#pentagram@irc.freenode.net logs for 9 Mar 2005 (GMT)

Archive Today Yesterday Tomorrow
Pentagram homepage


[03:15:32] --> Zain has joined #pentagram
[03:15:35] <Zain> hello
[03:15:46] <Zain> how did the scale z scale y commands work?
[03:15:49] <Zain> for the ninis?
[03:16:03] <Zain> Inis
[03:16:57] <Zain> hello?
[03:26:26] <Zain> how do i make so that i dont have that 320x200 resolution in inis?
[04:01:45] <-- oaQ^ has left IRC (leguin.freenode.net irc.freenode.net)
[04:01:45] <-- Darke has left IRC (leguin.freenode.net irc.freenode.net)
[04:01:51] --> oaQ^ has joined #pentagram
[04:01:51] --> Darke has joined #pentagram
[04:04:47] <watt> Woot. Looks like I got my math right in the Pool class. Now I just have to write a MemManager and Allocators.
[05:28:49] <Zain> nevermind fixed it ;)
[05:30:00] <-- Zain has left IRC ()
[05:44:33] <watt> I
[05:45:46] <watt> I'm not going to bother with a MemManager yet. It makes more sense to just have the allocator in the ObjectManager if only Objects are pooled.
[06:01:57] <watt> ah seg fault.. yup.
[06:06:44] <watt> well.. that's gonna be fairly hard to track down.. so, night.
[06:14:24] --> Colourless has joined #Pentagram
[06:14:24] --- ChanServ gives channel operator status to Colourless
[06:20:10] <Colourless> hiya
[06:32:45] <watt> hi..
[06:33:05] <watt> I was wrong.. It wasn't too hard to find the seg fault.. bah
[06:33:11] <Colourless> :-)
[06:34:23] <Colourless> what was the problem?
[06:34:55] <watt> a->allocate(size) instead of ptr = a->allocate(size)
[06:35:07] <watt> leaving ptr = 0 is bad.
[06:35:09] <Colourless> yes, that would do it :-)
[06:37:46] <watt> hmm.. although now assert(desktopGump->getObjId() == 256) is failing.... crap
[06:39:32] <Colourless> that would be a small problem :-)
[06:48:53] <watt> desktop gump and scaler gump got the same address :-(
[06:49:02] <watt> makes sense
[06:51:13] <Colourless> why?
[06:51:27] <Colourless> you allocated busted? not marking blocks used?
[06:51:53] <watt> well, I must have added the location to the freeNodes stack twice at the end
[06:59:17] <watt> bad order on free for the freeNodes stack increment, I believe
[07:02:30] --> Chetic has joined #pentagram
[07:03:24] <watt> seems to be working.. I'll test it and clean up a bit tomorrow.
[07:03:47] <watt> night
[07:04:43] <Colourless> night
[07:09:33] <-- Chetic_ has left IRC (Read error: 60 (Operation timed out))
[07:14:50] <-- wjp has left IRC (leguin.freenode.net irc.freenode.net)
[07:18:07] --> wjp has joined #pentagram
[10:12:33] --> Ember has joined #pentagram
[10:15:38] --- ChanServ gives channel operator status to wjp
[10:31:17] <-- Darke has left IRC (Read error: 110 (Connection timed out))
[12:04:48] <-- Lightkey has left IRC (Read error: 60 (Operation timed out))
[12:13:09] --> Lightkey has joined #Pentagram
[14:08:29] <-- Chetic has left IRC (Read error: 60 (Operation timed out))
[14:16:24] <-- Kirben has left IRC ("System Meltdown")
[15:44:32] --> megawatt has joined #pentagram
[15:48:10] <megawatt> blah.. ok, enough NickServ stuff for now...
[15:49:47] <megawatt> anyway... current plan I think is to make Allocator become FlatPoolAllocator and Pool to become FlatPool (better names for those are welcome)
[15:50:56] <megawatt> Allocator become a base class, which would only malloc and free. Not sure what the base Pool should do yet. (Simply pool any size?)
[15:52:23] <megawatt> a MemManager would hold the Allocators, but would also set the Allocator for Object (a static reference held in Object for speed purposes)
[15:52:42] <megawatt> anything anyone would like to add or change to that list?
[15:55:07] <megawatt> actually, methinks Pool would be best as a pure virtual class.
[15:58:38] <Colourless> all up to you :-)
[15:58:43] <Colourless> it's your little project
[16:28:43] <megawatt> sub-project. I'd consider it to be part of Pentagram, and therefore should be able to be scrutinized by any dev. :-)
[16:30:06] <Colourless> damn :-)
[16:33:39] <megawatt> Oh yeah, I need to include a process in there to remove pools from the allocator if more than one pool is empty.
[16:33:53] <megawatt> a very low priority process....
[16:36:18] <megawatt> Did we ever do anything about process rings?
[16:36:46] <Colourless> no
[16:36:48] <megawatt> Mentioned it way back... http://www.math.leidenuniv.nl/~wpalenst/pentagramlog.php?log=16May2004
[16:37:52] <Colourless> ring was only going to be something if we were going to write processes for pentagram in usecode
[16:39:13] <Colourless> no 'really' much point in them if coded in C++
[16:44:32] <megawatt> I see
[17:06:11] --> Fingolfin has joined #pentagram
[17:06:11] --- ChanServ gives channel operator status to Fingolfin
[17:09:09] <Colourless> hi max
[17:23:40] <Fingolfin> hi ryan, willem, all
[17:26:50] <megawatt> hiya
[17:53:27] --> Chetic has joined #pentagram
[18:10:03] <wjp> hi Max
[18:24:57] <-- Fingolfin has left IRC ("42")
[20:52:16] <Colourless> later
[20:52:18] <-- Colourless has left IRC ("casts improved invisibility")
[22:20:26] <-- watt has left IRC (Read error: 104 (Connection reset by peer))
[22:26:08] --> Kirben has joined #pentagram
[22:26:08] --- ChanServ gives channel operator status to Kirben
[23:35:16] <-- megawatt has left IRC ("Chatzilla 0.9.67 [Firefox 1.0/20041107]")