845513 - TSan: Thread data race in sctp_notify_assoc_change() vs. user_connect()

Status: RESOLVED FIXED

(Core :: WebRTC, defect, P2)

Description

[Christoph Diehl [:posidron]]

Attachment: trace

During initial tests with ThreadSanitizer (LLVM version), we get a data race reported as described in the attached log. Trace was created on mozilla-central with changeset 122820:c233837cce08.

According to the TSan devs, most of the reported traces should be real data races, even though they can be "benign". We need to determine if the race can/should be fixed, or put on the ignore list. Even for benign races, TSan devs suggest to fix them (second priority), as they can also cause problems [1].

[1] http://software.intel.com/en-us/blogs/2013/01/06/benign-data-races-what-could-possibly-go-wrong

Comment 1

[Randell Jesup [:jesup] (needinfo me)]

Michael: probable upstream issue with usrsctp

Comment 2

[Michael Tüxen]

Randell: OK. I see what happens. Let me think about if it is OK or how it has to be resolved. This is not userland specific, but general.

Best regards
Michael

Comment 3

[Michael Tüxen]

OK, the problem has been fixed upstream. The patch for the main sources is in
http://code.google.com/p/sctp-refimpl/source/detail?r=8427
and the patch for the userland stack is part of
http://code.google.com/p/sctp-refimpl/source/detail?r=8429
(the changes of /trunk/KERN/usrsctp/usrsctplib/netinet/sctputil.c are relevant).

Thank you very much for reporting!

Best regards
Michael

Comment 4

[Randell Jesup [:jesup] (needinfo me)]

Fixed by checkin for bug 855620