Skip to content

🧪 Fixup test marks #521

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

Open
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

🧪 Fixup test marks #521

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
6f2cf37
:art: sort markers, add more
joerunde Oct 10, 2025
b262b04
:fire: remove skip_global_cleanup
joerunde Oct 10, 2025
1f48d8b
:fire: rip out unused cleanup fixture
joerunde Oct 10, 2025
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
16 changes: 10 additions & 6 deletions pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -137,19 +137,23 @@ asyncio_default_fixture_loop_scope = "function"

# --8<-- [start:test-markers-definition]
markers = [
"skip_global_cleanup",
"e2e: Tests using end-to-end engine spin-up",
"basic: Basic correctness tests",
"cb: Continuous batching tests",
"cpu: Tests using CPU (i.e. eager) backend",
"compat: backward compatibility tests",
"spyre: Tests using Spyre hardware backend",
"compiler_support_32k: Tests for 32k compiler support",
"cpu: Tests using CPU (i.e. eager) backend",
"decoder: Tests for decoder models",
"e2e: Tests using end-to-end engine spin-up",
"embedding: Tests for embedding models",
"quantized: Tests for quantized models",
"full_model: Tests that use full-sized granite models",
"multi: Tests that require >1 cards",
"other_e2e: ??? Tests for sampling params that take a long time",
"precompilation: Tests the handling of precompiled model catalogs",
"quantized: Tests for quantized models",
"scoring: Tests for scoring and reranking models",
"spyre: Tests using Spyre hardware backend",
"utils: Tests for utility functions",
"worker: Tests for worker logic",
"worker: Tests for worker logic"
]
# --8<-- [end:test-markers-definition]

Expand Down
10 changes: 0 additions & 10 deletions tests/conftest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -176,16 +176,6 @@ def init_test_http_connection():
global_http_connection.reuse_client = False


@pytest.fixture()
def should_do_global_cleanup_after_test(request) -> bool:
"""Allow subdirectories to skip global cleanup by overriding this fixture.
This can provide a ~10x speedup for non-GPU unit tests since they don't need
to initialize torch.
"""

return not request.node.get_closest_marker("skip_global_cleanup")


@pytest.fixture(autouse=True)
def cleanup_fixture(should_do_global_cleanup_after_test: bool):
yield
Copy link
Collaborator

Choose a reason for hiding this comment

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

Shouldn't this be removed as well?

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

yeah- we require --forked anyway for any tests where we want to initialize something new on the spyre cards. As far as I know we've never relied on this fixture

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

oh interesting, though some of this cleanup may have been doing something for our cpu tests, even though most of them re-use models between tests 🤔

Will need to inspect further next week

Expand Down
Loading