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
|
//===-- AMDGPUAnnotateUniformValues.cpp - ---------------------------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
/// \file
/// This pass adds amdgpu.uniform metadata to IR values so this information
/// can be used during instruction selection.
//
//===----------------------------------------------------------------------===//
#include "AMDGPU.h"
#include "AMDGPUMemoryUtils.h"
#include "Utils/AMDGPUBaseInfo.h"
#include "llvm/Analysis/AliasAnalysis.h"
#include "llvm/Analysis/MemorySSA.h"
#include "llvm/Analysis/UniformityAnalysis.h"
#include "llvm/IR/InstVisitor.h"
#include "llvm/InitializePasses.h"
#define DEBUG_TYPE "amdgpu-annotate-uniform"
using namespace llvm;
namespace {
class AMDGPUAnnotateUniformValues
: public InstVisitor<AMDGPUAnnotateUniformValues> {
UniformityInfo *UA;
MemorySSA *MSSA;
AliasAnalysis *AA;
bool isEntryFunc;
bool Changed = false;
void setUniformMetadata(Instruction *I) {
I->setMetadata("amdgpu.uniform", MDNode::get(I->getContext(), {}));
Changed = true;
}
void setNoClobberMetadata(Instruction *I) {
I->setMetadata("amdgpu.noclobber", MDNode::get(I->getContext(), {}));
Changed = true;
}
public:
AMDGPUAnnotateUniformValues(UniformityInfo &UA, MemorySSA &MSSA,
AliasAnalysis &AA, const Function &F)
: UA(&UA), MSSA(&MSSA), AA(&AA),
isEntryFunc(AMDGPU::isEntryFunctionCC(F.getCallingConv())) {}
void visitBranchInst(BranchInst &I);
void visitLoadInst(LoadInst &I);
bool changed() const { return Changed; }
};
} // End anonymous namespace
void AMDGPUAnnotateUniformValues::visitBranchInst(BranchInst &I) {
if (UA->isUniform(&I))
setUniformMetadata(&I);
}
void AMDGPUAnnotateUniformValues::visitLoadInst(LoadInst &I) {
Value *Ptr = I.getPointerOperand();
if (!UA->isUniform(Ptr))
return;
Instruction *PtrI = dyn_cast<Instruction>(Ptr);
if (PtrI)
setUniformMetadata(PtrI);
// We're tracking up to the Function boundaries, and cannot go beyond because
// of FunctionPass restrictions. We can ensure that is memory not clobbered
// for memory operations that are live in to entry points only.
if (!isEntryFunc)
return;
bool GlobalLoad = I.getPointerAddressSpace() == AMDGPUAS::GLOBAL_ADDRESS;
if (GlobalLoad && !AMDGPU::isClobberedInFunction(&I, MSSA, AA))
setNoClobberMetadata(&I);
}
PreservedAnalyses
AMDGPUAnnotateUniformValuesPass::run(Function &F,
FunctionAnalysisManager &FAM) {
UniformityInfo &UI = FAM.getResult<UniformityInfoAnalysis>(F);
MemorySSA &MSSA = FAM.getResult<MemorySSAAnalysis>(F).getMSSA();
AAResults &AA = FAM.getResult<AAManager>(F);
AMDGPUAnnotateUniformValues Impl(UI, MSSA, AA, F);
Impl.visit(F);
if (!Impl.changed())
return PreservedAnalyses::all();
PreservedAnalyses PA = PreservedAnalyses::none();
// TODO: Should preserve nearly everything
PA.preserveSet<CFGAnalyses>();
return PA;
}
class AMDGPUAnnotateUniformValuesLegacy : public FunctionPass {
public:
static char ID;
AMDGPUAnnotateUniformValuesLegacy() : FunctionPass(ID) {}
bool doInitialization(Module &M) override { return false; }
bool runOnFunction(Function &F) override;
StringRef getPassName() const override {
return "AMDGPU Annotate Uniform Values";
}
void getAnalysisUsage(AnalysisUsage &AU) const override {
AU.addRequired<UniformityInfoWrapperPass>();
AU.addRequired<MemorySSAWrapperPass>();
AU.addRequired<AAResultsWrapperPass>();
AU.setPreservesAll();
}
};
bool AMDGPUAnnotateUniformValuesLegacy::runOnFunction(Function &F) {
if (skipFunction(F))
return false;
UniformityInfo &UI =
getAnalysis<UniformityInfoWrapperPass>().getUniformityInfo();
MemorySSA &MSSA = getAnalysis<MemorySSAWrapperPass>().getMSSA();
AliasAnalysis &AA = getAnalysis<AAResultsWrapperPass>().getAAResults();
AMDGPUAnnotateUniformValues Impl(UI, MSSA, AA, F);
Impl.visit(F);
return Impl.changed();
}
INITIALIZE_PASS_BEGIN(AMDGPUAnnotateUniformValuesLegacy, DEBUG_TYPE,
"Add AMDGPU uniform metadata", false, false)
INITIALIZE_PASS_DEPENDENCY(UniformityInfoWrapperPass)
INITIALIZE_PASS_DEPENDENCY(MemorySSAWrapperPass)
INITIALIZE_PASS_DEPENDENCY(AAResultsWrapperPass)
INITIALIZE_PASS_END(AMDGPUAnnotateUniformValuesLegacy, DEBUG_TYPE,
"Add AMDGPU uniform metadata", false, false)
char AMDGPUAnnotateUniformValuesLegacy::ID = 0;
FunctionPass *llvm::createAMDGPUAnnotateUniformValuesLegacy() {
return new AMDGPUAnnotateUniformValuesLegacy();
}
|
