[MachineLICM] Don't treat cross copies as cheap

[dmgreen]

Summary

This adjusts the cost of cross register bank copy instructions in MachineLICM, no longer treating them as expensive in order to allow them to be pulled out of inner loops. This can be especially important for MVE code where we sink VDUP's into blocks attempting to fold them into register variants of vector instructions. Where this scalar is a float value we are left with a COPY from SPR to GPR which then needs to be hoisted.

This is an alternative to D76024.
The isCrossCopy code was taken from DetectDeadLanes with some adjustment/cleanup. I don't feel like an expert on some of these areas like subreg copies.
I'm not 100% sure all these test diffs are better, but most of them look OK to me.

[arsenm]

llvm/lib/CodeGen/MachineLICM.cpp

1157–1187

This looks pretty much identical to TargetRegisterInfo::shouldRewriteCopySrc/shareSameRegisterFile

[dmgreen]

Thanks for the suggestion.

I made it call shouldRewriteCopySrc, which allows the target to modify things if it wishes. The AMDGPU/idiv-licm.ll test changed ever so slightly. Let me know if I should change things around to instead call shareSameRegisterFile more directly.

[arsenm]

In D76135#1921963, @dmgreen wrote:

Thanks for the suggestion.

I made it call shouldRewriteCopySrc, which allows the target to modify things if it wishes. The AMDGPU/idiv-licm.ll test changed ever so slightly. Let me know if I should change things around to instead call shareSameRegisterFile more directly.

If DetectDeadLanes also has the same code, then they should probably be merged and directly exposed

[dmgreen]

Now using TargetRegisterInfo::shareSameRegisterFile.

[dmgreen]

I have altered the way that MVE VDUP are lowered, and we end up with a VMOV now instead of using a COPY. That means I don't need this any more for MVE at least.

@arsenm, let me know whether you want me to continue with this for AMDGPU or not.

[arsenm]

llvm/include/llvm/CodeGen/TargetRegisterInfo.h

533 ↗

(On Diff #250503)

There's no reason for this to be virtual?

llvm/test/CodeGen/AMDGPU/idiv-licm.ll

276

This probably is an improvement

[dmgreen]

No longer a virtual method.

[arsenm]

llvm/lib/CodeGen/MachineLICM.cpp

1191–1193

I'd like to get somebody else's opinion on this too. Based on the isAsCheapAsAMove check, I'm not sure what the intended handling of copies is here. For AMDGPU for example, copies get more expensive with wider register classes, which isn't a property of the register bank

[qcolombet]

llvm/lib/CodeGen/MachineLICM.cpp

1173

I feel we should stick to shouldRewriteCopySrc to take into account the "cheapness" of the copy.

llvm/lib/CodeGen/MachineLICM.cpp

for (MachineInstr &UseMI : MRI->use_nodbg_instructions(Reg)) {
 
     // Only look at the first in loop use.
     break;
   }
 
   return false;
 }
 
+static bool isCrossCopy(const MachineInstr &MI) {
+  if (!MI.isCopy())
+    return false;
+  const MachineRegisterInfo &MRI = MI.getMF()->getRegInfo();
+
+  Register DstReg = MI.getOperand(0).getReg();
+  unsigned DstSubIdx = MI.getOperand(0).getSubReg();
+  Register SrcReg = MI.getOperand(1).getReg();
+  unsigned SrcSubIdx = MI.getOperand(1).getSubReg();
+
+  if (!SrcReg.isVirtual() || !DstReg.isVirtual())
+    return false;
+
+  const TargetRegisterClass *DstRC = MRI.getRegClass(DstReg);
+  const TargetRegisterClass *SrcRC = MRI.getRegClass(SrcReg);
+  if (DstRC == SrcRC)
+    return false;
+
+  const TargetRegisterInfo &TRI = *MRI.getTargetRegisterInfo();
+
+  unsigned PreA, PreB; // Unused.
+  if (SrcSubIdx && DstSubIdx)
+    return !TRI.getCommonSuperRegClass(SrcRC, SrcSubIdx, DstRC, DstSubIdx, PreA,
+                                       PreB);
+  if (SrcSubIdx)
+    return !TRI.getMatchingSuperRegClass(SrcRC, DstRC, SrcSubIdx);
+  if (DstSubIdx)
+    return !TRI.getMatchingSuperRegClass(DstRC, SrcRC, DstSubIdx);
+  return !TRI.getCommonSubClass(SrcRC, DstRC);
+}
+
 /// Return true if the instruction is marked "cheap" or the operand latency
 /// between its def and a use is one or less.
 bool MachineLICMBase::IsCheapInstruction(MachineInstr &MI) const {
+  if (isCrossCopy(MI))
+    return false;
   if (TII->isAsCheapAsAMove(MI) || MI.isCopyLike())
     return true;
 
   bool isCheap = false;
   unsigned NumDefs = MI.getDesc().getNumDefs();
   for (unsigned i = 0, e = MI.getNumOperands(); NumDefs && i != e; ++i) {
     MachineOperand &DefMO = MI.getOperand(i);
     if (!DefMO.isReg() || !DefMO.isDef())
       continue;

llvm/test/CodeGen/AMDGPU/idiv-licm.ll

 }
 
 define amdgpu_kernel void @udiv16_invariant_denom(i16 addrspace(1)* nocapture %arg, i16 %arg1) {
 ; GFX9-LABEL: udiv16_invariant_denom:
 ; GFX9:       ; %bb.0: ; %bb
 ; GFX9-NEXT:    s_load_dword s3, s[0:1], 0x2c
 ; GFX9-NEXT:    s_mov_b32 s2, 0xffff
 ; GFX9-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
-; GFX9-NEXT:    v_mov_b32_e32 v3, 0
-; GFX9-NEXT:    v_mov_b32_e32 v4, 0
+; GFX9-NEXT:    v_mov_b32_e32 v2, 0
+; GFX9-NEXT:    v_mov_b32_e32 v5, 0
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    s_and_b32 s3, s2, s3
 ; GFX9-NEXT:    v_cvt_f32_u32_e32 v0, s3
 ; GFX9-NEXT:    s_movk_i32 s3, 0x400
-; GFX9-NEXT:    v_rcp_iflag_f32_e32 v1, v0
+; GFX9-NEXT:    v_mov_b32_e32 v4, s5
+; GFX9-NEXT:    v_rcp_iflag_f32_e32 v3, v0
 ; GFX9-NEXT:  BB4_1: ; %bb3
 ; GFX9-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GFX9-NEXT:    v_and_b32_e32 v2, s2, v4
-; GFX9-NEXT:    v_cvt_f32_u32_e32 v8, v2
-; GFX9-NEXT:    v_lshlrev_b64 v[5:6], 1, v[2:3]
-; GFX9-NEXT:    v_mov_b32_e32 v7, s5
-; GFX9-NEXT:    v_add_co_u32_e64 v5, s[0:1], s4, v5
-; GFX9-NEXT:    v_mul_f32_e32 v2, v8, v1
-; GFX9-NEXT:    v_trunc_f32_e32 v2, v2
-; GFX9-NEXT:    v_addc_co_u32_e64 v6, s[0:1], v7, v6, s[0:1]
-; GFX9-NEXT:    v_cvt_u32_f32_e32 v7, v2
-; GFX9-NEXT:    v_add_u16_e32 v4, 1, v4
-; GFX9-NEXT:    v_mad_f32 v2, -v2, v0, v8
-; GFX9-NEXT:    v_cmp_ge_f32_e64 s[0:1], |v2|, v0
-; GFX9-NEXT:    v_cmp_eq_u16_e32 vcc, s3, v4
-; GFX9-NEXT:    v_addc_co_u32_e64 v2, s[0:1], 0, v7, s[0:1]
+; GFX9-NEXT:    v_and_b32_e32 v1, s2, v5
+; GFX9-NEXT:    v_cvt_f32_u32_e32 v8, v1
+; GFX9-NEXT:    v_lshlrev_b64 v[6:7], 1, v[1:2]
+; GFX9-NEXT:    v_add_u16_e32 v5, 1, v5
+; GFX9-NEXT:    v_add_co_u32_e64 v6, s[0:1], s4, v6
+; GFX9-NEXT:    v_mul_f32_e32 v1, v8, v3
+; GFX9-NEXT:    v_trunc_f32_e32 v1, v1
+; GFX9-NEXT:    v_cvt_u32_f32_e32 v9, v1
+; GFX9-NEXT:    v_addc_co_u32_e64 v7, s[0:1], v4, v7, s[0:1]
+; GFX9-NEXT:    v_mad_f32 v1, -v1, v0, v8
+; GFX9-NEXT:    v_cmp_ge_f32_e64 s[0:1], |v1|, v0
+; GFX9-NEXT:    v_cmp_eq_u16_e32 vcc, s3, v5
+; GFX9-NEXT:    v_addc_co_u32_e64 v1, s[0:1], 0, v9, s[0:1]
 ; GFX9-NEXT:    s_and_b64 vcc, exec, vcc
-; GFX9-NEXT:    global_store_short v[5:6], v2, off
+; GFX9-NEXT:    global_store_short v[6:7], v1, off
 ; GFX9-NEXT:    s_cbranch_vccz BB4_1
 ; GFX9-NEXT:  ; %bb.2: ; %bb2
 ; GFX9-NEXT:    s_endpgm
 bb:
   br label %bb3
 
 bb2:                                              ; preds = %bb3
   ret void

 }
 
 define amdgpu_kernel void @urem16_invariant_denom(i16 addrspace(1)* nocapture %arg, i16 %arg1) {
 ; GFX9-LABEL: urem16_invariant_denom:
 ; GFX9:       ; %bb.0: ; %bb
 ; GFX9-NEXT:    s_load_dword s3, s[0:1], 0x2c
 ; GFX9-NEXT:    s_mov_b32 s2, 0xffff
 ; GFX9-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
-; GFX9-NEXT:    v_mov_b32_e32 v3, 0
+; GFX9-NEXT:    v_mov_b32_e32 v2, 0
 ; GFX9-NEXT:    s_movk_i32 s6, 0x400
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    s_and_b32 s3, s2, s3
 ; GFX9-NEXT:    v_cvt_f32_u32_e32 v0, s3
-; GFX9-NEXT:    v_mov_b32_e32 v4, 0
-; GFX9-NEXT:    v_rcp_iflag_f32_e32 v1, v0
+; GFX9-NEXT:    v_mov_b32_e32 v4, s5
+; GFX9-NEXT:    v_mov_b32_e32 v5, 0
+; GFX9-NEXT:    v_rcp_iflag_f32_e32 v3, v0
 ; GFX9-NEXT:  BB5_1: ; %bb3
 ; GFX9-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GFX9-NEXT:    v_and_b32_e32 v2, s2, v4
-; GFX9-NEXT:    v_cvt_f32_u32_e32 v8, v2
-; GFX9-NEXT:    v_lshlrev_b64 v[5:6], 1, v[2:3]
-; GFX9-NEXT:    v_mov_b32_e32 v7, s5
-; GFX9-NEXT:    v_add_co_u32_e64 v5, s[0:1], s4, v5
-; GFX9-NEXT:    v_addc_co_u32_e64 v6, s[0:1], v7, v6, s[0:1]
-; GFX9-NEXT:    v_mul_f32_e32 v7, v8, v1
-; GFX9-NEXT:    v_trunc_f32_e32 v7, v7
-; GFX9-NEXT:    v_cvt_u32_f32_e32 v9, v7
-; GFX9-NEXT:    v_mad_f32 v7, -v7, v0, v8
-; GFX9-NEXT:    v_cmp_ge_f32_e64 s[0:1], |v7|, v0
-; GFX9-NEXT:    v_add_u16_e32 v4, 1, v4
-; GFX9-NEXT:    v_addc_co_u32_e64 v7, s[0:1], 0, v9, s[0:1]
-; GFX9-NEXT:    v_mul_lo_u32 v7, v7, s3
-; GFX9-NEXT:    v_cmp_eq_u16_e32 vcc, s6, v4
+; GFX9-NEXT:    v_and_b32_e32 v1, s2, v5
+; GFX9-NEXT:    v_cvt_f32_u32_e32 v8, v1
+; GFX9-NEXT:    v_lshlrev_b64 v[6:7], 1, v[1:2]
+; GFX9-NEXT:    v_add_u16_e32 v5, 1, v5
+; GFX9-NEXT:    v_add_co_u32_e64 v6, s[0:1], s4, v6
+; GFX9-NEXT:    v_mul_f32_e32 v9, v8, v3
+; GFX9-NEXT:    v_trunc_f32_e32 v9, v9
+; GFX9-NEXT:    v_cvt_u32_f32_e32 v10, v9
+; GFX9-NEXT:    v_addc_co_u32_e64 v7, s[0:1], v4, v7, s[0:1]
+; GFX9-NEXT:    v_mad_f32 v8, -v9, v0, v8
+; GFX9-NEXT:    v_cmp_ge_f32_e64 s[0:1], |v8|, v0
+; GFX9-NEXT:    v_addc_co_u32_e64 v8, s[0:1], 0, v10, s[0:1]
+; GFX9-NEXT:    v_mul_lo_u32 v8, v8, s3
+; GFX9-NEXT:    v_cmp_eq_u16_e32 vcc, s6, v5
 ; GFX9-NEXT:    s_and_b64 vcc, exec, vcc
-; GFX9-NEXT:    v_sub_u32_e32 v2, v2, v7
-; GFX9-NEXT:    global_store_short v[5:6], v2, off
+; GFX9-NEXT:    v_sub_u32_e32 v1, v1, v8
+; GFX9-NEXT:    global_store_short v[6:7], v1, off
 ; GFX9-NEXT:    s_cbranch_vccz BB5_1
 ; GFX9-NEXT:  ; %bb.2: ; %bb2
 ; GFX9-NEXT:    s_endpgm
 bb:
   br label %bb3
 
 bb2:                                              ; preds = %bb3
   ret void
bb3:                                              ; preds = %bb3, %bb
   br i1 %tmp8, label %bb2, label %bb3
 }
 
 define amdgpu_kernel void @sdiv16_invariant_denom(i16 addrspace(1)* nocapture %arg, i16 %arg1) {
 ; GFX9-LABEL: sdiv16_invariant_denom:
 ; GFX9:       ; %bb.0: ; %bb
 ; GFX9-NEXT:    s_load_dword s2, s[0:1], 0x2c
 ; GFX9-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
-; GFX9-NEXT:    v_mov_b32_e32 v3, 0
+; GFX9-NEXT:    v_mov_b32_e32 v2, 0
 ; GFX9-NEXT:    s_movk_i32 s3, 0x400
-; GFX9-NEXT:    v_mov_b32_e32 v4, 0
+; GFX9-NEXT:    v_mov_b32_e32 v5, 0
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    s_sext_i32_i16 s2, s2
 ; GFX9-NEXT:    v_cvt_f32_i32_e32 v0, s2
-; GFX9-NEXT:    v_rcp_iflag_f32_e32 v1, v0
+; GFX9-NEXT:    v_mov_b32_e32 v4, s5
+; GFX9-NEXT:    v_rcp_iflag_f32_e32 v3, v0
 ; GFX9-NEXT:  BB6_1: ; %bb3
 ; GFX9-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GFX9-NEXT:    v_bfe_i32 v5, v4, 0, 16
-; GFX9-NEXT:    v_and_b32_e32 v2, 0xffff, v4
-; GFX9-NEXT:    v_cvt_f32_i32_e32 v9, v5
-; GFX9-NEXT:    v_xor_b32_e32 v8, s2, v5
-; GFX9-NEXT:    v_lshlrev_b64 v[5:6], 1, v[2:3]
-; GFX9-NEXT:    v_mov_b32_e32 v7, s5
-; GFX9-NEXT:    v_add_co_u32_e64 v5, s[0:1], s4, v5
-; GFX9-NEXT:    v_addc_co_u32_e64 v6, s[0:1], v7, v6, s[0:1]
-; GFX9-NEXT:    v_mul_f32_e32 v7, v9, v1
-; GFX9-NEXT:    v_trunc_f32_e32 v7, v7
-; GFX9-NEXT:    v_ashrrev_i32_e32 v2, 30, v8
-; GFX9-NEXT:    v_cvt_i32_f32_e32 v8, v7
-; GFX9-NEXT:    v_mad_f32 v7, -v7, v0, v9
-; GFX9-NEXT:    v_add_u16_e32 v4, 1, v4
-; GFX9-NEXT:    v_or_b32_e32 v2, 1, v2
-; GFX9-NEXT:    v_cmp_ge_f32_e64 s[0:1], |v7|, |v0|
-; GFX9-NEXT:    v_cmp_eq_u16_e32 vcc, s3, v4
-; GFX9-NEXT:    v_cndmask_b32_e64 v2, 0, v2, s[0:1]
-; GFX9-NEXT:    v_add_u32_e32 v2, v8, v2
+; GFX9-NEXT:    v_bfe_i32 v6, v5, 0, 16
+; GFX9-NEXT:    v_cvt_f32_i32_e32 v9, v6
+; GFX9-NEXT:    v_and_b32_e32 v1, 0xffff, v5
+; GFX9-NEXT:    v_xor_b32_e32 v8, s2, v6
+; GFX9-NEXT:    v_lshlrev_b64 v[6:7], 1, v[1:2]
+; GFX9-NEXT:    v_ashrrev_i32_e32 v1, 30, v8
+; GFX9-NEXT:    v_mul_f32_e32 v8, v9, v3
+; GFX9-NEXT:    v_trunc_f32_e32 v8, v8
+; GFX9-NEXT:    v_add_co_u32_e64 v6, s[0:1], s4, v6
+; GFX9-NEXT:    v_cvt_i32_f32_e32 v10, v8
+; GFX9-NEXT:    v_addc_co_u32_e64 v7, s[0:1], v4, v7, s[0:1]
+; GFX9-NEXT:    v_mad_f32 v8, -v8, v0, v9
+; GFX9-NEXT:    v_add_u16_e32 v5, 1, v5
+; GFX9-NEXT:    v_or_b32_e32 v1, 1, v1
+; GFX9-NEXT:    v_cmp_ge_f32_e64 s[0:1], |v8|, |v0|
+; GFX9-NEXT:    v_cmp_eq_u16_e32 vcc, s3, v5
+; GFX9-NEXT:    v_cndmask_b32_e64 v1, 0, v1, s[0:1]
+; GFX9-NEXT:    v_add_u32_e32 v1, v10, v1
 ; GFX9-NEXT:    s_and_b64 vcc, exec, vcc
-; GFX9-NEXT:    global_store_short v[5:6], v2, off
+; GFX9-NEXT:    global_store_short v[6:7], v1, off
 ; GFX9-NEXT:    s_cbranch_vccz BB6_1
 ; GFX9-NEXT:  ; %bb.2: ; %bb2
 ; GFX9-NEXT:    s_endpgm
 bb:
   br label %bb3
 
 bb2:                                              ; preds = %bb3
   ret void
bb3:                                              ; preds = %bb3, %bb
   br i1 %tmp8, label %bb2, label %bb3
 }
 
 define amdgpu_kernel void @srem16_invariant_denom(i16 addrspace(1)* nocapture %arg, i16 %arg1) {
 ; GFX9-LABEL: srem16_invariant_denom:
 ; GFX9:       ; %bb.0: ; %bb
 ; GFX9-NEXT:    s_load_dword s2, s[0:1], 0x2c
 ; GFX9-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x24
-; GFX9-NEXT:    v_mov_b32_e32 v3, 0
+; GFX9-NEXT:    v_mov_b32_e32 v2, 0
 ; GFX9-NEXT:    s_movk_i32 s3, 0x400
-; GFX9-NEXT:    v_mov_b32_e32 v4, 0
+; GFX9-NEXT:    v_mov_b32_e32 v5, 0
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    s_sext_i32_i16 s2, s2
 ; GFX9-NEXT:    v_cvt_f32_i32_e32 v0, s2
-; GFX9-NEXT:    v_rcp_iflag_f32_e32 v1, v0
+; GFX9-NEXT:    v_mov_b32_e32 v4, s5
+; GFX9-NEXT:    v_rcp_iflag_f32_e32 v3, v0
 ; GFX9-NEXT:  BB7_1: ; %bb3
 ; GFX9-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GFX9-NEXT:    v_bfe_i32 v7, v4, 0, 16
-; GFX9-NEXT:    v_and_b32_e32 v2, 0xffff, v4
-; GFX9-NEXT:    v_cvt_f32_i32_e32 v10, v7
-; GFX9-NEXT:    v_lshlrev_b64 v[5:6], 1, v[2:3]
-; GFX9-NEXT:    v_mov_b32_e32 v8, s5
-; GFX9-NEXT:    v_add_co_u32_e64 v5, s[0:1], s4, v5
-; GFX9-NEXT:    v_addc_co_u32_e64 v6, s[0:1], v8, v6, s[0:1]
-; GFX9-NEXT:    v_mul_f32_e32 v8, v10, v1
-; GFX9-NEXT:    v_xor_b32_e32 v9, s2, v7
-; GFX9-NEXT:    v_trunc_f32_e32 v8, v8
-; GFX9-NEXT:    v_ashrrev_i32_e32 v2, 30, v9
-; GFX9-NEXT:    v_cvt_i32_f32_e32 v9, v8
-; GFX9-NEXT:    v_mad_f32 v8, -v8, v0, v10
-; GFX9-NEXT:    v_or_b32_e32 v2, 1, v2
-; GFX9-NEXT:    v_cmp_ge_f32_e64 s[0:1], |v8|, |v0|
-; GFX9-NEXT:    v_cndmask_b32_e64 v2, 0, v2, s[0:1]
-; GFX9-NEXT:    v_add_u32_e32 v2, v9, v2
-; GFX9-NEXT:    v_mul_lo_u32 v2, v2, s2
-; GFX9-NEXT:    v_add_u16_e32 v4, 1, v4
-; GFX9-NEXT:    v_cmp_eq_u16_e32 vcc, s3, v4
+; GFX9-NEXT:    v_bfe_i32 v8, v5, 0, 16
+; GFX9-NEXT:    v_cvt_f32_i32_e32 v10, v8
+; GFX9-NEXT:    v_and_b32_e32 v1, 0xffff, v5
+; GFX9-NEXT:    v_xor_b32_e32 v9, s2, v8
+; GFX9-NEXT:    v_lshlrev_b64 v[6:7], 1, v[1:2]
+; GFX9-NEXT:    v_ashrrev_i32_e32 v1, 30, v9
+; GFX9-NEXT:    v_mul_f32_e32 v9, v10, v3
+; GFX9-NEXT:    v_trunc_f32_e32 v9, v9
+; GFX9-NEXT:    v_add_co_u32_e64 v6, s[0:1], s4, v6
+; GFX9-NEXT:    v_cvt_i32_f32_e32 v11, v9
+; GFX9-NEXT:    v_addc_co_u32_e64 v7, s[0:1], v4, v7, s[0:1]
+; GFX9-NEXT:    v_mad_f32 v9, -v9, v0, v10
+; GFX9-NEXT:    v_or_b32_e32 v1, 1, v1
+; GFX9-NEXT:    v_cmp_ge_f32_e64 s[0:1], |v9|, |v0|
+; GFX9-NEXT:    v_cndmask_b32_e64 v1, 0, v1, s[0:1]
+; GFX9-NEXT:    v_add_u32_e32 v1, v11, v1
+; GFX9-NEXT:    v_mul_lo_u32 v1, v1, s2
+; GFX9-NEXT:    v_add_u16_e32 v5, 1, v5
+; GFX9-NEXT:    v_cmp_eq_u16_e32 vcc, s3, v5
 ; GFX9-NEXT:    s_and_b64 vcc, exec, vcc
-; GFX9-NEXT:    v_sub_u32_e32 v2, v7, v2
-; GFX9-NEXT:    global_store_short v[5:6], v2, off
+; GFX9-NEXT:    v_sub_u32_e32 v1, v8, v1
+; GFX9-NEXT:    global_store_short v[6:7], v1, off
 ; GFX9-NEXT:    s_cbranch_vccz BB7_1
 ; GFX9-NEXT:  ; %bb.2: ; %bb2
 ; GFX9-NEXT:    s_endpgm
 bb:
   br label %bb3
 
 bb2:                                              ; preds = %bb3
   ret void

llvm/test/CodeGen/AMDGPU/sdiv64.ll

 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], s[10:11], v0
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB0_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB0_6
 ; GCN-IR-NEXT:  BB0_4: ; %udiv-preheader
-; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], s[10:11], v4
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], s[10:11], v4
 ; GCN-IR-NEXT:    s_add_u32 s10, s6, -1
-; GCN-IR-NEXT:    v_add_i32_e32 v4, vcc, v2, v3
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
+; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
+; GCN-IR-NEXT:    v_add_i32_e32 v5, vcc, v2, v3
 ; GCN-IR-NEXT:    s_addc_u32 s11, s7, -1
-; GCN-IR-NEXT:    v_addc_u32_e64 v5, s[0:1], -1, 0, vcc
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_addc_u32_e64 v6, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s11
 ; GCN-IR-NEXT:  BB0_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s11
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s10, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s6, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s7, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s10, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s6, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s7, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB0_5
 ; GCN-IR-NEXT:  BB0_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB0_7: ; %udiv-end
 ; GCN-IR-NEXT:    s_xor_b64 s[0:1], s[8:9], s[2:3]
 ; GCN-IR-NEXT:    v_xor_b32_e32 v0, s0, v0

 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], s[10:11], v0
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB9_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB9_6
 ; GCN-IR-NEXT:  BB9_4: ; %udiv-preheader
-; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], s[10:11], v4
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], s[10:11], v4
 ; GCN-IR-NEXT:    s_add_u32 s10, s8, -1
-; GCN-IR-NEXT:    v_add_i32_e32 v4, vcc, v2, v3
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
+; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
+; GCN-IR-NEXT:    v_add_i32_e32 v5, vcc, v2, v3
 ; GCN-IR-NEXT:    s_addc_u32 s11, s9, -1
-; GCN-IR-NEXT:    v_addc_u32_e64 v5, s[0:1], -1, 0, vcc
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_addc_u32_e64 v6, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s11
 ; GCN-IR-NEXT:  BB9_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s11
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s10, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s8, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s9, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s10, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s8, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s9, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB9_5
 ; GCN-IR-NEXT:  BB9_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB9_7: ; %udiv-end
 ; GCN-IR-NEXT:    s_xor_b64 s[0:1], s[6:7], s[2:3]
 ; GCN-IR-NEXT:    v_xor_b32_e32 v0, s0, v0

 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB10_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB10_6
 ; GCN-IR-NEXT:  BB10_4: ; %udiv-preheader
 ; GCN-IR-NEXT:    s_add_u32 s7, s8, -1
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], 24, v3
-; GCN-IR-NEXT:    v_sub_i32_e32 v4, vcc, 58, v2
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], 24, v3
 ; GCN-IR-NEXT:    s_addc_u32 s10, s9, -1
-; GCN-IR-NEXT:    v_subb_u32_e64 v5, s[0:1], 0, 0, vcc
+; GCN-IR-NEXT:    v_sub_i32_e32 v5, vcc, 58, v2
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_subb_u32_e64 v6, s[0:1], 0, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s10
 ; GCN-IR-NEXT:  BB10_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s10
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s7, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s8, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s9, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_or_b32_e32 v8, v8, v3
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s7, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s8, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s9, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB10_5
 ; GCN-IR-NEXT:  BB10_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB10_7: ; %udiv-end
 ; GCN-IR-NEXT:    v_xor_b32_e32 v0, s2, v0
 ; GCN-IR-NEXT:    v_xor_b32_e32 v1, s3, v1

 ; GCN-IR-NEXT:    s_xor_b64 s[10:11], exec, s[4:5]
 ; GCN-IR-NEXT:    s_cbranch_execz BB13_5
 ; GCN-IR-NEXT:  ; %bb.2: ; %udiv-preheader
 ; GCN-IR-NEXT:    v_lshr_b64 v[9:10], v[7:8], v9
 ; GCN-IR-NEXT:    v_add_i32_e32 v7, vcc, 0xffffffcf, v0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v11, 0
 ; GCN-IR-NEXT:    v_addc_u32_e64 v8, s[4:5], 0, -1, vcc
 ; GCN-IR-NEXT:    v_mov_b32_e32 v12, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v0, 31
 ; GCN-IR-NEXT:    s_movk_i32 s12, 0x7fff
 ; GCN-IR-NEXT:  BB13_3: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
 ; GCN-IR-NEXT:    v_lshl_b64 v[9:10], v[9:10], 1
-; GCN-IR-NEXT:    v_lshrrev_b32_e32 v0, 31, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_lshr_b32_e32 v5, v4, v0
+; GCN-IR-NEXT:    v_or_b32_e32 v9, v9, v5
 ; GCN-IR-NEXT:    v_lshl_b64 v[3:4], v[3:4], 1
-; GCN-IR-NEXT:    v_sub_i32_e32 v5, vcc, s12, v0
+; GCN-IR-NEXT:    v_sub_i32_e32 v5, vcc, s12, v9
 ; GCN-IR-NEXT:    v_subb_u32_e32 v5, vcc, 0, v10, vcc
 ; GCN-IR-NEXT:    v_or_b32_e32 v3, v11, v3
+; GCN-IR-NEXT:    v_ashr_i32_e32 v11, v5, v0
+; GCN-IR-NEXT:    v_and_b32_e32 v13, 0x8000, v11
+; GCN-IR-NEXT:    v_and_b32_e32 v5, 1, v11
 ; GCN-IR-NEXT:    v_add_i32_e32 v11, vcc, 1, v7
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v5
 ; GCN-IR-NEXT:    v_or_b32_e32 v4, v12, v4
 ; GCN-IR-NEXT:    v_addc_u32_e32 v12, vcc, 0, v8, vcc
 ; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[11:12], v[7:8]
-; GCN-IR-NEXT:    v_and_b32_e32 v5, 1, v9
-; GCN-IR-NEXT:    v_and_b32_e32 v9, 0x8000, v9
 ; GCN-IR-NEXT:    v_mov_b32_e32 v7, v11
 ; GCN-IR-NEXT:    v_mov_b32_e32 v6, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v8, v12
 ; GCN-IR-NEXT:    v_mov_b32_e32 v12, v6
-; GCN-IR-NEXT:    v_mov_b32_e32 v13, 0
-; GCN-IR-NEXT:    v_sub_i32_e64 v9, s[4:5], v0, v9
-; GCN-IR-NEXT:    v_subb_u32_e64 v10, s[4:5], v10, v13, s[4:5]
+; GCN-IR-NEXT:    v_mov_b32_e32 v14, 0
+; GCN-IR-NEXT:    v_sub_i32_e64 v9, s[4:5], v9, v13
+; GCN-IR-NEXT:    v_subb_u32_e64 v10, s[4:5], v10, v14, s[4:5]
 ; GCN-IR-NEXT:    s_or_b64 s[8:9], vcc, s[8:9]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v11, v5
 ; GCN-IR-NEXT:    s_andn2_b64 exec, exec, s[8:9]
 ; GCN-IR-NEXT:    s_cbranch_execnz BB13_3
 ; GCN-IR-NEXT:  ; %bb.4: ; %Flow
 ; GCN-IR-NEXT:    s_or_b64 exec, exec, s[8:9]
 ; GCN-IR-NEXT:  BB13_5: ; %Flow3
 ; GCN-IR-NEXT:    s_or_b64 exec, exec, s[10:11]

llvm/test/CodeGen/AMDGPU/sdwa-peephole.ll

entry:
   store <8 x i8> %tmp19, <8 x i8> addrspace(1)* %arrayidx5, align 8
   ret void
 }
 
 ; GCN-LABEL: {{^}}sdwa_crash_inlineasm_def:
 ; GCN: s_mov_b32 s{{[0-9]+}}, 0xffff
 ; GCN: v_and_b32_e32 v{{[0-9]+}}, s{{[0-9]+}}, v{{[0-9]+}}
 ;
-; TODO: Why is the constant not peepholed into the v_or_b32_e32?
-;
-; NOSDWA: s_mov_b32 [[CONST:s[0-9]+]], 0x10000
-; NOSDWA: v_or_b32_e32 v{{[0-9]+}}, s0,
+; NOSDWA: v_or_b32_e32 v{{[0-9]+}}, 0x10000,
 ; SDWA: v_or_b32_e32 v{{[0-9]+}}, 0x10000,
 define amdgpu_kernel void @sdwa_crash_inlineasm_def() #0 {
 bb:
   br label %bb1
 
 bb1:                                              ; preds = %bb11, %bb
   %tmp = phi <2 x i32> [ %tmp12, %bb11 ], [ undef, %bb ]
   br i1 true, label %bb2, label %bb11

llvm/test/CodeGen/AMDGPU/srem64.ll

 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], s[6:7], v0
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB0_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB0_6
 ; GCN-IR-NEXT:  BB0_4: ; %udiv-preheader
-; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
 ; GCN-IR-NEXT:    s_add_u32 s8, s2, -1
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], s[6:7], v4
-; GCN-IR-NEXT:    v_add_i32_e32 v4, vcc, v2, v3
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
-; GCN-IR-NEXT:    s_addc_u32 s9, s3, -1
-; GCN-IR-NEXT:    v_addc_u32_e64 v5, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
+; GCN-IR-NEXT:    v_add_i32_e32 v5, vcc, v2, v3
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], s[6:7], v4
+; GCN-IR-NEXT:    s_addc_u32 s9, s3, -1
+; GCN-IR-NEXT:    v_addc_u32_e64 v6, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s9
 ; GCN-IR-NEXT:  BB0_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s9
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s8, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s2, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s3, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s8, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s2, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s3, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB0_5
 ; GCN-IR-NEXT:  BB0_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB0_7: ; %udiv-end
 ; GCN-IR-NEXT:    v_mul_lo_u32 v1, s2, v1
 ; GCN-IR-NEXT:    v_mul_hi_u32 v2, s2, v0

 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], s[8:9], v0
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB8_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB8_6
 ; GCN-IR-NEXT:  BB8_4: ; %udiv-preheader
-; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
 ; GCN-IR-NEXT:    s_add_u32 s6, s10, -1
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], s[8:9], v4
-; GCN-IR-NEXT:    v_add_i32_e32 v4, vcc, v2, v3
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
-; GCN-IR-NEXT:    s_addc_u32 s7, s11, -1
-; GCN-IR-NEXT:    v_addc_u32_e64 v5, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
+; GCN-IR-NEXT:    v_add_i32_e32 v5, vcc, v2, v3
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], s[8:9], v4
+; GCN-IR-NEXT:    s_addc_u32 s7, s11, -1
+; GCN-IR-NEXT:    v_addc_u32_e64 v6, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s7
 ; GCN-IR-NEXT:  BB8_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s7
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s6, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s10, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s11, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s6, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s10, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s11, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB8_5
 ; GCN-IR-NEXT:  BB8_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB8_7: ; %udiv-end
 ; GCN-IR-NEXT:    v_mul_lo_u32 v1, s10, v1
 ; GCN-IR-NEXT:    v_mul_hi_u32 v2, s10, v0

 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], s[6:7], v0
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB9_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB9_6
 ; GCN-IR-NEXT:  BB9_4: ; %udiv-preheader
-; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
 ; GCN-IR-NEXT:    s_add_u32 s10, s8, -1
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], s[6:7], v4
-; GCN-IR-NEXT:    v_add_i32_e32 v4, vcc, v2, v3
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
-; GCN-IR-NEXT:    s_addc_u32 s11, s9, -1
-; GCN-IR-NEXT:    v_addc_u32_e64 v5, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
+; GCN-IR-NEXT:    v_add_i32_e32 v5, vcc, v2, v3
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], s[6:7], v4
+; GCN-IR-NEXT:    s_addc_u32 s11, s9, -1
+; GCN-IR-NEXT:    v_addc_u32_e64 v6, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s11
 ; GCN-IR-NEXT:  BB9_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s11
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s10, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s8, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s9, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s10, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s8, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s9, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB9_5
 ; GCN-IR-NEXT:  BB9_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB9_7: ; %udiv-end
 ; GCN-IR-NEXT:    v_mul_lo_u32 v1, s8, v1
 ; GCN-IR-NEXT:    v_mul_hi_u32 v2, s8, v0

 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB10_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB10_6
 ; GCN-IR-NEXT:  BB10_4: ; %udiv-preheader
 ; GCN-IR-NEXT:    s_add_u32 s7, s2, -1
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], 24, v3
-; GCN-IR-NEXT:    v_sub_i32_e32 v4, vcc, 58, v2
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], 24, v3
 ; GCN-IR-NEXT:    s_addc_u32 s8, s3, -1
-; GCN-IR-NEXT:    v_subb_u32_e64 v5, s[0:1], 0, 0, vcc
+; GCN-IR-NEXT:    v_sub_i32_e32 v5, vcc, 58, v2
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_subb_u32_e64 v6, s[0:1], 0, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s8
 ; GCN-IR-NEXT:  BB10_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s8
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s7, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s2, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s3, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_or_b32_e32 v8, v8, v3
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s7, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s2, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s3, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB10_5
 ; GCN-IR-NEXT:  BB10_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB10_7: ; %udiv-end
 ; GCN-IR-NEXT:    v_mul_lo_u32 v1, s2, v1
 ; GCN-IR-NEXT:    v_mul_hi_u32 v2, s2, v0

 ; GCN-IR-NEXT:    v_mov_b32_e32 v6, 0
 ; GCN-IR-NEXT:    v_lshl_b64 v[4:5], v[0:1], v4
 ; GCN-IR-NEXT:    s_mov_b64 s[8:9], 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v7, 0
 ; GCN-IR-NEXT:    s_and_saveexec_b64 s[4:5], vcc
 ; GCN-IR-NEXT:    s_xor_b64 s[10:11], exec, s[4:5]
 ; GCN-IR-NEXT:    s_cbranch_execz BB13_5
 ; GCN-IR-NEXT:  ; %bb.2: ; %udiv-preheader
-; GCN-IR-NEXT:    v_mov_b32_e32 v12, 0
-; GCN-IR-NEXT:    v_lshr_b64 v[10:11], v[0:1], v9
+; GCN-IR-NEXT:    v_mov_b32_e32 v13, 0
+; GCN-IR-NEXT:    v_lshr_b64 v[11:12], v[0:1], v9
 ; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 0xffffffcf, v8
 ; GCN-IR-NEXT:    v_addc_u32_e64 v9, s[4:5], 0, -1, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v13, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v14, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 31
 ; GCN-IR-NEXT:    s_movk_i32 s12, 0x7fff
 ; GCN-IR-NEXT:  BB13_3: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[10:11], v[10:11], 1
-; GCN-IR-NEXT:    v_lshrrev_b32_e32 v6, 31, v5
-; GCN-IR-NEXT:    v_or_b32_e32 v10, v10, v6
+; GCN-IR-NEXT:    v_lshl_b64 v[11:12], v[11:12], 1
+; GCN-IR-NEXT:    v_lshr_b32_e32 v6, v5, v10
+; GCN-IR-NEXT:    v_or_b32_e32 v11, v11, v6
 ; GCN-IR-NEXT:    v_lshl_b64 v[4:5], v[4:5], 1
-; GCN-IR-NEXT:    v_sub_i32_e32 v6, vcc, s12, v10
-; GCN-IR-NEXT:    v_subb_u32_e32 v6, vcc, 0, v11, vcc
-; GCN-IR-NEXT:    v_or_b32_e32 v4, v12, v4
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v12, 31, v6
-; GCN-IR-NEXT:    v_and_b32_e32 v14, 0x8000, v12
-; GCN-IR-NEXT:    v_and_b32_e32 v6, 1, v12
-; GCN-IR-NEXT:    v_add_i32_e32 v12, vcc, 1, v8
-; GCN-IR-NEXT:    v_or_b32_e32 v5, v13, v5
-; GCN-IR-NEXT:    v_addc_u32_e32 v13, vcc, 0, v9, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[12:13], v[8:9]
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v12
+; GCN-IR-NEXT:    v_sub_i32_e32 v6, vcc, s12, v11
+; GCN-IR-NEXT:    v_subb_u32_e32 v6, vcc, 0, v12, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v4, v13, v4
+; GCN-IR-NEXT:    v_ashr_i32_e32 v13, v6, v10
+; GCN-IR-NEXT:    v_and_b32_e32 v15, 0x8000, v13
+; GCN-IR-NEXT:    v_and_b32_e32 v6, 1, v13
+; GCN-IR-NEXT:    v_add_i32_e32 v13, vcc, 1, v8
+; GCN-IR-NEXT:    v_or_b32_e32 v5, v14, v5
+; GCN-IR-NEXT:    v_addc_u32_e32 v14, vcc, 0, v9, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[13:14], v[8:9]
+; GCN-IR-NEXT:    v_mov_b32_e32 v8, v13
 ; GCN-IR-NEXT:    v_mov_b32_e32 v7, 0
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v13
-; GCN-IR-NEXT:    v_mov_b32_e32 v13, v7
-; GCN-IR-NEXT:    v_mov_b32_e32 v15, 0
-; GCN-IR-NEXT:    v_sub_i32_e64 v10, s[4:5], v10, v14
-; GCN-IR-NEXT:    v_subb_u32_e64 v11, s[4:5], v11, v15, s[4:5]
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v14
+; GCN-IR-NEXT:    v_mov_b32_e32 v14, v7
+; GCN-IR-NEXT:    v_mov_b32_e32 v16, 0
+; GCN-IR-NEXT:    v_sub_i32_e64 v11, s[4:5], v11, v15
+; GCN-IR-NEXT:    v_subb_u32_e64 v12, s[4:5], v12, v16, s[4:5]
 ; GCN-IR-NEXT:    s_or_b64 s[8:9], vcc, s[8:9]
-; GCN-IR-NEXT:    v_mov_b32_e32 v12, v6
+; GCN-IR-NEXT:    v_mov_b32_e32 v13, v6
 ; GCN-IR-NEXT:    s_andn2_b64 exec, exec, s[8:9]
 ; GCN-IR-NEXT:    s_cbranch_execnz BB13_3
 ; GCN-IR-NEXT:  ; %bb.4: ; %Flow
 ; GCN-IR-NEXT:    s_or_b64 exec, exec, s[8:9]
 ; GCN-IR-NEXT:  BB13_5: ; %Flow3
 ; GCN-IR-NEXT:    s_or_b64 exec, exec, s[10:11]
 ; GCN-IR-NEXT:    v_lshl_b64 v[4:5], v[4:5], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v5

llvm/test/CodeGen/AMDGPU/udiv64.ll

 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], s[6:7], v0
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB0_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB0_6
 ; GCN-IR-NEXT:  BB0_4: ; %udiv-preheader
-; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], s[6:7], v4
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], s[6:7], v4
 ; GCN-IR-NEXT:    s_add_u32 s6, s2, -1
-; GCN-IR-NEXT:    v_add_i32_e32 v4, vcc, v2, v3
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
+; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
+; GCN-IR-NEXT:    v_add_i32_e32 v5, vcc, v2, v3
 ; GCN-IR-NEXT:    s_addc_u32 s7, s3, -1
-; GCN-IR-NEXT:    v_addc_u32_e64 v5, s[0:1], -1, 0, vcc
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_addc_u32_e64 v6, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s7
 ; GCN-IR-NEXT:  BB0_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s7
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s6, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s2, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s3, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s6, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s2, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s3, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB0_5
 ; GCN-IR-NEXT:  BB0_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB0_7: ; %udiv-end
 ; GCN-IR-NEXT:    s_mov_b32 s7, 0xf000
 ; GCN-IR-NEXT:    s_mov_b32 s6, -1

 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], s[6:7], v0
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB7_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB7_6
 ; GCN-IR-NEXT:  BB7_4: ; %udiv-preheader
-; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], s[6:7], v4
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], s[6:7], v4
 ; GCN-IR-NEXT:    s_add_u32 s6, s2, -1
-; GCN-IR-NEXT:    v_add_i32_e32 v4, vcc, v2, v3
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
+; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
+; GCN-IR-NEXT:    v_add_i32_e32 v5, vcc, v2, v3
 ; GCN-IR-NEXT:    s_addc_u32 s7, s3, -1
-; GCN-IR-NEXT:    v_addc_u32_e64 v5, s[0:1], -1, 0, vcc
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_addc_u32_e64 v6, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s7
 ; GCN-IR-NEXT:  BB7_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s7
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s6, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s2, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s3, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s6, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s2, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s3, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB7_5
 ; GCN-IR-NEXT:  BB7_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB7_7: ; %udiv-end
 ; GCN-IR-NEXT:    s_mov_b32 s7, 0xf000
 ; GCN-IR-NEXT:    s_mov_b32 s6, -1

 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB8_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB8_6
 ; GCN-IR-NEXT:  BB8_4: ; %udiv-preheader
 ; GCN-IR-NEXT:    s_add_u32 s3, s6, -1
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], 24, v3
-; GCN-IR-NEXT:    v_sub_i32_e32 v4, vcc, 58, v2
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], 24, v3
 ; GCN-IR-NEXT:    s_addc_u32 s8, s7, -1
-; GCN-IR-NEXT:    v_subb_u32_e64 v5, s[0:1], 0, 0, vcc
+; GCN-IR-NEXT:    v_sub_i32_e32 v5, vcc, 58, v2
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_subb_u32_e64 v6, s[0:1], 0, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s8
 ; GCN-IR-NEXT:  BB8_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s8
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s3, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s6, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s7, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_or_b32_e32 v8, v8, v3
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s3, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s6, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s7, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB8_5
 ; GCN-IR-NEXT:  BB8_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB8_7: ; %udiv-end
 ; GCN-IR-NEXT:    s_mov_b32 s7, 0xf000
 ; GCN-IR-NEXT:    s_mov_b32 s6, s2

 ; GCN-IR-NEXT:    s_xor_b64 s[10:11], exec, s[4:5]
 ; GCN-IR-NEXT:    s_cbranch_execz BB10_5
 ; GCN-IR-NEXT:  ; %bb.2: ; %udiv-preheader
 ; GCN-IR-NEXT:    v_lshr_b64 v[7:8], v[0:1], v7
 ; GCN-IR-NEXT:    v_add_i32_e32 v0, vcc, 0xffffffcf, v6
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
 ; GCN-IR-NEXT:    v_addc_u32_e64 v1, s[4:5], 0, -1, vcc
 ; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, 31
 ; GCN-IR-NEXT:    s_movk_i32 s12, 0x7fff
 ; GCN-IR-NEXT:  BB10_3: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
 ; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
-; GCN-IR-NEXT:    v_lshrrev_b32_e32 v4, 31, v3
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v7, v4
+; GCN-IR-NEXT:    v_lshr_b32_e32 v4, v3, v6
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v4
 ; GCN-IR-NEXT:    v_lshl_b64 v[2:3], v[2:3], 1
-; GCN-IR-NEXT:    v_sub_i32_e32 v4, vcc, s12, v6
+; GCN-IR-NEXT:    v_sub_i32_e32 v4, vcc, s12, v7
 ; GCN-IR-NEXT:    v_subb_u32_e32 v4, vcc, 0, v8, vcc
 ; GCN-IR-NEXT:    v_or_b32_e32 v2, v9, v2
+; GCN-IR-NEXT:    v_ashr_i32_e32 v9, v4, v6
+; GCN-IR-NEXT:    v_and_b32_e32 v11, 0x8000, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v4, 1, v9
 ; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v0
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v7, 31, v4
 ; GCN-IR-NEXT:    v_or_b32_e32 v3, v10, v3
 ; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v1, vcc
 ; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[0:1]
-; GCN-IR-NEXT:    v_and_b32_e32 v4, 1, v7
-; GCN-IR-NEXT:    v_and_b32_e32 v7, 0x8000, v7
 ; GCN-IR-NEXT:    v_mov_b32_e32 v0, v9
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v1, v10
 ; GCN-IR-NEXT:    v_mov_b32_e32 v10, v5
-; GCN-IR-NEXT:    v_mov_b32_e32 v11, 0
-; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[4:5], v6, v7
-; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[4:5], v8, v11, s[4:5]
+; GCN-IR-NEXT:    v_mov_b32_e32 v12, 0
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[4:5], v7, v11
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[4:5], v8, v12, s[4:5]
 ; GCN-IR-NEXT:    s_or_b64 s[8:9], vcc, s[8:9]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, v4
 ; GCN-IR-NEXT:    s_andn2_b64 exec, exec, s[8:9]
 ; GCN-IR-NEXT:    s_cbranch_execnz BB10_3
 ; GCN-IR-NEXT:  ; %bb.4: ; %Flow
 ; GCN-IR-NEXT:    s_or_b64 exec, exec, s[8:9]
 ; GCN-IR-NEXT:  BB10_5: ; %Flow3
 ; GCN-IR-NEXT:    s_or_b64 exec, exec, s[10:11]

 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:  BB11_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
 ; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
 ; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v3
 ; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, 23, v6
 ; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, 0, v7, vcc
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
 ; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v10, 24, v8
 ; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v8, 24, v8
-; GCN-IR-NEXT:    v_sub_i32_e32 v6, vcc, v6, v8
-; GCN-IR-NEXT:    v_add_i32_e64 v8, s[0:1], 1, v4
+; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e64 v9, s[0:1], 0, v5, s[0:1]
-; GCN-IR-NEXT:    v_cmp_lt_u64_e64 s[0:1], v[8:9], v[4:5]
+; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_subbrev_u32_e32 v7, vcc, 0, v7, vcc
+; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
+; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v3, s[0:1]
+; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
 ; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB11_5
 ; GCN-IR-NEXT:  BB11_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB11_7: ; %udiv-end
 ; GCN-IR-NEXT:    s_mov_b32 s7, 0xf000

 ; GCN-IR-NEXT:    s_xor_b64 s[10:11], exec, s[4:5]
 ; GCN-IR-NEXT:    s_cbranch_execz BB12_5
 ; GCN-IR-NEXT:  ; %bb.2: ; %udiv-preheader
 ; GCN-IR-NEXT:    v_lshr_b64 v[7:8], v[0:1], v7
 ; GCN-IR-NEXT:    v_add_i32_e32 v0, vcc, 0xffffffc4, v6
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
 ; GCN-IR-NEXT:    v_addc_u32_e64 v1, s[4:5], 0, -1, vcc
 ; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, 31
 ; GCN-IR-NEXT:  BB12_3: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
 ; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
-; GCN-IR-NEXT:    v_lshrrev_b32_e32 v4, 31, v3
+; GCN-IR-NEXT:    v_lshr_b32_e32 v4, v3, v6
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v4
 ; GCN-IR-NEXT:    v_lshl_b64 v[2:3], v[2:3], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v7, v4
-; GCN-IR-NEXT:    v_sub_i32_e32 v4, vcc, 23, v6
-; GCN-IR-NEXT:    v_or_b32_e32 v2, v9, v2
+; GCN-IR-NEXT:    v_sub_i32_e32 v4, vcc, 23, v7
 ; GCN-IR-NEXT:    v_subb_u32_e32 v4, vcc, 0, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v2, v9, v2
+; GCN-IR-NEXT:    v_ashr_i32_e32 v9, v4, v6
+; GCN-IR-NEXT:    v_and_b32_e32 v4, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v9, 24, v9
+; GCN-IR-NEXT:    v_sub_i32_e32 v7, vcc, v7, v9
 ; GCN-IR-NEXT:    v_add_i32_e64 v9, s[4:5], 1, v0
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v7, 31, v4
 ; GCN-IR-NEXT:    v_or_b32_e32 v3, v10, v3
 ; GCN-IR-NEXT:    v_addc_u32_e64 v10, s[4:5], 0, v1, s[4:5]
 ; GCN-IR-NEXT:    v_cmp_lt_u64_e64 s[4:5], v[9:10], v[0:1]
-; GCN-IR-NEXT:    v_and_b32_e32 v4, 1, v7
-; GCN-IR-NEXT:    v_and_b32_e32 v7, 24, v7
 ; GCN-IR-NEXT:    v_mov_b32_e32 v0, v9
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, 0
-; GCN-IR-NEXT:    v_sub_i32_e32 v7, vcc, v6, v7
 ; GCN-IR-NEXT:    v_mov_b32_e32 v1, v10
 ; GCN-IR-NEXT:    v_mov_b32_e32 v10, v5
 ; GCN-IR-NEXT:    v_subbrev_u32_e32 v8, vcc, 0, v8, vcc
 ; GCN-IR-NEXT:    s_or_b64 s[8:9], s[4:5], s[8:9]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, v4
 ; GCN-IR-NEXT:    s_andn2_b64 exec, exec, s[8:9]
 ; GCN-IR-NEXT:    s_cbranch_execnz BB12_3
 ; GCN-IR-NEXT:  ; %bb.4: ; %Flow

llvm/test/CodeGen/AMDGPU/urem64.ll

 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], s[6:7], v0
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB0_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB0_6
 ; GCN-IR-NEXT:  BB0_4: ; %udiv-preheader
-; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
 ; GCN-IR-NEXT:    s_add_u32 s8, s2, -1
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], s[6:7], v4
-; GCN-IR-NEXT:    v_add_i32_e32 v4, vcc, v2, v3
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
-; GCN-IR-NEXT:    s_addc_u32 s9, s3, -1
-; GCN-IR-NEXT:    v_addc_u32_e64 v5, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_not_b32_e32 v2, v2
+; GCN-IR-NEXT:    v_add_i32_e32 v5, vcc, v2, v3
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], s[6:7], v4
+; GCN-IR-NEXT:    s_addc_u32 s9, s3, -1
+; GCN-IR-NEXT:    v_addc_u32_e64 v6, s[0:1], -1, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s9
 ; GCN-IR-NEXT:  BB0_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s9
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s8, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s2, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s3, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s8, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s2, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s3, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB0_5
 ; GCN-IR-NEXT:  BB0_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB0_7: ; %udiv-end
 ; GCN-IR-NEXT:    v_mul_lo_u32 v1, s2, v1
 ; GCN-IR-NEXT:    v_mul_hi_u32 v2, s2, v0

 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
 ; GCN-IR-NEXT:    s_cbranch_vccz BB6_4
 ; GCN-IR-NEXT:  ; %bb.3:
 ; GCN-IR-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:    s_branch BB6_6
 ; GCN-IR-NEXT:  BB6_4: ; %udiv-preheader
 ; GCN-IR-NEXT:    s_add_u32 s3, s6, -1
-; GCN-IR-NEXT:    v_lshr_b64 v[6:7], 24, v3
-; GCN-IR-NEXT:    v_sub_i32_e32 v4, vcc, 58, v2
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, 0
+; GCN-IR-NEXT:    v_lshr_b64 v[7:8], 24, v3
 ; GCN-IR-NEXT:    s_addc_u32 s8, s7, -1
-; GCN-IR-NEXT:    v_subb_u32_e64 v5, s[0:1], 0, 0, vcc
+; GCN-IR-NEXT:    v_sub_i32_e32 v5, vcc, 58, v2
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
+; GCN-IR-NEXT:    v_subb_u32_e64 v6, s[0:1], 0, 0, vcc
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v4, s8
 ; GCN-IR-NEXT:  BB6_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
+; GCN-IR-NEXT:    v_lshl_b64 v[7:8], v[7:8], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
-; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
-; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
-; GCN-IR-NEXT:    v_mov_b32_e32 v2, s8
-; GCN-IR-NEXT:    v_sub_i32_e32 v8, vcc, s3, v6
-; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v2, v7, vcc
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
-; GCN-IR-NEXT:    v_and_b32_e32 v10, s6, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v11, s7, v8
-; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
-; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
-; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
+; GCN-IR-NEXT:    v_or_b32_e32 v8, v8, v3
+; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, s3, v7
+; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, v4, v8, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v0, v9, v0
+; GCN-IR-NEXT:    v_ashrrev_i32_e32 v9, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v11, s6, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v9
+; GCN-IR-NEXT:    v_and_b32_e32 v12, s7, v9
+; GCN-IR-NEXT:    v_add_i32_e32 v9, vcc, 1, v5
+; GCN-IR-NEXT:    v_or_b32_e32 v1, v10, v1
+; GCN-IR-NEXT:    v_addc_u32_e32 v10, vcc, 0, v6, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[9:10], v[5:6]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
-; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v11, s[0:1]
+; GCN-IR-NEXT:    v_sub_i32_e64 v7, s[0:1], v7, v11
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_mov_b32_e32 v10, v3
+; GCN-IR-NEXT:    v_subb_u32_e64 v8, s[0:1], v8, v12, s[0:1]
 ; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
+; GCN-IR-NEXT:    v_mov_b32_e32 v9, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB6_5
 ; GCN-IR-NEXT:  BB6_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB6_7: ; %udiv-end
 ; GCN-IR-NEXT:    v_mul_lo_u32 v1, s6, v1
 ; GCN-IR-NEXT:    v_mul_hi_u32 v2, s6, v0

 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v3, 0
 ; GCN-IR-NEXT:  BB7_5: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
 ; GCN-IR-NEXT:    v_lshl_b64 v[6:7], v[6:7], 1
 ; GCN-IR-NEXT:    v_lshrrev_b32_e32 v2, 31, v1
 ; GCN-IR-NEXT:    v_or_b32_e32 v6, v6, v2
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
+; GCN-IR-NEXT:    v_or_b32_e32 v7, v7, v3
 ; GCN-IR-NEXT:    v_sub_i32_e32 v2, vcc, 23, v6
 ; GCN-IR-NEXT:    v_subb_u32_e32 v2, vcc, 0, v7, vcc
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v8, v0
 ; GCN-IR-NEXT:    v_ashrrev_i32_e32 v8, 31, v2
+; GCN-IR-NEXT:    v_and_b32_e32 v10, 24, v8
 ; GCN-IR-NEXT:    v_and_b32_e32 v2, 1, v8
-; GCN-IR-NEXT:    v_and_b32_e32 v8, 24, v8
-; GCN-IR-NEXT:    v_sub_i32_e32 v6, vcc, v6, v8
-; GCN-IR-NEXT:    v_add_i32_e64 v8, s[0:1], 1, v4
+; GCN-IR-NEXT:    v_add_i32_e32 v8, vcc, 1, v4
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v9, v1
-; GCN-IR-NEXT:    v_addc_u32_e64 v9, s[0:1], 0, v5, s[0:1]
-; GCN-IR-NEXT:    v_cmp_lt_u64_e64 s[0:1], v[8:9], v[4:5]
+; GCN-IR-NEXT:    v_addc_u32_e32 v9, vcc, 0, v5, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[8:9], v[4:5]
 ; GCN-IR-NEXT:    v_mov_b32_e32 v4, v8
-; GCN-IR-NEXT:    v_subbrev_u32_e32 v7, vcc, 0, v7, vcc
+; GCN-IR-NEXT:    v_sub_i32_e64 v6, s[0:1], v6, v10
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, v9
 ; GCN-IR-NEXT:    v_mov_b32_e32 v9, v3
-; GCN-IR-NEXT:    s_and_b64 vcc, exec, s[0:1]
+; GCN-IR-NEXT:    v_subb_u32_e64 v7, s[0:1], v7, v3, s[0:1]
+; GCN-IR-NEXT:    s_and_b64 vcc, exec, vcc
 ; GCN-IR-NEXT:    v_mov_b32_e32 v8, v2
 ; GCN-IR-NEXT:    s_cbranch_vccz BB7_5
 ; GCN-IR-NEXT:  BB7_6: ; %udiv-loop-exit
 ; GCN-IR-NEXT:    v_lshl_b64 v[0:1], v[0:1], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v0, v2, v0
 ; GCN-IR-NEXT:    v_or_b32_e32 v1, v3, v1
 ; GCN-IR-NEXT:  BB7_7: ; %udiv-end
 ; GCN-IR-NEXT:    v_mul_hi_u32 v2, v0, 24

 ; GCN-IR-NEXT:    v_mov_b32_e32 v4, 0
 ; GCN-IR-NEXT:    v_lshl_b64 v[2:3], v[0:1], v2
 ; GCN-IR-NEXT:    s_mov_b64 s[8:9], 0
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, 0
 ; GCN-IR-NEXT:    s_and_saveexec_b64 s[4:5], vcc
 ; GCN-IR-NEXT:    s_xor_b64 s[10:11], exec, s[4:5]
 ; GCN-IR-NEXT:    s_cbranch_execz BB9_5
 ; GCN-IR-NEXT:  ; %bb.2: ; %udiv-preheader
-; GCN-IR-NEXT:    v_mov_b32_e32 v10, 0
-; GCN-IR-NEXT:    v_lshr_b64 v[8:9], v[0:1], v7
+; GCN-IR-NEXT:    v_mov_b32_e32 v11, 0
+; GCN-IR-NEXT:    v_lshr_b64 v[9:10], v[0:1], v7
 ; GCN-IR-NEXT:    v_add_i32_e32 v6, vcc, 0xffffffcf, v6
 ; GCN-IR-NEXT:    v_addc_u32_e64 v7, s[4:5], 0, -1, vcc
-; GCN-IR-NEXT:    v_mov_b32_e32 v11, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v12, 0
+; GCN-IR-NEXT:    v_mov_b32_e32 v8, 31
 ; GCN-IR-NEXT:    s_movk_i32 s12, 0x7fff
 ; GCN-IR-NEXT:  BB9_3: ; %udiv-do-while
 ; GCN-IR-NEXT:    ; =>This Inner Loop Header: Depth=1
-; GCN-IR-NEXT:    v_lshl_b64 v[8:9], v[8:9], 1
-; GCN-IR-NEXT:    v_lshrrev_b32_e32 v4, 31, v3
-; GCN-IR-NEXT:    v_or_b32_e32 v8, v8, v4
+; GCN-IR-NEXT:    v_lshl_b64 v[9:10], v[9:10], 1
+; GCN-IR-NEXT:    v_lshr_b32_e32 v4, v3, v8
+; GCN-IR-NEXT:    v_or_b32_e32 v9, v9, v4
 ; GCN-IR-NEXT:    v_lshl_b64 v[2:3], v[2:3], 1
-; GCN-IR-NEXT:    v_sub_i32_e32 v4, vcc, s12, v8
-; GCN-IR-NEXT:    v_subb_u32_e32 v4, vcc, 0, v9, vcc
-; GCN-IR-NEXT:    v_or_b32_e32 v2, v10, v2
-; GCN-IR-NEXT:    v_ashrrev_i32_e32 v10, 31, v4
-; GCN-IR-NEXT:    v_and_b32_e32 v12, 0x8000, v10
-; GCN-IR-NEXT:    v_and_b32_e32 v4, 1, v10
-; GCN-IR-NEXT:    v_add_i32_e32 v10, vcc, 1, v6
-; GCN-IR-NEXT:    v_or_b32_e32 v3, v11, v3
-; GCN-IR-NEXT:    v_addc_u32_e32 v11, vcc, 0, v7, vcc
-; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[10:11], v[6:7]
-; GCN-IR-NEXT:    v_mov_b32_e32 v6, v10
+; GCN-IR-NEXT:    v_sub_i32_e32 v4, vcc, s12, v9
+; GCN-IR-NEXT:    v_subb_u32_e32 v4, vcc, 0, v10, vcc
+; GCN-IR-NEXT:    v_or_b32_e32 v2, v11, v2
+; GCN-IR-NEXT:    v_ashr_i32_e32 v11, v4, v8
+; GCN-IR-NEXT:    v_and_b32_e32 v13, 0x8000, v11
+; GCN-IR-NEXT:    v_and_b32_e32 v4, 1, v11
+; GCN-IR-NEXT:    v_add_i32_e32 v11, vcc, 1, v6
+; GCN-IR-NEXT:    v_or_b32_e32 v3, v12, v3
+; GCN-IR-NEXT:    v_addc_u32_e32 v12, vcc, 0, v7, vcc
+; GCN-IR-NEXT:    v_cmp_lt_u64_e32 vcc, v[11:12], v[6:7]
+; GCN-IR-NEXT:    v_mov_b32_e32 v6, v11
 ; GCN-IR-NEXT:    v_mov_b32_e32 v5, 0
-; GCN-IR-NEXT:    v_mov_b32_e32 v7, v11
-; GCN-IR-NEXT:    v_mov_b32_e32 v11, v5
-; GCN-IR-NEXT:    v_mov_b32_e32 v13, 0
-; GCN-IR-NEXT:    v_sub_i32_e64 v8, s[4:5], v8, v12
-; GCN-IR-NEXT:    v_subb_u32_e64 v9, s[4:5], v9, v13, s[4:5]
+; GCN-IR-NEXT:    v_mov_b32_e32 v7, v12
+; GCN-IR-NEXT:    v_mov_b32_e32 v12, v5
+; GCN-IR-NEXT:    v_mov_b32_e32 v14, 0
+; GCN-IR-NEXT:    v_sub_i32_e64 v9, s[4:5], v9, v13
+; GCN-IR-NEXT:    v_subb_u32_e64 v10, s[4:5], v10, v14, s[4:5]
 ; GCN-IR-NEXT:    s_or_b64 s[8:9], vcc, s[8:9]
-; GCN-IR-NEXT:    v_mov_b32_e32 v10, v4
+; GCN-IR-NEXT:    v_mov_b32_e32 v11, v4
 ; GCN-IR-NEXT:    s_andn2_b64 exec, exec, s[8:9]
 ; GCN-IR-NEXT:    s_cbranch_execnz BB9_3
 ; GCN-IR-NEXT:  ; %bb.4: ; %Flow
 ; GCN-IR-NEXT:    s_or_b64 exec, exec, s[8:9]
 ; GCN-IR-NEXT:  BB9_5: ; %Flow3
 ; GCN-IR-NEXT:    s_or_b64 exec, exec, s[10:11]
 ; GCN-IR-NEXT:    v_lshl_b64 v[2:3], v[2:3], 1
 ; GCN-IR-NEXT:    v_or_b32_e32 v5, v5, v3

llvm/test/CodeGen/Thumb2/mve-floatregloops.ll

 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -mtriple=thumbv8.1m.main-arm-none-eabi -mattr=+mve.fp -verify-machineinstrs %s -o - | FileCheck %s
 
 define arm_aapcs_vfpcc void @test_fadd(float* noalias nocapture readonly %A, float %B, float* noalias nocapture %C, i32 %n) {
 ; CHECK-LABEL: test_fadd:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r2, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r3, s0
 ; CHECK-NEXT:  .LBB0_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
-; CHECK-NEXT:    vmov r3, s0
+; CHECK-NEXT:    vldrw.u32 q0, [r0], #16
 ; CHECK-NEXT:    subs r2, #4
-; CHECK-NEXT:    vadd.f32 q1, q1, r3
-; CHECK-NEXT:    vstrb.8 q1, [r1], #16
+; CHECK-NEXT:    vadd.f32 q0, q0, r3
+; CHECK-NEXT:    vstrb.8 q0, [r1], #16
 ; CHECK-NEXT:    bne .LBB0_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp18 = icmp sgt i32 %n, 0

 }
 
 define arm_aapcs_vfpcc void @test_fadd_r(float* noalias nocapture readonly %A, float %B, float* noalias nocapture %C, i32 %n) {
 ; CHECK-LABEL: test_fadd_r:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r2, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r3, s0
 ; CHECK-NEXT:  .LBB1_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
-; CHECK-NEXT:    vmov r3, s0
+; CHECK-NEXT:    vldrw.u32 q0, [r0], #16
 ; CHECK-NEXT:    subs r2, #4
-; CHECK-NEXT:    vadd.f32 q1, q1, r3
-; CHECK-NEXT:    vstrb.8 q1, [r1], #16
+; CHECK-NEXT:    vadd.f32 q0, q0, r3
+; CHECK-NEXT:    vstrb.8 q0, [r1], #16
 ; CHECK-NEXT:    bne .LBB1_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp18 = icmp sgt i32 %n, 0

 }
 
 define arm_aapcs_vfpcc void @test_fmul(float* noalias nocapture readonly %A, float %B, float* noalias nocapture %C, i32 %n) {
 ; CHECK-LABEL: test_fmul:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r2, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r3, s0
 ; CHECK-NEXT:  .LBB2_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
-; CHECK-NEXT:    vmov r3, s0
+; CHECK-NEXT:    vldrw.u32 q0, [r0], #16
 ; CHECK-NEXT:    subs r2, #4
-; CHECK-NEXT:    vmul.f32 q1, q1, r3
-; CHECK-NEXT:    vstrb.8 q1, [r1], #16
+; CHECK-NEXT:    vmul.f32 q0, q0, r3
+; CHECK-NEXT:    vstrb.8 q0, [r1], #16
 ; CHECK-NEXT:    bne .LBB2_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp18 = icmp sgt i32 %n, 0

 }
 
 define arm_aapcs_vfpcc void @test_fmul_r(float* noalias nocapture readonly %A, float %B, float* noalias nocapture %C, i32 %n) {
 ; CHECK-LABEL: test_fmul_r:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r2, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r3, s0
 ; CHECK-NEXT:  .LBB3_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
-; CHECK-NEXT:    vmov r3, s0
+; CHECK-NEXT:    vldrw.u32 q0, [r0], #16
 ; CHECK-NEXT:    subs r2, #4
-; CHECK-NEXT:    vmul.f32 q1, q1, r3
-; CHECK-NEXT:    vstrb.8 q1, [r1], #16
+; CHECK-NEXT:    vmul.f32 q0, q0, r3
+; CHECK-NEXT:    vstrb.8 q0, [r1], #16
 ; CHECK-NEXT:    bne .LBB3_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp18 = icmp sgt i32 %n, 0

 }
 
 define arm_aapcs_vfpcc void @test_fsub(float* noalias nocapture readonly %A, float %B, float* noalias nocapture %C, i32 %n) {
 ; CHECK-LABEL: test_fsub:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r2, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r3, s0
 ; CHECK-NEXT:  .LBB4_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
-; CHECK-NEXT:    vmov r3, s0
+; CHECK-NEXT:    vldrw.u32 q0, [r0], #16
 ; CHECK-NEXT:    subs r2, #4
-; CHECK-NEXT:    vsub.f32 q1, q1, r3
-; CHECK-NEXT:    vstrb.8 q1, [r1], #16
+; CHECK-NEXT:    vsub.f32 q0, q0, r3
+; CHECK-NEXT:    vstrb.8 q0, [r1], #16
 ; CHECK-NEXT:    bne .LBB4_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp18 = icmp sgt i32 %n, 0

 
 
 define arm_aapcs_vfpcc void @test_fmas(float* noalias nocapture readonly %A, float* noalias nocapture readonly %B, float %C, float* noalias nocapture %D, i32 %n) {
 ; CHECK-LABEL: test_fmas:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r3, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r12, s0
 ; CHECK-NEXT:  .LBB6_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
-; CHECK-NEXT:    vldrw.u32 q2, [r1], #16
-; CHECK-NEXT:    vmov r12, s0
+; CHECK-NEXT:    vldrw.u32 q0, [r0], #16
+; CHECK-NEXT:    vldrw.u32 q1, [r1], #16
 ; CHECK-NEXT:    subs r3, #4
-; CHECK-NEXT:    vfmas.f32 q2, q1, r12
-; CHECK-NEXT:    vstrb.8 q2, [r2], #16
+; CHECK-NEXT:    vfmas.f32 q1, q0, r12
+; CHECK-NEXT:    vstrb.8 q1, [r2], #16
 ; CHECK-NEXT:    bne .LBB6_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp110 = icmp sgt i32 %n, 0

 }
 
 define arm_aapcs_vfpcc void @test_fmas_r(float* noalias nocapture readonly %A, float* noalias nocapture readonly %B, float %C, float* noalias nocapture %D, i32 %n) {
 ; CHECK-LABEL: test_fmas_r:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r3, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r12, s0
 ; CHECK-NEXT:  .LBB7_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
-; CHECK-NEXT:    vldrw.u32 q2, [r1], #16
-; CHECK-NEXT:    vmov r12, s0
+; CHECK-NEXT:    vldrw.u32 q0, [r0], #16
+; CHECK-NEXT:    vldrw.u32 q1, [r1], #16
 ; CHECK-NEXT:    subs r3, #4
-; CHECK-NEXT:    vfmas.f32 q2, q1, r12
-; CHECK-NEXT:    vstrb.8 q2, [r2], #16
+; CHECK-NEXT:    vfmas.f32 q1, q0, r12
+; CHECK-NEXT:    vstrb.8 q1, [r2], #16
 ; CHECK-NEXT:    bne .LBB7_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp110 = icmp sgt i32 %n, 0

 }
 
 define arm_aapcs_vfpcc void @test_fma(float* noalias nocapture readonly %A, float* noalias nocapture readonly %B, float %C, float* noalias nocapture %D, i32 %n) {
 ; CHECK-LABEL: test_fma:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r3, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r12, s0
 ; CHECK-NEXT:  .LBB8_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
-; CHECK-NEXT:    vldrw.u32 q2, [r1], #16
-; CHECK-NEXT:    vmov r12, s0
+; CHECK-NEXT:    vldrw.u32 q0, [r0], #16
+; CHECK-NEXT:    vldrw.u32 q1, [r1], #16
 ; CHECK-NEXT:    subs r3, #4
-; CHECK-NEXT:    vfma.f32 q2, q1, r12
-; CHECK-NEXT:    vstrb.8 q2, [r2], #16
+; CHECK-NEXT:    vfma.f32 q1, q0, r12
+; CHECK-NEXT:    vstrb.8 q1, [r2], #16
 ; CHECK-NEXT:    bne .LBB8_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp110 = icmp sgt i32 %n, 0

 }
 
 define arm_aapcs_vfpcc void @test_fma_r(float* noalias nocapture readonly %A, float* noalias nocapture readonly %B, float %C, float* noalias nocapture %D, i32 %n) {
 ; CHECK-LABEL: test_fma_r:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r3, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r12, s0
+; CHECK-NEXT:    vdup.32 q0, r12
 ; CHECK-NEXT:  .LBB9_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vmov r12, s0
-; CHECK-NEXT:    vldrw.u32 q2, [r0], #16
-; CHECK-NEXT:    vldrw.u32 q3, [r1], #16
-; CHECK-NEXT:    vdup.32 q1, r12
+; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
+; CHECK-NEXT:    vldrw.u32 q2, [r1], #16
 ; CHECK-NEXT:    subs r3, #4
-; CHECK-NEXT:    vfma.f32 q3, q1, q2
-; CHECK-NEXT:    vstrb.8 q3, [r2], #16
+; CHECK-NEXT:    vfma.f32 q2, q0, q1
+; CHECK-NEXT:    vstrb.8 q2, [r2], #16
 ; CHECK-NEXT:    bne .LBB9_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp110 = icmp sgt i32 %n, 0

 
 
 define arm_aapcs_vfpcc void @test_fmss(float* noalias nocapture readonly %A, float* noalias nocapture readonly %B, float %C, float* noalias nocapture %D, i32 %n) {
 ; CHECK-LABEL: test_fmss:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r3, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r12, s0
+; CHECK-NEXT:    vdup.32 q0, r12
+; CHECK-NEXT:    vneg.f32 q0, q0
 ; CHECK-NEXT:  .LBB10_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vmov r12, s0
-; CHECK-NEXT:    vldrw.u32 q2, [r0], #16
-; CHECK-NEXT:    vdup.32 q1, r12
-; CHECK-NEXT:    vldrw.u32 q3, [r1], #16
-; CHECK-NEXT:    vneg.f32 q1, q1
+; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
+; CHECK-NEXT:    vldrw.u32 q2, [r1], #16
+; CHECK-NEXT:    vmov q3, q0
 ; CHECK-NEXT:    subs r3, #4
-; CHECK-NEXT:    vfma.f32 q1, q3, q2
-; CHECK-NEXT:    vstrb.8 q1, [r2], #16
+; CHECK-NEXT:    vfma.f32 q3, q2, q1
+; CHECK-NEXT:    vstrb.8 q3, [r2], #16
 ; CHECK-NEXT:    bne .LBB10_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp110 = icmp sgt i32 %n, 0

 }
 
 define arm_aapcs_vfpcc void @test_fms(float* noalias nocapture readonly %A, float* noalias nocapture readonly %B, float %C, float* noalias nocapture %D, i32 %n) {
 ; CHECK-LABEL: test_fms:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r3, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r12, s0
 ; CHECK-NEXT:  .LBB12_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q1, [r1], #16
-; CHECK-NEXT:    vldrw.u32 q2, [r0], #16
-; CHECK-NEXT:    vmov r12, s0
+; CHECK-NEXT:    vldrw.u32 q0, [r1], #16
+; CHECK-NEXT:    vldrw.u32 q1, [r0], #16
 ; CHECK-NEXT:    subs r3, #4
-; CHECK-NEXT:    vneg.f32 q1, q1
-; CHECK-NEXT:    vfma.f32 q1, q2, r12
-; CHECK-NEXT:    vstrb.8 q1, [r2], #16
+; CHECK-NEXT:    vneg.f32 q0, q0
+; CHECK-NEXT:    vfma.f32 q0, q1, r12
+; CHECK-NEXT:    vstrb.8 q0, [r2], #16
 ; CHECK-NEXT:    bne .LBB12_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp110 = icmp sgt i32 %n, 0

 }
 
 define arm_aapcs_vfpcc void @test_fms_r(float* noalias nocapture readonly %A, float* noalias nocapture readonly %B, float %C, float* noalias nocapture %D, i32 %n) {
 ; CHECK-LABEL: test_fms_r:
 ; CHECK:       @ %bb.0: @ %entry
 ; CHECK-NEXT:    cmp r3, #1
 ; CHECK-NEXT:    it lt
 ; CHECK-NEXT:    bxlt lr
+; CHECK-NEXT:    vmov r12, s0
+; CHECK-NEXT:    vdup.32 q0, r12
 ; CHECK-NEXT:  .LBB13_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q2, [r1], #16
-; CHECK-NEXT:    vmov r12, s0
-; CHECK-NEXT:    vldrw.u32 q3, [r0], #16
-; CHECK-NEXT:    vdup.32 q1, r12
-; CHECK-NEXT:    vneg.f32 q2, q2
+; CHECK-NEXT:    vldrw.u32 q1, [r1], #16
+; CHECK-NEXT:    vldrw.u32 q2, [r0], #16
 ; CHECK-NEXT:    subs r3, #4
-; CHECK-NEXT:    vfma.f32 q2, q1, q3
-; CHECK-NEXT:    vstrb.8 q2, [r2], #16
+; CHECK-NEXT:    vneg.f32 q1, q1
+; CHECK-NEXT:    vfma.f32 q1, q0, q2
+; CHECK-NEXT:    vstrb.8 q1, [r2], #16
 ; CHECK-NEXT:    bne .LBB13_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    bx lr
 entry:
   %0 = and i32 %n, 7
   %cmp = icmp eq i32 %0, 0
   tail call void @llvm.assume(i1 %cmp)
   %cmp110 = icmp sgt i32 %n, 0

 ; CHECK-NEXT:    push.w {r4, r5, r6, r7, r8, r9, r10, r11, lr}
 ; CHECK-NEXT:    .pad #4
 ; CHECK-NEXT:    sub sp, #4
 ; CHECK-NEXT:    .vsave {d8, d9}
 ; CHECK-NEXT:    vpush {d8, d9}
 ; CHECK-NEXT:    .pad #16
 ; CHECK-NEXT:    sub sp, #16
 ; CHECK-NEXT:    ldrh r4, [r0]
-; CHECK-NEXT:    ldr.w r12, [r0, #4]
+; CHECK-NEXT:    ldr r5, [r0, #4]
 ; CHECK-NEXT:    subs r7, r4, #1
 ; CHECK-NEXT:    cmp r7, #3
 ; CHECK-NEXT:    bhi .LBB15_6
 ; CHECK-NEXT:  @ %bb.1: @ %if.then
 ; CHECK-NEXT:    ldr r6, [r0, #8]
-; CHECK-NEXT:    add.w r11, r12, r7, lsl #2
+; CHECK-NEXT:    add.w r11, r5, r7, lsl #2
 ; CHECK-NEXT:    lsr.w lr, r3, #2
 ; CHECK-NEXT:    vldr s0, [r6]
 ; CHECK-NEXT:    vldr s2, [r6, #4]
 ; CHECK-NEXT:    vldr s4, [r6, #8]
 ; CHECK-NEXT:    vldr s6, [r6, #12]
 ; CHECK-NEXT:    wls lr, lr, .LBB15_5
 ; CHECK-NEXT:  @ %bb.2: @ %while.body.lr.ph
-; CHECK-NEXT:    str r4, [sp, #12] @ 4-byte Spill
-; CHECK-NEXT:    bic r4, r3, #3
+; CHECK-NEXT:    strd r3, r4, [sp, #8] @ 8-byte Folded Spill
+; CHECK-NEXT:    vmov r10, s4
+; CHECK-NEXT:    vmov r12, s6
+; CHECK-NEXT:    bic r3, r3, #3
+; CHECK-NEXT:    vmov r4, s2
+; CHECK-NEXT:    str r3, [sp] @ 4-byte Spill
+; CHECK-NEXT:    vmov r8, s0
+; CHECK-NEXT:    add.w r3, r2, r3, lsl #2
+; CHECK-NEXT:    str r3, [sp, #4] @ 4-byte Spill
 ; CHECK-NEXT:    movs r6, #0
-; CHECK-NEXT:    str r4, [sp, #4] @ 4-byte Spill
-; CHECK-NEXT:    add.w r4, r2, r4, lsl #2
-; CHECK-NEXT:    str r4, [sp, #8] @ 4-byte Spill
+; CHECK-NEXT:    mov r3, r5
 ; CHECK-NEXT:  .LBB15_3: @ %while.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    adds r7, r1, r6
-; CHECK-NEXT:    add.w r5, r12, r6
-; CHECK-NEXT:    vldrw.u32 q2, [r7]
+; CHECK-NEXT:    add.w r9, r1, r6
 ; CHECK-NEXT:    add.w r7, r11, r6
-; CHECK-NEXT:    vmov r10, s0
+; CHECK-NEXT:    vldrw.u32 q2, [r9]
+; CHECK-NEXT:    adds r5, r3, r6
 ; CHECK-NEXT:    vstrw.32 q2, [r7]
-; CHECK-NEXT:    vmov r9, s2
-; CHECK-NEXT:    vldrw.u32 q2, [r5]
-; CHECK-NEXT:    vmov r4, s4
 ; CHECK-NEXT:    adds r7, r2, r6
+; CHECK-NEXT:    vldrw.u32 q2, [r5]
 ; CHECK-NEXT:    adds r6, #16
-; CHECK-NEXT:    vmul.f32 q2, q2, r10
+; CHECK-NEXT:    vmul.f32 q2, q2, r8
 ; CHECK-NEXT:    vldrw.u32 q3, [r5, #4]
-; CHECK-NEXT:    vmov r8, s6
-; CHECK-NEXT:    vfma.f32 q2, q3, r9
+; CHECK-NEXT:    vfma.f32 q2, q3, r4
 ; CHECK-NEXT:    vldrw.u32 q3, [r5, #8]
 ; CHECK-NEXT:    vldrw.u32 q4, [r5, #12]
-; CHECK-NEXT:    vfma.f32 q2, q3, r4
-; CHECK-NEXT:    vfma.f32 q2, q4, r8
+; CHECK-NEXT:    vfma.f32 q2, q3, r10
+; CHECK-NEXT:    vfma.f32 q2, q4, r12
 ; CHECK-NEXT:    vstrw.32 q2, [r7]
 ; CHECK-NEXT:    le lr, .LBB15_3
 ; CHECK-NEXT:  @ %bb.4: @ %while.end.loopexit
-; CHECK-NEXT:    ldr r2, [sp, #4] @ 4-byte Reload
+; CHECK-NEXT:    ldr r2, [sp] @ 4-byte Reload
+; CHECK-NEXT:    add r4, sp, #4
 ; CHECK-NEXT:    add r11, r6
-; CHECK-NEXT:    add.w r12, r12, r2, lsl #2
+; CHECK-NEXT:    add.w r5, r3, r2, lsl #2
 ; CHECK-NEXT:    add.w r1, r1, r2, lsl #2
-; CHECK-NEXT:    ldrd r2, r4, [sp, #8] @ 8-byte Folded Reload
+; CHECK-NEXT:    ldm r4, {r2, r3, r4} @ 12-byte Folded Reload
 ; CHECK-NEXT:  .LBB15_5: @ %while.end
 ; CHECK-NEXT:    and r7, r3, #3
 ; CHECK-NEXT:    vldrw.u32 q2, [r1]
 ; CHECK-NEXT:    vctp.32 r7
 ; CHECK-NEXT:    vpst
 ; CHECK-NEXT:    vstrwt.32 q2, [r11]
 ; CHECK-NEXT:    vmov r6, s2
-; CHECK-NEXT:    vmov r5, s0
-; CHECK-NEXT:    vldrw.u32 q0, [r12]
-; CHECK-NEXT:    vmov r1, s6
+; CHECK-NEXT:    vmov r1, s0
+; CHECK-NEXT:    vldrw.u32 q0, [r5]
+; CHECK-NEXT:    vmov r12, s6
 ; CHECK-NEXT:    vmov r7, s4
-; CHECK-NEXT:    vmul.f32 q0, q0, r5
-; CHECK-NEXT:    vldrw.u32 q1, [r12, #4]
+; CHECK-NEXT:    vmul.f32 q0, q0, r1
+; CHECK-NEXT:    vldrw.u32 q1, [r5, #4]
 ; CHECK-NEXT:    vfma.f32 q0, q1, r6
-; CHECK-NEXT:    vldrw.u32 q1, [r12, #8]
+; CHECK-NEXT:    vldrw.u32 q1, [r5, #8]
 ; CHECK-NEXT:    vfma.f32 q0, q1, r7
-; CHECK-NEXT:    vldrw.u32 q1, [r12, #12]
-; CHECK-NEXT:    vfma.f32 q0, q1, r1
+; CHECK-NEXT:    vldrw.u32 q1, [r5, #12]
+; CHECK-NEXT:    vfma.f32 q0, q1, r12
 ; CHECK-NEXT:    vpst
 ; CHECK-NEXT:    vstrwt.32 q0, [r2]
-; CHECK-NEXT:    ldr.w r12, [r0, #4]
+; CHECK-NEXT:    ldr r5, [r0, #4]
 ; CHECK-NEXT:  .LBB15_6: @ %if.end
-; CHECK-NEXT:    add.w r0, r12, r3, lsl #2
+; CHECK-NEXT:    add.w r0, r5, r3, lsl #2
+; CHECK-NEXT:    mov r2, r5
 ; CHECK-NEXT:    lsr.w lr, r4, #2
 ; CHECK-NEXT:    wls lr, lr, .LBB15_10
 ; CHECK-NEXT:  @ %bb.7: @ %while.body51.preheader
-; CHECK-NEXT:    bic r2, r4, #3
-; CHECK-NEXT:    adds r1, r2, r3
-; CHECK-NEXT:    mov r3, r12
-; CHECK-NEXT:    add.w r1, r12, r1, lsl #2
+; CHECK-NEXT:    bic r7, r4, #3
+; CHECK-NEXT:    adds r1, r7, r3
+; CHECK-NEXT:    mov r3, r2
+; CHECK-NEXT:    add.w r1, r2, r1, lsl #2
 ; CHECK-NEXT:  .LBB15_8: @ %while.body51
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    vldrw.u32 q0, [r0], #16
 ; CHECK-NEXT:    vstrb.8 q0, [r3], #16
 ; CHECK-NEXT:    le lr, .LBB15_8
 ; CHECK-NEXT:  @ %bb.9: @ %while.end55.loopexit
-; CHECK-NEXT:    add.w r12, r12, r2, lsl #2
+; CHECK-NEXT:    add.w r2, r2, r7, lsl #2
 ; CHECK-NEXT:    mov r0, r1
 ; CHECK-NEXT:  .LBB15_10: @ %while.end55
 ; CHECK-NEXT:    ands r1, r4, #3
 ; CHECK-NEXT:    beq .LBB15_12
 ; CHECK-NEXT:  @ %bb.11: @ %if.then59
 ; CHECK-NEXT:    vldrw.u32 q0, [r0]
 ; CHECK-NEXT:    vctp.32 r1
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vstrwt.32 q0, [r12]
+; CHECK-NEXT:    vstrwt.32 q0, [r2]
 ; CHECK-NEXT:  .LBB15_12: @ %if.end61
 ; CHECK-NEXT:    add sp, #16
 ; CHECK-NEXT:    vpop {d8, d9}
 ; CHECK-NEXT:    add sp, #4
 ; CHECK-NEXT:    pop.w {r4, r5, r6, r7, r8, r9, r10, r11, pc}
 entry:
   %pState1 = getelementptr inbounds %struct.arm_fir_instance_f32, %struct.arm_fir_instance_f32* %S, i32 0, i32 1
   %0 = load float*, float** %pState1, align 4

llvm/test/CodeGen/Thumb2/mve-pred-threshold.ll

 ; CHECK-NEXT:    push {r7, lr}
 ; CHECK-NEXT:    cmp r1, #0
 ; CHECK-NEXT:    it eq
 ; CHECK-NEXT:    popeq {r7, pc}
 ; CHECK-NEXT:    vneg.f32 s2, s0
 ; CHECK-NEXT:    mvn r2, #3
 ; CHECK-NEXT:    add.w r1, r2, r1, lsl #2
 ; CHECK-NEXT:    movs r2, #1
-; CHECK-NEXT:    vmov.i32 q1, #0x0
 ; CHECK-NEXT:    add.w lr, r2, r1, lsr #2
+; CHECK-NEXT:    vmov r1, s0
 ; CHECK-NEXT:    dls lr, lr
+; CHECK-NEXT:    vmov r2, s2
+; CHECK-NEXT:    vmov.i32 q0, #0x0
 ; CHECK-NEXT:  .LBB3_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q2, [r0]
-; CHECK-NEXT:    vmov r1, s0
-; CHECK-NEXT:    vcmp.f32 ge, q2, r1
-; CHECK-NEXT:    vmov r1, s2
-; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vcmpt.f32 le, q2, r1
+; CHECK-NEXT:    vldrw.u32 q1, [r0]
+; CHECK-NEXT:    vpt.f32 ge, q1, r1
+; CHECK-NEXT:    vcmpt.f32 le, q1, r2
 ; CHECK-NEXT:    vpnot
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vstrwt.32 q1, [r0], #16
+; CHECK-NEXT:    vstrwt.32 q0, [r0], #16
 ; CHECK-NEXT:    le lr, .LBB3_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    pop {r7, pc}
 entry:
   %conv = zext i16 %N to i32
   %mul = shl nuw nsw i32 %conv, 2
   %cmp15 = icmp eq i16 %N, 0
   br i1 %cmp15, label %for.cond.cleanup, label %vector.ph

 ; CHECK-NEXT:    cmp r1, #0
 ; CHECK-NEXT:    it eq
 ; CHECK-NEXT:    popeq {r7, pc}
 ; CHECK-NEXT:    mvn r2, #7
 ; CHECK-NEXT:    add.w r1, r2, r1, lsl #3
 ; CHECK-NEXT:    movs r2, #1
 ; CHECK-NEXT:    vneg.f16 s2, s0
 ; CHECK-NEXT:    add.w lr, r2, r1, lsr #3
-; CHECK-NEXT:    vmov.i32 q1, #0x0
+; CHECK-NEXT:    vmov r1, s2
+; CHECK-NEXT:    vmov r2, s0
+; CHECK-NEXT:    vmov.i32 q0, #0x0
 ; CHECK-NEXT:    dls lr, lr
 ; CHECK-NEXT:  .LBB4_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vmov r1, s2
-; CHECK-NEXT:    vldrh.u16 q2, [r0]
-; CHECK-NEXT:    vmov r2, s0
-; CHECK-NEXT:    vpt.f16 ge, q2, r2
-; CHECK-NEXT:    vcmpt.f16 le, q2, r1
+; CHECK-NEXT:    vldrh.u16 q1, [r0]
+; CHECK-NEXT:    vpt.f16 ge, q1, r2
+; CHECK-NEXT:    vcmpt.f16 le, q1, r1
 ; CHECK-NEXT:    vpnot
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vstrht.16 q1, [r0], #16
+; CHECK-NEXT:    vstrht.16 q0, [r0], #16
 ; CHECK-NEXT:    le lr, .LBB4_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    pop {r7, pc}
 entry:
   %0 = bitcast float %T.coerce to i32
   %tmp.0.extract.trunc = trunc i32 %0 to i16
   %1 = bitcast i16 %tmp.0.extract.trunc to half
   %conv = zext i16 %N to i32

 ; CHECK-NEXT:    push {r7, lr}
 ; CHECK-NEXT:    cmp r1, #0
 ; CHECK-NEXT:    it eq
 ; CHECK-NEXT:    popeq {r7, pc}
 ; CHECK-NEXT:    vneg.f32 s2, s0
 ; CHECK-NEXT:    mvn r2, #3
 ; CHECK-NEXT:    add.w r1, r2, r1, lsl #2
 ; CHECK-NEXT:    movs r2, #1
-; CHECK-NEXT:    vmov.i32 q1, #0x0
 ; CHECK-NEXT:    add.w lr, r2, r1, lsr #2
+; CHECK-NEXT:    vmov r1, s0
 ; CHECK-NEXT:    dls lr, lr
+; CHECK-NEXT:    vmov r2, s2
+; CHECK-NEXT:    vmov.i32 q0, #0x0
 ; CHECK-NEXT:  .LBB8_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vldrw.u32 q2, [r0]
-; CHECK-NEXT:    vmov r1, s0
-; CHECK-NEXT:    vcmp.f32 ge, q2, r1
-; CHECK-NEXT:    vmov r1, s2
-; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vcmpt.f32 le, q2, r1
+; CHECK-NEXT:    vldrw.u32 q1, [r0]
+; CHECK-NEXT:    vpt.f32 ge, q1, r1
+; CHECK-NEXT:    vcmpt.f32 le, q1, r2
 ; CHECK-NEXT:    vpnot
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vstrwt.32 q1, [r0], #16
+; CHECK-NEXT:    vstrwt.32 q0, [r0], #16
 ; CHECK-NEXT:    le lr, .LBB8_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    pop {r7, pc}
 entry:
   %conv = zext i16 %N to i32
   %mul = shl nuw nsw i32 %conv, 2
   %cmp15 = icmp eq i16 %N, 0
   br i1 %cmp15, label %for.cond.cleanup, label %vector.ph

 ; CHECK-NEXT:    cmp r1, #0
 ; CHECK-NEXT:    it eq
 ; CHECK-NEXT:    popeq {r7, pc}
 ; CHECK-NEXT:    mvn r2, #7
 ; CHECK-NEXT:    add.w r1, r2, r1, lsl #3
 ; CHECK-NEXT:    movs r2, #1
 ; CHECK-NEXT:    vneg.f16 s2, s0
 ; CHECK-NEXT:    add.w lr, r2, r1, lsr #3
-; CHECK-NEXT:    vmov.i32 q1, #0x0
+; CHECK-NEXT:    vmov r1, s2
+; CHECK-NEXT:    vmov r2, s0
+; CHECK-NEXT:    vmov.i32 q0, #0x0
 ; CHECK-NEXT:    dls lr, lr
 ; CHECK-NEXT:  .LBB9_1: @ %vector.body
 ; CHECK-NEXT:    @ =>This Inner Loop Header: Depth=1
-; CHECK-NEXT:    vmov r1, s2
-; CHECK-NEXT:    vldrh.u16 q2, [r0]
-; CHECK-NEXT:    vmov r2, s0
-; CHECK-NEXT:    vpt.f16 ge, q2, r2
-; CHECK-NEXT:    vcmpt.f16 le, q2, r1
+; CHECK-NEXT:    vldrh.u16 q1, [r0]
+; CHECK-NEXT:    vpt.f16 ge, q1, r2
+; CHECK-NEXT:    vcmpt.f16 le, q1, r1
 ; CHECK-NEXT:    vpnot
 ; CHECK-NEXT:    vpst
-; CHECK-NEXT:    vstrht.16 q1, [r0], #16
+; CHECK-NEXT:    vstrht.16 q0, [r0], #16
 ; CHECK-NEXT:    le lr, .LBB9_1
 ; CHECK-NEXT:  @ %bb.2: @ %for.cond.cleanup
 ; CHECK-NEXT:    pop {r7, pc}
 entry:
   %0 = bitcast float %T.coerce to i32
   %tmp.0.extract.trunc = trunc i32 %0 to i16
   %1 = bitcast i16 %tmp.0.extract.trunc to half
   %conv = zext i16 %N to i32