| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Comments:
<0> padded to what, default word size of the target machine?
<1> depends. but yea. that's usually what it is padded to
<2> grr I can't find the permission to open this door
<2> stupid tron 2.0
<1> so union { short s; char c[2]; } would probably end up being sizeof 4.
<0> the union is a faster/cleaer way to do it though right?
<2> rly?
<1> it's the least work way. if you can live with it's restrictions.
<1> s/it's/its
<2> only use I've ever had for a union was splitting registers up into pieces in an emulator
<1> i've used union hundreds of times.
<0> thats extactly
<0> what im doing lol
<1> struct object { enum object_type type; union value value; }
<2> union {unsigned short HL; struct {unsigned char L; unsigned char H;};};
<2> union {unsigned short AF; struct {unsigned char F; unsigned char A;};};
<2> :P
<1> or struct attribute { char *name; enum attr_type type; union attrvalue value; }
<1> LordZid, i'm lazy and just use an array. and enums to index the array if i must.
<0> i was just wondering if union was gonna be slower than using bitshift and or to get the hi/lo byte of the word
<1> the union should be as fast as bit shifts, if not faster.
<1> if you bitshift in the right way the compiler will optimize it out pretty well.
<1> and on ARM you get bitshifts that are multiples of 2 for free.
<0> its for ps1 so i dont think its compiler can optimise too well
<0> its some old gnu c compiler
<0> from like 1994
<1> if it's gcc for the mips4000 on the ps1, then the compiler will optimize it just fine.
<2> MIPS?
<2> yep
<0> yeah mips r3000a cpu
<2> I was going to do some psx dev
<2> couldn't find a decent toolchain
<2> dcc me ploz
<0> i have the psyq stuff for it
<1> i just tore apart a gamecube and use that for a little powerpc devel. i have a 1gb SD memory card in it, ethernet, etc.
<3> SgtUnix, man, they're really slow in getting all the results out
<3> SgtUnix, missing one exam
<4> OrngeTide: bitshifts that are multiples of two meaning you can multiply by 2^(2^k)?
<0> LordZid do you want the psx dev stuff? :s
<1> evilgeek, no 2^(2*k)
<4> oh, multiples. sorry, i'm stupid.
<1> so you can shift 0 to 30. but you only have 4 bits to encode that shift.
<4> ugh.
<1> it's like an extra addressing mode on arm. so many register and immediate parameters support it.
<1> if you want to do an odd shift you have to use an instruction for that.
<4> that's cool, though.
<0> the parity bit of a number should be the same on any machine right? :s
<1> yea. depending on if it's odd or even parity.
<1> and that the machines have the same number of bits in a char or word.
<1> crc is the same as well. it's just a mathematical calculation. even if you have different endian the crc is the same.
<5> http://thedailywtf.com/forums/thread/78254.aspx
<0> hmm psx is 32bit im pretty sure that everything else is the same but i'll have to check it
<1> the psx is a very modern big endian processor. it's going to be pretty much like any computer that is big endian.
<6> SgtUnix: funny stuff
<6> doesn't oracle have a daemon mode? yeesh...
<1> neat. OpenBFS (journaling filesystem) is MIT licensed. i wonder how hard it would be to stick into freebsd or netbsd.
<7> OrngeTide, what was it implemented for originally?
<5> CStubing: probably some other product from a vendor using oracle
<5> CStubing: i doubt oracle themselves would do such a thing.. but i dunno
<7> SgtUnix, I felt bad for the guy (thanks to some FreeBSD guy). :-P
<7> SgtUnix, since there's no turn-around for him, decided to drop it down to $200 with $40 an hour for support.
<7> I'm an idiot probably, oh well. :-)
<1> sbahra[], that's a good question. the filesystem was for BeOS. but this free version .. i don't know.
<7> "OpenBFS Becomes SkyFS for SkyOS - OSNews.com"
<7> OrngeTide, does BeOS use the vnode/VFS concept?
<1> sbahra[], it is for Haiku which is a free implementation of BeOS.
<1> sbahra[], i would think so. BeOS is posix-ish in a lot of ways. and vnodes/VFS are the easy solution.
<7> Ok.
<7> Downloading the source now.
<7> It's in a zip file :0
<7> and a dead link
<7> ehe
<1> awww
<7> Actually, none of them can be downloaded it seems
<7> http://www.bug-br.org.br/openbfs/index.phtml?section=testing
<1> that's really too bad. it looks like nobody has touched it in 3 years either.
<1> well if i find it i'll post it up some place so BSD'ers can take a peek at it at least. i'm going to go find some lunch though. adios.
<6> SgtUnix: yeah, I ***umed it was some other vendor.. I can't think of any viable reason to use sleep like that in a startup script (at least not any reason for which there's already a better solution).
<6> *not already
<5> CStubing: maybe it was some disgruntled employee who was quitting anyway, and saw it all coming.
<5> sabotage that hardly gets noticed
<5> but more likely it's just pure laziness
<6> SgtUnix: possibly, but it's too subtle to be believable sabotage ;)
<8> after hashing a string, and jumping to its index in a hashtable, is it advisible to then reverify the key stored at that location, or are some of the hashing functions good enough to just trust them that the hash was unique?
<5> CStubing: which makes it even more funny for that programmer to tell his friends at parties
<9> ncaller: Theoretically, no such hash function exists. In practice, it has to be long enough.
<9> Of course, a long and good enough hash function will take more time to evaluate than a simple comparison.
<10> anyoen here ever argued with LaRoache people?
<10> they use a lot of big worse
<10> and sound really sophisticated
<10> but, they say some really....dumb things
<10> linux has some way to prevent run-away memory allocations, apparently
<10> like, for(;;) (void)malloc(100);
<6> farhan: resource limits
<11> farhan: The memory's not actually being allocated because you're not using it. If you write to it, you'll probably see the process balloon.
<1> i love demand paging
<6> afaik, malloc will only allocate up to certain limits... shells can define these limits, and the kernel also has them... see RLIMIT_DATA
<1> for(;;) *(int*)malloc(sizeof(int))=42;
<1> malloc will alloc until there is no address space left.
<1> and on linux malloc will fail if you try to allocate a chunk larger than some size based on your memory size.
<7> OrngeTide, you do not need to cast malloc.
<7> #include <stdlib.h>
<7> farhan, tsk tsk.
<7> farhan, we had 2 meetings, another one is next Mondat at 5PM regarding exploit mitigation techniques (and will talk some about an IDS I was work on).
<7> farhan, we have another on UPC (upc.gwu.edu) about parallel programming techniques (UPC is supposedly better than mpi and any other of that stuff)
<7> farhan, and yet another coming up on embedded Linux used for fligh control software.
<11> You do need to cast malloc() to do this particular example....
<11> at least, to do it simply in one line.
<1> Xirtam, thank you
<11> I was going to give pretty much the same example but you beat me to it. :)
<1> sbahra[], did you want me to dereference a void* ? :)
<7> OrngeTide, oh dude, didn't realize you were doing " = 42"
<7> OrngeTide, my bad.
<7> ;-)
<1> ehe. it's okay. it was evil/bizarre code
<7> farhan, you should come. :-P
<7> Also, anyone in DC, we are trying to build a decent OSS club here.
<1> will there be an open bar?
<1> san jose could use an indecent OSS club.
<1> like you have to come in togos to the meetings.
<7> I am supposed to be there...
<1> san jose is kind of boring. but it's a good place if you want a wide selection of tech jobs
<7> I applied to every single UC, I was rejected by every single one (3% acceptance for international students iirc). :-/
<1> yea. UCs have gotten to be really tough to get into.
<7> OrngeTide, don't care for employment...just the "culture". It helps with ***uring you of purpose.
<7> OrngeTide, yeah. I should have went to community college honestly, then I could've transferred.
<1> my coworkers daughter has a 4.1 GPA. (wtf?!) and she's worried about getting into UC
<7> It was a big mistake to come here.
<7> OrngeTide, haha
<7> OrngeTide, how are her SAT scores?
<1> very high.
<1> i had a 2.8GPA and i'll probably always fetch a bigger salary than her. :P
<1> UCs are telling students with below 4.0 GPA to go to community college. and some 4.0s are going to CC just because the colleges have gotten that exclusive.
<1> i hate the state funding things unnecessarily. but i really do think UCs should be expanded so there is more room for students to go to them.
<1> what's the point of paying all these taxes if the state universities aren't for everyone.
<1> not being allowed to make any mistakes when you're 16-18 in high school seems pretty unreasonable to me.
<1> if you're going to make mistakes that's the best time to do it, imo.
<7> I mean, most people here think I had a 100.0 GPA or similar in high-school...
<7> but I was far from a "nerd" really (I mean, illegal activities all around).
<7> So, my options are far from open.
<7> Especially after my abysmal performance first semester here ;]
<12> i had a 2.57 GPA in highschool
<13> I knew quite a few people in my HS that had >4.0 gpa
<12> and even in univ i have a 2.0 now hehe
<1> i rarely did any work in high school. and mostly wrote code and did drugs.
<1> i had a 1.8GPA in my junior year :)
<12> drugs help for coding
<1> myrddin, i totally don't understand how you can get above 4.0.
Return to
#c
or
Go to some related
logs:
woori finance haha
ucomet
#heroin
#delphi
#politics
#heroin
#dreamweaver
#sex
#ubuntu
assholefeever
|
|