[llvm-bugs] [Bug 24421] Performance of unaligned NEON loads via intrinsics is poor

[via llvm-bugs]

https://llvm.org/bugs/show_bug.cgi?id=24421

Eli Friedman <efriedma at codeaurora.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |efriedma at codeaurora.org
         Resolution|---                         |FIXED

--- Comment #5 from Eli Friedman <efriedma at codeaurora.org> ---
Trunk clang generates the expected code given the "__attribute__((aligned(1)))"
annotation.  If you want to avoid gcc extensions, you can also memcpy to a
local uint32_t... but that doesn't optimize well using gcc.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20161208/77d92b69/attachment-0001.html>