[RISCV] Use temporary stack in expanding SPLAT_VECTOR_SPLIT_I64_VL node

There is an issue: https://github.com/llvm/llvm-project/issues/63515
The issue is because when expanding SPLAT_VECTOR_SPLIT_I64_VL node, only memoperand is used to create dependency.
However in ScheduleDAGNodes, dependency is checked with chain only, and breaks order of store/load instructions.
I think in llvm.bitreverse.nxv2i64 intrinsic SPLAT_VECTOR_SPLIT_I64_VL nodes are parallel processed,
so no chain should be add to these nodes.
Using temporary in expanding SPLAT_VECTOR_SPLIT_I64_VL node can keep vlse instruction get correct value
no matter order of store instructions is changed.

Differential Revision: https://reviews.llvm.org/D153743

diff --git a/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp b/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp
index bebafb7..e6f5621 100644 (file)
--- a/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp
@@ -98,13 +98,12 @@ void RISCVDAGToDAGISel::PreprocessISelDAG() {
           MF.getInfo<RISCVMachineFunctionInfo>();
       SDLoc DL(N);
 
-      // We use the same frame index we use for moving two i32s into 64-bit FPR.
-      // This is an analogous operation.
-      int FI = FuncInfo->getMoveF64FrameIndex(MF);
+      // Create temporary stack for each expanding node.
+      SDValue StackSlot =
+          CurDAG->CreateStackTemporary(TypeSize::Fixed(8), Align(4));
+      int FI = cast<FrameIndexSDNode>(StackSlot.getNode())->getIndex();
       MachinePointerInfo MPI = MachinePointerInfo::getFixedStack(MF, FI);
       const TargetLowering &TLI = CurDAG->getTargetLoweringInfo();
-      SDValue StackSlot =
-          CurDAG->getFrameIndex(FI, TLI.getPointerTy(CurDAG->getDataLayout()));
 
       SDValue Chain = CurDAG->getEntryNode();
       Lo = CurDAG->getStore(Chain, DL, Lo, StackSlot, MPI, Align(8));

diff --git a/llvm/test/CodeGen/RISCV/rvv/bitreverse-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/bitreverse-sdnode.ll
index 8fdf9c0..9e4bc82 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/bitreverse-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/bitreverse-sdnode.ll
@@ -967,19 +967,19 @@ declare <vscale x 16 x i32> @llvm.bitreverse.nxv16i32(<vscale x 16 x i32>)
 define <vscale x 1 x i64> @bitreverse_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32-LABEL: bitreverse_nxv1i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
+; RV32-NEXT:    sw zero, 4(sp)
 ; RV32-NEXT:    lui a0, 1044480
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 0(sp)
 ; RV32-NEXT:    lui a0, 61681
 ; RV32-NEXT:    addi a0, a0, -241
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 28(sp)
+; RV32-NEXT:    sw a0, 24(sp)
 ; RV32-NEXT:    lui a0, 209715
 ; RV32-NEXT:    addi a0, a0, 819
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 20(sp)
+; RV32-NEXT:    sw a0, 16(sp)
 ; RV32-NEXT:    lui a0, 349525
 ; RV32-NEXT:    addi a0, a0, 1365
 ; RV32-NEXT:    sw a0, 12(sp)
@@ -993,10 +993,10 @@ define <vscale x 1 x i64> @bitreverse_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32-NEXT:    li a2, 40
 ; RV32-NEXT:    vsll.vx v10, v10, a2
 ; RV32-NEXT:    vor.vv v9, v9, v10
-; RV32-NEXT:    addi a3, sp, 8
+; RV32-NEXT:    mv a3, sp
 ; RV32-NEXT:    vlse64.v v10, (a3), zero
-; RV32-NEXT:    lui a4, 4080
-; RV32-NEXT:    vand.vx v11, v8, a4
+; RV32-NEXT:    lui a3, 4080
+; RV32-NEXT:    vand.vx v11, v8, a3
 ; RV32-NEXT:    vsll.vi v11, v11, 24
 ; RV32-NEXT:    vand.vv v12, v8, v10
 ; RV32-NEXT:    vsll.vi v12, v12, 8
@@ -1007,23 +1007,26 @@ define <vscale x 1 x i64> @bitreverse_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32-NEXT:    vand.vx v12, v12, a1
 ; RV32-NEXT:    vor.vv v11, v12, v11
 ; RV32-NEXT:    vsrl.vi v12, v8, 24
-; RV32-NEXT:    vand.vx v12, v12, a4
+; RV32-NEXT:    vand.vx v12, v12, a3
 ; RV32-NEXT:    vsrl.vi v8, v8, 8
 ; RV32-NEXT:    vand.vv v8, v8, v10
 ; RV32-NEXT:    vor.vv v8, v8, v12
-; RV32-NEXT:    vlse64.v v10, (a3), zero
+; RV32-NEXT:    addi a0, sp, 24
+; RV32-NEXT:    vlse64.v v10, (a0), zero
 ; RV32-NEXT:    vor.vv v8, v8, v11
 ; RV32-NEXT:    vor.vv v8, v9, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 4
 ; RV32-NEXT:    vand.vv v9, v9, v10
 ; RV32-NEXT:    vand.vv v8, v8, v10
-; RV32-NEXT:    vlse64.v v10, (a3), zero
+; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    vlse64.v v10, (a0), zero
 ; RV32-NEXT:    vsll.vi v8, v8, 4
 ; RV32-NEXT:    vor.vv v8, v9, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 2
 ; RV32-NEXT:    vand.vv v9, v9, v10
 ; RV32-NEXT:    vand.vv v8, v8, v10
-; RV32-NEXT:    vlse64.v v10, (a3), zero
+; RV32-NEXT:    addi a0, sp, 8
+; RV32-NEXT:    vlse64.v v10, (a0), zero
 ; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vor.vv v8, v9, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 1
@@ -1031,7 +1034,7 @@ define <vscale x 1 x i64> @bitreverse_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32-NEXT:    vand.vv v8, v8, v10
 ; RV32-NEXT:    vadd.vv v8, v8, v8
 ; RV32-NEXT:    vor.vv v8, v9, v8
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: bitreverse_nxv1i64:
@@ -1101,19 +1104,19 @@ declare <vscale x 1 x i64> @llvm.bitreverse.nxv1i64(<vscale x 1 x i64>)
 define <vscale x 2 x i64> @bitreverse_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32-LABEL: bitreverse_nxv2i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
+; RV32-NEXT:    sw zero, 4(sp)
 ; RV32-NEXT:    lui a0, 1044480
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 0(sp)
 ; RV32-NEXT:    lui a0, 61681
 ; RV32-NEXT:    addi a0, a0, -241
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 28(sp)
+; RV32-NEXT:    sw a0, 24(sp)
 ; RV32-NEXT:    lui a0, 209715
 ; RV32-NEXT:    addi a0, a0, 819
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 20(sp)
+; RV32-NEXT:    sw a0, 16(sp)
 ; RV32-NEXT:    lui a0, 349525
 ; RV32-NEXT:    addi a0, a0, 1365
 ; RV32-NEXT:    sw a0, 12(sp)
@@ -1127,10 +1130,10 @@ define <vscale x 2 x i64> @bitreverse_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32-NEXT:    li a2, 40
 ; RV32-NEXT:    vsll.vx v12, v12, a2
 ; RV32-NEXT:    vor.vv v10, v10, v12
-; RV32-NEXT:    addi a3, sp, 8
+; RV32-NEXT:    mv a3, sp
 ; RV32-NEXT:    vlse64.v v12, (a3), zero
-; RV32-NEXT:    lui a4, 4080
-; RV32-NEXT:    vand.vx v14, v8, a4
+; RV32-NEXT:    lui a3, 4080
+; RV32-NEXT:    vand.vx v14, v8, a3
 ; RV32-NEXT:    vsll.vi v14, v14, 24
 ; RV32-NEXT:    vand.vv v16, v8, v12
 ; RV32-NEXT:    vsll.vi v16, v16, 8
@@ -1141,23 +1144,26 @@ define <vscale x 2 x i64> @bitreverse_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32-NEXT:    vand.vx v16, v16, a1
 ; RV32-NEXT:    vor.vv v14, v16, v14
 ; RV32-NEXT:    vsrl.vi v16, v8, 24
-; RV32-NEXT:    vand.vx v16, v16, a4
+; RV32-NEXT:    vand.vx v16, v16, a3
 ; RV32-NEXT:    vsrl.vi v8, v8, 8
 ; RV32-NEXT:    vand.vv v8, v8, v12
 ; RV32-NEXT:    vor.vv v8, v8, v16
-; RV32-NEXT:    vlse64.v v12, (a3), zero
+; RV32-NEXT:    addi a0, sp, 24
+; RV32-NEXT:    vlse64.v v12, (a0), zero
 ; RV32-NEXT:    vor.vv v8, v8, v14
 ; RV32-NEXT:    vor.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 4
 ; RV32-NEXT:    vand.vv v10, v10, v12
 ; RV32-NEXT:    vand.vv v8, v8, v12
-; RV32-NEXT:    vlse64.v v12, (a3), zero
+; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    vlse64.v v12, (a0), zero
 ; RV32-NEXT:    vsll.vi v8, v8, 4
 ; RV32-NEXT:    vor.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 2
 ; RV32-NEXT:    vand.vv v10, v10, v12
 ; RV32-NEXT:    vand.vv v8, v8, v12
-; RV32-NEXT:    vlse64.v v12, (a3), zero
+; RV32-NEXT:    addi a0, sp, 8
+; RV32-NEXT:    vlse64.v v12, (a0), zero
 ; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vor.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 1
@@ -1165,7 +1171,7 @@ define <vscale x 2 x i64> @bitreverse_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32-NEXT:    vand.vv v8, v8, v12
 ; RV32-NEXT:    vadd.vv v8, v8, v8
 ; RV32-NEXT:    vor.vv v8, v10, v8
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: bitreverse_nxv2i64:
@@ -1235,19 +1241,19 @@ declare <vscale x 2 x i64> @llvm.bitreverse.nxv2i64(<vscale x 2 x i64>)
 define <vscale x 4 x i64> @bitreverse_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32-LABEL: bitreverse_nxv4i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
+; RV32-NEXT:    sw zero, 4(sp)
 ; RV32-NEXT:    lui a0, 1044480
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 0(sp)
 ; RV32-NEXT:    lui a0, 61681
 ; RV32-NEXT:    addi a0, a0, -241
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 28(sp)
+; RV32-NEXT:    sw a0, 24(sp)
 ; RV32-NEXT:    lui a0, 209715
 ; RV32-NEXT:    addi a0, a0, 819
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 20(sp)
+; RV32-NEXT:    sw a0, 16(sp)
 ; RV32-NEXT:    lui a0, 349525
 ; RV32-NEXT:    addi a0, a0, 1365
 ; RV32-NEXT:    sw a0, 12(sp)
@@ -1261,10 +1267,10 @@ define <vscale x 4 x i64> @bitreverse_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32-NEXT:    li a2, 40
 ; RV32-NEXT:    vsll.vx v16, v16, a2
 ; RV32-NEXT:    vor.vv v12, v12, v16
-; RV32-NEXT:    addi a3, sp, 8
+; RV32-NEXT:    mv a3, sp
 ; RV32-NEXT:    vlse64.v v16, (a3), zero
-; RV32-NEXT:    lui a4, 4080
-; RV32-NEXT:    vand.vx v20, v8, a4
+; RV32-NEXT:    lui a3, 4080
+; RV32-NEXT:    vand.vx v20, v8, a3
 ; RV32-NEXT:    vsll.vi v20, v20, 24
 ; RV32-NEXT:    vand.vv v24, v8, v16
 ; RV32-NEXT:    vsll.vi v24, v24, 8
@@ -1275,23 +1281,26 @@ define <vscale x 4 x i64> @bitreverse_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32-NEXT:    vand.vx v24, v24, a1
 ; RV32-NEXT:    vor.vv v20, v24, v20
 ; RV32-NEXT:    vsrl.vi v24, v8, 24
-; RV32-NEXT:    vand.vx v24, v24, a4
+; RV32-NEXT:    vand.vx v24, v24, a3
 ; RV32-NEXT:    vsrl.vi v8, v8, 8
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    vor.vv v8, v8, v24
-; RV32-NEXT:    vlse64.v v16, (a3), zero
+; RV32-NEXT:    addi a0, sp, 24
+; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vor.vv v8, v8, v20
 ; RV32-NEXT:    vor.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 4
 ; RV32-NEXT:    vand.vv v12, v12, v16
 ; RV32-NEXT:    vand.vv v8, v8, v16
-; RV32-NEXT:    vlse64.v v16, (a3), zero
+; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vsll.vi v8, v8, 4
 ; RV32-NEXT:    vor.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 2
 ; RV32-NEXT:    vand.vv v12, v12, v16
 ; RV32-NEXT:    vand.vv v8, v8, v16
-; RV32-NEXT:    vlse64.v v16, (a3), zero
+; RV32-NEXT:    addi a0, sp, 8
+; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vor.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 1
@@ -1299,7 +1308,7 @@ define <vscale x 4 x i64> @bitreverse_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    vadd.vv v8, v8, v8
 ; RV32-NEXT:    vor.vv v8, v12, v8
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: bitreverse_nxv4i64:
@@ -1369,27 +1378,27 @@ declare <vscale x 4 x i64> @llvm.bitreverse.nxv4i64(<vscale x 4 x i64>)
 define <vscale x 8 x i64> @bitreverse_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32-LABEL: bitreverse_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    sub sp, sp, a0
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 8 * vlenb
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 8 * vlenb
+; RV32-NEXT:    sw zero, 20(sp)
 ; RV32-NEXT:    lui a0, 1044480
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 16(sp)
 ; RV32-NEXT:    lui a0, 61681
 ; RV32-NEXT:    addi a0, a0, -241
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 44(sp)
+; RV32-NEXT:    sw a0, 40(sp)
 ; RV32-NEXT:    lui a0, 209715
 ; RV32-NEXT:    addi a0, a0, 819
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 36(sp)
+; RV32-NEXT:    sw a0, 32(sp)
 ; RV32-NEXT:    lui a0, 349525
 ; RV32-NEXT:    addi a0, a0, 1365
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 28(sp)
+; RV32-NEXT:    sw a0, 24(sp)
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsll.vx v16, v8, a0
@@ -1399,20 +1408,20 @@ define <vscale x 8 x i64> @bitreverse_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32-NEXT:    li a2, 40
 ; RV32-NEXT:    vsll.vx v24, v24, a2
 ; RV32-NEXT:    vor.vv v16, v16, v24
-; RV32-NEXT:    addi a3, sp, 16
+; RV32-NEXT:    addi a3, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
-; RV32-NEXT:    addi a3, sp, 8
+; RV32-NEXT:    addi a3, sp, 16
 ; RV32-NEXT:    vlse64.v v24, (a3), zero
-; RV32-NEXT:    lui a4, 4080
-; RV32-NEXT:    vand.vx v0, v8, a4
+; RV32-NEXT:    lui a3, 4080
+; RV32-NEXT:    vand.vx v0, v8, a3
 ; RV32-NEXT:    vsll.vi v0, v0, 24
 ; RV32-NEXT:    vand.vv v16, v8, v24
 ; RV32-NEXT:    vsll.vi v16, v16, 8
 ; RV32-NEXT:    vor.vv v16, v0, v16
-; RV32-NEXT:    addi a5, sp, 16
-; RV32-NEXT:    vl8r.v v0, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    addi a4, sp, 48
+; RV32-NEXT:    vl8r.v v0, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v16, v0, v16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsrl.vx v0, v8, a2
 ; RV32-NEXT:    vand.vx v0, v0, a1
 ; RV32-NEXT:    vsrl.vx v16, v8, a0
@@ -1420,23 +1429,26 @@ define <vscale x 8 x i64> @bitreverse_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32-NEXT:    vsrl.vi v16, v8, 8
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsrl.vi v8, v8, 24
-; RV32-NEXT:    vand.vx v8, v8, a4
+; RV32-NEXT:    vand.vx v8, v8, a3
 ; RV32-NEXT:    vor.vv v8, v16, v8
-; RV32-NEXT:    vlse64.v v16, (a3), zero
+; RV32-NEXT:    addi a0, sp, 40
+; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vor.vv v8, v8, v0
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v24, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v24, v8, 4
 ; RV32-NEXT:    vand.vv v24, v24, v16
 ; RV32-NEXT:    vand.vv v8, v8, v16
-; RV32-NEXT:    vlse64.v v16, (a3), zero
+; RV32-NEXT:    addi a0, sp, 32
+; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vsll.vi v8, v8, 4
 ; RV32-NEXT:    vor.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v24, v8, 2
 ; RV32-NEXT:    vand.vv v24, v24, v16
 ; RV32-NEXT:    vand.vv v8, v8, v16
-; RV32-NEXT:    vlse64.v v16, (a3), zero
+; RV32-NEXT:    addi a0, sp, 24
+; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vor.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v24, v8, 1
@@ -1447,7 +1459,7 @@ define <vscale x 8 x i64> @bitreverse_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: bitreverse_nxv8i64:

diff --git a/llvm/test/CodeGen/RISCV/rvv/bitreverse-vp.ll b/llvm/test/CodeGen/RISCV/rvv/bitreverse-vp.ll
index 30f6002..d24abdc 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/bitreverse-vp.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/bitreverse-vp.ll
@@ -1969,55 +1969,56 @@ declare <vscale x 1 x i64> @llvm.vp.bitreverse.nxv1i64(<vscale x 1 x i64>, <vsca
 define <vscale x 1 x i64> @vp_bitreverse_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_bitreverse_nxv1i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
+; RV32-NEXT:    sw zero, 4(sp)
 ; RV32-NEXT:    lui a1, 1044480
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
-; RV32-NEXT:    li a2, 56
+; RV32-NEXT:    li a1, 56
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
-; RV32-NEXT:    vsll.vx v9, v8, a2, v0.t
-; RV32-NEXT:    lui a1, 16
-; RV32-NEXT:    addi a3, a1, -256
-; RV32-NEXT:    vand.vx v10, v8, a3, v0.t
-; RV32-NEXT:    li a4, 40
-; RV32-NEXT:    vsll.vx v10, v10, a4, v0.t
+; RV32-NEXT:    vsll.vx v9, v8, a1, v0.t
+; RV32-NEXT:    lui a2, 16
+; RV32-NEXT:    addi a2, a2, -256
+; RV32-NEXT:    vand.vx v10, v8, a2, v0.t
+; RV32-NEXT:    li a3, 40
+; RV32-NEXT:    vsll.vx v10, v10, a3, v0.t
 ; RV32-NEXT:    vor.vv v9, v9, v10, v0.t
-; RV32-NEXT:    lui a5, 4080
-; RV32-NEXT:    vand.vx v10, v8, a5, v0.t
+; RV32-NEXT:    lui a4, 4080
+; RV32-NEXT:    vand.vx v10, v8, a4, v0.t
 ; RV32-NEXT:    vsll.vi v10, v10, 24, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    mv a5, sp
 ; RV32-NEXT:    vsetvli a6, zero, e64, m1, ta, ma
-; RV32-NEXT:    vlse64.v v11, (a1), zero
+; RV32-NEXT:    vlse64.v v11, (a5), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v12, v8, v11, v0.t
 ; RV32-NEXT:    vsll.vi v12, v12, 8, v0.t
 ; RV32-NEXT:    vor.vv v10, v10, v12, v0.t
 ; RV32-NEXT:    vor.vv v9, v9, v10, v0.t
-; RV32-NEXT:    vsrl.vx v10, v8, a2, v0.t
-; RV32-NEXT:    vsrl.vx v12, v8, a4, v0.t
-; RV32-NEXT:    vand.vx v12, v12, a3, v0.t
+; RV32-NEXT:    vsrl.vx v10, v8, a1, v0.t
+; RV32-NEXT:    vsrl.vx v12, v8, a3, v0.t
+; RV32-NEXT:    vand.vx v12, v12, a2, v0.t
 ; RV32-NEXT:    vor.vv v10, v12, v10, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 24, v0.t
-; RV32-NEXT:    vand.vx v12, v12, a5, v0.t
+; RV32-NEXT:    vand.vx v12, v12, a4, v0.t
 ; RV32-NEXT:    vsrl.vi v8, v8, 8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v11, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    vor.vv v8, v9, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 4, v0.t
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -2026,6 +2027,7 @@ define <vscale x 1 x i64> @vp_bitreverse_nxv1i64(<vscale x 1 x i64> %va, <vscale
 ; RV32-NEXT:    vsll.vi v8, v8, 4, v0.t
 ; RV32-NEXT:    vor.vv v8, v9, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 2, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -2034,6 +2036,7 @@ define <vscale x 1 x i64> @vp_bitreverse_nxv1i64(<vscale x 1 x i64> %va, <vscale
 ; RV32-NEXT:    vsll.vi v8, v8, 2, v0.t
 ; RV32-NEXT:    vor.vv v8, v9, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 1, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -2041,7 +2044,7 @@ define <vscale x 1 x i64> @vp_bitreverse_nxv1i64(<vscale x 1 x i64> %va, <vscale
 ; RV32-NEXT:    vand.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    vsll.vi v8, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v9, v8, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_bitreverse_nxv1i64:
@@ -2110,19 +2113,19 @@ define <vscale x 1 x i64> @vp_bitreverse_nxv1i64(<vscale x 1 x i64> %va, <vscale
 define <vscale x 1 x i64> @vp_bitreverse_nxv1i64_unmasked(<vscale x 1 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_bitreverse_nxv1i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
+; RV32-NEXT:    sw zero, 4(sp)
 ; RV32-NEXT:    lui a1, 1044480
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
 ; RV32-NEXT:    sw a1, 12(sp)
@@ -2139,7 +2142,7 @@ define <vscale x 1 x i64> @vp_bitreverse_nxv1i64_unmasked(<vscale x 1 x i64> %va
 ; RV32-NEXT:    lui a4, 4080
 ; RV32-NEXT:    vand.vx v10, v8, a4
 ; RV32-NEXT:    vsll.vi v10, v10, 24
-; RV32-NEXT:    addi a5, sp, 8
+; RV32-NEXT:    mv a5, sp
 ; RV32-NEXT:    vsetvli a6, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v11, (a5), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -2159,30 +2162,33 @@ define <vscale x 1 x i64> @vp_bitreverse_nxv1i64_unmasked(<vscale x 1 x i64> %va
 ; RV32-NEXT:    vor.vv v8, v8, v10
 ; RV32-NEXT:    vor.vv v8, v9, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 4
-; RV32-NEXT:    vsetvli a1, zero, e64, m1, ta, ma
-; RV32-NEXT:    vlse64.v v10, (a5), zero
+; RV32-NEXT:    addi a1, sp, 24
+; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
+; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10
 ; RV32-NEXT:    vand.vv v8, v8, v10
 ; RV32-NEXT:    vsll.vi v8, v8, 4
 ; RV32-NEXT:    vor.vv v8, v9, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 2
-; RV32-NEXT:    vsetvli a1, zero, e64, m1, ta, ma
-; RV32-NEXT:    vlse64.v v10, (a5), zero
+; RV32-NEXT:    addi a1, sp, 16
+; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
+; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10
 ; RV32-NEXT:    vand.vv v8, v8, v10
 ; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vor.vv v8, v9, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 1
-; RV32-NEXT:    vsetvli a1, zero, e64, m1, ta, ma
-; RV32-NEXT:    vlse64.v v10, (a5), zero
+; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
+; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10
 ; RV32-NEXT:    vand.vv v8, v8, v10
 ; RV32-NEXT:    vadd.vv v8, v8, v8
 ; RV32-NEXT:    vor.vv v8, v9, v8
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_bitreverse_nxv1i64_unmasked:
@@ -2255,55 +2261,56 @@ declare <vscale x 2 x i64> @llvm.vp.bitreverse.nxv2i64(<vscale x 2 x i64>, <vsca
 define <vscale x 2 x i64> @vp_bitreverse_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_bitreverse_nxv2i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
+; RV32-NEXT:    sw zero, 4(sp)
 ; RV32-NEXT:    lui a1, 1044480
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
-; RV32-NEXT:    li a2, 56
+; RV32-NEXT:    li a1, 56
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
-; RV32-NEXT:    vsll.vx v10, v8, a2, v0.t
-; RV32-NEXT:    lui a1, 16
-; RV32-NEXT:    addi a3, a1, -256
-; RV32-NEXT:    vand.vx v12, v8, a3, v0.t
-; RV32-NEXT:    li a4, 40
-; RV32-NEXT:    vsll.vx v12, v12, a4, v0.t
+; RV32-NEXT:    vsll.vx v10, v8, a1, v0.t
+; RV32-NEXT:    lui a2, 16
+; RV32-NEXT:    addi a2, a2, -256
+; RV32-NEXT:    vand.vx v12, v8, a2, v0.t
+; RV32-NEXT:    li a3, 40
+; RV32-NEXT:    vsll.vx v12, v12, a3, v0.t
 ; RV32-NEXT:    vor.vv v10, v10, v12, v0.t
-; RV32-NEXT:    lui a5, 4080
-; RV32-NEXT:    vand.vx v12, v8, a5, v0.t
+; RV32-NEXT:    lui a4, 4080
+; RV32-NEXT:    vand.vx v12, v8, a4, v0.t
 ; RV32-NEXT:    vsll.vi v12, v12, 24, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    mv a5, sp
 ; RV32-NEXT:    vsetvli a6, zero, e64, m2, ta, ma
-; RV32-NEXT:    vlse64.v v14, (a1), zero
+; RV32-NEXT:    vlse64.v v14, (a5), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v16, v8, v14, v0.t
 ; RV32-NEXT:    vsll.vi v16, v16, 8, v0.t
 ; RV32-NEXT:    vor.vv v12, v12, v16, v0.t
 ; RV32-NEXT:    vor.vv v10, v10, v12, v0.t
-; RV32-NEXT:    vsrl.vx v12, v8, a2, v0.t
-; RV32-NEXT:    vsrl.vx v16, v8, a4, v0.t
-; RV32-NEXT:    vand.vx v16, v16, a3, v0.t
+; RV32-NEXT:    vsrl.vx v12, v8, a1, v0.t
+; RV32-NEXT:    vsrl.vx v16, v8, a3, v0.t
+; RV32-NEXT:    vand.vx v16, v16, a2, v0.t
 ; RV32-NEXT:    vor.vv v12, v16, v12, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 24, v0.t
-; RV32-NEXT:    vand.vx v16, v16, a5, v0.t
+; RV32-NEXT:    vand.vx v16, v16, a4, v0.t
 ; RV32-NEXT:    vsrl.vi v8, v8, 8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v14, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    vor.vv v8, v10, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 4, v0.t
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -2312,6 +2319,7 @@ define <vscale x 2 x i64> @vp_bitreverse_nxv2i64(<vscale x 2 x i64> %va, <vscale
 ; RV32-NEXT:    vsll.vi v8, v8, 4, v0.t
 ; RV32-NEXT:    vor.vv v8, v10, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 2, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -2320,6 +2328,7 @@ define <vscale x 2 x i64> @vp_bitreverse_nxv2i64(<vscale x 2 x i64> %va, <vscale
 ; RV32-NEXT:    vsll.vi v8, v8, 2, v0.t
 ; RV32-NEXT:    vor.vv v8, v10, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 1, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -2327,7 +2336,7 @@ define <vscale x 2 x i64> @vp_bitreverse_nxv2i64(<vscale x 2 x i64> %va, <vscale
 ; RV32-NEXT:    vand.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    vsll.vi v8, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v10, v8, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_bitreverse_nxv2i64:
@@ -2396,19 +2405,19 @@ define <vscale x 2 x i64> @vp_bitreverse_nxv2i64(<vscale x 2 x i64> %va, <vscale
 define <vscale x 2 x i64> @vp_bitreverse_nxv2i64_unmasked(<vscale x 2 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_bitreverse_nxv2i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
+; RV32-NEXT:    sw zero, 4(sp)
 ; RV32-NEXT:    lui a1, 1044480
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
 ; RV32-NEXT:    sw a1, 12(sp)
@@ -2425,7 +2434,7 @@ define <vscale x 2 x i64> @vp_bitreverse_nxv2i64_unmasked(<vscale x 2 x i64> %va
 ; RV32-NEXT:    lui a4, 4080
 ; RV32-NEXT:    vand.vx v12, v8, a4
 ; RV32-NEXT:    vsll.vi v12, v12, 24
-; RV32-NEXT:    addi a5, sp, 8
+; RV32-NEXT:    mv a5, sp
 ; RV32-NEXT:    vsetvli a6, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v14, (a5), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -2445,30 +2454,33 @@ define <vscale x 2 x i64> @vp_bitreverse_nxv2i64_unmasked(<vscale x 2 x i64> %va
 ; RV32-NEXT:    vor.vv v8, v8, v12
 ; RV32-NEXT:    vor.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 4
-; RV32-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
-; RV32-NEXT:    vlse64.v v12, (a5), zero
+; RV32-NEXT:    addi a1, sp, 24
+; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
+; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12
 ; RV32-NEXT:    vand.vv v8, v8, v12
 ; RV32-NEXT:    vsll.vi v8, v8, 4
 ; RV32-NEXT:    vor.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 2
-; RV32-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
-; RV32-NEXT:    vlse64.v v12, (a5), zero
+; RV32-NEXT:    addi a1, sp, 16
+; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
+; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12
 ; RV32-NEXT:    vand.vv v8, v8, v12
 ; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vor.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 1
-; RV32-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
-; RV32-NEXT:    vlse64.v v12, (a5), zero
+; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
+; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12
 ; RV32-NEXT:    vand.vv v8, v8, v12
 ; RV32-NEXT:    vadd.vv v8, v8, v8
 ; RV32-NEXT:    vor.vv v8, v10, v8
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_bitreverse_nxv2i64_unmasked:
@@ -2541,55 +2553,56 @@ declare <vscale x 4 x i64> @llvm.vp.bitreverse.nxv4i64(<vscale x 4 x i64>, <vsca
 define <vscale x 4 x i64> @vp_bitreverse_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_bitreverse_nxv4i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
+; RV32-NEXT:    sw zero, 4(sp)
 ; RV32-NEXT:    lui a1, 1044480
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
-; RV32-NEXT:    li a2, 56
+; RV32-NEXT:    li a1, 56
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
-; RV32-NEXT:    vsll.vx v12, v8, a2, v0.t
-; RV32-NEXT:    lui a1, 16
-; RV32-NEXT:    addi a3, a1, -256
-; RV32-NEXT:    vand.vx v16, v8, a3, v0.t
-; RV32-NEXT:    li a4, 40
-; RV32-NEXT:    vsll.vx v16, v16, a4, v0.t
+; RV32-NEXT:    vsll.vx v12, v8, a1, v0.t
+; RV32-NEXT:    lui a2, 16
+; RV32-NEXT:    addi a2, a2, -256
+; RV32-NEXT:    vand.vx v16, v8, a2, v0.t
+; RV32-NEXT:    li a3, 40
+; RV32-NEXT:    vsll.vx v16, v16, a3, v0.t
 ; RV32-NEXT:    vor.vv v16, v12, v16, v0.t
-; RV32-NEXT:    lui a5, 4080
-; RV32-NEXT:    vand.vx v12, v8, a5, v0.t
+; RV32-NEXT:    lui a4, 4080
+; RV32-NEXT:    vand.vx v12, v8, a4, v0.t
 ; RV32-NEXT:    vsll.vi v20, v12, 24, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    mv a5, sp
 ; RV32-NEXT:    vsetvli a6, zero, e64, m4, ta, ma
-; RV32-NEXT:    vlse64.v v12, (a1), zero
+; RV32-NEXT:    vlse64.v v12, (a5), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v24, v8, v12, v0.t
 ; RV32-NEXT:    vsll.vi v24, v24, 8, v0.t
 ; RV32-NEXT:    vor.vv v20, v20, v24, v0.t
 ; RV32-NEXT:    vor.vv v16, v16, v20, v0.t
-; RV32-NEXT:    vsrl.vx v20, v8, a2, v0.t
-; RV32-NEXT:    vsrl.vx v24, v8, a4, v0.t
-; RV32-NEXT:    vand.vx v24, v24, a3, v0.t
+; RV32-NEXT:    vsrl.vx v20, v8, a1, v0.t
+; RV32-NEXT:    vsrl.vx v24, v8, a3, v0.t
+; RV32-NEXT:    vand.vx v24, v24, a2, v0.t
 ; RV32-NEXT:    vor.vv v20, v24, v20, v0.t
 ; RV32-NEXT:    vsrl.vi v24, v8, 24, v0.t
-; RV32-NEXT:    vand.vx v24, v24, a5, v0.t
+; RV32-NEXT:    vand.vx v24, v24, a4, v0.t
 ; RV32-NEXT:    vsrl.vi v8, v8, 8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v24, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v20, v0.t
 ; RV32-NEXT:    vor.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 4, v0.t
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -2598,6 +2611,7 @@ define <vscale x 4 x i64> @vp_bitreverse_nxv4i64(<vscale x 4 x i64> %va, <vscale
 ; RV32-NEXT:    vsll.vi v8, v8, 4, v0.t
 ; RV32-NEXT:    vor.vv v8, v12, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 2, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -2606,6 +2620,7 @@ define <vscale x 4 x i64> @vp_bitreverse_nxv4i64(<vscale x 4 x i64> %va, <vscale
 ; RV32-NEXT:    vsll.vi v8, v8, 2, v0.t
 ; RV32-NEXT:    vor.vv v8, v12, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 1, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -2613,7 +2628,7 @@ define <vscale x 4 x i64> @vp_bitreverse_nxv4i64(<vscale x 4 x i64> %va, <vscale
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsll.vi v8, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v12, v8, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_bitreverse_nxv4i64:
@@ -2682,19 +2697,19 @@ define <vscale x 4 x i64> @vp_bitreverse_nxv4i64(<vscale x 4 x i64> %va, <vscale
 define <vscale x 4 x i64> @vp_bitreverse_nxv4i64_unmasked(<vscale x 4 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_bitreverse_nxv4i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
+; RV32-NEXT:    sw zero, 4(sp)
 ; RV32-NEXT:    lui a1, 1044480
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
 ; RV32-NEXT:    sw a1, 12(sp)
@@ -2711,7 +2726,7 @@ define <vscale x 4 x i64> @vp_bitreverse_nxv4i64_unmasked(<vscale x 4 x i64> %va
 ; RV32-NEXT:    lui a4, 4080
 ; RV32-NEXT:    vand.vx v16, v8, a4
 ; RV32-NEXT:    vsll.vi v16, v16, 24
-; RV32-NEXT:    addi a5, sp, 8
+; RV32-NEXT:    mv a5, sp
 ; RV32-NEXT:    vsetvli a6, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v20, (a5), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -2731,30 +2746,33 @@ define <vscale x 4 x i64> @vp_bitreverse_nxv4i64_unmasked(<vscale x 4 x i64> %va
 ; RV32-NEXT:    vor.vv v8, v8, v16
 ; RV32-NEXT:    vor.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 4
-; RV32-NEXT:    vsetvli a1, zero, e64, m4, ta, ma
-; RV32-NEXT:    vlse64.v v16, (a5), zero
+; RV32-NEXT:    addi a1, sp, 24
+; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    vsll.vi v8, v8, 4
 ; RV32-NEXT:    vor.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 2
-; RV32-NEXT:    vsetvli a1, zero, e64, m4, ta, ma
-; RV32-NEXT:    vlse64.v v16, (a5), zero
+; RV32-NEXT:    addi a1, sp, 16
+; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vor.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 1
-; RV32-NEXT:    vsetvli a1, zero, e64, m4, ta, ma
-; RV32-NEXT:    vlse64.v v16, (a5), zero
+; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    vadd.vv v8, v8, v8
 ; RV32-NEXT:    vor.vv v8, v12, v8
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_bitreverse_nxv4i64_unmasked:
@@ -2827,98 +2845,99 @@ declare <vscale x 7 x i64> @llvm.vp.bitreverse.nxv7i64(<vscale x 7 x i64>, <vsca
 define <vscale x 7 x i64> @vp_bitreverse_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_bitreverse_nxv7i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    li a2, 24
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x18, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 24 * vlenb
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x18, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 24 * vlenb
+; RV32-NEXT:    sw zero, 20(sp)
 ; RV32-NEXT:    lui a1, 1044480
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 44(sp)
+; RV32-NEXT:    sw a1, 40(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 36(sp)
+; RV32-NEXT:    sw a1, 32(sp)
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
-; RV32-NEXT:    li a2, 56
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
+; RV32-NEXT:    li a1, 56
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
-; RV32-NEXT:    vsll.vx v16, v8, a2, v0.t
-; RV32-NEXT:    lui a1, 16
-; RV32-NEXT:    addi a3, a1, -256
-; RV32-NEXT:    vand.vx v24, v8, a3, v0.t
-; RV32-NEXT:    li a4, 40
-; RV32-NEXT:    vsll.vx v24, v24, a4, v0.t
+; RV32-NEXT:    vsll.vx v16, v8, a1, v0.t
+; RV32-NEXT:    lui a2, 16
+; RV32-NEXT:    addi a2, a2, -256
+; RV32-NEXT:    vand.vx v24, v8, a2, v0.t
+; RV32-NEXT:    li a3, 40
+; RV32-NEXT:    vsll.vx v24, v24, a3, v0.t
 ; RV32-NEXT:    vor.vv v16, v16, v24, v0.t
-; RV32-NEXT:    addi a1, sp, 16
-; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    lui a5, 4080
-; RV32-NEXT:    vand.vx v16, v8, a5, v0.t
+; RV32-NEXT:    addi a4, sp, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    lui a4, 4080
+; RV32-NEXT:    vand.vx v16, v8, a4, v0.t
 ; RV32-NEXT:    vsll.vi v16, v16, 24, v0.t
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 4
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    slli a5, a5, 4
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    addi a5, a5, 48
+; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a5, sp, 16
 ; RV32-NEXT:    vsetvli a6, zero, e64, m8, ta, ma
-; RV32-NEXT:    vlse64.v v16, (a1), zero
-; RV32-NEXT:    csrr a6, vlenb
-; RV32-NEXT:    slli a6, a6, 3
-; RV32-NEXT:    add a6, sp, a6
-; RV32-NEXT:    addi a6, a6, 16
-; RV32-NEXT:    vs8r.v v16, (a6) # Unknown-size Folded Spill
+; RV32-NEXT:    vlse64.v v16, (a5), zero
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    slli a5, a5, 3
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    addi a5, a5, 48
+; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v8, v16, v0.t
 ; RV32-NEXT:    vsll.vi v16, v16, 8, v0.t
-; RV32-NEXT:    csrr a6, vlenb
-; RV32-NEXT:    slli a6, a6, 4
-; RV32-NEXT:    add a6, sp, a6
-; RV32-NEXT:    addi a6, a6, 16
-; RV32-NEXT:    vl8r.v v24, (a6) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    slli a5, a5, 4
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    addi a5, a5, 48
+; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v16, v24, v16, v0.t
-; RV32-NEXT:    addi a6, sp, 16
-; RV32-NEXT:    vl8r.v v24, (a6) # Unknown-size Folded Reload
+; RV32-NEXT:    addi a5, sp, 48
+; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v16, v24, v16, v0.t
-; RV32-NEXT:    csrr a6, vlenb
-; RV32-NEXT:    slli a6, a6, 4
-; RV32-NEXT:    add a6, sp, a6
-; RV32-NEXT:    addi a6, a6, 16
-; RV32-NEXT:    vs8r.v v16, (a6) # Unknown-size Folded Spill
-; RV32-NEXT:    vsrl.vx v24, v8, a2, v0.t
-; RV32-NEXT:    vsrl.vx v16, v8, a4, v0.t
-; RV32-NEXT:    vand.vx v16, v16, a3, v0.t
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    slli a5, a5, 4
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    addi a5, a5, 48
+; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    vsrl.vx v24, v8, a1, v0.t
+; RV32-NEXT:    vsrl.vx v16, v8, a3, v0.t
+; RV32-NEXT:    vand.vx v16, v16, a2, v0.t
 ; RV32-NEXT:    vor.vv v16, v16, v24, v0.t
-; RV32-NEXT:    addi a2, sp, 16
-; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a1, sp, 48
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsrl.vi v24, v8, 24, v0.t
-; RV32-NEXT:    vand.vx v24, v24, a5, v0.t
+; RV32-NEXT:    vand.vx v24, v24, a4, v0.t
 ; RV32-NEXT:    vsrl.vi v8, v8, 8, v0.t
-; RV32-NEXT:    csrr a2, vlenb
-; RV32-NEXT:    slli a2, a2, 3
-; RV32-NEXT:    add a2, sp, a2
-; RV32-NEXT:    addi a2, a2, 16
-; RV32-NEXT:    vl8r.v v16, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 48
+; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v24, v0.t
-; RV32-NEXT:    addi a2, sp, 16
-; RV32-NEXT:    vl8r.v v16, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    addi a1, sp, 48
+; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
-; RV32-NEXT:    csrr a2, vlenb
-; RV32-NEXT:    slli a2, a2, 4
-; RV32-NEXT:    add a2, sp, a2
-; RV32-NEXT:    addi a2, a2, 16
-; RV32-NEXT:    vl8r.v v16, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 4
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 48
+; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
+; RV32-NEXT:    addi a1, sp, 40
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2927,6 +2946,7 @@ define <vscale x 7 x i64> @vp_bitreverse_nxv7i64(<vscale x 7 x i64> %va, <vscale
 ; RV32-NEXT:    vsll.vi v8, v8, 4, v0.t
 ; RV32-NEXT:    vor.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 2, v0.t
+; RV32-NEXT:    addi a1, sp, 32
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2935,6 +2955,7 @@ define <vscale x 7 x i64> @vp_bitreverse_nxv7i64(<vscale x 7 x i64> %va, <vscale
 ; RV32-NEXT:    vsll.vi v8, v8, 2, v0.t
 ; RV32-NEXT:    vor.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2946,7 +2967,7 @@ define <vscale x 7 x i64> @vp_bitreverse_nxv7i64(<vscale x 7 x i64> %va, <vscale
 ; RV32-NEXT:    li a1, 24
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_bitreverse_nxv7i64:
@@ -3032,27 +3053,27 @@ define <vscale x 7 x i64> @vp_bitreverse_nxv7i64(<vscale x 7 x i64> %va, <vscale
 define <vscale x 7 x i64> @vp_bitreverse_nxv7i64_unmasked(<vscale x 7 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_bitreverse_nxv7i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    slli a1, a1, 3
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 8 * vlenb
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 8 * vlenb
+; RV32-NEXT:    sw zero, 20(sp)
 ; RV32-NEXT:    lui a1, 1044480
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 44(sp)
+; RV32-NEXT:    sw a1, 40(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 36(sp)
+; RV32-NEXT:    sw a1, 32(sp)
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    li a1, 56
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsll.vx v16, v8, a1
@@ -3062,22 +3083,22 @@ define <vscale x 7 x i64> @vp_bitreverse_nxv7i64_unmasked(<vscale x 7 x i64> %va
 ; RV32-NEXT:    li a3, 40
 ; RV32-NEXT:    vsll.vx v24, v24, a3
 ; RV32-NEXT:    vor.vv v16, v16, v24
-; RV32-NEXT:    addi a4, sp, 16
+; RV32-NEXT:    addi a4, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    lui a4, 4080
 ; RV32-NEXT:    vand.vx v16, v8, a4
 ; RV32-NEXT:    vsll.vi v0, v16, 24
-; RV32-NEXT:    addi a5, sp, 8
+; RV32-NEXT:    addi a5, sp, 16
 ; RV32-NEXT:    vsetvli a6, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a5), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v24, v8, v16
 ; RV32-NEXT:    vsll.vi v24, v24, 8
 ; RV32-NEXT:    vor.vv v24, v0, v24
-; RV32-NEXT:    addi a6, sp, 16
-; RV32-NEXT:    vl8r.v v0, (a6) # Unknown-size Folded Reload
+; RV32-NEXT:    addi a5, sp, 48
+; RV32-NEXT:    vl8r.v v0, (a5) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v24, v0, v24
-; RV32-NEXT:    vs8r.v v24, (a6) # Unknown-size Folded Spill
+; RV32-NEXT:    vs8r.v v24, (a5) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsrl.vx v0, v8, a3
 ; RV32-NEXT:    vand.vx v0, v0, a2
 ; RV32-NEXT:    vsrl.vx v24, v8, a1
@@ -3088,28 +3109,31 @@ define <vscale x 7 x i64> @vp_bitreverse_nxv7i64_unmasked(<vscale x 7 x i64> %va
 ; RV32-NEXT:    vand.vx v8, v8, a4
 ; RV32-NEXT:    vor.vv v8, v16, v8
 ; RV32-NEXT:    vor.vv v8, v8, v24
-; RV32-NEXT:    addi a1, sp, 16
+; RV32-NEXT:    addi a1, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
-; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
-; RV32-NEXT:    vlse64.v v24, (a5), zero
+; RV32-NEXT:    addi a1, sp, 40
+; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vand.vv v8, v8, v24
 ; RV32-NEXT:    vsll.vi v8, v8, 4
 ; RV32-NEXT:    vor.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 2
-; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
-; RV32-NEXT:    vlse64.v v24, (a5), zero
+; RV32-NEXT:    addi a1, sp, 32
+; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vand.vv v8, v8, v24
 ; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vor.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
-; RV32-NEXT:    vlse64.v v24, (a5), zero
+; RV32-NEXT:    addi a1, sp, 24
+; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vand.vv v8, v8, v24
@@ -3118,7 +3142,7 @@ define <vscale x 7 x i64> @vp_bitreverse_nxv7i64_unmasked(<vscale x 7 x i64> %va
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_bitreverse_nxv7i64_unmasked:
@@ -3191,98 +3215,99 @@ declare <vscale x 8 x i64> @llvm.vp.bitreverse.nxv8i64(<vscale x 8 x i64>, <vsca
 define <vscale x 8 x i64> @vp_bitreverse_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_bitreverse_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    li a2, 24
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x18, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 24 * vlenb
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x18, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 24 * vlenb
+; RV32-NEXT:    sw zero, 20(sp)
 ; RV32-NEXT:    lui a1, 1044480
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 44(sp)
+; RV32-NEXT:    sw a1, 40(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 36(sp)
+; RV32-NEXT:    sw a1, 32(sp)
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
-; RV32-NEXT:    li a2, 56
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
+; RV32-NEXT:    li a1, 56
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
-; RV32-NEXT:    vsll.vx v16, v8, a2, v0.t
-; RV32-NEXT:    lui a1, 16
-; RV32-NEXT:    addi a3, a1, -256
-; RV32-NEXT:    vand.vx v24, v8, a3, v0.t
-; RV32-NEXT:    li a4, 40
-; RV32-NEXT:    vsll.vx v24, v24, a4, v0.t
+; RV32-NEXT:    vsll.vx v16, v8, a1, v0.t
+; RV32-NEXT:    lui a2, 16
+; RV32-NEXT:    addi a2, a2, -256
+; RV32-NEXT:    vand.vx v24, v8, a2, v0.t
+; RV32-NEXT:    li a3, 40
+; RV32-NEXT:    vsll.vx v24, v24, a3, v0.t
 ; RV32-NEXT:    vor.vv v16, v16, v24, v0.t
-; RV32-NEXT:    addi a1, sp, 16
-; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    lui a5, 4080
-; RV32-NEXT:    vand.vx v16, v8, a5, v0.t
+; RV32-NEXT:    addi a4, sp, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    lui a4, 4080
+; RV32-NEXT:    vand.vx v16, v8, a4, v0.t
 ; RV32-NEXT:    vsll.vi v16, v16, 24, v0.t
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 4
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    slli a5, a5, 4
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    addi a5, a5, 48
+; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a5, sp, 16
 ; RV32-NEXT:    vsetvli a6, zero, e64, m8, ta, ma
-; RV32-NEXT:    vlse64.v v16, (a1), zero
-; RV32-NEXT:    csrr a6, vlenb
-; RV32-NEXT:    slli a6, a6, 3
-; RV32-NEXT:    add a6, sp, a6
-; RV32-NEXT:    addi a6, a6, 16
-; RV32-NEXT:    vs8r.v v16, (a6) # Unknown-size Folded Spill
+; RV32-NEXT:    vlse64.v v16, (a5), zero
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    slli a5, a5, 3
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    addi a5, a5, 48
+; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v8, v16, v0.t
 ; RV32-NEXT:    vsll.vi v16, v16, 8, v0.t
-; RV32-NEXT:    csrr a6, vlenb
-; RV32-NEXT:    slli a6, a6, 4
-; RV32-NEXT:    add a6, sp, a6
-; RV32-NEXT:    addi a6, a6, 16
-; RV32-NEXT:    vl8r.v v24, (a6) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    slli a5, a5, 4
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    addi a5, a5, 48
+; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v16, v24, v16, v0.t
-; RV32-NEXT:    addi a6, sp, 16
-; RV32-NEXT:    vl8r.v v24, (a6) # Unknown-size Folded Reload
+; RV32-NEXT:    addi a5, sp, 48
+; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v16, v24, v16, v0.t
-; RV32-NEXT:    csrr a6, vlenb
-; RV32-NEXT:    slli a6, a6, 4
-; RV32-NEXT:    add a6, sp, a6
-; RV32-NEXT:    addi a6, a6, 16
-; RV32-NEXT:    vs8r.v v16, (a6) # Unknown-size Folded Spill
-; RV32-NEXT:    vsrl.vx v24, v8, a2, v0.t
-; RV32-NEXT:    vsrl.vx v16, v8, a4, v0.t
-; RV32-NEXT:    vand.vx v16, v16, a3, v0.t
+; RV32-NEXT:    csrr a5, vlenb
+; RV32-NEXT:    slli a5, a5, 4
+; RV32-NEXT:    add a5, sp, a5
+; RV32-NEXT:    addi a5, a5, 48
+; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    vsrl.vx v24, v8, a1, v0.t
+; RV32-NEXT:    vsrl.vx v16, v8, a3, v0.t
+; RV32-NEXT:    vand.vx v16, v16, a2, v0.t
 ; RV32-NEXT:    vor.vv v16, v16, v24, v0.t
-; RV32-NEXT:    addi a2, sp, 16
-; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a1, sp, 48
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsrl.vi v24, v8, 24, v0.t
-; RV32-NEXT:    vand.vx v24, v24, a5, v0.t
+; RV32-NEXT:    vand.vx v24, v24, a4, v0.t
 ; RV32-NEXT:    vsrl.vi v8, v8, 8, v0.t
-; RV32-NEXT:    csrr a2, vlenb
-; RV32-NEXT:    slli a2, a2, 3
-; RV32-NEXT:    add a2, sp, a2
-; RV32-NEXT:    addi a2, a2, 16
-; RV32-NEXT:    vl8r.v v16, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 48
+; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v24, v0.t
-; RV32-NEXT:    addi a2, sp, 16
-; RV32-NEXT:    vl8r.v v16, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    addi a1, sp, 48
+; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
-; RV32-NEXT:    csrr a2, vlenb
-; RV32-NEXT:    slli a2, a2, 4
-; RV32-NEXT:    add a2, sp, a2
-; RV32-NEXT:    addi a2, a2, 16
-; RV32-NEXT:    vl8r.v v16, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 4
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 48
+; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
+; RV32-NEXT:    addi a1, sp, 40
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -3291,6 +3316,7 @@ define <vscale x 8 x i64> @vp_bitreverse_nxv8i64(<vscale x 8 x i64> %va, <vscale
 ; RV32-NEXT:    vsll.vi v8, v8, 4, v0.t
 ; RV32-NEXT:    vor.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 2, v0.t
+; RV32-NEXT:    addi a1, sp, 32
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -3299,6 +3325,7 @@ define <vscale x 8 x i64> @vp_bitreverse_nxv8i64(<vscale x 8 x i64> %va, <vscale
 ; RV32-NEXT:    vsll.vi v8, v8, 2, v0.t
 ; RV32-NEXT:    vor.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -3310,7 +3337,7 @@ define <vscale x 8 x i64> @vp_bitreverse_nxv8i64(<vscale x 8 x i64> %va, <vscale
 ; RV32-NEXT:    li a1, 24
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_bitreverse_nxv8i64:
@@ -3396,27 +3423,27 @@ define <vscale x 8 x i64> @vp_bitreverse_nxv8i64(<vscale x 8 x i64> %va, <vscale
 define <vscale x 8 x i64> @vp_bitreverse_nxv8i64_unmasked(<vscale x 8 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_bitreverse_nxv8i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    slli a1, a1, 3
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 8 * vlenb
-; RV32-NEXT:    sw zero, 12(sp)
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 8 * vlenb
+; RV32-NEXT:    sw zero, 20(sp)
 ; RV32-NEXT:    lui a1, 1044480
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 44(sp)
+; RV32-NEXT:    sw a1, 40(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 36(sp)
+; RV32-NEXT:    sw a1, 32(sp)
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    li a1, 56
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsll.vx v16, v8, a1
@@ -3426,22 +3453,22 @@ define <vscale x 8 x i64> @vp_bitreverse_nxv8i64_unmasked(<vscale x 8 x i64> %va
 ; RV32-NEXT:    li a3, 40
 ; RV32-NEXT:    vsll.vx v24, v24, a3
 ; RV32-NEXT:    vor.vv v16, v16, v24
-; RV32-NEXT:    addi a4, sp, 16
+; RV32-NEXT:    addi a4, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    lui a4, 4080
 ; RV32-NEXT:    vand.vx v16, v8, a4
 ; RV32-NEXT:    vsll.vi v0, v16, 24
-; RV32-NEXT:    addi a5, sp, 8
+; RV32-NEXT:    addi a5, sp, 16
 ; RV32-NEXT:    vsetvli a6, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a5), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v24, v8, v16
 ; RV32-NEXT:    vsll.vi v24, v24, 8
 ; RV32-NEXT:    vor.vv v24, v0, v24
-; RV32-NEXT:    addi a6, sp, 16
-; RV32-NEXT:    vl8r.v v0, (a6) # Unknown-size Folded Reload
+; RV32-NEXT:    addi a5, sp, 48
+; RV32-NEXT:    vl8r.v v0, (a5) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v24, v0, v24
-; RV32-NEXT:    vs8r.v v24, (a6) # Unknown-size Folded Spill
+; RV32-NEXT:    vs8r.v v24, (a5) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsrl.vx v0, v8, a3
 ; RV32-NEXT:    vand.vx v0, v0, a2
 ; RV32-NEXT:    vsrl.vx v24, v8, a1
@@ -3452,28 +3479,31 @@ define <vscale x 8 x i64> @vp_bitreverse_nxv8i64_unmasked(<vscale x 8 x i64> %va
 ; RV32-NEXT:    vand.vx v8, v8, a4
 ; RV32-NEXT:    vor.vv v8, v16, v8
 ; RV32-NEXT:    vor.vv v8, v8, v24
-; RV32-NEXT:    addi a1, sp, 16
+; RV32-NEXT:    addi a1, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vor.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
-; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
-; RV32-NEXT:    vlse64.v v24, (a5), zero
+; RV32-NEXT:    addi a1, sp, 40
+; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vand.vv v8, v8, v24
 ; RV32-NEXT:    vsll.vi v8, v8, 4
 ; RV32-NEXT:    vor.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 2
-; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
-; RV32-NEXT:    vlse64.v v24, (a5), zero
+; RV32-NEXT:    addi a1, sp, 32
+; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vand.vv v8, v8, v24
 ; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vor.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
-; RV32-NEXT:    vlse64.v v24, (a5), zero
+; RV32-NEXT:    addi a1, sp, 24
+; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vand.vv v8, v8, v24
@@ -3482,7 +3512,7 @@ define <vscale x 8 x i64> @vp_bitreverse_nxv8i64_unmasked(<vscale x 8 x i64> %va
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_bitreverse_nxv8i64_unmasked:

diff --git a/llvm/test/CodeGen/RISCV/rvv/ctlz-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/ctlz-sdnode.ll
index 4171934..04fcebc 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/ctlz-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/ctlz-sdnode.ll
@@ -1417,24 +1417,24 @@ declare <vscale x 16 x i32> @llvm.ctlz.nxv16i32(<vscale x 16 x i32>, i1)
 define <vscale x 1 x i64> @ctlz_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-LABEL: ctlz_nxv1i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
 ; RV32I-NEXT:    vsrl.vi v9, v8, 1
 ; RV32I-NEXT:    vor.vv v8, v8, v9
@@ -1449,9 +1449,10 @@ define <vscale x 1 x i64> @ctlz_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-NEXT:    li a0, 32
 ; RV32I-NEXT:    vsrl.vx v9, v8, a0
 ; RV32I-NEXT:    vor.vv v8, v8, v9
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v9, (a0), zero
 ; RV32I-NEXT:    vnot.v v8, v8
+; RV32I-NEXT:    addi a0, sp, 24
+; RV32I-NEXT:    vlse64.v v9, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v11, v8, 1
 ; RV32I-NEXT:    vand.vv v9, v11, v9
@@ -1460,7 +1461,9 @@ define <vscale x 1 x i64> @ctlz_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
 ; RV32I-NEXT:    vand.vv v8, v8, v10
 ; RV32I-NEXT:    vadd.vv v8, v9, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v9, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v11, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v11
@@ -1468,7 +1471,7 @@ define <vscale x 1 x i64> @ctlz_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v10
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: ctlz_nxv1i64:
@@ -1557,24 +1560,24 @@ declare <vscale x 1 x i64> @llvm.ctlz.nxv1i64(<vscale x 1 x i64>, i1)
 define <vscale x 2 x i64> @ctlz_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-LABEL: ctlz_nxv2i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
 ; RV32I-NEXT:    vsrl.vi v10, v8, 1
 ; RV32I-NEXT:    vor.vv v8, v8, v10
@@ -1589,9 +1592,10 @@ define <vscale x 2 x i64> @ctlz_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-NEXT:    li a0, 32
 ; RV32I-NEXT:    vsrl.vx v10, v8, a0
 ; RV32I-NEXT:    vor.vv v8, v8, v10
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vnot.v v8, v8
+; RV32I-NEXT:    addi a0, sp, 24
+; RV32I-NEXT:    vlse64.v v10, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v14, v8, 1
 ; RV32I-NEXT:    vand.vv v10, v14, v10
@@ -1600,7 +1604,9 @@ define <vscale x 2 x i64> @ctlz_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
 ; RV32I-NEXT:    vand.vv v8, v8, v12
 ; RV32I-NEXT:    vadd.vv v8, v10, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v14, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v14
@@ -1608,7 +1614,7 @@ define <vscale x 2 x i64> @ctlz_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v12
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: ctlz_nxv2i64:
@@ -1697,24 +1703,24 @@ declare <vscale x 2 x i64> @llvm.ctlz.nxv2i64(<vscale x 2 x i64>, i1)
 define <vscale x 4 x i64> @ctlz_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-LABEL: ctlz_nxv4i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    vsetvli a0, zero, e64, m4, ta, ma
 ; RV32I-NEXT:    vsrl.vi v12, v8, 1
 ; RV32I-NEXT:    vor.vv v8, v8, v12
@@ -1729,9 +1735,10 @@ define <vscale x 4 x i64> @ctlz_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-NEXT:    li a0, 32
 ; RV32I-NEXT:    vsrl.vx v12, v8, a0
 ; RV32I-NEXT:    vor.vv v8, v8, v12
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vnot.v v8, v8
+; RV32I-NEXT:    addi a0, sp, 24
+; RV32I-NEXT:    vlse64.v v12, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v20, v8, 1
 ; RV32I-NEXT:    vand.vv v12, v20, v12
@@ -1740,7 +1747,9 @@ define <vscale x 4 x i64> @ctlz_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
 ; RV32I-NEXT:    vand.vv v8, v8, v16
 ; RV32I-NEXT:    vadd.vv v8, v12, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v20, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v20
@@ -1748,7 +1757,7 @@ define <vscale x 4 x i64> @ctlz_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v16
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: ctlz_nxv4i64:
@@ -1837,24 +1846,24 @@ declare <vscale x 4 x i64> @llvm.ctlz.nxv4i64(<vscale x 4 x i64>, i1)
 define <vscale x 8 x i64> @ctlz_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-LABEL: ctlz_nxv8i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; RV32I-NEXT:    vsrl.vi v16, v8, 1
 ; RV32I-NEXT:    vor.vv v8, v8, v16
@@ -1869,9 +1878,10 @@ define <vscale x 8 x i64> @ctlz_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-NEXT:    li a0, 32
 ; RV32I-NEXT:    vsrl.vx v16, v8, a0
 ; RV32I-NEXT:    vor.vv v8, v8, v16
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vnot.v v8, v8
+; RV32I-NEXT:    addi a0, sp, 24
+; RV32I-NEXT:    vlse64.v v16, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
 ; RV32I-NEXT:    vlse64.v v24, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v0, v8, 1
 ; RV32I-NEXT:    vand.vv v16, v0, v16
@@ -1880,7 +1890,9 @@ define <vscale x 8 x i64> @ctlz_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
 ; RV32I-NEXT:    vand.vv v8, v8, v24
 ; RV32I-NEXT:    vadd.vv v8, v16, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v24, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v0, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v0
@@ -1888,7 +1900,7 @@ define <vscale x 8 x i64> @ctlz_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v24
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: ctlz_nxv8i64:
@@ -3307,24 +3319,24 @@ define <vscale x 16 x i32> @ctlz_zero_undef_nxv16i32(<vscale x 16 x i32> %va) {
 define <vscale x 1 x i64> @ctlz_zero_undef_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-LABEL: ctlz_zero_undef_nxv1i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
 ; RV32I-NEXT:    vsrl.vi v9, v8, 1
 ; RV32I-NEXT:    vor.vv v8, v8, v9
@@ -3339,9 +3351,10 @@ define <vscale x 1 x i64> @ctlz_zero_undef_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-NEXT:    li a0, 32
 ; RV32I-NEXT:    vsrl.vx v9, v8, a0
 ; RV32I-NEXT:    vor.vv v8, v8, v9
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v9, (a0), zero
 ; RV32I-NEXT:    vnot.v v8, v8
+; RV32I-NEXT:    addi a0, sp, 24
+; RV32I-NEXT:    vlse64.v v9, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v11, v8, 1
 ; RV32I-NEXT:    vand.vv v9, v11, v9
@@ -3350,7 +3363,9 @@ define <vscale x 1 x i64> @ctlz_zero_undef_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
 ; RV32I-NEXT:    vand.vv v8, v8, v10
 ; RV32I-NEXT:    vadd.vv v8, v9, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v9, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v11, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v11
@@ -3358,7 +3373,7 @@ define <vscale x 1 x i64> @ctlz_zero_undef_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v10
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: ctlz_zero_undef_nxv1i64:
@@ -3442,24 +3457,24 @@ define <vscale x 1 x i64> @ctlz_zero_undef_nxv1i64(<vscale x 1 x i64> %va) {
 define <vscale x 2 x i64> @ctlz_zero_undef_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-LABEL: ctlz_zero_undef_nxv2i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
 ; RV32I-NEXT:    vsrl.vi v10, v8, 1
 ; RV32I-NEXT:    vor.vv v8, v8, v10
@@ -3474,9 +3489,10 @@ define <vscale x 2 x i64> @ctlz_zero_undef_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-NEXT:    li a0, 32
 ; RV32I-NEXT:    vsrl.vx v10, v8, a0
 ; RV32I-NEXT:    vor.vv v8, v8, v10
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vnot.v v8, v8
+; RV32I-NEXT:    addi a0, sp, 24
+; RV32I-NEXT:    vlse64.v v10, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v14, v8, 1
 ; RV32I-NEXT:    vand.vv v10, v14, v10
@@ -3485,7 +3501,9 @@ define <vscale x 2 x i64> @ctlz_zero_undef_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
 ; RV32I-NEXT:    vand.vv v8, v8, v12
 ; RV32I-NEXT:    vadd.vv v8, v10, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v14, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v14
@@ -3493,7 +3511,7 @@ define <vscale x 2 x i64> @ctlz_zero_undef_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v12
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: ctlz_zero_undef_nxv2i64:
@@ -3577,24 +3595,24 @@ define <vscale x 2 x i64> @ctlz_zero_undef_nxv2i64(<vscale x 2 x i64> %va) {
 define <vscale x 4 x i64> @ctlz_zero_undef_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-LABEL: ctlz_zero_undef_nxv4i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    vsetvli a0, zero, e64, m4, ta, ma
 ; RV32I-NEXT:    vsrl.vi v12, v8, 1
 ; RV32I-NEXT:    vor.vv v8, v8, v12
@@ -3609,9 +3627,10 @@ define <vscale x 4 x i64> @ctlz_zero_undef_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-NEXT:    li a0, 32
 ; RV32I-NEXT:    vsrl.vx v12, v8, a0
 ; RV32I-NEXT:    vor.vv v8, v8, v12
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vnot.v v8, v8
+; RV32I-NEXT:    addi a0, sp, 24
+; RV32I-NEXT:    vlse64.v v12, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v20, v8, 1
 ; RV32I-NEXT:    vand.vv v12, v20, v12
@@ -3620,7 +3639,9 @@ define <vscale x 4 x i64> @ctlz_zero_undef_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
 ; RV32I-NEXT:    vand.vv v8, v8, v16
 ; RV32I-NEXT:    vadd.vv v8, v12, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v20, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v20
@@ -3628,7 +3649,7 @@ define <vscale x 4 x i64> @ctlz_zero_undef_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v16
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: ctlz_zero_undef_nxv4i64:
@@ -3712,24 +3733,24 @@ define <vscale x 4 x i64> @ctlz_zero_undef_nxv4i64(<vscale x 4 x i64> %va) {
 define <vscale x 8 x i64> @ctlz_zero_undef_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-LABEL: ctlz_zero_undef_nxv8i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; RV32I-NEXT:    vsrl.vi v16, v8, 1
 ; RV32I-NEXT:    vor.vv v8, v8, v16
@@ -3744,9 +3765,10 @@ define <vscale x 8 x i64> @ctlz_zero_undef_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-NEXT:    li a0, 32
 ; RV32I-NEXT:    vsrl.vx v16, v8, a0
 ; RV32I-NEXT:    vor.vv v8, v8, v16
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vnot.v v8, v8
+; RV32I-NEXT:    addi a0, sp, 24
+; RV32I-NEXT:    vlse64.v v16, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
 ; RV32I-NEXT:    vlse64.v v24, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v0, v8, 1
 ; RV32I-NEXT:    vand.vv v16, v0, v16
@@ -3755,7 +3777,9 @@ define <vscale x 8 x i64> @ctlz_zero_undef_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
 ; RV32I-NEXT:    vand.vv v8, v8, v24
 ; RV32I-NEXT:    vadd.vv v8, v16, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v24, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v0, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v0
@@ -3763,7 +3787,7 @@ define <vscale x 8 x i64> @ctlz_zero_undef_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v24
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: ctlz_zero_undef_nxv8i64:

diff --git a/llvm/test/CodeGen/RISCV/rvv/ctlz-vp.ll b/llvm/test/CodeGen/RISCV/rvv/ctlz-vp.ll
index d2dbc3a..d7faa8b 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/ctlz-vp.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/ctlz-vp.ll
@@ -2073,24 +2073,24 @@ declare <vscale x 1 x i64> @llvm.vp.ctlz.nxv1i64(<vscale x 1 x i64>, i1 immarg,
 define <vscale x 1 x i64> @vp_ctlz_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv1i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vsrl.vi v9, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v9, v0.t
@@ -2107,12 +2107,13 @@ define <vscale x 1 x i64> @vp_ctlz_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x
 ; RV32-NEXT:    vor.vv v8, v8, v9, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v9, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -2122,17 +2123,19 @@ define <vscale x 1 x i64> @vp_ctlz_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x
 ; RV32-NEXT:    vadd.vv v8, v10, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v9, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v9, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v9, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv1i64:
@@ -2189,24 +2192,24 @@ define <vscale x 1 x i64> @vp_ctlz_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x
 define <vscale x 1 x i64> @vp_ctlz_nxv1i64_unmasked(<vscale x 1 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv1i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vsrl.vi v9, v8, 1
 ; RV32-NEXT:    vor.vv v8, v8, v9
@@ -2223,12 +2226,13 @@ define <vscale x 1 x i64> @vp_ctlz_nxv1i64_unmasked(<vscale x 1 x i64> %va, i32
 ; RV32-NEXT:    vor.vv v8, v8, v9
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10
 ; RV32-NEXT:    vsub.vv v8, v8, v9
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -2238,17 +2242,19 @@ define <vscale x 1 x i64> @vp_ctlz_nxv1i64_unmasked(<vscale x 1 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v9
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v9
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v9
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv1i64_unmasked:
@@ -2309,24 +2315,24 @@ declare <vscale x 2 x i64> @llvm.vp.ctlz.nxv2i64(<vscale x 2 x i64>, i1 immarg,
 define <vscale x 2 x i64> @vp_ctlz_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv2i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vsrl.vi v10, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v10, v0.t
@@ -2343,12 +2349,13 @@ define <vscale x 2 x i64> @vp_ctlz_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x
 ; RV32-NEXT:    vor.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v10, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -2358,17 +2365,19 @@ define <vscale x 2 x i64> @vp_ctlz_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x
 ; RV32-NEXT:    vadd.vv v8, v12, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v10, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v10, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv2i64:
@@ -2425,24 +2434,24 @@ define <vscale x 2 x i64> @vp_ctlz_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x
 define <vscale x 2 x i64> @vp_ctlz_nxv2i64_unmasked(<vscale x 2 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv2i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vsrl.vi v10, v8, 1
 ; RV32-NEXT:    vor.vv v8, v8, v10
@@ -2459,12 +2468,13 @@ define <vscale x 2 x i64> @vp_ctlz_nxv2i64_unmasked(<vscale x 2 x i64> %va, i32
 ; RV32-NEXT:    vor.vv v8, v8, v10
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12
 ; RV32-NEXT:    vsub.vv v8, v8, v10
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -2474,17 +2484,19 @@ define <vscale x 2 x i64> @vp_ctlz_nxv2i64_unmasked(<vscale x 2 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v10
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v10
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v10
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv2i64_unmasked:
@@ -2545,24 +2557,24 @@ declare <vscale x 4 x i64> @llvm.vp.ctlz.nxv4i64(<vscale x 4 x i64>, i1 immarg,
 define <vscale x 4 x i64> @vp_ctlz_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv4i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vsrl.vi v12, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v12, v0.t
@@ -2579,12 +2591,13 @@ define <vscale x 4 x i64> @vp_ctlz_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x
 ; RV32-NEXT:    vor.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v12, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -2594,17 +2607,19 @@ define <vscale x 4 x i64> @vp_ctlz_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x
 ; RV32-NEXT:    vadd.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v12, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v12, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv4i64:
@@ -2661,24 +2676,24 @@ define <vscale x 4 x i64> @vp_ctlz_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x
 define <vscale x 4 x i64> @vp_ctlz_nxv4i64_unmasked(<vscale x 4 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv4i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vsrl.vi v12, v8, 1
 ; RV32-NEXT:    vor.vv v8, v8, v12
@@ -2695,12 +2710,13 @@ define <vscale x 4 x i64> @vp_ctlz_nxv4i64_unmasked(<vscale x 4 x i64> %va, i32
 ; RV32-NEXT:    vor.vv v8, v8, v12
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16
 ; RV32-NEXT:    vsub.vv v8, v8, v12
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -2710,17 +2726,19 @@ define <vscale x 4 x i64> @vp_ctlz_nxv4i64_unmasked(<vscale x 4 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v12
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v12
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v12
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv4i64_unmasked:
@@ -2781,24 +2799,24 @@ declare <vscale x 7 x i64> @llvm.vp.ctlz.nxv7i64(<vscale x 7 x i64>, i1 immarg,
 define <vscale x 7 x i64> @vp_ctlz_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv7i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
@@ -2815,12 +2833,13 @@ define <vscale x 7 x i64> @vp_ctlz_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2830,17 +2849,19 @@ define <vscale x 7 x i64> @vp_ctlz_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x
 ; RV32-NEXT:    vadd.vv v8, v24, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv7i64:
@@ -2897,24 +2918,24 @@ define <vscale x 7 x i64> @vp_ctlz_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x
 define <vscale x 7 x i64> @vp_ctlz_nxv7i64_unmasked(<vscale x 7 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv7i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
 ; RV32-NEXT:    vor.vv v8, v8, v16
@@ -2931,12 +2952,13 @@ define <vscale x 7 x i64> @vp_ctlz_nxv7i64_unmasked(<vscale x 7 x i64> %va, i32
 ; RV32-NEXT:    vor.vv v8, v8, v16
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2946,17 +2968,19 @@ define <vscale x 7 x i64> @vp_ctlz_nxv7i64_unmasked(<vscale x 7 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv7i64_unmasked:
@@ -3017,24 +3041,24 @@ declare <vscale x 8 x i64> @llvm.vp.ctlz.nxv8i64(<vscale x 8 x i64>, i1 immarg,
 define <vscale x 8 x i64> @vp_ctlz_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
@@ -3051,12 +3075,13 @@ define <vscale x 8 x i64> @vp_ctlz_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -3066,17 +3091,19 @@ define <vscale x 8 x i64> @vp_ctlz_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x
 ; RV32-NEXT:    vadd.vv v8, v24, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv8i64:
@@ -3133,24 +3160,24 @@ define <vscale x 8 x i64> @vp_ctlz_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x
 define <vscale x 8 x i64> @vp_ctlz_nxv8i64_unmasked(<vscale x 8 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv8i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
 ; RV32-NEXT:    vor.vv v8, v8, v16
@@ -3167,12 +3194,13 @@ define <vscale x 8 x i64> @vp_ctlz_nxv8i64_unmasked(<vscale x 8 x i64> %va, i32
 ; RV32-NEXT:    vor.vv v8, v8, v16
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -3182,17 +3210,19 @@ define <vscale x 8 x i64> @vp_ctlz_nxv8i64_unmasked(<vscale x 8 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv8i64_unmasked:
@@ -3253,13 +3283,13 @@ declare <vscale x 16 x i64> @llvm.vp.ctlz.nxv16i64(<vscale x 16 x i64>, i1 immar
 define <vscale x 16 x i64> @vp_ctlz_nxv16i64(<vscale x 16 x i64> %va, <vscale x 16 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv16i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    li a2, 40
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x28, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 40 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x28, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 40 * vlenb
 ; RV32-NEXT:    vmv1r.v v1, v0
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    srli a2, a1, 3
@@ -3267,20 +3297,20 @@ define <vscale x 16 x i64> @vp_ctlz_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    vslidedown.vx v0, v0, a2
 ; RV32-NEXT:    lui a2, 349525
 ; RV32-NEXT:    addi a2, a2, 1365
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 44(sp)
+; RV32-NEXT:    sw a2, 40(sp)
 ; RV32-NEXT:    lui a2, 209715
 ; RV32-NEXT:    addi a2, a2, 819
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 36(sp)
+; RV32-NEXT:    sw a2, 32(sp)
 ; RV32-NEXT:    lui a2, 61681
 ; RV32-NEXT:    addi a2, a2, -241
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 28(sp)
+; RV32-NEXT:    sw a2, 24(sp)
 ; RV32-NEXT:    lui a2, 4112
 ; RV32-NEXT:    addi a2, a2, 257
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 20(sp)
+; RV32-NEXT:    sw a2, 16(sp)
 ; RV32-NEXT:    sub a2, a0, a1
 ; RV32-NEXT:    sltu a3, a0, a2
 ; RV32-NEXT:    addi a3, a3, -1
@@ -3303,111 +3333,114 @@ define <vscale x 16 x i64> @vp_ctlz_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    csrr a4, vlenb
 ; RV32-NEXT:    slli a4, a4, 5
 ; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
+; RV32-NEXT:    addi a4, a4, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsrl.vi v24, v16, 1, v0.t
-; RV32-NEXT:    addi a4, sp, 8
+; RV32-NEXT:    addi a4, sp, 40
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 4
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsub.vv v16, v16, v24, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a4, sp, 32
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 3
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 3
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsrl.vi v16, v16, 2, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 3
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 3
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vadd.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v24, v16, 4, v0.t
 ; RV32-NEXT:    vadd.vv v24, v16, v24, v0.t
+; RV32-NEXT:    addi a4, sp, 24
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
+; RV32-NEXT:    addi a4, sp, 16
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
 ; RV32-NEXT:    csrr a3, vlenb
 ; RV32-NEXT:    slli a3, a3, 3
 ; RV32-NEXT:    add a3, sp, a3
-; RV32-NEXT:    addi a3, a3, 16
+; RV32-NEXT:    addi a3, a3, 48
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmul.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    li a3, 56
 ; RV32-NEXT:    vsrl.vx v16, v16, a3, v0.t
-; RV32-NEXT:    addi a4, sp, 16
+; RV32-NEXT:    addi a4, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    bltu a0, a1, .LBB46_2
 ; RV32-NEXT:  # %bb.1:
@@ -3432,14 +3465,14 @@ define <vscale x 16 x i64> @vp_ctlz_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 4
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 5
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v8, v8, 2, v0.t
@@ -3451,23 +3484,23 @@ define <vscale x 16 x i64> @vp_ctlz_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    li a1, 24
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vx v8, v8, a3, v0.t
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    li a1, 40
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv16i64:
@@ -3588,28 +3621,28 @@ define <vscale x 16 x i64> @vp_ctlz_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 define <vscale x 16 x i64> @vp_ctlz_nxv16i64_unmasked(<vscale x 16 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_nxv16i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    slli a1, a1, 5
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x20, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 32 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x20, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 32 * vlenb
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 44(sp)
+; RV32-NEXT:    sw a1, 40(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 36(sp)
+; RV32-NEXT:    sw a1, 32(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    sub a2, a0, a1
 ; RV32-NEXT:    sltu a3, a0, a2
@@ -3631,18 +3664,19 @@ define <vscale x 16 x i64> @vp_ctlz_nxv16i64_unmasked(<vscale x 16 x i64> %va, i
 ; RV32-NEXT:    vor.vv v16, v16, v24
 ; RV32-NEXT:    vnot.v v16, v16
 ; RV32-NEXT:    vsrl.vi v24, v16, 1
-; RV32-NEXT:    addi a4, sp, 8
+; RV32-NEXT:    addi a4, sp, 40
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v0, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v0, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v0, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v0
 ; RV32-NEXT:    vsub.vv v16, v16, v24
+; RV32-NEXT:    addi a4, sp, 32
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v0, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
@@ -3652,27 +3686,29 @@ define <vscale x 16 x i64> @vp_ctlz_nxv16i64_unmasked(<vscale x 16 x i64> %va, i
 ; RV32-NEXT:    vadd.vv v16, v24, v16
 ; RV32-NEXT:    vsrl.vi v24, v16, 4
 ; RV32-NEXT:    vadd.vv v16, v16, v24
+; RV32-NEXT:    addi a4, sp, 24
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 4
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v24, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v24, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v16, v24
+; RV32-NEXT:    addi a4, sp, 16
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
 ; RV32-NEXT:    csrr a3, vlenb
 ; RV32-NEXT:    slli a3, a3, 3
 ; RV32-NEXT:    add a3, sp, a3
-; RV32-NEXT:    addi a3, a3, 16
+; RV32-NEXT:    addi a3, a3, 48
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmul.vv v24, v24, v16
 ; RV32-NEXT:    li a3, 56
 ; RV32-NEXT:    vsrl.vx v16, v24, a3
-; RV32-NEXT:    addi a4, sp, 16
+; RV32-NEXT:    addi a4, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    bltu a0, a1, .LBB47_2
 ; RV32-NEXT:  # %bb.1:
@@ -3697,7 +3733,7 @@ define <vscale x 16 x i64> @vp_ctlz_nxv16i64_unmasked(<vscale x 16 x i64> %va, i
 ; RV32-NEXT:    li a1, 24
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v24, v16
 ; RV32-NEXT:    vsub.vv v8, v8, v24
@@ -3710,22 +3746,22 @@ define <vscale x 16 x i64> @vp_ctlz_nxv16i64_unmasked(<vscale x 16 x i64> %va, i
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 4
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    vsrl.vx v8, v8, a3
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 5
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_nxv16i64_unmasked:
@@ -5865,24 +5901,24 @@ define <vscale x 16 x i32> @vp_ctlz_zero_undef_nxv16i32_unmasked(<vscale x 16 x
 define <vscale x 1 x i64> @vp_ctlz_zero_undef_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv1i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vsrl.vi v9, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v9, v0.t
@@ -5899,12 +5935,13 @@ define <vscale x 1 x i64> @vp_ctlz_zero_undef_nxv1i64(<vscale x 1 x i64> %va, <v
 ; RV32-NEXT:    vor.vv v8, v8, v9, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v9, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -5914,17 +5951,19 @@ define <vscale x 1 x i64> @vp_ctlz_zero_undef_nxv1i64(<vscale x 1 x i64> %va, <v
 ; RV32-NEXT:    vadd.vv v8, v10, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v9, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v9, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v9, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv1i64:
@@ -5981,24 +6020,24 @@ define <vscale x 1 x i64> @vp_ctlz_zero_undef_nxv1i64(<vscale x 1 x i64> %va, <v
 define <vscale x 1 x i64> @vp_ctlz_zero_undef_nxv1i64_unmasked(<vscale x 1 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv1i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vsrl.vi v9, v8, 1
 ; RV32-NEXT:    vor.vv v8, v8, v9
@@ -6015,12 +6054,13 @@ define <vscale x 1 x i64> @vp_ctlz_zero_undef_nxv1i64_unmasked(<vscale x 1 x i64
 ; RV32-NEXT:    vor.vv v8, v8, v9
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10
 ; RV32-NEXT:    vsub.vv v8, v8, v9
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -6030,17 +6070,19 @@ define <vscale x 1 x i64> @vp_ctlz_zero_undef_nxv1i64_unmasked(<vscale x 1 x i64
 ; RV32-NEXT:    vadd.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v9
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v9
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v9
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv1i64_unmasked:
@@ -6100,24 +6142,24 @@ define <vscale x 1 x i64> @vp_ctlz_zero_undef_nxv1i64_unmasked(<vscale x 1 x i64
 define <vscale x 2 x i64> @vp_ctlz_zero_undef_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv2i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vsrl.vi v10, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v10, v0.t
@@ -6134,12 +6176,13 @@ define <vscale x 2 x i64> @vp_ctlz_zero_undef_nxv2i64(<vscale x 2 x i64> %va, <v
 ; RV32-NEXT:    vor.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v10, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -6149,17 +6192,19 @@ define <vscale x 2 x i64> @vp_ctlz_zero_undef_nxv2i64(<vscale x 2 x i64> %va, <v
 ; RV32-NEXT:    vadd.vv v8, v12, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v10, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v10, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv2i64:
@@ -6216,24 +6261,24 @@ define <vscale x 2 x i64> @vp_ctlz_zero_undef_nxv2i64(<vscale x 2 x i64> %va, <v
 define <vscale x 2 x i64> @vp_ctlz_zero_undef_nxv2i64_unmasked(<vscale x 2 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv2i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vsrl.vi v10, v8, 1
 ; RV32-NEXT:    vor.vv v8, v8, v10
@@ -6250,12 +6295,13 @@ define <vscale x 2 x i64> @vp_ctlz_zero_undef_nxv2i64_unmasked(<vscale x 2 x i64
 ; RV32-NEXT:    vor.vv v8, v8, v10
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12
 ; RV32-NEXT:    vsub.vv v8, v8, v10
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -6265,17 +6311,19 @@ define <vscale x 2 x i64> @vp_ctlz_zero_undef_nxv2i64_unmasked(<vscale x 2 x i64
 ; RV32-NEXT:    vadd.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v10
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v10
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v10
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv2i64_unmasked:
@@ -6335,24 +6383,24 @@ define <vscale x 2 x i64> @vp_ctlz_zero_undef_nxv2i64_unmasked(<vscale x 2 x i64
 define <vscale x 4 x i64> @vp_ctlz_zero_undef_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv4i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vsrl.vi v12, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v12, v0.t
@@ -6369,12 +6417,13 @@ define <vscale x 4 x i64> @vp_ctlz_zero_undef_nxv4i64(<vscale x 4 x i64> %va, <v
 ; RV32-NEXT:    vor.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v12, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -6384,17 +6433,19 @@ define <vscale x 4 x i64> @vp_ctlz_zero_undef_nxv4i64(<vscale x 4 x i64> %va, <v
 ; RV32-NEXT:    vadd.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v12, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v12, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv4i64:
@@ -6451,24 +6502,24 @@ define <vscale x 4 x i64> @vp_ctlz_zero_undef_nxv4i64(<vscale x 4 x i64> %va, <v
 define <vscale x 4 x i64> @vp_ctlz_zero_undef_nxv4i64_unmasked(<vscale x 4 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv4i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vsrl.vi v12, v8, 1
 ; RV32-NEXT:    vor.vv v8, v8, v12
@@ -6485,12 +6536,13 @@ define <vscale x 4 x i64> @vp_ctlz_zero_undef_nxv4i64_unmasked(<vscale x 4 x i64
 ; RV32-NEXT:    vor.vv v8, v8, v12
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16
 ; RV32-NEXT:    vsub.vv v8, v8, v12
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -6500,17 +6552,19 @@ define <vscale x 4 x i64> @vp_ctlz_zero_undef_nxv4i64_unmasked(<vscale x 4 x i64
 ; RV32-NEXT:    vadd.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v12
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v12
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v12
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv4i64_unmasked:
@@ -6570,24 +6624,24 @@ define <vscale x 4 x i64> @vp_ctlz_zero_undef_nxv4i64_unmasked(<vscale x 4 x i64
 define <vscale x 7 x i64> @vp_ctlz_zero_undef_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv7i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
@@ -6604,12 +6658,13 @@ define <vscale x 7 x i64> @vp_ctlz_zero_undef_nxv7i64(<vscale x 7 x i64> %va, <v
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -6619,17 +6674,19 @@ define <vscale x 7 x i64> @vp_ctlz_zero_undef_nxv7i64(<vscale x 7 x i64> %va, <v
 ; RV32-NEXT:    vadd.vv v8, v24, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv7i64:
@@ -6686,24 +6743,24 @@ define <vscale x 7 x i64> @vp_ctlz_zero_undef_nxv7i64(<vscale x 7 x i64> %va, <v
 define <vscale x 7 x i64> @vp_ctlz_zero_undef_nxv7i64_unmasked(<vscale x 7 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv7i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
 ; RV32-NEXT:    vor.vv v8, v8, v16
@@ -6720,12 +6777,13 @@ define <vscale x 7 x i64> @vp_ctlz_zero_undef_nxv7i64_unmasked(<vscale x 7 x i64
 ; RV32-NEXT:    vor.vv v8, v8, v16
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -6735,17 +6793,19 @@ define <vscale x 7 x i64> @vp_ctlz_zero_undef_nxv7i64_unmasked(<vscale x 7 x i64
 ; RV32-NEXT:    vadd.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv7i64_unmasked:
@@ -6805,24 +6865,24 @@ define <vscale x 7 x i64> @vp_ctlz_zero_undef_nxv7i64_unmasked(<vscale x 7 x i64
 define <vscale x 8 x i64> @vp_ctlz_zero_undef_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
@@ -6839,12 +6899,13 @@ define <vscale x 8 x i64> @vp_ctlz_zero_undef_nxv8i64(<vscale x 8 x i64> %va, <v
 ; RV32-NEXT:    vor.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -6854,17 +6915,19 @@ define <vscale x 8 x i64> @vp_ctlz_zero_undef_nxv8i64(<vscale x 8 x i64> %va, <v
 ; RV32-NEXT:    vadd.vv v8, v24, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv8i64:
@@ -6921,24 +6984,24 @@ define <vscale x 8 x i64> @vp_ctlz_zero_undef_nxv8i64(<vscale x 8 x i64> %va, <v
 define <vscale x 8 x i64> @vp_ctlz_zero_undef_nxv8i64_unmasked(<vscale x 8 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv8i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
 ; RV32-NEXT:    vor.vv v8, v8, v16
@@ -6955,12 +7018,13 @@ define <vscale x 8 x i64> @vp_ctlz_zero_undef_nxv8i64_unmasked(<vscale x 8 x i64
 ; RV32-NEXT:    vor.vv v8, v8, v16
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -6970,17 +7034,19 @@ define <vscale x 8 x i64> @vp_ctlz_zero_undef_nxv8i64_unmasked(<vscale x 8 x i64
 ; RV32-NEXT:    vadd.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv8i64_unmasked:
@@ -7039,13 +7105,13 @@ define <vscale x 8 x i64> @vp_ctlz_zero_undef_nxv8i64_unmasked(<vscale x 8 x i64
 define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64(<vscale x 16 x i64> %va, <vscale x 16 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv16i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    li a2, 40
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x28, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 40 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x28, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 40 * vlenb
 ; RV32-NEXT:    vmv1r.v v1, v0
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    srli a2, a1, 3
@@ -7053,20 +7119,20 @@ define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    vslidedown.vx v0, v0, a2
 ; RV32-NEXT:    lui a2, 349525
 ; RV32-NEXT:    addi a2, a2, 1365
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 44(sp)
+; RV32-NEXT:    sw a2, 40(sp)
 ; RV32-NEXT:    lui a2, 209715
 ; RV32-NEXT:    addi a2, a2, 819
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 36(sp)
+; RV32-NEXT:    sw a2, 32(sp)
 ; RV32-NEXT:    lui a2, 61681
 ; RV32-NEXT:    addi a2, a2, -241
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 28(sp)
+; RV32-NEXT:    sw a2, 24(sp)
 ; RV32-NEXT:    lui a2, 4112
 ; RV32-NEXT:    addi a2, a2, 257
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 20(sp)
+; RV32-NEXT:    sw a2, 16(sp)
 ; RV32-NEXT:    sub a2, a0, a1
 ; RV32-NEXT:    sltu a3, a0, a2
 ; RV32-NEXT:    addi a3, a3, -1
@@ -7089,111 +7155,114 @@ define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    csrr a4, vlenb
 ; RV32-NEXT:    slli a4, a4, 5
 ; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
+; RV32-NEXT:    addi a4, a4, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsrl.vi v24, v16, 1, v0.t
-; RV32-NEXT:    addi a4, sp, 8
+; RV32-NEXT:    addi a4, sp, 40
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 4
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsub.vv v16, v16, v24, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a4, sp, 32
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 3
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 3
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsrl.vi v16, v16, 2, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 3
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 3
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vadd.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v24, v16, 4, v0.t
 ; RV32-NEXT:    vadd.vv v24, v16, v24, v0.t
+; RV32-NEXT:    addi a4, sp, 24
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
+; RV32-NEXT:    addi a4, sp, 16
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
 ; RV32-NEXT:    csrr a3, vlenb
 ; RV32-NEXT:    slli a3, a3, 3
 ; RV32-NEXT:    add a3, sp, a3
-; RV32-NEXT:    addi a3, a3, 16
+; RV32-NEXT:    addi a3, a3, 48
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmul.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    li a3, 56
 ; RV32-NEXT:    vsrl.vx v16, v16, a3, v0.t
-; RV32-NEXT:    addi a4, sp, 16
+; RV32-NEXT:    addi a4, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    bltu a0, a1, .LBB94_2
 ; RV32-NEXT:  # %bb.1:
@@ -7218,14 +7287,14 @@ define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 4
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 5
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v8, v8, 2, v0.t
@@ -7237,23 +7306,23 @@ define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    li a1, 24
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vx v8, v8, a3, v0.t
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    li a1, 40
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv16i64:
@@ -7374,28 +7443,28 @@ define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64(<vscale x 16 x i64> %va,
 define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64_unmasked(<vscale x 16 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctlz_zero_undef_nxv16i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    slli a1, a1, 5
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x20, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 32 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x20, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 32 * vlenb
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 44(sp)
+; RV32-NEXT:    sw a1, 40(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 36(sp)
+; RV32-NEXT:    sw a1, 32(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    sub a2, a0, a1
 ; RV32-NEXT:    sltu a3, a0, a2
@@ -7417,18 +7486,19 @@ define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64_unmasked(<vscale x 16 x
 ; RV32-NEXT:    vor.vv v16, v16, v24
 ; RV32-NEXT:    vnot.v v16, v16
 ; RV32-NEXT:    vsrl.vi v24, v16, 1
-; RV32-NEXT:    addi a4, sp, 8
+; RV32-NEXT:    addi a4, sp, 40
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v0, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v0, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v0, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v0
 ; RV32-NEXT:    vsub.vv v16, v16, v24
+; RV32-NEXT:    addi a4, sp, 32
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v0, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
@@ -7438,27 +7508,29 @@ define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64_unmasked(<vscale x 16 x
 ; RV32-NEXT:    vadd.vv v16, v24, v16
 ; RV32-NEXT:    vsrl.vi v24, v16, 4
 ; RV32-NEXT:    vadd.vv v16, v16, v24
+; RV32-NEXT:    addi a4, sp, 24
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 4
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v24, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v24, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v16, v24
+; RV32-NEXT:    addi a4, sp, 16
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
 ; RV32-NEXT:    csrr a3, vlenb
 ; RV32-NEXT:    slli a3, a3, 3
 ; RV32-NEXT:    add a3, sp, a3
-; RV32-NEXT:    addi a3, a3, 16
+; RV32-NEXT:    addi a3, a3, 48
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmul.vv v24, v24, v16
 ; RV32-NEXT:    li a3, 56
 ; RV32-NEXT:    vsrl.vx v16, v24, a3
-; RV32-NEXT:    addi a4, sp, 16
+; RV32-NEXT:    addi a4, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    bltu a0, a1, .LBB95_2
 ; RV32-NEXT:  # %bb.1:
@@ -7483,7 +7555,7 @@ define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64_unmasked(<vscale x 16 x
 ; RV32-NEXT:    li a1, 24
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v24, v16
 ; RV32-NEXT:    vsub.vv v8, v8, v24
@@ -7496,22 +7568,22 @@ define <vscale x 16 x i64> @vp_ctlz_zero_undef_nxv16i64_unmasked(<vscale x 16 x
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 4
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    vsrl.vx v8, v8, a3
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 5
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctlz_zero_undef_nxv16i64_unmasked:

diff --git a/llvm/test/CodeGen/RISCV/rvv/ctpop-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/ctpop-sdnode.ll
index c72432d..f2cfd54 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/ctpop-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/ctpop-sdnode.ll
@@ -752,27 +752,28 @@ declare <vscale x 16 x i32> @llvm.ctpop.nxv16i32(<vscale x 16 x i32>)
 define <vscale x 1 x i64> @ctpop_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32-LABEL: ctpop_nxv1i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a0, 349525
 ; RV32-NEXT:    addi a0, a0, 1365
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 28(sp)
+; RV32-NEXT:    sw a0, 24(sp)
 ; RV32-NEXT:    lui a0, 209715
 ; RV32-NEXT:    addi a0, a0, 819
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 20(sp)
+; RV32-NEXT:    sw a0, 16(sp)
 ; RV32-NEXT:    lui a0, 61681
 ; RV32-NEXT:    addi a0, a0, -241
 ; RV32-NEXT:    sw a0, 12(sp)
 ; RV32-NEXT:    sw a0, 8(sp)
 ; RV32-NEXT:    lui a0, 4112
 ; RV32-NEXT:    addi a0, a0, 257
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 4(sp)
+; RV32-NEXT:    sw a0, 0(sp)
 ; RV32-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
-; RV32-NEXT:    addi a0, sp, 8
+; RV32-NEXT:    addi a0, sp, 24
 ; RV32-NEXT:    vlse64.v v9, (a0), zero
+; RV32-NEXT:    addi a0, sp, 16
 ; RV32-NEXT:    vlse64.v v10, (a0), zero
 ; RV32-NEXT:    vsrl.vi v11, v8, 1
 ; RV32-NEXT:    vand.vv v9, v11, v9
@@ -781,7 +782,9 @@ define <vscale x 1 x i64> @ctpop_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32-NEXT:    vsrl.vi v8, v8, 2
 ; RV32-NEXT:    vand.vv v8, v8, v10
 ; RV32-NEXT:    vadd.vv v8, v9, v8
+; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vlse64.v v9, (a0), zero
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v10, (a0), zero
 ; RV32-NEXT:    vsrl.vi v11, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v11
@@ -789,7 +792,7 @@ define <vscale x 1 x i64> @ctpop_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32-NEXT:    vmul.vv v8, v8, v10
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: ctpop_nxv1i64:
@@ -833,27 +836,28 @@ declare <vscale x 1 x i64> @llvm.ctpop.nxv1i64(<vscale x 1 x i64>)
 define <vscale x 2 x i64> @ctpop_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32-LABEL: ctpop_nxv2i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a0, 349525
 ; RV32-NEXT:    addi a0, a0, 1365
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 28(sp)
+; RV32-NEXT:    sw a0, 24(sp)
 ; RV32-NEXT:    lui a0, 209715
 ; RV32-NEXT:    addi a0, a0, 819
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 20(sp)
+; RV32-NEXT:    sw a0, 16(sp)
 ; RV32-NEXT:    lui a0, 61681
 ; RV32-NEXT:    addi a0, a0, -241
 ; RV32-NEXT:    sw a0, 12(sp)
 ; RV32-NEXT:    sw a0, 8(sp)
 ; RV32-NEXT:    lui a0, 4112
 ; RV32-NEXT:    addi a0, a0, 257
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 4(sp)
+; RV32-NEXT:    sw a0, 0(sp)
 ; RV32-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
-; RV32-NEXT:    addi a0, sp, 8
+; RV32-NEXT:    addi a0, sp, 24
 ; RV32-NEXT:    vlse64.v v10, (a0), zero
+; RV32-NEXT:    addi a0, sp, 16
 ; RV32-NEXT:    vlse64.v v12, (a0), zero
 ; RV32-NEXT:    vsrl.vi v14, v8, 1
 ; RV32-NEXT:    vand.vv v10, v14, v10
@@ -862,7 +866,9 @@ define <vscale x 2 x i64> @ctpop_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32-NEXT:    vsrl.vi v8, v8, 2
 ; RV32-NEXT:    vand.vv v8, v8, v12
 ; RV32-NEXT:    vadd.vv v8, v10, v8
+; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vlse64.v v10, (a0), zero
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v12, (a0), zero
 ; RV32-NEXT:    vsrl.vi v14, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v14
@@ -870,7 +876,7 @@ define <vscale x 2 x i64> @ctpop_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32-NEXT:    vmul.vv v8, v8, v12
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: ctpop_nxv2i64:
@@ -914,27 +920,28 @@ declare <vscale x 2 x i64> @llvm.ctpop.nxv2i64(<vscale x 2 x i64>)
 define <vscale x 4 x i64> @ctpop_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32-LABEL: ctpop_nxv4i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a0, 349525
 ; RV32-NEXT:    addi a0, a0, 1365
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 28(sp)
+; RV32-NEXT:    sw a0, 24(sp)
 ; RV32-NEXT:    lui a0, 209715
 ; RV32-NEXT:    addi a0, a0, 819
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 20(sp)
+; RV32-NEXT:    sw a0, 16(sp)
 ; RV32-NEXT:    lui a0, 61681
 ; RV32-NEXT:    addi a0, a0, -241
 ; RV32-NEXT:    sw a0, 12(sp)
 ; RV32-NEXT:    sw a0, 8(sp)
 ; RV32-NEXT:    lui a0, 4112
 ; RV32-NEXT:    addi a0, a0, 257
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 4(sp)
+; RV32-NEXT:    sw a0, 0(sp)
 ; RV32-NEXT:    vsetvli a0, zero, e64, m4, ta, ma
-; RV32-NEXT:    addi a0, sp, 8
+; RV32-NEXT:    addi a0, sp, 24
 ; RV32-NEXT:    vlse64.v v12, (a0), zero
+; RV32-NEXT:    addi a0, sp, 16
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vsrl.vi v20, v8, 1
 ; RV32-NEXT:    vand.vv v12, v20, v12
@@ -943,7 +950,9 @@ define <vscale x 4 x i64> @ctpop_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32-NEXT:    vsrl.vi v8, v8, 2
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    vadd.vv v8, v12, v8
+; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vlse64.v v12, (a0), zero
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vsrl.vi v20, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v20
@@ -951,7 +960,7 @@ define <vscale x 4 x i64> @ctpop_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: ctpop_nxv4i64:
@@ -995,27 +1004,28 @@ declare <vscale x 4 x i64> @llvm.ctpop.nxv4i64(<vscale x 4 x i64>)
 define <vscale x 8 x i64> @ctpop_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32-LABEL: ctpop_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a0, 349525
 ; RV32-NEXT:    addi a0, a0, 1365
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 28(sp)
+; RV32-NEXT:    sw a0, 24(sp)
 ; RV32-NEXT:    lui a0, 209715
 ; RV32-NEXT:    addi a0, a0, 819
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 20(sp)
+; RV32-NEXT:    sw a0, 16(sp)
 ; RV32-NEXT:    lui a0, 61681
 ; RV32-NEXT:    addi a0, a0, -241
 ; RV32-NEXT:    sw a0, 12(sp)
 ; RV32-NEXT:    sw a0, 8(sp)
 ; RV32-NEXT:    lui a0, 4112
 ; RV32-NEXT:    addi a0, a0, 257
-; RV32-NEXT:    sw a0, 12(sp)
-; RV32-NEXT:    sw a0, 8(sp)
+; RV32-NEXT:    sw a0, 4(sp)
+; RV32-NEXT:    sw a0, 0(sp)
 ; RV32-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
-; RV32-NEXT:    addi a0, sp, 8
+; RV32-NEXT:    addi a0, sp, 24
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
+; RV32-NEXT:    addi a0, sp, 16
 ; RV32-NEXT:    vlse64.v v24, (a0), zero
 ; RV32-NEXT:    vsrl.vi v0, v8, 1
 ; RV32-NEXT:    vand.vv v16, v0, v16
@@ -1024,7 +1034,9 @@ define <vscale x 8 x i64> @ctpop_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32-NEXT:    vsrl.vi v8, v8, 2
 ; RV32-NEXT:    vand.vv v8, v8, v24
 ; RV32-NEXT:    vadd.vv v8, v16, v8
+; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v24, (a0), zero
 ; RV32-NEXT:    vsrl.vi v0, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v0
@@ -1032,7 +1044,7 @@ define <vscale x 8 x i64> @ctpop_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32-NEXT:    vmul.vv v8, v8, v24
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: ctpop_nxv8i64:

diff --git a/llvm/test/CodeGen/RISCV/rvv/ctpop-vp.ll b/llvm/test/CodeGen/RISCV/rvv/ctpop-vp.ll
index f17ad49..3b33f00 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/ctpop-vp.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/ctpop-vp.ll
@@ -1539,32 +1539,33 @@ declare <vscale x 1 x i64> @llvm.vp.ctpop.nxv1i64(<vscale x 1 x i64>, <vscale x
 define <vscale x 1 x i64> @vp_ctpop_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv1i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vsrl.vi v9, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v9, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -1574,17 +1575,19 @@ define <vscale x 1 x i64> @vp_ctpop_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1
 ; RV32-NEXT:    vadd.vv v8, v10, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v9, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v9, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v9, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv1i64:
@@ -1627,32 +1630,33 @@ define <vscale x 1 x i64> @vp_ctpop_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1
 define <vscale x 1 x i64> @vp_ctpop_nxv1i64_unmasked(<vscale x 1 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv1i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vsrl.vi v9, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10
 ; RV32-NEXT:    vsub.vv v8, v8, v9
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -1662,17 +1666,19 @@ define <vscale x 1 x i64> @vp_ctpop_nxv1i64_unmasked(<vscale x 1 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v9
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v9
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v9
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv1i64_unmasked:
@@ -1719,32 +1725,33 @@ declare <vscale x 2 x i64> @llvm.vp.ctpop.nxv2i64(<vscale x 2 x i64>, <vscale x
 define <vscale x 2 x i64> @vp_ctpop_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv2i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vsrl.vi v10, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v10, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -1754,17 +1761,19 @@ define <vscale x 2 x i64> @vp_ctpop_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2
 ; RV32-NEXT:    vadd.vv v8, v12, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v10, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v10, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv2i64:
@@ -1807,32 +1816,33 @@ define <vscale x 2 x i64> @vp_ctpop_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2
 define <vscale x 2 x i64> @vp_ctpop_nxv2i64_unmasked(<vscale x 2 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv2i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vsrl.vi v10, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12
 ; RV32-NEXT:    vsub.vv v8, v8, v10
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -1842,17 +1852,19 @@ define <vscale x 2 x i64> @vp_ctpop_nxv2i64_unmasked(<vscale x 2 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v10
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v10
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v10
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv2i64_unmasked:
@@ -1899,32 +1911,33 @@ declare <vscale x 4 x i64> @llvm.vp.ctpop.nxv4i64(<vscale x 4 x i64>, <vscale x
 define <vscale x 4 x i64> @vp_ctpop_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv4i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vsrl.vi v12, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v12, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -1934,17 +1947,19 @@ define <vscale x 4 x i64> @vp_ctpop_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4
 ; RV32-NEXT:    vadd.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v12, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v12, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv4i64:
@@ -1987,32 +2002,33 @@ define <vscale x 4 x i64> @vp_ctpop_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4
 define <vscale x 4 x i64> @vp_ctpop_nxv4i64_unmasked(<vscale x 4 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv4i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vsrl.vi v12, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16
 ; RV32-NEXT:    vsub.vv v8, v8, v12
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -2022,17 +2038,19 @@ define <vscale x 4 x i64> @vp_ctpop_nxv4i64_unmasked(<vscale x 4 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v12
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v12
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v12
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv4i64_unmasked:
@@ -2079,32 +2097,33 @@ declare <vscale x 7 x i64> @llvm.vp.ctpop.nxv7i64(<vscale x 7 x i64>, <vscale x
 define <vscale x 7 x i64> @vp_ctpop_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv7i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2114,17 +2133,19 @@ define <vscale x 7 x i64> @vp_ctpop_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7
 ; RV32-NEXT:    vadd.vv v8, v24, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv7i64:
@@ -2167,32 +2188,33 @@ define <vscale x 7 x i64> @vp_ctpop_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7
 define <vscale x 7 x i64> @vp_ctpop_nxv7i64_unmasked(<vscale x 7 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv7i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2202,17 +2224,19 @@ define <vscale x 7 x i64> @vp_ctpop_nxv7i64_unmasked(<vscale x 7 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv7i64_unmasked:
@@ -2259,32 +2283,33 @@ declare <vscale x 8 x i64> @llvm.vp.ctpop.nxv8i64(<vscale x 8 x i64>, <vscale x
 define <vscale x 8 x i64> @vp_ctpop_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2294,17 +2319,19 @@ define <vscale x 8 x i64> @vp_ctpop_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8
 ; RV32-NEXT:    vadd.vv v8, v24, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv8i64:
@@ -2347,32 +2374,33 @@ define <vscale x 8 x i64> @vp_ctpop_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8
 define <vscale x 8 x i64> @vp_ctpop_nxv8i64_unmasked(<vscale x 8 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv8i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2382,17 +2410,19 @@ define <vscale x 8 x i64> @vp_ctpop_nxv8i64_unmasked(<vscale x 8 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv8i64_unmasked:
@@ -2439,19 +2469,19 @@ declare <vscale x 16 x i64> @llvm.vp.ctpop.nxv16i64(<vscale x 16 x i64>, <vscale
 define <vscale x 16 x i64> @vp_ctpop_nxv16i64(<vscale x 16 x i64> %va, <vscale x 16 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv16i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    li a2, 48
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x30, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 48 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x30, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 48 * vlenb
 ; RV32-NEXT:    vmv1r.v v1, v0
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a2, 40
+; RV32-NEXT:    li a2, 24
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    addi a1, a1, 48
 ; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    srli a2, a1, 3
@@ -2459,107 +2489,122 @@ define <vscale x 16 x i64> @vp_ctpop_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    vslidedown.vx v0, v0, a2
 ; RV32-NEXT:    lui a2, 349525
 ; RV32-NEXT:    addi a2, a2, 1365
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 44(sp)
+; RV32-NEXT:    sw a2, 40(sp)
 ; RV32-NEXT:    lui a2, 209715
 ; RV32-NEXT:    addi a2, a2, 819
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 36(sp)
+; RV32-NEXT:    sw a2, 32(sp)
 ; RV32-NEXT:    lui a2, 61681
 ; RV32-NEXT:    addi a2, a2, -241
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 28(sp)
+; RV32-NEXT:    sw a2, 24(sp)
 ; RV32-NEXT:    lui a2, 4112
 ; RV32-NEXT:    addi a2, a2, 257
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 20(sp)
+; RV32-NEXT:    sw a2, 16(sp)
 ; RV32-NEXT:    sub a2, a0, a1
 ; RV32-NEXT:    sltu a3, a0, a2
 ; RV32-NEXT:    addi a3, a3, -1
 ; RV32-NEXT:    and a2, a3, a2
 ; RV32-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v24, v16, 1, v0.t
-; RV32-NEXT:    addi a3, sp, 8
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 40
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vs8r.v v24, (a3) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a3, sp, 40
 ; RV32-NEXT:    vsetvli a4, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a3), zero
 ; RV32-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    slli a4, a4, 5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
-; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 40
-; RV32-NEXT:    mul a4, a4, a5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
-; RV32-NEXT:    vsub.vv v16, v16, v24, v0.t
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 40
-; RV32-NEXT:    mul a4, a4, a5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    slli a3, a3, 5
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 40
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vl8r.v v24, (a3) # Unknown-size Folded Reload
+; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 24
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vl8r.v v24, (a3) # Unknown-size Folded Reload
+; RV32-NEXT:    vsub.vv v16, v24, v16, v0.t
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 40
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a3, sp, 32
 ; RV32-NEXT:    vsetvli a4, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a3), zero
 ; RV32-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 40
-; RV32-NEXT:    mul a4, a4, a5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 40
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vl8r.v v24, (a3) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 24
-; RV32-NEXT:    mul a4, a4, a5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs8r.v v24, (a4) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 40
-; RV32-NEXT:    mul a4, a4, a5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 24
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vs8r.v v24, (a3) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 40
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vl8r.v v24, (a3) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsrl.vi v24, v24, 2, v0.t
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 40
-; RV32-NEXT:    mul a4, a4, a5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 40
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 24
-; RV32-NEXT:    mul a4, a4, a5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 24
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vl8r.v v24, (a3) # Unknown-size Folded Reload
 ; RV32-NEXT:    vadd.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v24, v16, 4, v0.t
 ; RV32-NEXT:    vadd.vv v24, v16, v24, v0.t
+; RV32-NEXT:    addi a3, sp, 24
 ; RV32-NEXT:    vsetvli a4, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a3), zero
 ; RV32-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 24
-; RV32-NEXT:    mul a4, a4, a5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 24
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
+; RV32-NEXT:    addi a3, sp, 16
 ; RV32-NEXT:    vsetvli a4, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a3), zero
-; RV32-NEXT:    addi a3, sp, 16
+; RV32-NEXT:    addi a3, sp, 48
 ; RV32-NEXT:    vs8r.v v8, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
 ; RV32-NEXT:    csrr a2, vlenb
 ; RV32-NEXT:    slli a2, a2, 4
 ; RV32-NEXT:    add a2, sp, a2
-; RV32-NEXT:    addi a2, a2, 16
+; RV32-NEXT:    addi a2, a2, 48
 ; RV32-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmul.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    li a2, 56
@@ -2567,7 +2612,7 @@ define <vscale x 16 x i64> @vp_ctpop_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    csrr a3, vlenb
 ; RV32-NEXT:    slli a3, a3, 3
 ; RV32-NEXT:    add a3, sp, a3
-; RV32-NEXT:    addi a3, a3, 16
+; RV32-NEXT:    addi a3, a3, 48
 ; RV32-NEXT:    vs8r.v v8, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    bltu a0, a1, .LBB46_2
 ; RV32-NEXT:  # %bb.1:
@@ -2575,13 +2620,13 @@ define <vscale x 16 x i64> @vp_ctpop_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:  .LBB46_2:
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmv1r.v v0, v1
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v8, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsrl.vi v24, v8, 1, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 5
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
@@ -2589,7 +2634,7 @@ define <vscale x 16 x i64> @vp_ctpop_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    li a1, 40
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v8, v8, 2, v0.t
@@ -2601,26 +2646,26 @@ define <vscale x 16 x i64> @vp_ctpop_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    li a1, 24
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 4
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vx v8, v8, a2, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    li a1, 48
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv16i64:
@@ -2714,28 +2759,28 @@ define <vscale x 16 x i64> @vp_ctpop_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 define <vscale x 16 x i64> @vp_ctpop_nxv16i64_unmasked(<vscale x 16 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_ctpop_nxv16i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    slli a1, a1, 5
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x20, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 32 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x20, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 32 * vlenb
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 44(sp)
+; RV32-NEXT:    sw a1, 40(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 36(sp)
+; RV32-NEXT:    sw a1, 32(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    sub a2, a0, a1
 ; RV32-NEXT:    sltu a3, a0, a2
@@ -2743,18 +2788,19 @@ define <vscale x 16 x i64> @vp_ctpop_nxv16i64_unmasked(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    and a2, a3, a2
 ; RV32-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
 ; RV32-NEXT:    vsrl.vi v24, v16, 1
-; RV32-NEXT:    addi a3, sp, 8
+; RV32-NEXT:    addi a3, sp, 40
 ; RV32-NEXT:    vsetvli a4, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v0, (a3), zero
 ; RV32-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 24
-; RV32-NEXT:    mul a4, a4, a5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs8r.v v0, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    li a4, 24
+; RV32-NEXT:    mul a3, a3, a4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vs8r.v v0, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v0
 ; RV32-NEXT:    vsub.vv v16, v16, v24
+; RV32-NEXT:    addi a3, sp, 32
 ; RV32-NEXT:    vsetvli a4, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v0, (a3), zero
 ; RV32-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
@@ -2764,27 +2810,29 @@ define <vscale x 16 x i64> @vp_ctpop_nxv16i64_unmasked(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    vadd.vv v16, v24, v16
 ; RV32-NEXT:    vsrl.vi v24, v16, 4
 ; RV32-NEXT:    vadd.vv v24, v16, v24
+; RV32-NEXT:    addi a3, sp, 24
 ; RV32-NEXT:    vsetvli a4, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a3), zero
 ; RV32-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    slli a4, a4, 4
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    slli a3, a3, 4
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 48
+; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v16, v24, v16
+; RV32-NEXT:    addi a3, sp, 16
 ; RV32-NEXT:    vsetvli a4, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a3), zero
 ; RV32-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
 ; RV32-NEXT:    csrr a2, vlenb
 ; RV32-NEXT:    slli a2, a2, 3
 ; RV32-NEXT:    add a2, sp, a2
-; RV32-NEXT:    addi a2, a2, 16
+; RV32-NEXT:    addi a2, a2, 48
 ; RV32-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmul.vv v16, v16, v24
 ; RV32-NEXT:    li a2, 56
 ; RV32-NEXT:    vsrl.vx v16, v16, a2
-; RV32-NEXT:    addi a3, sp, 16
+; RV32-NEXT:    addi a3, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    bltu a0, a1, .LBB47_2
 ; RV32-NEXT:  # %bb.1:
@@ -2796,7 +2844,7 @@ define <vscale x 16 x i64> @vp_ctpop_nxv16i64_unmasked(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    li a1, 24
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v24, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
@@ -2809,22 +2857,22 @@ define <vscale x 16 x i64> @vp_ctpop_nxv16i64_unmasked(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 4
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    vsrl.vx v8, v8, a2
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 5
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_ctpop_nxv16i64_unmasked:

diff --git a/llvm/test/CodeGen/RISCV/rvv/cttz-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/cttz-sdnode.ll
index 11c338e..77d0183 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/cttz-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/cttz-sdnode.ll
@@ -1356,40 +1356,43 @@ declare <vscale x 16 x i32> @llvm.cttz.nxv16i32(<vscale x 16 x i32>, i1)
 define <vscale x 1 x i64> @cttz_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-LABEL: cttz_nxv1i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    li a0, 1
 ; RV32I-NEXT:    vsetvli a1, zero, e64, m1, ta, ma
 ; RV32I-NEXT:    vsub.vx v9, v8, a0
 ; RV32I-NEXT:    vnot.v v8, v8
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vand.vv v8, v8, v9
+; RV32I-NEXT:    addi a0, sp, 24
 ; RV32I-NEXT:    vlse64.v v9, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
+; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v11, v8, 1
-; RV32I-NEXT:    vand.vv v10, v11, v10
-; RV32I-NEXT:    vsub.vv v8, v8, v10
-; RV32I-NEXT:    vand.vv v10, v8, v9
+; RV32I-NEXT:    vand.vv v9, v11, v9
+; RV32I-NEXT:    vsub.vv v8, v8, v9
+; RV32I-NEXT:    vand.vv v9, v8, v10
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
-; RV32I-NEXT:    vand.vv v8, v8, v9
-; RV32I-NEXT:    vadd.vv v8, v10, v8
+; RV32I-NEXT:    vand.vv v8, v8, v10
+; RV32I-NEXT:    vadd.vv v8, v9, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v9, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v11, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v11
@@ -1397,7 +1400,7 @@ define <vscale x 1 x i64> @cttz_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v10
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: cttz_nxv1i64:
@@ -1521,40 +1524,43 @@ declare <vscale x 1 x i64> @llvm.cttz.nxv1i64(<vscale x 1 x i64>, i1)
 define <vscale x 2 x i64> @cttz_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-LABEL: cttz_nxv2i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    li a0, 1
 ; RV32I-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
 ; RV32I-NEXT:    vsub.vx v10, v8, a0
 ; RV32I-NEXT:    vnot.v v8, v8
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vand.vv v8, v8, v10
+; RV32I-NEXT:    addi a0, sp, 24
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
+; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v14, v8, 1
-; RV32I-NEXT:    vand.vv v12, v14, v12
-; RV32I-NEXT:    vsub.vv v8, v8, v12
-; RV32I-NEXT:    vand.vv v12, v8, v10
+; RV32I-NEXT:    vand.vv v10, v14, v10
+; RV32I-NEXT:    vsub.vv v8, v8, v10
+; RV32I-NEXT:    vand.vv v10, v8, v12
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
-; RV32I-NEXT:    vand.vv v8, v8, v10
-; RV32I-NEXT:    vadd.vv v8, v12, v8
+; RV32I-NEXT:    vand.vv v8, v8, v12
+; RV32I-NEXT:    vadd.vv v8, v10, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v14, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v14
@@ -1562,7 +1568,7 @@ define <vscale x 2 x i64> @cttz_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v12
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: cttz_nxv2i64:
@@ -1686,40 +1692,43 @@ declare <vscale x 2 x i64> @llvm.cttz.nxv2i64(<vscale x 2 x i64>, i1)
 define <vscale x 4 x i64> @cttz_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-LABEL: cttz_nxv4i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    li a0, 1
 ; RV32I-NEXT:    vsetvli a1, zero, e64, m4, ta, ma
 ; RV32I-NEXT:    vsub.vx v12, v8, a0
 ; RV32I-NEXT:    vnot.v v8, v8
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vand.vv v8, v8, v12
+; RV32I-NEXT:    addi a0, sp, 24
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
+; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v20, v8, 1
-; RV32I-NEXT:    vand.vv v16, v20, v16
-; RV32I-NEXT:    vsub.vv v8, v8, v16
-; RV32I-NEXT:    vand.vv v16, v8, v12
+; RV32I-NEXT:    vand.vv v12, v20, v12
+; RV32I-NEXT:    vsub.vv v8, v8, v12
+; RV32I-NEXT:    vand.vv v12, v8, v16
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
-; RV32I-NEXT:    vand.vv v8, v8, v12
-; RV32I-NEXT:    vadd.vv v8, v16, v8
+; RV32I-NEXT:    vand.vv v8, v8, v16
+; RV32I-NEXT:    vadd.vv v8, v12, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v20, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v20
@@ -1727,7 +1736,7 @@ define <vscale x 4 x i64> @cttz_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v16
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: cttz_nxv4i64:
@@ -1851,40 +1860,43 @@ declare <vscale x 4 x i64> @llvm.cttz.nxv4i64(<vscale x 4 x i64>, i1)
 define <vscale x 8 x i64> @cttz_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-LABEL: cttz_nxv8i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    li a0, 1
 ; RV32I-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32I-NEXT:    vsub.vx v16, v8, a0
 ; RV32I-NEXT:    vnot.v v8, v8
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v24, (a0), zero
 ; RV32I-NEXT:    vand.vv v8, v8, v16
+; RV32I-NEXT:    addi a0, sp, 24
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
+; RV32I-NEXT:    vlse64.v v24, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v0, v8, 1
-; RV32I-NEXT:    vand.vv v24, v0, v24
-; RV32I-NEXT:    vsub.vv v8, v8, v24
-; RV32I-NEXT:    vand.vv v24, v8, v16
+; RV32I-NEXT:    vand.vv v16, v0, v16
+; RV32I-NEXT:    vsub.vv v8, v8, v16
+; RV32I-NEXT:    vand.vv v16, v8, v24
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
-; RV32I-NEXT:    vand.vv v8, v8, v16
-; RV32I-NEXT:    vadd.vv v8, v24, v8
+; RV32I-NEXT:    vand.vv v8, v8, v24
+; RV32I-NEXT:    vadd.vv v8, v16, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v24, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v0, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v0
@@ -1892,7 +1904,7 @@ define <vscale x 8 x i64> @cttz_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v24
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: cttz_nxv8i64:
@@ -3265,40 +3277,43 @@ define <vscale x 16 x i32> @cttz_zero_undef_nxv16i32(<vscale x 16 x i32> %va) {
 define <vscale x 1 x i64> @cttz_zero_undef_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-LABEL: cttz_zero_undef_nxv1i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    li a0, 1
 ; RV32I-NEXT:    vsetvli a1, zero, e64, m1, ta, ma
 ; RV32I-NEXT:    vsub.vx v9, v8, a0
 ; RV32I-NEXT:    vnot.v v8, v8
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vand.vv v8, v8, v9
+; RV32I-NEXT:    addi a0, sp, 24
 ; RV32I-NEXT:    vlse64.v v9, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
+; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v11, v8, 1
-; RV32I-NEXT:    vand.vv v10, v11, v10
-; RV32I-NEXT:    vsub.vv v8, v8, v10
-; RV32I-NEXT:    vand.vv v10, v8, v9
+; RV32I-NEXT:    vand.vv v9, v11, v9
+; RV32I-NEXT:    vsub.vv v8, v8, v9
+; RV32I-NEXT:    vand.vv v9, v8, v10
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
-; RV32I-NEXT:    vand.vv v8, v8, v9
-; RV32I-NEXT:    vadd.vv v8, v10, v8
+; RV32I-NEXT:    vand.vv v8, v8, v10
+; RV32I-NEXT:    vadd.vv v8, v9, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v9, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v11, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v11
@@ -3306,7 +3321,7 @@ define <vscale x 1 x i64> @cttz_zero_undef_nxv1i64(<vscale x 1 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v10
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: cttz_zero_undef_nxv1i64:
@@ -3385,40 +3400,43 @@ define <vscale x 1 x i64> @cttz_zero_undef_nxv1i64(<vscale x 1 x i64> %va) {
 define <vscale x 2 x i64> @cttz_zero_undef_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-LABEL: cttz_zero_undef_nxv2i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    li a0, 1
 ; RV32I-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
 ; RV32I-NEXT:    vsub.vx v10, v8, a0
 ; RV32I-NEXT:    vnot.v v8, v8
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vand.vv v8, v8, v10
+; RV32I-NEXT:    addi a0, sp, 24
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
+; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v14, v8, 1
-; RV32I-NEXT:    vand.vv v12, v14, v12
-; RV32I-NEXT:    vsub.vv v8, v8, v12
-; RV32I-NEXT:    vand.vv v12, v8, v10
+; RV32I-NEXT:    vand.vv v10, v14, v10
+; RV32I-NEXT:    vsub.vv v8, v8, v10
+; RV32I-NEXT:    vand.vv v10, v8, v12
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
-; RV32I-NEXT:    vand.vv v8, v8, v10
-; RV32I-NEXT:    vadd.vv v8, v12, v8
+; RV32I-NEXT:    vand.vv v8, v8, v12
+; RV32I-NEXT:    vadd.vv v8, v10, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v10, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v14, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v14
@@ -3426,7 +3444,7 @@ define <vscale x 2 x i64> @cttz_zero_undef_nxv2i64(<vscale x 2 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v12
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: cttz_zero_undef_nxv2i64:
@@ -3505,40 +3523,43 @@ define <vscale x 2 x i64> @cttz_zero_undef_nxv2i64(<vscale x 2 x i64> %va) {
 define <vscale x 4 x i64> @cttz_zero_undef_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-LABEL: cttz_zero_undef_nxv4i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    li a0, 1
 ; RV32I-NEXT:    vsetvli a1, zero, e64, m4, ta, ma
 ; RV32I-NEXT:    vsub.vx v12, v8, a0
 ; RV32I-NEXT:    vnot.v v8, v8
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vand.vv v8, v8, v12
+; RV32I-NEXT:    addi a0, sp, 24
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
+; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v20, v8, 1
-; RV32I-NEXT:    vand.vv v16, v20, v16
-; RV32I-NEXT:    vsub.vv v8, v8, v16
-; RV32I-NEXT:    vand.vv v16, v8, v12
+; RV32I-NEXT:    vand.vv v12, v20, v12
+; RV32I-NEXT:    vsub.vv v8, v8, v12
+; RV32I-NEXT:    vand.vv v12, v8, v16
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
-; RV32I-NEXT:    vand.vv v8, v8, v12
-; RV32I-NEXT:    vadd.vv v8, v16, v8
+; RV32I-NEXT:    vand.vv v8, v8, v16
+; RV32I-NEXT:    vadd.vv v8, v12, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v12, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v20, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v20
@@ -3546,7 +3567,7 @@ define <vscale x 4 x i64> @cttz_zero_undef_nxv4i64(<vscale x 4 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v16
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: cttz_zero_undef_nxv4i64:
@@ -3625,40 +3646,43 @@ define <vscale x 4 x i64> @cttz_zero_undef_nxv4i64(<vscale x 4 x i64> %va) {
 define <vscale x 8 x i64> @cttz_zero_undef_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-LABEL: cttz_zero_undef_nxv8i64:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    .cfi_def_cfa_offset 16
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    .cfi_def_cfa_offset 32
 ; RV32I-NEXT:    lui a0, 349525
 ; RV32I-NEXT:    addi a0, a0, 1365
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 28(sp)
+; RV32I-NEXT:    sw a0, 24(sp)
 ; RV32I-NEXT:    lui a0, 209715
 ; RV32I-NEXT:    addi a0, a0, 819
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 20(sp)
+; RV32I-NEXT:    sw a0, 16(sp)
 ; RV32I-NEXT:    lui a0, 61681
 ; RV32I-NEXT:    addi a0, a0, -241
 ; RV32I-NEXT:    sw a0, 12(sp)
 ; RV32I-NEXT:    sw a0, 8(sp)
 ; RV32I-NEXT:    lui a0, 4112
 ; RV32I-NEXT:    addi a0, a0, 257
-; RV32I-NEXT:    sw a0, 12(sp)
-; RV32I-NEXT:    sw a0, 8(sp)
+; RV32I-NEXT:    sw a0, 4(sp)
+; RV32I-NEXT:    sw a0, 0(sp)
 ; RV32I-NEXT:    li a0, 1
 ; RV32I-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32I-NEXT:    vsub.vx v16, v8, a0
 ; RV32I-NEXT:    vnot.v v8, v8
-; RV32I-NEXT:    addi a0, sp, 8
-; RV32I-NEXT:    vlse64.v v24, (a0), zero
 ; RV32I-NEXT:    vand.vv v8, v8, v16
+; RV32I-NEXT:    addi a0, sp, 24
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
+; RV32I-NEXT:    addi a0, sp, 16
+; RV32I-NEXT:    vlse64.v v24, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v0, v8, 1
-; RV32I-NEXT:    vand.vv v24, v0, v24
-; RV32I-NEXT:    vsub.vv v8, v8, v24
-; RV32I-NEXT:    vand.vv v24, v8, v16
+; RV32I-NEXT:    vand.vv v16, v0, v16
+; RV32I-NEXT:    vsub.vv v8, v8, v16
+; RV32I-NEXT:    vand.vv v16, v8, v24
 ; RV32I-NEXT:    vsrl.vi v8, v8, 2
-; RV32I-NEXT:    vand.vv v8, v8, v16
-; RV32I-NEXT:    vadd.vv v8, v24, v8
+; RV32I-NEXT:    vand.vv v8, v8, v24
+; RV32I-NEXT:    vadd.vv v8, v16, v8
+; RV32I-NEXT:    addi a0, sp, 8
 ; RV32I-NEXT:    vlse64.v v16, (a0), zero
+; RV32I-NEXT:    mv a0, sp
 ; RV32I-NEXT:    vlse64.v v24, (a0), zero
 ; RV32I-NEXT:    vsrl.vi v0, v8, 4
 ; RV32I-NEXT:    vadd.vv v8, v8, v0
@@ -3666,7 +3690,7 @@ define <vscale x 8 x i64> @cttz_zero_undef_nxv8i64(<vscale x 8 x i64> %va) {
 ; RV32I-NEXT:    vmul.vv v8, v8, v24
 ; RV32I-NEXT:    li a0, 56
 ; RV32I-NEXT:    vsrl.vx v8, v8, a0
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: cttz_zero_undef_nxv8i64:

diff --git a/llvm/test/CodeGen/RISCV/rvv/cttz-vp.ll b/llvm/test/CodeGen/RISCV/rvv/cttz-vp.ll
index b7b503c..ce0ce9d 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/cttz-vp.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/cttz-vp.ll
@@ -1771,36 +1771,37 @@ declare <vscale x 1 x i64> @llvm.vp.cttz.nxv1i64(<vscale x 1 x i64>, i1 immarg,
 define <vscale x 1 x i64> @vp_cttz_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv1i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vsub.vx v9, v8, a1, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v9, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v9, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -1810,17 +1811,19 @@ define <vscale x 1 x i64> @vp_cttz_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x
 ; RV32-NEXT:    vadd.vv v8, v10, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v9, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v9, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v9, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv1i64:
@@ -1867,36 +1870,37 @@ define <vscale x 1 x i64> @vp_cttz_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x
 define <vscale x 1 x i64> @vp_cttz_nxv1i64_unmasked(<vscale x 1 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv1i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vsub.vx v9, v8, a1
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vand.vv v8, v8, v9
 ; RV32-NEXT:    vsrl.vi v9, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10
 ; RV32-NEXT:    vsub.vv v8, v8, v9
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -1906,17 +1910,19 @@ define <vscale x 1 x i64> @vp_cttz_nxv1i64_unmasked(<vscale x 1 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v9
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v9
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v9
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv1i64_unmasked:
@@ -1967,36 +1973,37 @@ declare <vscale x 2 x i64> @llvm.vp.cttz.nxv2i64(<vscale x 2 x i64>, i1 immarg,
 define <vscale x 2 x i64> @vp_cttz_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv2i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vsub.vx v10, v8, a1, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v10, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -2006,17 +2013,19 @@ define <vscale x 2 x i64> @vp_cttz_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x
 ; RV32-NEXT:    vadd.vv v8, v12, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v10, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v10, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv2i64:
@@ -2063,36 +2072,37 @@ define <vscale x 2 x i64> @vp_cttz_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x
 define <vscale x 2 x i64> @vp_cttz_nxv2i64_unmasked(<vscale x 2 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv2i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vsub.vx v10, v8, a1
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vand.vv v8, v8, v10
 ; RV32-NEXT:    vsrl.vi v10, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12
 ; RV32-NEXT:    vsub.vv v8, v8, v10
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -2102,17 +2112,19 @@ define <vscale x 2 x i64> @vp_cttz_nxv2i64_unmasked(<vscale x 2 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v10
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v10
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v10
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv2i64_unmasked:
@@ -2163,36 +2175,37 @@ declare <vscale x 4 x i64> @llvm.vp.cttz.nxv4i64(<vscale x 4 x i64>, i1 immarg,
 define <vscale x 4 x i64> @vp_cttz_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv4i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vsub.vx v12, v8, a1, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v12, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -2202,17 +2215,19 @@ define <vscale x 4 x i64> @vp_cttz_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x
 ; RV32-NEXT:    vadd.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v12, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v12, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv4i64:
@@ -2259,36 +2274,37 @@ define <vscale x 4 x i64> @vp_cttz_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x
 define <vscale x 4 x i64> @vp_cttz_nxv4i64_unmasked(<vscale x 4 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv4i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vsub.vx v12, v8, a1
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vand.vv v8, v8, v12
 ; RV32-NEXT:    vsrl.vi v12, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16
 ; RV32-NEXT:    vsub.vv v8, v8, v12
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -2298,17 +2314,19 @@ define <vscale x 4 x i64> @vp_cttz_nxv4i64_unmasked(<vscale x 4 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v12
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v12
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v12
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv4i64_unmasked:
@@ -2359,36 +2377,37 @@ declare <vscale x 7 x i64> @llvm.vp.cttz.nxv7i64(<vscale x 7 x i64>, i1 immarg,
 define <vscale x 7 x i64> @vp_cttz_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv7i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsub.vx v16, v8, a1, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2398,17 +2417,19 @@ define <vscale x 7 x i64> @vp_cttz_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x
 ; RV32-NEXT:    vadd.vv v8, v24, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv7i64:
@@ -2455,36 +2476,37 @@ define <vscale x 7 x i64> @vp_cttz_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x
 define <vscale x 7 x i64> @vp_cttz_nxv7i64_unmasked(<vscale x 7 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv7i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsub.vx v16, v8, a1
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2494,17 +2516,19 @@ define <vscale x 7 x i64> @vp_cttz_nxv7i64_unmasked(<vscale x 7 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv7i64_unmasked:
@@ -2555,36 +2579,37 @@ declare <vscale x 8 x i64> @llvm.vp.cttz.nxv8i64(<vscale x 8 x i64>, i1 immarg,
 define <vscale x 8 x i64> @vp_cttz_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsub.vx v16, v8, a1, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2594,17 +2619,19 @@ define <vscale x 8 x i64> @vp_cttz_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x
 ; RV32-NEXT:    vadd.vv v8, v24, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv8i64:
@@ -2651,36 +2678,37 @@ define <vscale x 8 x i64> @vp_cttz_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x
 define <vscale x 8 x i64> @vp_cttz_nxv8i64_unmasked(<vscale x 8 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv8i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsub.vx v16, v8, a1
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -2690,17 +2718,19 @@ define <vscale x 8 x i64> @vp_cttz_nxv8i64_unmasked(<vscale x 8 x i64> %va, i32
 ; RV32-NEXT:    vadd.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv8i64_unmasked:
@@ -2751,13 +2781,13 @@ declare <vscale x 16 x i64> @llvm.vp.cttz.nxv16i64(<vscale x 16 x i64>, i1 immar
 define <vscale x 16 x i64> @vp_cttz_nxv16i64(<vscale x 16 x i64> %va, <vscale x 16 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv16i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    li a2, 40
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x28, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 40 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x28, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 40 * vlenb
 ; RV32-NEXT:    vmv1r.v v1, v0
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    srli a2, a1, 3
@@ -2765,20 +2795,20 @@ define <vscale x 16 x i64> @vp_cttz_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    vslidedown.vx v0, v0, a2
 ; RV32-NEXT:    lui a2, 349525
 ; RV32-NEXT:    addi a2, a2, 1365
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 44(sp)
+; RV32-NEXT:    sw a2, 40(sp)
 ; RV32-NEXT:    lui a2, 209715
 ; RV32-NEXT:    addi a2, a2, 819
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 36(sp)
+; RV32-NEXT:    sw a2, 32(sp)
 ; RV32-NEXT:    lui a2, 61681
 ; RV32-NEXT:    addi a2, a2, -241
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 28(sp)
+; RV32-NEXT:    sw a2, 24(sp)
 ; RV32-NEXT:    lui a2, 4112
 ; RV32-NEXT:    addi a2, a2, 257
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 20(sp)
+; RV32-NEXT:    sw a2, 16(sp)
 ; RV32-NEXT:    sub a2, a0, a1
 ; RV32-NEXT:    sltu a3, a0, a2
 ; RV32-NEXT:    addi a3, a3, -1
@@ -2789,113 +2819,101 @@ define <vscale x 16 x i64> @vp_cttz_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    vnot.v v16, v16, v0.t
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    slli a4, a4, 4
 ; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
+; RV32-NEXT:    addi a4, a4, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsrl.vi v24, v16, 1, v0.t
-; RV32-NEXT:    addi a4, sp, 8
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v24, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a4, sp, 40
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 4
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsub.vv v16, v16, v24, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a4, sp, 32
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 3
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    vsrl.vi v16, v16, 2, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v24, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    vsrl.vi v24, v24, 2, v0.t
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 3
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vadd.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v24, v16, 4, v0.t
 ; RV32-NEXT:    vadd.vv v24, v16, v24, v0.t
+; RV32-NEXT:    addi a4, sp, 24
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
+; RV32-NEXT:    addi a4, sp, 16
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
 ; RV32-NEXT:    csrr a3, vlenb
 ; RV32-NEXT:    slli a3, a3, 3
 ; RV32-NEXT:    add a3, sp, a3
-; RV32-NEXT:    addi a3, a3, 16
+; RV32-NEXT:    addi a3, a3, 48
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmul.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    li a3, 56
 ; RV32-NEXT:    vsrl.vx v16, v16, a3, v0.t
-; RV32-NEXT:    addi a4, sp, 16
+; RV32-NEXT:    addi a4, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    bltu a0, a1, .LBB46_2
 ; RV32-NEXT:  # %bb.1:
@@ -2908,16 +2926,17 @@ define <vscale x 16 x i64> @vp_cttz_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v24, v8, 1, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
-; RV32-NEXT:    slli a0, a0, 4
+; RV32-NEXT:    li a1, 24
+; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 5
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v8, v8, 2, v0.t
@@ -2926,26 +2945,25 @@ define <vscale x 16 x i64> @vp_cttz_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
-; RV32-NEXT:    li a1, 24
-; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    slli a0, a0, 4
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vx v8, v8, a3, v0.t
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    li a1, 40
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv16i64:
@@ -3046,28 +3064,28 @@ define <vscale x 16 x i64> @vp_cttz_nxv16i64(<vscale x 16 x i64> %va, <vscale x
 define <vscale x 16 x i64> @vp_cttz_nxv16i64_unmasked(<vscale x 16 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_nxv16i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    slli a1, a1, 5
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x20, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 32 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x20, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 32 * vlenb
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 44(sp)
+; RV32-NEXT:    sw a1, 40(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 36(sp)
+; RV32-NEXT:    sw a1, 32(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    sub a2, a0, a1
 ; RV32-NEXT:    sltu a3, a0, a2
@@ -3079,18 +3097,19 @@ define <vscale x 16 x i64> @vp_cttz_nxv16i64_unmasked(<vscale x 16 x i64> %va, i
 ; RV32-NEXT:    vnot.v v16, v16
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsrl.vi v24, v16, 1
-; RV32-NEXT:    addi a4, sp, 8
+; RV32-NEXT:    addi a4, sp, 40
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v0, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v0, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v0, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v0
 ; RV32-NEXT:    vsub.vv v16, v16, v24
+; RV32-NEXT:    addi a4, sp, 32
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v0, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
@@ -3100,27 +3119,29 @@ define <vscale x 16 x i64> @vp_cttz_nxv16i64_unmasked(<vscale x 16 x i64> %va, i
 ; RV32-NEXT:    vadd.vv v16, v24, v16
 ; RV32-NEXT:    vsrl.vi v24, v16, 4
 ; RV32-NEXT:    vadd.vv v16, v16, v24
+; RV32-NEXT:    addi a4, sp, 24
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 4
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v24, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v24, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v16, v24
+; RV32-NEXT:    addi a4, sp, 16
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
 ; RV32-NEXT:    csrr a3, vlenb
 ; RV32-NEXT:    slli a3, a3, 3
 ; RV32-NEXT:    add a3, sp, a3
-; RV32-NEXT:    addi a3, a3, 16
+; RV32-NEXT:    addi a3, a3, 48
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmul.vv v24, v24, v16
 ; RV32-NEXT:    li a3, 56
 ; RV32-NEXT:    vsrl.vx v16, v24, a3
-; RV32-NEXT:    addi a4, sp, 16
+; RV32-NEXT:    addi a4, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    bltu a0, a1, .LBB47_2
 ; RV32-NEXT:  # %bb.1:
@@ -3135,7 +3156,7 @@ define <vscale x 16 x i64> @vp_cttz_nxv16i64_unmasked(<vscale x 16 x i64> %va, i
 ; RV32-NEXT:    li a1, 24
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v24, v16
 ; RV32-NEXT:    vsub.vv v8, v8, v24
@@ -3148,22 +3169,22 @@ define <vscale x 16 x i64> @vp_cttz_nxv16i64_unmasked(<vscale x 16 x i64> %va, i
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 4
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    vsrl.vx v8, v8, a3
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 5
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_nxv16i64_unmasked:
@@ -4981,36 +5002,37 @@ define <vscale x 16 x i32> @vp_cttz_zero_undef_nxv16i32_unmasked(<vscale x 16 x
 define <vscale x 1 x i64> @vp_cttz_zero_undef_nxv1i64(<vscale x 1 x i64> %va, <vscale x 1 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv1i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vsub.vx v9, v8, a1, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v9, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v9, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -5020,17 +5042,19 @@ define <vscale x 1 x i64> @vp_cttz_zero_undef_nxv1i64(<vscale x 1 x i64> %va, <v
 ; RV32-NEXT:    vadd.vv v8, v10, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v9, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v9, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v9, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v9, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv1i64:
@@ -5077,36 +5101,37 @@ define <vscale x 1 x i64> @vp_cttz_zero_undef_nxv1i64(<vscale x 1 x i64> %va, <v
 define <vscale x 1 x i64> @vp_cttz_zero_undef_nxv1i64_unmasked(<vscale x 1 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv1i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vsub.vx v9, v8, a1
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vand.vv v8, v8, v9
 ; RV32-NEXT:    vsrl.vi v9, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v9, v9, v10
 ; RV32-NEXT:    vsub.vv v8, v8, v9
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
@@ -5116,17 +5141,19 @@ define <vscale x 1 x i64> @vp_cttz_zero_undef_nxv1i64_unmasked(<vscale x 1 x i64
 ; RV32-NEXT:    vadd.vv v8, v10, v8
 ; RV32-NEXT:    vsrl.vi v9, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v9
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v9
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
 ; RV32-NEXT:    vlse64.v v9, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v9
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv1i64_unmasked:
@@ -5176,36 +5203,37 @@ define <vscale x 1 x i64> @vp_cttz_zero_undef_nxv1i64_unmasked(<vscale x 1 x i64
 define <vscale x 2 x i64> @vp_cttz_zero_undef_nxv2i64(<vscale x 2 x i64> %va, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv2i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vsub.vx v10, v8, a1, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v10, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -5215,17 +5243,19 @@ define <vscale x 2 x i64> @vp_cttz_zero_undef_nxv2i64(<vscale x 2 x i64> %va, <v
 ; RV32-NEXT:    vadd.vv v8, v12, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v10, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v10, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v10, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v10, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv2i64:
@@ -5272,36 +5302,37 @@ define <vscale x 2 x i64> @vp_cttz_zero_undef_nxv2i64(<vscale x 2 x i64> %va, <v
 define <vscale x 2 x i64> @vp_cttz_zero_undef_nxv2i64_unmasked(<vscale x 2 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv2i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vsub.vx v10, v8, a1
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vand.vv v8, v8, v10
 ; RV32-NEXT:    vsrl.vi v10, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v10, v10, v12
 ; RV32-NEXT:    vsub.vv v8, v8, v10
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
@@ -5311,17 +5342,19 @@ define <vscale x 2 x i64> @vp_cttz_zero_undef_nxv2i64_unmasked(<vscale x 2 x i64
 ; RV32-NEXT:    vadd.vv v8, v12, v8
 ; RV32-NEXT:    vsrl.vi v10, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v10
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v10
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
 ; RV32-NEXT:    vlse64.v v10, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m2, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v10
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv2i64_unmasked:
@@ -5371,36 +5404,37 @@ define <vscale x 2 x i64> @vp_cttz_zero_undef_nxv2i64_unmasked(<vscale x 2 x i64
 define <vscale x 4 x i64> @vp_cttz_zero_undef_nxv4i64(<vscale x 4 x i64> %va, <vscale x 4 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv4i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vsub.vx v12, v8, a1, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v12, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -5410,17 +5444,19 @@ define <vscale x 4 x i64> @vp_cttz_zero_undef_nxv4i64(<vscale x 4 x i64> %va, <v
 ; RV32-NEXT:    vadd.vv v8, v16, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v12, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v12, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v12, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v12, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv4i64:
@@ -5467,36 +5503,37 @@ define <vscale x 4 x i64> @vp_cttz_zero_undef_nxv4i64(<vscale x 4 x i64> %va, <v
 define <vscale x 4 x i64> @vp_cttz_zero_undef_nxv4i64_unmasked(<vscale x 4 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv4i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vsub.vx v12, v8, a1
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vand.vv v8, v8, v12
 ; RV32-NEXT:    vsrl.vi v12, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v12, v12, v16
 ; RV32-NEXT:    vsub.vv v8, v8, v12
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
@@ -5506,17 +5543,19 @@ define <vscale x 4 x i64> @vp_cttz_zero_undef_nxv4i64_unmasked(<vscale x 4 x i64
 ; RV32-NEXT:    vadd.vv v8, v16, v8
 ; RV32-NEXT:    vsrl.vi v12, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v12
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v12
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; RV32-NEXT:    vlse64.v v12, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v12
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv4i64_unmasked:
@@ -5566,36 +5605,37 @@ define <vscale x 4 x i64> @vp_cttz_zero_undef_nxv4i64_unmasked(<vscale x 4 x i64
 define <vscale x 7 x i64> @vp_cttz_zero_undef_nxv7i64(<vscale x 7 x i64> %va, <vscale x 7 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv7i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsub.vx v16, v8, a1, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -5605,17 +5645,19 @@ define <vscale x 7 x i64> @vp_cttz_zero_undef_nxv7i64(<vscale x 7 x i64> %va, <v
 ; RV32-NEXT:    vadd.vv v8, v24, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv7i64:
@@ -5662,36 +5704,37 @@ define <vscale x 7 x i64> @vp_cttz_zero_undef_nxv7i64(<vscale x 7 x i64> %va, <v
 define <vscale x 7 x i64> @vp_cttz_zero_undef_nxv7i64_unmasked(<vscale x 7 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv7i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsub.vx v16, v8, a1
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -5701,17 +5744,19 @@ define <vscale x 7 x i64> @vp_cttz_zero_undef_nxv7i64_unmasked(<vscale x 7 x i64
 ; RV32-NEXT:    vadd.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv7i64_unmasked:
@@ -5761,36 +5806,37 @@ define <vscale x 7 x i64> @vp_cttz_zero_undef_nxv7i64_unmasked(<vscale x 7 x i64
 define <vscale x 8 x i64> @vp_cttz_zero_undef_nxv8i64(<vscale x 8 x i64> %va, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsub.vx v16, v8, a1, v0.t
 ; RV32-NEXT:    vnot.v v8, v8, v0.t
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 1, v0.t
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -5800,17 +5846,19 @@ define <vscale x 8 x i64> @vp_cttz_zero_undef_nxv8i64(<vscale x 8 x i64> %va, <v
 ; RV32-NEXT:    vadd.vv v8, v24, v8, v0.t
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0, v0.t
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv8i64:
@@ -5857,36 +5905,37 @@ define <vscale x 8 x i64> @vp_cttz_zero_undef_nxv8i64(<vscale x 8 x i64> %va, <v
 define <vscale x 8 x i64> @vp_cttz_zero_undef_nxv8i64_unmasked(<vscale x 8 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv8i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -32
+; RV32-NEXT:    .cfi_def_cfa_offset 32
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
 ; RV32-NEXT:    sw a1, 12(sp)
 ; RV32-NEXT:    sw a1, 8(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 4(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    li a1, 1
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vsub.vx v16, v8, a1
 ; RV32-NEXT:    vnot.v v8, v8
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    vsrl.vi v16, v8, 1
-; RV32-NEXT:    addi a1, sp, 8
+; RV32-NEXT:    addi a1, sp, 24
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsub.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
@@ -5896,17 +5945,19 @@ define <vscale x 8 x i64> @vp_cttz_zero_undef_nxv8i64_unmasked(<vscale x 8 x i64
 ; RV32-NEXT:    vadd.vv v8, v24, v8
 ; RV32-NEXT:    vsrl.vi v16, v8, 4
 ; RV32-NEXT:    vadd.vv v8, v8, v16
+; RV32-NEXT:    addi a1, sp, 8
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vand.vv v8, v8, v16
+; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a1), zero
 ; RV32-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    li a0, 56
 ; RV32-NEXT:    vsrl.vx v8, v8, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 32
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv8i64_unmasked:
@@ -5955,13 +6006,13 @@ define <vscale x 8 x i64> @vp_cttz_zero_undef_nxv8i64_unmasked(<vscale x 8 x i64
 define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64(<vscale x 16 x i64> %va, <vscale x 16 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv16i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    li a2, 40
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x28, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 40 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x28, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 40 * vlenb
 ; RV32-NEXT:    vmv1r.v v1, v0
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    srli a2, a1, 3
@@ -5969,20 +6020,20 @@ define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    vslidedown.vx v0, v0, a2
 ; RV32-NEXT:    lui a2, 349525
 ; RV32-NEXT:    addi a2, a2, 1365
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 44(sp)
+; RV32-NEXT:    sw a2, 40(sp)
 ; RV32-NEXT:    lui a2, 209715
 ; RV32-NEXT:    addi a2, a2, 819
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 36(sp)
+; RV32-NEXT:    sw a2, 32(sp)
 ; RV32-NEXT:    lui a2, 61681
 ; RV32-NEXT:    addi a2, a2, -241
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 28(sp)
+; RV32-NEXT:    sw a2, 24(sp)
 ; RV32-NEXT:    lui a2, 4112
 ; RV32-NEXT:    addi a2, a2, 257
-; RV32-NEXT:    sw a2, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a2, 20(sp)
+; RV32-NEXT:    sw a2, 16(sp)
 ; RV32-NEXT:    sub a2, a0, a1
 ; RV32-NEXT:    sltu a3, a0, a2
 ; RV32-NEXT:    addi a3, a3, -1
@@ -5993,113 +6044,101 @@ define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    vnot.v v16, v16, v0.t
 ; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
 ; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    slli a4, a4, 4
 ; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
+; RV32-NEXT:    addi a4, a4, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsrl.vi v24, v16, 1, v0.t
-; RV32-NEXT:    addi a4, sp, 8
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v24, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a4, sp, 40
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 4
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v16, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsub.vv v16, v16, v24, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    addi a4, sp, 32
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    vand.vv v16, v16, v24, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 3
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    vsrl.vi v16, v16, 2, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 5
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v16, (a5) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v24, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
+; RV32-NEXT:    vsrl.vi v24, v24, 2, v0.t
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 3
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vl8r.v v24, (a5) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vl8r.v v24, (a4) # Unknown-size Folded Reload
 ; RV32-NEXT:    vadd.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v24, v16, 4, v0.t
 ; RV32-NEXT:    vadd.vv v24, v16, v24, v0.t
+; RV32-NEXT:    addi a4, sp, 24
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v16, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v16, v0.t
+; RV32-NEXT:    addi a4, sp, 16
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
 ; RV32-NEXT:    csrr a3, vlenb
 ; RV32-NEXT:    slli a3, a3, 3
 ; RV32-NEXT:    add a3, sp, a3
-; RV32-NEXT:    addi a3, a3, 16
+; RV32-NEXT:    addi a3, a3, 48
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmul.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    li a3, 56
 ; RV32-NEXT:    vsrl.vx v16, v16, a3, v0.t
-; RV32-NEXT:    addi a4, sp, 16
+; RV32-NEXT:    addi a4, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    bltu a0, a1, .LBB94_2
 ; RV32-NEXT:  # %bb.1:
@@ -6112,16 +6151,17 @@ define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v24, v8, 1, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
-; RV32-NEXT:    slli a0, a0, 4
+; RV32-NEXT:    li a1, 24
+; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v16, v24, v16, v0.t
 ; RV32-NEXT:    vsub.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 5
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vi v8, v8, 2, v0.t
@@ -6130,26 +6170,25 @@ define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64(<vscale x 16 x i64> %va,
 ; RV32-NEXT:    vsrl.vi v16, v8, 4, v0.t
 ; RV32-NEXT:    vadd.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
-; RV32-NEXT:    li a1, 24
-; RV32-NEXT:    mul a0, a0, a1
+; RV32-NEXT:    slli a0, a0, 4
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmul.vv v8, v8, v16, v0.t
 ; RV32-NEXT:    vsrl.vx v8, v8, a3, v0.t
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    li a1, 40
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv16i64:
@@ -6250,28 +6289,28 @@ define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64(<vscale x 16 x i64> %va,
 define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64_unmasked(<vscale x 16 x i64> %va, i32 zeroext %evl) {
 ; RV32-LABEL: vp_cttz_zero_undef_nxv16i64_unmasked:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    addi sp, sp, -16
-; RV32-NEXT:    .cfi_def_cfa_offset 16
+; RV32-NEXT:    addi sp, sp, -48
+; RV32-NEXT:    .cfi_def_cfa_offset 48
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    slli a1, a1, 5
 ; RV32-NEXT:    sub sp, sp, a1
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x20, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 32 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x30, 0x22, 0x11, 0x20, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 48 + 32 * vlenb
 ; RV32-NEXT:    lui a1, 349525
 ; RV32-NEXT:    addi a1, a1, 1365
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 44(sp)
+; RV32-NEXT:    sw a1, 40(sp)
 ; RV32-NEXT:    lui a1, 209715
 ; RV32-NEXT:    addi a1, a1, 819
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 36(sp)
+; RV32-NEXT:    sw a1, 32(sp)
 ; RV32-NEXT:    lui a1, 61681
 ; RV32-NEXT:    addi a1, a1, -241
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 28(sp)
+; RV32-NEXT:    sw a1, 24(sp)
 ; RV32-NEXT:    lui a1, 4112
 ; RV32-NEXT:    addi a1, a1, 257
-; RV32-NEXT:    sw a1, 12(sp)
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 20(sp)
+; RV32-NEXT:    sw a1, 16(sp)
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    sub a2, a0, a1
 ; RV32-NEXT:    sltu a3, a0, a2
@@ -6283,18 +6322,19 @@ define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64_unmasked(<vscale x 16 x
 ; RV32-NEXT:    vnot.v v16, v16
 ; RV32-NEXT:    vand.vv v16, v16, v24
 ; RV32-NEXT:    vsrl.vi v24, v16, 1
-; RV32-NEXT:    addi a4, sp, 8
+; RV32-NEXT:    addi a4, sp, 40
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v0, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    li a6, 24
-; RV32-NEXT:    mul a5, a5, a6
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v0, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v0, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v24, v0
 ; RV32-NEXT:    vsub.vv v16, v16, v24
+; RV32-NEXT:    addi a4, sp, 32
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v0, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
@@ -6304,27 +6344,29 @@ define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64_unmasked(<vscale x 16 x
 ; RV32-NEXT:    vadd.vv v16, v24, v16
 ; RV32-NEXT:    vsrl.vi v24, v16, 4
 ; RV32-NEXT:    vadd.vv v16, v16, v24
+; RV32-NEXT:    addi a4, sp, 24
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v24, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
-; RV32-NEXT:    csrr a5, vlenb
-; RV32-NEXT:    slli a5, a5, 4
-; RV32-NEXT:    add a5, sp, a5
-; RV32-NEXT:    addi a5, a5, 16
-; RV32-NEXT:    vs8r.v v24, (a5) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 48
+; RV32-NEXT:    vs8r.v v24, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vand.vv v24, v16, v24
+; RV32-NEXT:    addi a4, sp, 16
 ; RV32-NEXT:    vsetvli a5, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v16, (a4), zero
 ; RV32-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
 ; RV32-NEXT:    csrr a3, vlenb
 ; RV32-NEXT:    slli a3, a3, 3
 ; RV32-NEXT:    add a3, sp, a3
-; RV32-NEXT:    addi a3, a3, 16
+; RV32-NEXT:    addi a3, a3, 48
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmul.vv v24, v24, v16
 ; RV32-NEXT:    li a3, 56
 ; RV32-NEXT:    vsrl.vx v16, v24, a3
-; RV32-NEXT:    addi a4, sp, 16
+; RV32-NEXT:    addi a4, sp, 48
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    bltu a0, a1, .LBB95_2
 ; RV32-NEXT:  # %bb.1:
@@ -6339,7 +6381,7 @@ define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64_unmasked(<vscale x 16 x
 ; RV32-NEXT:    li a1, 24
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v24, v24, v16
 ; RV32-NEXT:    vsub.vv v8, v8, v24
@@ -6352,22 +6394,22 @@ define <vscale x 16 x i64> @vp_cttz_zero_undef_nxv16i64_unmasked(<vscale x 16 x
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 4
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 3
 ; RV32-NEXT:    add a0, sp, a0
-; RV32-NEXT:    addi a0, a0, 16
+; RV32-NEXT:    addi a0, a0, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    vsrl.vx v8, v8, a3
-; RV32-NEXT:    addi a0, sp, 16
+; RV32-NEXT:    addi a0, sp, 48
 ; RV32-NEXT:    vl8r.v v16, (a0) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a0, vlenb
 ; RV32-NEXT:    slli a0, a0, 5
 ; RV32-NEXT:    add sp, sp, a0
-; RV32-NEXT:    addi sp, sp, 16
+; RV32-NEXT:    addi sp, sp, 48
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vp_cttz_zero_undef_nxv16i64_unmasked:

diff --git a/llvm/test/CodeGen/RISCV/rvv/splat-vector-split-i64-vl-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/splat-vector-split-i64-vl-sdnode.ll
new file mode 100644 (file)
index 0000000..0504163
--- /dev/null
+++ b/llvm/test/CodeGen/RISCV/rvv/splat-vector-split-i64-vl-sdnode.ll
@@ -0,0 +1,92 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+v -verify-machineinstrs < %s | FileCheck %s --check-prefixes=CHECK
+
+declare <vscale x 2 x i64> @llvm.bitreverse.nxv2i64(<vscale x 2 x i64>)
+
+define i32 @splat_vector_split_i64() {
+; CHECK-LABEL: splat_vector_split_i64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi sp, sp, -32
+; CHECK-NEXT:    .cfi_def_cfa_offset 32
+; CHECK-NEXT:    li a0, 3
+; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vmv.s.x v10, a0
+; CHECK-NEXT:    vmv.v.i v8, 0
+; CHECK-NEXT:    vsetivli zero, 4, e64, m2, tu, ma
+; CHECK-NEXT:    vslideup.vi v8, v10, 3
+; CHECK-NEXT:    sw zero, 4(sp)
+; CHECK-NEXT:    lui a0, 1044480
+; CHECK-NEXT:    sw a0, 0(sp)
+; CHECK-NEXT:    lui a0, 61681
+; CHECK-NEXT:    addi a0, a0, -241
+; CHECK-NEXT:    sw a0, 28(sp)
+; CHECK-NEXT:    sw a0, 24(sp)
+; CHECK-NEXT:    lui a0, 209715
+; CHECK-NEXT:    addi a0, a0, 819
+; CHECK-NEXT:    sw a0, 20(sp)
+; CHECK-NEXT:    sw a0, 16(sp)
+; CHECK-NEXT:    lui a0, 349525
+; CHECK-NEXT:    addi a0, a0, 1365
+; CHECK-NEXT:    sw a0, 12(sp)
+; CHECK-NEXT:    sw a0, 8(sp)
+; CHECK-NEXT:    li a0, 56
+; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vsrl.vx v10, v8, a0
+; CHECK-NEXT:    li a1, 40
+; CHECK-NEXT:    vsrl.vx v12, v8, a1
+; CHECK-NEXT:    lui a2, 16
+; CHECK-NEXT:    addi a2, a2, -256
+; CHECK-NEXT:    vand.vx v12, v12, a2
+; CHECK-NEXT:    vor.vv v10, v12, v10
+; CHECK-NEXT:    vsrl.vi v12, v8, 24
+; CHECK-NEXT:    mv a3, sp
+; CHECK-NEXT:    vlse64.v v14, (a3), zero
+; CHECK-NEXT:    lui a3, 4080
+; CHECK-NEXT:    vand.vx v12, v12, a3
+; CHECK-NEXT:    vsrl.vi v16, v8, 8
+; CHECK-NEXT:    vand.vv v16, v16, v14
+; CHECK-NEXT:    vor.vv v12, v16, v12
+; CHECK-NEXT:    vor.vv v10, v12, v10
+; CHECK-NEXT:    vand.vv v12, v8, v14
+; CHECK-NEXT:    vsll.vi v12, v12, 8
+; CHECK-NEXT:    vand.vx v14, v8, a3
+; CHECK-NEXT:    vsll.vi v14, v14, 24
+; CHECK-NEXT:    vor.vv v12, v14, v12
+; CHECK-NEXT:    vsll.vx v14, v8, a0
+; CHECK-NEXT:    vand.vx v8, v8, a2
+; CHECK-NEXT:    vsll.vx v8, v8, a1
+; CHECK-NEXT:    vor.vv v8, v14, v8
+; CHECK-NEXT:    addi a0, sp, 24
+; CHECK-NEXT:    vlse64.v v14, (a0), zero
+; CHECK-NEXT:    vor.vv v8, v8, v12
+; CHECK-NEXT:    vor.vv v8, v8, v10
+; CHECK-NEXT:    vsrl.vi v10, v8, 4
+; CHECK-NEXT:    vand.vv v10, v10, v14
+; CHECK-NEXT:    vand.vv v8, v8, v14
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vlse64.v v12, (a0), zero
+; CHECK-NEXT:    vsll.vi v8, v8, 4
+; CHECK-NEXT:    vor.vv v8, v10, v8
+; CHECK-NEXT:    vsrl.vi v10, v8, 2
+; CHECK-NEXT:    vand.vv v10, v10, v12
+; CHECK-NEXT:    vand.vv v8, v8, v12
+; CHECK-NEXT:    addi a0, sp, 8
+; CHECK-NEXT:    vlse64.v v12, (a0), zero
+; CHECK-NEXT:    vsll.vi v8, v8, 2
+; CHECK-NEXT:    vor.vv v8, v10, v8
+; CHECK-NEXT:    vsrl.vi v10, v8, 1
+; CHECK-NEXT:    vand.vv v10, v10, v12
+; CHECK-NEXT:    vand.vv v8, v8, v12
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    vor.vv v8, v10, v8
+; CHECK-NEXT:    vsetivli zero, 1, e64, m2, ta, ma
+; CHECK-NEXT:    vslidedown.vi v8, v8, 3
+; CHECK-NEXT:    vmv.x.s a0, v8
+; CHECK-NEXT:    addi sp, sp, 32
+; CHECK-NEXT:    ret
+    %1 = insertelement <vscale x 2 x i64> zeroinitializer, i64 3, i64 3
+    %2 = tail call <vscale x 2 x i64> @llvm.bitreverse.nxv2i64(<vscale x 2 x i64> %1)
+    %3 = extractelement <vscale x 2 x i64> %2, i32 3
+    %4 = trunc i64 %3 to i32
+    ret i32 %4
+}

diff --git a/llvm/test/CodeGen/RISCV/rvv/stepvector.ll b/llvm/test/CodeGen/RISCV/rvv/stepvector.ll
index e20783d..1f9f11a 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/stepvector.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/stepvector.ll
@@ -658,7 +658,7 @@ define <vscale x 16 x i64> @mul_bigimm_stepvector_nxv16i64() {
 ; RV32-NEXT:    lui a1, 11557
 ; RV32-NEXT:    addi a1, a1, -683
 ; RV32-NEXT:    mul a1, a0, a1
-; RV32-NEXT:    sw a1, 8(sp)
+; RV32-NEXT:    sw a1, 0(sp)
 ; RV32-NEXT:    srli a0, a0, 3
 ; RV32-NEXT:    li a1, 62
 ; RV32-NEXT:    mul a1, a0, a1
@@ -666,10 +666,11 @@ define <vscale x 16 x i64> @mul_bigimm_stepvector_nxv16i64() {
 ; RV32-NEXT:    addi a2, a2, -1368
 ; RV32-NEXT:    mulhu a0, a0, a2
 ; RV32-NEXT:    add a0, a0, a1
-; RV32-NEXT:    sw a0, 12(sp)
+; RV32-NEXT:    sw a0, 4(sp)
 ; RV32-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vlse64.v v8, (a0), zero
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vid.v v24
 ; RV32-NEXT:    vmul.vv v8, v24, v8

diff --git a/llvm/test/CodeGen/RISCV/rvv/vadd-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vadd-sdnode.ll
index cb30e1f..264cf2e 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/vadd-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vadd-sdnode.ll
@@ -870,8 +870,9 @@ define <vscale x 8 x i64> @vadd_xx_nxv8i64(i64 %a, i64 %b) nounwind {
 ; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v8, (a0), zero
-; RV32-NEXT:    sw a3, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a3, 4(sp)
+; RV32-NEXT:    sw a2, 0(sp)
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vadd.vv v8, v8, v16
 ; RV32-NEXT:    addi sp, sp, 16

diff --git a/llvm/test/CodeGen/RISCV/rvv/vand-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vand-sdnode.ll
index 2ac0b32..f10954b 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/vand-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vand-sdnode.ll
@@ -1361,8 +1361,9 @@ define <vscale x 8 x i64> @vand_xx_nxv8i64(i64 %a, i64 %b) nounwind {
 ; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v8, (a0), zero
-; RV32-NEXT:    sw a3, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a3, 4(sp)
+; RV32-NEXT:    sw a2, 0(sp)
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vand.vv v8, v8, v16
 ; RV32-NEXT:    addi sp, sp, 16

diff --git a/llvm/test/CodeGen/RISCV/rvv/vmul-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vmul-sdnode.ll
index 0394cf4..f1f7cdd 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/vmul-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmul-sdnode.ll
@@ -932,8 +932,9 @@ define <vscale x 8 x i64> @vmul_xx_nxv8i64(i64 %a, i64 %b) nounwind {
 ; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v8, (a0), zero
-; RV32-NEXT:    sw a3, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a3, 4(sp)
+; RV32-NEXT:    sw a2, 0(sp)
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vmul.vv v8, v8, v16
 ; RV32-NEXT:    addi sp, sp, 16

diff --git a/llvm/test/CodeGen/RISCV/rvv/vor-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vor-sdnode.ll
index bf6a8b9..3564676 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/vor-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vor-sdnode.ll
@@ -1154,8 +1154,9 @@ define <vscale x 8 x i64> @vor_xx_nxv8i64(i64 %a, i64 %b) nounwind {
 ; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v8, (a0), zero
-; RV32-NEXT:    sw a3, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a3, 4(sp)
+; RV32-NEXT:    sw a2, 0(sp)
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vor.vv v8, v8, v16
 ; RV32-NEXT:    addi sp, sp, 16

diff --git a/llvm/test/CodeGen/RISCV/rvv/vsub-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vsub-sdnode.ll
index 6a7b2a4..5ab028f 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/vsub-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vsub-sdnode.ll
@@ -848,8 +848,9 @@ define <vscale x 8 x i64> @vsub_xx_nxv8i64(i64 %a, i64 %b) nounwind {
 ; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v8, (a0), zero
-; RV32-NEXT:    sw a3, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a3, 4(sp)
+; RV32-NEXT:    sw a2, 0(sp)
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vsub.vv v8, v8, v16
 ; RV32-NEXT:    addi sp, sp, 16

diff --git a/llvm/test/CodeGen/RISCV/rvv/vxor-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vxor-sdnode.ll
index 8171dda..e8c4f1f 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/vxor-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vxor-sdnode.ll
@@ -1361,8 +1361,9 @@ define <vscale x 8 x i64> @vxor_xx_nxv8i64(i64 %a, i64 %b) nounwind {
 ; RV32-NEXT:    addi a0, sp, 8
 ; RV32-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vlse64.v v8, (a0), zero
-; RV32-NEXT:    sw a3, 12(sp)
-; RV32-NEXT:    sw a2, 8(sp)
+; RV32-NEXT:    sw a3, 4(sp)
+; RV32-NEXT:    sw a2, 0(sp)
+; RV32-NEXT:    mv a0, sp
 ; RV32-NEXT:    vlse64.v v16, (a0), zero
 ; RV32-NEXT:    vxor.vv v8, v8, v16
 ; RV32-NEXT:    addi sp, sp, 16