Skip to content

Dockerfile - add sm_103 support for cuda12.9 docker image #737

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
WenqingLan1 wants to merge 12 commits into from
Closed

Dockerfile - add sm_103 support for cuda12.9 docker image #737

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
4358fa7
add support for sm 103
WenqingLan1 Aug 1, 2025
ccee488
remove duplicate installation of hpc-x and nccl
WenqingLan1 Sep 8, 2025
a0ec607
Merge branch 'main' of github.com:microsoft/superbenchmark into wenqi…
WenqingLan1 Sep 8, 2025
f364c22
revert dist inference change
WenqingLan1 Sep 8, 2025
ace33ba
fix cuda12.8 dockerfile
WenqingLan1 Sep 9, 2025
22fa394
Fix formatting and indentation in Makefile
WenqingLan1 Sep 10, 2025
33c30b6
merge main
WenqingLan1 Oct 7, 2025
31191f5
add hpcx back
WenqingLan1 Oct 7, 2025
9b9a6cc
add hpcx back
WenqingLan1 Oct 7, 2025
67f6af9
correct hpcx version
WenqingLan1 Oct 7, 2025
9206e20
update
WenqingLan1 Oct 7, 2025
bb389ad
update
WenqingLan1 Oct 7, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 1 addition & 12 deletions dockerfile/cuda12.8.dockerfile
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ FROM nvcr.io/nvidia/pytorch:25.02-py3
# - TransformerEngine 2.0
# Mellanox:
# - MOFED_VERSION; 5.4-rdmacore39.0
# - HPC-X: v2.21.0-CUDA12.x
# - HPC-X: v2.21
# Intel:
# - mlc: v3.11

Expand Down Expand Up @@ -122,17 +122,6 @@ RUN if [ "$TARGETARCH" = "amd64" ]; then \
echo "Skipping Intel MLC, AOCC and AMD Bliss installations for non-amd64 architecture: $TARGETARCH"; \
fi

# Install NCCL 2.25.1
RUN cd /tmp && \
git clone -b v2.25.1-1 https://github.com/NVIDIA/nccl.git && \
cd nccl && \
make -j ${NUM_MAKE_JOBS} src.build \
NVCC_GENCODE="-gencode=arch=compute_100,code=sm_100 \
-gencode=arch=compute_90,code=sm_90 \
-gencode=arch=compute_80,code=sm_80" && \
make install && \
rm -rf /tmp/nccl

# Install UCX with multi-threading support
ENV UCX_VERSION=1.18.0
RUN cd /tmp && \
Expand Down
11 changes: 0 additions & 11 deletions dockerfile/cuda12.9.dockerfile
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -123,17 +123,6 @@ RUN if [ "$TARGETARCH" = "amd64" ]; then \
echo "Skipping Intel MLC, AOCC and AMD Bliss installations for non-amd64 architecture: $TARGETARCH"; \
fi

# Install NCCL 2.26.3
RUN cd /tmp && \
git clone -b v2.26.3-1 https://github.com/NVIDIA/nccl.git && \
cd nccl && \
make -j ${NUM_MAKE_JOBS} src.build \
NVCC_GENCODE="-gencode=arch=compute_100,code=sm_100 \
-gencode=arch=compute_90,code=sm_90 \
-gencode=arch=compute_80,code=sm_80" && \
make install && \
rm -rf /tmp/nccl

# Install UCX with multi-threading support
ENV UCX_VERSION=1.18.0
RUN cd /tmp && \
Expand Down
Loading