diff --git a/llvm/include/llvm/Target/Target.td b/llvm/include/llvm/Target/Target.td
--- a/llvm/include/llvm/Target/Target.td
+++ b/llvm/include/llvm/Target/Target.td
@@ -186,6 +186,10 @@
   bits<16> HWEncoding = 0;
 
   bit isArtificial = false;
+
+  // isConstant - This register always holds a constant value (e.g. the zero
+  // register in architectures such as MIPS)
+  bit isConstant = false;
 }
 
 // RegisterWithSubRegs - This can be used to define instances of Register which
diff --git a/llvm/lib/Target/AArch64/AArch64RegisterInfo.h b/llvm/lib/Target/AArch64/AArch64RegisterInfo.h
--- a/llvm/lib/Target/AArch64/AArch64RegisterInfo.h
+++ b/llvm/lib/Target/AArch64/AArch64RegisterInfo.h
@@ -91,7 +91,6 @@
   BitVector getReservedRegs(const MachineFunction &MF) const override;
   bool isAsmClobberable(const MachineFunction &MF,
                        MCRegister PhysReg) const override;
-  bool isConstantPhysReg(MCRegister PhysReg) const override;
   const TargetRegisterClass *
   getPointerRegClass(const MachineFunction &MF,
                      unsigned Kind = 0) const override;
diff --git a/llvm/lib/Target/AArch64/AArch64RegisterInfo.cpp b/llvm/lib/Target/AArch64/AArch64RegisterInfo.cpp
--- a/llvm/lib/Target/AArch64/AArch64RegisterInfo.cpp
+++ b/llvm/lib/Target/AArch64/AArch64RegisterInfo.cpp
@@ -366,10 +366,6 @@
   return !isReservedReg(MF, PhysReg);
 }
 
-bool AArch64RegisterInfo::isConstantPhysReg(MCRegister PhysReg) const {
-  return PhysReg == AArch64::WZR || PhysReg == AArch64::XZR;
-}
-
 const TargetRegisterClass *
 AArch64RegisterInfo::getPointerRegClass(const MachineFunction &MF,
                                       unsigned Kind) const {
diff --git a/llvm/lib/Target/AArch64/AArch64RegisterInfo.td b/llvm/lib/Target/AArch64/AArch64RegisterInfo.td
--- a/llvm/lib/Target/AArch64/AArch64RegisterInfo.td
+++ b/llvm/lib/Target/AArch64/AArch64RegisterInfo.td
@@ -97,6 +97,7 @@
 def W29   : AArch64Reg<29, "w29">, DwarfRegNum<[29]>;
 def W30   : AArch64Reg<30, "w30">, DwarfRegNum<[30]>;
 def WSP   : AArch64Reg<31, "wsp">, DwarfRegNum<[31]>;
+let isConstant = true in
 def WZR   : AArch64Reg<31, "wzr">, DwarfRegAlias<WSP>;
 
 let SubRegIndices = [sub_32] in {
@@ -132,6 +133,7 @@
 def FP    : AArch64Reg<29, "x29", [W29]>, DwarfRegAlias<W29>;
 def LR    : AArch64Reg<30, "x30", [W30]>, DwarfRegAlias<W30>;
 def SP    : AArch64Reg<31, "sp",  [WSP]>, DwarfRegAlias<WSP>;
+let isConstant = true in
 def XZR   : AArch64Reg<31, "xzr", [WZR]>, DwarfRegAlias<WSP>;
 }
 
diff --git a/llvm/lib/Target/AMDGPU/SIRegisterInfo.h b/llvm/lib/Target/AMDGPU/SIRegisterInfo.h
--- a/llvm/lib/Target/AMDGPU/SIRegisterInfo.h
+++ b/llvm/lib/Target/AMDGPU/SIRegisterInfo.h
@@ -289,8 +289,6 @@
     return isVGPR(MRI, Reg) || isAGPR(MRI, Reg);
   }
 
-  bool isConstantPhysReg(MCRegister PhysReg) const override;
-
   bool isDivergentRegClass(const TargetRegisterClass *RC) const override {
     return !isSGPRClass(RC);
   }
diff --git a/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp b/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
--- a/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
@@ -3083,20 +3083,6 @@
   return RC;
 }
 
-bool SIRegisterInfo::isConstantPhysReg(MCRegister PhysReg) const {
-  switch (PhysReg) {
-  case AMDGPU::SGPR_NULL:
-  case AMDGPU::SGPR_NULL64:
-  case AMDGPU::SRC_SHARED_BASE:
-  case AMDGPU::SRC_PRIVATE_BASE:
-  case AMDGPU::SRC_SHARED_LIMIT:
-  case AMDGPU::SRC_PRIVATE_LIMIT:
-    return true;
-  default:
-    return false;
-  }
-}
-
 ArrayRef<MCPhysReg>
 SIRegisterInfo::getAllSGPR128(const MachineFunction &MF) const {
   return makeArrayRef(AMDGPU::SGPR_128RegClass.begin(),
diff --git a/llvm/lib/Target/AMDGPU/SIRegisterInfo.td b/llvm/lib/Target/AMDGPU/SIRegisterInfo.td
--- a/llvm/lib/Target/AMDGPU/SIRegisterInfo.td
+++ b/llvm/lib/Target/AMDGPU/SIRegisterInfo.td
@@ -219,21 +219,25 @@
 
 defm SGPR_NULL_gfxpre11 : SIRegLoHi16 <"null", 125>;
 defm SGPR_NULL_gfx11plus : SIRegLoHi16 <"null", 124>;
+let isConstant = true in {
 defm SGPR_NULL : SIRegLoHi16 <"null", 0>;
 defm SGPR_NULL_HI : SIRegLoHi16 <"", 0>;
+} // isConstant = true
 
 def SGPR_NULL64 :
     RegisterWithSubRegs<"null", [SGPR_NULL, SGPR_NULL_HI]> {
   let Namespace = "AMDGPU";
   let SubRegIndices = [sub0, sub1];
   let HWEncoding = SGPR_NULL.HWEncoding;
+  let isConstant = true;
 }
-
+let isConstant = true in {
 defm SRC_SHARED_BASE : SIRegLoHi16<"src_shared_base", 235>;
 defm SRC_SHARED_LIMIT : SIRegLoHi16<"src_shared_limit", 236>;
 defm SRC_PRIVATE_BASE : SIRegLoHi16<"src_private_base", 237>;
 defm SRC_PRIVATE_LIMIT : SIRegLoHi16<"src_private_limit", 238>;
 defm SRC_POPS_EXITING_WAVE_ID : SIRegLoHi16<"src_pops_exiting_wave_id", 239>;
+} // isConstant = true
 
 // Not addressable
 def MODE : SIReg <"mode", 0>;
diff --git a/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.h b/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.h
--- a/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.h
+++ b/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.h
@@ -31,7 +31,6 @@
   const uint32_t *getNoPreservedMask() const override;
 
   BitVector getReservedRegs(const MachineFunction &MF) const override;
-  bool isConstantPhysReg(MCRegister PhysReg) const override;
 
   const TargetRegisterClass *
   getPointerRegClass(const MachineFunction &MF,
diff --git a/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.cpp b/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.cpp
--- a/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.cpp
+++ b/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.cpp
@@ -96,10 +96,6 @@
   return Reserved;
 }
 
-bool LoongArchRegisterInfo::isConstantPhysReg(MCRegister PhysReg) const {
-  return PhysReg == LoongArch::R0;
-}
-
 Register
 LoongArchRegisterInfo::getFrameRegister(const MachineFunction &MF) const {
   const TargetFrameLowering *TFI = getFrameLowering(MF);
diff --git a/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.td b/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.td
--- a/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.td
+++ b/llvm/lib/Target/LoongArch/LoongArchRegisterInfo.td
@@ -40,6 +40,7 @@
 // Integer registers
 
 let RegAltNameIndices = [RegAliasName] in {
+  let isConstant = true in
   def R0  : LoongArchReg<0,  "r0", ["zero"]>, DwarfRegNum<[0]>;
   def R1  : LoongArchReg<1,  "r1", ["ra"]>, DwarfRegNum<[1]>;
   def R2  : LoongArchReg<2,  "r2", ["tp"]>, DwarfRegNum<[2]>;
diff --git a/llvm/lib/Target/Mips/MipsRegisterInfo.h b/llvm/lib/Target/Mips/MipsRegisterInfo.h
--- a/llvm/lib/Target/Mips/MipsRegisterInfo.h
+++ b/llvm/lib/Target/Mips/MipsRegisterInfo.h
@@ -69,8 +69,6 @@
   /// Debug information queries.
   Register getFrameRegister(const MachineFunction &MF) const override;
 
-  bool isConstantPhysReg(MCRegister PhysReg) const override;
-
   /// Return GPR register class.
   virtual const TargetRegisterClass *intRegClass(unsigned Size) const = 0;
 
diff --git a/llvm/lib/Target/Mips/MipsRegisterInfo.cpp b/llvm/lib/Target/Mips/MipsRegisterInfo.cpp
--- a/llvm/lib/Target/Mips/MipsRegisterInfo.cpp
+++ b/llvm/lib/Target/Mips/MipsRegisterInfo.cpp
@@ -318,7 +318,3 @@
   // sized objects.
   return MF.getRegInfo().canReserveReg(BP);
 }
-
-bool MipsRegisterInfo::isConstantPhysReg(MCRegister PhysReg) const {
-  return PhysReg == Mips::ZERO_64 || PhysReg == Mips::ZERO;
-}
diff --git a/llvm/lib/Target/Mips/MipsRegisterInfo.td b/llvm/lib/Target/Mips/MipsRegisterInfo.td
--- a/llvm/lib/Target/Mips/MipsRegisterInfo.td
+++ b/llvm/lib/Target/Mips/MipsRegisterInfo.td
@@ -84,6 +84,7 @@
 
 let Namespace = "Mips" in {
   // General Purpose Registers
+  let isConstant = true in
   def ZERO : MipsGPRReg< 0, "zero">, DwarfRegNum<[0]>;
   def AT   : MipsGPRReg< 1, "1">,    DwarfRegNum<[1]>;
   def V0   : MipsGPRReg< 2, "2">,    DwarfRegNum<[2]>;
@@ -118,6 +119,7 @@
   def RA   : MipsGPRReg< 31, "ra">,  DwarfRegNum<[31]>;
 
   // General Purpose 64-bit Registers
+  let isConstant = true in
   def ZERO_64 : Mips64GPRReg< 0, "zero", [ZERO]>, DwarfRegNum<[0]>;
   def AT_64   : Mips64GPRReg< 1, "1",    [AT]>, DwarfRegNum<[1]>;
   def V0_64   : Mips64GPRReg< 2, "2",    [V0]>, DwarfRegNum<[2]>;
diff --git a/llvm/lib/Target/PowerPC/PPCRegisterInfo.td b/llvm/lib/Target/PowerPC/PPCRegisterInfo.td
--- a/llvm/lib/Target/PowerPC/PPCRegisterInfo.td
+++ b/llvm/lib/Target/PowerPC/PPCRegisterInfo.td
@@ -185,8 +185,10 @@
 }
 
 // The representation of r0 when treated as the constant 0.
+let isConstant = true in {
 def ZERO  : GPR<0, "0">,    DwarfRegAlias<R0>;
 def ZERO8 : GP8<ZERO, "0">, DwarfRegAlias<X0>;
+} // isConstant = true
 
 // Representations of the frame pointer used by ISD::FRAMEADDR.
 def FP   : GPR<0 /* arbitrary */, "**FRAME POINTER**">;
diff --git a/llvm/lib/Target/RISCV/RISCVRegisterInfo.h b/llvm/lib/Target/RISCV/RISCVRegisterInfo.h
--- a/llvm/lib/Target/RISCV/RISCVRegisterInfo.h
+++ b/llvm/lib/Target/RISCV/RISCVRegisterInfo.h
@@ -33,8 +33,6 @@
   bool isAsmClobberable(const MachineFunction &MF,
                         MCRegister PhysReg) const override;
 
-  bool isConstantPhysReg(MCRegister PhysReg) const override;
-
   const uint32_t *getNoPreservedMask() const override;
 
   bool hasReservedSpillSlot(const MachineFunction &MF, Register Reg,
diff --git a/llvm/lib/Target/RISCV/RISCVRegisterInfo.cpp b/llvm/lib/Target/RISCV/RISCVRegisterInfo.cpp
--- a/llvm/lib/Target/RISCV/RISCVRegisterInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVRegisterInfo.cpp
@@ -117,10 +117,6 @@
   return !MF.getSubtarget<RISCVSubtarget>().isRegisterReservedByUser(PhysReg);
 }
 
-bool RISCVRegisterInfo::isConstantPhysReg(MCRegister PhysReg) const {
-  return PhysReg == RISCV::X0 || PhysReg == RISCV::VLENB;
-}
-
 const uint32_t *RISCVRegisterInfo::getNoPreservedMask() const {
   return CSR_NoRegs_RegMask;
 }
diff --git a/llvm/lib/Target/RISCV/RISCVRegisterInfo.td b/llvm/lib/Target/RISCV/RISCVRegisterInfo.td
--- a/llvm/lib/Target/RISCV/RISCVRegisterInfo.td
+++ b/llvm/lib/Target/RISCV/RISCVRegisterInfo.td
@@ -77,6 +77,7 @@
 // instructions.
 
 let RegAltNameIndices = [ABIRegAltName] in {
+  let isConstant = true in
   def X0  : RISCVReg<0, "x0", ["zero"]>, DwarfRegNum<[0]>;
   let CostPerUse = [0, 1] in {
   def X1  : RISCVReg<1, "x1", ["ra"]>, DwarfRegNum<[1]>;
@@ -458,6 +459,7 @@
   def VL     : RISCVReg<0, "vl", ["vl"]>;
   def VXSAT  : RISCVReg<0, "vxsat", ["vxsat"]>;
   def VXRM   : RISCVReg<0, "vxrm", ["vxrm"]>;
+  let isConstant = true in
   def VLENB  : RISCVReg<0, "vlenb", ["vlenb"]>,
                DwarfRegNum<[!add(4096, SysRegVLENB.Encoding)]>;
 }
diff --git a/llvm/lib/Target/VE/VERegisterInfo.h b/llvm/lib/Target/VE/VERegisterInfo.h
--- a/llvm/lib/Target/VE/VERegisterInfo.h
+++ b/llvm/lib/Target/VE/VERegisterInfo.h
@@ -30,7 +30,6 @@
   const uint32_t *getNoPreservedMask() const override;
 
   BitVector getReservedRegs(const MachineFunction &MF) const override;
-  bool isConstantPhysReg(MCRegister PhysReg) const override;
 
   const TargetRegisterClass *getPointerRegClass(const MachineFunction &MF,
                                                 unsigned Kind) const override;
diff --git a/llvm/lib/Target/VE/VERegisterInfo.cpp b/llvm/lib/Target/VE/VERegisterInfo.cpp
--- a/llvm/lib/Target/VE/VERegisterInfo.cpp
+++ b/llvm/lib/Target/VE/VERegisterInfo.cpp
@@ -96,16 +96,6 @@
   return Reserved;
 }
 
-bool VERegisterInfo::isConstantPhysReg(MCRegister PhysReg) const {
-  switch (PhysReg) {
-  case VE::VM0:
-  case VE::VMP0:
-    return true;
-  default:
-    return false;
-  }
-}
-
 const TargetRegisterClass *
 VERegisterInfo::getPointerRegClass(const MachineFunction &MF,
                                    unsigned Kind) const {
diff --git a/llvm/lib/Target/VE/VERegisterInfo.td b/llvm/lib/Target/VE/VERegisterInfo.td
--- a/llvm/lib/Target/VE/VERegisterInfo.td
+++ b/llvm/lib/Target/VE/VERegisterInfo.td
@@ -148,10 +148,13 @@
 def VIX : VEVecReg<255, "vix", [], ["vix"]>;
 
 // Vector mask registers - 256 bits wide
-foreach I = 0-15 in
+let isConstant = true in
+def VM0 : VEMaskReg<0, "vm0", [], ["vm0"]>, DwarfRegNum<[128]>;
+foreach I = 1-15 in
   def VM#I : VEMaskReg<I, "vm"#I, [], ["vm"#I]>, DwarfRegNum<[!add(128,I)]>;
 
 // Aliases of VMs to use as a pair of two VM for packed instructions
+let isConstant = true in
 def VMP0 : VEMaskReg<0, "vm0", [], ["vm0"]>;
 
 let SubRegIndices = [sub_vm_even, sub_vm_odd], CoveredBySubRegs = 1 in
diff --git a/llvm/utils/TableGen/CodeGenRegisters.h b/llvm/utils/TableGen/CodeGenRegisters.h
--- a/llvm/utils/TableGen/CodeGenRegisters.h
+++ b/llvm/utils/TableGen/CodeGenRegisters.h
@@ -154,6 +154,7 @@
     bool CoveredBySubRegs;
     bool HasDisjunctSubRegs;
     bool Artificial;
+    bool Constant;
 
     // Map SubRegIndex -> Register.
     typedef std::map<CodeGenSubRegIndex *, CodeGenRegister *,
diff --git a/llvm/utils/TableGen/CodeGenRegisters.cpp b/llvm/utils/TableGen/CodeGenRegisters.cpp
--- a/llvm/utils/TableGen/CodeGenRegisters.cpp
+++ b/llvm/utils/TableGen/CodeGenRegisters.cpp
@@ -154,8 +154,8 @@
     : TheDef(R), EnumValue(Enum),
       CostPerUse(R->getValueAsListOfInts("CostPerUse")),
       CoveredBySubRegs(R->getValueAsBit("CoveredBySubRegs")),
-      HasDisjunctSubRegs(false), SubRegsComplete(false),
-      SuperRegsComplete(false), TopoSig(~0u) {
+      HasDisjunctSubRegs(false), Constant(R->getValueAsBit("isConstant")),
+      SubRegsComplete(false), SuperRegsComplete(false), TopoSig(~0u) {
   Artificial = R->getValueAsBit("isArtificial");
 }
 
diff --git a/llvm/utils/TableGen/RegisterInfoEmitter.cpp b/llvm/utils/TableGen/RegisterInfoEmitter.cpp
--- a/llvm/utils/TableGen/RegisterInfoEmitter.cpp
+++ b/llvm/utils/TableGen/RegisterInfoEmitter.cpp
@@ -1190,6 +1190,7 @@
      << "MCRegister) const override;\n"
      << "  bool isArgumentRegister(const MachineFunction &, "
      << "MCRegister) const override;\n"
+     << "  bool isConstantPhysReg(MCRegister PhysReg) const override final;\n"
      << "  /// Devirtualized TargetFrameLowering.\n"
      << "  static const " << TargetName << "FrameLowering *getFrameLowering(\n"
      << "      const MachineFunction &MF);\n"
@@ -1678,6 +1679,15 @@
   OS << "      false;\n";
   OS << "}\n\n";
 
+  OS << "bool " << ClassName << "::\n"
+     << "isConstantPhysReg(MCRegister PhysReg) const {\n"
+     << "  return\n";
+  for (const auto &Reg : Regs)
+    if (Reg.Constant)
+      OS << "      PhysReg == " << getQualifiedName(Reg.TheDef) << " ||\n";
+  OS << "      false;\n";
+  OS << "}\n\n";
+
   OS << "ArrayRef<const char *> " << ClassName
      << "::getRegMaskNames() const {\n";
   if (!CSRSets.empty()) {