00:00:00 --- log: started osdev/02.03.17 00:00:10 if anyone is interested, it's an interesting fix 00:00:14 doesn't seem to be documented 00:04:05 --- join: trans (~trans@a2a02274.intergate.bconnected.net) joined #osdev 00:27:40 --- join: zooy0rk (~zooy0rk@c-24-98-76-121.atl.client2.attbi.com) joined #osdev 00:45:05 --- quit: trans (Read error: 110 (Connection timed out)) 00:50:58 --- quit: zooy0rk (Remote closed the connection) 01:19:41 --- nick: geist -> geist-sleep 01:44:41 --- join: daxy (you@u212-239-163-15.adsl.pi.be) joined #osdev 01:44:45 heya 01:52:08 hmm i have some kind of bug in my msg calls... 01:58:29 Good for you.. I recommend the "RAID" spray to get rid of them. 01:58:53 --- nick: ZZzzzlowcoder -> slowcoder 02:02:48 --- join: rob_ert (~robert@h173n2fls33o898.telia.com) joined #osdev 02:04:42 --- join: trans (~trans@a2a02274.intergate.bconnected.net) joined #osdev 02:05:49 hmm... can't find the bug :( 02:05:54 :-( 02:05:59 Poor little dax 02:06:04 What does it cause? 02:06:15 something with my msg calls is fuxored 02:06:22 :-( 02:06:56 or maybe my ping-pong testing threads are bugged heh 02:08:39 meh can't find it... 02:08:47 one thread just blocks 02:12:40 my ping-pong threads have to be faulty, i don't really see what could be wrong otherwise 02:13:01 Threads playing ping-pong, aye? 02:13:25 yea 02:17:03 hmm 02:17:51 isnt there a program that displays X resources for programs? 02:19:45 hmm 02:19:45 THIS SUCKS. 02:27:19 --- quit: malenfant () 02:29:33 YES!! 02:29:36 editres rocks 02:58:36 anyone alive? 02:58:45 me 02:59:16 good 02:59:44 i'm stuck with an idiot bug 03:04:39 and? 03:05:02 i think i found it... hmm 03:10:32 wtf 03:10:36 this is weird... 03:11:26 thread 200 executes a send(100, msg) call and it sends a message to itself... ?? 03:15:17 faulty send function 03:15:41 prolly... 03:16:28 don't really see where the problem is though... 03:17:02 --- quit: trans (Read error: 110 (Connection timed out)) 03:17:07 somewhere in send 03:17:09 lets try to do it by calling my functions directly (no interrupts) 03:18:22 daxy: can i join yer project? 03:18:32 air, why? 03:18:49 but i want to be project leader :) 03:18:55 ... 03:19:16 air: wanna be project leader in my team? 03:19:23 I will leave 03:19:41 haha 03:19:42 --- nick: Mathis|afk -> Mathis 03:20:19 i would say my syscall interface is screwed. 03:22:03 dax: as would i :) 03:22:21 air2, and why would you say that? 03:23:47 04:11 thread 200 executes a send(100, msg) call and it sends a message to 03:24:52 actualy it's the other way around 03:30:37 anyone willing to take a look @ my code to help me out as i can't find teh bug 03:31:08 if you cant read the code, then why should others be able to read the code? 03:31:10 ;-) 03:31:25 well i CAN read it, but i CAN'T find the bug 03:31:27 that's a difference 03:32:18 oh 03:32:25 oooooohhhhh 03:32:28 aww 03:33:19 anyhow... would you mind taking a look @ my code? 03:35:27 sure, send it 03:36:02 lets hope dcc works.... 03:36:24 doesnt seems so 03:36:53 ftp://daxy.d2g.com:1337/upload/kernel.tar.bz2 03:37:57 passive mode prolly doesn't work though 03:40:08 arggg... 03:40:22 do you have my email address? 03:40:27 ah, mom 03:40:28 no. 03:42:36 have it 03:43:01 k 03:45:08 it goes wrong in kernel/main.cc line 82 03:49:36 --- join: [Mathis] (Mathis@gstw-d9b89f06.pool.mediaWays.net) joined #osdev 03:50:23 wb 03:50:25 <[Mathis]> re 03:50:51 Hi :) 03:52:31 Mathis, it goes wrong in kernel/main.cc line 82 03:53:07 <[Mathis]> I just wanted to type in Linux after it booted, then it crashed and rebooted 03:54:35 <[Mathis]> seems that my mainboard is buggy 03:58:00 <[Mathis]> ok, main.cc, 82? 03:58:06 yep 03:58:09 the send call 03:58:31 that send function is defined a bit higher in main.cc 03:58:43 and the syscall is located in arch/i386/traps.cc 03:59:44 <[Mathis]> tried to 'send' without int50 call? 04:00:55 --- join: dax (you@u212-239-163-15.adsl.pi.be) joined #osdev 04:00:56 --- quit: daxy (Read error: 104 (Connection reset by peer)) 04:00:58 meh 04:01:13 <[Mathis]> tried to 'send' without int50 call? 04:03:06 yes 04:03:08 that worked 04:03:14 so the syscall interface is faulty 04:03:19 but i can't find where it goes wrong 04:03:49 <[Mathis]> cant seen which parameter (W0, W1, W2_REG) is reg a, b, c 04:04:07 <[Mathis]> ah, see it 04:04:26 <[Mathis]> wth is 'frame'? 04:05:22 <[Mathis]> where is TrapHandler called? 04:05:32 traphandler is called in arch/interrupt.S 04:06:02 and frame is just all the regs pushed on the stack... they are later poped off again... (at the end of the interrupt) 04:06:33 <[Mathis]> and you're sure, that the regs are correct organized inside the struct? 04:06:58 yes 04:09:14 <[Mathis]> and that there is no other value between that doesnt belong to it 04:09:21 --- quit: Mathis (Read error: 113 (No route to host)) 04:09:27 --- nick: [Mathis] -> Mathis 04:09:32 yea pretty sure... 04:10:25 I dunno where exactly int50 jumps to 04:11:10 arch/interrupt.S, first the sys_call macro and then alltraps 04:11:31 and alltraps passes it to TrapHandler 04:12:15 what is 'ENTRY(syscall_ ## n) ? 04:13:22 for syscall 50 it works out to: .align 4; .globl syscall_50; syscall_50: 04:13:41 I c 04:14:21 lez list what we have on the stack 04:14:45 0, 50, 04:15:06 (all regs), ds, es, fs, gs 04:15:45 then jump to TrapHandler 04:16:52 as I can see, TrapHandler takes a parameter 04:17:13 yes... a structure containing all the regs and such on the stack... 04:17:20 but where in alltraps is this parameter pushed onto the stack? 04:17:59 well first of all the 2 pushes by that macro, and then the pushal and the other pushes 04:18:36 I dont understand... 04:18:55 I see a "call EXT(TrapHandler)" 04:19:06 but no parameter given to it before 04:19:36 pushal; pushl %ds; pushl %es; pushl %fs; pushl %gs 04:19:52 and the pushl 0 and pushl 50 before that 04:20:15 I c 04:20:23 and that works? 04:20:34 so many parameters? 04:20:38 i'm pretty sure (99.9%) that that part works 04:21:37 hmm, and 0.1% not? 04:22:04 you never know... heh 04:22:11 but no, i reallly think that part works 04:22:27 you need to be 102% sure that it works 04:22:40 102% out of 100% seems to be hard... 04:22:57 else there will be wrong values in (ExceptionFrame )frame 04:23:08 lemme write a small syscall to test it... 04:26:53 maybe the int50 call switches the stacks 04:29:02 the frame thinggy works 04:29:07 100.00% sure 04:29:51 ok 04:29:59 also in TrapHandler? 04:30:07 yes 04:30:50 so if you call int50, all 3 parameters are passed perfectly? 04:31:14 that's the next thing i'm going to check... 04:31:46 --- join: trans (~trans@a2a02274.intergate.bconnected.net) joined #osdev 04:31:54 coz I dont see in 'void send' that these parameters are pushed onto stack 04:32:21 they are 04:32:33 that's the inline asm syntax afaik 04:33:18 I see them only been assigned to the parameters, not more 04:34:04 "a"(msg->param[0]), "b"(msg->param[1]), "c"(msg->param[2]), "d"(dest), "D"(0), "S"((uint32_t)~0) <-- that puts them all in their correct registers 04:34:45 but they have to go to the stack and not into the registers, or? 04:35:13 the syscall pushes them on the stack (pushal) 04:35:23 gah 04:35:26 gv stinks 04:35:35 I c 04:35:42 it was one stinking widget that wont read the Xdefaults value 04:35:49 it has 04:37:05 so I dunno 04:39:18 brb 04:42:17 --- quit: Mathis ("connection reset by beer") 04:45:27 back 05:01:58 i can't figure this out... 05:18:02 --- nick: lynx_zzz -> lynx 05:18:05 MORNING! 05:19:28 --- quit: trans (Read error: 110 (Connection timed out)) 05:23:03 heya lynx 05:48:20 gah 05:48:26 mozilla is falling apart 05:52:42 --- join: Mathis (Mathis@gstw-d9b89f12.pool.mediaWays.net) joined #osdev 05:53:50 TIP: dont use cvs snapshots of mozilla 05:54:32 better tip: don't use mozilla at all 05:54:53 no other option 05:56:40 ie and netscape both have huge remote exploit holes in them 06:00:37 --- quit: Mathis ("connection reset by beer") 06:41:25 --- join: caereth (caereth@lgh022a.robackshus3.ac.se) joined #osdev 06:41:33 hello 06:45:59 --- join: trans (~trans@a2a02274.intergate.bconnected.net) joined #osdev 07:09:23 --- join: Aardappel (~Aardappel@pD9E10C7E.dip.t-dialin.net) joined #osdev 07:14:14 hm 07:14:20 anyone alive? 07:19:31 Njet. 07:19:41 aye 07:24:28 --- join: jukka (jukka@MMMCCXXXV.hdyn.saunalahti.fi) joined #osdev 07:28:36 --- quit: trans (Read error: 110 (Connection timed out)) 07:28:40 --- quit: lynx ("BitchX: use it, it makes you bulletproof") 07:33:37 --- quit: nbsp () 07:34:39 --- join: lynx (~lynx@pD9544DD9.dip.t-dialin.net) joined #osdev 08:05:33 --- join: jarno (NumeroUno@ppp-64-213-118.friaco.access.uk.tiscali.com) joined #osdev 08:09:36 --- part: jarno left #osdev 08:20:23 --- join: nbsp (g@ip68-14-60-7.no.no.cox.net) joined #osdev 08:21:16 m00 08:27:21 m00 08:28:18 heya spindle 08:29:25 hey 08:29:30 --- nick: Spindle[Learnin] -> Spindle 08:30:09 hmm i think i'm going to try to debug this a bit more... 08:31:57 --- nick: ZzZ_enton -> Zenton 08:32:25 hi all 08:32:41 hello 08:33:18 How the hell can I plot a pixel on the screen in pmode? vesa won't work vga won't work? 08:33:32 vesa works 08:33:45 seriously!!! 08:34:00 yea seriously 08:34:15 in the words of Ren, Happy Happy Joy Joy 08:34:29 well... not really 08:34:34 ???? 08:35:10 eh 08:35:12 sure 08:35:14 all works 08:35:22 you have to switch to the modes first , tough 08:35:22 there is an interrupt to fetch the required functions, so you can call them... but you need to figure out a way to call that int 08:35:31 or by using a realmode portal 08:35:38 i see 08:35:47 --- quit: ink (Read error: 110 (Connection timed out)) 08:35:53 becayse you can't call anything from the BIOS in pmode 08:35:59 if you switched to vesa(lfb) or vga you can work with it as usual 08:36:03 besides setting stuff up at the beggining 08:36:21 Spindle : code your own realmode portal 08:36:37 * Spindle has no idea how to do that 08:36:49 hrm 08:37:07 Spindle : what os are you coding for? 08:38:04 i think multiboot specs give a pointer to vesa bios at the begining 08:38:11 write now I am deving on Win98 Se 08:38:17 but i am not coding for an OS 08:38:20 and if anyone would be willing to help me find a bug in my messaging/syscalls i would really appreciate it as i've been trying to find it for the past day but i just can't find what's wrong... 08:38:22 I am making it all form scracth 08:40:46 Spindle, look @ this: http://www.delorie.com/djgpp/doc/ug/graphics/vesa.html 08:40:55 k 08:41:28 dax it doesn't show 08:41:38 i get that pfft ack crap 08:42:58 i am gonna restart 08:46:36 --- join: trans (~trans@a2a02274.intergate.bconnected.net) joined #osdev 08:46:36 --- quit: Spindle (Read error: 104 (Connection reset by peer)) 08:47:24 --- join: ink (~ink@user-vcauvnh.dsl.mindspring.com) joined #osdev 08:48:05 trans ink 08:51:50 --- quit: jukka ("mURR murr") 09:11:22 --- join: WiS (PiTfLuK@ACB629AA.ipt.aol.com) joined #osdev 09:11:34 hello 09:13:30 hello 09:14:02 i need informations about memory allocator 09:14:16 What kind of information ? 09:14:43 i don t understand 09:14:49 the utility 09:14:59 because 09:15:10 ? Do you need to know how to implement a memory allocator ? 09:15:10 i load my program always in the same segment 09:15:25 yes eventually 09:15:59 http://www.cs.colorado.edu/~zorn/Malloc.html#gnumalloc 09:16:04 There's a few of them.. 09:16:10 ok 09:16:13 thanks 09:16:14 :) 09:16:19 They will probably need some modification to work with segmented memory.. 09:16:42 my "OS" work in real mode 09:16:56 Exactly.. 09:16:57 i don t know if 09:17:03 i must have an allocator 09:17:09 So, your "page-size" would be 64K then. 09:17:13 yes 09:17:34 but it execute just .COM 09:17:44 Well, if it's an OS youre writing, you won't need an allocator until you decide that you need one.. 09:17:51 i need an elf loader... might code that soon 09:18:08 dax: Same here.. My ext2 code is starting to work now.. 09:18:28 i'm still trying to figure out that ipc bug... 09:18:30 well brb 09:18:32 homework 09:19:56 --- nick: slowcoder -> SlowPlayingGames 09:20:00 --- join: Spindle (~SpindleTe@12-249-117-17.client.attbi.com) joined #osdev 09:20:11 anyone know some vesa tutorials in pmode? 09:24:20 --- join: wossname (wossname@HSE-QuebecCity-ppp81128.qc.sympatico.ca) joined #osdev 09:29:44 --- quit: trans (Read error: 110 (Connection timed out)) 09:35:06 --- quit: Rico () 09:41:30 Does anyone know how hard it is to make OpenGL be your OS renderer 09:45:36 --- nick: geist-sleep -> geist 09:45:48 morning 09:47:56 morn 09:48:50 morrn 09:49:43 my eye hurts 09:50:04 not sure why. I went to sleep and woke up and my left eye hurts 09:50:06 :( 09:52:56 take it out and look at it 09:53:45 hmm, that may cause more harm than good 09:54:05 hmm, true 09:56:31 hmmm just had a publishing house send me an email 09:56:43 --- quit: WiS ("++") 09:57:09 sort of 'do you have a book in mind, or would you be thinking of teaming up with others to write books about stuff you know?' 10:09:33 GEIST 10:09:36 !!! 10:10:41 [18:39] Does anyone know how hard it is to make OpenGL be your OS renderer <-- going to be pretty hard... loads of work... (unless you port mesa or something like that) 10:12:43 you mean vesA? 10:13:06 no... mesa... a opengl library 10:13:15 heh 10:13:28 just a q.. spindle... how long have you been programming? in your favourite language? 10:15:56 2 years 10:15:59 I am a game programmer 10:16:10 I know everyhting about DirectX OpenGL C C++ 10:16:24 only 2 years? 10:16:26 but on the side i am developing a Graphical OS 10:16:30 only 2 years 10:16:32 --- join: zephir (~zephir@harr-c-165.resnet.purdue.edu) joined #osdev 10:16:39 and you know everything? isn't that a bit uhm... over-optimistic? 10:16:40 g'morning 10:16:48 howdy zephir 10:16:53 could you beat john carmack in a code deathmatch? 10:17:08 heh 10:17:09 i've been programming C/C++ for like 4 years, i wouldn't say i know everything 10:17:13 i hate cramcks code 10:17:16 its chicken shit 10:17:18 :D 10:17:26 geist: i figured the stupid problem out...it was linker script error ;) 10:17:30 the sleep helped 10:17:38 anyways I have my own Dev team thats making a futristic FPS based on the moon thats going to go commerical 10:17:40 ah, linker scripts will bite you in the ass 10:17:44 we are putting up our site any day now 10:17:49 cool 10:17:58 are you working on a design doc? 10:17:58 and you have time to mess with OSs? 10:18:42 Spindle, just being curious... how old are you? 10:18:49 no that stuff is all done 10:19:09 15 in july 10:19:32 2 years of c/c++ isn't that bad, then 10:19:41 I leanr stuff fast 10:19:46 wossname, well... i beat him :P 10:19:47 I have been alive 786wks 6days 19hrs 17mins 27secs 10:19:49 I am Now 15.092613 Years Old. 10:19:53 4 years of C++ 10:20:01 8 years of pascal :P 10:20:01 oh man, I'm around a bunch of kids 10:20:10 yea i know 10:20:10 hehe, I'm sixteen :) 10:20:12 I went to nehe and picked up OpenGL (the basics in a week) 10:20:26 you post as Sean on the 'cboard'? 10:20:30 i just know jack shit about OS dev and want to learn how stuff works 10:20:31 opengl is very easy to learn actualy 10:20:38 yeah 10:20:43 implementing it is not 10:20:44 i like it better than D3D 10:20:48 heh 10:20:51 Spindle, OS != game... this is teh real stuff mate 10:21:07 no I thought OpenGL can render the screen for me :P 10:21:16 it took me years and years to get enough know how to do anything 10:21:21 Spindle : right 10:21:31 I started my OS 4 days ago 10:21:35 you can type text :P 10:21:37 in any color 10:21:43 remember, when you code an os there is absolutely no support 10:21:48 and i want to skip this crappy mode 13 shit 10:21:49 you gots to do it all yourself 10:21:57 i know 10:22:01 you won't have mode 13h when you write a real os 10:22:13 you'll have to write code to setup video card registers yourself 10:22:27 I am gonna use vesa to start with 10:22:36 heh i've been working on my os (latest rewrite) for like a month or 2 and there still aren't any drivers 10:22:44 vesa is old 10:23:04 I've been working on mine for over a year and there are no graphics drivers 10:23:11 actually my os has no video drivers, it just writes directly to the video mem 10:23:11 of course, I'm putting that all off until it's time 10:23:32 i guess geist's os is by far the most developed? 10:23:36 amongst us 10:23:38 geist, same here... you have to set your priorities 10:23:38 if you wirte directly to the video mem your using A0000h 10:23:41 he was in dr. dobb's ;) 10:23:52 aw shucks 10:24:04 obvi, the man from be himself ;) 10:24:30 daxos has made quite some progress in the past weeks too... 10:24:33 I worked on be's kernel team a few years. thus the obvious similarity of the newos kernel to beos 10:24:44 still isn't anywhere near newos 10:24:48 actually cookin is crusing along too on rtmk 10:25:19 I'm developing mine a bit slower than most of em. I want to get it all right the first time. I'm trying to write production quality stuff 10:25:24 hey geist, would you mind helping me trying to find a bug? 10:25:28 really spending the time up front designing 10:25:34 I just want to start thing right and not re write a lot of the code 10:25:36 my question is if so many people develop their own little kernel, wouldnt it be better if we all worked on one? more ideas more dev porwer 10:25:45 everybody has their own religion 10:25:53 the religion of the micro kernel, the religion of the macro kernel 10:25:59 people dont really develop their own os to get anything done 10:26:04 they just dev it to do it 10:26:10 yea 10:26:11 yea i guess 10:26:18 therefore getting a bunch of folks together doesn't really work 10:26:20 just the desire to learn 10:26:27 yea it's fun... (although it's frustrating for me atm) 10:26:34 the better way is to just hack on yours and if it gets big enough some people will try to latch on 10:26:49 you can either kick em off the boat or find em someplace they can work on 10:27:05 I'm trying the latter, and I've gotten quite a few people that have written code for newos 10:27:27 i never liked working on larger projects with more people... dunno why... i worked on apostle, gemini and reactos before starting my own... 10:27:34 especially old friends from be, they get a lot of stuff done since I'm on their wavelength 10:28:04 well at least I have my Advance Hellow World OS down (all the files are C but the bootstrap loader) even the kernel is C :P 10:28:13 Hello* 10:28:15 good 10:28:20 damn it! i ran into trouble again 10:28:37 while doing it in asm is an admirable accomplishment, it really doesn't let you get that big 10:28:44 hehe my IO macros are fscked up 10:29:08 dax : sorry I didn't respond earlier. send me the code and I'll see what I can see 10:29:09 so were mine... i fixed them though 10:29:24 ok... 10:29:39 http://home.pi.be/~p4u02228/kernel.tar.bz2 10:30:06 ah, time for stealing 10:30:13 geist: ok 10:30:27 wossname, well not really... you said you only steal good code 10:30:51 i've got a small bug in the IPC syscall somewere... 10:30:54 but i can't find it 10:31:05 (or maybe the 2 test threads in main.cc are faulty...) 10:31:48 okay, so it context switches properly I assume? 10:31:52 you've tested that 10:31:58 yes 10:32:11 and the messages work too when i don't use interrupts 10:32:25 so my interrupt handler has to fubar something 10:34:28 still lookin... 10:36:12 maybe you'll have more success if i tell you what's wrong? heh 10:36:22 I guess so 10:36:28 just walking through the trap code 10:36:33 did you look at newos? 10:36:34 :) 10:36:38 heh 10:36:43 looks kind of similar. :) 10:36:51 newos, vsta, apostle, gemini, etc... 10:37:04 hey i DID wrote some code for apostle 10:37:27 i think i borrowed the paging code from newos too 10:37:32 i'll rewrite it later though 10:37:44 argh 10:37:45 it's a quite hack now... 10:37:49 I dont care at all 10:37:50 no wait it's l4 based 10:38:11 but i WILL rewrite it... (look @ it and you'll understand why lol) 10:38:19 seems reasonable 10:38:36 it's messy compared to other parts of my code... 10:41:13 hmmm, not sure 10:41:27 this is big enough that it's not readily apparent right away 10:42:07 the problem is that somehow i get a send call from thread id 200 which sends to thread id 200... (i never called such a function in my code) 10:42:34 is it right away, or after it's ping-ponged for a while? 10:43:14 hmm... good question 10:43:38 that'll tell you if it's a broken implementation, or it's some interrupt thing 10:44:08 the key to good debugging is to try to narrow it down, instead of just staring at it trying to figure it out 10:44:42 yea i already added quite some debugging output and such... but this isn't being easy to debug 10:45:56 and it could be something subtle, like the two stacks that are allocated are the same one so both threads are running on the same stack or something like that 10:46:13 I didn't dig into the heap code, but that'd be a good thing to verify 10:46:28 when you create the threads, print out the stack and all the data bout it, etc 10:47:14 basically at this stage you dont know if anything works quite right, so you have to start from the ground up and verify that it is working, until you work up to your problem 10:48:04 hmm 10:48:34 getting recv, send, recv, send, recv, and then i get a weirdo send from thread 200 to thread 200... 10:49:28 you code is big enough that it's too hard to eye verify it 10:49:33 start debugging it from the ground up 10:49:40 hmm 10:50:03 how? 10:50:06 when coding an OS, you'll get some seriously weirdo stuff if you dont cleanly take an int or context switch 10:50:09 the power of printf 10:50:15 I suggest writing to the serial port 10:50:53 basically it appears that you aren't getting anywhere doing what you are doing 10:50:59 so try to take it one step at a time 10:51:35 I refuse to debug it for you, but I'll poke you in the right direction. :) 10:51:50 yea that was kinda what i was hoping for 10:52:01 you have quite some experience with it i guess 10:52:13 I have a lot of experience debugging super low level fuckups 10:52:22 --- join: trans (~trans@a2a02274.intergate.bconnected.net) joined #osdev 10:52:24 take it to the baseline and make sure everything works 10:53:07 when debugging every problem, there are a lot of assumptions you make. in the normal case (like writing an app on linux or something) you can assume a lot of things (like the context switch works, malloc is okay, what I read from a file is good, etc) 10:53:14 in this case, you can't assume much 10:53:18 the weird thing is sending messages with the ints hasn't been a problem till i tried that ping-pong... 10:53:24 right 10:53:51 probably because it's actually broken, and you introduced another thread to the system, and something is getting clobbered on the context switch 10:54:35 well 1 thread sending messages to one isn't a problem, the replying seems to be one though 10:54:36 another idea that may be interesting to try is run two threads that dont ping-pong but operate on their own data and continally verify that their data is good 10:54:50 such as work some math over and over again 10:55:03 so you'll see if there's some issue context switching 10:55:07 then try to run 10 threads that way 10:55:24 then try to have a thread call a syscall over and over again 10:55:36 and then have 10 threads call a syscall over and over again 10:55:40 and then work back to what you have 10:55:51 you took a big step and it doesn't work, so scale back and find out what does 10:56:04 as in one thrad constantly receiving threads and one constantly sending threads? 10:56:10 dont necessarilly expect a magic bullet that will solve it. it may be multiple problems 10:56:10 thread* 10:56:20 think simpler 10:56:36 the point is to scale back and find what does work 10:56:37 well that was the last thing i recall which worked... 10:56:41 so take a big step back 10:57:40 again, scale back and run a couple of threads that use their own data 10:57:59 and make sure one of the thread's data doesn't get on top of the other one 10:58:04 to make sure the stack switch is okay 10:58:37 the threads you have here are simple enough that if one of them got switched to the other's stack, it'd probably still run but you'd get weird var pollution like what you may have 10:59:38 could be... hmm 10:59:57 dont make the mistake that i've made many times and assumed stuff works 11:00:13 you'll bang your head against the wall until you collapse in a bloody mess 11:00:36 only after you have looked back enough can you really say something works 11:00:50 I still find these bugs in core stuff where I'm wondering how it ever worked 11:01:00 because I'll stress the system in a new way and stuff will fail 11:01:14 so I contiually go back to the core stuff and re-verify it 11:01:20 hmm... i have 2 threads now, one sending, one receiving and verifying messages 11:01:27 it seems as they don't get corrupted 11:01:31 sigh 11:01:38 you keep trying the same things 11:01:41 that doesn't prove anythig 11:01:48 try something new, forget the messages stuff 11:01:56 or try 10 threads 11:01:58 or something like that 11:02:01 it proves my message passing isn't totaly fubar... hmm 11:02:06 sigh 11:02:15 10 threads? 11:02:17 well, I need to go. I can only help you so far 11:02:18 you got it... make it 50 11:02:32 well, that's probably a bit excessive 11:02:39 hehe 11:02:41 yea i know 11:02:47 but forget the message dea 11:02:54 I'm going to suggest this one more time: 11:03:06 have threads running that need their own data. dont do any syscalls 11:03:22 but have the threads constantly revalidate that the data they have is still good 11:03:45 ok... 11:03:45 like do some math goes through a set of operations that comes back to the start 11:03:56 hmm 11:04:03 example? 11:04:03 that'll make sure the stack switch is clean and no regs are getting clobbered 11:04:12 no example, you can figure that out 11:05:01 uhm 11:05:06 all I'm saying is that you are running some pretty trivial tests and then proclaiming stuff to be working 11:05:13 what kinda math? 11:05:19 a * b 11:05:20 --- quit: zephir ("Client Exiting") 11:05:20 etc 11:05:27 but I gotta go, spending too much time here 11:05:44 remember the words of eks, yes 11:05:55 --- nick: geist -> geist-away 11:06:07 gotta clean up the apartment before band practice today 11:06:09 like a = 2, b = 5 and then a while loop testing if a * b = 10 ? 11:06:24 needs to be more complex than that 11:06:32 something that needs the stack 11:06:49 sigh, I give up. keep at it 11:07:00 others here can help you too 11:07:08 oh well, thanks for the advice 11:07:17 no prob, sorry I can't help you more 11:07:24 BTW, good work 11:07:30 ? 11:07:33 hey i still have to do something myself heh 11:07:38 good work on what? 11:07:41 on daxos 11:07:47 it's coming along. pretty nice 11:07:57 i wasn't expecting you to solve the bug... just wanted some pointers 11:07:59 these are good bugs to be fixing 11:09:14 btw geist 11:09:19 since you complain about bx alot 11:09:23 try irssi, it's much nicer 11:09:24 Is anyone intrested in making a team from people 14-17 11:09:29 (console unix client just like bx) 11:09:39 actually I'm quitd happy with bx 11:09:55 i have found it to be generally better 11:09:59 the code is much cleaner, anyway :) 11:10:22 Spindle, for what cause? 11:10:29 to learn 11:11:10 This will be a graphical OS with a nice GUI 11:11:14 spindle: ever read the minix book? 11:11:21 --- join: zephir (~zephir@harr-c-165.resnet.purdue.edu) joined #osdev 11:11:23 nope 11:11:30 :)))) shawn is a happy man now 11:11:33 i strongly recommend you give it a try 11:11:48 wossname: the tanennbaum book? 11:12:01 Spindle, or the minix one, or his other one... "modern operating systems" 11:12:02 operating system design and principles 11:12:14 nothing: yes 11:13:28 wossname: why do you recommend that one? 11:13:46 nothing: because he is writing a graphical OS with a nice GUI 11:14:27 ney ways I got web space today for my Game Dev team 11:14:33 I can have an OS section up there 11:14:36 fourms etc 11:14:39 nothing: ^^ and that as well 11:14:53 uh 11:15:12 what does a nice GUI have to do with the necessary reading of that book? 11:15:26 . 11:15:30 i haven't read that one, i just want to know what is in that that is so neat 11:15:44 reality is contained within the book 11:15:45 reality 11:16:02 ... 11:16:13 --- nick: lynx -> lynx_somewhereinspace 11:17:04 seriously, does he talk about GUI design in that book? 11:17:12 i thought it was only basic priniciples of os design 11:17:33 Listen i'll get whatever book I need, but is anyone intrested? This will be started form scratch, we will make docs on how everythign works and then code 11:17:54 * geist-away giggles 11:18:26 Geist: ? 11:18:37 oh nothing 11:19:02 * dax is trying to stress his context-switch 11:19:19 did you verify that the stacks are seperate? 11:19:30 i'm trying to do that heh 11:19:30 did you print the thread structure when it is created? 11:19:37 did you verify that the heap works correctly? 11:19:54 i can't debug everything in 5 minutes heh 11:20:15 arg 11:20:21 geist is having fun at my expense 11:20:26 * nothing goes off and cries 11:20:32 ? 11:21:25 uhm i know that in bochs you can use a port for debugging output, which one is that? 11:21:33 e9 11:21:42 12:28 * geist-away giggles 11:21:42 12:29 < Spindle> Geist: ? 11:21:43 12:29 < geist-away> oh nothing 11:22:15 ? 11:22:27 you're the one who picked a nick that's also a common english word 11:22:30 n/m 11:22:56 arg 11:23:01 that was incredibly stupid on my part 11:24:04 maybe i should write a debug console... 11:24:30 just dump to the serial port or e9 11:25:34 dax what exactly are you trying to do? 11:26:07 Spindle, debug my code :P 11:26:21 --- nick: lynx_somewhereinspace -> lynxO 11:26:39 --- nick: lynxO -> lynx`O`Seven 11:32:47 i have a slight problem 11:33:02 for some reason bochs only displays the first line i write to port e9 11:33:58 --- quit: nbsp (carter.openprojects.net irc.openprojects.net) 11:35:03 --- join: nbsp (g@ip68-14-60-7.no.no.cox.net) joined #osdev 11:35:40 nm 11:36:16 yep all my stacks are different 11:36:41 nothing: there? 11:38:12 damn geist was right 11:38:22 already fixed a few bugs heh 11:41:01 --- nick: SlowPlayingGames -> SlowPlaingMGS2 11:44:40 yes witten 11:45:29 nothing: torsion 0.0.12 is out 11:45:33 in case you hadn't seen it 11:45:43 lots of changes 11:46:04 cool 11:46:09 i'll check that out 11:46:14 hopefully you have changed the dir structure :) 11:46:18 hmm 11:46:21 not yet :) 11:46:31 I'll do that once I factor out x86 code 11:46:39 but that's high on my list of things to do 11:46:51 you're welcome to help :) 11:47:03 :) 11:47:21 i need to get bochs installed on this box 11:47:24 anyone here who can come up with a good release name? 11:47:38 other than nothing ... 11:47:41 j/k 11:47:43 :) 11:47:49 --- nick: SlowPlaingMGS2 -> Slowcoder 11:47:54 hmm.... i'm trying to write a thread that would stress test if my stacks work decently and if nothing gets fubared during context switches... any ideas what operations whould be good for sucha purpose? 11:47:57 zephir: what?? 11:48:12 i am trying to think of a good releasename 11:48:22 zephir: why'd you say that ... other than nothing? :) 11:48:27 lol 11:48:32 your ability to come up with name ;) 11:48:59 s 11:49:10 --- quit: nbsp (carter.openprojects.net irc.openprojects.net) 11:49:11 anyone? 11:49:12 heh 11:49:15 zephir: Releasename for what ? 11:49:20 zephir: based on my nick? 11:49:22 the kernel i am wokring 11:49:34 nothing: no, i am making fun of you, and your name..geez man 11:49:49 lol 11:49:53 Slowcoder: i work by releases, and i name each release something 11:49:59 cant figure out a good release name 11:50:01 zephir: Still early in development? 11:50:02 for a kernel 11:50:04 yeah 11:50:10 "First Encounter" ? 11:50:13 i was thinking of darwin, buts its taken 11:50:20 zephir: how about SuckOS? 11:50:28 "Neanderthal" ? 11:50:29 OpenSTFU 11:50:35 OpenWTF 11:50:36 Slowcoder: whats neanderthal? 11:50:50 zephir: The predecessor of man. 11:50:56 ah ic 11:51:01 hehe i like taht one 11:51:04 mind if i use it? 11:51:11 Sure, np 11:51:13 thanks 11:51:19 i'm prolly going with OpenSTFU or something like that 11:51:38 or RTFMATIO 11:51:42 printf("Welcome to CRIX v 1.0 [Neanderthal]"); 11:51:49 (read the fucking manual although there isn't one) 11:51:58 dax; open* would be a good project name, not a good release name 11:52:20 zephir i was actualy talking about project names :P 11:52:22 zephir: where can ppl download your code? 11:52:30 release names like woody or boner are nice... 11:52:49 nothing: there's nothing yet, except for printfs 11:52:55 lol 11:52:59 pun inteded? 11:53:06 no i was serious this time 11:53:14 enough punishing the fellow :) 11:53:25 although I agree it is funny 11:53:45 yea... stop punishing him... return to your duty as a light breeze :P 11:53:46 nothing: still want it? 11:53:49 yes 11:54:00 zephir: Btw.. Window NT has been rumored to mean "Windows NeanderThal" 11:54:04 actually let me clean it up, visit http://www.osrg.org 11:54:10 get it like tomorrow or something 11:54:29 Slowcoder: didnt know that, i thought it was Networking Technologies 11:54:41 but everything is a possibility 11:54:46 untill MS states it 11:54:52 zephir: "New Technologies".. According to M$ 11:54:59 ah ok 11:55:01 anyone... 11:55:20 nothing: be here tonight or tomorrow, i will send you a tar.gz 11:55:20 i'm trying to write a thread that would stress test if my stacks work decently and if nothing gets fubared during context switches... any ideas what operations would be good for sucha purpose? 11:55:50 dax: Deep recursions ? 11:56:16 Slowcoder, example? 11:56:56 for(int i=0;i<100;i++) for(int j=0;j<100;j++) SomeFunction(); 11:56:58 ? 11:57:15 and pass some parameters with it? 11:57:15 --- join: nbsp (g@ip68-14-60-7.no.no.cox.net) joined #osdev 11:57:24 Yea.. 11:58:07 thats not recursion btw 11:58:16 heres my example 11:58:18 zephir: True indeed 11:58:42 int main() { foo(1000000); } 11:59:13 foo(int n) { if(n < 0) return; foo(n - 1); } 11:59:16 there ya go 11:59:31 That'll require at least one meg of stack.. 11:59:46 may be 10 mg? 11:59:55 no its 1 11:59:55 nm 12:00:00 too many zeks 12:01:02 Slowcoder, or the one you gave me isn't stressing anything, or maybe it really works heh 12:01:50 dax: Go with zephir's example.. I don't know what I was thinking with that.. I must be on drugs or something. 12:05:52 running 1 thread with foo(100), no problem... 12:06:25 2 thread with foo(100), no problem 12:07:05 5 threads with foo(100), no problem 12:07:38 10 thread with foo(100), no problem... 12:07:42 hmm seems good 12:08:17 20 threads work too 12:08:23 --- part: Spindle left #osdev 12:08:39 --- join: Moo (~SpindleTe@12-249-117-17.client.attbi.com) joined #osdev 12:08:41 Hey 12:09:35 hmm... is this a good sign? 12:09:47 yeah i guess 12:09:57 try larger numbers, will stress stack 12:10:06 Does anyone know tuts on implementing the keyboard? 12:10:18 trying 50 threads with foo(1000) 12:10:33 page fault :( 12:10:37 Moo: http://www.execpc.com/~geezer/osd/kbd/index.htm 12:10:50 thanks zephir 12:10:59 np 12:11:02 damn 12:11:04 this uhm sucks 12:11:07 there ya go 12:11:08 I saw that site already 12:11:12 oh 12:11:14 sorry then 12:11:19 But i thought i can't access bios in 32 pmode? 12:11:45 --- nick: Moo -> LiteOS 12:11:54 --- nick: LiteOS -> Moo 12:11:57 heh 12:12:10 25 threads with foo(1000) page faults too... damn 12:12:21 going to shower, and then i'm going to figure out what this problem is 12:12:22 --- nick: nothing -> rihpez 12:12:32 dax: can you send me the latest daxos src? 12:13:35 ... 12:16:00 --- nick: lynx`O`Seven -> lynx 12:16:27 grr 12:17:40 i think dax is in the shower 12:21:08 zephir: how do you like my nick now? 12:21:22 are you nothing? 12:21:54 no, i'm something 12:22:02 of course, didn't you see the nick-change? 12:25:31 back 12:26:07 one latest daxos sources comin' up... 12:26:34 rihpez: nope didnt see the nick change, busy working ;) and yeah its better i guess :) 12:26:42 now i can use nothing more often 12:26:45 :)) 12:26:47 lol 12:27:36 http://home.pi.be/~p4u02228/kernel.tar.gz 12:28:21 dammit 12:28:28 i really am horrible at coming up with names 12:28:33 i need a really good nick, a cool nick 12:28:37 can someone help me out ? 12:28:49 please suggest something ;( 12:28:57 nothing? 12:29:30 heh 12:29:46 what are your interests? 12:30:15 zephir: milk produckts and shiny bright objects 12:30:23 lol 12:30:24 hehe MilkMan 12:30:34 ShinyMilk ? 12:30:39 should i go back to 'morton'? 12:30:42 hehe 12:30:43 no 12:30:45 BrightCheese? 12:30:55 yea the morton one was good 12:30:56 btw, where did you get your nick from, dax? 12:30:56 Cheesy 12:31:02 mine? 12:31:04 thats a good one considering the recent events 12:31:06 i'm interested in teh origin of all your nicks 12:31:10 no rihpez's 12:31:16 honestly i don't recall where i got this one from anymore 12:31:27 zephir: how did you come up with yours? 12:31:28 my other nicks are suffocate & daripper 12:31:30 zephir has no origin its a derivative of some words that my gf threw out 12:31:39 daxx == mod musician? 12:31:39 heh 12:31:52 zephir is the roman god of the wind 12:32:10 rihpez, you can go with RedKing, opposite of RedQueen in Resident Evil movie 12:32:34 dax, thx now i have asource ;) + romans are cool 12:32:34 --- nick: rihpez -> da5id 12:32:39 anyone ever read snocrash? :) 12:33:24 nopr 12:33:59 anyone here know perl? 12:34:08 zephir: dax has mastered it 12:34:15 i know a bit perl... 12:34:21 mastered what? 12:34:25 perl 12:34:31 at least i thought you had 12:34:36 not really 12:34:36 i need to get command line args 12:34:42 --- quit: wossname ("you will all pay the terrible price! the terrible, terrible price!") 12:35:15 how do i know how many args has been passed in? 12:35:32 --- nick: da5id -> morton 12:35:41 $num = @ARGV? 12:36:33 uhm no idea... too long ago 12:36:47 k 12:36:59 zephir: what's your major @ purdue? 12:37:11 cs 12:37:14 ah 12:37:16 why? 12:37:17 that's cool 12:37:19 just wondering 12:37:24 thx 12:37:26 where you at? 12:37:26 hmm this pagefaults out... 12:37:29 pa 12:37:34 i am from de :) 12:37:38 zephir: is purdue a fun univ? :) 12:37:39 my gf is from PA 12:37:42 morton: yeah 12:37:58 or my ex-gf 12:38:01 --- nick: geist-away -> geist-band 12:38:02 zephir: what part of pa is she from? 12:38:03 where in PA? 12:38:19 zephir: reading, as my hostname implies 12:38:30 it's a city west of philadlephia 12:38:33 ok i am bad with this whois stuff 12:38:55 so where from pa was she? 12:39:00 west part, east part..? 12:39:10 pittsburg 12:39:24 i think i have been to reding 12:39:30 you in HS? 12:42:27 nope 12:42:52 college? 12:42:57 middle school? 12:43:05 college 12:43:08 which one? 12:43:19 branch campus of penn state 12:43:26 state park? 12:43:48 it sucks, but it's only because a) i blew off high school or b) i am stupid, take your pick 12:44:58 nothing, i was hoping to get into better schools, but landed at purdue, and trust me i hate this place,,,its just that not like the east coast 12:45:18 however, you learn to live with what you have, and thats whats important 12:45:36 all you gotta do is take whatever you have, and use it fully 12:46:10 hmm that foo() function gives me a pagefault... 12:46:33 it shouldnt 12:46:45 bah 12:46:51 maybe my stack is too small? 12:46:59 could be 12:47:12 --- join: lar1 (~Larman@adsl-63-204-133-235.dsl.snfc21.pacbell.net) joined #osdev 12:47:13 you need to talk to geist 12:47:24 is he in band? 12:47:28 yes 12:47:30 it is called Kissing Contest 12:47:34 indie rock band :) 12:47:36 why do i need to talk to geist? 12:47:49 dax: because he knows more about os dev than everyone in this channel combined :) 12:48:27 already talked to him today... he already helped me find several bugs heh 12:49:03 --- join: jukka (jukka@MMCLXIX.hdyn.saunalahti.fi) joined #osdev 12:55:55 lol i need to rewrite my scheduler 12:57:58 firing 75 threads @ it kills performance 12:57:59 --- join: Rico (Rico@node-c-1c6a.a2000.nl) joined #osdev 13:01:08 heh 13:01:57 yea i'm limited by stack size 13:02:20 increased stack size and now foo(220) works 13:04:26 100 threads... no problem 13:04:32 (a bit slow though) 13:06:37 somehow i think my context switch code works 13:06:46 just being limited by stack size here 13:06:53 i'll work that out tomorrow 13:06:54 cya 13:10:25 dax: you develope winodws, oh, it can run more than 70 threads!! 13:10:25 wow 13:10:25 amazing 13:10:26 must be 2000mhz 13:11:04 anyone good with emacs macros? 13:12:24 ok 13:12:26 that's it! 13:12:30 bah to all of you! 13:12:40 what? 13:12:44 i'm sticking with 'nothing' - it is a good nick 13:12:47 heh 13:12:57 yeah, occassional laughter is always good :) 13:13:08 heh 13:15:31 laters dude 13:15:53 --- quit: zephir ("Client Exiting") 13:18:44 --- quit: Aardappel ("http://wouter.fov120.com/") 13:26:47 --- quit: Rico (Read error: 104 (Connection reset by peer)) 13:29:27 --- join: malenfant (malenfant@bc-vic-a53-02-70.look.ca) joined #osdev 13:38:23 --- join: cookin (~jrydberg@d212-151-89-101.swipnet.se) joined #osdev 13:41:28 chef, 13:41:29 ey 14:02:18 chef? 14:02:35 --- quit: Moo (Read error: 104 (Connection reset by peer)) 14:03:08 --- join: wossname (wossname@HSE-QuebecCity-ppp81716.qc.sympatico.ca) joined #osdev 14:08:00 cookin = chef 14:08:03 obvious :) 14:08:58 --- join: Spindle (~SpindleTe@12-249-117-17.client.attbi.com) joined #osdev 14:09:09 --- part: Spindle left #osdev 14:09:13 --- join: Spindle (~SpindleTe@12-249-117-17.client.attbi.com) joined #osdev 14:09:42 hehe 14:09:44 :) 14:11:14 cooking = making food = chef's job 14:11:23 why is it so hard to understand :) 14:13:33 chef in swedish meen "the boss" 14:14:39 but we talk here enlgihs 14:14:42 besides i know fairly well 14:14:48 No you don't :) 14:14:49 jag kan ju tala svenska.. 14:14:53 Your swedish sucks. 14:14:54 oij 14:14:57 nej, jag kan inte 14:15:01 förlåt 14:15:16 men, rob_ert, jag kan INTE tala svenska. 14:21:01 --- quit: jukka ("zz") 14:21:20 cookin = the boss? :) 14:21:26 sure is 14:21:27 :) 14:21:31 hehe 14:21:41 so, did you see that girl you saw on the subway lately? 14:21:46 did she yell at you again? :) 14:21:51 that was on the bus. 14:21:54 we have no subway here. 14:22:05 no, i have not :) 14:22:07 lucky me. 14:24:12 was she hot? :) 14:26:44 don't remember 14:27:13 Was programming languages did she prefer? 14:27:16 er 14:27:17 heh 14:27:20 What 14:27:23 she preferred Python 14:27:30 big long sexy python :P 14:28:33 --- quit: caereth ("BitchX-1.0c18 -- just do it.") 14:32:12 --- quit: rob_ert ("My life is my life, and it's per definition a real life.") 14:33:13 damn this is good music 14:33:20 http://www.explorespacenotdrugs.com/ <-- nice one 14:35:19 is there a limit to kernel size? 14:36:17 spindle: yes 14:36:23 spindle: the limit is your imagination 14:36:25 ;) 14:36:31 hehj 14:36:44 my imaginaiton only goes 512 bytes 14:36:47 no but really, there is no practical limit, but it obiously has to be within 4gb in size on the intel arch. 14:37:21 i assume you are writing a 32bit kernel 14:37:33 yup 14:37:54 --- quit: wossname ("roofles") 14:44:52 ah 14:44:53 cool 14:44:59 i just booted the latest torsion 14:45:04 are you here witten? 14:46:58 --- join: zephir (~zephir@harr-c-165.resnet.purdue.edu) joined #osdev 14:53:14 * zephir is away: for dinner 14:59:55 --- quit: ink ("bbl") 15:04:27 --- quit: cookin ("Client Exiting") 15:09:51 can anyone tell me a good info place on programming interrupts in pmode? 15:35:41 --- join: corsairk8 (~localhost@pc-62-30-121-110-cr.blueyonder.co.uk) joined #osdev 15:38:43 --- quit: gab (Read error: 113 (No route to host)) 15:40:21 --- quit: corsairk8 () 15:40:33 Zenton: www.intel.com 15:40:50 search for intel software developer's manual 15:40:52 * zephir is back (gone 00:47:38) 15:41:46 thanks zephir 15:44:34 --- join: gab (~prfalken@gaia.chx-labs.org) joined #osdev 15:45:03 --- nick: Zenton -> ZzZ_enton 15:45:21 g'night all 15:54:08 --- quit: trans (Read error: 110 (Connection timed out)) 16:16:54 Anyone know a good OS beginner book? 16:35:26 --- join: LiteOS (~SpindleTe@12-249-117-17.client.attbi.com) joined #osdev 16:35:27 --- quit: Spindle (Read error: 104 (Connection reset by peer)) 16:44:06 --- quit: lynx ("BitchX-1.0c18 -- just do it.") 17:09:39 --- quit: dax (Read error: 110 (Connection timed out)) 17:10:23 --- join: trans (~trans@a2a02274.intergate.bconnected.net) joined #osdev 17:22:23 live crowd 17:32:13 shhh.. we're busy coding 17:43:59 --- quit: morton (carter.openprojects.net irc.openprojects.net) 17:43:59 --- quit: gremlin (carter.openprojects.net irc.openprojects.net) 17:43:59 --- quit: ZzZ_enton (carter.openprojects.net irc.openprojects.net) 17:45:13 --- join: ZzZ_enton (~vicente@8.Red-80-34-35.pooles.rima-tde.net) joined #osdev 17:45:13 --- join: gremlin (~gremlin@tethys.parkside.net) joined #osdev 17:45:13 --- join: morton (~nothing@pcp01518417pcs.reding01.pa.comcast.net) joined #osdev 18:09:29 hey morton 18:12:26 --- join: Spindle (~Harris@12-249-117-17.client.attbi.com) joined #osdev 18:12:30 Hey 18:12:52 Anyone know some good OS dev sites that are not OSDEV.ORG 18:13:17 google: search for triple fault club 18:13:35 we should have a google bot on this channel? anyone who can do that? 18:14:11 triple fault club eh? 18:14:43 eh man 18:28:36 --- quit: zephir ("Client Exiting") 18:29:16 --- quit: trans (Read error: 110 (Connection timed out)) 18:32:43 --- join: trans (~trans@a2a02274.intergate.bconnected.net) joined #osdev 18:36:16 eh 18:44:52 --- nick: geist-band -> geist 18:47:35 hiya 18:51:35 morton: I'm here now.. 19:07:30 zephir, an infobot? 19:08:00 I'd look in to running an infobot... 19:25:05 --- quit: trans (Read error: 110 (Connection timed out)) 19:26:06 --- quit: Spindle (Read error: 110 (Connection timed out)) 19:37:55 y is everyone so quiet 19:37:59 is it ones sleepy time 19:45:52 --- join: trans (~trans@a2a02274.intergate.bconnected.net) joined #osdev 19:55:26 Has anyone here made a simple GUI for their OS? 19:55:56 no but once i tried quickly 19:55:59 =) 19:59:00 heh 19:59:01 and? 19:59:18 oh i just sorta dropped it 19:59:55 just started to write a few include files 20:00:19 i see 20:00:25 when I start mine its gonna be vesa 20:00:35 800x600 in 32 bit color 20:00:45 then make a min Win32 type thing 20:00:48 for the windows 20:00:57 is it going to be fixed or something? 20:01:49 nah 20:02:12 mode 15h i believe is vesa 800x600x32 :P 20:02:26 I just don't want a crappy text based OS 20:02:33 cause there is only so much you can do with it 20:03:03 may i suggest laying it out so there is a small portion of code which interfaces to lowlevel stuff 20:03:38 that way it can be more independant of stuff like vesa 20:03:42 and even os 20:03:58 but can you use vesa in pmode!? 20:04:07 depends 20:04:33 i dont think you can use the interrupts 20:06:07 nope 20:06:18 so how the hell do you get graphics to da screen! :D 20:06:43 drivers 20:07:08 hence the advantage of indepedance 20:08:04 --- quit: malenfant () 20:08:12 any site on info for writing a driver? 20:08:39 check out freevga 20:08:52 i dont know the site but it is in google 20:09:34 --- join: malenfant (malenfant@bc-vic-a53-01-46.look.ca) joined #osdev 20:11:40 k, thanks 20:13:52 ok back 20:13:58 witten: are you here *now*? :) 20:14:44 gah 20:14:48 oh well, time to sleep 20:14:54 --- nick: morton -> n-zZz 20:18:35 morton: I am now :) 20:19:30 cool 20:19:37 i was playing w/ the new torsion build 20:19:41 got my bochs to work finally :) 20:19:54 i just click "make" in xemacs and it rebuilds torsion and runs bochs :) 20:20:04 now to do some actual hacking 20:20:09 well, tomorrow, have to sleep now 20:20:14 ok 20:20:23 are you working on it rite now? 20:20:29 nope, working on a final 20:20:32 brb phone 20:25:30 --- quit: gab (Read error: 113 (No route to host)) 20:27:37 ok back (temporarily) 20:27:57 n-zZz: you know about "make image" in my Makefile, right? 20:28:02 uh 20:28:05 of course 20:28:09 ok 20:28:10 why? 20:28:17 just wondering if that's what you're using :) 20:28:21 well 20:28:30 i do (make img && ./bochs) 20:28:35 that's an alias of mine in xemacs 20:28:40 ah ok 20:28:45 I do something similar from the commandline 20:28:52 yeah i did that too, but it was too much work :) 20:28:59 just one button click now :) 20:29:02 hehe 20:29:08 I just hit up, enter :) 20:29:09 anyway 20:29:17 I think I'll make an "x86" subdirectory 20:29:23 and try to do some refactoring 20:29:29 but not now 20:29:49 ok 20:29:51 cool 21:30:56 --- quit: trans (Read error: 110 (Connection timed out)) 21:42:48 --- quit: air ("[ http://www.qzx.com ] ") 21:42:55 --- join: air (brand@12-254-199-50.client.attbi.com) joined #osdev 21:55:40 --- quit: nbsp () 22:10:16 --- quit: air2 ("AIRC v0.1.8pre -- http://www.qzx.com/airc") 22:57:09 --- join: trans (~trans@a2a02274.intergate.bconnected.net) joined #osdev 23:59:34 --- quit: malenfant () 23:59:59 --- log: ended osdev/02.03.17