Skip to content

dpl-workflow.sh: add env variable to use full MI100 serialization in online #14699

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
ehellbar merged 1 commit into from
Sep 29, 2025
Merged

dpl-workflow.sh: add env variable to use full MI100 serialization in online #14699

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
9 changes: 7 additions & 2 deletions prodtests/full-system-test/dpl-workflow.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -278,8 +278,13 @@ if [[ $GPUTYPE == "HIP" ]]; then
GPU_CONFIG+=" --environment \"ROCR_VISIBLE_DEVICES={timeslice${TIMESLICEOFFSET}}\""
fi
# serialization workaround for MI100 nodes: remove it again if the problem will be fixed in ROCm, then also remove the DISABLE_MI100_SERIALIZATION flag in the O2DPG parse script
[[ $EPNSYNCMODE == 1 ]] && [[ ${EPN_NODE_MI100:-} == "1" ]] && [[ ${DISABLE_MI100_SERIALIZATION:-0} != 1 ]] && GPU_CONFIG_KEY+="GPU_proc.amdMI100SerializationWorkaround=1;"
[[ -n ${OPTIMIZED_PARALLEL_ASYNC:-} ]] && [[ ${EPN_NODE_MI100:-} == "1" ]] && [[ ${DISABLE_MI100_SERIALIZATION:-0} != 1 ]] && GPU_CONFIG_KEY+="GPU_proc.serializeGPU=3;"
if [[ ${EPN_NODE_MI100:-} == "1" && ${DISABLE_MI100_SERIALIZATION:-0} != 1 ]]; then
if [[ -n ${OPTIMIZED_PARALLEL_ASYNC:-} ]] || [[ $EPNSYNCMODE == 1 && ${FULL_MI100_SERIALIZATION:-0} == 1 ]]; then
GPU_CONFIG_KEY+="GPU_proc.serializeGPU=3;"
elif [[ $EPNSYNCMODE == 1 ]]; then
GPU_CONFIG_KEY+="GPU_proc.amdMI100SerializationWorkaround=1;"
fi
fi
#export HSA_TOOLS_LIB=/opt/rocm/lib/librocm-debug-agent.so.2
else
GPU_CONFIG_KEY+="GPU_proc.deviceNum=-2;"
Expand Down