Closed ArthurSonzogni closed 3 years ago
@stephenmcgruer Do you expect this code to be correct? Or there is something obviously wrong I don't understand? I have seen you made many changes about stash in the recent past.
I'm afraid I no longer work on WPT and do not have time to investigate this problem. cc @web-platform-tests/wpt-core-team
Thanks for you help so far!
@ArthurSonzogni when this happens, are there multiple instances of wptserve
running this test at the same time, or does it happen without anything happen concurrently at all?
I am not sure. I don't see any process/thread with this name. I guess this happens within python3, so I would say yes, given the output:
ps -avt | grep /usr/bin/python3
0.0 /usr/bin/python3 -c from multiprocessing.resource_tracker import main;main(5)
0.0 /usr/bin/python3 -c from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=6, pipe_handle=10) --multiprocessing-fork
0.1 /usr/bin/python3 -c from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=6, pipe_handle=12) --multiprocessing-fork
0.1 /usr/bin/python3 -c from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=6, pipe_handle=14) --multiprocessing-fork
0.1 /usr/bin/python3 -c from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=6, pipe_handle=16) --multiprocessing-fork
0.1 /usr/bin/python3 -c from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=6, pipe_handle=18) --multiprocessing-fork
0.1 /usr/bin/python3 -c from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=6, pipe_handle=20) --multiprocessing-fork
0.1 /usr/bin/python3 -c from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=6, pipe_handle=22) --multiprocessing-fork
0.1 /usr/bin/python3 -c from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=6, pipe_handle=24) --multiprocessing-fork
0.1 /usr/bin/python3 -c from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=6, pipe_handle=26) --multiprocessing-fork
0.1 /usr/bin/python3 -c from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=6, pipe_handle=35) --multiprocessing-fork
This is mostly failing during stability test. I am using:
./wpt run --repeat 20 --binary=/usr/bin/google-chrome-unstable chrome ./html/cross-origin-embedder-policy/credentialless/cors-or-credentialless/redirect.tentative.html
and can reliably reproduce every 10 runs.
In bug: https://bugs.chromium.org/p/chromium/issues/detail?id=1198255
I have issue with the file: html/cross-origin-embedder-policy/credentialless/resources/dispatcher.py
It can be summarized by:
Theoricaly, every
stash.put()
is preceded bystash.take()
.stash.lock()
prevents concurrent access.Despite this, there are errors sometimes: https://github.com/web-platform-tests/wpt/pull/28575/checks?check_run_id=2391872015
This sounds like a bug in
stash.lock
.@stephenmcgruer Do you expect this code to be correct? Or there is something obviously wrong I don't understand? I have seen you made many changes about stash in the recent past.