[PATCH] D32596: [DAGCombine] Transform (fmul X, -2.0) --> (fneg (fadd X, X)).

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

In https://reviews.llvm.org/D32596#743761, @arsenm wrote:

> This is worse for AMDGPU because it requires a larger instruction encoding for f16/f32. For f64 this is better


Specifically if the user has a neg source modifier. Otherwise it is always worse


https://reviews.llvm.org/D32596