PR #48618 Fix GPU memory teardown in CLI serve tests

Fix GPU memory teardown in CLI serve tests #48618

SunMarc merged 4 commits into main from fix-serve-gpu-teardown

[fix] Free GPU memory between CLI test classes to prevent meta-device…

577ad577

sergereview commented on 2026-09-08

[fix] Set expandable_segments via os.environ before CUDA init in CLI …

7527785f

[fix] Release InferenceThread closure references before blocking on q…

585e5ec1

[fix] Clean up: remove unneeded expandable_segments env var, add clea…

983f9a3a

sergereview commented on 2026-09-08

ydshieh requested a review from

SunMarc 32 days ago

SunMarc approved these changes on 2026-09-08

SunMarc merged 0a959de1 into main 32 days ago

SunMarc deleted the fix-serve-gpu-teardown branch 32 days ago

Reviewers

SunMarc

sergereview

Assignees

No one assigned

Labels

None yet

Milestone

No milestone