[PATCH] D57348: [CodeGen][X86] Don't scalarize vector saturating add/sub

[Nikita Popov via Phabricator via llvm-commits]

nikic created this revision.
nikic added reviewers: RKSimon, spatel.
Herald added a subscriber: llvm-commits.

This matches what the cost model already does.

This relies on vector SADDO/SSUBO though. I think this works fine because those are expanded on all archs (for vectors), but is possibly not entirely reliable.


Repository:
  rL LLVM

https://reviews.llvm.org/D57348

Files:
  lib/CodeGen/SelectionDAG/TargetLowering.cpp
  test/CodeGen/X86/sadd_sat.ll
  test/CodeGen/X86/sadd_sat_vec.ll
  test/CodeGen/X86/ssub_sat.ll
  test/CodeGen/X86/ssub_sat_vec.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D57348.183928.patch
Type: text/x-patch
Size: 432219 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190128/4f4e4aef/attachment-0001.bin>