Date: Sun, 8 Sep 2013 18:45:05 +0200 From: magnum <john.magnum@...hmail.com> To: john-dev@...ts.openwall.com Subject: Re: mask mode bug On 7 sep 2013, at 03:07, Solar Designer <solar@...nwall.com> wrote: > On Thu, Sep 05, 2013 at 08:35:18PM +0200, magnum wrote: >> On 5 Sep, 2013, at 19:44 , Solar Designer <solar@...nwall.com> wrote: >>> FWIW, when I run Sayantan's bleeding-mask from a week ago (didn't update >>> since then) on "well" against a single raw-md5 hash, it uses over 1300 MB >>> on the GTX TITAN card. (Perhaps the same per-GPU on 7990, but there's >>> no easy reporting of that.) >> >> Do you have any idea what GWS you were using at that time? I run very low, 64K or lower depending on mask. > > I did not adjust GWS, so: > > $ ./john pw -pla=2 -form=raw-md5-opencl -mask='?l?l?l?l?l?l?l?l?l?l' > Device 4: GeForce GTX TITAN > Local worksize (LWS) 64, global worksize (GWS) 4194304 > Loaded 1 password hash (Raw-MD5-opencl [MD5 OpenCL (inefficient, development use only)]) > Press 'q' or Ctrl-C to abort, almost any other key for status > Using kernel md5_ccc... > 0g 0:00:03:02 0g/s 2554Kp/s 1729Mc/s 1729MC/s aaabneuvrs On a side note the above must mean a kernel duration of 1.64 seconds. Targetting 200 ms, 512K would be sane. Are we losing a lot of performance at 512K? (testing that) OK I get 1693Mc/s so we lose 2%. > With GWS=65536, memory usage reduces to "740MB" - still too much for > cracking of one hash, but is more reasonable. Unfortunately, the speed > is worse: > > 0g 0:00:03:04 0g/s 1976Kp/s 1342Mc/s 1342MC/s aaabetnqcu..aaabetpmpb The key buffer should be 640KB (n.b. not MB) in this case. Considering we only have a single hash, I wonder what is using hundreds of megabytes of memory? BTW when testing a GWS of 512K only 773MB was used so the difference in memory footprint between 64K and 512K is much less than between 512K and 4M. I wonder why. magnum
Powered by blists - more mailing lists
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.