Blame mpn/x86_64/bobcat/copyi.asm
|
Packit |
5c3484 |
dnl AMD64 mpn_copyi optimised for AMD bobcat.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
dnl Copyright 2003, 2005, 2007, 2011, 2012 Free Software Foundation, Inc.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
dnl This file is part of the GNU MP Library.
|
|
Packit |
5c3484 |
dnl
|
|
Packit |
5c3484 |
dnl The GNU MP Library is free software; you can redistribute it and/or modify
|
|
Packit |
5c3484 |
dnl it under the terms of either:
|
|
Packit |
5c3484 |
dnl
|
|
Packit |
5c3484 |
dnl * the GNU Lesser General Public License as published by the Free
|
|
Packit |
5c3484 |
dnl Software Foundation; either version 3 of the License, or (at your
|
|
Packit |
5c3484 |
dnl option) any later version.
|
|
Packit |
5c3484 |
dnl
|
|
Packit |
5c3484 |
dnl or
|
|
Packit |
5c3484 |
dnl
|
|
Packit |
5c3484 |
dnl * the GNU General Public License as published by the Free Software
|
|
Packit |
5c3484 |
dnl Foundation; either version 2 of the License, or (at your option) any
|
|
Packit |
5c3484 |
dnl later version.
|
|
Packit |
5c3484 |
dnl
|
|
Packit |
5c3484 |
dnl or both in parallel, as here.
|
|
Packit |
5c3484 |
dnl
|
|
Packit |
5c3484 |
dnl The GNU MP Library is distributed in the hope that it will be useful, but
|
|
Packit |
5c3484 |
dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
|
|
Packit |
5c3484 |
dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
|
|
Packit |
5c3484 |
dnl for more details.
|
|
Packit |
5c3484 |
dnl
|
|
Packit |
5c3484 |
dnl You should have received copies of the GNU General Public License and the
|
|
Packit |
5c3484 |
dnl GNU Lesser General Public License along with the GNU MP Library. If not,
|
|
Packit |
5c3484 |
dnl see https://www.gnu.org/licenses/.
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
include(`../config.m4')
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
C cycles/limb
|
|
Packit |
5c3484 |
C AMD K8,K9 1
|
|
Packit |
5c3484 |
C AMD K10 1-2 (alignment fluctuations)
|
|
Packit |
5c3484 |
C AMD bd1 ?
|
|
Packit |
5c3484 |
C AMD bobcat 1.5
|
|
Packit |
5c3484 |
C Intel P4 2.8
|
|
Packit |
5c3484 |
C Intel core2 1
|
|
Packit |
5c3484 |
C Intel NHM 1-1.25
|
|
Packit |
5c3484 |
C Intel SBR 1
|
|
Packit |
5c3484 |
C Intel atom 2.87
|
|
Packit |
5c3484 |
C VIA nano 2
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
C INPUT PARAMETERS
|
|
Packit |
5c3484 |
C rp rdi
|
|
Packit |
5c3484 |
C up rsi
|
|
Packit |
5c3484 |
C n rdx
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
define(`rp',`%rdi')
|
|
Packit |
5c3484 |
define(`up',`%rsi')
|
|
Packit |
5c3484 |
define(`n',`%rdx')
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
ABI_SUPPORT(DOS64)
|
|
Packit |
5c3484 |
ABI_SUPPORT(STD64)
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
ASM_START()
|
|
Packit |
5c3484 |
TEXT
|
|
Packit |
5c3484 |
ALIGN(16)
|
|
Packit |
5c3484 |
PROLOGUE(mpn_copyi)
|
|
Packit |
5c3484 |
FUNC_ENTRY(3)
|
|
Packit |
5c3484 |
lea -32(up,n,8), up
|
|
Packit |
5c3484 |
lea -32(rp,n,8), rp
|
|
Packit |
5c3484 |
neg n
|
|
Packit |
5c3484 |
add $4, n
|
|
Packit |
5c3484 |
jg L(end)
|
|
Packit |
5c3484 |
ALIGN(16)
|
|
Packit |
5c3484 |
L(top): mov (up,n,8), %r8
|
|
Packit |
5c3484 |
mov %r8, (rp,n,8)
|
|
Packit |
5c3484 |
mov 8(up,n,8), %r8
|
|
Packit |
5c3484 |
mov %r8, 8(rp,n,8)
|
|
Packit |
5c3484 |
mov 16(up,n,8), %r8
|
|
Packit |
5c3484 |
mov %r8, 16(rp,n,8)
|
|
Packit |
5c3484 |
mov 24(up,n,8), %r8
|
|
Packit |
5c3484 |
mov %r8, 24(rp,n,8)
|
|
Packit |
5c3484 |
L(ent): add $4, n
|
|
Packit |
5c3484 |
jle L(top)
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
L(end): cmp $4, R32(n)
|
|
Packit |
5c3484 |
jz L(ret)
|
|
Packit |
5c3484 |
mov (up,n,8), %r8
|
|
Packit |
5c3484 |
mov %r8, (rp,n,8)
|
|
Packit |
5c3484 |
cmp $3, R32(n)
|
|
Packit |
5c3484 |
jz L(ret)
|
|
Packit |
5c3484 |
mov 8(up,n,8), %r8
|
|
Packit |
5c3484 |
mov %r8, 8(rp,n,8)
|
|
Packit |
5c3484 |
cmp $2, R32(n)
|
|
Packit |
5c3484 |
jz L(ret)
|
|
Packit |
5c3484 |
mov 16(up,n,8), %r8
|
|
Packit |
5c3484 |
mov %r8, 16(rp,n,8)
|
|
Packit |
5c3484 |
|
|
Packit |
5c3484 |
L(ret): FUNC_EXIT()
|
|
Packit |
5c3484 |
ret
|
|
Packit |
5c3484 |
EPILOGUE()
|
|
rpm-build |
01f633 |
CF_PROT
|