crypto: x86/aes-xts - additional optimizations

Browse source

Reduce latency by taking advantage of the property vaesenclast(key, a) ^
b == vaesenclast(key ^ b, a), like I did in the AES-GCM code.

Also replace a vpand and vpxor with a vpternlogd.

On AMD Zen 5 this improves performance by about 3%.  Intel performance
remains about the same, with a 0.1% improvement being seen on Icelake.

Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

...

This commit is contained in:

Eric Biggers 2024-12-12 13:28:45 -08:00 • committed by Herbert Xu

parent 68e95f5c64

commit 3cd46a78ee

1 changed files with 90 additions and 55 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

145

arch/x86/crypto/aes-xts-avx-x86_64.S

View file