[PATCH] D50167: RFC: [SCEV] Add explicit representations of umin/smin

[Keno Fischer via Phabricator via llvm-commits]

loladiro added inline comments.


================
Comment at: lib/Analysis/ScalarEvolution.cpp:10137
+        // min(A, ...) <= A
+        IsMinMaxConsistingOfByNegation<SCEVSMaxExpr>(SE, LHS, RHS) ||
+        IsMinMaxConsistingOf<SCEVSMinExpr>(LHS, RHS) ||
----------------
sanjoy wrote:
> Does this case even fire anymore?
I wanted to avoid regressing the (probably not very likely, but possible) case that somebody might construct `~umax(~x, ~y)` manually. However, in retrospect the better way to do that is probably to just pattern match that to `umin` on construction. Let me try that and remove this case.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D50167/new/

https://reviews.llvm.org/D50167