[PATCH] D141643: AMDGPU: Put un-initiaized enumerators together in an enum definition.

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/test/CodeGen/AMDGPU/permlane-op-sel.ll:2
+; RUN: llc -march=amdgcn -mcpu=gfx1030 -filetype=obj < %s | llvm-objdump --triple=amdgcn--amdhsa --mcpu=gfx1030 -d - | FileCheck %s
+
+declare i32 @llvm.amdgcn.permlane16(i32, i32, i32, i32, i1, i1)
----------------
Should also add a direct to asm line with -show-mc-encoding to be sure


================
Comment at: llvm/test/CodeGen/AMDGPU/permlane-op-sel.ll:6
+; CHECK-LABEL: <permlane_op_sel_obj>:
+; CHECK: v_permlane16_b32 v{{[0-9]+}}, v{{[0-9]+}}, s{{[0-9]+}}, s{{[0-9]+}} op_sel:[1,0]
+define amdgpu_kernel void @permlane_op_sel_obj(ptr addrspace(1) %out, i32 %src0, i32 %src1, i32 %src2) {
----------------
Should check the encoding comments


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D141643/new/

https://reviews.llvm.org/D141643