00:00:00 --- log: started forth/06.02.25 00:43:40 --- quit: rsyncx ("Leaving") 00:48:50 --- quit: Entopical () 01:29:09 --- join: Cheery (n=Henri@a81-197-45-47.elisa-laajakaista.fi) joined #forth 03:52:47 --- join: llama32 (n=zak@083.a.001.beg.iprimus.net.au) joined #forth 03:56:29 i dont get it... threaded code = the program is basically an array of pointers to subroutines? does the VM just go through these, calling whatever the point to (or jumping to that address in the aray of pointers)? is this actually fast enough? 03:57:28 thats pretty much what it is 03:57:43 its faster on some CPUs 03:58:13 x86 is more suited to subroutine threading = a bunch of call instructions 03:58:23 which particular forth are you looking at? 03:59:19 none, im designing my own VM (for a non-forth language) 04:00:31 are there any benchmarks for this stuff? 04:06:25 somewhere, but I wouldn'T know where 04:06:37 threaded code is faster than it seems 04:10:04 dunno how well it would fit with a non forth language 04:13:05 well im thinking of an OS really, single address space - no protection, so nothing that is not a method call is allowed anyway really. of course i could safely JIT some bytecode, though it is easier to let the compiler do all the compiling :) 04:16:59 --- join: ravenEx (i=ravenEx@VPN-239-248.aichyna.com) joined #forth 04:17:19 ah :) 04:17:28 well depends on the language 04:17:33 which language is it supposed to compile? 04:19:33 a home-brew one :) 04:25:02 well the code that is eventually to be executed can access all memory and everything else the machine has to offer, and the compiler has to produce this code, without being able to produce code that is not safe, so it must be restricted to calls and things like 'mov ebx, 50' (it cannot put raw data on the stack - it would then be used as a pointer, so it could effectively access any memory the machine has to offer - but it may do something like t 04:56:04 --- quit: ravenEx (Read error: 131 (Connection reset by peer)) 05:02:09 --- join: PoppaVic (n=pete@0-2pool236-150.nas22.chicago4.il.us.da.qwest.net) joined #forth 06:16:20 --- quit: llama32 ("Download Gaim: http://gaim.sourceforge.net/") 06:24:25 cvs, rcs, it's all a mulch 06:24:39 heh - yep 06:24:45 I've been playing with svn lately and it has treated me well 06:25:46 I just never learned to like them - too old to change now. 06:26:00 I can just about tolerate diff/patch ;-) 06:26:57 cvs/svn are very useful if you have multiple people working on the same code 06:28:19 Sure. I under stand that, but I was raised w/o that option, let alone folks that ever delivered. 06:28:57 I;m used to tarballs and mail-attachments and download-sites. 06:29:17 we used some hacked-up network-aware version of rcs at hp.. 06:29:18 http://docs.hp.com/en/B6454-97413/ch03s07.html 06:29:55 pass 06:30:12 * reuben grins 06:30:58 i still see folks reference comp.src from time to time 06:31:27 for? rvs or whatever? 06:32:14 oh, not for anything specific.. your comment about the old days of tarballs and mail attachments made me think about it 06:33:35 ahh 06:34:02 yeah, for masses of folks that are active, I can see an advantage.. Like a biz, specifically. 07:30:01 --- quit: PoppaVic ("Pulls the pin...") 07:31:42 --- join: PoppaVic (n=pete@0-2pool198-97.nas30.chicago4.il.us.da.qwest.net) joined #forth 17:51:21 --- log: started forth/06.02.25 17:51:21 --- join: clog (n=nef@bespin.org) joined #forth 17:51:21 --- topic: 'Welcome to #forth. We discuss the Forth programming language, simplicity, and a variety of technical subjects. The rules are simple: don't spam, don't troll, and please speak English.' 17:51:21 --- topic: set by Quartus on [Sat Jan 21 17:32:55 2006] 17:51:21 --- names: list (clog rsyncx Cheery Snoopy42 Quiznos uiuiuiu Quartus saon virl skylan madgarden madwork warpzero @crc OrngeTide Raystm2 Zymurgy @JasonWoof reuben ccfg Jim7J1AJH) 18:32:48 --- quit: rsyncx ("Leaving") 19:37:06 --- quit: JasonWoof (Read error: 104 (Connection reset by peer)) 19:37:24 --- join: JasonWoof (n=jason@pdpc/supporter/student/Herkamire) joined #forth 19:37:24 --- mode: ChanServ set +o JasonWoof 20:19:19 --- join: ravenEx (i=ravenEx@VPN-239-112.aichyna.com) joined #forth 21:10:38 --- join: arke (i=f2@bespin.org) joined #forth 21:17:22 --- join: tathi (n=josh@pdpc/supporter/bronze/tathi) joined #forth 22:44:17 --- join: PoppaVic (n=pete@0-1pool64-83.nas22.chicago4.il.us.da.qwest.net) joined #forth 23:33:00 --- quit: tathi ("leaving") 23:51:41 http://opensparc.sunsource.net/nonav/opensparct1.html 23:52:15 doesn't this sound like Chuck Moore's idea of lots of little processors? 23:55:09 somewhat 23:55:22 the T1 only has 1 core, but 32 threads 23:55:30 x25 has 25 cores, 1 thread each 23:55:48 i dunno which would be better though 23:59:59 --- log: ended forth/06.02.25