[llvm] [SelectionDAG] Legalize vector types for atomic load (PR #111414)

[via llvm-commits]

================
@@ -1,12 +1,18 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=x86_64-apple-macosx10.7.0 -verify-machineinstrs | FileCheck %s
 ; RUN: llc < %s -mtriple=x86_64-apple-macosx10.7.0 -verify-machineinstrs -O0 | FileCheck %s
+; RUN: llc %s --mtriple=x86_64 -o - | FileCheck %s --check-prefix=CHECK0
 
----------------
jofrn wrote:

This seems like a good preexisting test to add the <1 x n> cases to. It has a specific triple `apple`, and when generalizing it, the difference in the output tests is not much, and so it duplicates quite a bit. However, I do notice that bfloat passes only when `--mtriple=x86_64`, so we may want to add the additional run line.

https://github.com/llvm/llvm-project/pull/111414