[PATCH] D59006: [x86] improve the default expansion of uaddsat/usubsat

[Sanjay Patel via Phabricator via llvm-commits]

spatel updated this revision to Diff 191595.
spatel added a comment.

Patch updated:
We improved the generic expansion slightly with D59066 <https://reviews.llvm.org/D59066>. That leaves customization for x86 which is required because umin/umax are custom lowered even if we don't actually have the instructions pmaxud/pmaxuq. That's not a generic lowering problem; that's an x86 problem.

In the earlier draft, I had made this a combine, but that seems pretty clearly wrong. We're just custom lowering a few specific vector types. Test changes look pretty close what we had before.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D59006/new/

https://reviews.llvm.org/D59006

Files:
  llvm/lib/Target/X86/X86ISelLowering.cpp
  llvm/test/CodeGen/X86/uadd_sat.ll
  llvm/test/CodeGen/X86/uadd_sat_vec.ll
  llvm/test/CodeGen/X86/usub_sat.ll
  llvm/test/CodeGen/X86/usub_sat_vec.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D59006.191595.patch
Type: text/x-patch
Size: 79510 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190320/35e69f57/attachment.bin>