summaryrefslogtreecommitdiffstats
path: root/apps/plugins/mpegplayer/libmpeg2/motion_comp_arm_s.S
blob: 1ec1b0686ec6178056551b3eed5f6459e03692b6 (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
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
@ motion_comp_arm_s.S
@ Copyright (C) 2004 AGAWA Koji <i (AT) atty (DOT) jp>
@
@ This file is part of mpeg2dec, a free MPEG-2 video stream decoder.
@ See http://libmpeg2.sourceforge.net/ for updates.
@
@ mpeg2dec is free software; you can redistribute it and/or modify
@ it under the terms of the GNU General Public License as published by
@ the Free Software Foundation; either version 2 of the License, or
@ (at your option) any later version.
@
@ mpeg2dec 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 General Public License for more details.
@
@ You should have received a copy of the GNU General Public License
@ along with this program; if not, write to the Free Software
@ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
@
@ $Id$

#include "config.h" /* Rockbox: ARM architecture version */

        .text
        
@ ----------------------------------------------------------------
        .align
        .global MC_put_o_16
MC_put_o_16:
        @@ void func(uint8_t * dest, const uint8_t * ref, int stride, int height)
        @@ pld [r1]
        stmfd sp!, {r4-r7, lr} @ R14 is also called LR
        and r4, r1, #3
        ldr pc, [pc, r4, lsl #2]
        .word 0
        .word MC_put_o_16_align0
        .word MC_put_o_16_align1
        .word MC_put_o_16_align2
        .word MC_put_o_16_align3

MC_put_o_16_align0:
        ldmia r1, {r4-r7}
        add r1, r1, r2
        @@ pld [r1]
        stmia r0, {r4-r7}
        subs r3, r3, #1
        add r0, r0, r2
        bne MC_put_o_16_align0
        ldmpc regs=r4-r7 @@ update PC with LR content.

.macro  ADJ_ALIGN_QW shift, R0, R1, R2, R3, R4
        mov \R0, \R0, lsr #(\shift)
        orr \R0, \R0, \R1, lsl #(32 - \shift)
        mov \R1, \R1, lsr #(\shift)
        orr \R1, \R1, \R2, lsl #(32 - \shift)
        mov \R2, \R2, lsr #(\shift)
        orr \R2, \R2, \R3, lsl #(32 - \shift)
        mov \R3, \R3, lsr #(\shift)
        orr \R3, \R3, \R4, lsl #(32 - \shift)
        mov \R4, \R4, lsr #(\shift)
.endm

MC_put_o_16_align1:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r4-r7, r12}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_QW 8, r4, r5, r6, r7, r12
        stmia r0, {r4-r7}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r4-r7 @@ update PC with LR content.

MC_put_o_16_align2:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r4-r7, r12}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_QW 16, r4, r5, r6, r7, r12
        stmia r0, {r4-r7}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r4-r7 @@ update PC with LR content.

MC_put_o_16_align3:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r4-r7, r12}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_QW 24, r4, r5, r6, r7, r12
        stmia r0, {r4-r7}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r4-r7 @@ update PC with LR content.

@ ----------------------------------------------------------------
        .align
        .global MC_put_o_8
MC_put_o_8:
        @@ void func(uint8_t * dest, const uint8_t * ref, int stride, int height)
        @@ pld [r1]
        stmfd sp!, {r4, r5, lr} @ R14 is also called LR
        and r4, r1, #3
        ldr pc, [pc, r4, lsl #2]
        .word 0
        .word MC_put_o_8_align0
        .word MC_put_o_8_align1
        .word MC_put_o_8_align2
        .word MC_put_o_8_align3

MC_put_o_8_align0:
        ldmia r1, {r4, r5}
        add r1, r1, r2
        @@ pld [r1]
        stmia r0, {r4, r5}
        add r0, r0, r2
        subs r3, r3, #1
        bne MC_put_o_8_align0
        ldmpc regs=r4-r5 @@ update PC with LR content.

.macro  ADJ_ALIGN_DW shift, R0, R1, R2
        mov \R0, \R0, lsr #(\shift)
        orr \R0, \R0, \R1, lsl #(32 - \shift)
        mov \R1, \R1, lsr #(\shift)
        orr \R1, \R1, \R2, lsl #(32 - \shift)
        mov \R2, \R2, lsr #(\shift)
.endm

MC_put_o_8_align1:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r4, r5, r12}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_DW 8, r4, r5, r12
        stmia r0, {r4, r5}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r4-r5 @@ update PC with LR content.

MC_put_o_8_align2:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r4, r5, r12}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_DW 16, r4, r5, r12
        stmia r0, {r4, r5}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r4-r5 @@ update PC with LR content.

MC_put_o_8_align3:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r4, r5, r12}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_DW 24, r4, r5, r12
        stmia r0, {r4, r5}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r4-r5 @@ update PC with LR content.

@ ----------------------------------------------------------------
.macro  AVG_PW rW1, rW2
        mov \rW2, \rW2, lsl #24
        orr \rW2, \rW2, \rW1, lsr #8
        eor r9, \rW1, \rW2
#if ARM_ARCH >= 6
        uhadd8 \rW2, \rW1, \rW2
#else
        and \rW2, \rW1, \rW2
        and r10, r9, r11
        add \rW2, \rW2, r10, lsr #1
#endif
        and r9, r9, r12
        add \rW2, \rW2, r9
.endm

#if ARM_ARCH >= 6
#define HIGHEST_REG r9
#else
#define HIGHEST_REG r11
#endif

        .align
        .global MC_put_x_16
MC_put_x_16:
        @@ void func(uint8_t * dest, const uint8_t * ref, int stride, int height)
        @@ pld [r1]
        stmfd sp!, {r4-HIGHEST_REG, lr} @ R14 is also called LR
        and r4, r1, #3
        ldr r12, 2f
#if ARM_ARCH < 6
        mvn r11, r12
#endif
        ldr pc, [pc, r4, lsl #2]
2:      .word 0x01010101
        .word MC_put_x_16_align0
        .word MC_put_x_16_align1
        .word MC_put_x_16_align2
        .word MC_put_x_16_align3

MC_put_x_16_align0:
        ldmia r1, {r4-r8}
        add r1, r1, r2
        @@ pld [r1]
        AVG_PW r7, r8
        AVG_PW r6, r7
        AVG_PW r5, r6
        AVG_PW r4, r5
        stmia r0, {r5-r8}
        subs r3, r3, #1
        add r0, r0, r2
        bne MC_put_x_16_align0
        ldmpc regs=r4-HIGHEST_REG @@ update PC with LR content.

MC_put_x_16_align1:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r4-r8}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_QW 8, r4, r5, r6, r7, r8
        AVG_PW r7, r8
        AVG_PW r6, r7
        AVG_PW r5, r6
        AVG_PW r4, r5
        stmia r0, {r5-r8}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r4-HIGHEST_REG @@ update PC with LR content.

MC_put_x_16_align2:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r4-r8}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_QW 16, r4, r5, r6, r7, r8
        AVG_PW r7, r8
        AVG_PW r6, r7
        AVG_PW r5, r6
        AVG_PW r4, r5
        stmia r0, {r5-r8}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r4-HIGHEST_REG @@ update PC with LR content.

MC_put_x_16_align3:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r4-r8}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_QW 24, r4, r5, r6, r7, r8
        AVG_PW r7, r8
        AVG_PW r6, r7
        AVG_PW r5, r6
        AVG_PW r4, r5
        stmia r0, {r5-r8}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r4-HIGHEST_REG @@ update PC with LR content.

@ ----------------------------------------------------------------
        .align
        .global MC_put_x_8
MC_put_x_8:
        @@ void func(uint8_t * dest, const uint8_t * ref, int stride, int height)
        @@ pld [r1]
        stmfd sp!, {r6-HIGHEST_REG, lr} @ R14 is also called LR
        and r6, r1, #3
        ldr r12, 2f
#if ARM_ARCH < 6
        mvn r11, r12
#endif
        ldr pc, [pc, r6, lsl #2]
2:      .word 0x01010101
        .word MC_put_x_8_align0
        .word MC_put_x_8_align1
        .word MC_put_x_8_align2
        .word MC_put_x_8_align3

MC_put_x_8_align0:
        ldmia r1, {r6-r8}
        add r1, r1, r2
        @@ pld [r1]
        AVG_PW r7, r8
        AVG_PW r6, r7
        stmia r0, {r7-r8}
        subs r3, r3, #1
        add r0, r0, r2
        bne MC_put_x_8_align0
        ldmpc regs=r6-HIGHEST_REG @@ update PC with LR content.

MC_put_x_8_align1:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r6-r8}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_DW 8, r6, r7, r8
        AVG_PW r7, r8
        AVG_PW r6, r7
        stmia r0, {r7-r8}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r6-HIGHEST_REG @@ update PC with LR content.

MC_put_x_8_align2:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r6-r8}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_DW 16, r6, r7, r8
        AVG_PW r7, r8
        AVG_PW r6, r7
        stmia r0, {r7-r8}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r6-HIGHEST_REG @@ update PC with LR content.

MC_put_x_8_align3:
        and r1, r1, #0xFFFFFFFC
1:      ldmia r1, {r6-r8}
        add r1, r1, r2
        @@ pld [r1]
        ADJ_ALIGN_DW 24, r6, r7, r8
        AVG_PW r7, r8
        AVG_PW r6, r7
        stmia r0, {r7-r8}
        subs r3, r3, #1
        add r0, r0, r2
        bne 1b
        ldmpc regs=r6-HIGHEST_REG @@ update PC with LR content.