[llvm] [SelectionDAG] Faster generic expansion of llvm.pext and llvm.pdep (PR #227292)

[via llvm-commits]

================
@@ -0,0 +1,185 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
+; RUN: llc < %s -mtriple=aarch64 -mcpu=apple-m1 | FileCheck %s
----------------
mightsleep wrote:

llc doesn't derive the CPU from the triple (only arm64e defaults to apple-a12), so -mcpu is needed either way. It doesn't change the output, but I guess I could switch to -mtriple=arm64-apple-macosx -mcpu=apple-m1 like most Apple tests here.

https://github.com/llvm/llvm-project/pull/227292