[PATCH 6/7] block-sha1: Use '(B&C)+(D&(B^C))' instead of '(B&C)|(D&(B|C))' in round 3

From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Aug 6, 2009, 15:24 UTC
Message-ID: <alpine.LFD.2.01.0908060822390.3390@localhost.localdomain>
In-Reply-To: <alpine.LFD.2.01.0908060820110.3390@localhost.localdomain>

From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Thu, 6 Aug 2009 07:27:57 -0700
Subject: [PATCH 6/7] block-sha1: Use '(B&C)+(D&(B^C))' instead of '(B&C)|(D&(B|C))' in round 3

It's an equivalent expression, but the '+' gives us some freedom in instruction selection (for example, we can use 'lea' rather than 'add'), and associates with the other additions around it to give some minor scheduling freedom.

Suggested-by: linux@horizon.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---

A small micro-optimization. It may not matter hugely, but it's definitely the right thing to do. And with the new macroized thing, you can see clearly what part of the SHA1 rounds it affects, and how.

 block-sha1/sha1.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/block-sha1/sha1.c b/block-sha1/sha1.c
index 4837d58..9a060a6 100644
--- a/block-sha1/sha1.c
+++ b/block-sha1/sha1.c
@@ -112,7 +112,7 @@ void blk_SHA1_Final(unsigned char hashout[20], blk_SHA_CTX *ctx)
 #define T_0_15(t)  SHA_ROUND(t, SHA_SRC, (((C^D)&B)^D) , 0x5a827999 )
 #define T_16_19(t) SHA_ROUND(t, SHA_MIX, (((C^D)&B)^D) , 0x5a827999 )
 #define T_20_39(t) SHA_ROUND(t, SHA_MIX, (B^C^D) , 0x6ed9eba1 )
-#define T_40_59(t) SHA_ROUND(t, SHA_MIX, ((B&C)|(D&(B|C))) , 0x8f1bbcdc )
+#define T_40_59(t) SHA_ROUND(t, SHA_MIX, ((B&C)+(D&(B^C))) , 0x8f1bbcdc )
 #define T_60_79(t) SHA_ROUND(t, SHA_MIX, (B^C^D) ,  0xca62c1d6 )
 
 static void blk_SHA1Block(blk_SHA_CTX *ctx, const unsigned int *data)

-- 
1.6.4.31.g154b2.dirty