brl/mutter
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
f8e9565482
mutter/clutter/clutter-master-clock.c

413 lines
11 KiB
C
Raw Normal View History

Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
/*
* Clutter.
*
* An OpenGL based 'interactive canvas' library.
*
* Authored By: Emmanuele Bassi <ebassi@linux.intel.com>
*
* Copyright (C) 2009 Intel Corporation.
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library. If not, see <http://www.gnu.org/licenses/>.
*/
/*
* SECTION:clutter-master-clock
* @short_description: The master clock for all animations
*
* The #ClutterMasterClock class is responsible for advancing all
* #ClutterTimelines when a stage is being redrawn. The master clock
* makes sure that the scenegraph is always integrally updated before
* painting it.
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "clutter-master-clock.h"
#include "clutter-debug.h"
#include "clutter-private.h"
#define CLUTTER_MASTER_CLOCK_CLASS(klass) (G_TYPE_CHECK_CLASS_CAST ((klass), CLUTTER_TYPE_MASTER_CLOCK, ClutterMasterClockClass))
#define CLUTTER_IS_MASTER_CLOCK_CLASS(klass) (G_TYPE_CHECK_CLASS_TYPE ((klass), CLUTTER_TYPE_MASTER_CLOCK))
#define CLUTTER_MASTER_CLASS_GET_CLASS(obj) (G_TYPE_INSTANCE_GET_CLASS ((obj), CLUTTER_TYPE_MASTER_CLOCK, ClutterMasterClockClass))
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
typedef struct _ClutterClockSource ClutterClockSource;
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
typedef struct _ClutterMasterClockClass ClutterMasterClockClass;
struct _ClutterMasterClock
{
GObject parent_instance;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
/* the list of timelines handled by the clock */
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
GSList *timelines;
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
/* the current state of the clock
*/
GTimeVal cur_tick;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
/* the previous state of the clock, used to compute
* the delta
*/
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
GTimeVal prev_tick;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
/* an idle source, used by the Master Clock to queue
* a redraw on the stage and drive the animations
*/
GSource *source;
[master-clock] Throttle if no redraw was performed A flag in the master clock is now set whenever the dispatch caused an actual redraw of a stage. If this flag is not set during the prepare and check functions then it will resort to limiting the redraw attempts to the default frame rate as if vblank syncing was disabled. Otherwise if a timeline is running that does not cause the scene to change then it would busy-wait with 100% CPU until the next frame. This fix was suggested by Owen Taylor in: http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-10 14:50:27 +00:00
guint updated_stages : 1;
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
};
struct _ClutterMasterClockClass
{
GObjectClass parent_class;
};
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
struct _ClutterClockSource
{
GSource source;
ClutterMasterClock *master_clock;
};
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
static gboolean clutter_clock_prepare (GSource *source,
gint *timeout);
static gboolean clutter_clock_check (GSource *source);
static gboolean clutter_clock_dispatch (GSource *source,
GSourceFunc callback,
gpointer user_data);
static ClutterMasterClock *default_clock = NULL;
static GSourceFuncs clock_funcs = {
clutter_clock_prepare,
clutter_clock_check,
clutter_clock_dispatch,
NULL
};
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
G_DEFINE_TYPE (ClutterMasterClock, clutter_master_clock, G_TYPE_OBJECT);
[master clock] Handle Timeline::started signal correctly The "started" signal is sent first after the timeline has been set to the 'running' state. For this reason, checking if the clock has any running timelines running will always return true in the "started" signal handler: the timeline that sent the signal is running. What needs to be checked in the signal handler is if there are any timelines running other than the one that emitted the ::started signal, which we know is running anyway. This prevents frames from being lost at the beginning of an animation when a timeline is started after a quiescent period. Fixes bug: http://bugzilla.openedhand.com/show_bug.cgi?id=1617 Signed-off-by: Jonas Bonn <jonas@southpole.se> Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-05-29 13:25:19 +00:00
/*
Remove stage update idle and do updates from the master clock When a redraw is queued on a stage, simply set a flag; then in the check/prepare functions of the master clock source, check for stages that need redrawing. This avoids the complexity of having multiple competing sources at the same priority and makes the update ordering more reliable and understandable. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 22:22:51 +00:00
* master_clock_is_running:
[master clock] Handle Timeline::started signal correctly The "started" signal is sent first after the timeline has been set to the 'running' state. For this reason, checking if the clock has any running timelines running will always return true in the "started" signal handler: the timeline that sent the signal is running. What needs to be checked in the signal handler is if there are any timelines running other than the one that emitted the ::started signal, which we know is running anyway. This prevents frames from being lost at the beginning of an animation when a timeline is started after a quiescent period. Fixes bug: http://bugzilla.openedhand.com/show_bug.cgi?id=1617 Signed-off-by: Jonas Bonn <jonas@southpole.se> Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-05-29 13:25:19 +00:00
* @master_clock: a #ClutterMasterClock
*
Remove stage update idle and do updates from the master clock When a redraw is queued on a stage, simply set a flag; then in the check/prepare functions of the master clock source, check for stages that need redrawing. This avoids the complexity of having multiple competing sources at the same priority and makes the update ordering more reliable and understandable. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 22:22:51 +00:00
* Checks if we should currently be advancing timelines or redrawing
* stages.
[master clock] Handle Timeline::started signal correctly The "started" signal is sent first after the timeline has been set to the 'running' state. For this reason, checking if the clock has any running timelines running will always return true in the "started" signal handler: the timeline that sent the signal is running. What needs to be checked in the signal handler is if there are any timelines running other than the one that emitted the ::started signal, which we know is running anyway. This prevents frames from being lost at the beginning of an animation when a timeline is started after a quiescent period. Fixes bug: http://bugzilla.openedhand.com/show_bug.cgi?id=1617 Signed-off-by: Jonas Bonn <jonas@southpole.se> Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-05-29 13:25:19 +00:00
*
* Return value: %TRUE if the #ClutterMasterClock has at least
* one running timeline
*/
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
static gboolean
Remove stage update idle and do updates from the master clock When a redraw is queued on a stage, simply set a flag; then in the check/prepare functions of the master clock source, check for stages that need redrawing. This avoids the complexity of having multiple competing sources at the same priority and makes the update ordering more reliable and understandable. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 22:22:51 +00:00
master_clock_is_running (ClutterMasterClock *master_clock)
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
{
Simplify timeout list handling for the master clock Instead of keeping a list of all timelines, and connecting to signals and weak notifies, simply keep a list of running timelines; this greatly simplifies both the book-keeping, and also determining if there are any running timelines. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 21:44:40 +00:00
ClutterStageManager *stage_manager = clutter_stage_manager_get_default ();
const GSList *stages, *l;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
Simplify timeout list handling for the master clock Instead of keeping a list of all timelines, and connecting to signals and weak notifies, simply keep a list of running timelines; this greatly simplifies both the book-keeping, and also determining if there are any running timelines. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 21:44:40 +00:00
if (master_clock->timelines)
return TRUE;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
Remove stage update idle and do updates from the master clock When a redraw is queued on a stage, simply set a flag; then in the check/prepare functions of the master clock source, check for stages that need redrawing. This avoids the complexity of having multiple competing sources at the same priority and makes the update ordering more reliable and understandable. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 22:22:51 +00:00
stages = clutter_stage_manager_peek_stages (stage_manager);
for (l = stages; l; l = l->next)
Compress events as part of the frame cycle Instead of trying to guess about which motion events are extraneous, queue up all events until we process a frame. This allows us to look ahead and reliably compress consecutive sequence of motion events. clutter-main.c: Feed received events to the stage for queueing. Remove old compression code. Remove clutter_get_motion_events_frequency() clutter_set_motion_events_frequency() clutter-stage.c: Keep a queue of pending events. clutter-master-clock.c: Add processng of queued events to the clock source dispatch function. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 23:10:41 +00:00
if (_clutter_stage_has_queued_events (l->data) ||
_clutter_stage_needs_update (l->data))
Remove stage update idle and do updates from the master clock When a redraw is queued on a stage, simply set a flag; then in the check/prepare functions of the master clock source, check for stages that need redrawing. This avoids the complexity of having multiple competing sources at the same priority and makes the update ordering more reliable and understandable. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 22:22:51 +00:00
return TRUE;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
return FALSE;
}
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
/*
* master_clock_next_frame_delay:
* @master_clock: a #ClutterMasterClock
*
* Computes the number of delay before we need to draw the next frame.
*
* Return value: -1 if there is no next frame pending, otherwise the
* number of millseconds before the we need to draw the next frame
*/
static gint
master_clock_next_frame_delay (ClutterMasterClock *master_clock)
{
GTimeVal now;
GTimeVal next;
if (!master_clock_is_running (master_clock))
return -1;
[master-clock] Throttle if no redraw was performed A flag in the master clock is now set whenever the dispatch caused an actual redraw of a stage. If this flag is not set during the prepare and check functions then it will resort to limiting the redraw attempts to the default frame rate as if vblank syncing was disabled. Otherwise if a timeline is running that does not cause the scene to change then it would busy-wait with 100% CPU until the next frame. This fix was suggested by Owen Taylor in: http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-10 14:50:27 +00:00
if (clutter_feature_available (CLUTTER_FEATURE_SYNC_TO_VBLANK) &&
master_clock->updated_stages)
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
{
/* When we have sync-to-vblank, we count on that to throttle
* our frame rate, and otherwise draw frames as fast as possible.
*/
return 0;
}
if (master_clock->prev_tick.tv_sec == 0)
{
/* If we weren't previously running, then draw the next frame
* immediately
*/
return 0;
}
/* Otherwise, wait at least 1/frame_rate seconds since we last started a frame */
g_source_get_current_time (master_clock->source, &now);
next = master_clock->prev_tick;
g_time_val_add (&next, 1000000 / clutter_get_default_frame_rate ());
if (next.tv_sec < now.tv_sec ||
(next.tv_sec == now.tv_sec && next.tv_usec < now.tv_usec))
{
return 0;
}
else
{
return ((next.tv_sec - now.tv_sec) * 1000 +
(next.tv_usec - now.tv_usec) / 1000);
}
}
[master clock] Handle Timeline::started signal correctly The "started" signal is sent first after the timeline has been set to the 'running' state. For this reason, checking if the clock has any running timelines running will always return true in the "started" signal handler: the timeline that sent the signal is running. What needs to be checked in the signal handler is if there are any timelines running other than the one that emitted the ::started signal, which we know is running anyway. This prevents frames from being lost at the beginning of an animation when a timeline is started after a quiescent period. Fixes bug: http://bugzilla.openedhand.com/show_bug.cgi?id=1617 Signed-off-by: Jonas Bonn <jonas@southpole.se> Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-05-29 13:25:19 +00:00
/*
* clutter_clock_source_new:
* @master_clock: a #ClutterMasterClock for the source
*
* The #ClutterClockSource is an idle GSource that will queue a redraw
* if @master_clock has at least a running #ClutterTimeline. The redraw
* will cause @master_clock to advance all timelines, thus advancing all
* animations as well.
*
* Return value: the newly created #GSource
*/
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
static GSource *
clutter_clock_source_new (ClutterMasterClock *master_clock)
{
GSource *source = g_source_new (&clock_funcs, sizeof (ClutterClockSource));
ClutterClockSource *clock_source = (ClutterClockSource *) source;
clock_source->master_clock = master_clock;
return source;
}
static gboolean
clutter_clock_prepare (GSource *source,
gint *timeout)
{
ClutterClockSource *clock_source = (ClutterClockSource *) source;
ClutterMasterClock *master_clock = clock_source->master_clock;
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
int delay;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
delay = master_clock_next_frame_delay (master_clock);
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
*timeout = delay;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
return delay == 0;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
}
static gboolean
clutter_clock_check (GSource *source)
{
ClutterClockSource *clock_source = (ClutterClockSource *) source;
ClutterMasterClock *master_clock = clock_source->master_clock;
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
int delay;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
delay = master_clock_next_frame_delay (master_clock);
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
return delay == 0;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
}
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
static gboolean
clutter_clock_dispatch (GSource *source,
GSourceFunc callback,
gpointer user_data)
{
Remove stage update idle and do updates from the master clock When a redraw is queued on a stage, simply set a flag; then in the check/prepare functions of the master clock source, check for stages that need redrawing. This avoids the complexity of having multiple competing sources at the same priority and makes the update ordering more reliable and understandable. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 22:22:51 +00:00
ClutterClockSource *clock_source = (ClutterClockSource *) source;
ClutterMasterClock *master_clock = clock_source->master_clock;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
ClutterStageManager *stage_manager = clutter_stage_manager_get_default ();
Compress events as part of the frame cycle Instead of trying to guess about which motion events are extraneous, queue up all events until we process a frame. This allows us to look ahead and reliably compress consecutive sequence of motion events. clutter-main.c: Feed received events to the stage for queueing. Remove old compression code. Remove clutter_get_motion_events_frequency() clutter_set_motion_events_frequency() clutter-stage.c: Keep a queue of pending events. clutter-master-clock.c: Add processng of queued events to the clock source dispatch function. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 23:10:41 +00:00
GSList *stages, *l;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
CLUTTER_NOTE (SCHEDULER, "Master clock [tick]");
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
/* Get the time to use for this frame.
*/
g_source_get_current_time (source, &master_clock->cur_tick);
Compress events as part of the frame cycle Instead of trying to guess about which motion events are extraneous, queue up all events until we process a frame. This allows us to look ahead and reliably compress consecutive sequence of motion events. clutter-main.c: Feed received events to the stage for queueing. Remove old compression code. Remove clutter_get_motion_events_frequency() clutter_set_motion_events_frequency() clutter-stage.c: Keep a queue of pending events. clutter-master-clock.c: Add processng of queued events to the clock source dispatch function. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 23:10:41 +00:00
/* We need to protect ourselves against stages being destroyed during
* event handling
*/
stages = clutter_stage_manager_list_stages (stage_manager);
g_slist_foreach (stages, (GFunc)g_object_ref, NULL);
[master-clock] Throttle if no redraw was performed A flag in the master clock is now set whenever the dispatch caused an actual redraw of a stage. If this flag is not set during the prepare and check functions then it will resort to limiting the redraw attempts to the default frame rate as if vblank syncing was disabled. Otherwise if a timeline is running that does not cause the scene to change then it would busy-wait with 100% CPU until the next frame. This fix was suggested by Owen Taylor in: http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-10 14:50:27 +00:00
master_clock->updated_stages = FALSE;
Compress events as part of the frame cycle Instead of trying to guess about which motion events are extraneous, queue up all events until we process a frame. This allows us to look ahead and reliably compress consecutive sequence of motion events. clutter-main.c: Feed received events to the stage for queueing. Remove old compression code. Remove clutter_get_motion_events_frequency() clutter_set_motion_events_frequency() clutter-stage.c: Keep a queue of pending events. clutter-master-clock.c: Add processng of queued events to the clock source dispatch function. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 23:10:41 +00:00
/* Process queued events
*/
for (l = stages; l != NULL; l = l->next)
_clutter_stage_process_queued_events (l->data);
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
Run the repaint functions inside the redraw cycle Now that every redraw is performed within the master clock we need to run the repaint functions inside it.
2009-06-09 15:28:25 +00:00
_clutter_master_clock_advance (master_clock);
_clutter_run_repaint_functions ();
Remove stage update idle and do updates from the master clock When a redraw is queued on a stage, simply set a flag; then in the check/prepare functions of the master clock source, check for stages that need redrawing. This avoids the complexity of having multiple competing sources at the same priority and makes the update ordering more reliable and understandable. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 22:22:51 +00:00
/* Update any stage that needs redraw/relayout after the clock
* is advanced.
[master clock] Handle Timeline::started signal correctly The "started" signal is sent first after the timeline has been set to the 'running' state. For this reason, checking if the clock has any running timelines running will always return true in the "started" signal handler: the timeline that sent the signal is running. What needs to be checked in the signal handler is if there are any timelines running other than the one that emitted the ::started signal, which we know is running anyway. This prevents frames from being lost at the beginning of an animation when a timeline is started after a quiescent period. Fixes bug: http://bugzilla.openedhand.com/show_bug.cgi?id=1617 Signed-off-by: Jonas Bonn <jonas@southpole.se> Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-05-29 13:25:19 +00:00
*/
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
for (l = stages; l != NULL; l = l->next)
[master-clock] Throttle if no redraw was performed A flag in the master clock is now set whenever the dispatch caused an actual redraw of a stage. If this flag is not set during the prepare and check functions then it will resort to limiting the redraw attempts to the default frame rate as if vblank syncing was disabled. Otherwise if a timeline is running that does not cause the scene to change then it would busy-wait with 100% CPU until the next frame. This fix was suggested by Owen Taylor in: http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-10 14:50:27 +00:00
master_clock->updated_stages |= _clutter_stage_do_update (l->data);
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
Compress events as part of the frame cycle Instead of trying to guess about which motion events are extraneous, queue up all events until we process a frame. This allows us to look ahead and reliably compress consecutive sequence of motion events. clutter-main.c: Feed received events to the stage for queueing. Remove old compression code. Remove clutter_get_motion_events_frequency() clutter_set_motion_events_frequency() clutter-stage.c: Keep a queue of pending events. clutter-master-clock.c: Add processng of queued events to the clock source dispatch function. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 23:10:41 +00:00
g_slist_foreach (stages, (GFunc)g_object_unref, NULL);
g_slist_free (stages);
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
master_clock->prev_tick = master_clock->cur_tick;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
return TRUE;
}
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
static void
clutter_master_clock_finalize (GObject *gobject)
{
ClutterMasterClock *master_clock = CLUTTER_MASTER_CLOCK (gobject);
g_slist_free (master_clock->timelines);
G_OBJECT_CLASS (clutter_master_clock_parent_class)->finalize (gobject);
}
static void
clutter_master_clock_class_init (ClutterMasterClockClass *klass)
{
GObjectClass *gobject_class = G_OBJECT_CLASS (klass);
gobject_class->finalize = clutter_master_clock_finalize;
}
static void
clutter_master_clock_init (ClutterMasterClock *self)
{
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
GSource *source;
source = clutter_clock_source_new (self);
self->source = source;
[master-clock] Throttle if no redraw was performed A flag in the master clock is now set whenever the dispatch caused an actual redraw of a stage. If this flag is not set during the prepare and check functions then it will resort to limiting the redraw attempts to the default frame rate as if vblank syncing was disabled. Otherwise if a timeline is running that does not cause the scene to change then it would busy-wait with 100% CPU until the next frame. This fix was suggested by Owen Taylor in: http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-10 14:50:27 +00:00
self->updated_stages = TRUE;
[clock] Rework the master clock The master clock is currently advanced using a frame source driven by the default frame rate. This breaks the sync to vblank because the vblanking rate could be different than 60 Hz -- or it might be completely disabled (e.g. with CLUTTER_VBLANK=none). We should be using the main loop to check if we have timelines playing, and if so queue a redraw on the stages we own. We should also prepare the subsequent frame at the end of the redraw process, so if there are new redraw we will have the scene already in place. This makes Clutter redraw at the maximum frame rate, which is limited by the vblanking frequency.
2009-05-07 18:07:21 +00:00
g_source_set_priority (source, CLUTTER_PRIORITY_REDRAW);
g_source_set_can_recurse (source, FALSE);
g_source_attach (source, NULL);
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
}
/*
* _clutter_master_clock_get_default:
*
* Retrieves the default master clock. If this function has never
* been called before, the default master clock is created.
*
* Return value: the default master clock. The returned object is
* owned by Clutter and should not be modified or freed
*/
ClutterMasterClock *
_clutter_master_clock_get_default (void)
{
if (G_LIKELY (default_clock != NULL))
return default_clock;
default_clock = g_object_new (CLUTTER_TYPE_MASTER_CLOCK, NULL);
return default_clock;
}
/*
* _clutter_master_clock_add_timeline:
* @master_clock: a #ClutterMasterClock
* @timeline: a #ClutterTimeline
*
Simplify timeout list handling for the master clock Instead of keeping a list of all timelines, and connecting to signals and weak notifies, simply keep a list of running timelines; this greatly simplifies both the book-keeping, and also determining if there are any running timelines. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 21:44:40 +00:00
* Adds @timeline to the list of playing timelines held by the master
* clock.
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
*/
void
_clutter_master_clock_add_timeline (ClutterMasterClock *master_clock,
ClutterTimeline *timeline)
{
Simplify timeout list handling for the master clock Instead of keeping a list of all timelines, and connecting to signals and weak notifies, simply keep a list of running timelines; this greatly simplifies both the book-keeping, and also determining if there are any running timelines. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 21:44:40 +00:00
gboolean is_first;
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
if (g_slist_find (master_clock->timelines, timeline))
return;
Simplify timeout list handling for the master clock Instead of keeping a list of all timelines, and connecting to signals and weak notifies, simply keep a list of running timelines; this greatly simplifies both the book-keeping, and also determining if there are any running timelines. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 21:44:40 +00:00
is_first = master_clock->timelines == NULL;
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
master_clock->timelines = g_slist_prepend (master_clock->timelines,
timeline);
Simplify timeout list handling for the master clock Instead of keeping a list of all timelines, and connecting to signals and weak notifies, simply keep a list of running timelines; this greatly simplifies both the book-keeping, and also determining if there are any running timelines. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 21:44:40 +00:00
if (is_first)
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
_clutter_master_clock_start_running (master_clock);
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
}
/*
* _clutter_master_clock_remove_timeline:
* @master_clock: a #ClutterMasterClock
* @timeline: a #ClutterTimeline
*
Simplify timeout list handling for the master clock Instead of keeping a list of all timelines, and connecting to signals and weak notifies, simply keep a list of running timelines; this greatly simplifies both the book-keeping, and also determining if there are any running timelines. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-06 21:44:40 +00:00
* Removes @timeline from the list of playing timelines held by the
* master clock.
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
*/
void
_clutter_master_clock_remove_timeline (ClutterMasterClock *master_clock,
ClutterTimeline *timeline)
{
master_clock->timelines = g_slist_remove (master_clock->timelines,
timeline);
Limit the frame rate when not syncing to VBLANK clutter-master-clock.c clutter-master-clock.h: When the SYNC_TO_VBLANK feature is not available, wait for 1/frame_rate seconds since the start of the last frame before drawing the next frame. Add _clutter_master_clock_start_running() to abstract the usage of g_main_context_wakeup() clutter-stage.c: Add _clutter_master_clock_start_running() clutter-main.c: Update docs for clutter_set_default_frame_rate() clutter_get_default_frame_rate() to no longer talk about timeline frame rates. test-text-perf.c test-text.c: Set a frame rate of 1000fps so that frame-rate limiting doesn't affect the result. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-07 02:48:15 +00:00
}
/*
* _clutter_master_clock_start_running:
* @master_clock: a #ClutterMasterClock
*
* Called when we have events or redraws to process; if the clock
* is stopped, does the processing necessary to wake it up again.
*/
void
_clutter_master_clock_start_running (ClutterMasterClock *master_clock)
{
/* If called from a different thread, we need to wake up the
* main loop to start running the timelines
*/
g_main_context_wakeup (NULL);
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
}
/*
* _clutter_master_clock_advance:
* @master_clock: a #ClutterMasterClock
*
* Advances all the timelines held by the master clock. This function
* should be called before calling clutter_redraw() to make sure that
* all the timelines are advanced and the scene is updated.
*/
void
_clutter_master_clock_advance (ClutterMasterClock *master_clock)
{
Move elapsed-time calculations into ClutterTimeline Instead of calculating a delta in the master clock, and passing that into each timeline, make each timeline individually responsible for remembering the last time and computing the delta. This: - Fixes a problem where we could spin infinitely processing timeline-only frames with < 1msec differences. - Makes timelines consistently start timing on the first frame; instead of doing different things for the first started timeline and other timelines. - Improves accuracy of elapsed time computations by avoiding accumulating microsecond => millisecond truncation errors. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-08 17:00:09 +00:00
GSList *l, *next;
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
g_return_if_fail (CLUTTER_IS_MASTER_CLOCK (master_clock));
Move elapsed-time calculations into ClutterTimeline Instead of calculating a delta in the master clock, and passing that into each timeline, make each timeline individually responsible for remembering the last time and computing the delta. This: - Fixes a problem where we could spin infinitely processing timeline-only frames with < 1msec differences. - Makes timelines consistently start timing on the first frame; instead of doing different things for the first started timeline and other timelines. - Improves accuracy of elapsed time computations by avoiding accumulating microsecond => millisecond truncation errors. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-08 17:00:09 +00:00
for (l = master_clock->timelines; l != NULL; l = next)
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
{
Move elapsed-time calculations into ClutterTimeline Instead of calculating a delta in the master clock, and passing that into each timeline, make each timeline individually responsible for remembering the last time and computing the delta. This: - Fixes a problem where we could spin infinitely processing timeline-only frames with < 1msec differences. - Makes timelines consistently start timing on the first frame; instead of doing different things for the first started timeline and other timelines. - Improves accuracy of elapsed time computations by avoiding accumulating microsecond => millisecond truncation errors. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-08 17:00:09 +00:00
next = l->next;
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
Move elapsed-time calculations into ClutterTimeline Instead of calculating a delta in the master clock, and passing that into each timeline, make each timeline individually responsible for remembering the last time and computing the delta. This: - Fixes a problem where we could spin infinitely processing timeline-only frames with < 1msec differences. - Makes timelines consistently start timing on the first frame; instead of doing different things for the first started timeline and other timelines. - Improves accuracy of elapsed time computations by avoiding accumulating microsecond => millisecond truncation errors. http://bugzilla.openedhand.com/show_bug.cgi?id=1637 Signed-off-by: Emmanuele Bassi <ebassi@linux.intel.com>
2009-06-08 17:00:09 +00:00
clutter_timeline_do_tick (l->data, &master_clock->cur_tick);
Use a single master "clock" to drive timelines Currently, all timelines install a timeout inside the TimeoutPool they share. Every time the main loop spins, all the timeouts are updated. This, in turn, will usually lead to redraws being queued on the stages. This behaviour leads to the potential starvation of timelines and to excessive redraws. One lesson learned from the games developers is that the scenegraph should be prepared in its entirety before the GL paint sequence is initiated. This means making sure that every ::new-frame signal handler is called before clutter_redraw() is invoked. In order to do so a TimeoutPool is not enough: we need a master clock. The clock will be responsible for advancing all the active timelines created inside a scene, but only when the stage is being redrawn. The sequence is: + queue_redraw() is invoked on an actor and bubbles up to the stage + if no redraw() has already been scheduled, install an idle handler with a known priority + inside the idle handler: - advance the master clock, which will in turn advance every playing timeline by the amount of milliseconds elapsed since the last redraw; this will make every playing timeline emit the ::new-frame signal - queue a relayout - call the redraw() method of the backend This way we trade multiple timeouts with a single frame source that only runs if a timeline is playing and queues redraws on the various stages.
2009-03-21 20:39:32 +00:00
}
}
Reference in New Issue Copy Permalink