[PATCH] D63060: [InstCombine] usub.sat(a, b) + b => umax(a, b) (PR42178)

[Jon Roelofs via Phabricator via llvm-commits]

jroelofs added a comment.

I was *just* thinking about the inverse of this transformation as a way to increase available ILP on CPUs where there are more pipes that can do arithmetic than compares.... cool!



================
Comment at: llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp:1428
 
+  // usub.sat(A, B) + A => umax(A, B)
+  if (match(&I, m_c_BinOp(
----------------
Comment doesn't seem to match what the code is doing.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D63060/new/

https://reviews.llvm.org/D63060