anv: move preprocessing of NIR right before compilation

For graphics pipelines, we'll need to load NIR for retained shaders.
We want to avoid as much processing as possible while doing that when
we're able to load ISA from cache.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15637>

diff --git a/src/intel/vulkan/anv_pipeline.c b/src/intel/vulkan/anv_pipeline.c
index f2e4352..98a8943 100644 (file)
--- a/src/intel/vulkan/anv_pipeline.c
+++ b/src/intel/vulkan/anv_pipeline.c
@@ -242,33 +242,6 @@ anv_shader_stage_to_nir(struct anv_device *device,
    NIR_PASS_V(nir, nir_lower_io_to_temporaries,
               nir_shader_get_entrypoint(nir), true, false);
 
-   const struct nir_lower_sysvals_to_varyings_options sysvals_to_varyings = {
-      .point_coord = true,
-   };
-   NIR_PASS(_, nir, nir_lower_sysvals_to_varyings, &sysvals_to_varyings);
-
-   const nir_opt_access_options opt_access_options = {
-      .is_vulkan = true,
-   };
-   NIR_PASS(_, nir, nir_opt_access, &opt_access_options);
-
-   /* Vulkan uses the separate-shader linking model */
-   nir->info.separate_shader = true;
-
-   struct brw_nir_compiler_opts opts = {
-      .softfp64 = device->fp64_nir,
-      .robust_image_access =
-         device->vk.enabled_features.robustImageAccess ||
-         device->vk.enabled_features.robustImageAccess2,
-   };
-   brw_preprocess_nir(compiler, nir, &opts);
-
-   if (nir->info.stage == MESA_SHADER_MESH && !nir->info.mesh.nv) {
-      NIR_PASS(_, nir, anv_nir_lower_set_vtx_and_prim_count);
-      NIR_PASS(_, nir, nir_opt_dce);
-      NIR_PASS(_, nir, nir_remove_dead_variables, nir_var_shader_out, NULL);
-   }
-
    return nir;
 }
 
@@ -1673,6 +1646,40 @@ anv_fixup_subgroup_size(struct anv_device *device, struct shader_info *info)
       info->subgroup_size = BRW_SUBGROUP_SIZE;
 }
 
+static void
+anv_pipeline_nir_preprocess(struct anv_pipeline *pipeline, nir_shader *nir)
+{
+   struct anv_device *device = pipeline->device;
+   const struct brw_compiler *compiler = device->physical->compiler;
+
+   const struct nir_lower_sysvals_to_varyings_options sysvals_to_varyings = {
+      .point_coord = true,
+   };
+   NIR_PASS(_, nir, nir_lower_sysvals_to_varyings, &sysvals_to_varyings);
+
+   const nir_opt_access_options opt_access_options = {
+      .is_vulkan = true,
+   };
+   NIR_PASS(_, nir, nir_opt_access, &opt_access_options);
+
+   /* Vulkan uses the separate-shader linking model */
+   nir->info.separate_shader = true;
+
+   struct brw_nir_compiler_opts opts = {
+      .softfp64 = device->fp64_nir,
+      .robust_image_access =
+         device->vk.enabled_features.robustImageAccess ||
+         device->vk.enabled_features.robustImageAccess2,
+   };
+   brw_preprocess_nir(compiler, nir, &opts);
+
+   if (nir->info.stage == MESA_SHADER_MESH && !nir->info.mesh.nv) {
+      NIR_PASS(_, nir, anv_nir_lower_set_vtx_and_prim_count);
+      NIR_PASS(_, nir, nir_opt_dce);
+      NIR_PASS(_, nir, nir_remove_dead_variables, nir_var_shader_out, NULL);
+   }
+}
+
 static VkResult
 anv_graphics_pipeline_compile(struct anv_graphics_base_pipeline *pipeline,
                               struct vk_pipeline_cache *cache,
@@ -1730,6 +1737,13 @@ anv_graphics_pipeline_compile(struct anv_graphics_base_pipeline *pipeline,
    if (result != VK_SUCCESS)
       goto fail;
 
+   for (int s = 0; s < ARRAY_SIZE(pipeline->shaders); s++) {
+      if (stages[s].nir == NULL)
+         continue;
+
+      anv_pipeline_nir_preprocess(&pipeline->base, stages[s].nir);
+   }
+
    if (stages[MESA_SHADER_MESH].info && stages[MESA_SHADER_FRAGMENT].info) {
       anv_apply_per_prim_attr_wa(stages[MESA_SHADER_MESH].nir,
                                  stages[MESA_SHADER_FRAGMENT].nir,
@@ -2052,6 +2066,8 @@ anv_pipeline_compile_cs(struct anv_compute_pipeline *pipeline,
          return vk_error(pipeline, VK_ERROR_UNKNOWN);
       }
 
+      anv_pipeline_nir_preprocess(&pipeline->base, stage.nir);
+
       anv_pipeline_lower_nir(&pipeline->base, mem_ctx, &stage, layout,
                              0 /* view_mask */,
                              false /* use_primitive_replication */);
@@ -2746,6 +2762,8 @@ anv_pipeline_compile_ray_tracing(struct anv_ray_tracing_pipeline *pipeline,
          return vk_error(pipeline, VK_ERROR_OUT_OF_HOST_MEMORY);
       }
 
+      anv_pipeline_nir_preprocess(&pipeline->base, stages[i].nir);
+
       anv_pipeline_lower_nir(&pipeline->base, pipeline_ctx, &stages[i],
                              layout, 0 /* view_mask */,
                              false /* use_primitive_replication */);