| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Comments:
<0> mauke: im going to have to find a victim for that memmove memcpy idea
<1> i did a hearing test and got a 60% score
<2> terrible
<1> but I think it always used to be like that
<2> i have all kinds of ear and hearing problems. and get a hearing test every year.
<3> #define signal(n, f) signal(n, (void (*)(int))main)
<1> so I don't want to make it worse, even if it's just a tiny bit
<1> OrngeTide: what's your score?
<0> mauke: funny, i have a couple programs that do that on purpose
<0> mauke: ioccc entries that ive never finished, but still
<3> what, for all signal calls?
<4> main isn't guaranteed to be in the namespace.
<2> #define exit(n) main(0, 0)
<0> god i love my ioccc entry that ive been working on for years
<0> mauke: yeah, for all of them, it was an ioccc thing
<2> SgtUnix, i'm not sure what your 60% means. they aren't scored like that for me.
<3> abaddon: nice
<2> SgtUnix, but on a good day my right is 75% of normal
<5> SgtUnix, nah.. reading though stuff mostly... some tech docs for amd64
<2> usually worse when my middle ear has collapsed.
<5> SgtUnix, http://www.amd.com/us-en/Processors/ProductInformation/0,,30_118_4699_875%5E7044,00.html
<3> #define signal(n, f) signal(n, ({int main(); (void (*)(int))main; })) // GNU C
<0> mauke: thats nothing, the realy one ive been working on for 3 years now, wait, 4 years now is great, its a webserver, the c code is also valid parsable html, it is its own index.html, i wont be finished until its fully http 1.0 compliant
<5> if you want em, going to do some yasm
<0> mauke: but the size requirements make that hard
<2> mauke, better
<6> abaddon, your a sick, sad little man
<3> haha
<0> ophten: oh its worse than ive told them and you know it, but i dont want to spoil the real sick part until i enter it
<3> how do you embed HTML in C?
<3> or the other way around
<6> with a spatula
<0> mauke: think typedef a char *; a href="http://foo.org";
<1> Mad_Coder: heavy reading material. :)
<3> oh, right
<0> mauke: it follows like so, lots of creativity
<0> also the trigraph for { is ??<
<0> } is ??>
<0> and every browser ive seen ignores the ??
<7> i'm shopping for the receiver, which brand is better, kenwood or denon
<3> maybe I should add html to http://mauke.ath.cx/stuff/poly.sh.pl.tcl.cpp.bf.py.c.lhs.txt
<0> the compiler whines that there is nothing in ??<a href="foo"??> but it compiles
<2> what if your charset doesn't have < and > ? :)
<1> mauke: ah you were the one who made that? I've had that file in my ~ for 3 weeks now
<0> ot: i dont know man, i just know that ??< is the trigraph for when you dont have {
<5> SgtUnix, its kind of overwhelming to read, and patience is something I lack too..., also I'm brushing dust of my C skills(a lot of dust there)
<1> (for no particular reason, i have lots of temporary files in my ~)
<3> SgtUnix: yep
<0> ot: and i only care that it compiles ;)
<2> i get tons of trigraph warnings on my Mac apps because i have lots of character literals like '????'
<6> i forgot how much time #c ops had on their hands ;)
<4> mauke: Checking for C or C++ can be done by sizeof 'a', I think.
<2> abaddon, it's a sad sad world if you can't at least push USASCII
<3> hund: not if sizeof (int) == sizeof (char)
<2> oh! I want that. i want sizeof(int) to be sizeof(char)
<2> some DSPs are like that.
<4> OK, I thought int had to be bigger than char.
<0> mauke: yes but if sizeof(int) == sizeof(char) cry
<0> also, never bother with sizeof(char) it is always one
<0> by definition
<2> hund, char has to be at least 8 bits. but can be more. and int has to be at least 16 bits (to have the necessary range).
<1> ophten: we made clones of our selves that do all the real work, then we can just spend all our time ircing.
<3> yah, the point is that 'a' is an int in C but a char in C++
<4> OrngeTide: Right, so if the char is 16 bits, you're screwed.
<2> hund, why? there is no restriction on the sizes.
<4> OrngeTide: In my brave attempt. :)
<2> i think you could probably even have char be bigger than int.
<0> ot: no you cant
<2> at the very least the standard allows them to be the same size
<0> ot: char is always sizeof == 1
<2> abaddon, really?
<0> ot: it is the base unit
<8> OT, yeah.
<2> abaddon, oh. right right. good point
<0> ot: it is defined to always be 1
<0> so sizeof(int) would hav eto be less than 1
<2> yea. i know it's 1. i just didn't think that sizeof(int) would end up being less than 1.
<2> when i went to the apple store I had to register for the genius bar and when i went to they said the line was too long and told me to make an appointment.
<0> ot: well unless sizeof starts to be a floating point operator, or sizeof(int) could somehpow be 0, i think its going to have to be at least 1
<8> Hmm, is sizeof(int) guaranteed not to be fractional?
<0> kidart: its an operator, so yes
<2> abaddon, sizeof should return a float!
<2> i wish C had half floats.
<1> aedinius: the apple store here is *always* very busy
<2> short float x; :)
<8> abaddon, so what if it's an operator?
<0> ot: it doesnt return anything, anymore than + - ?: or * do
<1> aedinius: the employees are always pretty helpful though, despite getting very few breaks
<0> ot: its an integer operator
<2> SgtUnix, i have to shove people out of the way to get into the store. and it's in a mall so parking is a m***ive pain
<2> abaddon, i'm using return as a substitute for the word evaluates-to
<0> ot: i know, and im just being a **** ;)
<2> i wasn't sure if it was okay to use lisp-lingo
<2> abaddon, yes. that's typical of you
<0> ot: gotta go with what you're good at right ;)
<1> OrngeTide: good thing the Apple Store Online exists.
<7> linux kernel has def of 16-bit floats somewhere
<2> abaddon, true
<2> MishMan, where?!
<9> you might not want to buy a Dell laptop: http://www.theinquirer.net/?article=32550
<0> ot: dont do it, its a trick
<7> somewhere in accouting.h
<2> i don't believe you.
<0> ot: he just wants you to look at the linux kernel
<0> ot: once you do that, you'll never be the same again
<1> hahaa
<2> abaddon, that's why i'm making him do it
<0> ot: sometime died inside me when i did that
<1> abaddon: you're damaged for life?
<2> abaddon, i broke down crying last time I did that
<0> sgtunix: its worse than you know, ive been doing kernel devel in linux for a couple weeks now at work
<0> SgtUnix: hense my apearance on irc
<2> i tried to choke myself with my linux kernel device drivers book.
<1> abaddon: i had an idea that it might be a bit messy, but it's really that bad? so have them use FreeBSD instead for the product ? :)
<8> "The sizeof operator yields the size (in bytes) of its operand, which may be an expression or the parenthesized name of a type. The size is determined from the type of the operand. The result is an integer."
<8> Hmpf.
<0> no-one was immediately available to pull teeth or sell my internal organs to, so i figured irc was next in line
<7> OT: include/linux/acct.h, see for comp_t is a 16-bit "floating" point number with a 3-bit base 8
<2> YAY! IRC
<0> sgtunix: they do on some products
<2> MishMan, are you sure? I think that's fixed point.
<1> abaddon: which company was it again - I forget
<8> abaddon, you should thank the heavens you're not doing win32 driver developement.
<0> sgtunix: juniper
<1> ah, that's it.
<0> kidart: ive done that before many times, i much prefer win32 driver devel
<7> OT: is has exponent
<0> kidart: they have real kernel debuggers, they have a solid well defined kernel api that wont change (gets expanded but doesnt depricate) just because some guy felt like it
<7> fixed point would have no exponent
<2> MishMan, interesting.
<8> abaddon, true. But the model is horrible.
<8> abaddon, I had the distinct lack of pleasure of porting a linux driver to win32.
<1> pretty big company, 4,100+ employees, revenue of 2.1 bn. Hmm, JunOS
<10> if I create a hash table with hcreate_r() and add entries with hsearch_r() ... is there any way to print out the whole table without having to know each key in the table?
<1> abaddon: are they using linux because it's a technical advantage or becaused they need the buzzword factor? ;)
<8> abaddon, the amount of code grew by an unreasonable percentage.
<0> kidart: i dont care if the os ****s, so long as its api is stable (in that the api doesnt change every 20 seconds) i can actually do my job, on linux i dont care how spank-tastic the new code might be (its not in point of fact) i want my **** to work and keep working
<8> abaddon, the API is stable, but it's stably BAD.
<0> kidart: thats because they are very different worlds, portin gdrivers blows no matter where you port from, but unix to win32 in general is hard
<2> MishMan, it doesn't appear to have a nonnormalized form. but otherwise it's pretty good.
<2> http://lxr.linux.no/source/kernel/acct.c#L279
<2> sorry i didn't believe you. eheh
<0> kidart: i agree that its bad, but stable is more important to me than good, i can work around other peoples bad code, its what i do, but i cant work around it if i cant link with the ****ing kernel because i blinked and they changefd it all
<9> the Australian Government is giving out software to filter internet porn. I wonder what spy functions it also has
<8> abaddon, at some point, I wanted to do something, and did another port, from the linux source to Jungo WinDriver. Was almost line-for-line. Which was pretty amusing.
<0> sgtunix: they're usin git because they aquired people using it and they've not changed it to bsd yet
<8> s/do/check/
<8> Just goes to show how overengineered WDM is.
Return to
#c
or
Go to some related
logs:
#netbsd
site:www.quotesdb.info thabondgirl
#politics
#beginner
silicon image driver
local windows shutdownscript
philpins chat
#freebsd
#windows
nx delete available sessions
|
|