[AArch64][SVE] Add lowering for llvm fabs

[Asif]

Summary

Add the functionality to lower fabs for passthru variant

llvm/lib/Target/AArch64/AArch64ISelLowering.h

enum NodeType : unsigned {
   SRL_PRED,
   SUB_PRED,
   UDIV_PRED,
   UMAX_PRED,
   UMIN_PRED,
 
   // Predicated instructions with the result of inactive lanes provided by the
   // last operand.
+  FABS_MERGE_PASSTHRU,
   FCEIL_MERGE_PASSTHRU,
   FFLOOR_MERGE_PASSTHRU,
   FNEARBYINT_MERGE_PASSTHRU,
   FNEG_MERGE_PASSTHRU,
   FRECPX_MERGE_PASSTHRU,
   FRINT_MERGE_PASSTHRU,
   FROUND_MERGE_PASSTHRU,
   FROUNDEVEN_MERGE_PASSTHRU,

llvm/lib/Target/AArch64/AArch64ISelLowering.cpp

static bool isMergePassthruOpcode(unsigned Opc) {
   case AArch64ISD::FP_ROUND_MERGE_PASSTHRU:
   case AArch64ISD::FP_EXTEND_MERGE_PASSTHRU:
   case AArch64ISD::SINT_TO_FP_MERGE_PASSTHRU:
   case AArch64ISD::UINT_TO_FP_MERGE_PASSTHRU:
   case AArch64ISD::FCVTZU_MERGE_PASSTHRU:
   case AArch64ISD::FCVTZS_MERGE_PASSTHRU:
   case AArch64ISD::FSQRT_MERGE_PASSTHRU:
   case AArch64ISD::FRECPX_MERGE_PASSTHRU:
+  case AArch64ISD::FABS_MERGE_PASSTHRU:
     return true;
   }
 }
 
 AArch64TargetLowering::AArch64TargetLowering(const TargetMachine &TM,
                                              const AArch64Subtarget &STI)
     : TargetLowering(TM), Subtarget(&STI) {
   // AArch64 doesn't have comparisons which set GPRs or setcc instructions, so
for (auto VT : {MVT::nxv2f16, MVT::nxv4f16, MVT::nxv8f16, MVT::nxv2f32,
       setOperationAction(ISD::FCEIL, VT, Custom);
       setOperationAction(ISD::FFLOOR, VT, Custom);
       setOperationAction(ISD::FNEARBYINT, VT, Custom);
       setOperationAction(ISD::FRINT, VT, Custom);
       setOperationAction(ISD::FROUND, VT, Custom);
       setOperationAction(ISD::FROUNDEVEN, VT, Custom);
       setOperationAction(ISD::FTRUNC, VT, Custom);
       setOperationAction(ISD::FSQRT, VT, Custom);
+      setOperationAction(ISD::FABS, VT, Custom);
       setOperationAction(ISD::FP_EXTEND, VT, Custom);
       setOperationAction(ISD::FP_ROUND, VT, Custom);
     }
 
     setOperationAction(ISD::SPLAT_VECTOR, MVT::nxv8bf16, Custom);
 
     setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::i8, Custom);
     setOperationAction(ISD::INTRINSIC_WO_CHAIN, MVT::i16, Custom);
case AArch64ISD::FIRST_NUMBER:
     MAKE_CASE(AArch64ISD::FP_ROUND_MERGE_PASSTHRU)
     MAKE_CASE(AArch64ISD::FP_EXTEND_MERGE_PASSTHRU)
     MAKE_CASE(AArch64ISD::SINT_TO_FP_MERGE_PASSTHRU)
     MAKE_CASE(AArch64ISD::UINT_TO_FP_MERGE_PASSTHRU)
     MAKE_CASE(AArch64ISD::FCVTZU_MERGE_PASSTHRU)
     MAKE_CASE(AArch64ISD::FCVTZS_MERGE_PASSTHRU)
     MAKE_CASE(AArch64ISD::FSQRT_MERGE_PASSTHRU)
     MAKE_CASE(AArch64ISD::FRECPX_MERGE_PASSTHRU)
+    MAKE_CASE(AArch64ISD::FABS_MERGE_PASSTHRU)
     MAKE_CASE(AArch64ISD::SETCC_MERGE_ZERO)
     MAKE_CASE(AArch64ISD::ADC)
     MAKE_CASE(AArch64ISD::SBC)
     MAKE_CASE(AArch64ISD::ADDS)
     MAKE_CASE(AArch64ISD::SUBS)
     MAKE_CASE(AArch64ISD::ADCS)
     MAKE_CASE(AArch64ISD::SBCS)
     MAKE_CASE(AArch64ISD::ANDS)
return DAG.getNode(AArch64ISD::FCVTZS_MERGE_PASSTHRU, dl,
                        Op.getValueType(), Op.getOperand(2), Op.getOperand(3),
                        Op.getOperand(1));
   case Intrinsic::aarch64_sve_fsqrt:
     return DAG.getNode(AArch64ISD::FSQRT_MERGE_PASSTHRU, dl, Op.getValueType(),
                        Op.getOperand(2), Op.getOperand(3), Op.getOperand(1));
   case Intrinsic::aarch64_sve_frecpx:
     return DAG.getNode(AArch64ISD::FRECPX_MERGE_PASSTHRU, dl, Op.getValueType(),
                        Op.getOperand(2), Op.getOperand(3), Op.getOperand(1));
+  case Intrinsic::aarch64_sve_fabs:
+    return DAG.getNode(AArch64ISD::FABS_MERGE_PASSTHRU, dl, Op.getValueType(),
+                       Op.getOperand(2), Op.getOperand(3), Op.getOperand(1));
   case Intrinsic::aarch64_sve_convert_to_svbool: {
     EVT OutVT = Op.getValueType();
     EVT InVT = Op.getOperand(1).getValueType();
     // Return the operand if the cast isn't changing type,
     // i.e. <n x 16 x i1> -> <n x 16 x i1>
     if (InVT == OutVT)
       return Op.getOperand(1);
     // Otherwise, zero the newly introduced lanes.
SDValue AArch64TargetLowering::LowerOperation(SDValue Op,
   case ISD::FROUND:
     return LowerToPredicatedOp(Op, DAG, AArch64ISD::FROUND_MERGE_PASSTHRU);
   case ISD::FROUNDEVEN:
     return LowerToPredicatedOp(Op, DAG, AArch64ISD::FROUNDEVEN_MERGE_PASSTHRU);
   case ISD::FTRUNC:
     return LowerToPredicatedOp(Op, DAG, AArch64ISD::FTRUNC_MERGE_PASSTHRU);
   case ISD::FSQRT:
     return LowerToPredicatedOp(Op, DAG, AArch64ISD::FSQRT_MERGE_PASSTHRU);
+  case ISD::FABS:
+    return LowerToPredicatedOp(Op, DAG, AArch64ISD::FABS_MERGE_PASSTHRU);
   case ISD::FP_ROUND:
   case ISD::STRICT_FP_ROUND:
     return LowerFP_ROUND(Op, DAG);
   case ISD::FP_EXTEND:
     return LowerFP_EXTEND(Op, DAG);
   case ISD::FRAMEADDR:
     return LowerFRAMEADDR(Op, DAG);
   case ISD::SPONENTRY:

llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td

 def AArch64umin_p : SDNode<"AArch64ISD::UMIN_PRED", SDT_AArch64Arith>;
 
 def SDT_AArch64IntExtend : SDTypeProfile<1, 4, [
   SDTCisVec<0>, SDTCisVec<1>, SDTCisVec<2>, SDTCisVT<3, OtherVT>, SDTCisVec<4>,
   SDTCVecEltisVT<1,i1>, SDTCisSameAs<0,2>, SDTCisVTSmallerThanOp<3, 2>, SDTCisSameAs<0,4>
 ]>;
 
 // Predicated operations with the result of inactive lanes provided by the last operand.
-def AArch64fneg_mt : SDNode<"AArch64ISD::FNEG_MERGE_PASSTHRU", SDT_AArch64Arith>;
-def AArch64sxt_mt  : SDNode<"AArch64ISD::SIGN_EXTEND_INREG_MERGE_PASSTHRU", SDT_AArch64IntExtend>;
-def AArch64uxt_mt  : SDNode<"AArch64ISD::ZERO_EXTEND_INREG_MERGE_PASSTHRU", SDT_AArch64IntExtend>;
+def AArch64fneg_mt   : SDNode<"AArch64ISD::FNEG_MERGE_PASSTHRU", SDT_AArch64Arith>;
+def AArch64fabs_mt   : SDNode<"AArch64ISD::FABS_MERGE_PASSTHRU", SDT_AArch64Arith>;
+def AArch64sxt_mt    : SDNode<"AArch64ISD::SIGN_EXTEND_INREG_MERGE_PASSTHRU", SDT_AArch64IntExtend>;
+def AArch64uxt_mt    : SDNode<"AArch64ISD::ZERO_EXTEND_INREG_MERGE_PASSTHRU", SDT_AArch64IntExtend>;
 def AArch64frintp_mt : SDNode<"AArch64ISD::FCEIL_MERGE_PASSTHRU", SDT_AArch64Arith>;
 def AArch64frintm_mt : SDNode<"AArch64ISD::FFLOOR_MERGE_PASSTHRU", SDT_AArch64Arith>;
 def AArch64frinti_mt : SDNode<"AArch64ISD::FNEARBYINT_MERGE_PASSTHRU", SDT_AArch64Arith>;
 def AArch64frintx_mt : SDNode<"AArch64ISD::FRINT_MERGE_PASSTHRU", SDT_AArch64Arith>;
 def AArch64frinta_mt : SDNode<"AArch64ISD::FROUND_MERGE_PASSTHRU", SDT_AArch64Arith>;
 def AArch64frintn_mt : SDNode<"AArch64ISD::FROUNDEVEN_MERGE_PASSTHRU", SDT_AArch64Arith>;
 def AArch64frintz_mt : SDNode<"AArch64ISD::FTRUNC_MERGE_PASSTHRU", SDT_AArch64Arith>;
-def AArch64fsqrt_mt : SDNode<"AArch64ISD::FSQRT_MERGE_PASSTHRU", SDT_AArch64Arith>;
+def AArch64fsqrt_mt  : SDNode<"AArch64ISD::FSQRT_MERGE_PASSTHRU", SDT_AArch64Arith>;
 def AArch64frecpx_mt : SDNode<"AArch64ISD::FRECPX_MERGE_PASSTHRU", SDT_AArch64Arith>;
 
 def SDT_AArch64FCVT : SDTypeProfile<1, 3, [
   SDTCisVec<0>, SDTCisVec<1>, SDTCisVec<2>, SDTCisVec<3>,
   SDTCVecEltisVT<1,i1>
 ]>;
 
 def SDT_AArch64FCVTR : SDTypeProfile<1, 4, [
let Predicates = [HasSVE] in {
   defm CNT_ZPmZ  : sve_int_un_pred_arit_1<   0b010, "cnt",  int_aarch64_sve_cnt>;
 
  let Predicates = [HasSVE, HasBF16] in {
   def : SVE_3_Op_Pat<nxv8i16, int_aarch64_sve_cnt, nxv8i16, nxv8i1, nxv8bf16, !cast<Instruction>(CNT_ZPmZ_H)>;
  }
 
   defm CNOT_ZPmZ : sve_int_un_pred_arit_1<   0b011, "cnot", int_aarch64_sve_cnot>;
   defm NOT_ZPmZ  : sve_int_un_pred_arit_1<   0b110, "not",  int_aarch64_sve_not>;
-  defm FABS_ZPmZ : sve_int_un_pred_arit_1_fp<0b100, "fabs", int_aarch64_sve_fabs, null_frag>;
-  defm FNEG_ZPmZ : sve_int_un_pred_arit_1_fp<0b101, "fneg", null_frag, AArch64fneg_mt>;
+  defm FABS_ZPmZ : sve_int_un_pred_arit_1_fp<0b100, "fabs", AArch64fabs_mt>;
+  defm FNEG_ZPmZ : sve_int_un_pred_arit_1_fp<0b101, "fneg", AArch64fneg_mt>;
 
   defm SMAX_ZPmZ : sve_int_bin_pred_arit_1<0b000, "smax", "SMAX_ZPZZ", int_aarch64_sve_smax, DestructiveBinaryComm>;
   defm UMAX_ZPmZ : sve_int_bin_pred_arit_1<0b001, "umax", "UMAX_ZPZZ", int_aarch64_sve_umax, DestructiveBinaryComm>;
   defm SMIN_ZPmZ : sve_int_bin_pred_arit_1<0b010, "smin", "SMIN_ZPZZ", int_aarch64_sve_smin, DestructiveBinaryComm>;
   defm UMIN_ZPmZ : sve_int_bin_pred_arit_1<0b011, "umin", "UMIN_ZPZZ", int_aarch64_sve_umin, DestructiveBinaryComm>;
   defm SABD_ZPmZ : sve_int_bin_pred_arit_1<0b100, "sabd", "SABD_ZPZZ", int_aarch64_sve_sabd, DestructiveBinaryComm>;
   defm UABD_ZPmZ : sve_int_bin_pred_arit_1<0b101, "uabd", "UABD_ZPZZ", int_aarch64_sve_uabd, DestructiveBinaryComm>;
 

llvm/lib/Target/AArch64/SVEInstrFormats.td

multiclass sve_int_un_pred_arit_1<bits<3> opc, string asm,
   def : SVE_3_Op_Pat<nxv4i32, op, nxv4i32, nxv4i1,  nxv4i32, !cast<Instruction>(NAME # _S)>;
   def : SVE_3_Op_Pat<nxv2i64, op, nxv2i64, nxv2i1,  nxv2i64, !cast<Instruction>(NAME # _D)>;
 
   def : SVE_3_Op_Pat<nxv8i16, op, nxv8i16, nxv8i1, nxv8f16, !cast<Instruction>(NAME # _H)>;
   def : SVE_3_Op_Pat<nxv4i32, op, nxv4i32, nxv4i1, nxv4f32, !cast<Instruction>(NAME # _S)>;
   def : SVE_3_Op_Pat<nxv2i64, op, nxv2i64, nxv2i1, nxv2f64, !cast<Instruction>(NAME # _D)>;
 }
 
-// TODO: Remove int_op once its last use is converted to ir_op.
-multiclass sve_int_un_pred_arit_1_fp<bits<3> opc, string asm,
-                                     SDPatternOperator int_op,
-                                     SDPatternOperator ir_op> {
+multiclass sve_int_un_pred_arit_1_fp<bits<3> opc, string asm, SDPatternOperator op> {
   def _H : sve_int_un_pred_arit<0b01, { opc, 0b1 }, asm, ZPR16>;
   def _S : sve_int_un_pred_arit<0b10, { opc, 0b1 }, asm, ZPR32>;
   def _D : sve_int_un_pred_arit<0b11, { opc, 0b1 }, asm, ZPR64>;
 
-  def : SVE_3_Op_Pat<nxv8f16, int_op, nxv8f16, nxv8i1, nxv8f16, !cast<Instruction>(NAME # _H)>;
-  def : SVE_3_Op_Pat<nxv4f32, int_op, nxv4f32, nxv4i1, nxv4f32, !cast<Instruction>(NAME # _S)>;
-  def : SVE_3_Op_Pat<nxv2f64, int_op, nxv2f64, nxv2i1, nxv2f64, !cast<Instruction>(NAME # _D)>;
-
-  def : SVE_1_Op_Passthru_Pat<nxv8f16, ir_op, nxv8i1, nxv8f16, !cast<Instruction>(NAME # _H)>;
-  def : SVE_1_Op_Passthru_Pat<nxv4f16, ir_op, nxv4i1, nxv4f16, !cast<Instruction>(NAME # _H)>;
-  def : SVE_1_Op_Passthru_Pat<nxv2f16, ir_op, nxv2i1, nxv2f16, !cast<Instruction>(NAME # _H)>;
-  def : SVE_1_Op_Passthru_Pat<nxv4f32, ir_op, nxv4i1, nxv4f32, !cast<Instruction>(NAME # _S)>;
-  def : SVE_1_Op_Passthru_Pat<nxv2f32, ir_op, nxv2i1, nxv2f32, !cast<Instruction>(NAME # _S)>;
-  def : SVE_1_Op_Passthru_Pat<nxv2f64, ir_op, nxv2i1, nxv2f64, !cast<Instruction>(NAME # _D)>;
+  def : SVE_1_Op_Passthru_Pat<nxv8f16, op, nxv8i1, nxv8f16, !cast<Instruction>(NAME # _H)>;
+  def : SVE_1_Op_Passthru_Pat<nxv4f16, op, nxv4i1, nxv4f16, !cast<Instruction>(NAME # _H)>;
+  def : SVE_1_Op_Passthru_Pat<nxv2f16, op, nxv2i1, nxv2f16, !cast<Instruction>(NAME # _H)>;
+  def : SVE_1_Op_Passthru_Pat<nxv4f32, op, nxv4i1, nxv4f32, !cast<Instruction>(NAME # _S)>;
+  def : SVE_1_Op_Passthru_Pat<nxv2f32, op, nxv2i1, nxv2f32, !cast<Instruction>(NAME # _S)>;
+  def : SVE_1_Op_Passthru_Pat<nxv2f64, op, nxv2i1, nxv2f64, !cast<Instruction>(NAME # _D)>;
 }
 
 //===----------------------------------------------------------------------===//
 // SVE Integer Wide Immediate - Unpredicated Group
 //===----------------------------------------------------------------------===//
 class sve_int_dup_imm<bits<2> sz8_64, string asm,
                       ZPRRegOp zprty, Operand immtype>
 : I<(outs zprty:$Zd), (ins immtype:$imm),

llvm/test/CodeGen/AArch64/sve-fp.ll

 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ptrue p0.d
 ; CHECK-NEXT:    fsqrt z0.d, p0/m, z0.d
 ; CHECK-NEXT:    ret
   %res = call <vscale x 2 x double> @llvm.sqrt.nxv2f64(<vscale x 2 x double> %a)
   ret <vscale x 2 x double> %res
 }
 
+; FABS
+
+define <vscale x 8 x half> @fabs_nxv8f16(<vscale x 8 x half> %a) {
+; CHECK-LABEL: fabs_nxv8f16:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    ptrue p0.h
+; CHECK-NEXT:    fabs z0.h, p0/m, z0.h
+; CHECK-NEXT:    ret
+  %res = call <vscale x 8 x half> @llvm.fabs.nxv8f16(<vscale x 8 x half> %a)
+  ret <vscale x 8 x half> %res
+}
+
+define <vscale x 4 x half> @fabs_nxv4f16(<vscale x 4 x half> %a) {
+; CHECK-LABEL: fabs_nxv4f16:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    ptrue p0.s
+; CHECK-NEXT:    fabs z0.h, p0/m, z0.h
+; CHECK-NEXT:    ret
+  %res = call <vscale x 4 x half> @llvm.fabs.nxv4f16(<vscale x 4 x half> %a)
+  ret <vscale x 4 x half> %res
+}
+
+define <vscale x 2 x half> @fabs_nxv2f16(<vscale x 2 x half> %a) {
+; CHECK-LABEL: fabs_nxv2f16:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    ptrue p0.d
+; CHECK-NEXT:    fabs z0.h, p0/m, z0.h
+; CHECK-NEXT:    ret
+  %res = call <vscale x 2 x half> @llvm.fabs.nxv2f16(<vscale x 2 x half> %a)
+  ret <vscale x 2 x half> %res
+}
+
+define <vscale x 4 x float> @fabs_nxv4f32(<vscale x 4 x float> %a) {
+; CHECK-LABEL: fabs_nxv4f32:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    ptrue p0.s
+; CHECK-NEXT:    fabs z0.s, p0/m, z0.s
+; CHECK-NEXT:    ret
+  %res = call <vscale x 4 x float> @llvm.fabs.nxv4f32(<vscale x 4 x float> %a)
+  ret <vscale x 4 x float> %res
+}
+
+define <vscale x 2 x float> @fabs_nxv2f32(<vscale x 2 x float> %a) {
+; CHECK-LABEL: fabs_nxv2f32:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    ptrue p0.d
+; CHECK-NEXT:    fabs z0.s, p0/m, z0.s
+; CHECK-NEXT:    ret
+  %res = call <vscale x 2 x float> @llvm.fabs.nxv2f32(<vscale x 2 x float> %a)
+  ret <vscale x 2 x float> %res
+}
+
+define <vscale x 2 x double> @fabs_nxv2f64(<vscale x 2 x double> %a) {
+; CHECK-LABEL: fabs_nxv2f64:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    ptrue p0.d
+; CHECK-NEXT:    fabs z0.d, p0/m, z0.d
+; CHECK-NEXT:    ret
+  %res = call <vscale x 2 x double> @llvm.fabs.nxv2f64(<vscale x 2 x double> %a)
+  ret <vscale x 2 x double> %res
+}
+
 declare <vscale x 8 x half> @llvm.aarch64.sve.frecps.x.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>)
 declare <vscale x 4 x float>  @llvm.aarch64.sve.frecps.x.nxv4f32(<vscale x 4 x float> , <vscale x 4 x float>)
 declare <vscale x 2 x double> @llvm.aarch64.sve.frecps.x.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>)
 
 declare <vscale x 8 x half> @llvm.aarch64.sve.frsqrts.x.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>)
 declare <vscale x 4 x float> @llvm.aarch64.sve.frsqrts.x.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>)
 declare <vscale x 2 x double> @llvm.aarch64.sve.frsqrts.x.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>)
 
 declare <vscale x 2 x double> @llvm.fma.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>)
 declare <vscale x 4 x float> @llvm.fma.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>)
 declare <vscale x 2 x float> @llvm.fma.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, <vscale x 2 x float>)
 declare <vscale x 8 x half> @llvm.fma.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>)
 declare <vscale x 4 x half> @llvm.fma.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, <vscale x 4 x half>)
 declare <vscale x 2 x half> @llvm.fma.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, <vscale x 2 x half>)
 
 declare <vscale x 8 x half> @llvm.sqrt.nxv8f16( <vscale x 8 x half>)
 declare <vscale x 4 x half> @llvm.sqrt.nxv4f16( <vscale x 4 x half>)
 declare <vscale x 2 x half> @llvm.sqrt.nxv2f16( <vscale x 2 x half>)
 declare <vscale x 4 x float> @llvm.sqrt.nxv4f32(<vscale x 4 x float>)
 declare <vscale x 2 x float> @llvm.sqrt.nxv2f32(<vscale x 2 x float>)
 declare <vscale x 2 x double> @llvm.sqrt.nxv2f64(<vscale x 2 x double>)
 
+declare <vscale x 8 x half> @llvm.fabs.nxv8f16( <vscale x 8 x half>)
+declare <vscale x 4 x half> @llvm.fabs.nxv4f16( <vscale x 4 x half>)
+declare <vscale x 2 x half> @llvm.fabs.nxv2f16( <vscale x 2 x half>)
+declare <vscale x 4 x float> @llvm.fabs.nxv4f32(<vscale x 4 x float>)
+declare <vscale x 2 x float> @llvm.fabs.nxv2f32(<vscale x 2 x float>)
+declare <vscale x 2 x double> @llvm.fabs.nxv2f64(<vscale x 2 x double>)
+
 ; Function Attrs: nounwind readnone
 declare double @llvm.aarch64.sve.faddv.nxv2f64(<vscale x 2 x i1>, <vscale x 2 x double>) #2