[BOLT] Enable tail duplication const/copy propagation

[Amir]

Summary

Re-enable const/copy propagation after tail duplication.
Add tail-dup-peephole.s XFAIL test triggered by taildup.

bolt/lib/Passes/TailDuplication.cpp

 static cl::opt<unsigned> TailDuplicationMaximumDuplication(
     "tail-duplication-maximum-duplication",
     cl::desc("maximum size of duplicated blocks (in bytes)"), cl::ZeroOrMore,
     cl::ReallyHidden, cl::init(64), cl::cat(BoltOptCategory));
 
 static cl::opt<bool> TailDuplicationConstCopyPropagation(
     "tail-duplication-const-copy-propagation",
     cl::desc("enable const and copy propagation after tail duplication"),
-    cl::ReallyHidden, cl::init(false), cl::cat(BoltOptCategory));
+    cl::ReallyHidden, cl::init(true), cl::cat(BoltOptCategory));
 
 } // namespace opts
 
 namespace llvm {
 namespace bolt {
 void TailDuplication::getCallerSavedRegs(const MCInst &Inst, BitVector &Regs,
                                          BinaryContext &BC) const {
   if (!BC.MIB->isCall(Inst))

bolt/test/X86/tail-dup-peephole.s

+# This reproduces a bug in double-jumps peephole triggered by tail duplication.
+# XFAIL: *
+# REQUIRES: system-linux
+# RUN: llvm-mc -filetype=obj -triple x86_64-unknown-unknown %s -o %t.o
+# RUN: link_fdata %s %t.o %t.fdata
+# RUN: llvm-strip --strip-unneeded %t.o
+# RUN: %clang %cflags %t.o -o %t.exe -Wl,-q
+# RUN: llvm-bolt %t.exe -o %t.out -data %t.fdata -relocs -lite=0 -funcs=main \
+# RUN:   -reorder-blocks=cache -peepholes=double-jumps -tail-duplication=1
+
+  .globl main
+main:
+  .cfi_startproc
+a:
+  je  b
+# FDATA: 1 main #a# 1 main #b# 1 2
+  movq  0, %r11
+  testq %r11, %r11
+c:
+  jmp d
+e:
+  je  f
+# FDATA: 1 main #e# 1 main #f# 1 4
+d:
+  cmpl  %esi, %edi
+  leaq  0, %rsi
+  movq  %r12, %rdx
+  movq  0, %rdi
+  leaq  0, %r15
+  addq  8, %r15
+  cmpq  %r15, %rdi
+  je  g
+  movq  8, %r13
+h:
+  jne g
+# FDATA: 1 main #h# 1 main #i# 0 3
+i:
+  leaq  0, %rdi
+f:
+  cmpw  $0xae, 0x20(%rdi)
+b:
+  jmp c
+g:
+  jmp h
+  .cfi_endproc