[RISCV] Add SDNode patterns for vwsll.[vv,vx,vi]

[luke]

Summary

This reuses the patterns introduced to help lower vnsr[a,l].vx in D155698.

[craig.topper]

LGTM

llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td

defm _VV : VPseudoBinaryRoundingMode<m.wvrclass, m.vrclass, m.vrclass, m,
                                       "@earlyclobber $rd", UsesVXRM=0>;
 }
 
 multiclass VPseudoBinaryW_VX<LMULInfo m> {
   defm "_VX" : VPseudoBinary<m.wvrclass, m.vrclass, GPR, m,
                              "@earlyclobber $rd">;
 }
 
+multiclass VPseudoBinaryW_VI<Operand ImmType, LMULInfo m> {
+  defm "_VI" : VPseudoBinary<m.wvrclass, m.vrclass, ImmType, m,
+                             "@earlyclobber $rd">;
+}
+
 multiclass VPseudoBinaryW_VF<LMULInfo m, FPR_Info f> {
   defm "_V" # f.FX : VPseudoBinary<m.wvrclass, m.vrclass,
                                    f.fprclass, m,
                                    "@earlyclobber $rd">;
 }
 
 multiclass VPseudoBinaryW_VF_RM<LMULInfo m, FPR_Info f> {
   defm "_V" # f.FX : VPseudoBinaryRoundingMode<m.wvrclass, m.vrclass,
foreach m = MxListW in {
 
     defm "" : VPseudoBinaryW_VV<m>,
             Sched<[WriteVIWALUV_MX, ReadVIWALUV_MX, ReadVIWALUV_MX, ReadVMask]>;
     defm "" : VPseudoBinaryW_VX<m>,
             Sched<[WriteVIWALUX_MX, ReadVIWALUV_MX, ReadVIWALUX_MX, ReadVMask]>;
   }
 }
 
+multiclass VPseudoVWALU_VV_VX_VI<Operand ImmType> : VPseudoVWALU_VV_VX {
+  foreach m = MxListW in {
+    defm "" : VPseudoBinaryW_VI<ImmType, m>;
+  }
+}
+
 multiclass VPseudoVWMUL_VV_VX {
   foreach m = MxListW in {
     defvar mx = m.MX;
     defvar WriteVIWMulV_MX = !cast<SchedWrite>("WriteVIWMulV_" # mx);
     defvar WriteVIWMulX_MX = !cast<SchedWrite>("WriteVIWMulX_" # mx);
     defvar ReadVIWMulV_MX = !cast<SchedRead>("ReadVIWMulV_" # mx);
     defvar ReadVIWMulX_MX = !cast<SchedRead>("ReadVIWMulX_" # mx);
 

llvm/lib/Target/RISCV/RISCVInstrInfoZvk.td

 defm PseudoVREV8 : VPseudoUnaryV_V;
 defm PseudoVCLZ : VPseudoUnaryV_V;
 defm PseudoVCTZ : VPseudoUnaryV_V;
 defm PseudoVCPOP : VPseudoUnaryV_V;
 
 defm PseudoVROL : VPseudoVALU_VV_VX;
 defm PseudoVROR : VPseudoVALU_VV_VX_VI<uimm6>;
 
+defm PseudoVWSLL : VPseudoVWALU_VV_VX_VI<uimm5>;
+
 //===----------------------------------------------------------------------===//
 // SDNode patterns
 //===----------------------------------------------------------------------===//
 
 multiclass VPatUnarySDNode_V<SDPatternOperator op, string instruction_name> {
   foreach vti = AllIntegerVectors in {
     let Predicates = !listconcat([HasStdExtZvbb],
                                  GetVTypePredicates<vti>.Predicates) in {
def : Pat<(vti.Vector (rotl vti.RegClass:$rs2,
                  (vti.Vector (IMPLICIT_DEF)),
                  vti.RegClass:$rs2,
                  (NegImm64 uimm6:$rs1),
                  vti.AVL, vti.Log2SEW, TA_MA)>;
   }
 }
 defm : VPatBinarySDNode_VV_VX_VI<rotr, "PseudoVROR", uimm6>;
 
+foreach vtiToWti = AllWidenableIntVectors in {
+  defvar vti = vtiToWti.Vti;
+  defvar wti = vtiToWti.Wti;
+  let Predicates = !listconcat([HasStdExtZvbb],
+                               GetVTypePredicates<vti>.Predicates,
+                               GetVTypePredicates<wti>.Predicates) in {
+    def : Pat<(shl (wti.Vector (zext_oneuse (vti.Vector vti.RegClass:$rs2))),
+                   (wti.Vector (ext_oneuse (vti.Vector vti.RegClass:$rs1)))),
+              (!cast<Instruction>("PseudoVWSLL_VV_"#vti.LMul.MX)
+                 (wti.Vector (IMPLICIT_DEF)),
+                 vti.RegClass:$rs2, vti.RegClass:$rs1,
+                 vti.AVL, vti.Log2SEW, TA_MA)>;
+
+    def : Pat<(shl (wti.Vector (zext_oneuse (vti.Vector vti.RegClass:$rs2))),
+                   (wti.Vector (LowBitsSplatPat (XLenVT GPR:$rs1)))),
+              (!cast<Instruction>("PseudoVWSLL_VX_"#vti.LMul.MX)
+                 (wti.Vector (IMPLICIT_DEF)),
+                 vti.RegClass:$rs2, GPR:$rs1,
+                 vti.AVL, vti.Log2SEW, TA_MA)>;
+
+    def : Pat<(shl (wti.Vector (zext_oneuse (vti.Vector vti.RegClass:$rs2))),
+                   (wti.Vector (SplatPat_uimm5 uimm5:$rs1))),
+              (!cast<Instruction>("PseudoVWSLL_VI_"#vti.LMul.MX)
+                 (wti.Vector (IMPLICIT_DEF)),
+                 vti.RegClass:$rs2, uimm5:$rs1,
+                 vti.AVL, vti.Log2SEW, TA_MA)>;
+  }
+}
+
 //===----------------------------------------------------------------------===//
 // VL patterns
 //===----------------------------------------------------------------------===//
 
 multiclass VPatUnaryVL_V<SDPatternOperator op, string instruction_name> {
   foreach vti = AllIntegerVectors in {
     let Predicates = !listconcat([HasStdExtZvbb],
                                  GetVTypePredicates<vti>.Predicates) in {

llvm/test/CodeGen/RISCV/rvv/vwsll-sdnode.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
+; RUN: llc -mtriple=riscv32 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,CHECK-RV32
+; RUN: llc -mtriple=riscv64 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK,CHECK-RV64
+; RUN: llc -mtriple=riscv32 -mattr=+v,+experimental-zvbb -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK-ZVBB,CHECK-ZVBB32
+; RUN: llc -mtriple=riscv64 -mattr=+v,+experimental-zvbb -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK-ZVBB,CHECK-ZVBB64
+
+; ==============================================================================
+; i32 -> i64
+; ==============================================================================
+
+define <vscale x 2 x i64> @vwsll_vv_nxv2i64_sext(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b) {
+; CHECK-LABEL: vwsll_vv_nxv2i64_sext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vv_nxv2i64_sext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vv v10, v8, v9
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %x = zext <vscale x 2 x i32> %a to <vscale x 2 x i64>
+  %y = sext <vscale x 2 x i32> %b to <vscale x 2 x i64>
+  %z = shl <vscale x 2 x i64> %x, %y
+  ret <vscale x 2 x i64> %z
+}
+
+define <vscale x 2 x i64> @vwsll_vv_nxv2i64_zext(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b) {
+; CHECK-LABEL: vwsll_vv_nxv2i64_zext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vzext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vv_nxv2i64_zext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vv v10, v8, v9
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %x = zext <vscale x 2 x i32> %a to <vscale x 2 x i64>
+  %y = zext <vscale x 2 x i32> %b to <vscale x 2 x i64>
+  %z = shl <vscale x 2 x i64> %x, %y
+  ret <vscale x 2 x i64> %z
+}
+
+define <vscale x 2 x i64> @vwsll_vx_i64_nxv2i64(<vscale x 2 x i32> %a, i64 %b) {
+; CHECK-LABEL: vwsll_vx_i64_nxv2i64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsll.vx v8, v10, a0
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i64_nxv2i64:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 2 x i64> poison, i64 %b, i32 0
+  %splat = shufflevector <vscale x 2 x i64> %head, <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
+  %x = zext <vscale x 2 x i32> %a to <vscale x 2 x i64>
+  %z = shl <vscale x 2 x i64> %x, %splat
+  ret <vscale x 2 x i64> %z
+}
+
+define <vscale x 2 x i64> @vwsll_vx_i32_nxv2i64_sext(<vscale x 2 x i32> %a, i32 %b) {
+; CHECK-LABEL: vwsll_vx_i32_nxv2i64_sext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i32_nxv2i64_sext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
+  %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
+  %x = zext <vscale x 2 x i32> %a to <vscale x 2 x i64>
+  %y = sext <vscale x 2 x i32> %splat to <vscale x 2 x i64>
+  %z = shl <vscale x 2 x i64> %x, %y
+  ret <vscale x 2 x i64> %z
+}
+
+define <vscale x 2 x i64> @vwsll_vx_i32_nxv2i64_zext(<vscale x 2 x i32> %a, i32 %b) {
+; CHECK-LABEL: vwsll_vx_i32_nxv2i64_zext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vzext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i32_nxv2i64_zext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 2 x i32> poison, i32 %b, i32 0
+  %splat = shufflevector <vscale x 2 x i32> %head, <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer
+  %x = zext <vscale x 2 x i32> %a to <vscale x 2 x i64>
+  %y = zext <vscale x 2 x i32> %splat to <vscale x 2 x i64>
+  %z = shl <vscale x 2 x i64> %x, %y
+  ret <vscale x 2 x i64> %z
+}
+
+define <vscale x 2 x i64> @vwsll_vx_i16_nxv2i64_sext(<vscale x 2 x i32> %a, i16 %b) {
+; CHECK-LABEL: vwsll_vx_i16_nxv2i64_sext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsext.vf4 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i16_nxv2i64_sext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 2 x i16> poison, i16 %b, i32 0
+  %splat = shufflevector <vscale x 2 x i16> %head, <vscale x 2 x i16> poison, <vscale x 2 x i32> zeroinitializer
+  %x = zext <vscale x 2 x i32> %a to <vscale x 2 x i64>
+  %y = sext <vscale x 2 x i16> %splat to <vscale x 2 x i64>
+  %z = shl <vscale x 2 x i64> %x, %y
+  ret <vscale x 2 x i64> %z
+}
+
+define <vscale x 2 x i64> @vwsll_vx_i16_nxv2i64_zext(<vscale x 2 x i32> %a, i16 %b) {
+; CHECK-LABEL: vwsll_vx_i16_nxv2i64_zext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vzext.vf4 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i16_nxv2i64_zext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 2 x i16> poison, i16 %b, i32 0
+  %splat = shufflevector <vscale x 2 x i16> %head, <vscale x 2 x i16> poison, <vscale x 2 x i32> zeroinitializer
+  %x = zext <vscale x 2 x i32> %a to <vscale x 2 x i64>
+  %y = zext <vscale x 2 x i16> %splat to <vscale x 2 x i64>
+  %z = shl <vscale x 2 x i64> %x, %y
+  ret <vscale x 2 x i64> %z
+}
+
+define <vscale x 2 x i64> @vwsll_vx_i8_nxv2i64_sext(<vscale x 2 x i32> %a, i8 %b) {
+; CHECK-LABEL: vwsll_vx_i8_nxv2i64_sext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e8, mf4, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsext.vf8 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i8_nxv2i64_sext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 2 x i8> poison, i8 %b, i32 0
+  %splat = shufflevector <vscale x 2 x i8> %head, <vscale x 2 x i8> poison, <vscale x 2 x i32> zeroinitializer
+  %x = zext <vscale x 2 x i32> %a to <vscale x 2 x i64>
+  %y = sext <vscale x 2 x i8> %splat to <vscale x 2 x i64>
+  %z = shl <vscale x 2 x i64> %x, %y
+  ret <vscale x 2 x i64> %z
+}
+
+define <vscale x 2 x i64> @vwsll_vx_i8_nxv2i64_zext(<vscale x 2 x i32> %a, i8 %b) {
+; CHECK-LABEL: vwsll_vx_i8_nxv2i64_zext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e8, mf4, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vzext.vf8 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i8_nxv2i64_zext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 2 x i8> poison, i8 %b, i32 0
+  %splat = shufflevector <vscale x 2 x i8> %head, <vscale x 2 x i8> poison, <vscale x 2 x i32> zeroinitializer
+  %x = zext <vscale x 2 x i32> %a to <vscale x 2 x i64>
+  %y = zext <vscale x 2 x i8> %splat to <vscale x 2 x i64>
+  %z = shl <vscale x 2 x i64> %x, %y
+  ret <vscale x 2 x i64> %z
+}
+
+define <vscale x 2 x i64> @vwsll_vi_nxv2i64(<vscale x 2 x i32> %a) {
+; CHECK-LABEL: vwsll_vi_nxv2i64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsll.vi v8, v10, 2
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vi_nxv2i64:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vi v10, v8, 2
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %x = zext <vscale x 2 x i32> %a to <vscale x 2 x i64>
+  %z = shl <vscale x 2 x i64> %x, shufflevector(<vscale x 2 x i64> insertelement(<vscale x 2 x i64> poison, i64 2, i32 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
+  ret <vscale x 2 x i64> %z
+}
+
+; ==============================================================================
+; i16 -> i32
+; ==============================================================================
+
+define <vscale x 4 x i32> @vwsll_vv_nxv4i32_sext(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b) {
+; CHECK-LABEL: vwsll_vv_nxv4i32_sext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vv_nxv4i32_sext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vv v10, v8, v9
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %x = zext <vscale x 4 x i16> %a to <vscale x 4 x i32>
+  %y = sext <vscale x 4 x i16> %b to <vscale x 4 x i32>
+  %z = shl <vscale x 4 x i32> %x, %y
+  ret <vscale x 4 x i32> %z
+}
+
+define <vscale x 4 x i32> @vwsll_vv_nxv4i32_zext(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b) {
+; CHECK-LABEL: vwsll_vv_nxv4i32_zext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vzext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vv_nxv4i32_zext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vv v10, v8, v9
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %x = zext <vscale x 4 x i16> %a to <vscale x 4 x i32>
+  %y = zext <vscale x 4 x i16> %b to <vscale x 4 x i32>
+  %z = shl <vscale x 4 x i32> %x, %y
+  ret <vscale x 4 x i32> %z
+}
+
+define <vscale x 4 x i32> @vwsll_vx_i64_nxv4i32(<vscale x 4 x i16> %a, i64 %b) {
+; CHECK-RV32-LABEL: vwsll_vx_i64_nxv4i32:
+; CHECK-RV32:       # %bb.0:
+; CHECK-RV32-NEXT:    addi sp, sp, -16
+; CHECK-RV32-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-RV32-NEXT:    sw a1, 12(sp)
+; CHECK-RV32-NEXT:    sw a0, 8(sp)
+; CHECK-RV32-NEXT:    addi a0, sp, 8
+; CHECK-RV32-NEXT:    vsetvli a1, zero, e32, m2, ta, ma
+; CHECK-RV32-NEXT:    vlse64.v v12, (a0), zero
+; CHECK-RV32-NEXT:    vzext.vf2 v10, v8
+; CHECK-RV32-NEXT:    vnsrl.wi v8, v12, 0
+; CHECK-RV32-NEXT:    vsll.vv v8, v10, v8
+; CHECK-RV32-NEXT:    addi sp, sp, 16
+; CHECK-RV32-NEXT:    ret
+;
+; CHECK-RV64-LABEL: vwsll_vx_i64_nxv4i32:
+; CHECK-RV64:       # %bb.0:
+; CHECK-RV64-NEXT:    vsetvli a1, zero, e64, m4, ta, ma
+; CHECK-RV64-NEXT:    vmv.v.x v12, a0
+; CHECK-RV64-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
+; CHECK-RV64-NEXT:    vzext.vf2 v10, v8
+; CHECK-RV64-NEXT:    vnsrl.wi v8, v12, 0
+; CHECK-RV64-NEXT:    vsll.vv v8, v10, v8
+; CHECK-RV64-NEXT:    ret
+;
+; CHECK-ZVBB32-LABEL: vwsll_vx_i64_nxv4i32:
+; CHECK-ZVBB32:       # %bb.0:
+; CHECK-ZVBB32-NEXT:    addi sp, sp, -16
+; CHECK-ZVBB32-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-ZVBB32-NEXT:    sw a1, 12(sp)
+; CHECK-ZVBB32-NEXT:    sw a0, 8(sp)
+; CHECK-ZVBB32-NEXT:    addi a0, sp, 8
+; CHECK-ZVBB32-NEXT:    vsetvli a1, zero, e32, m2, ta, ma
+; CHECK-ZVBB32-NEXT:    vlse64.v v12, (a0), zero
+; CHECK-ZVBB32-NEXT:    vzext.vf2 v10, v8
+; CHECK-ZVBB32-NEXT:    vnsrl.wi v8, v12, 0
+; CHECK-ZVBB32-NEXT:    vsll.vv v8, v10, v8
+; CHECK-ZVBB32-NEXT:    addi sp, sp, 16
+; CHECK-ZVBB32-NEXT:    ret
+;
+; CHECK-ZVBB64-LABEL: vwsll_vx_i64_nxv4i32:
+; CHECK-ZVBB64:       # %bb.0:
+; CHECK-ZVBB64-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-ZVBB64-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB64-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB64-NEXT:    ret
+  %head = insertelement <vscale x 4 x i64> poison, i64 %b, i32 0
+  %splat = shufflevector <vscale x 4 x i64> %head, <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
+  %x = zext <vscale x 4 x i16> %a to <vscale x 4 x i32>
+  %y = trunc <vscale x 4 x i64> %splat to <vscale x 4 x i32>
+  %z = shl <vscale x 4 x i32> %x, %y
+  ret <vscale x 4 x i32> %z
+}
+
+define <vscale x 4 x i32> @vwsll_vx_i32_nxv4i32(<vscale x 4 x i16> %a, i32 %b) {
+; CHECK-LABEL: vwsll_vx_i32_nxv4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsll.vx v8, v10, a0
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i32_nxv4i32:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
+  %splat = shufflevector <vscale x 4 x i32> %head, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+  %x = zext <vscale x 4 x i16> %a to <vscale x 4 x i32>
+  %z = shl <vscale x 4 x i32> %x, %splat
+  ret <vscale x 4 x i32> %z
+}
+
+define <vscale x 4 x i32> @vwsll_vx_i16_nxv4i32_sext(<vscale x 4 x i16> %a, i16 %b) {
+; CHECK-LABEL: vwsll_vx_i16_nxv4i32_sext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i16_nxv4i32_sext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 4 x i16> poison, i16 %b, i32 0
+  %splat = shufflevector <vscale x 4 x i16> %head, <vscale x 4 x i16> poison, <vscale x 4 x i32> zeroinitializer
+  %x = zext <vscale x 4 x i16> %a to <vscale x 4 x i32>
+  %y = sext <vscale x 4 x i16> %splat to <vscale x 4 x i32>
+  %z = shl <vscale x 4 x i32> %x, %y
+  ret <vscale x 4 x i32> %z
+}
+
+define <vscale x 4 x i32> @vwsll_vx_i16_nxv4i32_zext(<vscale x 4 x i16> %a, i16 %b) {
+; CHECK-LABEL: vwsll_vx_i16_nxv4i32_zext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vzext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i16_nxv4i32_zext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 4 x i16> poison, i16 %b, i32 0
+  %splat = shufflevector <vscale x 4 x i16> %head, <vscale x 4 x i16> poison, <vscale x 4 x i32> zeroinitializer
+  %x = zext <vscale x 4 x i16> %a to <vscale x 4 x i32>
+  %y = zext <vscale x 4 x i16> %splat to <vscale x 4 x i32>
+  %z = shl <vscale x 4 x i32> %x, %y
+  ret <vscale x 4 x i32> %z
+}
+
+define <vscale x 4 x i32> @vwsll_vx_i8_nxv4i32_sext(<vscale x 4 x i16> %a, i8 %b) {
+; CHECK-LABEL: vwsll_vx_i8_nxv4i32_sext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e8, mf2, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsext.vf4 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i8_nxv4i32_sext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 4 x i8> poison, i8 %b, i32 0
+  %splat = shufflevector <vscale x 4 x i8> %head, <vscale x 4 x i8> poison, <vscale x 4 x i32> zeroinitializer
+  %x = zext <vscale x 4 x i16> %a to <vscale x 4 x i32>
+  %y = sext <vscale x 4 x i8> %splat to <vscale x 4 x i32>
+  %z = shl <vscale x 4 x i32> %x, %y
+  ret <vscale x 4 x i32> %z
+}
+
+define <vscale x 4 x i32> @vwsll_vx_i8_nxv4i32_zext(<vscale x 4 x i16> %a, i8 %b) {
+; CHECK-LABEL: vwsll_vx_i8_nxv4i32_zext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e8, mf2, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vzext.vf4 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i8_nxv4i32_zext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 4 x i8> poison, i8 %b, i32 0
+  %splat = shufflevector <vscale x 4 x i8> %head, <vscale x 4 x i8> poison, <vscale x 4 x i32> zeroinitializer
+  %x = zext <vscale x 4 x i16> %a to <vscale x 4 x i32>
+  %y = zext <vscale x 4 x i8> %splat to <vscale x 4 x i32>
+  %z = shl <vscale x 4 x i32> %x, %y
+  ret <vscale x 4 x i32> %z
+}
+
+define <vscale x 4 x i32> @vwsll_vi_nxv4i32(<vscale x 4 x i16> %a) {
+; CHECK-LABEL: vwsll_vi_nxv4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsll.vi v8, v10, 2
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vi_nxv4i32:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vi v10, v8, 2
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %x = zext <vscale x 4 x i16> %a to <vscale x 4 x i32>
+  %z = shl <vscale x 4 x i32> %x, shufflevector(<vscale x 4 x i32> insertelement(<vscale x 4 x i32> poison, i32 2, i32 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+  ret <vscale x 4 x i32> %z
+}
+
+; ==============================================================================
+; i8 -> i16
+; ==============================================================================
+
+define <vscale x 8 x i16> @vwsll_vv_nxv8i16_sext(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b) {
+; CHECK-LABEL: vwsll_vv_nxv8i16_sext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vv_nxv8i16_sext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vv v10, v8, v9
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %x = zext <vscale x 8 x i8> %a to <vscale x 8 x i16>
+  %y = sext <vscale x 8 x i8> %b to <vscale x 8 x i16>
+  %z = shl <vscale x 8 x i16> %x, %y
+  ret <vscale x 8 x i16> %z
+}
+
+define <vscale x 8 x i16> @vwsll_vv_nxv8i16_zext(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b) {
+; CHECK-LABEL: vwsll_vv_nxv8i16_zext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vzext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vv_nxv8i16_zext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vv v10, v8, v9
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %x = zext <vscale x 8 x i8> %a to <vscale x 8 x i16>
+  %y = zext <vscale x 8 x i8> %b to <vscale x 8 x i16>
+  %z = shl <vscale x 8 x i16> %x, %y
+  ret <vscale x 8 x i16> %z
+}
+
+define <vscale x 8 x i16> @vwsll_vx_i64_nxv8i16(<vscale x 8 x i8> %a, i64 %b) {
+; CHECK-RV32-LABEL: vwsll_vx_i64_nxv8i16:
+; CHECK-RV32:       # %bb.0:
+; CHECK-RV32-NEXT:    addi sp, sp, -16
+; CHECK-RV32-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-RV32-NEXT:    sw a1, 12(sp)
+; CHECK-RV32-NEXT:    sw a0, 8(sp)
+; CHECK-RV32-NEXT:    addi a0, sp, 8
+; CHECK-RV32-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
+; CHECK-RV32-NEXT:    vlse64.v v16, (a0), zero
+; CHECK-RV32-NEXT:    vzext.vf2 v10, v8
+; CHECK-RV32-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
+; CHECK-RV32-NEXT:    vnsrl.wi v12, v16, 0
+; CHECK-RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; CHECK-RV32-NEXT:    vnsrl.wi v8, v12, 0
+; CHECK-RV32-NEXT:    vsll.vv v8, v10, v8
+; CHECK-RV32-NEXT:    addi sp, sp, 16
+; CHECK-RV32-NEXT:    ret
+;
+; CHECK-RV64-LABEL: vwsll_vx_i64_nxv8i16:
+; CHECK-RV64:       # %bb.0:
+; CHECK-RV64-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
+; CHECK-RV64-NEXT:    vmv.v.x v16, a0
+; CHECK-RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; CHECK-RV64-NEXT:    vzext.vf2 v10, v8
+; CHECK-RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
+; CHECK-RV64-NEXT:    vnsrl.wi v12, v16, 0
+; CHECK-RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; CHECK-RV64-NEXT:    vnsrl.wi v8, v12, 0
+; CHECK-RV64-NEXT:    vsll.vv v8, v10, v8
+; CHECK-RV64-NEXT:    ret
+;
+; CHECK-ZVBB32-LABEL: vwsll_vx_i64_nxv8i16:
+; CHECK-ZVBB32:       # %bb.0:
+; CHECK-ZVBB32-NEXT:    addi sp, sp, -16
+; CHECK-ZVBB32-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-ZVBB32-NEXT:    sw a1, 12(sp)
+; CHECK-ZVBB32-NEXT:    sw a0, 8(sp)
+; CHECK-ZVBB32-NEXT:    addi a0, sp, 8
+; CHECK-ZVBB32-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
+; CHECK-ZVBB32-NEXT:    vlse64.v v16, (a0), zero
+; CHECK-ZVBB32-NEXT:    vzext.vf2 v10, v8
+; CHECK-ZVBB32-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
+; CHECK-ZVBB32-NEXT:    vnsrl.wi v12, v16, 0
+; CHECK-ZVBB32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; CHECK-ZVBB32-NEXT:    vnsrl.wi v8, v12, 0
+; CHECK-ZVBB32-NEXT:    vsll.vv v8, v10, v8
+; CHECK-ZVBB32-NEXT:    addi sp, sp, 16
+; CHECK-ZVBB32-NEXT:    ret
+;
+; CHECK-ZVBB64-LABEL: vwsll_vx_i64_nxv8i16:
+; CHECK-ZVBB64:       # %bb.0:
+; CHECK-ZVBB64-NEXT:    vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-ZVBB64-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB64-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB64-NEXT:    ret
+  %head = insertelement <vscale x 4 x i64> poison, i64 %b, i32 0
+  %splat = shufflevector <vscale x 4 x i64> %head, <vscale x 4 x i64> poison, <vscale x 8 x i32> zeroinitializer
+  %x = zext <vscale x 8 x i8> %a to <vscale x 8 x i16>
+  %y = trunc <vscale x 8 x i64> %splat to <vscale x 8 x i16>
+  %z = shl <vscale x 8 x i16> %x, %y
+  ret <vscale x 8 x i16> %z
+}
+
+define <vscale x 8 x i16> @vwsll_vx_i32_nxv8i16(<vscale x 8 x i8> %a, i32 %b) {
+; CHECK-LABEL: vwsll_vx_i32_nxv8i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, ma
+; CHECK-NEXT:    vmv.v.x v12, a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vnsrl.wi v8, v12, 0
+; CHECK-NEXT:    vsll.vv v8, v10, v8
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i32_nxv8i16:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 8 x i32> poison, i32 %b, i32 0
+  %splat = shufflevector <vscale x 8 x i32> %head, <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
+  %x = zext <vscale x 8 x i8> %a to <vscale x 8 x i16>
+  %y = trunc <vscale x 8 x i32> %splat to <vscale x 8 x i16>
+  %z = shl <vscale x 8 x i16> %x, %y
+  ret <vscale x 8 x i16> %z
+}
+
+define <vscale x 8 x i16> @vwsll_vx_i16_nxv8i16(<vscale x 8 x i8> %a, i16 %b) {
+; CHECK-LABEL: vwsll_vx_i16_nxv8i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsll.vx v8, v10, a0
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i16_nxv8i16:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 8 x i16> poison, i16 %b, i32 0
+  %splat = shufflevector <vscale x 8 x i16> %head, <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
+  %x = zext <vscale x 8 x i8> %a to <vscale x 8 x i16>
+  %z = shl <vscale x 8 x i16> %x, %splat
+  ret <vscale x 8 x i16> %z
+}
+
+define <vscale x 8 x i16> @vwsll_vx_i8_nxv8i16_sext(<vscale x 8 x i8> %a, i8 %b) {
+; CHECK-LABEL: vwsll_vx_i8_nxv8i16_sext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i8_nxv8i16_sext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 8 x i8> poison, i8 %b, i32 0
+  %splat = shufflevector <vscale x 8 x i8> %head, <vscale x 8 x i8> poison, <vscale x 8 x i32> zeroinitializer
+  %x = zext <vscale x 8 x i8> %a to <vscale x 8 x i16>
+  %y = sext <vscale x 8 x i8> %splat to <vscale x 8 x i16>
+  %z = shl <vscale x 8 x i16> %x, %y
+  ret <vscale x 8 x i16> %z
+}
+
+define <vscale x 8 x i16> @vwsll_vx_i8_nxv8i16_zext(<vscale x 8 x i8> %a, i8 %b) {
+; CHECK-LABEL: vwsll_vx_i8_nxv8i16_zext:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-NEXT:    vmv.v.x v9, a0
+; CHECK-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vzext.vf2 v12, v9
+; CHECK-NEXT:    vsll.vv v8, v10, v12
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vx_i8_nxv8i16_zext:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a1, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vx v10, v8, a0
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %head = insertelement <vscale x 8 x i8> poison, i8 %b, i32 0
+  %splat = shufflevector <vscale x 8 x i8> %head, <vscale x 8 x i8> poison, <vscale x 8 x i32> zeroinitializer
+  %x = zext <vscale x 8 x i8> %a to <vscale x 8 x i16>
+  %y = zext <vscale x 8 x i8> %splat to <vscale x 8 x i16>
+  %z = shl <vscale x 8 x i16> %x, %y
+  ret <vscale x 8 x i16> %z
+}
+
+define <vscale x 8 x i16> @vwsll_vi_nxv8i16(<vscale x 8 x i8> %a) {
+; CHECK-LABEL: vwsll_vi_nxv8i16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-NEXT:    vzext.vf2 v10, v8
+; CHECK-NEXT:    vsll.vi v8, v10, 2
+; CHECK-NEXT:    ret
+;
+; CHECK-ZVBB-LABEL: vwsll_vi_nxv8i16:
+; CHECK-ZVBB:       # %bb.0:
+; CHECK-ZVBB-NEXT:    vsetvli a0, zero, e8, m1, ta, ma
+; CHECK-ZVBB-NEXT:    vwsll.vi v10, v8, 2
+; CHECK-ZVBB-NEXT:    vmv2r.v v8, v10
+; CHECK-ZVBB-NEXT:    ret
+  %x = zext <vscale x 8 x i8> %a to <vscale x 8 x i16>
+  %z = shl <vscale x 8 x i16> %x, shufflevector(<vscale x 8 x i16> insertelement(<vscale x 8 x i16> poison, i16 2, i32 0), <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer)
+  ret <vscale x 8 x i16> %z
+}