[PATCH] D55601: [bugpoint] Add ability to reduce to zero elements

[Philip Reames via Phabricator via llvm-commits]

reames added a comment.

A question.  Why do we need a 1 limit at all?  Wouldn't reducing any list to zero when 1 element is required to produce the failure fail to reduce?  I'm wondering if what we want is essentially a base case for the reducer that tries removing the last element, and if it fails, backing off to the single element case.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D55601/new/

https://reviews.llvm.org/D55601