Skip to content

[ROCm] Add HIP support for AMD Instinct GPUs #31

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
andyluo7 wants to merge 1 commit into
base: main
Choose a base branch
from
Open

[ROCm] Add HIP support for AMD Instinct GPUs #31

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions setup.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -58,7 +58,7 @@
# CUDA / ROCm specific
# -------------------------------------------------
if IS_HIP:
archs = os.getenv("GPU_ARCHS", "native").split(";")
archs = os.getenv("GPU_ARCHS", "gfx942").split(";")
nvcc_flags += [f"--offload-arch={arch}" for arch in archs]
else:
# CUDA only
Expand Down Expand Up @@ -115,14 +115,14 @@
],
extra_compile_args={
"cxx": cxx_flags,
"nvcc": nvcc_flags + [
"nvcc": nvcc_flags + ([] if IS_HIP else [
# The following definitions must be undefined
# since we need half-precision operation.
"--extended-lambda",
"-U__CUDA_NO_HALF_OPERATORS__",
"-U__CUDA_NO_HALF_CONVERSIONS__",
"-U__CUDA_NO_HALF2_OPERATORS__",
],
]),
},
),

Expand Down
Loading