-
Notifications
You must be signed in to change notification settings - Fork 17.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
runtime: access violation in gentraceback on windows/386 #50144
Comments
CC @bufflig @alexbrainman @cherrymui At first I thought this might have something to do with #49681, but some of the failures are much older. |
Yeah, this looks different from #49681 . It looks like it panics again while printing traceback from a fault. |
Is the fact that is prints "Test killed: ran too long" relevant? I mean is this something that happens when we try to kill the test, or does the test end up in an infinite loop? |
Not clear to me. It may be that this failure mode tends to occur when the signal terminating the test arrives on top of another signal or vice-versa — that would at least explain the presence of the The That suggests that perhaps the binary did deadlock after printing the |
(CC @golang/windows @golang/runtime) |
It looks like the traceback is called with PC=SP=LR=0. It faults when it tries to read from the SP. Maybe the signal context is wrong? I search for exception 0x200c7 but couldn't find anything useful. So perhaps the exception code is also wrong? Perhaps the exception handler shouldn't be called in the first place? I don't know what caused it to be called (with seemingly wrong arguments). |
2022-08-08T14:11:19-6e9925c/windows-386-2012:
|
Found new dashboard test flakes for:
2022-09-02 19:09 windows-arm64-10 go@55ca6a20 runtime/trace.TestTraceCPUProfile (log)
2022-09-05 08:08 windows-amd64-longtest go@af7f4176 runtime/pprof.TestMathBigDivide (log)
2022-09-27 15:57 windows-386-2012 go@e978ef36 runtime (log)
|
Last watchflake report in previous comment seems to be conflating two different bugs. Both Line 165 in e978ef3
I don't know how |
Change https://go.dev/cl/444618 mentions this issue: |
Found new dashboard test flakes for:
2022-10-24 17:28 windows-386-2008 go@0cc20ecc runtime (log)
|
Hmm. Is it possible that the up-stack fix @cherrymui suggested in that code review actually happened at some point? |
greplogs --dashboard -md -l -e '(?ms)^fatal error: unexpected signal during runtime execution\n\[signal 0xc0000005.*^runtime\.gentraceback'
2021-12-13T18:45:54-2580d0e/windows-386-2012
2021-10-29T17:13:13-645d078/windows-386-2008
2020-12-22T20:00:00-c9fb4eb/windows-386-2008
2020-03-25T14:19:25-91b8b13/windows-386-2008
The text was updated successfully, but these errors were encountered: