IRC log started Thu Sep 16 00:00:00 1999 [msg(TUNES)] permlog 1999.0916 -:- eihrul [lee@usr5-ppp61.lvdi.net] has joined #tunes -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp61.lvdi.net]) -:- NaDdY-BoY [nope@207.217.203.208] has joined #Tunes umm? no? -:- NaDdY-BoY [nope@207.217.203.208] has left #Tunes [] 07:30am -:- Closing Link: TUNES[bespin.dhs.org] by adams.openprojects.net (Ping timeout for TUNES[bespin.dhs.org]) -:- Connection closed from irc.us.openprojects.net: Success -:- Connecting to port 6667 of server irc.us.openprojects.net [refnum 0] -:- Unable to connect to port 6667 of server irc.us.openprojects.net: No route to host -:- Connecting to port 6667 of server irc.us.openprojects.net [refnum 0] -:- Unable to connect to port 6667 of server irc.us.openprojects.net: Interrupted system call -:- Connecting to port 6667 of server irc.us.openprojects.net [refnum 0] -:- BitchX+Deb1an: Servers exhausted. Restarting. [1] -:- BitchX+Deb1an: For more information about BitchX type /about -:- Welcome to the Internet Relay Network TUNES -:- Your host is adams.openprojects.net, running version u2.10.05.18.(ipcheck4-5) -:- This server was cobbled together Wed Apr 28 1999 at 13 24:47 EDT -:- adams.openprojects.net u2.10.05.18.(ipcheck4-5) dioswkfcg biklmnopstv -:- [local users on irc(21)] 5% -:- [global users on irc(194)] 43% -:- [invisible users on irc(259)] 57% -:- [ircops on irc(18)] 4% -:- [total users on irc(453)] -:- [unknown connections(0)] -:- [total servers on irc(33)] (avg. 13 users per server) -:- [total channels created(140)] (avg. 3 users per channel) !adams.openprojects.net Highest connection count: 45 (44 clients) !adams.openprojects.net Welcome to Open Projects! You are on 3 ca 1(2) ft 14(14) tr. -:- Mode change [+f] for user TUNES -:- Mode change [+iws] for user TUNES -:- JOIN activated by "TUNES #tunes tunes@bespin.dhs.org " -:- TUNES [tunes@bespin.dhs.org] has joined #tunes -:- Topic for #TUNES: Free Reflective Computing System || http://www.qzx.com/bfe/ || http://www.hex.net/~cbbrowne/lisposes.html -:- topic set by hcf [Wed Sep 15 18:38:40 1999] -:- [Users(#Tunes:4)] [ TUNES ] [ abi ] [ Zhivago ] [@Tril ] -:- Channel #Tunes was created at Sun Feb 28 08:48:06 1999 -:- BitchX+Deb1an: Join to #tunes was synced in 6.668 secs!! -:- Mode change [-s] for user TUNES -:- hcf [nef@me-portland-us146.javanet.com] has joined #tunes -:- |trax| [jknox@216.26.22.210] has joined #tunes <|trax|> anyone alive -:- |trax| [jknox@216.26.22.210] has left #tunes [] 10:50am !devlin.openprojects.net!! Remote CONNECT brin.openprojects.net 8005 from lilo -:- NetSplit: clarke.openprojects.net split from sterling.openprojects.net [11:41am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [clarke.openprojects.net] -:- Netjoined: clarke.openprojects.net sterling.openprojects.net -:- Tril [dem@bespin.dhs.org] has joined #Tunes -:- ServerMode/#Tunes [+o Tril] by clarke.openprojects.net -:- bineng [bineng@b189.ryd.student.liu.se] has joined #tunes hey 01:10pm -:- SignOff hcf: #TUNES (Leaving) !NickServ:*! ^lilo used GETPASS on pythos -:- hcf [nef@me-portland-us329.javanet.com] has joined #tunes -:- SignOff abi: #TUNES (Ping timeout for abi[206.63.100.13]) -:- abi [nef@bespin.dhs.org] has joined #tunes -:- eihrul [lee@usr5-ppp11.lvdi.net] has joined #tunes -:- pyro [tcn@cci-209150250104.clarityconnect.net] has joined #tunes * pyro/#tunes breaks out the champagne sup? I heard the US dropped crypto export controls wasn't that a long time ago? or i remember seeing something about that on slashdot a few weeks back (may have just been "consideration" thereof) I don't watch slashdot much, but I guess it became official today or this week it came as a surprise to me We still have to worry about 1) key escrow, 2) the weakness of RSA 02:50pm -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp11.lvdi.net]) -:- eihrul [lee@usr5-ppp164.lvdi.net] has joined #tunes wb hey! the radio station I was listening to just disappeared! -:- SignOff hcf: #TUNES (Leaving) it's wet but it ain't that windy.. 03:10pm -:- SignOff pyro: #TUNES (Ping timeout for pyro[cci-209150250104.clarityconnect.net]) -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp164.lvdi.net]) -:- eihrul [lee@usr5-ppp164.lvdi.net] has joined #tunes -:- SignOff eihrul: #TUNES (Leaving) -:- eihrul [lee@usr5-ppp164.lvdi.net] has joined #tunes -:- lar1 [lar1@1Cust26.tnt20.sfo3.da.uu.net] has joined #tunes Yo y0 04:40pm Heh, article on Linchiks on /. eh? you can get on slashdot? Yes Slowly But yes 04:50pm hrmm... wtf is going on with sprint lately? i am getting > 1 second ping times Holy crap Earthlink accuired sprint, right? Earthlink isn't good anymore They suffer from the AOL effect well, routes keep going down at about sprint says traceroute and it's damned annoying Call yer ISP and be a badass but umm... they might kick me off or do something nasty Why whould they? _They_ are the ones providing _you_ with crappy service! 05:00pm yeah, and they might purposely make it crappier and then again, it could be sprint's problem... not necessarily theirs at which case even they can't do much about it Ohh, I thought you had sprint :) In that case, nevermind... your probly right, not much yer ISP can do i have LVDI, which goes through sprint just an itsy bitsy weenie local isp Hmm, yah small ISP I wounldn't get all up in their face we're getting a sprint connection on oct 1 :) i pity you... Tril: Uhh, who do you have now? s/uhh/ugg wa.com You can't just dial into bespin? bespin is on nas.com's ethernet. nas gets internet from wa.com, and is adding a sprint line on oct 1. To get to bespin you need to have internet from someone, nas or otherwise i dont provide any service on bespin that competes with nas such as dial in modems Ohh, I see 05:10pm -:- pyro [tcn@cci-209150250093.clarityconnect.net] has joined #tunes -:- liar [brand@p0wer.qzx.com] has joined #tunes hey liar yeah hey, u try the latest bfe? u need to uncomment the #ifdef in xbuttons.c:update_xforms() and all the registers will work s/uncomment/remove/ what's in the newest bfe? Yo liar -:- SignOff pyro: #TUNES (pyro has no reason) 05:50pm eihrul: registers unless you need me to add or debug anything, there's no reason to get new version yet :) i just need to rearrange the input boxes so the segment registers will have enuf room i need u to write the structure function :) Me writing the struct thing is all yer waiting for? lar1: ya man yer taking too long :)( liar: wha structure function? I'll do it right now to parse the .bferc config file Sorry, I was unable to do it last night... lemme see the config file :) its the easiest config file to parse anything outside {} it ignored Bah! Fine, eihrul you do it if you all don't think I can code it up fast i just wanna see it... :) Right { STACK 10 20 } would tell bfe to display 10 stack entries above esp and 20 below eihrul: its in the bfe.txt file eihrul: that comes with bfe.tgz :) eihrul: u've seen it before liar: So you still want me to do it? and i already have a getword() function so it makes it really easy 06:00pm lar1: ya tokenize :) strings sux0r getword is in the new bfe? dcc it mabye? lar1: its in the copy u have lar1: its in the structures.c file Ok You want me function in structures. c? c is the biggest hack of all more like it ya And you want me to check where for the file agian? The home dir then the dir bfe resides in? u just need to use getword until u find a '{' and then read the next word: STACK, LIST, STRUCT and then read the following words based on whatever the first word was. check in current dir first and then home Ok ack... what? having to read strings over and over there's betters ways to find strings in files :) u wanna do it? :) if you could give me better description of what exactly the format of the config file means Better yet, u wanna tell lar1 how to do it better? as what you have their is vague how is it vague? before i can do either i need to know what needs to get done... liar: one sec... i'll find what needs clarification first, is a LIST an array or a real list? if u find a { LIST blah... } or a { STRUCT blah... } then u need to use malloc to allocate space. if u find a { STACK x y } then u just need to set the stack variables stack_above and stack_below * lar1/#tunes thinks that what eihrul is wondering about is exactly what was fscking me up last night ok maybe array would be a better name stick to the questions at hand :) i have more i also want a linked list okay, so nodesize defines the size of each element of the array? yes AHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH! Linked list? You said the unholy word! and nodes is the number of elements now where do the val?text and size? come in liar: I assume eihrul is doing this now? lar1: dunno each element in the array is a valtext and valsize 06:10pm as u can see from those structs i just msg'd u so nodesize refers to the node element count not the size in bytes? therefor, being a misnomer right you really need to name things by what they do... :) it helps, really! ya ya here's a suggestion though... you can do evil things with structures... i.e. put dynamically sized arrays starting at the exact end of the structure thereby allowing memlist_struct and the number of value_structs to be allocated as one contiguous chunk of memory :) gcc bitches when i had [] there's a trick to it :) what? struct memlist_struct { ... struct value_struct values [0]; } 0? ptr = (struct memlist_struct *) malloc (sizeof (struct memlist_struct) + sizeof (struct value_struct) * value_count); also note... sizeof (long) == sizeof (int) :) Uhhh Liar: Ok to just do what you said? ya well if someone would pull their head outta their ass and define a consitent set of data types for C then i wouldnt have that problem :) they are consistent lar1: ? on most 32 bit machines, sizeof (char) == 1 actually, that's ANSI mandated... ha then you have sizeof (char) < sizeof (short) < sizeof (int) == sizeof (long) liar: Can I just code up the function the way you told me to? Or should I figure the way eihrul is saying? :) well whose doing it? me or lar1? :) or are we gonna race? lar1: what eihrul said :) ya race eihrul winds s/winds/wins can i just call it mem_struct? :) ok we will have one rule. eihrul must have both hands tied behind his back :) Heh liar: You sure there isn't _any_ more code to do? i've never typed with my other appendages before... though that would make for 4 in total... eihrul: ya, but they are all defined in xglobals.h :) i'm gonna change memlist_struct to mem_struct it's not like any code depends on it 06:20pm ok changed i'm changing it :P can i make it use ARRAY instead of LIST? also... what good are the value names in an array? yes oops its not an array then what is it? { LIST io_list 0x0f000 512 2 dn_io_base 4 dn_io_ports 4 } the iolist has 512 entries and each entry has 2 fields: base and ports and each field is 4 bytes long so it has 512 8 byte entries so it's an array of structures... ya so why not just express that directly :) eh? actually... there's no difference between LIST and STRUC, except one is an array of them and the other is just one of them... hence LIST with element_count == 1, is just STRUC brb well i wanted struc to be the same as list but it only has one entry no need really right or atleast in the code representation -:- SignOff lar1: #TUNES (Read error to lar1[1Cust26.tnt20.sfo3.da.uu.net]: Connection reset by peer) right so i don't need that stupid type field i just made struc so the user doesnt have to type that 1 06:30pm and feel free to rename anything :) i have already done so... actually, might be easier to not use dynamically sized structures in this situation :) why? just because... :) well the number of values can vary STRUC can have hundreds -:- lar1 [lar1@1Cust197.tnt20.sfo3.da.uu.net] has joined #tunes u cant use fixed size the head structure itself can be dynamically sized it actually amounts in the same amount of reallocation or so either way but just facilitates parsing in one direction specifically :) uh ok 06:40pm Hey, liar, how does BRiX pass messages? fork/call liar: How does that work? u create new threads Ok instead of using message queues it just creates a thread to process the message Uhh, then thread A can't talk to thread B? ya they can as long as they are both methods of the same object they can use shared memory Is shared memory the best way to pass messages? well even if they are methods of different objects they can still talk thread A just creates an object and forks thread B with a reference to that object then they can write messages to it butthey must both be in the same address space cuz shared memory in brix doesnt span address spaces Hmm im also against everything i just said too :) So am I What is a better way to pass messages? i dont want threads talking to each other i prefer the way i do it duh! You just said that you were aginst what you said! Whats wrong with threads yapping? no, im against using an object for threads to talk But yet you like it? that feature was added for a different purpose, and not so threads could talk 06:50pm thread A should say everything it needs to say when it creates B just like when u call printf() in C, u dont sit there and continue to feed it info So how do you pass all the messages in the first place? The way you just said? when the thread is created screen.printf(message goes here) screen is the object printf is the method and the message is in the ()'s just like C Can the message just be put on the stack and have the method pop it from the stack? ya What if app A needs to be sent a message by somthing? there are no apps Ok, thread A wich happens to be the main of mybrowser How does the message, "Hey! Quit NOW!" get to thread A? it doesnt Uhhh brix uses a persistent object store Whats a persistent object store? ugh ask me in a week after i finish bfe Hmmm, ok or u could always do what other ppl do and read the web page :) I am reading yer page But as you know... most of that goes over my head until someone explains it to me could u ask question when yer finished? Sure eihrul: u here? i got a problem You still have a weeks worth of bfe? Mabye there is some I can do? lar1: ok i got a perfect job for u liar: Yes? 07:00pm there is a stack list on the bfe window and i need to display the TOS value and the values above TOS and below Ok sounds simple right? wrong Why not? i just need to dump X amount of memory at starting at the top most address but paging screws us over again :) if that top value is outside of the stack space then... Wait, what do you mean TOS? bochs will tell u, and u need to add 4 to the address and try again and again and again until u finally get it but there is another problem too the top address might be in range but the bottom value is below stack space :( TOS = top of stack esp ok i got it How could the tos be out of the stack space? read a single value at a time :) Huh? say u want 20 entries above and 5 entries below displayed u need a total of (20+1+5)*4 bytes instead of reading in a block of 104 bytes and trying to figure out if its out of paging range below or above, just read it in 4 byte chunks at a time ok So what function is this gonna be? hold on kk string parsing is so annoying heh but yer so good at it :) 07:10pm okay, so far i can parse the structure name :) liar: Umm, when I go to the info stuff on brix it sends me the php source lar1: dont tell me that no it dont im in info right now Well its happning when u click on "Information"? When I click on any topic under information jeez u sure give a lot of details dont ya :) the first half of the links work the last half dont. and the page even says so Why do some work and some not? everything below untyped Everything I want to read! :) php3 got removed from apache so i was forced to make each file +x and i had A LOT of files and i got tired and stopped I'll do it for ya ha i'm gonna do something evil :) chmod +x blah ;) string_read () is very reuseable in this circumstance lar1: its not that simple liar: Why not? also gotta add the line to the top of each file so bash knows to run php3 Its only about 10 files, right? and u gotta change each file from *.php3 to *.cgi Ok or actually... it won't work, damn Give me telnet and I'll do it no can do i would have to let u login to my account with my passwd and u could change that passwd and i wouldnt be able to log back onto the net :) You can't just chmod a+w * to the files you need changed and give me an account? 07:20pm u dont seem to understand. i dont have root Ohhh I thought you did! almost done ugh, yer suppose to take yer time * eihrul/#tunes cackles. 07:30pm heh hmm, I don't suppose you people know of any decent generic ide driver source in a legible form? retro has one i dont know if its readable grrr... it's messing up somewhere hmm, retro has ide? I'll have to check again Tril: You there? I'd like to talk about the review project nevermind, I have to eat -:- lar1 is now known as lar_eatin okay... i think i'm kinda done :) it's probably buggy... heh so it does, I was busy looking at old versions :) why in the hell do u use const? once again retro to the rescue, just need to rewrite the asm into lisp hahaha it allows for compiler optimizations why anyone would write an os in lisp is beyond me in some circumstances the compiler has to re-read memory because it can't determine that the memory is NOT going to change eihrul: what did u change? liar: because its both a HLL and a LLL? word_read () and a little something :) er wait... there's one thing i forgot, doh! 07:40pm liar: and it means that I dont' need all this assembly that retro has :) liar: I can build smoothly down to the metal but actually... the code i sent you should work the same it will just be largely more syntactically tolerant i.e. it does not require you to put in closing '}'s heh but doesn't matter... because it just looks for opening '{'s ya so i just need to add the [0] to memlist and values right? er wait... i forgot the xglobals.h you still need to fix the array of mem_struct's itself :) what you want to basically do is: struct mem_struct tmp; while (structure_read (file, & tmp) != 0) { array [whatever] = tmp; } 07:50pm uhh eh? er... brain fart... while (structure_read (file, & tmp) == 0) structure_read should read the entire config file so make a config_read () function with that loop or something you can do it dude... that was yer job :) gah... i have to do everything huh? ya and did u check for { STACK x y } ? nope it can be read more than once, just keep setting stack_above and stack_below so the last time its used will be the values bfe uses nr_strucs should be only one of those *_struc variables i need to set right? aside from the obvious memlist yes/no? ya global variables == icky now where are stack_above and stack_below at? no they speed up the code cuz u dont have to pass them around extern unsigned stack_above, stack_below; i just added them * eihrul/#tunes sighs... what is so speed critical about BFE? nothing... we're not controlling nuclear power plants withi t it has to update an entire window everytime u press a button * eihrul/#tunes cackles. not to mention it has to update all the memory/structure windows too haha... you think argument passing is going to matter any? thats alot of shit going thru that pipe from bochs every little bit helps :) the input parsing is going to be you rbiggest bottleneck... no contest and the context switching... and the load bochs itself puts on the CPU 08:00pm and just why does bochs put a load on the cpu when its in a continuous loop like jmp short $??? i could never figure that out why shouldn't it? that's a continuous stream of instructions, albeit a special case it should go idle and wait for a timer to tick or something oh btw, u can rm any of those vars in globals:structures that u dont use