review.tizen.org Git - platform/upstream/llvm.git/commit

projects / platform / upstream / llvm.git / commit

author	Joseph Huber <jhuber6@vols.utk.edu>
	Thu, 5 Jan 2023 04:18:41 +0000 (22:18 -0600)
committer	Joseph Huber <jhuber6@vols.utk.edu>
	Wed, 11 Jan 2023 16:30:30 +0000 (10:30 -0600)
commit	56ebfca4bcc0e7a54989ad7aea5395b27f67e373
tree	5a930b0fd8748580ef408a2561415c67c5d46d4a	tree \| snapshot
parent	fbd207fd6f6bc76a41cbf90fb6d87c6fa627ddf7	commit \| diff

[CUDA][HIP] Add support for `--offload-arch=native` to CUDA and refactor

This patch adds basic support for `--offload-arch=native` to CUDA. This
is done using the `nvptx-arch` tool that was introduced previously. Some
of the logic for handling executing these tools was factored into a
common helper as well. This patch does not add support for OpenMP or the
"new" driver. That will be done later.

Reviewed By: yaxunl

Differential Revision: https://reviews.llvm.org/D141051

15 files changed:

clang/include/clang/Basic/DiagnosticDriverKinds.td		diff \| blob \| history
clang/include/clang/Driver/Options.td		diff \| blob \| history
clang/include/clang/Driver/ToolChain.h		diff \| blob \| history
clang/lib/Driver/Driver.cpp		diff \| blob \| history
clang/lib/Driver/ToolChain.cpp		diff \| blob \| history
clang/lib/Driver/ToolChains/AMDGPU.cpp		diff \| blob \| history
clang/lib/Driver/ToolChains/AMDGPU.h		diff \| blob \| history
clang/lib/Driver/ToolChains/AMDGPUOpenMP.cpp		diff \| blob \| history
clang/lib/Driver/ToolChains/Cuda.cpp		diff \| blob \| history
clang/lib/Driver/ToolChains/Cuda.h		diff \| blob \| history
clang/test/Driver/Inputs/nvptx-arch/nvptx_arch_fail	[new file with mode: 0755]	blob
clang/test/Driver/Inputs/nvptx-arch/nvptx_arch_sm_70	[new file with mode: 0755]	blob
clang/test/Driver/amdgpu-hip-system-arch.c	[new file with mode: 0644]	blob
clang/test/Driver/amdgpu-openmp-system-arch-fail.c		diff \| blob \| history
clang/test/Driver/nvptx-cuda-system-arch.c	[new file with mode: 0644]	blob

Domain: System / Toolchain;

RSS Atom