summaryrefslogtreecommitdiffstats
path: root/libavcodec/aarch64/h264cmc_neon.S
blob: d1025c7a254297b96a417c2d4395627b240dfae3 (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
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
/*
 * Copyright (c) 2008 Mans Rullgard <mans@mansr.com>
 * Copyright (c) 2013 Janne Grunau <janne-libav@jannau.net>
 *
 * 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/aarch64/asm.S"

/* chroma_mc8(uint8_t *dst, uint8_t *src, int stride, int h, int x, int y) */
.macro  h264_chroma_mc8 type, codec=h264
function ff_\type\()_\codec\()_chroma_mc8_neon, export=1
        sxtw            x2,  w2
  .ifc \type,avg
        mov             x8,  x0
  .endif
        prfm            pldl1strm, [x1]
        prfm            pldl1strm, [x1, x2]
  .ifc \codec,rv40
        movrel          x6,  rv40bias
        lsr             w9,  w5,  #1
        lsr             w10, w4,  #1
        lsl             w9,  w9,  #3
        lsl             w10, w10, #1
        add             w9,  w9,  w10
        add             x6,  x6,  w9, UXTW
        ld1r            {v22.8H}, [x6]
  .endif
  .ifc \codec,vc1
        movi            v22.8H,   #28
  .endif
        mul             w7,  w4,  w5
        lsl             w14, w5,  #3
        lsl             w13, w4,  #3
        cmp             w7,  #0
        sub             w6,  w14, w7
        sub             w12, w13, w7
        sub             w4,  w7,  w13
        sub             w4,  w4,  w14
        add             w4,  w4,  #64
        b.eq            2f

        dup             v0.8B,  w4
        dup             v1.8B,  w12
        ld1             {v4.8B, v5.8B}, [x1], x2
        dup             v2.8B,  w6
        dup             v3.8B,  w7
        ext             v5.8B,  v4.8B,  v5.8B,  #1
1:      ld1             {v6.8B, v7.8B}, [x1], x2
        umull           v16.8H, v4.8B,  v0.8B
        umlal           v16.8H, v5.8B,  v1.8B
        ext             v7.8B,  v6.8B,  v7.8B,  #1
        ld1             {v4.8B, v5.8B}, [x1], x2
        umlal           v16.8H, v6.8B,  v2.8B
        prfm            pldl1strm, [x1]
        ext             v5.8B,  v4.8B,  v5.8B,  #1
        umlal           v16.8H, v7.8B,  v3.8B
        umull           v17.8H, v6.8B,  v0.8B
        subs            w3,  w3,  #2
        umlal           v17.8H, v7.8B, v1.8B
        umlal           v17.8H, v4.8B, v2.8B
        umlal           v17.8H, v5.8B, v3.8B
        prfm            pldl1strm, [x1, x2]
  .ifc \codec,h264
        rshrn           v16.8B, v16.8H, #6
        rshrn           v17.8B, v17.8H, #6
  .else
        add             v16.8H, v16.8H, v22.8H
        add             v17.8H, v17.8H, v22.8H
        shrn            v16.8B, v16.8H, #6
        shrn            v17.8B, v17.8H, #6
  .endif
  .ifc \type,avg
        ld1             {v20.8B}, [x8], x2
        ld1             {v21.8B}, [x8], x2
        urhadd          v16.8B, v16.8B, v20.8B
        urhadd          v17.8B, v17.8B, v21.8B
  .endif
        st1             {v16.8B}, [x0], x2
        st1             {v17.8B}, [x0], x2
        b.gt            1b
        ret

2:      adds            w12, w12, w6
        dup             v0.8B, w4
        b.eq            5f
        tst             w6,  w6
        dup             v1.8B, w12
        b.eq            4f

        ld1             {v4.8B}, [x1], x2
3:      ld1             {v6.8B}, [x1], x2
        umull           v16.8H, v4.8B,  v0.8B
        umlal           v16.8H, v6.8B,  v1.8B
        ld1             {v4.8B}, [x1], x2
        umull           v17.8H, v6.8B,  v0.8B
        umlal           v17.8H, v4.8B,  v1.8B
        prfm            pldl1strm, [x1]
  .ifc \codec,h264
        rshrn           v16.8B, v16.8H, #6
        rshrn           v17.8B, v17.8H, #6
  .else
        add             v16.8H, v16.8H, v22.8H
        add             v17.8H, v17.8H, v22.8H
        shrn            v16.8B, v16.8H, #6
        shrn            v17.8B, v17.8H, #6
  .endif
        prfm            pldl1strm, [x1, x2]
  .ifc \type,avg
        ld1             {v20.8B}, [x8], x2
        ld1             {v21.8B}, [x8], x2
        urhadd          v16.8B, v16.8B, v20.8B
        urhadd          v17.8B, v17.8B, v21.8B
  .endif
        subs            w3,  w3,  #2
        st1             {v16.8B}, [x0], x2
        st1             {v17.8B}, [x0], x2
        b.gt            3b
        ret

4:      ld1             {v4.8B, v5.8B}, [x1], x2
        ld1             {v6.8B, v7.8B}, [x1], x2
        ext             v5.8B,  v4.8B,  v5.8B,  #1
        ext             v7.8B,  v6.8B,  v7.8B,  #1
        prfm            pldl1strm, [x1]
        subs            w3,  w3,  #2
        umull           v16.8H, v4.8B, v0.8B
        umlal           v16.8H, v5.8B, v1.8B
        umull           v17.8H, v6.8B, v0.8B
        umlal           v17.8H, v7.8B, v1.8B
        prfm            pldl1strm, [x1, x2]
  .ifc \codec,h264
        rshrn           v16.8B, v16.8H, #6
        rshrn           v17.8B, v17.8H, #6
  .else
        add             v16.8H, v16.8H, v22.8H
        add             v17.8H, v17.8H, v22.8H
        shrn            v16.8B, v16.8H, #6
        shrn            v17.8B, v17.8H, #6
  .endif
  .ifc \type,avg
        ld1             {v20.8B}, [x8], x2
        ld1             {v21.8B}, [x8], x2
        urhadd          v16.8B, v16.8B, v20.8B
        urhadd          v17.8B, v17.8B, v21.8B
  .endif
        st1             {v16.8B}, [x0], x2
        st1             {v17.8B}, [x0], x2
        b.gt            4b
        ret

5:      ld1             {v4.8B}, [x1], x2
        ld1             {v5.8B}, [x1], x2
        prfm            pldl1strm, [x1]
        subs            w3,  w3,  #2
        umull           v16.8H, v4.8B, v0.8B
        umull           v17.8H, v5.8B, v0.8B
        prfm            pldl1strm, [x1, x2]
  .ifc \codec,h264
        rshrn           v16.8B, v16.8H, #6
        rshrn           v17.8B, v17.8H, #6
  .else
        add             v16.8H, v16.8H, v22.8H
        add             v17.8H, v17.8H, v22.8H
        shrn            v16.8B, v16.8H, #6
        shrn            v17.8B, v17.8H, #6
  .endif
  .ifc \type,avg
        ld1             {v20.8B}, [x8], x2
        ld1             {v21.8B}, [x8], x2
        urhadd          v16.8B, v16.8B, v20.8B
        urhadd          v17.8B, v17.8B, v21.8B
  .endif
        st1             {v16.8B}, [x0], x2
        st1             {v17.8B}, [x0], x2
        b.gt            5b
        ret
endfunc
.endm

/* chroma_mc4(uint8_t *dst, uint8_t *src, int stride, int h, int x, int y) */
.macro  h264_chroma_mc4 type, codec=h264
function ff_\type\()_\codec\()_chroma_mc4_neon, export=1
        sxtw            x2,  w2
  .ifc \type,avg
        mov             x8,  x0
  .endif
        prfm            pldl1strm, [x1]
        prfm            pldl1strm, [x1, x2]
  .ifc \codec,rv40
        movrel          x6,  rv40bias
        lsr             w9,  w5,  #1
        lsr             w10, w4,  #1
        lsl             w9,  w9,  #3
        lsl             w10, w10, #1
        add             w9,  w9,  w10
        add             x6,  x6,  w9, UXTW
        ld1r            {v22.8H}, [x6]
  .endif
  .ifc \codec,vc1
        movi            v22.8H,   #28
  .endif
        mul             w7,  w4,  w5
        lsl             w14, w5,  #3
        lsl             w13, w4,  #3
        cmp             w7,  #0
        sub             w6,  w14, w7
        sub             w12, w13, w7
        sub             w4,  w7,  w13
        sub             w4,  w4,  w14
        add             w4,  w4,  #64
        b.eq            2f

        dup             v24.8B,  w4
        dup             v25.8B,  w12
        ld1             {v4.8B}, [x1], x2
        dup             v26.8B,  w6
        dup             v27.8B,  w7
        ext             v5.8B,  v4.8B,  v5.8B, #1
        trn1            v0.2S,  v24.2S, v25.2S
        trn1            v2.2S,  v26.2S, v27.2S
        trn1            v4.2S,  v4.2S,  v5.2S
1:      ld1             {v6.8B}, [x1], x2
        ext             v7.8B,  v6.8B,  v7.8B, #1
        trn1            v6.2S,  v6.2S,  v7.2S
        umull           v18.8H, v4.8B,  v0.8B
        umlal           v18.8H, v6.8B,  v2.8B
        ld1             {v4.8B}, [x1], x2
        ext             v5.8B,  v4.8B,  v5.8B, #1
        trn1            v4.2S,  v4.2S,  v5.2S
        prfm            pldl1strm, [x1]
        umull           v19.8H, v6.8B,  v0.8B
        umlal           v19.8H, v4.8B,  v2.8B
        trn1            v30.2D, v18.2D, v19.2D
        trn2            v31.2D, v18.2D, v19.2D
        add             v18.8H, v30.8H, v31.8H
  .ifc \codec,h264
        rshrn           v16.8B, v18.8H, #6
  .else
        add             v18.8H, v18.8H, v22.8H
        shrn            v16.8B, v18.8H, #6
  .endif
        subs            w3,  w3,  #2
        prfm            pldl1strm, [x1, x2]
  .ifc \type,avg
        ld1             {v20.S}[0], [x8], x2
        ld1             {v20.S}[1], [x8], x2
        urhadd          v16.8B, v16.8B, v20.8B
  .endif
        st1             {v16.S}[0], [x0], x2
        st1             {v16.S}[1], [x0], x2
        b.gt            1b
        ret

2:      adds            w12, w12, w6
        dup             v30.8B, w4
        b.eq            5f
        tst             w6,  w6
        dup             v31.8B, w12
        trn1            v0.2S,  v30.2S, v31.2S
        trn2            v1.2S,  v30.2S, v31.2S
        b.eq            4f

        ext             v1.8B,  v0.8B,  v1.8B, #4
        ld1             {v4.S}[0], [x1], x2
3:      ld1             {v4.S}[1], [x1], x2
        umull           v18.8H, v4.8B,  v0.8B
        ld1             {v4.S}[0], [x1], x2
        umull           v19.8H, v4.8B,  v1.8B
        trn1            v30.2D, v18.2D, v19.2D
        trn2            v31.2D, v18.2D, v19.2D
        add             v18.8H, v30.8H, v31.8H
        prfm            pldl1strm, [x1]
  .ifc \codec,h264
        rshrn           v16.8B, v18.8H, #6
  .else
        add             v18.8H, v18.8H, v22.8H
        shrn            v16.8B, v18.8H, #6
  .endif
  .ifc \type,avg
        ld1             {v20.S}[0], [x8], x2
        ld1             {v20.S}[1], [x8], x2
        urhadd          v16.8B, v16.8B, v20.8B
  .endif
        subs            w3,  w3,  #2
        prfm            pldl1strm, [x1, x2]
        st1             {v16.S}[0], [x0], x2
        st1             {v16.S}[1], [x0], x2
        b.gt            3b
        ret

4:      ld1             {v4.8B}, [x1], x2
        ld1             {v6.8B}, [x1], x2
        ext             v5.8B,  v4.8B,  v5.8B, #1
        ext             v7.8B,  v6.8B,  v7.8B, #1
        trn1            v4.2S,  v4.2S,  v5.2S
        trn1            v6.2S,  v6.2S,  v7.2S
        umull           v18.8H, v4.8B,  v0.8B
        umull           v19.8H, v6.8B,  v0.8B
        subs            w3,  w3,  #2
        trn1            v30.2D, v18.2D, v19.2D
        trn2            v31.2D, v18.2D, v19.2D
        add             v18.8H, v30.8H, v31.8H
        prfm            pldl1strm, [x1]
  .ifc \codec,h264
        rshrn           v16.8B, v18.8H, #6
  .else
        add             v18.8H, v18.8H, v22.8H
        shrn            v16.8B, v18.8H, #6
  .endif
  .ifc \type,avg
        ld1             {v20.S}[0], [x8], x2
        ld1             {v20.S}[1], [x8], x2
        urhadd          v16.8B, v16.8B, v20.8B
  .endif
        prfm            pldl1strm, [x1]
        st1             {v16.S}[0], [x0], x2
        st1             {v16.S}[1], [x0], x2
        b.gt            4b
        ret

5:      ld1             {v4.S}[0], [x1], x2
        ld1             {v4.S}[1], [x1], x2
        umull           v18.8H, v4.8B,  v30.8B
        subs            w3,  w3,  #2
        prfm            pldl1strm, [x1]
  .ifc \codec,h264
        rshrn           v16.8B, v18.8H, #6
  .else
        add             v18.8H, v18.8H, v22.8H
        shrn            v16.8B, v18.8H, #6
  .endif
  .ifc \type,avg
        ld1             {v20.S}[0], [x8], x2
        ld1             {v20.S}[1], [x8], x2
        urhadd          v16.8B, v16.8B, v20.8B
  .endif
        prfm            pldl1strm, [x1]
        st1             {v16.S}[0], [x0], x2
        st1             {v16.S}[1], [x0], x2
        b.gt            5b
        ret
endfunc
.endm

.macro  h264_chroma_mc2 type
function ff_\type\()_h264_chroma_mc2_neon, export=1
        sxtw            x2,  w2
        prfm            pldl1strm, [x1]
        prfm            pldl1strm, [x1, x2]
        orr             w7,  w4,  w5
        cbz             w7,  2f

        mul             w7,  w4,  w5
        lsl             w14, w5,  #3
        lsl             w13, w4,  #3
        sub             w6,  w14, w7
        sub             w12, w13, w7
        sub             w4,  w7,  w13
        sub             w4,  w4,  w14
        add             w4,  w4,  #64
        dup             v0.8B,  w4
        dup             v2.8B,  w12
        dup             v1.8B,  w6
        dup             v3.8B,  w7
        trn1            v0.4H,  v0.4H,  v2.4H
        trn1            v1.4H,  v1.4H,  v3.4H
1:
        ld1             {v4.S}[0],  [x1], x2
        ld1             {v4.S}[1],  [x1], x2
        rev64           v5.2S,  v4.2S
        ld1             {v5.S}[1],  [x1]
        ext             v6.8B,  v4.8B,  v5.8B,  #1
        ext             v7.8B,  v5.8B,  v4.8B,  #1
        trn1            v4.4H,  v4.4H,  v6.4H
        trn1            v5.4H,  v5.4H,  v7.4H
        umull           v16.8H, v4.8B,  v0.8B
        umlal           v16.8H, v5.8B,  v1.8B
  .ifc \type,avg
        ld1             {v18.H}[0], [x0], x2
        ld1             {v18.H}[2], [x0]
        sub             x0,  x0,  x2
  .endif
        rev64           v17.4S, v16.4S
        add             v16.8H, v16.8H, v17.8H
        rshrn           v16.8B, v16.8H, #6
  .ifc \type,avg
        urhadd          v16.8B, v16.8B, v18.8B
  .endif
        st1             {v16.H}[0], [x0], x2
        st1             {v16.H}[2], [x0], x2
        subs            w3,  w3,  #2
        b.gt            1b
        ret

2:
        ld1             {v16.H}[0], [x1], x2
        ld1             {v16.H}[1], [x1], x2
  .ifc \type,avg
        ld1             {v18.H}[0], [x0], x2
        ld1             {v18.H}[1], [x0]
        sub             x0,  x0,  x2
        urhadd          v16.8B, v16.8B, v18.8B
  .endif
        st1             {v16.H}[0], [x0], x2
        st1             {v16.H}[1], [x0], x2
        subs            w3,  w3,  #2
        b.gt            2b
        ret
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
OpenPOWER on IntegriCloud