vp8_memcpy_neon.asm 2.27 KB
Newer Older
John Koleszar's avatar
John Koleszar committed
1
;
2
;  Copyright (c) 2010 The WebM project authors. All Rights Reserved.
John Koleszar's avatar
John Koleszar committed
3
;
4
;  Use of this source code is governed by a BSD-style license
5
6
;  that can be found in the LICENSE file in the root of the source
;  tree. An additional intellectual property rights grant can be found
7
;  in the file PATENTS.  All contributing project authors may
8
;  be found in the AUTHORS file in the root of the source tree.
John Koleszar's avatar
John Koleszar committed
9
10
11
;


12
    EXPORT |vp8_memcpy_partial_neon|
John Koleszar's avatar
John Koleszar committed
13
14
15
16
17
18
19

    ARM
    REQUIRE8
    PRESERVE8

    AREA ||.text||, CODE, READONLY, ALIGN=2
;=========================================
20
21
22
23
;this is not a full memcpy function!!!
;void vp8_memcpy_partial_neon(unsigned char *dst_ptr, unsigned char *src_ptr,
;                             int sz);
|vp8_memcpy_partial_neon| PROC
John Koleszar's avatar
John Koleszar committed
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
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
    ;pld                [r1]                        ;preload pred data
    ;pld                [r1, #128]
    ;pld                [r1, #256]
    ;pld                [r1, #384]

    mov             r12, r2, lsr #8                 ;copy 256 bytes data at one time

memcpy_neon_loop
    vld1.8          {q0, q1}, [r1]!                 ;load src data
    subs            r12, r12, #1
    vld1.8          {q2, q3}, [r1]!
    vst1.8          {q0, q1}, [r0]!                 ;copy to dst_ptr
    vld1.8          {q4, q5}, [r1]!
    vst1.8          {q2, q3}, [r0]!
    vld1.8          {q6, q7}, [r1]!
    vst1.8          {q4, q5}, [r0]!
    vld1.8          {q8, q9}, [r1]!
    vst1.8          {q6, q7}, [r0]!
    vld1.8          {q10, q11}, [r1]!
    vst1.8          {q8, q9}, [r0]!
    vld1.8          {q12, q13}, [r1]!
    vst1.8          {q10, q11}, [r0]!
    vld1.8          {q14, q15}, [r1]!
    vst1.8          {q12, q13}, [r0]!
    vst1.8          {q14, q15}, [r0]!

    ;pld                [r1]                        ;preload pred data -- need to adjust for real device
    ;pld                [r1, #128]
    ;pld                [r1, #256]
    ;pld                [r1, #384]

    bne             memcpy_neon_loop

    ands            r3, r2, #0xff                   ;extra copy
    beq             done_copy_neon_loop

extra_copy_neon_loop
    vld1.8          {q0}, [r1]!                 ;load src data
    subs            r3, r3, #16
    vst1.8          {q0}, [r0]!
    bne             extra_copy_neon_loop

done_copy_neon_loop
    bx              lr
    ENDP

    END