aco: Implement memory_barrier_tcs_patch.

TCS outputs are going to be written to LDS, so it
has to use memory_barrier_shared in order to ensure
that it waits for LDS writes.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3964>

diff --git a/src/amd/compiler/aco_instruction_selection.cpp b/src/amd/compiler/aco_instruction_selection.cpp
index 069f3d24f62576d10196349ca80c5ec30cca1d40..0e8624cf51c35c2b5bd66e415e3f3e6cbd3c685c 100644 (file)
--- a/src/amd/compiler/aco_instruction_selection.cpp
+++ b/src/amd/compiler/aco_instruction_selection.cpp
@@ -5414,6 +5414,7 @@ void emit_memory_barrier(isel_context *ctx, nir_intrinsic_instr *instr) {
       case nir_intrinsic_memory_barrier_image:
          bld.barrier(aco_opcode::p_memory_barrier_image);
          break;
+      case nir_intrinsic_memory_barrier_tcs_patch:
       case nir_intrinsic_memory_barrier_shared:
          bld.barrier(aco_opcode::p_memory_barrier_shared);
          break;
@@ -6342,6 +6343,7 @@ void visit_intrinsic(isel_context *ctx, nir_intrinsic_instr *instr)
 
       break;
    }
+   case nir_intrinsic_memory_barrier_tcs_patch:
    case nir_intrinsic_group_memory_barrier:
    case nir_intrinsic_memory_barrier:
    case nir_intrinsic_memory_barrier_buffer:
@@ -6349,8 +6351,6 @@ void visit_intrinsic(isel_context *ctx, nir_intrinsic_instr *instr)
    case nir_intrinsic_memory_barrier_shared:
       emit_memory_barrier(ctx, instr);
       break;
-   case nir_intrinsic_memory_barrier_tcs_patch:
-      break;
    case nir_intrinsic_load_num_work_groups: {
       Temp dst = get_ssa_temp(ctx, &instr->dest.ssa);
       bld.copy(Definition(dst), Operand(get_arg(ctx, ctx->args->ac.num_work_groups)));