Herbert Xu 3a92d687c8 crypto: sha512 - Avoid stack bloat on i386
Unfortunately in reducing W from 80 to 16 we ended up unrolling
the loop twice.  As gcc has issues dealing with 64-bit ops on
i386 this means that we end up using even more stack space (>1K).

This patch solves the W reduction by moving LOAD_OP/BLEND_OP
into the loop itself, thus avoiding the need to duplicate it.

While the stack space still isn't great (>0.5K) it is at least
in the same ball park as the amount of stack used for our C sha1
implementation.

Note that this patch basically reverts to the original code so
the diff looks bigger than it really is.

Cc: stable@vger.kernel.org
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2012-02-05 15:09:28 +11:00
..
2011-07-26 16:49:47 -07:00
2010-02-16 20:26:46 +08:00
2010-12-02 14:47:16 +08:00
2010-11-13 21:47:55 +09:00
2010-02-16 20:29:45 +08:00
2010-05-26 10:36:51 +10:00
2008-06-02 15:46:51 +10:00
2010-02-16 20:33:49 +08:00
2010-12-02 14:47:16 +08:00
2008-08-29 15:50:02 +10:00
2010-12-02 14:47:16 +08:00
2011-10-21 14:24:03 +02:00
2008-04-21 10:19:34 +08:00
2008-04-21 10:19:34 +08:00
2011-08-06 18:32:45 -07:00
2007-10-10 16:55:38 -07:00
2008-04-21 10:19:34 +08:00
2008-12-25 11:02:21 +11:00