Index: llvm/trunk/test/CodeGen/MSP430/libcalls.ll
===================================================================
--- llvm/trunk/test/CodeGen/MSP430/libcalls.ll
+++ llvm/trunk/test/CodeGen/MSP430/libcalls.ll
@@ -433,9 +433,10 @@
 
 ; CHECK: call #__mspabi_divi
   %0 = load volatile i16, i16* @g_i16, align 8
-  %1 = sdiv i16 %0, %0
+  %1 = load volatile i16, i16* @g_i16, align 8
+  %2 = sdiv i16 %0, %1
 
-  ret i16 %1
+  ret i16 %2
 }
 
 define i32 @divli() #0 {
@@ -444,9 +445,10 @@
 
 ; CHECK: call #__mspabi_divli
   %0 = load volatile i32, i32* @g_i32, align 8
-  %1 = sdiv i32 %0, %0
+  %1 = load volatile i32, i32* @g_i32, align 8
+  %2 = sdiv i32 %0, %1
 
-  ret i32 %1
+  ret i32 %2
 }
 
 define i64 @divlli() #0 {
@@ -455,9 +457,10 @@
 
 ; CHECK: call #__mspabi_divlli
   %0 = load volatile i64, i64* @g_i64, align 8
-  %1 = sdiv i64 %0, %0
+  %1 = load volatile i64, i64* @g_i64, align 8
+  %2 = sdiv i64 %0, %1
 
-  ret i64 %1
+  ret i64 %2
 }
 
 define i16 @divu() #0 {
@@ -466,9 +469,10 @@
 
 ; CHECK: call #__mspabi_divu
   %0 = load volatile i16, i16* @g_i16, align 8
-  %1 = udiv i16 %0, %0
+  %1 = load volatile i16, i16* @g_i16, align 8
+  %2 = udiv i16 %0, %1
 
-  ret i16 %1
+  ret i16 %2
 }
 
 define i32 @divul() #0 {
@@ -477,9 +481,10 @@
 
 ; CHECK: call #__mspabi_divul
   %0 = load volatile i32, i32* @g_i32, align 8
-  %1 = udiv i32 %0, %0
+  %1 = load volatile i32, i32* @g_i32, align 8
+  %2 = udiv i32 %0, %1
 
-  ret i32 %1
+  ret i32 %2
 }
 
 define i64 @divull() #0 {
@@ -488,9 +493,10 @@
 
 ; CHECK: call #__mspabi_divull
   %0 = load volatile i64, i64* @g_i64, align 8
-  %1 = udiv i64 %0, %0
+  %1 = load volatile i64, i64* @g_i64, align 8
+  %2 = udiv i64 %0, %1
 
-  ret i64 %1
+  ret i64 %2
 }
 
 define i16 @remi() #0 {
@@ -499,9 +505,10 @@
 
 ; CHECK: call #__mspabi_remi
   %0 = load volatile i16, i16* @g_i16, align 8
-  %1 = srem i16 %0, %0
+  %1 = load volatile i16, i16* @g_i16, align 8
+  %2 = srem i16 %0, %1
 
-  ret i16 %1
+  ret i16 %2
 }
 
 define i32 @remli() #0 {
@@ -510,9 +517,10 @@
 
 ; CHECK: call #__mspabi_remli
   %0 = load volatile i32, i32* @g_i32, align 8
-  %1 = srem i32 %0, %0
+  %1 = load volatile i32, i32* @g_i32, align 8
+  %2 = srem i32 %0, %1
 
-  ret i32 %1
+  ret i32 %2
 }
 
 define i64 @remlli() #0 {
@@ -521,9 +529,10 @@
 
 ; CHECK: call #__mspabi_remlli
   %0 = load volatile i64, i64* @g_i64, align 8
-  %1 = srem i64 %0, %0
+  %1 = load volatile i64, i64* @g_i64, align 8
+  %2 = srem i64 %0, %1
 
-  ret i64 %1
+  ret i64 %2
 }
 
 define i16 @remu() #0 {
@@ -532,9 +541,10 @@
 
 ; CHECK: call #__mspabi_remu
   %0 = load volatile i16, i16* @g_i16, align 8
-  %1 = urem i16 %0, %0
+  %1 = load volatile i16, i16* @g_i16, align 8
+  %2 = urem i16 %0, %1
 
-  ret i16 %1
+  ret i16 %2
 }
 
 define i32 @remul() #0 {
@@ -543,9 +553,10 @@
 
 ; CHECK: call #__mspabi_remul
   %0 = load volatile i32, i32* @g_i32, align 8
-  %1 = urem i32 %0, %0
+  %1 = load volatile i32, i32* @g_i32, align 8
+  %2 = urem i32 %0, %1
 
-  ret i32 %1
+  ret i32 %2
 }
 
 define i64 @remull() #0 {
@@ -554,9 +565,10 @@
 
 ; CHECK: call #__mspabi_remull
   %0 = load volatile i64, i64* @g_i64, align 8
-  %1 = urem i64 %0, %0
+  %1 = load volatile i64, i64* @g_i64, align 8
+  %2 = urem i64 %0, %1
 
-  ret i64 %1
+  ret i64 %2
 }
 
 define i16 @mpyi() #0 {
Index: llvm/trunk/test/CodeGen/SystemZ/pr32372.ll
===================================================================
--- llvm/trunk/test/CodeGen/SystemZ/pr32372.ll
+++ llvm/trunk/test/CodeGen/SystemZ/pr32372.ll
@@ -4,10 +4,7 @@
 define void @pr32372(i8*) {
 ; CHECK-LABEL: pr32372:
 ; CHECK:       # %bb.0: # %BB
-; CHECK-NEXT:    llc %r1, 0(%r2)
 ; CHECK-NEXT:    mvhhi 0(%r1), -3825
-; CHECK-NEXT:    llill %r0, 0
-; CHECK-NEXT:    dlr %r0, %r1
 ; CHECK-NEXT:  .LBB0_1: # %CF251
 ; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    j .LBB0_1
@@ -15,7 +12,8 @@
   %L = load i8, i8* %0
   store i16 -3825, i16* undef
   %L5 = load i8, i8* %0
-  %B9 = urem i8 %L5, %L
+  %B8 = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 %L5, i8 %L)
+  %B9 = extractvalue {i8, i1} %B8, 0
   %I107 = insertelement <8 x i8> zeroinitializer, i8 %B9, i32 7
   %ZE141 = zext i8 %L5 to i16
   br label %CF251
@@ -29,3 +27,5 @@
   %Shuff230 = shufflevector <2 x i16> undef, <2 x i16> undef, <2 x i32> <i32 3, i32 1>
   br label %CF251
 }
+
+declare {i8, i1} @llvm.umul.with.overflow.i8(i8, i8) nounwind readnone
Index: llvm/trunk/test/CodeGen/X86/pr38539.ll
===================================================================
--- llvm/trunk/test/CodeGen/X86/pr38539.ll
+++ llvm/trunk/test/CodeGen/X86/pr38539.ll
@@ -187,7 +187,7 @@
 ; X64-NEXT:    movb (%rax), %al
 ; X64-NEXT:    movzbl %al, %eax
 ; X64-NEXT:    # kill: def $eax killed $eax def $ax
-; X64-NEXT:    divb %al
+; X64-NEXT:    divb (%rax)
 ; X64-NEXT:    movl %eax, %r8d
 ; X64-NEXT:    xorl %eax, %eax
 ; X64-NEXT:    xorl %edx, %edx
@@ -239,47 +239,47 @@
 ; X86-NEXT:    .cfi_offset %edi, -16
 ; X86-NEXT:    .cfi_offset %ebx, -12
 ; X86-NEXT:    movzbl {{[0-9]+}}(%esp), %esi
-; X86-NEXT:    movl %esi, %eax
-; X86-NEXT:    shll $30, %eax
-; X86-NEXT:    sarl $30, %eax
+; X86-NEXT:    movl %esi, %ecx
+; X86-NEXT:    shll $30, %ecx
+; X86-NEXT:    sarl $30, %ecx
 ; X86-NEXT:    movl (%esp), %edi
-; X86-NEXT:    movb (%eax), %bl
-; X86-NEXT:    pushl %eax
+; X86-NEXT:    movb (%eax), %al
+; X86-NEXT:    movzbl %al, %eax
+; X86-NEXT:    # kill: def $eax killed $eax def $ax
+; X86-NEXT:    divb (%eax)
+; X86-NEXT:    movl %eax, %ebx
+; X86-NEXT:    pushl %ecx
 ; X86-NEXT:    pushl %edi
 ; X86-NEXT:    pushl $0
 ; X86-NEXT:    pushl $0
 ; X86-NEXT:    calll __moddi3
 ; X86-NEXT:    addl $16, %esp
 ; X86-NEXT:    andl $3, %edx
+; X86-NEXT:    testb %al, %al
+; X86-NEXT:    setne (%eax)
 ; X86-NEXT:    cmpl %eax, %edi
 ; X86-NEXT:    sbbl %edx, %esi
-; X86-NEXT:    setb %dl
-; X86-NEXT:    setae %dh
+; X86-NEXT:    setae %dl
+; X86-NEXT:    sbbb %cl, %cl
 ; X86-NEXT:    testb %al, %al
-; X86-NEXT:    setne %bh
-; X86-NEXT:    setne (%eax)
-; X86-NEXT:    movzbl %bl, %eax
-; X86-NEXT:    xorl %ecx, %ecx
-; X86-NEXT:    subb %dl, %cl
-; X86-NEXT:    # kill: def $eax killed $eax def $ax
-; X86-NEXT:    divb %bl
-; X86-NEXT:    negb %dh
-; X86-NEXT:    cmpb %al, %al
+; X86-NEXT:    setne %ch
+; X86-NEXT:    negb %dl
+; X86-NEXT:    cmpb %bl, %al
 ; X86-NEXT:    setle %al
 ; X86-NEXT:    negb %al
 ; X86-NEXT:    cbtw
-; X86-NEXT:    idivb %dh
+; X86-NEXT:    idivb %dl
 ; X86-NEXT:    movsbl %ah, %eax
 ; X86-NEXT:    movzbl %al, %eax
 ; X86-NEXT:    andl $1, %eax
 ; X86-NEXT:    shll $3, %eax
 ; X86-NEXT:    negl %eax
-; X86-NEXT:    negb %bh
+; X86-NEXT:    negb %ch
 ; X86-NEXT:    leal -8(%esp,%eax), %eax
 ; X86-NEXT:    movl %eax, (%eax)
 ; X86-NEXT:    movl %ecx, %eax
 ; X86-NEXT:    cbtw
-; X86-NEXT:    idivb %bh
+; X86-NEXT:    idivb %ch
 ; X86-NEXT:    movsbl %ah, %eax
 ; X86-NEXT:    andb $1, %al
 ; X86-NEXT:    movb %al, (%eax)
@@ -295,8 +295,9 @@
   %L17 = load i34, i34* %A30
   %B20 = and i34 %L17, -1
   %G2 = getelementptr i34, i34* %A30, i1 true
-  %L10 = load i8, i8* undef
-  %B6 = udiv i8 %L10, %L10
+  %L10 = load volatile i8, i8* undef
+  %L11 = load volatile i8, i8* undef
+  %B6 = udiv i8 %L10, %L11
   %C15 = icmp eq i8 undef, 0
   %B8 = srem i34 0, %B20
   %C2 = icmp ule i34 %B8, %B20