Deadlocking problem related to Win clipboard thread that still exists

[Noel Grandin]

On Mon, 14 Dec 2020 at 08:34, Mike Kaganski <mikekaganski at hotmail.com>
wrote:

>
> I suppose we must check to release solar mutex before any blocking calls
> to clipboard thread, including explicit waits (like
> CMtaOleClipboard::getClipboard that posts a message and then waits) and
> implicit waits like described above, calling COM methods of objects
> known to be created in a different thread.
>
>
Probably that ^^ and also we could post events to the main thread from the
clipboard, rather than directly handling them in the clipboard thread - the
stack trace referenced above looks like it could do that.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/libreoffice/attachments/20201214/749e44eb/attachment.htm>