For the current Atom processor, the fastest way to handle a call
indirect through a memory address is to load the memory address into
a register and then call indirect through the register.
This patch implements this improvement by modifying SelectionDAG to
force a function address which is a memory reference to be loaded
into a virtual register.
Patch by Sriram Murali.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@178171 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/lib/Target/X86/X86Subtarget.cpp b/lib/Target/X86/X86Subtarget.cpp
index 4b368b4..6e66c1a 100644
--- a/lib/Target/X86/X86Subtarget.cpp
+++ b/lib/Target/X86/X86Subtarget.cpp
@@ -458,6 +458,7 @@
HasSlowDivide = false;
PostRAScheduler = false;
PadShortFunctions = false;
+ CallRegIndirect = false;
stackAlignment = 4;
// FIXME: this is a known good value for Yonah. How about others?
MaxInlineSizeThreshold = 128;