[PATCH] Disable loop unrolling in loop vectorization pass when VF is 1 on x86

[Xinliang David Li]

Or fix this in stages -- x86 first and other archs later when more data is
available (help with testing on other archs is needed).

David

On Wed, May 6, 2015 at 2:37 AM, Renato Golin <renato.golin at linaro.org>
wrote:

> Hi Wei,
>
> The example you have shown would produce bad vectorized code on any
> architecture, I don't think anything you said (multiple unrolling and
> prologue loops) would make much difference on other archs. Maybe you're
> trying to fix a global problem locally, and creating some unnecessary
> constraints for the cases that do work.
>
> However, your performance improvements are really impressive, so I think
> we ought to check other archs, and maybe try to detect the problematic case
> on a generic level?
>
> cheers,
> --renato
>
>
> REPOSITORY
>   rL LLVM
>
> http://reviews.llvm.org/D9515
>
> EMAIL PREFERENCES
>   http://reviews.llvm.org/settings/panel/emailpreferences/
>
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150506/1fa07927/attachment.html>