[PATCH] D83764: DAG: Try scalarizing when expanding saturating add/sub

[Matt Arsenault via Phabricator via llvm-commits]

arsenm created this revision.
arsenm added reviewers: greened, nikic, RKSimon.
Herald added subscribers: dmgreen, hiraditya, tpr, wdng.
Herald added a project: LLVM.

In an upcoming AMDGPU patch, the scalar cases will be legal and vector
ops should be scalarized, rather than producing a long sequence of
vector ops which will also need to be scalarized.

      

Use a lazy heuristic that seems to work and improves the thumb2 MVE
test.


https://reviews.llvm.org/D83764

Files:
  llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
  llvm/test/CodeGen/Thumb2/mve-saturating-arith.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D83764.277793.patch
Type: text/x-patch
Size: 15129 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200714/7872fa71/attachment.bin>