diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
index f5f4d71236fee..a48dd0e5fedba 100644
--- a/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
@@ -3356,6 +3356,32 @@ bool SelectionDAGLegalize::ExpandNode(SDNode *Node) {
     Results.push_back(Op);
     break;
   }
+  case ISD::FCANONICALIZE: {
+    // This implements llvm.canonicalize.f* by multiplication with 1.0, as
+    // suggested in
+    // https://llvm.org/docs/LangRef.html#llvm-canonicalize-intrinsic.
+    // It uses strict_fp operations even outside a strict_fp context in order
+    // to guarantee that the canonicalization is not optimized away by later
+    // passes. The result chain introduced by that is intentionally ignored
+    // since no ordering requirement is intended here.
+
+    // Create strict multiplication by 1.0.
+    SDValue Operand = Node->getOperand(0);
+    EVT VT = Operand.getValueType();
+    SDValue One = DAG.getConstantFP(1.0, dl, VT);
+    SDValue Chain = DAG.getEntryNode();
+    SDValue Mul = DAG.getNode(ISD::STRICT_FMUL, dl, {VT, MVT::Other},
+                              {Chain, Operand, One});
+
+    // Propagate existing flags on canonicalize, and additionally set
+    // NoFPExcept.
+    SDNodeFlags CanonicalizeFlags = Node->getFlags();
+    CanonicalizeFlags.setNoFPExcept(true);
+    Mul->setFlags(CanonicalizeFlags);
+
+    Results.push_back(Mul);
+    break;
+  }
   case ISD::SIGN_EXTEND_INREG: {
     EVT ExtraVT = cast<VTSDNode>(Node->getOperand(1))->getVT();
     EVT VT = Node->getValueType(0);
diff --git a/llvm/lib/CodeGen/TargetLoweringBase.cpp b/llvm/lib/CodeGen/TargetLoweringBase.cpp
index 4f722005f4056..c9f927ade4f1f 100644
--- a/llvm/lib/CodeGen/TargetLoweringBase.cpp
+++ b/llvm/lib/CodeGen/TargetLoweringBase.cpp
@@ -875,6 +875,10 @@ void TargetLoweringBase::initActions() {
                       ISD::FATAN2},
                      {MVT::f32, MVT::f64, MVT::f128}, Expand);
 
+  // Insert custom handling default for llvm.canonicalize.*.
+  setOperationAction(ISD::FCANONICALIZE,
+                     {MVT::f16, MVT::f32, MVT::f64, MVT::f128}, Expand);
+
   // FIXME: Query RuntimeLibCalls to make the decision.
   setOperationAction({ISD::LRINT, ISD::LLRINT, ISD::LROUND, ISD::LLROUND},
                      {MVT::f32, MVT::f64, MVT::f128}, LibCall);
diff --git a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
index 0d388fc3c787d..cdb68684b3856 100644
--- a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
+++ b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
@@ -768,6 +768,12 @@ AArch64TargetLowering::AArch64TargetLowering(const TargetMachine &TM,
     setOperationAction(Op, MVT::v8bf16, Expand);
   }
 
+  // Legalize fcanonicalize to circumvent default expansion
+  setOperationAction(ISD::FCANONICALIZE, {MVT::f32, MVT::f64}, Legal);
+  if (Subtarget->hasFullFP16()) {
+    setOperationAction(ISD::FCANONICALIZE, MVT::f16, Legal);
+  }
+
   // fpextend from f16 or bf16 to f32 is legal
   setOperationAction(ISD::FP_EXTEND, MVT::f32, Legal);
   setOperationAction(ISD::FP_EXTEND, MVT::v4f32, Legal);
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
index e32accaba85fc..26be37046e2e3 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
@@ -423,6 +423,11 @@ AMDGPUTargetLowering::AMDGPUTargetLowering(const TargetMachine &TM,
   setOperationAction({ISD::FLOG10, ISD::FLOG, ISD::FEXP, ISD::FEXP10}, MVT::f16,
                      Custom);
 
+  setOperationAction(ISD::FCANONICALIZE, {MVT::f32, MVT::f64}, Legal);
+  if (Subtarget->has16BitInsts()) {
+    setOperationAction(ISD::FCANONICALIZE, MVT::f16, Legal);
+  }
+
   // FIXME: These IS_FPCLASS vector fp types are marked custom so it reaches
   // scalarization code. Can be removed when IS_FPCLASS expand isn't called by
   // default unless marked custom/legal.
diff --git a/llvm/lib/Target/LoongArch/LoongArchISelLowering.cpp b/llvm/lib/Target/LoongArch/LoongArchISelLowering.cpp
index 602ce240814e9..72dbb44815657 100644
--- a/llvm/lib/Target/LoongArch/LoongArchISelLowering.cpp
+++ b/llvm/lib/Target/LoongArch/LoongArchISelLowering.cpp
@@ -195,6 +195,7 @@ LoongArchTargetLowering::LoongArchTargetLowering(const TargetMachine &TM,
     setOperationAction(ISD::FMINNUM, MVT::f32, Legal);
     setOperationAction(ISD::FMAXNUM_IEEE, MVT::f32, Legal);
     setOperationAction(ISD::FMAXNUM, MVT::f32, Legal);
+    setOperationAction(ISD::FCANONICALIZE, MVT::f32, Legal);
     setOperationAction(ISD::STRICT_FSETCCS, MVT::f32, Legal);
     setOperationAction(ISD::STRICT_FSETCC, MVT::f32, Legal);
     setOperationAction(ISD::IS_FPCLASS, MVT::f32, Legal);
@@ -242,6 +243,7 @@ LoongArchTargetLowering::LoongArchTargetLowering(const TargetMachine &TM,
     setOperationAction(ISD::FMINNUM_IEEE, MVT::f64, Legal);
     setOperationAction(ISD::FMINNUM, MVT::f64, Legal);
     setOperationAction(ISD::FMAXNUM_IEEE, MVT::f64, Legal);
+    setOperationAction(ISD::FCANONICALIZE, MVT::f64, Legal);
     setOperationAction(ISD::FMAXNUM, MVT::f64, Legal);
     setOperationAction(ISD::IS_FPCLASS, MVT::f64, Legal);
     setOperationAction(ISD::FSIN, MVT::f64, Expand);
diff --git a/llvm/lib/Target/Mips/MipsISelLowering.cpp b/llvm/lib/Target/Mips/MipsISelLowering.cpp
index 5a15b7d9849d3..74a256e9729b3 100644
--- a/llvm/lib/Target/Mips/MipsISelLowering.cpp
+++ b/llvm/lib/Target/Mips/MipsISelLowering.cpp
@@ -373,6 +373,8 @@ MipsTargetLowering::MipsTargetLowering(const MipsTargetMachine &TM,
     setOperationAction(ISD::FMAXNUM, MVT::f64, Legal);
     setOperationAction(ISD::IS_FPCLASS, MVT::f32, Legal);
     setOperationAction(ISD::IS_FPCLASS, MVT::f64, Legal);
+    setOperationAction(ISD::FCANONICALIZE, MVT::f32, Legal);
+    setOperationAction(ISD::FCANONICALIZE, MVT::f64, Legal);
   } else {
     setOperationAction(ISD::FCANONICALIZE, MVT::f32, Custom);
     setOperationAction(ISD::FCANONICALIZE, MVT::f64, Custom);
diff --git a/llvm/lib/Target/PowerPC/PPCISelLowering.cpp b/llvm/lib/Target/PowerPC/PPCISelLowering.cpp
index 3616b86a7fd3f..b96505816dee8 100644
--- a/llvm/lib/Target/PowerPC/PPCISelLowering.cpp
+++ b/llvm/lib/Target/PowerPC/PPCISelLowering.cpp
@@ -777,6 +777,8 @@ PPCTargetLowering::PPCTargetLowering(const PPCTargetMachine &TM,
     setOperationAction(ISD::FMAXNUM_IEEE, MVT::f32, Legal);
     setOperationAction(ISD::FMINNUM_IEEE, MVT::f64, Legal);
     setOperationAction(ISD::FMINNUM_IEEE, MVT::f32, Legal);
+    setOperationAction(ISD::FCANONICALIZE, MVT::f64, Legal);
+    setOperationAction(ISD::FCANONICALIZE, MVT::f32, Legal);
   }
 
   if (Subtarget.hasAltivec()) {
diff --git a/llvm/test/CodeGen/SystemZ/canonicalize-vars.ll b/llvm/test/CodeGen/SystemZ/canonicalize-vars.ll
new file mode 100644
index 0000000000000..e02f931c4d31e
--- /dev/null
+++ b/llvm/test/CodeGen/SystemZ/canonicalize-vars.ll
@@ -0,0 +1,405 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --default-march s390x-unknown-linux-gnu --version 5
+; RUN: llc -mtriple=s390x-linux-gnu -mcpu=z16 < %s | FileCheck %s -check-prefixes=Z16
+
+define half @canonicalize_fp16(half %a) nounwind {
+; Z16-LABEL: canonicalize_fp16:
+; Z16:       # %bb.0:
+; Z16-NEXT:    stmg %r14, %r15, 112(%r15)
+; Z16-NEXT:    aghi %r15, -160
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    lmg %r14, %r15, 272(%r15)
+; Z16-NEXT:    br %r14
+  %canonicalized = call half @llvm.canonicalize.f16(half %a)
+  ret half %canonicalized
+}
+define float @canonicalize_fp32(float %a) {
+; Z16-LABEL: canonicalize_fp32:
+; Z16:       # %bb.0:
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    br %r14
+  %canonicalized = call float @llvm.canonicalize.f32(float %a)
+  ret float %canonicalized
+}
+
+define double @canonicalize_fp64(double %a) {
+; Z16-LABEL: canonicalize_fp64:
+; Z16:       # %bb.0:
+; Z16-NEXT:    vgmg %v1, 2, 11
+; Z16-NEXT:    mdbr %f0, %f1
+; Z16-NEXT:    br %r14
+  %canonicalized = call double @llvm.canonicalize.f64(double %a)
+  ret double %canonicalized
+}
+
+define fp128 @canonicalize_fp128(fp128 %a) {
+; Z16-LABEL: canonicalize_fp128:
+; Z16:       # %bb.0:
+; Z16-NEXT:    larl %r1, .LCPI3_0
+; Z16-NEXT:    vl %v0, 0(%r3), 3
+; Z16-NEXT:    vl %v1, 0(%r1), 3
+; Z16-NEXT:    wfmxb %v0, %v0, %v1
+; Z16-NEXT:    vst %v0, 0(%r2), 3
+; Z16-NEXT:    br %r14
+  %canonicalized = call fp128 @llvm.canonicalize.f128(fp128 %a)
+  ret fp128 %canonicalized
+}
+
+define void @canonicalize_ptr_f16(ptr %out) nounwind {
+; Z16-LABEL: canonicalize_ptr_f16:
+; Z16:       # %bb.0:
+; Z16-NEXT:    stmg %r13, %r15, 104(%r15)
+; Z16-NEXT:    aghi %r15, -160
+; Z16-NEXT:    vlreph %v0, 0(%r2)
+; Z16-NEXT:    lgr %r13, %r2
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    vsteh %v0, 0(%r13), 0
+; Z16-NEXT:    lmg %r13, %r15, 264(%r15)
+; Z16-NEXT:    br %r14
+  %val = load half, ptr %out
+  %canonicalized = call half @llvm.canonicalize.f16(half %val)
+  store half %canonicalized, ptr %out
+  ret void
+}
+
+define void @canonicalize_ptr_f32(ptr %out) {
+; Z16-LABEL: canonicalize_ptr_f32:
+; Z16:       # %bb.0:
+; Z16-NEXT:    vgmf %v0, 2, 8
+; Z16-NEXT:    meeb %f0, 0(%r2)
+; Z16-NEXT:    ste %f0, 0(%r2)
+; Z16-NEXT:    br %r14
+  %val = load float, ptr %out
+  %canonicalized = call float @llvm.canonicalize.f32(float %val)
+  store float %canonicalized, ptr %out
+  ret void
+}
+
+define void @canonicalize_ptr_f64(ptr %out) {
+; Z16-LABEL: canonicalize_ptr_f64:
+; Z16:       # %bb.0:
+; Z16-NEXT:    vgmg %v0, 2, 11
+; Z16-NEXT:    mdb %f0, 0(%r2)
+; Z16-NEXT:    std %f0, 0(%r2)
+; Z16-NEXT:    br %r14
+  %val = load double, ptr %out
+  %canonicalized = call double @llvm.canonicalize.f64(double %val)
+  store double %canonicalized, ptr %out
+  ret void
+}
+
+define void @canonicalize_ptr_f128(ptr %out) {
+; Z16-LABEL: canonicalize_ptr_f128:
+; Z16:       # %bb.0:
+; Z16-NEXT:    larl %r1, .LCPI7_0
+; Z16-NEXT:    vl %v0, 0(%r2), 3
+; Z16-NEXT:    vl %v1, 0(%r1), 3
+; Z16-NEXT:    wfmxb %v0, %v0, %v1
+; Z16-NEXT:    vst %v0, 0(%r2), 3
+; Z16-NEXT:    br %r14
+  %val = load fp128, ptr %out
+  %canonicalized = call fp128 @llvm.canonicalize.f128(fp128 %val)
+  store fp128 %canonicalized, ptr %out
+  ret void
+}
+define <8 x half> @canonicalize_v8f16(<8 x half> %a) nounwind {
+; Z16-LABEL: canonicalize_v8f16:
+; Z16:       # %bb.0:
+; Z16-NEXT:    stmg %r13, %r15, 104(%r15)
+; Z16-NEXT:    aghi %r15, -224
+; Z16-NEXT:    std %f8, 216(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f9, 208(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f10, 200(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f11, 192(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f12, 184(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f13, 176(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f14, 168(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f15, 160(%r15) # 8-byte Spill
+; Z16-NEXT:    vlreph %v11, 414(%r15)
+; Z16-NEXT:    vlreph %v12, 406(%r15)
+; Z16-NEXT:    vlreph %v13, 398(%r15)
+; Z16-NEXT:    vlreph %v14, 390(%r15)
+; Z16-NEXT:    ldr %f8, %f6
+; Z16-NEXT:    ldr %f9, %f4
+; Z16-NEXT:    ldr %f10, %f2
+; Z16-NEXT:    lgr %r13, %r2
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f15, %f0
+; Z16-NEXT:    ldr %f0, %f10
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f10, %f0
+; Z16-NEXT:    ldr %f0, %f9
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f9, %f0
+; Z16-NEXT:    ldr %f0, %f8
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f8, %f0
+; Z16-NEXT:    ldr %f0, %f14
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f14, %f0
+; Z16-NEXT:    ldr %f0, %f13
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f13, %f0
+; Z16-NEXT:    ldr %f0, %f12
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f12, %f0
+; Z16-NEXT:    ldr %f0, %f11
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    vsteh %v0, 14(%r13), 0
+; Z16-NEXT:    vsteh %v12, 12(%r13), 0
+; Z16-NEXT:    vsteh %v13, 10(%r13), 0
+; Z16-NEXT:    vsteh %v14, 8(%r13), 0
+; Z16-NEXT:    vsteh %v8, 6(%r13), 0
+; Z16-NEXT:    vsteh %v9, 4(%r13), 0
+; Z16-NEXT:    vsteh %v10, 2(%r13), 0
+; Z16-NEXT:    vsteh %v15, 0(%r13), 0
+; Z16-NEXT:    ld %f8, 216(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f9, 208(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f10, 200(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f11, 192(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f12, 184(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f13, 176(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f14, 168(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f15, 160(%r15) # 8-byte Reload
+; Z16-NEXT:    lmg %r13, %r15, 328(%r15)
+; Z16-NEXT:    br %r14
+  %canonicalized = call <8 x half> @llvm.canonicalize.v8f16(<8 x half> %a)
+  ret <8 x half> %canonicalized
+}
+define <4 x float> @canonicalize_v4f32(<4 x float> %a) {
+; Z16-LABEL: canonicalize_v4f32:
+; Z16:       # %bb.0:
+; Z16-NEXT:    vrepf %v0, %v24, 3
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    vrepf %v2, %v24, 2
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    meebr %f2, %f1
+; Z16-NEXT:    vrepf %v3, %v24, 1
+; Z16-NEXT:    vmrhf %v0, %v2, %v0
+; Z16-NEXT:    wfmsb %f2, %v24, %f1
+; Z16-NEXT:    wfmsb %f1, %f3, %f1
+; Z16-NEXT:    vmrhf %v1, %v2, %v1
+; Z16-NEXT:    vmrhg %v24, %v1, %v0
+; Z16-NEXT:    br %r14
+  %canonicalized = call <4 x float> @llvm.canonicalize.v4f32(<4 x float> %a)
+  ret <4 x float> %canonicalized
+}
+
+define <4 x double> @canonicalize_v4f64(<4 x double> %a) {
+; Z16-LABEL: canonicalize_v4f64:
+; Z16:       # %bb.0:
+; Z16-NEXT:    vgmg %v0, 2, 11
+; Z16-NEXT:    vrepg %v2, %v24, 1
+; Z16-NEXT:    wfmdb %f1, %v24, %f0
+; Z16-NEXT:    mdbr %f2, %f0
+; Z16-NEXT:    vmrhg %v24, %v1, %v2
+; Z16-NEXT:    vrepg %v2, %v26, 1
+; Z16-NEXT:    wfmdb %f1, %v26, %f0
+; Z16-NEXT:    wfmdb %f0, %f2, %f0
+; Z16-NEXT:    vmrhg %v26, %v1, %v0
+; Z16-NEXT:    br %r14
+  %canonicalized = call <4 x double> @llvm.canonicalize.v4f64(<4 x double> %a)
+  ret <4 x double> %canonicalized
+}
+
+define <2 x fp128> @canonicalize_v2f128(<2 x fp128> %a) {
+; Z16-LABEL: canonicalize_v2f128:
+; Z16:       # %bb.0:
+; Z16-NEXT:    larl %r1, .LCPI11_0
+; Z16-NEXT:    vl %v0, 16(%r3), 3
+; Z16-NEXT:    vl %v1, 0(%r3), 3
+; Z16-NEXT:    vl %v2, 0(%r1), 3
+; Z16-NEXT:    wfmxb %v1, %v1, %v2
+; Z16-NEXT:    wfmxb %v0, %v0, %v2
+; Z16-NEXT:    vst %v0, 16(%r2), 4
+; Z16-NEXT:    vst %v1, 0(%r2), 4
+; Z16-NEXT:    br %r14
+  %canonicalized = call <2 x fp128> @llvm.canonicalize.v2f128(<2 x fp128> %a)
+  ret <2 x fp128> %canonicalized
+}
+
+define void @canonicalize_ptr_v8f16(ptr %out) nounwind {
+; Z16-LABEL: canonicalize_ptr_v8f16:
+; Z16:       # %bb.0:
+; Z16-NEXT:    stmg %r13, %r15, 104(%r15)
+; Z16-NEXT:    aghi %r15, -224
+; Z16-NEXT:    std %f8, 216(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f9, 208(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f10, 200(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f11, 192(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f12, 184(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f13, 176(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f14, 168(%r15) # 8-byte Spill
+; Z16-NEXT:    std %f15, 160(%r15) # 8-byte Spill
+; Z16-NEXT:    vlreph %v0, 0(%r2)
+; Z16-NEXT:    vlreph %v8, 14(%r2)
+; Z16-NEXT:    vlreph %v9, 12(%r2)
+; Z16-NEXT:    vlreph %v10, 10(%r2)
+; Z16-NEXT:    lgr %r13, %r2
+; Z16-NEXT:    vlreph %v11, 8(%r2)
+; Z16-NEXT:    vlreph %v12, 6(%r2)
+; Z16-NEXT:    vlreph %v13, 4(%r2)
+; Z16-NEXT:    vlreph %v14, 2(%r2)
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f15, %f0
+; Z16-NEXT:    ldr %f0, %f14
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f14, %f0
+; Z16-NEXT:    ldr %f0, %f13
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f13, %f0
+; Z16-NEXT:    ldr %f0, %f12
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f12, %f0
+; Z16-NEXT:    ldr %f0, %f11
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f11, %f0
+; Z16-NEXT:    ldr %f0, %f10
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f10, %f0
+; Z16-NEXT:    ldr %f0, %f9
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    ldr %f9, %f0
+; Z16-NEXT:    ldr %f0, %f8
+; Z16-NEXT:    brasl %r14, __extendhfsf2@PLT
+; Z16-NEXT:    vgmf %v1, 2, 8
+; Z16-NEXT:    meebr %f0, %f1
+; Z16-NEXT:    brasl %r14, __truncsfhf2@PLT
+; Z16-NEXT:    vsteh %v9, 12(%r13), 0
+; Z16-NEXT:    vsteh %v10, 10(%r13), 0
+; Z16-NEXT:    vsteh %v11, 8(%r13), 0
+; Z16-NEXT:    vsteh %v12, 6(%r13), 0
+; Z16-NEXT:    vsteh %v13, 4(%r13), 0
+; Z16-NEXT:    vsteh %v14, 2(%r13), 0
+; Z16-NEXT:    vsteh %v15, 0(%r13), 0
+; Z16-NEXT:    ld %f8, 216(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f9, 208(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f10, 200(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f11, 192(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f12, 184(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f13, 176(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f14, 168(%r15) # 8-byte Reload
+; Z16-NEXT:    ld %f15, 160(%r15) # 8-byte Reload
+; Z16-NEXT:    vsteh %v0, 14(%r13), 0
+; Z16-NEXT:    lmg %r13, %r15, 328(%r15)
+; Z16-NEXT:    br %r14
+  %val = load <8 x half>, ptr %out
+  %canonicalized = call <8 x half> @llvm.canonicalize.v8f16(<8 x half> %val)
+  store <8 x half> %canonicalized, ptr %out
+  ret void
+}
+
+define void @canonicalize_ptr_v4f32(ptr %out) {
+; Z16-LABEL: canonicalize_ptr_v4f32:
+; Z16:       # %bb.0:
+; Z16-NEXT:    vl %v0, 0(%r2), 3
+; Z16-NEXT:    vrepf %v1, %v0, 3
+; Z16-NEXT:    vgmf %v2, 2, 8
+; Z16-NEXT:    vrepf %v3, %v0, 2
+; Z16-NEXT:    meebr %f1, %f2
+; Z16-NEXT:    meebr %f3, %f2
+; Z16-NEXT:    vmrhf %v1, %v3, %v1
+; Z16-NEXT:    wfmsb %f3, %f0, %f2
+; Z16-NEXT:    vrepf %v0, %v0, 1
+; Z16-NEXT:    meebr %f0, %f2
+; Z16-NEXT:    vmrhf %v0, %v3, %v0
+; Z16-NEXT:    vmrhg %v0, %v0, %v1
+; Z16-NEXT:    vst %v0, 0(%r2), 3
+; Z16-NEXT:    br %r14
+  %val = load <4 x float>, ptr %out
+  %canonicalized = call <4 x float> @llvm.canonicalize.v4f32(<4 x float> %val)
+  store <4 x float> %canonicalized, ptr %out
+  ret void
+}
+
+define void @canonicalize_ptr_v4f64(ptr %out) {
+; Z16-LABEL: canonicalize_ptr_v4f64:
+; Z16:       # %bb.0:
+; Z16-NEXT:    vl %v1, 16(%r2), 4
+; Z16-NEXT:    vgmg %v2, 2, 11
+; Z16-NEXT:    wfmdb %f3, %f1, %f2
+; Z16-NEXT:    vrepg %v1, %v1, 1
+; Z16-NEXT:    mdbr %f1, %f2
+; Z16-NEXT:    vl %v0, 0(%r2), 4
+; Z16-NEXT:    vmrhg %v1, %v3, %v1
+; Z16-NEXT:    wfmdb %f3, %f0, %f2
+; Z16-NEXT:    vrepg %v0, %v0, 1
+; Z16-NEXT:    mdbr %f0, %f2
+; Z16-NEXT:    vmrhg %v0, %v3, %v0
+; Z16-NEXT:    vst %v0, 0(%r2), 4
+; Z16-NEXT:    vst %v1, 16(%r2), 4
+; Z16-NEXT:    br %r14
+  %val = load <4 x double>, ptr %out
+  %canonicalized = call <4 x double> @llvm.canonicalize.v4f64(<4 x double> %val)
+  store <4 x double> %canonicalized, ptr %out
+  ret void
+}
+
+define void @canonicalize_ptr_v2f128(ptr %out) {
+; Z16-LABEL: canonicalize_ptr_v2f128:
+; Z16:       # %bb.0:
+; Z16-NEXT:    larl %r1, .LCPI15_0
+; Z16-NEXT:    vl %v0, 16(%r2), 4
+; Z16-NEXT:    vl %v1, 0(%r2), 4
+; Z16-NEXT:    vl %v2, 0(%r1), 3
+; Z16-NEXT:    wfmxb %v1, %v1, %v2
+; Z16-NEXT:    wfmxb %v0, %v0, %v2
+; Z16-NEXT:    vst %v0, 16(%r2), 4
+; Z16-NEXT:    vst %v1, 0(%r2), 4
+; Z16-NEXT:    br %r14
+  %val = load <2 x fp128>, ptr %out
+  %canonicalized = call <2 x fp128> @llvm.canonicalize.v2f128(<2 x fp128> %val)
+  store <2 x fp128> %canonicalized, ptr %out
+  ret void
+}