00:00:00 --- log: started osdev/03.05.29 00:07:55 --- join: Boney_ (~paul@dsl-203-113-204-5.VIC.netspace.net.au) joined #osdev 00:08:06 --- quit: Boney ("leaving") 00:08:25 --- nick: Boney_ -> Boney 00:15:29 --- join: witten (~witten@adsl-gte-la-216-86-199-140.mminternet.com) joined #osdev 00:22:39 --- quit: bono ("leaving") 00:25:58 --- quit: trans (Read error: 110 (Connection timed out)) 00:31:49 --- join: wl (philipp@pD9E2D7A1.dip.t-dialin.net) joined #osdev 00:51:25 --- quit: CLxyz (Read error: 104 (Connection reset by peer)) 00:56:25 --- join: int13h (~no@calcium.nelte.lt) joined #osdev 00:58:32 --- join: CLxyz (CLxyz@00-01-03-d2-a4-07.bconnected.net) joined #osdev 01:17:52 --- quit: Mathis ("good night") 01:20:50 --- quit: int13h (Read error: 104 (Connection reset by peer)) 01:25:02 --- quit: witten ("Client exiting") 01:25:11 morning 01:26:31 Good morning and welcome to the world of inreality! 01:27:53 Hoi Rico, mijn mooi meisje! 01:56:35 --- join: trans (eravlw@fatwire-201-195.uniserve.ca) joined #osdev 02:02:44 lodda..... 02:03:25 * Rico undresses his sister's clothes he's wearing 02:04:48 ... 02:05:57 transvestite 02:06:03 ! 02:08:24 ...!!! 02:14:25 --- quit: kyelewis (Read error: 104 (Connection reset by peer)) 02:59:46 --- join: ReKleSS (~ReKleSS@c17565.mckinn1.vic.optusnet.com.au) joined #osdev 03:17:51 --- quit: trans (Read error: 110 (Connection timed out)) 03:26:01 --- join: codacola (~codacola@210-54-197-104.dialup.xtra.co.nz) joined #osdev 03:30:22 --- join: redblue (star@ppp028.216-96-207.sherb.mt.videotron.ca) joined #osdev 03:30:50 --- part: codacola left #osdev 04:00:32 No Conversation? :( 04:00:39 * lodda = bored 04:25:10 --- join: Mathis (irc@pD9EAB2DD.dip.t-dialin.net) joined #osdev 04:25:22 hi 04:31:55 --- join: trans (bcevmj@fatwire-201-195.uniserve.ca) joined #osdev 04:38:36 hi Mathis 04:39:05 hi lodda 04:39:13 meine Wasserkühlung ist nun komplett 04:39:19 und funzt wunderbar 04:39:49 Hm. Is this a German channel? :> 04:39:58 yep 04:40:06 Mein Deutsch ist nicht sehr gut... :( 04:40:07 since only german people are talking... 04:40:09 hehe 04:47:10 NUR DEUTSCH. WIR HASSEN DICH. 04:47:29 rofl 04:47:38 ? 04:47:45 gutes deutsch 04:47:48 mir doch egal 04:47:49 Jaja 04:48:06 Sehr gut... wasserstoff und kartoffelsalat 04:48:10 Ach so 04:49:47 babbel kein blödes Zeug 04:50:37 lol 04:50:47 du bist... erm... can't remember the word for irritating :p 04:51:05 Hört ihr Rammstein? :> 04:51:12 noeh 04:51:32 shit... and I've got a german exam on wednesday. 04:52:12 RekleSS: dict.leo.org 04:52:27 ? 04:52:40 if you don't know a german word =) 04:52:46 ah, ok 04:52:46 thx 04:53:20 wasserstoff und kartoffelsalat mmmmmmm 04:55:48 :D 05:04:37 --- join: alamar (alamar@im.alamar.eu.org) joined #osdev 05:19:13 the dutch police found a great amount of drugs wgich is bigger than thier whole collection of drugs they found last year, and it was found in my home village!!! 05:19:56 *g* 05:20:16 lol 05:20:27 which village? *g* 05:20:39 Oostzaan 05:21:16 Rico: why didn't you find it befor? 05:21:27 I tried to hide it! 05:22:09 aww 05:22:27 how much drugs were it in total? 05:22:27 my clients are mad!!! 05:22:50 ??????????? 05:23:01 267.000 XTC pills and 117kg amfetaminpasta 05:24:11 thats what the police confiscate on one day in hamburg ;-) 05:24:21 267 XTC pills? 05:24:28 267,000 05:24:28 that's not much :( 05:24:30 ah 05:24:33 better 05:24:40 but where is the heroin...? 05:24:47 and the cannabis...? 05:24:49 oh 05:24:52 holland is the country of xtc 05:25:00 that's ourt main drugs 05:25:22 wl: do you live in Hamburg? 05:25:29 pot is partly legal here 05:25:34 i lived there 05:26:03 I'll be playing a game of Urban Terror now :D 05:27:06 !!! 05:27:09 another UT player? 05:27:10 rare.... 05:27:15 Oj, no! It's teh Rico. 05:27:18 Hoi, Rico! 05:27:46 --- quit: trans (Read error: 60 (Operation timed out)) 05:28:44 --- quit: ReKleSS ("cat /dev/bullshit > homework") 05:30:45 --- join: kernel2420 (1000@ppp-251-162.25-151.libero.it) joined #osdev 05:31:38 --- quit: kernel2420 (Client Quit) 05:42:15 --- join: mrMister (~andri@ti122110a080-0648.bb.online.no) joined #osdev 05:42:16 --- quit: Odin- (Read error: 54 (Connection reset by peer)) 05:46:28 rare? 05:46:30 is it? 05:59:09 --- join: [Mathis] (irc@pD9EA8FB5.dip.t-dialin.net) joined #osdev 05:59:23 <[Mathis]> hi 06:06:34 --- join: Mathis_ (irc@pD9EAB70D.dip.t-dialin.net) joined #osdev 06:07:41 --- quit: Mathis (Read error: 60 (Operation timed out)) 06:07:58 --- nick: Mathis_ -> Mathis 06:22:07 --- quit: [Mathis] (Read error: 110 (Connection timed out)) 06:30:38 --- quit: Mathis ("good night") 06:46:26 --- join: durrow (~dcox@216.136.56.71) joined #osdev 06:46:29 Howdy all 06:46:54 Gday.. 06:48:43 --- quit: durrow (Client Quit) 06:53:28 --- quit: dax (Read error: 104 (Connection reset by peer)) 07:02:51 --- join: dax (dax@u212-239-163-28.adsl.pi.be) joined #osdev 07:04:41 --- join: trans (kzipjw@fatwire-201-195.uniserve.ca) joined #osdev 07:24:42 --- join: kernel2420 (1000@ppp-160-164.25-151.libero.it) joined #osdev 07:25:52 hello 07:25:52 i have some problem with ld script 07:28:50 --- join: zwane_ (User-10542@modemcable204.207-203-24.mtl.mc.videotron.ca) joined #osdev 07:30:57 I have win2k3 running on my laptop, it works nice, so far. 07:32:53 Rico: nice.. stolen? 07:34:13 I've got xp on another partition... rarely use it. 07:41:26 --- quit: zwane (Read error: 110 (Connection timed out)) 07:41:32 --- quit: kernel2420 ("www.italios.it") 07:42:12 kernel probably didnt know what his problem was 07:56:21 --- join: Mathis (irc@pD9EAB70D.dip.t-dialin.net) joined #osdev 07:56:33 hi 07:59:55 --- quit: zwane_ (Read error: 104 (Connection reset by peer)) 08:04:54 --- quit: trans (Read error: 60 (Operation timed out)) 08:14:57 --- quit: mrMister ("Client exiting") 08:14:58 --- join: mrMister (~andri@ti122110a080-0648.bb.online.no) joined #osdev 08:15:41 --- join: zwane_ (User-10542@modemcable204.207-203-24.mtl.mc.videotron.ca) joined #osdev 08:20:24 --- quit: mrMister ("Client exiting") 08:20:25 --- join: mrMister (~andri@ti122110a080-0648.bb.online.no) joined #osdev 08:29:52 --- join: wl_ (philipp@pD954E412.dip.t-dialin.net) joined #osdev 08:31:06 --- quit: zwane_ (Read error: 60 (Operation timed out)) 08:39:08 --- join: |Mandrake| (~EuEuEuEu@rdis.esec-manuel-fonseca.rcts.pt) joined #osdev 08:39:23 --- join: zwane_ (User-10542@modemcable204.207-203-24.mtl.mc.videotron.ca) joined #osdev 08:42:49 --- quit: |Mandrake| (Client Quit) 08:48:47 --- join: durrow (~dcox@216.136.56.71) joined #osdev 08:49:13 howdy all 08:49:59 Hi :) 08:50:46 --- quit: wl (Read error: 110 (Connection timed out)) 08:51:23 i was wondering if anybody in the channel has implemented stack based task switching 08:51:42 stack based? 08:51:48 err software based 08:51:50 sorry 08:52:16 not me 08:52:22 http://robert.zizi.org/ <-- check Primula. 08:52:29 It's crap, but the task switching works. :P 08:53:16 you can replace crap with asm :P 08:54:13 i guess i have more of a design issue 08:54:16 Hah., 08:54:21 Not really... 08:54:26 Crap == no design ;) 08:54:56 i am wondering how to save the stack off on the interrupt into C data structure 08:54:58 so i do a push 08:55:01 then mov eax, esp 08:55:07 call myFunction 08:55:16 myfunction now has the address of esp 08:55:24 how do i grab the rest of the data 08:57:34 Uhm.. 08:57:51 sorry - i should be more exact 08:57:53 Have several stacks. 08:58:04 Copying stack contents would take too much time. 08:58:12 pusha; mov eax, esp; call myFunction; mov esp, eax; popa; retd 08:58:13 Each process has its own stack. 08:58:28 Oh, so you want to save the registers that way? 08:58:33 yes 08:59:04 make sense? 08:59:15 Yes. 08:59:20 whew! 08:59:37 that's great news - i thought after all this time i still didn't get understand it 08:59:39 Check pit.asm in Primula. 08:59:55 There's where I save the registers. 09:04:10 yes in assembly it seems much easier 09:05:21 Things like these should be coded in assembly language. 09:05:40 Short snippets of code, which needs to be fast. 09:06:34 --- join: Ishq- (~Ishq@frm-64-4-102-68.access.ntelos.net) joined #osdev 09:06:43 --- quit: Ishq (Read error: 104 (Connection reset by peer)) 09:07:07 --- quit: Ishq- (Client Quit) 09:07:19 --- join: Ishq (~Ishq@frm-64-4-102-68.access.ntelos.net) joined #osdev 09:26:48 --- quit: redblue (Read error: 104 (Connection reset by peer)) 09:27:44 okay... let's try this line of thought 09:27:57 "pusha" "push esp" "call fptr" 09:28:06 then i should be able to print out the values of the stack 09:28:24 so i do printk("0x%x\n",passedValue) 09:28:29 that should be the value of esp right? 09:29:45 --- join: redblue (star@ppp033.216-96-207.sherb.mt.videotron.ca) joined #osdev 09:29:56 adding +4 to passedValue should lead to each registers value correct? 09:30:37 --- quit: HeavyJoost ("robot, err, reboot =)") 09:31:29 why arent you just trying it out? 09:32:58 i am 09:33:02 i swear i am 09:33:06 [crying] 09:33:44 the values i'm getting don't make a lot of sense 09:34:09 debug shell 09:34:33 whoops - wrong window :) 09:37:01 the values i get are: esp: 0x10a9e8 edi: 0x10a9f8 esi: 0x10aa08 ebp: 0x10aa18 esp: 0x10aa28 ebx: 0x10aa38 ebx: 0x10aa48 ecx: 0x10aa58 eax: 0x10aa68 09:37:28 durrow, what do you wanna tell that with us? 09:38:30 us with that... 09:41:13 mostly looking for help to make sure i'm not just being a putz 09:41:21 hm 09:41:22 i have a feeling i am just being a goof 09:41:51 maybe you have to learn first how to code yourself... 09:42:17 i have been struggling with the software based task switching for a while now and i have lost all confidence in what i had working 09:42:19 --- join: trans (nnjcvy@fatwire-201-195.uniserve.ca) joined #osdev 09:43:02 ouch 09:43:13 hehe durrow, we seem to be on same waters. but Im just trying to do normal kernel threads without even TSS 09:43:36 supposedly so simple, yet i havent had any success yet 09:43:38 bleh 09:48:11 it's interesting that so many folks choose TSS over software based 09:48:28 robert is only the second person i've seen that has used software based 09:48:35 which looks all the more complicated to me 09:48:45 :) 09:49:14 it's well documented however - therefore a bit easier. there's a whole book on it (protected mode software architecture) 09:52:51 --- join: jsr (www@visitor-225-200.birdie13.se) joined #osdev 09:54:39 --- join: HeavyJoost (~HeavyJoos@a213-84-139-110.adsl.xs4all.nl) joined #osdev 09:55:12 sure 09:56:44 did i miss something out? 09:56:51 why did you say "sure"? 09:57:20 :) guess he was telling me something in PM. wrong window prolly 09:58:44 ah ok 10:01:54 --- join: mur (jukka@baana-62-165-185-222.phnet.fi) joined #osdev 10:02:09 hey mur 10:02:41 * mur murr lodda 10:02:43 hello all 10:02:45 --- join: thib (~thib@bofh.bitcode.org) joined #osdev 10:15:56 --- join: Odin- (~sbkhh@adsl-2-216.du.snerpa.is) joined #osdev 10:15:56 --- quit: zwane_ (Read error: 104 (Connection reset by peer)) 10:16:10 odin snerppu 10:22:16 mawg. 10:22:43 hello Smaragdi 10:23:27 --- join: zwane_ (User-10542@modemcable204.207-203-24.mtl.mc.videotron.ca) joined #osdev 10:24:25 Smari, suP? 10:24:35 mur: fínt sko., you? :) 10:24:41 nice 10:24:50 * mur been doing animations today :) 10:24:57 nice is just like enrish 10:25:02 i'll use that :) 10:25:21 yesterday i read 2 books.. about 600 to 700 pages (i dont even know :) 10:26:57 * thib sparkar í smára 10:26:58 ;) 10:27:18 * Odin- segir Smára að svara í sömu mynt. :) 10:27:24 --- quit: HeavyJoost ("robot, err, reboot =)") 10:27:24 thib: Hóra. 10:27:28 ;) 10:27:30 Smari: Tussa 10:27:54 mur: Pay attention, we're playing the prophanity game. 10:28:09 (is it "proFanity"?) 10:28:17 thib: Þarmasleikja. ;) 10:28:33 nahh, none of this. It wastes time! 10:29:33 Smari: Ah. You, of course, have none of that, as is all too common with you. 10:31:54 * mur predicts that the game will be doomed sooner or later 10:33:00 --- quit: trans (Read error: 110 (Connection timed out)) 10:33:46 --- join: witten (~witten@adsl-gte-la-216-86-199-140.mminternet.com) joined #osdev 10:38:01 --- quit: thib ("Client Exiting") 10:41:12 --- quit: alamar (Read error: 110 (Connection timed out)) 10:41:58 folks, if i do a function(a,b), then how will the system stack change? first 'a' would be pushed, then 'b' and then the 'address of instruction following function(a,b)' will be pushed? is the order correct? 10:44:34 --- join: HeavyJoost (~HeavyJoos@a213-84-139-110.adsl.xs4all.nl) joined #osdev 10:55:40 mors: I would think it depends on the compiler 10:57:34 hmm. any idea about gcc? 10:57:48 nope, sorry 10:58:05 i think it does: 10:58:07 push b 10:58:10 push a 10:58:15 call function 10:58:20 but i am not sure 10:58:59 hmm. i was wondering if it was like that.. 10:59:27 i think lodda's right 10:59:29 anyway, gcc -s test.c gives me the assembler code. but in AT&T syntax. any idea how to generate intel syntax assembly code 10:59:40 mors: no 10:59:41 oh okay durrow 10:59:54 hmm. ok 10:59:58 --- quit: HeavyJoost ("robot, err, reboot =)") 11:00:20 push a 11:00:30 push b 11:00:30 call function 11:00:30 guess i have to live with AT&T, its there everywhere in gcc code 11:00:33 is the right order for gcc 11:00:42 oh, you sure? Mathis 11:00:46 sure 11:00:52 cool. thnx 11:01:13 you could also find it out by trying it out yourself 11:01:21 by playing around with gcc 11:01:36 yeah. was trying to do what... but this AT&T syntax kinda discouraged me 11:01:40 that^ 11:01:42 * mur has cool suggestion if someone wants to experience somethign new 11:01:52 you coudl .. 11:01:53 you have to learn that syntax 11:02:02 .. buy helsinki-moscow-bejing train ticket! 11:02:04 if its on the lines of Matrix , im game, mur :) 11:02:11 it takes only a week 11:02:15 by train 11:02:25 yeah mathis , looks like that .. 11:02:49 you get some kinda commission for tickets you sell? 11:03:45 no 11:03:58 but i could take consultation money if you have loose money 11:04:52 just checked it's B then A 11:05:11 hmm. you mean , push b, push a, call function? 11:05:25 yes 11:05:25 *blink* 11:05:35 okay.... 11:05:40 11:05:50 :P 11:06:12 thanks durrow. 11:06:14 pascal is a then b 11:06:18 oh 11:06:21 :) 11:06:24 C is B then A 11:06:47 wtf does the x86 popa instruction discard the stack pointer? 11:07:27 popa does edi, esi, ebp, esp, ebx, edx, ecx and eax 11:07:32 [i think] 11:07:54 according to http://people.freebsd.org/~jhb/386htm/POPA.htm it throws away the esp value 11:08:29 that explains a few things 11:09:04 heh 11:09:16 I think I should just do manual push'es of the registers I want to keep 11:09:23 kinda silly to push the stack pointer onto the stack, isn't it? 11:09:24 yup :) 11:09:35 I mean.. what would you do with it? 11:10:02 how would you retrieve it, assuming you place it there for safekeeping whilst storing something else there? ;) 11:10:22 * Smari is tired. 11:10:25 i guess your right... 11:10:26 :) 11:10:31 heh 11:10:33 Smari, why? 11:10:43 too much fish? 11:10:56 Smari: well then why does pusha even push the stack pointer to begin with 11:10:57 ? 11:11:06 witten: does it push it? 11:11:09 yes 11:11:22 witten: I thought you just said it threw it away? 11:11:26 ttp://people.freebsd.org/~jhb/386htm/PUSHA.htm 11:11:28 http://people.freebsd.org/~jhb/386htm/PUSHA.htm 11:11:40 pusha pushes esp, and then popa ignores what was pushed 11:11:51 mur: indeed. I was working this morning.. and I'm poofed. ;) 11:12:00 haha. :) 11:12:07 did you checked the url i gave? :) 11:12:09 witten: Somehow that makes very little sense.. ;) 11:12:21 which is why I'm asking here :) 11:12:27 mur: rentacoder? aye.. it's.. formidable. But I signed up anyway. ;) 11:12:32 witten: Wait a sec.. 11:13:52 witten: According to a book I have here pusha and popa save and restore the stack pointer as well, respectively... 11:14:03 * Smari laughs.. 11:15:11 ok, weird 11:15:14 --- join: kaze (Kaze@ca-bordeaux-16-28.w80-8.abo.wanadoo.fr) joined #osdev 11:19:55 wow.. 11:20:10 this is really silly... :> 11:20:40 unless somebody can give me a good explaination for the usefulness of this, I'm going to decree IA32 an even worse architecture than I had previously assumed. 11:22:32 IT WORKS! 11:22:37 [dancing up and down] 11:22:43 software based stack love 11:22:44 :) 11:22:51 figured it out.... took forever 11:22:59 =) 11:23:10 i think i will write a tutorial for the world on this one 11:25:00 --- join: Kurt (~kmw@oc-nas-11-s34.cinergycom.net) joined #osdev 11:25:19 yup yup , please do 11:25:33 with proper code examples too :P 11:26:16 --- join: I440r (~I440r@dialup-67.29.205.31.Dial1.Cincinnati1.Level3.net) joined #osdev 11:29:35 --- quit: witten ("Client exiting") 11:34:26 yeah! :) 11:34:41 durrow: and please have AT&T assembly code. :> 11:36:09 --- quit: jsr (Read error: 104 (Connection reset by peer)) 11:38:02 --- quit: Kurt ("Connection reset by ear") 11:43:23 --- join: jsr (www@visitor-225-200.birdie13.se) joined #osdev 11:48:38 :) 11:52:46 --- join: kernel2420 (1000@ppp-151-198.25-151.libero.it) joined #osdev 11:53:34 * kernel2420 saluta * 11:53:40 moi 11:54:02 hell mur 11:54:12 >:) 11:54:35 --- join: revanthn (revanthn@202.9.183.213) joined #osdev 11:54:46 --- quit: I440r ("abort" damn - not enough spam in my mail box!"") 11:55:05 hello revanthn 11:55:57 mur,hi 11:56:01 * mur goes 11:56:03 cya 11:56:20 --- quit: mur ("maybe more animations, maybe not.. reading more >:)") 12:02:50 --- join: trans (qqcqez@fatwire-201-195.uniserve.ca) joined #osdev 12:10:38 --- join: Ghiottone (~ale@ppp-169-17.27-151.libero.it) joined #osdev 12:14:02 --- join: Phactorial (~Phactoria@dial36-56.sbm.net.sa) joined #osdev 12:14:05 Hey 12:18:12 Hi Phactorial 12:18:57 oy Robert ;) 12:22:22 --- quit: kaze ("Leaving") 12:23:33 --- join: nullify (1000@pool-141-153-179-229.mad.east.verizon.net) joined #osdev 12:30:44 --- quit: revanthn () 12:33:10 --- quit: Boney (Read error: 110 (Connection timed out)) 12:35:03 i have more problem with ld script 12:35:19 OUTPUT_FORMAT("binary") 12:35:19 ENTRY(_start) 12:35:19 SECTIONS 12:35:19 { 12:35:19 .text 0x10000 : 12:35:20 { 12:35:22 code = .; _code = .; __code = .; 12:35:24 *(.text) 12:35:26 . = ALIGN(4096); 12:35:30 } 12:35:34 .data . : 12:35:36 { 12:35:38 data = .; _data = .; __data = .; 12:35:40 *(.data) 12:35:44 } 12:35:48 .bss : 12:35:50 { 12:35:52 bss = .; _bss = .; __bss = .; 12:35:54 *(.bss) 12:35:58 } 12:36:04 end = .; _end = .; __end = .; 12:36:06 } 12:36:33 well 12:36:43 i use this script with this command: 12:36:55 $(LD) -Tlink.ld -o$(KERNEL) -L./kernel -L./i386 -L./hardware -L./libc -L./fpu_emu --whole-archive -t -lkernel -lhardware -llibc -li386 -lfpu_emu -Map kernel.map 12:38:10 and i want to have the binary that is egual with binary result of this comamnd: 12:38:21 $(LD) -Bstatic --oformat binary -o$(KERNEL) -L./kernel -L./i386 -L./hardware -L./libc -L./fpu_emu --whole-archive -t -lkernel -lhardware -llibc -li386 -lfpu_emu -Ttext 0x10000 -Map kernel.map 12:38:30 but the binary are different 12:38:33 why? 12:41:39 parameters are different 12:42:49 which? 12:43:19 if you cant see it you have a bigger problem that different binaries... 12:43:24 than 12:44:06 --- join: Factorial (~Phactoria@cache4-1.ruh.isu.net.sa) joined #osdev 12:44:23 --- quit: Phactorial (Read error: 54 (Connection reset by peer)) 12:45:08 --- nick: Factorial -> Phactorial 12:48:45 Mathis: you'll say to help me? 12:50:28 brb 12:52:58 --- part: Ghiottone left #osdev 12:55:24 --- quit: redblue (Read error: 104 (Connection reset by peer)) 12:55:39 --- join: int13h (~no@calcium.nelte.lt) joined #osdev 13:08:20 --- join: wossname (wossname@HSE-QuebecCity-ppp81628.qc.sympatico.ca) joined #osdev 13:08:30 kernel2420: no, just giving hints 13:08:54 --- quit: Phactorial (Remote closed the connection) 13:09:10 ok Mathis, thanks ^___^ 13:09:10 if you dont learn helping yourself you wont survive in this world 13:09:24 ehehehhee 13:12:27 --- join: voider|NA (~voider@ppp219-234.lino.sympatico.ca) joined #osdev 13:12:35 to demonstrate you the real world we will leave you alone with your problem 13:13:48 --- join: DRF (Daniel@host81-7-58-193.surfport24.v21.co.uk) joined #osdev 13:16:42 --- part: nullify left #osdev 13:18:35 --- join: witten (~witten@ip-64-32-131-193.dsl.lax.megapath.net) joined #osdev 13:25:25 --- quit: trans (Read error: 110 (Connection timed out)) 13:27:07 --- quit: durrow (Read error: 60 (Operation timed out)) 13:27:48 --- join: durrow (~dcox@216.136.56.71) joined #osdev 13:30:01 --- join: pavlovski (~tim@host217-44-184-5.range217-44.btcentralplus.com) joined #osdev 13:31:20 hello 13:32:09 Hi pavlovski 13:32:54 --- quit: voider|NA ("cya all") 13:36:50 --- join: thib (~thib@bofh.bitcode.org) joined #osdev 13:38:10 . 13:38:15 , 13:38:53 ` 13:39:15 ; 13:39:18 ~ 13:39:31 ¬ 13:39:38 nice 13:39:42 good pull 13:39:51 hi 13:39:57 hi int 13:40:07 can bochs emulate vesa lfb? 13:40:21 hmmm on megatokyo board there is a thread on this 13:40:30 i think it has issues 13:40:30 let me see if i can find it 13:40:41 yes, it can, but it can't go beyond 256 colours 13:41:55 can emulate 800x600 resolution? 13:42:29 I think so 13:43:41 it can 13:43:46 --- join: HeavyJoost (~HeavyJoos@a213-84-139-110.adsl.xs4all.nl) joined #osdev 13:43:57 http://www.mega-tokyo.com/forum/index.php?board=1;action=display;threadid=3725 13:44:03 that's the thread 13:44:06 OH MY GOSH! 13:44:10 Mathis gave a straight answer 13:44:22 lol 13:44:23 didn't tell you to find out yourself or tell you to jump off a bridge 13:44:35 the end of the world is coming 13:48:37 durrow: wow, that thread is a replica of the last 5 minutes here :) 13:48:47 someone: can Bochs do VBE 13:48:51 me: yes, but only in 256 colours 13:51:58 argh! my memory doesn't work 13:52:43 ? 13:53:09 what I'm doing is: 13:53:17 allocating some physical memory and mapping it 13:53:26 strcpy(virtual_address, "blah"); 13:53:35 getting the physical addresses of the block 13:53:38 passing them to the ATA driver 13:53:48 mapping them to a temporary region 13:53:53 reading a sector into the temporary region 13:53:59 unmapping the temporary region 13:54:02 reading from the original buffer 13:54:22 I'm sure that the sector is being read, but nothing is being written to memory 13:54:26 argh! 13:54:47 have fun debugging. :P 13:55:16 pff. 13:55:36 hm 13:56:33 hehe 13:57:12 --- quit: int13h (Read error: 104 (Connection reset by peer)) 13:59:08 ps this is a dual CPU system :o 14:00:33 --- quit: DRF (Read error: 110 (Connection timed out)) 14:02:17 --- nick: wossname -> wossname|terrist 14:03:23 --- quit: thib (Read error: 104 (Connection reset by peer)) 14:05:13 * pavlovski explodes 14:05:41 I would too. 14:05:44 Sounds like fun. 14:06:20 hmm 14:06:36 pavlovski: how do you look now? 14:06:39 it's only happening on this particular system 14:06:42 lodda: exploded 14:08:24 --- join: thib (~thib@bofh.bitcode.org) joined #osdev 14:14:16 --- join: Ubel (~jonn@arnarson.is) joined #osdev 14:41:56 rétt! 14:42:40 pavlovski: how does that look exactly? 14:42:50 and what kind of explosion? 14:50:01 --- join: damadman (~kdehl@as3-2-3.sgp.lk.bonet.se) joined #osdev 14:50:12 --- quit: damadman (Remote closed the connection) 14:53:19 stupid that is 14:55:43 --- join: trans (hbptkq@fatwire-201-195.uniserve.ca) joined #osdev 15:09:06 geist: you there? 15:09:16 yes 15:14:59 great... 15:15:04 hey! you're not geist! 15:15:05 --- join: ralix (Iceburg@kmp53-01-p114.gt.saix.net) joined #osdev 15:15:50 morning :) 15:15:59 Hi 15:16:14 heya Robert how you doing? 15:17:12 Not too bad. :) 15:17:23 Reading a bit about 3D clipping. 15:17:32 thats good to hear, its been ages since last i've been here 15:17:50 I started learning OpenGL programming since yesterday 15:17:54 :D 15:18:00 I've never done 3D programming before. 15:18:17 ralix: OpenGL will fry thy brain 15:18:39 Hooray for fried brains. 15:18:48 Robert: Indeed ;) 15:19:13 --- quit: thib ("Client Exiting") 15:20:40 hehe, yea opengl is pretty ugly thing 15:20:56 but its interesting 15:21:33 the opengl books are pretty good though 15:21:44 [i am thinking of the two sgi books - can't remember the title] 15:22:00 --- join: Ishq- (~Ishq@frm-64-4-102-68.access.ntelos.net) joined #osdev 15:22:16 --- quit: Ishq (Killed (NickServ (Ghost: Ishq-!~Ishq@frm-64-4-102-68.access.ntelos.net))) 15:22:22 the redbook i think it's called 15:22:52 * debug usually thinks of Red Bible == the 4.4BSD book 15:23:14 http://fly.cc.fer.hr/~unreal/theredbook/ 15:23:16 there 15:23:29 or updated for 1.1 15:23:30 http://www.dcc.unicamp.br/~lmarcos/courses/mc603/redbook/ 15:23:51 I have the redbook I am actually looking for the bluebook, anyone? 15:24:43 OpenGL blue book? 15:24:44 what's that? 15:26:20 yep, I read about it NeHe gamdev site 15:26:25 its one of the recommended books 15:26:35 --- quit: HeavyJoost ("r38007") 15:27:54 where is it recommended? 15:29:01 oh, seen it 15:29:39 oh - yes - i have both the red and book blue on the shelf 15:29:46 one is the programming guide the other is the api guide 15:30:08 sorry - but my bookshelf is now in the next room 15:30:19 ok, found them on Amazon 15:30:33 the Red Book is "The Open GL Programming Guide: The Official Guide to Learning OpenGL, Release 1.1 (OTL)" 15:30:56 the blue is the reference guide 15:30:57 the Blue Book is "OpenGL Reference Manual: The Official Reference Document for OpenGL, Release 1.1 (OTL) " 15:31:22 nehe is a great site for opengl work! 15:32:52 --- join: HeavyJoost (~HeavyJoos@a213-84-139-110.adsl.xs4all.nl) joined #osdev 15:34:18 durrow: yep 15:34:27 their tutorials are good aswell 15:48:04 --- nick: wossname|terrist -> wossname 15:48:40 --- quit: wl_ ("Quit") 15:48:40 --- quit: wossname ("=___=") 15:56:36 --- quit: trans (Read error: 60 (Operation timed out)) 16:12:22 --- part: kernel2420 left #osdev 16:14:27 --- quit: dh (Remote closed the connection) 16:15:18 * geist is here finally 16:15:25 good night 16:16:03 later lodda 16:19:49 wb geist, cheers lodda 16:20:51 --- join: dh (~gfafgawrg@213-35-248-12-dsl.kvm.estpak.ee) joined #osdev 16:20:53 here I am too 16:26:18 --- quit: Rico (Read error: 104 (Connection reset by peer)) 16:27:07 hi 16:28:35 werd 16:32:21 --- quit: Mathis ("good night") 16:39:22 --- quit: debug ("Client Exiting") 16:42:27 --- quit: Ishq- ("Sic Semper Tyrannis!") 16:42:31 --- join: Ishq (~Ishq@frm-64-4-102-68.access.ntelos.net) joined #osdev 16:57:01 can give me link/info/discussion/research regarding why C++ is a bad object orientated programming language? because although I don't program in the language I would just like to know why its not widey used in kernel/device/etc development and what so really bad about the language? 16:58:01 its used alot in commercial institutes with regards to commercial software, but not used alot in embedded development 16:58:03 it's usually not used in kernel/device development for a variety of reasons 16:58:17 1) kernel stuff was traditionally done in C, old habits are hard to kill 16:58:20 object-orientatedness is not the reason why it tends not to be used for system programming 16:58:37 2) requires more runtime to get to all of the features, hard to do some of the features in a really low level environment 16:59:05 3) requires a more in depth level of knowledge to understand all of the implicit stuff that goes on, C is extremely explicit 16:59:15 I don't think (2) is valid, at least not today 16:59:18 which is very important to low level coding where you have to watch absolutely everything you do 16:59:20 I'm doing file with C++ at the moment 16:59:23 *fine 16:59:23 you bet it is 16:59:32 try to do rtti and exceptions from interrupt context 16:59:37 so turn them off 16:59:45 re-read #2 16:59:50 RTTI is part of the 'bad' O-O programming that ralix mentioned 17:00:07 and the NT kernel, for one, is full of exception handling in C 17:00:08 "requires more runtime to get to *all* of the features" 17:00:25 you don't need to use all the features 17:00:32 but that's not the spirit of #2 17:00:38 I agree 17:00:49 anyway, silly argument 17:00:51 I wouldn't use RTTI even in user mode, because I don't believe it's ever necessary 17:01:01 point is it's hard to use some of the features of C++ in low level environments 17:01:12 what about the low-level code generated by C++ Compiler? 17:01:13 I believe (1) is the overriding reason, with a bit of (3) 17:01:14 whether or not pavlovski would use them even in user space isn't relevant to the discussion 17:01:29 ralix: more or less the same 17:01:41 if you compile C code as C++ it should be identical 17:01:56 features like templates will add more code, as will inling 17:02:07 so C has no advantages over C in code generation? 17:02:14 or rather no real advantages 17:02:16 4) there is a perception that c++ is slower, and most kernel people value speed a lot 17:02:19 also, the compiler will go off and generate various constructions, destructors and displacement method prologues automatically 17:02:20 err C++ 17:02:35 ralix: could you say that again? 17:03:02 so C has no advantages over C++ in regards to low-level code generation? 17:03:02 experienced devs know that c++ isn't by default any 'slower' than C given the same set of functionality 17:03:10 but it's a bit easier to write inefficient code in c++ 17:03:35 if you're writing C code in an object-orientated style (which IMHO you ought to be), you end up with the same object code as what a C++ compiler would give you for free 17:03:40 right 17:03:50 C++ makes object-orientated programming less hard work, if used correctly 17:04:02 if used correctly is the key phrase 17:04:07 C++ provides some features which I don't believe are necessary 17:04:19 and makes some O-O design patterns non-obvious 17:04:28 anyway, the end answer is that you need to be really proficient in c++ before you use it to write a kernel 17:04:38 e.g. I avoid function overloading, and I don't find many uses for operator overloading 17:04:38 on the other hand, you'd better be pretty damn proficient anyway to write a kernel 17:04:41 right 17:04:45 so that doesn't really hold too much water to me 17:05:02 I wrote a lot of O-O code in C the way I thought it ought to be, then started writing C++ code to do the same thing 17:05:15 but the C++ I write now is different from the C++ I wrote before I started writing O-O code in C 17:05:21 if you understand what I'm saying 17:05:28 simple OO is pretty easy to do in C, it gets quite a bit trickier to write complex inheiritance, etc in C 17:05:34 --- quit: zwane_ (Remote closed the connection) 17:05:37 basically, don't try to go writing MFC-style code in the kernel 17:05:52 ofcourse not! 17:05:53 :) 17:05:58 geist: I just find O-O in C to be too much typing 17:06:12 also, for the most part traditional kernels dont really need super complex OO or for that matter super complex code 17:06:27 I'd prefer to put the 'virtual' keyword next to a method declaration than go off and define a vtbl structure, then embed a pointer to it in my class struct 17:06:34 professional kernel code tends to be pretty simple in implementation, extremely complex in algorithm 17:06:48 NT is heavily object-orientated in places 17:07:07 look at the Object Manager, the IO Manager (device drivers) or GDI for examples 17:07:08 which is why lots of people look at kernel code and think it's much easier to write a professional quality kernel than it is 17:08:04 hmm 17:08:10 anyway did we answer your question ralix ? 17:08:14 so at the end of the day, C does have any major advantages over C++? and then again it boils down to the developer and the code written by that specific person 17:08:15 I guess C++ vs. C is a different way of expressing the same algorithms 17:08:38 ralix: I believe that C's main advantage is that it's better established in this area than C++ 17:09:23 --- join: revanthn (revanthn@202.9.183.189) joined #osdev 17:09:24 most professional and established kernels were started and the core was written at least 10 years ago 17:09:29 and c++ was much less mature then 17:09:47 pavlovski: so could it then be said that for example, Win32 Commercial development C++ would be have a better advantage? 17:10:11 ralix: you mean, is it a better idea to learn how to apply C++ to Win32? 17:10:59 no, I mean because C++ is so widey used in Win32 Development these days, more than C is, does this give C++ the advantage? 17:11:12 what advantage? 17:11:15 ie. it is better established than C 17:11:31 I thought we were talking about kernels, not Win32 17:11:42 C is better established in the systems programming area than C++ 17:11:44 pavlovski: I was just making an example 17:12:02 C was desinged to be a system programming language in the first place 17:12:08 Win32 is a C API, so it's equally easy programming Win32 in either 17:12:23 Dennis Ritche made up all the things that lacked in other languages and created C 17:12:34 and Windows class libraries only exist for C++, so you can't make the comparison there 17:12:57 --- join: zwane (User-10542@modemcable204.207-203-24.mtl.mc.videotron.ca) joined #osdev 17:13:16 i agree, but we going off the original topic 17:13:33 silly example, my excuses to all 17:14:24 anyway I think I have an understanding now 17:14:24 --- quit: Ubel ("Client exiting") 17:14:25 geist: I've got a funny bug that you might be able to shed some light on 17:14:29 or rather a basic understanding 17:14:33 pavlovski: sure 17:14:46 thanks geist, pavlovski, revanthn 17:14:54 no prob ralix 17:15:33 geist: ok, I'm trying to read a sector from the hard drive 17:15:48 what I'm doing is allocating a buffer on the kernel heap, and writing some text to it ("blah" or something) 17:16:03 then passing the physical addresses of that buffer's page to the ATA driver 17:16:29 it reads from the hard disk, maps the physical address into a temporary mapping region, and PIO's into that temporary region 17:16:36 it then unmaps and returns to the caller 17:16:51 the caller then prints out the contents of the buffer using the virtual address it got when it allocated the buffer 17:16:54 follow me so far? 17:17:30 okay 17:18:11 this works fine on single-CPU machines (including Bochs in 1-CPU mode) 17:18:41 but on my dual machine (either with both CPUs started, or limited to one) and on Bochs in 2-CPU mode, nothing gets written to memory 17:18:59 the ATA driver can read from the disk and write to the buffer via its temporary mapping 17:19:06 it can even read the bytes back that it's just written 17:19:16 but by the time the caller tries to print the buffer, all you see is "blah" followed by zeroes 17:19:34 all the code at this point is running on the same CPU (nothing jumps CPUs half-way through) 17:19:47 I wondered what you thought 17:23:37 back, sorry was talking to coworkers 17:24:39 * geist reads this carefully 17:24:44 --- join: revanth (revanthn@202.9.183.143) joined #osdev 17:26:01 --- join: madMilo (~trillian@plms16748-96.pool.007mundo.com) joined #osdev 17:26:01 so when you say pio it into that region you mean you take an interrupt or otherwise manually read from a io port or whatnot and write it out? 17:26:20 the BSP gets IRQ 14 and sets a flag 17:26:44 --- part: madMilo left #osdev 17:26:46 the processor which is doing the reading sees the flag and read 256 shorts from the controller 17:26:56 so you said it actually gets written to the second mapping of the page, but the original mapping doesnt see anything 17:27:00 hrm 17:27:07 it plugs each of these into the caller's buffer, which it has mapped to a different virtual address than it was originally 17:27:07 --- join: madMilo (~trillian@plms16748-96.pool.007mundo.com) joined #osdev 17:27:16 then unmaps this temporary mapping when it's finished 17:27:18 since you said you're sure it's running on the same cpu, it's not a TLB out of sync thing 17:27:39 it's as if the CPU isn't noticing the data when they're read from a different virtual address 17:27:57 I thought maybe it was a cache problem, but (a) I added __asm__("wbinvd") to the unmapping routine 17:28:08 nah, the cache on x86 is totally coherant 17:28:10 and (b) it's all on the same CPU anyway, especially on the dual machine with the AP disabled 17:28:54 so if you just try to do it outside of the hard drive, what do you get 17:28:57 ie, grab a page, map it 17:29:10 map it again, write to one, make sure it shows up on the second one 17:29:13 you mean fill the buffer with garbage instead of a sector? 17:29:19 --- quit: madMilo (Client Quit) 17:29:24 haven't tried that, but sensible bytes are coming out of the hard drive 17:29:36 just to establish that your mmu mapping code definitely works 17:29:44 I even printed the bytes coming out of the port, wrote them to the screen, copied them into DEBUG and disassembled them 17:29:47 or the code to lookup the address of the heap entry 17:29:48 saw the NT MBR code 17:29:57 --- join: madMilo (~trillian@plms16748-96.pool.007mundo.com) joined #osdev 17:30:04 try to make sure 'blah' is in the temporary mapping before you do PIO 17:30:10 I'm printing the phys addresses of heap pages (which is where the buffer comes from) and the pages being mapped temporarily 17:30:16 geist: good idea 17:30:17 to try to absolutely without a doubt verify that the two pages are really the same 17:30:31 cause it sure looks like they aren't 17:30:51 what kind of TLB synchronization are you doing? 17:31:02 invlpg after each page is mapped or unmapped 17:31:14 no inter-CPU TLB syncing 17:31:26 try doing a global tlb invalidate after ecah mapping and see if it goes away 17:31:33 and yeah, you'll have to do inter cpu tlb soon 17:31:39 aha 17:31:51 before buffer at e8000758: 00 00 00 00 ... 17:32:01 after buffer at e8000758: 33 c0 8e d0 ... 17:32:14 ...further down... 17:32:24 buffer at 90002758: 62 6c 61 68 ... 17:32:32 first line is ATA before PIO 17:32:38 second is ATA after PIO 17:32:43 last is caller after disk read 17:32:46 --- part: madMilo left #osdev 17:32:54 get the hard drive out of the picture for now 17:32:57 so the mapping is bollocks 17:33:05 that's just complicating the whole thing, verify the mapping is correct 17:33:19 scale back and make sure the basics are working 17:33:23 hmm 17:33:30 thats an extremely important skill to have when doing os work 17:33:42 because you can never really verify that something works without a shadow of a doubt if you started over 17:33:56 so its easy to just assume the mapping works and the problem is the pio code, for example 17:34:20 but you have to go back and re verify that the stuff you're depending on works to see if you just didn't hit an edge case that you hadn't seen before 17:34:31 --- join: trans (ncohaq@fatwire-201-195.uniserve.ca) joined #osdev 17:34:43 I *constantly* scale back and verify that lower level stuff works when I hit a strange problem like this 17:35:47 ok, probably too late to go investigating now 17:35:56 but it's good to know where the problem lies 17:36:02 cool 17:36:14 keep that debugging mentality in mind when you hit stuff like this in the future 17:36:24 never assume that everything works 17:36:30 be paranoid 17:36:49 my tendency is to think, "it's only a minor thing, I'll put in some printfs and I'll be on my way" 17:37:44 yeah, that'll bite you in the ass later 17:38:09 you need to be thinking 'oh wow an opportunity to reproduce something that is probably broken, better fix it now' 17:38:20 damn right 17:38:33 the worst thing at work is somebody saying, "It crashes!" 17:38:33 --- quit: mrMister ("gone") 17:38:43 then me running the software again and again and having it work perfectly 17:39:06 on Tuesday, the renderer would crash consistently on the third run with this particular demo 17:39:15 on Wednesday, and ever since, it's been fine 17:40:32 hmm 17:40:37 * pavlovski could be onto something 17:40:44 yep 17:40:53 with os work you have to be extremely thourough 17:40:54 this is what I think the problem is: 17:41:05 a bit of confusion over gcc's "m" asm constraint 17:41:12 I have: 17:41:13 bool ArchMapMemoryRange(addr_t virt, addr_t phys, addr_t virt_end, uint16_t priv) 17:43:00 --- join: pavlovskii (~tim@host217-44-176-38.range217-44.btcentralplus.com) joined #osdev 17:43:18 weird, got cut off 17:43:22 anyway, it's doing what I want now 17:43:29 goodnight! 17:43:42 --- quit: pavlovskii (Client Quit) 17:44:19 --- join: thib (~thib@bofh.bitcode.org) joined #osdev 17:44:20 --- quit: thib (Client Quit) 17:46:19 --- quit: revanthn (Connection timed out) 17:47:05 cool 17:47:20 --- nick: revanth -> revanthn 17:48:20 --- quit: Odin- (Read error: 104 (Connection reset by peer)) 17:53:47 Unix and Windows Compared 17:54:03 http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnucmg/html/UCMGch02.asp 17:54:33 multics wins! 17:56:04 geist,do you know about Van Jacobson 17:57:13 heard of it, dont know off the top of my head 17:58:39 he started our as Experimental Physicst and now he is a network expert 17:58:57 see how carriers chnaged when computers came in 17:59:02 er *change 17:59:22 he worked as a cheif scientist at Cisco System 17:59:39 oh 17:59:41 --- join: Boney (~paul@dsl-203-113-204-5.VIC.netspace.net.au) joined #osdev 18:00:40 --- quit: pavlovski (Read error: 110 (Connection timed out)) 18:05:11 what's the diff between a txt file in unix and a txt file from winows 18:05:21 end of line characters 18:05:24 I only ^M 18:05:27 unix uses just \n 18:05:27 have you guys heard about SCO sue'ing Linux, IBM? 18:05:30 windows uses \r\n 18:05:32 mac uses \r 18:05:44 ralix: yeah, lets not bring it here 18:05:51 oh sorry 18:06:07 well, you can if you want, I just dont want to hear about it 18:06:22 geist,thanx 18:06:28 nah I just wanted to inform those who havn't heard about it :) 18:06:59 that whole thing is full of so much misinformation and press hype that it's not worth discussing anywhere 18:07:48 yea I can imagine 18:09:00 anyway, geist I see NewOS is doing well :) 18:09:33 well, stuff is happening currently, and that's progress 18:09:41 been hacking on a ppc port 18:10:04 cool 18:10:14 every time I port it to a new platform or try to it shows some deficiencies in the design 18:10:19 which is cool, it gets better every time 18:10:45 --- join: Odin- (~sbkhh@adsl-2-216.du.snerpa.is) joined #osdev 18:11:30 well alot of my buddies have been telling me about it over the past couple of weeks, and I am was like you guys only hearing about it now 18:11:43 so the name is generally growing day by day 18:12:43 heh 18:12:57 it has all the hype, none of the progress of any true open source project 18:13:24 it's great to see it enter the prestigious ranks of other 'waste of time' hobby OS's 18:13:43 geist,no offense but true open source projects need teams working on them 18:13:46 yea its quite an archievment 18:14:13 revanthn: yep, and I'm 100% open to any help 18:14:27 in fact, big chunks of it are written by other folks 18:14:58 but frankly not many people with the necessary skills have tried to help 18:16:05 but anyway, I'm totally open to help 18:16:31 that's cuz most of the other skilled people are working on their projects 18:16:35 :) 18:16:35 right 18:16:40 I would help, but I am a far away from the required skills 18:17:36 gotta start somewhere 18:17:50 I dont discriminate based on skills, but I dont submit crappy code to the tree 18:17:53 :) 18:18:05 geist,what kinda memory allocator does newos use for kernel memory requests 18:18:14 standard heap 18:18:26 you mean for malloc style memory requests? 18:18:33 yeah 18:18:36 the VM deals in regions 18:18:44 some bigger kernel stuff uses regions 18:18:50 like the heap, for example 18:19:37 I've thought about writing a slab allocator, but haven't gotten around to it 18:20:15 what do you mean by "the VM deals in regions" 18:20:44 are they some kinda fixes size buckets 18:21:04 regions are chunks of virtual address space, multiples of page size 18:21:29 they map a backing store, like anonymous memory (zero filled backed by a swap file), a file, or physical memory 18:21:49 they can be cloned (contents of one can be mapped again) 18:21:59 ok g2g 18:21:59 ttyl 18:21:59 --- quit: revanthn () 18:22:03 heh 18:29:13 have you checked out yellowtab yet geist? 18:30:17 no 18:33:41 --- join: Iceburg (~hi@kmp53-01-p57.gt.saix.net) joined #osdev 18:33:59 --- quit: ralix (Killed (NickServ (Ghost: Iceburg!~hi@kmp53-01-p57.gt.saix.net))) 18:34:07 --- nick: Iceburg -> ralix 18:47:07 --- quit: zwane (Remote closed the connection) 18:55:40 --- quit: trans (Read error: 60 (Operation timed out)) 19:04:35 --- quit: coredump (Read error: 104 (Connection reset by peer)) 19:14:39 geist: you can steal my slab allocator 19:14:42 since novo wrote it for me :) 19:15:01 hmm, but it's gpl 19:15:03 oh well 19:17:25 --- join: zwane (User-10542@modemcable204.207-203-24.mtl.mc.videotron.ca) joined #osdev 19:18:52 --- quit: zwane (Remote closed the connection) 19:31:41 --- quit: ralix () 19:32:57 --- nick: msa -> matt___ 19:49:03 --- quit: sayke (No route to host) 19:49:20 --- join: zwane (User-10542@modemcable204.207-203-24.mtl.mc.videotron.ca) joined #osdev 19:50:19 anyone awake? 20:15:55 me 20:15:59 --- join: air (~brand@12-210-162-121.client.attbi.com) joined #osdev 20:16:14 --- quit: durrow ("BitchX: use it, it makes you bulletproof") 20:16:18 i had a great today, tell me if u can see any flaws 20:17:49 a little keychain device with a small processor that holds a credit card number, the device would have a couple contacts for something like serial and would plug into a credit card reader device 20:19:09 the device would power up the keychain device and send it a merchant id number, the device would then encrypt the credit card number and merchant number and send data back to reader 20:19:56 the reader would then send data to credit card company and they would decrypt and pay merchant, assuming sender was the merchant in the data 20:20:06 oh and price would also be in the encrypted data 20:20:21 and u could also have a browser plugin that did the same thing 20:20:45 air: if anyone steals the device, they can buy stuff as you 20:20:47 --- join: Kurt (~kmw@oc-nas-11-s232.cinergycom.net) joined #osdev 20:20:48 online merchant sends u an id, plugin encrypts price, merchant id and yer cc and sends to online merchant 20:21:00 witten: same with cc 20:21:08 yeah, so what's the advantage? 20:21:28 the merchant cant make multiple charges 20:21:39 yer cc number wont be sent over net 20:22:03 u could even add a fingerprint system if u wanted 20:22:16 but as is it is better than what we currently have 20:22:32 this would require both the merchant and the CC processor to be compatible 20:22:35 which is a huge change 20:22:43 i know 20:23:13 the device wouldnt cost that much 20:23:33 how would the encryption work? symmetric? public key? 20:23:42 how would key exchange / key distribution work? 20:23:54 no keys 20:24:03 the payment processor has to be able to decrypt the encrypted data! 20:24:34 the merchant has no keys 20:24:41 only u and the cc company 20:25:05 the merchant only ever sees an ecrypted packet 20:25:09 how does the merchant know that your device is sending the right amount? 20:25:19 because the cc company replies 20:25:26 ok 20:26:00 how do you know how much the merchant is proposing to charge you? a display on the device? 20:26:15 that would be smart too 20:26:26 and buttons on the device to approve the transaction? 20:26:28 cuz u cant trust what their machine says 20:26:44 would only need one button 20:26:54 ok, well this is getting more and more expensive :) 20:26:58 heh 20:27:17 would also be kewl if device held multiple cc numbers 20:27:37 would require second button to toggle them 20:27:49 visa and mastercard have similar programs, except without special-purpose devices.. when you make an online purchse, a java applet pops up, asking if you approve the charge. if so, you enter a special password that your cc issuer then verifies 20:27:55 but those would be deluxe models 20:28:05 heh 20:28:43 basic model has approve button, deluxe model has price lcd and second button for multi-cc 20:28:47 :) 20:28:58 ok :) 20:29:12 good luck getting anyone to implement this.. the industry is extremely slow to change 20:29:13 with current ccs u have no idea what the merchant charges 20:29:24 right 20:29:33 i was hoping one of u would steal my idea and get it done :) 20:29:43 well I work for a credit card payment gateway 20:29:53 but that doesn't mean shit if the merchants don't implement it too 20:30:02 ok, talk to yer boss 20:30:15 heh, we're not a hardware company :) 20:31:08 so, just pass it along 20:31:18 upto visa and mc 20:31:43 i thought u made the muppets? 20:31:47 they're huge slow to change companies as well 20:31:53 that was my last job 20:32:08 oh :) 20:33:03 we're just one little company.. not a lot we can do to get both merchants and cc issuers to change 20:33:28 but I admit I'd like it if I had to explicitly authorize every charge to my credit card 20:33:43 both the actual charge and the amount 20:33:58 and nobody ever sees yer cc number 20:34:11 --- join: trans (trmtpw@fatwire-201-195.uniserve.ca) joined #osdev 20:34:12 unless u run windows and yer machine is hacked 20:34:20 but thats outta my hands :) 20:34:46 well normally when you submit your cc number to an online store, it's encrypted via https 20:34:55 but yeah, the merchant can see the cc number 20:35:23 and then the merchant can be hacked and someone has yer cc number 20:36:27 yeah, which is why we offer a recurring billing service that lets merchants just store a 6-digit alphanumeric "billing ID" to stand in place for the CC number 20:36:46 oh and this would need to be transmitted from device to reader via copper or fiber for the ultra deluxe models :) 20:36:57 and then they just hit our service with 6FQ23P1 and we know to charge CC # 4444-1111-2222-3333 20:37:07 haha 20:37:11 wireless is not secure enuf, not that it matters cuz the packet is encrypted but... 20:47:05 time to go home 20:47:07 later 20:47:09 --- quit: witten ("Client exiting") 21:04:38 ohhh 21:04:46 * air wants to move to india 21:05:10 what language do they speak there, must learn it before the move 21:33:12 --- join: pengo (xtofu@p211-tnt3.mel.ihug.com.au) joined #osdev 21:34:16 hindi, english 21:36:05 why do you want to move to india, anyway 21:39:58 --- nick: Kurt -> CheapWhore 21:41:27 --- nick: CheapWhore -> Kurt 21:50:06 --- quit: trans (Read error: 110 (Connection timed out)) 22:00:49 --- quit: Kurt ("Connection reset by beer") 22:23:32 --- quit: pengo (Read error: 104 (Connection reset by peer)) 22:23:35 --- join: peng (xtofu@p211-tnt3.mel.ihug.com.au) joined #osdev 22:25:24 mors: the president of india dont like bill gates and he openly told developers over there to work on OSS and to use OSS 22:28:25 oh, a Braveheart 22:43:19 eh? 22:53:42 --- join: air- (~brand@12-210-162-121.client.attbi.com) joined #osdev 22:53:42 --- quit: air (Read error: 104 (Connection reset by peer)) 23:02:17 --- nick: air- -> air 23:05:13 --- join: trans (bnrlbe@fatwire-201-195.uniserve.ca) joined #osdev 23:08:45 --- join: sayke (~abuse@sttldslgw29poolA154.sttl.uswest.net) joined #osdev 23:50:16 --- join: ruiner (ruiner@ppp490.lax.centurytel.net) joined #osdev 23:58:49 --- quit: trans (Read error: 60 (Operation timed out)) 23:59:56 --- join: witten (~witten@adsl-gte-la-216-86-199-140.mminternet.com) joined #osdev 23:59:59 --- log: ended osdev/03.05.29