diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index dffc836144cc..f4dc9b7a3831 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -296,6 +296,18 @@ int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
 	case DRM_CAP_ASYNC_PAGE_FLIP:
 		req->value = dev->mode_config.async_page_flip;
 		break;
+	case DRM_CAP_CURSOR_WIDTH:
+		if (dev->mode_config.cursor_width)
+			req->value = dev->mode_config.cursor_width;
+		else
+			req->value = 64;
+		break;
+	case DRM_CAP_CURSOR_HEIGHT:
+		if (dev->mode_config.cursor_height)
+			req->value = dev->mode_config.cursor_height;
+		else
+			req->value = 64;
+		break;
 	default:
 		return -EINVAL;
 	}
diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index fa18cf374470..faa77f543a07 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -1151,8 +1151,10 @@ tda998x_encoder_init(struct i2c_client *client,
 
 	priv->current_page = 0xff;
 	priv->cec = i2c_new_dummy(client->adapter, 0x34);
-	if (!priv->cec)
+	if (!priv->cec) {
+		kfree(priv);
 		return -ENODEV;
+	}
 	priv->dpms = DRM_MODE_DPMS_OFF;
 
 	encoder_slave->slave_priv = priv;
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 9fa24347963a..4c1672809493 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -8586,6 +8586,20 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 	if (ring->id == RCS)
 		len += 6;
 
+	/*
+	 * BSpec MI_DISPLAY_FLIP for IVB:
+	 * "The full packet must be contained within the same cache line."
+	 *
+	 * Currently the LRI+SRM+MI_DISPLAY_FLIP all fit within the same
+	 * cacheline, if we ever start emitting more commands before
+	 * the MI_DISPLAY_FLIP we may need to first emit everything else,
+	 * then do the cacheline alignment, and finally emit the
+	 * MI_DISPLAY_FLIP.
+	 */
+	ret = intel_ring_cacheline_align(ring);
+	if (ret)
+		goto err_unpin;
+
 	ret = intel_ring_begin(ring, len);
 	if (ret)
 		goto err_unpin;
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 2f517b85b3f4..57552eb386b0 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -537,6 +537,7 @@ intel_dp_aux_native_write(struct intel_dp *intel_dp,
 	uint8_t	msg[20];
 	int msg_bytes;
 	uint8_t	ack;
+	int retry;
 
 	if (WARN_ON(send_bytes > 16))
 		return -E2BIG;
@@ -548,19 +549,21 @@ intel_dp_aux_native_write(struct intel_dp *intel_dp,
 	msg[3] = send_bytes - 1;
 	memcpy(&msg[4], send, send_bytes);
 	msg_bytes = send_bytes + 4;
-	for (;;) {
+	for (retry = 0; retry < 7; retry++) {
 		ret = intel_dp_aux_ch(intel_dp, msg, msg_bytes, &ack, 1);
 		if (ret < 0)
 			return ret;
 		ack >>= 4;
 		if ((ack & DP_AUX_NATIVE_REPLY_MASK) == DP_AUX_NATIVE_REPLY_ACK)
-			break;
+			return send_bytes;
 		else if ((ack & DP_AUX_NATIVE_REPLY_MASK) == DP_AUX_NATIVE_REPLY_DEFER)
-			udelay(100);
+			usleep_range(400, 500);
 		else
 			return -EIO;
 	}
-	return send_bytes;
+
+	DRM_ERROR("too many retries, giving up\n");
+	return -EIO;
 }
 
 /* Write a single byte to the aux channel in native mode */
@@ -582,6 +585,7 @@ intel_dp_aux_native_read(struct intel_dp *intel_dp,
 	int reply_bytes;
 	uint8_t ack;
 	int ret;
+	int retry;
 
 	if (WARN_ON(recv_bytes > 19))
 		return -E2BIG;
@@ -595,7 +599,7 @@ intel_dp_aux_native_read(struct intel_dp *intel_dp,
 	msg_bytes = 4;
 	reply_bytes = recv_bytes + 1;
 
-	for (;;) {
+	for (retry = 0; retry < 7; retry++) {
 		ret = intel_dp_aux_ch(intel_dp, msg, msg_bytes,
 				      reply, reply_bytes);
 		if (ret == 0)
@@ -608,10 +612,13 @@ intel_dp_aux_native_read(struct intel_dp *intel_dp,
 			return ret - 1;
 		}
 		else if ((ack & DP_AUX_NATIVE_REPLY_MASK) == DP_AUX_NATIVE_REPLY_DEFER)
-			udelay(100);
+			usleep_range(400, 500);
 		else
 			return -EIO;
 	}
+
+	DRM_ERROR("too many retries, giving up\n");
+	return -EIO;
 }
 
 static int
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index b7f1742caf87..31b36c5ac894 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -1653,6 +1653,27 @@ int intel_ring_begin(struct intel_ring_buffer *ring,
 	return 0;
 }
 
+/* Align the ring tail to a cacheline boundary */
+int intel_ring_cacheline_align(struct intel_ring_buffer *ring)
+{
+	int num_dwords = (64 - (ring->tail & 63)) / sizeof(uint32_t);
+	int ret;
+
+	if (num_dwords == 0)
+		return 0;
+
+	ret = intel_ring_begin(ring, num_dwords);
+	if (ret)
+		return ret;
+
+	while (num_dwords--)
+		intel_ring_emit(ring, MI_NOOP);
+
+	intel_ring_advance(ring);
+
+	return 0;
+}
+
 void intel_ring_init_seqno(struct intel_ring_buffer *ring, u32 seqno)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 71a73f4fe252..0b243ce33714 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -233,6 +233,7 @@ intel_write_status_page(struct intel_ring_buffer *ring,
 void intel_cleanup_ring_buffer(struct intel_ring_buffer *ring);
 
 int __must_check intel_ring_begin(struct intel_ring_buffer *ring, int n);
+int __must_check intel_ring_cacheline_align(struct intel_ring_buffer *ring);
 static inline void intel_ring_emit(struct intel_ring_buffer *ring,
 				   u32 data)
 {
diff --git a/drivers/gpu/drm/nouveau/Makefile b/drivers/gpu/drm/nouveau/Makefile
index e88145ba1bf5..d310c195bdfe 100644
--- a/drivers/gpu/drm/nouveau/Makefile
+++ b/drivers/gpu/drm/nouveau/Makefile
@@ -141,6 +141,7 @@ nouveau-y += core/subdev/mc/base.o
 nouveau-y += core/subdev/mc/nv04.o
 nouveau-y += core/subdev/mc/nv40.o
 nouveau-y += core/subdev/mc/nv44.o
+nouveau-y += core/subdev/mc/nv4c.o
 nouveau-y += core/subdev/mc/nv50.o
 nouveau-y += core/subdev/mc/nv94.o
 nouveau-y += core/subdev/mc/nv98.o
diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nv40.c b/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
index 1b653dd74a70..08b88591ed60 100644
--- a/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
+++ b/drivers/gpu/drm/nouveau/core/engine/device/nv40.c
@@ -311,7 +311,7 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv4c_mc_oclass;
 		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] =  nv46_fb_oclass;
@@ -334,7 +334,7 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv4c_mc_oclass;
 		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] =  nv4e_fb_oclass;
@@ -357,7 +357,7 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv4c_mc_oclass;
 		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] =  nv46_fb_oclass;
@@ -380,7 +380,7 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv4c_mc_oclass;
 		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] =  nv46_fb_oclass;
@@ -403,7 +403,7 @@ nv40_identify(struct nouveau_device *device)
 		device->oclass[NVDEV_SUBDEV_CLOCK  ] = &nv40_clock_oclass;
 		device->oclass[NVDEV_SUBDEV_THERM  ] = &nv40_therm_oclass;
 		device->oclass[NVDEV_SUBDEV_DEVINIT] =  nv1a_devinit_oclass;
-		device->oclass[NVDEV_SUBDEV_MC     ] =  nv44_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  nv4c_mc_oclass;
 		device->oclass[NVDEV_SUBDEV_BUS    ] =  nv31_bus_oclass;
 		device->oclass[NVDEV_SUBDEV_TIMER  ] = &nv04_timer_oclass;
 		device->oclass[NVDEV_SUBDEV_FB     ] =  nv46_fb_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
index 940eaa5d8b9a..9ad722e4e087 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nv50.c
@@ -1142,7 +1142,7 @@ nv50_disp_intr_unk20_2(struct nv50_disp_priv *priv, int head)
 	if (conf != ~0) {
 		if (outp.location == 0 && outp.type == DCB_OUTPUT_DP) {
 			u32 soff = (ffs(outp.or) - 1) * 0x08;
-			u32 ctrl = nv_rd32(priv, 0x610798 + soff);
+			u32 ctrl = nv_rd32(priv, 0x610794 + soff);
 			u32 datarate;
 
 			switch ((ctrl & 0x000f0000) >> 16) {
diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
index 9a850fe19515..54c1b5b471cd 100644
--- a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c
@@ -112,7 +112,7 @@ nve0_fifo_runlist_update(struct nve0_fifo_priv *priv, u32 engine)
 
 	nv_wr32(priv, 0x002270, cur->addr >> 12);
 	nv_wr32(priv, 0x002274, (engine << 20) | (p >> 3));
-	if (!nv_wait(priv, 0x002284 + (engine * 4), 0x00100000, 0x00000000))
+	if (!nv_wait(priv, 0x002284 + (engine * 8), 0x00100000, 0x00000000))
 		nv_error(priv, "runlist %d update timeout\n", engine);
 	mutex_unlock(&nv_subdev(priv)->mutex);
 }
diff --git a/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c b/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
index 30ed19c52e05..7a367c402978 100644
--- a/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
+++ b/drivers/gpu/drm/nouveau/core/engine/graph/nv50.c
@@ -539,7 +539,7 @@ nv50_priv_tp_trap(struct nv50_graph_priv *priv, int type, u32 ustatus_old,
 				ustatus &= ~0x04030000;
 			}
 			if (ustatus && display) {
-				nv_error("%s - TP%d:", name, i);
+				nv_error(priv, "%s - TP%d:", name, i);
 				nouveau_bitfield_print(nv50_mpc_traps, ustatus);
 				pr_cont("\n");
 				ustatus = 0;
diff --git a/drivers/gpu/drm/nouveau/core/include/subdev/mc.h b/drivers/gpu/drm/nouveau/core/include/subdev/mc.h
index adc88b73d911..3c6738edd127 100644
--- a/drivers/gpu/drm/nouveau/core/include/subdev/mc.h
+++ b/drivers/gpu/drm/nouveau/core/include/subdev/mc.h
@@ -47,6 +47,7 @@ struct nouveau_mc_oclass {
 extern struct nouveau_oclass *nv04_mc_oclass;
 extern struct nouveau_oclass *nv40_mc_oclass;
 extern struct nouveau_oclass *nv44_mc_oclass;
+extern struct nouveau_oclass *nv4c_mc_oclass;
 extern struct nouveau_oclass *nv50_mc_oclass;
 extern struct nouveau_oclass *nv94_mc_oclass;
 extern struct nouveau_oclass *nv98_mc_oclass;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/base.c b/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
index aa0fbbec7f08..ef0c9c4a8cc3 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/bios/base.c
@@ -130,6 +130,10 @@ nouveau_bios_shadow_prom(struct nouveau_bios *bios)
 	u16 pcir;
 	int i;
 
+	/* there is no prom on nv4x IGP's */
+	if (device->card_type == NV_40 && device->chipset >= 0x4c)
+		return;
+
 	/* enable access to rom */
 	if (device->card_type >= NV_50)
 		pcireg = 0x088050;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/nv1a.c b/drivers/gpu/drm/nouveau/core/subdev/fb/nv1a.c
index 9159a5ccee93..265d1253624a 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/fb/nv1a.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/fb/nv1a.c
@@ -36,7 +36,7 @@ nv1a_fb_oclass = &(struct nv04_fb_impl) {
 		.fini = _nouveau_fb_fini,
 	},
 	.base.memtype = nv04_fb_memtype_valid,
-	.base.ram = &nv10_ram_oclass,
+	.base.ram = &nv1a_ram_oclass,
 	.tile.regions = 8,
 	.tile.init = nv10_fb_tile_init,
 	.tile.fini = nv10_fb_tile_fini,
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.h b/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.h
index b0d5c31606c1..81a408e7d034 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.h
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv04.h
@@ -14,6 +14,7 @@ int  nv04_mc_ctor(struct nouveau_object *, struct nouveau_object *,
 extern const struct nouveau_mc_intr nv04_mc_intr[];
 int  nv04_mc_init(struct nouveau_object *);
 void nv40_mc_msi_rearm(struct nouveau_mc *);
+int  nv44_mc_init(struct nouveau_object *object);
 int  nv50_mc_init(struct nouveau_object *);
 extern const struct nouveau_mc_intr nv50_mc_intr[];
 extern const struct nouveau_mc_intr nvc0_mc_intr[];
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv44.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv44.c
index 3bfee5c6c4f2..cc4d0d2d886e 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv44.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv44.c
@@ -24,7 +24,7 @@
 
 #include "nv04.h"
 
-static int
+int
 nv44_mc_init(struct nouveau_object *object)
 {
 	struct nv04_mc_priv *priv = (void *)object;
diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv4c.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv4c.c
new file mode 100644
index 000000000000..a75c35ccf25c
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv4c.c
@@ -0,0 +1,45 @@
+/*
+ * Copyright 2014 Ilia Mirkin
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ilia Mirkin
+ */
+
+#include "nv04.h"
+
+static void
+nv4c_mc_msi_rearm(struct nouveau_mc *pmc)
+{
+	struct nv04_mc_priv *priv = (void *)pmc;
+	nv_wr08(priv, 0x088050, 0xff);
+}
+
+struct nouveau_oclass *
+nv4c_mc_oclass = &(struct nouveau_mc_oclass) {
+	.base.handle = NV_SUBDEV(MC, 0x4c),
+	.base.ofuncs = &(struct nouveau_ofuncs) {
+		.ctor = nv04_mc_ctor,
+		.dtor = _nouveau_mc_dtor,
+		.init = nv44_mc_init,
+		.fini = _nouveau_mc_fini,
+	},
+	.intr = nv04_mc_intr,
+	.msi_rearm = nv4c_mc_msi_rearm,
+}.base;
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 488686d490c0..4aed1714b9ab 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -1249,7 +1249,7 @@ nouveau_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
 			mem->bus.is_iomem = !dev->agp->cant_use_aperture;
 		}
 #endif
-		if (!node->memtype)
+		if (nv_device(drm->device)->card_type < NV_50 || !node->memtype)
 			/* untiled */
 			break;
 		/* fallthrough, tiled memory */
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 78c8e7146d56..89c484d8ac26 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -376,6 +376,8 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags)
 	if (ret)
 		goto fail_device;
 
+	dev->irq_enabled = true;
+
 	/* 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...
@@ -475,6 +477,7 @@ nouveau_drm_remove(struct pci_dev *pdev)
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nouveau_object *device;
 
+	dev->irq_enabled = false;
 	device = drm->client.base.device;
 	drm_put_dev(dev);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_vga.c b/drivers/gpu/drm/nouveau/nouveau_vga.c
index 81638d7f2eff..471347edc27e 100644
--- a/drivers/gpu/drm/nouveau/nouveau_vga.c
+++ b/drivers/gpu/drm/nouveau/nouveau_vga.c
@@ -14,7 +14,9 @@ nouveau_vga_set_decode(void *priv, bool state)
 {
 	struct nouveau_device *device = nouveau_dev(priv);
 
-	if (device->chipset >= 0x40)
+	if (device->card_type == NV_40 && device->chipset >= 0x4c)
+		nv_wr32(device, 0x088060, state);
+	else if (device->chipset >= 0x40)
 		nv_wr32(device, 0x088054, state);
 	else
 		nv_wr32(device, 0x001854, state);
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index a9338c85630f..0d19f4f94d5a 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -559,7 +559,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc,
 	u32 adjusted_clock = mode->clock;
 	int encoder_mode = atombios_get_encoder_mode(encoder);
 	u32 dp_clock = mode->clock;
-	int bpc = radeon_get_monitor_bpc(connector);
+	int bpc = radeon_crtc->bpc;
 	bool is_duallink = radeon_dig_monitor_is_duallink(encoder, mode->clock);
 
 	/* reset the pll flags */
@@ -1176,7 +1176,7 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
 		evergreen_tiling_fields(tiling_flags, &bankw, &bankh, &mtaspect, &tile_split);
 
 		/* Set NUM_BANKS. */
-		if (rdev->family >= CHIP_BONAIRE) {
+		if (rdev->family >= CHIP_TAHITI) {
 			unsigned tileb, index, num_banks, tile_split_bytes;
 
 			/* Calculate the macrotile mode index. */
@@ -1194,13 +1194,14 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc,
 				return -EINVAL;
 			}
 
-			num_banks = (rdev->config.cik.macrotile_mode_array[index] >> 6) & 0x3;
+			if (rdev->family >= CHIP_BONAIRE)
+				num_banks = (rdev->config.cik.macrotile_mode_array[index] >> 6) & 0x3;
+			else
+				num_banks = (rdev->config.si.tile_mode_array[index] >> 20) & 0x3;
 			fb_format |= EVERGREEN_GRPH_NUM_BANKS(num_banks);
 		} else {
-			/* SI and older. */
-			if (rdev->family >= CHIP_TAHITI)
-				tmp = rdev->config.si.tile_config;
-			else if (rdev->family >= CHIP_CAYMAN)
+			/* NI and older. */
+			if (rdev->family >= CHIP_CAYMAN)
 				tmp = rdev->config.cayman.tile_config;
 			else
 				tmp = rdev->config.evergreen.tile_config;
diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c
index a42d61571f49..2cec2ab02f80 100644
--- a/drivers/gpu/drm/radeon/atombios_encoders.c
+++ b/drivers/gpu/drm/radeon/atombios_encoders.c
@@ -464,11 +464,12 @@ atombios_tv_setup(struct drm_encoder *encoder, int action)
 
 static u8 radeon_atom_get_bpc(struct drm_encoder *encoder)
 {
-	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 	int bpc = 8;
 
-	if (connector)
-		bpc = radeon_get_monitor_bpc(connector);
+	if (encoder->crtc) {
+		struct radeon_crtc *radeon_crtc = to_radeon_crtc(encoder->crtc);
+		bpc = radeon_crtc->bpc;
+	}
 
 	switch (bpc) {
 	case 0:
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index f2b9e21ce4da..5623e7542d99 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -1680,7 +1680,7 @@ bool evergreen_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd)
 	case RADEON_HPD_6:
 		if (RREG32(DC_HPD6_INT_STATUS) & DC_HPDx_SENSE)
 			connected = true;
-			break;
+		break;
 	default:
 		break;
 	}
diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c
index 1217fbcbdcca..ca814276b075 100644
--- a/drivers/gpu/drm/radeon/ni_dpm.c
+++ b/drivers/gpu/drm/radeon/ni_dpm.c
@@ -2588,7 +2588,7 @@ static int ni_populate_sq_ramping_values(struct radeon_device *rdev,
 	if (NISLANDS_DPM2_SQ_RAMP_STI_SIZE > (STI_SIZE_MASK >> STI_SIZE_SHIFT))
 		enable_sq_ramping = false;
 
-	if (NISLANDS_DPM2_SQ_RAMP_LTI_RATIO <= (LTI_RATIO_MASK >> LTI_RATIO_SHIFT))
+	if (NISLANDS_DPM2_SQ_RAMP_LTI_RATIO > (LTI_RATIO_MASK >> LTI_RATIO_SHIFT))
 		enable_sq_ramping = false;
 
 	for (i = 0; i < state->performance_level_count; i++) {
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 4a8ac1cd6b4c..024db37b1832 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -135,6 +135,9 @@ extern int radeon_hard_reset;
 /* R600+ */
 #define R600_RING_TYPE_UVD_INDEX	5
 
+/* number of hw syncs before falling back on blocking */
+#define RADEON_NUM_SYNCS			4
+
 /* hardcode those limit for now */
 #define RADEON_VA_IB_OFFSET			(1 << 20)
 #define RADEON_VA_RESERVED_SIZE			(8 << 20)
@@ -554,7 +557,6 @@ int radeon_mode_dumb_mmap(struct drm_file *filp,
 /*
  * Semaphores.
  */
-/* everything here is constant */
 struct radeon_semaphore {
 	struct radeon_sa_bo		*sa_bo;
 	signed				waiters;
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index d680608f6f5b..fbd8b930f2be 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -571,6 +571,8 @@ static void radeon_crtc_init(struct drm_device *dev, int index)
 		radeon_crtc->max_cursor_width = CURSOR_WIDTH;
 		radeon_crtc->max_cursor_height = CURSOR_HEIGHT;
 	}
+	dev->mode_config.cursor_width = radeon_crtc->max_cursor_width;
+	dev->mode_config.cursor_height = radeon_crtc->max_cursor_height;
 
 #if 0
 	radeon_crtc->mode_set.crtc = &radeon_crtc->base;
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index 1b783f0e6d3a..15e44a7281ab 100644
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -139,7 +139,7 @@ int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
 	}
 
 	/* 64 dwords should be enough for fence too */
-	r = radeon_ring_lock(rdev, ring, 64 + RADEON_NUM_RINGS * 8);
+	r = radeon_ring_lock(rdev, ring, 64 + RADEON_NUM_SYNCS * 8);
 	if (r) {
 		dev_err(rdev->dev, "scheduling IB failed (%d).\n", r);
 		return r;
diff --git a/drivers/gpu/drm/radeon/radeon_semaphore.c b/drivers/gpu/drm/radeon/radeon_semaphore.c
index 2b42aa1914f2..9006b32d5eed 100644
--- a/drivers/gpu/drm/radeon/radeon_semaphore.c
+++ b/drivers/gpu/drm/radeon/radeon_semaphore.c
@@ -34,14 +34,15 @@
 int radeon_semaphore_create(struct radeon_device *rdev,
 			    struct radeon_semaphore **semaphore)
 {
+	uint32_t *cpu_addr;
 	int i, r;
 
 	*semaphore = kmalloc(sizeof(struct radeon_semaphore), GFP_KERNEL);
 	if (*semaphore == NULL) {
 		return -ENOMEM;
 	}
-	r = radeon_sa_bo_new(rdev, &rdev->ring_tmp_bo,
-			     &(*semaphore)->sa_bo, 8, 8, true);
+	r = radeon_sa_bo_new(rdev, &rdev->ring_tmp_bo, &(*semaphore)->sa_bo,
+			     8 * RADEON_NUM_SYNCS, 8, true);
 	if (r) {
 		kfree(*semaphore);
 		*semaphore = NULL;
@@ -49,7 +50,10 @@ int radeon_semaphore_create(struct radeon_device *rdev,
 	}
 	(*semaphore)->waiters = 0;
 	(*semaphore)->gpu_addr = radeon_sa_bo_gpu_addr((*semaphore)->sa_bo);
-	*((uint64_t*)radeon_sa_bo_cpu_addr((*semaphore)->sa_bo)) = 0;
+
+	cpu_addr = radeon_sa_bo_cpu_addr((*semaphore)->sa_bo);
+	for (i = 0; i < RADEON_NUM_SYNCS; ++i)
+		cpu_addr[i] = 0;
 
 	for (i = 0; i < RADEON_NUM_RINGS; ++i)
 		(*semaphore)->sync_to[i] = NULL;
@@ -125,6 +129,7 @@ int radeon_semaphore_sync_rings(struct radeon_device *rdev,
 				struct radeon_semaphore *semaphore,
 				int ring)
 {
+	unsigned count = 0;
 	int i, r;
 
         for (i = 0; i < RADEON_NUM_RINGS; ++i) {
@@ -140,6 +145,12 @@ int radeon_semaphore_sync_rings(struct radeon_device *rdev,
 			return -EINVAL;
 		}
 
+		if (++count > RADEON_NUM_SYNCS) {
+			/* not enough room, wait manually */
+			radeon_fence_wait_locked(fence);
+			continue;
+		}
+
 		/* allocate enough space for sync command */
 		r = radeon_ring_alloc(rdev, &rdev->ring[i], 16);
 		if (r) {
@@ -164,6 +175,8 @@ int radeon_semaphore_sync_rings(struct radeon_device *rdev,
 
 		radeon_ring_commit(rdev, &rdev->ring[i]);
 		radeon_fence_note_sync(fence, ring);
+
+		semaphore->gpu_addr += 8;
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/radeon/rv770_dpm.c b/drivers/gpu/drm/radeon/rv770_dpm.c
index 5b2ea8ac0731..b5f63f5e22a3 100644
--- a/drivers/gpu/drm/radeon/rv770_dpm.c
+++ b/drivers/gpu/drm/radeon/rv770_dpm.c
@@ -2526,14 +2526,7 @@ u32 rv770_dpm_get_mclk(struct radeon_device *rdev, bool low)
 bool rv770_dpm_vblank_too_short(struct radeon_device *rdev)
 {
 	u32 vblank_time = r600_dpm_get_vblank_time(rdev);
-	u32 switch_limit = 300;
-
-	/* quirks */
-	/* ASUS K70AF */
-	if ((rdev->pdev->device == 0x9553) &&
-	    (rdev->pdev->subsystem_vendor == 0x1043) &&
-	    (rdev->pdev->subsystem_device == 0x1c42))
-		switch_limit = 200;
+	u32 switch_limit = 200; /* 300 */
 
 	/* RV770 */
 	/* mclk switching doesn't seem to work reliably on desktop RV770s */
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index eafb0e6bc67e..0a2f5b4bca43 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -2395,7 +2395,7 @@ static int si_populate_sq_ramping_values(struct radeon_device *rdev,
 	if (SISLANDS_DPM2_SQ_RAMP_STI_SIZE > (STI_SIZE_MASK >> STI_SIZE_SHIFT))
 		enable_sq_ramping = false;
 
-	if (SISLANDS_DPM2_SQ_RAMP_LTI_RATIO <= (LTI_RATIO_MASK >> LTI_RATIO_SHIFT))
+	if (SISLANDS_DPM2_SQ_RAMP_LTI_RATIO > (LTI_RATIO_MASK >> LTI_RATIO_SHIFT))
 		enable_sq_ramping = false;
 
 	for (i = 0; i < state->performance_level_count; i++) {
diff --git a/drivers/gpu/drm/ttm/ttm_agp_backend.c b/drivers/gpu/drm/ttm/ttm_agp_backend.c
index 3302f99e7497..764be36397fd 100644
--- a/drivers/gpu/drm/ttm/ttm_agp_backend.c
+++ b/drivers/gpu/drm/ttm/ttm_agp_backend.c
@@ -126,6 +126,7 @@ struct ttm_tt *ttm_agp_tt_create(struct ttm_bo_device *bdev,
 	agp_be->ttm.func = &ttm_agp_func;
 
 	if (ttm_tt_init(&agp_be->ttm, bdev, size, page_flags, dummy_read_page)) {
+		kfree(agp_be);
 		return NULL;
 	}
 
diff --git a/drivers/gpu/drm/vmwgfx/svga3d_reg.h b/drivers/gpu/drm/vmwgfx/svga3d_reg.h
index b645647b7776..bb594c11605e 100644
--- a/drivers/gpu/drm/vmwgfx/svga3d_reg.h
+++ b/drivers/gpu/drm/vmwgfx/svga3d_reg.h
@@ -1223,9 +1223,19 @@ typedef enum {
 #define SVGA_3D_CMD_INVALIDATE_GB_IMAGE_PARTIAL 1129
 
 #define SVGA_3D_CMD_SET_GB_SHADERCONSTS_INLINE  1130
-
+#define SVGA_3D_CMD_GB_SCREEN_DMA               1131
+#define SVGA_3D_CMD_BIND_GB_SURFACE_WITH_PITCH  1132
+#define SVGA_3D_CMD_GB_MOB_FENCE                1133
+#define SVGA_3D_CMD_DEFINE_GB_SURFACE_V2        1134
 #define SVGA_3D_CMD_DEFINE_GB_MOB64          1135
 #define SVGA_3D_CMD_REDEFINE_GB_MOB64        1136
+#define SVGA_3D_CMD_NOP_ERROR                1137
+
+#define SVGA_3D_CMD_RESERVED1                1138
+#define SVGA_3D_CMD_RESERVED2                1139
+#define SVGA_3D_CMD_RESERVED3                1140
+#define SVGA_3D_CMD_RESERVED4                1141
+#define SVGA_3D_CMD_RESERVED5                1142
 
 #define SVGA_3D_CMD_MAX                      1142
 #define SVGA_3D_CMD_FUTURE_MAX               3000
@@ -1973,8 +1983,7 @@ struct {
    uint32 sizeInBytes;
    uint32 validSizeInBytes;
    SVGAMobFormat ptDepth;
-}
-__attribute__((__packed__))
+} __packed
 SVGA3dCmdSetOTableBase;  /* SVGA_3D_CMD_SET_OTABLE_BASE */
 
 typedef
@@ -1984,15 +1993,13 @@ struct {
    uint32 sizeInBytes;
    uint32 validSizeInBytes;
    SVGAMobFormat ptDepth;
-}
-__attribute__((__packed__))
+} __packed
 SVGA3dCmdSetOTableBase64;  /* SVGA_3D_CMD_SET_OTABLE_BASE64 */
 
 typedef
 struct {
    SVGAOTableType type;
-}
-__attribute__((__packed__))
+} __packed
 SVGA3dCmdReadbackOTable;  /* SVGA_3D_CMD_READBACK_OTABLE */
 
 /*
@@ -2005,8 +2012,7 @@ struct SVGA3dCmdDefineGBMob {
    SVGAMobFormat ptDepth;
    PPN base;
    uint32 sizeInBytes;
-}
-__attribute__((__packed__))
+} __packed
 SVGA3dCmdDefineGBMob;   /* SVGA_3D_CMD_DEFINE_GB_MOB */
 
 
@@ -2017,8 +2023,7 @@ SVGA3dCmdDefineGBMob;   /* SVGA_3D_CMD_DEFINE_GB_MOB */
 typedef
 struct SVGA3dCmdDestroyGBMob {
    SVGAMobId mobid;
-}
-__attribute__((__packed__))
+} __packed
 SVGA3dCmdDestroyGBMob;   /* SVGA_3D_CMD_DESTROY_GB_MOB */
 
 /*
@@ -2031,8 +2036,7 @@ struct SVGA3dCmdRedefineGBMob {
    SVGAMobFormat ptDepth;
    PPN base;
    uint32 sizeInBytes;
-}
-__attribute__((__packed__))
+} __packed
 SVGA3dCmdRedefineGBMob;   /* SVGA_3D_CMD_REDEFINE_GB_MOB */
 
 /*
@@ -2045,8 +2049,7 @@ struct SVGA3dCmdDefineGBMob64 {
    SVGAMobFormat ptDepth;
    PPN64 base;
    uint32 sizeInBytes;
-}
-__attribute__((__packed__))
+} __packed
 SVGA3dCmdDefineGBMob64;   /* SVGA_3D_CMD_DEFINE_GB_MOB64 */
 
 /*
@@ -2059,8 +2062,7 @@ struct SVGA3dCmdRedefineGBMob64 {
    SVGAMobFormat ptDepth;
    PPN64 base;
    uint32 sizeInBytes;
-}
-__attribute__((__packed__))
+} __packed
 SVGA3dCmdRedefineGBMob64;   /* SVGA_3D_CMD_REDEFINE_GB_MOB64 */
 
 /*
@@ -2070,8 +2072,7 @@ SVGA3dCmdRedefineGBMob64;   /* SVGA_3D_CMD_REDEFINE_GB_MOB64 */
 typedef
 struct SVGA3dCmdUpdateGBMobMapping {
    SVGAMobId mobid;
-}
-__attribute__((__packed__))
+} __packed
 SVGA3dCmdUpdateGBMobMapping;   /* SVGA_3D_CMD_UPDATE_GB_MOB_MAPPING */
 
 /*
@@ -2087,7 +2088,8 @@ struct SVGA3dCmdDefineGBSurface {
    uint32 multisampleCount;
    SVGA3dTextureFilter autogenFilter;
    SVGA3dSize size;
-} SVGA3dCmdDefineGBSurface;   /* SVGA_3D_CMD_DEFINE_GB_SURFACE */
+} __packed
+SVGA3dCmdDefineGBSurface;   /* SVGA_3D_CMD_DEFINE_GB_SURFACE */
 
 /*
  * Destroy a guest-backed surface.
@@ -2096,7 +2098,8 @@ struct SVGA3dCmdDefineGBSurface {
 typedef
 struct SVGA3dCmdDestroyGBSurface {
    uint32 sid;
-} SVGA3dCmdDestroyGBSurface;   /* SVGA_3D_CMD_DESTROY_GB_SURFACE */
+} __packed
+SVGA3dCmdDestroyGBSurface;   /* SVGA_3D_CMD_DESTROY_GB_SURFACE */
 
 /*
  * Bind a guest-backed surface to an object.
@@ -2106,7 +2109,8 @@ typedef
 struct SVGA3dCmdBindGBSurface {
    uint32 sid;
    SVGAMobId mobid;
-} SVGA3dCmdBindGBSurface;   /* SVGA_3D_CMD_BIND_GB_SURFACE */
+} __packed
+SVGA3dCmdBindGBSurface;   /* SVGA_3D_CMD_BIND_GB_SURFACE */
 
 /*
  * Conditionally bind a mob to a guest backed surface if testMobid
@@ -2123,7 +2127,7 @@ struct{
    SVGAMobId testMobid;
    SVGAMobId mobid;
    uint32 flags;
-}
+} __packed
 SVGA3dCmdCondBindGBSurface;          /* SVGA_3D_CMD_COND_BIND_GB_SURFACE */
 
 /*
@@ -2135,7 +2139,8 @@ typedef
 struct SVGA3dCmdUpdateGBImage {
    SVGA3dSurfaceImageId image;
    SVGA3dBox box;
-} SVGA3dCmdUpdateGBImage;   /* SVGA_3D_CMD_UPDATE_GB_IMAGE */
+} __packed
+SVGA3dCmdUpdateGBImage;   /* SVGA_3D_CMD_UPDATE_GB_IMAGE */
 
 /*
  * Update an entire guest-backed surface.
@@ -2145,7 +2150,8 @@ struct SVGA3dCmdUpdateGBImage {
 typedef
 struct SVGA3dCmdUpdateGBSurface {
    uint32 sid;
-} SVGA3dCmdUpdateGBSurface;   /* SVGA_3D_CMD_UPDATE_GB_SURFACE */
+} __packed
+SVGA3dCmdUpdateGBSurface;   /* SVGA_3D_CMD_UPDATE_GB_SURFACE */
 
 /*
  * Readback an image in a guest-backed surface.
@@ -2155,7 +2161,8 @@ struct SVGA3dCmdUpdateGBSurface {
 typedef
 struct SVGA3dCmdReadbackGBImage {
    SVGA3dSurfaceImageId image;
-} SVGA3dCmdReadbackGBImage;   /* SVGA_3D_CMD_READBACK_GB_IMAGE*/
+} __packed
+SVGA3dCmdReadbackGBImage;   /* SVGA_3D_CMD_READBACK_GB_IMAGE*/
 
 /*
  * Readback an entire guest-backed surface.
@@ -2165,7 +2172,8 @@ struct SVGA3dCmdReadbackGBImage {
 typedef
 struct SVGA3dCmdReadbackGBSurface {
    uint32 sid;
-} SVGA3dCmdReadbackGBSurface;   /* SVGA_3D_CMD_READBACK_GB_SURFACE */
+} __packed
+SVGA3dCmdReadbackGBSurface;   /* SVGA_3D_CMD_READBACK_GB_SURFACE */
 
 /*
  * Readback a sub rect of an image in a guest-backed surface.  After
@@ -2179,7 +2187,7 @@ struct SVGA3dCmdReadbackGBImagePartial {
    SVGA3dSurfaceImageId image;
    SVGA3dBox box;
    uint32 invertBox;
-}
+} __packed
 SVGA3dCmdReadbackGBImagePartial; /* SVGA_3D_CMD_READBACK_GB_IMAGE_PARTIAL */
 
 /*
@@ -2190,7 +2198,8 @@ SVGA3dCmdReadbackGBImagePartial; /* SVGA_3D_CMD_READBACK_GB_IMAGE_PARTIAL */
 typedef
 struct SVGA3dCmdInvalidateGBImage {
    SVGA3dSurfaceImageId image;
-} SVGA3dCmdInvalidateGBImage;   /* SVGA_3D_CMD_INVALIDATE_GB_IMAGE */
+} __packed
+SVGA3dCmdInvalidateGBImage;   /* SVGA_3D_CMD_INVALIDATE_GB_IMAGE */
 
 /*
  * Invalidate an entire guest-backed surface.
@@ -2200,7 +2209,8 @@ struct SVGA3dCmdInvalidateGBImage {
 typedef
 struct SVGA3dCmdInvalidateGBSurface {
    uint32 sid;
-} SVGA3dCmdInvalidateGBSurface; /* SVGA_3D_CMD_INVALIDATE_GB_SURFACE */
+} __packed
+SVGA3dCmdInvalidateGBSurface; /* SVGA_3D_CMD_INVALIDATE_GB_SURFACE */
 
 /*
  * Invalidate a sub rect of an image in a guest-backed surface.  After
@@ -2214,7 +2224,7 @@ struct SVGA3dCmdInvalidateGBImagePartial {
    SVGA3dSurfaceImageId image;
    SVGA3dBox box;
    uint32 invertBox;
-}
+} __packed
 SVGA3dCmdInvalidateGBImagePartial; /* SVGA_3D_CMD_INVALIDATE_GB_IMAGE_PARTIAL */
 
 /*
@@ -2224,7 +2234,8 @@ SVGA3dCmdInvalidateGBImagePartial; /* SVGA_3D_CMD_INVALIDATE_GB_IMAGE_PARTIAL */
 typedef
 struct SVGA3dCmdDefineGBContext {
    uint32 cid;
-} SVGA3dCmdDefineGBContext;   /* SVGA_3D_CMD_DEFINE_GB_CONTEXT */
+} __packed
+SVGA3dCmdDefineGBContext;   /* SVGA_3D_CMD_DEFINE_GB_CONTEXT */
 
 /*
  * Destroy a guest-backed context.
@@ -2233,7 +2244,8 @@ struct SVGA3dCmdDefineGBContext {
 typedef
 struct SVGA3dCmdDestroyGBContext {
    uint32 cid;
-} SVGA3dCmdDestroyGBContext;   /* SVGA_3D_CMD_DESTROY_GB_CONTEXT */
+} __packed
+SVGA3dCmdDestroyGBContext;   /* SVGA_3D_CMD_DESTROY_GB_CONTEXT */
 
 /*
  * Bind a guest-backed context.
@@ -2252,7 +2264,8 @@ struct SVGA3dCmdBindGBContext {
    uint32 cid;
    SVGAMobId mobid;
    uint32 validContents;
-} SVGA3dCmdBindGBContext;   /* SVGA_3D_CMD_BIND_GB_CONTEXT */
+} __packed
+SVGA3dCmdBindGBContext;   /* SVGA_3D_CMD_BIND_GB_CONTEXT */
 
 /*
  * Readback a guest-backed context.
@@ -2262,7 +2275,8 @@ struct SVGA3dCmdBindGBContext {
 typedef
 struct SVGA3dCmdReadbackGBContext {
    uint32 cid;
-} SVGA3dCmdReadbackGBContext;   /* SVGA_3D_CMD_READBACK_GB_CONTEXT */
+} __packed
+SVGA3dCmdReadbackGBContext;   /* SVGA_3D_CMD_READBACK_GB_CONTEXT */
 
 /*
  * Invalidate a guest-backed context.
@@ -2270,7 +2284,8 @@ struct SVGA3dCmdReadbackGBContext {
 typedef
 struct SVGA3dCmdInvalidateGBContext {
    uint32 cid;
-} SVGA3dCmdInvalidateGBContext;   /* SVGA_3D_CMD_INVALIDATE_GB_CONTEXT */
+} __packed
+SVGA3dCmdInvalidateGBContext;   /* SVGA_3D_CMD_INVALIDATE_GB_CONTEXT */
 
 /*
  * Define a guest-backed shader.
@@ -2281,7 +2296,8 @@ struct SVGA3dCmdDefineGBShader {
    uint32 shid;
    SVGA3dShaderType type;
    uint32 sizeInBytes;
-} SVGA3dCmdDefineGBShader;   /* SVGA_3D_CMD_DEFINE_GB_SHADER */
+} __packed
+SVGA3dCmdDefineGBShader;   /* SVGA_3D_CMD_DEFINE_GB_SHADER */
 
 /*
  * Bind a guest-backed shader.
@@ -2291,7 +2307,8 @@ typedef struct SVGA3dCmdBindGBShader {
    uint32 shid;
    SVGAMobId mobid;
    uint32 offsetInBytes;
-} SVGA3dCmdBindGBShader;   /* SVGA_3D_CMD_BIND_GB_SHADER */
+} __packed
+SVGA3dCmdBindGBShader;   /* SVGA_3D_CMD_BIND_GB_SHADER */
 
 /*
  * Destroy a guest-backed shader.
@@ -2299,7 +2316,8 @@ typedef struct SVGA3dCmdBindGBShader {
 
 typedef struct SVGA3dCmdDestroyGBShader {
    uint32 shid;
-} SVGA3dCmdDestroyGBShader;   /* SVGA_3D_CMD_DESTROY_GB_SHADER */
+} __packed
+SVGA3dCmdDestroyGBShader;   /* SVGA_3D_CMD_DESTROY_GB_SHADER */
 
 typedef
 struct {
@@ -2314,14 +2332,16 @@ struct {
     * Note that FLOAT and INT constants are 4-dwords in length, while
     * BOOL constants are 1-dword in length.
     */
-} SVGA3dCmdSetGBShaderConstInline;
+} __packed
+SVGA3dCmdSetGBShaderConstInline;
 /* SVGA_3D_CMD_SET_GB_SHADERCONSTS_INLINE */
 
 typedef
 struct {
    uint32               cid;
    SVGA3dQueryType      type;
-} SVGA3dCmdBeginGBQuery;           /* SVGA_3D_CMD_BEGIN_GB_QUERY */
+} __packed
+SVGA3dCmdBeginGBQuery;           /* SVGA_3D_CMD_BEGIN_GB_QUERY */
 
 typedef
 struct {
@@ -2329,7 +2349,8 @@ struct {
    SVGA3dQueryType      type;
    SVGAMobId mobid;
    uint32 offset;
-} SVGA3dCmdEndGBQuery;                  /* SVGA_3D_CMD_END_GB_QUERY */
+} __packed
+SVGA3dCmdEndGBQuery;                  /* SVGA_3D_CMD_END_GB_QUERY */
 
 
 /*
@@ -2346,21 +2367,22 @@ struct {
    SVGA3dQueryType      type;
    SVGAMobId mobid;
    uint32 offset;
-} SVGA3dCmdWaitForGBQuery;          /* SVGA_3D_CMD_WAIT_FOR_GB_QUERY */
+} __packed
+SVGA3dCmdWaitForGBQuery;          /* SVGA_3D_CMD_WAIT_FOR_GB_QUERY */
 
 typedef
 struct {
    SVGAMobId mobid;
    uint32 fbOffset;
    uint32 initalized;
-}
+} __packed
 SVGA3dCmdEnableGart;              /* SVGA_3D_CMD_ENABLE_GART */
 
 typedef
 struct {
    SVGAMobId mobid;
    uint32 gartOffset;
-}
+} __packed
 SVGA3dCmdMapMobIntoGart;          /* SVGA_3D_CMD_MAP_MOB_INTO_GART */
 
 
@@ -2368,7 +2390,7 @@ typedef
 struct {
    uint32 gartOffset;
    uint32 numPages;
-}
+} __packed
 SVGA3dCmdUnmapGartRange;          /* SVGA_3D_CMD_UNMAP_GART_RANGE */
 
 
@@ -2385,27 +2407,27 @@ struct {
    int32 xRoot;
    int32 yRoot;
    uint32 flags;
-}
+} __packed
 SVGA3dCmdDefineGBScreenTarget;    /* SVGA_3D_CMD_DEFINE_GB_SCREENTARGET */
 
 typedef
 struct {
    uint32 stid;
-}
+} __packed
 SVGA3dCmdDestroyGBScreenTarget;  /* SVGA_3D_CMD_DESTROY_GB_SCREENTARGET */
 
 typedef
 struct {
    uint32 stid;
    SVGA3dSurfaceImageId image;
-}
+} __packed
 SVGA3dCmdBindGBScreenTarget;  /* SVGA_3D_CMD_BIND_GB_SCREENTARGET */
 
 typedef
 struct {
    uint32 stid;
    SVGA3dBox box;
-}
+} __packed
 SVGA3dCmdUpdateGBScreenTarget;  /* SVGA_3D_CMD_UPDATE_GB_SCREENTARGET */
 
 /*
diff --git a/drivers/gpu/drm/vmwgfx/svga3d_surfacedefs.h b/drivers/gpu/drm/vmwgfx/svga3d_surfacedefs.h
index 8369c3ba10fe..ef3385096145 100644
--- a/drivers/gpu/drm/vmwgfx/svga3d_surfacedefs.h
+++ b/drivers/gpu/drm/vmwgfx/svga3d_surfacedefs.h
@@ -38,8 +38,11 @@
 
 #define DIV_ROUND_UP(x, y)  (((x) + (y) - 1) / (y))
 #define max_t(type, x, y)  ((x) > (y) ? (x) : (y))
+#define min_t(type, x, y)  ((x) < (y) ? (x) : (y))
 #define surf_size_struct SVGA3dSize
 #define u32 uint32
+#define u64 uint64_t
+#define U32_MAX ((u32)~0U)
 
 #endif /* __KERNEL__ */
 
@@ -704,8 +707,8 @@ static const struct svga3d_surface_desc svga3d_surface_descs[] = {
 
 static inline u32 clamped_umul32(u32 a, u32 b)
 {
-	uint64_t tmp = (uint64_t) a*b;
-	return (tmp > (uint64_t) ((u32) -1)) ? (u32) -1 : tmp;
+	u64 tmp = (u64) a*b;
+	return (tmp > (u64) U32_MAX) ? U32_MAX : tmp;
 }
 
 static inline const struct svga3d_surface_desc *
@@ -834,7 +837,7 @@ svga3dsurface_get_serialized_size(SVGA3dSurfaceFormat format,
 				  bool cubemap)
 {
 	const struct svga3d_surface_desc *desc = svga3dsurface_get_desc(format);
-	u32 total_size = 0;
+	u64 total_size = 0;
 	u32 mip;
 
 	for (mip = 0; mip < num_mip_levels; mip++) {
@@ -847,7 +850,7 @@ svga3dsurface_get_serialized_size(SVGA3dSurfaceFormat format,
 	if (cubemap)
 		total_size *= SVGA3D_MAX_SURFACE_FACES;
 
-	return total_size;
+	return (u32) min_t(u64, total_size, (u64) U32_MAX);
 }
 
 
diff --git a/drivers/gpu/drm/vmwgfx/svga_reg.h b/drivers/gpu/drm/vmwgfx/svga_reg.h
index 71defa4d2d75..11323dd5196f 100644
--- a/drivers/gpu/drm/vmwgfx/svga_reg.h
+++ b/drivers/gpu/drm/vmwgfx/svga_reg.h
@@ -169,10 +169,17 @@ enum {
    SVGA_REG_TRACES = 45,            /* Enable trace-based updates even when FIFO is on */
    SVGA_REG_GMRS_MAX_PAGES = 46,    /* Maximum number of 4KB pages for all GMRs */
    SVGA_REG_MEMORY_SIZE = 47,       /* Total dedicated device memory excluding FIFO */
+   SVGA_REG_COMMAND_LOW = 48,       /* Lower 32 bits and submits commands */
+   SVGA_REG_COMMAND_HIGH = 49,      /* Upper 32 bits of command buffer PA */
    SVGA_REG_MAX_PRIMARY_BOUNDING_BOX_MEM = 50,   /* Max primary memory */
    SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB = 51, /* Suggested limit on mob mem */
    SVGA_REG_DEV_CAP = 52,           /* Write dev cap index, read value */
-   SVGA_REG_TOP = 53,               /* Must be 1 more than the last register */
+   SVGA_REG_CMD_PREPEND_LOW = 53,
+   SVGA_REG_CMD_PREPEND_HIGH = 54,
+   SVGA_REG_SCREENTARGET_MAX_WIDTH = 55,
+   SVGA_REG_SCREENTARGET_MAX_HEIGHT = 56,
+   SVGA_REG_MOB_MAX_SIZE = 57,
+   SVGA_REG_TOP = 58,               /* Must be 1 more than the last register */
 
    SVGA_PALETTE_BASE = 1024,        /* Base of SVGA color map */
    /* Next 768 (== 256*3) registers exist for colormap */
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c
index 9426c53fb483..1e80152674b5 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_context.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_context.c
@@ -551,8 +551,7 @@ static int vmw_context_scrub_shader(struct vmw_ctx_bindinfo *bi, bool rebind)
 	cmd->header.size = sizeof(cmd->body);
 	cmd->body.cid = bi->ctx->id;
 	cmd->body.type = bi->i1.shader_type;
-	cmd->body.shid =
-		cpu_to_le32((rebind) ? bi->res->id : SVGA3D_INVALID_ID);
+	cmd->body.shid = ((rebind) ? bi->res->id : SVGA3D_INVALID_ID);
 	vmw_fifo_commit(dev_priv, sizeof(*cmd));
 
 	return 0;
@@ -585,8 +584,7 @@ static int vmw_context_scrub_render_target(struct vmw_ctx_bindinfo *bi,
 	cmd->header.size = sizeof(cmd->body);
 	cmd->body.cid = bi->ctx->id;
 	cmd->body.type = bi->i1.rt_type;
-	cmd->body.target.sid =
-		cpu_to_le32((rebind) ? bi->res->id : SVGA3D_INVALID_ID);
+	cmd->body.target.sid = ((rebind) ? bi->res->id : SVGA3D_INVALID_ID);
 	cmd->body.target.face = 0;
 	cmd->body.target.mipmap = 0;
 	vmw_fifo_commit(dev_priv, sizeof(*cmd));
@@ -628,8 +626,7 @@ static int vmw_context_scrub_texture(struct vmw_ctx_bindinfo *bi,
 	cmd->body.c.cid = bi->ctx->id;
 	cmd->body.s1.stage = bi->i1.texture_stage;
 	cmd->body.s1.name = SVGA3D_TS_BIND_TEXTURE;
-	cmd->body.s1.value =
-		cpu_to_le32((rebind) ? bi->res->id : SVGA3D_INVALID_ID);
+	cmd->body.s1.value = ((rebind) ? bi->res->id : SVGA3D_INVALID_ID);
 	vmw_fifo_commit(dev_priv, sizeof(*cmd));
 
 	return 0;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 3bdc0adc656d..0083cbf99edf 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -667,6 +667,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 		dev_priv->memory_size = 512*1024*1024;
 	}
 	dev_priv->max_mob_pages = 0;
+	dev_priv->max_mob_size = 0;
 	if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS) {
 		uint64_t mem_size =
 			vmw_read(dev_priv,
@@ -676,6 +677,8 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 		dev_priv->prim_bb_mem =
 			vmw_read(dev_priv,
 				 SVGA_REG_MAX_PRIMARY_BOUNDING_BOX_MEM);
+		dev_priv->max_mob_size =
+			vmw_read(dev_priv, SVGA_REG_MOB_MAX_SIZE);
 	} else
 		dev_priv->prim_bb_mem = dev_priv->vram_size;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index ecaa302a6154..9e4be1725985 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -386,6 +386,7 @@ struct vmw_private {
 	uint32_t max_gmr_ids;
 	uint32_t max_gmr_pages;
 	uint32_t max_mob_pages;
+	uint32_t max_mob_size;
 	uint32_t memory_size;
 	bool has_gmr;
 	bool has_mob;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
index 269b85cc875a..efb575a7996c 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
@@ -602,7 +602,7 @@ static int vmw_cmd_cid_check(struct vmw_private *dev_priv,
 {
 	struct vmw_cid_cmd {
 		SVGA3dCmdHeader header;
-		__le32 cid;
+		uint32_t cid;
 	} *cmd;
 
 	cmd = container_of(header, struct vmw_cid_cmd, header);
@@ -1835,7 +1835,7 @@ static int vmw_cmd_check_not_3d(struct vmw_private *dev_priv,
 	return 0;
 }
 
-static const struct vmw_cmd_entry const vmw_cmd_entries[SVGA_3D_CMD_MAX] = {
+static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = {
 	VMW_CMD_DEF(SVGA_3D_CMD_SURFACE_DEFINE, &vmw_cmd_invalid,
 		    false, false, false),
 	VMW_CMD_DEF(SVGA_3D_CMD_SURFACE_DESTROY, &vmw_cmd_invalid,
@@ -2032,6 +2032,9 @@ static int vmw_cmd_check(struct vmw_private *dev_priv,
 		goto out_invalid;
 
 	entry = &vmw_cmd_entries[cmd_id];
+	if (unlikely(!entry->func))
+		goto out_invalid;
+
 	if (unlikely(!entry->user_allow && !sw_context->kernel))
 		goto out_privileged;
 
@@ -2469,7 +2472,7 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 	if (dev_priv->has_mob) {
 		ret = vmw_rebind_contexts(sw_context);
 		if (unlikely(ret != 0))
-			goto out_err;
+			goto out_unlock_binding;
 	}
 
 	cmd = vmw_fifo_reserve(dev_priv, command_size);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
index f9881f9e62bd..47b70949bf3a 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
@@ -102,6 +102,9 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data,
 		vmw_fp->gb_aware = true;
 		param->value = dev_priv->max_mob_pages * PAGE_SIZE;
 		break;
+	case DRM_VMW_PARAM_MAX_MOB_SIZE:
+		param->value = dev_priv->max_mob_size;
+		break;
 	default:
 		DRM_ERROR("Illegal vmwgfx get param request: %d\n",
 			  param->param);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c
index 217d941b8176..ee3856578a12 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c
@@ -371,13 +371,13 @@ int vmw_shader_destroy_ioctl(struct drm_device *dev, void *data,
 					 TTM_REF_USAGE);
 }
 
-int vmw_shader_alloc(struct vmw_private *dev_priv,
-		     struct vmw_dma_buffer *buffer,
-		     size_t shader_size,
-		     size_t offset,
-		     SVGA3dShaderType shader_type,
-		     struct ttm_object_file *tfile,
-		     u32 *handle)
+static int vmw_shader_alloc(struct vmw_private *dev_priv,
+			    struct vmw_dma_buffer *buffer,
+			    size_t shader_size,
+			    size_t offset,
+			    SVGA3dShaderType shader_type,
+			    struct ttm_object_file *tfile,
+			    u32 *handle)
 {
 	struct vmw_user_shader *ushader;
 	struct vmw_resource *res, *tmp;
@@ -779,6 +779,8 @@ vmw_compat_shader_man_create(struct vmw_private *dev_priv)
 	int ret;
 
 	man = kzalloc(sizeof(*man), GFP_KERNEL);
+	if (man == NULL)
+		return ERR_PTR(-ENOMEM);
 
 	man->dev_priv = dev_priv;
 	INIT_LIST_HEAD(&man->list);
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 71727b6210ae..8f3dee097579 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -907,6 +907,9 @@ struct drm_mode_config {
 
 	/* whether async page flip is supported or not */
 	bool async_page_flip;
+
+	/* cursor size */
+	uint32_t cursor_width, cursor_height;
 };
 
 #define obj_to_crtc(x) container_of(x, struct drm_crtc, base)
diff --git a/include/drm/ttm/ttm_page_alloc.h b/include/drm/ttm/ttm_page_alloc.h
index d1f61bfe0ebe..49a828425fa2 100644
--- a/include/drm/ttm/ttm_page_alloc.h
+++ b/include/drm/ttm/ttm_page_alloc.h
@@ -29,6 +29,8 @@
 #include <drm/ttm/ttm_bo_driver.h>
 #include <drm/ttm/ttm_memory.h>
 
+struct device;
+
 /**
  * Initialize pool allocator.
  */
diff --git a/include/uapi/drm/drm.h b/include/uapi/drm/drm.h
index 3c9a833992e8..b06c8ed68707 100644
--- a/include/uapi/drm/drm.h
+++ b/include/uapi/drm/drm.h
@@ -619,6 +619,8 @@ struct drm_gem_open {
 #define  DRM_PRIME_CAP_EXPORT		0x2
 #define DRM_CAP_TIMESTAMP_MONOTONIC	0x6
 #define DRM_CAP_ASYNC_PAGE_FLIP		0x7
+#define DRM_CAP_CURSOR_WIDTH		0x8
+#define DRM_CAP_CURSOR_HEIGHT		0x9
 
 /** DRM_IOCTL_GET_CAP ioctl argument type */
 struct drm_get_cap {
diff --git a/include/uapi/drm/vmwgfx_drm.h b/include/uapi/drm/vmwgfx_drm.h
index 9971c560ed9a..87792a5fee3b 100644
--- a/include/uapi/drm/vmwgfx_drm.h
+++ b/include/uapi/drm/vmwgfx_drm.h
@@ -87,6 +87,7 @@
 #define DRM_VMW_PARAM_MAX_SURF_MEMORY  7
 #define DRM_VMW_PARAM_3D_CAPS_SIZE     8
 #define DRM_VMW_PARAM_MAX_MOB_MEMORY   9
+#define DRM_VMW_PARAM_MAX_MOB_SIZE     10
 
 /**
  * struct drm_vmw_getparam_arg