CUDA supported bruteforcers

Moderator: BarsMonster

IvanG
Posts: 89
Joined: Wed May 06, 2009 12:42 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by IvanG » Wed Jan 13, 2010 5:24 pm

vampyr wrote:Also: ihashgpu appears to be a direct copy of some of my old code, as it has exactly the same bugs, and around the same speed (!) Because, hey, look at the post date of my govsec cracker, then at ihashgpu.
You're funny boy. How comparing 550M with 720M you ends as "around the same speed"? :P Not mention the latest v0.7 where SHA1 should hit 870M on4870x2. You'll never be able to reach this with crappy Brook+. And especially this (single 5870):
Hash type: SHA1, Hash: a72befac3e58eb24d559d9fe0045cfdf090782e2
Salt: e1 6b ed 51
Device #0: [RV870] 850.00 Mhz 1600 SP
Hardware monitoring enabled, threshold temperature is 90°C.
CURPWD: UjBhpI DONE: 79.21% ETA: 3s AVRSPD: 1368.6M

vampyr
Posts: 21
Joined: Sun Mar 29, 2009 9:09 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by vampyr » Wed Jan 13, 2010 6:05 pm

Your cracker, on my box with a single ATI4870x2: 601mhashes/s.
Speed of the previous (single hash version) of my cracker: 600mhashes/s
Speed of the current (multihash+client/server version): 550mhashes/s, but stable up to 100k hashes.

I'd say that its pretty decent, for some 'crappy' BROOK+ app.
Then again, my app is not focused on the cracking of single hashes, it's more about pwning large hash lists, quickly.

But i do like your status screen tho, flashy for a console app. Also, gotta give you this, nice fast startup time, no unit tests or benchmarks.
But i guess i can't skip those, because of my networking shit.

IvanG
Posts: 89
Joined: Wed May 06, 2009 12:42 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by IvanG » Wed Jan 13, 2010 6:32 pm

vampyr wrote:Your cracker, on my box with a single ATI4870x2: 601mhashes/s.
Which is weird actually. I had 590M=290M+300M with 4770+4850, so at 4870x2 it must be at least 20% faster. Which Catalyst version do you use and which ighashgpu version?

vampyr
Posts: 21
Joined: Sun Mar 29, 2009 9:09 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by vampyr » Wed Jan 13, 2010 6:36 pm

Latest version of both the driver and ihashgpu, but this rig runs vista, so the 2nd core is always in 2d mode, with my cracker too.
Guess it's just the drivers or something being crap.

But hey, if you want a really fair comparison, i could hack up a single hash version of sha1 and mysql5 for the next version of my cracker;)
Because i can guarantee you, that single hash version is completely ALU bound, unlike the version that works on lists, which is somewhat memory bould (10% speed loss).

vampyr
Posts: 21
Joined: Sun Mar 29, 2009 9:09 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by vampyr » Wed Jan 13, 2010 6:38 pm

Also "I’ve removed last 4 rounds from SHA-1 (“reversed” in other words). Last optimization was already done earlier for CUDA code, " This line makes me wonder..
Because, well, i did that for my NVIDIA kernel, but not for my ATI kernel, in the previous versions of my cracker.
You also have all the old bugs i had.
That plus having comparable speeds makes this likely to be more than just a coincidence.

IvanG
Posts: 89
Joined: Wed May 06, 2009 12:42 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by IvanG » Wed Jan 13, 2010 7:23 pm

:roll:
Seriously, :roll: :lol: .

vampyr
Posts: 21
Joined: Sun Mar 29, 2009 9:09 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by vampyr » Wed Jan 13, 2010 8:52 pm

Hehe, sorry man, but that shit just looked SO suspicious. Been dissasembling and reading your executable for the past hour or so, and it appears to have been written directly in GPU assembler;) Funny, how close the compiler can come to highly optimized assembly, not?

D3ad0ne
Posts: 111
Joined: Fri Sep 04, 2009 5:37 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by D3ad0ne » Thu Jan 14, 2010 9:07 pm

vampyr, I'll do some independant testing later this week.

vampyr
Posts: 21
Joined: Sun Mar 29, 2009 9:09 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by vampyr » Fri Jan 15, 2010 11:40 am

Thanks bro. I'm working on hacking up the brook+ compiler to support the bitalign and bytealign instructions atm.

IvanG
Posts: 89
Joined: Wed May 06, 2009 12:42 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by IvanG » Fri Jan 15, 2010 1:47 pm

vampyr wrote:Funny, how close the compiler can come to highly optimized assembly, not?
The problem (for me at least) that it isn't possible to write directly in GPU assembly. Brook+ taking C-like code and translate it into ATI's IL (intermediate language), then this IL compiled by calclCompile function from ATI's drivers pack into GPU ISA and finally used by GPU device. I've just get rid of Brook+ -> IL step. Because Brook itself is really crappy thing and I see no point to fight with Brook to force it produce good IL code rather than writing IL code from scratch by my own. All optimizations performed by calclCompile() function not Brook, so it's kinda no point to optimize kernels one step away from actual optimizer.

Anyway, for me it doesn't looks so weird that SHA1 speed nearly the same for different methods. After all it's absolutely the same algorithm compiled and optimized by absolutely the same CAL layer. The only differences can be in initial "thread id to ansi/unicode password" transformation, sha1 body function is exactly the same and there can be some variations among passwords per thread, threads per block distribution. However, my latest sha1 version got nice speed-up even without bitalign instructions (about 20%) and yours dropped by 10% (though it supports multihashing). So I see a lot of fights "you vs Brook" in future :P.

Unfortunately, I wasn't able to test your program at all. It simply producing a lot of crash messages and pseudographics but nothing useful. May be this caused by my 5770+4770/Vista64 config, may be not but I'm just lazy to figure out what's wrong. I can't call your readme.txt a documentation neither :). May be D3ad0ne will be more lucky.

vampyr
Posts: 21
Joined: Sun Mar 29, 2009 9:09 pm
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Re: CUDA supported bruteforcers

Post by vampyr » Fri Jan 15, 2010 3:52 pm

Hey, the point isn't a full documentation, the point is enough information to get it up and running.
What are the error messages it produces? The application should output some .log files, for both kernels, and the client.
Because this might just be caused by your weird setup :P Which, is all the more interesting :)

Also, i agree on being tired of fighting with BROOK+. I'm rolling my own compiler atm, and i can do in ~200 lines what brook+ can do in ~20k.
The thing i'm writing also translates more directly to IL, kind of like a glorified find and replace;)

Post Reply
[phpBB Debug] PHP Warning: in file [ROOT]/vendor/twig/twig/lib/Twig/Extension/Core.php on line 1266: count(): Parameter must be an array or an object that implements Countable

Who is online

Users browsing this forum: No registered users and 1 guest