Skip to content

ci: mitigate OOM kills on large-model export tests #244

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
regisss merged 2 commits into from
Jul 22, 2026
Merged

ci: mitigate OOM kills on large-model export tests #244

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f7f957a
ci: mitigate OOM kills on large-model export tests
JakeStevens Jul 21, 2026
a205292
ci: drop disk-cleanup step that wiped the Python toolcache
JakeStevens Jul 21, 2026
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
18 changes: 18 additions & 0 deletions .github/workflows/test_models.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,9 @@ jobs:
# Custom job name, now shortened and cleaner
name: ${{ matrix.test-modeling }} (et=${{ matrix.executorch-version }}, py=${{ matrix.python-version }}, ${{ matrix.os }})
runs-on: ${{ matrix.os }}
# Fail cleanly on a hang instead of dragging the run for hours; large-model
# export can be slow (especially once paging to swap), so keep this generous.
timeout-minutes: 180
env:
MODEL_NAME: ${{ matrix.test-modeling }}
steps:
Expand All @@ -50,6 +53,21 @@ jobs:
uses: actions/setup-python@v2
with:
python-version: ${{ matrix.python-version }}
- name: Increase swap space
run: |
# Large models (e.g. gemma-3-4b, Phi-4-mini) push past the 16GB RAM on
# standard runners and get OOM-killed during load/quantization
# (exit 143). A big swapfile lets them page to disk instead of being killed.
echo "Memory before:"
free -h
sudo swapoff -a || true
sudo rm -f /mnt/swapfile || true
sudo fallocate -l 24G /mnt/swapfile || sudo dd if=/dev/zero of=/mnt/swapfile bs=1M count=24576
sudo chmod 600 /mnt/swapfile
sudo mkswap /mnt/swapfile
sudo swapon /mnt/swapfile
echo "Memory after:"
free -h
- name: Install dependencies for ExecuTorch
run: |
# Clean up cache to save space
Expand Down
18 changes: 18 additions & 0 deletions .github/workflows/test_models_nightly.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,6 +38,9 @@ jobs:

name: ${{ matrix.test-modeling }} (et=nightly, py=${{ matrix.python-version }}, ${{ matrix.os }})
runs-on: ${{ matrix.os }}
# Fail cleanly on a hang instead of dragging the run for hours; large-model
# export can be slow (especially once paging to swap), so keep this generous.
timeout-minutes: 180
env:
MODEL_NAME: ${{ matrix.test-modeling }}
steps:
Expand All @@ -46,6 +49,21 @@ jobs:
uses: actions/setup-python@v2
with:
python-version: ${{ matrix.python-version }}
- name: Increase swap space
run: |
# Large models (e.g. gemma-3-4b, Phi-4-mini) push past the 16GB RAM on
# standard runners and get OOM-killed during load/quantization
# (exit 143). A big swapfile lets them page to disk instead of being killed.
echo "Memory before:"
free -h
sudo swapoff -a || true
sudo rm -f /mnt/swapfile || true
sudo fallocate -l 24G /mnt/swapfile || sudo dd if=/dev/zero of=/mnt/swapfile bs=1M count=24576
sudo chmod 600 /mnt/swapfile
sudo mkswap /mnt/swapfile
sudo swapon /mnt/swapfile
echo "Memory after:"
free -h
- name: Install dependencies for ExecuTorch nightly
run: |
pip cache purge || true
Expand Down
Loading