Skip to content

[TTI] Add cost model support for [u|s]cmp #106824

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 2 commits into from
Aug 31, 2024
Merged

[TTI] Add cost model support for [u|s]cmp #106824

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7aea58f
[TTI] Add cost model support for [u|s]cmp
dtcxzyw Aug 31, 2024
0d04e20
[AArch64][Analysis] Add cost model tests for AArch64
dtcxzyw Aug 31, 2024
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
33 changes: 33 additions & 0 deletions llvm/include/llvm/CodeGen/BasicTTIImpl.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2196,6 +2196,12 @@ class BasicTTIImplBase : public TargetTransformInfoImplCRTPBase<T> {
case Intrinsic::bitreverse:
ISD = ISD::BITREVERSE;
break;
case Intrinsic::ucmp:
ISD = ISD::UCMP;
break;
case Intrinsic::scmp:
ISD = ISD::SCMP;
break;
}

auto *ST = dyn_cast<StructType>(RetTy);
Expand Down Expand Up @@ -2433,6 +2439,33 @@ class BasicTTIImplBase : public TargetTransformInfoImplCRTPBase<T> {
}
return Cost;
}
case Intrinsic::ucmp:
case Intrinsic::scmp: {
Type *CmpTy = Tys[0];
Type *CondTy = RetTy->getWithNewBitWidth(1);
InstructionCost Cost =
thisT()->getCmpSelInstrCost(BinaryOperator::ICmp, CmpTy, CondTy,
CmpIntrinsic::getGTPredicate(IID),
CostKind) +
thisT()->getCmpSelInstrCost(BinaryOperator::ICmp, CmpTy, CondTy,
CmpIntrinsic::getLTPredicate(IID),
CostKind);

if (TLI->shouldExpandCmpUsingSelects()) {
// x < y ? -1 : (x > y ? 1 : 0)
Cost += 2 * thisT()->getCmpSelInstrCost(
BinaryOperator::Select, RetTy, CondTy,
ICmpInst::BAD_ICMP_PREDICATE, CostKind);
} else {
// zext(x > y) - zext(x < y)
Cost +=
2 * thisT()->getCastInstrCost(CastInst::ZExt, RetTy, CondTy,
TTI::CastContextHint::None, CostKind);
Cost += thisT()->getArithmeticInstrCost(BinaryOperator::Sub, RetTy,
CostKind);
}
return Cost;
}
default:
break;
}
Expand Down
52 changes: 52 additions & 0 deletions llvm/test/Analysis/CostModel/AArch64/cmp.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -121,3 +121,55 @@ define void @andcmp() {
%c64sle = icmp sle i64 %a64, 0
ret void
}

define void @uscmp() {
; CHECK-THROUGHPUT-LABEL: 'uscmp'
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %u8 = call i8 @llvm.ucmp.i8.i8(i8 undef, i8 undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %u16 = call i16 @llvm.ucmp.i16.i16(i16 undef, i16 undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %u32 = call i32 @llvm.ucmp.i32.i32(i32 undef, i32 undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %u64 = call i64 @llvm.ucmp.i64.i64(i64 undef, i64 undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 98 for instruction: %uv16i8 = call <16 x i8> @llvm.ucmp.v16i8.v16i8(<16 x i8> undef, <16 x i8> undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %uv8i16 = call <8 x i16> @llvm.ucmp.v8i16.v8i16(<8 x i16> undef, <8 x i16> undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %uv4i32 = call <4 x i32> @llvm.ucmp.v4i32.v4i32(<4 x i32> undef, <4 x i32> undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %s8 = call i8 @llvm.scmp.i8.i8(i8 undef, i8 undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %s16 = call i16 @llvm.scmp.i16.i16(i16 undef, i16 undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %s32 = call i32 @llvm.scmp.i32.i32(i32 undef, i32 undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %s64 = call i64 @llvm.scmp.i64.i64(i64 undef, i64 undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 98 for instruction: %sv16i8 = call <16 x i8> @llvm.scmp.v16i8.v16i8(<16 x i8> undef, <16 x i8> undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 50 for instruction: %sv8i16 = call <8 x i16> @llvm.scmp.v8i16.v8i16(<8 x i16> undef, <8 x i16> undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 26 for instruction: %sv4i32 = call <4 x i32> @llvm.scmp.v4i32.v4i32(<4 x i32> undef, <4 x i32> undef)
; CHECK-THROUGHPUT-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; CHECK-SIZE-LABEL: 'uscmp'
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %u8 = call i8 @llvm.ucmp.i8.i8(i8 undef, i8 undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %u16 = call i16 @llvm.ucmp.i16.i16(i16 undef, i16 undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %u32 = call i32 @llvm.ucmp.i32.i32(i32 undef, i32 undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %u64 = call i64 @llvm.ucmp.i64.i64(i64 undef, i64 undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %uv16i8 = call <16 x i8> @llvm.ucmp.v16i8.v16i8(<16 x i8> undef, <16 x i8> undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %uv8i16 = call <8 x i16> @llvm.ucmp.v8i16.v8i16(<8 x i16> undef, <8 x i16> undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %uv4i32 = call <4 x i32> @llvm.ucmp.v4i32.v4i32(<4 x i32> undef, <4 x i32> undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %s8 = call i8 @llvm.scmp.i8.i8(i8 undef, i8 undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %s16 = call i16 @llvm.scmp.i16.i16(i16 undef, i16 undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %s32 = call i32 @llvm.scmp.i32.i32(i32 undef, i32 undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %s64 = call i64 @llvm.scmp.i64.i64(i64 undef, i64 undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %sv16i8 = call <16 x i8> @llvm.scmp.v16i8.v16i8(<16 x i8> undef, <16 x i8> undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %sv8i16 = call <8 x i16> @llvm.scmp.v8i16.v8i16(<8 x i16> undef, <8 x i16> undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %sv4i32 = call <4 x i32> @llvm.scmp.v4i32.v4i32(<4 x i32> undef, <4 x i32> undef)
; CHECK-SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret void
;
%u8 = call i8 @llvm.ucmp(i8 undef, i8 undef)
%u16 = call i16 @llvm.ucmp(i16 undef, i16 undef)
%u32 = call i32 @llvm.ucmp(i32 undef, i32 undef)
%u64 = call i64 @llvm.ucmp(i64 undef, i64 undef)
%uv16i8 = call <16 x i8> @llvm.ucmp(<16 x i8> undef, <16 x i8> undef)
%uv8i16 = call <8 x i16> @llvm.ucmp(<8 x i16> undef, <8 x i16> undef)
%uv4i32 = call <4 x i32> @llvm.ucmp(<4 x i32> undef, <4 x i32> undef)
%s8 = call i8 @llvm.scmp(i8 undef, i8 undef)
%s16 = call i16 @llvm.scmp(i16 undef, i16 undef)
%s32 = call i32 @llvm.scmp(i32 undef, i32 undef)
%s64 = call i64 @llvm.scmp(i64 undef, i64 undef)
%sv16i8 = call <16 x i8> @llvm.scmp(<16 x i8> undef, <16 x i8> undef)
%sv8i16 = call <8 x i16> @llvm.scmp(<8 x i16> undef, <8 x i16> undef)
%sv4i32 = call <4 x i32> @llvm.scmp(<4 x i32> undef, <4 x i32> undef)
ret void
}
Loading
Loading