[SelectionDAG] Handle inverted conditions when splitting into multiple branches.

Summary:
When conditional branches with complex conditions are split into
multiple branches in SelectionDAGBuilder::FindMergedConditions, also
handle inverted conditions.  These may sometimes appear without having
been optimized by InstCombine when CodeGenPrepare decides to sink and
duplicate cmp instructions, causing them to have only one use.  This
problem can be increased by e.g. GVNHoist hiding more cmps from
InstCombine by combining equivalent cmps from different blocks.

For example codegen X & !(Y | Z) as:
    jmp_if_X TmpBB
    jmp FBB
  TmpBB:
    jmp_if_notY Tmp2BB
    jmp FBB
  Tmp2BB:
    jmp_if_notZ TBB
    jmp FBB

Reviewers: bogner, MatzeB, qcolombet

Subscribers: llvm-commits, hiraditya, mcrosier, sebpop

Differential Revision: https://reviews.llvm.org/D28380

llvm-svn: 292944
This commit is contained in:
Geoff Berry 2017-01-24 16:36:07 +00:00
parent f56e3cdd5d
commit 92a286ae5a
5 changed files with 88 additions and 20 deletions

View File

@ -1584,7 +1584,8 @@ SelectionDAGBuilder::EmitBranchForMergedCondition(const Value *Cond,
MachineBasicBlock *CurBB,
MachineBasicBlock *SwitchBB,
BranchProbability TProb,
BranchProbability FProb) {
BranchProbability FProb,
bool InvertCond) {
const BasicBlock *BB = CurBB->getBasicBlock();
// If the leaf of the tree is a comparison, merge the condition into
@ -1598,10 +1599,14 @@ SelectionDAGBuilder::EmitBranchForMergedCondition(const Value *Cond,
isExportableFromCurrentBlock(BOp->getOperand(1), BB))) {
ISD::CondCode Condition;
if (const ICmpInst *IC = dyn_cast<ICmpInst>(Cond)) {
Condition = getICmpCondCode(IC->getPredicate());
ICmpInst::Predicate Pred =
InvertCond ? IC->getInversePredicate() : IC->getPredicate();
Condition = getICmpCondCode(Pred);
} else {
const FCmpInst *FC = cast<FCmpInst>(Cond);
Condition = getFCmpCondCode(FC->getPredicate());
FCmpInst::Predicate Pred =
InvertCond ? FC->getInversePredicate() : FC->getPredicate();
Condition = getFCmpCondCode(Pred);
if (TM.Options.NoNaNsFPMath)
Condition = getFCmpCodeWithoutNaN(Condition);
}
@ -1614,7 +1619,8 @@ SelectionDAGBuilder::EmitBranchForMergedCondition(const Value *Cond,
}
// Create a CaseBlock record representing this branch.
CaseBlock CB(ISD::SETEQ, Cond, ConstantInt::getTrue(*DAG.getContext()),
ISD::CondCode Opc = InvertCond ? ISD::SETNE : ISD::SETEQ;
CaseBlock CB(Opc, Cond, ConstantInt::getTrue(*DAG.getContext()),
nullptr, TBB, FBB, CurBB, TProb, FProb);
SwitchCases.push_back(CB);
}
@ -1627,16 +1633,42 @@ void SelectionDAGBuilder::FindMergedConditions(const Value *Cond,
MachineBasicBlock *SwitchBB,
Instruction::BinaryOps Opc,
BranchProbability TProb,
BranchProbability FProb) {
// If this node is not part of the or/and tree, emit it as a branch.
BranchProbability FProb,
bool InvertCond) {
// Skip over not part of the tree and remember to invert op and operands at
// next level.
if (BinaryOperator::isNot(Cond) && Cond->hasOneUse()) {
Cond = cast<Instruction>(Cond)->getOperand(0);
FindMergedConditions(Cond, TBB, FBB, CurBB, SwitchBB, Opc, TProb, FProb,
!InvertCond);
return;
}
const Instruction *BOp = dyn_cast<Instruction>(Cond);
// Compute the effective opcode for Cond, taking into account whether it needs
// to be inverted, e.g.
// and (not (or A, B)), C
// gets lowered as
// and (and (not A, not B), C)
unsigned BOpc = 0;
if (BOp) {
BOpc = BOp->getOpcode();
if (InvertCond) {
if (BOpc == Instruction::And)
BOpc = Instruction::Or;
else if (BOpc == Instruction::Or)
BOpc = Instruction::And;
}
}
// If this node is not part of the or/and tree, emit it as a branch.
if (!BOp || !(isa<BinaryOperator>(BOp) || isa<CmpInst>(BOp)) ||
(unsigned)BOp->getOpcode() != Opc || !BOp->hasOneUse() ||
BOpc != Opc || !BOp->hasOneUse() ||
BOp->getParent() != CurBB->getBasicBlock() ||
!InBlock(BOp->getOperand(0), CurBB->getBasicBlock()) ||
!InBlock(BOp->getOperand(1), CurBB->getBasicBlock())) {
EmitBranchForMergedCondition(Cond, TBB, FBB, CurBB, SwitchBB,
TProb, FProb);
TProb, FProb, InvertCond);
return;
}
@ -1671,14 +1703,14 @@ void SelectionDAGBuilder::FindMergedConditions(const Value *Cond,
auto NewFalseProb = TProb / 2 + FProb;
// Emit the LHS condition.
FindMergedConditions(BOp->getOperand(0), TBB, TmpBB, CurBB, SwitchBB, Opc,
NewTrueProb, NewFalseProb);
NewTrueProb, NewFalseProb, InvertCond);
// Normalize A/2 and B to get A/(1+B) and 2B/(1+B).
SmallVector<BranchProbability, 2> Probs{TProb / 2, FProb};
BranchProbability::normalizeProbabilities(Probs.begin(), Probs.end());
// Emit the RHS condition into TmpBB.
FindMergedConditions(BOp->getOperand(1), TBB, FBB, TmpBB, SwitchBB, Opc,
Probs[0], Probs[1]);
Probs[0], Probs[1], InvertCond);
} else {
assert(Opc == Instruction::And && "Unknown merge op!");
// Codegen X & Y as:
@ -1704,14 +1736,14 @@ void SelectionDAGBuilder::FindMergedConditions(const Value *Cond,
auto NewFalseProb = FProb / 2;
// Emit the LHS condition.
FindMergedConditions(BOp->getOperand(0), TmpBB, FBB, CurBB, SwitchBB, Opc,
NewTrueProb, NewFalseProb);
NewTrueProb, NewFalseProb, InvertCond);
// Normalize A and B/2 to get 2A/(1+A) and B/(1+A).
SmallVector<BranchProbability, 2> Probs{TProb, FProb / 2};
BranchProbability::normalizeProbabilities(Probs.begin(), Probs.end());
// Emit the RHS condition into TmpBB.
FindMergedConditions(BOp->getOperand(1), TBB, FBB, TmpBB, SwitchBB, Opc,
Probs[0], Probs[1]);
Probs[0], Probs[1], InvertCond);
}
}
@ -1795,7 +1827,8 @@ void SelectionDAGBuilder::visitBr(const BranchInst &I) {
FindMergedConditions(BOp, Succ0MBB, Succ1MBB, BrMBB, BrMBB,
Opcode,
getEdgeProbability(BrMBB, Succ0MBB),
getEdgeProbability(BrMBB, Succ1MBB));
getEdgeProbability(BrMBB, Succ1MBB),
/*InvertCond=*/false);
// If the compares in later blocks need to use values not currently
// exported from this block, export them now. This block should always
// be the first entry.

View File

@ -688,12 +688,13 @@ public:
MachineBasicBlock *FBB, MachineBasicBlock *CurBB,
MachineBasicBlock *SwitchBB,
Instruction::BinaryOps Opc, BranchProbability TW,
BranchProbability FW);
BranchProbability FW, bool InvertCond);
void EmitBranchForMergedCondition(const Value *Cond, MachineBasicBlock *TBB,
MachineBasicBlock *FBB,
MachineBasicBlock *CurBB,
MachineBasicBlock *SwitchBB,
BranchProbability TW, BranchProbability FW);
BranchProbability TW, BranchProbability FW,
bool InvertCond);
bool ShouldEmitAsBranches(const std::vector<CaseBlock> &Cases);
bool isExportableFromCurrentBlock(const Value *V, const BasicBlock *FromBB);
void CopyToExportRegsIfNeeded(const Value *V);

View File

@ -0,0 +1,32 @@
; RUN: llc -mtriple=aarch64 -verify-machineinstrs < %s | FileCheck %s
declare void @foo()
; Check that the inverted or doesn't inhibit the splitting of the
; complex conditional into three branch instructions.
; CHECK-LABEL: test_and_not
; CHECK: cbz w0, [[L:\.LBB[0-9_]+]]
; CHECK: cmp w1, #2
; CHECK: b.lo [[L]]
; CHECK: cmp w2, #2
; CHECK: b.hi [[L]]
define void @test_and_not(i32 %a, i32 %b, i32 %c) {
bb1:
%cmp1 = icmp ult i32 %a, 1
%cmp2 = icmp ult i32 %b, 2
%cmp3 = icmp ult i32 %c, 3
%or = or i1 %cmp1, %cmp2
%not.or = xor i1 %or, -1
%and = and i1 %not.or, %cmp3
br i1 %and, label %bb2, label %bb3
bb2:
ret void
bb3:
call void @foo()
ret void
}

View File

@ -70,7 +70,7 @@ define void @test3(i64 %a, i64 %b, i1 %p) nounwind {
@g_100 = external global i8 ; <i8*> [#uses=2]
@_2E_str = external constant [15 x i8], align 1 ; <[15 x i8]*> [#uses=1]
define i32 @test4() nounwind {
define i1 @test4() nounwind {
entry:
%0 = load i8, i8* @g_3, align 1 ; <i8> [#uses=2]
%1 = sext i8 %0 to i32 ; <i32> [#uses=1]
@ -107,10 +107,11 @@ bb.i.i: ; preds = %func_4.exit.i
func_1.exit: ; preds = %bb.i.i, %func_4.exit.i
%g_96.tmp.0.i = phi i8 [ %g_96.promoted.i, %bb.i.i ], [ %.mux.i, %func_4.exit.i ] ; <i8> [#uses=2]
%ret = phi i1 [ 0, %bb.i.i ], [ %.not.i, %func_4.exit.i ]
store i8 %g_96.tmp.0.i, i8* @g_96
%6 = zext i8 %g_96.tmp.0.i to i32 ; <i32> [#uses=1]
%7 = tail call i32 (i8*, ...) @printf(i8* noalias getelementptr ([15 x i8], [15 x i8]* @_2E_str, i64 0, i64 0), i32 %6) nounwind ; <i32> [#uses=0]
ret i32 0
ret i1 %ret
}
declare i32 @printf(i8* nocapture, ...) nounwind

View File

@ -12,10 +12,11 @@ declare i32 @printf(i8* nocapture readonly, ...)
;CHECK: cmpl
;CHECK: setg
;CHECK: setl
;CHECK: cmpl
;CHECK: setg
;CHECK: andb
;CHECK: setl
;CHECK: orb
;CHECK: je
@.str = private unnamed_addr constant [4 x i8] c"%d\0A\00", align 1
; Function Attrs: optsize ssp uwtable