[PATCH 0/1] MR5257: winex11: Set the correct visual even if alpha matches.

[Gabriel Ivăncescu]

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/5257

[PATCH 1/1] winex11: Set the correct visual even if alpha matches.

[Gabriel Ivăncescu]

From: Gabriel Ivăncescu <gabrielopcode(a)gmail.com>

8a71a4a30495635bdc905af41869d8ed9e43009f caused us to set a different visual
but with the correct alpha prior to actually setting it to the systray's
visual. In this case, the alpha would match and not set it to the systray's
visual. This caused the Ubuntu 23.10.1 default WM to crash.

Signed-off-by: Gabriel Ivăncescu <gabrielopcode(a)gmail.com>
---
 dlls/winex11.drv/window.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/dlls/winex11.drv/window.c b/dlls/winex11.drv/window.c
index afa13f01277..5a014c9080d 100644
--- a/dlls/winex11.drv/window.c
+++ b/dlls/winex11.drv/window.c
@@ -1798,15 +1798,16 @@ static void destroy_whole_window( struct x11drv_win_data *data, BOOL already_des
  */
 void set_window_visual( struct x11drv_win_data *data, const XVisualInfo *vis, BOOL use_alpha )
 {
+    BOOL same_visual = (data->vis.visualid == vis->visualid);
     Window client_window = data->client_window;
     Window whole_window = data->whole_window;
 
-    if (!data->use_alpha == !use_alpha) return;
+    if (!data->use_alpha == !use_alpha && same_visual) return;
     if (data->surface) window_surface_release( data->surface );
     data->surface = NULL;
     data->use_alpha = use_alpha;
 
-    if (data->vis.visualid == vis->visualid) return;
+    if (same_visual) return;
     data->client_window = 0;
     destroy_whole_window( data, client_window != 0 /* don't destroy whole_window until reparented */ );
     data->vis = *vis;
-- 
GitLab

https://gitlab.winehq.org/wine/wine/-/merge_requests/5257

[Rémi Bernon]

I don't think the test failures are related. The d3d10_1 and d3dx9_36:effect failures come from VKD3D 1.11 merge, the ddraw one looks to be some returning failure and urlmon has been failing recently for some reason.

The global timeout after user32:input (when it doesn't seem related to the amount of time spent in previous tests) is more worrying, but whatever is going on there I feel like I'm maybe to blame.

-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/5257#note_63887

[PATCH 0/1] MR5257: winex11: Set the correct visual even if alpha matches. - approved

[Rémi Bernon]

This merge request was approved by Rémi Bernon.
-- 
https://gitlab.winehq.org/wine/wine/-/merge_requests/5257