[PATCH] D132477: Improve cost model for some 128-bit vector operations that use SVE

[David Sherwood via Phabricator via llvm-commits]

david-arm added inline comments.


================
Comment at: llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp:2047
+          Cost *= 4;
+      } else {
+        // On AArch64, vector divisions are not supported natively and are
----------------
I just realised that we also need to multiply by the legalisation cost here too, i.e.

 Cost *= LT.first

so that we calculate the costs correctly for types such as <8 x i64>, etc.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D132477/new/

https://reviews.llvm.org/D132477