RegAllocFast: Remove early selection loop, the spill calculation will report cost...

The 2nd loop calculates spill costs but reports free registers as cost
0 anyway, so there is little benefit from having a separate early
loop.

Surprisingly this is not NFC, as many register are marked regDisabled
so the first loop often picks up later registers unnecessarily instead
of the first one available in the allocation order...

Patch by Matthias Braun

llvm-svn: 356499

diff --git a/llvm/lib/CodeGen/RegAllocFast.cpp b/llvm/lib/CodeGen/RegAllocFast.cpp
index d3e52f9..46620d8 100644 (file)
--- a/llvm/lib/CodeGen/RegAllocFast.cpp
+++ b/llvm/lib/CodeGen/RegAllocFast.cpp
@@ -582,17 +582,9 @@ void RegAllocFast::allocVirtReg(MachineInstr &MI, LiveReg &LR, unsigned Hint) {
     }
   }
 
-  // First try to find a completely free register.
-  ArrayRef<MCPhysReg> AllocationOrder = RegClassInfo.getOrder(&RC);
-  for (MCPhysReg PhysReg : AllocationOrder) {
-    if (PhysRegState[PhysReg] == regFree && !isRegUsedInInstr(PhysReg)) {
-      assignVirtToPhysReg(LR, PhysReg);
-      return;
-    }
-  }
-
   MCPhysReg BestReg = 0;
   unsigned BestCost = spillImpossible;
+  ArrayRef<MCPhysReg> AllocationOrder = RegClassInfo.getOrder(&RC);
   for (MCPhysReg PhysReg : AllocationOrder) {
     LLVM_DEBUG(dbgs() << "\tRegister: " << printReg(PhysReg, TRI) << ' ');
     unsigned Cost = calcSpillCost(PhysReg);

diff --git a/llvm/test/CodeGen/AArch64/arm64-fast-isel-br.ll b/llvm/test/CodeGen/AArch64/arm64-fast-isel-br.ll
index 87d6811..4c952c6 100644 (file)
--- a/llvm/test/CodeGen/AArch64/arm64-fast-isel-br.ll
+++ b/llvm/test/CodeGen/AArch64/arm64-fast-isel-br.ll
@@ -94,7 +94,7 @@ entry:
   store i32 %c, i32* %c.addr, align 4
   store i64 %d, i64* %d.addr, align 8
   %0 = load i16, i16* %b.addr, align 2
-; CHECK: tbz w0, #0, LBB4_2
+; CHECK: tbz w8, #0, LBB4_2
   %conv = trunc i16 %0 to i1
   br i1 %conv, label %if.then, label %if.end
 

diff --git a/llvm/test/CodeGen/AArch64/arm64-fast-isel-conversion-fallback.ll b/llvm/test/CodeGen/AArch64/arm64-fast-isel-conversion-fallback.ll
index b3e649c..392af06 100644 (file)
--- a/llvm/test/CodeGen/AArch64/arm64-fast-isel-conversion-fallback.ll
+++ b/llvm/test/CodeGen/AArch64/arm64-fast-isel-conversion-fallback.ll
@@ -26,8 +26,8 @@ entry:
 define half @sitofp_hw_i1(i1 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: sitofp_hw_i1
-; CHECK: sbfx w0, w0, #0, #1
-; CHECK: scvtf s0, w0
+; CHECK: sbfx w8, w0, #0, #1
+; CHECK: scvtf s0, w8
 ; CHECK: fcvt  h0, s0
   %conv = sitofp i1 %a to half
   ret half %conv
@@ -37,8 +37,8 @@ entry:
 define half @sitofp_hw_i8(i8 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: sitofp_hw_i8
-; CHECK: sxtb w0, w0
-; CHECK: scvtf s0, w0
+; CHECK: sxtb w8, w0
+; CHECK: scvtf s0, w8
 ; CHECK: fcvt  h0, s0
   %conv = sitofp i8 %a to half
   ret half %conv
@@ -48,8 +48,8 @@ entry:
 define half @sitofp_hw_i16(i16 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: sitofp_hw_i16
-; CHECK: sxth w0, w0
-; CHECK: scvtf s0, w0
+; CHECK: sxth w8, w0
+; CHECK: scvtf s0, w8
 ; CHECK: fcvt  h0, s0
   %conv = sitofp i16 %a to half
   ret half %conv
@@ -79,8 +79,8 @@ entry:
 define half @uitofp_hw_i1(i1 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: uitofp_hw_i1
-; CHECK: and w0, w0, #0x1
-; CHECK: ucvtf s0, w0
+; CHECK: and w8, w0, #0x1
+; CHECK: ucvtf s0, w8
 ; CHECK: fcvt  h0, s0
   %conv = uitofp i1 %a to half
   ret half %conv
@@ -90,8 +90,8 @@ entry:
 define half @uitofp_hw_i8(i8 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: uitofp_hw_i8
-; CHECK: and w0, w0, #0xff
-; CHECK: ucvtf s0, w0
+; CHECK: and w8, w0, #0xff
+; CHECK: ucvtf s0, w8
 ; CHECK: fcvt  h0, s0
   %conv = uitofp i8 %a to half
   ret half %conv
@@ -101,8 +101,8 @@ entry:
 define half @uitofp_hw_i16(i16 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: uitofp_hw_i16
-; CHECK: and w0, w0, #0xffff
-; CHECK: ucvtf s0, w0
+; CHECK: and w8, w0, #0xffff
+; CHECK: ucvtf s0, w8
 ; CHECK: fcvt  h0, s0
   %conv = uitofp i16 %a to half
   ret half %conv

diff --git a/llvm/test/CodeGen/AArch64/arm64-fast-isel-conversion.ll b/llvm/test/CodeGen/AArch64/arm64-fast-isel-conversion.ll
index 7b208cc..7a97a40 100644 (file)
--- a/llvm/test/CodeGen/AArch64/arm64-fast-isel-conversion.ll
+++ b/llvm/test/CodeGen/AArch64/arm64-fast-isel-conversion.ll
@@ -9,13 +9,13 @@ entry:
 ; CHECK: strh w1, [sp, #12]
 ; CHECK: str w2, [sp, #8]
 ; CHECK: str x3, [sp]
-; CHECK: ldr x3, [sp]
-; CHECK: mov x0, x3
-; CHECK: str w0, [sp, #8]
-; CHECK: ldr w0, [sp, #8]
-; CHECK: strh w0, [sp, #12]
-; CHECK: ldrh w0, [sp, #12]
-; CHECK: strb w0, [sp, #15]
+; CHECK: ldr x8, [sp]
+; CHECK: mov x9, x8
+; CHECK: str w9, [sp, #8]
+; CHECK: ldr w9, [sp, #8]
+; CHECK: strh w9, [sp, #12]
+; CHECK: ldrh w9, [sp, #12]
+; CHECK: strb w9, [sp, #15]
 ; CHECK: ldrb w0, [sp, #15]
 ; CHECK: add sp, sp, #16
 ; CHECK: ret
@@ -49,13 +49,13 @@ entry:
 ; CHECK: strh w1, [sp, #12]
 ; CHECK: str w2, [sp, #8]
 ; CHECK: str x3, [sp]
-; CHECK: ldrb w0, [sp, #15]
-; CHECK: strh w0, [sp, #12]
-; CHECK: ldrh w0, [sp, #12]
-; CHECK: str w0, [sp, #8]
-; CHECK: ldr w0, [sp, #8]
-; CHECK: mov x3, x0
-; CHECK: str x3, [sp]
+; CHECK: ldrb w8, [sp, #15]
+; CHECK: strh w8, [sp, #12]
+; CHECK: ldrh w8, [sp, #12]
+; CHECK: str w8, [sp, #8]
+; CHECK: ldr w8, [sp, #8]
+; CHECK: mov x9, x8
+; CHECK: str x9, [sp]
 ; CHECK: ldr x0, [sp]
 ; CHECK: ret
   %a.addr = alloca i8, align 1
@@ -105,12 +105,12 @@ entry:
 ; CHECK: strh w1, [sp, #12]
 ; CHECK: str w2, [sp, #8]
 ; CHECK: str x3, [sp]
-; CHECK: ldrsb w0, [sp, #15]
-; CHECK: strh w0, [sp, #12]
-; CHECK: ldrsh w0, [sp, #12]
-; CHECK: str w0, [sp, #8]
-; CHECK: ldrsw x3, [sp, #8]
-; CHECK: str x3, [sp]
+; CHECK: ldrsb w8, [sp, #15]
+; CHECK: strh w8, [sp, #12]
+; CHECK: ldrsh w8, [sp, #12]
+; CHECK: str w8, [sp, #8]
+; CHECK: ldrsw x9, [sp, #8]
+; CHECK: str x9, [sp]
 ; CHECK: ldr x0, [sp]
 ; CHECK: ret
   %a.addr = alloca i8, align 1
@@ -166,7 +166,8 @@ entry:
 define signext i16 @sext_i1_i16(i1 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: sext_i1_i16
-; CHECK: sbfx w0, w0, #0, #1
+; CHECK: sbfx w8, w0, #0, #1
+; CHECK-NEXT: sxth     w0, w8
   %conv = sext i1 %a to i16
   ret i16 %conv
 }
@@ -175,7 +176,8 @@ entry:
 define signext i8 @sext_i1_i8(i1 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: sext_i1_i8
-; CHECK: sbfx w0, w0, #0, #1
+; CHECK: sbfx w8, w0, #0, #1
+; CHECK-NEXT: sxtb w0, w8
   %conv = sext i1 %a to i8
   ret i8 %conv
 }
@@ -238,8 +240,8 @@ entry:
 define float @sitofp_sw_i1(i1 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: sitofp_sw_i1
-; CHECK: sbfx w0, w0, #0, #1
-; CHECK: scvtf s0, w0
+; CHECK: sbfx w8, w0, #0, #1
+; CHECK: scvtf s0, w8
   %conv = sitofp i1 %a to float
   ret float %conv
 }
@@ -248,8 +250,8 @@ entry:
 define float @sitofp_sw_i8(i8 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: sitofp_sw_i8
-; CHECK: sxtb w0, w0
-; CHECK: scvtf s0, w0
+; CHECK: sxtb w8, w0
+; CHECK: scvtf s0, w8
   %conv = sitofp i8 %a to float
   ret float %conv
 }
@@ -302,8 +304,8 @@ entry:
 define float @uitofp_sw_i1(i1 %a) nounwind ssp {
 entry:
 ; CHECK-LABEL: uitofp_sw_i1
-; CHECK: and w0, w0, #0x1
-; CHECK: ucvtf s0, w0
+; CHECK: and w8, w0, #0x1
+; CHECK: ucvtf s0, w8
   %conv = uitofp i1 %a to float
   ret float %conv
 }

diff --git a/llvm/test/CodeGen/AArch64/swift-return.ll b/llvm/test/CodeGen/AArch64/swift-return.ll
index b909482..3d0bed4 100644 (file)
--- a/llvm/test/CodeGen/AArch64/swift-return.ll
+++ b/llvm/test/CodeGen/AArch64/swift-return.ll
@@ -8,7 +8,7 @@
 ; CHECK-O0-LABEL: test1
 ; CHECK-O0: bl      _gen
 ; CHECK-O0: sxth    [[TMP:w.*]], w0
-; CHECK-O0: add     w0, [[TMP]], w1, sxtb
+; CHECK-O0: add     w8, [[TMP]], w1, sxtb
 define i16 @test1(i32) {
 entry:
   %call = call swiftcc { i16, i8 } @gen(i32 %0)

diff --git a/llvm/test/CodeGen/AArch64/swifterror.ll b/llvm/test/CodeGen/AArch64/swifterror.ll
index 3c3ab60..55237b2 100644 (file)
--- a/llvm/test/CodeGen/AArch64/swifterror.ll
+++ b/llvm/test/CodeGen/AArch64/swifterror.ll
@@ -19,12 +19,11 @@ define float @foo(%swift_error** swifterror %error_ptr_ref) {
 ; CHECK-O0-LABEL: foo:
 ; CHECK-O0: orr w{{.*}}, wzr, #0x10
 ; CHECK-O0: malloc
-; CHECK-O0: mov x21, x0
-; CHECK-O0-NOT: x21
+; CHECK-O0: mov x1, x0
+; CHECK-O0-NOT: x1
 ; CHECK-O0: orr [[ID:w[0-9]+]], wzr, #0x1
-; CHECK-O0-NOT: x21
 ; CHECK-O0: strb [[ID]], [x0, #8]
-; CHECK-O0-NOT: x21
+; CHECK-O0: mov x21, x1
 entry:
   %call = call i8* @malloc(i64 16)
   %call.0 = bitcast i8* %call to %swift_error*

diff --git a/llvm/test/CodeGen/ARM/Windows/alloca.ll b/llvm/test/CodeGen/ARM/Windows/alloca.ll
index 203333e..7db854d 100644 (file)
--- a/llvm/test/CodeGen/ARM/Windows/alloca.ll
+++ b/llvm/test/CodeGen/ARM/Windows/alloca.ll
@@ -17,7 +17,7 @@ entry:
 ; CHECK: bl num_entries
 ; Any register is actually valid here, but turns out we use lr,
 ; because we do not have the kill flag on R0.
-; CHECK: mov.w [[R1:lr]], #7
+; CHECK: movs [[R1:r1]], #7
 ; CHECK: add.w [[R0:r[0-9]+]], [[R1]], [[R0]], lsl #2
 ; CHECK: bic [[R0]], [[R0]], #4
 ; CHECK: lsrs r4, [[R0]], #2

diff --git a/llvm/test/CodeGen/ARM/thumb-big-stack.ll b/llvm/test/CodeGen/ARM/thumb-big-stack.ll
index e5cbb97..954c126 100644 (file)
--- a/llvm/test/CodeGen/ARM/thumb-big-stack.ll
+++ b/llvm/test/CodeGen/ARM/thumb-big-stack.ll
@@ -12,7 +12,7 @@ target triple = "thumbv7s-apple-ios"
 ; CHECK: movw [[ADDR:(r[0-9]+|lr)]], #
 ; CHECK-NEXT: add [[ADDR]], sp
 ; CHECK-NEXT: vst1.64 {d{{[0-9]+}}, d{{[0-9]+}}}, {{\[}}[[ADDR]]:128]
-define <4 x float> @f(<4 x float> %x) {
+define <4 x float> @f(<4 x float> %x, float %val) {
 entry:
   %.compoundliteral7837 = alloca <4 x float>, align 16
   %.compoundliteral7865 = alloca <4 x float>, align 16
@@ -143,9 +143,9 @@ entry:
   %.compoundliteral13969 = alloca <4 x float>, align 16
   %.compoundliteral13983 = alloca <4 x float>, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40746999A0000000, float 0xC0719B3340000000, float 0xC070B66660000000, float 0xC07404CCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40746999A0000000, float 0xC0719B3340000000, float 0xC070B66660000000, float 0xC07404CCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40701B3340000000, float 0x405B866660000000, float 0xC0763999A0000000, float 4.895000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40701B3340000000, float 0x405B866660000000, float 0xC0763999A0000000, float 4.895000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -153,17 +153,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add68 = fadd <4 x float> %tmp1, %tmp
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add68, <4 x float>* undef, align 16
+  store volatile <4 x float> %add68, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp2 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add76 = fadd float undef, 0x4074C999A0000000
+  %add76 = fadd float %val, 0x4074C999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp3 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins77 = insertelement <4 x float> %tmp3, float %add76, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins77, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins77, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp4 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -175,15 +175,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins80 = insertelement <4 x float> %tmp5, float %add79, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins80, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins80, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40678CCCC0000000, float 0xC03E4CCCC0000000, float -4.170000e+02, float -1.220000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40678CCCC0000000, float 0xC03E4CCCC0000000, float -4.170000e+02, float -1.220000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp6 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add82 = fadd <4 x float> undef, %tmp6
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add82, <4 x float>* undef, align 16
+  store volatile <4 x float> %add82, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp7 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -195,19 +195,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins85 = insertelement <4 x float> %tmp8, float %add84, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins85, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins85, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp9 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext86 = extractelement <4 x float> %tmp9, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add93 = fadd float undef, 0xC076C66660000000
+  %add93 = fadd float %val, 0xC076C66660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp10 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins94 = insertelement <4 x float> %tmp10, float %add93, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x406C2999A0000000, float 8.050000e+01, float 0xC0794999A0000000, float 0xC073E4CCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x406C2999A0000000, float 8.050000e+01, float 0xC0794999A0000000, float 0xC073E4CCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp11 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -223,17 +223,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp14 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins102 = insertelement <4 x float> undef, float undef, i32 1
+  %vecins102 = insertelement <4 x float> undef, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins102, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins102, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp15 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add104 = fadd float undef, 0x406AB999A0000000
+  %add104 = fadd float %val, 0x406AB999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp16 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC0531999A0000000, float 0xC0737999A0000000, float 0x407CB33340000000, float 0xC06DCCCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC0531999A0000000, float 0xC0737999A0000000, float 0x407CB33340000000, float 0xC06DCCCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext579 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -243,7 +243,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins581 = insertelement <4 x float> %tmp17, float %add580, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins581, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins581, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp18 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -251,7 +251,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add583 = fadd float %vecext582, 0x40444CCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp19 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -261,25 +261,25 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins592 = insertelement <4 x float> undef, float %add591, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins592, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins592, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp20 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add594 = fadd float undef, 0xC05B466660000000
+  %add594 = fadd float %val, 0xC05B466660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add605 = fadd float undef, 0x407164CCC0000000
+  %add605 = fadd float %val, 0x407164CCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp21 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add616 = fadd float undef, 1.885000e+02
+  %add616 = fadd float %val, 1.885000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp22 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp23 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins620 = insertelement <4 x float> undef, float undef, i32 1
+  %vecins620 = insertelement <4 x float> undef, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins620, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins620, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext621 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -287,7 +287,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins623 = insertelement <4 x float> undef, float %add622, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins623, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins623, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp24 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -299,9 +299,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins626 = insertelement <4 x float> %tmp25, float %add625, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins626, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins626, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x404D0CCCC0000000, float 3.955000e+02, float 0xC0334CCCC0000000, float 0x40754E6660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x404D0CCCC0000000, float 3.955000e+02, float 0xC0334CCCC0000000, float 0x40754E6660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp26 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -309,7 +309,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add628 = fadd <4 x float> %tmp27, %tmp26
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add628, <4 x float>* undef, align 16
+  store volatile <4 x float> %add628, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp28 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -321,7 +321,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins631 = insertelement <4 x float> %tmp29, float %add630, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins631, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins631, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp30 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -333,7 +333,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins634 = insertelement <4 x float> %tmp31, float %add633, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins634, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins634, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp32 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -347,13 +347,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp35 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add658 = fadd float undef, 0xC04A4CCCC0000000
+  %add658 = fadd float %val, 0xC04A4CCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext663 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp36 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins665 = insertelement <4 x float> %tmp36, float undef, i32 2
+  %vecins665 = insertelement <4 x float> %tmp36, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext694 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -363,31 +363,31 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins696 = insertelement <4 x float> %tmp37, float %add695, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins696, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins696, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC069FCCCC0000000, float 0xC07C6E6660000000, float 0x4067E33340000000, float 0x4078DB3340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC069FCCCC0000000, float 0xC07C6E6660000000, float 0x4067E33340000000, float 0x4078DB3340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp38 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext699 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add703 = fadd float undef, 0x4068F33340000000
+  %add703 = fadd float %val, 0x4068F33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins704 = insertelement <4 x float> undef, float %add703, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins704, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins704, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp39 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp40 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins710 = insertelement <4 x float> %tmp40, float undef, i32 3
+  %vecins710 = insertelement <4 x float> %tmp40, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins710, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins710, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC05D9999A0000000, float 0x405D6CCCC0000000, float 0x40765CCCC0000000, float 0xC07C64CCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC05D9999A0000000, float 0x405D6CCCC0000000, float 0x40765CCCC0000000, float 0xC07C64CCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp41 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -395,7 +395,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add712 = fadd <4 x float> %tmp42, %tmp41
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add712, <4 x float>* undef, align 16
+  store volatile <4 x float> %add712, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp43 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -403,7 +403,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp44 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins715 = insertelement <4 x float> %tmp44, float undef, i32 0
+  %vecins715 = insertelement <4 x float> %tmp44, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp45 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -415,19 +415,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins718 = insertelement <4 x float> %tmp46, float %add717, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins718, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins718, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp47 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext719 = extractelement <4 x float> %tmp47, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add723 = fadd float undef, 0xC06A6CCCC0000000
+  %add723 = fadd float %val, 0xC06A6CCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins724 = insertelement <4 x float> undef, float %add723, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add726 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext730 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -437,19 +437,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins732 = insertelement <4 x float> %tmp48, float %add731, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins732, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins732, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp49 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext733 = extractelement <4 x float> %tmp49, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp50 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins738 = insertelement <4 x float> %tmp50, float undef, i32 3
+  %vecins738 = insertelement <4 x float> %tmp50, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x406E6CCCC0000000, float 0xC07A766660000000, float 0xC0608CCCC0000000, float 0xC063333340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x406E6CCCC0000000, float 0xC07A766660000000, float 0xC0608CCCC0000000, float 0xC063333340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp51 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -465,7 +465,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins743 = insertelement <4 x float> %tmp53, float %add742, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins743, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins743, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp54 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -473,7 +473,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add754 = fadd <4 x float> %tmp55, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add754, <4 x float>* undef, align 16
+  store volatile <4 x float> %add754, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp56 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -485,7 +485,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins757 = insertelement <4 x float> %tmp57, float %add756, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add765 = fadd float undef, 0x405BA66660000000
+  %add765 = fadd float %val, 0x405BA66660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp58 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -501,11 +501,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins771 = insertelement <4 x float> %tmp60, float %add770, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins771, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins771, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp61 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add776 = fadd float undef, 0xC055F33340000000
+  %add776 = fadd float %val, 0xC055F33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins777 = insertelement <4 x float> undef, float %add776, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -515,7 +515,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add782 = fadd <4 x float> %tmp63, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add782, <4 x float>* undef, align 16
+  store volatile <4 x float> %add782, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp64 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -523,25 +523,25 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add784 = fadd float %vecext783, -3.455000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC07A866660000000, float 0xC05CF999A0000000, float 0xC0757199A0000000, float -3.845000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC07A866660000000, float 0xC05CF999A0000000, float 0xC0757199A0000000, float -3.845000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add796 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add796, <4 x float>* undef, align 16
+  store volatile <4 x float> %add796, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp65 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add801 = fadd float undef, 3.045000e+02
+  %add801 = fadd float %val, 3.045000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp66 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins802 = insertelement <4 x float> %tmp66, float %add801, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins802, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins802, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext803 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp67 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -549,7 +549,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add810 = fadd <4 x float> undef, %tmp68
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add810, <4 x float>* undef, align 16
+  store volatile <4 x float> %add810, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp69 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -557,17 +557,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp70 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins813 = insertelement <4 x float> %tmp70, float undef, i32 0
+  %vecins813 = insertelement <4 x float> %tmp70, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext817 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add818 = fadd float %vecext817, -4.830000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins822 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins822 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins822, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins822, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 2.700000e+01, float 0xC05F666660000000, float 0xC07D0199A0000000, float 0x407A6CCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 2.700000e+01, float 0xC05F666660000000, float 0xC07D0199A0000000, float 0x407A6CCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp71 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -577,17 +577,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add838 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add838, <4 x float>* undef, align 16
+  store volatile <4 x float> %add838, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp73 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext839 = extractelement <4 x float> %tmp73, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add849 = fadd float undef, 0xC07C266660000000
+  %add849 = fadd float %val, 0xC07C266660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC07D566660000000, float 0xC06D233340000000, float 0x4068B33340000000, float 0xC07ADCCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC07D566660000000, float 0xC06D233340000000, float 0x4068B33340000000, float 0xC07ADCCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp74 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -609,9 +609,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins861 = insertelement <4 x float> %tmp77, float %add860, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins889 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins889 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins889, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins889, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp78 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -623,9 +623,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins892 = insertelement <4 x float> %tmp79, float %add891, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins892, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins892, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4063D33340000000, float 0xC076433340000000, float 0x407C966660000000, float 0xC07B5199A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4063D33340000000, float 0xC076433340000000, float 0x407C966660000000, float 0xC07B5199A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp80 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -633,7 +633,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add894 = fadd <4 x float> %tmp81, %tmp80
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add894, <4 x float>* undef, align 16
+  store volatile <4 x float> %add894, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext895 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -659,7 +659,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins903 = insertelement <4 x float> %tmp84, float %add902, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins903, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins903, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext904 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -669,7 +669,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins906 = insertelement <4 x float> %tmp85, float %add905, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC07EFCCCC0000000, float 1.795000e+02, float 0x407E3E6660000000, float 0x4070633340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC07EFCCCC0000000, float 1.795000e+02, float 0x407E3E6660000000, float 0x4070633340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp86 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -677,13 +677,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add908 = fadd <4 x float> %tmp87, %tmp86
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add908, <4 x float>* undef, align 16
+  store volatile <4 x float> %add908, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp88 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp89 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp90 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -703,7 +703,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins917 = insertelement <4 x float> %tmp92, float %add916, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins917, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins917, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp93 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -715,17 +715,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins920 = insertelement <4 x float> %tmp94, float %add919, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins920, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins920, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp95 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins925 = insertelement <4 x float> %tmp95, float undef, i32 0
+  %vecins925 = insertelement <4 x float> %tmp95, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins925, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins925, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp96 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add927 = fadd float undef, 0xC0501999A0000000
+  %add927 = fadd float %val, 0xC0501999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp97 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -739,7 +739,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins931 = insertelement <4 x float> %tmp98, float %add930, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC047B33340000000, float 0x404ACCCCC0000000, float 0x40708E6660000000, float 0x4060F999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC047B33340000000, float 0x404ACCCCC0000000, float 0x40708E6660000000, float 0x4060F999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp99 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -747,11 +747,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext937 = extractelement <4 x float> %tmp100, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add941 = fadd float undef, -4.665000e+02
+  %add941 = fadd float %val, -4.665000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins942 = insertelement <4 x float> undef, float %add941, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins942, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins942, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp101 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -763,29 +763,29 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins945 = insertelement <4 x float> %tmp102, float %add944, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins945, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins945, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp103 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add947 = fadd float undef, 0xC051933340000000
+  %add947 = fadd float %val, 0xC051933340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp104 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins948 = insertelement <4 x float> %tmp104, float %add947, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins948, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins948, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4060CCCCC0000000, float 0xC07BAB3340000000, float 0xC061233340000000, float 0xC076C199A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4060CCCCC0000000, float 0xC07BAB3340000000, float 0xC061233340000000, float 0xC076C199A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp105 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add955 = fadd float undef, 0x4077F4CCC0000000
+  %add955 = fadd float %val, 0x4077F4CCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp106 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins956 = insertelement <4 x float> %tmp106, float %add955, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins956, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins956, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext971 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -795,17 +795,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins973 = insertelement <4 x float> %tmp107, float %add972, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins973, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins973, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp108 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext974 = extractelement <4 x float> %tmp108, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins976 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins976 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins976, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins976, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x407E266660000000, float -1.225000e+02, float 0x407EB199A0000000, float 0x407BA199A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x407E266660000000, float -1.225000e+02, float 0x407EB199A0000000, float 0x407BA199A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp109 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -817,7 +817,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp112 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext982 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -825,7 +825,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins984 = insertelement <4 x float> undef, float %add983, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins984, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins984, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp113 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -837,25 +837,25 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins987 = insertelement <4 x float> %tmp114, float %add986, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins987, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins987, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp115 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp116 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins995 = insertelement <4 x float> %tmp116, float undef, i32 0
+  %vecins995 = insertelement <4 x float> %tmp116, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins995, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins995, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp117 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add997 = fadd float undef, 0xC0798999A0000000
+  %add997 = fadd float %val, 0xC0798999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp118 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins998 = insertelement <4 x float> %tmp118, float %add997, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins998, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins998, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp119 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -865,7 +865,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp120 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp121 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -879,13 +879,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1031 = fadd float %vecext1030, 2.010000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp123 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp124 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1085 = insertelement <4 x float> %tmp124, float undef, i32 2
+  %vecins1085 = insertelement <4 x float> %tmp124, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp125 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -897,13 +897,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1088 = insertelement <4 x float> %tmp126, float %add1087, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1088, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1088, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp127 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1090 = fadd <4 x float> undef, %tmp127
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp128 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -915,7 +915,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1096 = insertelement <4 x float> %tmp129, float %add1095, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1096, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1096, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp130 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -927,7 +927,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1099 = insertelement <4 x float> %tmp131, float %add1098, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1099, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1099, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp132 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -939,9 +939,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1102 = insertelement <4 x float> %tmp133, float %add1101, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1102, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1102, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4059866660000000, float 0x4072466660000000, float 0xC078FE6660000000, float 0xC058ACCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4059866660000000, float 0x4072466660000000, float 0xC078FE6660000000, float 0xC058ACCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp134 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -961,9 +961,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp137 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1110 = insertelement <4 x float> %tmp137, float undef, i32 1
+  %vecins1110 = insertelement <4 x float> %tmp137, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1110, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1110, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp138 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -975,21 +975,21 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1113 = insertelement <4 x float> %tmp139, float %add1112, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1113, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1113, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1115 = fadd float undef, 0x4072B33340000000
+  %add1115 = fadd float %val, 0x4072B33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1116 = insertelement <4 x float> undef, float %add1115, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1116, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1116, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC0721999A0000000, float 0x4075633340000000, float 0x40794199A0000000, float 0x4061066660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC0721999A0000000, float 0x4075633340000000, float 0x40794199A0000000, float 0x4061066660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp140 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1118 = fadd <4 x float> %tmp140, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add1118, <4 x float>* undef, align 16
+  store volatile <4 x float> %add1118, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp141 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -999,7 +999,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1121 = insertelement <4 x float> undef, float %add1120, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1121, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1121, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp142 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1013,9 +1013,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1125 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1127 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins1127 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1127, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1127, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp144 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1027,7 +1027,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1130 = insertelement <4 x float> %tmp145, float %add1129, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC06D6CCCC0000000, float 0xC032E66660000000, float -1.005000e+02, float 0x40765B3340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC06D6CCCC0000000, float 0xC032E66660000000, float -1.005000e+02, float 0x40765B3340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp146 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1045,7 +1045,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1135 = insertelement <4 x float> %tmp149, float %add1134, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1135, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1135, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp150 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1053,13 +1053,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp151 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1138 = insertelement <4 x float> %tmp151, float undef, i32 1
+  %vecins1138 = insertelement <4 x float> %tmp151, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1138, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1138, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp152 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1140 = fadd float undef, 0x407AE999A0000000
+  %add1140 = fadd float %val, 0x407AE999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp153 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1073,7 +1073,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1144 = insertelement <4 x float> %tmp154, float %add1143, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1144, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1144, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp155 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1081,27 +1081,27 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1146 = fadd <4 x float> %tmp156, %tmp155
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add1146, <4 x float>* undef, align 16
+  store volatile <4 x float> %add1146, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp157 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1148 = fadd float undef, 4.145000e+02
+  %add1148 = fadd float %val, 4.145000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp158 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1158 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins1158 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1158, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1158, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40603999A0000000, float -9.150000e+01, float 0xC051E66660000000, float -4.825000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40603999A0000000, float -9.150000e+01, float 0xC051E66660000000, float -4.825000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1218 = fadd float undef, 0xC078733340000000
+  %add1218 = fadd float %val, 0xC078733340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1219 = insertelement <4 x float> undef, float %add1218, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC0655CCCC0000000, float -4.900000e+01, float -4.525000e+02, float 4.205000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC0655CCCC0000000, float -4.900000e+01, float -4.525000e+02, float 4.205000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp159 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1113,7 +1113,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1281 = insertelement <4 x float> %tmp160, float %add1280, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1281, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1281, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp161 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1125,7 +1125,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1284 = insertelement <4 x float> %tmp162, float %add1283, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1284, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1284, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp163 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1133,27 +1133,27 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1286 = fadd <4 x float> %tmp164, %tmp163
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add1286, <4 x float>* undef, align 16
+  store volatile <4 x float> %add1286, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp165 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1288 = fadd float undef, 0xC0731199A0000000
+  %add1288 = fadd float %val, 0xC0731199A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp166 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp167 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1444 = extractelement <4 x float> %tmp167, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1460 = insertelement <4 x float> undef, float undef, i32 1
+  %vecins1460 = insertelement <4 x float> undef, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1460, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1460, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp168 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1462 = fadd float undef, -1.670000e+02
+  %add1462 = fadd float %val, -1.670000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1463 = insertelement <4 x float> undef, float %add1462, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1167,9 +1167,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1466 = insertelement <4 x float> %tmp170, float %add1465, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1466, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1466, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 3.885000e+02, float 0x4054266660000000, float -9.500000e+01, float 8.500000e+01>, <4 x float>* undef
+  store volatile <4 x float> <float 3.885000e+02, float 0x4054266660000000, float -9.500000e+01, float 8.500000e+01>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp171 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1177,17 +1177,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1468 = fadd <4 x float> %tmp172, %tmp171
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add1468, <4 x float>* undef, align 16
+  store volatile <4 x float> %add1468, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp173 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1470 = fadd float undef, 0x4033B33340000000
+  %add1470 = fadd float %val, 0x4033B33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp174 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1471 = insertelement <4 x float> %tmp174, float %add1470, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1471, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1471, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp175 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1205,9 +1205,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp178 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1477 = insertelement <4 x float> %tmp178, float undef, i32 2
+  %vecins1477 = insertelement <4 x float> %tmp178, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1477, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1477, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp179 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1219,15 +1219,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1480 = insertelement <4 x float> %tmp180, float %add1479, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1480, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1480, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC061B33340000000, float 3.290000e+02, float 0xC067766660000000, float 0x407DB33340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC061B33340000000, float 3.290000e+02, float 0xC067766660000000, float 0x407DB33340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp181 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp182 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp183 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1241,9 +1241,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1486 = extractelement <4 x float> %tmp185, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1502 = insertelement <4 x float> undef, float undef, i32 1
+  %vecins1502 = insertelement <4 x float> undef, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1502, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1502, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1503 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1253,7 +1253,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1505 = insertelement <4 x float> %tmp186, float %add1504, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1505, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1505, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp187 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1265,9 +1265,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1508 = insertelement <4 x float> %tmp188, float %add1507, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1508, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1508, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40762B3340000000, float 0xC074566660000000, float 0xC07C74CCC0000000, float 0xC053F999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40762B3340000000, float 0xC074566660000000, float 0xC07C74CCC0000000, float 0xC053F999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp189 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1275,7 +1275,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1510 = fadd <4 x float> %tmp190, %tmp189
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add1510, <4 x float>* undef, align 16
+  store volatile <4 x float> %add1510, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp191 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1289,13 +1289,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1656 = insertelement <4 x float> %tmp193, float %add1655, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1656, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1656, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1658 = fadd float undef, 0x40709999A0000000
+  %add1658 = fadd float %val, 0x40709999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp194 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1660 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1305,19 +1305,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1662 = insertelement <4 x float> %tmp195, float %add1661, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1662, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1662, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC075266660000000, float 0xC072C4CCC0000000, float 0x407C4E6660000000, float -4.485000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC075266660000000, float 0xC072C4CCC0000000, float 0x407C4E6660000000, float -4.485000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1676 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins1676 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp196 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1692 = fadd <4 x float> %tmp196, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add1692, <4 x float>* undef, align 16
+  store volatile <4 x float> %add1692, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp197 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1329,7 +1329,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1695 = insertelement <4 x float> %tmp198, float %add1694, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1695, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1695, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp199 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1341,7 +1341,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1698 = insertelement <4 x float> %tmp200, float %add1697, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1698, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1698, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp201 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1349,15 +1349,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp202 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1701 = insertelement <4 x float> %tmp202, float undef, i32 2
+  %vecins1701 = insertelement <4 x float> %tmp202, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1701, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1701, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp203 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1704 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins1704 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC075933340000000, float 0xC0489999A0000000, float 0xC078AB3340000000, float 0x406DFCCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC075933340000000, float 0xC0489999A0000000, float 0xC078AB3340000000, float 0x406DFCCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp204 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1365,9 +1365,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp206 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1709 = insertelement <4 x float> %tmp206, float undef, i32 0
+  %vecins1709 = insertelement <4 x float> %tmp206, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1709, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1709, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp207 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1375,11 +1375,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1714 = fadd float %vecext1713, 0xC0703199A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1723 = insertelement <4 x float> undef, float undef, i32 0
+  %vecins1723 = insertelement <4 x float> undef, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp208 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1730 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1389,9 +1389,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1732 = insertelement <4 x float> %tmp209, float %add1731, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1732, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1732, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40551999A0000000, float 0xC0708999A0000000, float 0xC054F33340000000, float 0xC07C5999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40551999A0000000, float 0xC0708999A0000000, float 0xC054F33340000000, float 0xC07C5999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp210 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1399,7 +1399,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp211 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1736 = fadd float undef, 0x407C3999A0000000
+  %add1736 = fadd float %val, 0x407C3999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp212 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1415,7 +1415,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1740 = insertelement <4 x float> %tmp214, float %add1739, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1740, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1740, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp215 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1427,25 +1427,25 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1743 = insertelement <4 x float> %tmp216, float %add1742, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1743, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1743, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1744 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp217 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1746 = insertelement <4 x float> %tmp217, float undef, i32 3
+  %vecins1746 = insertelement <4 x float> %tmp217, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC076466660000000, float 0x4060BCCCC0000000, float 0x405EF999A0000000, float 0x4074766660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC076466660000000, float 0x4060BCCCC0000000, float 0x405EF999A0000000, float 0x4074766660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp218 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1748 = fadd <4 x float> undef, %tmp218
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add1748, <4 x float>* undef, align 16
+  store volatile <4 x float> %add1748, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp219 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1750 = fadd float undef, 0x407C6B3340000000
+  %add1750 = fadd float %val, 0x407C6B3340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1751 = insertelement <4 x float> undef, float %add1750, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1467,21 +1467,21 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp223 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1759 = fadd float undef, 0x40678999A0000000
+  %add1759 = fadd float %val, 0x40678999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp224 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1760 = insertelement <4 x float> %tmp224, float %add1759, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1760, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1760, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x405E333340000000, float 0x40571999A0000000, float 0xC02E333340000000, float 0x4053A66660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x405E333340000000, float 0x40571999A0000000, float 0xC02E333340000000, float 0x4053A66660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp225 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1762 = fadd <4 x float> undef, %tmp225
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add1762, <4 x float>* undef, align 16
+  store volatile <4 x float> %add1762, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp226 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1493,7 +1493,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1765 = insertelement <4 x float> %tmp227, float %add1764, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1765, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1765, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp228 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1505,7 +1505,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1768 = insertelement <4 x float> %tmp229, float %add1767, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1768, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1768, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1769 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1515,7 +1515,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1771 = insertelement <4 x float> %tmp230, float %add1770, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1771, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1771, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp231 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1525,13 +1525,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp234 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1779 = insertelement <4 x float> %tmp234, float undef, i32 0
+  %vecins1779 = insertelement <4 x float> %tmp234, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1779, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1779, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp235 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp236 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1541,9 +1541,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1785 = insertelement <4 x float> undef, float %add1784, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1785, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1785, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC07074CCC0000000, float 0xC04D666660000000, float 3.235000e+02, float 0xC0724199A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC07074CCC0000000, float 0xC04D666660000000, float 3.235000e+02, float 0xC0724199A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp237 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1559,25 +1559,25 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1793 = insertelement <4 x float> %tmp239, float %add1792, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1793, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1793, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp240 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1795 = fadd float undef, 0x4055266660000000
+  %add1795 = fadd float %val, 0x4055266660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp241 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1796 = insertelement <4 x float> %tmp241, float %add1795, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1799 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins1799 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1800 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp242 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float -6.600000e+01, float 0xC07B2199A0000000, float 0x4011333340000000, float 0xC0635CCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float -6.600000e+01, float 0xC07B2199A0000000, float 0x4011333340000000, float 0xC0635CCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp243 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1587,7 +1587,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp246 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1865 = fadd float undef, -2.235000e+02
+  %add1865 = fadd float %val, -2.235000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp247 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1597,33 +1597,33 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp249 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1872 = insertelement <4 x float> %tmp249, float undef, i32 3
+  %vecins1872 = insertelement <4 x float> %tmp249, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x406B8999A0000000, float 0xC0696CCCC0000000, float 0xC07A34CCC0000000, float 0x407654CCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x406B8999A0000000, float 0xC0696CCCC0000000, float 0xC07A34CCC0000000, float 0x407654CCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp250 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1874 = fadd <4 x float> %tmp250, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add1874, <4 x float>* undef, align 16
+  store volatile <4 x float> %add1874, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1875 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp251 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1894 = insertelement <4 x float> %tmp251, float undef, i32 1
+  %vecins1894 = insertelement <4 x float> %tmp251, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp252 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1895 = extractelement <4 x float> %tmp252, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1900 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins1900 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1900, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1900, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1905 = insertelement <4 x float> undef, float undef, i32 0
+  %vecins1905 = insertelement <4 x float> undef, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1905, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1905, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp253 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1633,7 +1633,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1908 = insertelement <4 x float> undef, float %add1907, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1908, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1908, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1909 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1649,23 +1649,23 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1916 = fadd <4 x float> %tmp256, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add1916, <4 x float>* undef, align 16
+  store volatile <4 x float> %add1916, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1923 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp257 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add1927 = fadd float undef, 0x40761999A0000000
+  %add1927 = fadd float %val, 0x40761999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp258 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1928 = insertelement <4 x float> %tmp258, float %add1927, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1928, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1928, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 7.100000e+01, float 0xC0634999A0000000, float 0x407B0B3340000000, float 0xC07DE999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 7.100000e+01, float 0xC0634999A0000000, float 0x407B0B3340000000, float 0xC07DE999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp259 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1677,9 +1677,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp262 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1933 = insertelement <4 x float> %tmp262, float undef, i32 0
+  %vecins1933 = insertelement <4 x float> %tmp262, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1933, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1933, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp263 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1693,15 +1693,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1940 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1942 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins1942 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float -8.200000e+01, float 0xC04C733340000000, float 0xC077ACCCC0000000, float 0x4074566660000000>, <4 x float>* undef
+  store volatile <4 x float> <float -8.200000e+01, float 0xC04C733340000000, float 0xC077ACCCC0000000, float 0x4074566660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp265 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp266 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp267 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1709,13 +1709,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add1946 = fadd float %vecext1945, 0xC074866660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1953 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins1953 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1953, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1953, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp268 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp269 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1737,15 +1737,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1964 = insertelement <4 x float> %tmp272, float %add1963, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1964, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1964, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1965 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp273 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1967 = insertelement <4 x float> %tmp273, float undef, i32 2
+  %vecins1967 = insertelement <4 x float> %tmp273, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1967, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1967, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp274 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1757,9 +1757,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1970 = insertelement <4 x float> %tmp275, float %add1969, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1970, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1970, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x402E9999A0000000, float 0x407344CCC0000000, float -4.165000e+02, float 0x4078FCCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x402E9999A0000000, float 0x407344CCC0000000, float -4.165000e+02, float 0x4078FCCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp276 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1767,31 +1767,31 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp278 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1975 = insertelement <4 x float> %tmp278, float undef, i32 0
+  %vecins1975 = insertelement <4 x float> %tmp278, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1975, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1975, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp279 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1976 = extractelement <4 x float> %tmp279, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1978 = insertelement <4 x float> undef, float undef, i32 1
+  %vecins1978 = insertelement <4 x float> undef, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1978, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1978, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1979 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1981 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins1981 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1981, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1981, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins1984 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins1984 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1984, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1984, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC06A766660000000, float 0xC07CE4CCC0000000, float -1.055000e+02, float 0x40786E6660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC06A766660000000, float 0xC07CE4CCC0000000, float -1.055000e+02, float 0x40786E6660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext1990 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1803,11 +1803,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins1998 = insertelement <4 x float> %tmp280, float %add1997, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins1998, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins1998, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC0794E6660000000, float 0xC073CCCCC0000000, float 0x407994CCC0000000, float 6.500000e+01>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC0794E6660000000, float 0xC073CCCCC0000000, float 0x407994CCC0000000, float 6.500000e+01>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2004 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1817,7 +1817,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2006 = insertelement <4 x float> %tmp281, float %add2005, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2006, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2006, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp282 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1825,7 +1825,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp283 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins2009 = insertelement <4 x float> %tmp283, float undef, i32 2
+  %vecins2009 = insertelement <4 x float> %tmp283, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp284 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1837,15 +1837,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2012 = insertelement <4 x float> %tmp285, float %add2011, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2012, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2012, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC04E733340000000, float 0xC074566660000000, float 0x4079F66660000000, float 0xC0705B3340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC04E733340000000, float 0xC074566660000000, float 0x4079F66660000000, float 0xC0705B3340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp286 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp287 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp288 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1857,7 +1857,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2017 = insertelement <4 x float> %tmp289, float %add2016, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add2022 = fadd float undef, 8.350000e+01
+  %add2022 = fadd float %val, 8.350000e+01
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp290 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1871,7 +1871,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add2028 = fadd <4 x float> %tmp292, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add2028, <4 x float>* undef, align 16
+  store volatile <4 x float> %add2028, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2029 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1879,11 +1879,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp293 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp294 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add2036 = fadd float undef, 0x407DE66660000000
+  %add2036 = fadd float %val, 0x407DE66660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp295 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1895,9 +1895,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp299 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins2045 = insertelement <4 x float> %tmp299, float undef, i32 0
+  %vecins2045 = insertelement <4 x float> %tmp299, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2045, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2045, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp300 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1905,35 +1905,35 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add2047 = fadd float %vecext2046, 0xC065433340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2052 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp301 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins2054 = insertelement <4 x float> %tmp301, float undef, i32 3
+  %vecins2054 = insertelement <4 x float> %tmp301, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2054, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2054, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4024666660000000, float 0x4079366660000000, float 0x40721B3340000000, float 0x406E533340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4024666660000000, float 0x4079366660000000, float 0x40721B3340000000, float 0x406E533340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp302 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add2056 = fadd <4 x float> undef, %tmp302
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add2056, <4 x float>* undef, align 16
+  store volatile <4 x float> %add2056, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp303 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp304 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins2062 = insertelement <4 x float> %tmp304, float undef, i32 1
+  %vecins2062 = insertelement <4 x float> %tmp304, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2062, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2062, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp305 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp306 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1943,9 +1943,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2068 = insertelement <4 x float> undef, float %add2067, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2068, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2068, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC07EFCCCC0000000, float -3.420000e+02, float 0xC07BC999A0000000, float 0x40751999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC07EFCCCC0000000, float -3.420000e+02, float 0xC07BC999A0000000, float 0x40751999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp307 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1953,7 +1953,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add2070 = fadd <4 x float> %tmp308, %tmp307
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add2070, <4 x float>* undef, align 16
+  store volatile <4 x float> %add2070, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp309 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1965,7 +1965,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2073 = insertelement <4 x float> %tmp310, float %add2072, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2073, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2073, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp311 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1973,7 +1973,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp312 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins2076 = insertelement <4 x float> %tmp312, float undef, i32 1
+  %vecins2076 = insertelement <4 x float> %tmp312, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp313 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1985,7 +1985,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2079 = insertelement <4 x float> %tmp314, float %add2078, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2079, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2079, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp315 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -1997,15 +1997,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2082 = insertelement <4 x float> %tmp316, float %add2081, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2082, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2082, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40764E6660000000, float 0x40501999A0000000, float 0xC079A4CCC0000000, float 0x4050533340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40764E6660000000, float 0x40501999A0000000, float 0xC079A4CCC0000000, float 0x4050533340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp317 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp318 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp319 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2015,7 +2015,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2087 = insertelement <4 x float> undef, float %add2086, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2087, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2087, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2480 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2029,23 +2029,23 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2485 = insertelement <4 x float> %tmp320, float %add2484, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2485, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2485, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp321 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add2487 = fadd float undef, 2.030000e+02
+  %add2487 = fadd float %val, 2.030000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp322 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4073DE6660000000, float 0x4067CCCCC0000000, float 0xC03F1999A0000000, float 4.350000e+01>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4073DE6660000000, float 0x4067CCCCC0000000, float 0xC03F1999A0000000, float 4.350000e+01>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2491 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp323 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp324 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2055,9 +2055,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp325 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins2499 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins2499 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2499, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2499, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2500 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2079,7 +2079,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp329 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add2534 = fadd float undef, 0x4072C66660000000
+  %add2534 = fadd float %val, 0x4072C66660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2536 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2089,15 +2089,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2538 = insertelement <4 x float> %tmp330, float %add2537, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2538, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2538, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2539 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add2540 = fadd float %vecext2539, 0x406F9999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins2580 = insertelement <4 x float> undef, float undef, i32 1
+  %vecins2580 = insertelement <4 x float> undef, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2580, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2580, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp331 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2107,7 +2107,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2583 = insertelement <4 x float> undef, float %add2582, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2583, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2583, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2584 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2115,21 +2115,21 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp332 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40773199A0000000, float 0x407D7999A0000000, float 0xC0717199A0000000, float 0xC07E9CCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40773199A0000000, float 0x407D7999A0000000, float 0xC0717199A0000000, float 0xC07E9CCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add2590 = fadd float undef, 0x407B1999A0000000
+  %add2590 = fadd float %val, 0x407B1999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp333 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp334 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add2672 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add2672, <4 x float>* undef, align 16
+  store volatile <4 x float> %add2672, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp335 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2141,37 +2141,37 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2678 = insertelement <4 x float> %tmp336, float %add2677, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2678, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2678, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp337 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2679 = extractelement <4 x float> %tmp337, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins2681 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins2681 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2681, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2681, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp338 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext2682 = extractelement <4 x float> %tmp338, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins2684 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins2684 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp339 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp340 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp341 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add2688 = fadd float undef, 0x4063266660000000
+  %add2688 = fadd float %val, 0x4063266660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins2692 = insertelement <4 x float> undef, float undef, i32 1
+  %vecins2692 = insertelement <4 x float> undef, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2692, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2692, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp342 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2183,9 +2183,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins2698 = insertelement <4 x float> %tmp343, float %add2697, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins2698, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins2698, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40547999A0000000, float 0xC060633340000000, float 0x4075766660000000, float 0x4072D33340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40547999A0000000, float 0xC060633340000000, float 0x4075766660000000, float 0x4072D33340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp344 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2193,7 +2193,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add2700 = fadd <4 x float> %tmp345, %tmp344
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add2700, <4 x float>* undef, align 16
+  store volatile <4 x float> %add2700, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp346 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2207,15 +2207,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp349 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3121 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3125 = fadd float undef, 0xC06F266660000000
+  %add3125 = fadd float %val, 0xC06F266660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3126 = insertelement <4 x float> undef, float %add3125, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3126, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3126, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp350 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2227,11 +2227,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3129 = insertelement <4 x float> %tmp351, float %add3128, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3129, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3129, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp352 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3131 = fadd float undef, 3.215000e+02
+  %add3131 = fadd float %val, 3.215000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp353 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2239,15 +2239,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add3134 = fadd <4 x float> %tmp354, %tmp353
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add3134, <4 x float>* undef, align 16
+  store volatile <4 x float> %add3134, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp355 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3136 = fadd float undef, 0x4074333340000000
+  %add3136 = fadd float %val, 0x4074333340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins3140 = insertelement <4 x float> undef, float undef, i32 1
+  %vecins3140 = insertelement <4 x float> undef, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3140, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3140, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp356 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2259,7 +2259,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3143 = insertelement <4 x float> %tmp357, float %add3142, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3143, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3143, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp358 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2271,15 +2271,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3146 = insertelement <4 x float> %tmp359, float %add3145, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3146, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3146, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp360 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins3272 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins3272 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3272, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3272, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x407B4999A0000000, float 0x40695CCCC0000000, float 0xC05C0CCCC0000000, float 0x407EB33340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x407B4999A0000000, float 0x40695CCCC0000000, float 0xC05C0CCCC0000000, float 0x407EB33340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp361 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2287,7 +2287,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add3274 = fadd <4 x float> %tmp362, %tmp361
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add3274, <4 x float>* undef, align 16
+  store volatile <4 x float> %add3274, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp363 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2299,7 +2299,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3277 = insertelement <4 x float> %tmp364, float %add3276, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3277, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3277, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp365 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2309,7 +2309,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3280 = insertelement <4 x float> undef, float %add3279, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3280, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3280, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp366 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2321,7 +2321,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3283 = insertelement <4 x float> %tmp367, float %add3282, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3283, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3283, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp368 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2333,7 +2333,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp369 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp370 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2345,7 +2345,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3291 = insertelement <4 x float> %tmp371, float %add3290, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3291, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3291, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3292 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2353,11 +2353,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp373 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins3328 = insertelement <4 x float> %tmp373, float undef, i32 3
+  %vecins3328 = insertelement <4 x float> %tmp373, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add3330 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add3330, <4 x float>* undef, align 16
+  store volatile <4 x float> %add3330, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3331 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2367,7 +2367,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3333 = insertelement <4 x float> %tmp374, float %add3332, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3333, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3333, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3334 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2385,7 +2385,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3339 = insertelement <4 x float> %tmp376, float %add3338, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3339, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3339, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp377 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2393,13 +2393,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp378 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins3342 = insertelement <4 x float> %tmp378, float undef, i32 3
+  %vecins3342 = insertelement <4 x float> %tmp378, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp379 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add3344 = fadd <4 x float> %tmp379, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add3344, <4 x float>* undef, align 16
+  store volatile <4 x float> %add3344, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp380 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2419,15 +2419,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3350 = insertelement <4 x float> %tmp382, float %add3349, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3350, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3350, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3352 = fadd float undef, 0xC06ACCCCC0000000
+  %add3352 = fadd float %val, 0xC06ACCCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp383 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins3423 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins3423 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3423, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3423, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3424 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2437,9 +2437,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3426 = insertelement <4 x float> %tmp384, float %add3425, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3426, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3426, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 2.795000e+02, float -4.065000e+02, float 0xC05CD999A0000000, float 1.825000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 2.795000e+02, float -4.065000e+02, float 0xC05CD999A0000000, float 1.825000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp385 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2457,7 +2457,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3431 = insertelement <4 x float> %tmp388, float %add3430, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3431, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3431, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp389 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2469,15 +2469,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3434 = insertelement <4 x float> %tmp390, float %add3433, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3434, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3434, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3435 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp391 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins3437 = insertelement <4 x float> %tmp391, float undef, i32 2
+  %vecins3437 = insertelement <4 x float> %tmp391, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3437, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3437, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp392 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2485,7 +2485,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add3439 = fadd float %vecext3438, 0xC071D999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC0798199A0000000, float -3.385000e+02, float 0xC050066660000000, float 0xC075E999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC0798199A0000000, float -3.385000e+02, float 0xC050066660000000, float 0xC075E999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp393 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2493,7 +2493,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add3442 = fadd <4 x float> %tmp394, %tmp393
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add3442, <4 x float>* undef, align 16
+  store volatile <4 x float> %add3442, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3443 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2509,7 +2509,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3448 = insertelement <4 x float> %tmp396, float %add3447, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3448, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3448, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp397 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2521,15 +2521,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3451 = insertelement <4 x float> %tmp398, float %add3450, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3451, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3451, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3453 = fadd float undef, 0xC07ADCCCC0000000
+  %add3453 = fadd float %val, 0xC07ADCCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp399 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3454 = insertelement <4 x float> %tmp399, float %add3453, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3454, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3454, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp400 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2539,7 +2539,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3459 = insertelement <4 x float> undef, float %add3458, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3459, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3459, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp401 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2547,19 +2547,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp402 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins3462 = insertelement <4 x float> %tmp402, float undef, i32 1
+  %vecins3462 = insertelement <4 x float> %tmp402, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3462, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3462, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp403 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3464 = fadd float undef, 0xC057B999A0000000
+  %add3464 = fadd float %val, 0xC057B999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp404 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3465 = insertelement <4 x float> %tmp404, float %add3464, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3465, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3465, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp405 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2569,21 +2569,21 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp406 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x405C3999A0000000, float 0xC07C6B3340000000, float 0x407ACB3340000000, float 0xC06E0999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x405C3999A0000000, float 0xC07C6B3340000000, float 0x407ACB3340000000, float 0xC06E0999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp407 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp408 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3477 = extractelement <4 x float> %tmp408, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins3479 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins3479 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3479, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3479, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3480 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2593,23 +2593,23 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3482 = insertelement <4 x float> %tmp409, float %add3481, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3482, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3482, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 3.565000e+02, float 0xC0464CCCC0000000, float 0x4037666660000000, float 0xC0788CCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 3.565000e+02, float 0xC0464CCCC0000000, float 0x4037666660000000, float 0xC0788CCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp410 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add3484 = fadd <4 x float> %tmp410, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add3484, <4 x float>* undef, align 16
+  store volatile <4 x float> %add3484, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp411 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3486 = fadd float undef, -1.415000e+02
+  %add3486 = fadd float %val, -1.415000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3487 = insertelement <4 x float> undef, float %add3486, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3487, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3487, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp412 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2621,25 +2621,25 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3490 = insertelement <4 x float> %tmp413, float %add3489, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3490, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3490, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3492 = fadd float undef, 0x4078066660000000
+  %add3492 = fadd float %val, 0x4078066660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp414 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3493 = insertelement <4 x float> %tmp414, float %add3492, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3493, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3493, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp415 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3495 = fadd float undef, 0xC0798999A0000000
+  %add3495 = fadd float %val, 0xC0798999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp416 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3496 = insertelement <4 x float> %tmp416, float %add3495, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3496, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3496, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp417 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2647,7 +2647,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add3498 = fadd <4 x float> %tmp418, %tmp417
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add3498, <4 x float>* undef, align 16
+  store volatile <4 x float> %add3498, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3499 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2663,25 +2663,25 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp420 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3506 = fadd float undef, 0xC074DB3340000000
+  %add3506 = fadd float %val, 0xC074DB3340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp421 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins3507 = insertelement <4 x float> %tmp421, float %add3506, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins3507, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins3507, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add3509 = fadd float undef, 0xC066033340000000
+  %add3509 = fadd float %val, 0xC066033340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp422 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x404B333340000000, float 4.680000e+02, float 0x40577999A0000000, float 0xC07D9999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x404B333340000000, float 4.680000e+02, float 0x40577999A0000000, float 0xC07D9999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp423 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3513 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2693,9 +2693,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext3516 = extractelement <4 x float> %tmp425, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5414 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins5414 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5414, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5414, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp426 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2703,33 +2703,33 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5416 = fadd <4 x float> %tmp427, %tmp426
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add5416, <4 x float>* undef, align 16
+  store volatile <4 x float> %add5416, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp428 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add5418 = fadd float undef, 0xC07ED999A0000000
+  %add5418 = fadd float %val, 0xC07ED999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp429 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5419 = insertelement <4 x float> %tmp429, float %add5418, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5624 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins5624 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5624, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5624, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC07B4999A0000000, float 0x4078B33340000000, float 0xC07674CCC0000000, float 0xC07C533340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC07B4999A0000000, float 0x4078B33340000000, float 0xC07674CCC0000000, float 0xC07C533340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5626 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add5626, <4 x float>* undef, align 16
+  store volatile <4 x float> %add5626, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext5627 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp430 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5629 = insertelement <4 x float> %tmp430, float undef, i32 0
+  %vecins5629 = insertelement <4 x float> %tmp430, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5629, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5629, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp431 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2739,13 +2739,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5632 = insertelement <4 x float> undef, float %add5631, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5632, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5632, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp432 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5688 = insertelement <4 x float> %tmp432, float undef, i32 1
+  %vecins5688 = insertelement <4 x float> %tmp432, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5688, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5688, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp433 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2753,35 +2753,35 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp434 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5691 = insertelement <4 x float> %tmp434, float undef, i32 2
+  %vecins5691 = insertelement <4 x float> %tmp434, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5691, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5691, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext5692 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float -4.350000e+02, float 0xC0775CCCC0000000, float 0xC0714999A0000000, float 0xC0661999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float -4.350000e+02, float 0xC0775CCCC0000000, float 0xC0714999A0000000, float 0xC0661999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp435 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5696 = fadd <4 x float> undef, %tmp435
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add5696, <4 x float>* undef, align 16
+  store volatile <4 x float> %add5696, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add5701 = fadd float undef, 0x4077D4CCC0000000
+  %add5701 = fadd float %val, 0x4077D4CCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp436 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5702 = insertelement <4 x float> %tmp436, float %add5701, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5702, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5702, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp437 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp438 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5705 = insertelement <4 x float> %tmp438, float undef, i32 2
+  %vecins5705 = insertelement <4 x float> %tmp438, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5705, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5705, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp439 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2793,9 +2793,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5708 = insertelement <4 x float> %tmp440, float %add5707, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5708, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5708, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x405D666660000000, float 0xC069333340000000, float 0x407B6B3340000000, float 0xC06EB33340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x405D666660000000, float 0xC069333340000000, float 0x407B6B3340000000, float 0xC06EB33340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp441 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2803,7 +2803,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5710 = fadd <4 x float> %tmp442, %tmp441
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add5710, <4 x float>* undef, align 16
+  store volatile <4 x float> %add5710, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp443 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2815,19 +2815,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5713 = insertelement <4 x float> %tmp444, float %add5712, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5713, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5713, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp445 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp446 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5716 = insertelement <4 x float> %tmp446, float undef, i32 1
+  %vecins5716 = insertelement <4 x float> %tmp446, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp447 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5724 = fadd <4 x float> %tmp447, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add5724, <4 x float>* undef, align 16
+  store volatile <4 x float> %add5724, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp448 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2835,21 +2835,21 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp449 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5750 = insertelement <4 x float> %tmp449, float undef, i32 3
+  %vecins5750 = insertelement <4 x float> %tmp449, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40692999A0000000, float 0xC07C4CCCC0000000, float 0x407D1E6660000000, float 0x407B4199A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40692999A0000000, float 0xC07C4CCCC0000000, float 0x407D1E6660000000, float 0x407B4199A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp450 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5752 = fadd <4 x float> undef, %tmp450
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add5754 = fadd float undef, 0xC064033340000000
+  %add5754 = fadd float %val, 0xC064033340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp451 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5755 = insertelement <4 x float> %tmp451, float %add5754, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5755, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5755, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp452 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2861,7 +2861,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5758 = insertelement <4 x float> %tmp453, float %add5757, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5758, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5758, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp454 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2869,9 +2869,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp455 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5761 = insertelement <4 x float> %tmp455, float undef, i32 2
+  %vecins5761 = insertelement <4 x float> %tmp455, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5761, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5761, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp456 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2883,13 +2883,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5764 = insertelement <4 x float> %tmp457, float %add5763, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5764, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5764, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x407A6B3340000000, float 0x40470CCCC0000000, float 0xC076F4CCC0000000, float 0x40791999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x407A6B3340000000, float 0x40470CCCC0000000, float 0xC076F4CCC0000000, float 0x40791999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5766 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add5766, <4 x float>* undef, align 16
+  store volatile <4 x float> %add5766, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp458 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2901,9 +2901,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5769 = insertelement <4 x float> %tmp459, float %add5768, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5769, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5769, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add5771 = fadd float undef, 8.000000e+00
+  %add5771 = fadd float %val, 8.000000e+00
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp460 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2911,11 +2911,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp461 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add5796 = fadd float undef, 0x4058ECCCC0000000
+  %add5796 = fadd float %val, 0x4058ECCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5797 = insertelement <4 x float> undef, float %add5796, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5797, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5797, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp462 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2923,7 +2923,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp463 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5800 = insertelement <4 x float> %tmp463, float undef, i32 1
+  %vecins5800 = insertelement <4 x float> %tmp463, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp464 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2935,7 +2935,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5803 = insertelement <4 x float> %tmp465, float %add5802, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5803, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5803, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp466 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2947,11 +2947,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5806 = insertelement <4 x float> %tmp467, float %add5805, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5806, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5806, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp468 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp469 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2961,7 +2961,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp470 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp471 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2973,9 +2973,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5820 = insertelement <4 x float> %tmp472, float %add5819, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5820, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5820, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40514CCCC0000000, float 0x406A7999A0000000, float 0xC078766660000000, float 0xC0522CCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40514CCCC0000000, float 0x406A7999A0000000, float 0xC078766660000000, float 0xC0522CCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp473 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2983,7 +2983,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5822 = fadd <4 x float> %tmp474, %tmp473
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add5822, <4 x float>* undef, align 16
+  store volatile <4 x float> %add5822, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp475 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -2991,7 +2991,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp476 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5825 = insertelement <4 x float> %tmp476, float undef, i32 0
+  %vecins5825 = insertelement <4 x float> %tmp476, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp477 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3003,7 +3003,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5828 = insertelement <4 x float> %tmp478, float %add5827, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5828, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5828, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp479 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3015,19 +3015,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5831 = insertelement <4 x float> %tmp480, float %add5830, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float -3.370000e+02, float 0xC072DE6660000000, float -2.670000e+02, float 0x4062333340000000>, <4 x float>* undef
+  store volatile <4 x float> <float -3.370000e+02, float 0xC072DE6660000000, float -2.670000e+02, float 0x4062333340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp481 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext5837 = extractelement <4 x float> %tmp481, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5839 = insertelement <4 x float> undef, float undef, i32 0
+  %vecins5839 = insertelement <4 x float> undef, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5839, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5839, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp482 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3035,33 +3035,33 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp483 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5842 = insertelement <4 x float> %tmp483, float undef, i32 1
+  %vecins5842 = insertelement <4 x float> %tmp483, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5842, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5842, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp484 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp485 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5845 = insertelement <4 x float> %tmp485, float undef, i32 2
+  %vecins5845 = insertelement <4 x float> %tmp485, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5845, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5845, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC06EC999A0000000, float 0x406D5999A0000000, float 0x4056F33340000000, float 0xC07E14CCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC06EC999A0000000, float 0x406D5999A0000000, float 0x4056F33340000000, float 0xC07E14CCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5850 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add5850, <4 x float>* undef, align 16
+  store volatile <4 x float> %add5850, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp486 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add5852 = fadd float undef, 2.985000e+02
+  %add5852 = fadd float %val, 2.985000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp487 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5853 = insertelement <4 x float> %tmp487, float %add5852, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5853, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5853, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp488 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3073,17 +3073,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5856 = insertelement <4 x float> %tmp489, float %add5855, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5856, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5856, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp490 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add5858 = fadd float undef, 0x4071666660000000
+  %add5858 = fadd float %val, 0x4071666660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp491 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5859 = insertelement <4 x float> %tmp491, float %add5858, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5859, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5859, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp492 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3099,19 +3099,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5901 = insertelement <4 x float> %tmp494, float %add5900, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5901, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5901, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add5914 = fadd float undef, 0x40786E6660000000
+  %add5914 = fadd float %val, 0x40786E6660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins5918 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins5918 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5918, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5918, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x406F266660000000, float 7.900000e+01, float -4.695000e+02, float -4.880000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0x406F266660000000, float 7.900000e+01, float -4.695000e+02, float -4.880000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5920 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add5920, <4 x float>* undef, align 16
+  store volatile <4 x float> %add5920, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add5934 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3121,7 +3121,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp495 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp496 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3131,13 +3131,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins5996 = insertelement <4 x float> undef, float %add5995, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins5996, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins5996, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp497 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext5997 = extractelement <4 x float> %tmp497, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp498 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3149,15 +3149,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6002 = insertelement <4 x float> %tmp499, float %add6001, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6002, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6002, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC07EA199A0000000, float 0x407DC33340000000, float 0xC0753199A0000000, float -3.895000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC07EA199A0000000, float 0x407DC33340000000, float 0xC0753199A0000000, float -3.895000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp500 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6004 = fadd <4 x float> undef, %tmp500
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6004, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6004, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp501 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3165,7 +3165,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp502 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6007 = insertelement <4 x float> %tmp502, float undef, i32 0
+  %vecins6007 = insertelement <4 x float> %tmp502, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp503 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3173,9 +3173,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp504 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6024 = insertelement <4 x float> %tmp504, float undef, i32 1
+  %vecins6024 = insertelement <4 x float> %tmp504, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6024, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6024, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp505 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3187,7 +3187,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6027 = insertelement <4 x float> %tmp506, float %add6026, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6027, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6027, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6028 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3197,15 +3197,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6030 = insertelement <4 x float> %tmp507, float %add6029, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6030, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6030, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC0527999A0000000, float 0xC06AD999A0000000, float 0x3FF6666660000000, float 0xC03F666660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC0527999A0000000, float 0xC06AD999A0000000, float 0x3FF6666660000000, float 0xC03F666660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp508 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp509 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp510 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3213,7 +3213,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp511 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6036 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3221,17 +3221,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6038 = insertelement <4 x float> undef, float %add6037, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6038, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6038, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp512 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6040 = fadd float undef, 0x4071ECCCC0000000
+  %add6040 = fadd float %val, 0x4071ECCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp513 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6041 = insertelement <4 x float> %tmp513, float %add6040, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6041, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6041, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp514 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3243,9 +3243,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6044 = insertelement <4 x float> %tmp515, float %add6043, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6044, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6044, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC065FCCCC0000000, float 0x40767CCCC0000000, float 0x4079D4CCC0000000, float 0xC07314CCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC065FCCCC0000000, float 0x40767CCCC0000000, float 0x4079D4CCC0000000, float 0xC07314CCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp516 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3253,15 +3253,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6046 = fadd <4 x float> %tmp517, %tmp516
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6046, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6046, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6047 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp518 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6049 = insertelement <4 x float> %tmp518, float undef, i32 0
+  %vecins6049 = insertelement <4 x float> %tmp518, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6049, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6049, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp519 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3269,19 +3269,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6051 = fadd float %vecext6050, 0x407E4E6660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6055 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins6055 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6056 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp520 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6061 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp521 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp522 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3295,9 +3295,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6072 = insertelement <4 x float> undef, float %add6071, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6072, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6072, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40546CCCC0000000, float 0x4067D66660000000, float 0xC060E33340000000, float 0x4061533340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40546CCCC0000000, float 0x4067D66660000000, float 0xC060E33340000000, float 0x4061533340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp523 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3305,7 +3305,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6074 = fadd <4 x float> %tmp524, %tmp523
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6074, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6074, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp525 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3317,23 +3317,23 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6077 = insertelement <4 x float> %tmp526, float %add6076, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6077, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6077, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp527 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6079 = fadd float undef, 0xC07E9B3340000000
+  %add6079 = fadd float %val, 0xC07E9B3340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp528 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp529 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6082 = fadd float undef, 0x407DCE6660000000
+  %add6082 = fadd float %val, 0x407DCE6660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6083 = insertelement <4 x float> undef, float %add6082, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6083, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6083, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp530 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3343,9 +3343,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6086 = insertelement <4 x float> undef, float %add6085, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6086, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6086, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4055C66660000000, float 0x40735199A0000000, float 0xC0713199A0000000, float 0x40729B3340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4055C66660000000, float 0x40735199A0000000, float 0xC0713199A0000000, float 0x40729B3340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp531 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3353,19 +3353,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6088 = fadd <4 x float> %tmp532, %tmp531
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6088, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6088, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp533 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6089 = extractelement <4 x float> %tmp533, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6107 = fadd float undef, 0xC06A166660000000
+  %add6107 = fadd float %val, 0xC06A166660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp534 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6108 = insertelement <4 x float> %tmp534, float %add6107, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6108, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6108, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp535 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3375,7 +3375,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp536 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp537 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3395,7 +3395,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6119 = insertelement <4 x float> %tmp540, float %add6118, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6119, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6119, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp541 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3407,7 +3407,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6122 = insertelement <4 x float> %tmp542, float %add6121, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6122, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6122, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6123 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3415,17 +3415,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp543 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6126 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp544 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6128 = insertelement <4 x float> %tmp544, float undef, i32 3
+  %vecins6128 = insertelement <4 x float> %tmp544, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6128, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6128, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float -2.980000e+02, float 0xC06F0CCCC0000000, float 0xC054A66660000000, float 0xC040CCCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float -2.980000e+02, float 0xC06F0CCCC0000000, float 0xC054A66660000000, float 0xC040CCCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp545 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3441,7 +3441,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6133 = insertelement <4 x float> undef, float %add6132, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6133, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6133, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6134 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3463,9 +3463,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp551 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6178 = insertelement <4 x float> %tmp551, float undef, i32 1
+  %vecins6178 = insertelement <4 x float> %tmp551, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6178, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6178, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp552 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3487,13 +3487,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6184 = insertelement <4 x float> %tmp555, float %add6183, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6184, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6184, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp556 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6189 = insertelement <4 x float> undef, float undef, i32 0
+  %vecins6189 = insertelement <4 x float> undef, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6189, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6189, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp557 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3505,7 +3505,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6192 = insertelement <4 x float> %tmp558, float %add6191, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6192, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6192, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp559 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3519,7 +3519,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6198 = insertelement <4 x float> %tmp561, float %add6197, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x407904CCC0000000, float 0x406A833340000000, float 4.895000e+02, float 0x40648999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x407904CCC0000000, float 0x406A833340000000, float 4.895000e+02, float 0x40648999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp562 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3527,7 +3527,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6200 = fadd <4 x float> %tmp563, %tmp562
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6200, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6200, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp564 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3535,7 +3535,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp565 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6203 = insertelement <4 x float> %tmp565, float undef, i32 0
+  %vecins6203 = insertelement <4 x float> %tmp565, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp566 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3549,9 +3549,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp568 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6209 = insertelement <4 x float> %tmp568, float undef, i32 2
+  %vecins6209 = insertelement <4 x float> %tmp568, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6209, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6209, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp569 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3559,7 +3559,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp570 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6219 = fadd float undef, 0xC0596CCCC0000000
+  %add6219 = fadd float %val, 0xC0596CCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp571 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3573,7 +3573,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6228 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6228, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6228, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6229 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3583,7 +3583,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6231 = insertelement <4 x float> %tmp573, float %add6230, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6231, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6231, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp574 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3595,7 +3595,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6234 = insertelement <4 x float> %tmp575, float %add6233, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6234, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6234, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6235 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3603,13 +3603,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6237 = insertelement <4 x float> undef, float %add6236, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6237, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6237, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp576 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6245 = insertelement <4 x float> undef, float undef, i32 0
+  %vecins6245 = insertelement <4 x float> undef, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6245, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6245, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp577 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3619,17 +3619,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp578 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6251 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins6251 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp579 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6253 = fadd float undef, 0xC0692999A0000000
+  %add6253 = fadd float %val, 0xC0692999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6254 = insertelement <4 x float> undef, float %add6253, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6254, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6254, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 4.600000e+02, float 0xC0777B3340000000, float 0x40351999A0000000, float 0xC06E433340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 4.600000e+02, float 0xC0777B3340000000, float 0x40351999A0000000, float 0xC06E433340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp580 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3637,7 +3637,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6256 = fadd <4 x float> %tmp581, %tmp580
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6256, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6256, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp582 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3649,7 +3649,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6259 = insertelement <4 x float> %tmp583, float %add6258, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6259, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6259, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp584 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3661,7 +3661,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6262 = insertelement <4 x float> %tmp585, float %add6261, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6262, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6262, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp586 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3669,9 +3669,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp587 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6265 = insertelement <4 x float> %tmp587, float undef, i32 2
+  %vecins6265 = insertelement <4 x float> %tmp587, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6265, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6265, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp588 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3683,9 +3683,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6268 = insertelement <4 x float> %tmp589, float %add6267, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6268, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6268, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float -3.130000e+02, float 0xC079733340000000, float -4.660000e+02, float 0xC064E66660000000>, <4 x float>* undef
+  store volatile <4 x float> <float -3.130000e+02, float 0xC079733340000000, float -4.660000e+02, float 0xC064E66660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp590 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3693,7 +3693,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6270 = fadd <4 x float> %tmp591, %tmp590
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6270, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6270, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp592 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3705,7 +3705,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6273 = insertelement <4 x float> %tmp593, float %add6272, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6273, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6273, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp594 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3717,7 +3717,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6276 = insertelement <4 x float> %tmp595, float %add6275, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6276, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6276, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp596 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3729,7 +3729,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6279 = insertelement <4 x float> %tmp597, float %add6278, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6279, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6279, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp598 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3739,21 +3739,21 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6282 = insertelement <4 x float> undef, float %add6281, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6282, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6282, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4067ECCCC0000000, float 0xC040CCCCC0000000, float 0xC0762E6660000000, float -4.750000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4067ECCCC0000000, float 0xC040CCCCC0000000, float 0xC0762E6660000000, float -4.750000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6284 = fadd <4 x float> undef, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6285 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6289 = fadd float undef, 0xC0738999A0000000
+  %add6289 = fadd float %val, 0xC0738999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp599 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6293 = insertelement <4 x float> %tmp599, float undef, i32 2
+  %vecins6293 = insertelement <4 x float> %tmp599, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6293, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6293, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp600 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3763,15 +3763,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6296 = insertelement <4 x float> undef, float %add6295, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6296, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6296, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40704199A0000000, float 0x40753CCCC0000000, float 0xC07E2199A0000000, float 0xC068833340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40704199A0000000, float 0x40753CCCC0000000, float 0xC07E2199A0000000, float 0xC068833340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp601 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6298 = fadd <4 x float> undef, %tmp601
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6298, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6298, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp602 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3783,7 +3783,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6301 = insertelement <4 x float> %tmp603, float %add6300, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6301, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6301, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp604 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3795,7 +3795,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6304 = insertelement <4 x float> %tmp605, float %add6303, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6304, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6304, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp606 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3805,7 +3805,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6307 = insertelement <4 x float> undef, float %add6306, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6307, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6307, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp607 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3817,9 +3817,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6310 = insertelement <4 x float> %tmp608, float %add6309, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6310, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6310, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x407A233340000000, float 0x406DA33340000000, float 3.725000e+02, float 0x40761199A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x407A233340000000, float 0x406DA33340000000, float 3.725000e+02, float 0x40761199A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp609 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3827,7 +3827,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6312 = fadd <4 x float> %tmp610, %tmp609
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6312, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6312, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp611 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3849,13 +3849,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6657 = insertelement <4 x float> %tmp614, float %add6656, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6657, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6657, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6660 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins6660 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6660, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6660, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC064E33340000000, float 0xC064833340000000, float 0xC0673CCCC0000000, float 0xC074266660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC064E33340000000, float 0xC064833340000000, float 0xC0673CCCC0000000, float 0xC074266660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp615 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3867,7 +3867,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6665 = insertelement <4 x float> %tmp616, float %add6664, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp617 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3875,15 +3875,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp618 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC07CC4CCC0000000, float 0x404EE66660000000, float 0xC0754CCCC0000000, float 0xC0744B3340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC07CC4CCC0000000, float 0x404EE66660000000, float 0xC0754CCCC0000000, float 0xC0744B3340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp619 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6676 = fadd <4 x float> %tmp619, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6676, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6676, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp620 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3901,7 +3901,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp622 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp623 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3913,7 +3913,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6685 = insertelement <4 x float> %tmp624, float %add6684, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6685, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6685, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp625 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3925,15 +3925,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6688 = insertelement <4 x float> %tmp626, float %add6687, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6688, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6688, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 7.500000e+00, float 0x4077E33340000000, float 0xC0596CCCC0000000, float 0xC07D4E6660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 7.500000e+00, float 0x4077E33340000000, float 0xC0596CCCC0000000, float 0xC07D4E6660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp627 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6690 = fadd <4 x float> undef, %tmp627
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6690, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6690, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp628 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3945,7 +3945,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6693 = insertelement <4 x float> %tmp629, float %add6692, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6693, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6693, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp630 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3957,7 +3957,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6696 = insertelement <4 x float> %tmp631, float %add6695, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6696, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6696, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp632 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3969,7 +3969,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6699 = insertelement <4 x float> %tmp633, float %add6698, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6699, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6699, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp634 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3981,17 +3981,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6702 = insertelement <4 x float> %tmp635, float %add6701, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6702, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6702, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40772CCCC0000000, float 0xC0625CCCC0000000, float 6.200000e+01, float 0xC06ADCCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40772CCCC0000000, float 0xC0625CCCC0000000, float 6.200000e+01, float 0xC06ADCCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp636 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp637 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6707 = insertelement <4 x float> undef, float undef, i32 0
+  %vecins6707 = insertelement <4 x float> undef, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6707, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6707, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp638 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -3999,7 +3999,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp639 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp640 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4031,21 +4031,21 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp645 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6726 = fadd float undef, 0x4059B999A0000000
+  %add6726 = fadd float %val, 0x4059B999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp646 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6727 = insertelement <4 x float> %tmp646, float %add6726, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6727, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6727, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6728 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6729 = fadd float %vecext6728, 0xC073466660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC0309999A0000000, float -2.715000e+02, float 1.620000e+02, float 0x40674CCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC0309999A0000000, float -2.715000e+02, float 1.620000e+02, float 0x40674CCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp647 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4053,7 +4053,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6732 = fadd <4 x float> %tmp648, %tmp647
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6732, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6732, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp649 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4065,7 +4065,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6735 = insertelement <4 x float> %tmp650, float %add6734, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6735, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6735, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp651 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4077,7 +4077,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6738 = insertelement <4 x float> %tmp652, float %add6737, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6738, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6738, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp653 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4089,7 +4089,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6741 = insertelement <4 x float> %tmp654, float %add6740, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6741, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6741, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp655 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4101,7 +4101,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6744 = insertelement <4 x float> %tmp656, float %add6743, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6744, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6744, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp657 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4109,21 +4109,21 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6746 = fadd <4 x float> %tmp658, %tmp657
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6746, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6746, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp659 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6749 = insertelement <4 x float> undef, float undef, i32 0
+  %vecins6749 = insertelement <4 x float> undef, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6749, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6749, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp660 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6751 = fadd float undef, 0x4075DE6660000000
+  %add6751 = fadd float %val, 0x4075DE6660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6752 = insertelement <4 x float> undef, float %add6751, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6752, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6752, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp661 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4133,7 +4133,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6755 = insertelement <4 x float> undef, float %add6754, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6755, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6755, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp662 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4145,15 +4145,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6758 = insertelement <4 x float> %tmp663, float %add6757, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6758, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6758, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x403D1999A0000000, float 0xC05F533340000000, float 3.945000e+02, float 3.950000e+01>, <4 x float>* undef
+  store volatile <4 x float> <float 0x403D1999A0000000, float 0xC05F533340000000, float 3.945000e+02, float 3.950000e+01>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp664 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6760 = fadd <4 x float> undef, %tmp664
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6760, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6760, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp665 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4165,9 +4165,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp666 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC079BE6660000000, float 4.930000e+02, float 0x406CC33340000000, float 0xC062E999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC079BE6660000000, float 4.930000e+02, float 0x406CC33340000000, float 0xC062E999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp667 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4183,7 +4183,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6777 = insertelement <4 x float> %tmp669, float %add6776, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6777, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6777, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp670 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4195,9 +4195,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6784 = extractelement <4 x float> %tmp671, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6875 = insertelement <4 x float> undef, float undef, i32 0
+  %vecins6875 = insertelement <4 x float> undef, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6875, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6875, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp672 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4207,15 +4207,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6878 = insertelement <4 x float> undef, float %add6877, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6878, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6878, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6888 = fadd float undef, 0x4057CCCCC0000000
+  %add6888 = fadd float %val, 0x4057CCCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp673 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6889 = insertelement <4 x float> %tmp673, float %add6888, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6889, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6889, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp674 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4227,7 +4227,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6892 = insertelement <4 x float> %tmp675, float %add6891, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6892, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6892, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp676 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4239,7 +4239,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6895 = insertelement <4 x float> %tmp677, float %add6894, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6895, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6895, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp678 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4249,7 +4249,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6900 = fadd <4 x float> %tmp680, %tmp679
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6900, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6900, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp681 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4261,9 +4261,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6903 = insertelement <4 x float> %tmp682, float %add6902, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6903, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6903, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6905 = fadd float undef, 0x4031B33340000000
+  %add6905 = fadd float %val, 0x4031B33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp683 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4271,9 +4271,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp684 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6912 = insertelement <4 x float> %tmp684, float undef, i32 3
+  %vecins6912 = insertelement <4 x float> %tmp684, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 3.315000e+02, float 0xC066C999A0000000, float 0xC061F33340000000, float 0x4071166660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 3.315000e+02, float 0xC066C999A0000000, float 0xC061F33340000000, float 0x4071166660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp685 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4281,13 +4281,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6914 = fadd <4 x float> %tmp686, %tmp685
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6914, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6914, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6915 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6920 = insertelement <4 x float> undef, float undef, i32 1
+  %vecins6920 = insertelement <4 x float> undef, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6920, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6920, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext6921 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4295,11 +4295,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp687 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6926 = insertelement <4 x float> %tmp687, float undef, i32 3
+  %vecins6926 = insertelement <4 x float> %tmp687, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6926, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6926, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC03C4CCCC0000000, float 0xC07E5199A0000000, float -8.250000e+01, float 0xC043B33340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC03C4CCCC0000000, float 0xC07E5199A0000000, float -8.250000e+01, float 0xC043B33340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp688 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4307,13 +4307,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6928 = fadd <4 x float> %tmp689, %tmp688
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6928, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6928, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6930 = fadd float undef, -4.590000e+02
+  %add6930 = fadd float %val, -4.590000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6931 = insertelement <4 x float> undef, float %add6930, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6931, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6931, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp690 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4323,7 +4323,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp691 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp692 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4349,15 +4349,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp695 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6950 = fadd float undef, 0xC078F33340000000
+  %add6950 = fadd float %val, 0xC078F33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp696 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6951 = insertelement <4 x float> %tmp696, float %add6950, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6951, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6951, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp697 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4369,7 +4369,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6954 = insertelement <4 x float> %tmp698, float %add6953, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6954, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6954, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp699 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4377,7 +4377,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6956 = fadd <4 x float> %tmp700, %tmp699
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6956, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6956, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp701 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4389,7 +4389,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6959 = insertelement <4 x float> %tmp702, float %add6958, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6959, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6959, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp703 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4401,15 +4401,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6965 = insertelement <4 x float> %tmp704, float %add6964, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6965, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6965, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add6975 = fadd float undef, 0x406AF33340000000
+  %add6975 = fadd float %val, 0x406AF33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp705 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6976 = insertelement <4 x float> %tmp705, float %add6975, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6976, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6976, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp706 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4417,7 +4417,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6984 = fadd <4 x float> %tmp707, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6984, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6984, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp708 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4429,7 +4429,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins6987 = insertelement <4 x float> %tmp709, float %add6986, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6987, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6987, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp710 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4439,11 +4439,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp711 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins6996 = insertelement <4 x float> %tmp711, float undef, i32 3
+  %vecins6996 = insertelement <4 x float> %tmp711, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins6996, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins6996, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4077A4CCC0000000, float 0xC0757199A0000000, float 0xC072F4CCC0000000, float 0xC071DCCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4077A4CCC0000000, float 0xC0757199A0000000, float 0xC072F4CCC0000000, float 0xC071DCCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp712 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4451,7 +4451,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add6998 = fadd <4 x float> %tmp713, %tmp712
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add6998, <4 x float>* undef, align 16
+  store volatile <4 x float> %add6998, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp714 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4463,7 +4463,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7001 = insertelement <4 x float> %tmp715, float %add7000, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7001, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7001, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp716 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4475,11 +4475,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7004 = insertelement <4 x float> %tmp717, float %add7003, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp718 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add7140 = fadd float undef, 0x403D333340000000
+  %add7140 = fadd float %val, 0x403D333340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7141 = insertelement <4 x float> undef, float %add7140, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4489,7 +4489,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7144 = insertelement <4 x float> undef, float %add7143, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp719 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4501,15 +4501,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7150 = insertelement <4 x float> %tmp720, float %add7149, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7150, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7150, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 1.700000e+02, float 0xC077B4CCC0000000, float 0x40625999A0000000, float 0x406C166660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 1.700000e+02, float 0xC077B4CCC0000000, float 0x40625999A0000000, float 0x406C166660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp721 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add7152 = fadd <4 x float> %tmp721, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add7152, <4 x float>* undef, align 16
+  store volatile <4 x float> %add7152, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext7156 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4519,7 +4519,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7158 = insertelement <4 x float> %tmp722, float %add7157, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7158, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7158, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp723 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4531,13 +4531,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7161 = insertelement <4 x float> %tmp724, float %add7160, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7161, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7161, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add7168 = fadd float undef, 0xC072F199A0000000
+  %add7168 = fadd float %val, 0xC072F199A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp725 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext7170 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4545,11 +4545,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7172 = insertelement <4 x float> undef, float %add7171, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7172, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7172, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext7173 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp726 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4559,7 +4559,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7421 = insertelement <4 x float> undef, float %add7420, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7421, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7421, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp727 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4571,7 +4571,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7424 = insertelement <4 x float> %tmp728, float %add7423, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7424, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7424, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp729 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4583,11 +4583,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7427 = insertelement <4 x float> %tmp730, float %add7426, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7427, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7427, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext7428 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp731 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4599,9 +4599,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7570 = insertelement <4 x float> %tmp732, float %add7569, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7570, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7570, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40745199A0000000, float 0xC0411999A0000000, float -5.650000e+01, float -4.005000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40745199A0000000, float 0xC0411999A0000000, float -5.650000e+01, float -4.005000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp733 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4609,7 +4609,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add7572 = fadd <4 x float> %tmp734, %tmp733
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add7572, <4 x float>* undef, align 16
+  store volatile <4 x float> %add7572, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext7573 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4619,11 +4619,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7575 = insertelement <4 x float> %tmp735, float %add7574, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7575, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7575, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp736 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add7577 = fadd float undef, 0xC051666660000000
+  %add7577 = fadd float %val, 0xC051666660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp737 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4635,7 +4635,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7581 = insertelement <4 x float> undef, float %add7580, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7581, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7581, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp739 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4647,7 +4647,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7584 = insertelement <4 x float> %tmp740, float %add7583, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC057533340000000, float 0x4060A33340000000, float 0x40791E6660000000, float 2.455000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC057533340000000, float 0x4060A33340000000, float 0x40791E6660000000, float 2.455000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp741 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4655,7 +4655,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add7586 = fadd <4 x float> %tmp742, %tmp741
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add7586, <4 x float>* undef, align 16
+  store volatile <4 x float> %add7586, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp743 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4665,7 +4665,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp744 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp745 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4677,15 +4677,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7592 = insertelement <4 x float> %tmp746, float %add7591, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7592, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7592, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp747 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext7593 = extractelement <4 x float> %tmp747, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins7595 = insertelement <4 x float> undef, float undef, i32 2
+  %vecins7595 = insertelement <4 x float> undef, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7595, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7595, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp748 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4693,17 +4693,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add7597 = fadd float %vecext7596, 0x407E666660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x406A766660000000, float 0xBFC99999A0000000, float 0xC0751B3340000000, float -4.075000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0x406A766660000000, float 0xBFC99999A0000000, float 0xC0751B3340000000, float -4.075000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp749 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add7616 = fadd float undef, 0xC04DE66660000000
+  %add7616 = fadd float %val, 0xC04DE66660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp750 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7617 = insertelement <4 x float> %tmp750, float %add7616, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7617, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7617, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp751 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4715,17 +4715,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7620 = insertelement <4 x float> %tmp752, float %add7619, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7620, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7620, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp753 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add7622 = fadd float undef, 0xC054B999A0000000
+  %add7622 = fadd float %val, 0xC054B999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp754 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins7626 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins7626 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7626, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7626, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp755 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4733,7 +4733,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add7628 = fadd <4 x float> %tmp756, %tmp755
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add7628, <4 x float>* undef, align 16
+  store volatile <4 x float> %add7628, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp757 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4745,13 +4745,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7631 = insertelement <4 x float> %tmp758, float %add7630, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add7639 = fadd float undef, 0x407C5999A0000000
+  %add7639 = fadd float %val, 0x407C5999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp759 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7640 = insertelement <4 x float> %tmp759, float %add7639, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x406AA66660000000, float 0x4067C66660000000, float 0xC054866660000000, float -2.400000e+01>, <4 x float>* undef
+  store volatile <4 x float> <float 0x406AA66660000000, float 0x4067C66660000000, float 0xC054866660000000, float -2.400000e+01>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp760 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4759,9 +4759,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp761 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add7644 = fadd float undef, 0xC0758999A0000000
+  %add7644 = fadd float %val, 0xC0758999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp762 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4773,7 +4773,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7648 = insertelement <4 x float> %tmp763, float %add7647, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7648, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7648, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp764 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4785,7 +4785,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7651 = insertelement <4 x float> %tmp765, float %add7650, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7651, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7651, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp766 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4797,7 +4797,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7654 = insertelement <4 x float> %tmp767, float %add7653, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7654, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7654, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp768 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4805,7 +4805,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add7656 = fadd <4 x float> %tmp769, %tmp768
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add7656, <4 x float>* undef, align 16
+  store volatile <4 x float> %add7656, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp770 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4817,7 +4817,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7659 = insertelement <4 x float> %tmp771, float %add7658, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7659, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7659, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp772 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4829,7 +4829,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7662 = insertelement <4 x float> %tmp773, float %add7661, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7662, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7662, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp774 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4841,7 +4841,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7665 = insertelement <4 x float> %tmp775, float %add7664, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7665, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7665, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp776 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4851,7 +4851,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7668 = insertelement <4 x float> undef, float %add7667, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7668, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7668, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp777 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4873,23 +4873,23 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp781 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp782 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add7731 = fadd float undef, 1.900000e+02
+  %add7731 = fadd float %val, 1.900000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp783 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins7732 = insertelement <4 x float> %tmp783, float %add7731, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7732, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7732, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp784 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins7735 = insertelement <4 x float> %tmp784, float undef, i32 2
+  %vecins7735 = insertelement <4 x float> %tmp784, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7735, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7735, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp785 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4897,11 +4897,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add7737 = fadd float %vecext7736, 0xC06AF66660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins7850 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins7850 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins7850, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins7850, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4062A33340000000, float 2.290000e+02, float 0x40509999A0000000, float 0xC078BE6660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4062A33340000000, float 2.290000e+02, float 0x40509999A0000000, float 0xC078BE6660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp786 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4909,7 +4909,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add7852 = fadd <4 x float> %tmp787, %tmp786
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add7852, <4 x float>* undef, align 16
+  store volatile <4 x float> %add7852, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp788 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4921,13 +4921,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9398 = insertelement <4 x float> %tmp789, float %add9397, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9398, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9398, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext9399 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp790 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9401 = insertelement <4 x float> %tmp790, float undef, i32 2
+  %vecins9401 = insertelement <4 x float> %tmp790, float %val, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp791 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4939,11 +4939,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9404 = insertelement <4 x float> %tmp792, float %add9403, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9404, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9404, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp793 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp794 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4959,7 +4959,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp796 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp797 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4971,7 +4971,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9415 = insertelement <4 x float> %tmp798, float %add9414, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9415, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9415, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp799 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4983,9 +4983,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9418 = insertelement <4 x float> %tmp800, float %add9417, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9418, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9418, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 3.555000e+02, float 0xC062E33340000000, float 0x4065C66660000000, float -3.645000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 3.555000e+02, float 0xC062E33340000000, float 0x4065C66660000000, float -3.645000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp801 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -4993,7 +4993,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9420 = fadd <4 x float> %tmp802, %tmp801
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add9420, <4 x float>* undef, align 16
+  store volatile <4 x float> %add9420, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp803 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5001,9 +5001,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp804 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9423 = insertelement <4 x float> %tmp804, float undef, i32 0
+  %vecins9423 = insertelement <4 x float> %tmp804, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9423, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9423, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp805 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5015,17 +5015,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9426 = insertelement <4 x float> %tmp806, float %add9425, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9426, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9426, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp807 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9428 = fadd float undef, 0xC065466660000000
+  %add9428 = fadd float %val, 0xC065466660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp808 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9429 = insertelement <4 x float> %tmp808, float %add9428, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9429, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9429, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp809 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5037,7 +5037,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9432 = insertelement <4 x float> %tmp810, float %add9431, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC07C7E6660000000, float 1.205000e+02, float 0x4050D999A0000000, float 0xC06B233340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC07C7E6660000000, float 1.205000e+02, float 0x4050D999A0000000, float 0xC06B233340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp811 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5045,7 +5045,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9434 = fadd <4 x float> %tmp812, %tmp811
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9436 = fadd float undef, -3.185000e+02
+  %add9436 = fadd float %val, -3.185000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp813 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5053,7 +5053,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp814 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp815 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5065,7 +5065,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9443 = insertelement <4 x float> %tmp816, float %add9442, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9443, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9443, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp817 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5077,7 +5077,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9446 = insertelement <4 x float> %tmp818, float %add9445, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9446, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9446, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp819 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5085,23 +5085,23 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9448 = fadd <4 x float> %tmp820, %tmp819
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add9448, <4 x float>* undef, align 16
+  store volatile <4 x float> %add9448, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9450 = fadd float undef, 0xC0718199A0000000
+  %add9450 = fadd float %val, 0xC0718199A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp821 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9451 = insertelement <4 x float> %tmp821, float %add9450, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9451, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9451, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp822 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp823 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9454 = insertelement <4 x float> %tmp823, float undef, i32 1
+  %vecins9454 = insertelement <4 x float> %tmp823, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9454, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9454, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp824 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5113,23 +5113,23 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9457 = insertelement <4 x float> %tmp825, float %add9456, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9457, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9457, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext9458 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp826 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9460 = insertelement <4 x float> %tmp826, float undef, i32 3
+  %vecins9460 = insertelement <4 x float> %tmp826, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9460, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9460, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x407B5E6660000000, float 0x40648999A0000000, float 0xC06B966660000000, float 0x40341999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x407B5E6660000000, float 0x40648999A0000000, float 0xC06B966660000000, float 0x40341999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp827 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9462 = fadd <4 x float> %tmp827, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add9462, <4 x float>* undef, align 16
+  store volatile <4 x float> %add9462, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp828 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5137,23 +5137,23 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp829 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9465 = insertelement <4 x float> %tmp829, float undef, i32 0
+  %vecins9465 = insertelement <4 x float> %tmp829, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9467 = fadd float undef, 0x405D666660000000
+  %add9467 = fadd float %val, 0x405D666660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp830 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9468 = insertelement <4 x float> %tmp830, float %add9467, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9468, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9468, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp831 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9470 = fadd float undef, 0x4077033340000000
+  %add9470 = fadd float %val, 0x4077033340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp832 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext9472 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5163,9 +5163,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9474 = insertelement <4 x float> %tmp833, float %add9473, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9474, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9474, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x404F733340000000, float 0x407AB4CCC0000000, float 0x40605999A0000000, float 0xC03E4CCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x404F733340000000, float 0x407AB4CCC0000000, float 0x40605999A0000000, float 0xC03E4CCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp834 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5173,7 +5173,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9476 = fadd <4 x float> %tmp835, %tmp834
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add9476, <4 x float>* undef, align 16
+  store volatile <4 x float> %add9476, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp836 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5185,17 +5185,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9479 = insertelement <4 x float> %tmp837, float %add9478, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9479, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9479, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp838 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9481 = fadd float undef, 0x407BE33340000000
+  %add9481 = fadd float %val, 0x407BE33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp839 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9482 = insertelement <4 x float> %tmp839, float %add9481, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9482, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9482, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext9483 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5205,7 +5205,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9485 = insertelement <4 x float> %tmp840, float %add9484, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9485, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9485, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp841 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5215,13 +5215,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp842 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC076B999A0000000, float 0xC0706CCCC0000000, float 0x407904CCC0000000, float 0x407EE199A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC076B999A0000000, float 0xC0706CCCC0000000, float 0x407904CCC0000000, float 0x407EE199A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp843 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp844 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5229,15 +5229,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9492 = fadd float %vecext9491, 0x407C166660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9495 = fadd float undef, 0x407DBB3340000000
+  %add9495 = fadd float %val, 0x407DBB3340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp845 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9496 = insertelement <4 x float> %tmp845, float %add9495, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9496, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9496, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp846 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5249,41 +5249,41 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9499 = insertelement <4 x float> %tmp847, float %add9498, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9499, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9499, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp848 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9501 = fadd float undef, 0x407D5CCCC0000000
+  %add9501 = fadd float %val, 0x407D5CCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp849 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9502 = insertelement <4 x float> %tmp849, float %add9501, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9502, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9502, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp850 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9504 = fadd <4 x float> %tmp850, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add9504, <4 x float>* undef, align 16
+  store volatile <4 x float> %add9504, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp851 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9506 = fadd float undef, 0x4076EE6660000000
+  %add9506 = fadd float %val, 0x4076EE6660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp852 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9507 = insertelement <4 x float> %tmp852, float %add9506, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9507, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9507, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp853 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9509 = fadd float undef, 0xC0535999A0000000
+  %add9509 = fadd float %val, 0xC0535999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp854 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp855 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5295,7 +5295,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9513 = insertelement <4 x float> %tmp856, float %add9512, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9513, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9513, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp857 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5303,11 +5303,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp858 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9516 = insertelement <4 x float> %tmp858, float undef, i32 3
+  %vecins9516 = insertelement <4 x float> %tmp858, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9516, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9516, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x407254CCC0000000, float 0x407844CCC0000000, float 0xC04D9999A0000000, float 0xC0550CCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x407254CCC0000000, float 0x407844CCC0000000, float 0xC04D9999A0000000, float 0xC0550CCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp859 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5319,9 +5319,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp862 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9521 = insertelement <4 x float> %tmp862, float undef, i32 0
+  %vecins9521 = insertelement <4 x float> %tmp862, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9521, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9521, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp863 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5333,25 +5333,25 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9524 = insertelement <4 x float> %tmp864, float %add9523, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9524, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9524, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp865 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9526 = fadd float undef, 0x4072833340000000
+  %add9526 = fadd float %val, 0x4072833340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp866 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9527 = insertelement <4 x float> %tmp866, float %add9526, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9527, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9527, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp867 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9530 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins9530 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9530, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9530, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4072F4CCC0000000, float 0x4065CCCCC0000000, float 0x4051D33340000000, float 0x40680CCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4072F4CCC0000000, float 0x4065CCCCC0000000, float 0x4051D33340000000, float 0x40680CCCC0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp868 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5363,9 +5363,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp870 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9535 = insertelement <4 x float> %tmp870, float undef, i32 0
+  %vecins9535 = insertelement <4 x float> %tmp870, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9535, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9535, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp871 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5377,7 +5377,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9538 = insertelement <4 x float> %tmp872, float %add9537, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9538, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9538, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp873 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5385,17 +5385,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9543 = fadd float %vecext9542, 0x4050D999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9576 = fadd float undef, 0x40219999A0000000
+  %add9576 = fadd float %val, 0x40219999A0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9577 = insertelement <4 x float> undef, float %add9576, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9577, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9577, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp874 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9580 = insertelement <4 x float> undef, float undef, i32 1
+  %vecins9580 = insertelement <4 x float> undef, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9580, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9580, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp875 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5407,11 +5407,11 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9583 = insertelement <4 x float> %tmp876, float %add9582, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9583, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9583, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp877 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext9673 = extractelement <4 x float> undef, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5421,7 +5421,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9675 = insertelement <4 x float> %tmp878, float %add9674, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9675, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9675, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext9676 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5441,7 +5441,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9681 = insertelement <4 x float> %tmp881, float %add9680, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9681, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9681, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp882 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5451,7 +5451,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9686 = fadd <4 x float> %tmp883, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add9686, <4 x float>* undef, align 16
+  store volatile <4 x float> %add9686, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp884 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5481,19 +5481,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9695 = insertelement <4 x float> %tmp888, float %add9694, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9695, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9695, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp889 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9697 = fadd float undef, 0x4058D33340000000
+  %add9697 = fadd float %val, 0x4058D33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp890 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9698 = insertelement <4 x float> %tmp890, float %add9697, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9698, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9698, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4062CCCCC0000000, float 0x407AD999A0000000, float 0x40582CCCC0000000, float 0xC0712B3340000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4062CCCCC0000000, float 0x407AD999A0000000, float 0x40582CCCC0000000, float 0xC0712B3340000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp891 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5509,7 +5509,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9703 = insertelement <4 x float> %tmp893, float %add9702, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9703, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9703, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp894 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5521,7 +5521,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9706 = insertelement <4 x float> %tmp895, float %add9705, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9706, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9706, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext9707 = extractelement <4 x float> undef, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5531,23 +5531,23 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9709 = insertelement <4 x float> %tmp896, float %add9708, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9709, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9709, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp897 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext9710 = extractelement <4 x float> %tmp897, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9712 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins9712 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9712, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9712, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4069F33340000000, float 0xC048266660000000, float 0x40638CCCC0000000, float 0xC07EC199A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4069F33340000000, float 0xC048266660000000, float 0x40638CCCC0000000, float 0xC07EC199A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp898 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9714 = fadd <4 x float> undef, %tmp898
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add9714, <4 x float>* undef, align 16
+  store volatile <4 x float> %add9714, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp899 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5555,9 +5555,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp900 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9717 = insertelement <4 x float> %tmp900, float undef, i32 0
+  %vecins9717 = insertelement <4 x float> %tmp900, float %val, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9717, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9717, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp901 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5569,7 +5569,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9720 = insertelement <4 x float> %tmp902, float %add9719, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9720, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9720, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp903 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5581,7 +5581,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9723 = insertelement <4 x float> %tmp904, float %add9722, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9723, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9723, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp905 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5593,15 +5593,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9726 = insertelement <4 x float> %tmp906, float %add9725, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9726, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9726, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float -4.575000e+02, float 0x40713E6660000000, float 0x407D133340000000, float -1.425000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float -4.575000e+02, float 0x40713E6660000000, float 0x407D133340000000, float -1.425000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp907 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9728 = fadd <4 x float> %tmp907, undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add9728, <4 x float>* undef, align 16
+  store volatile <4 x float> %add9728, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp908 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5613,17 +5613,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9731 = insertelement <4 x float> %tmp909, float %add9730, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9731, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9731, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp910 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9733 = fadd float undef, 0xC050F33340000000
+  %add9733 = fadd float %val, 0xC050F33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp911 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9734 = insertelement <4 x float> %tmp911, float %add9733, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9734, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9734, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp912 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5635,23 +5635,23 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9737 = insertelement <4 x float> %tmp913, float %add9736, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9737, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9737, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp914 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext9738 = extractelement <4 x float> %tmp914, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9740 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins9740 = insertelement <4 x float> undef, float %val, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9740, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9740, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 2.150000e+02, float 0x405A2CCCC0000000, float 2.310000e+02, float 0x404E1999A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 2.150000e+02, float 0x405A2CCCC0000000, float 2.310000e+02, float 0x404E1999A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp915 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp916 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp917 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5661,7 +5661,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9745 = insertelement <4 x float> undef, float %add9744, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9745, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9745, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp918 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5673,7 +5673,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9748 = insertelement <4 x float> %tmp919, float %add9747, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9748, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9748, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp920 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5685,7 +5685,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9751 = insertelement <4 x float> %tmp921, float %add9750, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9751, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9751, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp922 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5697,9 +5697,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9754 = insertelement <4 x float> %tmp923, float %add9753, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9754, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9754, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 2.590000e+02, float 0x407B7199A0000000, float 0xC07ED199A0000000, float 0xC064FCCCC0000000>, <4 x float>* %.compoundliteral9755
+  store volatile <4 x float> <float 2.590000e+02, float 0x407B7199A0000000, float 0xC07ED199A0000000, float 0xC064FCCCC0000000>, <4 x float>* %.compoundliteral9755
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp924 = load <4 x float>, <4 x float>* %.compoundliteral9755
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5717,7 +5717,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9759 = insertelement <4 x float> %tmp927, float %add9758, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9759, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9759, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp928 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5729,17 +5729,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9762 = insertelement <4 x float> %tmp929, float %add9761, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9762, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9762, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp930 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add9764 = fadd float undef, 0xC060E66660000000
+  %add9764 = fadd float %val, 0xC060E66660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp931 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9765 = insertelement <4 x float> %tmp931, float %add9764, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9765, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9765, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp932 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5751,9 +5751,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9768 = insertelement <4 x float> %tmp933, float %add9767, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9768, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9768, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4032CCCCC0000000, float -9.600000e+01, float -5.000000e+02, float 0x4078EE6660000000>, <4 x float>* %.compoundliteral9769
+  store volatile <4 x float> <float 0x4032CCCCC0000000, float -9.600000e+01, float -5.000000e+02, float 0x4078EE6660000000>, <4 x float>* %.compoundliteral9769
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp934 = load <4 x float>, <4 x float>* %.compoundliteral9769
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5761,7 +5761,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add9770 = fadd <4 x float> %tmp935, %tmp934
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add9770, <4 x float>* undef, align 16
+  store volatile <4 x float> %add9770, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp936 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5773,7 +5773,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9773 = insertelement <4 x float> %tmp937, float %add9772, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9773, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9773, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp938 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5785,25 +5785,25 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins9776 = insertelement <4 x float> %tmp939, float %add9775, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins9776, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins9776, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext9816 = extractelement <4 x float> undef, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp940 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %vecins9818 = insertelement <4 x float> %tmp940, float undef, i32 1
+  %vecins9818 = insertelement <4 x float> %tmp940, float %val, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp941 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add10388 = fadd float undef, 4.755000e+02
+  %add10388 = fadd float %val, 4.755000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp942 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10389 = insertelement <4 x float> %tmp942, float %add10388, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10389, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10389, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp943 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5815,19 +5815,19 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10392 = insertelement <4 x float> %tmp944, float %add10391, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10392, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10392, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp945 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp946 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add10405 = fadd float undef, -5.650000e+01
+  %add10405 = fadd float %val, -5.650000e+01
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp947 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10406 = insertelement <4 x float> %tmp947, float %add10405, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10406, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10406, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp948 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5839,7 +5839,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10409 = insertelement <4 x float> %tmp949, float %add10408, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10409, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10409, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp950 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5849,9 +5849,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp951 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float -2.340000e+02, float -4.720000e+02, float 4.350000e+02, float 0xC059A66660000000>, <4 x float>* %.compoundliteral10413
+  store volatile <4 x float> <float -2.340000e+02, float -4.720000e+02, float 4.350000e+02, float 0xC059A66660000000>, <4 x float>* %.compoundliteral10413
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp952 = load <4 x float>, <4 x float>* %.compoundliteral10413
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5859,7 +5859,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add10414 = fadd <4 x float> %tmp953, %tmp952
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add10414, <4 x float>* undef, align 16
+  store volatile <4 x float> %add10414, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp954 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5871,7 +5871,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10417 = insertelement <4 x float> %tmp955, float %add10416, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10417, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10417, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp956 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5883,15 +5883,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10420 = insertelement <4 x float> %tmp957, float %add10419, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10420, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10420, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add10422 = fadd float undef, 0xC0662CCCC0000000
+  %add10422 = fadd float %val, 0xC0662CCCC0000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext10424 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x402B333340000000, float 0x40735E6660000000, float 0xC0567999A0000000, float 2.050000e+02>, <4 x float>* undef
+  store volatile <4 x float> <float 0x402B333340000000, float 0x40735E6660000000, float 0xC0567999A0000000, float 2.050000e+02>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp958 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5899,7 +5899,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add10428 = fadd <4 x float> %tmp959, %tmp958
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add10428, <4 x float>* undef, align 16
+  store volatile <4 x float> %add10428, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp960 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5909,13 +5909,13 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp961 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add10436 = fadd float undef, 0xC06AF33340000000
+  %add10436 = fadd float %val, 0xC06AF33340000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp962 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10437 = insertelement <4 x float> %tmp962, float %add10436, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10437, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10437, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecext10438 = extractelement <4 x float> undef, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5925,9 +5925,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10440 = insertelement <4 x float> %tmp963, float %add10439, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10440, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10440, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC065E999A0000000, float 0x4067D33340000000, float 0xC070133340000000, float 0x406B666660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0xC065E999A0000000, float 0x4067D33340000000, float 0xC070133340000000, float 0x406B666660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp964 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5941,7 +5941,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10445 = insertelement <4 x float> %tmp966, float %add10444, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10445, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10445, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp967 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5953,7 +5953,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10448 = insertelement <4 x float> %tmp968, float %add10447, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10448, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10448, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp969 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5965,7 +5965,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10451 = insertelement <4 x float> %tmp970, float %add10450, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10451, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10451, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp971 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5975,7 +5975,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10454 = insertelement <4 x float> undef, float %add10453, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x406AFCCCC0000000, float 0xC07604CCC0000000, float 6.900000e+01, float 0xC060A66660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x406AFCCCC0000000, float 0xC07604CCC0000000, float 6.900000e+01, float 0xC060A66660000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp972 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5983,7 +5983,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %add10456 = fadd <4 x float> %tmp973, %tmp972
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %add10456, <4 x float>* undef, align 16
+  store volatile <4 x float> %add10456, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp974 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -5993,7 +5993,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10459 = insertelement <4 x float> undef, float %add10458, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10459, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10459, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp975 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -6015,7 +6015,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10465 = insertelement <4 x float> %tmp978, float %add10464, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10465, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10465, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp979 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -6027,9 +6027,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10468 = insertelement <4 x float> %tmp980, float %add10467, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10468, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10468, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x4078833340000000, float 0x40786CCCC0000000, float 0xC0468CCCC0000000, float 0xC0793199A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x4078833340000000, float 0x40786CCCC0000000, float 0xC0468CCCC0000000, float 0xC0793199A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp981 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -6045,7 +6045,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10473 = insertelement <4 x float> %tmp983, float %add10472, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10473, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10473, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp984 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -6057,15 +6057,15 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10476 = insertelement <4 x float> %tmp985, float %add10475, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10476, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10476, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add10489 = fadd float undef, 0x4074666660000000
+  %add10489 = fadd float %val, 0x4074666660000000
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp986 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10490 = insertelement <4 x float> %tmp986, float %add10489, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10490, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10490, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp987 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -6079,9 +6079,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10510 = insertelement <4 x float> %tmp989, float %add10509, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10510, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10510, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0x40656999A0000000, float 0xC073766660000000, float 1.685000e+02, float 0x40765199A0000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x40656999A0000000, float 0xC073766660000000, float 1.685000e+02, float 0x40765199A0000000>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp990 = load <4 x float>, <4 x float>* undef
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -6097,17 +6097,17 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10515 = insertelement <4 x float> %tmp992, float %add10514, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10515, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10515, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp993 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  %add10562 = fadd float undef, 2.035000e+02
+  %add10562 = fadd float %val, 2.035000e+02
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp994 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10563 = insertelement <4 x float> %tmp994, float %add10562, i32 2
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10563, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10563, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp995 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -6119,9 +6119,9 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10566 = insertelement <4 x float> %tmp996, float %add10565, i32 3
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10566, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10566, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> <float 0xC068B999A0000000, float 0xC050E66660000000, float 0xC0725999A0000000, float 0xC054D33340000000>, <4 x float>* %.compoundliteral10567
+  store volatile <4 x float> <float 0xC068B999A0000000, float 0xC050E66660000000, float 0xC0725999A0000000, float 0xC054D33340000000>, <4 x float>* %.compoundliteral10567
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp997 = load <4 x float>, <4 x float>* %.compoundliteral10567
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -6139,7 +6139,7 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10571 = insertelement <4 x float> %tmp1000, float %add10570, i32 0
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10571, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10571, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %tmp1001 = load <4 x float>, <4 x float>* undef, align 16
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
@@ -6151,56 +6151,56 @@ entry:
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
   %vecins10574 = insertelement <4 x float> %tmp1002, float %add10573, i32 1
   tail call void asm sideeffect "", "~{q0}{q1}{q2}{q3}{q4}{q5}{q6}{q7}{q8}{q9}{q10}{q11}{q12}{q13}{q14}{q15}"()
-  store <4 x float> %vecins10574, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10574, <4 x float>* undef, align 16
   %tmp1003 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10575 = extractelement <4 x float> %tmp1003, i32 2
   %tmp1004 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins10577 = insertelement <4 x float> %tmp1004, float undef, i32 2
-  store <4 x float> %vecins10577, <4 x float>* undef, align 16
+  %vecins10577 = insertelement <4 x float> %tmp1004, float %val, i32 2
+  store volatile <4 x float> %vecins10577, <4 x float>* undef, align 16
   %tmp1005 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10578 = extractelement <4 x float> %tmp1005, i32 3
   %add10579 = fadd float %vecext10578, 0x4076566660000000
   %tmp1006 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10580 = insertelement <4 x float> %tmp1006, float %add10579, i32 3
-  store <4 x float> %vecins10580, <4 x float>* undef, align 16
-  store <4 x float> <float 0x407CAB3340000000, float 1.685000e+02, float 0xC07B866660000000, float 0xC061ACCCC0000000>, <4 x float>* %.compoundliteral10581
+  store volatile <4 x float> %vecins10580, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x407CAB3340000000, float 1.685000e+02, float 0xC07B866660000000, float 0xC061ACCCC0000000>, <4 x float>* %.compoundliteral10581
   %tmp1007 = load <4 x float>, <4 x float>* %.compoundliteral10581
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1008 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10583 = extractelement <4 x float> %tmp1008, i32 0
   %add10584 = fadd float %vecext10583, 0xC060533340000000
   %tmp1009 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10585 = insertelement <4 x float> %tmp1009, float %add10584, i32 0
-  store <4 x float> %vecins10585, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10585, <4 x float>* undef, align 16
   %tmp1010 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10586 = extractelement <4 x float> %tmp1010, i32 1
   %add10587 = fadd float %vecext10586, 0xC0694CCCC0000000
   %tmp1011 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10588 = insertelement <4 x float> %tmp1011, float %add10587, i32 1
-  store <4 x float> %vecins10588, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10588, <4 x float>* undef, align 16
   %tmp1012 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10589 = extractelement <4 x float> %tmp1012, i32 2
   %add10590 = fadd float %vecext10589, 0xC0541999A0000000
   %tmp1013 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10591 = insertelement <4 x float> %tmp1013, float %add10590, i32 2
-  store <4 x float> %vecins10591, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10591, <4 x float>* undef, align 16
   %tmp1014 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10592 = extractelement <4 x float> %tmp1014, i32 3
   %add10593 = fadd float %vecext10592, 0xC06C566660000000
   %tmp1015 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10594 = insertelement <4 x float> %tmp1015, float %add10593, i32 3
-  store <4 x float> %vecins10594, <4 x float>* undef, align 16
-  store <4 x float> <float 0x407A3199A0000000, float 0xC0659999A0000000, float 0x407E0999A0000000, float 0xC0334CCCC0000000>, <4 x float>* %.compoundliteral10595
+  store volatile <4 x float> %vecins10594, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x407A3199A0000000, float 0xC0659999A0000000, float 0x407E0999A0000000, float 0xC0334CCCC0000000>, <4 x float>* %.compoundliteral10595
   %tmp1016 = load <4 x float>, <4 x float>* %.compoundliteral10595
   %tmp1017 = load <4 x float>, <4 x float>* undef, align 16
   %add10596 = fadd <4 x float> %tmp1017, %tmp1016
-  store <4 x float> %add10596, <4 x float>* undef, align 16
+  store volatile <4 x float> %add10596, <4 x float>* undef, align 16
   %tmp1018 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10597 = extractelement <4 x float> %tmp1018, i32 0
   %add10598 = fadd float %vecext10597, 0x40640999A0000000
   %tmp1019 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10599 = insertelement <4 x float> %tmp1019, float %add10598, i32 0
-  store <4 x float> %vecins10599, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10599, <4 x float>* undef, align 16
   %tmp1020 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10600 = extractelement <4 x float> %tmp1020, i32 1
   %add10601 = fadd float %vecext10600, 0xC073966660000000
@@ -6211,48 +6211,48 @@ entry:
   %add10604 = fadd float %vecext10603, 1.780000e+02
   %tmp1023 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10605 = insertelement <4 x float> %tmp1023, float %add10604, i32 2
-  store <4 x float> %vecins10605, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10605, <4 x float>* undef, align 16
   %tmp1024 = load <4 x float>, <4 x float>* undef, align 16
-  %add10607 = fadd float undef, 0x4070A33340000000
+  %add10607 = fadd float %val, 0x4070A33340000000
   %tmp1025 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> <float 0x407C5999A0000000, float 0x4046733340000000, float 0xC06E6CCCC0000000, float 0xC063C33340000000>, <4 x float>* %.compoundliteral10609
+  store volatile <4 x float> <float 0x407C5999A0000000, float 0x4046733340000000, float 0xC06E6CCCC0000000, float 0xC063C33340000000>, <4 x float>* %.compoundliteral10609
   %tmp1026 = load <4 x float>, <4 x float>* %.compoundliteral10609
   %tmp1027 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1028 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10611 = extractelement <4 x float> %tmp1028, i32 0
   %add10612 = fadd float %vecext10611, 0x40757199A0000000
   %vecins10613 = insertelement <4 x float> undef, float %add10612, i32 0
-  store <4 x float> %vecins10613, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10613, <4 x float>* undef, align 16
   %tmp1029 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10614 = extractelement <4 x float> %tmp1029, i32 1
   %add10615 = fadd float %vecext10614, 0x40740CCCC0000000
   %tmp1030 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10616 = insertelement <4 x float> %tmp1030, float %add10615, i32 1
-  store <4 x float> %vecins10616, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10616, <4 x float>* undef, align 16
   %tmp1031 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10617 = extractelement <4 x float> %tmp1031, i32 2
   %add10618 = fadd float %vecext10617, 0xC012CCCCC0000000
   %tmp1032 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10619 = insertelement <4 x float> %tmp1032, float %add10618, i32 2
-  store <4 x float> %vecins10619, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10619, <4 x float>* undef, align 16
   %tmp1033 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10620 = extractelement <4 x float> %tmp1033, i32 3
   %add10621 = fadd float %vecext10620, 0x406E566660000000
   %tmp1034 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> <float 0x407B2199A0000000, float 0xC07D9CCCC0000000, float -4.350000e+01, float 0xC07D3B3340000000>, <4 x float>* %.compoundliteral10623
+  store volatile <4 x float> <float 0x407B2199A0000000, float 0xC07D9CCCC0000000, float -4.350000e+01, float 0xC07D3B3340000000>, <4 x float>* %.compoundliteral10623
   %tmp1035 = load <4 x float>, <4 x float>* %.compoundliteral10623
   %add10624 = fadd <4 x float> undef, %tmp1035
   %tmp1036 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10625 = extractelement <4 x float> %tmp1036, i32 0
   %tmp1037 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins10627 = insertelement <4 x float> %tmp1037, float undef, i32 0
-  store <4 x float> %vecins10627, <4 x float>* undef, align 16
+  %vecins10627 = insertelement <4 x float> %tmp1037, float %val, i32 0
+  store volatile <4 x float> %vecins10627, <4 x float>* undef, align 16
   %tmp1038 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10628 = extractelement <4 x float> %tmp1038, i32 1
   %add10629 = fadd float %vecext10628, 0x407E3CCCC0000000
   %tmp1039 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10630 = insertelement <4 x float> %tmp1039, float %add10629, i32 1
-  store <4 x float> %vecins10630, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10630, <4 x float>* undef, align 16
   %tmp1040 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10631 = extractelement <4 x float> %tmp1040, i32 2
   %tmp1041 = load <4 x float>, <4 x float>* undef, align 16
@@ -6261,8 +6261,8 @@ entry:
   %add10635 = fadd float %vecext10634, 0xC067533340000000
   %tmp1043 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10636 = insertelement <4 x float> %tmp1043, float %add10635, i32 3
-  store <4 x float> %vecins10636, <4 x float>* undef, align 16
-  store <4 x float> <float 1.950000e+02, float 0x407E8E6660000000, float 0x407D7CCCC0000000, float 0x407E166660000000>, <4 x float>* %.compoundliteral10637
+  store volatile <4 x float> %vecins10636, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 1.950000e+02, float 0x407E8E6660000000, float 0x407D7CCCC0000000, float 0x407E166660000000>, <4 x float>* %.compoundliteral10637
   %tmp1044 = load <4 x float>, <4 x float>* undef, align 16
   %add10638 = fadd <4 x float> %tmp1044, undef
   %tmp1045 = load <4 x float>, <4 x float>* undef, align 16
@@ -6270,94 +6270,94 @@ entry:
   %add10640 = fadd float %vecext10639, 0x406CA33340000000
   %tmp1046 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10641 = insertelement <4 x float> %tmp1046, float %add10640, i32 0
-  store <4 x float> %vecins10641, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10641, <4 x float>* undef, align 16
   %tmp1047 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10642 = extractelement <4 x float> %tmp1047, i32 1
   %add10643 = fadd float %vecext10642, 0xC07C8999A0000000
   %tmp1048 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10644 = insertelement <4 x float> %tmp1048, float %add10643, i32 1
-  store <4 x float> %vecins10644, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10644, <4 x float>* undef, align 16
   %tmp1049 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10645 = extractelement <4 x float> %tmp1049, i32 2
   %tmp1050 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1051 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins10748 = insertelement <4 x float> undef, float undef, i32 3
+  %vecins10748 = insertelement <4 x float> undef, float %val, i32 3
   %tmp1052 = load <4 x float>, <4 x float>* %.compoundliteral10749
   %add10750 = fadd <4 x float> undef, %tmp1052
-  store <4 x float> %add10750, <4 x float>* undef, align 16
+  store volatile <4 x float> %add10750, <4 x float>* undef, align 16
   %tmp1053 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10751 = extractelement <4 x float> %tmp1053, i32 0
   %add10752 = fadd float %vecext10751, 0x4071B33340000000
   %tmp1054 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10753 = insertelement <4 x float> %tmp1054, float %add10752, i32 0
-  store <4 x float> %vecins10753, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10753, <4 x float>* undef, align 16
   %tmp1055 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10754 = extractelement <4 x float> %tmp1055, i32 1
   %add10755 = fadd float %vecext10754, 0xC076A66660000000
   %tmp1056 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10756 = insertelement <4 x float> %tmp1056, float %add10755, i32 1
-  store <4 x float> %vecins10756, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10756, <4 x float>* undef, align 16
   %tmp1057 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10757 = extractelement <4 x float> %tmp1057, i32 2
   %add10758 = fadd float %vecext10757, 3.800000e+01
   %tmp1058 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10759 = insertelement <4 x float> %tmp1058, float %add10758, i32 2
-  store <4 x float> %vecins10759, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10759, <4 x float>* undef, align 16
   %tmp1059 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10760 = extractelement <4 x float> %tmp1059, i32 3
-  store <4 x float> undef, <4 x float>* undef, align 16
-  store <4 x float> <float 0xC075BB3340000000, float 0x4074D4CCC0000000, float 0xC07A466660000000, float 0xC0691CCCC0000000>, <4 x float>* %.compoundliteral10763
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0xC075BB3340000000, float 0x4074D4CCC0000000, float 0xC07A466660000000, float 0xC0691CCCC0000000>, <4 x float>* %.compoundliteral10763
   %tmp1060 = load <4 x float>, <4 x float>* %.compoundliteral10763
   %tmp1061 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1062 = load <4 x float>, <4 x float>* undef, align 16
-  %add10985 = fadd float undef, 0x405E933340000000
+  %add10985 = fadd float %val, 0x405E933340000000
   %tmp1063 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10986 = insertelement <4 x float> %tmp1063, float %add10985, i32 3
-  store <4 x float> %vecins10986, <4 x float>* undef, align 16
-  store <4 x float> <float 0xC0721E6660000000, float -4.180000e+02, float 0x406F366660000000, float 0xC055F999A0000000>, <4 x float>* %.compoundliteral10987
+  store volatile <4 x float> %vecins10986, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0xC0721E6660000000, float -4.180000e+02, float 0x406F366660000000, float 0xC055F999A0000000>, <4 x float>* %.compoundliteral10987
   %tmp1064 = load <4 x float>, <4 x float>* %.compoundliteral10987
   %tmp1065 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins10994 = insertelement <4 x float> %tmp1065, float undef, i32 1
+  %vecins10994 = insertelement <4 x float> %tmp1065, float %val, i32 1
   %tmp1066 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10995 = extractelement <4 x float> %tmp1066, i32 2
   %add10996 = fadd float %vecext10995, 0x406F9999A0000000
   %tmp1067 = load <4 x float>, <4 x float>* undef, align 16
   %vecins10997 = insertelement <4 x float> %tmp1067, float %add10996, i32 2
-  store <4 x float> %vecins10997, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins10997, <4 x float>* undef, align 16
   %tmp1068 = load <4 x float>, <4 x float>* undef, align 16
   %vecext10998 = extractelement <4 x float> %tmp1068, i32 3
   %add10999 = fadd float %vecext10998, -2.765000e+02
   %tmp1069 = load <4 x float>, <4 x float>* undef, align 16
   %vecins11000 = insertelement <4 x float> %tmp1069, float %add10999, i32 3
-  store <4 x float> %vecins11000, <4 x float>* undef, align 16
-  store <4 x float> <float 0x4078F999A0000000, float 0xC06D166660000000, float 0x40501999A0000000, float 0x406FC999A0000000>, <4 x float>* %.compoundliteral11001
+  store volatile <4 x float> %vecins11000, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x4078F999A0000000, float 0xC06D166660000000, float 0x40501999A0000000, float 0x406FC999A0000000>, <4 x float>* %.compoundliteral11001
   %tmp1070 = load <4 x float>, <4 x float>* undef, align 16
   %add11002 = fadd <4 x float> %tmp1070, undef
   %vecext11003 = extractelement <4 x float> undef, i32 0
   %vecext11009 = extractelement <4 x float> undef, i32 2
   %tmp1071 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins11033 = insertelement <4 x float> %tmp1071, float undef, i32 0
-  store <4 x float> %vecins11033, <4 x float>* undef, align 16
+  %vecins11033 = insertelement <4 x float> %tmp1071, float %val, i32 0
+  store volatile <4 x float> %vecins11033, <4 x float>* undef, align 16
   %tmp1072 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11034 = extractelement <4 x float> %tmp1072, i32 1
   %add11035 = fadd float %vecext11034, 0x4056D33340000000
   %tmp1073 = load <4 x float>, <4 x float>* undef, align 16
   %vecins11036 = insertelement <4 x float> %tmp1073, float %add11035, i32 1
-  store <4 x float> %vecins11036, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins11036, <4 x float>* undef, align 16
   %tmp1074 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11037 = extractelement <4 x float> %tmp1074, i32 2
   %add11038 = fadd float %vecext11037, 0xC06EA33340000000
   %tmp1075 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1076 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11040 = extractelement <4 x float> %tmp1076, i32 3
   %add11041 = fadd float %vecext11040, 0x40746CCCC0000000
   %tmp1077 = load <4 x float>, <4 x float>* undef, align 16
   %vecins11042 = insertelement <4 x float> %tmp1077, float %add11041, i32 3
-  store <4 x float> <float 0x405DD999A0000000, float -3.775000e+02, float -1.265000e+02, float 0xC065C66660000000>, <4 x float>* undef
+  store volatile <4 x float> <float 0x405DD999A0000000, float -3.775000e+02, float -1.265000e+02, float 0xC065C66660000000>, <4 x float>* undef
   %tmp1078 = load <4 x float>, <4 x float>* undef, align 16
   %add11044 = fadd <4 x float> %tmp1078, undef
-  store <4 x float> %add11044, <4 x float>* undef, align 16
+  store volatile <4 x float> %add11044, <4 x float>* undef, align 16
   %tmp1079 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11045 = extractelement <4 x float> %tmp1079, i32 0
   %add11046 = fadd float %vecext11045, 0xC076E66660000000
@@ -6366,58 +6366,58 @@ entry:
   %tmp1081 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11048 = extractelement <4 x float> %tmp1081, i32 1
   %add11049 = fadd float %vecext11048, 4.100000e+02
-  %vecins11064 = insertelement <4 x float> undef, float undef, i32 1
-  %add11074 = fadd float undef, 0xC06FF999A0000000
+  %vecins11064 = insertelement <4 x float> undef, float %val, i32 1
+  %add11074 = fadd float %val, 0xC06FF999A0000000
   %tmp1082 = load <4 x float>, <4 x float>* undef, align 16
   %vecins11075 = insertelement <4 x float> %tmp1082, float %add11074, i32 0
-  store <4 x float> %vecins11075, <4 x float>* undef, align 16
-  %add11077 = fadd float undef, 0xC075D33340000000
+  store volatile <4 x float> %vecins11075, <4 x float>* undef, align 16
+  %add11077 = fadd float %val, 0xC075D33340000000
   %tmp1083 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1084 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1085 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11093 = extractelement <4 x float> %tmp1085, i32 2
   %add11094 = fadd float %vecext11093, 0xC07CD66660000000
   %tmp1086 = load <4 x float>, <4 x float>* undef, align 16
   %vecins11095 = insertelement <4 x float> %tmp1086, float %add11094, i32 2
-  store <4 x float> %vecins11095, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
-  store <4 x float> <float 0x4061F66660000000, float 0xC076DB3340000000, float 0xC055A66660000000, float 2.415000e+02>, <4 x float>* undef
+  store volatile <4 x float> %vecins11095, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x4061F66660000000, float 0xC076DB3340000000, float 0xC055A66660000000, float 2.415000e+02>, <4 x float>* undef
   %tmp1087 = load <4 x float>, <4 x float>* undef
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1088 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11513 = extractelement <4 x float> %tmp1088, i32 2
   %add11514 = fadd float %vecext11513, 0xC07C7199A0000000
   %vecins11515 = insertelement <4 x float> undef, float %add11514, i32 2
-  store <4 x float> %vecins11515, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins11515, <4 x float>* undef, align 16
   %add11520 = fadd <4 x float> undef, undef
-  store <4 x float> %add11520, <4 x float>* undef, align 16
+  store volatile <4 x float> %add11520, <4 x float>* undef, align 16
   %vecext11521 = extractelement <4 x float> undef, i32 0
   %add11522 = fadd float %vecext11521, 0x4041733340000000
   %tmp1089 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1090 = load <4 x float>, <4 x float>* undef
   %tmp1091 = load <4 x float>, <4 x float>* undef, align 16
   %add11562 = fadd <4 x float> %tmp1091, %tmp1090
   %tmp1092 = load <4 x float>, <4 x float>* undef, align 16
-  %add11564 = fadd float undef, 0xC0411999A0000000
+  %add11564 = fadd float %val, 0xC0411999A0000000
   %tmp1093 = load <4 x float>, <4 x float>* undef, align 16
   %vecins11565 = insertelement <4 x float> %tmp1093, float %add11564, i32 0
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %vecext11586 = extractelement <4 x float> undef, i32 3
   %add11587 = fadd float %vecext11586, 3.760000e+02
   %tmp1094 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
-  store <4 x float> <float 0xC06ED999A0000000, float 1.380000e+02, float 0xC073AB3340000000, float 0x4078A66660000000>, <4 x float>* undef
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0xC06ED999A0000000, float 1.380000e+02, float 0xC073AB3340000000, float 0x4078A66660000000>, <4 x float>* undef
   %tmp1095 = load <4 x float>, <4 x float>* undef
   %tmp1096 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1097 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1098 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins11593 = insertelement <4 x float> %tmp1098, float undef, i32 0
+  %vecins11593 = insertelement <4 x float> %tmp1098, float %val, i32 0
   %vecext11594 = extractelement <4 x float> undef, i32 1
   %tmp1099 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins11596 = insertelement <4 x float> %tmp1099, float undef, i32 1
-  store <4 x float> %vecins11596, <4 x float>* undef, align 16
+  %vecins11596 = insertelement <4 x float> %tmp1099, float %val, i32 1
+  store volatile <4 x float> %vecins11596, <4 x float>* undef, align 16
   %tmp1100 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11597 = extractelement <4 x float> %tmp1100, i32 2
   %add11598 = fadd float %vecext11597, 0x40430CCCC0000000
@@ -6426,34 +6426,34 @@ entry:
   %tmp1102 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11600 = extractelement <4 x float> %tmp1102, i32 3
   %tmp1103 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins11602 = insertelement <4 x float> %tmp1103, float undef, i32 3
-  store <4 x float> %vecins11602, <4 x float>* undef, align 16
+  %vecins11602 = insertelement <4 x float> %tmp1103, float %val, i32 3
+  store volatile <4 x float> %vecins11602, <4 x float>* undef, align 16
   %tmp1104 = load <4 x float>, <4 x float>* undef
   %tmp1105 = load <4 x float>, <4 x float>* undef, align 16
   %add11604 = fadd <4 x float> %tmp1105, %tmp1104
   %tmp1106 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11605 = extractelement <4 x float> %tmp1106, i32 0
   %tmp1107 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins11607 = insertelement <4 x float> %tmp1107, float undef, i32 0
-  %vecins11621 = insertelement <4 x float> undef, float undef, i32 0
-  %vecins11630 = insertelement <4 x float> undef, float undef, i32 3
-  store <4 x float> %vecins11630, <4 x float>* undef, align 16
-  store <4 x float> <float -1.190000e+02, float 0x402F666660000000, float 0xC07BD33340000000, float -1.595000e+02>, <4 x float>* %.compoundliteral11631
+  %vecins11607 = insertelement <4 x float> %tmp1107, float %val, i32 0
+  %vecins11621 = insertelement <4 x float> undef, float %val, i32 0
+  %vecins11630 = insertelement <4 x float> undef, float %val, i32 3
+  store volatile <4 x float> %vecins11630, <4 x float>* undef, align 16
+  store volatile <4 x float> <float -1.190000e+02, float 0x402F666660000000, float 0xC07BD33340000000, float -1.595000e+02>, <4 x float>* %.compoundliteral11631
   %tmp1108 = load <4 x float>, <4 x float>* %.compoundliteral11631
   %tmp1109 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
-  %add11634 = fadd float undef, -1.075000e+02
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  %add11634 = fadd float %val, -1.075000e+02
   %vecext11647 = extractelement <4 x float> undef, i32 0
   %add11648 = fadd float %vecext11647, 0x40775999A0000000
   %tmp1110 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11650 = extractelement <4 x float> undef, i32 1
   %tmp1111 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins11784 = insertelement <4 x float> %tmp1111, float undef, i32 3
-  store <4 x float> %vecins11784, <4 x float>* undef, align 16
-  store <4 x float> <float 1.605000e+02, float 0x4068366660000000, float 2.820000e+02, float 0x407CF66660000000>, <4 x float>* %.compoundliteral11785
+  %vecins11784 = insertelement <4 x float> %tmp1111, float %val, i32 3
+  store volatile <4 x float> %vecins11784, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 1.605000e+02, float 0x4068366660000000, float 2.820000e+02, float 0x407CF66660000000>, <4 x float>* %.compoundliteral11785
   %tmp1112 = load <4 x float>, <4 x float>* %.compoundliteral11785
   %add11786 = fadd <4 x float> undef, %tmp1112
-  store <4 x float> %add11786, <4 x float>* undef, align 16
+  store volatile <4 x float> %add11786, <4 x float>* undef, align 16
   %tmp1113 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11787 = extractelement <4 x float> %tmp1113, i32 0
   %vecext11807 = extractelement <4 x float> undef, i32 2
@@ -6463,60 +6463,60 @@ entry:
   %add11811 = fadd float %vecext11810, 0x4068F66660000000
   %tmp1115 = load <4 x float>, <4 x float>* undef, align 16
   %vecins11812 = insertelement <4 x float> %tmp1115, float %add11811, i32 3
-  store <4 x float> %vecins11812, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins11812, <4 x float>* undef, align 16
   %tmp1116 = load <4 x float>, <4 x float>* undef
   %tmp1117 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11958 = extractelement <4 x float> undef, i32 1
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %vecext11961 = extractelement <4 x float> undef, i32 2
   %add11962 = fadd float %vecext11961, -3.680000e+02
   %tmp1118 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
-  %add11965 = fadd float undef, 0x4061133340000000
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  %add11965 = fadd float %val, 0x4061133340000000
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1119 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11975 = extractelement <4 x float> %tmp1119, i32 2
   %tmp1120 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins11977 = insertelement <4 x float> %tmp1120, float undef, i32 2
-  store <4 x float> %vecins11977, <4 x float>* undef, align 16
+  %vecins11977 = insertelement <4 x float> %tmp1120, float %val, i32 2
+  store volatile <4 x float> %vecins11977, <4 x float>* undef, align 16
   %vecext11978 = extractelement <4 x float> undef, i32 3
   %add11979 = fadd float %vecext11978, 0xC0688999A0000000
   %tmp1121 = load <4 x float>, <4 x float>* undef, align 16
   %vecins11980 = insertelement <4 x float> %tmp1121, float %add11979, i32 3
-  store <4 x float> %vecins11980, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins11980, <4 x float>* undef, align 16
   %add11982 = fadd <4 x float> undef, undef
-  store <4 x float> %add11982, <4 x float>* undef, align 16
+  store volatile <4 x float> %add11982, <4 x float>* undef, align 16
   %tmp1122 = load <4 x float>, <4 x float>* undef, align 16
   %vecext11983 = extractelement <4 x float> %tmp1122, i32 0
   %add11984 = fadd float %vecext11983, 0xC075966660000000
   %tmp1123 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins12005 = insertelement <4 x float> undef, float undef, i32 2
-  store <4 x float> %vecins12005, <4 x float>* undef, align 16
+  %vecins12005 = insertelement <4 x float> undef, float %val, i32 2
+  store volatile <4 x float> %vecins12005, <4 x float>* undef, align 16
   %tmp1124 = load <4 x float>, <4 x float>* undef, align 16
-  %add12007 = fadd float undef, 0xC07124CCC0000000
+  %add12007 = fadd float %val, 0xC07124CCC0000000
   %vecins12008 = insertelement <4 x float> undef, float %add12007, i32 3
-  store <4 x float> %vecins12008, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins12008, <4 x float>* undef, align 16
   %tmp1125 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1126 = load <4 x float>, <4 x float>* undef, align 16
-  %add12012 = fadd float undef, 0xC0750CCCC0000000
+  %add12012 = fadd float %val, 0xC0750CCCC0000000
   %tmp1127 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12013 = insertelement <4 x float> %tmp1127, float %add12012, i32 0
-  store <4 x float> %vecins12013, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins12013, <4 x float>* undef, align 16
   %tmp1128 = load <4 x float>, <4 x float>* undef, align 16
-  %add12015 = fadd float undef, 0x4079CE6660000000
+  %add12015 = fadd float %val, 0x4079CE6660000000
   %tmp1129 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12016 = insertelement <4 x float> %tmp1129, float %add12015, i32 1
-  store <4 x float> %vecins12016, <4 x float>* undef, align 16
-  %add12018 = fadd float undef, 3.555000e+02
+  store volatile <4 x float> %vecins12016, <4 x float>* undef, align 16
+  %add12018 = fadd float %val, 3.555000e+02
   %tmp1130 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12019 = insertelement <4 x float> %tmp1130, float %add12018, i32 2
   %tmp1131 = load <4 x float>, <4 x float>* undef, align 16
   %vecext12020 = extractelement <4 x float> %tmp1131, i32 3
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %vecext12028 = extractelement <4 x float> undef, i32 1
-  store <4 x float> undef, <4 x float>* undef, align 16
-  store <4 x float> <float 0x40791999A0000000, float 0x407C7CCCC0000000, float 0x4070F33340000000, float 0xC056ECCCC0000000>, <4 x float>* undef
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x40791999A0000000, float 0x407C7CCCC0000000, float 0x4070F33340000000, float 0xC056ECCCC0000000>, <4 x float>* undef
   %tmp1132 = load <4 x float>, <4 x float>* undef, align 16
   %add12038 = fadd <4 x float> %tmp1132, undef
   %tmp1133 = load <4 x float>, <4 x float>* undef, align 16
@@ -6524,27 +6524,27 @@ entry:
   %add12043 = fadd float %vecext12042, 0x402F9999A0000000
   %tmp1134 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12044 = insertelement <4 x float> %tmp1134, float %add12043, i32 1
-  store <4 x float> %vecins12044, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins12044, <4 x float>* undef, align 16
   %vecext12045 = extractelement <4 x float> undef, i32 2
   %add12046 = fadd float %vecext12045, 0xC07EF33340000000
   %tmp1135 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12047 = insertelement <4 x float> %tmp1135, float %add12046, i32 2
-  store <4 x float> %vecins12047, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins12047, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1136 = load <4 x float>, <4 x float>* undef, align 16
   %vecext12112 = extractelement <4 x float> %tmp1136, i32 1
   %tmp1137 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
-  %add12116 = fadd float undef, 0xC074F4CCC0000000
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  %add12116 = fadd float %val, 0xC074F4CCC0000000
   %tmp1138 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12117 = insertelement <4 x float> %tmp1138, float %add12116, i32 2
-  store <4 x float> %vecins12117, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins12117, <4 x float>* undef, align 16
   %tmp1139 = load <4 x float>, <4 x float>* undef, align 16
   %vecext12118 = extractelement <4 x float> %tmp1139, i32 3
   %add12119 = fadd float %vecext12118, 0xC0638CCCC0000000
   %tmp1140 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12120 = insertelement <4 x float> %tmp1140, float %add12119, i32 3
-  %add12152 = fadd float undef, 0x4039333340000000
+  %add12152 = fadd float %val, 0x4039333340000000
   %tmp1141 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12153 = insertelement <4 x float> %tmp1141, float %add12152, i32 0
   %vecext12154 = extractelement <4 x float> undef, i32 1
@@ -6561,67 +6561,67 @@ entry:
   %add12161 = fadd float %vecext12160, 0x407B1999A0000000
   %tmp1146 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12162 = insertelement <4 x float> %tmp1146, float %add12161, i32 3
-  store <4 x float> %vecins12162, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins12162, <4 x float>* undef, align 16
   %tmp1147 = load <4 x float>, <4 x float>* undef
   %tmp1148 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1149 = load <4 x float>, <4 x float>* undef, align 16
   %vecext12182 = extractelement <4 x float> %tmp1149, i32 1
   %tmp1150 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
-  store <4 x float> <float 0x4061833340000000, float 0x405CA66660000000, float -1.275000e+02, float 0x405BC66660000000>, <4 x float>* undef
-  %add12208 = fadd float undef, 0x407854CCC0000000
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x4061833340000000, float 0x405CA66660000000, float -1.275000e+02, float 0x405BC66660000000>, <4 x float>* undef
+  %add12208 = fadd float %val, 0x407854CCC0000000
   %tmp1151 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1152 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1153 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins12218 = insertelement <4 x float> undef, float undef, i32 3
-  store <4 x float> %vecins12218, <4 x float>* undef, align 16
-  store <4 x float> <float 0x407C3CCCC0000000, float 0xC057C66660000000, float 2.605000e+02, float 0xC07974CCC0000000>, <4 x float>* undef
+  %vecins12218 = insertelement <4 x float> undef, float %val, i32 3
+  store volatile <4 x float> %vecins12218, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x407C3CCCC0000000, float 0xC057C66660000000, float 2.605000e+02, float 0xC07974CCC0000000>, <4 x float>* undef
   %tmp1154 = load <4 x float>, <4 x float>* undef
   %tmp1155 = load <4 x float>, <4 x float>* undef, align 16
   %add12220 = fadd <4 x float> %tmp1155, %tmp1154
   %tmp1156 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1157 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins12223 = insertelement <4 x float> %tmp1157, float undef, i32 0
-  store <4 x float> %vecins12223, <4 x float>* undef, align 16
+  %vecins12223 = insertelement <4 x float> %tmp1157, float %val, i32 0
+  store volatile <4 x float> %vecins12223, <4 x float>* undef, align 16
   %tmp1158 = load <4 x float>, <4 x float>* undef, align 16
-  %add12242 = fadd float undef, 0x4067E33340000000
+  %add12242 = fadd float %val, 0x4067E33340000000
   %tmp1159 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12243 = insertelement <4 x float> %tmp1159, float %add12242, i32 2
-  store <4 x float> %vecins12243, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins12243, <4 x float>* undef, align 16
   %tmp1160 = load <4 x float>, <4 x float>* undef, align 16
   %vecext12244 = extractelement <4 x float> %tmp1160, i32 3
   %add12245 = fadd float %vecext12244, 0x4071AE6660000000
   %tmp1161 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12246 = insertelement <4 x float> %tmp1161, float %add12245, i32 3
-  store <4 x float> %vecins12246, <4 x float>* undef, align 16
-  store <4 x float> <float -4.880000e+02, float 0xC079966660000000, float -8.450000e+01, float 0xC0464CCCC0000000>, <4 x float>* %.compoundliteral12247
+  store volatile <4 x float> %vecins12246, <4 x float>* undef, align 16
+  store volatile <4 x float> <float -4.880000e+02, float 0xC079966660000000, float -8.450000e+01, float 0xC0464CCCC0000000>, <4 x float>* %.compoundliteral12247
   %tmp1162 = load <4 x float>, <4 x float>* %.compoundliteral12247
   %tmp1163 = load <4 x float>, <4 x float>* undef, align 16
   %add12248 = fadd <4 x float> %tmp1163, %tmp1162
-  store <4 x float> %add12248, <4 x float>* undef, align 16
+  store volatile <4 x float> %add12248, <4 x float>* undef, align 16
   %tmp1164 = load <4 x float>, <4 x float>* undef, align 16
   %vecext12249 = extractelement <4 x float> %tmp1164, i32 0
   %add12250 = fadd float %vecext12249, 1.075000e+02
   %tmp1165 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1166 = load <4 x float>, <4 x float>* undef, align 16
   %vecext12252 = extractelement <4 x float> %tmp1166, i32 1
   %add12253 = fadd float %vecext12252, 0xC0662CCCC0000000
   %tmp1167 = load <4 x float>, <4 x float>* undef, align 16
   %vecins12254 = insertelement <4 x float> %tmp1167, float %add12253, i32 1
-  store <4 x float> %vecins12254, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins12254, <4 x float>* undef, align 16
   %tmp1168 = load <4 x float>, <4 x float>* undef, align 16
   %vecext12255 = extractelement <4 x float> %tmp1168, i32 2
   %add12256 = fadd float %vecext12255, 0x40554CCCC0000000
-  store <4 x float> undef, <4 x float>* undef, align 16
-  %add13141 = fadd float undef, 0x40768999A0000000
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  %add13141 = fadd float %val, 0x40768999A0000000
   %tmp1169 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13142 = insertelement <4 x float> %tmp1169, float %add13141, i32 3
-  store <4 x float> %vecins13142, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins13142, <4 x float>* undef, align 16
   %tmp1170 = load <4 x float>, <4 x float>* undef
   %add13144 = fadd <4 x float> undef, %tmp1170
-  store <4 x float> %add13144, <4 x float>* undef, align 16
+  store volatile <4 x float> %add13144, <4 x float>* undef, align 16
   %tmp1171 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13145 = extractelement <4 x float> %tmp1171, i32 0
   %add13146 = fadd float %vecext13145, 3.975000e+02
@@ -6630,137 +6630,137 @@ entry:
   %add13379 = fadd float %vecext13378, 0xC053B33340000000
   %tmp1173 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13380 = insertelement <4 x float> %tmp1173, float %add13379, i32 3
-  store <4 x float> %vecins13380, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins13380, <4 x float>* undef, align 16
   %tmp1174 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins13408 = insertelement <4 x float> %tmp1174, float undef, i32 3
-  store <4 x float> %vecins13408, <4 x float>* undef, align 16
-  store <4 x float> <float 0xC0455999A0000000, float 0xC07D366660000000, float 4.240000e+02, float -1.670000e+02>, <4 x float>* undef
+  %vecins13408 = insertelement <4 x float> %tmp1174, float %val, i32 3
+  store volatile <4 x float> %vecins13408, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0xC0455999A0000000, float 0xC07D366660000000, float 4.240000e+02, float -1.670000e+02>, <4 x float>* undef
   %tmp1175 = load <4 x float>, <4 x float>* undef
   %tmp1176 = load <4 x float>, <4 x float>* undef, align 16
   %add13410 = fadd <4 x float> %tmp1176, %tmp1175
-  store <4 x float> %add13410, <4 x float>* undef, align 16
+  store volatile <4 x float> %add13410, <4 x float>* undef, align 16
   %tmp1177 = load <4 x float>, <4 x float>* undef, align 16
-  %add13412 = fadd float undef, 0xC0708999A0000000
+  %add13412 = fadd float %val, 0xC0708999A0000000
   %tmp1178 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13413 = insertelement <4 x float> %tmp1178, float %add13412, i32 0
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %vecext13428 = extractelement <4 x float> undef, i32 1
   %add13429 = fadd float %vecext13428, 0xC063BCCCC0000000
   %tmp1179 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13430 = insertelement <4 x float> %tmp1179, float %add13429, i32 1
-  store <4 x float> %vecins13430, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins13430, <4 x float>* undef, align 16
   %tmp1180 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13431 = extractelement <4 x float> %tmp1180, i32 2
-  %vecins13433 = insertelement <4 x float> undef, float undef, i32 2
-  store <4 x float> undef, <4 x float>* undef, align 16
-  %add13449 = fadd float undef, 4.590000e+02
+  %vecins13433 = insertelement <4 x float> undef, float %val, i32 2
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  %add13449 = fadd float %val, 4.590000e+02
   %tmp1181 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13450 = insertelement <4 x float> %tmp1181, float %add13449, i32 3
-  store <4 x float> %vecins13450, <4 x float>* undef, align 16
-  store <4 x float> <float 0xC073A66660000000, float 0xC041B33340000000, float 0x4066233340000000, float 0x4071C33340000000>, <4 x float>* undef
+  store volatile <4 x float> %vecins13450, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0xC073A66660000000, float 0xC041B33340000000, float 0x4066233340000000, float 0x4071C33340000000>, <4 x float>* undef
   %tmp1182 = load <4 x float>, <4 x float>* undef
   %tmp1183 = load <4 x float>, <4 x float>* undef, align 16
   %add13452 = fadd <4 x float> %tmp1183, %tmp1182
-  store <4 x float> %add13452, <4 x float>* undef, align 16
+  store volatile <4 x float> %add13452, <4 x float>* undef, align 16
   %tmp1184 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13453 = extractelement <4 x float> %tmp1184, i32 0
   %add13454 = fadd float %vecext13453, 0xC072866660000000
   %tmp1185 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13455 = insertelement <4 x float> %tmp1185, float %add13454, i32 0
-  %add13471 = fadd float undef, 0xC0556CCCC0000000
+  %add13471 = fadd float %val, 0xC0556CCCC0000000
   %tmp1186 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13472 = insertelement <4 x float> %tmp1186, float %add13471, i32 1
-  store <4 x float> %vecins13472, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins13472, <4 x float>* undef, align 16
   %tmp1187 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13473 = extractelement <4 x float> %tmp1187, i32 2
   %add13474 = fadd float %vecext13473, 0xC0786999A0000000
   %tmp1188 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13475 = insertelement <4 x float> %tmp1188, float %add13474, i32 2
-  store <4 x float> %vecins13475, <4 x float>* undef, align 16
-  %add13477 = fadd float undef, 0xC07C3E6660000000
+  store volatile <4 x float> %vecins13475, <4 x float>* undef, align 16
+  %add13477 = fadd float %val, 0xC07C3E6660000000
   %tmp1189 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13478 = insertelement <4 x float> %tmp1189, float %add13477, i32 3
-  store <4 x float> %vecins13478, <4 x float>* undef, align 16
-  store <4 x float> <float -4.740000e+02, float 0x4023CCCCC0000000, float 0xC05C266660000000, float 0x407B7199A0000000>, <4 x float>* undef
+  store volatile <4 x float> %vecins13478, <4 x float>* undef, align 16
+  store volatile <4 x float> <float -4.740000e+02, float 0x4023CCCCC0000000, float 0xC05C266660000000, float 0x407B7199A0000000>, <4 x float>* undef
   %tmp1190 = load <4 x float>, <4 x float>* undef, align 16
   %add13480 = fadd <4 x float> %tmp1190, undef
-  store <4 x float> %add13480, <4 x float>* undef, align 16
+  store volatile <4 x float> %add13480, <4 x float>* undef, align 16
   %tmp1191 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13481 = extractelement <4 x float> %tmp1191, i32 0
   %add13482 = fadd float %vecext13481, 0xC07BA4CCC0000000
   %tmp1192 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13483 = insertelement <4 x float> %tmp1192, float %add13482, i32 0
-  store <4 x float> %vecins13483, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins13483, <4 x float>* undef, align 16
   %tmp1193 = load <4 x float>, <4 x float>* undef, align 16
-  %add13485 = fadd float undef, 0x406B1999A0000000
+  %add13485 = fadd float %val, 0x406B1999A0000000
   %tmp1194 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13486 = insertelement <4 x float> %tmp1194, float %add13485, i32 1
-  store <4 x float> %vecins13486, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins13486, <4 x float>* undef, align 16
   %tmp1195 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13487 = extractelement <4 x float> %tmp1195, i32 2
   %add13488 = fadd float %vecext13487, 0x40647999A0000000
   %tmp1196 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13489 = insertelement <4 x float> %tmp1196, float %add13488, i32 2
-  store <4 x float> %vecins13489, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins13489, <4 x float>* undef, align 16
   %tmp1197 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13490 = extractelement <4 x float> %tmp1197, i32 3
   %tmp1198 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins13492 = insertelement <4 x float> %tmp1198, float undef, i32 3
-  store <4 x float> %vecins13492, <4 x float>* undef, align 16
+  %vecins13492 = insertelement <4 x float> %tmp1198, float %val, i32 3
+  store volatile <4 x float> %vecins13492, <4 x float>* undef, align 16
   %tmp1199 = load <4 x float>, <4 x float>* %.compoundliteral13493
   %tmp1200 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
-  %vecins13548 = insertelement <4 x float> undef, float undef, i32 3
-  store <4 x float> <float 4.540000e+02, float 3.760000e+02, float 0x406EA33340000000, float 0x405AACCCC0000000>, <4 x float>* %.compoundliteral13549
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
+  %vecins13548 = insertelement <4 x float> undef, float %val, i32 3
+  store volatile <4 x float> <float 4.540000e+02, float 3.760000e+02, float 0x406EA33340000000, float 0x405AACCCC0000000>, <4 x float>* %.compoundliteral13549
   %tmp1201 = load <4 x float>, <4 x float>* undef, align 16
-  %add13552 = fadd float undef, 3.230000e+02
+  %add13552 = fadd float %val, 3.230000e+02
   %tmp1202 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13553 = insertelement <4 x float> %tmp1202, float %add13552, i32 0
   %tmp1203 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13554 = extractelement <4 x float> %tmp1203, i32 1
   %tmp1204 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins13556 = insertelement <4 x float> %tmp1204, float undef, i32 1
-  store <4 x float> %vecins13556, <4 x float>* undef, align 16
+  %vecins13556 = insertelement <4 x float> %tmp1204, float %val, i32 1
+  store volatile <4 x float> %vecins13556, <4 x float>* undef, align 16
   %tmp1205 = load <4 x float>, <4 x float>* undef, align 16
-  %add13558 = fadd float undef, 2.625000e+02
+  %add13558 = fadd float %val, 2.625000e+02
   %tmp1206 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13559 = insertelement <4 x float> %tmp1206, float %add13558, i32 2
-  store <4 x float> %vecins13559, <4 x float>* undef, align 16
-  %add13575 = fadd float undef, -4.725000e+02
+  store volatile <4 x float> %vecins13559, <4 x float>* undef, align 16
+  %add13575 = fadd float %val, -4.725000e+02
   %tmp1207 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13576 = insertelement <4 x float> %tmp1207, float %add13575, i32 3
-  store <4 x float> %vecins13576, <4 x float>* undef, align 16
-  store <4 x float> <float 0x40334CCCC0000000, float 0xC0785CCCC0000000, float 0xC078D66660000000, float 3.745000e+02>, <4 x float>* undef
+  store volatile <4 x float> %vecins13576, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x40334CCCC0000000, float 0xC0785CCCC0000000, float 0xC078D66660000000, float 3.745000e+02>, <4 x float>* undef
   %tmp1208 = load <4 x float>, <4 x float>* undef
   %tmp1209 = load <4 x float>, <4 x float>* undef, align 16
   %add13578 = fadd <4 x float> %tmp1209, %tmp1208
-  store <4 x float> %add13578, <4 x float>* undef, align 16
+  store volatile <4 x float> %add13578, <4 x float>* undef, align 16
   %tmp1210 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1211 = load <4 x float>, <4 x float>* undef, align 16
   %add13592 = fadd <4 x float> %tmp1211, undef
-  store <4 x float> %add13592, <4 x float>* undef, align 16
+  store volatile <4 x float> %add13592, <4 x float>* undef, align 16
   %tmp1212 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13593 = extractelement <4 x float> %tmp1212, i32 0
   %add13594 = fadd float %vecext13593, 0xC0708B3340000000
   %tmp1213 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1214 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13596 = extractelement <4 x float> %tmp1214, i32 1
   %add13597 = fadd float %vecext13596, 0x40660999A0000000
-  %vecins13604 = insertelement <4 x float> undef, float undef, i32 3
-  store <4 x float> %vecins13604, <4 x float>* undef, align 16
-  store <4 x float> <float 0x407B4999A0000000, float 0xC067F66660000000, float 0xC068F999A0000000, float 0xC079233340000000>, <4 x float>* undef
+  %vecins13604 = insertelement <4 x float> undef, float %val, i32 3
+  store volatile <4 x float> %vecins13604, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x407B4999A0000000, float 0xC067F66660000000, float 0xC068F999A0000000, float 0xC079233340000000>, <4 x float>* undef
   %tmp1215 = load <4 x float>, <4 x float>* undef, align 16
   %add13606 = fadd <4 x float> %tmp1215, undef
   %tmp1216 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13607 = extractelement <4 x float> %tmp1216, i32 0
-  %vecins13609 = insertelement <4 x float> undef, float undef, i32 0
+  %vecins13609 = insertelement <4 x float> undef, float %val, i32 0
   %tmp1217 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1218 = load <4 x float>, <4 x float>* undef, align 16
-  %add13622 = fadd float undef, -3.390000e+02
+  %add13622 = fadd float %val, -3.390000e+02
   %vecins13623 = insertelement <4 x float> undef, float %add13622, i32 0
-  store <4 x float> %vecins13623, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins13623, <4 x float>* undef, align 16
   %tmp1219 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13624 = extractelement <4 x float> %tmp1219, i32 1
   %add13625 = fadd float %vecext13624, 0x405C3999A0000000
@@ -6772,41 +6772,41 @@ entry:
   %add13631 = fadd float %vecext13630, 0xC060333340000000
   %tmp1222 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13632 = insertelement <4 x float> %tmp1222, float %add13631, i32 3
-  store <4 x float> %vecins13632, <4 x float>* undef, align 16
-  store <4 x float> <float 0x4078D66660000000, float 0x4048B33340000000, float 0x4051466660000000, float -2.965000e+02>, <4 x float>* undef
+  store volatile <4 x float> %vecins13632, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x4078D66660000000, float 0x4048B33340000000, float 0x4051466660000000, float -2.965000e+02>, <4 x float>* undef
   %tmp1223 = load <4 x float>, <4 x float>* undef
   %tmp1224 = load <4 x float>, <4 x float>* undef, align 16
   %add13634 = fadd <4 x float> %tmp1224, %tmp1223
-  store <4 x float> %add13634, <4 x float>* undef, align 16
+  store volatile <4 x float> %add13634, <4 x float>* undef, align 16
   %vecext13635 = extractelement <4 x float> undef, i32 0
   %add13636 = fadd float %vecext13635, 0x406A5999A0000000
   %tmp1225 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13637 = insertelement <4 x float> %tmp1225, float %add13636, i32 0
-  store <4 x float> %vecins13637, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins13637, <4 x float>* undef, align 16
   %tmp1226 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1227 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins13643 = insertelement <4 x float> %tmp1227, float undef, i32 2
-  store <4 x float> undef, <4 x float>* undef, align 16
+  %vecins13643 = insertelement <4 x float> %tmp1227, float %val, i32 2
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1228 = load <4 x float>, <4 x float>* undef, align 16
-  %add13785 = fadd float undef, 0x4068866660000000
+  %add13785 = fadd float %val, 0x4068866660000000
   %tmp1229 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13786 = insertelement <4 x float> %tmp1229, float %add13785, i32 3
-  store <4 x float> %vecins13786, <4 x float>* undef, align 16
-  store <4 x float> <float 0x407704CCC0000000, float 0x4047B33340000000, float 0x40797B3340000000, float 0xC0652CCCC0000000>, <4 x float>* %.compoundliteral13787
+  store volatile <4 x float> %vecins13786, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x407704CCC0000000, float 0x4047B33340000000, float 0x40797B3340000000, float 0xC0652CCCC0000000>, <4 x float>* %.compoundliteral13787
   %tmp1230 = load <4 x float>, <4 x float>* undef, align 16
   %add13788 = fadd <4 x float> %tmp1230, undef
   %tmp1231 = load <4 x float>, <4 x float>* undef
   %tmp1232 = load <4 x float>, <4 x float>* undef, align 16
   %add13802 = fadd <4 x float> %tmp1232, %tmp1231
-  store <4 x float> %add13802, <4 x float>* undef, align 16
+  store volatile <4 x float> %add13802, <4 x float>* undef, align 16
   %tmp1233 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13803 = extractelement <4 x float> %tmp1233, i32 0
   %add13804 = fadd float %vecext13803, -2.900000e+01
   %tmp1234 = load <4 x float>, <4 x float>* undef, align 16
   %vecins13805 = insertelement <4 x float> %tmp1234, float %add13804, i32 0
-  store <4 x float> %vecins13805, <4 x float>* undef, align 16
+  store volatile <4 x float> %vecins13805, <4 x float>* undef, align 16
   %tmp1235 = load <4 x float>, <4 x float>* undef, align 16
-  %add13807 = fadd float undef, 6.400000e+01
+  %add13807 = fadd float %val, 6.400000e+01
   %tmp1236 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1237 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13809 = extractelement <4 x float> %tmp1237, i32 2
@@ -6814,28 +6814,28 @@ entry:
   %vecext13812 = extractelement <4 x float> %tmp1238, i32 3
   %add13813 = fadd float %vecext13812, -3.615000e+02
   %vecins13814 = insertelement <4 x float> undef, float %add13813, i32 3
-  store <4 x float> %vecins13814, <4 x float>* undef, align 16
-  store <4 x float> <float -2.270000e+02, float -1.500000e+01, float 0x407084CCC0000000, float -1.425000e+02>, <4 x float>* undef
+  store volatile <4 x float> %vecins13814, <4 x float>* undef, align 16
+  store volatile <4 x float> <float -2.270000e+02, float -1.500000e+01, float 0x407084CCC0000000, float -1.425000e+02>, <4 x float>* undef
   %tmp1239 = load <4 x float>, <4 x float>* undef
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1240 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13817 = extractelement <4 x float> %tmp1240, i32 0
-  %vecins13856 = insertelement <4 x float> undef, float undef, i32 3
-  store <4 x float> %vecins13856, <4 x float>* undef, align 16
-  store <4 x float> <float 0x40656CCCC0000000, float 0xC0656999A0000000, float 0x40778E6660000000, float 0x407ECE6660000000>, <4 x float>* undef
+  %vecins13856 = insertelement <4 x float> undef, float %val, i32 3
+  store volatile <4 x float> %vecins13856, <4 x float>* undef, align 16
+  store volatile <4 x float> <float 0x40656CCCC0000000, float 0xC0656999A0000000, float 0x40778E6660000000, float 0x407ECE6660000000>, <4 x float>* undef
   %tmp1241 = load <4 x float>, <4 x float>* undef
   %tmp1242 = load <4 x float>, <4 x float>* undef, align 16
-  store <4 x float> undef, <4 x float>* undef, align 16
+  store volatile <4 x float> undef, <4 x float>* undef, align 16
   %tmp1243 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13859 = extractelement <4 x float> %tmp1243, i32 0
   %tmp1244 = load <4 x float>, <4 x float>* undef, align 16
-  %vecins13861 = insertelement <4 x float> %tmp1244, float undef, i32 0
+  %vecins13861 = insertelement <4 x float> %tmp1244, float %val, i32 0
   %tmp1245 = load <4 x float>, <4 x float>* undef, align 16
   %vecext13862 = extractelement <4 x float> %tmp1245, i32 1
   %add13863 = fadd float %vecext13862, -1.380000e+02
   %vecins13864 = insertelement <4 x float> undef, float %add13863, i32 1
-  %vecins13867 = insertelement <4 x float> undef, float undef, i32 2
-  store <4 x float> %vecins13867, <4 x float>* undef, align 16
+  %vecins13867 = insertelement <4 x float> undef, float %val, i32 2
+  store volatile <4 x float> %vecins13867, <4 x float>* undef, align 16
   %tmp1246 = load <4 x float>, <4 x float>* undef, align 16
   %tmp1247 = load <4 x float>, <4 x float>* undef, align 16
   ret <4 x float> undef

diff --git a/llvm/test/CodeGen/Mips/Fast-ISel/callabi.ll b/llvm/test/CodeGen/Mips/Fast-ISel/callabi.ll
index f22fbcc..9025af9 100644 (file)
--- a/llvm/test/CodeGen/Mips/Fast-ISel/callabi.ll
+++ b/llvm/test/CodeGen/Mips/Fast-ISel/callabi.ll
@@ -244,12 +244,12 @@ define void @cxiiiiconv() {
   ; ALL-DAG:        lw      $[[REG_C1_ADDR:[0-9]+]], %got(c1)($[[REG_GP]])
   ; ALL-DAG:        lbu     $[[REG_C1:[0-9]+]], 0($[[REG_C1_ADDR]])
   ; 32R1-DAG:       sll     $[[REG_C1_1:[0-9]+]], $[[REG_C1]], 24
-  ; 32R1-DAG:       sra     $4, $[[REG_C1_1]], 24
-  ; 32R2-DAG:       seb     $4, $[[REG_C1]]
+  ; 32R1-DAG:       sra     $5, $[[REG_C1_1]], 24
+  ; 32R2-DAG:       seb     $5, $[[REG_C1]]
   ; FIXME: andi is superfulous
   ; ALL-DAG:        lw      $[[REG_UC1_ADDR:[0-9]+]], %got(uc1)($[[REG_GP]])
   ; ALL-DAG:        lbu     $[[REG_UC1:[0-9]+]], 0($[[REG_UC1_ADDR]])
-  ; ALL-DAG:        andi    $5, $[[REG_UC1]], 255
+  ; ALL-DAG:        andi    $4, $[[REG_UC1]], 255
   ; ALL-DAG:        lw      $[[REG_S1_ADDR:[0-9]+]], %got(s1)($[[REG_GP]])
   ; ALL-DAG:        lhu     $[[REG_S1:[0-9]+]], 0($[[REG_S1_ADDR]])
   ; 32R1-DAG:       sll     $[[REG_S1_1:[0-9]+]], $[[REG_S1]], 16

diff --git a/llvm/test/CodeGen/Mips/Fast-ISel/pr40325.ll b/llvm/test/CodeGen/Mips/Fast-ISel/pr40325.ll
index a9ce70f..6befe70 100644 (file)
--- a/llvm/test/CodeGen/Mips/Fast-ISel/pr40325.ll
+++ b/llvm/test/CodeGen/Mips/Fast-ISel/pr40325.ll
@@ -5,8 +5,8 @@ define void @test(i32 %x, i1* %p) nounwind {
 ; CHECK-LABEL: test:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    move $1, $4
-; CHECK-NEXT:    andi $4, $4, 1
-; CHECK-NEXT:    sb $4, 0($5)
+; CHECK-NEXT:    andi $2, $4, 1
+; CHECK-NEXT:    sb $2, 0($5)
 ; CHECK-NEXT:    andi $1, $1, 1
 ; CHECK-NEXT:    bgtz $1, $BB0_1
 ; CHECK-NEXT:    nop

diff --git a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/add.ll b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/add.ll
index 2ecc5ca..57dbddc 100644 (file)
--- a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/add.ll
+++ b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/add.ll
@@ -14,9 +14,9 @@ entry:
 define signext i8 @add_i8_sext(i8 signext %a, i8 signext %b) {
 ; MIPS32-LABEL: add_i8_sext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    addu $4, $5, $4
-; MIPS32-NEXT:    sll $4, $4, 24
-; MIPS32-NEXT:    sra $2, $4, 24
+; MIPS32-NEXT:    addu $1, $5, $4
+; MIPS32-NEXT:    sll $1, $1, 24
+; MIPS32-NEXT:    sra $2, $1, 24
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -27,9 +27,9 @@ entry:
 define zeroext i8 @add_i8_zext(i8 zeroext %a, i8 zeroext %b) {
 ; MIPS32-LABEL: add_i8_zext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    addu $4, $5, $4
-; MIPS32-NEXT:    ori $5, $zero, 255
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    addu $1, $5, $4
+; MIPS32-NEXT:    ori $2, $zero, 255
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -51,9 +51,9 @@ entry:
 define signext i16 @add_i16_sext(i16 signext %a, i16 signext %b) {
 ; MIPS32-LABEL: add_i16_sext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    addu $4, $5, $4
-; MIPS32-NEXT:    sll $4, $4, 16
-; MIPS32-NEXT:    sra $2, $4, 16
+; MIPS32-NEXT:    addu $1, $5, $4
+; MIPS32-NEXT:    sll $1, $1, 16
+; MIPS32-NEXT:    sra $2, $1, 16
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -64,9 +64,9 @@ entry:
 define zeroext i16 @add_i16_zext(i16 zeroext %a, i16 zeroext %b) {
 ; MIPS32-LABEL: add_i16_zext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    addu $4, $5, $4
-; MIPS32-NEXT:    ori $5, $zero, 65535
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    addu $1, $5, $4
+; MIPS32-NEXT:    ori $2, $zero, 65535
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -89,15 +89,15 @@ define i64 @add_i64(i64 %a, i64 %b) {
 ; MIPS32-LABEL: add_i64:
 ; MIPS32:       # %bb.0: # %entry
 ; MIPS32-NEXT:    ori $1, $zero, 0
-; MIPS32-NEXT:    addu $4, $6, $4
-; MIPS32-NEXT:    ori $2, $zero, 1
-; MIPS32-NEXT:    and $1, $1, $2
-; MIPS32-NEXT:    addu $1, $4, $1
+; MIPS32-NEXT:    addu $2, $6, $4
+; MIPS32-NEXT:    ori $3, $zero, 1
+; MIPS32-NEXT:    and $1, $1, $3
+; MIPS32-NEXT:    addu $1, $2, $1
 ; MIPS32-NEXT:    sltu $2, $1, $6
-; MIPS32-NEXT:    addu $4, $7, $5
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $2, $5
-; MIPS32-NEXT:    addu $3, $4, $2
+; MIPS32-NEXT:    addu $3, $7, $5
+; MIPS32-NEXT:    ori $4, $zero, 1
+; MIPS32-NEXT:    and $2, $2, $4
+; MIPS32-NEXT:    addu $3, $3, $2
 ; MIPS32-NEXT:    move $2, $1
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
@@ -155,12 +155,12 @@ declare { i32, i1 } @llvm.uadd.with.overflow.i32(i32, i32)
 define void @uadd_with_overflow(i32 %lhs, i32 %rhs, i32* %padd, i1* %pcarry_flag) {
 ; MIPS32-LABEL: uadd_with_overflow:
 ; MIPS32:       # %bb.0:
-; MIPS32-NEXT:    addu $4, $4, $5
-; MIPS32-NEXT:    sltu $5, $4, $5
-; MIPS32-NEXT:    ori $1, $zero, 1
-; MIPS32-NEXT:    and $1, $5, $1
-; MIPS32-NEXT:    sb $1, 0($7)
-; MIPS32-NEXT:    sw $4, 0($6)
+; MIPS32-NEXT:    addu $1, $4, $5
+; MIPS32-NEXT:    sltu $2, $1, $5
+; MIPS32-NEXT:    ori $3, $zero, 1
+; MIPS32-NEXT:    and $2, $2, $3
+; MIPS32-NEXT:    sb $2, 0($7)
+; MIPS32-NEXT:    sw $1, 0($6)
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
   %res = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 %lhs, i32 %rhs)

diff --git a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/icmp.ll b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/icmp.ll
index c9ff477..fc5f397 100644 (file)
--- a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/icmp.ll
+++ b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/icmp.ll
@@ -4,10 +4,10 @@
 define i32 @eq(i32 %a, i32 %b){
 ; MIPS32-LABEL: eq:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    xor $4, $4, $5
-; MIPS32-NEXT:    sltiu $4, $4, 1
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    xor $1, $4, $5
+; MIPS32-NEXT:    sltiu $1, $1, 1
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -19,10 +19,10 @@ entry:
 define i32 @ne(i32 %a, i32 %b) {
 ; MIPS32-LABEL: ne:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    xor $4, $4, $5
-; MIPS32-NEXT:    sltu $4, $zero, $4
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    xor $1, $4, $5
+; MIPS32-NEXT:    sltu $1, $zero, $1
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -34,9 +34,9 @@ entry:
 define i32 @sgt(i32 %a, i32 %b) {
 ; MIPS32-LABEL: sgt:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    slt $4, $5, $4
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    slt $1, $5, $4
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -48,10 +48,10 @@ entry:
 define i32 @sge(i32 %a, i32 %b) {
 ; MIPS32-LABEL: sge:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    slt $4, $4, $5
-; MIPS32-NEXT:    xori $4, $4, 1
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    slt $1, $4, $5
+; MIPS32-NEXT:    xori $1, $1, 1
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -63,9 +63,9 @@ entry:
 define i32 @slt(i32 %a, i32 %b) {
 ; MIPS32-LABEL: slt:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    slt $4, $4, $5
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    slt $1, $4, $5
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -77,10 +77,10 @@ entry:
 define i32 @sle(i32 %a, i32 %b) {
 ; MIPS32-LABEL: sle:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    slt $4, $5, $4
-; MIPS32-NEXT:    xori $4, $4, 1
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    slt $1, $5, $4
+; MIPS32-NEXT:    xori $1, $1, 1
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -92,9 +92,9 @@ entry:
 define i32 @ugt(i32 %a, i32 %b) {
 ; MIPS32-LABEL: ugt:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    sltu $4, $5, $4
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    sltu $1, $5, $4
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -106,10 +106,10 @@ entry:
 define i32 @uge(i32 %a, i32 %b) {
 ; MIPS32-LABEL: uge:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    sltu $4, $4, $5
-; MIPS32-NEXT:    xori $4, $4, 1
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    sltu $1, $4, $5
+; MIPS32-NEXT:    xori $1, $1, 1
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -121,9 +121,9 @@ entry:
 define i32 @ult(i32 %a, i32 %b) {
 ; MIPS32-LABEL: ult:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    sltu $4, $4, $5
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    sltu $1, $4, $5
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -135,10 +135,10 @@ entry:
 define i32 @ule(i32 %a, i32 %b) {
 ; MIPS32-LABEL: ule:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    sltu $4, $5, $4
-; MIPS32-NEXT:    xori $4, $4, 1
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    sltu $1, $5, $4
+; MIPS32-NEXT:    xori $1, $1, 1
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:

diff --git a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/mul.ll b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/mul.ll
index c7dc2f8..7e5fc78 100644 (file)
--- a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/mul.ll
+++ b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/mul.ll
@@ -14,9 +14,9 @@ entry:
 define signext i8 @mul_i8_sext(i8 signext %a, i8 signext %b) {
 ; MIPS32-LABEL: mul_i8_sext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    mul $4, $5, $4
-; MIPS32-NEXT:    sll $4, $4, 24
-; MIPS32-NEXT:    sra $2, $4, 24
+; MIPS32-NEXT:    mul $1, $5, $4
+; MIPS32-NEXT:    sll $1, $1, 24
+; MIPS32-NEXT:    sra $2, $1, 24
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -27,9 +27,9 @@ entry:
 define zeroext i8 @mul_i8_zext(i8 zeroext %a, i8 zeroext %b) {
 ; MIPS32-LABEL: mul_i8_zext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    mul $4, $5, $4
-; MIPS32-NEXT:    ori $5, $zero, 255
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    mul $1, $5, $4
+; MIPS32-NEXT:    ori $2, $zero, 255
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -51,9 +51,9 @@ entry:
 define signext i16 @mul_i16_sext(i16 signext %a, i16 signext %b) {
 ; MIPS32-LABEL: mul_i16_sext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    mul $4, $5, $4
-; MIPS32-NEXT:    sll $4, $4, 16
-; MIPS32-NEXT:    sra $2, $4, 16
+; MIPS32-NEXT:    mul $1, $5, $4
+; MIPS32-NEXT:    sll $1, $1, 16
+; MIPS32-NEXT:    sra $2, $1, 16
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -64,9 +64,9 @@ entry:
 define zeroext i16 @mul_i16_zext(i16 zeroext %a, i16 zeroext %b) {
 ; MIPS32-LABEL: mul_i16_zext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    mul $4, $5, $4
-; MIPS32-NEXT:    ori $5, $zero, 65535
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    mul $1, $5, $4
+; MIPS32-NEXT:    ori $2, $zero, 65535
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -89,12 +89,12 @@ define i64 @mul_i64(i64 %a, i64 %b) {
 ; MIPS32-LABEL: mul_i64:
 ; MIPS32:       # %bb.0: # %entry
 ; MIPS32-NEXT:    mul $2, $6, $4
-; MIPS32-NEXT:    mul $7, $7, $4
-; MIPS32-NEXT:    mul $5, $6, $5
+; MIPS32-NEXT:    mul $1, $7, $4
+; MIPS32-NEXT:    mul $3, $6, $5
 ; MIPS32-NEXT:    multu $6, $4
 ; MIPS32-NEXT:    mfhi $4
-; MIPS32-NEXT:    addu $5, $7, $5
-; MIPS32-NEXT:    addu $3, $5, $4
+; MIPS32-NEXT:    addu $1, $1, $3
+; MIPS32-NEXT:    addu $3, $1, $4
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -191,13 +191,13 @@ define void @umul_with_overflow(i32 %lhs, i32 %rhs, i32* %pmul, i1* %pcarry_flag
 ; MIPS32:       # %bb.0:
 ; MIPS32-NEXT:    mul $1, $4, $5
 ; MIPS32-NEXT:    multu $4, $5
-; MIPS32-NEXT:    mfhi $4
-; MIPS32-NEXT:    ori $5, $zero, 0
-; MIPS32-NEXT:    xor $4, $4, $5
-; MIPS32-NEXT:    sltu $4, $zero, $4
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $4, $4, $5
-; MIPS32-NEXT:    sb $4, 0($7)
+; MIPS32-NEXT:    mfhi $2
+; MIPS32-NEXT:    ori $3, $zero, 0
+; MIPS32-NEXT:    xor $2, $2, $3
+; MIPS32-NEXT:    sltu $2, $zero, $2
+; MIPS32-NEXT:    ori $3, $zero, 1
+; MIPS32-NEXT:    and $2, $2, $3
+; MIPS32-NEXT:    sb $2, 0($7)
 ; MIPS32-NEXT:    sw $1, 0($6)
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop

diff --git a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/rem_and_div.ll b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/rem_and_div.ll
index 0b2d774..33149fe 100644 (file)
--- a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/rem_and_div.ll
+++ b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/rem_and_div.ll
@@ -5,15 +5,15 @@
 define signext i8 @sdiv_i8(i8 signext %a, i8 signext %b) {
 ; MIPS32-LABEL: sdiv_i8:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    sll $5, $5, 24
-; MIPS32-NEXT:    sra $5, $5, 24
-; MIPS32-NEXT:    sll $4, $4, 24
-; MIPS32-NEXT:    sra $4, $4, 24
-; MIPS32-NEXT:    div $zero, $5, $4
-; MIPS32-NEXT:    teq $4, $zero, 7
-; MIPS32-NEXT:    mflo $4
-; MIPS32-NEXT:    sll $4, $4, 24
-; MIPS32-NEXT:    sra $2, $4, 24
+; MIPS32-NEXT:    sll $1, $5, 24
+; MIPS32-NEXT:    sra $1, $1, 24
+; MIPS32-NEXT:    sll $2, $4, 24
+; MIPS32-NEXT:    sra $2, $2, 24
+; MIPS32-NEXT:    div $zero, $1, $2
+; MIPS32-NEXT:    teq $2, $zero, 7
+; MIPS32-NEXT:    mflo $1
+; MIPS32-NEXT:    sll $1, $1, 24
+; MIPS32-NEXT:    sra $2, $1, 24
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -24,15 +24,15 @@ entry:
 define signext i16 @sdiv_i16(i16 signext %a, i16 signext %b) {
 ; MIPS32-LABEL: sdiv_i16:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    sll $5, $5, 16
-; MIPS32-NEXT:    sra $5, $5, 16
-; MIPS32-NEXT:    sll $4, $4, 16
-; MIPS32-NEXT:    sra $4, $4, 16
-; MIPS32-NEXT:    div $zero, $5, $4
-; MIPS32-NEXT:    teq $4, $zero, 7
-; MIPS32-NEXT:    mflo $4
-; MIPS32-NEXT:    sll $4, $4, 16
-; MIPS32-NEXT:    sra $2, $4, 16
+; MIPS32-NEXT:    sll $1, $5, 16
+; MIPS32-NEXT:    sra $1, $1, 16
+; MIPS32-NEXT:    sll $2, $4, 16
+; MIPS32-NEXT:    sra $2, $2, 16
+; MIPS32-NEXT:    div $zero, $1, $2
+; MIPS32-NEXT:    teq $2, $zero, 7
+; MIPS32-NEXT:    mflo $1
+; MIPS32-NEXT:    sll $1, $1, 16
+; MIPS32-NEXT:    sra $2, $1, 16
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -81,15 +81,15 @@ entry:
 define signext i8 @srem_i8(i8 signext %a, i8 signext %b) {
 ; MIPS32-LABEL: srem_i8:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    sll $5, $5, 24
-; MIPS32-NEXT:    sra $5, $5, 24
-; MIPS32-NEXT:    sll $4, $4, 24
-; MIPS32-NEXT:    sra $4, $4, 24
-; MIPS32-NEXT:    div $zero, $5, $4
-; MIPS32-NEXT:    teq $4, $zero, 7
-; MIPS32-NEXT:    mflo $4
-; MIPS32-NEXT:    sll $4, $4, 24
-; MIPS32-NEXT:    sra $2, $4, 24
+; MIPS32-NEXT:    sll $1, $5, 24
+; MIPS32-NEXT:    sra $1, $1, 24
+; MIPS32-NEXT:    sll $2, $4, 24
+; MIPS32-NEXT:    sra $2, $2, 24
+; MIPS32-NEXT:    div $zero, $1, $2
+; MIPS32-NEXT:    teq $2, $zero, 7
+; MIPS32-NEXT:    mflo $1
+; MIPS32-NEXT:    sll $1, $1, 24
+; MIPS32-NEXT:    sra $2, $1, 24
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -100,15 +100,15 @@ entry:
 define signext i16 @srem_i16(i16 signext %a, i16 signext %b) {
 ; MIPS32-LABEL: srem_i16:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    sll $5, $5, 16
-; MIPS32-NEXT:    sra $5, $5, 16
-; MIPS32-NEXT:    sll $4, $4, 16
-; MIPS32-NEXT:    sra $4, $4, 16
-; MIPS32-NEXT:    div $zero, $5, $4
-; MIPS32-NEXT:    teq $4, $zero, 7
-; MIPS32-NEXT:    mfhi $4
-; MIPS32-NEXT:    sll $4, $4, 16
-; MIPS32-NEXT:    sra $2, $4, 16
+; MIPS32-NEXT:    sll $1, $5, 16
+; MIPS32-NEXT:    sra $1, $1, 16
+; MIPS32-NEXT:    sll $2, $4, 16
+; MIPS32-NEXT:    sra $2, $2, 16
+; MIPS32-NEXT:    div $zero, $1, $2
+; MIPS32-NEXT:    teq $2, $zero, 7
+; MIPS32-NEXT:    mfhi $1
+; MIPS32-NEXT:    sll $1, $1, 16
+; MIPS32-NEXT:    sra $2, $1, 16
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -159,10 +159,10 @@ define signext i8 @udiv_i8(i8 signext %a, i8 signext %b) {
 ; MIPS32:       # %bb.0: # %entry
 ; MIPS32-NEXT:    ori $1, $zero, 255
 ; MIPS32-NEXT:    and $1, $5, $1
-; MIPS32-NEXT:    ori $5, $zero, 255
-; MIPS32-NEXT:    and $4, $4, $5
-; MIPS32-NEXT:    divu $zero, $1, $4
-; MIPS32-NEXT:    teq $4, $zero, 7
+; MIPS32-NEXT:    ori $2, $zero, 255
+; MIPS32-NEXT:    and $2, $4, $2
+; MIPS32-NEXT:    divu $zero, $1, $2
+; MIPS32-NEXT:    teq $2, $zero, 7
 ; MIPS32-NEXT:    mflo $1
 ; MIPS32-NEXT:    sll $1, $1, 24
 ; MIPS32-NEXT:    sra $2, $1, 24
@@ -178,10 +178,10 @@ define signext i16 @udiv_i16(i16 signext %a, i16 signext %b) {
 ; MIPS32:       # %bb.0: # %entry
 ; MIPS32-NEXT:    ori $1, $zero, 65535
 ; MIPS32-NEXT:    and $1, $5, $1
-; MIPS32-NEXT:    ori $5, $zero, 65535
-; MIPS32-NEXT:    and $4, $4, $5
-; MIPS32-NEXT:    divu $zero, $1, $4
-; MIPS32-NEXT:    teq $4, $zero, 7
+; MIPS32-NEXT:    ori $2, $zero, 65535
+; MIPS32-NEXT:    and $2, $4, $2
+; MIPS32-NEXT:    divu $zero, $1, $2
+; MIPS32-NEXT:    teq $2, $zero, 7
 ; MIPS32-NEXT:    mflo $1
 ; MIPS32-NEXT:    sll $1, $1, 16
 ; MIPS32-NEXT:    sra $2, $1, 16
@@ -235,10 +235,10 @@ define signext i8 @urem_i8(i8 signext %a, i8 signext %b) {
 ; MIPS32:       # %bb.0: # %entry
 ; MIPS32-NEXT:    ori $1, $zero, 255
 ; MIPS32-NEXT:    and $1, $5, $1
-; MIPS32-NEXT:    ori $5, $zero, 255
-; MIPS32-NEXT:    and $4, $4, $5
-; MIPS32-NEXT:    divu $zero, $1, $4
-; MIPS32-NEXT:    teq $4, $zero, 7
+; MIPS32-NEXT:    ori $2, $zero, 255
+; MIPS32-NEXT:    and $2, $4, $2
+; MIPS32-NEXT:    divu $zero, $1, $2
+; MIPS32-NEXT:    teq $2, $zero, 7
 ; MIPS32-NEXT:    mfhi $1
 ; MIPS32-NEXT:    sll $1, $1, 24
 ; MIPS32-NEXT:    sra $2, $1, 24
@@ -254,10 +254,10 @@ define signext i16 @urem_i16(i16 signext %a, i16 signext %b) {
 ; MIPS32:       # %bb.0: # %entry
 ; MIPS32-NEXT:    ori $1, $zero, 65535
 ; MIPS32-NEXT:    and $1, $5, $1
-; MIPS32-NEXT:    ori $5, $zero, 65535
-; MIPS32-NEXT:    and $4, $4, $5
-; MIPS32-NEXT:    divu $zero, $1, $4
-; MIPS32-NEXT:    teq $4, $zero, 7
+; MIPS32-NEXT:    ori $2, $zero, 65535
+; MIPS32-NEXT:    and $2, $4, $2
+; MIPS32-NEXT:    divu $zero, $1, $2
+; MIPS32-NEXT:    teq $2, $zero, 7
 ; MIPS32-NEXT:    mfhi $1
 ; MIPS32-NEXT:    sll $1, $1, 16
 ; MIPS32-NEXT:    sra $2, $1, 16

diff --git a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/select.ll b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/select.ll
index 4babe89..1aa2e42 100644 (file)
--- a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/select.ll
+++ b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/select.ll
@@ -60,11 +60,11 @@ entry:
 define i32 @select_with_negation(i32 %a, i32 %b, i32 %x, i32 %y) {
 ; MIPS32-LABEL: select_with_negation:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    slt $4, $4, $5
-; MIPS32-NEXT:    not $4, $4
-; MIPS32-NEXT:    ori $5, $zero, 1
-; MIPS32-NEXT:    and $4, $4, $5
-; MIPS32-NEXT:    movn $7, $6, $4
+; MIPS32-NEXT:    slt $1, $4, $5
+; MIPS32-NEXT:    not $1, $1
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $1, $1, $2
+; MIPS32-NEXT:    movn $7, $6, $1
 ; MIPS32-NEXT:    move $2, $7
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop

diff --git a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/stack_args.ll b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/stack_args.ll
index 47c11e6..0fe1b7c 100644 (file)
--- a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/stack_args.ll
+++ b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/stack_args.ll
@@ -13,7 +13,7 @@ define  i32 @g(i32  %x1, i32 %x2, i32 %x3, i32 %x4, i32 %x5){
 ; MIPS32-NEXT:    addiu $1, $sp, 48
 ; MIPS32-NEXT:    lw $1, 0($1)
 ; MIPS32-NEXT:    move $2, $sp
-; MIPS32-NEXT:    ori $3, $zero, 1
+; MIPS32-NEXT:    ori $3, $zero, 16
 ; MIPS32-NEXT:    addu $2, $2, $3
 ; MIPS32-NEXT:    sw $1, 0($2)
 ; MIPS32-NEXT:    jal f

diff --git a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/sub.ll b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/sub.ll
index 7086686..7e96449 100644 (file)
--- a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/sub.ll
+++ b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/sub.ll
@@ -15,9 +15,9 @@ entry:
 define signext i8 @sub_i8_sext(i8 signext %a, i8 signext %b) {
 ; MIPS32-LABEL: sub_i8_sext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    subu $4, $5, $4
-; MIPS32-NEXT:    sll $4, $4, 24
-; MIPS32-NEXT:    sra $2, $4, 24
+; MIPS32-NEXT:    subu $1, $5, $4
+; MIPS32-NEXT:    sll $1, $1, 24
+; MIPS32-NEXT:    sra $2, $1, 24
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -28,9 +28,9 @@ entry:
 define zeroext i8 @sub_i8_zext(i8 zeroext %a, i8 zeroext %b) {
 ; MIPS32-LABEL: sub_i8_zext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    subu $4, $5, $4
-; MIPS32-NEXT:    ori $5, $zero, 255
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    subu $1, $5, $4
+; MIPS32-NEXT:    ori $2, $zero, 255
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -52,9 +52,9 @@ entry:
 define signext i16 @sub_i16_sext(i16 signext %a, i16 signext %b) {
 ; MIPS32-LABEL: sub_i16_sext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    subu $4, $5, $4
-; MIPS32-NEXT:    sll $4, $4, 16
-; MIPS32-NEXT:    sra $2, $4, 16
+; MIPS32-NEXT:    subu $1, $5, $4
+; MIPS32-NEXT:    sll $1, $1, 16
+; MIPS32-NEXT:    sra $2, $1, 16
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -65,9 +65,9 @@ entry:
 define zeroext i16 @sub_i16_zext(i16 zeroext %a, i16 zeroext %b) {
 ; MIPS32-LABEL: sub_i16_zext:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    subu $4, $5, $4
-; MIPS32-NEXT:    ori $5, $zero, 65535
-; MIPS32-NEXT:    and $2, $4, $5
+; MIPS32-NEXT:    subu $1, $5, $4
+; MIPS32-NEXT:    ori $2, $zero, 65535
+; MIPS32-NEXT:    and $2, $1, $2
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -90,11 +90,11 @@ define i64 @sub_i64(i64 %a, i64 %b) {
 ; MIPS32-LABEL: sub_i64:
 ; MIPS32:       # %bb.0: # %entry
 ; MIPS32-NEXT:    subu $2, $6, $4
-; MIPS32-NEXT:    sltu $4, $6, $4
-; MIPS32-NEXT:    subu $5, $7, $5
-; MIPS32-NEXT:    ori $6, $zero, 1
-; MIPS32-NEXT:    and $4, $4, $6
-; MIPS32-NEXT:    subu $3, $5, $4
+; MIPS32-NEXT:    sltu $1, $6, $4
+; MIPS32-NEXT:    subu $3, $7, $5
+; MIPS32-NEXT:    ori $4, $zero, 1
+; MIPS32-NEXT:    and $1, $1, $4
+; MIPS32-NEXT:    subu $3, $3, $1
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:

diff --git a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/truncStore_and_aExtLoad.ll b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/truncStore_and_aExtLoad.ll
index 3fc7e63..e051b30 100644 (file)
--- a/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/truncStore_and_aExtLoad.ll
+++ b/llvm/test/CodeGen/Mips/GlobalISel/llvm-ir/truncStore_and_aExtLoad.ll
@@ -26,9 +26,9 @@ entry:
 define void @load_store_i1(i1* %px, i1* %py) {
 ; MIPS32-LABEL: load_store_i1:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    lbu $5, 0($5)
-; MIPS32-NEXT:    ori $1, $zero, 1
-; MIPS32-NEXT:    and $1, $5, $1
+; MIPS32-NEXT:    lbu $1, 0($5)
+; MIPS32-NEXT:    ori $2, $zero, 1
+; MIPS32-NEXT:    and $1, $1, $2
 ; MIPS32-NEXT:    sb $1, 0($4)
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
@@ -41,8 +41,8 @@ entry:
 define void @load_store_i8(i8* %px, i8* %py) {
 ; MIPS32-LABEL: load_store_i8:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    lbu $5, 0($5)
-; MIPS32-NEXT:    sb $5, 0($4)
+; MIPS32-NEXT:    lbu $1, 0($5)
+; MIPS32-NEXT:    sb $1, 0($4)
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -54,8 +54,8 @@ entry:
 define void @load_store_i16(i16* %px, i16* %py) {
 ; MIPS32-LABEL: load_store_i16:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    lhu $5, 0($5)
-; MIPS32-NEXT:    sh $5, 0($4)
+; MIPS32-NEXT:    lhu $1, 0($5)
+; MIPS32-NEXT:    sh $1, 0($4)
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:
@@ -67,8 +67,8 @@ entry:
 define void @load_store_i32(i32* %px, i32* %py) {
 ; MIPS32-LABEL: load_store_i32:
 ; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    lw $5, 0($5)
-; MIPS32-NEXT:    sw $5, 0($4)
+; MIPS32-NEXT:    lw $1, 0($5)
+; MIPS32-NEXT:    sw $1, 0($4)
 ; MIPS32-NEXT:    jr $ra
 ; MIPS32-NEXT:    nop
 entry:

diff --git a/llvm/test/CodeGen/Mips/atomic.ll b/llvm/test/CodeGen/Mips/atomic.ll
index 3d516ea..b0207ff 100644 (file)
--- a/llvm/test/CodeGen/Mips/atomic.ll
+++ b/llvm/test/CodeGen/Mips/atomic.ll
@@ -56,17 +56,16 @@ define i32 @AtomicLoadAdd32(i32 signext %incr) nounwind {
 ; MIPS32O0:       # %bb.0: # %entry
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32O0-NEXT:  $BB0_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $25, 0($2)
-; MIPS32O0-NEXT:    addu $1, $25, $4
-; MIPS32O0-NEXT:    sc $1, 0($2)
-; MIPS32O0-NEXT:    beqz $1, $BB0_1
+; MIPS32O0-NEXT:    ll $2, 0($1)
+; MIPS32O0-NEXT:    addu $3, $2, $4
+; MIPS32O0-NEXT:    sc $3, 0($1)
+; MIPS32O0-NEXT:    beqz $3, $BB0_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    move $2, $25
 ; MIPS32O0-NEXT:    jr $ra
 ; MIPS32O0-NEXT:    nop
 ;
@@ -108,18 +107,18 @@ define i32 @AtomicLoadAdd32(i32 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32R6O0-NEXT:  $BB0_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $1, 0($2)
-; MIPS32R6O0-NEXT:    addu $3, $1, $4
-; MIPS32R6O0-NEXT:    sc $3, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $3, $BB0_1
+; MIPS32R6O0-NEXT:    ll $3, 0($1)
+; MIPS32R6O0-NEXT:    addu $5, $3, $4
+; MIPS32R6O0-NEXT:    sc $5, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $5, $BB0_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    move $2, $1
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    move $2, $3
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;
@@ -317,17 +316,16 @@ define i32 @AtomicLoadSub32(i32 signext %incr) nounwind {
 ; MIPS32O0:       # %bb.0: # %entry
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32O0-NEXT:  $BB1_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $25, 0($2)
-; MIPS32O0-NEXT:    subu $1, $25, $4
-; MIPS32O0-NEXT:    sc $1, 0($2)
-; MIPS32O0-NEXT:    beqz $1, $BB1_1
+; MIPS32O0-NEXT:    ll $2, 0($1)
+; MIPS32O0-NEXT:    subu $3, $2, $4
+; MIPS32O0-NEXT:    sc $3, 0($1)
+; MIPS32O0-NEXT:    beqz $3, $BB1_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    move $2, $25
 ; MIPS32O0-NEXT:    jr $ra
 ; MIPS32O0-NEXT:    nop
 ;
@@ -369,18 +367,18 @@ define i32 @AtomicLoadSub32(i32 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32R6O0-NEXT:  $BB1_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $1, 0($2)
-; MIPS32R6O0-NEXT:    subu $3, $1, $4
-; MIPS32R6O0-NEXT:    sc $3, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $3, $BB1_1
+; MIPS32R6O0-NEXT:    ll $3, 0($1)
+; MIPS32R6O0-NEXT:    subu $5, $3, $4
+; MIPS32R6O0-NEXT:    sc $5, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $5, $BB1_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    move $2, $1
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    move $2, $3
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;
@@ -578,17 +576,16 @@ define i32 @AtomicLoadXor32(i32 signext %incr) nounwind {
 ; MIPS32O0:       # %bb.0: # %entry
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32O0-NEXT:  $BB2_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $25, 0($2)
-; MIPS32O0-NEXT:    xor $1, $25, $4
-; MIPS32O0-NEXT:    sc $1, 0($2)
-; MIPS32O0-NEXT:    beqz $1, $BB2_1
+; MIPS32O0-NEXT:    ll $2, 0($1)
+; MIPS32O0-NEXT:    xor $3, $2, $4
+; MIPS32O0-NEXT:    sc $3, 0($1)
+; MIPS32O0-NEXT:    beqz $3, $BB2_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    move $2, $25
 ; MIPS32O0-NEXT:    jr $ra
 ; MIPS32O0-NEXT:    nop
 ;
@@ -630,18 +627,18 @@ define i32 @AtomicLoadXor32(i32 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32R6O0-NEXT:  $BB2_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $1, 0($2)
-; MIPS32R6O0-NEXT:    xor $3, $1, $4
-; MIPS32R6O0-NEXT:    sc $3, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $3, $BB2_1
+; MIPS32R6O0-NEXT:    ll $3, 0($1)
+; MIPS32R6O0-NEXT:    xor $5, $3, $4
+; MIPS32R6O0-NEXT:    sc $5, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $5, $BB2_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    move $2, $1
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    move $2, $3
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;
@@ -838,17 +835,16 @@ define i32 @AtomicLoadOr32(i32 signext %incr) nounwind {
 ; MIPS32O0:       # %bb.0: # %entry
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32O0-NEXT:  $BB3_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $25, 0($2)
-; MIPS32O0-NEXT:    or $1, $25, $4
-; MIPS32O0-NEXT:    sc $1, 0($2)
-; MIPS32O0-NEXT:    beqz $1, $BB3_1
+; MIPS32O0-NEXT:    ll $2, 0($1)
+; MIPS32O0-NEXT:    or $3, $2, $4
+; MIPS32O0-NEXT:    sc $3, 0($1)
+; MIPS32O0-NEXT:    beqz $3, $BB3_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    move $2, $25
 ; MIPS32O0-NEXT:    jr $ra
 ; MIPS32O0-NEXT:    nop
 ;
@@ -890,18 +886,18 @@ define i32 @AtomicLoadOr32(i32 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32R6O0-NEXT:  $BB3_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $1, 0($2)
-; MIPS32R6O0-NEXT:    or $3, $1, $4
-; MIPS32R6O0-NEXT:    sc $3, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $3, $BB3_1
+; MIPS32R6O0-NEXT:    ll $3, 0($1)
+; MIPS32R6O0-NEXT:    or $5, $3, $4
+; MIPS32R6O0-NEXT:    sc $5, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $5, $BB3_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    move $2, $1
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    move $2, $3
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;
@@ -1098,17 +1094,16 @@ define i32 @AtomicLoadAnd32(i32 signext %incr) nounwind {
 ; MIPS32O0:       # %bb.0: # %entry
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32O0-NEXT:  $BB4_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $25, 0($2)
-; MIPS32O0-NEXT:    and $1, $25, $4
-; MIPS32O0-NEXT:    sc $1, 0($2)
-; MIPS32O0-NEXT:    beqz $1, $BB4_1
+; MIPS32O0-NEXT:    ll $2, 0($1)
+; MIPS32O0-NEXT:    and $3, $2, $4
+; MIPS32O0-NEXT:    sc $3, 0($1)
+; MIPS32O0-NEXT:    beqz $3, $BB4_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    move $2, $25
 ; MIPS32O0-NEXT:    jr $ra
 ; MIPS32O0-NEXT:    nop
 ;
@@ -1150,18 +1145,18 @@ define i32 @AtomicLoadAnd32(i32 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32R6O0-NEXT:  $BB4_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $1, 0($2)
-; MIPS32R6O0-NEXT:    and $3, $1, $4
-; MIPS32R6O0-NEXT:    sc $3, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $3, $BB4_1
+; MIPS32R6O0-NEXT:    ll $3, 0($1)
+; MIPS32R6O0-NEXT:    and $5, $3, $4
+; MIPS32R6O0-NEXT:    sc $5, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $5, $BB4_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    move $2, $1
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    move $2, $3
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;
@@ -1359,18 +1354,17 @@ define i32 @AtomicLoadNand32(i32 signext %incr) nounwind {
 ; MIPS32O0:       # %bb.0: # %entry
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32O0-NEXT:  $BB5_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $25, 0($2)
-; MIPS32O0-NEXT:    and $1, $25, $4
-; MIPS32O0-NEXT:    nor $1, $zero, $1
-; MIPS32O0-NEXT:    sc $1, 0($2)
-; MIPS32O0-NEXT:    beqz $1, $BB5_1
+; MIPS32O0-NEXT:    ll $2, 0($1)
+; MIPS32O0-NEXT:    and $3, $2, $4
+; MIPS32O0-NEXT:    nor $3, $zero, $3
+; MIPS32O0-NEXT:    sc $3, 0($1)
+; MIPS32O0-NEXT:    beqz $3, $BB5_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    move $2, $25
 ; MIPS32O0-NEXT:    jr $ra
 ; MIPS32O0-NEXT:    nop
 ;
@@ -1414,19 +1408,19 @@ define i32 @AtomicLoadNand32(i32 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32R6O0-NEXT:  $BB5_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $1, 0($2)
-; MIPS32R6O0-NEXT:    and $3, $1, $4
-; MIPS32R6O0-NEXT:    nor $3, $zero, $3
-; MIPS32R6O0-NEXT:    sc $3, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $3, $BB5_1
+; MIPS32R6O0-NEXT:    ll $3, 0($1)
+; MIPS32R6O0-NEXT:    and $5, $3, $4
+; MIPS32R6O0-NEXT:    nor $5, $zero, $5
+; MIPS32R6O0-NEXT:    sc $5, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $5, $BB5_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    move $2, $1
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    move $2, $3
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;
@@ -1637,19 +1631,19 @@ define i32 @AtomicSwap32(i32 signext %newval) nounwind {
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32O0-NEXT:    addu $2, $2, $25
+; MIPS32O0-NEXT:    addu $1, $2, $25
 ; MIPS32O0-NEXT:    sw $4, 4($sp)
-; MIPS32O0-NEXT:    lw $4, 4($sp)
-; MIPS32O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32O0-NEXT:    lw $2, 4($sp)
+; MIPS32O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32O0-NEXT:  $BB6_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $25, 0($2)
-; MIPS32O0-NEXT:    move $1, $4
-; MIPS32O0-NEXT:    sc $1, 0($2)
-; MIPS32O0-NEXT:    beqz $1, $BB6_1
+; MIPS32O0-NEXT:    ll $3, 0($1)
+; MIPS32O0-NEXT:    move $4, $2
+; MIPS32O0-NEXT:    sc $4, 0($1)
+; MIPS32O0-NEXT:    beqz $4, $BB6_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    move $2, $25
+; MIPS32O0-NEXT:    move $2, $3
 ; MIPS32O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32O0-NEXT:    jr $ra
 ; MIPS32O0-NEXT:    nop
@@ -1697,20 +1691,20 @@ define i32 @AtomicSwap32(i32 signext %newval) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
 ; MIPS32R6O0-NEXT:    sw $4, 4($sp)
-; MIPS32R6O0-NEXT:    lw $4, 4($sp)
-; MIPS32R6O0-NEXT:    lw $2, %got(x)($2)
+; MIPS32R6O0-NEXT:    lw $3, 4($sp)
+; MIPS32R6O0-NEXT:    lw $1, %got(x)($1)
 ; MIPS32R6O0-NEXT:  $BB6_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $1, 0($2)
-; MIPS32R6O0-NEXT:    move $3, $4
-; MIPS32R6O0-NEXT:    sc $3, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $3, $BB6_1
+; MIPS32R6O0-NEXT:    ll $4, 0($1)
+; MIPS32R6O0-NEXT:    move $5, $3
+; MIPS32R6O0-NEXT:    sc $5, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $5, $BB6_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    move $2, $1
-; MIPS32R6O0-NEXT:    sw $25, 0($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 0($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    move $2, $4
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;
@@ -1942,29 +1936,29 @@ define i32 @AtomicCmpSwap32(i32 signext %oldval, i32 signext %newval) nounwind {
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -16
-; MIPS32O0-NEXT:    addu $2, $2, $25
+; MIPS32O0-NEXT:    addu $1, $2, $25
 ; MIPS32O0-NEXT:    sw $5, 12($sp)
-; MIPS32O0-NEXT:    lw $5, 12($sp)
-; MIPS32O0-NEXT:    lw $2, %got(x)($2)
-; MIPS32O0-NEXT:    lw $25, 8($sp) # 4-byte Folded Reload
-; MIPS32O0-NEXT:    move $1, $4
+; MIPS32O0-NEXT:    lw $2, 12($sp)
+; MIPS32O0-NEXT:    lw $1, %got(x)($1)
+; MIPS32O0-NEXT:    lw $3, 8($sp) # 4-byte Folded Reload
+; MIPS32O0-NEXT:    move $5, $4
 ; MIPS32O0-NEXT:  $BB7_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $3, 0($2)
-; MIPS32O0-NEXT:    bne $3, $1, $BB7_3
+; MIPS32O0-NEXT:    ll $6, 0($1)
+; MIPS32O0-NEXT:    bne $6, $5, $BB7_3
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
 ; MIPS32O0-NEXT:    # in Loop: Header=BB7_1 Depth=1
-; MIPS32O0-NEXT:    move $6, $5
-; MIPS32O0-NEXT:    sc $6, 0($2)
-; MIPS32O0-NEXT:    beqz $6, $BB7_1
+; MIPS32O0-NEXT:    move $7, $2
+; MIPS32O0-NEXT:    sc $7, 0($1)
+; MIPS32O0-NEXT:    beqz $7, $BB7_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  $BB7_3: # %entry
-; MIPS32O0-NEXT:    xor $1, $3, $4
+; MIPS32O0-NEXT:    xor $1, $6, $4
 ; MIPS32O0-NEXT:    sltiu $1, $1, 1
-; MIPS32O0-NEXT:    move $2, $3
-; MIPS32O0-NEXT:    sw $3, 8($sp) # 4-byte Folded Spill
-; MIPS32O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    move $2, $6
+; MIPS32O0-NEXT:    sw $6, 8($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    sw $3, 4($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:    sw $1, 0($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:    addiu $sp, $sp, 16
 ; MIPS32O0-NEXT:    jr $ra
@@ -2020,28 +2014,28 @@ define i32 @AtomicCmpSwap32(i32 signext %oldval, i32 signext %newval) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -24
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $5
-; MIPS32R6O0-NEXT:    move $1, $4
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $5
+; MIPS32R6O0-NEXT:    move $3, $4
 ; MIPS32R6O0-NEXT:    sw $5, 20($sp)
 ; MIPS32R6O0-NEXT:    lw $5, 20($sp)
-; MIPS32R6O0-NEXT:    lw $2, %got(x)($2)
-; MIPS32R6O0-NEXT:    lw $3, 16($sp) # 4-byte Folded Reload
+; MIPS32R6O0-NEXT:    lw $1, %got(x)($1)
+; MIPS32R6O0-NEXT:    lw $6, 16($sp) # 4-byte Folded Reload
 ; MIPS32R6O0-NEXT:  $BB7_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $6, 0($2)
-; MIPS32R6O0-NEXT:    bnec $6, $4, $BB7_3
+; MIPS32R6O0-NEXT:    ll $7, 0($1)
+; MIPS32R6O0-NEXT:    bnec $7, $4, $BB7_3
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
 ; MIPS32R6O0-NEXT:    # in Loop: Header=BB7_1 Depth=1
-; MIPS32R6O0-NEXT:    move $7, $5
-; MIPS32R6O0-NEXT:    sc $7, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $7, $BB7_1
+; MIPS32R6O0-NEXT:    move $8, $5
+; MIPS32R6O0-NEXT:    sc $8, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $8, $BB7_1
 ; MIPS32R6O0-NEXT:  $BB7_3: # %entry
-; MIPS32R6O0-NEXT:    move $2, $6
-; MIPS32R6O0-NEXT:    sw $25, 12($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $1, 8($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $6, 16($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $3, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 12($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    move $2, $7
+; MIPS32R6O0-NEXT:    sw $3, 8($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $7, 16($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $6, 4($sp) # 4-byte Folded Spill
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 24
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;
@@ -2329,33 +2323,33 @@ define signext i8 @AtomicLoadAdd8(i8 signext %incr) nounwind {
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(y)($2)
-; MIPS32O0-NEXT:    addiu $25, $zero, -4
-; MIPS32O0-NEXT:    and $25, $2, $25
-; MIPS32O0-NEXT:    andi $2, $2, 3
-; MIPS32O0-NEXT:    sll $2, $2, 3
-; MIPS32O0-NEXT:    ori $1, $zero, 255
-; MIPS32O0-NEXT:    sllv $1, $1, $2
-; MIPS32O0-NEXT:    nor $3, $zero, $1
-; MIPS32O0-NEXT:    sllv $4, $4, $2
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(y)($1)
+; MIPS32O0-NEXT:    addiu $2, $zero, -4
+; MIPS32O0-NEXT:    and $2, $1, $2
+; MIPS32O0-NEXT:    andi $1, $1, 3
+; MIPS32O0-NEXT:    sll $1, $1, 3
+; MIPS32O0-NEXT:    ori $3, $zero, 255
+; MIPS32O0-NEXT:    sllv $3, $3, $1
+; MIPS32O0-NEXT:    nor $5, $zero, $3
+; MIPS32O0-NEXT:    sllv $4, $4, $1
 ; MIPS32O0-NEXT:  $BB8_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $6, 0($25)
-; MIPS32O0-NEXT:    addu $7, $6, $4
-; MIPS32O0-NEXT:    and $7, $7, $1
-; MIPS32O0-NEXT:    and $8, $6, $3
-; MIPS32O0-NEXT:    or $8, $8, $7
-; MIPS32O0-NEXT:    sc $8, 0($25)
-; MIPS32O0-NEXT:    beqz $8, $BB8_1
+; MIPS32O0-NEXT:    ll $7, 0($2)
+; MIPS32O0-NEXT:    addu $8, $7, $4
+; MIPS32O0-NEXT:    and $8, $8, $3
+; MIPS32O0-NEXT:    and $9, $7, $5
+; MIPS32O0-NEXT:    or $9, $9, $8
+; MIPS32O0-NEXT:    sc $9, 0($2)
+; MIPS32O0-NEXT:    beqz $9, $BB8_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    and $5, $6, $1
-; MIPS32O0-NEXT:    srlv $5, $5, $2
-; MIPS32O0-NEXT:    sll $5, $5, 24
-; MIPS32O0-NEXT:    sra $5, $5, 24
+; MIPS32O0-NEXT:    and $6, $7, $3
+; MIPS32O0-NEXT:    srlv $6, $6, $1
+; MIPS32O0-NEXT:    sll $6, $6, 24
+; MIPS32O0-NEXT:    sra $6, $6, 24
 ; MIPS32O0-NEXT:  # %bb.3: # %entry
-; MIPS32O0-NEXT:    sw $5, 4($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    sw $6, 4($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:  # %bb.4: # %entry
 ; MIPS32O0-NEXT:    lw $1, 4($sp) # 4-byte Folded Reload
 ; MIPS32O0-NEXT:    sll $2, $1, 24
@@ -2432,33 +2426,33 @@ define signext i8 @AtomicLoadAdd8(i8 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(y)($2)
-; MIPS32R6O0-NEXT:    addiu $1, $zero, -4
-; MIPS32R6O0-NEXT:    and $1, $2, $1
-; MIPS32R6O0-NEXT:    andi $2, $2, 3
-; MIPS32R6O0-NEXT:    sll $2, $2, 3
-; MIPS32R6O0-NEXT:    ori $3, $zero, 255
-; MIPS32R6O0-NEXT:    sllv $3, $3, $2
-; MIPS32R6O0-NEXT:    nor $5, $zero, $3
-; MIPS32R6O0-NEXT:    sllv $4, $4, $2
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(y)($1)
+; MIPS32R6O0-NEXT:    addiu $3, $zero, -4
+; MIPS32R6O0-NEXT:    and $3, $1, $3
+; MIPS32R6O0-NEXT:    andi $1, $1, 3
+; MIPS32R6O0-NEXT:    sll $1, $1, 3
+; MIPS32R6O0-NEXT:    ori $5, $zero, 255
+; MIPS32R6O0-NEXT:    sllv $5, $5, $1
+; MIPS32R6O0-NEXT:    nor $6, $zero, $5
+; MIPS32R6O0-NEXT:    sllv $4, $4, $1
 ; MIPS32R6O0-NEXT:  $BB8_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $7, 0($1)
-; MIPS32R6O0-NEXT:    addu $8, $7, $4
-; MIPS32R6O0-NEXT:    and $8, $8, $3
-; MIPS32R6O0-NEXT:    and $9, $7, $5
-; MIPS32R6O0-NEXT:    or $9, $9, $8
-; MIPS32R6O0-NEXT:    sc $9, 0($1)
-; MIPS32R6O0-NEXT:    beqzc $9, $BB8_1
+; MIPS32R6O0-NEXT:    ll $8, 0($3)
+; MIPS32R6O0-NEXT:    addu $9, $8, $4
+; MIPS32R6O0-NEXT:    and $9, $9, $5
+; MIPS32R6O0-NEXT:    and $10, $8, $6
+; MIPS32R6O0-NEXT:    or $10, $10, $9
+; MIPS32R6O0-NEXT:    sc $10, 0($3)
+; MIPS32R6O0-NEXT:    beqzc $10, $BB8_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    and $6, $7, $3
-; MIPS32R6O0-NEXT:    srlv $6, $6, $2
-; MIPS32R6O0-NEXT:    seb $6, $6
+; MIPS32R6O0-NEXT:    and $7, $8, $5
+; MIPS32R6O0-NEXT:    srlv $7, $7, $1
+; MIPS32R6O0-NEXT:    seb $7, $7
 ; MIPS32R6O0-NEXT:  # %bb.3: # %entry
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $6, 0($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $7, 0($sp) # 4-byte Folded Spill
 ; MIPS32R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS32R6O0-NEXT:    lw $1, 0($sp) # 4-byte Folded Reload
 ; MIPS32R6O0-NEXT:    seb $2, $1
@@ -2604,30 +2598,30 @@ define signext i8 @AtomicLoadAdd8(i8 signext %incr) nounwind {
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicLoadAdd8)))
 ; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(y)($1)
-; MIPS64R6O0-NEXT:    daddiu $4, $zero, -4
-; MIPS64R6O0-NEXT:    and $4, $1, $4
-; MIPS64R6O0-NEXT:    andi $3, $1, 3
-; MIPS64R6O0-NEXT:    xori $3, $3, 3
-; MIPS64R6O0-NEXT:    sll $3, $3, 3
-; MIPS64R6O0-NEXT:    ori $5, $zero, 255
-; MIPS64R6O0-NEXT:    sllv $5, $5, $3
-; MIPS64R6O0-NEXT:    nor $6, $zero, $5
-; MIPS64R6O0-NEXT:    sllv $2, $2, $3
+; MIPS64R6O0-NEXT:    daddiu $3, $zero, -4
+; MIPS64R6O0-NEXT:    and $3, $1, $3
+; MIPS64R6O0-NEXT:    andi $5, $1, 3
+; MIPS64R6O0-NEXT:    xori $5, $5, 3
+; MIPS64R6O0-NEXT:    sll $5, $5, 3
+; MIPS64R6O0-NEXT:    ori $6, $zero, 255
+; MIPS64R6O0-NEXT:    sllv $6, $6, $5
+; MIPS64R6O0-NEXT:    nor $7, $zero, $6
+; MIPS64R6O0-NEXT:    sllv $2, $2, $5
 ; MIPS64R6O0-NEXT:  .LBB8_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    ll $8, 0($4)
-; MIPS64R6O0-NEXT:    addu $9, $8, $2
-; MIPS64R6O0-NEXT:    and $9, $9, $5
-; MIPS64R6O0-NEXT:    and $10, $8, $6
-; MIPS64R6O0-NEXT:    or $10, $10, $9
-; MIPS64R6O0-NEXT:    sc $10, 0($4)
-; MIPS64R6O0-NEXT:    beqzc $10, .LBB8_1
+; MIPS64R6O0-NEXT:    ll $9, 0($3)
+; MIPS64R6O0-NEXT:    addu $10, $9, $2
+; MIPS64R6O0-NEXT:    and $10, $10, $6
+; MIPS64R6O0-NEXT:    and $11, $9, $7
+; MIPS64R6O0-NEXT:    or $11, $11, $10
+; MIPS64R6O0-NEXT:    sc $11, 0($3)
+; MIPS64R6O0-NEXT:    beqzc $11, .LBB8_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    and $7, $8, $5
-; MIPS64R6O0-NEXT:    srlv $7, $7, $3
-; MIPS64R6O0-NEXT:    seb $7, $7
+; MIPS64R6O0-NEXT:    and $8, $9, $6
+; MIPS64R6O0-NEXT:    srlv $8, $8, $5
+; MIPS64R6O0-NEXT:    seb $8, $8
 ; MIPS64R6O0-NEXT:  # %bb.3: # %entry
-; MIPS64R6O0-NEXT:    sw $7, 12($sp) # 4-byte Folded Spill
+; MIPS64R6O0-NEXT:    sw $8, 12($sp) # 4-byte Folded Spill
 ; MIPS64R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS64R6O0-NEXT:    lw $1, 12($sp) # 4-byte Folded Reload
 ; MIPS64R6O0-NEXT:    seb $2, $1
@@ -2846,33 +2840,33 @@ define signext i8 @AtomicLoadSub8(i8 signext %incr) nounwind {
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(y)($2)
-; MIPS32O0-NEXT:    addiu $25, $zero, -4
-; MIPS32O0-NEXT:    and $25, $2, $25
-; MIPS32O0-NEXT:    andi $2, $2, 3
-; MIPS32O0-NEXT:    sll $2, $2, 3
-; MIPS32O0-NEXT:    ori $1, $zero, 255
-; MIPS32O0-NEXT:    sllv $1, $1, $2
-; MIPS32O0-NEXT:    nor $3, $zero, $1
-; MIPS32O0-NEXT:    sllv $4, $4, $2
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(y)($1)
+; MIPS32O0-NEXT:    addiu $2, $zero, -4
+; MIPS32O0-NEXT:    and $2, $1, $2
+; MIPS32O0-NEXT:    andi $1, $1, 3
+; MIPS32O0-NEXT:    sll $1, $1, 3
+; MIPS32O0-NEXT:    ori $3, $zero, 255
+; MIPS32O0-NEXT:    sllv $3, $3, $1
+; MIPS32O0-NEXT:    nor $5, $zero, $3
+; MIPS32O0-NEXT:    sllv $4, $4, $1
 ; MIPS32O0-NEXT:  $BB9_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $6, 0($25)
-; MIPS32O0-NEXT:    subu $7, $6, $4
-; MIPS32O0-NEXT:    and $7, $7, $1
-; MIPS32O0-NEXT:    and $8, $6, $3
-; MIPS32O0-NEXT:    or $8, $8, $7
-; MIPS32O0-NEXT:    sc $8, 0($25)
-; MIPS32O0-NEXT:    beqz $8, $BB9_1
+; MIPS32O0-NEXT:    ll $7, 0($2)
+; MIPS32O0-NEXT:    subu $8, $7, $4
+; MIPS32O0-NEXT:    and $8, $8, $3
+; MIPS32O0-NEXT:    and $9, $7, $5
+; MIPS32O0-NEXT:    or $9, $9, $8
+; MIPS32O0-NEXT:    sc $9, 0($2)
+; MIPS32O0-NEXT:    beqz $9, $BB9_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    and $5, $6, $1
-; MIPS32O0-NEXT:    srlv $5, $5, $2
-; MIPS32O0-NEXT:    sll $5, $5, 24
-; MIPS32O0-NEXT:    sra $5, $5, 24
+; MIPS32O0-NEXT:    and $6, $7, $3
+; MIPS32O0-NEXT:    srlv $6, $6, $1
+; MIPS32O0-NEXT:    sll $6, $6, 24
+; MIPS32O0-NEXT:    sra $6, $6, 24
 ; MIPS32O0-NEXT:  # %bb.3: # %entry
-; MIPS32O0-NEXT:    sw $5, 4($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    sw $6, 4($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:  # %bb.4: # %entry
 ; MIPS32O0-NEXT:    lw $1, 4($sp) # 4-byte Folded Reload
 ; MIPS32O0-NEXT:    sll $2, $1, 24
@@ -2949,33 +2943,33 @@ define signext i8 @AtomicLoadSub8(i8 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(y)($2)
-; MIPS32R6O0-NEXT:    addiu $1, $zero, -4
-; MIPS32R6O0-NEXT:    and $1, $2, $1
-; MIPS32R6O0-NEXT:    andi $2, $2, 3
-; MIPS32R6O0-NEXT:    sll $2, $2, 3
-; MIPS32R6O0-NEXT:    ori $3, $zero, 255
-; MIPS32R6O0-NEXT:    sllv $3, $3, $2
-; MIPS32R6O0-NEXT:    nor $5, $zero, $3
-; MIPS32R6O0-NEXT:    sllv $4, $4, $2
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(y)($1)
+; MIPS32R6O0-NEXT:    addiu $3, $zero, -4
+; MIPS32R6O0-NEXT:    and $3, $1, $3
+; MIPS32R6O0-NEXT:    andi $1, $1, 3
+; MIPS32R6O0-NEXT:    sll $1, $1, 3
+; MIPS32R6O0-NEXT:    ori $5, $zero, 255
+; MIPS32R6O0-NEXT:    sllv $5, $5, $1
+; MIPS32R6O0-NEXT:    nor $6, $zero, $5
+; MIPS32R6O0-NEXT:    sllv $4, $4, $1
 ; MIPS32R6O0-NEXT:  $BB9_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $7, 0($1)
-; MIPS32R6O0-NEXT:    subu $8, $7, $4
-; MIPS32R6O0-NEXT:    and $8, $8, $3
-; MIPS32R6O0-NEXT:    and $9, $7, $5
-; MIPS32R6O0-NEXT:    or $9, $9, $8
-; MIPS32R6O0-NEXT:    sc $9, 0($1)
-; MIPS32R6O0-NEXT:    beqzc $9, $BB9_1
+; MIPS32R6O0-NEXT:    ll $8, 0($3)
+; MIPS32R6O0-NEXT:    subu $9, $8, $4
+; MIPS32R6O0-NEXT:    and $9, $9, $5
+; MIPS32R6O0-NEXT:    and $10, $8, $6
+; MIPS32R6O0-NEXT:    or $10, $10, $9
+; MIPS32R6O0-NEXT:    sc $10, 0($3)
+; MIPS32R6O0-NEXT:    beqzc $10, $BB9_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    and $6, $7, $3
-; MIPS32R6O0-NEXT:    srlv $6, $6, $2
-; MIPS32R6O0-NEXT:    seb $6, $6
+; MIPS32R6O0-NEXT:    and $7, $8, $5
+; MIPS32R6O0-NEXT:    srlv $7, $7, $1
+; MIPS32R6O0-NEXT:    seb $7, $7
 ; MIPS32R6O0-NEXT:  # %bb.3: # %entry
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $6, 0($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $7, 0($sp) # 4-byte Folded Spill
 ; MIPS32R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS32R6O0-NEXT:    lw $1, 0($sp) # 4-byte Folded Reload
 ; MIPS32R6O0-NEXT:    seb $2, $1
@@ -3121,30 +3115,30 @@ define signext i8 @AtomicLoadSub8(i8 signext %incr) nounwind {
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicLoadSub8)))
 ; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(y)($1)
-; MIPS64R6O0-NEXT:    daddiu $4, $zero, -4
-; MIPS64R6O0-NEXT:    and $4, $1, $4
-; MIPS64R6O0-NEXT:    andi $3, $1, 3
-; MIPS64R6O0-NEXT:    xori $3, $3, 3
-; MIPS64R6O0-NEXT:    sll $3, $3, 3
-; MIPS64R6O0-NEXT:    ori $5, $zero, 255
-; MIPS64R6O0-NEXT:    sllv $5, $5, $3
-; MIPS64R6O0-NEXT:    nor $6, $zero, $5
-; MIPS64R6O0-NEXT:    sllv $2, $2, $3
+; MIPS64R6O0-NEXT:    daddiu $3, $zero, -4
+; MIPS64R6O0-NEXT:    and $3, $1, $3
+; MIPS64R6O0-NEXT:    andi $5, $1, 3
+; MIPS64R6O0-NEXT:    xori $5, $5, 3
+; MIPS64R6O0-NEXT:    sll $5, $5, 3
+; MIPS64R6O0-NEXT:    ori $6, $zero, 255
+; MIPS64R6O0-NEXT:    sllv $6, $6, $5
+; MIPS64R6O0-NEXT:    nor $7, $zero, $6
+; MIPS64R6O0-NEXT:    sllv $2, $2, $5
 ; MIPS64R6O0-NEXT:  .LBB9_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    ll $8, 0($4)
-; MIPS64R6O0-NEXT:    subu $9, $8, $2
-; MIPS64R6O0-NEXT:    and $9, $9, $5
-; MIPS64R6O0-NEXT:    and $10, $8, $6
-; MIPS64R6O0-NEXT:    or $10, $10, $9
-; MIPS64R6O0-NEXT:    sc $10, 0($4)
-; MIPS64R6O0-NEXT:    beqzc $10, .LBB9_1
+; MIPS64R6O0-NEXT:    ll $9, 0($3)
+; MIPS64R6O0-NEXT:    subu $10, $9, $2
+; MIPS64R6O0-NEXT:    and $10, $10, $6
+; MIPS64R6O0-NEXT:    and $11, $9, $7
+; MIPS64R6O0-NEXT:    or $11, $11, $10
+; MIPS64R6O0-NEXT:    sc $11, 0($3)
+; MIPS64R6O0-NEXT:    beqzc $11, .LBB9_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    and $7, $8, $5
-; MIPS64R6O0-NEXT:    srlv $7, $7, $3
-; MIPS64R6O0-NEXT:    seb $7, $7
+; MIPS64R6O0-NEXT:    and $8, $9, $6
+; MIPS64R6O0-NEXT:    srlv $8, $8, $5
+; MIPS64R6O0-NEXT:    seb $8, $8
 ; MIPS64R6O0-NEXT:  # %bb.3: # %entry
-; MIPS64R6O0-NEXT:    sw $7, 12($sp) # 4-byte Folded Spill
+; MIPS64R6O0-NEXT:    sw $8, 12($sp) # 4-byte Folded Spill
 ; MIPS64R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS64R6O0-NEXT:    lw $1, 12($sp) # 4-byte Folded Reload
 ; MIPS64R6O0-NEXT:    seb $2, $1
@@ -3365,34 +3359,34 @@ define signext i8 @AtomicLoadNand8(i8 signext %incr) nounwind {
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(y)($2)
-; MIPS32O0-NEXT:    addiu $25, $zero, -4
-; MIPS32O0-NEXT:    and $25, $2, $25
-; MIPS32O0-NEXT:    andi $2, $2, 3
-; MIPS32O0-NEXT:    sll $2, $2, 3
-; MIPS32O0-NEXT:    ori $1, $zero, 255
-; MIPS32O0-NEXT:    sllv $1, $1, $2
-; MIPS32O0-NEXT:    nor $3, $zero, $1
-; MIPS32O0-NEXT:    sllv $4, $4, $2
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(y)($1)
+; MIPS32O0-NEXT:    addiu $2, $zero, -4
+; MIPS32O0-NEXT:    and $2, $1, $2
+; MIPS32O0-NEXT:    andi $1, $1, 3
+; MIPS32O0-NEXT:    sll $1, $1, 3
+; MIPS32O0-NEXT:    ori $3, $zero, 255
+; MIPS32O0-NEXT:    sllv $3, $3, $1
+; MIPS32O0-NEXT:    nor $5, $zero, $3
+; MIPS32O0-NEXT:    sllv $4, $4, $1
 ; MIPS32O0-NEXT:  $BB10_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $6, 0($25)
-; MIPS32O0-NEXT:    and $7, $6, $4
-; MIPS32O0-NEXT:    nor $7, $zero, $7
-; MIPS32O0-NEXT:    and $7, $7, $1
-; MIPS32O0-NEXT:    and $8, $6, $3
-; MIPS32O0-NEXT:    or $8, $8, $7
-; MIPS32O0-NEXT:    sc $8, 0($25)
-; MIPS32O0-NEXT:    beqz $8, $BB10_1
+; MIPS32O0-NEXT:    ll $7, 0($2)
+; MIPS32O0-NEXT:    and $8, $7, $4
+; MIPS32O0-NEXT:    nor $8, $zero, $8
+; MIPS32O0-NEXT:    and $8, $8, $3
+; MIPS32O0-NEXT:    and $9, $7, $5
+; MIPS32O0-NEXT:    or $9, $9, $8
+; MIPS32O0-NEXT:    sc $9, 0($2)
+; MIPS32O0-NEXT:    beqz $9, $BB10_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    and $5, $6, $1
-; MIPS32O0-NEXT:    srlv $5, $5, $2
-; MIPS32O0-NEXT:    sll $5, $5, 24
-; MIPS32O0-NEXT:    sra $5, $5, 24
+; MIPS32O0-NEXT:    and $6, $7, $3
+; MIPS32O0-NEXT:    srlv $6, $6, $1
+; MIPS32O0-NEXT:    sll $6, $6, 24
+; MIPS32O0-NEXT:    sra $6, $6, 24
 ; MIPS32O0-NEXT:  # %bb.3: # %entry
-; MIPS32O0-NEXT:    sw $5, 4($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    sw $6, 4($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:  # %bb.4: # %entry
 ; MIPS32O0-NEXT:    lw $1, 4($sp) # 4-byte Folded Reload
 ; MIPS32O0-NEXT:    sll $2, $1, 24
@@ -3471,34 +3465,34 @@ define signext i8 @AtomicLoadNand8(i8 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(y)($2)
-; MIPS32R6O0-NEXT:    addiu $1, $zero, -4
-; MIPS32R6O0-NEXT:    and $1, $2, $1
-; MIPS32R6O0-NEXT:    andi $2, $2, 3
-; MIPS32R6O0-NEXT:    sll $2, $2, 3
-; MIPS32R6O0-NEXT:    ori $3, $zero, 255
-; MIPS32R6O0-NEXT:    sllv $3, $3, $2
-; MIPS32R6O0-NEXT:    nor $5, $zero, $3
-; MIPS32R6O0-NEXT:    sllv $4, $4, $2
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(y)($1)
+; MIPS32R6O0-NEXT:    addiu $3, $zero, -4
+; MIPS32R6O0-NEXT:    and $3, $1, $3
+; MIPS32R6O0-NEXT:    andi $1, $1, 3
+; MIPS32R6O0-NEXT:    sll $1, $1, 3
+; MIPS32R6O0-NEXT:    ori $5, $zero, 255
+; MIPS32R6O0-NEXT:    sllv $5, $5, $1
+; MIPS32R6O0-NEXT:    nor $6, $zero, $5
+; MIPS32R6O0-NEXT:    sllv $4, $4, $1
 ; MIPS32R6O0-NEXT:  $BB10_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $7, 0($1)
-; MIPS32R6O0-NEXT:    and $8, $7, $4
-; MIPS32R6O0-NEXT:    nor $8, $zero, $8
-; MIPS32R6O0-NEXT:    and $8, $8, $3
-; MIPS32R6O0-NEXT:    and $9, $7, $5
-; MIPS32R6O0-NEXT:    or $9, $9, $8
-; MIPS32R6O0-NEXT:    sc $9, 0($1)
-; MIPS32R6O0-NEXT:    beqzc $9, $BB10_1
+; MIPS32R6O0-NEXT:    ll $8, 0($3)
+; MIPS32R6O0-NEXT:    and $9, $8, $4
+; MIPS32R6O0-NEXT:    nor $9, $zero, $9
+; MIPS32R6O0-NEXT:    and $9, $9, $5
+; MIPS32R6O0-NEXT:    and $10, $8, $6
+; MIPS32R6O0-NEXT:    or $10, $10, $9
+; MIPS32R6O0-NEXT:    sc $10, 0($3)
+; MIPS32R6O0-NEXT:    beqzc $10, $BB10_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    and $6, $7, $3
-; MIPS32R6O0-NEXT:    srlv $6, $6, $2
-; MIPS32R6O0-NEXT:    seb $6, $6
+; MIPS32R6O0-NEXT:    and $7, $8, $5
+; MIPS32R6O0-NEXT:    srlv $7, $7, $1
+; MIPS32R6O0-NEXT:    seb $7, $7
 ; MIPS32R6O0-NEXT:  # %bb.3: # %entry
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $6, 0($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $7, 0($sp) # 4-byte Folded Spill
 ; MIPS32R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS32R6O0-NEXT:    lw $1, 0($sp) # 4-byte Folded Reload
 ; MIPS32R6O0-NEXT:    seb $2, $1
@@ -3648,31 +3642,31 @@ define signext i8 @AtomicLoadNand8(i8 signext %incr) nounwind {
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicLoadNand8)))
 ; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(y)($1)
-; MIPS64R6O0-NEXT:    daddiu $4, $zero, -4
-; MIPS64R6O0-NEXT:    and $4, $1, $4
-; MIPS64R6O0-NEXT:    andi $3, $1, 3
-; MIPS64R6O0-NEXT:    xori $3, $3, 3
-; MIPS64R6O0-NEXT:    sll $3, $3, 3
-; MIPS64R6O0-NEXT:    ori $5, $zero, 255
-; MIPS64R6O0-NEXT:    sllv $5, $5, $3
-; MIPS64R6O0-NEXT:    nor $6, $zero, $5
-; MIPS64R6O0-NEXT:    sllv $2, $2, $3
+; MIPS64R6O0-NEXT:    daddiu $3, $zero, -4
+; MIPS64R6O0-NEXT:    and $3, $1, $3
+; MIPS64R6O0-NEXT:    andi $5, $1, 3
+; MIPS64R6O0-NEXT:    xori $5, $5, 3
+; MIPS64R6O0-NEXT:    sll $5, $5, 3
+; MIPS64R6O0-NEXT:    ori $6, $zero, 255
+; MIPS64R6O0-NEXT:    sllv $6, $6, $5
+; MIPS64R6O0-NEXT:    nor $7, $zero, $6
+; MIPS64R6O0-NEXT:    sllv $2, $2, $5
 ; MIPS64R6O0-NEXT:  .LBB10_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    ll $8, 0($4)
-; MIPS64R6O0-NEXT:    and $9, $8, $2
-; MIPS64R6O0-NEXT:    nor $9, $zero, $9
-; MIPS64R6O0-NEXT:    and $9, $9, $5
-; MIPS64R6O0-NEXT:    and $10, $8, $6
-; MIPS64R6O0-NEXT:    or $10, $10, $9
-; MIPS64R6O0-NEXT:    sc $10, 0($4)
-; MIPS64R6O0-NEXT:    beqzc $10, .LBB10_1
+; MIPS64R6O0-NEXT:    ll $9, 0($3)
+; MIPS64R6O0-NEXT:    and $10, $9, $2
+; MIPS64R6O0-NEXT:    nor $10, $zero, $10
+; MIPS64R6O0-NEXT:    and $10, $10, $6
+; MIPS64R6O0-NEXT:    and $11, $9, $7
+; MIPS64R6O0-NEXT:    or $11, $11, $10
+; MIPS64R6O0-NEXT:    sc $11, 0($3)
+; MIPS64R6O0-NEXT:    beqzc $11, .LBB10_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    and $7, $8, $5
-; MIPS64R6O0-NEXT:    srlv $7, $7, $3
-; MIPS64R6O0-NEXT:    seb $7, $7
+; MIPS64R6O0-NEXT:    and $8, $9, $6
+; MIPS64R6O0-NEXT:    srlv $8, $8, $5
+; MIPS64R6O0-NEXT:    seb $8, $8
 ; MIPS64R6O0-NEXT:  # %bb.3: # %entry
-; MIPS64R6O0-NEXT:    sw $7, 12($sp) # 4-byte Folded Spill
+; MIPS64R6O0-NEXT:    sw $8, 12($sp) # 4-byte Folded Spill
 ; MIPS64R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS64R6O0-NEXT:    lw $1, 12($sp) # 4-byte Folded Reload
 ; MIPS64R6O0-NEXT:    seb $2, $1
@@ -3896,32 +3890,32 @@ define signext i8 @AtomicSwap8(i8 signext %newval) nounwind {
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(y)($2)
-; MIPS32O0-NEXT:    addiu $25, $zero, -4
-; MIPS32O0-NEXT:    and $25, $2, $25
-; MIPS32O0-NEXT:    andi $2, $2, 3
-; MIPS32O0-NEXT:    sll $2, $2, 3
-; MIPS32O0-NEXT:    ori $1, $zero, 255
-; MIPS32O0-NEXT:    sllv $1, $1, $2
-; MIPS32O0-NEXT:    nor $3, $zero, $1
-; MIPS32O0-NEXT:    sllv $4, $4, $2
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(y)($1)
+; MIPS32O0-NEXT:    addiu $2, $zero, -4
+; MIPS32O0-NEXT:    and $2, $1, $2
+; MIPS32O0-NEXT:    andi $1, $1, 3
+; MIPS32O0-NEXT:    sll $1, $1, 3
+; MIPS32O0-NEXT:    ori $3, $zero, 255
+; MIPS32O0-NEXT:    sllv $3, $3, $1
+; MIPS32O0-NEXT:    nor $5, $zero, $3
+; MIPS32O0-NEXT:    sllv $4, $4, $1
 ; MIPS32O0-NEXT:  $BB11_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $6, 0($25)
-; MIPS32O0-NEXT:    and $7, $4, $1
-; MIPS32O0-NEXT:    and $8, $6, $3
-; MIPS32O0-NEXT:    or $8, $8, $7
-; MIPS32O0-NEXT:    sc $8, 0($25)
-; MIPS32O0-NEXT:    beqz $8, $BB11_1
+; MIPS32O0-NEXT:    ll $7, 0($2)
+; MIPS32O0-NEXT:    and $8, $4, $3
+; MIPS32O0-NEXT:    and $9, $7, $5
+; MIPS32O0-NEXT:    or $9, $9, $8
+; MIPS32O0-NEXT:    sc $9, 0($2)
+; MIPS32O0-NEXT:    beqz $9, $BB11_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    and $5, $6, $1
-; MIPS32O0-NEXT:    srlv $5, $5, $2
-; MIPS32O0-NEXT:    sll $5, $5, 24
-; MIPS32O0-NEXT:    sra $5, $5, 24
+; MIPS32O0-NEXT:    and $6, $7, $3
+; MIPS32O0-NEXT:    srlv $6, $6, $1
+; MIPS32O0-NEXT:    sll $6, $6, 24
+; MIPS32O0-NEXT:    sra $6, $6, 24
 ; MIPS32O0-NEXT:  # %bb.3: # %entry
-; MIPS32O0-NEXT:    sw $5, 4($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    sw $6, 4($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:  # %bb.4: # %entry
 ; MIPS32O0-NEXT:    lw $1, 4($sp) # 4-byte Folded Reload
 ; MIPS32O0-NEXT:    sll $2, $1, 24
@@ -3996,32 +3990,32 @@ define signext i8 @AtomicSwap8(i8 signext %newval) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(y)($2)
-; MIPS32R6O0-NEXT:    addiu $1, $zero, -4
-; MIPS32R6O0-NEXT:    and $1, $2, $1
-; MIPS32R6O0-NEXT:    andi $2, $2, 3
-; MIPS32R6O0-NEXT:    sll $2, $2, 3
-; MIPS32R6O0-NEXT:    ori $3, $zero, 255
-; MIPS32R6O0-NEXT:    sllv $3, $3, $2
-; MIPS32R6O0-NEXT:    nor $5, $zero, $3
-; MIPS32R6O0-NEXT:    sllv $4, $4, $2
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(y)($1)
+; MIPS32R6O0-NEXT:    addiu $3, $zero, -4
+; MIPS32R6O0-NEXT:    and $3, $1, $3
+; MIPS32R6O0-NEXT:    andi $1, $1, 3
+; MIPS32R6O0-NEXT:    sll $1, $1, 3
+; MIPS32R6O0-NEXT:    ori $5, $zero, 255
+; MIPS32R6O0-NEXT:    sllv $5, $5, $1
+; MIPS32R6O0-NEXT:    nor $6, $zero, $5
+; MIPS32R6O0-NEXT:    sllv $4, $4, $1
 ; MIPS32R6O0-NEXT:  $BB11_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $7, 0($1)
-; MIPS32R6O0-NEXT:    and $8, $4, $3
-; MIPS32R6O0-NEXT:    and $9, $7, $5
-; MIPS32R6O0-NEXT:    or $9, $9, $8
-; MIPS32R6O0-NEXT:    sc $9, 0($1)
-; MIPS32R6O0-NEXT:    beqzc $9, $BB11_1
+; MIPS32R6O0-NEXT:    ll $8, 0($3)
+; MIPS32R6O0-NEXT:    and $9, $4, $5
+; MIPS32R6O0-NEXT:    and $10, $8, $6
+; MIPS32R6O0-NEXT:    or $10, $10, $9
+; MIPS32R6O0-NEXT:    sc $10, 0($3)
+; MIPS32R6O0-NEXT:    beqzc $10, $BB11_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    and $6, $7, $3
-; MIPS32R6O0-NEXT:    srlv $6, $6, $2
-; MIPS32R6O0-NEXT:    seb $6, $6
+; MIPS32R6O0-NEXT:    and $7, $8, $5
+; MIPS32R6O0-NEXT:    srlv $7, $7, $1
+; MIPS32R6O0-NEXT:    seb $7, $7
 ; MIPS32R6O0-NEXT:  # %bb.3: # %entry
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $6, 0($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $7, 0($sp) # 4-byte Folded Spill
 ; MIPS32R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS32R6O0-NEXT:    lw $1, 0($sp) # 4-byte Folded Reload
 ; MIPS32R6O0-NEXT:    seb $2, $1
@@ -4163,29 +4157,29 @@ define signext i8 @AtomicSwap8(i8 signext %newval) nounwind {
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicSwap8)))
 ; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(y)($1)
-; MIPS64R6O0-NEXT:    daddiu $4, $zero, -4
-; MIPS64R6O0-NEXT:    and $4, $1, $4
-; MIPS64R6O0-NEXT:    andi $3, $1, 3
-; MIPS64R6O0-NEXT:    xori $3, $3, 3
-; MIPS64R6O0-NEXT:    sll $3, $3, 3
-; MIPS64R6O0-NEXT:    ori $5, $zero, 255
-; MIPS64R6O0-NEXT:    sllv $5, $5, $3
-; MIPS64R6O0-NEXT:    nor $6, $zero, $5
-; MIPS64R6O0-NEXT:    sllv $2, $2, $3
+; MIPS64R6O0-NEXT:    daddiu $3, $zero, -4
+; MIPS64R6O0-NEXT:    and $3, $1, $3
+; MIPS64R6O0-NEXT:    andi $5, $1, 3
+; MIPS64R6O0-NEXT:    xori $5, $5, 3
+; MIPS64R6O0-NEXT:    sll $5, $5, 3
+; MIPS64R6O0-NEXT:    ori $6, $zero, 255
+; MIPS64R6O0-NEXT:    sllv $6, $6, $5
+; MIPS64R6O0-NEXT:    nor $7, $zero, $6
+; MIPS64R6O0-NEXT:    sllv $2, $2, $5
 ; MIPS64R6O0-NEXT:  .LBB11_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    ll $8, 0($4)
-; MIPS64R6O0-NEXT:    and $9, $2, $5
-; MIPS64R6O0-NEXT:    and $10, $8, $6
-; MIPS64R6O0-NEXT:    or $10, $10, $9
-; MIPS64R6O0-NEXT:    sc $10, 0($4)
-; MIPS64R6O0-NEXT:    beqzc $10, .LBB11_1
+; MIPS64R6O0-NEXT:    ll $9, 0($3)
+; MIPS64R6O0-NEXT:    and $10, $2, $6
+; MIPS64R6O0-NEXT:    and $11, $9, $7
+; MIPS64R6O0-NEXT:    or $11, $11, $10
+; MIPS64R6O0-NEXT:    sc $11, 0($3)
+; MIPS64R6O0-NEXT:    beqzc $11, .LBB11_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    and $7, $8, $5
-; MIPS64R6O0-NEXT:    srlv $7, $7, $3
-; MIPS64R6O0-NEXT:    seb $7, $7
+; MIPS64R6O0-NEXT:    and $8, $9, $6
+; MIPS64R6O0-NEXT:    srlv $8, $8, $5
+; MIPS64R6O0-NEXT:    seb $8, $8
 ; MIPS64R6O0-NEXT:  # %bb.3: # %entry
-; MIPS64R6O0-NEXT:    sw $7, 12($sp) # 4-byte Folded Spill
+; MIPS64R6O0-NEXT:    sw $8, 12($sp) # 4-byte Folded Spill
 ; MIPS64R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS64R6O0-NEXT:    lw $1, 12($sp) # 4-byte Folded Reload
 ; MIPS64R6O0-NEXT:    seb $2, $1
@@ -4404,38 +4398,38 @@ define signext i8 @AtomicCmpSwap8(i8 signext %oldval, i8 signext %newval) nounwi
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(y)($2)
-; MIPS32O0-NEXT:    addiu $25, $zero, -4
-; MIPS32O0-NEXT:    and $25, $2, $25
-; MIPS32O0-NEXT:    andi $2, $2, 3
-; MIPS32O0-NEXT:    sll $2, $2, 3
-; MIPS32O0-NEXT:    ori $1, $zero, 255
-; MIPS32O0-NEXT:    sllv $1, $1, $2
-; MIPS32O0-NEXT:    nor $3, $zero, $1
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(y)($1)
+; MIPS32O0-NEXT:    addiu $2, $zero, -4
+; MIPS32O0-NEXT:    and $2, $1, $2
+; MIPS32O0-NEXT:    andi $1, $1, 3
+; MIPS32O0-NEXT:    sll $1, $1, 3
+; MIPS32O0-NEXT:    ori $3, $zero, 255
+; MIPS32O0-NEXT:    sllv $3, $3, $1
+; MIPS32O0-NEXT:    nor $6, $zero, $3
 ; MIPS32O0-NEXT:    andi $4, $4, 255
-; MIPS32O0-NEXT:    sllv $4, $4, $2
+; MIPS32O0-NEXT:    sllv $4, $4, $1
 ; MIPS32O0-NEXT:    andi $5, $5, 255
-; MIPS32O0-NEXT:    sllv $5, $5, $2
+; MIPS32O0-NEXT:    sllv $5, $5, $1
 ; MIPS32O0-NEXT:  $BB12_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $7, 0($25)
-; MIPS32O0-NEXT:    and $8, $7, $1
-; MIPS32O0-NEXT:    bne $8, $4, $BB12_3
+; MIPS32O0-NEXT:    ll $8, 0($2)
+; MIPS32O0-NEXT:    and $9, $8, $3
+; MIPS32O0-NEXT:    bne $9, $4, $BB12_3
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
 ; MIPS32O0-NEXT:    # in Loop: Header=BB12_1 Depth=1
-; MIPS32O0-NEXT:    and $7, $7, $3
-; MIPS32O0-NEXT:    or $7, $7, $5
-; MIPS32O0-NEXT:    sc $7, 0($25)
-; MIPS32O0-NEXT:    beqz $7, $BB12_1
+; MIPS32O0-NEXT:    and $8, $8, $6
+; MIPS32O0-NEXT:    or $8, $8, $5
+; MIPS32O0-NEXT:    sc $8, 0($2)
+; MIPS32O0-NEXT:    beqz $8, $BB12_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  $BB12_3: # %entry
-; MIPS32O0-NEXT:    srlv $6, $8, $2
-; MIPS32O0-NEXT:    sll $6, $6, 24
-; MIPS32O0-NEXT:    sra $6, $6, 24
+; MIPS32O0-NEXT:    srlv $7, $9, $1
+; MIPS32O0-NEXT:    sll $7, $7, 24
+; MIPS32O0-NEXT:    sra $7, $7, 24
 ; MIPS32O0-NEXT:  # %bb.4: # %entry
-; MIPS32O0-NEXT:    sw $6, 4($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    sw $7, 4($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:  # %bb.5: # %entry
 ; MIPS32O0-NEXT:    lw $1, 4($sp) # 4-byte Folded Reload
 ; MIPS32O0-NEXT:    sll $2, $1, 24
@@ -4520,39 +4514,39 @@ define signext i8 @AtomicCmpSwap8(i8 signext %oldval, i8 signext %newval) nounwi
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -16
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $5
-; MIPS32R6O0-NEXT:    move $1, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(y)($2)
-; MIPS32R6O0-NEXT:    addiu $3, $zero, -4
-; MIPS32R6O0-NEXT:    and $3, $2, $3
-; MIPS32R6O0-NEXT:    andi $2, $2, 3
-; MIPS32R6O0-NEXT:    sll $2, $2, 3
-; MIPS32R6O0-NEXT:    ori $6, $zero, 255
-; MIPS32R6O0-NEXT:    sllv $6, $6, $2
-; MIPS32R6O0-NEXT:    nor $7, $zero, $6
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $5
+; MIPS32R6O0-NEXT:    move $3, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(y)($1)
+; MIPS32R6O0-NEXT:    addiu $6, $zero, -4
+; MIPS32R6O0-NEXT:    and $6, $1, $6
+; MIPS32R6O0-NEXT:    andi $1, $1, 3
+; MIPS32R6O0-NEXT:    sll $1, $1, 3
+; MIPS32R6O0-NEXT:    ori $7, $zero, 255
+; MIPS32R6O0-NEXT:    sllv $7, $7, $1
+; MIPS32R6O0-NEXT:    nor $8, $zero, $7
 ; MIPS32R6O0-NEXT:    andi $4, $4, 255
-; MIPS32R6O0-NEXT:    sllv $4, $4, $2
+; MIPS32R6O0-NEXT:    sllv $4, $4, $1
 ; MIPS32R6O0-NEXT:    andi $5, $5, 255
-; MIPS32R6O0-NEXT:    sllv $5, $5, $2
+; MIPS32R6O0-NEXT:    sllv $5, $5, $1
 ; MIPS32R6O0-NEXT:  $BB12_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $9, 0($3)
-; MIPS32R6O0-NEXT:    and $10, $9, $6
-; MIPS32R6O0-NEXT:    bnec $10, $4, $BB12_3
+; MIPS32R6O0-NEXT:    ll $10, 0($6)
+; MIPS32R6O0-NEXT:    and $11, $10, $7
+; MIPS32R6O0-NEXT:    bnec $11, $4, $BB12_3
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
 ; MIPS32R6O0-NEXT:    # in Loop: Header=BB12_1 Depth=1
-; MIPS32R6O0-NEXT:    and $9, $9, $7
-; MIPS32R6O0-NEXT:    or $9, $9, $5
-; MIPS32R6O0-NEXT:    sc $9, 0($3)
-; MIPS32R6O0-NEXT:    beqzc $9, $BB12_1
+; MIPS32R6O0-NEXT:    and $10, $10, $8
+; MIPS32R6O0-NEXT:    or $10, $10, $5
+; MIPS32R6O0-NEXT:    sc $10, 0($6)
+; MIPS32R6O0-NEXT:    beqzc $10, $BB12_1
 ; MIPS32R6O0-NEXT:  $BB12_3: # %entry
-; MIPS32R6O0-NEXT:    srlv $8, $10, $2
-; MIPS32R6O0-NEXT:    seb $8, $8
+; MIPS32R6O0-NEXT:    srlv $9, $11, $1
+; MIPS32R6O0-NEXT:    seb $9, $9
 ; MIPS32R6O0-NEXT:  # %bb.4: # %entry
-; MIPS32R6O0-NEXT:    sw $25, 12($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $1, 8($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $8, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 12($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $3, 8($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $9, 4($sp) # 4-byte Folded Spill
 ; MIPS32R6O0-NEXT:  # %bb.5: # %entry
 ; MIPS32R6O0-NEXT:    lw $2, 4($sp) # 4-byte Folded Reload
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 16
@@ -4989,30 +4983,30 @@ define i1 @AtomicCmpSwapRes8(i8* %ptr, i8 signext %oldval, i8 signext %newval) n
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -8
 ; MIPS32O0-NEXT:    addiu $1, $zero, -4
 ; MIPS32O0-NEXT:    and $1, $4, $1
-; MIPS32O0-NEXT:    andi $4, $4, 3
-; MIPS32O0-NEXT:    sll $4, $4, 3
-; MIPS32O0-NEXT:    ori $2, $zero, 255
-; MIPS32O0-NEXT:    sllv $2, $2, $4
-; MIPS32O0-NEXT:    nor $3, $zero, $2
+; MIPS32O0-NEXT:    andi $2, $4, 3
+; MIPS32O0-NEXT:    sll $2, $2, 3
+; MIPS32O0-NEXT:    ori $3, $zero, 255
+; MIPS32O0-NEXT:    sllv $3, $3, $2
+; MIPS32O0-NEXT:    nor $4, $zero, $3
 ; MIPS32O0-NEXT:    andi $7, $5, 255
-; MIPS32O0-NEXT:    sllv $7, $7, $4
+; MIPS32O0-NEXT:    sllv $7, $7, $2
 ; MIPS32O0-NEXT:    andi $6, $6, 255
-; MIPS32O0-NEXT:    sllv $6, $6, $4
+; MIPS32O0-NEXT:    sllv $6, $6, $2
 ; MIPS32O0-NEXT:  $BB13_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
 ; MIPS32O0-NEXT:    ll $9, 0($1)
-; MIPS32O0-NEXT:    and $10, $9, $2
+; MIPS32O0-NEXT:    and $10, $9, $3
 ; MIPS32O0-NEXT:    bne $10, $7, $BB13_3
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
 ; MIPS32O0-NEXT:    # in Loop: Header=BB13_1 Depth=1
-; MIPS32O0-NEXT:    and $9, $9, $3
+; MIPS32O0-NEXT:    and $9, $9, $4
 ; MIPS32O0-NEXT:    or $9, $9, $6
 ; MIPS32O0-NEXT:    sc $9, 0($1)
 ; MIPS32O0-NEXT:    beqz $9, $BB13_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  $BB13_3: # %entry
-; MIPS32O0-NEXT:    srlv $8, $10, $4
+; MIPS32O0-NEXT:    srlv $8, $10, $2
 ; MIPS32O0-NEXT:    sll $8, $8, 24
 ; MIPS32O0-NEXT:    sra $8, $8, 24
 ; MIPS32O0-NEXT:  # %bb.4: # %entry
@@ -5285,37 +5279,37 @@ define i1 @AtomicCmpSwapRes8(i8* %ptr, i8 signext %oldval, i8 signext %newval) n
 ; MIPS64R6O0-NEXT:    daddiu $sp, $sp, -32
 ; MIPS64R6O0-NEXT:    move $1, $6
 ; MIPS64R6O0-NEXT:    move $2, $5
-; MIPS64R6O0-NEXT:    move $5, $4
-; MIPS64R6O0-NEXT:    daddiu $6, $zero, -4
-; MIPS64R6O0-NEXT:    and $6, $4, $6
-; MIPS64R6O0-NEXT:    andi $3, $4, 3
-; MIPS64R6O0-NEXT:    xori $3, $3, 3
-; MIPS64R6O0-NEXT:    sll $3, $3, 3
-; MIPS64R6O0-NEXT:    ori $7, $zero, 255
-; MIPS64R6O0-NEXT:    sllv $7, $7, $3
-; MIPS64R6O0-NEXT:    nor $8, $zero, $7
-; MIPS64R6O0-NEXT:    andi $9, $2, 255
-; MIPS64R6O0-NEXT:    sllv $9, $9, $3
+; MIPS64R6O0-NEXT:    move $3, $4
+; MIPS64R6O0-NEXT:    daddiu $5, $zero, -4
+; MIPS64R6O0-NEXT:    and $5, $4, $5
+; MIPS64R6O0-NEXT:    andi $7, $4, 3
+; MIPS64R6O0-NEXT:    xori $7, $7, 3
+; MIPS64R6O0-NEXT:    sll $7, $7, 3
+; MIPS64R6O0-NEXT:    ori $8, $zero, 255
+; MIPS64R6O0-NEXT:    sllv $8, $8, $7
+; MIPS64R6O0-NEXT:    nor $9, $zero, $8
+; MIPS64R6O0-NEXT:    andi $10, $2, 255
+; MIPS64R6O0-NEXT:    sllv $10, $10, $7
 ; MIPS64R6O0-NEXT:    andi $1, $1, 255
-; MIPS64R6O0-NEXT:    sllv $1, $1, $3
+; MIPS64R6O0-NEXT:    sllv $1, $1, $7
 ; MIPS64R6O0-NEXT:  .LBB13_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    ll $11, 0($6)
-; MIPS64R6O0-NEXT:    and $12, $11, $7
-; MIPS64R6O0-NEXT:    bnec $12, $9, .LBB13_3
+; MIPS64R6O0-NEXT:    ll $12, 0($5)
+; MIPS64R6O0-NEXT:    and $13, $12, $8
+; MIPS64R6O0-NEXT:    bnec $13, $10, .LBB13_3
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
 ; MIPS64R6O0-NEXT:    # in Loop: Header=BB13_1 Depth=1
-; MIPS64R6O0-NEXT:    and $11, $11, $8
-; MIPS64R6O0-NEXT:    or $11, $11, $1
-; MIPS64R6O0-NEXT:    sc $11, 0($6)
-; MIPS64R6O0-NEXT:    beqzc $11, .LBB13_1
+; MIPS64R6O0-NEXT:    and $12, $12, $9
+; MIPS64R6O0-NEXT:    or $12, $12, $1
+; MIPS64R6O0-NEXT:    sc $12, 0($5)
+; MIPS64R6O0-NEXT:    beqzc $12, .LBB13_1
 ; MIPS64R6O0-NEXT:  .LBB13_3: # %entry
-; MIPS64R6O0-NEXT:    srlv $10, $12, $3
-; MIPS64R6O0-NEXT:    seb $10, $10
+; MIPS64R6O0-NEXT:    srlv $11, $13, $7
+; MIPS64R6O0-NEXT:    seb $11, $11
 ; MIPS64R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS64R6O0-NEXT:    sw $2, 28($sp) # 4-byte Folded Spill
-; MIPS64R6O0-NEXT:    sd $5, 16($sp) # 8-byte Folded Spill
-; MIPS64R6O0-NEXT:    sw $10, 12($sp) # 4-byte Folded Spill
+; MIPS64R6O0-NEXT:    sd $3, 16($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    sw $11, 12($sp) # 4-byte Folded Spill
 ; MIPS64R6O0-NEXT:  # %bb.5: # %entry
 ; MIPS64R6O0-NEXT:    lw $1, 12($sp) # 4-byte Folded Reload
 ; MIPS64R6O0-NEXT:    lw $2, 28($sp) # 4-byte Folded Reload
@@ -5557,33 +5551,33 @@ define signext i16 @AtomicLoadAdd16(i16 signext %incr) nounwind {
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(z)($2)
-; MIPS32O0-NEXT:    addiu $25, $zero, -4
-; MIPS32O0-NEXT:    and $25, $2, $25
-; MIPS32O0-NEXT:    andi $2, $2, 3
-; MIPS32O0-NEXT:    sll $2, $2, 3
-; MIPS32O0-NEXT:    ori $1, $zero, 65535
-; MIPS32O0-NEXT:    sllv $1, $1, $2
-; MIPS32O0-NEXT:    nor $3, $zero, $1
-; MIPS32O0-NEXT:    sllv $4, $4, $2
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(z)($1)
+; MIPS32O0-NEXT:    addiu $2, $zero, -4
+; MIPS32O0-NEXT:    and $2, $1, $2
+; MIPS32O0-NEXT:    andi $1, $1, 3
+; MIPS32O0-NEXT:    sll $1, $1, 3
+; MIPS32O0-NEXT:    ori $3, $zero, 65535
+; MIPS32O0-NEXT:    sllv $3, $3, $1
+; MIPS32O0-NEXT:    nor $5, $zero, $3
+; MIPS32O0-NEXT:    sllv $4, $4, $1
 ; MIPS32O0-NEXT:  $BB14_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $6, 0($25)
-; MIPS32O0-NEXT:    addu $7, $6, $4
-; MIPS32O0-NEXT:    and $7, $7, $1
-; MIPS32O0-NEXT:    and $8, $6, $3
-; MIPS32O0-NEXT:    or $8, $8, $7
-; MIPS32O0-NEXT:    sc $8, 0($25)
-; MIPS32O0-NEXT:    beqz $8, $BB14_1
+; MIPS32O0-NEXT:    ll $7, 0($2)
+; MIPS32O0-NEXT:    addu $8, $7, $4
+; MIPS32O0-NEXT:    and $8, $8, $3
+; MIPS32O0-NEXT:    and $9, $7, $5
+; MIPS32O0-NEXT:    or $9, $9, $8
+; MIPS32O0-NEXT:    sc $9, 0($2)
+; MIPS32O0-NEXT:    beqz $9, $BB14_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    and $5, $6, $1
-; MIPS32O0-NEXT:    srlv $5, $5, $2
-; MIPS32O0-NEXT:    sll $5, $5, 16
-; MIPS32O0-NEXT:    sra $5, $5, 16
+; MIPS32O0-NEXT:    and $6, $7, $3
+; MIPS32O0-NEXT:    srlv $6, $6, $1
+; MIPS32O0-NEXT:    sll $6, $6, 16
+; MIPS32O0-NEXT:    sra $6, $6, 16
 ; MIPS32O0-NEXT:  # %bb.3: # %entry
-; MIPS32O0-NEXT:    sw $5, 4($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    sw $6, 4($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:  # %bb.4: # %entry
 ; MIPS32O0-NEXT:    lw $1, 4($sp) # 4-byte Folded Reload
 ; MIPS32O0-NEXT:    sll $2, $1, 16
@@ -5660,33 +5654,33 @@ define signext i16 @AtomicLoadAdd16(i16 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(z)($2)
-; MIPS32R6O0-NEXT:    addiu $1, $zero, -4
-; MIPS32R6O0-NEXT:    and $1, $2, $1
-; MIPS32R6O0-NEXT:    andi $2, $2, 3
-; MIPS32R6O0-NEXT:    sll $2, $2, 3
-; MIPS32R6O0-NEXT:    ori $3, $zero, 65535
-; MIPS32R6O0-NEXT:    sllv $3, $3, $2
-; MIPS32R6O0-NEXT:    nor $5, $zero, $3
-; MIPS32R6O0-NEXT:    sllv $4, $4, $2
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(z)($1)
+; MIPS32R6O0-NEXT:    addiu $3, $zero, -4
+; MIPS32R6O0-NEXT:    and $3, $1, $3
+; MIPS32R6O0-NEXT:    andi $1, $1, 3
+; MIPS32R6O0-NEXT:    sll $1, $1, 3
+; MIPS32R6O0-NEXT:    ori $5, $zero, 65535
+; MIPS32R6O0-NEXT:    sllv $5, $5, $1
+; MIPS32R6O0-NEXT:    nor $6, $zero, $5
+; MIPS32R6O0-NEXT:    sllv $4, $4, $1
 ; MIPS32R6O0-NEXT:  $BB14_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $7, 0($1)
-; MIPS32R6O0-NEXT:    addu $8, $7, $4
-; MIPS32R6O0-NEXT:    and $8, $8, $3
-; MIPS32R6O0-NEXT:    and $9, $7, $5
-; MIPS32R6O0-NEXT:    or $9, $9, $8
-; MIPS32R6O0-NEXT:    sc $9, 0($1)
-; MIPS32R6O0-NEXT:    beqzc $9, $BB14_1
+; MIPS32R6O0-NEXT:    ll $8, 0($3)
+; MIPS32R6O0-NEXT:    addu $9, $8, $4
+; MIPS32R6O0-NEXT:    and $9, $9, $5
+; MIPS32R6O0-NEXT:    and $10, $8, $6
+; MIPS32R6O0-NEXT:    or $10, $10, $9
+; MIPS32R6O0-NEXT:    sc $10, 0($3)
+; MIPS32R6O0-NEXT:    beqzc $10, $BB14_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    and $6, $7, $3
-; MIPS32R6O0-NEXT:    srlv $6, $6, $2
-; MIPS32R6O0-NEXT:    seh $6, $6
+; MIPS32R6O0-NEXT:    and $7, $8, $5
+; MIPS32R6O0-NEXT:    srlv $7, $7, $1
+; MIPS32R6O0-NEXT:    seh $7, $7
 ; MIPS32R6O0-NEXT:  # %bb.3: # %entry
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $6, 0($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $7, 0($sp) # 4-byte Folded Spill
 ; MIPS32R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS32R6O0-NEXT:    lw $1, 0($sp) # 4-byte Folded Reload
 ; MIPS32R6O0-NEXT:    seh $2, $1
@@ -5832,30 +5826,30 @@ define signext i16 @AtomicLoadAdd16(i16 signext %incr) nounwind {
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicLoadAdd16)))
 ; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(z)($1)
-; MIPS64R6O0-NEXT:    daddiu $4, $zero, -4
-; MIPS64R6O0-NEXT:    and $4, $1, $4
-; MIPS64R6O0-NEXT:    andi $3, $1, 3
-; MIPS64R6O0-NEXT:    xori $3, $3, 2
-; MIPS64R6O0-NEXT:    sll $3, $3, 3
-; MIPS64R6O0-NEXT:    ori $5, $zero, 65535
-; MIPS64R6O0-NEXT:    sllv $5, $5, $3
-; MIPS64R6O0-NEXT:    nor $6, $zero, $5
-; MIPS64R6O0-NEXT:    sllv $2, $2, $3
+; MIPS64R6O0-NEXT:    daddiu $3, $zero, -4
+; MIPS64R6O0-NEXT:    and $3, $1, $3
+; MIPS64R6O0-NEXT:    andi $5, $1, 3
+; MIPS64R6O0-NEXT:    xori $5, $5, 2
+; MIPS64R6O0-NEXT:    sll $5, $5, 3
+; MIPS64R6O0-NEXT:    ori $6, $zero, 65535
+; MIPS64R6O0-NEXT:    sllv $6, $6, $5
+; MIPS64R6O0-NEXT:    nor $7, $zero, $6
+; MIPS64R6O0-NEXT:    sllv $2, $2, $5
 ; MIPS64R6O0-NEXT:  .LBB14_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    ll $8, 0($4)
-; MIPS64R6O0-NEXT:    addu $9, $8, $2
-; MIPS64R6O0-NEXT:    and $9, $9, $5
-; MIPS64R6O0-NEXT:    and $10, $8, $6
-; MIPS64R6O0-NEXT:    or $10, $10, $9
-; MIPS64R6O0-NEXT:    sc $10, 0($4)
-; MIPS64R6O0-NEXT:    beqzc $10, .LBB14_1
+; MIPS64R6O0-NEXT:    ll $9, 0($3)
+; MIPS64R6O0-NEXT:    addu $10, $9, $2
+; MIPS64R6O0-NEXT:    and $10, $10, $6
+; MIPS64R6O0-NEXT:    and $11, $9, $7
+; MIPS64R6O0-NEXT:    or $11, $11, $10
+; MIPS64R6O0-NEXT:    sc $11, 0($3)
+; MIPS64R6O0-NEXT:    beqzc $11, .LBB14_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    and $7, $8, $5
-; MIPS64R6O0-NEXT:    srlv $7, $7, $3
-; MIPS64R6O0-NEXT:    seh $7, $7
+; MIPS64R6O0-NEXT:    and $8, $9, $6
+; MIPS64R6O0-NEXT:    srlv $8, $8, $5
+; MIPS64R6O0-NEXT:    seh $8, $8
 ; MIPS64R6O0-NEXT:  # %bb.3: # %entry
-; MIPS64R6O0-NEXT:    sw $7, 12($sp) # 4-byte Folded Spill
+; MIPS64R6O0-NEXT:    sw $8, 12($sp) # 4-byte Folded Spill
 ; MIPS64R6O0-NEXT:  # %bb.4: # %entry
 ; MIPS64R6O0-NEXT:    lw $1, 12($sp) # 4-byte Folded Reload
 ; MIPS64R6O0-NEXT:    seh $2, $1
@@ -6084,36 +6078,36 @@ define {i16, i1} @foo(i16* %addr, i16 %l, i16 %r, i16 %new) {
 ; MIPS32O0:       # %bb.0:
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -8
 ; MIPS32O0-NEXT:    .cfi_def_cfa_offset 8
-; MIPS32O0-NEXT:    addu $5, $5, $6
+; MIPS32O0-NEXT:    addu $1, $5, $6
 ; MIPS32O0-NEXT:    sync
-; MIPS32O0-NEXT:    addiu $6, $zero, -4
-; MIPS32O0-NEXT:    and $6, $4, $6
-; MIPS32O0-NEXT:    andi $4, $4, 3
-; MIPS32O0-NEXT:    sll $4, $4, 3
-; MIPS32O0-NEXT:    ori $1, $zero, 65535
-; MIPS32O0-NEXT:    sllv $1, $1, $4
-; MIPS32O0-NEXT:    nor $2, $zero, $1
-; MIPS32O0-NEXT:    andi $3, $5, 65535
-; MIPS32O0-NEXT:    sllv $3, $3, $4
+; MIPS32O0-NEXT:    addiu $2, $zero, -4
+; MIPS32O0-NEXT:    and $2, $4, $2
+; MIPS32O0-NEXT:    andi $3, $4, 3
+; MIPS32O0-NEXT:    sll $3, $3, 3
+; MIPS32O0-NEXT:    ori $4, $zero, 65535
+; MIPS32O0-NEXT:    sllv $4, $4, $3
+; MIPS32O0-NEXT:    nor $5, $zero, $4
+; MIPS32O0-NEXT:    andi $6, $1, 65535
+; MIPS32O0-NEXT:    sllv $6, $6, $3
 ; MIPS32O0-NEXT:    andi $7, $7, 65535
-; MIPS32O0-NEXT:    sllv $7, $7, $4
+; MIPS32O0-NEXT:    sllv $7, $7, $3
 ; MIPS32O0-NEXT:  $BB15_1: # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $9, 0($6)
-; MIPS32O0-NEXT:    and $10, $9, $1
-; MIPS32O0-NEXT:    bne $10, $3, $BB15_3
+; MIPS32O0-NEXT:    ll $9, 0($2)
+; MIPS32O0-NEXT:    and $10, $9, $4
+; MIPS32O0-NEXT:    bne $10, $6, $BB15_3
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # in Loop: Header=BB15_1 Depth=1
-; MIPS32O0-NEXT:    and $9, $9, $2
+; MIPS32O0-NEXT:    and $9, $9, $5
 ; MIPS32O0-NEXT:    or $9, $9, $7
-; MIPS32O0-NEXT:    sc $9, 0($6)
+; MIPS32O0-NEXT:    sc $9, 0($2)
 ; MIPS32O0-NEXT:    beqz $9, $BB15_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  $BB15_3:
-; MIPS32O0-NEXT:    srlv $8, $10, $4
+; MIPS32O0-NEXT:    srlv $8, $10, $3
 ; MIPS32O0-NEXT:    sll $8, $8, 16
 ; MIPS32O0-NEXT:    sra $8, $8, 16
 ; MIPS32O0-NEXT:  # %bb.4:
-; MIPS32O0-NEXT:    sw $5, 4($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    sw $1, 4($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:    sw $8, 0($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:  # %bb.5:
 ; MIPS32O0-NEXT:    lw $1, 4($sp) # 4-byte Folded Reload
@@ -6684,19 +6678,18 @@ define i32 @CheckSync(i32 signext %v) nounwind noinline {
 ; MIPS32O0:       # %bb.0: # %entry
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32O0-NEXT:    addu $2, $2, $25
+; MIPS32O0-NEXT:    addu $1, $2, $25
 ; MIPS32O0-NEXT:    sync
-; MIPS32O0-NEXT:    lw $2, %got(countsint)($2)
+; MIPS32O0-NEXT:    lw $1, %got(countsint)($1)
 ; MIPS32O0-NEXT:  $BB16_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $25, 0($2)
-; MIPS32O0-NEXT:    addu $1, $25, $4
-; MIPS32O0-NEXT:    sc $1, 0($2)
-; MIPS32O0-NEXT:    beqz $1, $BB16_1
+; MIPS32O0-NEXT:    ll $2, 0($1)
+; MIPS32O0-NEXT:    addu $3, $2, $4
+; MIPS32O0-NEXT:    sc $3, 0($1)
+; MIPS32O0-NEXT:    beqz $3, $BB16_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
 ; MIPS32O0-NEXT:    sync
-; MIPS32O0-NEXT:    move $2, $25
 ; MIPS32O0-NEXT:    jr $ra
 ; MIPS32O0-NEXT:    nop
 ;
@@ -6741,20 +6734,20 @@ define i32 @CheckSync(i32 signext %v) nounwind noinline {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
 ; MIPS32R6O0-NEXT:    sync
-; MIPS32R6O0-NEXT:    lw $2, %got(countsint)($2)
+; MIPS32R6O0-NEXT:    lw $1, %got(countsint)($1)
 ; MIPS32R6O0-NEXT:  $BB16_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $1, 0($2)
-; MIPS32R6O0-NEXT:    addu $3, $1, $4
-; MIPS32R6O0-NEXT:    sc $3, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $3, $BB16_1
+; MIPS32R6O0-NEXT:    ll $3, 0($1)
+; MIPS32R6O0-NEXT:    addu $5, $3, $4
+; MIPS32R6O0-NEXT:    sc $5, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $5, $BB16_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
 ; MIPS32R6O0-NEXT:    sync
-; MIPS32R6O0-NEXT:    move $2, $1
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    move $2, $3
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;
@@ -6990,34 +6983,34 @@ define i32 @zeroreg() nounwind {
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $sp, $sp, -16
-; MIPS32O0-NEXT:    addu $2, $2, $25
+; MIPS32O0-NEXT:    addu $1, $2, $25
 ; MIPS32O0-NEXT:    sync
-; MIPS32O0-NEXT:    lw $2, %got(a)($2)
-; MIPS32O0-NEXT:    addiu $25, $zero, 0
-; MIPS32O0-NEXT:    addiu $1, $zero, 1
-; MIPS32O0-NEXT:    lw $3, 12($sp) # 4-byte Folded Reload
-; MIPS32O0-NEXT:    move $4, $1
+; MIPS32O0-NEXT:    lw $1, %got(a)($1)
+; MIPS32O0-NEXT:    addiu $2, $zero, 0
+; MIPS32O0-NEXT:    addiu $3, $zero, 1
+; MIPS32O0-NEXT:    lw $4, 12($sp) # 4-byte Folded Reload
+; MIPS32O0-NEXT:    move $5, $3
 ; MIPS32O0-NEXT:  $BB17_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $5, 0($2)
-; MIPS32O0-NEXT:    bne $5, $4, $BB17_3
+; MIPS32O0-NEXT:    ll $6, 0($1)
+; MIPS32O0-NEXT:    bne $6, $5, $BB17_3
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
 ; MIPS32O0-NEXT:    # in Loop: Header=BB17_1 Depth=1
-; MIPS32O0-NEXT:    move $6, $25
-; MIPS32O0-NEXT:    sc $6, 0($2)
-; MIPS32O0-NEXT:    beqz $6, $BB17_1
+; MIPS32O0-NEXT:    move $7, $2
+; MIPS32O0-NEXT:    sc $7, 0($1)
+; MIPS32O0-NEXT:    beqz $7, $BB17_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  $BB17_3: # %entry
-; MIPS32O0-NEXT:    xor $1, $5, $1
+; MIPS32O0-NEXT:    xor $1, $6, $3
 ; MIPS32O0-NEXT:    sltiu $1, $1, 1
 ; MIPS32O0-NEXT:    sync
 ; MIPS32O0-NEXT:    addiu $2, $zero, 1
-; MIPS32O0-NEXT:    xor $2, $5, $2
+; MIPS32O0-NEXT:    xor $2, $6, $2
 ; MIPS32O0-NEXT:    sltiu $2, $2, 1
 ; MIPS32O0-NEXT:    andi $2, $2, 1
-; MIPS32O0-NEXT:    sw $5, 12($sp) # 4-byte Folded Spill
-; MIPS32O0-NEXT:    sw $3, 8($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    sw $6, 12($sp) # 4-byte Folded Spill
+; MIPS32O0-NEXT:    sw $4, 8($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:    sw $1, 4($sp) # 4-byte Folded Spill
 ; MIPS32O0-NEXT:    addiu $sp, $sp, 16
 ; MIPS32O0-NEXT:    jr $ra
@@ -7079,28 +7072,28 @@ define i32 @zeroreg() nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
 ; MIPS32R6O0-NEXT:    sync
-; MIPS32R6O0-NEXT:    lw $2, %got(a)($2)
-; MIPS32R6O0-NEXT:    addiu $25, $zero, 0
-; MIPS32R6O0-NEXT:    addiu $1, $zero, 1
-; MIPS32R6O0-NEXT:    lw $3, 4($sp) # 4-byte Folded Reload
-; MIPS32R6O0-NEXT:    move $4, $1
+; MIPS32R6O0-NEXT:    lw $1, %got(a)($1)
+; MIPS32R6O0-NEXT:    addiu $2, $zero, 0
+; MIPS32R6O0-NEXT:    addiu $3, $zero, 1
+; MIPS32R6O0-NEXT:    lw $4, 4($sp) # 4-byte Folded Reload
+; MIPS32R6O0-NEXT:    move $5, $3
 ; MIPS32R6O0-NEXT:  $BB17_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $5, 0($2)
-; MIPS32R6O0-NEXT:    bnec $5, $4, $BB17_3
+; MIPS32R6O0-NEXT:    ll $6, 0($1)
+; MIPS32R6O0-NEXT:    bnec $6, $5, $BB17_3
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
 ; MIPS32R6O0-NEXT:    # in Loop: Header=BB17_1 Depth=1
-; MIPS32R6O0-NEXT:    move $6, $25
-; MIPS32R6O0-NEXT:    sc $6, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $6, $BB17_1
+; MIPS32R6O0-NEXT:    move $7, $2
+; MIPS32R6O0-NEXT:    sc $7, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $7, $BB17_1
 ; MIPS32R6O0-NEXT:  $BB17_3: # %entry
-; MIPS32R6O0-NEXT:    xor $1, $5, $1
+; MIPS32R6O0-NEXT:    xor $1, $6, $3
 ; MIPS32R6O0-NEXT:    sltiu $2, $1, 1
 ; MIPS32R6O0-NEXT:    sync
-; MIPS32R6O0-NEXT:    sw $5, 4($sp) # 4-byte Folded Spill
-; MIPS32R6O0-NEXT:    sw $3, 0($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $6, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $4, 0($sp) # 4-byte Folded Spill
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;
@@ -7403,18 +7396,17 @@ define i32 @AtomicLoadAdd32_OffGt9Bit(i32 signext %incr) nounwind {
 ; MIPS32O0:       # %bb.0: # %entry
 ; MIPS32O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32O0-NEXT:    addu $2, $2, $25
-; MIPS32O0-NEXT:    lw $2, %got(x)($2)
-; MIPS32O0-NEXT:    addiu $2, $2, 1024
+; MIPS32O0-NEXT:    addu $1, $2, $25
+; MIPS32O0-NEXT:    lw $1, %got(x)($1)
+; MIPS32O0-NEXT:    addiu $1, $1, 1024
 ; MIPS32O0-NEXT:  $BB18_1: # %entry
 ; MIPS32O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32O0-NEXT:    ll $25, 0($2)
-; MIPS32O0-NEXT:    addu $1, $25, $4
-; MIPS32O0-NEXT:    sc $1, 0($2)
-; MIPS32O0-NEXT:    beqz $1, $BB18_1
+; MIPS32O0-NEXT:    ll $2, 0($1)
+; MIPS32O0-NEXT:    addu $3, $2, $4
+; MIPS32O0-NEXT:    sc $3, 0($1)
+; MIPS32O0-NEXT:    beqz $3, $BB18_1
 ; MIPS32O0-NEXT:    nop
 ; MIPS32O0-NEXT:  # %bb.2: # %entry
-; MIPS32O0-NEXT:    move $2, $25
 ; MIPS32O0-NEXT:    jr $ra
 ; MIPS32O0-NEXT:    nop
 ;
@@ -7458,19 +7450,19 @@ define i32 @AtomicLoadAdd32_OffGt9Bit(i32 signext %incr) nounwind {
 ; MIPS32R6O0-NEXT:    lui $2, %hi(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $2, $2, %lo(_gp_disp)
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, -8
-; MIPS32R6O0-NEXT:    addu $2, $2, $25
-; MIPS32R6O0-NEXT:    move $25, $4
-; MIPS32R6O0-NEXT:    lw $2, %got(x)($2)
-; MIPS32R6O0-NEXT:    addiu $2, $2, 1024
+; MIPS32R6O0-NEXT:    addu $1, $2, $25
+; MIPS32R6O0-NEXT:    move $2, $4
+; MIPS32R6O0-NEXT:    lw $1, %got(x)($1)
+; MIPS32R6O0-NEXT:    addiu $1, $1, 1024
 ; MIPS32R6O0-NEXT:  $BB18_1: # %entry
 ; MIPS32R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS32R6O0-NEXT:    ll $1, 0($2)
-; MIPS32R6O0-NEXT:    addu $3, $1, $4
-; MIPS32R6O0-NEXT:    sc $3, 0($2)
-; MIPS32R6O0-NEXT:    beqzc $3, $BB18_1
+; MIPS32R6O0-NEXT:    ll $3, 0($1)
+; MIPS32R6O0-NEXT:    addu $5, $3, $4
+; MIPS32R6O0-NEXT:    sc $5, 0($1)
+; MIPS32R6O0-NEXT:    beqzc $5, $BB18_1
 ; MIPS32R6O0-NEXT:  # %bb.2: # %entry
-; MIPS32R6O0-NEXT:    move $2, $1
-; MIPS32R6O0-NEXT:    sw $25, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    sw $2, 4($sp) # 4-byte Folded Spill
+; MIPS32R6O0-NEXT:    move $2, $3
 ; MIPS32R6O0-NEXT:    addiu $sp, $sp, 8
 ; MIPS32R6O0-NEXT:    jrc $ra
 ;

diff --git a/llvm/test/CodeGen/Mips/atomic64.ll b/llvm/test/CodeGen/Mips/atomic64.ll
index 8e5002b..c8f08df 100644 (file)
--- a/llvm/test/CodeGen/Mips/atomic64.ll
+++ b/llvm/test/CodeGen/Mips/atomic64.ll
@@ -96,16 +96,17 @@ define i64 @AtomicLoadAdd(i64 signext %incr) nounwind {
 ; MIPS64R6O0-NEXT:    lui $1, %hi(%neg(%gp_rel(AtomicLoadAdd)))
 ; MIPS64R6O0-NEXT:    daddu $1, $1, $25
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicLoadAdd)))
-; MIPS64R6O0-NEXT:    move $25, $4
+; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(x)($1)
 ; MIPS64R6O0-NEXT:  .LBB0_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    lld $2, 0($1)
-; MIPS64R6O0-NEXT:    daddu $3, $2, $4
-; MIPS64R6O0-NEXT:    scd $3, 0($1)
-; MIPS64R6O0-NEXT:    beqzc $3, .LBB0_1
+; MIPS64R6O0-NEXT:    lld $3, 0($1)
+; MIPS64R6O0-NEXT:    daddu $5, $3, $4
+; MIPS64R6O0-NEXT:    scd $5, 0($1)
+; MIPS64R6O0-NEXT:    beqzc $5, .LBB0_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    sd $25, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    sd $2, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    move $2, $3
 ; MIPS64R6O0-NEXT:    daddiu $sp, $sp, 16
 ; MIPS64R6O0-NEXT:    jrc $ra
 ;
@@ -256,16 +257,17 @@ define i64 @AtomicLoadSub(i64 signext %incr) nounwind {
 ; MIPS64R6O0-NEXT:    lui $1, %hi(%neg(%gp_rel(AtomicLoadSub)))
 ; MIPS64R6O0-NEXT:    daddu $1, $1, $25
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicLoadSub)))
-; MIPS64R6O0-NEXT:    move $25, $4
+; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(x)($1)
 ; MIPS64R6O0-NEXT:  .LBB1_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    lld $2, 0($1)
-; MIPS64R6O0-NEXT:    dsubu $3, $2, $4
-; MIPS64R6O0-NEXT:    scd $3, 0($1)
-; MIPS64R6O0-NEXT:    beqzc $3, .LBB1_1
+; MIPS64R6O0-NEXT:    lld $3, 0($1)
+; MIPS64R6O0-NEXT:    dsubu $5, $3, $4
+; MIPS64R6O0-NEXT:    scd $5, 0($1)
+; MIPS64R6O0-NEXT:    beqzc $5, .LBB1_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    sd $25, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    sd $2, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    move $2, $3
 ; MIPS64R6O0-NEXT:    daddiu $sp, $sp, 16
 ; MIPS64R6O0-NEXT:    jrc $ra
 ;
@@ -416,16 +418,17 @@ define i64 @AtomicLoadAnd(i64 signext %incr) nounwind {
 ; MIPS64R6O0-NEXT:    lui $1, %hi(%neg(%gp_rel(AtomicLoadAnd)))
 ; MIPS64R6O0-NEXT:    daddu $1, $1, $25
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicLoadAnd)))
-; MIPS64R6O0-NEXT:    move $25, $4
+; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(x)($1)
 ; MIPS64R6O0-NEXT:  .LBB2_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    lld $2, 0($1)
-; MIPS64R6O0-NEXT:    and $3, $2, $4
-; MIPS64R6O0-NEXT:    scd $3, 0($1)
-; MIPS64R6O0-NEXT:    beqzc $3, .LBB2_1
+; MIPS64R6O0-NEXT:    lld $3, 0($1)
+; MIPS64R6O0-NEXT:    and $5, $3, $4
+; MIPS64R6O0-NEXT:    scd $5, 0($1)
+; MIPS64R6O0-NEXT:    beqzc $5, .LBB2_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    sd $25, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    sd $2, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    move $2, $3
 ; MIPS64R6O0-NEXT:    daddiu $sp, $sp, 16
 ; MIPS64R6O0-NEXT:    jrc $ra
 ;
@@ -576,16 +579,17 @@ define i64 @AtomicLoadOr(i64 signext %incr) nounwind {
 ; MIPS64R6O0-NEXT:    lui $1, %hi(%neg(%gp_rel(AtomicLoadOr)))
 ; MIPS64R6O0-NEXT:    daddu $1, $1, $25
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicLoadOr)))
-; MIPS64R6O0-NEXT:    move $25, $4
+; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(x)($1)
 ; MIPS64R6O0-NEXT:  .LBB3_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    lld $2, 0($1)
-; MIPS64R6O0-NEXT:    or $3, $2, $4
-; MIPS64R6O0-NEXT:    scd $3, 0($1)
-; MIPS64R6O0-NEXT:    beqzc $3, .LBB3_1
+; MIPS64R6O0-NEXT:    lld $3, 0($1)
+; MIPS64R6O0-NEXT:    or $5, $3, $4
+; MIPS64R6O0-NEXT:    scd $5, 0($1)
+; MIPS64R6O0-NEXT:    beqzc $5, .LBB3_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    sd $25, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    sd $2, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    move $2, $3
 ; MIPS64R6O0-NEXT:    daddiu $sp, $sp, 16
 ; MIPS64R6O0-NEXT:    jrc $ra
 ;
@@ -736,16 +740,17 @@ define i64 @AtomicLoadXor(i64 signext %incr) nounwind {
 ; MIPS64R6O0-NEXT:    lui $1, %hi(%neg(%gp_rel(AtomicLoadXor)))
 ; MIPS64R6O0-NEXT:    daddu $1, $1, $25
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicLoadXor)))
-; MIPS64R6O0-NEXT:    move $25, $4
+; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(x)($1)
 ; MIPS64R6O0-NEXT:  .LBB4_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    lld $2, 0($1)
-; MIPS64R6O0-NEXT:    xor $3, $2, $4
-; MIPS64R6O0-NEXT:    scd $3, 0($1)
-; MIPS64R6O0-NEXT:    beqzc $3, .LBB4_1
+; MIPS64R6O0-NEXT:    lld $3, 0($1)
+; MIPS64R6O0-NEXT:    xor $5, $3, $4
+; MIPS64R6O0-NEXT:    scd $5, 0($1)
+; MIPS64R6O0-NEXT:    beqzc $5, .LBB4_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    sd $25, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    sd $2, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    move $2, $3
 ; MIPS64R6O0-NEXT:    daddiu $sp, $sp, 16
 ; MIPS64R6O0-NEXT:    jrc $ra
 ;
@@ -900,17 +905,18 @@ define i64 @AtomicLoadNand(i64 signext %incr) nounwind {
 ; MIPS64R6O0-NEXT:    lui $1, %hi(%neg(%gp_rel(AtomicLoadNand)))
 ; MIPS64R6O0-NEXT:    daddu $1, $1, $25
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicLoadNand)))
-; MIPS64R6O0-NEXT:    move $25, $4
+; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(x)($1)
 ; MIPS64R6O0-NEXT:  .LBB5_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    lld $2, 0($1)
-; MIPS64R6O0-NEXT:    and $3, $2, $4
-; MIPS64R6O0-NEXT:    nor $3, $zero, $3
-; MIPS64R6O0-NEXT:    scd $3, 0($1)
-; MIPS64R6O0-NEXT:    beqzc $3, .LBB5_1
+; MIPS64R6O0-NEXT:    lld $3, 0($1)
+; MIPS64R6O0-NEXT:    and $5, $3, $4
+; MIPS64R6O0-NEXT:    nor $5, $zero, $5
+; MIPS64R6O0-NEXT:    scd $5, 0($1)
+; MIPS64R6O0-NEXT:    beqzc $5, .LBB5_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    sd $25, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    sd $2, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    move $2, $3
 ; MIPS64R6O0-NEXT:    daddiu $sp, $sp, 16
 ; MIPS64R6O0-NEXT:    jrc $ra
 ;
@@ -1074,18 +1080,19 @@ define i64 @AtomicSwap64(i64 signext %newval) nounwind {
 ; MIPS64R6O0-NEXT:    lui $1, %hi(%neg(%gp_rel(AtomicSwap64)))
 ; MIPS64R6O0-NEXT:    daddu $1, $1, $25
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicSwap64)))
-; MIPS64R6O0-NEXT:    move $25, $4
+; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    sd $4, 8($sp)
-; MIPS64R6O0-NEXT:    ld $4, 8($sp)
+; MIPS64R6O0-NEXT:    ld $3, 8($sp)
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(x)($1)
 ; MIPS64R6O0-NEXT:  .LBB6_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    lld $2, 0($1)
-; MIPS64R6O0-NEXT:    move $3, $4
-; MIPS64R6O0-NEXT:    scd $3, 0($1)
-; MIPS64R6O0-NEXT:    beqzc $3, .LBB6_1
+; MIPS64R6O0-NEXT:    lld $4, 0($1)
+; MIPS64R6O0-NEXT:    move $5, $3
+; MIPS64R6O0-NEXT:    scd $5, 0($1)
+; MIPS64R6O0-NEXT:    beqzc $5, .LBB6_1
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
-; MIPS64R6O0-NEXT:    sd $25, 0($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    sd $2, 0($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    move $2, $4
 ; MIPS64R6O0-NEXT:    daddiu $sp, $sp, 16
 ; MIPS64R6O0-NEXT:    jrc $ra
 ;
@@ -1271,27 +1278,27 @@ define i64 @AtomicCmpSwap64(i64 signext %oldval, i64 signext %newval) nounwind {
 ; MIPS64R6O0-NEXT:    lui $1, %hi(%neg(%gp_rel(AtomicCmpSwap64)))
 ; MIPS64R6O0-NEXT:    daddu $1, $1, $25
 ; MIPS64R6O0-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(AtomicCmpSwap64)))
-; MIPS64R6O0-NEXT:    move $25, $5
-; MIPS64R6O0-NEXT:    move $2, $4
+; MIPS64R6O0-NEXT:    move $2, $5
+; MIPS64R6O0-NEXT:    move $3, $4
 ; MIPS64R6O0-NEXT:    sd $5, 40($sp)
 ; MIPS64R6O0-NEXT:    ld $5, 40($sp)
 ; MIPS64R6O0-NEXT:    ld $1, %got_disp(x)($1)
-; MIPS64R6O0-NEXT:    ld $3, 32($sp) # 8-byte Folded Reload
+; MIPS64R6O0-NEXT:    ld $6, 32($sp) # 8-byte Folded Reload
 ; MIPS64R6O0-NEXT:  .LBB7_1: # %entry
 ; MIPS64R6O0-NEXT:    # =>This Inner Loop Header: Depth=1
-; MIPS64R6O0-NEXT:    lld $6, 0($1)
-; MIPS64R6O0-NEXT:    bnec $6, $4, .LBB7_3
+; MIPS64R6O0-NEXT:    lld $7, 0($1)
+; MIPS64R6O0-NEXT:    bnec $7, $4, .LBB7_3
 ; MIPS64R6O0-NEXT:  # %bb.2: # %entry
 ; MIPS64R6O0-NEXT:    # in Loop: Header=BB7_1 Depth=1
-; MIPS64R6O0-NEXT:    move $7, $5
-; MIPS64R6O0-NEXT:    scd $7, 0($1)
-; MIPS64R6O0-NEXT:    beqzc $7, .LBB7_1
+; MIPS64R6O0-NEXT:    move $8, $5
+; MIPS64R6O0-NEXT:    scd $8, 0($1)
+; MIPS64R6O0-NEXT:    beqzc $8, .LBB7_1
 ; MIPS64R6O0-NEXT:  .LBB7_3: # %entry
 ; MIPS64R6O0-NEXT:    sd $2, 24($sp) # 8-byte Folded Spill
-; MIPS64R6O0-NEXT:    move $2, $6
-; MIPS64R6O0-NEXT:    sd $25, 16($sp) # 8-byte Folded Spill
-; MIPS64R6O0-NEXT:    sd $6, 32($sp) # 8-byte Folded Spill
-; MIPS64R6O0-NEXT:    sd $3, 8($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    move $2, $7
+; MIPS64R6O0-NEXT:    sd $3, 16($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    sd $7, 32($sp) # 8-byte Folded Spill
+; MIPS64R6O0-NEXT:    sd $6, 8($sp) # 8-byte Folded Spill
 ; MIPS64R6O0-NEXT:    daddiu $sp, $sp, 48
 ; MIPS64R6O0-NEXT:    jrc $ra
 ;

diff --git a/llvm/test/CodeGen/Mips/atomicCmpSwapPW.ll b/llvm/test/CodeGen/Mips/atomicCmpSwapPW.ll
index 973f3a5..bc4b489 100644 (file)
--- a/llvm/test/CodeGen/Mips/atomicCmpSwapPW.ll
+++ b/llvm/test/CodeGen/Mips/atomicCmpSwapPW.ll
@@ -79,29 +79,29 @@ define void @foo(i32 %new, i32 %old) {
 ; N64-NEXT:    sll $1, $1, 0
 ; N64-NEXT:    move $2, $4
 ; N64-NEXT:    sll $2, $2, 0
-; N64-NEXT:    lui $4, %highest(sym)
-; N64-NEXT:    daddiu $4, $4, %higher(sym)
-; N64-NEXT:    dsll $4, $4, 16
-; N64-NEXT:    daddiu $4, $4, %hi(sym)
-; N64-NEXT:    dsll $4, $4, 16
-; N64-NEXT:    ld $4, %lo(sym)($4)
+; N64-NEXT:    lui $3, %highest(sym)
+; N64-NEXT:    daddiu $3, $3, %higher(sym)
+; N64-NEXT:    dsll $3, $3, 16
+; N64-NEXT:    daddiu $3, $3, %hi(sym)
+; N64-NEXT:    dsll $3, $3, 16
+; N64-NEXT:    ld $3, %lo(sym)($3)
 ; N64-NEXT:    sync
-; N64-NEXT:    lw $3, 12($sp) # 4-byte Folded Reload
+; N64-NEXT:    lw $6, 12($sp) # 4-byte Folded Reload
 ; N64-NEXT:  .LBB0_1: # %entry
 ; N64-NEXT:    # =>This Inner Loop Header: Depth=1
-; N64-NEXT:    ll $6, 0($4)
-; N64-NEXT:    bne $6, $2, .LBB0_3
+; N64-NEXT:    ll $7, 0($3)
+; N64-NEXT:    bne $7, $2, .LBB0_3
 ; N64-NEXT:    nop
 ; N64-NEXT:  # %bb.2: # %entry
 ; N64-NEXT:    # in Loop: Header=BB0_1 Depth=1
-; N64-NEXT:    move $7, $1
-; N64-NEXT:    sc $7, 0($4)
-; N64-NEXT:    beqz $7, .LBB0_1
+; N64-NEXT:    move $8, $1
+; N64-NEXT:    sc $8, 0($3)
+; N64-NEXT:    beqz $8, .LBB0_1
 ; N64-NEXT:    nop
 ; N64-NEXT:  .LBB0_3: # %entry
 ; N64-NEXT:    sync
-; N64-NEXT:    sw $6, 12($sp) # 4-byte Folded Spill
-; N64-NEXT:    sw $3, 8($sp) # 4-byte Folded Spill
+; N64-NEXT:    sw $7, 12($sp) # 4-byte Folded Spill
+; N64-NEXT:    sw $6, 8($sp) # 4-byte Folded Spill
 ; N64-NEXT:    daddiu $sp, $sp, 16
 ; N64-NEXT:    jr $ra
 ; N64-NEXT:    nop

diff --git a/llvm/test/CodeGen/PowerPC/anon_aggr.ll b/llvm/test/CodeGen/PowerPC/anon_aggr.ll
index f5eb601..17f4ed4 100644 (file)
--- a/llvm/test/CodeGen/PowerPC/anon_aggr.ll
+++ b/llvm/test/CodeGen/PowerPC/anon_aggr.ll
@@ -38,11 +38,11 @@ unequal:
   ret i8* %array2_ptr
 }
 ; CHECK-LABEL: func2:
-; CHECK-DAG: cmpld {{([0-9]+,)?}}4, 6
+; CHECK-DAG: cmpld {{([0-9]+,)?}}4, 5
 ; CHECK-DAG: std 6, 72(1)
 ; CHECK-DAG: std 5, 64(1)
-; CHECK-DAG: std 6, -[[OFFSET1:[0-9]+]]
-; CHECK-DAG: std 5, -[[OFFSET2:[0-9]+]]
+; CHECK-DAG: std 5, -[[OFFSET1:[0-9]+]]
+; CHECK-DAG: std 3, -[[OFFSET2:[0-9]+]]
 ; CHECK: ld 3, -[[OFFSET2]](1)
 ; CHECK: ld 3, -[[OFFSET1]](1)
 

diff --git a/llvm/test/CodeGen/PowerPC/fp-int128-fp-combine.ll b/llvm/test/CodeGen/PowerPC/fp-int128-fp-combine.ll
index 1b403a6..6c96905 100644 (file)
--- a/llvm/test/CodeGen/PowerPC/fp-int128-fp-combine.ll
+++ b/llvm/test/CodeGen/PowerPC/fp-int128-fp-combine.ll
@@ -31,7 +31,8 @@ entry:
 define float @f_i128_fi_nsz(float %v) #0 {
 ; CHECK-LABEL: f_i128_fi_nsz:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    friz 1, 1
+; CHECK-NEXT:    friz 0, 1
+; CHECK-NEXT:    fmr 1, 0
 ; CHECK-NEXT:    blr
 entry:
   %a = fptosi float %v to i128

diff --git a/llvm/test/CodeGen/PowerPC/fp64-to-int16.ll b/llvm/test/CodeGen/PowerPC/fp64-to-int16.ll
index 8f11603..0048f91 100644 (file)
--- a/llvm/test/CodeGen/PowerPC/fp64-to-int16.ll
+++ b/llvm/test/CodeGen/PowerPC/fp64-to-int16.ll
@@ -5,8 +5,8 @@ target triple = "powerpc64le--linux-gnu"
 define i1 @Test(double %a) {
 ; CHECK-LABEL: Test:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    xscvdpsxws 1, 1
-; CHECK-NEXT:    mfvsrwz 3, 1
+; CHECK-NEXT:    xscvdpsxws 0, 1
+; CHECK-NEXT:    mfvsrwz 3, 0
 ; CHECK-NEXT:    xori 3, 3, 65534
 ; CHECK-NEXT:    cntlzw 3, 3
 ; CHECK-NEXT:    srwi 3, 3, 5

diff --git a/llvm/test/CodeGen/PowerPC/vsx.ll b/llvm/test/CodeGen/PowerPC/vsx.ll
index 4b1c2ed..e4e1ad6 100644 (file)
--- a/llvm/test/CodeGen/PowerPC/vsx.ll
+++ b/llvm/test/CodeGen/PowerPC/vsx.ll
@@ -1,3 +1,4 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -relocation-model=static -verify-machineinstrs -mcpu=pwr7 \
 ; RUN:     -mtriple=powerpc64-unknown-linux-gnu -mattr=+vsx \
 ; RUN:     -ppc-vsr-nums-as-vr -ppc-asm-full-reg-names < %s | FileCheck %s
@@ -15,1189 +16,2526 @@
 ; RUN:     -check-prefix=CHECK-LE %s
 
 define double @test1(double %a, double %b) {
+; CHECK-LABEL: test1:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xsmuldp f1, f1, f2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test1:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xsmuldp f1, f1, f2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test1:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xsmuldp f1, f1, f2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test1:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xsmuldp f1, f1, f2
+; CHECK-LE-NEXT:    blr
 entry:
   %v = fmul double %a, %b
   ret double %v
 
-; CHECK-LABEL: @test1
-; CHECK: xsmuldp f1, f1, f2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test1
-; CHECK-LE: xsmuldp f1, f1, f2
-; CHECK-LE: blr
 }
 
 define double @test2(double %a, double %b) {
+; CHECK-LABEL: test2:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xsdivdp f1, f1, f2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test2:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xsdivdp f1, f1, f2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test2:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xsdivdp f1, f1, f2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test2:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xsdivdp f1, f1, f2
+; CHECK-LE-NEXT:    blr
 entry:
   %v = fdiv double %a, %b
   ret double %v
 
-; CHECK-LABEL: @test2
-; CHECK: xsdivdp f1, f1, f2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test2
-; CHECK-LE: xsdivdp f1, f1, f2
-; CHECK-LE: blr
 }
 
 define double @test3(double %a, double %b) {
+; CHECK-LABEL: test3:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xsadddp f1, f1, f2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test3:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xsadddp f1, f1, f2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test3:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xsadddp f1, f1, f2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test3:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xsadddp f1, f1, f2
+; CHECK-LE-NEXT:    blr
 entry:
   %v = fadd double %a, %b
   ret double %v
 
-; CHECK-LABEL: @test3
-; CHECK: xsadddp f1, f1, f2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test3
-; CHECK-LE: xsadddp f1, f1, f2
-; CHECK-LE: blr
 }
 
 define <2 x double> @test4(<2 x double> %a, <2 x double> %b) {
+; CHECK-LABEL: test4:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xvadddp v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test4:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xvadddp v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test4:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xvadddp v2, v2, v3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test4:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xvadddp v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = fadd <2 x double> %a, %b
   ret <2 x double> %v
 
-; CHECK-LABEL: @test4
-; CHECK: xvadddp v2, v2, v3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test4
-; CHECK-LE: xvadddp v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <4 x i32> @test5(<4 x i32> %a, <4 x i32> %b) {
+; CHECK-LABEL: test5:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlxor v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test5:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlxor v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test5:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlxor v2, v2, v3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test5:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlxor v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = xor <4 x i32> %a, %b
   ret <4 x i32> %v
 
-; CHECK-REG-LABEL: @test5
-; CHECK-REG: xxlxor v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test5
-; CHECK-FISL: xxlxor v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test5
-; CHECK-LE: xxlxor v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <8 x i16> @test6(<8 x i16> %a, <8 x i16> %b) {
+; CHECK-LABEL: test6:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlxor v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test6:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlxor v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test6:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlxor vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test6:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlxor v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = xor <8 x i16> %a, %b
   ret <8 x i16> %v
 
-; CHECK-REG-LABEL: @test6
-; CHECK-REG: xxlxor v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test6
-; CHECK-FISL: xxlxor v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test6
-; CHECK-LE: xxlxor v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <16 x i8> @test7(<16 x i8> %a, <16 x i8> %b) {
+; CHECK-LABEL: test7:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlxor v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test7:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlxor v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test7:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlxor vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test7:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlxor v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = xor <16 x i8> %a, %b
   ret <16 x i8> %v
 
-; CHECK-REG-LABEL: @test7
-; CHECK-REG: xxlxor v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test7
-; CHECK-FISL: xxlxor v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test7
-; CHECK-LE: xxlxor v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <4 x i32> @test8(<4 x i32> %a, <4 x i32> %b) {
+; CHECK-LABEL: test8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlor v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test8:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlor v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test8:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlor v2, v2, v3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test8:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlor v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = or <4 x i32> %a, %b
   ret <4 x i32> %v
 
-; CHECK-REG-LABEL: @test8
-; CHECK-REG: xxlor v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test8
-; CHECK-FISL: xxlor v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test8
-; CHECK-LE: xxlor v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <8 x i16> @test9(<8 x i16> %a, <8 x i16> %b) {
+; CHECK-LABEL: test9:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlor v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test9:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlor v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test9:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlor vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test9:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlor v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = or <8 x i16> %a, %b
   ret <8 x i16> %v
 
-; CHECK-REG-LABEL: @test9
-; CHECK-REG: xxlor v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test9
-; CHECK-FISL: xxlor v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test9
-; CHECK-LE: xxlor v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <16 x i8> @test10(<16 x i8> %a, <16 x i8> %b) {
+; CHECK-LABEL: test10:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlor v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test10:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlor v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test10:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlor vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test10:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlor v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = or <16 x i8> %a, %b
   ret <16 x i8> %v
 
-; CHECK-REG-LABEL: @test10
-; CHECK-REG: xxlor v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test10
-; CHECK-FISL: xxlor v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test10
-; CHECK-LE: xxlor v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <4 x i32> @test11(<4 x i32> %a, <4 x i32> %b) {
+; CHECK-LABEL: test11:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxland v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test11:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxland v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test11:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxland v2, v2, v3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test11:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxland v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = and <4 x i32> %a, %b
   ret <4 x i32> %v
 
-; CHECK-REG-LABEL: @test11
-; CHECK-REG: xxland v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test11
-; CHECK-FISL: xxland v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test11
-; CHECK-LE: xxland v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <8 x i16> @test12(<8 x i16> %a, <8 x i16> %b) {
+; CHECK-LABEL: test12:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxland v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test12:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxland v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test12:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxland vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test12:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxland v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = and <8 x i16> %a, %b
   ret <8 x i16> %v
 
-; CHECK-REG-LABEL: @test12
-; CHECK-REG: xxland v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test12
-; CHECK-FISL: xxland v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test12
-; CHECK-LE: xxland v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <16 x i8> @test13(<16 x i8> %a, <16 x i8> %b) {
+; CHECK-LABEL: test13:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxland v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test13:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxland v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test13:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxland vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test13:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxland v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = and <16 x i8> %a, %b
   ret <16 x i8> %v
 
-; CHECK-REG-LABEL: @test13
-; CHECK-REG: xxland v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test13
-; CHECK-FISL: xxland v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test13
-; CHECK-LE: xxland v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <4 x i32> @test14(<4 x i32> %a, <4 x i32> %b) {
+; CHECK-LABEL: test14:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlnor v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test14:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlnor v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test14:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlor vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlnor v2, v2, v3
+; CHECK-FISL-NEXT:    li r3, -16
+; CHECK-FISL-NEXT:    stxvd2x vs0, r1, r3 # 16-byte Folded Spill
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test14:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlnor v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = or <4 x i32> %a, %b
   %w = xor <4 x i32> %v, <i32 -1, i32 -1, i32 -1, i32 -1>
   ret <4 x i32> %w
 
-; CHECK-REG-LABEL: @test14
-; CHECK-REG: xxlnor v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test14
-; CHECK-FISL: xxlor vs0, v2, v3
-; CHECK-FISL: xxlnor v2, v2, v3
-; CHECK-FISL-NOT: lis
-; CHECK-FISL-NOT: ori
-; CHECK-FISL: li r3, -16
-; CHECK-FISL-NOT: lis
-; CHECK-FISL-NOT: ori
-; CHECK-FISL: stxvd2x vs0, r1, r3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test14
-; CHECK-LE: xxlnor v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <8 x i16> @test15(<8 x i16> %a, <8 x i16> %b) {
+; CHECK-LABEL: test15:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlnor v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test15:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlnor v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test15:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlor vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v4, vs0, vs0
+; CHECK-FISL-NEXT:    xxlnor vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    li r3, -16
+; CHECK-FISL-NEXT:    stxvd2x v4, r1, r3 # 16-byte Folded Spill
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test15:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlnor v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = or <8 x i16> %a, %b
   %w = xor <8 x i16> %v, <i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1>
   ret <8 x i16> %w
 
-; CHECK-REG-LABEL: @test15
-; CHECK-REG: xxlnor v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test15
-; CHECK-FISL: xxlor vs0, v2, v3
-; CHECK-FISL: xxlor v4, vs0, vs0
-; CHECK-FISL: xxlnor vs0, v2, v3
-; CHECK-FISL: xxlor v2, vs0, vs0
-; CHECK-FISL-NOT: lis
-; CHECK-FISL-NOT: ori
-; CHECK-FISL: li r3, -16
-; CHECK-FISL-NOT: lis
-; CHECK-FISL-NOT: ori
-; CHECK-FISL: stxvd2x v4, r1, r3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test15
-; CHECK-LE: xxlnor v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <16 x i8> @test16(<16 x i8> %a, <16 x i8> %b) {
+; CHECK-LABEL: test16:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlnor v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test16:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlnor v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test16:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlor vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v4, vs0, vs0
+; CHECK-FISL-NEXT:    xxlnor vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    li r3, -16
+; CHECK-FISL-NEXT:    stxvd2x v4, r1, r3 # 16-byte Folded Spill
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test16:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlnor v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %v = or <16 x i8> %a, %b
   %w = xor <16 x i8> %v, <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
   ret <16 x i8> %w
 
-; CHECK-REG-LABEL: @test16
-; CHECK-REG: xxlnor v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test16
-; CHECK-FISL: xxlor vs0, v2, v3
-; CHECK-FISL: xxlor v4, vs0, vs0
-; CHECK-FISL: xxlnor vs0, v2, v3
-; CHECK-FISL: xxlor v2, vs0, vs0
-; CHECK-FISL-NOT: lis
-; CHECK-FISL-NOT: ori
-; CHECK-FISL: li r3, -16
-; CHECK-FISL-NOT: lis
-; CHECK-FISL-NOT: ori
-; CHECK-FISL: stxvd2x v4, r1, r3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test16
-; CHECK-LE: xxlnor v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <4 x i32> @test17(<4 x i32> %a, <4 x i32> %b) {
+; CHECK-LABEL: test17:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlandc v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test17:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlandc v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test17:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlnor vs0, v3, v3
+; CHECK-FISL-NEXT:    xxland v2, v2, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test17:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlandc v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %w = xor <4 x i32> %b, <i32 -1, i32 -1, i32 -1, i32 -1>
   %v = and <4 x i32> %a, %w
   ret <4 x i32> %v
 
-; CHECK-REG-LABEL: @test17
-; CHECK-REG: xxlandc v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test17
-; CHECK-FISL: xxlnor v3, v3, v3
-; CHECK-FISL: xxland v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test17
-; CHECK-LE: xxlandc v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <8 x i16> @test18(<8 x i16> %a, <8 x i16> %b) {
+; CHECK-LABEL: test18:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlandc v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test18:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlandc v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test18:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlnor vs0, v3, v3
+; CHECK-FISL-NEXT:    xxlor v4, vs0, vs0
+; CHECK-FISL-NEXT:    xxlandc vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    li r3, -16
+; CHECK-FISL-NEXT:    stxvd2x v4, r1, r3 # 16-byte Folded Spill
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test18:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlandc v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %w = xor <8 x i16> %b, <i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1>
   %v = and <8 x i16> %a, %w
   ret <8 x i16> %v
 
-; CHECK-REG-LABEL: @test18
-; CHECK-REG: xxlandc v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test18
-; CHECK-FISL: xxlnor vs0, v3, v3
-; CHECK-FISL: xxlor v4, vs0, vs0
-; CHECK-FISL: xxlandc vs0, v2, v3
-; CHECK-FISL: xxlor v2, vs0, vs0
-; CHECK-FISL-NOT: lis
-; CHECK-FISL-NOT: ori
-; CHECK-FISL: li r3, -16
-; CHECK-FISL-NOT: lis
-; CHECK-FISL-NOT: ori
-; CHECK-FISL: stxvd2x v4, r1, r3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test18
-; CHECK-LE: xxlandc v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <16 x i8> @test19(<16 x i8> %a, <16 x i8> %b) {
+; CHECK-LABEL: test19:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xxlandc v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test19:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xxlandc v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test19:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xxlnor vs0, v3, v3
+; CHECK-FISL-NEXT:    xxlor v4, vs0, vs0
+; CHECK-FISL-NEXT:    xxlandc vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    li r3, -16
+; CHECK-FISL-NEXT:    stxvd2x v4, r1, r3 # 16-byte Folded Spill
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test19:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xxlandc v2, v2, v3
+; CHECK-LE-NEXT:    blr
 entry:
   %w = xor <16 x i8> %b, <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
   %v = and <16 x i8> %a, %w
   ret <16 x i8> %v
 
-; CHECK-REG-LABEL: @test19
-; CHECK-REG: xxlandc v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test19
-; CHECK-FISL: xxlnor vs0, v3, v3
-; CHECK-FISL: xxlor v4, vs0, vs0
-; CHECK-FISL: xxlandc vs0, v2, v3
-; CHECK-FISL: xxlor v2, vs0, vs0
-; CHECK-FISL-NOT: lis
-; CHECK-FISL-NOT: ori
-; CHECK-FISL: li r3, -16
-; CHECK-FISL-NOT: lis
-; CHECK-FISL-NOT: ori
-; CHECK-FISL: stxvd2x v4, r1, r3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test19
-; CHECK-LE: xxlandc v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <4 x i32> @test20(<4 x i32> %a, <4 x i32> %b, <4 x i32> %c, <4 x i32> %d) {
+; CHECK-LABEL: test20:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vcmpequw v4, v4, v5
+; CHECK-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test20:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    vcmpequw v4, v4, v5
+; CHECK-REG-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test20:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    vcmpequw v4, v4, v5
+; CHECK-FISL-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test20:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    vcmpequw v4, v4, v5
+; CHECK-LE-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-LE-NEXT:    blr
 entry:
   %m = icmp eq <4 x i32> %c, %d
   %v = select <4 x i1> %m, <4 x i32> %a, <4 x i32> %b
   ret <4 x i32> %v
 
-; CHECK-REG-LABEL: @test20
-; CHECK-REG: vcmpequw v4, v4, v5
-; CHECK-REG: xxsel v2, v3, v2, v4
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test20
-; CHECK-FISL: vcmpequw v4, v4, v5
-; CHECK-FISL: xxsel v2, v3, v2, v4
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test20
-; CHECK-LE: vcmpequw v4, v4, v5
-; CHECK-LE: xxsel v2, v3, v2, v4
-; CHECK-LE: blr
 }
 
 define <4 x float> @test21(<4 x float> %a, <4 x float> %b, <4 x float> %c, <4 x float> %d) {
+; CHECK-LABEL: test21:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xvcmpeqsp vs0, v4, v5
+; CHECK-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test21:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xvcmpeqsp vs0, v4, v5
+; CHECK-REG-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test21:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xvcmpeqsp vs0, v4, v5
+; CHECK-FISL-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test21:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xvcmpeqsp vs0, v4, v5
+; CHECK-LE-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-LE-NEXT:    blr
 entry:
   %m = fcmp oeq <4 x float> %c, %d
   %v = select <4 x i1> %m, <4 x float> %a, <4 x float> %b
   ret <4 x float> %v
 
-; CHECK-REG-LABEL: @test21
-; CHECK-REG: xvcmpeqsp vs0, v4, v5
-; CHECK-REG: xxsel v2, v3, v2, vs0
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test21
-; CHECK-FISL: xvcmpeqsp v4, v4, v5
-; CHECK-FISL: xxsel v2, v3, v2, v4
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test21
-; CHECK-LE: xvcmpeqsp vs0, v4, v5
-; CHECK-LE: xxsel v2, v3, v2, vs0
-; CHECK-LE: blr
 }
 
 define <4 x float> @test22(<4 x float> %a, <4 x float> %b, <4 x float> %c, <4 x float> %d) {
+; CHECK-LABEL: test22:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xvcmpeqsp vs0, v5, v5
+; CHECK-NEXT:    xvcmpeqsp vs1, v4, v4
+; CHECK-NEXT:    xvcmpeqsp vs2, v4, v5
+; CHECK-NEXT:    xxlnor vs0, vs0, vs0
+; CHECK-NEXT:    xxlnor vs1, vs1, vs1
+; CHECK-NEXT:    xxlor vs0, vs1, vs0
+; CHECK-NEXT:    xxlor vs0, vs2, vs0
+; CHECK-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test22:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xvcmpeqsp vs0, v5, v5
+; CHECK-REG-NEXT:    xvcmpeqsp vs1, v4, v4
+; CHECK-REG-NEXT:    xvcmpeqsp vs2, v4, v5
+; CHECK-REG-NEXT:    xxlnor vs0, vs0, vs0
+; CHECK-REG-NEXT:    xxlnor vs1, vs1, vs1
+; CHECK-REG-NEXT:    xxlor vs0, vs1, vs0
+; CHECK-REG-NEXT:    xxlor vs0, vs2, vs0
+; CHECK-REG-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test22:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xvcmpeqsp vs0, v4, v5
+; CHECK-FISL-NEXT:    xvcmpeqsp vs1, v5, v5
+; CHECK-FISL-NEXT:    xxlnor vs1, vs1, vs1
+; CHECK-FISL-NEXT:    xvcmpeqsp vs2, v4, v4
+; CHECK-FISL-NEXT:    xxlnor vs2, vs2, vs2
+; CHECK-FISL-NEXT:    xxlor vs1, vs2, vs1
+; CHECK-FISL-NEXT:    xxlor vs0, vs0, vs1
+; CHECK-FISL-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test22:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xvcmpeqsp vs0, v5, v5
+; CHECK-LE-NEXT:    xvcmpeqsp vs1, v4, v4
+; CHECK-LE-NEXT:    xvcmpeqsp vs2, v4, v5
+; CHECK-LE-NEXT:    xxlnor vs0, vs0, vs0
+; CHECK-LE-NEXT:    xxlnor vs1, vs1, vs1
+; CHECK-LE-NEXT:    xxlor vs0, vs1, vs0
+; CHECK-LE-NEXT:    xxlor vs0, vs2, vs0
+; CHECK-LE-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-LE-NEXT:    blr
 entry:
   %m = fcmp ueq <4 x float> %c, %d
   %v = select <4 x i1> %m, <4 x float> %a, <4 x float> %b
   ret <4 x float> %v
 
-; CHECK-REG-LABEL: @test22
-; CHECK-REG-DAG: xvcmpeqsp vs0, v5, v5
-; CHECK-REG-DAG: xvcmpeqsp vs1, v4, v4
-; CHECK-REG-DAG: xvcmpeqsp vs2, v4, v5
-; CHECK-REG-DAG: xxlnor vs0, vs0, vs0
-; CHECK-REG-DAG: xxlnor vs1, vs1, vs1
-; CHECK-REG-DAG: xxlor vs0, vs1, vs0
-; CHECK-REG-DAG: xxlor vs0, vs2, vs0
-; CHECK-REG: xxsel v2, v3, v2, vs0
-; CHECK-REG: blr
-
-; CHECK-FISL-LABEL: @test22
-; CHECK-FISL-DAG: xvcmpeqsp vs0, v4, v5
-; CHECK-FISL-DAG: xvcmpeqsp v5, v5, v5
-; CHECK-FISL-DAG: xvcmpeqsp v4, v4, v4
-; CHECK-FISL-DAG: xxlnor v5, v5, v5
-; CHECK-FISL-DAG: xxlnor v4, v4, v4
-; CHECK-FISL-DAG: xxlor v4, v4, v5
-; CHECK-FISL-DAG: xxlor vs0, vs0, v4
-; CHECK-FISL: xxsel v2, v3, v2, vs0
-; CHECK-FISL: blr
-
-; CHECK-LE-LABEL: @test22
-; CHECK-LE-DAG: xvcmpeqsp vs0, v5, v5
-; CHECK-LE-DAG: xvcmpeqsp vs1, v4, v4
-; CHECK-LE-DAG: xvcmpeqsp vs2, v4, v5
-; CHECK-LE-DAG: xxlnor vs0, vs0, vs0
-; CHECK-LE-DAG: xxlnor vs1, vs1, vs1
-; CHECK-LE-DAG: xxlor vs0, vs1, vs0
-; CHECK-LE-DAG: xxlor vs0, vs2, vs0
-; CHECK-LE: xxsel v2, v3, v2, vs0
-; CHECK-LE: blr
+
+
 }
 
 define <8 x i16> @test23(<8 x i16> %a, <8 x i16> %b, <8 x i16> %c, <8 x i16> %d) {
+; CHECK-LABEL: test23:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vcmpequh v4, v4, v5
+; CHECK-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test23:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    vcmpequh v4, v4, v5
+; CHECK-REG-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test23:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    vcmpequh v4, v4, v5
+; CHECK-FISL-NEXT:    xxsel vs0, v3, v2, v4
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test23:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    vcmpequh v4, v4, v5
+; CHECK-LE-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-LE-NEXT:    blr
 entry:
   %m = icmp eq <8 x i16> %c, %d
   %v = select <8 x i1> %m, <8 x i16> %a, <8 x i16> %b
   ret <8 x i16> %v
 
-; CHECK-REG-LABEL: @test23
-; CHECK-REG: vcmpequh v4, v4, v5
-; CHECK-REG: xxsel v2, v3, v2, v4
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test23
-; CHECK-FISL: vcmpequh v4, v4, v5
-; CHECK-FISL: xxsel v2, v3, v2, v4
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test23
-; CHECK-LE: vcmpequh v4, v4, v5
-; CHECK-LE: xxsel v2, v3, v2, v4
-; CHECK-LE: blr
 }
 
 define <16 x i8> @test24(<16 x i8> %a, <16 x i8> %b, <16 x i8> %c, <16 x i8> %d) {
+; CHECK-LABEL: test24:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vcmpequb v4, v4, v5
+; CHECK-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test24:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    vcmpequb v4, v4, v5
+; CHECK-REG-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test24:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    vcmpequb v4, v4, v5
+; CHECK-FISL-NEXT:    xxsel vs0, v3, v2, v4
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test24:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    vcmpequb v4, v4, v5
+; CHECK-LE-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-LE-NEXT:    blr
 entry:
   %m = icmp eq <16 x i8> %c, %d
   %v = select <16 x i1> %m, <16 x i8> %a, <16 x i8> %b
   ret <16 x i8> %v
 
-; CHECK-REG-LABEL: @test24
-; CHECK-REG: vcmpequb v4, v4, v5
-; CHECK-REG: xxsel v2, v3, v2, v4
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test24
-; CHECK-FISL: vcmpequb v4, v4, v5
-; CHECK-FISL: xxsel v2, v3, v2, v4
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test24
-; CHECK-LE: vcmpequb v4, v4, v5
-; CHECK-LE: xxsel v2, v3, v2, v4
-; CHECK-LE: blr
 }
 
 define <2 x double> @test25(<2 x double> %a, <2 x double> %b, <2 x double> %c, <2 x double> %d) {
+; CHECK-LABEL: test25:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xvcmpeqdp vs0, v4, v5
+; CHECK-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test25:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xvcmpeqdp vs0, v4, v5
+; CHECK-REG-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test25:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xvcmpeqdp vs0, v4, v5
+; CHECK-FISL-NEXT:    xxsel v2, v3, v2, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test25:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xvcmpeqdp v4, v4, v5
+; CHECK-LE-NEXT:    xxsel v2, v3, v2, v4
+; CHECK-LE-NEXT:    blr
 entry:
   %m = fcmp oeq <2 x double> %c, %d
   %v = select <2 x i1> %m, <2 x double> %a, <2 x double> %b
   ret <2 x double> %v
 
-; CHECK-LABEL: @test25
-; CHECK: xvcmpeqdp vs0, v4, v5
-; CHECK: xxsel v2, v3, v2, vs0
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test25
-; CHECK-LE: xvcmpeqdp v4, v4, v5
-; CHECK-LE: xxsel v2, v3, v2, v4
-; CHECK-LE: blr
 }
 
 define <2 x i64> @test26(<2 x i64> %a, <2 x i64> %b) {
+; CHECK-LABEL: test26:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi r3, r1, -32
+; CHECK-NEXT:    addi r4, r1, -48
+; CHECK-NEXT:    stxvd2x v3, 0, r3
+; CHECK-NEXT:    stxvd2x v2, 0, r4
+; CHECK-NEXT:    ld r3, -24(r1)
+; CHECK-NEXT:    ld r4, -40(r1)
+; CHECK-NEXT:    add r3, r4, r3
+; CHECK-NEXT:    std r3, -8(r1)
+; CHECK-NEXT:    ld r3, -32(r1)
+; CHECK-NEXT:    ld r4, -48(r1)
+; CHECK-NEXT:    add r3, r4, r3
+; CHECK-NEXT:    std r3, -16(r1)
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test26:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addi r3, r1, -32
+; CHECK-REG-NEXT:    addi r4, r1, -48
+; CHECK-REG-NEXT:    stxvd2x v3, 0, r3
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r4
+; CHECK-REG-NEXT:    ld r3, -24(r1)
+; CHECK-REG-NEXT:    ld r4, -40(r1)
+; CHECK-REG-NEXT:    add r3, r4, r3
+; CHECK-REG-NEXT:    std r3, -8(r1)
+; CHECK-REG-NEXT:    ld r3, -32(r1)
+; CHECK-REG-NEXT:    ld r4, -48(r1)
+; CHECK-REG-NEXT:    add r3, r4, r3
+; CHECK-REG-NEXT:    std r3, -16(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test26:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addi r3, r1, -32
+; CHECK-FISL-NEXT:    stxvd2x v3, 0, r3
+; CHECK-FISL-NEXT:    addi r3, r1, -48
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    ld r3, -24(r1)
+; CHECK-FISL-NEXT:    ld r4, -40(r1)
+; CHECK-FISL-NEXT:    add r3, r4, r3
+; CHECK-FISL-NEXT:    std r3, -8(r1)
+; CHECK-FISL-NEXT:    ld r3, -32(r1)
+; CHECK-FISL-NEXT:    ld r4, -48(r1)
+; CHECK-FISL-NEXT:    add r3, r4, r3
+; CHECK-FISL-NEXT:    std r3, -16(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test26:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    vaddudm v2, v2, v3
+; CHECK-LE-NEXT:    blr
   %v = add <2 x i64> %a, %b
   ret <2 x i64> %v
 
-; CHECK-LABEL: @test26
 
 ; Make sure we use only two stores (one for each operand).
-; CHECK: stxvd2x v3, 0, r3
-; CHECK: stxvd2x v2, 0, r4
-; CHECK-NOT: stxvd2x
 
 ; FIXME: The code quality here is not good; just make sure we do something for now.
-; CHECK: add r3, r4, r3
-; CHECK: add r3, r4, r3
-; CHECK: blr
 
-; CHECK-LE: vaddudm v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <2 x i64> @test27(<2 x i64> %a, <2 x i64> %b) {
+; CHECK-LABEL: test27:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xxland v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test27:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xxland v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test27:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xxland vs0, v2, v3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test27:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxland v2, v2, v3
+; CHECK-LE-NEXT:    blr
   %v = and <2 x i64> %a, %b
   ret <2 x i64> %v
 
-; CHECK-LABEL: @test27
-; CHECK: xxland v2, v2, v3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test27
-; CHECK-LE: xxland v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <2 x double> @test28(<2 x double>* %a) {
+; CHECK-LABEL: test28:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test28:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test28:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    lxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test28:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-LE-NEXT:    xxswapd v2, vs0
+; CHECK-LE-NEXT:    blr
   %v = load <2 x double>, <2 x double>* %a, align 16
   ret <2 x double> %v
 
-; CHECK-LABEL: @test28
-; CHECK: lxvd2x v2, 0, r3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test28
-; CHECK-LE: lxvd2x vs0, 0, r3
-; CHECK-LE: xxswapd v2, vs0
-; CHECK-LE: blr
 }
 
 define void @test29(<2 x double>* %a, <2 x double> %b) {
+; CHECK-LABEL: test29:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    stxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test29:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test29:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test29:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxswapd vs0, v2
+; CHECK-LE-NEXT:    stxvd2x vs0, 0, r3
+; CHECK-LE-NEXT:    blr
   store <2 x double> %b, <2 x double>* %a, align 16
   ret void
 
-; CHECK-LABEL: @test29
-; CHECK: stxvd2x v2, 0, r3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test29
-; CHECK-LE: xxswapd vs0, v2
-; CHECK-LE: stxvd2x vs0, 0, r3
-; CHECK-LE: blr
 }
 
 define <2 x double> @test28u(<2 x double>* %a) {
+; CHECK-LABEL: test28u:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test28u:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test28u:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    lxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test28u:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-LE-NEXT:    xxswapd v2, vs0
+; CHECK-LE-NEXT:    blr
   %v = load <2 x double>, <2 x double>* %a, align 8
   ret <2 x double> %v
 
-; CHECK-LABEL: @test28u
-; CHECK: lxvd2x v2, 0, r3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test28u
-; CHECK-LE: lxvd2x vs0, 0, r3
-; CHECK-LE: xxswapd v2, vs0
-; CHECK-LE: blr
 }
 
 define void @test29u(<2 x double>* %a, <2 x double> %b) {
+; CHECK-LABEL: test29u:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    stxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test29u:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test29u:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test29u:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxswapd vs0, v2
+; CHECK-LE-NEXT:    stxvd2x vs0, 0, r3
+; CHECK-LE-NEXT:    blr
   store <2 x double> %b, <2 x double>* %a, align 8
   ret void
 
-; CHECK-LABEL: @test29u
-; CHECK: stxvd2x v2, 0, r3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test29u
-; CHECK-LE: xxswapd vs0, v2
-; CHECK-LE: stxvd2x vs0, 0, r3
-; CHECK-LE: blr
 }
 
 define <2 x i64> @test30(<2 x i64>* %a) {
+; CHECK-LABEL: test30:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test30:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test30:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test30:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-LE-NEXT:    xxswapd v2, vs0
+; CHECK-LE-NEXT:    blr
   %v = load <2 x i64>, <2 x i64>* %a, align 16
   ret <2 x i64> %v
 
-; CHECK-REG-LABEL: @test30
-; CHECK-REG: lxvd2x v2, 0, r3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test30
-; CHECK-FISL: lxvd2x vs0, 0, r3
-; CHECK-FISL: xxlor v2, vs0, vs0
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test30
-; CHECK-LE: lxvd2x vs0, 0, r3
-; CHECK-LE: xxswapd v2, vs0
-; CHECK-LE: blr
 }
 
 define void @test31(<2 x i64>* %a, <2 x i64> %b) {
+; CHECK-LABEL: test31:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    stxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test31:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test31:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test31:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxswapd vs0, v2
+; CHECK-LE-NEXT:    stxvd2x vs0, 0, r3
+; CHECK-LE-NEXT:    blr
   store <2 x i64> %b, <2 x i64>* %a, align 16
   ret void
 
-; CHECK-LABEL: @test31
-; CHECK: stxvd2x v2, 0, r3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test31
-; CHECK-LE: xxswapd vs0, v2
-; CHECK-LE: stxvd2x vs0, 0, r3
-; CHECK-LE: blr
 }
 
 define <4 x float> @test32(<4 x float>* %a) {
+; CHECK-LABEL: test32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lxvw4x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test32:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    lxvw4x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test32:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    lxvw4x v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test32:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    lvx v2, 0, r3
+; CHECK-LE-NEXT:    blr
   %v = load <4 x float>, <4 x float>* %a, align 16
   ret <4 x float> %v
 
-; CHECK-REG-LABEL: @test32
-; CHECK-REG: lxvw4x v2, 0, r3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test32
-; CHECK-FISL: lxvw4x v2, 0, r3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test32
-; CHECK-LE: lvx v2, 0, r3
-; CHECK-LE-NOT: xxswapd
-; CHECK-LE: blr
 }
 
 define void @test33(<4 x float>* %a, <4 x float> %b) {
+; CHECK-LABEL: test33:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    stxvw4x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test33:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    stxvw4x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test33:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    stxvw4x v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test33:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    stvx v2, 0, r3
+; CHECK-LE-NEXT:    blr
   store <4 x float> %b, <4 x float>* %a, align 16
   ret void
 
-; CHECK-REG-LABEL: @test33
-; CHECK-REG: stxvw4x v2, 0, r3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test33
-; CHECK-FISL: stxvw4x v2, 0, r3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test33
-; CHECK-LE-NOT: xxswapd
-; CHECK-LE: stvx v2, 0, r3
-; CHECK-LE: blr
 }
 
 define <4 x float> @test32u(<4 x float>* %a) {
+; CHECK-LABEL: test32u:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    li r4, 15
+; CHECK-NEXT:    lvsl v3, 0, r3
+; CHECK-NEXT:    lvx v2, r3, r4
+; CHECK-NEXT:    lvx v4, 0, r3
+; CHECK-NEXT:    vperm v2, v4, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test32u:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    li r4, 15
+; CHECK-REG-NEXT:    lvsl v3, 0, r3
+; CHECK-REG-NEXT:    lvx v2, r3, r4
+; CHECK-REG-NEXT:    lvx v4, 0, r3
+; CHECK-REG-NEXT:    vperm v2, v4, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test32u:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    li r4, 15
+; CHECK-FISL-NEXT:    lvx v2, r3, r4
+; CHECK-FISL-NEXT:    lvsl v3, 0, r3
+; CHECK-FISL-NEXT:    lvx v4, 0, r3
+; CHECK-FISL-NEXT:    vperm v2, v4, v2, v3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test32u:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-LE-NEXT:    xxswapd v2, vs0
+; CHECK-LE-NEXT:    blr
   %v = load <4 x float>, <4 x float>* %a, align 8
   ret <4 x float> %v
 
-; CHECK-LABEL: @test32u
-; CHECK-DAG: lvsl v3, 0, r3
-; CHECK-DAG: lvx v2, r3, r4
-; CHECK-DAG: lvx v4, 0, r3
-; CHECK: vperm v2, v4, v2, v3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test32u
-; CHECK-LE: lxvd2x vs0, 0, r3
-; CHECK-LE: xxswapd v2, vs0
-; CHECK-LE: blr
 }
 
 define void @test33u(<4 x float>* %a, <4 x float> %b) {
+; CHECK-LABEL: test33u:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    stxvw4x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test33u:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    stxvw4x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test33u:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    stxvw4x v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test33u:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxswapd vs0, v2
+; CHECK-LE-NEXT:    stxvd2x vs0, 0, r3
+; CHECK-LE-NEXT:    blr
   store <4 x float> %b, <4 x float>* %a, align 8
   ret void
 
-; CHECK-REG-LABEL: @test33u
-; CHECK-REG: stxvw4x v2, 0, r3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test33u
-; CHECK-FISL: stxvw4x v2, 0, r3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test33u
-; CHECK-LE: xxswapd vs0, v2
-; CHECK-LE: stxvd2x vs0, 0, r3
-; CHECK-LE: blr
 }
 
 define <4 x i32> @test34(<4 x i32>* %a) {
+; CHECK-LABEL: test34:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lxvw4x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test34:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    lxvw4x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test34:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    lxvw4x v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test34:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    lvx v2, 0, r3
+; CHECK-LE-NEXT:    blr
   %v = load <4 x i32>, <4 x i32>* %a, align 16
   ret <4 x i32> %v
 
-; CHECK-REG-LABEL: @test34
-; CHECK-REG: lxvw4x v2, 0, r3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test34
-; CHECK-FISL: lxvw4x v2, 0, r3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test34
-; CHECK-LE: lvx v2, 0, r3
-; CHECK-LE-NOT: xxswapd
-; CHECK-LE: blr
 }
 
 define void @test35(<4 x i32>* %a, <4 x i32> %b) {
+; CHECK-LABEL: test35:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    stxvw4x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test35:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    stxvw4x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test35:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    stxvw4x v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test35:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    stvx v2, 0, r3
+; CHECK-LE-NEXT:    blr
   store <4 x i32> %b, <4 x i32>* %a, align 16
   ret void
 
-; CHECK-REG-LABEL: @test35
-; CHECK-REG: stxvw4x v2, 0, r3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test35
-; CHECK-FISL: stxvw4x v2, 0, r3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test35
-; CHECK-LE-NOT: xxswapd
-; CHECK-LE: stvx v2, 0, r3
-; CHECK-LE: blr
 }
 
 define <2 x double> @test40(<2 x i64> %a) {
+; CHECK-LABEL: test40:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xvcvuxddp v2, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test40:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xvcvuxddp v2, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test40:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xvcvuxddp v2, v2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test40:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xvcvuxddp v2, v2
+; CHECK-LE-NEXT:    blr
   %v = uitofp <2 x i64> %a to <2 x double>
   ret <2 x double> %v
 
-; CHECK-LABEL: @test40
-; CHECK: xvcvuxddp v2, v2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test40
-; CHECK-LE: xvcvuxddp v2, v2
-; CHECK-LE: blr
 }
 
 define <2 x double> @test41(<2 x i64> %a) {
+; CHECK-LABEL: test41:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xvcvsxddp v2, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test41:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xvcvsxddp v2, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test41:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xvcvsxddp v2, v2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test41:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xvcvsxddp v2, v2
+; CHECK-LE-NEXT:    blr
   %v = sitofp <2 x i64> %a to <2 x double>
   ret <2 x double> %v
 
-; CHECK-LABEL: @test41
-; CHECK: xvcvsxddp v2, v2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test41
-; CHECK-LE: xvcvsxddp v2, v2
-; CHECK-LE: blr
 }
 
 define <2 x i64> @test42(<2 x double> %a) {
+; CHECK-LABEL: test42:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xvcvdpuxds v2, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test42:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xvcvdpuxds v2, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test42:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xvcvdpuxds v2, v2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test42:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xvcvdpuxds v2, v2
+; CHECK-LE-NEXT:    blr
   %v = fptoui <2 x double> %a to <2 x i64>
   ret <2 x i64> %v
 
-; CHECK-LABEL: @test42
-; CHECK: xvcvdpuxds v2, v2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test42
-; CHECK-LE: xvcvdpuxds v2, v2
-; CHECK-LE: blr
 }
 
 define <2 x i64> @test43(<2 x double> %a) {
+; CHECK-LABEL: test43:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xvcvdpsxds v2, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test43:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xvcvdpsxds v2, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test43:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xvcvdpsxds v2, v2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test43:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xvcvdpsxds v2, v2
+; CHECK-LE-NEXT:    blr
   %v = fptosi <2 x double> %a to <2 x i64>
   ret <2 x i64> %v
 
-; CHECK-LABEL: @test43
-; CHECK: xvcvdpsxds v2, v2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test43
-; CHECK-LE: xvcvdpsxds v2, v2
-; CHECK-LE: blr
 }
 
 define <2 x float> @test44(<2 x i64> %a) {
+; CHECK-LABEL: test44:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    addi r4, r1, -64
+; CHECK-NEXT:    stxvd2x v2, 0, r3
+; CHECK-NEXT:    ld r3, -8(r1)
+; CHECK-NEXT:    std r3, -24(r1)
+; CHECK-NEXT:    ld r3, -16(r1)
+; CHECK-NEXT:    std r3, -32(r1)
+; CHECK-NEXT:    lfd f0, -24(r1)
+; CHECK-NEXT:    fcfidus f0, f0
+; CHECK-NEXT:    stfs f0, -48(r1)
+; CHECK-NEXT:    lfd f0, -32(r1)
+; CHECK-NEXT:    addi r3, r1, -48
+; CHECK-NEXT:    fcfidus f0, f0
+; CHECK-NEXT:    stfs f0, -64(r1)
+; CHECK-NEXT:    lxvw4x v2, 0, r3
+; CHECK-NEXT:    lxvw4x v3, 0, r4
+; CHECK-NEXT:    vmrghw v2, v3, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test44:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    addi r4, r1, -64
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    ld r3, -8(r1)
+; CHECK-REG-NEXT:    std r3, -24(r1)
+; CHECK-REG-NEXT:    ld r3, -16(r1)
+; CHECK-REG-NEXT:    std r3, -32(r1)
+; CHECK-REG-NEXT:    lfd f0, -24(r1)
+; CHECK-REG-NEXT:    fcfidus f0, f0
+; CHECK-REG-NEXT:    stfs f0, -48(r1)
+; CHECK-REG-NEXT:    lfd f0, -32(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -48
+; CHECK-REG-NEXT:    fcfidus f0, f0
+; CHECK-REG-NEXT:    stfs f0, -64(r1)
+; CHECK-REG-NEXT:    lxvw4x v2, 0, r3
+; CHECK-REG-NEXT:    lxvw4x v3, 0, r4
+; CHECK-REG-NEXT:    vmrghw v2, v3, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test44:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    ld r3, -8(r1)
+; CHECK-FISL-NEXT:    std r3, -24(r1)
+; CHECK-FISL-NEXT:    ld r3, -16(r1)
+; CHECK-FISL-NEXT:    std r3, -32(r1)
+; CHECK-FISL-NEXT:    lfd f0, -24(r1)
+; CHECK-FISL-NEXT:    fcfidus f0, f0
+; CHECK-FISL-NEXT:    stfs f0, -48(r1)
+; CHECK-FISL-NEXT:    lfd f0, -32(r1)
+; CHECK-FISL-NEXT:    fcfidus f0, f0
+; CHECK-FISL-NEXT:    stfs f0, -64(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -48
+; CHECK-FISL-NEXT:    lxvw4x v2, 0, r3
+; CHECK-FISL-NEXT:    addi r3, r1, -64
+; CHECK-FISL-NEXT:    lxvw4x v3, 0, r3
+; CHECK-FISL-NEXT:    vmrghw v2, v3, v2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test44:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxswapd vs0, v2
+; CHECK-LE-NEXT:    xxlor vs1, v2, v2
+; CHECK-LE-NEXT:    xscvuxdsp f1, f1
+; CHECK-LE-NEXT:    xscvuxdsp f0, f0
+; CHECK-LE-NEXT:    xscvdpspn vs1, f1
+; CHECK-LE-NEXT:    xscvdpspn vs0, f0
+; CHECK-LE-NEXT:    xxsldwi v3, vs1, vs1, 1
+; CHECK-LE-NEXT:    xxsldwi v2, vs0, vs0, 1
+; CHECK-LE-NEXT:    vmrglw v2, v3, v2
+; CHECK-LE-NEXT:    blr
   %v = uitofp <2 x i64> %a to <2 x float>
   ret <2 x float> %v
 
-; CHECK-LABEL: @test44
 ; FIXME: The code quality here looks pretty bad.
-; CHECK: blr
 }
 
 define <2 x float> @test45(<2 x i64> %a) {
+; CHECK-LABEL: test45:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    addi r4, r1, -64
+; CHECK-NEXT:    stxvd2x v2, 0, r3
+; CHECK-NEXT:    ld r3, -8(r1)
+; CHECK-NEXT:    std r3, -24(r1)
+; CHECK-NEXT:    ld r3, -16(r1)
+; CHECK-NEXT:    std r3, -32(r1)
+; CHECK-NEXT:    lfd f0, -24(r1)
+; CHECK-NEXT:    fcfids f0, f0
+; CHECK-NEXT:    stfs f0, -48(r1)
+; CHECK-NEXT:    lfd f0, -32(r1)
+; CHECK-NEXT:    addi r3, r1, -48
+; CHECK-NEXT:    fcfids f0, f0
+; CHECK-NEXT:    stfs f0, -64(r1)
+; CHECK-NEXT:    lxvw4x v2, 0, r3
+; CHECK-NEXT:    lxvw4x v3, 0, r4
+; CHECK-NEXT:    vmrghw v2, v3, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test45:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    addi r4, r1, -64
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    ld r3, -8(r1)
+; CHECK-REG-NEXT:    std r3, -24(r1)
+; CHECK-REG-NEXT:    ld r3, -16(r1)
+; CHECK-REG-NEXT:    std r3, -32(r1)
+; CHECK-REG-NEXT:    lfd f0, -24(r1)
+; CHECK-REG-NEXT:    fcfids f0, f0
+; CHECK-REG-NEXT:    stfs f0, -48(r1)
+; CHECK-REG-NEXT:    lfd f0, -32(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -48
+; CHECK-REG-NEXT:    fcfids f0, f0
+; CHECK-REG-NEXT:    stfs f0, -64(r1)
+; CHECK-REG-NEXT:    lxvw4x v2, 0, r3
+; CHECK-REG-NEXT:    lxvw4x v3, 0, r4
+; CHECK-REG-NEXT:    vmrghw v2, v3, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test45:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    ld r3, -8(r1)
+; CHECK-FISL-NEXT:    std r3, -24(r1)
+; CHECK-FISL-NEXT:    ld r3, -16(r1)
+; CHECK-FISL-NEXT:    std r3, -32(r1)
+; CHECK-FISL-NEXT:    lfd f0, -24(r1)
+; CHECK-FISL-NEXT:    fcfids f0, f0
+; CHECK-FISL-NEXT:    stfs f0, -48(r1)
+; CHECK-FISL-NEXT:    lfd f0, -32(r1)
+; CHECK-FISL-NEXT:    fcfids f0, f0
+; CHECK-FISL-NEXT:    stfs f0, -64(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -48
+; CHECK-FISL-NEXT:    lxvw4x v2, 0, r3
+; CHECK-FISL-NEXT:    addi r3, r1, -64
+; CHECK-FISL-NEXT:    lxvw4x v3, 0, r3
+; CHECK-FISL-NEXT:    vmrghw v2, v3, v2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test45:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxswapd vs0, v2
+; CHECK-LE-NEXT:    xxlor vs1, v2, v2
+; CHECK-LE-NEXT:    xscvsxdsp f1, f1
+; CHECK-LE-NEXT:    xscvsxdsp f0, f0
+; CHECK-LE-NEXT:    xscvdpspn vs1, f1
+; CHECK-LE-NEXT:    xscvdpspn vs0, f0
+; CHECK-LE-NEXT:    xxsldwi v3, vs1, vs1, 1
+; CHECK-LE-NEXT:    xxsldwi v2, vs0, vs0, 1
+; CHECK-LE-NEXT:    vmrglw v2, v3, v2
+; CHECK-LE-NEXT:    blr
   %v = sitofp <2 x i64> %a to <2 x float>
   ret <2 x float> %v
 
-; CHECK-LABEL: @test45
 ; FIXME: The code quality here looks pretty bad.
-; CHECK: blr
 }
 
 define <2 x i64> @test46(<2 x float> %a) {
+; CHECK-LABEL: test46:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi r3, r1, -48
+; CHECK-NEXT:    stxvw4x v2, 0, r3
+; CHECK-NEXT:    lfs f0, -44(r1)
+; CHECK-NEXT:    xscvdpuxds f0, f0
+; CHECK-NEXT:    stfd f0, -32(r1)
+; CHECK-NEXT:    lfs f0, -48(r1)
+; CHECK-NEXT:    xscvdpuxds f0, f0
+; CHECK-NEXT:    stfd f0, -24(r1)
+; CHECK-NEXT:    ld r3, -32(r1)
+; CHECK-NEXT:    std r3, -8(r1)
+; CHECK-NEXT:    ld r3, -24(r1)
+; CHECK-NEXT:    std r3, -16(r1)
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test46:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addi r3, r1, -48
+; CHECK-REG-NEXT:    stxvw4x v2, 0, r3
+; CHECK-REG-NEXT:    lfs f0, -44(r1)
+; CHECK-REG-NEXT:    xscvdpuxds f0, f0
+; CHECK-REG-NEXT:    stfd f0, -32(r1)
+; CHECK-REG-NEXT:    lfs f0, -48(r1)
+; CHECK-REG-NEXT:    xscvdpuxds f0, f0
+; CHECK-REG-NEXT:    stfd f0, -24(r1)
+; CHECK-REG-NEXT:    ld r3, -32(r1)
+; CHECK-REG-NEXT:    std r3, -8(r1)
+; CHECK-REG-NEXT:    ld r3, -24(r1)
+; CHECK-REG-NEXT:    std r3, -16(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test46:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addi r3, r1, -48
+; CHECK-FISL-NEXT:    stxvw4x v2, 0, r3
+; CHECK-FISL-NEXT:    lfs f0, -44(r1)
+; CHECK-FISL-NEXT:    xscvdpuxds f0, f0
+; CHECK-FISL-NEXT:    stfd f0, -32(r1)
+; CHECK-FISL-NEXT:    lfs f0, -48(r1)
+; CHECK-FISL-NEXT:    xscvdpuxds f0, f0
+; CHECK-FISL-NEXT:    stfd f0, -24(r1)
+; CHECK-FISL-NEXT:    ld r3, -32(r1)
+; CHECK-FISL-NEXT:    std r3, -8(r1)
+; CHECK-FISL-NEXT:    ld r3, -24(r1)
+; CHECK-FISL-NEXT:    std r3, -16(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    lxvd2x vs1, 0, r3
+; CHECK-FISL-NEXT:    xxlor v2, vs1, vs1
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test46:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxsldwi vs0, v2, v2, 3
+; CHECK-LE-NEXT:    xxswapd vs1, v2
+; CHECK-LE-NEXT:    xscvspdpn f0, vs0
+; CHECK-LE-NEXT:    xscvspdpn f1, vs1
+; CHECK-LE-NEXT:    xxmrghd vs0, vs1, vs0
+; CHECK-LE-NEXT:    xvcvdpuxds v2, vs0
+; CHECK-LE-NEXT:    blr
   %v = fptoui <2 x float> %a to <2 x i64>
   ret <2 x i64> %v
 
-; CHECK-LABEL: @test46
 ; FIXME: The code quality here looks pretty bad.
-; CHECK: blr
 }
 
 define <2 x i64> @test47(<2 x float> %a) {
+; CHECK-LABEL: test47:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi r3, r1, -48
+; CHECK-NEXT:    stxvw4x v2, 0, r3
+; CHECK-NEXT:    lfs f0, -44(r1)
+; CHECK-NEXT:    xscvdpsxds f0, f0
+; CHECK-NEXT:    stfd f0, -32(r1)
+; CHECK-NEXT:    lfs f0, -48(r1)
+; CHECK-NEXT:    xscvdpsxds f0, f0
+; CHECK-NEXT:    stfd f0, -24(r1)
+; CHECK-NEXT:    ld r3, -32(r1)
+; CHECK-NEXT:    std r3, -8(r1)
+; CHECK-NEXT:    ld r3, -24(r1)
+; CHECK-NEXT:    std r3, -16(r1)
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test47:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addi r3, r1, -48
+; CHECK-REG-NEXT:    stxvw4x v2, 0, r3
+; CHECK-REG-NEXT:    lfs f0, -44(r1)
+; CHECK-REG-NEXT:    xscvdpsxds f0, f0
+; CHECK-REG-NEXT:    stfd f0, -32(r1)
+; CHECK-REG-NEXT:    lfs f0, -48(r1)
+; CHECK-REG-NEXT:    xscvdpsxds f0, f0
+; CHECK-REG-NEXT:    stfd f0, -24(r1)
+; CHECK-REG-NEXT:    ld r3, -32(r1)
+; CHECK-REG-NEXT:    std r3, -8(r1)
+; CHECK-REG-NEXT:    ld r3, -24(r1)
+; CHECK-REG-NEXT:    std r3, -16(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test47:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addi r3, r1, -48
+; CHECK-FISL-NEXT:    stxvw4x v2, 0, r3
+; CHECK-FISL-NEXT:    lfs f0, -44(r1)
+; CHECK-FISL-NEXT:    xscvdpsxds f0, f0
+; CHECK-FISL-NEXT:    stfd f0, -32(r1)
+; CHECK-FISL-NEXT:    lfs f0, -48(r1)
+; CHECK-FISL-NEXT:    xscvdpsxds f0, f0
+; CHECK-FISL-NEXT:    stfd f0, -24(r1)
+; CHECK-FISL-NEXT:    ld r3, -32(r1)
+; CHECK-FISL-NEXT:    std r3, -8(r1)
+; CHECK-FISL-NEXT:    ld r3, -24(r1)
+; CHECK-FISL-NEXT:    std r3, -16(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    lxvd2x vs1, 0, r3
+; CHECK-FISL-NEXT:    xxlor v2, vs1, vs1
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test47:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxsldwi vs0, v2, v2, 3
+; CHECK-LE-NEXT:    xxswapd vs1, v2
+; CHECK-LE-NEXT:    xscvspdpn f0, vs0
+; CHECK-LE-NEXT:    xscvspdpn f1, vs1
+; CHECK-LE-NEXT:    xxmrghd vs0, vs1, vs0
+; CHECK-LE-NEXT:    xvcvdpsxds v2, vs0
+; CHECK-LE-NEXT:    blr
   %v = fptosi <2 x float> %a to <2 x i64>
   ret <2 x i64> %v
 
-; CHECK-LABEL: @test47
 ; FIXME: The code quality here looks pretty bad.
-; CHECK: blr
 }
 
 define <2 x double> @test50(double* %a) {
+; CHECK-LABEL: test50:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    lxvdsx v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test50:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    lxvdsx v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test50:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    lxvdsx v2, 0, r3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test50:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    lxvdsx v2, 0, r3
+; CHECK-LE-NEXT:    blr
   %v = load double, double* %a, align 8
   %w = insertelement <2 x double> undef, double %v, i32 0
   %x = insertelement <2 x double> %w, double %v, i32 1
   ret <2 x double> %x
 
-; CHECK-LABEL: @test50
-; CHECK: lxvdsx v2, 0, r3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test50
-; CHECK-LE: lxvdsx v2, 0, r3
-; CHECK-LE: blr
 }
 
 define <2 x double> @test51(<2 x double> %a, <2 x double> %b) {
+; CHECK-LABEL: test51:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xxspltd v2, v2, 0
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test51:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xxspltd v2, v2, 0
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test51:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xxspltd v2, v2, 0
+; CHECK-FISL-NEXT:    li r3, -16
+; CHECK-FISL-NEXT:    stxvd2x v3, r1, r3 # 16-byte Folded Spill
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test51:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxspltd v2, v2, 1
+; CHECK-LE-NEXT:    blr
   %v = shufflevector <2 x double> %a, <2 x double> %b, <2 x i32> <i32 0, i32 0>
   ret <2 x double> %v
 
-; CHECK-LABEL: @test51
-; CHECK: xxspltd v2, v2, 0
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test51
-; CHECK-LE: xxspltd v2, v2, 1
-; CHECK-LE: blr
 }
 
 define <2 x double> @test52(<2 x double> %a, <2 x double> %b) {
+; CHECK-LABEL: test52:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xxmrghd v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test52:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xxmrghd v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test52:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xxmrghd v2, v2, v3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test52:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxmrgld v2, v3, v2
+; CHECK-LE-NEXT:    blr
   %v = shufflevector <2 x double> %a, <2 x double> %b, <2 x i32> <i32 0, i32 2>
   ret <2 x double> %v
 
-; CHECK-LABEL: @test52
-; CHECK: xxmrghd v2, v2, v3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test52
-; CHECK-LE: xxmrgld v2, v3, v2
-; CHECK-LE: blr
 }
 
 define <2 x double> @test53(<2 x double> %a, <2 x double> %b) {
+; CHECK-LABEL: test53:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xxmrghd v2, v3, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test53:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xxmrghd v2, v3, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test53:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xxmrghd v2, v3, v2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test53:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxmrgld v2, v2, v3
+; CHECK-LE-NEXT:    blr
   %v = shufflevector <2 x double> %a, <2 x double> %b, <2 x i32> <i32 2, i32 0>
   ret <2 x double> %v
 
-; CHECK-LABEL: @test53
-; CHECK: xxmrghd v2, v3, v2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test53
-; CHECK-LE: xxmrgld v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <2 x double> @test54(<2 x double> %a, <2 x double> %b) {
+; CHECK-LABEL: test54:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xxpermdi v2, v2, v3, 2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test54:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xxpermdi v2, v2, v3, 2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test54:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xxpermdi v2, v2, v3, 2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test54:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxpermdi v2, v3, v2, 2
+; CHECK-LE-NEXT:    blr
   %v = shufflevector <2 x double> %a, <2 x double> %b, <2 x i32> <i32 1, i32 2>
   ret <2 x double> %v
 
-; CHECK-LABEL: @test54
-; CHECK: xxpermdi v2, v2, v3, 2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test54
-; CHECK-LE: xxpermdi v2, v3, v2, 2
-; CHECK-LE: blr
 }
 
 define <2 x double> @test55(<2 x double> %a, <2 x double> %b) {
+; CHECK-LABEL: test55:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xxmrgld v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test55:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xxmrgld v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test55:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xxmrgld v2, v2, v3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test55:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxmrghd v2, v3, v2
+; CHECK-LE-NEXT:    blr
   %v = shufflevector <2 x double> %a, <2 x double> %b, <2 x i32> <i32 1, i32 3>
   ret <2 x double> %v
 
-; CHECK-LABEL: @test55
-; CHECK: xxmrgld v2, v2, v3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test55
-; CHECK-LE: xxmrghd v2, v3, v2
-; CHECK-LE: blr
 }
 
 define <2 x i64> @test56(<2 x i64> %a, <2 x i64> %b) {
+; CHECK-LABEL: test56:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xxmrgld v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test56:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xxmrgld v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test56:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xxmrgld v2, v2, v3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test56:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxmrghd v2, v3, v2
+; CHECK-LE-NEXT:    blr
   %v = shufflevector <2 x i64> %a, <2 x i64> %b, <2 x i32> <i32 1, i32 3>
   ret <2 x i64> %v
 
-; CHECK-LABEL: @test56
-; CHECK: xxmrgld v2, v2, v3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test56
-; CHECK-LE: xxmrghd v2, v3, v2
-; CHECK-LE: blr
 }
 
 define <2 x i64> @test60(<2 x i64> %a, <2 x i64> %b) {
+; CHECK-LABEL: test60:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi r3, r1, -32
+; CHECK-NEXT:    addi r4, r1, -48
+; CHECK-NEXT:    stxvd2x v3, 0, r3
+; CHECK-NEXT:    stxvd2x v2, 0, r4
+; CHECK-NEXT:    lwz r3, -20(r1)
+; CHECK-NEXT:    ld r4, -40(r1)
+; CHECK-NEXT:    sld r3, r4, r3
+; CHECK-NEXT:    std r3, -8(r1)
+; CHECK-NEXT:    lwz r3, -28(r1)
+; CHECK-NEXT:    ld r4, -48(r1)
+; CHECK-NEXT:    sld r3, r4, r3
+; CHECK-NEXT:    std r3, -16(r1)
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test60:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addi r3, r1, -32
+; CHECK-REG-NEXT:    addi r4, r1, -48
+; CHECK-REG-NEXT:    stxvd2x v3, 0, r3
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r4
+; CHECK-REG-NEXT:    lwz r3, -20(r1)
+; CHECK-REG-NEXT:    ld r4, -40(r1)
+; CHECK-REG-NEXT:    sld r3, r4, r3
+; CHECK-REG-NEXT:    std r3, -8(r1)
+; CHECK-REG-NEXT:    lwz r3, -28(r1)
+; CHECK-REG-NEXT:    ld r4, -48(r1)
+; CHECK-REG-NEXT:    sld r3, r4, r3
+; CHECK-REG-NEXT:    std r3, -16(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test60:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addi r3, r1, -32
+; CHECK-FISL-NEXT:    stxvd2x v3, 0, r3
+; CHECK-FISL-NEXT:    addi r3, r1, -48
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    lwz r4, -20(r1)
+; CHECK-FISL-NEXT:    ld r3, -40(r1)
+; CHECK-FISL-NEXT:    sld r3, r3, r4
+; CHECK-FISL-NEXT:    std r3, -8(r1)
+; CHECK-FISL-NEXT:    lwz r4, -28(r1)
+; CHECK-FISL-NEXT:    ld r3, -48(r1)
+; CHECK-FISL-NEXT:    sld r3, r3, r4
+; CHECK-FISL-NEXT:    std r3, -16(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test60:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    vsld v2, v2, v3
+; CHECK-LE-NEXT:    blr
   %v = shl <2 x i64> %a, %b
   ret <2 x i64> %v
 
-; CHECK-LABEL: @test60
 ; This should scalarize, and the current code quality is not good.
-; CHECK: stxvd2x v3, 0, r3
-; CHECK: stxvd2x v2, 0, r4
-; CHECK: sld r3, r4, r3
-; CHECK: sld r3, r4, r3
-; CHECK: lxvd2x v2, 0, r3
-; CHECK: blr
 }
 
 define <2 x i64> @test61(<2 x i64> %a, <2 x i64> %b) {
+; CHECK-LABEL: test61:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi r3, r1, -32
+; CHECK-NEXT:    addi r4, r1, -48
+; CHECK-NEXT:    stxvd2x v3, 0, r3
+; CHECK-NEXT:    stxvd2x v2, 0, r4
+; CHECK-NEXT:    lwz r3, -20(r1)
+; CHECK-NEXT:    ld r4, -40(r1)
+; CHECK-NEXT:    srd r3, r4, r3
+; CHECK-NEXT:    std r3, -8(r1)
+; CHECK-NEXT:    lwz r3, -28(r1)
+; CHECK-NEXT:    ld r4, -48(r1)
+; CHECK-NEXT:    srd r3, r4, r3
+; CHECK-NEXT:    std r3, -16(r1)
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test61:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addi r3, r1, -32
+; CHECK-REG-NEXT:    addi r4, r1, -48
+; CHECK-REG-NEXT:    stxvd2x v3, 0, r3
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r4
+; CHECK-REG-NEXT:    lwz r3, -20(r1)
+; CHECK-REG-NEXT:    ld r4, -40(r1)
+; CHECK-REG-NEXT:    srd r3, r4, r3
+; CHECK-REG-NEXT:    std r3, -8(r1)
+; CHECK-REG-NEXT:    lwz r3, -28(r1)
+; CHECK-REG-NEXT:    ld r4, -48(r1)
+; CHECK-REG-NEXT:    srd r3, r4, r3
+; CHECK-REG-NEXT:    std r3, -16(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test61:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addi r3, r1, -32
+; CHECK-FISL-NEXT:    stxvd2x v3, 0, r3
+; CHECK-FISL-NEXT:    addi r3, r1, -48
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    lwz r4, -20(r1)
+; CHECK-FISL-NEXT:    ld r3, -40(r1)
+; CHECK-FISL-NEXT:    srd r3, r3, r4
+; CHECK-FISL-NEXT:    std r3, -8(r1)
+; CHECK-FISL-NEXT:    lwz r4, -28(r1)
+; CHECK-FISL-NEXT:    ld r3, -48(r1)
+; CHECK-FISL-NEXT:    srd r3, r3, r4
+; CHECK-FISL-NEXT:    std r3, -16(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test61:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    vsrd v2, v2, v3
+; CHECK-LE-NEXT:    blr
   %v = lshr <2 x i64> %a, %b
   ret <2 x i64> %v
 
-; CHECK-LABEL: @test61
 ; This should scalarize, and the current code quality is not good.
-; CHECK: stxvd2x v3, 0, r3
-; CHECK: stxvd2x v2, 0, r4
-; CHECK: srd r3, r4, r3
-; CHECK: srd r3, r4, r3
-; CHECK: lxvd2x v2, 0, r3
-; CHECK: blr
 }
 
 define <2 x i64> @test62(<2 x i64> %a, <2 x i64> %b) {
+; CHECK-LABEL: test62:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi r3, r1, -32
+; CHECK-NEXT:    addi r4, r1, -48
+; CHECK-NEXT:    stxvd2x v3, 0, r3
+; CHECK-NEXT:    stxvd2x v2, 0, r4
+; CHECK-NEXT:    lwz r3, -20(r1)
+; CHECK-NEXT:    ld r4, -40(r1)
+; CHECK-NEXT:    srad r3, r4, r3
+; CHECK-NEXT:    std r3, -8(r1)
+; CHECK-NEXT:    lwz r3, -28(r1)
+; CHECK-NEXT:    ld r4, -48(r1)
+; CHECK-NEXT:    srad r3, r4, r3
+; CHECK-NEXT:    std r3, -16(r1)
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test62:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addi r3, r1, -32
+; CHECK-REG-NEXT:    addi r4, r1, -48
+; CHECK-REG-NEXT:    stxvd2x v3, 0, r3
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r4
+; CHECK-REG-NEXT:    lwz r3, -20(r1)
+; CHECK-REG-NEXT:    ld r4, -40(r1)
+; CHECK-REG-NEXT:    srad r3, r4, r3
+; CHECK-REG-NEXT:    std r3, -8(r1)
+; CHECK-REG-NEXT:    lwz r3, -28(r1)
+; CHECK-REG-NEXT:    ld r4, -48(r1)
+; CHECK-REG-NEXT:    srad r3, r4, r3
+; CHECK-REG-NEXT:    std r3, -16(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test62:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addi r3, r1, -32
+; CHECK-FISL-NEXT:    stxvd2x v3, 0, r3
+; CHECK-FISL-NEXT:    addi r3, r1, -48
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    lwz r4, -20(r1)
+; CHECK-FISL-NEXT:    ld r3, -40(r1)
+; CHECK-FISL-NEXT:    srad r3, r3, r4
+; CHECK-FISL-NEXT:    std r3, -8(r1)
+; CHECK-FISL-NEXT:    lwz r4, -28(r1)
+; CHECK-FISL-NEXT:    ld r3, -48(r1)
+; CHECK-FISL-NEXT:    srad r3, r3, r4
+; CHECK-FISL-NEXT:    std r3, -16(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test62:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    vsrad v2, v2, v3
+; CHECK-LE-NEXT:    blr
   %v = ashr <2 x i64> %a, %b
   ret <2 x i64> %v
 
-; CHECK-LABEL: @test62
 ; This should scalarize, and the current code quality is not good.
-; CHECK: stxvd2x v3, 0, r3
-; CHECK: stxvd2x v2, 0, r4
-; CHECK: srad r3, r4, r3
-; CHECK: srad r3, r4, r3
-; CHECK: lxvd2x v2, 0, r3
-; CHECK: blr
 }
 
 define double @test63(<2 x double> %a) {
+; CHECK-LABEL: test63:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xxlor f1, v2, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test63:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xxlor f1, v2, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test63:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xxlor f0, v2, v2
+; CHECK-FISL-NEXT:    fmr f1, f0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test63:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxswapd vs1, v2
+; CHECK-LE-NEXT:    # kill: def $f1 killed $f1 killed $vsl1
+; CHECK-LE-NEXT:    blr
   %v = extractelement <2 x double> %a, i32 0
   ret double %v
 
-; CHECK-REG-LABEL: @test63
-; CHECK-REG: xxlor f1, v2, v2
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test63
-; CHECK-FISL: xxlor f0, v2, v2
-; CHECK-FISL: fmr f1, f0
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test63
-; CHECK-LE: xxswapd vs1, v2
-; CHECK-LE: blr
 }
 
 define double @test64(<2 x double> %a) {
+; CHECK-LABEL: test64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xxswapd vs1, v2
+; CHECK-NEXT:    # kill: def $f1 killed $f1 killed $vsl1
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test64:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xxswapd vs1, v2
+; CHECK-REG-NEXT:    # kill: def $f1 killed $f1 killed $vsl1
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test64:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xxswapd vs0, v2
+; CHECK-FISL-NEXT:    fmr f1, f0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test64:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxlor f1, v2, v2
+; CHECK-LE-NEXT:    blr
   %v = extractelement <2 x double> %a, i32 1
   ret double %v
 
-; CHECK-REG-LABEL: @test64
-; CHECK-REG: xxswapd vs1, v2
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test64
-; CHECK-FISL: xxswapd  v2, v2
-; CHECK-FISL: xxlor f0, v2, v2
-; CHECK-FISL: fmr f1, f0
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test64
-; CHECK-LE: xxlor f1, v2, v2
 }
 
 define <2 x i1> @test65(<2 x i64> %a, <2 x i64> %b) {
+; CHECK-LABEL: test65:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vcmpequw v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test65:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    vcmpequw v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test65:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    vcmpequw v2, v2, v3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test65:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    vcmpequd v2, v2, v3
+; CHECK-LE-NEXT:    blr
   %w = icmp eq <2 x i64> %a, %b
   ret <2 x i1> %w
 
-; CHECK-REG-LABEL: @test65
-; CHECK-REG: vcmpequw v2, v2, v3
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test65
-; CHECK-FISL: vcmpequw v2, v2, v3
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test65
-; CHECK-LE: vcmpequd v2, v2, v3
-; CHECK-LE: blr
 }
 
 define <2 x i1> @test66(<2 x i64> %a, <2 x i64> %b) {
+; CHECK-LABEL: test66:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vcmpequw v2, v2, v3
+; CHECK-NEXT:    xxlnor v2, v2, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test66:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    vcmpequw v2, v2, v3
+; CHECK-REG-NEXT:    xxlnor v2, v2, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test66:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    vcmpequw v2, v2, v3
+; CHECK-FISL-NEXT:    xxlnor vs0, v2, v2
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test66:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    vcmpequd v2, v2, v3
+; CHECK-LE-NEXT:    xxlnor v2, v2, v2
+; CHECK-LE-NEXT:    blr
   %w = icmp ne <2 x i64> %a, %b
   ret <2 x i1> %w
 
-; CHECK-REG-LABEL: @test66
-; CHECK-REG: vcmpequw v2, v2, v3
-; CHECK-REG: xxlnor v2, v2, v2
-; CHECK-REG: blr
 
-; CHECK-FISL-LABEL: @test66
-; CHECK-FISL: vcmpequw v2, v2, v3
-; CHECK-FISL: xxlnor v2, v2, v2
-; CHECK-FISL: blr
 
-; CHECK-LE-LABEL: @test66
-; CHECK-LE: vcmpequd v2, v2, v3
-; CHECK-LE: xxlnor v2, v2, v2
-; CHECK-LE: blr
 }
 
 define <2 x i1> @test67(<2 x i64> %a, <2 x i64> %b) {
+; CHECK-LABEL: test67:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi r3, r1, -32
+; CHECK-NEXT:    addi r4, r1, -48
+; CHECK-NEXT:    stxvd2x v3, 0, r3
+; CHECK-NEXT:    stxvd2x v2, 0, r4
+; CHECK-NEXT:    ld r3, -24(r1)
+; CHECK-NEXT:    ld r4, -40(r1)
+; CHECK-NEXT:    cmpld r4, r3
+; CHECK-NEXT:    li r3, 0
+; CHECK-NEXT:    li r4, -1
+; CHECK-NEXT:    isel r5, r4, r3, lt
+; CHECK-NEXT:    std r5, -8(r1)
+; CHECK-NEXT:    ld r5, -32(r1)
+; CHECK-NEXT:    ld r6, -48(r1)
+; CHECK-NEXT:    cmpld r6, r5
+; CHECK-NEXT:    isel r3, r4, r3, lt
+; CHECK-NEXT:    std r3, -16(r1)
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test67:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addi r3, r1, -32
+; CHECK-REG-NEXT:    addi r4, r1, -48
+; CHECK-REG-NEXT:    stxvd2x v3, 0, r3
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r4
+; CHECK-REG-NEXT:    ld r3, -24(r1)
+; CHECK-REG-NEXT:    ld r4, -40(r1)
+; CHECK-REG-NEXT:    cmpld r4, r3
+; CHECK-REG-NEXT:    li r3, 0
+; CHECK-REG-NEXT:    li r4, -1
+; CHECK-REG-NEXT:    isel r5, r4, r3, lt
+; CHECK-REG-NEXT:    std r5, -8(r1)
+; CHECK-REG-NEXT:    ld r5, -32(r1)
+; CHECK-REG-NEXT:    ld r6, -48(r1)
+; CHECK-REG-NEXT:    cmpld r6, r5
+; CHECK-REG-NEXT:    isel r3, r4, r3, lt
+; CHECK-REG-NEXT:    std r3, -16(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test67:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addi r3, r1, -32
+; CHECK-FISL-NEXT:    stxvd2x v3, 0, r3
+; CHECK-FISL-NEXT:    addi r3, r1, -48
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    ld r3, -24(r1)
+; CHECK-FISL-NEXT:    ld r4, -40(r1)
+; CHECK-FISL-NEXT:    cmpld r4, r3
+; CHECK-FISL-NEXT:    li r3, 0
+; CHECK-FISL-NEXT:    li r4, -1
+; CHECK-FISL-NEXT:    isel r5, r4, r3, lt
+; CHECK-FISL-NEXT:    std r5, -8(r1)
+; CHECK-FISL-NEXT:    ld r5, -32(r1)
+; CHECK-FISL-NEXT:    ld r6, -48(r1)
+; CHECK-FISL-NEXT:    cmpld r6, r5
+; CHECK-FISL-NEXT:    isel r3, r4, r3, lt
+; CHECK-FISL-NEXT:    std r3, -16(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test67:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    vcmpgtud v2, v3, v2
+; CHECK-LE-NEXT:    blr
   %w = icmp ult <2 x i64> %a, %b
   ret <2 x i1> %w
 
-; CHECK-LABEL: @test67
 ; This should scalarize, and the current code quality is not good.
-; CHECK: stxvd2x v3, 0, r3
-; CHECK: stxvd2x v2, 0, r4
-; CHECK: cmpld r4, r3
-; CHECK: cmpld r6, r5
-; CHECK: lxvd2x v2, 0, r3
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test67
-; CHECK-LE: vcmpgtud v2, v3, v2
-; CHECK-LE: blr
 }
 
 define <2 x double> @test68(<2 x i32> %a) {
+; CHECK-LABEL: test68:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    xxmrghw vs0, v2, v2
+; CHECK-NEXT:    xvcvsxwdp v2, vs0
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test68:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    xxmrghw vs0, v2, v2
+; CHECK-REG-NEXT:    xvcvsxwdp v2, vs0
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test68:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    xxmrghw vs0, v2, v2
+; CHECK-FISL-NEXT:    xvcvsxwdp v2, vs0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test68:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    xxmrglw v2, v2, v2
+; CHECK-LE-NEXT:    xvcvsxwdp v2, v2
+; CHECK-LE-NEXT:    blr
   %w = sitofp <2 x i32> %a to <2 x double>
   ret <2 x double> %w
 
-; CHECK-LABEL: @test68
-; CHECK: xxmrghw vs0, v2, v2
-; CHECK: xvcvsxwdp v2, vs0
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test68
-; CHECK-LE: xxmrglw v2, v2, v2
-; CHECK-LE: xvcvsxwdp v2, v2
-; CHECK-LE: blr
 }
 
 ; This gets scalarized so the code isn't great
 define <2 x double> @test69(<2 x i16> %a) {
+; CHECK-LABEL: test69:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addis r3, r2, .LCPI63_0@toc@ha
+; CHECK-NEXT:    addi r3, r3, .LCPI63_0@toc@l
+; CHECK-NEXT:    lxvw4x v3, 0, r3
+; CHECK-NEXT:    addi r3, r1, -32
+; CHECK-NEXT:    vperm v2, v2, v2, v3
+; CHECK-NEXT:    stxvd2x v2, 0, r3
+; CHECK-NEXT:    lha r3, -18(r1)
+; CHECK-NEXT:    std r3, -8(r1)
+; CHECK-NEXT:    lha r3, -26(r1)
+; CHECK-NEXT:    std r3, -16(r1)
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    xvcvsxddp v2, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test69:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addis r3, r2, .LCPI63_0@toc@ha
+; CHECK-REG-NEXT:    addi r3, r3, .LCPI63_0@toc@l
+; CHECK-REG-NEXT:    lxvw4x v3, 0, r3
+; CHECK-REG-NEXT:    addi r3, r1, -32
+; CHECK-REG-NEXT:    vperm v2, v2, v2, v3
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    lha r3, -18(r1)
+; CHECK-REG-NEXT:    std r3, -8(r1)
+; CHECK-REG-NEXT:    lha r3, -26(r1)
+; CHECK-REG-NEXT:    std r3, -16(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    xvcvsxddp v2, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test69:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addis r3, r2, .LCPI63_0@toc@ha
+; CHECK-FISL-NEXT:    addi r3, r3, .LCPI63_0@toc@l
+; CHECK-FISL-NEXT:    lxvw4x v3, 0, r3
+; CHECK-FISL-NEXT:    vperm v2, v2, v2, v3
+; CHECK-FISL-NEXT:    addi r3, r1, -32
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    lha r3, -18(r1)
+; CHECK-FISL-NEXT:    std r3, -8(r1)
+; CHECK-FISL-NEXT:    lha r3, -26(r1)
+; CHECK-FISL-NEXT:    std r3, -16(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    xvcvsxddp v2, v2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test69:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    addis r3, r2, .LCPI63_0@toc@ha
+; CHECK-LE-NEXT:    addi r3, r3, .LCPI63_0@toc@l
+; CHECK-LE-NEXT:    lvx v3, 0, r3
+; CHECK-LE-NEXT:    addis r3, r2, .LCPI63_1@toc@ha
+; CHECK-LE-NEXT:    addi r3, r3, .LCPI63_1@toc@l
+; CHECK-LE-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-LE-NEXT:    vperm v2, v2, v2, v3
+; CHECK-LE-NEXT:    xxswapd v3, vs0
+; CHECK-LE-NEXT:    vsld v2, v2, v3
+; CHECK-LE-NEXT:    vsrad v2, v2, v3
+; CHECK-LE-NEXT:    xvcvsxddp v2, v2
+; CHECK-LE-NEXT:    blr
   %w = sitofp <2 x i16> %a to <2 x double>
   ret <2 x double> %w
 
-; CHECK-LABEL: @test69
-; CHECK-DAG: lxvd2x v2, 0, r3
-; CHECK-DAG: xvcvsxddp v2, v2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test69
-; CHECK-LE: vperm
-; CHECK-LE: vsld
-; CHECK-LE: vsrad
-; CHECK-LE: xvcvsxddp v2, v2
-; CHECK-LE: blr
 }
 
 ; This gets scalarized so the code isn't great
 define <2 x double> @test70(<2 x i8> %a) {
+; CHECK-LABEL: test70:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addis r3, r2, .LCPI64_0@toc@ha
+; CHECK-NEXT:    addi r3, r3, .LCPI64_0@toc@l
+; CHECK-NEXT:    lxvw4x v3, 0, r3
+; CHECK-NEXT:    addi r3, r1, -32
+; CHECK-NEXT:    vperm v2, v2, v2, v3
+; CHECK-NEXT:    stxvd2x v2, 0, r3
+; CHECK-NEXT:    ld r3, -24(r1)
+; CHECK-NEXT:    extsb r3, r3
+; CHECK-NEXT:    std r3, -8(r1)
+; CHECK-NEXT:    ld r3, -32(r1)
+; CHECK-NEXT:    extsb r3, r3
+; CHECK-NEXT:    std r3, -16(r1)
+; CHECK-NEXT:    addi r3, r1, -16
+; CHECK-NEXT:    lxvd2x v2, 0, r3
+; CHECK-NEXT:    xvcvsxddp v2, v2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test70:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addis r3, r2, .LCPI64_0@toc@ha
+; CHECK-REG-NEXT:    addi r3, r3, .LCPI64_0@toc@l
+; CHECK-REG-NEXT:    lxvw4x v3, 0, r3
+; CHECK-REG-NEXT:    addi r3, r1, -32
+; CHECK-REG-NEXT:    vperm v2, v2, v2, v3
+; CHECK-REG-NEXT:    stxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    ld r3, -24(r1)
+; CHECK-REG-NEXT:    extsb r3, r3
+; CHECK-REG-NEXT:    std r3, -8(r1)
+; CHECK-REG-NEXT:    ld r3, -32(r1)
+; CHECK-REG-NEXT:    extsb r3, r3
+; CHECK-REG-NEXT:    std r3, -16(r1)
+; CHECK-REG-NEXT:    addi r3, r1, -16
+; CHECK-REG-NEXT:    lxvd2x v2, 0, r3
+; CHECK-REG-NEXT:    xvcvsxddp v2, v2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test70:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    addis r3, r2, .LCPI64_0@toc@ha
+; CHECK-FISL-NEXT:    addi r3, r3, .LCPI64_0@toc@l
+; CHECK-FISL-NEXT:    lxvw4x v3, 0, r3
+; CHECK-FISL-NEXT:    vperm v2, v2, v2, v3
+; CHECK-FISL-NEXT:    addi r3, r1, -32
+; CHECK-FISL-NEXT:    stxvd2x v2, 0, r3
+; CHECK-FISL-NEXT:    ld r3, -24(r1)
+; CHECK-FISL-NEXT:    extsb r3, r3
+; CHECK-FISL-NEXT:    std r3, -8(r1)
+; CHECK-FISL-NEXT:    ld r3, -32(r1)
+; CHECK-FISL-NEXT:    extsb r3, r3
+; CHECK-FISL-NEXT:    std r3, -16(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-FISL-NEXT:    xxlor v2, vs0, vs0
+; CHECK-FISL-NEXT:    xvcvsxddp v2, v2
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test70:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    addis r3, r2, .LCPI64_0@toc@ha
+; CHECK-LE-NEXT:    addi r3, r3, .LCPI64_0@toc@l
+; CHECK-LE-NEXT:    lvx v3, 0, r3
+; CHECK-LE-NEXT:    addis r3, r2, .LCPI64_1@toc@ha
+; CHECK-LE-NEXT:    addi r3, r3, .LCPI64_1@toc@l
+; CHECK-LE-NEXT:    lxvd2x vs0, 0, r3
+; CHECK-LE-NEXT:    vperm v2, v2, v2, v3
+; CHECK-LE-NEXT:    xxswapd v3, vs0
+; CHECK-LE-NEXT:    vsld v2, v2, v3
+; CHECK-LE-NEXT:    vsrad v2, v2, v3
+; CHECK-LE-NEXT:    xvcvsxddp v2, v2
+; CHECK-LE-NEXT:    blr
   %w = sitofp <2 x i8> %a to <2 x double>
   ret <2 x double> %w
 
-; CHECK-LABEL: @test70
-; CHECK-DAG: lxvd2x v2, 0, r3
-; CHECK-DAG: xvcvsxddp v2, v2
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test70
-; CHECK-LE: vperm
-; CHECK-LE: vsld
-; CHECK-LE: vsrad
-; CHECK-LE: xvcvsxddp v2, v2
-; CHECK-LE: blr
 }
 
 ; This gets scalarized so the code isn't great
 define <2 x i32> @test80(i32 %v) {
+; CHECK-LABEL: test80:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi r4, r1, -16
+; CHECK-NEXT:    stw r3, -16(r1)
+; CHECK-NEXT:    addis r3, r2, .LCPI65_0@toc@ha
+; CHECK-NEXT:    lxvw4x vs0, 0, r4
+; CHECK-NEXT:    addi r3, r3, .LCPI65_0@toc@l
+; CHECK-NEXT:    lxvw4x v3, 0, r3
+; CHECK-NEXT:    xxspltw v2, vs0, 0
+; CHECK-NEXT:    vadduwm v2, v2, v3
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test80:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    addi r4, r1, -16
+; CHECK-REG-NEXT:    stw r3, -16(r1)
+; CHECK-REG-NEXT:    addis r3, r2, .LCPI65_0@toc@ha
+; CHECK-REG-NEXT:    lxvw4x vs0, 0, r4
+; CHECK-REG-NEXT:    addi r3, r3, .LCPI65_0@toc@l
+; CHECK-REG-NEXT:    lxvw4x v3, 0, r3
+; CHECK-REG-NEXT:    xxspltw v2, vs0, 0
+; CHECK-REG-NEXT:    vadduwm v2, v2, v3
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test80:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    mr r4, r3
+; CHECK-FISL-NEXT:    stw r4, -16(r1)
+; CHECK-FISL-NEXT:    addi r3, r1, -16
+; CHECK-FISL-NEXT:    lxvw4x vs0, 0, r3
+; CHECK-FISL-NEXT:    xxspltw v2, vs0, 0
+; CHECK-FISL-NEXT:    addis r3, r2, .LCPI65_0@toc@ha
+; CHECK-FISL-NEXT:    addi r3, r3, .LCPI65_0@toc@l
+; CHECK-FISL-NEXT:    lxvw4x v3, 0, r3
+; CHECK-FISL-NEXT:    vadduwm v2, v2, v3
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test80:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    mtvsrd f0, r3
+; CHECK-LE-NEXT:    addis r4, r2, .LCPI65_0@toc@ha
+; CHECK-LE-NEXT:    addi r3, r4, .LCPI65_0@toc@l
+; CHECK-LE-NEXT:    xxswapd vs0, vs0
+; CHECK-LE-NEXT:    lvx v3, 0, r3
+; CHECK-LE-NEXT:    xxspltw v2, vs0, 3
+; CHECK-LE-NEXT:    vadduwm v2, v2, v3
+; CHECK-LE-NEXT:    blr
   %b1 = insertelement <2 x i32> undef, i32 %v, i32 0
   %b2 = shufflevector <2 x i32> %b1, <2 x i32> undef, <2 x i32> zeroinitializer
   %i = add <2 x i32> %b2, <i32 2, i32 3>
   ret <2 x i32> %i
 
-; CHECK-REG-LABEL: @test80
-; CHECK-REG-DAG: stw r3, -16(r1)
-; CHECK-REG-DAG: addi r4, r1, -16
-; CHECK-REG: addis r3, r2, .LCPI65_0@toc@ha
-; CHECK-REG-DAG: addi r3, r3, .LCPI65_0@toc@l
-; CHECK-REG-DAG: lxvw4x vs0, 0, r4
-; CHECK-REG-DAG: lxvw4x v3, 0, r3
-; CHECK-REG: xxspltw v2, vs0, 0
-; CHECK-REG: vadduwm v2, v2, v3
-; CHECK-REG-NOT: stxvw4x
-; CHECK-REG: blr
-
-; CHECK-FISL-LABEL: @test80
-; CHECK-FISL: mr r4, r3
-; CHECK-FISL: stw r4, -16(r1)
-; CHECK-FISL: addi r3, r1, -16
-; CHECK-FISL-DAG: lxvw4x vs0, 0, r3
-; CHECK-FISL-DAG: xxspltw v2, vs0, 0
-; CHECK-FISL: addis r3, r2, .LCPI65_0@toc@ha
-; CHECK-FISL: addi r3, r3, .LCPI65_0@toc@l
-; CHECK-FISL-DAG: lxvw4x v3, 0, r3
-; CHECK-FISL: vadduwm
-; CHECK-FISL-NOT: stxvw4x
-; CHECK-FISL: blr
-
-; CHECK-LE-LABEL: @test80
-; CHECK-LE-DAG: mtvsrd f0, r3
-; CHECK-LE-DAG: xxswapd  vs0, vs0
-; CHECK-LE-DAG: addi r3, r4, .LCPI65_0@toc@l
-; CHECK-LE-DAG: lvx v3, 0, r3
-; CHECK-LE-DAG: xxspltw v2, vs0, 3
-; CHECK-LE-NOT: xxswapd v3,
-; CHECK-LE: vadduwm v2, v2, v3
-; CHECK-LE: blr
+
+
 }
 
 define <2 x double> @test81(<4 x float> %b) {
+; CHECK-LABEL: test81:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test81:
+; CHECK-REG:       # %bb.0:
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test81:
+; CHECK-FISL:       # %bb.0:
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test81:
+; CHECK-LE:       # %bb.0:
+; CHECK-LE-NEXT:    blr
   %w = bitcast <4 x float> %b to <2 x double>
   ret <2 x double> %w
 
-; CHECK-LABEL: @test81
-; CHECK: blr
 
-; CHECK-LE-LABEL: @test81
-; CHECK-LE: blr
 }
 
 define double @test82(double %a, double %b, double %c, double %d) {
+; CHECK-LABEL: test82:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xscmpudp cr0, f3, f4
+; CHECK-NEXT:    beqlr cr0
+; CHECK-NEXT:  # %bb.1: # %entry
+; CHECK-NEXT:    fmr f1, f2
+; CHECK-NEXT:    blr
+;
+; CHECK-REG-LABEL: test82:
+; CHECK-REG:       # %bb.0: # %entry
+; CHECK-REG-NEXT:    xscmpudp cr0, f3, f4
+; CHECK-REG-NEXT:    beqlr cr0
+; CHECK-REG-NEXT:  # %bb.1: # %entry
+; CHECK-REG-NEXT:    fmr f1, f2
+; CHECK-REG-NEXT:    blr
+;
+; CHECK-FISL-LABEL: test82:
+; CHECK-FISL:       # %bb.0: # %entry
+; CHECK-FISL-NEXT:    xscmpudp cr0, f3, f4
+; CHECK-FISL-NEXT:    stfd f2, -8(r1) # 8-byte Folded Spill
+; CHECK-FISL-NEXT:    stfd f1, -16(r1) # 8-byte Folded Spill
+; CHECK-FISL-NEXT:    beq cr0, .LBB67_2
+; CHECK-FISL-NEXT:  # %bb.1: # %entry
+; CHECK-FISL-NEXT:    lfd f0, -8(r1) # 8-byte Folded Reload
+; CHECK-FISL-NEXT:    stfd f0, -16(r1) # 8-byte Folded Spill
+; CHECK-FISL-NEXT:  .LBB67_2: # %entry
+; CHECK-FISL-NEXT:    lfd f0, -16(r1) # 8-byte Folded Reload
+; CHECK-FISL-NEXT:    fmr f1, f0
+; CHECK-FISL-NEXT:    blr
+;
+; CHECK-LE-LABEL: test82:
+; CHECK-LE:       # %bb.0: # %entry
+; CHECK-LE-NEXT:    xscmpudp cr0, f3, f4
+; CHECK-LE-NEXT:    beqlr cr0
+; CHECK-LE-NEXT:  # %bb.1: # %entry
+; CHECK-LE-NEXT:    fmr f1, f2
+; CHECK-LE-NEXT:    blr
 entry:
   %m = fcmp oeq double %c, %d
   %v = select i1 %m, double %a, double %b
   ret double %v
 
-; CHECK-REG-LABEL: @test82
-; CHECK-REG: xscmpudp cr0, f3, f4
-; CHECK-REG: beqlr cr0
 
-; CHECK-FISL-LABEL: @test82
-; CHECK-FISL: xscmpudp cr0, f3, f4
-; CHECK-FISL: beq cr0
 
-; CHECK-LE-LABEL: @test82
-; CHECK-LE: xscmpudp cr0, f3, f4
-; CHECK-LE: beqlr cr0
 }

diff --git a/llvm/test/CodeGen/SystemZ/swifterror.ll b/llvm/test/CodeGen/SystemZ/swifterror.ll
index eac84ef..613e869 100644 (file)
--- a/llvm/test/CodeGen/SystemZ/swifterror.ll
+++ b/llvm/test/CodeGen/SystemZ/swifterror.ll
@@ -16,7 +16,7 @@ define float @foo(%swift_error** swifterror %error_ptr_ref) {
 ; CHECK-O0-LABEL: foo:
 ; CHECK-O0: lghi %r2, 16
 ; CHECK-O0: brasl %r14, malloc
-; CHECK-O0: lgr %r9, %r2
+; CHECK-O0: lgr %r0, %r2
 ; CHECK-O0: mvi 8(%r2), 1
 entry:
   %call = call i8* @malloc(i64 16)

diff --git a/llvm/test/CodeGen/X86/atomic-unordered.ll b/llvm/test/CodeGen/X86/atomic-unordered.ll
index a85ddf7..430e45b 100644 (file)
--- a/llvm/test/CodeGen/X86/atomic-unordered.ll
+++ b/llvm/test/CodeGen/X86/atomic-unordered.ll
@@ -604,9 +604,8 @@ define void @widen_zero_init_unaligned(i32* %p0, i32 %v1, i32 %v2) {
 define i64 @load_fold_add1(i64* %p) {
 ; CHECK-O0-LABEL: load_fold_add1:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    addq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    addq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_add1:
@@ -639,9 +638,8 @@ define i64 @load_fold_add2(i64* %p, i64 %v2) {
 define i64 @load_fold_add3(i64* %p1, i64* %p2) {
 ; CHECK-O0-LABEL: load_fold_add3:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    addq (%rsi), %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    addq (%rsi), %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_add3:
@@ -659,9 +657,8 @@ define i64 @load_fold_add3(i64* %p1, i64* %p2) {
 define i64 @load_fold_sub1(i64* %p) {
 ; CHECK-O0-LABEL: load_fold_sub1:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    subq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    subq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_sub1:
@@ -677,9 +674,8 @@ define i64 @load_fold_sub1(i64* %p) {
 define i64 @load_fold_sub2(i64* %p, i64 %v2) {
 ; CHECK-O0-LABEL: load_fold_sub2:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    subq %rsi, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    subq %rsi, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_sub2:
@@ -695,9 +691,8 @@ define i64 @load_fold_sub2(i64* %p, i64 %v2) {
 define i64 @load_fold_sub3(i64* %p1, i64* %p2) {
 ; CHECK-O0-LABEL: load_fold_sub3:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    subq (%rsi), %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    subq (%rsi), %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_sub3:
@@ -749,9 +744,8 @@ define i64 @load_fold_mul2(i64* %p, i64 %v2) {
 define i64 @load_fold_mul3(i64* %p1, i64* %p2) {
 ; CHECK-O0-LABEL: load_fold_mul3:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    imulq (%rsi), %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    imulq (%rsi), %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_mul3:
@@ -771,8 +765,8 @@ define i64 @load_fold_sdiv1(i64* %p) {
 ; CHECK-O0:       # %bb.0:
 ; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    cqto
-; CHECK-O0-NEXT:    movl $15, %edi
-; CHECK-O0-NEXT:    idivq %rdi
+; CHECK-O0-NEXT:    movl $15, %ecx
+; CHECK-O0-NEXT:    idivq %rcx
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_sdiv1:
@@ -862,8 +856,8 @@ define i64 @load_fold_udiv1(i64* %p) {
 ; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    xorl %ecx, %ecx
 ; CHECK-O0-NEXT:    movl %ecx, %edx
-; CHECK-O0-NEXT:    movl $15, %edi
-; CHECK-O0-NEXT:    divq %rdi
+; CHECK-O0-NEXT:    movl $15, %esi
+; CHECK-O0-NEXT:    divq %rsi
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_udiv1:
@@ -949,8 +943,8 @@ define i64 @load_fold_srem1(i64* %p) {
 ; CHECK-O0:       # %bb.0:
 ; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    cqto
-; CHECK-O0-NEXT:    movl $15, %edi
-; CHECK-O0-NEXT:    idivq %rdi
+; CHECK-O0-NEXT:    movl $15, %ecx
+; CHECK-O0-NEXT:    idivq %rcx
 ; CHECK-O0-NEXT:    movq %rdx, %rax
 ; CHECK-O0-NEXT:    retq
 ;
@@ -1049,8 +1043,8 @@ define i64 @load_fold_urem1(i64* %p) {
 ; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    xorl %ecx, %ecx
 ; CHECK-O0-NEXT:    movl %ecx, %edx
-; CHECK-O0-NEXT:    movl $15, %edi
-; CHECK-O0-NEXT:    divq %rdi
+; CHECK-O0-NEXT:    movl $15, %esi
+; CHECK-O0-NEXT:    divq %rsi
 ; CHECK-O0-NEXT:    movq %rdx, %rax
 ; CHECK-O0-NEXT:    retq
 ;
@@ -1144,9 +1138,8 @@ define i64 @load_fold_urem3(i64* %p1, i64* %p2) {
 define i64 @load_fold_shl1(i64* %p) {
 ; CHECK-O0-LABEL: load_fold_shl1:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    shlq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    shlq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_shl1:
@@ -1162,11 +1155,10 @@ define i64 @load_fold_shl1(i64* %p) {
 define i64 @load_fold_shl2(i64* %p, i64 %v2) {
 ; CHECK-O0-LABEL: load_fold_shl2:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
+; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    movq %rsi, %rcx
 ; CHECK-O0-NEXT:    # kill: def $cl killed $rcx
-; CHECK-O0-NEXT:    shlq %cl, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    shlq %cl, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_shl2:
@@ -1181,11 +1173,10 @@ define i64 @load_fold_shl2(i64* %p, i64 %v2) {
 define i64 @load_fold_shl3(i64* %p1, i64* %p2) {
 ; CHECK-O0-LABEL: load_fold_shl3:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
+; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    movq (%rsi), %rcx
 ; CHECK-O0-NEXT:    # kill: def $cl killed $rcx
-; CHECK-O0-NEXT:    shlq %cl, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    shlq %cl, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_shl3:
@@ -1203,9 +1194,8 @@ define i64 @load_fold_shl3(i64* %p1, i64* %p2) {
 define i64 @load_fold_lshr1(i64* %p) {
 ; CHECK-O0-LABEL: load_fold_lshr1:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    shrq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    shrq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_lshr1:
@@ -1221,11 +1211,10 @@ define i64 @load_fold_lshr1(i64* %p) {
 define i64 @load_fold_lshr2(i64* %p, i64 %v2) {
 ; CHECK-O0-LABEL: load_fold_lshr2:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
+; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    movq %rsi, %rcx
 ; CHECK-O0-NEXT:    # kill: def $cl killed $rcx
-; CHECK-O0-NEXT:    shrq %cl, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    shrq %cl, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_lshr2:
@@ -1240,11 +1229,10 @@ define i64 @load_fold_lshr2(i64* %p, i64 %v2) {
 define i64 @load_fold_lshr3(i64* %p1, i64* %p2) {
 ; CHECK-O0-LABEL: load_fold_lshr3:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
+; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    movq (%rsi), %rcx
 ; CHECK-O0-NEXT:    # kill: def $cl killed $rcx
-; CHECK-O0-NEXT:    shrq %cl, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    shrq %cl, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_lshr3:
@@ -1262,9 +1250,8 @@ define i64 @load_fold_lshr3(i64* %p1, i64* %p2) {
 define i64 @load_fold_ashr1(i64* %p) {
 ; CHECK-O0-LABEL: load_fold_ashr1:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    sarq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    sarq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_ashr1:
@@ -1280,11 +1267,10 @@ define i64 @load_fold_ashr1(i64* %p) {
 define i64 @load_fold_ashr2(i64* %p, i64 %v2) {
 ; CHECK-O0-LABEL: load_fold_ashr2:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
+; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    movq %rsi, %rcx
 ; CHECK-O0-NEXT:    # kill: def $cl killed $rcx
-; CHECK-O0-NEXT:    sarq %cl, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    sarq %cl, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_ashr2:
@@ -1299,11 +1285,10 @@ define i64 @load_fold_ashr2(i64* %p, i64 %v2) {
 define i64 @load_fold_ashr3(i64* %p1, i64* %p2) {
 ; CHECK-O0-LABEL: load_fold_ashr3:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
+; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    movq (%rsi), %rcx
 ; CHECK-O0-NEXT:    # kill: def $cl killed $rcx
-; CHECK-O0-NEXT:    sarq %cl, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    sarq %cl, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_ashr3:
@@ -1321,9 +1306,8 @@ define i64 @load_fold_ashr3(i64* %p1, i64* %p2) {
 define i64 @load_fold_and1(i64* %p) {
 ; CHECK-O0-LABEL: load_fold_and1:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    andq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    andq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_and1:
@@ -1356,9 +1340,8 @@ define i64 @load_fold_and2(i64* %p, i64 %v2) {
 define i64 @load_fold_and3(i64* %p1, i64* %p2) {
 ; CHECK-O0-LABEL: load_fold_and3:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    andq (%rsi), %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    andq (%rsi), %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_and3:
@@ -1376,9 +1359,8 @@ define i64 @load_fold_and3(i64* %p1, i64* %p2) {
 define i64 @load_fold_or1(i64* %p) {
 ; CHECK-O0-LABEL: load_fold_or1:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    orq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    orq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_or1:
@@ -1411,9 +1393,8 @@ define i64 @load_fold_or2(i64* %p, i64 %v2) {
 define i64 @load_fold_or3(i64* %p1, i64* %p2) {
 ; CHECK-O0-LABEL: load_fold_or3:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    orq (%rsi), %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    orq (%rsi), %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_or3:
@@ -1431,9 +1412,8 @@ define i64 @load_fold_or3(i64* %p1, i64* %p2) {
 define i64 @load_fold_xor1(i64* %p) {
 ; CHECK-O0-LABEL: load_fold_xor1:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    xorq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    xorq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_xor1:
@@ -1466,9 +1446,8 @@ define i64 @load_fold_xor2(i64* %p, i64 %v2) {
 define i64 @load_fold_xor3(i64* %p1, i64* %p2) {
 ; CHECK-O0-LABEL: load_fold_xor3:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    xorq (%rsi), %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    xorq (%rsi), %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_xor3:
@@ -1485,10 +1464,11 @@ define i64 @load_fold_xor3(i64* %p1, i64* %p2) {
 define i1 @load_fold_icmp1(i64* %p) {
 ; CHECK-O0-LABEL: load_fold_icmp1:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    subq $15, %rdi
-; CHECK-O0-NEXT:    sete %al
-; CHECK-O0-NEXT:    movq %rdi, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    subq $15, %rax
+; CHECK-O0-NEXT:    sete %cl
+; CHECK-O0-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
+; CHECK-O0-NEXT:    movb %cl, %al
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_icmp1:
@@ -1504,10 +1484,11 @@ define i1 @load_fold_icmp1(i64* %p) {
 define i1 @load_fold_icmp2(i64* %p, i64 %v2) {
 ; CHECK-O0-LABEL: load_fold_icmp2:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    subq %rsi, %rdi
-; CHECK-O0-NEXT:    sete %al
-; CHECK-O0-NEXT:    movq %rdi, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    subq %rsi, %rax
+; CHECK-O0-NEXT:    sete %cl
+; CHECK-O0-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
+; CHECK-O0-NEXT:    movb %cl, %al
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_icmp2:
@@ -1523,11 +1504,12 @@ define i1 @load_fold_icmp2(i64* %p, i64 %v2) {
 define i1 @load_fold_icmp3(i64* %p1, i64* %p2) {
 ; CHECK-O0-LABEL: load_fold_icmp3:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    movq (%rsi), %rsi
-; CHECK-O0-NEXT:    subq %rsi, %rdi
-; CHECK-O0-NEXT:    sete %al
-; CHECK-O0-NEXT:    movq %rdi, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    movq (%rsi), %rcx
+; CHECK-O0-NEXT:    subq %rcx, %rax
+; CHECK-O0-NEXT:    sete %dl
+; CHECK-O0-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
+; CHECK-O0-NEXT:    movb %dl, %al
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: load_fold_icmp3:
@@ -1981,9 +1963,9 @@ define void @rmw_fold_shl2(i64* %p, i64 %v) {
 ; CHECK-O0:       # %bb.0:
 ; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    movb %sil, %cl
-; CHECK-O0-NEXT:    # implicit-def: $rsi
-; CHECK-O0-NEXT:    movb %cl, %sil
-; CHECK-O0-NEXT:    shlxq %rsi, %rax, %rax
+; CHECK-O0-NEXT:    # implicit-def: $rdx
+; CHECK-O0-NEXT:    movb %cl, %dl
+; CHECK-O0-NEXT:    shlxq %rdx, %rax, %rax
 ; CHECK-O0-NEXT:    movq %rax, (%rdi)
 ; CHECK-O0-NEXT:    retq
 ;
@@ -2026,9 +2008,9 @@ define void @rmw_fold_lshr2(i64* %p, i64 %v) {
 ; CHECK-O0:       # %bb.0:
 ; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    movb %sil, %cl
-; CHECK-O0-NEXT:    # implicit-def: $rsi
-; CHECK-O0-NEXT:    movb %cl, %sil
-; CHECK-O0-NEXT:    shrxq %rsi, %rax, %rax
+; CHECK-O0-NEXT:    # implicit-def: $rdx
+; CHECK-O0-NEXT:    movb %cl, %dl
+; CHECK-O0-NEXT:    shrxq %rdx, %rax, %rax
 ; CHECK-O0-NEXT:    movq %rax, (%rdi)
 ; CHECK-O0-NEXT:    retq
 ;
@@ -2071,9 +2053,9 @@ define void @rmw_fold_ashr2(i64* %p, i64 %v) {
 ; CHECK-O0:       # %bb.0:
 ; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    movb %sil, %cl
-; CHECK-O0-NEXT:    # implicit-def: $rsi
-; CHECK-O0-NEXT:    movb %cl, %sil
-; CHECK-O0-NEXT:    sarxq %rsi, %rax, %rax
+; CHECK-O0-NEXT:    # implicit-def: $rdx
+; CHECK-O0-NEXT:    movb %cl, %dl
+; CHECK-O0-NEXT:    sarxq %rdx, %rax, %rax
 ; CHECK-O0-NEXT:    movq %rax, (%rdi)
 ; CHECK-O0-NEXT:    retq
 ;
@@ -2214,8 +2196,9 @@ define void @rmw_fold_xor2(i64* %p, i64 %v) {
 define i32 @fold_trunc(i64* %p) {
 ; CHECK-O0-LABEL: fold_trunc:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    movl %edi, %eax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    movl %eax, %ecx
+; CHECK-O0-NEXT:    movl %ecx, %eax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: fold_trunc:
@@ -2232,9 +2215,10 @@ define i32 @fold_trunc(i64* %p) {
 define i32 @fold_trunc_add(i64* %p, i32 %v2) {
 ; CHECK-O0-LABEL: fold_trunc_add:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    movl %edi, %eax
-; CHECK-O0-NEXT:    addl %esi, %eax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    movl %eax, %ecx
+; CHECK-O0-NEXT:    addl %esi, %ecx
+; CHECK-O0-NEXT:    movl %ecx, %eax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: fold_trunc_add:
@@ -2253,9 +2237,10 @@ define i32 @fold_trunc_add(i64* %p, i32 %v2) {
 define i32 @fold_trunc_and(i64* %p, i32 %v2) {
 ; CHECK-O0-LABEL: fold_trunc_and:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    movl %edi, %eax
-; CHECK-O0-NEXT:    andl %esi, %eax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    movl %eax, %ecx
+; CHECK-O0-NEXT:    andl %esi, %ecx
+; CHECK-O0-NEXT:    movl %ecx, %eax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: fold_trunc_and:
@@ -2274,9 +2259,10 @@ define i32 @fold_trunc_and(i64* %p, i32 %v2) {
 define i32 @fold_trunc_or(i64* %p, i32 %v2) {
 ; CHECK-O0-LABEL: fold_trunc_or:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    movl %edi, %eax
-; CHECK-O0-NEXT:    orl %esi, %eax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    movl %eax, %ecx
+; CHECK-O0-NEXT:    orl %esi, %ecx
+; CHECK-O0-NEXT:    movl %ecx, %eax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: fold_trunc_or:
@@ -2296,12 +2282,12 @@ define i32 @fold_trunc_or(i64* %p, i32 %v2) {
 define i32 @split_load(i64* %p) {
 ; CHECK-O0-LABEL: split_load:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
-; CHECK-O0-NEXT:    movb %dil, %al
-; CHECK-O0-NEXT:    shrq $32, %rdi
-; CHECK-O0-NEXT:    movb %dil, %cl
-; CHECK-O0-NEXT:    orb %cl, %al
-; CHECK-O0-NEXT:    movzbl %al, %eax
+; CHECK-O0-NEXT:    movq (%rdi), %rax
+; CHECK-O0-NEXT:    movb %al, %cl
+; CHECK-O0-NEXT:    shrq $32, %rax
+; CHECK-O0-NEXT:    movb %al, %dl
+; CHECK-O0-NEXT:    orb %dl, %cl
+; CHECK-O0-NEXT:    movzbl %cl, %eax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: split_load:
@@ -2424,10 +2410,9 @@ define void @dead_store(i64* %p, i64 %v) {
 define i64 @nofold_fence(i64* %p) {
 ; CHECK-O0-LABEL: nofold_fence:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
+; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    mfence
-; CHECK-O0-NEXT:    addq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    addq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: nofold_fence:
@@ -2445,10 +2430,9 @@ define i64 @nofold_fence(i64* %p) {
 define i64 @nofold_fence_acquire(i64* %p) {
 ; CHECK-O0-LABEL: nofold_fence_acquire:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
+; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    #MEMBARRIER
-; CHECK-O0-NEXT:    addq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    addq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: nofold_fence_acquire:
@@ -2467,10 +2451,9 @@ define i64 @nofold_fence_acquire(i64* %p) {
 define i64 @nofold_stfence(i64* %p) {
 ; CHECK-O0-LABEL: nofold_stfence:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
+; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    #MEMBARRIER
-; CHECK-O0-NEXT:    addq $15, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    addq $15, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: nofold_stfence:
@@ -2570,10 +2553,9 @@ define i64 @fold_invariant_clobber(i64* dereferenceable(8) %p, i64 %arg) {
 define i64 @fold_invariant_fence(i64* dereferenceable(8) %p, i64 %arg) {
 ; CHECK-O0-LABEL: fold_invariant_fence:
 ; CHECK-O0:       # %bb.0:
-; CHECK-O0-NEXT:    movq (%rdi), %rdi
+; CHECK-O0-NEXT:    movq (%rdi), %rax
 ; CHECK-O0-NEXT:    mfence
-; CHECK-O0-NEXT:    addq %rsi, %rdi
-; CHECK-O0-NEXT:    movq %rdi, %rax
+; CHECK-O0-NEXT:    addq %rsi, %rax
 ; CHECK-O0-NEXT:    retq
 ;
 ; CHECK-O3-LABEL: fold_invariant_fence:

diff --git a/llvm/test/CodeGen/X86/avx-load-store.ll b/llvm/test/CodeGen/X86/avx-load-store.ll
index 1b28c72..7b18b9c 100644 (file)
--- a/llvm/test/CodeGen/X86/avx-load-store.ll
+++ b/llvm/test/CodeGen/X86/avx-load-store.ll
@@ -45,15 +45,15 @@ define void @test_256_load(double* nocapture %d, float* nocapture %f, <4 x i64>*
 ; CHECK_O0-NEXT:    movq %rsi, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; CHECK_O0-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; CHECK_O0-NEXT:    callq dummy
-; CHECK_O0-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rdx # 8-byte Reload
+; CHECK_O0-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rax # 8-byte Reload
 ; CHECK_O0-NEXT:    vmovups {{[-0-9]+}}(%r{{[sb]}}p), %ymm0 # 32-byte Reload
-; CHECK_O0-NEXT:    vmovapd %ymm0, (%rdx)
-; CHECK_O0-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rsi # 8-byte Reload
+; CHECK_O0-NEXT:    vmovapd %ymm0, (%rax)
+; CHECK_O0-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rcx # 8-byte Reload
 ; CHECK_O0-NEXT:    vmovups {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
-; CHECK_O0-NEXT:    vmovaps %ymm1, (%rsi)
-; CHECK_O0-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rdi # 8-byte Reload
+; CHECK_O0-NEXT:    vmovaps %ymm1, (%rcx)
+; CHECK_O0-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rdx # 8-byte Reload
 ; CHECK_O0-NEXT:    vmovups {{[-0-9]+}}(%r{{[sb]}}p), %ymm2 # 32-byte Reload
-; CHECK_O0-NEXT:    vmovdqa %ymm2, (%rdi)
+; CHECK_O0-NEXT:    vmovdqa %ymm2, (%rdx)
 ; CHECK_O0-NEXT:    addq $152, %rsp
 ; CHECK_O0-NEXT:    vzeroupper
 ; CHECK_O0-NEXT:    retq

diff --git a/llvm/test/CodeGen/X86/extend-set-cc-uses-dbg.ll b/llvm/test/CodeGen/X86/extend-set-cc-uses-dbg.ll
index 149ef7f..7d05a86 100644 (file)
--- a/llvm/test/CodeGen/X86/extend-set-cc-uses-dbg.ll
+++ b/llvm/test/CodeGen/X86/extend-set-cc-uses-dbg.ll
@@ -7,8 +7,8 @@ define void @foo(i32* %p) !dbg !4 {
 bb:
   %tmp = load i32, i32* %p, align 4, !dbg !7
   ; CHECK: $eax = MOV32rm killed {{.*}} $rdi, {{.*}} debug-location !7 :: (load 4 from %ir.p)
-  ; CHECK-NEXT: $edi = MOV32rr killed $eax, implicit-def $rdi, debug-location !7
-  ; CHECK-NEXT: $rcx = MOV64rr $rdi, debug-location !7
+  ; CHECK-NEXT: $ecx = MOV32rr killed $eax, implicit-def $rcx, debug-location !7
+  ; CHECK-NEXT: $rdx = MOV64rr $rcx, debug-location !7
 
   switch i32 %tmp, label %bb7 [
     i32 0, label %bb1

diff --git a/llvm/test/CodeGen/X86/fast-isel-select.ll b/llvm/test/CodeGen/X86/fast-isel-select.ll
index cf459f8..88cb4e0 100644 (file)
--- a/llvm/test/CodeGen/X86/fast-isel-select.ll
+++ b/llvm/test/CodeGen/X86/fast-isel-select.ll
@@ -11,12 +11,12 @@ define i32 @fastisel_select(i1 %exchSub2211_, i1 %trunc_8766) {
 ; CHECK:       ## %bb.0:
 ; CHECK-NEXT:    movb %sil, %al
 ; CHECK-NEXT:    movb %dil, %cl
-; CHECK-NEXT:    xorl %esi, %esi
+; CHECK-NEXT:    xorl %edx, %edx
 ; CHECK-NEXT:    subb %al, %cl
 ; CHECK-NEXT:    testb $1, %cl
-; CHECK-NEXT:    movl $1204476887, %edi ## imm = 0x47CADBD7
-; CHECK-NEXT:    cmovnel %edi, %esi
-; CHECK-NEXT:    movl %esi, %eax
+; CHECK-NEXT:    movl $1204476887, %esi ## imm = 0x47CADBD7
+; CHECK-NEXT:    cmovnel %esi, %edx
+; CHECK-NEXT:    movl %edx, %eax
 ; CHECK-NEXT:    retq
   %shuffleInternal15257_8932 = sub i1 %exchSub2211_, %trunc_8766
   %counter_diff1345 = select i1 %shuffleInternal15257_8932, i32 1204476887, i32 0

diff --git a/llvm/test/CodeGen/X86/fast-isel-x86-64.ll b/llvm/test/CodeGen/X86/fast-isel-x86-64.ll
index b1f380e..30c8af2 100644 (file)
--- a/llvm/test/CodeGen/X86/fast-isel-x86-64.ll
+++ b/llvm/test/CodeGen/X86/fast-isel-x86-64.ll
@@ -13,7 +13,7 @@ define i32 @test1(i32 %i) nounwind ssp {
 }
 
 ; CHECK-LABEL: test1:
-; CHECK: andl  $8, 
+; CHECK: andl  $8,
 
 
 ; rdar://9289512 - The load should fold into the compare.
@@ -119,14 +119,14 @@ define i32 @test10(i32 %X) nounwind {
   %Y = udiv i32 %X, 8
   ret i32 %Y
 ; CHECK-LABEL: test10:
-; CHECK: shrl  $3, 
+; CHECK: shrl  $3,
 }
 
 define i32 @test11(i32 %X) nounwind {
   %Y = sdiv exact i32 %X, 8
   ret i32 %Y
 ; CHECK-LABEL: test11:
-; CHECK: sarl  $3, 
+; CHECK: sarl  $3,
 }
 
 
@@ -168,7 +168,7 @@ entry:
   call void @test13f(i1 zeroext %tobool) noredzone
   ret void
 ; CHECK-LABEL: test14:
-; CHECK: andb  $1, 
+; CHECK: andb  $1,
 ; CHECK: callq
 }
 
@@ -227,7 +227,7 @@ if.else:                                          ; preds = %entry
 ; CHECK: movl  (%rdi), %eax
 ; CHECK: callq _foo
 ; CHECK: cmpl  $5, %eax
-; CHECK-NEXT: je 
+; CHECK-NEXT: je
 }
 
 ; Check that 0.0 is materialized using xorps
@@ -299,8 +299,8 @@ define void @test23(i8* noalias sret %result) {
 ; CHECK-LABEL: test23:
 ; CHECK: movq %rdi, [[STACK:[0-9]+\(%rsp\)]]
 ; CHECK: call
-; CHECK: movq [[STACK]], %rdi
-; CHECK: movq %rdi, %rax
+; CHECK: movq [[STACK]], %rcx
+; CHECK: movq %rcx, %rax
 ; CHECK: ret
 }
 

diff --git a/llvm/test/CodeGen/X86/pr27591.ll b/llvm/test/CodeGen/X86/pr27591.ll
index a925bb8..97ad681 100644 (file)
--- a/llvm/test/CodeGen/X86/pr27591.ll
+++ b/llvm/test/CodeGen/X86/pr27591.ll
@@ -9,8 +9,9 @@ define void @test1(i32 %x) #0 {
 ; CHECK-NEXT:    pushq %rax
 ; CHECK-NEXT:    cmpl $0, %edi
 ; CHECK-NEXT:    setne %al
-; CHECK-NEXT:    movzbl %al, %edi
-; CHECK-NEXT:    andl $1, %edi
+; CHECK-NEXT:    movzbl %al, %ecx
+; CHECK-NEXT:    andl $1, %ecx
+; CHECK-NEXT:    movl %ecx, %edi
 ; CHECK-NEXT:    callq callee1
 ; CHECK-NEXT:    popq %rax
 ; CHECK-NEXT:    retq
@@ -26,9 +27,10 @@ define void @test2(i32 %x) #0 {
 ; CHECK-NEXT:    pushq %rax
 ; CHECK-NEXT:    cmpl $0, %edi
 ; CHECK-NEXT:    setne %al
-; CHECK-NEXT:    movzbl %al, %edi
-; CHECK-NEXT:    andl $1, %edi
-; CHECK-NEXT:    negl %edi
+; CHECK-NEXT:    movzbl %al, %ecx
+; CHECK-NEXT:    andl $1, %ecx
+; CHECK-NEXT:    negl %ecx
+; CHECK-NEXT:    movl %ecx, %edi
 ; CHECK-NEXT:    callq callee2
 ; CHECK-NEXT:    popq %rax
 ; CHECK-NEXT:    retq

diff --git a/llvm/test/CodeGen/X86/swift-return.ll b/llvm/test/CodeGen/X86/swift-return.ll
index 77591d3..fb504ca 100644 (file)
--- a/llvm/test/CodeGen/X86/swift-return.ll
+++ b/llvm/test/CodeGen/X86/swift-return.ll
@@ -28,10 +28,10 @@ define i16 @test(i32 %key) {
 ; CHECK-O0-NEXT:    movl %edi, {{[0-9]+}}(%rsp)
 ; CHECK-O0-NEXT:    movl {{[0-9]+}}(%rsp), %edi
 ; CHECK-O0-NEXT:    callq gen
-; CHECK-O0-NEXT:    movswl %ax, %edi
-; CHECK-O0-NEXT:    movsbl %dl, %ecx
-; CHECK-O0-NEXT:    addl %ecx, %edi
-; CHECK-O0-NEXT:    movw %di, %ax
+; CHECK-O0-NEXT:    movswl %ax, %ecx
+; CHECK-O0-NEXT:    movsbl %dl, %esi
+; CHECK-O0-NEXT:    addl %esi, %ecx
+; CHECK-O0-NEXT:    movw %cx, %ax
 ; CHECK-O0-NEXT:    popq %rcx
 ; CHECK-O0-NEXT:    .cfi_def_cfa_offset 8
 ; CHECK-O0-NEXT:    retq
@@ -79,16 +79,16 @@ define i32 @test2(i32 %key) #0 {
 ; CHECK-O0-NEXT:    movl {{[0-9]+}}(%rsp), %edi
 ; CHECK-O0-NEXT:    movq %rsp, %rax
 ; CHECK-O0-NEXT:    callq gen2
-; CHECK-O0-NEXT:    movl {{[0-9]+}}(%rsp), %edi
 ; CHECK-O0-NEXT:    movl {{[0-9]+}}(%rsp), %ecx
 ; CHECK-O0-NEXT:    movl {{[0-9]+}}(%rsp), %edx
-; CHECK-O0-NEXT:    movl (%rsp), %esi
+; CHECK-O0-NEXT:    movl {{[0-9]+}}(%rsp), %esi
+; CHECK-O0-NEXT:    movl (%rsp), %edi
 ; CHECK-O0-NEXT:    movl {{[0-9]+}}(%rsp), %r8d
-; CHECK-O0-NEXT:    addl %r8d, %esi
-; CHECK-O0-NEXT:    addl %edx, %esi
-; CHECK-O0-NEXT:    addl %ecx, %esi
-; CHECK-O0-NEXT:    addl %edi, %esi
-; CHECK-O0-NEXT:    movl %esi, %eax
+; CHECK-O0-NEXT:    addl %r8d, %edi
+; CHECK-O0-NEXT:    addl %esi, %edi
+; CHECK-O0-NEXT:    addl %edx, %edi
+; CHECK-O0-NEXT:    addl %ecx, %edi
+; CHECK-O0-NEXT:    movl %edi, %eax
 ; CHECK-O0-NEXT:    addq $24, %rsp
 ; CHECK-O0-NEXT:    .cfi_def_cfa_offset 8
 ; CHECK-O0-NEXT:    retq

diff --git a/llvm/test/DebugInfo/X86/op_deref.ll b/llvm/test/DebugInfo/X86/op_deref.ll
index 1b49dc5..5de9976 100644 (file)
--- a/llvm/test/DebugInfo/X86/op_deref.ll
+++ b/llvm/test/DebugInfo/X86/op_deref.ll
@@ -6,10 +6,10 @@
 ; RUN:     | FileCheck %s -check-prefix=CHECK -check-prefix=DWARF3
 
 ; DWARF4: DW_AT_location [DW_FORM_sec_offset]                      (0x00000000
-; DWARF4-NEXT:  {{.*}}: DW_OP_breg2 RCX+0, DW_OP_deref
+; DWARF4-NEXT:  {{.*}}: DW_OP_breg1 RDX+0, DW_OP_deref
 
 ; DWARF3: DW_AT_location [DW_FORM_data4]                      (0x00000000
-; DWARF3-NEXT:  {{.*}}: DW_OP_breg2 RCX+0, DW_OP_deref
+; DWARF3-NEXT:  {{.*}}: DW_OP_breg1 RDX+0, DW_OP_deref
 
 ; CHECK-NOT: DW_TAG
 ; CHECK: DW_AT_name [DW_FORM_strp]  ( .debug_str[0x00000067] = "vla")
@@ -17,8 +17,8 @@
 ; Check the DEBUG_VALUE comments for good measure.
 ; RUN: llc -O0 -mtriple=x86_64-apple-darwin %s -o - -filetype=asm | FileCheck %s -check-prefix=ASM-CHECK
 ; vla should have a register-indirect address at one point.
-; ASM-CHECK: DEBUG_VALUE: vla <- [DW_OP_deref] [$rcx+0]
-; ASM-CHECK: DW_OP_breg2
+; ASM-CHECK: DEBUG_VALUE: vla <- [DW_OP_deref] [$rdx+0]
+; ASM-CHECK: DW_OP_breg1
 
 ; RUN: llvm-as %s -o - | llvm-dis - | FileCheck %s --check-prefix=PRETTY-PRINT
 ; PRETTY-PRINT: DIExpression(DW_OP_deref)