[Openmp-commits] [PATCH] D101326: [OpenMP][libomptarget][test] Separate lit tests for different offloading targets (2/2)

[Joachim Protze via Phabricator via Openmp-commits]

protze.joachim added a comment.

https://reviews.llvm.org/D101498 will fix this by allowing to distinguish
platforms supporting unified shared memory from platforms that don't.
Do you think, that looking at the gpu generation is sufficient, or does the
host architecture matter as well?

Kelvin Li via Phabricator <reviews at reviews.llvm.org> schrieb am Do., 29.
Apr. 2021, 20:02:

> kkwli0 added a comment.
>
> The following regressions seem to be caused by this patch on Power9+V100.
>
>   Unexpectedly Passed Tests (4):
>     libomptarget :: nvptx64-nvidia-cuda :: unified_shared_memory/api.c
>     libomptarget :: nvptx64-nvidia-cuda ::
>
> unified_shared_memory/close_enter_exit.c
>
>   libomptarget :: nvptx64-nvidia-cuda ::
>
> unified_shared_memory/close_modifier.c
>
>   libomptarget :: nvptx64-nvidia-cuda ::
>
> unified_shared_memory/shared_update.c
>
> Repository:
>
>   rG LLVM Github Monorepo
>
> CHANGES SINCE LAST ACTION
>
>   https://reviews.llvm.org/D101326/new/
>
> https://reviews.llvm.org/D101326


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D101326/new/

https://reviews.llvm.org/D101326