| commit | c04d241d13820b33224b5cbd89a427fc08e5d1d9 | [log] [tgz] |
|---|---|---|
| author | Arnold Schwaighofer <aschwaighofer@apple.com> | Tue Oct 29 01:33:57 2013 +0000 |
| committer | Arnold Schwaighofer <aschwaighofer@apple.com> | Tue Oct 29 01:33:57 2013 +0000 |
| tree | dbf40b8b49ab3278f68615ad0af8db89758d9f07 | |
| parent | 7e8cebf22d170769b0bf0c2a69309faa0e36ac4c [diff] |
ARM cost model: Unaligned vectorized double stores are expensive Updated a test case that assumed that <2 x double> would vectorize to use <4 x float>. radar://15338229 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193574 91177308-0d34-0410-b5e6-96231b3b80d8