[PATCH 2/3] Specify stack_burn for SHA256 cipher

Dmitry Eremin-Solenikov dbaryshkov at gmail.com
Mon Jul 15 11:26:59 CEST 2013


On Mon, Jul 15, 2013 at 11:53 AM, Werner Koch <wk at gnupg.org> wrote:
> On Sat, 13 Jul 2013 16:50, dbaryshkov at gmail.com said:
>
>>    hd->bctx.count = 0;
>>    hd->bctx.blocksize = 64;
>> +  hd->bctx.stack_burn = 74*4+32;
>
> I can't find the original code.  Is that based on a private branch?

Ah... It looks so. I refactored most of hash functions working with
blocks of data, but did it so long ago, that I forgot about that.

Would you be interested in a patch that separates common
block-handling code from hash functions (except whirpool and
sha512)?

-- 
With best wishes
Dmitry



More information about the Gcrypt-devel mailing list