Provide an accelerated SHA-256 assembly implementation for aarch64.

This provides a SHA-256 assembly implementation that makes use of the ARM
Cryptographic Extension (CE), which is found on many arm64 CPUs. This gives
a performance gain of up to 7.5x on an Apple M2 (dependent on block size).
If an aarch64 machine does not have SHA2 support, then we'll fall back to
using the existing C implementation.

ok kettenis@ tb@

2 files changed, 9 insertions, 2 deletions

diff --git a/src/lib/libcrypto/arch/aarch64/Makefile.inc b/src/lib/libcrypto/arch/aarch64/Makefile.inc
index 20a634dc99..41198b069f 100644
--- a/src/lib/libcrypto/arch/aarch64/Makefile.inc
+++ b/src/lib/libcrypto/arch/aarch64/Makefile.inc
@@ -1,7 +1,10 @@
-# $OpenBSD: Makefile.inc,v 1.14 2024/11/08 13:34:24 jsing Exp $
+# $OpenBSD: Makefile.inc,v 1.15 2025/03/07 14:21:22 jsing Exp $
 
 # aarch64-specific libcrypto build rules
 
 SRCS += crypto_cpu_caps.c
 
+SRCS += sha256_aarch64.c
+SRCS += sha256_aarch64_ce.S
+
 AFLAGS+= -mmark-bti-property
diff --git a/src/lib/libcrypto/arch/aarch64/crypto_arch.h b/src/lib/libcrypto/arch/aarch64/crypto_arch.h
index b0188c498a..adc91cd19f 100644
--- a/src/lib/libcrypto/arch/aarch64/crypto_arch.h
+++ b/src/lib/libcrypto/arch/aarch64/crypto_arch.h
@@ -1,4 +1,4 @@
-/*      $OpenBSD: crypto_arch.h,v 1.2 2024/11/08 13:34:24 jsing Exp $ */
+/*      $OpenBSD: crypto_arch.h,v 1.3 2025/03/07 14:21:22 jsing Exp $ */
 /*
  * Copyright (c) 2024 Joel Sing <jsing@openbsd.org>
  *
@@ -33,4 +33,8 @@ extern uint64_t crypto_cpu_caps_aarch64;
 #define CRYPTO_CPU_CAPS_AARCH64_SHA512  (1ULL << 4)
 #define CRYPTO_CPU_CAPS_AARCH64_SHA3    (1ULL << 5)
 
+#ifndef OPENSSL_NO_ASM
+#define HAVE_SHA256_BLOCK_DATA_ORDER
+#endif
+
 #endif