while debugging several instances of my application (which is built on libevent)
I got a SEGV in one instance  down in  event_tree_RB_REMOVE_COLOR...
This apparently came from timeout_process... in event.c:700

It seems these SEGV would happen soon after my app had been interrupted while in gdb by a sigpipe... such that the time for the TIMED event had already passed... while the app was interrupted in gdb, (I just cont'd to resume, soon followed by SEGV)

Is this a possible explanation?  Or am I grasping at straws... ?!
These are the first errors I've run into actually inside event code itself... although I know triggered by something I've done, or failed to do...

I haven't had any success duplicating the SEGV when running the apps normally, no gdb or purify...

I'm hoping this is just an artifact of running in gdb... otherwise I'm a little intimidated trying to debug these RB trees!?  Any suggestions or hints, if I do have to wade into the event code...
it seems either I'm deleting the memory for an event, but not doing an event_del??  or some such??

MDJ

Thoughts?
MDJ


Yahoo! Music Unlimited - Access over 1 million songs. Try it free.
_______________________________________________
Libevent-users mailing list
Libevent-users@monkey.org
http://monkey.org/mailman/listinfo/libevent-users

Reply via email to