1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
| /* SPDX-License-Identifier: GPL-2.0 */
| /*
| * arch/alpha/lib/ev67-strlen.S
| * 21264 version by Rick Gorton <rick.gorton@alpha-processor.com>
| *
| * Finds length of a 0-terminated string. Optimized for the
| * Alpha architecture:
| *
| * - memory accessed as aligned quadwords only
| * - uses bcmpge to compare 8 bytes in parallel
| *
| * Much of the information about 21264 scheduling/coding comes from:
| * Compiler Writer's Guide for the Alpha 21264
| * abbreviated as 'CWG' in other comments here
| * ftp.digital.com/pub/Digital/info/semiconductor/literature/dsc-library.html
| * Scheduling notation:
| * E - either cluster
| * U - upper subcluster; U0 - subcluster U0; U1 - subcluster U1
| * L - lower subcluster; L0 - subcluster L0; L1 - subcluster L1
| */
| #include <asm/export.h>
| .set noreorder
| .set noat
|
| .globl strlen
| .ent strlen
| .align 4
| strlen:
| ldq_u $1, 0($16) # L : load first quadword ($16 may be misaligned)
| lda $2, -1($31) # E :
| insqh $2, $16, $2 # U :
| andnot $16, 7, $0 # E :
|
| or $2, $1, $1 # E :
| cmpbge $31, $1, $2 # E : $2 <- bitmask: bit i == 1 <==> i-th byte == 0
| nop # E :
| bne $2, $found # U :
|
| $loop: ldq $1, 8($0) # L :
| addq $0, 8, $0 # E : addr += 8
| cmpbge $31, $1, $2 # E :
| beq $2, $loop # U :
|
| $found:
| cttz $2, $3 # U0 :
| addq $0, $3, $0 # E :
| subq $0, $16, $0 # E :
| ret $31, ($26) # L0 :
|
| .end strlen
| EXPORT_SYMBOL(strlen)
|
|
