[Mesa-dev] [PATCH 02/12] nir: use the same driver location for packed varyings

[Timothy Arceri]

---
 src/compiler/nir/nir.h                    |  4 ++--
 src/compiler/nir/nir_lower_io.c           | 25 ++++++++++++++++++++++++-
 src/mesa/drivers/dri/i965/brw_nir.c       | 12 +++++++-----
 src/mesa/state_tracker/st_glsl_to_nir.cpp |  3 +++
 4 files changed, 36 insertions(+), 8 deletions(-)

diff --git a/src/compiler/nir/nir.h b/src/compiler/nir/nir.h
index 7ddad73..4dd38a7 100644
--- a/src/compiler/nir/nir.h
+++ b/src/compiler/nir/nir.h
@@ -2309,8 +2309,8 @@ void nir_lower_io_to_temporaries(nir_shader *shader, nir_function *entrypoint,
 
 void nir_shader_gather_info(nir_shader *shader, nir_function_impl *entrypoint);
 
-void nir_assign_var_locations(struct exec_list *var_list,
-                              unsigned *size,
+void nir_assign_var_locations(struct exec_list *var_list, unsigned *size,
+                              unsigned base_offset,
                               int (*type_size)(const struct glsl_type *));
 
 void nir_lower_io(nir_shader *shader,
diff --git a/src/compiler/nir/nir_lower_io.c b/src/compiler/nir/nir_lower_io.c
index 0d6d8e4..2079004 100644
--- a/src/compiler/nir/nir_lower_io.c
+++ b/src/compiler/nir/nir_lower_io.c
@@ -43,10 +43,18 @@ struct lower_io_state {
 
 void
 nir_assign_var_locations(struct exec_list *var_list, unsigned *size,
+                         unsigned base_offset,
                          int (*type_size)(const struct glsl_type *))
 {
    unsigned location = 0;
 
+   /* There are 32 regular and 32 patch varyings allowed */
+   int locations[64][2];
+   for (unsigned i = 0; i < 64; i++) {
+      for (unsigned j = 0; j < 2; j++)
+         locations[i][j] = -1;
+   }
+
    nir_foreach_variable(var, var_list) {
       /*
        * UBO's have their own address spaces, so don't count them towards the
@@ -56,7 +64,22 @@ nir_assign_var_locations(struct exec_list *var_list, unsigned *size,
           var->interface_type != NULL)
          continue;
 
-      var->data.driver_location = location;
+      /* Make sure we give the same location to varyings packed with
+       * ARB_enhanced_layouts.
+       */
+      int idx = var->data.location - base_offset;
+      if (base_offset && idx >= 0) {
+         assert(idx < ARRAY_SIZE(locations));
+
+         if (locations[idx][var->data.index] == -1) {
+            var->data.driver_location = location;
+            locations[idx][var->data.index] = location;
+         } else {
+            var->data.driver_location = locations[idx][var->data.index];
+         }
+      } else {
+         var->data.driver_location = location;
+      }
       location += type_size(var->type);
    }
 
diff --git a/src/mesa/drivers/dri/i965/brw_nir.c b/src/mesa/drivers/dri/i965/brw_nir.c
index 9274f2e..e60d398 100644
--- a/src/mesa/drivers/dri/i965/brw_nir.c
+++ b/src/mesa/drivers/dri/i965/brw_nir.c
@@ -285,7 +285,8 @@ brw_nir_lower_tes_inputs(nir_shader *nir, const struct brw_vue_map *vue_map)
 void
 brw_nir_lower_fs_inputs(nir_shader *nir)
 {
-   nir_assign_var_locations(&nir->inputs, &nir->num_inputs, type_size_scalar);
+   nir_assign_var_locations(&nir->inputs, &nir->num_inputs, VARYING_SLOT_VAR0,
+                            type_size_scalar);
    nir_lower_io(nir, nir_var_shader_in, type_size_scalar);
 }
 
@@ -295,6 +296,7 @@ brw_nir_lower_vue_outputs(nir_shader *nir,
 {
    if (is_scalar) {
       nir_assign_var_locations(&nir->outputs, &nir->num_outputs,
+                               VARYING_SLOT_VAR0,
                                type_size_vec4_times_4);
       nir_lower_io(nir, nir_var_shader_out, type_size_vec4_times_4);
    } else {
@@ -333,7 +335,7 @@ void
 brw_nir_lower_fs_outputs(nir_shader *nir)
 {
    nir_assign_var_locations(&nir->outputs, &nir->num_outputs,
-                            type_size_scalar);
+                            FRAG_RESULT_DATA0, type_size_scalar);
    nir_lower_io(nir, nir_var_shader_out, type_size_scalar);
 }
 
@@ -353,11 +355,11 @@ static void
 brw_nir_lower_uniforms(nir_shader *nir, bool is_scalar)
 {
    if (is_scalar) {
-      nir_assign_var_locations(&nir->uniforms, &nir->num_uniforms,
+      nir_assign_var_locations(&nir->uniforms, &nir->num_uniforms, 0,
                                type_size_scalar_bytes);
       nir_lower_io(nir, nir_var_uniform, type_size_scalar_bytes);
    } else {
-      nir_assign_var_locations(&nir->uniforms, &nir->num_uniforms,
+      nir_assign_var_locations(&nir->uniforms, &nir->num_uniforms, 0,
                                type_size_vec4_bytes);
       nir_lower_io(nir, nir_var_uniform, type_size_vec4_bytes);
    }
@@ -366,7 +368,7 @@ brw_nir_lower_uniforms(nir_shader *nir, bool is_scalar)
 void
 brw_nir_lower_cs_shared(nir_shader *nir)
 {
-   nir_assign_var_locations(&nir->shared, &nir->num_shared,
+   nir_assign_var_locations(&nir->shared, &nir->num_shared, 0,
                             type_size_scalar_bytes);
    nir_lower_io(nir, nir_var_shared, type_size_scalar_bytes);
 }
diff --git a/src/mesa/state_tracker/st_glsl_to_nir.cpp b/src/mesa/state_tracker/st_glsl_to_nir.cpp
index 6cfbb8e..4fd8666 100644
--- a/src/mesa/state_tracker/st_glsl_to_nir.cpp
+++ b/src/mesa/state_tracker/st_glsl_to_nir.cpp
@@ -308,16 +308,19 @@ st_finalize_nir(struct st_context *st, struct gl_program *prog, nir_shader *nir)
       sort_varyings(&nir->outputs);
       nir_assign_var_locations(&nir->outputs,
                                &nir->num_outputs,
+                               VARYING_SLOT_VAR0,
                                st_glsl_type_size);
       st_nir_fixup_varying_slots(st, &nir->outputs);
    } else if (nir->stage == MESA_SHADER_FRAGMENT) {
       sort_varyings(&nir->inputs);
       nir_assign_var_locations(&nir->inputs,
                                &nir->num_inputs,
+                               VARYING_SLOT_VAR0,
                                st_glsl_type_size);
       st_nir_fixup_varying_slots(st, &nir->inputs);
       nir_assign_var_locations(&nir->outputs,
                                &nir->num_outputs,
+                               FRAG_RESULT_DATA0,
                                st_glsl_type_size);
    } else {
       unreachable("invalid shader type for tgsi bypass\n");
-- 
2.5.5