diff options
| author | Renaud Kauffmann <rkauffmann@nvidia.com> | 2025-01-22 19:06:28 -0800 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-01-22 19:06:28 -0800 |
| commit | 652ff20140d79544db4dfa21314fc62c3c9182e5 (patch) | |
| tree | 3047a840b40aaa427eefc5a710c20053a760025d /llvm/lib/CodeGen/MachineFunction.cpp | |
| parent | b46fcb9fa32f24660b1b8858d5c4cbdb76ef9d8b (diff) | |
| download | llvm-652ff20140d79544db4dfa21314fc62c3c9182e5.zip llvm-652ff20140d79544db4dfa21314fc62c3c9182e5.tar.gz llvm-652ff20140d79544db4dfa21314fc62c3c9182e5.tar.bz2 | |
[flang][cuda] Adding atomicadd as a cudadevice intrinsic and converting it LLVM dialect (#123840)
With these changes, CUF atomic operations are handled as cudadevice
intrinsics and are converted straight to the LLVM dialect with the
`llvm.atomicrw` operation.
I am only submitting changes for `atomicadd` to gather feedback. If we
are to proceed with these changes I will add support for all other
applicable atomic operations following this pattern.
Diffstat (limited to 'llvm/lib/CodeGen/MachineFunction.cpp')
0 files changed, 0 insertions, 0 deletions