#exult@irc.freenode.net logs for 12 Jun 2016 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage


[01:21:06] --> DominusExult has joined #exult
[01:21:06] --- ChanServ gives channel operator status to DominusExult
[01:24:33] <-- Dominus has left IRC (Ping timeout: 246 seconds)
[01:24:33] --- DominusExult is now known as Dominus
[08:42:49] <Dominus> same thing happens when I move mtgm.mid up in the index (to number 25, for example)
[08:51:45] <Dominus> The problem should also be there when you use a real MT32 device on other OS...
[08:53:15] --> GitHub has joined #exult
[08:53:15] <GitHub> [exult] DominusExult pushed 2 new commits to master: https://git.io/von2f
[08:53:15] <GitHub> exult/master 2accb88 Alagner: bug 1936/added sound for "too heavy" message
[08:53:15] <GitHub> exult/master e50cfdf DominusExult: Circumventing the crash problem with a real MT32 device when mtgm.mid is not at index #26 in exult.flx.
[08:53:15] <-- GitHub has left #exult
[12:57:34] --> DominusExult has joined #exult
[12:57:34] --- ChanServ gives channel operator status to DominusExult
[12:58:46] <-- Dominus has left IRC (Ping timeout: 244 seconds)
[12:58:47] --- DominusExult is now known as Dominus
[14:30:11] <Marzo> Dominus: can you undo your workaround locally, compile clean with heavy GDB debug, configure Exult to use point scaler at 1x scale and no aspect correction, run Exult through valgrind with the command "valgrind --tool=memcheck --leak-check=full --show-leak-kinds=all --track-origins=yes exult --fov --nomenu" and post the log?
[14:30:46] <Marzo> (putting in point 1x and no aspect correction is to make Exult run faster, to compensate a bit for the performance hit of valgrind)
[14:35:31] <Marzo> (oh, and you definitely want to redirect the log into a file)
[15:00:11] <Dominus> Marzo: will do in a couple of minutes. Unfortunately I read a couple of notes that valgrind and OS X > 10.7 don't work well together but let's see first...
[15:16:58] <Dominus> Marzo: http://pastebin.com/vg2mbrKe
[15:43:46] <Dominus> Marzo AFAICT without having another OS, you should be able to reproduce with a midi driver and conversion set to mt32
[17:04:03] <Marzo> Dominus: sadly, the log was useless
[17:04:14] <Marzo> I will have to try and see if the crash also happens on Linux
[17:52:15] <Dominus> I feared that ;(