Skip to content

Update amd-docker.Dockerfile #458

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

Merged
msaroufim merged 1 commit into from
Mar 5, 2026
Merged

Update amd-docker.Dockerfile #458

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
7 changes: 6 additions & 1 deletion docker/amd-docker.Dockerfile
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,9 @@ FROM ghcr.io/actions/actions-runner:latest

ENV CXX=clang++

ARG GPU_ARCH="gfx950"
ENV GPU_ARCH_LIST=$GPU_ARCH
Comment on lines +5 to +6

Copilot AI Mar 5, 2026

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ARG GPU_ARCH="gfx950" will include the quote characters in the default value (Dockerfile instructions don’t do shell-style unquoting). That can propagate to GPU_ARCH_LIST/GPU_ARCHS and break downstream parsing. Use an unquoted default (e.g., ARG GPU_ARCH=gfx950) and reference it via ${GPU_ARCH} when setting the ENV to avoid accidental literal characters.

Suggested change
ARG GPU_ARCH="gfx950"
ENV GPU_ARCH_LIST=$GPU_ARCH
ARG GPU_ARCH=gfx950
ENV GPU_ARCH_LIST=${GPU_ARCH}

Copilot uses AI. Check for mistakes.

RUN sudo apt-get update -y \
&& sudo apt-get install -y --no-install-recommends \
software-properties-common \
Expand Down Expand Up @@ -41,11 +44,13 @@ ENV ROCM_PATH=/opt/rocm

RUN sudo pip install --break-system-packages --no-cache-dir torch==2.10.0+rocm7.1 --index-url https://download.pytorch.org/whl/rocm7.1

ARG PREBUILD_KERNELS=1

RUN git clone --recursive https://github.com/ROCm/aiter.git \
&& cd aiter \
&& git checkout f3be04a12a0cfd6b5e2c7a94edc774f1bc24460d \
&& sudo pip install --break-system-packages -r requirements.txt \
&& sudo python3 setup.py develop
&& sudo MAX_JOBS=64 PREBUILD_KERNELS=$PREBUILD_KERNELS GPU_ARCHS=$GPU_ARCH_LIST python3 setup.py develop

Copilot AI Mar 5, 2026

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

sudo MAX_JOBS=64 PREBUILD_KERNELS=... GPU_ARCHS=... python3 ... won’t set those variables for the python3 process; the assignments are treated as arguments to sudo, which typically causes sudo: MAX_JOBS=64: command not found and breaks the image build. Set the environment on the command side (e.g., MAX_JOBS=64 ... sudo python3 ...) or use sudo env MAX_JOBS=64 PREBUILD_KERNELS=... GPU_ARCHS=... python3 ....

Suggested change
&& sudo MAX_JOBS=64 PREBUILD_KERNELS=$PREBUILD_KERNELS GPU_ARCHS=$GPU_ARCH_LIST python3 setup.py develop
&& sudo env MAX_JOBS=64 PREBUILD_KERNELS=$PREBUILD_KERNELS GPU_ARCHS=$GPU_ARCH_LIST python3 setup.py develop

Copilot uses AI. Check for mistakes.

RUN sudo mkdir -p /home/runner/aiter/aiter/jit/build \
&& sudo chown -R runner:runner /home/runner/aiter/aiter/jit/build
Expand Down
Loading