[PATCH] D68019: [ARM][CGP] Allow signext arguments

[Eli Friedman via Phabricator via llvm-commits]

efriedma added a comment.

Obviously ignoring the attribute is correct; signext is just an ABI attribute which has no semantic effect at the IR level.  But why was the code to check it added in https://reviews.llvm.org/D61381?  I assume it was supposed to be some sort of profitability check.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D68019/new/

https://reviews.llvm.org/D68019