xref: /netbsd-src/external/lgpl3/gmp/dist/mpn/arm/v6/submul_1.asm (revision ce54336801cf28877c3414aa2fcb251dddd543a2)
1dnl  ARM mpn_submul_1.
2
3dnl  Copyright 2012, 2013 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
31include(`../config.m4')
32
33C	     cycles/limb
34C StrongARM:	 -
35C XScale	 -
36C Cortex-A7	 ?
37C Cortex-A8	 ?
38C Cortex-A9	 3.75
39C Cortex-A15	 4.0
40
41C This loop complements U on the fly,
42C   U' = B^n - 1 - U
43C and then uses that
44C   R - U*v = R + U'*v + v - B^n v
45
46C TODO
47C  * Micro-optimise feed-in code.
48C  * Optimise for n=1,2 by delaying register saving.
49C  * Try using ldm/stm.
50
51define(`rp',`r0')
52define(`up',`r1')
53define(`n', `r2')
54define(`v0',`r3')
55
56ASM_START()
57PROLOGUE(mpn_submul_1)
58	stmfd	sp!, { r4, r5, r6, r7 }
59
60	ands	r6, n, #3
61	mov	r12, v0
62	beq	L(fi0)
63	cmp	r6, #2
64	bcc	L(fi1)
65	beq	L(fi2)
66
67L(fi3):	ldr	r4, [up], #12
68	mvn	r4, r4
69	ldr	r6, [rp, #0]
70	ldr	r5, [up, #-8]
71	b	L(lo3)
72
73L(fi0):	ldr	r5, [up], #16
74	mvn	r5, r5
75	ldr	r7, [rp], #4
76	ldr	r4, [up, #-12]
77	b	L(lo0)
78
79L(fi1):	ldr	r4, [up], #4
80	mvn	r4, r4
81	ldr	r6, [rp], #8
82	subs	n, n, #1
83	beq	L(1)
84	ldr	r5, [up]
85	b	L(lo1)
86
87L(fi2):	ldr	r5, [up], #8
88	mvn	r5, r5
89	ldr	r7, [rp], #12
90	ldr	r4, [up, #-4]
91	b	L(lo2)
92
93	ALIGN(16)
94L(top):	ldr	r6, [rp, #-8]
95	ldr	r5, [up]
96	str	r7, [rp, #-12]
97L(lo1):	umaal	r6, r12, r4, v0
98	add	up, up, #16
99	mvn	r5, r5
100	ldr	r7, [rp, #-4]
101	ldr	r4, [up, #-12]
102	str	r6, [rp, #-8]
103L(lo0):	umaal	r7, r12, r5, v0
104	mvn	r4, r4
105	ldr	r6, [rp, #0]
106	ldr	r5, [up, #-8]
107	str	r7, [rp, #-4]
108L(lo3):	umaal	r6, r12, r4, v0
109	mvn	r5, r5
110	ldr	r7, [rp, #4]
111	ldr	r4, [up, #-4]
112	str	r6, [rp], #16
113L(lo2):	umaal	r7, r12, r5, v0
114	mvn	r4, r4
115	subs	n, n, #4
116	bhi	L(top)
117
118	ldr	r6, [rp, #-8]
119	str	r7, [rp, #-12]
120L(1):	umaal	r6, r12, r4, v0
121	str	r6, [rp, #-8]
122	sub	r0, v0, r12
123	ldmfd	sp!, { r4, r5, r6, r7 }
124	bx	lr
125EPILOGUE()
126