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
|
//===-- AMDGPUTargetTransformInfo.cpp - AMDGPU specific TTI pass ---------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// \file
// This file implements a TargetTransformInfo analysis pass specific to the
// AMDGPU target machine. It uses the target's detailed information to provide
// more precise answers to certain TTI queries, while letting the target
// independent and default TTI implementations handle the rest.
//
//===----------------------------------------------------------------------===//
#include "AMDGPU.h"
#include "AMDGPUTargetMachine.h"
#include "llvm/Analysis/LoopInfo.h"
#include "llvm/Analysis/TargetTransformInfo.h"
#include "llvm/Analysis/ValueTracking.h"
#include "llvm/Support/Debug.h"
#include "llvm/Target/CostTable.h"
#include "llvm/Target/TargetLowering.h"
using namespace llvm;
#define DEBUG_TYPE "AMDGPUtti"
// Declare the pass initialization routine locally as target-specific passes
// don't have a target-wide initialization entry point, and so we rely on the
// pass constructor initialization.
namespace llvm {
void initializeAMDGPUTTIPass(PassRegistry &);
}
namespace {
class AMDGPUTTI final : public ImmutablePass, public TargetTransformInfo {
const AMDGPUTargetMachine *TM;
const AMDGPUSubtarget *ST;
const AMDGPUTargetLowering *TLI;
/// Estimate the overhead of scalarizing an instruction. Insert and Extract
/// are set if the result needs to be inserted and/or extracted from vectors.
unsigned getScalarizationOverhead(Type *Ty, bool Insert, bool Extract) const;
public:
AMDGPUTTI() : ImmutablePass(ID), TM(nullptr), ST(nullptr), TLI(nullptr) {
llvm_unreachable("This pass cannot be directly constructed");
}
AMDGPUTTI(const AMDGPUTargetMachine *TM)
: ImmutablePass(ID), TM(TM), ST(TM->getSubtargetImpl()),
TLI(TM->getTargetLowering()) {
initializeAMDGPUTTIPass(*PassRegistry::getPassRegistry());
}
void initializePass() override { pushTTIStack(this); }
void getAnalysisUsage(AnalysisUsage &AU) const override {
TargetTransformInfo::getAnalysisUsage(AU);
}
/// Pass identification.
static char ID;
/// Provide necessary pointer adjustments for the two base classes.
void *getAdjustedAnalysisPointer(const void *ID) override {
if (ID == &TargetTransformInfo::ID)
return (TargetTransformInfo *)this;
return this;
}
bool hasBranchDivergence() const override;
void getUnrollingPreferences(Loop *L,
UnrollingPreferences &UP) const override;
/// @}
};
} // end anonymous namespace
INITIALIZE_AG_PASS(AMDGPUTTI, TargetTransformInfo, "AMDGPUtti",
"AMDGPU Target Transform Info", true, true, false)
char AMDGPUTTI::ID = 0;
ImmutablePass *
llvm::createAMDGPUTargetTransformInfoPass(const AMDGPUTargetMachine *TM) {
return new AMDGPUTTI(TM);
}
bool AMDGPUTTI::hasBranchDivergence() const { return true; }
void AMDGPUTTI::getUnrollingPreferences(Loop *L,
UnrollingPreferences &UP) const {
for (Loop::block_iterator BI = L->block_begin(), BE = L->block_end();
BI != BE; ++BI) {
BasicBlock *BB = *BI;
for (BasicBlock::const_iterator I = BB->begin(), E = BB->end();
I != E; ++I) {
const GetElementPtrInst *GEP = dyn_cast<GetElementPtrInst>(I);
if (!GEP)
continue;
const Value *Ptr = GEP->getPointerOperand();
const AllocaInst *Alloca = dyn_cast<AllocaInst>(GetUnderlyingObject(Ptr));
if (Alloca) {
// We want to do whatever we can to limit the number of alloca
// instructions that make it through to the code generator. allocas
// require us to use indirect addressing, which is slow and prone to
// compiler bugs. If this loop does an address calculation on an
// alloca ptr, then we want to use a higher than normal loop unroll
// threshold. This will give SROA a better chance to eliminate these
// allocas.
//
// Don't use the maximum allowed value here as it will make some
// programs way too big.
UP.Threshold = 500;
}
}
}
}
|
