Add move vector to the integrated ARM assembler.

src/IceInstARM32.cpp

 //===- subzero/src/IceInstARM32.cpp - ARM32 instruction implementation ----===//
 //
 //                        The Subzero Code Generator
 //
 // This file is distributed under the University of Illinois Open Source
 // License. See LICENSE.TXT for details.
 //
 //===----------------------------------------------------------------------===//
 ///
 /// \file
@@ skipping 1548 matching lines @@
       }
     }
     break; // Error
   case IceType_v4i1:
   case IceType_v8i1:
   case IceType_v16i1:
   case IceType_v16i8:
   case IceType_v8i16:
   case IceType_v4i32:
   case IceType_v4f32:
-    // TODO(kschimpf): Add vector moves.
-    emitUsingTextFixup(Func);
+    assert(SrcTy == DestTy && "Mov on different vector types");
+    Asm->vorrq(Dest, Src0, Src0);

[Eric Holk, 2016/02/11 19:24:22]

The other moves in this function pass the result of getPredicate() to the
assembler. Do we need to handle conditional vector moves? We should probably at
least assert that it's an unconditional move in case some higher level code
tries to generate one.

[Karl, 2016/02/11 22:10:28]

The same can be said for the emit() method. Adding TODO to fix that case as
well!

I talked to John and (at least currently) we shouldn't be generating conditions
for these vector moves.

Done

     return;
   }
   llvm::report_fatal_error("Mov: don't know how to move " +
                            typeIceString(SrcTy) + " to " +
                            typeIceString(DestTy));
 }
 
 void InstARM32Mov::dump(const Cfg *Func) const {
   if (!BuildDefs::dump())
     return;
@@ skipping 1188 matching lines @@
 
 template class InstARM32FourAddrGPR<InstARM32::Mla>;
 template class InstARM32FourAddrGPR<InstARM32::Mls>;
 
 template class InstARM32CmpLike<InstARM32::Cmn>;
 template class InstARM32CmpLike<InstARM32::Cmp>;
 template class InstARM32CmpLike<InstARM32::Tst>;
 
 } // end of namespace ARM32
 } // end of namespace Ice