[PATCH] D142682: AMDGPU: Combine down fcopysign f64 magnitude

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

In D142682#4084919 <https://reviews.llvm.org/D142682#4084919>, @foad wrote:

> Looks OK but does it actually improve codegen? It is hard to tell from the diffs.

This mostly improves the cases involving constants. The actual width reduction is effectively what the codegen already, so there's a narrow range of second order improvements available from other combines


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D142682/new/

https://reviews.llvm.org/D142682