diff --git a/llvm/test/CodeGen/X86/bt.ll b/llvm/test/CodeGen/X86/bt.ll
@@ -611,23 +611,21 @@ define void @query3(i32 %x, i32 %n) nounwind {
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    btl %ecx, %eax
-; X86-NEXT:    jb .LBB16_1
-; X86-NEXT:  # %bb.2: # %UnifiedReturnBlock
-; X86-NEXT:    retl
-; X86-NEXT:  .LBB16_1: # %bb
+; X86-NEXT:    jae .LBB16_2
+; X86-NEXT:  # %bb.1: # %bb
 ; X86-NEXT:    calll foo
+; X86-NEXT:  .LBB16_2: # %UnifiedReturnBlock
 ; X86-NEXT:    retl
 ;
 ; X64-LABEL: query3:
 ; X64:       # %bb.0: # %entry
 ; X64-NEXT:    btl %esi, %edi
-; X64-NEXT:    jb .LBB16_1
-; X64-NEXT:  # %bb.2: # %UnifiedReturnBlock
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB16_1: # %bb
+; X64-NEXT:    jae .LBB16_2
+; X64-NEXT:  # %bb.1: # %bb
 ; X64-NEXT:    pushq %rax
 ; X64-NEXT:    callq foo
 ; X64-NEXT:    popq %rax
+; X64-NEXT:  .LBB16_2: # %UnifiedReturnBlock
 ; X64-NEXT:    retq
 entry:
   %tmp29 = shl i32 1, %n
@@ -649,23 +647,21 @@ define void @query3b(i32 %x, i32 %n) nounwind {
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    btl %ecx, %eax
-; X86-NEXT:    jb .LBB17_1
-; X86-NEXT:  # %bb.2: # %UnifiedReturnBlock
-; X86-NEXT:    retl
-; X86-NEXT:  .LBB17_1: # %bb
+; X86-NEXT:    jae .LBB17_2
+; X86-NEXT:  # %bb.1: # %bb
 ; X86-NEXT:    calll foo
+; X86-NEXT:  .LBB17_2: # %UnifiedReturnBlock
 ; X86-NEXT:    retl
 ;
 ; X64-LABEL: query3b:
 ; X64:       # %bb.0: # %entry
 ; X64-NEXT:    btl %esi, %edi
-; X64-NEXT:    jb .LBB17_1
-; X64-NEXT:  # %bb.2: # %UnifiedReturnBlock
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB17_1: # %bb
+; X64-NEXT:    jae .LBB17_2
+; X64-NEXT:  # %bb.1: # %bb
 ; X64-NEXT:    pushq %rax
 ; X64-NEXT:    callq foo
 ; X64-NEXT:    popq %rax
+; X64-NEXT:  .LBB17_2: # %UnifiedReturnBlock
 ; X64-NEXT:    retq
 entry:
   %tmp29 = shl i32 1, %n
@@ -687,23 +683,21 @@ define void @query3x(i32 %x, i32 %n) nounwind {
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    btl %ecx, %eax
-; X86-NEXT:    jb .LBB18_1
-; X86-NEXT:  # %bb.2: # %UnifiedReturnBlock
-; X86-NEXT:    retl
-; X86-NEXT:  .LBB18_1: # %bb
+; X86-NEXT:    jae .LBB18_2
+; X86-NEXT:  # %bb.1: # %bb
 ; X86-NEXT:    calll foo
+; X86-NEXT:  .LBB18_2: # %UnifiedReturnBlock
 ; X86-NEXT:    retl
 ;
 ; X64-LABEL: query3x:
 ; X64:       # %bb.0: # %entry
 ; X64-NEXT:    btl %esi, %edi
-; X64-NEXT:    jb .LBB18_1
-; X64-NEXT:  # %bb.2: # %UnifiedReturnBlock
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB18_1: # %bb
+; X64-NEXT:    jae .LBB18_2
+; X64-NEXT:  # %bb.1: # %bb
 ; X64-NEXT:    pushq %rax
 ; X64-NEXT:    callq foo
 ; X64-NEXT:    popq %rax
+; X64-NEXT:  .LBB18_2: # %UnifiedReturnBlock
 ; X64-NEXT:    retq
 entry:
   %tmp29 = shl i32 1, %n
@@ -725,23 +719,21 @@ define void @query3bx(i32 %x, i32 %n) nounwind {
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    btl %ecx, %eax
-; X86-NEXT:    jb .LBB19_1
-; X86-NEXT:  # %bb.2: # %UnifiedReturnBlock
-; X86-NEXT:    retl
-; X86-NEXT:  .LBB19_1: # %bb
+; X86-NEXT:    jae .LBB19_2
+; X86-NEXT:  # %bb.1: # %bb
 ; X86-NEXT:    calll foo
+; X86-NEXT:  .LBB19_2: # %UnifiedReturnBlock
 ; X86-NEXT:    retl
 ;
 ; X64-LABEL: query3bx:
 ; X64:       # %bb.0: # %entry
 ; X64-NEXT:    btl %esi, %edi
-; X64-NEXT:    jb .LBB19_1
-; X64-NEXT:  # %bb.2: # %UnifiedReturnBlock
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB19_1: # %bb
+; X64-NEXT:    jae .LBB19_2
+; X64-NEXT:  # %bb.1: # %bb
 ; X64-NEXT:    pushq %rax
 ; X64-NEXT:    callq foo
 ; X64-NEXT:    popq %rax
+; X64-NEXT:  .LBB19_2: # %UnifiedReturnBlock
 ; X64-NEXT:    retq
 entry:
   %tmp29 = shl i32 1, %n

diff --git a/llvm/test/CodeGen/X86/conditional-tailcall-pgso.ll b/llvm/test/CodeGen/X86/conditional-tailcall-pgso.ll
@@ -63,21 +63,21 @@ define void @f_non_leaf(i32 %x, i32 %y) !prof !14 {
 ; CHECK32-NEXT:    #APP
 ; CHECK32-NEXT:    #NO_APP
 ; CHECK32-NEXT:    cmpl {{[0-9]+}}(%esp), %eax # encoding: [0x3b,0x44,0x24,0x0c]
-; CHECK32-NEXT:    je .LBB1_1 # encoding: [0x74,A]
-; CHECK32-NEXT:    # fixup A - offset: 1, value: .LBB1_1-1, kind: FK_PCRel_1
-; CHECK32-NEXT:  # %bb.2: # %bb2
+; CHECK32-NEXT:    jne .LBB1_2 # encoding: [0x75,A]
+; CHECK32-NEXT:    # fixup A - offset: 1, value: .LBB1_2-1, kind: FK_PCRel_1
+; CHECK32-NEXT:  # %bb.1: # %bb1
 ; CHECK32-NEXT:    popl %ebx # encoding: [0x5b]
 ; CHECK32-NEXT:    .cfi_def_cfa_offset 4
-; CHECK32-NEXT:    jmp bar # TAILCALL
+; CHECK32-NEXT:    jmp foo # TAILCALL
 ; CHECK32-NEXT:    # encoding: [0xeb,A]
-; CHECK32-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
-; CHECK32-NEXT:  .LBB1_1: # %bb1
+; CHECK32-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; CHECK32-NEXT:  .LBB1_2: # %bb2
 ; CHECK32-NEXT:    .cfi_def_cfa_offset 8
 ; CHECK32-NEXT:    popl %ebx # encoding: [0x5b]
 ; CHECK32-NEXT:    .cfi_def_cfa_offset 4
-; CHECK32-NEXT:    jmp foo # TAILCALL
+; CHECK32-NEXT:    jmp bar # TAILCALL
 ; CHECK32-NEXT:    # encoding: [0xeb,A]
-; CHECK32-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; CHECK32-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
 ;
 ; CHECK64-LABEL: f_non_leaf:
 ; CHECK64:       # %bb.0: # %entry
@@ -87,21 +87,21 @@ define void @f_non_leaf(i32 %x, i32 %y) !prof !14 {
 ; CHECK64-NEXT:    #APP
 ; CHECK64-NEXT:    #NO_APP
 ; CHECK64-NEXT:    cmpl %esi, %edi # encoding: [0x39,0xf7]
-; CHECK64-NEXT:    je .LBB1_1 # encoding: [0x74,A]
-; CHECK64-NEXT:    # fixup A - offset: 1, value: .LBB1_1-1, kind: FK_PCRel_1
-; CHECK64-NEXT:  # %bb.2: # %bb2
+; CHECK64-NEXT:    jne .LBB1_2 # encoding: [0x75,A]
+; CHECK64-NEXT:    # fixup A - offset: 1, value: .LBB1_2-1, kind: FK_PCRel_1
+; CHECK64-NEXT:  # %bb.1: # %bb1
 ; CHECK64-NEXT:    popq %rbx # encoding: [0x5b]
 ; CHECK64-NEXT:    .cfi_def_cfa_offset 8
-; CHECK64-NEXT:    jmp bar # TAILCALL
+; CHECK64-NEXT:    jmp foo # TAILCALL
 ; CHECK64-NEXT:    # encoding: [0xeb,A]
-; CHECK64-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
-; CHECK64-NEXT:  .LBB1_1: # %bb1
+; CHECK64-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; CHECK64-NEXT:  .LBB1_2: # %bb2
 ; CHECK64-NEXT:    .cfi_def_cfa_offset 16
 ; CHECK64-NEXT:    popq %rbx # encoding: [0x5b]
 ; CHECK64-NEXT:    .cfi_def_cfa_offset 8
-; CHECK64-NEXT:    jmp foo # TAILCALL
+; CHECK64-NEXT:    jmp bar # TAILCALL
 ; CHECK64-NEXT:    # encoding: [0xeb,A]
-; CHECK64-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; CHECK64-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
 ;
 ; WIN64-LABEL: f_non_leaf:
 ; WIN64:       # %bb.0: # %entry
@@ -111,19 +111,19 @@ define void @f_non_leaf(i32 %x, i32 %y) !prof !14 {
 ; WIN64-NEXT:    #APP
 ; WIN64-NEXT:    #NO_APP
 ; WIN64-NEXT:    cmpl %edx, %ecx # encoding: [0x39,0xd1]
-; WIN64-NEXT:    je .LBB1_1 # encoding: [0x74,A]
-; WIN64-NEXT:    # fixup A - offset: 1, value: .LBB1_1-1, kind: FK_PCRel_1
-; WIN64-NEXT:  # %bb.2: # %bb2
+; WIN64-NEXT:    jne .LBB1_2 # encoding: [0x75,A]
+; WIN64-NEXT:    # fixup A - offset: 1, value: .LBB1_2-1, kind: FK_PCRel_1
+; WIN64-NEXT:  # %bb.1: # %bb1
 ; WIN64-NEXT:    popq %rbx # encoding: [0x5b]
-; WIN64-NEXT:    jmp bar # TAILCALL
+; WIN64-NEXT:    jmp foo # TAILCALL
 ; WIN64-NEXT:    # encoding: [0xeb,A]
-; WIN64-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
-; WIN64-NEXT:  .LBB1_1: # %bb1
+; WIN64-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; WIN64-NEXT:  .LBB1_2: # %bb2
 ; WIN64-NEXT:    nop # encoding: [0x90]
 ; WIN64-NEXT:    popq %rbx # encoding: [0x5b]
-; WIN64-NEXT:    jmp foo # TAILCALL
+; WIN64-NEXT:    jmp bar # TAILCALL
 ; WIN64-NEXT:    # encoding: [0xeb,A]
-; WIN64-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; WIN64-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
 ; WIN64-NEXT:    .seh_handlerdata
 ; WIN64-NEXT:    .text
 ; WIN64-NEXT:    .seh_endproc

diff --git a/llvm/test/CodeGen/X86/conditional-tailcall.ll b/llvm/test/CodeGen/X86/conditional-tailcall.ll
@@ -63,21 +63,21 @@ define void @f_non_leaf(i32 %x, i32 %y) optsize {
 ; CHECK32-NEXT:    #APP
 ; CHECK32-NEXT:    #NO_APP
 ; CHECK32-NEXT:    cmpl {{[0-9]+}}(%esp), %eax # encoding: [0x3b,0x44,0x24,0x0c]
-; CHECK32-NEXT:    je .LBB1_1 # encoding: [0x74,A]
-; CHECK32-NEXT:    # fixup A - offset: 1, value: .LBB1_1-1, kind: FK_PCRel_1
-; CHECK32-NEXT:  # %bb.2: # %bb2
+; CHECK32-NEXT:    jne .LBB1_2 # encoding: [0x75,A]
+; CHECK32-NEXT:    # fixup A - offset: 1, value: .LBB1_2-1, kind: FK_PCRel_1
+; CHECK32-NEXT:  # %bb.1: # %bb1
 ; CHECK32-NEXT:    popl %ebx # encoding: [0x5b]
 ; CHECK32-NEXT:    .cfi_def_cfa_offset 4
-; CHECK32-NEXT:    jmp bar # TAILCALL
+; CHECK32-NEXT:    jmp foo # TAILCALL
 ; CHECK32-NEXT:    # encoding: [0xeb,A]
-; CHECK32-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
-; CHECK32-NEXT:  .LBB1_1: # %bb1
+; CHECK32-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; CHECK32-NEXT:  .LBB1_2: # %bb2
 ; CHECK32-NEXT:    .cfi_def_cfa_offset 8
 ; CHECK32-NEXT:    popl %ebx # encoding: [0x5b]
 ; CHECK32-NEXT:    .cfi_def_cfa_offset 4
-; CHECK32-NEXT:    jmp foo # TAILCALL
+; CHECK32-NEXT:    jmp bar # TAILCALL
 ; CHECK32-NEXT:    # encoding: [0xeb,A]
-; CHECK32-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; CHECK32-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
 ;
 ; CHECK64-LABEL: f_non_leaf:
 ; CHECK64:       # %bb.0: # %entry
@@ -87,21 +87,21 @@ define void @f_non_leaf(i32 %x, i32 %y) optsize {
 ; CHECK64-NEXT:    #APP
 ; CHECK64-NEXT:    #NO_APP
 ; CHECK64-NEXT:    cmpl %esi, %edi # encoding: [0x39,0xf7]
-; CHECK64-NEXT:    je .LBB1_1 # encoding: [0x74,A]
-; CHECK64-NEXT:    # fixup A - offset: 1, value: .LBB1_1-1, kind: FK_PCRel_1
-; CHECK64-NEXT:  # %bb.2: # %bb2
+; CHECK64-NEXT:    jne .LBB1_2 # encoding: [0x75,A]
+; CHECK64-NEXT:    # fixup A - offset: 1, value: .LBB1_2-1, kind: FK_PCRel_1
+; CHECK64-NEXT:  # %bb.1: # %bb1
 ; CHECK64-NEXT:    popq %rbx # encoding: [0x5b]
 ; CHECK64-NEXT:    .cfi_def_cfa_offset 8
-; CHECK64-NEXT:    jmp bar # TAILCALL
+; CHECK64-NEXT:    jmp foo # TAILCALL
 ; CHECK64-NEXT:    # encoding: [0xeb,A]
-; CHECK64-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
-; CHECK64-NEXT:  .LBB1_1: # %bb1
+; CHECK64-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; CHECK64-NEXT:  .LBB1_2: # %bb2
 ; CHECK64-NEXT:    .cfi_def_cfa_offset 16
 ; CHECK64-NEXT:    popq %rbx # encoding: [0x5b]
 ; CHECK64-NEXT:    .cfi_def_cfa_offset 8
-; CHECK64-NEXT:    jmp foo # TAILCALL
+; CHECK64-NEXT:    jmp bar # TAILCALL
 ; CHECK64-NEXT:    # encoding: [0xeb,A]
-; CHECK64-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; CHECK64-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
 ;
 ; WIN64-LABEL: f_non_leaf:
 ; WIN64:       # %bb.0: # %entry
@@ -111,19 +111,19 @@ define void @f_non_leaf(i32 %x, i32 %y) optsize {
 ; WIN64-NEXT:    #APP
 ; WIN64-NEXT:    #NO_APP
 ; WIN64-NEXT:    cmpl %edx, %ecx # encoding: [0x39,0xd1]
-; WIN64-NEXT:    je .LBB1_1 # encoding: [0x74,A]
-; WIN64-NEXT:    # fixup A - offset: 1, value: .LBB1_1-1, kind: FK_PCRel_1
-; WIN64-NEXT:  # %bb.2: # %bb2
+; WIN64-NEXT:    jne .LBB1_2 # encoding: [0x75,A]
+; WIN64-NEXT:    # fixup A - offset: 1, value: .LBB1_2-1, kind: FK_PCRel_1
+; WIN64-NEXT:  # %bb.1: # %bb1
 ; WIN64-NEXT:    popq %rbx # encoding: [0x5b]
-; WIN64-NEXT:    jmp bar # TAILCALL
+; WIN64-NEXT:    jmp foo # TAILCALL
 ; WIN64-NEXT:    # encoding: [0xeb,A]
-; WIN64-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
-; WIN64-NEXT:  .LBB1_1: # %bb1
+; WIN64-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; WIN64-NEXT:  .LBB1_2: # %bb2
 ; WIN64-NEXT:    nop # encoding: [0x90]
 ; WIN64-NEXT:    popq %rbx # encoding: [0x5b]
-; WIN64-NEXT:    jmp foo # TAILCALL
+; WIN64-NEXT:    jmp bar # TAILCALL
 ; WIN64-NEXT:    # encoding: [0xeb,A]
-; WIN64-NEXT:    # fixup A - offset: 1, value: foo-1, kind: FK_PCRel_1
+; WIN64-NEXT:    # fixup A - offset: 1, value: bar-1, kind: FK_PCRel_1
 ; WIN64-NEXT:    .seh_handlerdata
 ; WIN64-NEXT:    .text
 ; WIN64-NEXT:    .seh_endproc

diff --git a/llvm/test/CodeGen/X86/fast-isel-cmp-branch2.ll b/llvm/test/CodeGen/X86/fast-isel-cmp-branch2.ll
diff --git a/llvm/test/CodeGen/X86/funnel-shift.ll b/llvm/test/CodeGen/X86/funnel-shift.ll
@@ -941,17 +941,17 @@ define void @PR45265(i32 %0, %struct.S* nocapture readonly %1) nounwind {
 ; X32-SSE2-NEXT:    shldl $24, %ebx, %edi
 ; X32-SSE2-NEXT:    xorl %eax, %edi
 ; X32-SSE2-NEXT:    orl %edi, %ecx
-; X32-SSE2-NEXT:    je .LBB44_2
-; X32-SSE2-NEXT:  # %bb.1:
+; X32-SSE2-NEXT:    jne .LBB44_1
+; X32-SSE2-NEXT:  # %bb.2:
 ; X32-SSE2-NEXT:    popl %esi
 ; X32-SSE2-NEXT:    popl %edi
 ; X32-SSE2-NEXT:    popl %ebx
-; X32-SSE2-NEXT:    retl
-; X32-SSE2-NEXT:  .LBB44_2:
+; X32-SSE2-NEXT:    jmp _Z3foov # TAILCALL
+; X32-SSE2-NEXT:  .LBB44_1:
 ; X32-SSE2-NEXT:    popl %esi
 ; X32-SSE2-NEXT:    popl %edi
 ; X32-SSE2-NEXT:    popl %ebx
-; X32-SSE2-NEXT:    jmp _Z3foov # TAILCALL
+; X32-SSE2-NEXT:    retl
 ;
 ; X64-AVX2-LABEL: PR45265:
 ; X64-AVX2:       # %bb.0:
@@ -964,11 +964,11 @@ define void @PR45265(i32 %0, %struct.S* nocapture readonly %1) nounwind {
 ; X64-AVX2-NEXT:    movq (%rsi,%rcx,4), %rcx
 ; X64-AVX2-NEXT:    shrdq $40, %rdi, %rcx
 ; X64-AVX2-NEXT:    cmpq %rax, %rcx
-; X64-AVX2-NEXT:    je .LBB44_2
-; X64-AVX2-NEXT:  # %bb.1:
-; X64-AVX2-NEXT:    retq
-; X64-AVX2-NEXT:  .LBB44_2:
+; X64-AVX2-NEXT:    jne .LBB44_1
+; X64-AVX2-NEXT:  # %bb.2:
 ; X64-AVX2-NEXT:    jmp _Z3foov # TAILCALL
+; X64-AVX2-NEXT:  .LBB44_1:
+; X64-AVX2-NEXT:    retq
   %3 = sext i32 %0 to i64
   %4 = getelementptr inbounds %struct.S, %struct.S* %1, i64 %3
   %5 = bitcast %struct.S* %4 to i88*

diff --git a/llvm/test/CodeGen/X86/indirect-branch-tracking-eh2.ll b/llvm/test/CodeGen/X86/indirect-branch-tracking-eh2.ll
@@ -1,183 +1,59 @@
-; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -mtriple x86_64-unknown-unknown -exception-model sjlj -verify-machineinstrs=0 < %s | FileCheck %s --check-prefix=NUM
 ; RUN: llc -mtriple x86_64-unknown-unknown -exception-model sjlj -verify-machineinstrs=0 < %s | FileCheck %s --check-prefix=SJLJ
 
+; NUM-COUNT-3: endbr64
+
+;SJLJ:       main:                                  # @main
+;SJLJ-NEXT: .Lfunc_begin0:
+;SJLJ-NEXT: # %bb.0:                                # %entry
+;SJLJ-NEXT:         endbr64
+;SJLJ-NEXT:         pushq   %rbp
+;SJLJ:               callq   _Unwind_SjLj_Register
+;SJLJ-NEXT: .Ltmp0:
+;SJLJ-NEXT:         callq   _Z3foov
+;SJLJ-NEXT: .Ltmp1:
+;SJLJ-NEXT: # %bb.1:                                # %invoke.cont
+;SJLJ-NEXT:         movl
+;SJLJ-NEXT: .LBB0_7:                                # %return
+;SJLJ:               callq   _Unwind_SjLj_Unregister
+;SJLJ:               retq
+;SJLJ-NEXT: .LBB0_9:
+;SJLJ-NEXT:         endbr64
+;SJLJ-NEXT:         movl
+;SJLJ-NEXT:         cmpl
+;SJLJ-NEXT:         jb      .LBB0_10
+;SJLJ-NEXT: # %bb.11:
+;SJLJ-NEXT:         ud2
+;SJLJ-NEXT: .LBB0_10:
+;SJLJ-NEXT:         leaq    .LJTI0_0(%rip), %rcx
+;SJLJ-NEXT:         jmpq    *(%rcx,%rax,8)
+;SJLJ-NEXT: .LBB0_2:                                # %lpad
+;SJLJ-NEXT: .Ltmp2:
+;SJLJ-NEXT:         endbr64
+;SJLJ:               jne     .LBB0_4
+;SJLJ-NEXT: # %bb.3:                                # %catch3
+;SJLJ:               callq   __cxa_begin_catch
+;SJLJ:               jmp     .LBB0_6
+;SJLJ-NEXT: .LBB0_4:                                # %catch.fallthrough
+;SJLJ-NEXT:         cmpl
+;SJLJ-NEXT:         jne     .LBB0_8
+;SJLJ-NEXT: # %bb.5:                                # %catch
+;SJLJ:               callq   __cxa_begin_catch
+;SJLJ:               cmpb
+;SJLJ-NEXT: .LBB0_6:                                # %return
+;SJLJ:               callq   __cxa_end_catch
+;SJLJ-NEXT:         jmp     .LBB0_7
+;SJLJ-NEXT: .LBB0_8:                                # %eh.resume
+;SJLJ-NEXT:         movl
+;SJLJ-NEXT: .Lfunc_end0:
+;SJLJ:      .LJTI0_0:
+;SJLJ-NEXT:         .quad   .LBB0_2
+
 @_ZTIi = external dso_local constant i8*
 @_ZTIc = external dso_local constant i8*
 
 ; Function Attrs: noinline norecurse optnone uwtable
 define dso_local i32 @main() #0 personality i8* bitcast (i32 (...)* @__gxx_personality_sj0 to i8*) {
-; NUM-LABEL: main:
-; NUM:       # %bb.0: # %entry
-; NUM-NEXT:    endbr64
-; NUM-NEXT:    pushq %rbp
-; NUM-NEXT:    movq %rsp, %rbp
-; NUM-NEXT:    pushq %r15
-; NUM-NEXT:    pushq %r14
-; NUM-NEXT:    pushq %r13
-; NUM-NEXT:    pushq %r12
-; NUM-NEXT:    pushq %rbx
-; NUM-NEXT:    subq $120, %rsp
-; NUM-NEXT:    movl $0, -44(%rbp)
-; NUM-NEXT:    movq $__gxx_personality_sj0, -120(%rbp)
-; NUM-NEXT:    movq $GCC_except_table0, -112(%rbp)
-; NUM-NEXT:    movq %rbp, -104(%rbp)
-; NUM-NEXT:    movq %rsp, -88(%rbp)
-; NUM-NEXT:    movq $.LBB0_9, -96(%rbp)
-; NUM-NEXT:    movl $1, -144(%rbp)
-; NUM-NEXT:    leaq -152(%rbp), %rdi
-; NUM-NEXT:    callq _Unwind_SjLj_Register
-; NUM-NEXT:  .Ltmp0:
-; NUM-NEXT:    callq _Z3foov
-; NUM-NEXT:  .Ltmp1:
-; NUM-NEXT:  # %bb.1: # %invoke.cont
-; NUM-NEXT:    movl $1, -44(%rbp)
-; NUM-NEXT:  .LBB0_7: # %return
-; NUM-NEXT:    movl -44(%rbp), %ebx
-; NUM-NEXT:    leaq -152(%rbp), %rdi
-; NUM-NEXT:    callq _Unwind_SjLj_Unregister
-; NUM-NEXT:    movl %ebx, %eax
-; NUM-NEXT:    addq $120, %rsp
-; NUM-NEXT:    popq %rbx
-; NUM-NEXT:    popq %r12
-; NUM-NEXT:    popq %r13
-; NUM-NEXT:    popq %r14
-; NUM-NEXT:    popq %r15
-; NUM-NEXT:    popq %rbp
-; NUM-NEXT:    retq
-; NUM-NEXT:  .LBB0_9:
-; NUM-NEXT:    endbr64
-; NUM-NEXT:    movl -144(%rbp), %eax
-; NUM-NEXT:    cmpl $1, %eax
-; NUM-NEXT:    jb .LBB0_10
-; NUM-NEXT:  # %bb.11:
-; NUM-NEXT:    ud2
-; NUM-NEXT:  .LBB0_10:
-; NUM-NEXT:    leaq {{.*}}(%rip), %rcx
-; NUM-NEXT:    jmpq *(%rcx,%rax,8)
-; NUM-NEXT:  .LBB0_2: # %lpad
-; NUM-NEXT:  .Ltmp2:
-; NUM-NEXT:    endbr64
-; NUM-NEXT:    movl -140(%rbp), %ecx
-; NUM-NEXT:    movl -136(%rbp), %eax
-; NUM-NEXT:    movq %rcx, -56(%rbp)
-; NUM-NEXT:    movl %eax, -64(%rbp)
-; NUM-NEXT:    cmpl $2, %eax
-; NUM-NEXT:    je .LBB0_3
-; NUM-NEXT:  # %bb.4: # %catch.fallthrough
-; NUM-NEXT:    cmpl $1, %eax
-; NUM-NEXT:    jne .LBB0_8
-; NUM-NEXT:  # %bb.5: # %catch
-; NUM-NEXT:    movq -56(%rbp), %rdi
-; NUM-NEXT:    movl $-1, -144(%rbp)
-; NUM-NEXT:    callq __cxa_begin_catch
-; NUM-NEXT:    movb (%rax), %al
-; NUM-NEXT:    movb %al, -45(%rbp)
-; NUM-NEXT:    xorl %ecx, %ecx
-; NUM-NEXT:    cmpb $3, %al
-; NUM-NEXT:    jmp .LBB0_6
-; NUM-NEXT:  .LBB0_3: # %catch3
-; NUM-NEXT:    movq -56(%rbp), %rdi
-; NUM-NEXT:    movl $-1, -144(%rbp)
-; NUM-NEXT:    callq __cxa_begin_catch
-; NUM-NEXT:    movl (%rax), %eax
-; NUM-NEXT:    movl %eax, -60(%rbp)
-; NUM-NEXT:    xorl %ecx, %ecx
-; NUM-NEXT:    cmpl $5, %eax
-; NUM-NEXT:  .LBB0_6: # %return
-; NUM-NEXT:    setne %cl
-; NUM-NEXT:    movl %ecx, -44(%rbp)
-; NUM-NEXT:    movl $-1, -144(%rbp)
-; NUM-NEXT:    callq __cxa_end_catch
-; NUM-NEXT:    jmp .LBB0_7
-; NUM-NEXT:  .LBB0_8: # %eh.resume
-; NUM-NEXT:    movl $-1, -144(%rbp)
-;
-; SJLJ-LABEL: main:
-; SJLJ:       # %bb.0: # %entry
-; SJLJ-NEXT:    endbr64
-; SJLJ-NEXT:    pushq %rbp
-; SJLJ-NEXT:    movq %rsp, %rbp
-; SJLJ-NEXT:    pushq %r15
-; SJLJ-NEXT:    pushq %r14
-; SJLJ-NEXT:    pushq %r13
-; SJLJ-NEXT:    pushq %r12
-; SJLJ-NEXT:    pushq %rbx
-; SJLJ-NEXT:    subq $120, %rsp
-; SJLJ-NEXT:    movl $0, -44(%rbp)
-; SJLJ-NEXT:    movq $__gxx_personality_sj0, -120(%rbp)
-; SJLJ-NEXT:    movq $GCC_except_table0, -112(%rbp)
-; SJLJ-NEXT:    movq %rbp, -104(%rbp)
-; SJLJ-NEXT:    movq %rsp, -88(%rbp)
-; SJLJ-NEXT:    movq $.LBB0_9, -96(%rbp)
-; SJLJ-NEXT:    movl $1, -144(%rbp)
-; SJLJ-NEXT:    leaq -152(%rbp), %rdi
-; SJLJ-NEXT:    callq _Unwind_SjLj_Register
-; SJLJ-NEXT:  .Ltmp0:
-; SJLJ-NEXT:    callq _Z3foov
-; SJLJ-NEXT:  .Ltmp1:
-; SJLJ-NEXT:  # %bb.1: # %invoke.cont
-; SJLJ-NEXT:    movl $1, -44(%rbp)
-; SJLJ-NEXT:  .LBB0_7: # %return
-; SJLJ-NEXT:    movl -44(%rbp), %ebx
-; SJLJ-NEXT:    leaq -152(%rbp), %rdi
-; SJLJ-NEXT:    callq _Unwind_SjLj_Unregister
-; SJLJ-NEXT:    movl %ebx, %eax
-; SJLJ-NEXT:    addq $120, %rsp
-; SJLJ-NEXT:    popq %rbx
-; SJLJ-NEXT:    popq %r12
-; SJLJ-NEXT:    popq %r13
-; SJLJ-NEXT:    popq %r14
-; SJLJ-NEXT:    popq %r15
-; SJLJ-NEXT:    popq %rbp
-; SJLJ-NEXT:    retq
-; SJLJ-NEXT:  .LBB0_9:
-; SJLJ-NEXT:    endbr64
-; SJLJ-NEXT:    movl -144(%rbp), %eax
-; SJLJ-NEXT:    cmpl $1, %eax
-; SJLJ-NEXT:    jb .LBB0_10
-; SJLJ-NEXT:  # %bb.11:
-; SJLJ-NEXT:    ud2
-; SJLJ-NEXT:  .LBB0_10:
-; SJLJ-NEXT:    leaq {{.*}}(%rip), %rcx
-; SJLJ-NEXT:    jmpq *(%rcx,%rax,8)
-; SJLJ-NEXT:  .LBB0_2: # %lpad
-; SJLJ-NEXT:  .Ltmp2:
-; SJLJ-NEXT:    endbr64
-; SJLJ-NEXT:    movl -140(%rbp), %ecx
-; SJLJ-NEXT:    movl -136(%rbp), %eax
-; SJLJ-NEXT:    movq %rcx, -56(%rbp)
-; SJLJ-NEXT:    movl %eax, -64(%rbp)
-; SJLJ-NEXT:    cmpl $2, %eax
-; SJLJ-NEXT:    je .LBB0_3
-; SJLJ-NEXT:  # %bb.4: # %catch.fallthrough
-; SJLJ-NEXT:    cmpl $1, %eax
-; SJLJ-NEXT:    jne .LBB0_8
-; SJLJ-NEXT:  # %bb.5: # %catch
-; SJLJ-NEXT:    movq -56(%rbp), %rdi
-; SJLJ-NEXT:    movl $-1, -144(%rbp)
-; SJLJ-NEXT:    callq __cxa_begin_catch
-; SJLJ-NEXT:    movb (%rax), %al
-; SJLJ-NEXT:    movb %al, -45(%rbp)
-; SJLJ-NEXT:    xorl %ecx, %ecx
-; SJLJ-NEXT:    cmpb $3, %al
-; SJLJ-NEXT:    jmp .LBB0_6
-; SJLJ-NEXT:  .LBB0_3: # %catch3
-; SJLJ-NEXT:    movq -56(%rbp), %rdi
-; SJLJ-NEXT:    movl $-1, -144(%rbp)
-; SJLJ-NEXT:    callq __cxa_begin_catch
-; SJLJ-NEXT:    movl (%rax), %eax
-; SJLJ-NEXT:    movl %eax, -60(%rbp)
-; SJLJ-NEXT:    xorl %ecx, %ecx
-; SJLJ-NEXT:    cmpl $5, %eax
-; SJLJ-NEXT:  .LBB0_6: # %return
-; SJLJ-NEXT:    setne %cl
-; SJLJ-NEXT:    movl %ecx, -44(%rbp)
-; SJLJ-NEXT:    movl $-1, -144(%rbp)
-; SJLJ-NEXT:    callq __cxa_end_catch
-; SJLJ-NEXT:    jmp .LBB0_7
-; SJLJ-NEXT:  .LBB0_8: # %eh.resume
-; SJLJ-NEXT:    movl $-1, -144(%rbp)
 entry:
   %retval = alloca i32, align 4
   %exn.slot = alloca i8*

diff --git a/llvm/test/CodeGen/X86/jump_sign.ll b/llvm/test/CodeGen/X86/jump_sign.ll
@@ -139,12 +139,11 @@ define i32 @func_l2(i32 %a, i32 %b) nounwind {
 ; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; CHECK-NEXT:    movl %ecx, %eax
 ; CHECK-NEXT:    subl %edx, %eax
-; CHECK-NEXT:    je .LBB8_1
-; CHECK-NEXT:  # %bb.2: # %if.else
-; CHECK-NEXT:    retl
-; CHECK-NEXT:  .LBB8_1: # %if.then
+; CHECK-NEXT:    jne .LBB8_2
+; CHECK-NEXT:  # %bb.1: # %if.then
 ; CHECK-NEXT:    cmpl %ecx, %edx
 ; CHECK-NEXT:    cmovlel %ecx, %eax
+; CHECK-NEXT:  .LBB8_2: # %if.else
 ; CHECK-NEXT:    retl
   %cmp = icmp eq i32 %b, %a
   %sub = sub nsw i32 %a, %b
@@ -330,13 +329,12 @@ define i8* @func_r(i8* %base, i32* nocapture %offset, i32 %size) nounwind {
 ; CHECK-NEXT:    movl (%edx), %ecx
 ; CHECK-NEXT:    xorl %eax, %eax
 ; CHECK-NEXT:    subl {{[0-9]+}}(%esp), %ecx
-; CHECK-NEXT:    jge .LBB15_1
-; CHECK-NEXT:  # %bb.2: # %return
-; CHECK-NEXT:    retl
-; CHECK-NEXT:  .LBB15_1: # %if.end
+; CHECK-NEXT:    jl .LBB15_2
+; CHECK-NEXT:  # %bb.1: # %if.end
 ; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; CHECK-NEXT:    movl %ecx, (%edx)
 ; CHECK-NEXT:    addl %ecx, %eax
+; CHECK-NEXT:  .LBB15_2: # %return
 ; CHECK-NEXT:    retl
 entry:
   %0 = load i32, i32* %offset, align 8

diff --git a/llvm/test/CodeGen/X86/lsr-negative-stride.ll b/llvm/test/CodeGen/X86/lsr-negative-stride.ll
@@ -19,7 +19,11 @@ define i32 @t(i32 %a, i32 %b) nounwind {
 ; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %edx
 ; CHECK-NEXT:    cmpl %ecx, %edx
-; CHECK-NEXT:    je .LBB0_1
+; CHECK-NEXT:    jne .LBB0_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    movl %edx, %eax
+; CHECK-NEXT:    retl
+; CHECK-NEXT:    .p2align 4, 0x90
 ; CHECK-NEXT:  .LBB0_2: # %bb.outer
 ; CHECK-NEXT:    # =>This Loop Header: Depth=1
 ; CHECK-NEXT:    # Child Loop BB0_3 Depth 2
@@ -45,9 +49,6 @@ define i32 @t(i32 %a, i32 %b) nounwind {
 ; CHECK-NEXT:    jne .LBB0_2
 ; CHECK-NEXT:  .LBB0_6: # %bb17
 ; CHECK-NEXT:    retl
-; CHECK-NEXT:  .LBB0_1:
-; CHECK-NEXT:    movl %edx, %eax
-; CHECK-NEXT:    retl
 entry:
 	%tmp1434 = icmp eq i32 %a, %b		; <i1> [#uses=1]
 	br i1 %tmp1434, label %bb17, label %bb.outer

diff --git a/llvm/test/CodeGen/X86/machine-cse.ll b/llvm/test/CodeGen/X86/machine-cse.ll
@@ -110,11 +110,10 @@ define i32 @cross_mbb_phys_cse(i32 %a, i32 %b) nounwind ssp {
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    movl $1, %eax
 ; CHECK-NEXT:    cmpl %esi, %edi
-; CHECK-NEXT:    jbe .LBB2_1
-; CHECK-NEXT:  # %bb.2: # %return
-; CHECK-NEXT:    retq
-; CHECK-NEXT:  .LBB2_1: # %if.end
+; CHECK-NEXT:    ja .LBB2_2
+; CHECK-NEXT:  # %bb.1: # %if.end
 ; CHECK-NEXT:    sbbl %eax, %eax
+; CHECK-NEXT:  .LBB2_2: # %return
 ; CHECK-NEXT:    retq
 entry:
   %cmp = icmp ugt i32 %a, %b

diff --git a/llvm/test/CodeGen/X86/memcmp-more-load-pairs.ll b/llvm/test/CodeGen/X86/memcmp-more-load-pairs.ll
diff --git a/llvm/test/CodeGen/X86/memcmp-optsize.ll b/llvm/test/CodeGen/X86/memcmp-optsize.ll
@@ -120,16 +120,16 @@ define i32 @length3(i8* %X, i8* %Y) nounwind optsize {
 ; X86-NEXT:    rolw $8, %dx
 ; X86-NEXT:    rolw $8, %si
 ; X86-NEXT:    cmpw %si, %dx
-; X86-NEXT:    je .LBB4_1
-; X86-NEXT:  # %bb.3: # %res_block
-; X86-NEXT:    setae %al
-; X86-NEXT:    movzbl %al, %eax
-; X86-NEXT:    leal -1(%eax,%eax), %eax
-; X86-NEXT:    jmp .LBB4_2
-; X86-NEXT:  .LBB4_1: # %loadbb1
+; X86-NEXT:    jne .LBB4_3
+; X86-NEXT:  # %bb.1: # %loadbb1
 ; X86-NEXT:    movzbl 2(%eax), %eax
 ; X86-NEXT:    movzbl 2(%ecx), %ecx
 ; X86-NEXT:    subl %ecx, %eax
+; X86-NEXT:    jmp .LBB4_2
+; X86-NEXT:  .LBB4_3: # %res_block
+; X86-NEXT:    setae %al
+; X86-NEXT:    movzbl %al, %eax
+; X86-NEXT:    leal -1(%eax,%eax), %eax
 ; X86-NEXT:  .LBB4_2: # %endblock
 ; X86-NEXT:    popl %esi
 ; X86-NEXT:    retl
@@ -141,16 +141,16 @@ define i32 @length3(i8* %X, i8* %Y) nounwind optsize {
 ; X64-NEXT:    rolw $8, %ax
 ; X64-NEXT:    rolw $8, %cx
 ; X64-NEXT:    cmpw %cx, %ax
-; X64-NEXT:    je .LBB4_1
-; X64-NEXT:  # %bb.3: # %res_block
-; X64-NEXT:    setae %al
-; X64-NEXT:    movzbl %al, %eax
-; X64-NEXT:    leal -1(%rax,%rax), %eax
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB4_1: # %loadbb1
+; X64-NEXT:    jne .LBB4_3
+; X64-NEXT:  # %bb.1: # %loadbb1
 ; X64-NEXT:    movzbl 2(%rdi), %eax
 ; X64-NEXT:    movzbl 2(%rsi), %ecx
 ; X64-NEXT:    subl %ecx, %eax
+; X64-NEXT:    retq
+; X64-NEXT:  .LBB4_3: # %res_block
+; X64-NEXT:    setae %al
+; X64-NEXT:    movzbl %al, %eax
+; X64-NEXT:    leal -1(%rax,%rax), %eax
 ; X64-NEXT:    retq
   %m = tail call i32 @memcmp(i8* %X, i8* %Y, i64 3) nounwind
   ret i32 %m
@@ -265,16 +265,16 @@ define i32 @length5(i8* %X, i8* %Y) nounwind optsize {
 ; X86-NEXT:    bswapl %edx
 ; X86-NEXT:    bswapl %esi
 ; X86-NEXT:    cmpl %esi, %edx
-; X86-NEXT:    je .LBB9_1
-; X86-NEXT:  # %bb.3: # %res_block
-; X86-NEXT:    setae %al
-; X86-NEXT:    movzbl %al, %eax
-; X86-NEXT:    leal -1(%eax,%eax), %eax
-; X86-NEXT:    jmp .LBB9_2
-; X86-NEXT:  .LBB9_1: # %loadbb1
+; X86-NEXT:    jne .LBB9_3
+; X86-NEXT:  # %bb.1: # %loadbb1
 ; X86-NEXT:    movzbl 4(%eax), %eax
 ; X86-NEXT:    movzbl 4(%ecx), %ecx
 ; X86-NEXT:    subl %ecx, %eax
+; X86-NEXT:    jmp .LBB9_2
+; X86-NEXT:  .LBB9_3: # %res_block
+; X86-NEXT:    setae %al
+; X86-NEXT:    movzbl %al, %eax
+; X86-NEXT:    leal -1(%eax,%eax), %eax
 ; X86-NEXT:  .LBB9_2: # %endblock
 ; X86-NEXT:    popl %esi
 ; X86-NEXT:    retl
@@ -286,16 +286,16 @@ define i32 @length5(i8* %X, i8* %Y) nounwind optsize {
 ; X64-NEXT:    bswapl %eax
 ; X64-NEXT:    bswapl %ecx
 ; X64-NEXT:    cmpl %ecx, %eax
-; X64-NEXT:    je .LBB9_1
-; X64-NEXT:  # %bb.3: # %res_block
-; X64-NEXT:    setae %al
-; X64-NEXT:    movzbl %al, %eax
-; X64-NEXT:    leal -1(%rax,%rax), %eax
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB9_1: # %loadbb1
+; X64-NEXT:    jne .LBB9_3
+; X64-NEXT:  # %bb.1: # %loadbb1
 ; X64-NEXT:    movzbl 4(%rdi), %eax
 ; X64-NEXT:    movzbl 4(%rsi), %ecx
 ; X64-NEXT:    subl %ecx, %eax
+; X64-NEXT:    retq
+; X64-NEXT:  .LBB9_3: # %res_block
+; X64-NEXT:    setae %al
+; X64-NEXT:    movzbl %al, %eax
+; X64-NEXT:    leal -1(%rax,%rax), %eax
 ; X64-NEXT:    retq
   %m = tail call i32 @memcmp(i8* %X, i8* %Y, i64 5) nounwind
   ret i32 %m
@@ -334,32 +334,30 @@ define i32 @length8(i8* %X, i8* %Y) nounwind optsize {
 ; X86-LABEL: length8:
 ; X86:       # %bb.0:
 ; X86-NEXT:    pushl %esi
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
-; X86-NEXT:    movl (%esi), %eax
-; X86-NEXT:    movl (%edx), %ecx
-; X86-NEXT:    bswapl %eax
+; X86-NEXT:    movl (%esi), %ecx
+; X86-NEXT:    movl (%eax), %edx
 ; X86-NEXT:    bswapl %ecx
-; X86-NEXT:    cmpl %ecx, %eax
-; X86-NEXT:    jne .LBB11_4
+; X86-NEXT:    bswapl %edx
+; X86-NEXT:    cmpl %edx, %ecx
+; X86-NEXT:    jne .LBB11_2
 ; X86-NEXT:  # %bb.1: # %loadbb1
-; X86-NEXT:    movl 4(%esi), %eax
-; X86-NEXT:    movl 4(%edx), %ecx
-; X86-NEXT:    bswapl %eax
+; X86-NEXT:    movl 4(%esi), %ecx
+; X86-NEXT:    movl 4(%eax), %edx
 ; X86-NEXT:    bswapl %ecx
-; X86-NEXT:    cmpl %ecx, %eax
-; X86-NEXT:    je .LBB11_2
-; X86-NEXT:  .LBB11_4: # %res_block
-; X86-NEXT:    xorl %edx, %edx
-; X86-NEXT:    cmpl %ecx, %eax
-; X86-NEXT:    setae %dl
-; X86-NEXT:    leal -1(%edx,%edx), %eax
+; X86-NEXT:    bswapl %edx
+; X86-NEXT:    xorl %eax, %eax
+; X86-NEXT:    cmpl %edx, %ecx
+; X86-NEXT:    je .LBB11_3
+; X86-NEXT:  .LBB11_2: # %res_block
+; X86-NEXT:    xorl %eax, %eax
+; X86-NEXT:    cmpl %edx, %ecx
+; X86-NEXT:    setae %al
+; X86-NEXT:    leal -1(%eax,%eax), %eax
 ; X86-NEXT:  .LBB11_3: # %endblock
 ; X86-NEXT:    popl %esi
 ; X86-NEXT:    retl
-; X86-NEXT:  .LBB11_2:
-; X86-NEXT:    xorl %eax, %eax
-; X86-NEXT:    jmp .LBB11_3
 ;
 ; X64-LABEL: length8:
 ; X64:       # %bb.0:
@@ -463,27 +461,26 @@ define i32 @length12(i8* %X, i8* %Y) nounwind optsize {
 ;
 ; X64-LABEL: length12:
 ; X64:       # %bb.0:
-; X64-NEXT:    movq (%rdi), %rax
-; X64-NEXT:    movq (%rsi), %rcx
-; X64-NEXT:    bswapq %rax
+; X64-NEXT:    movq (%rdi), %rcx
+; X64-NEXT:    movq (%rsi), %rdx
 ; X64-NEXT:    bswapq %rcx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    jne .LBB15_4
+; X64-NEXT:    bswapq %rdx
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    jne .LBB15_2
 ; X64-NEXT:  # %bb.1: # %loadbb1
-; X64-NEXT:    movl 8(%rdi), %eax
-; X64-NEXT:    movl 8(%rsi), %ecx
-; X64-NEXT:    bswapl %eax
+; X64-NEXT:    movl 8(%rdi), %ecx
+; X64-NEXT:    movl 8(%rsi), %edx
 ; X64-NEXT:    bswapl %ecx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    je .LBB15_2
-; X64-NEXT:  .LBB15_4: # %res_block
-; X64-NEXT:    xorl %edx, %edx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    setae %dl
-; X64-NEXT:    leal -1(%rdx,%rdx), %eax
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB15_2:
+; X64-NEXT:    bswapl %edx
+; X64-NEXT:    xorl %eax, %eax
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    je .LBB15_3
+; X64-NEXT:  .LBB15_2: # %res_block
 ; X64-NEXT:    xorl %eax, %eax
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    setae %al
+; X64-NEXT:    leal -1(%rax,%rax), %eax
+; X64-NEXT:  .LBB15_3: # %endblock
 ; X64-NEXT:    retq
   %m = tail call i32 @memcmp(i8* %X, i8* %Y, i64 12) nounwind
   ret i32 %m
@@ -504,27 +501,26 @@ define i32 @length16(i8* %X, i8* %Y) nounwind optsize {
 ;
 ; X64-LABEL: length16:
 ; X64:       # %bb.0:
-; X64-NEXT:    movq (%rdi), %rax
-; X64-NEXT:    movq (%rsi), %rcx
-; X64-NEXT:    bswapq %rax
+; X64-NEXT:    movq (%rdi), %rcx
+; X64-NEXT:    movq (%rsi), %rdx
 ; X64-NEXT:    bswapq %rcx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    jne .LBB16_4
+; X64-NEXT:    bswapq %rdx
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    jne .LBB16_2
 ; X64-NEXT:  # %bb.1: # %loadbb1
-; X64-NEXT:    movq 8(%rdi), %rax
-; X64-NEXT:    movq 8(%rsi), %rcx
-; X64-NEXT:    bswapq %rax
+; X64-NEXT:    movq 8(%rdi), %rcx
+; X64-NEXT:    movq 8(%rsi), %rdx
 ; X64-NEXT:    bswapq %rcx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    je .LBB16_2
-; X64-NEXT:  .LBB16_4: # %res_block
-; X64-NEXT:    xorl %edx, %edx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    setae %dl
-; X64-NEXT:    leal -1(%rdx,%rdx), %eax
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB16_2:
+; X64-NEXT:    bswapq %rdx
 ; X64-NEXT:    xorl %eax, %eax
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    je .LBB16_3
+; X64-NEXT:  .LBB16_2: # %res_block
+; X64-NEXT:    xorl %eax, %eax
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    setae %al
+; X64-NEXT:    leal -1(%rax,%rax), %eax
+; X64-NEXT:  .LBB16_3: # %endblock
 ; X64-NEXT:    retq
   %m = tail call i32 @memcmp(i8* %X, i8* %Y, i64 16) nounwind
   ret i32 %m

diff --git a/llvm/test/CodeGen/X86/memcmp-pgso.ll b/llvm/test/CodeGen/X86/memcmp-pgso.ll
@@ -120,19 +120,19 @@ define i32 @length3(i8* %X, i8* %Y) nounwind !prof !14 {
 ; X86-NEXT:    rolw $8, %dx
 ; X86-NEXT:    rolw $8, %si
 ; X86-NEXT:    cmpw %si, %dx
-; X86-NEXT:    je .LBB4_1
-; X86-NEXT:  # %bb.3: # %res_block
+; X86-NEXT:    jne .LBB4_3
+; X86-NEXT:  # %bb.1: # %loadbb1
+; X86-NEXT:    movzbl 2(%eax), %eax
+; X86-NEXT:    movzbl 2(%ecx), %ecx
+; X86-NEXT:    subl %ecx, %eax
+; X86-NEXT:    jmp .LBB4_2
+; X86-NEXT:  .LBB4_3: # %res_block
 ; X86-NEXT:    setae %al
 ; X86-NEXT:    movzbl %al, %eax
 ; X86-NEXT:    leal -1(%eax,%eax), %eax
 ; X86-NEXT:  .LBB4_2: # %endblock
 ; X86-NEXT:    popl %esi
 ; X86-NEXT:    retl
-; X86-NEXT:  .LBB4_1: # %loadbb1
-; X86-NEXT:    movzbl 2(%eax), %eax
-; X86-NEXT:    movzbl 2(%ecx), %ecx
-; X86-NEXT:    subl %ecx, %eax
-; X86-NEXT:    jmp .LBB4_2
 ;
 ; X64-LABEL: length3:
 ; X64:       # %bb.0:
@@ -141,16 +141,16 @@ define i32 @length3(i8* %X, i8* %Y) nounwind !prof !14 {
 ; X64-NEXT:    rolw $8, %ax
 ; X64-NEXT:    rolw $8, %cx
 ; X64-NEXT:    cmpw %cx, %ax
-; X64-NEXT:    je .LBB4_1
-; X64-NEXT:  # %bb.3: # %res_block
-; X64-NEXT:    setae %al
-; X64-NEXT:    movzbl %al, %eax
-; X64-NEXT:    leal -1(%rax,%rax), %eax
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB4_1: # %loadbb1
+; X64-NEXT:    jne .LBB4_3
+; X64-NEXT:  # %bb.1: # %loadbb1
 ; X64-NEXT:    movzbl 2(%rdi), %eax
 ; X64-NEXT:    movzbl 2(%rsi), %ecx
 ; X64-NEXT:    subl %ecx, %eax
+; X64-NEXT:    retq
+; X64-NEXT:  .LBB4_3: # %res_block
+; X64-NEXT:    setae %al
+; X64-NEXT:    movzbl %al, %eax
+; X64-NEXT:    leal -1(%rax,%rax), %eax
 ; X64-NEXT:    retq
   %m = tail call i32 @memcmp(i8* %X, i8* %Y, i64 3) nounwind
   ret i32 %m
@@ -265,19 +265,19 @@ define i32 @length5(i8* %X, i8* %Y) nounwind !prof !14 {
 ; X86-NEXT:    bswapl %edx
 ; X86-NEXT:    bswapl %esi
 ; X86-NEXT:    cmpl %esi, %edx
-; X86-NEXT:    je .LBB9_1
-; X86-NEXT:  # %bb.3: # %res_block
+; X86-NEXT:    jne .LBB9_3
+; X86-NEXT:  # %bb.1: # %loadbb1
+; X86-NEXT:    movzbl 4(%eax), %eax
+; X86-NEXT:    movzbl 4(%ecx), %ecx
+; X86-NEXT:    subl %ecx, %eax
+; X86-NEXT:    jmp .LBB9_2
+; X86-NEXT:  .LBB9_3: # %res_block
 ; X86-NEXT:    setae %al
 ; X86-NEXT:    movzbl %al, %eax
 ; X86-NEXT:    leal -1(%eax,%eax), %eax
 ; X86-NEXT:  .LBB9_2: # %endblock
 ; X86-NEXT:    popl %esi
 ; X86-NEXT:    retl
-; X86-NEXT:  .LBB9_1: # %loadbb1
-; X86-NEXT:    movzbl 4(%eax), %eax
-; X86-NEXT:    movzbl 4(%ecx), %ecx
-; X86-NEXT:    subl %ecx, %eax
-; X86-NEXT:    jmp .LBB9_2
 ;
 ; X64-LABEL: length5:
 ; X64:       # %bb.0:
@@ -286,16 +286,16 @@ define i32 @length5(i8* %X, i8* %Y) nounwind !prof !14 {
 ; X64-NEXT:    bswapl %eax
 ; X64-NEXT:    bswapl %ecx
 ; X64-NEXT:    cmpl %ecx, %eax
-; X64-NEXT:    je .LBB9_1
-; X64-NEXT:  # %bb.3: # %res_block
-; X64-NEXT:    setae %al
-; X64-NEXT:    movzbl %al, %eax
-; X64-NEXT:    leal -1(%rax,%rax), %eax
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB9_1: # %loadbb1
+; X64-NEXT:    jne .LBB9_3
+; X64-NEXT:  # %bb.1: # %loadbb1
 ; X64-NEXT:    movzbl 4(%rdi), %eax
 ; X64-NEXT:    movzbl 4(%rsi), %ecx
 ; X64-NEXT:    subl %ecx, %eax
+; X64-NEXT:    retq
+; X64-NEXT:  .LBB9_3: # %res_block
+; X64-NEXT:    setae %al
+; X64-NEXT:    movzbl %al, %eax
+; X64-NEXT:    leal -1(%rax,%rax), %eax
 ; X64-NEXT:    retq
   %m = tail call i32 @memcmp(i8* %X, i8* %Y, i64 5) nounwind
   ret i32 %m
@@ -334,32 +334,30 @@ define i32 @length8(i8* %X, i8* %Y) nounwind !prof !14 {
 ; X86-LABEL: length8:
 ; X86:       # %bb.0:
 ; X86-NEXT:    pushl %esi
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
-; X86-NEXT:    movl (%esi), %eax
-; X86-NEXT:    movl (%edx), %ecx
-; X86-NEXT:    bswapl %eax
+; X86-NEXT:    movl (%esi), %ecx
+; X86-NEXT:    movl (%eax), %edx
+; X86-NEXT:    bswapl %ecx
+; X86-NEXT:    bswapl %edx
+; X86-NEXT:    cmpl %edx, %ecx
+; X86-NEXT:    jne .LBB11_2
+; X86-NEXT:  # %bb.1: # %loadbb1
+; X86-NEXT:    movl 4(%esi), %ecx
+; X86-NEXT:    movl 4(%eax), %edx
 ; X86-NEXT:    bswapl %ecx
-; X86-NEXT:    cmpl %ecx, %eax
-; X86-NEXT:    je .LBB11_1
-; X86-NEXT:  .LBB11_4: # %res_block
-; X86-NEXT:    xorl %edx, %edx
-; X86-NEXT:    cmpl %ecx, %eax
-; X86-NEXT:    setae %dl
-; X86-NEXT:    leal -1(%edx,%edx), %eax
+; X86-NEXT:    bswapl %edx
+; X86-NEXT:    xorl %eax, %eax
+; X86-NEXT:    cmpl %edx, %ecx
+; X86-NEXT:    je .LBB11_3
+; X86-NEXT:  .LBB11_2: # %res_block
+; X86-NEXT:    xorl %eax, %eax
+; X86-NEXT:    cmpl %edx, %ecx
+; X86-NEXT:    setae %al
+; X86-NEXT:    leal -1(%eax,%eax), %eax
 ; X86-NEXT:  .LBB11_3: # %endblock
 ; X86-NEXT:    popl %esi
 ; X86-NEXT:    retl
-; X86-NEXT:  .LBB11_1: # %loadbb1
-; X86-NEXT:    movl 4(%esi), %eax
-; X86-NEXT:    movl 4(%edx), %ecx
-; X86-NEXT:    bswapl %eax
-; X86-NEXT:    bswapl %ecx
-; X86-NEXT:    cmpl %ecx, %eax
-; X86-NEXT:    jne .LBB11_4
-; X86-NEXT:  # %bb.2:
-; X86-NEXT:    xorl %eax, %eax
-; X86-NEXT:    jmp .LBB11_3
 ;
 ; X64-LABEL: length8:
 ; X64:       # %bb.0:
@@ -463,27 +461,26 @@ define i32 @length12(i8* %X, i8* %Y) nounwind !prof !14 {
 ;
 ; X64-LABEL: length12:
 ; X64:       # %bb.0:
-; X64-NEXT:    movq (%rdi), %rax
-; X64-NEXT:    movq (%rsi), %rcx
-; X64-NEXT:    bswapq %rax
+; X64-NEXT:    movq (%rdi), %rcx
+; X64-NEXT:    movq (%rsi), %rdx
 ; X64-NEXT:    bswapq %rcx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    je .LBB15_1
-; X64-NEXT:  .LBB15_4: # %res_block
-; X64-NEXT:    xorl %edx, %edx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    setae %dl
-; X64-NEXT:    leal -1(%rdx,%rdx), %eax
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB15_1: # %loadbb1
-; X64-NEXT:    movl 8(%rdi), %eax
-; X64-NEXT:    movl 8(%rsi), %ecx
-; X64-NEXT:    bswapl %eax
+; X64-NEXT:    bswapq %rdx
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    jne .LBB15_2
+; X64-NEXT:  # %bb.1: # %loadbb1
+; X64-NEXT:    movl 8(%rdi), %ecx
+; X64-NEXT:    movl 8(%rsi), %edx
 ; X64-NEXT:    bswapl %ecx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    jne .LBB15_4
-; X64-NEXT:  # %bb.2:
+; X64-NEXT:    bswapl %edx
+; X64-NEXT:    xorl %eax, %eax
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    je .LBB15_3
+; X64-NEXT:  .LBB15_2: # %res_block
 ; X64-NEXT:    xorl %eax, %eax
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    setae %al
+; X64-NEXT:    leal -1(%rax,%rax), %eax
+; X64-NEXT:  .LBB15_3: # %endblock
 ; X64-NEXT:    retq
   %m = tail call i32 @memcmp(i8* %X, i8* %Y, i64 12) nounwind
   ret i32 %m
@@ -504,27 +501,26 @@ define i32 @length16(i8* %X, i8* %Y) nounwind !prof !14 {
 ;
 ; X64-LABEL: length16:
 ; X64:       # %bb.0:
-; X64-NEXT:    movq (%rdi), %rax
-; X64-NEXT:    movq (%rsi), %rcx
-; X64-NEXT:    bswapq %rax
+; X64-NEXT:    movq (%rdi), %rcx
+; X64-NEXT:    movq (%rsi), %rdx
 ; X64-NEXT:    bswapq %rcx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    je .LBB16_1
-; X64-NEXT:  .LBB16_4: # %res_block
-; X64-NEXT:    xorl %edx, %edx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    setae %dl
-; X64-NEXT:    leal -1(%rdx,%rdx), %eax
-; X64-NEXT:    retq
-; X64-NEXT:  .LBB16_1: # %loadbb1
-; X64-NEXT:    movq 8(%rdi), %rax
-; X64-NEXT:    movq 8(%rsi), %rcx
-; X64-NEXT:    bswapq %rax
+; X64-NEXT:    bswapq %rdx
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    jne .LBB16_2
+; X64-NEXT:  # %bb.1: # %loadbb1
+; X64-NEXT:    movq 8(%rdi), %rcx
+; X64-NEXT:    movq 8(%rsi), %rdx
 ; X64-NEXT:    bswapq %rcx
-; X64-NEXT:    cmpq %rcx, %rax
-; X64-NEXT:    jne .LBB16_4
-; X64-NEXT:  # %bb.2:
+; X64-NEXT:    bswapq %rdx
 ; X64-NEXT:    xorl %eax, %eax
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    je .LBB16_3
+; X64-NEXT:  .LBB16_2: # %res_block
+; X64-NEXT:    xorl %eax, %eax
+; X64-NEXT:    cmpq %rdx, %rcx
+; X64-NEXT:    setae %al
+; X64-NEXT:    leal -1(%rax,%rax), %eax
+; X64-NEXT:  .LBB16_3: # %endblock
 ; X64-NEXT:    retq
   %m = tail call i32 @memcmp(i8* %X, i8* %Y, i64 16) nounwind
   ret i32 %m

diff --git a/llvm/test/CodeGen/X86/memcmp.ll b/llvm/test/CodeGen/X86/memcmp.ll
diff --git a/llvm/test/CodeGen/X86/neg_cmp.ll b/llvm/test/CodeGen/X86/neg_cmp.ll
@@ -10,11 +10,11 @@ define void @neg_cmp(i32 %x, i32 %y) nounwind {
 ; CHECK-LABEL: neg_cmp:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    addl %esi, %edi
-; CHECK-NEXT:    je .LBB0_2
-; CHECK-NEXT:  # %bb.1: # %if.end
-; CHECK-NEXT:    retq
-; CHECK-NEXT:  .LBB0_2: # %if.then
+; CHECK-NEXT:    jne .LBB0_1
+; CHECK-NEXT:  # %bb.2: # %if.then
 ; CHECK-NEXT:    jmp g # TAILCALL
+; CHECK-NEXT:  .LBB0_1: # %if.end
+; CHECK-NEXT:    retq
   %sub = sub i32 0, %y
   %cmp = icmp eq i32 %x, %sub
   br i1 %cmp, label %if.then, label %if.end
@@ -31,11 +31,11 @@ define void @neg_cmp_commuted(i32 %x, i32 %y) nounwind {
 ; CHECK-LABEL: neg_cmp_commuted:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    addl %esi, %edi
-; CHECK-NEXT:    je .LBB1_2
-; CHECK-NEXT:  # %bb.1: # %if.end
-; CHECK-NEXT:    retq
-; CHECK-NEXT:  .LBB1_2: # %if.then
+; CHECK-NEXT:    jne .LBB1_1
+; CHECK-NEXT:  # %bb.2: # %if.then
 ; CHECK-NEXT:    jmp g # TAILCALL
+; CHECK-NEXT:  .LBB1_1: # %if.end
+; CHECK-NEXT:    retq
   %sub = sub i32 0, %y
   %cmp = icmp eq i32 %sub, %x
   br i1 %cmp, label %if.then, label %if.end

diff --git a/llvm/test/CodeGen/X86/nobt.ll b/llvm/test/CodeGen/X86/nobt.ll
@@ -9,11 +9,10 @@ define void @test2(i32 %x, i32 %n) nounwind {
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    xorl %eax, %eax
 ; CHECK-NEXT:    testb %al, %al
-; CHECK-NEXT:    je .LBB0_1
-; CHECK-NEXT:  # %bb.2: # %UnifiedReturnBlock
-; CHECK-NEXT:    retl
-; CHECK-NEXT:  .LBB0_1: # %bb
+; CHECK-NEXT:    jne .LBB0_2
+; CHECK-NEXT:  # %bb.1: # %bb
 ; CHECK-NEXT:    calll foo
+; CHECK-NEXT:  .LBB0_2: # %UnifiedReturnBlock
 ; CHECK-NEXT:    retl
 entry:
         %tmp1 = and i32 %x, 1
@@ -35,11 +34,10 @@ define void @test3(i32 %x, i32 %n) nounwind {
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    xorl %eax, %eax
 ; CHECK-NEXT:    testb %al, %al
-; CHECK-NEXT:    je .LBB1_1
-; CHECK-NEXT:  # %bb.2: # %UnifiedReturnBlock
-; CHECK-NEXT:    retl
-; CHECK-NEXT:  .LBB1_1: # %bb
+; CHECK-NEXT:    jne .LBB1_2
+; CHECK-NEXT:  # %bb.1: # %bb
 ; CHECK-NEXT:    calll foo
+; CHECK-NEXT:  .LBB1_2: # %UnifiedReturnBlock
 ; CHECK-NEXT:    retl
 entry:
         %tmp1 = and i32 %x, 1

diff --git a/llvm/test/CodeGen/X86/pr29170.ll b/llvm/test/CodeGen/X86/pr29170.ll
@@ -11,11 +11,8 @@ define i32 @main() {
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    xorl %eax, %eax
 ; CHECK-NEXT:    testb %al, %al
-; CHECK-NEXT:    je .LBB0_1
-; CHECK-NEXT:  .LBB0_3: # %if.else
-; CHECK-NEXT:    xorl %eax, %eax
-; CHECK-NEXT:    retl
-; CHECK-NEXT:  .LBB0_1: # %go
+; CHECK-NEXT:    jne .LBB0_3
+; CHECK-NEXT:  # %bb.1: # %go
 ; CHECK-NEXT:    movl $-1, %ecx
 ; CHECK-NEXT:    movsbl b, %edx
 ; CHECK-NEXT:    notl %ecx
@@ -26,6 +23,9 @@ define i32 @main() {
 ; CHECK-NEXT:  # %bb.2: # %if.then
 ; CHECK-NEXT:    movl $42, %eax
 ; CHECK-NEXT:    retl
+; CHECK-NEXT:  .LBB0_3: # %if.else
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    retl
 entry:
   %true = icmp eq i32 0, 0
   %const = bitcast i64 -4294967296 to i64

diff --git a/llvm/test/CodeGen/X86/wide-integer-cmp.ll b/llvm/test/CodeGen/X86/wide-integer-cmp.ll
@@ -9,13 +9,13 @@ define i32 @branch_eq(i64 %a, i64 %b) {
 ; CHECK-NEXT:    xorl {{[0-9]+}}(%esp), %ecx
 ; CHECK-NEXT:    xorl {{[0-9]+}}(%esp), %eax
 ; CHECK-NEXT:    orl %ecx, %eax
-; CHECK-NEXT:    je .LBB0_1
-; CHECK-NEXT:  # %bb.2: # %bb2
-; CHECK-NEXT:    movl $2, %eax
-; CHECK-NEXT:    retl
-; CHECK-NEXT:  .LBB0_1: # %bb1
+; CHECK-NEXT:    jne .LBB0_2
+; CHECK-NEXT:  # %bb.1: # %bb1
 ; CHECK-NEXT:    movl $1, %eax
 ; CHECK-NEXT:    retl
+; CHECK-NEXT:  .LBB0_2: # %bb2
+; CHECK-NEXT:    movl $2, %eax
+; CHECK-NEXT:    retl
 entry:
   %cmp = icmp eq i64 %a, %b
 	br i1 %cmp, label %bb1, label %bb2
@@ -55,13 +55,13 @@ define i32 @branch_ule(i64 %a, i64 %b) {
 ; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; CHECK-NEXT:    cmpl {{[0-9]+}}(%esp), %eax
 ; CHECK-NEXT:    sbbl {{[0-9]+}}(%esp), %ecx
-; CHECK-NEXT:    jae .LBB2_1
-; CHECK-NEXT:  # %bb.2: # %bb2
-; CHECK-NEXT:    movl $2, %eax
-; CHECK-NEXT:    retl
-; CHECK-NEXT:  .LBB2_1: # %bb1
+; CHECK-NEXT:    jb .LBB2_2
+; CHECK-NEXT:  # %bb.1: # %bb1
 ; CHECK-NEXT:    movl $1, %eax
 ; CHECK-NEXT:    retl
+; CHECK-NEXT:  .LBB2_2: # %bb2
+; CHECK-NEXT:    movl $2, %eax
+; CHECK-NEXT:    retl
 entry:
   %cmp = icmp ule i64 %a, %b
 	br i1 %cmp, label %bb1, label %bb2

diff --git a/llvm/test/DebugInfo/COFF/fpo-shrink-wrap.ll b/llvm/test/DebugInfo/COFF/fpo-shrink-wrap.ll
@@ -16,11 +16,8 @@
 ; ASM:         .cv_loc 0 1 3 9                 # t.c:3:9
 ; ASM:         movl    %ecx, %eax
 ; ASM:         cmpl    %edx, %ecx
-; ASM:         jge     LBB0_1
+; ASM:         jl      [[EPILOGUE:LBB0_[0-9]+]]
 
-; ASM:         retl    $8
-
-; ASM: LBB0_1: 
 ; ASM:         pushl   %ebx
 ; ASM:         .cv_fpo_pushreg %ebx
 ; ASM:         pushl   %edi
@@ -34,24 +31,26 @@
 ; ASM:         popl    %esi
 ; ASM:         popl    %edi
 ; ASM:         popl    %ebx
-; ASM: Ltmp11:
+; ASM: [[EPILOGUE]]:                                 # %return
+; ASM:         retl    $8
+; ASM: Ltmp10:
 ; ASM:         .cv_fpo_endproc
 
 ; Note how RvaStart advances 7 bytes to skip the shrink-wrapped portion.
 ; OBJ: SubSectionType: FrameData (0xF5)
 ; OBJ:    FrameData {
 ; OBJ:      RvaStart: 0x0
 ; OBJ:      CodeSize: 0x36
-; OBJ:     PrologSize: 0xC
+; OBJ:      PrologSize: 0x9
 ; OBJ:      FrameFunc [
 ; OBJ-NEXT:   $T0 .raSearch =
 ; OBJ-NEXT:   $eip $T0 ^ =
 ; OBJ-NEXT:   $esp $T0 4 + =
 ; OBJ-NEXT: ]
 ; OBJ:    }
 ; OBJ:    FrameData {
-; OBJ:      RvaStart: 0xA
-; OBJ:      CodeSize: 0x2C
+; OBJ:      RvaStart: 0x7
+; OBJ:      CodeSize: 0x2F
 ; OBJ:      PrologSize: 0x2
 ; OBJ:      FrameFunc [
 ; OBJ-NEXT:   $T0 .raSearch =
@@ -61,8 +60,8 @@
 ; OBJ-NEXT: ]
 ; OBJ:    }
 ; OBJ:    FrameData {
-; OBJ:      RvaStart: 0xB
-; OBJ:      CodeSize: 0x2B
+; OBJ:      RvaStart: 0x8
+; OBJ:      CodeSize: 0x2E
 ; OBJ:      PrologSize: 0x1
 ; OBJ:      FrameFunc [
 ; OBJ-NEXT:   $T0 .raSearch =
@@ -73,8 +72,8 @@
 ; OBJ-NEXT: ]
 ; OBJ:    }
 ; OBJ:    FrameData {
-; OBJ:      RvaStart: 0xC
-; OBJ:      CodeSize: 0x2A
+; OBJ:      RvaStart: 0x9
+; OBJ:      CodeSize: 0x2D
 ; OBJ:      PrologSize: 0x0
 ; OBJ:      FrameFunc [
 ; OBJ-NEXT:   $T0 .raSearch =

diff --git a/llvm/test/Transforms/PGOProfile/counter_promo_mexits.ll b/llvm/test/Transforms/PGOProfile/counter_promo_mexits.ll
@@ -1,95 +1,27 @@
-; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: opt < %s -pgo-instr-gen -instrprof -do-counter-promotion=true -speculative-counter-promotion-max-exiting=3 -S | FileCheck %s --check-prefix=PROMO
-; RUN: opt < %s --passes=pgo-instr-gen,instrprof -do-counter-promotion=true -speculative-counter-promotion-max-exiting=3 -S | FileCheck %s --check-prefix=PROMO
+; RUN: opt < %s -pgo-instr-gen -instrprof -do-counter-promotion=true -speculative-counter-promotion-max-exiting=3 -S | FileCheck --check-prefix=PROMO %s
+; RUN: opt < %s --passes=pgo-instr-gen,instrprof -do-counter-promotion=true -speculative-counter-promotion-max-exiting=3 -S | FileCheck --check-prefix=PROMO %s
 
 @g = common local_unnamed_addr global i32 0, align 4
 
 define void @foo(i32 %arg) local_unnamed_addr {
-; PROMO-LABEL: @foo(
-; PROMO-NEXT:  bb:
-; PROMO-NEXT:    [[T:%.*]] = add nsw i32 [[ARG:%.*]], -1
-; PROMO-NEXT:    br label [[BB1:%.*]]
-; PROMO:       bb1:
-; PROMO-NEXT:    [[PGOCOUNT213:%.*]] = phi i64 [ 0, [[BB:%.*]] ], [ [[PGOCOUNT212:%.*]], [[BB11:%.*]] ]
-; PROMO-NEXT:    [[PGOCOUNT19:%.*]] = phi i64 [ 0, [[BB]] ], [ [[PGOCOUNT18:%.*]], [[BB11]] ]
-; PROMO-NEXT:    [[PGOCOUNT6:%.*]] = phi i64 [ 0, [[BB]] ], [ [[PGOCOUNT5:%.*]], [[BB11]] ]
-; PROMO-NEXT:    [[T2:%.*]] = phi i32 [ 0, [[BB]] ], [ [[T12:%.*]], [[BB11]] ]
-; PROMO-NEXT:    [[T3:%.*]] = icmp sgt i32 [[T2]], [[ARG]]
-; PROMO-NEXT:    br i1 [[T3]], label [[BB7:%.*]], label [[BB4:%.*]]
-; PROMO:       bb4:
-; PROMO-NEXT:    [[TMP0:%.*]] = add i64 [[PGOCOUNT6]], 1
-; PROMO-NEXT:    tail call void @bar(i32 1)
-; PROMO-NEXT:    [[T5:%.*]] = load i32, i32* @g, align 4
-; PROMO-NEXT:    [[T6:%.*]] = icmp sgt i32 [[T5]], 100
-; PROMO-NEXT:    br i1 [[T6]], label [[BB15_0:%.*]], label [[BB11]]
-; PROMO:       bb7:
-; PROMO-NEXT:    [[T8:%.*]] = icmp slt i32 [[T2]], [[T]]
-; PROMO-NEXT:    br i1 [[T8]], label [[BB9:%.*]], label [[BB10:%.*]]
-; PROMO:       bb9:
-; PROMO-NEXT:    [[TMP1:%.*]] = add i64 [[PGOCOUNT19]], 1
-; PROMO-NEXT:    tail call void @bar(i32 2)
-; PROMO-NEXT:    br label [[BB11]]
-; PROMO:       bb10:
-; PROMO-NEXT:    [[TMP2:%.*]] = add i64 [[PGOCOUNT213]], 1
-; PROMO-NEXT:    tail call void @bar(i32 3)
-; PROMO-NEXT:    br label [[BB11]]
-; PROMO:       bb11:
-; PROMO-NEXT:    [[PGOCOUNT212]] = phi i64 [ [[TMP2]], [[BB10]] ], [ [[PGOCOUNT213]], [[BB9]] ], [ [[PGOCOUNT213]], [[BB4]] ]
-; PROMO-NEXT:    [[PGOCOUNT18]] = phi i64 [ [[PGOCOUNT19]], [[BB10]] ], [ [[TMP1]], [[BB9]] ], [ [[PGOCOUNT19]], [[BB4]] ]
-; PROMO-NEXT:    [[PGOCOUNT5]] = phi i64 [ [[PGOCOUNT6]], [[BB10]] ], [ [[PGOCOUNT6]], [[BB9]] ], [ [[TMP0]], [[BB4]] ]
-; PROMO-NEXT:    [[T12]] = add nuw nsw i32 [[T2]], 1
-; PROMO-NEXT:    [[T13:%.*]] = icmp slt i32 [[T2]], 99
-; PROMO-NEXT:    br i1 [[T13]], label [[BB1]], label [[BB14:%.*]]
-; PROMO:       bb14:
-; PROMO-NEXT:    [[PGOCOUNT_PROMOTED7:%.*]] = load i64, i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 1), align 4
-; PROMO-NEXT:    [[TMP3:%.*]] = add i64 [[PGOCOUNT_PROMOTED7]], [[PGOCOUNT5]]
-; PROMO-NEXT:    store i64 [[TMP3]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 1), align 4
-; PROMO-NEXT:    [[PGOCOUNT_PROMOTED11:%.*]] = load i64, i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 0), align 4
-; PROMO-NEXT:    [[TMP4:%.*]] = add i64 [[PGOCOUNT_PROMOTED11]], [[PGOCOUNT18]]
-; PROMO-NEXT:    store i64 [[TMP4]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 0), align 4
-; PROMO-NEXT:    [[PGOCOUNT_PROMOTED15:%.*]] = load i64, i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 2), align 4
-; PROMO-NEXT:    [[TMP5:%.*]] = add i64 [[PGOCOUNT_PROMOTED15]], [[PGOCOUNT212]]
-; PROMO-NEXT:    store i64 [[TMP5]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 2), align 4
-; PROMO-NEXT:    [[PGOCOUNT3:%.*]] = load i64, i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 3), align 4
-; PROMO-NEXT:    [[TMP6:%.*]] = add i64 [[PGOCOUNT3]], 1
-; PROMO-NEXT:    store i64 [[TMP6]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 3), align 4
-; PROMO-NEXT:    tail call void @bar(i32 0)
-; PROMO-NEXT:    br label [[BB15:%.*]]
-; PROMO:       bb15_0:
-; PROMO-NEXT:    [[PGOCOUNT_PROMOTED:%.*]] = load i64, i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 1), align 4
-; PROMO-NEXT:    [[TMP7:%.*]] = add i64 [[PGOCOUNT_PROMOTED]], [[TMP0]]
-; PROMO-NEXT:    store i64 [[TMP7]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 1), align 4
-; PROMO-NEXT:    [[PGOCOUNT_PROMOTED10:%.*]] = load i64, i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 0), align 4
-; PROMO-NEXT:    [[TMP8:%.*]] = add i64 [[PGOCOUNT_PROMOTED10]], [[PGOCOUNT19]]
-; PROMO-NEXT:    store i64 [[TMP8]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 0), align 4
-; PROMO-NEXT:    [[PGOCOUNT_PROMOTED14:%.*]] = load i64, i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 2), align 4
-; PROMO-NEXT:    [[TMP9:%.*]] = add i64 [[PGOCOUNT_PROMOTED14]], [[PGOCOUNT213]]
-; PROMO-NEXT:    store i64 [[TMP9]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 2), align 4
-; PROMO-NEXT:    [[PGOCOUNT4:%.*]] = load i64, i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 4), align 4
-; PROMO-NEXT:    [[TMP10:%.*]] = add i64 [[PGOCOUNT4]], 1
-; PROMO-NEXT:    store i64 [[TMP10]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* @__profc_foo, i64 0, i64 4), align 4
-; PROMO-NEXT:    br label [[BB15]]
-; PROMO:       bb15:
-; PROMO-NEXT:    tail call void @bar(i32 1)
-; PROMO-NEXT:    ret void
-;
+; PROMO-LABEL: @foo
 bb:
-  %t = add nsw i32 %arg, -1
+  %tmp = add nsw i32 %arg, -1
   br label %bb1
 bb1:                                              ; preds = %bb11, %bb
-  %t2 = phi i32 [ 0, %bb ], [ %t12, %bb11 ]
-  %t3 = icmp sgt i32 %t2, %arg
-  br i1 %t3, label %bb7, label %bb4
+  %tmp2 = phi i32 [ 0, %bb ], [ %tmp12, %bb11 ]
+  %tmp3 = icmp sgt i32 %tmp2, %arg
+  br i1 %tmp3, label %bb7, label %bb4
 
 bb4:                                              ; preds = %bb1
   tail call void @bar(i32 1)
-  %t5 = load i32, i32* @g, align 4
-  %t6 = icmp sgt i32 %t5, 100
-  br i1 %t6, label %bb15_0, label %bb11
+  %tmp5 = load i32, i32* @g, align 4
+  %tmp6 = icmp sgt i32 %tmp5, 100
+  br i1 %tmp6, label %bb15_0, label %bb11
 
 bb7:                                              ; preds = %bb1
-  %t8 = icmp slt i32 %t2, %t
-  br i1 %t8, label %bb9, label %bb10
+  %tmp8 = icmp slt i32 %tmp2, %tmp
+  br i1 %tmp8, label %bb9, label %bb10
 
 bb9:                                              ; preds = %bb7
   tail call void @bar(i32 2)
@@ -100,18 +32,45 @@ bb10:                                             ; preds = %bb7
   br label %bb11
 
 bb11:                                             ; preds = %bb10, %bb9, %bb4
-  %t12 = add nuw nsw i32 %t2, 1
-  %t13 = icmp slt i32 %t2, 99
-  br i1 %t13, label %bb1, label %bb14
+  %tmp12 = add nuw nsw i32 %tmp2, 1
+  %tmp13 = icmp slt i32 %tmp2, 99
+  br i1 %tmp13, label %bb1, label %bb14
 
 bb14:                                             ; preds = %bb11
+; PROMO-LABEL: bb14:
   tail call void @bar(i32 0)
   br label %bb15
-
+; PROMO:  %pgocount.promoted{{.*}} = load {{.*}} @__profc_foo{{.*}} 0)
+; PROMO-NEXT: add 
+; PROMO-NEXT: store {{.*}}@__profc_foo{{.*}}0)
+; PROMO-NEXT:  %pgocount.promoted{{.*}} = load {{.*}} @__profc_foo{{.*}} 1)
+; PROMO-NEXT: add 
+; PROMO-NEXT: store {{.*}}@__profc_foo{{.*}}1)
+; PROMO-NEXT:  %pgocount.promoted{{.*}} = load {{.*}} @__profc_foo{{.*}} 2)
+; PROMO-NEXT: add 
+; PROMO-NEXT: store {{.*}}@__profc_foo{{.*}}2)
+; PROMO-NEXT:  %pgocount{{.*}} = load {{.*}} @__profc_foo{{.*}} 3)
+; PROMO-NEXT: add 
+; PROMO-NEXT: store {{.*}}@__profc_foo{{.*}}3)
 
 
 bb15_0:                                             ; preds = %bb11
+; PROMO-LABEL: bb15_0:
   br label %bb15
+; PROMO:  %pgocount.promoted{{.*}} = load {{.*}} @__profc_foo{{.*}} 0)
+; PROMO-NEXT: add 
+; PROMO-NEXT: store {{.*}}@__profc_foo{{.*}}0)
+; PROMO-NEXT:  %pgocount.promoted{{.*}} = load {{.*}} @__profc_foo{{.*}} 1)
+; PROMO-NEXT: add 
+; PROMO-NEXT: store {{.*}}@__profc_foo{{.*}}1)
+; PROMO-NEXT:  %pgocount.promoted{{.*}} = load {{.*}} @__profc_foo{{.*}} 2)
+; PROMO-NEXT: add 
+; PROMO-NEXT: store {{.*}}@__profc_foo{{.*}}2)
+; PROMO-NEXT:  %pgocount{{.*}} = load {{.*}} @__profc_foo{{.*}} 4)
+; PROMO-NEXT: add 
+; PROMO-NEXT: store {{.*}}@__profc_foo{{.*}}4)
+; PROMO-NOT: @__profc_foo
+
 
 bb15:                                             ; preds = %bb14, %bb4
   tail call void @bar(i32 1)

diff --git a/llvm/test/Transforms/PGOProfile/landingpad.ll b/llvm/test/Transforms/PGOProfile/landingpad.ll
@@ -85,8 +85,7 @@ catch.dispatch:
 ; GEN: catch.dispatch:
 ; GEN-NOT: call void @llvm.instrprof.increment
   %tmp3 = call i32 @llvm.eh.typeid.for(i8* bitcast (i8** @_ZTIi to i8*))
-  %c = icmp ne i32 %tmp2, %tmp3
-  %matches = xor i1 %c, -1
+  %matches = icmp eq i32 %tmp2, %tmp3
   br i1 %matches, label %catch, label %eh.resume
 ; USE: br i1 %matches, label %catch, label %eh.resume
 ; USE-SAME: !prof ![[BW_CATCH_DISPATCH:[0-9]+]]