Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20111128011228.GA21755@openwall.com>
Date: Mon, 28 Nov 2011 05:12:28 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: .bss size in -jumbo

On Mon, Nov 28, 2011 at 05:07:29AM +0400, Solar Designer wrote:
> Since there are no format destructors - I mean, we have init(), but
> not the opposite - the entire 140 MB will end up getting allocated by a
> full --test run anyway, ...

Correction: in formats where we scale the number of key slots based on
actual thread count, the actual memory usage on most systems will be
substantially less.  It's just the when those arrays are in .bss,
they're pre-allocated for the maximum supported key slot count, even if
a lot fewer slots are used with typical thread counts.  With dynamic
allocation, we'd be able to allocate memory for fewer key slots.

Alexander

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.