[SDag][ARM][RISCV] Allow lowering CTPOP into a libcall (#101786)

This is a reland of #99752 with the bug fixed (see test diff in the
third commit in this PR).
All `popcount` libcalls return `int`, but `ISD::CTPOP` returns the type
of the argument, which can be wider than `int`. The fix is to make DAG
legalizer pass the correct return type to `makeLibCall` and sign-extend
the result afterwards.

Original commit message:
The main change is adding CTPOP to `RuntimeLibcalls.def` to allow
targets to use LibCall action for CTPOP. DAG legalizers are changed
accordingly.

Pull Request: https://github.com/llvm/llvm-project/pull/101786
diff --git a/llvm/include/llvm/CodeGen/BasicTTIImpl.h b/llvm/include/llvm/CodeGen/BasicTTIImpl.h
index 80df6d7..edb7701 100644
--- a/llvm/include/llvm/CodeGen/BasicTTIImpl.h
+++ b/llvm/include/llvm/CodeGen/BasicTTIImpl.h
@@ -2503,7 +2503,7 @@
         return (LT.first * 2);
       else
         return (LT.first * 1);
-    } else if (!TLI->isOperationExpand(ISD, LT.second)) {
+    } else if (TLI->isOperationCustom(ISD, LT.second)) {
       // If the operation is custom lowered then assume
       // that the code is twice as expensive.
       return (LT.first * 2);
diff --git a/llvm/include/llvm/IR/RuntimeLibcalls.def b/llvm/include/llvm/IR/RuntimeLibcalls.def
index 2545aebc..cd8e9b5 100644
--- a/llvm/include/llvm/IR/RuntimeLibcalls.def
+++ b/llvm/include/llvm/IR/RuntimeLibcalls.def
@@ -85,6 +85,9 @@
 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")
diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
index b596f6b..3e47136 100644
--- a/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
@@ -129,7 +129,8 @@
                                      ArrayRef<int> Mask) const;
 
   std::pair<SDValue, SDValue> ExpandLibCall(RTLIB::Libcall LC, SDNode *Node,
-                        TargetLowering::ArgListTy &&Args, bool isSigned);
+                                            TargetLowering::ArgListTy &&Args,
+                                            bool IsSigned, EVT RetVT);
   std::pair<SDValue, SDValue> ExpandLibCall(RTLIB::Libcall LC, SDNode *Node, bool isSigned);
 
   void ExpandFPLibCall(SDNode *Node, RTLIB::Libcall LC,
@@ -150,6 +151,9 @@
                           RTLIB::Libcall Call_F80, RTLIB::Libcall Call_F128,
                           RTLIB::Libcall Call_PPCF128,
                           SmallVectorImpl<SDValue> &Results);
+  SDValue ExpandBitCountingLibCall(SDNode *Node, RTLIB::Libcall CallI32,
+                                   RTLIB::Libcall CallI64,
+                                   RTLIB::Libcall CallI128);
   void ExpandDivRemLibCall(SDNode *Node, SmallVectorImpl<SDValue> &Results);
   void ExpandSinCosLibCall(SDNode *Node, SmallVectorImpl<SDValue> &Results);
 
@@ -2114,9 +2118,10 @@
 // register, return the lo part and set the hi part to the by-reg argument in
 // the first.  If it does fit into a single register, return the result and
 // leave the Hi part unset.
-std::pair<SDValue, SDValue> SelectionDAGLegalize::ExpandLibCall(RTLIB::Libcall LC, SDNode *Node,
-                                            TargetLowering::ArgListTy &&Args,
-                                            bool isSigned) {
+std::pair<SDValue, SDValue>
+SelectionDAGLegalize::ExpandLibCall(RTLIB::Libcall LC, SDNode *Node,
+                                    TargetLowering::ArgListTy &&Args,
+                                    bool IsSigned, EVT RetVT) {
   EVT CodePtrTy = TLI.getPointerTy(DAG.getDataLayout());
   SDValue Callee;
   if (const char *LibcallName = TLI.getLibcallName(LC))
@@ -2127,7 +2132,6 @@
                                 Node->getOperationName(&DAG));
   }
 
-  EVT RetVT = Node->getValueType(0);
   Type *RetTy = RetVT.getTypeForEVT(*DAG.getContext());
 
   // By default, the input chain to this libcall is the entry node of the
@@ -2147,7 +2151,7 @@
     InChain = TCChain;
 
   TargetLowering::CallLoweringInfo CLI(DAG);
-  bool signExtend = TLI.shouldSignExtendTypeInLibCall(RetTy, isSigned);
+  bool signExtend = TLI.shouldSignExtendTypeInLibCall(RetTy, IsSigned);
   CLI.setDebugLoc(SDLoc(Node))
       .setChain(InChain)
       .setLibCallee(TLI.getLibcallCallingConv(LC), RetTy, Callee,
@@ -2183,7 +2187,8 @@
     Args.push_back(Entry);
   }
 
-  return ExpandLibCall(LC, Node, std::move(Args), isSigned);
+  return ExpandLibCall(LC, Node, std::move(Args), isSigned,
+                       Node->getValueType(0));
 }
 
 void SelectionDAGLegalize::ExpandFPLibCall(SDNode* Node,
@@ -2259,6 +2264,50 @@
   ExpandFPLibCall(Node, LC, Results);
 }
 
+SDValue SelectionDAGLegalize::ExpandBitCountingLibCall(
+    SDNode *Node, RTLIB::Libcall CallI32, RTLIB::Libcall CallI64,
+    RTLIB::Libcall CallI128) {
+  RTLIB::Libcall LC;
+  switch (Node->getSimpleValueType(0).SimpleTy) {
+  default:
+    llvm_unreachable("Unexpected request for libcall!");
+  case MVT::i32:
+    LC = CallI32;
+    break;
+  case MVT::i64:
+    LC = CallI64;
+    break;
+  case MVT::i128:
+    LC = CallI128;
+    break;
+  }
+
+  // Bit-counting libcalls have one unsigned argument and return `int`.
+  // Note that `int` may be illegal on this target; ExpandLibCall will
+  // take care of promoting it to a legal type.
+  SDValue Op = Node->getOperand(0);
+  EVT IntVT =
+      EVT::getIntegerVT(*DAG.getContext(), DAG.getLibInfo().getIntSize());
+
+  TargetLowering::ArgListEntry Arg;
+  EVT ArgVT = Op.getValueType();
+  Type *ArgTy = ArgVT.getTypeForEVT(*DAG.getContext());
+  Arg.Node = Op;
+  Arg.Ty = ArgTy;
+  Arg.IsSExt = TLI.shouldSignExtendTypeInLibCall(ArgTy, /*IsSigned=*/false);
+  Arg.IsZExt = !Arg.IsSExt;
+
+  SDValue Res = ExpandLibCall(LC, Node, TargetLowering::ArgListTy{Arg},
+                              /*IsSigned=*/true, IntVT)
+                    .first;
+
+  // If ExpandLibCall created a tail call, the result was already
+  // of the correct type. Otherwise, we need to sign extend it.
+  if (Res.getValueType() != MVT::Other)
+    Res = DAG.getSExtOrTrunc(Res, SDLoc(Node), Node->getValueType(0));
+  return Res;
+}
+
 /// Issue libcalls to __{u}divmod to compute div / rem pairs.
 void
 SelectionDAGLegalize::ExpandDivRemLibCall(SDNode *Node,
@@ -4993,19 +5042,12 @@
                                        RTLIB::MUL_I64, RTLIB::MUL_I128));
     break;
   case ISD::CTLZ_ZERO_UNDEF:
-    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;
-    }
+    Results.push_back(ExpandBitCountingLibCall(
+        Node, RTLIB::CTLZ_I32, RTLIB::CTLZ_I64, RTLIB::CTLZ_I128));
+    break;
+  case ISD::CTPOP:
+    Results.push_back(ExpandBitCountingLibCall(
+        Node, RTLIB::CTPOP_I32, RTLIB::CTPOP_I64, RTLIB::CTPOP_I128));
     break;
   case ISD::RESET_FPENV: {
     // It is legalized to call 'fesetenv(FE_DFL_ENV)'. On most targets
diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp b/llvm/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp
index 53244a9..83dd519 100644
--- a/llvm/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeIntegerTypes.cpp
@@ -4012,15 +4012,35 @@
   SplitInteger(Result, Lo, Hi);
 }
 
-void DAGTypeLegalizer::ExpandIntRes_CTPOP(SDNode *N,
-                                          SDValue &Lo, SDValue &Hi) {
-  SDLoc dl(N);
+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;
+    EVT IntVT =
+        EVT::getIntegerVT(*DAG.getContext(), DAG.getLibInfo().getIntSize());
+    SDValue Res = TLI.makeLibCall(DAG, LC, IntVT, Op, CallOptions, DL).first;
+    SplitInteger(DAG.getSExtOrTrunc(Res, DL, VT), Lo, Hi);
+    return;
+  }
+
   // ctpop(HiLo) -> ctpop(Hi)+ctpop(Lo)
-  GetExpandedInteger(N->getOperand(0), Lo, Hi);
+  GetExpandedInteger(Op, 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,
diff --git a/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp b/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
index 3995216..3362677 100644
--- a/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
@@ -9390,8 +9390,9 @@
                         !isOperationLegalOrCustomOrPromote(ISD::XOR, VT)))
     return SDValue();
 
-  // Emit Table Lookup if ISD::CTLZ and ISD::CTPOP are not legal.
-  if (!VT.isVector() && isOperationExpand(ISD::CTPOP, VT) &&
+  // 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) &&
       !isOperationLegal(ISD::CTLZ, VT))
     if (SDValue V = CTTZTableLookup(Node, DAG, dl, VT, Op, NumBitsPerElt))
       return V;
diff --git a/llvm/lib/Target/ARM/ARMISelLowering.cpp b/llvm/lib/Target/ARM/ARMISelLowering.cpp
index 2290ac2..bdebd84 100644
--- a/llvm/lib/Target/ARM/ARMISelLowering.cpp
+++ b/llvm/lib/Target/ARM/ARMISelLowering.cpp
@@ -1221,7 +1221,8 @@
     setOperationAction(ISD::ROTR, VT, Expand);
   }
   setOperationAction(ISD::CTTZ,  MVT::i32, Custom);
-  setOperationAction(ISD::CTPOP, MVT::i32, Expand);
+  setOperationAction(ISD::CTPOP, MVT::i32, LibCall);
+  setOperationAction(ISD::CTPOP, MVT::i64, LibCall);
   if (!Subtarget->hasV5TOps() || Subtarget->isThumb1Only()) {
     setOperationAction(ISD::CTLZ, MVT::i32, Expand);
     setOperationAction(ISD::CTLZ_ZERO_UNDEF, MVT::i32, LibCall);
diff --git a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
index dadae2e..a75bd54 100644
--- a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
@@ -395,7 +395,12 @@
     if (Subtarget.is64Bit())
       setOperationAction({ISD::CTTZ, ISD::CTTZ_ZERO_UNDEF}, MVT::i32, Custom);
   } else {
-    setOperationAction({ISD::CTTZ, ISD::CTPOP}, XLenVT, Expand);
+    setOperationAction(ISD::CTTZ, XLenVT, Expand);
+    if (Subtarget.is64Bit())
+      setOperationAction(ISD::CTPOP, MVT::i128, LibCall);
+    else
+      setOperationAction(ISD::CTPOP, MVT::i32, LibCall);
+    setOperationAction(ISD::CTPOP, MVT::i64, LibCall);
   }
 
   if (Subtarget.hasStdExtZbb() || Subtarget.hasVendorXTHeadBb() ||
diff --git a/llvm/test/CodeGen/ARM/popcnt.ll b/llvm/test/CodeGen/ARM/popcnt.ll
index edcae5e..fc43873 100644
--- a/llvm/test/CodeGen/ARM/popcnt.ll
+++ b/llvm/test/CodeGen/ARM/popcnt.ll
@@ -324,76 +324,22 @@
 define i32 @ctpop32(i32 %x) nounwind readnone {
 ; CHECK-LABEL: ctpop32:
 ; CHECK:       @ %bb.0:
-; CHECK-NEXT:    ldr r1, .LCPI22_0
-; CHECK-NEXT:    ldr r2, .LCPI22_3
-; CHECK-NEXT:    and r1, r1, r0, lsr #1
-; CHECK-NEXT:    ldr r12, .LCPI22_1
-; CHECK-NEXT:    sub r0, r0, r1
-; CHECK-NEXT:    ldr r3, .LCPI22_2
-; CHECK-NEXT:    and r1, r0, r2
-; CHECK-NEXT:    and r0, r2, r0, lsr #2
-; CHECK-NEXT:    add r0, r1, r0
-; CHECK-NEXT:    add r0, r0, r0, lsr #4
-; CHECK-NEXT:    and r0, r0, r12
-; CHECK-NEXT:    mul r1, r0, r3
-; CHECK-NEXT:    lsr r0, r1, #24
-; CHECK-NEXT:    mov pc, lr
-; CHECK-NEXT:    .p2align 2
-; CHECK-NEXT:  @ %bb.1:
-; CHECK-NEXT:  .LCPI22_0:
-; CHECK-NEXT:    .long 1431655765 @ 0x55555555
-; CHECK-NEXT:  .LCPI22_1:
-; CHECK-NEXT:    .long 252645135 @ 0xf0f0f0f
-; CHECK-NEXT:  .LCPI22_2:
-; CHECK-NEXT:    .long 16843009 @ 0x1010101
-; CHECK-NEXT:  .LCPI22_3:
-; CHECK-NEXT:    .long 858993459 @ 0x33333333
+; CHECK-NEXT:    b __popcountsi2
   %count = tail call i32 @llvm.ctpop.i32(i32 %x)
   ret i32 %count
 }
 
-define i32 @ctpop64(i64 %x) nounwind readnone {
+define i64 @ctpop64(i64 %x) nounwind readnone {
 ; CHECK-LABEL: ctpop64:
 ; CHECK:       @ %bb.0:
-; CHECK-NEXT:    .save {r4, lr}
-; CHECK-NEXT:    push {r4, lr}
-; CHECK-NEXT:    ldr r2, .LCPI23_0
-; CHECK-NEXT:    ldr r3, .LCPI23_3
-; CHECK-NEXT:    and r4, r2, r0, lsr #1
-; CHECK-NEXT:    and r2, r2, r1, lsr #1
-; CHECK-NEXT:    sub r0, r0, r4
-; CHECK-NEXT:    sub r1, r1, r2
-; CHECK-NEXT:    and r4, r0, r3
-; CHECK-NEXT:    and r2, r1, r3
-; CHECK-NEXT:    and r0, r3, r0, lsr #2
-; CHECK-NEXT:    and r1, r3, r1, lsr #2
-; CHECK-NEXT:    add r0, r4, r0
-; CHECK-NEXT:    ldr lr, .LCPI23_1
-; CHECK-NEXT:    add r1, r2, r1
-; CHECK-NEXT:    ldr r12, .LCPI23_2
-; CHECK-NEXT:    add r0, r0, r0, lsr #4
-; CHECK-NEXT:    and r0, r0, lr
-; CHECK-NEXT:    add r1, r1, r1, lsr #4
-; CHECK-NEXT:    mul r2, r0, r12
-; CHECK-NEXT:    and r0, r1, lr
-; CHECK-NEXT:    mul r1, r0, r12
-; CHECK-NEXT:    lsr r0, r2, #24
-; CHECK-NEXT:    add r0, r0, r1, lsr #24
-; CHECK-NEXT:    pop {r4, lr}
+; CHECK-NEXT:    .save {r11, lr}
+; CHECK-NEXT:    push {r11, lr}
+; CHECK-NEXT:    bl __popcountdi2
+; CHECK-NEXT:    asr r1, r0, #31
+; CHECK-NEXT:    pop {r11, lr}
 ; CHECK-NEXT:    mov pc, lr
-; CHECK-NEXT:    .p2align 2
-; CHECK-NEXT:  @ %bb.1:
-; CHECK-NEXT:  .LCPI23_0:
-; CHECK-NEXT:    .long 1431655765 @ 0x55555555
-; CHECK-NEXT:  .LCPI23_1:
-; CHECK-NEXT:    .long 252645135 @ 0xf0f0f0f
-; CHECK-NEXT:  .LCPI23_2:
-; CHECK-NEXT:    .long 16843009 @ 0x1010101
-; CHECK-NEXT:  .LCPI23_3:
-; CHECK-NEXT:    .long 858993459 @ 0x33333333
   %count = tail call i64 @llvm.ctpop.i64(i64 %x)
-  %conv = trunc i64 %count to i32
-  ret i32 %conv
+  ret i64 %count
 }
 
 define i32 @ctpop_eq_one(i64 %x) nounwind readnone {
diff --git a/llvm/test/CodeGen/RISCV/ctlz-cttz-ctpop.ll b/llvm/test/CodeGen/RISCV/ctlz-cttz-ctpop.ll
index a46168f..f8c3a75 100644
--- a/llvm/test/CodeGen/RISCV/ctlz-cttz-ctpop.ll
+++ b/llvm/test/CodeGen/RISCV/ctlz-cttz-ctpop.ll
@@ -1156,46 +1156,30 @@
 }
 
 define i32 @test_ctlz_i32(i32 %a) nounwind {
-; RV32I-LABEL: test_ctlz_i32:
-; RV32I:       # %bb.0:
-; RV32I-NEXT:    beqz a0, .LBB10_2
-; RV32I-NEXT:  # %bb.1: # %cond.false
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    lui a2, 349525
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, 1365
-; RV32I-NEXT:    srli a2, a0, 2
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 4
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 8
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 16
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a2, a0, 1
-; RV32I-NEXT:    and a1, a2, a1
-; RV32I-NEXT:    lui a2, 209715
-; RV32I-NEXT:    addi a2, a2, 819
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a2
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    lui a2, 61681
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, -241
-; RV32I-NEXT:    and a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    ret
-; RV32I-NEXT:  .LBB10_2:
-; RV32I-NEXT:    li a0, 32
-; RV32I-NEXT:    ret
+; RV32_NOZBB-LABEL: test_ctlz_i32:
+; RV32_NOZBB:       # %bb.0:
+; RV32_NOZBB-NEXT:    beqz a0, .LBB10_2
+; RV32_NOZBB-NEXT:  # %bb.1: # %cond.false
+; RV32_NOZBB-NEXT:    addi sp, sp, -16
+; RV32_NOZBB-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32_NOZBB-NEXT:    srli a1, a0, 1
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 2
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 4
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 8
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 16
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    not a0, a0
+; RV32_NOZBB-NEXT:    call __popcountsi2
+; RV32_NOZBB-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32_NOZBB-NEXT:    addi sp, sp, 16
+; RV32_NOZBB-NEXT:    ret
+; RV32_NOZBB-NEXT:  .LBB10_2:
+; RV32_NOZBB-NEXT:    li a0, 32
+; RV32_NOZBB-NEXT:    ret
 ;
 ; RV64I-LABEL: test_ctlz_i32:
 ; RV64I:       # %bb.0:
@@ -1239,46 +1223,6 @@
 ; RV64I-NEXT:    li a0, 32
 ; RV64I-NEXT:    ret
 ;
-; RV32M-LABEL: test_ctlz_i32:
-; RV32M:       # %bb.0:
-; RV32M-NEXT:    beqz a0, .LBB10_2
-; RV32M-NEXT:  # %bb.1: # %cond.false
-; RV32M-NEXT:    srli a1, a0, 1
-; RV32M-NEXT:    lui a2, 349525
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    addi a1, a2, 1365
-; RV32M-NEXT:    srli a2, a0, 2
-; RV32M-NEXT:    or a0, a0, a2
-; RV32M-NEXT:    srli a2, a0, 4
-; RV32M-NEXT:    or a0, a0, a2
-; RV32M-NEXT:    srli a2, a0, 8
-; RV32M-NEXT:    or a0, a0, a2
-; RV32M-NEXT:    srli a2, a0, 16
-; RV32M-NEXT:    or a0, a0, a2
-; RV32M-NEXT:    not a0, a0
-; RV32M-NEXT:    srli a2, a0, 1
-; RV32M-NEXT:    and a1, a2, a1
-; RV32M-NEXT:    lui a2, 209715
-; RV32M-NEXT:    addi a2, a2, 819
-; RV32M-NEXT:    sub a0, a0, a1
-; RV32M-NEXT:    and a1, a0, a2
-; RV32M-NEXT:    srli a0, a0, 2
-; RV32M-NEXT:    and a0, a0, a2
-; RV32M-NEXT:    lui a2, 61681
-; RV32M-NEXT:    add a0, a1, a0
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    add a0, a0, a1
-; RV32M-NEXT:    lui a1, 4112
-; RV32M-NEXT:    addi a2, a2, -241
-; RV32M-NEXT:    and a0, a0, a2
-; RV32M-NEXT:    addi a1, a1, 257
-; RV32M-NEXT:    mul a0, a0, a1
-; RV32M-NEXT:    srli a0, a0, 24
-; RV32M-NEXT:    ret
-; RV32M-NEXT:  .LBB10_2:
-; RV32M-NEXT:    li a0, 32
-; RV32M-NEXT:    ret
-;
 ; RV64M-LABEL: test_ctlz_i32:
 ; RV64M:       # %bb.0:
 ; RV64M-NEXT:    sext.w a1, a0
@@ -1346,240 +1290,75 @@
 }
 
 define i64 @test_ctlz_i64(i64 %a) nounwind {
-; RV32I-LABEL: test_ctlz_i64:
-; RV32I:       # %bb.0:
-; RV32I-NEXT:    lui a2, 349525
-; RV32I-NEXT:    lui a3, 209715
-; RV32I-NEXT:    lui a5, 61681
-; RV32I-NEXT:    addi a4, a2, 1365
-; RV32I-NEXT:    addi a3, a3, 819
-; RV32I-NEXT:    addi a2, a5, -241
-; RV32I-NEXT:    bnez a1, .LBB11_2
-; RV32I-NEXT:  # %bb.1:
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 2
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 8
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 16
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a3
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a3
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    addi a0, a0, 32
-; RV32I-NEXT:    li a1, 0
-; RV32I-NEXT:    ret
-; RV32I-NEXT:  .LBB11_2:
-; RV32I-NEXT:    srli a0, a1, 1
-; RV32I-NEXT:    or a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 2
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 8
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 16
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a3
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a3
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    li a1, 0
-; RV32I-NEXT:    ret
+; RV32_NOZBB-LABEL: test_ctlz_i64:
+; RV32_NOZBB:       # %bb.0:
+; RV32_NOZBB-NEXT:    addi sp, sp, -16
+; RV32_NOZBB-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32_NOZBB-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
+; RV32_NOZBB-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
+; RV32_NOZBB-NEXT:    mv s1, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 1
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 2
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 4
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 8
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 16
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    not a0, a0
+; RV32_NOZBB-NEXT:    call __popcountsi2
+; RV32_NOZBB-NEXT:    mv s0, a0
+; RV32_NOZBB-NEXT:    srli a0, s1, 1
+; RV32_NOZBB-NEXT:    or a0, s1, a0
+; RV32_NOZBB-NEXT:    srli a1, a0, 2
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 4
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 8
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 16
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    not a0, a0
+; RV32_NOZBB-NEXT:    call __popcountsi2
+; RV32_NOZBB-NEXT:    bnez s1, .LBB11_2
+; RV32_NOZBB-NEXT:  # %bb.1:
+; RV32_NOZBB-NEXT:    addi a0, s0, 32
+; RV32_NOZBB-NEXT:  .LBB11_2:
+; RV32_NOZBB-NEXT:    li a1, 0
+; RV32_NOZBB-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32_NOZBB-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
+; RV32_NOZBB-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
+; RV32_NOZBB-NEXT:    addi sp, sp, 16
+; RV32_NOZBB-NEXT:    ret
 ;
-; RV64I-LABEL: test_ctlz_i64:
-; RV64I:       # %bb.0:
-; RV64I-NEXT:    beqz a0, .LBB11_2
-; RV64I-NEXT:  # %bb.1: # %cond.false
-; RV64I-NEXT:    srli a1, a0, 1
-; RV64I-NEXT:    lui a2, 349525
-; RV64I-NEXT:    lui a3, 209715
-; RV64I-NEXT:    or a0, a0, a1
-; RV64I-NEXT:    addiw a1, a2, 1365
-; RV64I-NEXT:    addiw a2, a3, 819
-; RV64I-NEXT:    srli a3, a0, 2
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    slli a3, a1, 32
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    slli a3, a2, 32
-; RV64I-NEXT:    add a2, a2, a3
-; RV64I-NEXT:    srli a3, a0, 4
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 8
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 16
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 32
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    not a0, a0
-; RV64I-NEXT:    srli a3, a0, 1
-; RV64I-NEXT:    and a1, a3, a1
-; RV64I-NEXT:    lui a3, 61681
-; RV64I-NEXT:    addiw a3, a3, -241
-; RV64I-NEXT:    sub a0, a0, a1
-; RV64I-NEXT:    and a1, a0, a2
-; RV64I-NEXT:    srli a0, a0, 2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a2, a3, 32
-; RV64I-NEXT:    add a0, a1, a0
-; RV64I-NEXT:    srli a1, a0, 4
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    add a2, a3, a2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a1, a0, 8
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 16
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 32
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    srli a0, a0, 56
-; RV64I-NEXT:    ret
-; RV64I-NEXT:  .LBB11_2:
-; RV64I-NEXT:    li a0, 64
-; RV64I-NEXT:    ret
-;
-; RV32M-LABEL: test_ctlz_i64:
-; RV32M:       # %bb.0:
-; RV32M-NEXT:    lui a2, 349525
-; RV32M-NEXT:    lui a3, 209715
-; RV32M-NEXT:    lui a6, 61681
-; RV32M-NEXT:    lui a7, 4112
-; RV32M-NEXT:    addi a5, a2, 1365
-; RV32M-NEXT:    addi a4, a3, 819
-; RV32M-NEXT:    addi a3, a6, -241
-; RV32M-NEXT:    addi a2, a7, 257
-; RV32M-NEXT:    bnez a1, .LBB11_2
-; RV32M-NEXT:  # %bb.1:
-; RV32M-NEXT:    srli a1, a0, 1
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 2
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 8
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 16
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    not a0, a0
-; RV32M-NEXT:    srli a1, a0, 1
-; RV32M-NEXT:    and a1, a1, a5
-; RV32M-NEXT:    sub a0, a0, a1
-; RV32M-NEXT:    and a1, a0, a4
-; RV32M-NEXT:    srli a0, a0, 2
-; RV32M-NEXT:    and a0, a0, a4
-; RV32M-NEXT:    add a0, a1, a0
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    add a0, a0, a1
-; RV32M-NEXT:    and a0, a0, a3
-; RV32M-NEXT:    mul a0, a0, a2
-; RV32M-NEXT:    srli a0, a0, 24
-; RV32M-NEXT:    addi a0, a0, 32
-; RV32M-NEXT:    li a1, 0
-; RV32M-NEXT:    ret
-; RV32M-NEXT:  .LBB11_2:
-; RV32M-NEXT:    srli a0, a1, 1
-; RV32M-NEXT:    or a0, a1, a0
-; RV32M-NEXT:    srli a1, a0, 2
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 8
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 16
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    not a0, a0
-; RV32M-NEXT:    srli a1, a0, 1
-; RV32M-NEXT:    and a1, a1, a5
-; RV32M-NEXT:    sub a0, a0, a1
-; RV32M-NEXT:    and a1, a0, a4
-; RV32M-NEXT:    srli a0, a0, 2
-; RV32M-NEXT:    and a0, a0, a4
-; RV32M-NEXT:    add a0, a1, a0
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    add a0, a0, a1
-; RV32M-NEXT:    and a0, a0, a3
-; RV32M-NEXT:    mul a0, a0, a2
-; RV32M-NEXT:    srli a0, a0, 24
-; RV32M-NEXT:    li a1, 0
-; RV32M-NEXT:    ret
-;
-; RV64M-LABEL: test_ctlz_i64:
-; RV64M:       # %bb.0:
-; RV64M-NEXT:    beqz a0, .LBB11_2
-; RV64M-NEXT:  # %bb.1: # %cond.false
-; RV64M-NEXT:    srli a1, a0, 1
-; RV64M-NEXT:    lui a2, 349525
-; RV64M-NEXT:    lui a3, 209715
-; RV64M-NEXT:    lui a4, 61681
-; RV64M-NEXT:    or a0, a0, a1
-; RV64M-NEXT:    addiw a1, a2, 1365
-; RV64M-NEXT:    addiw a2, a3, 819
-; RV64M-NEXT:    addiw a3, a4, -241
-; RV64M-NEXT:    srli a4, a0, 2
-; RV64M-NEXT:    or a0, a0, a4
-; RV64M-NEXT:    slli a4, a1, 32
-; RV64M-NEXT:    add a1, a1, a4
-; RV64M-NEXT:    slli a4, a2, 32
-; RV64M-NEXT:    add a2, a2, a4
-; RV64M-NEXT:    slli a4, a3, 32
-; RV64M-NEXT:    add a3, a3, a4
-; RV64M-NEXT:    srli a4, a0, 4
-; RV64M-NEXT:    or a0, a0, a4
-; RV64M-NEXT:    srli a4, a0, 8
-; RV64M-NEXT:    or a0, a0, a4
-; RV64M-NEXT:    srli a4, a0, 16
-; RV64M-NEXT:    or a0, a0, a4
-; RV64M-NEXT:    srli a4, a0, 32
-; RV64M-NEXT:    or a0, a0, a4
-; RV64M-NEXT:    not a0, a0
-; RV64M-NEXT:    srli a4, a0, 1
-; RV64M-NEXT:    and a1, a4, a1
-; RV64M-NEXT:    sub a0, a0, a1
-; RV64M-NEXT:    and a1, a0, a2
-; RV64M-NEXT:    srli a0, a0, 2
-; RV64M-NEXT:    and a0, a0, a2
-; RV64M-NEXT:    lui a2, 4112
-; RV64M-NEXT:    addiw a2, a2, 257
-; RV64M-NEXT:    add a0, a1, a0
-; RV64M-NEXT:    srli a1, a0, 4
-; RV64M-NEXT:    add a0, a0, a1
-; RV64M-NEXT:    slli a1, a2, 32
-; RV64M-NEXT:    and a0, a0, a3
-; RV64M-NEXT:    add a1, a2, a1
-; RV64M-NEXT:    mul a0, a0, a1
-; RV64M-NEXT:    srli a0, a0, 56
-; RV64M-NEXT:    ret
-; RV64M-NEXT:  .LBB11_2:
-; RV64M-NEXT:    li a0, 64
-; RV64M-NEXT:    ret
+; RV64NOZBB-LABEL: test_ctlz_i64:
+; RV64NOZBB:       # %bb.0:
+; RV64NOZBB-NEXT:    beqz a0, .LBB11_2
+; RV64NOZBB-NEXT:  # %bb.1: # %cond.false
+; RV64NOZBB-NEXT:    addi sp, sp, -16
+; RV64NOZBB-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
+; RV64NOZBB-NEXT:    srli a1, a0, 1
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    srli a1, a0, 2
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    srli a1, a0, 4
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    srli a1, a0, 8
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    srli a1, a0, 16
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    srli a1, a0, 32
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    not a0, a0
+; RV64NOZBB-NEXT:    call __popcountdi2
+; RV64NOZBB-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; RV64NOZBB-NEXT:    addi sp, sp, 16
+; RV64NOZBB-NEXT:    ret
+; RV64NOZBB-NEXT:  .LBB11_2:
+; RV64NOZBB-NEXT:    li a0, 64
+; RV64NOZBB-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: test_ctlz_i64:
 ; RV32ZBB:       # %bb.0:
@@ -1793,41 +1572,20 @@
 }
 
 define i32 @test_ctlz_i32_zero_undef(i32 %a) nounwind {
-; RV32I-LABEL: test_ctlz_i32_zero_undef:
-; RV32I:       # %bb.0:
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    lui a2, 349525
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, 1365
-; RV32I-NEXT:    srli a2, a0, 2
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 4
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 8
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 16
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a2, a0, 1
-; RV32I-NEXT:    and a1, a2, a1
-; RV32I-NEXT:    lui a2, 209715
-; RV32I-NEXT:    addi a2, a2, 819
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a2
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    lui a2, 61681
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, -241
-; RV32I-NEXT:    and a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    ret
+; RV32_NOZBB-LABEL: test_ctlz_i32_zero_undef:
+; RV32_NOZBB:       # %bb.0:
+; RV32_NOZBB-NEXT:    srli a1, a0, 1
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 2
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 4
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 8
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 16
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    not a0, a0
+; RV32_NOZBB-NEXT:    tail __popcountsi2
 ;
 ; RV64I-LABEL: test_ctlz_i32_zero_undef:
 ; RV64I:       # %bb.0:
@@ -1865,41 +1623,6 @@
 ; RV64I-NEXT:    srliw a0, a0, 24
 ; RV64I-NEXT:    ret
 ;
-; RV32M-LABEL: test_ctlz_i32_zero_undef:
-; RV32M:       # %bb.0:
-; RV32M-NEXT:    srli a1, a0, 1
-; RV32M-NEXT:    lui a2, 349525
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    addi a1, a2, 1365
-; RV32M-NEXT:    srli a2, a0, 2
-; RV32M-NEXT:    or a0, a0, a2
-; RV32M-NEXT:    srli a2, a0, 4
-; RV32M-NEXT:    or a0, a0, a2
-; RV32M-NEXT:    srli a2, a0, 8
-; RV32M-NEXT:    or a0, a0, a2
-; RV32M-NEXT:    srli a2, a0, 16
-; RV32M-NEXT:    or a0, a0, a2
-; RV32M-NEXT:    not a0, a0
-; RV32M-NEXT:    srli a2, a0, 1
-; RV32M-NEXT:    and a1, a2, a1
-; RV32M-NEXT:    lui a2, 209715
-; RV32M-NEXT:    addi a2, a2, 819
-; RV32M-NEXT:    sub a0, a0, a1
-; RV32M-NEXT:    and a1, a0, a2
-; RV32M-NEXT:    srli a0, a0, 2
-; RV32M-NEXT:    and a0, a0, a2
-; RV32M-NEXT:    lui a2, 61681
-; RV32M-NEXT:    add a0, a1, a0
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    add a0, a0, a1
-; RV32M-NEXT:    lui a1, 4112
-; RV32M-NEXT:    addi a2, a2, -241
-; RV32M-NEXT:    and a0, a0, a2
-; RV32M-NEXT:    addi a1, a1, 257
-; RV32M-NEXT:    mul a0, a0, a1
-; RV32M-NEXT:    srli a0, a0, 24
-; RV32M-NEXT:    ret
-;
 ; RV64M-LABEL: test_ctlz_i32_zero_undef:
 ; RV64M:       # %bb.0:
 ; RV64M-NEXT:    srliw a1, a0, 1
@@ -1961,230 +1684,70 @@
 }
 
 define i64 @test_ctlz_i64_zero_undef(i64 %a) nounwind {
-; RV32I-LABEL: test_ctlz_i64_zero_undef:
-; RV32I:       # %bb.0:
-; RV32I-NEXT:    lui a2, 349525
-; RV32I-NEXT:    lui a3, 209715
-; RV32I-NEXT:    lui a5, 61681
-; RV32I-NEXT:    addi a4, a2, 1365
-; RV32I-NEXT:    addi a3, a3, 819
-; RV32I-NEXT:    addi a2, a5, -241
-; RV32I-NEXT:    bnez a1, .LBB15_2
-; RV32I-NEXT:  # %bb.1:
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 2
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 8
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 16
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a3
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a3
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    addi a0, a0, 32
-; RV32I-NEXT:    li a1, 0
-; RV32I-NEXT:    ret
-; RV32I-NEXT:  .LBB15_2:
-; RV32I-NEXT:    srli a0, a1, 1
-; RV32I-NEXT:    or a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 2
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 8
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 16
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a3
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a3
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    li a1, 0
-; RV32I-NEXT:    ret
+; RV32_NOZBB-LABEL: test_ctlz_i64_zero_undef:
+; RV32_NOZBB:       # %bb.0:
+; RV32_NOZBB-NEXT:    addi sp, sp, -16
+; RV32_NOZBB-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32_NOZBB-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
+; RV32_NOZBB-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
+; RV32_NOZBB-NEXT:    mv s1, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 1
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 2
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 4
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 8
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 16
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    not a0, a0
+; RV32_NOZBB-NEXT:    call __popcountsi2
+; RV32_NOZBB-NEXT:    mv s0, a0
+; RV32_NOZBB-NEXT:    srli a0, s1, 1
+; RV32_NOZBB-NEXT:    or a0, s1, a0
+; RV32_NOZBB-NEXT:    srli a1, a0, 2
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 4
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 8
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    srli a1, a0, 16
+; RV32_NOZBB-NEXT:    or a0, a0, a1
+; RV32_NOZBB-NEXT:    not a0, a0
+; RV32_NOZBB-NEXT:    call __popcountsi2
+; RV32_NOZBB-NEXT:    bnez s1, .LBB15_2
+; RV32_NOZBB-NEXT:  # %bb.1:
+; RV32_NOZBB-NEXT:    addi a0, s0, 32
+; RV32_NOZBB-NEXT:  .LBB15_2:
+; RV32_NOZBB-NEXT:    li a1, 0
+; RV32_NOZBB-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32_NOZBB-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
+; RV32_NOZBB-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
+; RV32_NOZBB-NEXT:    addi sp, sp, 16
+; RV32_NOZBB-NEXT:    ret
 ;
-; RV64I-LABEL: test_ctlz_i64_zero_undef:
-; RV64I:       # %bb.0:
-; RV64I-NEXT:    srli a1, a0, 1
-; RV64I-NEXT:    lui a2, 349525
-; RV64I-NEXT:    lui a3, 209715
-; RV64I-NEXT:    or a0, a0, a1
-; RV64I-NEXT:    addiw a1, a2, 1365
-; RV64I-NEXT:    addiw a2, a3, 819
-; RV64I-NEXT:    srli a3, a0, 2
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    slli a3, a1, 32
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    slli a3, a2, 32
-; RV64I-NEXT:    add a2, a2, a3
-; RV64I-NEXT:    srli a3, a0, 4
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 8
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 16
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 32
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    not a0, a0
-; RV64I-NEXT:    srli a3, a0, 1
-; RV64I-NEXT:    and a1, a3, a1
-; RV64I-NEXT:    lui a3, 61681
-; RV64I-NEXT:    addiw a3, a3, -241
-; RV64I-NEXT:    sub a0, a0, a1
-; RV64I-NEXT:    and a1, a0, a2
-; RV64I-NEXT:    srli a0, a0, 2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a2, a3, 32
-; RV64I-NEXT:    add a0, a1, a0
-; RV64I-NEXT:    srli a1, a0, 4
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    add a2, a3, a2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a1, a0, 8
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 16
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 32
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    srli a0, a0, 56
-; RV64I-NEXT:    ret
-;
-; RV32M-LABEL: test_ctlz_i64_zero_undef:
-; RV32M:       # %bb.0:
-; RV32M-NEXT:    lui a2, 349525
-; RV32M-NEXT:    lui a3, 209715
-; RV32M-NEXT:    lui a6, 61681
-; RV32M-NEXT:    lui a7, 4112
-; RV32M-NEXT:    addi a5, a2, 1365
-; RV32M-NEXT:    addi a4, a3, 819
-; RV32M-NEXT:    addi a3, a6, -241
-; RV32M-NEXT:    addi a2, a7, 257
-; RV32M-NEXT:    bnez a1, .LBB15_2
-; RV32M-NEXT:  # %bb.1:
-; RV32M-NEXT:    srli a1, a0, 1
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 2
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 8
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 16
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    not a0, a0
-; RV32M-NEXT:    srli a1, a0, 1
-; RV32M-NEXT:    and a1, a1, a5
-; RV32M-NEXT:    sub a0, a0, a1
-; RV32M-NEXT:    and a1, a0, a4
-; RV32M-NEXT:    srli a0, a0, 2
-; RV32M-NEXT:    and a0, a0, a4
-; RV32M-NEXT:    add a0, a1, a0
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    add a0, a0, a1
-; RV32M-NEXT:    and a0, a0, a3
-; RV32M-NEXT:    mul a0, a0, a2
-; RV32M-NEXT:    srli a0, a0, 24
-; RV32M-NEXT:    addi a0, a0, 32
-; RV32M-NEXT:    li a1, 0
-; RV32M-NEXT:    ret
-; RV32M-NEXT:  .LBB15_2:
-; RV32M-NEXT:    srli a0, a1, 1
-; RV32M-NEXT:    or a0, a1, a0
-; RV32M-NEXT:    srli a1, a0, 2
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 8
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    srli a1, a0, 16
-; RV32M-NEXT:    or a0, a0, a1
-; RV32M-NEXT:    not a0, a0
-; RV32M-NEXT:    srli a1, a0, 1
-; RV32M-NEXT:    and a1, a1, a5
-; RV32M-NEXT:    sub a0, a0, a1
-; RV32M-NEXT:    and a1, a0, a4
-; RV32M-NEXT:    srli a0, a0, 2
-; RV32M-NEXT:    and a0, a0, a4
-; RV32M-NEXT:    add a0, a1, a0
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    add a0, a0, a1
-; RV32M-NEXT:    and a0, a0, a3
-; RV32M-NEXT:    mul a0, a0, a2
-; RV32M-NEXT:    srli a0, a0, 24
-; RV32M-NEXT:    li a1, 0
-; RV32M-NEXT:    ret
-;
-; RV64M-LABEL: test_ctlz_i64_zero_undef:
-; RV64M:       # %bb.0:
-; RV64M-NEXT:    srli a1, a0, 1
-; RV64M-NEXT:    lui a2, 349525
-; RV64M-NEXT:    lui a3, 209715
-; RV64M-NEXT:    lui a4, 61681
-; RV64M-NEXT:    or a0, a0, a1
-; RV64M-NEXT:    addiw a1, a2, 1365
-; RV64M-NEXT:    addiw a2, a3, 819
-; RV64M-NEXT:    addiw a3, a4, -241
-; RV64M-NEXT:    srli a4, a0, 2
-; RV64M-NEXT:    or a0, a0, a4
-; RV64M-NEXT:    slli a4, a1, 32
-; RV64M-NEXT:    add a1, a1, a4
-; RV64M-NEXT:    slli a4, a2, 32
-; RV64M-NEXT:    add a2, a2, a4
-; RV64M-NEXT:    slli a4, a3, 32
-; RV64M-NEXT:    add a3, a3, a4
-; RV64M-NEXT:    srli a4, a0, 4
-; RV64M-NEXT:    or a0, a0, a4
-; RV64M-NEXT:    srli a4, a0, 8
-; RV64M-NEXT:    or a0, a0, a4
-; RV64M-NEXT:    srli a4, a0, 16
-; RV64M-NEXT:    or a0, a0, a4
-; RV64M-NEXT:    srli a4, a0, 32
-; RV64M-NEXT:    or a0, a0, a4
-; RV64M-NEXT:    not a0, a0
-; RV64M-NEXT:    srli a4, a0, 1
-; RV64M-NEXT:    and a1, a4, a1
-; RV64M-NEXT:    sub a0, a0, a1
-; RV64M-NEXT:    and a1, a0, a2
-; RV64M-NEXT:    srli a0, a0, 2
-; RV64M-NEXT:    and a0, a0, a2
-; RV64M-NEXT:    lui a2, 4112
-; RV64M-NEXT:    addiw a2, a2, 257
-; RV64M-NEXT:    add a0, a1, a0
-; RV64M-NEXT:    srli a1, a0, 4
-; RV64M-NEXT:    add a0, a0, a1
-; RV64M-NEXT:    slli a1, a2, 32
-; RV64M-NEXT:    and a0, a0, a3
-; RV64M-NEXT:    add a1, a2, a1
-; RV64M-NEXT:    mul a0, a0, a1
-; RV64M-NEXT:    srli a0, a0, 56
-; RV64M-NEXT:    ret
+; RV64NOZBB-LABEL: test_ctlz_i64_zero_undef:
+; RV64NOZBB:       # %bb.0:
+; RV64NOZBB-NEXT:    addi sp, sp, -16
+; RV64NOZBB-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
+; RV64NOZBB-NEXT:    srli a1, a0, 1
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    srli a1, a0, 2
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    srli a1, a0, 4
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    srli a1, a0, 8
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    srli a1, a0, 16
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    srli a1, a0, 32
+; RV64NOZBB-NEXT:    or a0, a0, a1
+; RV64NOZBB-NEXT:    not a0, a0
+; RV64NOZBB-NEXT:    call __popcountdi2
+; RV64NOZBB-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; RV64NOZBB-NEXT:    addi sp, sp, 16
+; RV64NOZBB-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: test_ctlz_i64_zero_undef:
 ; RV32ZBB:       # %bb.0:
@@ -2396,30 +1959,9 @@
 }
 
 define i32 @test_ctpop_i32(i32 %a) nounwind {
-; RV32I-LABEL: test_ctpop_i32:
-; RV32I:       # %bb.0:
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    lui a2, 349525
-; RV32I-NEXT:    addi a2, a2, 1365
-; RV32I-NEXT:    and a1, a1, a2
-; RV32I-NEXT:    lui a2, 209715
-; RV32I-NEXT:    addi a2, a2, 819
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a2
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    lui a2, 61681
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, -241
-; RV32I-NEXT:    and a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    ret
+; RV32_NOZBB-LABEL: test_ctpop_i32:
+; RV32_NOZBB:       # %bb.0:
+; RV32_NOZBB-NEXT:    tail __popcountsi2
 ;
 ; RV64I-LABEL: test_ctpop_i32:
 ; RV64I:       # %bb.0:
@@ -2446,30 +1988,6 @@
 ; RV64I-NEXT:    srliw a0, a0, 24
 ; RV64I-NEXT:    ret
 ;
-; RV32M-LABEL: test_ctpop_i32:
-; RV32M:       # %bb.0:
-; RV32M-NEXT:    srli a1, a0, 1
-; RV32M-NEXT:    lui a2, 349525
-; RV32M-NEXT:    addi a2, a2, 1365
-; RV32M-NEXT:    and a1, a1, a2
-; RV32M-NEXT:    lui a2, 209715
-; RV32M-NEXT:    addi a2, a2, 819
-; RV32M-NEXT:    sub a0, a0, a1
-; RV32M-NEXT:    and a1, a0, a2
-; RV32M-NEXT:    srli a0, a0, 2
-; RV32M-NEXT:    and a0, a0, a2
-; RV32M-NEXT:    lui a2, 61681
-; RV32M-NEXT:    add a0, a1, a0
-; RV32M-NEXT:    srli a1, a0, 4
-; RV32M-NEXT:    add a0, a0, a1
-; RV32M-NEXT:    lui a1, 4112
-; RV32M-NEXT:    addi a2, a2, -241
-; RV32M-NEXT:    and a0, a0, a2
-; RV32M-NEXT:    addi a1, a1, 257
-; RV32M-NEXT:    mul a0, a0, a1
-; RV32M-NEXT:    srli a0, a0, 24
-; RV32M-NEXT:    ret
-;
 ; RV64M-LABEL: test_ctpop_i32:
 ; RV64M:       # %bb.0:
 ; RV64M-NEXT:    srli a1, a0, 1
@@ -2506,28 +2024,7 @@
 ;
 ; RV32XTHEADBB-LABEL: test_ctpop_i32:
 ; RV32XTHEADBB:       # %bb.0:
-; RV32XTHEADBB-NEXT:    srli a1, a0, 1
-; RV32XTHEADBB-NEXT:    lui a2, 349525
-; RV32XTHEADBB-NEXT:    addi a2, a2, 1365
-; RV32XTHEADBB-NEXT:    and a1, a1, a2
-; RV32XTHEADBB-NEXT:    lui a2, 209715
-; RV32XTHEADBB-NEXT:    addi a2, a2, 819
-; RV32XTHEADBB-NEXT:    sub a0, a0, a1
-; RV32XTHEADBB-NEXT:    and a1, a0, a2
-; RV32XTHEADBB-NEXT:    srli a0, a0, 2
-; RV32XTHEADBB-NEXT:    and a0, a0, a2
-; RV32XTHEADBB-NEXT:    lui a2, 61681
-; RV32XTHEADBB-NEXT:    add a0, a1, a0
-; RV32XTHEADBB-NEXT:    srli a1, a0, 4
-; RV32XTHEADBB-NEXT:    add a0, a0, a1
-; RV32XTHEADBB-NEXT:    addi a1, a2, -241
-; RV32XTHEADBB-NEXT:    and a0, a0, a1
-; RV32XTHEADBB-NEXT:    slli a1, a0, 8
-; RV32XTHEADBB-NEXT:    add a0, a0, a1
-; RV32XTHEADBB-NEXT:    slli a1, a0, 16
-; RV32XTHEADBB-NEXT:    add a0, a0, a1
-; RV32XTHEADBB-NEXT:    srli a0, a0, 24
-; RV32XTHEADBB-NEXT:    ret
+; RV32XTHEADBB-NEXT:    tail __popcountsi2
 ;
 ; RV64XTHEADBB-LABEL: test_ctpop_i32:
 ; RV64XTHEADBB:       # %bb.0:
@@ -2558,150 +2055,24 @@
 }
 
 define i64 @test_ctpop_i64(i64 %a) nounwind {
-; RV32I-LABEL: test_ctpop_i64:
-; RV32I:       # %bb.0:
-; RV32I-NEXT:    srli a2, a1, 1
-; RV32I-NEXT:    lui a3, 349525
-; RV32I-NEXT:    lui a4, 209715
-; RV32I-NEXT:    srli a5, a0, 1
-; RV32I-NEXT:    addi a3, a3, 1365
-; RV32I-NEXT:    and a2, a2, a3
-; RV32I-NEXT:    and a3, a5, a3
-; RV32I-NEXT:    lui a5, 61681
-; RV32I-NEXT:    addi a4, a4, 819
-; RV32I-NEXT:    addi a5, a5, -241
-; RV32I-NEXT:    sub a1, a1, a2
-; RV32I-NEXT:    sub a0, a0, a3
-; RV32I-NEXT:    and a2, a1, a4
-; RV32I-NEXT:    srli a1, a1, 2
-; RV32I-NEXT:    and a3, a0, a4
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    and a0, a0, a4
-; RV32I-NEXT:    add a1, a2, a1
-; RV32I-NEXT:    add a0, a3, a0
-; RV32I-NEXT:    srli a2, a1, 4
-; RV32I-NEXT:    srli a3, a0, 4
-; RV32I-NEXT:    add a1, a1, a2
-; RV32I-NEXT:    add a0, a0, a3
-; RV32I-NEXT:    and a1, a1, a5
-; RV32I-NEXT:    and a0, a0, a5
-; RV32I-NEXT:    slli a2, a1, 8
-; RV32I-NEXT:    slli a3, a0, 8
-; RV32I-NEXT:    add a1, a1, a2
-; RV32I-NEXT:    add a0, a0, a3
-; RV32I-NEXT:    slli a2, a1, 16
-; RV32I-NEXT:    slli a3, a0, 16
-; RV32I-NEXT:    add a1, a1, a2
-; RV32I-NEXT:    add a0, a0, a3
-; RV32I-NEXT:    srli a1, a1, 24
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    li a1, 0
-; RV32I-NEXT:    ret
+; RV32_NOZBB-LABEL: test_ctpop_i64:
+; RV32_NOZBB:       # %bb.0:
+; RV32_NOZBB-NEXT:    addi sp, sp, -16
+; RV32_NOZBB-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32_NOZBB-NEXT:    call __popcountdi2
+; RV32_NOZBB-NEXT:    srai a1, a0, 31
+; RV32_NOZBB-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32_NOZBB-NEXT:    addi sp, sp, 16
+; RV32_NOZBB-NEXT:    ret
 ;
-; RV64I-LABEL: test_ctpop_i64:
-; RV64I:       # %bb.0:
-; RV64I-NEXT:    lui a1, 349525
-; RV64I-NEXT:    lui a2, 209715
-; RV64I-NEXT:    addiw a1, a1, 1365
-; RV64I-NEXT:    addiw a2, a2, 819
-; RV64I-NEXT:    slli a3, a1, 32
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    slli a3, a2, 32
-; RV64I-NEXT:    add a2, a2, a3
-; RV64I-NEXT:    srli a3, a0, 1
-; RV64I-NEXT:    and a1, a3, a1
-; RV64I-NEXT:    lui a3, 61681
-; RV64I-NEXT:    addiw a3, a3, -241
-; RV64I-NEXT:    sub a0, a0, a1
-; RV64I-NEXT:    and a1, a0, a2
-; RV64I-NEXT:    srli a0, a0, 2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a2, a3, 32
-; RV64I-NEXT:    add a0, a1, a0
-; RV64I-NEXT:    srli a1, a0, 4
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    add a2, a3, a2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a1, a0, 8
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 16
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 32
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    srli a0, a0, 56
-; RV64I-NEXT:    ret
-;
-; RV32M-LABEL: test_ctpop_i64:
-; RV32M:       # %bb.0:
-; RV32M-NEXT:    srli a2, a1, 1
-; RV32M-NEXT:    lui a3, 349525
-; RV32M-NEXT:    lui a4, 209715
-; RV32M-NEXT:    lui a5, 61681
-; RV32M-NEXT:    srli a6, a0, 1
-; RV32M-NEXT:    addi a3, a3, 1365
-; RV32M-NEXT:    and a2, a2, a3
-; RV32M-NEXT:    and a3, a6, a3
-; RV32M-NEXT:    lui a6, 4112
-; RV32M-NEXT:    addi a4, a4, 819
-; RV32M-NEXT:    addi a5, a5, -241
-; RV32M-NEXT:    addi a6, a6, 257
-; RV32M-NEXT:    sub a1, a1, a2
-; RV32M-NEXT:    sub a0, a0, a3
-; RV32M-NEXT:    and a2, a1, a4
-; RV32M-NEXT:    srli a1, a1, 2
-; RV32M-NEXT:    and a3, a0, a4
-; RV32M-NEXT:    srli a0, a0, 2
-; RV32M-NEXT:    and a1, a1, a4
-; RV32M-NEXT:    and a0, a0, a4
-; RV32M-NEXT:    add a1, a2, a1
-; RV32M-NEXT:    add a0, a3, a0
-; RV32M-NEXT:    srli a2, a1, 4
-; RV32M-NEXT:    srli a3, a0, 4
-; RV32M-NEXT:    add a1, a1, a2
-; RV32M-NEXT:    add a0, a0, a3
-; RV32M-NEXT:    and a1, a1, a5
-; RV32M-NEXT:    and a0, a0, a5
-; RV32M-NEXT:    mul a1, a1, a6
-; RV32M-NEXT:    mul a0, a0, a6
-; RV32M-NEXT:    srli a1, a1, 24
-; RV32M-NEXT:    srli a0, a0, 24
-; RV32M-NEXT:    add a0, a0, a1
-; RV32M-NEXT:    li a1, 0
-; RV32M-NEXT:    ret
-;
-; RV64M-LABEL: test_ctpop_i64:
-; RV64M:       # %bb.0:
-; RV64M-NEXT:    lui a1, 349525
-; RV64M-NEXT:    lui a2, 209715
-; RV64M-NEXT:    lui a3, 61681
-; RV64M-NEXT:    addiw a1, a1, 1365
-; RV64M-NEXT:    addiw a2, a2, 819
-; RV64M-NEXT:    addiw a3, a3, -241
-; RV64M-NEXT:    slli a4, a1, 32
-; RV64M-NEXT:    add a1, a1, a4
-; RV64M-NEXT:    slli a4, a2, 32
-; RV64M-NEXT:    add a2, a2, a4
-; RV64M-NEXT:    slli a4, a3, 32
-; RV64M-NEXT:    add a3, a3, a4
-; RV64M-NEXT:    srli a4, a0, 1
-; RV64M-NEXT:    and a1, a4, a1
-; RV64M-NEXT:    sub a0, a0, a1
-; RV64M-NEXT:    and a1, a0, a2
-; RV64M-NEXT:    srli a0, a0, 2
-; RV64M-NEXT:    and a0, a0, a2
-; RV64M-NEXT:    lui a2, 4112
-; RV64M-NEXT:    addiw a2, a2, 257
-; RV64M-NEXT:    add a0, a1, a0
-; RV64M-NEXT:    srli a1, a0, 4
-; RV64M-NEXT:    add a0, a0, a1
-; RV64M-NEXT:    slli a1, a2, 32
-; RV64M-NEXT:    and a0, a0, a3
-; RV64M-NEXT:    add a1, a2, a1
-; RV64M-NEXT:    mul a0, a0, a1
-; RV64M-NEXT:    srli a0, a0, 56
-; RV64M-NEXT:    ret
+; RV64NOZBB-LABEL: test_ctpop_i64:
+; RV64NOZBB:       # %bb.0:
+; RV64NOZBB-NEXT:    addi sp, sp, -16
+; RV64NOZBB-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
+; RV64NOZBB-NEXT:    call __popcountdi2
+; RV64NOZBB-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; RV64NOZBB-NEXT:    addi sp, sp, 16
+; RV64NOZBB-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: test_ctpop_i64:
 ; RV32ZBB:       # %bb.0:
@@ -2718,77 +2089,21 @@
 ;
 ; RV32XTHEADBB-LABEL: test_ctpop_i64:
 ; RV32XTHEADBB:       # %bb.0:
-; RV32XTHEADBB-NEXT:    srli a2, a1, 1
-; RV32XTHEADBB-NEXT:    lui a3, 349525
-; RV32XTHEADBB-NEXT:    lui a4, 209715
-; RV32XTHEADBB-NEXT:    srli a5, a0, 1
-; RV32XTHEADBB-NEXT:    addi a3, a3, 1365
-; RV32XTHEADBB-NEXT:    and a2, a2, a3
-; RV32XTHEADBB-NEXT:    and a3, a5, a3
-; RV32XTHEADBB-NEXT:    lui a5, 61681
-; RV32XTHEADBB-NEXT:    addi a4, a4, 819
-; RV32XTHEADBB-NEXT:    addi a5, a5, -241
-; RV32XTHEADBB-NEXT:    sub a1, a1, a2
-; RV32XTHEADBB-NEXT:    sub a0, a0, a3
-; RV32XTHEADBB-NEXT:    and a2, a1, a4
-; RV32XTHEADBB-NEXT:    srli a1, a1, 2
-; RV32XTHEADBB-NEXT:    and a3, a0, a4
-; RV32XTHEADBB-NEXT:    srli a0, a0, 2
-; RV32XTHEADBB-NEXT:    and a1, a1, a4
-; RV32XTHEADBB-NEXT:    and a0, a0, a4
-; RV32XTHEADBB-NEXT:    add a1, a2, a1
-; RV32XTHEADBB-NEXT:    add a0, a3, a0
-; RV32XTHEADBB-NEXT:    srli a2, a1, 4
-; RV32XTHEADBB-NEXT:    srli a3, a0, 4
-; RV32XTHEADBB-NEXT:    add a1, a1, a2
-; RV32XTHEADBB-NEXT:    add a0, a0, a3
-; RV32XTHEADBB-NEXT:    and a1, a1, a5
-; RV32XTHEADBB-NEXT:    and a0, a0, a5
-; RV32XTHEADBB-NEXT:    slli a2, a1, 8
-; RV32XTHEADBB-NEXT:    slli a3, a0, 8
-; RV32XTHEADBB-NEXT:    add a1, a1, a2
-; RV32XTHEADBB-NEXT:    add a0, a0, a3
-; RV32XTHEADBB-NEXT:    slli a2, a1, 16
-; RV32XTHEADBB-NEXT:    slli a3, a0, 16
-; RV32XTHEADBB-NEXT:    add a1, a1, a2
-; RV32XTHEADBB-NEXT:    add a0, a0, a3
-; RV32XTHEADBB-NEXT:    srli a1, a1, 24
-; RV32XTHEADBB-NEXT:    srli a0, a0, 24
-; RV32XTHEADBB-NEXT:    add a0, a0, a1
-; RV32XTHEADBB-NEXT:    li a1, 0
+; RV32XTHEADBB-NEXT:    addi sp, sp, -16
+; RV32XTHEADBB-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32XTHEADBB-NEXT:    call __popcountdi2
+; RV32XTHEADBB-NEXT:    srai a1, a0, 31
+; RV32XTHEADBB-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32XTHEADBB-NEXT:    addi sp, sp, 16
 ; RV32XTHEADBB-NEXT:    ret
 ;
 ; RV64XTHEADBB-LABEL: test_ctpop_i64:
 ; RV64XTHEADBB:       # %bb.0:
-; RV64XTHEADBB-NEXT:    lui a1, 349525
-; RV64XTHEADBB-NEXT:    lui a2, 209715
-; RV64XTHEADBB-NEXT:    addiw a1, a1, 1365
-; RV64XTHEADBB-NEXT:    addiw a2, a2, 819
-; RV64XTHEADBB-NEXT:    slli a3, a1, 32
-; RV64XTHEADBB-NEXT:    add a1, a1, a3
-; RV64XTHEADBB-NEXT:    slli a3, a2, 32
-; RV64XTHEADBB-NEXT:    add a2, a2, a3
-; RV64XTHEADBB-NEXT:    srli a3, a0, 1
-; RV64XTHEADBB-NEXT:    and a1, a3, a1
-; RV64XTHEADBB-NEXT:    lui a3, 61681
-; RV64XTHEADBB-NEXT:    addiw a3, a3, -241
-; RV64XTHEADBB-NEXT:    sub a0, a0, a1
-; RV64XTHEADBB-NEXT:    and a1, a0, a2
-; RV64XTHEADBB-NEXT:    srli a0, a0, 2
-; RV64XTHEADBB-NEXT:    and a0, a0, a2
-; RV64XTHEADBB-NEXT:    slli a2, a3, 32
-; RV64XTHEADBB-NEXT:    add a0, a1, a0
-; RV64XTHEADBB-NEXT:    srli a1, a0, 4
-; RV64XTHEADBB-NEXT:    add a0, a0, a1
-; RV64XTHEADBB-NEXT:    add a2, a3, a2
-; RV64XTHEADBB-NEXT:    and a0, a0, a2
-; RV64XTHEADBB-NEXT:    slli a1, a0, 8
-; RV64XTHEADBB-NEXT:    add a0, a0, a1
-; RV64XTHEADBB-NEXT:    slli a1, a0, 16
-; RV64XTHEADBB-NEXT:    add a0, a0, a1
-; RV64XTHEADBB-NEXT:    slli a1, a0, 32
-; RV64XTHEADBB-NEXT:    add a0, a0, a1
-; RV64XTHEADBB-NEXT:    srli a0, a0, 56
+; RV64XTHEADBB-NEXT:    addi sp, sp, -16
+; RV64XTHEADBB-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
+; RV64XTHEADBB-NEXT:    call __popcountdi2
+; RV64XTHEADBB-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; RV64XTHEADBB-NEXT:    addi sp, sp, 16
 ; RV64XTHEADBB-NEXT:    ret
   %1 = call i64 @llvm.ctpop.i64(i64 %a)
   ret i64 %1
diff --git a/llvm/test/CodeGen/RISCV/ctz_zero_return_test.ll b/llvm/test/CodeGen/RISCV/ctz_zero_return_test.ll
index 03a6a6b..d57c4d6 100644
--- a/llvm/test/CodeGen/RISCV/ctz_zero_return_test.ll
+++ b/llvm/test/CodeGen/RISCV/ctz_zero_return_test.ll
@@ -602,118 +602,68 @@
 ;
 ; RV32I-LABEL: ctlz:
 ; RV32I:       # %bb.0: # %entry
-; RV32I-NEXT:    lui a2, 349525
-; RV32I-NEXT:    lui a3, 209715
-; RV32I-NEXT:    lui a5, 61681
-; RV32I-NEXT:    addi a4, a2, 1365
-; RV32I-NEXT:    addi a3, a3, 819
-; RV32I-NEXT:    addi a2, a5, -241
-; RV32I-NEXT:    bnez a1, .LBB7_2
+; RV32I-NEXT:    addi sp, sp, -16
+; RV32I-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    mv s1, a1
+; RV32I-NEXT:    srli a1, a0, 1
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 4
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 16
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    not a0, a0
+; RV32I-NEXT:    call __popcountsi2
+; RV32I-NEXT:    mv s0, a0
+; RV32I-NEXT:    srli a0, s1, 1
+; RV32I-NEXT:    or a0, s1, a0
+; RV32I-NEXT:    srli a1, a0, 2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 4
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 16
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    not a0, a0
+; RV32I-NEXT:    call __popcountsi2
+; RV32I-NEXT:    bnez s1, .LBB7_2
 ; RV32I-NEXT:  # %bb.1: # %entry
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 2
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 8
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 16
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a3
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a3
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    addi a0, a0, 32
+; RV32I-NEXT:    addi a0, s0, 32
+; RV32I-NEXT:  .LBB7_2: # %entry
 ; RV32I-NEXT:    andi a0, a0, 63
-; RV32I-NEXT:    ret
-; RV32I-NEXT:  .LBB7_2:
-; RV32I-NEXT:    srli a0, a1, 1
-; RV32I-NEXT:    or a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 2
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 8
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 16
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a3
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a3
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    andi a0, a0, 63
+; RV32I-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    addi sp, sp, 16
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: ctlz:
 ; RV64I:       # %bb.0: # %entry
+; RV64I-NEXT:    addi sp, sp, -16
+; RV64I-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
 ; RV64I-NEXT:    srli a1, a0, 1
-; RV64I-NEXT:    lui a2, 349525
-; RV64I-NEXT:    lui a3, 209715
 ; RV64I-NEXT:    or a0, a0, a1
-; RV64I-NEXT:    addiw a1, a2, 1365
-; RV64I-NEXT:    addiw a2, a3, 819
-; RV64I-NEXT:    srli a3, a0, 2
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    slli a3, a1, 32
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    slli a3, a2, 32
-; RV64I-NEXT:    add a2, a2, a3
-; RV64I-NEXT:    srli a3, a0, 4
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 8
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 16
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 32
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    not a0, a0
-; RV64I-NEXT:    srli a3, a0, 1
-; RV64I-NEXT:    and a1, a3, a1
-; RV64I-NEXT:    lui a3, 61681
-; RV64I-NEXT:    addiw a3, a3, -241
-; RV64I-NEXT:    sub a0, a0, a1
-; RV64I-NEXT:    and a1, a0, a2
-; RV64I-NEXT:    srli a0, a0, 2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a2, a3, 32
-; RV64I-NEXT:    add a0, a1, a0
+; RV64I-NEXT:    srli a1, a0, 2
+; RV64I-NEXT:    or a0, a0, a1
 ; RV64I-NEXT:    srli a1, a0, 4
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    add a2, a3, a2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a1, a0, 8
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 16
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 32
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a0, a0, 2
-; RV64I-NEXT:    srli a0, a0, 58
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srli a1, a0, 8
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srli a1, a0, 16
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srli a1, a0, 32
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    not a0, a0
+; RV64I-NEXT:    call __popcountdi2
+; RV64I-NEXT:    andi a0, a0, 63
+; RV64I-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; RV64I-NEXT:    addi sp, sp, 16
 ; RV64I-NEXT:    ret
 
 
diff --git a/llvm/test/CodeGen/RISCV/pr56457.ll b/llvm/test/CodeGen/RISCV/pr56457.ll
index cf518b3..19cc8b3 100644
--- a/llvm/test/CodeGen/RISCV/pr56457.ll
+++ b/llvm/test/CodeGen/RISCV/pr56457.ll
@@ -9,46 +9,25 @@
 ; CHECK-NEXT:    slli a1, a0, 49
 ; CHECK-NEXT:    beqz a1, .LBB0_2
 ; CHECK-NEXT:  # %bb.1: # %cond.false
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
 ; CHECK-NEXT:    srli a1, a1, 50
-; CHECK-NEXT:    lui a2, 1
-; CHECK-NEXT:    lui a3, 209715
-; CHECK-NEXT:    lui a4, 61681
 ; CHECK-NEXT:    or a0, a0, a1
-; CHECK-NEXT:    addiw a1, a2, 1365
-; CHECK-NEXT:    addiw a2, a3, 819
-; CHECK-NEXT:    addiw a3, a4, -241
-; CHECK-NEXT:    slli a4, a2, 32
-; CHECK-NEXT:    add a2, a2, a4
-; CHECK-NEXT:    slli a4, a3, 32
-; CHECK-NEXT:    add a3, a3, a4
-; CHECK-NEXT:    slli a4, a0, 49
-; CHECK-NEXT:    srli a4, a4, 51
-; CHECK-NEXT:    or a0, a0, a4
-; CHECK-NEXT:    slli a4, a0, 49
-; CHECK-NEXT:    srli a4, a4, 53
-; CHECK-NEXT:    or a0, a0, a4
-; CHECK-NEXT:    slli a4, a0, 49
-; CHECK-NEXT:    srli a4, a4, 57
-; CHECK-NEXT:    or a0, a0, a4
+; CHECK-NEXT:    slli a1, a0, 49
+; CHECK-NEXT:    srli a1, a1, 51
+; CHECK-NEXT:    or a0, a0, a1
+; CHECK-NEXT:    slli a1, a0, 49
+; CHECK-NEXT:    srli a1, a1, 53
+; CHECK-NEXT:    or a0, a0, a1
+; CHECK-NEXT:    slli a1, a0, 49
+; CHECK-NEXT:    srli a1, a1, 57
+; CHECK-NEXT:    or a0, a0, a1
 ; CHECK-NEXT:    not a0, a0
-; CHECK-NEXT:    srli a4, a0, 1
-; CHECK-NEXT:    and a1, a4, a1
 ; CHECK-NEXT:    slli a0, a0, 49
 ; CHECK-NEXT:    srli a0, a0, 49
-; CHECK-NEXT:    sub a0, a0, a1
-; CHECK-NEXT:    and a1, a0, a2
-; CHECK-NEXT:    srli a0, a0, 2
-; CHECK-NEXT:    and a0, a0, a2
-; CHECK-NEXT:    add a0, a1, a0
-; CHECK-NEXT:    srli a1, a0, 4
-; CHECK-NEXT:    add a0, a0, a1
-; CHECK-NEXT:    lui a1, 4112
-; CHECK-NEXT:    addiw a1, a1, 257
-; CHECK-NEXT:    and a0, a0, a3
-; CHECK-NEXT:    slli a2, a1, 32
-; CHECK-NEXT:    add a1, a1, a2
-; CHECK-NEXT:    mul a0, a0, a1
-; CHECK-NEXT:    srli a0, a0, 56
+; CHECK-NEXT:    call __popcountdi2
+; CHECK-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; CHECK-NEXT:    addi sp, sp, 16
 ; CHECK-NEXT:    ret
 ; CHECK-NEXT:  .LBB0_2:
 ; CHECK-NEXT:    li a0, 15
diff --git a/llvm/test/CodeGen/RISCV/pr95271.ll b/llvm/test/CodeGen/RISCV/pr95271.ll
index aa941cb..46e9a19 100644
--- a/llvm/test/CodeGen/RISCV/pr95271.ll
+++ b/llvm/test/CodeGen/RISCV/pr95271.ll
@@ -6,29 +6,8 @@
 ; RV32I-LABEL: PR95271:
 ; RV32I:       # %bb.0:
 ; RV32I-NEXT:    lw a0, 0(a0)
-; RV32I-NEXT:    lui a1, 349525
-; RV32I-NEXT:    addi a1, a1, 1365
 ; RV32I-NEXT:    addi a0, a0, 1
-; RV32I-NEXT:    srli a2, a0, 1
-; RV32I-NEXT:    and a1, a2, a1
-; RV32I-NEXT:    lui a2, 209715
-; RV32I-NEXT:    addi a2, a2, 819
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a2
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    lui a2, 61681
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, -241
-; RV32I-NEXT:    and a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    ret
+; RV32I-NEXT:    tail __popcountsi2
 ;
 ; RV64I-LABEL: PR95271:
 ; RV64I:       # %bb.0:
diff --git a/llvm/test/CodeGen/RISCV/rv32xtheadbb.ll b/llvm/test/CodeGen/RISCV/rv32xtheadbb.ll
index 04a2f67..e783421 100644
--- a/llvm/test/CodeGen/RISCV/rv32xtheadbb.ll
+++ b/llvm/test/CodeGen/RISCV/rv32xtheadbb.ll
@@ -11,38 +11,22 @@
 ; RV32I:       # %bb.0:
 ; RV32I-NEXT:    beqz a0, .LBB0_2
 ; RV32I-NEXT:  # %bb.1: # %cond.false
+; RV32I-NEXT:    addi sp, sp, -16
+; RV32I-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    lui a2, 349525
 ; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, 1365
-; RV32I-NEXT:    srli a2, a0, 2
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 4
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 8
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 16
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a2, a0, 1
-; RV32I-NEXT:    and a1, a2, a1
-; RV32I-NEXT:    lui a2, 209715
-; RV32I-NEXT:    addi a2, a2, 819
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a2
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    lui a2, 61681
-; RV32I-NEXT:    add a0, a1, a0
+; RV32I-NEXT:    srli a1, a0, 2
+; RV32I-NEXT:    or a0, a0, a1
 ; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, -241
-; RV32I-NEXT:    and a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 16
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    not a0, a0
+; RV32I-NEXT:    call __popcountsi2
+; RV32I-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    addi sp, sp, 16
 ; RV32I-NEXT:    ret
 ; RV32I-NEXT:  .LBB0_2:
 ; RV32I-NEXT:    li a0, 32
@@ -61,71 +45,45 @@
 define i64 @ctlz_i64(i64 %a) nounwind {
 ; RV32I-LABEL: ctlz_i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    lui a2, 349525
-; RV32I-NEXT:    lui a3, 209715
-; RV32I-NEXT:    lui a5, 61681
-; RV32I-NEXT:    addi a4, a2, 1365
-; RV32I-NEXT:    addi a3, a3, 819
-; RV32I-NEXT:    addi a2, a5, -241
-; RV32I-NEXT:    bnez a1, .LBB1_2
+; RV32I-NEXT:    addi sp, sp, -16
+; RV32I-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    mv s1, a1
+; RV32I-NEXT:    srli a1, a0, 1
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 4
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 16
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    not a0, a0
+; RV32I-NEXT:    call __popcountsi2
+; RV32I-NEXT:    mv s0, a0
+; RV32I-NEXT:    srli a0, s1, 1
+; RV32I-NEXT:    or a0, s1, a0
+; RV32I-NEXT:    srli a1, a0, 2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 4
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 16
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    not a0, a0
+; RV32I-NEXT:    call __popcountsi2
+; RV32I-NEXT:    bnez s1, .LBB1_2
 ; RV32I-NEXT:  # %bb.1:
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 2
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 8
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 16
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a3
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a3
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    addi a0, a0, 32
-; RV32I-NEXT:    li a1, 0
-; RV32I-NEXT:    ret
+; RV32I-NEXT:    addi a0, s0, 32
 ; RV32I-NEXT:  .LBB1_2:
-; RV32I-NEXT:    srli a0, a1, 1
-; RV32I-NEXT:    or a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 2
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 8
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 16
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a3
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a3
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
 ; RV32I-NEXT:    li a1, 0
+; RV32I-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    addi sp, sp, 16
 ; RV32I-NEXT:    ret
 ;
 ; RV32XTHEADBB-LABEL: ctlz_i64:
diff --git a/llvm/test/CodeGen/RISCV/rv32zbb.ll b/llvm/test/CodeGen/RISCV/rv32zbb.ll
index 98c86da..acfdff8 100644
--- a/llvm/test/CodeGen/RISCV/rv32zbb.ll
+++ b/llvm/test/CodeGen/RISCV/rv32zbb.ll
@@ -11,38 +11,22 @@
 ; RV32I:       # %bb.0:
 ; RV32I-NEXT:    beqz a0, .LBB0_2
 ; RV32I-NEXT:  # %bb.1: # %cond.false
+; RV32I-NEXT:    addi sp, sp, -16
+; RV32I-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    lui a2, 349525
 ; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, 1365
-; RV32I-NEXT:    srli a2, a0, 2
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 4
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 8
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    srli a2, a0, 16
-; RV32I-NEXT:    or a0, a0, a2
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a2, a0, 1
-; RV32I-NEXT:    and a1, a2, a1
-; RV32I-NEXT:    lui a2, 209715
-; RV32I-NEXT:    addi a2, a2, 819
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a2
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    lui a2, 61681
-; RV32I-NEXT:    add a0, a1, a0
+; RV32I-NEXT:    srli a1, a0, 2
+; RV32I-NEXT:    or a0, a0, a1
 ; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, -241
-; RV32I-NEXT:    and a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 16
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    not a0, a0
+; RV32I-NEXT:    call __popcountsi2
+; RV32I-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    addi sp, sp, 16
 ; RV32I-NEXT:    ret
 ; RV32I-NEXT:  .LBB0_2:
 ; RV32I-NEXT:    li a0, 32
@@ -61,71 +45,45 @@
 define i64 @ctlz_i64(i64 %a) nounwind {
 ; RV32I-LABEL: ctlz_i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    lui a2, 349525
-; RV32I-NEXT:    lui a3, 209715
-; RV32I-NEXT:    lui a5, 61681
-; RV32I-NEXT:    addi a4, a2, 1365
-; RV32I-NEXT:    addi a3, a3, 819
-; RV32I-NEXT:    addi a2, a5, -241
-; RV32I-NEXT:    bnez a1, .LBB1_2
+; RV32I-NEXT:    addi sp, sp, -16
+; RV32I-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    mv s1, a1
+; RV32I-NEXT:    srli a1, a0, 1
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 4
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 16
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    not a0, a0
+; RV32I-NEXT:    call __popcountsi2
+; RV32I-NEXT:    mv s0, a0
+; RV32I-NEXT:    srli a0, s1, 1
+; RV32I-NEXT:    or a0, s1, a0
+; RV32I-NEXT:    srli a1, a0, 2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 4
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    srli a1, a0, 16
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    not a0, a0
+; RV32I-NEXT:    call __popcountsi2
+; RV32I-NEXT:    bnez s1, .LBB1_2
 ; RV32I-NEXT:  # %bb.1:
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 2
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 8
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 16
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a3
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a3
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    addi a0, a0, 32
-; RV32I-NEXT:    li a1, 0
-; RV32I-NEXT:    ret
+; RV32I-NEXT:    addi a0, s0, 32
 ; RV32I-NEXT:  .LBB1_2:
-; RV32I-NEXT:    srli a0, a1, 1
-; RV32I-NEXT:    or a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 2
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 8
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    srli a1, a0, 16
-; RV32I-NEXT:    or a0, a0, a1
-; RV32I-NEXT:    not a0, a0
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a3
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a3
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
 ; RV32I-NEXT:    li a1, 0
+; RV32I-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    addi sp, sp, 16
 ; RV32I-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: ctlz_i64:
@@ -253,28 +211,7 @@
 define i32 @ctpop_i32(i32 %a) nounwind {
 ; RV32I-LABEL: ctpop_i32:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    srli a1, a0, 1
-; RV32I-NEXT:    lui a2, 349525
-; RV32I-NEXT:    addi a2, a2, 1365
-; RV32I-NEXT:    and a1, a1, a2
-; RV32I-NEXT:    lui a2, 209715
-; RV32I-NEXT:    addi a2, a2, 819
-; RV32I-NEXT:    sub a0, a0, a1
-; RV32I-NEXT:    and a1, a0, a2
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a0, a0, a2
-; RV32I-NEXT:    lui a2, 61681
-; RV32I-NEXT:    add a0, a1, a0
-; RV32I-NEXT:    srli a1, a0, 4
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    addi a1, a2, -241
-; RV32I-NEXT:    and a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 8
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    slli a1, a0, 16
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    ret
+; RV32I-NEXT:    tail __popcountsi2
 ;
 ; RV32ZBB-LABEL: ctpop_i32:
 ; RV32ZBB:       # %bb.0:
@@ -365,42 +302,21 @@
 define <2 x i32> @ctpop_v2i32(<2 x i32> %a) nounwind {
 ; RV32I-LABEL: ctpop_v2i32:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    srli a2, a0, 1
-; RV32I-NEXT:    lui a3, 349525
-; RV32I-NEXT:    lui a4, 209715
-; RV32I-NEXT:    srli a5, a1, 1
-; RV32I-NEXT:    addi a3, a3, 1365
-; RV32I-NEXT:    and a2, a2, a3
-; RV32I-NEXT:    and a3, a5, a3
-; RV32I-NEXT:    lui a5, 61681
-; RV32I-NEXT:    addi a4, a4, 819
-; RV32I-NEXT:    addi a5, a5, -241
-; RV32I-NEXT:    sub a0, a0, a2
-; RV32I-NEXT:    sub a1, a1, a3
-; RV32I-NEXT:    and a2, a0, a4
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a3, a1, a4
-; RV32I-NEXT:    srli a1, a1, 2
-; RV32I-NEXT:    and a0, a0, a4
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    add a0, a2, a0
-; RV32I-NEXT:    add a1, a3, a1
-; RV32I-NEXT:    srli a2, a0, 4
-; RV32I-NEXT:    srli a3, a1, 4
-; RV32I-NEXT:    add a0, a0, a2
-; RV32I-NEXT:    add a1, a1, a3
-; RV32I-NEXT:    and a0, a0, a5
-; RV32I-NEXT:    and a1, a1, a5
-; RV32I-NEXT:    slli a2, a0, 8
-; RV32I-NEXT:    slli a3, a1, 8
-; RV32I-NEXT:    add a0, a0, a2
-; RV32I-NEXT:    add a1, a1, a3
-; RV32I-NEXT:    slli a2, a0, 16
-; RV32I-NEXT:    slli a3, a1, 16
-; RV32I-NEXT:    add a0, a0, a2
-; RV32I-NEXT:    add a1, a1, a3
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    srli a1, a1, 24
+; RV32I-NEXT:    addi sp, sp, -16
+; RV32I-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    mv s0, a1
+; RV32I-NEXT:    call __popcountsi2
+; RV32I-NEXT:    mv s1, a0
+; RV32I-NEXT:    mv a0, s0
+; RV32I-NEXT:    call __popcountsi2
+; RV32I-NEXT:    mv a1, a0
+; RV32I-NEXT:    mv a0, s1
+; RV32I-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    addi sp, sp, 16
 ; RV32I-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: ctpop_v2i32:
@@ -517,44 +433,12 @@
 define i64 @ctpop_i64(i64 %a) nounwind {
 ; RV32I-LABEL: ctpop_i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    srli a2, a1, 1
-; RV32I-NEXT:    lui a3, 349525
-; RV32I-NEXT:    lui a4, 209715
-; RV32I-NEXT:    srli a5, a0, 1
-; RV32I-NEXT:    addi a3, a3, 1365
-; RV32I-NEXT:    and a2, a2, a3
-; RV32I-NEXT:    and a3, a5, a3
-; RV32I-NEXT:    lui a5, 61681
-; RV32I-NEXT:    addi a4, a4, 819
-; RV32I-NEXT:    addi a5, a5, -241
-; RV32I-NEXT:    sub a1, a1, a2
-; RV32I-NEXT:    sub a0, a0, a3
-; RV32I-NEXT:    and a2, a1, a4
-; RV32I-NEXT:    srli a1, a1, 2
-; RV32I-NEXT:    and a3, a0, a4
-; RV32I-NEXT:    srli a0, a0, 2
-; RV32I-NEXT:    and a1, a1, a4
-; RV32I-NEXT:    and a0, a0, a4
-; RV32I-NEXT:    add a1, a2, a1
-; RV32I-NEXT:    add a0, a3, a0
-; RV32I-NEXT:    srli a2, a1, 4
-; RV32I-NEXT:    srli a3, a0, 4
-; RV32I-NEXT:    add a1, a1, a2
-; RV32I-NEXT:    add a0, a0, a3
-; RV32I-NEXT:    and a1, a1, a5
-; RV32I-NEXT:    and a0, a0, a5
-; RV32I-NEXT:    slli a2, a1, 8
-; RV32I-NEXT:    slli a3, a0, 8
-; RV32I-NEXT:    add a1, a1, a2
-; RV32I-NEXT:    add a0, a0, a3
-; RV32I-NEXT:    slli a2, a1, 16
-; RV32I-NEXT:    slli a3, a0, 16
-; RV32I-NEXT:    add a1, a1, a2
-; RV32I-NEXT:    add a0, a0, a3
-; RV32I-NEXT:    srli a1, a1, 24
-; RV32I-NEXT:    srli a0, a0, 24
-; RV32I-NEXT:    add a0, a0, a1
-; RV32I-NEXT:    li a1, 0
+; RV32I-NEXT:    addi sp, sp, -16
+; RV32I-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    call __popcountdi2
+; RV32I-NEXT:    srai a1, a0, 31
+; RV32I-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    addi sp, sp, 16
 ; RV32I-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: ctpop_i64:
@@ -682,82 +566,38 @@
 define <2 x i64> @ctpop_v2i64(<2 x i64> %a) nounwind {
 ; RV32I-LABEL: ctpop_v2i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    lw a3, 0(a1)
-; RV32I-NEXT:    lw a4, 4(a1)
-; RV32I-NEXT:    lw a2, 8(a1)
-; RV32I-NEXT:    lw a1, 12(a1)
-; RV32I-NEXT:    lui a5, 349525
-; RV32I-NEXT:    addi a5, a5, 1365
-; RV32I-NEXT:    srli a6, a4, 1
-; RV32I-NEXT:    srli a7, a3, 1
-; RV32I-NEXT:    srli t0, a1, 1
-; RV32I-NEXT:    srli t1, a2, 1
-; RV32I-NEXT:    and a6, a6, a5
-; RV32I-NEXT:    and a7, a7, a5
-; RV32I-NEXT:    and t0, t0, a5
-; RV32I-NEXT:    and a5, t1, a5
-; RV32I-NEXT:    lui t1, 209715
-; RV32I-NEXT:    addi t1, t1, 819
-; RV32I-NEXT:    sub a4, a4, a6
-; RV32I-NEXT:    sub a3, a3, a7
-; RV32I-NEXT:    sub a1, a1, t0
-; RV32I-NEXT:    sub a2, a2, a5
-; RV32I-NEXT:    and a5, a4, t1
-; RV32I-NEXT:    srli a4, a4, 2
-; RV32I-NEXT:    and a6, a3, t1
-; RV32I-NEXT:    srli a3, a3, 2
-; RV32I-NEXT:    and a7, a1, t1
-; RV32I-NEXT:    srli a1, a1, 2
-; RV32I-NEXT:    and t0, a2, t1
-; RV32I-NEXT:    srli a2, a2, 2
-; RV32I-NEXT:    and a4, a4, t1
-; RV32I-NEXT:    and a3, a3, t1
-; RV32I-NEXT:    and a1, a1, t1
-; RV32I-NEXT:    and a2, a2, t1
-; RV32I-NEXT:    add a4, a5, a4
-; RV32I-NEXT:    lui a5, 61681
-; RV32I-NEXT:    addi a5, a5, -241
-; RV32I-NEXT:    add a3, a6, a3
-; RV32I-NEXT:    add a1, a7, a1
-; RV32I-NEXT:    add a2, t0, a2
-; RV32I-NEXT:    srli a6, a4, 4
-; RV32I-NEXT:    srli a7, a3, 4
-; RV32I-NEXT:    srli t0, a1, 4
-; RV32I-NEXT:    add a4, a4, a6
-; RV32I-NEXT:    srli a6, a2, 4
-; RV32I-NEXT:    add a3, a3, a7
-; RV32I-NEXT:    add a1, a1, t0
-; RV32I-NEXT:    add a2, a2, a6
-; RV32I-NEXT:    and a4, a4, a5
-; RV32I-NEXT:    and a3, a3, a5
-; RV32I-NEXT:    and a1, a1, a5
-; RV32I-NEXT:    and a2, a2, a5
-; RV32I-NEXT:    slli a5, a4, 8
-; RV32I-NEXT:    slli a6, a3, 8
-; RV32I-NEXT:    slli a7, a1, 8
-; RV32I-NEXT:    slli t0, a2, 8
-; RV32I-NEXT:    add a4, a4, a5
-; RV32I-NEXT:    add a3, a3, a6
-; RV32I-NEXT:    add a1, a1, a7
-; RV32I-NEXT:    add a2, a2, t0
-; RV32I-NEXT:    slli a5, a4, 16
-; RV32I-NEXT:    slli a6, a3, 16
-; RV32I-NEXT:    slli a7, a1, 16
-; RV32I-NEXT:    slli t0, a2, 16
-; RV32I-NEXT:    add a4, a4, a5
-; RV32I-NEXT:    add a3, a3, a6
-; RV32I-NEXT:    add a1, a1, a7
-; RV32I-NEXT:    add a2, a2, t0
-; RV32I-NEXT:    srli a4, a4, 24
-; RV32I-NEXT:    srli a3, a3, 24
-; RV32I-NEXT:    srli a1, a1, 24
-; RV32I-NEXT:    srli a2, a2, 24
-; RV32I-NEXT:    add a3, a3, a4
-; RV32I-NEXT:    add a1, a2, a1
-; RV32I-NEXT:    sw a3, 0(a0)
-; RV32I-NEXT:    sw zero, 4(a0)
-; RV32I-NEXT:    sw a1, 8(a0)
-; RV32I-NEXT:    sw zero, 12(a0)
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    sw ra, 28(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s0, 24(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s1, 20(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s2, 16(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s3, 12(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s4, 8(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    lw a2, 0(a1)
+; RV32I-NEXT:    lw a3, 4(a1)
+; RV32I-NEXT:    lw s0, 8(a1)
+; RV32I-NEXT:    lw s1, 12(a1)
+; RV32I-NEXT:    mv s2, a0
+; RV32I-NEXT:    mv a0, a2
+; RV32I-NEXT:    mv a1, a3
+; RV32I-NEXT:    call __popcountdi2
+; RV32I-NEXT:    mv s3, a0
+; RV32I-NEXT:    srai s4, a0, 31
+; RV32I-NEXT:    mv a0, s0
+; RV32I-NEXT:    mv a1, s1
+; RV32I-NEXT:    call __popcountdi2
+; RV32I-NEXT:    srai a1, a0, 31
+; RV32I-NEXT:    sw s3, 0(s2)
+; RV32I-NEXT:    sw s4, 4(s2)
+; RV32I-NEXT:    sw a0, 8(s2)
+; RV32I-NEXT:    sw a1, 12(s2)
+; RV32I-NEXT:    lw ra, 28(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s0, 24(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s1, 20(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s2, 16(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s3, 12(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s4, 8(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: ctpop_v2i64:
diff --git a/llvm/test/CodeGen/RISCV/rv64xtheadbb.ll b/llvm/test/CodeGen/RISCV/rv64xtheadbb.ll
index d9f7d36..a11bc09 100644
--- a/llvm/test/CodeGen/RISCV/rv64xtheadbb.ll
+++ b/llvm/test/CodeGen/RISCV/rv64xtheadbb.ll
@@ -295,48 +295,24 @@
 ; RV64I:       # %bb.0:
 ; RV64I-NEXT:    beqz a0, .LBB5_2
 ; RV64I-NEXT:  # %bb.1: # %cond.false
+; RV64I-NEXT:    addi sp, sp, -16
+; RV64I-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
 ; RV64I-NEXT:    srli a1, a0, 1
-; RV64I-NEXT:    lui a2, 349525
-; RV64I-NEXT:    lui a3, 209715
 ; RV64I-NEXT:    or a0, a0, a1
-; RV64I-NEXT:    addiw a1, a2, 1365
-; RV64I-NEXT:    addiw a2, a3, 819
-; RV64I-NEXT:    srli a3, a0, 2
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    slli a3, a1, 32
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    slli a3, a2, 32
-; RV64I-NEXT:    add a2, a2, a3
-; RV64I-NEXT:    srli a3, a0, 4
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 8
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 16
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 32
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    not a0, a0
-; RV64I-NEXT:    srli a3, a0, 1
-; RV64I-NEXT:    and a1, a3, a1
-; RV64I-NEXT:    lui a3, 61681
-; RV64I-NEXT:    addiw a3, a3, -241
-; RV64I-NEXT:    sub a0, a0, a1
-; RV64I-NEXT:    and a1, a0, a2
-; RV64I-NEXT:    srli a0, a0, 2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a2, a3, 32
-; RV64I-NEXT:    add a0, a1, a0
+; RV64I-NEXT:    srli a1, a0, 2
+; RV64I-NEXT:    or a0, a0, a1
 ; RV64I-NEXT:    srli a1, a0, 4
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    add a2, a3, a2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a1, a0, 8
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 16
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 32
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    srli a0, a0, 56
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srli a1, a0, 8
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srli a1, a0, 16
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srli a1, a0, 32
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    not a0, a0
+; RV64I-NEXT:    call __popcountdi2
+; RV64I-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; RV64I-NEXT:    addi sp, sp, 16
 ; RV64I-NEXT:    ret
 ; RV64I-NEXT:  .LBB5_2:
 ; RV64I-NEXT:    li a0, 64
diff --git a/llvm/test/CodeGen/RISCV/rv64zbb.ll b/llvm/test/CodeGen/RISCV/rv64zbb.ll
index 17eb081..bb70784 100644
--- a/llvm/test/CodeGen/RISCV/rv64zbb.ll
+++ b/llvm/test/CodeGen/RISCV/rv64zbb.ll
@@ -285,48 +285,24 @@
 ; RV64I:       # %bb.0:
 ; RV64I-NEXT:    beqz a0, .LBB5_2
 ; RV64I-NEXT:  # %bb.1: # %cond.false
+; RV64I-NEXT:    addi sp, sp, -16
+; RV64I-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
 ; RV64I-NEXT:    srli a1, a0, 1
-; RV64I-NEXT:    lui a2, 349525
-; RV64I-NEXT:    lui a3, 209715
 ; RV64I-NEXT:    or a0, a0, a1
-; RV64I-NEXT:    addiw a1, a2, 1365
-; RV64I-NEXT:    addiw a2, a3, 819
-; RV64I-NEXT:    srli a3, a0, 2
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    slli a3, a1, 32
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    slli a3, a2, 32
-; RV64I-NEXT:    add a2, a2, a3
-; RV64I-NEXT:    srli a3, a0, 4
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 8
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 16
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    srli a3, a0, 32
-; RV64I-NEXT:    or a0, a0, a3
-; RV64I-NEXT:    not a0, a0
-; RV64I-NEXT:    srli a3, a0, 1
-; RV64I-NEXT:    and a1, a3, a1
-; RV64I-NEXT:    lui a3, 61681
-; RV64I-NEXT:    addiw a3, a3, -241
-; RV64I-NEXT:    sub a0, a0, a1
-; RV64I-NEXT:    and a1, a0, a2
-; RV64I-NEXT:    srli a0, a0, 2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a2, a3, 32
-; RV64I-NEXT:    add a0, a1, a0
+; RV64I-NEXT:    srli a1, a0, 2
+; RV64I-NEXT:    or a0, a0, a1
 ; RV64I-NEXT:    srli a1, a0, 4
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    add a2, a3, a2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a1, a0, 8
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 16
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 32
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    srli a0, a0, 56
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srli a1, a0, 8
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srli a1, a0, 16
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srli a1, a0, 32
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    not a0, a0
+; RV64I-NEXT:    call __popcountdi2
+; RV64I-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; RV64I-NEXT:    addi sp, sp, 16
 ; RV64I-NEXT:    ret
 ; RV64I-NEXT:  .LBB5_2:
 ; RV64I-NEXT:    li a0, 64
@@ -828,35 +804,11 @@
 define i64 @ctpop_i64(i64 %a) nounwind {
 ; RV64I-LABEL: ctpop_i64:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    lui a1, 349525
-; RV64I-NEXT:    lui a2, 209715
-; RV64I-NEXT:    addiw a1, a1, 1365
-; RV64I-NEXT:    addiw a2, a2, 819
-; RV64I-NEXT:    slli a3, a1, 32
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    slli a3, a2, 32
-; RV64I-NEXT:    add a2, a2, a3
-; RV64I-NEXT:    srli a3, a0, 1
-; RV64I-NEXT:    and a1, a3, a1
-; RV64I-NEXT:    lui a3, 61681
-; RV64I-NEXT:    addiw a3, a3, -241
-; RV64I-NEXT:    sub a0, a0, a1
-; RV64I-NEXT:    and a1, a0, a2
-; RV64I-NEXT:    srli a0, a0, 2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a2, a3, 32
-; RV64I-NEXT:    add a0, a1, a0
-; RV64I-NEXT:    srli a1, a0, 4
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    add a2, a3, a2
-; RV64I-NEXT:    and a0, a0, a2
-; RV64I-NEXT:    slli a1, a0, 8
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 16
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    slli a1, a0, 32
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    srli a0, a0, 56
+; RV64I-NEXT:    addi sp, sp, -16
+; RV64I-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
+; RV64I-NEXT:    call __popcountdi2
+; RV64I-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; RV64I-NEXT:    addi sp, sp, 16
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: ctpop_i64:
@@ -948,52 +900,21 @@
 define <2 x i64> @ctpop_v2i64(<2 x i64> %a) nounwind {
 ; RV64I-LABEL: ctpop_v2i64:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    srli a2, a0, 1
-; RV64I-NEXT:    lui a3, 349525
-; RV64I-NEXT:    lui a4, 209715
-; RV64I-NEXT:    lui a5, 61681
-; RV64I-NEXT:    addiw a3, a3, 1365
-; RV64I-NEXT:    addiw a4, a4, 819
-; RV64I-NEXT:    addiw a5, a5, -241
-; RV64I-NEXT:    slli a6, a3, 32
-; RV64I-NEXT:    add a3, a3, a6
-; RV64I-NEXT:    slli a6, a4, 32
-; RV64I-NEXT:    add a4, a4, a6
-; RV64I-NEXT:    slli a6, a5, 32
-; RV64I-NEXT:    add a5, a5, a6
-; RV64I-NEXT:    srli a6, a1, 1
-; RV64I-NEXT:    and a2, a2, a3
-; RV64I-NEXT:    and a3, a6, a3
-; RV64I-NEXT:    sub a0, a0, a2
-; RV64I-NEXT:    sub a1, a1, a3
-; RV64I-NEXT:    and a2, a0, a4
-; RV64I-NEXT:    srli a0, a0, 2
-; RV64I-NEXT:    and a3, a1, a4
-; RV64I-NEXT:    srli a1, a1, 2
-; RV64I-NEXT:    and a0, a0, a4
-; RV64I-NEXT:    and a1, a1, a4
-; RV64I-NEXT:    add a0, a2, a0
-; RV64I-NEXT:    add a1, a3, a1
-; RV64I-NEXT:    srli a2, a0, 4
-; RV64I-NEXT:    srli a3, a1, 4
-; RV64I-NEXT:    add a0, a0, a2
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    and a0, a0, a5
-; RV64I-NEXT:    and a1, a1, a5
-; RV64I-NEXT:    slli a2, a0, 8
-; RV64I-NEXT:    slli a3, a1, 8
-; RV64I-NEXT:    add a0, a0, a2
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    slli a2, a0, 16
-; RV64I-NEXT:    slli a3, a1, 16
-; RV64I-NEXT:    add a0, a0, a2
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    slli a2, a0, 32
-; RV64I-NEXT:    slli a3, a1, 32
-; RV64I-NEXT:    add a0, a0, a2
-; RV64I-NEXT:    add a1, a1, a3
-; RV64I-NEXT:    srli a0, a0, 56
-; RV64I-NEXT:    srli a1, a1, 56
+; RV64I-NEXT:    addi sp, sp, -32
+; RV64I-NEXT:    sd ra, 24(sp) # 8-byte Folded Spill
+; RV64I-NEXT:    sd s0, 16(sp) # 8-byte Folded Spill
+; RV64I-NEXT:    sd s1, 8(sp) # 8-byte Folded Spill
+; RV64I-NEXT:    mv s0, a1
+; RV64I-NEXT:    call __popcountdi2
+; RV64I-NEXT:    mv s1, a0
+; RV64I-NEXT:    mv a0, s0
+; RV64I-NEXT:    call __popcountdi2
+; RV64I-NEXT:    mv a1, a0
+; RV64I-NEXT:    mv a0, s1
+; RV64I-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
+; RV64I-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
+; RV64I-NEXT:    ld s1, 8(sp) # 8-byte Folded Reload
+; RV64I-NEXT:    addi sp, sp, 32
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: ctpop_v2i64:
diff --git a/llvm/test/CodeGen/RISCV/sextw-removal.ll b/llvm/test/CodeGen/RISCV/sextw-removal.ll
index 4949460..4494d9b8 100644
--- a/llvm/test/CodeGen/RISCV/sextw-removal.ll
+++ b/llvm/test/CodeGen/RISCV/sextw-removal.ll
@@ -316,52 +316,18 @@
 define void @test7(i32 signext %arg, i32 signext %arg1) nounwind {
 ; RV64I-LABEL: test7:
 ; RV64I:       # %bb.0: # %bb
-; RV64I-NEXT:    addi sp, sp, -48
-; RV64I-NEXT:    sd ra, 40(sp) # 8-byte Folded Spill
-; RV64I-NEXT:    sd s0, 32(sp) # 8-byte Folded Spill
-; RV64I-NEXT:    sd s1, 24(sp) # 8-byte Folded Spill
-; RV64I-NEXT:    sd s2, 16(sp) # 8-byte Folded Spill
-; RV64I-NEXT:    sd s3, 8(sp) # 8-byte Folded Spill
+; RV64I-NEXT:    addi sp, sp, -16
+; RV64I-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
 ; RV64I-NEXT:    sraw a0, a0, a1
-; RV64I-NEXT:    lui a1, 349525
-; RV64I-NEXT:    lui a2, 209715
-; RV64I-NEXT:    lui a3, 61681
-; RV64I-NEXT:    lui a4, 4112
-; RV64I-NEXT:    addiw s0, a1, 1365
-; RV64I-NEXT:    addiw s1, a2, 819
-; RV64I-NEXT:    addiw s2, a3, -241
-; RV64I-NEXT:    addiw s3, a4, 257
-; RV64I-NEXT:    slli a1, s0, 32
-; RV64I-NEXT:    add s0, s0, a1
-; RV64I-NEXT:    slli a1, s1, 32
-; RV64I-NEXT:    add s1, s1, a1
-; RV64I-NEXT:    slli a1, s2, 32
-; RV64I-NEXT:    add s2, s2, a1
-; RV64I-NEXT:    slli a1, s3, 32
-; RV64I-NEXT:    add s3, s3, a1
 ; RV64I-NEXT:  .LBB6_1: # %bb2
 ; RV64I-NEXT:    # =>This Inner Loop Header: Depth=1
+; RV64I-NEXT:    sext.w a0, a0
 ; RV64I-NEXT:    call foo
-; RV64I-NEXT:    srli a1, a0, 1
-; RV64I-NEXT:    and a1, a1, s0
-; RV64I-NEXT:    sub a0, a0, a1
-; RV64I-NEXT:    and a1, a0, s1
-; RV64I-NEXT:    srli a0, a0, 2
-; RV64I-NEXT:    and a0, a0, s1
-; RV64I-NEXT:    add a0, a1, a0
-; RV64I-NEXT:    srli a1, a0, 4
-; RV64I-NEXT:    add a0, a0, a1
-; RV64I-NEXT:    and a0, a0, s2
-; RV64I-NEXT:    mul a0, a0, s3
-; RV64I-NEXT:    srli a0, a0, 56
+; RV64I-NEXT:    call __popcountdi2
 ; RV64I-NEXT:    bnez a0, .LBB6_1
 ; RV64I-NEXT:  # %bb.2: # %bb7
-; RV64I-NEXT:    ld ra, 40(sp) # 8-byte Folded Reload
-; RV64I-NEXT:    ld s0, 32(sp) # 8-byte Folded Reload
-; RV64I-NEXT:    ld s1, 24(sp) # 8-byte Folded Reload
-; RV64I-NEXT:    ld s2, 16(sp) # 8-byte Folded Reload
-; RV64I-NEXT:    ld s3, 8(sp) # 8-byte Folded Reload
-; RV64I-NEXT:    addi sp, sp, 48
+; RV64I-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; RV64I-NEXT:    addi sp, sp, 16
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: test7:
diff --git a/llvm/test/CodeGen/Thumb2/mve-ctpop.ll b/llvm/test/CodeGen/Thumb2/mve-ctpop.ll
index 724bd4f..9f89b2c 100644
--- a/llvm/test/CodeGen/Thumb2/mve-ctpop.ll
+++ b/llvm/test/CodeGen/Thumb2/mve-ctpop.ll
@@ -1,5 +1,4 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; NOTE: Assertions have been autoenerated by utils/update_llc_test_checks.py
 ; RUN: llc -mtriple=thumbv8.1m.main-none-none-eabi -mattr=+mve.fp -verify-machineinstrs %s -o - | FileCheck %s --check-prefix=CHECK
 
 define arm_aapcs_vfpcc <2 x i64> @ctpop_2i64_t(<2 x i64> %src){
@@ -7,56 +6,20 @@
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    .save {r4, r5, r7, lr}
 ; CHECK-NEXT:    push {r4, r5, r7, lr}
-; CHECK-NEXT:    vmov r1, r2, d1
-; CHECK-NEXT:    mov.w lr, #1431655765
-; CHECK-NEXT:    vmov r3, r4, d0
-; CHECK-NEXT:    mov.w r12, #858993459
-; CHECK-NEXT:    vldr s1, .LCPI0_0
-; CHECK-NEXT:    vmov.f32 s3, s1
-; CHECK-NEXT:    and.w r0, lr, r2, lsr #1
-; CHECK-NEXT:    subs r0, r2, r0
-; CHECK-NEXT:    and.w r2, r12, r0, lsr #2
-; CHECK-NEXT:    bic r0, r0, #-858993460
-; CHECK-NEXT:    add r0, r2
-; CHECK-NEXT:    and.w r2, lr, r1, lsr #1
-; CHECK-NEXT:    subs r1, r1, r2
-; CHECK-NEXT:    add.w r0, r0, r0, lsr #4
-; CHECK-NEXT:    and.w r2, r12, r1, lsr #2
-; CHECK-NEXT:    bic r1, r1, #-858993460
-; CHECK-NEXT:    add r1, r2
-; CHECK-NEXT:    and.w r2, lr, r3, lsr #1
-; CHECK-NEXT:    subs r2, r3, r2
-; CHECK-NEXT:    bic r5, r0, #-252645136
-; CHECK-NEXT:    add.w r1, r1, r1, lsr #4
-; CHECK-NEXT:    mov.w r0, #16843009
-; CHECK-NEXT:    and.w r3, r12, r2, lsr #2
-; CHECK-NEXT:    bic r2, r2, #-858993460
-; CHECK-NEXT:    add r2, r3
-; CHECK-NEXT:    and.w r3, lr, r4, lsr #1
-; CHECK-NEXT:    subs r3, r4, r3
-; CHECK-NEXT:    bic r1, r1, #-252645136
-; CHECK-NEXT:    add.w r2, r2, r2, lsr #4
-; CHECK-NEXT:    muls r5, r0, r5
-; CHECK-NEXT:    and.w r4, r12, r3, lsr #2
-; CHECK-NEXT:    bic r3, r3, #-858993460
-; CHECK-NEXT:    bic r2, r2, #-252645136
-; CHECK-NEXT:    add r3, r4
-; CHECK-NEXT:    muls r1, r0, r1
-; CHECK-NEXT:    add.w r3, r3, r3, lsr #4
-; CHECK-NEXT:    muls r2, r0, r2
-; CHECK-NEXT:    bic r3, r3, #-252645136
-; CHECK-NEXT:    muls r0, r3, r0
-; CHECK-NEXT:    lsrs r1, r1, #24
-; CHECK-NEXT:    add.w r1, r1, r5, lsr #24
-; CHECK-NEXT:    lsrs r2, r2, #24
-; CHECK-NEXT:    vmov s2, r1
-; CHECK-NEXT:    add.w r0, r2, r0, lsr #24
-; CHECK-NEXT:    vmov s0, r0
+; CHECK-NEXT:    .vsave {d8, d9}
+; CHECK-NEXT:    vpush {d8, d9}
+; CHECK-NEXT:    vmov q4, q0
+; CHECK-NEXT:    vmov r0, r1, d9
+; CHECK-NEXT:    bl __popcountdi2
+; CHECK-NEXT:    mov r4, r0
+; CHECK-NEXT:    vmov r0, r1, d8
+; CHECK-NEXT:    asrs r5, r4, #31
+; CHECK-NEXT:    bl __popcountdi2
+; CHECK-NEXT:    asrs r1, r0, #31
+; CHECK-NEXT:    vmov q0[2], q0[0], r0, r4
+; CHECK-NEXT:    vmov q0[3], q0[1], r1, r5
+; CHECK-NEXT:    vpop {d8, d9}
 ; CHECK-NEXT:    pop {r4, r5, r7, pc}
-; CHECK-NEXT:    .p2align 2
-; CHECK-NEXT:  @ %bb.1:
-; CHECK-NEXT:  .LCPI0_0:
-; CHECK-NEXT:    .long 0x00000000 @ float 0
 entry:
   %0 = call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %src)
   ret <2 x i64> %0