[PATCH] D76515: [ARM] Fix incorrect handling of big-endian vmov.i64

[John Brawn via Phabricator via llvm-commits]

john.brawn created this revision.
john.brawn added reviewers: dmgreen, simon_tatham, SjoerdMeijer.
Herald added subscribers: hiraditya, kristof.beyls.
Herald added a project: LLVM.
john.brawn added a parent revision: D76514: [ARM] Avoid pointless vrev of element-wise vmov.

Currently when the target is big-endian vmov.i64 reverses the order of the two words of the vector. This is correct only when the underlying element type is 32-bit, as actually what it should be doing is considering it a vector of the underlying type and reversing the elements of that.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D76515

Files:
  llvm/lib/Target/ARM/ARMISelLowering.cpp
  llvm/test/CodeGen/ARM/big-endian-neon-fp16-bitconv.ll
  llvm/test/CodeGen/ARM/big-endian-vmov.ll
  llvm/test/CodeGen/ARM/vmov.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D76515.251693.patch
Type: text/x-patch
Size: 8090 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200320/b4ba0803/attachment.bin>