Chris Murphy
2018-09-04 19:05:37 UTC
I've come across a regression from Fedora 28 to Fedora 29. Problem
happens with combination of:
1. Wayland session
2. Chrome with hardware acceleration enabled; and also glxgears
3. multiple displays
4. moving the application window from one display to another or
disconnecting the external display
5. nVidia Quadro M1000M
So if any one of those things is not true, problem doesn't happen
(e.g. Fedora QA hasn't yet heard of users with multiple displays using
i915 graphics having this problem)
Bug 1616269 - [abrt] xorg-x11-server-Xwayland: OsLookupColor():
Display server crashed (edit)
https://bugzilla.redhat.com/show_bug.cgi?id=1616269
Bug includes a 'coredumpctl gdb' processing of the crashed XWayland
process. But it's unclear to me whether that's sufficient info to
track down this regression; and whether that bug needs a
bugzilla.freedesktop.org placeholder bug to point to the Fedora one or
if the Fedora one is sufficient.
Also are XWayland crashers properly assigned to mutter on GNOME? Or XOrg?
Thanks,
happens with combination of:
1. Wayland session
2. Chrome with hardware acceleration enabled; and also glxgears
3. multiple displays
4. moving the application window from one display to another or
disconnecting the external display
5. nVidia Quadro M1000M
So if any one of those things is not true, problem doesn't happen
(e.g. Fedora QA hasn't yet heard of users with multiple displays using
i915 graphics having this problem)
Bug 1616269 - [abrt] xorg-x11-server-Xwayland: OsLookupColor():
Display server crashed (edit)
https://bugzilla.redhat.com/show_bug.cgi?id=1616269
Bug includes a 'coredumpctl gdb' processing of the crashed XWayland
process. But it's unclear to me whether that's sufficient info to
track down this regression; and whether that bug needs a
bugzilla.freedesktop.org placeholder bug to point to the Fedora one or
if the Fedora one is sufficient.
Also are XWayland crashers properly assigned to mutter on GNOME? Or XOrg?
Thanks,
--
Chris Murphy
Chris Murphy