[all-commits] [llvm/llvm-project] 7f1525: [LV] Apply sink-after & interleave-groups as VPlan...

[Gil Rapaport via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 7f152543e4ff91f2ce8578a1b3b3ec18f580345a
      https://github.com/llvm/llvm-project/commit/7f152543e4ff91f2ce8578a1b3b3ec18f580345a
  Author: Gil Rapaport <gil.rapaport at intel.com>
  Date:   2019-11-09 (Sat, 09 Nov 2019)

  Changed paths:
    M llvm/include/llvm/Analysis/VectorUtils.h
    M llvm/lib/Transforms/Vectorize/LoopVectorizationPlanner.h
    M llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
    M llvm/lib/Transforms/Vectorize/VPRecipeBuilder.h
    M llvm/lib/Transforms/Vectorize/VPlan.cpp
    M llvm/lib/Transforms/Vectorize/VPlan.h
    M llvm/test/Transforms/LoopVectorize/first-order-recurrence.ll
    A llvm/test/Transforms/LoopVectorize/interleaved-accesses-uniform-load.ll
    M llvm/unittests/Transforms/Vectorize/VPlanTest.cpp

  Log Message:
  -----------
  [LV] Apply sink-after & interleave-groups as VPlan transformations (NFCI)

This recommits 11ed1c0239fd51fd2f064311dc7725277ed0a994 (reverted in
9f08ce0d2197d4f163dfa4633eae2347ce8fc881 for failing an assert) with a fix:
tryToWidenMemory() now first checks if the widening decision is to interleave,
thus maintaining previous behavior where tryToInterleaveMemory() was called
first, giving priority to interleave decisions over widening/scalarization. This
commit adds the test case that exposed this bug as a LIT.