mirror of https://github.com/wolfSSL/wolfssl.git
SHA-3, ARM64: add assembly support for crypto instructions
Add ability to compile ARM assembly from inline C code.pull/4833/head
Sean Parkinson
parent
59ea65bad3
commit
0042a2594c
38
configure.ac
38
configure.ac
|
|
@ -1687,31 +1687,38 @@ then
|
|||
fi
|
||||
|
||||
|
||||
ENABLED_ARMASM_INLINE="no"
|
||||
ENABLED_ARMASM_SHA3="no"
|
||||
# ARM Assembly
|
||||
# Both SHA3 and SHA512 instructions available with ARMV8.2-a
|
||||
AC_ARG_ENABLE([armasm],
|
||||
[AS_HELP_STRING([--enable-armasm],[Enable wolfSSL ARMv8 ASM support (default: disabled). Set to sha512-crypto to use SHA512 instructions with Aarch64 CPU.])],
|
||||
[AS_HELP_STRING([--enable-armasm],[Enable wolfSSL ARMv8 ASM support (default: disabled). Set to sha512-crypto or sha3-crypto to use SHA512 and SHA3 instructions with Aarch64 CPU.])],
|
||||
[ ENABLED_ARMASM=$enableval ],
|
||||
[ ENABLED_ARMASM=no ]
|
||||
)
|
||||
if test "$ENABLED_ARMASM" != "no" && test "$ENABLED_ASM" = "yes"
|
||||
then
|
||||
|
||||
for v in `echo $ENABLED_ARMASM | tr "," " "`
|
||||
do
|
||||
case $v in
|
||||
yes)
|
||||
;;
|
||||
sha512-crypto)
|
||||
inline)
|
||||
ENABLED_ARMASM_INLINE=yes
|
||||
;;
|
||||
sha512-crypto | sha3-crypto)
|
||||
case $host_cpu in
|
||||
*aarch64*)
|
||||
;;
|
||||
*)
|
||||
AC_MSG_ERROR([SHA512 instructions only available on Aarch64 CPU.])
|
||||
AC_MSG_ERROR([SHA512/SHA3 instructions only available on Aarch64 CPU.])
|
||||
break;;
|
||||
esac
|
||||
ENABLED_ARMASM_SHA512=yes
|
||||
ENABLED_ARMASM_SHA3=yes
|
||||
;;
|
||||
*)
|
||||
AC_MSG_ERROR([Invalid choice of ARM asm inclusions (yes, sha512-crypto): $ENABLED_ARMASM.])
|
||||
AC_MSG_ERROR([Invalid choice of ARM asm inclusions (yes, sha512-crypto, sha3-crypto): $ENABLED_ARMASM.])
|
||||
break;;
|
||||
esac
|
||||
done
|
||||
|
|
@ -1729,11 +1736,15 @@ then
|
|||
case $host_os in
|
||||
*darwin*)
|
||||
# All known Aarch64 Mac computers support SHA-512 instructions
|
||||
ENABLED_ARMASM_SHA512=yes
|
||||
ENABLED_ARMASM_SHA3=yes
|
||||
;;
|
||||
*)
|
||||
# +crypto needed for hardware acceleration
|
||||
AM_CPPFLAGS="$AM_CPPFLAGS -mcpu=generic+crypto"
|
||||
if test "$ENABLED_ARMASM_SHA3" = "yes"; then
|
||||
AM_CPPFLAGS="$AM_CPPFLAGS -march=armv8.2-a+crypto+sha3"
|
||||
else
|
||||
AM_CPPFLAGS="$AM_CPPFLAGS -mcpu=generic+crypto"
|
||||
fi
|
||||
;;
|
||||
esac
|
||||
# Include options.h
|
||||
|
|
@ -1764,9 +1775,9 @@ then
|
|||
esac
|
||||
fi
|
||||
|
||||
if test "$ENABLED_ARMASM_SHA512" = "yes"; then
|
||||
AM_CFLAGS="$AM_CFLAGS -DWOLFSSL_ARMASM_CRYPTO_SHA512"
|
||||
AM_CCASFLAGS="$AM_CCASFLAGS -DWOLFSSL_ARMASM_CRYPTO_SHA512"
|
||||
if test "$ENABLED_ARMASM_SHA3" = "yes"; then
|
||||
AM_CFLAGS="$AM_CFLAGS -DWOLFSSL_ARMASM_CRYPTO_SHA512 -DWOLFSSL_ARMASM_CRYPTO_SHA3"
|
||||
AM_CCASFLAGS="$AM_CCASFLAGS -DWOLFSSL_ARMASM_CRYPTO_SHA512 -DWOLFSSL_ARMASM_CRYPTO_SHA3"
|
||||
fi
|
||||
|
||||
# Xilinx hardened crypto
|
||||
|
|
@ -7389,6 +7400,7 @@ AM_CONDITIONAL([BUILD_SNIFFTEST],[ test "x$ENABLED_SNIFFTEST" = "xyes"])
|
|||
AM_CONDITIONAL([BUILD_AESGCM],[test "x$ENABLED_AESGCM" = "xyes" || test "x$ENABLED_USERSETTINGS" = "xyes"])
|
||||
AM_CONDITIONAL([BUILD_AESCCM],[test "x$ENABLED_AESCCM" = "xyes" || test "x$ENABLED_USERSETTINGS" = "xyes"])
|
||||
AM_CONDITIONAL([BUILD_ARMASM],[test "x$ENABLED_ARMASM" = "xyes"])
|
||||
AM_CONDITIONAL([BUILD_ARMASM_INLINE],[test "x$ENABLED_ARMASM_INLINE" = "xyes"])
|
||||
AM_CONDITIONAL([BUILD_XILINX],[test "x$ENABLED_XILINX" = "xyes"])
|
||||
AM_CONDITIONAL([BUILD_AESNI],[test "x$ENABLED_AESNI" = "xyes"])
|
||||
AM_CONDITIONAL([BUILD_INTELASM],[test "x$ENABLED_INTELASM" = "xyes"])
|
||||
|
|
@ -7897,8 +7909,12 @@ echo " * PKCS#12: $ENABLED_PKCS12"
|
|||
echo " * Cavium Nitrox: $ENABLED_CAVIUM"
|
||||
echo " * Cavium Octeon (Sync): $ENABLED_OCTEON_SYNC"
|
||||
echo " * Intel Quick Assist: $ENABLED_INTEL_QA"
|
||||
if test "$ENABLED_ARMASM_INLINE" = "yes"
|
||||
then
|
||||
ENABLED_ARMASM="inline C"
|
||||
fi
|
||||
echo " * ARM ASM: $ENABLED_ARMASM"
|
||||
echo " * ARM ASM SHA512 Crypto $ENABLED_ARMASM_SHA512"
|
||||
echo " * ARM ASM SHA512/SHA3 Crypto $ENABLED_ARMASM_SHA3"
|
||||
echo " * AES Key Wrap: $ENABLED_AESKEYWRAP"
|
||||
echo " * Write duplicate: $ENABLED_WRITEDUP"
|
||||
echo " * Xilinx Hardware Acc.: $ENABLED_XILINX"
|
||||
|
|
|
|||
|
|
@ -293,9 +293,14 @@ if BUILD_SHA512
|
|||
src_libwolfssl_la_SOURCES += wolfcrypt/src/sha512.c
|
||||
if BUILD_ARMASM
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-sha512.c
|
||||
if BUILD_ARMASM_INLINE
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-sha512-asm_c.c
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-32-sha512-asm_c.c
|
||||
else
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-sha512-asm.S
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-32-sha512-asm.S
|
||||
endif
|
||||
endif
|
||||
if BUILD_INTELASM
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/sha512_asm.S
|
||||
endif
|
||||
|
|
@ -303,6 +308,13 @@ endif
|
|||
|
||||
if BUILD_SHA3
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/sha3.c
|
||||
if BUILD_ARMASM
|
||||
if BUILD_ARMASM_INLINE
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-sha3-asm_c.c
|
||||
else
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-sha3-asm.S
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
|
||||
if BUILD_DH
|
||||
|
|
@ -479,8 +491,13 @@ if !BUILD_FIPS_CURRENT
|
|||
if BUILD_SHA512
|
||||
if BUILD_ARMASM
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-sha512.c
|
||||
if BUILD_ARMASM_INLINE
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-sha512-asm_c.c
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-32-sha512-asm_c.c
|
||||
else
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-sha512-asm.S
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-32-sha512-asm.S
|
||||
endif
|
||||
else
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/sha512.c
|
||||
if BUILD_INTELASM
|
||||
|
|
@ -493,6 +510,13 @@ endif !BUILD_FIPS_CURRENT
|
|||
if !BUILD_FIPS_CURRENT
|
||||
if BUILD_SHA3
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/sha3.c
|
||||
if BUILD_ARMASM
|
||||
if BUILD_ARMASM_INLINE
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-sha3-asm_c.c
|
||||
else
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-sha3-asm.S
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
endif !BUILD_FIPS_CURRENT
|
||||
|
||||
|
|
@ -653,8 +677,13 @@ if BUILD_INTELASM
|
|||
src_libwolfssl_la_SOURCES += wolfcrypt/src/fe_x25519_asm.S
|
||||
else
|
||||
if BUILD_ARMASM
|
||||
if BUILD_ARMASM_INLINE
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-32-curve25519_c.c
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-curve25519_c.c
|
||||
else
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-32-curve25519.S
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-curve25519.S
|
||||
endif
|
||||
else
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/fe_operations.c
|
||||
endif
|
||||
|
|
@ -672,7 +701,11 @@ if BUILD_INTELASM
|
|||
src_libwolfssl_la_SOURCES += wolfcrypt/src/fe_x25519_asm.S
|
||||
else
|
||||
if BUILD_ARMASM
|
||||
if BUILD_ARMASM_INLINE
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-curve25519_c.c
|
||||
else
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/port/arm/armv8-curve25519.S
|
||||
endif
|
||||
else
|
||||
src_libwolfssl_la_SOURCES += wolfcrypt/src/fe_operations.c
|
||||
endif
|
||||
|
|
|
|||
|
|
@ -62,10 +62,6 @@ EXTRA_DIST += wolfcrypt/src/port/ti/ti-aes.c \
|
|||
wolfcrypt/src/port/arm/armv8-aes.c \
|
||||
wolfcrypt/src/port/arm/armv8-sha256.c \
|
||||
wolfcrypt/src/port/arm/armv8-chacha.c \
|
||||
wolfcrypt/src/port/arm/armv8-curve25519.c \
|
||||
wolfcrypt/src/port/arm/armv8-32-curve25519.c \
|
||||
wolfcrypt/src/port/arm/armv8-sha512-asm.c \
|
||||
wolfcrypt/src/port/arm/armv8-32-sha512-asm.c \
|
||||
wolfcrypt/src/port/nxp/ksdk_port.c \
|
||||
wolfcrypt/src/port/nxp/dcp_port.c \
|
||||
wolfcrypt/src/port/nxp/se050_port.c \
|
||||
|
|
|
|||
|
|
@ -68,10 +68,10 @@ void fe_frombytes(fe out, const unsigned char* in)
|
|||
void fe_tobytes(unsigned char* out, const fe n)
|
||||
{
|
||||
__asm__ __volatile__ (
|
||||
"ldrd r2, r3, [%[in]]\n\t"
|
||||
"ldrd r12, lr, [%[in], #8]\n\t"
|
||||
"ldrd r4, r5, [%[in], #16]\n\t"
|
||||
"ldrd r6, r7, [%[in], #24]\n\t"
|
||||
"ldrd r2, r3, [%[n]]\n\t"
|
||||
"ldrd r12, lr, [%[n], #8]\n\t"
|
||||
"ldrd r4, r5, [%[n], #16]\n\t"
|
||||
"ldrd r6, r7, [%[n], #24]\n\t"
|
||||
"adds r8, r2, #19\n\t"
|
||||
"adcs r8, r3, #0\n\t"
|
||||
"adcs r8, r12, #0\n\t"
|
||||
|
|
@ -3861,23 +3861,26 @@ void fe_ge_to_p2(fe rx, fe ry, fe rz, const fe px, const fe py, const fe pz, con
|
|||
"str %[ry], [sp, #4]\n\t"
|
||||
"str %[rz], [sp, #8]\n\t"
|
||||
"str %[px], [sp, #12]\n\t"
|
||||
"ldr r2, [sp, #32]\n\t"
|
||||
"ldr r2, [sp, #28]\n\t"
|
||||
"ldr r1, [sp, #12]\n\t"
|
||||
"ldr r0, [sp]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #28]\n\t"
|
||||
"ldr r1, [sp, #24]\n\t"
|
||||
"ldr r2, [sp, #24]\n\t"
|
||||
"ldr r1, [sp, #20]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #32]\n\t"
|
||||
"ldr r1, [sp, #28]\n\t"
|
||||
"ldr r2, [sp, #28]\n\t"
|
||||
"ldr r1, [sp, #24]\n\t"
|
||||
"ldr r0, [sp, #8]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"add sp, sp, #16\n\t"
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [px] "+r" (px), [py] "+r" (py), [pz] "+r" (pz), [pt] "+r" (pt)
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [px] "+r" (px)
|
||||
:
|
||||
: "memory", "lr"
|
||||
);
|
||||
(void)py;
|
||||
(void)pz;
|
||||
(void)pt;
|
||||
}
|
||||
|
||||
void fe_ge_to_p3(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz, const fe pt)
|
||||
|
|
@ -3888,27 +3891,31 @@ void fe_ge_to_p3(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe
|
|||
"str %[ry], [sp, #4]\n\t"
|
||||
"str %[rz], [sp, #8]\n\t"
|
||||
"str %[rt], [sp, #12]\n\t"
|
||||
"ldr r2, [sp, #36]\n\t"
|
||||
"ldr r1, [sp, #24]\n\t"
|
||||
"ldr r0, [sp]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #32]\n\t"
|
||||
"ldr r1, [sp, #28]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #36]\n\t"
|
||||
"ldr r1, [sp, #32]\n\t"
|
||||
"ldr r0, [sp, #8]\n\t"
|
||||
"ldr r1, [sp, #20]\n\t"
|
||||
"ldr r0, [sp]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #28]\n\t"
|
||||
"ldr r1, [sp, #24]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #32]\n\t"
|
||||
"ldr r1, [sp, #28]\n\t"
|
||||
"ldr r0, [sp, #8]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #24]\n\t"
|
||||
"ldr r1, [sp, #20]\n\t"
|
||||
"ldr r0, [sp, #12]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"add sp, sp, #16\n\t"
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt), [px] "+r" (px), [py] "+r" (py), [pz] "+r" (pz), [pt] "+r" (pt)
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt)
|
||||
:
|
||||
: "memory", "lr"
|
||||
);
|
||||
(void)px;
|
||||
(void)py;
|
||||
(void)pz;
|
||||
(void)pt;
|
||||
}
|
||||
|
||||
void fe_ge_dbl(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz)
|
||||
|
|
@ -3919,15 +3926,15 @@ void fe_ge_dbl(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"str %[ry], [sp, #4]\n\t"
|
||||
"str %[rz], [sp, #8]\n\t"
|
||||
"str %[rt], [sp, #12]\n\t"
|
||||
"ldr r1, [sp, #88]\n\t"
|
||||
"ldr r1, [sp, #52]\n\t"
|
||||
"ldr r0, [sp]\n\t"
|
||||
"bl fe_sq\n\t"
|
||||
"ldr r1, [sp, #92]\n\t"
|
||||
"ldr r1, [sp, #56]\n\t"
|
||||
"ldr r0, [sp, #8]\n\t"
|
||||
"bl fe_sq\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"ldr r1, [sp, #88]\n\t"
|
||||
"ldr r2, [sp, #92]\n\t"
|
||||
"ldr r1, [sp, #52]\n\t"
|
||||
"ldr r2, [sp, #56]\n\t"
|
||||
/* Add */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -4114,7 +4121,7 @@ void fe_ge_dbl(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"strd r5, r6, [r0, #8]\n\t"
|
||||
"strd r7, r8, [r0, #16]\n\t"
|
||||
"strd r9, r10, [r0, #24]\n\t"
|
||||
"ldr r1, [sp, #96]\n\t"
|
||||
"ldr r1, [sp, #60]\n\t"
|
||||
"ldr r0, [sp, #12]\n\t"
|
||||
"bl fe_sq2\n\t"
|
||||
"ldr r0, [sp, #12]\n\t"
|
||||
|
|
@ -4159,10 +4166,13 @@ void fe_ge_dbl(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"strd r7, r8, [r0, #16]\n\t"
|
||||
"strd r9, r10, [r0, #24]\n\t"
|
||||
"add sp, sp, #16\n\t"
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt), [px] "+r" (px), [py] "+r" (py), [pz] "+r" (pz)
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt)
|
||||
:
|
||||
: "memory", "r12", "lr", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "r11"
|
||||
);
|
||||
(void)px;
|
||||
(void)py;
|
||||
(void)pz;
|
||||
}
|
||||
|
||||
void fe_ge_madd(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz, const fe pt, const fe qxy2d, const fe qyplusx, const fe qyminusx)
|
||||
|
|
@ -4174,8 +4184,8 @@ void fe_ge_madd(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe p
|
|||
"str %[rz], [sp, #8]\n\t"
|
||||
"str %[rt], [sp, #12]\n\t"
|
||||
"ldr r0, [sp]\n\t"
|
||||
"ldr r1, [sp, #108]\n\t"
|
||||
"ldr r2, [sp, #104]\n\t"
|
||||
"ldr r1, [sp, #72]\n\t"
|
||||
"ldr r2, [sp, #68]\n\t"
|
||||
/* Add */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -4216,8 +4226,8 @@ void fe_ge_madd(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe p
|
|||
"strd r7, r8, [r0, #16]\n\t"
|
||||
"strd r9, r10, [r0, #24]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"ldr r1, [sp, #108]\n\t"
|
||||
"ldr r2, [sp, #104]\n\t"
|
||||
"ldr r1, [sp, #72]\n\t"
|
||||
"ldr r2, [sp, #68]\n\t"
|
||||
/* Sub */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -4257,16 +4267,16 @@ void fe_ge_madd(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe p
|
|||
"strd r5, r6, [r0, #8]\n\t"
|
||||
"strd r7, r8, [r0, #16]\n\t"
|
||||
"strd r9, r10, [r0, #24]\n\t"
|
||||
"ldr r2, [sp, #124]\n\t"
|
||||
"ldr r2, [sp, #88]\n\t"
|
||||
"ldr r1, [sp]\n\t"
|
||||
"ldr r0, [sp, #8]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #128]\n\t"
|
||||
"ldr r2, [sp, #92]\n\t"
|
||||
"ldr r1, [sp, #4]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #116]\n\t"
|
||||
"ldr r1, [sp, #120]\n\t"
|
||||
"ldr r2, [sp, #80]\n\t"
|
||||
"ldr r1, [sp, #84]\n\t"
|
||||
"ldr r0, [sp, #12]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
|
|
@ -4372,7 +4382,7 @@ void fe_ge_madd(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe p
|
|||
"adc r10, r10, lr\n\t"
|
||||
"strd r9, r10, [r1, #24]\n\t"
|
||||
"ldr r0, [sp, #8]\n\t"
|
||||
"ldr r1, [sp, #112]\n\t"
|
||||
"ldr r1, [sp, #76]\n\t"
|
||||
/* Double */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -4506,10 +4516,14 @@ void fe_ge_madd(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe p
|
|||
"adc r10, r10, lr\n\t"
|
||||
"strd r9, r10, [r1, #24]\n\t"
|
||||
"add sp, sp, #32\n\t"
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt), [px] "+r" (px), [py] "+r" (py), [pz] "+r" (pz), [pt] "+r" (pt)
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt)
|
||||
:
|
||||
: "memory", "r12", "lr", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "r11"
|
||||
);
|
||||
(void)px;
|
||||
(void)py;
|
||||
(void)pz;
|
||||
(void)pt;
|
||||
(void)qxy2d;
|
||||
(void)qyplusx;
|
||||
(void)qyminusx;
|
||||
|
|
@ -4524,8 +4538,8 @@ void fe_ge_msub(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe p
|
|||
"str %[rz], [sp, #8]\n\t"
|
||||
"str %[rt], [sp, #12]\n\t"
|
||||
"ldr r0, [sp]\n\t"
|
||||
"ldr r1, [sp, #108]\n\t"
|
||||
"ldr r2, [sp, #104]\n\t"
|
||||
"ldr r1, [sp, #72]\n\t"
|
||||
"ldr r2, [sp, #68]\n\t"
|
||||
/* Add */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -4566,8 +4580,8 @@ void fe_ge_msub(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe p
|
|||
"strd r7, r8, [r0, #16]\n\t"
|
||||
"strd r9, r10, [r0, #24]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"ldr r1, [sp, #108]\n\t"
|
||||
"ldr r2, [sp, #104]\n\t"
|
||||
"ldr r1, [sp, #72]\n\t"
|
||||
"ldr r2, [sp, #68]\n\t"
|
||||
/* Sub */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -4607,16 +4621,16 @@ void fe_ge_msub(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe p
|
|||
"strd r5, r6, [r0, #8]\n\t"
|
||||
"strd r7, r8, [r0, #16]\n\t"
|
||||
"strd r9, r10, [r0, #24]\n\t"
|
||||
"ldr r2, [sp, #128]\n\t"
|
||||
"ldr r2, [sp, #92]\n\t"
|
||||
"ldr r1, [sp]\n\t"
|
||||
"ldr r0, [sp, #8]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #124]\n\t"
|
||||
"ldr r2, [sp, #88]\n\t"
|
||||
"ldr r1, [sp, #4]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #116]\n\t"
|
||||
"ldr r1, [sp, #120]\n\t"
|
||||
"ldr r2, [sp, #80]\n\t"
|
||||
"ldr r1, [sp, #84]\n\t"
|
||||
"ldr r0, [sp, #12]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
|
|
@ -4722,7 +4736,7 @@ void fe_ge_msub(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe p
|
|||
"adc r10, r10, lr\n\t"
|
||||
"strd r9, r10, [r1, #24]\n\t"
|
||||
"ldr r0, [sp, #8]\n\t"
|
||||
"ldr r1, [sp, #112]\n\t"
|
||||
"ldr r1, [sp, #76]\n\t"
|
||||
/* Double */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -4856,10 +4870,14 @@ void fe_ge_msub(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe p
|
|||
"adc r10, r10, lr\n\t"
|
||||
"strd r9, r10, [r1, #24]\n\t"
|
||||
"add sp, sp, #32\n\t"
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt), [px] "+r" (px), [py] "+r" (py), [pz] "+r" (pz), [pt] "+r" (pt)
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt)
|
||||
:
|
||||
: "memory", "r12", "lr", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "r11"
|
||||
);
|
||||
(void)px;
|
||||
(void)py;
|
||||
(void)pz;
|
||||
(void)pt;
|
||||
(void)qxy2d;
|
||||
(void)qyplusx;
|
||||
(void)qyminusx;
|
||||
|
|
@ -4874,8 +4892,8 @@ void fe_ge_add(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"str %[rz], [sp, #8]\n\t"
|
||||
"str %[rt], [sp, #12]\n\t"
|
||||
"ldr r0, [sp]\n\t"
|
||||
"ldr r1, [sp, #172]\n\t"
|
||||
"ldr r2, [sp, #168]\n\t"
|
||||
"ldr r1, [sp, #136]\n\t"
|
||||
"ldr r2, [sp, #132]\n\t"
|
||||
/* Add */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -4916,8 +4934,8 @@ void fe_ge_add(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"strd r7, r8, [r0, #16]\n\t"
|
||||
"strd r9, r10, [r0, #24]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"ldr r1, [sp, #172]\n\t"
|
||||
"ldr r2, [sp, #168]\n\t"
|
||||
"ldr r1, [sp, #136]\n\t"
|
||||
"ldr r2, [sp, #132]\n\t"
|
||||
/* Sub */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -4957,20 +4975,20 @@ void fe_ge_add(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"strd r5, r6, [r0, #8]\n\t"
|
||||
"strd r7, r8, [r0, #16]\n\t"
|
||||
"strd r9, r10, [r0, #24]\n\t"
|
||||
"ldr r2, [sp, #192]\n\t"
|
||||
"ldr r2, [sp, #156]\n\t"
|
||||
"ldr r1, [sp]\n\t"
|
||||
"ldr r0, [sp, #8]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #196]\n\t"
|
||||
"ldr r2, [sp, #160]\n\t"
|
||||
"ldr r1, [sp, #4]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #180]\n\t"
|
||||
"ldr r1, [sp, #188]\n\t"
|
||||
"ldr r2, [sp, #144]\n\t"
|
||||
"ldr r1, [sp, #152]\n\t"
|
||||
"ldr r0, [sp, #12]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #184]\n\t"
|
||||
"ldr r1, [sp, #176]\n\t"
|
||||
"ldr r2, [sp, #148]\n\t"
|
||||
"ldr r1, [sp, #140]\n\t"
|
||||
"ldr r0, [sp]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"add r0, sp, #16\n\t"
|
||||
|
|
@ -5211,10 +5229,14 @@ void fe_ge_add(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"adc r10, r10, lr\n\t"
|
||||
"strd r9, r10, [r1, #24]\n\t"
|
||||
"add sp, sp, #0x60\n\t"
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt), [px] "+r" (px), [py] "+r" (py), [pz] "+r" (pz), [pt] "+r" (pt)
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt)
|
||||
:
|
||||
: "memory", "r12", "lr", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "r11"
|
||||
);
|
||||
(void)px;
|
||||
(void)py;
|
||||
(void)pz;
|
||||
(void)pt;
|
||||
(void)qz;
|
||||
(void)qt2d;
|
||||
(void)qyplusx;
|
||||
|
|
@ -5230,8 +5252,8 @@ void fe_ge_sub(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"str %[rz], [sp, #8]\n\t"
|
||||
"str %[rt], [sp, #12]\n\t"
|
||||
"ldr r0, [sp]\n\t"
|
||||
"ldr r1, [sp, #172]\n\t"
|
||||
"ldr r2, [sp, #168]\n\t"
|
||||
"ldr r1, [sp, #136]\n\t"
|
||||
"ldr r2, [sp, #132]\n\t"
|
||||
/* Add */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -5272,8 +5294,8 @@ void fe_ge_sub(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"strd r7, r8, [r0, #16]\n\t"
|
||||
"strd r9, r10, [r0, #24]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"ldr r1, [sp, #172]\n\t"
|
||||
"ldr r2, [sp, #168]\n\t"
|
||||
"ldr r1, [sp, #136]\n\t"
|
||||
"ldr r2, [sp, #132]\n\t"
|
||||
/* Sub */
|
||||
"ldrd %[rt], r4, [r1]\n\t"
|
||||
"ldrd r5, r6, [r1, #8]\n\t"
|
||||
|
|
@ -5313,20 +5335,20 @@ void fe_ge_sub(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"strd r5, r6, [r0, #8]\n\t"
|
||||
"strd r7, r8, [r0, #16]\n\t"
|
||||
"strd r9, r10, [r0, #24]\n\t"
|
||||
"ldr r2, [sp, #196]\n\t"
|
||||
"ldr r2, [sp, #160]\n\t"
|
||||
"ldr r1, [sp]\n\t"
|
||||
"ldr r0, [sp, #8]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #192]\n\t"
|
||||
"ldr r2, [sp, #156]\n\t"
|
||||
"ldr r1, [sp, #4]\n\t"
|
||||
"ldr r0, [sp, #4]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #180]\n\t"
|
||||
"ldr r1, [sp, #188]\n\t"
|
||||
"ldr r2, [sp, #144]\n\t"
|
||||
"ldr r1, [sp, #152]\n\t"
|
||||
"ldr r0, [sp, #12]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"ldr r2, [sp, #184]\n\t"
|
||||
"ldr r1, [sp, #176]\n\t"
|
||||
"ldr r2, [sp, #148]\n\t"
|
||||
"ldr r1, [sp, #140]\n\t"
|
||||
"ldr r0, [sp]\n\t"
|
||||
"bl fe_mul\n\t"
|
||||
"add r0, sp, #16\n\t"
|
||||
|
|
@ -5567,10 +5589,14 @@ void fe_ge_sub(fe rx, fe ry, fe rz, fe rt, const fe px, const fe py, const fe pz
|
|||
"adc r10, r10, lr\n\t"
|
||||
"strd r9, r10, [r1, #24]\n\t"
|
||||
"add sp, sp, #0x60\n\t"
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt), [px] "+r" (px), [py] "+r" (py), [pz] "+r" (pz), [pt] "+r" (pt)
|
||||
: [rx] "+r" (rx), [ry] "+r" (ry), [rz] "+r" (rz), [rt] "+r" (rt)
|
||||
:
|
||||
: "memory", "r12", "lr", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "r11"
|
||||
);
|
||||
(void)px;
|
||||
(void)py;
|
||||
(void)pz;
|
||||
(void)pt;
|
||||
(void)qz;
|
||||
(void)qt2d;
|
||||
(void)qyplusx;
|
||||
|
|
@ -120,6 +120,7 @@ static const uint64_t L_SHA512_transform_len_k[] = {
|
|||
0x6c44198c4a475817UL,
|
||||
};
|
||||
|
||||
void Transform_Sha512_Len();
|
||||
void Transform_Sha512_Len(wc_Sha512* sha512, const byte* data, word32 len)
|
||||
{
|
||||
__asm__ __volatile__ (
|
||||
|
|
@ -3650,6 +3651,7 @@ static const uint64_t L_SHA512_transform_neon_len_k[] = {
|
|||
0x6c44198c4a475817UL,
|
||||
};
|
||||
|
||||
void Transform_Sha512_Len(wc_Sha512* sha512, const byte* data, word32 len);
|
||||
void Transform_Sha512_Len(wc_Sha512* sha512, const byte* data, word32 len)
|
||||
{
|
||||
__asm__ __volatile__ (
|
||||
|
|
@ -4773,7 +4775,7 @@ void Transform_Sha512_Len(wc_Sha512* sha512, const byte* data, word32 len)
|
|||
"subs %[len], %[len], #0x80\n\t"
|
||||
"bne L_sha512_len_neon_begin_%=\n\t"
|
||||
: [sha512] "+r" (sha512), [data] "+r" (data), [len] "+r" (len)
|
||||
: [L_SHA512_transform_len_k] "r" (L_SHA512_transform_len_k), [L_SHA512_transform_neon_len_k] "r" (L_SHA512_transform_neon_len_k)
|
||||
: [L_SHA512_transform_neon_len_k] "r" (L_SHA512_transform_neon_len_k)
|
||||
: "memory", "r3", "r12", "d0", "d1", "d2", "d3", "d4", "d5", "d6", "d7", "d8", "d9", "d10", "d11", "d12", "d13", "d14", "d15", "q8", "q9", "q10", "q11", "q12", "q13", "q14", "q15"
|
||||
);
|
||||
}
|
||||
|
|
@ -0,0 +1,215 @@
|
|||
/* armv8-sha3-asm
|
||||
*
|
||||
* Copyright (C) 2006-2021 wolfSSL Inc.
|
||||
*
|
||||
* This file is part of wolfSSL.
|
||||
*
|
||||
* wolfSSL is free software; you can redistribute it and/or modify
|
||||
* it under the terms of the GNU General Public License as published by
|
||||
* the Free Software Foundation; either version 2 of the License, or
|
||||
* (at your option) any later version.
|
||||
*
|
||||
* wolfSSL is distributed in the hope that it will be useful,
|
||||
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
||||
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
||||
* GNU General Public License for more details.
|
||||
*
|
||||
* You should have received a copy of the GNU General Public License
|
||||
* along with this program; if not, write to the Free Software
|
||||
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1335, USA
|
||||
*/
|
||||
|
||||
#ifdef HAVE_CONFIG_H
|
||||
#include <config.h>
|
||||
#endif /* HAVE_CONFIG_H */
|
||||
#include <wolfssl/wolfcrypt/settings.h>
|
||||
|
||||
/* Generated using (from wolfssl):
|
||||
* cd ../scripts
|
||||
* ruby ./sha3/sha3.rb arm64 ../wolfssl/wolfcrypt/src/port/arm/armv8-sha3-asm.S
|
||||
*/
|
||||
#ifdef WOLFSSL_ARMASM
|
||||
#ifdef __aarch64__
|
||||
#ifdef WOLFSSL_SHA3
|
||||
#ifdef WOLFSSL_ARMASM_CRYPTO_SHA3
|
||||
#ifndef __APPLE__
|
||||
.text
|
||||
.type L_SHA3_transform_crypto_r, %object
|
||||
.section .rodata
|
||||
.size L_SHA3_transform_crypto_r, 192
|
||||
#else
|
||||
.section __DATA,__data
|
||||
#endif /* __APPLE__ */
|
||||
#ifndef __APPLE__
|
||||
.align 3
|
||||
#else
|
||||
.p2align 3
|
||||
#endif /* __APPLE__ */
|
||||
L_SHA3_transform_crypto_r:
|
||||
.xword 0x1
|
||||
.xword 0x8082
|
||||
.xword 0x800000000000808a
|
||||
.xword 0x8000000080008000
|
||||
.xword 0x808b
|
||||
.xword 0x80000001
|
||||
.xword 0x8000000080008081
|
||||
.xword 0x8000000000008009
|
||||
.xword 0x8a
|
||||
.xword 0x88
|
||||
.xword 0x80008009
|
||||
.xword 0x8000000a
|
||||
.xword 0x8000808b
|
||||
.xword 0x800000000000008b
|
||||
.xword 0x8000000000008089
|
||||
.xword 0x8000000000008003
|
||||
.xword 0x8000000000008002
|
||||
.xword 0x8000000000000080
|
||||
.xword 0x800a
|
||||
.xword 0x800000008000000a
|
||||
.xword 0x8000000080008081
|
||||
.xword 0x8000000000008080
|
||||
.xword 0x80000001
|
||||
.xword 0x8000000080008008
|
||||
#ifndef __APPLE__
|
||||
.text
|
||||
.globl BlockSha3
|
||||
.type BlockSha3,@function
|
||||
.align 2
|
||||
BlockSha3:
|
||||
#else
|
||||
.section __TEXT,__text
|
||||
.globl _BlockSha3
|
||||
.p2align 2
|
||||
_BlockSha3:
|
||||
#endif /* __APPLE__ */
|
||||
stp x29, x30, [sp, #-80]!
|
||||
add x29, sp, #0
|
||||
stp d8, d9, [x29, #16]
|
||||
stp d10, d11, [x29, #32]
|
||||
stp d12, d13, [x29, #48]
|
||||
stp d14, d15, [x29, #64]
|
||||
#ifdef __APPLE__
|
||||
.arch_extension sha3
|
||||
#endif /* __APPLE__ */
|
||||
#ifndef __APPLE__
|
||||
adrp x1, L_SHA3_transform_crypto_r
|
||||
add x1, x1, :lo12:L_SHA3_transform_crypto_r
|
||||
#else
|
||||
adrp x1, L_SHA3_transform_crypto_r@PAGE
|
||||
add x1, x1, :lo12:L_SHA3_transform_crypto_r@PAGEOFF
|
||||
#endif /* __APPLE__ */
|
||||
ld4 {v0.d, v1.d, v2.d, v3.d}[0], [x0], #32
|
||||
ld4 {v4.d, v5.d, v6.d, v7.d}[0], [x0], #32
|
||||
ld4 {v8.d, v9.d, v10.d, v11.d}[0], [x0], #32
|
||||
ld4 {v12.d, v13.d, v14.d, v15.d}[0], [x0], #32
|
||||
ld4 {v16.d, v17.d, v18.d, v19.d}[0], [x0], #32
|
||||
ld4 {v20.d, v21.d, v22.d, v23.d}[0], [x0], #32
|
||||
ld1 {v24.1d}, [x0]
|
||||
sub x0, x0, #0xc0
|
||||
mov x2, #24
|
||||
# Start of 24 rounds
|
||||
L_sha3_crypto_begin:
|
||||
# Col Mix
|
||||
eor3 v31.16b, v0.16b, v5.16b, v10.16b
|
||||
eor3 v27.16b, v1.16b, v6.16b, v11.16b
|
||||
eor3 v28.16b, v2.16b, v7.16b, v12.16b
|
||||
eor3 v29.16b, v3.16b, v8.16b, v13.16b
|
||||
eor3 v30.16b, v4.16b, v9.16b, v14.16b
|
||||
eor3 v31.16b, v31.16b, v15.16b, v20.16b
|
||||
eor3 v27.16b, v27.16b, v16.16b, v21.16b
|
||||
eor3 v28.16b, v28.16b, v17.16b, v22.16b
|
||||
eor3 v29.16b, v29.16b, v18.16b, v23.16b
|
||||
eor3 v30.16b, v30.16b, v19.16b, v24.16b
|
||||
rax1 v25.2d, v30.2d, v27.2d
|
||||
rax1 v26.2d, v31.2d, v28.2d
|
||||
rax1 v27.2d, v27.2d, v29.2d
|
||||
rax1 v28.2d, v28.2d, v30.2d
|
||||
rax1 v29.2d, v29.2d, v31.2d
|
||||
eor v0.16b, v0.16b, v25.16b
|
||||
xar v30.2d, v1.2d, v26.2d, #63
|
||||
xar v1.2d, v6.2d, v26.2d, #20
|
||||
xar v6.2d, v9.2d, v29.2d, #44
|
||||
xar v9.2d, v22.2d, v27.2d, #3
|
||||
xar v22.2d, v14.2d, v29.2d, #25
|
||||
xar v14.2d, v20.2d, v25.2d, #46
|
||||
xar v20.2d, v2.2d, v27.2d, #2
|
||||
xar v2.2d, v12.2d, v27.2d, #21
|
||||
xar v12.2d, v13.2d, v28.2d, #39
|
||||
xar v13.2d, v19.2d, v29.2d, #56
|
||||
xar v19.2d, v23.2d, v28.2d, #8
|
||||
xar v23.2d, v15.2d, v25.2d, #23
|
||||
xar v15.2d, v4.2d, v29.2d, #37
|
||||
xar v4.2d, v24.2d, v29.2d, #50
|
||||
xar v24.2d, v21.2d, v26.2d, #62
|
||||
xar v21.2d, v8.2d, v28.2d, #9
|
||||
xar v8.2d, v16.2d, v26.2d, #19
|
||||
xar v16.2d, v5.2d, v25.2d, #28
|
||||
xar v5.2d, v3.2d, v28.2d, #36
|
||||
xar v3.2d, v18.2d, v28.2d, #43
|
||||
xar v18.2d, v17.2d, v27.2d, #49
|
||||
xar v17.2d, v11.2d, v26.2d, #54
|
||||
xar v11.2d, v7.2d, v27.2d, #58
|
||||
xar v7.2d, v10.2d, v25.2d, #61
|
||||
# Row Mix
|
||||
mov v25.16b, v0.16b
|
||||
mov v26.16b, v1.16b
|
||||
bcax v0.16b, v25.16b, v2.16b, v26.16b
|
||||
bcax v1.16b, v26.16b, v3.16b, v2.16b
|
||||
bcax v2.16b, v2.16b, v4.16b, v3.16b
|
||||
bcax v3.16b, v3.16b, v25.16b, v4.16b
|
||||
bcax v4.16b, v4.16b, v26.16b, v25.16b
|
||||
mov v25.16b, v5.16b
|
||||
mov v26.16b, v6.16b
|
||||
bcax v5.16b, v25.16b, v7.16b, v26.16b
|
||||
bcax v6.16b, v26.16b, v8.16b, v7.16b
|
||||
bcax v7.16b, v7.16b, v9.16b, v8.16b
|
||||
bcax v8.16b, v8.16b, v25.16b, v9.16b
|
||||
bcax v9.16b, v9.16b, v26.16b, v25.16b
|
||||
mov v26.16b, v11.16b
|
||||
bcax v10.16b, v30.16b, v12.16b, v26.16b
|
||||
bcax v11.16b, v26.16b, v13.16b, v12.16b
|
||||
bcax v12.16b, v12.16b, v14.16b, v13.16b
|
||||
bcax v13.16b, v13.16b, v30.16b, v14.16b
|
||||
bcax v14.16b, v14.16b, v26.16b, v30.16b
|
||||
mov v25.16b, v15.16b
|
||||
mov v26.16b, v16.16b
|
||||
bcax v15.16b, v25.16b, v17.16b, v26.16b
|
||||
bcax v16.16b, v26.16b, v18.16b, v17.16b
|
||||
bcax v17.16b, v17.16b, v19.16b, v18.16b
|
||||
bcax v18.16b, v18.16b, v25.16b, v19.16b
|
||||
bcax v19.16b, v19.16b, v26.16b, v25.16b
|
||||
mov v25.16b, v20.16b
|
||||
mov v26.16b, v21.16b
|
||||
bcax v20.16b, v25.16b, v22.16b, v26.16b
|
||||
bcax v21.16b, v26.16b, v23.16b, v22.16b
|
||||
bcax v22.16b, v22.16b, v24.16b, v23.16b
|
||||
bcax v23.16b, v23.16b, v25.16b, v24.16b
|
||||
bcax v24.16b, v24.16b, v26.16b, v25.16b
|
||||
ld1r {v30.2d}, [x1], #8
|
||||
subs x2, x2, #1
|
||||
eor v0.16b, v0.16b, v30.16b
|
||||
bne L_sha3_crypto_begin
|
||||
st4 {v0.d, v1.d, v2.d, v3.d}[0], [x0], #32
|
||||
st4 {v4.d, v5.d, v6.d, v7.d}[0], [x0], #32
|
||||
st4 {v8.d, v9.d, v10.d, v11.d}[0], [x0], #32
|
||||
st4 {v12.d, v13.d, v14.d, v15.d}[0], [x0], #32
|
||||
st4 {v16.d, v17.d, v18.d, v19.d}[0], [x0], #32
|
||||
st4 {v20.d, v21.d, v22.d, v23.d}[0], [x0], #32
|
||||
st1 {v24.1d}, [x0]
|
||||
ldp d8, d9, [x29, #16]
|
||||
ldp d10, d11, [x29, #32]
|
||||
ldp d12, d13, [x29, #48]
|
||||
ldp d14, d15, [x29, #64]
|
||||
ldp x29, x30, [sp], #0x50
|
||||
ret
|
||||
#ifndef __APPLE__
|
||||
.size BlockSha3,.-BlockSha3
|
||||
#endif /* __APPLE__ */
|
||||
#endif /* WOLFSSL_ARMASM_CRYPTO_SHA3 */
|
||||
#endif /* WOLFSSL_SHA3 */
|
||||
#endif /* __aarch64__ */
|
||||
#endif /* WOLFSSL_ARMASM */
|
||||
|
||||
#if defined(__linux__) && defined(__ELF__)
|
||||
.section .note.GNU-stack,"",%progbits
|
||||
#endif
|
||||
|
|
@ -0,0 +1,185 @@
|
|||
/* armv8-sha3-asm
|
||||
*
|
||||
* Copyright (C) 2006-2021 wolfSSL Inc.
|
||||
*
|
||||
* This file is part of wolfSSL.
|
||||
*
|
||||
* wolfSSL is free software; you can redistribute it and/or modify
|
||||
* it under the terms of the GNU General Public License as published by
|
||||
* the Free Software Foundation; either version 2 of the License, or
|
||||
* (at your option) any later version.
|
||||
*
|
||||
* wolfSSL is distributed in the hope that it will be useful,
|
||||
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
||||
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
||||
* GNU General Public License for more details.
|
||||
*
|
||||
* You should have received a copy of the GNU General Public License
|
||||
* along with this program; if not, write to the Free Software
|
||||
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1335, USA
|
||||
*/
|
||||
|
||||
#ifdef HAVE_CONFIG_H
|
||||
#include <config.h>
|
||||
#endif /* HAVE_CONFIG_H */
|
||||
#include <wolfssl/wolfcrypt/settings.h>
|
||||
|
||||
/* Generated using (from wolfssl):
|
||||
* cd ../scripts
|
||||
* ruby ./sha3/sha3.rb arm64 ../wolfssl/wolfcrypt/src/port/arm/armv8-sha3-asm.c
|
||||
*/
|
||||
#ifdef WOLFSSL_ARMASM
|
||||
#ifdef __aarch64__
|
||||
#include <wolfssl/wolfcrypt/sha3.h>
|
||||
|
||||
#ifdef WOLFSSL_SHA3
|
||||
#ifdef WOLFSSL_ARMASM_CRYPTO_SHA3
|
||||
static const uint64_t L_SHA3_transform_crypto_r[] = {
|
||||
0x1UL,
|
||||
0x8082UL,
|
||||
0x800000000000808aUL,
|
||||
0x8000000080008000UL,
|
||||
0x808bUL,
|
||||
0x80000001UL,
|
||||
0x8000000080008081UL,
|
||||
0x8000000000008009UL,
|
||||
0x8aUL,
|
||||
0x88UL,
|
||||
0x80008009UL,
|
||||
0x8000000aUL,
|
||||
0x8000808bUL,
|
||||
0x800000000000008bUL,
|
||||
0x8000000000008089UL,
|
||||
0x8000000000008003UL,
|
||||
0x8000000000008002UL,
|
||||
0x8000000000000080UL,
|
||||
0x800aUL,
|
||||
0x800000008000000aUL,
|
||||
0x8000000080008081UL,
|
||||
0x8000000000008080UL,
|
||||
0x80000001UL,
|
||||
0x8000000080008008UL,
|
||||
};
|
||||
|
||||
void BlockSha3(unsigned long* state)
|
||||
{
|
||||
__asm__ __volatile__ (
|
||||
#ifdef __APPLE__
|
||||
".arch_extension sha3\n\t"
|
||||
#endif /* __APPLE__ */
|
||||
#ifndef __APPLE__
|
||||
"adrp x1, %[L_SHA3_transform_crypto_r]\n\t"
|
||||
"add x1, x1, :lo12:%[L_SHA3_transform_crypto_r]\n\t"
|
||||
#else
|
||||
"adrp x1, %[L_SHA3_transform_crypto_r]@PAGE\n\t"
|
||||
"add x1, x1, %[L_SHA3_transform_crypto_r]@PAGEOFF\n\t"
|
||||
#endif /* __APPLE__ */
|
||||
"ld4 {v0.d, v1.d, v2.d, v3.d}[0], [%x[state]], #32\n\t"
|
||||
"ld4 {v4.d, v5.d, v6.d, v7.d}[0], [%x[state]], #32\n\t"
|
||||
"ld4 {v8.d, v9.d, v10.d, v11.d}[0], [%x[state]], #32\n\t"
|
||||
"ld4 {v12.d, v13.d, v14.d, v15.d}[0], [%x[state]], #32\n\t"
|
||||
"ld4 {v16.d, v17.d, v18.d, v19.d}[0], [%x[state]], #32\n\t"
|
||||
"ld4 {v20.d, v21.d, v22.d, v23.d}[0], [%x[state]], #32\n\t"
|
||||
"ld1 {v24.1d}, [%x[state]]\n\t"
|
||||
"sub %x[state], %x[state], #0xc0\n\t"
|
||||
"mov x2, #24\n\t"
|
||||
/* Start of 24 rounds */
|
||||
"\n"
|
||||
"L_sha3_crypto_begin_%=: \n\t"
|
||||
/* Col Mix */
|
||||
"eor3 v31.16b, v0.16b, v5.16b, v10.16b\n\t"
|
||||
"eor3 v27.16b, v1.16b, v6.16b, v11.16b\n\t"
|
||||
"eor3 v28.16b, v2.16b, v7.16b, v12.16b\n\t"
|
||||
"eor3 v29.16b, v3.16b, v8.16b, v13.16b\n\t"
|
||||
"eor3 v30.16b, v4.16b, v9.16b, v14.16b\n\t"
|
||||
"eor3 v31.16b, v31.16b, v15.16b, v20.16b\n\t"
|
||||
"eor3 v27.16b, v27.16b, v16.16b, v21.16b\n\t"
|
||||
"eor3 v28.16b, v28.16b, v17.16b, v22.16b\n\t"
|
||||
"eor3 v29.16b, v29.16b, v18.16b, v23.16b\n\t"
|
||||
"eor3 v30.16b, v30.16b, v19.16b, v24.16b\n\t"
|
||||
"rax1 v25.2d, v30.2d, v27.2d\n\t"
|
||||
"rax1 v26.2d, v31.2d, v28.2d\n\t"
|
||||
"rax1 v27.2d, v27.2d, v29.2d\n\t"
|
||||
"rax1 v28.2d, v28.2d, v30.2d\n\t"
|
||||
"rax1 v29.2d, v29.2d, v31.2d\n\t"
|
||||
"eor v0.16b, v0.16b, v25.16b\n\t"
|
||||
"xar v30.2d, v1.2d, v26.2d, #63\n\t"
|
||||
"xar v1.2d, v6.2d, v26.2d, #20\n\t"
|
||||
"xar v6.2d, v9.2d, v29.2d, #44\n\t"
|
||||
"xar v9.2d, v22.2d, v27.2d, #3\n\t"
|
||||
"xar v22.2d, v14.2d, v29.2d, #25\n\t"
|
||||
"xar v14.2d, v20.2d, v25.2d, #46\n\t"
|
||||
"xar v20.2d, v2.2d, v27.2d, #2\n\t"
|
||||
"xar v2.2d, v12.2d, v27.2d, #21\n\t"
|
||||
"xar v12.2d, v13.2d, v28.2d, #39\n\t"
|
||||
"xar v13.2d, v19.2d, v29.2d, #56\n\t"
|
||||
"xar v19.2d, v23.2d, v28.2d, #8\n\t"
|
||||
"xar v23.2d, v15.2d, v25.2d, #23\n\t"
|
||||
"xar v15.2d, v4.2d, v29.2d, #37\n\t"
|
||||
"xar v4.2d, v24.2d, v29.2d, #50\n\t"
|
||||
"xar v24.2d, v21.2d, v26.2d, #62\n\t"
|
||||
"xar v21.2d, v8.2d, v28.2d, #9\n\t"
|
||||
"xar v8.2d, v16.2d, v26.2d, #19\n\t"
|
||||
"xar v16.2d, v5.2d, v25.2d, #28\n\t"
|
||||
"xar v5.2d, v3.2d, v28.2d, #36\n\t"
|
||||
"xar v3.2d, v18.2d, v28.2d, #43\n\t"
|
||||
"xar v18.2d, v17.2d, v27.2d, #49\n\t"
|
||||
"xar v17.2d, v11.2d, v26.2d, #54\n\t"
|
||||
"xar v11.2d, v7.2d, v27.2d, #58\n\t"
|
||||
"xar v7.2d, v10.2d, v25.2d, #61\n\t"
|
||||
/* Row Mix */
|
||||
"mov v25.16b, v0.16b\n\t"
|
||||
"mov v26.16b, v1.16b\n\t"
|
||||
"bcax v0.16b, v25.16b, v2.16b, v26.16b\n\t"
|
||||
"bcax v1.16b, v26.16b, v3.16b, v2.16b\n\t"
|
||||
"bcax v2.16b, v2.16b, v4.16b, v3.16b\n\t"
|
||||
"bcax v3.16b, v3.16b, v25.16b, v4.16b\n\t"
|
||||
"bcax v4.16b, v4.16b, v26.16b, v25.16b\n\t"
|
||||
"mov v25.16b, v5.16b\n\t"
|
||||
"mov v26.16b, v6.16b\n\t"
|
||||
"bcax v5.16b, v25.16b, v7.16b, v26.16b\n\t"
|
||||
"bcax v6.16b, v26.16b, v8.16b, v7.16b\n\t"
|
||||
"bcax v7.16b, v7.16b, v9.16b, v8.16b\n\t"
|
||||
"bcax v8.16b, v8.16b, v25.16b, v9.16b\n\t"
|
||||
"bcax v9.16b, v9.16b, v26.16b, v25.16b\n\t"
|
||||
"mov v26.16b, v11.16b\n\t"
|
||||
"bcax v10.16b, v30.16b, v12.16b, v26.16b\n\t"
|
||||
"bcax v11.16b, v26.16b, v13.16b, v12.16b\n\t"
|
||||
"bcax v12.16b, v12.16b, v14.16b, v13.16b\n\t"
|
||||
"bcax v13.16b, v13.16b, v30.16b, v14.16b\n\t"
|
||||
"bcax v14.16b, v14.16b, v26.16b, v30.16b\n\t"
|
||||
"mov v25.16b, v15.16b\n\t"
|
||||
"mov v26.16b, v16.16b\n\t"
|
||||
"bcax v15.16b, v25.16b, v17.16b, v26.16b\n\t"
|
||||
"bcax v16.16b, v26.16b, v18.16b, v17.16b\n\t"
|
||||
"bcax v17.16b, v17.16b, v19.16b, v18.16b\n\t"
|
||||
"bcax v18.16b, v18.16b, v25.16b, v19.16b\n\t"
|
||||
"bcax v19.16b, v19.16b, v26.16b, v25.16b\n\t"
|
||||
"mov v25.16b, v20.16b\n\t"
|
||||
"mov v26.16b, v21.16b\n\t"
|
||||
"bcax v20.16b, v25.16b, v22.16b, v26.16b\n\t"
|
||||
"bcax v21.16b, v26.16b, v23.16b, v22.16b\n\t"
|
||||
"bcax v22.16b, v22.16b, v24.16b, v23.16b\n\t"
|
||||
"bcax v23.16b, v23.16b, v25.16b, v24.16b\n\t"
|
||||
"bcax v24.16b, v24.16b, v26.16b, v25.16b\n\t"
|
||||
"ld1r {v30.2d}, [x1], #8\n\t"
|
||||
"subs x2, x2, #1\n\t"
|
||||
"eor v0.16b, v0.16b, v30.16b\n\t"
|
||||
"bne L_sha3_crypto_begin_%=\n\t"
|
||||
"st4 {v0.d, v1.d, v2.d, v3.d}[0], [%x[state]], #32\n\t"
|
||||
"st4 {v4.d, v5.d, v6.d, v7.d}[0], [%x[state]], #32\n\t"
|
||||
"st4 {v8.d, v9.d, v10.d, v11.d}[0], [%x[state]], #32\n\t"
|
||||
"st4 {v12.d, v13.d, v14.d, v15.d}[0], [%x[state]], #32\n\t"
|
||||
"st4 {v16.d, v17.d, v18.d, v19.d}[0], [%x[state]], #32\n\t"
|
||||
"st4 {v20.d, v21.d, v22.d, v23.d}[0], [%x[state]], #32\n\t"
|
||||
"st1 {v24.1d}, [%x[state]]\n\t"
|
||||
: [state] "+r" (state)
|
||||
: [L_SHA3_transform_crypto_r] "S" (L_SHA3_transform_crypto_r)
|
||||
: "memory", "x1", "x2", "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9", "v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17", "v18", "v19", "v20", "v21", "v22", "v23", "v24", "v25", "v26", "v27", "v28", "v29", "v30", "v31"
|
||||
);
|
||||
}
|
||||
|
||||
#endif /* WOLFSSL_ARMASM_CRYPTO_SHA3 */
|
||||
#endif /* WOLFSSL_SHA3 */
|
||||
#endif /* __aarch64__ */
|
||||
#endif /* WOLFSSL_ARMASM */
|
||||
|
|
@ -1201,20 +1201,12 @@ Transform_Sha512_Len_crypto:
|
|||
.p2align 2
|
||||
_Transform_Sha512_Len_crypto:
|
||||
#endif /* __APPLE__ */
|
||||
stp x29, x30, [sp, #-208]!
|
||||
stp x29, x30, [sp, #-80]!
|
||||
add x29, sp, #0
|
||||
stp d8, d9, [x29, #16]
|
||||
stp d10, d11, [x29, #32]
|
||||
stp d12, d13, [x29, #48]
|
||||
stp d14, d15, [x29, #64]
|
||||
stp d16, d17, [x29, #80]
|
||||
stp d18, d19, [x29, #96]
|
||||
stp d20, d21, [x29, #112]
|
||||
stp d22, d23, [x29, #128]
|
||||
stp d24, d25, [x29, #144]
|
||||
stp d26, d27, [x29, #160]
|
||||
stp d28, d29, [x29, #176]
|
||||
stp d30, d31, [x29, #192]
|
||||
#ifdef __APPLE__
|
||||
.arch_extension sha3
|
||||
#endif /* __APPLE__ */
|
||||
|
|
@ -1734,15 +1726,7 @@ L_sha512_len_crypto_begin:
|
|||
ldp d10, d11, [x29, #32]
|
||||
ldp d12, d13, [x29, #48]
|
||||
ldp d14, d15, [x29, #64]
|
||||
ldp d16, d17, [x29, #80]
|
||||
ldp d18, d19, [x29, #96]
|
||||
ldp d20, d21, [x29, #112]
|
||||
ldp d22, d23, [x29, #128]
|
||||
ldp d24, d25, [x29, #144]
|
||||
ldp d26, d27, [x29, #160]
|
||||
ldp d28, d29, [x29, #176]
|
||||
ldp d30, d31, [x29, #192]
|
||||
ldp x29, x30, [sp], #0xd0
|
||||
ldp x29, x30, [sp], #0x50
|
||||
ret
|
||||
#ifndef __APPLE__
|
||||
.size Transform_Sha512_Len_crypto,.-Transform_Sha512_Len_crypto
|
||||
|
|
|
|||
|
|
@ -51,6 +51,7 @@
|
|||
#endif
|
||||
|
||||
|
||||
#ifndef WOLFSSL_ARMASM
|
||||
#ifdef WOLFSSL_SHA3_SMALL
|
||||
/* Rotate a 64-bit value left.
|
||||
*
|
||||
|
|
@ -137,8 +138,7 @@ static const word64 hash_keccak_r[24] =
|
|||
* i The index of the loop.
|
||||
*/
|
||||
#define SWAP_ROTL(s, t1, t2, i) \
|
||||
do \
|
||||
{ \
|
||||
do { \
|
||||
t2 = s[K_I_##i]; s[K_I_##i] = ROTL64(t1, K_R_##i); \
|
||||
} \
|
||||
while (0)
|
||||
|
|
@ -151,12 +151,10 @@ while (0)
|
|||
* t Temporary variable.
|
||||
*/
|
||||
#define COL_MIX(s, b, x, t) \
|
||||
do \
|
||||
{ \
|
||||
do { \
|
||||
for (x = 0; x < 5; x++) \
|
||||
b[x] = s[x + 0] ^ s[x + 5] ^ s[x + 10] ^ s[x + 15] ^ s[x + 20]; \
|
||||
for (x = 0; x < 5; x++) \
|
||||
{ \
|
||||
for (x = 0; x < 5; x++) { \
|
||||
t = b[(x + 4) % 5] ^ ROTL64(b[(x + 1) % 5], 1); \
|
||||
s[x + 0] ^= t; \
|
||||
s[x + 5] ^= t; \
|
||||
|
|
@ -179,14 +177,12 @@ while (0)
|
|||
* t1 Temporary variable.
|
||||
*/
|
||||
#define ROW_MIX(s, b, y, x, t0, t1) \
|
||||
do \
|
||||
{ \
|
||||
for (y = 0; y < 5; y++) \
|
||||
{ \
|
||||
do { \
|
||||
for (y = 0; y < 5; y++) { \
|
||||
for (x = 0; x < 5; x++) \
|
||||
b[x] = s[y * 5 + x]; \
|
||||
for (x = 0; x < 5; x++) \
|
||||
s[y * 5 + x] = b[x] ^ (~b[(x + 1) % 5] & b[(x + 2) % 5]); \
|
||||
s[y * 5 + x] = b[x] ^ (~b[(x + 1) % 5] & b[(x + 2) % 5]); \
|
||||
} \
|
||||
} \
|
||||
while (0)
|
||||
|
|
@ -202,10 +198,8 @@ while (0)
|
|||
* t1 Temporary variable.
|
||||
*/
|
||||
#define ROW_MIX(s, b, y, x, t12, t34) \
|
||||
do \
|
||||
{ \
|
||||
for (y = 0; y < 5; y++) \
|
||||
{ \
|
||||
do { \
|
||||
for (y = 0; y < 5; y++) { \
|
||||
for (x = 0; x < 5; x++) \
|
||||
b[x] = s[y * 5 + x]; \
|
||||
t12 = (b[1] ^ b[2]); t34 = (b[3] ^ b[4]); \
|
||||
|
|
@ -351,8 +345,7 @@ static const word64 hash_keccak_r[24] =
|
|||
* t Temporary variable.
|
||||
*/
|
||||
#define COL_MIX(s, b, x, t) \
|
||||
do \
|
||||
{ \
|
||||
do { \
|
||||
(b)[0] = (s)[0] ^ (s)[5] ^ (s)[10] ^ (s)[15] ^ (s)[20]; \
|
||||
(b)[1] = (s)[1] ^ (s)[6] ^ (s)[11] ^ (s)[16] ^ (s)[21]; \
|
||||
(b)[2] = (s)[2] ^ (s)[7] ^ (s)[12] ^ (s)[17] ^ (s)[22]; \
|
||||
|
|
@ -384,8 +377,7 @@ while (0)
|
|||
* t1 Temporary variable. (Unused)
|
||||
*/
|
||||
#define ROW_MIX(s2, s1, b, t0, t1) \
|
||||
do \
|
||||
{ \
|
||||
do { \
|
||||
(b)[0] = (s1)[0]; \
|
||||
(b)[1] = S((s1), 0); \
|
||||
(b)[2] = S((s1), 1); \
|
||||
|
|
@ -449,8 +441,7 @@ while (0)
|
|||
* t34 Temporary variable.
|
||||
*/
|
||||
#define ROW_MIX(s2, s1, b, t12, t34) \
|
||||
do \
|
||||
{ \
|
||||
do { \
|
||||
(b)[0] = (s1)[0]; \
|
||||
(b)[1] = S((s1), 0); \
|
||||
(b)[2] = S((s1), 1); \
|
||||
|
|
@ -536,6 +527,7 @@ static void BlockSha3(word64 *s)
|
|||
}
|
||||
}
|
||||
#endif /* WOLFSSL_SHA3_SMALL */
|
||||
#endif /* !WOLFSSL_ARMASM */
|
||||
|
||||
static WC_INLINE word64 Load64Unaligned(const unsigned char *a)
|
||||
{
|
||||
|
|
@ -618,8 +610,7 @@ static int Sha3Update(wc_Sha3* sha3, const byte* data, word32 len, byte p)
|
|||
byte l;
|
||||
byte *t;
|
||||
|
||||
if (sha3->i > 0)
|
||||
{
|
||||
if (sha3->i > 0) {
|
||||
l = p * 8 - sha3->i;
|
||||
if (l > len) {
|
||||
l = (byte)len;
|
||||
|
|
@ -632,16 +623,14 @@ static int Sha3Update(wc_Sha3* sha3, const byte* data, word32 len, byte p)
|
|||
len -= i;
|
||||
sha3->i += i;
|
||||
|
||||
if (sha3->i == p * 8)
|
||||
{
|
||||
if (sha3->i == p * 8) {
|
||||
for (i = 0; i < p; i++)
|
||||
sha3->s[i] ^= Load64BitBigEndian(sha3->t + 8 * i);
|
||||
BlockSha3(sha3->s);
|
||||
sha3->i = 0;
|
||||
}
|
||||
}
|
||||
while (len >= ((word32)(p * 8)))
|
||||
{
|
||||
while (len >= ((word32)(p * 8))) {
|
||||
for (i = 0; i < p; i++)
|
||||
sha3->s[i] ^= Load64Unaligned(data + 8 * i);
|
||||
BlockSha3(sha3->s);
|
||||
|
|
|
|||
|
|
@ -128,6 +128,9 @@ struct wc_Sha3 {
|
|||
typedef wc_Sha3 wc_Shake;
|
||||
#endif
|
||||
|
||||
#ifdef WOLFSSL_ARMASM
|
||||
WOLFSSL_LOCAL void BlockSha3(word64 *s);
|
||||
#endif
|
||||
|
||||
WOLFSSL_API int wc_InitSha3_224(wc_Sha3* sha3, void* heap, int devId);
|
||||
WOLFSSL_API int wc_Sha3_224_Update(wc_Sha3* sha3, const byte* data, word32 len);
|
||||
|
|
|
|||
Loading…
Reference in New Issue