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
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
|
//===-- X86ShuffleDecode.cpp - X86 shuffle decode logic -------------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// Define several functions to decode x86 specific shuffle semantics into a
// generic vector mask.
//
//===----------------------------------------------------------------------===//
#include "X86ShuffleDecode.h"
//===----------------------------------------------------------------------===//
// Vector Mask Decoding
//===----------------------------------------------------------------------===//
namespace llvm {
void DecodeINSERTPSMask(unsigned Imm, SmallVectorImpl<int> &ShuffleMask) {
// Defaults the copying the dest value.
ShuffleMask.push_back(0);
ShuffleMask.push_back(1);
ShuffleMask.push_back(2);
ShuffleMask.push_back(3);
// Decode the immediate.
unsigned ZMask = Imm & 15;
unsigned CountD = (Imm >> 4) & 3;
unsigned CountS = (Imm >> 6) & 3;
// CountS selects which input element to use.
unsigned InVal = 4+CountS;
// CountD specifies which element of destination to update.
ShuffleMask[CountD] = InVal;
// ZMask zaps values, potentially overriding the CountD elt.
if (ZMask & 1) ShuffleMask[0] = SM_SentinelZero;
if (ZMask & 2) ShuffleMask[1] = SM_SentinelZero;
if (ZMask & 4) ShuffleMask[2] = SM_SentinelZero;
if (ZMask & 8) ShuffleMask[3] = SM_SentinelZero;
}
// <3,1> or <6,7,2,3>
void DecodeMOVHLPSMask(unsigned NElts, SmallVectorImpl<int> &ShuffleMask) {
for (unsigned i = NElts/2; i != NElts; ++i)
ShuffleMask.push_back(NElts+i);
for (unsigned i = NElts/2; i != NElts; ++i)
ShuffleMask.push_back(i);
}
// <0,2> or <0,1,4,5>
void DecodeMOVLHPSMask(unsigned NElts, SmallVectorImpl<int> &ShuffleMask) {
for (unsigned i = 0; i != NElts/2; ++i)
ShuffleMask.push_back(i);
for (unsigned i = 0; i != NElts/2; ++i)
ShuffleMask.push_back(NElts+i);
}
/// DecodePSHUFMask - This decodes the shuffle masks for pshufd, and vpermilp*.
/// VT indicates the type of the vector allowing it to handle different
/// datatypes and vector widths.
void DecodePSHUFMask(EVT VT, unsigned Imm, SmallVectorImpl<int> &ShuffleMask) {
unsigned NumElts = VT.getVectorNumElements();
unsigned NumLanes = VT.getSizeInBits() / 128;
unsigned NumLaneElts = NumElts / NumLanes;
int NewImm = Imm;
for (unsigned l = 0; l != NumElts; l += NumLaneElts) {
for (unsigned i = 0; i != NumLaneElts; ++i) {
ShuffleMask.push_back(NewImm % NumLaneElts + l);
NewImm /= NumLaneElts;
}
if (NumLaneElts == 4) NewImm = Imm; // reload imm
}
}
void DecodePSHUFHWMask(unsigned Imm, SmallVectorImpl<int> &ShuffleMask) {
ShuffleMask.push_back(0);
ShuffleMask.push_back(1);
ShuffleMask.push_back(2);
ShuffleMask.push_back(3);
for (unsigned i = 0; i != 4; ++i) {
ShuffleMask.push_back(4+(Imm & 3));
Imm >>= 2;
}
}
void DecodePSHUFLWMask(unsigned Imm, SmallVectorImpl<int> &ShuffleMask) {
for (unsigned i = 0; i != 4; ++i) {
ShuffleMask.push_back((Imm & 3));
Imm >>= 2;
}
ShuffleMask.push_back(4);
ShuffleMask.push_back(5);
ShuffleMask.push_back(6);
ShuffleMask.push_back(7);
}
/// DecodeSHUFPMask - This decodes the shuffle masks for shufp*. VT indicates
/// the type of the vector allowing it to handle different datatypes and vector
/// widths.
void DecodeSHUFPMask(EVT VT, unsigned Imm, SmallVectorImpl<int> &ShuffleMask) {
unsigned NumElts = VT.getVectorNumElements();
unsigned NumLanes = VT.getSizeInBits() / 128;
unsigned NumLaneElts = NumElts / NumLanes;
int NewImm = Imm;
for (unsigned l = 0; l != NumElts; l += NumLaneElts) {
// Part that reads from dest.
for (unsigned i = 0; i != NumLaneElts/2; ++i) {
ShuffleMask.push_back(NewImm % NumLaneElts + l);
NewImm /= NumLaneElts;
}
// Part that reads from src.
for (unsigned i = 0; i != NumLaneElts/2; ++i) {
ShuffleMask.push_back(NewImm % NumLaneElts + NumElts + l);
NewImm /= NumLaneElts;
}
if (NumLaneElts == 4) NewImm = Imm; // reload imm
}
}
/// DecodeUNPCKHMask - This decodes the shuffle masks for unpckhps/unpckhpd
/// and punpckh*. VT indicates the type of the vector allowing it to handle
/// different datatypes and vector widths.
void DecodeUNPCKHMask(EVT VT, SmallVectorImpl<int> &ShuffleMask) {
unsigned NumElts = VT.getVectorNumElements();
// Handle 128 and 256-bit vector lengths. AVX defines UNPCK* to operate
// independently on 128-bit lanes.
unsigned NumLanes = VT.getSizeInBits() / 128;
if (NumLanes == 0 ) NumLanes = 1; // Handle MMX
unsigned NumLaneElts = NumElts / NumLanes;
for (unsigned l = 0; l != NumElts; l += NumLaneElts) {
for (unsigned i = l + NumLaneElts/2, e = l + NumLaneElts; i != e; ++i) {
ShuffleMask.push_back(i); // Reads from dest/src1
ShuffleMask.push_back(i+NumElts); // Reads from src/src2
}
}
}
/// DecodeUNPCKLMask - This decodes the shuffle masks for unpcklps/unpcklpd
/// and punpckl*. VT indicates the type of the vector allowing it to handle
/// different datatypes and vector widths.
void DecodeUNPCKLMask(EVT VT, SmallVectorImpl<int> &ShuffleMask) {
unsigned NumElts = VT.getVectorNumElements();
// Handle 128 and 256-bit vector lengths. AVX defines UNPCK* to operate
// independently on 128-bit lanes.
unsigned NumLanes = VT.getSizeInBits() / 128;
if (NumLanes == 0 ) NumLanes = 1; // Handle MMX
unsigned NumLaneElts = NumElts / NumLanes;
for (unsigned l = 0; l != NumElts; l += NumLaneElts) {
for (unsigned i = l, e = l + NumLaneElts/2; i != e; ++i) {
ShuffleMask.push_back(i); // Reads from dest/src1
ShuffleMask.push_back(i+NumElts); // Reads from src/src2
}
}
}
void DecodeVPERM2X128Mask(EVT VT, unsigned Imm,
SmallVectorImpl<int> &ShuffleMask) {
if (Imm & 0x88)
return; // Not a shuffle
unsigned HalfSize = VT.getVectorNumElements()/2;
unsigned FstHalfBegin = (Imm & 0x3) * HalfSize;
unsigned SndHalfBegin = ((Imm >> 4) & 0x3) * HalfSize;
for (int i = FstHalfBegin, e = FstHalfBegin+HalfSize; i != e; ++i)
ShuffleMask.push_back(i);
for (int i = SndHalfBegin, e = SndHalfBegin+HalfSize; i != e; ++i)
ShuffleMask.push_back(i);
}
} // llvm namespace
|
