xref: /netbsd-src/external/lgpl3/gmp/dist/mpn/x86/pentium4/copyi.asm (revision f3cfa6f6ce31685c6c4a758bc430e69eb99f50a4)
1dnl  Pentium-4 mpn_copyi -- copy limb vector, incrementing.
2
3dnl  Copyright 1999-2001 Free Software Foundation, Inc.
4
5dnl  This file is part of the GNU MP Library.
6dnl
7dnl  The GNU MP Library is free software; you can redistribute it and/or modify
8dnl  it under the terms of either:
9dnl
10dnl    * the GNU Lesser General Public License as published by the Free
11dnl      Software Foundation; either version 3 of the License, or (at your
12dnl      option) any later version.
13dnl
14dnl  or
15dnl
16dnl    * the GNU General Public License as published by the Free Software
17dnl      Foundation; either version 2 of the License, or (at your option) any
18dnl      later version.
19dnl
20dnl  or both in parallel, as here.
21dnl
22dnl  The GNU MP Library is distributed in the hope that it will be useful, but
23dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
24dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
25dnl  for more details.
26dnl
27dnl  You should have received copies of the GNU General Public License and the
28dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
29dnl  see https://www.gnu.org/licenses/.
30
31
32dnl  The rep/movsl is very slow for small blocks on pentium4.  Its startup
33dnl  time seems to be about 110 cycles.  It then copies at a rate of one
34dnl  limb per cycle.  We therefore fall back to an open-coded 2 c/l copying
35dnl  loop for smaller sizes.
36
37dnl  Ultimately, we may want to use 64-bit movd or 128-bit movdqu in some
38dnl  nifty unrolled arrangement.  Clearly, that could reach much higher
39dnl  speeds, at least for large blocks.
40
41include(`../config.m4')
42
43
44defframe(PARAM_SIZE, 12)
45defframe(PARAM_SRC, 8)
46defframe(PARAM_DST,  4)
47
48	TEXT
49	ALIGN(8)
50
51PROLOGUE(mpn_copyi)
52deflit(`FRAME',0)
53
54	movl	PARAM_SIZE, %ecx
55	cmpl	$150, %ecx
56	jg	L(replmovs)
57
58	movl	PARAM_SRC, %eax
59	movl	PARAM_DST, %edx
60	movl	%ebx, PARAM_SIZE
61	testl	%ecx, %ecx
62	jz	L(end)
63
64L(loop):
65	movl	(%eax), %ebx
66	leal	4(%eax), %eax
67	addl	$-1, %ecx
68	movl	%ebx, (%edx)
69	leal	4(%edx), %edx
70
71	jnz	L(loop)
72
73L(end):
74	movl	PARAM_SIZE, %ebx
75	ret
76
77L(replmovs):
78	cld	C better safe than sorry, see mpn/x86/README
79
80	movl	%esi, %eax
81	movl	PARAM_SRC, %esi
82	movl	%edi, %edx
83	movl	PARAM_DST, %edi
84
85	rep
86	movsl
87
88	movl	%eax, %esi
89	movl	%edx, %edi
90
91	ret
92
93EPILOGUE()
94