Skip to content

fix: remove cutlass-dsl stub that crashes TRT-LLM on CUDA 13.1 (#7372) #7393

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
tanmayv25 wants to merge 1 commit into
base: release/1.0.0-post.1
Choose a base branch
from
+5 −1
Open

fix: remove cutlass-dsl stub that crashes TRT-LLM on CUDA 13.1 (#7372) #7393

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: 5 additions & 1 deletion container/templates/trtllm_runtime.Dockerfile
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -252,7 +252,11 @@ RUN --mount=type=bind,source=./container/deps/requirements.txt,target=/tmp/requi
--extra-index-url https://download.pytorch.org/whl/cu130 \
--requirement /tmp/requirements.txt \
--requirement /tmp/requirements.test.txt \
cupy-cuda13x
cupy-cuda13x && \
# nvidia-cutlass-dsl-libs-base==4.4.1 (transitive dep) ships a stub cute/experimental/__init__.py
# that unconditionally raises NotImplementedError, crashing TRT-LLM on import. cutlass-dsl==4.3.4
# (pinned by TRT-LLM) works without cute/experimental/. Remove the stub to fix the NotImplementedError.
rm -rf ${VIRTUAL_ENV}/lib/python${PYTHON_VERSION}/site-packages/nvidia_cutlass_dsl/python_packages/cutlass/cute/experimental/

# Copy tests, deploy and components for CI with correct ownership
# Pattern: COPY --chmod=775 <path>; chmod g+w <path> done later as root because COPY --chmod only affects <path>/*, not <path>
Expand Down
Loading