commit 5ae3656890fc65a2a44f5795749f265612ace5b6
author Timur Kristóf <timur.kristof@gmail.com> Mon Oct 05 15:30:18 2020 +0200
committer Timur Kristóf <timur.kristof@gmail.com> Fri Oct 09 15:26:15 2020 +0200
tree d4504976de9500eb8cf53a7667d3daa6757c552b
parent 61280bb4b66e4ff7db393f7c4db368b65bc391fa

aco/ngg: Calculate workgroup size of NGG shaders.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6964>

src/amd/compiler/aco_instruction_selection_setup.cpp

diff --git a/src/amd/compiler/aco_instruction_selection_setup.cpp b/src/amd/compiler/aco_instruction_selection_setup.cpp
index 352f40b..b69b362 100644
--- a/src/amd/compiler/aco_instruction_selection_setup.cpp
+++ b/src/amd/compiler/aco_instruction_selection_setup.cpp
@@ -1261,8 +1261,18 @@
       setup_tcs_info(&ctx, shaders[1], shaders[0]);
       program->workgroup_size = ctx.tcs_num_patches * MAX2(shaders[1]->info.tess.tcs_vertices_out, ctx.args->options->key.tcs.input_vertices);
    } else if (program->stage & hw_ngg_gs) {
-      /* TODO: Calculate workgroup size of NGG shaders. */
-      program->workgroup_size = UINT_MAX;
+      gfx10_ngg_info &ngg_info = args->shader_info->ngg_info;
+
+      /* Max ES (SW VS) threads */
+      uint32_t max_esverts = ngg_info.hw_max_esverts;
+      /* Max GS input primitives = max GS threads */
+      uint32_t max_gs_input_prims = ngg_info.max_gsprims;
+      /* Maximum output vertices -- each thread can export only 1 vertex */
+      uint32_t max_out_vtx = ngg_info.max_out_verts;
+      /* Maximum output primitives -- each thread can export only 1 or 0 primitive */
+      uint32_t max_out_prm = ngg_info.max_gsprims * ngg_info.prim_amp_factor;
+
+      program->workgroup_size = MAX4(max_esverts, max_gs_input_prims, max_out_vtx, max_out_prm);
    } else {
       unreachable("Unsupported shader stage.");
    }