xref: /netbsd-src/external/lgpl3/gmp/dist/mpn/pa32/hppa1_1/addmul_1.asm (revision ce54336801cf28877c3414aa2fcb251dddd543a2)
1dnl  HP-PA 1.1 mpn_addmul_1 -- Multiply a limb vector with a limb and add the
2dnl  result to a second limb vector.
3
4dnl  Copyright 1992-1994, 2000-2002 Free Software Foundation, Inc.
5
6dnl  This file is part of the GNU MP Library.
7dnl
8dnl  The GNU MP Library is free software; you can redistribute it and/or modify
9dnl  it under the terms of either:
10dnl
11dnl    * the GNU Lesser General Public License as published by the Free
12dnl      Software Foundation; either version 3 of the License, or (at your
13dnl      option) any later version.
14dnl
15dnl  or
16dnl
17dnl    * the GNU General Public License as published by the Free Software
18dnl      Foundation; either version 2 of the License, or (at your option) any
19dnl      later version.
20dnl
21dnl  or both in parallel, as here.
22dnl
23dnl  The GNU MP Library is distributed in the hope that it will be useful, but
24dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
25dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
26dnl  for more details.
27dnl
28dnl  You should have received copies of the GNU General Public License and the
29dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
30dnl  see https://www.gnu.org/licenses/.
31
32include(`../config.m4')
33
34C INPUT PARAMETERS
35C res_ptr	r26
36C s1_ptr	r25
37C size		r24
38C s2_limb	r23
39
40C This runs at 11 cycles/limb on a PA7000.  With the used instructions, it can
41C not become faster due to data cache contention after a store.  On the PA7100
42C it runs at 10 cycles/limb.
43
44C There are some ideas described in mul_1.asm that applies to this code too.
45
46ASM_START()
47PROLOGUE(mpn_addmul_1)
48C	.callinfo	frame=64,no_calls
49
50	ldo		64(%r30),%r30
51	fldws,ma	4(%r25),%fr5
52	stw		%r23,-16(%r30)		C move s2_limb ...
53	addib,=		-1,%r24,L(just_one_limb)
54	 fldws		-16(%r30),%fr4		C ... into fr4
55	add		%r0,%r0,%r0		C clear carry
56	xmpyu		%fr4,%fr5,%fr6
57	fldws,ma	4(%r25),%fr7
58	fstds		%fr6,-16(%r30)
59	xmpyu		%fr4,%fr7,%fr8
60	ldw		-12(%r30),%r19		C least significant limb in product
61	ldw		-16(%r30),%r28
62
63	fstds		%fr8,-16(%r30)
64	addib,=		-1,%r24,L(end)
65	 ldw		-12(%r30),%r1
66
67C Main loop
68LDEF(loop)
69	ldws		0(%r26),%r29
70	fldws,ma	4(%r25),%fr5
71	add		%r29,%r19,%r19
72	stws,ma		%r19,4(%r26)
73	addc		%r28,%r1,%r19
74	xmpyu		%fr4,%fr5,%fr6
75	ldw		-16(%r30),%r28
76	fstds		%fr6,-16(%r30)
77	addc		%r0,%r28,%r28
78	addib,<>	-1,%r24,L(loop)
79	 ldw		-12(%r30),%r1
80
81LDEF(end)
82	ldw		0(%r26),%r29
83	add		%r29,%r19,%r19
84	stws,ma		%r19,4(%r26)
85	addc		%r28,%r1,%r19
86	ldw		-16(%r30),%r28
87	ldws		0(%r26),%r29
88	addc		%r0,%r28,%r28
89	add		%r29,%r19,%r19
90	stws,ma		%r19,4(%r26)
91	addc		%r0,%r28,%r28
92	bv		0(%r2)
93	 ldo		-64(%r30),%r30
94
95LDEF(just_one_limb)
96	xmpyu		%fr4,%fr5,%fr6
97	ldw		0(%r26),%r29
98	fstds		%fr6,-16(%r30)
99	ldw		-12(%r30),%r1
100	ldw		-16(%r30),%r28
101	add		%r29,%r1,%r19
102	stw		%r19,0(%r26)
103	addc		%r0,%r28,%r28
104	bv		0(%r2)
105	 ldo		-64(%r30),%r30
106EPILOGUE()
107