[LLVMdev] the clang 3.5 loop optimizer seems to jump in unintentional for simple loops

[Chandler Carruth]

On Tue, Jul 22, 2014 at 8:51 PM, Dennis Luehring <dl.soluz at gmx.net> wrote:

> with clang3.5 + #define ITER the_func contains masses of code
> the code in main is also sometimes different (not just inlined) to the_func
>

It looks like it has vectorized the code which is usually good? I'm
actually surprised it failed to vectorize when using the integer indices.

Did you benchmark the performance?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20140722/fad33969/attachment.html>