4
* Optimized RAID-5 checksumming functions for IA-64.
6
* This program is free software; you can redistribute it and/or modify
7
* it under the terms of the GNU General Public License as published by
8
* the Free Software Foundation; either version 2, or (at your option)
11
* You should have received a copy of the GNU General Public License
12
* (for example /usr/src/linux/COPYING); if not, write to the Free
13
* Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
16
#include <asm/asmmacro.h>
18
GLOBAL_ENTRY(xor_ia64_2)
22
alloc r31 = ar.pfs, 3, 0, 13, 16
40
.rotr s1[6+1], s2[6+1], d[2]
43
(p[0]) ld8.nta s1[0] = [r16], 8
44
(p[0]) ld8.nta s2[0] = [r17], 8
45
(p[6]) xor d[0] = s1[6], s2[6]
46
(p[6+1])st8.nta [r8] = d[1], 8
55
GLOBAL_ENTRY(xor_ia64_3)
59
alloc r31 = ar.pfs, 4, 0, 20, 24
78
.rotr s1[6+1], s2[6+1], s3[6+1], d[2]
81
(p[0]) ld8.nta s1[0] = [r16], 8
82
(p[0]) ld8.nta s2[0] = [r17], 8
83
(p[6]) xor d[0] = s1[6], s2[6]
85
(p[0]) ld8.nta s3[0] = [r18], 8
86
(p[6+1])st8.nta [r8] = d[1], 8
87
(p[6]) xor d[0] = d[0], s3[6]
95
GLOBAL_ENTRY(xor_ia64_4)
99
alloc r31 = ar.pfs, 5, 0, 27, 32
119
.rotr s1[6+1], s2[6+1], s3[6+1], s4[6+1], d[2]
122
(p[0]) ld8.nta s1[0] = [r16], 8
123
(p[0]) ld8.nta s2[0] = [r17], 8
124
(p[6]) xor d[0] = s1[6], s2[6]
125
(p[0]) ld8.nta s3[0] = [r18], 8
126
(p[0]) ld8.nta s4[0] = [r19], 8
127
(p[6]) xor r20 = s3[6], s4[6]
129
(p[6+1])st8.nta [r8] = d[1], 8
130
(p[6]) xor d[0] = d[0], r20
138
GLOBAL_ENTRY(xor_ia64_5)
142
alloc r31 = ar.pfs, 6, 0, 34, 40
163
.rotr s1[6+1], s2[6+1], s3[6+1], s4[6+1], s5[6+1], d[2]
166
(p[0]) ld8.nta s1[0] = [r16], 8
167
(p[0]) ld8.nta s2[0] = [r17], 8
168
(p[6]) xor d[0] = s1[6], s2[6]
169
(p[0]) ld8.nta s3[0] = [r18], 8
170
(p[0]) ld8.nta s4[0] = [r19], 8
171
(p[6]) xor r21 = s3[6], s4[6]
173
(p[0]) ld8.nta s5[0] = [r20], 8
174
(p[6+1])st8.nta [r8] = d[1], 8
175
(p[6]) xor d[0] = d[0], r21
177
(p[6]) xor d[0] = d[0], s5[6]