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
|
//===-- AMDGPUSubtarget.cpp - AMDGPU Subtarget Information ----------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
/// \file
/// \brief Implements the AMDGPU specific subclass of TargetSubtarget.
//
//===----------------------------------------------------------------------===//
#include "AMDGPUSubtarget.h"
#include "R600ISelLowering.h"
#include "R600InstrInfo.h"
#include "R600MachineScheduler.h"
#include "SIISelLowering.h"
#include "SIInstrInfo.h"
#include "SIMachineFunctionInfo.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/CodeGen/MachineScheduler.h"
using namespace llvm;
#define DEBUG_TYPE "amdgpu-subtarget"
#define GET_SUBTARGETINFO_ENUM
#define GET_SUBTARGETINFO_TARGET_DESC
#define GET_SUBTARGETINFO_CTOR
#include "AMDGPUGenSubtargetInfo.inc"
AMDGPUSubtarget &
AMDGPUSubtarget::initializeSubtargetDependencies(const Triple &TT,
StringRef GPU, StringRef FS) {
// Determine default and user-specified characteristics
// On SI+, we want FP64 denormals to be on by default. FP32 denormals can be
// enabled, but some instructions do not respect them and they run at the
// double precision rate, so don't enable by default.
//
// We want to be able to turn these off, but making this a subtarget feature
// for SI has the unhelpful behavior that it unsets everything else if you
// disable it.
SmallString<256> FullFS("+promote-alloca,+fp64-denormals,");
FullFS += FS;
if (GPU == "" && TT.getArch() == Triple::amdgcn)
GPU = "SI";
ParseSubtargetFeatures(GPU, FullFS);
// FIXME: I don't think think Evergreen has any useful support for
// denormals, but should be checked. Should we issue a warning somewhere
// if someone tries to enable these?
if (getGeneration() <= AMDGPUSubtarget::NORTHERN_ISLANDS) {
FP32Denormals = false;
FP64Denormals = false;
}
return *this;
}
AMDGPUSubtarget::AMDGPUSubtarget(const Triple &TT, StringRef GPU, StringRef FS,
TargetMachine &TM)
: AMDGPUGenSubtargetInfo(TT, GPU, FS), DevName(GPU), Is64bit(false),
DumpCode(false), R600ALUInst(false), HasVertexCache(false),
TexVTXClauseSize(0), Gen(AMDGPUSubtarget::R600), FP64(false),
FP64Denormals(false), FP32Denormals(false), FastFMAF32(false),
CaymanISA(false), FlatAddressSpace(false), EnableIRStructurizer(true),
EnablePromoteAlloca(false), EnableIfCvt(true), EnableLoadStoreOpt(false),
EnableUnsafeDSOffsetFolding(false),
WavefrontSize(0), CFALUBug(false), LocalMemorySize(0),
EnableVGPRSpilling(false), SGPRInitBug(false), IsGCN(false),
GCN1Encoding(false), GCN3Encoding(false), CIInsts(false), LDSBankCount(0),
IsaVersion(ISAVersion0_0_0), EnableHugeScratchBuffer(false),
FrameLowering(TargetFrameLowering::StackGrowsUp,
64 * 16, // Maximum stack alignment (long16)
0),
InstrItins(getInstrItineraryForCPU(GPU)), TargetTriple(TT) {
initializeSubtargetDependencies(TT, GPU, FS);
if (getGeneration() <= AMDGPUSubtarget::NORTHERN_ISLANDS) {
InstrInfo.reset(new R600InstrInfo(*this));
TLInfo.reset(new R600TargetLowering(TM, *this));
} else {
InstrInfo.reset(new SIInstrInfo(*this));
TLInfo.reset(new SITargetLowering(TM, *this));
}
}
unsigned AMDGPUSubtarget::getStackEntrySize() const {
assert(getGeneration() <= NORTHERN_ISLANDS);
switch(getWavefrontSize()) {
case 16:
return 8;
case 32:
return hasCaymanISA() ? 4 : 8;
case 64:
return 4;
default:
llvm_unreachable("Illegal wavefront size.");
}
}
unsigned AMDGPUSubtarget::getAmdKernelCodeChipID() const {
switch(getGeneration()) {
default: llvm_unreachable("ChipID unknown");
case SEA_ISLANDS: return 12;
}
}
AMDGPU::IsaVersion AMDGPUSubtarget::getIsaVersion() const {
return AMDGPU::getIsaVersion(getFeatureBits());
}
bool AMDGPUSubtarget::isVGPRSpillingEnabled(
const SIMachineFunctionInfo *MFI) const {
return MFI->getShaderType() == ShaderType::COMPUTE || EnableVGPRSpilling;
}
void AMDGPUSubtarget::overrideSchedPolicy(MachineSchedPolicy &Policy,
MachineInstr *begin,
MachineInstr *end,
unsigned NumRegionInstrs) const {
if (getGeneration() >= SOUTHERN_ISLANDS) {
// Track register pressure so the scheduler can try to decrease
// pressure once register usage is above the threshold defined by
// SIRegisterInfo::getRegPressureSetLimit()
Policy.ShouldTrackPressure = true;
// Enabling both top down and bottom up scheduling seems to give us less
// register spills than just using one of these approaches on its own.
Policy.OnlyTopDown = false;
Policy.OnlyBottomUp = false;
}
}
|
