Index: llvm/lib/Target/ARM/ARMBaseInstrInfo.cpp
===================================================================
--- llvm/lib/Target/ARM/ARMBaseInstrInfo.cpp
+++ llvm/lib/Target/ARM/ARMBaseInstrInfo.cpp
@@ -5391,7 +5391,9 @@
 
   switch (MI.getOpcode()) {
   case ARM::VSETLNi32:
+  case ARM::MVE_VMOV_to_lane_32:
     // dX = VSETLNi32 dY, rZ, imm
+    // qX = MVE_VMOV_to_lane_32 qY, rZ, imm
     const MachineOperand &MOBaseReg = MI.getOperand(1);
     const MachineOperand &MOInsertedReg = MI.getOperand(2);
     if (MOInsertedReg.isUndef())
@@ -5402,7 +5404,7 @@
 
     InsertedReg.Reg = MOInsertedReg.getReg();
     InsertedReg.SubReg = MOInsertedReg.getSubReg();
-    InsertedReg.SubIdx = MOIndex.getImm() == 0 ? ARM::ssub_0 : ARM::ssub_1;
+    InsertedReg.SubIdx = ARM::ssub_0 + MOIndex.getImm();
     return true;
   }
   llvm_unreachable("Target dependent opcode missing");
Index: llvm/lib/Target/ARM/ARMInstrMVE.td
===================================================================
--- llvm/lib/Target/ARM/ARMInstrMVE.td
+++ llvm/lib/Target/ARM/ARMInstrMVE.td
@@ -1825,12 +1825,13 @@
 }
 
 def MVE_VMOV_from_lane_32  : MVE_VMOV_lane_32<            MVE_VMOV_from_lane>;
-def MVE_VMOV_to_lane_32    : MVE_VMOV_lane_32<            MVE_VMOV_to_lane>;
 def MVE_VMOV_from_lane_s16 : MVE_VMOV_lane_16<"s16", 0b0, MVE_VMOV_from_lane>;
 def MVE_VMOV_from_lane_u16 : MVE_VMOV_lane_16<"u16", 0b1, MVE_VMOV_from_lane>;
-def MVE_VMOV_to_lane_16    : MVE_VMOV_lane_16< "16", 0b0, MVE_VMOV_to_lane>;
 def MVE_VMOV_from_lane_s8  : MVE_VMOV_lane_8 < "s8", 0b0, MVE_VMOV_from_lane>;
 def MVE_VMOV_from_lane_u8  : MVE_VMOV_lane_8 < "u8", 0b1, MVE_VMOV_from_lane>;
+let isInsertSubreg = 1 in
+def MVE_VMOV_to_lane_32    : MVE_VMOV_lane_32<            MVE_VMOV_to_lane>;
+def MVE_VMOV_to_lane_16    : MVE_VMOV_lane_16< "16", 0b0, MVE_VMOV_to_lane>;
 def MVE_VMOV_to_lane_8     : MVE_VMOV_lane_8 <  "8", 0b0, MVE_VMOV_to_lane>;
 
 // This is the same as insertelt but allows the inserted value to be an i32 as
Index: llvm/test/CodeGen/Thumb2/mve-vecreduce-add.ll
===================================================================
--- llvm/test/CodeGen/Thumb2/mve-vecreduce-add.ll
+++ llvm/test/CodeGen/Thumb2/mve-vecreduce-add.ll
@@ -274,11 +274,8 @@
 ; CHECK-NEXT:    vmov r0, s0
 ; CHECK-NEXT:    vmov r2, s2
 ; CHECK-NEXT:    sxth r0, r0
-; CHECK-NEXT:    vmov.32 q1[0], r0
 ; CHECK-NEXT:    asrs r1, r0, #31
-; CHECK-NEXT:    vmov.32 q1[1], r1
 ; CHECK-NEXT:    sxth r2, r2
-; CHECK-NEXT:    vmov r0, s4
 ; CHECK-NEXT:    adds r0, r0, r2
 ; CHECK-NEXT:    adc.w r1, r1, r2, asr #31
 ; CHECK-NEXT:    bx lr
@@ -798,11 +795,8 @@
 ; CHECK-NEXT:    vmov r0, s0
 ; CHECK-NEXT:    vmov r2, s2
 ; CHECK-NEXT:    sxtb r0, r0
-; CHECK-NEXT:    vmov.32 q1[0], r0
 ; CHECK-NEXT:    asrs r1, r0, #31
-; CHECK-NEXT:    vmov.32 q1[1], r1
 ; CHECK-NEXT:    sxtb r2, r2
-; CHECK-NEXT:    vmov r0, s4
 ; CHECK-NEXT:    adds r0, r0, r2
 ; CHECK-NEXT:    adc.w r1, r1, r2, asr #31
 ; CHECK-NEXT:    bx lr
@@ -1131,15 +1125,12 @@
 ; CHECK-NEXT:    vmov r2, s0
 ; CHECK-NEXT:    vmov r3, s2
 ; CHECK-NEXT:    sxth r2, r2
-; CHECK-NEXT:    vmov.32 q1[0], r2
-; CHECK-NEXT:    asrs r2, r2, #31
-; CHECK-NEXT:    vmov.32 q1[1], r2
+; CHECK-NEXT:    asr.w r12, r2, #31
 ; CHECK-NEXT:    sxth r3, r3
-; CHECK-NEXT:    vmov r12, s4
-; CHECK-NEXT:    adds.w r12, r12, r3
-; CHECK-NEXT:    adc.w r2, r2, r3, asr #31
-; CHECK-NEXT:    adds.w r0, r0, r12
-; CHECK-NEXT:    adcs r1, r2
+; CHECK-NEXT:    adds r2, r2, r3
+; CHECK-NEXT:    adc.w r3, r12, r3, asr #31
+; CHECK-NEXT:    adds r0, r0, r2
+; CHECK-NEXT:    adcs r1, r3
 ; CHECK-NEXT:    bx lr
 entry:
   %xx = sext <2 x i16> %x to <2 x i64>
@@ -1694,15 +1685,12 @@
 ; CHECK-NEXT:    vmov r2, s0
 ; CHECK-NEXT:    vmov r3, s2
 ; CHECK-NEXT:    sxtb r2, r2
-; CHECK-NEXT:    vmov.32 q1[0], r2
-; CHECK-NEXT:    asrs r2, r2, #31
-; CHECK-NEXT:    vmov.32 q1[1], r2
+; CHECK-NEXT:    asr.w r12, r2, #31
 ; CHECK-NEXT:    sxtb r3, r3
-; CHECK-NEXT:    vmov r12, s4
-; CHECK-NEXT:    adds.w r12, r12, r3
-; CHECK-NEXT:    adc.w r2, r2, r3, asr #31
-; CHECK-NEXT:    adds.w r0, r0, r12
-; CHECK-NEXT:    adcs r1, r2
+; CHECK-NEXT:    adds r2, r2, r3
+; CHECK-NEXT:    adc.w r3, r12, r3, asr #31
+; CHECK-NEXT:    adds r0, r0, r2
+; CHECK-NEXT:    adcs r1, r3
 ; CHECK-NEXT:    bx lr
 entry:
   %xx = sext <2 x i8> %x to <2 x i64>
Index: llvm/test/CodeGen/Thumb2/mve-vecreduce-mla.ll
===================================================================
--- llvm/test/CodeGen/Thumb2/mve-vecreduce-mla.ll
+++ llvm/test/CodeGen/Thumb2/mve-vecreduce-mla.ll
@@ -970,23 +970,20 @@
 ; CHECK-NEXT:    push {r4, lr}
 ; CHECK-NEXT:    vmov r0, s4
 ; CHECK-NEXT:    vmov r1, s0
-; CHECK-NEXT:    vmov r2, s5
+; CHECK-NEXT:    vmov r3, s5
 ; CHECK-NEXT:    vmov r4, s7
-; CHECK-NEXT:    umull r12, r3, r1, r0
-; CHECK-NEXT:    mla r1, r1, r2, r3
+; CHECK-NEXT:    umull r12, r2, r1, r0
+; CHECK-NEXT:    mla r1, r1, r3, r2
 ; CHECK-NEXT:    vmov r2, s1
-; CHECK-NEXT:    vmov r3, s2
-; CHECK-NEXT:    vmov.32 q2[0], r12
+; CHECK-NEXT:    mla lr, r2, r0, r1
+; CHECK-NEXT:    vmov r0, s6
+; CHECK-NEXT:    vmov r2, s2
+; CHECK-NEXT:    umull r3, r1, r2, r0
+; CHECK-NEXT:    mla r1, r2, r4, r1
+; CHECK-NEXT:    vmov r2, s3
 ; CHECK-NEXT:    mla r1, r2, r0, r1
-; CHECK-NEXT:    vmov r2, s6
-; CHECK-NEXT:    vmov.32 q2[1], r1
-; CHECK-NEXT:    vmov r12, s8
-; CHECK-NEXT:    umull lr, r0, r3, r2
-; CHECK-NEXT:    mla r0, r3, r4, r0
-; CHECK-NEXT:    vmov r3, s3
-; CHECK-NEXT:    mla r2, r3, r2, r0
-; CHECK-NEXT:    adds.w r0, r12, lr
-; CHECK-NEXT:    adcs r1, r2
+; CHECK-NEXT:    adds.w r0, r12, r3
+; CHECK-NEXT:    adc.w r1, r1, lr
 ; CHECK-NEXT:    pop {r4, pc}
 entry:
   %m = mul <2 x i64> %x, %y
@@ -1843,23 +1840,20 @@
 ; CHECK-NEXT:    vmov r3, s0
 ; CHECK-NEXT:    vmov r4, s5
 ; CHECK-NEXT:    vmov r6, s7
-; CHECK-NEXT:    umull r12, lr, r3, r2
-; CHECK-NEXT:    mla r3, r3, r4, lr
+; CHECK-NEXT:    umull lr, r12, r3, r2
+; CHECK-NEXT:    mla r3, r3, r4, r12
 ; CHECK-NEXT:    vmov r4, s1
-; CHECK-NEXT:    vmov.32 q2[0], r12
-; CHECK-NEXT:    mla r2, r4, r2, r3
-; CHECK-NEXT:    vmov r4, s6
-; CHECK-NEXT:    vmov r3, s2
-; CHECK-NEXT:    vmov.32 q2[1], r2
-; CHECK-NEXT:    vmov r12, s8
-; CHECK-NEXT:    umull lr, r5, r3, r4
-; CHECK-NEXT:    mla r3, r3, r6, r5
+; CHECK-NEXT:    mla r12, r4, r2, r3
+; CHECK-NEXT:    vmov r3, s6
+; CHECK-NEXT:    vmov r4, s2
+; CHECK-NEXT:    umull r2, r5, r4, r3
+; CHECK-NEXT:    mla r4, r4, r6, r5
 ; CHECK-NEXT:    vmov r5, s3
-; CHECK-NEXT:    adds.w r6, r12, lr
-; CHECK-NEXT:    mla r3, r5, r4, r3
-; CHECK-NEXT:    adcs r2, r3
-; CHECK-NEXT:    adds r0, r0, r6
-; CHECK-NEXT:    adcs r1, r2
+; CHECK-NEXT:    adds.w r2, r2, lr
+; CHECK-NEXT:    mla r3, r5, r3, r4
+; CHECK-NEXT:    adc.w r3, r3, r12
+; CHECK-NEXT:    adds r0, r0, r2
+; CHECK-NEXT:    adcs r1, r3
 ; CHECK-NEXT:    pop {r4, r5, r6, pc}
 entry:
   %m = mul <2 x i64> %x, %y
Index: llvm/test/CodeGen/Thumb2/mve-vld3.ll
===================================================================
--- llvm/test/CodeGen/Thumb2/mve-vld3.ll
+++ llvm/test/CodeGen/Thumb2/mve-vld3.ll
@@ -1132,20 +1132,18 @@
 ; CHECK-NEXT:    ldrd r2, r3, [r0]
 ; CHECK-NEXT:    ldr r0, [r0, #8]
 ; CHECK-NEXT:    vmov.32 q0[0], r2
+; CHECK-NEXT:    vmov.16 q2[0], r2
 ; CHECK-NEXT:    vmov.32 q0[1], r3
 ; CHECK-NEXT:    vmov.32 q0[2], r0
-; CHECK-NEXT:    vmov r0, s0
 ; CHECK-NEXT:    vmovx.f16 s12, s0
-; CHECK-NEXT:    vmov.16 q2[0], r0
 ; CHECK-NEXT:    vmovx.f16 s4, s1
 ; CHECK-NEXT:    vins.f16 s12, s2
 ; CHECK-NEXT:    vins.f16 s8, s4
-; CHECK-NEXT:    vmov r0, s1
 ; CHECK-NEXT:    vadd.f16 q1, q2, q3
-; CHECK-NEXT:    vmovx.f16 s8, s2
-; CHECK-NEXT:    vmov.16 q0[0], r0
-; CHECK-NEXT:    vins.f16 s0, s8
-; CHECK-NEXT:    vadd.f16 q0, q1, q0
+; CHECK-NEXT:    vmovx.f16 s0, s2
+; CHECK-NEXT:    vmov.16 q2[0], r3
+; CHECK-NEXT:    vins.f16 s8, s0
+; CHECK-NEXT:    vadd.f16 q0, q1, q2
 ; CHECK-NEXT:    vmov r0, s0
 ; CHECK-NEXT:    str r0, [r1]
 ; CHECK-NEXT:    bx lr
@@ -1179,16 +1177,15 @@
 ; CHECK-NEXT:    vmov.32 q2[0], r2
 ; CHECK-NEXT:    vmovx.f16 s13, s3
 ; CHECK-NEXT:    vmov.32 q2[1], r0
-; CHECK-NEXT:    vmov r2, s1
+; CHECK-NEXT:    vmov r0, s1
 ; CHECK-NEXT:    vmovx.f16 s16, s8
 ; CHECK-NEXT:    vins.f16 s13, s9
 ; CHECK-NEXT:    vins.f16 s5, s16
-; CHECK-NEXT:    vmov r0, s8
 ; CHECK-NEXT:    vadd.f16 q1, q1, q3
-; CHECK-NEXT:    vmov.16 q3[0], r2
+; CHECK-NEXT:    vmov.16 q3[0], r0
 ; CHECK-NEXT:    vins.f16 s12, s0
 ; CHECK-NEXT:    vmovx.f16 s0, s9
-; CHECK-NEXT:    vmov.16 q3[2], r0
+; CHECK-NEXT:    vmov.16 q3[2], r2
 ; CHECK-NEXT:    vins.f16 s13, s0
 ; CHECK-NEXT:    vadd.f16 q0, q1, q3
 ; CHECK-NEXT:    vmov r2, s1
Index: llvm/test/CodeGen/Thumb2/mve-vst3.ll
===================================================================
--- llvm/test/CodeGen/Thumb2/mve-vst3.ll
+++ llvm/test/CodeGen/Thumb2/mve-vst3.ll
@@ -18,7 +18,6 @@
 ; CHECK-NEXT:    vmov.f32 s8, s7
 ; CHECK-NEXT:    vmov.f32 s10, s1
 ; CHECK-NEXT:    vmov r2, s8
-; CHECK-NEXT:    vmov r0, s10
 ; CHECK-NEXT:    vmov.f64 d4, d2
 ; CHECK-NEXT:    vmov.f32 s9, s6
 ; CHECK-NEXT:    vmov.f32 s10, s0
@@ -1359,15 +1358,14 @@
 ; CHECK-NEXT:    ldr r2, [r0, #8]
 ; CHECK-NEXT:    vmovx.f16 s12, s8
 ; CHECK-NEXT:    vins.f16 s8, s9
-; CHECK-NEXT:    vmov.32 q0[0], r2
 ; CHECK-NEXT:    vmov q1, q2
-; CHECK-NEXT:    vmov r2, s0
 ; CHECK-NEXT:    vmovx.f16 s8, s9
 ; CHECK-NEXT:    vmov.16 q1[2], r2
+; CHECK-NEXT:    vmov.32 q0[0], r2
 ; CHECK-NEXT:    vmov r0, s8
 ; CHECK-NEXT:    vins.f16 s5, s12
-; CHECK-NEXT:    vmovx.f16 s0, s0
 ; CHECK-NEXT:    vmov.16 q1[4], r0
+; CHECK-NEXT:    vmovx.f16 s0, s0
 ; CHECK-NEXT:    vins.f16 s6, s0
 ; CHECK-NEXT:    vmov r0, s4
 ; CHECK-NEXT:    vmov r2, s5
@@ -1406,26 +1404,24 @@
 ; CHECK-NEXT:    vmov.f32 s7, s9
 ; CHECK-NEXT:    vmovx.f16 s12, s4
 ; CHECK-NEXT:    vins.f16 s4, s6
-; CHECK-NEXT:    vmov r0, s0
-; CHECK-NEXT:    vmov q2, q1
 ; CHECK-NEXT:    vmovx.f16 s14, s6
-; CHECK-NEXT:    vmov.16 q2[2], r0
-; CHECK-NEXT:    vmov r0, s14
+; CHECK-NEXT:    vmov q2, q1
+; CHECK-NEXT:    vmov r2, s14
+; CHECK-NEXT:    vmov.16 q2[2], r4
 ; CHECK-NEXT:    vins.f16 s9, s12
 ; CHECK-NEXT:    vmovx.f16 s12, s0
-; CHECK-NEXT:    vmov.16 q2[4], r0
-; CHECK-NEXT:    vmov r2, s1
+; CHECK-NEXT:    vmov.16 q2[4], r2
+; CHECK-NEXT:    vmovx.f16 s0, s1
 ; CHECK-NEXT:    vins.f16 s10, s12
 ; CHECK-NEXT:    vmovx.f16 s12, s5
 ; CHECK-NEXT:    vins.f16 s5, s7
-; CHECK-NEXT:    vmovx.f16 s0, s1
 ; CHECK-NEXT:    vmovx.f16 s4, s7
 ; CHECK-NEXT:    vmov.f32 s11, s5
-; CHECK-NEXT:    vmov r0, s4
-; CHECK-NEXT:    vmov.16 q1[0], r2
+; CHECK-NEXT:    vmov r2, s4
+; CHECK-NEXT:    vmov.16 q1[0], r0
 ; CHECK-NEXT:    vins.f16 s4, s12
 ; CHECK-NEXT:    vstrw.32 q2, [r1]
-; CHECK-NEXT:    vmov.16 q1[2], r0
+; CHECK-NEXT:    vmov.16 q1[2], r2
 ; CHECK-NEXT:    vins.f16 s5, s0
 ; CHECK-NEXT:    vmov r2, s4
 ; CHECK-NEXT:    vmov r0, s5