Files
archived-llvm/test/Transforms/LoopVectorize
Jonas Paulsson ccd4f446eb [LoopVectorizer] Fix for cost values of memory accesses.
This commit is a combination of two patches:

* "Fix in getScalarizationOverhead()"

   If target returns false in TTI.prefersVectorizedAddressing(), it means the
   address registers will not need to be extracted. Therefore, there should
   be no operands scalarization overhead for a load instruction.

* "Don't pass the instruction pointer from getMemInstScalarizationCost."

   Since VF is always > 1, this is a cost query for an instruction in the
   vectorized loop and it should not be evaluated within the scalar
   context of the instruction.

Review: Ulrich Weigand, Hal Finkel
https://reviews.llvm.org/D52351
https://reviews.llvm.org/D52417

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@345603 91177308-0d34-0410-b5e6-96231b3b80d8
2018-10-30 14:34:15 +00:00
..
2018-04-13 20:16:32 +00:00
2018-02-05 08:09:49 +00:00
2018-03-09 18:02:36 +00:00
2016-06-29 10:01:06 +00:00