mpir/mpn/x86_64w/copyd.asm
2009-02-22 21:03:08 +00:00

105 lines
3.1 KiB
NASM

; Copyright 1999, 2000, 2001, 2002 Free Software Foundation, Inc.
;
; Copyright 2008 Brian Gladman
;
; 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.
;
; AMD64 mpn_copyd -- decrementing copy limb vector
;
; Calling interface:
;
; void mpn_copyd(
; mp_ptr dst, rcx
; mp_srcptr src, rdx
; mp_size_t size r8
; )
;
; This is an SEH Leaf Function (no unwind support needed)
%define UNROLL_THRESHOLD 16
%define d_ptr rcx
%define s_ptr rdx
%define s_len r8
bits 64
section .text
global __gmpn_copyd:function
%ifdef DLL
export __gmpn_copyd
%endif
__gmpn_copyd:
movsxd s_len, r8d
cmp s_len, byte UNROLL_THRESHOLD
jge .2 ; if many limbs to move
dec s_len
jl .1
.0: mov rax, [s_ptr+s_len*8] ; short move via rax
mov [d_ptr+s_len*8], rax
dec s_len
jge .0 ; avoid single byte ret that
.1: rep ret ; interferes with branch prediction
.2: mov rax, s_ptr ; find relative alignment of
xor rax, d_ptr ; source and destination (min
test al, 8
jnz .7 ; not 16 byte aligned
lea rax,[s_ptr+s_len*8]
test al, 8 ; see if src is on 16 byte boundary
jz .3
dec s_len
mov rax, [rax-8] ; if not do a one limb copy
mov [d_ptr+s_len*8], rax
.3: lea s_len, [s_len-4] ; now 16 byte aligned
.4: prefetchnta [s_ptr+s_len*8+16-3*64] ; should this be -4*64 ??
movdqa xmm0, [s_ptr+s_len*8+16] ; move 32 bytes at a time
movntdq [d_ptr+s_len*8+16], xmm0
movdqa xmm0, [s_ptr+s_len*8]
movntdq [d_ptr+s_len*8], xmm0
sub s_len, 4
jge .4
sfence
test s_len, 2
jz .5
movdqa xmm0, [s_ptr+s_len*8+16] ; move 16 bytes if necessary
movdqa [d_ptr+s_len*8+16], xmm0
.5: test s_len, 1
jz .6
movq xmm0, [s_ptr] ; move 8 bytes if necessary
movq [d_ptr], xmm0
.6: ret
.7: lea s_len, [s_len-2] ; move 8 bytes at a time
.8: movq xmm0, [s_ptr+s_len*8+8]
movq xmm1, [s_ptr+s_len*8]
movq [d_ptr+s_len*8+8], xmm0
movq [d_ptr+s_len*8], xmm1
sub s_len, 2
jge .8
test s_len, 1
jz .9
movq xmm0, [s_ptr]
movq [d_ptr], xmm0
.9: ret
end