Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: (42 commits)
  drm/radeon/kms/pm: switch to dynamically allocating clock mode array
  drm/radeon/kms: optimize r600_pm_profile_init
  drm/radeon/kms/pm: add a proper pm profile init function for fusion
  drm/radeon/kms: remove extraneous calls to radeon_pm_compute_clocks()
  drm/exynos: added padding to be 64-bit align.
  drm: fix kconfig unmet dependency warning
  drm: add some comments to drm_wait_vblank and drm_queue_vblank_event
  drm/radeon/benchmark: signedness bug in radeon_benchmark_move()
  drm: do not sleep on vblank while holding a mutex
  MAINTAINERS: exynos: Add EXYNOS DRM maintainer entry
  drm: try to restore previous CRTC config if mode set fails
  drm/radeon/kms: make an aux failure debug only
  drm: drop select of SLOW_WORK
  drm: serialize access to list of debugfs files
  drm/radeon/kms: fix use of vram scratch page on evergreen/ni
  drm/radeon: Make sure CS mutex is held across GPU reset.
  drm: Ensure string is null terminated.
  vmwgfx: Only allow 64x64 cursors
  vmwgfx: Initialize clip rect loop correctly in surface dirty
  vmwgfx: Close screen object system
  ...

42 files changed, 441 insertions, 234 deletions

diff --git a/MAINTAINERS b/MAINTAINERS
index 4808256446f..c1b1f3e40c3 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2342,6 +2342,13 @@ S:	Supported
 F:	drivers/gpu/drm/i915
 F:	include/drm/i915*
 
+DRM DRIVERS FOR EXYNOS
+M:	Inki Dae <inki.dae@samsung.com>
+L:	dri-devel@lists.freedesktop.org
+S:	Supported
+F:	drivers/gpu/drm/exynos
+F:	include/drm/exynos*
+
 DSCC4 DRIVER
 M:	Francois Romieu <romieu@fr.zoreil.com>
 L:	netdev@vger.kernel.org
diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c
index 66cd0b8096c..c92424ca1a5 100644
--- a/drivers/char/agp/intel-gtt.c
+++ b/drivers/char/agp/intel-gtt.c
@@ -1186,10 +1186,11 @@ static void gen6_cleanup(void)
 /* Certain Gen5 chipsets require require idling the GPU before
  * unmapping anything from the GTT when VT-d is enabled.
  */
-extern int intel_iommu_gfx_mapped;
 static inline int needs_idle_maps(void)
 {
+#ifdef CONFIG_INTEL_IOMMU
 	const unsigned short gpu_devid = intel_private.pcidev->device;
+	extern int intel_iommu_gfx_mapped;
 
 	/* Query intel_iommu to see if we need the workaround. Presumably that
 	 * was loaded first.
@@ -1198,7 +1199,7 @@ static inline int needs_idle_maps(void)
 	     gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_M_IG) &&
 	     intel_iommu_gfx_mapped)
 		return 1;
-
+#endif
 	return 0;
 }
 
@@ -1236,7 +1237,7 @@ static int i9xx_setup(void)
 		intel_private.gtt_bus_addr = reg_addr + gtt_offset;
 	}
 
-	if (needs_idle_maps());
+	if (needs_idle_maps())
 		intel_private.base.do_idle_maps = 1;
 
 	intel_i9xx_setup_flush();
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 785127cb281..1368826ef28 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -9,7 +9,6 @@ menuconfig DRM
 	depends on (AGP || AGP=n) && !EMULATED_CMPXCHG && MMU
 	select I2C
 	select I2C_ALGOBIT
-	select SLOW_WORK
 	help
 	  Kernel-level support for the Direct Rendering Infrastructure (DRI)
 	  introduced in XFree86 4.0. If you say Y here, you need to select
@@ -96,6 +95,7 @@ config DRM_I915
 	select FB_CFB_IMAGEBLIT
 	# i915 depends on ACPI_VIDEO when ACPI is enabled
 	# but for select to work, need to select ACPI_VIDEO's dependencies, ick
+	select BACKLIGHT_LCD_SUPPORT if ACPI
 	select BACKLIGHT_CLASS_DEVICE if ACPI
 	select VIDEO_OUTPUT_CONTROL if ACPI
 	select INPUT if ACPI
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 9a2e2a14b3b..405c63b9d53 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -2118,8 +2118,10 @@ struct drm_property *drm_property_create(struct drm_device *dev, int flags,
 	property->num_values = num_values;
 	INIT_LIST_HEAD(&property->enum_blob_list);
 
-	if (name)
+	if (name) {
 		strncpy(property->name, name, DRM_PROP_NAME_LEN);
+		property->name[DRM_PROP_NAME_LEN-1] = '\0';
+	}
 
 	list_add_tail(&property->head, &dev->mode_config.property_list);
 	return property;
diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index 2957636161e..3969f7553fe 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -484,6 +484,7 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 	struct drm_connector *save_connectors, *connector;
 	int count = 0, ro, fail = 0;
 	struct drm_crtc_helper_funcs *crtc_funcs;
+	struct drm_mode_set save_set;
 	int ret = 0;
 	int i;
 
@@ -556,6 +557,12 @@ int drm_crtc_helper_set_config(struct drm_mode_set *set)
 		save_connectors[count++] = *connector;
 	}
 
+	save_set.crtc = set->crtc;
+	save_set.mode = &set->crtc->mode;
+	save_set.x = set->crtc->x;
+	save_set.y = set->crtc->y;
+	save_set.fb = set->crtc->fb;
+
 	/* We should be able to check here if the fb has the same properties
 	 * and then just flip_or_move it */
 	if (set->crtc->fb != set->fb) {
@@ -721,6 +728,12 @@ fail:
 		*connector = save_connectors[count++];
 	}
 
+	/* Try to restore the config */
+	if (mode_changed &&
+	    !drm_crtc_helper_set_mode(save_set.crtc, save_set.mode, save_set.x,
+				      save_set.y, save_set.fb))
+		DRM_ERROR("failed to restore config after modeset failure\n");
+
 	kfree(save_connectors);
 	kfree(save_encoders);
 	kfree(save_crtcs);
diff --git a/drivers/gpu/drm/drm_debugfs.c b/drivers/gpu/drm/drm_debugfs.c
index d067c12ba94..1c7a1c0d3ed 100644
--- a/drivers/gpu/drm/drm_debugfs.c
+++ b/drivers/gpu/drm/drm_debugfs.c
@@ -118,7 +118,10 @@ int drm_debugfs_create_files(struct drm_info_list *files, int count,
 		tmp->minor = minor;
 		tmp->dent = ent;
 		tmp->info_ent = &files[i];
-		list_add(&(tmp->list), &(minor->debugfs_nodes.list));
+
+		mutex_lock(&minor->debugfs_lock);
+		list_add(&tmp->list, &minor->debugfs_list);
+		mutex_unlock(&minor->debugfs_lock);
 	}
 	return 0;
 
@@ -146,7 +149,8 @@ int drm_debugfs_init(struct drm_minor *minor, int minor_id,
 	char name[64];
 	int ret;
 
-	INIT_LIST_HEAD(&minor->debugfs_nodes.list);
+	INIT_LIST_HEAD(&minor->debugfs_list);
+	mutex_init(&minor->debugfs_lock);
 	sprintf(name, "%d", minor_id);
 	minor->debugfs_root = debugfs_create_dir(name, root);
 	if (!minor->debugfs_root) {
@@ -192,8 +196,9 @@ int drm_debugfs_remove_files(struct drm_info_list *files, int count,
 	struct drm_info_node *tmp;
 	int i;
 
+	mutex_lock(&minor->debugfs_lock);
 	for (i = 0; i < count; i++) {
-		list_for_each_safe(pos, q, &minor->debugfs_nodes.list) {
+		list_for_each_safe(pos, q, &minor->debugfs_list) {
 			tmp = list_entry(pos, struct drm_info_node, list);
 			if (tmp->info_ent == &files[i]) {
 				debugfs_remove(tmp->dent);
@@ -202,6 +207,7 @@ int drm_debugfs_remove_files(struct drm_info_list *files, int count,
 			}
 		}
 	}
+	mutex_unlock(&minor->debugfs_lock);
 	return 0;
 }
 EXPORT_SYMBOL(drm_debugfs_remove_files);
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index fc81af9dbf4..40c187c60f4 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -125,7 +125,7 @@ static struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_SG_ALLOC, drm_sg_alloc_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF(DRM_IOCTL_SG_FREE, drm_sg_free, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 
-	DRM_IOCTL_DEF(DRM_IOCTL_WAIT_VBLANK, drm_wait_vblank, 0),
+	DRM_IOCTL_DEF(DRM_IOCTL_WAIT_VBLANK, drm_wait_vblank, DRM_UNLOCKED),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_MODESET_CTL, drm_modeset_ctl, 0),
 
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index cb3794a00f9..68b756253f9 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -407,13 +407,16 @@ int drm_irq_uninstall(struct drm_device *dev)
 	/*
 	 * Wake up any waiters so they don't hang.
 	 */
-	spin_lock_irqsave(&dev->vbl_lock, irqflags);
-	for (i = 0; i < dev->num_crtcs; i++) {
-		DRM_WAKEUP(&dev->vbl_queue[i]);
-		dev->vblank_enabled[i] = 0;
-		dev->last_vblank[i] = dev->driver->get_vblank_counter(dev, i);
+	if (dev->num_crtcs) {
+		spin_lock_irqsave(&dev->vbl_lock, irqflags);
+		for (i = 0; i < dev->num_crtcs; i++) {
+			DRM_WAKEUP(&dev->vbl_queue[i]);
+			dev->vblank_enabled[i] = 0;
+			dev->last_vblank[i] =
+				dev->driver->get_vblank_counter(dev, i);
+		}
+		spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 	}
-	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 
 	if (!irq_enabled)
 		return -EINVAL;
@@ -1125,6 +1128,7 @@ static int drm_queue_vblank_event(struct drm_device *dev, int pipe,
 		trace_drm_vblank_event_delivered(current->pid, pipe,
 						 vblwait->request.sequence);
 	} else {
+		/* drm_handle_vblank_events will call drm_vblank_put */
 		list_add_tail(&e->base.link, &dev->vblank_event_list);
 		vblwait->reply.sequence = vblwait->request.sequence;
 	}
@@ -1205,8 +1209,12 @@ int drm_wait_vblank(struct drm_device *dev, void *data,
 		goto done;
 	}
 
-	if (flags & _DRM_VBLANK_EVENT)
+	if (flags & _DRM_VBLANK_EVENT) {
+		/* must hold on to the vblank ref until the event fires
+		 * drm_vblank_put will be called asynchronously
+		 */
 		return drm_queue_vblank_event(dev, crtc, vblwait, file_priv);
+	}
 
 	if ((flags & _DRM_VBLANK_NEXTONMISS) &&
 	    (seq - vblwait->request.sequence) <= (1<<23)) {
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index d14b44e13f5..4f40f1ce1d8 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -1506,7 +1506,10 @@ drm_add_fake_info_node(struct drm_minor *minor,
 	node->minor = minor;
 	node->dent = ent;
 	node->info_ent = (void *) key;
-	list_add(&node->list, &minor->debugfs_nodes.list);
+
+	mutex_lock(&minor->debugfs_lock);
+	list_add(&node->list, &minor->debugfs_list);
+	mutex_unlock(&minor->debugfs_lock);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 6651c36b6e8..d18b07adcff 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1396,7 +1396,7 @@ i915_gem_mmap_gtt(struct drm_file *file,
 
 	if (obj->base.size > dev_priv->mm.gtt_mappable_end) {
 		ret = -E2BIG;
-		goto unlock;
+		goto out;
 	}
 
 	if (obj->madv != I915_MADV_WILLNEED) {
diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.c b/drivers/gpu/drm/nouveau/nouveau_bios.c
index 032a8209813..5fc201b49d3 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bios.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bios.c
@@ -640,10 +640,9 @@ static int
 nv50_pll_set(struct drm_device *dev, uint32_t reg, uint32_t clk)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
-	uint32_t reg0 = nv_rd32(dev, reg + 0);
-	uint32_t reg1 = nv_rd32(dev, reg + 4);
 	struct nouveau_pll_vals pll;
 	struct pll_lims pll_limits;
+	u32 ctrl, mask, coef;
 	int ret;
 
 	ret = get_pll_limits(dev, reg, &pll_limits);
@@ -654,15 +653,20 @@ nv50_pll_set(struct drm_device *dev, uint32_t reg, uint32_t clk)
 	if (!clk)
 		return -ERANGE;
 
-	reg0 = (reg0 & 0xfff8ffff) | (pll.log2P << 16);
-	reg1 = (reg1 & 0xffff0000) | (pll.N1 << 8) | pll.M1;
-
-	if (dev_priv->vbios.execute) {
-		still_alive();
-		nv_wr32(dev, reg + 4, reg1);
-		nv_wr32(dev, reg + 0, reg0);
+	coef = pll.N1 << 8 | pll.M1;
+	ctrl = pll.log2P << 16;
+	mask = 0x00070000;
+	if (reg == 0x004008) {
+		mask |= 0x01f80000;
+		ctrl |= (pll_limits.log2p_bias << 19);
+		ctrl |= (pll.log2P << 22);
 	}
 
+	if (!dev_priv->vbios.execute)
+		return 0;
+
+	nv_mask(dev, reg + 0, mask, ctrl);
+	nv_wr32(dev, reg + 4, coef);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 7226f419e17..7cc37e69086 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -148,7 +148,7 @@ set_placement_range(struct nouveau_bo *nvbo, uint32_t type)
 
 	if (dev_priv->card_type == NV_10 &&
 	    nvbo->tile_mode && (type & TTM_PL_FLAG_VRAM) &&
-	    nvbo->bo.mem.num_pages < vram_pages / 2) {
+	    nvbo->bo.mem.num_pages < vram_pages / 4) {
 		/*
 		 * Make sure that the color and depth buffers are handled
 		 * by independent memory controller units. Up to a 9x
diff --git a/drivers/gpu/drm/nouveau/nouveau_channel.c b/drivers/gpu/drm/nouveau/nouveau_channel.c
index a319d5646ea..bb6ec9ef867 100644
--- a/drivers/gpu/drm/nouveau/nouveau_channel.c
+++ b/drivers/gpu/drm/nouveau/nouveau_channel.c
@@ -158,6 +158,7 @@ nouveau_channel_alloc(struct drm_device *dev, struct nouveau_channel **chan_ret,
 	INIT_LIST_HEAD(&chan->nvsw.vbl_wait);
 	INIT_LIST_HEAD(&chan->nvsw.flip);
 	INIT_LIST_HEAD(&chan->fence.pending);
+	spin_lock_init(&chan->fence.lock);
 
 	/* setup channel's memory and vm */
 	ret = nouveau_gpuobj_channel_init(chan, vram_handle, gart_handle);
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index e0d275e1c96..cea6696b190 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -710,7 +710,7 @@ nouveau_connector_mode_valid(struct drm_connector *connector,
 	case OUTPUT_DP:
 		max_clock  = nv_encoder->dp.link_nr;
 		max_clock *= nv_encoder->dp.link_bw;
-		clock = clock * nouveau_connector_bpp(connector) / 8;
+		clock = clock * nouveau_connector_bpp(connector) / 10;
 		break;
 	default:
 		BUG_ON(1);
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index 14a8627efe4..3a4cc32b9e4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -487,6 +487,7 @@ int nouveau_fbcon_init(struct drm_device *dev)
 {
 	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	struct nouveau_fbdev *nfbdev;
+	int preferred_bpp;
 	int ret;
 
 	nfbdev = kzalloc(sizeof(struct nouveau_fbdev), GFP_KERNEL);
@@ -505,7 +506,15 @@ int nouveau_fbcon_init(struct drm_device *dev)
 	}
 
 	drm_fb_helper_single_add_all_connectors(&nfbdev->helper);
-	drm_fb_helper_initial_config(&nfbdev->helper, 32);
+
+	if (dev_priv->vram_size <= 32 * 1024 * 1024)
+		preferred_bpp = 8;
+	else if (dev_priv->vram_size <= 64 * 1024 * 1024)
+		preferred_bpp = 16;
+	else
+		preferred_bpp = 32;
+
+	drm_fb_helper_initial_config(&nfbdev->helper, preferred_bpp);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c
index 81116cfea27..2f6daae68b9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fence.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fence.c
@@ -539,8 +539,6 @@ nouveau_fence_channel_init(struct nouveau_channel *chan)
 			return ret;
 	}
 
-	INIT_LIST_HEAD(&chan->fence.pending);
-	spin_lock_init(&chan->fence.lock);
 	atomic_set(&chan->fence.last_sequence_irq, 0);
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_i2c.c b/drivers/gpu/drm/nouveau/nouveau_i2c.c
index c6143df48b9..d39b2202b19 100644
--- a/drivers/gpu/drm/nouveau/nouveau_i2c.c
+++ b/drivers/gpu/drm/nouveau/nouveau_i2c.c
@@ -333,7 +333,7 @@ nouveau_i2c_identify(struct drm_device *dev, const char *what,
 
 	NV_DEBUG(dev, "Probing %ss on I2C bus: %d\n", what, index);
 
-	for (i = 0; info[i].addr; i++) {
+	for (i = 0; i2c && info[i].addr; i++) {
 		if (nouveau_probe_i2c_addr(i2c, info[i].addr) &&
 		    (!match || match(i2c, &info[i]))) {
 			NV_INFO(dev, "Detected %s: %s\n", what, info[i].type);
diff --git a/drivers/gpu/drm/nouveau/nouveau_perf.c b/drivers/gpu/drm/nouveau/nouveau_perf.c
index 9f178aa9416..33d03fbf00d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_perf.c
+++ b/drivers/gpu/drm/nouveau/nouveau_perf.c
@@ -239,7 +239,7 @@ nouveau_perf_init(struct drm_device *dev)
 	if(version == 0x15) {
 		memtimings->timing =
 				kcalloc(entries, sizeof(*memtimings->timing), GFP_KERNEL);
-		if(!memtimings) {
+		if (!memtimings->timing) {
 			NV_WARN(dev,"Could not allocate memtiming table\n");
 			return;
 		}
diff --git a/drivers/gpu/drm/nouveau/nouveau_state.c b/drivers/gpu/drm/nouveau/nouveau_state.c
index 82478e0998e..d8831ab42bb 100644
--- a/drivers/gpu/drm/nouveau/nouveau_state.c
+++ b/drivers/gpu/drm/nouveau/nouveau_state.c
@@ -579,6 +579,14 @@ nouveau_card_init(struct drm_device *dev)
 	if (ret)
 		goto out_display_early;
 
+	/* workaround an odd issue on nvc1 by disabling the device's
+	 * nosnoop capability.  hopefully won't cause issues until a
+	 * better fix is found - assuming there is one...
+	 */
+	if (dev_priv->chipset == 0xc1) {
+		nv_mask(dev, 0x00088080, 0x00000800, 0x00000000);
+	}
+
 	nouveau_pm_init(dev);
 
 	ret = engine->vram.init(dev);
@@ -1102,12 +1110,13 @@ int nouveau_load(struct drm_device *dev, unsigned long flags)
 	dev_priv->noaccel = !!nouveau_noaccel;
 	if (nouveau_noaccel == -1) {
 		switch (dev_priv->chipset) {
-		case 0xc1: /* known broken */
-		case 0xc8: /* never tested */
+#if 0
+		case 0xXX: /* known broken */
 			NV_INFO(dev, "acceleration disabled by default, pass "
 				     "noaccel=0 to force enable\n");
 			dev_priv->noaccel = true;
 			break;
+#endif
 		default:
 			dev_priv->noaccel = false;
 			break;
diff --git a/drivers/gpu/drm/nouveau/nv40_pm.c b/drivers/gpu/drm/nouveau/nv40_pm.c
index bbc0b9c7e1f..e676b0d5347 100644
--- a/drivers/gpu/drm/nouveau/nv40_pm.c
+++ b/drivers/gpu/drm/nouveau/nv40_pm.c
@@ -57,12 +57,14 @@ read_pll_2(struct drm_device *dev, u32 reg)
 	int P = (ctrl & 0x00070000) >> 16;
 	u32 ref = 27000, clk = 0;
 
-	if (ctrl & 0x80000000)
+	if ((ctrl & 0x80000000) && M1) {
 		clk = ref * N1 / M1;
-
-	if (!(ctrl & 0x00000100)) {
-		if (ctrl & 0x40000000)
-			clk = clk * N2 / M2;
+		if ((ctrl & 0x40000100) == 0x40000000) {
+			if (M2)
+				clk = clk * N2 / M2;
+			else
+				clk = 0;
+		}
 	}
 
 	return clk >> P;
@@ -177,6 +179,11 @@ nv40_pm_clocks_pre(struct drm_device *dev, struct nouveau_pm_level *perflvl)
 	}
 
 	/* memory clock */
+	if (!perflvl->memory) {
+		info->mpll_ctrl = 0x00000000;
+		goto out;
+	}
+
 	ret = nv40_calc_pll(dev, 0x004020, &pll, perflvl->memory,
 			    &N1, &M1, &N2, &M2, &log2P);
 	if (ret < 0)
@@ -264,6 +271,9 @@ nv40_pm_clocks_set(struct drm_device *dev, void *pre_state)
 	mdelay(5);
 	nv_mask(dev, 0x00c040, 0x00000333, info->ctrl);
 
+	if (!info->mpll_ctrl)
+		goto resume;
+
 	/* wait for vblank start on active crtcs, disable memory access */
 	for (i = 0; i < 2; i++) {
 		if (!(crtc_mask & (1 << i)))
diff --git a/drivers/gpu/drm/nouveau/nv50_graph.c b/drivers/gpu/drm/nouveau/nv50_graph.c
index 8c979b31ff6..ac601f7c4e1 100644
--- a/drivers/gpu/drm/nouveau/nv50_graph.c
+++ b/drivers/gpu/drm/nouveau/nv50_graph.c
@@ -131,8 +131,8 @@ nv50_graph_init(struct drm_device *dev, int engine)
 	NV_DEBUG(dev, "\n");
 
 	/* master reset */
-	nv_mask(dev, 0x000200, 0x00200100, 0x00000000);
-	nv_mask(dev, 0x000200, 0x00200100, 0x00200100);
+	nv_mask(dev, 0x000200, 0x00201000, 0x00000000);
+	nv_mask(dev, 0x000200, 0x00201000, 0x00201000);
 	nv_wr32(dev, 0x40008c, 0x00000004); /* HW_CTX_SWITCH_ENABLED */
 
 	/* reset/enable traps and interrupts */
diff --git a/drivers/gpu/drm/nouveau/nv50_grctx.c b/drivers/gpu/drm/nouveau/nv50_grctx.c
index d05c2c3b244..4b46d696856 100644
--- a/drivers/gpu/drm/nouveau/nv50_grctx.c
+++ b/drivers/gpu/drm/nouveau/nv50_grctx.c
@@ -601,7 +601,7 @@ nv50_graph_construct_mmio(struct nouveau_grctx *ctx)
 					gr_def(ctx, offset + 0x1c, 0x00880000);
 					break;
 				case 0x86:
-					gr_def(ctx, offset + 0x1c, 0x008c0000);
+					gr_def(ctx, offset + 0x1c, 0x018c0000);
 					break;
 				case 0x92:
 				case 0x96:
diff --git a/drivers/gpu/drm/nouveau/nv50_vram.c b/drivers/gpu/drm/nouveau/nv50_vram.c
index 9da23838e63..2e45e57fd86 100644
--- a/drivers/gpu/drm/nouveau/nv50_vram.c
+++ b/drivers/gpu/drm/nouveau/nv50_vram.c
@@ -160,7 +160,7 @@ nv50_vram_rblock(struct drm_device *dev)
 	colbits  =  (r4 & 0x0000f000) >> 12;
 	rowbitsa = ((r4 & 0x000f0000) >> 16) + 8;
 	rowbitsb = ((r4 & 0x00f00000) >> 20) + 8;
-	banks    = ((r4 & 0x01000000) ? 8 : 4);
+	banks    = 1 << (((r4 & 0x03000000) >> 24) + 2);
 
 	rowsize = parts * banks * (1 << colbits) * 8;
 	predicted = rowsize << rowbitsa;
diff --git a/drivers/gpu/drm/nouveau/nvc0_graph.c b/drivers/gpu/drm/nouveau/nvc0_graph.c
index bbdbc51830c..a74e501afd2 100644
--- a/drivers/gpu/drm/nouveau/nvc0_graph.c
+++ b/drivers/gpu/drm/nouveau/nvc0_graph.c
@@ -157,8 +157,8 @@ nvc0_graph_create_context_mmio_list(struct nouveau_channel *chan)
 	struct nvc0_graph_priv *priv = nv_engine(chan->dev, NVOBJ_ENGINE_GR);
 	struct nvc0_graph_chan *grch = chan->engctx[NVOBJ_ENGINE_GR];
 	struct drm_device *dev = chan->dev;
+	struct drm_nouveau_private *dev_priv = dev->dev_private;
 	int i = 0, gpc, tp, ret;
-	u32 magic;
 
 	ret = nouveau_gpuobj_new(dev, chan, 0x2000, 256, NVOBJ_FLAG_VM,
 				 &grch->unk408004);
@@ -207,14 +207,37 @@ nvc0_graph_create_context_mmio_list(struct nouveau_channel *chan)
 	nv_wo32(grch->mmio, i++ * 4, 0x0041880c);
 	nv_wo32(grch->mmio, i++ * 4, 0x80000018);
 
-	magic = 0x02180000;
-	nv_wo32(grch->mmio, i++ * 4, 0x00405830);
-	nv_wo32(grch->mmio, i++ * 4, magic);
-	for (gpc = 0; gpc < priv->gpc_nr; gpc++) {
-		for (tp = 0; tp < priv->tp_nr[gpc]; tp++, magic += 0x0324) {
-			u32 reg = 0x504520 + (gpc * 0x8000) + (tp * 0x0800);
-			nv_wo32(grch->mmio, i++ * 4, reg);
-			nv_wo32(grch->mmio, i++ * 4, magic);
+	if (dev_priv->chipset != 0xc1) {
+		u32 magic = 0x02180000;
+		nv_wo32(grch->mmio, i++ * 4, 0x00405830);
+		nv_wo32(grch->mmio, i++ * 4, magic);
+		for (gpc = 0; gpc < priv->gpc_nr; gpc++) {
+			for (tp = 0; tp < priv->tp_nr[gpc]; tp++) {
+				u32 reg = TP_UNIT(gpc, tp, 0x520);
+				nv_wo32(grch->mmio, i++ * 4, reg);
+				nv_wo32(grch->mmio, i++ * 4, magic);
+				magic += 0x0324;
+			}
+		}
+	} else {
+		u32 magic = 0x02180000;
+		nv_wo32(grch->mmio, i++ * 4, 0x00405830);
+		nv_wo32(grch->mmio, i++ * 4, magic | 0x0000218);
+		nv_wo32(grch->mmio, i++ * 4, 0x004064c4);
+		nv_wo32(grch->mmio, i++ * 4, 0x0086ffff);
+		for (gpc = 0; gpc < priv->gpc_nr; gpc++) {
+			for (tp = 0; tp < priv->tp_nr[gpc]; tp++) {
+				u32 reg = TP_UNIT(gpc, tp, 0x520);
+				nv_wo32(grch->mmio, i++ * 4, reg);
+				nv_wo32(grch->mmio, i++ * 4, (1 << 28) | magic);
+				magic += 0x0324;
+			}
+			for (tp = 0; tp < priv->tp_nr[gpc]; tp++) {
+				u32 reg = TP_UNIT(gpc, tp, 0x544);
+				nv_wo32(grch->mmio, i++ * 4, reg);
+				nv_wo32(grch->mmio, i++ * 4, magic);
+				magic += 0x0324;
+			}
 		}
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nvc0_grctx.c b/drivers/gpu/drm/nouveau/nvc0_grctx.c
index dd0e6a736b3..96b0b93d94c 100644
--- a/drivers/gpu/drm/nouveau/nvc0_grctx.c
+++ b/drivers/gpu/drm/nouveau/nvc0_grctx.c
@@ -1812,6 +1812,7 @@ nvc0_grctx_generate(struct nouveau_channel *chan)
 		/* calculate first set of magics */
 		memcpy(tpnr, priv->tp_nr, sizeof(priv->tp_nr));
 
+		gpc = -1;
 		for (tp = 0; tp < priv->