[PATCH] D57059: [SLP] Initial support for the vectorization of the non-power-of-2 vectors.

[Nikita Popov via Phabricator via llvm-commits]

nikic added a comment.

In D57059#2442527 <https://reviews.llvm.org/D57059#2442527>, @anton-afanasyev wrote:

> In D57059#2442286 <https://reviews.llvm.org/D57059#2442286>, @RKSimon wrote:
>
>> AFAICT the only outstanding question is whether the compile time increase is acceptable?
>
> I'd agree that geomean = 0.2% is acceptable for the change with such awesome performance impact, just noted that changed time compilation is significant in comparision with other changes. Generally it looks good to me apart from one minor unaddressed comment.

Could the summary of the revision be updated with the performance data? The thread is very long and I didn't spot where the measurements are, so it's hard to say what we're trading off here...

Generally though this level of geomean regression is fine, the issue is usually in the outliers. For example, I spot this this 4-5% regression:

  CMakeFiles/lencod.dir/transform8x8.c.o 	3053M 	3188M (+4.41%)

It may be worthwhile to briefly check it in case something can be improved.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D57059/new/

https://reviews.llvm.org/D57059