A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 16, 22:00 UTC
Fix release pipeline's -e flag (#6390)
Simon Mo committed
Jul 12, 2024 at 14:08 UTC
21b2dcedabb84332a73e8cda3886a6988c52a7df
1 file changed
+1
-1
.buildkite/release-pipeline.yaml
+1
-1
@@ -3,7 +3,7 @@ steps:
3
agents:
4
queue: cpu_queue
5
commands:
6
- - "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg USE_SCCACHE=1 --build-arg CUDA_VERSION={{matrix.cuda_version}} --build-arg PYTHON_VERSION={{matrix.python_version}} --tag vllm-ci:build-image -e CMAKE_BUILD_TYPE=Release --target build --progress plain ."
6
+ - "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg USE_SCCACHE=1 --build-arg CUDA_VERSION={{matrix.cuda_version}} --build-arg PYTHON_VERSION={{matrix.python_version}} --tag vllm-ci:build-image --target build --progress plain ."
7
- "mkdir artifacts"
8
- "docker run --rm -v $(pwd)/artifacts:/artifacts_host vllm-ci:build-image cp -r dist /artifacts_host"
9
# rename the files to change linux -> manylinux1