[PATCH] D141870: [DAG] Fold fadd(vecreduce(a), vecreduce(b)) into vecreduce(fadd(a,b))

[Sanjay Patel via Phabricator via llvm-commits]

spatel added a comment.

We probably want this for all associate reductions - fmul, min/max, integer ops, etc. Generalize into a helper function that takes an opcode and maps it to the corresponding reduction opcode? Subsequent patches then just need to add set of tests and another case into a switch or something like that.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D141870/new/

https://reviews.llvm.org/D141870