[PATCH] D109500: [openmp] Fix 51647, corrupt bitcode on amdgpu

Johannes Doerfert via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Mon Sep 13 08:59:44 PDT 2021


jdoerfert added a comment.

In D109500#2997666 <https://reviews.llvm.org/D109500#2997666>, @JonChesterfield wrote:

> Having trouble applying this to llvm-13 release, 'opt: Unknown command line argument '-openmp-opt-disable-state-machine-rewrite'
>
> @jdoerfert do we need to apply some state machine changes to llvm-13?
>
> diff vs 13 looks like
>
>   ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --function-signature --check-attributes --include-generated-funcs
>   -; RUN: opt -S -passes=openmp-opt < %s | FileCheck %s
>   +; RUN: opt --mtriple=amdgcn-amd-amdhsa --data-layout=A5 -S -passes=openmp-opt < %s | FileCheck %s --check-prefixes=AMDGPU
>   +; RUN: opt --mtriple=nvptx64--         -S -passes=openmp-opt < %s | FileCheck %s --check-prefixes=NVPTX
>   +; RUN: opt --mtriple=amdgcn-amd-amdhsa --data-layout=A5 -openmp-opt-disable-state-machine-rewrite -S -passes=openmp-opt < %s | FileCheck %s --check-prefixes=AMDGPU-DISABLED
>   +; RUN: opt --mtriple=nvptx64--         -openmp-opt-disable-state-machine-rewrite -S -passes=openmp-opt < %s | FileCheck %s --check-prefixes=NVPTX-DISABLED
>    
>
> so I expect I can apply the same change manually, without the --disabled test lines

Yes, remove the run lines for disable for back porting.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D109500/new/

https://reviews.llvm.org/D109500



More information about the llvm-commits mailing list