diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -24698,6 +24698,8 @@
   // (select (x != 0), -1, y) -> ~(sign_bit (x - 1)) | y
   // (select (and (x , 0x1) == 0), y, (z ^ y) ) -> (-(and (x , 0x1)) & z ) ^ y
   // (select (and (x , 0x1) == 0), y, (z | y) ) -> (-(and (x , 0x1)) & z ) | y
+  // (select (x > 0), x, 0) -> (~(x << (size_in_bits(x)-1))) & x
+  // (select (x < 0), x, 0) -> ((x << (size_in_bits(x)-1))) & x
   if (Cond.getOpcode() == X86ISD::SETCC &&
       Cond.getOperand(1).getOpcode() == X86ISD::CMP &&
       isNullConstant(Cond.getOperand(1).getOperand(1))) {
@@ -24778,6 +24780,21 @@
         SDValue And = DAG.getNode(ISD::AND, DL, VT, Mask, Src1); // Mask & z
         return DAG.getNode(Op2.getOpcode(), DL, VT, And, Src2);  // And Op y
       }
+    } else if ((VT == MVT::i32 || VT == MVT::i64) && isNullConstant(Op2) &&
+        (CmpOp0 == Op1) && ((CondCode == X86::COND_S) ||    // smin(x, 0)
+            (CondCode == X86::COND_G && hasAndNot(Op1)))) { // smax(x, 0)
+      // (select (x < 0), x, 0) -> ((x << (size_in_bits(x)-1))) & x
+      //
+      // If the comparison is testing for a positive value, we have to invert
+      // the sign bit mask, so only do that transform if the target has a
+      // bitwise 'and not' instruction (the invert is free).
+      // (select (x > 0), x, 0) -> (~(x << (size_in_bits(x)-1))) & x
+      unsigned ShCt = VT.getSizeInBits() - 1;
+      SDValue ShiftAmt = DAG.getConstant(ShCt, DL, VT);
+      SDValue Shift = DAG.getNode(ISD::SRA, DL, VT, Op1, ShiftAmt);
+      if (CondCode == X86::COND_G)
+        Shift = DAG.getNOT(DL, Shift, VT);
+      return DAG.getNode(ISD::AND, DL, VT, Shift, Op1);
     }
   }
 
diff --git a/llvm/test/CodeGen/X86/fpclamptosat_vec.ll b/llvm/test/CodeGen/X86/fpclamptosat_vec.ll
--- a/llvm/test/CodeGen/X86/fpclamptosat_vec.ll
+++ b/llvm/test/CodeGen/X86/fpclamptosat_vec.ll
@@ -2974,33 +2974,36 @@
 ; CHECK-NEXT:    cmpq %rcx, %rax
 ; CHECK-NEXT:    movq %rcx, %rsi
 ; CHECK-NEXT:    cmovbq %rax, %rsi
-; CHECK-NEXT:    xorl %edi, %edi
 ; CHECK-NEXT:    testq %rdx, %rdx
 ; CHECK-NEXT:    cmovnsq %rcx, %rax
 ; CHECK-NEXT:    cmoveq %rsi, %rax
-; CHECK-NEXT:    cmovnsq %rdi, %rdx
 ; CHECK-NEXT:    cmpq %rcx, %rbx
 ; CHECK-NEXT:    movq %rcx, %rsi
 ; CHECK-NEXT:    cmovbq %rbx, %rsi
 ; CHECK-NEXT:    testq %r14, %r14
 ; CHECK-NEXT:    cmovsq %rbx, %rcx
 ; CHECK-NEXT:    cmoveq %rsi, %rcx
-; CHECK-NEXT:    cmovsq %r14, %rdi
-; CHECK-NEXT:    testq %rdi, %rdi
+; CHECK-NEXT:    movq %rdx, %rsi
+; CHECK-NEXT:    sarq $63, %rsi
+; CHECK-NEXT:    andq %rdx, %rsi
+; CHECK-NEXT:    movq %r14, %rdx
+; CHECK-NEXT:    sarq $63, %rdx
+; CHECK-NEXT:    andq %r14, %rdx
 ; CHECK-NEXT:    movabsq $-9223372036854775808, %rbx # imm = 0x8000000000000000
-; CHECK-NEXT:    movq %rbx, %rsi
-; CHECK-NEXT:    cmovnsq %rcx, %rsi
 ; CHECK-NEXT:    cmpq %rbx, %rcx
-; CHECK-NEXT:    cmovbeq %rbx, %rcx
-; CHECK-NEXT:    cmpq $-1, %rdi
-; CHECK-NEXT:    cmovneq %rsi, %rcx
+; CHECK-NEXT:    movq %rbx, %rdi
+; CHECK-NEXT:    cmovaq %rcx, %rdi
 ; CHECK-NEXT:    testq %rdx, %rdx
-; CHECK-NEXT:    movq %rbx, %rsi
-; CHECK-NEXT:    cmovnsq %rax, %rsi
-; CHECK-NEXT:    cmpq %rbx, %rax
-; CHECK-NEXT:    cmovbeq %rbx, %rax
+; CHECK-NEXT:    cmovsq %rbx, %rcx
 ; CHECK-NEXT:    cmpq $-1, %rdx
-; CHECK-NEXT:    cmovneq %rsi, %rax
+; CHECK-NEXT:    cmoveq %rdi, %rcx
+; CHECK-NEXT:    cmpq %rbx, %rax
+; CHECK-NEXT:    movq %rbx, %rdx
+; CHECK-NEXT:    cmovaq %rax, %rdx
+; CHECK-NEXT:    testq %rsi, %rsi
+; CHECK-NEXT:    cmovsq %rbx, %rax
+; CHECK-NEXT:    cmpq $-1, %rsi
+; CHECK-NEXT:    cmoveq %rdx, %rax
 ; CHECK-NEXT:    movq %rax, %xmm0
 ; CHECK-NEXT:    movq %rcx, %xmm1
 ; CHECK-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
@@ -3138,33 +3141,36 @@
 ; CHECK-NEXT:    cmpq %rcx, %rax
 ; CHECK-NEXT:    movq %rcx, %rsi
 ; CHECK-NEXT:    cmovbq %rax, %rsi
-; CHECK-NEXT:    xorl %edi, %edi
 ; CHECK-NEXT:    testq %rdx, %rdx
 ; CHECK-NEXT:    cmovnsq %rcx, %rax
 ; CHECK-NEXT:    cmoveq %rsi, %rax
-; CHECK-NEXT:    cmovnsq %rdi, %rdx
 ; CHECK-NEXT:    cmpq %rcx, %rbx
 ; CHECK-NEXT:    movq %rcx, %rsi
 ; CHECK-NEXT:    cmovbq %rbx, %rsi
 ; CHECK-NEXT:    testq %r14, %r14
 ; CHECK-NEXT:    cmovsq %rbx, %rcx
 ; CHECK-NEXT:    cmoveq %rsi, %rcx
-; CHECK-NEXT:    cmovsq %r14, %rdi
-; CHECK-NEXT:    testq %rdi, %rdi
+; CHECK-NEXT:    movq %rdx, %rsi
+; CHECK-NEXT:    sarq $63, %rsi
+; CHECK-NEXT:    andq %rdx, %rsi
+; CHECK-NEXT:    movq %r14, %rdx
+; CHECK-NEXT:    sarq $63, %rdx
+; CHECK-NEXT:    andq %r14, %rdx
 ; CHECK-NEXT:    movabsq $-9223372036854775808, %rbx # imm = 0x8000000000000000
-; CHECK-NEXT:    movq %rbx, %rsi
-; CHECK-NEXT:    cmovnsq %rcx, %rsi
 ; CHECK-NEXT:    cmpq %rbx, %rcx
-; CHECK-NEXT:    cmovbeq %rbx, %rcx
-; CHECK-NEXT:    cmpq $-1, %rdi
-; CHECK-NEXT:    cmovneq %rsi, %rcx
+; CHECK-NEXT:    movq %rbx, %rdi
+; CHECK-NEXT:    cmovaq %rcx, %rdi
 ; CHECK-NEXT:    testq %rdx, %rdx
-; CHECK-NEXT:    movq %rbx, %rsi
-; CHECK-NEXT:    cmovnsq %rax, %rsi
-; CHECK-NEXT:    cmpq %rbx, %rax
-; CHECK-NEXT:    cmovbeq %rbx, %rax
+; CHECK-NEXT:    cmovsq %rbx, %rcx
 ; CHECK-NEXT:    cmpq $-1, %rdx
-; CHECK-NEXT:    cmovneq %rsi, %rax
+; CHECK-NEXT:    cmoveq %rdi, %rcx
+; CHECK-NEXT:    cmpq %rbx, %rax
+; CHECK-NEXT:    movq %rbx, %rdx
+; CHECK-NEXT:    cmovaq %rax, %rdx
+; CHECK-NEXT:    testq %rsi, %rsi
+; CHECK-NEXT:    cmovsq %rbx, %rax
+; CHECK-NEXT:    cmpq $-1, %rsi
+; CHECK-NEXT:    cmoveq %rdx, %rax
 ; CHECK-NEXT:    movq %rax, %xmm0
 ; CHECK-NEXT:    movq %rcx, %xmm1
 ; CHECK-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
@@ -3292,46 +3298,49 @@
 ; CHECK-NEXT:    .cfi_offset %rbx, -32
 ; CHECK-NEXT:    .cfi_offset %r14, -24
 ; CHECK-NEXT:    .cfi_offset %rbp, -16
-; CHECK-NEXT:    movl %esi, %ebp
+; CHECK-NEXT:    movl %esi, %r14d
 ; CHECK-NEXT:    movzwl %di, %edi
 ; CHECK-NEXT:    callq __gnu_h2f_ieee@PLT
 ; CHECK-NEXT:    callq __fixsfti@PLT
-; CHECK-NEXT:    movq %rax, %rbx
-; CHECK-NEXT:    movq %rdx, %r14
-; CHECK-NEXT:    movzwl %bp, %edi
+; CHECK-NEXT:    movq %rax, %rbp
+; CHECK-NEXT:    movq %rdx, %rbx
+; CHECK-NEXT:    movzwl %r14w, %edi
 ; CHECK-NEXT:    callq __gnu_h2f_ieee@PLT
 ; CHECK-NEXT:    callq __fixsfti@PLT
 ; CHECK-NEXT:    movabsq $9223372036854775807, %rcx # imm = 0x7FFFFFFFFFFFFFFF
 ; CHECK-NEXT:    cmpq %rcx, %rax
 ; CHECK-NEXT:    movq %rcx, %rsi
 ; CHECK-NEXT:    cmovbq %rax, %rsi
-; CHECK-NEXT:    xorl %edi, %edi
 ; CHECK-NEXT:    testq %rdx, %rdx
 ; CHECK-NEXT:    cmovnsq %rcx, %rax
 ; CHECK-NEXT:    cmoveq %rsi, %rax
-; CHECK-NEXT:    cmovnsq %rdi, %rdx
-; CHECK-NEXT:    cmpq %rcx, %rbx
+; CHECK-NEXT:    cmpq %rcx, %rbp
 ; CHECK-NEXT:    movq %rcx, %rsi
-; CHECK-NEXT:    cmovbq %rbx, %rsi
-; CHECK-NEXT:    testq %r14, %r14
-; CHECK-NEXT:    cmovsq %rbx, %rcx
+; CHECK-NEXT:    cmovbq %rbp, %rsi
+; CHECK-NEXT:    testq %rbx, %rbx
+; CHECK-NEXT:    cmovsq %rbp, %rcx
 ; CHECK-NEXT:    cmoveq %rsi, %rcx
-; CHECK-NEXT:    cmovsq %r14, %rdi
-; CHECK-NEXT:    testq %rdi, %rdi
+; CHECK-NEXT:    movq %rdx, %rsi
+; CHECK-NEXT:    sarq $63, %rsi
+; CHECK-NEXT:    andq %rdx, %rsi
+; CHECK-NEXT:    movq %rbx, %rdx
+; CHECK-NEXT:    sarq $63, %rdx
+; CHECK-NEXT:    andq %rbx, %rdx
 ; CHECK-NEXT:    movabsq $-9223372036854775808, %rbp # imm = 0x8000000000000000
-; CHECK-NEXT:    movq %rbp, %rsi
-; CHECK-NEXT:    cmovnsq %rcx, %rsi
 ; CHECK-NEXT:    cmpq %rbp, %rcx
-; CHECK-NEXT:    cmovbeq %rbp, %rcx
-; CHECK-NEXT:    cmpq $-1, %rdi
-; CHECK-NEXT:    cmovneq %rsi, %rcx
+; CHECK-NEXT:    movq %rbp, %rdi
+; CHECK-NEXT:    cmovaq %rcx, %rdi
 ; CHECK-NEXT:    testq %rdx, %rdx
-; CHECK-NEXT:    movq %rbp, %rsi
-; CHECK-NEXT:    cmovnsq %rax, %rsi
-; CHECK-NEXT:    cmpq %rbp, %rax
-; CHECK-NEXT:    cmovbeq %rbp, %rax
+; CHECK-NEXT:    cmovsq %rbp, %rcx
 ; CHECK-NEXT:    cmpq $-1, %rdx
-; CHECK-NEXT:    cmovneq %rsi, %rax
+; CHECK-NEXT:    cmoveq %rdi, %rcx
+; CHECK-NEXT:    cmpq %rbp, %rax
+; CHECK-NEXT:    movq %rbp, %rdx
+; CHECK-NEXT:    cmovaq %rax, %rdx
+; CHECK-NEXT:    testq %rsi, %rsi
+; CHECK-NEXT:    cmovsq %rbp, %rax
+; CHECK-NEXT:    cmpq $-1, %rsi
+; CHECK-NEXT:    cmoveq %rdx, %rax
 ; CHECK-NEXT:    movq %rax, %xmm1
 ; CHECK-NEXT:    movq %rcx, %xmm0
 ; CHECK-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
diff --git a/llvm/test/CodeGen/X86/sdiv_fix_sat.ll b/llvm/test/CodeGen/X86/sdiv_fix_sat.ll
--- a/llvm/test/CodeGen/X86/sdiv_fix_sat.ll
+++ b/llvm/test/CodeGen/X86/sdiv_fix_sat.ll
@@ -315,48 +315,49 @@
 ; X64-NEXT:    movq %rsi, (%rsp) # 8-byte Spill
 ; X64-NEXT:    movq %rdi, %r15
 ; X64-NEXT:    leaq (%rdi,%rdi), %rax
-; X64-NEXT:    movq %rdi, %r12
-; X64-NEXT:    sarq $63, %r12
-; X64-NEXT:    shldq $31, %rax, %r12
+; X64-NEXT:    movq %rdi, %rbx
+; X64-NEXT:    sarq $63, %rbx
+; X64-NEXT:    shldq $31, %rax, %rbx
 ; X64-NEXT:    shlq $32, %r15
-; X64-NEXT:    movq %rsi, %r13
-; X64-NEXT:    sarq $63, %r13
+; X64-NEXT:    movq %rsi, %r12
+; X64-NEXT:    sarq $63, %r12
 ; X64-NEXT:    movq %r15, %rdi
-; X64-NEXT:    movq %r12, %rsi
-; X64-NEXT:    movq %r13, %rcx
+; X64-NEXT:    movq %rbx, %rsi
+; X64-NEXT:    movq %r12, %rcx
 ; X64-NEXT:    callq __divti3@PLT
-; X64-NEXT:    movq %rax, %rbx
+; X64-NEXT:    movq %rax, %r13
 ; X64-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; X64-NEXT:    movq %rdx, %rbp
 ; X64-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    subq $1, %rbx
+; X64-NEXT:    subq $1, %r13
 ; X64-NEXT:    sbbq $0, %rbp
-; X64-NEXT:    testq %r12, %r12
+; X64-NEXT:    testq %rbx, %rbx
 ; X64-NEXT:    sets %al
-; X64-NEXT:    testq %r13, %r13
+; X64-NEXT:    testq %r12, %r12
 ; X64-NEXT:    sets %r14b
 ; X64-NEXT:    xorb %al, %r14b
 ; X64-NEXT:    movq %r15, %rdi
-; X64-NEXT:    movq %r12, %rsi
+; X64-NEXT:    movq %rbx, %rsi
 ; X64-NEXT:    movq (%rsp), %rdx # 8-byte Reload
-; X64-NEXT:    movq %r13, %rcx
+; X64-NEXT:    movq %r12, %rcx
 ; X64-NEXT:    callq __modti3@PLT
 ; X64-NEXT:    orq %rax, %rdx
 ; X64-NEXT:    setne %al
 ; X64-NEXT:    testb %r14b, %al
 ; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %rbp # 8-byte Folded Reload
-; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %rbx # 8-byte Folded Reload
-; X64-NEXT:    xorl %eax, %eax
-; X64-NEXT:    testq %rbp, %rbp
-; X64-NEXT:    cmovnsq %rax, %rbp
-; X64-NEXT:    movq $-1, %rcx
-; X64-NEXT:    cmovgq %rcx, %rbx
+; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r13 # 8-byte Folded Reload
+; X64-NEXT:    movq %rbp, %rcx
+; X64-NEXT:    sarq $63, %rcx
+; X64-NEXT:    andq %rbp, %rcx
+; X64-NEXT:    movq $-1, %rax
+; X64-NEXT:    movq $-1, %rdx
+; X64-NEXT:    cmovnsq %rcx, %rdx
 ; X64-NEXT:    testq %rbp, %rbp
-; X64-NEXT:    cmovnsq %rbp, %rcx
-; X64-NEXT:    cmpq $-1, %rbp
-; X64-NEXT:    cmovlq %rax, %rbx
-; X64-NEXT:    shrdq $1, %rcx, %rbx
-; X64-NEXT:    movq %rbx, %rax
+; X64-NEXT:    cmovgq %rax, %r13
+; X64-NEXT:    xorl %eax, %eax
+; X64-NEXT:    cmpq $-1, %rcx
+; X64-NEXT:    cmovgeq %r13, %rax
+; X64-NEXT:    shrdq $1, %rdx, %rax
 ; X64-NEXT:    addq $24, %rsp
 ; X64-NEXT:    popq %rbx
 ; X64-NEXT:    popq %r12
@@ -374,123 +375,119 @@
 ; X86-NEXT:    pushl %edi
 ; X86-NEXT:    pushl %esi
 ; X86-NEXT:    andl $-8, %esp
-; X86-NEXT:    subl $88, %esp
+; X86-NEXT:    subl $56, %esp
 ; X86-NEXT:    movl 8(%ebp), %ecx
-; X86-NEXT:    movl 12(%ebp), %eax
-; X86-NEXT:    movl 20(%ebp), %edi
-; X86-NEXT:    sarl $31, %edi
-; X86-NEXT:    movl %eax, %ebx
-; X86-NEXT:    sarl $31, %ebx
-; X86-NEXT:    movl %ebx, %edx
-; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    shldl $31, %eax, %edx
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    shldl $31, %ecx, %eax
-; X86-NEXT:    movl %eax, %esi
+; X86-NEXT:    movl 12(%ebp), %ebx
+; X86-NEXT:    movl 20(%ebp), %edx
+; X86-NEXT:    movl %edx, %esi
+; X86-NEXT:    sarl $31, %esi
+; X86-NEXT:    movl %ebx, %eax
+; X86-NEXT:    sarl $31, %eax
+; X86-NEXT:    movl %eax, %edi
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    shldl $31, %ebx, %edi
+; X86-NEXT:    shldl $31, %ecx, %ebx
 ; X86-NEXT:    shll $31, %ecx
 ; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    pushl %edi
-; X86-NEXT:    pushl %edi
-; X86-NEXT:    pushl 20(%ebp)
+; X86-NEXT:    pushl %esi
+; X86-NEXT:    pushl %esi
+; X86-NEXT:    pushl %edx
 ; X86-NEXT:    pushl 16(%ebp)
+; X86-NEXT:    pushl %eax
+; X86-NEXT:    pushl %edi
 ; X86-NEXT:    pushl %ebx
-; X86-NEXT:    pushl %edx
-; X86-NEXT:    pushl %esi
 ; X86-NEXT:    pushl %ecx
+; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    pushl %eax
-; X86-NEXT:    calll __divti3
+; X86-NEXT:    calll __modti3
 ; X86-NEXT:    addl $32, %esp
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %edx
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    subl $1, %esi
-; X86-NEXT:    movl %ecx, %eax
-; X86-NEXT:    sbbl $0, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %edx, %eax
-; X86-NEXT:    sbbl $0, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %ebx
-; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    sbbl $0, %ebx
-; X86-NEXT:    testl %edi, %edi
+; X86-NEXT:    testl %esi, %esi
 ; X86-NEXT:    sets %al
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    testl %ecx, %ecx
-; X86-NEXT:    sets %dl
-; X86-NEXT:    xorb %al, %dl
-; X86-NEXT:    movb %dl, {{[-0-9]+}}(%e{{[sb]}}p) # 1-byte Spill
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    testl %edx, %edx
+; X86-NEXT:    sets %cl
+; X86-NEXT:    xorb %al, %cl
+; X86-NEXT:    movb %cl, {{[-0-9]+}}(%e{{[sb]}}p) # 1-byte Spill
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
+; X86-NEXT:    orl {{[0-9]+}}(%esp), %ecx
+; X86-NEXT:    orl {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    orl %ecx, %eax
 ; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    pushl %edi
-; X86-NEXT:    pushl %edi
+; X86-NEXT:    setne (%esp) # 1-byte Folded Spill
+; X86-NEXT:    pushl %esi
+; X86-NEXT:    pushl %esi
 ; X86-NEXT:    pushl 20(%ebp)
 ; X86-NEXT:    pushl 16(%ebp)
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
-; X86-NEXT:    pushl {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
+; X86-NEXT:    pushl %edx
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %ebx
 ; X86-NEXT:    pushl {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
 ; X86-NEXT:    pushl %eax
-; X86-NEXT:    calll __modti3
+; X86-NEXT:    calll __divti3
 ; X86-NEXT:    addl $32, %esp
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    orl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
-; X86-NEXT:    orl {{[0-9]+}}(%esp), %ecx
-; X86-NEXT:    orl %eax, %ecx
-; X86-NEXT:    setne %al
-; X86-NEXT:    testb %al, {{[-0-9]+}}(%e{{[sb]}}p) # 1-byte Folded Reload
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
-; X86-NEXT:    testl %ebx, %ebx
-; X86-NEXT:    movl $0, %eax
-; X86-NEXT:    cmovsl %ebx, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl $2147483647, %eax # imm = 0x7FFFFFFF
-; X86-NEXT:    cmovsl %edx, %eax
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
+; X86-NEXT:    movl %eax, %ebx
+; X86-NEXT:    subl $1, %ebx
+; X86-NEXT:    movl %esi, %edx
+; X86-NEXT:    sbbl $0, %edx
+; X86-NEXT:    setb %cl
+; X86-NEXT:    movb (%esp), %ch # 1-byte Reload
+; X86-NEXT:    testb %ch, {{[-0-9]+}}(%e{{[sb]}}p) # 1-byte Folded Reload
+; X86-NEXT:    cmovel %eax, %ebx
+; X86-NEXT:    cmovel %esi, %edx
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl $-1, %eax
-; X86-NEXT:    cmovsl %esi, %eax
-; X86-NEXT:    movl %ebx, %edi
-; X86-NEXT:    sarl $31, %edi
-; X86-NEXT:    andl %ecx, %edi
-; X86-NEXT:    testl %ebx, %ebx
-; X86-NEXT:    cmovel %ebx, %edi
-; X86-NEXT:    movl %edx, %ecx
+; X86-NEXT:    addb $255, %cl
+; X86-NEXT:    sbbl $0, %eax
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    movl %edi, %esi
+; X86-NEXT:    sbbl $0, %esi
+; X86-NEXT:    movb (%esp), %cl # 1-byte Reload
+; X86-NEXT:    testb %cl, {{[-0-9]+}}(%e{{[sb]}}p) # 1-byte Folded Reload
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
+; X86-NEXT:    cmovel %edi, %esi
 ; X86-NEXT:    cmpl $2147483647, %edx # imm = 0x7FFFFFFF
-; X86-NEXT:    movl $2147483647, %edx # imm = 0x7FFFFFFF
-; X86-NEXT:    cmovbl %ecx, %edx
-; X86-NEXT:    testl %ecx, %ecx
+; X86-NEXT:    movl $2147483647, %ecx # imm = 0x7FFFFFFF
+; X86-NEXT:    cmovbl %edx, %ecx
+; X86-NEXT:    movl %ecx, (%esp) # 4-byte Spill
+; X86-NEXT:    testl %edx, %edx
 ; X86-NEXT:    movl $-1, %ecx
-; X86-NEXT:    cmovsl %ecx, %esi
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
-; X86-NEXT:    cmovnel %eax, %esi
-; X86-NEXT:    cmovnel {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
+; X86-NEXT:    cmovnsl %ebx, %ecx
+; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %esi, %edi
+; X86-NEXT:    sarl $31, %edi
+; X86-NEXT:    movl %edi, %ecx
+; X86-NEXT:    andl %eax, %ecx
+; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    testl %esi, %esi
+; X86-NEXT:    movl $2147483647, %ecx # imm = 0x7FFFFFFF
+; X86-NEXT:    cmovnsl %ecx, %edx
+; X86-NEXT:    movl $-1, %ecx
+; X86-NEXT:    cmovnsl %ecx, %ebx
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    cmovel %esi, %ecx
+; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    orl %esi, %eax
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
+; X86-NEXT:    cmovel (%esp), %edx # 4-byte Folded Reload
 ; X86-NEXT:    cmpl $-2147483647, %edx # imm = 0x80000001
 ; X86-NEXT:    movl $-2147483648, %eax # imm = 0x80000000
 ; X86-NEXT:    cmovael %edx, %eax
+; X86-NEXT:    movl %eax, (%esp) # 4-byte Spill
 ; X86-NEXT:    movl %edx, %ecx
 ; X86-NEXT:    sarl $31, %ecx
-; X86-NEXT:    andl %esi, %ecx
-; X86-NEXT:    cmpl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
-; X86-NEXT:    movl $-2147483648, %ebx # imm = 0x80000000
-; X86-NEXT:    cmovsl %ebx, %edx
-; X86-NEXT:    movl $0, %ebx
-; X86-NEXT:    cmovsl %ebx, %esi
+; X86-NEXT:    andl %ebx, %ecx
+; X86-NEXT:    xorl %eax, %eax
+; X86-NEXT:    andl %esi, %edi
+; X86-NEXT:    movl $-2147483648, %esi # imm = 0x80000000
+; X86-NEXT:    cmovsl %esi, %edx
+; X86-NEXT:    cmovnsl %ebx, %eax
 ; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Folded Reload
 ; X86-NEXT:    cmpl $-1, %edi
-; X86-NEXT:    cmovel %ecx, %esi
-; X86-NEXT:    cmovel %eax, %edx
-; X86-NEXT:    movl %esi, %eax
+; X86-NEXT:    cmovel %ecx, %eax
+; X86-NEXT:    cmovel (%esp), %edx # 4-byte Folded Reload
 ; X86-NEXT:    leal -12(%ebp), %esp
 ; X86-NEXT:    popl %esi
 ; X86-NEXT:    popl %edi
@@ -622,22 +619,22 @@
 ; X64-NEXT:    testb %bl, %al
 ; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r14 # 8-byte Folded Reload
 ; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r13 # 8-byte Folded Reload
-; X64-NEXT:    movl $4294967295, %edx # imm = 0xFFFFFFFF
-; X64-NEXT:    cmpq %rdx, %r13
+; X64-NEXT:    movl $4294967295, %ecx # imm = 0xFFFFFFFF
+; X64-NEXT:    cmpq %rcx, %r13
 ; X64-NEXT:    movl $4294967295, %eax # imm = 0xFFFFFFFF
 ; X64-NEXT:    cmovbq %r13, %rax
-; X64-NEXT:    xorl %ecx, %ecx
 ; X64-NEXT:    testq %r14, %r14
-; X64-NEXT:    cmovnsq %rdx, %r13
+; X64-NEXT:    cmovnsq %rcx, %r13
 ; X64-NEXT:    cmoveq %rax, %r13
-; X64-NEXT:    cmovnsq %rcx, %r14
-; X64-NEXT:    movabsq $-4294967296, %rcx # imm = 0xFFFFFFFF00000000
-; X64-NEXT:    cmpq %rcx, %r13
-; X64-NEXT:    movq %rcx, %rax
+; X64-NEXT:    movabsq $-4294967296, %rdx # imm = 0xFFFFFFFF00000000
+; X64-NEXT:    cmpq %rdx, %r13
+; X64-NEXT:    movq %rdx, %rax
 ; X64-NEXT:    cmovaq %r13, %rax
-; X64-NEXT:    testq %r14, %r14
-; X64-NEXT:    cmovsq %rcx, %r13
-; X64-NEXT:    cmpq $-1, %r14
+; X64-NEXT:    movq %r14, %rcx
+; X64-NEXT:    sarq $63, %rcx
+; X64-NEXT:    andq %r14, %rcx
+; X64-NEXT:    cmovsq %rdx, %r13
+; X64-NEXT:    cmpq $-1, %rcx
 ; X64-NEXT:    cmoveq %rax, %r13
 ; X64-NEXT:    movq %r13, %xmm0
 ; X64-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
@@ -684,15 +681,15 @@
 ; X64-NEXT:    testq %r14, %r14
 ; X64-NEXT:    cmovnsq %rcx, %r12
 ; X64-NEXT:    cmoveq %rax, %r12
-; X64-NEXT:    movl $0, %eax
-; X64-NEXT:    cmovnsq %rax, %r14
-; X64-NEXT:    movabsq $-4294967296, %rcx # imm = 0xFFFFFFFF00000000
-; X64-NEXT:    cmpq %rcx, %r12
-; X64-NEXT:    movq %rcx, %rax
+; X64-NEXT:    movabsq $-4294967296, %rdx # imm = 0xFFFFFFFF00000000
+; X64-NEXT:    cmpq %rdx, %r12
+; X64-NEXT:    movq %rdx, %rax
 ; X64-NEXT:    cmovaq %r12, %rax
-; X64-NEXT:    testq %r14, %r14
-; X64-NEXT:    cmovsq %rcx, %r12
-; X64-NEXT:    cmpq $-1, %r14
+; X64-NEXT:    movq %r14, %rcx
+; X64-NEXT:    sarq $63, %rcx
+; X64-NEXT:    andq %r14, %rcx
+; X64-NEXT:    cmovsq %rdx, %r12
+; X64-NEXT:    cmpq $-1, %rcx
 ; X64-NEXT:    cmoveq %rax, %r12
 ; X64-NEXT:    movq %r12, %xmm0
 ; X64-NEXT:    movdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
@@ -707,9 +704,9 @@
 ; X64-NEXT:    paddq %xmm1, %xmm1
 ; X64-NEXT:    movdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
 ; X64-NEXT:    movq %xmm1, %rbx
-; X64-NEXT:    movq %rbx, %r12
-; X64-NEXT:    sarq $63, %r12
-; X64-NEXT:    shldq $31, %rbx, %r12
+; X64-NEXT:    movq %rbx, %r13
+; X64-NEXT:    sarq $63, %r13
+; X64-NEXT:    shldq $31, %rbx, %r13
 ; X64-NEXT:    pshufd $238, {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Folded Reload
 ; X64-NEXT:    # xmm1 = mem[2,3,2,3]
 ; X64-NEXT:    pxor %xmm0, %xmm0
@@ -723,19 +720,19 @@
 ; X64-NEXT:    movq %rbx, %r15
 ; X64-NEXT:    shlq $31, %r15
 ; X64-NEXT:    movq %r15, %rdi
-; X64-NEXT:    movq %r12, %rsi
+; X64-NEXT:    movq %r13, %rsi
 ; X64-NEXT:    movq %rbp, %rcx
 ; X64-NEXT:    callq __divti3@PLT
-; X64-NEXT:    movq %rax, %r13
+; X64-NEXT:    movq %rax, %r12
 ; X64-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; X64-NEXT:    movq %rdx, %r14
 ; X64-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    subq $1, %r13
+; X64-NEXT:    subq $1, %r12
 ; X64-NEXT:    sbbq $0, %r14
 ; X64-NEXT:    shrq $63, %rbx
 ; X64-NEXT:    xorl %ebp, %ebx
 ; X64-NEXT:    movq %r15, %rdi
-; X64-NEXT:    movq %r12, %rsi
+; X64-NEXT:    movq %r13, %rsi
 ; X64-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rdx # 8-byte Reload
 ; X64-NEXT:    movq %rbp, %rcx
 ; X64-NEXT:    callq __modti3@PLT
@@ -743,25 +740,25 @@
 ; X64-NEXT:    setne %al
 ; X64-NEXT:    testb %bl, %al
 ; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r14 # 8-byte Folded Reload
-; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r13 # 8-byte Folded Reload
+; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r12 # 8-byte Folded Reload
 ; X64-NEXT:    movl $4294967295, %ecx # imm = 0xFFFFFFFF
-; X64-NEXT:    cmpq %rcx, %r13
+; X64-NEXT:    cmpq %rcx, %r12
 ; X64-NEXT:    movl $4294967295, %eax # imm = 0xFFFFFFFF
-; X64-NEXT:    cmovbq %r13, %rax
-; X64-NEXT:    testq %r14, %r14
-; X64-NEXT:    cmovnsq %rcx, %r13
-; X64-NEXT:    cmoveq %rax, %r13
-; X64-NEXT:    movl $0, %eax
-; X64-NEXT:    cmovnsq %rax, %r14
-; X64-NEXT:    movabsq $-4294967296, %rcx # imm = 0xFFFFFFFF00000000
-; X64-NEXT:    cmpq %rcx, %r13
-; X64-NEXT:    movq %rcx, %rax
-; X64-NEXT:    cmovaq %r13, %rax
+; X64-NEXT:    cmovbq %r12, %rax
 ; X64-NEXT:    testq %r14, %r14
-; X64-NEXT:    cmovsq %rcx, %r13
-; X64-NEXT:    cmpq $-1, %r14
-; X64-NEXT:    cmoveq %rax, %r13
-; X64-NEXT:    movq %r13, %xmm0
+; X64-NEXT:    cmovnsq %rcx, %r12
+; X64-NEXT:    cmoveq %rax, %r12
+; X64-NEXT:    movabsq $-4294967296, %rdx # imm = 0xFFFFFFFF00000000
+; X64-NEXT:    cmpq %rdx, %r12
+; X64-NEXT:    movq %rdx, %rax
+; X64-NEXT:    cmovaq %r12, %rax
+; X64-NEXT:    movq %r14, %rcx
+; X64-NEXT:    sarq $63, %rcx
+; X64-NEXT:    andq %r14, %rcx
+; X64-NEXT:    cmovsq %rdx, %r12
+; X64-NEXT:    cmpq $-1, %rcx
+; X64-NEXT:    cmoveq %rax, %r12
+; X64-NEXT:    movq %r12, %xmm0
 ; X64-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
 ; X64-NEXT:    pshufd $238, {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Folded Reload
 ; X64-NEXT:    # xmm0 = mem[2,3,2,3]
@@ -806,15 +803,16 @@
 ; X64-NEXT:    testq %r14, %r14
 ; X64-NEXT:    cmovnsq %rcx, %r12
 ; X64-NEXT:    cmoveq %rax, %r12
-; X64-NEXT:    movl $0, %eax
-; X64-NEXT:    cmovnsq %rax, %r14
 ; X64-NEXT:    movabsq $-4294967296, %rcx # imm = 0xFFFFFFFF00000000
 ; X64-NEXT:    cmpq %rcx, %r12
 ; X64-NEXT:    movq %rcx, %rax
+; X64-NEXT:    movq %rcx, %rdx
 ; X64-NEXT:    cmovaq %r12, %rax
-; X64-NEXT:    testq %r14, %r14
-; X64-NEXT:    cmovsq %rcx, %r12
-; X64-NEXT:    cmpq $-1, %r14
+; X64-NEXT:    movq %r14, %rcx
+; X64-NEXT:    sarq $63, %rcx
+; X64-NEXT:    andq %r14, %rcx
+; X64-NEXT:    cmovsq %rdx, %r12
+; X64-NEXT:    cmpq $-1, %rcx
 ; X64-NEXT:    cmoveq %rax, %r12
 ; X64-NEXT:    movq %r12, %xmm0
 ; X64-NEXT:    movdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
@@ -839,41 +837,94 @@
 ; X86-NEXT:    pushl %edi
 ; X86-NEXT:    pushl %esi
 ; X86-NEXT:    andl $-8, %esp
-; X86-NEXT:    subl $256, %esp # imm = 0x100
-; X86-NEXT:    movl 24(%ebp), %edx
-; X86-NEXT:    movl 40(%ebp), %edi
-; X86-NEXT:    leal {{[0-9]+}}(%esp), %ebx
-; X86-NEXT:    movl %edi, %eax
-; X86-NEXT:    sarl $31, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    subl $192, %esp
+; X86-NEXT:    movl 24(%ebp), %esi
+; X86-NEXT:    movl 40(%ebp), %ecx
+; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    movl %ecx, %edi
+; X86-NEXT:    movl %ecx, %edx
+; X86-NEXT:    sarl $31, %edi
+; X86-NEXT:    movl %esi, %ecx
+; X86-NEXT:    sarl $31, %ecx
+; X86-NEXT:    addl %esi, %esi
+; X86-NEXT:    adcl %ecx, %ecx
+; X86-NEXT:    andl $1, %ecx
+; X86-NEXT:    negl %ecx
+; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %ecx, %ebx
+; X86-NEXT:    shldl $31, %esi, %ebx
+; X86-NEXT:    shll $31, %esi
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %edx
+; X86-NEXT:    pushl %ecx
+; X86-NEXT:    pushl %ecx
+; X86-NEXT:    pushl %ebx
+; X86-NEXT:    pushl %esi
+; X86-NEXT:    pushl %eax
+; X86-NEXT:    calll __modti3
+; X86-NEXT:    addl $32, %esp
+; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl 40(%ebp)
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    pushl %ecx
+; X86-NEXT:    pushl %ecx
+; X86-NEXT:    pushl %ebx
+; X86-NEXT:    pushl %esi
+; X86-NEXT:    pushl %eax
+; X86-NEXT:    calll __divti3
+; X86-NEXT:    addl $32, %esp
+; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    movl 36(%ebp), %ebx
+; X86-NEXT:    movl %ebx, %edi
+; X86-NEXT:    sarl $31, %edi
+; X86-NEXT:    movl 20(%ebp), %edx
 ; X86-NEXT:    movl %edx, %ecx
 ; X86-NEXT:    sarl $31, %ecx
 ; X86-NEXT:    addl %edx, %edx
 ; X86-NEXT:    adcl %ecx, %ecx
 ; X86-NEXT:    andl $1, %ecx
 ; X86-NEXT:    negl %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl %ecx, %esi
 ; X86-NEXT:    shldl $31, %edx, %esi
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    shll $31, %edx
 ; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    pushl %eax
-; X86-NEXT:    pushl %eax
-; X86-NEXT:    pushl %eax
 ; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %ebx
 ; X86-NEXT:    pushl %ecx
 ; X86-NEXT:    pushl %ecx
+; X86-NEXT:    movl %ecx, %ebx
 ; X86-NEXT:    pushl %esi
 ; X86-NEXT:    pushl %edx
-; X86-NEXT:    pushl %ebx
+; X86-NEXT:    pushl %eax
 ; X86-NEXT:    calll __modti3
 ; X86-NEXT:    addl $32, %esp
 ; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl 36(%ebp), %edx
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl 36(%ebp)
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    pushl %ebx
+; X86-NEXT:    pushl %ebx
+; X86-NEXT:    pushl %esi
+; X86-NEXT:    pushl {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
+; X86-NEXT:    pushl %eax
+; X86-NEXT:    calll __divti3
+; X86-NEXT:    addl $32, %esp
+; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    movl 32(%ebp), %edx
 ; X86-NEXT:    movl %edx, %edi
 ; X86-NEXT:    sarl $31, %edi
-; X86-NEXT:    movl 20(%ebp), %ecx
+; X86-NEXT:    movl 16(%ebp), %ecx
 ; X86-NEXT:    movl %ecx, %esi
 ; X86-NEXT:    sarl $31, %esi
 ; X86-NEXT:    addl %ecx, %ecx
@@ -898,40 +949,11 @@
 ; X86-NEXT:    pushl %eax
 ; X86-NEXT:    calll __modti3
 ; X86-NEXT:    addl $32, %esp
-; X86-NEXT:    leal {{[0-9]+}}(%esp), %edi
-; X86-NEXT:    movl 28(%ebp), %ebx
-; X86-NEXT:    movl %ebx, %edx
-; X86-NEXT:    sarl $31, %edx
-; X86-NEXT:    movl 12(%ebp), %esi
-; X86-NEXT:    movl %esi, %ecx
-; X86-NEXT:    sarl $31, %ecx
-; X86-NEXT:    addl %esi, %esi
-; X86-NEXT:    adcl %ecx, %ecx
-; X86-NEXT:    andl $1, %ecx
-; X86-NEXT:    negl %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %ecx, %eax
-; X86-NEXT:    shldl $31, %esi, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    shll $31, %esi
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    pushl %edx
-; X86-NEXT:    pushl %edx
-; X86-NEXT:    pushl %edx
-; X86-NEXT:    pushl %ebx
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl %eax
-; X86-NEXT:    pushl %esi
-; X86-NEXT:    pushl %edi
-; X86-NEXT:    calll __divti3
-; X86-NEXT:    addl $32, %esp
 ; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl 32(%ebp), %edx
+; X86-NEXT:    movl 28(%ebp), %edx
 ; X86-NEXT:    movl %edx, %esi
 ; X86-NEXT:    sarl $31, %esi
-; X86-NEXT:    movl 16(%ebp), %ecx
+; X86-NEXT:    movl 12(%ebp), %ecx
 ; X86-NEXT:    movl %ecx, %ebx
 ; X86-NEXT:    sarl $31, %ebx
 ; X86-NEXT:    addl %ecx, %ecx
@@ -954,11 +976,10 @@
 ; X86-NEXT:    calll __modti3
 ; X86-NEXT:    addl $32, %esp
 ; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    pushl %esi
 ; X86-NEXT:    pushl %esi
 ; X86-NEXT:    pushl %esi
-; X86-NEXT:    pushl 32(%ebp)
+; X86-NEXT:    pushl 28(%ebp)
 ; X86-NEXT:    pushl %ebx
 ; X86-NEXT:    pushl %ebx
 ; X86-NEXT:    pushl %edi
@@ -966,358 +987,283 @@
 ; X86-NEXT:    pushl %eax
 ; X86-NEXT:    calll __divti3
 ; X86-NEXT:    addl $32, %esp
-; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl 40(%ebp)
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
-; X86-NEXT:    pushl {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
-; X86-NEXT:    pushl %eax
-; X86-NEXT:    calll __divti3
-; X86-NEXT:    addl $32, %esp
-; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl %ecx
-; X86-NEXT:    pushl 36(%ebp)
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Reload
-; X86-NEXT:    pushl %edi
-; X86-NEXT:    pushl %edi
-; X86-NEXT:    pushl {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
-; X86-NEXT:    pushl {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
-; X86-NEXT:    pushl %eax
-; X86-NEXT:    calll __divti3
-; X86-NEXT:    addl $32, %esp
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    subl $1, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
-; X86-NEXT:    movl %esi, %eax
-; X86-NEXT:    sbbl $0, %eax
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    sbbl $0, %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    sbbl $0, %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    testl %edi, %edi
-; X86-NEXT:    sets %cl
-; X86-NEXT:    cmpl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
-; X86-NEXT:    sets %ch
-; X86-NEXT:    xorb %cl, %ch
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
-; X86-NEXT:    orl {{[0-9]+}}(%esp), %edi
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %edx
-; X86-NEXT:    orl {{[0-9]+}}(%esp), %edx
-; X86-NEXT:    orl %edi, %edx
-; X86-NEXT:    setne %cl
-; X86-NEXT:    testb %ch, %cl
-; X86-NEXT:    cmovel %esi, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    subl $1, %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
-; X86-NEXT:    movl %edi, %ecx
 ; X86-NEXT:    sbbl $0, %ecx
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %edx
-; X86-NEXT:    movl %edx, %eax
-; X86-NEXT:    sbbl $0, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl $0, %edx
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    sbbl $0, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    testl %ebx, %ebx
 ; X86-NEXT:    sets %bl
-; X86-NEXT:    cmpl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
+; X86-NEXT:    testl %esi, %esi
 ; X86-NEXT:    sets %bh
 ; X86-NEXT:    xorb %bl, %bh
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
 ; X86-NEXT:    orl {{[0-9]+}}(%esp), %esi
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    orl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    orl %esi, %eax
-; X86-NEXT:    setne %al
-; X86-NEXT:    testb %bh, %al
-; X86-NEXT:    cmovel %edi, %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    cmovel %edx, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    subl $1, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
-; X86-NEXT:    movl %edi, %esi
-; X86-NEXT:    sbbl $0, %esi
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    sbbl $0, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    sbbl $0, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    testl %ecx, %ecx
-; X86-NEXT:    sets %al
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
-; X86-NEXT:    testl %edx, %edx
-; X86-NEXT:    sets %bl
-; X86-NEXT:    xorb %al, %bl
+; X86-NEXT:    orl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    orl %esi, %edi
+; X86-NEXT:    setne %bl
+; X86-NEXT:    testb %bh, %bl
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
+; X86-NEXT:    movl %eax, %ebx
+; X86-NEXT:    sarl $31, %ebx
+; X86-NEXT:    movl %ebx, %esi
+; X86-NEXT:    andl %ecx, %esi
+; X86-NEXT:    movl %ebx, %edi
+; X86-NEXT:    andl %edx, %edi
+; X86-NEXT:    testl %eax, %eax
+; X86-NEXT:    cmovel %eax, %edi
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl $-1, %edi
+; X86-NEXT:    cmovsl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Folded Reload
+; X86-NEXT:    negl %ecx
+; X86-NEXT:    movl $0, %ecx
+; X86-NEXT:    sbbl %ecx, %ecx
+; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
+; X86-NEXT:    orl %eax, %edx
+; X86-NEXT:    cmovnel %edi, %ecx
+; X86-NEXT:    movl $0, %edi
+; X86-NEXT:    cmovel %edi, %esi
+; X86-NEXT:    cmpl $-1, %esi
+; X86-NEXT:    movl $0, %edx
+; X86-NEXT:    cmovel %ecx, %edx
+; X86-NEXT:    andl %eax, %ebx
+; X86-NEXT:    cmovsl %edi, %ecx
+; X86-NEXT:    movl $-1, %eax
+; X86-NEXT:    cmovsl %eax, %esi
+; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
+; X86-NEXT:    cmpl $-1, %ebx
+; X86-NEXT:    cmovel %edx, %ecx
+; X86-NEXT:    cmovnel %esi, %ebx
+; X86-NEXT:    shldl $31, %ecx, %ebx
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    leal {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    pushl %edx
-; X86-NEXT:    pushl %edx
-; X86-NEXT:    pushl %edx
-; X86-NEXT:    pushl 28(%ebp)
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    pushl %ecx
 ; X86-NEXT:    pushl %ecx
 ; X86-NEXT:    pushl %ecx
+; X86-NEXT:    pushl 32(%ebp)
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
+; X86-NEXT:    pushl %esi
+; X86-NEXT:    pushl %esi
 ; X86-NEXT:    pushl {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
 ; X86-NEXT:    pushl {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
 ; X86-NEXT:    pushl %eax
-; X86-NEXT:    calll __modti3
+; X86-NEXT:    calll __divti3
 ; X86-NEXT:    addl $32, %esp
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    orl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
-; X86-NEXT:    orl {{[0-9]+}}(%esp), %ecx
-; X86-NEXT:    orl %eax, %ecx
-; X86-NEXT:    setne %al
-; X86-NEXT:    testb %bl, %al
-; X86-NEXT:    cmovel %edi, %esi
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
+; X86-NEXT:    subl $1, %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl $0, %edi
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %ebx
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl $0, %ebx
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    subl $1, %ecx
+; X86-NEXT:    sbbl $0, %ecx
+; X86-NEXT:    testl %esi, %esi
+; X86-NEXT:    sets %al
+; X86-NEXT:    cmpl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
+; X86-NEXT:    sets %dl
+; X86-NEXT:    xorb %al, %dl
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    orl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
+; X86-NEXT:    orl {{[0-9]+}}(%esp), %esi
+; X86-NEXT:    orl %eax, %esi
+; X86-NEXT:    setne %al
+; X86-NEXT:    testb %dl, %al
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
+; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Folded Reload
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
+; X86-NEXT:    movl %ecx, %esi
+; X86-NEXT:    sarl $31, %esi
 ; X86-NEXT:    movl %esi, %eax
-; X86-NEXT:    sbbl $0, %eax
+; X86-NEXT:    andl %edi, %eax
+; X86-NEXT:    movl %esi, %edx
+; X86-NEXT:    andl %ebx, %edx
+; X86-NEXT:    testl %ecx, %ecx
+; X86-NEXT:    cmovel %ecx, %edx
+; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl $-1, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    cmovsl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
+; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    negl %edi
+; X86-NEXT:    movl $0, %edi
+; X86-NEXT:    sbbl %edi, %edi
+; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Folded Reload
+; X86-NEXT:    orl %ecx, %ebx
+; X86-NEXT:    cmovnel {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Folded Reload
+; X86-NEXT:    movl $0, %ebx
+; X86-NEXT:    cmovel %ebx, %eax
+; X86-NEXT:    cmpl $-1, %eax
+; X86-NEXT:    movl $0, %ebx
+; X86-NEXT:    cmovel %edi, %ebx
+; X86-NEXT:    andl %ecx, %esi
+; X86-NEXT:    movl $0, %ecx
+; X86-NEXT:    cmovsl %ecx, %edi
+; X86-NEXT:    movl $-1, %ecx
+; X86-NEXT:    cmovsl %ecx, %eax
+; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
+; X86-NEXT:    cmpl $-1, %esi
+; X86-NEXT:    cmovel %ebx, %edi
+; X86-NEXT:    cmovnel %eax, %esi
+; X86-NEXT:    shldl $31, %edi, %esi
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    subl $1, %eax
+; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %ebx
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl $0, %ebx
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    sbbl $0, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %eax, %esi
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %edx
 ; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    sbbl $0, %edx
 ; X86-NEXT:    cmpl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
-; X86-NEXT:    sets %bl
+; X86-NEXT:    sets %cl
 ; X86-NEXT:    cmpl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
-; X86-NEXT:    sets %bh
-; X86-NEXT:    xorb %bl, %bh
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
-; X86-NEXT:    orl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    sets %ch
+; X86-NEXT:    xorb %cl, %ch
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    orl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    orl %edi, %eax
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    orl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    orl %eax, %edi
 ; X86-NEXT:    setne %al
-; X86-NEXT:    testb %bh, %al
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    cmovel %esi, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    testb %ch, %al
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
 ; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
 ; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %edx, %esi
+; X86-NEXT:    sarl $31, %esi
+; X86-NEXT:    movl %esi, %ecx
+; X86-NEXT:    andl %ebx, %ecx
+; X86-NEXT:    movl %esi, %edi
+; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Folded Reload
 ; X86-NEXT:    testl %edx, %edx
+; X86-NEXT:    cmovel %edx, %edi
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl $-1, %edi
+; X86-NEXT:    cmovsl %eax, %edi
+; X86-NEXT:    negl %ebx
 ; X86-NEXT:    movl $0, %eax
-; X86-NEXT:    cmovsl %edx, %eax
+; X86-NEXT:    sbbl %eax, %eax
+; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
+; X86-NEXT:    orl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
+; X86-NEXT:    cmovnel %edi, %eax
+; X86-NEXT:    movl $0, %edi
+; X86-NEXT:    cmovel %edi, %ecx
+; X86-NEXT:    cmpl $-1, %ecx
+; X86-NEXT:    movl $0, %ebx
+; X86-NEXT:    cmovel %eax, %ebx
+; X86-NEXT:    andl %edx, %esi
+; X86-NEXT:    cmovsl %edi, %eax
+; X86-NEXT:    movl $-1, %edx
+; X86-NEXT:    cmovsl %edx, %ecx
+; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
+; X86-NEXT:    cmpl $-1, %esi
+; X86-NEXT:    cmovel %ebx, %eax
+; X86-NEXT:    cmovnel %ecx, %esi
+; X86-NEXT:    shldl $31, %eax, %esi
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl $-1, %eax
-; X86-NEXT:    cmovsl %ecx, %eax
+; X86-NEXT:    subl $1, %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    movl %eax, %ecx
-; X86-NEXT:    sarl $31, %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %ecx, %edx
-; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
-; X86-NEXT:    testl %eax, %eax
-; X86-NEXT:    cmovel %eax, %edx
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl $0, %ecx
-; X86-NEXT:    cmovsl %eax, %ecx
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl $0, %esi
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %ebx
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl $0, %ebx
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl $-1, %eax
-; X86-NEXT:    cmovsl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl $0, %ecx
+; X86-NEXT:    cmpl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
+; X86-NEXT:    sets %al
+; X86-NEXT:    cmpl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
+; X86-NEXT:    sets %dl
+; X86-NEXT:    xorb %al, %dl
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    orl {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    orl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    orl %eax, %edi
+; X86-NEXT:    setne %al
+; X86-NEXT:    testb %dl, %al
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    movl %eax, %edi
-; X86-NEXT:    sarl $31, %edi
-; X86-NEXT:    movl %edi, %ecx
-; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    testl %eax, %eax
-; X86-NEXT:    cmovel %eax, %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl $0, %ecx
-; X86-NEXT:    cmovsl %eax, %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl $-1, %eax
-; X86-NEXT:    cmovsl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    movl %ecx, %ebx
-; X86-NEXT:    sarl $31, %ebx
-; X86-NEXT:    movl %ebx, %eax
-; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    testl %ecx, %ecx
-; X86-NEXT:    cmovel %ecx, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl $0, %eax
-; X86-NEXT:    cmovsl %ecx, %eax
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
 ; X86-NEXT:    movl %ecx, %edx
-; X86-NEXT:    movl $-1, %esi
-; X86-NEXT:    cmovsl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    movl %eax, %ecx
-; X86-NEXT:    sarl $31, %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    testl %eax, %eax
-; X86-NEXT:    cmovel %eax, %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    andl %eax, %ebx
-; X86-NEXT:    negl %eax
-; X86-NEXT:    movl $0, %ecx
-; X86-NEXT:    sbbl %ecx, %ecx
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
-; X86-NEXT:    cmovnel %esi, %ecx
-; X86-NEXT:    movl $0, %edx
-; X86-NEXT:    cmovel %edx, %ebx
-; X86-NEXT:    cmpl $-1, %ebx
-; X86-NEXT:    movl $0, %esi
-; X86-NEXT:    cmovel %ecx, %esi
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    testl %eax, %eax
-; X86-NEXT:    cmovsl %edx, %ecx
-; X86-NEXT:    movl $-1, %edx
-; X86-NEXT:    cmovsl %edx, %ebx
-; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    cmpl $-1, %eax
-; X86-NEXT:    cmovel %esi, %ecx
-; X86-NEXT:    cmovnel %ebx, %eax
-; X86-NEXT:    shldl $31, %ecx, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    andl %eax, %edi
-; X86-NEXT:    negl %eax
-; X86-NEXT:    movl $0, %eax
-; X86-NEXT:    sbbl %eax, %eax
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    cmovnel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl $0, %esi
-; X86-NEXT:    cmovel %esi, %edi
-; X86-NEXT:    cmpl $-1, %edi
-; X86-NEXT:    movl $0, %edx
-; X86-NEXT:    cmovel %eax, %edx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    sarl $31, %edx
+; X86-NEXT:    movl %edx, %eax
+; X86-NEXT:    andl %esi, %eax
+; X86-NEXT:    movl %edx, %edi
+; X86-NEXT:    andl %ebx, %edi
 ; X86-NEXT:    testl %ecx, %ecx
-; X86-NEXT:    cmovsl %esi, %eax
-; X86-NEXT:    movl $0, %esi
-; X86-NEXT:    movl $-1, %ebx
-; X86-NEXT:    cmovsl %ebx, %edi
-; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    cmpl $-1, %ecx
-; X86-NEXT:    cmovel %edx, %eax
-; X86-NEXT:    cmovnel %edi, %ecx
-; X86-NEXT:    shldl $31, %eax, %ecx
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
-; X86-NEXT:    andl %eax, %edx
-; X86-NEXT:    negl %eax
-; X86-NEXT:    movl $0, %eax
-; X86-NEXT:    sbbl %eax, %eax
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    cmovnel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    cmovel %esi, %edx
-; X86-NEXT:    cmpl $-1, %edx
-; X86-NEXT:    movl $0, %ecx
-; X86-NEXT:    cmovel %eax, %ecx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
-; X86-NEXT:    testl %ebx, %ebx
-; X86-NEXT:    cmovsl %esi, %eax
-; X86-NEXT:    movl $-1, %edi
-; X86-NEXT:    cmovsl %edi, %edx
-; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
-; X86-NEXT:    cmpl $-1, %ebx
-; X86-NEXT:    cmovel %ecx, %eax
-; X86-NEXT:    cmovnel %edx, %ebx
-; X86-NEXT:    shldl $31, %eax, %ebx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT:    cmovel %ecx, %edi
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl $-1, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Reload
-; X86-NEXT:    andl %eax, %edi
-; X86-NEXT:    negl %eax
-; X86-NEXT:    movl $0, %eax
-; X86-NEXT:    sbbl %eax, %eax
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    cmovnel {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    cmovel %esi, %edi
-; X86-NEXT:    cmpl $-1, %edi
+; X86-NEXT:    cmovsl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Folded Reload
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    negl %esi
+; X86-NEXT:    movl $0, %esi
+; X86-NEXT:    sbbl %esi, %esi
+; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
+; X86-NEXT:    orl %ecx, %ebx
+; X86-NEXT:    cmovnel {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
+; X86-NEXT:    movl $0, %ebx
+; X86-NEXT:    cmovel %ebx, %eax
+; X86-NEXT:    cmpl $-1, %eax
+; X86-NEXT:    movl $0, %ebx
+; X86-NEXT:    cmovel %esi, %ebx
+; X86-NEXT:    andl %ecx, %edx
 ; X86-NEXT:    movl $0, %ecx
-; X86-NEXT:    cmovel %eax, %ecx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
-; X86-NEXT:    testl %edx, %edx
-; X86-NEXT:    cmovsl %esi, %eax
-; X86-NEXT:    movl $-1, %esi
-; X86-NEXT:    cmovsl %esi, %edi
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
-; X86-NEXT:    andl %edx, %esi
-; X86-NEXT:    cmpl $-1, %esi
-; X86-NEXT:    cmovel %ecx, %eax
-; X86-NEXT:    cmovnel %edi, %esi
-; X86-NEXT:    shldl $31, %eax, %esi
+; X86-NEXT:    cmovsl %ecx, %esi
+; X86-NEXT:    movl $-1, %ecx
+; X86-NEXT:    cmovsl %ecx, %eax
+; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
+; X86-NEXT:    cmpl $-1, %edx
+; X86-NEXT:    cmovel %ebx, %esi
+; X86-NEXT:    cmovnel %eax, %edx
+; X86-NEXT:    shldl $31, %esi, %edx
 ; X86-NEXT:    movl 8(%ebp), %eax
-; X86-NEXT:    movl %esi, 12(%eax)
-; X86-NEXT:    movl %ebx, 8(%eax)
+; X86-NEXT:    movl %edx, 12(%eax)
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    movl %ecx, 8(%eax)
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
 ; X86-NEXT:    movl %ecx, 4(%eax)
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload