mutter/cogl/winsys/cogl-winsys-egl-kms.c

1304 lines
38 KiB
C
Raw Normal View History

/*
* Cogl
*
This re-licenses Cogl 1.18 under the MIT license Since the Cogl 1.18 branch is actively maintained in parallel with the master branch; this is a counter part to commit 1b83ef938fc16b which re-licensed the master branch to use the MIT license. This re-licensing is a follow up to the proposal that was sent to the Cogl mailing list: http://lists.freedesktop.org/archives/cogl/2013-December/001465.html Note: there was a copyright assignment policy in place for Clutter (and therefore Cogl which was part of Clutter at the time) until the 11th of June 2010 and so we only checked the details after that point (commit 0bbf50f905) For each file, authors were identified via this Git command: $ git blame -p -C -C -C20 -M -M10 0bbf50f905..HEAD We received blanket approvals for re-licensing all Red Hat and Collabora contributions which reduced how many people needed to be contacted individually: - http://lists.freedesktop.org/archives/cogl/2013-December/001470.html - http://lists.freedesktop.org/archives/cogl/2014-January/001536.html Individual approval requests were sent to all the other identified authors who all confirmed the re-license on the Cogl mailinglist: http://lists.freedesktop.org/archives/cogl/2014-January As well as updating the copyright header in all sources files, the COPYING file has been updated to reflect the license change and also document the other licenses used in Cogl such as the SGI Free Software License B, version 2.0 and the 3-clause BSD license. This patch was not simply cherry-picked from master; but the same methodology was used to check the source files.
2014-02-21 20:28:54 -05:00
* A Low Level GPU Graphics and Utilities API
*
* Copyright (C) 2011 Intel Corporation.
*
This re-licenses Cogl 1.18 under the MIT license Since the Cogl 1.18 branch is actively maintained in parallel with the master branch; this is a counter part to commit 1b83ef938fc16b which re-licensed the master branch to use the MIT license. This re-licensing is a follow up to the proposal that was sent to the Cogl mailing list: http://lists.freedesktop.org/archives/cogl/2013-December/001465.html Note: there was a copyright assignment policy in place for Clutter (and therefore Cogl which was part of Clutter at the time) until the 11th of June 2010 and so we only checked the details after that point (commit 0bbf50f905) For each file, authors were identified via this Git command: $ git blame -p -C -C -C20 -M -M10 0bbf50f905..HEAD We received blanket approvals for re-licensing all Red Hat and Collabora contributions which reduced how many people needed to be contacted individually: - http://lists.freedesktop.org/archives/cogl/2013-December/001470.html - http://lists.freedesktop.org/archives/cogl/2014-January/001536.html Individual approval requests were sent to all the other identified authors who all confirmed the re-license on the Cogl mailinglist: http://lists.freedesktop.org/archives/cogl/2014-January As well as updating the copyright header in all sources files, the COPYING file has been updated to reflect the license change and also document the other licenses used in Cogl such as the SGI Free Software License B, version 2.0 and the 3-clause BSD license. This patch was not simply cherry-picked from master; but the same methodology was used to check the source files.
2014-02-21 20:28:54 -05:00
* Permission is hereby granted, free of charge, to any person
* obtaining a copy of this software and associated documentation
* files (the "Software"), to deal in the Software without
* restriction, including without limitation the rights to use, copy,
* modify, merge, publish, distribute, sublicense, and/or sell copies
* of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
This re-licenses Cogl 1.18 under the MIT license Since the Cogl 1.18 branch is actively maintained in parallel with the master branch; this is a counter part to commit 1b83ef938fc16b which re-licensed the master branch to use the MIT license. This re-licensing is a follow up to the proposal that was sent to the Cogl mailing list: http://lists.freedesktop.org/archives/cogl/2013-December/001465.html Note: there was a copyright assignment policy in place for Clutter (and therefore Cogl which was part of Clutter at the time) until the 11th of June 2010 and so we only checked the details after that point (commit 0bbf50f905) For each file, authors were identified via this Git command: $ git blame -p -C -C -C20 -M -M10 0bbf50f905..HEAD We received blanket approvals for re-licensing all Red Hat and Collabora contributions which reduced how many people needed to be contacted individually: - http://lists.freedesktop.org/archives/cogl/2013-December/001470.html - http://lists.freedesktop.org/archives/cogl/2014-January/001536.html Individual approval requests were sent to all the other identified authors who all confirmed the re-license on the Cogl mailinglist: http://lists.freedesktop.org/archives/cogl/2014-January As well as updating the copyright header in all sources files, the COPYING file has been updated to reflect the license change and also document the other licenses used in Cogl such as the SGI Free Software License B, version 2.0 and the 3-clause BSD license. This patch was not simply cherry-picked from master; but the same methodology was used to check the source files.
2014-02-21 20:28:54 -05:00
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
This re-licenses Cogl 1.18 under the MIT license Since the Cogl 1.18 branch is actively maintained in parallel with the master branch; this is a counter part to commit 1b83ef938fc16b which re-licensed the master branch to use the MIT license. This re-licensing is a follow up to the proposal that was sent to the Cogl mailing list: http://lists.freedesktop.org/archives/cogl/2013-December/001465.html Note: there was a copyright assignment policy in place for Clutter (and therefore Cogl which was part of Clutter at the time) until the 11th of June 2010 and so we only checked the details after that point (commit 0bbf50f905) For each file, authors were identified via this Git command: $ git blame -p -C -C -C20 -M -M10 0bbf50f905..HEAD We received blanket approvals for re-licensing all Red Hat and Collabora contributions which reduced how many people needed to be contacted individually: - http://lists.freedesktop.org/archives/cogl/2013-December/001470.html - http://lists.freedesktop.org/archives/cogl/2014-January/001536.html Individual approval requests were sent to all the other identified authors who all confirmed the re-license on the Cogl mailinglist: http://lists.freedesktop.org/archives/cogl/2014-January As well as updating the copyright header in all sources files, the COPYING file has been updated to reflect the license change and also document the other licenses used in Cogl such as the SGI Free Software License B, version 2.0 and the 3-clause BSD license. This patch was not simply cherry-picked from master; but the same methodology was used to check the source files.
2014-02-21 20:28:54 -05:00
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*
*
* Authors:
* Rob Bradford <rob@linux.intel.com>
* Kristian Høgsberg (from eglkms.c)
* Benjamin Franzke (from eglkms.c)
* Robert Bragg <robert@linux.intel.com>
* Neil Roberts <neil@linux.intel.com>
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include <EGL/egl.h>
#include <EGL/eglext.h>
#include <stddef.h>
#include <drm.h>
#include <xf86drm.h>
#include <xf86drmMode.h>
#include <gbm.h>
#include <glib.h>
#include <sys/fcntl.h>
#include <unistd.h>
#include <string.h>
#include <stdlib.h>
#include "cogl-winsys-egl-kms-private.h"
#include "cogl-winsys-egl-private.h"
#include "cogl-renderer-private.h"
#include "cogl-framebuffer-private.h"
#include "cogl-onscreen-private.h"
Add -Wmissing-declarations to maintainer flags and fix problems This option to GCC makes it give a warning whenever a global function is defined without a declaration. This should catch cases were we've defined a function but forgot to put it in a header. In that case it is either only used within one file so we should make it static or we should declare it in a header. The following changes where made to fix problems: • Some functions were made static • cogl-path.h (the one containing the 1.0 API) was split into two files, one defining the functions and one defining the enums so that cogl-path.c can include the enum and function declarations from the 2.0 API as well as the function declarations from the 1.0 API. • cogl2-clip-state has been removed. This only had one experimental function called cogl_clip_push_from_path but as this is unstable we might as well remove it favour of the equivalent cogl_framebuffer_* API. • The GLX, SDL and WGL winsys's now have a private header to define their get_vtable function instead of directly declaring in the C file where it is called. • All places that were calling COGL_OBJECT_DEFINE need to have the cogl_is_whatever function declared so these have been added either as a public function or in a private header. • Some files that were not including the header containing their function declarations have been fixed to do so. • Any unused error quark functions have been removed. If we later want them we should add them back one by one and add a declaration for them in a header. • _cogl_is_framebuffer has been renamed to cogl_is_framebuffer and made a public function with a declaration in cogl-framebuffer.h • Similarly for CoglOnscreen. • cogl_vdraw_indexed_attributes is called cogl_framebuffer_vdraw_indexed_attributes in the header. The definition has been changed to match the header. • cogl_index_buffer_allocate has been removed. This had no declaration and I'm not sure what it's supposed to do. • CoglJournal has been changed to use the internal CoglObject macro so that it won't define an exported cogl_is_journal symbol. • The _cogl_blah_pointer_from_handle functions have been removed. CoglHandle isn't used much anymore anyway and in the few places where it is used I think it's safe to just use the implicit cast from void* to the right type. • The test-utils.h header for the conformance tests explicitly disables the -Wmissing-declaration option using a pragma because all of the tests declare their main function without a header. Any mistakes relating to missing declarations aren't really important for the tests. • cogl_quaternion_init_from_quaternion and init_from_matrix have been given declarations in cogl-quaternion.h Reviewed-by: Robert Bragg <robert@linux.intel.com>
2012-03-06 13:21:28 -05:00
#include "cogl-kms-renderer.h"
#include "cogl-kms-display.h"
#include "cogl-version.h"
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
#include "cogl-error-private.h"
#include "cogl-poll-private.h"
static const CoglWinsysEGLVtable _cogl_winsys_egl_vtable;
static const CoglWinsysVtable *parent_vtable;
typedef struct _CoglRendererKMS
{
int fd;
int opened_fd;
struct gbm_device *gbm;
CoglClosure *swap_notify_idle;
CoglBool page_flips_not_supported;
} CoglRendererKMS;
typedef struct _CoglOutputKMS
{
drmModeConnector *connector;
drmModeEncoder *encoder;
drmModeCrtc *saved_crtc;
drmModeModeInfo *modes;
int n_modes;
drmModeModeInfo mode;
} CoglOutputKMS;
typedef struct _CoglDisplayKMS
{
GList *outputs;
GList *crtcs;
int width, height;
CoglBool pending_set_crtc;
struct gbm_surface *dummy_gbm_surface;
CoglOnscreen *onscreen;
} CoglDisplayKMS;
typedef struct _CoglFlipKMS
{
CoglOnscreen *onscreen;
int pending;
} CoglFlipKMS;
typedef struct _CoglOnscreenKMS
{
struct gbm_surface *surface;
uint32_t current_fb_id;
uint32_t next_fb_id;
struct gbm_bo *current_bo;
struct gbm_bo *next_bo;
CoglBool pending_swap_notify;
EGLSurface *pending_egl_surface;
struct gbm_surface *pending_surface;
} CoglOnscreenKMS;
static const char device_name[] = "/dev/dri/card0";
static void
_cogl_winsys_renderer_disconnect (CoglRenderer *renderer)
{
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
eglTerminate (egl_renderer->edpy);
if (kms_renderer->opened_fd >= 0)
close (kms_renderer->opened_fd);
g_slice_free (CoglRendererKMS, kms_renderer);
g_slice_free (CoglRendererEGL, egl_renderer);
}
static void
flush_pending_swap_notify_cb (void *data,
void *user_data)
{
CoglFramebuffer *framebuffer = data;
if (framebuffer->type == COGL_FRAMEBUFFER_TYPE_ONSCREEN)
{
CoglOnscreen *onscreen = COGL_ONSCREEN (framebuffer);
CoglOnscreenEGL *egl_onscreen = onscreen->winsys;
CoglOnscreenKMS *kms_onscreen = egl_onscreen->platform;
if (kms_onscreen->pending_swap_notify)
{
CoglFrameInfo *info = g_queue_pop_head (&onscreen->pending_frame_infos);
_cogl_onscreen_notify_frame_sync (onscreen, info);
_cogl_onscreen_notify_complete (onscreen, info);
kms_onscreen->pending_swap_notify = FALSE;
cogl_object_unref (info);
}
}
}
static void
flush_pending_swap_notify_idle (void *user_data)
{
CoglContext *context = user_data;
CoglRendererEGL *egl_renderer = context->display->renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
/* This needs to be disconnected before invoking the callbacks in
* case the callbacks cause it to be queued again */
_cogl_closure_disconnect (kms_renderer->swap_notify_idle);
kms_renderer->swap_notify_idle = NULL;
g_list_foreach (context->framebuffers,
flush_pending_swap_notify_cb,
NULL);
}
static void
free_current_bo (CoglOnscreen *onscreen)
{
CoglOnscreenEGL *egl_onscreen = onscreen->winsys;
CoglOnscreenKMS *kms_onscreen = egl_onscreen->platform;
CoglContext *context = COGL_FRAMEBUFFER (onscreen)->context;
CoglRenderer *renderer = context->display->renderer;
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
if (kms_onscreen->current_fb_id)
{
drmModeRmFB (kms_renderer->fd,
kms_onscreen->current_fb_id);
kms_onscreen->current_fb_id = 0;
}
if (kms_onscreen->current_bo)
{
gbm_surface_release_buffer (kms_onscreen->surface,
kms_onscreen->current_bo);
kms_onscreen->current_bo = NULL;
}
}
static void
queue_swap_notify_for_onscreen (CoglOnscreen *onscreen)
{
CoglOnscreenEGL *egl_onscreen = onscreen->winsys;
CoglOnscreenKMS *kms_onscreen = egl_onscreen->platform;
CoglContext *context = COGL_FRAMEBUFFER (onscreen)->context;
CoglRenderer *renderer = context->display->renderer;
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
/* We only want to notify that the swap is complete when the
* application calls cogl_context_dispatch so instead of
* immediately notifying we queue an idle callback */
if (!kms_renderer->swap_notify_idle)
{
kms_renderer->swap_notify_idle =
_cogl_poll_renderer_add_idle (renderer,
flush_pending_swap_notify_idle,
context,
NULL);
}
kms_onscreen->pending_swap_notify = TRUE;
}
static void
process_flip (CoglFlipKMS *flip)
{
/* We're only ready to dispatch a swap notification once all outputs
* have flipped... */
flip->pending--;
if (flip->pending == 0)
{
CoglOnscreen *onscreen = flip->onscreen;
CoglOnscreenEGL *egl_onscreen = onscreen->winsys;
CoglOnscreenKMS *kms_onscreen = egl_onscreen->platform;
queue_swap_notify_for_onscreen (onscreen);
free_current_bo (onscreen);
kms_onscreen->current_fb_id = kms_onscreen->next_fb_id;
kms_onscreen->next_fb_id = 0;
kms_onscreen->current_bo = kms_onscreen->next_bo;
kms_onscreen->next_bo = NULL;
cogl_object_unref (flip->onscreen);
g_slice_free (CoglFlipKMS, flip);
}
}
static void
page_flip_handler (int fd,
unsigned int frame,
unsigned int sec,
unsigned int usec,
void *data)
{
CoglFlipKMS *flip = data;
process_flip (flip);
}
static void
handle_drm_event (CoglRendererKMS *kms_renderer)
{
drmEventContext evctx;
if (kms_renderer->page_flips_not_supported)
return;
memset (&evctx, 0, sizeof evctx);
evctx.version = DRM_EVENT_CONTEXT_VERSION;
evctx.page_flip_handler = page_flip_handler;
drmHandleEvent (kms_renderer->fd, &evctx);
}
static void
dispatch_kms_events (void *user_data, int revents)
{
CoglRenderer *renderer = user_data;
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
if (!revents)
return;
handle_drm_event (kms_renderer);
}
static CoglBool
_cogl_winsys_renderer_connect (CoglRenderer *renderer,
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
CoglError **error)
{
CoglRendererEGL *egl_renderer;
CoglRendererKMS *kms_renderer;
renderer->winsys = g_slice_new0 (CoglRendererEGL);
egl_renderer = renderer->winsys;
egl_renderer->platform_vtable = &_cogl_winsys_egl_vtable;
egl_renderer->platform = g_slice_new0 (CoglRendererKMS);
kms_renderer = egl_renderer->platform;
kms_renderer->fd = -1;
kms_renderer->opened_fd = -1;
if (renderer->kms_fd >= 0)
{
kms_renderer->fd = renderer->kms_fd;
}
else
{
kms_renderer->opened_fd = open (device_name, O_RDWR);
kms_renderer->fd = kms_renderer->opened_fd;
if (kms_renderer->fd < 0)
{
/* Probably permissions error */
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_INIT,
"Couldn't open %s", device_name);
return FALSE;
}
}
kms_renderer->gbm = gbm_create_device (kms_renderer->fd);
if (kms_renderer->gbm == NULL)
{
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_INIT,
"Couldn't create gbm device");
goto close_fd;
}
egl_renderer->edpy = eglGetDisplay ((EGLNativeDisplayType)kms_renderer->gbm);
if (egl_renderer->edpy == EGL_NO_DISPLAY)
{
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_INIT,
"Couldn't get eglDisplay");
goto destroy_gbm_device;
}
if (!_cogl_winsys_egl_renderer_connect_common (renderer, error))
goto egl_terminate;
_cogl_poll_renderer_add_fd (renderer,
kms_renderer->fd,
COGL_POLL_FD_EVENT_IN,
NULL, /* no prepare callback */
dispatch_kms_events,
renderer);
return TRUE;
egl_terminate:
eglTerminate (egl_renderer->edpy);
destroy_gbm_device:
gbm_device_destroy (kms_renderer->gbm);
close_fd:
if (kms_renderer->opened_fd >= 0)
close (kms_renderer->opened_fd);
_cogl_winsys_renderer_disconnect (renderer);
return FALSE;
}
static CoglBool
is_connector_excluded (int id,
int *excluded_connectors,
int n_excluded_connectors)
{
int i;
for (i = 0; i < n_excluded_connectors; i++)
if (excluded_connectors[i] == id)
return TRUE;
return FALSE;
}
static drmModeConnector *
find_connector (int fd,
drmModeRes *resources,
int *excluded_connectors,
int n_excluded_connectors)
{
int i;
for (i = 0; i < resources->count_connectors; i++)
{
drmModeConnector *connector =
drmModeGetConnector (fd, resources->connectors[i]);
if (connector &&
connector->connection == DRM_MODE_CONNECTED &&
connector->count_modes > 0 &&
!is_connector_excluded (connector->connector_id,
excluded_connectors,
n_excluded_connectors))
return connector;
drmModeFreeConnector(connector);
}
return NULL;
}
static CoglBool
find_mirror_modes (drmModeModeInfo *modes0,
int n_modes0,
drmModeModeInfo *modes1,
int n_modes1,
drmModeModeInfo *mode1_out,
drmModeModeInfo *mode0_out)
{
int i;
for (i = 0; i < n_modes0; i++)
{
int j;
drmModeModeInfo *mode0 = &modes0[i];
for (j = 0; j < n_modes1; j++)
{
drmModeModeInfo *mode1 = &modes1[j];
if (mode1->hdisplay == mode0->hdisplay &&
mode1->vdisplay == mode0->vdisplay)
{
*mode0_out = *mode0;
*mode1_out = *mode1;
return TRUE;
}
}
}
return FALSE;
}
static drmModeModeInfo builtin_1024x768 =
{
63500, /* clock */
1024, 1072, 1176, 1328, 0,
768, 771, 775, 798, 0,
59920,
DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_PVSYNC,
0,
"1024x768"
};
static CoglBool
is_panel (int type)
{
return (type == DRM_MODE_CONNECTOR_LVDS ||
type == DRM_MODE_CONNECTOR_eDP);
}
static CoglOutputKMS *
find_output (int _index,
int fd,
drmModeRes *resources,
int *excluded_connectors,
int n_excluded_connectors,
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
CoglError **error)
{
char *connector_env_name = g_strdup_printf ("COGL_KMS_CONNECTOR%d", _index);
char *mode_env_name;
drmModeConnector *connector;
drmModeEncoder *encoder;
CoglOutputKMS *output;
drmModeModeInfo *modes;
int n_modes;
if (getenv (connector_env_name))
{
unsigned long id = strtoul (getenv (connector_env_name), NULL, 10);
connector = drmModeGetConnector (fd, id);
}
else
connector = NULL;
g_free (connector_env_name);
if (connector == NULL)
connector = find_connector (fd, resources,
excluded_connectors, n_excluded_connectors);
if (connector == NULL)
{
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_INIT,
"No currently active connector found");
return NULL;
}
/* XXX: At this point it seems connector->encoder_id may be an invalid id of 0
* even though the connector is marked as connected. Referencing ->encoders[0]
* seems more reliable. */
encoder = drmModeGetEncoder (fd, connector->encoders[0]);
output = g_slice_new0 (CoglOutputKMS);
output->connector = connector;
output->encoder = encoder;
output->saved_crtc = drmModeGetCrtc (fd, encoder->crtc_id);
if (is_panel (connector->connector_type))
{
n_modes = connector->count_modes + 1;
modes = g_new (drmModeModeInfo, n_modes);
memcpy (modes, connector->modes,
sizeof (drmModeModeInfo) * connector->count_modes);
/* TODO: parse EDID */
modes[n_modes - 1] = builtin_1024x768;
}
else
{
n_modes = connector->count_modes;
modes = g_new (drmModeModeInfo, n_modes);
memcpy (modes, connector->modes,
sizeof (drmModeModeInfo) * n_modes);
}
mode_env_name = g_strdup_printf ("COGL_KMS_CONNECTOR%d_MODE", _index);
if (getenv (mode_env_name))
{
const char *name = getenv (mode_env_name);
int i;
CoglBool found = FALSE;
drmModeModeInfo mode;
for (i = 0; i < n_modes; i++)
{
if (strcmp (modes[i].name, name) == 0)
{
found = TRUE;
break;
}
}
if (!found)
{
g_free (mode_env_name);
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_INIT,
"COGL_KMS_CONNECTOR%d_MODE of %s could not be found",
_index, name);
return NULL;
}
n_modes = 1;
mode = modes[i];
g_free (modes);
modes = g_new (drmModeModeInfo, 1);
modes[0] = mode;
}
g_free (mode_env_name);
output->modes = modes;
output->n_modes = n_modes;
return output;
}
static void
setup_crtc_modes (CoglDisplay *display, int fb_id)
{
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
CoglRendererEGL *egl_renderer = display->renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
GList *l;
for (l = kms_display->crtcs; l; l = l->next)
{
CoglKmsCrtc *crtc = l->data;
int ret = drmModeSetCrtc (kms_renderer->fd,
crtc->id,
fb_id, crtc->x, crtc->y,
crtc->connectors, crtc->count,
crtc->count ? &crtc->mode : NULL);
if (ret)
g_warning ("Failed to set crtc mode %s: %m", crtc->mode.name);
}
}
static void
flip_all_crtcs (CoglDisplay *display, CoglFlipKMS *flip, int fb_id)
{
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
CoglRendererEGL *egl_renderer = display->renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
GList *l;
gboolean needs_flip = FALSE;
for (l = kms_display->crtcs; l; l = l->next)
{
CoglKmsCrtc *crtc = l->data;
int ret;
if (crtc->count == 0 || crtc->ignore)
continue;
needs_flip = TRUE;
if (!kms_renderer->page_flips_not_supported)
{
ret = drmModePageFlip (kms_renderer->fd,
crtc->id, fb_id,
DRM_MODE_PAGE_FLIP_EVENT, flip);
if (ret)
{
g_warning ("Failed to flip: %m");
kms_renderer->page_flips_not_supported = TRUE;
break;
}
}
flip->pending++;
}
if (kms_renderer->page_flips_not_supported && needs_flip)
flip->pending = 1;
}
static void
crtc_free (CoglKmsCrtc *crtc)
{
g_free (crtc->connectors);
g_slice_free (CoglKmsCrtc, crtc);
}
static CoglKmsCrtc *
crtc_copy (CoglKmsCrtc *from)
{
CoglKmsCrtc *new;
new = g_slice_new (CoglKmsCrtc);
*new = *from;
new->connectors = g_memdup (from->connectors, from->count * sizeof(uint32_t));
return new;
}
static CoglBool
_cogl_winsys_egl_display_setup (CoglDisplay *display,
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
CoglError **error)
{
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display;
CoglRendererEGL *egl_renderer = display->renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
drmModeRes *resources;
CoglOutputKMS *output0, *output1;
CoglBool mirror;
CoglKmsCrtc *crtc0, *crtc1;
kms_display = g_slice_new0 (CoglDisplayKMS);
egl_display->platform = kms_display;
resources = drmModeGetResources (kms_renderer->fd);
if (!resources)
{
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_INIT,
"drmModeGetResources failed");
return FALSE;
}
output0 = find_output (0,
kms_renderer->fd,
resources,
NULL,
0, /* n excluded connectors */
error);
kms_display->outputs = g_list_append (kms_display->outputs, output0);
if (!output0)
return FALSE;
if (getenv ("COGL_KMS_MIRROR"))
mirror = TRUE;
else
mirror = FALSE;
if (mirror)
{
int exclude_connector = output0->connector->connector_id;
output1 = find_output (1,
kms_renderer->fd,
resources,
&exclude_connector,
1, /* n excluded connectors */
error);
if (!output1)
return FALSE;
kms_display->outputs = g_list_append (kms_display->outputs, output1);
if (!find_mirror_modes (output0->modes, output0->n_modes,
output1->modes, output1->n_modes,
&output0->mode,
&output1->mode))
{
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_INIT,
"Failed to find matching modes for mirroring");
return FALSE;
}
}
else
{
output0->mode = output0->modes[0];
output1 = NULL;
}
crtc0 = g_slice_new (CoglKmsCrtc);
crtc0->id = output0->encoder->crtc_id;
crtc0->x = 0;
crtc0->y = 0;
crtc0->mode = output0->mode;
crtc0->connectors = g_new (uint32_t, 1);
crtc0->connectors[0] = output0->connector->connector_id;
crtc0->count = 1;
kms_display->crtcs = g_list_prepend (kms_display->crtcs, crtc0);
if (output1)
{
crtc1 = g_slice_new (CoglKmsCrtc);
crtc1->id = output1->encoder->crtc_id;
crtc1->x = 0;
crtc1->y = 0;
crtc1->mode = output1->mode;
crtc1->connectors = g_new (uint32_t, 1);
crtc1->connectors[0] = output1->connector->connector_id;
crtc1->count = 1;
kms_display->crtcs = g_list_prepend (kms_display->crtcs, crtc1);
}
kms_display->width = output0->mode.hdisplay;
kms_display->height = output0->mode.vdisplay;
/* We defer setting the crtc modes until the first swap_buffers request of a
* CoglOnscreen framebuffer. */
kms_display->pending_set_crtc = TRUE;
return TRUE;
}
static void
output_free (int fd, CoglOutputKMS *output)
{
if (output->modes)
g_free (output->modes);
if (output->encoder)
drmModeFreeEncoder (output->encoder);
if (output->connector)
{
if (output->saved_crtc)
{
int ret = drmModeSetCrtc (fd,
output->saved_crtc->crtc_id,
output->saved_crtc->buffer_id,
output->saved_crtc->x,
output->saved_crtc->y,
&output->connector->connector_id, 1,
&output->saved_crtc->mode);
if (ret)
g_warning (G_STRLOC ": Error restoring saved CRTC");
}
drmModeFreeConnector (output->connector);
}
g_slice_free (CoglOutputKMS, output);
}
static void
_cogl_winsys_egl_display_destroy (CoglDisplay *display)
{
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
CoglRenderer *renderer = display->renderer;
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
GList *l;
for (l = kms_display->outputs; l; l = l->next)
output_free (kms_renderer->fd, l->data);
g_list_free (kms_display->outputs);
kms_display->outputs = NULL;
g_list_free_full (kms_display->crtcs, (GDestroyNotify) crtc_free);
g_slice_free (CoglDisplayKMS, egl_display->platform);
}
static CoglBool
_cogl_winsys_egl_context_created (CoglDisplay *display,
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
CoglError **error)
{
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
CoglRenderer *renderer = display->renderer;
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
if ((egl_renderer->private_features &
COGL_EGL_WINSYS_FEATURE_SURFACELESS_CONTEXT) == 0)
{
kms_display->dummy_gbm_surface =
gbm_surface_create (kms_renderer->gbm,
16, 16,
GBM_FORMAT_XRGB8888,
GBM_BO_USE_RENDERING);
if (!kms_display->dummy_gbm_surface)
{
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_CREATE_CONTEXT,
"Failed to create dummy GBM surface");
return FALSE;
}
egl_display->dummy_surface =
eglCreateWindowSurface (egl_renderer->edpy,
egl_display->egl_config,
(EGLNativeWindowType)
kms_display->dummy_gbm_surface,
NULL);
if (egl_display->dummy_surface == EGL_NO_SURFACE)
{
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_CREATE_CONTEXT,
"Failed to create dummy EGL surface");
return FALSE;
}
}
if (!_cogl_winsys_egl_make_current (display,
egl_display->dummy_surface,
egl_display->dummy_surface,
egl_display->egl_context))
{
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_CREATE_CONTEXT,
"Failed to make context current");
return FALSE;
}
return TRUE;
}
static void
_cogl_winsys_egl_cleanup_context (CoglDisplay *display)
{
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
CoglRenderer *renderer = display->renderer;
CoglRendererEGL *egl_renderer = renderer->winsys;
if (egl_display->dummy_surface != EGL_NO_SURFACE)
{
eglDestroySurface (egl_renderer->edpy, egl_display->dummy_surface);
egl_display->dummy_surface = EGL_NO_SURFACE;
}
if (kms_display->dummy_gbm_surface != NULL)
{
gbm_surface_destroy (kms_display->dummy_gbm_surface);
kms_display->dummy_gbm_surface = NULL;
}
}
static void
_cogl_winsys_onscreen_swap_buffers_with_damage (CoglOnscreen *onscreen,
const int *rectangles,
int n_rectangles)
{
CoglContext *context = COGL_FRAMEBUFFER (onscreen)->context;
CoglDisplayEGL *egl_display = context->display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
CoglRenderer *renderer = context->display->renderer;
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
CoglOnscreenEGL *egl_onscreen = onscreen->winsys;
CoglOnscreenKMS *kms_onscreen = egl_onscreen->platform;
uint32_t handle, stride;
CoglFlipKMS *flip;
/* If we already have a pending swap then block until it completes */
while (kms_onscreen->next_fb_id != 0)
handle_drm_event (kms_renderer);
if (kms_onscreen->pending_egl_surface)
{
eglDestroySurface (egl_renderer->edpy, egl_onscreen->egl_surface);
egl_onscreen->egl_surface = kms_onscreen->pending_egl_surface;
kms_onscreen->pending_egl_surface = NULL;
_cogl_framebuffer_winsys_update_size (COGL_FRAMEBUFFER (kms_display->onscreen),
kms_display->width, kms_display->height);
context->current_draw_buffer_changes |= COGL_FRAMEBUFFER_STATE_BIND;
}
parent_vtable->onscreen_swap_buffers_with_damage (onscreen,
rectangles,
n_rectangles);
if (kms_onscreen->pending_surface)
{
free_current_bo (onscreen);
gbm_surface_destroy (kms_onscreen->surface);
kms_onscreen->surface = kms_onscreen->pending_surface;
kms_onscreen->pending_surface = NULL;
}
/* Now we need to set the CRTC to whatever is the front buffer */
kms_onscreen->next_bo = gbm_surface_lock_front_buffer (kms_onscreen->surface);
#if (COGL_VERSION_ENCODE (COGL_GBM_MAJOR, COGL_GBM_MINOR, COGL_GBM_MICRO) >= \
COGL_VERSION_ENCODE (8, 1, 0))
stride = gbm_bo_get_stride (kms_onscreen->next_bo);
#else
stride = gbm_bo_get_pitch (kms_onscreen->next_bo);
#endif
handle = gbm_bo_get_handle (kms_onscreen->next_bo).u32;
if (drmModeAddFB (kms_renderer->fd,
kms_display->width,
kms_display->height,
24, /* depth */
32, /* bpp */
stride,
handle,
&kms_onscreen->next_fb_id))
{
g_warning ("Failed to create new back buffer handle: %m");
gbm_surface_release_buffer (kms_onscreen->surface,
kms_onscreen->next_bo);
kms_onscreen->next_bo = NULL;
kms_onscreen->next_fb_id = 0;
return;
}
/* If this is the first framebuffer to be presented then we now setup the
* crtc modes, else we flip from the previous buffer */
if (kms_display->pending_set_crtc)
{
setup_crtc_modes (context->display, kms_onscreen->next_fb_id);
kms_display->pending_set_crtc = FALSE;
}
flip = g_slice_new0 (CoglFlipKMS);
flip->onscreen = onscreen;
flip_all_crtcs (context->display, flip, kms_onscreen->next_fb_id);
if (flip->pending == 0)
{
drmModeRmFB (kms_renderer->fd, kms_onscreen->next_fb_id);
gbm_surface_release_buffer (kms_onscreen->surface,
kms_onscreen->next_bo);
kms_onscreen->next_bo = NULL;
kms_onscreen->next_fb_id = 0;
g_slice_free (CoglFlipKMS, flip);
flip = NULL;
queue_swap_notify_for_onscreen (onscreen);
}
else
{
/* Ensure the onscreen remains valid while it has any pending flips... */
cogl_object_ref (flip->onscreen);
/* Process flip right away if we can't wait for vblank */
if (kms_renderer->page_flips_not_supported)
{
setup_crtc_modes (context->display, kms_onscreen->next_fb_id);
process_flip (flip);
}
}
}
static CoglBool
_cogl_winsys_egl_context_init (CoglContext *context,
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
CoglError **error)
{
COGL_FLAGS_SET (context->features,
COGL_FEATURE_ID_SWAP_BUFFERS_EVENT, TRUE);
/* TODO: remove this deprecated feature */
COGL_FLAGS_SET (context->winsys_features,
COGL_WINSYS_FEATURE_SWAP_BUFFERS_EVENT,
TRUE);
COGL_FLAGS_SET (context->winsys_features,
COGL_WINSYS_FEATURE_SYNC_AND_COMPLETE_EVENT,
TRUE);
return TRUE;
}
static CoglBool
_cogl_winsys_onscreen_init (CoglOnscreen *onscreen,
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
CoglError **error)
{
CoglFramebuffer *framebuffer = COGL_FRAMEBUFFER (onscreen);
CoglContext *context = framebuffer->context;
CoglDisplay *display = context->display;
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
CoglRenderer *renderer = display->renderer;
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
CoglOnscreenEGL *egl_onscreen;
CoglOnscreenKMS *kms_onscreen;
_COGL_RETURN_VAL_IF_FAIL (egl_display->egl_context, FALSE);
if (kms_display->onscreen)
{
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_CREATE_ONSCREEN,
"Cannot have multiple onscreens in the KMS platform");
return FALSE;
}
kms_display->onscreen = onscreen;
onscreen->winsys = g_slice_new0 (CoglOnscreenEGL);
egl_onscreen = onscreen->winsys;
kms_onscreen = g_slice_new0 (CoglOnscreenKMS);
egl_onscreen->platform = kms_onscreen;
kms_onscreen->surface =
gbm_surface_create (kms_renderer->gbm,
kms_display->width,
kms_display->height,
GBM_BO_FORMAT_XRGB8888,
GBM_BO_USE_SCANOUT |
GBM_BO_USE_RENDERING);
if (!kms_onscreen->surface)
{
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_CREATE_ONSCREEN,
"Failed to allocate surface");
return FALSE;
}
egl_onscreen->egl_surface =
eglCreateWindowSurface (egl_renderer->edpy,
egl_display->egl_config,
(EGLNativeWindowType) kms_onscreen->surface,
NULL);
if (egl_onscreen->egl_surface == EGL_NO_SURFACE)
{
Adds CoglError api Although we use GLib internally in Cogl we would rather not leak GLib api through Cogl's own api, except through explicitly namespaced cogl_glib_ / cogl_gtype_ feature apis. One of the benefits we see to not leaking GLib through Cogl's public API is that documentation for Cogl won't need to first introduce the Glib API to newcomers, thus hopefully lowering the barrier to learning Cogl. This patch provides a Cogl specific typedef for reporting runtime errors which by no coincidence matches the typedef for GError exactly. If Cogl is built with --enable-glib (default) then developers can even safely assume that a CoglError is a GError under the hood. This patch also enforces a consistent policy for when NULL is passed as an error argument and an error is thrown. In this case we log the error and abort the application, instead of silently ignoring it. In common cases where nothing has been implemented to handle a particular error and/or where applications are just printing the error and aborting themselves then this saves some typing. This also seems more consistent with language based exceptions which usually cause a program to abort if they are not explicitly caught (which passing a non-NULL error signifies in this case) Since this policy for NULL error pointers is stricter than the standard GError convention, there is a clear note in the documentation to warn developers that are used to using the GError api. Reviewed-by: Neil Roberts <neil@linux.intel.com> (cherry picked from commit b068d5ea09ab32c37e8c965fc8582c85d1b2db46) Note: Since we can't change the Cogl 1.x api the patch was changed to not rename _error_quark() functions to be _error_domain() functions and although it's a bit ugly, instead of providing our own CoglError type that's compatible with GError we simply #define CoglError to GError unless Cogl is built with glib disabled. Note: this patch does technically introduce an API break since it drops the cogl_error_get_type() symbol generated by glib-mkenum (Since the CoglError enum was replaced by a CoglSystemError enum) but for now we are assuming that this will not affect anyone currently using the Cogl API. If this does turn out to be a problem in practice then we would be able to fix this my manually copying an implementation of cogl_error_get_type() generated by glib-mkenum into a compatibility source file and we could also define the original COGL_ERROR_ enums for compatibility too. Note: another minor concern with cherry-picking this patch to the 1.14 branch is that an api scanner would be lead to believe that some APIs have changed, and for example the gobject-introspection parser which understands the semantics of GError will not understand the semantics of CoglError. We expect most people that have tried to use gobject-introspection with Cogl already understand though that it is not well suited to generating bindings of the Cogl api anyway and we aren't aware or anyone depending on such bindings for apis involving GErrors. (GnomeShell only makes very-very minimal use of Cogl via the gjs bindings for the cogl_rectangle and cogl_color apis.) The main reason we have cherry-picked this patch to the 1.14 branch even given the above concerns is that without it it would become very awkward for us to cherry-pick other beneficial patches from master.
2012-08-31 14:28:27 -04:00
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_CREATE_ONSCREEN,
"Failed to allocate surface");
return FALSE;
}
_cogl_framebuffer_winsys_update_size (framebuffer,
kms_display->width,
kms_display->height);
return TRUE;
}
static void
_cogl_winsys_onscreen_deinit (CoglOnscreen *onscreen)
{
CoglFramebuffer *framebuffer = COGL_FRAMEBUFFER (onscreen);
CoglContext *context = framebuffer->context;
CoglDisplay *display = context->display;
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
CoglRenderer *renderer = context->display->renderer;
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglOnscreenEGL *egl_onscreen = onscreen->winsys;
CoglOnscreenKMS *kms_onscreen;
/* If we never successfully allocated then there's nothing to do */
if (egl_onscreen == NULL)
return;
kms_display->onscreen = NULL;
kms_onscreen = egl_onscreen->platform;
/* flip state takes a reference on the onscreen so there should
* never be outstanding flips when we reach here. */
g_return_if_fail (kms_onscreen->next_fb_id == 0);
free_current_bo (onscreen);
if (egl_onscreen->egl_surface != EGL_NO_SURFACE)
{
eglDestroySurface (egl_renderer->edpy, egl_onscreen->egl_surface);
egl_onscreen->egl_surface = EGL_NO_SURFACE;
}
if (kms_onscreen->surface)
{
gbm_surface_destroy (kms_onscreen->surface);
kms_onscreen->surface = NULL;
}
g_slice_free (CoglOnscreenKMS, kms_onscreen);
g_slice_free (CoglOnscreenEGL, onscreen->winsys);
onscreen->winsys = NULL;
}
static const CoglWinsysEGLVtable
_cogl_winsys_egl_vtable =
{
.display_setup = _cogl_winsys_egl_display_setup,
.display_destroy = _cogl_winsys_egl_display_destroy,
.context_created = _cogl_winsys_egl_context_created,
.cleanup_context = _cogl_winsys_egl_cleanup_context,
.context_init = _cogl_winsys_egl_context_init
};
const CoglWinsysVtable *
_cogl_winsys_egl_kms_get_vtable (void)
{
static CoglBool vtable_inited = FALSE;
static CoglWinsysVtable vtable;
if (!vtable_inited)
{
/* The EGL_KMS winsys is a subclass of the EGL winsys so we
start by copying its vtable */
parent_vtable = _cogl_winsys_egl_get_vtable ();
vtable = *parent_vtable;
vtable.id = COGL_WINSYS_ID_EGL_KMS;
vtable.name = "EGL_KMS";
vtable.renderer_connect = _cogl_winsys_renderer_connect;
vtable.renderer_disconnect = _cogl_winsys_renderer_disconnect;
vtable.onscreen_init = _cogl_winsys_onscreen_init;
vtable.onscreen_deinit = _cogl_winsys_onscreen_deinit;
/* The KMS winsys doesn't support swap region */
vtable.onscreen_swap_region = NULL;
vtable.onscreen_swap_buffers_with_damage =
_cogl_winsys_onscreen_swap_buffers_with_damage;
vtable_inited = TRUE;
}
return &vtable;
}
void
cogl_kms_renderer_set_kms_fd (CoglRenderer *renderer,
int fd)
{
_COGL_RETURN_IF_FAIL (cogl_is_renderer (renderer));
/* NB: Renderers are considered immutable once connected */
_COGL_RETURN_IF_FAIL (!renderer->connected);
renderer->kms_fd = fd;
}
struct gbm_device *
cogl_kms_renderer_get_gbm (CoglRenderer *renderer)
{
_COGL_RETURN_VAL_IF_FAIL (cogl_is_renderer (renderer), NULL);
if (renderer->connected)
{
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
return kms_renderer->gbm;
}
else
return NULL;
}
int
cogl_kms_renderer_get_kms_fd (CoglRenderer *renderer)
{
_COGL_RETURN_VAL_IF_FAIL (cogl_is_renderer (renderer), -1);
if (renderer->connected)
{
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
return kms_renderer->fd;
}
else
return -1;
}
void
cogl_kms_display_queue_modes_reset (CoglDisplay *display)
{
if (display->setup)
{
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
kms_display->pending_set_crtc = TRUE;
}
}
CoglBool
cogl_kms_display_set_layout (CoglDisplay *display,
int width,
int height,
CoglKmsCrtc **crtcs,
int n_crtcs,
CoglError **error)
{
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
CoglRenderer *renderer = display->renderer;
CoglRendererEGL *egl_renderer = renderer->winsys;
CoglRendererKMS *kms_renderer = egl_renderer->platform;
GList *crtc_list;
int i;
if ((width != kms_display->width ||
height != kms_display->height) &&
kms_display->onscreen)
{
CoglOnscreenEGL *egl_onscreen = kms_display->onscreen->winsys;
CoglOnscreenKMS *kms_onscreen = egl_onscreen->platform;
struct gbm_surface *new_surface;
EGLSurface new_egl_surface;
/* Need to drop the GBM surface and create a new one */
new_surface = gbm_surface_create (kms_renderer->gbm,
width, height,
GBM_BO_FORMAT_XRGB8888,
GBM_BO_USE_SCANOUT |
GBM_BO_USE_RENDERING);
if (!new_surface)
{
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_CREATE_ONSCREEN,
"Failed to allocate new surface");
return FALSE;
}
new_egl_surface =
eglCreateWindowSurface (egl_renderer->edpy,
egl_display->egl_config,
(EGLNativeWindowType) new_surface,
NULL);
if (new_egl_surface == EGL_NO_SURFACE)
{
_cogl_set_error (error, COGL_WINSYS_ERROR,
COGL_WINSYS_ERROR_CREATE_ONSCREEN,
"Failed to allocate new surface");
gbm_surface_destroy (new_surface);
return FALSE;
}
if (kms_onscreen->pending_egl_surface)
eglDestroySurface (egl_renderer->edpy, kms_onscreen->pending_egl_surface);
if (kms_onscreen->pending_surface)
gbm_surface_destroy (kms_onscreen->pending_surface);
kms_onscreen->pending_surface = new_surface;
kms_onscreen->pending_egl_surface = new_egl_surface;
}
kms_display->width = width;
kms_display->height = height;
g_list_free_full (kms_display->crtcs, (GDestroyNotify) crtc_free);
crtc_list = NULL;
for (i = 0; i < n_crtcs; i++)
{
crtc_list = g_list_prepend (crtc_list, crtc_copy (crtcs[i]));
}
crtc_list = g_list_reverse (crtc_list);
kms_display->crtcs = crtc_list;
kms_display->pending_set_crtc = TRUE;
return TRUE;
}
void
cogl_kms_display_set_ignore_crtc (CoglDisplay *display,
uint32_t id,
CoglBool ignore)
{
CoglDisplayEGL *egl_display = display->winsys;
CoglDisplayKMS *kms_display = egl_display->platform;
GList *l;
for (l = kms_display->crtcs; l; l = l->next)
{
CoglKmsCrtc *crtc = l->data;
if (crtc->id == id)
{
crtc->ignore = ignore;
break;
}
}
}