[llvm] [AArch64] Use brk{a, b} for a lane mask from cttz.elts (PR #178674)

[David Sherwood via llvm-commits]

================
@@ -0,0 +1,228 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc -mtriple=aarch64 -mattr=+sve < %s | FileCheck %s
+
+;; Scalable
+define <vscale x 16 x i1> @mask_exclude_active_nxv16(<vscale x 16 x i1> %mask.in) {
----------------
david-arm wrote:

I think it's worth having a couple of tests for wider types too, even though we will expand the cttz.elts intrinsic in those cases. For example, a test for <vscale x 32 x i1> and a test for <32 x i1>.

https://github.com/llvm/llvm-project/pull/178674