[Adium-devl] Ready for Adium 1.3.3?

David Smith catfish.man at gmail.com
Sun Jan 25 20:55:04 UTC 2009


Nevermind. I fail at being up to date on branch.

On Jan 25, 2009, at 12:13 PM, Evan Schoenberg wrote:

>
> On Jan 25, 2009, at 12:17 PM, Eric Richie wrote:
>
>> Right before it dies is always:
>>
>> 10:36:04: (Libpurple: gg) ** gg_watch_fd(0x193243d0);
>> 10:36:04: (Libpurple: gg) // gg_watch_fd() GG_STATE_CONNECTED
>> 10:36:04: (Libpurple: gg) ** gg_watch_fd_connected(0x193243d0,  
>> 0x1afaead0);
>> 10:36:04: (Libpurple: gg) ** gg_recv_packet(0x193243d0);
>> 10:36:04: (Libpurple: gg) // gg_recv_packet() header  
>> recv(11,0xbfffe170,8) = 8
>> 10:36:04: (Libpurple: gg) // gg_watch_fd_connected() received a  
>> notify reply
>>
>> I'll send you the full logs directly.
>
> Thanks for the log.  It's crashing at
> 0   libSystem.B.dylib             	0x93086e70 strlen + 16
> 1   libSystem.B.dylib             	0x9307ef30 vasprintf + 491
> 2   libglib                       	0x17f92f68 g_vasprintf + 87
> 3   libglib                       	0x17f77ed9 g_strdup_vprintf + 38
> 4 libpurple 0x180cec9b purple_debug_vargs + 206
> 5   libpurple                     	0x180ceec9 purple_debug_info + 92
> 6   libpurple                     	0x18125bd5 ggp_callback_recv + 785
> ggp_callback_recv + 785 is the 4th invocation of purple_debug_info  
> (per otx diassembly of libpurple.framework). That crash then should  
> be coming from this debug line:
> 		case GG_EVENT_NOTIFY60:
> 			purple_debug_info("gg",
> 				"notify60_pre: (%d) status=%d; version=%d; descr=%s\n",
> 				ev->event.notify60->uin, ev->event.notify60->status,
> 				ev->event.notify60->version,
> 				ev->event.notify60->descr ? ev->event.notify60->descr : "(null)");
>
> I think the only way that would ultimately crash in strlen(), since  
> it's protected against NULL, is if ev->event.notify60->descr points  
> to freed memory or isn't NULL-terminated.  Perusing the gadu-gadu  
> code, I don't see anywhere that this happens... but I don't know  
> anything about the (rather cumbersome) gadu-gadu code.
> This crash is therefore 'waiting on libpurple' - specifically,  
> waiting on one of the few people who care about gadu-gadu to take a  
> look - and we shouldn't block on it.
> Cheers,
> Evan
> _______________________________________________
> Adium-devl mailing list
> Adium-devl at adiumx.com
> http://adiumx.com/mailman/listinfo/adium-devl_adiumx.com





More information about the devel mailing list