[PATCH] D75157: [DAGCombine] Perform the fold of A - (-B) -> A + B only when it is cheaper

[Roman Lebedev via Phabricator via llvm-commits]

lebedev.ri added a comment.

In D75157#1892994 <https://reviews.llvm.org/D75157#1892994>, @steven.zhang wrote:

> In D75157#1892866 <https://reviews.llvm.org/D75157#1892866>, @lebedev.ri wrote:
>
> > The test changes don't immediately stand out to me as improvements.
> >  I'd think the fix should go in other direction.
>
>
> I want to make them consistent to avoid the confusion no matter which direction.




In D75157#1893099 <https://reviews.llvm.org/D75157#1893099>, @RKSimon wrote:

> Loss of commutativity by going from fadd to fsub will likely cause register pressure regressions.


I'm not very active wrt fp side of things, but i would almost think we need something like this instead

  // unfold (fsub A, B) -> (fadd A, (fneg B))
  if ((!LegalOperations || TLI.isOperationLegalOrCustom(ISD::FADD, VT)) &&
      TLI.getNegatibleCost(N1, DAG, LegalOperations, ForCodeSize) !=
          TargetLowering::NegatibleCost::Expensive)
    return DAG.getNode(
        ISD::FADD, DL, VT, N0,
        TLI.getNegatedExpression(N1, DAG, LegalOperations, ForCodeSize), Flags);


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D75157/new/

https://reviews.llvm.org/D75157