i965/vec4: Always use NIR

GLSL IR vs. NIR shader-db results for vec4 programs on i965:

   total instructions in shared programs: 1499328 -> 1388354 (-7.40%)
   instructions in affected programs:     1245199 -> 1134225 (-8.91%)
   helped:                                7469
   HURT:                                  2440

GLSL IR vs. NIR shader-db results for vec4 programs on G4x:

   total instructions in shared programs: 1436799 -> 1325825 (-7.72%)
   instructions in affected programs:     1205599 -> 1094625 (-9.20%)
   helped:                                7469
   HURT:                                  2440

GLSL IR vs. NIR shader-db results for vec4 programs on Iron Lake:

   total instructions in shared programs: 1436654 -> 1325682 (-7.72%)
   instructions in affected programs:     1205503 -> 1094531 (-9.21%)
   helped:                                7468
   HURT:                                  2440

GLSL IR vs. NIR shader-db results for vec4 programs on Sandy Bridge:

   total instructions in shared programs: 2016249 -> 1787033 (-11.37%)
   instructions in affected programs:     1850547 -> 1621331 (-12.39%)
   helped:                                14856
   HURT:                                  1481

GLSL IR vs. NIR shader-db results for vec4 programs on Ivy Bridge:

   total instructions in shared programs: 1848027 -> 1648216 (-10.81%)
   instructions in affected programs:     1660279 -> 1460468 (-12.03%)
   helped:                                14668
   HURT:                                  1369

GLSL IR vs. NIR shader-db results for vec4 programs on Bay Trail:

   total instructions in shared programs: 1848027 -> 1648216 (-10.81%)
   instructions in affected programs:     1660279 -> 1460468 (-12.03%)
   helped:                                14668
   HURT:                                  1369

GLSL IR vs. NIR shader-db results for vec4 programs on Haswell:

   total instructions in shared programs: 1848027 -> 1648216 (-10.81%)
   instructions in affected programs:     1660279 -> 1460468 (-12.03%)
   helped:                                14668
   HURT:                                  1369

I also ran our full suite of benchmarks on a Haswell and had the following
statistically significant (according to ministat) changes:

   Test                        master-glsl     master-nir     diff
   bench_OglGeomPoint          461.556         463.006        1.450
   bench_OglTerrainFlyInst     184.484         187.574        3.090
   bench_OglTerrainPanInst     132.412         136.307        3.895
   bench_OglTexFilterAniso     19.653          19.645         -0.008
   bench_OglTexFilterTri       58.333          58.009         -0.324
   bench_OglVSInstancing       65.049          65.327         0.278
   bench_trexoff               69.474          69.694         0.220
   bench_valley                40.708          41.125         0.417

v2 (Jason Ekstrand):
 - Remove more uses of NirOptions as a switch
 - New shader-db numbers
 - Added benchmark numbers

Reviewed-by: Matt Turner <mattst88@gmail.com>

3 files changed, 14 insertions, 49 deletions

diff --git a/src/mesa/drivers/dri/i965/brw_program.c b/src/mesa/drivers/dri/i965/brw_program.c
index 0e4b823..a034dac 100644
--- a/src/mesa/drivers/dri/i965/brw_program.c
+++ b/src/mesa/drivers/dri/i965/brw_program.c
@@ -142,9 +142,7 @@ brwProgramStringNotify(struct gl_context *ctx,
 
       brw_add_texrect_params(prog);
 
-      if (ctx->Const.ShaderCompilerOptions[MESA_SHADER_FRAGMENT].NirOptions) {
-         prog->nir = brw_create_nir(brw, NULL, prog, MESA_SHADER_FRAGMENT, true);
-      }
+      prog->nir = brw_create_nir(brw, NULL, prog, MESA_SHADER_FRAGMENT, true);
 
       brw_fs_precompile(ctx, NULL, prog);
       break;
@@ -168,10 +166,8 @@ brwProgramStringNotify(struct gl_context *ctx,
 
       brw_add_texrect_params(prog);
 
-      if (ctx->Const.ShaderCompilerOptions[MESA_SHADER_VERTEX].NirOptions) {
-         prog->nir = brw_create_nir(brw, NULL, prog, MESA_SHADER_VERTEX,
-                                    brw->intelScreen->compiler->scalar_vs);
-      }
+      prog->nir = brw_create_nir(brw, NULL, prog, MESA_SHADER_VERTEX,
+                                 brw->intelScreen->compiler->scalar_vs);
 
       brw_vs_precompile(ctx, NULL, prog);
       break;
diff --git a/src/mesa/drivers/dri/i965/brw_shader.cpp b/src/mesa/drivers/dri/i965/brw_shader.cpp
index af85bd2..2c23f53 100644
--- a/src/mesa/drivers/dri/i965/brw_shader.cpp
+++ b/src/mesa/drivers/dri/i965/brw_shader.cpp
@@ -142,8 +142,7 @@ brw_compiler_create(void *mem_ctx, const struct brw_device_info *devinfo)
       if (devinfo->gen < 7)
          compiler->glsl_compiler_options[i].EmitNoIndirectSampler = true;
 
-      if (is_scalar || brw_env_var_as_boolean("INTEL_USE_NIR", true))
-         compiler->glsl_compiler_options[i].NirOptions = nir_options;
+      compiler->glsl_compiler_options[i].NirOptions = nir_options;
    }
 
    return compiler;
@@ -271,8 +270,6 @@ process_glsl_ir(gl_shader_stage stage,
    brw_lower_texture_gradients(brw, shader->ir);
    do_vec_index_to_cond_assign(shader->ir);
    lower_vector_insert(shader->ir, true);
-   if (options->NirOptions == NULL)
-      brw_do_cubemap_normalize(shader->ir);
    lower_offset_arrays(shader->ir);
    brw_do_lower_unnormalized_offset(shader->ir);
    lower_noise(shader->ir);
@@ -341,9 +338,6 @@ brw_link_shader(struct gl_context *ctx, struct gl_shader_program *shProg)
 
    for (stage = 0; stage < ARRAY_SIZE(shProg->_LinkedShaders); stage++) {
       struct gl_shader *shader = shProg->_LinkedShaders[stage];
-      const struct gl_shader_compiler_options *options =
-         &ctx->Const.ShaderCompilerOptions[stage];
-
       if (!shader)
 	 continue;
 
@@ -391,10 +385,8 @@ brw_link_shader(struct gl_context *ctx, struct gl_shader_program *shProg)
 
       brw_add_texrect_params(prog);
 
-      if (options->NirOptions) {
-         prog->nir = brw_create_nir(brw, shProg, prog, (gl_shader_stage) stage,
-                                    is_scalar_shader_stage(compiler, stage));
-      }
+      prog->nir = brw_create_nir(brw, shProg, prog, (gl_shader_stage) stage,
+                                 is_scalar_shader_stage(compiler, stage));
 
       _mesa_reference_program(ctx, &prog, NULL);
    }
diff --git a/src/mesa/drivers/dri/i965/brw_vec4.cpp b/src/mesa/drivers/dri/i965/brw_vec4.cpp
index c61b385..056ce39 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4.cpp
@@ -1786,9 +1786,6 @@ vec4_visitor::emit_shader_time_write(int shader_time_subindex, src_reg value)
 bool
 vec4_visitor::run()
 {
-   bool use_vec4_nir =
-      compiler->glsl_compiler_options[stage].NirOptions != NULL;
-
    sanity_param_count = prog->Parameters->NumParameters;
 
    if (shader_time_index >= 0)
@@ -1798,19 +1795,10 @@ vec4_visitor::run()
 
    emit_prolog();
 
-   if (use_vec4_nir) {
-      assert(prog->nir != NULL);
-      emit_nir_code();
-      if (failed)
-         return false;
-   } else if (shader) {
-      /* Generate VS IR for main().  (the visitor only descends into
-       * functions called "main").
-       */
-      visit_instructions(shader->base.ir);
-   } else {
-      emit_program_code();
-   }
+   assert(prog->nir != NULL);
+   emit_nir_code();
+   if (failed)
+      return false;
    base_ir = NULL;
 
    emit_thread_end();
@@ -1823,18 +1811,9 @@ vec4_visitor::run()
     * that we have reladdr computations available for CSE, since we'll
     * often do repeated subexpressions for those.
     */
-   if (shader || use_vec4_nir) {
-      move_grf_array_access_to_scratch();
-      move_uniform_array_access_to_pull_constants();
-   } else {
-      /* The ARB_vertex_program frontend emits pull constant loads directly
-       * rather than using reladdr, so we don't need to walk through all the
-       * instructions looking for things to move.  There isn't anything.
-       *
-       * We do still need to split things to vec4 size.
-       */
-      split_uniform_registers();
-   }
+   move_grf_array_access_to_scratch();
+   move_uniform_array_access_to_pull_constants();
+
    pack_uniform_registers();
    move_push_constants_to_pull_constants();
    split_virtual_grfs();
@@ -1974,9 +1953,7 @@ brw_vs_emit(struct brw_context *brw,
    if (unlikely(INTEL_DEBUG & DEBUG_VS))
       brw_dump_ir("vertex", prog, &shader->base, &vp->Base);
 
-   if (!vp->Base.nir &&
-       (brw->intelScreen->compiler->scalar_vs ||
-        brw->intelScreen->compiler->glsl_compiler_options[MESA_SHADER_VERTEX].NirOptions != NULL)) {
+   if (!vp->Base.nir) {
       /* Normally we generate NIR in LinkShader() or
        * ProgramStringNotify(), but Mesa's fixed-function vertex program
        * handling doesn't notify the driver at all.  Just do it here, at