1//===- AVRShift.cpp - Shift Expansion Pass --------------------------------===//
2//
3// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6//
7//===----------------------------------------------------------------------===//
8//
9/// \file
10/// Expand non-8-bit and non-16-bit shift instructions (shl, lshr, ashr) to
11/// inline loops, just like avr-gcc. This must be done in IR because otherwise
12/// the type legalizer will turn 32-bit shifts into (non-existing) library calls
13/// such as __ashlsi3.
14//
15//===----------------------------------------------------------------------===//
16
17#include "AVR.h"
18#include "llvm/IR/IRBuilder.h"
19#include "llvm/IR/InstIterator.h"
20
21using namespace llvm;
22
23namespace {
24
25class AVRShiftExpand : public FunctionPass {
26public:
27 static char ID;
28
29 AVRShiftExpand() : FunctionPass(ID) {}
30
31 bool runOnFunction(Function &F) override;
32
33 StringRef getPassName() const override { return "AVR Shift Expansion"; }
34
35private:
36 void expand(BinaryOperator *BI);
37};
38
39} // end of anonymous namespace
40
41char AVRShiftExpand::ID = 0;
42
43INITIALIZE_PASS(AVRShiftExpand, "avr-shift-expand", "AVR Shift Expansion",
44 false, false)
45
46Pass *llvm::createAVRShiftExpandPass() { return new AVRShiftExpand(); }
47
48bool AVRShiftExpand::runOnFunction(Function &F) {
49 SmallVector<BinaryOperator *, 1> ShiftInsts;
50 auto &Ctx = F.getContext();
51 for (Instruction &I : instructions(F)) {
52 if (!I.isShift())
53 // Only expand shift instructions (shl, lshr, ashr).
54 continue;
55 if (I.getType() == Type::getInt8Ty(C&: Ctx) || I.getType() == Type::getInt16Ty(C&: Ctx))
56 // Only expand non-8-bit and non-16-bit shifts, since those are expanded
57 // directly during isel.
58 continue;
59 if (isa<ConstantInt>(Val: I.getOperand(i: 1)))
60 // Only expand when the shift amount is not known.
61 // Known shift amounts are (currently) better expanded inline.
62 continue;
63 ShiftInsts.push_back(Elt: cast<BinaryOperator>(Val: &I));
64 }
65
66 // The expanding itself needs to be done separately as expand() will remove
67 // these instructions. Removing instructions while iterating over a basic
68 // block is not a great idea.
69 for (auto *I : ShiftInsts) {
70 expand(BI: I);
71 }
72
73 // Return whether this function expanded any shift instructions.
74 return ShiftInsts.size() > 0;
75}
76
77void AVRShiftExpand::expand(BinaryOperator *BI) {
78 auto &Ctx = BI->getContext();
79 IRBuilder<> Builder(BI);
80 Type *InputTy = cast<Instruction>(Val: BI)->getType();
81 Type *Int8Ty = Type::getInt8Ty(C&: Ctx);
82 Value *Int8Zero = ConstantInt::get(Ty: Int8Ty, V: 0);
83
84 // Split the current basic block at the point of the existing shift
85 // instruction and insert a new basic block for the loop.
86 BasicBlock *BB = BI->getParent();
87 Function *F = BB->getParent();
88 BasicBlock *EndBB = BB->splitBasicBlock(I: BI, BBName: "shift.done");
89 BasicBlock *LoopBB = BasicBlock::Create(Context&: Ctx, Name: "shift.loop", Parent: F, InsertBefore: EndBB);
90
91 // Truncate the shift amount to i8, which is trivially lowered to a single
92 // AVR register.
93 Builder.SetInsertPoint(&BB->back());
94 Value *ShiftAmount = Builder.CreateTrunc(V: BI->getOperand(i_nocapture: 1), DestTy: Int8Ty);
95
96 // Replace the unconditional branch that splitBasicBlock created with a
97 // conditional branch.
98 Value *Cmp1 = Builder.CreateICmpEQ(LHS: ShiftAmount, RHS: Int8Zero);
99 Builder.CreateCondBr(Cond: Cmp1, True: EndBB, False: LoopBB);
100 BB->back().eraseFromParent();
101
102 // Create the loop body starting with PHI nodes.
103 Builder.SetInsertPoint(LoopBB);
104 PHINode *ShiftAmountPHI = Builder.CreatePHI(Ty: Int8Ty, NumReservedValues: 2);
105 ShiftAmountPHI->addIncoming(V: ShiftAmount, BB);
106 PHINode *ValuePHI = Builder.CreatePHI(Ty: InputTy, NumReservedValues: 2);
107 ValuePHI->addIncoming(V: BI->getOperand(i_nocapture: 0), BB);
108
109 // Subtract the shift amount by one, as we're shifting one this loop
110 // iteration.
111 Value *ShiftAmountSub =
112 Builder.CreateSub(LHS: ShiftAmountPHI, RHS: ConstantInt::get(Ty: Int8Ty, V: 1));
113 ShiftAmountPHI->addIncoming(V: ShiftAmountSub, BB: LoopBB);
114
115 // Emit the actual shift instruction. The difference is that this shift
116 // instruction has a constant shift amount, which can be emitted inline
117 // without a library call.
118 Value *ValueShifted;
119 switch (BI->getOpcode()) {
120 case Instruction::Shl:
121 ValueShifted = Builder.CreateShl(LHS: ValuePHI, RHS: ConstantInt::get(Ty: InputTy, V: 1));
122 break;
123 case Instruction::LShr:
124 ValueShifted = Builder.CreateLShr(LHS: ValuePHI, RHS: ConstantInt::get(Ty: InputTy, V: 1));
125 break;
126 case Instruction::AShr:
127 ValueShifted = Builder.CreateAShr(LHS: ValuePHI, RHS: ConstantInt::get(Ty: InputTy, V: 1));
128 break;
129 default:
130 llvm_unreachable("asked to expand an instruction that is not a shift");
131 }
132 ValuePHI->addIncoming(V: ValueShifted, BB: LoopBB);
133
134 // Branch to either the loop again (if there is more to shift) or to the
135 // basic block after the loop (if all bits are shifted).
136 Value *Cmp2 = Builder.CreateICmpEQ(LHS: ShiftAmountSub, RHS: Int8Zero);
137 Builder.CreateCondBr(Cond: Cmp2, True: EndBB, False: LoopBB);
138
139 // Collect the resulting value. This is necessary in the IR but won't produce
140 // any actual instructions.
141 Builder.SetInsertPoint(BI);
142 PHINode *Result = Builder.CreatePHI(Ty: InputTy, NumReservedValues: 2);
143 Result->addIncoming(V: BI->getOperand(i_nocapture: 0), BB);
144 Result->addIncoming(V: ValueShifted, BB: LoopBB);
145
146 // Replace the original shift instruction.
147 BI->replaceAllUsesWith(V: Result);
148 BI->eraseFromParent();
149}
150