From: Christian König <deathsimple@vodafone.de>
Date: Sat, 7 May 2011 12:11:40 +0000 (+0200)
Subject: Merge remote-tracking branch 'origin/master' into pipe-video
X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=213b9004a6ee033a16af3dcd187aa68b56c39858;p=mesa.git

Merge remote-tracking branch 'origin/master' into pipe-video

Conflicts:
	src/gallium/drivers/r600/r600_state.c
---

213b9004a6ee033a16af3dcd187aa68b56c39858
diff --cc src/gallium/drivers/r600/r600_state.c
index 8646b9f4905,3f979cf170f..b9084c953ee
--- a/src/gallium/drivers/r600/r600_state.c
+++ b/src/gallium/drivers/r600/r600_state.c
@@@ -780,17 -769,36 +774,36 @@@ static void r600_cb(struct r600_pipe_co
  		S_0280A0_NUMBER_TYPE(ntype) |
  		S_0280A0_ENDIAN(endian);
  
- 	color_info_mask = 0xFFFFFFFF & ~S_0280A0_BLEND_CLAMP(1);
- 
- 	/* on R600 this can't be set if BLEND_CLAMP isn't set,
- 	   if BLEND_FLOAT32 is set of > 11 bits in a UNORM or SNORM */
- 	if (desc->colorspace != UTIL_FORMAT_COLORSPACE_ZS && desc->channel[i].size < 12) {
- 		//TODO: Seems to work on RV710, but i have no idea what to do between R600-RV710
- 		if (rctx->family < CHIP_RV710) {
- 			color_info |= S_0280A0_BLEND_CLAMP(1);
- 			color_info_mask |= S_0280A0_BLEND_CLAMP(1);
- 		}
- 		color_info |= S_0280A0_SOURCE_FORMAT(V_0280A0_EXPORT_NORM);
+ 	/* EXPORT_NORM is an optimzation that can be enabled for better
+ 	 * performance in certain cases
+ 	 */
+ 	if (rctx->family < CHIP_RV770) {
+ 		/* EXPORT_NORM can be enabled if:
+ 		 * - 11-bit or smaller UNORM/SNORM/SRGB
+ 		 * - BLEND_CLAMP is enabled
+ 		 * - BLEND_FLOAT32 is disabled
+ 		 */
++		// TODO get BLEND_CLAMP state from rasterizer state
+ 		if (desc->colorspace != UTIL_FORMAT_COLORSPACE_ZS &&
+ 		    (desc->channel[i].size < 12 &&
+ 		     desc->channel[i].type != UTIL_FORMAT_TYPE_FLOAT &&
+ 		     ntype != V_0280A0_NUMBER_UINT &&
+ 		     ntype != V_0280A0_NUMBER_SINT) &&
+ 		    G_0280A0_BLEND_CLAMP(color_info) &&
+ 		    !G_0280A0_BLEND_FLOAT32(color_info))
+ 			color_info |= S_0280A0_SOURCE_FORMAT(V_0280A0_EXPORT_NORM);
+ 	} else {
+ 		/* EXPORT_NORM can be enabled if:
+ 		 * - 11-bit or smaller UNORM/SNORM/SRGB
+ 		 * - 16-bit or smaller FLOAT
+ 		 */
+ 		if (desc->colorspace != UTIL_FORMAT_COLORSPACE_ZS &&
+ 		    ((desc->channel[i].size < 12 &&
+ 		      desc->channel[i].type != UTIL_FORMAT_TYPE_FLOAT &&
+ 		      ntype != V_0280A0_NUMBER_UINT && ntype != V_0280A0_NUMBER_SINT) ||
+ 		    (desc->channel[i].size < 17 &&
+ 		     desc->channel[i].type == UTIL_FORMAT_TYPE_FLOAT)))
+ 			color_info |= S_0280A0_SOURCE_FORMAT(V_0280A0_EXPORT_NORM);
  	}
  
  	r600_pipe_state_add_reg(rstate,
@@@ -798,7 -806,7 +811,7 @@@
  				(offset + r600_bo_offset(bo[0])) >> 8, 0xFFFFFFFF, bo[0]);
  	r600_pipe_state_add_reg(rstate,
  				R_0280A0_CB_COLOR0_INFO + cb * 4,
- 				color_info, color_info_mask, NULL);
 -				color_info, 0xFFFFFFFF, bo[0]);
++				color_info, ~S_0280A0_BLEND_CLAMP(1), NULL);
  	r600_pipe_state_add_reg(rstate,
  				R_028060_CB_COLOR0_SIZE + cb * 4,
  				S_028060_PITCH_TILE_MAX(pitch) |
diff --cc src/gallium/winsys/r600/drm/r600_hw_context.c
index 935ac07802e,0514bbeaba3..34ea2a6bacb
--- a/src/gallium/winsys/r600/drm/r600_hw_context.c
+++ b/src/gallium/winsys/r600/drm/r600_hw_context.c
@@@ -1124,17 -1145,16 +1145,18 @@@ void r600_context_block_emit_dirty(stru
  		if (block->pm4_bo_index[j]) {
  			/* find relocation */
  			id = block->pm4_bo_index[j];
 -			r600_context_bo_reloc(ctx,
 -					&block->pm4[block->reloc[id].bo_pm4_index],
 -					block->reloc[id].bo);
 -			r600_context_bo_flush(ctx,
 -					block->reloc[id].flush_flags,
 -					block->reloc[id].flush_mask,
 -					block->reloc[id].bo);
 +			if (block->reloc[id].bo) {
 +				r600_context_bo_reloc(ctx,
 +						&block->pm4[block->reloc[id].bo_pm4_index],
 +						block->reloc[id].bo);
 +				r600_context_bo_flush(ctx,
 +						block->reloc[id].flush_flags,
 +						block->reloc[id].flush_mask,
 +						block->reloc[id].bo);
 +			}
  		}
  	}
+ 	ctx->flags &= ~R600_CONTEXT_CHECK_EVENT_FLUSH;
  	memcpy(&ctx->pm4[ctx->pm4_cdwords], block->pm4, block->pm4_ndwords * 4);
  	ctx->pm4_cdwords += block->pm4_ndwords;