ci: align MUSA Docker image with upstream SDK 5.2.0 and mp_31

This commit is contained in:
leejet
2026-10-17 14:14:29 +08:00
parent 68285df319
commit f0aac524d3
2 changed files with 9 additions and 8 deletions
+5 -6
View File
@@ -1,7 +1,6 @@
ARG MUSA_VERSION=rc4.2.0
ARG UBUNTU_VERSION=22.04
ARG MUSA_IMAGE=registry.mthreads.com/mcconline/inference/pytorch:2.9.1.post1-py3.10-musa5.2.0-mp31-devel-ubuntu22.04-amd64
FROM mthreads/musa:${MUSA_VERSION}-devel-ubuntu${UBUNTU_VERSION}-amd64 as build
FROM ${MUSA_IMAGE} AS build
# sd-server embeds the web UI at build time, so the build image needs Node/pnpm.
RUN apt-get update && apt-get install -y --no-install-recommends ccache cmake git ca-certificates curl gnupg && \
@@ -22,9 +21,8 @@ COPY . .
RUN mkdir build && cd build && \
cmake .. -DCMAKE_C_COMPILER=clang -DCMAKE_CXX_COMPILER=clang++ \
-DCMAKE_C_FLAGS="${CMAKE_C_FLAGS} -fopenmp -I/usr/lib/llvm-14/lib/clang/14.0.0/include -L/usr/lib/llvm-14/lib" \
-DCMAKE_CXX_FLAGS="${CMAKE_CXX_FLAGS} -fopenmp -I/usr/lib/llvm-14/lib/clang/14.0.0/include -L/usr/lib/llvm-14/lib" \
-DSD_MUSA=ON \
-DMUSA_ARCHITECTURES=31 \
-DSD_BUILD_SHARED_LIBS=ON \
-DGGML_NATIVE=OFF \
-DSD_BUILD_SHARED_GGML_LIB=ON \
@@ -35,7 +33,7 @@ RUN mkdir build && cd build && \
-DCMAKE_BUILD_TYPE=Release && \
cmake --build . --config Release
FROM mthreads/musa:${MUSA_VERSION}-runtime-ubuntu${UBUNTU_VERSION}-amd64 as runtime
FROM ${MUSA_IMAGE} AS runtime
COPY --from=build /sd.cpp/build/bin /sd.cpp/bin
RUN printf '#!/bin/sh\nexec /sd.cpp/bin/sd-cli "$@"\n' > /sd-cli && \
@@ -45,3 +43,4 @@ RUN printf '#!/bin/sh\nexec /sd.cpp/bin/sd-cli "$@"\n' > /sd-cli && \
WORKDIR /sd.cpp
ENTRYPOINT [ "/sd-cli" ]
CMD []
+4 -2
View File
@@ -119,14 +119,16 @@ cmake --build . --config Release
## Build with MUSA
This provides GPU acceleration using Moore Threads GPU. Make sure to have the MUSA toolkit installed.
This provides GPU acceleration using Moore Threads GPUs. The configuration below follows upstream validation with MUSA SDK 5.2.0 and MTT S5000 (PH1, `mp_31`). Install that toolkit before building. Older SDKs such as 4.2 may lack the CUB headers required by the current backend.
```shell
mkdir build && cd build
cmake .. -DCMAKE_C_COMPILER=/usr/local/musa/bin/clang -DCMAKE_CXX_COMPILER=/usr/local/musa/bin/clang++ -DSD_MUSA=ON -DCMAKE_BUILD_TYPE=Release
cmake .. -DCMAKE_C_COMPILER=/usr/local/musa/bin/clang -DCMAKE_CXX_COMPILER=/usr/local/musa/bin/clang++ -DSD_MUSA=ON -DMUSA_ARCHITECTURES=31 -DCMAKE_BUILD_TYPE=Release
cmake --build . --config Release
```
The MUSA Docker image uses the same SDK 5.2.0 development image as upstream for both build and runtime stages, and targets `mp_31` only. It includes the development tools and PyTorch from that base image, so it is larger than a minimal runtime image. Other GPU architectures require separate validation.
## Build with Metal
Using Metal makes the computation run on the GPU. Currently, there are some issues with Metal when performing operations on very large matrices, making it highly inefficient at the moment. Performance improvements are expected in the near future.