Teach ComputeNumSignBits about signed reminder.

This optimization a continuation of r231140 that reasoned about signed div.

llvm-svn: 231433
This commit is contained in:
Nadav Rotem 2015-03-06 00:23:58 +00:00
parent 25d5abdb3a
commit c99a38796c
2 changed files with 48 additions and 1 deletions

View File

@ -1723,7 +1723,7 @@ unsigned ComputeNumSignBits(Value *V, const DataLayout *TD,
Tmp = TyBits - U->getOperand(0)->getType()->getScalarSizeInBits();
return ComputeNumSignBits(U->getOperand(0), TD, Depth+1, Q) + Tmp;
case Instruction::SDiv:
case Instruction::SDiv: {
const APInt *Denominator;
// sdiv X, C -> adds log(C) sign bits.
if (match(U->getOperand(1), m_APInt(Denominator))) {
@ -1739,6 +1739,32 @@ unsigned ComputeNumSignBits(Value *V, const DataLayout *TD,
return std::min(TyBits, NumBits + Denominator->logBase2());
}
break;
}
case Instruction::SRem: {
const APInt *Denominator;
// srem X, C -> we know that the result is within 0..C-1 when C is a
// positive constant and the sign bits are at most TypeBits - log2(C).
if (match(U->getOperand(1), m_APInt(Denominator))) {
// Ignore non-positive denominator.
if (!Denominator->isStrictlyPositive())
break;
// Calculate the incoming numerator bits. SRem by a positive constant
// can't lower the number of sign bits.
unsigned NumrBits = ComputeNumSignBits(U->getOperand(0), TD, Depth+1, Q);
// Calculate the leading sign bit constraints by examining the
// denominator. The remainder is in the range 0..C-1, which is
// calculated by the log2(denominator). The sign bits are the bit-width
// minus this value. The result of this subtraction has to be positive.
unsigned ResBits = TyBits - Denominator->logBase2();
return std::max(NumrBits, ResBits);
}
break;
}
case Instruction::AShr: {
Tmp = ComputeNumSignBits(U->getOperand(0), TD, Depth+1, Q);

View File

@ -386,3 +386,24 @@ entry:
ret i1 %obit
}
; CHECK-LABEL: @overflow_mod_mul(
; CHECK: ret i1 false
define i1 @overflow_mod_mul(i32 %v1, i32 %v2) nounwind {
entry:
%rem = srem i32 %v1, 1000
%t = call %ov.result.32 @llvm.smul.with.overflow.i32(i32 %rem, i32 %rem)
%obit = extractvalue %ov.result.32 %t, 1
ret i1 %obit
}
; CHECK-LABEL: @overflow_mod_overflow_mul(
; CHECK-NOT: ret i1 false
define i1 @overflow_mod_overflow_mul(i32 %v1, i32 %v2) nounwind {
entry:
%rem = srem i32 %v1, 65537
; This may overflow because the result of the mul operands may be greater than 16bits
; and the result greater than 32.
%t = call %ov.result.32 @llvm.smul.with.overflow.i32(i32 %rem, i32 %rem)
%obit = extractvalue %ov.result.32 %t, 1
ret i1 %obit
}