crypto: aesni - AVX and AVX2 version of AESNI-GCM encode and decode
authorTim Chen <tim.c.chen@linux.intel.com>
Wed, 11 Dec 2013 22:28:41 +0000 (14:28 -0800)
committerHerbert Xu <herbert@gondor.apana.org.au>
Fri, 20 Dec 2013 12:06:24 +0000 (20:06 +0800)
commitd764593af924930d5c15685bc5946cb943da1a55
tree5f01056b5662ba704c85274f79c70e8cd6972bda
parentfed286110f4bab01f93f06c32951fbc120fb71b1
crypto: aesni - AVX and AVX2 version of AESNI-GCM encode and decode

We have added AVX and AVX2 routines that optimize AESNI-GCM encode/decode.
These routines are optimized for encrypt and decrypt of large buffers.
In tests we have seen up to 6% speedup for 1K, 11% speedup for 2K and
18% speedup for 8K buffer over the existing SSE version.  These routines
should provide even better speedup for future Intel x86_64 cpus.

Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
arch/x86/crypto/Makefile
arch/x86/crypto/aesni-intel_avx.S [new file with mode: 0644]
arch/x86/crypto/aesni-intel_glue.c