[PATCH] D36113: [Loop Vectorize] Vectorize Loops with Backward Dependence

[Daniel Berlin via llvm-commits]

I'm definitely with Hal on the second point. We alreayd know there are
deficiencies there, and the right way is to improve LICM.
If the goal is "move loop invariants", we want one pass that does that, not
6.

It's fine to have multiple passes when the job is so large that it can't
sanely be done optimally by one thing, but i don't think this is ojne of
those cases ....


On Tue, Aug 22, 2017 at 7:11 PM, Hal Finkel via Phabricator <
reviews at reviews.llvm.org> wrote:

> hfinkel added a comment.
>
> Two high-level thoughts:
>
> 1. Is this transformation always profitable even if we don't later
> vectorize? I worry that you're taking a stream of accesses that generally
> appear in order and making them appear out of order. That could have odd
> effects. Maybe this should all be a utility that the vectorizer uses more
> directly?
> 2. I don't think a goal of this should be to provide backup handling of
> things that LICM misses but MemorySSA can prove are loop invariant. We
> should improve LICM by making it use MemorySSA (see work on this in
> https://reviews.llvm.org/D35741 - pushing this along might help?). We
> could run LICM right before we vectorize too.
>
>
> https://reviews.llvm.org/D36113
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170822/d031e61c/attachment.html>