[PATCH] D148999: [X86] X86FixupInstTunings - add VPERMILPDri -> VSHUFPDrri mapping

[Phoebe Wang via Phabricator via llvm-commits]

pengfei added inline comments.


================
Comment at: llvm/lib/Target/X86/X86FixupInstTuning.cpp:132
+  // `vshufpd` is always as fast or faster than `vpermilpd` and takes
+  // 1 less byte of code size for VEX and SSE encoding.
+  auto ProcessVPERMILPDri = [&](unsigned NewOpc) -> bool {
----------------
Nit: EVEX and VEX? There's no SSE vpermilpd. Applies to vpermilps too.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D148999/new/

https://reviews.llvm.org/D148999