| 1 |
/* |
| 2 |
* All Video Processing kernels |
| 3 |
* Copyright © <2010>, Intel Corporation. |
| 4 |
* |
| 5 |
* Permission is hereby granted, free of charge, to any person obtaining a |
| 6 |
* copy of this software and associated documentation files (the |
| 7 |
* "Software"), to deal in the Software without restriction, including |
| 8 |
* without limitation the rights to use, copy, modify, merge, publish, |
| 9 |
* distribute, sub license, and/or sell copies of the Software, and to |
| 10 |
* permit persons to whom the Software is furnished to do so, subject to |
| 11 |
* the following conditions: |
| 12 |
* |
| 13 |
* The above copyright notice and this permission notice (including the |
| 14 |
* next paragraph) shall be included in all copies or substantial portions |
| 15 |
* of the Software. |
| 16 |
* |
| 17 |
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS |
| 18 |
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF |
| 19 |
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. |
| 20 |
* IN NO EVENT SHALL PRECISION INSIGHT AND/OR ITS SUPPLIERS BE LIABLE FOR |
| 21 |
* ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, |
| 22 |
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE |
| 23 |
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. |
| 24 |
* |
| 25 |
* This file was originally licensed under the following license |
| 26 |
* |
| 27 |
* Licensed under the Apache License, Version 2.0 (the "License"); |
| 28 |
* you may not use this file except in compliance with the License. |
| 29 |
* You may obtain a copy of the License at |
| 30 |
* |
| 31 |
* http://www.apache.org/licenses/LICENSE-2.0 |
| 32 |
* |
| 33 |
* Unless required by applicable law or agreed to in writing, software |
| 34 |
* distributed under the License is distributed on an "AS IS" BASIS, |
| 35 |
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 36 |
* See the License for the specific language governing permissions and |
| 37 |
* limitations under the License. |
| 38 |
* |
| 39 |
*/ |
| 40 |
|
| 41 |
|
| 42 |
#include "PL4x8_Save_NV11.inc" |
| 43 |
|
| 44 |
mov (8) mMSGHDR<1>:ud rMSGSRC<8;8,1>:ud |
| 45 |
|
| 46 |
#if !defined(SAVE_UV_ONLY) |
| 47 |
// Save current planar frame Y block data (16x8) ------------------------------- |
| 48 |
|
| 49 |
mov (2) mMSGHDR.0<1>:d wORIX<2;2,1>:w // Block origin |
| 50 |
mov (1) mMSGHDR.2<1>:ud nDPW_BLOCK_SIZE_Y:ud // Block width and height (16x8) |
| 51 |
|
| 52 |
///* Yoni - masking is not relevant for ILK?!? |
| 53 |
// |
| 54 |
// //Use the mask to determine which pixels shouldn't be over-written |
| 55 |
// cmp.ge.f0.0 (1) NULLREG BLOCK_MASK_D:ud 0x00FFFFFF:ud //Check if all pixels in the block need to be modified |
| 56 |
// (f0.0) jmpi WritePlanarToDataPort |
| 57 |
// |
| 58 |
// //If mask is not all 1's, then load the entire 16x8 block |
| 59 |
// //so that only those bytes may be modified that need to be (using the mask) |
| 60 |
// send (8) SRC_YD(0)<1> MSGHDR MSGSRC<8;8,1>:ud DWBRMSGDSC+0x00040000+BI_DEST_Y:ud //16x8 |
| 61 |
// |
| 62 |
// asr (2) MSGSRC.0<1>:ud ORIX<2;2,1>:w 1:w // U/V block origin should be half of Y's |
| 63 |
// mov (1) MSGSRC.2<1>:ud 0x00030007:ud // Block width and height (8x4) |
| 64 |
// send (8) SRC_UD(0)<1> MSGHDR MSGSRC<8;8,1>:ud DWBRMSGDSC+0x00010000+BI_DEST_U:ud |
| 65 |
// send (8) SRC_VD(0)<1> MSGHDR MSGSRC<8;8,1>:ud DWBRMSGDSC+0x00010000+BI_DEST_V:ud |
| 66 |
// |
| 67 |
// //Restore the origin information |
| 68 |
// mov (2) MSGSRC.0<1>:ud ORIX<2;2,1>:w // Block origin |
| 69 |
// mov (1) MSGSRC.2<1>:ud 0x0007000F:ud // Block width and height (16x8) |
| 70 |
// |
| 71 |
// //expand U and V to be aligned on word boundary |
| 72 |
// mov (16) SRC_UW(1)<1> SRC_U(0,16) |
| 73 |
// mov (16) SRC_UW(0)<1> SRC_U(0, 0) |
| 74 |
// mov (16) SRC_VW(1)<1> SRC_V(0,16) |
| 75 |
// mov (16) SRC_VW(0)<1> SRC_V(0, 0) |
| 76 |
// |
| 77 |
// //Merge the data |
| 78 |
// mov (1) f0.1:uw BLOCK_MASK_V:uw //Load the mask on flag reg |
| 79 |
// (f0.1) mov (8) TEMP0<1>:uw BLOCK_MASK_H:uw |
| 80 |
// (-f0.1) mov (8) TEMP0<1>:uw 0:uw |
| 81 |
// |
| 82 |
// // Destination is Word aligned |
| 83 |
// $for(0; <Y_ROW_SIZE; 2) { |
| 84 |
// mov (1) f0.1:uw TEMP(0,%1)<0;1,0> |
| 85 |
// (-f0.1) mov (16) DEST_Y(0, %1*32)<2> SRC_Y(0, %1*16) |
| 86 |
// (-f0.1) mov (16) DEST_U(0, %1*8)<1> SRC_U(0, %1*8) //only works for Word aligned Byte data |
| 87 |
// (-f0.1) mov (16) DEST_V(0, %1*8)<1> SRC_V(0, %1*8) //only works for Word aligned Byte data |
| 88 |
// |
| 89 |
// mov (1) f0.1:uw TEMP(0,1+%1)<0;1,0> |
| 90 |
// (-f0.1) mov (16) DEST_Y(0, 1+%1*32)<2> SRC_Y(0, 1+%1*16) |
| 91 |
// |
| 92 |
// } |
| 93 |
// |
| 94 |
//*/ Yoni - masking is not relevant for ILK?!? |
| 95 |
|
| 96 |
WritePlanarToDataPort: |
| 97 |
$for(0,0; <nY_NUM_OF_ROWS; 2,1) { |
| 98 |
mov (16) mubMSGPAYLOAD(%2,0)<1> ub2DEST_Y(%1)REGION(16,2) |
| 99 |
mov (16) mubMSGPAYLOAD(%2,16)<1> ub2DEST_Y(%1+1)REGION(16,2) |
| 100 |
} |
| 101 |
send (8) dNULLREG mMSGHDR udDUMMY_NULL nDATAPORT_WRITE nDPMW_MSGDSC+nDPW_MSG_SIZE_Y+nBI_DESTINATION_Y:ud |
| 102 |
#endif |
| 103 |
|
| 104 |
// Save U/V data block in planar format (4x8) ---------------------------------- |
| 105 |
mov (2) rMSGSRC.0<1>:d wORIX<2;2,1>:w // Block origin |
| 106 |
asr (1) rMSGSRC.0<1>:d rMSGSRC.0<0;1,0>:d 1:w // U/V block origin should be half of Y's |
| 107 |
mov (8) mMSGHDR<1>:ud rMSGSRC<8;8,1>:ud |
| 108 |
|
| 109 |
$for(0,0; <nY_NUM_OF_ROWS;4,1) { |
| 110 |
mov (16) mubMSGPAYLOAD(%2,0)<2> ub2DEST_U(%2)REGION(16,2) |
| 111 |
mov (16) mubMSGPAYLOAD(%2,1)<2> ub2DEST_V(%2)REGION(16,2) |
| 112 |
} |
| 113 |
send (8) dNULLREG mMSGHDR udDUMMY_NULL nDATAPORT_WRITE nDPMW_MSGDSC+nDPW_MSG_SIZE_UV+nBI_DESTINATION_UV:ud |
| 114 |
|
| 115 |
// End of PL4x8_Save_NV11 |
| 116 |
|