IRC log started Tue Sep 14 00:00:02 1999
[msg(TUNES)] permlog 1999.0914
-:- SignOff liar: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep)
-:- SignOff Crimson: #TUNES (http://www.mistik.net)
-:- Zhivago [brian@th.merddin.com.au] has joined #tunes
<Zhivago> small crowd today :)
-:- hcf [nef@me-portland-us641.javanet.com] has joined #tunes
07:30am
<Zhivago> hello
-:- SignOff hcf: #TUNES (Leaving)
07:40am
-:- NetSplit: devlin.openprojects.net split from sterling.openprojects.net [08:06am]
-:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [devlin.openprojects.net]
-:- Netjoined: devlin.openprojects.net sterling.openprojects.net
-:- abi [nef@bespin.dhs.org] has joined #Tunes
-:- MAxD [m.dentico@a86.teseo.it] has joined #tunes
-:- MAxD [m.dentico@a86.teseo.it] has left #tunes []
-:- Connection closed from irc.us.openprojects.net: Success
-:- Use /Server to connect to a server
-:- 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]
-:- BitchX+Deb1an: For more information about BitchX type /about
-:- Welcome to the Internet Relay Network TUNES (from adams.openprojects.net)
-:- Your host is adams.openprojects.net, running version u2.10.05.18.(ipcheck4-5) (from adams.openprojects.net)
-:- This server was cobbled together Wed Apr 28 1999 at 13 24:47 EDT(from adams.openprojects.net)
-:- adams.openprojects.net u2.10.05.18.(ipcheck4-5) dioswkfcg biklmnopstv 
-:- [local users on irc(26)] 5%
-:- [global users on irc(207)] 42%
-:- [invisible users on irc(282)] 58%
-:- [ircops on irc(12)] 2%
-:- [total users on irc(489)]
-:- [unknown connections(0)]
-:- [total servers on irc(31)] (avg. 15 users per server)
-:- [total channels created(139)] (avg. 3 users per channel)
!adams.openprojects.net Highest connection count: 38 (37 clients)
!adams.openprojects.net Welcome to Open Projects!  You are on 3 ca 1(2) ft 14(14).
-:- Mode change [+f] for user TUNES
-:- lackey.openprojects.net  That server doesn't exist
-:- Mode change [+iws] for user TUNES
-:- JOIN activated by "TUNES #tunes tunes@bespin.dhs.org "
-:- TUNES [tunes@bespin.dhs.org] has joined #tunes
-:- [Users(#Tunes:6)] 
[ TUNES     ] [ liar      ] [ eihrul    ] [ abi       ] [ Zhivago   ] 
[@Tril      ] 
-:- Channel #Tunes was created at Sun Feb 28 08:48:06 1999
-:- BitchX+Deb1an: Join to #tunes was synced in 6.262 secs!!
-:- Mode change [-s] for user TUNES
<Tril> because it wont respnod to ctrl-alt-backspace?
!irq:*! well it doesn't look like there were any odd characters in that quit string.. so the problem must be elsewhere
<liar> i just got home and it was sitting on the gdm login prompt
<liar> is there a log file that xf86 would write to
<liar> ?
<eihrul> liar: hmm... i just had a nifty idea in school today
03:00pm
<Tril> i dont think X has logs
<eihrul> was pondering how i wanted to lay out my reserved address space and how i was going to map page tables in
<eihrul> liar: and then it dawned on me... all i had to do was  directory [0] = directory
<Tril> reflection
<eihrul> yep... it's cool because:
<Tril> err, i mean, self reference
<eihrul> the PDE of any address is   (address >> 22) * 4, the PTE of any address is (address >> 12) * 4
<Tril> it's not necessarily clean reflection
<eihrul> but it's efficient :)
03:10pm
!lilo:*! isho 3
!netgod:*! answr 42
!lilo:*! oops, my bad
<liar> eihrul: what are u talking about?
<eihrul> page directories!
-:- ChanServ has changed the topic on channel #Tunes to: www.tunes.org
<liar> what about them?
<eihrul> about mapping all the page tables into the address space
<liar> haha chanserv screwed over the topic :)
<eihrul> all you have to do is make that one link and it all falls into place
<liar> i dont know what yer talking about
<eihrul> remember when we were talking about disabling paging
<liar> ya
<eihrul> and how it flushed the whole tlb on pents?
<liar> ppro's
<eihrul> yeah... :)
!irq:*! lilo any idea what happened?
<eihrul> but i found an easy way to map all the page tables into the address space transparently
<eihrul> all ya gotta go is map a PDE refer to the page directory itself
<eihrul> er make
<liar> i dont get it
<eihrul> how are you mapping page tables into your address space now?
<liar> i add them to the pgdir
!lilo:*! looking at it
<eihrul> you have a page dir reserved for mapping them?
<eihrul> er page table?
<liar> what
!irq:*! go lilo, go lilo (:
<eihrul> what what?
<eihrul> how exactly are you doing it with the pgdir? :)
<liar> there is a single pgdir per address space and it points to 1024 pgtbls
<eihrul> okay...
<liar> and each pgtbl points to 1024 pages
03:30pm
<eihrul> here's the trick: page tables and page directories have exactly the same format
<liar> duh
<eihrul> and all page tables in your page directory need to be mapped into address space
<eihrul> so instead of using a whole new page to do that... you use the page directory itself
<liar> oh
<eihrul> that way you don't have to maintain two diff tables
<eihrul> just the one page directory
<liar> hmm
<liar> so all the pgtbls for an address space are say in the last 4meg of the space
<eihrul> i think i'm going to put mine starting at 0...
!devlin.openprojects.net!! Remote CONNECT dickson.openprojects.net 8005 from lilo
!irc.linux.com!! Received :tolkien.openprojects.net SERVER dickson.openprojects.net from sterling.openprojects.net !?!
<liar> my kernel area is in the first 4meg
<eihrul> well... regardless, all you have to do to find a given PTE is    start + (address >> 12) * 4
<eihrul> and start + (address >> 22) * 4 for PDEs
<liar> hey have u heard of "btb web" or "beyond the basics"
<eihrul> nope
<Tril> (address>>X) * 4 is wasteful isnt it
<eihrul> shl r1, 22; lea r1, [r2 + r1 * 4]
<eihrul> hardly wasteful :)
<Tril> >> is not shl
<eihrul> er shr
<eihrul> brain fart...
<eihrul> and once you find the initial PDE or PTE or whatever, finding the next is just an add r, 4 away
<liar> http://yesfree.com/cam1.htm
<liar> go there
<Tril> what i am saying (and i could be wrong) is that >>22 and then *4 seems wasteful, why dont you just >>20
<eihrul> because...
<eihrul> PDE/PTE's are 4 bytes in length
<eihrul> so you need to mask off the last 2 bits, address needs to be dword aligned
<Tril> you wan to zero those bits?
<eihrul> i could use and, but lea is cheaper in this case since i need to do an add too
03:40pm
<Tril> yes, ok. But the problem is that low level programming is WRONG :-)
<liar> eihrul: those bits are already zero
<eihrul> only if the address is aligned
<liar> it has to be zligned
<liar> page talbes MUST be on page boundaries
<eihrul> you're nitpicking right now :)
<eihrul> the mechanism as a whole is the important part
<eihrul> tril: sure it is... but it's fun
<eihrul> 3,000 piece jig saw puzzles are wrong... but people do them
-:- SignOff abi: #TUNES (Read error to abi[bespin.dhs.org]: EOF from client)
<liar> Tril: good luck writing yer os without using anycode written for low level :)
<Tril> what os
<liar> tunes or anything else u work on
<Tril> the concept of os is prejudiced.
<Tril> it presupposes a separate program that is favored in some way
<Tril> which inevitably leads to favoring certain people (i.e. system programmers) in the same way.
<eihrul> so how does lack of an os favor not system programmers?
<Tril> because all apps are equal
<Tril> the distinction of special app is made by the user ONLY
<eihrul> systems programmers would be the only ones capable of really making those decisions
<eihrul> and in the end you have OS functionality provided SOMEWHERE
<eihrul> so you will have some privileged elements to it... no matter what
<Tril> what functionality?
03:50pm
<eihrul> services to interface with the user and the hardware
<eihrul> "what user?"
04:00pm
<liar> where is abi?
<liar> abi has my bookmark for xforms and i need it
<eihrul> me too
<Tril> holdon
<Tril> xforms => a widget toolkit at http://bragg.phys.uwm.edu/xforms
<Tril> is that all you need?
<liar> thanks
<liar> ya
<eihrul> liar: i need your new bfe code too :)
<liar> give me a minute and the breakpoint code will be working
<eihrul> actually... i wonder if FLTK will work with BFE
<liar> fltk?
<eihrul> fltk.easysoftware.com
04:10pm
!irq:*! <node> [node] when I set channel options | it updated when i /msg chanserv info #chan, but the it didnt _do_ anything | eg. didnt remember topic, didnt set modes
<liar> oh that
<eihrul> tis a nice toolkit with FL emulation
<liar> what is fl?
<eihrul> i thinks xforms :)
-:- SignOff liar: #TUNES (BitchX-74a14 -- just do it.)
<eihrul> though i may be confusing it with somethin else
-:- liar [brand@p0wer.qzx.com] has joined #tunes
<liar> OH YA NOW IM PISSED
<eihrul> so, can i get your new bfe code?
<liar> i was just booted out of x again
<liar> maybe rebooting will fix it
<eihrul> why would it?
<liar> it has never happened before but has now happened twice today
<liar> brb
-:- SignOff liar: #TUNES (BRiX [http://www.qzx.com/brix] :: )
04:20pm
-:- liar [brand@p0wer.qzx.com] has joined #tunes
<liar> can i get that url for xforms again?
<Tril> xforms => a widget toolkit at http://bragg.phys.uwm.edu/xforms
<liar> thanks
04:40pm
<liar> hmm i can only enter numbers in the breakpoint entry box
<eihrul> hmmish... 1 more month to finish this dern thing
04:50pm
<liar> what?
<eihrul> gotta finish my kernel
<liar> why?
<eihrul> i want to include it in my portfolio of work when i apply to colleges
<liar> uhh ok :)
<liar> and yer getting out of high school in a month?
<eihrul> no... but i need to apply by next month :)
05:00pm
<eihrul> bfe just locks up
<liar> when?
!irq:*! * irq wonders if lilo found the problem
<eihrul> before doing anything
!irq:*! if not, i can break out my codehacking cap this weekend...
<eihrul> i'll run it through gdb
<liar> works for me
!spice:*! maybe an irq conflict *g*
!irq:*! give me gdb and a 1984 IBM keyboard, i'll debug windows!
!spice:*! I use a 1991 Compaq keyboard
<eihrul> doh...
<eihrul> i forgot to change the popen2 () call
!irq:*! pff, only a real 1984 microswitch keyboard with detachable cable is the best
<liar> huh?
<eihrul> you have the location hard coded, i.e. "bochs"
<eihrul> when i have to keep changing it to "./bochs"
<eihrul> you should make a config file or environment variable or sumfin :)
<liar> i have an idea, INSTALL BOCHS :)
<eihrul> no... i like running it from the distribution dir
<eihrul> i have having to remove tarballs once they're installed
<liar> whatever 
<eihrul> if there was a debian package... :)
<eihrul> there... works
05:10pm
<eihrul> seems to lock up nicely when you hit stop
!spice:*! f00  <----  blatant wallop abuse!
<liar> ya dont hit stop
<liar> i need to find out howto send a ctrl-c to bochs
<eihrul> hrmm... i think i know how... i'll try hacking it in
<liar> k
<eihrul> need to change popen2 () to return the pid though
<liar> crap, im gonna need to have instr_go set a variable so instr_stop wont run the update_xforms() if the program wasnt running
<liar> we have the pid
<liar> global var pid
<liar> :)
<eihrul> oh... doh
<eihrul> and that is why global variables suck!
<liar> no, that is why they rock. u didnt have to modify popen2() :)
<eihrul> but i didn't exactly see pid as a global variable
<liar> everything is global :)
<eihrul> well, it crashed bochs
<liar> what did?
<eihrul> kill () >:)
<eihrul> ah... SIGINT == ctl-c
<liar> ah kewl
<eihrul> oh wait... i think it might have worked...
<eihrul> just your brix image exits too soon
<eihrul> also, you probably want to catch SIGCHLD
<liar> this image has a jmp short $
<liar> why?
<eihrul> so that when bochs exits you can exit gracefully rathre than have it crash
<liar> how do i catch signals?
<eihrul> man 2 signal
05:20pm
<eihrul> signal (SIGCHLD, sigchld_handler);   /* void sigchld_handler (int); */
<liar> ok done
<eihrul> the kill (pid, SIGINT) working?
<liar> oh crap i sent u the full image :)
<liar> void (*)(int)
<liar> wtf is that
<liar> void *sigchld_handler(int) ??
<eihrul> no...
<eihrul> the (*) denotes a pointer to
<eihrul> rather than returning a pointer
<eihrul> void (*pointer_to_function_accepting_int_and_returning_void) (int) = NULL;
<liar> well it works
<liar> didnt update tho
<eihrul> okay... next order of business is that when you hit quit, bochs doesn't go down :)
<liar> eh?
<eihrul> or rather, if you hit Go, then hit Quit
<liar> why?
<eihrul> you should send a SIGINT to stop execution if it's running then quit...
<liar> oh ya
05:30pm
<liar> ok fixed
05:40pm
<eihrul> hrmm... what to do...
<eihrul> A) have an array of pointers to tasks (to map task id's to tasks), or B) have a list of tasks and a bitmap of the allocated tasks
<Tril> that's the kind of decision that should be made at runtime
06:10pm
<liar> eihrul: eh?
<liar> eihrul: how about a list of tasks and when adding a new task just scan the list for an empty task
06:30pm
-:- lar1 [LarMan@1Cust135.tnt31.sfo3.da.uu.net] has joined #tunes
<lar1> Hey
<liar> hey
<liar> u in linux?
<lar1> Nope
<lar1> SHould I be?
<liar> get in it
<lar1> Ok... That meas I must give up my mp3 listning
<lar1> Oh well
<liar> why?
<lar1> I don't have xmms
<liar> xmms?
<lar1> The new version of x11amp
<liar> why do u need that?
<liar> x11amp is just fine
<lar1> I don't have x11amp wither
<liar> it comes with rh6
<lar1> Yah, I didn't install it
<liar> bummer
<liar> do u have rh6 installed?
<lar1> Plus E looks like crap in 16bpp (thanks for helping w/ that one) and I can't fiure out the the vert for my LCD at 800x600x24bpp
<lar1> Yes
<lar1> I got rid of gnome
06:40pm
<liar> u sure it didnt installe x11amp?
<lar1> Yes, I explicitly told it not to
<lar1> I only installed what I knew I needed
<liar> do u have the cd?
<lar1> Yes
<lar1> I could install
<liar> stick in the main cd and goto its RPMS dir
<lar1> Yah
<liar> and type 'rpm -Uvh whatever_the_hell_its_called.rpm
<lar1> Hrmmm, moms home with the food
<lar1> Not rpm -i blah?
<liar> _uvh is better cuz i will install or upgrade
<lar1> Ok, cool
<lar1> Be back - food
<liar> u can win a merceded slk in this international programming contest
<liar> www.ari.de
06:50pm
-:- SignOff Zhivago: #TUNES (Ping timeout for Zhivago[th.merddin.com.au])
-:- Zhivago [brian@th.merddin.com.au] has joined #tunes
<eihrul> liar: cuz that would suck :)
<eihrul> and task structures take up memory
07:10pm
-:- NetSplit: forward.openprojects.net split from irc.linux.com [07:10pm]
-:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [forward.openprojects.net]
-:- Netjoined: forward.openprojects.net irc.linux.com
-:- Zhivago [brian@th.merddin.com.au] has joined #Tunes
<lar1> liar: I have to go, I might be back in an hour or so
<lar1> liar:  If not, I'll be here tommarow... _in_ Linux :)
-:- hcf [nef@me-portland-us936.javanet.com] has joined #tunes
<lar1> Hey hcf!
<hcf> hey
<hcf> yes, i removed the nickometer thing
<lar1> Grumble... grumble
<hcf> just caused noise, which this channel has enuf of
<lar1> heh
<lar1> You read the logs in their entirety?
<hcf> just grep'd for 'hcf'
* lar1/#tunes can't stand more then 1 or 2 pages of chat logs
<lar1> Ohh :)
<lar1> Thats a really good idea, I think I'll implement it
08:30pm
-:- SignOff lar1: #TUNES (Leaving)
-:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp83.lvdi.net])
-:- SignOff hcf: #TUNES (Leaving)
-:- SignOff Zhivago: #TUNES (Ping timeout for Zhivago[th.merddin.com.au])
-:- Zhivago [brian@th.merddin.com.au] has joined #tunes
-:- hcf [nef@me-portland-us204.javanet.com] has joined #tunes
-:- SignOff hcf: #TUNES (Leaving)
-:- liar has changed the topic on channel #Tunes to: www.tunes.org || http://www.qzx.com/bfe/
-:- SignOff liar: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep)
[msg(TUNES)] newlog 1999.0915
IRC log ended Wed Sep 15 00:00:01 1999