Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
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
26 changes: 26 additions & 0 deletions .github/workflows/bench.yml
Original file line number Diff line number Diff line change
Expand Up @@ -92,6 +92,31 @@ jobs:
(cd pr && . ./mfc.sh load -c ${{ matrix.flag }} -m g)
(cd pr && ./mfc.sh bench_diff ../master/bench-${{ matrix.device }}.yaml ../pr/bench-${{ matrix.device }}.yaml)

- name: Process Nsight Profiling Report
run: |
if [ -f "pr/report.nsys-rep" ]; then
echo "=== Nsight Profiling Summary ==="
echo "Master"
(cd master && nsys stats --report nvtx_sum report.nsys-rep)
echo "Pr"
(cd pr && nsys stats --report nvtx_sum report.nsys-rep)

echo "=== CUDA API CALLS ==="
echo "Master"
(cd master && nsys stats --report cuda_api_sum --format table report.nsys-rep | head -100)
echo "Pr"
(cd pr && nsys stats --report cuda_api_sum --format table report.nsys-rep | head -100)

echo "=== GPU KERNELS ==="
echo "Master"
(cd master && nsys stats --report cuda_gpu_kern_sum --format table report.nsys-rep | head -100)
echo "Pr"
(cd pr && nsys stats --report cuda_gpu_kern_sum --format table report.nsys-rep | head -100)

else
echo "No Nsight report found, skipping profiling analysis"
fi

- name: Print Logs
if: always()
run: |
Expand All @@ -107,5 +132,6 @@ jobs:
path: |
pr/bench-${{ matrix.device }}.*
pr/build/benchmarks/*
pr/report.nsys-rep
master/bench-${{ matrix.device }}.*
master/build/benchmarks/*
4 changes: 2 additions & 2 deletions .github/workflows/phoenix/bench.sh
Original file line number Diff line number Diff line change
Expand Up @@ -17,9 +17,9 @@ mkdir -p $currentdir
export TMPDIR=$currentdir

if [ "$device" = "gpu" ]; then
./mfc.sh bench --mem 12 -j $(nproc) -o "$job_slug.yaml" -- -c phoenix-bench $device_opts -n $n_ranks
nsys profile -o report ./mfc.sh bench --mem 12 -j $(nproc) -o "$job_slug.yaml" -- -c phoenix-bench $device_opts -n $n_ranks
else
./mfc.sh bench --mem 1 -j $(nproc) -o "$job_slug.yaml" -- -c phoenix-bench $device_opts -n $n_ranks
nsys profile -o report ./mfc.sh bench --mem 1 -j $(nproc) -o "$job_slug.yaml" -- -c phoenix-bench $device_opts -n $n_ranks
fi

sleep 10
Expand Down
Loading