#include "brw_ff_gs.h"
-static bool
-do_gs_prog(struct brw_context *brw,
- struct gl_shader_program *prog,
- struct brw_geometry_program *gp,
- struct brw_gs_prog_key *key)
+bool
+brw_codegen_gs_prog(struct brw_context *brw,
+ struct gl_shader_program *prog,
+ struct brw_geometry_program *gp,
+ struct brw_gs_prog_key *key)
{
struct brw_stage_state *stage_state = &brw->gs.base;
struct brw_gs_compile c;
rzalloc_array(NULL, const gl_constant_value *, param_count);
c.prog_data.base.base.pull_param =
rzalloc_array(NULL, const gl_constant_value *, param_count);
- /* Setting nr_params here NOT to the size of the param and pull_param
- * arrays, but to the number of uniform components vec4_visitor
- * needs. vec4_visitor::setup_uniforms() will set it back to a proper value.
- */
- c.prog_data.base.base.nr_params = ALIGN(param_count, 4) / 4 + gs->num_samplers;
+ c.prog_data.base.base.nr_params = param_count;
if (brw->gen >= 7) {
if (gp->program.OutputType == GL_POINTS) {
outputs_written |= BITFIELD64_BIT(VARYING_SLOT_CLIP_DIST1);
}
- brw_compute_vue_map(brw, &c.prog_data.base.vue_map, outputs_written);
+ brw_compute_vue_map(brw->intelScreen->devinfo,
+ &c.prog_data.base.vue_map, outputs_written);
/* Compute the output vertex size.
*
c.prog_data.output_topology =
get_hw_prim_for_gl_prim(gp->program.OutputType);
- brw_compute_vue_map(brw, &c.input_vue_map, c.key.input_varyings);
+ brw_compute_vue_map(brw->intelScreen->devinfo,
+ &c.input_vue_map, c.key.input_varyings);
/* GS inputs are read from the VUE 256 bits (2 vec4's) at a time, so we
* need to program a URB read length of ceiling(num_slots / 2).
}
/* Scratch space is used for register spilling */
- if (c.base.last_scratch) {
- perf_debug("Geometry shader triggered register spilling. "
- "Try reducing the number of live vec4 values to "
- "improve performance.\n");
-
- c.prog_data.base.base.total_scratch
- = brw_get_scratch_size(c.base.last_scratch*REG_SIZE);
-
+ if (c.prog_data.base.base.total_scratch) {
brw_get_scratch_bo(brw, &stage_state->scratch_bo,
c.prog_data.base.base.total_scratch *
brw->max_gs_threads);
return true;
}
+static bool
+brw_gs_state_dirty(struct brw_context *brw)
+{
+ return brw_state_dirty(brw,
+ _NEW_TEXTURE,
+ BRW_NEW_GEOMETRY_PROGRAM |
+ BRW_NEW_TRANSFORM_FEEDBACK |
+ BRW_NEW_VUE_MAP_VS);
+}
static void
+brw_gs_populate_key(struct brw_context *brw,
+ struct brw_gs_prog_key *key)
+{
+ struct gl_context *ctx = &brw->ctx;
+ struct brw_stage_state *stage_state = &brw->gs.base;
+ struct brw_geometry_program *gp =
+ (struct brw_geometry_program *) brw->geometry_program;
+ struct gl_program *prog = &gp->program.Base;
+
+ memset(key, 0, sizeof(*key));
+
+ key->base.program_string_id = gp->id;
+ brw_setup_vue_key_clip_info(brw, &key->base,
+ gp->program.Base.UsesClipDistanceOut);
+
+ /* _NEW_TEXTURE */
+ brw_populate_sampler_prog_key_data(ctx, prog, stage_state->sampler_count,
+ &key->base.tex);
+
+ /* BRW_NEW_VUE_MAP_VS */
+ key->input_varyings = brw->vue_map_vs.slots_valid;
+}
+
+void
brw_upload_gs_prog(struct brw_context *brw)
{
struct gl_context *ctx = &brw->ctx;
+ struct gl_shader_program **current = ctx->_Shader->CurrentProgram;
struct brw_stage_state *stage_state = &brw->gs.base;
struct brw_gs_prog_key key;
/* BRW_NEW_GEOMETRY_PROGRAM */
struct brw_geometry_program *gp =
(struct brw_geometry_program *) brw->geometry_program;
+ if (!brw_gs_state_dirty(brw))
+ return;
+
if (gp == NULL) {
/* No geometry shader. Vertex data just passes straight through. */
- if (brw->state.dirty.brw & BRW_NEW_VUE_MAP_VS) {
+ if (brw->ctx.NewDriverState & BRW_NEW_VUE_MAP_VS) {
brw->vue_map_geom_out = brw->vue_map_vs;
- brw->state.dirty.brw |= BRW_NEW_VUE_MAP_GEOM_OUT;
+ brw->ctx.NewDriverState |= BRW_NEW_VUE_MAP_GEOM_OUT;
}
if (brw->gen == 6 &&
- (brw->state.dirty.brw & BRW_NEW_TRANSFORM_FEEDBACK)) {
+ (brw->ctx.NewDriverState & BRW_NEW_TRANSFORM_FEEDBACK)) {
gen6_brw_upload_ff_gs_prog(brw);
return;
}
return;
}
- struct gl_program *prog = &gp->program.Base;
-
- memset(&key, 0, sizeof(key));
-
- key.base.program_string_id = gp->id;
- brw_setup_vec4_key_clip_info(brw, &key.base,
- gp->program.Base.UsesClipDistanceOut);
-
- /* _NEW_TEXTURE */
- brw_populate_sampler_prog_key_data(ctx, prog, stage_state->sampler_count,
- &key.base.tex);
-
- /* BRW_NEW_VUE_MAP_VS */
- key.input_varyings = brw->vue_map_vs.slots_valid;
+ brw_gs_populate_key(brw, &key);
if (!brw_search_cache(&brw->cache, BRW_CACHE_GS_PROG,
&key, sizeof(key),
&stage_state->prog_offset, &brw->gs.prog_data)) {
- bool success =
- do_gs_prog(brw, ctx->_Shader->CurrentProgram[MESA_SHADER_GEOMETRY], gp,
- &key);
+ bool success = brw_codegen_gs_prog(brw, current[MESA_SHADER_GEOMETRY],
+ gp, &key);
assert(success);
(void)success;
}
brw->gs.base.prog_data = &brw->gs.prog_data->base.base;
- if (memcmp(&brw->vs.prog_data->base.vue_map, &brw->vue_map_geom_out,
+ if (memcmp(&brw->gs.prog_data->base.vue_map, &brw->vue_map_geom_out,
sizeof(brw->vue_map_geom_out)) != 0) {
brw->vue_map_geom_out = brw->gs.prog_data->base.vue_map;
- brw->state.dirty.brw |= BRW_NEW_VUE_MAP_GEOM_OUT;
+ brw->ctx.NewDriverState |= BRW_NEW_VUE_MAP_GEOM_OUT;
}
}
-
-const struct brw_tracked_state brw_gs_prog = {
- .dirty = {
- .mesa = _NEW_TEXTURE,
- .brw = BRW_NEW_GEOMETRY_PROGRAM |
- BRW_NEW_TRANSFORM_FEEDBACK |
- BRW_NEW_VUE_MAP_VS,
- },
- .emit = brw_upload_gs_prog
-};
-
-
bool
brw_gs_precompile(struct gl_context *ctx,
struct gl_shader_program *shader_prog,
memset(&key, 0, sizeof(key));
- brw_vec4_setup_prog_key_for_precompile(ctx, &key.base, bgp->id, &gp->Base);
+ brw_vue_setup_prog_key_for_precompile(ctx, &key.base, bgp->id, &gp->Base);
/* Assume that the set of varyings coming in from the vertex shader exactly
* matches what the geometry shader requires.
*/
key.input_varyings = gp->Base.InputsRead;
- success = do_gs_prog(brw, shader_prog, bgp, &key);
+ success = brw_codegen_gs_prog(brw, shader_prog, bgp, &key);
brw->gs.base.prog_offset = old_prog_offset;
brw->gs.prog_data = old_prog_data;