[PATCH] D90070: [CostModel] Add optional VecPred argument to getCmpSelInstrCost.

[Dave Green via Phabricator via llvm-commits]

dmgreen added a comment.

AArch64 costs sound OK, especially if we can fix NE codegen to be more likely to be a single instruction.

I would personally keep the args in (Opcode, Types, CondCode, CostKind, CtxI) order, to keep them more consistent with the other cost functions. Now that I look at them though I don't think they are as consistent as I believed they were!



================
Comment at: llvm/test/Analysis/CostModel/AArch64/vector-select.ll:43
 ; CODE-NEXT:    cmeq  v{{.+}}.4h, v{{.+}}.4h, v{{.+}}.4h
 ; CODE-NEXT:    mvn   v{{.+}}.8b, v{{.+}}.8b
 ; CODE-NEXT:    bif   v{{.+}}.8b, v{{.+}}.8b, v{{.+}}.8b
----------------
Huh.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D90070/new/

https://reviews.llvm.org/D90070