mutter/clutter/glx/clutter-backend-glx.c

846 lines
27 KiB
C
Raw Normal View History

2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
/* Clutter.
* An OpenGL based 'interactive canvas' library.
* Authored By Matthew Allum <mallum@openedhand.com>
* Copyright (C) 2006-2007 OpenedHand
*
* 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/>.
*
*
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include <string.h>
#include <sys/types.h>
#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
#include <fcntl.h>
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
#include <glib/gi18n-lib.h>
#include <GL/glx.h>
#include <GL/glxext.h>
#include <GL/gl.h>
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
#include "clutter-backend-glx.h"
#include "clutter-event-glx.h"
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
#include "clutter-stage-glx.h"
#include "clutter-glx.h"
2009-03-30 11:41:02 -04:00
#include "clutter-profile.h"
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
#include "../clutter-event.h"
#include "../clutter-main.h"
#include "../clutter-debug.h"
#include "../clutter-private.h"
#include "../clutter-version.h"
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
#include "cogl/cogl.h"
2009-03-30 11:41:02 -04:00
G_DEFINE_TYPE (ClutterBackendGLX, clutter_backend_glx, CLUTTER_TYPE_BACKEND_X11);
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
/* singleton object */
static ClutterBackendGLX *backend_singleton = NULL;
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
static gchar *clutter_vblank_name = NULL;
G_CONST_RETURN gchar*
clutter_backend_glx_get_vblank_method (void)
{
return clutter_vblank_name;
}
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
static gboolean
clutter_backend_glx_pre_parse (ClutterBackend *backend,
GError **error)
{
const gchar *env_string;
env_string = g_getenv ("CLUTTER_VBLANK");
if (env_string)
{
clutter_vblank_name = g_strdup (env_string);
env_string = NULL;
}
return clutter_backend_x11_pre_parse (backend, error);
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
}
static gboolean
clutter_backend_glx_post_parse (ClutterBackend *backend,
GError **error)
{
ClutterBackendX11 *backend_x11 = CLUTTER_BACKEND_X11 (backend);
ClutterBackendGLX *backend_glx = CLUTTER_BACKEND_GLX (backend);
ClutterBackendClass *backend_class =
CLUTTER_BACKEND_CLASS (clutter_backend_glx_parent_class);
int glx_major, glx_minor;
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
if (!backend_class->post_parse (backend, error))
return FALSE;
if (!glXQueryExtension (backend_x11->xdpy,
&backend_glx->error_base,
&backend_glx->event_base))
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
{
g_set_error (error, CLUTTER_INIT_ERROR,
CLUTTER_INIT_ERROR_BACKEND,
"XServer appears to lack required GLX support");
return FALSE;
}
/* XXX: Technically we should require >= GLX 1.3 support but for a long
* time Mesa has exported a hybrid GLX, exporting extensions specified
* to require GLX 1.3, but still reporting 1.2 via glXQueryVersion. */
if (!glXQueryVersion (backend_x11->xdpy, &glx_major, &glx_minor)
|| !(glx_major == 1 && glx_minor >= 2))
{
g_set_error (error, CLUTTER_INIT_ERROR,
CLUTTER_INIT_ERROR_BACKEND,
"XServer appears to lack required GLX 1.2 support");
return FALSE;
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
}
return TRUE;
}
static const GOptionEntry entries[] =
{
{ "vblank", 0,
0,
G_OPTION_ARG_STRING, &clutter_vblank_name,
N_("VBlank method to be used (none, dri or glx)"), "METHOD"
},
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
{ NULL }
};
static void
clutter_backend_glx_add_options (ClutterBackend *backend,
GOptionGroup *group)
{
g_option_group_add_entries (group, entries);
clutter_backend_x11_add_options (backend, group);
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
}
static void
clutter_backend_glx_finalize (GObject *gobject)
{
if (backend_singleton)
backend_singleton = NULL;
G_OBJECT_CLASS (clutter_backend_glx_parent_class)->finalize (gobject);
}
static void
clutter_backend_glx_dispose (GObject *gobject)
{
ClutterBackendGLX *backend_glx = CLUTTER_BACKEND_GLX (gobject);
ClutterBackendX11 *backend_x11 = CLUTTER_BACKEND_X11 (gobject);
/* Unrealize all shaders, since the GL context is going away */
_clutter_shader_release_all ();
if (backend_glx->gl_context)
{
glXMakeContextCurrent (backend_x11->xdpy, None, None, NULL);
glXDestroyContext (backend_x11->xdpy, backend_glx->gl_context);
backend_glx->gl_context = NULL;
}
if (backend_glx->dummy_glxwin)
{
glXDestroyWindow (backend_x11->xdpy, backend_glx->dummy_glxwin);
backend_glx->dummy_glxwin = None;
}
if (backend_glx->dummy_xwin)
{
XDestroyWindow (backend_x11->xdpy, backend_glx->dummy_xwin);
backend_glx->dummy_xwin = None;
}
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
G_OBJECT_CLASS (clutter_backend_glx_parent_class)->dispose (gobject);
}
static GObject *
clutter_backend_glx_constructor (GType gtype,
guint n_params,
GObjectConstructParam *params)
{
GObjectClass *parent_class;
GObject *retval;
if (!backend_singleton)
{
parent_class = G_OBJECT_CLASS (clutter_backend_glx_parent_class);
retval = parent_class->constructor (gtype, n_params, params);
backend_singleton = CLUTTER_BACKEND_GLX (retval);
return retval;
}
g_warning ("Attempting to create a new backend object. This should "
"never happen, so we return the singleton instance.");
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
return g_object_ref (backend_singleton);
}
static gboolean
check_vblank_env (const char *name)
{
if (clutter_vblank_name && !g_ascii_strcasecmp (clutter_vblank_name, name))
return TRUE;
return FALSE;
}
static ClutterFeatureFlags
clutter_backend_glx_get_features (ClutterBackend *backend)
{
ClutterBackendGLX *backend_glx = CLUTTER_BACKEND_GLX (backend);
const gchar *glx_extensions = NULL;
const gchar *gl_extensions = NULL;
ClutterFeatureFlags flags;
gboolean use_dri = FALSE;
flags = clutter_backend_x11_get_features (backend);
flags |= CLUTTER_FEATURE_STAGE_MULTIPLE;
/* this will make sure that the GL context exists */
g_assert (backend_glx->gl_context != NULL);
g_assert (glXGetCurrentDrawable () != None);
CLUTTER_NOTE (BACKEND,
"Checking features\n"
" GL_VENDOR: %s\n"
" GL_RENDERER: %s\n"
" GL_VERSION: %s\n"
" GL_EXTENSIONS: %s",
glGetString (GL_VENDOR),
glGetString (GL_RENDERER),
glGetString (GL_VERSION),
glGetString (GL_EXTENSIONS));
glx_extensions =
glXQueryExtensionsString (clutter_x11_get_default_display (),
clutter_x11_get_default_screen ());
CLUTTER_NOTE (BACKEND, " GLX Extensions: %s", glx_extensions);
gl_extensions = (const gchar *)glGetString (GL_EXTENSIONS);
/* When using glBlitFramebuffer or glXCopySubBufferMESA for sub stage
* redraws, we cannot rely on glXSwapIntervalSGI to throttle the blits
* so we need to resort to manually synchronizing with the vblank so we
* always check for the video_sync extension...
*/
if (_cogl_check_extension ("GLX_SGI_video_sync", glx_extensions) &&
/* Note: the GLX_SGI_video_sync spec explicitly states this extension
* only works for direct contexts. */
glXIsDirect (clutter_x11_get_default_display (),
backend_glx->gl_context))
{
backend_glx->get_video_sync =
(GetVideoSyncProc) cogl_get_proc_address ("glXGetVideoSyncSGI");
backend_glx->wait_video_sync =
(WaitVideoSyncProc) cogl_get_proc_address ("glXWaitVideoSyncSGI");
}
use_dri = check_vblank_env ("dri");
/* First check for explicit disabling or it set elsewhere (eg NVIDIA) */
if (check_vblank_env ("none"))
{
CLUTTER_NOTE (BACKEND, "vblank sync: disabled at user request");
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
goto vblank_setup_done;
}
if (g_getenv ("__GL_SYNC_TO_VBLANK") != NULL)
{
backend_glx->vblank_type = CLUTTER_VBLANK_GLX_SWAP;
flags |= CLUTTER_FEATURE_SYNC_TO_VBLANK;
CLUTTER_NOTE (BACKEND, "Using __GL_SYNC_TO_VBLANK hint");
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
goto vblank_setup_done;
}
/* We try two GL vblank syncing mechanisms.
* glXSwapIntervalSGI is tried first, then glXGetVideoSyncSGI.
*
* glXSwapIntervalSGI is known to work with Mesa and in particular
* the Intel drivers. glXGetVideoSyncSGI has serious problems with
* Intel drivers causing terrible frame rate so it only tried as a
* fallback.
*
* How well glXGetVideoSyncSGI works with other driver (ATI etc) needs
* to be investigated. glXGetVideoSyncSGI on ATI at least seems to have
* no effect.
*/
if (!use_dri &&
_cogl_check_extension ("GLX_SGI_swap_control", glx_extensions))
{
backend_glx->swap_interval =
(SwapIntervalProc) cogl_get_proc_address ("glXSwapIntervalSGI");
CLUTTER_NOTE (BACKEND, "attempting glXSwapIntervalSGI vblank setup");
if (backend_glx->swap_interval != NULL &&
backend_glx->swap_interval (1) == 0)
{
backend_glx->vblank_type = CLUTTER_VBLANK_GLX_SWAP;
flags |= CLUTTER_FEATURE_SYNC_TO_VBLANK;
CLUTTER_NOTE (BACKEND, "glXSwapIntervalSGI setup success");
#ifdef GLX_INTEL_swap_event
/* GLX_INTEL_swap_event allows us to avoid blocking the CPU
* while we wait for glXSwapBuffers to complete, and instead
* we get an X event notifying us of completion...
*/
if (!(clutter_paint_debug_flags & CLUTTER_DEBUG_DISABLE_SWAP_EVENTS) &&
_cogl_check_extension ("GLX_INTEL_swap_event", glx_extensions))
{
flags |= CLUTTER_FEATURE_SWAP_EVENTS;
}
#endif /* GLX_INTEL_swap_event */
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
goto vblank_setup_done;
}
CLUTTER_NOTE (BACKEND, "glXSwapIntervalSGI vblank setup failed");
}
if (!use_dri &&
!(flags & CLUTTER_FEATURE_SYNC_TO_VBLANK) &&
_cogl_check_extension ("GLX_SGI_video_sync", glx_extensions))
{
CLUTTER_NOTE (BACKEND, "attempting glXGetVideoSyncSGI vblank setup");
if ((backend_glx->get_video_sync != NULL) &&
(backend_glx->wait_video_sync != NULL))
{
CLUTTER_NOTE (BACKEND, "glXGetVideoSyncSGI vblank setup success");
backend_glx->vblank_type = CLUTTER_VBLANK_GLX;
flags |= CLUTTER_FEATURE_SYNC_TO_VBLANK;
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
goto vblank_setup_done;
}
CLUTTER_NOTE (BACKEND, "glXGetVideoSyncSGI vblank setup failed");
}
#ifdef __linux__
/*
* DRI is really an extreme fallback -rumoured to work with Via chipsets
*/
if (!(flags & CLUTTER_FEATURE_SYNC_TO_VBLANK))
{
CLUTTER_NOTE (BACKEND, "attempting DRI vblank setup");
backend_glx->dri_fd = open("/dev/dri/card0", O_RDWR);
if (backend_glx->dri_fd >= 0)
{
CLUTTER_NOTE (BACKEND, "DRI vblank setup success");
backend_glx->vblank_type = CLUTTER_VBLANK_DRI;
flags |= CLUTTER_FEATURE_SYNC_TO_VBLANK;
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
goto vblank_setup_done;
}
CLUTTER_NOTE (BACKEND, "DRI vblank setup failed");
}
#endif /* __linux__ */
CLUTTER_NOTE (BACKEND, "no use-able vblank mechanism found");
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
vblank_setup_done:
if (_cogl_check_extension ("GLX_MESA_copy_sub_buffer", glx_extensions))
{
backend_glx->copy_sub_buffer =
(CopySubBufferProc) cogl_get_proc_address ("glXCopySubBufferMESA");
backend_glx->can_blit_sub_buffer = TRUE;
backend_glx->blit_sub_buffer_is_synchronized = TRUE;
}
else if (_cogl_check_extension ("GL_EXT_framebuffer_blit", gl_extensions))
{
CLUTTER_NOTE (BACKEND,
"Using glBlitFramebuffer fallback for sub_buffer copies");
backend_glx->blit_framebuffer =
(BlitFramebufferProc) cogl_get_proc_address ("glBlitFramebuffer");
backend_glx->can_blit_sub_buffer = TRUE;
backend_glx->blit_sub_buffer_is_synchronized = FALSE;
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
}
CLUTTER_NOTE (BACKEND, "backend features checked");
return flags;
}
/* It seems the GLX spec never defined an invalid GLXFBConfig that
* we could overload as an indication of error, so we have to return
* an explicit boolean status. */
gboolean
_clutter_backend_glx_get_fbconfig (ClutterBackendGLX *backend_glx,
GLXFBConfig *config)
{
ClutterBackendX11 *backend_x11 = CLUTTER_BACKEND_X11 (backend_glx);
GLXFBConfig *configs = NULL;
gboolean use_argb = clutter_x11_get_use_argb_visual ();
int n_configs, i;
static const int attributes[] = {
GLX_DRAWABLE_TYPE, GLX_WINDOW_BIT,
GLX_RENDER_TYPE, GLX_RGBA_BIT,
GLX_DOUBLEBUFFER, GL_TRUE,
GLX_RED_SIZE, 1,
GLX_GREEN_SIZE, 1,
GLX_BLUE_SIZE, 1,
GLX_ALPHA_SIZE, 1,
GLX_DEPTH_SIZE, 1,
GLX_STENCIL_SIZE, 1,
None
};
if (backend_x11->xdpy == NULL || backend_x11->xscreen == NULL)
return FALSE;
/* If we don't already have a cached config then try to get one */
if (!backend_glx->found_fbconfig)
{
CLUTTER_NOTE (BACKEND,
"Retrieving GL fbconfig, dpy: %p, xscreen; %p (%d)",
backend_x11->xdpy,
backend_x11->xscreen,
backend_x11->xscreen_num);
configs = glXChooseFBConfig (backend_x11->xdpy,
backend_x11->xscreen_num,
attributes,
&n_configs);
if (configs)
{
if (use_argb)
{
for (i = 0; i < n_configs; i++)
{
XVisualInfo *vinfo;
vinfo = glXGetVisualFromFBConfig (backend_x11->xdpy,
configs[i]);
if (vinfo == NULL)
continue;
if (vinfo->depth == 32 &&
(vinfo->red_mask == 0xff0000 &&
vinfo->green_mask == 0x00ff00 &&
vinfo->blue_mask == 0x0000ff))
{
CLUTTER_NOTE (BACKEND,
"Found GLX visual ARGB [index:%d]", i);
backend_glx->found_fbconfig = TRUE;
backend_glx->fbconfig = configs[i];
goto out;
}
}
/* If we make it here then we didn't find an RGBA config so
we'll fall back to using an RGB config */
CLUTTER_NOTE (BACKEND, "ARGB visual requested, but none found");
}
if (n_configs >= 1)
{
backend_glx->found_fbconfig = TRUE;
backend_glx->fbconfig = configs[0];
}
out:
XFree (configs);
}
}
if (backend_glx->found_fbconfig)
{
*config = backend_glx->fbconfig;
return TRUE;
}
else
return FALSE;
}
void
_clutter_backend_glx_blit_sub_buffer (ClutterBackendGLX *backend_glx,
GLXDrawable drawable,
int x, int y, int width, int height)
{
ClutterBackendX11 *backend_x11 = CLUTTER_BACKEND_X11 (backend_glx);
if (backend_glx->copy_sub_buffer)
{
backend_glx->copy_sub_buffer (backend_x11->xdpy, drawable,
x, y, width, height);
}
else if (backend_glx->blit_framebuffer)
{
glDrawBuffer (GL_FRONT);
backend_glx->blit_framebuffer (x, y, x + width, y + height,
x, y, x + width, y + height,
GL_COLOR_BUFFER_BIT, GL_NEAREST);
glDrawBuffer (GL_BACK);
glFlush();
}
}
static XVisualInfo *
clutter_backend_glx_get_visual_info (ClutterBackendX11 *backend_x11)
{
ClutterBackendGLX *backend_glx = CLUTTER_BACKEND_GLX (backend_x11);
GLXFBConfig config;
if (!_clutter_backend_glx_get_fbconfig (backend_glx, &config))
return NULL;
return glXGetVisualFromFBConfig (backend_x11->xdpy, config);
}
static gboolean
clutter_backend_glx_create_context (ClutterBackend *backend,
GError **error)
{
ClutterBackendGLX *backend_glx = CLUTTER_BACKEND_GLX (backend);
ClutterBackendX11 *backend_x11 = CLUTTER_BACKEND_X11 (backend);
GLXFBConfig config;
gboolean is_direct;
Window root_xwin;
XSetWindowAttributes attrs;
XVisualInfo *xvisinfo;
Display *xdisplay;
int major;
int minor;
GLXDrawable dummy_drawable;
if (backend_glx->gl_context != NULL)
return TRUE;
xdisplay = clutter_x11_get_default_display ();
root_xwin = clutter_x11_get_root_window ();
if (!_clutter_backend_glx_get_fbconfig (backend_glx, &config))
{
g_set_error (error, CLUTTER_INIT_ERROR,
CLUTTER_INIT_ERROR_BACKEND,
"Unable to find suitable fbconfig for the GLX context");
return FALSE;
}
CLUTTER_NOTE (BACKEND, "Creating GLX Context (display: %p)", xdisplay);
backend_glx->gl_context = glXCreateNewContext (xdisplay,
config,
GLX_RGBA_TYPE,
NULL,
True);
if (backend_glx->gl_context == NULL)
{
g_set_error (error, CLUTTER_INIT_ERROR,
CLUTTER_INIT_ERROR_BACKEND,
"Unable to create suitable GL context");
return FALSE;
}
is_direct = glXIsDirect (xdisplay, backend_glx->gl_context);
CLUTTER_NOTE (GL, "Setting %s context",
is_direct ? "direct"
: "indirect");
_cogl_set_indirect_context (!is_direct);
/* COGL assumes that there is always a GL context selected; in order
* to make sure that a GLX context exists and is made current, we use
* a dummy, offscreen override-redirect window to which we can always
* fall back if no stage is available
*
* XXX - we need to do this dance because GLX does not allow creating
* a context and querying it for basic information (even the function
* pointers) unless it's made current to a real Drawable. it should be
* possible to avoid this in future releases of Mesa and X11, but right
* now this is the best solution available.
*/
xvisinfo = glXGetVisualFromFBConfig (xdisplay, config);
if (xvisinfo == NULL)
{
g_set_error (error, CLUTTER_INIT_ERROR,
CLUTTER_INIT_ERROR_BACKEND,
"Unable to retrieve the X11 visual");
return FALSE;
}
clutter_x11_trap_x_errors ();
attrs.override_redirect = True;
attrs.colormap = XCreateColormap (xdisplay,
root_xwin,
xvisinfo->visual,
AllocNone);
attrs.border_pixel = 0;
backend_glx->dummy_xwin = XCreateWindow (xdisplay, root_xwin,
-100, -100, 1, 1,
0,
xvisinfo->depth,
CopyFromParent,
xvisinfo->visual,
CWOverrideRedirect | CWColormap | CWBorderPixel,
&attrs);
/* Try and create a GLXWindow to use with extensions dependent on
* GLX versions >= 1.3 that don't accept regular X Windows as GLX
* drawables. */
if (glXQueryVersion (backend_x11->xdpy, &major, &minor) &&
major == 1 && minor >= 3)
{
backend_glx->dummy_glxwin = glXCreateWindow (backend_x11->xdpy,
config,
backend_glx->dummy_xwin,
NULL);
}
if (backend_glx->dummy_glxwin)
dummy_drawable = backend_glx->dummy_glxwin;
else
dummy_drawable = backend_glx->dummy_xwin;
CLUTTER_NOTE (BACKEND, "Selecting dummy 0x%x for the GLX context",
(unsigned int) dummy_drawable);
glXMakeContextCurrent (xdisplay,
dummy_drawable,
dummy_drawable,
backend_glx->gl_context);
XFree (xvisinfo);
if (clutter_x11_untrap_x_errors ())
{
g_set_error (error, CLUTTER_INIT_ERROR,
CLUTTER_INIT_ERROR_BACKEND,
"Unable to select the newly created GLX context");
return FALSE;
}
return TRUE;
}
/* TODO: remove this interface in favour of
* _clutter_stage_window_make_current () */
static void
clutter_backend_glx_ensure_context (ClutterBackend *backend,
ClutterStage *stage)
{
ClutterStageWindow *impl;
/* if there is no stage, the stage is being destroyed or it has no
* implementation attached to it then we clear the GL context
*/
if (stage == NULL ||
(CLUTTER_PRIVATE_FLAGS (stage) & CLUTTER_ACTOR_IN_DESTRUCTION) ||
((impl = _clutter_stage_get_window (stage)) == NULL))
{
ClutterBackendX11 *backend_x11;
backend_x11 = CLUTTER_BACKEND_X11 (backend);
CLUTTER_NOTE (MULTISTAGE, "Clearing all context");
glXMakeContextCurrent (backend_x11->xdpy, None, None, NULL);
}
else
{
ClutterBackendGLX *backend_glx;
ClutterBackendX11 *backend_x11;
ClutterStageGLX *stage_glx;
ClutterStageX11 *stage_x11;
GLXDrawable drawable;
g_assert (impl != NULL);
stage_glx = CLUTTER_STAGE_GLX (impl);
stage_x11 = CLUTTER_STAGE_X11 (impl);
backend_glx = CLUTTER_BACKEND_GLX (backend);
backend_x11 = CLUTTER_BACKEND_X11 (backend);
drawable = stage_glx->glxwin ? stage_glx->glxwin : stage_x11->xwin;
CLUTTER_NOTE (BACKEND,
"Setting context for stage of type %s, window: 0x%x",
G_OBJECT_TYPE_NAME (impl),
(unsigned int) drawable);
/* no GL context to set */
if (backend_glx->gl_context == NULL)
return;
2008-05-12 Emmanuele Bassi <ebassi@openedhand.com> Rework the stage wrapper/implementation relation: remove duplicated code and all the bookkeeping from the backends into ClutterStage whenever possible, to reduce the amount of work a backend must do (and possibly get wrong). Thanks to Tommi Komulainen. * clutter/clutter-main.c: (clutter_init_with_args), (clutter_init): Realize the default stage after creation. The default stage is special, because we use it in the initialization sequence. This removes the burden from the backends and reduces the things a backend can get wrong. * clutter/clutter-stage.c: (clutter_stage_show): Make sure to realize the implementation if it hasn't been realized yet. (clutter_stage_realize): Set the REALIZED flag and call clutter_stage_ensure_current() if the implementation was successfully realized. (clutter_stage_unrealized): Call clutter_stage_ensure_current() on unrealize. * clutter/glx/clutter-backend-glx.c: (clutter_backend_glx_create_stage): Do not realize the stage anymore when creating it, and let the normal realization sequence take place. (clutter_backend_glx_ensure_context): Trap for X11 errors. * clutter/glx/clutter-stage-glx.c: (clutter_stage_glx_realize): Chain up to the X11 implementation so that we can set up the window state (title, cursor visibility) when we actually have a X window. Also, do not call clutter_stage_ensure_current(), and rely on the wrapper to do it for us. This means we can drop setting the REALIZED flag on the wrapper. (clutter_stage_glx_unrealize): Do not call clutter_stage_ensure_current() ourselves, and rely on the wrapper to do it for us. * clutter/x11/clutter-stage-x11.c: (set_wm_title), (set_cursor_visible): Move the WM title and cursor visibility code inside their own functions. (clutter_stage_x11_realize): Set the window title and whether the cursor is visible or not after realizing the stage. (clutter_stage_x11_set_cursor_visible), (clutter_stage_x11_set_title): Call set_wm_title() and set_cursor_visible(). (clutter_stage_x11_finalize): Free the title string. * clutter/x11/clutter-stage-x11.h: Save more of the stage state, so that we can set it even when the stage hasn't been realized yet. * clutter/eglnative/clutter-backend-egl.c: (clutter_backend_egl_create_stage): * clutter/eglnative/clutter-stage-egl.c: (clutter_stage_egl_unrealize), (clutter_stage_egl_realize): Update the eglnative backend. * clutter/eglx/clutter-backend-egl.c: (clutter_backend_egl_ensure_context), (clutter_backend_egl_create_stage): * clutter/eglx/clutter-stage-egl.c: (clutter_stage_egl_unrealize), (clutter_stage_egl_realize): Update the eglx backend. * clutter/sdl/clutter-backend-sdl.c: (clutter_backend_sdl_create_stage): * clutter/sdl/clutter-stage-sdl.c: (clutter_stage_sdl_realize): Update the sdl backend. * clutter/fruity/clutter-backend-fruity.c: (clutter_backend_fruity_create_stage): * clutter/sdl/clutter-stage-fruity.c: (clutter_stage_fruity_realize): Update the fruity backend. * tests/test-multistage.c (on_button_press): Bail out if clutter_stage_new() returns NULL. * HACKING.backends: Update backend writing documentation.
2008-05-12 11:26:37 -04:00
clutter_x11_trap_x_errors ();
/* we might get here inside the final dispose cycle, so we
* need to handle this gracefully
*/
if (drawable == None)
{
GLXDrawable dummy_drawable;
CLUTTER_NOTE (BACKEND,
"Received a stale stage, clearing all context");
if (backend_glx->dummy_glxwin)
dummy_drawable = backend_glx->dummy_glxwin;
else
dummy_drawable = backend_glx->dummy_xwin;
if (dummy_drawable == None)
glXMakeContextCurrent (backend_x11->xdpy, None, None, NULL);
else
{
glXMakeContextCurrent (backend_x11->xdpy,
dummy_drawable,
dummy_drawable,
backend_glx->gl_context);
}
}
else
{
CLUTTER_NOTE (BACKEND,
"MakeContextCurrent dpy: %p, window: 0x%x (%s), context: %p",
backend_x11->xdpy,
(unsigned int) drawable,
stage_x11->is_foreign_xwin ? "foreign" : "native",
backend_glx->gl_context);
glXMakeContextCurrent (backend_x11->xdpy,
drawable,
drawable,
backend_glx->gl_context);
/*
* In case we are using GLX_SGI_swap_control for vblank syncing we need call
* glXSwapIntervalSGI here to make sure that it affects the current drawable.
*/
if (backend_glx->vblank_type == CLUTTER_VBLANK_GLX_SWAP && backend_glx->swap_interval != NULL)
backend_glx->swap_interval (1);
}
2008-05-12 Emmanuele Bassi <ebassi@openedhand.com> Rework the stage wrapper/implementation relation: remove duplicated code and all the bookkeeping from the backends into ClutterStage whenever possible, to reduce the amount of work a backend must do (and possibly get wrong). Thanks to Tommi Komulainen. * clutter/clutter-main.c: (clutter_init_with_args), (clutter_init): Realize the default stage after creation. The default stage is special, because we use it in the initialization sequence. This removes the burden from the backends and reduces the things a backend can get wrong. * clutter/clutter-stage.c: (clutter_stage_show): Make sure to realize the implementation if it hasn't been realized yet. (clutter_stage_realize): Set the REALIZED flag and call clutter_stage_ensure_current() if the implementation was successfully realized. (clutter_stage_unrealized): Call clutter_stage_ensure_current() on unrealize. * clutter/glx/clutter-backend-glx.c: (clutter_backend_glx_create_stage): Do not realize the stage anymore when creating it, and let the normal realization sequence take place. (clutter_backend_glx_ensure_context): Trap for X11 errors. * clutter/glx/clutter-stage-glx.c: (clutter_stage_glx_realize): Chain up to the X11 implementation so that we can set up the window state (title, cursor visibility) when we actually have a X window. Also, do not call clutter_stage_ensure_current(), and rely on the wrapper to do it for us. This means we can drop setting the REALIZED flag on the wrapper. (clutter_stage_glx_unrealize): Do not call clutter_stage_ensure_current() ourselves, and rely on the wrapper to do it for us. * clutter/x11/clutter-stage-x11.c: (set_wm_title), (set_cursor_visible): Move the WM title and cursor visibility code inside their own functions. (clutter_stage_x11_realize): Set the window title and whether the cursor is visible or not after realizing the stage. (clutter_stage_x11_set_cursor_visible), (clutter_stage_x11_set_title): Call set_wm_title() and set_cursor_visible(). (clutter_stage_x11_finalize): Free the title string. * clutter/x11/clutter-stage-x11.h: Save more of the stage state, so that we can set it even when the stage hasn't been realized yet. * clutter/eglnative/clutter-backend-egl.c: (clutter_backend_egl_create_stage): * clutter/eglnative/clutter-stage-egl.c: (clutter_stage_egl_unrealize), (clutter_stage_egl_realize): Update the eglnative backend. * clutter/eglx/clutter-backend-egl.c: (clutter_backend_egl_ensure_context), (clutter_backend_egl_create_stage): * clutter/eglx/clutter-stage-egl.c: (clutter_stage_egl_unrealize), (clutter_stage_egl_realize): Update the eglx backend. * clutter/sdl/clutter-backend-sdl.c: (clutter_backend_sdl_create_stage): * clutter/sdl/clutter-stage-sdl.c: (clutter_stage_sdl_realize): Update the sdl backend. * clutter/fruity/clutter-backend-fruity.c: (clutter_backend_fruity_create_stage): * clutter/sdl/clutter-stage-fruity.c: (clutter_stage_fruity_realize): Update the fruity backend. * tests/test-multistage.c (on_button_press): Bail out if clutter_stage_new() returns NULL. * HACKING.backends: Update backend writing documentation.
2008-05-12 11:26:37 -04:00
if (clutter_x11_untrap_x_errors ())
g_critical ("Unable to make the stage window 0x%x the current "
"GLX drawable",
(unsigned int) drawable);
}
}
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
/*
* FIXME: we should remove backend_class->redraw() and just
* have stage_window_iface->redraw()
*/
static void
clutter_backend_glx_redraw (ClutterBackend *backend,
ClutterStage *stage)
{
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
ClutterStageWindow *impl = _clutter_stage_get_window (stage);
if (G_UNLIKELY (impl == NULL))
{
CLUTTER_NOTE (BACKEND, "Stage [%p] has no implementation", stage);
return;
}
g_assert (CLUTTER_IS_STAGE_GLX (impl));
Adds initial clipped redraw support to Clutter A new (internal only currently) API, _clutter_actor_queue_clipped_redraw can be used to queue a redraw along with a clip rectangle in actor coordinates. This clip rectangle propagates up to the stage and clutter backend which may optionally use the information to optimize stage redraws. The GLX backend in particular may scissor the next redraw to the clip rectangle and use GLX_MESA_copy_sub_buffer to present the stage subregion. The intention is that any actors that can naturally determine the bounds of updates should queue clipped redraws to reduce the cost of updating small regions of the screen. Notes: » If GLX_MESA_copy_sub_buffer isn't available then the GLX backend ignores any clip rectangles. » queuing multiple clipped redraws will result in the bounding box of each clip rectangle being used. » If a clipped redraw has a height > 300 pixels then it's promoted into a full stage redraw, so that the GPU doesn't end up blocking too long waiting for the vsync to reach the optimal position to avoid tearing. » Note: no empirical data was used to come up with this threshold so we may need to tune this. » Currently only ClutterX11TexturePixmap makes use of this new API. This is done via a new "queue-damage-redraw" signal that is emitted when the pixmap is updated. The default handler queues a clipped redraw with the assumption that the pixmap is being painted as a rectangle covering the actors transformed allocation. If you subclass ClutterX11TexturePixmap and change how it's painted you now also need to override the signal handler and queue your own redraw. Technically this is a semantic break, but it's assumed that no one is currently doing this. This still leaves a few unsolved issues with regards to optimizing sub stage redraws that need to be addressed in further work so this can only be considered a stepping stone a this point: » Because we have no reliable way to determine if the painting of any given actor is being modified any optimizations implemented using _clutter_actor_queue_redraw_with_clip must be overridable by a subclass, and technically must be opt-in for existing classes to avoid a change in semantics. E.g. consider that a user connects to the paint signal for ClutterTexture and paints a circle instead of a rectangle. In this case any original logic to queue clipped redraws would be incorrect. » Currently only the implementation of an actor has enough information with which to queue clipped redraws. E.g. It is not possible for generic code in clutter-actor.c to queue a clipped redraw when hiding an actor because actors have no way to report a "paint box". (remember actors can draw outside their allocation and actors with depth may also be projected outside of their allocation) » The current plan is to add a actor_class->get_paint_cuboid() virtual so actors can report a bounding cube for everything they would draw in their current state and use that to queue clipped redraws against the stage by projecting the paint cube into stage coordinates. » Our heuristics for promoting clipped redraws into full redraws to avoid blocking the GPU while we wait for the vsync need improving: » vsync issues aren't relevant for redirected/composited applications so they should use different heuristics. In this case we instead need to trade off the cost of blitting when using glXCopySubBuffer vs promoting to a full redraw and flipping instead.
2009-11-30 12:47:55 -05:00
clutter_stage_glx_redraw (CLUTTER_STAGE_GLX (impl),
stage);
}
static ClutterStageWindow *
clutter_backend_glx_create_stage (ClutterBackend *backend,
ClutterStage *wrapper,
GError **error)
{
ClutterBackendX11 *backend_x11 = CLUTTER_BACKEND_X11 (backend);
ClutterStageWindow *stage_window;
ClutterStageX11 *stage_x11;
CLUTTER_NOTE (BACKEND, "Creating stage of type '%s'",
g_type_name (CLUTTER_STAGE_TYPE));
stage_window = g_object_new (CLUTTER_TYPE_STAGE_GLX, NULL);
/* copy backend data into the stage */
stage_x11 = CLUTTER_STAGE_X11 (stage_window);
stage_x11->wrapper = wrapper;
CLUTTER_NOTE (BACKEND,
"GLX stage created[%p] (dpy:%p, screen:%d, root:%u, wrap:%p)",
stage_window,
backend_x11->xdpy,
backend_x11->xscreen_num,
(unsigned int) backend_x11->xwin_root,
wrapper);
return stage_window;
}
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
static void
clutter_backend_glx_class_init (ClutterBackendGLXClass *klass)
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
{
GObjectClass *gobject_class = G_OBJECT_CLASS (klass);
ClutterBackendClass *backend_class = CLUTTER_BACKEND_CLASS (klass);
ClutterBackendX11Class *backendx11_class = CLUTTER_BACKEND_X11_CLASS (klass);
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
gobject_class->constructor = clutter_backend_glx_constructor;
gobject_class->dispose = clutter_backend_glx_dispose;
gobject_class->finalize = clutter_backend_glx_finalize;
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
backend_class->pre_parse = clutter_backend_glx_pre_parse;
backend_class->post_parse = clutter_backend_glx_post_parse;
backend_class->create_stage = clutter_backend_glx_create_stage;
backend_class->add_options = clutter_backend_glx_add_options;
backend_class->get_features = clutter_backend_glx_get_features;
backend_class->redraw = clutter_backend_glx_redraw;
backend_class->create_context = clutter_backend_glx_create_context;
backend_class->ensure_context = clutter_backend_glx_ensure_context;
backendx11_class->get_visual_info = clutter_backend_glx_get_visual_info;
backendx11_class->handle_event = clutter_backend_glx_handle_event;
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
}
static void
clutter_backend_glx_init (ClutterBackendGLX *backend_glx)
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
{
2007-03-22 Emmanuele Bassi <ebassi@openedhand.com> * clutter/clutter-private.h: Remove inclusion of backend-specific headers; update the main context object; add the declarations for the event queue functions. * clutter/clutter-backend.[ch]: Add the abstract ClutterBackend object, which holds backend-specific settings, the main stage, and the event queue. Every backend must implement a subclass of ClutterBackend and ClutterStage. * clutter/clutter-feature.c: Protect the GLX specific calls behing #ifdef HAVE_CLUTTER_GLX. * clutter/clutter-actor.c: * clutter/clutter-group.c: * clutter/clutter-clone-texture.c: Include GL/gl.h * clutter/clutter-event.[ch]: Update public API and implement the event queue private API; hold a reference on the event objects; move out the keysym-to-unicode table; add the new event types. * clutter/clutter-color.h: Include clutter-fixed.h * clutter/clutter-main.c: Update API; get the main stage from the backend object; process the event received from the queue; lock/unlock the main mutex if we have one; move the initialisation process sooner in the init sequence, in order to have the backend object when we check for options; call the backed vfuncs in the pre/post parse hooks. * clutter/clutter-stage.c: Make ClutterStage and abstract class, implemented by the backends. * clutter/clutter/glx/clutter-glx.h: * clutter/clutter/glx/clutter-backend-glx.[ch]: * clutter/clutter/glx/clutter-event-glx.c: * clutter/clutter/glx/clutter-stage-glx.[ch]: * clutter/clutter/glx/Makefile.am: Add the GLX backend. * clutter/clutter/egl/clutter-backend-egl.[ch]: * clutter/clutter/egl/clutter-event-egl.c: * clutter/clutter/egl/clutter-stage-egl.[ch]: * clutter/clutter/egl/Makefile.am: Add the stub for a EGL backend. * examples/*.c: Update for the new API.
2007-03-22 14:21:59 -04:00
}
/* every backend must implement this function */
GType
_clutter_backend_impl_get_type (void)
{
return clutter_backend_glx_get_type ();
}