================
@@ -3141,6 +3148,10 @@ def warn_attribute_ignored_no_calls_in_stmt: Warning<
   "statement">,
   InGroup<IgnoredAttributes>;
 
+def warn_attribute_needs_aggregate : Warning<
----------------
higher-performance wrote:

I think I disagree. It's not really erroneous to do this, conceptually 
speaking. It's just a no-op, because in these cases there is never a situation 
where the user directly initializes the fields to begin with.

Moreover, if in the future versions of the compiler we decide to allow it in 
some fashion, then suddenly earlier compiler versions would produce a hard 
error, making it impossible for users to switch back to old compilers with 
their new code. I don't see any gain for us in making life hard for users like 
that; we might as well make it a warning.

https://github.com/llvm/llvm-project/pull/102040
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to