[PATCH] D41522: [Inliner] Restrict soft-float inlining penalty.

[Eli Friedman via Phabricator via llvm-commits]

efriedma created this revision.
efriedma added reviewers: haicheng, chandlerc, mcrosier.
Herald added a subscriber: eraman.

The penalty is currently getting applied in a bunch of places where it doesn't make sense, like bitcasts (which are free) and calls (which were getting the call penalty applied twice). Instead, just apply the penalty to binary operators and floating-point casts.

(Not sure if I should also apply this to fcmp instructions.)


Repository:
  rL LLVM

https://reviews.llvm.org/D41522

Files:
  lib/Analysis/InlineCost.cpp
  test/Transforms/Inline/inline-fp.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D41522.127945.patch
Type: text/x-patch
Size: 8761 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20171221/de61ef21/attachment.bin>