intel/cs: Rework the way thread local ID is handled

Previously, brw_nir_lower_intrinsics added the param and then emitted a
load_uniform intrinsic to load it directly.  This commit switches things
over to use a specific NIR intrinsic for the thread id.  The one thing I
don't like about this approach is that we have to copy thread_local_id
over to the new visitor in import_uniforms.

Reviewed-by: Iago Toral Quiroga <[email protected]>

1 files changed, 14 insertions, 0 deletions

diff --git a/src/intel/compiler/brw_fs_nir.cpp b/src/intel/compiler/brw_fs_nir.cpp
index 04b6e5119a2..77d8bae4db6 100644
--- a/src/intel/compiler/brw_fs_nir.cpp
+++ b/src/intel/compiler/brw_fs_nir.cpp
@@ -88,6 +88,16 @@ fs_visitor::nir_setup_uniforms()
    }
 
    uniforms = nir->num_uniforms / 4;
+
+   if (stage == MESA_SHADER_COMPUTE) {
+      /* Add a uniform for the thread local id.  It must be the last uniform
+       * on the list.
+       */
+      assert(uniforms == prog_data->nr_params);
+      uint32_t *param = brw_stage_prog_data_add_params(prog_data, 1);
+      *param = BRW_PARAM_BUILTIN_THREAD_LOCAL_ID;
+      thread_local_id = fs_reg(UNIFORM, uniforms++, BRW_REGISTER_TYPE_UD);
+   }
 }
 
 static bool
@@ -3412,6 +3422,10 @@ fs_visitor::nir_emit_cs_intrinsic(const fs_builder &bld,
       cs_prog_data->uses_barrier = true;
       break;
 
+   case nir_intrinsic_load_intel_thread_local_id:
+      bld.MOV(retype(dest, BRW_REGISTER_TYPE_UD), thread_local_id);
+      break;
+
    case nir_intrinsic_load_local_invocation_id:
    case nir_intrinsic_load_work_group_id: {
       gl_system_value sv = nir_system_value_from_intrinsic(instr->intrinsic);