Date: Sun, 27 Mar 2011 01:13:23 +0100 From: magnum <rawsmooth@...dband.net> To: john-dev@...ts.openwall.com Subject: Re: john scalability On 2011-03-26 20:59, magnum wrote: > Do the self tests check all five (ten) functions? I'll experiment a little. Sure enough they do. I see now the hash table size are chosen per salt, which is obvious when you think about it. So with a skewed salt distribution you may have some binaries not in a hash table, and some in each of every supported size tables. But the added sizes will mostly be needed for the unsalted formats, ie nsldap, mysql-sha1, mysql-fast. And these are not super-fast anyway. How about formats like DMD5 or MYSQL, which are unsalted but only define fmt_default stubs of the functions. This means hash tables are not used at all, right? DMD5 is very slow though. I'm not sure MYSQL is fast enough to benefit very much either. Maybe this was all moot but I learned a lot. The format the needs this most (apart from NT) is probably raw-sha1, and this is taken care of by intrinsics-2. 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.