[PATCH] LLVM targeting aarch64 doesn't correctly produce aligned accesses for non-aligned data at -O0/fast-isel (-mno-unaligned-access)

[Chad Rosier]

Please address the comments about the test case and then commit.


REPOSITORY
  rL LLVM

================
Comment at: test/CodeGen/AArch64/arm64-strict-align.ll:4
@@ -3,2 +3,3 @@
 ; RUN: llc < %s -mtriple=arm64-apple-darwin -aarch64-strict-align | FileCheck %s --check-prefix=CHECK-STRICT
+; RUN: llc < %s -mtriple=arm64-apple-darwin -aarch64-strict-align -fast-isel | FileCheck %s --check-prefix=CHECK-STRICT-FASTISEL
 
----------------
Please use the "CHECK-STRICT" prefix; the output is exactly the same for fast-isel and non-fast-isel.  No need for the redundancy.

http://reviews.llvm.org/D10360

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/