Skip to content

[JAX] Fix L0_jax_unittest docs example test to enforce single-GPU #3059

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
jberchtold-nvidia merged 2 commits into from
May 29, 2026
+1 −1
Merged

[JAX] Fix L0_jax_unittest docs example test to enforce single-GPU #3059

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
14f10a2
Update test.sh
jberchtold-nvidia May 28, 2026
1220b35
Merge branch 'main' into jberchtold/fix-l0-docs-test
tdophung May 29, 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
2 changes: 1 addition & 1 deletion qa/L0_jax_unittest/test.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,7 +45,7 @@ NVTE_JAX_CUSTOM_CALLS="false" python3 -m pytest -c $TE_PATH/tests/jax/pytest.ini
# Exercise the docs/examples/jax tutorials. The multi-GPU tests are
# skipped at runtime when fewer than 4 devices are visible, so this is safe on
# single-GPU runners.
python3 -m pytest -c $TE_PATH/tests/jax/pytest.ini -v --junitxml=$XML_LOG_DIR/pytest_docs_examples_jax.xml $TE_PATH/docs/examples/jax/ || test_fail "docs/examples/jax"
CUDA_VISIBLE_DEVICES=0 python3 -m pytest -c $TE_PATH/tests/jax/pytest.ini -v --junitxml=$XML_LOG_DIR/pytest_docs_examples_jax.xml $TE_PATH/docs/examples/jax/ || test_fail "docs/examples/jax"
Comment on lines 45 to +48
Copy link
Copy Markdown
Contributor

@greptile-apps greptile-apps Bot May 28, 2026

Choose a reason for hiding this comment

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

P2 The comment above still reflects the old rationale — that multi-GPU tests would self-skip when fewer than 4 devices are visible. Now that CUDA_VISIBLE_DEVICES=0 actively enforces single-GPU visibility, that explanation is stale and slightly misleading. Updating it prevents future readers from removing the env var under the mistaken belief that runtime skipping is a sufficient guard.

Suggested change
# Exercise the docs/examples/jax tutorials. The multi-GPU tests are
# skipped at runtime when fewer than 4 devices are visible, so this is safe on
# single-GPU runners.
python3 -m pytest -c $TE_PATH/tests/jax/pytest.ini -v --junitxml=$XML_LOG_DIR/pytest_docs_examples_jax.xml $TE_PATH/docs/examples/jax/ || test_fail "docs/examples/jax"
CUDA_VISIBLE_DEVICES=0 python3 -m pytest -c $TE_PATH/tests/jax/pytest.ini -v --junitxml=$XML_LOG_DIR/pytest_docs_examples_jax.xml $TE_PATH/docs/examples/jax/ || test_fail "docs/examples/jax"
# Exercise the docs/examples/jax tutorials. CUDA_VISIBLE_DEVICES=0 restricts
# JAX to a single GPU so that distributed tests are not attempted on
# multi-GPU runners.
CUDA_VISIBLE_DEVICES=0 python3 -m pytest -c $TE_PATH/tests/jax/pytest.ini -v --junitxml=$XML_LOG_DIR/pytest_docs_examples_jax.xml $TE_PATH/docs/examples/jax/ || test_fail "docs/examples/jax"

Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!


if [ $RET -ne 0 ]; then
echo "Error: some sub-tests failed: $FAILED_CASES"
Expand Down
Loading