#gemrb@irc.freenode.net logs for 22 Apr 2013 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[15:55:00] --> exultbot has joined #gemrb
[15:55:00] --- Topic for #gemrb is: GemRB 0.7.2 | http://gemrb.org | Something wrong? State your exact version and CHECK THE GEMRB LOG | Be wary of your thoughts for there are Illithid present: http://log.usecode.org/gemrblog.php | import pdb; pdb.set_trace()
[15:55:00] --- Topic for #gemrb set by lynxlynxlynx!~quassel@sourcemage/warlock/lynxlynxlynx at Tue Nov 20 22:36:44 2012
[16:01:45] --> Coriander has joined #gemrb
[16:04:41] <-- brada has left IRC (Quit: brada)
[16:05:15] --> brada has joined #gemrb
[16:17:00] <-- brada has left IRC (Quit: brada)
[16:24:08] --> rocket_hamster has joined #gemrb
[16:27:03] --> brada has joined #gemrb
[16:28:03] --- ermo^ is now known as ermo
[16:41:58] <-- brada has left IRC (Quit: brada)
[16:44:00] --> nutron has joined #gemrb
[17:01:05] --> fizzle has joined #gemrb
[17:11:11] --> Yoshimo has joined #gemrb
[17:11:56] <-- Yoshimo has left IRC (Client Quit)
[17:21:32] <-- rocket_hamster has left IRC (Remote host closed the connection)
[17:56:39] <Pepelka> [commit] lynxlynxlynx: added a gameflag for the slight difference in continuation handling https://github.com/gemrb/gemrb/commit/8776b90fddc16b606b29352d03593ef4ed15534f
[17:56:40] <Pepelka> [commit] lynxlynxlynx: fx_play_visual_effect: check for null target earlier (potential crash) https://github.com/gemrb/gemrb/commit/7206fda80df61a4dd01a09e271f3aedf09c1560c
[17:56:41] <Pepelka> [commit] lynxlynxlynx: MapControl::OnMouseDown: make doubleclicking possible https://github.com/gemrb/gemrb/commit/2bf3615020bcb81cc30106f1cb1ed0cd3cf771b3
[17:56:43] <Pepelka> [commit] lynxlynxlynx: Actor::PlayHitSound: moved the armor check safety, so it's not dead code https://github.com/gemrb/gemrb/commit/90999f6af0b736b13dd2dd1126420946ee1fa022
[17:56:43] <Pepelka> [commit] lynxlynxlynx: GameScript::SetPlayerSound: avoid out-of-bounds write https://github.com/gemrb/gemrb/commit/b643f53db667faa20af0be129414eca55f1028ec
[17:56:45] <Pepelka> [commit] lynxlynxlynx: GameScript::SetMusic: avoid out-of-bounds write https://github.com/gemrb/gemrb/commit/948cfe18793994e51124e45d6f45c015ffc66cc7
[18:16:46] --> Yoshimo has joined #gemrb
[18:40:55] <fizzle> I can pretty reliably crash gemrb in the circus
[18:41:26] <fizzle> looks similar to what we talked about yesterday
[18:41:34] <fizzle> dangling map references
[18:41:41] <fizzle> just the other way around today
[18:46:38] <Pepelka> [commit] lynxlynxlynx: GameControl: don't leak effects https://github.com/gemrb/gemrb/commit/282eab5fc56dfcb93da34ebf00fd59774819428c
[18:46:39] <Pepelka> [commit] lynxlynxlynx: GameScript: don't leak effects https://github.com/gemrb/gemrb/commit/7667e000a3d1c0488b0767298420af91f0aa190c
[18:46:40] <Pepelka> [commit] lynxlynxlynx: PutMapNotes: pst coordinates are stored as dwords, so don't cast one to a word https://github.com/gemrb/gemrb/commit/bd94fbf09a0b1c86b99bdb330088569e7271b39f
[18:46:41] <Pepelka> [commit] lynxlynxlynx: AREImporter::PutMapnotes: according to iesdp, pst color is a dword https://github.com/gemrb/gemrb/commit/b5787bce59e45699ddd9e66be0078ac4caf3a2ff
[18:46:42] <Pepelka> [commit] lynxlynxlynx: GameScript::UsedExit: removed logically dead code https://github.com/gemrb/gemrb/commit/327740d9095fedfbe6336eeec2938ec583947c90
[18:46:44] <Pepelka> [commit] lynxlynxlynx: GameScript::StartMusic: removed potential out-of-bounds access https://github.com/gemrb/gemrb/commit/8d0d66e79bafe6fad688151a729cb7dc2b6d33c8
[18:46:45] <Pepelka> [commit] lynxlynxlynx: Game::GetXPFromCR: avoid unlikely div-by-0 https://github.com/gemrb/gemrb/commit/22a0c872c15713d2c4a2977982ba4ac41343f54d
[18:47:36] <lynxlynxlynx> when i played through the last time, i got a memory corruption with kalah
[18:47:41] <lynxlynxlynx> but it wasn't reproducible
[18:47:46] <lynxlynxlynx> valgrind was silent too
[18:48:14] <fizzle> aerie's where I'm crashing
[18:48:30] <fizzle> her Map reference is obviously corrupt
[18:48:52] <lynxlynxlynx> don't remember seeing that, but this time i just ignored her anyway
[18:48:58] <fizzle> kalah probably has the same issue if I'd get past aerie for some reason
[18:49:27] <lynxlynxlynx> i'm not sure, this sounds like it would crash on entry
[18:49:34] <fizzle> did you get the problem after defeating kalah?
[18:51:55] --> brada has joined #gemrb
[18:52:08] <lynxlynxlynx> i think it was in the middle of the fight
[18:52:17] <gembot> build #554 of cmake g++-4.2 is complete: Failure [4failed] Build details are at http://buildbot.gemrb.org/builders/cmake%20g%2B%2B-4.2/builds/554 blamelist: Jaka Kranjc <lynxlupodian@users.sourceforge.net>
[18:52:18] <fuzzie> placepersistents looks kind of odd
[18:52:42] <fizzle> MoveGlobalsTo looks to be causing the crash for me, ie. the start of the ending cutscene
[18:52:48] <lynxlynxlynx> 403
[18:53:07] <fuzzie> 1e78f9e16d98bb82d902dfab180bd0e401333b7f in particular, by lynx
[18:54:08] <lynxlynxlynx> are you sure that code is still that way?
[18:54:11] <fuzzie> yes
[18:54:54] <lynxlynxlynx> it is indeed
[18:55:00] <fuzzie> i don't remember what we were fixing there
[18:55:42] <fuzzie> i do remember some loop like that
[18:55:48] <lynxlynxlynx> i started with an even worse implementation
[18:55:58] <lynxlynxlynx> we were skipping whole actors
[18:56:20] <lynxlynxlynx> took a whole release for anyone to notice a glaring bug
[18:56:28] <fizzle> I actually believe PurgeArea is to blame here
[18:56:44] <fuzzie> lynxlynxlynx: it just looks like it'll currently not work for i==last
[18:57:42] <fuzzie> fizzle: it seems clearly bad for the same reason as this code was, actually
[18:58:09] <fuzzie> oh, no, reverse loop
[18:59:42] <fuzzie> fizzle: so why do you think it's to blame? :)
[19:00:09] <fizzle> because it doesn't clear the reference to the map we're just about to remove from memory
[19:00:26] <fuzzie> should it?
[19:00:51] <fizzle> MoveGlobalsTo for example tries to reference it
[19:00:53] <-- brada has left IRC (Ping timeout: 240 seconds)
[19:01:04] <fuzzie> yes..
[19:01:06] <fizzle> and if the map is gone, poof
[19:01:10] <fuzzie> yes
[19:01:14] <fuzzie> because you purged an in-use area
[19:01:38] <fuzzie> oh, no, icky
[19:01:59] <fuzzie> ok, better question: does it fix it if you fix that?
[19:02:11] <fizzle> seems to, yes
[19:02:23] <fizzle> at least I haven't crashed the last two times
[19:02:26] <fuzzie> +1 then
[19:02:33] <fuzzie> i mean, if you're just setting map to null
[19:02:35] <fizzle> which is a better record than I had before
[19:02:40] <fizzle> yes
[19:03:02] <fizzle> well SetMap(NULL)
[19:03:23] <fizzle> obviously, we can't also clear Area in that case...
[19:03:34] <fuzzie> yes
[19:03:47] <fuzzie> which is why I asked
[19:04:00] <fizzle> I moved that part out
[19:04:12] <fuzzie> ok.
[19:04:41] <fuzzie> you clearly understand what's going on better than I do now, so great :)
[19:05:06] <fizzle> I wouldn't bet on that
[19:07:32] <-- Yoshimo has left IRC (Quit: Yoshimo)
[19:21:20] <Pepelka> [commit] fizzet: Map::DeleteActor/RemoveActor: also clear the area resref https://github.com/gemrb/gemrb/commit/a57efa9c8b61c925bae5385077fa79e26c9e2777
[19:21:21] <Pepelka> [commit] fizzet: Map::PurgeArea: clear the map reference for all actors https://github.com/gemrb/gemrb/commit/21819f86f77292585a7e6d82d558285334bb6e9c
[19:29:48] <gembot> build #555 of cmake g++-4.2 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/cmake%20g%2B%2B-4.2/builds/555
[19:34:53] --> brada has joined #gemrb
[19:34:56] --> edheldil_ has joined #gemrb
[19:36:01] --> vampi-the-frog has joined #gemrb
[19:37:18] <brada> i have no idea why the xcode build is failing
[19:37:31] <brada> the libraries/headers are in the same location as always
[19:37:40] <brada> and buildbot has readpermission
[19:38:42] <brada> unless that include line was add/changed recently
[19:39:04] <brada> but my fs isnt case sensitive so its not python.h vs Python.h problem
[19:46:42] <brada> so did a buildbot configuration change happen recently that broke this?
[19:47:09] <brada> because looking at the output the PWD path and the GEM_IOS_XXX paths are clearly wrong now
[19:47:16] <brada> but nothing changed on the machine
[19:48:50] <brada> ill jsut fix it on my end
[20:02:41] <-- edheldil has left IRC (Remote host closed the connection)
[20:40:07] <-- brada has left IRC (Quit: brada)
[20:41:14] <-- vampi-the-frog has left IRC (Quit: Leaving)
[20:51:07] <-- fizzle has left #gemrb
[21:30:15] <-- Cable_ has left IRC (Ping timeout: 276 seconds)
[21:42:26] --> Cable_ has joined #gemrb
[22:17:02] <-- Coriander has left IRC (*.net *.split)
[22:18:09] --> Coriander has joined #gemrb
[22:24:10] --> lynxlynxlynx_ has joined #gemrb
[22:24:10] <-- lynxlynxlynx has left IRC (Read error: Connection reset by peer)
[22:43:11] <-- lynxlynxlynx_ has left IRC (Remote host closed the connection)
[22:51:56] <-- edheldil_ has left IRC (Ping timeout: 255 seconds)
[23:11:55] <-- Coriander has left IRC (*.net *.split)
[23:12:46] --> Coriander has joined #gemrb