cheng/mpir
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

make sure fat dir has every file in fat structure

Browse Source
This commit is contained in:
jasonmoxham 2011-04-07 14:19:31 +00:00
parent b5c5bd605f
commit 3f15c30522
16 changed files with 161 additions and 848 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

87
mpn/x86_64/add_n.as
View File

@ -1,87 +0,0 @@
; AMD64 mpn_add_n
; Copyright 2008 Jason Moxham
; This file is part of the MPIR Library.
; The MPIR Library is free software; you can redistribute it and/or modify
; it under the terms of the GNU Lesser General Public License as published
; by the Free Software Foundation; either version 2.1 of the License, or (at
; your option) any later version.
; The MPIR Library 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 Lesser General Public
; License for more details.
; You should have received a copy of the GNU Lesser General Public License
; along with the MPIR Library; see the file COPYING.LIB. If not, write
; to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
; Boston, MA 02110-1301, USA.
; (rdi,rcx) = (rsi,rcx)+(rdx,rcx)
; rax = carry
%include 'yasm_mac.inc'
BITS 64
GLOBAL_FUNC mpn_add_n
mov rax, rcx
and rax, 3
shr rcx, 2
cmp rcx, 0
; carry flag is clear here
jnz loop1
mov r11, [rsi]
add r11, [rdx]
mov [rdi], r11
dec rax
jz end1
mov r11, [rsi+8]
adc r11, [rdx+8]
mov [rdi+8], r11
dec rax
jz end1
mov r11, [rsi+16]
adc r11, [rdx+16]
mov [rdi+16], r11
dec rax
end1:
adc rax, rax
ret
align 8
loop1
mov r11, [rsi]
mov r8, [rsi+8]
lea rsi, [rsi+32]
adc r11, [rdx]
adc r8, [rdx+8]
lea rdx, [rdx+32]
mov [rdi], r11
mov [rdi+8], r8
lea rdi, [rdi+32]
mov r9, [rsi-16]
mov r10, [rsi-8]
adc r9, [rdx-16]
adc r10, [rdx-8]
mov [rdi-16], r9
dec rcx
mov [rdi-8], r10
jnz loop1
inc rax
dec rax
jz end
mov r11, [rsi]
adc r11, [rdx]
mov [rdi], r11
dec rax
jz end
mov r11, [rsi+8]
adc r11, [rdx+8]
mov [rdi+8], r11
dec rax
jz end
mov r11, [rsi+16]
adc r11, [rdx+16]
mov [rdi+16], r11
dec rax
end:
adc rax, rax
ret

145
mpn/x86_64/addmul_1.asm
View File

@ -1,145 +0,0 @@
dnl mpn_addmul_1
dnl Copyright 2011 The Code Cavern
dnl This file is part of the MPIR Library.
dnl The MPIR Library is free software; you can redistribute it and/or modify
dnl it under the terms of the GNU Lesser General Public License as published
dnl by the Free Software Foundation; either version 2.1 of the License, or (at
dnl your option) any later version.
dnl The MPIR Library is distributed in the hope that it will be useful, but
dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
dnl License for more details.
dnl You should have received a copy of the GNU Lesser General Public License
dnl along with the MPIR Library; see the file COPYING.LIB. If not, write
dnl to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
dnl Boston, MA 02110-1301, USA.
include(`../config.m4')
ASM_START()
PROLOGUE(mpn_addmul_1)
mov (%rsi),%rax
cmp $1,%rdx
je one
mov $5,%r11
lea -40(%rsi,%rdx,8),%rsi
lea -40(%rdi,%rdx,8),%rdi
sub %rdx,%r11
mul %rcx
.byte 0x26
mov %rax,%r8
.byte 0x26
mov 8(%rsi,%r11,8),%rax
.byte 0x26
mov %rdx,%r9
.byte 0x26
cmp $0,%r11
.byte 0x26
mov %r12,-8(%rsp)
.byte 0x26
jge skiplp
lp: xor %r10,%r10
mul %rcx
add %r8,(%rdi,%r11,8)
adc %rax,%r9
adc %rdx,%r10
mov 16(%rsi,%r11,8),%rax
mul %rcx
add %r9,8(%rdi,%r11,8)
adc %rax,%r10
mov $0,%r12d
adc %rdx,%r12
mov 24(%rsi,%r11,8),%rax
xor %r8,%r8
xor %r9,%r9
mul %rcx
add %r10,16(%rdi,%r11,8)
adc %rax,%r12
adc %rdx,%r8
mov 32(%rsi,%r11,8),%rax
mul %rcx
add %r12,24(%rdi,%r11,8)
adc %rax,%r8
adc %rdx,%r9
add $4,%r11
mov 8(%rsi,%r11,8),%rax
jnc lp
skiplp:
xor %r10,%r10
mul %rcx
add %r8,(%rdi,%r11,8)
adc %rax,%r9
adc %rdx,%r10
cmp $2,%r11
ja case0
jz case1
jp case2
case3: mov 16(%rsi),%rax
mul %rcx
add %r9,8(%rdi)
adc %rax,%r10
mov $0,%r12d
adc %rdx,%r12
mov 24(%rsi),%rax
xor %r8,%r8
xor %r9,%r9
mul %rcx
add %r10,16(%rdi)
adc %rax,%r12
adc %rdx,%r8
mov 32(%rsi),%rax
mul %rcx
add %r12,24(%rdi)
adc %rax,%r8
adc %rdx,%r9
add %r8,32(%rdi)
adc $0,%r9
mov -8(%rsp),%r12
mov %r9,%rax
ret
case2: mov 24(%rsi),%rax
mul %rcx
add %r9,16(%rdi)
adc %rax,%r10
mov $0,%r12d
adc %rdx,%r12
mov 32(%rsi),%rax
xor %r8,%r8
#xor %r9,%r9
mul %rcx
add %r10,24(%rdi)
adc %rax,%r12
adc %rdx,%r8
add %r12,32(%rdi)
adc $0,%r8
mov -8(%rsp),%r12
mov %r8,%rax
ret
ALIGN(16)
case1: mov 32(%rsi),%rax
mul %rcx
add %r9,24(%rdi)
adc %rax,%r10
mov $0,%r12d
adc %rdx,%r12
add %r10,32(%rdi)
adc $0,%r12
mov %r12,%rax
mov -8(%rsp),%r12
ret
one: mul %rcx
add %rax,(%rdi)
adc $0,%rdx
mov %rdx,%rax
ret
case0: add %r9,32(%rdi)
adc $0,%r10
mov -8(%rsp),%r12
mov %r10,%rax
ret
EPILOGUE()

23
mpn/x86_64/fat/add_n.c Normal file
View File

@ -0,0 +1,23 @@
/* Fat binary fallback mpn_add_n
Copyright 2011 Jason Moxham
This file is part of the MPIR Library.
The MPIR Library is free software; you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation; either version 2.1 of the License, or (at your
option) any later version.
The MPIR Library 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 Lesser General Public
License for more details.
You should have received a copy of the GNU Lesser General Public License
along with the GNU MP Library; see the file COPYING.LIB. If not, write to
the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
MA 02110-1301, USA. */
#include "mpn/generic/add_n.c"

23
mpn/x86_64/fat/addmul_1.c Normal file
View File

@ -0,0 +1,23 @@
/* Fat binary fallback mpn_addmul_1
Copyright 2011 Jason Moxham
This file is part of the MPIR Library.
The MPIR Library is free software; you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation; either version 2.1 of the License, or (at your
option) any later version.
The MPIR Library 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 Lesser General Public
License for more details.
You should have received a copy of the GNU Lesser General Public License
along with the GNU MP Library; see the file COPYING.LIB. If not, write to
the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
MA 02110-1301, USA. */
#include "mpn/generic/addmul_1.c"

2
mpn/x86_64/fat/copyd.c
View File

@ -22,7 +22,5 @@ MA 02110-1301, USA. */
#include "mpir.h"
#include "gmp-impl.h"
//#ifndef mpn_copyd
void mpn_copyd(mp_ptr rp,mp_srcptr sp,mp_size_t n)
{MPN_COPY_DECR(rp,sp,n);return;}
//#endif

2
mpn/x86_64/fat/copyi.c
View File

@ -22,7 +22,5 @@ MA 02110-1301, USA. */
#include "mpir.h"
#include "gmp-impl.h"
//#ifndef mpn_copyi
void mpn_copyi(mp_ptr rp,mp_srcptr sp,mp_size_t n)
{MPN_COPY_INCR(rp,sp,n);return;}
//#endif

23
mpn/x86_64/fat/lshift.c Normal file
View File

@ -0,0 +1,23 @@
/* Fat binary fallback mpn_lshift
Copyright 2011 Jason Moxham
This file is part of the MPIR Library.
The MPIR Library is free software; you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation; either version 2.1 of the License, or (at your
option) any later version.
The MPIR Library 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 Lesser General Public
License for more details.
You should have received a copy of the GNU Lesser General Public License
along with the GNU MP Library; see the file COPYING.LIB. If not, write to
the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
MA 02110-1301, USA. */
#include "mpn/generic/lshift.c"

23
mpn/x86_64/fat/mul_1.c Normal file
View File

@ -0,0 +1,23 @@
/* Fat binary fallback mpn_mul_1
Copyright 2011 Jason Moxham
This file is part of the MPIR Library.
The MPIR Library is free software; you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation; either version 2.1 of the License, or (at your
option) any later version.
The MPIR Library 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 Lesser General Public
License for more details.
You should have received a copy of the GNU Lesser General Public License
along with the GNU MP Library; see the file COPYING.LIB. If not, write to
the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
MA 02110-1301, USA. */
#include "mpn/generic/mul_1.c"

23
mpn/x86_64/fat/rshift.c Normal file
View File

@ -0,0 +1,23 @@
/* Fat binary fallback mpn_rshift
Copyright 2011 Jason Moxham
This file is part of the MPIR Library.
The MPIR Library is free software; you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation; either version 2.1 of the License, or (at your
option) any later version.
The MPIR Library 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 Lesser General Public
License for more details.
You should have received a copy of the GNU Lesser General Public License
along with the GNU MP Library; see the file COPYING.LIB. If not, write to
the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
MA 02110-1301, USA. */
#include "mpn/generic/rshift.c"

23
mpn/x86_64/fat/sub_n.c Normal file
View File

@ -0,0 +1,23 @@
/* Fat binary fallback mpn_sub_n
Copyright 2011 Jason Moxham
This file is part of the MPIR Library.
The MPIR Library is free software; you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation; either version 2.1 of the License, or (at your
option) any later version.
The MPIR Library 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 Lesser General Public
License for more details.
You should have received a copy of the GNU Lesser General Public License
along with the GNU MP Library; see the file COPYING.LIB. If not, write to
the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
MA 02110-1301, USA. */
#include "mpn/generic/sub_n.c"

23
mpn/x86_64/fat/submul_1.c Normal file
View File

@ -0,0 +1,23 @@
/* Fat binary fallback mpn_submul_1
Copyright 2011 Jason Moxham
This file is part of the MPIR Library.
The MPIR Library is free software; you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation; either version 2.1 of the License, or (at your
option) any later version.
The MPIR Library 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 Lesser General Public
License for more details.
You should have received a copy of the GNU Lesser General Public License
along with the GNU MP Library; see the file COPYING.LIB. If not, write to
the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
MA 02110-1301, USA. */
#include "mpn/generic/submul_1.c"

96
mpn/x86_64/lshift.as
View File

@ -1,96 +0,0 @@
; AMD64 mpn_lshift
; Copyright 2008 Jason Moxham
; This file is part of the MPIR Library.
; The MPIR Library is free software; you can redistribute it and/or modify
; it under the terms of the GNU Lesser General Public License as published
; by the Free Software Foundation; either version 2.1 of the License, or (at
; your option) any later version.
; The MPIR Library 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 Lesser General Public
; License for more details.
; You should have received a copy of the GNU Lesser General Public License
; along with the MPIR Library; see the file COPYING.LIB. If not, write
; to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
; Boston, MA 02110-1301, USA.
; (rdi,rdx) = (rsi,rdx)<<rcx
; rax = carry
%include 'yasm_mac.inc'
BITS 64
GLOBAL_FUNC mpn_lshift
mov eax, 64
sub rax, rcx
movq mm0, rcx
sub rdx, 4
movq mm1, rax
movq mm5, [rsi+rdx*8+24]
movq mm3, mm5
psrlq mm5, mm1
movq rax, mm5
psllq mm3, mm0
jbe skiploop
align 16
loop1
movq mm2, [rsi+rdx*8+16]
movq mm4, mm2
psrlq mm2, mm1
por mm3, mm2
movq [rdi+rdx*8+24], mm3
psllq mm4, mm0
movq mm5, [rsi+rdx*8+8]
movq mm3, mm5
psrlq mm5, mm1
por mm4, mm5
movq [rdi+rdx*8+16], mm4
psllq mm3, mm0
movq mm2, [rsi+rdx*8]
movq mm4, mm2
psrlq mm2, mm1
por mm3, mm2
movq [rdi+rdx*8+8], mm3
psllq mm4, mm0
movq mm5, [rsi+rdx*8-8]
movq mm3, mm5
psrlq mm5, mm1
por mm4, mm5
movq [rdi+rdx*8], mm4
psllq mm3, mm0
sub rdx, 4
ja loop1
skiploop:
cmp rdx, -1
jl next
movq mm2, [rsi+rdx*8+16]
movq mm4, mm2
psrlq mm2, mm1
por mm3, mm2
movq [rdi+rdx*8+24], mm3
psllq mm4, mm0
movq mm5, [rsi+rdx*8+8]
movq mm3, mm5
psrlq mm5, mm1
por mm4, mm5
movq [rdi+rdx*8+16], mm4
psllq mm3, mm0
sub rdx, 2
next:
test rdx, 1
jnz end
movq mm2, [rsi+rdx*8+16]
movq mm4, mm2
psrlq mm2, mm1
por mm3, mm2
movq [rdi+rdx*8+24], mm3
psllq mm4, mm0
movq [rdi+rdx*8+16], mm4
emms
ret
end:
movq [rdi+rdx*8+24], mm3
emms
ret

145
mpn/x86_64/mul_1.asm
View File

@ -1,145 +0,0 @@
dnl mpn_mul_1
dnl Copyright 2011 The Code Cavern
dnl This file is part of the MPIR Library.
dnl The MPIR Library is free software; you can redistribute it and/or modify
dnl it under the terms of the GNU Lesser General Public License as published
dnl by the Free Software Foundation; either version 2.1 of the License, or (at
dnl your option) any later version.
dnl The MPIR Library is distributed in the hope that it will be useful, but
dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
dnl License for more details.
dnl You should have received a copy of the GNU Lesser General Public License
dnl along with the MPIR Library; see the file COPYING.LIB. If not, write
dnl to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
dnl Boston, MA 02110-1301, USA.
include(`../config.m4')
ASM_START()
PROLOGUE(mpn_mul_1)
mov (%rsi),%rax
cmp $1,%rdx
je one
mov $5,%r11
lea -40(%rsi,%rdx,8),%rsi
lea -40(%rdi,%rdx,8),%rdi
sub %rdx,%r11
mul %rcx
.byte 0x26
mov %rax,%r8
.byte 0x26
mov 8(%rsi,%r11,8),%rax
.byte 0x26
mov %rdx,%r9
.byte 0x26
cmp $0,%r11
.byte 0x26
mov %r12,-8(%rsp)
.byte 0x26
jge skiplp
lp: xor %r10,%r10
mul %rcx
mov %r8,(%rdi,%r11,8)
add %rax,%r9
adc %rdx,%r10
mov 16(%rsi,%r11,8),%rax
mul %rcx
mov %r9,8(%rdi,%r11,8)
add %rax,%r10
mov $0,%r12d
adc %rdx,%r12
mov 24(%rsi,%r11,8),%rax
xor %r8,%r8
xor %r9,%r9
mul %rcx
mov %r10,16(%rdi,%r11,8)
add %rax,%r12
adc %rdx,%r8
mov 32(%rsi,%r11,8),%rax
mul %rcx
mov %r12,24(%rdi,%r11,8)
add %rax,%r8
adc %rdx,%r9
add $4,%r11
mov 8(%rsi,%r11,8),%rax
jnc lp
skiplp:
xor %r10,%r10
mul %rcx
mov %r8,(%rdi,%r11,8)
add %rax,%r9
adc %rdx,%r10
cmp $2,%r11
ja case0
jz case1
jp case2
case3: mov 16(%rsi),%rax
mul %rcx
mov %r9,8(%rdi)
add %rax,%r10
mov $0,%r12d
adc %rdx,%r12
mov 24(%rsi),%rax
xor %r8,%r8
xor %r9,%r9
mul %rcx
mov %r10,16(%rdi)
add %rax,%r12
adc %rdx,%r8
mov 32(%rsi),%rax
mul %rcx
mov %r12,24(%rdi)
add %rax,%r8
adc %rdx,%r9
mov %r8,32(%rdi)
#add $0,%r9
mov -8(%rsp),%r12
mov %r9,%rax
ret
case2: mov 24(%rsi),%rax
mul %rcx
mov %r9,16(%rdi)
add %rax,%r10
mov $0,%r12d
adc %rdx,%r12
mov 32(%rsi),%rax
xor %r8,%r8
#xor %r9,%r9
mul %rcx
mov %r10,24(%rdi)
add %rax,%r12
adc %rdx,%r8
mov %r12,32(%rdi)
#add $0,%r8
mov -8(%rsp),%r12
mov %r8,%rax
ret
ALIGN(16)
case1: mov 32(%rsi),%rax
mul %rcx
mov %r9,24(%rdi)
add %rax,%r10
mov $0,%r12d
adc %rdx,%r12
mov %r10,32(%rdi)
#add $0,%r12
mov %r12,%rax
mov -8(%rsp),%r12
ret
one: mul %rcx
mov %rax,(%rdi)
#add $0,%rdx
mov %rdx,%rax
ret
case0: mov %r9,32(%rdi)
#add $0,%r10
mov -8(%rsp),%r12
mov %r10,%rax
ret
EPILOGUE()

139
mpn/x86_64/rshift.asm
View File

@ -1,139 +0,0 @@
dnl mpn_rshift
dnl Copyright 2009 Jason Moxham
dnl This file is part of the MPIR Library.
dnl The MPIR Library is free software; you can redistribute it and/or modify
dnl it under the terms of the GNU Lesser General Public License as published
dnl by the Free Software Foundation; either version 2.1 of the License, or (at
dnl your option) any later version.
dnl The MPIR Library is distributed in the hope that it will be useful, but
dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
dnl License for more details.
dnl You should have received a copy of the GNU Lesser General Public License
dnl along with the MPIR Library; see the file COPYING.LIB. If not, write
dnl to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
dnl Boston, MA 02110-1301, USA.
include(`../config.m4')
C ret mpn_rshift(mp_ptr,mp_ptr,mp_size_t,ul)
C rax rdi, rsi, rdx
ASM_START()
PROLOGUE(mpn_rshift)
C below really a movq
movd %rcx,%mm0
mov $64,%rax
sub %rcx,%rax
C below really a movq
movd %rax,%mm1
mov $4,%rcx
lea -32(%rsi,%rdx,8),%rsi
lea -32(%rdi,%rdx,8),%rdi
sub %rdx,%rcx
movq (%rsi,%rcx,8),%mm5
movq %mm5,%mm3
psllq %mm1,%mm5
C below really a movq
movd %mm5,%rax
psrlq %mm0,%mm3
jge skiplp
ALIGN(16)
lp:
movq 8(%rsi,%rcx,8),%mm2
movq %mm2,%mm4
psllq %mm1,%mm2
por %mm2,%mm3
movq %mm3,(%rdi,%rcx,8)
psrlq %mm0,%mm4
movq 16(%rsi,%rcx,8),%mm5
movq %mm5,%mm3
psllq %mm1,%mm5
por %mm5,%mm4
movq %mm4,8(%rdi,%rcx,8)
psrlq %mm0,%mm3
movq 24(%rsi,%rcx,8),%mm2
movq %mm2,%mm4
psllq %mm1,%mm2
por %mm2,%mm3
movq 32(%rsi,%rcx,8),%mm5
movq %mm3,16(%rdi,%rcx,8)
psrlq %mm0,%mm4
movq %mm5,%mm3
psllq %mm1,%mm5
por %mm5,%mm4
movq %mm4,24(%rdi,%rcx,8)
psrlq %mm0,%mm3
add $4,%rcx
jnc lp
skiplp:
cmp $2,%rcx
ja case0
jz case1
jp case2
case3:
movq 8(%rsi,%rcx,8),%mm2
movq %mm2,%mm4
psllq %mm1,%mm2
por %mm2,%mm3
movq %mm3,(%rdi,%rcx,8)
psrlq %mm0,%mm4
movq 16(%rsi,%rcx,8),%mm5
movq %mm5,%mm3
psllq %mm1,%mm5
por %mm5,%mm4
movq %mm4,8(%rdi,%rcx,8)
psrlq %mm0,%mm3
movq 24(%rsi,%rcx,8),%mm2
movq %mm2,%mm4
psllq %mm1,%mm2
por %mm2,%mm3
movq %mm3,16(%rdi,%rcx,8)
psrlq %mm0,%mm4
movq %mm4,24(%rdi,%rcx,8)
emms
ret
ALIGN(16)
case2:
movq 8(%rsi,%rcx,8),%mm2
movq %mm2,%mm4
psllq %mm1,%mm2
por %mm2,%mm3
movq %mm3,(%rdi,%rcx,8)
psrlq %mm0,%mm4
movq 16(%rsi,%rcx,8),%mm5
movq %mm5,%mm3
psllq %mm1,%mm5
por %mm5,%mm4
movq %mm4,8(%rdi,%rcx,8)
psrlq %mm0,%mm3
movq %mm3,16(%rdi,%rcx,8)
emms
ret
ALIGN(16)
case1:
movq 8(%rsi,%rcx,8),%mm2
movq %mm2,%mm4
psllq %mm1,%mm2
por %mm2,%mm3
movq %mm3,(%rdi,%rcx,8)
psrlq %mm0,%mm4
movq %mm4,8(%rdi,%rcx,8)
emms
ret
ALIGN(16)
case0:
movq %mm3,(%rdi,%rcx,8)
emms
ret
EPILOGUE()

87
mpn/x86_64/sub_n.as
View File

@ -1,87 +0,0 @@
; AMD64 mpn_sub_n
; Copyright 2008 Jason Moxham
; This file is part of the MPIR Library.
; The MPIR Library is free software; you can redistribute it and/or modify
; it under the terms of the GNU Lesser General Public License as published
; by the Free Software Foundation; either version 2.1 of the License, or (at
; your option) any later version.
; The MPIR Library 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 Lesser General Public
; License for more details.
; You should have received a copy of the GNU Lesser General Public License
; along with the MPIR Library; see the file COPYING.LIB. If not, write
; to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
; Boston, MA 02110-1301, USA.
; (rdi,rcx)=(rsi,rcx)-(rdx,rcx)
; rax=borrow
%include 'yasm_mac.inc'
BITS 64
GLOBAL_FUNC mpn_sub_n
mov rax, rcx
and rax, 3
shr rcx, 2
cmp rcx, 0
; carry flag is clear here
jnz loop1
mov r11, [rsi]
sub r11, [rdx]
mov [rdi], r11
dec rax
jz end1
mov r11, [rsi+8]
sbb r11, [rdx+8]
mov [rdi+8], r11
dec rax
jz end1
mov r11, [rsi+16]
sbb r11, [rdx+16]
mov [rdi+16], r11
dec rax
end1:
adc rax, rax
ret
align 8
loop1:
mov r11, [rsi]
mov r8, [rsi+8]
lea rsi, [rsi+32]
sbb r11, [rdx]
sbb r8, [rdx+8]
lea rdx, [rdx+32]
mov [rdi], r11
mov [rdi+8], r8
lea rdi, [rdi+32]
mov r9, [rsi-16]
mov r10, [rsi-8]
sbb r9, [rdx-16]
sbb r10, [rdx-8]
mov [rdi-16], r9
dec rcx
mov [rdi-8], r10
jnz loop1
inc rax
dec rax
jz end
mov r11, [rsi]
sbb r11, [rdx]
mov [rdi], r11
dec rax
jz end
mov r11, [rsi+8]
sbb r11, [rdx+8]
mov [rdi+8], r11
dec rax
jz end
mov r11, [rsi+16]
sbb r11, [rdx+16]
mov [rdi+16], r11
dec rax
end:
adc rax, rax
ret

145
mpn/x86_64/submul_1.asm
View File

@ -1,145 +0,0 @@
dnl mpn_submul_1
dnl Copyright 2011 The Code Cavern
dnl This file is part of the MPIR Library.
dnl The MPIR Library is free software; you can redistribute it and/or modify
dnl it under the terms of the GNU Lesser General Public License as published
dnl by the Free Software Foundation; either version 2.1 of the License, or (at
dnl your option) any later version.
dnl The MPIR Library is distributed in the hope that it will be useful, but
dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
dnl License for more details.
dnl You should have received a copy of the GNU Lesser General Public License
dnl along with the MPIR Library; see the file COPYING.LIB. If not, write
dnl to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
dnl Boston, MA 02110-1301, USA.
include(`../config.m4')
ASM_START()
PROLOGUE(mpn_submul_1)
mov (%rsi),%rax
cmp $1,%rdx
je one
mov $5,%r11
lea -40(%rsi,%rdx,8),%rsi
lea -40(%rdi,%rdx,8),%rdi
sub %rdx,%r11
mul %rcx
.byte 0x26
mov %rax,%r8
.byte 0x26
mov 8(%rsi,%r11,8),%rax
.byte 0x26
mov %rdx,%r9
.byte 0x26
cmp $0,%r11
.byte 0x26
mov %r12,-8(%rsp)
.byte 0x26
jge skiplp
lp: xor %r10,%r10
mul %rcx
sub %r8,(%rdi,%r11,8)
adc %rax,%r9
adc %rdx,%r10
mov 16(%rsi,%r11,8),%rax
mul %rcx
sub %r9,8(%rdi,%r11,8)
adc %rax,%r10
mov $0,%r12d
adc %rdx,%r12
mov 24(%rsi,%r11,8),%rax
xor %r8,%r8
xor %r9,%r9
mul %rcx
sub %r10,16(%rdi,%r11,8)
adc %rax,%r12
adc %rdx,%r8
mov 32(%rsi,%r11,8),%rax
mul %rcx
sub %r12,24(%rdi,%r11,8)
adc %rax,%r8
adc %rdx,%r9
add $4,%r11
mov 8(%rsi,%r11,8),%rax
jnc lp
skiplp:
xor %r10,%r10
mul %rcx
sub %r8,(%rdi,%r11,8)
adc %rax,%r9
adc %rdx,%r10
cmp $2,%r11
ja case0
jz case1
jp case2
case3: mov 16(%rsi),%rax
mul %rcx
sub %r9,8(%rdi)
adc %rax,%r10
mov $0,%r12d
adc %rdx,%r12
mov 24(%rsi),%rax
xor %r8,%r8
xor %r9,%r9
mul %rcx
sub %r10,16(%rdi)
adc %rax,%r12
adc %rdx,%r8
mov 32(%rsi),%rax
mul %rcx
sub %r12,24(%rdi)
adc %rax,%r8
adc %rdx,%r9
sub %r8,32(%rdi)
adc $0,%r9
mov -8(%rsp),%r12
mov %r9,%rax
ret
case2: mov 24(%rsi),%rax
mul %rcx
sub %r9,16(%rdi)
adc %rax,%r10
mov $0,%r12d
adc %rdx,%r12
mov 32(%rsi),%rax
xor %r8,%r8
#xor %r9,%r9
mul %rcx
sub %r10,24(%rdi)
adc %rax,%r12
adc %rdx,%r8
sub %r12,32(%rdi)
adc $0,%r8
mov -8(%rsp),%r12
mov %r8,%rax
ret
ALIGN(16)
case1: mov 32(%rsi),%rax
mul %rcx
sub %r9,24(%rdi)
adc %rax,%r10
mov $0,%r12d
adc %rdx,%r12
sub %r10,32(%rdi)
adc $0,%r12
mov %r12,%rax
mov -8(%rsp),%r12
ret
one: mul %rcx
sub %rax,(%rdi)
adc $0,%rdx
mov %rdx,%rax
ret
case0: sub %r9,32(%rdi)
adc $0,%r10
mov -8(%rsp),%r12
mov %r10,%rax
ret
EPILOGUE()