blob: d5aa552a62b7a04e56211b03ba694eaffea81690 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
|
/*
* Copyright © <2010>, Intel Corporation.
*
* This program is licensed under the terms and conditions of the
* Eclipse Public License (EPL), version 1.0. The full text of the EPL is at
* http://www.opensource.org/licenses/eclipse-1.0.php.
*
*/
// Module name: loadNV12_16x16T.asm
//
// Load and transpose NV12 16x16 block
//
//----------------------------------------------------------------
// Symbols need to be defined before including this module
//
// Source region in :ud
// SRC_YD: SRC_YD Base=rxx ElementSize=4 SrcRegion=REGION(8,1) Type=ud // 8 GRFs
// SRC_UD: SRC_UD Base=rxx ElementSize=4 SrcRegion=REGION(8,1) Type=ud (U+V for NV12) // 4 GRFs
//
// Source region is :ub. The same region as :ud region
// SRC_YB: SRC_YB Base=rxx ElementSize=1 SrcRegion=REGION(16,1) Type=ub // 8 GRFs
// SRC_UB: SRC_UB Base=rxx ElementSize=1 SrcRegion=REGION(16,1) Type=ub // 2 GRFs
// SRC_VB: SRC_VB Base=rxx ElementSize=1 SrcRegion=REGION(16,1) Type=ub // 2 GRFs
//
// Binding table index:
// BI_SRC_Y: Binding table index of Y surface
// BI_SRC_UV: Binding table index of UV surface (NV12)
//
// Temp buffer:
// BUF_B: BUF_B Base=rxx ElementSize=1 SrcRegion=REGION(16,1) Type=ub
//
//----------------------------------------------------------------
#if defined(_DEBUG)
mov (1) EntrySignatureC:w 0xDDD1:w
#endif
// Read Y
mov (2) MSGSRC.0<1>:ud ORIX_CUR<2;2,1>:w // Block origin
mov (1) MSGSRC.2<1>:ud 0x000F000F:ud // Block width and height (16x16)
send (8) SRC_YD(0)<1> MSGHDRY MSGSRC<8;8,1>:ud DAPREAD RESP_LEN(8)+DWBRMSGDSC_RC+BI_SRC_Y // Read 8 GRFs
// Read U+V
asr (1) MSGSRC.1:ud MSGSRC.1:ud 1:w // NV12 U+V block origin y = half of Y comp
mov (1) MSGSRC.2<1>:ud 0x0007000F:ud // NV12 U+V block width and height (16x8)
send (8) SRC_UD(0)<1> MSGHDRU MSGSRC<8;8,1>:ud DAPREAD RESP_LEN(4)+DWBRMSGDSC_RC+BI_SRC_UV // Read 4 GRFs
#include "TransposeNV12_16x16.asm"
// #include "Transpose_Y_16x16.asm"
// #include "Transpose_NV12_UV_16x8.asm"
// End of loadNV12_16x16T
|