huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 12:24:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge 4e1c68e6f8 into c1cf36c0dc

Browse Source
This commit is contained in:
Funtowicz Morgan 2025-02-05 09:24:02 +01:00 committed by GitHub
commit c837843264
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 17 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
.github/workflows/build.yaml vendored
View File

@ -156,6 +156,15 @@ jobs:
username: ${{ secrets.AZURE_DOCKER_USERNAME }}
password: ${{ secrets.AZURE_DOCKER_PASSWORD }}
registry: db4c2190dd824d1f950f5d1555fbadf0.azurecr.io
- name: configure aws credentials
id: aws-creds
uses: aws-actions/configure-aws-credentials@e3dd6a429d7300a6a4c196c26e071d42e0343502
with:
role-to-assume: ${{ secrets.AWS_ROLE_GITHUB_BUILDX_CACHE }}
role-duration-seconds: 18000
aws-region: us-east-1
output-credentials: true
# If pull request
- name: Extract metadata (tags, labels) for Docker
if: ${{ github.event_name == 'pull_request' }}
@ -186,6 +195,8 @@ jobs:
- name: Build and push Docker image
id: build-and-push
uses: docker/build-push-action@v4
env:
DOCKER_BUILD_SUMMARY: false
with:
context: .
file: ${{ env.DOCKERFILE }}
@ -197,13 +208,14 @@ jobs:
PLATFORM=${{ env.PLATFORM }}
build_type=${{ env.BUILD_TYPE }}
sccache_gha_enabled=on
secrets: |
actions_cache_url=${{ env.ACTIONS_CACHE_URL }}
actions_runtime_token=${{ env.ACTIONS_RUNTIME_TOKEN }}
target: ${{ env.TARGET }}
tags: ${{ steps.meta.outputs.tags || steps.meta-pr.outputs.tags }}
labels: ${{ steps.meta.outputs.labels || steps.meta-pr.outputs.labels }}
cache-from: type=s3,region=us-east-1,bucket=ci-docker-buildx-cache,name=text-generation-inference-cache${{ env.LABEL }},mode=min,access_key_id=${{ secrets.S3_CI_DOCKER_BUILDX_CACHE_ACCESS_KEY_ID }},secret_access_key=${{ secrets.S3_CI_DOCKER_BUILDX_CACHE_SECRET_ACCESS_KEY }},mode=max
cache-to: type=s3,region=us-east-1,bucket=ci-docker-buildx-cache,name=text-generation-inference-cache${{ env.LABEL }},mode=min,access_key_id=${{ secrets.S3_CI_DOCKER_BUILDX_CACHE_ACCESS_KEY_ID }},secret_access_key=${{ secrets.S3_CI_DOCKER_BUILDX_CACHE_SECRET_ACCESS_KEY }},mode=max
cache-from: type=s3,region=us-east-1,bucket=${{ vars.AWS_S3BUCKET_GITHUB_BUILDX_CACHE }},name=text-generation-inference-cache${{ env.LABEL }},mode=min,access_key_id=${{ steps.aws-creds.outputs.aws-access-key-id }},secret_access_key=${{ steps.aws-creds.outputs.aws-secret-access-key }},session_token=${{ steps.aws-creds.outputs.aws-session-token }},mode=max
cache-to: type=s3,region=us-east-1,bucket=${{ vars.AWS_S3BUCKET_GITHUB_BUILDX_CACHE }},name=text-generation-inference-cache${{ env.LABEL }},mode=min,access_key_id=${{ steps.aws-creds.outputs.aws-access-key-id }},secret_access_key=${{ steps.aws-creds.outputs.aws-secret-access-key }},session_token=${{ steps.aws-creds.outputs.aws-session-token }},mode=max
- name: Final
id: final
run: |

8
Dockerfile_trtllm
View File

@ -64,8 +64,6 @@ WORKDIR /usr/src/text-generation-inference
ARG cuda_arch_list
ARG build_type
ARG sccache_gha_enabled
ARG actions_cache_url
ARG actions_runtime_token
# Install Rust
ENV PATH="/root/.cargo/bin:$PATH"
@ -83,8 +81,6 @@ ENV CUDA_ARCH_LIST=${cuda_arch_list}
# SCCACHE Specifics args - before finding a better, more generic, way...
ENV SCCACHE_GHA_ENABLED=${sccache_gha_enabled}
ENV ACTIONS_CACHE_URL=${actions_cache_url}
ENV ACTIONS_RUNTIME_TOKEN=${actions_runtime_token}
COPY Cargo.lock Cargo.lock
COPY Cargo.toml Cargo.toml
@ -98,7 +94,9 @@ COPY --from=mpi-builder /usr/local/mpi /usr/local/mpi
ENV RUSTC_WRAPPER=sccache
ENV CMAKE_INSTALL_PREFIX=$TGI_INSTALL_PREFIX
RUN export CMAKE_C_COMPILER_LAUNCHER=sccache && \
RUN --mount=type=secret,id=actions_cache_url,env=ACTIONS_CACHE_URL \
--mount=type=secret,id=actions_runtime_token,env=ACTIONS_RUNTIME_TOKEN \
export CMAKE_C_COMPILER_LAUNCHER=sccache && \
export CMAKE_CXX_COMPILER_LAUNCHER=sccache && \
export CMAKE_CUDA_COMPILER_LAUNCHER=sccache && \
mkdir $TGI_INSTALL_PREFIX && mkdir "$TGI_INSTALL_PREFIX/include" && mkdir "$TGI_INSTALL_PREFIX/lib" && \