[PATCH] D67344: [ARM][MVE] VCTP instruction selection

[Dave Green via Phabricator via llvm-commits]

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

LGTM



================
Comment at: test/CodeGen/Thumb2/mve-vctp.ll:14
+  %pred = call <16 x i1> @llvm.arm.vctp8(i32 %arg)
+  %ld = load <16 x i8>, <16 x i8>* %in
+  %res = select <16 x i1> %pred, <16 x i8> %ld, <16 x i8> zeroinitializer
----------------
If you add arm_aapcs_vfpcc, you can pass the <16 x i8> vectors sanely are args/return values, if you want to remove the extra vldrs and vstrs.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D67344/new/

https://reviews.llvm.org/D67344