Summary: | Crash in clientshape | ||
---|---|---|---|
Product: | [Plasma] kwin | Reporter: | David Edmundson <kde> |
Component: | wayland-generic | Assignee: | KWin default assignee <kwin-bugs-null> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | lewis, nate |
Priority: | NOR | ||
Version: | git master | ||
Target Milestone: | --- | ||
Platform: | Other | ||
OS: | Linux | ||
Latest Commit: | https://invent.kde.org/plasma/kwin/-/commit/00ce98b7a25f04f9efb402e86681514c6d232be8 | Version Fixed In: | 5.20.3 |
Description
David Edmundson
2020-09-15 15:25:19 UTC
Hmm, this crash is really strange. My assumption was that the surface tree never contains inert sub-surfaces. As for the WindowPixmap::shape() method, it returns the shape of the pixmap. For X11 windows, it corresponds to the bounding shape. For Wayland surfaces, it is QRect(QPoint(0, 0), size) since the Shape extension is X11-only. My backtrace for https://invent.kde.org/plasma/kwin/-/merge_requests/408 is identical to David's. FWIW, I found that checking only for surface() without subSurface() lead to black window decorations on XWayland clients. |