diff --git a/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp b/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp
index a9e5bb6ecd9b8..ed9b17716b359 100644
--- a/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp
+++ b/llvm/lib/Target/RISCV/RISCVVLOptimizer.cpp
@@ -270,6 +270,43 @@ static OperandInfo getOperandInfo(const MachineOperand &MO,
   case RISCV::VSSE64_V:
     return OperandInfo(RISCVVType::getEMULEqualsEEWDivSEWTimesLMUL(6, MI), 6);
 
+  // Vector Indexed Instructions
+  // vs(o|u)xei<eew>.v
+  // Dest/Data (operand 0) EEW=SEW, EMUL=LMUL. Source EEW=<eew> and
+  // EMUL=(EEW/SEW)*LMUL.
+  case RISCV::VLUXEI8_V:
+  case RISCV::VLOXEI8_V:
+  case RISCV::VSUXEI8_V:
+  case RISCV::VSOXEI8_V: {
+    if (MO.getOperandNo() == 0)
+      return OperandInfo(MIVLMul, MILog2SEW);
+    return OperandInfo(RISCVVType::getEMULEqualsEEWDivSEWTimesLMUL(3, MI), 3);
+  }
+  case RISCV::VLUXEI16_V:
+  case RISCV::VLOXEI16_V:
+  case RISCV::VSUXEI16_V:
+  case RISCV::VSOXEI16_V: {
+    if (MO.getOperandNo() == 0)
+      return OperandInfo(MIVLMul, MILog2SEW);
+    return OperandInfo(RISCVVType::getEMULEqualsEEWDivSEWTimesLMUL(4, MI), 4);
+  }
+  case RISCV::VLUXEI32_V:
+  case RISCV::VLOXEI32_V:
+  case RISCV::VSUXEI32_V:
+  case RISCV::VSOXEI32_V: {
+    if (MO.getOperandNo() == 0)
+      return OperandInfo(MIVLMul, MILog2SEW);
+    return OperandInfo(RISCVVType::getEMULEqualsEEWDivSEWTimesLMUL(5, MI), 5);
+  }
+  case RISCV::VLUXEI64_V:
+  case RISCV::VLOXEI64_V:
+  case RISCV::VSUXEI64_V:
+  case RISCV::VSOXEI64_V: {
+    if (MO.getOperandNo() == 0)
+      return OperandInfo(MIVLMul, MILog2SEW);
+    return OperandInfo(RISCVVType::getEMULEqualsEEWDivSEWTimesLMUL(6, MI), 6);
+  }
+
   // Vector Integer Arithmetic Instructions
   // Vector Single-Width Integer Add and Subtract
   case RISCV::VADD_VI:
diff --git a/llvm/test/CodeGen/RISCV/rvv/narrow-shift-extend.ll b/llvm/test/CodeGen/RISCV/rvv/narrow-shift-extend.ll
index e47517abacb4d..3fbe635576c9b 100644
--- a/llvm/test/CodeGen/RISCV/rvv/narrow-shift-extend.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/narrow-shift-extend.ll
@@ -10,10 +10,10 @@ declare <vscale x 4 x i32> @llvm.riscv.vloxei.nxv4i32.nxv4i64(
 define <vscale x 4 x i32> @test_vloxei(ptr %ptr, <vscale x 4 x i8> %offset, i64 %vl) {
 ; CHECK-LABEL: test_vloxei:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v8
 ; CHECK-NEXT:    vsll.vi v12, v12, 4
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vloxei64.v v8, (a0), v12
 ; CHECK-NEXT:    ret
 entry:
@@ -30,10 +30,10 @@ entry:
 define <vscale x 4 x i32> @test_vloxei2(ptr %ptr, <vscale x 4 x i8> %offset, i64 %vl) {
 ; CHECK-LABEL: test_vloxei2:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v8
 ; CHECK-NEXT:    vsll.vi v12, v12, 14
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vloxei64.v v8, (a0), v12
 ; CHECK-NEXT:    ret
 entry:
@@ -50,10 +50,10 @@ entry:
 define <vscale x 4 x i32> @test_vloxei3(ptr %ptr, <vscale x 4 x i8> %offset, i64 %vl) {
 ; CHECK-LABEL: test_vloxei3:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v8
 ; CHECK-NEXT:    vsll.vi v12, v12, 26
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vloxei64.v v8, (a0), v12
 ; CHECK-NEXT:    ret
 entry:
@@ -74,9 +74,8 @@ define <vscale x 4 x i32> @test_vloxei4(ptr %ptr, <vscale x 4 x i8> %offset, <vs
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v8, v0.t
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vsll.vi v12, v12, 4
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vloxei64.v v8, (a0), v12
 ; CHECK-NEXT:    ret
 entry:
@@ -100,10 +99,10 @@ declare <vscale x 4 x i32> @llvm.riscv.vloxei.nxv4i32.nxv4i16(
 define <vscale x 4 x i32> @test_vloxei5(ptr %ptr, <vscale x 4 x i8> %offset, i64 %vl) {
 ; CHECK-LABEL: test_vloxei5:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e16, m1, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e16, m1, ta, ma
 ; CHECK-NEXT:    vzext.vf2 v9, v8
 ; CHECK-NEXT:    vsll.vi v10, v9, 12
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vloxei16.v v8, (a0), v10
 ; CHECK-NEXT:    ret
 entry:
@@ -121,12 +120,12 @@ define <vscale x 4 x i32> @test_vloxei6(ptr %ptr, <vscale x 4 x i7> %offset, i64
 ; CHECK-LABEL: test_vloxei6:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    li a2, 127
-; CHECK-NEXT:    vsetvli a3, zero, e8, mf2, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e8, mf2, ta, ma
 ; CHECK-NEXT:    vand.vx v8, v8, a2
 ; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v8
 ; CHECK-NEXT:    vsll.vi v12, v12, 4
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vloxei64.v v8, (a0), v12
 ; CHECK-NEXT:    ret
 entry:
@@ -146,8 +145,9 @@ define <vscale x 4 x i32> @test_vloxei7(ptr %ptr, <vscale x 4 x i1> %offset, i64
 ; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vmv.v.i v8, 0
 ; CHECK-NEXT:    vmerge.vim v8, v8, 1, v0
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vsll.vi v12, v8, 2
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vloxei64.v v8, (a0), v12
 ; CHECK-NEXT:    ret
 entry:
@@ -172,10 +172,10 @@ declare <vscale x 4 x i32> @llvm.riscv.vloxei.mask.nxv4i32.nxv4i64(
 define <vscale x 4 x i32> @test_vloxei_mask(ptr %ptr, <vscale x 4 x i8> %offset, <vscale x 4 x i1> %m, i64 %vl) {
 ; CHECK-LABEL: test_vloxei_mask:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v8
 ; CHECK-NEXT:    vsll.vi v12, v12, 4
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vloxei64.v v8, (a0), v12, v0.t
 ; CHECK-NEXT:    ret
 entry:
@@ -199,10 +199,10 @@ declare <vscale x 4 x i32> @llvm.riscv.vluxei.nxv4i32.nxv4i64(
 define <vscale x 4 x i32> @test_vluxei(ptr %ptr, <vscale x 4 x i8> %offset, i64 %vl) {
 ; CHECK-LABEL: test_vluxei:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v8
 ; CHECK-NEXT:    vsll.vi v12, v12, 4
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vluxei64.v v8, (a0), v12
 ; CHECK-NEXT:    ret
 entry:
@@ -227,10 +227,10 @@ declare <vscale x 4 x i32> @llvm.riscv.vluxei.mask.nxv4i32.nxv4i64(
 define <vscale x 4 x i32> @test_vluxei_mask(ptr %ptr, <vscale x 4 x i8> %offset, <vscale x 4 x i1> %m, i64 %vl) {
 ; CHECK-LABEL: test_vluxei_mask:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v8
 ; CHECK-NEXT:    vsll.vi v12, v12, 4
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vluxei64.v v8, (a0), v12, v0.t
 ; CHECK-NEXT:    ret
 entry:
@@ -254,10 +254,10 @@ declare void @llvm.riscv.vsoxei.nxv4i32.nxv4i64(
 define void @test_vsoxei(<vscale x 4 x i32> %val, ptr %ptr, <vscale x 4 x i8> %offset, i64 %vl) {
 ; CHECK-LABEL: test_vsoxei:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v10
 ; CHECK-NEXT:    vsll.vi v12, v12, 4
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vsoxei64.v v8, (a0), v12
 ; CHECK-NEXT:    ret
 entry:
@@ -281,10 +281,10 @@ declare void @llvm.riscv.vsoxei.mask.nxv4i32.nxv4i64(
 define void @test_vsoxei_mask(<vscale x 4 x i32> %val, ptr %ptr, <vscale x 4 x i8> %offset, <vscale x 4 x i1> %m, i64 %vl) {
 ; CHECK-LABEL: test_vsoxei_mask:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v10
 ; CHECK-NEXT:    vsll.vi v12, v12, 4
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vsoxei64.v v8, (a0), v12, v0.t
 ; CHECK-NEXT:    ret
 entry:
@@ -308,10 +308,10 @@ declare void @llvm.riscv.vsuxei.nxv4i32.nxv4i64(
 define void @test_vsuxei(<vscale x 4 x i32> %val, ptr %ptr, <vscale x 4 x i8> %offset, i64 %vl) {
 ; CHECK-LABEL: test_vsuxei:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v10
 ; CHECK-NEXT:    vsll.vi v12, v12, 4
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vsuxei64.v v8, (a0), v12
 ; CHECK-NEXT:    ret
 entry:
@@ -335,10 +335,10 @@ declare void @llvm.riscv.vsuxei.mask.nxv4i32.nxv4i64(
 define void @test_vsuxei_mask(<vscale x 4 x i32> %val, ptr %ptr, <vscale x 4 x i8> %offset, <vscale x 4 x i1> %m, i64 %vl) {
 ; CHECK-LABEL: test_vsuxei_mask:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli a2, zero, e64, m4, ta, ma
+; CHECK-NEXT:    vsetvli zero, a1, e64, m4, ta, ma
 ; CHECK-NEXT:    vzext.vf8 v12, v10
 ; CHECK-NEXT:    vsll.vi v12, v12, 4
-; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vsuxei64.v v8, (a0), v12, v0.t
 ; CHECK-NEXT:    ret
 entry:
diff --git a/llvm/test/CodeGen/RISCV/rvv/pr63459.ll b/llvm/test/CodeGen/RISCV/rvv/pr63459.ll
index 5ef8e18bb2641..08a2e588330b5 100644
--- a/llvm/test/CodeGen/RISCV/rvv/pr63459.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/pr63459.ll
@@ -5,11 +5,10 @@ define void @snork(ptr %arg, <vscale x 2 x i64> %arg1) {
 ; CHECK-LABEL: snork:
 ; CHECK:       # %bb.0: # %bb
 ; CHECK-NEXT:    csrr a1, vlenb
-; CHECK-NEXT:    vsetvli a2, zero, e64, m2, ta, ma
+; CHECK-NEXT:    vsetivli zero, 4, e64, m2, ta, ma
 ; CHECK-NEXT:    vmul.vx v8, v8, a1
 ; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, ma
 ; CHECK-NEXT:    vmv.v.i v10, 1
-; CHECK-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
 ; CHECK-NEXT:    vsoxei64.v v10, (a0), v8
 ; CHECK-NEXT:    ret
 bb:
diff --git a/llvm/test/CodeGen/RISCV/rvv/vl-opt-op-info.mir b/llvm/test/CodeGen/RISCV/rvv/vl-opt-op-info.mir
index 8587ec136afd8..79772b2ff91e6 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vl-opt-op-info.mir
+++ b/llvm/test/CodeGen/RISCV/rvv/vl-opt-op-info.mir
@@ -573,6 +573,108 @@ body: |
     PseudoVSSE8_V_MF2 %x, $noreg, $noreg, 1, 3 /* e8 */
 ...
 ---
+name: vsuxeiN_v_data
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: vsuxeiN_v_data
+    ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, 1, 3 /* e8 */, 0 /* tu, mu */
+    ; CHECK-NEXT: PseudoVSUXEI8_V_M1_M1 %x, $noreg, $noreg, 1, 3 /* e8 */
+    %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0
+    PseudoVSUXEI8_V_M1_M1 %x, $noreg, $noreg, 1, 3 /* e8 */
+...
+---
+name: vsuxeiN_v_data_incompatible_eew
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: vsuxeiN_v_data_incompatible_eew
+    ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0 /* tu, mu */
+    ; CHECK-NEXT: PseudoVSUXEI8_V_M1_M1 %x, $noreg, $noreg, 1, 3 /* e8 */
+    %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0
+    PseudoVSUXEI8_V_M1_M1 %x, $noreg, $noreg, 1, 3 /* e8 */
+...
+---
+name: vsuxeiN_v_data_incompatible_emul
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: vsuxeiN_v_data_incompatible_emul
+    ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0 /* tu, mu */
+    ; CHECK-NEXT: PseudoVSUXEI8_V_MF2_MF2 %x, $noreg, $noreg, 1, 3 /* e8 */
+    %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0
+    PseudoVSUXEI8_V_MF2_MF2 %x, $noreg, $noreg, 1, 3 /* e8 */
+...
+---
+name: vsuxeiN_v_idx
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: vsuxeiN_v_idx
+    ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, 1, 3 /* e8 */, 0 /* tu, mu */
+    ; CHECK-NEXT: PseudoVSUXEI8_V_M1_M1 $noreg, $noreg, %x, 1, 3 /* e8 */
+    %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0
+    PseudoVSUXEI8_V_M1_M1 $noreg, $noreg, %x, 1, 3 /* e8 */
+...
+---
+name: vsuxeiN_v_idx_incompatible_eew
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: vsuxeiN_v_idx_incompatible_eew
+    ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0 /* tu, mu */
+    ; CHECK-NEXT: %y:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0 /* tu, mu */
+    ; CHECK-NEXT: PseudoVSUXEI8_V_M1_M1 $noreg, $noreg, %x, 1, 3 /* e8 */
+    %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0
+    %y:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0
+    PseudoVSUXEI8_V_M1_M1 $noreg, $noreg, %x, 1, 3 /* e8 */
+...
+---
+name: vsuxeiN_v_idx_incompatible_emul
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: vsuxeiN_v_idx_incompatible_emul
+    ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0 /* tu, mu */
+    ; CHECK-NEXT: PseudoVSUXEI8_V_MF2_MF2 $noreg, $noreg, %x, 1, 3 /* e8 */
+    %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0
+    PseudoVSUXEI8_V_MF2_MF2 $noreg, $noreg, %x, 1, 3 /* e8 */
+...
+---
+name: vluxeiN_v_data
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: vluxeiN_v_data
+    ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, 1, 3 /* e8 */, 0 /* tu, mu */
+    ; CHECK-NEXT: %y:vr = PseudoVLUXEI8_V_M1_M1 $noreg, $noreg, %x, 1, 3 /* e8 */, 0 /* tu, mu */
+    %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0
+    %y:vr = PseudoVLUXEI8_V_M1_M1 $noreg, $noreg, %x, 1, 3 /* e8 */, 0
+...
+---
+name: vluxeiN_v_incompatible_eew
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: vluxeiN_v_incompatible_eew
+    ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0 /* tu, mu */
+    ; CHECK-NEXT: %y:vr = PseudoVLUXEI8_V_M1_M1 $noreg, $noreg, %x, 1, 3 /* e8 */, 0 /* tu, mu */
+    %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 4 /* e16 */, 0
+    %y:vr = PseudoVLUXEI8_V_M1_M1 $noreg, $noreg, %x, 1, 3 /* e8 */, 0
+...
+---
+name: vluxeiN_v_data_incompatible_emul
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: vluxeiN_v_data_incompatible_emul
+    ; CHECK: %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0 /* tu, mu */
+    ; CHECK-NEXT: %y:vr = PseudoVLUXEI8_V_MF2_MF2 $noreg, $noreg, %x, 1, 3 /* e8 */, 0 /* tu, mu */
+    %x:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0
+    %y:vr = PseudoVLUXEI8_V_MF2_MF2 $noreg, $noreg, %x, 1, 3 /* e8 */, 0
+...
+---
+name: vluxeiN_v_idx
+body: |
+  bb.0:
+    ; CHECK-LABEL: name: vluxeiN_v_idx
+    ; CHECK: %x:vr = PseudoVADD_VV_MF2 $noreg, $noreg, $noreg, 1, 3 /* e8 */, 0 /* tu, mu */
+    ; CHECK-NEXT: early-clobber %y:vr = PseudoVLUXEI8_V_MF2_M1 $noreg, $noreg, %x, 1, 4 /* e16 */, 0 /* tu, mu */
+    %x:vr = PseudoVADD_VV_MF2 $noreg, $noreg, $noreg, -1, 3 /* e8 */, 0
+    %y:vr = PseudoVLUXEI8_V_MF2_M1 $noreg, $noreg, %x, 1, 4 /* e16 */, 0
+...
+---
 name: vmop_mm
 body: |
   bb.0:
diff --git a/llvm/test/CodeGen/RISCV/rvv/vpgather-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vpgather-sdnode.ll
index ca52ce6e2c4a1..abe7bdad8125a 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vpgather-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vpgather-sdnode.ll
@@ -235,17 +235,17 @@ define <vscale x 8 x i8> @vpgather_nxv8i8(<vscale x 8 x ptr> %ptrs, <vscale x 8
 define <vscale x 8 x i8> @vpgather_baseidx_nxv8i8(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i8:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
-; RV32-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e8, m1, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i8:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
-; RV64-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e8, m1, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i8, ptr %base, <vscale x 8 x i8> %idxs
@@ -264,22 +264,22 @@ define <vscale x 32 x i8> @vpgather_baseidx_nxv32i8(ptr %base, <vscale x 32 x i8
 ; RV32-NEXT:    slli a2, a3, 1
 ; RV32-NEXT:    srli a3, a3, 2
 ; RV32-NEXT:    sub a4, a1, a2
+; RV32-NEXT:    sltu a5, a1, a4
+; RV32-NEXT:    addi a5, a5, -1
+; RV32-NEXT:    and a4, a5, a4
 ; RV32-NEXT:    vslidedown.vx v0, v0, a3
-; RV32-NEXT:    sltu a3, a1, a4
-; RV32-NEXT:    addi a3, a3, -1
-; RV32-NEXT:    and a3, a3, a4
-; RV32-NEXT:    vsetvli a4, zero, e32, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, a4, e32, m8, ta, ma
 ; RV32-NEXT:    vsext.vf4 v16, v10
-; RV32-NEXT:    vsetvli zero, a3, e8, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e8, m2, ta, ma
 ; RV32-NEXT:    vluxei32.v v10, (a0), v16, v0.t
 ; RV32-NEXT:    bltu a1, a2, .LBB12_2
 ; RV32-NEXT:  # %bb.1:
 ; RV32-NEXT:    mv a1, a2
 ; RV32-NEXT:  .LBB12_2:
-; RV32-NEXT:    vsetvli a2, zero, e32, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m8, ta, ma
 ; RV32-NEXT:    vsext.vf4 v16, v8
 ; RV32-NEXT:    vmv1r.v v0, v12
-; RV32-NEXT:    vsetvli zero, a1, e8, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e8, m2, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
@@ -302,41 +302,41 @@ define <vscale x 32 x i8> @vpgather_baseidx_nxv32i8(ptr %base, <vscale x 32 x i8
 ; RV64-NEXT:    srli a4, a2, 3
 ; RV64-NEXT:    vsetvli a7, zero, e8, mf4, ta, ma
 ; RV64-NEXT:    vslidedown.vx v0, v13, a4
-; RV64-NEXT:    vsetvli a7, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a6, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v11
-; RV64-NEXT:    vsetvli zero, a6, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e8, m1, ta, ma
 ; RV64-NEXT:    vluxei64.v v11, (a0), v16, v0.t
 ; RV64-NEXT:    bltu a5, a2, .LBB12_2
 ; RV64-NEXT:  # %bb.1:
 ; RV64-NEXT:    mv a5, a2
 ; RV64-NEXT:  .LBB12_2:
-; RV64-NEXT:    vsetvli a6, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a5, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v10
 ; RV64-NEXT:    vmv1r.v v0, v13
-; RV64-NEXT:    vsetvli zero, a5, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e8, m1, ta, ma
 ; RV64-NEXT:    vluxei64.v v10, (a0), v16, v0.t
 ; RV64-NEXT:    bltu a1, a3, .LBB12_4
 ; RV64-NEXT:  # %bb.3:
 ; RV64-NEXT:    mv a1, a3
 ; RV64-NEXT:  .LBB12_4:
 ; RV64-NEXT:    sub a3, a1, a2
+; RV64-NEXT:    sltu a5, a1, a3
+; RV64-NEXT:    addi a5, a5, -1
+; RV64-NEXT:    and a3, a5, a3
 ; RV64-NEXT:    vsetvli a5, zero, e8, mf4, ta, ma
 ; RV64-NEXT:    vslidedown.vx v0, v12, a4
-; RV64-NEXT:    sltu a4, a1, a3
-; RV64-NEXT:    addi a4, a4, -1
-; RV64-NEXT:    and a3, a4, a3
-; RV64-NEXT:    vsetvli a4, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a3, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v9
-; RV64-NEXT:    vsetvli zero, a3, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e8, m1, ta, ma
 ; RV64-NEXT:    vluxei64.v v9, (a0), v16, v0.t
 ; RV64-NEXT:    bltu a1, a2, .LBB12_6
 ; RV64-NEXT:  # %bb.5:
 ; RV64-NEXT:    mv a1, a2
 ; RV64-NEXT:  .LBB12_6:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vmv1r.v v0, v12
-; RV64-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e8, m1, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i8, ptr %base, <vscale x 32 x i8> %idxs
@@ -529,19 +529,19 @@ define <vscale x 8 x i16> @vpgather_nxv8i16(<vscale x 8 x ptr> %ptrs, <vscale x
 define <vscale x 8 x i16> @vpgather_baseidx_nxv8i8_nxv8i16(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i8_nxv8i16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i8_nxv8i16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <vscale x 8 x i8> %idxs
@@ -552,19 +552,19 @@ define <vscale x 8 x i16> @vpgather_baseidx_nxv8i8_nxv8i16(ptr %base, <vscale x
 define <vscale x 8 x i16> @vpgather_baseidx_sext_nxv8i8_nxv8i16(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8i16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8i16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -576,17 +576,17 @@ define <vscale x 8 x i16> @vpgather_baseidx_sext_nxv8i8_nxv8i16(ptr %base, <vsca
 define <vscale x 8 x i16> @vpgather_baseidx_zext_nxv8i8_nxv8i16(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8i16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV32-NEXT:    vwaddu.vv v10, v8, v8
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vluxei16.v v8, (a0), v10, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8i16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV64-NEXT:    vwaddu.vv v10, v8, v8
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei16.v v8, (a0), v10, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -598,18 +598,17 @@ define <vscale x 8 x i16> @vpgather_baseidx_zext_nxv8i8_nxv8i16(ptr %base, <vsca
 define <vscale x 8 x i16> @vpgather_baseidx_nxv8i16(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
-; RV32-NEXT:    vwadd.vv v12, v8, v8
 ; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vwadd.vv v12, v8, v8
 ; RV32-NEXT:    vluxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <vscale x 8 x i16> %idxs
@@ -755,19 +754,18 @@ define <vscale x 8 x i32> @vpgather_nxv8i32(<vscale x 8 x ptr> %ptrs, <vscale x
 define <vscale x 8 x i32> @vpgather_baseidx_nxv8i8_nxv8i32(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i8_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vsll.vi v8, v12, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i8_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i32, ptr %base, <vscale x 8 x i8> %idxs
@@ -778,19 +776,18 @@ define <vscale x 8 x i32> @vpgather_baseidx_nxv8i8_nxv8i32(ptr %base, <vscale x
 define <vscale x 8 x i32> @vpgather_baseidx_sext_nxv8i8_nxv8i32(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vsll.vi v8, v12, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i32>
@@ -802,19 +799,19 @@ define <vscale x 8 x i32> @vpgather_baseidx_sext_nxv8i8_nxv8i32(ptr %base, <vsca
 define <vscale x 8 x i32> @vpgather_baseidx_zext_nxv8i8_nxv8i32(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV32-NEXT:    vzext.vf2 v10, v8
 ; RV32-NEXT:    vsll.vi v12, v10, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei16.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV64-NEXT:    vzext.vf2 v10, v8
 ; RV64-NEXT:    vsll.vi v12, v10, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei16.v v8, (a0), v12, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i32>
@@ -826,19 +823,18 @@ define <vscale x 8 x i32> @vpgather_baseidx_zext_nxv8i8_nxv8i32(ptr %base, <vsca
 define <vscale x 8 x i32> @vpgather_baseidx_nxv8i16_nxv8i32(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i16_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v8, v12, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i16_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i32, ptr %base, <vscale x 8 x i16> %idxs
@@ -849,19 +845,18 @@ define <vscale x 8 x i32> @vpgather_baseidx_nxv8i16_nxv8i32(ptr %base, <vscale x
 define <vscale x 8 x i32> @vpgather_baseidx_sext_nxv8i16_nxv8i32(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i16_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v8, v12, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i16_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i16> %idxs to <vscale x 8 x i32>
@@ -873,19 +868,17 @@ define <vscale x 8 x i32> @vpgather_baseidx_sext_nxv8i16_nxv8i32(ptr %base, <vsc
 define <vscale x 8 x i32> @vpgather_baseidx_zext_nxv8i16_nxv8i32(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i16_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vzext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v8, v12, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i16_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v12, v8
 ; RV64-NEXT:    vsll.vi v8, v12, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i16> %idxs to <vscale x 8 x i32>
@@ -897,18 +890,17 @@ define <vscale x 8 x i32> @vpgather_baseidx_zext_nxv8i16_nxv8i32(ptr %base, <vsc
 define <vscale x 8 x i32> @vpgather_baseidx_nxv8i32(ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
-; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i32, ptr %base, <vscale x 8 x i32> %idxs
@@ -1012,19 +1004,18 @@ define <vscale x 8 x i64> @vpgather_nxv8i64(<vscale x 8 x ptr> %ptrs, <vscale x
 define <vscale x 8 x i64> @vpgather_baseidx_nxv8i8_nxv8i64(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i8_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i8_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i64, ptr %base, <vscale x 8 x i8> %idxs
@@ -1035,19 +1026,18 @@ define <vscale x 8 x i64> @vpgather_baseidx_nxv8i8_nxv8i64(ptr %base, <vscale x
 define <vscale x 8 x i64> @vpgather_baseidx_sext_nxv8i8_nxv8i64(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i64>
@@ -1059,19 +1049,19 @@ define <vscale x 8 x i64> @vpgather_baseidx_sext_nxv8i8_nxv8i64(ptr %base, <vsca
 define <vscale x 8 x i64> @vpgather_baseidx_zext_nxv8i8_nxv8i64(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV32-NEXT:    vzext.vf2 v10, v8
 ; RV32-NEXT:    vsll.vi v16, v10, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei16.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV64-NEXT:    vzext.vf2 v10, v8
 ; RV64-NEXT:    vsll.vi v16, v10, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei16.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i64>
@@ -1083,19 +1073,18 @@ define <vscale x 8 x i64> @vpgather_baseidx_zext_nxv8i8_nxv8i64(ptr %base, <vsca
 define <vscale x 8 x i64> @vpgather_baseidx_nxv8i16_nxv8i64(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i16_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i16_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i64, ptr %base, <vscale x 8 x i16> %idxs
@@ -1106,19 +1095,18 @@ define <vscale x 8 x i64> @vpgather_baseidx_nxv8i16_nxv8i64(ptr %base, <vscale x
 define <vscale x 8 x i64> @vpgather_baseidx_sext_nxv8i16_nxv8i64(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i16_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i16_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i16> %idxs to <vscale x 8 x i64>
@@ -1130,19 +1118,19 @@ define <vscale x 8 x i64> @vpgather_baseidx_sext_nxv8i16_nxv8i64(ptr %base, <vsc
 define <vscale x 8 x i64> @vpgather_baseidx_zext_nxv8i16_nxv8i64(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i16_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vzext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i16_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v12, v8
 ; RV64-NEXT:    vsll.vi v16, v12, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i16> %idxs to <vscale x 8 x i64>
@@ -1154,18 +1142,17 @@ define <vscale x 8 x i64> @vpgather_baseidx_zext_nxv8i16_nxv8i64(ptr %base, <vsc
 define <vscale x 8 x i64> @vpgather_baseidx_nxv8i32_nxv8i64(ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i32_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v8, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i32_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i64, ptr %base, <vscale x 8 x i32> %idxs
@@ -1176,18 +1163,17 @@ define <vscale x 8 x i64> @vpgather_baseidx_nxv8i32_nxv8i64(ptr %base, <vscale x
 define <vscale x 8 x i64> @vpgather_baseidx_sext_nxv8i32_nxv8i64(ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i32_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v8, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i32_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i32> %idxs to <vscale x 8 x i64>
@@ -1199,18 +1185,17 @@ define <vscale x 8 x i64> @vpgather_baseidx_sext_nxv8i32_nxv8i64(ptr %base, <vsc
 define <vscale x 8 x i64> @vpgather_baseidx_zext_nxv8i32_nxv8i64(ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i32_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v8, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i32_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vzext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i32> %idxs to <vscale x 8 x i64>
@@ -1222,18 +1207,17 @@ define <vscale x 8 x i64> @vpgather_baseidx_zext_nxv8i32_nxv8i64(ptr %base, <vsc
 define <vscale x 8 x i64> @vpgather_baseidx_nxv8i64(ptr %base, <vscale x 8 x i64> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vnsrl.wi v16, v8, 0
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
-; RV64-NEXT:    vsll.vi v8, v8, 3
 ; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsll.vi v8, v8, 3
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i64, ptr %base, <vscale x 8 x i64> %idxs
@@ -1342,19 +1326,19 @@ define <vscale x 8 x bfloat> @vpgather_nxv8bf16(<vscale x 8 x ptr> %ptrs, <vscal
 define <vscale x 8 x bfloat> @vpgather_baseidx_nxv8i8_nxv8bf16(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i8_nxv8bf16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i8_nxv8bf16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds bfloat, ptr %base, <vscale x 8 x i8> %idxs
@@ -1365,19 +1349,19 @@ define <vscale x 8 x bfloat> @vpgather_baseidx_nxv8i8_nxv8bf16(ptr %base, <vscal
 define <vscale x 8 x bfloat> @vpgather_baseidx_sext_nxv8i8_nxv8bf16(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8bf16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8bf16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -1389,17 +1373,17 @@ define <vscale x 8 x bfloat> @vpgather_baseidx_sext_nxv8i8_nxv8bf16(ptr %base, <
 define <vscale x 8 x bfloat> @vpgather_baseidx_zext_nxv8i8_nxv8bf16(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8bf16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV32-NEXT:    vwaddu.vv v10, v8, v8
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vluxei16.v v8, (a0), v10, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8bf16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV64-NEXT:    vwaddu.vv v10, v8, v8
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei16.v v8, (a0), v10, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -1411,18 +1395,17 @@ define <vscale x 8 x bfloat> @vpgather_baseidx_zext_nxv8i8_nxv8bf16(ptr %base, <
 define <vscale x 8 x bfloat> @vpgather_baseidx_nxv8bf16(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8bf16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
-; RV32-NEXT:    vwadd.vv v12, v8, v8
 ; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vwadd.vv v12, v8, v8
 ; RV32-NEXT:    vluxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8bf16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds bfloat, ptr %base, <vscale x 8 x i16> %idxs
@@ -1531,19 +1514,19 @@ define <vscale x 8 x half> @vpgather_nxv8f16(<vscale x 8 x ptr> %ptrs, <vscale x
 define <vscale x 8 x half> @vpgather_baseidx_nxv8i8_nxv8f16(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i8_nxv8f16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i8_nxv8f16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds half, ptr %base, <vscale x 8 x i8> %idxs
@@ -1554,19 +1537,19 @@ define <vscale x 8 x half> @vpgather_baseidx_nxv8i8_nxv8f16(ptr %base, <vscale x
 define <vscale x 8 x half> @vpgather_baseidx_sext_nxv8i8_nxv8f16(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8f16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8f16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -1578,17 +1561,17 @@ define <vscale x 8 x half> @vpgather_baseidx_sext_nxv8i8_nxv8f16(ptr %base, <vsc
 define <vscale x 8 x half> @vpgather_baseidx_zext_nxv8i8_nxv8f16(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8f16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV32-NEXT:    vwaddu.vv v10, v8, v8
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vluxei16.v v8, (a0), v10, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8f16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV64-NEXT:    vwaddu.vv v10, v8, v8
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei16.v v8, (a0), v10, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -1600,18 +1583,17 @@ define <vscale x 8 x half> @vpgather_baseidx_zext_nxv8i8_nxv8f16(ptr %base, <vsc
 define <vscale x 8 x half> @vpgather_baseidx_nxv8f16(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8f16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
-; RV32-NEXT:    vwadd.vv v12, v8, v8
 ; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vwadd.vv v12, v8, v8
 ; RV32-NEXT:    vluxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8f16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds half, ptr %base, <vscale x 8 x i16> %idxs
@@ -1715,19 +1697,18 @@ define <vscale x 8 x float> @vpgather_nxv8f32(<vscale x 8 x ptr> %ptrs, <vscale
 define <vscale x 8 x float> @vpgather_baseidx_nxv8i8_nxv8f32(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i8_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vsll.vi v8, v12, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i8_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds float, ptr %base, <vscale x 8 x i8> %idxs
@@ -1738,19 +1719,18 @@ define <vscale x 8 x float> @vpgather_baseidx_nxv8i8_nxv8f32(ptr %base, <vscale
 define <vscale x 8 x float> @vpgather_baseidx_sext_nxv8i8_nxv8f32(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vsll.vi v8, v12, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i32>
@@ -1762,19 +1742,19 @@ define <vscale x 8 x float> @vpgather_baseidx_sext_nxv8i8_nxv8f32(ptr %base, <vs
 define <vscale x 8 x float> @vpgather_baseidx_zext_nxv8i8_nxv8f32(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV32-NEXT:    vzext.vf2 v10, v8
 ; RV32-NEXT:    vsll.vi v12, v10, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei16.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV64-NEXT:    vzext.vf2 v10, v8
 ; RV64-NEXT:    vsll.vi v12, v10, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei16.v v8, (a0), v12, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i32>
@@ -1786,19 +1766,18 @@ define <vscale x 8 x float> @vpgather_baseidx_zext_nxv8i8_nxv8f32(ptr %base, <vs
 define <vscale x 8 x float> @vpgather_baseidx_nxv8i16_nxv8f32(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i16_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v8, v12, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i16_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds float, ptr %base, <vscale x 8 x i16> %idxs
@@ -1809,19 +1788,18 @@ define <vscale x 8 x float> @vpgather_baseidx_nxv8i16_nxv8f32(ptr %base, <vscale
 define <vscale x 8 x float> @vpgather_baseidx_sext_nxv8i16_nxv8f32(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i16_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v8, v12, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i16_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i16> %idxs to <vscale x 8 x i32>
@@ -1833,19 +1811,17 @@ define <vscale x 8 x float> @vpgather_baseidx_sext_nxv8i16_nxv8f32(ptr %base, <v
 define <vscale x 8 x float> @vpgather_baseidx_zext_nxv8i16_nxv8f32(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i16_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vzext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v8, v12, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i16_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v12, v8
 ; RV64-NEXT:    vsll.vi v8, v12, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i16> %idxs to <vscale x 8 x i32>
@@ -1857,18 +1833,17 @@ define <vscale x 8 x float> @vpgather_baseidx_zext_nxv8i16_nxv8f32(ptr %base, <v
 define <vscale x 8 x float> @vpgather_baseidx_nxv8f32(ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
-; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV32-NEXT:    vsll.vi v8, v8, 2
 ; RV32-NEXT:    vluxei32.v v8, (a0), v8, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds float, ptr %base, <vscale x 8 x i32> %idxs
@@ -1972,19 +1947,18 @@ define <vscale x 6 x double> @vpgather_nxv6f64(<vscale x 6 x ptr> %ptrs, <vscale
 define <vscale x 6 x double> @vpgather_baseidx_nxv6i8_nxv6f64(ptr %base, <vscale x 6 x i8> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv6i8_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv6i8_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 6 x i8> %idxs
@@ -1995,19 +1969,18 @@ define <vscale x 6 x double> @vpgather_baseidx_nxv6i8_nxv6f64(ptr %base, <vscale
 define <vscale x 6 x double> @vpgather_baseidx_sext_nxv6i8_nxv6f64(ptr %base, <vscale x 6 x i8> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv6i8_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv6i8_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 6 x i8> %idxs to <vscale x 6 x i64>
@@ -2019,19 +1992,19 @@ define <vscale x 6 x double> @vpgather_baseidx_sext_nxv6i8_nxv6f64(ptr %base, <v
 define <vscale x 6 x double> @vpgather_baseidx_zext_nxv6i8_nxv6f64(ptr %base, <vscale x 6 x i8> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv6i8_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV32-NEXT:    vzext.vf2 v10, v8
 ; RV32-NEXT:    vsll.vi v16, v10, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei16.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv6i8_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV64-NEXT:    vzext.vf2 v10, v8
 ; RV64-NEXT:    vsll.vi v16, v10, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei16.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 6 x i8> %idxs to <vscale x 6 x i64>
@@ -2043,19 +2016,18 @@ define <vscale x 6 x double> @vpgather_baseidx_zext_nxv6i8_nxv6f64(ptr %base, <v
 define <vscale x 6 x double> @vpgather_baseidx_nxv6i16_nxv6f64(ptr %base, <vscale x 6 x i16> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv6i16_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv6i16_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 6 x i16> %idxs
@@ -2066,19 +2038,18 @@ define <vscale x 6 x double> @vpgather_baseidx_nxv6i16_nxv6f64(ptr %base, <vscal
 define <vscale x 6 x double> @vpgather_baseidx_sext_nxv6i16_nxv6f64(ptr %base, <vscale x 6 x i16> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv6i16_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv6i16_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 6 x i16> %idxs to <vscale x 6 x i64>
@@ -2090,19 +2061,19 @@ define <vscale x 6 x double> @vpgather_baseidx_sext_nxv6i16_nxv6f64(ptr %base, <
 define <vscale x 6 x double> @vpgather_baseidx_zext_nxv6i16_nxv6f64(ptr %base, <vscale x 6 x i16> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv6i16_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vzext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv6i16_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v12, v8
 ; RV64-NEXT:    vsll.vi v16, v12, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 6 x i16> %idxs to <vscale x 6 x i64>
@@ -2114,18 +2085,17 @@ define <vscale x 6 x double> @vpgather_baseidx_zext_nxv6i16_nxv6f64(ptr %base, <
 define <vscale x 6 x double> @vpgather_baseidx_nxv6i32_nxv6f64(ptr %base, <vscale x 6 x i32> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv6i32_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v8, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv6i32_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 6 x i32> %idxs
@@ -2136,18 +2106,17 @@ define <vscale x 6 x double> @vpgather_baseidx_nxv6i32_nxv6f64(ptr %base, <vscal
 define <vscale x 6 x double> @vpgather_baseidx_sext_nxv6i32_nxv6f64(ptr %base, <vscale x 6 x i32> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv6i32_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v8, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv6i32_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 6 x i32> %idxs to <vscale x 6 x i64>
@@ -2159,18 +2128,17 @@ define <vscale x 6 x double> @vpgather_baseidx_sext_nxv6i32_nxv6f64(ptr %base, <
 define <vscale x 6 x double> @vpgather_baseidx_zext_nxv6i32_nxv6f64(ptr %base, <vscale x 6 x i32> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv6i32_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v8, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv6i32_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vzext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 6 x i32> %idxs to <vscale x 6 x i64>
@@ -2182,18 +2150,17 @@ define <vscale x 6 x double> @vpgather_baseidx_zext_nxv6i32_nxv6f64(ptr %base, <
 define <vscale x 6 x double> @vpgather_baseidx_nxv6f64(ptr %base, <vscale x 6 x i64> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vnsrl.wi v16, v8, 0
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
-; RV64-NEXT:    vsll.vi v8, v8, 3
 ; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsll.vi v8, v8, 3
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 6 x i64> %idxs
@@ -2223,19 +2190,18 @@ define <vscale x 8 x double> @vpgather_nxv8f64(<vscale x 8 x ptr> %ptrs, <vscale
 define <vscale x 8 x double> @vpgather_baseidx_nxv8i8_nxv8f64(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i8_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i8_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 8 x i8> %idxs
@@ -2246,19 +2212,18 @@ define <vscale x 8 x double> @vpgather_baseidx_nxv8i8_nxv8f64(ptr %base, <vscale
 define <vscale x 8 x double> @vpgather_baseidx_sext_nxv8i8_nxv8f64(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i8_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i64>
@@ -2270,19 +2235,19 @@ define <vscale x 8 x double> @vpgather_baseidx_sext_nxv8i8_nxv8f64(ptr %base, <v
 define <vscale x 8 x double> @vpgather_baseidx_zext_nxv8i8_nxv8f64(ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV32-NEXT:    vzext.vf2 v10, v8
 ; RV32-NEXT:    vsll.vi v16, v10, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei16.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i8_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV64-NEXT:    vzext.vf2 v10, v8
 ; RV64-NEXT:    vsll.vi v16, v10, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei16.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i64>
@@ -2294,19 +2259,18 @@ define <vscale x 8 x double> @vpgather_baseidx_zext_nxv8i8_nxv8f64(ptr %base, <v
 define <vscale x 8 x double> @vpgather_baseidx_nxv8i16_nxv8f64(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i16_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i16_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 8 x i16> %idxs
@@ -2317,19 +2281,18 @@ define <vscale x 8 x double> @vpgather_baseidx_nxv8i16_nxv8f64(ptr %base, <vscal
 define <vscale x 8 x double> @vpgather_baseidx_sext_nxv8i16_nxv8f64(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i16_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i16_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i16> %idxs to <vscale x 8 x i64>
@@ -2341,19 +2304,19 @@ define <vscale x 8 x double> @vpgather_baseidx_sext_nxv8i16_nxv8f64(ptr %base, <
 define <vscale x 8 x double> @vpgather_baseidx_zext_nxv8i16_nxv8f64(ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i16_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vzext.vf2 v12, v8
 ; RV32-NEXT:    vsll.vi v16, v12, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i16_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v12, v8
 ; RV64-NEXT:    vsll.vi v16, v12, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i16> %idxs to <vscale x 8 x i64>
@@ -2365,18 +2328,17 @@ define <vscale x 8 x double> @vpgather_baseidx_zext_nxv8i16_nxv8f64(ptr %base, <
 define <vscale x 8 x double> @vpgather_baseidx_nxv8i32_nxv8f64(ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8i32_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v8, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8i32_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 8 x i32> %idxs
@@ -2387,18 +2349,17 @@ define <vscale x 8 x double> @vpgather_baseidx_nxv8i32_nxv8f64(ptr %base, <vscal
 define <vscale x 8 x double> @vpgather_baseidx_sext_nxv8i32_nxv8f64(ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_sext_nxv8i32_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v8, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_sext_nxv8i32_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i32> %idxs to <vscale x 8 x i64>
@@ -2410,18 +2371,17 @@ define <vscale x 8 x double> @vpgather_baseidx_sext_nxv8i32_nxv8f64(ptr %base, <
 define <vscale x 8 x double> @vpgather_baseidx_zext_nxv8i32_nxv8f64(ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_zext_nxv8i32_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v8, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_zext_nxv8i32_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vzext.vf2 v16, v8
 ; RV64-NEXT:    vsll.vi v8, v16, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i32> %idxs to <vscale x 8 x i64>
@@ -2433,18 +2393,17 @@ define <vscale x 8 x double> @vpgather_baseidx_zext_nxv8i32_nxv8f64(ptr %base, <
 define <vscale x 8 x double> @vpgather_baseidx_nxv8f64(ptr %base, <vscale x 8 x i64> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpgather_baseidx_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vnsrl.wi v16, v8, 0
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vluxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpgather_baseidx_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
-; RV64-NEXT:    vsll.vi v8, v8, 3
 ; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsll.vi v8, v8, 3
 ; RV64-NEXT:    vluxei64.v v8, (a0), v8, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 8 x i64> %idxs
diff --git a/llvm/test/CodeGen/RISCV/rvv/vpscatter-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vpscatter-sdnode.ll
index 6f74d8cd7e01a..647e3965b7ec2 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vpscatter-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vpscatter-sdnode.ll
@@ -47,17 +47,15 @@ define void @vpscatter_nxv2i8(<vscale x 2 x i8> %val, <vscale x 2 x ptr> %ptrs,
 define void @vpscatter_nxv2i16_truncstore_nxv2i8(<vscale x 2 x i16> %val, <vscale x 2 x ptr> %ptrs, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_nxv2i16_truncstore_nxv2i8:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a1, zero, e8, mf4, ta, ma
-; RV32-NEXT:    vnsrl.wi v8, v8, 0
 ; RV32-NEXT:    vsetvli zero, a0, e8, mf4, ta, ma
+; RV32-NEXT:    vnsrl.wi v8, v8, 0
 ; RV32-NEXT:    vsoxei32.v v8, (zero), v9, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_nxv2i16_truncstore_nxv2i8:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a1, zero, e8, mf4, ta, ma
-; RV64-NEXT:    vnsrl.wi v8, v8, 0
 ; RV64-NEXT:    vsetvli zero, a0, e8, mf4, ta, ma
+; RV64-NEXT:    vnsrl.wi v8, v8, 0
 ; RV64-NEXT:    vsoxei64.v v8, (zero), v10, v0.t
 ; RV64-NEXT:    ret
   %tval = trunc <vscale x 2 x i16> %val to <vscale x 2 x i8>
@@ -68,21 +66,19 @@ define void @vpscatter_nxv2i16_truncstore_nxv2i8(<vscale x 2 x i16> %val, <vscal
 define void @vpscatter_nxv2i32_truncstore_nxv2i8(<vscale x 2 x i32> %val, <vscale x 2 x ptr> %ptrs, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_nxv2i32_truncstore_nxv2i8:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
+; RV32-NEXT:    vsetvli zero, a0, e16, mf2, ta, ma
 ; RV32-NEXT:    vnsrl.wi v8, v8, 0
 ; RV32-NEXT:    vsetvli zero, zero, e8, mf4, ta, ma
 ; RV32-NEXT:    vnsrl.wi v8, v8, 0
-; RV32-NEXT:    vsetvli zero, a0, e8, mf4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (zero), v9, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_nxv2i32_truncstore_nxv2i8:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
+; RV64-NEXT:    vsetvli zero, a0, e16, mf2, ta, ma
 ; RV64-NEXT:    vnsrl.wi v8, v8, 0
 ; RV64-NEXT:    vsetvli zero, zero, e8, mf4, ta, ma
 ; RV64-NEXT:    vnsrl.wi v8, v8, 0
-; RV64-NEXT:    vsetvli zero, a0, e8, mf4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (zero), v10, v0.t
 ; RV64-NEXT:    ret
   %tval = trunc <vscale x 2 x i32> %val to <vscale x 2 x i8>
@@ -93,25 +89,23 @@ define void @vpscatter_nxv2i32_truncstore_nxv2i8(<vscale x 2 x i32> %val, <vscal
 define void @vpscatter_nxv2i64_truncstore_nxv2i8(<vscale x 2 x i64> %val, <vscale x 2 x ptr> %ptrs, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_nxv2i64_truncstore_nxv2i8:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; RV32-NEXT:    vsetvli zero, a0, e32, m1, ta, ma
 ; RV32-NEXT:    vnsrl.wi v11, v8, 0
 ; RV32-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
 ; RV32-NEXT:    vnsrl.wi v8, v11, 0
 ; RV32-NEXT:    vsetvli zero, zero, e8, mf4, ta, ma
 ; RV32-NEXT:    vnsrl.wi v8, v8, 0
-; RV32-NEXT:    vsetvli zero, a0, e8, mf4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (zero), v10, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_nxv2i64_truncstore_nxv2i8:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, a0, e32, m1, ta, ma
 ; RV64-NEXT:    vnsrl.wi v12, v8, 0
 ; RV64-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
 ; RV64-NEXT:    vnsrl.wi v8, v12, 0
 ; RV64-NEXT:    vsetvli zero, zero, e8, mf4, ta, ma
 ; RV64-NEXT:    vnsrl.wi v8, v8, 0
-; RV64-NEXT:    vsetvli zero, a0, e8, mf4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (zero), v10, v0.t
 ; RV64-NEXT:    ret
   %tval = trunc <vscale x 2 x i64> %val to <vscale x 2 x i8>
@@ -174,17 +168,17 @@ define void @vpscatter_nxv8i8(<vscale x 8 x i8> %val, <vscale x 8 x ptr> %ptrs,
 define void @vpscatter_baseidx_nxv8i8(<vscale x 8 x i8> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i8:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v9
-; RV32-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e8, m1, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i8:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v9
-; RV64-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e8, m1, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i8, ptr %base, <vscale x 8 x i8> %idxs
@@ -231,17 +225,15 @@ define void @vpscatter_nxv2i16(<vscale x 2 x i16> %val, <vscale x 2 x ptr> %ptrs
 define void @vpscatter_nxv2i32_truncstore_nxv2i16(<vscale x 2 x i32> %val, <vscale x 2 x ptr> %ptrs, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_nxv2i32_truncstore_nxv2i16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
-; RV32-NEXT:    vnsrl.wi v8, v8, 0
 ; RV32-NEXT:    vsetvli zero, a0, e16, mf2, ta, ma
+; RV32-NEXT:    vnsrl.wi v8, v8, 0
 ; RV32-NEXT:    vsoxei32.v v8, (zero), v9, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_nxv2i32_truncstore_nxv2i16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
-; RV64-NEXT:    vnsrl.wi v8, v8, 0
 ; RV64-NEXT:    vsetvli zero, a0, e16, mf2, ta, ma
+; RV64-NEXT:    vnsrl.wi v8, v8, 0
 ; RV64-NEXT:    vsoxei64.v v8, (zero), v10, v0.t
 ; RV64-NEXT:    ret
   %tval = trunc <vscale x 2 x i32> %val to <vscale x 2 x i16>
@@ -252,21 +244,19 @@ define void @vpscatter_nxv2i32_truncstore_nxv2i16(<vscale x 2 x i32> %val, <vsca
 define void @vpscatter_nxv2i64_truncstore_nxv2i16(<vscale x 2 x i64> %val, <vscale x 2 x ptr> %ptrs, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_nxv2i64_truncstore_nxv2i16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; RV32-NEXT:    vsetvli zero, a0, e32, m1, ta, ma
 ; RV32-NEXT:    vnsrl.wi v11, v8, 0
 ; RV32-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
 ; RV32-NEXT:    vnsrl.wi v8, v11, 0
-; RV32-NEXT:    vsetvli zero, a0, e16, mf2, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (zero), v10, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_nxv2i64_truncstore_nxv2i16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, a0, e32, m1, ta, ma
 ; RV64-NEXT:    vnsrl.wi v12, v8, 0
 ; RV64-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
 ; RV64-NEXT:    vnsrl.wi v8, v12, 0
-; RV64-NEXT:    vsetvli zero, a0, e16, mf2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (zero), v10, v0.t
 ; RV64-NEXT:    ret
   %tval = trunc <vscale x 2 x i64> %val to <vscale x 2 x i16>
@@ -329,19 +319,19 @@ define void @vpscatter_nxv8i16(<vscale x 8 x i16> %val, <vscale x 8 x ptr> %ptrs
 define void @vpscatter_baseidx_nxv8i8_nxv8i16(<vscale x 8 x i16> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i8_nxv8i16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v10
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i8_nxv8i16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v10
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <vscale x 8 x i8> %idxs
@@ -352,19 +342,19 @@ define void @vpscatter_baseidx_nxv8i8_nxv8i16(<vscale x 8 x i16> %val, ptr %base
 define void @vpscatter_baseidx_sext_nxv8i8_nxv8i16(<vscale x 8 x i16> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8i16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v10
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8i16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v10
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -376,17 +366,17 @@ define void @vpscatter_baseidx_sext_nxv8i8_nxv8i16(<vscale x 8 x i16> %val, ptr
 define void @vpscatter_baseidx_zext_nxv8i8_nxv8i16(<vscale x 8 x i16> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8i16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV32-NEXT:    vwaddu.vv v12, v10, v10
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei16.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8i16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV64-NEXT:    vwaddu.vv v12, v10, v10
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei16.v v8, (a0), v12, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -398,18 +388,17 @@ define void @vpscatter_baseidx_zext_nxv8i8_nxv8i16(<vscale x 8 x i16> %val, ptr
 define void @vpscatter_baseidx_nxv8i16(<vscale x 8 x i16> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
-; RV32-NEXT:    vwadd.vv v12, v10, v10
 ; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vwadd.vv v12, v10, v10
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v10
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <vscale x 8 x i16> %idxs
@@ -430,9 +419,8 @@ define void @vpscatter_baseidx_vpsext_nxv8i16_nxv8i16(<vscale x 8 x i16> %val, p
 ; RV64:       # %bb.0:
 ; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vsext.vf2 v12, v10, v0.t
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vwadd.vv v16, v12, v12
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = call <vscale x 8 x i32> @llvm.vp.sext.nxv8i16.nxv8i32(<vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 %evl)
@@ -454,9 +442,8 @@ define void @vpscatter_baseidx_vpzext_nxv8i16_nxv8i16(<vscale x 8 x i16> %val, p
 ; RV64:       # %bb.0:
 ; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v12, v10, v0.t
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vwadd.vv v16, v12, v12
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = call <vscale x 8 x i32> @llvm.vp.zext.nxv8i16.nxv8i32(<vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 %evl)
@@ -471,10 +458,10 @@ define void @vpscatter_baseidx_vpsext_nxv8i32_nxv8i16(<vscale x 8 x i16> %val, p
 ; RV32:       # %bb.0:
 ; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV32-NEXT:    vsext.vf2 v16, v12, v0.t
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV32-NEXT:    vnsrl.wi v12, v16, 0
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
@@ -497,10 +484,10 @@ define void @vpscatter_baseidx_vpzext_nxv8i32_nxv8i16(<vscale x 8 x i16> %val, p
 ; RV32:       # %bb.0:
 ; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV32-NEXT:    vzext.vf2 v16, v12, v0.t
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV32-NEXT:    vnsrl.wi v12, v16, 0
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
@@ -556,17 +543,15 @@ define void @vpscatter_nxv2i32(<vscale x 2 x i32> %val, <vscale x 2 x ptr> %ptrs
 define void @vpscatter_nxv2i64_truncstore_nxv2i32(<vscale x 2 x i64> %val, <vscale x 2 x ptr> %ptrs, <vscale x 2 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_nxv2i64_truncstore_nxv2i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
-; RV32-NEXT:    vnsrl.wi v11, v8, 0
 ; RV32-NEXT:    vsetvli zero, a0, e32, m1, ta, ma
+; RV32-NEXT:    vnsrl.wi v11, v8, 0
 ; RV32-NEXT:    vsoxei32.v v11, (zero), v10, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_nxv2i64_truncstore_nxv2i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
-; RV64-NEXT:    vnsrl.wi v12, v8, 0
 ; RV64-NEXT:    vsetvli zero, a0, e32, m1, ta, ma
+; RV64-NEXT:    vnsrl.wi v12, v8, 0
 ; RV64-NEXT:    vsoxei64.v v12, (zero), v10, v0.t
 ; RV64-NEXT:    ret
   %tval = trunc <vscale x 2 x i64> %val to <vscale x 2 x i32>
@@ -629,19 +614,18 @@ define void @vpscatter_nxv8i32(<vscale x 8 x i32> %val, <vscale x 8 x ptr> %ptrs
 define void @vpscatter_baseidx_nxv8i8_nxv8i32(<vscale x 8 x i32> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i8_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v16, v12
 ; RV32-NEXT:    vsll.vi v12, v16, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i8_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v12
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i32, ptr %base, <vscale x 8 x i8> %idxs
@@ -652,19 +636,18 @@ define void @vpscatter_baseidx_nxv8i8_nxv8i32(<vscale x 8 x i32> %val, ptr %base
 define void @vpscatter_baseidx_sext_nxv8i8_nxv8i32(<vscale x 8 x i32> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v16, v12
 ; RV32-NEXT:    vsll.vi v12, v16, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v12
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i32>
@@ -676,19 +659,19 @@ define void @vpscatter_baseidx_sext_nxv8i8_nxv8i32(<vscale x 8 x i32> %val, ptr
 define void @vpscatter_baseidx_zext_nxv8i8_nxv8i32(<vscale x 8 x i32> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV32-NEXT:    vzext.vf2 v14, v12
 ; RV32-NEXT:    vsll.vi v12, v14, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei16.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV64-NEXT:    vzext.vf2 v14, v12
 ; RV64-NEXT:    vsll.vi v12, v14, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei16.v v8, (a0), v12, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i32>
@@ -700,19 +683,18 @@ define void @vpscatter_baseidx_zext_nxv8i8_nxv8i32(<vscale x 8 x i32> %val, ptr
 define void @vpscatter_baseidx_nxv8i16_nxv8i32(<vscale x 8 x i32> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i16_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v16, v12
 ; RV32-NEXT:    vsll.vi v12, v16, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i16_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v12
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i32, ptr %base, <vscale x 8 x i16> %idxs
@@ -723,19 +705,18 @@ define void @vpscatter_baseidx_nxv8i16_nxv8i32(<vscale x 8 x i32> %val, ptr %bas
 define void @vpscatter_baseidx_sext_nxv8i16_nxv8i32(<vscale x 8 x i32> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i16_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v16, v12
 ; RV32-NEXT:    vsll.vi v12, v16, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i16_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v12
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i16> %idxs to <vscale x 8 x i32>
@@ -747,19 +728,17 @@ define void @vpscatter_baseidx_sext_nxv8i16_nxv8i32(<vscale x 8 x i32> %val, ptr
 define void @vpscatter_baseidx_zext_nxv8i16_nxv8i32(<vscale x 8 x i32> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i16_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vzext.vf2 v16, v12
 ; RV32-NEXT:    vsll.vi v12, v16, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i16_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v16, v12
 ; RV64-NEXT:    vsll.vi v12, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i16> %idxs to <vscale x 8 x i32>
@@ -771,18 +750,17 @@ define void @vpscatter_baseidx_zext_nxv8i16_nxv8i32(<vscale x 8 x i32> %val, ptr
 define void @vpscatter_baseidx_nxv8i32(<vscale x 8 x i32> %val, ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
-; RV32-NEXT:    vsll.vi v12, v12, 2
 ; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV32-NEXT:    vsll.vi v12, v12, 2
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v16, v12
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i32, ptr %base, <vscale x 8 x i32> %idxs
@@ -881,19 +859,18 @@ define void @vpscatter_nxv8i64(<vscale x 8 x i64> %val, <vscale x 8 x ptr> %ptrs
 define void @vpscatter_baseidx_nxv8i8_nxv8i64(<vscale x 8 x i64> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i8_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i8_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i64, ptr %base, <vscale x 8 x i8> %idxs
@@ -904,19 +881,18 @@ define void @vpscatter_baseidx_nxv8i8_nxv8i64(<vscale x 8 x i64> %val, ptr %base
 define void @vpscatter_baseidx_sext_nxv8i8_nxv8i64(<vscale x 8 x i64> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i64>
@@ -928,19 +904,19 @@ define void @vpscatter_baseidx_sext_nxv8i8_nxv8i64(<vscale x 8 x i64> %val, ptr
 define void @vpscatter_baseidx_zext_nxv8i8_nxv8i64(<vscale x 8 x i64> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV32-NEXT:    vzext.vf2 v18, v16
 ; RV32-NEXT:    vsll.vi v16, v18, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei16.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV64-NEXT:    vzext.vf2 v18, v16
 ; RV64-NEXT:    vsll.vi v16, v18, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei16.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i64>
@@ -952,19 +928,18 @@ define void @vpscatter_baseidx_zext_nxv8i8_nxv8i64(<vscale x 8 x i64> %val, ptr
 define void @vpscatter_baseidx_nxv8i16_nxv8i64(<vscale x 8 x i64> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i16_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i16_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i64, ptr %base, <vscale x 8 x i16> %idxs
@@ -975,19 +950,18 @@ define void @vpscatter_baseidx_nxv8i16_nxv8i64(<vscale x 8 x i64> %val, ptr %bas
 define void @vpscatter_baseidx_sext_nxv8i16_nxv8i64(<vscale x 8 x i64> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i16_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i16_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i16> %idxs to <vscale x 8 x i64>
@@ -999,19 +973,19 @@ define void @vpscatter_baseidx_sext_nxv8i16_nxv8i64(<vscale x 8 x i64> %val, ptr
 define void @vpscatter_baseidx_zext_nxv8i16_nxv8i64(<vscale x 8 x i64> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i16_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vzext.vf2 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i16_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v20, v16
 ; RV64-NEXT:    vsll.vi v16, v20, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i16> %idxs to <vscale x 8 x i64>
@@ -1023,18 +997,17 @@ define void @vpscatter_baseidx_zext_nxv8i16_nxv8i64(<vscale x 8 x i64> %val, ptr
 define void @vpscatter_baseidx_nxv8i32_nxv8i64(<vscale x 8 x i64> %val, ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i32_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i32_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i64, ptr %base, <vscale x 8 x i32> %idxs
@@ -1045,18 +1018,17 @@ define void @vpscatter_baseidx_nxv8i32_nxv8i64(<vscale x 8 x i64> %val, ptr %bas
 define void @vpscatter_baseidx_sext_nxv8i32_nxv8i64(<vscale x 8 x i64> %val, ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i32_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i32_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i32> %idxs to <vscale x 8 x i64>
@@ -1068,18 +1040,17 @@ define void @vpscatter_baseidx_sext_nxv8i32_nxv8i64(<vscale x 8 x i64> %val, ptr
 define void @vpscatter_baseidx_zext_nxv8i32_nxv8i64(<vscale x 8 x i64> %val, ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i32_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i32_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vzext.vf2 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i32> %idxs to <vscale x 8 x i64>
@@ -1091,18 +1062,17 @@ define void @vpscatter_baseidx_zext_nxv8i32_nxv8i64(<vscale x 8 x i64> %val, ptr
 define void @vpscatter_baseidx_nxv8i64(<vscale x 8 x i64> %val, ptr %base, <vscale x 8 x i64> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vnsrl.wi v24, v16, 0
 ; RV32-NEXT:    vsll.vi v16, v24, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
-; RV64-NEXT:    vsll.vi v16, v16, 3
 ; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsll.vi v16, v16, 3
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds i64, ptr %base, <vscale x 8 x i64> %idxs
@@ -1201,19 +1171,19 @@ define void @vpscatter_nxv8bf16(<vscale x 8 x bfloat> %val, <vscale x 8 x ptr> %
 define void @vpscatter_baseidx_nxv8i8_nxv8bf16(<vscale x 8 x bfloat> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i8_nxv8bf16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v10
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i8_nxv8bf16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v10
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds bfloat, ptr %base, <vscale x 8 x i8> %idxs
@@ -1224,19 +1194,19 @@ define void @vpscatter_baseidx_nxv8i8_nxv8bf16(<vscale x 8 x bfloat> %val, ptr %
 define void @vpscatter_baseidx_sext_nxv8i8_nxv8bf16(<vscale x 8 x bfloat> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8bf16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v10
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8bf16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v10
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -1248,17 +1218,17 @@ define void @vpscatter_baseidx_sext_nxv8i8_nxv8bf16(<vscale x 8 x bfloat> %val,
 define void @vpscatter_baseidx_zext_nxv8i8_nxv8bf16(<vscale x 8 x bfloat> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8bf16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV32-NEXT:    vwaddu.vv v12, v10, v10
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei16.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8bf16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV64-NEXT:    vwaddu.vv v12, v10, v10
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei16.v v8, (a0), v12, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -1270,18 +1240,17 @@ define void @vpscatter_baseidx_zext_nxv8i8_nxv8bf16(<vscale x 8 x bfloat> %val,
 define void @vpscatter_baseidx_nxv8bf16(<vscale x 8 x bfloat> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8bf16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
-; RV32-NEXT:    vwadd.vv v12, v10, v10
 ; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vwadd.vv v12, v10, v10
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8bf16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v10
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds bfloat, ptr %base, <vscale x 8 x i16> %idxs
@@ -1380,19 +1349,19 @@ define void @vpscatter_nxv8f16(<vscale x 8 x half> %val, <vscale x 8 x ptr> %ptr
 define void @vpscatter_baseidx_nxv8i8_nxv8f16(<vscale x 8 x half> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i8_nxv8f16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v10
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i8_nxv8f16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v10
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds half, ptr %base, <vscale x 8 x i8> %idxs
@@ -1403,19 +1372,19 @@ define void @vpscatter_baseidx_nxv8i8_nxv8f16(<vscale x 8 x half> %val, ptr %bas
 define void @vpscatter_baseidx_sext_nxv8i8_nxv8f16(<vscale x 8 x half> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8f16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v12, v10
 ; RV32-NEXT:    vadd.vv v12, v12, v12
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8f16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v10
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -1427,17 +1396,17 @@ define void @vpscatter_baseidx_sext_nxv8i8_nxv8f16(<vscale x 8 x half> %val, ptr
 define void @vpscatter_baseidx_zext_nxv8i8_nxv8f16(<vscale x 8 x half> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8f16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV32-NEXT:    vwaddu.vv v12, v10, v10
-; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV32-NEXT:    vsoxei16.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8f16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e8, m1, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e8, m1, ta, ma
 ; RV64-NEXT:    vwaddu.vv v12, v10, v10
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei16.v v8, (a0), v12, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i16>
@@ -1449,18 +1418,17 @@ define void @vpscatter_baseidx_zext_nxv8i8_nxv8f16(<vscale x 8 x half> %val, ptr
 define void @vpscatter_baseidx_nxv8f16(<vscale x 8 x half> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8f16:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
-; RV32-NEXT:    vwadd.vv v12, v10, v10
 ; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV32-NEXT:    vwadd.vv v12, v10, v10
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8f16:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v10
 ; RV64-NEXT:    vadd.vv v16, v16, v16
-; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds half, ptr %base, <vscale x 8 x i16> %idxs
@@ -1559,19 +1527,18 @@ define void @vpscatter_nxv8f32(<vscale x 8 x float> %val, <vscale x 8 x ptr> %pt
 define void @vpscatter_baseidx_nxv8i8_nxv8f32(<vscale x 8 x float> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i8_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v16, v12
 ; RV32-NEXT:    vsll.vi v12, v16, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i8_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v12
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds float, ptr %base, <vscale x 8 x i8> %idxs
@@ -1582,19 +1549,18 @@ define void @vpscatter_baseidx_nxv8i8_nxv8f32(<vscale x 8 x float> %val, ptr %ba
 define void @vpscatter_baseidx_sext_nxv8i8_nxv8f32(<vscale x 8 x float> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v16, v12
 ; RV32-NEXT:    vsll.vi v12, v16, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v16, v12
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i32>
@@ -1606,19 +1572,19 @@ define void @vpscatter_baseidx_sext_nxv8i8_nxv8f32(<vscale x 8 x float> %val, pt
 define void @vpscatter_baseidx_zext_nxv8i8_nxv8f32(<vscale x 8 x float> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV32-NEXT:    vzext.vf2 v14, v12
 ; RV32-NEXT:    vsll.vi v12, v14, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei16.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV64-NEXT:    vzext.vf2 v14, v12
 ; RV64-NEXT:    vsll.vi v12, v14, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei16.v v8, (a0), v12, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i32>
@@ -1630,19 +1596,18 @@ define void @vpscatter_baseidx_zext_nxv8i8_nxv8f32(<vscale x 8 x float> %val, pt
 define void @vpscatter_baseidx_nxv8i16_nxv8f32(<vscale x 8 x float> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i16_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v16, v12
 ; RV32-NEXT:    vsll.vi v12, v16, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i16_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v12
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds float, ptr %base, <vscale x 8 x i16> %idxs
@@ -1653,19 +1618,18 @@ define void @vpscatter_baseidx_nxv8i16_nxv8f32(<vscale x 8 x float> %val, ptr %b
 define void @vpscatter_baseidx_sext_nxv8i16_nxv8f32(<vscale x 8 x float> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i16_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v16, v12
 ; RV32-NEXT:    vsll.vi v12, v16, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i16_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v16, v12
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i16> %idxs to <vscale x 8 x i32>
@@ -1677,19 +1641,17 @@ define void @vpscatter_baseidx_sext_nxv8i16_nxv8f32(<vscale x 8 x float> %val, p
 define void @vpscatter_baseidx_zext_nxv8i16_nxv8f32(<vscale x 8 x float> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i16_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vzext.vf2 v16, v12
 ; RV32-NEXT:    vsll.vi v12, v16, 2
-; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i16_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v16, v12
 ; RV64-NEXT:    vsll.vi v12, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i16> %idxs to <vscale x 8 x i32>
@@ -1701,18 +1663,17 @@ define void @vpscatter_baseidx_zext_nxv8i16_nxv8f32(<vscale x 8 x float> %val, p
 define void @vpscatter_baseidx_nxv8f32(<vscale x 8 x float> %val, ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8f32:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
-; RV32-NEXT:    vsll.vi v12, v12, 2
 ; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV32-NEXT:    vsll.vi v12, v12, 2
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v12, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8f32:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v16, v12
 ; RV64-NEXT:    vsll.vi v16, v16, 2
-; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds float, ptr %base, <vscale x 8 x i32> %idxs
@@ -1811,19 +1772,18 @@ define void @vpscatter_nxv6f64(<vscale x 6 x double> %val, <vscale x 6 x ptr> %p
 define void @vpscatter_baseidx_nxv6i8_nxv6f64(<vscale x 6 x double> %val, ptr %base, <vscale x 6 x i8> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv6i8_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv6i8_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 6 x i8> %idxs
@@ -1834,19 +1794,18 @@ define void @vpscatter_baseidx_nxv6i8_nxv6f64(<vscale x 6 x double> %val, ptr %b
 define void @vpscatter_baseidx_sext_nxv6i8_nxv6f64(<vscale x 6 x double> %val, ptr %base, <vscale x 6 x i8> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv6i8_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv6i8_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 6 x i8> %idxs to <vscale x 6 x i64>
@@ -1858,19 +1817,19 @@ define void @vpscatter_baseidx_sext_nxv6i8_nxv6f64(<vscale x 6 x double> %val, p
 define void @vpscatter_baseidx_zext_nxv6i8_nxv6f64(<vscale x 6 x double> %val, ptr %base, <vscale x 6 x i8> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv6i8_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV32-NEXT:    vzext.vf2 v18, v16
 ; RV32-NEXT:    vsll.vi v16, v18, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei16.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv6i8_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV64-NEXT:    vzext.vf2 v18, v16
 ; RV64-NEXT:    vsll.vi v16, v18, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei16.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 6 x i8> %idxs to <vscale x 6 x i64>
@@ -1882,19 +1841,18 @@ define void @vpscatter_baseidx_zext_nxv6i8_nxv6f64(<vscale x 6 x double> %val, p
 define void @vpscatter_baseidx_nxv6i16_nxv6f64(<vscale x 6 x double> %val, ptr %base, <vscale x 6 x i16> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv6i16_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv6i16_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 6 x i16> %idxs
@@ -1905,19 +1863,18 @@ define void @vpscatter_baseidx_nxv6i16_nxv6f64(<vscale x 6 x double> %val, ptr %
 define void @vpscatter_baseidx_sext_nxv6i16_nxv6f64(<vscale x 6 x double> %val, ptr %base, <vscale x 6 x i16> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv6i16_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv6i16_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 6 x i16> %idxs to <vscale x 6 x i64>
@@ -1929,19 +1886,19 @@ define void @vpscatter_baseidx_sext_nxv6i16_nxv6f64(<vscale x 6 x double> %val,
 define void @vpscatter_baseidx_zext_nxv6i16_nxv6f64(<vscale x 6 x double> %val, ptr %base, <vscale x 6 x i16> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv6i16_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vzext.vf2 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv6i16_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v20, v16
 ; RV64-NEXT:    vsll.vi v16, v20, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 6 x i16> %idxs to <vscale x 6 x i64>
@@ -1953,18 +1910,17 @@ define void @vpscatter_baseidx_zext_nxv6i16_nxv6f64(<vscale x 6 x double> %val,
 define void @vpscatter_baseidx_nxv6i32_nxv6f64(<vscale x 6 x double> %val, ptr %base, <vscale x 6 x i32> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv6i32_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv6i32_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 6 x i32> %idxs
@@ -1975,18 +1931,17 @@ define void @vpscatter_baseidx_nxv6i32_nxv6f64(<vscale x 6 x double> %val, ptr %
 define void @vpscatter_baseidx_sext_nxv6i32_nxv6f64(<vscale x 6 x double> %val, ptr %base, <vscale x 6 x i32> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv6i32_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv6i32_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 6 x i32> %idxs to <vscale x 6 x i64>
@@ -1998,18 +1953,17 @@ define void @vpscatter_baseidx_sext_nxv6i32_nxv6f64(<vscale x 6 x double> %val,
 define void @vpscatter_baseidx_zext_nxv6i32_nxv6f64(<vscale x 6 x double> %val, ptr %base, <vscale x 6 x i32> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv6i32_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv6i32_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vzext.vf2 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 6 x i32> %idxs to <vscale x 6 x i64>
@@ -2021,18 +1975,17 @@ define void @vpscatter_baseidx_zext_nxv6i32_nxv6f64(<vscale x 6 x double> %val,
 define void @vpscatter_baseidx_nxv6f64(<vscale x 6 x double> %val, ptr %base, <vscale x 6 x i64> %idxs, <vscale x 6 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv6f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vnsrl.wi v24, v16, 0
 ; RV32-NEXT:    vsll.vi v16, v24, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv6f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
-; RV64-NEXT:    vsll.vi v16, v16, 3
 ; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsll.vi v16, v16, 3
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 6 x i64> %idxs
@@ -2061,19 +2014,18 @@ define void @vpscatter_nxv8f64(<vscale x 8 x double> %val, <vscale x 8 x ptr> %p
 define void @vpscatter_baseidx_nxv8i8_nxv8f64(<vscale x 8 x double> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i8_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i8_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 8 x i8> %idxs
@@ -2084,19 +2036,18 @@ define void @vpscatter_baseidx_nxv8i8_nxv8f64(<vscale x 8 x double> %val, ptr %b
 define void @vpscatter_baseidx_sext_nxv8i8_nxv8f64(<vscale x 8 x double> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf4 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i8_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf8 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i8> %idxs to <vscale x 8 x i64>
@@ -2108,19 +2059,19 @@ define void @vpscatter_baseidx_sext_nxv8i8_nxv8f64(<vscale x 8 x double> %val, p
 define void @vpscatter_baseidx_zext_nxv8i8_nxv8f64(<vscale x 8 x double> %val, ptr %base, <vscale x 8 x i8> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV32-NEXT:    vzext.vf2 v18, v16
 ; RV32-NEXT:    vsll.vi v16, v18, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei16.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i8_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e16, m2, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e16, m2, ta, ma
 ; RV64-NEXT:    vzext.vf2 v18, v16
 ; RV64-NEXT:    vsll.vi v16, v18, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei16.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i8> %idxs to <vscale x 8 x i64>
@@ -2132,19 +2083,18 @@ define void @vpscatter_baseidx_zext_nxv8i8_nxv8f64(<vscale x 8 x double> %val, p
 define void @vpscatter_baseidx_nxv8i16_nxv8f64(<vscale x 8 x double> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i16_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i16_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 8 x i16> %idxs
@@ -2155,19 +2105,18 @@ define void @vpscatter_baseidx_nxv8i16_nxv8f64(<vscale x 8 x double> %val, ptr %
 define void @vpscatter_baseidx_sext_nxv8i16_nxv8f64(<vscale x 8 x double> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i16_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsext.vf2 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i16_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf4 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i16> %idxs to <vscale x 8 x i64>
@@ -2179,19 +2128,19 @@ define void @vpscatter_baseidx_sext_nxv8i16_nxv8f64(<vscale x 8 x double> %val,
 define void @vpscatter_baseidx_zext_nxv8i16_nxv8f64(<vscale x 8 x double> %val, ptr %base, <vscale x 8 x i16> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i16_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vzext.vf2 v20, v16
 ; RV32-NEXT:    vsll.vi v16, v20, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i16_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV64-NEXT:    vzext.vf2 v20, v16
 ; RV64-NEXT:    vsll.vi v16, v20, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i16> %idxs to <vscale x 8 x i64>
@@ -2203,18 +2152,17 @@ define void @vpscatter_baseidx_zext_nxv8i16_nxv8f64(<vscale x 8 x double> %val,
 define void @vpscatter_baseidx_nxv8i32_nxv8f64(<vscale x 8 x double> %val, ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8i32_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8i32_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 8 x i32> %idxs
@@ -2225,18 +2173,17 @@ define void @vpscatter_baseidx_nxv8i32_nxv8f64(<vscale x 8 x double> %val, ptr %
 define void @vpscatter_baseidx_sext_nxv8i32_nxv8f64(<vscale x 8 x double> %val, ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_sext_nxv8i32_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_sext_nxv8i32_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsext.vf2 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = sext <vscale x 8 x i32> %idxs to <vscale x 8 x i64>
@@ -2248,18 +2195,17 @@ define void @vpscatter_baseidx_sext_nxv8i32_nxv8f64(<vscale x 8 x double> %val,
 define void @vpscatter_baseidx_zext_nxv8i32_nxv8f64(<vscale x 8 x double> %val, ptr %base, <vscale x 8 x i32> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_zext_nxv8i32_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vsll.vi v16, v16, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_zext_nxv8i32_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
+; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vzext.vf2 v24, v16
 ; RV64-NEXT:    vsll.vi v16, v24, 3
-; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %eidxs = zext <vscale x 8 x i32> %idxs to <vscale x 8 x i64>
@@ -2271,18 +2217,17 @@ define void @vpscatter_baseidx_zext_nxv8i32_nxv8f64(<vscale x 8 x double> %val,
 define void @vpscatter_baseidx_nxv8f64(<vscale x 8 x double> %val, ptr %base, <vscale x 8 x i64> %idxs, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; RV32-LABEL: vpscatter_baseidx_nxv8f64:
 ; RV32:       # %bb.0:
-; RV32-NEXT:    vsetvli a2, zero, e32, m4, ta, ma
+; RV32-NEXT:    vsetvli zero, a1, e32, m4, ta, ma
 ; RV32-NEXT:    vnsrl.wi v24, v16, 0
 ; RV32-NEXT:    vsll.vi v16, v24, 3
-; RV32-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV32-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV32-NEXT:    vsoxei32.v v8, (a0), v16, v0.t
 ; RV32-NEXT:    ret
 ;
 ; RV64-LABEL: vpscatter_baseidx_nxv8f64:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    vsetvli a2, zero, e64, m8, ta, ma
-; RV64-NEXT:    vsll.vi v16, v16, 3
 ; RV64-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; RV64-NEXT:    vsll.vi v16, v16, 3
 ; RV64-NEXT:    vsoxei64.v v8, (a0), v16, v0.t
 ; RV64-NEXT:    ret
   %ptrs = getelementptr inbounds double, ptr %base, <vscale x 8 x i64> %idxs