[PATCH] D146287: [AMDGPU][GISel] Add inverse ballot intrinsic

[Thomas Symalla via Phabricator via llvm-commits]

tsymalla added inline comments.


================
Comment at: llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.inverse.ballot.i32.ll:2
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -march=amdgcn -mcpu=gfx1010 -mattr=+wavefrontsize32,-wavefrontsize64 -global-isel -amdgpu-global-isel-risky-select -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX10 %s
+; RUN: llc -march=amdgcn -mcpu=gfx1100 -amdgpu-enable-delay-alu=0 -mattr=+wavefrontsize32,-wavefrontsize64 -global-isel -amdgpu-global-isel-risky-select -verify-machineinstrs < %s | FileCheck  -check-prefixes=GFX11 %s
----------------
Why do you require -risky-select?


================
Comment at: llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.inverse.ballot.i32.ll:21
+; GFX11-NEXT:    global_store_b32 v[0:1], v2, off
+; GFX11-NEXT:    s_sendmsg sendmsg(MSG_DEALLOC_VGPRS)
+; GFX11-NEXT:    s_endpgm
----------------
OutOfCache wrote:
> This `s_sendmsg` instruction is the only difference between gfx10 and gfx11 in every test. Should I remove the gfx10 test?
I'd add just a single test run if the differences are minor.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D146287/new/

https://reviews.llvm.org/D146287