[PATCH] D142065: [SVE] Fix incorrect lowering of predicate permute builtins.

[Paul Walker via Phabricator via llvm-commits]

paulwalker-arm added inline comments.


================
Comment at: llvm/lib/Target/AArch64/SVEInstrFormats.td:6361
+  def : SVE_2_Op_Pat<nxv4i1, ir_op, nxv4i1, nxv4i1, !cast<Instruction>(NAME # _S)>;
+  def : SVE_2_Op_Pat<nxv2i1, ir_op, nxv2i1, nxv2i1, !cast<Instruction>(NAME # _D)>;
 }
----------------
peterwaller-arm wrote:
> Out of interest, is there a good reason to handle the nxv16 pattern case differently in the `I` multiclass args? Written this way at a glance it looks like it is missing.
My reasoning was the pattern within the instruction class is mandatory for the correct clang builtin support so I figured that should take priority.  That means extra patterns are only required for the unpacked cases, which are optional based on them having value during code generation.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D142065/new/

https://reviews.llvm.org/D142065