00:00:00 --- log: started osdev/03.05.13 00:20:24 --- quit: zhware (Read error: 104 (Connection reset by peer)) 00:20:25 --- join: zhware_ (~zhware@daidomon.com) joined #osdev 00:21:43 --- quit: zhware_ (Client Quit) 00:22:09 --- join: zhware (~zhware@daidomon.com) joined #osdev 00:23:45 --- quit: Gila (Killed (NickServ (Ghost: gila_!~gila@fia160-4.dsl.hccnet.nl))) 00:23:57 --- join: gila_ (~gila@cc16711-a.delfz1.gr.home.nl) joined #osdev 00:24:23 --- join: Gila (~gila@fia160-4.dsl.hccnet.nl) joined #osdev 01:17:10 --- join: trans (seqkwn@fatwire-201-205.uniserve.ca) joined #osdev 01:37:40 --- join: Mathis (irc@pD9EA96BD.dip.t-dialin.net) joined #osdev 01:37:52 hi 01:45:45 Hey Mathis 01:48:34 Hey Robert 01:54:50 --- join: revanthn (hla@202.9.183.151) joined #osdev 01:55:53 --- quit: trans (Read error: 110 (Connection timed out)) 01:57:31 --- quit: minddog ("adios") 01:58:04 --- quit: air ("cria 0.2.8cvs15 -- http://cria.sf.net") 02:17:23 --- quit: revanthn () 02:21:57 --- quit: Mathis ("good night") 02:34:46 --- join: minddog (~minddog@fleenode.net) joined #osdev 02:47:08 --- quit: Phactorial (Read error: 60 (Operation timed out)) 03:17:17 --- join: trans (bhwltx@fatwire-201-205.uniserve.ca) joined #osdev 03:21:50 --- join: darkito (~darkito@80-25-82-102.uc.nombres.ttd.es) joined #osdev 03:21:51 --- join: sayke (~abuse@sttldslgw29poolF187.sttl.uswest.net) joined #osdev 03:22:08 hola 03:22:29 * sayke collapse 03:22:39 bah wrong chan 03:22:41 not that it matters 03:22:56 :) 03:22:59 Hey darkito 03:23:09 Robert ! 03:23:11 :D) 03:23:19 xD 03:23:22 How's my darkito? 03:23:27 heh 03:23:42 doing some homeworks 03:23:48 so not too fine 03:23:49 :P 03:24:14 --- join: mur (murr@baana-62-165-187-42.phnet.fi) joined #osdev 03:24:34 hello 03:25:03 hello mur my snowy friend 03:25:04 :P 03:25:44 Murr! 03:25:54 Robert : and you? 03:26:07 Well.. I've got a class in about 2 hours. 03:26:11 But otherwise, just fine :) 03:26:12 :) 03:29:50 >:) darkito, Robert 03:29:54 * mur is sick :( 03:30:05 :\ 03:30:15 I hope you get getter better :\ 03:30:25 s/getter// 03:30:44 * mur saw weird dreams last night >:P .. almost like hallucinations 03:31:39 Oh, you've got some experience of that! 03:37:36 --- join: file (intranet@mctn1-7636.nb.aliant.net) joined #osdev 03:38:06 Hey file 03:38:23 /home/file 03:40:05 hi 03:40:42 yawn 03:40:46 hiya folks 03:40:46 Morning 03:40:54 heh, actually I did just wake up 03:40:57 it's 3:40 am here 03:40:58 hello geist 03:41:05 Oh :( 03:41:07 fun fun! 03:41:23 Insomnia, or do you have to go somewhere this early? 03:42:27 I went to France this weekend 03:42:41 got back yesterday afternoon, I'm still halfway on France time 03:43:01 I'm not fighting it too much, I'll just go into work really early 03:45:00 You go to another continent, for a _weekend_? 03:45:06 well 5 days 03:45:09 long weekend 03:45:31 Oh, OK. 03:52:06 --- quit: mur ("foo") 04:11:24 woOoOoOoo 04:20:23 yep 04:28:34 --- quit: trans (Read error: 110 (Connection timed out)) 04:30:53 --- join: kyelewis (~w00t@203.221.42.101) joined #osdev 04:30:58 hey kye 04:31:04 heya mors 04:31:22 hiya folks 04:31:29 heya geist 04:31:30 hey geist 04:36:59 soooo.... what's happening? 04:37:38 I need to get ready and go to work in a bit 04:37:54 though it's 4:30 am, but I'm awake so I may as well use it as an opportunity to get some work done 04:38:02 hehe 04:40:56 --- join: wl (philipp@p50865E83.dip.t-dialin.net) joined #osdev 04:56:27 lol stupid junk mail 04:56:49 04:56:50 I always remember my mother telling me... 04:56:50 "Mark, if Brian (my best friend as a child) jumped off a bridge, would you do it too?" 04:56:50 Well, obviously not- But, if my best friend was following in the footsteps of Thousands of other Internet Entrepreneurs, using a simple system to make a killing online... 04:56:52 04:57:13 04:57:14 I KNOW I'D DO THAT... 04:57:14 In fact, I've already done it! 04:57:16 04:57:23 (sorry about the extra snips :P ) 04:58:11 --- nick: kyelewis -> kyelewis_z||z 04:58:42 * kyelewis_z||z is away: yeah, yeah, you know the drill... sleep and all that - and i know you're thinking "pfft... who needs sleep" :P 04:58:53 --- join: Kurt (~kmw@oc-nas-11-s193.cinergycom.net) joined #osdev 04:59:13 night geist, mors, all 05:04:23 night 05:09:26 hi all 05:09:43 hi geist, though asleep := 05:09:47 :) 05:09:51 --- quit: Kurt ("Connection reset by rear") 05:10:03 oops 05:10:17 kyelewis was the one going asleep := 05:10:19 ah 05:10:21 :) 05:11:16 heh 05:24:13 --- join: mrMister (~andri@ti122110a080-0648.bb.online.no) joined #osdev 05:26:03 Hei 05:26:11 Nu går jag till skolan.. 05:26:13 See you :) 05:26:24 lol 05:26:46 see you, Robert 05:26:46 jah. 05:28:32 skolangen 05:28:49 hehe 05:28:56 barneskole? 05:52:49 --- join: trans (vhpzdx@fatwire-201-205.uniserve.ca) joined #osdev 05:55:00 --- quit: sayke (Read error: 113 (No route to host)) 06:01:37 --- join: sayke (~abuse@sttldslgw29poolF193.sttl.uswest.net) joined #osdev 06:03:48 --- quit: darkito ("Connection reset by Michael McGreen") 06:22:15 hm 06:24:09 hm? 06:26:58 yeah 06:31:28 hmmm 06:32:01 * lodda listens to FFVII music 06:33:59 --- join: wl_ (philipp@p50865B43.dip.t-dialin.net) joined #osdev 06:41:30 it is a wonderful day 06:42:07 awesome! 06:42:15 it was a nice ride in this morning 06:43:27 :) 06:44:33 --- join: kemu (~jonas@183.150-136-217.adsl.skynet.be) joined #osdev 06:45:39 muprhy is against me 06:46:22 murphy 06:46:53 * debug was supposed to do a message passing thingy on the school machines during the weekend, but obviously they had to reinstall the machines this particular weekend 06:47:37 heh 06:47:40 the deadline is still a day or two from now, but it feels like Murphy will come up with something else in time for that 06:48:03 of course it is a woderful day! 06:48:07 :) 06:49:41 how's everyone doing? 06:51:09 fine basicly, if i hadn't tomorrow three lessons with a teacher, i dislike :/ 06:53:10 --- quit: wl (Read error: 110 (Connection timed out)) 06:53:33 --- join: quantis_work (quantis@cg166.halls.umist.ac.uk) joined #osdev 06:54:28 ello 06:54:54 yo 06:58:38 hows you 06:58:59 it is a wonderful day 06:59:09 * debug fills up another glass 06:59:14 (coca cola :-) 07:00:26 coca cola is loopey juice.... 07:02:15 --- quit: nutrageous ("Sic Semper Tyrannis!") 07:03:12 --- quit: green_ (Read error: 104 (Connection reset by peer)) 07:03:45 --- join: Ishq (~Ishq@frm-64-4-102-202.access.ntelos.net) joined #osdev 07:03:47 --- nick: Ishq -> nutrageous 07:04:04 --- nick: nutrageous -> Ishq 07:04:19 you alright ishq? 07:04:23 --- join: bono (~bono@modemcable027.101-200-24.mtl.mc.videotron.ca) joined #osdev 07:04:26 --- nick: quantis_work -> quantis 07:04:29 this script acting crazy 07:04:39 never botehre with them 07:04:49 how goeth your schoolith 07:05:23 im at uni 07:05:28 supposedly revising for exams 07:05:38 fun fun 07:06:30 Ive got an OS related question... In a microkernel design, are the servers allowed a static amount of memory, or can they ask for more (sbrk) like any other process? 07:07:47 most servers should be working in user spae 07:07:55 so they can ask for more memory if they want 07:08:39 bono: you designing an os? 07:09:16 yes, another one! :-) 07:09:49 cool 07:09:56 quantis: servers in user space... many pple put them in ring 2, whereas user space processes reside in ring 3... 07:10:34 thats pretty ineffcient to be honest 07:10:42 and i havent heard of anyone actualyl doing it 07:10:59 thing is you then have to sets fo protection to go through for nearly every call 07:11:29 true 07:11:49 most services just provide wrappers for the underlying kernel, especially memory management, doing that means every call you make to a service involves to protection traps (or whatever its called when you cross levels) 07:12:24 plus generally the service ill be fairly safe from memory areas if you code the protection right in the first place 07:12:52 are qnx and others designed with only 2 protection levels? 07:13:01 dunno about qnx 07:13:09 but just about every os ive looked at is 07:13:17 even NT and linux 07:13:27 i mean, microkernels... 07:13:41 NT was to begin with 07:14:13 but they noticed while stability was good, performance was too poor compared to all the monolitihic unix kernels they were tryign to compete with 07:14:18 now its sort of a hybrid 07:14:34 and how do you guys know about windows internals? its scheduling policy? etc... where is it documented? 07:14:54 there are a couple of books (i havent read about them) 07:15:14 what i know is from old BYTE magazines which used to be pretty technial 07:15:20 and explained pretty well 07:15:39 ill try and find the book on amazon for you wait up a min 07:16:26 was just wondering... dont bother... 07:16:37 --- join: lodsb (~lodsb@p50809820.dip.t-dialin.net) joined #osdev 07:16:48 hey lynx 07:16:50 arggh it lynx!!!! 07:16:53 Hey. 07:17:05 rob: hows teh avr's going ? 07:17:29 Not too bad... I've been working a little on a Forth compiler to be run inside the VM. 07:18:26 cool 07:20:34 shpulrnt take to long 07:20:41 ive seen some forth compilers that are like 8k 07:23:10 That big? ;) 07:23:11 which should fit quite niclely in and avr memory 07:23:25 i was going to say 4k but that was only on 8bit machines.... 07:23:44 Depends on how good you want to make it.. 07:24:04 Simple and basic one should take more than 1kB, but if you want some library functions, a bit more :) 07:26:15 really!!! that small!!! 07:26:24 hmmm 07:26:39 * quantis starts contemplating an object orientated forth.... 07:28:15 quantis: am currently reading taunenbaum's book... minix uses all of the 4 layers... 07:28:32 * Robert pokes Minix. 07:28:35 really? 07:28:42 * quantis reaches for his copy 07:28:59 quantis: page 94 of the 2nd edition 07:30:18 arse 07:30:23 i never noticed that 07:30:28 could be why it so slow 07:32:59 i dont think thats how its actually implmented (though im probably wrong) 07:34:53 quantis: in a microkernel, are the processes of equal level aligned, followed by all processes of the next level, etc... or are all processes of all level mixed up? 07:35:29 ("followed by" in the memory alignment sense) 07:38:00 think its up to the implmentor 07:38:19 it normally better to handle memory protection in software 07:38:44 risc chips just throw exceptions and traps, and let the os provides the algorithms 07:38:55 intel have slightly higher level mechanism 07:42:36 --- join: kernel2420 (1000@ppp-159-166.25-151.libero.it) joined #osdev 07:43:38 --- quit: kernel2420 (Client Quit) 07:43:46 --- join: kernel2420 (1000@ppp-159-166.25-151.libero.it) joined #osdev 07:44:57 --- quit: kernel2420 (Client Quit) 07:56:19 bono: I'm a bit late 07:56:34 but virtually all modern operating systems only use 2 protection levels 07:57:11 * quantis was starting to sweat a little there that hed got something drasticaly wrong 07:57:41 yeah you were pretty close 07:58:02 a little noncommital, but thats okay 07:58:22 :) 07:59:54 geist: even moderm microkernel-designed OS? 08:00:03 --- quit: Gila (Remote closed the connection) 08:03:36 --- nick: lodsb -> lynx__ 08:07:46 yes 08:07:56 bono: no other modern cpu has more than 2 levels of protection 08:08:11 and virtually all modern operating systems are cross-platform 08:08:25 therefore you can't design one around >2 protection levels 08:09:15 what you saying MULTICS isnt modern, withs it 8levels!!!1 08:09:48 no, that was eight priority levels I think 08:10:22 thats what i said wasnt it? 08:10:34 it had eight rings on its system 08:10:44 but they only ever really used 3 or something 08:10:47 we were talking about priviledge levels at the cpu 08:10:56 not scheduler priorities 08:11:01 multics and its hardware were pretty united 08:11:07 could be 08:11:10 no the hardware was modified 08:11:13 anyway, point is it's all 2 now 08:11:20 there is very little use of > 2 08:11:24 dont forget they were all using transistors back in the early 70's 08:11:38 I can't really think of a way to use >2 that would really be worth the trouble 08:11:44 geist: there would be use, but the performance hit isnt worth it 08:11:54 multiple system images? 08:12:04 even that, no one has ever come up with a reasonable use of it 08:12:31 I've had many people try to find a problem that fits the feature 08:12:32 suppose 08:12:45 instead of the other way, which is to have a problem that would be solved with >2 08:13:14 like the SAS systems ? 08:13:41 in what way? 08:14:55 not i ment with less then 2 levels 08:15:06 Single Addrees Space systems 08:38:51 --- quit: quantis (Read error: 54 (Connection reset by peer)) 08:43:14 Question-> what gcc version should i be using to develop my OS? 08:43:35 the devel tree? the latest stable? ? 08:44:05 write your own C Compiler ;P 08:56:51 --- quit: trans (Read error: 110 (Connection timed out)) 08:59:23 --- quit: Boney ("Sleep") 09:04:55 --- quit: wl_ ("Quit") 09:05:40 --- join: SLACKo (~SLACKo@62.114.149.214) joined #osdev 09:08:14 --- quit: SLACKo (Client Quit) 09:16:58 --- quit: lynx__ ("bong timeout") 09:24:17 --- join: SLACKo (~SLACKo@65.199.128.66) joined #osdev 09:27:30 --- join: sxz (~sixer@212.59.18.2) joined #osdev 09:27:33 hi 09:29:29 any ideas why all linux tlb flush defines are like: do { } while(0) ? 09:30:02 sxz: kernelnewbies.org answers that question... 09:30:13 sxz: (no offense) ;-) 09:30:18 heh 09:30:19 :) 09:30:29 kthnx anyway 09:30:52 it's so the macro works "properly" 09:30:53 --- join: WeleaseBwian (WeleaseBwi@133.43-200-80.adsl.skynet.be) joined #osdev 09:31:32 mOO 09:32:55 LOEHL 09:33:18 hehe 09:33:31 JEAN 09:33:36 two belgians, cute 09:33:36 'allo 09:33:45 bonjour madame 09:33:59 je ne suis pas un belge 09:34:07 heh 09:34:14 je suis né en france 09:34:28 *urinates on the keyboard* 09:34:34 lol 09:34:38 do so 09:34:45 then type on it 09:34:52 i did 09:35:44 --- quit: WeleaseBwian () 09:35:51 *g* 09:42:32 --- join: wl (philipp@p50865B43.dip.t-dialin.net) joined #osdev 09:46:07 --- quit: SLACKo ("[x]chat") 10:11:21 --- join: silvio (1000@ppp-142-161.25-151.libero.it) joined #osdev 10:11:52 --- quit: silvio (Client Quit) 10:15:19 --- join: trans (xulsfx@fatwire-201-205.uniserve.ca) joined #osdev 10:20:52 --- quit: dh ("Fhtagn-Nagh Yog Sothoth") 10:28:26 --- quit: Divine (Read error: 110 (Connection timed out)) 10:38:22 Ive got problems linking my C++ kernel (compiled with g++ 3.2)... I get "undefined reference to `__gxx_personality_v0" . What am I forgetting? ANyone? 10:43:43 nope 10:43:53 define a function somewhere 10:44:00 void __gxx_personality_v0(void) 10:44:54 if you dont use C++ and exceptions, it'll never get called 10:45:57 --- quit: kemu ("Lost terminal") 10:48:39 you were right on... -fno-exceptions solves the problem! :-) 10:50:12 geist: should i be lonking my kernel with ld or with g++? is g++ only a wrapper to ld? 10:52:33 use ld 10:52:45 though gcc if it sees a bunch of .o files on the command line will figure out what to do 10:52:53 gcc is a wrapper to lots of things, ld is one of them 10:53:02 but in general doing it in seperate steps is a little clearer 11:01:07 --- quit: minddog ("leaving") 11:01:10 --- join: minddog (~minddog@fleenode.net) joined #osdev 11:01:22 --- nick: minddog -> minddog[away] 11:10:45 --- join: Divine (~john@12-246-112-182.client.attbi.com) joined #osdev 11:12:33 --- join: kreep (~Mental@65.169.21.243) joined #osdev 11:13:18 howdy 11:22:50 --- join: mur (murr@baana-62-165-189-56.phnet.fi) joined #osdev 11:22:50 --- quit: wl (Read error: 104 (Connection reset by peer)) 11:23:48 * mur ! 11:24:40 bbl 11:25:31 am i that scary? :P 11:25:47 lol 11:25:51 not by far ;) 11:26:03 mur: yes 11:26:14 do you know how hard it is to get 386 manuals in hardcopy 11:26:28 you order them at intel.com 11:26:37 * lodda is waiting for them since 2 weeks 11:26:38 you cant 11:26:47 --- join: wl (philipp@p50865B43.dip.t-dialin.net) joined #osdev 11:26:51 386 manuals are "obsolete" 11:26:55 according to intel 11:27:09 ugh 11:27:10 but they are still in high demand 11:27:15 why do you need PAPER copy of them? 11:27:27 the guy i talked to said in just that day they had 17 thousand requests 11:28:22 i asked if they are so requested, why not keep it in print, and the guy said because they are 10 years old... 11:29:24 and mur, because i want it in hardcopy :) 11:30:34 why = 11:30:35 ? 11:31:11 because of bad intercade design and poor monitor quality? :P 11:31:19 interface 11:31:36 no 11:31:43 because i like it in paper form 11:31:59 why? 11:32:16 because i do 11:32:28 there is no "why" to that answer 11:35:53 --- quit: Divine (Read error: 104 (Connection reset by peer)) 11:36:49 --- join: Divine (~john@12-246-112-182.client.attbi.com) joined #osdev 11:42:38 --- join: Zenton (~vicente@8.Red-80-34-35.pooles.rima-tde.net) joined #osdev 11:43:41 Xerroz: you could print it out yourself :D 11:44:01 Xerroz: are you writing an OS for only 386? 11:45:16 no 11:45:16 http://www.choppingblock.org/d/20020816.html 11:45:28 i want the 386 manuals so i know where to start 11:45:53 sigh, another 'I need the manual printed out' person 11:45:56 then build optimizations for higher level processors 11:46:00 I went through that phase 11:46:11 * debug reappears 11:46:18 i dont need it printed, i would prefere it printed.... 11:46:25 half of a 10 week OS course done in 3 hours 11:46:29 * debug is happy 11:46:29 order one from intel 11:46:34 and use the pdf's until then 11:46:38 or print out a chapter at a time 11:46:49 i dont even own a printer 11:46:54 you want the most recent ones for sure 11:47:01 it'll fix errata in older ones 11:47:16 im building my os on a 486 11:47:50 and? 11:48:02 so a p4 manual wouldnt do me any good 11:48:08 yes it would 11:48:11 get the most recent one 11:48:17 skip the chapters that are p4 specific 11:48:22 you really think things have changed that much? heh 11:48:24 of which there are probably 1 11:48:44 90% of it's the same, and any stuff you'll mess with in the next year or so is the same 11:48:57 yea well how many of the instructions would be the same on a 486? 11:49:00 p4's are pretty much 100% backwards compatible with a 386 11:49:17 they're all the same 11:49:22 cmov will work on a 386? 11:49:24 and the intel manuals are pretty good about noting where something isn't supported on older processoras 11:49:28 no, that was added in ppro 11:49:29 http://www.choppingblock.org/d/20020827.html 11:49:34 thats what i mean 11:49:42 but it says that it was added in ppro 11:49:44 so then you dont use it 11:49:57 you'd be suprised how little was added actually 11:50:32 cmov is probably about 50% of the instructions that were added since 386 that aren't in a specific set of instructions (like MMX, SSE, etc) 11:50:40 and where does it say it was added in ppro? 11:50:45 it does 11:51:05 point is, the newer manuals are better 11:51:05 i asked where.... 11:51:11 in the fucking manual 11:51:21 do you want me to tell you the fucking page? 11:51:24 probably in the instruction set manual.. 11:51:24 you want exact page and line? =P 11:51:25 im looking at it 11:51:38 and it doesnt say anything about when it was added 11:51:52 look at the exceptions it can cause 11:52:07 specifically the #UD exception 11:52:31 2002 ia-32 architecture software developer's manual 11:52:32 The CMOVcc instructions were introduced in the P6 family processors 11:52:34 volume 2 11:52:38 3-81 11:52:47 "The CMOVcc instructions were introduced in the P6 family processors" 11:52:55 im tgere 11:53:01 there* 11:53:01 page 123 11:53:09 trans are you here? 11:53:20 there isnt a UD exception 11:53:31 what? 11:53:47 there is no UD exception 11:54:10 what are you talking about? where? why are you telling us this? 11:54:18 point is use a current manual 11:54:26 it's better, it'll say if something is newer than 386 11:54:39 for the simple fact that they'll fix earlier erratta 11:54:44 look at the exceptions it can cause 11:54:44 specifically the #UD exception 11:54:46 errors in the docs 11:54:53 I dont see it either 11:54:54 move on 11:55:04 wcstok is probably on crack 11:55:11 heh 11:55:17 either way i dont want a current document 11:55:26 i dont care about current documents 11:55:28 ARGH 11:55:45 fine 11:55:46 it happens..rather odd they dont have that listed though 11:55:47 i want to use a 383 document 11:55:52 and thats what im going to use 11:55:55 then use a 386 document 11:56:03 have fun, dont mention it again 11:56:35 at the minimum order a new one from intel 11:56:49 then when it shows up in a few months, you'll have another source 11:56:51 i already have the p4 manuals in hardcopy form 11:56:56 WHAT? 11:57:03 so then... why... never mind 11:57:05 back to work 11:57:24 lol 11:58:02 i guess Xerroz is a Intel-hater and wants to destroy intel by ordering Hard Copies 11:58:10 heh 11:58:23 I ordered docs from AMD 11:58:27 I'm a everyone hater 11:58:31 lol 11:58:44 the AMD ones take a bit... 11:59:02 that makes no sense, how would ordering hard copies destroy intel... 11:59:10 uhm 11:59:40 and how would that constitute hating intel? 11:59:45 they are free, they have to spend money, tada, if you order like one billion of it 11:59:53 heh, they took so long sending the AMD docs, I went and ordered anther copy..and then 3 months later another..and then I ended up with 3 sets of em 12:00:22 i got my amd manuals pretty quickly 12:00:25 2 weeks 12:01:43 of which? Intel or AMD? 12:01:54 AMD 12:01:55 I ordered the x86-64 docs like a year ago 12:02:01 then 3 months later I got a few of them 12:02:07 then about 3 months later I got all 5 12:02:12 i have all 5 volumes of x86-64 12:02:14 so I have duplicates of 3 of them 12:02:14 hehe 12:02:37 i guess i was lucky to get them so quickly 12:03:52 --- join: kernel2420 (1000@ppp-176-162.25-151.libero.it) joined #osdev 12:03:52 --- quit: kernel2420 (Client Quit) 12:04:52 you progbably ordered 'em when they were finally going through the backlogs 12:05:44 i guess so 12:08:45 bbl 12:08:48 --- quit: Divine (Read error: 110 (Connection timed out)) 12:09:13 --- join: revanthn (pbhgd@202.9.183.146) joined #osdev 12:09:24 --- nick: minddog[away] -> minddog 12:10:01 --- join: cookin (~jrydberg@213.204.157.23) joined #osdev 12:20:28 --- join: Divine (~john@12-246-112-182.client.attbi.com) joined #osdev 12:24:08 back 12:28:10 --- quit: gila_ ("Client Exiting") 12:28:33 --- join: debug_ (~debug@meego.hemmet.chalmers.se) joined #osdev 12:29:44 --- join: kemu (~jonas@183.150-136-217.adsl.skynet.be) joined #osdev 12:44:48 --- quit: sxz (calvino.freenode.net irc.freenode.net) 12:44:48 --- quit: HeavyJoost_ (calvino.freenode.net irc.freenode.net) 12:44:48 --- quit: dax (calvino.freenode.net irc.freenode.net) 12:44:48 --- quit: notten (calvino.freenode.net irc.freenode.net) 12:44:48 --- quit: lodda (calvino.freenode.net irc.freenode.net) 12:44:48 --- quit: mors (calvino.freenode.net irc.freenode.net) 12:44:48 --- quit: wcstok (calvino.freenode.net irc.freenode.net) 12:44:51 --- quit: zhware (Dead socket) 12:44:51 --- quit: debug_ (Excess Flood) 12:44:51 --- quit: kemu (Dead socket) 12:44:52 --- quit: Hirogen2 (calvino.freenode.net irc.freenode.net) 12:44:52 --- quit: Apophis (calvino.freenode.net irc.freenode.net) 12:44:52 --- quit: Robert (calvino.freenode.net irc.freenode.net) 12:44:52 --- quit: revenger (calvino.freenode.net irc.freenode.net) 12:44:52 --- quit: Matzon (calvino.freenode.net irc.freenode.net) 12:44:52 --- quit: Xerroz (calvino.freenode.net irc.freenode.net) 12:44:53 --- quit: cookin (calvino.freenode.net irc.freenode.net) 12:44:53 --- quit: revanthn (calvino.freenode.net irc.freenode.net) 12:44:53 --- quit: mur (calvino.freenode.net irc.freenode.net) 12:44:54 --- quit: sayke (calvino.freenode.net irc.freenode.net) 12:44:54 --- quit: file (calvino.freenode.net irc.freenode.net) 12:44:54 --- quit: coredump (calvino.freenode.net irc.freenode.net) 12:44:54 --- quit: EtherNet (calvino.freenode.net irc.freenode.net) 12:44:54 --- quit: Divine (calvino.freenode.net irc.freenode.net) 12:44:54 --- quit: trans (calvino.freenode.net irc.freenode.net) 12:44:54 --- quit: bono (calvino.freenode.net irc.freenode.net) 12:44:54 --- quit: voider|NA (calvino.freenode.net irc.freenode.net) 12:44:54 --- quit: geist (calvino.freenode.net irc.freenode.net) 12:44:54 --- quit: Rico (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: debug (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: gab (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: mrMister (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: Zenton (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: wl (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: kreep (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: minddog (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: kyelewis_z||z (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: jaf (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: zwanem (calvino.freenode.net irc.freenode.net) 12:44:59 --- quit: asmodeus (calvino.freenode.net irc.freenode.net) 12:45:00 --- quit: jonaslund (calvino.freenode.net irc.freenode.net) 12:45:00 --- quit: wli (calvino.freenode.net irc.freenode.net) 12:45:14 --- join: debug_ (~debug@meego.hemmet.chalmers.se) joined #osdev 12:45:17 --- join: voider|NA (voider@206.167.166.74) joined #osdev 12:45:24 --- join: minddog (~minddog@67.128.0.30) joined #osdev 12:47:45 --- join: Matzon (Mazon@0x50a1b5bf.unknown.tele.dk) joined #osdev 12:47:45 --- join: revenger (revenger@chello080108014242.36.11.vie.surfer.at) joined #osdev 12:47:45 --- join: Robert (~snofs@h138n2fls31o965.telia.com) joined #osdev 12:47:45 --- join: Apophis (~kn64@ppp1609.nsw.padsl.internode.on.net) joined #osdev 12:47:45 --- join: Hirogen2 (jengelh@linux01.gwdg.de) joined #osdev 12:47:45 --- join: file (~jcolp@mctn1-3230.nb.aliant.net) joined #osdev 12:47:45 --- join: Divine (~john@12-246-112-182.client.attbi.com) joined #osdev 12:47:45 --- join: z3r0_one (~z3r0_one@lsanca1-ar40-4-33-138-004.lsanca1.dsl-verizon.net) joined #osdev 12:47:45 --- join: I440r (~I440r@dialup-67.29.214.168.Dial1.Cincinnati1.Level3.net) joined #osdev 12:47:45 --- join: sayke (~abuse@sttldslgw29poolF193.sttl.uswest.net) joined #osdev 12:47:45 --- join: cookin (~jrydberg@213.204.157.23) joined #osdev 12:47:45 --- join: coredump (~ben@h0020af25039b.ne.client2.attbi.com) joined #osdev 12:47:45 --- join: zhware (~zhware@external.silveregg.co.jp) joined #osdev 12:47:45 --- join: zwanem (User-10528@modemcable204.207-203-24.mtl.mc.videotron.ca) joined #osdev 12:47:45 --- join: Zenton (~vicente@8.Red-80-34-35.pooles.rima-tde.net) joined #osdev 12:47:45 --- join: kreep (~Mental@65.169.21.243) joined #osdev 12:47:45 --- join: mrMister (~andri@ti122110a080-0648.bb.online.no) joined #osdev 12:47:45 --- join: kyelewis_z||z (~w00t@203.221.42.101) joined #osdev 12:47:45 --- join: jaf (~sbox@200.146.20.59) joined #osdev 12:47:45 --- join: jonaslund (whizzter@as2-5-1.sbn.s.bonet.se) joined #osdev 12:47:45 --- join: wli (wli@holomorphy.com) joined #osdev 12:47:45 --- join: geist (~geist@tkgeisel.com) joined #osdev 12:47:45 --- join: EtherNet- (~askbill@200.45.181.96) joined #osdev 12:47:53 --- join: revanthn (omsakg@202.9.183.195) joined #osdev 12:48:09 --- join: Rico (~Rico@node-c-1c6a.a2000.nl) joined #osdev 12:48:09 --- join: bono (~bono@modemcable027.101-200-24.mtl.mc.videotron.ca) joined #osdev 12:48:09 --- join: gab (~prfalken@gaia.chx-labs.org) joined #osdev 12:48:09 --- join: sxz (~sixer@212.59.18.2) joined #osdev 12:48:09 --- join: HeavyJoost_ (~HeavyJoos@a213-84-139-110.adsl.xs4all.nl) joined #osdev 12:48:09 --- join: dax (dax@u212-239-165-167.adsl.pi.be) joined #osdev 12:48:09 --- join: mors (~skywalker@64.104.136.140) joined #osdev 12:48:09 --- join: notten (~root@bgp423523bgs.union01.nj.comcast.net) joined #osdev 12:48:09 --- join: lodda (~playgroun@h138n2fls31o965.telia.com) joined #osdev 12:48:09 --- join: wcstok (jwhite@cse-old.unl.edu) joined #osdev 12:48:14 --- join: Ghiottone (~ale@151.27.17.45) joined #osdev 12:48:17 --- join: kemu (~jonas@183.150-136-217.adsl.skynet.be) joined #osdev 12:48:23 --- quit: mrMister (calvino.freenode.net irc.freenode.net) 12:48:23 --- quit: jonaslund (calvino.freenode.net irc.freenode.net) 12:48:23 --- quit: jaf (calvino.freenode.net irc.freenode.net) 12:48:23 --- quit: kreep (calvino.freenode.net irc.freenode.net) 12:48:23 --- quit: zwanem (calvino.freenode.net irc.freenode.net) 12:48:23 --- quit: geist (calvino.freenode.net irc.freenode.net) 12:48:23 --- quit: EtherNet- (calvino.freenode.net irc.freenode.net) 12:48:23 --- quit: wli (calvino.freenode.net irc.freenode.net) 12:48:23 --- quit: zhware (calvino.freenode.net irc.freenode.net) 12:48:23 --- quit: kyelewis_z||z (calvino.freenode.net irc.freenode.net) 12:48:23 --- quit: Zenton (calvino.freenode.net irc.freenode.net) 12:48:25 --- join: kyelewis (~w00t@203.221.42.101) joined #osdev 12:48:30 --- join: jonaslund (whizzter@c07984a.sbn.bostream.se) joined #osdev 12:48:34 --- quit: cookin (calvino.freenode.net irc.freenode.net) 12:48:34 --- quit: coredump (calvino.freenode.net irc.freenode.net) 12:48:43 --- join: cookin (~jrydberg@213.204.157.23) joined #osdev 12:48:43 --- join: coredump (~ben@h0020af25039b.ne.client2.attbi.com) joined #osdev 12:48:50 --- join: kreep (~Mental@65.169.21.243) joined #osdev 12:48:53 --- join: EtherNet- (~askbill@200.45.181.96) joined #osdev 12:49:00 --- join: zwanem (User-10528@modemcable204.207-203-24.mtl.mc.videotron.ca) joined #osdev 12:49:32 --- join: Zenton (~vicente@80.34.35.8) joined #osdev 12:49:46 --- join: mrMister (~andri@ti122110a080-0648.bb.online.no) joined #osdev 12:49:53 --- join: geist (~geist@tkgeisel.com) joined #osdev 12:51:18 --- join: wl (philipp@p50865B43.dip.t-dialin.net) joined #osdev 12:51:24 --- join: zhware (~zhware@external.silveregg.co.jp) joined #osdev 12:52:16 --- join: Mathis (irc@pD9EA9601.dip.t-dialin.net) joined #osdev 12:52:58 hi 12:53:05 --- join: asmodeus (~www@h125n2fls33o867.telia.com) joined #osdev 12:53:27 howdy Mathis 12:53:43 howdy geist 12:54:13 Hi 12:54:35 hi Robert 12:54:59 Mathis :) 12:55:07 Robert :-) 12:58:09 --- quit: wl ("Quit") 13:04:53 --- nick: cookin -> cookin-awk 13:05:14 --- nick: cookin-awk -> cock-afk 13:09:18 wow 13:09:36 Awww, cock. 13:09:39 I love you so much. 13:09:42 these netsplits....... 13:09:44 And you yoo, lodda :) 13:09:47 --- quit: bono ("Lost terminal") 13:09:50 yooyoo 13:09:53 yoyo 13:09:53 Er, too 13:09:57 >:) 13:13:46 --- quit: Zenton ("Client Exiting") 13:15:00 --- join: Kurt (~kmw@oc-nas-11-s91.cinergycom.net) joined #osdev 13:15:23 --- quit: z3r0_one ("Now committing seppuku daily for the last time...") 13:18:02 --- quit: revanthn () 13:18:40 --- join: kernel2420 (1000@ppp-176-162.25-151.libero.it) joined #osdev 13:19:18 hello 13:19:58 --- join: revanthn (oonrle@202.9.183.150) joined #osdev 13:19:58 hello 13:21:04 i have problem with floating point rappresentation 13:21:12 why is long double 96bit size? 13:21:35 in the intel reference long double must be 80 bit 13:21:42 what is 'rappresentation'? 13:21:43 i use gcc compile 13:23:12 mmmhh 13:23:43 mument 13:23:49 check your spell 13:25:17 configuration 13:25:30 how the bit is store in memory 13:27:18 am i stupid or my mm doesn`t work properly ? 13:27:18 char* virtual_address = (char*)alloc_page(); 13:27:18 map_page((ulong*)kPageDir, (ulong)virtual_address, (ulong)virtual_address); 13:27:18 for(i=0; i<4096; i++) 13:27:18 printf("%lu", virtual_address[i]); 13:27:23 page fault 13:27:58 ulong = japanese tea? :D 13:28:04 yeah 13:28:11 somethin like that =] 13:28:26 --- join: mur (murr@baana-62-165-189-56.phnet.fi) joined #osdev 13:28:37 mur from banana 13:28:48 hello playground 13:29:05 sings "i am the banana and you are the blahblahblah" 13:29:37 a lot of children play on me and do evil tricks on me *g* 13:30:19 hrm 13:30:53 --- nick: cock-afk -> cookin 13:31:05 teh cock is back to cook 13:31:12 to be cooked 13:31:19 :D 13:31:58 bananas to the beat... 13:32:41 cookin with banana 13:32:50 yum 13:33:50 heh, cock-afk 13:37:24 love, heaven, submission and more... 13:37:34 love, hell, in room 904 13:37:39 tremendous pain... 13:38:02 kernel2420: the double may be 80 bits on the cpu, but C can only allocate in powers of two 13:38:05 eight, six, five, four, oh baby, gimme some more... 13:38:08 so it probably gives it 128 bits 13:38:22 printf("%d", sizeof(double)); should tell you 13:38:47 geist: maybe you have to write like he does ;-) 13:39:08 doesn't the FPU work with 80-bits, but a double is just 64-bits? 13:39:20 dunno, should be easy to find out 13:39:35 FPU works with 80 bits, yep 13:39:44 double = 64 bit 13:40:02 edx:eax, ecx:ebx 13:40:12 oh and 'long double' != double 13:40:14 didn't know that 13:40:38 --- quit: sxz () 13:40:39 looks like there is a 128bit long double, but it's emulated in software 13:40:40 sure? 13:40:46 x86 has a 80 bit long double 13:40:55 ecx:ebx:edx:eax? 13:41:43 for returns? maybe in a fp or mmx reg 13:41:45 looking 13:42:03 mmx may have other jobs 13:42:41 also I wouldnt use MMX for return values 13:42:42 probably 13:42:55 maybe memory location would be a better choice 13:42:57 well, easy enough to find out 13:43:04 write a program, test it 13:43:06 geist: good, but i think that the gcc reppresentation follow an other representation then int format 13:43:12 who the hell needs 128bit values?!? 13:44:34 i want understand the standard used by gcc 13:46:27 hello kernel2420! 13:46:30 saluta ti! 13:46:44 trying to figure it out right now... 13:46:46 hello mur 13:47:15 hmm, dont know the fp instructions on x86 at all 13:47:48 on this box(not even pentium, so no MMX and stuff) it uses the FPU for returning long doubles 13:48:56 interesting, sizeof(long double) == 12 13:49:03 yea 13:49:12 so it must be that long double == 80 bits + some crap 13:49:19 since x86 has a special >64 bit mode 13:50:11 --- join: witten (~witten@ip-64-32-131-193.dsl.lax.megapath.net) joined #osdev 13:50:57 looks like it's returned on the fp stack 13:51:08 fadd %st(0),%st 13:51:20 I did long double res = foo() * 2; 13:51:25 where foo() returns a long double 13:51:52 oh yeah, that's right, the x86 fpu regs are 80bits long anyway 13:51:54 maybe it's M$, which has smuggled a developer into gcc team, spying us coders out with 16 bit! 13:52:02 what does the ABI say? 13:52:22 a bastards inferface 13:52:32 oh, I dont wanna look that up 13:53:03 who does. 13:54:44 --- join: dh (~gfafgawrg@213-35-248-212-dsl.kvm.estpak.ee) joined #osdev 13:54:57 I never look it up anyway, I just look at the config.h file for the particular architecture in gcc 13:55:03 and write test programs 13:56:12 --- join: keyhack (keyhack@216-190-244-176.nrp3.mon.ny.frontiernet.net) joined #osdev 13:56:45 --- quit: mrMister ("kvennkyn!") 14:00:02 linux do not wanna boot in guss :/ 14:00:30 --- part: revanthn left #osdev 14:02:24 oh 14:02:36 i just made an optimization that increased the speed of GUSS with 10% 14:03:23 hmm, you made some random loops into it? *G* 14:03:57 s/made/removed 14:04:54 instead of doing a mmu translation on the program counter on a on-page branch i always cache the physical address of the page that the program counter is on 14:05:11 --- quit: keyhack () 14:05:27 since the virtual-physical translation cache is implemented using a splay-tree, it got quite fast when the tree lookup was removed 14:06:16 that makes sense 14:06:16 --- quit: Ghiottone ("Client exiting") 14:06:36 so how do you check that it has crossed a page? 14:06:51 check on every branch and then on each sequential fetch? 14:07:24 well, I guess checking to see if you crossed a page isn't the expensive part 14:07:27 it's the translation 14:07:31 when the insn is decoded i check if it is a on-page or off-page branch 14:07:39 good night 14:07:43 later lodda 14:07:48 and then have different service routines that does the insn semantics 14:08:48 hmm, I wonder if bochs does that 14:09:06 do not think so 14:09:12 how fast i bochs really? 14:09:28 reasonably quick considering what it does 14:09:36 but it doesn't do translation, and it's not asm optimized 14:09:44 so actually it' 14:09:52 it's probably no where near something like virtualpc 14:10:43 well 14:10:48 bochs could be faster i guess 14:10:56 if they had done it the GUSS way :) 14:11:38 anyway, gotta go. see ya later 14:11:42 --- quit: cookin ("Client exiting") 14:41:48 --- quit: witten ("Client exiting") 14:44:48 --- join: jsr (www@du-12-105.ppp.telenordia.se) joined #osdev 14:52:35 --- quit: kreep ("Client Exiting") 15:01:30 --- join: xeon (~xeon@cs6668188-147.austin.rr.com) joined #osdev 15:01:41 Hey every1 15:01:46 --- quit: file (Read error: 113 (No route to host)) 15:03:40 --- part: kernel2420 left #osdev 15:11:13 Is anyone in here implementing an OO C++ kernel? 15:16:40 I tried writing in C++ 15:17:39 writing it completely with classes and so on is impossible by that design I wanted to use 15:31:36 --- join: int13h (~no@calcium.nelte.lt) joined #osdev 15:32:55 gtg, sleep required 15:32:56 --- quit: Mathis ("good night") 15:34:32 --- join: jaf (~sbox@200.146.20.59) joined #osdev 15:37:52 --- quit: I440r (No route to host) 16:03:46 --- quit: Kurt ("Connection reset by rear") 16:03:50 --- quit: jsr (Read error: 104 (Connection reset by peer)) 16:04:00 Good night 16:06:03 'night 16:07:19 --- quit: mur ("MURR!") 16:22:35 --- join: minddog_ (~minddog@fleenode.net) joined #osdev 16:27:02 --- quit: int13h () 16:28:52 --- join: file (intranet@mctn1-2143.nb.aliant.net) joined #osdev 16:33:11 --- join: zwane (User-10528@modemcable204.207-203-24.mtl.mc.videotron.ca) joined #osdev 16:33:57 hm 16:37:12 --- join: nutrageous (~Ishq@frm-64-4-102-202.access.ntelos.net) joined #osdev 16:37:39 --- quit: Ishq (Read error: 104 (Connection reset by peer)) 16:37:40 --- nick: nutrageous -> Ishq 16:47:21 --- join: I440r (~I440r@dialup-67.29.205.44.Dial1.Cincinnati1.Level3.net) joined #osdev 16:49:13 Has anyone researched the Windows NT OS design? I don't know about you, but to me it looks super advanced and well thought-out... 16:55:02 yes 16:55:08 there is much to learn from it 16:59:54 I spent a lot of time looking at NT 17:11:30 geist: I've gained a new found respect for it. 17:12:03 I should have gotten a book earlier, I'm really thinking about my design now, I might even read a few more books before I even start coding. 17:12:15 good 17:13:37 God *shakes head* I look back now and want to slap myself for some of my naiveness 17:15:24 never underestimate how complex a modern os is 17:15:34 I know that now. 17:15:43 we're talking about *millions* of lines of code 17:15:57 I know 17:16:25 that doesn't mean it's not a worthy subject of research 17:16:26 I heard Windows XP was like more than 30 million lines. 17:16:52 last numbers I heard was win2k was around 50, and I'm sure XP is much more than that 17:17:12 I can see it too, I personally ran 'wc' on the beos source tree when I worked there, and it was a little over 6 million lines 17:17:21 and it wasn't anywhere nearly as complex as windows NT 17:17:37 Who knows, I just know that it's a lot, and I'm never going to be able to produce anything even remotely as large and complex as that. 17:18:25 Would you like to see what kind of design I've thought of so far? 17:19:07 sure 17:20:15 --- quit: I440r () 17:20:22 Well to begin with I wanted to use all 4 of the x86 protection levels. 17:20:43 looks like linux has around 4.5 million lines of .c .h files in the kernel alone 17:20:54 But, some people advised against it, so I tried to rethink my design. 17:21:05 yeah, that's not needed 17:21:53 I looked at how Windows 2000 does it and it looks like they have a semi-microkernel almost 17:22:17 Actually it looks just like a microkernel, but they deny it, so I'll assume they're telling the truth 17:22:33 about as much as osx with mach is a microkernel 17:23:03 Windows 2000 has the HAL, Kernel, Servers, and services in supervisor mode, and everything else in usermode 17:23:19 not all servers 17:23:52 but a lot of the time critical stuff 17:23:59 --- join: Zenton (~vicente@8.Red-80-34-35.pooles.rima-tde.net) joined #osdev 17:24:01 But, I want to Have only the HAL and Microkernel in supervisor mode, and all the servers and user stuff in user mode 17:24:15 okay, what about device drivers? 17:24:17 I'm doing a Microkernel design too 17:24:24 user mode too 17:24:32 where is the vm located? 17:24:33 That's the hope, though 17:24:37 As a server 17:24:40 oh? 17:24:46 how will you implement that? 17:24:53 Through the HAL 17:25:01 hrm 17:25:29 so would putting the vm in a server not have some reentrancy issues? 17:25:35 what happens when you take a page fault? 17:25:43 do you do a quick context switch to the vm server? 17:25:55 how do you do any demand paging within the vm server? 17:26:24 Reentrancy issues? Such as efficiency? 17:26:30 no 17:27:00 so if you page fault inside a server you'll have to ask the vm server to do some work for you right? 17:27:21 Yes. 17:27:38 so then what if there is a page fault in the vm server? 17:27:44 will it not be demand paged? 17:27:53 if so, how will you handle memory allocation within the vm server 17:28:09 or what about any memory allocation along the critical path of the swap file 17:28:20 ie, the disk drivers, fs servers, network servers, etc 17:28:34 Ok, ok 17:28:37 that's a very tough problem to solve in a monolithic system, and even harder with a ukernel 17:29:16 That's a sign that I need to go research some more ;) 17:29:17 mind you what you suggest can be done 17:29:37 it has been done even, it's just a *very* complex problem 17:29:59 also note that a microkernel is more of a design philosophy than anything else 17:30:09 it's a continuum of design, you can do more of less of a pure design 17:30:25 ie, vm in the kernel, drivers, but fs outside, network outside for example 17:30:36 or gui inside but a seperate module, like NT 17:30:53 or kernel has vm and threading, and nothing else 17:31:07 or kernel is just a task switcher, everything else outside 17:31:31 in general as you get more towards a ukernel, you gain flexiblity and a more modular design, but lose speed 17:31:40 Kernel just a message passer and supervisor, everything else outside? 17:31:48 and in some ways things get more complex 17:32:12 geist: Have you heard of sysenter/sysleave x86 instructions? 17:32:19 no I haven't actually 17:32:29 I'll eventually get around to it, but that's just an optimization 17:32:39 there are a few different ones 17:33:29 That's the only thing I can think of to try to improve the speed. 17:33:39 It's going to be tough... 17:34:00 People are saying that I should make a monolithic kernel first, but I don't want to poison my mind... 17:34:13 What will NewOS be? 17:34:15 poison your mind? 17:34:55 Poison my mind. 17:35:01 huh? 17:35:15 As in learn something that will be hard to unlearn 17:35:49 what garbage is that? 17:35:55 unlearn what? 17:37:11 Unlearn the design. 17:37:11 That way when I later try to make a microkernel, I don't reuse the design 17:37:22 uhh okay 17:39:36 --- quit: minddog_ (calvino.freenode.net irc.freenode.net) 17:39:36 --- quit: jaf (calvino.freenode.net irc.freenode.net) 17:39:36 --- quit: xeon (calvino.freenode.net irc.freenode.net) 17:39:36 --- quit: dh (calvino.freenode.net irc.freenode.net) 17:39:36 --- quit: Divine (calvino.freenode.net irc.freenode.net) 17:39:36 --- quit: sayke (calvino.freenode.net irc.freenode.net) 17:39:43 xeon: I don't think it's a problem. 17:39:48 hm. netsplit? 17:39:51 yeah 17:40:10 --- join: minddog_ (~minddog@fleenode.net) joined #osdev 17:40:10 --- join: jaf (~sbox@200.146.20.59) joined #osdev 17:40:10 --- join: xeon (~xeon@cs6668188-147.austin.rr.com) joined #osdev 17:40:10 --- join: dh (~gfafgawrg@213-35-248-212-dsl.kvm.estpak.ee) joined #osdev 17:40:10 --- join: Divine (~john@12-246-112-182.client.attbi.com) joined #osdev 17:40:10 --- join: sayke (~abuse@sttldslgw29poolF193.sttl.uswest.net) joined #osdev 17:40:13 'unlearn' that's pretty funny 17:40:13 looks like he's read the microkernel propaganda 17:40:14 xeon: I don't think it's a problem. 17:40:16 oh well, it's progress 17:40:29 at least he's thinking about design 17:41:00 I'm taking my first OS course now... we don't really talk much about micro vs monolithic 17:41:12 that's considered an implementation issue 17:41:17 geist: I realize it's propaganda, but don't they kind of have a point? 17:41:56 xeon: you cannot really do anything 'pure' these days anyway 17:43:20 I wrote a monolithic kernel before I begun my current (micro-)kernel, and I don't think that it is problematic. in fact, I think the more you code the better 17:45:40 How far did you get with your monolithic kernel? 17:46:48 it ran some bsd binaries 17:47:04 but not complex stuff like gcc or anything like that 17:47:34 I never supported writes to the filesystem, networking was non-existant... lots of stuff missing 17:47:44 but it was fun while it lasted :) 17:47:55 fun 17:48:43 * debug_ has learned by design mistakes from that kernel 17:48:49 s/by/from/ 17:51:07 http://www.mdstud.chalmers.se/~md1gavan/ycx2/yoctix-0.0.85-build_1223-boot.img <-- the monolithic thingy 17:52:19 it doesn't work in bochs though 17:54:33 OH! 17:54:49 You Do you do YCX? 17:55:04 You made YCX?* 17:55:55 yoctix, yes 17:56:30 heh, read about the theft 17:56:32 bummer 17:57:11 debug: I like your kernel code, it's very clean and understandable. 17:57:29 the only time when I had more adrenaline in my body, except for when I saw that my code had been stolen, was when I had accidentally ran into another car while working. that was a really weird feeling 17:57:41 xeon: I hope you don't mean the old yoctix. it's awful :) 17:57:48 wow, it's not worth that 17:57:53 Nope, the new 17:57:58 it's just source code man 17:58:01 What's this about code stealing? 17:58:11 if someone stole it, write em a nasty mail otherwise what can you do 17:58:26 not worth killing the project over it 17:58:40 geist: I had more or less moved over to ycx2 anyway, so... 17:58:44 yeah 17:58:59 basically that's the risk you take when you put source online 17:59:04 I had a page up, comparing my code to his, but I've removed that page now 17:59:06 debug: Who stole your code? 17:59:14 even something like GPL doesn't really protect you 17:59:16 some guy, I think he was just learning how to code 17:59:26 what are you gonna do, sue the person that stole it? 17:59:27 but it made me really upset 17:59:37 yah 17:59:56 he appologized, both by mail and on his OS project home page, so it's pretty much ok now 18:00:51 What was his excuse? 18:02:02 that he was going to rewrite those parts completely anyway, and he didn't realize that he had his sources available for download before he had time to rewrite it 18:02:02 :) 18:02:05 * debug_ remembers "interesting" bugs in yoctix 18:02:21 when I did the openbsd binary emulation, there were some weird things regarding mmap() and fork() 18:02:39 mmap takes a dummy int, to pad out to 64-bit. I didn't see that at first, so many programs bugged out at random 18:02:57 and fork() doesn't simply return a value in eax, it uses edx too. it made me really confussed 18:02:59 -s 18:03:25 writing your own syscalls and so on is probably a lot easier, then you have complete control/understanding of what's going on 18:05:44 http://www.openbsd.org/cgi-bin/cvsweb/src/lib/libc/arch/i386/sys/fork.S?rev=1.2&content-type=text/x-cvsweb-markup 18:06:40 I think i'm going to make my own executable format :P 18:08:10 I wish I had old machines with different architectures 18:08:41 I wish I had time to work more on ycx2 18:10:12 Me too, only not on ycx2 :P 18:10:26 :) 18:14:43 yeah, I dont have enough time myself 18:14:49 busy all the time\ 18:15:32 * debug_ dreams about rainy summer days 18:17:19 it's past 03 am here, so I think I'll go to sleep 18:17:21 gnite 18:18:03 nite 18:19:18 mmm 18:32:17 --- quit: Zenton ("Client Exiting") 18:35:17 --- quit: kyelewis (Read error: 60 (Operation timed out)) 18:39:20 xeon: sup 18:40:06 --- join: kyelewis (~w00t@220.240.64.150) joined #osdev 18:50:09 Has anyone noticed how echo only keeps one space if there are two in a sentence... 19:02:19 xeon: whaddya mean? 19:17:46 --- join: redblue (star@ppp039.216-96-207.sherb.mt.videotron.ca) joined #osdev 19:27:10 --- join: kreep (~kreep@CPE-65-26-15-71.kc.rr.com) joined #osdev 19:27:27 howdy 19:46:08 Hey kreep 19:46:48 notten: wow, 45 minutes later: I mean echo hello dude. I am cool --- echoes: hello dude. I am cool 20:21:36 --- join: xeon` (~xeon@cs6668188-147.austin.rr.com) joined #osdev 20:38:22 --- quit: xeon (Read error: 110 (Connection timed out)) 20:38:22 --- nick: xeon` -> xeon 20:51:15 --- quit: xeon (" HydraIRC -> http://hydrairc.sf.net <- irc client ownage!") 20:58:23 --- join: Boney (~paul@dsl-203-113-198-87.VIC.netspace.net.au) joined #osdev 21:01:56 --- quit: minddog_ ("leaving") 21:02:05 --- quit: minddog ("Lost terminal") 21:40:20 --- quit: redblue ("Happiness isn't something you experience; it's something you remember. -OL") 21:44:08 --- join: cmorris (~morris@kirbydhcp-966.eks.lsu.edu) joined #osdev 21:59:16 --- quit: kreep ("hehehehe hahahahaha ahhehehehahahehahaeh haehae hehea ahehehe heh heh hrrmmmm....") 22:00:23 --- join: minddog (minddog@fleenode.net) joined #osdev 22:01:03 (UnderNet) [ TCA ] hey voider 22:01:04 (UnderNet) [ voider ] test 22:01:14 --- quit: minddog (Client Quit) 22:01:26 (UnderNet) [ TCA ] !google voider 22:01:30 (UnderNet) [ @[neuron] ] Result: accretions...artists...zo voider - [ http://www.accretions.com/artists/zovoider.asp ] 22:01:31 (UnderNet) [ TCA ] hm. 22:01:39 (UnderNet) [ voider ] :P 22:02:01 (UnderNet) [ voider ] i added a relay between both #osdev from freenode and here 22:02:12 (UnderNet) [ voider ] simply a test 22:02:15 (UnderNet) [ TCA ] How does that work? 22:02:23 (UnderNet) [ voider ] a simple script 22:02:27 (UnderNet) [ TCA ] And, uhm.. wanna op me? 22:02:38 (UnderNet) [ voider ] ya 22:02:47 (UnderNet) [ @TCA ] thanx 22:02:53 (UnderNet) [ @TCA ] Did you see my shining new box? 22:03:00 (UnderNet) [ voider ] no 22:03:01 (UnderNet) [ voider ] url? 22:03:03 (UnderNet) [ @TCA ] Did you see my old crappy box? 22:03:07 (UnderNet) [ @TCA ] no url 22:03:09 (UnderNet) [ @TCA ] just the stats 22:03:10 (UnderNet) [ @TCA ] os[Windows 98 (4.10 - 2222)] uptime[6m 47s] cpu[1-Intel Pentium 4, 3073MHz, 512KB] mem[Usage: 157/255MB (61.57%) [||||||----]] 22:03:21 hey guys, what do you think of this (the relay) ? :P 22:03:26 (UnderNet) [ @TCA ] I was on a 475mhz box before 22:03:31 (UnderNet) [ voider ] lol 22:03:34 (UnderNet) [ voider ] very nice 22:03:43 (UnderNet) [ voider ] all is ok but the win98 :P 22:03:51 (UnderNet) [ @TCA ] I like win98 22:04:02 (UnderNet) [ voider ] i prefer win2k 22:04:47 (UnderNet) [ voider ] win98 is so buggy 22:04:59 (UnderNet) [ @TCA ] Weird.. this doesn't behave the same way my old one did 22:05:08 (UnderNet) [ voider ] i dunno if ppl from ppl like the relau 22:05:11 (UnderNet) [ voider ] relay 22:05:18 (UnderNet) [ @TCA ] I'm not sure what the relay does? 22:05:23 (UnderNet) [ @TCA ] Can they read what I'm saying? 22:05:26 (UnderNet) [ @TCA ] cuz, that would freak me out 22:05:37 (UnderNet) [ voider ] it make both of us see what happen on the #osdev channel of the other network 22:05:43 (UnderNet) [ voider ] yea 22:05:45 (UnderNet) [ voider ] :P 22:05:50 (UnderNet) [ @TCA ] heh 22:06:27 --- join: TCA (~monsters@12-224-105-86.client.attbi.com) joined #osdev 22:07:44 (UnderNet) [ voider ] ok 22:07:50 (UnderNet) [ voider ] do you see it? 22:07:51 (UnderNet) [ @TCA ] alright 22:08:07 (UnderNet) [ @TCA ] heh. That's eerie 22:08:13 (UnderNet) [ voider ] eerie? 22:08:24 (UnderNet) [ @TCA ] Seeing it on both ends like that. 22:08:37 (UnderNet) [ voider ] i dont unserstand this word :P 22:08:40 (UnderNet) [ voider ] 'eerie' 22:08:42 (UnderNet) [ @TCA ] You can turn that off.. I'd prefer not to bother them. 22:08:49 (UnderNet) [ voider ] hum, maybe 22:08:53 after timestamps, voider's name, the network tag, and the speaker's name, the text doesn't start until halfway across the window 22:08:55 (UnderNet) [ voider ] it must like a test 22:08:57 (UnderNet) [ @TCA ] I mean, it's interesting.. 22:09:16 cmorris: anyway, i think i should remove it 22:09:16 (UnderNet) [ @TCA ] But the crosstalk could be horrid 22:09:30 (UnderNet) [ @TCA ] Especially since they have so many more people. 22:09:35 yeah, no good ever comes of a relay script O_o 22:09:37 (UnderNet) [ voider ] yeah, its cool to be able to join 2 channel like this 22:09:57 --- join: minddog (minddog@fleenode.net) joined #osdev 22:10:00 (UnderNet) [ voider ] but i dont know it the will like that :P 22:11:58 here's an interesting question... I'm want to develop a dhcp client for an os that doesn't support my nic, so I would have to use bochs 22:12:13 will dhcp requests/replies make it through windows to bochs? 22:12:34 --- quit: Matzon () 22:12:45 --- nick: voider|NA -> voider 22:15:12 --- part: TCA left #osdev 22:38:16 --- quit: dh (Read error: 104 (Connection reset by peer)) 22:39:33 --- join: bono (~bono@modemcable027.101-200-24.mtl.mc.videotron.ca) joined #osdev 22:39:44 anyone here? 22:40:40 Im trying to call a function from my kernel. This function is located in a module i loaded with grub's "module" command. thing is... how do I know at compile time where grub will put the module in memory? 22:43:32 --- join: TCA (~monsters@12-224-105-86.client.attbi.com) joined #osdev 22:43:48 --- quit: TCA (Client Quit) 22:53:13 guess ill have to create a new address space for the module... 22:53:19 --- quit: bono ("leaving") 23:02:14 --- quit: zwane (Read error: 104 (Connection reset by peer)) 23:21:08 --- quit: dax (Read error: 54 (Connection reset by peer)) 23:21:47 --- join: green_ (~green@adsl-64-172-59-120.dsl.snfc21.pacbell.net) joined #osdev 23:35:42 --- quit: kemu (Read error: 104 (Connection reset by peer)) 23:59:59 --- log: ended osdev/03.05.13