[Mesa-dev] [PATCH] intel/compiler: Make brw_nir_lower_intrinsics compute-specific

[Jordan Justen]

Reviewed-by: Jordan Justen <jordan.l.justen at intel.com>

On 2017-10-06 10:20:28, Jason Ekstrand wrote:
> It's already only ever called from brw_compile_cs and only handles
> compute intrinsics.  Let's just make it CS-specific.  We can always
> make it handle other stages again later if we want.
> 
> Cc: Jordan Justen <jordan.l.justen at intel.com>
> 
> ---
>  src/intel/Makefile.sources                         |  2 +-
>  src/intel/compiler/brw_fs.cpp                      |  2 +-
>  src/intel/compiler/brw_nir.h                       |  4 ++--
>  ..._intrinsics.c => brw_nir_lower_cs_intrinsics.c} | 22 ++++++++--------------
>  4 files changed, 12 insertions(+), 18 deletions(-)
>  rename src/intel/compiler/{brw_nir_intrinsics.c => brw_nir_lower_cs_intrinsics.c} (89%)
> 
> diff --git a/src/intel/Makefile.sources b/src/intel/Makefile.sources
> index 9672dcc..b835533 100644
> --- a/src/intel/Makefile.sources
> +++ b/src/intel/Makefile.sources
> @@ -75,7 +75,7 @@ COMPILER_FILES = \
>         compiler/brw_nir_analyze_boolean_resolves.c \
>         compiler/brw_nir_analyze_ubo_ranges.c \
>         compiler/brw_nir_attribute_workarounds.c \
> -       compiler/brw_nir_intrinsics.c \
> +       compiler/brw_nir_lower_cs_intrinsics.c \
>         compiler/brw_nir_opt_peephole_ffma.c \
>         compiler/brw_nir_tcs_workarounds.c \
>         compiler/brw_packed_float.c \
> diff --git a/src/intel/compiler/brw_fs.cpp b/src/intel/compiler/brw_fs.cpp
> index 6f5f21d..371df71 100644
> --- a/src/intel/compiler/brw_fs.cpp
> +++ b/src/intel/compiler/brw_fs.cpp
> @@ -6766,7 +6766,7 @@ brw_compile_cs(const struct brw_compiler *compiler, void *log_data,
>        MAX2(shader->num_uniforms,
>             (unsigned)4 * (prog_data->thread_local_id_index + 1));
>  
> -   brw_nir_lower_intrinsics(shader, &prog_data->base);
> +   brw_nir_lower_cs_intrinsics(shader, prog_data);
>     shader = brw_postprocess_nir(shader, compiler, true);
>  
>     prog_data->local_size[0] = shader->info.cs.local_size[0];
> diff --git a/src/intel/compiler/brw_nir.h b/src/intel/compiler/brw_nir.h
> index f4b13b1..6f64a73 100644
> --- a/src/intel/compiler/brw_nir.h
> +++ b/src/intel/compiler/brw_nir.h
> @@ -95,8 +95,8 @@ void brw_nir_analyze_boolean_resolves(nir_shader *nir);
>  nir_shader *brw_preprocess_nir(const struct brw_compiler *compiler,
>                                 nir_shader *nir);
>  
> -bool brw_nir_lower_intrinsics(nir_shader *nir,
> -                              struct brw_stage_prog_data *prog_data);
> +bool brw_nir_lower_cs_intrinsics(nir_shader *nir,
> +                                 struct brw_cs_prog_data *prog_data);
>  void brw_nir_lower_vs_inputs(nir_shader *nir,
>                               bool use_legacy_snorm_formula,
>                               const uint8_t *vs_attrib_wa_flags);
> diff --git a/src/intel/compiler/brw_nir_intrinsics.c b/src/intel/compiler/brw_nir_lower_cs_intrinsics.c
> similarity index 89%
> rename from src/intel/compiler/brw_nir_intrinsics.c
> rename to src/intel/compiler/brw_nir_lower_cs_intrinsics.c
> index abbbc6f..602ef2e 100644
> --- a/src/intel/compiler/brw_nir_intrinsics.c
> +++ b/src/intel/compiler/brw_nir_lower_cs_intrinsics.c
> @@ -26,10 +26,7 @@
>  
>  struct lower_intrinsics_state {
>     nir_shader *nir;
> -   union {
> -      struct brw_stage_prog_data *prog_data;
> -      struct brw_cs_prog_data *cs_prog_data;
> -   };
> +   struct brw_cs_prog_data *prog_data;
>     nir_function_impl *impl;
>     bool progress;
>     nir_builder builder;
> @@ -50,9 +47,9 @@ read_thread_local_id(struct lower_intrinsics_state *state)
>     if (group_size <= 8)
>        return nir_imm_int(b, 0);
>  
> -   assert(state->cs_prog_data->thread_local_id_index >= 0);
> +   assert(state->prog_data->thread_local_id_index >= 0);
>     state->cs_thread_id_used = true;
> -   const int id_index = state->cs_prog_data->thread_local_id_index;
> +   const int id_index = state->prog_data->thread_local_id_index;
>  
>     nir_intrinsic_instr *load =
>        nir_intrinsic_instr_create(nir, nir_intrinsic_load_uniform);
> @@ -84,7 +81,6 @@ lower_cs_intrinsics_convert_block(struct lower_intrinsics_state *state,
>        nir_ssa_def *sysval;
>        switch (intrinsic->intrinsic) {
>        case nir_intrinsic_load_local_invocation_index: {
> -         assert(nir->stage == MESA_SHADER_COMPUTE);
>           /* We construct the local invocation index from:
>            *
>            *    gl_LocalInvocationIndex =
> @@ -97,7 +93,6 @@ lower_cs_intrinsics_convert_block(struct lower_intrinsics_state *state,
>        }
>  
>        case nir_intrinsic_load_local_invocation_id: {
> -         assert(nir->stage == MESA_SHADER_COMPUTE);
>           /* We lower gl_LocalInvocationID from gl_LocalInvocationIndex based
>            * on this formula:
>            *
> @@ -156,11 +151,10 @@ lower_cs_intrinsics_convert_impl(struct lower_intrinsics_state *state)
>  }
>  
>  bool
> -brw_nir_lower_intrinsics(nir_shader *nir, struct brw_stage_prog_data *prog_data)
> +brw_nir_lower_cs_intrinsics(nir_shader *nir,
> +                            struct brw_cs_prog_data *prog_data)
>  {
> -   /* Currently we only lower intrinsics for compute shaders */
> -   if (nir->stage != MESA_SHADER_COMPUTE)
> -      return false;
> +   assert(nir->stage == MESA_SHADER_COMPUTE);
>  
>     bool progress = false;
>     struct lower_intrinsics_state state;
> @@ -179,8 +173,8 @@ brw_nir_lower_intrinsics(nir_shader *nir, struct brw_stage_prog_data *prog_data)
>        progress |= state.progress;
>     } while (state.progress);
>  
> -   if (nir->stage == MESA_SHADER_COMPUTE && !state.cs_thread_id_used)
> -      state.cs_prog_data->thread_local_id_index = -1;
> +   if (!state.cs_thread_id_used)
> +      state.prog_data->thread_local_id_index = -1;
>  
>     return progress;
>  }
> -- 
> 2.5.0.400.gff86faf
>