[llvm] [AArch64][GlobalISel] Fix incorrect codegen for FPR16/FPR8 to GPR copies (PR #171499)

[Matt Arsenault via llvm-commits]

================
@@ -568,6 +568,10 @@ AArch64LegalizerInfo::AArch64LegalizerInfo(const AArch64Subtarget &ST)
         return Query.Types[0] == s128 &&
                Query.MMODescrs[0].Ordering != AtomicOrdering::NotAtomic;
       })
+      .widenScalarIf(
----------------
arsenm wrote:

No, that shouldn't be an issue 

https://github.com/llvm/llvm-project/pull/171499