mirror of
https://git.datalinker.icu/vllm-project/vllm.git
synced 2026-04-17 04:17:04 +08:00
refactor
This commit is contained in:
atalman
parent
0b9c701495
commit
69b9a83985
@ -133,19 +133,25 @@ RUN ldconfig /usr/local/cuda-$(echo $CUDA_VERSION | cut -d. -f1,2)/compat/
|
||||
ARG PYTORCH_CUDA_INDEX_BASE_URL
|
||||
ARG USE_TORCH_NIGHTLY
|
||||
|
||||
# Set PyTorch index URL and prerelease flag based on USE_TORCH_NIGHTLY
|
||||
RUN if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=--prerelease=allow" >> /etc/environment; \
|
||||
else \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=" >> /etc/environment; \
|
||||
fi
|
||||
ENV PYTORCH_INDEX=""
|
||||
ENV PRERELEASE_FLAG=""
|
||||
RUN . /etc/environment && echo "PYTORCH_INDEX=${PYTORCH_INDEX}" && echo "PRERELEASE_FLAG=${PRERELEASE_FLAG}"
|
||||
|
||||
WORKDIR /workspace
|
||||
|
||||
# install build and runtime dependencies
|
||||
COPY requirements/common.txt requirements/common.txt
|
||||
COPY requirements/cuda.txt requirements/cuda.txt
|
||||
RUN --mount=type=cache,target=/root/.cache/uv \
|
||||
if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG="--prerelease=allow"; \
|
||||
else \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG=""; \
|
||||
fi && \
|
||||
. /etc/environment && \
|
||||
uv pip install --python /opt/venv/bin/python3 -r requirements/cuda.txt \
|
||||
--extra-index-url ${PYTORCH_INDEX} ${PRERELEASE_FLAG}
|
||||
|
||||
@ -175,14 +181,19 @@ ENV UV_INDEX_STRATEGY="unsafe-best-match"
|
||||
# Use copy mode to avoid hardlink failures with Docker cache mounts
|
||||
ENV UV_LINK_MODE=copy
|
||||
|
||||
RUN --mount=type=cache,target=/root/.cache/uv \
|
||||
if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG="--prerelease=allow"; \
|
||||
# Set PyTorch index URL and prerelease flag based on USE_TORCH_NIGHTLY
|
||||
RUN if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=--prerelease=allow" >> /etc/environment; \
|
||||
else \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG=""; \
|
||||
fi && \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=" >> /etc/environment; \
|
||||
fi
|
||||
ENV PYTORCH_INDEX=""
|
||||
ENV PRERELEASE_FLAG=""
|
||||
|
||||
RUN --mount=type=cache,target=/root/.cache/uv \
|
||||
. /etc/environment && \
|
||||
uv pip install --python /opt/venv/bin/python3 -r requirements/build.txt \
|
||||
--extra-index-url ${PYTORCH_INDEX} ${PRERELEASE_FLAG}
|
||||
|
||||
@ -316,14 +327,19 @@ ENV UV_INDEX_STRATEGY="unsafe-best-match"
|
||||
# Use copy mode to avoid hardlink failures with Docker cache mounts
|
||||
ENV UV_LINK_MODE=copy
|
||||
|
||||
RUN --mount=type=cache,target=/root/.cache/uv \
|
||||
if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG="--prerelease=allow"; \
|
||||
# Set PyTorch index URL and prerelease flag based on USE_TORCH_NIGHTLY
|
||||
RUN if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=--prerelease=allow" >> /etc/environment; \
|
||||
else \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG=""; \
|
||||
fi && \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=" >> /etc/environment; \
|
||||
fi
|
||||
ENV PYTORCH_INDEX=""
|
||||
ENV PRERELEASE_FLAG=""
|
||||
|
||||
RUN --mount=type=cache,target=/root/.cache/uv \
|
||||
. /etc/environment && \
|
||||
uv pip install --python /opt/venv/bin/python3 -r requirements/build.txt \
|
||||
--extra-index-url ${PYTORCH_INDEX} ${PRERELEASE_FLAG}
|
||||
|
||||
@ -383,19 +399,24 @@ ENV UV_INDEX_STRATEGY="unsafe-best-match"
|
||||
# Use copy mode to avoid hardlink failures with Docker cache mounts
|
||||
ENV UV_LINK_MODE=copy
|
||||
|
||||
# Set PyTorch index URL and prerelease flag based on USE_TORCH_NIGHTLY
|
||||
RUN if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=--prerelease=allow" >> /etc/environment; \
|
||||
else \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=" >> /etc/environment; \
|
||||
fi
|
||||
ENV PYTORCH_INDEX=""
|
||||
ENV PRERELEASE_FLAG=""
|
||||
|
||||
# Install libnuma-dev, required by fastsafetensors (fixes #20384)
|
||||
RUN apt-get update && apt-get install -y --no-install-recommends libnuma-dev && rm -rf /var/lib/apt/lists/*
|
||||
COPY requirements/lint.txt requirements/lint.txt
|
||||
COPY requirements/test.txt requirements/test.txt
|
||||
COPY requirements/dev.txt requirements/dev.txt
|
||||
RUN --mount=type=cache,target=/root/.cache/uv \
|
||||
if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG="--prerelease=allow"; \
|
||||
else \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG=""; \
|
||||
fi && \
|
||||
. /etc/environment && \
|
||||
uv pip install --python /opt/venv/bin/python3 -r requirements/dev.txt \
|
||||
--extra-index-url ${PYTORCH_INDEX} ${PRERELEASE_FLAG}
|
||||
#################### DEV IMAGE ####################
|
||||
@ -492,16 +513,22 @@ RUN ldconfig /usr/local/cuda-$(echo $CUDA_VERSION | cut -d. -f1,2)/compat/
|
||||
# This is ~2GB and rarely changes
|
||||
ARG PYTORCH_CUDA_INDEX_BASE_URL
|
||||
ARG USE_TORCH_NIGHTLY
|
||||
|
||||
# Set PyTorch index URL and prerelease flag based on USE_TORCH_NIGHTLY
|
||||
RUN if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=--prerelease=allow" >> /etc/environment; \
|
||||
else \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=" >> /etc/environment; \
|
||||
fi
|
||||
ENV PYTORCH_INDEX=""
|
||||
ENV PRERELEASE_FLAG=""
|
||||
|
||||
COPY requirements/common.txt /tmp/common.txt
|
||||
COPY requirements/cuda.txt /tmp/requirements-cuda.txt
|
||||
RUN --mount=type=cache,target=/root/.cache/uv \
|
||||
if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG="--prerelease=allow"; \
|
||||
else \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG=""; \
|
||||
fi && \
|
||||
. /etc/environment && \
|
||||
uv pip install --system -r /tmp/requirements-cuda.txt \
|
||||
--extra-index-url ${PYTORCH_INDEX} ${PRERELEASE_FLAG} && \
|
||||
rm /tmp/requirements-cuda.txt /tmp/common.txt
|
||||
@ -560,13 +587,7 @@ ARG USE_TORCH_NIGHTLY
|
||||
# Install vllm wheel first, so that torch etc will be installed.
|
||||
RUN --mount=type=bind,from=build,src=/workspace/dist,target=/vllm-workspace/dist \
|
||||
--mount=type=cache,target=/root/.cache/uv \
|
||||
if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG="--prerelease=allow"; \
|
||||
else \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG=""; \
|
||||
fi && \
|
||||
. /etc/environment && \
|
||||
uv pip install --system dist/*.whl --verbose \
|
||||
--extra-index-url ${PYTORCH_INDEX} ${PRERELEASE_FLAG}
|
||||
|
||||
@ -589,13 +610,7 @@ ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
|
||||
# Install EP kernels wheels (pplx-kernels and DeepEP) that have been built in the `build` stage
|
||||
RUN --mount=type=bind,from=build,src=/tmp/ep_kernels_workspace/dist,target=/vllm-workspace/ep_kernels/dist \
|
||||
--mount=type=cache,target=/root/.cache/uv \
|
||||
if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG="--prerelease=allow"; \
|
||||
else \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG=""; \
|
||||
fi && \
|
||||
. /etc/environment && \
|
||||
uv pip install --system ep_kernels/dist/*.whl --verbose \
|
||||
--extra-index-url ${PYTORCH_INDEX} ${PRERELEASE_FLAG}
|
||||
|
||||
@ -631,6 +646,17 @@ ENV UV_INDEX_STRATEGY="unsafe-best-match"
|
||||
# Use copy mode to avoid hardlink failures with Docker cache mounts
|
||||
ENV UV_LINK_MODE=copy
|
||||
|
||||
# Set PyTorch index URL and prerelease flag based on USE_TORCH_NIGHTLY
|
||||
RUN if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=--prerelease=allow" >> /etc/environment; \
|
||||
else \
|
||||
echo "PYTORCH_INDEX=${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')" >> /etc/environment; \
|
||||
echo "PRERELEASE_FLAG=" >> /etc/environment; \
|
||||
fi
|
||||
ENV PYTORCH_INDEX=""
|
||||
ENV PRERELEASE_FLAG=""
|
||||
|
||||
RUN echo 'tzdata tzdata/Areas select America' | debconf-set-selections \
|
||||
&& echo 'tzdata tzdata/Zones/America select Los_Angeles' | debconf-set-selections \
|
||||
&& apt-get update -y \
|
||||
@ -640,13 +666,7 @@ RUN echo 'tzdata tzdata/Areas select America' | debconf-set-selections \
|
||||
RUN --mount=type=cache,target=/root/.cache/uv \
|
||||
CUDA_MAJOR="${CUDA_VERSION%%.*}"; \
|
||||
if [ "$CUDA_MAJOR" -ge 12 ]; then \
|
||||
if [ "$USE_TORCH_NIGHTLY" = "true" ]; then \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG="--prerelease=allow"; \
|
||||
else \
|
||||
PYTORCH_INDEX="${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.')"; \
|
||||
PRERELEASE_FLAG=""; \
|
||||
fi && \
|
||||
. /etc/environment && \
|
||||
uv pip install --system -r requirements/dev.txt \
|
||||
--extra-index-url ${PYTORCH_INDEX} ${PRERELEASE_FLAG}; \
|
||||
fi
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user