Bug 455573 - Valgrind crashes when running KStars test
Summary: Valgrind crashes when running KStars test
Status: RESOLVED NOT A BUG
Alias: None
Product: valgrind
Classification: Developer tools
Component: memcheck (show other bugs)
Version: 3.17.0
Platform: Ubuntu Linux
: NOR crash
Target Milestone: ---
Assignee: Julian Seward
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2022-06-18 23:30 UTC by Hy
Modified: 2022-06-20 05:20 UTC (History)
1 user (show)

See Also:
Latest Commit:
Version Fixed In:
Sentry Crash Report:


Attachments
this is the stderr/stdout for the crash, which seems to include a stack trace (349.50 KB, text/plain)
2022-06-18 23:30 UTC, Hy
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Hy 2022-06-18 23:30:22 UTC
Created attachment 149910 [details]
this is the stderr/stdout for the crash, which seems to include a stack trace

SUMMARY

When I try to run valgrind to find memory issues in a test for KStars (a KDE planetarium program) I get a crash.
I believe I'm using the latest valgrind release for Ubuntu 20.04
The test runs several parallel threads to test thread-safeness of some KStars code.
I've attached the output.

STEPS TO REPRODUCE
1.  You'd need the compiled KStars test, if you'd like to reproduce, let me know and I can help you compile it.
2. 
3. 

OBSERVED RESULT
crash

EXPECTED RESULT
no crash

SOFTWARE/OS VERSIONS
Linux/KDE Plasma:  Just tried on Ubuntu 20.04


ADDITIONAL INFORMATION
Comment 1 Paul Floyd 2022-06-19 06:36:11 UTC
I don't see any Valgrind errors. There's a guest 'abort()' at the end.

All of the test framework noise makes it very hard to read.

Can you run the test on its own and attach the Valgrind-only output?

You may also need a debug build (or debuginfo) for Qt to see the locations of this part of the stack

==72018==    by 0x4885AA1: ??? (in /usr/lib/x86_64-linux-gnu/libQt5Test.so.5.12.8)
==72018==    by 0x74759D1: ??? (in /usr/lib/x86_64-linux-gnu/libQt5Core.so.5.12.8)
Comment 2 Hy 2022-06-20 05:20:54 UTC
Thanks you Paul. I think you are right. I confused the stack-trace output there (which I guess was a kill timeout) with a stack-trace from ValGrind. I'll resolve this, look further, and get back if it turns out there really is an issue.

Appreciate your help,
Hy