Daniel van Vugt [Fri, 6 Apr 2018 10:26:58 +0000 (18:26 +0800)]
js/ui: Choose some actors to cache on the GPU
Flag some actors that are good candidates for caching in texture memory
(what Clutter calls "offscreen redirect"), thereby mostly eliminating
their repaint overhead.
This isn't exactly groundbreaking, it's how you're meant to use
OpenGL in the first place. But the difficulty is in the design of
Clutter which has some peculiarities making universal caching
inefficient at the moment:
* Repainting an offscreen actor is measurably slower than repainting
the same actor if it was uncached. But only by less than 100%,
so if an actor can avoid changing every frame then caching is usually
more efficient over that timeframe.
* The cached painting from a container typically includes its children,
so you can't cache containers whose children are usually animating at
full frame rate. That results in a performance loss.
This could be remedied in future by Clutter explicitly separating a
container's background painting from its child painting and always
caching the background (as StWidget tries to in some cases already).
So this commit selects just a few areas where caching has been verified
to be beneficial, and many use cases now see their CPU usage halved:
One small window active...... 10% -> 7% (-30%)
...under a panel menu........ 23% -> 9% (-61%)
One maximized window active.. 12% -> 9% (-25%)
...under a panel menu........ 23% -> 11% (-52%)
...under a shell dialog...... 22% -> 12% (-45%)
...in activities overview.... 32% -> 17% (-47%)
(on an i7-7700)
Revert "workspaceThumbnail: rebuild thumbnails if workareas size changed"
It is unclear what the change was supposed to be fixing, but it
broke animations of workspace additions and removals, as those
events trigger the ::workareas-changed signal.
Carlos Garnacho [Wed, 15 Aug 2018 13:39:58 +0000 (15:39 +0200)]
runDialog: Use ClutterText::activate for enter handling
Instead of consuming the event in front of the input method. Enter
is sometimes overriden by those, so it seems better to let the IM
handle the key event, and react later to it if it got propagated
anyway. That is what ::activate does, so use this signal.
This used to work before ClutterInputMethod/InputFocus because the
IM received the events directly from stage captured events. This
is not the case anymore.
We used to keep the workspace switcher slid out when the user made use
of workspaces. This was changed in commit 2d84975 to give more space
to window previews, but it turned out to make the switcher quite a lot
more difficult to interact with (rather than only being a question of
discoverability). So go back to the previous behavior.
overviewControls: Sync hover after drag operations
During global grabs, actors miss enter and leave events required
for correct hover tracking. This can cause the workspace switcher
to get stuck while slid out, so ensure the actor's hover state is
synced after drag operations.
Florian Müllner [Fri, 10 Aug 2018 11:56:07 +0000 (13:56 +0200)]
build: Use dedicated resources for helper programs
Using a single resource file for all JS sources saves a couple of
build system instructions, but has some serious downsides:
- bundling the entire shell code with the tools blows
up their size unnecessarily
- the tools are rebuilt unnecessarily for any shell
code change
Autotools was painful enough to let this slip, but with meson we
don't have any excuses - using the actual dependencies speeds up
the build a tiny bit and reduces the tools' sizes from over 2M
to about 50k.
Florian Müllner [Fri, 10 Aug 2018 11:43:59 +0000 (13:43 +0200)]
build: Clean up tools' dependencies
Simply reusing the same dependencies as gnome-shell itself not only
means that we link tons of stuff unnecessarily, but also that we
have to do the whole mutter rpath dance for nothing. Just use the
dependencies those executables actually need for a nice cleanup.
Florian Müllner [Fri, 10 Aug 2018 11:34:15 +0000 (13:34 +0200)]
build: Don't modify typelib search path in tools
We need this in the main gnome-shell executable in order to locate
the private Shell and St typelibs, but those aren't useful or even
usable in the extension-prefs/portal helper tools.
Carlos Garnacho [Fri, 10 Aug 2018 19:37:48 +0000 (21:37 +0200)]
layout: Mark chrome container as NO_LAYOUT
Showing and hiding children does not affect the allocation of the uiGroup
nor its other children. We can avoid full relayout/redraw in those
situations.
Pascal Nowack [Mon, 6 Aug 2018 12:21:02 +0000 (14:21 +0200)]
build: Fix build with meson >= 47.0
With meson >= 47.0, building gnome-shell will fail
when --rebuild-types is used as an option for gtk-doc,
so stop using that option and bump the meson requirement.
We show a cover pane on top of the overview during transitions to
prevent issues caused by clicks and mouseover events when the overview
is not ready. Right now, this pane is only being shown on the primary
monitor, which obviosly allows interactions to happen before the
animations are finished on the secondary monitors.
To fix this, use the size of the whole stage for the cover pane.
Carlos Garnacho [Fri, 13 Jul 2018 22:14:07 +0000 (00:14 +0200)]
keyboard: defer position-changed till we have a rect
Emitting it that soon results in JS warnings, as we don't have
everything in place yet. The position-changed signal will be
emitted from other locations as soon as we have it.
Olivier Blin [Thu, 2 Aug 2018 22:35:24 +0000 (00:35 +0200)]
tools: fix XDG desktop syntax in gnome-shell-overrides-migration
Reported by desktop-file-validate:
error: value "True" for boolean key "NoDisplay" in group "Desktop
Entry" contains invalid characters, boolean values must be "false" or
"true"
verdre [Tue, 19 Jun 2018 10:45:18 +0000 (12:45 +0200)]
workspace: Simplify detecting added dialogs after closing a window
When trying to close a window in the overview by clicking the close
button and the window doesn't get closed but a dialog is added to the
window afterwards, we close the overview and show the dialog.
Instead of adding a separate listener for the window-added signal to the
WindowOverlay, let the WindowClones remember that the close button was
pressed and activate themselves if a dialog is added after that.
workspace: Fix infinite loop when finding parent window of dialogs
When a dialog is added to a window while the overview is shown, we get
its parent using get_transient_for() so we can add it to the right
window clone.
If we have multiple layers of dialogs we have to do this recursively
until we find the root ancestor. This case currently results in an
infinite loop: Since parent is always set to the same window, the
while-condition will always be true.
Florian Müllner [Wed, 1 Aug 2018 11:27:07 +0000 (13:27 +0200)]
showOSD: Fix handling of defined 'falsy' parameters
For the OSD, all parameters except for the icon are optional - if the
caller doesn't include the 'label' option, the OSD won't show a label
etc.
While this makes sense for an API, it means that we have to be careful
to correctly differentiate an option that was omitted and an option
that has a 'falsy' value like false or 0.
Unfortunately since commit ccaae5d3c we no longer do, with the result
that OSDs meant for the first monitor will show up on all, and a level
of 0 is presented as no level bar instead of an empty one, whoops.
While the new per-desktop overrides in GIO are easier to use for
both developers and users, it is still inconvenient for everyone
who changed the defaults using the old overrides hack to lose
their settings. Address this by running a small script on startup
that migrates existing settings.
Florian Müllner [Wed, 16 Aug 2017 19:15:48 +0000 (21:15 +0200)]
Replace custom override schema with per-desktop override
GSettings now recognizes per-desktop overrides that can be used
to change schemas' default values for a particular desktop. This
is not entirely unlike our existing custom override mechanism in
mutter, except that it is not limited to keys in org.gnome.mutter,
and it doesn't require a separate schema - the latter means that
we (and gnome-teak-tool) no longer have to figure out the correct
schema for the current login session and just use the original one.
Didier Roche [Fri, 9 Feb 2018 14:10:16 +0000 (15:10 +0100)]
volume: Show overamplified icon when in overdrive
Show an overamplified volume icon if volume is louder the max normalized one.
Use a similar logic as gnome-settings-daemon to delimit values, restricted
to output.
The purpose is to help users remember that visiting some websites or
using some apps can get LOUD.
Implement for barLevel an overdrive area. This is a zone represented via a
different styling to indicate that you are bypassing the normal zone of
a given level, without reaching yet the maximum limit.
Depending on hardware and recorded volume level, turning up the speakers
to the maximum volume may not be enough and the user will want to amplify
the volume above 100%. Currently this requires opening the sound Settings
panel which gets cumbersome when required repeatedly.
To support this case better, allow raising the sound volume above 100%
directly from the system menu if the feature is enabled via the
`allow-volume-above-100-percent` key in `org.gnome.desktop.sound`.
Allow osd representing levels that can be more than 100% by accepting
an optional parameter setting that maximum level.
gnome-settings-daemon will use this to indicate volume levels above 100%,
which our own volume indicator will soon support as well.
Didier Roche [Thu, 8 Feb 2018 18:22:29 +0000 (19:22 +0100)]
osdWindow: Reuse BarLevel drawing functionality
Reuse the BarLevel class to get similar drawing behavior as Slider.
Rename theme css impacted properties and ensure that the osdWindow
remains accessible.
Ensure we don't force setting a custom border color like on the OSD.
As strings are guaranteed to use UTF-8 in the GNOME platform, generic
file APIs like g_file_load_contents() return raw data instead. Since
gjs' recent update to mozjs60, this data is now returns as Uint8Array
which cannot simply be treated as string - its toString() method boils
down to arr.join(',') - so use gjs' new ByteArray module to explicitly
convert the data.
The settings action button in the system menu simply launches
gnome-control-center, so we want its icon (and accessible name)
to always match the app. So instead of keeping the button in-sync
with Settings, just look up that information from the app itself.
We can simply request the symbolic variant from CSS so that we don't
have to append '-symbolic' to all the names. This will always make
it easier to pick up that information from external sources (like
.desktop files).
windowManager: Fix switch animation between non-adjacent workspaces
Commit e5c95b910d refactored the workspace animation to also handle
animations that involve all surrounding workspaces, but due to an
ill-advised review comment (guess whose) it broke the animation
for non-neighboring workspaces.
Update the code to handle correctly whether in a given direction:
- we have the target workspace of a given index
- we have a neighboring workspace
- we don't need to animate anything
Daniel Drake [Fri, 27 Jul 2018 16:30:22 +0000 (11:30 -0500)]
Add check_cloexec_fds debug command
Add a debug command (to be executed manually via Alt+F2) to check
that all of gnome-shell's file descriptors have the CLOEXEC flag set.
This is important so that internal file descriptors do not get passed
to apps when they are launched.
It prints a warning message for every fd that does not have the flag set.
fdwalk() is used from the standard library if available (it is not
available in glibc), otherwise we use the same implementation as glib
has internally.
Daniel Drake [Wed, 6 Jun 2018 13:07:17 +0000 (07:07 -0600)]
shell-app: remove child_setup from app launching
When the amount of free memory on the system is somewhat low, gnome-shell
will sometimes fail to launch apps, reporting the error:
fork(): Cannot allocate memory
fork() is failing here because while cloning the process virtual address
space, Linux worries that the thread being forked may end up COWing the
entire address space of the parent process (gnome-shell, which is
memory-hungry), and there is not enough free memory to permit that to
happen. This check is somewhat irrelevant because we are only forking
to immediately exec(), which will discard the whole virtual address
space anyway.
This issue can be avoided by using a new optimized gspawn codepath in
the latest glib development version, which uses posix_spawn() internally.
For the optimized codepath to be used, we must not pass a child_setup
function, so the the file descriptor management is reimplemented here
using new glib API to pass fds to the child process. The old API will
continue to be used on older glib versions.
We must also change the spawn flags for this code path to be hit.
I checked that gnome-shell's open file descriptors are all CLOEXEC
so using G_SPAWN_LEAVE_DESCRIPTORS_OPEN should be safe.
This will result in more resilient app launching when memory is low,
since the optimized spawn path avoids cloning the virtual address
space of the parent process (gnome-shell) and avoids the irrelevant
memory overcommit check.
Carlos Garnacho [Fri, 13 Jul 2018 21:38:03 +0000 (23:38 +0200)]
windowManager: Make workspace switching gestures follow motion
When 4fg swipe motion happens, set up early the workspace switching
animation with all surrounding workspaces. This allows us to move
all content back and forth in any direction. This works on both
touchcreens and touchpads.
When the gesture is activated, the same data is reused to follow
up with the tween animation.
The threshold has been also doubled, it was fairly small to start
with, and feels better now that workspaces stick to fingers.
Besides the separation into distinct functions, the stored data has
been made able to generically store windows from all surrounding
workspaces. All while keeping a special mode to animate between two
workspaces (The usual till now), this is the only mode exercised so
far.
In order to ease animations, all window groups are now children of
a common container, which is then animated.
app: Consider "new-window" action for opening new windows
While can_open_new_window() uses some elaborate heuristics to predict
whether an application can open multiple windows, open_new_window()
will always simply relaunch the application. This is often the best
we can do, but when an application provides a "new-window" action in
its .desktop file or on the bus, it is much more likely to work as
expected than blindly activating the app and hoping for a particular
behavior.
Graphical applications like GIMP or GIMP allow picking colors from
any location on-screen. In order to keep supporting this feature
on wayland and in sandboxed apps, we will expose an appropriate
method in the Screenshot interface, so first add a corresponding
method to ShellScreenshot.
A custom callback type is more convenient, but only as long as no
other callback type is required. We are about to add functionality
that does not return the filename to a screenshot saved on disk, so
prepare for that by moving to GIO's generic async callback pattern.
windowManager: Add top edge drag gesture to unmake fullscreen window
Fullscreen windows cannot be restored by touch device users unless the
application adds support for it.
As it is unlikely to change all application lets introduce a top edge
drag gesture which unmakes fullscreen windows.
gdm, util: Always allow to retry login in unlock mode
When in lockscreen mode there's no point of resetting the auth login as there's
no welcome screen, and that would just cause the UI to freeze, with no reason.
This could have been useful if we were stopping the user to login for a given
time after ALLOWED_FAILURES attempts, but this is not the case yet.
When we get a reset signal the preemptiveAnswer should be also unset or it will
be used next time the user authPrompt will be activated, even without any further
user interaction.
Jonas Ådahl [Fri, 20 Jul 2018 14:50:50 +0000 (16:50 +0200)]
Add remote access indication and control
Add an indicator for when there is something access the display server
remotely. This could be 1) remote desktop, 2) screen cast or 3) remote
control, but all effectively applications using
org.freedesktop.portal.ScreenCast or org.gnome.portal.RemoteDesktop as
well as gnome-remote-desktop using the corresponding org.gnome.Mutter
APIs directly.
As it is now, it'll simply show a single icon for when anything is
having an active session, and a single action "Turn off" that'll close
every active session.
Some systems define it to /usr/lib64 on 64-bit systems, which is not
where systemd looks for unit files. Just hardcode 'lib' in the install
prefix until we get to use the pkg-config file.
The input method may hint that certain keycodes should be pressed/released
besides the textual information in ::commit. An example is hitting space
in some IMs to commit text, where both ::commit happens, and an space is
visibly inserted. In order to handle this properly, we must honor
::forward-key-press.
In order to cater for the case that a keypress is forwarded while handling
that same keypress in a physical keyboard, check the current event being
handled and just forward it as-is if it matches. This is necessary to
prevent state from being doubly set, and the second event silenced away.
Iain Lane [Wed, 17 Jan 2018 15:56:17 +0000 (15:56 +0000)]
Add systemd units
So that we can be started by systemd --user, instead of gnome-session.
There are three units:
- gnome-shell.service: Start gnome-shell itself.
- gnome-shell-x11.target, gnome-shell-wayland.target: Sync points for
units that need to care if x11 or wayland is in use.
gnome-settings-daemon will use these, for example.
Iain Lane [Mon, 8 Jan 2018 14:23:52 +0000 (14:23 +0000)]
loginManager: Get the session ID from logind if XDG_SESSION_ID unset
If we're started by systemd, we won't be in the user's display session.
However, this is still the session that will get locked & unlocked. Ask
logind what the 'display' or 'greeter' session is, and watch for the
Unlock signal for that session to know when to unlock.
For windows, the cursor location needs to be adjusted by the frame
offsets. However we cannot assume that there is a window, as the
shell itself can have the key focus.
Florian Müllner [Thu, 24 May 2018 15:28:36 +0000 (17:28 +0200)]
panel: Allow restoring maximized/tiled windows by touch
Maximized and tiled windows can be restored with a drag gesture,
not only from their titlebars, but also from any non-reactive
parts of the top bar above the window. Currently this only works
for actual pointer devices, extend the behavior to handle touch
as well.
Xavier Johnson [Tue, 29 May 2018 04:52:24 +0000 (00:52 -0400)]
switcherPopup: Cancel window cycling with Tab
If the Escape key is used for a window/app cycler/switcher shortcut
(such as "Switch windows directly"), then there is no way to cancel
the switching/cycling operation with the keyboard.
This change allows cancelling such an operation by pressing the Tab
key, but only if Tab is not already being used by the current
switcher/cycler shortcut.
Andrea Azzarone [Thu, 21 Jun 2018 16:40:43 +0000 (18:40 +0200)]
popupMenu: Don't handle key presses directly if there are modifiers
Key events involved in a keyboard shortcut are not completely consumed by
Mutter. That means that if the popupMenu is bound to a shortcut (e.g.
Alt<Space>) and the user keeps the keys pressed, the same key-event will be
delivered to the popupMenu. We can workaround this issue filtering out all the
events where a a modifier is down (except capslock).
Florian Müllner [Sat, 9 Jun 2018 11:23:35 +0000 (13:23 +0200)]
events: Re-use event messages where possible
Destroying and recreating the entire events list on every change is not only
wasteful, it also breaks the clear functionality as messages scheduled for
removal are replaced with "new" messages after the first message has been
removed.
Address both issues by keeping track of all messages and re-use them
whenever possible.
Carlos Garnacho [Thu, 28 Jun 2018 18:22:40 +0000 (20:22 +0200)]
st-adjustment: Mark all properties as EXPLICIT_NOTIFY
All adjustment setter functions take good care of avoiding emission of
notify:: when it's not needed. The set_property() implementation already
calls into the setter functions, so mark the properties as EXPLICITY_NOTIFY
in order to optimize notify:: emission away through g_object_set (rather
common from JS code).
Carlos Garnacho [Thu, 28 Jun 2018 14:41:07 +0000 (16:41 +0200)]
st-box-layout: Queue relayout on adjustment changes
The actor allocation doesn't change per-se, but apply_transform()
will practically transform it. In order to have the paint volume
update accordingly, queue a relayout.
Carlos Garnacho [Fri, 15 Jun 2018 12:23:21 +0000 (14:23 +0200)]
st: Clip StEntry to allocation
The default get_paint_volume() implementation will do the union
of children, and the child ClutterText paint volume may expand
beyond StEntry size when text overflows.
We actually want all content to be clipped to the StEntry, so
implement get_paint_volume() and tell it so.
Carlos Garnacho [Wed, 16 May 2018 21:57:38 +0000 (23:57 +0200)]
st: Make StScrollables' paint volume reflect the unconstrained view
And constrain it in StScrollView instead (instead of falling back to an
infinite paint volume, as the actor as paint/pick impls, but no
corresponding get_paint_volume one).
Fixes artifacts with the AppView (and possibly other places) when paint
volumes are aggressively cached.
Carlos Garnacho [Fri, 25 May 2018 09:35:49 +0000 (11:35 +0200)]
keyboard: Implement standalone FocusTracker
And stop using FocusCaretTracker for caret position purposes. This
new object uses 1) the text-input protocol in wayland and 2) Info
from IBusPanelService for X11 (which is meant to work for XIM too).
This drops the usage of AtspiEventListener for OSK purposes, which
is best to avoid.