[PowerPC 3/4] Little-endian adjustments for VSX vector shuffle

When performing instruction selection for ISD::VECTOR_SHUFFLE, there
is special code for handling v2f64 and v2i64 using VSX instructions.
This code must be adjusted for little-endian.  Because the two inputs
are treated as a double-wide register, we must swap their order for
little endian.  To get the appropriate mask elements to use with the
big-endian biased XXPERMDI instruction, we must reverse their order
and invert the bits.

A new test is added to test the 16 possible values of the shuffle
mask.  It is initially disabled for reasons specified in the test.  It
is re-enabled by patch 4/4.

llvm-svn: 223791
diff --git a/llvm/lib/Target/PowerPC/PPCISelDAGToDAG.cpp b/llvm/lib/Target/PowerPC/PPCISelDAGToDAG.cpp
index e04d7e1..595052e 100644
--- a/llvm/lib/Target/PowerPC/PPCISelDAGToDAG.cpp
+++ b/llvm/lib/Target/PowerPC/PPCISelDAGToDAG.cpp
@@ -1380,6 +1380,15 @@
         else
           DM[i] = 1;
 
+      // For little endian, we must swap the input operands and adjust
+      // the mask elements (reverse and invert them).
+      if (PPCSubTarget->isLittleEndian()) {
+        std::swap(Op1, Op2);
+        unsigned tmp = DM[0];
+        DM[0] = 1 - DM[1];
+        DM[1] = 1 - tmp;
+      }
+
       SDValue DMV = CurDAG->getTargetConstant(DM[1] | (DM[0] << 1), MVT::i32);
 
       if (Op1 == Op2 && DM[0] == 0 && DM[1] == 0 &&