IRC log started Fri Oct 15 00:00:01 1999 [msg(TUNES)] permlog 1999.1015 -:- SignOff eihrul: #TUNES ([x]chat) -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[11-134.006.popsite.net]) -:- fare [fare@quatramaran.ens.fr] has joined #Tunes morning fare 02:40am night, Tril! still up? sort of 02:50am playing War2 Battle.net edition a new game to try out: http://www.woodsoup.org/projs/tux_aqfh/ 03:00am -:- fare is now known as Fare61453 -:- Syntax [dma@04-202.006.popsite.net] has joined #Tunes -:- IvanJoeI1 [user7653@ctv21225131051.ctv.es] has joined #tunes hola colegas, alguien habla español? -:- IvanJoeI1 [user7653@ctv21225131051.ctv.es] has left #tunes [] 04:20am -:- SignOff smkl: #TUNES (Ping timeout for smkl[glubimox.yok.utu.fi]) -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- NetSplit: bear.openprojects.net split from varley.openprojects.net [05:40am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [bear.openprojects.net] -:- Netjoined: bear.openprojects.net varley.openprojects.net -:- Fare61453 [fare@quatramaran.ens.fr] has joined #Tunes -:- fire [no@209-68-229-53.dialup.cust.tfb.com] has joined #tunes anyone awake? well i am ah good morning from california 06:30am good afternoon from finland moi how's the weather there? hmm it's fine now but it has been raining a lot it has been very hot here for the past couple days :( 06:40am -:- Synner [dma@10-070.006.popsite.net] has joined #Tunes -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[04-202.006.popsite.net]) -:- Synner is now known as Syntax bye -:- SignOff fire: #TUNES (Leaving) 07:00am -:- IMPO [ihatebob@p06-35.hartford.dialin.ntplx.com] has joined #tunes -:- Fare61453 is now known as Fare -:- Kaufmann [newbie@dial857.infolink.com.br] has joined #tunes Hallo! hi lo Anyone want to talk? 09:10am -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes -:- SignOff Kaufmann: #TUNES (Ping timeout for Kaufmann[dial857.infolink.com.br]) -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes -:- hcf [nef@me-portland-us208.javanet.com] has joined #tunes -:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us208.javanet.com]) -:- hcf [nef@me-portland-us208.javanet.com] has joined #tunes * Fare/#Tunes , having mirrored the yonezawa labs site and printed papers, reads on -:- SignOff smkl: #TUNES (No windows for this server) -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- Fare is now known as Fare61453 -:- SignOff IMPO: #TUNES (Leaving) -:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us208.javanet.com]) -:- hcf [nef@me-portland-us208.javanet.com] has joined #tunes -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes -:- Fare61453 is now known as Fare * Tril/#TUNES is back from the dead. Gone 115 hrs 25 min 40 secs evening, fare 01:10pm -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes Hi, Tril! * Tril/#TUNES goes to eat breakfast isn't it a bit late for breakfast? 01:20pm -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) Not since I stayed up to 4AM playing War2 -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes 01:30pm -:- eihrul [eihrul@usr5-ppp192.lvdi.net] has joined #tunes * Tril/#TUNES is away: (afk) [BX-MsgLog Off] -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes ugh ugh? well, gakuk, then kukag -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) I used to use "kukag" with an meaning opposite to that of "gakuk" Is Syntax a bot? no clue * eihrul/#tunes is pondering how to allow more than one address space to map in a given page. if i only have one single structure representing a page that is nearly an impossibility eihrul: what do you mean? as i can only record a static number of mapped addresses/address spaces i have a single structure for every page (for memory allocation) ever heard of chained structures? or of trees? but it also needs to do bookkeeping on where its mapped and who -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes if by chained structures, you mean lists, then yes... trees yes but i'd like to prevent having to dynamically allocate a bookkeeping record for every single mapping of a page into an address space Chris Okasaki's book on purely functional data structures is a great one -- lots of interesting stuff why dynamically allocate? Just dynamically point. 02:20pm any information on it? (a web page perhaps) after all, the number of descriptors is proportional to the number of pages, at all times... the bookkeeping records have to come from somewhere a single page can be mapped into multiple address spaces with multiple addresses and this all has to be recorded eihrul: do a web search; Chris did his PhD at CMU, an improved version of which was printed and is sold. eihrul: so you have to keep a reverse page table; "easy". reverse page table...? you mean, from physical pages to logical mappings? the structures represent a given physical page frame well, it's a usual problem of implementing sets of pages. -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) they record the address space the frame is mapped into and where in the address space it is mapped i don't have sets of pages... i got rid of them and went with allocation of individual pages it just makes it harder to manage shared memory * eihrul/#tunes found a listing of osaka's papers out cmu. no, it's about implementing *sets* btw, why do you need know who uses a page? don't you just need to know whether or not a page is used? when the address space is destroyed i have to be able to clean it up it's just refcount GC, isn't it? * eihrul/#tunes ponders... maybe i don't... it would just make cleanup a bit tricky -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes as i'd have to walk the whole page table and free any pages there but since this only happens at the destruction of an address space maybe it wouldn't be too harsh the next problem becomes: how do i track who can map pages and who can't? my only thoughts thus far is to restrict a task's ability to map stuff into its address space completely if you want reference listing, just have a set of references and only have one actual task that can map into other address spaces and it does all the verification however it wants... (all optional of course) there are *many* well-known implementations of sets, depending on what operations you want to do on them if all you need is constant time insertion/deletion, something like a doubly-linked list might do that's not the problem... the problem is verifying a given task has the right to map a page in some tasks shouldn't be able to map an arbitrary page into their address space and read it 02:30pm that wouldn't be very secure at all eihrul: based on what criterion? why not use an implicit capability model a la W7? that's what i'm deciding... how to represent rights to access/map that specific page I'd say go for it W7-like where at? with objects instead of lambdas, if you wan eihrul: VMs.html#lambda leading url please? :) -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) http://www.tunes.org/Review/VMs.html#lambda -:- Kaufmann [Kaufmann@dial858.infolink.com.br] has joined #tunes -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes Shalom! Salaam! ni2 hao4 Shit... Microsoft is lobbying the gov't to cut the budget for the DoJ's antitrust division haha irony -:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us208.javanet.com]) -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) -:- SignOff Kaufmann: #TUNES (Ping timeout for Kaufmann[dial858.infolink.com.br]) -:- Syntax [dma@10-070.006.popsite.net] has joined #Tunes -:- Kaufmann [Kaufmann@dial837.infolink.com.br] has joined #tunes Anyway, as I was saying... "The nonprofit groups [who support Microsoft] made their request in a letter last month after an all-expenses-paid trip to Microsoft headquarters in Redmond, Wash., where they were entertained and briefed on an array of issues facing the company." 02:40pm -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[10-070.006.popsite.net]) -:- hcf [nef@me-portland-us1024.javanet.com] has joined #tunes -:- SignOff Kaufmann: #TUNES ("Jesus would make a good main character for a Seinfeld spinoff" - Amber) -:- fire [no@209-68-229-215.dialup.cust.tfb.com] has joined #tunes hi lo my parents say my fellow athletes on the cross country team cannot drive me out to dinner i am thoughroly pissed (shit spelling) atheles? er athletes? that's worse than being a child rapist ... than i am the next hitler fire: what's your parents' pretext? hey, they let you talk to wierd people on the internet, but won't let you have normal social relations? my mom is very paranoid about me talking to people on the internet she thinks you guys are all weirdos lemme phone her, to tell her about the dangers of the internet... * eihrul/#tunes is a weirdo. 03:50pm * Fare/#Tunes is two weirdos hey im making a pointer to a pointer to a function * eihrul/#tunes nods. does that make you horny? * eihrul/#tunes is an infinite number of weirdos stretched out along a hypercube. pointer to pointers? well, pointers mostly suck, anyway yes well references are better i like the idea of every variable being referenced by a pointer like scheme but this code is just plain kinky Tril: for high level programming any way, with very very few tasks needing to be low level at all * Fare/#Tunes must write a review of ABCL/R for yesterday... eihrul : pointers are a special case of references Tril: i understand that, it's just that degenerate cases are usually more efficient :) thanks to tril and eihrul my lengua interpreter is coming along nicely what is degenerate what did I do to help lengua give an account? you mean as in what do i mean by 'degenerate'? that little getchar problem i was having :) no, what's the word mean fire: written in which language? C! ouch i don't know the webster's definition :) eirhul: apt-get install dict oh yeah baby but i use it in the sense of a system in which certain variables are held constant and assumed fare: i just can't resist fare: but im gonna bootstrap onto itself having sunk to a lower and usually corrupt and vicious stat e being mathematically simpler (as by having a factor or constant equal to zero) than the typical case i was close that proves it, corruption and viciousness is mathematically simpler than its opposite Tril: no it isn't. Tril: dumbness is simpler which is one you should never ascribe to malice what can be explained by greed and stupidity s/one/why/ 04:00pm damnit there are some weird bugs in my code i enter in "44444444444444444444" and it outputs it as "444444445444445444" or you have bad RAM bugs in code more probable now that RAM is so expensive, maybe software programmers will write less bloated code it's also *slow* what's that have to do with RAM price hrm we'll be back to the days of offline sorting and multiple-tape algorithms! this is very odd * eihrul/#tunes never experienced those d ays the first time didn't you read the knuth book? or old literature about sorting algorithms or batch processing? s/k/ks/ yes the bubble sory makes me horny sort rather but i've never seen batch processing machines in the like er and the like eih: even IBM PCs or apple ][s, with the floppy being bigger than the RAM, were often batch processors * eihrul/#tunes never had such a computer. i got my first computer very late and apple 2s have been long sinced booted from most american schools 04:10pm does anyone master kali? or abcl? fare: has anyone you mean past tense you're using present tense oh and mastered cause they must agree "has anyone mastered kali?" mastered kali has anyone? yoda speak! FORTH" anyone kali mastered has ?" scheme" (has (mastered anyone kali))" s/has/has?/ whose going to do it in intercal? (LISP (HAS-P (PAST-TENSE (MASTER (ANYONE) KALI)))) * fire/#tunes ph34rs * Fare/#Tunes ph1r3s * eihrul/#tunes phr33s eihrul: your solution stopped memory leakage but you sorta formed it to those printf's that are just there for testing i actually need to push those words onto a stack so i need to keep the '\0' so fix it... ok you just told me to get it worked for the code which you gave me, and i did :) fire: either stop using C or cope with it, but don't make us suffer from your poor or masochistic choices! 04:20pm how about i put the nulls where the whitespace is? put push those? eh, why? you have to guarentee you only have one whitespace between words then so i make mini strings out of the big one and that it gets replaced by the null delimiter no i just put the whitespace checker after the rest so whatever whitespace is left goes bye bye you don't want to have multiple whitespace characters between a word stop doing micro-optimizations! s/a words/words/g or else, program in assembly eihrul: it's the programmer's own fault if he puts in multiple whitespace C is only suited for micro-optimization it has little or no high level programming structures :) then don't use C use Haskell its a portable assembler no, it isn't. C is well, a very restricted version of it not portable. Just somehow ported. ladys ladys and the restricted version that's portable is really lame. i never said C wasn't lame that's why i'm not using it fare: wouldn't you like to tell people that they should use the TUNES HLL instead ? ;) :/ go code my son ip chains can wait ipchains are done it kinda works, although not in the exact expected way then code thine HLL no, I must write my ABCL review for yesterday and then I must rewrite my article on free software for yesterday ok write it then code the HLL uh fare: you need to understand priority fire: my internal scheduler is broken if you haven't noticed tunes really isn't going anywhere :( just go write a barely functional HLL interpreter and add onto it 04:30pm Fare: external scheduler eihrul: who? me and hcf :) someone who can physically harm you c'mon fare just go hack something up -:- SignOff hcf: #TUNES (Leaving) that'd be my sister, but I won't let her hypnotist then? i can hurt your feelings >=) I wouldn't hurt the hypnotist fire: no you can't :( hum. time to GOTO 3053 * Fare/#Tunes GOTO 3053 -:- Fare is now known as Fare3053 uh -:- IMPO [ihatebob@p09-35.hartford.dialin.ntplx.com] has joined #tunes CHEROKEE PEOPLE do do do do do do CHEROKEE TRIBE i want your simple interpreter in my email in the morning ;) hi impo hi fire: call his sister hehe i don't speak french :/ find someone who does hmmm i may know some people 04:40pm tag spoke french fluently so does her sister nina who incidentally was the lead singer for veruca salt :D does she speak german too? i believe so she's cool :) eihrul: i tried my idea but for some reason the next word always looses a character you have to put the '\0's exactly where the whitespace is not AFTER well the way i figure... after every word comes whitespace or null '\0' = '\0' doesn't change anything brb hrm in theory this should work abp: i got that queer diamond program to work :D ^5!!! -:- Kaufmann [Kaufmann@dial712.infolink.com.br] has joined #tunes So what's up? i ended up using like six counters my code doesn't work :D how about you? my code doesn't work either :) 04:50pm But that's normal for me :) as with me I'm listening to Black Sabbath... _Reunion_ ok thisis too messed it works for numbers but not letters What? Perhaps I can help you out well i see where the problem is i think brb -:- Syntax [dma@04-161.006.popsite.net] has joined #Tunes 05:00pm -:- SignOff Kaufmann: #TUNES (Ping timeout for Kaufmann[dial712.infolink.com.br]) -:- Kaufmann [Kaufmann@dial594.infolink.com.br] has joined #tunes wb 05:20pm -:- air [brand@p0wer.qzx.com] has joined #tunes -:- SignOff Kaufmann: #TUNES ("Jesus would make a good main character for a Seinfeld spinoff" - Amber) hi air * fire/#tunes breaths air hey anyone know about slash notation for netmasks what about it? what range of addresses is meant by x.x.80.0/23 ? 23 or 24? 31 whoa! well, if it was 24 I know what it would be x.x.80.* is easy abi: way to decide!!! abi: ab or cd? i don't know, tril abi: (3 + 5) & 2 zero well 23 is 10111 23 has nothing to do with that.. it means the number of bits in the netmask oh I got it 05:50pm oh crap, now that i dont change cr3 i have to invalidate page table entries or the stupid cpu uses old values and i get bugs yah.... ive been trying to find a couple bugs for days nw\ow and just remembered that so flush the TLb it'll probably be faster than invalidating tons of entries one by one as its 22 cycles to flush whole thing, 22 cycles to flush a single page but actually, didn't you say that each object had its own page table/ so why you having problems? well i map all stacks in the same spot bbiaf bad idea all my kernel stacks are in a diff location so i never have to switch address spaces to access any of them but that's a necessity of my design 06:10pm YES! my code works :D i love you all keep your head up... cause that new day... is gonna come... mamma mamma! you know you raised without a father either i wanna take this time to thank you... for given me life :D 06:40pm impo: http://www.wyclef.com/video/NewDayFullVideo.ram 06:50pm eihrul: so are my kernel stacks eihrul: my user stacks are all mapped into the same page woe is you, eh? :) 07:00pm bye -:- SignOff fire: #TUNES (Leaving) -:- lar1 [lar1@sdn-ar-014casfrMP093.dialsprint.net] has joined #tunes Hey 07:10pm Anyone know what kind of stuff they do on 'supercomputers' nowadays? download porno at blazing speeds compete with other supercomputers to find who can calculate pi to the furthest digit Heh :) 3.1415926535897 is as far as I go If I set up my sun 3-50 as an xterm... what could I do on it? vim stuff? 07:30pm Hmm, can it do distribued.net stuff? apperently not There is no client 07:40pm haha this guy got his nt account terminated because we found back orifice in his directory and he was like I DON'T EVEN KNOW HOW TO SAVE SOMETHING Hehe oh i guess my life revolves around things like that woo the party never stops In NASM, if I include, will it include the file before assembly? have u guys heard of shasta mcnasty? air: Yes, never seen it though fat chick: "does it bother u that im a few pounds overweight?" guy: "no i just think of u as 3 chicks rolled into one" Heheh it shows him imagining these these chicks in bathing suits and they all suck together into this blob and out pops this fat chick :) I gotta see this show! that was just the commercial, i havent ever seen it 08:00pm How does the %include directive work in nasm? it includes other files just like it does in C It includes it right where the include directive is? 08:10pm uhh ya Ok, cool 08:20pm air: How does one get the location of the hardware cursor? 08:30pm hold on k i dont think u can u can set it but i think u need to track it in software Hrmm, that sucks in brix i just keep the x,y in software and tell the hardware to move the cursor when it needs to move and i update x,y so i never need to read the hardware to find it Yah Where do you store that x,y an arbitrary location? 08:40pm Ohh! I know! I'll put it in that free spot in my GDT! :) uhh, no i keep the x,y in my screen object Well, I don't have a screen object, and I have a free word in my GDT... 08:50pm are u gonna stuff everything else in there too? :) why dont u just create some global kernel variables? Once I get the second stage loader in memory, I'll move it to the Kernel Data Area 09:00pm And how do you make a global in NASM? 09:10pm dunno -:- eihrul [eihrul@usr5-ppp192.lvdi.net] has left #tunes [] -:- SignOff IMPO: #TUNES (Ping timeout for IMPO[p09-35.hartford.dialin.ntplx.com]) im just talking about data that has a fixed physical address