|
but the standard libraOHHHHH I GET IT
|
# ? Feb 27, 2013 06:54 |
|
|
# ? Jun 11, 2024 10:56 |
|
page 420
|
# ? Feb 27, 2013 07:21 |
|
i'm curious when/if we'll see a serious garbage collected implementation of c++ also smoke piff get piffy
|
# ? Feb 27, 2013 10:38 |
|
FamDav posted:i'm curious when/if we'll see a serious garbage collected implementation of c++
|
# ? Feb 27, 2013 11:25 |
|
Not sure if this counts as 'serious', but it is a C/C++ garbage collector: http://www.hpl.hp.com/personal/Hans_Boehm/gc/ http://en.wikipedia.org/wiki/Boehm_garbage_collector It's used by Mozilla, Mono, irssi, Scheme implementations, etc. too.
|
# ? Feb 27, 2013 14:23 |
|
FamDav posted:i'm curious when/if we'll see a serious garbage collected implementation of c++ I think u need c++/cli
|
# ? Feb 27, 2013 14:26 |
|
There are conservative GCs, you can't get a real GC because the C++ type system is optional.
|
# ? Feb 27, 2013 16:18 |
|
Nomnom Cookie posted:Java gets within spitting distance of c++ for performance so unless you're writing kernel code or need real-time guarantees I fail to see the point of using c++. C is the obvious choice of greybeards and Linux spergs but there's no niche for c++
|
# ? Feb 27, 2013 16:40 |
|
have things changed, can you get high performance graphics out of java now I wouldn't wanna write that poo poo in bare C
|
# ? Feb 27, 2013 16:49 |
|
hotspots gc is fine. gc pausing is one of those non-issues that greybeards bring up when they talk about savage optimizations they can only do in c. there are sometimes specific reasons to pick a different gc mechanism or to tweak the settings but those are only after you've actually determined theres a performance problem caused by gc which is almost never the case. its another case of guesswork theoretical analysis where autismals will try to pre-optimize the system instead of looking at the real world metrics. you are not writing perfect code that is somehow beyond the capabilities of the gc. you're either doing dumb poo poo like destroying and recreating the same objects over and over OR (and this is the case in 99.99999% of cases) you have a loving memory leak. fix the leak you shithead.
|
# ? Feb 27, 2013 16:57 |
|
Mr Dog posted:There are conservative GCs, you can't get a real GC because the C++ type system is optional. all GC is conservative, the question whether a piece of memory will ever be used again is undecidable
|
# ? Feb 27, 2013 16:57 |
|
GC pausing is an issue as soon as you have any hard real time constraints (and some soft ones).
|
# ? Feb 27, 2013 16:58 |
|
otoh many things are an issue when you have real time constraints
|
# ? Feb 27, 2013 17:01 |
|
Internaut! posted:have things changed, can you get high performance graphics out of java now insert minecraft joke
|
# ? Feb 27, 2013 17:01 |
|
i sort of said this spread out across a couple of posts, but reference counting plus everything-is-a-table is a neat language model. only reference count on the table level, and encourage large tables rather than a billion atomic objects floating around. combines beautifully with column-orientation basically kx k/q is the holy land for scaling
|
# ? Feb 27, 2013 17:04 |
|
Shaggar posted:fix the leak you shithead. but this is why someone is using java instead of c!
|
# ? Feb 27, 2013 17:04 |
|
Shaggar posted:hotspots gc is fine. gc pausing is one of those non-issues that greybeards bring up when they talk about savage optimizations they can only do in c. there are sometimes specific reasons to pick a different gc mechanism or to tweak the settings but those are only after you've actually determined theres a performance problem caused by gc which is almost never the case. its another case of guesswork theoretical analysis where autismals will try to pre-optimize the system instead of looking at the real world metrics. You're only saying this because you're so used to GC pauses your brain is literally editing them out http://www.eaglemanlab.net/papers/StetsonetalNeuron2006.pdf
|
# ? Feb 27, 2013 17:05 |
|
Shaggar posted:hotspots gc is fine. gc pausing is one of those non-issues that greybeards bring up when they talk about savage optimizations they can only do in c. there are sometimes specific reasons to pick a different gc mechanism or to tweak the settings but those are only after you've actually determined theres a performance problem caused by gc which is almost never the case. its another case of guesswork theoretical analysis where autismals will try to pre-optimize the system instead of looking at the real world metrics. hahahahaha
|
# ? Feb 27, 2013 17:05 |
|
MononcQc posted:GC pausing is an issue as soon as you have any hard real time constraints (and some soft ones). and for the 1% of users where that's the case then they can do some real analysis of the running system to figure out how to tune it. for the rest, its not an issue
|
# ? Feb 27, 2013 17:06 |
|
sorry officer I rear-ended the other driver because someone didn't tune the GC of my braking system correctly and it skipped on the signal I sent
|
# ? Feb 27, 2013 17:08 |
|
I lo ve it w hen my garba ge collec tor kic ks in in m y bro wser
|
# ? Feb 27, 2013 17:21 |
|
gc pausing is a non issue! *display ad bidding fleet pauses the world for 50s for GC, literally six figures go straight down the drain*
|
# ? Feb 27, 2013 17:28 |
|
not the gcs fault ur code is bad
|
# ? Feb 27, 2013 17:42 |
|
lol okay
|
# ? Feb 27, 2013 17:43 |
|
shaggar!!!!!
|
# ? Feb 27, 2013 17:45 |
|
every time the world stops for gc i forget where my goalposts are
|
# ? Feb 27, 2013 17:46 |
|
moving forward using all my mem allocating with you was never second best i saw the program crashing all around your stack never knowing who's right in the forum black i'll stop the world and gc with you
|
# ? Feb 27, 2013 17:51 |
|
horse mans posted:gc pausing is a non issue! use non-pausing/concurrent gc for ad bidding
|
# ? Feb 27, 2013 17:52 |
|
Surely not every last line of your codebase is performance critical Do the performance critical stuff in C, layer some plang interpreter on top of it and export your high performance primitives as a dsl i still think code where you don't explicitly qualify your types is needlessly obnoxious to maintain for anything bigger than like a college project though
|
# ? Feb 27, 2013 17:56 |
|
^^^^: FINE WHATEVER If gc is so great then why does my jvm binary depend on libc.so.6?
|
# ? Feb 27, 2013 17:57 |
|
MononcQc posted:Not sure if this counts as 'serious', but it is a C/C++ garbage collector: lol mozilla and mono both use it as their "dumb" gc issues intended for debugging their complex gc
|
# ? Feb 27, 2013 18:08 |
whoops
|
|
# ? Feb 27, 2013 18:18 |
|
its page 420, let us all hate on php and smoke a drug together
|
# ? Feb 27, 2013 18:22 |
|
KBlair posted:its page 420, let us all hate on php and smoke a drug together
|
# ? Feb 27, 2013 18:25 |
|
|
# ? Feb 27, 2013 18:26 |
|
i once tried php, wasnt a fan, gave up web programming forever
|
# ? Feb 27, 2013 18:47 |
|
Yakattak posted:gave up web programming forever more people should do this
|
# ? Feb 27, 2013 18:48 |
|
horse mans posted:
|
# ? Feb 27, 2013 18:48 |
|
haha you suck atp osting
|
# ? Feb 27, 2013 18:50 |
|
|
# ? Jun 11, 2024 10:56 |
|
no chocckakacokacoka just sucks at quote nesting
|
# ? Feb 27, 2013 18:51 |