[X86] Fix emitEpilogue() to make less assumptions about pops

This is the mirror image of r242395.
When X86FrameLowering::emitEpilogue() looks for where to insert the %esp addition that
deallocates stack space used for local allocations, it assumes that any sequence of pop
instructions from function exit backwards consists purely of restoring callee-save registers.

This may be false, since from some point backward, the pops may be clean-up of stack space
allocated for arguments to a call.

Patch by: amjad.aboud@intel.com
Differential Revision: http://reviews.llvm.org/D12688

llvm-svn: 247784
diff --git a/llvm/lib/CodeGen/MachineInstr.cpp b/llvm/lib/CodeGen/MachineInstr.cpp
index 036965b..f2984b2 100644
--- a/llvm/lib/CodeGen/MachineInstr.cpp
+++ b/llvm/lib/CodeGen/MachineInstr.cpp
@@ -1735,13 +1735,16 @@
   }
 
   bool HaveSemi = false;
-  const unsigned PrintableFlags = FrameSetup;
+  const unsigned PrintableFlags = FrameSetup | FrameDestroy;
   if (Flags & PrintableFlags) {
     if (!HaveSemi) OS << ";"; HaveSemi = true;
     OS << " flags: ";
 
     if (Flags & FrameSetup)
       OS << "FrameSetup";
+
+    if (Flags & FrameDestroy)
+      OS << "FrameDestroy";
   }
 
   if (!memoperands_empty()) {
diff --git a/llvm/lib/Target/X86/X86FrameLowering.cpp b/llvm/lib/Target/X86/X86FrameLowering.cpp
index 9e09810..edc4114 100644
--- a/llvm/lib/Target/X86/X86FrameLowering.cpp
+++ b/llvm/lib/Target/X86/X86FrameLowering.cpp
@@ -286,6 +286,8 @@
           .addReg(Reg, getDefRegState(!isSub) | getUndefRegState(isSub));
         if (isSub)
           MI->setFlag(MachineInstr::FrameSetup);
+        else
+          MI->setFlag(MachineInstr::FrameDestroy);
         Offset -= ThisVal;
         continue;
       }
@@ -295,6 +297,8 @@
         MBB, MBBI, DL, isSub ? -ThisVal : ThisVal, InEpilogue);
     if (isSub)
       MI.setMIFlag(MachineInstr::FrameSetup);
+    else
+      MI.setMIFlag(MachineInstr::FrameDestroy);
 
     Offset -= ThisVal;
   }
@@ -1075,7 +1079,7 @@
 
     // Pop EBP.
     BuildMI(MBB, MBBI, DL, TII.get(Is64Bit ? X86::POP64r : X86::POP32r),
-            MachineFramePtr);
+            MachineFramePtr).setMIFlag(MachineInstr::FrameDestroy);
   } else if (hasFP(MF)) {
     // Calculate required stack adjustment.
     uint64_t FrameSize = StackSize - SlotSize;
@@ -1088,7 +1092,8 @@
 
     // Pop EBP.
     BuildMI(MBB, MBBI, DL,
-            TII.get(Is64Bit ? X86::POP64r : X86::POP32r), MachineFramePtr);
+            TII.get(Is64Bit ? X86::POP64r : X86::POP32r), MachineFramePtr)
+        .setMIFlag(MachineInstr::FrameDestroy);
   } else {
     NumBytes = StackSize - CSSize;
   }
@@ -1099,8 +1104,9 @@
     MachineBasicBlock::iterator PI = std::prev(MBBI);
     unsigned Opc = PI->getOpcode();
 
-    if (Opc != X86::POP32r && Opc != X86::POP64r && Opc != X86::DBG_VALUE &&
-        !PI->isTerminator())
+    if ((Opc != X86::POP32r || !PI->getFlag(MachineInstr::FrameDestroy)) &&
+        (Opc != X86::POP64r || !PI->getFlag(MachineInstr::FrameDestroy)) &&
+        Opc != X86::DBG_VALUE && !PI->isTerminator())
       break;
 
     --MBBI;
@@ -1463,7 +1469,8 @@
         !X86::GR32RegClass.contains(Reg))
       continue;
 
-    BuildMI(MBB, MI, DL, TII.get(Opc), Reg);
+    BuildMI(MBB, MI, DL, TII.get(Opc), Reg)
+        .setMIFlag(MachineInstr::FrameDestroy);
   }
   return true;
 }