Debian Codegraph @

src/shaders/post_processing/gen5_6/Common/RGBX_Save_YUV_Fix.asm

raw
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 * Authors:
40 * Halley Zhao <[email protected]>
41 */
42
43 // Module name: PL16x8_PL8x4.asm
44 //----------------------------------------------------------------
45
46 #include "RGBX_Load_16x8.inc"
47
48 #if (0)
49 #define nTEMP0 34 // transformation coefficient
50 #define nTEMP1 35 // one row of Y (first half register is used)
51 #define nTEMP2 36 // first half of one row
52 #define nTEMP3 37 // second half of one row
53 #define nTEMP4 38 // mul and add
54 #define nTEMP5 39 // mul and add
55 #define nTEMP6 40 // mul and add
56 #define nTEMP7 41 // mul and add
57 #define nTEMP8 42 // sum of mul
58 #define nTEMP10 44
59 #define nTEMP12 46
60 #define nTEMP14 48
61 #define nTEMP16 50
62 #define nTEMP17 51
63 #define nTEMP18 52
64
65 #define nTEMP24 58
66 #endif
67
68 $for(0; <nY_NUM_OF_ROWS; 1) {
69 // BGRX | B | G | R | X |
70 // ###### do on row for Y
71 // #### mul and add
72 mul (16) REG2(r, nTEMP4, 0)<1>:uw r[SRC_RGBA_OFFSET_1, %1*32 + 0]<0; 16,1>:ub ubRGB_to_Y_Coef_Fix<0;4,1>:ub
73 mul (16) REG2(r, nTEMP5, 0)<1>:uw r[SRC_RGBA_OFFSET_1, %1*32 + 16]<0; 16,1>:ub ubRGB_to_Y_Coef_Fix<0;4,1>:ub
74 mul (16) REG2(r, nTEMP6, 0)<1>:uw r[SRC_RGBA_OFFSET_2, %1*32 + 0]<0; 16,1>:ub ubRGB_to_Y_Coef_Fix<0;4,1>:ub
75 mul (16) REG2(r, nTEMP7, 0)<1>:uw r[SRC_RGBA_OFFSET_2, %1*32 + 16]<0; 16,1>:ub ubRGB_to_Y_Coef_Fix<0;4,1>:ub
76
77 add (4) REG2(r, nTEMP4, 0)<4>:uw REG2(r, nTEMP4, 0)<0;4,4>:uw REG2(r, nTEMP4, 1)<0;4,4>:uw
78 add (4) REG2(r, nTEMP5, 0)<4>:uw REG2(r, nTEMP5, 0)<0;4,4>:uw REG2(r, nTEMP5, 1)<0;4,4>:uw
79 add (4) REG2(r, nTEMP6, 0)<4>:uw REG2(r, nTEMP6, 0)<0;4,4>:uw REG2(r, nTEMP6, 1)<0;4,4>:uw
80 add (4) REG2(r, nTEMP7, 0)<4>:uw REG2(r, nTEMP7, 0)<0;4,4>:uw REG2(r, nTEMP7, 1)<0;4,4>:uw
81 add (4) REG2(r, nTEMP4, 0)<4>:uw REG2(r, nTEMP4, 0)<0;4,4>:uw REG2(r, nTEMP4, 2)<0;4,4>:uw
82 add (4) REG2(r, nTEMP5, 0)<4>:uw REG2(r, nTEMP5, 0)<0;4,4>:uw REG2(r, nTEMP5, 2)<0;4,4>:uw
83 add (4) REG2(r, nTEMP6, 0)<4>:uw REG2(r, nTEMP6, 0)<0;4,4>:uw REG2(r, nTEMP6, 2)<0;4,4>:uw
84 add (4) REG2(r, nTEMP7, 0)<4>:uw REG2(r, nTEMP7, 0)<0;4,4>:uw REG2(r, nTEMP7, 2)<0;4,4>:uw
85
86 // #### write Y to the 1 row
87 mov (4) REG2(r, nTEMP8, 0)<1>:uw REG2(r, nTEMP4, 0)<0; 4, 4>:uw
88 mov (4) REG2(r, nTEMP8, 4)<1>:uw REG2(r, nTEMP5, 0)<0; 4, 4>:uw
89 mov (4) REG2(r, nTEMP8, 8)<1>:uw REG2(r, nTEMP6, 0)<0; 4, 4>:uw
90 mov (4) REG2(r, nTEMP8, 12)<1>:uw REG2(r, nTEMP7, 0)<0; 4, 4>:uw
91 add (16) REG2(r, nTEMP8, 0)<1>:uw REG2(r, nTEMP8, 0)<0; 16, 1>:uw 0x1080:uw
92 mov (16) REG2(r, nTEMP8, 0)<1>:ub REG2(r, nTEMP8, 1)<0; 16, 2>:ub
93 mov (16) uwDEST_Y(%1)<1> REG2(r,nTEMP8, 0)<0;16,1>:ub
94
95 // ###### do one row for U
96 // #### mul and add
97 mul (16) REG2(r, nTEMP4, 0)<1>:w r[SRC_RGBA_OFFSET_1, %1*32 + 0]<0; 16,1>:ub bRGB_to_U_Coef_Fix<0;4,1>:b
98 mul (16) REG2(r, nTEMP5, 0)<1>:w r[SRC_RGBA_OFFSET_1, %1*32 + 16]<0; 16,1>:ub bRGB_to_U_Coef_Fix<0;4,1>:b
99 mul (16) REG2(r, nTEMP6, 0)<1>:w r[SRC_RGBA_OFFSET_2, %1*32 + 0]<0; 16,1>:ub bRGB_to_U_Coef_Fix<0;4,1>:b
100 mul (16) REG2(r, nTEMP7, 0)<1>:w r[SRC_RGBA_OFFSET_2, %1*32 + 16]<0; 16,1>:ub bRGB_to_U_Coef_Fix<0;4,1>:b
101
102 add (4) REG2(r, nTEMP4, 0)<4>:w REG2(r, nTEMP4, 0)<0;4,4>:w REG2(r, nTEMP4, 1)<0;4,4>:w
103 add (4) REG2(r, nTEMP5, 0)<4>:w REG2(r, nTEMP5, 0)<0;4,4>:w REG2(r, nTEMP5, 1)<0;4,4>:w
104 add (4) REG2(r, nTEMP6, 0)<4>:w REG2(r, nTEMP6, 0)<0;4,4>:w REG2(r, nTEMP6, 1)<0;4,4>:w
105 add (4) REG2(r, nTEMP7, 0)<4>:w REG2(r, nTEMP7, 0)<0;4,4>:w REG2(r, nTEMP7, 1)<0;4,4>:w
106 add (4) REG2(r, nTEMP4, 0)<4>:w REG2(r, nTEMP4, 0)<0;4,4>:w REG2(r, nTEMP4, 2)<0;4,4>:w
107 add (4) REG2(r, nTEMP5, 0)<4>:w REG2(r, nTEMP5, 0)<0;4,4>:w REG2(r, nTEMP5, 2)<0;4,4>:w
108 add (4) REG2(r, nTEMP6, 0)<4>:w REG2(r, nTEMP6, 0)<0;4,4>:w REG2(r, nTEMP6, 2)<0;4,4>:w
109 add (4) REG2(r, nTEMP7, 0)<4>:w REG2(r, nTEMP7, 0)<0;4,4>:w REG2(r, nTEMP7, 2)<0;4,4>:w
110
111 // #### write U to the 1 row
112 mov (4) REG2(r, nTEMP8, 0)<1>:w REG2(r, nTEMP4, 0)<0; 4, 4>:w
113 mov (4) REG2(r, nTEMP8, 4)<1>:w REG2(r, nTEMP5, 0)<0; 4, 4>:w
114 mov (4) REG2(r, nTEMP8, 8)<1>:w REG2(r, nTEMP6, 0)<0; 4, 4>:w
115 mov (4) REG2(r, nTEMP8, 12)<1>:w REG2(r, nTEMP7, 0)<0; 4, 4>:w
116 add (16) REG2(r, nTEMP8, 0)<1>:uw REG2(r, nTEMP8, 0)<0; 16, 1>:w 0x8080:uw
117 mov (16) REG2(r, nTEMP8, 0)<1>:ub REG2(r, nTEMP8, 1)<0; 16, 2>:ub
118 mov (16) uwDEST_U(%1)<1> REG2(r,nTEMP8, 0)<0;16,1>:ub
119
120 // ###### do one row for V
121 // #### mul and add
122 mul (16) REG2(r, nTEMP4, 0)<1>:w r[SRC_RGBA_OFFSET_1, %1*32 + 0]<0; 16,1>:ub bRGB_to_V_Coef_Fix<0;4,1>:b
123 mul (16) REG2(r, nTEMP5, 0)<1>:w r[SRC_RGBA_OFFSET_1, %1*32 + 16]<0; 16,1>:ub bRGB_to_V_Coef_Fix<0;4,1>:b
124 mul (16) REG2(r, nTEMP6, 0)<1>:w r[SRC_RGBA_OFFSET_2, %1*32 + 0]<0; 16,1>:ub bRGB_to_V_Coef_Fix<0;4,1>:b
125 mul (16) REG2(r, nTEMP7, 0)<1>:w r[SRC_RGBA_OFFSET_2, %1*32 + 16]<0; 16,1>:ub bRGB_to_V_Coef_Fix<0;4,1>:b
126
127 add (4) REG2(r, nTEMP4, 0)<4>:w REG2(r, nTEMP4, 0)<0;4,4>:w REG2(r, nTEMP4, 1)<0;4,4>:w
128 add (4) REG2(r, nTEMP5, 0)<4>:w REG2(r, nTEMP5, 0)<0;4,4>:w REG2(r, nTEMP5, 1)<0;4,4>:w
129 add (4) REG2(r, nTEMP6, 0)<4>:w REG2(r, nTEMP6, 0)<0;4,4>:w REG2(r, nTEMP6, 1)<0;4,4>:w
130 add (4) REG2(r, nTEMP7, 0)<4>:w REG2(r, nTEMP7, 0)<0;4,4>:w REG2(r, nTEMP7, 1)<0;4,4>:w
131 add (4) REG2(r, nTEMP4, 0)<4>:w REG2(r, nTEMP4, 0)<0;4,4>:w REG2(r, nTEMP4, 2)<0;4,4>:w
132 add (4) REG2(r, nTEMP5, 0)<4>:w REG2(r, nTEMP5, 0)<0;4,4>:w REG2(r, nTEMP5, 2)<0;4,4>:w
133 add (4) REG2(r, nTEMP6, 0)<4>:w REG2(r, nTEMP6, 0)<0;4,4>:w REG2(r, nTEMP6, 2)<0;4,4>:w
134 add (4) REG2(r, nTEMP7, 0)<4>:w REG2(r, nTEMP7, 0)<0;4,4>:w REG2(r, nTEMP7, 2)<0;4,4>:w
135
136 // #### write V to the 1 row
137 mov (4) REG2(r, nTEMP8, 0)<1>:w REG2(r, nTEMP4, 0)<0; 4, 4>:w
138 mov (4) REG2(r, nTEMP8, 4)<1>:w REG2(r, nTEMP5, 0)<0; 4, 4>:w
139 mov (4) REG2(r, nTEMP8, 8)<1>:w REG2(r, nTEMP6, 0)<0; 4, 4>:w
140 mov (4) REG2(r, nTEMP8, 12)<1>:w REG2(r, nTEMP7, 0)<0; 4, 4>:w
141 add (16) REG2(r, nTEMP8, 0)<1>:uw REG2(r, nTEMP8, 0)<0; 16, 1>:w 0x8080:uw
142 mov (16) REG2(r, nTEMP8, 0)<1>:ub REG2(r, nTEMP8, 1)<0; 16, 2>:ub
143 mov (16) uwDEST_V(%1)<1> REG2(r,nTEMP8, 0)<0;16,1>:ub
144 }
145