From 933fc63a1d230896bc09a08cf08dde4ac5b51703 Mon Sep 17 00:00:00 2001
From: Luke Lau <luke@igalia.com>
Date: Wed, 11 Sep 2024 09:44:57 +0800
Subject: [PATCH] [RISCV] Rematerialize vmv.s.x and vfmv.s.f (#108012)

Continuing with #107993 and #108007, this handles the last of the main
rematerializable vector instructions.

There's an extra spill in one of the test cases, but it's likely noise
from the spill weights and isn't an issue in practice.
---
 llvm/lib/Target/RISCV/RISCVInstrInfo.cpp      |   2 +
 .../Target/RISCV/RISCVInstrInfoVPseudos.td    |   4 +-
 .../rvv/fixed-vectors-interleaved-access.ll   | 912 +++++++++---------
 llvm/test/CodeGen/RISCV/rvv/remat.ll          | 130 +++
 4 files changed, 600 insertions(+), 448 deletions(-)

diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfo.cpp b/llvm/lib/Target/RISCV/RISCVInstrInfo.cpp
index a805c68e7795..13212c2aea5d 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfo.cpp
@@ -172,6 +172,8 @@ bool RISCVInstrInfo::isReallyTriviallyReMaterializable(
   case RISCV::VMV_V_X:
   case RISCV::VFMV_V_F:
   case RISCV::VMV_V_I:
+  case RISCV::VMV_S_X:
+  case RISCV::VFMV_S_F:
   case RISCV::VID_V:
     if (MI.getOperand(1).isUndef() &&
         /* After RISCVInsertVSETVLI most pseudos will have implicit uses on vl
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td b/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
index 2eceef5066f7..430e09fd834b 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
@@ -6764,7 +6764,7 @@ let mayLoad = 0, mayStore = 0, hasSideEffects = 0 in {
     Pseudo<(outs GPR:$rd), (ins VR:$rs2, ixlenimm:$sew), []>,
     Sched<[WriteVMovXS, ReadVMovXS]>,
     RISCVVPseudo;
-  let HasVLOp = 1, HasSEWOp = 1, BaseInstr = VMV_S_X,
+  let HasVLOp = 1, HasSEWOp = 1, BaseInstr = VMV_S_X, isReMaterializable = 1,
       Constraints = "$rd = $rs1" in
   def PseudoVMV_S_X: Pseudo<(outs VR:$rd),
                             (ins VR:$rs1, GPR:$rs2, AVL:$vl, ixlenimm:$sew),
@@ -6787,7 +6787,7 @@ let mayLoad = 0, mayStore = 0, hasSideEffects = 0 in {
              (ins VR:$rs2, ixlenimm:$sew), []>,
       Sched<[WriteVMovFS, ReadVMovFS]>,
       RISCVVPseudo;
-    let HasVLOp = 1, HasSEWOp = 1, BaseInstr = VFMV_S_F,
+    let HasVLOp = 1, HasSEWOp = 1, BaseInstr = VFMV_S_F, isReMaterializable = 1,
         Constraints = "$rd = $rs1" in
     def "PseudoVFMV_S_" # f.FX :
       Pseudo<(outs VR:$rd),
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll
index bc3e135a588a..eff56e408d6d 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll
@@ -159,296 +159,308 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    addi sp, sp, -16
 ; RV32-NEXT:    .cfi_def_cfa_offset 16
 ; RV32-NEXT:    csrr a2, vlenb
-; RV32-NEXT:    li a3, 80
+; RV32-NEXT:    li a3, 84
 ; RV32-NEXT:    mul a2, a2, a3
 ; RV32-NEXT:    sub sp, sp, a2
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0e, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0xd0, 0x00, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 80 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0e, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0xd4, 0x00, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 84 * vlenb
 ; RV32-NEXT:    addi a3, a1, 256
 ; RV32-NEXT:    li a2, 32
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, ma
-; RV32-NEXT:    vle32.v v16, (a3)
+; RV32-NEXT:    vle32.v v8, (a3)
 ; RV32-NEXT:    csrr a3, vlenb
-; RV32-NEXT:    slli a3, a3, 6
+; RV32-NEXT:    li a4, 76
+; RV32-NEXT:    mul a3, a3, a4
 ; RV32-NEXT:    add a3, sp, a3
 ; RV32-NEXT:    addi a3, a3, 16
-; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
+; RV32-NEXT:    vs8r.v v8, (a3) # Unknown-size Folded Spill
 ; RV32-NEXT:    addi a3, a1, 128
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, ma
-; RV32-NEXT:    vslideup.vi v8, v16, 4
+; RV32-NEXT:    vslideup.vi v4, v8, 4
 ; RV32-NEXT:    csrr a4, vlenb
 ; RV32-NEXT:    li a5, 40
 ; RV32-NEXT:    mul a4, a4, a5
 ; RV32-NEXT:    add a4, sp, a4
 ; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs4r.v v8, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    vs4r.v v4, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    lui a4, 12
 ; RV32-NEXT:    vmv.s.x v0, a4
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 16
+; RV32-NEXT:    vs1r.v v0, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetivli zero, 16, e32, m8, ta, ma
-; RV32-NEXT:    vslidedown.vi v16, v16, 16
+; RV32-NEXT:    vslidedown.vi v8, v8, 16
 ; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 56
+; RV32-NEXT:    li a5, 48
 ; RV32-NEXT:    mul a4, a4, a5
 ; RV32-NEXT:    add a4, sp, a4
 ; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
-; RV32-NEXT:    vmv1r.v v3, v0
+; RV32-NEXT:    vs8r.v v8, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
-; RV32-NEXT:    vslideup.vi v8, v16, 10, v0.t
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 44
-; RV32-NEXT:    mul a4, a4, a5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs4r.v v8, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    vslideup.vi v4, v8, 10, v0.t
 ; RV32-NEXT:    lui a4, %hi(.LCPI6_0)
 ; RV32-NEXT:    addi a4, a4, %lo(.LCPI6_0)
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    vle16.v v8, (a4)
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    slli a4, a4, 5
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs4r.v v8, (a4) # Unknown-size Folded Spill
+; RV32-NEXT:    vle16.v v0, (a4)
 ; RV32-NEXT:    lui a4, %hi(.LCPI6_1)
 ; RV32-NEXT:    addi a4, a4, %lo(.LCPI6_1)
 ; RV32-NEXT:    lui a5, 1
 ; RV32-NEXT:    vle16.v v8, (a4)
 ; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a6, 24
+; RV32-NEXT:    li a6, 56
 ; RV32-NEXT:    mul a4, a4, a6
 ; RV32-NEXT:    add a4, sp, a4
 ; RV32-NEXT:    addi a4, a4, 16
 ; RV32-NEXT:    vs4r.v v8, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vle32.v v8, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a4, 72
+; RV32-NEXT:    li a4, 68
 ; RV32-NEXT:    mul a1, a1, a4
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    vle32.v v24, (a3)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 48
+; RV32-NEXT:    li a3, 60
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    addi a1, a5, -64
-; RV32-NEXT:    vmv.s.x v0, a1
+; RV32-NEXT:    vmv.s.x v16, a1
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 36
+; RV32-NEXT:    li a3, 44
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vs1r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vrgatherei16.vv v16, v8, v0
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 5
+; RV32-NEXT:    li a3, 44
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl4r.v v4, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v16, v8, v4
+; RV32-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 24
+; RV32-NEXT:    li a3, 56
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vrgatherei16.vv v16, v24, v8, v0.t
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 44
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsetivli zero, 12, e32, m4, tu, ma
-; RV32-NEXT:    vmv.v.v v8, v16
+; RV32-NEXT:    vmv.v.v v4, v16
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 44
+; RV32-NEXT:    li a3, 36
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vs4r.v v4, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 6
+; RV32-NEXT:    li a3, 76
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
 ; RV32-NEXT:    vslideup.vi v12, v8, 2
-; RV32-NEXT:    vmv1r.v v8, v3
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 4
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs1r.v v3, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    vmv1r.v v0, v3
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 56
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vslideup.vi v12, v16, 8, v0.t
-; RV32-NEXT:    lui a1, %hi(.LCPI6_2)
-; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_2)
-; RV32-NEXT:    lui a3, %hi(.LCPI6_3)
-; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_3)
-; RV32-NEXT:    vsetvli zero, a2, e16, m4, ta, ma
-; RV32-NEXT:    vle16.v v0, (a1)
-; RV32-NEXT:    vle16.v v4, (a3)
-; RV32-NEXT:    lui a1, %hi(.LCPI6_4)
-; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_4)
-; RV32-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV32-NEXT:    vle16.v v10, (a1)
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 72
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    vrgatherei16.vv v24, v16, v0
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 36
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 48
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v24, v16, v4, v0.t
-; RV32-NEXT:    vsetivli zero, 12, e32, m4, tu, ma
-; RV32-NEXT:    vmv.v.v v12, v24
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 36
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 6
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
-; RV32-NEXT:    vrgatherei16.vv v12, v24, v10
-; RV32-NEXT:    vmv1r.v v0, v8
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 56
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vslideup.vi v12, v24, 6, v0.t
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 5
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    lui a1, %hi(.LCPI6_5)
-; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_5)
-; RV32-NEXT:    lui a3, %hi(.LCPI6_6)
-; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_6)
-; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    vle16.v v12, (a1)
-; RV32-NEXT:    vle16.v v8, (a3)
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 12
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    li a1, 960
-; RV32-NEXT:    vmv.s.x v8, a1
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 72
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v24, v0, v12
-; RV32-NEXT:    vmv1r.v v3, v8
-; RV32-NEXT:    vmv1r.v v0, v8
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 12
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v24, v16, v8, v0.t
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    li a3, 24
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    lui a1, %hi(.LCPI6_7)
-; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_7)
-; RV32-NEXT:    lui a3, %hi(.LCPI6_8)
-; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_8)
-; RV32-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV32-NEXT:    vle16.v v8, (a1)
-; RV32-NEXT:    lui a1, %hi(.LCPI6_9)
-; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_9)
-; RV32-NEXT:    vsetvli zero, a2, e16, m4, ta, ma
-; RV32-NEXT:    vle16.v v4, (a3)
-; RV32-NEXT:    vle16.v v12, (a1)
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 6
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
-; RV32-NEXT:    vrgatherei16.vv v12, v24, v8
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 4
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 56
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vmv4r.v v24, v16
-; RV32-NEXT:    vslideup.vi v12, v16, 4, v0.t
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 12
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 72
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    vrgatherei16.vv v8, v16, v4
-; RV32-NEXT:    vmv1r.v v0, v3
+; RV32-NEXT:    vl1r.v v1, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv1r.v v0, v1
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    li a3, 48
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vslideup.vi v12, v16, 8, v0.t
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 56
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    lui a1, %hi(.LCPI6_2)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_2)
+; RV32-NEXT:    lui a3, %hi(.LCPI6_3)
+; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_3)
+; RV32-NEXT:    vsetvli zero, a2, e16, m4, ta, ma
+; RV32-NEXT:    vle16.v v12, (a1)
+; RV32-NEXT:    vle16.v v8, (a3)
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 28
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    lui a1, %hi(.LCPI6_4)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_4)
+; RV32-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
+; RV32-NEXT:    vle16.v v2, (a1)
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 68
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
+; RV32-NEXT:    vrgatherei16.vv v24, v16, v12
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 44
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 60
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 28
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl4r.v v4, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vrgatherei16.vv v24, v8, v4, v0.t
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 56
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vsetivli zero, 12, e32, m4, tu, ma
+; RV32-NEXT:    vmv.v.v v8, v24
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 56
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 76
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
+; RV32-NEXT:    vrgatherei16.vv v8, v24, v2
+; RV32-NEXT:    vmv1r.v v0, v1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 48
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vslideup.vi v8, v24, 6, v0.t
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 44
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    lui a1, %hi(.LCPI6_5)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_5)
+; RV32-NEXT:    lui a3, %hi(.LCPI6_6)
+; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_6)
+; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
+; RV32-NEXT:    vle16.v v24, (a1)
+; RV32-NEXT:    vle16.v v4, (a3)
+; RV32-NEXT:    li a1, 960
+; RV32-NEXT:    vmv.s.x v0, a1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 4
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vrgatherei16.vv v8, v16, v24
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 60
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vrgatherei16.vv v8, v24, v4, v0.t
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 28
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    lui a1, %hi(.LCPI6_7)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_7)
+; RV32-NEXT:    lui a3, %hi(.LCPI6_8)
+; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_8)
+; RV32-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
+; RV32-NEXT:    vle16.v v16, (a1)
+; RV32-NEXT:    lui a1, %hi(.LCPI6_9)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_9)
+; RV32-NEXT:    vsetvli zero, a2, e16, m4, ta, ma
+; RV32-NEXT:    vle16.v v8, (a3)
+; RV32-NEXT:    csrr a3, vlenb
+; RV32-NEXT:    slli a3, a3, 2
+; RV32-NEXT:    add a3, sp, a3
+; RV32-NEXT:    addi a3, a3, 16
+; RV32-NEXT:    vs4r.v v8, (a3) # Unknown-size Folded Spill
+; RV32-NEXT:    vle16.v v8, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    slli a1, a1, 3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl4r.v v28, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v8, v16, v28, v0.t
+; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 76
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
+; RV32-NEXT:    vrgatherei16.vv v12, v8, v16
+; RV32-NEXT:    vmv1r.v v0, v1
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 48
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vslideup.vi v12, v16, 4, v0.t
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 24
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 68
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl4r.v v20, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
+; RV32-NEXT:    vrgatherei16.vv v8, v0, v20
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 4
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl4r.v v20, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vrgatherei16.vv v8, v24, v20, v0.t
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    slli a1, a1, 4
 ; RV32-NEXT:    add a1, sp, a1
@@ -461,48 +473,51 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    lui a1, 15
 ; RV32-NEXT:    vmv.s.x v3, a1
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 6
+; RV32-NEXT:    li a3, 76
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vslideup.vi v8, v16, 6
+; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vslideup.vi v8, v24, 6
 ; RV32-NEXT:    vmv1r.v v0, v3
-; RV32-NEXT:    vrgatherei16.vv v8, v24, v12, v0.t
+; RV32-NEXT:    vrgatherei16.vv v8, v16, v12, v0.t
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 2
+; RV32-NEXT:    li a3, 76
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vmv4r.v v24, v16
 ; RV32-NEXT:    lui a1, %hi(.LCPI6_11)
 ; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_11)
 ; RV32-NEXT:    lui a3, %hi(.LCPI6_12)
 ; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_12)
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    vle16.v v24, (a1)
+; RV32-NEXT:    vle16.v v28, (a1)
 ; RV32-NEXT:    vle16.v v4, (a3)
 ; RV32-NEXT:    li a1, 1008
 ; RV32-NEXT:    vmv.s.x v0, a1
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    slli a1, a1, 2
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 72
+; RV32-NEXT:    li a3, 68
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v8, v16, v24
+; RV32-NEXT:    vrgatherei16.vv v8, v16, v28
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 48
+; RV32-NEXT:    li a3, 60
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vrgatherei16.vv v8, v16, v4, v0.t
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 6
+; RV32-NEXT:    slli a1, a1, 3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
@@ -511,14 +526,14 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    lui a3, %hi(.LCPI6_14)
 ; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_14)
 ; RV32-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV32-NEXT:    vle16.v v20, (a1)
+; RV32-NEXT:    vle16.v v8, (a1)
 ; RV32-NEXT:    lui a1, %hi(.LCPI6_15)
 ; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_15)
 ; RV32-NEXT:    vsetvli zero, a2, e16, m4, ta, ma
-; RV32-NEXT:    vle16.v v24, (a3)
-; RV32-NEXT:    vle16.v v8, (a1)
+; RV32-NEXT:    vle16.v v28, (a3)
+; RV32-NEXT:    vle16.v v12, (a1)
 ; RV32-NEXT:    addi a1, sp, 16
-; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    vmv1r.v v0, v3
 ; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    li a3, 40
@@ -526,21 +541,16 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl4r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 56
-; RV32-NEXT:    mul a1, a1, a3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
-; RV32-NEXT:    vrgatherei16.vv v16, v8, v20, v0.t
+; RV32-NEXT:    vrgatherei16.vv v16, v24, v8, v0.t
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 5
+; RV32-NEXT:    li a3, 44
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl4r.v v20, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 24
+; RV32-NEXT:    li a3, 28
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
@@ -548,20 +558,20 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    vsetivli zero, 10, e32, m4, tu, ma
 ; RV32-NEXT:    vmv.v.v v20, v8
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 72
+; RV32-NEXT:    li a3, 68
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    vrgatherei16.vv v8, v0, v24
+; RV32-NEXT:    vrgatherei16.vv v8, v0, v28
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    slli a1, a1, 2
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a2, 48
+; RV32-NEXT:    li a2, 60
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
@@ -570,56 +580,57 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    vl4r.v v4, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vrgatherei16.vv v8, v24, v4, v0.t
 ; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a2, 24
+; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl4r.v v28, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a1, vlenb
 ; RV32-NEXT:    slli a1, a1, 4
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vsetivli zero, 10, e32, m4, tu, ma
+; RV32-NEXT:    vmv.v.v v28, v0
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a2, 12
+; RV32-NEXT:    li a2, 76
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl4r.v v24, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vsetivli zero, 10, e32, m4, tu, ma
-; RV32-NEXT:    vmv.v.v v24, v0
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 6
+; RV32-NEXT:    slli a1, a1, 3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 2
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl4r.v v28, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vmv.v.v v28, v0
+; RV32-NEXT:    vmv.v.v v24, v0
 ; RV32-NEXT:    vmv.v.v v16, v8
 ; RV32-NEXT:    addi a1, a0, 320
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, ma
 ; RV32-NEXT:    vse32.v v16, (a1)
 ; RV32-NEXT:    addi a1, a0, 256
-; RV32-NEXT:    vse32.v v28, (a1)
-; RV32-NEXT:    addi a1, a0, 192
 ; RV32-NEXT:    vse32.v v24, (a1)
+; RV32-NEXT:    addi a1, a0, 192
+; RV32-NEXT:    vse32.v v28, (a1)
 ; RV32-NEXT:    addi a1, a0, 128
 ; RV32-NEXT:    vse32.v v20, (a1)
 ; RV32-NEXT:    addi a1, a0, 64
 ; RV32-NEXT:    csrr a2, vlenb
-; RV32-NEXT:    li a3, 36
+; RV32-NEXT:    li a3, 56
 ; RV32-NEXT:    mul a2, a2, a3
 ; RV32-NEXT:    add a2, sp, a2
 ; RV32-NEXT:    addi a2, a2, 16
 ; RV32-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
 ; RV32-NEXT:    vse32.v v8, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a2, 44
+; RV32-NEXT:    li a2, 36
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vse32.v v8, (a0)
 ; RV32-NEXT:    csrr a0, vlenb
-; RV32-NEXT:    li a1, 80
+; RV32-NEXT:    li a1, 84
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add sp, sp, a0
 ; RV32-NEXT:    addi sp, sp, 16
@@ -630,15 +641,15 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV64-NEXT:    addi sp, sp, -16
 ; RV64-NEXT:    .cfi_def_cfa_offset 16
 ; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    li a3, 74
+; RV64-NEXT:    li a3, 66
 ; RV64-NEXT:    mul a2, a2, a3
 ; RV64-NEXT:    sub sp, sp, a2
-; RV64-NEXT:    .cfi_escape 0x0f, 0x0e, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0xca, 0x00, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 74 * vlenb
+; RV64-NEXT:    .cfi_escape 0x0f, 0x0e, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0xc2, 0x00, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 66 * vlenb
 ; RV64-NEXT:    addi a2, a1, 256
 ; RV64-NEXT:    vsetivli zero, 16, e64, m8, ta, ma
 ; RV64-NEXT:    vle64.v v16, (a2)
 ; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    li a3, 25
+; RV64-NEXT:    li a3, 21
 ; RV64-NEXT:    mul a2, a2, a3
 ; RV64-NEXT:    add a2, sp, a2
 ; RV64-NEXT:    addi a2, a2, 16
@@ -646,76 +657,85 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV64-NEXT:    addi a2, a1, 128
 ; RV64-NEXT:    vle64.v v8, (a1)
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a3, a1, 6
-; RV64-NEXT:    add a1, a3, a1
+; RV64-NEXT:    li a3, 57
+; RV64-NEXT:    mul a1, a1, a3
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, ma
 ; RV64-NEXT:    vrgather.vi v12, v16, 4
 ; RV64-NEXT:    li a1, 128
-; RV64-NEXT:    vmv.s.x v8, a1
+; RV64-NEXT:    vmv.s.x v0, a1
+; RV64-NEXT:    addi a1, sp, 16
+; RV64-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    vsetivli zero, 8, e64, m8, ta, ma
 ; RV64-NEXT:    vslidedown.vi v16, v16, 8
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a3, 49
+; RV64-NEXT:    li a3, 37
 ; RV64-NEXT:    mul a1, a1, a3
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    vmv1r.v v0, v8
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
 ; RV64-NEXT:    vrgather.vi v12, v16, 2, v0.t
 ; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
 ; RV64-NEXT:    vid.v v10
 ; RV64-NEXT:    li a1, 6
-; RV64-NEXT:    vmul.vx v2, v10, a1
+; RV64-NEXT:    vmul.vx v8, v10, a1
 ; RV64-NEXT:    li a1, 56
-; RV64-NEXT:    vle64.v v16, (a2)
+; RV64-NEXT:    vle64.v v24, (a2)
 ; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    li a3, 57
+; RV64-NEXT:    li a3, 45
 ; RV64-NEXT:    mul a2, a2, a3
 ; RV64-NEXT:    add a2, sp, a2
 ; RV64-NEXT:    addi a2, a2, 16
-; RV64-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
-; RV64-NEXT:    vmv.s.x v7, a1
-; RV64-NEXT:    vadd.vi v10, v2, -16
+; RV64-NEXT:    vs8r.v v24, (a2) # Unknown-size Folded Spill
+; RV64-NEXT:    vmv.s.x v10, a1
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 6
-; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    li a2, 53
+; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vrgatherei16.vv v16, v24, v2
-; RV64-NEXT:    vmv1r.v v0, v7
+; RV64-NEXT:    vs1r.v v10, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vadd.vi v10, v8, -16
 ; RV64-NEXT:    csrr a1, vlenb
 ; RV64-NEXT:    li a2, 57
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; RV64-NEXT:    vrgatherei16.vv v16, v0, v8
+; RV64-NEXT:    vmv2r.v v4, v8
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 53
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl1r.v v6, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vmv1r.v v0, v6
 ; RV64-NEXT:    vrgatherei16.vv v16, v24, v10, v0.t
 ; RV64-NEXT:    vsetivli zero, 6, e64, m4, tu, ma
 ; RV64-NEXT:    vmv.v.v v12, v16
 ; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    slli a2, a1, 4
+; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    csrr a1, vlenb
 ; RV64-NEXT:    li a2, 21
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 25
-; RV64-NEXT:    mul a1, a1, a2
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
-; RV64-NEXT:    vrgather.vi v12, v16, 5
-; RV64-NEXT:    vmv1r.v v0, v8
-; RV64-NEXT:    vmv1r.v v6, v8
+; RV64-NEXT:    vrgather.vi v12, v8, 5
+; RV64-NEXT:    addi a1, sp, 16
+; RV64-NEXT:    vl1r.v v1, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vmv1r.v v0, v1
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 49
+; RV64-NEXT:    li a2, 37
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
@@ -723,19 +743,19 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV64-NEXT:    vrgather.vi v12, v16, 3, v0.t
 ; RV64-NEXT:    vmv.v.v v28, v12
 ; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v24, v2, 1
-; RV64-NEXT:    vadd.vi v26, v2, -15
+; RV64-NEXT:    vadd.vi v24, v4, 1
+; RV64-NEXT:    vadd.vi v26, v4, -15
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 6
-; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    li a2, 57
+; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
 ; RV64-NEXT:    vrgatherei16.vv v16, v8, v24
-; RV64-NEXT:    vmv1r.v v0, v7
+; RV64-NEXT:    vmv1r.v v0, v6
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 57
+; RV64-NEXT:    li a2, 45
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
@@ -744,8 +764,8 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV64-NEXT:    vsetivli zero, 6, e64, m4, tu, ma
 ; RV64-NEXT:    vmv.v.v v28, v16
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 4
-; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    li a2, 13
+; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vs4r.v v28, (a1) # Unknown-size Folded Spill
@@ -755,7 +775,7 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV64-NEXT:    vmv.v.i v9, 6
 ; RV64-NEXT:    vmv.v.x v10, a1
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 25
+; RV64-NEXT:    li a2, 21
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
@@ -763,259 +783,253 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, ma
 ; RV64-NEXT:    vrgatherei16.vv v12, v16, v9
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 45
+; RV64-NEXT:    li a2, 53
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    vrgatherei16.vv v12, v16, v10
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 41
-; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    slli a2, a1, 3
+; RV64-NEXT:    add a1, a2, a1
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    vmv4r.v v8, v16
 ; RV64-NEXT:    vrgather.vi v12, v16, 2
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 37
-; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    slli a2, a1, 5
+; RV64-NEXT:    add a1, a2, a1
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    vrgather.vi v12, v16, 3
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 5
-; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    li a2, 29
+; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    li a1, 24
-; RV64-NEXT:    vmv.s.x v1, a1
-; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v24, v2, 2
-; RV64-NEXT:    vadd.vi v4, v2, -14
+; RV64-NEXT:    vmv.s.x v0, a1
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 6
-; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    li a2, 21
+; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vrgatherei16.vv v8, v16, v24
-; RV64-NEXT:    vmv1r.v v0, v1
+; RV64-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
+; RV64-NEXT:    vadd.vi v16, v4, 2
+; RV64-NEXT:    vadd.vi v2, v4, -14
 ; RV64-NEXT:    csrr a1, vlenb
 ; RV64-NEXT:    li a2, 57
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v8, v24, v4, v0.t
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; RV64-NEXT:    vrgatherei16.vv v8, v24, v16
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 25
+; RV64-NEXT:    li a2, 45
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v8, v16, v2, v0.t
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    vmv1r.v v0, v6
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 49
-; RV64-NEXT:    mul a1, a1, a2
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 45
-; RV64-NEXT:    mul a1, a1, a2
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl4r.v v20, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
-; RV64-NEXT:    vrgather.vi v20, v16, 4, v0.t
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 45
-; RV64-NEXT:    mul a1, a1, a2
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs4r.v v20, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v4, v2, 3
-; RV64-NEXT:    vadd.vi v8, v2, -13
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs2r.v v8, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 6
-; RV64-NEXT:    add a1, a2, a1
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vrgatherei16.vv v8, v16, v4
 ; RV64-NEXT:    vmv1r.v v0, v1
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl2r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v8, v24, v16, v0.t
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 3
-; RV64-NEXT:    add a1, a2, a1
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    vmv1r.v v0, v6
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 49
+; RV64-NEXT:    li a2, 37
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 41
+; RV64-NEXT:    li a2, 53
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vl4r.v v28, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
-; RV64-NEXT:    vrgather.vi v8, v24, 5, v0.t
+; RV64-NEXT:    vrgather.vi v28, v24, 4, v0.t
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 41
+; RV64-NEXT:    li a2, 53
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    lui a1, 96
-; RV64-NEXT:    li a2, 192
-; RV64-NEXT:    vmv.s.x v28, a2
-; RV64-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
-; RV64-NEXT:    vmv.v.x v8, a1
-; RV64-NEXT:    vmv1r.v v0, v28
+; RV64-NEXT:    vs4r.v v28, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vmv2r.v v8, v4
+; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
+; RV64-NEXT:    vadd.vi v4, v4, 3
+; RV64-NEXT:    vadd.vi v6, v8, -13
+; RV64-NEXT:    vmv2r.v v2, v8
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 57
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; RV64-NEXT:    vrgatherei16.vv v8, v24, v4
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 21
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v8, v16, v6, v0.t
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 21
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vmv1r.v v0, v1
 ; RV64-NEXT:    csrr a1, vlenb
 ; RV64-NEXT:    li a2, 37
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    slli a2, a1, 3
+; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl4r.v v4, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
+; RV64-NEXT:    vrgather.vi v4, v16, 5, v0.t
+; RV64-NEXT:    lui a1, 96
+; RV64-NEXT:    li a2, 192
+; RV64-NEXT:    vmv.s.x v1, a2
+; RV64-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
+; RV64-NEXT:    vmv.v.x v8, a1
+; RV64-NEXT:    vmv1r.v v0, v1
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    slli a2, a1, 5
+; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl4r.v v12, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
-; RV64-NEXT:    vrgatherei16.vv v12, v24, v8, v0.t
+; RV64-NEXT:    vrgatherei16.vv v12, v16, v8, v0.t
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 37
-; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    slli a2, a1, 5
+; RV64-NEXT:    add a1, a2, a1
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    li a1, 28
 ; RV64-NEXT:    vmv.s.x v0, a1
-; RV64-NEXT:    addi a1, sp, 16
-; RV64-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v30, v2, 4
-; RV64-NEXT:    vadd.vi v6, v2, -12
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 6
+; RV64-NEXT:    slli a2, a1, 3
 ; RV64-NEXT:    add a1, a2, a1
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vrgatherei16.vv v16, v8, v30
+; RV64-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
+; RV64-NEXT:    vadd.vi v22, v2, 4
+; RV64-NEXT:    vadd.vi v20, v2, -12
 ; RV64-NEXT:    csrr a1, vlenb
 ; RV64-NEXT:    li a2, 57
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v16, v8, v6, v0.t
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    lui a1, 112
-; RV64-NEXT:    addi a1, a1, 1
-; RV64-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
-; RV64-NEXT:    vmv.v.x v12, a1
-; RV64-NEXT:    vmv1r.v v0, v28
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 5
-; RV64-NEXT:    add a1, a2, a1
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl4r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
-; RV64-NEXT:    vrgatherei16.vv v16, v24, v12, v0.t
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 5
-; RV64-NEXT:    add a1, a2, a1
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs4r.v v16, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; RV64-NEXT:    vrgatherei16.vv v8, v24, v22
 ; RV64-NEXT:    csrr a1, vlenb
 ; RV64-NEXT:    li a2, 45
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl4r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 25
-; RV64-NEXT:    mul a1, a1, a2
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vsetivli zero, 5, e64, m4, tu, ma
-; RV64-NEXT:    vmv.v.v v16, v24
-; RV64-NEXT:    vmv2r.v v8, v2
-; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v12, v2, 5
+; RV64-NEXT:    vrgatherei16.vv v8, v24, v20, v0.t
+; RV64-NEXT:    lui a1, 112
+; RV64-NEXT:    addi a1, a1, 1
+; RV64-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
+; RV64-NEXT:    vmv.v.x v12, a1
+; RV64-NEXT:    vmv1r.v v0, v1
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 6
-; RV64-NEXT:    add a1, a2, a1
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
-; RV64-NEXT:    vrgatherei16.vv v24, v0, v12
-; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v2, v8, -11
-; RV64-NEXT:    addi a1, sp, 16
-; RV64-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 57
+; RV64-NEXT:    li a2, 29
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vrgatherei16.vv v24, v8, v2, v0.t
+; RV64-NEXT:    vl4r.v v20, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
+; RV64-NEXT:    vrgatherei16.vv v20, v16, v12, v0.t
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 41
+; RV64-NEXT:    li a2, 29
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vs4r.v v20, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 53
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl4r.v v12, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vsetivli zero, 5, e64, m4, tu, ma
+; RV64-NEXT:    vmv.v.v v12, v24
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 53
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
+; RV64-NEXT:    vadd.vi v12, v2, 5
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 57
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; RV64-NEXT:    vrgatherei16.vv v24, v16, v12
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; RV64-NEXT:    vadd.vi v12, v2, -11
+; RV64-NEXT:    csrr a1, vlenb
 ; RV64-NEXT:    slli a2, a1, 3
 ; RV64-NEXT:    add a1, a2, a1
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 45
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
+; RV64-NEXT:    vrgatherei16.vv v24, v16, v12, v0.t
+; RV64-NEXT:    vmv4r.v v12, v4
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 21
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    vsetivli zero, 5, e64, m4, tu, ma
 ; RV64-NEXT:    vmv.v.v v12, v0
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 37
-; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    slli a2, a1, 5
+; RV64-NEXT:    add a1, a2, a1
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl4r.v v20, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vmv.v.v v20, v8
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vmv.v.v v20, v0
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 5
-; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    li a2, 29
+; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
@@ -1028,24 +1042,30 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV64-NEXT:    addi a1, a0, 192
 ; RV64-NEXT:    vse64.v v12, (a1)
 ; RV64-NEXT:    addi a1, a0, 128
-; RV64-NEXT:    vse64.v v16, (a1)
+; RV64-NEXT:    csrr a2, vlenb
+; RV64-NEXT:    li a3, 53
+; RV64-NEXT:    mul a2, a2, a3
+; RV64-NEXT:    add a2, sp, a2
+; RV64-NEXT:    addi a2, a2, 16
+; RV64-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
+; RV64-NEXT:    vse64.v v8, (a1)
 ; RV64-NEXT:    addi a1, a0, 64
 ; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    slli a3, a2, 4
-; RV64-NEXT:    add a2, a3, a2
+; RV64-NEXT:    li a3, 13
+; RV64-NEXT:    mul a2, a2, a3
 ; RV64-NEXT:    add a2, sp, a2
 ; RV64-NEXT:    addi a2, a2, 16
 ; RV64-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
 ; RV64-NEXT:    vse64.v v8, (a1)
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 21
-; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    slli a2, a1, 4
+; RV64-NEXT:    add a1, a2, a1
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    vse64.v v8, (a0)
 ; RV64-NEXT:    csrr a0, vlenb
-; RV64-NEXT:    li a1, 74
+; RV64-NEXT:    li a1, 66
 ; RV64-NEXT:    mul a0, a0, a1
 ; RV64-NEXT:    add sp, sp, a0
 ; RV64-NEXT:    addi sp, sp, 16
diff --git a/llvm/test/CodeGen/RISCV/rvv/remat.ll b/llvm/test/CodeGen/RISCV/rvv/remat.ll
index 343b086898c1..4f58ccb5188d 100644
--- a/llvm/test/CodeGen/RISCV/rvv/remat.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/remat.ll
@@ -377,3 +377,133 @@ define void @vfmv.v.f(ptr %p, double %x) {
   store volatile double %x, ptr %p
   ret void
 }
+
+define void @vmv.s.x(ptr %p, i64 %x) {
+; POSTRA-LABEL: vmv.s.x:
+; POSTRA:       # %bb.0:
+; POSTRA-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
+; POSTRA-NEXT:    vmv.s.x v8, a1
+; POSTRA-NEXT:    vs8r.v v8, (a0)
+; POSTRA-NEXT:    vl8re64.v v16, (a0)
+; POSTRA-NEXT:    vl8re64.v v24, (a0)
+; POSTRA-NEXT:    vl8re64.v v0, (a0)
+; POSTRA-NEXT:    vl8re64.v v8, (a0)
+; POSTRA-NEXT:    vs8r.v v8, (a0)
+; POSTRA-NEXT:    vs8r.v v0, (a0)
+; POSTRA-NEXT:    vs8r.v v24, (a0)
+; POSTRA-NEXT:    vs8r.v v16, (a0)
+; POSTRA-NEXT:    vmv.s.x v8, a1
+; POSTRA-NEXT:    vs8r.v v8, (a0)
+; POSTRA-NEXT:    sd a1, 0(a0)
+; POSTRA-NEXT:    ret
+;
+; PRERA-LABEL: vmv.s.x:
+; PRERA:       # %bb.0:
+; PRERA-NEXT:    addi sp, sp, -16
+; PRERA-NEXT:    .cfi_def_cfa_offset 16
+; PRERA-NEXT:    csrr a2, vlenb
+; PRERA-NEXT:    slli a2, a2, 3
+; PRERA-NEXT:    sub sp, sp, a2
+; PRERA-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 8 * vlenb
+; PRERA-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
+; PRERA-NEXT:    vmv.s.x v8, a1
+; PRERA-NEXT:    vs8r.v v8, (a0)
+; PRERA-NEXT:    vl8re64.v v16, (a0)
+; PRERA-NEXT:    addi a2, sp, 16
+; PRERA-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
+; PRERA-NEXT:    vl8re64.v v24, (a0)
+; PRERA-NEXT:    vl8re64.v v0, (a0)
+; PRERA-NEXT:    vl8re64.v v16, (a0)
+; PRERA-NEXT:    vs8r.v v16, (a0)
+; PRERA-NEXT:    vs8r.v v0, (a0)
+; PRERA-NEXT:    vs8r.v v24, (a0)
+; PRERA-NEXT:    vl8r.v v16, (a2) # Unknown-size Folded Reload
+; PRERA-NEXT:    vs8r.v v16, (a0)
+; PRERA-NEXT:    vs8r.v v8, (a0)
+; PRERA-NEXT:    sd a1, 0(a0)
+; PRERA-NEXT:    csrr a0, vlenb
+; PRERA-NEXT:    slli a0, a0, 3
+; PRERA-NEXT:    add sp, sp, a0
+; PRERA-NEXT:    addi sp, sp, 16
+; PRERA-NEXT:    ret
+  %vmv.s.x = call <vscale x 8 x i64> @llvm.riscv.vmv.s.x.nxv8i64(<vscale x 8 x i64> poison, i64 %x, i64 -1)
+  store volatile <vscale x 8 x i64> %vmv.s.x, ptr %p
+
+  %a = load volatile <vscale x 8 x i64>, ptr %p
+  %b = load volatile <vscale x 8 x i64>, ptr %p
+  %c = load volatile <vscale x 8 x i64>, ptr %p
+  %d = load volatile <vscale x 8 x i64>, ptr %p
+  store volatile <vscale x 8 x i64> %d, ptr %p
+  store volatile <vscale x 8 x i64> %c, ptr %p
+  store volatile <vscale x 8 x i64> %b, ptr %p
+  store volatile <vscale x 8 x i64> %a, ptr %p
+
+  store volatile <vscale x 8 x i64> %vmv.s.x, ptr %p
+  store volatile i64 %x, ptr %p
+  ret void
+}
+
+define void @vfmv.s.f(ptr %p, double %x) {
+; POSTRA-LABEL: vfmv.s.f:
+; POSTRA:       # %bb.0:
+; POSTRA-NEXT:    vsetvli a1, zero, e64, m1, ta, ma
+; POSTRA-NEXT:    vfmv.s.f v8, fa0
+; POSTRA-NEXT:    vs8r.v v8, (a0)
+; POSTRA-NEXT:    vl8re64.v v16, (a0)
+; POSTRA-NEXT:    vl8re64.v v24, (a0)
+; POSTRA-NEXT:    vl8re64.v v0, (a0)
+; POSTRA-NEXT:    vl8re64.v v8, (a0)
+; POSTRA-NEXT:    vs8r.v v8, (a0)
+; POSTRA-NEXT:    vs8r.v v0, (a0)
+; POSTRA-NEXT:    vs8r.v v24, (a0)
+; POSTRA-NEXT:    vs8r.v v16, (a0)
+; POSTRA-NEXT:    vfmv.s.f v8, fa0
+; POSTRA-NEXT:    vs8r.v v8, (a0)
+; POSTRA-NEXT:    fsd fa0, 0(a0)
+; POSTRA-NEXT:    ret
+;
+; PRERA-LABEL: vfmv.s.f:
+; PRERA:       # %bb.0:
+; PRERA-NEXT:    addi sp, sp, -16
+; PRERA-NEXT:    .cfi_def_cfa_offset 16
+; PRERA-NEXT:    csrr a1, vlenb
+; PRERA-NEXT:    slli a1, a1, 3
+; PRERA-NEXT:    sub sp, sp, a1
+; PRERA-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 8 * vlenb
+; PRERA-NEXT:    vsetvli a1, zero, e64, m1, ta, ma
+; PRERA-NEXT:    vfmv.s.f v8, fa0
+; PRERA-NEXT:    vs8r.v v8, (a0)
+; PRERA-NEXT:    vl8re64.v v16, (a0)
+; PRERA-NEXT:    addi a1, sp, 16
+; PRERA-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; PRERA-NEXT:    vl8re64.v v24, (a0)
+; PRERA-NEXT:    vl8re64.v v0, (a0)
+; PRERA-NEXT:    vl8re64.v v16, (a0)
+; PRERA-NEXT:    vs8r.v v16, (a0)
+; PRERA-NEXT:    vs8r.v v0, (a0)
+; PRERA-NEXT:    vs8r.v v24, (a0)
+; PRERA-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; PRERA-NEXT:    vs8r.v v16, (a0)
+; PRERA-NEXT:    vs8r.v v8, (a0)
+; PRERA-NEXT:    fsd fa0, 0(a0)
+; PRERA-NEXT:    csrr a0, vlenb
+; PRERA-NEXT:    slli a0, a0, 3
+; PRERA-NEXT:    add sp, sp, a0
+; PRERA-NEXT:    addi sp, sp, 16
+; PRERA-NEXT:    ret
+  %vfmv.s.f = call <vscale x 8 x double> @llvm.riscv.vfmv.s.f.nxv8f64(<vscale x 8 x double> poison, double %x, i64 -1)
+  store volatile <vscale x 8 x double> %vfmv.s.f, ptr %p
+
+  %a = load volatile <vscale x 8 x double>, ptr %p
+  %b = load volatile <vscale x 8 x double>, ptr %p
+  %c = load volatile <vscale x 8 x double>, ptr %p
+  %d = load volatile <vscale x 8 x double>, ptr %p
+  store volatile <vscale x 8 x double> %d, ptr %p
+  store volatile <vscale x 8 x double> %c, ptr %p
+  store volatile <vscale x 8 x double> %b, ptr %p
+  store volatile <vscale x 8 x double> %a, ptr %p
+
+  store volatile <vscale x 8 x double> %vfmv.s.f, ptr %p
+  store volatile double %x, ptr %p
+  ret void
+}