crypto: aegis128 - duplicate init() and final() hooks in SIMD code

In order to speed up aegis128 processing even more, duplicate the init()
and final() routines as SIMD versions in their entirety. This results
in a 2x speedup on ARM Cortex-A57 for ~1500 byte packets (using AES
instructions).

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

1 files changed, 21 insertions, 0 deletions

diff --git a/crypto/aegis128-neon.c b/crypto/aegis128-neon.c
index 751f9c19..8271b1fa 100644
--- a/crypto/aegis128-neon.c
+++ b/crypto/aegis128-neon.c
@@ -8,11 +8,14 @@
 
 #include "aegis.h"
 
+void crypto_aegis128_init_neon(void *state, const void *key, const void *iv);
 void crypto_aegis128_update_neon(void *state, const void *msg);
 void crypto_aegis128_encrypt_chunk_neon(void *state, void *dst, const void *src,
 					unsigned int size);
 void crypto_aegis128_decrypt_chunk_neon(void *state, void *dst, const void *src,
 					unsigned int size);
+void crypto_aegis128_final_neon(void *state, void *tag_xor, uint64_t assoclen,
+				uint64_t cryptlen);
 
 int aegis128_have_aes_insn __ro_after_init;
 
@@ -25,6 +28,15 @@ bool crypto_aegis128_have_simd(void)
 	return IS_ENABLED(CONFIG_ARM64);
 }
 
+void crypto_aegis128_init_simd(union aegis_block *state,
+			       const union aegis_block *key,
+			       const u8 *iv)
+{
+	kernel_neon_begin();
+	crypto_aegis128_init_neon(state, key, iv);
+	kernel_neon_end();
+}
+
 void crypto_aegis128_update_simd(union aegis_block *state, const void *msg)
 {
 	kernel_neon_begin();
@@ -47,3 +59,12 @@ void crypto_aegis128_decrypt_chunk_simd(union aegis_block *state, u8 *dst,
 	crypto_aegis128_decrypt_chunk_neon(state, dst, src, size);
 	kernel_neon_end();
 }
+
+void crypto_aegis128_final_simd(union aegis_block *state,
+				union aegis_block *tag_xor,
+				u64 assoclen, u64 cryptlen)
+{
+	kernel_neon_begin();
+	crypto_aegis128_final_neon(state, tag_xor, assoclen, cryptlen);
+	kernel_neon_end();
+}