cor3ntin wrote:

> Why doesn't this break the 'deferred instantiation/checking' rules? My 
> understanding is that we're not allowed to instantiate these until we need 
> them for checking, since they are allowed to Change up until the first time 
> they are checked.

The thing we should not do here - and I need to check we are not doing the 
wrong thing is failing when the substitution in the concept arguments - which 
might not be used in the atomic constraint - fails,

I guess there is a question of what happens if there is an error in a non 
sfinae context in the concept arguments.

https://github.com/llvm/llvm-project/pull/199671
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to