May 25
https://issues.dlang.org/show_bug.cgi?id=24562

Jonathan M Davis <issues.dlang@jmdavisProg.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|blocker                     |major

--- Comment #1 from Jonathan M Davis <issues.dlang@jmdavisProg.com> ---
Okay. In our current situation we were able to work around the issue via explicit opAssigns, so I reduce this from being marked as a blocker, but in cases where you're dealing with templated code outside your control, that won't work. So, this is still a major issue.

--
May 27
https://issues.dlang.org/show_bug.cgi?id=24562

RazvanN <razvan.nitu1305@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |razvan.nitu1305@gmail.com

--- Comment #2 from RazvanN <razvan.nitu1305@gmail.com> ---
(In reply to Jonathan M Davis from comment #1)
> Okay. In our current situation we were able to work around the issue via explicit opAssigns, so I reduce this from being marked as a blocker, but in cases where you're dealing with templated code outside your control, that won't work. So, this is still a major issue.

My hunch is that the generated opAssign is probably using some code internally that copy constructs an instance of S1 but it mismatches some qualifiers. Deleting the copy constructor probably fixes this because it can then do some implicit blitting. Will check now.

--