00:00:00 --- log: started osdev/02.12.05 00:08:43 --- quit: Rico () 00:08:46 --- join: Rico (Rico@node-c-1c6a.a2000.nl) joined #osdev 00:46:13 --- quit: zhware (Read error: 104 (Connection reset by peer)) 01:04:23 --- quit: trans (Read error: 110 (Connection timed out)) 01:09:19 --- join: corsairk8 (~Stealthy@pc-62-31-1-150-cr.blueyonder.co.uk) joined #osdev 01:26:20 --- quit: zwanem (Excess Flood) 01:26:36 --- join: zwanem (User-10390@modemcable017.51-203-24.mtl.mc.videotron.ca) joined #osdev 02:03:17 --- quit: HeavyJoost ("Client Exiting") 02:09:19 --- join: trans (~trans@fatwire-201-211.uniserve.ca) joined #osdev 02:25:52 --- join: lix (~liux@AMontsouris-105-1-1-130.abo.wanadoo.fr) joined #osdev 02:26:21 --- quit: lix (Client Quit) 02:34:23 --- quit: trans (Read error: 60 (Operation timed out)) 02:50:47 --- join: zensox (~zensox@pD954B7C5.dip.t-dialin.net) joined #osdev 02:51:55 --- nick: Robert_ -> Robert 03:05:27 --- join: bcc32 (Telonir@212.199.253.251) joined #osdev 03:07:32 --- join: mmh (~mmh@203.200.62.89) joined #osdev 03:11:00 --- quit: mmh (Client Quit) 03:36:14 i'm baaaack 03:36:20 omg that exam was soooo easy 03:36:59 hehe 03:37:30 took me bout 1 hour... that must be my average time for my exams up to now :D 03:41:01 bleh another physics exam tomorrow 03:48:10 Quack 03:56:48 I should go 03:57:06 school... 03:57:13 subsitute for english probably, thankfully 03:57:15 so I'll bbl 04:03:18 --- join: draq (ident@ESS-p-144-138-43-102.mega.tmns.net.au) joined #osdev 04:09:56 --- join: trans (~trans@fatwire-201-211.uniserve.ca) joined #osdev 04:10:05 --- quit: stormbind (Read error: 110 (Connection timed out)) 04:10:32 --- join: stormbind (~stormbind@pD9E61E86.dip.t-dialin.net) joined #osdev 04:21:11 --- join: lodda (Lothar@p508FCB13.dip.t-dialin.net) joined #osdev 04:21:39 heyhey 04:21:49 hey lod.da 04:22:11 lod.da? 04:22:15 yes. 04:22:45 hum? 04:22:52 *huch 04:22:55 hummin' bird stroke again? 04:22:56 hmm 04:23:00 BOOO 04:23:03 SAID MOO 04:23:05 THE COW. 04:23:06 aha 04:23:16 hummmmmmm 04:23:50 bah, latin classtest makes me silly 04:24:20 latin owns. 04:24:25 not really 04:25:02 latin is fun but it makes me slow and tired 04:53:13 --- quit: zensox (Read error: 110 (Connection timed out)) 04:56:25 lol 04:58:00 what? 04:58:07 --- nick: lynx_school -> lynx 04:58:13 socially 04:58:16 disaster 04:58:26 hmm? 04:58:29 and lingua latina is sucky, too 04:58:39 :( 04:58:52 why socially disaster? 04:59:14 well 04:59:21 just look at the ppl in the channel here... 04:59:54 most of them are pretty on their own.... most dont talk to others in reallife or have sexual problems... lol... some are 24/7 on and have no time for RL 05:00:35 but you have a sex life? :) except with yourself 05:03:21 i talk to other people in rl 05:03:26 and i don't have sexual problems 05:03:29 well not really 05:03:41 i talk to other people in RL 05:03:55 but not often ;) 05:10:28 --- quit: trans (Read error: 110 (Connection timed out)) 05:10:30 Swiftech's web site shows a torture test in which a a barebones system with a 462 equipped CPU was dropped 3 times from a height of 30 feet without any damage to the CPU, motherboard or HS. That suggests a 462-installed system might even survive freight shipping. 05:10:32 lol 05:10:33 funny 05:11:13 lol 05:18:37 hey, 0xA0000 is the base address of VESA vidememory, is it? 05:21:24 --- quit: draq (Remote closed the connection) 05:21:26 --- join: Mathis (~Mathis@manz-d9b950ad.pool.mediaWays.net) joined #osdev 05:21:41 hey Mathis 05:21:48 ho lodda 05:27:58 --- join: quantis (~darth@cg166.halls.umist.ac.uk) joined #osdev 05:28:02 elo 05:28:10 the biatch is back... 05:28:13 :) 05:28:18 lodda : no 05:28:20 heyhey quantis 05:28:23 it is vga memory 05:28:28 hi lodda 05:28:43 my pc is driving me crazy 05:29:00 dax:why? 05:29:04 noise 05:29:10 * quantis goes and hugs lynx i luv you.... 05:29:17 yeah that gets on my nerves as well 05:29:26 this one is also loud 05:29:34 my rooms horrible i can hear the fridge fan as well most of the time 05:30:19 my psu been making weird sounds a s well (not jsutthe fan) like when im downloading stuff, think its getting interference from the ethernet card :( 05:30:28 i just turned off 2 fans 05:30:36 and undervolted another one 05:30:43 wont it overheat ? 05:31:26 good question 05:31:31 brb 05:31:36 going to lower my VCORE a bit 05:31:41 --- quit: dax (Remote closed the connection) 05:35:25 jesus, osnews is getting busy since that new guy joined 05:58:11 --- join: dax (~dax@u212-239-163-49.adsl.pi.be) joined #osdev 06:04:15 --- quit: darkito ("damn! loads of sugar!") 06:08:55 --- quit: lodda () 06:25:35 --- join: trans (~trans@fatwire-201-211.uniserve.ca) joined #osdev 06:28:30 hmm my pc is a bit more quiet now 06:28:39 :) 06:28:43 dropped the VCORE by 0.1V 06:28:49 removed 2 fans 06:28:56 and undervolted another one 06:28:57 how many megahertz you lose ? 06:29:01 uhm non 06:29:05 :) 06:29:10 i really wub this chip 06:29:18 athlon? 06:29:18 if i push it i get 1.6-1.7 out of it 06:29:21 yea 06:29:35 it's 1.1ghz now 06:29:44 and it's supposed to be a 1ghz tbird :D 06:29:54 --- join: Javanx (~javanx@host167-105.pool80180.interbusiness.it) joined #osdev 06:29:57 bah soon noise will be no more! 06:30:37 ahhh 06:30:46 get a pda ... 06:30:56 going to move everything rackmounted and watercool it 06:31:14 and reduce to 1 or 2 quiet fans 06:31:33 your not just going to seal you computer and put a water pipe in are you!!! 06:32:03 ... 06:32:06 uhm no 06:32:27 that was funny, 06:32:28 i just have to find an apropriate radiator and some time 06:32:51 was it you that posted the link to the site about the guy who did that ? 06:33:46 yes 06:33:57 that was damned retarded 06:34:15 guess he didnt know water conducts... 06:34:29 you cold buy a couple of dreamcasts 06:34:41 there liquad/gas cooled 06:34:51 cant rember which 06:35:05 heatpipes prolly 06:35:15 which is indeed gas/liquid :) 06:35:38 my pc is running hot atm 06:35:46 but atleast it's less noisy 06:36:07 till it goes pop... 06:36:16 hmm temps aren't that bad 06:36:22 cpu stays below 50°C 06:37:58 hmm nice to see the chip run at 1.69V (default of 1.75V) 06:39:20 omg i hate this photo-related-shop 06:39:42 they said they would have the printer in max 1.5 week 06:39:52 after 1.5 weeks they said it will take another 1.5 weeks 06:40:03 oh boy i'm going to kill em 06:45:58 --- join: HeavyJoost (~HeavyJoos@a213-84-139-110.adsl.xs4all.nl) joined #osdev 06:47:21 daxos :D 06:47:59 hows daxos going npw 06:48:13 started rewriting it like 2 days ago 06:48:25 and i don't have alot of time atm 06:48:28 exams et al 06:48:47 yeah i always find a start new project when i really really shouldnt be starting em 06:50:38 hehe :) 06:50:47 but progress has been good 06:50:53 and pretty fast too 06:51:18 Titanium daxos # scripts/line_count 06:51:23 7554 total 06:51:27 ought to clean it 06:54:33 and the scheduler is crap 06:54:55 why? 06:55:03 just basic round-robin? 06:55:17 yes 06:55:29 fast priority based one though 06:55:33 but it's a bit sucky 06:55:49 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 06:56:05 no one seems to like there own code in here 06:56:46 heyyy 06:57:05 uhm i like parts of my code 06:57:12 ello novice 06:57:14 whats up 06:57:16 especialy the clean stuff :D 06:57:28 dax: if you dont like it dont write it :).... 06:57:35 novice: not a lot really 07:10:02 me off to bed me thinks 07:10:17 cya in an hour or two 07:10:23 --- quit: quantis ("Client Exiting") 07:10:24 later 07:11:34 --- quit: HeavyJoost (Remote closed the connection) 07:13:34 --- join: HeavyJoost (~HeavyJoos@a213-84-139-110.adsl.xs4all.nl) joined #osdev 07:16:18 whichq = ffs(run_map) - 1; 07:16:18 if (rand() < 0x3000) { 07:16:18 whichq = ffs(run_map & ~(whichq + 1)) - 1; 07:16:20 hmm 07:18:34 --- quit: trans (Read error: 60 (Operation timed out)) 07:29:48 beh 07:30:02 the exam tomorrow is gonna be a bit harder then the previous ones :( 07:30:08 cough "a bit" 07:30:49 (read: A MUTHARFUCKIN' HUGE (read: XBOX HUGE) LOAD HARDER) 07:41:45 "Oh my god!" - "Yes?" *g* 07:47:55 --- join: beefcommando (~beefcomma@spkdsl-116-101.cet.com) joined #osdev 07:50:20 --- join: Kurt (~kmw@gc-nas-01-s221.cinergycom.net) joined #osdev 07:53:56 --- join: stormbind_ (~stormbind@pD9E61E86.dip.t-dialin.net) joined #osdev 07:54:19 --- quit: stormbind (Read error: 113 (No route to host)) 07:56:22 --- join: lodda (htjonhigfg@p508FE232.dip.t-dialin.net) joined #osdev 07:56:31 heyhey 07:56:47 --- quit: corsairk8 (Read error: 54 (Connection reset by peer)) 08:00:04 --- nick: Kurt -> DukeEllington 08:05:15 * lynx tickles quantis 08:05:47 * Robert bites lynx 08:05:57 --- join: acme (acme@p50809F57.dip.t-dialin.net) joined #osdev 08:06:00 * lynx is sexually aroused 08:06:36 :D 08:06:38 Hey acme 08:21:15 YO 08:21:20 robert has da big skills 08:22:36 Yes! 08:25:00 --- join: lix (~liux@AMontsouris-105-1-2-10.abo.wanadoo.fr) joined #osdev 08:25:13 --- part: lix left #osdev 08:29:52 no i have DA HUGE SKILLZ 08:29:58 LIKE XBOX LOL@ME 08:30:13 shall i start reading some of my poetry again? 08:30:18 IK BEN HET DOMME ROBERT LOL@ROBERT 08:30:27 No! 08:30:44 why not? 08:31:02 it suxx lol!!! roflmao omfg!!! 08:31:03 Robert: ... ;) 08:31:04 hrm 08:31:13 Hrm! 08:32:24 what's wrong with poetry 08:32:49 and especialy MINE! 08:32:50 hmm 08:32:59 It sucks. 08:33:02 Period. 08:33:03 GNAGNA 08:33:09 SAID MISTER COW 08:33:15 FOR HE COULDN'T SAY 08:33:18 GNUGNU 08:33:22 AS HE WAS A COW 08:33:25 NOT A GNU 08:33:35 * dax bows 08:33:44 >:D 08:34:09 thank you thank you 08:34:13 another one? 08:34:27 why all uppercase? 08:34:41 cause capslock was on and i didn't notice. 08:35:04 <:D <--- LOOK I HAVE A HAT! 08:35:33 robert you shouldn't wear your hat like that btw 08:35:35 j00 r teh c0wz0r 08:35:55 =#[:-) 08:35:58 I have a hat, too! 08:36:34 ****=#[:-) <-- look a chimney on my head! 08:37:56 ANOTHERH POEM? 08:38:07 gnu gnu said the gnu 08:38:16 sitting upon my chamber door 08:38:21 crack crack said the door 08:38:28 and there fell the gnu 08:38:37 ... 08:39:43 omg that was a good one 08:39:57 brb 08:40:10 mount /dev/food /home/dax/mouth 08:41:51 --- quit: minddog[zz] (Remote closed the connection) 08:42:30 --- join: minddog (~minddog@ip68-98-85-105.ph.ph.cox.net) joined #osdev 08:44:22 --- nick: DukeEllington -> Kurt 08:50:05 --- quit: Robert (forward.freenode.net irc.freenode.net) 08:51:49 --- join: Robert (~Robert@robost86.tsps1.freenet6.net) joined #osdev 08:56:29 BACK 08:56:32 poem? 08:58:09 --- join: trans (~trans@fatwire-201-211.uniserve.ca) joined #osdev 09:00:26 yes? no? 09:00:33 COMe ON YOU DIDN'T ALL DIE DID YA 09:00:43 me did 09:00:59 Mathis: you read my poems? 09:01:01 right in front of my Cygwin/XFree86 server... 09:01:08 what po? 09:01:18 GNAGNA 09:01:18 SAID MISTER COW 09:01:18 FOR HE COULDN'T SAY 09:01:18 GNUGNU 09:01:18 AS HE WAS A COW 09:01:18 NOT A GNU 09:01:21 and 09:01:28 gnu gnu said the gnu 09:01:28 sitting upon my chamber door 09:01:28 crack crack said the door 09:01:28 and there fell the gnu 09:02:03 hmmm 09:02:11 I'll laugh later... 09:03:38 :/ 09:03:55 are you an animal hater? 09:04:59 crack crack said the door and with falling on the floor , it beheaded the gnu IN NU! 09:05:06 IM NU? 09:05:46 ts 09:05:58 zzrsh zrsh said the plug and the gnu willing to f*** was electrocuted :/ 09:06:06 roflmao 09:06:12 good one 09:06:15 :) 09:06:29 now one about cows biting you 09:06:57 hrm 09:07:12 moo moo there is a cow, is it you? 09:07:22 it sits on a lovely site 09:07:28 noo noo there is a gnuu, is it youuuu? 09:07:31 nooo nooo blablabla 09:07:33 song :) 09:07:41 and wants lynx to bite :( 09:07:45 that is a cool poem! 09:08:59 moo moo , is it a cow, is it you? it sits on a lovely site ; merley wondering if it exists, the gnu? ; instead of thinking further it prefered lynx to bite! 09:09:01 now 09:09:06 it is a cross-rhyme! 09:09:13 sexeh man 09:09:15 moo/you/gnu , site/bite 09:14:21 dax+leisbeth were playin on the ground \n suddenly a stranger sound \n and with one apace lunge \n a cat sprang out and ate the bunch 09:14:31 if lunge and bunch rhymes :P 09:14:35 and he held forth a gnu 09:14:37 and said 09:14:47 behold, this is a gnu, for it loves penguins 09:14:54 and all were amazed 09:15:00 because they knew that it was so 09:15:12 and they asked to the lord 09:15:16 do you do windows too? 09:15:20 and the lord said 09:15:21 no. 09:15:34 hrm 09:15:36 :D 09:15:42 mine was better 09:15:42 rowan owns. 09:17:13 for all the wisdom that we have \n penguins dont. (period) 09:17:19 wow 09:17:25 this is nearly... senseful 09:18:19 and here a nifty one 09:18:21 hrm 09:18:27 a gnu was in the wood 09:18:41 a penguin came and everywhere was blood! 09:18:48 HA! 09:18:53 penguins kill gnus! 09:19:04 hrm 09:19:15 or 09:19:17 heiku style 09:19:27 a gnu was in the wood 09:19:36 a penguine came full of blood 09:19:41 it was the devil. 09:19:43 ---- 09:19:45 AMEN! 09:19:51 *bsd rulz :P 09:20:01 it has a double meaning! 09:20:42 dax ? 09:20:46 uhm no 09:20:47 ? 09:20:59 yes it does 09:21:16 watch it or i'm gonna quote my deathmetal lyrics if i find them 09:21:23 bah 09:21:29 i am the better poet! 09:21:48 dax is not a weasel, nor devils are penguins! 09:22:00 neither are* 09:22:30 "Convertion to homosexual", "The homo i will slaughter", "Necrophelic Homophobe" and damn i don't have"... separated & torn apart" anymore :( 09:22:34 lost my best 09:22:51 and "Curse upon thee" is still in progress 09:22:53 bah 09:23:24 when designing an FS.. where do you start? 09:23:35 goals 09:23:51 and then 09:23:59 hrm 09:24:07 i want to find some kind of a "todo" list 09:24:15 damn i lost my best song 09:24:16 :( 09:24:42 dax is addicted to porn, yet he does scorn \n goatse.cx :( 09:25:03 and i'm not going to quote the other ones, or you guys are going to start thinking that i'm a weirdo 09:26:05 if you don't already :( 09:27:01 i leave 09:27:12 --- nick: lynx -> lynx|out 09:27:26 oh noo i cannot breath outside 09:27:36 breathesd 09:27:36 g LOCKED UP 09:28:27 SMALL CHAMBER 09:28:27 GAS RELEASED 09:28:30 etc etc etc 09:29:17 GAS RELEASED, CAR FULL ! DAX DRIVE HOME! 09:29:26 ... 09:29:32 kickban? 09:29:38 who? 09:29:56 gas as in gasoline... 09:30:02 uhm no 09:30:08 that would be petrol. 09:30:11 JUST ONE MORE TIME 09:30:11 WHILEST HE WAS BURNING 09:30:11 FROM THE INSIDE OUT 09:30:11 BEING SLAUGHTERED BY 09:30:11 THE PURIFYING GAS 09:30:39 etc etc etc 09:30:42 wait 09:30:51 hrm 09:30:54 SMALL CHAMBER 09:30:59 LOCKED IN 09:31:03 POWER ON 09:31:03 THIGHT PUSSY 09:31:05 oops 09:31:06 TV ON 09:31:16 USER NUMBED TO DEATH. 09:31:27 another haiku style one 09:31:55 bah 09:32:01 SMALL CHAMBER\n LOCKED IN\n POWER ON\n eleectriiiicciiittyyyy floowwwwinnggg\n THROUGH MY BRAIN BURNING THE REMAINS 09:32:08 bleh 09:32:15 blahblah 09:32:17 bluhuh 09:32:24 greehehe 09:32:30 small chamber locked in power on dax horny sextoys move. 09:32:31 rehehe 09:32:35 :) 09:32:48 hrm 09:32:58 damn i got ZE EXZAM tomorrow 09:33:02 physics (again) 09:33:04 ZE ? 09:33:05 ohh 09:33:07 nice 09:33:11 not really 09:33:12 about which topic? 09:33:17 i had another physics exam today 09:33:24 basic thermodynamics 09:34:05 ah 09:34:05 ok 09:34:11 heating of gasses, fluids, solids, melting/condensing/sublimating/etc etc etc all that stuff 09:34:46 easy stuff 09:34:54 SMALL CHAMBER \N LOCKED IN \n POWER ON \n 650 WATTS \n 10 MINUTED \n DAX DONE. 09:35:24 bye 09:35:40 :'( 09:35:54 gtg too... i have to study a bit or my mom will start naggin me 09:36:05 although i'ld prolly pass the exam without studying 09:36:09 just too easy 09:37:31 bbl 09:40:16 --- quit: Robert (Remote closed the connection) 09:40:22 --- join: Robert (~Robert@robost86.tsps1.freenet6.net) joined #osdev 09:50:16 --- join: quantis (~darth@cg166.halls.umist.ac.uk) joined #osdev 09:50:25 evening 09:54:38 --- quit: Robert (Remote closed the connection) 09:55:23 --- join: Robert (~Robert@robost86.tsps1.freenet6.net) joined #osdev 10:02:58 Welcome back Robert! 10:10:46 --- nick: geist-sleep -> geist 10:12:06 howdy folks 10:21:28 ello there 10:21:34 --- join: dax__ (~dax@u212-239-207-42.adsl.pi.be) joined #osdev 10:22:02 wb dax 10:22:08 puter not blown up yet ? 10:22:08 sup 10:22:26 I'm so pumped, first day of motorcycle class today 10:22:28 geist: done any more stuff with newos recently? 10:22:33 wow 10:22:39 guess you enjoyed it then ? 10:23:03 haven't been yet, it's later on 10:23:09 ahh right 10:23:11 (10am here now, class at 6pm) 10:23:27 funny that, its like 6 here :) 10:23:37 quantis: yes and no, I've been pretty busy. But I've been interested in getting rid of a few old bugs, in general tighten up the ship 10:23:45 should be in uk you couldbe having you lesson any minute now... 10:24:22 --- quit: Mathis ("User pushed the X - because it's Xtra, baby") 10:24:27 there's a particular crasher that's always existed that I think I have a way to track down, but I haven't had a chance to try it 10:24:28 me going start my kernel afreash in a couple of days when i get the draft interim report for my project finished and can relax a bit hopefully 10:24:30 tooo busy 10:24:45 what causes it ? 10:24:53 if I knew, I'd fix it 10:24:54 is this one of the triple fault ones ? 10:25:02 but I can reproduce it by running serious load 10:25:03 good point 10:25:09 like thousands of processes 10:25:13 oh yeah i rember you saying about that 10:25:21 starting and stopping 10:25:23 well i listened in more like... 10:25:51 it takes a while to reproduce, so running under bochs isn't much of an option, since it takes like 8 hours to do it, and bochs would produce about 50gigs of logs by then 10:25:59 is it that the process table runs out of memory? 10:26:09 shouldn't be 10:26:24 simple things like that are taken care of 10:26:31 :) 10:26:34 at the minimum I'll panic gracefully 10:26:49 and thankfully it'll crash on single proc machines as well 10:26:59 so it's not some terrible SMP race, thouse really really suck to debug 10:27:21 it's probably something vm related 10:27:44 very hard to recreate all the time, cause you got to hope both processors are working exactly the same way at the same time :( 10:27:44 like the page directory gets freed and reused for something else but is still in use as a page dir 10:28:07 that shouldnt be hard to fix, doesnt you vm support regions ? 10:28:11 well, if you could recreate it, lots of times you'll see a cpu crash and how it got there was completely inexplicable 10:28:23 quantis: sure, but that doesn't have a lot to do with it 10:28:45 and hopefully it's not a tlb problem, those are really really tough to debug 10:28:46 yeah just mark the region fixed in the page direcectory, so as not to swap it 10:29:03 only have a very basic grasp of tlb's 10:29:04 since you'll almost always have some completely inexplicable crash, and by the time you inspect the corpse, it's fine 10:29:07 they scare me 10:29:23 arse i got to go 10:29:26 kickboxing calls 10:29:29 quantis: no, I mean the physical page that a page dir is gets recycled and assigned to another region 10:29:30 enjoy the cyckleing 10:29:38 ahh right 10:29:40 hmm 10:29:43 anyway, later 10:29:53 --- quit: quantis (Remote closed the connection) 10:33:35 --- quit: acme (Read error: 54 (Connection reset by peer)) 10:34:25 --- join: gianluca (~kernel@ppp-60-139.28-151.libero.it) joined #osdev 10:34:28 hello 10:34:58 sup 10:35:08 me? 10:35:25 sure 10:35:35 hmmm 10:35:41 nothing :) 10:35:49 cool, that's not bad 10:35:55 I gots too much stuff goin on here 10:36:01 but it'll be quieter this weekend 10:36:20 yes yes, just think if i'd start with "HEY WHY MY KERNEL DOESN'T WORK?!??" :) 10:36:28 oh what are you doing? 10:36:40 hey, i tried newos some day ago :) 10:36:58 it saw my dual :) 10:36:59 yeah, not very useful 10:37:04 but it booted? 10:37:09 yes 10:37:12 cool 10:37:26 if you had a realtek 8139, you can telnet into it as well 10:37:33 do it use all two processors? 10:37:37 sure 10:37:41 oh i have it :) 10:38:09 yeah, if you have the 8139, it should be currently hard coded to be at 192.168.0.99 netmask 255.255.255.0 10:38:11 ops, not in this computer :P 10:38:17 with the gateway at 192.168.0.1 10:38:23 --- quit: dax (Read error: 110 (Connection timed out)) 10:38:38 well, actually was it with that old floppy on the web page? not sure what that's at 10:38:48 I need to stick up a new one, but I want to fix a couple of stability bugs first 10:39:05 you should sync up with my source server and build it yourself 10:39:06 ok, i'll try cvs version then 10:39:14 oh you use no cvs 10:39:15 yeah, it's not cvs though, it's perforce 10:39:24 but I have a page for it, hold on 10:39:25 hmm free sw? 10:39:47 perforce will give you a free license if you're working on an open source project 10:39:47 --- join: muphicks (~Gazza@earthforce.freeserve.co.uk) joined #osdev 10:39:56 oh 10:39:57 http://sourceforge.net/docman/display_doc.php?docid=5692&group_id=20140 10:40:04 bitkeeper stile 10:40:13 style :p 10:40:14 I guess, I dont know anything about bitkepper 10:40:35 except a bunch of losers are bitching at linus about it 10:40:46 hmmm well bitkeeper give you free licence but you have to public your stats 10:40:54 into its webpage 10:41:04 yeah, perforce will but you have to allow everyone to sync against it 10:41:18 oh, so they give server space for it_ 10:41:19 ? 10:41:28 nah, I'm running my own server 10:41:38 I just have a 20 user license for free 10:41:47 normally that's about $600/yr/user 10:42:19 the problem with bitkeeper is that if you want to publish it in their server space 10:42:46 you will see your project near of main linux kernel projects, and that's no fine :D 10:42:58 not fine 10:43:54 oh so you can't install it on your own server? 10:44:02 yes you can 10:44:05 oh 10:44:11 it's only if you want a free server 10:44:14 but i have no server space 10:44:16 yes 10:44:28 yeah, I dont know of any perforce free servers, you normally just run your own 10:44:31 and it's really cool to use 10:44:44 which I'd want to do anyway for any of them 10:44:58 well if you have a server, why not 10:45:09 right 10:45:19 that way I can control the backups 10:45:21 but think to people living with 56k's on year 2003! 10:46:31 oh and if you have a private project, you can use bitkeeper without sending your project stat to bitmover.com 10:47:35 ah, well if perforce didn't have an open source thing, I'd consider paying the $600/year 10:47:51 so is perforce so cool? 10:47:57 what makes it special? 10:48:00 I'm soo used to it. I've used it at every company I've ever worked for 10:48:22 and as far as I know it's used at every company I've ever heard of, except maybe microsoft, but they have their own system 10:48:33 and yeah I totally like it 10:48:42 I'm pretty good with it too, lots of merging tricks 10:48:46 --- quit: Kurt ("Connection reset by ear") 10:50:05 --- quit: Javanx () 10:50:51 drex now has 5 syscalls :D 10:51:33 drex? 10:51:59 http://drex.sf.net , my little kernel project 10:54:22 --- quit: Robert (Remote closed the connection) 10:54:28 --- join: Robert (~Robert@robost86.tsps1.freenet6.net) joined #osdev 10:55:30 --- quit: zwanem (Read error: 110 (Connection timed out)) 11:00:03 ok i am leaving 11:00:11 see you 11:00:24 --- join: liesbeth (Simon@223.124-136-217.adsl.skynet.be) joined #osdev 11:00:29 ello 11:00:48 hi liesbeth 11:01:10 how's everybody 11:01:26 i am going! 11:01:28 bye! 11:01:33 toodeloo 11:01:40 --- quit: gianluca ("ircII EPIC4-1.1.7 -- Are we there yet?") 11:02:29 does anyone know if therapy? is going to tour in belgium next year? 11:03:19 nope 11:03:57 no tour or you don't know? 11:07:44 dont know 11:07:50 --- join: Aardappel (wvo96r@p508C5C78.dip.t-dialin.net) joined #osdev 11:09:18 YESSS, THERAPY? IN BELGIUM IN AUGUST, goodygoody 11:09:22 wahooo 11:10:25 sorry 11:11:03 ok, and now tell me what therapy? is :) 11:12:16 music 11:13:28 hmm.well, even i understand that :) but what kind of music? 11:14:12 eeh, guitars 11:14:16 loads 11:14:20 and drums 11:14:25 and a drunk singer 11:14:32 hrmm 11:14:38 classic 11:15:07 starting kazaa... 11:15:30 i'm not allowed to go to concerts, yet, because i'm a girl 11:15:34 boohoohoo 11:15:41 ??? 11:16:09 but you'll be 16 then, right? 11:16:14 no 15 11:16:16 about 11:16:23 i'll ask my dad to come with me 11:16:36 he can't say no 11:16:48 i hope 11:16:51 nevermind 11:17:00 what's with daxy? can'the come with you? 11:17:02 i hope they tour here 11:17:15 eeh, don't think he's into concerts 11:17:22 i do'nt know 11:17:29 i'll ask him, nicely 11:19:00 ooh, i could take my brother 11:19:02 good 11:20:20 this monkey is gone to heaven tralalla 11:20:26 ^ 11:20:33 | 11:20:42 steps in the thinking of a woman 11:20:54 qué 11:21:10 i'm not a woman 11:21:14 i don't know french 11:21:28 que is french, qué is spanish 11:21:31 no? 11:21:34 hmmm 11:21:34 :) 11:21:47 i don't know one of them 11:21:58 ooh, neither do i 11:22:08 (is that a correct sentince?) 11:22:17 i think so 11:22:24 but's sentence i guess 11:22:24 great 11:22:34 well, my english sux 11:22:37 :( 11:22:59 mine too! 11:23:07 as you see in "but's" which is wrong i think 11:23:41 i've got a huuuuuuuuuge exam chemistry tomorrow hiphiphooray 11:23:48 i'm nervous 11:23:48 therapy? sounds funny 11:23:55 yes? 11:24:00 i like it 11:24:17 i wrote latin & german this week! 11:24:18 i put it on and than jump round the room 11:24:46 it's not a jump-around music i think 11:25:00 ;) 11:25:03 it is. 11:25:22 i heard diane 11:25:34 ooh, thats a soft song 11:25:40 CHURCH OF NOISE 11:25:45 SCREAMAGER 11:25:59 EVIL ELVIS is good too 11:26:06 six mile water is soft 11:26:18 WHERE EAGLES DARE is good 11:26:22 ango Romeo 11:26:28 *Tango Romeo 11:26:33 Stalk and slash 11:26:39 do all belgians like writing in uppercase? 11:27:16 HE S NOT THAT KINDA GIRL is good too 11:27:35 do not insult my people :) 11:27:54 nowhere is a jumpsong 11:28:19 ooh, STORIES is a scary song 11:28:44 DIE LAUGHING is a romantic one, not really, nearly 11:29:14 the guitarsolo in THIS ONE'S FOR YOU is fantastic 11:29:34 bodybag girl is not that good 11:29:50 blahblahblah, i'll stop 11:29:56 :) 11:30:15 church of noise sounds strange as diane does 11:30:16 you like it? 11:30:25 a bit? 11:30:29 say yes 11:30:33 it's ok 11:30:36 oof 11:30:43 oof? 11:30:48 dax doesn't like it that much 11:30:54 i'm crazy bout it 11:31:26 ok, it _is_ a jump-around music 11:31:51 and if i sound very drunk, it sounds almost like it 11:31:57 :) 11:32:11 "if i sound" or if you are? 11:33:05 i sound, i never drank alcohol before, except champagne for little children 0.5° 11:33:37 kinda silly 11:34:21 0.5 :) 11:34:59 i'm also not drinking any alcohol 11:35:08 at the moment? 11:35:13 no, ever 11:35:20 (except silvester) 11:35:25 good to know i'm not alone 11:36:04 after 2 bottles of kidschampagne i still can get quite tipsy ;) 11:36:31 hrmm...my code does not work :( 11:36:42 i'm gonna looking some TV 11:37:05 ok 11:37:52 hi 11:38:28 bonjour! 11:40:56 i'm seeing dax tomorrow, joy! 11:42:23 --- quit: n0vice ("I like core dumps") 11:42:54 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 11:45:10 --- nick: dax__ -> dax 11:45:24 dax! 11:45:34 lilo and stitch dvd, is al uit!!!!!!!!!! 11:46:09 can someone say my nick real quick 11:46:49 no. 11:49:46 |\|0\/1C3 11:50:21 come on 11:51:39 --- join: ChillySpy (ChillySpy@ppp110.ppp3.cleveland.nccw.net) joined #osdev 11:53:04 and now - time to do homework 11:59:28 n0vice 11:59:51 ahh thank you 12:01:03 --- quit: ChillySpy () 12:03:22 --- quit: liesbeth () 12:03:44 Sounds of 165 dB would cause a person's hair to catch fire from the frictional heating caused by air undergoing such intense compression and expansion. 12:04:41 Wednesday, 4 December, 2002, 14:28 GMT 12:04:41 Novel fridge cools with sound 12:04:41 A fridge, PA 12:04:41 Food could soon be chilled with sound 12:04:41 Some of the coolest sounds in the world are coming out of a laboratory in Pennsylvania. 12:04:44 At the lab, two US researchers are working on ways to use sound to chill food. 12:04:50 They have produced a sonic fridge that converts very loud sounds to directly cool a fridge containing ice cream. 12:04:55 The researchers hope that their work will end reliance on gases that can contribute to global warming. 12:04:59 Louder than rock 12:05:03 The researchers have exploited the fact that sound waves travel by compressing and expanding the gas that they are generated in. 12:05:08 In the 1980s, Scott Backhaus and Greg Swift at the Los Alamos National Laboratory realised that this compression/expansion action could be used to cool and heat metal plates placed in the path of the sound wave. 12:05:13 A temperature gradient can be generated by putting a stack of plates in the right place in a tube in which the sound wave is bouncing around. Some plates in the stack will get hotter and others colder. 12:05:20 Ice cream, BBC 12:05:22 The novel fridge will be used to cool ice cream 12:05:24 All it takes to make a refrigerator out of this system is to attach heat exchangers to the ends of the stack. 12:05:29 Now, Matt Poese and Steve Garratt of the Applied Research Laboratory at Penn State University are doing work on a small chiller cabinet that will be used to keep ice cream cool. 12:05:35 The research is being sponsored by ice cream makers Ben & Jerry's and Unilever. 12:05:39 The pair are using enormously loud sounds to keep their chiller cabinet cool. 12:05:42 Humans feel pain when they hear sounds of 120 decibels, a level typically reached next to the speakers at a rock concert. 12:05:49 The sounds pumped through the Penn State fridge reach 173 dB, tens of thousands of times more intense than any rock concert. 12:05:52 oops 12:05:54 lol 12:05:56 too much in clipboard 12:05:58 nm the flood. 12:12:09 --- part: muphicks left #osdev 12:13:52 --- join: Ghiottone (~alex@ppp-23-16.27-151.libero.it) joined #osdev 12:14:44 interesting stuff though 12:19:31 --- join: quantis (~darth@cg166.halls.umist.ac.uk) joined #osdev 12:19:35 ello 12:19:48 yes interesting 12:19:53 http://news.bbc.co.uk/2/hi/technology/2543085.stm 12:19:55 that the article 12:23:56 freebsd 5.0 is getting close to release :) 12:24:04 cool 12:24:11 so almost time to update 12:31:06 moo 12:34:49 --- quit: HeavyJoost ("Client Exiting") 12:35:18 * file works 12:35:26 I need to cover over my work projects for tomorrow 12:35:40 I get two whole periods of nothing, so I thought I'd rewrite my HTML parser and graphics engine 12:37:26 --- join: Kurt (~kmw@gc-nas-01-s273.cinergycom.net) joined #osdev 12:40:22 file whats this for ?, some app or your os ? 12:41:45 quantis: contract work 12:42:31 --- join: HeavyJoost (~HeavyJoos@a213-84-139-110.adsl.xs4all.nl) joined #osdev 12:44:42 but your doing this at school? 12:44:49 is it renta coder stuff ? 12:45:06 what's renta coder? 12:45:40 ppl put up reqwuest for ppl to code them a project, and then coders bid to do the project 12:46:08 its like ebay but matching coders to programs rather then buyers to sellers 12:46:25 ic 12:46:26 no 12:46:30 hmm oop business practises... 12:46:36 * file applied for a job as a coder and got accepted 12:46:44 cool 12:46:47 :/ 12:46:53 what language you using 12:46:58 i wanna earn some moneh with me skillz too 12:47:05 which exe would be faster? a normal or a UPX compressed one? 12:47:06 C 12:47:09 * file does all C 12:47:20 me has no skillz, me has no money...:( 12:47:58 im off to spar back in a bit 12:47:59 i surely have the skills 12:48:39 well till you show us daxos how can we believe you :) 12:49:06 i have showed fragments of daxos 12:49:16 oh okay 12:49:18 :( 12:49:21 and the performance of previous daxos versions was decent 12:49:25 aka better than mach 12:49:32 bit worse than l4-ka v4 12:49:33 Skills to do what, daxchen? 12:49:40 code stuff 12:49:50 design skills are a bit less developed though 12:49:53 dax: Who wants a little kiddie to try to code his new OS? ;) 12:50:04 well l4 ment to be the current pinacle of microkernel performance aint it ? 12:50:22 yes 12:50:49 i got very close to l4ka rc2 performance 12:50:51 damn your aiming high 12:50:57 but can't get close to v4 12:51:12 im still trying to see how many ways i can print hello world from a kernel 12:51:24 i'm not aiming high 12:51:37 if you don't aim high enough, your project will turn into crap. 12:51:44 you must demand perfection. 12:52:01 of the design, of the code, of the layout, e-v-e-r-y-t-h-i-n-g 12:52:43 allright then, im going to write an exokernel, with posix libos and gtk on top.... 12:52:57 been thinking about exokernels too 12:53:01 or not as the case will probably turn out... 12:53:02 couple of nice ideas 12:53:29 nice ideas i just still havent got around to look at the kernel to see what makes it different from a micro fully yet 12:53:37 i think i understand the principles 12:53:49 just not howd they could implment it much different 12:54:00 alot of differences from a pure micro 12:54:14 but pure micro isn't really the way to go... ipc is too limiting 12:58:05 splat 12:58:07 i'm not saying daxos is perfect, it's far from perfect 12:58:18 but it will be close to the way i want it 12:59:45 --- join: mik42 (~mik@me-augustacuda1cable3a-68.agstme.adelphia.net) joined #osdev 13:02:27 --- quit: Ghiottone ("Uscita dal client") 13:04:39 fixed a bug with spinlocks 13:04:52 --- quit: stormbind_ (Remote closed the connection) 13:04:58 still have a crappy scheduler though 13:05:06 needs to be improved really urgently 13:05:13 as it starves lower priority tasks 13:05:16 and threads 13:08:40 test your might 13:08:47 okay? 13:09:37 anyone know of a little program that will let me put in an assembler instruction, and it spits the opcodes out in hex?? 13:09:41 i don't want to assemble/disassemble every time... 13:09:47 hmm nope 13:09:50 and i don't know assembler good enough yet to do it by hand 13:09:54 the intel manuals contail lists though 13:10:30 yeah, i was just wondering if anyone had automated it... 13:10:56 dax: in most systems lower priority threads get starved if a higher priority is cpu bound 13:11:15 er better put. lower priority cpu bound threads get slaughtered by higher priority cpu bound processes 13:13:16 well this scheduler never gives a slice to a lower priority one :( 13:13:57 lots of real ones dont 13:14:37 and i also have some doubts with my way of deleting threads 13:14:48 --- join: violat0r (gigax@63.74.24.35) joined #osdev 13:14:50 i "kill" them and put them on a reaper queue and wake up the reaper 13:14:59 but in general the running threads are usually within a few priorities of each other and real systems tend to dynamically boost or not the threads by a few points 13:15:11 reaper then gets a timeslice and frees all the memory 13:15:21 okay, what's the problem with that? 13:15:29 hmm i dunno 13:15:44 lets just hope the reaper doesn't decide to crash 13:15:56 well, shit if it crashes that's a kernel panic 13:16:06 uhm yea kinda 13:16:16 how about 'lets hope doesn't decide to crash' 13:16:19 and it's so simple that if it crashes the rest of the kernel should be already dead 13:16:34 lets hope the whole system won't crash 13:16:52 well atm it's perfectly stable and i can get huge uptimes :D 13:17:07 i have no memory management though :( 13:17:16 mine is too (the "Hello, World!" one) 13:17:24 ;) 13:17:37 hey mine atleast has basic ipc and tasks & threads 13:17:49 IRQ handling is a mess and slow, should be rewritten 13:17:52 GOOD FOR YOU! 13:18:33 and a nicely working slab allocator 13:18:39 but i wonder if a slaw allocator is worth it 13:18:40 git 13:18:45 dax: it is 13:18:47 I need to write one 13:19:04 even for a small kernel that doesn't allocate alot? 13:19:10 but the basic slab allocator is just a pain to use, where it gets really useful is when you have a caching slab allocator 13:19:53 --- join: gianluca (~kernel@ppp-88-135.28-151.libero.it) joined #osdev 13:19:57 bz 13:20:05 i have a small 600 line one 13:20:12 hi dax 13:20:24 it can't free memory or shrink caches though 13:20:37 --- quit: lodda ("User pushed the X - because it's Xtra, baby") 13:20:41 hi gianluca 13:20:43 I need to do a big solaris-style slab allocator, but there are some terrible vm reentrancy issues to deal with 13:20:57 hi quantis, geist 13:21:09 drex has slab allocator :) 13:21:10 but in general any sort of kernel memory allocator that can resize has terrible vm reentrancy probs 13:21:13 every allocation done by my kernel is performed with my slab allocator 13:21:25 mine too 13:21:27 the vm issues are why I haven't done one yet 13:21:32 --- quit: Aardappel (Connection timed out) 13:21:41 I just use a generic malloc-style heap 13:21:44 it creates a few caches for a few common sizes, so each malloc uses the slabs :D 13:21:55 yah 13:22:04 and threads/tasks/... all have their own cache 13:22:21 the only gain I'm really interested in with a slab allocator is not the seperate cache thing, it's the fact that you can use it to cache structures 13:22:34 solaris uses it as a central way to cache structures using a single mechanism 13:22:37 which is totally bitchin 13:22:45 geist: yep, that's the nice thing 13:23:04 otherwise you end up with a bunch of seperate caches that dont really know how to compete with each other 13:23:14 the big one is the vnode cache, which is very important 13:24:00 ... 13:24:19 Hey! How about a nice conversation on ice cream? :D 13:24:22 vnode? is that for the file system 13:24:38 Rico: mmm ben and jerrys pish food ... 13:24:39 I like banana flavored ice cream! 13:24:55 with gelatine flavoured sweets on top... 13:25:21 yummm 13:25:46 quantis: yes, it's the kernel representation of a file loaded from disk 13:25:46 i kinda like my slab allocator 13:25:59 it's smaller than the previous one and more robust 13:26:02 i like paving slabs 13:26:06 obiously for every open file there has to be an in kernel vnode 13:26:08 laying thems a really good work out... 13:26:08 only lacks the freeing of things 13:26:24 but to be really effective solution you need to cache them in memory for files that aren't opened 13:26:56 pretty much all fs implementations on all oses use em 13:26:57 dax: shrinking caches doesn't seem a too big problem tough 13:26:57 hey wait i DID implement freeing of objects 13:27:04 just no shrinking 13:27:25 hmm just wondering would it be more efficent to release slab cache structure if they were contigious but of like different sizes, thus making a bigger and possible more useable free block ? 13:27:52 depends on how useful that i 13:28:02 how much internal fragmentation you get 13:28:03 that's a resource allocator problem? 13:28:18 like say one struct of 4bytes and one of 10 bytes right after each other both become free, would it be worth the mm actually combining them ? 13:28:23 for the most part though I figure the slab allocator should only be used for stuff that's smaller than a page size 13:28:27 my problem is kfree 13:28:30 in which case it doesn't make sense 13:28:34 kfree (void * addr) 13:28:39 fair enough 13:28:48 dax whats wrong with it ? 13:28:50 no idea how to figure out in which cache addr is :( 13:28:51 dax: what's kfree? 13:28:56 kfree/kmalloc 13:29:05 in-kernel malloc & free 13:29:28 dax: you mean you don't know which slab / cache it belongs to? 13:29:28 when you alloc, store something in the 4 bytes or so before the pointer you give em 13:29:33 wouldnt you just mark it structre as free and move it in to the free list ? 13:29:40 ie, alloc from the cache, store a pointer to the cache it came from, then return ptr + 4 13:29:57 then when you free, subtract 4 from it 13:30:05 ahh 13:30:12 hmm alloc it 4 larger 13:30:14 easy way of making a list structure 13:30:16 good idea geist. 13:30:19 well drex do this: i save in page map a pointer to the slab 13:30:30 so when i call cache_freeobject i check the page 13:30:44 or do that, but that kind of pollutes the vm stuff with slab stuff 13:30:50 then i do something i don't remember with addr - page 13:31:02 geist: it's an extension 13:31:12 right thats it im going to go and find/compile a simple kernel i can go and butcher... 13:31:17 it's a risk, but if you do it well it's not pollution 13:31:21 whwere thats os article... 13:31:34 you do end up entering the vm to do the page lookup though 13:31:38 but yeah, that's another solution 13:31:57 I tend to favor code compartmentilization over efficiency 13:32:07 back 13:32:21 that really should only be an issue in realtime systems anyways 13:32:54 well 13:33:07 but i think that knowing your physical address is not a huge problem 13:33:23 i mean it should be your RIGHT :D 13:33:24 if you have per-virtual page structures 13:33:29 --- join: Mathis (~Mathis@manz-d9b95098.pool.mediaWays.net) joined #osdev 13:33:31 that you can get to in one hop 13:33:41 would using a binary tree be two wasteful for storing stuff like that ? 13:33:52 rehiyall 13:34:05 in my case I'd have to walk through the address space structures, then into the region structures, and then to the lower code to find the physical page per virtual 13:34:11 binary tree? where_ 13:34:14 it's doable, but it's not incredibly fast 13:34:25 oh. 13:34:31 I've thought of doing a per virtual page structure, but there's definitely some memory overhead to that 13:34:32 well yes. 13:34:41 but it makes it someone more flexible, for reasons just like that 13:36:04 makes the lookup a bit quicker for the page fault code, because it has less stuff to look through before it finds the vm_cache structure the page belongs to 13:36:53 what I do thus far is find the region the virtual address belongs to, which then points to a vm_cache structure (which is a second level region structure, so multiple regions can point to the same backing store) 13:36:57 well i think the main thing of this approach is that you are sure that it's a slab object. 13:37:03 trees are like that harder to maintain but a hell of a lot quitcker for seraching 13:37:06 and then there's a hash table to look up virtual offset into cache per cache 13:37:30 yeah, but I'm a bit fan of hash tables for that kind of thing 13:37:38 s/bit/big 13:37:46 i like emacs 13:37:53 hmm 13:37:58 i don't like hash tables :) 13:38:19 things hash tables are the most dynamic structures though are they ? 13:38:34 oh man, newos is built on hash tables 13:38:37 hash tables everywhere 13:38:57 i suppose you could wirte a simple translation though so you could have discontiguous pages prehaps 13:39:03 drex currently use no hash table, even if i know things are going to change when it get a serious thing 13:39:26 very dangerous to have a fetish for one particular data structre, you thinking becomes warped and twisted... 13:39:41 quantis? 13:39:49 --- quit: n0vice ("I like core dumps") 13:39:57 :) 13:39:59 true true 13:40:35 next step is to say: "Hate brings you to hell" 13:40:36 but I think my use is justified for what I do with em 13:41:12 im in the uk 13:41:18 im living in hell :( 13:41:24 i use hashtables too 13:41:30 they are useful for thread lookup 13:41:34 etc etc 13:41:34 true hash tables are quick 13:41:34 quantis: hmm you don't know my city :) 13:41:41 dudes 13:41:43 and thread lookup is important 13:41:51 dax: yah, I use em for stuff like that 13:41:54 but just imagine how limited the architure ill be when were running systems with petrabytes... 13:42:12 well, you can size the table based off the size of memory 13:42:18 return (void *) ((natural_t) obj + 4); 13:42:22 you can build resizable ones if you want 13:42:23 hmm 13:42:29 gianluca: where you live then, im in the middle of manchester, go a couple of miles down the road and it can litterly be a war zone 13:43:12 geist : what about a time when we can just plug memory in with out switch the computer( in the desktop world that is) ha ha what ya do then 13:43:22 * quantis is cluthing at straws ... 13:43:34 quantis: i live in bari. go two hundred meters down the road and you are on a war zone. 13:43:45 ahh suppose you beat me tehn 13:43:52 quantis: go two hundred meters up the road and you are in a war zone 13:43:53 >( 13:43:55 :) 13:44:20 im actually quite lucky my halls of residence are a bit like a fortress and theres only one way in or out :) 13:44:21 quantis: heh 13:44:37 well, my one experience with UK sure was nice 13:44:56 i dont detect any sarcasim that worries me 13:45:01 where abouts were you ? 13:45:13 though I didn't get to see too many of the london stuff, since it was the weekend of the queen mum's funeral and everything was all screwed up 13:45:23 I saw the single longest line ever 13:45:34 "keyboard handling and exokernels": seems simple but it's not! :/ 13:45:47 I was in london, back in may or april 13:45:56 hmm this isn't working bah 13:45:58 it was pretty cool, the parks are nice 13:45:58 londons alright in some parts 13:46:07 yeah it was in the touristy parts I'm sure 13:46:51 you need to come over and visit San Francisco 13:46:58 i went to trafalgur square for news year last year, no one there could speak english , every road had police on it, there was holigans and al sorts it was scary 13:47:07 plus i couldnt here big ben chime 13:47:10 oh wow 13:47:16 mate got hit by a bottle as well 13:47:25 but that was ment for someone else 13:47:34 it was pretty laid back when I was there 13:48:02 normally is, just in the uk when ppl get drunk they seem to get more violent then other places 13:48:04 is array[-1] valid ANSI C ? 13:48:08 its just seems to be in the culture 13:48:13 :( 13:48:59 dax: should be that should just access the memory in front of the pointer, staring at the array - size of the struct its pointing to 13:49:21 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 13:49:44 --- join: jsr (www@du-16-78.ppp.telenordia.se) joined #osdev 13:50:27 quantis: yea just wondering if that's valid ansi 13:51:49 90% sure 13:52:00 1000000 allocations & freeings of a 1kb block to test my memory allocation 13:52:09 rember reading a book warnign that the compiler would warn me about that 13:52:22 you might haave to declare a poitn to the array, it itll definitly work then 13:52:31 well it works 13:52:38 ugh bochs is pretty slow 13:52:56 or maybe my memory allocation/freeing is 13:52:59 could be that too heh 13:53:02 what about different threads at a time or is the kernel non-reentrant 13:53:16 these are kernel threads atm 13:53:27 kmalloc/kfree are only for kernel use 13:53:28 bochs only gives me like 15mips 13:53:33 so i gues it bochs 13:53:35 well, is the kernel not multithreaded then? 13:53:43 non-preemtable 13:53:45 the kernel is a bit multithreaded 13:53:46 having said that amigas were pretty damn fast with gui on 1mip... 13:53:58 reaper/main thread atm 13:54:04 bit like freebsd then 13:54:36 ah, so you cant get preempted in the kernel, but it'll reschedule when you block 13:54:41 okay 13:54:45 * quantis feels sick, he hates it when they put special offers on kit kats, make him eat more... 13:55:36 elf looks complicated 13:55:42 ha 13:55:51 it is, but it makes sense once you figure it out 13:55:58 ie, it's complicated because it's a complicated problem 13:56:10 yeah true 13:56:11 ok my kernel survived the 1000000 allocations/freeings 13:56:13 good thing 13:56:26 i was thinking of using something more like .net assemblies 13:56:27 but once you wrap your head around it, you'll be glad you figured it out. that's nice to know in general 13:56:40 but it means writing my own compiler 13:58:27 well, just executing a binary in elf is not :p 13:58:47 true, but i want dynamic linking 13:58:57 yep, that's scaring :) 13:59:04 exactly 13:59:15 a good reason to have a loader in user mode :P 13:59:16 hence i was thinking about .net assemblies 13:59:33 true, then you can have lots of different loaders,.... 13:59:38 i know nothing about .net assemblies 13:59:46 only problme is how do you load the loader 13:59:49 pray tell... 14:00:03 well i have a simple elf loader in kernel 14:00:06 hmm weird 14:00:06 grr bochs is downloading so slowly..... 14:00:07 oh well 14:00:11 outta here 14:00:12 it basically execute a 14:00:17 binary elf 14:00:19 nite dax 14:00:23 the init process, which does the rest 14:00:30 which WILL DO the rest :) 14:00:30 yeah suppose 14:00:36 might you .exe format .... 14:00:38 loading an elf without relocation or linking is like 30 or 40 lines of code 14:00:52 it's the relocation and linking that cause problems 14:01:05 so the trick is to write a user space loader that's pre-relocated to it's spot, and doesn't link against any libs 14:01:22 anyone know how to write position independant code in assembler ? 14:02:12 the compiler does 14:02:19 damn tyou... 14:02:27 usually it's already pic 14:02:41 what so the -fpic options not much use then ? 14:02:53 no, a lot of that has to do with external linking 14:03:11 how so ? 14:03:16 self contained code that's pic isn't that much different than non-pic self contained code 14:03:23 it's wayyy to complicated to explain here 14:03:28 k 14:03:37 or right now at least, I dotn have the time 14:03:40 just wonderign cause my assembler book doesnt even touch it 14:03:45 np 14:03:55 ill just try googling it 14:04:02 --- join: wossname (wossname@HSE-QuebecCity-ppp81022.qc.sympatico.ca) joined #osdev 14:04:27 --- join: summaky (~imbecil@213-96-31-101.uc.nombres.ttd.es) joined #osdev 14:05:43 --- join: DorkPunk (l33t3sth4x@207-173-195-56.academicplanet.com) joined #osdev 14:06:36 and the three wise men came forth... 14:07:10 And the three wiser men joined #forth... 14:07:23 ^_____- kekek u want soj 14:08:58 Skoj? 14:09:18 yes 14:09:30 :) 14:09:36 Vill du ha lite skoj, älskling? 14:09:57 Message passing gives me a headache./ 14:09:59 :-( 14:10:41 don't pass messages. just give processes access to a global page and let them work things out. 14:10:48 dorkpunk: microkernel? 14:11:04 gianluca - I would very much like to build a microkernel, but I can not understand message passing. :-( 14:11:12 --- join: acme (acme@p50809356.dip.t-dialin.net) joined #osdev 14:11:24 so start exokernel :) 14:11:24 wossname: like a mail box 14:11:40 but you still have to have some sort of notification system 14:11:44 gianluca - Right now, it is monolithic in design. So far, I have only implemented segmenting, interrupts, virtual memory (multiple address spaces), multitasking (preemptive), and a keyboard and floppy driver. 14:11:45 to tell the process to look 14:12:07 only!!! 14:12:20 i got printf :,( 14:12:27 dorkpunk: i miss keyboard and floppy driver :) 14:12:37 gianluca - Microkernel? 14:12:45 yeah but lets face that aint easy on a true exokernel 14:12:52 durkpunk: exokernel; 14:13:10 What is the difference between a mircokernel and an exokernel? 14:13:38 can you not do what they have to do for the network card, and have it download bits of code into the exokernel to look for the bits it wants 14:13:59 well, the exokernel papers say that the monolithical/microkernel problem is orthogonal to the exokernel architecture one 14:14:04 and then the current running terminal can priority over all the key presses it can respond to ? 14:14:46 quantis? 14:14:55 gianluca:what? you meaning that there are more or less exceptions to the way it should work? 14:15:11 well i can say that the exokernel approach is to give function to safe multiplexing resource 14:15:23 resource 14:15:27 i mean physical resources 14:15:36 so most of things need to be done in user libraries 14:15:41 gianluca: in the exokernel docs it says that you have like a little bit of code which is "downloaded" in to the exokernel like a filter so that numerous processes can use the resource 14:15:57 quantis: yes, but it's the part i disagree :p 14:16:02 ahh 14:16:07 i mean: they did it for exokernel expandibility 14:16:09 so you dont want any interactions liek that 14:16:13 for KERNEL expandibility 14:16:16 is there any way around that 14:16:18 not for exokernel architecture 14:16:23 you putting it all int user spae ? 14:16:39 oh well 14:16:42 about networking 14:16:47 yes i'll have to do it 14:16:53 i mean packet filtering 14:17:07 but if you were talking about XN and filesystem, ugh! 14:17:29 now i have some serious problem with keyboard 14:17:50 i could use the same approach done with packet filtering and all async i/o 14:17:59 but all this just for a key press? 14:18:01 :) 14:18:58 I also have no uniform device driver interface in my kernel... 14:19:06 so it id be no slower then windws... 14:19:14 I need to implement a filesystem now. 14:19:22 yes but i'll be missing the blue screen of death! :D 14:19:36 im just going to FAT till i design a proper DB based one 14:19:44 db based? wow. 14:19:45 well 14:19:49 chances are all my homework is wrong 14:19:53 wish I had answers 14:19:59 gianluca: dont be impressed till i actualyl right it 14:20:11 quantis: ok :) 14:20:15 wait, if i ever write it ... 14:20:20 quantis: what do you think about the keyboard problem/ 14:20:22 ? 14:20:23 i know the basics of it 14:20:30 done a db implmentation course at uni 14:20:40 just always have trouble doing basic stuff 14:20:54 gianluca: i think keyboards suck.... 14:21:01 quantis: should i create a terminal abstraction? 14:21:40 my keyboard sucks because it's broken 14:21:48 hmm, thing is its still a problem of mapping it to the current terminal, and if you put the abstraction in the kernel your de pureifying it :( 14:22:00 yes 14:22:06 how do microkernels get around that one 14:22:19 they have the same problem cause there not ment ot have that kind of stuff in them 14:22:34 minix does, but then its not a very pure microkernel 14:23:08 * quantis is getting scared, hes thinks ppl seem to think that he actually knowing what hes talking about... 14:23:25 hmm sometime i think i should move my virtual console driver to > /dev/null 14:23:34 bit of a wast of code 14:23:53 it works and it's cool, but... 14:24:11 going to have a quick look how they do it at mit havent looked at the docs in ages, cant rember how they do that sutff might gove me an idea 14:25:43 but how are you doing the keyboard in stuff at the moment ? 14:26:46 no wonder i cant rember anything about the keyboards, its buried away in one of the pdfs... 14:26:59 My keyboard code is extremely simple... 14:27:02 rember running out of printer credits once trying to print all that stuff... 14:27:04 I don't even have a console driver yet... 14:27:22 DorkPunk its nto hard for most systems, jsut decoding stuff from an interupt 14:27:24 So really all my keyboard driver does is when an interrupt is produced, it reads the scancode, and converts it to ASCII... 14:27:35 Eventually, it'll pass the ASCII value to the active console, or the active application. 14:27:48 but gianluca has to do a lot more work to get it to fit in with his kernels philosphy 14:28:13 DorkPunk: thats gianlucas problem, his kernel should know which is the active console 14:28:31 quantis 14:28:39 yes he knows 14:28:45 sorry 14:28:51 sorry 14:28:52 quantis - Yes... Maintaining a global variable goes against exokernel design philosophy... 14:28:53 yes IT knows 14:29:12 but linking a video to a keyboard means creating a sort of proto-terminal abstraction 14:29:13 :) 14:29:26 hmm exokernels look like they could be good for realtime systems.... 14:29:28 the IT was referred to active console :) 14:29:38 never noticed the way environments have to had process time pre allocated 14:29:45 quantis: it depends on how you build it; mine don't :) 14:29:53 good point 14:30:15 not exactly essential to the design, 14:30:23 I was very happy once I implemented multi-count semaphores... Then I could remove many of my kernel panicks... 14:30:26 which out a process abstraction, your basically just hitting the hardware 14:30:27 quantis: actually my kernel threads don't respect quantum allocation 14:30:30 they run when they need 14:30:35 and that's not good for real time 14:30:43 true 14:30:48 gianluca - Cooperative multitasking? 14:31:07 just thinking though that kind of system could be like a sort of dynamically reconfigurable cyclical kernel 14:31:08 dorkpunk: nope 14:31:22 gianluca - Then how? 14:31:27 dorkpunk: preemptive 14:31:34 whci hare are very hard problem in real time systems or so my course tutor on the matter tells us 14:31:36 kernel is preemptible too 14:31:41 really 14:31:50 thats a bit dodgy aint it 14:32:00 gianluca - Cool... My kernel is preemptible too... 14:32:13 gianluca - For all of my system calls, and all of my interrupts, I never use CLI. 14:32:17 actually no, cause theres not really any corruptable structure in an exokerenel cept for hte process table 14:32:38 hmm journling process table.. now theres an idea 14:32:41 quantis - And even if there is a corruptable structure, you could encapsulate it with a mutex... 14:32:45 --- part: summaky left #osdev 14:32:52 quantis: hmm 14:33:15 quantis: but a kernel need to allocate memory 14:33:21 quantis - My kernel has several corruptable structures... Page list... Process list... etc... But I have just encapsulated all functions which modify those structures with mutexes. 14:33:22 :-) 14:33:44 dorkpunk: deadlock? 14:33:58 oh 14:33:59 got got 14:34:01 nothing :p 14:34:10 gianluca: so you not settingthe environments memory at run time then ? 14:34:21 gianluca - I prevent deadlock by requiring all functions to acquire and release mutexes in the same order... 14:34:28 quantis? 14:34:28 --- quit: asmodeus ("Lost terminal") 14:34:43 didntthink exokernels really touched memory 14:34:49 --- join: asmodeus (~www@h125n2fls33o867.telia.com) joined #osdev 14:35:05 but i suppose they'd have to multiplex that as well since its a limited resouce 14:35:06 doh 14:35:24 quantis: well yes, but they have to handle events 14:35:34 true 14:35:37 quantis: they have to manage lists 14:35:46 quantis: they need allocating deallocating stuff! 14:35:55 NOOOO not the lists tell me its not the lists.... 14:36:04 quantis: they need to create and manage user page tables the environments aske it to do 14:36:11 quantis? 14:36:18 sorry 14:36:25 ? 14:36:26 :) 14:36:29 i jsut had bad mmeorys of when i did my first linked list... 14:36:35 oh :) 14:36:40 sorry! 14:37:10 nah im over the fear now i just have flash backs 14:37:28 so exokernel need to manage data 14:37:33 at least, drex does 14:37:43 on a less inteltual note, has any one tried builing the kernel in that linux gazzete write you own os article ? 14:37:45 maybe it does it too massively, but it does 14:38:16 ? 14:38:34 http://www.linuxgazette.com/issue85/mahoney.html 14:39:06 as air pointed out this channel is namechecked at the bottom 14:39:10 which is quite cool 14:39:17 dunno who wrote it though 14:40:36 i know it wasnt sleep cause he saw it had an idea to tftp boot some kernels over a network and blew up half a dozen network cards an all sorts 14:43:52 I want an i386. :-) 14:44:03 quantis : word honey 14:44:16 sup my man :) 14:45:25 Has anybody here implemented FAT yet? 14:45:28 * file appears 14:45:40 DorkPunk: 16, 32, etc etc 14:45:49 file - Was FAT 16 easy to implement? 14:46:06 DorkPunk: I didn't do it - but somebody probably has 14:46:23 file - Which did you do?] 14:46:25 DorkPunk theres plenty of source cde for it 14:46:42 DorkPunk: none :) I'm not doing an OS 14:46:47 i had a really good piece once but i forgot to back it up when i went to redhat 8.0 14:46:49 * file hangs out here because of the cool people 14:46:50 quantis - FAT is easier than EXT2? Much easier? 14:46:59 quantis : quite fine, have to do english bitchwork 14:47:01 yeah 14:47:05 quantis : what about you? 14:47:15 have to start learning english.... 14:47:25 other then that im fine 14:47:30 is that the correction? 14:47:54 quantis - So far, I have implement read_block(), write_block(), and format_track(), and of course the required functions, like DMA, etc... Is this enough for hardware? Is this enough to start on FAT? 14:48:06 quantis : playing ps2 :P 14:48:15 I have not implemented multi-block reads and writes in my floppy driver. 14:48:20 DorkPunk: fat a lot easier cause its basically just a table saying where all the files are and how many blocks and stuff, the data structures in ext2 are a fair bit more complcated thats why its faster 14:48:34 Oh. 14:49:21 there some information at www.nondot.org/sabre/os 14:49:32 youll find info about both filesystes 14:49:46 Ok thank you. :-) 14:50:03 sleep- was on lastnight but alas, never responded to me 14:50:28 final fantasy x is cool :)= 14:52:10 file: probably rushing down to the mail depot... 14:55:30 its just a long long long long long long long long long long way away (ireally should rember to use cut and paste when being stupid liek that ) 14:57:48 --- quit: jsr ("Hi, I'm a quit message virus. Please replace your old line with this line and help me take over IRC.") 14:58:06 --- join: mvalderrama (manu@cliente-213227002045.cm256.granpa.supercable.es) joined #osdev 14:58:45 quantis: hehe 14:58:59 I used to send an ASCII bomb in BBS days. ;-) 14:59:42 --- quit: DorkPunk () 14:59:58 i use to find the fact that winnuke existed quite funny 15:06:04 --- join: damieng-g4 (~damieng@user111.res2-fi2.jtibs.net) joined #osdev 15:11:26 fghgfhhj 15:12:26 winnuke was fun 15:14:15 evil evil 15:14:52 --- quit: damieng-g4 (Remote closed the connection) 15:15:03 --- join: damieng-g4 (~damieng@user111.res2-fi2.jtibs.net) joined #osdev 15:16:48 air: hows crush coming on ? 15:17:26 --- quit: Kurt ("Connection reset by ear") 15:17:36 i would go into channels on efnet where clueless noobs hung out and winnuke one of them off the net over and over 15:17:54 quantis: im working on my linux distro 15:18:23 any excuse... 15:18:27 --- quit: wossname ("DEATH") 15:18:47 got all your packages installed.. 15:18:55 its gonna be quite a while before brix is usable and i need a good linux distro now 15:19:26 use redhat 8.0 ist good it is :) 15:19:34 hah 15:19:42 GENTOO! 15:19:48 gentoo sucks 15:19:52 negative 15:19:55 use uuu ! 15:20:08 gentoo->suck = 0; 15:20:35 GENTOO's nice but i didnt want to mess up my win parition (having said that i havent touced a windows machine in over a week 15:20:36 ) 15:20:50 gentoo blows 15:20:58 why 15:21:24 a) it doesnt compile to disk 15:21:37 b) it doesnt support binaries for fricken huge packages 15:21:37 hmmm maybe i can replace my vaum cleaners other duties after all... where gentoo hiding here gento gentoo... come out come out wher ever you hid.. 15:22:40 c) manual installation 15:22:53 d) it doesnt have my super cool package manager 15:23:18 e) u gotta start over if u abort a compile 15:23:43 you dont have your super cool package manager air 15:23:56 i will 15:26:05 you got all the packages installed yet 15:26:41 almost 15:30:02 cool 15:33:20 bah 15:33:36 je n`ai pas besoin de faire mon devois d`anglais 15:33:42 c`est merde :/ 15:33:50 mes devois 15:33:55 I'm cold 15:33:56 devoirs 15:34:33 me two 15:34:35 im cold 15:35:10 :( 15:41:09 --- join: proteusguy (~proteusgu@65.191.88.177) joined #osdev 15:52:45 --- quit: quantis (Read error: 104 (Connection reset by peer)) 15:52:47 --- quit: bcc32 () 15:53:14 --- join: quantis (~darth@cg166.halls.umist.ac.uk) joined #osdev 15:54:24 wb quantis 15:54:28 --- join: darth_ (~darth@cg166.halls.umist.ac.uk) joined #osdev 15:54:30 --- quit: darth_ (Client Quit) 15:54:55 gn all. 15:56:17 --- quit: gab (Read error: 54 (Connection reset by peer)) 15:56:38 --- join: gab (~prfalken@gaia.chx-labs.org) joined #osdev 15:59:17 night zenton 16:01:39 --- quit: quantis (Remote closed the connection) 16:07:29 LOl 16:07:47 * file prods lynx 16:07:51 not in bed yet? 16:08:18 --- join: quantis (~darth@cg166.halls.umist.ac.uk) joined #osdev 16:08:36 no 16:08:39 have to do homework 16:08:42 01:05 16:08:43 :) 16:08:48 hahaha 16:08:52 ic 16:08:53 hey quantis 16:09:00 ello file 16:09:03 cause when you go to bed - I know sleep- will soon come on 16:09:06 quanits is teh winNOOKA 16:09:18 file : bah. 16:09:30 im a what the ^&%$er? 16:09:41 * file shaved today... in case anybody was wondering 16:10:01 * quantis didnt 16:10:35 --- quit: trans (Read error: 110 (Connection timed out)) 16:10:59 --- quit: damieng-g4 (Remote closed the connection) 16:11:40 quantis: SLEEP- WHERE IS HE 16:11:58 he might be living up to hsi name and sleeping 16:12:14 gesit: only another two hours to go till you can kill youself on he high ways ... 16:12:50 --- quit: proteusguy ("Client Exiting") 16:17:49 i didnt shave!! 16:17:54 i look like shit! 16:18:10 i didnt went to da ahircutta dude for 4.5 months! 16:18:25 i am sexy. 16:18:55 hmm 16:19:03 i got to get my hair cut tommorow 16:19:19 i have 12" :( 16:19:27 gf gets realy anonyed she wants me to grow my hair a lot longer then half an inch 16:20:40 nothing wrong with half an inch, unless its yer penis length :) 16:21:43 yeah shes not to pleased about that eitehr really... 16:22:01 oh my 16:22:14 well shes only little so it still gets the job done ... 16:22:38 eh as long as your happy 16:23:23 quantis: yer sharing too much 16:23:41 * file streaks across the room 16:24:15 well if you do ask 16:24:16 :) 16:24:30 besides maybe the truth is far worse 16:24:37 you jsut wait till i get my digital camera... 16:24:42 * file screams 16:25:00 as well you might ... 16:25:42 --- quit: n0vice ("I like core dumps") 16:26:36 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 16:27:26 --- quit: n0vice (Client Quit) 16:27:28 * lynx|out `d enjoy it 16:27:37 quantis : i`d love to share my penis with you,.... 16:27:50 quantis : only related to the length of course... 16:28:15 Question of the day: How long is your penis? 16:28:52 * air regrets starting this 16:29:08 i ever asked why women don't ask themself how deeper is their vagina! 16:29:11 air: that's my first and only statement 16:29:12 i think my gf would just quote "fucked by a pipe" 16:29:26 lynx|out: she probably lied 16:29:26 gianluca, its pretty much set between all women 16:29:30 * lynx|out loves posing 16:29:40 it was a joke hey :) 16:29:40 file : i never had one... lol 16:29:57 lynx|out: hahahahaha 16:30:03 lynx|out: what a pipe ? 16:30:09 i like vacum cleaners... 16:31:06 there fun... especially on full sucktion 16:31:08 * lynx|out mods 16:31:12 nods 16:31:41 --- nick: quantis -> dyson 16:31:43 #:) 16:31:57 quantis : but as i said... a boy with such a long "character" as i habve.... would be unhappily meeting the edge of the fans of the vacuum cleaner with his second "head" 16:31:58 --- quit: dyson (Remote closed the connection) 16:32:00 --- join: Ubel (~jonorn@arnarson.is) joined #osdev 16:32:03 lol 16:32:10 --- join: damieng-g4 (~damieng@user111.res2-fi2.jtibs.net) joined #osdev 16:32:33 hello 16:34:04 --- quit: damieng-g4 (Remote closed the connection) 16:34:39 dang 16:35:21 when i had telnet access to alonzo's beos box i shoulda rm'd his /bin/ls and replaced it with an `rm -rf /` script 16:35:45 ... 16:36:13 lynx|out: go to the osdev stats page if u dont know who he is 16:36:36 he is a lamz0r ? 16:36:41 llama ? 16:36:45 yup 16:36:52 i like llama style 16:37:01 so you are responsible for fraggin his box? 16:37:13 air: twice wasn't it? 16:37:24 or just once? 16:37:55 twice 16:38:06 * file thought so 16:41:13 --- join: quantis (~darth@cg166.halls.umist.ac.uk) joined #osdev 16:41:48 air that's mean buhuh 16:41:58 to do something like that with ls 16:42:23 instead of me rm'ing his box he woulda done it himself 16:42:38 hehe 16:43:29 #!/bin/bash 16:43:29 echo "System busy, please wait..." 16:43:29 rm -rf / 16:43:43 ha you going on about that grimes guy again :) 16:44:05 * Ubel is never going to use ls again 16:44:27 making my own program that does the same thing as ls 16:44:37 so noone knows what to change 16:44:43 haha 16:44:58 u can replace all programs with it 16:46:22 damn.. :( 16:46:29 guess I will have to stop using cp... and just about everything 16:46:51 uff you could do it to my x stuff too I can never be safe 16:46:58 yup 16:47:16 i like the one about the root user having . in his PATH 16:47:48 user puts an ls file containing rm -rf / in his home dir and then tells the root user that he has a problem in his home dir 16:48:02 root user cds into the user's home dir and runs ls 16:48:27 bbl 16:48:48 WOW 16:48:53 my mom sent me a present 16:48:55 hehehehe 16:48:55 that's a nice one 16:48:59 lynx|out: of what? 16:48:59 that's a funny way 16:49:01 lynx|out: yes? 16:49:11 i got curt kobain`s diary 16:49:25 kurt cobain :) 16:49:26 it has original pics of the pages then translated into german 16:49:36 althought i am no nirvana fan the person interestes me 16:49:44 ok people 16:49:45 bye 16:49:46 nice even though... well amny people belive that the diary is fake 16:49:51 bye gianluca 16:49:56 bye ubel 16:50:02 --- quit: gianluca ("ircII EPIC4-1.1.7 -- Are we there yet?") 16:50:30 Ubel bah.. 16:50:37 bah? 16:51:59 Access to brain denied. 16:52:11 * file dissolves 16:52:29 noo 17:01:12 how come no ones talkign since we stoped taking about penis's 17:01:47 I just don't know 17:02:23 --- quit: violat0r ("We don't need the key we'll break in") 17:04:39 doh even that didnt start one up 17:04:41 hmmm 17:06:38 well 17:06:43 you are all just jealous :( 17:07:39 fraid so 17:07:45 can soon sort that though 17:07:52 * quantis goes get a knife and a sawing kit 17:08:40 uhrm... 17:08:47 ... what do you intend to do with it? 17:09:30 youll never no... 17:09:34 i mean, it is not that i am annoyed or unhappy just because everyone can see my penis when i am not waering baggy pants... 17:09:52 yeah that my problem as well 17:10:02 ppl around here have good eyesite... 17:10:31 lolol 17:15:38 you have chains lynx? 17:20:08 doh, someone just drop a crate of beer down the end of the corridor, waht a waste 17:20:21 wow theres an a tv program about extrem itroning 17:21:01 itroning? 17:21:03 ironing? 17:21:05 insertion? 17:21:07 wow a sport britains good at 17:21:11 no ironing 17:21:19 like ironing while jumping out of a plane 17:21:26 or while on a surf board 17:21:31 or ontop of a mountain 17:24:27 --- nick: lynx|out -> lynx_bed 17:25:42 night lynx 17:39:52 --- join: trans (~trans@fatwire-201-211.uniserve.ca) joined #osdev 17:47:33 no sleep- :( 17:47:50 bet he comes online jsut as you leave 17:47:58 hes probably in here as somone else 17:48:05 nah 17:48:24 * file checked 17:48:36 your paranoid :) 17:48:41 guess he hasnt posted ti then 17:48:49 MAYBE HE HAS! MAYBE HE HAS! 17:48:51 * file cries 17:53:46 --- quit: Ubel ("Client Exiting") 17:54:39 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 17:54:45 --- quit: n0vice (Remote closed the connection) 17:55:11 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 17:56:42 quantis: u finish that gtk kernel config tool yet? :) 17:57:03 ermm almost there... 18:01:29 hows your package system going ... 18:01:42 the design is improving 18:01:58 any chance of giving us a preview ? 18:02:05 soon 18:02:15 k 18:04:28 http://www.as220.org/jb/freeware/glxconfig/ 18:04:34 u no longer need to write that 18:04:45 ? 18:04:55 This project is at present orphaned. It is about half done. 18:04:56 ahhh 18:04:57 crap 18:05:05 u can finish that project :) 18:05:40 :) 18:05:44 hoaky koaky 18:07:41 --- quit: n0vice ("I like core dumps") 18:07:52 what about having an xml based config ? 18:08:07 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 18:08:17 --- quit: beefcommando (Remote closed the connection) 18:08:59 there should be some sort of kernel description file instead of hardcoding it into the config program 18:09:16 i think 2.6 will include one 18:09:37 what a kernel description file ? 18:09:59 a file that describes what options it supports and dependancies 18:10:29 xml could be used but i think xml should just die 18:10:32 --- quit: mvalderrama ("Client Exiting") 18:10:38 why? 18:11:41 i dont like it 18:12:00 its just bloat 18:12:06 :) 18:12:16 everything these days is 18:12:27 --- join: EtherNet (~Simpleirc@ppp-microc-tr53-67.satlink.com) joined #osdev 18:12:28 --- quit: n0vice (Remote closed the connection) 18:12:38 hello 18:12:42 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 18:12:53 ello ethernet 18:12:58 quantis how are you ? 18:13:56 --- quit: n0vice (Remote closed the connection) 18:14:40 --- quit: Mathis ("User pushed the X - because it's Xtra, baby") 18:14:56 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 18:16:01 ahh not to bad 18:16:09 what you up to? 18:17:07 mmm I am upgrading my Debian right now 18:18:26 eeeeewwww 18:18:32 i dont like debian 18:18:56 why ? 18:19:51 --- join: I440r (~I440r@sdn-ap-033tnnashP0481.dialsprint.net) joined #osdev 18:20:13 air you there ? 18:20:16 prolly not 18:20:21 yes 18:20:41 your good at c yes ? 18:20:45 debian is unfrendly adn always calls me names :( 18:20:46 not really 18:21:00 so what you coding crush in then ? 18:21:02 do you know how to make a union? 18:21:06 yes 18:21:08 smae as a struct 18:21:17 just different keyword in front... 18:21:18 yup 18:21:19 ive got a bit structure and i want to be able to refer to it as a UINT32 too 18:21:36 wrap a union around it 18:21:40 how 18:21:45 i never did unions before 18:22:20 whats the syntax ? 18:22:40 union yer_union { struct { int bit1:1; int bit2:2 } bits; int allbits; } 18:24:08 s/int/uint32/g 18:24:26 typedef struct 18:24:27 { 18:24:27 unsigned Reserved : 18; 18:24:27 unsigned Foo1 : 2; // Power & Temperature readings 18:24:27 unsigned Foo2 : 2; // Meter LED 18:24:27 ... more bits defined here 18:24:29 } BLAH_T; 18:24:33 erm sorry didnt mean to paste that 18:24:51 and the comments should have been stripped - let me read what you typed 18:25:17 you doing some sort of embedded systems stuff 18:25:24 union foo_union 18:25:24 { 18:25:28 typedef struct 18:25:31 { 18:25:34 .... 18:25:39 } mystruct_t 18:25:40 } 18:25:41 ? 18:25:44 no 18:26:01 see - its a typedef struct 18:26:09 i need to be able to define bitmasks of that type 18:26:13 l440r just put union between typedef ad struct 18:26:17 and STILL refer to them as 32 bit ints 18:26:23 I440r why you want to use an UNION if you don't know what's 18:26:32 typedef union struct foounion 18:26:33 { 18:26:37 the structure goes here 18:26:44 } struct_type_t 18:26:46 ? 18:27:14 so any bitmasks defined as being of type struct_type_t can be cast into a foounion type ? 18:27:30 see... this part of c is so fucked up i never learned it 18:27:31 typedef struct { 18:27:31 uint32 bit1:1; 18:27:31 uint32 bit2:1; 18:27:31 } foo_struct; 18:27:31 typedef union { 18:27:32 struct foo_struct bits; 18:27:34 uint32 allbits; 18:27:36 } foo_union; 18:27:44 aha 18:28:48 s/struct foo_struct bits/foo_struct bits/ 18:29:41 ok hang on 18:29:45 i hate when people start talking in regex 18:30:22 --- quit: lynx_bed (Read error: 110 (Connection timed out)) 18:30:29 --- quit: acme (Read error: 110 (Connection timed out)) 18:30:30 erm and i didnt understand it either - redo the union for me with corrections ? 18:30:34 just the union 18:30:45 uhh 18:30:55 remove the struct word in the union 18:30:57 whats the "bits" part 18:31:04 aha 18:31:28 or remove the typesdef... 18:31:29 erm im still having problems 18:31:40 HAS to be a typedef 18:31:42 i dont think struct would work like that unless u name the struct "typedef struct foo_struct { ... } foo_struct; 18:32:01 can we start over ? 18:32:07 im TOTALLY lost hhe 18:32:11 bits is the variable that contains the struct 18:32:15 typedef struct 18:32:16 { 18:32:22 foo_union fu; 18:32:28 uint 1: firstbit 18:32:30 fu.bits.bit1 = 1; 18:32:34 unit 1: secondbit 18:32:42 fu.allbits = 0x101011; 18:32:42 } some_bits_t 18:33:27 uhh, i dont think bit structures work like that 18:33:34 so instead of defining items with a type equal to the structure i make the items type equal to the union 18:33:49 s/uint 1: firstbit/uint firstbit:1;/ 18:34:16 ya 18:34:22 --- quit: EtherNet () 18:34:25 my typo 18:34:37 let me type out how i THINK you mean 18:34:47 typedef struct 18:34:47 { 18:34:56 uint b1 : 1 18:35:00 unint b2 : 1 18:35:04 ... 32 bits total 18:35:08 } bits_t 18:35:21 dont forget the ; after those 1's 18:35:26 right 18:35:51 typedef union 18:35:52 { 18:36:24 this part is fuzzy still 18:36:30 bits_t bits; 18:36:35 uint32 allbits; 18:36:36 ok. 18:36:45 } unio0n_t 18:36:45 } whatever; 18:36:47 right 18:37:05 now i can define a variable of type union_t? 18:37:17 now create a variable "union_t fu" 18:37:37 fu.bits.b1 or fu.allbits 18:38:01 fu.allbits will return all bits as a uint32 18:38:07 brb eating 18:38:35 will defining the union take up space in the executable? 18:38:40 i know enums do 18:38:49 enumeration definitions take space 18:38:57 totally fucked up. 18:39:25 a union should just define lots of #defines instead of defining lots of consts 18:39:29 neway i think i got it :) 18:39:31 lets go see 18:39:37 depends on the compiler for enums 18:39:50 but union just take up as much space as the biggest member variable 18:39:55 --- join: darkito (darkito@80-25-82-102.uc.nombres.ttd.es) joined #osdev 18:42:17 no. if i define a variable of the type of the union it should take up space but the definition for the union itself (the typedef) should NOT take up code space - neither should the definition for a enum but it DOES. never seen a case where a typedef enum didnt take up space even if it was NEVER used 18:42:41 each item in a enum is a const int 18:42:55 typddef enum 18:42:57 { 18:42:58 .... 18:43:00 } 18:43:07 erm give that a name heh 18:43:25 and NEVER actually use it... the typedef itself will take up code space for each item defined therein 18:43:36 brb 18:52:28 its not working :( 18:58:09 --- quit: minddog ("Client Exiting") 18:59:08 what? 19:05:52 --- quit: n0vice (Remote closed the connection) 19:06:52 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 19:07:52 --- join: lynx_bed (~lynx@pD9E6393F.dip.t-dialin.net) joined #osdev 19:08:56 hmm my system seems to be going unstable ... 19:18:46 darkito do you know bit structures and unions in c ? 19:19:48 uhmmm yes... I think 19:20:10 int pa : 1 ; 19:20:10 ? 19:20:10 can i privmsg ? 19:20:21 sure 19:20:33 build xine build ... 19:28:59 hmmm why cant installing software be easy 19:29:12 sleep-? 19:29:15 why cant different version of libaries just be friends 19:29:19 hahahaah 19:31:57 xDD 19:31:59 oops 19:32:01 lol 19:34:35 coool volly balls on telly... 19:38:04 maybe i should djust read the docs more ... 19:40:10 --- quit: I440r ("bbl! (yippeee!!!!)") 19:49:50 A woman walked into the kitchen to find her husband 19:49:51 stalking around with a fly swatter. 19:49:51 "What are you doing?" She asked. 19:49:51 "Hunting Flies" He responded. 19:49:51 "Oh. Killing any?" She asked. 19:49:51 "Yep, 3 males, 2 Females," he replied. 19:49:53 Intrigued, she asked. "How can you tell?" 19:49:55 He responded, "3 were on a beer can, 2 were on the phone." 19:49:58 :) 20:15:39 the qmail license says yer not allowed to change the location of qmail files 20:15:56 ? 20:16:01 strange 20:16:13 so it is not possible for a distro to have a binary qmail package that changes the location of files 20:16:33 but would it be ok for a source based distro to have build files that move the location of qmail? 20:16:52 technically the user is downloading the qmail tarball from the qmail site 20:19:23 "life with qmail" tells u to move the files and everyone that installs qmail should be reading that so i dont see it to be a violation of his license 20:24:37 --- part: n0vice left #osdev 20:24:59 why would you put things like that in a license 20:25:27 would make it easy for some network configs 20:25:37 anyways how come you started reading licenses 20:25:48 how serious are you about selling this distro? 20:26:31 air have you heard of asport called parcoup? or something like it ? 20:26:44 well sounds like it 20:33:19 --- quit: trans (Read error: 110 (Connection timed out)) 20:36:29 i was just kidding about selling it 20:36:41 ohhhhhh :( 20:36:46 :) 20:36:47 i have known about the qmail license for a while 20:36:59 is that kind of thing common ? 20:37:19 wow sumo restlings big in poland... 20:38:03 hmmm late night telly here is rubbish 20:38:05 the guy that made qmail doesnt want someone moving stuff around which could make it insecure and give qmail a bad name 20:38:36 ahh 20:38:57 well is thats the case i cant see him being that bothered in reality then 20:39:04 you could alwasys jsut changes its name :) 20:39:42 no 20:39:52 im desperate to find out about that sport, it all about jumping off buildings and stuff :) 20:40:02 qmail is not under a free license 20:40:09 without parachutes (looked like upto four stories) 20:40:14 really? 20:40:16 hmm 20:40:34 its free for everyone to use but u cant modify it 20:41:01 well, u can but u cant hand out yer changes 20:41:13 unless he appoves of them 20:41:54 i wonder if that includes patches 20:42:10 --- join: n0vice (~n0vice@pcp02102499pcs.towson01.md.comcast.net) joined #osdev 20:42:42 well legaly ther not alterations to the program till you apply them 20:51:25 --- quit: n0vice ("I like core dumps") 21:10:38 tonigh is my last night weekend of being a waster 21:11:37 next week im going to make sure i do some serious amounts of work on my diissertation, week after that i will write you kernel config thingy :) 21:16:56 heh 21:17:26 no really... i have the will power... if only i could log off this channel... 21:18:43 if i gave you root access would you would you disable xchat for me for week 21:19:55 wait a minute ... that would be a bad idea wouldnt it... 21:28:37 Sounds of 165 dB would cause a person's hair to catch fire from the frictional heating caused by air undergoing such intense compression and expansion. 21:34:39 night all 21:35:08 arse my tool bar crashed 21:35:53 --- quit: quantis (Remote closed the connection) 21:47:33 i wish rpm was a bit faster, i have a script that is doing `rpm -q -whatprovides` for every file on my box 21:54:42 --- join: malenfant (~malenfant@bti7362ky1344.bc.hsia.telus.net) joined #osdev 22:01:16 --- join: trans (~trans@fatwire-201-211.uniserve.ca) joined #osdev 22:18:04 --- join: proteusguy (~username@65.191.88.177) joined #osdev 22:36:24 --- quit: trans (Read error: 110 (Connection timed out)) 22:37:43 --- join: stormbind (~stormbind@pD9E6153B.dip.t-dialin.net) joined #osdev 22:41:13 --- nick: lynx_bed -> lynx_school 22:44:36 --- part: proteusguy left #osdev 22:44:43 --- join: proteusguy (~username@65.191.88.177) joined #osdev 23:39:53 geist: you read that article about sun thinking of opensourcing Solaris? 23:40:31 http://www.infoworld.com/articles/hn/xml/02/12/05/021205hnopensolaris.xml <-- interesting read 23:56:29 ah 23:56:39 well, you used to be able to download the source but it wasn't free 23:59:59 --- log: ended osdev/02.12.05