/* * Copyright 2000-2013 Intel Corporation All Rights Reserved * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. * Authors: * Zhao Yakui */ // Module name: common.inc // // Common header file for all Video-Processing kernels // .default_execution_size (16) .default_register_type :ub .reg_count_total 128 .reg_count_payload 7 //========== Common constants ========== //========== Macros ========== //Fast Jump, For more details see "Set_Layer_N.asm" //========== Defines ==================== //========== Static Parameters (Common To All) ========== //r1 //r2 // e.g. byte0 byte1 byte2 // YUYV 0 1 3 // YVYU 0 3 1 //Color Pipe (IECP) parameters //ByteCopy //r4 // e.g. byte0 byte1 byte2 // YUYV 0 1 3 // YVYU 0 3 1 //========== Inline parameters (Common To All) =========== //============== Binding Index Table=========== //Common between DNDI and DNUV //================= Common Message Descriptor ===== // Message descriptor for thread spawning // Message Descriptors // = 000 0001 (min message len 1 ) 0,0000 (resp len 0 -add later) // 0000,0000,0000 // 0001(Spawn a root thread),0001 (Root thread spawn thread) // = 0x02000011 // Thread Spawner Message Descriptor // Message descriptor for atomic operation add // Message Descriptors // = 000 0110 (min message len 6 ) 0,0000 (resp len 0 -add later) // 1(header present)001,10(typed atomic operation)0(return enabled)0(slot group, low 8 bits),0111 (AOP_Add) // 0000,0000 (Binding table index, added later) // = 0x02000011 // Atomic Operation Add Message Descriptor // Message descriptor for dataport media write // Message Descriptors // = 000 0001 (min message len 1 - add later) 00000 (resp len 0) // 1 (header present 1) 0 1010 (media block write) 000000 // 00000000 (binding table index - set later) // = 0x020A8000 // Message Length defines // Response Length defines // Block Width and Height Size defines // Extended Message Descriptors // Common message descriptors: //===================== Math Function Control =================================== //============ Message Registers =============== // buf4 starts from r28 //#define mMSGHDR_EOT r43 // Dummy Message Register for EOT .declare mubMSGPAYLOAD Base=r30 ElementSize=1 SrcRegion=<16;16,1> Type=ub .declare muwMSGPAYLOAD Base=r30 ElementSize=2 SrcRegion=<16;16,1> Type=uw .declare mudMSGPAYLOAD Base=r30 ElementSize=4 SrcRegion=<8;8,1> Type=ud .declare mfMSGPAYLOAD Base=r30 ElementSize=4 SrcRegion=<8;8,1> Type=f //=================== End of thread instruction =========================== //=====================Pointers Used===================================== //======================================================================= //r9-r17 // Define temp space for any usages // Common Buffers // temp space for rotation .declare fROBUF Base=r9.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f .declare udROBUF Base=r9.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=ud .declare uwROBUF Base=r9.0 ElementSize=2 SrcRegion=<16;16,1> DstRegion=<1> Type=uw .declare ubROBUF Base=r9.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub .declare ub4ROBUF Base=r9.0 ElementSize=1 SrcRegion=<32;8,4> DstRegion=<4> Type=ub // End of common.inc // Module name: YUV_to_RGB.asm // // Convert YUV to RGB, handle it by 16x4 block // // Description: Includes all definitions explicit to Fast Composite. //========== GRF partition ========== // r0 header : r0 (1 GRF) // Static parameters : r1 - r6 (6 GRFS) // Inline parameters : r7 - r8 (2 GRFs) // MSGSRC : r27 (1 GRF) //=================================== //Interface: //========== Static Parameters (Explicit To Fast Composite) ========== //r1 //CSC Set 0 .declare udCSC_CURBE Base=r1.0 ElementSize=4 Type=ud //Constant alpha //r2 // Gen7 AVS WA // WiDi Definitions //Colorfill // 0: 0-degree, 1: 90, 2: 180, 3: 270-degree, clockwise. .declare ubCOLOR_PIXEL_VAL Base=r2.20 ElementSize=1 SrcRegion=<0;1,0> DstRegion=<1> Type=ub //r3 //Normalised Ratio of Horizontal step size with main video for all layers //Normalised Ratio of Horizontal step size with main video for all layers becomes //Normalised Horizontal step size for all layers in VP_Setup.asm //r4 //Normalised Vertical step size for all layers //r5 //Normalised Vertical Frame Origin for all layers //r6 //Normalised Horizontal Frame Origin for all layers //========== Inline Parameters (Explicit To Fast Composite) ========== //Main video Step X //====================== Binding table (Explicit To Fast Composite)========================================= //Used by Interlaced Scaling Kernels //========== Sampler State Table Index (Explicit To Fast Composite)========== //Sampler Index for AVS/IEF messages //Sampler Index for SIMD16 sampler messages //============================================================================= .declare fBUFFER_0 Base=r64.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f .declare fBUFFER_1 Base=r80.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f .declare fBUFFER_2 Base=r96.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f .declare fBUFFER_3 Base=r112.0 ElementSize=4 SrcRegion=<8;8,1> DstRegion=<1> Type=f .declare bBUFFER_0 Base=r64.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub .declare bBUFFER_1 Base=r80.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub .declare bBUFFER_2 Base=r96.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub .declare bBUFFER_3 Base=r112.0 ElementSize=1 SrcRegion=<16;16,1> DstRegion=<1> Type=ub //Pointer to mask reg //r18 //Always keep Cannel Pointers and Offsets in same GRF, so that we can use // NODDCLR, NODDCHK flags. -rT .declare udCSC_COEFF_0 Base=r18.0 ElementSize=4 Type=ud // 1 GRF //r19 .declare udCSC_COEFF_1 Base=r19.0 ElementSize=4 Type=ud // 1 GRF //r20 .declare uwALPHA_MASK_REG_TEMP Base=r20.0 ElementSize=2 SrcRegion=<16;16,1> Type=uw // 1 GRF //r21 .declare uwALPHA_MASK_REG Base=r21.0 ElementSize=2 SrcRegion=<16;16,1> Type=uw // 1 GRF //r22 //Always keep Cannel Pointers and Offsets in same GRF, so that we can use // NODDCLR, NODDCHK flags. -rT //Keep fORIGIN_X_NLAS, fY_OFFSET_2ND_BLOCK, fSTEP_X_NLAS, pMSGDSC_COPY, ubCONST_ALPHA_COPY as //sub registers of same GRF to enable using NODDCLR NODDCHK. -rT //r23 //Lumakey //r24 //r25 //r26 //defines to generate LABELS during compile time. //Msg payload buffers; upto 4 full-size messages can be written //Unnecessary to use the MSGPayLoad, So it is temporiarily used for conversion of YUV->RGB .declare fBUFFER_R Base=r28.0 ElementSize=4 SrcRegion=<8;8,1> Type=f .declare fBUFFER_G Base=r30.0 ElementSize=4 SrcRegion=<8;8,1> Type=f .declare fBUFFER_B Base=r32.0 ElementSize=4 SrcRegion=<8;8,1> Type=f .declare fBUFFER_Y Base=r36.0 ElementSize=4 SrcRegion=<8;8,1> Type=f .declare fBUFFER_U Base=r38.0 ElementSize=4 SrcRegion=<8;8,1> Type=f .declare fBUFFER_V Base=r40.0 ElementSize=4 SrcRegion=<8;8,1> Type=f .declare wTempR Base=r42.0 ElementSize=2 Type=w .declare wTempG Base=r44.0 ElementSize=2 Type=w .declare wTempB Base=r46.0 ElementSize=2 Type=w .declare ubTempR Base=r42.0 ElementSize=1 Type=ub .declare ubTempG Base=r44.0 ElementSize=1 Type=ub .declare ubTempB Base=r46.0 ElementSize=1 Type=ub // the r17 register (nTEMP0) is originally defined from "Common.inc" // instead of re-defining a nTEMP0 here, we use "SAVE_RGB" suffix for its naming .declare wTemp0 Base=r17.0 ElementSize=2 Type=uw //_SAVE_INC_ // NTSC standard // R = Clamp ( 1.164(Y-16/255) + 1.596(Cr-128/255)) // G = Clamp ( 1.164(Y-16/255) - 0.813(Cr-128/255) - 0.392(Cb-128/255)) // B = Clamp ( 1.164(Y-16/255) + 2.017(Cb-128/255)) // ITU-R conversion, Now we are using ITU-R conversion // R = clip( Y + 1.402*(Cr-128)) // ITU-R // G = clip( Y - 0.344*(Cb-128) - 0.714*(Cr-128)) // B = clip( Y + 1.772*(Cb-128)) // At the save module we have all 8 address sub-registers available. // So we will use PING-PONG type of scheme to save the data using // pointers pBUF_CHNL_TOP_8x4 and pBUF_CHNL_BOT_8x4. This will help // reduce dependency. - rT //wBUFF_CHNL_PTR points to either buffer 0 or buffer 4. //Add appropriate offsets to get pointers for all buffers (1,2,3 or 5). //Offsets are zero for buffer 0 and buffer 4. //Y/U/V is also stored as R/G/B for the internal purpose //for BUFFER_0 mov (4) a0.0<1>:uw r22.0<4;4,1>:uw //the first line in the block 0 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 1]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 17]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 1]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 17]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 1]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 17]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,1]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,1]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,1]<2>:ub ubTempB(0, 0)<32;8,4> //the second line in the block 0 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 33]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 49]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 33]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 49]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 33]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 49]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,33]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,33]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,33]<2>:ub ubTempB(0, 0)<32;8,4> //the third line in the block 0 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 65]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 81]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 65]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 81]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 65]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 81]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,65]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,65]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,65]<2>:ub ubTempB(0, 0)<32;8,4> //the fourth line in the block 0 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 97]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 113]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 97]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 113]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 97]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 113]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,97]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,97]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,97]<2>:ub ubTempB(0, 0)<32;8,4> //for BUFFER_1 add (4) a0.0<1>:uw r22.0<4;4,1>:uw 512:uw //the first line in the block 1 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 1]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 17]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 1]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 17]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 1]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 17]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,1]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,1]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,1]<2>:ub ubTempB(0, 0)<32;8,4> //the second line in the block 1 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 33]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 49]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 33]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 49]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 33]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 49]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,33]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,33]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,33]<2>:ub ubTempB(0, 0)<32;8,4> //the third line in the block 1 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 65]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 81]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 65]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 81]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 65]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 81]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,65]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,65]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,65]<2>:ub ubTempB(0, 0)<32;8,4> //the fourth line in the block 1 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 97]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 113]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 97]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 113]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 97]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 113]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,97]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,97]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,97]<2>:ub ubTempB(0, 0)<32;8,4> //for BUFFER_2 add (4) a0.0<1>:uw r22.0<4;4,1>:uw 1024:uw //the first line in the block 2 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 1]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 17]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 1]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 17]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 1]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 17]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,1]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,1]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,1]<2>:ub ubTempB(0, 0)<32;8,4> //the second line in the block 2 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 33]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 49]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 33]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 49]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 33]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 49]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,33]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,33]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,33]<2>:ub ubTempB(0, 0)<32;8,4> //the third line in the block 2 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 65]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 81]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 65]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 81]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 65]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 81]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,65]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,65]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,65]<2>:ub ubTempB(0, 0)<32;8,4> //the fourth line in the block 2 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 97]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 113]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 97]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 113]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 97]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 113]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,97]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,97]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,97]<2>:ub ubTempB(0, 0)<32;8,4> //for BUFFER_3 add (4) a0.0<1>:uw r22.0<4;4,1>:uw 1536:uw //the first line in the block 3 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 1]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 17]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 1]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 17]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 1]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 17]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,1]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,1]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,1]<2>:ub ubTempB(0, 0)<32;8,4> //the second line in the block 3 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 33]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 49]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 33]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 49]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 33]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 49]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,33]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,33]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,33]<2>:ub ubTempB(0, 0)<32;8,4> //the third line in the block 3 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 65]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 81]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 65]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 81]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 65]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 81]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,65]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,65]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,65]<2>:ub ubTempB(0, 0)<32;8,4> //the fourth line in the block 3 mov (8) fBUFFER_Y(0, 0)<1> r[a0.1, 97]<16;8,2>:ub mov (8) fBUFFER_Y(1, 0)<1> r[a0.1, 113]<16;8,2>:ub mov (8) fBUFFER_U(0, 0)<1> r[a0.2, 97]<16;8,2>:ub mov (8) fBUFFER_U(1, 0)<1> r[a0.2, 113]<16;8,2>:ub mov (8) fBUFFER_V(0, 0)<1> r[a0.0, 97]<16;8,2>:ub mov (8) fBUFFER_V(1, 0)<1> r[a0.0, 113]<16;8,2>:ub add (16) fBUFFER_U(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> -128.0f add (16) fBUFFER_V(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -128.0f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_R(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 1.402f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (8) acc0.0<1>:f fBUFFER_U(0, 0)<8;8,1> -0.344f mac (8) acc1.0<1>:f fBUFFER_U(1, 0)<8;8,1> -0.344f mac (16) fBUFFER_G(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> -0.714f mov (16) acc0.0<1>:f fBUFFER_Y(0, 0)<8;8,1> mac (16) fBUFFER_B(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 1.772f mul.sat (16) fBUFFER_Y(0, 0)<1> fBUFFER_R(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_U(0, 0)<1> fBUFFER_G(0, 0)<8;8,1> 0.0039216f mul.sat (16) fBUFFER_V(0, 0)<1> fBUFFER_B(0, 0)<8;8,1> 0.0039216f mul (16) fBUFFER_R(0, 0)<1> fBUFFER_Y(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_G(0, 0)<1> fBUFFER_U(0, 0)<8;8,1> 255.0f mul (16) fBUFFER_B(0, 0)<1> fBUFFER_V(0, 0)<8;8,1> 255.0f mov (16) wTempR(0, 0)<2> fBUFFER_R(0, 0)<8;8,1> mov (16) wTempG(0, 0)<2> fBUFFER_G(0, 0)<8;8,1> mov (16) wTempB(0, 0)<2> fBUFFER_B(0, 0)<8;8,1> mov (16) r[a0.1,97]<2>:ub ubTempR(0, 0)<32;8,4> mov (16) r[a0.2,97]<2>:ub ubTempG(0, 0)<32;8,4> mov (16) r[a0.0,97]<2>:ub ubTempB(0, 0)<32;8,4>