[llvm] [AMDGPU][NFC] Test subword loads (PR #220977)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,653 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
+;
+; Test scalar subword load optimization for GFX12+
+;
+; RUN: llc -mtriple=amdgpu12.50 < %s | FileCheck %s
+
+declare i32 @llvm.amdgcn.s.buffer.load.i32(<4 x i32>, i32, i32 immarg)
+
+;===----------------------------------------------------------------------===;
+; S_LOAD variants with inttoptr
+;===----------------------------------------------------------------------===;
+
+define amdgpu_kernel void @s_load_u8_inttoptr(ptr addrspace(1) %out) {
----------------
arsenm wrote:

this was a comment on the test naming, not a request to remove it 

https://github.com/llvm/llvm-project/pull/220977