|
clang snype all hail our dark master C++14
|
# ? Jul 27, 2014 17:57 |
|
|
# ? Mar 29, 2024 14:24 |
|
Avenging Dentist posted:all hail our dark master C++14
|
# ? Jul 27, 2014 18:24 |
|
c++17 will be the basest of c++s
|
# ? Jul 27, 2014 18:26 |
|
MIPSPro++
|
# ? Jul 27, 2014 19:38 |
|
rjmccall posted:my guess, without reading any gcc code (which i'm not allowed to do, lol), is that linus has actually diagnosed the problem correctly, but it's buried under a long discussion of what's probably a red herring (also under a giant pile of assholeishness). the red zone is almost certainly a distraction here. gcc appears to be spilling a simple constant (which is definitely screwy) to a slot which should be well within the stack frame, except unfortunately it's doing that spill just before the stack frame has actually been established. there isn't normally a reason you'd need to materialize a constant in the prologue; wild guess, there's something weird (inline assembly?) at the start of the function, and some piece of code is trying to position a "hey, i need this constant now" assertion before that and isn't being careful enough to avoid the prologue. that mistake is also confusing the register allocator, hence the spill u could write pretty much anything about this & that'd work for me
|
# ? Jul 27, 2014 22:47 |
|
lunis
|
# ? Jul 27, 2014 22:47 |
|
remember back when SGI was still A Company That's In Business (as opposed to existing in name only, like they do now) and they were gonna donate XFS to linux, but the linux devs were huge assholes to the SGI devs assigned to it?
|
# ? Jul 28, 2014 01:22 |
|
Doc Block posted:remember when linux devs were huge assholes?
|
# ? Jul 28, 2014 01:35 |
|
Doc Block posted:remember linux?
|
# ? Jul 28, 2014 01:40 |
|
also the funniest part is that you can still say "hey gcc is bein real fukken dumb!!" without coming across like a whiny little baby but linus and his fans don't seem to get that but then what do i know, i'm not the person who ~*~invented an operating system~*~ (that only even gained a foothold because bsd was mired in legal issues at the time)
|
# ? Jul 28, 2014 01:42 |
|
i think you mean gnu/linus
|
# ? Jul 28, 2014 01:44 |
|
excuse you, linux made a kernel, which is only half of the gnu/linux operating system hurd will be finished any day now
|
# ? Jul 28, 2014 01:44 |
|
When Linus passed away, he was buried in a computer cemetary. A couple days later, the town drunk was walking through the cemetery and heard profanity and fart noises coming from the area where Linus was buried. Terrified, the drunk ran and got the village technophile to come and listen to it. The technophile bent close to the grave and heard some faint, unrecognizable beeping and the unmistakable sound of farts coming from the grave. Frightened, the technophile ran and got the town programmer. When the programmer arrived, he bent his ear to the grave, listened for a moment, and said, "Ah, yes, that's the Linux Kernel, being played backwards." He listened a while longer, and said, "There's the Git source code, and it's backwards, too. Most puzzling." Suddenly the realization of what was happening dawned on the programmer; he stood up and announced to the crowd that had gathered in the cemetery, "My fellow citizens, there's nothing to worry about. It's just Linus decompiling!"
|
# ? Jul 28, 2014 01:46 |
|
pram posted:i think you mean gnu/linus if you say gnu/linux three times in a dark room, rms will appear in your emacs buffer and sing the free software song to end the world
|
# ? Jul 28, 2014 01:53 |
|
Avenging Dentist posted:clang snype #include <gas>
|
# ? Jul 28, 2014 21:05 |
|
USSMICHELLEBACHMAN posted:When Linus passed away, he was buried in a computer cemetary. A couple days later, the town drunk was walking through the cemetery and heard profanity and fart noises coming from the area where Linus was buried.
|
# ? Jul 29, 2014 05:27 |
|
Linus is (somewhat) Finnish and he is neither throwing a tantrum or being rude here. This is how Finns normally communicate and if some stupid third-worlders can't deal with that well then maybe they should just stop being so loving dumb and backwards
|
# ? Jul 29, 2014 12:50 |
|
hey Jerry, are you not accepting ethnic Swedes as Finns?
|
# ? Jul 29, 2014 13:31 |
|
or is it just that Linus is a traitor to his country and became a US citizen?
|
# ? Jul 29, 2014 13:32 |
|
axolotl farmer posted:hey Jerry, are you not accepting ethnic Swedes as Finns? He was born in Hel-Looksfors, not in Egentliga Finland
|
# ? Jul 29, 2014 13:34 |
|
icc or eat a dick
|
# ? Jul 29, 2014 13:45 |
|
.lnk to the past posted:icc or eat a dick
|
# ? Jul 29, 2014 18:30 |
|
vc++ 6 is and will remain the greatest piece of software ever designed.
|
# ? Jul 29, 2014 18:53 |
|
Egan Yardley posted:vc++ 6 is and will remain the greatest piece of software ever designed. it's the greatest IDE, I'll grant you that. how is Xcode still so bad after all these years?
|
# ? Jul 29, 2014 20:14 |
|
Doc Block posted:MIPSPro++ turns out the old compiler had bugs with a few math operations that the new ones fixed users don't care
|
# ? Jul 29, 2014 20:30 |
|
do new compilers make assembler that is better than handcrafted poo poo
|
# ? Jul 29, 2014 21:17 |
|
depends on who's handcrafting the poo poo. a modern compiler is probably smarter than you about basic integer/pointer arithmetic, and it knows a lot of subtle crap about instruction ordering and microarchitectural latencies. on the other hand, it's required to follow the platform abi, can easily be blocked by false dependencies, doesn't always know when it's worthwhile to chase trade-offs, doesn't necessarily realize that there's this one perfect instruction for summing the odd bits of a floating-point mantissa, and occasionally just makes bad decisions
|
# ? Jul 29, 2014 21:41 |
|
a piece of poo poo evidently
|
# ? Jul 29, 2014 21:58 |
|
echinopsis posted:a piece of poo poo evidently or is it!!!!!!
|
# ? Jul 29, 2014 22:05 |
|
it is possible to assembler better than a computer it is possible for a computer to better than you both are possible one way to find out: do both and profile
|
# ? Jul 29, 2014 22:51 |
|
Bloody posted:it is possible to assembler better than a computer i would let the computer do it first and then handcode if performance becomes an issue but it is your decision do not let me impede your creativity!
|
# ? Jul 29, 2014 22:55 |
|
Bloody posted:it is possible to assembler better than a computer Ditto. I'll often drop down to node.js if I really need to be close to the metal.
|
# ? Jul 29, 2014 22:57 |
|
Pham Nuwen posted:i would let the computer do it first and then handcode if performance becomes an issue but it is your decision do not let me impede your creativity! same also lmao at doing this outside of an embedded platform or i guess algotrading but lmao if you algotrade on x86 anyways
|
# ? Jul 29, 2014 22:58 |
|
if you arent algotrading on asics you're just a loving joke, get out of my face
|
# ? Jul 29, 2014 22:59 |
|
Bloody posted:if you arent algotrading on asics you're just a loving joke, get out of my face and none of that standard cell or gate array garbage either!
|
# ? Jul 29, 2014 23:00 |
|
Avenging Dentist posted:Ditto. I'll often drop down to node.js if I really need to be close to the metal.
|
# ? Jul 29, 2014 23:00 |
|
hobbesmaster posted:and none of that standard cell or gate array garbage either! ure aloud to use fpgas as a stopgap while you wait for your 14nm fab run to come back i wonder if anybody actually does these things anybody want to invest in my new fund
|
# ? Jul 29, 2014 23:02 |
|
sometimes i load up vs and I'm all like now for some assembly then I'm like wtf am i doing
|
# ? Jul 29, 2014 23:38 |
|
Bloody posted:ure aloud to use fpgas as a stopgap while you wait for your 14nm fab run to come back bitcoiners.
|
# ? Jul 29, 2014 23:57 |
|
|
# ? Mar 29, 2024 14:24 |
|
a human can outperform a compiler in tiny chunks and specific optimizations cases. a compiler will do better with general purpose code and problems that require complex register allocation and branching. basically the larger the code, the better the compiler will do. humans also have more information available to them than the compiler knows about the intention of the code and can do a better job optimizing for the intended purpose.
|
# ? Jul 30, 2014 01:45 |