break;
+	}
+
+	DRM_DEBUG_DRIVER("setting pipe bpc to %d (max display bpc %d)\n",
+			 bpc, display_bpc);
+
+	*pipe_bpp = bpc * 3;
+
+	return display_bpc != bpc;
+}
+
 static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 			      struct drm_display_mode *mode,
 			      struct drm_display_mode *adjusted_mode,
@@ -4720,7 +5008,9 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 	struct fdi_m_n m_n = {0};
 	u32 temp;
 	u32 lvds_sync = 0;
-	int target_clock, pixel_multiplier, lane, link_bw, bpp, factor;
+	int target_clock, pixel_multiplier, lane, link_bw, factor;
+	unsigned int pipe_bpp;
+	bool dither;
 
 	list_for_each_entry(encoder, &mode_config->encoder_list, base.head) {
 		if (encoder->base.crtc != crtc)
@@ -4847,56 +5137,37 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 	/* determine panel color depth */
 	temp = I915_READ(PIPECONF(pipe));
 	temp &= ~PIPE_BPC_MASK;
-	if (is_lvds) {
-		/* the BPC will be 6 if it is 18-bit LVDS panel */
-		if ((I915_READ(PCH_LVDS) & LVDS_A3_POWER_MASK) == LVDS_A3_POWER_UP)
-			temp |= PIPE_8BPC;
-		else
-			temp |= PIPE_6BPC;
-	} else if (has_edp_encoder) {
-		switch (dev_priv->edp.bpp/3) {
-		case 8:
-			temp |= PIPE_8BPC;
-			break;
-		case 10:
-			temp |= PIPE_10BPC;
-			break;
-		case 6:
-			temp |= PIPE_6BPC;
-			break;
-		case 12:
-			temp |= PIPE_12BPC;
-			break;
-		}
-	} else
-		temp |= PIPE_8BPC;
-	I915_WRITE(PIPECONF(pipe), temp);
-
-	switch (temp & PIPE_BPC_MASK) {
-	case PIPE_8BPC:
-		bpp = 24;
+	dither = intel_choose_pipe_bpp_dither(crtc, &pipe_bpp);
+	switch (pipe_bpp) {
+	case 18:
+		temp |= PIPE_6BPC;
 		break;
-	case PIPE_10BPC:
-		bpp = 30;
+	case 24:
+		temp |= PIPE_8BPC;
 		break;
-	case PIPE_6BPC:
-		bpp = 18;
+	case 30:
+		temp |= PIPE_10BPC;
 		break;
-	case PIPE_12BPC:
-		bpp = 36;
+	case 36:
+		temp |= PIPE_12BPC;
 		break;
 	default:
-		DRM_ERROR("unknown pipe bpc value\n");
-		bpp = 24;
+		WARN(1, "intel_choose_pipe_bpp returned invalid value\n");
+		temp |= PIPE_8BPC;
+		pipe_bpp = 24;
+		break;
 	}
 
+	intel_crtc->bpp = pipe_bpp;
+	I915_WRITE(PIPECONF(pipe), temp);
+
 	if (!lane) {
 		/*
 		 * Account for spread spectrum to avoid
 		 * oversubscribing the link. Max center spread
 		 * is 2.5%; use 5% for safety's sake.
 		 */
-		u32 bps = target_clock * bpp * 21 / 20;
+		u32 bps = target_clock * intel_crtc->bpp * 21 / 20;
 		lane = bps / (link_bw * 8) + 1;
 	}
 
@@ -4904,7 +5175,8 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 
 	if (pixel_multiplier > 1)
 		link_bw *= pixel_multiplier;
-	ironlake_compute_m_n(bpp, lane, target_clock, link_bw, &m_n);
+	ironlake_compute_m_n(intel_crtc->bpp, lane, target_clock, link_bw,
+			     &m_n);
 
 	/* Ironlake: try to setup display ref clock before DPLL
 	 * enabling. This is only under driver's control after
@@ -5107,14 +5379,12 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
 		I915_WRITE(PCH_LVDS, temp);
 	}
 
-	/* set the dithering flag and clear for anything other than a panel. */
 	pipeconf &= ~PIPECONF_DITHER_EN;
 	pipeconf &= ~PIPECONF_DITHER_TYPE_MASK;
-	if (dev_priv->lvds_dither && (is_lvds || has_edp_encoder)) {
+	if ((is_lvds && dev_priv->lvds_dither) || dither) {
 		pipeconf |= PIPECONF_DITHER_EN;
 		pipeconf |= PIPECONF_DITHER_TYPE_ST1;
 	}
-
 	if (is_dp || intel_encoder_is_pch_edp(&has_edp_encoder->base)) {
 		intel_dp_set_m_n(crtc, mode, adjusted_mode);
 	} else {
@@ -5434,21 +5704,15 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc,
 			goto fail_locked;
 		}
 
-		ret = i915_gem_object_pin(obj, PAGE_SIZE, true);
-		if (ret) {
-			DRM_ERROR("failed to pin cursor bo\n");
-			goto fail_locked;
-		}
-
-		ret = i915_gem_object_set_to_gtt_domain(obj, 0);
+		ret = i915_gem_object_pin_to_display_plane(obj, 0, NULL);
 		if (ret) {
 			DRM_ERROR("failed to move cursor bo into the GTT\n");
-			goto fail_unpin;
+			goto fail_locked;
 		}
 
 		ret = i915_gem_object_put_fence(obj);
 		if (ret) {
-			DRM_ERROR("failed to move cursor bo into the GTT\n");
+			DRM_ERROR("failed to release fence for cursor");
 			goto fail_unpin;
 		}
 
@@ -6151,6 +6415,7 @@ static void intel_unpin_work_fn(struct work_struct *__work)
 	drm_gem_object_unreference(&work->pending_flip_obj->base);
 	drm_gem_object_unreference(&work->old_fb_obj->base);
 
+	intel_update_fbc(work->dev);
 	mutex_unlock(&work->dev->struct_mutex);
 	kfree(work);
 }
@@ -6515,6 +6780,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	if (ret)
 		goto cleanup_pending;
 
+	intel_disable_fbc(dev);
 	mutex_unlock(&dev->struct_mutex);
 
 	trace_i915_flip_request(intel_crtc->plane, obj);
@@ -6643,6 +6909,7 @@ static void intel_crtc_init(struct drm_device *dev, int pipe)
 
 	intel_crtc_reset(&intel_crtc->base);
 	intel_crtc->active = true; /* force the pipe off on setup_init_config */
+	intel_crtc->bpp = 24; /* default for pre-Ironlake */
 
 	if (HAS_PCH_SPLIT(dev)) {
 		intel_helper_funcs.prepare = ironlake_crtc_prepare;
@@ -6869,6 +7136,11 @@ int intel_framebuffer_init(struct drm_device *dev,
 	switch (mode_cmd->bpp) {
 	case 8:
 	case 16:
+		/* Only pre-ILK can handle 5:5:5 */
+		if (mode_cmd->depth == 15 && !HAS_PCH_SPLIT(dev))
+			return -EINVAL;
+		break;
+
 	case 24:
 	case 32:
 		break;
@@ -7283,6 +7555,59 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
 	mutex_unlock(&dev_priv->dev->struct_mutex);
 }
 
+void gen6_update_ring_freq(struct drm_i915_private *dev_priv)
+{
+	int min_freq = 15;
+	int gpu_freq, ia_freq, max_ia_freq;
+	int scaling_factor = 180;
+
+	max_ia_freq = cpufreq_quick_get_max(0);
+	/*
+	 * Default to measured freq if none found, PCU will ensure we don't go
+	 * over
+	 */
+	if (!max_ia_freq)
+		max_ia_freq = tsc_khz;
+
+	/* Convert from kHz to MHz */
+	max_ia_freq /= 1000;
+
+	mutex_lock(&dev_priv->dev->struct_mutex);
+
+	/*
+	 * For each potential GPU frequency, load a ring frequency we'd like
+	 * to use for memory access.  We do this by specifying the IA frequency
+	 * the PCU should use as a reference to determine the ring frequency.
+	 */
+	for (gpu_freq = dev_priv->max_delay; gpu_freq >= dev_priv->min_delay;
+	     gpu_freq--) {
+		int diff = dev_priv->max_delay - gpu_freq;
+
+		/*
+		 * For GPU frequencies less than 750MHz, just use the lowest
+		 * ring freq.
+		 */
+		if (gpu_freq < min_freq)
+			ia_freq = 800;
+		else
+			ia_freq = max_ia_freq - ((diff * scaling_factor) / 2);
+		ia_freq = DIV_ROUND_CLOSEST(ia_freq, 100);
+
+		I915_WRITE(GEN6_PCODE_DATA,
+			   (ia_freq << GEN6_PCODE_FREQ_IA_RATIO_SHIFT) |
+			   gpu_freq);
+		I915_WRITE(GEN6_PCODE_MAILBOX, GEN6_PCODE_READY |
+			   GEN6_PCODE_WRITE_MIN_FREQ_TABLE);
+		if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) &
+			      GEN6_PCODE_READY) == 0, 10)) {
+			DRM_ERROR("pcode write of freq table timed out\n");
+			continue;
+		}
+	}
+
+	mutex_unlock(&dev_priv->dev->struct_mutex);
+}
+
 static void ironlake_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -7639,9 +7964,11 @@ static void intel_init_display(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev)) {
 		dev_priv->display.dpms = ironlake_crtc_dpms;
 		dev_priv->display.crtc_mode_set = ironlake_crtc_mode_set;
+		dev_priv->display.update_plane = ironlake_update_plane;
 	} else {
 		dev_priv->display.dpms = i9xx_crtc_dpms;
 		dev_priv->display.crtc_mode_set = i9xx_crtc_mode_set;
+		dev_priv->display.update_plane = i9xx_update_plane;
 	}
 
 	if (I915_HAS_FBC(dev)) {
@@ -7926,8 +8253,10 @@ void intel_modeset_init(struct drm_device *dev)
 		intel_init_emon(dev);
 	}
 
-	if (IS_GEN6(dev))
+	if (IS_GEN6(dev) || IS_GEN7(dev)) {
 		gen6_enable_rps(dev_priv);
+		gen6_update_ring_freq(dev_priv);
+	}
 
 	INIT_WORK(&dev_priv->idle_work, intel_idle_update);
 	setup_timer(&dev_priv->idle_timer, intel_gpu_idle_timer,
@@ -7963,12 +8292,11 @@ void intel_modeset_cleanup(struct drm_device *dev)
 		intel_increase_pllclock(crtc);
 	}
 
-	if (dev_priv->display.disable_fbc)
-		dev_priv->display.disable_fbc(dev);
+	intel_disable_fbc(dev);
 
 	if (IS_IRONLAKE_M(dev))
 		ironlake_disable_drps(dev);
-	if (IS_GEN6(dev))
+	if (IS_GEN6(dev) || IS_GEN7(dev))
 		gen6_disable_rps(dev);
 
 	if (IS_IRONLAKE_M(dev))
@@ -7981,6 +8309,9 @@ void intel_modeset_cleanup(struct drm_device *dev)
 	drm_irq_uninstall(dev);
 	cancel_work_sync(&dev_priv->hotplug_work);
 
+	/* flush any delayed tasks or pending work */
+	flush_scheduled_work();
+
 	/* Shut off idle work before the crtcs get freed. */
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		intel_crtc = to_intel_crtc(crtc);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 391b55f1cc7..f797fb58ba9 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -50,7 +50,6 @@ struct intel_dp {
 	bool has_audio;
 	int force_audio;
 	uint32_t color_range;
-	int dpms_mode;
 	uint8_t link_bw;
 	uint8_t lane_count;
 	uint8_t dpcd[4];
@@ -138,8 +137,8 @@ intel_dp_max_lane_count(struct intel_dp *intel_dp)
 {
 	int max_lane_count = 4;
 
-	if (intel_dp->dpcd[0] >= 0x11) {
-		max_lane_count = intel_dp->dpcd[2] & 0x1f;
+	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;
@@ -153,7 +152,7 @@ intel_dp_max_lane_count(struct intel_dp *intel_dp)
 static int
 intel_dp_max_link_bw(struct intel_dp *intel_dp)
 {
-	int max_link_bw = intel_dp->dpcd[1];
+	int max_link_bw = intel_dp->dpcd[DP_MAX_LINK_RATE];
 
 	switch (max_link_bw) {
 	case DP_LINK_BW_1_62:
@@ -179,12 +178,14 @@ intel_dp_link_clock(uint8_t link_bw)
 static int
 intel_dp_link_required(struct drm_device *dev, struct intel_dp *intel_dp, int pixel_clock)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_crtc *crtc = intel_dp->base.base.crtc;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int bpp = 24;
 
-	if (is_edp(intel_dp))
-		return (pixel_clock * dev_priv->edp.bpp + 7) / 8;
-	else
-		return pixel_clock * 3;
+	if (intel_crtc)
+		bpp = intel_crtc->bpp;
+
+	return (pixel_clock * bpp + 7) / 8;
 }
 
 static int
@@ -682,7 +683,7 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
 	struct drm_encoder *encoder;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int lane_count = 4, bpp = 24;
+	int lane_count = 4;
 	struct intel_dp_m_n m_n;
 	int pipe = intel_crtc->pipe;
 
@@ -701,7 +702,6 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
 			break;
 		} else if (is_edp(intel_dp)) {
 			lane_count = dev_priv->edp.lanes;
-			bpp = dev_priv->edp.bpp;
 			break;
 		}
 	}
@@ -711,7 +711,7 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
 	 * the number of bytes_per_pixel post-LUT, which we always
 	 * set up for 8-bits of R/G/B, or 3 bytes total.
 	 */
-	intel_dp_compute_m_n(bpp, lane_count,
+	intel_dp_compute_m_n(intel_crtc->bpp, lane_count,
 			     mode->clock, adjusted_mode->clock, &m_n);
 
 	if (HAS_PCH_SPLIT(dev)) {
@@ -774,7 +774,8 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
 	/*
 	 * Check for DPCD version > 1.1 and enhanced framing support
 	 */
-	if (intel_dp->dpcd[0] >= 0x11 && (intel_dp->dpcd[2] & DP_ENHANCED_FRAME_CAP)) {
+	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;
 	}
@@ -942,11 +943,44 @@ static void ironlake_edp_pll_off(struct drm_encoder *encoder)
 	udelay(200);
 }
 
+/* If the sink supports it, try to set the power state appropriately */
+static void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode)
+{
+	int ret, i;
+
+	/* Should have a valid DPCD by this point */
+	if (intel_dp->dpcd[DP_DPCD_REV] < 0x11)
+		return;
+
+	if (mode != DRM_MODE_DPMS_ON) {
+		ret = intel_dp_aux_native_write_1(intel_dp, DP_SET_POWER,
+						  DP_SET_POWER_D3);
+		if (ret != 1)
+			DRM_DEBUG_DRIVER("failed to write sink power state\n");
+	} else {
+		/*
+		 * When turning on, we need to retry for 1ms to give the sink
+		 * time to wake up.
+		 */
+		for (i = 0; i < 3; i++) {
+			ret = intel_dp_aux_native_write_1(intel_dp,
+							  DP_SET_POWER,
+							  DP_SET_POWER_D0);
+			if (ret == 1)
+				break;
+			msleep(1);
+		}
+	}
+}
+
 static void intel_dp_prepare(struct drm_encoder *encoder)
 {
 	struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 	struct drm_device *dev = encoder->dev;
 
+	/* Wake up the sink first */
+	intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
+
 	if (is_edp(intel_dp)) {
 		ironlake_edp_backlight_off(dev);
 		ironlake_edp_panel_off(dev);
@@ -990,6 +1024,7 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 	if (mode != DRM_MODE_DPMS_ON) {
 		if (is_edp(intel_dp))
 			ironlake_edp_backlight_off(dev);
+		intel_dp_sink_dpms(intel_dp, mode);
 		intel_dp_link_down(intel_dp);
 		if (is_edp(intel_dp))
 			ironlake_edp_panel_off(dev);
@@ -998,6 +1033,7 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 	} else {
 		if (is_edp(intel_dp))
 			ironlake_edp_panel_vdd_on(intel_dp);
+		intel_dp_sink_dpms(intel_dp, mode);
 		if (!(dp_reg & DP_PORT_EN)) {
 			intel_dp_start_link_train(intel_dp);
 			if (is_edp(intel_dp)) {
@@ -1009,7 +1045,31 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 		if (is_edp(intel_dp))
 			ironlake_edp_backlight_on(dev);
 	}
-	intel_dp->dpms_mode = mode;
+}
+
+/*
+ * Native read with retry for link status and receiver capability reads for
+ * cases where the sink may still be asleep.
+ */
+static bool
+intel_dp_aux_native_read_retry(struct intel_dp *intel_dp, uint16_t address,
+			       uint8_t *recv, int recv_bytes)
+{
+	int ret, i;
+
+	/*
+	 * Sinks are *supposed* to come up within 1ms from an off state,
+	 * but we're also supposed to retry 3 times per the spec.
+	 */
+	for (i = 0; i < 3; i++) {
+		ret = intel_dp_aux_native_read(intel_dp, address, recv,
+					       recv_bytes);
+		if (ret == recv_bytes)
+			return true;
+		msleep(1);
+	}
+
+	return false;
 }
 
 /*
@@ -1019,14 +1079,10 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 static bool
 intel_dp_get_link_status(struct intel_dp *intel_dp)
 {
-	int ret;
-
-	ret = intel_dp_aux_native_read(intel_dp,
-				       DP_LANE0_1_STATUS,
-				       intel_dp->link_status, DP_LINK_STATUS_SIZE);
-	if (ret != DP_LINK_STATUS_SIZE)
-		return false;
-	return true;
+	return intel_dp_aux_native_read_retry(intel_dp,
+					      DP_LANE0_1_STATUS,
+					      intel_dp->link_status,
+					      DP_LINK_STATUS_SIZE);
 }
 
 static uint8_t
@@ -1515,6 +1571,8 @@ intel_dp_link_down(struct intel_dp *intel_dp)
 static void
 intel_dp_check_link_status(struct intel_dp *intel_dp)
 {
+	int ret;
+
 	if (!intel_dp->base.base.crtc)
 		return;
 
@@ -1523,6 +1581,15 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 		return;
 	}
 
+	/* Try to read receiver status if the link appears to be up */
+	ret = intel_dp_aux_native_read(intel_dp,
+				       0x000, intel_dp->dpcd,
+				       sizeof (intel_dp->dpcd));
+	if (ret != sizeof(intel_dp->dpcd)) {
+		intel_dp_link_down(intel_dp);
+		return;
+	}
+
 	if (!intel_channel_eq_ok(intel_dp)) {
 		intel_dp_start_link_train(intel_dp);
 		intel_dp_complete_link_train(intel_dp);
@@ -1533,6 +1600,7 @@ static enum drm_connector_status
 ironlake_dp_detect(struct intel_dp *intel_dp)
 {
 	enum drm_connector_status status;
+	bool ret;
 
 	/* Can't disconnect eDP, but you can close the lid... */
 	if (is_edp(intel_dp)) {
@@ -1543,13 +1611,11 @@ ironlake_dp_detect(struct intel_dp *intel_dp)
 	}
 
 	status = connector_status_disconnected;
-	if (intel_dp_aux_native_read(intel_dp,
-				     0x000, intel_dp->dpcd,
-				     sizeof (intel_dp->dpcd))
-	    == sizeof(intel_dp->dpcd)) {
-		if (intel_dp->dpcd[0] != 0)
-			status = connector_status_connected;
-	}
+	ret = intel_dp_aux_native_read_retry(intel_dp,
+					     0x000, intel_dp->dpcd,
+					     sizeof (intel_dp->dpcd));
+	if (ret && intel_dp->dpcd[DP_DPCD_REV] != 0)
+		status = connector_status_connected;
 	DRM_DEBUG_KMS("DPCD: %hx%hx%hx%hx\n", intel_dp->dpcd[0],
 		      intel_dp->dpcd[1], intel_dp->dpcd[2], intel_dp->dpcd[3]);
 	return status;
@@ -1586,7 +1652,7 @@ g4x_dp_detect(struct intel_dp *intel_dp)
 	if (intel_dp_aux_native_read(intel_dp, 0x000, intel_dp->dpcd,
 				     sizeof (intel_dp->dpcd)) == sizeof (intel_dp->dpcd))
 	{
-		if (intel_dp->dpcd[0] != 0)
+		if (intel_dp->dpcd[DP_DPCD_REV] != 0)
 			status = connector_status_connected;
 	}
 
@@ -1790,8 +1856,7 @@ intel_dp_hot_plug(struct intel_encoder *intel_encoder)
 {
 	struct intel_dp *intel_dp = container_of(intel_encoder, struct intel_dp, base);
 
-	if (intel_dp->dpms_mode == DRM_MODE_DPMS_ON)
-		intel_dp_check_link_status(intel_dp);
+	intel_dp_check_link_status(intel_dp);
 }
 
 /* Return which DP Port should be selected for Transcoder DP control */
@@ -1859,7 +1924,6 @@ intel_dp_init(struct drm_device *dev, int output_reg)
 		return;
 
 	intel_dp->output_reg = output_reg;
-	intel_dp->dpms_mode = -1;
 
 	intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL);
 	if (!intel_connector) {
@@ -1954,8 +2018,9 @@ intel_dp_init(struct drm_device *dev, int output_reg)
 					       sizeof(intel_dp->dpcd));
 		ironlake_edp_panel_vdd_off(intel_dp);
 		if (ret == sizeof(intel_dp->dpcd)) {
-			if (intel_dp->dpcd[0] >= 0x11)
-				dev_priv->no_aux_handshake = intel_dp->dpcd[3] &
+			if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11)
+				dev_priv->no_aux_handshake =
+					intel_dp->dpcd[DP_MAX_DOWNSPREAD] &
 					DP_NO_AUX_HANDSHAKE_LINK_TRAINING;
 		} else {
 			/* if this fails, presume the device is a ghost */
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 9ffa61eb4d7..6e990f9760e 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -170,6 +170,7 @@ struct intel_crtc {
 	int16_t cursor_x, cursor_y;
 	int16_t cursor_width, cursor_height;
 	bool cursor_visible;
+	unsigned int bpp;
 };
 
 #define to_intel_crtc(x) container_of(x, struct intel_crtc, base)
@@ -233,6 +234,13 @@ struct intel_unpin_work {
 	bool enable_stall_check;
 };
 
+struct intel_fbc_work {
+	struct delayed_work work;
+	struct drm_crtc *crtc;
+	struct drm_framebuffer *fb;
+	int interval;
+};
+
 int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter);
 extern bool intel_ddc_probe(struct intel_encoder *intel_encoder, int ddc_bus);
 
@@ -317,6 +325,7 @@ extern void intel_enable_clock_gating(struct drm_device *dev);
 extern void ironlake_enable_drps(struct drm_device *dev);
 extern void ironlake_disable_drps(struct drm_device *dev);
 extern void gen6_enable_rps(struct drm_i915_private *dev_priv);
+extern void gen6_update_ring_freq(struct drm_i915_private *dev_priv);
 extern void gen6_disable_rps(struct drm_device *dev);
 extern void intel_init_emon(struct drm_device *dev);
 
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index aa0a8e83142..1ed8e690391 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -124,12 +124,18 @@ static void intel_hdmi_mode_set(struct drm_encoder *encoder,
 	u32 sdvox;
 
 	sdvox = SDVO_ENCODING_HDMI | SDVO_BORDER_ENABLE;
-	sdvox |= intel_hdmi->color_range;
+	if (!HAS_PCH_SPLIT(dev))
+		sdvox |= intel_hdmi->color_range;
 	if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
 		sdvox |= SDVO_VSYNC_ACTIVE_HIGH;
 	if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
 		sdvox |= SDVO_HSYNC_ACTIVE_HIGH;
 
+	if (intel_crtc->bpp > 24)
+		sdvox |= COLOR_FORMAT_12bpc;
+	else
+		sdvox |= COLOR_FORMAT_8bpc;
+
 	/* Required on CPT */
 	if (intel_hdmi->has_hdmi_sink && HAS_PCH_CPT(dev))
 		sdvox |= HDMI_MODE_SELECT;
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index d2c71042290..b7c5ddb564d 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -297,19 +297,26 @@ static int intel_opregion_video_event(struct notifier_block *nb,
 	/* The only video events relevant to opregion are 0x80. These indicate
 	   either a docking event, lid switch or display switch request. In
 	   Linux, these are handled by the dock, button and video drivers.
-	   We might want to fix the video driver to be opregion-aware in
-	   future, but right now we just indicate to the firmware that the
-	   request has been handled */
+	*/
 
 	struct opregion_acpi *acpi;
+	struct acpi_bus_event *event = data;
+	int ret = NOTIFY_OK;
+
+	if (strcmp(event->device_class, ACPI_VIDEO_CLASS) != 0)
+		return NOTIFY_DONE;
 
 	if (!system_opregion)
 		return NOTIFY_DONE;
 
 	acpi = system_opregion->acpi;
+
+	if (event->type == 0x80 && !(acpi->cevt & 0x1))
+		ret = NOTIFY_BAD;
+
 	acpi->csts = 0;
 
-	return NOTIFY_OK;
+	return ret;
 }
 
 static struct notifier_block intel_opregion_notifier = {
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index 9e2959bc91c..d3603808682 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -773,14 +773,10 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay,
 	if (ret != 0)
 		return ret;
 
-	ret = i915_gem_object_pin(new_bo, PAGE_SIZE, true);
+	ret = i915_gem_object_pin_to_display_plane(new_bo, 0, NULL);
 	if (ret != 0)
 		return ret;
 
-	ret = i915_gem_object_set_to_gtt_domain(new_bo, 0);
-	if (ret != 0)
-		goto out_unpin;
-
 	ret = i915_gem_object_put_fence(new_bo);
 	if (ret)
 		goto out_unpin;
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 95c4b142993..e9615685a39 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -236,7 +236,8 @@ init_pipe_control(struct intel_ring_buffer *ring)
 		ret = -ENOMEM;
 		goto err;
 	}
-	obj->cache_level = I915_CACHE_LLC;
+
+	i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
 
 	ret = i915_gem_object_pin(obj, 4096, true);
 	if (ret)
@@ -776,7 +777,8 @@ static int init_status_page(struct intel_ring_buffer *ring)
 		ret = -ENOMEM;
 		goto err;
 	}
-	obj->cache_level = I915_CACHE_LLC;
+
+	i915_gem_object_set_cache_level(obj, I915_CACHE_LLC);
 
 	ret = i915_gem_object_pin(obj, 4096, true);
 	if (ret != 0) {
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index c0e0ee63fbf..39ac2b634ae 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -165,7 +165,7 @@ void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring);
 int __must_check intel_wait_ring_buffer(struct intel_ring_buffer *ring, int n);
 static inline int intel_wait_ring_idle(struct intel_ring_buffer *ring)
 {
-	return intel_wait_ring_buffer(ring, ring->space - 8);
+	return intel_wait_ring_buffer(ring, ring->size - 8);
 }
 
 int __must_check intel_ring_begin(struct intel_ring_buffer *ring, int n);
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index 113e4e7264c..210d570fd51 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1236,6 +1236,8 @@ intel_tv_detect_type (struct intel_tv *intel_tv,
 		      struct drm_connector *connector)
 {
 	struct drm_encoder *encoder = &intel_tv->base.base;
+	struct drm_crtc *crtc = encoder->crtc;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
@@ -1258,6 +1260,10 @@ intel_tv_detect_type (struct intel_tv *intel_tv,
 	/* Poll for TV detection */
 	tv_ctl &= ~(TV_ENC_ENABLE | TV_TEST_MODE_MASK);
 	tv_ctl |= TV_TEST_MODE_MONITOR_DETECT;
+	if (intel_crtc->pipe == 1)
+		tv_ctl |= TV_ENC_PIPEB_SELECT;
+	else
+		tv_ctl &= ~TV_ENC_PIPEB_SELECT;
 
 	tv_dac &= ~(TVDAC_SENSE_MASK | DAC_A_MASK | DAC_B_MASK | DAC_C_MASK);
 	tv_dac |= (TVDAC_STATE_CHG_EN |
@@ -1277,26 +1283,26 @@ intel_tv_detect_type (struct intel_tv *intel_tv,
 			      to_intel_crtc(intel_tv->base.base.crtc)->pipe);
 
 	type = -1;
-	if (wait_for((tv_dac = I915_READ(TV_DAC)) & TVDAC_STATE_CHG, 20) == 0) {
-		DRM_DEBUG_KMS("TV detected: %x, %x\n", tv_ctl, tv_dac);
-		/*
-		 *  A B C
-		 *  0 1 1 Composite
-		 *  1 0 X svideo
-		 *  0 0 0 Component
-		 */
-		if ((tv_dac & TVDAC_SENSE_MASK) == (TVDAC_B_SENSE | TVDAC_C_SENSE)) {
-			DRM_DEBUG_KMS("Detected Composite TV connection\n");
-			type = DRM_MODE_CONNECTOR_Composite;
-		} else if ((tv_dac & (TVDAC_A_SENSE|TVDAC_B_SENSE)) == TVDAC_A_SENSE) {
-			DRM_DEBUG_KMS("Detected S-Video TV connection\n");
-			type = DRM_MODE_CONNECTOR_SVIDEO;
-		} else if ((tv_dac & TVDAC_SENSE_MASK) == 0) {
-			DRM_DEBUG_KMS("Detected Component TV connection\n");
-			type = DRM_MODE_CONNECTOR_Component;
-		} else {
-			DRM_DEBUG_KMS("Unrecognised TV connection\n");
-		}
+	tv_dac = I915_READ(TV_DAC);
+	DRM_DEBUG_KMS("TV detected: %x, %x\n", tv_ctl, tv_dac);
+	/*
+	 *  A B C
+	 *  0 1 1 Composite
+	 *  1 0 X svideo
+	 *  0 0 0 Component
+	 */
+	if ((tv_dac & TVDAC_SENSE_MASK) == (TVDAC_B_SENSE | TVDAC_C_SENSE)) {
+		DRM_DEBUG_KMS("Detected Composite TV connection\n");
+		type = DRM_MODE_CONNECTOR_Composite;
+	} else if ((tv_dac & (TVDAC_A_SENSE|TVDAC_B_SENSE)) == TVDAC_A_SENSE) {
+		DRM_DEBUG_KMS("Detected S-Video TV connection\n");
+		type = DRM_MODE_CONNECTOR_SVIDEO;
+	} else if ((tv_dac & TVDAC_SENSE_MASK) == 0) {
+		DRM_DEBUG_KMS("Detected Component TV connection\n");
+		type = DRM_MODE_CONNECTOR_Component;
+	} else {
+		DRM_DEBUG_KMS("Unrecognised TV connection\n");
+		type = -1;
 	}
 
 	I915_WRITE(TV_DAC, save_tv_dac & ~TVDAC_STATE_CHG_EN);