1dnl  SPARC mpn_submul_1 -- Multiply a limb vector with a limb and subtract
2dnl  the result from a second limb vector.
3
4dnl  Copyright 1992, 1993, 1994, 2000 Free Software Foundation, Inc.
5
6dnl  This file is part of the GNU MP Library.
7
8dnl  The GNU MP Library is free software; you can redistribute it and/or modify
9dnl  it under the terms of the GNU Lesser General Public License as published
10dnl  by the Free Software Foundation; either version 3 of the License, or (at
11dnl  your option) any later version.
12
13dnl  The GNU MP Library is distributed in the hope that it will be useful, but
14dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
15dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public
16dnl  License for more details.
17
18dnl  You should have received a copy of the GNU Lesser General Public License
19dnl  along with the GNU MP Library.  If not, see http://www.gnu.org/licenses/.
20
21
22include(`../config.m4')
23
24C INPUT PARAMETERS
25C res_ptr	o0
26C s1_ptr	o1
27C size		o2
28C s2_limb	o3
29
30ASM_START()
31PROLOGUE(mpn_submul_1)
32	C Make S1_PTR and RES_PTR point at the end of their blocks
33	C and put (- 4 x SIZE) in index/loop counter.
34	sll	%o2,2,%o2
35	add	%o0,%o2,%o4	C RES_PTR in o4 since o0 is retval
36	add	%o1,%o2,%o1
37	sub	%g0,%o2,%o2
38
39	cmp	%o3,0xfff
40	bgu	L(large)
41	nop
42
43	ld	[%o1+%o2],%o5
44	mov	0,%o0
45	b	L(0)
46	 add	%o4,-4,%o4
47L(loop0):
48	subcc	%o5,%g1,%g1
49	ld	[%o1+%o2],%o5
50	addx	%o0,%g0,%o0
51	st	%g1,[%o4+%o2]
52L(0):	wr	%g0,%o3,%y
53	sra	%o5,31,%g2
54	and	%o3,%g2,%g2
55	andcc	%g1,0,%g1
56	mulscc	%g1,%o5,%g1
57	mulscc	%g1,%o5,%g1
58	mulscc	%g1,%o5,%g1
59	mulscc	%g1,%o5,%g1
60	mulscc	%g1,%o5,%g1
61	mulscc	%g1,%o5,%g1
62	mulscc	%g1,%o5,%g1
63	mulscc	%g1,%o5,%g1
64	mulscc	%g1,%o5,%g1
65	mulscc	%g1,%o5,%g1
66	mulscc	%g1,%o5,%g1
67	mulscc	%g1,%o5,%g1
68	mulscc	%g1,0,%g1
69	sra	%g1,20,%g4
70	sll	%g1,12,%g1
71	rd	%y,%g3
72	srl	%g3,20,%g3
73	or	%g1,%g3,%g1
74
75	addcc	%g1,%o0,%g1
76	addx	%g2,%g4,%o0	C add sign-compensation and cy to hi limb
77	addcc	%o2,4,%o2	C loop counter
78	bne	L(loop0)
79	 ld	[%o4+%o2],%o5
80
81	subcc	%o5,%g1,%g1
82	addx	%o0,%g0,%o0
83	retl
84	st	%g1,[%o4+%o2]
85
86L(large):
87	ld	[%o1+%o2],%o5
88	mov	0,%o0
89	sra	%o3,31,%g4	C g4 = mask of ones iff S2_LIMB < 0
90	b	L(1)
91	 add	%o4,-4,%o4
92L(loop):
93	subcc	%o5,%g3,%g3
94	ld	[%o1+%o2],%o5
95	addx	%o0,%g0,%o0
96	st	%g3,[%o4+%o2]
97L(1):	wr	%g0,%o5,%y
98	and	%o5,%g4,%g2
99	andcc	%g0,%g0,%g1
100	mulscc	%g1,%o3,%g1
101	mulscc	%g1,%o3,%g1
102	mulscc	%g1,%o3,%g1
103	mulscc	%g1,%o3,%g1
104	mulscc	%g1,%o3,%g1
105	mulscc	%g1,%o3,%g1
106	mulscc	%g1,%o3,%g1
107	mulscc	%g1,%o3,%g1
108	mulscc	%g1,%o3,%g1
109	mulscc	%g1,%o3,%g1
110	mulscc	%g1,%o3,%g1
111	mulscc	%g1,%o3,%g1
112	mulscc	%g1,%o3,%g1
113	mulscc	%g1,%o3,%g1
114	mulscc	%g1,%o3,%g1
115	mulscc	%g1,%o3,%g1
116	mulscc	%g1,%o3,%g1
117	mulscc	%g1,%o3,%g1
118	mulscc	%g1,%o3,%g1
119	mulscc	%g1,%o3,%g1
120	mulscc	%g1,%o3,%g1
121	mulscc	%g1,%o3,%g1
122	mulscc	%g1,%o3,%g1
123	mulscc	%g1,%o3,%g1
124	mulscc	%g1,%o3,%g1
125	mulscc	%g1,%o3,%g1
126	mulscc	%g1,%o3,%g1
127	mulscc	%g1,%o3,%g1
128	mulscc	%g1,%o3,%g1
129	mulscc	%g1,%o3,%g1
130	mulscc	%g1,%o3,%g1
131	mulscc	%g1,%o3,%g1
132	mulscc	%g1,%g0,%g1
133	rd	%y,%g3
134	addcc	%g3,%o0,%g3
135	addx	%g2,%g1,%o0
136	addcc	%o2,4,%o2
137	bne	L(loop)
138	 ld	[%o4+%o2],%o5
139
140	subcc	%o5,%g3,%g3
141	addx	%o0,%g0,%o0
142	retl
143	st	%g3,[%o4+%o2]
144EPILOGUE(mpn_submul_1)
145