huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 04:14:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

backend(trtllm): what if we expose ENV instead of inline?

Browse Source
This commit is contained in:
Morgan Funtowicz 2025-01-24 17:47:49 +01:00
parent 556a61d143
commit a8a9168065
1 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
Dockerfile_trtllm
View File

@ -61,6 +61,7 @@ FROM cuda-builder AS tgi-builder
WORKDIR /usr/src/text-generation-inference
# Scoped global args reuse
ARG cuda_arch_list
ARG build_type
ARG sccache_gha_enabled
ARG actions_cache_url
@ -74,8 +75,8 @@ RUN curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | bash -s -- -y &&
cargo install sccache --locked
ENV LD_LIBRARY_PATH="/usr/local/mpi/lib:$LD_LIBRARY_PATH"
ENV PKG_CONFIG_PATH="/usr/local/mpi/lib/pkgconfig:$PKG_CONFIG_PATH"
ENV CMAKE_PREFIX_PATH="/usr/local/mpi:/usr/local/tensorrt:$CMAKE_PREFIX_PATH"
ENV PKG_CONFIG_PATH="/usr/local/mpi/lib/pkgconfig"
ENV CMAKE_PREFIX_PATH="/usr/local/mpi:/usr/local/tensorrt"
ENV USE_LLD_LINKER=ON
ENV CUDA_ARCH_LIST=${cuda_arch_list}
@ -95,9 +96,9 @@ COPY launcher launcher
COPY --from=trt-builder /usr/local/tensorrt /usr/local/tensorrt
COPY --from=mpi-builder /usr/local/mpi /usr/local/mpi
ENV RUSTC_WRAPPER=sccache
ENV CMAKE_INSTALL_PREFIX=$TGI_INSTALL_PREFIX
RUN mkdir $TGI_INSTALL_PREFIX && mkdir "$TGI_INSTALL_PREFIX/include" && mkdir "$TGI_INSTALL_PREFIX/lib" && \
CMAKE_INSTALL_PREFIX=$TGI_INSTALL_PREFIX \
RUSTC_WRAPPER=sccache \
cargo build --profile ${build_type} --package text-generation-backends-trtllm --bin text-generation-backends-trtllm && \
sccache --show-stats