00:00:00 --- log: started forth/05.06.05
00:00:01 <Frek> so it can hold all swapfiles :)
00:00:38 <Frek> anyway, i'm seriously wonder what the hell most modern software does with all ram ?
00:22:07 <arke> With firefox and netscape my guess is that it needs all that because it does its own widget code
00:22:12 <arke> dont know about other things though
01:54:28 --- quit: Raystm2 (Read error: 110 (Connection timed out))
03:42:05 --- join: ICaMonkey (~chatzilla@80.122.33.114) joined #forth
04:20:48 --- join: Topaz (~top@sown-88.ecs.soton.ac.uk) joined #forth
04:23:05 --- join: Robert (~snofs@c-f778e055.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
05:20:02 --- quit: Robert ("leaving")
05:20:34 --- join: Robert (~snofs@c-f778e055.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
05:52:08 --- quit: ICaMonkey ("ChatZilla 0.9.61 [Mozilla rv:1.7.7/20050502]")
06:34:50 --- join: hrmpf (~obi@gw.mastmoen.no) joined #forth
06:37:12 <Robert> Hi.
06:37:40 <hrmpf> howdy
07:31:51 --- join: mark4 (~mark4@rrcs-24-242-160-169.sw.biz.rr.com) joined #forth
07:32:07 --- nick: mark4 -> I440r_
07:33:11 <Robert> Hi
08:55:37 --- quit: saon ("disconnecting from inet")
08:55:45 --- quit: saon|smgl ("following saon")
09:28:07 <crc> hi all
10:08:07 --- quit: Topaz ("Leaving")
10:08:55 <Robert> Hi, crc.
10:10:53 --- quit: hrmpf ("whatever u do, do it right!")
10:21:54 --- join: hrmpf (~obi@gw.mastmoen.no) joined #forth
10:29:09 --- join: madgarden (~madgarden@Kitchener-HSE-ppp3577323.sympatico.ca) joined #forth
10:34:44 <Robert> Hi
10:37:11 <madgarden> Hi.
10:43:49 --- join: tathi (~josh@pcp01375108pcs.milfrd01.pa.comcast.net) joined #forth
10:51:47 <arke> teh hies
10:52:46 <madgarden> heh ties
10:52:54 <hrmpf> ditto
10:52:58 <arke> :)
10:53:16 <Robert> Ted heils.
10:54:09 <madgarden> I know Ted. Good guy.
10:54:51 <Robert> Until he went nazi.
10:55:21 <arke> I know Ted. What an asshole.
10:56:01 <madgarden> Ted knows your asshole.
10:56:10 <madgarden> :O
10:57:22 --- join: qFox (~C00K13S@92pc222.sshunet.nl) joined #forth
10:58:16 <Robert> Hi, qFox 
10:58:55 <madgarden> Hike you, fox!! >:(
11:15:33 --- quit: tathi ("leaving")
11:27:36 <alexander_> hey
11:33:03 <Robert> Hi
11:36:20 <alexander_> I started hacking a forth IRC client last night
11:36:33 <alexander_> I was pretty impressed how quickly it came up, with isforth
11:40:53 <Robert> Neat. Will you be done soon?
11:42:34 <alexander_> I can connect and talk on channel, but I'm working on the windowing right now
11:42:42 <alexander_> so not for a while
11:43:09 <alexander_> the nifty thing I've got so far, are / commands are just calls to Forth words
11:43:09 <Robert> Still neat.
11:43:27 <alexander_> so if you define /join .. and the user types it, it'll just call the word
11:43:29 <Robert> Oh :)
11:43:41 <Robert> So /segfault will do bad things?
11:43:57 <Robert> : segfault 0 @ recurse ;
11:44:30 <arke> : segfault  0 0 ! ;
11:44:32 <arke> easier
11:44:34 <arke> :)
11:45:12 <alexander_> Robert: no, you have to name your words with a / 
11:45:15 <alexander_> so : /join ... ;
11:45:26 <alexander_> I thought about just letting you call whatever, but that seemed like a bad idea. ;)
11:45:43 <alexander_> : /forth quit ;
11:45:43 <alexander_> : /quit bye ;
11:46:02 <Robert> Ah.
11:54:07 --- join: Raystm2 (~vircuser@adsl-68-93-40-211.dsl.rcsntx.swbell.net) joined #forth
11:57:23 <Robert> Hi Ray
12:01:54 <alexander_> okay, I'll give someone a cookie if they come up with a cleaner version of a word than what i've got
12:03:22 <crc> jkl
12:03:22 <crc> ;'8
12:03:28 <crc> sorry
12:03:40 * crc dropped stuff on the keyboard
12:04:06 <crc> alexander_: are you connecting using raw syscalls?
12:04:41 <alexander_> crc: yah
12:05:05 <alexander_> with some assistance from I440r's socket calls
12:05:08 <crc> can I see the code?
12:05:34 * crc has written an irc client that relys on tcpclient to make the connection, I'd like to drop that dependency
12:06:06 <alexander_> ah
12:06:24 <alexander_> sure, you can see it, but it's not pretty. :)
12:06:29 <crc> that's ok
12:06:59 <alexander_> also, the polling is going to be different.. right now, there's one poll for the network, and a key?
12:28:21 <arke> alexander_: whats the word?
12:33:10 <arke> aaah
12:33:11 <arke> lalala
12:33:15 <arke> alexander_: I think I can
12:33:17 <arke> alexander_: :)
12:34:09 <alexander_> arke: awesome
12:34:30 <arke> and yes, it is quite ugly
12:34:51 <arke> :)
12:35:20 <alexander_> text processing just doesn't seem like it has much silver lining.
12:36:01 <arke> :)
12:36:05 <arke> please answer questions in /query :)
12:36:38 <alexander_> answerning
12:37:01 <arke> thanks :)
12:56:09 --- join: slava (~slava@CPE0080ad77a020-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
12:56:11 --- join: tathi (~josh@pcp01375108pcs.milfrd01.pa.comcast.net) joined #forth
12:56:21 <slava> hi tathi
12:56:30 <tathi> hi slava
12:56:53 <slava> so i've noticed on powerpc, there are no immedaite add/sub/mul instructions that set the overflow flag
12:56:56 <slava> is this correct?
12:57:09 <tathi> hmm
12:57:12 <slava> there is andi.
12:57:20 <tathi> I'm not sure actually.
12:57:23 <slava> but not addi. or mullwi. for instance
12:57:35 <slava> well, all this means is i have to remove an optimization :)
12:57:55 <slava> where literal values appearing before a word are put in an immediate operand if possible
12:58:19 <tathi> ah.
12:58:20 <slava> i need to always test the overflow flag to because of language semantics
12:59:01 <slava> heap-allocate an integer if the value no longer fits in a cell
12:59:10 <tathi> right.
12:59:23 <tathi> have you looked at addic and addic. ?
12:59:35 <tathi> PPC has this odd distinction between carry and overflow
12:59:52 <slava> i'll look
12:59:53 <slava> hmm
12:59:55 <tathi> (I have no idea if those would actually be helpful to you)
13:00:01 <slava> i thought i got it working to handle + and - overflow
13:00:03 <slava> i'll explain
13:00:10 <slava> i just generate code like this
13:00:20 <slava> ADDO. 14,15,16 ; whatever
13:00:34 <slava> BF SO end  ; jump to end if no overflow
13:00:53 <slava> ... code to allocate a bignum ...
13:00:56 <slava> end: rest of code
13:01:11 <slava> should i be using 'carry' here instead?
13:01:24 <tathi> no, that should be right.
13:01:33 <slava> hmm
13:01:42 <slava> see the low 3 bits of integers are always 0
13:01:46 <slava> so in effect they are 29 bits
13:02:08 <slava> if i do addc. instead and there's an overflow, will i be able to shift the result left by 3 and shift in the carry>
13:02:20 <tathi> hmm, you probably just have to remove the optimization.
13:02:29 <slava> ok
13:03:01 <slava> the problem is that it should be enabled on x86
13:03:07 <tathi> carry is an internal flag that's just used for extended-precision arithmetic.
13:03:17 <slava> ok
13:04:59 <slava> tathi, do you think its worth eliminating stack usage on powerpc, since it has registers to spare?
13:05:16 <slava> tathi, i'm not sure if the overhead of register shuffling between word calls will be higher
13:06:11 <tathi> yeah, I'm not sure either.
13:06:31 <tathi> a year or two back, I did a bit of running through algorithms by hand...
13:06:43 <tathi> but never did an implementation to see how well it really works in practice.
13:06:53 <slava> i'm not trying to beat gcc here, just something fast but still simple...
13:06:58 <qFox> oh hey
13:06:58 <qFox> :)
13:07:22 <qFox> madgarden> hike me?
13:07:47 <tathi> the impression I got was that a simple thing could generate shorter code fairly easily, so it might be enough to offset the extra register shuffling...
13:08:39 <tathi> qFox: say it out loud -- it sounds almost like "Hi qFox" which Robert had said just before.
13:08:54 <qFox> oh
13:08:59 <qFox> :p
13:09:57 <tathi> Also, I haven't really looked at factor to see how good your code generation is now.
13:10:02 <slava> its not great :)
13:10:21 <slava> basically STC with various inlining and peephole optimizations
13:45:37 --- quit: virl ("Verlassend")
14:02:52 --- quit: easy4 ()
14:09:12 <madgarden> qFox, :D :P
14:23:18 <arke> madgarden: does trillian supprt tabs?
14:23:40 <madgarden> I dunno, not using Trillian anymore.
14:31:19 <arke> Oh.
14:35:12 --- join: TheDarkBlueWiz (TheBlueWiz@ts001d0871.wdc-dc.xod.concentric.net) joined #forth
14:39:26 --- join: Topaz (~top@sown-88.ecs.soton.ac.uk) joined #forth
14:48:16 --- join: Raystm2_ (~vircuser@adsl-68-93-40-211.dsl.rcsntx.swbell.net) joined #forth
14:49:01 --- part: Raystm2_ left #forth
15:00:10 --- quit: Topaz (Read error: 104 (Connection reset by peer))
15:26:47 --- quit: TheDarkBlueWiz (Read error: 54 (Connection reset by peer))
15:38:43 --- quit: qFox ("this quit is sponsored by somebody!")
15:49:48 --- join: saon (1000@c-24-129-91-106.hsd1.fl.comcast.net) joined #forth
15:50:48 <slava> hi saon
15:51:22 <saon> hey slava
15:53:05 * slava notices that his coding productivity is inversely proportional to the temperature outside
15:53:09 <slava> it was 30 degrees today...
15:53:16 <slava> i don't function
15:55:23 <Robert> Heh
15:55:32 <Robert> You're using C, right?
15:55:49 <slava> a little, yes
15:56:26 <Robert> Bah.
15:56:49 <Robert> C as in celcius, in case you aren't joking.
15:56:57 <slava> oh
15:57:08 <arke> ahahaaha
15:57:09 <arke> ol
15:57:09 <slava> i thought you asked about C as in Unix
15:57:09 <arke> ol
15:57:10 <arke> lol
15:57:10 <arke> lol
15:57:12 <arke> lol
15:57:15 <arke> lol
15:57:24 <Robert> arke: STFU PLZ.
16:13:14 <alexander_> sup
16:13:47 --- join: saon|smgl (~saon@c-24-129-91-106.hsd1.fl.comcast.net) joined #forth
16:15:02 * Robert is writing call/return functions.
16:15:37 <slava> what cpu?
16:18:04 <Robert> The one I'm working with, http://robos.org/wiki.pl?Computer
16:21:20 <slava> cool
16:21:34 <slava> a few hundred khz? nice
16:21:39 <slava> bbl
16:21:52 <Robert> Bye.
16:25:59 --- quit: saon|smgl (Remote closed the connection)
16:29:24 --- join: saon_ (~saon@c-24-129-91-106.hsd1.fl.comcast.net) joined #forth
16:29:35 --- nick: saon_ -> saon|smgl
16:32:43 --- quit: saon|smgl (Client Quit)
16:33:18 --- quit: KB1FYR (Read error: 104 (Connection reset by peer))
16:37:32 --- join: KB1FYR (~Alex@196-220.suscom-maine.net) joined #forth
16:38:15 --- join: saon_ (~saon@c-24-129-91-106.hsd1.fl.comcast.net) joined #forth
16:38:23 --- nick: saon_ -> saon|smgl
17:51:12 --- join: saon_ (1000@c-24-129-91-106.hsd1.fl.comcast.net) joined #forth
17:51:44 --- quit: saon (Nick collision from services.)
17:51:52 --- nick: saon_ -> saon
17:57:48 --- join: Raystm2_ (~vircuser@adsl-68-93-114-104.dsl.rcsntx.swbell.net) joined #forth
17:57:54 --- join: Raystm2- (~vircuser@adsl-68-93-114-104.dsl.rcsntx.swbell.net) joined #forth
17:58:50 --- part: Raystm2_ left #forth
17:59:45 --- quit: Raystm2- (Client Quit)
17:59:56 --- join: Raystm2_ (~vircuser@adsl-68-93-114-104.dsl.rcsntx.swbell.net) joined #forth
18:00:46 --- quit: tathi ("leaving")
18:04:36 --- quit: KB1FYR (Read error: 104 (Connection reset by peer))
18:04:49 --- join: KB1FYR (~Alex@196-220.suscom-maine.net) joined #forth
18:06:21 --- quit: Raystm2 (Read error: 60 (Operation timed out))
18:06:58 --- join: Al2O3 (~Al2O3@c-24-1-126-202.hsd1.tx.comcast.net) joined #forth
18:09:10 --- nick: Raystm2_ -> Raystm2
18:22:53 --- quit: Al2O3 ("Leaving")
19:28:36 --- join: AlexF (~Alex@196-220.suscom-maine.net) joined #forth
19:28:36 --- quit: KB1FYR (Read error: 104 (Connection reset by peer))
19:57:00 --- quit: Raystm2 (Read error: 60 (Operation timed out))
20:01:28 --- join: KB1FYR (~Alex@196-220.suscom-maine.net) joined #forth
20:01:28 --- quit: AlexF (Read error: 104 (Connection reset by peer))
20:24:19 --- join: Raystm2 (~vircuser@adsl-68-93-114-104.dsl.rcsntx.swbell.net) joined #forth
20:39:38 <alexander_> cross compiler setup sucks
20:49:29 --- join: snoopy_1711 (snoopy_161@dsl-084-058-133-133.arcor-ip.net) joined #forth
20:58:02 --- quit: Snoopy42 (Read error: 145 (Connection timed out))
20:58:02 --- nick: snoopy_1711 -> Snoopy42
21:07:43 <alexander_> success
21:07:53 <alexander_> I just cross compiled my first word. :)
21:26:20 <Herkamire> alexander_: cross from what to what?
21:38:47 <alexander_> Herkamire: isForth on my x86 laptop to avrforth on my new board
21:39:16 <Herkamire> cool 
21:40:03 <alexander_> the cross compiler is cheese,.. but it'll do what I need
21:40:11 --- join: sproingie (~chuck@64-121-15-14.c3-0.sfrn-ubr8.sfrn.ca.cable.rcn.com) joined #forth
21:40:14 <alexander_> so now I can at least start programming forth, rather than the odd assembling I was doing before
21:40:35 <Herkamire> cool
21:40:49 <Herkamire> so you write code for it in isforth, which compiles to a binary you can load onto your board?
21:40:55 <alexander_> yah
21:41:28 <Herkamire> cool
21:41:31 <alexander_> I should even be able to upload the dictionary from isforth directly to the board, and control it with an umbilical
21:41:36 <Herkamire> I've started thinking a bit about cross-compiling
21:41:39 <Herkamire> porting and such
21:41:44 <Herkamire> but that'll be a ways off if ever
21:41:48 <alexander_> what kind of hardware?
21:42:02 <Herkamire> herkforth for ppc
21:42:08 <alexander_> ah
21:42:17 <saon> hrmm, is it considered bad practice to write to screen memory instead of using the cursor move calls and such?
21:42:32 <saon> it seems both are equally unportable
21:42:37 <Herkamire> I have support for 3 host systems (open firmware, linux and Mac), but it only supports ppc cpu
21:42:40 <saon> and screen memory writing is easier
21:43:56 <Herkamire> I'm working on a better way to switch between host systems, and manage the source code for each. and I think my new way of handling it would work fine for multiple architectures as well
21:44:12 <Herkamire> not that I know asm for any other architectures or anything...
21:44:23 <Herkamire> but I've started thinking about it
21:45:44 <sproingie> saon: whatever works.  if you're using "screen memory", you're probably not aiming at portability anyway
21:45:48 <alexander_> Herkamire: nod.
21:46:06 <alexander_> Herkamire: the AVR is a harvard architecture.. so it pretty much requires an umbilical forth to be useful
21:46:46 <saon> sproingie: yeah, it's for a null-modem version of boggle
21:47:06 <saon> between two comps running rf native, so, not really an issue
21:47:44 <sproingie> saon: any terminal prog is going to hit screen memory at some point anyway.  so if your direct approach doesnt work, move it up a layer of abstraction
21:50:15 <Herkamire> alexander_: I see. there are threading models that work perfectly well on harvard though
21:50:53 <alexander_> Herkamire: sure, but the AVR's code is all in Flash.. you can't really run any software from RAM
21:51:18 <Herkamire> there's ton's of forths that don't write any machine code
21:51:32 <Herkamire> like the vast majority of the ones written in C or other high level languages
21:51:58 <alexander_> sure,.. but I don't want to have my to mix RAM and flash interaction.. the handling of things would suck
21:52:16 <alexander_> I could do a token forth or something
21:52:26 <alexander_> but I wanted to keep things dead simple for the application.
23:03:59 --- quit: Herkamire ("off to bed")
23:36:52 --- quit: sproingie (Remote closed the connection)
23:51:27 <slava> anybody know how long longs are passed on powerpc?
23:56:04 <slava> n/m
23:58:00 --- join: Topaz (~top@sown-88.ecs.soton.ac.uk) joined #forth
23:59:59 --- log: ended forth/05.06.05