queue_task cannot free the struct because it doesn't know whether the
struct contains pointers to other dynamically allocated memory that also
needs to be freed (as is the case with struct send_request).
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/3015
If read_identity returns FALSE, it has already called clear_identity.
This is the same problem as in merge request !2970. Unfortunately I did not notice at first that the problem was present in two places.
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/3013
That'll mean that the host's `TxViewChange` method would get called - we'd need a test to see if that's the case. It still seems like we're hiding the problem. What crash did you see when using `ME_WrapMarkedParagraphs()`?
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/2941#note_35104
I wasn't completely sure what to do with the registry key. Keeping "Wine/X11 Driver" is a bit odd, but it also makes sure we stay backward compatible. Changing it would require more changes elsewhere and maybe it could be done later, if / when more options are moved.
--
v2: win32u: Move fullscreen window cursor clipping from winex11.
winex11: Move clip_fullscreen_window foreground check inside it.
win32u: Add a clipping_cursor member to user_thread_info.
win32u: Add a clipping_reset member to user_thread_info.
win32u: Move grab_pointer registry option from winex11.
winex11: Remove now unnecessary WM_X11DRV_CLIP_CURSOR_NOTIFY.
https://gitlab.winehq.org/wine/wine/-/merge_requests/3021
I wasn't completely sure what to do with the registry key. Keeping "Wine/X11 Driver" is a bit odd, but it also makes sure we stay backward compatible. Changing it would require more changes elsewhere and maybe it could be done later, if / when more options are moved.
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/3021