dgit.raspbian.org Git - mutter.git/log

Merge version 3.28.1-1+rpi1 and 3.28.1-2 to produce 3.28.1-2+rpi1

Add build-depends on libgles2-mesa-dev

Merge version 3.28.0-2+rpi1 and 3.28.1-1 to produce 3.28.1-1+rpi1

Merge mutter (3.28.1-2) import into refs/heads/workingbranch

xwayland: Don't abort if Xwayland crashes

Right now if Xwayland crashes, we crash, too.

On some level that makes sense, since we're supposed to control the
lifecycle of Xwayland, and by it crashing we've lost that control.

But practically speaking, the knock-on crash adds noise to the logs,
bug trackers, and retrace servers that only makes debugging harder.
And the crash isn't something mutter can "fix", since it's
ultimately from a bug in Xwayland anyway.

This commit makes mutter exit instead of crash if Xwayland goes away
unexpectedly.

Origin: https://gitlab.gnome.org/GNOME/mutter/commit/2d80fd0

Gbp-Pq: Name xwayland-Don-t-abort-if-Xwayland-crashes.patch

xwayland: use g_autoptr for GError in xserver_died

Right now we explicitly g_clear_error any error we find, but that
makes it tricky to return early from the function, which a
subsequent commit will want to to do.

This commit switches GError to use g_autoptr so the error clearing
happens automatically.

Origin: https://gitlab.gnome.org/GNOME/mutter/commit/bb65854

Gbp-Pq: Name xwayland-use-g_autoptr-for-GError-in-xserver_died.patch

backends: Add logical monitor -> monitor -> output -> crtc ref chain

Bug-Ubuntu: https://bugs.launchpad.net/ubuntu/+source/mutter/+bug/1703668
Bug-GNOME: https://bugzilla.gnome.org/show_bug.cgi?id=786929
Forwarded: yes, https://gitlab.gnome.org/GNOME/mutter/merge_requests/81

Make it so that each logical monitor has a reference to all the
monitors that are assigned to it.

All monitors has a reference to each output that belongs to it.

Each output has a reference to any CRTC it has been assigned.

https://bugzilla.gnome.org/show_bug.cgi?id=786929

Gbp-Pq: Name backends-Add-logical-monitor-monitor-output-crtc-ref-chai.patch

backends: Move MetaOutput::crtc field into private struct

Bug-Ubuntu: https://bugs.launchpad.net/ubuntu/+source/mutter/+bug/1703668
Bug-GNOME: https://bugzilla.gnome.org/show_bug.cgi?id=786929
Forwarded: yes, https://gitlab.gnome.org/GNOME/mutter/merge_requests/81

No functional changes. This is only done so that changes to reference
counting can done more reliably.

https://bugzilla.gnome.org/show_bug.cgi?id=786929

Gbp-Pq: Name backends-Move-MetaOutput-crtc-field-into-private-struct.patch

core: Return -1 if meta_window_get_monitor is called on an unmanaged window

Forwarded: yes
Bug-GNOME: https://bugzilla.gnome.org/show_bug.cgi?id=788834
Bug-Ubuntu: https://bugs.launchpad.net/ubuntu/bionic/+source/gnome-shell/+bug/1724439

As opposed to crashing. In this case, letting the caller deal with
it is the best policy, since this is public API.

Fixes #78834

Gbp-Pq: Name core-Return-1-if-meta_window_get_monitor-is-called-on-an-.patch

clutter: Smooth out master clock to smooth visuals

Bug-GNOME: https://gitlab.gnome.org/GNOME/mutter/issues/25
Forwarded, yes: https://gitlab.gnome.org/GNOME/mutter/merge_requests/70

Clutter's master clock was jittery because it included errors in cur_tick
such as dispatch delays due to other sources. Dispatch could also occur up
to 1ms early since GSource can only be timed to the millisecond. All of this
could impact the visual smoothness of animations as they are displayed on
the steady interval of the monitor, but spacially moving in less regular
steps derived from the dispatch times.

The simple fix is to ignore any jitter in dispatch timing. Try a little
bit harder to use a precise interval that will better match the display
hardware, and smoother visuals will follow.

https://gitlab.gnome.org/GNOME/mutter/issues/25

Gbp-Pq: Name clutter-Smooth-out-master-clock-to-smooth-visuals.patch

theme: load icons as Gtk does with fallback and RTL support

Forwarded: yes, https://gitlab.gnome.org/GNOME/mutter/merge_requests/62

Gbp-Pq: Name theme-load-icons-as-Gtk-does-with-fallback-and-RTL-suppor.patch

theme: use gtk_render_icon_suface to paint button icon

Bug-Ubuntu: https://bugs.launchpad.net/ubuntu/+source/mutter/+bug/1764558
Bug-GNOME: https://gitlab.gnome.org/GNOME/mutter/issues/100
Forwarded: yes, https://gitlab.gnome.org/GNOME/mutter/merge_requests/62

This will properly take care of the icon transformations
and of the shadow.

Fixes #100

Gbp-Pq: Name theme-use-gtk_render_icon_suface-to-paint-button-icon.patch

theme, frames: Use surface device scale instead of cairo_scale

Bug-Ubuntu: https://bugs.launchpad.net/ubuntu/+source/mutter/+bug/1764554
Bug-GNOME: https://gitlab.gnome.org/GNOME/mutter/issues/99
Forwarded: yes, https://gitlab.gnome.org/GNOME/mutter/merge_requests/61

Gtk now is caching the themed cairo surfaces, then as per
commit gtk@e36b629c the surface device scale is used to figure
out the current paint scaling.

Without this when using background-image's for window buttons
the -gtk-scaled icons isn't properly resized.

Fixes #99

Gbp-Pq: Name theme-frames-Use-surface-device-scale-instead-of-cairo_sc.patch

backends/x11: Support input-synaptics, if present.

Add support for configuring the Xorg synaptics touchpad driver.

Turns out it's very simple to support both libinput and synaptics
simultaneously, both under the heading of XI2.

Bug-Ubuntu: https://launchpad.net/bugs/1686081
Forwarded: https://gitlab.gnome.org/GNOME/mutter/merge_requests/37
Last-Update: 2018-03-16

Gbp-Pq: Name synaptics-support.patch

Skip test that started failing with 3.25

That way we can still benefit by running all the other tests.

Bug: https://bugzilla.gnome.org/785319
Applied-upstream: no

Gbp-Pq: Name skip-failing-tests-325.patch

Don't run failing tests

That way we can still benefit by running all the other tests.

Bug: https://bugzilla.gnome.org/782215
Bug: https://bugzilla.gnome.org/753836
Applied-upstream: no

Gbp-Pq: Name skip-failing-tests.patch

gtk-shell: Work around non-working startup notifications

GNOME Shell relies on the MetaScreen::startup-sequence-changed signal,
which is tied to (lib)startup-notification and therefore X11. As a result,
when we remove the startup sequence of a wayland client, GNOME Shell will
not be notified about this until startup-notification's timeout is hit.
As a temporary stop-gap, go through XWayland even for wayland clients,
so that the signal is emitted when expected.

Bug: https://bugzilla.gnome.org/show_bug.cgi?id=768531
Applied-upstream: no

Gbp-Pq: Name bgo768531_workaround-startup-notifications.patch

mutter (3.28.1-2) unstable; urgency=medium

  * Add xwayland-use-g_autoptr-for-GError-in-xserver_died.patch,
    xwayland-Don-t-abort-if-Xwayland-crashes.patch:
    - Cherry picked from upstream, to reduce noise of mutter on crashes
      which are actually caused by XWayland (LP: #1748450)

[dgit import unpatched mutter 3.28.1-2]

Import mutter_3.28.1-2.debian.tar.xz

[dgit import tarball mutter 3.28.1-2 mutter_3.28.1-2.debian.tar.xz]

Merge mutter (3.28.1-1) import into refs/heads/workingbranch

Import mutter_3.28.1.orig.tar.xz

[dgit import orig mutter_3.28.1.orig.tar.xz]