[llvm-commits] [llvm] r89865 - /llvm/trunk/lib/Target/X86/X86InstrInfo.h

[Bob Wilson]

Author: bwilson
Date: Wed Nov 25 11:27:53 2009
New Revision: 89865

URL: http://llvm.org/viewvc/llvm-project?rev=89865&view=rev
Log:
Based on the testcase for pr3120, running on my MacPro with Xeon processors,
it is definitely profitable to tail duplicate indirect branches for x86.
This is likely to be true to various degrees for all modern x86 processors.

Modified:
    llvm/trunk/lib/Target/X86/X86InstrInfo.h

Modified: llvm/trunk/lib/Target/X86/X86InstrInfo.h
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86InstrInfo.h?rev=89865&r1=89864&r2=89865&view=diff

==============================================================================
--- llvm/trunk/lib/Target/X86/X86InstrInfo.h (original)
+++ llvm/trunk/lib/Target/X86/X86InstrInfo.h Wed Nov 25 11:27:53 2009
@@ -632,6 +632,8 @@
   ///
   unsigned getGlobalBaseReg(MachineFunction *MF) const;
 
+  virtual bool isProfitableToDuplicateIndirectBranch() const { return true; }
+
 private:
   MachineInstr* foldMemoryOperandImpl(MachineFunction &MF,
                                      MachineInstr* MI,