RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2025-01-31 12:41:49 +01:00
Code Issues Projects Releases Wiki Activity

[BDCE] Remove dead uses of arguments

Browse Source 
In addition to finding dead uses of instructions, also find dead uses
of function arguments, and replace them with zero as well.

I'm changing the way the known bits are computed here to remove the
coupling between the transfer function and the algorithm. It previously
relied on the first op being visited first and computing known bits --
unless the first op is not an instruction, in which case they're computed
on the second op. I could have adjusted this to check for "instruction
or argument", but I think it's better to avoid the repeated calculation
with an explicit flag.

Differential Revision: https://reviews.llvm.org/D56247

llvm-svn: 350435
This commit is contained in:
Nikita Popov 2019-01-04 21:21:43 +00:00
parent 30be00653a
commit 61d02a0896
4 changed files with 51 additions and 47 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
include/llvm/Analysis/DemandedBits.h
View File

@ -65,9 +65,9 @@ public:
private:
void performAnalysis();
void determineLiveOperandBits(const Instruction *UserI,
const Instruction *I, unsigned OperandNo,
const Value *Val, unsigned OperandNo,
const APInt &AOut, APInt &AB,
KnownBits &Known, KnownBits &Known2);
KnownBits &Known, KnownBits &Known2, bool &KnownBitsComputed);
Function &F;
AssumptionCache ∾

86
lib/Analysis/DemandedBits.cpp
View File

@ -85,8 +85,9 @@ static bool isAlwaysLive(Instruction *I) {
}
void DemandedBits::determineLiveOperandBits(
const Instruction *UserI, const Instruction *I, unsigned OperandNo,
const APInt &AOut, APInt &AB, KnownBits &Known, KnownBits &Known2) {
const Instruction *UserI, const Value *Val, unsigned OperandNo,
const APInt &AOut, APInt &AB, KnownBits &Known, KnownBits &Known2,
bool &KnownBitsComputed) {
unsigned BitWidth = AB.getBitWidth();
// We're called once per operand, but for some instructions, we need to
@ -97,7 +98,11 @@ void DemandedBits::determineLiveOperandBits(
// provided here.
auto ComputeKnownBits =
[&](unsigned BitWidth, const Value *V1, const Value *V2) {
const DataLayout &DL = I->getModule()->getDataLayout();
if (KnownBitsComputed)
return;
KnownBitsComputed = true;
const DataLayout &DL = UserI->getModule()->getDataLayout();
Known = KnownBits(BitWidth);
computeKnownBits(V1, Known, DL, 0, &AC, UserI, &DT);
@ -129,7 +134,7 @@ void DemandedBits::determineLiveOperandBits(
// We need some output bits, so we need all bits of the
// input to the left of, and including, the leftmost bit
// known to be one.
ComputeKnownBits(BitWidth, I, nullptr);
ComputeKnownBits(BitWidth, Val, nullptr);
AB = APInt::getHighBitsSet(BitWidth,
std::min(BitWidth, Known.countMaxLeadingZeros()+1));
}
@ -139,7 +144,7 @@ void DemandedBits::determineLiveOperandBits(
// We need some output bits, so we need all bits of the
// input to the right of, and including, the rightmost bit
// known to be one.
ComputeKnownBits(BitWidth, I, nullptr);
ComputeKnownBits(BitWidth, Val, nullptr);
AB = APInt::getLowBitsSet(BitWidth,
std::min(BitWidth, Known.countMaxTrailingZeros()+1));
}
@ -234,14 +239,11 @@ void DemandedBits::determineLiveOperandBits(
// other operand are dead (unless they're both zero, in which
// case they can't both be dead, so just mark the LHS bits as
// dead).
if (OperandNo == 0) {
ComputeKnownBits(BitWidth, I, UserI->getOperand(1));
ComputeKnownBits(BitWidth, UserI->getOperand(0), UserI->getOperand(1));
if (OperandNo == 0)
AB &= ~Known2.Zero;
} else {
if (!isa<Instruction>(UserI->getOperand(0)))
ComputeKnownBits(BitWidth, UserI->getOperand(0), I);
else
AB &= ~(Known.Zero & ~Known2.Zero);
}
break;
case Instruction::Or:
AB = AOut;
@ -250,14 +252,11 @@ void DemandedBits::determineLiveOperandBits(
// other operand are dead (unless they're both one, in which
// case they can't both be dead, so just mark the LHS bits as
// dead).
if (OperandNo == 0) {
ComputeKnownBits(BitWidth, I, UserI->getOperand(1));
ComputeKnownBits(BitWidth, UserI->getOperand(0), UserI->getOperand(1));
if (OperandNo == 0)
AB &= ~Known2.One;
} else {
if (!isa<Instruction>(UserI->getOperand(0)))
ComputeKnownBits(BitWidth, UserI->getOperand(0), I);
else
AB &= ~(Known.One & ~Known2.One);
}
break;
case Instruction::Xor:
case Instruction::PHI:
@ -368,33 +367,40 @@ void DemandedBits::performAnalysis() {
Visited.insert(UserI);
KnownBits Known, Known2;
bool KnownBitsComputed = false;
// Compute the set of alive bits for each operand. These are anded into the
// existing set, if any, and if that changes the set of alive bits, the
// operand is added to the work-list.
for (Use &OI : UserI->operands()) {
if (Instruction *I = dyn_cast<Instruction>(OI)) {
Type *T = I->getType();
if (T->isIntOrIntVectorTy()) {
unsigned BitWidth = T->getScalarSizeInBits();
APInt AB = APInt::getAllOnesValue(BitWidth);
if (UserI->getType()->isIntOrIntVectorTy() && !AOut &&
!isAlwaysLive(UserI)) {
// If all bits of the output are dead, then all bits of the input
// are also dead.
AB = APInt(BitWidth, 0);
} else {
// Bits of each operand that are used to compute alive bits of the
// output are alive, all others are dead.
determineLiveOperandBits(UserI, I, OI.getOperandNo(), AOut, AB,
Known, Known2);
// We also want to detect dead uses of arguments, but will only store
// demanded bits for instructions.
Instruction *I = dyn_cast<Instruction>(OI);
if (!I && !isa<Argument>(OI))
continue;
// Keep track of uses which have no demanded bits.
if (AB.isNullValue())
DeadUses.insert(&OI);
else
DeadUses.erase(&OI);
}
Type *T = OI->getType();
if (T->isIntOrIntVectorTy()) {
unsigned BitWidth = T->getScalarSizeInBits();
APInt AB = APInt::getAllOnesValue(BitWidth);
if (UserI->getType()->isIntOrIntVectorTy() && !AOut &&
!isAlwaysLive(UserI)) {
// If all bits of the output are dead, then all bits of the input
// are also dead.
AB = APInt(BitWidth, 0);
} else {
// Bits of each operand that are used to compute alive bits of the
// output are alive, all others are dead.
determineLiveOperandBits(UserI, OI, OI.getOperandNo(), AOut, AB,
Known, Known2, KnownBitsComputed);
// Keep track of uses which have no demanded bits.
if (AB.isNullValue())
DeadUses.insert(&OI);
else
DeadUses.erase(&OI);
}
if (I) {
// If we've added to the set of alive bits (or the operand has not
// been previously visited), then re-queue the operand to be visited
// again.
@ -408,9 +414,9 @@ void DemandedBits::performAnalysis() {
AliveBits[I] = std::move(ABNew);
Worklist.push_back(I);
}
} else if (!Visited.count(I)) {
Worklist.push_back(I);
}
} else if (I && !Visited.count(I)) {
Worklist.push_back(I);
}
}
}

3
lib/Transforms/Scalar/BDCE.cpp
View File

@ -114,8 +114,7 @@ static bool bitTrackingDCE(Function &F, DemandedBits &DB) {
if (!U->getType()->isIntOrIntVectorTy())
continue;
// TODO: We could also find dead non-instruction uses, e.g. arguments.
if (!isa<Instruction>(U))
if (!isa<Instruction>(U) && !isa<Argument>(U))
continue;
if (!DB.isUseDead(&U))

5
test/Transforms/BDCE/dead-uses.ll
View File

@ -45,7 +45,7 @@ define <2 x i32> @pr39771_fshr_multi_use_instr_vec(<2 x i32> %a) {
; First fshr operand is dead, but it comes from an argument, not instruction.
define i32 @pr39771_fshr_multi_use_arg(i32 %a) {
; CHECK-LABEL: @pr39771_fshr_multi_use_arg(
; CHECK-NEXT: [[B:%.*]] = tail call i32 @llvm.fshr.i32(i32 [[A:%.*]], i32 [[A]], i32 1)
; CHECK-NEXT: [[B:%.*]] = tail call i32 @llvm.fshr.i32(i32 0, i32 [[A:%.*]], i32 1)
; CHECK-NEXT: [[C:%.*]] = lshr i32 [[B]], 23
; CHECK-NEXT: [[D:%.*]] = xor i32 [[C]], [[B]]
; CHECK-NEXT: [[E:%.*]] = and i32 [[D]], 31
@ -58,11 +58,10 @@ define i32 @pr39771_fshr_multi_use_arg(i32 %a) {
ret i32 %e
}
; Second or operand is dead, but BDCE does not realize this.
define i32 @pr39771_expanded_fshr_multi_use(i32 %a) {
; CHECK-LABEL: @pr39771_expanded_fshr_multi_use(
; CHECK-NEXT: [[TMP:%.*]] = lshr i32 [[A:%.*]], 1
; CHECK-NEXT: [[TMP2:%.*]] = shl i32 [[A]], 31
; CHECK-NEXT: [[TMP2:%.*]] = shl i32 0, 31
; CHECK-NEXT: [[B:%.*]] = or i32 [[TMP]], [[TMP2]]
; CHECK-NEXT: [[C:%.*]] = lshr i32 [[B]], 23
; CHECK-NEXT: [[D:%.*]] = xor i32 [[C]], [[B]]