youngkingdom/vllm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[Build/CI] Fix CUDA 11.8 build (#17679)

Browse Source 
Signed-off-by: Tyler Michael Smith <tyler@neuralmagic.com>
Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
Signed-off-by: Tyler Michael Smith <tysmith@redhat.com>
Co-authored-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
This commit is contained in:
Tyler Michael Smith
2025-05-22 15:13:54 -04:00
committed by GitHub
parent f8d2cc5f55
commit 6e588da0f4
9 changed files with 78 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
docker/Dockerfile
View File

@ -263,8 +263,11 @@ if [ "$TARGETPLATFORM" != "linux/arm64" ]; then \
export TORCH_CUDA_ARCH_LIST='7.5 8.0 8.9 9.0 10.0+PTX'; \
else \
export TORCH_CUDA_ARCH_LIST='7.5 8.0 8.9 9.0+PTX'; \
fi && \
export FLASHINFER_ENABLE_AOT=1; \
fi; \
CUDA_MAJOR="${CUDA_VERSION%%.*}"; \
if [ "$CUDA_MAJOR" -lt 12 ]; then \
export FLASHINFER_ENABLE_SM90=0; \
fi; \
uv pip install --system --no-build-isolation "git+https://github.com/flashinfer-ai/flashinfer@21ea1d2545f74782b91eb8c08fd503ac4c0743fc" ; \
fi
COPY examples examples
@ -275,7 +278,7 @@ RUN --mount=type=cache,target=/root/.cache/uv \
. /etc/environment && \
uv pip list
# Although we build Flashinfer with AOT mode, there's still
# Even when we build Flashinfer with AOT mode, there's still
# some issues w.r.t. JIT compilation. Therefore we need to
# install build dependencies for JIT compilation.
# TODO: Remove this once FlashInfer AOT wheel is fixed
@ -303,8 +306,11 @@ RUN --mount=type=cache,target=/root/.cache/uv \
uv pip install --system --no-build-isolation "git+https://github.com/state-spaces/mamba@v2.2.4"
# install development dependencies (for testing)
RUN --mount=type=cache,target=/root/.cache/uv \
uv pip install --system -r requirements/dev.txt
RUN --mount=type=cache,target=/root/.cache/uv \
CUDA_MAJOR="${CUDA_VERSION%%.*}"; \
if [ "$CUDA_MAJOR" -ge 12 ]; then \
uv pip install --system -r requirements/dev.txt; \
fi
# install development dependencies (for testing)
RUN --mount=type=cache,target=/root/.cache/uv \