[NFC][OpenMP][Flang] Add smoke test for omp target parallel #77579
Conversation
Added test which proves that end-to-end compilation of omp target parallel costruct is successful for Flang compiler.
DominikAdamski
requested review from
kiranchandramohan,
kiranktp,
dpalermo,
jsjodin,
agozillon,
skatrak and
ergawy
| ! REQUIRES: flang | ||
| ! UNSUPPORTED: nvptx64-nvidia-cuda-LTO | ||
| ! UNSUPPORTED: aarch64-unknown-linux-gnu | ||
| ! UNSUPPORTED: aarch64-unknown-linux-gnu-LTO | ||
| ! UNSUPPORTED: x86_64-pc-linux-gnu | ||
| ! UNSUPPORTED: x86_64-pc-linux-gnu-LTO |
There was a problem hiding this comment.
nit: can we replace this with:
! REQUIRES: flang, amdgcn-amd-amdhsa
?
There was a problem hiding this comment.
If we specify requirements as follows: REQUIRES: flang, amdgcn-amd-amdhsa then we will launch this test only for flang and AMDGPU (source: https://llvm.org/docs/TestingGuide.html ) . If we skip amdgcn-amd-amdhsa requirement, then we can launch the test with Flang + nvptx64-nvidia-cuda.
There was a problem hiding this comment.
If we specify requirements as follows:
REQUIRES: flang, amdgcn-amd-amdhsathen we will launch this test only for flang and AMDGPU (source: https://llvm.org/docs/TestingGuide.html ) . If we skipamdgcn-amd-amdhsarequirement, then we can launch the test with Flang +nvptx64-nvidia-cuda.
Ah, my bad! I didn't notice the LTO part in nvptx64-nvidia-cuda-LTO!
Added test which proves that end-to-end compilation of omp target parallel costruct is successful for Flang compiler.
Added test which proves that end-to-end compilation of omp target parallel costruct is successful for Flang compiler.