Michael137 wrote:
> The libstdc++ test fails. Maybe we don't even have a formatter for
> std::atomic there? It currently looks like:
>
> ```
>
> (std::atomic<int>) i = {
>
> std::__atomic_base<int> = {
>
> _M_i = 47
>
> }
>
> }
>
> ```
>
> .. which isn't that bad. Much better than libc++'s
>
> ```
>
> (std::__1::atomic<int>) i = {
>
> std::__1::__atomic_base<int, true> = {
>
> std::__1::__atomic_base<int, false> = {
>
> __a_ = {
>
> std::__1::__cxx_atomic_base_impl<int> = {
>
> __a_value = 47
>
> }
>
> }
>
> }
>
> }
>
> }
>
> ```
>
> We may still want to have a data formatter for consistency, but I suspect
> that's not what you're looking to do now. Maybe just move this part into the
> generic folder, but don't add the libstdc++ test case yet?
Huh I'm surprised that's what it looks like for libc++ tbh. Not sure what our
formatter is doing.
For libstdc++ we dont have a formatter. I think that's the case for a decent
number of our libcxx formatters. My plan was to just XFAIL. But just omitting
the libstdc++ category entirely for the unsupported formatters seems better
https://github.com/llvm/llvm-project/pull/146843
_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
