[PATCH] D68337: [ARM][MVE] Enable extending masked loads

[Sam Parker via Phabricator via llvm-commits]

samparker created this revision.
samparker added reviewers: dmgreen, efriedma, SjoerdMeijer.
Herald added a subscriber: kristof.beyls.

Allow us to generate sext/zext masked loads which can access v4i8, v8i8 and v4i16 memory to produce v4i32, v8i16 and v4i32 respectively.

My little (endian) brain only really works in that mode, so I'm dubious about the big endian support here.


https://reviews.llvm.org/D68337

Files:
  lib/Target/ARM/ARMISelLowering.cpp
  lib/Target/ARM/ARMInstrMVE.td
  lib/Target/ARM/ARMTargetTransformInfo.cpp
  test/CodeGen/Thumb2/LowOverheadLoops/mve-tail-data-types.ll
  test/CodeGen/Thumb2/mve-masked-ldst.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D68337.222830.patch
Type: text/x-patch
Size: 63078 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20191002/45725803/attachment.bin>