1dnl  Alpha mpn_add_n -- Add two limb vectors of the same length > 0 and
2dnl  store sum in a third limb vector.
3
4dnl  Copyright 1995, 1999, 2000, 2005, 2011 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      cycles/limb
35C EV4:     ?
36C EV5:     4.75
37C EV6:     3
38
39dnl  INPUT PARAMETERS
40dnl  res_ptr	r16
41dnl  s1_ptr	r17
42dnl  s2_ptr	r18
43dnl  size	r19
44
45ASM_START()
46PROLOGUE(mpn_add_nc)
47	bis	r20,r31,r25
48	br	L(com)
49EPILOGUE()
50PROLOGUE(mpn_add_n)
51	bis	r31,r31,r25		C clear cy
52L(com):	subq	r19,4,r19		C decr loop cnt
53	blt	r19,$Lend2		C if less than 4 limbs, goto 2nd loop
54C Start software pipeline for 1st loop
55	ldq	r0,0(r18)
56	ldq	r4,0(r17)
57	ldq	r1,8(r18)
58	ldq	r5,8(r17)
59	addq	r17,32,r17		C update s1_ptr
60	addq	r0,r4,r28		C 1st main add
61	ldq	r2,16(r18)
62	addq	r25,r28,r20		C 1st carry add
63	ldq	r3,24(r18)
64	cmpult	r28,r4,r8		C compute cy from last add
65	ldq	r6,-16(r17)
66	cmpult	r20,r28,r25		C compute cy from last add
67	ldq	r7,-8(r17)
68	bis	r8,r25,r25		C combine cy from the two adds
69	subq	r19,4,r19		C decr loop cnt
70	addq	r1,r5,r28		C 2nd main add
71	addq	r18,32,r18		C update s2_ptr
72	addq	r28,r25,r21		C 2nd carry add
73	cmpult	r28,r5,r8		C compute cy from last add
74	blt	r19,$Lend1		C if less than 4 limbs remain, jump
75C 1st loop handles groups of 4 limbs in a software pipeline
76	ALIGN(16)
77$Loop:	cmpult	r21,r28,r25		C compute cy from last add
78	ldq	r0,0(r18)
79	bis	r8,r25,r25		C combine cy from the two adds
80	ldq	r1,8(r18)
81	addq	r2,r6,r28		C 3rd main add
82	ldq	r4,0(r17)
83	addq	r28,r25,r22		C 3rd carry add
84	ldq	r5,8(r17)
85	cmpult	r28,r6,r8		C compute cy from last add
86	cmpult	r22,r28,r25		C compute cy from last add
87	stq	r20,0(r16)
88	bis	r8,r25,r25		C combine cy from the two adds
89	stq	r21,8(r16)
90	addq	r3,r7,r28		C 4th main add
91	addq	r28,r25,r23		C 4th carry add
92	cmpult	r28,r7,r8		C compute cy from last add
93	cmpult	r23,r28,r25		C compute cy from last add
94		addq	r17,32,r17		C update s1_ptr
95	bis	r8,r25,r25		C combine cy from the two adds
96		addq	r16,32,r16		C update res_ptr
97	addq	r0,r4,r28		C 1st main add
98	ldq	r2,16(r18)
99	addq	r25,r28,r20		C 1st carry add
100	ldq	r3,24(r18)
101	cmpult	r28,r4,r8		C compute cy from last add
102	ldq	r6,-16(r17)
103	cmpult	r20,r28,r25		C compute cy from last add
104	ldq	r7,-8(r17)
105	bis	r8,r25,r25		C combine cy from the two adds
106	subq	r19,4,r19		C decr loop cnt
107	stq	r22,-16(r16)
108	addq	r1,r5,r28		C 2nd main add
109	stq	r23,-8(r16)
110	addq	r25,r28,r21		C 2nd carry add
111		addq	r18,32,r18		C update s2_ptr
112	cmpult	r28,r5,r8		C compute cy from last add
113	bge	r19,$Loop
114C Finish software pipeline for 1st loop
115$Lend1:	cmpult	r21,r28,r25		C compute cy from last add
116	bis	r8,r25,r25		C combine cy from the two adds
117	addq	r2,r6,r28		C 3rd main add
118	addq	r28,r25,r22		C 3rd carry add
119	cmpult	r28,r6,r8		C compute cy from last add
120	cmpult	r22,r28,r25		C compute cy from last add
121	stq	r20,0(r16)
122	bis	r8,r25,r25		C combine cy from the two adds
123	stq	r21,8(r16)
124	addq	r3,r7,r28		C 4th main add
125	addq	r28,r25,r23		C 4th carry add
126	cmpult	r28,r7,r8		C compute cy from last add
127	cmpult	r23,r28,r25		C compute cy from last add
128	bis	r8,r25,r25		C combine cy from the two adds
129	addq	r16,32,r16		C update res_ptr
130	stq	r22,-16(r16)
131	stq	r23,-8(r16)
132$Lend2:	addq	r19,4,r19		C restore loop cnt
133	beq	r19,$Lret
134C Start software pipeline for 2nd loop
135	ldq	r0,0(r18)
136	ldq	r4,0(r17)
137	subq	r19,1,r19
138	beq	r19,$Lend0
139C 2nd loop handles remaining 1-3 limbs
140	ALIGN(16)
141$Loop0:	addq	r0,r4,r28		C main add
142	ldq	r0,8(r18)
143	cmpult	r28,r4,r8		C compute cy from last add
144	ldq	r4,8(r17)
145	addq	r28,r25,r20		C carry add
146	addq	r18,8,r18
147	addq	r17,8,r17
148	stq	r20,0(r16)
149	cmpult	r20,r28,r25		C compute cy from last add
150	subq	r19,1,r19		C decr loop cnt
151	bis	r8,r25,r25		C combine cy from the two adds
152	addq	r16,8,r16
153	bne	r19,$Loop0
154$Lend0:	addq	r0,r4,r28		C main add
155	addq	r28,r25,r20		C carry add
156	cmpult	r28,r4,r8		C compute cy from last add
157	cmpult	r20,r28,r25		C compute cy from last add
158	stq	r20,0(r16)
159	bis	r8,r25,r25		C combine cy from the two adds
160
161$Lret:	bis	r25,r31,r0		C return cy
162	ret	r31,(r26),1
163EPILOGUE()
164ASM_END()
165