diff --git a/src/amd/compiler/aco_instruction_selection.cpp b/src/amd/compiler/aco_instruction_selection.cpp
index f34b5691a781a3758ffa56993ab1b889674a2c4c..27da22fd2e8350456d327ec03fd0da0fd817c06b 100644
--- a/src/amd/compiler/aco_instruction_selection.cpp
+++ b/src/amd/compiler/aco_instruction_selection.cpp
@@ -10865,7 +10865,7 @@ void ngg_emit_sendmsg_gs_alloc_req(isel_context *ctx, Temp vtx_cnt = Temp(), Tem
/* VS/TES: we infer the vertex and primitive count from arguments
* GS: the caller needs to supply them
*/
- assert(ctx->shader->info.stage == MESA_SHADER_GEOMETRY
+ assert((ctx->stage & sw_gs)
? (vtx_cnt.id() && prm_cnt.id())
: (!vtx_cnt.id() && !prm_cnt.id()));
@@ -11330,9 +11330,18 @@ void ngg_gs_export_vertices(isel_context *ctx, Temp wg_vtx_cnt, Temp tid_in_tg,
begin_divergent_if_then(ctx, &ic, is_vtx_export_thread);
bld.reset(ctx->block);
- /* Vertex compaction: read stream 1 of the primitive flags to see which vertex the current thread needs to export */
- Operand m = load_lds_size_m0(bld);
- Temp exported_vtx_idx = bld.ds(aco_opcode::ds_read_u8, bld.def(v1), vertex_lds_addr, m, ctx->ngg_gs_primflags_offset + 1);
+ /* The index of the vertex that the current thread will export. */
+ Temp exported_vtx_idx;
+
+ if (ctx->ngg_gs_early_alloc) {
+ /* No vertex compaction necessary, the thread can export its own vertex. */
+ exported_vtx_idx = tid_in_tg;
+ } else {
+ /* Vertex compaction: read stream 1 of the primitive flags to see which vertex the current thread needs to export */
+ Operand m = load_lds_size_m0(bld);
+ exported_vtx_idx = bld.ds(aco_opcode::ds_read_u8, bld.def(v1), vertex_lds_addr, m, ctx->ngg_gs_primflags_offset + 1);
+ }
+
/* Get the LDS address of the vertex that the current thread must export. */
Temp exported_vtx_addr = ngg_gs_vertex_lds_addr(ctx, exported_vtx_idx);
@@ -11367,6 +11376,19 @@ void ngg_gs_export_vertices(isel_context *ctx, Temp wg_vtx_cnt, Temp tid_in_tg,
end_divergent_if(ctx, &ic);
}
+void ngg_gs_prelude(isel_context *ctx)
+{
+ if (!ctx->ngg_gs_early_alloc)
+ return;
+
+ /* We know the GS writes the maximum possible number of vertices, so
+ * it's likely that most threads need to export a primitive, too.
+ * Thus, we won't have to worry about primitive compaction here.
+ */
+ Temp num_max_vertices = ngg_max_vertex_count(ctx);
+ ngg_emit_sendmsg_gs_alloc_req(ctx, num_max_vertices, num_max_vertices);
+}
+
void ngg_gs_finale(isel_context *ctx)
{
if_context ic;
@@ -11391,19 +11413,33 @@ void ngg_gs_finale(isel_context *ctx)
*/
Temp vertex_live = bld.vopc(aco_opcode::v_cmp_lg_u32, bld.def(bld.lm), Operand(0u), Operand(prim_flag_0));
- /* Perform a workgroup reduction and exclusive scan. */
- std::pair<Temp, Temp> wg_scan = ngg_gs_workgroup_reduce_and_scan(ctx, vertex_live);
- bld.reset(ctx->block);
/* Total number of vertices emitted by the workgroup. */
- Temp wg_vtx_cnt = wg_scan.first;
+ Temp wg_vtx_cnt;
/* ID of the thread which will export the current thread's vertex. */
- Temp exporter_tid_in_tg = wg_scan.second;
- /* Skip all exports when possible. */
- Temp have_exports = bld.sopc(aco_opcode::s_cmp_lg_u32, bld.def(s1, scc), wg_vtx_cnt, Operand(0u));
- max_vtxcnt = bld.sop2(aco_opcode::s_cselect_b32, bld.def(s1), max_vtxcnt, Operand(0u), bld.scc(have_exports));
+ Temp exporter_tid_in_tg;
+
+ if (ctx->ngg_gs_early_alloc) {
+ /* There is no need for a scan or vertex compaction, we know that
+ * the GS writes all possible vertices so each thread can export its own vertex.
+ */
+ wg_vtx_cnt = max_vtxcnt;
+ exporter_tid_in_tg = tid_in_tg;
+ } else {
+ /* Perform a workgroup reduction and exclusive scan. */
+ std::pair<Temp, Temp> wg_scan = ngg_gs_workgroup_reduce_and_scan(ctx, vertex_live);
+ bld.reset(ctx->block);
+ /* Total number of vertices emitted by the workgroup. */
+ wg_vtx_cnt = wg_scan.first;
+ /* ID of the thread which will export the current thread's vertex. */
+ exporter_tid_in_tg = wg_scan.second;
+ /* Skip all exports when possible. */
+ Temp have_exports = bld.sopc(aco_opcode::s_cmp_lg_u32, bld.def(s1, scc), wg_vtx_cnt, Operand(0u));
+ max_vtxcnt = bld.sop2(aco_opcode::s_cselect_b32, bld.def(s1), max_vtxcnt, Operand(0u), bld.scc(have_exports));
+
+ ngg_emit_sendmsg_gs_alloc_req(ctx, wg_vtx_cnt, max_vtxcnt);
+ ngg_gs_setup_vertex_compaction(ctx, vertex_live, tid_in_tg, exporter_tid_in_tg);
+ }
- ngg_emit_sendmsg_gs_alloc_req(ctx, wg_vtx_cnt, max_vtxcnt);
- ngg_gs_setup_vertex_compaction(ctx, vertex_live, tid_in_tg, exporter_tid_in_tg);
ngg_gs_export_primitives(ctx, max_vtxcnt, tid_in_tg, exporter_tid_in_tg, prim_flag_0);
ngg_gs_export_vertices(ctx, wg_vtx_cnt, tid_in_tg, vertex_lds_addr);
}
@@ -11440,6 +11476,8 @@ void select_program(Program *program,
if (ngg_no_gs)
ngg_nogs_prelude(&ctx);
+ else if (!i && ngg_gs)
+ ngg_gs_prelude(&ctx);
/* In a merged VS+TCS HS, the VS implementation can be completely empty. */
nir_function_impl *func = nir_shader_get_entrypoint(nir);
diff --git a/src/amd/compiler/aco_instruction_selection.h b/src/amd/compiler/aco_instruction_selection.h
index 69ef7809405c8144f4f60dd2978cbfbabaa68d92..360f7db38b39cabc222ea7b0dc41962bcb65f15a 100644
--- a/src/amd/compiler/aco_instruction_selection.h
+++ b/src/amd/compiler/aco_instruction_selection.h
@@ -94,11 +94,14 @@ struct isel_context {
/* GS inputs */
bool ngg_nogs_early_prim_export = false;
+ bool ngg_gs_early_alloc = false;
Temp gs_wave_id;
unsigned ngg_gs_emit_addr = 0;
unsigned ngg_gs_emit_vtx_bytes = 0;
unsigned ngg_gs_scratch_addr = 0;
unsigned ngg_gs_primflags_offset = 0;
+ int ngg_gs_const_vtxcnt[4];
+ int ngg_gs_const_prmcnt[4];
/* VS output information */
bool export_clip_dists;
diff --git a/src/amd/compiler/aco_instruction_selection_setup.cpp b/src/amd/compiler/aco_instruction_selection_setup.cpp
index a3030346c4e77e83e49a304ab9d33f44c0c53657..352f40b94696ddd29e96f3f438ef1e6b032100e9 100644
--- a/src/amd/compiler/aco_instruction_selection_setup.cpp
+++ b/src/amd/compiler/aco_instruction_selection_setup.cpp
@@ -516,6 +516,10 @@ void setup_gs_variables(isel_context *ctx, nir_shader *nir)
/* Make sure we have enough room for emitted GS vertices */
assert((ngg_emit_bytes % (ctx->ngg_gs_emit_vtx_bytes * nir->info.gs.vertices_out)) == 0);
+
+ /* See if the number of vertices and primitives are compile-time known */
+ nir_gs_count_vertices_and_primitives(nir, ctx->ngg_gs_const_vtxcnt, ctx->ngg_gs_const_prmcnt, 4u);
+ ctx->ngg_gs_early_alloc = ctx->ngg_gs_const_vtxcnt[0] == nir->info.gs.vertices_out && ctx->ngg_gs_const_prmcnt[0] != -1;
}
if (ctx->stage & sw_vs)