[VP][RISCV] Add vp.fmuladd.

[craig.topper]

Summary

Expanded in SelectionDAGBuilder similar to llvm.fmuladd.

[craig.topper]

I forgot the LangRef change.

[craig.topper]

Add LangRef.

[frasercrmck]

LGTM but I have seen that we have some %t and %also.r mismatches throughout the LangRef that were introduced with the more recent VP intrinsics that have probably since been copy/pasted.

llvm/docs/LangRef.rst

19442

%also.t?

[frasercrmck]

In D134474#3817397, @frasercrmck wrote:

LGTM but I have seen that we have some %t and %also.r mismatches throughout the LangRef that were introduced with the more recent VP intrinsics that have probably since been copy/pasted.

Ignore me, I dunno what I was thinking. I think I was getting confused due to the similarity between the %r and %t lines, both being intrinsics

llvm/docs/LangRef.rst

 
 The canonicalization operation may be optimized away if:
 
 - The input is known to be canonical. For example, it was produced by a
   floating-point operation that is required by the standard to be canonical.
 - The result is consumed only by (or fused with) other floating-point
   operations. That is, the bits of the floating-point value are not examined.
 
+.. _int_fmuladd:
+
 '``llvm.fmuladd.*``' Intrinsic
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Syntax:
 """""""
 
 ::
 
.. code-block:: llvm
 
       %r = call <4 x float> @llvm.vp.fma.v4f32(<4 x float> %a, <4 x float> %b, <4 x float> %c, <4 x i1> %mask, i32 %evl)
       ;; For all lanes below %evl, %r is lane-wise equivalent to %also.r
 
       %t = call <4 x float> @llvm.fma(<4 x float> %a, <4 x float> %b, <4 x float> %c)
       %also.r = select <4 x i1> %mask, <4 x float> %t, <4 x float> poison
 
 
+.. _int_vp_fmuladd:
+
+'``llvm.vp.fmuladd.*``' Intrinsics
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+This is an overloaded intrinsic.
+
+::
+
+      declare <16 x float>  @llvm.vp.fmuladd.v16f32 (<16 x float> <left_op>, <16 x float> <middle_op>, <16 x float> <right_op>, <16 x i1> <mask>, i32 <vector_length>)
+      declare <vscale x 4 x float>  @llvm.vp.fmuladd.nxv4f32 (<vscale x 4 x float> <left_op>, <vscale x 4 x float> <middle_op>, <vscale x 4 x float> <right_op>, <vscale x 4 x i1> <mask>, i32 <vector_length>)
+      declare <256 x double>  @llvm.vp.fmuladd.v256f64 (<256 x double> <left_op>, <256 x double> <middle_op>, <256 x double> <right_op>, <256 x i1> <mask>, i32 <vector_length>)
+
+Overview:
+"""""""""
+
+Predicated floating-point multiply-add of two vectors of floating-point values
+that can be fused if code generator determines that (a) the target instruction
+set has support for a fused operation, and (b) that the fused operation is more
+efficient than the equivalent, separate pair of mul and add instructions.
+
+Arguments:
+""""""""""
+
+The first three operands and the result have the same vector of floating-point
+type. The fourth operand is the vector mask and has the same number of elements
+as the result vector type. The fifth operand is the explicit vector length of
+the operation.
+
+Semantics:
+""""""""""
+
+The '``llvm.vp.fmuladd``' intrinsic performs floating-point multiply-add (:ref:`llvm.fuladd <int_fmuladd>`)
+of the first, second, and third vector operand on each enabled lane.  The result
+on disabled lanes is a :ref:`poison value <poisonvalues>`.  The operation is
+performed in the default floating-point environment.
+
+Examples:
+"""""""""
+
+.. code-block:: llvm
+
+      %r = call <4 x float> @llvm.vp.fmuladd.v4f32(<4 x float> %a, <4 x float> %b, <4 x float> %c, <4 x i1> %mask, i32 %evl)
+      ;; For all lanes below %evl, %r is lane-wise equivalent to %also.r
+
+      %t = call <4 x float> @llvm.fmuladd(<4 x float> %a, <4 x float> %b, <4 x float> %c)
+      %also.r = select <4 x i1> %mask, <4 x float> %t, <4 x float> poison
+
+
 .. _int_vp_reduce_add:
 
 '``llvm.vp.reduce.add.*``' Intrinsics
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Syntax:
 """""""
 This is an overloaded intrinsic.

llvm/include/llvm/IR/Intrinsics.td

def int_vp_sqrt : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
                                LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                llvm_i32_ty]>;
   def int_vp_fma : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
                              [ LLVMMatchType<0>,
                                LLVMMatchType<0>,
                                LLVMMatchType<0>,
                                LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                llvm_i32_ty]>;
+  def int_vp_fmuladd : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
+                             [ LLVMMatchType<0>,
+                               LLVMMatchType<0>,
+                               LLVMMatchType<0>,
+                               LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
+                               llvm_i32_ty]>;
   def int_vp_minnum : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
                              [ LLVMMatchType<0>,
                                LLVMMatchType<0>,
                                LLVMScalarOrSameVectorWidth<0, llvm_i1_ty>,
                                llvm_i32_ty]>;
   def int_vp_maxnum : DefaultAttrsIntrinsic<[ llvm_anyvector_ty ],
                              [ LLVMMatchType<0>,
                                LLVMMatchType<0>,

llvm/include/llvm/IR/VPIntrinsics.def

 BEGIN_REGISTER_VP(vp_sqrt, 1, 2, VP_SQRT, -1)
 END_REGISTER_VP(vp_sqrt, VP_SQRT)
 
 // llvm.vp.fma(x,y,z,mask,vlen)
 BEGIN_REGISTER_VP(vp_fma, 3, 4, VP_FMA, -1)
 VP_PROPERTY_CONSTRAINEDFP(1, 1, experimental_constrained_fma)
 END_REGISTER_VP(vp_fma, VP_FMA)
 
+// llvm.vp.fmuladd(x,y,z,mask,vlen)
+BEGIN_REGISTER_VP(vp_fmuladd, 3, 4, VP_FMULADD, -1)
+VP_PROPERTY_CONSTRAINEDFP(1, 1, experimental_constrained_fmuladd)
+END_REGISTER_VP(vp_fmuladd, VP_FMULADD)
+
 // llvm.vp.minnum(x, y, mask,vlen)
 BEGIN_REGISTER_VP(vp_minnum, 2, 3, VP_FMINNUM, -1)
 END_REGISTER_VP(vp_minnum, VP_FMINNUM)
 
 // llvm.vp.maxnum(x, y, mask,vlen)
 BEGIN_REGISTER_VP(vp_maxnum, 2, 3, VP_FMAXNUM, -1)
 END_REGISTER_VP(vp_maxnum, VP_FMAXNUM)
 ///// } Floating-Point Arithmetic

llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp

case ISD::VP_STORE:
     visitVPStore(VPIntrin, OpValues);
     break;
   case ISD::VP_SCATTER:
     visitVPScatter(VPIntrin, OpValues);
     break;
   case ISD::EXPERIMENTAL_VP_STRIDED_STORE:
     visitVPStridedStore(VPIntrin, OpValues);
     break;
+  case ISD::VP_FMULADD: {
+    assert(OpValues.size() == 5 && "Unexpected number of operands");
+    SDNodeFlags SDFlags;
+    if (auto *FPMO = dyn_cast<FPMathOperator>(&VPIntrin))
+      SDFlags.copyFMF(*FPMO);
+    if (TM.Options.AllowFPOpFusion != FPOpFusion::Strict &&
+        TLI.isFMAFasterThanFMulAndFAdd(DAG.getMachineFunction(), ValueVTs[0])) {
+      setValue(&VPIntrin, DAG.getNode(ISD::VP_FMA, DL, VTs, OpValues, SDFlags));
+    } else {
+      SDValue Mul = DAG.getNode(
+          ISD::VP_FMUL, DL, VTs,
+          {OpValues[0], OpValues[1], OpValues[3], OpValues[4]}, SDFlags);
+      SDValue Add =
+          DAG.getNode(ISD::VP_FADD, DL, VTs,
+                      {Mul, OpValues[2], OpValues[3], OpValues[4]}, SDFlags);
+      setValue(&VPIntrin, Add);
+    }
+    break;
+  }
   }
 }
 
 SDValue SelectionDAGBuilder::lowerStartEH(SDValue Chain,
                                           const BasicBlock *EHPadBB,
                                           MCSymbol *&BeginLabel) {
   MachineFunction &MF = DAG.getMachineFunction();
   MachineModuleInfo &MMI = MF.getMMI();

llvm/test/CodeGen/RISCV/rvv/fixed-vectors-vfmuladd-vp.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+experimental-zvfh,+v,+m -target-abi=ilp32d -riscv-v-vector-bits-min=128 \
+; RUN:   -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+experimental-zvfh,+v,+m -target-abi=lp64d -riscv-v-vector-bits-min=128 \
+; RUN:   -verify-machineinstrs < %s | FileCheck %s
+
+declare <2 x half> @llvm.vp.fmuladd.v2f16(<2 x half>, <2 x half>, <2 x half>, <2 x i1>, i32)
+
+define <2 x half> @vfma_vv_v2f16(<2 x half> %va, <2 x half> %b, <2 x half> %c, <2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v2f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv1r.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <2 x half> @llvm.vp.fmuladd.v2f16(<2 x half> %va, <2 x half> %b, <2 x half> %c, <2 x i1> %m, i32 %evl)
+  ret <2 x half> %v
+}
+
+define <2 x half> @vfma_vv_v2f16_unmasked(<2 x half> %va, <2 x half> %b, <2 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v2f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <2 x i1> %head, <2 x i1> poison, <2 x i32> zeroinitializer
+  %v = call <2 x half> @llvm.vp.fmuladd.v2f16(<2 x half> %va, <2 x half> %b, <2 x half> %c, <2 x i1> %m, i32 %evl)
+  ret <2 x half> %v
+}
+
+define <2 x half> @vfma_vf_v2f16(<2 x half> %va, half %b, <2 x half> %vc, <2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v2f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <2 x half> poison, half %b, i32 0
+  %vb = shufflevector <2 x half> %elt.head, <2 x half> poison, <2 x i32> zeroinitializer
+  %v = call <2 x half> @llvm.vp.fmuladd.v2f16(<2 x half> %va, <2 x half> %vb, <2 x half> %vc, <2 x i1> %m, i32 %evl)
+  ret <2 x half> %v
+}
+
+define <2 x half> @vfma_vf_v2f16_unmasked(<2 x half> %va, half %b, <2 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v2f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <2 x half> poison, half %b, i32 0
+  %vb = shufflevector <2 x half> %elt.head, <2 x half> poison, <2 x i32> zeroinitializer
+  %head = insertelement <2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <2 x i1> %head, <2 x i1> poison, <2 x i32> zeroinitializer
+  %v = call <2 x half> @llvm.vp.fmuladd.v2f16(<2 x half> %va, <2 x half> %vb, <2 x half> %vc, <2 x i1> %m, i32 %evl)
+  ret <2 x half> %v
+}
+
+declare <4 x half> @llvm.vp.fmuladd.v4f16(<4 x half>, <4 x half>, <4 x half>, <4 x i1>, i32)
+
+define <4 x half> @vfma_vv_v4f16(<4 x half> %va, <4 x half> %b, <4 x half> %c, <4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v4f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv1r.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <4 x half> @llvm.vp.fmuladd.v4f16(<4 x half> %va, <4 x half> %b, <4 x half> %c, <4 x i1> %m, i32 %evl)
+  ret <4 x half> %v
+}
+
+define <4 x half> @vfma_vv_v4f16_unmasked(<4 x half> %va, <4 x half> %b, <4 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v4f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <4 x i1> %head, <4 x i1> poison, <4 x i32> zeroinitializer
+  %v = call <4 x half> @llvm.vp.fmuladd.v4f16(<4 x half> %va, <4 x half> %b, <4 x half> %c, <4 x i1> %m, i32 %evl)
+  ret <4 x half> %v
+}
+
+define <4 x half> @vfma_vf_v4f16(<4 x half> %va, half %b, <4 x half> %vc, <4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v4f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <4 x half> poison, half %b, i32 0
+  %vb = shufflevector <4 x half> %elt.head, <4 x half> poison, <4 x i32> zeroinitializer
+  %v = call <4 x half> @llvm.vp.fmuladd.v4f16(<4 x half> %va, <4 x half> %vb, <4 x half> %vc, <4 x i1> %m, i32 %evl)
+  ret <4 x half> %v
+}
+
+define <4 x half> @vfma_vf_v4f16_unmasked(<4 x half> %va, half %b, <4 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v4f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <4 x half> poison, half %b, i32 0
+  %vb = shufflevector <4 x half> %elt.head, <4 x half> poison, <4 x i32> zeroinitializer
+  %head = insertelement <4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <4 x i1> %head, <4 x i1> poison, <4 x i32> zeroinitializer
+  %v = call <4 x half> @llvm.vp.fmuladd.v4f16(<4 x half> %va, <4 x half> %vb, <4 x half> %vc, <4 x i1> %m, i32 %evl)
+  ret <4 x half> %v
+}
+
+declare <8 x half> @llvm.vp.fmuladd.v8f16(<8 x half>, <8 x half>, <8 x half>, <8 x i1>, i32)
+
+define <8 x half> @vfma_vv_v8f16(<8 x half> %va, <8 x half> %b, <8 x half> %c, <8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v8f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <8 x half> @llvm.vp.fmuladd.v8f16(<8 x half> %va, <8 x half> %b, <8 x half> %c, <8 x i1> %m, i32 %evl)
+  ret <8 x half> %v
+}
+
+define <8 x half> @vfma_vv_v8f16_unmasked(<8 x half> %va, <8 x half> %b, <8 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v8f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <8 x i1> %head, <8 x i1> poison, <8 x i32> zeroinitializer
+  %v = call <8 x half> @llvm.vp.fmuladd.v8f16(<8 x half> %va, <8 x half> %b, <8 x half> %c, <8 x i1> %m, i32 %evl)
+  ret <8 x half> %v
+}
+
+define <8 x half> @vfma_vf_v8f16(<8 x half> %va, half %b, <8 x half> %vc, <8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v8f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <8 x half> poison, half %b, i32 0
+  %vb = shufflevector <8 x half> %elt.head, <8 x half> poison, <8 x i32> zeroinitializer
+  %v = call <8 x half> @llvm.vp.fmuladd.v8f16(<8 x half> %va, <8 x half> %vb, <8 x half> %vc, <8 x i1> %m, i32 %evl)
+  ret <8 x half> %v
+}
+
+define <8 x half> @vfma_vf_v8f16_unmasked(<8 x half> %va, half %b, <8 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v8f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <8 x half> poison, half %b, i32 0
+  %vb = shufflevector <8 x half> %elt.head, <8 x half> poison, <8 x i32> zeroinitializer
+  %head = insertelement <8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <8 x i1> %head, <8 x i1> poison, <8 x i32> zeroinitializer
+  %v = call <8 x half> @llvm.vp.fmuladd.v8f16(<8 x half> %va, <8 x half> %vb, <8 x half> %vc, <8 x i1> %m, i32 %evl)
+  ret <8 x half> %v
+}
+
+declare <16 x half> @llvm.vp.fmuladd.v16f16(<16 x half>, <16 x half>, <16 x half>, <16 x i1>, i32)
+
+define <16 x half> @vfma_vv_v16f16(<16 x half> %va, <16 x half> %b, <16 x half> %c, <16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v16f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v10, v8, v12, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v10
+; CHECK-NEXT:    ret
+  %v = call <16 x half> @llvm.vp.fmuladd.v16f16(<16 x half> %va, <16 x half> %b, <16 x half> %c, <16 x i1> %m, i32 %evl)
+  ret <16 x half> %v
+}
+
+define <16 x half> @vfma_vv_v16f16_unmasked(<16 x half> %va, <16 x half> %b, <16 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v16f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v10, v12
+; CHECK-NEXT:    ret
+  %head = insertelement <16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <16 x i1> %head, <16 x i1> poison, <16 x i32> zeroinitializer
+  %v = call <16 x half> @llvm.vp.fmuladd.v16f16(<16 x half> %va, <16 x half> %b, <16 x half> %c, <16 x i1> %m, i32 %evl)
+  ret <16 x half> %v
+}
+
+define <16 x half> @vfma_vf_v16f16(<16 x half> %va, half %b, <16 x half> %vc, <16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v16f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <16 x half> poison, half %b, i32 0
+  %vb = shufflevector <16 x half> %elt.head, <16 x half> poison, <16 x i32> zeroinitializer
+  %v = call <16 x half> @llvm.vp.fmuladd.v16f16(<16 x half> %va, <16 x half> %vb, <16 x half> %vc, <16 x i1> %m, i32 %evl)
+  ret <16 x half> %v
+}
+
+define <16 x half> @vfma_vf_v16f16_unmasked(<16 x half> %va, half %b, <16 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v16f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <16 x half> poison, half %b, i32 0
+  %vb = shufflevector <16 x half> %elt.head, <16 x half> poison, <16 x i32> zeroinitializer
+  %head = insertelement <16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <16 x i1> %head, <16 x i1> poison, <16 x i32> zeroinitializer
+  %v = call <16 x half> @llvm.vp.fmuladd.v16f16(<16 x half> %va, <16 x half> %vb, <16 x half> %vc, <16 x i1> %m, i32 %evl)
+  ret <16 x half> %v
+}
+
+declare <2 x float> @llvm.vp.fmuladd.v2f32(<2 x float>, <2 x float>, <2 x float>, <2 x i1>, i32)
+
+define <2 x float> @vfma_vv_v2f32(<2 x float> %va, <2 x float> %b, <2 x float> %c, <2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v2f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv1r.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <2 x float> @llvm.vp.fmuladd.v2f32(<2 x float> %va, <2 x float> %b, <2 x float> %c, <2 x i1> %m, i32 %evl)
+  ret <2 x float> %v
+}
+
+define <2 x float> @vfma_vv_v2f32_unmasked(<2 x float> %va, <2 x float> %b, <2 x float> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v2f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <2 x i1> %head, <2 x i1> poison, <2 x i32> zeroinitializer
+  %v = call <2 x float> @llvm.vp.fmuladd.v2f32(<2 x float> %va, <2 x float> %b, <2 x float> %c, <2 x i1> %m, i32 %evl)
+  ret <2 x float> %v
+}
+
+define <2 x float> @vfma_vf_v2f32(<2 x float> %va, float %b, <2 x float> %vc, <2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v2f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <2 x float> poison, float %b, i32 0
+  %vb = shufflevector <2 x float> %elt.head, <2 x float> poison, <2 x i32> zeroinitializer
+  %v = call <2 x float> @llvm.vp.fmuladd.v2f32(<2 x float> %va, <2 x float> %vb, <2 x float> %vc, <2 x i1> %m, i32 %evl)
+  ret <2 x float> %v
+}
+
+define <2 x float> @vfma_vf_v2f32_unmasked(<2 x float> %va, float %b, <2 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v2f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <2 x float> poison, float %b, i32 0
+  %vb = shufflevector <2 x float> %elt.head, <2 x float> poison, <2 x i32> zeroinitializer
+  %head = insertelement <2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <2 x i1> %head, <2 x i1> poison, <2 x i32> zeroinitializer
+  %v = call <2 x float> @llvm.vp.fmuladd.v2f32(<2 x float> %va, <2 x float> %vb, <2 x float> %vc, <2 x i1> %m, i32 %evl)
+  ret <2 x float> %v
+}
+
+declare <4 x float> @llvm.vp.fmuladd.v4f32(<4 x float>, <4 x float>, <4 x float>, <4 x i1>, i32)
+
+define <4 x float> @vfma_vv_v4f32(<4 x float> %va, <4 x float> %b, <4 x float> %c, <4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v4f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <4 x float> @llvm.vp.fmuladd.v4f32(<4 x float> %va, <4 x float> %b, <4 x float> %c, <4 x i1> %m, i32 %evl)
+  ret <4 x float> %v
+}
+
+define <4 x float> @vfma_vv_v4f32_unmasked(<4 x float> %va, <4 x float> %b, <4 x float> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v4f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <4 x i1> %head, <4 x i1> poison, <4 x i32> zeroinitializer
+  %v = call <4 x float> @llvm.vp.fmuladd.v4f32(<4 x float> %va, <4 x float> %b, <4 x float> %c, <4 x i1> %m, i32 %evl)
+  ret <4 x float> %v
+}
+
+define <4 x float> @vfma_vf_v4f32(<4 x float> %va, float %b, <4 x float> %vc, <4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v4f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <4 x float> poison, float %b, i32 0
+  %vb = shufflevector <4 x float> %elt.head, <4 x float> poison, <4 x i32> zeroinitializer
+  %v = call <4 x float> @llvm.vp.fmuladd.v4f32(<4 x float> %va, <4 x float> %vb, <4 x float> %vc, <4 x i1> %m, i32 %evl)
+  ret <4 x float> %v
+}
+
+define <4 x float> @vfma_vf_v4f32_unmasked(<4 x float> %va, float %b, <4 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v4f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <4 x float> poison, float %b, i32 0
+  %vb = shufflevector <4 x float> %elt.head, <4 x float> poison, <4 x i32> zeroinitializer
+  %head = insertelement <4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <4 x i1> %head, <4 x i1> poison, <4 x i32> zeroinitializer
+  %v = call <4 x float> @llvm.vp.fmuladd.v4f32(<4 x float> %va, <4 x float> %vb, <4 x float> %vc, <4 x i1> %m, i32 %evl)
+  ret <4 x float> %v
+}
+
+declare <8 x float> @llvm.vp.fmuladd.v8f32(<8 x float>, <8 x float>, <8 x float>, <8 x i1>, i32)
+
+define <8 x float> @vfma_vv_v8f32(<8 x float> %va, <8 x float> %b, <8 x float> %c, <8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v8f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v10, v8, v12, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v10
+; CHECK-NEXT:    ret
+  %v = call <8 x float> @llvm.vp.fmuladd.v8f32(<8 x float> %va, <8 x float> %b, <8 x float> %c, <8 x i1> %m, i32 %evl)
+  ret <8 x float> %v
+}
+
+define <8 x float> @vfma_vv_v8f32_unmasked(<8 x float> %va, <8 x float> %b, <8 x float> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v8f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v10, v12
+; CHECK-NEXT:    ret
+  %head = insertelement <8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <8 x i1> %head, <8 x i1> poison, <8 x i32> zeroinitializer
+  %v = call <8 x float> @llvm.vp.fmuladd.v8f32(<8 x float> %va, <8 x float> %b, <8 x float> %c, <8 x i1> %m, i32 %evl)
+  ret <8 x float> %v
+}
+
+define <8 x float> @vfma_vf_v8f32(<8 x float> %va, float %b, <8 x float> %vc, <8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v8f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <8 x float> poison, float %b, i32 0
+  %vb = shufflevector <8 x float> %elt.head, <8 x float> poison, <8 x i32> zeroinitializer
+  %v = call <8 x float> @llvm.vp.fmuladd.v8f32(<8 x float> %va, <8 x float> %vb, <8 x float> %vc, <8 x i1> %m, i32 %evl)
+  ret <8 x float> %v
+}
+
+define <8 x float> @vfma_vf_v8f32_unmasked(<8 x float> %va, float %b, <8 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v8f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <8 x float> poison, float %b, i32 0
+  %vb = shufflevector <8 x float> %elt.head, <8 x float> poison, <8 x i32> zeroinitializer
+  %head = insertelement <8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <8 x i1> %head, <8 x i1> poison, <8 x i32> zeroinitializer
+  %v = call <8 x float> @llvm.vp.fmuladd.v8f32(<8 x float> %va, <8 x float> %vb, <8 x float> %vc, <8 x i1> %m, i32 %evl)
+  ret <8 x float> %v
+}
+
+declare <16 x float> @llvm.vp.fmuladd.v16f32(<16 x float>, <16 x float>, <16 x float>, <16 x i1>, i32)
+
+define <16 x float> @vfma_vv_v16f32(<16 x float> %va, <16 x float> %b, <16 x float> %c, <16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v16f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v12, v8, v16, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v12
+; CHECK-NEXT:    ret
+  %v = call <16 x float> @llvm.vp.fmuladd.v16f32(<16 x float> %va, <16 x float> %b, <16 x float> %c, <16 x i1> %m, i32 %evl)
+  ret <16 x float> %v
+}
+
+define <16 x float> @vfma_vv_v16f32_unmasked(<16 x float> %va, <16 x float> %b, <16 x float> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v16f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v12, v16
+; CHECK-NEXT:    ret
+  %head = insertelement <16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <16 x i1> %head, <16 x i1> poison, <16 x i32> zeroinitializer
+  %v = call <16 x float> @llvm.vp.fmuladd.v16f32(<16 x float> %va, <16 x float> %b, <16 x float> %c, <16 x i1> %m, i32 %evl)
+  ret <16 x float> %v
+}
+
+define <16 x float> @vfma_vf_v16f32(<16 x float> %va, float %b, <16 x float> %vc, <16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v16f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <16 x float> poison, float %b, i32 0
+  %vb = shufflevector <16 x float> %elt.head, <16 x float> poison, <16 x i32> zeroinitializer
+  %v = call <16 x float> @llvm.vp.fmuladd.v16f32(<16 x float> %va, <16 x float> %vb, <16 x float> %vc, <16 x i1> %m, i32 %evl)
+  ret <16 x float> %v
+}
+
+define <16 x float> @vfma_vf_v16f32_unmasked(<16 x float> %va, float %b, <16 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v16f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <16 x float> poison, float %b, i32 0
+  %vb = shufflevector <16 x float> %elt.head, <16 x float> poison, <16 x i32> zeroinitializer
+  %head = insertelement <16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <16 x i1> %head, <16 x i1> poison, <16 x i32> zeroinitializer
+  %v = call <16 x float> @llvm.vp.fmuladd.v16f32(<16 x float> %va, <16 x float> %vb, <16 x float> %vc, <16 x i1> %m, i32 %evl)
+  ret <16 x float> %v
+}
+
+declare <2 x double> @llvm.vp.fmuladd.v2f64(<2 x double>, <2 x double>, <2 x double>, <2 x i1>, i32)
+
+define <2 x double> @vfma_vv_v2f64(<2 x double> %va, <2 x double> %b, <2 x double> %c, <2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v2f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <2 x double> @llvm.vp.fmuladd.v2f64(<2 x double> %va, <2 x double> %b, <2 x double> %c, <2 x i1> %m, i32 %evl)
+  ret <2 x double> %v
+}
+
+define <2 x double> @vfma_vv_v2f64_unmasked(<2 x double> %va, <2 x double> %b, <2 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v2f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <2 x i1> %head, <2 x i1> poison, <2 x i32> zeroinitializer
+  %v = call <2 x double> @llvm.vp.fmuladd.v2f64(<2 x double> %va, <2 x double> %b, <2 x double> %c, <2 x i1> %m, i32 %evl)
+  ret <2 x double> %v
+}
+
+define <2 x double> @vfma_vf_v2f64(<2 x double> %va, double %b, <2 x double> %vc, <2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v2f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <2 x double> poison, double %b, i32 0
+  %vb = shufflevector <2 x double> %elt.head, <2 x double> poison, <2 x i32> zeroinitializer
+  %v = call <2 x double> @llvm.vp.fmuladd.v2f64(<2 x double> %va, <2 x double> %vb, <2 x double> %vc, <2 x i1> %m, i32 %evl)
+  ret <2 x double> %v
+}
+
+define <2 x double> @vfma_vf_v2f64_unmasked(<2 x double> %va, double %b, <2 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v2f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <2 x double> poison, double %b, i32 0
+  %vb = shufflevector <2 x double> %elt.head, <2 x double> poison, <2 x i32> zeroinitializer
+  %head = insertelement <2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <2 x i1> %head, <2 x i1> poison, <2 x i32> zeroinitializer
+  %v = call <2 x double> @llvm.vp.fmuladd.v2f64(<2 x double> %va, <2 x double> %vb, <2 x double> %vc, <2 x i1> %m, i32 %evl)
+  ret <2 x double> %v
+}
+
+declare <4 x double> @llvm.vp.fmuladd.v4f64(<4 x double>, <4 x double>, <4 x double>, <4 x i1>, i32)
+
+define <4 x double> @vfma_vv_v4f64(<4 x double> %va, <4 x double> %b, <4 x double> %c, <4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v4f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v10, v8, v12, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v10
+; CHECK-NEXT:    ret
+  %v = call <4 x double> @llvm.vp.fmuladd.v4f64(<4 x double> %va, <4 x double> %b, <4 x double> %c, <4 x i1> %m, i32 %evl)
+  ret <4 x double> %v
+}
+
+define <4 x double> @vfma_vv_v4f64_unmasked(<4 x double> %va, <4 x double> %b, <4 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v4f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v10, v12
+; CHECK-NEXT:    ret
+  %head = insertelement <4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <4 x i1> %head, <4 x i1> poison, <4 x i32> zeroinitializer
+  %v = call <4 x double> @llvm.vp.fmuladd.v4f64(<4 x double> %va, <4 x double> %b, <4 x double> %c, <4 x i1> %m, i32 %evl)
+  ret <4 x double> %v
+}
+
+define <4 x double> @vfma_vf_v4f64(<4 x double> %va, double %b, <4 x double> %vc, <4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v4f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <4 x double> poison, double %b, i32 0
+  %vb = shufflevector <4 x double> %elt.head, <4 x double> poison, <4 x i32> zeroinitializer
+  %v = call <4 x double> @llvm.vp.fmuladd.v4f64(<4 x double> %va, <4 x double> %vb, <4 x double> %vc, <4 x i1> %m, i32 %evl)
+  ret <4 x double> %v
+}
+
+define <4 x double> @vfma_vf_v4f64_unmasked(<4 x double> %va, double %b, <4 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v4f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <4 x double> poison, double %b, i32 0
+  %vb = shufflevector <4 x double> %elt.head, <4 x double> poison, <4 x i32> zeroinitializer
+  %head = insertelement <4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <4 x i1> %head, <4 x i1> poison, <4 x i32> zeroinitializer
+  %v = call <4 x double> @llvm.vp.fmuladd.v4f64(<4 x double> %va, <4 x double> %vb, <4 x double> %vc, <4 x i1> %m, i32 %evl)
+  ret <4 x double> %v
+}
+
+declare <8 x double> @llvm.vp.fmuladd.v8f64(<8 x double>, <8 x double>, <8 x double>, <8 x i1>, i32)
+
+define <8 x double> @vfma_vv_v8f64(<8 x double> %va, <8 x double> %b, <8 x double> %c, <8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v8f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v12, v8, v16, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v12
+; CHECK-NEXT:    ret
+  %v = call <8 x double> @llvm.vp.fmuladd.v8f64(<8 x double> %va, <8 x double> %b, <8 x double> %c, <8 x i1> %m, i32 %evl)
+  ret <8 x double> %v
+}
+
+define <8 x double> @vfma_vv_v8f64_unmasked(<8 x double> %va, <8 x double> %b, <8 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v8f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v12, v16
+; CHECK-NEXT:    ret
+  %head = insertelement <8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <8 x i1> %head, <8 x i1> poison, <8 x i32> zeroinitializer
+  %v = call <8 x double> @llvm.vp.fmuladd.v8f64(<8 x double> %va, <8 x double> %b, <8 x double> %c, <8 x i1> %m, i32 %evl)
+  ret <8 x double> %v
+}
+
+define <8 x double> @vfma_vf_v8f64(<8 x double> %va, double %b, <8 x double> %vc, <8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v8f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <8 x double> poison, double %b, i32 0
+  %vb = shufflevector <8 x double> %elt.head, <8 x double> poison, <8 x i32> zeroinitializer
+  %v = call <8 x double> @llvm.vp.fmuladd.v8f64(<8 x double> %va, <8 x double> %vb, <8 x double> %vc, <8 x i1> %m, i32 %evl)
+  ret <8 x double> %v
+}
+
+define <8 x double> @vfma_vf_v8f64_unmasked(<8 x double> %va, double %b, <8 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v8f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <8 x double> poison, double %b, i32 0
+  %vb = shufflevector <8 x double> %elt.head, <8 x double> poison, <8 x i32> zeroinitializer
+  %head = insertelement <8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <8 x i1> %head, <8 x i1> poison, <8 x i32> zeroinitializer
+  %v = call <8 x double> @llvm.vp.fmuladd.v8f64(<8 x double> %va, <8 x double> %vb, <8 x double> %vc, <8 x i1> %m, i32 %evl)
+  ret <8 x double> %v
+}
+
+declare <15 x double> @llvm.vp.fmuladd.v15f64(<15 x double>, <15 x double>, <15 x double>, <15 x i1>, i32)
+
+define <15 x double> @vfma_vv_v15f64(<15 x double> %va, <15 x double> %b, <15 x double> %c, <15 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v15f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, mu
+; CHECK-NEXT:    vle64.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v16, v8, v24, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v16
+; CHECK-NEXT:    ret
+  %v = call <15 x double> @llvm.vp.fmuladd.v15f64(<15 x double> %va, <15 x double> %b, <15 x double> %c, <15 x i1> %m, i32 %evl)
+  ret <15 x double> %v
+}
+
+define <15 x double> @vfma_vv_v15f64_unmasked(<15 x double> %va, <15 x double> %b, <15 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v15f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, mu
+; CHECK-NEXT:    vle64.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v16, v24
+; CHECK-NEXT:    ret
+  %head = insertelement <15 x i1> poison, i1 true, i32 0
+  %m = shufflevector <15 x i1> %head, <15 x i1> poison, <15 x i32> zeroinitializer
+  %v = call <15 x double> @llvm.vp.fmuladd.v15f64(<15 x double> %va, <15 x double> %b, <15 x double> %c, <15 x i1> %m, i32 %evl)
+  ret <15 x double> %v
+}
+
+declare <16 x double> @llvm.vp.fmuladd.v16f64(<16 x double>, <16 x double>, <16 x double>, <16 x i1>, i32)
+
+define <16 x double> @vfma_vv_v16f64(<16 x double> %va, <16 x double> %b, <16 x double> %c, <16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v16f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, mu
+; CHECK-NEXT:    vle64.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v16, v8, v24, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v16
+; CHECK-NEXT:    ret
+  %v = call <16 x double> @llvm.vp.fmuladd.v16f64(<16 x double> %va, <16 x double> %b, <16 x double> %c, <16 x i1> %m, i32 %evl)
+  ret <16 x double> %v
+}
+
+define <16 x double> @vfma_vv_v16f64_unmasked(<16 x double> %va, <16 x double> %b, <16 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v16f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, mu
+; CHECK-NEXT:    vle64.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v16, v24
+; CHECK-NEXT:    ret
+  %head = insertelement <16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <16 x i1> %head, <16 x i1> poison, <16 x i32> zeroinitializer
+  %v = call <16 x double> @llvm.vp.fmuladd.v16f64(<16 x double> %va, <16 x double> %b, <16 x double> %c, <16 x i1> %m, i32 %evl)
+  ret <16 x double> %v
+}
+
+define <16 x double> @vfma_vf_v16f64(<16 x double> %va, double %b, <16 x double> %vc, <16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v16f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <16 x double> poison, double %b, i32 0
+  %vb = shufflevector <16 x double> %elt.head, <16 x double> poison, <16 x i32> zeroinitializer
+  %v = call <16 x double> @llvm.vp.fmuladd.v16f64(<16 x double> %va, <16 x double> %vb, <16 x double> %vc, <16 x i1> %m, i32 %evl)
+  ret <16 x double> %v
+}
+
+define <16 x double> @vfma_vf_v16f64_unmasked(<16 x double> %va, double %b, <16 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_v16f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <16 x double> poison, double %b, i32 0
+  %vb = shufflevector <16 x double> %elt.head, <16 x double> poison, <16 x i32> zeroinitializer
+  %head = insertelement <16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <16 x i1> %head, <16 x i1> poison, <16 x i32> zeroinitializer
+  %v = call <16 x double> @llvm.vp.fmuladd.v16f64(<16 x double> %va, <16 x double> %vb, <16 x double> %vc, <16 x i1> %m, i32 %evl)
+  ret <16 x double> %v
+}
+
+declare <32 x double> @llvm.vp.fmuladd.v32f64(<32 x double>, <32 x double>, <32 x double>, <32 x i1>, i32)
+
+define <32 x double> @vfma_vv_v32f64(<32 x double> %va, <32 x double> %b, <32 x double> %c, <32 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v32f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a3, 48
+; CHECK-NEXT:    mul a1, a1, a3
+; CHECK-NEXT:    sub sp, sp, a1
+; CHECK-NEXT:    vmv1r.v v1, v0
+; CHECK-NEXT:    vsetivli zero, 2, e8, mf4, ta, mu
+; CHECK-NEXT:    vslidedown.vi v0, v0, 2
+; CHECK-NEXT:    addi a1, a2, 128
+; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, mu
+; CHECK-NEXT:    vle64.v v24, (a1)
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a3, 24
+; CHECK-NEXT:    mul a1, a1, a3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    addi a1, a0, 128
+; CHECK-NEXT:    vle64.v v24, (a1)
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a3, 40
+; CHECK-NEXT:    mul a1, a1, a3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    addi a3, a4, -16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 5
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    li a1, 0
+; CHECK-NEXT:    bltu a4, a3, .LBB50_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv a1, a3
+; CHECK-NEXT:  .LBB50_2:
+; CHECK-NEXT:    vle64.v v8, (a2)
+; CHECK-NEXT:    csrr a2, vlenb
+; CHECK-NEXT:    slli a2, a2, 4
+; CHECK-NEXT:    add a2, sp, a2
+; CHECK-NEXT:    addi a2, a2, 16
+; CHECK-NEXT:    vs8r.v v8, (a2) # Unknown-size Folded Spill
+; CHECK-NEXT:    vle64.v v8, (a0)
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
+; CHECK-NEXT:    li a0, 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a2, 24
+; CHECK-NEXT:    mul a1, a1, a2
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vl8re8.v v8, (a1) # Unknown-size Folded Reload
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vl8re8.v v24, (a1) # Unknown-size Folded Reload
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a2, 40
+; CHECK-NEXT:    mul a1, a1, a2
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vl8re8.v v16, (a1) # Unknown-size Folded Reload
+; CHECK-NEXT:    vfmadd.vv v16, v24, v8, v0.t
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a2, 40
+; CHECK-NEXT:    mul a1, a1, a2
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    bltu a4, a0, .LBB50_4
+; CHECK-NEXT:  # %bb.3:
+; CHECK-NEXT:    li a4, 16
+; CHECK-NEXT:  .LBB50_4:
+; CHECK-NEXT:    vsetvli zero, a4, e64, m8, ta, mu
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 5
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v8, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 4
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v24, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vl8re8.v v16, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vfmadd.vv v16, v8, v24, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v16
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a1, 40
+; CHECK-NEXT:    mul a0, a0, a1
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v16, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a1, 48
+; CHECK-NEXT:    mul a0, a0, a1
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    addi sp, sp, 16
+; CHECK-NEXT:    ret
+  %v = call <32 x double> @llvm.vp.fmuladd.v32f64(<32 x double> %va, <32 x double> %b, <32 x double> %c, <32 x i1> %m, i32 %evl)
+  ret <32 x double> %v
+}
+
+define <32 x double> @vfma_vv_v32f64_unmasked(<32 x double> %va, <32 x double> %b, <32 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_v32f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a3, 24
+; CHECK-NEXT:    mul a1, a1, a3
+; CHECK-NEXT:    sub sp, sp, a1
+; CHECK-NEXT:    addi a1, a2, 128
+; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, mu
+; CHECK-NEXT:    vle64.v v24, (a1)
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    addi a1, a0, 128
+; CHECK-NEXT:    vle64.v v24, (a1)
+; CHECK-NEXT:    addi a3, a4, -16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 4
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    li a1, 0
+; CHECK-NEXT:    bltu a4, a3, .LBB51_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv a1, a3
+; CHECK-NEXT:  .LBB51_2:
+; CHECK-NEXT:    vle64.v v8, (a2)
+; CHECK-NEXT:    addi a2, sp, 16
+; CHECK-NEXT:    vs8r.v v8, (a2) # Unknown-size Folded Spill
+; CHECK-NEXT:    vle64.v v0, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
+; CHECK-NEXT:    li a0, 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vl8re8.v v8, (a1) # Unknown-size Folded Reload
+; CHECK-NEXT:    vfmadd.vv v24, v16, v8
+; CHECK-NEXT:    bltu a4, a0, .LBB51_4
+; CHECK-NEXT:  # %bb.3:
+; CHECK-NEXT:    li a4, 16
+; CHECK-NEXT:  .LBB51_4:
+; CHECK-NEXT:    vsetvli zero, a4, e64, m8, ta, mu
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 4
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v16, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vl8re8.v v8, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vfmadd.vv v0, v16, v8
+; CHECK-NEXT:    vmv.v.v v8, v0
+; CHECK-NEXT:    vmv8r.v v16, v24
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a1, 24
+; CHECK-NEXT:    mul a0, a0, a1
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    addi sp, sp, 16
+; CHECK-NEXT:    ret
+  %head = insertelement <32 x i1> poison, i1 true, i32 0
+  %m = shufflevector <32 x i1> %head, <32 x i1> poison, <32 x i32> zeroinitializer
+  %v = call <32 x double> @llvm.vp.fmuladd.v32f64(<32 x double> %va, <32 x double> %b, <32 x double> %c, <32 x i1> %m, i32 %evl)
+  ret <32 x double> %v
+}
+

llvm/test/CodeGen/RISCV/rvv/vfmuladd-vp.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+experimental-zvfh,+v,+m -target-abi=ilp32d \
+; RUN:     -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+experimental-zvfh,+v,+m -target-abi=lp64d \
+; RUN:     -verify-machineinstrs < %s | FileCheck %s
+
+declare <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half>, <vscale x 1 x half>, <vscale x 1 x half>, <vscale x 1 x i1>, i32)
+
+define <vscale x 1 x half> @vfma_vv_nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv1r.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfma_vv_nxv1f16_unmasked(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv1f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfma_vf_nxv1f16(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %vb, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfma_vf_nxv1f16_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f16_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x half> %va, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfma_vf_nxv1f16_unmasked(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %vb, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfma_vf_nxv1f16_unmasked_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f16_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x half> %va, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+declare <vscale x 2 x half> @llvm.vp.fmuladd.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, <vscale x 2 x half>, <vscale x 2 x i1>, i32)
+
+define <vscale x 2 x half> @vfma_vv_nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x half> %b, <vscale x 2 x half> %c, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv2f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv1r.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <vscale x 2 x half> @llvm.vp.fmuladd.nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x half> %b, <vscale x 2 x half> %c, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x half> %v
+}
+
+define <vscale x 2 x half> @vfma_vv_nxv2f16_unmasked(<vscale x 2 x half> %va, <vscale x 2 x half> %b, <vscale x 2 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv2f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x half> @llvm.vp.fmuladd.nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x half> %b, <vscale x 2 x half> %c, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x half> %v
+}
+
+define <vscale x 2 x half> @vfma_vf_nxv2f16(<vscale x 2 x half> %va, half %b, <vscale x 2 x half> %vc, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 2 x half> %elt.head, <vscale x 2 x half> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x half> @llvm.vp.fmuladd.nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x half> %vb, <vscale x 2 x half> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x half> %v
+}
+
+define <vscale x 2 x half> @vfma_vf_nxv2f16_commute(<vscale x 2 x half> %va, half %b, <vscale x 2 x half> %vc, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f16_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 2 x half> %elt.head, <vscale x 2 x half> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x half> @llvm.vp.fmuladd.nxv2f16(<vscale x 2 x half> %vb, <vscale x 2 x half> %va, <vscale x 2 x half> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x half> %v
+}
+
+define <vscale x 2 x half> @vfma_vf_nxv2f16_unmasked(<vscale x 2 x half> %va, half %b, <vscale x 2 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 2 x half> %elt.head, <vscale x 2 x half> poison, <vscale x 2 x i32> zeroinitializer
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x half> @llvm.vp.fmuladd.nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x half> %vb, <vscale x 2 x half> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x half> %v
+}
+
+define <vscale x 2 x half> @vfma_vf_nxv2f16_unmasked_commute(<vscale x 2 x half> %va, half %b, <vscale x 2 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f16_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 2 x half> %elt.head, <vscale x 2 x half> poison, <vscale x 2 x i32> zeroinitializer
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x half> @llvm.vp.fmuladd.nxv2f16(<vscale x 2 x half> %vb, <vscale x 2 x half> %va, <vscale x 2 x half> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x half> %v
+}
+
+declare <vscale x 4 x half> @llvm.vp.fmuladd.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, <vscale x 4 x half>, <vscale x 4 x i1>, i32)
+
+define <vscale x 4 x half> @vfma_vv_nxv4f16(<vscale x 4 x half> %va, <vscale x 4 x half> %b, <vscale x 4 x half> %c, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv4f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <vscale x 4 x half> @llvm.vp.fmuladd.nxv4f16(<vscale x 4 x half> %va, <vscale x 4 x half> %b, <vscale x 4 x half> %c, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x half> %v
+}
+
+define <vscale x 4 x half> @vfma_vv_nxv4f16_unmasked(<vscale x 4 x half> %va, <vscale x 4 x half> %b, <vscale x 4 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv4f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x half> @llvm.vp.fmuladd.nxv4f16(<vscale x 4 x half> %va, <vscale x 4 x half> %b, <vscale x 4 x half> %c, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x half> %v
+}
+
+define <vscale x 4 x half> @vfma_vf_nxv4f16(<vscale x 4 x half> %va, half %b, <vscale x 4 x half> %vc, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 4 x half> %elt.head, <vscale x 4 x half> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x half> @llvm.vp.fmuladd.nxv4f16(<vscale x 4 x half> %va, <vscale x 4 x half> %vb, <vscale x 4 x half> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x half> %v
+}
+
+define <vscale x 4 x half> @vfma_vf_nxv4f16_commute(<vscale x 4 x half> %va, half %b, <vscale x 4 x half> %vc, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f16_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 4 x half> %elt.head, <vscale x 4 x half> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x half> @llvm.vp.fmuladd.nxv4f16(<vscale x 4 x half> %vb, <vscale x 4 x half> %va, <vscale x 4 x half> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x half> %v
+}
+
+define <vscale x 4 x half> @vfma_vf_nxv4f16_unmasked(<vscale x 4 x half> %va, half %b, <vscale x 4 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 4 x half> %elt.head, <vscale x 4 x half> poison, <vscale x 4 x i32> zeroinitializer
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x half> @llvm.vp.fmuladd.nxv4f16(<vscale x 4 x half> %va, <vscale x 4 x half> %vb, <vscale x 4 x half> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x half> %v
+}
+
+define <vscale x 4 x half> @vfma_vf_nxv4f16_unmasked_commute(<vscale x 4 x half> %va, half %b, <vscale x 4 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f16_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 4 x half> %elt.head, <vscale x 4 x half> poison, <vscale x 4 x i32> zeroinitializer
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x half> @llvm.vp.fmuladd.nxv4f16(<vscale x 4 x half> %vb, <vscale x 4 x half> %va, <vscale x 4 x half> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x half> %v
+}
+
+declare <vscale x 8 x half> @llvm.vp.fmuladd.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x i1>, i32)
+
+define <vscale x 8 x half> @vfma_vv_nxv8f16(<vscale x 8 x half> %va, <vscale x 8 x half> %b, <vscale x 8 x half> %c, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv8f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v10, v8, v12, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v10
+; CHECK-NEXT:    ret
+  %v = call <vscale x 8 x half> @llvm.vp.fmuladd.nxv8f16(<vscale x 8 x half> %va, <vscale x 8 x half> %b, <vscale x 8 x half> %c, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x half> %v
+}
+
+define <vscale x 8 x half> @vfma_vv_nxv8f16_unmasked(<vscale x 8 x half> %va, <vscale x 8 x half> %b, <vscale x 8 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv8f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v10, v12
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x half> @llvm.vp.fmuladd.nxv8f16(<vscale x 8 x half> %va, <vscale x 8 x half> %b, <vscale x 8 x half> %c, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x half> %v
+}
+
+define <vscale x 8 x half> @vfma_vf_nxv8f16(<vscale x 8 x half> %va, half %b, <vscale x 8 x half> %vc, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 8 x half> %elt.head, <vscale x 8 x half> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x half> @llvm.vp.fmuladd.nxv8f16(<vscale x 8 x half> %va, <vscale x 8 x half> %vb, <vscale x 8 x half> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x half> %v
+}
+
+define <vscale x 8 x half> @vfma_vf_nxv8f16_commute(<vscale x 8 x half> %va, half %b, <vscale x 8 x half> %vc, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f16_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 8 x half> %elt.head, <vscale x 8 x half> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x half> @llvm.vp.fmuladd.nxv8f16(<vscale x 8 x half> %vb, <vscale x 8 x half> %va, <vscale x 8 x half> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x half> %v
+}
+
+define <vscale x 8 x half> @vfma_vf_nxv8f16_unmasked(<vscale x 8 x half> %va, half %b, <vscale x 8 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 8 x half> %elt.head, <vscale x 8 x half> poison, <vscale x 8 x i32> zeroinitializer
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x half> @llvm.vp.fmuladd.nxv8f16(<vscale x 8 x half> %va, <vscale x 8 x half> %vb, <vscale x 8 x half> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x half> %v
+}
+
+define <vscale x 8 x half> @vfma_vf_nxv8f16_unmasked_commute(<vscale x 8 x half> %va, half %b, <vscale x 8 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f16_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 8 x half> %elt.head, <vscale x 8 x half> poison, <vscale x 8 x i32> zeroinitializer
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x half> @llvm.vp.fmuladd.nxv8f16(<vscale x 8 x half> %vb, <vscale x 8 x half> %va, <vscale x 8 x half> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x half> %v
+}
+
+declare <vscale x 16 x half> @llvm.vp.fmuladd.nxv16f16(<vscale x 16 x half>, <vscale x 16 x half>, <vscale x 16 x half>, <vscale x 16 x i1>, i32)
+
+define <vscale x 16 x half> @vfma_vv_nxv16f16(<vscale x 16 x half> %va, <vscale x 16 x half> %b, <vscale x 16 x half> %c, <vscale x 16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv16f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v12, v8, v16, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v12
+; CHECK-NEXT:    ret
+  %v = call <vscale x 16 x half> @llvm.vp.fmuladd.nxv16f16(<vscale x 16 x half> %va, <vscale x 16 x half> %b, <vscale x 16 x half> %c, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x half> %v
+}
+
+define <vscale x 16 x half> @vfma_vv_nxv16f16_unmasked(<vscale x 16 x half> %va, <vscale x 16 x half> %b, <vscale x 16 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv16f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v12, v16
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 16 x i1> %head, <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x half> @llvm.vp.fmuladd.nxv16f16(<vscale x 16 x half> %va, <vscale x 16 x half> %b, <vscale x 16 x half> %c, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x half> %v
+}
+
+define <vscale x 16 x half> @vfma_vf_nxv16f16(<vscale x 16 x half> %va, half %b, <vscale x 16 x half> %vc, <vscale x 16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv16f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 16 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 16 x half> %elt.head, <vscale x 16 x half> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x half> @llvm.vp.fmuladd.nxv16f16(<vscale x 16 x half> %va, <vscale x 16 x half> %vb, <vscale x 16 x half> %vc, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x half> %v
+}
+
+define <vscale x 16 x half> @vfma_vf_nxv16f16_commute(<vscale x 16 x half> %va, half %b, <vscale x 16 x half> %vc, <vscale x 16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv16f16_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 16 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 16 x half> %elt.head, <vscale x 16 x half> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x half> @llvm.vp.fmuladd.nxv16f16(<vscale x 16 x half> %vb, <vscale x 16 x half> %va, <vscale x 16 x half> %vc, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x half> %v
+}
+
+define <vscale x 16 x half> @vfma_vf_nxv16f16_unmasked(<vscale x 16 x half> %va, half %b, <vscale x 16 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv16f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 16 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 16 x half> %elt.head, <vscale x 16 x half> poison, <vscale x 16 x i32> zeroinitializer
+  %head = insertelement <vscale x 16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 16 x i1> %head, <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x half> @llvm.vp.fmuladd.nxv16f16(<vscale x 16 x half> %va, <vscale x 16 x half> %vb, <vscale x 16 x half> %vc, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x half> %v
+}
+
+define <vscale x 16 x half> @vfma_vf_nxv16f16_unmasked_commute(<vscale x 16 x half> %va, half %b, <vscale x 16 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv16f16_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 16 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 16 x half> %elt.head, <vscale x 16 x half> poison, <vscale x 16 x i32> zeroinitializer
+  %head = insertelement <vscale x 16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 16 x i1> %head, <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x half> @llvm.vp.fmuladd.nxv16f16(<vscale x 16 x half> %vb, <vscale x 16 x half> %va, <vscale x 16 x half> %vc, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x half> %v
+}
+
+declare <vscale x 32 x half> @llvm.vp.fmuladd.nxv32f16(<vscale x 32 x half>, <vscale x 32 x half>, <vscale x 32 x half>, <vscale x 32 x i1>, i32)
+
+define <vscale x 32 x half> @vfma_vv_nxv32f16(<vscale x 32 x half> %va, <vscale x 32 x half> %b, <vscale x 32 x half> %c, <vscale x 32 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv32f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vl8re16.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e16, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v16, v8, v24, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v16
+; CHECK-NEXT:    ret
+  %v = call <vscale x 32 x half> @llvm.vp.fmuladd.nxv32f16(<vscale x 32 x half> %va, <vscale x 32 x half> %b, <vscale x 32 x half> %c, <vscale x 32 x i1> %m, i32 %evl)
+  ret <vscale x 32 x half> %v
+}
+
+define <vscale x 32 x half> @vfma_vv_nxv32f16_unmasked(<vscale x 32 x half> %va, <vscale x 32 x half> %b, <vscale x 32 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv32f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vl8re16.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e16, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v16, v24
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 32 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 32 x i1> %head, <vscale x 32 x i1> poison, <vscale x 32 x i32> zeroinitializer
+  %v = call <vscale x 32 x half> @llvm.vp.fmuladd.nxv32f16(<vscale x 32 x half> %va, <vscale x 32 x half> %b, <vscale x 32 x half> %c, <vscale x 32 x i1> %m, i32 %evl)
+  ret <vscale x 32 x half> %v
+}
+
+define <vscale x 32 x half> @vfma_vf_nxv32f16(<vscale x 32 x half> %va, half %b, <vscale x 32 x half> %vc, <vscale x 32 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv32f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 32 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 32 x half> %elt.head, <vscale x 32 x half> poison, <vscale x 32 x i32> zeroinitializer
+  %v = call <vscale x 32 x half> @llvm.vp.fmuladd.nxv32f16(<vscale x 32 x half> %va, <vscale x 32 x half> %vb, <vscale x 32 x half> %vc, <vscale x 32 x i1> %m, i32 %evl)
+  ret <vscale x 32 x half> %v
+}
+
+define <vscale x 32 x half> @vfma_vf_nxv32f16_commute(<vscale x 32 x half> %va, half %b, <vscale x 32 x half> %vc, <vscale x 32 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv32f16_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 32 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 32 x half> %elt.head, <vscale x 32 x half> poison, <vscale x 32 x i32> zeroinitializer
+  %v = call <vscale x 32 x half> @llvm.vp.fmuladd.nxv32f16(<vscale x 32 x half> %vb, <vscale x 32 x half> %va, <vscale x 32 x half> %vc, <vscale x 32 x i1> %m, i32 %evl)
+  ret <vscale x 32 x half> %v
+}
+
+define <vscale x 32 x half> @vfma_vf_nxv32f16_unmasked(<vscale x 32 x half> %va, half %b, <vscale x 32 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv32f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 32 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 32 x half> %elt.head, <vscale x 32 x half> poison, <vscale x 32 x i32> zeroinitializer
+  %head = insertelement <vscale x 32 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 32 x i1> %head, <vscale x 32 x i1> poison, <vscale x 32 x i32> zeroinitializer
+  %v = call <vscale x 32 x half> @llvm.vp.fmuladd.nxv32f16(<vscale x 32 x half> %va, <vscale x 32 x half> %vb, <vscale x 32 x half> %vc, <vscale x 32 x i1> %m, i32 %evl)
+  ret <vscale x 32 x half> %v
+}
+
+define <vscale x 32 x half> @vfma_vf_nxv32f16_unmasked_commute(<vscale x 32 x half> %va, half %b, <vscale x 32 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv32f16_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 32 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 32 x half> %elt.head, <vscale x 32 x half> poison, <vscale x 32 x i32> zeroinitializer
+  %head = insertelement <vscale x 32 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 32 x i1> %head, <vscale x 32 x i1> poison, <vscale x 32 x i32> zeroinitializer
+  %v = call <vscale x 32 x half> @llvm.vp.fmuladd.nxv32f16(<vscale x 32 x half> %vb, <vscale x 32 x half> %va, <vscale x 32 x half> %vc, <vscale x 32 x i1> %m, i32 %evl)
+  ret <vscale x 32 x half> %v
+}
+
+declare <vscale x 1 x float> @llvm.vp.fmuladd.nxv1f32(<vscale x 1 x float>, <vscale x 1 x float>, <vscale x 1 x float>, <vscale x 1 x i1>, i32)
+
+define <vscale x 1 x float> @vfma_vv_nxv1f32(<vscale x 1 x float> %va, <vscale x 1 x float> %b, <vscale x 1 x float> %c, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv1f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv1r.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <vscale x 1 x float> @llvm.vp.fmuladd.nxv1f32(<vscale x 1 x float> %va, <vscale x 1 x float> %b, <vscale x 1 x float> %c, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x float> %v
+}
+
+define <vscale x 1 x float> @vfma_vv_nxv1f32_unmasked(<vscale x 1 x float> %va, <vscale x 1 x float> %b, <vscale x 1 x float> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv1f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x float> @llvm.vp.fmuladd.nxv1f32(<vscale x 1 x float> %va, <vscale x 1 x float> %b, <vscale x 1 x float> %c, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x float> %v
+}
+
+define <vscale x 1 x float> @vfma_vf_nxv1f32(<vscale x 1 x float> %va, float %b, <vscale x 1 x float> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 1 x float> %elt.head, <vscale x 1 x float> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x float> @llvm.vp.fmuladd.nxv1f32(<vscale x 1 x float> %va, <vscale x 1 x float> %vb, <vscale x 1 x float> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x float> %v
+}
+
+define <vscale x 1 x float> @vfma_vf_nxv1f32_commute(<vscale x 1 x float> %va, float %b, <vscale x 1 x float> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f32_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 1 x float> %elt.head, <vscale x 1 x float> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x float> @llvm.vp.fmuladd.nxv1f32(<vscale x 1 x float> %vb, <vscale x 1 x float> %va, <vscale x 1 x float> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x float> %v
+}
+
+define <vscale x 1 x float> @vfma_vf_nxv1f32_unmasked(<vscale x 1 x float> %va, float %b, <vscale x 1 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 1 x float> %elt.head, <vscale x 1 x float> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x float> @llvm.vp.fmuladd.nxv1f32(<vscale x 1 x float> %va, <vscale x 1 x float> %vb, <vscale x 1 x float> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x float> %v
+}
+
+define <vscale x 1 x float> @vfma_vf_nxv1f32_unmasked_commute(<vscale x 1 x float> %va, float %b, <vscale x 1 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f32_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 1 x float> %elt.head, <vscale x 1 x float> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x float> @llvm.vp.fmuladd.nxv1f32(<vscale x 1 x float> %vb, <vscale x 1 x float> %va, <vscale x 1 x float> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x float> %v
+}
+
+declare <vscale x 2 x float> @llvm.vp.fmuladd.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, <vscale x 2 x float>, <vscale x 2 x i1>, i32)
+
+define <vscale x 2 x float> @vfma_vv_nxv2f32(<vscale x 2 x float> %va, <vscale x 2 x float> %b, <vscale x 2 x float> %c, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv2f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <vscale x 2 x float> @llvm.vp.fmuladd.nxv2f32(<vscale x 2 x float> %va, <vscale x 2 x float> %b, <vscale x 2 x float> %c, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x float> %v
+}
+
+define <vscale x 2 x float> @vfma_vv_nxv2f32_unmasked(<vscale x 2 x float> %va, <vscale x 2 x float> %b, <vscale x 2 x float> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv2f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x float> @llvm.vp.fmuladd.nxv2f32(<vscale x 2 x float> %va, <vscale x 2 x float> %b, <vscale x 2 x float> %c, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x float> %v
+}
+
+define <vscale x 2 x float> @vfma_vf_nxv2f32(<vscale x 2 x float> %va, float %b, <vscale x 2 x float> %vc, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 2 x float> %elt.head, <vscale x 2 x float> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x float> @llvm.vp.fmuladd.nxv2f32(<vscale x 2 x float> %va, <vscale x 2 x float> %vb, <vscale x 2 x float> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x float> %v
+}
+
+define <vscale x 2 x float> @vfma_vf_nxv2f32_commute(<vscale x 2 x float> %va, float %b, <vscale x 2 x float> %vc, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f32_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 2 x float> %elt.head, <vscale x 2 x float> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x float> @llvm.vp.fmuladd.nxv2f32(<vscale x 2 x float> %vb, <vscale x 2 x float> %va, <vscale x 2 x float> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x float> %v
+}
+
+define <vscale x 2 x float> @vfma_vf_nxv2f32_unmasked(<vscale x 2 x float> %va, float %b, <vscale x 2 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 2 x float> %elt.head, <vscale x 2 x float> poison, <vscale x 2 x i32> zeroinitializer
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x float> @llvm.vp.fmuladd.nxv2f32(<vscale x 2 x float> %va, <vscale x 2 x float> %vb, <vscale x 2 x float> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x float> %v
+}
+
+define <vscale x 2 x float> @vfma_vf_nxv2f32_unmasked_commute(<vscale x 2 x float> %va, float %b, <vscale x 2 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f32_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 2 x float> %elt.head, <vscale x 2 x float> poison, <vscale x 2 x i32> zeroinitializer
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x float> @llvm.vp.fmuladd.nxv2f32(<vscale x 2 x float> %vb, <vscale x 2 x float> %va, <vscale x 2 x float> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x float> %v
+}
+
+declare <vscale x 4 x float> @llvm.vp.fmuladd.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x i1>, i32)
+
+define <vscale x 4 x float> @vfma_vv_nxv4f32(<vscale x 4 x float> %va, <vscale x 4 x float> %b, <vscale x 4 x float> %c, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv4f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v10, v8, v12, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v10
+; CHECK-NEXT:    ret
+  %v = call <vscale x 4 x float> @llvm.vp.fmuladd.nxv4f32(<vscale x 4 x float> %va, <vscale x 4 x float> %b, <vscale x 4 x float> %c, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x float> %v
+}
+
+define <vscale x 4 x float> @vfma_vv_nxv4f32_unmasked(<vscale x 4 x float> %va, <vscale x 4 x float> %b, <vscale x 4 x float> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv4f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v10, v12
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x float> @llvm.vp.fmuladd.nxv4f32(<vscale x 4 x float> %va, <vscale x 4 x float> %b, <vscale x 4 x float> %c, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x float> %v
+}
+
+define <vscale x 4 x float> @vfma_vf_nxv4f32(<vscale x 4 x float> %va, float %b, <vscale x 4 x float> %vc, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 4 x float> %elt.head, <vscale x 4 x float> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x float> @llvm.vp.fmuladd.nxv4f32(<vscale x 4 x float> %va, <vscale x 4 x float> %vb, <vscale x 4 x float> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x float> %v
+}
+
+define <vscale x 4 x float> @vfma_vf_nxv4f32_commute(<vscale x 4 x float> %va, float %b, <vscale x 4 x float> %vc, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f32_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 4 x float> %elt.head, <vscale x 4 x float> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x float> @llvm.vp.fmuladd.nxv4f32(<vscale x 4 x float> %vb, <vscale x 4 x float> %va, <vscale x 4 x float> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x float> %v
+}
+
+define <vscale x 4 x float> @vfma_vf_nxv4f32_unmasked(<vscale x 4 x float> %va, float %b, <vscale x 4 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 4 x float> %elt.head, <vscale x 4 x float> poison, <vscale x 4 x i32> zeroinitializer
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x float> @llvm.vp.fmuladd.nxv4f32(<vscale x 4 x float> %va, <vscale x 4 x float> %vb, <vscale x 4 x float> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x float> %v
+}
+
+define <vscale x 4 x float> @vfma_vf_nxv4f32_unmasked_commute(<vscale x 4 x float> %va, float %b, <vscale x 4 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f32_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 4 x float> %elt.head, <vscale x 4 x float> poison, <vscale x 4 x i32> zeroinitializer
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x float> @llvm.vp.fmuladd.nxv4f32(<vscale x 4 x float> %vb, <vscale x 4 x float> %va, <vscale x 4 x float> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x float> %v
+}
+
+declare <vscale x 8 x float> @llvm.vp.fmuladd.nxv8f32(<vscale x 8 x float>, <vscale x 8 x float>, <vscale x 8 x float>, <vscale x 8 x i1>, i32)
+
+define <vscale x 8 x float> @vfma_vv_nxv8f32(<vscale x 8 x float> %va, <vscale x 8 x float> %b, <vscale x 8 x float> %c, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv8f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v12, v8, v16, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v12
+; CHECK-NEXT:    ret
+  %v = call <vscale x 8 x float> @llvm.vp.fmuladd.nxv8f32(<vscale x 8 x float> %va, <vscale x 8 x float> %b, <vscale x 8 x float> %c, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x float> %v
+}
+
+define <vscale x 8 x float> @vfma_vv_nxv8f32_unmasked(<vscale x 8 x float> %va, <vscale x 8 x float> %b, <vscale x 8 x float> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv8f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v12, v16
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x float> @llvm.vp.fmuladd.nxv8f32(<vscale x 8 x float> %va, <vscale x 8 x float> %b, <vscale x 8 x float> %c, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x float> %v
+}
+
+define <vscale x 8 x float> @vfma_vf_nxv8f32(<vscale x 8 x float> %va, float %b, <vscale x 8 x float> %vc, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 8 x float> %elt.head, <vscale x 8 x float> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x float> @llvm.vp.fmuladd.nxv8f32(<vscale x 8 x float> %va, <vscale x 8 x float> %vb, <vscale x 8 x float> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x float> %v
+}
+
+define <vscale x 8 x float> @vfma_vf_nxv8f32_commute(<vscale x 8 x float> %va, float %b, <vscale x 8 x float> %vc, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f32_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 8 x float> %elt.head, <vscale x 8 x float> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x float> @llvm.vp.fmuladd.nxv8f32(<vscale x 8 x float> %vb, <vscale x 8 x float> %va, <vscale x 8 x float> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x float> %v
+}
+
+define <vscale x 8 x float> @vfma_vf_nxv8f32_unmasked(<vscale x 8 x float> %va, float %b, <vscale x 8 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 8 x float> %elt.head, <vscale x 8 x float> poison, <vscale x 8 x i32> zeroinitializer
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x float> @llvm.vp.fmuladd.nxv8f32(<vscale x 8 x float> %va, <vscale x 8 x float> %vb, <vscale x 8 x float> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x float> %v
+}
+
+define <vscale x 8 x float> @vfma_vf_nxv8f32_unmasked_commute(<vscale x 8 x float> %va, float %b, <vscale x 8 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f32_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 8 x float> %elt.head, <vscale x 8 x float> poison, <vscale x 8 x i32> zeroinitializer
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x float> @llvm.vp.fmuladd.nxv8f32(<vscale x 8 x float> %vb, <vscale x 8 x float> %va, <vscale x 8 x float> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x float> %v
+}
+
+declare <vscale x 16 x float> @llvm.vp.fmuladd.nxv16f32(<vscale x 16 x float>, <vscale x 16 x float>, <vscale x 16 x float>, <vscale x 16 x i1>, i32)
+
+define <vscale x 16 x float> @vfma_vv_nxv16f32(<vscale x 16 x float> %va, <vscale x 16 x float> %b, <vscale x 16 x float> %c, <vscale x 16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv16f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vl8re32.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e32, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v16, v8, v24, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v16
+; CHECK-NEXT:    ret
+  %v = call <vscale x 16 x float> @llvm.vp.fmuladd.nxv16f32(<vscale x 16 x float> %va, <vscale x 16 x float> %b, <vscale x 16 x float> %c, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x float> %v
+}
+
+define <vscale x 16 x float> @vfma_vv_nxv16f32_unmasked(<vscale x 16 x float> %va, <vscale x 16 x float> %b, <vscale x 16 x float> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv16f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vl8re32.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e32, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v16, v24
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 16 x i1> %head, <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x float> @llvm.vp.fmuladd.nxv16f32(<vscale x 16 x float> %va, <vscale x 16 x float> %b, <vscale x 16 x float> %c, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x float> %v
+}
+
+define <vscale x 16 x float> @vfma_vf_nxv16f32(<vscale x 16 x float> %va, float %b, <vscale x 16 x float> %vc, <vscale x 16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv16f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 16 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 16 x float> %elt.head, <vscale x 16 x float> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x float> @llvm.vp.fmuladd.nxv16f32(<vscale x 16 x float> %va, <vscale x 16 x float> %vb, <vscale x 16 x float> %vc, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x float> %v
+}
+
+define <vscale x 16 x float> @vfma_vf_nxv16f32_commute(<vscale x 16 x float> %va, float %b, <vscale x 16 x float> %vc, <vscale x 16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv16f32_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 16 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 16 x float> %elt.head, <vscale x 16 x float> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x float> @llvm.vp.fmuladd.nxv16f32(<vscale x 16 x float> %vb, <vscale x 16 x float> %va, <vscale x 16 x float> %vc, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x float> %v
+}
+
+define <vscale x 16 x float> @vfma_vf_nxv16f32_unmasked(<vscale x 16 x float> %va, float %b, <vscale x 16 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv16f32_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 16 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 16 x float> %elt.head, <vscale x 16 x float> poison, <vscale x 16 x i32> zeroinitializer
+  %head = insertelement <vscale x 16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 16 x i1> %head, <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x float> @llvm.vp.fmuladd.nxv16f32(<vscale x 16 x float> %va, <vscale x 16 x float> %vb, <vscale x 16 x float> %vc, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x float> %v
+}
+
+define <vscale x 16 x float> @vfma_vf_nxv16f32_unmasked_commute(<vscale x 16 x float> %va, float %b, <vscale x 16 x float> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv16f32_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 16 x float> poison, float %b, i32 0
+  %vb = shufflevector <vscale x 16 x float> %elt.head, <vscale x 16 x float> poison, <vscale x 16 x i32> zeroinitializer
+  %head = insertelement <vscale x 16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 16 x i1> %head, <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x float> @llvm.vp.fmuladd.nxv16f32(<vscale x 16 x float> %vb, <vscale x 16 x float> %va, <vscale x 16 x float> %vc, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x float> %v
+}
+
+declare <vscale x 1 x double> @llvm.vp.fmuladd.nxv1f64(<vscale x 1 x double>, <vscale x 1 x double>, <vscale x 1 x double>, <vscale x 1 x i1>, i32)
+
+define <vscale x 1 x double> @vfma_vv_nxv1f64(<vscale x 1 x double> %va, <vscale x 1 x double> %b, <vscale x 1 x double> %c, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv1f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v9
+; CHECK-NEXT:    ret
+  %v = call <vscale x 1 x double> @llvm.vp.fmuladd.nxv1f64(<vscale x 1 x double> %va, <vscale x 1 x double> %b, <vscale x 1 x double> %c, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x double> %v
+}
+
+define <vscale x 1 x double> @vfma_vv_nxv1f64_unmasked(<vscale x 1 x double> %va, <vscale x 1 x double> %b, <vscale x 1 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv1f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x double> @llvm.vp.fmuladd.nxv1f64(<vscale x 1 x double> %va, <vscale x 1 x double> %b, <vscale x 1 x double> %c, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x double> %v
+}
+
+define <vscale x 1 x double> @vfma_vf_nxv1f64(<vscale x 1 x double> %va, double %b, <vscale x 1 x double> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 1 x double> %elt.head, <vscale x 1 x double> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x double> @llvm.vp.fmuladd.nxv1f64(<vscale x 1 x double> %va, <vscale x 1 x double> %vb, <vscale x 1 x double> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x double> %v
+}
+
+define <vscale x 1 x double> @vfma_vf_nxv1f64_commute(<vscale x 1 x double> %va, double %b, <vscale x 1 x double> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f64_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 1 x double> %elt.head, <vscale x 1 x double> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x double> @llvm.vp.fmuladd.nxv1f64(<vscale x 1 x double> %vb, <vscale x 1 x double> %va, <vscale x 1 x double> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x double> %v
+}
+
+define <vscale x 1 x double> @vfma_vf_nxv1f64_unmasked(<vscale x 1 x double> %va, double %b, <vscale x 1 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 1 x double> %elt.head, <vscale x 1 x double> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x double> @llvm.vp.fmuladd.nxv1f64(<vscale x 1 x double> %va, <vscale x 1 x double> %vb, <vscale x 1 x double> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x double> %v
+}
+
+define <vscale x 1 x double> @vfma_vf_nxv1f64_unmasked_commute(<vscale x 1 x double> %va, double %b, <vscale x 1 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv1f64_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 1 x double> %elt.head, <vscale x 1 x double> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %v = call <vscale x 1 x double> @llvm.vp.fmuladd.nxv1f64(<vscale x 1 x double> %vb, <vscale x 1 x double> %va, <vscale x 1 x double> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x double> %v
+}
+
+declare <vscale x 2 x double> @llvm.vp.fmuladd.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x i1>, i32)
+
+define <vscale x 2 x double> @vfma_vv_nxv2f64(<vscale x 2 x double> %va, <vscale x 2 x double> %b, <vscale x 2 x double> %c, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv2f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v10, v8, v12, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v10
+; CHECK-NEXT:    ret
+  %v = call <vscale x 2 x double> @llvm.vp.fmuladd.nxv2f64(<vscale x 2 x double> %va, <vscale x 2 x double> %b, <vscale x 2 x double> %c, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x double> %v
+}
+
+define <vscale x 2 x double> @vfma_vv_nxv2f64_unmasked(<vscale x 2 x double> %va, <vscale x 2 x double> %b, <vscale x 2 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv2f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v10, v12
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x double> @llvm.vp.fmuladd.nxv2f64(<vscale x 2 x double> %va, <vscale x 2 x double> %b, <vscale x 2 x double> %c, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x double> %v
+}
+
+define <vscale x 2 x double> @vfma_vf_nxv2f64(<vscale x 2 x double> %va, double %b, <vscale x 2 x double> %vc, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 2 x double> %elt.head, <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x double> @llvm.vp.fmuladd.nxv2f64(<vscale x 2 x double> %va, <vscale x 2 x double> %vb, <vscale x 2 x double> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x double> %v
+}
+
+define <vscale x 2 x double> @vfma_vf_nxv2f64_commute(<vscale x 2 x double> %va, double %b, <vscale x 2 x double> %vc, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f64_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 2 x double> %elt.head, <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x double> @llvm.vp.fmuladd.nxv2f64(<vscale x 2 x double> %vb, <vscale x 2 x double> %va, <vscale x 2 x double> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x double> %v
+}
+
+define <vscale x 2 x double> @vfma_vf_nxv2f64_unmasked(<vscale x 2 x double> %va, double %b, <vscale x 2 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 2 x double> %elt.head, <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x double> @llvm.vp.fmuladd.nxv2f64(<vscale x 2 x double> %va, <vscale x 2 x double> %vb, <vscale x 2 x double> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x double> %v
+}
+
+define <vscale x 2 x double> @vfma_vf_nxv2f64_unmasked_commute(<vscale x 2 x double> %va, double %b, <vscale x 2 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv2f64_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v10
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 2 x double> %elt.head, <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %v = call <vscale x 2 x double> @llvm.vp.fmuladd.nxv2f64(<vscale x 2 x double> %vb, <vscale x 2 x double> %va, <vscale x 2 x double> %vc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x double> %v
+}
+
+declare <vscale x 4 x double> @llvm.vp.fmuladd.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>, <vscale x 4 x double>, <vscale x 4 x i1>, i32)
+
+define <vscale x 4 x double> @vfma_vv_nxv4f64(<vscale x 4 x double> %va, <vscale x 4 x double> %b, <vscale x 4 x double> %c, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv4f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v12, v8, v16, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v12
+; CHECK-NEXT:    ret
+  %v = call <vscale x 4 x double> @llvm.vp.fmuladd.nxv4f64(<vscale x 4 x double> %va, <vscale x 4 x double> %b, <vscale x 4 x double> %c, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x double> %v
+}
+
+define <vscale x 4 x double> @vfma_vv_nxv4f64_unmasked(<vscale x 4 x double> %va, <vscale x 4 x double> %b, <vscale x 4 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv4f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v12, v16
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x double> @llvm.vp.fmuladd.nxv4f64(<vscale x 4 x double> %va, <vscale x 4 x double> %b, <vscale x 4 x double> %c, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x double> %v
+}
+
+define <vscale x 4 x double> @vfma_vf_nxv4f64(<vscale x 4 x double> %va, double %b, <vscale x 4 x double> %vc, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 4 x double> %elt.head, <vscale x 4 x double> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x double> @llvm.vp.fmuladd.nxv4f64(<vscale x 4 x double> %va, <vscale x 4 x double> %vb, <vscale x 4 x double> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x double> %v
+}
+
+define <vscale x 4 x double> @vfma_vf_nxv4f64_commute(<vscale x 4 x double> %va, double %b, <vscale x 4 x double> %vc, <vscale x 4 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f64_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 4 x double> %elt.head, <vscale x 4 x double> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x double> @llvm.vp.fmuladd.nxv4f64(<vscale x 4 x double> %vb, <vscale x 4 x double> %va, <vscale x 4 x double> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x double> %v
+}
+
+define <vscale x 4 x double> @vfma_vf_nxv4f64_unmasked(<vscale x 4 x double> %va, double %b, <vscale x 4 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 4 x double> %elt.head, <vscale x 4 x double> poison, <vscale x 4 x i32> zeroinitializer
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x double> @llvm.vp.fmuladd.nxv4f64(<vscale x 4 x double> %va, <vscale x 4 x double> %vb, <vscale x 4 x double> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x double> %v
+}
+
+define <vscale x 4 x double> @vfma_vf_nxv4f64_unmasked_commute(<vscale x 4 x double> %va, double %b, <vscale x 4 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv4f64_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v12
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 4 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 4 x double> %elt.head, <vscale x 4 x double> poison, <vscale x 4 x i32> zeroinitializer
+  %head = insertelement <vscale x 4 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 4 x i1> %head, <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer
+  %v = call <vscale x 4 x double> @llvm.vp.fmuladd.nxv4f64(<vscale x 4 x double> %vb, <vscale x 4 x double> %va, <vscale x 4 x double> %vc, <vscale x 4 x i1> %m, i32 %evl)
+  ret <vscale x 4 x double> %v
+}
+
+declare <vscale x 7 x double> @llvm.vp.fmuladd.nxv7f64(<vscale x 7 x double>, <vscale x 7 x double>, <vscale x 7 x double>, <vscale x 7 x i1>, i32)
+
+define <vscale x 7 x double> @vfma_vv_nxv7f64(<vscale x 7 x double> %va, <vscale x 7 x double> %b, <vscale x 7 x double> %c, <vscale x 7 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv7f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v16, v8, v24, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v16
+; CHECK-NEXT:    ret
+  %v = call <vscale x 7 x double> @llvm.vp.fmuladd.nxv7f64(<vscale x 7 x double> %va, <vscale x 7 x double> %b, <vscale x 7 x double> %c, <vscale x 7 x i1> %m, i32 %evl)
+  ret <vscale x 7 x double> %v
+}
+
+define <vscale x 7 x double> @vfma_vv_nxv7f64_unmasked(<vscale x 7 x double> %va, <vscale x 7 x double> %b, <vscale x 7 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv7f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v16, v24
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 7 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 7 x i1> %head, <vscale x 7 x i1> poison, <vscale x 7 x i32> zeroinitializer
+  %v = call <vscale x 7 x double> @llvm.vp.fmuladd.nxv7f64(<vscale x 7 x double> %va, <vscale x 7 x double> %b, <vscale x 7 x double> %c, <vscale x 7 x i1> %m, i32 %evl)
+  ret <vscale x 7 x double> %v
+}
+
+declare <vscale x 8 x double> @llvm.vp.fmuladd.nxv8f64(<vscale x 8 x double>, <vscale x 8 x double>, <vscale x 8 x double>, <vscale x 8 x i1>, i32)
+
+define <vscale x 8 x double> @vfma_vv_nxv8f64(<vscale x 8 x double> %va, <vscale x 8 x double> %b, <vscale x 8 x double> %c, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv8f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v16, v8, v24, v0.t
+; CHECK-NEXT:    vmv.v.v v8, v16
+; CHECK-NEXT:    ret
+  %v = call <vscale x 8 x double> @llvm.vp.fmuladd.nxv8f64(<vscale x 8 x double> %va, <vscale x 8 x double> %b, <vscale x 8 x double> %c, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x double> %v
+}
+
+define <vscale x 8 x double> @vfma_vv_nxv8f64_unmasked(<vscale x 8 x double> %va, <vscale x 8 x double> %b, <vscale x 8 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv8f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vl8re64.v v24, (a0)
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vv v8, v16, v24
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x double> @llvm.vp.fmuladd.nxv8f64(<vscale x 8 x double> %va, <vscale x 8 x double> %b, <vscale x 8 x double> %c, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x double> %v
+}
+
+define <vscale x 8 x double> @vfma_vf_nxv8f64(<vscale x 8 x double> %va, double %b, <vscale x 8 x double> %vc, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 8 x double> %elt.head, <vscale x 8 x double> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x double> @llvm.vp.fmuladd.nxv8f64(<vscale x 8 x double> %va, <vscale x 8 x double> %vb, <vscale x 8 x double> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x double> %v
+}
+
+define <vscale x 8 x double> @vfma_vf_nxv8f64_commute(<vscale x 8 x double> %va, double %b, <vscale x 8 x double> %vc, <vscale x 8 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f64_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 8 x double> %elt.head, <vscale x 8 x double> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x double> @llvm.vp.fmuladd.nxv8f64(<vscale x 8 x double> %vb, <vscale x 8 x double> %va, <vscale x 8 x double> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x double> %v
+}
+
+define <vscale x 8 x double> @vfma_vf_nxv8f64_unmasked(<vscale x 8 x double> %va, double %b, <vscale x 8 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 8 x double> %elt.head, <vscale x 8 x double> poison, <vscale x 8 x i32> zeroinitializer
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x double> @llvm.vp.fmuladd.nxv8f64(<vscale x 8 x double> %va, <vscale x 8 x double> %vb, <vscale x 8 x double> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x double> %v
+}
+
+define <vscale x 8 x double> @vfma_vf_nxv8f64_unmasked_commute(<vscale x 8 x double> %va, double %b, <vscale x 8 x double> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vf_nxv8f64_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, mu
+; CHECK-NEXT:    vfmadd.vf v8, fa0, v16
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 8 x double> poison, double %b, i32 0
+  %vb = shufflevector <vscale x 8 x double> %elt.head, <vscale x 8 x double> poison, <vscale x 8 x i32> zeroinitializer
+  %head = insertelement <vscale x 8 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 8 x i1> %head, <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer
+  %v = call <vscale x 8 x double> @llvm.vp.fmuladd.nxv8f64(<vscale x 8 x double> %vb, <vscale x 8 x double> %va, <vscale x 8 x double> %vc, <vscale x 8 x i1> %m, i32 %evl)
+  ret <vscale x 8 x double> %v
+}
+
+declare <vscale x 16 x double> @llvm.vp.fmuladd.nxv16f64(<vscale x 16 x double>, <vscale x 16 x double>, <vscale x 16 x double>, <vscale x 16 x i1>, i32)
+
+define <vscale x 16 x double> @vfma_vv_nxv16f64(<vscale x 16 x double> %va, <vscale x 16 x double> %b, <vscale x 16 x double> %c, <vscale x 16 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv16f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a3, 48
+; CHECK-NEXT:    mul a1, a1, a3
+; CHECK-NEXT:    sub sp, sp, a1
+; CHECK-NEXT:    vmv1r.v v1, v0
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a3, 24
+; CHECK-NEXT:    mul a1, a1, a3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 5
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    li a3, 0
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    vsetvli a5, zero, e8, mf4, ta, mu
+; CHECK-NEXT:    slli a5, a1, 3
+; CHECK-NEXT:    add a6, a2, a5
+; CHECK-NEXT:    vl8re64.v v8, (a6)
+; CHECK-NEXT:    csrr a6, vlenb
+; CHECK-NEXT:    slli a6, a6, 3
+; CHECK-NEXT:    add a6, sp, a6
+; CHECK-NEXT:    addi a6, a6, 16
+; CHECK-NEXT:    vs8r.v v8, (a6) # Unknown-size Folded Spill
+; CHECK-NEXT:    add a5, a0, a5
+; CHECK-NEXT:    vl8re64.v v8, (a5)
+; CHECK-NEXT:    csrr a5, vlenb
+; CHECK-NEXT:    li a6, 40
+; CHECK-NEXT:    mul a5, a5, a6
+; CHECK-NEXT:    add a5, sp, a5
+; CHECK-NEXT:    addi a5, a5, 16
+; CHECK-NEXT:    vs8r.v v8, (a5) # Unknown-size Folded Spill
+; CHECK-NEXT:    srli a6, a1, 3
+; CHECK-NEXT:    sub a5, a4, a1
+; CHECK-NEXT:    vslidedown.vx v0, v0, a6
+; CHECK-NEXT:    bltu a4, a5, .LBB92_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv a3, a5
+; CHECK-NEXT:  .LBB92_2:
+; CHECK-NEXT:    vl8re64.v v8, (a2)
+; CHECK-NEXT:    csrr a2, vlenb
+; CHECK-NEXT:    slli a2, a2, 4
+; CHECK-NEXT:    add a2, sp, a2
+; CHECK-NEXT:    addi a2, a2, 16
+; CHECK-NEXT:    vs8r.v v8, (a2) # Unknown-size Folded Spill
+; CHECK-NEXT:    vl8re64.v v8, (a0)
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli zero, a3, e64, m8, ta, mu
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a2, 24
+; CHECK-NEXT:    mul a0, a0, a2
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v24, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v16, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a2, 40
+; CHECK-NEXT:    mul a0, a0, a2
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v8, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vfmadd.vv v8, v24, v16, v0.t
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a2, 40
+; CHECK-NEXT:    mul a0, a0, a2
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
+; CHECK-NEXT:    bltu a4, a1, .LBB92_4
+; CHECK-NEXT:  # %bb.3:
+; CHECK-NEXT:    mv a4, a1
+; CHECK-NEXT:  .LBB92_4:
+; CHECK-NEXT:    vsetvli zero, a4, e64, m8, ta, mu
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 5
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v24, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 4
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v16, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vl8re8.v v8, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vfmadd.vv v8, v24, v16, v0.t
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a1, 40
+; CHECK-NEXT:    mul a0, a0, a1
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v16, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a1, 48
+; CHECK-NEXT:    mul a0, a0, a1
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    addi sp, sp, 16
+; CHECK-NEXT:    ret
+  %v = call <vscale x 16 x double> @llvm.vp.fmuladd.nxv16f64(<vscale x 16 x double> %va, <vscale x 16 x double> %b, <vscale x 16 x double> %c, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x double> %v
+}
+
+define <vscale x 16 x double> @vfma_vv_nxv16f64_unmasked(<vscale x 16 x double> %va, <vscale x 16 x double> %b, <vscale x 16 x double> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfma_vv_nxv16f64_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a3, 24
+; CHECK-NEXT:    mul a1, a1, a3
+; CHECK-NEXT:    sub sp, sp, a1
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a3, a1, 3
+; CHECK-NEXT:    add a5, a2, a3
+; CHECK-NEXT:    vl8re64.v v24, (a5)
+; CHECK-NEXT:    csrr a5, vlenb
+; CHECK-NEXT:    slli a5, a5, 3
+; CHECK-NEXT:    add a5, sp, a5
+; CHECK-NEXT:    addi a5, a5, 16
+; CHECK-NEXT:    vs8r.v v24, (a5) # Unknown-size Folded Spill
+; CHECK-NEXT:    add a3, a0, a3
+; CHECK-NEXT:    vl8re64.v v24, (a3)
+; CHECK-NEXT:    sub a5, a4, a1
+; CHECK-NEXT:    csrr a3, vlenb
+; CHECK-NEXT:    slli a3, a3, 4
+; CHECK-NEXT:    add a3, sp, a3
+; CHECK-NEXT:    addi a3, a3, 16
+; CHECK-NEXT:    vs8r.v v8, (a3) # Unknown-size Folded Spill
+; CHECK-NEXT:    li a3, 0
+; CHECK-NEXT:    bltu a4, a5, .LBB93_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv a3, a5
+; CHECK-NEXT:  .LBB93_2:
+; CHECK-NEXT:    vl8re64.v v8, (a2)
+; CHECK-NEXT:    addi a2, sp, 16
+; CHECK-NEXT:    vs8r.v v8, (a2) # Unknown-size Folded Spill
+; CHECK-NEXT:    vl8re64.v v0, (a0)
+; CHECK-NEXT:    vsetvli zero, a3, e64, m8, ta, mu
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v8, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vfmadd.vv v24, v16, v8
+; CHECK-NEXT:    bltu a4, a1, .LBB93_4
+; CHECK-NEXT:  # %bb.3:
+; CHECK-NEXT:    mv a4, a1
+; CHECK-NEXT:  .LBB93_4:
+; CHECK-NEXT:    vsetvli zero, a4, e64, m8, ta, mu
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 4
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v16, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vl8re8.v v8, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vfmadd.vv v0, v16, v8
+; CHECK-NEXT:    vmv.v.v v8, v0
+; CHECK-NEXT:    vmv8r.v v16, v24
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a1, 24
+; CHECK-NEXT:    mul a0, a0, a1
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    addi sp, sp, 16
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 16 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 16 x i1> %head, <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer
+  %v = call <vscale x 16 x double> @llvm.vp.fmuladd.nxv16f64(<vscale x 16 x double> %va, <vscale x 16 x double> %b, <vscale x 16 x double> %c, <vscale x 16 x i1> %m, i32 %evl)
+  ret <vscale x 16 x double> %v
+}
+
+declare <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half>, <vscale x 1 x i1>, i32)
+
+define <vscale x 1 x half> @vfmsub_vv_nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfmsub_vv_nxv1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmsub.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv1r.v v8, v9
+; CHECK-NEXT:    ret
+  %negc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %negc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfmsub_vv_nxv1f16_unmasked(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfmsub_vv_nxv1f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmsub.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %negc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfmsub_vf_nxv1f16(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfmsub_vf_nxv1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmsub.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %vb, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfmsub_vf_nxv1f16_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfmsub_vf_nxv1f16_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmsub.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x half> %va, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfmsub_vf_nxv1f16_unmasked(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfmsub_vf_nxv1f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmsub.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %vb, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfmsub_vf_nxv1f16_unmasked_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfmsub_vf_nxv1f16_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfmsub.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x half> %va, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vv_nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vv_nxv1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv1r.v v8, v9
+; CHECK-NEXT:    ret
+  %negb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %b, <vscale x 1 x i1> %m, i32 %evl)
+  %negc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %negb, <vscale x 1 x half> %negc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vv_nxv1f16_commuted(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vv_nxv1f16_commuted:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    ret
+  %negb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %b, <vscale x 1 x i1> %m, i32 %evl)
+  %negc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negb, <vscale x 1 x half> %va, <vscale x 1 x half> %negc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vv_nxv1f16_unmasked(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vv_nxv1f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %b, <vscale x 1 x i1> %m, i32 %evl)
+  %negc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %negb, <vscale x 1 x half> %negc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vv_nxv1f16_unmasked_commuted(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vv_nxv1f16_unmasked_commuted:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %b, <vscale x 1 x i1> %m, i32 %evl)
+  %negc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negb, <vscale x 1 x half> %va, <vscale x 1 x half> %negc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vf_nxv1f16(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vf_nxv1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %negva = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 %evl)
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negva, <vscale x 1 x half> %vb, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vf_nxv1f16_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vf_nxv1f16_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %negva = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 %evl)
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x half> %negva, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vf_nxv1f16_unmasked(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vf_nxv1f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negva = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 %evl)
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negva, <vscale x 1 x half> %vb, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vf_nxv1f16_unmasked_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vf_nxv1f16_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negva = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 %evl)
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x half> %negva, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vf_nxv1f16_neg_splat(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vf_nxv1f16_neg_splat:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %negvb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x i1> %m, i32 %evl)
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %negvb, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vf_nxv1f16_neg_splat_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vf_nxv1f16_neg_splat_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %negvb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x i1> %m, i32 %evl)
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negvb, <vscale x 1 x half> %va, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vf_nxv1f16_neg_splat_unmasked(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vf_nxv1f16_neg_splat_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negvb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x i1> %m, i32 %evl)
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %negvb, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmadd_vf_nxv1f16_neg_splat_unmasked_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmadd_vf_nxv1f16_neg_splat_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negvb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x i1> %m, i32 %evl)
+  %negvc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negvb, <vscale x 1 x half> %va, <vscale x 1 x half> %negvc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vv_nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vv_nxv1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv1r.v v8, v9
+; CHECK-NEXT:    ret
+  %negb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %b, <vscale x 1 x i1> %m, i32 %evl)
+  %negc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %negb, <vscale x 1 x half> %negc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vv_nxv1f16_commuted(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vv_nxv1f16_commuted:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    ret
+  %negb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %b, <vscale x 1 x i1> %m, i32 %evl)
+  %negc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negb, <vscale x 1 x half> %va, <vscale x 1 x half> %negc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vv_nxv1f16_unmasked(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vv_nxv1f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %b, <vscale x 1 x i1> %m, i32 %evl)
+  %negc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %negb, <vscale x 1 x half> %negc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vv_nxv1f16_unmasked_commuted(<vscale x 1 x half> %va, <vscale x 1 x half> %b, <vscale x 1 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vv_nxv1f16_unmasked_commuted:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %b, <vscale x 1 x i1> %m, i32 %evl)
+  %negc = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %c, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negb, <vscale x 1 x half> %va, <vscale x 1 x half> %negc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vf_nxv1f16(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vf_nxv1f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmsub.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %negva = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negva, <vscale x 1 x half> %vb, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vf_nxv1f16_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vf_nxv1f16_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmsub.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %negva = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x half> %negva, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vf_nxv1f16_unmasked(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vf_nxv1f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmsub.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negva = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negva, <vscale x 1 x half> %vb, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vf_nxv1f16_unmasked_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vf_nxv1f16_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmsub.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negva = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x half> %negva, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vf_nxv1f16_neg_splat(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vf_nxv1f16_neg_splat:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmsub.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %negvb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %negvb, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vf_nxv1f16_neg_splat_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vf_nxv1f16_neg_splat_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmsub.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %negvb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negvb, <vscale x 1 x half> %va, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vf_nxv1f16_neg_splat_unmasked(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vf_nxv1f16_neg_splat_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmsub.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negvb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %negvb, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+define <vscale x 1 x half> @vfnmsub_vf_nxv1f16_neg_splat_unmasked_commute(<vscale x 1 x half> %va, half %b, <vscale x 1 x half> %vc, i32 zeroext %evl) {
+; CHECK-LABEL: vfnmsub_vf_nxv1f16_neg_splat_unmasked_commute:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vfnmsub.vf v8, fa0, v9
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
+  %head = insertelement <vscale x 1 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 1 x i1> %head, <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer
+  %negvb = call <vscale x 1 x half> @llvm.vp.fneg.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x i1> %m, i32 %evl)
+  %v = call <vscale x 1 x half> @llvm.vp.fmuladd.nxv1f16(<vscale x 1 x half> %negvb, <vscale x 1 x half> %va, <vscale x 1 x half> %vc, <vscale x 1 x i1> %m, i32 %evl)
+  ret <vscale x 1 x half> %v
+}
+
+declare <vscale x 2 x half> @llvm.vp.fneg.nxv2f16(<vscale x 2 x half>, <vscale x 2 x i1>, i32)
+
+define <vscale x 2 x half> @vfmsub_vv_nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x half> %b, <vscale x 2 x half> %c, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfmsub_vv_nxv2f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmsub.vv v9, v8, v10, v0.t
+; CHECK-NEXT:    vmv1r.v v8, v9
+; CHECK-NEXT:    ret
+  %negc = call <vscale x 2 x half> @llvm.vp.fneg.nxv2f16(<vscale x 2 x half> %c, <vscale x 2 x i1> %m, i32 %evl)
+  %v = call <vscale x 2 x half> @llvm.vp.fmuladd.nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x half> %b, <vscale x 2 x half> %negc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x half> %v
+}
+
+define <vscale x 2 x half> @vfmsub_vv_nxv2f16_unmasked(<vscale x 2 x half> %va, <vscale x 2 x half> %b, <vscale x 2 x half> %c, i32 zeroext %evl) {
+; CHECK-LABEL: vfmsub_vv_nxv2f16_unmasked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmsub.vv v8, v9, v10
+; CHECK-NEXT:    ret
+  %head = insertelement <vscale x 2 x i1> poison, i1 true, i32 0
+  %m = shufflevector <vscale x 2 x i1> %head, <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer
+  %negc = call <vscale x 2 x half> @llvm.vp.fneg.nxv2f16(<vscale x 2 x half> %c, <vscale x 2 x i1> %m, i32 %evl)
+  %v = call <vscale x 2 x half> @llvm.vp.fmuladd.nxv2f16(<vscale x 2 x half> %va, <vscale x 2 x half> %b, <vscale x 2 x half> %negc, <vscale x 2 x i1> %m, i32 %evl)
+  ret <vscale x 2 x half> %v
+}
+
+define <vscale x 2 x half> @vfmsub_vf_nxv2f16(<vscale x 2 x half> %va, half %b, <vscale x 2 x half> %vc, <vscale x 2 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: vfmsub_vf_nxv2f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vfmsub.vf v8, fa0, v9, v0.t
+; CHECK-NEXT:    ret
+  %elt.head = insertelement <vscale x 2 x half> poison, half %b, i32 0
+  %vb = shufflevector <vscale x 2 x half> %elt.head, <vscale x 2 x half> poison, <vscale x 2 x i32> zeroinitializer
+  %negvc = call <vscale x 2 x half> @llvm.vp.fneg.nxv