r600g: avoid unnecessary shader exports v2

In some cases TGSI shader has more color outputs than the number of CBs,
so it seems we need to limit the number of color exports. This requires
different shader variants depending on the nr_cbufs, but on the other hand
we are doing less exports, which are very costly.

v2: fix various piglit regressions

Signed-off-by: Vadim Girlin <[email protected]>
Signed-off-by: Jerome Glisse <[email protected]>

6 files changed, 41 insertions, 28 deletions

diff --git a/src/gallium/drivers/r600/evergreen_state.c b/src/gallium/drivers/r600/evergreen_state.c
index 4b82b06f177..895d8a2c4d4 100644
--- a/src/gallium/drivers/r600/evergreen_state.c
+++ b/src/gallium/drivers/r600/evergreen_state.c
@@ -2646,18 +2646,14 @@ void evergreen_pipe_shader_ps(struct pipe_context *ctx, struct r600_pipe_shader
 		db_shader_control |= S_02880C_KILL_ENABLE(1);
 
 	exports_ps = 0;
-	num_cout = 0;
 	for (i = 0; i < rshader->noutput; i++) {
 		if (rshader->output[i].name == TGSI_SEMANTIC_POSITION ||
 		    rshader->output[i].name == TGSI_SEMANTIC_STENCIL)
 			exports_ps |= 1;
-		else if (rshader->output[i].name == TGSI_SEMANTIC_COLOR) {
-			if (rshader->fs_write_all)
-				num_cout = rshader->nr_cbufs;
-			else
-				num_cout++;
-		}
 	}
+
+	num_cout = rshader->nr_ps_color_exports;
+
 	exports_ps |= S_02884C_EXPORT_COLORS(num_cout);
 	if (!exports_ps) {
 		/* always at least export 1 component per pixel */
diff --git a/src/gallium/drivers/r600/r600_pipe.h b/src/gallium/drivers/r600/r600_pipe.h
index 5d194e325f7..380fd660c29 100644
--- a/src/gallium/drivers/r600/r600_pipe.h
+++ b/src/gallium/drivers/r600/r600_pipe.h
@@ -198,10 +198,7 @@ struct r600_pipe_shader_selector {
 	/* PIPE_SHADER_[VERTEX|FRAGMENT|...] */
 	unsigned	type;
 
-	/* 1 on evergreen+ when the shader contains
-	 * TGSI_PROPERTY_FS_COLOR0_WRITES_ALL_CBUFS, otherwise it's 0.
-	 * Used to determine whether we need to include nr_cbufs in the key */
-	unsigned	eg_fs_write_all;
+	unsigned	nr_ps_max_color_exports;
 };
 
 struct r600_pipe_shader {
diff --git a/src/gallium/drivers/r600/r600_shader.c b/src/gallium/drivers/r600/r600_shader.c
index f690c10fd7c..32d5a78758f 100644
--- a/src/gallium/drivers/r600/r600_shader.c
+++ b/src/gallium/drivers/r600/r600_shader.c
@@ -800,6 +800,12 @@ static int tgsi_declaration(struct r600_shader_ctx *ctx)
 				ctx->cv_output = i;
 				break;
 			}
+		} else if (ctx->type == TGSI_PROCESSOR_FRAGMENT) {
+			switch (d->Semantic.Name) {
+			case TGSI_SEMANTIC_COLOR:
+				ctx->shader->nr_ps_max_color_exports++;
+				break;
+			}
 		}
 		break;
 	case TGSI_FILE_CONSTANT:
@@ -1152,8 +1158,10 @@ static int r600_shader_from_tgsi(struct r600_context * rctx, struct r600_pipe_sh
 	ctx.colors_used = 0;
 	ctx.clip_vertex_write = 0;
 
+	shader->nr_ps_color_exports = 0;
+	shader->nr_ps_max_color_exports = 0;
+
 	shader->two_side = (ctx.type == TGSI_PROCESSOR_FRAGMENT) && rctx->two_side;
-	shader->nr_cbufs = rctx->nr_cbufs;
 
 	/* register allocations */
 	/* Values [0,127] correspond to GPR[0..127].
@@ -1288,6 +1296,9 @@ static int r600_shader_from_tgsi(struct r600_context * rctx, struct r600_pipe_sh
 		}
 	}
 
+	if (shader->fs_write_all && rctx->chip_class >= EVERGREEN)
+		shader->nr_ps_max_color_exports = 8;
+
 	if (ctx.fragcoord_input >= 0) {
 		if (ctx.bc->chip_class == CAYMAN) {
 			for (j = 0 ; j < 4; j++) {
@@ -1527,10 +1538,17 @@ static int r600_shader_from_tgsi(struct r600_context * rctx, struct r600_pipe_sh
 			break;
 		case TGSI_PROCESSOR_FRAGMENT:
 			if (shader->output[i].name == TGSI_SEMANTIC_COLOR) {
+				/* never export more colors than the number of CBs */
+				if (next_pixel_base && next_pixel_base >= (rctx->nr_cbufs + rctx->dual_src_blend * 1)) {
+					/* skip export */
+					j--;
+					continue;
+				}
 				output[j].array_base = next_pixel_base++;
 				output[j].type = V_SQ_CF_ALLOC_EXPORT_WORD0_SQ_EXPORT_PIXEL;
+				shader->nr_ps_color_exports++;
 				if (shader->fs_write_all && (rctx->chip_class >= EVERGREEN)) {
-					for (k = 1; k < shader->nr_cbufs; k++) {
+					for (k = 1; k < rctx->nr_cbufs; k++) {
 						j++;
 						memset(&output[j], 0, sizeof(struct r600_bytecode_output));
 						output[j].gpr = shader->output[i].gpr;
@@ -1544,6 +1562,7 @@ static int r600_shader_from_tgsi(struct r600_context * rctx, struct r600_pipe_sh
 						output[j].array_base = next_pixel_base++;
 						output[j].inst = BC_INST(ctx.bc, V_SQ_CF_ALLOC_EXPORT_WORD1_SQ_CF_INST_EXPORT);
 						output[j].type = V_SQ_CF_ALLOC_EXPORT_WORD0_SQ_EXPORT_PIXEL;
+						shader->nr_ps_color_exports++;
 					}
 				}
 			} else if (shader->output[i].name == TGSI_SEMANTIC_POSITION) {
@@ -1594,7 +1613,7 @@ static int r600_shader_from_tgsi(struct r600_context * rctx, struct r600_pipe_sh
 	}
 
 	/* add fake pixel export */
-	if (ctx.type == TGSI_PROCESSOR_FRAGMENT && j == 0) {
+	if (ctx.type == TGSI_PROCESSOR_FRAGMENT && next_pixel_base == 0) {
 		memset(&output[j], 0, sizeof(struct r600_bytecode_output));
 		output[j].gpr = 0;
 		output[j].elem_size = 3;
diff --git a/src/gallium/drivers/r600/r600_shader.h b/src/gallium/drivers/r600/r600_shader.h
index 2d35e770070..eb0bbf6ebb7 100644
--- a/src/gallium/drivers/r600/r600_shader.h
+++ b/src/gallium/drivers/r600/r600_shader.h
@@ -49,7 +49,12 @@ struct r600_shader {
 	boolean			fs_write_all;
 	boolean			vs_prohibit_ucps;
 	boolean			two_side;
-	unsigned		nr_cbufs;
+	/* Number of color outputs in the TGSI shader,
+	 * sometimes it could be higher than nr_cbufs (bug?).
+	 * Also with writes_all property on eg+ it will be set to max CB number */
+	unsigned		nr_ps_max_color_exports;
+	/* Real number of ps color exports compiled in the bytecode */
+	unsigned		nr_ps_color_exports;
 	/* bit n is set if the shader writes gl_ClipDistance[n] */
 	unsigned		clip_dist_write;
 	/* flag is set if the shader writes VS_OUT_MISC_VEC (e.g. for PSIZE) */
diff --git a/src/gallium/drivers/r600/r600_state.c b/src/gallium/drivers/r600/r600_state.c
index 295453a5aef..d621aee9201 100644
--- a/src/gallium/drivers/r600/r600_state.c
+++ b/src/gallium/drivers/r600/r600_state.c
@@ -1644,6 +1644,8 @@ static void r600_set_framebuffer_state(struct pipe_context *ctx,
 
 	/* build states */
 	rctx->have_depth_fb = 0;
+	rctx->nr_cbufs = state->nr_cbufs;
+
 	for (int i = 0; i < state->nr_cbufs; i++) {
 		r600_cb(rctx, rstate, state, i);
 	}
diff --git a/src/gallium/drivers/r600/r600_state_common.c b/src/gallium/drivers/r600/r600_state_common.c
index 879dcc50edb..6050fa04972 100644
--- a/src/gallium/drivers/r600/r600_state_common.c
+++ b/src/gallium/drivers/r600/r600_state_common.c
@@ -437,9 +437,8 @@ static INLINE unsigned r600_shader_selector_key(struct pipe_context * ctx,
 	unsigned key;
 
 	if (sel->type == PIPE_SHADER_FRAGMENT) {
-		key = rctx->two_side;
-		if (sel->eg_fs_write_all)
-			key |= rctx->nr_cbufs << 1;
+		key = rctx->two_side |
+				MIN2(sel->nr_ps_max_color_exports, rctx->nr_cbufs + rctx->dual_src_blend) << 1;
 	} else
 		key = 0;
 
@@ -494,14 +493,12 @@ static int r600_shader_select(struct pipe_context *ctx,
 			return r;
 		}
 
-		/* We don't know the value of eg_fs_write_all property until we built
-		 * at least one variant, so we may need to recompute the key (include
-		 * rctx->nr_cbufs) after building first variant. */
+		/* We don't know the value of nr_ps_max_color_exports until we built
+		 * at least one variant, so we may need to recompute the key after
+		 * building first variant. */
 		if (sel->type == PIPE_SHADER_FRAGMENT &&
-				sel->num_shaders == 0 &&
-				rctx->chip_class >= EVERGREEN &&
-				shader->shader.fs_write_all) {
-			sel->eg_fs_write_all = 1;
+				sel->num_shaders == 0) {
+			sel->nr_ps_max_color_exports = shader->shader.nr_ps_max_color_exports;
 			key = r600_shader_selector_key(ctx, sel);
 		}
 
@@ -515,9 +512,6 @@ static int r600_shader_select(struct pipe_context *ctx,
 	shader->next_variant = sel->current;
 	sel->current = shader;
 
-	/* Moved from r600_bind_ps_shader, different shader variants
-	 * may use different number of GPRs, so we need to update it. */
-	/* FIXME: we never did it after rebuilding the shaders, is it required? */
 	if (rctx->chip_class < EVERGREEN && rctx->ps_shader && rctx->vs_shader) {
 		r600_adjust_gprs(rctx);
 	}