[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1aed7f4d705f7de7d118b50f9a78816b@smtp.hushmail.com>
Date: Sun, 17 Jun 2012 00:38:10 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Re: [patch] optional new raw sha1 implemetation
On 2012-06-17 00:15, magnum wrote:
> Tavis,
>
> I see a problem in binary(). You alloc space for a binary and return it
> to john, but john will only copy it so this is a memory leak. We could
> change it to a one-time alloc:
>
> static uint32_t *result;
> if (!result)
> result = mem_alloc_tiny(SHA1_DIGEST_SIZE, MEM_ALIGN_SIMD);
>
> However, as the alignment requirement is not really in binary() itself,
> we could just as well do this:
>
> static uint32_t result[SHA1_DIGEST_SIZE/sizeof(uint32_t)];
>
> And actually no matter how we do this, john will (currently) copy it
> with MEM_ALIGN_WORD in loader.c.
After reading the rest of the code, MEM_ALIGN_WORD seem to be OK. I'll
commit the static fix.
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.
