You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
1006 lines
28 KiB
1006 lines
28 KiB
########################################################################
|
|
# Implement fast SHA-512 with AVX2 instructions. (x86_64)
|
|
#
|
|
# Copyright (C) 2013 Intel Corporation.
|
|
#
|
|
# Authors:
|
|
# James Guilford <james.guilford@intel.com>
|
|
# Kirk Yap <kirk.s.yap@intel.com>
|
|
# David Cote <david.m.cote@intel.com>
|
|
# Tim Chen <tim.c.chen@linux.intel.com>
|
|
#
|
|
# This software is available to you under a choice of one of two
|
|
# licenses. You may choose to be licensed under the terms of the GNU
|
|
# General Public License (GPL) Version 2, available from the file
|
|
# COPYING in the main directory of this source tree, or the
|
|
# OpenIB.org BSD license below:
|
|
#
|
|
# Redistribution and use in source and binary forms, with or
|
|
# without modification, are permitted provided that the following
|
|
# conditions are met:
|
|
#
|
|
# - Redistributions of source code must retain the above
|
|
# copyright notice, this list of conditions and the following
|
|
# disclaimer.
|
|
#
|
|
# - Redistributions in binary form must reproduce the above
|
|
# copyright notice, this list of conditions and the following
|
|
# disclaimer in the documentation and/or other materials
|
|
# provided with the distribution.
|
|
#
|
|
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
|
# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
|
# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
|
# NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
|
|
# BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
|
|
# ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
|
|
# CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
# SOFTWARE.
|
|
#
|
|
########################################################################
|
|
#
|
|
# This code is described in an Intel White-Paper:
|
|
# "Fast SHA-512 Implementations on Intel Architecture Processors"
|
|
#
|
|
# To find it, surf to http://www.intel.com/p/en_US/embedded
|
|
# and search for that title.
|
|
#
|
|
########################################################################
|
|
# This code schedules 1 blocks at a time, with 4 lanes per block
|
|
########################################################################
|
|
# Using this part of Minio codebase under the license
|
|
# Apache License Version 2.0 with modifications
|
|
##
|
|
|
|
#ifdef HAS_AVX2
|
|
#ifndef ENTRY
|
|
#define ENTRY(name) \
|
|
.globl name ; \
|
|
.align 4,0x90 ; \
|
|
name:
|
|
#endif
|
|
|
|
#ifndef END
|
|
#define END(name) \
|
|
.size name, .-name
|
|
#endif
|
|
|
|
#ifndef ENDPROC
|
|
#define ENDPROC(name) \
|
|
.type name, @function ; \
|
|
END(name)
|
|
#endif
|
|
|
|
#define NUM_INVALID 100
|
|
|
|
#define TYPE_R32 0
|
|
#define TYPE_R64 1
|
|
#define TYPE_XMM 2
|
|
#define TYPE_INVALID 100
|
|
|
|
.macro R32_NUM opd r32
|
|
\opd = NUM_INVALID
|
|
.ifc \r32,%eax
|
|
\opd = 0
|
|
.endif
|
|
.ifc \r32,%ecx
|
|
\opd = 1
|
|
.endif
|
|
.ifc \r32,%edx
|
|
\opd = 2
|
|
.endif
|
|
.ifc \r32,%ebx
|
|
\opd = 3
|
|
.endif
|
|
.ifc \r32,%esp
|
|
\opd = 4
|
|
.endif
|
|
.ifc \r32,%ebp
|
|
\opd = 5
|
|
.endif
|
|
.ifc \r32,%esi
|
|
\opd = 6
|
|
.endif
|
|
.ifc \r32,%edi
|
|
\opd = 7
|
|
.endif
|
|
#ifdef X86_64
|
|
.ifc \r32,%r8d
|
|
\opd = 8
|
|
.endif
|
|
.ifc \r32,%r9d
|
|
\opd = 9
|
|
.endif
|
|
.ifc \r32,%r10d
|
|
\opd = 10
|
|
.endif
|
|
.ifc \r32,%r11d
|
|
\opd = 11
|
|
.endif
|
|
.ifc \r32,%r12d
|
|
\opd = 12
|
|
.endif
|
|
.ifc \r32,%r13d
|
|
\opd = 13
|
|
.endif
|
|
.ifc \r32,%r14d
|
|
\opd = 14
|
|
.endif
|
|
.ifc \r32,%r15d
|
|
\opd = 15
|
|
.endif
|
|
#endif
|
|
.endm
|
|
|
|
.macro R64_NUM opd r64
|
|
\opd = NUM_INVALID
|
|
#ifdef X86_64
|
|
.ifc \r64,%rax
|
|
\opd = 0
|
|
.endif
|
|
.ifc \r64,%rcx
|
|
\opd = 1
|
|
.endif
|
|
.ifc \r64,%rdx
|
|
\opd = 2
|
|
.endif
|
|
.ifc \r64,%rbx
|
|
\opd = 3
|
|
.endif
|
|
.ifc \r64,%rsp
|
|
\opd = 4
|
|
.endif
|
|
.ifc \r64,%rbp
|
|
\opd = 5
|
|
.endif
|
|
.ifc \r64,%rsi
|
|
\opd = 6
|
|
.endif
|
|
.ifc \r64,%rdi
|
|
\opd = 7
|
|
.endif
|
|
.ifc \r64,%r8
|
|
\opd = 8
|
|
.endif
|
|
.ifc \r64,%r9
|
|
\opd = 9
|
|
.endif
|
|
.ifc \r64,%r10
|
|
\opd = 10
|
|
.endif
|
|
.ifc \r64,%r11
|
|
\opd = 11
|
|
.endif
|
|
.ifc \r64,%r12
|
|
\opd = 12
|
|
.endif
|
|
.ifc \r64,%r13
|
|
\opd = 13
|
|
.endif
|
|
.ifc \r64,%r14
|
|
\opd = 14
|
|
.endif
|
|
.ifc \r64,%r15
|
|
\opd = 15
|
|
.endif
|
|
#endif
|
|
.endm
|
|
|
|
.macro XMM_NUM opd xmm
|
|
\opd = NUM_INVALID
|
|
.ifc \xmm,%xmm0
|
|
\opd = 0
|
|
.endif
|
|
.ifc \xmm,%xmm1
|
|
\opd = 1
|
|
.endif
|
|
.ifc \xmm,%xmm2
|
|
\opd = 2
|
|
.endif
|
|
.ifc \xmm,%xmm3
|
|
\opd = 3
|
|
.endif
|
|
.ifc \xmm,%xmm4
|
|
\opd = 4
|
|
.endif
|
|
.ifc \xmm,%xmm5
|
|
\opd = 5
|
|
.endif
|
|
.ifc \xmm,%xmm6
|
|
\opd = 6
|
|
.endif
|
|
.ifc \xmm,%xmm7
|
|
\opd = 7
|
|
.endif
|
|
.ifc \xmm,%xmm8
|
|
\opd = 8
|
|
.endif
|
|
.ifc \xmm,%xmm9
|
|
\opd = 9
|
|
.endif
|
|
.ifc \xmm,%xmm10
|
|
\opd = 10
|
|
.endif
|
|
.ifc \xmm,%xmm11
|
|
\opd = 11
|
|
.endif
|
|
.ifc \xmm,%xmm12
|
|
\opd = 12
|
|
.endif
|
|
.ifc \xmm,%xmm13
|
|
\opd = 13
|
|
.endif
|
|
.ifc \xmm,%xmm14
|
|
\opd = 14
|
|
.endif
|
|
.ifc \xmm,%xmm15
|
|
\opd = 15
|
|
.endif
|
|
.endm
|
|
|
|
.macro TYPE type reg
|
|
R32_NUM reg_type_r32 \reg
|
|
R64_NUM reg_type_r64 \reg
|
|
XMM_NUM reg_type_xmm \reg
|
|
.if reg_type_r64 <> NUM_INVALID
|
|
\type = TYPE_R64
|
|
.elseif reg_type_r32 <> NUM_INVALID
|
|
\type = TYPE_R32
|
|
.elseif reg_type_xmm <> NUM_INVALID
|
|
\type = TYPE_XMM
|
|
.else
|
|
\type = TYPE_INVALID
|
|
.endif
|
|
.endm
|
|
|
|
.macro PFX_OPD_SIZE
|
|
.byte 0x66
|
|
.endm
|
|
|
|
.macro PFX_REX opd1 opd2 W=0
|
|
.if ((\opd1 | \opd2) & 8) || \W
|
|
.byte 0x40 | ((\opd1 & 8) >> 3) | ((\opd2 & 8) >> 1) | (\W << 3)
|
|
.endif
|
|
.endm
|
|
|
|
.macro MODRM mod opd1 opd2
|
|
.byte \mod | (\opd1 & 7) | ((\opd2 & 7) << 3)
|
|
.endm
|
|
|
|
.macro PSHUFB_XMM xmm1 xmm2
|
|
XMM_NUM pshufb_opd1 \xmm1
|
|
XMM_NUM pshufb_opd2 \xmm2
|
|
PFX_OPD_SIZE
|
|
PFX_REX pshufb_opd1 pshufb_opd2
|
|
.byte 0x0f, 0x38, 0x00
|
|
MODRM 0xc0 pshufb_opd1 pshufb_opd2
|
|
.endm
|
|
|
|
.macro PCLMULQDQ imm8 xmm1 xmm2
|
|
XMM_NUM clmul_opd1 \xmm1
|
|
XMM_NUM clmul_opd2 \xmm2
|
|
PFX_OPD_SIZE
|
|
PFX_REX clmul_opd1 clmul_opd2
|
|
.byte 0x0f, 0x3a, 0x44
|
|
MODRM 0xc0 clmul_opd1 clmul_opd2
|
|
.byte \imm8
|
|
.endm
|
|
|
|
.macro PEXTRD imm8 xmm gpr
|
|
R32_NUM extrd_opd1 \gpr
|
|
XMM_NUM extrd_opd2 \xmm
|
|
PFX_OPD_SIZE
|
|
PFX_REX extrd_opd1 extrd_opd2
|
|
.byte 0x0f, 0x3a, 0x16
|
|
MODRM 0xc0 extrd_opd1 extrd_opd2
|
|
.byte \imm8
|
|
.endm
|
|
|
|
.macro MOVQ_R64_XMM opd1 opd2
|
|
TYPE movq_r64_xmm_opd1_type \opd1
|
|
.if movq_r64_xmm_opd1_type == TYPE_XMM
|
|
XMM_NUM movq_r64_xmm_opd1 \opd1
|
|
R64_NUM movq_r64_xmm_opd2 \opd2
|
|
.else
|
|
R64_NUM movq_r64_xmm_opd1 \opd1
|
|
XMM_NUM movq_r64_xmm_opd2 \opd2
|
|
.endif
|
|
PFX_OPD_SIZE
|
|
PFX_REX movq_r64_xmm_opd1 movq_r64_xmm_opd2 1
|
|
.if movq_r64_xmm_opd1_type == TYPE_XMM
|
|
.byte 0x0f, 0x7e
|
|
.else
|
|
.byte 0x0f, 0x6e
|
|
.endif
|
|
MODRM 0xc0 movq_r64_xmm_opd1 movq_r64_xmm_opd2
|
|
.endm
|
|
|
|
.text
|
|
|
|
# Virtual Registers
|
|
Y_0 = %ymm4
|
|
Y_1 = %ymm5
|
|
Y_2 = %ymm6
|
|
Y_3 = %ymm7
|
|
|
|
YTMP0 = %ymm0
|
|
YTMP1 = %ymm1
|
|
YTMP2 = %ymm2
|
|
YTMP3 = %ymm3
|
|
YTMP4 = %ymm8
|
|
XFER = YTMP0
|
|
|
|
BYTE_FLIP_MASK = %ymm9
|
|
|
|
# 1st arg
|
|
INP = %rdi
|
|
# 2nd arg
|
|
CTX = %rsi
|
|
# 3rd arg
|
|
NUM_BLKS = %rdx
|
|
|
|
c = %rcx
|
|
d = %r8
|
|
e = %rdx
|
|
y3 = %rdi
|
|
|
|
TBL = %rbp
|
|
|
|
a = %rax
|
|
b = %rbx
|
|
|
|
f = %r9
|
|
g = %r10
|
|
h = %r11
|
|
old_h = %r11
|
|
|
|
T1 = %r12
|
|
y0 = %r13
|
|
y1 = %r14
|
|
y2 = %r15
|
|
|
|
y4 = %r12
|
|
|
|
# Local variables (stack frame)
|
|
XFER_SIZE = 4*8
|
|
SRND_SIZE = 1*8
|
|
INP_SIZE = 1*8
|
|
INPEND_SIZE = 1*8
|
|
RSPSAVE_SIZE = 1*8
|
|
GPRSAVE_SIZE = 6*8
|
|
|
|
frame_XFER = 0
|
|
frame_SRND = frame_XFER + XFER_SIZE
|
|
frame_INP = frame_SRND + SRND_SIZE
|
|
frame_INPEND = frame_INP + INP_SIZE
|
|
frame_RSPSAVE = frame_INPEND + INPEND_SIZE
|
|
frame_GPRSAVE = frame_RSPSAVE + RSPSAVE_SIZE
|
|
frame_size = frame_GPRSAVE + GPRSAVE_SIZE
|
|
|
|
## assume buffers not aligned
|
|
#define VMOVDQ vmovdqu
|
|
|
|
# addm [mem], reg
|
|
# Add reg to mem using reg-mem add and store
|
|
.macro addm p1 p2
|
|
add \p1, \p2
|
|
mov \p2, \p1
|
|
.endm
|
|
|
|
|
|
# COPY_YMM_AND_BSWAP ymm, [mem], byte_flip_mask
|
|
# Load ymm with mem and byte swap each dword
|
|
.macro COPY_YMM_AND_BSWAP p1 p2 p3
|
|
VMOVDQ \p2, \p1
|
|
vpshufb \p3, \p1, \p1
|
|
.endm
|
|
# rotate_Ys
|
|
# Rotate values of symbols Y0...Y3
|
|
.macro rotate_Ys
|
|
Y_ = Y_0
|
|
Y_0 = Y_1
|
|
Y_1 = Y_2
|
|
Y_2 = Y_3
|
|
Y_3 = Y_
|
|
.endm
|
|
|
|
# RotateState
|
|
.macro RotateState
|
|
# Rotate symbols a..h right
|
|
old_h = h
|
|
TMP_ = h
|
|
h = g
|
|
g = f
|
|
f = e
|
|
e = d
|
|
d = c
|
|
c = b
|
|
b = a
|
|
a = TMP_
|
|
.endm
|
|
|
|
# macro MY_VPALIGNR YDST, YSRC1, YSRC2, RVAL
|
|
# YDST = {YSRC1, YSRC2} >> RVAL*8
|
|
.macro MY_VPALIGNR YDST YSRC1 YSRC2 RVAL
|
|
vperm2f128 $0x3, \YSRC2, \YSRC1, \YDST # YDST = {YS1_LO, YS2_HI}
|
|
vpalignr $\RVAL, \YSRC2, \YDST, \YDST # YDST = {YDS1, YS2} >> RVAL*8
|
|
.endm
|
|
|
|
.macro FOUR_ROUNDS_AND_SCHED
|
|
################################### RND N + 0 #########################################
|
|
|
|
# Extract w[t-7]
|
|
MY_VPALIGNR YTMP0, Y_3, Y_2, 8 # YTMP0 = W[-7]
|
|
# Calculate w[t-16] + w[t-7]
|
|
vpaddq Y_0, YTMP0, YTMP0 # YTMP0 = W[-7] + W[-16]
|
|
# Extract w[t-15]
|
|
MY_VPALIGNR YTMP1, Y_1, Y_0, 8 # YTMP1 = W[-15]
|
|
|
|
# Calculate sigma0
|
|
|
|
# Calculate w[t-15] ror 1
|
|
vpsrlq $1, YTMP1, YTMP2
|
|
vpsllq $(64-1), YTMP1, YTMP3
|
|
vpor YTMP2, YTMP3, YTMP3 # YTMP3 = W[-15] ror 1
|
|
# Calculate w[t-15] shr 7
|
|
vpsrlq $7, YTMP1, YTMP4 # YTMP4 = W[-15] >> 7
|
|
|
|
mov a, y3 # y3 = a # MAJA
|
|
rorx $41, e, y0 # y0 = e >> 41 # S1A
|
|
rorx $18, e, y1 # y1 = e >> 18 # S1B
|
|
add frame_XFER(%rsp),h # h = k + w + h # --
|
|
or c, y3 # y3 = a|c # MAJA
|
|
mov f, y2 # y2 = f # CH
|
|
rorx $34, a, T1 # T1 = a >> 34 # S0B
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) # S1
|
|
xor g, y2 # y2 = f^g # CH
|
|
rorx $14, e, y1 # y1 = (e >> 14) # S1
|
|
|
|
and e, y2 # y2 = (f^g)&e # CH
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) ^ (e>>14) # S1
|
|
rorx $39, a, y1 # y1 = a >> 39 # S0A
|
|
add h, d # d = k + w + h + d # --
|
|
|
|
and b, y3 # y3 = (a|c)&b # MAJA
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) # S0
|
|
rorx $28, a, T1 # T1 = (a >> 28) # S0
|
|
|
|
xor g, y2 # y2 = CH = ((f^g)&e)^g # CH
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) ^ (a>>28) # S0
|
|
mov a, T1 # T1 = a # MAJB
|
|
and c, T1 # T1 = a&c # MAJB
|
|
|
|
add y0, y2 # y2 = S1 + CH # --
|
|
or T1, y3 # y3 = MAJ = (a|c)&b)|(a&c) # MAJ
|
|
add y1, h # h = k + w + h + S0 # --
|
|
|
|
add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
|
|
|
|
add y2, h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
|
|
add y3, h # h = t1 + S0 + MAJ # --
|
|
|
|
RotateState
|
|
|
|
################################### RND N + 1 #########################################
|
|
|
|
# Calculate w[t-15] ror 8
|
|
vpsrlq $8, YTMP1, YTMP2
|
|
vpsllq $(64-8), YTMP1, YTMP1
|
|
vpor YTMP2, YTMP1, YTMP1 # YTMP1 = W[-15] ror 8
|
|
# XOR the three components
|
|
vpxor YTMP4, YTMP3, YTMP3 # YTMP3 = W[-15] ror 1 ^ W[-15] >> 7
|
|
vpxor YTMP1, YTMP3, YTMP1 # YTMP1 = s0
|
|
|
|
|
|
# Add three components, w[t-16], w[t-7] and sigma0
|
|
vpaddq YTMP1, YTMP0, YTMP0 # YTMP0 = W[-16] + W[-7] + s0
|
|
# Move to appropriate lanes for calculating w[16] and w[17]
|
|
vperm2f128 $0x0, YTMP0, YTMP0, Y_0 # Y_0 = W[-16] + W[-7] + s0 {BABA}
|
|
# Move to appropriate lanes for calculating w[18] and w[19]
|
|
vpand MASK_YMM_LO(%rip), YTMP0, YTMP0 # YTMP0 = W[-16] + W[-7] + s0 {DC00}
|
|
|
|
# Calculate w[16] and w[17] in both 128 bit lanes
|
|
|
|
# Calculate sigma1 for w[16] and w[17] on both 128 bit lanes
|
|
vperm2f128 $0x11, Y_3, Y_3, YTMP2 # YTMP2 = W[-2] {BABA}
|
|
vpsrlq $6, YTMP2, YTMP4 # YTMP4 = W[-2] >> 6 {BABA}
|
|
|
|
|
|
mov a, y3 # y3 = a # MAJA
|
|
rorx $41, e, y0 # y0 = e >> 41 # S1A
|
|
rorx $18, e, y1 # y1 = e >> 18 # S1B
|
|
add 1*8+frame_XFER(%rsp), h # h = k + w + h # --
|
|
or c, y3 # y3 = a|c # MAJA
|
|
|
|
|
|
mov f, y2 # y2 = f # CH
|
|
rorx $34, a, T1 # T1 = a >> 34 # S0B
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) # S1
|
|
xor g, y2 # y2 = f^g # CH
|
|
|
|
|
|
rorx $14, e, y1 # y1 = (e >> 14) # S1
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) ^ (e>>14) # S1
|
|
rorx $39, a, y1 # y1 = a >> 39 # S0A
|
|
and e, y2 # y2 = (f^g)&e # CH
|
|
add h, d # d = k + w + h + d # --
|
|
|
|
and b, y3 # y3 = (a|c)&b # MAJA
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) # S0
|
|
|
|
rorx $28, a, T1 # T1 = (a >> 28) # S0
|
|
xor g, y2 # y2 = CH = ((f^g)&e)^g # CH
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) ^ (a>>28) # S0
|
|
mov a, T1 # T1 = a # MAJB
|
|
and c, T1 # T1 = a&c # MAJB
|
|
add y0, y2 # y2 = S1 + CH # --
|
|
|
|
or T1, y3 # y3 = MAJ = (a|c)&b)|(a&c) # MAJ
|
|
add y1, h # h = k + w + h + S0 # --
|
|
|
|
add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
|
|
add y2, h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
|
|
add y3, h # h = t1 + S0 + MAJ # --
|
|
|
|
RotateState
|
|
|
|
|
|
################################### RND N + 2 #########################################
|
|
|
|
vpsrlq $19, YTMP2, YTMP3 # YTMP3 = W[-2] >> 19 {BABA}
|
|
vpsllq $(64-19), YTMP2, YTMP1 # YTMP1 = W[-2] << 19 {BABA}
|
|
vpor YTMP1, YTMP3, YTMP3 # YTMP3 = W[-2] ror 19 {BABA}
|
|
vpxor YTMP3, YTMP4, YTMP4 # YTMP4 = W[-2] ror 19 ^ W[-2] >> 6 {BABA}
|
|
vpsrlq $61, YTMP2, YTMP3 # YTMP3 = W[-2] >> 61 {BABA}
|
|
vpsllq $(64-61), YTMP2, YTMP1 # YTMP1 = W[-2] << 61 {BABA}
|
|
vpor YTMP1, YTMP3, YTMP3 # YTMP3 = W[-2] ror 61 {BABA}
|
|
vpxor YTMP3, YTMP4, YTMP4 # YTMP4 = s1 = (W[-2] ror 19) ^
|
|
# (W[-2] ror 61) ^ (W[-2] >> 6) {BABA}
|
|
|
|
# Add sigma1 to the other compunents to get w[16] and w[17]
|
|
vpaddq YTMP4, Y_0, Y_0 # Y_0 = {W[1], W[0], W[1], W[0]}
|
|
|
|
# Calculate sigma1 for w[18] and w[19] for upper 128 bit lane
|
|
vpsrlq $6, Y_0, YTMP4 # YTMP4 = W[-2] >> 6 {DC--}
|
|
|
|
mov a, y3 # y3 = a # MAJA
|
|
rorx $41, e, y0 # y0 = e >> 41 # S1A
|
|
add 2*8+frame_XFER(%rsp), h # h = k + w + h # --
|
|
|
|
rorx $18, e, y1 # y1 = e >> 18 # S1B
|
|
or c, y3 # y3 = a|c # MAJA
|
|
mov f, y2 # y2 = f # CH
|
|
xor g, y2 # y2 = f^g # CH
|
|
|
|
rorx $34, a, T1 # T1 = a >> 34 # S0B
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) # S1
|
|
and e, y2 # y2 = (f^g)&e # CH
|
|
|
|
rorx $14, e, y1 # y1 = (e >> 14) # S1
|
|
add h, d # d = k + w + h + d # --
|
|
and b, y3 # y3 = (a|c)&b # MAJA
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) ^ (e>>14) # S1
|
|
rorx $39, a, y1 # y1 = a >> 39 # S0A
|
|
xor g, y2 # y2 = CH = ((f^g)&e)^g # CH
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) # S0
|
|
rorx $28, a, T1 # T1 = (a >> 28) # S0
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) ^ (a>>28) # S0
|
|
mov a, T1 # T1 = a # MAJB
|
|
and c, T1 # T1 = a&c # MAJB
|
|
add y0, y2 # y2 = S1 + CH # --
|
|
|
|
or T1, y3 # y3 = MAJ = (a|c)&b)|(a&c) # MAJ
|
|
add y1, h # h = k + w + h + S0 # --
|
|
add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
|
|
add y2, h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
|
|
|
|
add y3, h # h = t1 + S0 + MAJ # --
|
|
|
|
RotateState
|
|
|
|
################################### RND N + 3 #########################################
|
|
|
|
vpsrlq $19, Y_0, YTMP3 # YTMP3 = W[-2] >> 19 {DC--}
|
|
vpsllq $(64-19), Y_0, YTMP1 # YTMP1 = W[-2] << 19 {DC--}
|
|
vpor YTMP1, YTMP3, YTMP3 # YTMP3 = W[-2] ror 19 {DC--}
|
|
vpxor YTMP3, YTMP4, YTMP4 # YTMP4 = W[-2] ror 19 ^ W[-2] >> 6 {DC--}
|
|
vpsrlq $61, Y_0, YTMP3 # YTMP3 = W[-2] >> 61 {DC--}
|
|
vpsllq $(64-61), Y_0, YTMP1 # YTMP1 = W[-2] << 61 {DC--}
|
|
vpor YTMP1, YTMP3, YTMP3 # YTMP3 = W[-2] ror 61 {DC--}
|
|
vpxor YTMP3, YTMP4, YTMP4 # YTMP4 = s1 = (W[-2] ror 19) ^
|
|
# (W[-2] ror 61) ^ (W[-2] >> 6) {DC--}
|
|
|
|
# Add the sigma0 + w[t-7] + w[t-16] for w[18] and w[19]
|
|
# to newly calculated sigma1 to get w[18] and w[19]
|
|
vpaddq YTMP4, YTMP0, YTMP2 # YTMP2 = {W[3], W[2], --, --}
|
|
|
|
# Form w[19, w[18], w17], w[16]
|
|
vpblendd $0xF0, YTMP2, Y_0, Y_0 # Y_0 = {W[3], W[2], W[1], W[0]}
|
|
|
|
mov a, y3 # y3 = a # MAJA
|
|
rorx $41, e, y0 # y0 = e >> 41 # S1A
|
|
rorx $18, e, y1 # y1 = e >> 18 # S1B
|
|
add 3*8+frame_XFER(%rsp), h # h = k + w + h # --
|
|
or c, y3 # y3 = a|c # MAJA
|
|
|
|
|
|
mov f, y2 # y2 = f # CH
|
|
rorx $34, a, T1 # T1 = a >> 34 # S0B
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) # S1
|
|
xor g, y2 # y2 = f^g # CH
|
|
|
|
|
|
rorx $14, e, y1 # y1 = (e >> 14) # S1
|
|
and e, y2 # y2 = (f^g)&e # CH
|
|
add h, d # d = k + w + h + d # --
|
|
and b, y3 # y3 = (a|c)&b # MAJA
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) ^ (e>>14) # S1
|
|
xor g, y2 # y2 = CH = ((f^g)&e)^g # CH
|
|
|
|
rorx $39, a, y1 # y1 = a >> 39 # S0A
|
|
add y0, y2 # y2 = S1 + CH # --
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) # S0
|
|
add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
|
|
|
|
rorx $28, a, T1 # T1 = (a >> 28) # S0
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) ^ (a>>28) # S0
|
|
mov a, T1 # T1 = a # MAJB
|
|
and c, T1 # T1 = a&c # MAJB
|
|
or T1, y3 # y3 = MAJ = (a|c)&b)|(a&c) # MAJ
|
|
|
|
add y1, h # h = k + w + h + S0 # --
|
|
add y2, h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
|
|
add y3, h # h = t1 + S0 + MAJ # --
|
|
|
|
RotateState
|
|
|
|
rotate_Ys
|
|
.endm
|
|
|
|
.macro DO_4ROUNDS
|
|
|
|
################################### RND N + 0 #########################################
|
|
|
|
mov f, y2 # y2 = f # CH
|
|
rorx $41, e, y0 # y0 = e >> 41 # S1A
|
|
rorx $18, e, y1 # y1 = e >> 18 # S1B
|
|
xor g, y2 # y2 = f^g # CH
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) # S1
|
|
rorx $14, e, y1 # y1 = (e >> 14) # S1
|
|
and e, y2 # y2 = (f^g)&e # CH
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) ^ (e>>14) # S1
|
|
rorx $34, a, T1 # T1 = a >> 34 # S0B
|
|
xor g, y2 # y2 = CH = ((f^g)&e)^g # CH
|
|
rorx $39, a, y1 # y1 = a >> 39 # S0A
|
|
mov a, y3 # y3 = a # MAJA
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) # S0
|
|
rorx $28, a, T1 # T1 = (a >> 28) # S0
|
|
add frame_XFER(%rsp), h # h = k + w + h # --
|
|
or c, y3 # y3 = a|c # MAJA
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) ^ (a>>28) # S0
|
|
mov a, T1 # T1 = a # MAJB
|
|
and b, y3 # y3 = (a|c)&b # MAJA
|
|
and c, T1 # T1 = a&c # MAJB
|
|
add y0, y2 # y2 = S1 + CH # --
|
|
|
|
add h, d # d = k + w + h + d # --
|
|
or T1, y3 # y3 = MAJ = (a|c)&b)|(a&c) # MAJ
|
|
add y1, h # h = k + w + h + S0 # --
|
|
|
|
add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
|
|
|
|
RotateState
|
|
|
|
################################### RND N + 1 #########################################
|
|
|
|
add y2, old_h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
|
|
mov f, y2 # y2 = f # CH
|
|
rorx $41, e, y0 # y0 = e >> 41 # S1A
|
|
rorx $18, e, y1 # y1 = e >> 18 # S1B
|
|
xor g, y2 # y2 = f^g # CH
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) # S1
|
|
rorx $14, e, y1 # y1 = (e >> 14) # S1
|
|
and e, y2 # y2 = (f^g)&e # CH
|
|
add y3, old_h # h = t1 + S0 + MAJ # --
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) ^ (e>>14) # S1
|
|
rorx $34, a, T1 # T1 = a >> 34 # S0B
|
|
xor g, y2 # y2 = CH = ((f^g)&e)^g # CH
|
|
rorx $39, a, y1 # y1 = a >> 39 # S0A
|
|
mov a, y3 # y3 = a # MAJA
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) # S0
|
|
rorx $28, a, T1 # T1 = (a >> 28) # S0
|
|
add 8*1+frame_XFER(%rsp), h # h = k + w + h # --
|
|
or c, y3 # y3 = a|c # MAJA
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) ^ (a>>28) # S0
|
|
mov a, T1 # T1 = a # MAJB
|
|
and b, y3 # y3 = (a|c)&b # MAJA
|
|
and c, T1 # T1 = a&c # MAJB
|
|
add y0, y2 # y2 = S1 + CH # --
|
|
|
|
add h, d # d = k + w + h + d # --
|
|
or T1, y3 # y3 = MAJ = (a|c)&b)|(a&c) # MAJ
|
|
add y1, h # h = k + w + h + S0 # --
|
|
|
|
add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
|
|
|
|
RotateState
|
|
|
|
################################### RND N + 2 #########################################
|
|
|
|
add y2, old_h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
|
|
mov f, y2 # y2 = f # CH
|
|
rorx $41, e, y0 # y0 = e >> 41 # S1A
|
|
rorx $18, e, y1 # y1 = e >> 18 # S1B
|
|
xor g, y2 # y2 = f^g # CH
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) # S1
|
|
rorx $14, e, y1 # y1 = (e >> 14) # S1
|
|
and e, y2 # y2 = (f^g)&e # CH
|
|
add y3, old_h # h = t1 + S0 + MAJ # --
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) ^ (e>>14) # S1
|
|
rorx $34, a, T1 # T1 = a >> 34 # S0B
|
|
xor g, y2 # y2 = CH = ((f^g)&e)^g # CH
|
|
rorx $39, a, y1 # y1 = a >> 39 # S0A
|
|
mov a, y3 # y3 = a # MAJA
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) # S0
|
|
rorx $28, a, T1 # T1 = (a >> 28) # S0
|
|
add 8*2+frame_XFER(%rsp), h # h = k + w + h # --
|
|
or c, y3 # y3 = a|c # MAJA
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) ^ (a>>28) # S0
|
|
mov a, T1 # T1 = a # MAJB
|
|
and b, y3 # y3 = (a|c)&b # MAJA
|
|
and c, T1 # T1 = a&c # MAJB
|
|
add y0, y2 # y2 = S1 + CH # --
|
|
|
|
add h, d # d = k + w + h + d # --
|
|
or T1, y3 # y3 = MAJ = (a|c)&b)|(a&c) # MAJ
|
|
add y1, h # h = k + w + h + S0 # --
|
|
|
|
add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
|
|
|
|
RotateState
|
|
|
|
################################### RND N + 3 #########################################
|
|
|
|
add y2, old_h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
|
|
mov f, y2 # y2 = f # CH
|
|
rorx $41, e, y0 # y0 = e >> 41 # S1A
|
|
rorx $18, e, y1 # y1 = e >> 18 # S1B
|
|
xor g, y2 # y2 = f^g # CH
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) # S1
|
|
rorx $14, e, y1 # y1 = (e >> 14) # S1
|
|
and e, y2 # y2 = (f^g)&e # CH
|
|
add y3, old_h # h = t1 + S0 + MAJ # --
|
|
|
|
xor y1, y0 # y0 = (e>>41) ^ (e>>18) ^ (e>>14) # S1
|
|
rorx $34, a, T1 # T1 = a >> 34 # S0B
|
|
xor g, y2 # y2 = CH = ((f^g)&e)^g # CH
|
|
rorx $39, a, y1 # y1 = a >> 39 # S0A
|
|
mov a, y3 # y3 = a # MAJA
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) # S0
|
|
rorx $28, a, T1 # T1 = (a >> 28) # S0
|
|
add 8*3+frame_XFER(%rsp), h # h = k + w + h # --
|
|
or c, y3 # y3 = a|c # MAJA
|
|
|
|
xor T1, y1 # y1 = (a>>39) ^ (a>>34) ^ (a>>28) # S0
|
|
mov a, T1 # T1 = a # MAJB
|
|
and b, y3 # y3 = (a|c)&b # MAJA
|
|
and c, T1 # T1 = a&c # MAJB
|
|
add y0, y2 # y2 = S1 + CH # --
|
|
|
|
|
|
add h, d # d = k + w + h + d # --
|
|
or T1, y3 # y3 = MAJ = (a|c)&b)|(a&c) # MAJ
|
|
add y1, h # h = k + w + h + S0 # --
|
|
|
|
add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
|
|
|
|
add y2, h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
|
|
|
|
add y3, h # h = t1 + S0 + MAJ # --
|
|
|
|
RotateState
|
|
|
|
.endm
|
|
|
|
########################################################################
|
|
# void sha512_transform_rorx(const void* M, void* D, uint64_t L)#
|
|
# Purpose: Updates the SHA512 digest stored at D with the message stored in M.
|
|
# The size of the message pointed to by M must be an integer multiple of SHA512
|
|
# message blocks.
|
|
# L is the message length in SHA512 blocks
|
|
########################################################################
|
|
ENTRY(sha512_transform_rorx)
|
|
# Allocate Stack Space
|
|
mov %rsp, %rax
|
|
sub $frame_size, %rsp
|
|
and $~(0x20 - 1), %rsp
|
|
mov %rax, frame_RSPSAVE(%rsp)
|
|
|
|
# Save GPRs
|
|
mov %rbp, frame_GPRSAVE(%rsp)
|
|
mov %rbx, 8*1+frame_GPRSAVE(%rsp)
|
|
mov %r12, 8*2+frame_GPRSAVE(%rsp)
|
|
mov %r13, 8*3+frame_GPRSAVE(%rsp)
|
|
mov %r14, 8*4+frame_GPRSAVE(%rsp)
|
|
mov %r15, 8*5+frame_GPRSAVE(%rsp)
|
|
|
|
shl $7, NUM_BLKS # convert to bytes
|
|
jz done_hash
|
|
add INP, NUM_BLKS # pointer to end of data
|
|
mov NUM_BLKS, frame_INPEND(%rsp)
|
|
|
|
## load initial digest
|
|
mov 8*0(CTX),a
|
|
mov 8*1(CTX),b
|
|
mov 8*2(CTX),c
|
|
mov 8*3(CTX),d
|
|
mov 8*4(CTX),e
|
|
mov 8*5(CTX),f
|
|
mov 8*6(CTX),g
|
|
mov 8*7(CTX),h
|
|
|
|
vmovdqa PSHUFFLE_BYTE_FLIP_MASK(%rip), BYTE_FLIP_MASK
|
|
|
|
loop0:
|
|
lea K512(%rip), TBL
|
|
|
|
## byte swap first 16 dwords
|
|
COPY_YMM_AND_BSWAP Y_0, (INP), BYTE_FLIP_MASK
|
|
COPY_YMM_AND_BSWAP Y_1, 1*32(INP), BYTE_FLIP_MASK
|
|
COPY_YMM_AND_BSWAP Y_2, 2*32(INP), BYTE_FLIP_MASK
|
|
COPY_YMM_AND_BSWAP Y_3, 3*32(INP), BYTE_FLIP_MASK
|
|
|
|
mov INP, frame_INP(%rsp)
|
|
|
|
## schedule 64 input dwords, by doing 12 rounds of 4 each
|
|
movq $4, frame_SRND(%rsp)
|
|
|
|
.align 16
|
|
loop1:
|
|
vpaddq (TBL), Y_0, XFER
|
|
vmovdqa XFER, frame_XFER(%rsp)
|
|
FOUR_ROUNDS_AND_SCHED
|
|
|
|
vpaddq 1*32(TBL), Y_0, XFER
|
|
vmovdqa XFER, frame_XFER(%rsp)
|
|
FOUR_ROUNDS_AND_SCHED
|
|
|
|
vpaddq 2*32(TBL), Y_0, XFER
|
|
vmovdqa XFER, frame_XFER(%rsp)
|
|
FOUR_ROUNDS_AND_SCHED
|
|
|
|
vpaddq 3*32(TBL), Y_0, XFER
|
|
vmovdqa XFER, frame_XFER(%rsp)
|
|
add $(4*32), TBL
|
|
FOUR_ROUNDS_AND_SCHED
|
|
|
|
subq $1, frame_SRND(%rsp)
|
|
jne loop1
|
|
|
|
movq $2, frame_SRND(%rsp)
|
|
loop2:
|
|
vpaddq (TBL), Y_0, XFER
|
|
vmovdqa XFER, frame_XFER(%rsp)
|
|
DO_4ROUNDS
|
|
vpaddq 1*32(TBL), Y_1, XFER
|
|
vmovdqa XFER, frame_XFER(%rsp)
|
|
add $(2*32), TBL
|
|
DO_4ROUNDS
|
|
|
|
vmovdqa Y_2, Y_0
|
|
vmovdqa Y_3, Y_1
|
|
|
|
subq $1, frame_SRND(%rsp)
|
|
jne loop2
|
|
|
|
addm 8*0(CTX),a
|
|
addm 8*1(CTX),b
|
|
addm 8*2(CTX),c
|
|
addm 8*3(CTX),d
|
|
addm 8*4(CTX),e
|
|
addm 8*5(CTX),f
|
|
addm 8*6(CTX),g
|
|
addm 8*7(CTX),h
|
|
|
|
mov frame_INP(%rsp), INP
|
|
add $128, INP
|
|
cmp frame_INPEND(%rsp), INP
|
|
jne loop0
|
|
|
|
done_hash:
|
|
|
|
# Restore GPRs
|
|
mov frame_GPRSAVE(%rsp) ,%rbp
|
|
mov 8*1+frame_GPRSAVE(%rsp) ,%rbx
|
|
mov 8*2+frame_GPRSAVE(%rsp) ,%r12
|
|
mov 8*3+frame_GPRSAVE(%rsp) ,%r13
|
|
mov 8*4+frame_GPRSAVE(%rsp) ,%r14
|
|
mov 8*5+frame_GPRSAVE(%rsp) ,%r15
|
|
|
|
# Restore Stack Pointer
|
|
mov frame_RSPSAVE(%rsp), %rsp
|
|
ret
|
|
ENDPROC(sha512_transform_rorx)
|
|
|
|
########################################################################
|
|
### Binary Data
|
|
|
|
.data
|
|
|
|
.align 64
|
|
# K[t] used in SHA512 hashing
|
|
K512:
|
|
.quad 0x428a2f98d728ae22,0x7137449123ef65cd
|
|
.quad 0xb5c0fbcfec4d3b2f,0xe9b5dba58189dbbc
|
|
.quad 0x3956c25bf348b538,0x59f111f1b605d019
|
|
.quad 0x923f82a4af194f9b,0xab1c5ed5da6d8118
|
|
.quad 0xd807aa98a3030242,0x12835b0145706fbe
|
|
.quad 0x243185be4ee4b28c,0x550c7dc3d5ffb4e2
|
|
.quad 0x72be5d74f27b896f,0x80deb1fe3b1696b1
|
|
.quad 0x9bdc06a725c71235,0xc19bf174cf692694
|
|
.quad 0xe49b69c19ef14ad2,0xefbe4786384f25e3
|
|
.quad 0x0fc19dc68b8cd5b5,0x240ca1cc77ac9c65
|
|
.quad 0x2de92c6f592b0275,0x4a7484aa6ea6e483
|
|
.quad 0x5cb0a9dcbd41fbd4,0x76f988da831153b5
|
|
.quad 0x983e5152ee66dfab,0xa831c66d2db43210
|
|
.quad 0xb00327c898fb213f,0xbf597fc7beef0ee4
|
|
.quad 0xc6e00bf33da88fc2,0xd5a79147930aa725
|
|
.quad 0x06ca6351e003826f,0x142929670a0e6e70
|
|
.quad 0x27b70a8546d22ffc,0x2e1b21385c26c926
|
|
.quad 0x4d2c6dfc5ac42aed,0x53380d139d95b3df
|
|
.quad 0x650a73548baf63de,0x766a0abb3c77b2a8
|
|
.quad 0x81c2c92e47edaee6,0x92722c851482353b
|
|
.quad 0xa2bfe8a14cf10364,0xa81a664bbc423001
|
|
.quad 0xc24b8b70d0f89791,0xc76c51a30654be30
|
|
.quad 0xd192e819d6ef5218,0xd69906245565a910
|
|
.quad 0xf40e35855771202a,0x106aa07032bbd1b8
|
|
.quad 0x19a4c116b8d2d0c8,0x1e376c085141ab53
|
|
.quad 0x2748774cdf8eeb99,0x34b0bcb5e19b48a8
|
|
.quad 0x391c0cb3c5c95a63,0x4ed8aa4ae3418acb
|
|
.quad 0x5b9cca4f7763e373,0x682e6ff3d6b2b8a3
|
|
.quad 0x748f82ee5defb2fc,0x78a5636f43172f60
|
|
.quad 0x84c87814a1f0ab72,0x8cc702081a6439ec
|
|
.quad 0x90befffa23631e28,0xa4506cebde82bde9
|
|
.quad 0xbef9a3f7b2c67915,0xc67178f2e372532b
|
|
.quad 0xca273eceea26619c,0xd186b8c721c0c207
|
|
.quad 0xeada7dd6cde0eb1e,0xf57d4f7fee6ed178
|
|
.quad 0x06f067aa72176fba,0x0a637dc5a2c898a6
|
|
.quad 0x113f9804bef90dae,0x1b710b35131c471b
|
|
.quad 0x28db77f523047d84,0x32caab7b40c72493
|
|
.quad 0x3c9ebe0a15c9bebc,0x431d67c49c100d4c
|
|
.quad 0x4cc5d4becb3e42b6,0x597f299cfc657e2a
|
|
.quad 0x5fcb6fab3ad6faec,0x6c44198c4a475817
|
|
|
|
.align 32
|
|
|
|
# Mask for byte-swapping a couple of qwords in an XMM register using (v)pshufb.
|
|
PSHUFFLE_BYTE_FLIP_MASK:
|
|
.octa 0x08090a0b0c0d0e0f0001020304050607
|
|
.octa 0x18191a1b1c1d1e1f1011121314151617
|
|
|
|
MASK_YMM_LO:
|
|
.octa 0x00000000000000000000000000000000
|
|
.octa 0xFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF
|
|
#endif
|
|
|