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
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
|
<!---======================= begin_copyright_notice ============================
Copyright (C) 2020-2022 Intel Corporation
SPDX-License-Identifier: MIT
============================= end_copyright_notice ==========================-->
## Opcode
WAIT = 0x5b
## Format
| | |
| --- | --- |
| 0x5b(WAIT) | Mask |
## Semantics
```
//Each TDR is a struct with two fields: a bool valid bit and a thread id.
//It is initialized during thread dispatch
for (i = 0; i < 8; ++i) {
if (mask & (1<< i)) {
TDR[i].valid = false;
}
if (TDR[i].valid) {
wait for TDR[i].thread to finish;
}
}
```
## Description
If a thread dependency pattern is specified during the creation of the
kernel's thread space, this instruction will cause the thread to wait
until all of its dependency threads have finished their execution. The
8-bit thread dependency clear mask provides finer-grained dependency
control. Each bit corresponds to one of the eight threads this thread
may depend on; if set, dependency is cleared, and this thread will not
wait for the corresponding thread's termination.
- **Mask(scalar):** The thread dependency clear mask. Must have type UB
#### Properties
## Text
```
WAIT <Mask>
```
## Notes
This instruction may only be used with the global thread space execution model.
|
