1 /* SPDX-License-Identifier: GPL-2.0-or-later */
5 * Optimized RAID-5 checksumming functions for IA-64.
8 #include <asm/asmmacro.h>
9 #include <asm/export.h>
11 GLOBAL_ENTRY(xor_ia64_2)
15 alloc r31 = ar.pfs, 3, 0, 13, 16
33 .rotr s1[6+1], s2[6+1], d[2]
36 (p[0]) ld8.nta s1[0] = [r16], 8
37 (p[0]) ld8.nta s2[0] = [r17], 8
38 (p[6]) xor d[0] = s1[6], s2[6]
39 (p[6+1])st8.nta [r8] = d[1], 8
47 EXPORT_SYMBOL(xor_ia64_2)
49 GLOBAL_ENTRY(xor_ia64_3)
53 alloc r31 = ar.pfs, 4, 0, 20, 24
72 .rotr s1[6+1], s2[6+1], s3[6+1], d[2]
75 (p[0]) ld8.nta s1[0] = [r16], 8
76 (p[0]) ld8.nta s2[0] = [r17], 8
77 (p[6]) xor d[0] = s1[6], s2[6]
79 (p[0]) ld8.nta s3[0] = [r18], 8
80 (p[6+1])st8.nta [r8] = d[1], 8
81 (p[6]) xor d[0] = d[0], s3[6]
88 EXPORT_SYMBOL(xor_ia64_3)
90 GLOBAL_ENTRY(xor_ia64_4)
94 alloc r31 = ar.pfs, 5, 0, 27, 32
114 .rotr s1[6+1], s2[6+1], s3[6+1], s4[6+1], d[2]
117 (p[0]) ld8.nta s1[0] = [r16], 8
118 (p[0]) ld8.nta s2[0] = [r17], 8
119 (p[6]) xor d[0] = s1[6], s2[6]
120 (p[0]) ld8.nta s3[0] = [r18], 8
121 (p[0]) ld8.nta s4[0] = [r19], 8
122 (p[6]) xor r20 = s3[6], s4[6]
124 (p[6+1])st8.nta [r8] = d[1], 8
125 (p[6]) xor d[0] = d[0], r20
132 EXPORT_SYMBOL(xor_ia64_4)
134 GLOBAL_ENTRY(xor_ia64_5)
138 alloc r31 = ar.pfs, 6, 0, 34, 40
159 .rotr s1[6+1], s2[6+1], s3[6+1], s4[6+1], s5[6+1], d[2]
162 (p[0]) ld8.nta s1[0] = [r16], 8
163 (p[0]) ld8.nta s2[0] = [r17], 8
164 (p[6]) xor d[0] = s1[6], s2[6]
165 (p[0]) ld8.nta s3[0] = [r18], 8
166 (p[0]) ld8.nta s4[0] = [r19], 8
167 (p[6]) xor r21 = s3[6], s4[6]
169 (p[0]) ld8.nta s5[0] = [r20], 8
170 (p[6+1])st8.nta [r8] = d[1], 8
171 (p[6]) xor d[0] = d[0], r21
173 (p[6]) xor d[0] = d[0], s5[6]
181 EXPORT_SYMBOL(xor_ia64_5)