text-generation-inference/server/Makefile-vllm
Mohit Sharma 8f66d323d0
Update vllm kernels for ROCM (#2826)
* (vllm) updated vllm rocm kernels

* revert silu

* update partition size

* remove grouped_topk

* (nit) remove log

* update moe-kernels commit
2024-12-18 12:44:42 +01:00

14 lines
480 B
Plaintext

commit_rocm := de990cd12537f78f74e40b5c8ee1a62d63d734dd
build-vllm-rocm:
if [ ! -d 'vllm' ]; then \
pip install -U ninja packaging --no-cache-dir && \
git clone https://github.com/mht-sharma/vllm.git vllm; \
fi
cd vllm && git fetch && git checkout $(commit_rocm) && \
PYTORCH_ROCM_ARCH="gfx90a;gfx942" python setup.py build
install-vllm-rocm: build-vllm-rocm
cd vllm && git fetch && git checkout $(commit_rocm) && \
PYTORCH_ROCM_ARCH="gfx90a;gfx942" pip install -e .