Skip to content

[InstCombine] Funnel shift with negative amount folds to funnel shift in opposite direction (#138334) #138763

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.

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
May 7, 2025
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
12 changes: 12 additions & 0 deletions llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -2306,6 +2306,18 @@ Instruction *InstCombinerImpl::visitCallInst(CallInst &CI) {
return BitOp;
}

// fshl(X, X, Neg(Y)) --> fshr(X, X, Y)
// fshr(X, X, Neg(Y)) --> fshl(X, X, Y)
// if BitWidth is a power-of-2
Value *Y;
if (Op0 == Op1 && isPowerOf2_32(BitWidth) &&
match(II->getArgOperand(2), m_Neg(m_Value(Y)))) {
Module *Mod = II->getModule();
Function *OppositeShift = Intrinsic::getOrInsertDeclaration(
Mod, IID == Intrinsic::fshl ? Intrinsic::fshr : Intrinsic::fshl, Ty);
return CallInst::Create(OppositeShift, {Op0, Op1, Y});
}

// fshl(X, 0, Y) --> shl(X, and(Y, BitWidth - 1)) if bitwidth is a
// power-of-2
if (IID == Intrinsic::fshl && isPowerOf2_32(BitWidth) &&
Expand Down
69 changes: 69 additions & 0 deletions llvm/test/Transforms/InstCombine/fsh.ll
Copy link
Contributor

Choose a reason for hiding this comment

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

Please also add a negative test where the funnel shift is not a rotate (operands not the same).

Original file line number Diff line number Diff line change
Expand Up @@ -1084,3 +1084,72 @@ define i8 @fshl_range_trunc(i1 %x) {
%tr = trunc nsw i32 %fshl to i8
ret i8 %tr
}

;; Issue #138334 negative rotate amounts can be folded into the opposite direction
define i32 @fshl_neg_amount(i32 %x, i32 %y) {
; CHECK-LABEL: @fshl_neg_amount(
; CHECK-NEXT: [[R:%.*]] = call i32 @llvm.fshr.i32(i32 [[X:%.*]], i32 [[X]], i32 [[Y:%.*]])
; CHECK-NEXT: ret i32 [[R]]
;
%n = sub i32 0, %y
%r = call i32 @llvm.fshl.i32(i32 %x, i32 %x, i32 %n)
ret i32 %r
}

define i32 @fshr_neg_amount(i32 %x, i32 %y) {
; CHECK-LABEL: @fshr_neg_amount(
; CHECK-NEXT: [[R:%.*]] = call i32 @llvm.fshl.i32(i32 [[X:%.*]], i32 [[X]], i32 [[Y:%.*]])
; CHECK-NEXT: ret i32 [[R]]
;
%n = sub i32 0, %y
%r = call i32 @llvm.fshr.i32(i32 %x, i32 %x, i32 %n)
ret i32 %r
}

;; negative test, funnel shift is not a rotate

define i32 @fshl_neg_amount_non_rotate(i32 %x, i32 %y, i32 %z) {
; CHECK-LABEL: @fshl_neg_amount_non_rotate(
; CHECK-NEXT: [[N:%.*]] = sub i32 0, [[Y:%.*]]
; CHECK-NEXT: [[R:%.*]] = call i32 @llvm.fshl.i32(i32 [[X:%.*]], i32 [[Z:%.*]], i32 [[N]])
; CHECK-NEXT: ret i32 [[R]]
;
%n = sub i32 0, %y
%r = call i32 @llvm.fshl.i32(i32 %x, i32 %z, i32 %n)
ret i32 %r
}

define i32 @fshr_neg_amount_non_rotate(i32 %x, i32 %y, i32 %z) {
; CHECK-LABEL: @fshr_neg_amount_non_rotate(
; CHECK-NEXT: [[N:%.*]] = sub i32 0, [[Y:%.*]]
; CHECK-NEXT: [[R:%.*]] = call i32 @llvm.fshr.i32(i32 [[X:%.*]], i32 [[Z:%.*]], i32 [[N]])
; CHECK-NEXT: ret i32 [[R]]
;
%n = sub i32 0, %y
%r = call i32 @llvm.fshr.i32(i32 %x, i32 %z, i32 %n)
ret i32 %r
}

;; negative test, bitwidth is not a power of two

define i31 @fshl_neg_amount_non_power_two(i31 %x, i31 %y) {
; CHECK-LABEL: @fshl_neg_amount_non_power_two(
; CHECK-NEXT: [[N:%.*]] = sub i31 0, [[Y:%.*]]
; CHECK-NEXT: [[R:%.*]] = call i31 @llvm.fshl.i31(i31 [[X:%.*]], i31 [[X]], i31 [[N]])
; CHECK-NEXT: ret i31 [[R]]
;
%n = sub i31 0, %y
%r = call i31 @llvm.fshl.i31(i31 %x, i31 %x, i31 %n)
ret i31 %r
}

define i31 @fshr_neg_amount_non_power_two(i31 %x, i31 %y) {
; CHECK-LABEL: @fshr_neg_amount_non_power_two(
; CHECK-NEXT: [[N:%.*]] = sub i31 0, [[Y:%.*]]
; CHECK-NEXT: [[R:%.*]] = call i31 @llvm.fshr.i31(i31 [[X:%.*]], i31 [[X]], i31 [[N]])
; CHECK-NEXT: ret i31 [[R]]
;
%n = sub i31 0, %y
%r = call i31 @llvm.fshr.i31(i31 %x, i31 %x, i31 %n)
ret i31 %r
}
Loading