Merge branch 'master' into for-next

Fast-forward merge with Linus to be able to merge patches
based on more recent version of the tree.

34 files changed, 471 insertions, 151 deletions

diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile
index 3896ef811102..9f363e0c4b60 100644
--- a/drivers/gpu/drm/radeon/Makefile
+++ b/drivers/gpu/drm/radeon/Makefile
@@ -5,6 +5,7 @@
 ccflags-y := -Iinclude/drm
 
 hostprogs-y := mkregtable
+clean-files := rn50_reg_safe.h r100_reg_safe.h r200_reg_safe.h rv515_reg_safe.h r300_reg_safe.h r420_reg_safe.h rs600_reg_safe.h r600_reg_safe.h evergreen_reg_safe.h cayman_reg_safe.h
 
 quiet_cmd_mkregtable = MKREGTABLE $@
       cmd_mkregtable = $(obj)/mkregtable $< > $@
diff --git a/drivers/gpu/drm/radeon/atom.c b/drivers/gpu/drm/radeon/atom.c
index ebdb0fdb8348..e88c64417a8a 100644
--- a/drivers/gpu/drm/radeon/atom.c
+++ b/drivers/gpu/drm/radeon/atom.c
@@ -1245,6 +1245,9 @@ struct atom_context *atom_parse(struct card_info *card, void *bios)
 	char name[512];
 	int i;
 
+	if (!ctx)
+		return NULL;
+
 	ctx->card = card;
 	ctx->bios = bios;
 
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index 9541995e4b21..c742944d3805 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -764,7 +764,7 @@ static void atombios_crtc_set_dcpll(struct drm_crtc *crtc,
 }
 
 static void atombios_crtc_program_pll(struct drm_crtc *crtc,
-				      int crtc_id,
+				      u32 crtc_id,
 				      int pll_id,
 				      u32 encoder_mode,
 				      u32 encoder_id,
@@ -851,8 +851,7 @@ static void atombios_crtc_program_pll(struct drm_crtc *crtc,
 			args.v5.ucPpll = pll_id;
 			break;
 		case 6:
-			args.v6.ulCrtcPclkFreq.ucCRTC = crtc_id;
-			args.v6.ulCrtcPclkFreq.ulPixelClock = cpu_to_le32(clock / 10);
+			args.v6.ulDispEngClkFreq = cpu_to_le32(crtc_id << 24 | clock / 10);
 			args.v6.ucRefDiv = ref_div;
 			args.v6.usFbDiv = cpu_to_le16(fb_div);
 			args.v6.ulFbDivDecFrac = cpu_to_le32(frac_fb_div * 100000);
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 8c0f9e36ff8e..7ad43c6b1db7 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -613,6 +613,18 @@ 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;
@@ -627,6 +639,7 @@ struct radeon_dp_link_train_info {
 	u8 train_set[4];
 	u8 link_status[DP_LINK_STATUS_SIZE];
 	u8 tries;
+	bool use_dpencoder;
 };
 
 static void radeon_dp_update_vs_emph(struct radeon_dp_link_train_info *dp_info)
@@ -646,7 +659,7 @@ static void radeon_dp_set_tp(struct radeon_dp_link_train_info *dp_info, int tp)
 	int rtp = 0;
 
 	/* set training pattern on the source */
-	if (ASIC_IS_DCE4(dp_info->rdev)) {
+	if (ASIC_IS_DCE4(dp_info->rdev) || !dp_info->use_dpencoder) {
 		switch (tp) {
 		case DP_TRAINING_PATTERN_1:
 			rtp = ATOM_ENCODER_CMD_DP_LINK_TRAINING_PATTERN1;
@@ -706,7 +719,7 @@ static int radeon_dp_link_train_init(struct radeon_dp_link_train_info *dp_info)
 	radeon_write_dpcd_reg(dp_info->radeon_connector, DP_LINK_BW_SET, tmp);
 
 	/* start training on the source */
-	if (ASIC_IS_DCE4(dp_info->rdev))
+	if (ASIC_IS_DCE4(dp_info->rdev) || !dp_info->use_dpencoder)
 		atombios_dig_encoder_setup(dp_info->encoder,
 					   ATOM_ENCODER_CMD_DP_LINK_TRAINING_START, 0);
 	else
@@ -731,7 +744,7 @@ static int radeon_dp_link_train_finish(struct radeon_dp_link_train_info *dp_info
 			      DP_TRAINING_PATTERN_DISABLE);
 
 	/* disable the training pattern on the source */
-	if (ASIC_IS_DCE4(dp_info->rdev))
+	if (ASIC_IS_DCE4(dp_info->rdev) || !dp_info->use_dpencoder)
 		atombios_dig_encoder_setup(dp_info->encoder,
 					   ATOM_ENCODER_CMD_DP_LINK_TRAINING_COMPLETE, 0);
 	else
@@ -869,7 +882,8 @@ void radeon_dp_link_train(struct drm_encoder *encoder,
 	struct radeon_connector *radeon_connector;
 	struct radeon_connector_atom_dig *dig_connector;
 	struct radeon_dp_link_train_info dp_info;
- 	u8 tmp;
+	int index;
+	u8 tmp, frev, crev;
 
 	if (!radeon_encoder->enc_priv)
 		return;
@@ -884,6 +898,18 @@ void radeon_dp_link_train(struct drm_encoder *encoder,
 	    (dig_connector->dp_sink_type != CONNECTOR_OBJECT_ID_eDP))
 		return;
 
+	/* DPEncoderService newer than 1.1 can't program properly the
+	 * training pattern. When facing such version use the
+	 * DIGXEncoderControl (X== 1 | 2)
+	 */
+	dp_info.use_dpencoder = true;
+	index = GetIndexIntoMasterTable(COMMAND, DPEncoderService);
+	if (atom_parse_cmd_header(rdev->mode_info.atom_context, index, &frev, &crev)) {
+		if (crev > 1) {
+			dp_info.use_dpencoder = false;
+		}
+	}
+
 	dp_info.enc_id = 0;
 	if (dig->dig_encoder)
 		dp_info.enc_id |= ATOM_DP_CONFIG_DIG2_ENCODER;
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 15bd0477a3e8..dc0a5b56c81a 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);
@@ -1382,9 +1408,6 @@ int evergreen_cp_resume(struct radeon_device *rdev)
 
 	/* set the wb address wether it's enabled or not */
 	WREG32(CP_RB_RPTR_ADDR,
-#ifdef __BIG_ENDIAN
-	       RB_RPTR_SWAP(2) |
-#endif
 	       ((rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC));
 	WREG32(CP_RB_RPTR_ADDR_HI, upper_32_bits(rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFF);
 	WREG32(SCRATCH_ADDR, ((rdev->wb.gpu_addr + RADEON_WB_SCRATCH_OFFSET) >> 8) & 0xFFFFFFFF);
@@ -1865,6 +1888,8 @@ 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 |=
@@ -2047,6 +2072,7 @@ static void evergreen_gpu_init(struct radeon_device *rdev)
 	rdev->config.evergreen.tile_config |=
 		((gb_addr_config & 0x30000000) >> 28) << 12;
 
+	rdev->config.evergreen.backend_map = gb_backend_map;
 	WREG32(GB_BACKEND_MAP, gb_backend_map);
 	WREG32(GB_ADDR_CONFIG, gb_addr_config);
 	WREG32(DMIF_ADDR_CONFIG, gb_addr_config);
@@ -2761,6 +2787,9 @@ int evergreen_irq_process(struct radeon_device *rdev)
 		return IRQ_NONE;
 	}
 restart_ih:
+	/* Order reading of wptr vs. reading of IH ring data */
+	rmb();
+
 	/* display interrupts */
 	evergreen_irq_ack(rdev);
 
diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index 23d36417158d..a134790903d3 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -428,7 +428,7 @@ static inline int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u3
 		last_reg = ARRAY_SIZE(evergreen_reg_safe_bm);
 
 	i = (reg >> 7);
-	if (i > last_reg) {
+	if (i >= last_reg) {
 		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return -EINVAL;
 	}
@@ -856,7 +856,6 @@ static inline int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u3
 	case SQ_PGM_START_PS:
 	case SQ_PGM_START_HS:
 	case SQ_PGM_START_LS:
-	case GDS_ADDR_BASE:
 	case SQ_CONST_MEM_BASE:
 	case SQ_ALU_CONST_CACHE_GS_0:
 	case SQ_ALU_CONST_CACHE_GS_1:
@@ -946,6 +945,34 @@ static inline int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u3
 		}
 		ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
 		break;
+	case SX_MEMORY_EXPORT_BASE:
+		if (p->rdev->family >= CHIP_CAYMAN) {
+			dev_warn(p->dev, "bad SET_CONFIG_REG "
+				 "0x%04X\n", reg);
+			return -EINVAL;
+		}
+		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		if (r) {
+			dev_warn(p->dev, "bad SET_CONFIG_REG "
+					"0x%04X\n", reg);
+			return -EINVAL;
+		}
+		ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
+		break;
+	case CAYMAN_SX_SCATTER_EXPORT_BASE:
+		if (p->rdev->family < CHIP_CAYMAN) {
+			dev_warn(p->dev, "bad SET_CONTEXT_REG "
+				 "0x%04X\n", reg);
+			return -EINVAL;
+		}
+		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		if (r) {
+			dev_warn(p->dev, "bad SET_CONTEXT_REG "
+					"0x%04X\n", reg);
+			return -EINVAL;
+		}
+		ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
+		break;
 	default:
 		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return -EINVAL;
@@ -1153,6 +1180,34 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			return r;
 		}
 		break;
+	case PACKET3_DISPATCH_DIRECT:
+		if (pkt->count != 3) {
+			DRM_ERROR("bad DISPATCH_DIRECT\n");
+			return -EINVAL;
+		}
+		r = evergreen_cs_track_check(p);
+		if (r) {
+			dev_warn(p->dev, "%s:%d invalid cmd stream %d\n", __func__, __LINE__, idx);
+			return r;
+		}
+		break;
+	case PACKET3_DISPATCH_INDIRECT:
+		if (pkt->count != 1) {
+			DRM_ERROR("bad DISPATCH_INDIRECT\n");
+			return -EINVAL;
+		}
+		r = evergreen_cs_packet_next_reloc(p, &reloc);
+		if (r) {
+			DRM_ERROR("bad DISPATCH_INDIRECT\n");
+			return -EINVAL;
+		}
+		ib[idx+0] = idx_value + (u32)(reloc->lobj.gpu_offset & 0xffffffff);
+		r = evergreen_cs_track_check(p);
+		if (r) {
+			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			return r;
+		}
+		break;
 	case PACKET3_WAIT_REG_MEM:
 		if (pkt->count != 5) {
 			DRM_ERROR("bad WAIT_REG_MEM\n");
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index b7b2714f0b32..7363d9dec909 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -351,6 +351,7 @@
 #define		COLOR_BUFFER_SIZE(x)				((x) << 0)
 #define		POSITION_BUFFER_SIZE(x)				((x) << 8)
 #define		SMX_BUFFER_SIZE(x)				((x) << 16)
+#define	SX_MEMORY_EXPORT_BASE				0x9010
 #define	SX_MISC						0x28350
 
 #define CB_PERF_CTR0_SEL_0				0x9A20
@@ -1122,6 +1123,7 @@
 #define CAYMAN_PA_SC_AA_CONFIG				0x28BE0
 #define         CAYMAN_MSAA_NUM_SAMPLES_SHIFT           0
 #define         CAYMAN_MSAA_NUM_SAMPLES_MASK            0x7
+#define CAYMAN_SX_SCATTER_EXPORT_BASE			0x28358
 /* cayman packet3 addition */
 #define	CAYMAN_PACKET3_DEALLOC_STATE			0x14
 
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index 559dbd412906..cbf57d75d925 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
@@ -669,6 +670,8 @@ 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);
 
@@ -833,6 +836,7 @@ static void cayman_gpu_init(struct radeon_device *rdev)
 	rdev->config.cayman.tile_config |=
 		((gb_addr_config & ROW_SIZE_MASK) >> ROW_SIZE_SHIFT) << 12;
 
+	rdev->config.cayman.backend_map = gb_backend_map;
 	WREG32(GB_BACKEND_MAP, gb_backend_map);
 	WREG32(GB_ADDR_CONFIG, gb_addr_config);
 	WREG32(DMIF_ADDR_CONFIG, gb_addr_config);
@@ -1158,6 +1162,7 @@ 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);
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index bc54b26cb32f..aa5571b73aa0 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -1662,6 +1662,7 @@ void r600_gpu_init(struct radeon_device *rdev)
 									       R6XX_MAX_BACKENDS_MASK) >> 16)),
 							(cc_rb_backend_disable >> 16));
 	rdev->config.r600.tile_config = tiling_config;
+	rdev->config.r600.backend_map = backend_map;
 	tiling_config |= BACKEND_MAP(backend_map);
 	WREG32(GB_TILING_CONFIG, tiling_config);
 	WREG32(DCP_TILING_CONFIG, tiling_config & 0xffff);
@@ -2212,9 +2213,6 @@ int r600_cp_resume(struct radeon_device *rdev)
 
 	/* set the wb address whether it's enabled or not */
 	WREG32(CP_RB_RPTR_ADDR,
-#ifdef __BIG_ENDIAN
-	       RB_RPTR_SWAP(2) |
-#endif
 	       ((rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC));
 	WREG32(CP_RB_RPTR_ADDR_HI, upper_32_bits(rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFF);
 	WREG32(SCRATCH_ADDR, ((rdev->wb.gpu_addr + RADEON_WB_SCRATCH_OFFSET) >> 8) & 0xFFFFFFFF);
@@ -2994,10 +2992,6 @@ int r600_irq_init(struct radeon_device *rdev)
 	/* RPTR_REARM only works if msi's are enabled */
 	if (rdev->msi_enabled)
 		ih_cntl |= RPTR_REARM;
-
-#ifdef __BIG_ENDIAN
-	ih_cntl |= IH_MC_SWAP(IH_MC_SWAP_32BIT);
-#endif
 	WREG32(IH_CNTL, ih_cntl);
 
 	/* force the active interrupt state to all disabled */
@@ -3308,6 +3302,10 @@ int r600_irq_process(struct radeon_device *rdev)
 	if (!rdev->ih.enabled || rdev->shutdown)
 		return IRQ_NONE;
 
+	/* No MSIs, need a dummy read to flush PCI DMAs */
+	if (!rdev->msi_enabled)
+		RREG32(IH_RB_WPTR);
+
 	wptr = r600_get_ih_wptr(rdev);
 	rptr = rdev->ih.rptr;
 	DRM_DEBUG("r600_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
@@ -3320,6 +3318,9 @@ int r600_irq_process(struct radeon_device *rdev)
 	}
 
 restart_ih:
+	/* Order reading of wptr vs. reading of IH ring data */
+	rmb();
+
 	/* display interrupts */
 	r600_irq_ack(rdev);
 
diff --git a/drivers/gpu/drm/radeon/r600_cp.c b/drivers/gpu/drm/radeon/r600_cp.c
index c3ab959bdc7c..45fd592f9606 100644
--- a/drivers/gpu/drm/radeon/r600_cp.c
+++ b/drivers/gpu/drm/radeon/r600_cp.c
@@ -1802,8 +1802,8 @@ static void r600_cp_init_ring_buffer(struct drm_device *dev,
 	/* Set ring buffer size */
 #ifdef __BIG_ENDIAN
 	RADEON_WRITE(R600_CP_RB_CNTL,
-		     RADEON_BUF_SWAP_32BIT |
-		     RADEON_RB_NO_UPDATE |
+		     R600_BUF_SWAP_32BIT |
+		     R600_RB_NO_UPDATE |
 		     (dev_priv->ring.rptr_update_l2qw << 8) |
 		     dev_priv->ring.size_l2qw);
 #else
@@ -1820,15 +1820,15 @@ static void r600_cp_init_ring_buffer(struct drm_device *dev,
 
 #ifdef __BIG_ENDIAN
 	RADEON_WRITE(R600_CP_RB_CNTL,
-		     RADEON_BUF_SWAP_32BIT |
-		     RADEON_RB_NO_UPDATE |
-		     RADEON_RB_RPTR_WR_ENA |
+		     R600_BUF_SWAP_32BIT |
+		     R600_RB_NO_UPDATE |
+		     R600_RB_RPTR_WR_ENA |
 		     (dev_priv->ring.rptr_update_l2qw << 8) |
 		     dev_priv->ring.size_l2qw);
 #else
 	RADEON_WRITE(R600_CP_RB_CNTL,
-		     RADEON_RB_NO_UPDATE |
-		     RADEON_RB_RPTR_WR_ENA |
+		     R600_RB_NO_UPDATE |
+		     R600_RB_RPTR_WR_ENA |
 		     (dev_priv->ring.rptr_update_l2qw << 8) |
 		     dev_priv->ring.size_l2qw);
 #endif
@@ -1851,13 +1851,8 @@ static void r600_cp_init_ring_buffer(struct drm_device *dev,
 			- ((unsigned long) dev->sg->virtual)
 			+ dev_priv->gart_vm_start;
 	}
-	RADEON_WRITE(R600_CP_RB_RPTR_ADDR,
-#ifdef __BIG_ENDIAN
-		     (2 << 0) |
-#endif
-		     (rptr_addr & 0xfffffffc));
-	RADEON_WRITE(R600_CP_RB_RPTR_ADDR_HI,
-		     upper_32_bits(rptr_addr));
+	RADEON_WRITE(R600_CP_RB_RPTR_ADDR, (rptr_addr & 0xfffffffc));
+	RADEON_WRITE(R600_CP_RB_RPTR_ADDR_HI, upper_32_bits(rptr_addr));
 
 #ifdef __BIG_ENDIAN
 	RADEON_WRITE(R600_CP_RB_CNTL,
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 909bda8dd550..cf83aa05a684 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -915,12 +915,11 @@ static inline int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx
 {
 	struct r600_cs_track *track = (struct r600_cs_track *)p->track;
 	struct radeon_cs_reloc *reloc;
-	u32 last_reg = ARRAY_SIZE(r600_reg_safe_bm);
 	u32 m, i, tmp, *ib;
 	int r;
 
 	i = (reg >> 7);
-	if (i > last_reg) {
+	if (i >= ARRAY_SIZE(r600_reg_safe_bm)) {
 		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return -EINVAL;
 	}
@@ -1200,6 +1199,15 @@ static inline int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx
 		}
 		ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
 		break;
+	case SX_MEMORY_EXPORT_BASE:
+		r = r600_cs_packet_next_reloc(p, &reloc);
+		if (r) {
+			dev_warn(p->dev, "bad SET_CONFIG_REG "
+					"0x%04X\n", reg);
+			return -EINVAL;
+		}
+		ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
+		break;
 	default:
 		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return -EINVAL;
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index ef0e0e016914..32807baf55e2 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -60,7 +60,7 @@
  *                          are considered as fatal)
  */
 
-#include <asm/atomic.h>
+#include <linux/atomic.h>
 #include <linux/wait.h>
 #include <linux/list.h>
 #include <linux/kref.h>
@@ -1003,6 +1003,7 @@ struct r600_asic {
 	unsigned		tiling_npipes;
 	unsigned		tiling_group_size;
 	unsigned		tile_config;
+	unsigned		backend_map;
 	struct r100_gpu_lockup	lockup;
 };
 
@@ -1028,6 +1029,7 @@ struct rv770_asic {
 	unsigned		tiling_npipes;
 	unsigned		tiling_group_size;
 	unsigned		tile_config;
+	unsigned		backend_map;
 	struct r100_gpu_lockup	lockup;
 };
 
@@ -1054,6 +1056,7 @@ struct evergreen_asic {
 	unsigned tiling_npipes;
 	unsigned tiling_group_size;
 	unsigned tile_config;
+	unsigned backend_map;
 	struct r100_gpu_lockup	lockup;
 };
 
@@ -1174,7 +1177,7 @@ struct radeon_device {
 	/* Register mmio */
 	resource_size_t			rmmio_base;
 	resource_size_t			rmmio_size;
-	void				*rmmio;
+	void __iomem			*rmmio;
 	radeon_rreg_t			mc_rreg;
 	radeon_wreg_t			mc_wreg;
 	radeon_rreg_t			pll_rreg;
@@ -1251,20 +1254,20 @@ int radeon_gpu_wait_for_idle(struct radeon_device *rdev);
 static inline uint32_t r100_mm_rreg(struct radeon_device *rdev, uint32_t reg)
 {
 	if (reg < rdev->rmmio_size)
-		return readl(((void __iomem *)rdev->rmmio) + reg);
+		return readl((rdev->rmmio) + reg);
 	else {
-		writel(reg, ((void __iomem *)rdev->rmmio) + RADEON_MM_INDEX);
-		return readl(((void __iomem *)rdev->rmmio) + RADEON_MM_DATA);
+		writel(reg, (rdev->rmmio) + RADEON_MM_INDEX);
+		return readl((rdev->rmmio) + RADEON_MM_DATA);
 	}
 }
 
 static inline void r100_mm_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v)
 {
 	if (reg < rdev->rmmio_size)
-		writel(v, ((void __iomem *)rdev->rmmio) + reg);
+		writel(v, (rdev->rmmio) + reg);
 	else {
-		writel(reg, ((void __iomem *)rdev->rmmio) + RADEON_MM_INDEX);
-		writel(v, ((void __iomem *)rdev->rmmio) + RADEON_MM_DATA);
+		writel(reg, (rdev->rmmio) + RADEON_MM_INDEX);
+		writel(v, (rdev->rmmio) + RADEON_MM_DATA);
 	}
 }
 
@@ -1296,10 +1299,10 @@ static inline void r100_io_wreg(struct radeon_device *rdev, u32 reg, u32 v)
 /*
  * Registers read & write functions.
  */
-#define RREG8(reg) readb(((void __iomem *)rdev->rmmio) + (reg))
-#define WREG8(reg, v) writeb(v, ((void __iomem *)rdev->rmmio) + (reg))
-#define RREG16(reg) readw(((void __iomem *)rdev->rmmio) + (reg))
-#define WREG16(reg, v) writew(v, ((void __iomem *)rdev->rmmio) + (reg))
+#define RREG8(reg) readb((rdev->rmmio) + (reg))
+#define WREG8(reg, v) writeb(v, (rdev->rmmio) + (reg))
+#define RREG16(reg) readw((rdev->rmmio) + (reg))
+#define WREG16(reg, v) writew(v, (rdev->rmmio) + (reg))
 #define RREG32(reg) r100_mm_rreg(rdev, (reg))
 #define DREG32(reg) printk(KERN_INFO "REGISTER: " #reg " : 0x%08X\n", r100_mm_rreg(rdev, (reg)))
 #define WREG32(reg, v) r100_mm_wreg(rdev, (reg), (v))
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index b2449629537d..df8218bb83a6 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -625,7 +625,7 @@ static struct radeon_asic r600_asic = {
 	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &r600_cs_parse,
 	.copy_blit = &r600_copy_blit,
-	.copy_dma = &r600_copy_blit,
+	.copy_dma = NULL,
 	.copy = &r600_copy_blit,
 	.get_engine_clock = &radeon_atom_get_engine_clock,
 	.set_engine_clock = &radeon_atom_set_engine_clock,
@@ -672,7 +672,7 @@ static struct radeon_asic rs780_asic = {
 	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &r600_cs_parse,
 	.copy_blit = &r600_copy_blit,
-	.copy_dma = &r600_copy_blit,
+	.copy_dma = NULL,
 	.copy = &r600_copy_blit,
 	.get_engine_clock = &radeon_atom_get_engine_clock,
 	.set_engine_clock = &radeon_atom_set_engine_clock,
@@ -719,7 +719,7 @@ static struct radeon_asic rv770_asic = {
 	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &r600_cs_parse,
 	.copy_blit = &r600_copy_blit,
-	.copy_dma = &r600_copy_blit,
+	.copy_dma = NULL,
 	.copy = &r600_copy_blit,
 	.get_engine_clock = &radeon_atom_get_engine_clock,
 	.set_engine_clock = &radeon_atom_set_engine_clock,
@@ -766,7 +766,7 @@ static struct radeon_asic evergreen_asic = {
 	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &evergreen_cs_parse,
 	.copy_blit = &evergreen_copy_blit,
-	.copy_dma = &evergreen_copy_blit,
+	.copy_dma = NULL,
 	.copy = &evergreen_copy_blit,
 	.get_engine_clock = &radeon_atom_get_engine_clock,
 	.set_engine_clock = &radeon_atom_set_engine_clock,
@@ -813,7 +813,7 @@ static struct radeon_asic sumo_asic = {
 	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &evergreen_cs_parse,
 	.copy_blit = &evergreen_copy_blit,
-	.copy_dma = &evergreen_copy_blit,
+	.copy_dma = NULL,
 	.copy = &evergreen_copy_blit,
 	.get_engine_clock = &radeon_atom_get_engine_clock,
 	.set_engine_clock = &radeon_atom_set_engine_clock,
@@ -860,7 +860,7 @@ static struct radeon_asic btc_asic = {
 	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &evergreen_cs_parse,
 	.copy_blit = &evergreen_copy_blit,
-	.copy_dma = &evergreen_copy_blit,
+	.copy_dma = NULL,
 	.copy = &evergreen_copy_blit,
 	.get_engine_clock = &radeon_atom_get_engine_clock,
 	.set_engine_clock = &radeon_atom_set_engine_clock,
@@ -907,7 +907,7 @@ static struct radeon_asic cayman_asic = {
 	.fence_ring_emit = &r600_fence_ring_emit,
 	.cs_parse = &evergreen_cs_parse,
 	.copy_blit = &evergreen_copy_blit,
-	.copy_dma = &evergreen_copy_blit,
+	.copy_dma = NULL,
 	.copy = &evergreen_copy_blit,
 	.get_engine_clock = &radeon_atom_get_engine_clock,
 	.set_engine_clock = &radeon_atom_set_engine_clock,
diff --git a/drivers/gpu/drm/radeon/radeon_clocks.c b/drivers/gpu/drm/radeon/radeon_clocks.c
index 2d48e7a1474b..b6e18c8db9f5 100644
--- a/drivers/gpu/drm/radeon/radeon_clocks.c
+++ b/drivers/gpu/drm/radeon/radeon_clocks.c
@@ -96,7 +96,7 @@ uint32_t radeon_legacy_get_memory_clock(struct radeon_device *rdev)
  * Read XTAL (ref clock), SCLK and MCLK from Open Firmware device
  * tree. Hopefully, ATI OF driver is kind enough to fill these
  */
-static bool __devinit radeon_read_clocks_OF(struct drm_device *dev)
+static bool radeon_read_clocks_OF(struct drm_device *dev)
 {
 	struct radeon_device *rdev = dev->dev_private;
 	struct device_node *dp = rdev->pdev->dev.of_node;
@@ -166,7 +166,7 @@ static bool __devinit radeon_read_clocks_OF(struct drm_device *dev)
 	return true;
 }
 #else
-static bool __devinit radeon_read_clocks_OF(struct drm_device *dev)
+static bool radeon_read_clocks_OF(struct drm_device *dev)
 {
 	return false;
 }
@@ -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 e4594676a07c..63675241c7ff 100644
--- a/drivers/gpu/drm/radeon/radeon_combios.c
+++ b/drivers/gpu/drm/radeon/radeon_combios.c
@@ -779,7 +779,8 @@ void radeon_combios_i2c_init(struct radeon_device *rdev)
 				}
 			}
 		}
-	} else if (rdev->family >= CHIP_R200) {
+	} else if ((rdev->family == CHIP_R200) ||
+		   (rdev->family >= CHIP_R300)) {
 		/* 0x68 */
 		i2c = combios_setup_i2c_bus(rdev, DDC_MONID, 0, 0);
 		rdev->i2c_bus[3] = radeon_i2c_create(dev, &i2c, "MONID");
@@ -2556,6 +2557,7 @@ void radeon_combios_get_power_modes(struct radeon_device *rdev)
 	u16 offset, misc, misc2 = 0;
 	u8 rev, blocks, tmp;
 	int state_index = 0;
+	struct radeon_i2c_bus_rec i2c_bus;
 
 	rdev->pm.default_power_state_index = -1;
 
@@ -2574,7 +2576,6 @@ void radeon_combios_get_power_modes(struct radeon_device *rdev)
 	offset = combios_get_table_offset(dev, COMBIOS_OVERDRIVE_INFO_TABLE);
 	if (offset) {
 		u8 thermal_controller = 0, gpio = 0, i2c_addr = 0, clk_bit = 0, data_bit = 0;
-		struct radeon_i2c_bus_rec i2c_bus;
 
 		rev = RBIOS8(offset);
 
@@ -2616,6 +2617,25 @@ void radeon_combios_get_power_modes(struct radeon_device *rdev)
 				i2c_new_device(&rdev->pm.i2c_bus->adapter, &info);
 			}
 		}
+	} else {
+		/* boards with a thermal chip, but no overdrive table */
+
+		/* Asus 9600xt has an f75375 on the monid bus */
+		if ((dev->pdev->device == 0x4152) &&
+		    (dev->pdev->subsystem_vendor == 0x1043) &&
+		    (dev->pdev->subsystem_device == 0xc002)) {
+			i2c_bus = combios_setup_i2c_bus(rdev, DDC_MONID, 0, 0);
+			rdev->pm.i2c_bus = radeon_i2c_lookup(rdev, &i2c_bus);
+			if (rdev->pm.i2c_bus) {
+				struct i2c_board_info info = { };
+				const char *name = "f75375";
+				info.addr = 0x28;
+				strlcpy(info.type, name, sizeof(info.type));
+				i2c_new_device(&rdev->pm.i2c_bus->adapter, &info);
+				DRM_INFO("Possible %s thermal controller at 0x%02x\n",
+					 name, info.addr);
+			}
+		}
 	}
 
 	if (rdev->flags & RADEON_IS_MOBILITY) {
@@ -3278,6 +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 9792d4ffdc86..c4b8741dbf58 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))
+	/* just deal with DP (not eDP) here. */
+	if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) {
+		int saved_dpms = connector->dpms;
+
+		if (radeon_hpd_sense(rdev, radeon_connector->hpd.hpd) &&
+		    radeon_dp_needs_link_train(radeon_connector))
 			drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
 		else
 			drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
+		connector->dpms = saved_dpms;
 	}
 }
 
@@ -430,16 +432,73 @@ int radeon_connector_set_property(struct drm_connector *connector, struct drm_pr
 	return 0;
 }
 
+/*
+ * Some integrated ATI Radeon chipset implementations (e. g.
+ * Asus M2A-VM HDMI) may indicate the availability of a DDC,
+ * even when there's no monitor connected. For these connectors
+ * following DDC probe extension will be applied: check also for the
+ * availability of EDID with at least a correct EDID header. Only then,
+ * DDC is assumed to be available. This prevents drm_get_edid() and
+ * drm_edid_block_valid() from periodically dumping data and kernel
+ * errors into the logs and onto the terminal.
+ */
+static bool radeon_connector_needs_extended_probe(struct radeon_device *dev,
+				     uint32_t supported_device,
+				     int connector_type)
+{
+	/* Asus M2A-VM HDMI board sends data to i2c bus even,
+	 * if HDMI add-on card is not plugged in or HDMI is disabled in
+	 * BIOS. Valid DDC can only be assumed, if also a valid EDID header
+	 * can be retrieved via i2c bus during DDC probe */
+	if ((dev->pdev->device == 0x791e) &&
+	    (dev->pdev->subsystem_vendor == 0x1043) &&
+	    (dev->pdev->subsystem_device == 0x826d)) {
+		if ((connector_type == DRM_MODE_CONNECTOR_HDMIA) &&
+		    (supported_device == ATOM_DEVICE_DFP2_SUPPORT))
+			return true;
+	}
+	/* ECS A740GM-M with ATI RADEON 2100 sends data to i2c bus
+	 * for a DVI connector that is not implemented */
+	if ((dev->pdev->device == 0x796e) &&
+	    (dev->pdev->subsystem_vendor == 0x1019) &&
+	    (dev->pdev->subsystem_device == 0x2615)) {
+		if ((connector_type == DRM_MODE_CONNECTOR_DVID) &&
+		    (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;
+}
+
 static void radeon_fixup_lvds_native_mode(struct drm_encoder *encoder,
 					  struct drm_connector *connector)
 {
 	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) {
@@ -450,6 +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;
@@ -661,7 +721,8 @@ radeon_vga_detect(struct drm_connector *connector, bool force)
 		ret = connector_status_disconnected;
 
 	if (radeon_connector->ddc_bus)
-		dret = radeon_ddc_probe(radeon_connector);
+		dret = radeon_ddc_probe(radeon_connector,
+					radeon_connector->requires_extended_probe);
 	if (dret) {
 		if (radeon_connector->edid) {
 			kfree(radeon_connector->edid);
@@ -833,7 +894,8 @@ radeon_dvi_detect(struct drm_connector *connector, bool force)
 	bool dret = false;
 
 	if (radeon_connector->ddc_bus)
-		dret = radeon_ddc_probe(radeon_connector);
+		dret = radeon_ddc_probe(radeon_connector,
+					radeon_connector->requires_extended_probe);
 	if (dret) {
 		if (radeon_connector->edid) {
 			kfree(radeon_connector->edid);
@@ -1235,12 +1297,33 @@ 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)) {
+	} 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 (radeon_hpd_sense(rdev, radeon_connector->hpd.hpd))
+			ret = connector_status_connected;
+		else {
+			/* need to setup ddc on the bridge */
 			if (encoder)
 				radeon_atom_ext_encoder_setup_ddc(encoder);
+			if (radeon_ddc_probe(radeon_connector,
+					     radeon_connector->requires_extended_probe))
+				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);
+			}
 		}
+	} 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;
@@ -1251,20 +1334,11 @@ radeon_dp_detect(struct drm_connector *connector, bool force)
 				if (radeon_dp_getdpcd(radeon_connector))
 					ret = connector_status_connected;
 			} else {
-				if (radeon_ddc_probe(radeon_connector))
+				if (radeon_ddc_probe(radeon_connector,
+						     radeon_connector->requires_extended_probe))
 					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);
@@ -1406,6 +1480,9 @@ radeon_add_atom_connector(struct drm_device *dev,
 	radeon_connector->shared_ddc = shared_ddc;
 	radeon_connector->connector_object_id = connector_object_id;
 	radeon_connector->hpd = *hpd;
+	radeon_connector->requires_extended_probe =
+		radeon_connector_needs_extended_probe(rdev, supported_device,
+							connector_type);
 	radeon_connector->router = *router;
 	if (router->ddc_valid || router->cd_valid) {
 		radeon_connector->router_bus = radeon_i2c_lookup(rdev, &router->i2c_info);
@@ -1752,6 +1829,9 @@ radeon_add_legacy_connector(struct drm_device *dev,
 	radeon_connector->devices = supported_device;
 	radeon_connector->connector_object_id = connector_object_id;
 	radeon_connector->hpd = *hpd;
+	radeon_connector->requires_extended_probe =
+		radeon_connector_needs_extended_probe(rdev, supported_device,
+							connector_type);
 	switch (connector_type) {
 	case DRM_MODE_CONNECTOR_VGA:
 		drm_connector_init(dev, &radeon_connector->base, &radeon_vga_connector_funcs, connector_type);
diff --git a/drivers/gpu/drm/radeon/radeon_cp.c b/drivers/gpu/drm/radeon/radeon_cp.c
index 75867792a4e2..045ec59478f9 100644
--- a/drivers/gpu/drm/radeon/radeon_cp.c
+++ b/drivers/gpu/drm/radeon/radeon_cp.c
@@ -2115,7 +2115,7 @@ int radeon_driver_load(struct drm_device *dev, unsigned long flags)
 
 	if (drm_pci_device_is_agp(dev))
 		dev_priv->flags |= RADEON_IS_AGP;
-	else if (drm_pci_device_is_pcie(dev))
+	else if (pci_is_pcie(dev->pdev))
 		dev_priv->flags |= RADEON_IS_PCIE;
 	else
 		dev_priv->flags |= RADEON_IS_PCI;
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 7cfaa7e2f3b5..b51e15725c6e 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) |
@@ -704,8 +710,9 @@ int radeon_device_init(struct radeon_device *rdev,
 	rdev->gpu_lockup = false;
 	rdev->accel_working = false;
 
-	DRM_INFO("initializing kernel modesetting (%s 0x%04X:0x%04X).\n",
-		radeon_family_name[rdev->family], pdev->vendor, pdev->device);
+	DRM_INFO("initializing kernel modesetting (%s 0x%04X:0x%04X 0x%04X:0x%04X).\n",
+		radeon_family_name[rdev->family], pdev->vendor, pdev->device,
+		pdev->subsystem_vendor, pdev->subsystem_device);
 
 	/* mutex initialization are all done here so we
 	 * can recall function without having locking issues */
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 292f73f0ddbd..6cc17fb96a57 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -282,7 +282,7 @@ void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id)
 	spin_lock_irqsave(&rdev->ddev->event_lock, flags);
 	work = radeon_crtc->unpin_work;
 	if (work == NULL ||
-	    !radeon_fence_signaled(work->fence)) {
+	    (work->fence && !radeon_fence_signaled(work->fence))) {
 		spin_unlock_irqrestore(&rdev->ddev->event_lock, flags);
 		return;
 	}
@@ -348,7 +348,6 @@ static int radeon_crtc_page_flip(struct drm_crtc *crtc,
 	struct radeon_framebuffer *new_radeon_fb;
 	struct drm_gem_object *obj;
 	struct radeon_bo *rbo;
-	struct radeon_fence *fence;
 	struct radeon_unpin_work *work;
 	unsigned long flags;
 	u32 tiling_flags, pitch_pixels;
@@ -359,16 +358,9 @@ static int radeon_crtc_page_flip(struct drm_crtc *crtc,
 	if (work == NULL)
 		return -ENOMEM;
 
-	r = radeon_fence_create(rdev, &fence);
-	if (unlikely(r != 0)) {
-		kfree(work);
-		DRM_ERROR("flip queue: failed to create fence.\n");
-		return -ENOMEM;
-	}
 	work->event = event;
 	work->rdev = rdev;
 	work->crtc_id = radeon_crtc->crtc_id;
-	work->fence = radeon_fence_ref(fence);
 	old_radeon_fb = to_radeon_framebuffer(crtc->fb);
 	new_radeon_fb = to_radeon_framebuffer(fb);
 	/* schedule unpin of the old buffer */
@@ -377,6 +369,10 @@ static int radeon_crtc_page_flip(struct drm_crtc *crtc,
 	drm_gem_object_reference(obj);
 	rbo = gem_to_radeon_bo(obj);
 	work->old_rbo = rbo;
+	obj = new_radeon_fb->obj;
+	rbo = gem_to_radeon_bo(obj);
+	if (rbo->tbo.sync_obj)
+		work->fence = radeon_fence_ref(rbo->tbo.sync_obj);
 	INIT_WORK(&work->work, radeon_unpin_work_func);
 
 	/* We borrow the event spin lock for protecting unpin_work */
@@ -391,9 +387,6 @@ static int radeon_crtc_page_flip(struct drm_crtc *crtc,
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 
 	/* pin the new buffer */
-	obj = new_radeon_fb->obj;
-	rbo = gem_to_radeon_bo(obj);
-
 	DRM_DEBUG_DRIVER("flip-ioctl() cur_fbo = %p, cur_bbo = %p\n",
 			 work->old_rbo, rbo);
 
@@ -461,37 +454,18 @@ static int radeon_crtc_page_flip(struct drm_crtc *crtc,
 		goto pflip_cleanup1;
 	}
 
-	/* 32 ought to cover us */
-	r = radeon_ring_lock(rdev, 32);
-	if (r) {
-		DRM_ERROR("failed to lock the ring before flip\n");
-		goto pflip_cleanup2;
-	}
-
-	/* emit the fence */
-	radeon_fence_emit(rdev, fence);
 	/* set the proper interrupt */
 	radeon_pre_page_flip(rdev, radeon_crtc->crtc_id);
-	/* fire the ring */
-	radeon_ring_unlock_commit(rdev);
 
 	return 0;
 
-pflip_cleanup2:
-	drm_vblank_put(dev, radeon_crtc->crtc_id);
-
 pflip_cleanup1:
-	r = radeon_bo_reserve(rbo, false);
-	if (unlikely(r != 0)) {
+	if (unlikely(radeon_bo_reserve(rbo, false) != 0)) {
 		DRM_ERROR("failed to reserve new rbo in error path\n");
 		goto pflip_cleanup;
 	}
-	r = radeon_bo_unpin(rbo);
-	if (unlikely(r != 0)) {
-		radeon_bo_unreserve(rbo);
-		r = -EINVAL;
+	if (unlikely(radeon_bo_unpin(rbo) != 0)) {
 		DRM_ERROR("failed to unpin new rbo in error path\n");
-		goto pflip_cleanup;
 	}
 	radeon_bo_unreserve(rbo);
 
@@ -501,7 +475,7 @@ pflip_cleanup:
 unlock_free:
 	drm_gem_object_unreference_unlocked(old_radeon_fb->obj);
 	spin_unlock_irqrestore(&dev->event_lock, flags);
-	radeon_fence_unref(&fence);
+	radeon_fence_unref(&work->fence);
 	kfree(work);
 
 	return r;
@@ -733,16 +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) {
@@ -777,8 +756,17 @@ static int radeon_ddc_dump(struct drm_connector *connector)
 	if (!radeon_connector->ddc_bus)
 		return -1;
 	edid = drm_get_edid(connector, &radeon_connector->ddc_bus->adapter);
+	/* Log EDID retrieval status here. In particular with regard to
+	 * connectors with requires_extended_probe flag set, that will prevent
+	 * function radeon_dvi_detect() to fetch EDID on this connector,
+	 * as long as there is no valid EDID header found */
 	if (edid) {
+		DRM_INFO("Radeon display connector %s: Found valid EDID",
+				drm_get_connector_name(connector));
 		kfree(edid);
+	} else {
+		DRM_INFO("Radeon display connector %s: No monitor connected or invalid EDID",
+				drm_get_connector_name(connector));
 	}
 	return ret;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 73dfbe8e5f9e..e71d2ed7fa11 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -51,9 +51,10 @@
  *   2.8.0 - pageflip support, r500 US_FORMAT regs. r500 ARGB2101010 colorbuf, r300->r500 CMASK, clock crystal query
  *   2.9.0 - r600 tiling (s3tc,rgtc) working, SET_PREDICATION packet 3 on r600 + eg, backend query
  *   2.10.0 - fusion 2D tiling
+ *   2.11.0 - backend map, initial compute support for the CS checker
  */
 #define KMS_DRIVER_MAJOR	2
-#define KMS_DRIVER_MINOR	10
+#define KMS_DRIVER_MINOR	11
 #define KMS_DRIVER_PATCHLEVEL	0
 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags);
 int radeon_driver_unload_kms(struct drm_device *dev);
diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c
index b293487e5aa3..319d85d7e759 100644
--- a/drivers/gpu/drm/radeon/radeon_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_encoders.c
@@ -2323,6 +2323,9 @@ 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_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c
index 9a9f9fcde378..3475a09f946b 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -29,7 +29,7 @@
  *    Dave Airlie
  */
 #include <linux/seq_file.h>
-#include <asm/atomic.h>
+#include <linux/atomic.h>
 #include <linux/wait.h>
 #include <linux/list.h>
 #include <linux/kref.h>
diff --git a/drivers/gpu/drm/radeon/radeon_i2c.c b/drivers/gpu/drm/radeon/radeon_i2c.c
index 781196db792f..6c111c1fa3f9 100644
--- a/drivers/gpu/drm/radeon/radeon_i2c.c
+++ b/drivers/gpu/drm/radeon/radeon_i2c.c
@@ -32,17 +32,17 @@
  * radeon_ddc_probe
  *
  */
-bool radeon_ddc_probe(struct radeon_connector *radeon_connector)
+bool radeon_ddc_probe(struct radeon_connector *radeon_connector, bool requires_extended_probe)
 {
-	u8 out_buf[] = { 0x0, 0x0};
-	u8 buf[2];
+	u8 out = 0x0;
+	u8 buf[8];
 	int ret;
 	struct i2c_msg msgs[] = {
 		{
 			.addr = 0x50,
 			.flags = 0,
 			.len = 1,
-			.buf = out_buf,
+			.buf = &out,
 		},
 		{
 			.addr = 0x50,
@@ -52,15 +52,31 @@ bool radeon_ddc_probe(struct radeon_connector *radeon_connector)
 		}
 	};
 
+	/* Read 8 bytes from i2c for extended probe of EDID header */
+	if (requires_extended_probe)
+		msgs[1].len = 8;
+
 	/* on hw with routers, select right port */
 	if (radeon_connector->router.ddc_valid)
 		radeon_router_select_ddc_port(radeon_connector);
 
 	ret = i2c_transfer(&radeon_connector->ddc_bus->adapter, msgs, 2);
-	if (ret == 2)
-		return true;
-
-	return false;
+	if (ret != 2)
+		/* Couldn't find an accessible DDC on this connector */
+		return false;
+	if (requires_extended_probe) {
+		/* Probe also for valid EDID header
+		 * EDID header starts with:
+		 * 0x00,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0x00.
+		 * Only the first 6 bytes must be valid as
+		 * drm_edid_block_valid() can fix the last 2 bytes */
+		if (drm_edid_header_is_valid(buf) < 6) {
+			/* Couldn't find an accessible EDID on this
+			 * connector */
+			return false;
+		}
+	}
+	return true;
 }
 
 /* bit banging i2c */
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index bd58af658581..be2c1224e68a 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -60,7 +60,7 @@ int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags)
 	/* update BUS flag */
 	if (drm_pci_device_is_agp(dev)) {
 		flags |= RADEON_IS_AGP;
-	} else if (drm_pci_device_is_pcie(dev)) {
+	} else if (pci_is_pcie(dev->pdev)) {
 		flags |= RADEON_IS_PCIE;
 	} else {
 		flags |= RADEON_IS_PCI;
@@ -237,6 +237,19 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	case RADEON_INFO_FUSION_GART_WORKING:
 		value = 1;
 		break;
+	case RADEON_INFO_BACKEND_MAP:
+		if (rdev->family >= CHIP_CAYMAN)
+			value = rdev->config.cayman.backend_map;
+		else if (rdev->family >= CHIP_CEDAR)
+			value = rdev->config.evergreen.backend_map;
+		else if (rdev->family >= CHIP_RV770)
+			value = rdev->config.rv770.backend_map;
+		else if (rdev->family >= CHIP_R600)
+			value = rdev->config.r600.backend_map;
+		else {
+			return -EINVAL;
+		}
+		break;
 	default:
 		DRM_DEBUG_KMS("Invalid request %d\n", info->request);
 		return -EINVAL;
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 6df4e3cec0c2..68820f5f6303 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -438,6 +438,9 @@ struct radeon_connector {
 	struct radeon_i2c_chan *ddc_bus;
 	/* some systems have an hdmi and vga port with a shared ddc line */
 	bool shared_ddc;
+	/* for some Radeon chip families we apply an additional EDID header
+	   check as part of the DDC probe */
+	bool requires_extended_probe;
 	bool use_digital;
 	/* we need to mind the EDID between detect
 	   and get modes due to analog/digital/tvencoder */
@@ -476,6 +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);
@@ -514,7 +518,8 @@ extern void radeon_i2c_put_byte(struct radeon_i2c_chan *i2c,
 				u8 val);
 extern void radeon_router_select_ddc_port(struct radeon_connector *radeon_connector);
 extern void radeon_router_select_cd_port(struct radeon_connector *radeon_connector);
-extern bool radeon_ddc_probe(struct radeon_connector *radeon_connector);
+extern bool radeon_ddc_probe(struct radeon_connector *radeon_connector,
+			bool requires_extended_probe);
 extern int radeon_ddc_get_modes(struct radeon_connector *radeon_connector);
 
 extern struct drm_encoder *radeon_best_encoder(struct drm_connector *connector);
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index aaa19dc418a0..6fabe89fa6a1 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -594,6 +594,9 @@ int radeon_pm_init(struct radeon_device *rdev)
 			if (rdev->pm.default_vddc)
 				radeon_atom_set_voltage(rdev, rdev->pm.default_vddc,
 							SET_VOLTAGE_TYPE_ASIC_VDDC);
+			if (rdev->pm.default_vddci)
+				radeon_atom_set_voltage(rdev, rdev->pm.default_vddci,
+							SET_VOLTAGE_TYPE_ASIC_VDDCI);
 			if (rdev->pm.default_sclk)
 				radeon_set_engine_clock(rdev, rdev->pm.default_sclk);
 			if (rdev->pm.default_mclk)
diff --git a/drivers/gpu/drm/radeon/radeon_reg.h b/drivers/gpu/drm/radeon/radeon_reg.h
index bc44a3d35ec6..b4ce86455707 100644
--- a/drivers/gpu/drm/radeon/radeon_reg.h
+++ b/drivers/gpu/drm/radeon/radeon_reg.h
@@ -3295,7 +3295,7 @@
 #	define RADEON_RB_BUFSZ_MASK		(0x3f << 0)
 #	define RADEON_RB_BLKSZ_SHIFT		8
 #	define RADEON_RB_BLKSZ_MASK		(0x3f << 8)
-#	define RADEON_BUF_SWAP_32BIT		(1 << 17)
+#	define RADEON_BUF_SWAP_32BIT		(2 << 16)
 #	define RADEON_MAX_FETCH_SHIFT		18
 #	define RADEON_MAX_FETCH_MASK		(0x3 << 18)
 #	define RADEON_RB_NO_UPDATE		(1 << 27)
diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c
index dee4a0c1b4b2..602fa3541c45 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 60125ddba1e9..9b86fb0e4122 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -450,6 +450,29 @@ 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/reg_srcs/cayman b/drivers/gpu/drm/radeon/reg_srcs/cayman
index 0aa8e85a9457..2316977eb924 100644
--- a/drivers/gpu/drm/radeon/reg_srcs/cayman
+++ b/drivers/gpu/drm/radeon/reg_srcs/cayman
@@ -208,6 +208,7 @@ cayman 0x9400
 0x0002834C PA_SC_VPORT_ZMAX_15
 0x00028350 SX_MISC
 0x00028354 SX_SURFACE_SYNC
+0x0002835C SX_SCATTER_EXPORT_SIZE
 0x00028380 SQ_VTX_SEMANTIC_0
 0x00028384 SQ_VTX_SEMANTIC_1
 0x00028388 SQ_VTX_SEMANTIC_2
@@ -432,6 +433,7 @@ cayman 0x9400
 0x00028700 SPI_STACK_MGMT
 0x00028704 SPI_WAVE_MGMT_1
 0x00028708 SPI_WAVE_MGMT_2
+0x00028720 GDS_ADDR_BASE
 0x00028724 GDS_ADDR_SIZE
 0x00028780 CB_BLEND0_CONTROL
 0x00028784 CB_BLEND1_CONTROL
diff --git a/drivers/gpu/drm/radeon/reg_srcs/evergreen b/drivers/gpu/drm/radeon/reg_srcs/evergreen
index 0e28cae7ea43..161737a28c23 100644
--- a/drivers/gpu/drm/radeon/reg_srcs/evergreen
+++ b/drivers/gpu/drm/radeon/reg_srcs/evergreen
@@ -44,6 +44,7 @@ evergreen 0x9400
 0x00008E28 SQ_STATIC_THREAD_MGMT_3
 0x00008E2C SQ_LDS_RESOURCE_MGMT
 0x00008E48 SQ_EX_ALLOC_TABLE_SLOTS
+0x00009014 SX_MEMORY_EXPORT_SIZE
 0x00009100 SPI_CONFIG_CNTL
 0x0000913C SPI_CONFIG_CNTL_1
 0x00009508 TA_CNTL_AUX
@@ -442,7 +443,9 @@ evergreen 0x9400
 0x000286EC SPI_COMPUTE_NUM_THREAD_X
 0x000286F0 SPI_COMPUTE_NUM_THREAD_Y
 0x000286F4 SPI_COMPUTE_NUM_THREAD_Z
+0x00028720 GDS_ADDR_BASE
 0x00028724 GDS_ADDR_SIZE
+0x00028728 GDS_ORDERED_WAVE_PER_SE
 0x00028780 CB_BLEND0_CONTROL
 0x00028784 CB_BLEND1_CONTROL
 0x00028788 CB_BLEND2_CONTROL
diff --git a/drivers/gpu/drm/radeon/reg_srcs/r600 b/drivers/gpu/drm/radeon/reg_srcs/r600
index ea49752ee99c..0380c5c15f80 100644
--- a/drivers/gpu/drm/radeon/reg_srcs/r600
+++ b/drivers/gpu/drm/radeon/reg_srcs/r600
@@ -429,6 +429,7 @@ r600 0x9400
 0x00028438 SX_ALPHA_REF
 0x00028410 SX_ALPHA_TEST_CONTROL
 0x00028350 SX_MISC
+0x00009014 SX_MEMORY_EXPORT_SIZE
 0x00009604 TC_INVALIDATE
 0x00009400 TD_FILTER4
 0x00009404 TD_FILTER4_1
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index 1f5850e473cc..4b5d0e6974a8 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -530,7 +530,7 @@ int rs600_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr)
 	addr = addr & 0xFFFFFFFFFFFFF000ULL;
 	addr |= R600_PTE_VALID | R600_PTE_SYSTEM | R600_PTE_SNOOPED;
 	addr |= R600_PTE_READABLE | R600_PTE_WRITEABLE;
-	writeq(addr, ((void __iomem *)ptr) + (i * 8));
+	writeq(addr, ptr + (i * 8));
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index 4de51891aa6d..4720d000d440 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -778,6 +778,7 @@ static void rv770_gpu_init(struct radeon_device *rdev)
 								(cc_rb_backend_disable >> 16));
 
 	rdev->config.rv770.tile_config = gb_tiling_config;
+	rdev->config.rv770.backend_map = backend_map;
 	gb_tiling_config |= BACKEND_MAP(backend_map);
 
 	WREG32(GB_TILING_CONFIG, gb_tiling_config);