[PATCH] D50480: [LV] Vectorizing loops of arbitrary trip count without remainder under opt for size

[Ayal Zaks via Phabricator via llvm-commits]

Ayal created this revision.
Ayal added reviewers: mkuper, hsaito, dcaballe, fhahn, rengolin, hfinkel.
Herald added subscribers: llvm-commits, rogfer01.

When optimizing for size, a loop is vectorized only if the resulting vector loop completely replaces the original scalar loop. This holds if no runtime guards are needed, if the original trip-count TC does not overflow, if TC is a known constant and if TC is a multiple of the VF. Targets with efficient vector masking can thereby overcome the last three TC-related conditions: see “Direction #1” in [[ http://lists.llvm.org/pipermail/llvm-dev/2018-August/125042.html | [llvm-dev] Vectorizing remainder loop ]] - this patch applies that transformation of setting the trip-count of the vector loop to be TC rounded-up to a multiple of VF while masking the vector body under a newly introduced "if (i < TC)" condition; or rather "if (i <= TC-1)" to overcome the aforementioned overflow hazard.

The patch allows loops with arbitrary trip counts to be vectorized under -Os, subject to the existing cost model considerations. It also applies to loops with small trip counts (under -O2) which are currently handled as if under -Os.

Handling loops with reductions and live-outs are marked as TODOs for subsequent extensions.


Repository:
  rL LLVM

https://reviews.llvm.org/D50480

Files:
  include/llvm/Transforms/Vectorize/LoopVectorizationLegality.h
  lib/Transforms/Vectorize/LoopVectorizationLegality.cpp
  lib/Transforms/Vectorize/LoopVectorize.cpp
  lib/Transforms/Vectorize/VPlan.cpp
  lib/Transforms/Vectorize/VPlan.h
  test/Transforms/LoopVectorize/X86/optsize.ll
  test/Transforms/LoopVectorize/X86/small-size.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D50480.159798.patch
Type: text/x-patch
Size: 23032 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180808/e23de011/attachment.bin>