summaryrefslogtreecommitdiff
path: root/libavcodec/x86/qpel.asm
blob: 4e72d5084ff6ba0757c8924d0acf1b950a545e8b (plain)
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
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
;******************************************************************************
;* SIMD-optimized quarterpel functions
;* Copyright (c) 2008 Loren Merritt
;* Copyright (c) 2003-2013 Michael Niedermayer
;* Copyright (c) 2013 Daniel Kang
;*
;* This file is part of FFmpeg.
;*
;* FFmpeg is free software; you can redistribute it and/or
;* modify it under the terms of the GNU Lesser General Public
;* License as published by the Free Software Foundation; either
;* version 2.1 of the License, or (at your option) any later version.
;*
;* FFmpeg is distributed in the hope that it will be useful,
;* but WITHOUT ANY WARRANTY; without even the implied warranty of
;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
;* Lesser General Public License for more details.
;*
;* You should have received a copy of the GNU Lesser General Public
;* License along with FFmpeg; if not, write to the Free Software
;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
;******************************************************************************

%include "libavutil/x86/x86util.asm"

SECTION .text

%macro op_avgh 3
    movh   %3, %2
    pavgb  %1, %3
    movh   %2, %1
%endmacro

%macro op_avg 2
    pavgb  %1, %2
    mova   %2, %1
%endmacro

%macro op_puth 2-3
    movh   %2, %1
%endmacro

%macro op_put 2
    mova   %2, %1
%endmacro

; void ff_put/avg_pixels4_l2_mmxext(uint8_t *dst, uint8_t *src1, uint8_t *src2,
;                                   int dstStride, int src1Stride, int h)
%macro PIXELS4_L2 1
%define OP op_%1h
cglobal %1_pixels4_l2, 6,6
    movsxdifnidn r3, r3d
    movsxdifnidn r4, r4d
    test        r5d, 1
    je        .loop
    movd         m0, [r1]
    movd         m1, [r2]
    add          r1, r4
    add          r2, 4
    pavgb        m0, m1
    OP           m0, [r0], m3
    add          r0, r3
    dec         r5d
.loop:
    mova         m0, [r1]
    mova         m1, [r1+r4]
    lea          r1, [r1+2*r4]
    pavgb        m0, [r2]
    pavgb        m1, [r2+4]
    OP           m0, [r0], m3
    OP           m1, [r0+r3], m3
    lea          r0, [r0+2*r3]
    mova         m0, [r1]
    mova         m1, [r1+r4]
    lea          r1, [r1+2*r4]
    pavgb        m0, [r2+8]
    pavgb        m1, [r2+12]
    OP           m0, [r0], m3
    OP           m1, [r0+r3], m3
    lea          r0, [r0+2*r3]
    add          r2, 16
    sub         r5d, 4
    jne       .loop
    REP_RET
%endmacro

INIT_MMX mmxext
PIXELS4_L2 put
PIXELS4_L2 avg

; void ff_put/avg_pixels8_l2_mmxext(uint8_t *dst, uint8_t *src1, uint8_t *src2,
;                                   int dstStride, int src1Stride, int h)
%macro PIXELS8_L2 1
%define OP op_%1
cglobal %1_pixels8_l2, 6,6
    movsxdifnidn r3, r3d
    movsxdifnidn r4, r4d
    test        r5d, 1
    je        .loop
    mova         m0, [r1]
    mova         m1, [r2]
    add          r1, r4
    add          r2, 8
    pavgb        m0, m1
    OP           m0, [r0]
    add          r0, r3
    dec         r5d
.loop:
    mova         m0, [r1]
    mova         m1, [r1+r4]
    lea          r1, [r1+2*r4]
    pavgb        m0, [r2]
    pavgb        m1, [r2+8]
    OP           m0, [r0]
    OP           m1, [r0+r3]
    lea          r0, [r0+2*r3]
    mova         m0, [r1]
    mova         m1, [r1+r4]
    lea          r1, [r1+2*r4]
    pavgb        m0, [r2+16]
    pavgb        m1, [r2+24]
    OP           m0, [r0]
    OP           m1, [r0+r3]
    lea          r0, [r0+2*r3]
    add          r2, 32
    sub         r5d, 4
    jne       .loop
    REP_RET
%endmacro

INIT_MMX mmxext
PIXELS8_L2 put
PIXELS8_L2 avg

; void ff_put/avg_pixels16_l2_mmxext(uint8_t *dst, uint8_t *src1, uint8_t *src2,
;                                    int dstStride, int src1Stride, int h)
%macro PIXELS16_L2 1
%define OP op_%1
cglobal %1_pixels16_l2, 6,6
    movsxdifnidn r3, r3d
    movsxdifnidn r4, r4d
    test        r5d, 1
    je        .loop
    mova         m0, [r1]
    mova         m1, [r1+8]
    pavgb        m0, [r2]
    pavgb        m1, [r2+8]
    add          r1, r4
    add          r2, 16
    OP           m0, [r0]
    OP           m1, [r0+8]
    add          r0, r3
    dec         r5d
.loop:
    mova         m0, [r1]
    mova         m1, [r1+8]
    add          r1, r4
    pavgb        m0, [r2]
    pavgb        m1, [r2+8]
    OP           m0, [r0]
    OP           m1, [r0+8]
    add          r0, r3
    mova         m0, [r1]
    mova         m1, [r1+8]
    add          r1, r4
    pavgb        m0, [r2+16]
    pavgb        m1, [r2+24]
    OP           m0, [r0]
    OP           m1, [r0+8]
    add          r0, r3
    add          r2, 32
    sub         r5d, 2
    jne       .loop
    REP_RET
%endmacro

INIT_MMX mmxext
PIXELS16_L2 put
PIXELS16_L2 avg