[NVBUG-6327718][test] Unwaive test_disaggregated_videomme[nemotron_na… - #17031

Closed

aswinvisva wants to merge 13 commits into

NVIDIA:mainfrom

aswinvisva:avisva/unwaive-nvbug-6327718

Commits on Aug 19, 2026

[NVBUG-6327718][test] Stress-test videomme EPD 100x per run (rep160-259)
aswinvisva
committed
2dfc88b
[NVBUG-6327718][fix] Remove unparametrized videomme entries from qa test list
aswinvisva
committed
5a1db50
[NVBUG-6327718][test] Stress-test videomme EPD 100x per run (rep260-359, faulthandler.enable)
aswinvisva
committed
d3d25f5
[NVBUG-6327718][test] Write faulthandler dumps to a real file, not sys.stderr (rep360-459)
aswinvisva
committed
26258b6
[NVBUG-6327718][test] Stress-test videomme EPD 100x per run (rep460-559)
aswinvisva
committed
e9e3dc1
[NVBUG-6327718][test] Bump videomme stress to 500x per run (rep560-1059)
aswinvisva
committed
a203099
[NVBUG-6327718][test] Re-trigger CI to re-sample crash rate
aswinvisva
committed
ab6aa1a
[NVBUG-6327718][test] Instrument videomme test for full-coverage hang capture
aswinvisva
committed
c3130bb
[NVBUG-6327718][test] Drop noisy ZMQ/executor-loop trace env vars
aswinvisva
committed
7e2ca8e
[NVBUG-6327718][test] Add py-spy, RLIMIT_CORE, native-crash forensics
aswinvisva
committed
3b4daa6
[NVBUG-6327718][test] Detached reaper subprocess for artifact archival
aswinvisva
committed
a3a283f
[https://nvbugs/6327718][test] Drain E/PD thread pool before shutting down LLMs
aswinvisva
committed
eac49b0
[https://nvbugs/6327718][test] A/B arm 2: add max_workers=16 to nvfp4
aswinvisva
committed
5f5b31a