AMDGPU: Make default AMDHSA Code Object Version to be 5

[cfang]

Summary

Also update LIT tests and docs.

[arsenm]

Are we really ready to switch the default? We aren't even optimizing the new locations yet?

[arsenm]

Typo in patch title

[kzhuravl]

Need to also update: https://github.com/llvm/llvm-project/blob/main/clang/include/clang/Driver/Options.td#L3626

[cfang]

Update https://github.com/llvm/llvm-project/blob/main/clang/include/clang/Driver/Options.td#L3626
as well as the corresponding LIT tests.

[arsenm]

LGTM. I thought this already happened?

[ronlieb]

lets hold off landing for a few more days?
checking some internal related testing results

Thanks,

[lamb-j]

clang/test/Driver/hip-device-libs.hip

144

500?

I think we may also need to update the comment here: https://github.com/llvm/llvm-project/blob/e9569748def85a2f6bb8066e516a84e6933ef407/clang/test/Driver/hip-device-libs.hip#LL196C1-L196C83

154

Should this be ABI5? That is, if CodeObjectVersion < 4, should it be set to 5 (instead of 4)?

May also be related to a needed change here: https://github.com/llvm/llvm-project/blob/6db007a0654ed7a6ed5c3aa3b61a937c19a6bc6b/clang/lib/Driver/ToolChains/ROCm.h#L29

llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp

31–34

We may also want to update this file here:

https://github.com/llvm/llvm-project/blob/e9569748def85a2f6bb8066e516a84e6933ef407/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp#L162