IRC log started Tue Nov 23 00:00:00 1999 -:- air [brand@p0wer.qzx.com] has joined #tunes [msg(TUNES)] permlog 1999.1123 -:- SignOff air: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep) -:- NetSplit: tolkien.openprojects.net split from varley.openprojects.net [01:07am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [tolkien.openprojects.net] -:- SignOff abi: #TUNES (Ping timeout for abi[bespin.dhs.org]) -:- abi [nef@bespin.dhs.org] has joined #tunes -:- Netjoined: tolkien.openprojects.net varley.openprojects.net -:- zarq [zarq@9dyn96.delft.casema.net] has joined #tunes -:- SignOff Fare: #TUNES (Read error to Fare[quatramaran.ens.fr]: Connection reset by peer) -:- Espresso^ [Espresso@212.150.33.102] has joined #tunes -:- Espresso^ [Espresso@212.150.33.102] has left #tunes [OmegaScript] -:- SignOff abi: #TUNES (Ping timeout for abi[bespin.dhs.org]) -:- SignOff zarq: #TUNES (Ping timeout for zarq[9dyn96.delft.casema.net]) -:- abi [nef@bespin.dhs.org] has joined #tunes -:- zarq [zarq@9dyn137.delft.casema.net] has joined #tunes -:- Fare [fare@quatramaran.ens.fr] has joined #Tunes -:- NetSplit: varley.openprojects.net split from asimov.openprojects.net [04:56am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [varley.openprojects.net] -:- Netjoined: varley.openprojects.net asimov.openprojects.net -:- abi [nef@bespin.dhs.org] has joined #tunes -:- zarq [zarq@9dyn137.delft.casema.net] has joined #tunes -:- SignOff Fufie: #TUNES (Ping timeout for Fufie[tunnel-44-13.vpn.uib.no]) -:- SignOff abi: #TUNES (Ping timeout for abi[bespin.dhs.org]) -:- NetSplit: tolkien.openprojects.net split from varley.openprojects.net [05:41am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [tolkien.openprojects.net] -:- abi [nef@bespin.dhs.org] has joined #tunes -:- Netjoined: tolkien.openprojects.net varley.openprojects.net -:- zarq [zarq@9dyn137.delft.casema.net] has joined #tunes -:- FareWell [fare@quatramaran.ens.fr] has joined #Tunes -:- FareWell [fare@quatramaran.ens.fr] has left #Tunes [] -:- epsas [epsas@aether.inflicted.net] has joined #tunes -:- AlonzoTG [Alonzo@209-122-203-68.s322.tnt6.lnhva.md.dialup.rcn.com] has joined #tunes -:- epsas [epsas@aether.inflicted.net] has left #tunes [] -:- hcf [nef@me-portland-us321.javanet.com] has joined #tunes hcf Fare Fare: whatcha want? a kick in my ass 09:30am for what? for not being the least productive whats not getting done? 09:40am TUNES, my thesis, etc, etc how can i "kick u in the ass"? 09:50am Fare: hello? Fare is buying you a ticket to France i dont travel in meatspace 10:10am -:- _ruiner_ [blah@207.230.192.153] has joined #tunes <_ruiner_> 'ello? hello _ruiner_ -:- SignOff AlonzoTG: #TUNES (Ping timeout for AlonzoTG[209-122-203-68.s322.tnt6.lnhva.md.dialup.rcn.com]) 10:20am <_ruiner_> whats up chief/ <_ruiner_> ? hmm we actually had some discussion yesterday smkl: about what? about functional programminging argh 10:30am <_ruiner_> hmmmmm.... -:- SignOff _ruiner_: #TUNES (Leaving) 10:40am -:- SignOff hcf: #TUNES (Leaving) -:- NetSplit: fontana.openprojects.net split from tolkien.openprojects.net [11:07am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [fontana.openprojects.net] -:- Netjoined: fontana.openprojects.net tolkien.openprojects.net -:- zarq [zarq@9dyn137.delft.casema.net] has joined #tunes -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) -:- Fare [fare@quatramaran.ens.fr] has joined #Tunes -:- smoke [smoke@15dyn120.delft.casema.net] has joined #tunes -:- SignOff smoke: #TUNES (Ping timeout for smoke[15dyn120.delft.casema.net]) -:- carlito [sabanmr@137.28.109.90] has joined #tunes abi: seen _ruiner_ _ruiner_ was last seen on IRC 3 hours, 15 minutes and 23 seconds ago, saying: hmmmmm.... [Tue Nov 23 10:30:25 1999] what a punk!!!!!!!!!! misoaa dmnoa -Gv-9qa guQ\ -:- SignOff carlito: #TUNES (Leaving) 01:50pm -:- binEng [Anders@j141.ryd.student.liu.se] has joined #tunes -:- smoke [smoke@15dyn175.delft.casema.net] has joined #tunes -:- SignOff smoke: #TUNES (Ping timeout for smoke[15dyn175.delft.casema.net]) -:- eihrul [lee@usr5-ppp200.lvdi.net] has joined #tunes -:- smoke [smoke@15dyn175.delft.casema.net] has joined #tunes -:- carlito [sabanmr@137.28.109.90] has joined #tunes abi: seen _ruiner_ _ruiner_ was last seen on IRC 4 hours, 2 minutes and 19 seconds ago, saying: hmmmmm.... [Tue Nov 23 10:30:25 1999] abi: seen carlito carlito was last seen on IRC 0 seconds ago, saying: abi: seen carlito [Tue Nov 23 14:32:53 1999] bye -:- SignOff carlito: #TUNES (Leaving) abi: seen core core was last seen on IRC 28 days, 5 hours, 22 minutes and 59 seconds ago, saying: hm, well, i'll be back :) you have a good day :) [Tue Oct 26 10:10:49 1999] abi: seen water water was last seen on IRC 22 days, 3 hours, 14 minutes and 5 seconds ago, saying: good point [Mon Nov 1 11:19:50 1999] 02:40pm seen eihrul -:- SignOff smoke: #TUNES (One day sheep will rule the world) -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp200.lvdi.net]) -:- AlonzoTG [Alonzo@207-172-49-70.s70.tnt7.lnhva.md.dialup.rcn.com] has joined #tunes -:- eihrul [lee@usr5-ppp200.lvdi.net] has joined #tunes 02:50pm * zarq/#tunes is back from being away: trug -:- _ruiner_ [blah@ppp145.wi.centurytel.net] has joined #tunes -:- SignOff _ruiner_: #TUNES (Leaving) -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp200.lvdi.net]) -:- eihrul [lee@usr5-ppp200.lvdi.net] has joined #tunes -:- SignOff binEng: #TUNES ( ) -:- eihrul_ [lee@usr5-ppp200.lvdi.net] has joined #tunes -:- NetSplit: adams.openprojects.net split from sterling.openprojects.net [03:30pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [adams.openprojects.net] -:- eihrul_ is now known as Leftix -:- Leftix is now known as eihrul -:- hcf [nef@me-portland-us800.javanet.com] has joined #tunes * zarq/#tunes is away: zzz * AlonzoTG/#tunes hurls a laser guided flaming woodchuck at zarq what's a woodchuck? (i understand i may get a big hit of it anyway) :) 04:10pm I'm not sure either... I tihnk its a cross between a beever and a squiril... om anyway... wetscrape crashed... -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) 04:20pm -:- air [brand@p0wer.qzx.com] has joined #tunes eihrul: u here? 04:50pm -:- Tril [dem@bespin.dhs.org] has joined #TUNES -:- mode/#tunes [+o Tril] by ChanServ air: yes i need to figure out some way to fix a problem when brix gets an irq it forks a thread and places it to run next s/places/sets/ why not just preempt instead? ya it does that so what's the problem? it adds it next in line and then calls the scheduler but the keyboard handler forks a thread for whatever is hooked to it that handles key input so umm, what's the problem? but if u press 2 keys really fast the second key will preempt that key hanlder 05:00pm why u have 1 thread for eahc key press? one key handler should always run, no? no air: guess you have to let enough of the handler run atomically to atleast register that the second thread should be buffered instead of just immediately sent either that or assign the second thread a lower priority so it can't preempt the first but it only affects bochs cuz bochs is so slow but it is still a problem that shouldnt be pushed aside it could plausably affect real hardware its the irq that is preempting the handler the irq calls the handler a second time? i dont use priorities in brix and even if i did they wouldnt help the keyboard irq forks the handler thread to run next that way the handler has higher priority than all other threads well... you're going to need some sort of synchronization... best way i can think of doing this: as the first thread enters the handler, set some sort of mutex when next thread comes in if the mutex is set, *yield* press 2 keys, irq thread 1, handler thread 1, irq 2, handler 2 that one yield should preserve ordering irq 2 is preempting handler 1 and not necessarily have troubles associated with a full mutex just need to worry if it gets preempted in the small period of time before you set the mutex... ya so that wont work welp, unless you let enough of it run atomically to set the mutex... then you're fucked :) unless you sort active interrupt handler threads.... i.e. if a second interrupt happens before first handler is done, shove it behind second could be done easily just maintain a pointer to the last interrupt task still active... eh? er thread and insert the new interrupt-invoked thread after it. hmm 05:10pm that would work that's actually necessary assuming your threads run to completion its the only way to preserve ordering :) one pointer per int and just update it to the new int they do should need just one pointer globally very few threads should run past one timeslice unless they are waiting on somethin one per irq why? i don't see any particular benefit for the added complexity would probably be slower as well one interrupt is jsut as important as another so interrupts really shouldn't be preempted but then the handler to a higher priority int could be set to run after a lower priority int if its higher priority that's not a problem... my scheduler handled these automatically all newly created threads were scheduled to run *LAST* within their priority that way you couldn't starve other threads by repeatedly destroying and creating a thread it also ensured highest priority thread ran first i assume you're going to have trouble without a priority scheduler, such is life well only drivers can fork threads next, normal threads fork last shrug, i'd put in a real priority scheduler if i were you... nah with realtime features! any thing you make is going to amount to a crude priority scheduler with all the apparent complexities of one without necessarily being a priority scheduler air: i.e. your one-pointer-per-irq sounds remarkably like my priority scheduler implementation :P 05:20pm the two parts are orthogonal and can be just as easily separated as combined -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) all i did was have one pointer per priority level damn, how many levels? shrug, you could just put as many levels as interrupts 05:30pm -:- yoo [ultima@user-38lcn72.dialup.mindspring.com] has joined #Tunes EiHrUl -:- _ruiner_ [DIY@ppp441.wi.centurytel.net] has joined #tunes hey yoo 06:20pm eihrul: that pointer thing wont work -:- SignOff _ruiner_: #TUNES (Ping timeout for _ruiner_[ppp441.wi.centurytel.net]) air: why not? its priority scheduling damn it when u say pointer do u mean a pointer to a list or pointer into a queue -:- AlonzoTG [Alonzo@216-164-136-139.s139.tnt4.lnhva.md.dialup.rcn.com] has joined #tunes oops nevermind 06:30pm -:- SignOff Tril: #TUNES (Tril has no reason) -:- SignOff hcf: #TUNES (Leaving) -:- lar1 [lar1@dialup-209.245.138.198.SanJose1.Level3.net] has joined #tunes Hey this is so much fun writing drivers im actualy coding shit that does stuff now Drivers are fun!!! 07:20pm -:- SignOff zarq: #TUNES (Ping timeout for zarq[9dyn137.delft.casema.net]) air: gah... -:- eihrul_ [lee@usr5-ppp66.lvdi.net] has joined #tunes -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp200.lvdi.net]) -:- eihrul_ is now known as eihrul gah? think about it, i've been writing a kernel that is hard to test and does nothing now im writing drivers that do something and the more drivers i write the larger base of code i have to work with so i can do even more stuff that's why i said gah i like to see results i'm jealous :P writing a kernel produces no immediate results 07:40pm si and once i finish this driver i will have a working store which will make it tons easier to write stuff si? grr -:- lar1 is now known as lar_eating its 9pm and yer eating? time zones... 8 pm where he is i believe ya so thats too late to be eating shrug... says you no time to burn it off 07:50pm -:- _ruiner_ [DIY@ppp324.wi.centurytel.net] has joined #tunes -:- SignOff _ruiner_: #TUNES (destroy what destroys you) -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) What? 08:00pm -:- Fare [fare@quatramaran.ens.fr] has joined #Tunes -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) -:- SignOff yoo: #TUNES (Ping timeout for yoo[user-38lcn72.dialup.mindspring.com]) -:- yoo [ultima@user-37kbah3.dialup.mindspring.com] has joined #Tunes -:- lar_eating is now known as lar1 -:- zarq [zarq@10dyn191.delft.casema.net] has joined #tunes -:- core [core@core.suntech.fr] has joined #tunes people hi CORE!!! hey mr. air :-) Core!! wow, red carpet and all hey lar1 ;) wow, you're alive! uh, yeah :) busy and annoyed that days are only 32 hours long, but alive :) 24 hours, core ;) oh? that'd be why i always have to cut out my 8 hours of sleep then :) eh heh core: I need help lar1: about ? they should have made days 32 hours long, easier to mask. core: Air gave up. Fixing Bochs bochs says he dont have X devel installed lar1: fixing bochs ? what's broken? core: they being? oh, it can't find xlib? Warn -- large paste-- c++ -o bochs -g -O2 main.o load32bitOShack.o state_file.o pc_system.o \ \ iodev/libiodev.a \ cpu/libcpu.a memory/libmemory.a gui/libgui.a \ debug/libdebug.a disasm/libdisasm.a \ fpu/libfpu.a \ \ -lX11 \ i don't know, "they" who decided on our time layout ;) \ checking for X... (cached) no collect2: ld returned 1 exit status make: *** [bochs] Error 1 lar1: remove config.{status,cache} first lar1: and it doesn't look like it's including any library path in that core: Ok, I rmed the files you specifyed lar1: ./configure again, i don't know if there's an option to add libraries dirs or to pass X11's core: Ok how does configure find X11? It found my X libs now (configure did) must be looking for a default path or accepting new ones somehow lar1: ah, cool. core: Doing that is the last thing I would have thought of lar1: did u install the x devel stuff after u first configured bochs? air: Yes oh *SMACK* lar1: the "(cached)" is what made me tell you to remove that :) That was stupid 10:50pm i need to tell kevin to get rid of the read only files in bochs Why not just make an open source version of bochs? bochs is open source I mean not costing money version or technically costing money kevin has a commercial license so u cant do that his license is cheap, come on :) u would need to rewrite it from scratch air: Thats what I am talking about lar1: have fun the x86 emulation will be fun to write, i think; the rest (reverse emulation of all ISA/PCI legacy devices) will be hell. True, I forgot about doing all the ISA and PCI crap i will pay him my $25 when he fixed "jmp short $+2" x86 would be fun though s/fixed/fixes/ 'when he fixed' made sense too :) er jmp short $ air: You should get a free licence for making BFE what's wrong with jmp short $ ? i used that quite a lot in the past. Do I have to have a "hard drive" file for bochs to use? bochss doesnt idle down why would it idle down? lar1: u can have a zero length file that's what hlt is for air: Ok lar1: just 'touch hdfilename' * yoo/#Tunes wonders if bfe does anything core: ok u tell me howto use hlt in ring 3 code air: in one word? don't :) yoo: have u used bfe? air: use a yield() or wait() of sorts air: Nope air: I remember when you were writing it and it did nothing though =) core: its for testing purposes What does touch do? im not adding another kernel function lar1: creates a zero length file air: you will need a kernel function to suspend a thread, _anyway_ air: Ah core: i have one air: a thread shouldn't spin; it should just relinquish control, and your idle thread running in kernel mode will loop on hlt air: why don't you use it then :) sleep, event_wait, thread_wait, suspend lar1: touch overwrites the date on a file with current data... kinda a nil write effect, which can also create a zero length file air: use one of those then :) lar1: er current date core: like i said above, its for testing core: i CANT use one of those 11:00pm air: you want your code to spin somewhere forever? *shrug* trap the gpf on hlt in kernel mode and execute it then, i don't know :) core: and for some funny reason when i run bochs and it sits in that jmp for about 30 seconds it crashes linux air: it crashes *linux* ? what do you mean? smp bugs linux freezes up solid air: hrm; i had jmp $ 's in lots of places a year and some ago while i was early into devel stages, and it just sat there happily. i cant ping or telnet to it, i cant move the mouse or type air: oh, your OC'ed CPU doesn't like it? :P hey its bochs fault Its yer cpu no it aint it can't be an application's fault if the entire OS freezes; it's either the OS or the hardware not doing its work :) well its linux but bochs made it happen whats the current version of linux? 1.2.13 I think What do I put for the chs of an empty file? 2.2.13 13? did linus die? i downloaded 12 months ago HE was at comdex no.. it's in 14pre375really_we_aren't_shitting_you_14_will_be_out_soon Wait latest is 3.3 wonder if there are any smp updates going from 5 to 12 did wonders in removing smp bugs er 7 to 12 core: You like anime? try clocking your cpus to their expected speed and try again :) core: 366? dont think so lar1: it depends; i like very early anime like goldorak or cobra (if that ever played in the US), or very recent that i saw in theaters, like porco rosso, jin-roh, etc :) air: buy some 466 ers they are cheap air: try it :) lar1: im running at 550 air: clock them down and run that again, see if it hangs. if it doesn't.. buy cheap 466's :) core: Hmm, well check http://www.tunes.org/~lar1/art2.gif anyways... I drew it lar1: do you know any of those, or did they only come out in france :) im not running on a 66mhz bus core: I haven't heard of them core: But I am not a very athoritative sources and i dont run cpus at the suggested mhz i thought celerons only worked fine on a 66 mhz bus lar1: ah ok :) well, expect to run into problems then :) air run 100x3.5 in a single celery system u can run these at 567/103 11:10pm beos can run dual at 460/83 and linux can run dual at 550/100 *shrug* i run cpus at the speed i bought them for and system has been up for 80 days ;) lar1: cant change multiplier the only problem i have is at work with a 13.5 GB IBM disk that doesn't seem to want to work with linux :( air: WHAT?!?! air: Multiplier locked? When did this happen? lar1: all p2's are core: you have 2gb fiber so don't complain ;) lar1: but i can't download all that stuff without a large disk ;) air: the celery has always been multiplier locked? lar1: yes lar1: thats why u buy lower mhz celeries air: what multiplier are they locked at? like the 300A and 366 air: the 466 air: what does it bring you to run them higher than their speed tho? :) mhz/66 7 knowing that my system is running much faster than some idiot that isnt oc'ing his box the hacker side of me that has to tap every drop of power from the hardware? So I coudn't run 466s at 100mhz bus? lar1: uhhh no :) 700 mhz could happen ;) I was planning on that! Dammit air: i'm some idiot not OC'ing his box :) intel had a hard enuf time getting the p3 to do 700 air: well, i understand - it's just, i have a visceral fear of the hardware locking on me and having to redo my work. so i just try not to do that :) yesterday i almost broke a project manager in half at suntech; he managed to drop a fucking jumper in my power supply ventilation. (while i was installing the disk) Hehe wow you should have plugged the supply in and made him get it out =) What is up with Xlib: connection to ":0.0" refused by server Xlib: Invalid MIT-MAGIC-COOKIE-1 key repeated twice? (starting Bochs, or vi) 11:20pm lar1: did you su? are u running it locally? air: Yes lar1: unless you use xhost, you can only start programs with the same user that started X. core: Ah let me guess yer running x as root? air: No air: I was doing some stuff as root and forgot to exit air: So I tryed to run bochs as root i can run bochs as root Crap Who feels like installing my vga font? thats easy Yay! bochs comes with a howto install it file :) It doens't work I tryed it that same file tells u howto make the hd image or just use a post-jurassic install of X and you'll have the vga font already. core: the bochs vga font? air: or another, but i don't remember installing the vga font on any XFree 3.3.5 boxes air: Could you tell me how to install the font? last time I folled the directions and it got all messed up and you had to telnet in to fix it This time I have telent disabled and I don't know how do undisable it :) -:- rix [RiX@dyn1-tnt10-45.detroit.mi.ameritech.net] has joined #tunes ummm copy it to /usr/lib/X11/fonts/misc/ a post-jurassic install of X? X has never included the vga font afaik then gzip it then run mkfontdir from that dir air: vga.pcf? ya im not seeing how anything could go wrong there :) oh ya MESSAGE (yoo ultima@user-37kbah3.dialup.mindspring.com) xset fp rehash ;) * yoo/#Tunes hides Now all I need is BFE and I am back to coding RavenOs Oh, how do I "install" bochs? i think bfe might be used to develope lynxOS 11:30pm There is no make install and there should be air: lynx os? www.lynx.com its a realtime os commercial os why do you think? the lynx website isn't lynx friendly cuz they emailed me about it ironically enough hehe what'd they say? Ah fsck I don't have xforms lar1: goto the bfe page eihrul: they said it kicked ass air: The link is dead oh no its just down it goes down alot air: Ok i will dcc u the file Thats ok I got the .deb 0.88? Yes I don't need the doc or devel, right? uhh