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
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
|
<!---======================= begin_copyright_notice ============================
Copyright (C) 2020-2022 Intel Corporation
SPDX-License-Identifier: MIT
============================= end_copyright_notice ==========================-->
## Opcode
LRP = 0x0e
## Format
| | | | | | | |
| --- | --- | --- | --- | --- | --- | --- |
| 0x0e(LRP) | Exec_size | Pred | Dst | Src0 | Src1 | Src2 |
## Semantics
```
for (i = 0; i < exec_size; ++i) {
if (ChEn[i]) {
dst[i] = src1[i] * src0[i] + src2[i] * (1.0 - src0[i]);
}
}
```
## Description
```
For each enabled channel, add the product of <src0> and <src1> to the product of <src2> and (1.0 - <src0>).
```
- **Exec_size(ub):** Execution size
- Bit[2..0]: size of the region for source and destination operands
- 0b000: 1 element (scalar)
- 0b001: 2 elements
- 0b010: 4 elements
- 0b011: 8 elements
- 0b100: 16 elements
- 0b101: 32 elements
- Bit[7..4]: execution mask (explicit control over the enabled channels)
- 0b0000: M1
- 0b0001: M2
- 0b0010: M3
- 0b0011: M4
- 0b0100: M5
- 0b0101: M6
- 0b0110: M7
- 0b0111: M8
- 0b1000: M1_NM
- 0b1001: M2_NM
- 0b1010: M3_NM
- 0b1011: M4_NM
- 0b1100: M5_NM
- 0b1101: M6_NM
- 0b1110: M7_NM
- 0b1111: M8_NM
- **Pred(uw):** Predication control
- **Dst(vec_operand):** The destination operand. Operand class: general
- **Src0(vec_operand):** The first source operand. Operand class: general,immediate
- **Src1(vec_operand):** The second source operand. Operand class: general,immediate
- **Src2(vec_operand):** The third source operand. Operand class: general,immediate
#### Properties
- **Supported Types:** F
- **Saturation:** Yes
- **Source Modifier:** arithmetic
## Text
```
[(<P>)] LRP[.sat] (<exec_size>) <dst> <src0> <src1> <src2>
```
## Notes
Register region is ignored for Dst, and contiguous elements will be updated. Src0, Src1, and Src2 support scalar region, in which case the single value is broadcast to all channels; other regions are ignored and contiguous elements will be accessed. Dst, Src0, Src1, and Src2 must be 16 byte aligned, except for scalar source operands.
|
