aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/arm/h264cmc_neon.S
blob: 779dc0bba6eb8ef9d21f38deeddd3834f5e7d31e (plain) (blame)
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
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
/*
 * Copyright (c) 2008 Mans Rullgard <mans@mansr.com>
 *
 * This file is part of Libav.
 *
 * Libav 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.
 *
 * Libav 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 Libav; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

#include "libavutil/arm/asm.S"

/* chroma_mc8(uint8_t *dst, uint8_t *src, ptrdiff_t stride, int h, int x, int y) */
.macro  h264_chroma_mc8 type, codec=h264
function ff_\type\()_\codec\()_chroma_mc8_neon, export=1
        push            {r4-r7, lr}
        ldrd            r4,  r5,  [sp, #20]
  .ifc \type,avg
        mov             lr,  r0
  .endif
        pld             [r1]
        pld             [r1, r2]

  .ifc \codec,rv40
        movrel          r6,  rv40bias
        lsr             r7,  r5,  #1
        add             r6,  r6,  r7,  lsl #3
        lsr             r7,  r4,  #1
        add             r6,  r6,  r7,  lsl #1
        vld1.16         {d22[],d23[]}, [r6,:16]
  .endif
  .ifc \codec,vc1
        vmov.u16        q11, #28
  .endif

A       muls            r7,  r4,  r5
T       mul             r7,  r4,  r5
T       cmp             r7,  #0
        rsb             r6,  r7,  r5,  lsl #3
        rsb             r12, r7,  r4,  lsl #3
        sub             r4,  r7,  r4,  lsl #3
        sub             r4,  r4,  r5,  lsl #3
        add             r4,  r4,  #64

        beq             2f

        vdup.8          d0,  r4
        vdup.8          d1,  r12
        vld1.8          {d4, d5}, [r1], r2
        vdup.8          d2,  r6
        vdup.8          d3,  r7
        vext.8          d5,  d4,  d5,  #1

1:      vld1.8          {d6, d7}, [r1], r2
        vmull.u8        q8,  d4,  d0
        vmlal.u8        q8,  d5,  d1
        vext.8          d7,  d6,  d7,  #1
        vld1.8          {d4, d5}, [r1], r2
        vmlal.u8        q8,  d6,  d2
        pld             [r1]
        vext.8          d5,  d4,  d5,  #1
        vmlal.u8        q8,  d7,  d3
        vmull.u8        q9,  d6,  d0
        subs            r3,  r3,  #2
        vmlal.u8        q9,  d7,  d1
        vmlal.u8        q9,  d4,  d2
        vmlal.u8        q9,  d5,  d3
        pld             [r1, r2]
  .ifc \codec,h264
        vrshrn.u16      d16, q8,  #6
        vrshrn.u16      d17, q9,  #6
  .else
        vadd.u16        q8,  q8,  q11
        vadd.u16        q9,  q9,  q11
        vshrn.u16       d16, q8,  #6
        vshrn.u16       d17, q9,  #6
  .endif
  .ifc \type,avg
        vld1.8          {d20}, [lr,:64], r2
        vld1.8          {d21}, [lr,:64], r2
        vrhadd.u8       q8,  q8,  q10
  .endif
        vst1.8          {d16}, [r0,:64], r2
        vst1.8          {d17}, [r0,:64], r2
        bgt             1b

        pop             {r4-r7, pc}

2:      adds            r12, r12, r6
        vdup.8          d0,  r4
        beq             5f
        tst             r6,  r6
        vdup.8          d1,  r12

        beq             4f

        vld1.8          {d4}, [r1], r2

3:      vld1.8          {d6}, [r1], r2
        vmull.u8        q8,  d4,  d0
        vmlal.u8        q8,  d6,  d1
        vld1.8          {d4}, [r1], r2
        vmull.u8        q9,  d6,  d0
        vmlal.u8        q9,  d4,  d1
        pld             [r1]
  .ifc \codec,h264
        vrshrn.u16      d16, q8,  #6
        vrshrn.u16      d17, q9,  #6
  .else
        vadd.u16        q8,  q8,  q11
        vadd.u16        q9,  q9,  q11
        vshrn.u16       d16, q8,  #6
        vshrn.u16       d17, q9,  #6
  .endif
        pld             [r1, r2]
  .ifc \type,avg
        vld1.8          {d20}, [lr,:64], r2
        vld1.8          {d21}, [lr,:64], r2
        vrhadd.u8       q8,  q8,  q10
  .endif
        subs            r3,  r3,  #2
        vst1.8          {d16}, [r0,:64], r2
        vst1.8          {d17}, [r0,:64], r2
        bgt             3b

        pop             {r4-r7, pc}

4:      vld1.8          {d4, d5}, [r1], r2
        vld1.8          {d6, d7}, [r1], r2
        vext.8          d5,  d4,  d5,  #1
        vext.8          d7,  d6,  d7,  #1
        pld             [r1]
        subs            r3,  r3,  #2
        vmull.u8        q8,  d4,  d0
        vmlal.u8        q8,  d5,  d1
        vmull.u8        q9,  d6,  d0
        vmlal.u8        q9,  d7,  d1
        pld             [r1, r2]
  .ifc \codec,h264
        vrshrn.u16      d16, q8,  #6
        vrshrn.u16      d17, q9,  #6
  .else
        vadd.u16        q8,  q8,  q11
        vadd.u16        q9,  q9,  q11
        vshrn.u16       d16, q8,  #6
        vshrn.u16       d17, q9,  #6
  .endif
  .ifc \type,avg
        vld1.8          {d20}, [lr,:64], r2
        vld1.8          {d21}, [lr,:64], r2
        vrhadd.u8       q8,  q8,  q10
  .endif
        vst1.8          {d16}, [r0,:64], r2
        vst1.8          {d17}, [r0,:64], r2
        bgt             4b

        pop             {r4-r7, pc}

5:      vld1.8          {d4}, [r1], r2
        vld1.8          {d5}, [r1], r2
        pld             [r1]
        subs            r3,  r3,  #2
        vmull.u8        q8,  d4,  d0
        vmull.u8        q9,  d5,  d0
        pld             [r1, r2]
  .ifc \codec,h264
        vrshrn.u16      d16, q8,  #6
        vrshrn.u16      d17, q9,  #6
  .else
        vadd.u16        q8,  q8,  q11
        vadd.u16        q9,  q9,  q11
        vshrn.u16       d16, q8,  #6
        vshrn.u16       d17, q9,  #6
  .endif
  .ifc \type,avg
        vld1.8          {d20}, [lr,:64], r2
        vld1.8          {d21}, [lr,:64], r2
        vrhadd.u8       q8,  q8,  q10
  .endif
        vst1.8          {d16}, [r0,:64], r2
        vst1.8          {d17}, [r0,:64], r2
        bgt             5b

        pop             {r4-r7, pc}
endfunc
.endm

/* chroma_mc4(uint8_t *dst, uint8_t *src, ptrdiff_t stride, int h, int x, int y) */
.macro  h264_chroma_mc4 type, codec=h264
function ff_\type\()_\codec\()_chroma_mc4_neon, export=1
        push            {r4-r7, lr}
        ldrd            r4,  r5,  [sp, #20]
  .ifc \type,avg
        mov             lr,  r0
  .endif
        pld             [r1]
        pld             [r1, r2]

  .ifc \codec,rv40
        movrel          r6,  rv40bias
        lsr             r7,  r5,  #1
        add             r6,  r6,  r7,  lsl #3
        lsr             r7,  r4,  #1
        add             r6,  r6,  r7,  lsl #1
        vld1.16         {d22[],d23[]}, [r6,:16]
  .endif
  .ifc \codec,vc1
        vmov.u16        q11, #28
  .endif

A       muls            r7,  r4,  r5
T       mul             r7,  r4,  r5
T       cmp             r7,  #0
        rsb             r6,  r7,  r5,  lsl #3
        rsb             r12, r7,  r4,  lsl #3
        sub             r4,  r7,  r4,  lsl #3
        sub             r4,  r4,  r5,  lsl #3
        add             r4,  r4,  #64

        beq             2f

        vdup.8          d0,  r4
        vdup.8          d1,  r12
        vld1.8          {d4},     [r1], r2
        vdup.8          d2,  r6
        vdup.8          d3,  r7

        vext.8          d5,  d4,  d5,  #1
        vtrn.32         d4,  d5

        vtrn.32         d0,  d1
        vtrn.32         d2,  d3

1:      vld1.8          {d6},     [r1], r2
        vext.8          d7,  d6,  d7,  #1
        vtrn.32         d6,  d7
        vmull.u8        q8,  d4,  d0
        vmlal.u8        q8,  d6,  d2
        vld1.8          {d4},     [r1], r2
        vext.8          d5,  d4,  d5,  #1
        vtrn.32         d4,  d5
        pld             [r1]
        vmull.u8        q9,  d6,  d0
        vmlal.u8        q9,  d4,  d2
        vadd.i16        d16, d16, d17
        vadd.i16        d17, d18, d19
  .ifc \codec,h264
        vrshrn.u16      d16, q8,  #6
  .else
        vadd.u16        q8,  q8,  q11
        vshrn.u16       d16, q8,  #6
  .endif
        subs            r3,  r3,  #2
        pld             [r1, r2]
  .ifc \type,avg
        vld1.32         {d20[0]}, [lr,:32], r2
        vld1.32         {d20[1]}, [lr,:32], r2
        vrhadd.u8       d16, d16, d20
  .endif
        vst1.32         {d16[0]}, [r0,:32], r2
        vst1.32         {d16[1]}, [r0,:32], r2
        bgt             1b

        pop             {r4-r7, pc}

2:      adds            r12, r12, r6
        vdup.8          d0,  r4
        beq             5f
        tst             r6,  r6
        vdup.8          d1,  r12
        vtrn.32         d0,  d1

        beq             4f

        vext.32         d1,  d0,  d1,  #1
        vld1.32         {d4[0]},  [r1], r2

3:      vld1.32         {d4[1]},  [r1], r2
        vmull.u8        q8,  d4,  d0
        vld1.32         {d4[0]},  [r1], r2
        vmull.u8        q9,  d4,  d1
        vadd.i16        d16, d16, d17
        vadd.i16        d17, d18, d19
        pld             [r1]
  .ifc \codec,h264
        vrshrn.u16      d16, q8,  #6
  .else
        vadd.u16        q8,  q8,  q11
        vshrn.u16       d16, q8,  #6
  .endif
  .ifc \type,avg
        vld1.32         {d20[0]}, [lr,:32], r2
        vld1.32         {d20[1]}, [lr,:32], r2
        vrhadd.u8       d16, d16, d20
  .endif
        subs            r3,  r3,  #2
        pld             [r1, r2]
        vst1.32         {d16[0]}, [r0,:32], r2
        vst1.32         {d16[1]}, [r0,:32], r2
        bgt             3b

        pop             {r4-r7, pc}

4:      vld1.8          {d4},     [r1], r2
        vld1.8          {d6},     [r1], r2
        vext.8          d5,  d4,  d5,  #1
        vext.8          d7,  d6,  d7,  #1
        vtrn.32         d4,  d5
        vtrn.32         d6,  d7
        vmull.u8        q8,  d4,  d0
        vmull.u8        q9,  d6,  d0
        subs            r3,  r3,  #2
        vadd.i16        d16, d16, d17
        vadd.i16        d17, d18, d19
        pld             [r1]
  .ifc \codec,h264
        vrshrn.u16      d16, q8,  #6
  .else
        vadd.u16        q8,  q8,  q11
        vshrn.u16       d16, q8,  #6
  .endif
  .ifc \type,avg
        vld1.32         {d20[0]}, [lr,:32], r2
        vld1.32         {d20[1]}, [lr,:32], r2
        vrhadd.u8       d16, d16, d20
  .endif
        pld             [r1]
        vst1.32         {d16[0]}, [r0,:32], r2
        vst1.32         {d16[1]}, [r0,:32], r2
        bgt             4b

        pop             {r4-r7, pc}

5:      vld1.32         {d4[0]},  [r1], r2
        vld1.32         {d4[1]},  [r1], r2
        vmull.u8        q8,  d4,  d0
        subs            r3,  r3,  #2
        pld             [r1]
  .ifc \codec,h264
        vrshrn.u16      d16, q8,  #6
  .else
        vadd.u16        q8,  q8,  q11
        vshrn.u16       d16, q8,  #6
  .endif
  .ifc \type,avg
        vld1.32         {d20[0]}, [lr,:32], r2
        vld1.32         {d20[1]}, [lr,:32], r2
        vrhadd.u8       d16, d16, d20
  .endif
        pld             [r1]
        vst1.32         {d16[0]}, [r0,:32], r2
        vst1.32         {d16[1]}, [r0,:32], r2
        bgt             5b

        pop             {r4-r7, pc}
endfunc
.endm

.macro  h264_chroma_mc2 type
function ff_\type\()_h264_chroma_mc2_neon, export=1
        push            {r4-r6, lr}
        ldr             r4,  [sp, #16]
        ldr             lr,  [sp, #20]
        pld             [r1]
        pld             [r1, r2]
        orrs            r5,  r4,  lr
        beq             2f

        mul             r5,  r4,  lr
        rsb             r6,  r5,  lr,  lsl #3
        rsb             r12, r5,  r4,  lsl #3
        sub             r4,  r5,  r4,  lsl #3
        sub             r4,  r4,  lr,  lsl #3
        add             r4,  r4,  #64
        vdup.8          d0,  r4
        vdup.8          d2,  r12
        vdup.8          d1,  r6
        vdup.8          d3,  r5
        vtrn.16         q0,  q1
1:
        vld1.32         {d4[0]},  [r1], r2
        vld1.32         {d4[1]},  [r1], r2
        vrev64.32       d5,  d4
        vld1.32         {d5[1]},  [r1]
        vext.8          q3,  q2,  q2,  #1
        vtrn.16         q2,  q3
        vmull.u8        q8,  d4,  d0
        vmlal.u8        q8,  d5,  d1
  .ifc \type,avg
        vld1.16         {d18[0]}, [r0,:16], r2
        vld1.16         {d18[1]}, [r0,:16]
        sub             r0,  r0,  r2
  .endif
        vtrn.32         d16, d17
        vadd.i16        d16, d16, d17
        vrshrn.u16      d16, q8,  #6
  .ifc \type,avg
        vrhadd.u8       d16, d16, d18
  .endif
        vst1.16         {d16[0]}, [r0,:16], r2
        vst1.16         {d16[1]}, [r0,:16], r2
        subs            r3,  r3,  #2
        bgt             1b
        pop             {r4-r6, pc}
2:
  .ifc \type,put
        ldrh_post       r5,  r1,  r2
        strh_post       r5,  r0,  r2
        ldrh_post       r6,  r1,  r2
        strh_post       r6,  r0,  r2
  .else
        vld1.16         {d16[0]}, [r1], r2
        vld1.16         {d16[1]}, [r1], r2
        vld1.16         {d18[0]}, [r0,:16], r2
        vld1.16         {d18[1]}, [r0,:16]
        sub             r0,  r0,  r2
        vrhadd.u8       d16, d16, d18
        vst1.16         {d16[0]}, [r0,:16], r2
        vst1.16         {d16[1]}, [r0,:16], r2
  .endif
        subs            r3,  r3,  #2
        bgt             2b
        pop             {r4-r6, pc}
endfunc
.endm

        h264_chroma_mc8 put
        h264_chroma_mc8 avg
        h264_chroma_mc4 put
        h264_chroma_mc4 avg
        h264_chroma_mc2 put
        h264_chroma_mc2 avg

#if CONFIG_RV40_DECODER
const   rv40bias
        .short           0, 16, 32, 16
        .short          32, 28, 32, 28
        .short           0, 32, 16, 32
        .short          32, 28, 32, 28
endconst

        h264_chroma_mc8 put, rv40
        h264_chroma_mc8 avg, rv40
        h264_chroma_mc4 put, rv40
        h264_chroma_mc4 avg, rv40
#endif

#if CONFIG_VC1_DECODER
        h264_chroma_mc8 put, vc1
        h264_chroma_mc8 avg, vc1
        h264_chroma_mc4 put, vc1
        h264_chroma_mc4 avg, vc1
#endif