[PATCH] D56011: [x86] lower extracted fadd/fsub to horizontal vector math

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon requested changes to this revision.
RKSimon added a comment.
This revision now requires changes to proceed.

In D56011#1345978 <https://reviews.llvm.org/D56011#1345978>, @spatel wrote:

> Patch updated:
>  Include cost model and SLP test changes. Should we be overriding the cost of scalar FADD/FSUB, so we don't have these diffs?


Yes - please add the relevant cost changes so the slp/cost test changes go away.

> Without that, it looks like we fall back to this questionable logic:
> 
>   // If the operation is custom lowered, then assume that the code is twice
>   // as expensive.

That is some VERY questionable logic.....


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D56011/new/

https://reviews.llvm.org/D56011