I have unpushed the f38 build.
https://bodhi.fedoraproject.org/updates/FEDORA-2024-68a2dba357
On Wed, 2024-01-24 at 11:34 +0100, Leigh Scott wrote:
see https://github.com/GNOME/gtk/commit/77ebdd85091833a7869ece48c3114fa6d9966321
Hi, all the bugs you referenced crash in X11 code. The above NEWS file commit specifically says it's for Wayland.
What am I missing here, please? Did you pick a wrong commit? What was the commit supposed to proof, please?
Bye, Milan
P.S.: opening three threads for a single thing with identical text is not nice. It can confuse archive readers, not finding answers they might be looking for. Try to avoid that in the future, please.
Am Mi., 24. Jan. 2024 um 12:40 Uhr schrieb Milan Crha mcrha@redhat.com:
On Wed, 2024-01-24 at 11:34 +0100, Leigh Scott wrote:
see https://github.com/GNOME/gtk/commit/77ebdd85091833a7869ece48c3114fa6d9966321
Hi,all the bugs you referenced crash in X11 code. The above NEWS file commit specifically says it's for Wayland.
What am I missing here, please? Did you pick a wrong commit? What was the commit supposed to proof, please?
You missed reading the first NEWS entry ("* Fix a crash introduced in the X11 changes in 3.24.40") and quoted the third only.
Now, whether that diff fixes our problems can only be "proved" by rebasing our package or picking those commits and testing.
Michael
On Wed, 2024-01-24 at 12:53 +0100, Michael J Gruber wrote:
You missed reading the first NEWS entry ("* Fix a crash introduced in the X11 changes in 3.24.40") and quoted the third only.
Hi, you are right. That's quite embarrassing. I'm sorry about that. No idea how I could overlook it. My fault. Bye, Milan
On Wed, Jan 24, 2024 at 5:25 AM Leigh Scott leigh123linux@gmail.com wrote:
Why isn't gtk3 a critical path package?
https://bugzilla.redhat.com/show_bug.cgi?id=2260068 https://bugzilla.redhat.com/show_bug.cgi?id=2260073 https://bugzilla.redhat.com/show_bug.cgi?id=2260074
https://bodhi.fedoraproject.org/updates/FEDORA-2024-5d8ac182bf
If it was, how would we test this? We need a GTK 3 desktop that has OpenQA tests so the affected commit would have been caught earlier: https://gitlab.gnome.org/GNOME/gtk/-/commit/5de743572ed0d41ab881816c06bc918c...
We don't currently have tests for desktops other than GNOME and KDE Plasma, so there's no way we would have known until after the fact. OpenQA tests are how we gate these kinds of things. While KDE Plasma has one GTK3 application tested (abrt), it is not affected by this issue since it doesn't interact with the gtk-layer-shell stuff.
-- 真実はいつも一つ！/ Always, there's only one truth!
-
Leigh Scott -
Michael J Gruber -
Milan Crha -
Neal Gompa