ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

musa: Upgrade MUSA SDK version to rc4.0.1 and use mudnn::Unary::IDENTITY op to accelerate D2D memory copy (#13647)

Browse source 
* musa: fix build warning (unused parameter)

Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>

* musa: upgrade MUSA SDK version to rc4.0.1

Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>

* musa: use mudnn::Unary::IDENTITY op to accelerate D2D memory copy

Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>

* Update ggml/src/ggml-cuda/cpy.cu

Co-authored-by: Johannes Gäßler <johannesg@5d6.de>

* musa: remove MUDNN_CHECK_GEN and use CUDA_CHECK_GEN instead in MUDNN_CHECK

Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>

---------

Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
This commit is contained in:
R0CKSTAR 2025-05-21 09:58:49 +08:00 • committed by GitHub
parent fb1cab201c
commit 33983057d0
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
10 changed files with 153 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

15
.devops/musa.Dockerfile
View file

@ -1,10 +1,10 @@
ARG UBUNTU_VERSION=22.04
# This needs to generally match the container host's environment.
ARG MUSA_VERSION=rc3.1.1
ARG MUSA_VERSION=rc4.0.1
# Target the MUSA build image
ARG BASE_MUSA_DEV_CONTAINER=mthreads/musa:${MUSA_VERSION}-devel-ubuntu${UBUNTU_VERSION}
ARG BASE_MUSA_DEV_CONTAINER=mthreads/musa:${MUSA_VERSION}-mudnn-devel-ubuntu${UBUNTU_VERSION}
ARG BASE_MUSA_RUN_CONTAINER=mthreads/musa:${MUSA_VERSION}-runtime-ubuntu${UBUNTU_VERSION}
ARG BASE_MUSA_RUN_CONTAINER=mthreads/musa:${MUSA_VERSION}-mudnn-runtime-ubuntu${UBUNTU_VERSION}
FROM ${BASE_MUSA_DEV_CONTAINER} AS build
@ -21,21 +21,14 @@ RUN apt-get update && \
libcurl4-openssl-dev \
libgomp1
COPY requirements.txt requirements.txt
COPY requirements requirements
RUN pip install --upgrade pip setuptools wheel \
&& pip install -r requirements.txt
WORKDIR /app
COPY . .
# Use the default MUSA archs if not specified
RUN if [ "${MUSA_DOCKER_ARCH}" != "default" ]; then \
export CMAKE_ARGS="-DMUSA_ARCHITECTURES=${MUSA_DOCKER_ARCH}"; \
fi && \
cmake -B build -DGGML_NATIVE=OFF -DGGML_MUSA=ON -DLLAMA_BUILD_TESTS=OFF -DGGML_BACKEND_DL=ON -DGGML_CPU_ALL_VARIANTS=ON ${CMAKE_ARGS} -DCMAKE_EXE_LINKER_FLAGS=-Wl,--allow-shlib-undefined . && \
cmake -B build -DGGML_NATIVE=OFF -DGGML_MUSA=ON -DGGML_BACKEND_DL=ON -DGGML_CPU_ALL_VARIANTS=ON -DLLAMA_BUILD_TESTS=OFF ${CMAKE_ARGS} -DCMAKE_EXE_LINKER_FLAGS=-Wl,--allow-shlib-undefined . && \
cmake --build build --config Release -j$(nproc)
RUN mkdir -p /app/lib && \