Skip to content

🩹 Fix *_nvptx toolchains after bf16e6a #261

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

Merged
merged 1 commit into from
Jan 2, 2025
Merged

🩹 Fix *_nvptx toolchains after bf16e6a #261

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
9 changes: 7 additions & 2 deletions ll/args.bzl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -228,6 +228,11 @@ def compile_object_args(
"cuda_nvptx",
"hip_nvptx",
]:
# Enable the use of libc++ for the clang-based device compilation.
# This is only used with the clang-based device compilation. The
# cuda_nvptx_nvcc toolchain uses libstdc++ from Nix.
args.add("-D_ALLOW_UNSUPPORTED_LIBCPP")

args.add("-Wno-unknown-cuda-version") # Will always be unknown.
args.add("-xcuda")
if toolchain.LL_CUDA_TOOLKIT != "":
Expand Down Expand Up @@ -528,8 +533,8 @@ def link_executable_args(ctx, in_files, out_file, mode):
args.add(toolchain.LL_CUDA_TOOLKIT, format = "-L%s/lib/stubs")

args.add("-lcuda")
args.add("-lcudart_static")
args.add("-lcupti_static")
args.add("-lcudart")
args.add("-lcupti")
if ctx.attr.compilation_mode == "hip_amdgpu":
args.add(toolchain.hip_runtime[0].dirname, format = "-L%s")
args.add(toolchain.hip_runtime[0].basename, format = "-l:%s")
Expand Down