matt/ollama37
1
0
Fork 0
mirror of https://github.com/dogkeeper886/ollama37.git synced 2025-12-10 07:46:59 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add cuda v12 variant and selection logic

Browse Source 
Based on compute capability and driver version, pick
v12 or v11 cuda variants.
This commit is contained in:
Daniel Hiltgen
2024-06-13 20:46:14 -07:00
parent fc3b4cda89
commit 4fe3a556fa
4 changed files with 84 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files

43
Dockerfile
View File

@@ -1,7 +1,7 @@
ARG GOLANG_VERSION=1.22.5
ARG CMAKE_VERSION=3.22.1
# this CUDA_VERSION corresponds with the one specified in docs/gpu.md
ARG CUDA_VERSION=11.3.1
ARG CUDA_VERSION_11=11.3.1
ARG CUDA_VERSION_12=12.4.0
ARG ROCM_VERSION=6.1.2
ARG JETPACK_6=r36.2.0
ARG JETPACK_5=r35.4.1
@@ -13,7 +13,7 @@ COPY .git .git
COPY .gitmodules .gitmodules
COPY llm llm
FROM --platform=linux/amd64 nvidia/cuda:$CUDA_VERSION-devel-centos7 AS cuda-build-amd64
FROM --platform=linux/amd64 nvidia/cuda:$CUDA_VERSION_11-devel-centos7 AS cuda-11-build-amd64
ARG CMAKE_VERSION
COPY ./scripts/rh_linux_deps.sh /
RUN CMAKE_VERSION=${CMAKE_VERSION} sh /rh_linux_deps.sh
@@ -23,9 +23,29 @@ WORKDIR /go/src/github.com/ollama/ollama/llm/generate
ARG CGO_CFLAGS
ENV GOARCH amd64
RUN --mount=type=cache,target=/root/.ccache \
OLLAMA_SKIP_STATIC_GENERATE=1 OLLAMA_SKIP_CPU_GENERATE=1 bash gen_linux.sh
OLLAMA_SKIP_STATIC_GENERATE=1 \
OLLAMA_SKIP_CPU_GENERATE=1 \
CMAKE_CUDA_ARCHITECTURES="50;52;53;60;61;62;70;72;75;80;86" \
CUDA_VARIANT="_v11" \
bash gen_linux.sh
FROM --platform=linux/arm64 nvidia/cuda:$CUDA_VERSION-devel-rockylinux8 AS cuda-build-server-arm64
FROM --platform=linux/amd64 nvidia/cuda:$CUDA_VERSION_12-devel-centos7 AS cuda-12-build-amd64
ARG CMAKE_VERSION
COPY ./scripts/rh_linux_deps.sh /
RUN CMAKE_VERSION=${CMAKE_VERSION} sh /rh_linux_deps.sh
ENV PATH /opt/rh/devtoolset-10/root/usr/bin:$PATH
COPY --from=llm-code / /go/src/github.com/ollama/ollama/
WORKDIR /go/src/github.com/ollama/ollama/llm/generate
ARG CGO_CFLAGS
ENV GOARCH amd64
RUN --mount=type=cache,target=/root/.ccache \
OLLAMA_SKIP_STATIC_GENERATE=1 \
OLLAMA_SKIP_CPU_GENERATE=1 \
CMAKE_CUDA_ARCHITECTURES="60;61;62;70;72;75;80;86;87;89;90;90a" \
CUDA_VARIANT="_v12" \
bash gen_linux.sh
FROM --platform=linux/arm64 nvidia/cuda:$CUDA_VERSION_11-devel-rockylinux8 AS cuda-11-build-server-arm64
ARG CMAKE_VERSION
COPY ./scripts/rh_linux_deps.sh /
RUN CMAKE_VERSION=${CMAKE_VERSION} sh /rh_linux_deps.sh
@@ -34,7 +54,8 @@ COPY --from=llm-code / /go/src/github.com/ollama/ollama/
WORKDIR /go/src/github.com/ollama/ollama/llm/generate
ARG CGO_CFLAGS
ENV GOARCH arm64
RUN OLLAMA_SKIP_STATIC_GENERATE=1 OLLAMA_SKIP_CPU_GENERATE=1 bash gen_linux.sh
RUN --mount=type=cache,target=/root/.ccache \
OLLAMA_SKIP_STATIC_GENERATE=1 OLLAMA_SKIP_CPU_GENERATE=1 bash gen_linux.sh
FROM --platform=linux/arm64 nvcr.io/nvidia/l4t-jetpack:${JETPACK_6} AS cuda-build-jetpack6-arm64
ARG CMAKE_VERSION
@@ -139,8 +160,10 @@ COPY . .
COPY --from=static-build-amd64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
COPY --from=cpu_avx-build-amd64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
COPY --from=cpu_avx2-build-amd64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
COPY --from=cuda-build-amd64 /go/src/github.com/ollama/ollama/dist/ dist/
COPY --from=cuda-build-amd64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
COPY --from=cuda-11-build-amd64 /go/src/github.com/ollama/ollama/dist/ dist/
COPY --from=cuda-11-build-amd64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
COPY --from=cuda-12-build-amd64 /go/src/github.com/ollama/ollama/dist/ dist/
COPY --from=cuda-12-build-amd64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
COPY --from=rocm-build-amd64 /go/src/github.com/ollama/ollama/dist/ dist/
COPY --from=rocm-build-amd64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
ARG GOFLAGS
@@ -155,8 +178,8 @@ ARG GOLANG_VERSION
WORKDIR /go/src/github.com/ollama/ollama
COPY . .
COPY --from=static-build-arm64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
COPY --from=cuda-build-server-arm64 /go/src/github.com/ollama/ollama/dist/ dist/
COPY --from=cuda-build-server-arm64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
COPY --from=cuda-11-build-server-arm64 /go/src/github.com/ollama/ollama/dist/ dist/
COPY --from=cuda-11-build-server-arm64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
## arm binary += 381M
COPY --from=cuda-build-jetpack6-arm64 /go/src/github.com/ollama/ollama/llm/build/linux/ llm/build/linux/
COPY --from=cuda-build-jetpack6-arm64 /go/src/github.com/ollama/ollama/dist/ dist/