[llvm] [AMDGPU] NFC. Add opt instcombine pass for llvm.amdgcn.ballot.i64.wave32.ll test. (PR #73779)

[Valery Pykhtin via llvm-commits]

================
@@ -3,6 +3,10 @@
 ; RUN: llc -march=amdgcn -global-isel=0 -mcpu=gfx1100 -amdgpu-enable-delay-alu=0 < %s | FileCheck %s --check-prefixes=CHECK,DAGISEL
 ; RUN: llc -march=amdgcn -global-isel -mcpu=gfx1010 < %s | FileCheck %s --check-prefixes=CHECK,GISEL
 ; RUN: llc -march=amdgcn -global-isel -mcpu=gfx1100 -amdgpu-enable-delay-alu=0 < %s | FileCheck %s --check-prefixes=CHECK,GISEL
+; RUN: opt -mtriple=amdgcn-- -mcpu=gfx1010 -mattr=+wavefrontsize32,-wavefrontsize64 -passes=instcombine -o - < %s | llc -march=amdgcn -global-isel=0 -mcpu=gfx1010 - | FileCheck %s --check-prefixes=CHECK-OPT,DAGISEL-OPT
----------------
vpykhtin wrote:

I see that tests in _test/Transforms/InstCombine/AMDGPU_ expectedly run only _opt_ but I would like to test the behavior of the entire compilation in -O0 and higher modes. In particular I would like to check if the _branch on ballot.i64!=0_ selection works correctly in wave32 mode. So this is more a codegen test. We have _llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-intrinsics.ll_ that checks ballot.

https://github.com/llvm/llvm-project/pull/73779