diff --git a/llvm/include/llvm/CodeGen/TargetLowering.h b/llvm/include/llvm/CodeGen/TargetLowering.h
--- a/llvm/include/llvm/CodeGen/TargetLowering.h
+++ b/llvm/include/llvm/CodeGen/TargetLowering.h
@@ -651,6 +651,11 @@
     return false;
   }
 
+  /// Return true if ctpop instruction is fast.
+  virtual bool isCtpopFast(EVT VT) const {
+    return isOperationLegal(ISD::CTPOP, VT);
+  }
+
   /// Return the maximum number of "x & (x - 1)" operations that can be done
   /// instead of deferring to a custom CTPOP.
   virtual unsigned getCustomCtpopCost(EVT VT, ISD::CondCode Cond) const {
diff --git a/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp b/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
--- a/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
@@ -4088,8 +4088,8 @@
   // (ctpop x) u< 2 -> (x & x-1) == 0
   // (ctpop x) u> 1 -> (x & x-1) != 0
   if (Cond == ISD::SETULT || Cond == ISD::SETUGT) {
-    // Keep the CTPOP if it is a legal vector op.
-    if (CTVT.isVector() && TLI.isOperationLegal(ISD::CTPOP, CTVT))
+    // Keep the CTPOP if it is a cheap vector op.
+    if (CTVT.isVector() && TLI.isCtpopFast(CTVT))
       return SDValue();
 
     unsigned CostLimit = TLI.getCustomCtpopCost(CTVT, Cond);
@@ -4114,8 +4114,8 @@
   // (ctpop x) == 1 --> (x != 0) && ((x & x-1) == 0)
   // (ctpop x) != 1 --> (x == 0) || ((x & x-1) != 0)
   if ((Cond == ISD::SETEQ || Cond == ISD::SETNE) && C1 == 1) {
-    // Keep the CTPOP if it is legal.
-    if (TLI.isOperationLegal(ISD::CTPOP, CTVT))
+    // Keep the CTPOP if it is cheap.
+    if (TLI.isCtpopFast(CTVT))
       return SDValue();
 
     SDValue Zero = DAG.getConstant(0, dl, CTVT);
diff --git a/llvm/lib/Target/RISCV/RISCVISelLowering.h b/llvm/lib/Target/RISCV/RISCVISelLowering.h
--- a/llvm/lib/Target/RISCV/RISCVISelLowering.h
+++ b/llvm/lib/Target/RISCV/RISCVISelLowering.h
@@ -592,6 +592,10 @@
   }
   bool convertSelectOfConstantsToMath(EVT VT) const override { return true; }
 
+  bool isCtpopFast(EVT VT) const override;
+
+  unsigned getCustomCtpopCost(EVT VT, ISD::CondCode Cond) const override;
+
   bool preferZeroCompareBranch() const override { return true; }
 
   bool shouldInsertFencesForAtomic(const Instruction *I) const override {
diff --git a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
--- a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
@@ -17497,6 +17497,20 @@
   return getTargetMMOFlags(NodeX) == getTargetMMOFlags(NodeY);
 }
 
+bool RISCVTargetLowering::isCtpopFast(EVT VT) const {
+  if (VT.isVector())
+    return isTypeLegal(VT) && Subtarget.hasStdExtZvbb();
+  return Subtarget.hasStdExtZbb() && (VT == MVT::i32 || VT == MVT::i64);
+}
+
+unsigned RISCVTargetLowering::getCustomCtpopCost(EVT VT,
+                                                 ISD::CondCode Cond) const {
+  if (Cond == ISD::SETUGT &&
+      (VT == MVT::i32 || (Subtarget.is64Bit() && VT == MVT::i64)))
+    return 1;
+  return isCtpopFast(VT) ? 0 : 1;
+}
+
 namespace llvm::RISCVVIntrinsicsTable {
 
 #define GET_RISCVVIntrinsicsTable_IMPL
diff --git a/llvm/test/CodeGen/RISCV/rv32zbb.ll b/llvm/test/CodeGen/RISCV/rv32zbb.ll
--- a/llvm/test/CodeGen/RISCV/rv32zbb.ll
+++ b/llvm/test/CodeGen/RISCV/rv32zbb.ll
@@ -309,6 +309,78 @@
   ret i32 %1
 }
 
+define i1 @ctpop_i32_ult_two(i32 signext %a) nounwind {
+; RV32I-LABEL: ctpop_i32_ult_two:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    addi a1, a0, -1
+; RV32I-NEXT:    and a0, a0, a1
+; RV32I-NEXT:    seqz a0, a0
+; RV32I-NEXT:    ret
+;
+; RV32ZBB-LABEL: ctpop_i32_ult_two:
+; RV32ZBB:       # %bb.0:
+; RV32ZBB-NEXT:    cpop a0, a0
+; RV32ZBB-NEXT:    sltiu a0, a0, 2
+; RV32ZBB-NEXT:    ret
+  %1 = call i32 @llvm.ctpop.i32(i32 %a)
+  %2 = icmp ult i32 %1, 2
+  ret i1 %2
+}
+
+define i1 @ctpop_i32_ugt_one(i32 signext %a) nounwind {
+; CHECK-LABEL: ctpop_i32_ugt_one:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi a1, a0, -1
+; CHECK-NEXT:    and a0, a0, a1
+; CHECK-NEXT:    snez a0, a0
+; CHECK-NEXT:    ret
+  %1 = call i32 @llvm.ctpop.i32(i32 %a)
+  %2 = icmp ugt i32 %1, 1
+  ret i1 %2
+}
+
+define i1 @ctpop_i32_eq_one(i32 signext %a) nounwind {
+; RV32I-LABEL: ctpop_i32_eq_one:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    addi a1, a0, -1
+; RV32I-NEXT:    and a1, a0, a1
+; RV32I-NEXT:    seqz a1, a1
+; RV32I-NEXT:    snez a0, a0
+; RV32I-NEXT:    and a0, a0, a1
+; RV32I-NEXT:    ret
+;
+; RV32ZBB-LABEL: ctpop_i32_eq_one:
+; RV32ZBB:       # %bb.0:
+; RV32ZBB-NEXT:    cpop a0, a0
+; RV32ZBB-NEXT:    addi a0, a0, -1
+; RV32ZBB-NEXT:    seqz a0, a0
+; RV32ZBB-NEXT:    ret
+  %1 = call i32 @llvm.ctpop.i32(i32 %a)
+  %2 = icmp eq i32 %1, 1
+  ret i1 %2
+}
+
+define i1 @ctpop_i32_ne_one(i32 signext %a) nounwind {
+; RV32I-LABEL: ctpop_i32_ne_one:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    addi a1, a0, -1
+; RV32I-NEXT:    and a1, a0, a1
+; RV32I-NEXT:    snez a1, a1
+; RV32I-NEXT:    seqz a0, a0
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    ret
+;
+; RV32ZBB-LABEL: ctpop_i32_ne_one:
+; RV32ZBB:       # %bb.0:
+; RV32ZBB-NEXT:    cpop a0, a0
+; RV32ZBB-NEXT:    addi a0, a0, -1
+; RV32ZBB-NEXT:    snez a0, a0
+; RV32ZBB-NEXT:    ret
+  %1 = call i32 @llvm.ctpop.i32(i32 %a)
+  %2 = icmp ne i32 %1, 1
+  ret i1 %2
+}
+
 declare i64 @llvm.ctpop.i64(i64)
 
 define i64 @ctpop_i64(i64 %a) nounwind {
@@ -380,6 +452,111 @@
   ret i64 %1
 }
 
+define i1 @ctpop_i64_ugt_two(i64 %a) nounwind {
+; RV32I-LABEL: ctpop_i64_ugt_two:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    addi a2, a0, -1
+; RV32I-NEXT:    and a2, a0, a2
+; RV32I-NEXT:    seqz a0, a0
+; RV32I-NEXT:    sub a0, a1, a0
+; RV32I-NEXT:    and a0, a1, a0
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    seqz a0, a0
+; RV32I-NEXT:    ret
+;
+; RV32ZBB-LABEL: ctpop_i64_ugt_two:
+; RV32ZBB:       # %bb.0:
+; RV32ZBB-NEXT:    cpop a1, a1
+; RV32ZBB-NEXT:    cpop a0, a0
+; RV32ZBB-NEXT:    add a0, a0, a1
+; RV32ZBB-NEXT:    sltiu a0, a0, 2
+; RV32ZBB-NEXT:    ret
+  %1 = call i64 @llvm.ctpop.i64(i64 %a)
+  %2 = icmp ult i64 %1, 2
+  ret i1 %2
+}
+
+define i1 @ctpop_i64_ugt_one(i64 %a) nounwind {
+; RV32I-LABEL: ctpop_i64_ugt_one:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    addi a2, a0, -1
+; RV32I-NEXT:    and a2, a0, a2
+; RV32I-NEXT:    seqz a0, a0
+; RV32I-NEXT:    sub a0, a1, a0
+; RV32I-NEXT:    and a0, a1, a0
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    snez a0, a0
+; RV32I-NEXT:    ret
+;
+; RV32ZBB-LABEL: ctpop_i64_ugt_one:
+; RV32ZBB:       # %bb.0:
+; RV32ZBB-NEXT:    cpop a1, a1
+; RV32ZBB-NEXT:    cpop a0, a0
+; RV32ZBB-NEXT:    add a0, a0, a1
+; RV32ZBB-NEXT:    sltiu a0, a0, 2
+; RV32ZBB-NEXT:    xori a0, a0, 1
+; RV32ZBB-NEXT:    ret
+  %1 = call i64 @llvm.ctpop.i64(i64 %a)
+  %2 = icmp ugt i64 %1, 1
+  ret i1 %2
+}
+
+define i1 @ctpop_i64_eq_one(i64 %a) nounwind {
+; RV32I-LABEL: ctpop_i64_eq_one:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    addi a2, a0, -1
+; RV32I-NEXT:    and a2, a0, a2
+; RV32I-NEXT:    seqz a3, a0
+; RV32I-NEXT:    sub a3, a1, a3
+; RV32I-NEXT:    and a3, a1, a3
+; RV32I-NEXT:    or a2, a2, a3
+; RV32I-NEXT:    seqz a2, a2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    snez a0, a0
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    ret
+;
+; RV32ZBB-LABEL: ctpop_i64_eq_one:
+; RV32ZBB:       # %bb.0:
+; RV32ZBB-NEXT:    cpop a1, a1
+; RV32ZBB-NEXT:    cpop a0, a0
+; RV32ZBB-NEXT:    add a0, a0, a1
+; RV32ZBB-NEXT:    addi a0, a0, -1
+; RV32ZBB-NEXT:    seqz a0, a0
+; RV32ZBB-NEXT:    ret
+  %1 = call i64 @llvm.ctpop.i64(i64 %a)
+  %2 = icmp eq i64 %1, 1
+  ret i1 %2
+}
+
+define i1 @ctpop_i64_ne_one(i64 %a) nounwind {
+; RV32I-LABEL: ctpop_i64_ne_one:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    addi a2, a0, -1
+; RV32I-NEXT:    and a2, a0, a2
+; RV32I-NEXT:    seqz a3, a0
+; RV32I-NEXT:    sub a3, a1, a3
+; RV32I-NEXT:    and a3, a1, a3
+; RV32I-NEXT:    or a2, a2, a3
+; RV32I-NEXT:    snez a2, a2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    seqz a0, a0
+; RV32I-NEXT:    or a0, a0, a2
+; RV32I-NEXT:    ret
+;
+; RV32ZBB-LABEL: ctpop_i64_ne_one:
+; RV32ZBB:       # %bb.0:
+; RV32ZBB-NEXT:    cpop a1, a1
+; RV32ZBB-NEXT:    cpop a0, a0
+; RV32ZBB-NEXT:    add a0, a0, a1
+; RV32ZBB-NEXT:    addi a0, a0, -1
+; RV32ZBB-NEXT:    snez a0, a0
+; RV32ZBB-NEXT:    ret
+  %1 = call i64 @llvm.ctpop.i64(i64 %a)
+  %2 = icmp ne i64 %1, 1
+  ret i1 %2
+}
+
 define i32 @sextb_i32(i32 %a) nounwind {
 ; RV32I-LABEL: sextb_i32:
 ; RV32I:       # %bb.0:
@@ -451,10 +628,10 @@
 define i32 @min_i32(i32 %a, i32 %b) nounwind {
 ; RV32I-LABEL: min_i32:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    blt a0, a1, .LBB10_2
+; RV32I-NEXT:    blt a0, a1, .LBB18_2
 ; RV32I-NEXT:  # %bb.1:
 ; RV32I-NEXT:    mv a0, a1
-; RV32I-NEXT:  .LBB10_2:
+; RV32I-NEXT:  .LBB18_2:
 ; RV32I-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: min_i32:
@@ -474,18 +651,18 @@
 define i64 @min_i64(i64 %a, i64 %b) nounwind {
 ; CHECK-LABEL: min_i64:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    beq a1, a3, .LBB11_2
+; CHECK-NEXT:    beq a1, a3, .LBB19_2
 ; CHECK-NEXT:  # %bb.1:
 ; CHECK-NEXT:    slt a4, a1, a3
-; CHECK-NEXT:    beqz a4, .LBB11_3
-; CHECK-NEXT:    j .LBB11_4
-; CHECK-NEXT:  .LBB11_2:
+; CHECK-NEXT:    beqz a4, .LBB19_3
+; CHECK-NEXT:    j .LBB19_4
+; CHECK-NEXT:  .LBB19_2:
 ; CHECK-NEXT:    sltu a4, a0, a2
-; CHECK-NEXT:    bnez a4, .LBB11_4
-; CHECK-NEXT:  .LBB11_3:
+; CHECK-NEXT:    bnez a4, .LBB19_4
+; CHECK-NEXT:  .LBB19_3:
 ; CHECK-NEXT:    mv a0, a2
 ; CHECK-NEXT:    mv a1, a3
-; CHECK-NEXT:  .LBB11_4:
+; CHECK-NEXT:  .LBB19_4:
 ; CHECK-NEXT:    ret
   %cmp = icmp slt i64 %a, %b
   %cond = select i1 %cmp, i64 %a, i64 %b
@@ -495,10 +672,10 @@
 define i32 @max_i32(i32 %a, i32 %b) nounwind {
 ; RV32I-LABEL: max_i32:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    blt a1, a0, .LBB12_2
+; RV32I-NEXT:    blt a1, a0, .LBB20_2
 ; RV32I-NEXT:  # %bb.1:
 ; RV32I-NEXT:    mv a0, a1
-; RV32I-NEXT:  .LBB12_2:
+; RV32I-NEXT:  .LBB20_2:
 ; RV32I-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: max_i32:
@@ -518,18 +695,18 @@
 define i64 @max_i64(i64 %a, i64 %b) nounwind {
 ; CHECK-LABEL: max_i64:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    beq a1, a3, .LBB13_2
+; CHECK-NEXT:    beq a1, a3, .LBB21_2
 ; CHECK-NEXT:  # %bb.1:
 ; CHECK-NEXT:    slt a4, a3, a1
-; CHECK-NEXT:    beqz a4, .LBB13_3
-; CHECK-NEXT:    j .LBB13_4
-; CHECK-NEXT:  .LBB13_2:
+; CHECK-NEXT:    beqz a4, .LBB21_3
+; CHECK-NEXT:    j .LBB21_4
+; CHECK-NEXT:  .LBB21_2:
 ; CHECK-NEXT:    sltu a4, a2, a0
-; CHECK-NEXT:    bnez a4, .LBB13_4
-; CHECK-NEXT:  .LBB13_3:
+; CHECK-NEXT:    bnez a4, .LBB21_4
+; CHECK-NEXT:  .LBB21_3:
 ; CHECK-NEXT:    mv a0, a2
 ; CHECK-NEXT:    mv a1, a3
-; CHECK-NEXT:  .LBB13_4:
+; CHECK-NEXT:  .LBB21_4:
 ; CHECK-NEXT:    ret
   %cmp = icmp sgt i64 %a, %b
   %cond = select i1 %cmp, i64 %a, i64 %b
@@ -539,10 +716,10 @@
 define i32 @minu_i32(i32 %a, i32 %b) nounwind {
 ; RV32I-LABEL: minu_i32:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    bltu a0, a1, .LBB14_2
+; RV32I-NEXT:    bltu a0, a1, .LBB22_2
 ; RV32I-NEXT:  # %bb.1:
 ; RV32I-NEXT:    mv a0, a1
-; RV32I-NEXT:  .LBB14_2:
+; RV32I-NEXT:  .LBB22_2:
 ; RV32I-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: minu_i32:
@@ -562,18 +739,18 @@
 define i64 @minu_i64(i64 %a, i64 %b) nounwind {
 ; CHECK-LABEL: minu_i64:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    beq a1, a3, .LBB15_2
+; CHECK-NEXT:    beq a1, a3, .LBB23_2
 ; CHECK-NEXT:  # %bb.1:
 ; CHECK-NEXT:    sltu a4, a1, a3
-; CHECK-NEXT:    beqz a4, .LBB15_3
-; CHECK-NEXT:    j .LBB15_4
-; CHECK-NEXT:  .LBB15_2:
+; CHECK-NEXT:    beqz a4, .LBB23_3
+; CHECK-NEXT:    j .LBB23_4
+; CHECK-NEXT:  .LBB23_2:
 ; CHECK-NEXT:    sltu a4, a0, a2
-; CHECK-NEXT:    bnez a4, .LBB15_4
-; CHECK-NEXT:  .LBB15_3:
+; CHECK-NEXT:    bnez a4, .LBB23_4
+; CHECK-NEXT:  .LBB23_3:
 ; CHECK-NEXT:    mv a0, a2
 ; CHECK-NEXT:    mv a1, a3
-; CHECK-NEXT:  .LBB15_4:
+; CHECK-NEXT:  .LBB23_4:
 ; CHECK-NEXT:    ret
   %cmp = icmp ult i64 %a, %b
   %cond = select i1 %cmp, i64 %a, i64 %b
@@ -583,10 +760,10 @@
 define i32 @maxu_i32(i32 %a, i32 %b) nounwind {
 ; RV32I-LABEL: maxu_i32:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    bltu a1, a0, .LBB16_2
+; RV32I-NEXT:    bltu a1, a0, .LBB24_2
 ; RV32I-NEXT:  # %bb.1:
 ; RV32I-NEXT:    mv a0, a1
-; RV32I-NEXT:  .LBB16_2:
+; RV32I-NEXT:  .LBB24_2:
 ; RV32I-NEXT:    ret
 ;
 ; RV32ZBB-LABEL: maxu_i32:
@@ -606,18 +783,18 @@
 define i64 @maxu_i64(i64 %a, i64 %b) nounwind {
 ; CHECK-LABEL: maxu_i64:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    beq a1, a3, .LBB17_2
+; CHECK-NEXT:    beq a1, a3, .LBB25_2
 ; CHECK-NEXT:  # %bb.1:
 ; CHECK-NEXT:    sltu a4, a3, a1
-; CHECK-NEXT:    beqz a4, .LBB17_3
-; CHECK-NEXT:    j .LBB17_4
-; CHECK-NEXT:  .LBB17_2:
+; CHECK-NEXT:    beqz a4, .LBB25_3
+; CHECK-NEXT:    j .LBB25_4
+; CHECK-NEXT:  .LBB25_2:
 ; CHECK-NEXT:    sltu a4, a2, a0
-; CHECK-NEXT:    bnez a4, .LBB17_4
-; CHECK-NEXT:  .LBB17_3:
+; CHECK-NEXT:    bnez a4, .LBB25_4
+; CHECK-NEXT:  .LBB25_3:
 ; CHECK-NEXT:    mv a0, a2
 ; CHECK-NEXT:    mv a1, a3
-; CHECK-NEXT:  .LBB17_4:
+; CHECK-NEXT:  .LBB25_4:
 ; CHECK-NEXT:    ret
   %cmp = icmp ugt i64 %a, %b
   %cond = select i1 %cmp, i64 %a, i64 %b
@@ -648,13 +825,13 @@
 define i64 @abs_i64(i64 %x) {
 ; CHECK-LABEL: abs_i64:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    bgez a1, .LBB19_2
+; CHECK-NEXT:    bgez a1, .LBB27_2
 ; CHECK-NEXT:  # %bb.1:
 ; CHECK-NEXT:    snez a2, a0
 ; CHECK-NEXT:    neg a0, a0
 ; CHECK-NEXT:    neg a1, a1
 ; CHECK-NEXT:    sub a1, a1, a2
-; CHECK-NEXT:  .LBB19_2:
+; CHECK-NEXT:  .LBB27_2:
 ; CHECK-NEXT:    ret
   %abs = tail call i64 @llvm.abs.i64(i64 %x, i1 true)
   ret i64 %abs
diff --git a/llvm/test/CodeGen/RISCV/rv64zbb.ll b/llvm/test/CodeGen/RISCV/rv64zbb.ll
--- a/llvm/test/CodeGen/RISCV/rv64zbb.ll
+++ b/llvm/test/CodeGen/RISCV/rv64zbb.ll
@@ -578,6 +578,85 @@
   ret i32 %1
 }
 
+define i1 @ctpop_i32_ult_two(i32 signext %a) nounwind {
+; RV64I-LABEL: ctpop_i32_ult_two:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addiw a1, a0, -1
+; RV64I-NEXT:    and a0, a0, a1
+; RV64I-NEXT:    seqz a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64ZBB-LABEL: ctpop_i32_ult_two:
+; RV64ZBB:       # %bb.0:
+; RV64ZBB-NEXT:    cpopw a0, a0
+; RV64ZBB-NEXT:    sltiu a0, a0, 2
+; RV64ZBB-NEXT:    ret
+  %1 = call i32 @llvm.ctpop.i32(i32 %a)
+  %2 = icmp ult i32 %1, 2
+  ret i1 %2
+}
+
+define i1 @ctpop_i32_ugt_one(i32 signext %a) nounwind {
+; RV64I-LABEL: ctpop_i32_ugt_one:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addiw a1, a0, -1
+; RV64I-NEXT:    and a0, a0, a1
+; RV64I-NEXT:    snez a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64ZBB-LABEL: ctpop_i32_ugt_one:
+; RV64ZBB:       # %bb.0:
+; RV64ZBB-NEXT:    addiw a1, a0, -1
+; RV64ZBB-NEXT:    and a0, a0, a1
+; RV64ZBB-NEXT:    snez a0, a0
+; RV64ZBB-NEXT:    ret
+  %1 = call i32 @llvm.ctpop.i32(i32 %a)
+  %2 = icmp ugt i32 %1, 1
+  ret i1 %2
+}
+
+define i1 @ctpop_i32_eq_one(i32 signext %a) nounwind {
+; RV64I-LABEL: ctpop_i32_eq_one:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addiw a1, a0, -1
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    seqz a1, a1
+; RV64I-NEXT:    snez a0, a0
+; RV64I-NEXT:    and a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBB-LABEL: ctpop_i32_eq_one:
+; RV64ZBB:       # %bb.0:
+; RV64ZBB-NEXT:    cpopw a0, a0
+; RV64ZBB-NEXT:    addi a0, a0, -1
+; RV64ZBB-NEXT:    seqz a0, a0
+; RV64ZBB-NEXT:    ret
+  %1 = call i32 @llvm.ctpop.i32(i32 %a)
+  %2 = icmp eq i32 %1, 1
+  ret i1 %2
+}
+
+define i1 @ctpop_i32_ne_one(i32 signext %a) nounwind {
+; RV64I-LABEL: ctpop_i32_ne_one:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addiw a1, a0, -1
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    snez a1, a1
+; RV64I-NEXT:    seqz a0, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBB-LABEL: ctpop_i32_ne_one:
+; RV64ZBB:       # %bb.0:
+; RV64ZBB-NEXT:    cpopw a0, a0
+; RV64ZBB-NEXT:    addi a0, a0, -1
+; RV64ZBB-NEXT:    snez a0, a0
+; RV64ZBB-NEXT:    ret
+  %1 = call i32 @llvm.ctpop.i32(i32 %a)
+  %2 = icmp ne i32 %1, 1
+  ret i1 %2
+}
+
 define signext i32 @ctpop_i32_load(ptr %p) nounwind {
 ; RV64I-LABEL: ctpop_i32_load:
 ; RV64I:       # %bb.0:
@@ -665,6 +744,85 @@
   ret i64 %1
 }
 
+define i1 @ctpop_i64_ugt_two(i64 %a) nounwind {
+; RV64I-LABEL: ctpop_i64_ugt_two:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addi a1, a0, -1
+; RV64I-NEXT:    and a0, a0, a1
+; RV64I-NEXT:    seqz a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64ZBB-LABEL: ctpop_i64_ugt_two:
+; RV64ZBB:       # %bb.0:
+; RV64ZBB-NEXT:    cpop a0, a0
+; RV64ZBB-NEXT:    sltiu a0, a0, 2
+; RV64ZBB-NEXT:    ret
+  %1 = call i64 @llvm.ctpop.i64(i64 %a)
+  %2 = icmp ult i64 %1, 2
+  ret i1 %2
+}
+
+define i1 @ctpop_i64_ugt_one(i64 %a) nounwind {
+; RV64I-LABEL: ctpop_i64_ugt_one:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addi a1, a0, -1
+; RV64I-NEXT:    and a0, a0, a1
+; RV64I-NEXT:    snez a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64ZBB-LABEL: ctpop_i64_ugt_one:
+; RV64ZBB:       # %bb.0:
+; RV64ZBB-NEXT:    addi a1, a0, -1
+; RV64ZBB-NEXT:    and a0, a0, a1
+; RV64ZBB-NEXT:    snez a0, a0
+; RV64ZBB-NEXT:    ret
+  %1 = call i64 @llvm.ctpop.i64(i64 %a)
+  %2 = icmp ugt i64 %1, 1
+  ret i1 %2
+}
+
+define i1 @ctpop_i64_eq_one(i64 %a) nounwind {
+; RV64I-LABEL: ctpop_i64_eq_one:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addi a1, a0, -1
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    seqz a1, a1
+; RV64I-NEXT:    snez a0, a0
+; RV64I-NEXT:    and a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBB-LABEL: ctpop_i64_eq_one:
+; RV64ZBB:       # %bb.0:
+; RV64ZBB-NEXT:    cpop a0, a0
+; RV64ZBB-NEXT:    addi a0, a0, -1
+; RV64ZBB-NEXT:    seqz a0, a0
+; RV64ZBB-NEXT:    ret
+  %1 = call i64 @llvm.ctpop.i64(i64 %a)
+  %2 = icmp eq i64 %1, 1
+  ret i1 %2
+}
+
+define i1 @ctpop_i64_ne_one(i64 %a) nounwind {
+; RV64I-LABEL: ctpop_i64_ne_one:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addi a1, a0, -1
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    snez a1, a1
+; RV64I-NEXT:    seqz a0, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBB-LABEL: ctpop_i64_ne_one:
+; RV64ZBB:       # %bb.0:
+; RV64ZBB-NEXT:    cpop a0, a0
+; RV64ZBB-NEXT:    addi a0, a0, -1
+; RV64ZBB-NEXT:    snez a0, a0
+; RV64ZBB-NEXT:    ret
+  %1 = call i64 @llvm.ctpop.i64(i64 %a)
+  %2 = icmp ne i64 %1, 1
+  ret i1 %2
+}
+
 define signext i32 @sextb_i32(i32 signext %a) nounwind {
 ; RV64I-LABEL: sextb_i32:
 ; RV64I:       # %bb.0:
@@ -732,10 +890,10 @@
 define signext i32 @min_i32(i32 signext %a, i32 signext %b) nounwind {
 ; RV64I-LABEL: min_i32:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    blt a0, a1, .LBB18_2
+; RV64I-NEXT:    blt a0, a1, .LBB26_2
 ; RV64I-NEXT:  # %bb.1:
 ; RV64I-NEXT:    mv a0, a1
-; RV64I-NEXT:  .LBB18_2:
+; RV64I-NEXT:  .LBB26_2:
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: min_i32:
@@ -750,10 +908,10 @@
 define i64 @min_i64(i64 %a, i64 %b) nounwind {
 ; RV64I-LABEL: min_i64:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    blt a0, a1, .LBB19_2
+; RV64I-NEXT:    blt a0, a1, .LBB27_2
 ; RV64I-NEXT:  # %bb.1:
 ; RV64I-NEXT:    mv a0, a1
-; RV64I-NEXT:  .LBB19_2:
+; RV64I-NEXT:  .LBB27_2:
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: min_i64:
@@ -768,10 +926,10 @@
 define signext i32 @max_i32(i32 signext %a, i32 signext %b) nounwind {
 ; RV64I-LABEL: max_i32:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    blt a1, a0, .LBB20_2
+; RV64I-NEXT:    blt a1, a0, .LBB28_2
 ; RV64I-NEXT:  # %bb.1:
 ; RV64I-NEXT:    mv a0, a1
-; RV64I-NEXT:  .LBB20_2:
+; RV64I-NEXT:  .LBB28_2:
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: max_i32:
@@ -786,10 +944,10 @@
 define i64 @max_i64(i64 %a, i64 %b) nounwind {
 ; RV64I-LABEL: max_i64:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    blt a1, a0, .LBB21_2
+; RV64I-NEXT:    blt a1, a0, .LBB29_2
 ; RV64I-NEXT:  # %bb.1:
 ; RV64I-NEXT:    mv a0, a1
-; RV64I-NEXT:  .LBB21_2:
+; RV64I-NEXT:  .LBB29_2:
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: max_i64:
@@ -804,10 +962,10 @@
 define signext i32 @minu_i32(i32 signext %a, i32 signext %b) nounwind {
 ; RV64I-LABEL: minu_i32:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    bltu a0, a1, .LBB22_2
+; RV64I-NEXT:    bltu a0, a1, .LBB30_2
 ; RV64I-NEXT:  # %bb.1:
 ; RV64I-NEXT:    mv a0, a1
-; RV64I-NEXT:  .LBB22_2:
+; RV64I-NEXT:  .LBB30_2:
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: minu_i32:
@@ -822,10 +980,10 @@
 define i64 @minu_i64(i64 %a, i64 %b) nounwind {
 ; RV64I-LABEL: minu_i64:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    bltu a0, a1, .LBB23_2
+; RV64I-NEXT:    bltu a0, a1, .LBB31_2
 ; RV64I-NEXT:  # %bb.1:
 ; RV64I-NEXT:    mv a0, a1
-; RV64I-NEXT:  .LBB23_2:
+; RV64I-NEXT:  .LBB31_2:
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: minu_i64:
@@ -840,10 +998,10 @@
 define signext i32 @maxu_i32(i32 signext %a, i32 signext %b) nounwind {
 ; RV64I-LABEL: maxu_i32:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    bltu a1, a0, .LBB24_2
+; RV64I-NEXT:    bltu a1, a0, .LBB32_2
 ; RV64I-NEXT:  # %bb.1:
 ; RV64I-NEXT:    mv a0, a1
-; RV64I-NEXT:  .LBB24_2:
+; RV64I-NEXT:  .LBB32_2:
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: maxu_i32:
@@ -858,10 +1016,10 @@
 define i64 @maxu_i64(i64 %a, i64 %b) nounwind {
 ; RV64I-LABEL: maxu_i64:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    bltu a1, a0, .LBB25_2
+; RV64I-NEXT:    bltu a1, a0, .LBB33_2
 ; RV64I-NEXT:  # %bb.1:
 ; RV64I-NEXT:    mv a0, a1
-; RV64I-NEXT:  .LBB25_2:
+; RV64I-NEXT:  .LBB33_2:
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-LABEL: maxu_i64:
diff --git a/llvm/test/CodeGen/RISCV/rvv/ctpop-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/ctpop-sdnode.ll
--- a/llvm/test/CodeGen/RISCV/rvv/ctpop-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/ctpop-sdnode.ll
@@ -857,6 +857,91 @@
   %a = call <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32> %va)
   ret <vscale x 16 x i32> %a
 }
+
+define <vscale x 16 x i1> @ctpop_nxv16i32_ult_two(<vscale x 16 x i32> %va) {
+; CHECK-LABEL: ctpop_nxv16i32_ult_two:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-NEXT:    vadd.vi v16, v8, -1
+; CHECK-NEXT:    vand.vv v8, v8, v16
+; CHECK-NEXT:    vmseq.vi v0, v8, 0
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: ctpop_nxv16i32_ult_two:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-ZVBB-NEXT:    vcpop.v v8, v8
+; CHECK-ZVBB-NEXT:    vmsleu.vi v0, v8, 1
+; CHECK-ZVBB-NEXT:    ret
+  %a = call <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32> %va)
+  %cmp = icmp ult <vscale x 16 x i32> %a, shufflevector (<vscale x 16 x i32> insertelement (<vscale x 16 x i32> poison, i32 2, i64 0), <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer)
+  ret <vscale x 16 x i1> %cmp
+}
+
+define <vscale x 16 x i1> @ctpop_nxv16i32_ugt_one(<vscale x 16 x i32> %va) {
+; CHECK-LABEL: ctpop_nxv16i32_ugt_one:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-NEXT:    vadd.vi v16, v8, -1
+; CHECK-NEXT:    vand.vv v8, v8, v16
+; CHECK-NEXT:    vmsne.vi v0, v8, 0
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: ctpop_nxv16i32_ugt_one:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-ZVBB-NEXT:    vcpop.v v8, v8
+; CHECK-ZVBB-NEXT:    vmsgtu.vi v0, v8, 1
+; CHECK-ZVBB-NEXT:    ret
+  %a = call <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32> %va)
+  %cmp = icmp ugt <vscale x 16 x i32> %a, shufflevector (<vscale x 16 x i32> insertelement (<vscale x 16 x i32> poison, i32 1, i64 0), <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer)
+  ret <vscale x 16 x i1> %cmp
+}
+
+define <vscale x 16 x i1> @ctpop_nxv16i32_eq_one(<vscale x 16 x i32> %va) {
+; CHECK-LABEL: ctpop_nxv16i32_eq_one:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-NEXT:    vadd.vi v16, v8, -1
+; CHECK-NEXT:    vand.vv v16, v8, v16
+; CHECK-NEXT:    vmseq.vi v24, v16, 0
+; CHECK-NEXT:    vmsne.vi v16, v8, 0
+; CHECK-NEXT:    vmand.mm v0, v16, v24
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: ctpop_nxv16i32_eq_one:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-ZVBB-NEXT:    vcpop.v v8, v8
+; CHECK-ZVBB-NEXT:    vmseq.vi v0, v8, 1
+; CHECK-ZVBB-NEXT:    ret
+  %a = call <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32> %va)
+  %cmp = icmp eq <vscale x 16 x i32> %a, shufflevector (<vscale x 16 x i32> insertelement (<vscale x 16 x i32> poison, i32 1, i64 0), <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer)
+  ret <vscale x 16 x i1> %cmp
+}
+
+define <vscale x 16 x i1> @ctpop_nxv16i32_ne_one(<vscale x 16 x i32> %va) {
+; CHECK-LABEL: ctpop_nxv16i32_ne_one:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-NEXT:    vadd.vi v16, v8, -1
+; CHECK-NEXT:    vand.vv v16, v8, v16
+; CHECK-NEXT:    vmsne.vi v24, v16, 0
+; CHECK-NEXT:    vmseq.vi v16, v8, 0
+; CHECK-NEXT:    vmor.mm v0, v16, v24
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: ctpop_nxv16i32_ne_one:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-ZVBB-NEXT:    vcpop.v v8, v8
+; CHECK-ZVBB-NEXT:    vmsne.vi v0, v8, 1
+; CHECK-ZVBB-NEXT:    ret
+  %a = call <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32> %va)
+  %cmp = icmp ne <vscale x 16 x i32> %a, shufflevector (<vscale x 16 x i32> insertelement (<vscale x 16 x i32> poison, i32 1, i64 0), <vscale x 16 x i32> poison, <vscale x 16 x i32> zeroinitializer)
+  ret <vscale x 16 x i1> %cmp
+}
+
 declare <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32>)
 
 define <vscale x 1 x i64> @ctpop_nxv1i64(<vscale x 1 x i64> %va) {
@@ -1217,4 +1302,89 @@
   %a = call <vscale x 8 x i64> @llvm.ctpop.nxv8i64(<vscale x 8 x i64> %va)
   ret <vscale x 8 x i64> %a
 }
+
+define <vscale x 8 x i1> @ctpop_nxv8i64_ult_two(<vscale x 8 x i64> %va) {
+; CHECK-LABEL: ctpop_nxv8i64_ult_two:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vadd.vi v16, v8, -1
+; CHECK-NEXT:    vand.vv v8, v8, v16
+; CHECK-NEXT:    vmseq.vi v0, v8, 0
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: ctpop_nxv8i64_ult_two:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-ZVBB-NEXT:    vcpop.v v8, v8
+; CHECK-ZVBB-NEXT:    vmsleu.vi v0, v8, 1
+; CHECK-ZVBB-NEXT:    ret
+  %a = call <vscale x 8 x i64> @llvm.ctpop.nxv8i64(<vscale x 8 x i64> %va)
+  %cmp = icmp ult <vscale x 8 x i64> %a, shufflevector (<vscale x 8 x i64> insertelement (<vscale x 8 x i64> poison, i64 2, i64 0), <vscale x 8 x i64> poison, <vscale x 8 x i32> zeroinitializer)
+  ret <vscale x 8 x i1> %cmp
+}
+
+define <vscale x 8 x i1> @ctpop_nxv8i64_ugt_one(<vscale x 8 x i64> %va) {
+; CHECK-LABEL: ctpop_nxv8i64_ugt_one:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vadd.vi v16, v8, -1
+; CHECK-NEXT:    vand.vv v8, v8, v16
+; CHECK-NEXT:    vmsne.vi v0, v8, 0
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: ctpop_nxv8i64_ugt_one:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-ZVBB-NEXT:    vcpop.v v8, v8
+; CHECK-ZVBB-NEXT:    vmsgtu.vi v0, v8, 1
+; CHECK-ZVBB-NEXT:    ret
+  %a = call <vscale x 8 x i64> @llvm.ctpop.nxv8i64(<vscale x 8 x i64> %va)
+  %cmp = icmp ugt <vscale x 8 x i64> %a, shufflevector (<vscale x 8 x i64> insertelement (<vscale x 8 x i64> poison, i64 1, i64 0), <vscale x 8 x i64> poison, <vscale x 8 x i32> zeroinitializer)
+  ret <vscale x 8 x i1> %cmp
+}
+
+define <vscale x 8 x i1> @ctpop_nxv8i64_eq_one(<vscale x 8 x i64> %va) {
+; CHECK-LABEL: ctpop_nxv8i64_eq_one:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vadd.vi v16, v8, -1
+; CHECK-NEXT:    vand.vv v16, v8, v16
+; CHECK-NEXT:    vmseq.vi v24, v16, 0
+; CHECK-NEXT:    vmsne.vi v16, v8, 0
+; CHECK-NEXT:    vmand.mm v0, v16, v24
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: ctpop_nxv8i64_eq_one:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-ZVBB-NEXT:    vcpop.v v8, v8
+; CHECK-ZVBB-NEXT:    vmseq.vi v0, v8, 1
+; CHECK-ZVBB-NEXT:    ret
+  %a = call <vscale x 8 x i64> @llvm.ctpop.nxv8i64(<vscale x 8 x i64> %va)
+  %cmp = icmp eq <vscale x 8 x i64> %a, shufflevector (<vscale x 8 x i64> insertelement (<vscale x 8 x i64> poison, i64 1, i64 0), <vscale x 8 x i64> poison, <vscale x 8 x i32> zeroinitializer)
+  ret <vscale x 8 x i1> %cmp
+}
+
+define <vscale x 8 x i1> @ctpop_nxv8i64_ne_one(<vscale x 8 x i64> %va) {
+; CHECK-LABEL: ctpop_nxv8i64_ne_one:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-NEXT:    vadd.vi v16, v8, -1
+; CHECK-NEXT:    vand.vv v16, v8, v16
+; CHECK-NEXT:    vmsne.vi v24, v16, 0
+; CHECK-NEXT:    vmseq.vi v16, v8, 0
+; CHECK-NEXT:    vmor.mm v0, v16, v24
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: ctpop_nxv8i64_ne_one:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-ZVBB-NEXT:    vcpop.v v8, v8
+; CHECK-ZVBB-NEXT:    vmsne.vi v0, v8, 1
+; CHECK-ZVBB-NEXT:    ret
+  %a = call <vscale x 8 x i64> @llvm.ctpop.nxv8i64(<vscale x 8 x i64> %va)
+  %cmp = icmp ne <vscale x 8 x i64> %a, shufflevector (<vscale x 8 x i64> insertelement (<vscale x 8 x i64> poison, i64 1, i64 0), <vscale x 8 x i64> poison, <vscale x 8 x i32> zeroinitializer)
+  ret <vscale x 8 x i1> %cmp
+}
+
 declare <vscale x 8 x i64> @llvm.ctpop.nxv8i64(<vscale x 8 x i64>)