Skip to content

Clarify MathDx usage with load_inline #517

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

Draft
msaroufim wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

Clarify MathDx usage with load_inline #517

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
5 changes: 3 additions & 2 deletions src/runners/modal_runner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,10 +33,11 @@
# cd src/runners && modal run test_cutlass_image.py
#
# For users writing submissions with torch.utils.cpp_extension.load_inline:
# C++ headers installed on the image (like CUTLASS) require explicit include paths:
# MathDx is already on CPLUS_INCLUDE_PATH. Include cuBLASDx in cuda_sources,
# which load_inline compiles with nvcc, rather than in cpp_sources:
# load_inline(
# ...
# extra_include_paths=["/opt/cutlass/include", "/opt/cutlass/tools/util/include"],
# cuda_sources=cuda_source, # #include <cublasdx.hpp> goes here
# )
# For raw nvcc compilation, CPLUS_INCLUDE_PATH is set so includes work automatically.
#
Expand Down