[gst-devel] Fun crash in opt

[in7y118 at public.uni-hamburg.de]

It should be known by now that opt is broken ;)
It's the common problem of not properly refcounting inside his code but not 
commenting it...

Feel free to fix it. Or just use basicomega like I do. As always valgrind 2.0 
is your friend...

Benjamin


Quoting iain <iain at prettypeople.org>:

> I know I should bugzilla this, but ...
> 
> Pipeline is {filesrc ! typefind ! spider ! custom-sink}
> The pipeline of {filesrc ! mad ! custom-sink} works fine.
> When I call g_object_unref on the thread I get this crash
> 
> Program received signal SIGSEGV, Segmentation fault.
> g_slist_remove (list=0x68637320, data=0x85b2bb0) at gslist.c:370
> 370           if (tmp->data == data)
> (gdb) print tmp
> $1 = (struct _GSList *) 0x68637320
> (gdb) print tmp->data
> Cannot access memory at address 0x68637320
> (gdb) print data
> $2 = (const void *) 0x85b2bb0
> (gdb) bt
> #0  g_slist_remove (list=0x68637320, data=0x85b2bb0) at gslist.c:370
> #1  0x40fb9a59 in remove_from_chain (chain=0x85f8c08, group=0x85b2bb0)
>     at gstoptimalscheduler.c:476
> #2  0x40fb9f2d in chain_recursively_migrate_group (chain=0x444021e8,
>     group=0x85b2bb0) at gstoptimalscheduler.c:574
> #3  0x40fb9f62 in chain_recursively_migrate_group (chain=0x444021e8,
>     group=0x85b2b00) at gstoptimalscheduler.c:584
> #4  0x40fbcf91 in gst_opt_scheduler_pad_unlink (sched=0x85ad070,
>     srcpad=0x85d7ed8, sinkpad=0x82c53f0) at gstoptimalscheduler.c:1967
> #5  0x401a5600 in gst_scheduler_pad_unlink (sched=0x85ad070,
> srcpad=0x85d7ed8,
>     sinkpad=0x82c53f0) at gstscheduler.c:186
> #6  0x40182dd3 in gst_bin_unset_element_sched (element=0x85b7ec0,
>     sched=0x85ad070) at gstbin.c:390
> #7  0x40c03a18 in g_list_foreach (list=0x85b2bb0,
>     func=0x40182b70 <gst_bin_unset_element_sched>, user_data=0x85ad070)
>     at glist.c:669
> #8  0x40182f31 in gst_bin_unset_element_sched (element=0x85b58c8,
>     sched=0x85ad070) at gstbin.c:362
> #9  0x401833fe in gst_bin_remove_func (bin=0x82cf3c0, element=0x85b58c8)
>     at gstbin.c:516
> #10 0x40183612 in gst_bin_remove (bin=0x82cf3c0, element=0x85b58c8)
>     at gstbin.c:570
> #11 0x401840a7 in gst_bin_dispose (object=0x82cf3c0) at gstbin.c:794
> #12 0x401ad8e6 in gst_thread_dispose (object=0x82cf3c0) at gstthread.
> c:209
> #13 0x40bb0f45 in g_object_last_unref (object=0x82cf3c0) at gobject.
> c:550
> #14 0x4013d0b1 in finalize (object=0x85acf18) at marlin-pipeline.c:83
> #15 0x4013a151 in finalize (object=0x85acf18) at marlin-load-pipeline.
> c:79
> 
> marlin-pipeline.c:83 is the g_object_unref call.
> Any ideas for debuggin it further?
> iain
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: Perforce Software.
> Perforce is the Fast Software Configuration Management System offering
> advanced branching capabilities and atomic changes on 50+ platforms.
> Free Eval! http://www.perforce.com/perforce/loadprog.html
> _______________________________________________
> gstreamer-devel mailing list
> gstreamer-devel at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/gstreamer-devel
>