xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-04-02 20:17:06 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix Dockerfile.rocm to set VLLM_TARGET_DEVICE=rocm

Browse Source 
Without this environment variable, the build defaults to VLLM_TARGET_DEVICE=cuda
as defined in vllm/envs.py, causing cmake to configure with -DVLLM_TARGET_DEVICE=cuda
instead of -DVLLM_TARGET_DEVICE=rocm.

This fix ensures ROCm Docker builds use the correct target device.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

Signed-off-by: westers <steve.westerhouse@origami-analytics.com>
This commit is contained in:
westers 2025-12-20 10:21:34 -06:00
parent 560ae9638c
commit ff00ef309d
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docker/Dockerfile.rocm
View File

@ -51,6 +51,8 @@ FROM fetch_vllm_${REMOTE_VLLM} AS fetch_vllm
# -----------------------
# vLLM build stages
FROM fetch_vllm AS build_vllm
# Set target device to ROCm for build
ENV VLLM_TARGET_DEVICE=rocm
# Build vLLM
RUN cd vllm \
&& python3 -m pip install -r requirements/rocm.txt \