Merge remote branch 'korg/drm-radeon-next' into drm-linus

* korg/drm-radeon-next:
  drm/radeon/kms: fix legacy get_engine/memory clock
  drm/radeon/kms/atom: atom parser fixes
  drm/radeon/kms: clean up atombios pll code
  drm/radeon/kms: clean up pll struct
  drm/radeon/kms/atom: fix crtc lock ordering
  drm/radeon: r6xx/r7xx possible security issue, system ram access
  drm/radeon/kms: r600/r700 don't test ib if ib initialization fails
  drm/radeon/kms: Forbid creation of framebuffer with no valid GEM object
  drm/radeon/kms: r600 handle irq vector ring overflow
  drm/radeon/kms: r600/r700 don't process IRQ if not initialized
  drm/radeon/kms: r600/r700 disable irq at suspend
  drm/radeon/kms/r4xx: cleanup atom path
  drm/radeon/kms: fix atombios_crtc_set_base
  drm/radeon/kms/atom: upstream parser updates
  drm/radeon/kms/atom: fix some parser bugs
  drm/radeon/kms: fix hardcoded mmio size in register functions
  drm/radeon/kms/r100: fix bug in CS parser
  drm/radeon/kms/r200: fix bug in CS parser
  drm/radeon/kms/r200: fix bug in CS parser

16 files changed, 472 insertions, 241 deletions

diff --git a/drivers/gpu/drm/radeon/atom.c b/drivers/gpu/drm/radeon/atom.c
index 388140a7e651..e3b44562d265 100644
--- a/drivers/gpu/drm/radeon/atom.c
+++ b/drivers/gpu/drm/radeon/atom.c
@@ -246,6 +246,9 @@ static uint32_t atom_get_src_int(atom_exec_context *ctx, uint8_t attr,
 		case ATOM_WS_ATTRIBUTES:
 			val = gctx->io_attr;
 			break;
+		case ATOM_WS_REGPTR:
+			val = gctx->reg_block;
+			break;
 		default:
 			val = ctx->ws[idx];
 		}
@@ -385,6 +388,32 @@ static uint32_t atom_get_src(atom_exec_context *ctx, uint8_t attr, int *ptr)
 	return atom_get_src_int(ctx, attr, ptr, NULL, 1);
 }
 
+static uint32_t atom_get_src_direct(atom_exec_context *ctx, uint8_t align, int *ptr)
+{
+	uint32_t val = 0xCDCDCDCD;
+
+	switch (align) {
+	case ATOM_SRC_DWORD:
+		val = U32(*ptr);
+		(*ptr) += 4;
+		break;
+	case ATOM_SRC_WORD0:
+	case ATOM_SRC_WORD8:
+	case ATOM_SRC_WORD16:
+		val = U16(*ptr);
+		(*ptr) += 2;
+		break;
+	case ATOM_SRC_BYTE0:
+	case ATOM_SRC_BYTE8:
+	case ATOM_SRC_BYTE16:
+	case ATOM_SRC_BYTE24:
+		val = U8(*ptr);
+		(*ptr)++;
+		break;
+	}
+	return val;
+}
+
 static uint32_t atom_get_dst(atom_exec_context *ctx, int arg, uint8_t attr,
 			     int *ptr, uint32_t *saved, int print)
 {
@@ -482,6 +511,9 @@ static void atom_put_dst(atom_exec_context *ctx, int arg, uint8_t attr,
 		case ATOM_WS_ATTRIBUTES:
 			gctx->io_attr = val;
 			break;
+		case ATOM_WS_REGPTR:
+			gctx->reg_block = val;
+			break;
 		default:
 			ctx->ws[idx] = val;
 		}
@@ -677,7 +709,7 @@ static void atom_op_mask(atom_exec_context *ctx, int *ptr, int arg)
 	SDEBUG("   dst: ");
 	dst = atom_get_dst(ctx, arg, attr, ptr, &saved, 1);
 	SDEBUG("   src1: ");
-	src1 = atom_get_src(ctx, attr, ptr);
+	src1 = atom_get_src_direct(ctx, ((attr >> 3) & 7), ptr);
 	SDEBUG("   src2: ");
 	src2 = atom_get_src(ctx, attr, ptr);
 	dst &= src1;
@@ -809,6 +841,38 @@ static void atom_op_setregblock(atom_exec_context *ctx, int *ptr, int arg)
 	SDEBUG("   base: 0x%04X\n", ctx->ctx->reg_block);
 }
 
+static void atom_op_shift_left(atom_exec_context *ctx, int *ptr, int arg)
+{
+	uint8_t attr = U8((*ptr)++), shift;
+	uint32_t saved, dst;
+	int dptr = *ptr;
+	attr &= 0x38;
+	attr |= atom_def_dst[attr >> 3] << 6;
+	SDEBUG("   dst: ");
+	dst = atom_get_dst(ctx, arg, attr, ptr, &saved, 1);
+	shift = atom_get_src_direct(ctx, ATOM_SRC_BYTE0, ptr);
+	SDEBUG("   shift: %d\n", shift);
+	dst <<= shift;
+	SDEBUG("   dst: ");
+	atom_put_dst(ctx, arg, attr, &dptr, dst, saved);
+}
+
+static void atom_op_shift_right(atom_exec_context *ctx, int *ptr, int arg)
+{
+	uint8_t attr = U8((*ptr)++), shift;
+	uint32_t saved, dst;
+	int dptr = *ptr;
+	attr &= 0x38;
+	attr |= atom_def_dst[attr >> 3] << 6;
+	SDEBUG("   dst: ");
+	dst = atom_get_dst(ctx, arg, attr, ptr, &saved, 1);
+	shift = atom_get_src_direct(ctx, ATOM_SRC_BYTE0, ptr);
+	SDEBUG("   shift: %d\n", shift);
+	dst >>= shift;
+	SDEBUG("   dst: ");
+	atom_put_dst(ctx, arg, attr, &dptr, dst, saved);
+}
+
 static void atom_op_shl(atom_exec_context *ctx, int *ptr, int arg)
 {
 	uint8_t attr = U8((*ptr)++), shift;
@@ -818,7 +882,7 @@ static void atom_op_shl(atom_exec_context *ctx, int *ptr, int arg)
 	attr |= atom_def_dst[attr >> 3] << 6;
 	SDEBUG("   dst: ");
 	dst = atom_get_dst(ctx, arg, attr, ptr, &saved, 1);
-	shift = U8((*ptr)++);
+	shift = atom_get_src(ctx, attr, ptr);
 	SDEBUG("   shift: %d\n", shift);
 	dst <<= shift;
 	SDEBUG("   dst: ");
@@ -834,7 +898,7 @@ static void atom_op_shr(atom_exec_context *ctx, int *ptr, int arg)
 	attr |= atom_def_dst[attr >> 3] << 6;
 	SDEBUG("   dst: ");
 	dst = atom_get_dst(ctx, arg, attr, ptr, &saved, 1);
-	shift = U8((*ptr)++);
+	shift = atom_get_src(ctx, attr, ptr);
 	SDEBUG("   shift: %d\n", shift);
 	dst >>= shift;
 	SDEBUG("   dst: ");
@@ -937,18 +1001,18 @@ static struct {
 	atom_op_or, ATOM_ARG_FB}, {
 	atom_op_or, ATOM_ARG_PLL}, {
 	atom_op_or, ATOM_ARG_MC}, {
-	atom_op_shl, ATOM_ARG_REG}, {
-	atom_op_shl, ATOM_ARG_PS}, {
-	atom_op_shl, ATOM_ARG_WS}, {
-	atom_op_shl, ATOM_ARG_FB}, {
-	atom_op_shl, ATOM_ARG_PLL}, {
-	atom_op_shl, ATOM_ARG_MC}, {
-	atom_op_shr, ATOM_ARG_REG}, {
-	atom_op_shr, ATOM_ARG_PS}, {
-	atom_op_shr, ATOM_ARG_WS}, {
-	atom_op_shr, ATOM_ARG_FB}, {
-	atom_op_shr, ATOM_ARG_PLL}, {
-	atom_op_shr, ATOM_ARG_MC}, {
+	atom_op_shift_left, ATOM_ARG_REG}, {
+	atom_op_shift_left, ATOM_ARG_PS}, {
+	atom_op_shift_left, ATOM_ARG_WS}, {
+	atom_op_shift_left, ATOM_ARG_FB}, {
+	atom_op_shift_left, ATOM_ARG_PLL}, {
+	atom_op_shift_left, ATOM_ARG_MC}, {
+	atom_op_shift_right, ATOM_ARG_REG}, {
+	atom_op_shift_right, ATOM_ARG_PS}, {
+	atom_op_shift_right, ATOM_ARG_WS}, {
+	atom_op_shift_right, ATOM_ARG_FB}, {
+	atom_op_shift_right, ATOM_ARG_PLL}, {
+	atom_op_shift_right, ATOM_ARG_MC}, {
 	atom_op_mul, ATOM_ARG_REG}, {
 	atom_op_mul, ATOM_ARG_PS}, {
 	atom_op_mul, ATOM_ARG_WS}, {
@@ -1058,8 +1122,6 @@ static void atom_execute_table_locked(struct atom_context *ctx, int index, uint3
 
 	SDEBUG(">> execute %04X (len %d, WS %d, PS %d)\n", base, len, ws, ps);
 
-	/* reset reg block */
-	ctx->reg_block = 0;
 	ectx.ctx = ctx;
 	ectx.ps_shift = ps / 4;
 	ectx.start = base;
@@ -1096,6 +1158,12 @@ static void atom_execute_table_locked(struct atom_context *ctx, int index, uint3
 void atom_execute_table(struct atom_context *ctx, int index, uint32_t * params)
 {
 	mutex_lock(&ctx->mutex);
+	/* reset reg block */
+	ctx->reg_block = 0;
+	/* reset fb window */
+	ctx->fb_base = 0;
+	/* reset io mode */
+	ctx->io_mode = ATOM_IO_MM;
 	atom_execute_table_locked(ctx, index, params);
 	mutex_unlock(&ctx->mutex);
 }
diff --git a/drivers/gpu/drm/radeon/atom.h b/drivers/gpu/drm/radeon/atom.h
index 47fd943f6d14..bc73781423a1 100644
--- a/drivers/gpu/drm/radeon/atom.h
+++ b/drivers/gpu/drm/radeon/atom.h
@@ -91,6 +91,7 @@
 #define ATOM_WS_AND_MASK	0x45
 #define ATOM_WS_FB_WINDOW	0x46
 #define ATOM_WS_ATTRIBUTES	0x47
+#define ATOM_WS_REGPTR  	0x48
 
 #define ATOM_IIO_NOP		0
 #define ATOM_IIO_START		1
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index 260fcf59f00c..e619aca5423f 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -409,59 +409,57 @@ static void atombios_set_ss(struct drm_crtc *crtc, int enable)
 	}
 }
 
-void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode *mode)
+union adjust_pixel_clock {
+	ADJUST_DISPLAY_PLL_PS_ALLOCATION v1;
+};
+
+static u32 atombios_adjust_pll(struct drm_crtc *crtc,
+			       struct drm_display_mode *mode,
+			       struct radeon_pll *pll)
 {
-	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
 	struct radeon_device *rdev = dev->dev_private;
 	struct drm_encoder *encoder = NULL;
 	struct radeon_encoder *radeon_encoder = NULL;
-	uint8_t frev, crev;
-	int index;
-	SET_PIXEL_CLOCK_PS_ALLOCATION args;
-	PIXEL_CLOCK_PARAMETERS *spc1_ptr;
-	PIXEL_CLOCK_PARAMETERS_V2 *spc2_ptr;
-	PIXEL_CLOCK_PARAMETERS_V3 *spc3_ptr;
-	uint32_t pll_clock = mode->clock;
-	uint32_t adjusted_clock;
-	uint32_t ref_div = 0, fb_div = 0, frac_fb_div = 0, post_div = 0;
-	struct radeon_pll *pll;
-	int pll_flags = 0;
+	u32 adjusted_clock = mode->clock;
 
-	memset(&args, 0, sizeof(args));
+	/* reset the pll flags */
+	pll->flags = 0;
 
 	if (ASIC_IS_AVIVO(rdev)) {
 		if ((rdev->family == CHIP_RS600) ||
 		    (rdev->family == CHIP_RS690) ||
 		    (rdev->family == CHIP_RS740))
-			pll_flags |= (RADEON_PLL_USE_FRAC_FB_DIV |
-				      RADEON_PLL_PREFER_CLOSEST_LOWER);
+			pll->flags |= (RADEON_PLL_USE_FRAC_FB_DIV |
+				       RADEON_PLL_PREFER_CLOSEST_LOWER);
 
 		if (ASIC_IS_DCE32(rdev) && mode->clock > 200000)	/* range limits??? */
-			pll_flags |= RADEON_PLL_PREFER_HIGH_FB_DIV;
+			pll->flags |= RADEON_PLL_PREFER_HIGH_FB_DIV;
 		else
-			pll_flags |= RADEON_PLL_PREFER_LOW_REF_DIV;
+			pll->flags |= RADEON_PLL_PREFER_LOW_REF_DIV;
 	} else {
-		pll_flags |= RADEON_PLL_LEGACY;
+		pll->flags |= RADEON_PLL_LEGACY;
 
 		if (mode->clock > 200000)	/* range limits??? */
-			pll_flags |= RADEON_PLL_PREFER_HIGH_FB_DIV;
+			pll->flags |= RADEON_PLL_PREFER_HIGH_FB_DIV;
 		else
-			pll_flags |= RADEON_PLL_PREFER_LOW_REF_DIV;
+			pll->flags |= RADEON_PLL_PREFER_LOW_REF_DIV;
 
 	}
 
 	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
 		if (encoder->crtc == crtc) {
-			if (!ASIC_IS_AVIVO(rdev)) {
-				if (encoder->encoder_type !=
-				    DRM_MODE_ENCODER_DAC)
-					pll_flags |= RADEON_PLL_NO_ODD_POST_DIV;
-				if (encoder->encoder_type ==
-					DRM_MODE_ENCODER_LVDS)
-					pll_flags |= RADEON_PLL_USE_REF_DIV;
-			}
 			radeon_encoder = to_radeon_encoder(encoder);
+			if (ASIC_IS_AVIVO(rdev)) {
+				/* DVO wants 2x pixel clock if the DVO chip is in 12 bit mode */
+				if (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1)
+					adjusted_clock = mode->clock * 2;
+			} else {
+				if (encoder->encoder_type != DRM_MODE_ENCODER_DAC)
+					pll->flags |= RADEON_PLL_NO_ODD_POST_DIV;
+				if (encoder->encoder_type == DRM_MODE_ENCODER_LVDS)
+					pll->flags |= RADEON_PLL_USE_REF_DIV;
+			}
 			break;
 		}
 	}
@@ -471,46 +469,101 @@ void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode *mode)
 	 * special hw requirements.
 	 */
 	if (ASIC_IS_DCE3(rdev)) {
-		ADJUST_DISPLAY_PLL_PS_ALLOCATION adjust_pll_args;
+		union adjust_pixel_clock args;
+		struct radeon_encoder_atom_dig *dig;
+		u8 frev, crev;
+		int index;
 
-		if (!encoder)
-			return;
-
-		memset(&adjust_pll_args, 0, sizeof(adjust_pll_args));
-		adjust_pll_args.usPixelClock = cpu_to_le16(mode->clock / 10);
-		adjust_pll_args.ucTransmitterID = radeon_encoder->encoder_id;
-		adjust_pll_args.ucEncodeMode = atombios_get_encoder_mode(encoder);
+		if (!radeon_encoder->enc_priv)
+			return adjusted_clock;
+		dig = radeon_encoder->enc_priv;
 
 		index = GetIndexIntoMasterTable(COMMAND, AdjustDisplayPll);
-		atom_execute_table(rdev->mode_info.atom_context,
-				   index, (uint32_t *)&adjust_pll_args);
-		adjusted_clock = le16_to_cpu(adjust_pll_args.usPixelClock) * 10;
-	} else {
-		/* DVO wants 2x pixel clock if the DVO chip is in 12 bit mode */
-		if (ASIC_IS_AVIVO(rdev) &&
-		    (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1))
-			adjusted_clock = mode->clock * 2;
-		else
-			adjusted_clock = mode->clock;
+		atom_parse_cmd_header(rdev->mode_info.atom_context, index, &frev,
+				      &crev);
+
+		memset(&args, 0, sizeof(args));
+
+		switch (frev) {
+		case 1:
+			switch (crev) {
+			case 1:
+			case 2:
+				args.v1.usPixelClock = cpu_to_le16(mode->clock / 10);
+				args.v1.ucTransmitterID = radeon_encoder->encoder_id;
+				args.v1.ucEncodeMode = atombios_get_encoder_mode(encoder);
+
+				atom_execute_table(rdev->mode_info.atom_context,
+						   index, (uint32_t *)&args);
+				adjusted_clock = le16_to_cpu(args.v1.usPixelClock) * 10;
+				break;
+			default:
+				DRM_ERROR("Unknown table version %d %d\n", frev, crev);
+				return adjusted_clock;
+			}
+			break;
+		default:
+			DRM_ERROR("Unknown table version %d %d\n", frev, crev);
+			return adjusted_clock;
+		}
 	}
+	return adjusted_clock;
+}
+
+union set_pixel_clock {
+	SET_PIXEL_CLOCK_PS_ALLOCATION base;
+	PIXEL_CLOCK_PARAMETERS v1;
+	PIXEL_CLOCK_PARAMETERS_V2 v2;
+	PIXEL_CLOCK_PARAMETERS_V3 v3;
+};
+
+void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode *mode)
+{
+	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct drm_encoder *encoder = NULL;
+	struct radeon_encoder *radeon_encoder = NULL;
+	u8 frev, crev;
+	int index;
+	union set_pixel_clock args;
+	u32 pll_clock = mode->clock;
+	u32 ref_div = 0, fb_div = 0, frac_fb_div = 0, post_div = 0;
+	struct radeon_pll *pll;
+	u32 adjusted_clock;
+
+	memset(&args, 0, sizeof(args));
+
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
+		if (encoder->crtc == crtc) {
+			radeon_encoder = to_radeon_encoder(encoder);
+			break;
+		}
+	}
+
+	if (!radeon_encoder)
+		return;
 
 	if (radeon_crtc->crtc_id == 0)
 		pll = &rdev->clock.p1pll;
 	else
 		pll = &rdev->clock.p2pll;
 
+	/* adjust pixel clock as needed */
+	adjusted_clock = atombios_adjust_pll(crtc, mode, pll);
+
 	if (ASIC_IS_AVIVO(rdev)) {
 		if (radeon_new_pll)
 			radeon_compute_pll_avivo(pll, adjusted_clock, &pll_clock,
 						 &fb_div, &frac_fb_div,
-						 &ref_div, &post_div, pll_flags);
+						 &ref_div, &post_div);
 		else
 			radeon_compute_pll(pll, adjusted_clock, &pll_clock,
 					   &fb_div, &frac_fb_div,
-					   &ref_div, &post_div, pll_flags);
+					   &ref_div, &post_div);
 	} else
 		radeon_compute_pll(pll, adjusted_clock, &pll_clock, &fb_div, &frac_fb_div,
-				   &ref_div, &post_div, pll_flags);
+				   &ref_div, &post_div);
 
 	index = GetIndexIntoMasterTable(COMMAND, SetPixelClock);
 	atom_parse_cmd_header(rdev->mode_info.atom_context, index, &frev,
@@ -520,45 +573,38 @@ void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode *mode)
 	case 1:
 		switch (crev) {
 		case 1:
-			spc1_ptr = (PIXEL_CLOCK_PARAMETERS *) & args.sPCLKInput;
-			spc1_ptr->usPixelClock = cpu_to_le16(mode->clock / 10);
-			spc1_ptr->usRefDiv = cpu_to_le16(ref_div);
-			spc1_ptr->usFbDiv = cpu_to_le16(fb_div);
-			spc1_ptr->ucFracFbDiv = frac_fb_div;
-			spc1_ptr->ucPostDiv = post_div;
-			spc1_ptr->ucPpll =
+			args.v1.usPixelClock = cpu_to_le16(mode->clock / 10);
+			args.v1.usRefDiv = cpu_to_le16(ref_div);
+			args.v1.usFbDiv = cpu_to_le16(fb_div);
+			args.v1.ucFracFbDiv = frac_fb_div;
+			args.v1.ucPostDiv = post_div;
+			args.v1.ucPpll =
 			    radeon_crtc->crtc_id ? ATOM_PPLL2 : ATOM_PPLL1;
-			spc1_ptr->ucCRTC = radeon_crtc->crtc_id;
-			spc1_ptr->ucRefDivSrc = 1;
+			args.v1.ucCRTC = radeon_crtc->crtc_id;
+			args.v1.ucRefDivSrc = 1;
 			break;
 		case 2:
-			spc2_ptr =
-			    (PIXEL_CLOCK_PARAMETERS_V2 *) & args.sPCLKInput;
-			spc2_ptr->usPixelClock = cpu_to_le16(mode->clock / 10);
-			spc2_ptr->usRefDiv = cpu_to_le16(ref_div);
-			spc2_ptr->usFbDiv = cpu_to_le16(fb_div);
-			spc2_ptr->ucFracFbDiv = frac_fb_div;
-			spc2_ptr->ucPostDiv = post_div;
-			spc2_ptr->ucPpll =
+			args.v2.usPixelClock = cpu_to_le16(mode->clock / 10);
+			args.v2.usRefDiv = cpu_to_le16(ref_div);
+			args.v2.usFbDiv = cpu_to_le16(fb_div);
+			args.v2.ucFracFbDiv = frac_fb_div;
+			args.v2.ucPostDiv = post_div;
+			args.v2.ucPpll =
 			    radeon_crtc->crtc_id ? ATOM_PPLL2 : ATOM_PPLL1;
-			spc2_ptr->ucCRTC = radeon_crtc->crtc_id;
-			spc2_ptr->ucRefDivSrc = 1;
+			args.v2.ucCRTC = radeon_crtc->crtc_id;
+			args.v2.ucRefDivSrc = 1;
 			break;
 		case 3:
-			if (!encoder)
-				return;
-			spc3_ptr =
-			    (PIXEL_CLOCK_PARAMETERS_V3 *) & args.sPCLKInput;
-			spc3_ptr->usPixelClock = cpu_to_le16(mode->clock / 10);
-			spc3_ptr->usRefDiv = cpu_to_le16(ref_div);
-			spc3_ptr->usFbDiv = cpu_to_le16(fb_div);
-			spc3_ptr->ucFracFbDiv = frac_fb_div;
-			spc3_ptr->ucPostDiv = post_div;
-			spc3_ptr->ucPpll =
+			args.v3.usPixelClock = cpu_to_le16(mode->clock / 10);
+			args.v3.usRefDiv = cpu_to_le16(ref_div);
+			args.v3.usFbDiv = cpu_to_le16(fb_div);
+			args.v3.ucFracFbDiv = frac_fb_div;
+			args.v3.ucPostDiv = post_div;
+			args.v3.ucPpll =
 			    radeon_crtc->crtc_id ? ATOM_PPLL2 : ATOM_PPLL1;
-			spc3_ptr->ucMiscInfo = (radeon_crtc->crtc_id << 2);
-			spc3_ptr->ucTransmitterId = radeon_encoder->encoder_id;
-			spc3_ptr->ucEncoderMode =
+			args.v3.ucMiscInfo = (radeon_crtc->crtc_id << 2);
+			args.v3.ucTransmitterId = radeon_encoder->encoder_id;
+			args.v3.ucEncoderMode =
 			    atombios_get_encoder_mode(encoder);
 			break;
 		default:
@@ -571,12 +617,11 @@ void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode *mode)
 		return;
 	}
 
-	printk("executing set pll\n");
 	atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args);
 }
 
-int atombios_crtc_set_base(struct drm_crtc *crtc, int x, int y,
-			   struct drm_framebuffer *old_fb)
+static int avivo_crtc_set_base(struct drm_crtc *crtc, int x, int y,
+			       struct drm_framebuffer *old_fb)
 {
 	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
@@ -706,6 +751,42 @@ int atombios_crtc_set_base(struct drm_crtc *crtc, int x, int y,
 	return 0;
 }
 
+int atombios_crtc_set_base(struct drm_crtc *crtc, int x, int y,
+			   struct drm_framebuffer *old_fb)
+{
+	struct drm_device *dev = crtc->dev;
+	struct radeon_device *rdev = dev->dev_private;
+
+	if (ASIC_IS_AVIVO(rdev))
+		return avivo_crtc_set_base(crtc, x, y, old_fb);
+	else
+		return radeon_crtc_set_base(crtc, x, y, old_fb);
+}
+
+/* properly set additional regs when using atombios */
+static void radeon_legacy_atom_fixup(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc);
+	u32 disp_merge_cntl;
+
+	switch (radeon_crtc->crtc_id) {
+	case 0:
+		disp_merge_cntl = RREG32(RADEON_DISP_MERGE_CNTL);
+		disp_merge_cntl &= ~RADEON_DISP_RGB_OFFSET_EN;
+		WREG32(RADEON_DISP_MERGE_CNTL, disp_merge_cntl);
+		break;
+	case 1:
+		disp_merge_cntl = RREG32(RADEON_DISP2_MERGE_CNTL);
+		disp_merge_cntl &= ~RADEON_DISP2_RGB_OFFSET_EN;
+		WREG32(RADEON_DISP2_MERGE_CNTL, disp_merge_cntl);
+		WREG32(RADEON_FP_H2_SYNC_STRT_WID,   RREG32(RADEON_CRTC2_H_SYNC_STRT_WID));
+		WREG32(RADEON_FP_V2_SYNC_STRT_WID,   RREG32(RADEON_CRTC2_V_SYNC_STRT_WID));
+		break;
+	}
+}
+
 int atombios_crtc_mode_set(struct drm_crtc *crtc,
 			   struct drm_display_mode *mode,
 			   struct drm_display_mode *adjusted_mode,
@@ -727,8 +808,8 @@ int atombios_crtc_mode_set(struct drm_crtc *crtc,
 	else {
 		if (radeon_crtc->crtc_id == 0)
 			atombios_set_crtc_dtd_timing(crtc, adjusted_mode);
-		radeon_crtc_set_base(crtc, x, y, old_fb);
-		radeon_legacy_atom_set_surface(crtc);
+		atombios_crtc_set_base(crtc, x, y, old_fb);
+		radeon_legacy_atom_fixup(crtc);
 	}
 	atombios_overscan_setup(crtc, mode, adjusted_mode);
 	atombios_scaler_setup(crtc);
@@ -746,8 +827,8 @@ static bool atombios_crtc_mode_fixup(struct drm_crtc *crtc,
 
 static void atombios_crtc_prepare(struct drm_crtc *crtc)
 {
-	atombios_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
 	atombios_lock_crtc(crtc, 1);
+	atombios_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
 }
 
 static void atombios_crtc_commit(struct drm_crtc *crtc)
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 62b27bede651..11c9a3fe6810 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -1504,6 +1504,7 @@ static int r100_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("PRIM_WALK must be 3 for IMMD draw\n");
 			return -EINVAL;
 		}
+		track->vtx_size = r100_get_vtx_size(radeon_get_ib_value(p, idx + 0));
 		track->vap_vf_cntl = radeon_get_ib_value(p, idx + 1);
 		track->immd_dwords = pkt->count - 1;
 		r = r100_cs_track_check(p->rdev, track);
diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c
index 20942127c46b..ff1e0cd608bf 100644
--- a/drivers/gpu/drm/radeon/r200.c
+++ b/drivers/gpu/drm/radeon/r200.c
@@ -371,13 +371,16 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 		case 5:
 		case 6:
 		case 7:
+			/* 1D/2D */
 			track->textures[i].tex_coord_type = 0;
 			break;
 		case 1:
-			track->textures[i].tex_coord_type = 1;
+			/* CUBE */
+			track->textures[i].tex_coord_type = 2;
 			break;
 		case 2:
-			track->textures[i].tex_coord_type = 2;
+			/* 3D */
+			track->textures[i].tex_coord_type = 1;
 			break;
 		}
 		break;
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index b3713f61964c..d0bd117a463a 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -1954,6 +1954,7 @@ int r600_suspend(struct radeon_device *rdev)
 	/* FIXME: we should wait for ring to be empty */
 	r600_cp_stop(rdev);
 	rdev->cp.ready = false;
+	r600_irq_suspend(rdev);
 	r600_wb_disable(rdev);
 	r600_pcie_gart_disable(rdev);
 	/* unpin shaders bo */
@@ -2063,13 +2064,14 @@ int r600_init(struct radeon_device *rdev)
 	if (rdev->accel_working) {
 		r = radeon_ib_pool_init(rdev);
 		if (r) {
-			DRM_ERROR("radeon: failed initializing IB pool (%d).\n", r);
-			rdev->accel_working = false;
-		}
-		r = r600_ib_test(rdev);
-		if (r) {
-			DRM_ERROR("radeon: failed testing IB (%d).\n", r);
+			dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
 			rdev->accel_working = false;
+		} else {
+			r = r600_ib_test(rdev);
+			if (r) {
+				dev_err(rdev->dev, "IB test failed (%d).\n", r);
+				rdev->accel_working = false;
+			}
 		}
 	}
 
@@ -2200,14 +2202,14 @@ void r600_ih_ring_init(struct radeon_device *rdev, unsigned ring_size)
 	rb_bufsz = drm_order(ring_size / 4);
 	ring_size = (1 << rb_bufsz) * 4;
 	rdev->ih.ring_size = ring_size;
-	rdev->ih.align_mask = 4 - 1;
+	rdev->ih.ptr_mask = rdev->ih.ring_size - 1;
+	rdev->ih.rptr = 0;
 }
 
-static int r600_ih_ring_alloc(struct radeon_device *rdev, unsigned ring_size)
+static int r600_ih_ring_alloc(struct radeon_device *rdev)
 {
 	int r;
 
-	rdev->ih.ring_size = ring_size;
 	/* Allocate ring buffer */
 	if (rdev->ih.ring_obj == NULL) {
 		r = radeon_bo_create(rdev, NULL, rdev->ih.ring_size,
@@ -2237,9 +2239,6 @@ static int r600_ih_ring_alloc(struct radeon_device *rdev, unsigned ring_size)
 			return r;
 		}
 	}
-	rdev->ih.ptr_mask = (rdev->cp.ring_size / 4) - 1;
-	rdev->ih.rptr = 0;
-
 	return 0;
 }
 
@@ -2389,7 +2388,7 @@ int r600_irq_init(struct radeon_device *rdev)
 	u32 interrupt_cntl, ih_cntl, ih_rb_cntl;
 
 	/* allocate ring */
-	ret = r600_ih_ring_alloc(rdev, rdev->ih.ring_size);
+	ret = r600_ih_ring_alloc(rdev);
 	if (ret)
 		return ret;
 
@@ -2452,10 +2451,15 @@ int r600_irq_init(struct radeon_device *rdev)
 	return ret;
 }
 
-void r600_irq_fini(struct radeon_device *rdev)
+void r600_irq_suspend(struct radeon_device *rdev)
 {
 	r600_disable_interrupts(rdev);
 	r600_rlc_stop(rdev);
+}
+
+void r600_irq_fini(struct radeon_device *rdev)
+{
+	r600_irq_suspend(rdev);
 	r600_ih_ring_fini(rdev);
 }
 
@@ -2470,8 +2474,12 @@ int r600_irq_set(struct radeon_device *rdev)
 		return -EINVAL;
 	}
 	/* don't enable anything if the ih is disabled */
-	if (!rdev->ih.enabled)
+	if (!rdev->ih.enabled) {
+		r600_disable_interrupts(rdev);
+		/* force the active interrupt state to all disabled */
+		r600_disable_interrupt_state(rdev);
 		return 0;
+	}
 
 	if (ASIC_IS_DCE3(rdev)) {
 		hpd1 = RREG32(DC_HPD1_INT_CONTROL) & ~DC_HPDx_INT_EN;
@@ -2641,16 +2649,18 @@ static inline u32 r600_get_ih_wptr(struct radeon_device *rdev)
 	wptr = RREG32(IH_RB_WPTR);
 
 	if (wptr & RB_OVERFLOW) {
-		WARN_ON(1);
-		/* XXX deal with overflow */
-		DRM_ERROR("IH RB overflow\n");
+		/* When a ring buffer overflow happen start parsing interrupt
+		 * from the last not overwritten vector (wptr + 16). Hopefully
+		 * this should allow us to catchup.
+		 */
+		dev_warn(rdev->dev, "IH ring buffer overflow (0x%08X, %d, %d)\n",
+			wptr, rdev->ih.rptr, (wptr + 16) + rdev->ih.ptr_mask);
+		rdev->ih.rptr = (wptr + 16) & rdev->ih.ptr_mask;
 		tmp = RREG32(IH_RB_CNTL);
 		tmp |= IH_WPTR_OVERFLOW_CLEAR;
 		WREG32(IH_RB_CNTL, tmp);
 	}
-	wptr = wptr & WPTR_OFFSET_MASK;
-
-	return wptr;
+	return (wptr & rdev->ih.ptr_mask);
 }
 
 /*        r600 IV Ring
@@ -2686,12 +2696,13 @@ int r600_irq_process(struct radeon_device *rdev)
 	u32 wptr = r600_get_ih_wptr(rdev);
 	u32 rptr = rdev->ih.rptr;
 	u32 src_id, src_data;
-	u32 last_entry = rdev->ih.ring_size - 16;
 	u32 ring_index, disp_int, disp_int_cont, disp_int_cont2;
 	unsigned long flags;
 	bool queue_hotplug = false;
 
 	DRM_DEBUG("r600_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
+	if (!rdev->ih.enabled)
+		return IRQ_NONE;
 
 	spin_lock_irqsave(&rdev->ih.lock, flags);
 
@@ -2820,10 +2831,8 @@ restart_ih:
 		}
 
 		/* wptr/rptr are in bytes! */
-		if (rptr == last_entry)
-			rptr = 0;
-		else
-			rptr += 16;
+		rptr += 16;
+		rptr &= rdev->ih.ptr_mask;
 	}
 	/* make sure wptr hasn't changed while processing */
 	wptr = r600_get_ih_wptr(rdev);
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 44060b92d9e6..9cc4ba8d03d5 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -36,6 +36,10 @@ static int r600_cs_packet_next_reloc_nomm(struct radeon_cs_parser *p,
 typedef int (*next_reloc_t)(struct radeon_cs_parser*, struct radeon_cs_reloc**);
 static next_reloc_t r600_cs_packet_next_reloc = &r600_cs_packet_next_reloc_mm;
 
+struct r600_cs_track {
+	u32	cb_color0_base_last;
+};
+
 /**
  * r600_cs_packet_parse() - parse cp packet and point ib index to next packet
  * @parser:	parser structure holding parsing context.
@@ -177,6 +181,28 @@ static int r600_cs_packet_next_reloc_nomm(struct radeon_cs_parser *p,
 }
 
 /**
+ * r600_cs_packet_next_is_pkt3_nop() - test if next packet is packet3 nop for reloc
+ * @parser:		parser structure holding parsing context.
+ *
+ * Check next packet is relocation packet3, do bo validation and compute
+ * GPU offset using the provided start.
+ **/
+static inline int r600_cs_packet_next_is_pkt3_nop(struct radeon_cs_parser *p)
+{
+	struct radeon_cs_packet p3reloc;
+	int r;
+
+	r = r600_cs_packet_parse(p, &p3reloc, p->idx);
+	if (r) {
+		return 0;
+	}
+	if (p3reloc.type != PACKET_TYPE3 || p3reloc.opcode != PACKET3_NOP) {
+		return 0;
+	}
+	return 1;
+}
+
+/**
  * r600_cs_packet_next_vline() - parse userspace VLINE packet
  * @parser:		parser structure holding parsing context.
  *
@@ -337,6 +363,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 				struct radeon_cs_packet *pkt)
 {
 	struct radeon_cs_reloc *reloc;
+	struct r600_cs_track *track;
 	volatile u32 *ib;
 	unsigned idx;
 	unsigned i;
@@ -344,6 +371,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 	int r;
 	u32 idx_value;
 
+	track = (struct r600_cs_track *)p->track;
 	ib = p->ib->ptr;
 	idx = pkt->idx + 1;
 	idx_value = radeon_get_ib_value(p, idx);
@@ -503,9 +531,60 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		for (i = 0; i < pkt->count; i++) {
 			reg = start_reg + (4 * i);
 			switch (reg) {
+			/* This register were added late, there is userspace
+			 * which does provide relocation for those but set
+			 * 0 offset. In order to avoid breaking old userspace
+			 * we detect this and set address to point to last
+			 * CB_COLOR0_BASE, note that if userspace doesn't set
+			 * CB_COLOR0_BASE before this register we will report
+			 * error. Old userspace always set CB_COLOR0_BASE
+			 * before any of this.
+			 */
+			case R_0280E0_CB_COLOR0_FRAG:
+			case R_0280E4_CB_COLOR1_FRAG:
+			case R_0280E8_CB_COLOR2_FRAG:
+			case R_0280EC_CB_COLOR3_FRAG:
+			case R_0280F0_CB_COLOR4_FRAG:
+			case R_0280F4_CB_COLOR5_FRAG:
+			case R_0280F8_CB_COLOR6_FRAG:
+			case R_0280FC_CB_COLOR7_FRAG:
+			case R_0280C0_CB_COLOR0_TILE:
+			case R_0280C4_CB_COLOR1_TILE:
+			case R_0280C8_CB_COLOR2_TILE:
+			case R_0280CC_CB_COLOR3_TILE:
+			case R_0280D0_CB_COLOR4_TILE:
+			case R_0280D4_CB_COLOR5_TILE:
+			case R_0280D8_CB_COLOR6_TILE:
+			case R_0280DC_CB_COLOR7_TILE:
+				if (!r600_cs_packet_next_is_pkt3_nop(p)) {
+					if (!track->cb_color0_base_last) {
+						dev_err(p->dev, "Broken old userspace ? no cb_color0_base supplied before trying to write 0x%08X\n", reg);
+						return -EINVAL;
+					}
+					ib[idx+1+i] = track->cb_color0_base_last;
+					printk_once(KERN_WARNING "You have old & broken userspace "
+						"please consider updating mesa & xf86-video-ati\n");
+				} else {
+					r = r600_cs_packet_next_reloc(p, &reloc);
+					if (r) {
+						dev_err(p->dev, "bad SET_CONTEXT_REG 0x%04X\n", reg);
+						return -EINVAL;
+					}
+					ib[idx+1+i] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
+				}
+				break;
 			case DB_DEPTH_BASE:
 			case DB_HTILE_DATA_BASE:
 			case CB_COLOR0_BASE:
+				r = r600_cs_packet_next_reloc(p, &reloc);
+				if (r) {
+					DRM_ERROR("bad SET_CONTEXT_REG "
+							"0x%04X\n", reg);
+					return -EINVAL;
+				}
+				ib[idx+1+i] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
+				track->cb_color0_base_last = ib[idx+1+i];
+				break;
 			case CB_COLOR1_BASE:
 			case CB_COLOR2_BASE:
 			case CB_COLOR3_BASE:
@@ -678,8 +757,11 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 int r600_cs_parse(struct radeon_cs_parser *p)
 {
 	struct radeon_cs_packet pkt;
+	struct r600_cs_track *track;
 	int r;
 
+	track = kzalloc(sizeof(*track), GFP_KERNEL);
+	p->track = track;
 	do {
 		r = r600_cs_packet_parse(p, &pkt, p->idx);
 		if (r) {
@@ -757,6 +839,7 @@ int r600_cs_legacy(struct drm_device *dev, void *data, struct drm_file *filp,
 	/* initialize parser */
 	memset(&parser, 0, sizeof(struct radeon_cs_parser));
 	parser.filp = filp;
+	parser.dev = &dev->pdev->dev;
 	parser.rdev = NULL;
 	parser.family = family;
 	parser.ib = &fake_ib;
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index 05894edadab4..30480881aed1 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -882,4 +882,29 @@
 #define		S_000E60_SOFT_RESET_VMC(x)		(((x) & 1) << 17)
 
 #define R_005480_HDP_MEM_COHERENCY_FLUSH_CNTL		0x5480
+
+#define R_0280E0_CB_COLOR0_FRAG                      0x0280E0
+#define   S_0280E0_BASE_256B(x)                        (((x) & 0xFFFFFFFF) << 0)
+#define   G_0280E0_BASE_256B(x)                        (((x) >> 0) & 0xFFFFFFFF)
+#define   C_0280E0_BASE_256B                           0x00000000
+#define R_0280E4_CB_COLOR1_FRAG                      0x0280E4
+#define R_0280E8_CB_COLOR2_FRAG                      0x0280E8
+#define R_0280EC_CB_COLOR3_FRAG