| Summary: | Valgrind got SIGSEGV while doing synchronization of multi-threaded tree and copy audio buffers | ||
|---|---|---|---|
| Product: | [Developer tools] valgrind | Reporter: | Joël Krähemann <jkraehemann> |
| Component: | general | Assignee: | Julian Seward <jseward> |
| Status: | RESOLVED NOT A BUG | ||
| Severity: | crash | CC: | jkraehemann, tom |
| Priority: | NOR | ||
| Version First Reported In: | 3.11.0 | ||
| Target Milestone: | --- | ||
| Platform: | Debian unstable | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed/Implemented In: | ||
| Sentry Crash Report: | |||
| Attachments: |
linux kernel configuration
/etc/initramfs-tools/modules |
||
|
Description
Joël Krähemann
2016-04-26 06:07:29 UTC
https://github.com/gsequencer/gsequencer/tree/0.7.0 commit 2be65f0caadcba4f0e9bee744dd5fe8c8242e15e Author: Joël Krähemann <jkraehemann-guest@users.alioth.debian.org> Date: Tue Apr 26 08:08:05 2016 +0200 valgrind crash Created attachment 98602 [details]
linux kernel configuration
linux realtime configuration with firmware build enabled
Created attachment 98603 [details]
/etc/initramfs-tools/modules
This makes my system having nouveau drivers
Were there any warnings from memcheck before valgrind crashed? Although that crash appears as a valgrind internal error really it is most likely just a result of your malloc heap being corrupt given that it occurs in valgrind's malloc implementation. So the first thing to do is to make sure that any issues valgrind has already reported before that are fixed. Yes, there where many of them invalid read of size 2. This is probably since an mistake within a function causing buffer-overflow. ags_recycling_create_audio_signal_with_frame_count() was affected but it is fixed for now in GSequencer. Valgrind doesn't crash anymore. Thank you, you're doing a great job. |