[PATCH] D61430: AMDGPU: Remove redundant patterns for shifts

[Matt Arsenault via Phabricator via llvm-commits]

arsenm marked an inline comment as done.
arsenm added inline comments.


================
Comment at: lib/Target/AMDGPU/VOP2Instructions.td:522
 
-let AddedComplexity = 1 in {
-  def : DivergentBinOp<srl, V_LSHRREV_B32_e64>;
----------------
rampitec wrote:
> Did you find out why added complexity was used?
I assume this was to give priority to these over the _e32 ones for some reason. Nothing broke when I removed either set


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D61430/new/

https://reviews.llvm.org/D61430