test: Suppress epoll_ctl data race#20218
Conversation
|
|
|
@hebasto I agree, but until someone figured out where the issue happens and how to fix it, we should suppress the intermittent failures, because they are hurting the project. |
hebasto
left a comment
hebasto
left a comment
There was a problem hiding this comment.
Approach ACK fa5c8f4b7027200d8f54aad83ab5c1bfc5a42c90.
It seems the data race in epoll_ctl could be false positive, and be caused by clang tsan instrumentation code (e.g., https://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20131007/190642.html). But not for sure, of course.
Also there is a possibility that this issue is/will be fixed in newer libevent versions.
maflcko
changed the title
maflcko
force-pushed
the
2010-testSuppWalletRace
branch
from
fa5c8f4 to
fa949b3
Compare
|
The following sections might be updated with supplementary metadata relevant to reviewers and maintainers. ConflictsReviewers, this pull request conflicts with the following ones:
If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first. |
|
Not sure why, but this doesn't work: https://cirrus-ci.com/task/5734386425397248?command=ci#L5451 |
Indeed ( |
Fixup of bitcoin#20218. Comments must start from the beginning of the line.
See #20745. |
d71e29e qa: Correct epoll_ctl data race suppression (Hennadii Stepanov) Pull request description: Fixup of #20218. Comments must start from the beginning of the line. ACKs for top commit: MarcoFalke: review ACK d71e29e Tree-SHA512: 4d8663ab505c347bcb62c2f118656e3343d5179825be0d1b86761ffdfdae1e7462002bf226a54dfc94be5885ce7f2633abaf70421ea35bf06eddad8e99fb9683
Summary: > It seems the data race in epoll_ctl could be false positive, and be caused by clang tsan instrumentation code (e.g., https://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20131007/190642.html). But not for sure, of course. This is a backport of [[bitcoin/bitcoin#20218 | core#20218]] and [[bitcoin/bitcoin#20745 | core#20745]] Note: I wasn't able to reproduce it today in a loop of 10 `ninja check-functional`, but I have seen this race in the past. I have a log file locally with a `ThreadSanitizer` data race involving `epoll_ctl`. Test Plan: With TSAN: ``` for i in `seq 10`; do ninja check-functional; done ``` Reviewers: #bitcoin_abc, Fabien Reviewed By: #bitcoin_abc, Fabien Differential Revision: https://reviews.bitcoinabc.org/D10722
3 participants
Happens intermittently: https://cirrus-ci.com/task/5462892373868544?command=ci#L5385