Teach DAGCombine to fold constant offsets into GlobalAddress nodes,

and add a TargetLowering hook for it to use to determine when this
is legal (i.e. not in PIC mode, etc.)

This allows instruction selection to emit folded constant offsets
in more cases, such as the included testcase, eliminating the need
for explicit arithmetic instructions.

This eliminates the need for the C++ code in X86ISelDAGToDAG.cpp
that attempted to achieve the same effect, but wasn't as effective.

Also, fix handling of offsets in GlobalAddressSDNodes in several
places, including changing GlobalAddressSDNode's offset from
int to int64_t.

The Mips, Alpha, Sparc, and CellSPU targets appear to be
unaware of GlobalAddress offsets currently, so set the hook to
false on those targets.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57748 91177308-0d34-0410-b5e6-96231b3b80d8

2 files changed, 8 insertions, 0 deletions

diff --git a/lib/Target/Sparc/SparcISelLowering.cpp b/lib/Target/Sparc/SparcISelLowering.cpp
index e81688e4e8..59e19d5466 100644
--- a/lib/Target/Sparc/SparcISelLowering.cpp
+++ b/lib/Target/Sparc/SparcISelLowering.cpp
@@ -996,3 +996,9 @@ getRegClassForInlineAsmConstraint(const std::string &Constraint,
 
   return std::vector<unsigned>();
 }
+
+bool
+SparcTargetLowering::isOffsetFoldingLegal(const GlobalAddressSDNode *GA) const {
+  // The Sparc target isn't yet aware of offsets.
+  return false;
+}
diff --git a/lib/Target/Sparc/SparcISelLowering.h b/lib/Target/Sparc/SparcISelLowering.h
index ea3171d027..a0e3c65344 100644
--- a/lib/Target/Sparc/SparcISelLowering.h
+++ b/lib/Target/Sparc/SparcISelLowering.h
@@ -70,6 +70,8 @@ namespace llvm {
     std::vector<unsigned>
     getRegClassForInlineAsmConstraint(const std::string &Constraint,
                                       MVT VT) const;
+
+    virtual bool isOffsetFoldingLegal(const GlobalAddressSDNode *GA) const;
   };
 } // end namespace llvm