[PATCH] [ARM64] Adds Cortex-A53 scheduling support for vector load/store post.

[Andrew Trick]

LGTM

================
Comment at: test/CodeGen/ARM64/misched-basic-A53.ll:113
@@ +112,3 @@
+
+; Regression Test for Bug 19761
+; - [ARM64] Cortex-a53 schedule mode can't handle NEON post-increment load
----------------
Chad Rosier wrote:
> Lines 114-116 aren't necessary.  The git log should retain such history.
I'm not sure what the official stance is now, but I like extra comments in test cases, referring to the source change and bug numbers (you can just say PR19761: title).

We avoid bug numbers in source code, but I think it's ok in test cases.

Codegen test cases can be very long-lived and difficult to maintain. It speeds things up a lot to explain what we're testing and why.

http://reviews.llvm.org/D3829