[PATCH] D137913: [X86] Rewrite `getScalarizationOverhead()`

[Roman Lebedev via Phabricator via llvm-commits]

lebedev.ri marked an inline comment as not done.
lebedev.ri added inline comments.


================
Comment at: llvm/lib/Target/X86/X86TargetTransformInfo.cpp:4440
+          APInt LaneEltMask = WidenedDemandedElts.extractBits(Scale, Scale * I);
+          unsigned Population = LaneEltMask.countPopulation();
+          bool SubvectorIsAffected = Population > 0;
----------------
lebedev.ri wrote:
> RKSimon wrote:
> > Do we actually need to do a costly popcnt? Are we doing anything other than checking for zero / allones?
> This whole logic is structured around the fact that we might only insert *some* elements.
Err, actually, right, we don't. Let me fix that...


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D137913/new/

https://reviews.llvm.org/D137913