john-dev - source() called twice for each crack

Follow @Openwall on Twitter for new release announcements and other news

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <2ca4786135bcdc0de7653e033a8a2a03@smtp.hushmail.com>
Date: Sun, 15 Nov 2015 02:03:40 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: source() called twice for each crack

Solar,

The source() function is always called twice in a row for no good 
reason. On a successful guess, we always have something like this (in 
various code paths):

	if (crk_methods.cmp_exact(crk_methods.source(
	    pw->source, pw->binary), index))
	if (crk_process_guess(salt, pw, index))
		return 1;

Then within crk_process_guess() we call it again for passing to 
log_guess(). A trivial fix would be to add a parameter to 
crk_process_guess() so we'd have:

	char *source = crk_methods.source(pw->source, pw->binary);
	if (crk_methods.cmp_exact(source, index))
	if (crk_process_guess(salt, pw, index, source))
		return 1;

Then in crk_process_guess() the source would be readily available.

I noticed this while working on something else and just wanted to 
mention it before I forget all about it. Perhaps the above change this 
would slightly improve the 29m_md5 tests you were focusing on a while 
ago, for example. The source function in raw-md5 is not terribly slow 
but in other cases I guess it could be.

magnum

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.