[PATCH] D25963: [LoopUnroll] Implement profile-based loop peeling

Xinliang David Li via llvm-commits llvm-commits at lists.llvm.org
Thu Oct 27 13:23:15 PDT 2016


On Thu, Oct 27, 2016 at 1:17 PM, Gerolf Hoflehner <ghoflehner at apple.com>
wrote:

>
> On Oct 26, 2016, at 12:39 PM, Xinliang David Li <davidxl at google.com>
> wrote:
>
>
>
> On Wed, Oct 26, 2016 at 10:08 AM, Gerolf Hoflehner <ghoflehner at apple.com>
> wrote:
>
>> Gerolf added a comment.
>>
>> Hi,
>>
>> Could you provide more background on this idea?  What is your
>> motivational use case? When the trip count is low why optimize?
>
>
> Low trip count loop can be hot too.
>
>
> But then you would optimize differently. Can you share the example where
> you actually see gains?
>

See Michael's examples - typical ones are short trip count inner loops in a
outer loop with high trip count.

David


>
> Thanks
> Gerolf
>
>
> David
>
>
>
>> If the profile is wrong and it actually is a hot loop for a
>> regular/different input set peeling could hurt. There are also side effects
>> on code size, register pressure etc. that could hurt performance.
>>
>> Thanks
>> Gerolf
>>
>>
>> https://reviews.llvm.org/D25963
>>
>>
>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20161027/e9636667/attachment.html>


More information about the llvm-commits mailing list