Backport MR #2353 to fix overview search with input methods (#2062660)

0001-clutter-Refactor-code-marking-actors-dirty-for-paint.patch

@@ -0,0 +1,65 @@
+From ce7f606d48f1a422465fd597c33047727993988a Mon Sep 17 00:00:00 2001
+From: Carlos Garnacho <carlosg@gnome.org>
+Date: Wed, 30 Mar 2022 20:59:37 +0200
+Subject: [PATCH 1/2] clutter: Refactor code marking actors dirty for paint()
+
+Simplify the function arguments (the origin is just the actor that
+the function is originally called from), and make it also handle
+marking as dirty the actor that got the redraw queued up explicitly.
+
+This makes it a single place where priv->is_dirty is being enabled.
+
+Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2353>
+---
+ clutter/clutter/clutter-actor.c | 16 +++++++---------
+ 1 file changed, 7 insertions(+), 9 deletions(-)
+
+diff --git a/clutter/clutter/clutter-actor.c b/clutter/clutter/clutter-actor.c
+index 59345a7252..2e4b30effc 100644
+--- a/clutter/clutter/clutter-actor.c
++++ b/clutter/clutter/clutter-actor.c
+@@ -2646,9 +2646,10 @@ _clutter_actor_queue_redraw_on_clones (ClutterActor *self)
+ }
+ 
+ static void
+-_clutter_actor_propagate_queue_redraw (ClutterActor *self,
+-                                       ClutterActor *origin)
++_clutter_actor_propagate_queue_redraw (ClutterActor *self)
+ {
++  ClutterActor *origin = self;
++
+   while (self)
+     {
+       /* no point in queuing a redraw on a destroyed actor */
+@@ -2657,13 +2658,12 @@ _clutter_actor_propagate_queue_redraw (ClutterActor *self,
+ 
+       _clutter_actor_queue_redraw_on_clones (self);
+ 
++      self->priv->is_dirty = TRUE;
++
+       /* If the queue redraw is coming from a child then the actor has
+          become dirty and any queued effect is no longer valid */
+       if (self != origin)
+-        {
+-          self->priv->is_dirty = TRUE;
+-          self->priv->effect_to_redraw = NULL;
+-        }
++        self->priv->effect_to_redraw = NULL;
+ 
+       /* If the actor isn't visible, we still had to emit the signal
+        * to allow for a ClutterClone, but the appearance of the parent
+@@ -8105,10 +8105,8 @@ _clutter_actor_queue_redraw_full (ClutterActor             *self,
+       priv->effect_to_redraw = NULL;
+     }
+ 
+-  priv->is_dirty = TRUE;
+-
+   if (!priv->propagated_one_redraw)
+-    _clutter_actor_propagate_queue_redraw (self, self);
++    _clutter_actor_propagate_queue_redraw (self);
+ }
+ 
+ /**
+-- 
+2.35.1
+

0002-clutter-Keep-actors-dirty-if-a-redraw-was-queued-up-.patch

@@ -0,0 +1,59 @@
+From f820bb35067f1e6b54d56f7652ee333ac8c8c35b Mon Sep 17 00:00:00 2001
+From: Carlos Garnacho <carlosg@gnome.org>
+Date: Wed, 30 Mar 2022 21:02:18 +0200
+Subject: [PATCH 2/2] clutter: Keep actors dirty if a redraw was queued up
+ during paint()
+
+In the right combination of circumstances, and given 2 actors (parent
+actor P with an offscreen effect and child actor C), we may have the
+following situation happening:
+
+- A redraw is queued on the actor C, actors C and P are marked as
+  priv->is_dirty and priv->propagated_one_redraw.
+- During paint() handling we paint actor P, priv->propagated_one_redraw
+  is turned off.
+- We recurse into child actor C, priv->propagated_one_redraw is turned
+  off.
+- A new redraw is queued on actor C, actors C and P are marked as
+  priv->is_dirty and priv->propagated_one_redraw.
+- The paint() method recurses back, actors C and P get priv->is_dirty
+  disabled, priv->propagated_one_redraw remains set.
+- At this point queueing up more redraws on actor C will not propagate
+  up, because actor C has priv->propagated_one_redraw set, but the
+  parent actor P has priv->is_dirty unset, so the offscreen effect will
+  not get CLUTTER_EFFECT_PAINT_ACTOR_DIRTY and will avoid repainting
+  actor C.
+
+The end result is that actor C does not redraw again, despite requesting
+redraws. This situation eventually resolves itself through e.g. relayouts
+on actor P, but may take some time to happen.
+
+In order to fix this, consider actors that did get a further redraw
+request still dirty after paint().
+
+Fixes: https://gitlab.gnome.org/GNOME/mutter/-/issues/2188
+Part-of: <https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2353>
+---
+ clutter/clutter/clutter-actor.c | 6 ++++--
+ 1 file changed, 4 insertions(+), 2 deletions(-)
+
+diff --git a/clutter/clutter/clutter-actor.c b/clutter/clutter/clutter-actor.c
+index 2e4b30effc..23ef03dc1f 100644
+--- a/clutter/clutter/clutter-actor.c
++++ b/clutter/clutter/clutter-actor.c
+@@ -3840,8 +3840,10 @@ clutter_actor_paint (ClutterActor        *self,
+   clutter_paint_node_paint (root_node, paint_context);
+ 
+   /* If we make it here then the actor has run through a complete
+-     paint run including all the effects so it's no longer dirty */
+-  priv->is_dirty = FALSE;
++   * paint run including all the effects so it's no longer dirty,
++   * unless a new redraw was queued up.
++   */
++  priv->is_dirty = priv->propagated_one_redraw;
+ }
+ 
+ /**
+-- 
+2.35.1
+

mutter.spec

@@ -10,7 +10,7 @@
 
 Name:          mutter
 Version:       42.0
-Release:       2%{?dist}
+Release:       3%{?dist}
 Summary:       Window and compositing manager based on Clutter
 
 License:       GPLv2+
@@ -38,6 +38,13 @@ Patch3:        2331.patch
 # https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2299
 Patch4:        2299.patch
 
+# Fix input to overview search freezing when using input method
+# https://gitlab.gnome.org/GNOME/mutter/-/issues/2188
+# https://bugzilla.redhat.com/show_bug.cgi?id=2062660
+# https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/2353
+Patch5:        0001-clutter-Refactor-code-marking-actors-dirty-for-paint.patch
+Patch6:        0002-clutter-Keep-actors-dirty-if-a-redraw-was-queued-up-.patch
+
 BuildRequires: pkgconfig(gobject-introspection-1.0) >= 1.41.0
 BuildRequires: pkgconfig(sm)
 BuildRequires: pkgconfig(libwacom)
@@ -181,6 +188,9 @@ desktop-file-validate %{buildroot}/%{_datadir}/applications/%{name}.desktop
 %{_datadir}/mutter-%{mutter_api_version}/tests
 
 %changelog
+* Thu Mar 31 2022 Adam Williamson <awilliam@redhat.com> - 42.0-3
+- Backport MR #2353 to fix overview search with input methods (#2062660)
+
 * Mon Mar 14 2022 Adam Williamson <awilliam@redhat.com> - 42.0-2
 - Backport MR #2299 to avoid a commonly-encountered crash (#2063381)