Skip to content

Revert "[SDag][ARM][RISCV] Allow lowering CTPOP into a libcall" #101740

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
Aug 2, 2024
Merged

Revert "[SDag][ARM][RISCV] Allow lowering CTPOP into a libcall" #101740

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
3 changes: 0 additions & 3 deletions llvm/include/llvm/IR/RuntimeLibcalls.def
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -85,9 +85,6 @@ HANDLE_LIBCALL(NEG_I64, "__negdi2")
HANDLE_LIBCALL(CTLZ_I32, "__clzsi2")
HANDLE_LIBCALL(CTLZ_I64, "__clzdi2")
HANDLE_LIBCALL(CTLZ_I128, "__clzti2")
HANDLE_LIBCALL(CTPOP_I32, "__popcountsi2")
HANDLE_LIBCALL(CTPOP_I64, "__popcountdi2")
HANDLE_LIBCALL(CTPOP_I128, "__popcountti2")

// Floating-point
HANDLE_LIBCALL(ADD_F32, "__addsf3")
Expand Down
38 changes: 21 additions & 17 deletions llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -140,9 +140,12 @@ class SelectionDAGLegalize {
RTLIB::Libcall Call_F128,
RTLIB::Libcall Call_PPCF128,
SmallVectorImpl<SDValue> &Results);
SDValue ExpandIntLibCall(SDNode *Node, bool IsSigned, RTLIB::Libcall Call_I8,
RTLIB::Libcall Call_I16, RTLIB::Libcall Call_I32,
RTLIB::Libcall Call_I64, RTLIB::Libcall Call_I128);
SDValue ExpandIntLibCall(SDNode *Node, bool isSigned,
RTLIB::Libcall Call_I8,
RTLIB::Libcall Call_I16,
RTLIB::Libcall Call_I32,
RTLIB::Libcall Call_I64,
RTLIB::Libcall Call_I128);
void ExpandArgFPLibCall(SDNode *Node,
RTLIB::Libcall Call_F32, RTLIB::Libcall Call_F64,
RTLIB::Libcall Call_F80, RTLIB::Libcall Call_F128,
Expand Down Expand Up @@ -2206,7 +2209,7 @@ void SelectionDAGLegalize::ExpandFPLibCall(SDNode* Node,
ExpandFPLibCall(Node, LC, Results);
}

SDValue SelectionDAGLegalize::ExpandIntLibCall(SDNode *Node, bool IsSigned,
SDValue SelectionDAGLegalize::ExpandIntLibCall(SDNode* Node, bool isSigned,
RTLIB::Libcall Call_I8,
RTLIB::Libcall Call_I16,
RTLIB::Libcall Call_I32,
Expand All @@ -2221,9 +2224,7 @@ SDValue SelectionDAGLegalize::ExpandIntLibCall(SDNode *Node, bool IsSigned,
case MVT::i64: LC = Call_I64; break;
case MVT::i128: LC = Call_I128; break;
}
assert(LC != RTLIB::UNKNOWN_LIBCALL &&
"LibCall explicitly requested, but not available");
return ExpandLibCall(LC, Node, IsSigned).first;
return ExpandLibCall(LC, Node, isSigned).first;
}

/// Expand the node to a libcall based on first argument type (for instance
Expand Down Expand Up @@ -4999,16 +5000,19 @@ void SelectionDAGLegalize::ConvertNodeToLibcall(SDNode *Node) {
RTLIB::MUL_I64, RTLIB::MUL_I128));
break;
case ISD::CTLZ_ZERO_UNDEF:
Results.push_back(ExpandIntLibCall(Node, /*IsSigned=*/false,
RTLIB::UNKNOWN_LIBCALL,
RTLIB::UNKNOWN_LIBCALL, RTLIB::CTLZ_I32,
RTLIB::CTLZ_I64, RTLIB::CTLZ_I128));
break;
case ISD::CTPOP:
Results.push_back(ExpandIntLibCall(Node, /*IsSigned=*/false,
RTLIB::UNKNOWN_LIBCALL,
RTLIB::UNKNOWN_LIBCALL, RTLIB::CTPOP_I32,
RTLIB::CTPOP_I64, RTLIB::CTPOP_I128));
switch (Node->getSimpleValueType(0).SimpleTy) {
default:
llvm_unreachable("LibCall explicitly requested, but not available");
case MVT::i32:
Results.push_back(ExpandLibCall(RTLIB::CTLZ_I32, Node, false).first);
break;
case MVT::i64:
Results.push_back(ExpandLibCall(RTLIB::CTLZ_I64, Node, false).first);
break;
case MVT::i128:
Results.push_back(ExpandLibCall(RTLIB::CTLZ_I128, Node, false).first);
break;
}
break;
case ISD::RESET_FPENV: {
// It is legalized to call 'fesetenv(FE_DFL_ENV)'. On most targets
Expand Down
32 changes: 7 additions & 25 deletions llvm/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3850,33 +3850,15 @@ void DAGTypeLegalizer::ExpandIntRes_CTLZ(SDNode *N,
Hi = DAG.getConstant(0, dl, NVT);
}

void DAGTypeLegalizer::ExpandIntRes_CTPOP(SDNode *N, SDValue &Lo, SDValue &Hi) {
SDValue Op = N->getOperand(0);
EVT VT = N->getValueType(0);
SDLoc DL(N);

if (TLI.getOperationAction(ISD::CTPOP, VT) == TargetLoweringBase::LibCall) {
RTLIB::Libcall LC = RTLIB::UNKNOWN_LIBCALL;
if (VT == MVT::i32)
LC = RTLIB::CTPOP_I32;
else if (VT == MVT::i64)
LC = RTLIB::CTPOP_I64;
else if (VT == MVT::i128)
LC = RTLIB::CTPOP_I128;
assert(LC != RTLIB::UNKNOWN_LIBCALL && TLI.getLibcallName(LC) &&
"LibCall explicitly requested, but not available");
TargetLowering::MakeLibCallOptions CallOptions;
SDValue Res = TLI.makeLibCall(DAG, LC, VT, Op, CallOptions, DL).first;
SplitInteger(Res, Lo, Hi);
return;
}

void DAGTypeLegalizer::ExpandIntRes_CTPOP(SDNode *N,
SDValue &Lo, SDValue &Hi) {
SDLoc dl(N);
// ctpop(HiLo) -> ctpop(Hi)+ctpop(Lo)
GetExpandedInteger(Op, Lo, Hi);
GetExpandedInteger(N->getOperand(0), Lo, Hi);
EVT NVT = Lo.getValueType();
Lo = DAG.getNode(ISD::ADD, DL, NVT, DAG.getNode(ISD::CTPOP, DL, NVT, Lo),
DAG.getNode(ISD::CTPOP, DL, NVT, Hi));
Hi = DAG.getConstant(0, DL, NVT);
Lo = DAG.getNode(ISD::ADD, dl, NVT, DAG.getNode(ISD::CTPOP, dl, NVT, Lo),
DAG.getNode(ISD::CTPOP, dl, NVT, Hi));
Hi = DAG.getConstant(0, dl, NVT);
}

void DAGTypeLegalizer::ExpandIntRes_CTTZ(SDNode *N,
Expand Down
5 changes: 2 additions & 3 deletions llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9171,9 +9171,8 @@ SDValue TargetLowering::expandCTTZ(SDNode *Node, SelectionDAG &DAG) const {
!isOperationLegalOrCustomOrPromote(ISD::XOR, VT)))
return SDValue();

// Emit Table Lookup if ISD::CTPOP used in the fallback path below is going
// to be expanded or converted to a libcall.
if (!VT.isVector() && !isOperationLegalOrCustomOrPromote(ISD::CTPOP, VT) &&
// Emit Table Lookup if ISD::CTLZ and ISD::CTPOP are not legal.
if (!VT.isVector() && isOperationExpand(ISD::CTPOP, VT) &&
!isOperationLegal(ISD::CTLZ, VT))
if (SDValue V = CTTZTableLookup(Node, DAG, dl, VT, Op, NumBitsPerElt))
return V;
Expand Down
5 changes: 1 addition & 4 deletions llvm/lib/Target/RISCV/RISCVISelLowering.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -393,10 +393,7 @@ RISCVTargetLowering::RISCVTargetLowering(const TargetMachine &TM,
setOperationAction({ISD::CTTZ, ISD::CTTZ_ZERO_UNDEF}, MVT::i32, Custom);
}
} else {
setOperationAction(ISD::CTTZ, XLenVT, Expand);
if (!Subtarget.is64Bit())
setOperationAction(ISD::CTPOP, MVT::i32, LibCall);
setOperationAction(ISD::CTPOP, MVT::i64, LibCall);
setOperationAction({ISD::CTTZ, ISD::CTPOP}, XLenVT, Expand);
if (RV64LegalI32 && Subtarget.is64Bit())
setOperationAction({ISD::CTTZ, ISD::CTPOP}, MVT::i32, Expand);
}
Expand Down
16 changes: 8 additions & 8 deletions llvm/test/Analysis/CostModel/RISCV/int-bit-manip.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -159,7 +159,7 @@ define void @bitreverse() {

define void @ctpop() {
; NOZVBB-LABEL: 'ctpop'
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %1 = call i8 @llvm.ctpop.i8(i8 undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %1 = call i8 @llvm.ctpop.i8(i8 undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %2 = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %3 = call <4 x i8> @llvm.ctpop.v4i8(<4 x i8> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %4 = call <8 x i8> @llvm.ctpop.v8i8(<8 x i8> undef)
Expand All @@ -169,7 +169,7 @@ define void @ctpop() {
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %8 = call <vscale x 4 x i8> @llvm.ctpop.nxv4i8(<vscale x 4 x i8> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %9 = call <vscale x 8 x i8> @llvm.ctpop.nxv8i8(<vscale x 8 x i8> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %10 = call <vscale x 16 x i8> @llvm.ctpop.nxv16i8(<vscale x 16 x i8> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %11 = call i16 @llvm.ctpop.i16(i16 undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %11 = call i16 @llvm.ctpop.i16(i16 undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %12 = call <2 x i16> @llvm.ctpop.v2i16(<2 x i16> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %13 = call <4 x i16> @llvm.ctpop.v4i16(<4 x i16> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %14 = call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> undef)
Expand All @@ -179,7 +179,7 @@ define void @ctpop() {
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %18 = call <vscale x 4 x i16> @llvm.ctpop.nxv4i16(<vscale x 4 x i16> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %19 = call <vscale x 8 x i16> @llvm.ctpop.nxv8i16(<vscale x 8 x i16> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %20 = call <vscale x 16 x i16> @llvm.ctpop.nxv16i16(<vscale x 16 x i16> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %21 = call i32 @llvm.ctpop.i32(i32 undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %21 = call i32 @llvm.ctpop.i32(i32 undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %22 = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %23 = call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %24 = call <8 x i32> @llvm.ctpop.v8i32(<8 x i32> undef)
Expand All @@ -189,7 +189,7 @@ define void @ctpop() {
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %28 = call <vscale x 4 x i32> @llvm.ctpop.nxv4i32(<vscale x 4 x i32> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %29 = call <vscale x 8 x i32> @llvm.ctpop.nxv8i32(<vscale x 8 x i32> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %30 = call <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %31 = call i64 @llvm.ctpop.i64(i64 undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %31 = call i64 @llvm.ctpop.i64(i64 undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %32 = call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %33 = call <4 x i64> @llvm.ctpop.v4i64(<4 x i64> undef)
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 21 for instruction: %34 = call <8 x i64> @llvm.ctpop.v8i64(<8 x i64> undef)
Expand All @@ -202,7 +202,7 @@ define void @ctpop() {
; NOZVBB-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; ZVBB-LABEL: 'ctpop'
; ZVBB-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %1 = call i8 @llvm.ctpop.i8(i8 undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %1 = call i8 @llvm.ctpop.i8(i8 undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x i8> @llvm.ctpop.v4i8(<4 x i8> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x i8> @llvm.ctpop.v8i8(<8 x i8> undef)
Expand All @@ -212,7 +212,7 @@ define void @ctpop() {
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x i8> @llvm.ctpop.nxv4i8(<vscale x 4 x i8> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x i8> @llvm.ctpop.nxv8i8(<vscale x 8 x i8> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x i8> @llvm.ctpop.nxv16i8(<vscale x 16 x i8> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %11 = call i16 @llvm.ctpop.i16(i16 undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %11 = call i16 @llvm.ctpop.i16(i16 undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i16> @llvm.ctpop.v2i16(<2 x i16> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i16> @llvm.ctpop.v4i16(<4 x i16> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> undef)
Expand All @@ -222,7 +222,7 @@ define void @ctpop() {
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %18 = call <vscale x 4 x i16> @llvm.ctpop.nxv4i16(<vscale x 4 x i16> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %19 = call <vscale x 8 x i16> @llvm.ctpop.nxv8i16(<vscale x 8 x i16> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %20 = call <vscale x 16 x i16> @llvm.ctpop.nxv16i16(<vscale x 16 x i16> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %21 = call i32 @llvm.ctpop.i32(i32 undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %21 = call i32 @llvm.ctpop.i32(i32 undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %22 = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %23 = call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %24 = call <8 x i32> @llvm.ctpop.v8i32(<8 x i32> undef)
Expand All @@ -232,7 +232,7 @@ define void @ctpop() {
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %28 = call <vscale x 4 x i32> @llvm.ctpop.nxv4i32(<vscale x 4 x i32> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %29 = call <vscale x 8 x i32> @llvm.ctpop.nxv8i32(<vscale x 8 x i32> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %30 = call <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %31 = call i64 @llvm.ctpop.i64(i64 undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %31 = call i64 @llvm.ctpop.i64(i64 undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %32 = call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %33 = call <4 x i64> @llvm.ctpop.v4i64(<4 x i64> undef)
; ZVBB-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %34 = call <8 x i64> @llvm.ctpop.v8i64(<8 x i64> undef)
Expand Down
Loading
Loading