Skip to content

[InstCombine] Skip division sinking with allowReciprocal flag #167633

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

Closed
wants to merge 1 commit into from
+8 −1
Closed

[InstCombine] Skip division sinking with allowReciprocal flag #167633

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
9 changes: 8 additions & 1 deletion llvm/lib/Transforms/InstCombine/InstCombineMulDivRem.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -795,7 +795,8 @@ Instruction *InstCombinerImpl::foldFMulReassoc(BinaryOperator &I) {
return BinaryOperator::CreateFDivFMF(CC1, X, FMF);
}
if (match(Op0, m_FDiv(m_Value(X), m_Constant(C1)))) {
// FIXME: This seems like it should also be checking for arcp
// Constant folding is not affected by allowReciprocal,
// since constants are calculated at compile time.
Comment on lines +798 to +799
Copy link
Contributor

@arsenm arsenm Nov 12, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This isn't exactly how semantics works

Copy link
Contributor

@jcranmer-intel jcranmer-intel Nov 13, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Also, despite the original comment, this really needs to be matching against the reassoc flag, not the arcp flag (since arcp explicitly doesn't allow reassociation of the division operation, only the inversion of the division).

// (X / C1) * C --> X * (C / C1)
Constant *CDivC1 =
ConstantFoldBinaryOpOperands(Instruction::FDiv, C, C1, DL);
Expand Down Expand Up @@ -837,6 +838,12 @@ Instruction *InstCombinerImpl::foldFMulReassoc(BinaryOperator &I) {
m_Value(Z)))) {
BinaryOperator *DivOp = cast<BinaryOperator>(((Z == Op0) ? Op1 : Op0));
FastMathFlags FMF = I.getFastMathFlags() & DivOp->getFastMathFlags();
// Preserve (X / Y) * Z when reciprocal math is enabled to allow backend
// rcp instruction emission. Sinking to (X * Z) / Y would prevent this
// optimization, alter computation order and potentially affecting
// precision.
if (FMF.allowReciprocal())
return nullptr;
if (FMF.allowReassoc()) {
// Sink division: (X / Y) * Z --> (X * Z) / Y
auto *NewFMul = Builder.CreateFMulFMF(X, Z, FMF);
Expand Down