Debian Codegraph @

src/shaders/post_processing/gen7/Save_AVS_PL3.g4a

raw
1 /*
2 * Copyright 2000-2011 Intel Corporation All Rights Reserved
3 *
4 * Permission is hereby granted, free of charge, to any person obtaining a
5 * copy of this software and associated documentation files (the
6 * "Software"), to deal in the Software without restriction, including
7 * without limitation the rights to use, copy, modify, merge, publish,
8 * distribute, sub license, and/or sell copies of the Software, and to
9 * permit persons to whom the Software is furnished to do so, subject to
10 * the following conditions:
11 *
12 * The above copyright notice and this permission notice (including the
13 * next paragraph) shall be included in all copies or substantial portions
14 * of the Software.
15 *
16 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
17 * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
18 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.
19 * IN NO EVENT SHALL PRECISION INSIGHT AND/OR ITS SUPPLIERS BE LIABLE FOR
20 * ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
21 * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
22 * SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
23 *
24 * This file was originally licensed under the following license
25 *
26 * Licensed under the Apache License, Version 2.0 (the "License");
27 * you may not use this file except in compliance with the License.
28 * You may obtain a copy of the License at
29 *
30 * http://www.apache.org/licenses/LICENSE-2.0
31 *
32 * Unless required by applicable law or agreed to in writing, software
33 * distributed under the License is distributed on an "AS IS" BASIS,
34 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
35 * See the License for the specific language governing permissions and
36 * limitations under the License.
37 */
38 // 84 // Total instruction count
39 // 1 // Total kernel count
40
41
42
43 // Module name: common.inc
44 //
45 // Common header file for all Video-Processing kernels
46 //
47
48 .default_execution_size (16)
49 .default_register_type :ub
50
51 .reg_count_total 128
52 .reg_count_payload 7
53
54 //========== Common constants ==========
55
56
57 //========== Macros ==========
58
59
60 //Fast Jump, For more details see "Set_Layer_N.asm"
61
62
63 //========== Defines ====================
64
65 //========== Static Parameters (Common To All) ==========
66 //r1
67
68
69 //r2
70
71 // e.g. byte0 byte1 byte2
72 // YUYV 0 1 3
73 // YVYU 0 3 1
74
75 //Color Pipe (IECP) parameters
76
77
78 //ByteCopy
79
80
81 //r4
82
83 // e.g. byte0 byte1 byte2
84 // YUYV 0 1 3
85 // YVYU 0 3 1
86
87
88 //========== Inline parameters (Common To All) ===========
89
90
91 //============== Binding Index Table===========
92 //Common between DNDI and DNUV
93
94
95 //================= Common Message Descriptor =====
96 // Message descriptor for thread spawning
97 // Message Descriptors
98 // = 000 0001 (min message len 1 ) 0,0000 (resp len 0 -add later)
99 // 0000,0000,0000
100 // 0001(Spawn a root thread),0001 (Root thread spawn thread)
101 // = 0x02000011
102 // Thread Spawner Message Descriptor
103
104
105 // Message descriptor for atomic operation add
106 // Message Descriptors
107 // = 000 0110 (min message len 6 ) 0,0000 (resp len 0 -add later)
108 // 1(header present)001,10(typed atomic operation)0(return enabled)0(slot group, low 8 bits),0111 (AOP_Add)
109 // 0000,0000 (Binding table index, added later)
110 // = 0x02000011
111
112 // Atomic Operation Add Message Descriptor
113
114
115 // Message descriptor for dataport media write
116 // Message Descriptors
117 // = 000 0001 (min message len 1 - add later) 00000 (resp len 0)
118 // 1 (header present 1) 0 1010 (media block write) 000000
119 // 00000000 (binding table index - set later)
120 // = 0x020A8000
121
122
123 // Message Length defines
124
125
126 // Response Length defines
127
128
129 // Block Width and Height Size defines
130
131
132 // Extended Message Descriptors
133
134
135 // Common message descriptors:
136
137
138 //===================== Math Function Control ===================================
139
140
141 //============ Message Registers ===============
142 // buf4 starts from r28
143
144
145 //#define mMSGHDR_EOT r43 // Dummy Message Register for EOT
146
147
148 .declare mubMSGPAYLOAD Base=r30 ElementSize=1 SrcRegion=<16;16,1> Type=ub
149 .declare muwMSGPAYLOAD Base=r30 ElementSize=2 SrcRegion=<16;16,1> Type=uw
150 .declare mudMSGPAYLOAD Base=r30 ElementSize=4 SrcRegion=<8;8,1> Type=ud
151 .declare mfMSGPAYLOAD Base=r30 ElementSize=4 SrcRegion=<8;8,1> Type=f
152
153 //=================== End of thread instruction ===========================
154
155
156 //=====================Pointers Used=====================================
157
158
159 //=======================================================================
160
161
162 //r11-r17
163 // Define temp space for any usages
164
165
166 // Common Buffers
167
168
169 // temp space for rotation
170
171 .declare fROBUF Base=r11.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f
172
173 .declare udROBUF Base=r11.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=ud
174
175 .declare uwROBUF Base=r11.0 ElementSize=2 SrcRegion=<16;16,1> DstRegion=<1> Type=uw
176
177 .declare ubROBUF Base=r11.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub
178
179 .declare ub4ROBUF Base=r11.0 ElementSize=1 SrcRegion=<32;8,4> DstRegion=<4> Type=ub
180
181
182 // End of common.inc
183
184
185 // Module name: Save_AVS_PL3.asm
186 //
187 // Save PL3 420 frame data block of size 16x16
188 //
189 // To save 16x16 block (16x16 byte of Y and 8x8 byte of U and V each) we need 3 send instructions with one of size 16x16 and two of size 8x8.
190 // -----------------
191 // | 16x16 Y |
192 // | |
193 // -----------------
194 // | 8x8 U |
195 // ---------
196 // | 8x8 V |
197 // ---------
198
199 //-----------------------------------------------------------------
200 //The layout of data is as follows:
201 //mMSGHDR0 : Y data header (16x16)
202 //mubMSGPAYLOAD0 : Y data payload (8 GRFs)
203 //mMSGHDR1 : U data header (8x8)
204 //mubMSGPAYLOAD1 : U data payload (2 GRFs)
205 //mMSGHDR2 : V data header (8x8)
206 //mubMSGPAYLOAD2 : V data payload (2 GRFs)
207 //------------------------------------------------------------------
208
209
210
211 // Module name: Save.inc
212
213
214
215
216 // Description: Includes all definitions explicit to Fast Composite.
217
218
219
220
221 // End of common.inc
222
223
224 //========== GRF partition ==========
225 // r0 header : r0 (1 GRF)
226 // Static parameters : r1 - r6 (6 GRFS)
227 // Inline parameters : r7 - r8 (2 GRFs)
228 // MSGSRC : r27 (1 GRF)
229 //===================================
230
231 //Interface:
232 //========== Static Parameters (Explicit To Fast Composite) ==========
233 //r1
234 //CSC Set 0
235
236
237 .declare udCSC_CURBE Base=r1.0 ElementSize=4 Type=ud
238
239 //Constant alpha
240
241
242 //r2
243
244
245 // Gen7 AVS WA
246
247
248 // WiDi Definitions
249
250
251 //Colorfill
252
253
254 // 0: 0-degree, 1: 90, 2: 180, 3: 270-degree, clockwise.
255
256 .declare ubCOLOR_PIXEL_VAL Base=r2.20 ElementSize=1 SrcRegion=<0;1,0> DstRegion=<1> Type=ub
257
258 //r3
259 //Normalised Ratio of Horizontal step size with main video for all layers
260
261
262 //Normalised Ratio of Horizontal step size with main video for all layers becomes
263 //Normalised Horizontal step size for all layers in VP_Setup.asm
264
265
266 //r4
267 //Normalised Vertical step size for all layers
268
269
270 //r5
271 //Normalised Vertical Frame Origin for all layers
272
273
274 //r6
275 //Normalised Horizontal Frame Origin for all layers
276
277
278 //========== Inline Parameters (Explicit To Fast Composite) ==========
279
280
281 //Main video Step X
282
283
284 //====================== Binding table (Explicit To Fast Composite)=========================================
285
286
287 //Used by Interlaced Scaling Kernels
288
289
290 //========== Sampler State Table Index (Explicit To Fast Composite)==========
291 //Sampler Index for AVS/IEF messages
292
293
294 //Sampler Index for SIMD16 sampler messages
295
296
297 //=============================================================================
298
299 .declare fBUFFER_0 Base=r64.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f
300 .declare fBUFFER_1 Base=r80.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f
301 .declare fBUFFER_2 Base=r96.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f
302 .declare fBUFFER_3 Base=r112.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f
303 .declare fBUFFER_4 Base=r28.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f
304 .declare fBUFFER_5 Base=r46.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f
305
306 .declare udBUFFER_0 Base=r64.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=ud
307 .declare udBUFFER_1 Base=r80.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=ud
308 .declare udBUFFER_2 Base=r96.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=ud
309 .declare udBUFFER_3 Base=r112.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=ud
310 .declare udBUFFER_4 Base=r28.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=ud
311 .declare udBUFFER_5 Base=r46.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=ud
312
313 .declare uwBUFFER_0 Base=r64.0 ElementSize=2 SrcRegion=<16;16,1> DstRegion=<1> Type=uw
314 .declare uwBUFFER_1 Base=r80.0 ElementSize=2 SrcRegion=<16;16,1> DstRegion=<1> Type=uw
315 .declare uwBUFFER_2 Base=r96.0 ElementSize=2 SrcRegion=<16;16,1> DstRegion=<1> Type=uw
316 .declare uwBUFFER_3 Base=r112.0 ElementSize=2 SrcRegion=<16;16,1> DstRegion=<1> Type=uw
317 .declare uwBUFFER_4 Base=r28.0 ElementSize=2 SrcRegion=<16;16,1> DstRegion=<1> Type=uw
318 .declare uwBUFFER_5 Base=r46.0 ElementSize=2 SrcRegion=<16;16,1> DstRegion=<1> Type=uw
319
320 .declare ubBUFFER_0 Base=r64.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub
321 .declare ubBUFFER_1 Base=r80.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub
322 .declare ubBUFFER_2 Base=r96.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub
323 .declare ubBUFFER_3 Base=r112.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub
324 .declare ubBUFFER_4 Base=r28.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub
325 .declare ubBUFFER_5 Base=r46.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub
326
327 .declare ub4BUFFER_0 Base=r64.0 ElementSize=1 SrcRegion=<32;8,4> DstRegion=<4> Type=ub
328 .declare ub4BUFFER_1 Base=r80.0 ElementSize=1 SrcRegion=<32;8,4> DstRegion=<4> Type=ub
329 .declare ub4BUFFER_2 Base=r96.0 ElementSize=1 SrcRegion=<32;8,4> DstRegion=<4> Type=ub
330 .declare ub4BUFFER_3 Base=r112.0 ElementSize=1 SrcRegion=<32;8,4> DstRegion=<4> Type=ub
331 .declare ub4BUFFER_4 Base=r28.0 ElementSize=1 SrcRegion=<32;8,4> DstRegion=<4> Type=ub
332 .declare ub4BUFFER_5 Base=r46.0 ElementSize=1 SrcRegion=<32;8,4> DstRegion=<4> Type=ub
333
334 //Pointer to mask reg
335
336
337 //r18
338
339
340 //Always keep Cannel Pointers and Offsets in same GRF, so that we can use
341 // NODDCLR, NODDCHK flags. -rT
342
343
344 .declare udCSC_COEFF_0 Base=r18.0 ElementSize=4 Type=ud // 1 GRF
345
346 //r19
347
348
349 .declare udCSC_COEFF_1 Base=r19.0 ElementSize=4 Type=ud // 1 GRF
350
351
352 //r20
353
354 .declare uwALPHA_MASK_REG_TEMP Base=r20.0 ElementSize=2 SrcRegion=<16;16,1> Type=uw // 1 GRF
355
356 //r21
357
358 .declare uwALPHA_MASK_REG Base=r21.0 ElementSize=2 SrcRegion=<16;16,1> Type=uw // 1 GRF
359
360 //r22
361
362
363 //Always keep Cannel Pointers and Offsets in same GRF, so that we can use
364 // NODDCLR, NODDCHK flags. -rT
365
366
367 //Keep fORIGIN_X_NLAS, fY_OFFSET_2ND_BLOCK, fSTEP_X_NLAS, pMSGDSC_COPY, ubCONST_ALPHA_COPY as
368 //sub registers of same GRF to enable using NODDCLR NODDCHK. -rT
369
370 //r23
371
372
373 //Lumakey
374
375
376 //r24
377
378
379 //r25
380
381
382 //r26
383
384
385 //defines to generate LABELS during compile time.
386
387
388 //Msg payload buffers; upto 4 full-size messages can be written
389
390
391 .declare mudMSGPAYLOAD0 Base=r29.0 ElementSize=4 SrcRegion=<8;8,1> Type=ud
392 .declare mudMSGPAYLOAD1 Base=r38.0 ElementSize=4 SrcRegion=<8;8,1> Type=ud
393 .declare mudMSGPAYLOAD2 Base=r47.0 ElementSize=4 SrcRegion=<8;8,1> Type=ud
394 .declare mudMSGPAYLOAD3 Base=r56.0 ElementSize=4 SrcRegion=<8;8,1> Type=ud
395
396 .declare muwMSGPAYLOAD0 Base=r29.0 ElementSize=2 SrcRegion=<16;16,1> Type=uw
397 .declare muwMSGPAYLOAD1 Base=r38.0 ElementSize=2 SrcRegion=<16;16,1> Type=uw
398 .declare muwMSGPAYLOAD2 Base=r47.0 ElementSize=2 SrcRegion=<16;16,1> Type=uw
399 .declare muwMSGPAYLOAD3 Base=r56.0 ElementSize=2 SrcRegion=<16;16,1> Type=uw
400
401 .declare mubMSGPAYLOAD0 Base=r29.0 ElementSize=1 SrcRegion=<16;16,1> Type=ub
402 .declare mubMSGPAYLOAD1 Base=r38.0 ElementSize=1 SrcRegion=<16;16,1> Type=ub
403 .declare mubMSGPAYLOAD2 Base=r47.0 ElementSize=1 SrcRegion=<16;16,1> Type=ub
404 .declare mubMSGPAYLOAD3 Base=r56.0 ElementSize=1 SrcRegion=<16;16,1> Type=ub
405 .declare mubMSGPAYLOAD4 Base=r32.0 ElementSize=1 SrcRegion=<16;16,1> Type=ub
406 .declare mubMSGPAYLOAD5 Base=r41.0 ElementSize=1 SrcRegion=<16;16,1> Type=ub
407 .declare mubMSGPAYLOAD6 Base=r50.0 ElementSize=1 SrcRegion=<16;16,1> Type=ub
408 .declare mubMSGPAYLOAD7 Base=r59.0 ElementSize=1 SrcRegion=<16;16,1> Type=ub
409
410
411 // the r17 register (nTEMP0) is originally defined from "Common.inc"
412 // instead of re-defining a nTEMP0 here, we use "SAVE_RGB" suffix for its naming
413
414 .declare uwTemp0 Base=r17.0 ElementSize=2 Type=uw
415
416
417 //_SAVE_INC_
418
419
420 // At the save module we have all 8 address sub-registers available.
421 // So we will use PING-PONG type of scheme to save the data using
422 // pointers pBUF_CHNL_TOP_8x4 and pBUF_CHNL_BOT_8x4. This will help
423 // reduce dependency. - rT
424
425 //wBUFF_CHNL_PTR points to either buffer 0 or buffer 4.
426 //Add appropriate offsets to get pointers for all buffers (1,2,3 or 5).
427 //Offsets are zero for buffer 0 and buffer 4.
428 add (4) a0.0:uw r22.0<4;4,1>:w 0:uw
429 add (4) a0.4:uw r22.0<4;4,1>:w 512:uw
430
431 //Set up header for Y,U and V data
432 mov (8) r28<1>:ud r27<8;8,1>:ud
433 mov (8) r37<1>:ud r27<8;8,1>:ud
434 mov (8) r46<1>:ud r27<8;8,1>:ud
435
436 mov (2) r28.0<1>:d r9.0<2;2,1>:w { NoDDClr } //ORI Y (LUMA) = ORI
437 shr (2) r37.0<1>:d r9.0<2;2,1>:w 1:w { NoDDClr } //H/V ORI U = H/V ORI/2
438 shr (2) r46.0<1>:d r9.0<2;2,1>:w 1:w { NoDDClr } //H/V ORI V = H/V ORI/2
439
440 mov (1) r28.2<1>:ud 0xF000F:ud { NoDDChk } // Y Block width and height (16x16)
441 mov (1)r37.2<1>:ud 0x70007:ud { NoDDChk } // U Block width and height (8x8)
442 mov (1)r46.2<1>:ud 0x70007:ud { NoDDChk } // V Block width and height (8x8)
443
444 // Unscramble, and pack data directly to MRFs
445
446 // Data 16x16 block is divided as -
447 // ---------
448 // | 0 |
449 // ---------
450 // | 1 |
451 // ---------
452 // | 2 |
453 // ---------
454 // | 3 |
455 // ---------
456 // All sub-blocks are of size 16x4
457 // 0: ubBUFFER_0
458 // 1: ubBUFFER_1, ubBUFFER_0+16
459 // 2: ubBUFFER_2
460 // 3: ubBUFFER_3, ubBUFFER_2+16
461
462 //Y Rounding, first
463 add.sat (16) r[a0.1,0]<1>:uw r[a0.1,0]<16;16,1>:uw 0x0080:uw
464 add.sat (16) r[a0.1,32]<1>:uw r[a0.1,32]<16;16,1>:uw 0x0080:uw
465 add.sat (16) r[a0.1,64]<1>:uw r[a0.1,64]<16;16,1>:uw 0x0080:uw
466 add.sat (16) r[a0.1,96]<1>:uw r[a0.1,96]<16;16,1>:uw 0x0080:uw
467
468 // U rounding
469 add.sat (8) r[a0.2,0]<2>:uw r[a0.2,0]<16;8,2>:uw 0x0080:uw
470 add.sat (8) r[a0.2,64]<2>:uw r[a0.2,64]<16;8,2>:uw 0x0080:uw
471
472 // V rounding
473 add.sat (8) r[a0.0,0]<2>:uw r[a0.0,0]<16;8,2>:uw 0x0080:uw
474 add.sat (8) r[a0.0,64]<2>:uw r[a0.0,64]<16;8,2>:uw 0x0080:uw
475
476 add (4) a0.0:uw r22.0<4;4,1>:w 1024:uw //Update Buffer 2 pointers
477
478 //Y Rounding, second
479 add.sat (16) r[a0.5,0]<1>:uw r[a0.5,0]<16;16,1>:uw 0x0080:uw
480 add.sat (16) r[a0.5,32]<1>:uw r[a0.5,32]<16;16,1>:uw 0x0080:uw
481 add.sat (16) r[a0.5,64]<1>:uw r[a0.5,64]<16;16,1>:uw 0x0080:uw
482 add.sat (16) r[a0.5,96]<1>:uw r[a0.5,96]<16;16,1>:uw 0x0080:uw
483
484 // U rounding
485 add.sat (8) r[a0.6,0]<2>:uw r[a0.6,0]<16;8,2>:uw 0x0080:uw
486 add.sat (8) r[a0.6,64]<2>:uw r[a0.6,64]<16;8,2>:uw 0x0080:uw
487
488 // V rounding
489 add.sat (8) r[a0.4,0]<2>:uw r[a0.4,0]<16;8,2>:uw 0x0080:uw
490 add.sat (8) r[a0.4,64]<2>:uw r[a0.4,64]<16;8,2>:uw 0x0080:uw
491
492 add (4) a0.4:uw r22.0<4;4,1>:w 1536:uw //Update Buffer 3 pointers
493
494 //Y Rounding, third
495 add.sat (16) r[a0.1,0]<1>:uw r[a0.1,0]<16;16,1>:uw 0x0080:uw
496 add.sat (16) r[a0.1,32]<1>:uw r[a0.1,32]<16;16,1>:uw 0x0080:uw
497 add.sat (16) r[a0.1,64]<1>:uw r[a0.1,64]<16;16,1>:uw 0x0080:uw
498 add.sat (16) r[a0.1,96]<1>:uw r[a0.1,96]<16;16,1>:uw 0x0080:uw
499
500 // U rounding
501 add.sat (8) r[a0.2,0]<2>:uw r[a0.2,0]<16;8,2>:uw 0x0080:uw
502 add.sat (8) r[a0.2,64]<2>:uw r[a0.2,64]<16;8,2>:uw 0x0080:uw
503
504 // V rounding
505 add.sat (8) r[a0.0,0]<2>:uw r[a0.0,0]<16;8,2>:uw 0x0080:uw
506 add.sat (8) r[a0.0,64]<2>:uw r[a0.0,64]<16;8,2>:uw 0x0080:uw
507
508
509 //Y Rounding, fourth
510 add.sat (16) r[a0.5,0]<1>:uw r[a0.5,0]<16;16,1>:uw 0x0080:uw
511 add.sat (16) r[a0.5,32]<1>:uw r[a0.5,32]<16;16,1>:uw 0x0080:uw
512 add.sat (16) r[a0.5,64]<1>:uw r[a0.5,64]<16;16,1>:uw 0x0080:uw
513 add.sat (16) r[a0.5,96]<1>:uw r[a0.5,96]<16;16,1>:uw 0x0080:uw
514
515 // U rounding
516 add.sat (8) r[a0.6,0]<2>:uw r[a0.6,0]<16;8,2>:uw 0x0080:uw
517 add.sat (8) r[a0.6,64]<2>:uw r[a0.6,64]<16;8,2>:uw 0x0080:uw
518
519 // V rounding
520 add.sat (8) r[a0.4,0]<2>:uw r[a0.4,0]<16;8,2>:uw 0x0080:uw
521 add.sat (8) r[a0.4,64]<2>:uw r[a0.4,64]<16;8,2>:uw 0x0080:uw
522
523 // restore the TOP and BOT pointers
524 add (4) a0.0:uw r22.0<4;4,1>:w 0:uw
525 add (4) a0.4:uw r22.0<4;4,1>:w 512:uw
526
527 //Buffer 0
528 //Move Y to msg payload
529 mov (16) mubMSGPAYLOAD0(0,0)<1> r[a0.1, 1]<32;16,2>:ub { NoDDClr }
530 mov (16) mubMSGPAYLOAD0(0,16)<1> r[a0.1, 33]<32;16,2>:ub { NoDDChk }
531 mov (16) mubMSGPAYLOAD0(1,0)<1> r[a0.1, 65]<32;16,2>:ub { NoDDClr }
532 mov (16) mubMSGPAYLOAD0(1,16)<1> r[a0.1, 97]<32;16,2>:ub { NoDDChk }
533
534 //Move U to msg payload
535 mov (8) mubMSGPAYLOAD1(0,0)<1> r[a0.2, 1]<32;8,4>:ub { NoDDClr }
536 mov (8) mubMSGPAYLOAD1(0,8)<1> r[a0.2, 65]<32;8,4>:ub { NoDDClr, NoDDChk }
537
538 //Move V to msg payload
539 mov (8) mubMSGPAYLOAD2(0,0)<1> r[a0.0, 1]<32;8,4>:ub { NoDDClr }
540 mov (8) mubMSGPAYLOAD2(0,8)<1> r[a0.0, 65]<32;8,4>:ub { NoDDClr, NoDDChk }
541
542 add (4) a0.0:uw r22.0<4;4,1>:w 1024:uw //Update Buffer 2 pointers
543
544 //Buffer 1
545 mov (16) mubMSGPAYLOAD0(2,0)<1> r[a0.5, 1]<32;16,2>:ub { NoDDClr }
546 mov (16) mubMSGPAYLOAD0(2,16)<1> r[a0.5, 33]<32;16,2>:ub { NoDDChk }
547 mov (16) mubMSGPAYLOAD0(3,0)<1> r[a0.5, 65]<32;16,2>:ub { NoDDClr }
548 mov (16) mubMSGPAYLOAD0(3,16)<1> r[a0.5, 97]<32;16,2>:ub { NoDDChk }
549
550 mov (8) mubMSGPAYLOAD1(0,16)<1> r[a0.6, 1]<32;8,4>:ub { NoDDClr, NoDDChk }
551 mov (8) mubMSGPAYLOAD1(0,24)<1> r[a0.6, 65]<32;8,4>:ub { NoDDChk }
552
553 mov (8) mubMSGPAYLOAD2(0,16)<1> r[a0.4, 1]<32;8,4>:ub { NoDDClr, NoDDChk }
554 mov (8) mubMSGPAYLOAD2(0,24)<1> r[a0.4, 65]<32;8,4>:ub { NoDDChk }
555
556 add (4) a0.4:uw r22.0<4;4,1>:w 1536:uw //Update Buffer 3 pointers
557
558 //Buffer 2
559 mov (16) mubMSGPAYLOAD0(4,0)<1> r[a0.1, 1]<32;16,2>:ub { NoDDClr }
560 mov (16) mubMSGPAYLOAD0(4,16)<1> r[a0.1, 33]<32;16,2>:ub { NoDDChk }
561 mov (16) mubMSGPAYLOAD0(5,0)<1> r[a0.1, 65]<32;16,2>:ub { NoDDClr }
562 mov (16) mubMSGPAYLOAD0(5,16)<1> r[a0.1, 97]<32;16,2>:ub { NoDDChk }
563
564 mov (8) mubMSGPAYLOAD1(1,0)<1> r[a0.2, 1]<32;8,4>:ub { NoDDClr }
565 mov (8) mubMSGPAYLOAD1(1,8)<1> r[a0.2, 65]<32;8,4>:ub { NoDDClr, NoDDChk }
566
567 mov (8) mubMSGPAYLOAD2(1,0)<1> r[a0.0, 1]<32;8,4>:ub { NoDDClr }
568 mov (8) mubMSGPAYLOAD2(1,8)<1> r[a0.0, 65]<32;8,4>:ub { NoDDClr, NoDDChk }
569
570 //Buffer 3
571 mov (16) mubMSGPAYLOAD0(6,0)<1> r[a0.5, 1]<32;16,2>:ub { NoDDClr }
572 mov (16) mubMSGPAYLOAD0(6,16)<1> r[a0.5, 33]<32;16,2>:ub { NoDDChk }
573 mov (16) mubMSGPAYLOAD0(7,0)<1> r[a0.5, 65]<32;16,2>:ub { NoDDClr }
574 mov (16) mubMSGPAYLOAD0(7,16)<1> r[a0.5, 97]<32;16,2>:ub { NoDDChk }
575
576 mov (8) mubMSGPAYLOAD1(1,16)<1> r[a0.6, 1]<32;8,4>:ub { NoDDClr, NoDDChk }
577 mov (8) mubMSGPAYLOAD1(1,24)<1> r[a0.6, 65]<32;8,4>:ub { NoDDChk }
578
579 mov (8) mubMSGPAYLOAD2(1,16)<1> r[a0.4, 1]<32;8,4>:ub { NoDDClr, NoDDChk }
580 mov (8) mubMSGPAYLOAD2(1,24)<1> r[a0.4, 65]<32;8,4>:ub { NoDDChk }
581
582 //===========================================================================
583
584 send (1) null<1>:d r28 0x5 0x120A8018:ud
585 send (1) null<1>:d r37 0x5 0x60A8019:ud
586 send (1) null<1>:d r46 0x5 0x60A801A:ud