Skip to content

Revert "Fix rsqrt inaccuracies." #88705

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 15, 2024
Merged

Revert "Fix rsqrt inaccuracies." #88705

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
88 changes: 12 additions & 76 deletions mlir/lib/Conversion/ComplexToStandard/ComplexToStandard.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,11 +27,9 @@ using namespace mlir;

namespace {

enum class AbsFn { abs, sqrt, rsqrt };

// Returns the absolute value, its square root or its reciprocal square root.
// Returns the absolute value or its square root.
Value computeAbs(Value real, Value imag, arith::FastMathFlags fmf,
ImplicitLocOpBuilder &b, AbsFn fn = AbsFn::abs) {
ImplicitLocOpBuilder &b, bool returnSqrt = false) {
Value one = b.create<arith::ConstantOp>(real.getType(),
b.getFloatAttr(real.getType(), 1.0));

Expand All @@ -45,13 +43,7 @@ Value computeAbs(Value real, Value imag, arith::FastMathFlags fmf,
Value ratioSqPlusOne = b.create<arith::AddFOp>(ratioSq, one, fmf);
Value result;

if (fn == AbsFn::rsqrt) {
ratioSqPlusOne = b.create<math::RsqrtOp>(ratioSqPlusOne, fmf);
min = b.create<math::RsqrtOp>(min, fmf);
max = b.create<math::RsqrtOp>(max, fmf);
}

if (fn == AbsFn::sqrt) {
if (returnSqrt) {
Value quarter = b.create<arith::ConstantOp>(
real.getType(), b.getFloatAttr(real.getType(), 0.25));
// sqrt(sqrt(a*b)) would avoid the pow, but will overflow more easily.
Expand Down Expand Up @@ -871,7 +863,7 @@ struct SqrtOpConversion : public OpConversionPattern<complex::SqrtOp> {

Value real = b.create<complex::ReOp>(elementType, adaptor.getComplex());
Value imag = b.create<complex::ImOp>(elementType, adaptor.getComplex());
Value absSqrt = computeAbs(real, imag, fmf, b, AbsFn::sqrt);
Value absSqrt = computeAbs(real, imag, fmf, b, /*returnSqrt=*/true);
Value argArg = b.create<math::Atan2Op>(imag, real, fmf);
Value sqrtArg = b.create<arith::MulFOp>(argArg, half, fmf);
Value cos = b.create<math::CosOp>(sqrtArg, fmf);
Expand Down Expand Up @@ -1155,74 +1147,18 @@ struct RsqrtOpConversion : public OpConversionPattern<complex::RsqrtOp> {
LogicalResult
matchAndRewrite(complex::RsqrtOp op, OpAdaptor adaptor,
ConversionPatternRewriter &rewriter) const override {
mlir::ImplicitLocOpBuilder b(op.getLoc(), rewriter);
mlir::ImplicitLocOpBuilder builder(op.getLoc(), rewriter);
auto type = cast<ComplexType>(adaptor.getComplex().getType());
auto elementType = cast<FloatType>(type.getElementType());

arith::FastMathFlags fmf = op.getFastMathFlagsAttr().getValue();

auto cst = [&](APFloat v) {
return b.create<arith::ConstantOp>(elementType,
b.getFloatAttr(elementType, v));
};
const auto &floatSemantics = elementType.getFloatSemantics();
Value zero = cst(APFloat::getZero(floatSemantics));
Value inf = cst(APFloat::getInf(floatSemantics));
Value negHalf = b.create<arith::ConstantOp>(
elementType, b.getFloatAttr(elementType, -0.5));
Value nan = cst(APFloat::getNaN(floatSemantics));

Value real = b.create<complex::ReOp>(elementType, adaptor.getComplex());
Value imag = b.create<complex::ImOp>(elementType, adaptor.getComplex());
Value absRsqrt = computeAbs(real, imag, fmf, b, AbsFn::rsqrt);
Value argArg = b.create<math::Atan2Op>(imag, real, fmf);
Value rsqrtArg = b.create<arith::MulFOp>(argArg, negHalf, fmf);
Value cos = b.create<math::CosOp>(rsqrtArg, fmf);
Value sin = b.create<math::SinOp>(rsqrtArg, fmf);

Value resultReal = b.create<arith::MulFOp>(absRsqrt, cos, fmf);
Value resultImag = b.create<arith::MulFOp>(absRsqrt, sin, fmf);

if (!arith::bitEnumContainsAll(fmf, arith::FastMathFlags::nnan |
arith::FastMathFlags::ninf)) {
Value negOne = b.create<arith::ConstantOp>(
elementType, b.getFloatAttr(elementType, -1));

Value realSignedZero = b.create<math::CopySignOp>(zero, real, fmf);
Value imagSignedZero = b.create<math::CopySignOp>(zero, imag, fmf);
Value negImagSignedZero =
b.create<arith::MulFOp>(negOne, imagSignedZero, fmf);
Value c = builder.create<arith::ConstantOp>(
elementType, builder.getFloatAttr(elementType, -0.5));
Value d = builder.create<arith::ConstantOp>(
elementType, builder.getFloatAttr(elementType, 0));

Value absReal = b.create<math::AbsFOp>(real, fmf);
Value absImag = b.create<math::AbsFOp>(imag, fmf);

Value absImagIsInf =
b.create<arith::CmpFOp>(arith::CmpFPredicate::OEQ, absImag, inf, fmf);
Value realIsNan =
b.create<arith::CmpFOp>(arith::CmpFPredicate::UNO, real, real, fmf);
Value realIsInf =
b.create<arith::CmpFOp>(arith::CmpFPredicate::OEQ, absReal, inf, fmf);
Value inIsNanInf = b.create<arith::AndIOp>(absImagIsInf, realIsNan);

Value resultIsZero = b.create<arith::OrIOp>(inIsNanInf, realIsInf);

resultReal =
b.create<arith::SelectOp>(resultIsZero, realSignedZero, resultReal);
resultImag = b.create<arith::SelectOp>(resultIsZero, negImagSignedZero,
resultImag);
}

Value isRealZero =
b.create<arith::CmpFOp>(arith::CmpFPredicate::OEQ, real, zero, fmf);
Value isImagZero =
b.create<arith::CmpFOp>(arith::CmpFPredicate::OEQ, imag, zero, fmf);
Value isZero = b.create<arith::AndIOp>(isRealZero, isImagZero);

resultReal = b.create<arith::SelectOp>(isZero, inf, resultReal);
resultImag = b.create<arith::SelectOp>(isZero, nan, resultImag);

rewriter.replaceOpWithNewOp<complex::CreateOp>(op, type, resultReal,
resultImag);
rewriter.replaceOp(op,
{powOpConversionImpl(builder, type, adaptor.getComplex(),
c, d, op.getFastmath())});
return success();
}
};
Expand Down
17 changes: 1 addition & 16 deletions mlir/test/Conversion/ComplexToStandard/convert-to-standard.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -837,21 +837,6 @@ func.func @complex_rsqrt(%arg: complex<f32>) -> complex<f32> {
return %rsqrt : complex<f32>
}

// CHECK-COUNT-5: arith.select
// CHECK-NOT: arith.select

// -----

// CHECK-LABEL: func @complex_rsqrt_nnan_ninf
// CHECK-SAME: %[[ARG:.*]]: complex<f32>
func.func @complex_rsqrt_nnan_ninf(%arg: complex<f32>) -> complex<f32> {
%sqrt = complex.rsqrt %arg fastmath<nnan,ninf> : complex<f32>
return %sqrt : complex<f32>
}

// CHECK-COUNT-3: arith.select
// CHECK-NOT: arith.select

// -----

// CHECK-LABEL: func.func @complex_angle
Expand Down Expand Up @@ -2118,4 +2103,4 @@ func.func @complex_tanh_with_fmf(%arg: complex<f32>) -> complex<f32> {
// CHECK: %[[NUM:.*]] = complex.create %[[TANH_A]], %[[TAN_B]] : complex<f32>
// CHECK: %[[ONE:.*]] = arith.constant 1.000000e+00 : f32
// CHECK: %[[MUL:.*]] = arith.mulf %[[TANH_A]], %[[TAN_B]] fastmath<nnan,contract> : f32
// CHECK: %[[DENOM:.*]] = complex.create %[[ONE]], %[[MUL]] : complex<f32>
// CHECK: %[[DENOM:.*]] = complex.create %[[ONE]], %[[MUL]] : complex<f32>