| 1 |
/* |
| 2 |
* Copyright © <2010>, Intel Corporation. |
| 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 |
*/ |
| 39 |
// Module name: saveNV12_16x4T.asm |
| 40 |
// |
| 41 |
// Transpose 16x4 to 4x16 YNV12 data and write to memory |
| 42 |
// |
| 43 |
//---------------------------------------------------------------- |
| 44 |
// Symbols need to be defined before including this module |
| 45 |
// |
| 46 |
// Left MB region: |
| 47 |
// PREV_MB_YB: Base=rxx ElementSize=1 SrcRegion=REGION(16,1) Type=ub |
| 48 |
// PREV_MB_UW: Base=ryy ElementSize=2 SrcRegion=REGION(8,1) Type=uw |
| 49 |
|
| 50 |
// Binding table index: |
| 51 |
// BI_SRC_Y: Binding table index of Y surface |
| 52 |
// BI_SRC_UV: Binding table index of UV surface (NV12) |
| 53 |
// |
| 54 |
// Temp buffer: |
| 55 |
// BUF_B: BUF_B Base=rxx ElementSize=1 SrcRegion=REGION(16,1) Type=ub |
| 56 |
// BUF_W: BUF_W Base=rxx ElementSize=1 SrcRegion=REGION(8,1) Type=uw |
| 57 |
// |
| 58 |
// |
| 59 |
#if defined(_DEBUG) |
| 60 |
mov (1) EntrySignatureC:w 0xDDD6:w |
| 61 |
#endif |
| 62 |
|
| 63 |
mov (2) MSGSRC.0<1>:ud ORIX_LEFT<2;2,1>:w // Block origin |
| 64 |
mov (1) MSGSRC.2<1>:ud 0x000F0003:ud // 4x16 |
| 65 |
|
| 66 |
// Transpose Y, save them to MRFs |
| 67 |
|
| 68 |
// 16x4 Y src in GRF (each pix is specified as yx) |
| 69 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 70 |
// |f1 e1 d1 c1 b1 a1 91 81 71 61 51 41 31 21 11 01 f0 e0 d0 c0 b0 a0 90 80 70 60 50 40 30 20 10 00| |
| 71 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 72 |
// |f3 e3 d3 c3 b3 a3 93 83 73 63 53 43 33 23 13 03 f2 e2 d2 c2 b2 a2 92 82 72 62 52 42 32 22 12 02| |
| 73 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 74 |
// |
| 75 |
// First step (16) <1> <=== <16;4,1> |
| 76 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 77 |
// |73 63 53 43 72 62 52 42 71 61 51 41 70 60 50 40 33 23 13 03 32 22 12 02 31 21 11 01 30 20 10 00| |
| 78 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 79 |
// |f3 e3 d3 c3 f2 e2 d2 c2 f1 e1 d1 c1 f0 e0 d0 c0 b3 a3 93 83 b2 a2 92 82 b1 a1 91 81 b0 a0 90 80| |
| 80 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 81 |
|
| 82 |
// The first step |
| 83 |
mov (16) BUF_B(0,0)<1> PREV_MB_YB(0,0)<16;4,1> |
| 84 |
mov (16) BUF_B(0,16)<1> PREV_MB_YB(0,4)<16;4,1> |
| 85 |
mov (16) BUF_B(1,0)<1> PREV_MB_YB(0,8)<16;4,1> |
| 86 |
mov (16) BUF_B(1,16)<1> PREV_MB_YB(0,12)<16;4,1> |
| 87 |
|
| 88 |
// |
| 89 |
// Second step (16) <1> <=== <1;4,4> |
| 90 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 91 |
// |73 72 71 70 63 62 61 60 53 52 51 50 43 42 41 40 33 32 31 30 23 22 21 20 13 12 11 10 03 02 01 00| |
| 92 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 93 |
// |f3 f2 f1 f0 e3 e2 e1 e0 d3 d2 d1 d0 c3 c2 c1 c0 b3 b2 b1 b0 a3 a2 a1 a0 93 92 91 90 83 82 81 80| |
| 94 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 95 |
// |
| 96 |
// The second step |
| 97 |
// mov (16) MSGPAYLOADB(0,0)<1> BUF_B(0,0)<32;8,4> // Read 2 rows, write 1 row |
| 98 |
// mov (16) MSGPAYLOADB(0,16)<1> BUF_B(0,1)<32;8,4> |
| 99 |
// mov (16) MSGPAYLOADB(1,0)<1> BUF_B(0,2)<32;8,4> |
| 100 |
// mov (16) MSGPAYLOADB(1,16)<1> BUF_B(0,3)<32;8,4> |
| 101 |
|
| 102 |
mov (16) MSGPAYLOADB(0,0)<1> BUF_B(0,0)<1;4,4> |
| 103 |
mov (16) MSGPAYLOADB(0,16)<1> BUF_B(0,16)<1;4,4> |
| 104 |
mov (16) MSGPAYLOADB(1,0)<1> BUF_B(1,0)<1;4,4> |
| 105 |
mov (16) MSGPAYLOADB(1,16)<1> BUF_B(1,16)<1;4,4> |
| 106 |
|
| 107 |
// Transposed Y in 4x16 is ready for writting to dataport. |
| 108 |
// |
| 109 |
send (8) NULLREG MSGHDR MSGSRC<8;8,1>:ud DAPWRITE MSG_LEN(2)+DWBWMSGDSC+BI_DEST_Y // Write 2 GRFs |
| 110 |
|
| 111 |
|
| 112 |
|
| 113 |
///////////////////////////////////////////////////////////////////////////////////////////////////// |
| 114 |
|
| 115 |
// Transpose U/V, save them to MRFs in NV12 format |
| 116 |
asr (1) MSGSRC.1:ud MSGSRC.1:ud 1:w // NV12 U+V block origin y = half of Y comp |
| 117 |
mov (1) MSGSRC.2<1>:ud 0x00070003:ud // NV12 U+V block width and height (4x8) |
| 118 |
|
| 119 |
|
| 120 |
// 16x2 UV src in GRF (each pix is specified as yx) |
| 121 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 122 |
// |71 71 61 61 51 51 41 41 31 31 21 21 11 11 01 01 70 70 60 60 50 50 40 40 30 30 20 20 10 10 00 00| |
| 123 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 124 |
// |
| 125 |
// First step (8) <1> <=== <8;4,1>:w |
| 126 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 127 |
// |71 71 61 61 51 51 41 41 70 70 60 60 50 50 40 40 31 31 21 21 11 11 01 01 30 30 20 20 10 10 00 00| |
| 128 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 129 |
mov (8) BUF_W(0,0)<1> PREV_MB_UW(0,0)<8;4,1> |
| 130 |
mov (8) BUF_W(0,8)<1> PREV_MB_UW(0,4)<8;4,1> |
| 131 |
|
| 132 |
// Second step (8) <1> <=== <1;2,4> |
| 133 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 134 |
// |71 71 70 70 61 61 60 60 51 51 50 50 41 41 40 40 31 31 30 30 21 21 20 20 11 11 10 10 01 01 00 00| |
| 135 |
// +-----------------------+-----------------------+-----------------------+-----------------------+ |
| 136 |
mov (8) MSGPAYLOADW(0,0)<1> BUF_W(0,0)<1;2,4> |
| 137 |
mov (8) MSGPAYLOADW(0,8)<1> BUF_W(0,8)<1;2,4> |
| 138 |
|
| 139 |
// Transposed U+V in NV12 in 4x8 is ready for writting to dataport. |
| 140 |
|
| 141 |
send (8) NULLREG MSGHDR MSGSRC<8;8,1>:ud DAPWRITE MSG_LEN(1)+DWBWMSGDSC+BI_DEST_UV // Write 1 GRF |
| 142 |
|
| 143 |
|