[LLVMbugs] [Bug 16716] char-copying loop should optimize to memcpy

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=16716

Richard Smith <richard-llvm at metafoo.co.uk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
                 CC|                            |richard-llvm at metafoo.co.uk
         Resolution|INVALID                     |---

--- Comment #3 from Richard Smith <richard-llvm at metafoo.co.uk> ---
(In reply to comment #2)
> Specifically, I'm saying that we can't idiom recognize the function based on
> its definition alone.  We can inline it into call sites and gain aliasing
> information from it, of course.

OK, but we're failing to do so in the first 'foo' function. This looks like a
legitimate bug, and seems to be a pass ordering issue: running 'opt -O2' on the
output of 'clang -O2' is enough to get the function to be optimized to a
bitcast.

-- 
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/20130727/e0dcb923/attachment.html>