00:00:00 --- log: started forth/05.07.06
00:15:53 <kunphuzil> slava: ah, i didn't know you wrote jEdit, i used to use that 
00:42:52 --- quit: slava (Remote closed the connection)
02:06:25 --- quit: virl (Read error: 131 (Connection reset by peer))
02:06:46 --- join: virl (icechat5@chello062178085149.1.12.vie.surfer.at) joined #forth
02:42:34 --- join: Minataku (~Ed@payphoneed.user) joined #forth
02:53:59 --- join: Topaz (~top@host81-157-97-213.range81-157.btcentralplus.com) joined #forth
02:55:17 --- quit: Topaz (Client Quit)
02:58:01 --- quit: kunphuzil ("Download Gaim: http://gaim.sourceforge.net/")
03:05:42 --- quit: PayphoneEd (Read error: 110 (Connection timed out))
05:01:09 --- join: PoppaVic (~pete@0-2pool198-72.nas30.chicago4.il.us.da.qwest.net) joined #forth
05:01:22 <virl> hi PoppaVic
05:01:40 <PoppaVic> hiya
05:59:41 --- join: yGREK (~ygrekhere@ip.85.202.152.96.dyn.pool-3.broadband.voliacable.com) joined #forth
06:00:26 --- quit: yGREK (Client Quit)
06:00:41 --- nick: Raystm2 -> nanstm
06:13:55 --- join: Robert (~snofs@c-f778e055.17-1-64736c10.cust.bredbandsbolaget.se) joined #forth
06:25:46 --- join: PoppaVic_ (~pete@0-1pool75-5.nas24.chicago4.il.us.da.qwest.net) joined #forth
06:26:39 --- quit: PoppaVic (Nick collision from services.)
06:26:42 --- nick: PoppaVic_ -> PoppaVic
06:26:54 <PoppaVic> damned telco
06:31:16 --- join: JasonWoof (~jason@Herkamire.student.supporter.pdpc) joined #forth
06:31:16 --- mode: ChanServ set +o JasonWoof
06:33:30 <PoppaVic> lo, herk
06:40:36 <JasonWoof> mornin
06:40:47 <JasonWoof> I'm up unpleasantly early
06:40:59 <JasonWoof> but hopefully some good breakfast will make it all better
06:41:05 <PoppaVic> It's "one of those days" with the telco *sigh* I'll prolly be in and out
06:41:46 <JasonWoof> I wanted to sleep right until 10am and get almost 8 hours
06:41:57 <PoppaVic> heh - good luck
06:42:03 <JasonWoof> but rain blew into the windows onto me around 7
06:42:19 <PoppaVic> I'm on the goddamned east side of the house, in the country - I suffer it every morning.
06:42:26 <JasonWoof> and it was a bit hot after that when I closed the window mostly
06:42:44 * PoppaVic hugs central air
06:42:54 <JasonWoof> then my mom called shortly after 9
06:43:04 <PoppaVic> oy, geezus - PHD day
06:43:05 <JasonWoof> anyway.
06:43:08 <JasonWoof> afk cooking
06:43:14 <PoppaVic> np
06:47:06 <PoppaVic> Man, I am in dire need of an asm-whiz for at least one set of funcs.. SHitbags
07:02:27 --- quit: PoppaVic ("brb")
07:06:30 --- join: PoppaVic (~pete@0-1pool46-38.nas30.chicago4.il.us.da.qwest.net) joined #forth
07:09:46 --- join: sproingie (foobar@64-121-15-14.c3-0.sfrn-ubr8.sfrn.ca.cable.rcn.com) joined #forth
07:10:18 --- join: PoppaVic_ (~pete@0-1pool46-67.nas30.chicago4.il.us.da.qwest.net) joined #forth
07:10:47 --- quit: PoppaVic (Nick collision from services.)
07:10:51 --- nick: PoppaVic_ -> PoppaVic
07:12:08 --- quit: saon|smgl ("Lost terminal")
07:58:58 --- join: saon_ (~saon@c-66-177-224-222.hsd1.fl.comcast.net) joined #forth
07:59:13 --- nick: saon_ -> saon|smgl
08:14:13 <PoppaVic> arrg
08:14:30 <PoppaVic> herk? you alive?
08:14:53 <JasonWoof> still breathing
08:14:57 <JasonWoof> fighting a virus
08:15:40 <JasonWoof> I think I got it off
08:15:41 <PoppaVic> OK... query... Given that libc is around, how would you set up for calls upon the C lib?
08:15:47 <JasonWoof> now I have to make it so it can't get on again
08:16:05 <JasonWoof> I only know the PPC calling conventions
08:16:14 <PoppaVic> ahhhhhh
08:16:28 <sproingie> rf has a FFI in asm you could use
08:16:31 <PoppaVic> See, that just won't help - I've been cogitating this for 2 days
08:16:45 <PoppaVic> FFI I just downloaded - still not universal, though
08:17:02 <sproingie> there's no such thing as a "universal ffi"
08:17:16 <PoppaVic> I know - cute, ain't it?
08:17:28 <sproingie> rf can invoke c functions.  structs are still up to you
08:17:36 <PoppaVic> yup
08:18:00 <sproingie> structs are pretty simple, just a lot of busywork to get a nice syntax
08:18:18 <sproingie> i'm a fan of factor's ffi.  looks almost exactly like C
08:18:33 <PoppaVic> gimme an url, in case I got something else
08:18:43 <sproingie> for rf or factor?
08:18:51 <PoppaVic> whomever
08:19:05 <sproingie> factor's ffi docs are out of date
08:19:17 <sproingie> http://factor.sourceforge.net/handbook.pdf
08:19:23 <sproingie> search down for "alien"
08:19:37 <sproingie> alien is what factor calls most things associated with ffi
08:19:56 <sproingie> rf's ffi is much simpler.
08:19:59 <sproingie> from libc.so
08:20:02 <sproingie> 1 cimport system
08:20:09 <sproingie> " /bin/sh" system
08:21:07 <sproingie> the implementation is all in rf.asm
08:21:47 <sproingie> it's your basic "calling c functions in asm" tutorial stuff, using dlopen and dlsym
08:21:49 <PoppaVic> well, if you know the home url for that particular mess, I'd love to see it.
08:22:18 <sproingie> dont think the source is browsable online.
08:22:19 <sproingie> http://retroforth.org/release/rf8-generic_ffi.tar.gz
08:22:33 <PoppaVic> sure. and I can do that now.. What I can't do well is demarcate the dstack args I need to get over before the call
08:22:55 <PoppaVic> grabbing
08:23:55 <sproingie> rf requires you to specify how many args the c function takes when importing
08:24:30 <sproingie> nothing to demarcate.  it just pushes eax down, the rest are already on the stack, and then pops off that many when done
08:25:26 <PoppaVic> eax is asm/reg; nonportable in the extreme
08:26:22 <sproingie> rf's FFI is in asm
08:26:26 <sproingie> rf is written in asm
08:26:31 <PoppaVic> sure, I expect that
08:26:42 <PoppaVic> however, it's also prolly for one cpu
08:27:08 <PoppaVic> SO, I end up back in the same hole
08:27:12 <sproingie> intel, of course.  however, C on most CPU's will have you push args onto the stack
08:27:29 <PoppaVic> no
08:27:51 <PoppaVic> we can't insure they push in a given order, or the number (with ...)
08:28:08 <PoppaVic> we rely on C to "know" all this stuff
08:28:19 <sproingie> the c standard specifies a stack, and there is a standard order
08:28:37 <sproingie> compilers can implement other calling conventions, but there is a standard one
08:28:39 <PoppaVic> ahh, is there an url I should know of?
08:29:04 <sproingie> well there's the c00 standard, which is impenetrably dense
08:29:06 <PoppaVic> I know we've cdecl and pascal, of course
08:29:22 <sproingie> right.  cdecl is quite standard about order of args
08:29:28 <sproingie> i believe it's right to left
08:29:50 <PoppaVic> sure... it parses r->l and pushes l->r
08:30:03 <PoppaVic> ..at least, last I looked
08:30:37 <PoppaVic> BUT, you cannot insure this: there is nothing about a compiler like that you can insure
08:30:49 <sproingie> ah yes, it is pushed left to right
08:30:56 <sproingie> yes you can.  cdecl is cdecl
08:31:22 <sproingie> some embedded systems might not bother with cdecl, those you're going to just have to puzzle out
08:31:28 <PoppaVic> once again: think (fmt, ...) - where is the 'end'?
08:31:41 <sproingie> oh you're talking about varargs
08:31:46 <PoppaVic> yah
08:31:51 <sproingie> lots of luck.  varargs suck heavily
08:31:56 <PoppaVic> we've a lot of c-func with them
08:32:23 <PoppaVic> they are great, as far as PASSING args - but they ain't pretty in terms of stack/limits
08:32:23 <sproingie> rf doesn't have any way to do varargs, nor does factor.  i believe it's still the same order tho, just push the format and all the args
08:32:37 <sproingie> you'll have to either parse the format string yourself or require the caller to specify how many args they use
08:32:49 <PoppaVic> yeah, I was afraid of that
08:32:52 <sproingie> the C function will figure out how many args to use, it parses the format string
08:33:13 <PoppaVic> so, let's say...
08:33:21 <sproingie> a lot of ffi's never bother to support varargs
08:34:06 <PoppaVic> ." %s %d %p" foo bar snafu 4 CCALL printf
08:34:13 <sproingie> if it's printf, you could always count the number of %'s in the string
08:34:21 <sproingie> you might overallocate, but that won't hurt anything
08:34:26 <PoppaVic> we'd push as we pop, sans limiter
08:34:52 <sproingie> oh come to think of it, overallocating might underflow the dstack, so i guess that could "hurt" :)
08:35:02 <PoppaVic> yep
08:35:23 <PoppaVic> remember: I leave the real SP alone, and use a struct/queue
08:35:30 <sproingie> in rf it wouldn't, but that's owing to rf's one-section architecture
08:36:30 <PoppaVic> I'm trying to see a downside to "CCALL <name> ( ... args -- ret )"
08:37:11 <sproingie> for declaring them?
08:37:32 <PoppaVic> I can even see: "(CCALL" .... "CCALL) foo"
08:38:10 <PoppaVic> yeah, demarcation/number/etc
08:39:00 <PoppaVic> we ALMOST need to keep a fmt-string for input/output for these calls.
08:39:32 <sproingie> i support.  i personally like factor's syntax the best.
08:39:34 <sproingie> FUNCTION: int   PQgetlength ( PGresult* res, int tup_num, int field_num ) ;
08:39:49 <PoppaVic> wtf?
08:40:07 <sproingie> that's factor's FFI declaration syntax
08:40:31 <sproingie> FUNCTION: void glRasterPos3d ( GLdouble x, GLdouble y, GLdouble z ) ;
08:40:33 <sproingie> and so on
08:40:36 <PoppaVic> hmm... That was almost amusing for C, how about for forthish?
08:41:02 <sproingie> you want to call a C function, you use a C style declaration.  makes perfect sense to me
08:41:18 <sproingie> x y z glRasterPos3d
08:41:44 <PoppaVic> now try a varg
08:41:47 <sproingie> if you want, you could always make it parse stack comments
08:42:04 <sproingie> slava wanted something you could copy and paste a .h file into and tweak a little
08:42:05 <PoppaVic> yes, I am thinking about it
08:43:30 <sproingie> factor doesn't do varargs afaik.  nor does retro.  the principle's straightforward tho, you have a ... type.  you approach delimiting the same way you do in C
08:43:39 <sproingie> parse a format string, use a count, or use a sentinel
08:43:44 <PoppaVic> The only prob I see with "(CCALL" is the marker.. Icould be wrong
08:44:07 <PoppaVic> yeah, the sentinel
08:44:21 <sproingie> syntax is entirely up to you
08:44:37 <sproingie> as for how, varargs are pretty ad hoc, you'll probably end up doing all three
08:45:09 <sproingie> obviously if you want to support printf, you'll need to parse a format string. or require delimiters or a count in the caller
08:45:30 <sproingie> there's a reason pascal doesn't do varargs
08:45:37 <PoppaVic> Yeah, I think I have the right "flavor"
08:46:05 <sproingie> varargs are the bane of many a compiler writer's existence
08:46:15 <sproingie> anyway, gotta go
08:46:23 <PoppaVic> yep, and lord are they necessary - stay well
08:47:42 <PoppaVic> madworks ideas are going to simplify THIS, while complicating the universe
08:50:37 --- quit: PoppaVic ("Pulls the pin...")
09:28:20 --- nick: arke -> grammar_nazi
09:29:37 --- nick: grammar_nazi -> ENGLISH_SUCKS
09:30:03 --- nick: ENGLISH_SUCKS -> arke
10:42:01 --- quit: madwork (Read error: 104 (Connection reset by peer))
10:43:32 --- join: kunphuzil (~Jason@ip-69-10-102-124.cableaz.net) joined #forth
10:44:00 --- join: madwork (~madgarden@derby.metrics.com) joined #forth
10:45:50 --- quit: arke ("OFF TO GERMANY!!!")
11:27:22 --- join: KB1FYR (~Alex@d-66-63-85-222.suscom-maine.net) joined #forth
11:36:40 --- join: tathi (~josh@tathi.bronze.supporter.pdpc) joined #forth
12:01:10 --- quit: KB1FYR ()
13:23:47 --- join: slava (~slava@CPE0080ad77a020-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth
13:37:55 --- quit: virl (Read error: 104 (Connection reset by peer))
13:38:52 --- join: virl (icechat5@chello062178085149.1.12.vie.surfer.at) joined #forth
13:59:17 --- join: KB1FYR (~Alex@d-66-63-85-222.suscom-maine.net) joined #forth
14:07:43 --- quit: KB1FYR ()
14:07:54 --- join: KB1FYR (~Alex@d-66-63-85-222.suscom-maine.net) joined #forth
14:37:44 --- quit: KB1FYR ()
15:01:12 <OrngeTide> I440r, hay
15:03:10 <OrngeTide> I440r, what ARM are you targeting?
15:10:16 <JasonWoof> my LEFT ARM
15:20:28 * OrngeTide shoots JasonWoof in the LEFT ARM
15:21:33 <tathi> I440r is doing stuff with ARMs?
16:18:56 <nanstm> tathi thanks for helping me see virl's question ;)
16:19:15 <tathi> sure
16:19:54 --- nick: nanstm -> Raystm2
17:07:39 --- quit: virl (Read error: 104 (Connection reset by peer))
17:07:50 --- join: virl (icechat5@chello062178085149.1.12.vie.surfer.at) joined #forth
18:11:32 <sproingie> http://home.pipeline.com/~hbaker1/ForthStack.html
18:15:37 --- quit: virl (Read error: 104 (Connection reset by peer))
18:18:40 <sproingie> interestingly it comes to the defense of "roll" over 
18:18:50 <sproingie> er instead of "pick" or "over"
18:33:48 --- nick: kunphuzil -> Tervekalroghklat
18:35:38 --- nick: Tervekalroghklat -> Gloeshaffen
18:40:43 --- nick: Gloeshaffen -> Garjktav
18:46:55 <Raystm2> sproingie: thanks for the link. I think the author was using `roll' as a stand-in for `bring-to-the-front-of-stack' type words that would included pick, over, swap, what-not.
18:48:13 <Raystm2> In the appendix he codes several examples of roll(n) in a lisp.
19:10:59 --- nick: Garjktav -> kunphuzil
19:24:03 --- join: kunphuzi1 (~Jason@ip-69-10-102-124.cableaz.net) joined #forth
19:24:22 --- join: snoopy_16 (snoopy_161@dsl-084-058-132-083.arcor-ip.net) joined #forth
19:30:18 --- quit: danniken (Read error: 54 (Connection reset by peer))
19:32:23 --- quit: Snoopy42 (Read error: 145 (Connection timed out))
19:32:35 --- nick: snoopy_16 -> Snoopy42
19:39:27 --- quit: kunphuzil (Read error: 110 (Connection timed out))
20:09:51 --- nick: kunphuzi1 -> kunphuzil
21:16:59 --- join: LOOP-HOG (~chatzilla@sub22-119.member.dsl-only.net) joined #forth
21:19:21 <slava> hi
21:28:15 <Raystm2> howdy slava, how are you ?
21:28:23 <slava> fine
21:38:30 --- join: danniken (CapStone@ppp-70-249-186-85.dsl.ltrkar.swbell.net) joined #forth
22:00:44 --- quit: LOOP-HOG ("ChatZilla 0.9.61 [Mozilla rv:1.7.1/20040707]")
23:59:59 --- log: ended forth/05.07.06