[PATCH] D36792: [AArch64] v8.3-a complex number support

[Sjoerd Meijer via Phabricator via llvm-commits]

SjoerdMeijer accepted this revision.
SjoerdMeijer added a comment.
This revision is now accepted and ready to land.

Hi Sam, many thanks for refactoring this. Looks really good now! Just 2 nits inlined, but no need for another review. 
Cheers, Sjoerd.



================
Comment at: lib/Target/AArch64/AArch64InstrFormats.td:9426
+                                          string asm, SDPatternOperator OpNode>{
+  let Predicates = [HasV8_3a,HasNEON,HasFullFP16] in {
+  def v4f16 : BaseSIMDThreeSameVectorComplex<0, U, 0b010, opcode, V64, rottype,
----------------
Nit: missing space after commas (also in the other predicates below)


================
Comment at: lib/Target/AArch64/AArch64InstrInfo.td:462
+// v8.3a complex add and multiply-accumulate. No predicate here, that is done
+// inside the multiclass as the F16 versions need different predicates.
+defm FCMLA : SIMDThreeSameVectorTiedComplexHSD<1, 0b110, complexrotateop,
----------------
Nit: F16 => FP16?


https://reviews.llvm.org/D36792