Merge branch 'v3.1-rc10' into for-3.2/core

Conflicts:
	block/blk-core.c
	include/linux/blkdev.h

Signed-off-by: Jens Axboe <axboe@kernel.dk>

19 files changed, 276 insertions, 245 deletions

diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 645b84b3d20..79e8ebc0530 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -115,6 +115,7 @@ static int radeon_dp_aux_native_write(struct radeon_connector *radeon_connector,
 	u8 msg[20];
 	int msg_bytes = send_bytes + 4;
 	u8 ack;
+	unsigned retry;
 
 	if (send_bytes > 16)
 		return -1;
@@ -125,20 +126,22 @@ static int radeon_dp_aux_native_write(struct radeon_connector *radeon_connector,
 	msg[3] = (msg_bytes << 4) | (send_bytes - 1);
 	memcpy(&msg[4], send, send_bytes);
 
-	while (1) {
+	for (retry = 0; retry < 4; retry++) {
 		ret = radeon_process_aux_ch(dig_connector->dp_i2c_bus,
 					    msg, msg_bytes, NULL, 0, delay, &ack);
-		if (ret < 0)
+		if (ret == -EBUSY)
+			continue;
+		else if (ret < 0)
 			return ret;
 		if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_ACK)
-			break;
+			return send_bytes;
 		else if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_DEFER)
 			udelay(400);
 		else
 			return -EIO;
 	}
 
-	return send_bytes;
+	return -EIO;
 }
 
 static int radeon_dp_aux_native_read(struct radeon_connector *radeon_connector,
@@ -149,26 +152,31 @@ static int radeon_dp_aux_native_read(struct radeon_connector *radeon_connector,
 	int msg_bytes = 4;
 	u8 ack;
 	int ret;
+	unsigned retry;
 
 	msg[0] = address;
 	msg[1] = address >> 8;
 	msg[2] = AUX_NATIVE_READ << 4;
 	msg[3] = (msg_bytes << 4) | (recv_bytes - 1);
 
-	while (1) {
+	for (retry = 0; retry < 4; retry++) {
 		ret = radeon_process_aux_ch(dig_connector->dp_i2c_bus,
 					    msg, msg_bytes, recv, recv_bytes, delay, &ack);
-		if (ret == 0)
-			return -EPROTO;
-		if (ret < 0)
+		if (ret == -EBUSY)
+			continue;
+		else if (ret < 0)
 			return ret;
 		if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_ACK)
 			return ret;
 		else if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_DEFER)
 			udelay(400);
+		else if (ret == 0)
+			return -EPROTO;
 		else
 			return -EIO;
 	}
+
+	return -EIO;
 }
 
 static void radeon_write_dpcd_reg(struct radeon_connector *radeon_connector,
@@ -232,7 +240,9 @@ int radeon_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode,
 	for (retry = 0; retry < 4; retry++) {
 		ret = radeon_process_aux_ch(auxch,
 					    msg, msg_bytes, reply, reply_bytes, 0, &ack);
-		if (ret < 0) {
+		if (ret == -EBUSY)
+			continue;
+		else if (ret < 0) {
 			DRM_DEBUG_KMS("aux_ch failed %d\n", ret);
 			return ret;
 		}
@@ -613,6 +623,18 @@ static bool radeon_dp_get_link_status(struct radeon_connector *radeon_connector,
 	return true;
 }
 
+bool radeon_dp_needs_link_train(struct radeon_connector *radeon_connector)
+{
+	u8 link_status[DP_LINK_STATUS_SIZE];
+	struct radeon_connector_atom_dig *dig = radeon_connector->con_priv;
+
+	if (!radeon_dp_get_link_status(radeon_connector, link_status))
+		return false;
+	if (dp_channel_eq_ok(link_status, dig->dp_lane_count))
+		return false;
+	return true;
+}
+
 struct radeon_dp_link_train_info {
 	struct radeon_device *rdev;
 	struct drm_encoder *encoder;
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 14dce9f2217..c4ffa14fb2f 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -41,6 +41,31 @@ static void evergreen_gpu_init(struct radeon_device *rdev);
 void evergreen_fini(struct radeon_device *rdev);
 static void evergreen_pcie_gen2_enable(struct radeon_device *rdev);
 
+void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev)
+{
+	u16 ctl, v;
+	int cap, err;
+
+	cap = pci_pcie_cap(rdev->pdev);
+	if (!cap)
+		return;
+
+	err = pci_read_config_word(rdev->pdev, cap + PCI_EXP_DEVCTL, &ctl);
+	if (err)
+		return;
+
+	v = (ctl & PCI_EXP_DEVCTL_READRQ) >> 12;
+
+	/* if bios or OS sets MAX_READ_REQUEST_SIZE to an invalid value, fix it
+	 * to avoid hangs or perfomance issues
+	 */
+	if ((v == 0) || (v == 6) || (v == 7)) {
+		ctl &= ~PCI_EXP_DEVCTL_READRQ;
+		ctl |= (2 << 12);
+		pci_write_config_word(rdev->pdev, cap + PCI_EXP_DEVCTL, ctl);
+	}
+}
+
 void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc)
 {
 	/* enable the pflip int */
@@ -743,7 +768,7 @@ static void evergreen_program_watermarks(struct radeon_device *rdev,
 		    !evergreen_average_bandwidth_vs_available_bandwidth(&wm) ||
 		    !evergreen_check_latency_hiding(&wm) ||
 		    (rdev->disp_priority == 2)) {
-			DRM_INFO("force priority to high\n");
+			DRM_DEBUG_KMS("force priority to high\n");
 			priority_a_cnt |= PRIORITY_ALWAYS_ON;
 			priority_b_cnt |= PRIORITY_ALWAYS_ON;
 		}
@@ -1357,6 +1382,7 @@ int evergreen_cp_resume(struct radeon_device *rdev)
 				 SOFT_RESET_PA |
 				 SOFT_RESET_SH |
 				 SOFT_RESET_VGT |
+				 SOFT_RESET_SPI |
 				 SOFT_RESET_SX));
 	RREG32(GRBM_SOFT_RESET);
 	mdelay(15);
@@ -1378,7 +1404,8 @@ int evergreen_cp_resume(struct radeon_device *rdev)
 	/* Initialize the ring buffer's read and write pointers */
 	WREG32(CP_RB_CNTL, tmp | RB_RPTR_WR_ENA);
 	WREG32(CP_RB_RPTR_WR, 0);
-	WREG32(CP_RB_WPTR, 0);
+	rdev->cp.wptr = 0;
+	WREG32(CP_RB_WPTR, rdev->cp.wptr);
 
 	/* set the wb address wether it's enabled or not */
 	WREG32(CP_RB_RPTR_ADDR,
@@ -1400,7 +1427,6 @@ int evergreen_cp_resume(struct radeon_device *rdev)
 	WREG32(CP_DEBUG, (1 << 27) | (1 << 28));
 
 	rdev->cp.rptr = RREG32(CP_RB_RPTR);
-	rdev->cp.wptr = RREG32(CP_RB_WPTR);
 
 	evergreen_cp_start(rdev);
 	rdev->cp.ready = true;
@@ -1564,48 +1590,6 @@ static u32 evergreen_get_tile_pipe_to_backend_map(struct radeon_device *rdev,
 	return backend_map;
 }
 
-static void evergreen_program_channel_remap(struct radeon_device *rdev)
-{
-	u32 tcp_chan_steer_lo, tcp_chan_steer_hi, mc_shared_chremap, tmp;
-
-	tmp = RREG32(MC_SHARED_CHMAP);
-	switch ((tmp & NOOFCHAN_MASK) >> NOOFCHAN_SHIFT) {
-	case 0:
-	case 1:
-	case 2:
-	case 3:
-	default:
-		/* default mapping */
-		mc_shared_chremap = 0x00fac688;
-		break;
-	}
-
-	switch (rdev->family) {
-	case CHIP_HEMLOCK:
-	case CHIP_CYPRESS:
-	case CHIP_BARTS:
-		tcp_chan_steer_lo = 0x54763210;
-		tcp_chan_steer_hi = 0x0000ba98;
-		break;
-	case CHIP_JUNIPER:
-	case CHIP_REDWOOD:
-	case CHIP_CEDAR:
-	case CHIP_PALM:
-	case CHIP_SUMO:
-	case CHIP_SUMO2:
-	case CHIP_TURKS:
-	case CHIP_CAICOS:
-	default:
-		tcp_chan_steer_lo = 0x76543210;
-		tcp_chan_steer_hi = 0x0000ba98;
-		break;
-	}
-
-	WREG32(TCP_CHAN_STEER_LO, tcp_chan_steer_lo);
-	WREG32(TCP_CHAN_STEER_HI, tcp_chan_steer_hi);
-	WREG32(MC_SHARED_CHREMAP, mc_shared_chremap);
-}
-
 static void evergreen_gpu_init(struct radeon_device *rdev)
 {
 	u32 cc_rb_backend_disable = 0;
@@ -1862,6 +1846,8 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 
 	WREG32(GRBM_CNTL, GRBM_READ_TIMEOUT(0xff));
 
+	evergreen_fix_pci_max_read_req_size(rdev);
+
 	cc_gc_shader_pipe_config = RREG32(CC_GC_SHADER_PIPE_CONFIG) & ~2;
 
 	cc_gc_shader_pipe_config |=
@@ -2050,8 +2036,6 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 	WREG32(DMIF_ADDR_CONFIG, gb_addr_config);
 	WREG32(HDP_ADDR_CONFIG, gb_addr_config);
 
-	evergreen_program_channel_remap(rdev);
-
 	num_shader_engines = ((RREG32(GB_ADDR_CONFIG) & NUM_SHADER_ENGINES(3)) >> 12) + 1;
 	grbm_gfx_index = INSTANCE_BROADCAST_WRITES;
 
@@ -3143,21 +3127,23 @@ int evergreen_suspend(struct radeon_device *rdev)
 }
 
 int evergreen_copy_blit(struct radeon_device *rdev,
-			uint64_t src_offset, uint64_t dst_offset,
-			unsigned num_pages, struct radeon_fence *fence)
+			uint64_t src_offset,
+			uint64_t dst_offset,
+			unsigned num_gpu_pages,
+			struct radeon_fence *fence)
 {
 	int r;
 
 	mutex_lock(&rdev->r600_blit.mutex);
 	rdev->r600_blit.vb_ib = NULL;
-	r = evergreen_blit_prepare_copy(rdev, num_pages * RADEON_GPU_PAGE_SIZE);
+	r = evergreen_blit_prepare_copy(rdev, num_gpu_pages * RADEON_GPU_PAGE_SIZE);
 	if (r) {
 		if (rdev->r600_blit.vb_ib)
 			radeon_ib_free(rdev, &rdev->r600_blit.vb_ib);
 		mutex_unlock(&rdev->r600_blit.mutex);
 		return r;
 	}
-	evergreen_kms_blit_copy(rdev, src_offset, dst_offset, num_pages * RADEON_GPU_PAGE_SIZE);
+	evergreen_kms_blit_copy(rdev, src_offset, dst_offset, num_gpu_pages * RADEON_GPU_PAGE_SIZE);
 	evergreen_blit_done_copy(rdev, fence);
 	mutex_unlock(&rdev->r600_blit.mutex);
 	return 0;
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index 44c4750f451..8c79ca97753 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -39,6 +39,7 @@ extern int evergreen_mc_wait_for_idle(struct radeon_device *rdev);
 extern void evergreen_mc_program(struct radeon_device *rdev);
 extern void evergreen_irq_suspend(struct radeon_device *rdev);
 extern int evergreen_mc_init(struct radeon_device *rdev);
+extern void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev);
 
 #define EVERGREEN_PFP_UCODE_SIZE 1120
 #define EVERGREEN_PM4_UCODE_SIZE 1376
@@ -568,36 +569,6 @@ static u32 cayman_get_tile_pipe_to_backend_map(struct radeon_device *rdev,
 	return backend_map;
 }
 
-static void cayman_program_channel_remap(struct radeon_device *rdev)
-{
-	u32 tcp_chan_steer_lo, tcp_chan_steer_hi, mc_shared_chremap, tmp;
-
-	tmp = RREG32(MC_SHARED_CHMAP);
-	switch ((tmp & NOOFCHAN_MASK) >> NOOFCHAN_SHIFT) {
-	case 0:
-	case 1:
-	case 2:
-	case 3:
-	default:
-		/* default mapping */
-		mc_shared_chremap = 0x00fac688;
-		break;
-	}
-
-	switch (rdev->family) {
-	case CHIP_CAYMAN:
-	default:
-		//tcp_chan_steer_lo = 0x54763210
-		tcp_chan_steer_lo = 0x76543210;
-		tcp_chan_steer_hi = 0x0000ba98;
-		break;
-	}
-
-	WREG32(TCP_CHAN_STEER_LO, tcp_chan_steer_lo);
-	WREG32(TCP_CHAN_STEER_HI, tcp_chan_steer_hi);
-	WREG32(MC_SHARED_CHREMAP, mc_shared_chremap);
-}
-
 static u32 cayman_get_disable_mask_per_asic(struct radeon_device *rdev,
 					    u32 disable_mask_per_se,
 					    u32 max_disable_mask_per_se,
@@ -669,6 +640,8 @@ static void cayman_gpu_init(struct radeon_device *rdev)
 
 	WREG32(GRBM_CNTL, GRBM_READ_TIMEOUT(0xff));
 
+	evergreen_fix_pci_max_read_req_size(rdev);
+
 	mc_shared_chmap = RREG32(MC_SHARED_CHMAP);
 	mc_arb_ramcfg = RREG32(MC_ARB_RAMCFG);
 
@@ -839,8 +812,6 @@ static void cayman_gpu_init(struct radeon_device *rdev)
 	WREG32(DMIF_ADDR_CONFIG, gb_addr_config);
 	WREG32(HDP_ADDR_CONFIG, gb_addr_config);
 
-	cayman_program_channel_remap(rdev);
-
 	/* primary versions */
 	WREG32(CC_RB_BACKEND_DISABLE, cc_rb_backend_disable);
 	WREG32(CC_SYS_RB_BACKEND_DISABLE, cc_rb_backend_disable);
@@ -1159,6 +1130,7 @@ int cayman_cp_resume(struct radeon_device *rdev)
 				 SOFT_RESET_PA |
 				 SOFT_RESET_SH |
 				 SOFT_RESET_VGT |
+				 SOFT_RESET_SPI |
 				 SOFT_RESET_SX));
 	RREG32(GRBM_SOFT_RESET);
 	mdelay(15);
@@ -1183,7 +1155,8 @@ int cayman_cp_resume(struct radeon_device *rdev)
 
 	/* Initialize the ring buffer's read and write pointers */
 	WREG32(CP_RB0_CNTL, tmp | RB_RPTR_WR_ENA);
-	WREG32(CP_RB0_WPTR, 0);
+	rdev->cp.wptr = 0;
+	WREG32(CP_RB0_WPTR, rdev->cp.wptr);
 
 	/* set the wb address wether it's enabled or not */
 	WREG32(CP_RB0_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC);
@@ -1203,7 +1176,6 @@ int cayman_cp_resume(struct radeon_device *rdev)
 	WREG32(CP_RB0_BASE, rdev->cp.gpu_addr >> 8);
 
 	rdev->cp.rptr = RREG32(CP_RB0_RPTR);
-	rdev->cp.wptr = RREG32(CP_RB0_WPTR);
 
 	/* ring1  - compute only */
 	/* Set ring buffer size */
@@ -1216,7 +1188,8 @@ int cayman_cp_resume(struct radeon_device *rdev)
 
 	/* Initialize the ring buffer's read and write pointers */
 	WREG32(CP_RB1_CNTL, tmp | RB_RPTR_WR_ENA);
-	WREG32(CP_RB1_WPTR, 0);
+	rdev->cp1.wptr = 0;
+	WREG32(CP_RB1_WPTR, rdev->cp1.wptr);
 
 	/* set the wb address wether it's enabled or not */
 	WREG32(CP_RB1_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP1_RPTR_OFFSET) & 0xFFFFFFFC);
@@ -1228,7 +1201,6 @@ int cayman_cp_resume(struct radeon_device *rdev)
 	WREG32(CP_RB1_BASE, rdev->cp1.gpu_addr >> 8);
 
 	rdev->cp1.rptr = RREG32(CP_RB1_RPTR);
-	rdev->cp1.wptr = RREG32(CP_RB1_WPTR);
 
 	/* ring2 - compute only */
 	/* Set ring buffer size */
@@ -1241,7 +1213,8 @@ int cayman_cp_resume(struct radeon_device *rdev)
 
 	/* Initialize the ring buffer's read and write pointers */
 	WREG32(CP_RB2_CNTL, tmp | RB_RPTR_WR_ENA);
-	WREG32(CP_RB2_WPTR, 0);
+	rdev->cp2.wptr = 0;
+	WREG32(CP_RB2_WPTR, rdev->cp2.wptr);
 
 	/* set the wb address wether it's enabled or not */
 	WREG32(CP_RB2_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP2_RPTR_OFFSET) & 0xFFFFFFFC);
@@ -1253,7 +1226,6 @@ int cayman_cp_resume(struct radeon_device *rdev)
 	WREG32(CP_RB2_BASE, rdev->cp2.gpu_addr >> 8);
 
 	rdev->cp2.rptr = RREG32(CP_RB2_RPTR);
-	rdev->cp2.wptr = RREG32(CP_RB2_WPTR);
 
 	/* start the rings */
 	cayman_cp_start(rdev);
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index f2204cb1ccd..7fcdbbbf297 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -721,11 +721,11 @@ void r100_fence_ring_emit(struct radeon_device *rdev,
 int r100_copy_blit(struct radeon_device *rdev,
 		   uint64_t src_offset,
 		   uint64_t dst_offset,
-		   unsigned num_pages,
+		   unsigned num_gpu_pages,
 		   struct radeon_fence *fence)
 {
 	uint32_t cur_pages;
-	uint32_t stride_bytes = PAGE_SIZE;
+	uint32_t stride_bytes = RADEON_GPU_PAGE_SIZE;
 	uint32_t pitch;
 	uint32_t stride_pixels;
 	unsigned ndw;
@@ -737,7 +737,7 @@ int r100_copy_blit(struct radeon_device *rdev,
 	/* radeon pitch is /64 */
 	pitch = stride_bytes / 64;
 	stride_pixels = stride_bytes / 4;
-	num_loops = DIV_ROUND_UP(num_pages, 8191);
+	num_loops = DIV_ROUND_UP(num_gpu_pages, 8191);
 
 	/* Ask for enough room for blit + flush + fence */
 	ndw = 64 + (10 * num_loops);
@@ -746,12 +746,12 @@ int r100_copy_blit(struct radeon_device *rdev,
 		DRM_ERROR("radeon: moving bo (%d) asking for %u dw.\n", r, ndw);
 		return -EINVAL;
 	}
-	while (num_pages > 0) {
-		cur_pages = num_pages;
+	while (num_gpu_pages > 0) {
+		cur_pages = num_gpu_pages;
 		if (cur_pages > 8191) {
 			cur_pages = 8191;
 		}
-		num_pages -= cur_pages;
+		num_gpu_pages -= cur_pages;
 
 		/* pages are in Y direction - height
 		   page width in X direction - width */
@@ -773,8 +773,8 @@ int r100_copy_blit(struct radeon_device *rdev,
 		radeon_ring_write(rdev, (0x1fff) | (0x1fff << 16));
 		radeon_ring_write(rdev, 0);
 		radeon_ring_write(rdev, (0x1fff) | (0x1fff << 16));
-		radeon_ring_write(rdev, num_pages);
-		radeon_ring_write(rdev, num_pages);
+		radeon_ring_write(rdev, num_gpu_pages);
+		radeon_ring_write(rdev, num_gpu_pages);
 		radeon_ring_write(rdev, cur_pages | (stride_pixels << 16));
 	}
 	radeon_ring_write(rdev, PACKET0(RADEON_DSTCACHE_CTLSTAT, 0));
@@ -990,7 +990,8 @@ int r100_cp_init(struct radeon_device *rdev, unsigned ring_size)
 	/* Force read & write ptr to 0 */
 	WREG32(RADEON_CP_RB_CNTL, tmp | RADEON_RB_RPTR_WR_ENA | RADEON_RB_NO_UPDATE);
 	WREG32(RADEON_CP_RB_RPTR_WR, 0);
-	WREG32(RADEON_CP_RB_WPTR, 0);
+	rdev->cp.wptr = 0;
+	WREG32(RADEON_CP_RB_WPTR, rdev->cp.wptr);
 
 	/* set the wb address whether it's enabled or not */
 	WREG32(R_00070C_CP_RB_RPTR_ADDR,
@@ -1007,9 +1008,6 @@ int r100_cp_init(struct radeon_device *rdev, unsigned ring_size)
 	WREG32(RADEON_CP_RB_CNTL, tmp);
 	udelay(10);
 	rdev->cp.rptr = RREG32(RADEON_CP_RB_RPTR);
-	rdev->cp.wptr = RREG32(RADEON_CP_RB_WPTR);
-	/* protect against crazy HW on resume */
-	rdev->cp.wptr &= rdev->cp.ptr_mask;
 	/* Set cp mode to bus mastering & enable cp*/
 	WREG32(RADEON_CP_CSQ_MODE,
 	       REG_SET(RADEON_INDIRECT2_START, indirect2_start) |
diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c
index f2405830041..a1f3ba063c2 100644
--- a/drivers/gpu/drm/radeon/r200.c
+++ b/drivers/gpu/drm/radeon/r200.c
@@ -84,7 +84,7 @@ static int r200_get_vtx_size_0(uint32_t vtx_fmt_0)
 int r200_copy_dma(struct radeon_device *rdev,
 		  uint64_t src_offset,
 		  uint64_t dst_offset,
-		  unsigned num_pages,
+		  unsigned num_gpu_pages,
 		  struct radeon_fence *fence)
 {
 	uint32_t size;
@@ -93,7 +93,7 @@ int r200_copy_dma(struct radeon_device *rdev,
 	int r = 0;
 
 	/* radeon pitch is /64 */
-	size = num_pages << PAGE_SHIFT;
+	size = num_gpu_pages << RADEON_GPU_PAGE_SHIFT;
 	num_loops = DIV_ROUND_UP(size, 0x1FFFFF);
 	r = radeon_ring_lock(rdev, num_loops * 4 + 64);
 	if (r) {
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index aa5571b73aa..720dd99163f 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -2209,7 +2209,8 @@ int r600_cp_resume(struct radeon_device *rdev)
 	/* Initialize the ring buffer's read and write pointers */
 	WREG32(CP_RB_CNTL, tmp | RB_RPTR_WR_ENA);
 	WREG32(CP_RB_RPTR_WR, 0);
-	WREG32(CP_RB_WPTR, 0);
+	rdev->cp.wptr = 0;
+	WREG32(CP_RB_WPTR, rdev->cp.wptr);
 
 	/* set the wb address whether it's enabled or not */
 	WREG32(CP_RB_RPTR_ADDR,
@@ -2231,7 +2232,6 @@ int r600_cp_resume(struct radeon_device *rdev)
 	WREG32(CP_DEBUG, (1 << 27) | (1 << 28));
 
 	rdev->cp.rptr = RREG32(CP_RB_RPTR);
-	rdev->cp.wptr = RREG32(CP_RB_WPTR);
 
 	r600_cp_start(rdev);
 	rdev->cp.ready = true;
@@ -2353,21 +2353,23 @@ void r600_fence_ring_emit(struct radeon_device *rdev,
 }
 
 int r600_copy_blit(struct radeon_device *rdev,
-		   uint64_t src_offset, uint64_t dst_offset,
-		   unsigned num_pages, struct radeon_fence *fence)
+		   uint64_t src_offset,
+		   uint64_t dst_offset,
+		   unsigned num_gpu_pages,
+		   struct radeon_fence *fence)
 {
 	int r;
 
 	mutex_lock(&rdev->r600_blit.mutex);
 	rdev->r600_blit.vb_ib = NULL;
-	r = r600_blit_prepare_copy(rdev, num_pages * RADEON_GPU_PAGE_SIZE);
+	r = r600_blit_prepare_copy(rdev, num_gpu_pages * RADEON_GPU_PAGE_SIZE);
 	if (r) {
 		if (rdev->r600_blit.vb_ib)
 			radeon_ib_free(rdev, &rdev->r600_blit.vb_ib);
 		mutex_unlock(&rdev->r600_blit.mutex);
 		return r;
 	}
-	r600_kms_blit_copy(rdev, src_offset, dst_offset, num_pages * RADEON_GPU_PAGE_SIZE);
+	r600_kms_blit_copy(rdev, src_offset, dst_offset, num_gpu_pages * RADEON_GPU_PAGE_SIZE);
 	r600_blit_done_copy(rdev, fence);
 	mutex_unlock(&rdev->r600_blit.mutex);
 	return 0;
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 32807baf55e..c1e056b35b2 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -322,6 +322,7 @@ union radeon_gart_table {
 
 #define RADEON_GPU_PAGE_SIZE 4096
 #define RADEON_GPU_PAGE_MASK (RADEON_GPU_PAGE_SIZE - 1)
+#define RADEON_GPU_PAGE_SHIFT 12
 
 struct radeon_gart {
 	dma_addr_t			table_addr;
@@ -914,17 +915,17 @@ struct radeon_asic {
 	int (*copy_blit)(struct radeon_device *rdev,
 			 uint64_t src_offset,
 			 uint64_t dst_offset,
-			 unsigned num_pages,
+			 unsigned num_gpu_pages,
 			 struct radeon_fence *fence);
 	int (*copy_dma)(struct radeon_device *rdev,
 			uint64_t src_offset,
 			uint64_t dst_offset,
-			unsigned num_pages,
+			unsigned num_gpu_pages,
 			struct radeon_fence *fence);
 	int (*copy)(struct radeon_device *rdev,
 		    uint64_t src_offset,
 		    uint64_t dst_offset,
-		    unsigned num_pages,
+		    unsigned num_gpu_pages,
 		    struct radeon_fence *fence);
 	uint32_t (*get_engine_clock)(struct radeon_device *rdev);
 	void (*set_engine_clock)(struct radeon_device *rdev, uint32_t eng_clock);
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 3d7a0d7c6a9..3dedaa07aac 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -75,7 +75,7 @@ uint32_t r100_pll_rreg(struct radeon_device *rdev, uint32_t reg);
 int r100_copy_blit(struct radeon_device *rdev,
 		   uint64_t src_offset,
 		   uint64_t dst_offset,
-		   unsigned num_pages,
+		   unsigned num_gpu_pages,
 		   struct radeon_fence *fence);
 int r100_set_surface_reg(struct radeon_device *rdev, int reg,
 			 uint32_t tiling_flags, uint32_t pitch,
@@ -143,7 +143,7 @@ extern void r100_post_page_flip(struct radeon_device *rdev, int crtc);
 extern int r200_copy_dma(struct radeon_device *rdev,
 			 uint64_t src_offset,
 			 uint64_t dst_offset,
-			 unsigned num_pages,
+			 unsigned num_gpu_pages,
 			 struct radeon_fence *fence);
 void r200_set_safe_registers(struct radeon_device *rdev);
 
@@ -311,7 +311,7 @@ void r600_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
 int r600_ring_test(struct radeon_device *rdev);
 int r600_copy_blit(struct radeon_device *rdev,
 		   uint64_t src_offset, uint64_t dst_offset,
-		   unsigned num_pages, struct radeon_fence *fence);
+		   unsigned num_gpu_pages, struct radeon_fence *fence);
 void r600_hpd_init(struct radeon_device *rdev);
 void r600_hpd_fini(struct radeon_device *rdev);
 bool r600_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd);
@@ -403,7 +403,7 @@ void evergreen_bandwidth_update(struct radeon_device *rdev);
 void evergreen_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
 int evergreen_copy_blit(struct radeon_device *rdev,
 			uint64_t src_offset, uint64_t dst_offset,
-			unsigned num_pages, struct radeon_fence *fence);
+			unsigned num_gpu_pages, struct radeon_fence *fence);
 void evergreen_hpd_init(struct radeon_device *rdev);
 void evergreen_hpd_fini(struct radeon_device *rdev);
 bool evergreen_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd);
diff --git a/drivers/gpu/drm/radeon/radeon_clocks.c b/drivers/gpu/drm/radeon/radeon_clocks.c
index dcd0863e31a..b6e18c8db9f 100644
--- a/drivers/gpu/drm/radeon/radeon_clocks.c
+++ b/drivers/gpu/drm/radeon/radeon_clocks.c
@@ -219,6 +219,9 @@ void radeon_get_clock_info(struct drm_device *dev)
 		} else {
 			DRM_INFO("Using generic clock info\n");
 
+			/* may need to be per card */
+			rdev->clock.max_pixel_clock = 35000;
+
 			if (rdev->flags & RADEON_IS_IGP) {
 				p1pll->reference_freq = 1432;
 				p2pll->reference_freq = 1432;
diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c
index e0138b674ac..63675241c7f 100644
--- a/drivers/gpu/drm/radeon/radeon_combios.c
+++ b/drivers/gpu/drm/radeon/radeon_combios.c
@@ -3298,6 +3298,14 @@ void radeon_combios_asic_init(struct drm_device *dev)
 	    rdev->pdev->subsystem_device == 0x30a4)
 		return;
 
+	/* quirk for rs4xx Compaq Presario V5245EU laptop to make it resume
+	 * - it hangs on resume inside the dynclk 1 table.
+	 */
+	if (rdev->family == CHIP_RS480 &&
+	    rdev->pdev->subsystem_vendor == 0x103c &&
+	    rdev->pdev->subsystem_device == 0x30ae)
+		return;
+
 	/* DYN CLK 1 */
 	table = combios_get_table_offset(dev, COMBIOS_DYN_CLK_1_TABLE);
 	if (table)
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index 6d6b5f16bc0..449c3d8c683 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -60,18 +60,20 @@ void radeon_connector_hotplug(struct drm_connector *connector)
 
 	radeon_hpd_set_polarity(rdev, radeon_connector->hpd.hpd);
 
-	/* powering up/down the eDP panel generates hpd events which
-	 * can interfere with modesetting.
-	 */
-	if (connector->connector_type == DRM_MODE_CONNECTOR_eDP)
+	/* if the connector is already off, don't turn it back on */
+	if (connector->dpms != DRM_MODE_DPMS_ON)
 		return;
 
-	/* pre-r600 did not always have the hpd pins mapped accurately to connectors */
-	if (rdev->family >= CHIP_R600) {
-		if (radeon_hpd_sense(rdev, radeon_connector->hpd.hpd))
-			drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
-		else
+	/* just deal with DP (not eDP) here. */
+	if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) {
+		int saved_dpms = connector->dpms;
+
+		/* Only turn off the display it it's physically disconnected */
+		if (!radeon_hpd_sense(rdev, radeon_connector->hpd.hpd))
 			drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
+		else if (radeon_dp_needs_link_train(radeon_connector))
+			drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
+		connector->dpms = saved_dpms;
 	}
 }
 
@@ -464,6 +466,16 @@ static bool radeon_connector_needs_extended_probe(struct radeon_device *dev,
 		    (supported_device == ATOM_DEVICE_DFP2_SUPPORT))
 			return true;
 	}
+	/* TOSHIBA Satellite L300D with ATI Mobility Radeon x1100
+	 * (RS690M) sends data to i2c bus for a HDMI connector that
+	 * is not implemented */
+	if ((dev->pdev->device == 0x791f) &&
+	    (dev->pdev->subsystem_vendor == 0x1179) &&
+	    (dev->pdev->subsystem_device == 0xff68)) {
+		if ((connector_type == DRM_MODE_CONNECTOR_HDMIA) &&
+		    (supported_device == ATOM_DEVICE_DFP2_SUPPORT))
+			return true;
+	}
 
 	/* Default: no EDID header probe required for DDC probing */
 	return false;
@@ -474,11 +486,19 @@ static void radeon_fixup_lvds_native_mode(struct drm_encoder *encoder,
 {
 	struct radeon_encoder *radeon_encoder =	to_radeon_encoder(encoder);
 	struct drm_display_mode *native_mode = &radeon_encoder->native_mode;
+	struct drm_display_mode *t, *mode;
+
+	/* If the EDID preferred mode doesn't match the native mode, use it */
+	list_for_each_entry_safe(mode, t, &connector->probed_modes, head) {
+		if (mode->type & DRM_MODE_TYPE_PREFERRED) {
+			if (mode->hdisplay != native_mode->hdisplay ||
+			    mode->vdisplay != native_mode->vdisplay)
+				memcpy(native_mode, mode, sizeof(*mode));
+		}
+	}
 
 	/* Try to get native mode details from EDID if necessary */
 	if (!native_mode->clock) {
-		struct drm_display_mode *t, *mode;
-
 		list_for_each_entry_safe(mode, t, &connector->probed_modes, head) {
 			if (mode->hdisplay == native_mode->hdisplay &&
 			    mode->vdisplay == native_mode->vdisplay) {
@@ -489,6 +509,7 @@ static void radeon_fixup_lvds_native_mode(struct drm_encoder *encoder,
 			}
 		}
 	}
+
 	if (!native_mode->clock) {
 		DRM_DEBUG_KMS("No LVDS native mode details, disabling RMX\n");
 		radeon_encoder->rmx_type = RMX_OFF;
@@ -1276,12 +1297,24 @@ radeon_dp_detect(struct drm_connector *connector, bool force)
 		if (!radeon_dig_connector->edp_on)
 			atombios_set_edp_panel_power(connector,
 						     ATOM_TRANSMITTER_ACTION_POWER_OFF);
-	} else {
-		/* need to setup ddc on the bridge */
-		if (radeon_connector_encoder_is_dp_bridge(connector)) {
-			if (encoder)
-				radeon_atom_ext_encoder_setup_ddc(encoder);
+	} else if (radeon_connector_encoder_is_dp_bridge(connector)) {
+		/* DP bridges are always DP */
+		radeon_dig_connector->dp_sink_type = CONNECTOR_OBJECT_ID_DISPLAYPORT;
+		/* get the DPCD from the bridge */
+		radeon_dp_getdpcd(radeon_connector);
+
+		if (encoder) {
+			/* setup ddc on the bridge */
+			radeon_atom_ext_encoder_setup_ddc(encoder);
+			if (radeon_ddc_probe(radeon_connector,
+					     radeon_connector->requires_extended_probe)) /* try DDC */
+				ret = connector_status_connected;
+			else if (radeon_connector->dac_load_detect) { /* try load detection */
+				struct drm_encoder_helper_funcs *encoder_funcs = encoder->helper_private;
+				ret = encoder_funcs->detect(encoder, connector);
+			}
 		}
+	} else {
 		radeon_dig_connector->dp_sink_type = radeon_dp_getsinktype(radeon_connector);
 		if (radeon_hpd_sense(rdev, radeon_connector->hpd.hpd)) {
 			ret = connector_status_connected;
@@ -1297,16 +1330,6 @@ radeon_dp_detect(struct drm_connector *connector, bool force)
 					ret = connector_status_connected;
 			}
 		}
-
-		if ((ret == connector_status_disconnected) &&
-		    radeon_connector->dac_load_detect) {
-			struct drm_encoder *encoder = radeon_best_single_encoder(connector);
-			struct drm_encoder_helper_funcs *encoder_funcs;
-			if (encoder) {
-				encoder_funcs = encoder->helper_private;
-				ret = encoder_funcs->detect(encoder, connector);
-			}
-		}
 	}
 
 	radeon_connector_update_scratch_regs(connector, ret);
diff --git a/drivers/gpu/drm/radeon/radeon_cursor.c b/drivers/gpu/drm/radeon/radeon_cursor.c
index 3189a7efb2e..fde25c0d65a 100644
--- a/drivers/gpu/drm/radeon/radeon_cursor.c
+++ b/drivers/gpu/drm/radeon/radeon_cursor.c
@@ -208,23 +208,25 @@ int radeon_crtc_cursor_move(struct drm_crtc *crtc,
 	int xorigin = 0, yorigin = 0;
 	int w = radeon_crtc->cursor_width;
 
-	if (x < 0)
-		xorigin = -x + 1;
-	if (y < 0)
-		yorigin = -y + 1;
-	if (xorigin >= CURSOR_WIDTH)
-		xorigin = CURSOR_WIDTH - 1;
-	if (yorigin >= CURSOR_HEIGHT)
-		yorigin = CURSOR_HEIGHT - 1;
-
 	if (ASIC_IS_AVIVO(rdev)) {
-		int i = 0;
-		struct drm_crtc *crtc_p;
-
 		/* avivo cursor are offset into the total surface */
 		x += crtc->x;
 		y += crtc->y;
-		DRM_DEBUG("x %d y %d c->x %d c->y %d\n", x, y, crtc->x, crtc->y);
+	}
+	DRM_DEBUG("x %d y %d c->x %d c->y %d\n", x, y, crtc->x, crtc->y);
+
+	if (x < 0) {
+		xorigin = min(-x, CURSOR_WIDTH - 1);
+		x = 0;
+	}
+	if (y < 0) {
+		yorigin = min(-y, CURSOR_HEIGHT - 1);
+		y = 0;
+	}
+
+	if (ASIC_IS_AVIVO(rdev)) {
+		int i = 0;
+		struct drm_crtc *crtc_p;
 
 		/* avivo cursor image can't end on 128 pixel boundary or
 		 * go past the end of the frame if both crtcs are enabled
@@ -253,16 +255,12 @@ int radeon_crtc_cursor_move(struct drm_crtc *crtc,
 
 	radeon_lock_cursor(crtc, true);
 	if (ASIC_IS_DCE4(rdev)) {
-		WREG32(EVERGREEN_CUR_POSITION + radeon_crtc->crtc_offset,
-		       ((xorigin ? 0 : x) << 16) |
-		       (yorigin ? 0 : y));
+		WREG32(EVERGREEN_CUR_POSITION + radeon_crtc->crtc_offset, (x << 16) | y);
 		WREG32(EVERGREEN_CUR_HOT_SPOT + radeon_crtc->crtc_offset, (xorigin << 16) | yorigin);
 		WREG32(EVERGREEN_CUR_SIZE + radeon_crtc->crtc_offset,
 		       ((w - 1) << 16) | (radeon_crtc->cursor_height - 1));
 	} else if (ASIC_IS_AVIVO(rdev)) {
-		WREG32(AVIVO_D1CUR_POSITION + radeon_crtc->crtc_offset,
-			     ((xorigin ? 0 : x) << 16) |
-			     (yorigin ? 0 : y));
+		WREG32(AVIVO_D1CUR_POSITION + radeon_crtc->crtc_offset, (x << 16) | y);
 		WREG32(AVIVO_D1CUR_HOT_SPOT + radeon_crtc->crtc_offset, (xorigin << 16) | yorigin);
 		WREG32(AVIVO_D1CUR_SIZE + radeon_crtc->crtc_offset,
 		       ((w - 1) << 16) | (radeon_crtc->cursor_height - 1));
@@ -276,8 +274,8 @@ int radeon_crtc_cursor_move(struct drm_crtc *crtc,
 			| yorigin));
 		WREG32(RADEON_CUR_HORZ_VERT_POSN + radeon_crtc->crtc_offset,
 		       (RADEON_CUR_LOCK
-			| ((xorigin ? 0 : x) << 16)
-			| (yorigin ? 0 : y)));
+			| (x << 16)
+			| y));
 		/* offset is from DISP(2)_BASE_ADDRESS */
 		WREG32(RADEON_CUR_OFFSET + radeon_crtc->crtc_offset, (radeon_crtc->legacy_cursor_offset +
 								      (yorigin * 256)));
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 440e6ecccc4..b51e15725c6 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -32,6 +32,7 @@
 #include <drm/radeon_drm.h>
 #include <linux/vgaarb.h>
 #include <linux/vga_switcheroo.h>
+#include <linux/efi.h>
 #include "radeon_reg.h"
 #include "radeon.h"
 #include "atom.h"
@@ -300,6 +301,8 @@ void radeon_vram_location(struct radeon_device *rdev, struct radeon_mc *mc, u64
 		mc->mc_vram_size = mc->aper_size;
 	}
 	mc->vram_end = mc->vram_start + mc->mc_vram_size - 1;
+	if (radeon_vram_limit && radeon_vram_limit < mc->real_vram_size)
+		mc->real_vram_size = radeon_vram_limit;
 	dev_info(rdev->dev, "VRAM: %lluM 0x%016llX - 0x%016llX (%lluM used)\n",
 			mc->mc_vram_size >> 20, mc->vram_start,
 			mc->vram_end, mc->real_vram_size >> 20);
@@ -348,6 +351,9 @@ bool radeon_card_posted(struct radeon_device *rdev)
 {
 	uint32_t reg;
 
+	if (efi_enabled && rdev->pdev->subsystem_vendor == PCI_VENDOR_ID_APPLE)
+		return false;
+
 	/* first check CRTCs */
 	if (ASIC_IS_DCE41(rdev)) {
 		reg = RREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET) |
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 1a858944e4f..6adb3e58aff 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -473,8 +473,8 @@ pflip_cleanup:
 	spin_lock_irqsave(&dev->event_lock, flags);
 	radeon_crtc->unpin_work = NULL;
 unlock_free:
-	drm_gem_object_unreference_unlocked(old_radeon_fb->obj);
 	spin_unlock_irqrestore(&dev->event_lock, flags);
+	drm_gem_object_unreference_unlocked(old_radeon_fb->obj);
 	radeon_fence_unref(&work->fence);
 	kfree(work);
 
@@ -707,16 +707,21 @@ int radeon_ddc_get_modes(struct radeon_connector *radeon_connector)
 		radeon_router_select_ddc_port(radeon_connector);
 
 	if ((radeon_connector->base.connector_type == DRM_MODE_CONNECTOR_DisplayPort) ||
-	    (radeon_connector->base.connector_type == DRM_MODE_CONNECTOR_eDP)) {
+	    (radeon_connector->base.connector_type == DRM_MODE_CONNECTOR_eDP) ||
+	    radeon_connector_encoder_is_dp_bridge(&radeon_connector->base)) {
 		struct radeon_connector_atom_dig *dig = radeon_connector->con_priv;
+
 		if ((dig->dp_sink_type == CONNECTOR_OBJECT_ID_DISPLAYPORT ||
 		     dig->dp_sink_type == CONNECTOR_OBJECT_ID_eDP) && dig->dp_i2c_bus)
-			radeon_connector->edid = drm_get_edid(&radeon_connector->base, &dig->dp_i2c_bus->adapter);
-	}
-	if (!radeon_connector->ddc_bus)
-		return -1;
-	if (!radeon_connector->edid) {
-		radeon_connector->edid = drm_get_edid(&radeon_connector->base, &radeon_connector->ddc_bus->adapter);
+			radeon_connector->edid = drm_get_edid(&radeon_connector->base,
+							      &dig->dp_i2c_bus->adapter);
+		else if (radeon_connector->ddc_bus && !radeon_connector->edid)
+			radeon_connector->edid = drm_get_edid(&radeon_connector->base,
+							      &radeon_connector->ddc_bus->adapter);
+	} else {
+		if (radeon_connector->ddc_bus && !radeon_connector->edid)
+			radeon_connector->edid = drm_get_edid(&radeon_connector->base,
+							      &radeon_connector->ddc_bus->adapter);
 	}
 
 	if (!radeon_connector->edid) {
diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c
index b293487e5aa..8a171b21b45 100644
--- a/drivers/gpu/drm/radeon/radeon_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_encoders.c
@@ -1507,7 +1507,14 @@ radeon_atom_encoder_dpms(struct drm_encoder *encoder, int mode)
 		switch (mode) {
 		case DRM_MODE_DPMS_ON:
 			args.ucAction = ATOM_ENABLE;
-			atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
+			/* workaround for DVOOutputControl on some RS690 systems */
+			if (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_DDI) {
+				u32 reg = RREG32(RADEON_BIOS_3_SCRATCH);
+				WREG32(RADEON_BIOS_3_SCRATCH, reg & ~ATOM_S3_DFP2I_ACTIVE);
+				atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
+				WREG32(RADEON_BIOS_3_SCRATCH, reg);
+			} else
+				atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
 			if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) {
 				args.ucAction = ATOM_LCD_BLON;
 				atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
@@ -1748,9 +1755,12 @@ static int radeon_atom_pick_dig_encoder(struct drm_encoder *encoder)
 	/* DCE4/5 */
 	if (ASIC_IS_DCE4(rdev)) {
 		dig = radeon_encoder->enc_priv;
-		if (ASIC_IS_DCE41(rdev))
-			return radeon_crtc->crtc_id;
-		else {
+		if (ASIC_IS_DCE41(rdev)) {
+			if (dig->linkb)
+				return 1;
+			else
+				return 0;
+		} else {
 			switch (radeon_encoder->encoder_id) {
 			case ENCODER_OBJECT_ID_INTERNAL_UNIPHY:
 				if (dig->linkb)
@@ -2323,6 +2333,9 @@ radeon_add_atom_encoder(struct drm_device *dev,
 	default:
 		encoder->possible_crtcs = 0x3;
 		break;
+	case 4:
+		encoder->possible_crtcs = 0xf;
+		break;
 	case 6:
 		encoder->possible_crtcs = 0x3f;
 		break;
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index d09031c03e2..68820f5f630 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -479,6 +479,7 @@ extern void radeon_dp_set_link_config(struct drm_connector *connector,
 				      struct drm_display_mode *mode);
 extern void radeon_dp_link_train(struct drm_encoder *encoder,
 				 struct drm_connector *connector);
+extern bool radeon_dp_needs_link_train(struct radeon_connector *radeon_connector);
 extern u8 radeon_dp_getsinktype(struct radeon_connector *radeon_connector);
 extern bool radeon_dp_getdpcd(struct radeon_connector *radeon_connector);
 extern void atombios_dig_encoder_setup(struct drm_encoder *encoder, int action, int panel_mode);
diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c
index dee4a0c1b4b..602fa3541c4 100644
--- a/drivers/gpu/drm/radeon/radeon_test.c
+++ b/drivers/gpu/drm/radeon/radeon_test.c
@@ -40,10 +40,14 @@ void radeon_test_moves(struct radeon_device *rdev)
 	size = 1024 * 1024;
 
 	/* Number of tests =
-	 * (Total GTT - IB pool - writeback page - ring buffer) / test size
+	 * (Total GTT - IB pool - writeback page - ring buffers) / test size
 	 */
-	n = ((u32)(rdev->mc.gtt_size - RADEON_IB_POOL_SIZE*64*1024 - RADEON_GPU_PAGE_SIZE -
-	     rdev->cp.ring_size)) / size;
+	n = rdev->mc.gtt_size - RADEON_IB_POOL_SIZE*64*1024 - rdev->cp.ring_size;
+	if (rdev->wb.wb_obj)
+		n -= RADEON_GPU_PAGE_SIZE;
+	if (rdev->ih.ring_obj)
+		n -= rdev->ih.ring_size;
+	n /= size;
 
 	gtt_obj = kzalloc(n * sizeof(*gtt_obj), GFP_KERNEL);
 	if (!gtt_obj) {
@@ -132,9 +136,15 @@ void radeon_test_moves(struct radeon_device *rdev)
 		     gtt_start++, vram_start++) {
 			if (*vram_start != gtt_start) {
 				DRM_ERROR("Incorrect GTT->VRAM copy %d: Got 0x%p, "
-					  "expected 0x%p (GTT map 0x%p-0x%p)\n",
-					  i, *vram_start, gtt_start, gtt_map,
-					  gtt_end);
+					  "expected 0x%p (GTT/VRAM offset "
+					  "0x%16llx/0x%16llx)\n",
+					  i, *vram_start, gtt_start,
+					  (unsigned long long)
+					  (gtt_addr - rdev->mc.gtt_start +
+					   (void*)gtt_start - gtt_map),
+					  (unsigned long long)
+					  (vram_addr - rdev->mc.vram_start +
+					   (void*)gtt_start - gtt_map));
 				radeon_bo_kunmap(vram_obj);
 				goto out_cleanup;
 			}
@@ -175,9 +185,15 @@ void radeon_test_moves(struct radeon_device *rdev)
 		     gtt_start++, vram_start++) {
 			if (*gtt_start != vram_start) {
 				DRM_ERROR("Incorrect VRAM->GTT copy %d: Got 0x%p, "
-					  "expected 0x%p (VRAM map 0x%p-0x%p)\n",
-					  i, *gtt_start, vram_start, vram_map,
-					  vram_end);
+					  "expected 0x%p (VRAM/GTT offset "
+					  "0x%16llx/0x%16llx)\n",
+					  i, *gtt_start, vram_start,
+					  (unsigned long long)
+					  (vram_addr - rdev->mc.vram_start +
+					   (void*)vram_start - vram_map),
+					  (unsigned long long)
+					  (gtt_addr - rdev->mc.gtt_start +
+					   (void*)vram_start - vram_map));
 				radeon_bo_kunmap(gtt_obj[i]);
 				goto out_cleanup;
 			}
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 60125ddba1e..0b5468bfaf5 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -277,7 +277,12 @@ static int radeon_move_blit(struct ttm_buffer_object *bo,
 		DRM_ERROR("Trying to move memory with CP turned off.\n");
 		return -EINVAL;
 	}
-	r = radeon_copy(rdev, old_start, new_start, new_mem->num_pages, fence);
+
+	BUILD_BUG_ON((PAGE_SIZE % RADEON_GPU_PAGE_SIZE) != 0);
+
+	r = radeon_copy(rdev, old_start, new_start,
+			new_mem->num_pages * (PAGE_SIZE / RADEON_GPU_PAGE_SIZE), /* GPU pages */
+			fence);
 	/* FIXME: handle copy error */
 	r = ttm_bo_move_accel_cleanup(bo, (void *)fence, NULL,
 				      evict, no_wait_reserve, no_wait_gpu, new_mem);
@@ -450,6 +455,29 @@ static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
 			return -EINVAL;
 		mem->bus.base = rdev->mc.aper_base;
 		mem->bus.is_iomem = true;
+#ifdef __alpha__
+		/*
+		 * Alpha: use bus.addr to hold the ioremap() return,
+		 * so we can modify bus.base below.
+		 */
+		if (mem->placement & TTM_PL_FLAG_WC)
+			mem->bus.addr =
+				ioremap_wc(mem->bus.base + mem->bus.offset,
+					   mem->bus.size);
+		else
+			mem->bus.addr =
+				ioremap_nocache(mem->bus.base + mem->bus.offset,
+						mem->bus.size);
+
+		/*
+		 * Alpha: Use just the bus offset plus
+		 * the hose/domain memory base for bus.base.
+		 * It then can be used to build PTEs for VRAM
+		 * access, as done in ttm_bo_vm_fault().
+		 */
+		mem->bus.base = (mem->bus.base & 0x0ffffffffUL) +
+			rdev->ddev->hose->dense_mem_base;
+#endif
 		break;
 	default:
 		return -EINVAL;
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index 4720d000d44..b13c2eedc32 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -536,55 +536,6 @@ static u32 r700_get_tile_pipe_to_backend_map(struct radeon_device *rdev,
 	return backend_map;
 }
 
-static void rv770_program_channel_remap(struct radeon_device *rdev)
-{
-	u32 tcp_chan_steer, mc_shared_chremap, tmp;
-	bool force_no_swizzle;
-
-	switch (rdev->family) {
-	case CHIP_RV770:
-	case CHIP_RV730:
-		force_no_swizzle = false;
-		break;
-	case CHIP_RV710:
-	case CHIP_RV740:
-	default:
-		force_no_swizzle = true;
-		break;
-	}
-
-	tmp = RREG32(MC_SHARED_CHMAP);
-	switch ((tmp & NOOFCHAN_MASK) >> NOOFCHAN_SHIFT) {
-	case 0:
-	case 1:
-	default:
-		/* default mapping */
-		mc_shared_chremap = 0x00fac688;
-		break;
-	case 2:
-	case 3:
-		if (force_no_swizzle)
-			mc_shared_chremap = 0x00fac688;
-		else
-			mc_shared_chremap = 0x00bbc298;
-		break;
-	}
-
-	if (rdev->family == CHIP_RV740)
-		tcp_chan_steer = 0x00ef2a60;
-	else
-		tcp_chan_steer = 0x00fac688;
-
-	/* RV770 CE has special chremap setup */
-	if (rdev->pdev->device == 0x944e) {
-		tcp_chan_steer = 0x00b08b08;
-		mc_shared_chremap = 0x00b08b08;
-	}
-
-	WREG32(TCP_CHAN_STEER, tcp_chan_steer);
-	WREG32(MC_SHARED_CHREMAP, mc_shared_chremap);
-}
-
 static void rv770_gpu_init(struct radeon_device *rdev)
 {
 	int i, j, num_qd_pipes;
@@ -785,8 +736,6 @@ static void rv770_gpu_init(struct radeon_device *rdev)
 	WREG32(DCP_TILING_CONFIG, (gb_tiling_config & 0xffff));
 	WREG32(HDP_TILING_CONFIG, (gb_tiling_config & 0xffff));
 
-	rv770_program_channel_remap(rdev);
-
 	WREG32(CC_RB_BACKEND_DISABLE,      cc_rb_backend_disable);
 	WREG32(CC_GC_SHADER_PIPE_CONFIG,   cc_gc_shader_pipe_config);
 	WREG32(GC_USER_SHADER_PIPE_CONFIG, cc_gc_shader_pipe_config);