glsl: use linked_shaders bitmask to iterate stages for subroutine fields

This should be faster than looping over every stage and null checking, but
will also make the code a bit cleaner when we switch to getting more fields
from gl_program rather than from gl_linked_shader as we can just copy the
pointer and not need to worry about null checking then copying.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>

src/compiler/glsl/link_uniforms.cpp

@@ -1140,10 +1140,10 @@ link_setup_uniform_remap_tables(struct gl_context *ctx,
       const unsigned entries =
          MAX2(1, prog->data->UniformStorage[i].array_elements);
 
-      for (unsigned j = 0; j < MESA_SHADER_STAGES; j++) {
+      unsigned mask = prog->data->linked_stages;
+      while (mask) {
+         const int j = u_bit_scan(&mask);
          struct gl_linked_shader *sh = prog->_LinkedShaders[j];
-         if (!sh)
-            continue;
 
          if (!prog->data->UniformStorage[i].opaque[j].active)
             continue;
@@ -1172,10 +1172,10 @@ link_setup_uniform_remap_tables(struct gl_context *ctx,
       const unsigned entries =
          MAX2(1, prog->data->UniformStorage[i].array_elements);
 
-      for (unsigned j = 0; j < MESA_SHADER_STAGES; j++) {
+      unsigned mask = prog->data->linked_stages;
+      while (mask) {
+         const int j = u_bit_scan(&mask);
          struct gl_linked_shader *sh = prog->_LinkedShaders[j];
-         if (!sh)
-            continue;
 
          if (!prog->data->UniformStorage[i].opaque[j].active)
             continue;

src/compiler/glsl/linker.cpp

@@ -3144,11 +3144,10 @@ check_resources(struct gl_context *ctx, struct gl_shader_program *prog)
 static void
 link_calculate_subroutine_compat(struct gl_shader_program *prog)
 {
-   for (unsigned i = 0; i < MESA_SHADER_STAGES; i++) {
+   unsigned mask = prog->data->linked_stages;
+   while (mask) {
+      const int i = u_bit_scan(&mask);
       struct gl_linked_shader *sh = prog->_LinkedShaders[i];
-      int count;
-      if (!sh)
-         continue;
 
       for (unsigned j = 0; j < sh->NumSubroutineUniformRemapTable; j++) {
          if (sh->SubroutineUniformRemapTable[j] == INACTIVE_UNIFORM_EXPLICIT_LOCATION)
@@ -3159,7 +3158,7 @@ link_calculate_subroutine_compat(struct gl_shader_program *prog)
          if (!uni)
             continue;
 
-         count = 0;
+         int count = 0;
          if (sh->NumSubroutineFunctions == 0) {
             linker_error(prog, "subroutine uniform %s defined but no valid functions found\n", uni->type->name);
             continue;
@@ -3181,16 +3180,16 @@ link_calculate_subroutine_compat(struct gl_shader_program *prog)
 static void
 check_subroutine_resources(struct gl_shader_program *prog)
 {
-   for (unsigned i = 0; i < MESA_SHADER_STAGES; i++) {
+   unsigned mask = prog->data->linked_stages;
+   while (mask) {
+      const int i = u_bit_scan(&mask);
       struct gl_linked_shader *sh = prog->_LinkedShaders[i];
 
-      if (sh) {
       if (sh->NumSubroutineUniformRemapTable > MAX_SUBROUTINE_UNIFORM_LOCATIONS)
          linker_error(prog, "Too many %s shader subroutine uniforms\n",
                       _mesa_shader_stage_to_string(i));
    }
 }
-}
 /**
  * Validate shader image resources.
  */
@@ -3383,12 +3382,11 @@ check_explicit_uniform_locations(struct gl_context *ctx,
    }
 
    unsigned entries_total = 0;
-   for (unsigned i = 0; i < MESA_SHADER_STAGES; i++) {
+   unsigned mask = prog->data->linked_stages;
+   while (mask) {
+      const int i = u_bit_scan(&mask);
       struct gl_linked_shader *sh = prog->_LinkedShaders[i];
 
-      if (!sh)
-         continue;
-
       foreach_in_list(ir_instruction, node, sh->ir) {
          ir_variable *var = node->as_variable();
          if (!var || var->data.mode != ir_var_uniform)
@@ -4317,14 +4315,12 @@ build_program_resource_list(struct gl_context *ctx,
       }
    }
 
-   for (unsigned i = 0; i < MESA_SHADER_STAGES; i++) {
+   unsigned mask = shProg->data->linked_stages;
+   while (mask) {
+      const int i = u_bit_scan(&mask);
       struct gl_linked_shader *sh = shProg->_LinkedShaders[i];
-      GLuint type;
 
-      if (!sh)
+      GLuint type = _mesa_shader_stage_to_subroutine((gl_shader_stage)i);
-         continue;
-
-      type = _mesa_shader_stage_to_subroutine((gl_shader_stage)i);
       for (unsigned j = 0; j < sh->NumSubroutineFunctions; j++) {
          if (!add_program_resource(shProg, resource_set,
                                    type, &sh->SubroutineFunctions[j], 0))
@@ -4372,12 +4368,11 @@ validate_sampler_array_indexing(struct gl_context *ctx,
 static void
 link_assign_subroutine_types(struct gl_shader_program *prog)
 {
-   for (unsigned i = 0; i < MESA_SHADER_STAGES; i++) {
+   unsigned mask = prog->data->linked_stages;
+   while (mask) {
+      const int i = u_bit_scan(&mask);
       gl_linked_shader *sh = prog->_LinkedShaders[i];
 
-      if (sh == NULL)
-         continue;
-
       sh->MaxSubroutineFunctionIndex = 0;
       foreach_in_list(ir_instruction, node, sh->ir) {
          ir_function *fn = node->as_function();