[PATCH] D106679: [WIP][X86] Move shl(x, 1) -> add(x, x) vector fold to ISEL (PR50468)

[Craig Topper via Phabricator via llvm-commits]

craig.topper added inline comments.


================
Comment at: llvm/lib/Target/X86/X86InstrAVX512.td:5925
+// Prefer add to shl-by-one, on many targets, ADD is faster than SHL.
+let AddedComplexity = 400 in {
+  let Predicates = [HasAVX512] in {
----------------
Do we really need AddedComplexity? Especially as high as 400. The explicit immediate should automatically rank higher than any immediate.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D106679/new/

https://reviews.llvm.org/D106679