Commit Graph

6407 Commits

Author SHA1 Message Date
Jonas Ådahl
c7be8fd398 wayland: Don't set cursor surface scale when we don't have a surface
The setting of the the surface scale even when the surface is not
created from a surface was introduced due to a crash when getting the
buffers when dividing by the scale. The only reason I can see this is
that we get the buffer from a non-existing surface when the wl_cursor
has not yet been set.

Instead, use the name field to avoid trying to use the non-existing
surface, effectively avoiding the division-by-zero that way.

https://bugzilla.gnome.org/show_bug.cgi?id=746141
2015-03-16 19:53:37 +08:00
Jonas Ådahl
b4344861a0 wayland: Only bind to supported gtk-shell versions
The gtk-shell Wayland protocol extension is not meant to be backward
compatible right now, so avoid binding to any version that is not the
one supported.

https://bugzilla.gnome.org/show_bug.cgi?id=745721
2015-03-11 11:37:37 +08:00
Philip Withnall
586240d2f8 gdkwindow: Fix potential NULL pointer dereference
gdk_window_ensure_native() can end up with a NULL parent pointer, which
it passes to find_native_parent_above()…but that expects a non-NULL
parent.

Found with scan-build.

https://bugzilla.gnome.org/show_bug.cgi?id=712760
2015-03-09 13:41:37 +00:00
Eric Le Bihan
8ecc51ad8a Fix undefined reference to get_xkb()
When configuring Gtk+ with --disable-xkb, the build fails because of an
undefined reference to get_xkb().

This patch fixes this issue.

Signed-off-by: Eric Le Bihan <eric.le.bihan.dev@free.fr>

https://bugzilla.gnome.org/show_bug.cgi?id=739070
2015-03-07 18:42:27 -05:00
Niels Nesse
b87715973f gl: Warn the user if they request a GL context version less than 3.2
If the user requests a version less than 3.2 the version is forced to 3.2.
Previous checking code have an inconsistent behavior depending on which
minor version number was specified. This is avoided now by temporarily
converting the major/minor pair into a single integer.

https://bugzilla.gnome.org/show_bug.cgi?id=744288
2015-03-06 14:02:55 +01:00
Carlos Garnacho
4cae9bdd05 x11: Store last axes from device
And use these for the missing axes if the valuator mask is incomplete.
This used to work fine on tablets because the Wacom driver ensures all
valuators are sent, which is not true if using the evdev driver.

https://bugzilla.gnome.org/show_bug.cgi?id=703610
2015-03-02 18:57:04 +01:00
Carlos Garnacho
eaebc2cb4b wayland: Delay wl_subsurface interface creation until the window is shown
It isn't really necessary anytime before, so just make sure it's there only
when the window is visible.

https://bugzilla.gnome.org/show_bug.cgi?id=743427
2015-03-02 12:11:59 +01:00
Carlos Garnacho
c55fcd35fe wayland: Thaw the clock if we hide a GdkWindow mid-frame
When a window is hidden, its surface and all its roles are destroyed,
if this happens when we already issued a wl_surface_commit and are
awaiting for a frame callback, the clock will remain frozen for the
next time the window is shown.

To avoid this, keep track of the wl_surface_frame() calls issued,
and ensure the clock is thawed after hiding. If we happen to receive
the frame callback, it is just ignored.

https://bugzilla.gnome.org/show_bug.cgi?id=743427
2015-03-02 12:11:59 +01:00
Ryan Lortie
c91076f64c gdkglcontext-win32: fix variable names in GDK_NOTE
These probably weren't caught because GDK_NOTE was disabled when the
change was tested.

https://bugzilla.gnome.org/show_bug.cgi?id=745400
2015-03-01 21:38:19 -05:00
Ryan Lortie
4f0ab5656b gdkglcontext-win32: remove unused variables
https://bugzilla.gnome.org/show_bug.cgi?id=745400
2015-03-01 21:38:19 -05:00
Matthias Clasen
af86bc58b9 wayland: Add debug spew 2015-02-28 00:58:05 -05:00
Matthias Clasen
94c5d691df wayland: Formatting fixes 2015-02-28 00:02:30 -05:00
Matthias Clasen
a0eb0e2346 wayland: Apply maximized and fullscreen state
We were just throwing the request away if the app asks to
fullscreen or maximize a window before it has been mapped.
This is something the GdkWindow API explicitly supports,
so make it work by saving the state until the surface exists.

This fixes things under weston. There are bugs in mutter
that keep this from working correctly with gnome-shell.

https://bugzilla.gnome.org/show_bug.cgi?id=745303
2015-02-27 19:09:03 -05:00
Matthias Clasen
3d205a2037 wayland: Drop some pointless lines
No need for these parameter checks, really.
2015-02-27 19:09:03 -05:00
Matthias Clasen
6ccef21d49 Fix a typo
Follow-up fix for commit 7cee6c5cc6.
2015-02-27 17:00:08 -05:00
Olivier Fourdan
53028ff3bb wayland: Don't use g_error() on connection lost
When the Wayland compositor vanishes, all applications connected will
receive a SIGPIPE as soon as they try to use wl_display_dispatch().

Do not use g_error() to terminate the applications when this occurs,
g_error() means an error in the application while here it's not truly
the case.

Use g_warning() and exit() instead.

Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>

https://bugzilla.gnome.org/show_bug.cgi?id=745289
2015-02-27 16:26:39 -05:00
Emmanuele Bassi
7cee6c5cc6 gdk: Use a better error message
Having "Assertion failed: success" on the console is not going to help
anyone.
2015-02-27 21:02:16 +00:00
Jonas Ådahl
f13157f4bb wayland: Don't allocate a full size SHM buffer when drawing using OpenGL
Before this patch, we'd always allocate a full size SHM buffer via
the wl_shm_pool, even though it would never be used. Instead allocate a
logical 1x1 cairo image surface.

https://bugzilla.gnome.org/show_bug.cgi?id=745076
2015-02-24 20:22:03 -05:00
Jonas Ådahl
aa81262f11 wayland: Scale up wl_egl_window according to window scale
In order to support window scales for EGL windows, resize the
wl_egl_window to the window dimension multiplied with the window scale,
just as with SHM window buffers.

https://bugzilla.gnome.org/show_bug.cgi?id=745076
2015-02-24 20:22:03 -05:00
Jonas Ådahl
56c9eac16e wayland: Properly support changing of surface scale
When the preferred surface scale changes, for example when entering a
wl_output with a higher scale than any previous entered output, recreate
the shm surface and redraw the window content with the new window scale.

Before this patch, the internal scale would be changed, but the shm
surface would not be recreated given the new scale, i.e. we'd attach a
buffer for a different scale than wl_surface.set_scale specified.

https://bugzilla.gnome.org/show_bug.cgi?id=745076
2015-02-24 20:22:02 -05:00
Jonas Ådahl
a70b478a5c wayland: Don't ever try to set surface buffer scale when not supported
If the compositor is too old for handling surface buffer scales, never
tyr to set change it. This will effectively always leave it to its
initial state, i.e. 1.

https://bugzilla.gnome.org/show_bug.cgi?id=745076
2015-02-24 20:22:02 -05:00
Benjamin Otte
40f2af665d gdkcairo: Bail if surface is in error
Don't try to paint onto an error surface. This happens for example when
gdk_cairo_set_source_pixbuf() is called with a pixbuf that is too big
for Cairo to handle.

Spotted by Christian Boxdörfer
2015-02-25 01:55:28 +01:00
Thibault Saunier
fb581a1084 gdk/quartz: Fix mixup in attributes usage creating window implementation
First, attributes can be NULL (which is always the case when calling
gdk_window_ensure_native) so do not unconditionally dereference it.

Then the window_type should be taken directly from the GdkWindow as
in other backends (such as the X11 one for example).

https://bugzilla.gnome.org/show_bug.cgi?id=744942
2015-02-22 15:50:39 +01:00
Philip Withnall
09bb109f01 docs: Expand input handling documentation to cover event masks
Also try and clarify a few things about event propagation. Move
input-handling.xml into gtk-doc’s expand_content_files variable so it
automatically links to widget documentation. Add links from
gtk_widget_add_events() and friends to the new documentation.

https://bugzilla.gnome.org/show_bug.cgi?id=744054
2015-02-19 12:48:42 +00:00
Carlos Garnacho
e43ec48e83 wayland: Use display from GdkDragContext device
It will be useless to check the source window on the destination side,
it's at the moment always NULL. Fetch the display from the device instead,
which will be set for every GdkDragContext.
2015-02-17 19:50:05 +01:00
Paolo Borelli
f4510c93b7 Do not use "return" when chaining up to a void function 2015-02-17 14:54:51 +01:00
Jonas Ådahl
5889905d1d wayland: Update to xdg-shell unstable v5
Only trivial changes needed.

https://bugzilla.gnome.org/show_bug.cgi?id=744455
2015-02-14 20:54:47 -05:00
Pier Luigi Fiorini
6ae521bb32 wayland: Do not require latest wl_seat
Some compositors might not offer wl_seat 4 resulting in GTK+ clients not
working on that compositor.

wl_seat 4 introduces keyboard repeat information, but when that information
is missing it is retrieved from settings, hence there's no reason to
require wl_seat 4.

This patch was tested against QtCompositor (5.5, dev branch)
and Weston 1.6.1.

Reviewed-by: Daniel Stone <daniels@collabora.com>

https://bugzilla.gnome.org/show_bug.cgi?id=744172
2015-02-13 19:05:32 -05:00
Chun-wei Fan
933ee3af56 gdkglcontext-win32.c: Fix Build on C89 Compilers
Make sure variables are declared at the top of the block.
2015-02-13 18:36:34 +08:00
Emmanuele Bassi
d066e7541e Remove GdkGLProfile
The existence of OpenGL implementations that do not provide the full
core profile compatibility because of reasons beyond the technical, like
llvmpipe not implementing floating point buffers, makes the existence of
GdkGLProfile and documenting the fact that we use core profiles a bit
harder.

Since we do not have any existing profile except the default, we can
remove the GdkGLProfile and its related API from GDK and GTK+, and sweep
the whole thing under the carpet, while we wait for an extension that
lets us ask for the most compatible profile possible.

https://bugzilla.gnome.org/show_bug.cgi?id=744407
2015-02-12 17:51:31 +00:00
Emmanuele Bassi
390a42fa60 gl: Add gdk_gl_context_get_version()
Store the OpenGL version when we first do the extensions check; this
allows client code to check the available GL version without requiring a
call to gdk_gl_context_make_current() and epoxy_gl_version().
2015-02-12 14:29:44 +00:00
Emmanuele Bassi
f52a59d414 docs: Specify the minimum version of GL provided by the core profile
When using GDK_GL_PROFILE_3_2_CORE, we are not only specifying that the
GDK should create a core profile; we are also specifying that the
minimum required version of OpenGL is set to 3.2.

We should also specify that the GDK_GL_PROFILE_DEFAULT profile is an
alias for GDK_GL_PROFILE_3_2_CORE.
2015-02-12 13:32:02 +00:00
Emmanuele Bassi
022724aa69 gl: Drop profile for gdk_window_create_gl_context()
Now that we have a two-stages GL context creation sequence, we can move
the profile to a pre-realize option, like the debug and forward
compatibility bits, or the GL version to use.
2015-02-12 12:34:28 +00:00
Armin K
6886f479fe wayland: Implement minimize support
Use the existing xdg_surface API for minimization.

https://bugzilla.gnome.org/show_bug.cgi?id=695945
2015-02-11 22:01:38 -05:00
Marco Trevisan (Treviño)
c82ad2c129 mir, gl: Ensure we use the 3.2 core profile
Emit an error if the profile is different.
2015-02-12 03:34:05 +01:00
Marco Trevisan (Treviño)
ce31df3238 mir: Switch GDK_GL_PROFILE_DEFAULT to mean 3_2_CORE
Instead of LEGACY.
2015-02-12 03:32:11 +01:00
Marco Trevisan (Treviño)
f9503ac189 mir, GL: Split GL context creation in two phases
Move egl_context initialization in gdk_gl_context_realize

 • gdk_window_create_gl_context() creates a GdkGLContext
 • gdk_gl_context_realize() creates the underlying resources

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-12 03:03:18 +01:00
Emmanuele Bassi
5666522ab3 gl: Fix the required version setter
We don't allow a version tuple to be lower than (3, 2), but we should
allow versions above that.

https://bugzilla.gnome.org/show_bug.cgi?id=744212
2015-02-10 10:16:53 +00:00
Chun-wei Fan
60195ab021 win32/gl: Ensure we use the 3.2 core profile
Emit an error if the profile is different.

This is a follow-up commit to commits cc45e82 (x11/gl: Ensure we use the
3.2 core profile) and 2d9081d (wayland/gl: Ensure we use the 3.2 core
profile), so that we do the same on GDK-Win32.  Update variable names and
comments so that the code is clearer to people, as we still need a
temporary legacy WGL context first before we can use
wglCreateContextAttribsARB() to create a WGL core (3.2+) context.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-10 16:21:29 +08:00
Emmanuele Bassi
03f5d5a5e4 gl: Put shaders in static storage
We are not changing the shader code.

The compiler will hopefully put them in the .rodata section.
2015-02-10 03:33:02 +00:00
Niels Nesse
50187ae86d gdkgl should use shader version 1.50
Using version 1.20 with a forward compatible 3.2 core context is incorrect
since OpenGL 3.2 deprecates shader version 1.20 (See section E.2). The latest
fglrx drivers will not compile these shaders.

https://bugzilla.gnome.org/show_bug.cgi?id=744203
2015-02-10 03:30:09 +00:00
Emmanuele Bassi
843475bd2e gl: Drop GdkGLContextClass.upload_texture()
It's unnecessary to allow per-backend overrides.
2015-02-09 19:10:31 +00:00
Emmanuele Bassi
2d35053d97 gl: Clean up the required version accessors
We can simplify the code, since we only have core GL profiles in GDK.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Emmanuele Bassi
4ad887f4bf docs: We do not support non-core GL profiles
No need to mention that some API only works on core GL profiles in the
GdkGLContext docs.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Emmanuele Bassi
601c49ef39 gl: Clean up pre-requisite checks for GdkGLContext setters
We don't support non-core profiles.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Emmanuele Bassi
2d9081d117 wayland/gl: Ensure we use the 3.2 core profile
Emit an error if the profile is different.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Emmanuele Bassi
cc45e828df x11/gl: Ensure we use the 3.2 core profile
Drop the dead code for the legacy profile, and return an error when
realizing if the profile is incorrect.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Emmanuele Bassi
1d3cc65e89 gl: Add more debugging notes
https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Emmanuele Bassi
395125bcfb gl: Do not use the extension API for core GL
Since we are using a Core GL profile, we need to drop the
extension-based API.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Emmanuele Bassi
4b8b3b431b gl: Drop OpenGL legacy profile
We simply don't want to care about legacy OpenGL.

All supported platforms also have support for OpenGL ≥ 3.2; it would
complicate the internal code; and would force us to use legacy GL
contexts internally if the first context created by the user is a legacy
GL context, and disable creation of core-3.2 contexts after that.

We will need to fix all our code examples to use the Core 3.2 profile.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Emmanuele Bassi
6aaa6c337b gl: Move getters for context options to the public API
They can be useful for third party code as well.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Emmanuele Bassi
f7497daee9 Switch GDK_GL_PROFILE_DEFAULT to mean 3_2_CORE
Instead of LEGACY.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Chun-wei Fan
4c091db6c2 win32/gl: Use the GdkGLContext options
https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:30 +00:00
Chun-wei Fan
ba56f11702 GDK-Win32: Split GL context creation in two phases
Like what is being done in the X11 and Wayland backends, create the
GdkWin32GLContext in 2 steps, where we only create the actual WGL context
in _gdk_win32_gl_context_realize().

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:28 +00:00
Emmanuele Bassi
42a895e85e wayland/gl: Use the GdkGLContext options
https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:05 +00:00
Emmanuele Bassi
3425f7fe5a x11/gl: Use the GdkGLContext options
When creating an OpenGL context using the glXCreateContextAttribs()
function.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:05 +00:00
Emmanuele Bassi
fa90052299 gl: Add context options
Users of the GdkGLContext API should be allowed to set properties on the
shim GdkGLContext instance prior to realization, so that the
backend-specific implementation can use the value of those properties
when creating the windowing system specific resources.

The main three options are:

 • a major/minor version tuple, to request a specific GL version
 • a debug bit, to request a "debug context", which provides additional
   validation and run time checking
 • a forward compatibility bit, to request a context that does not
   have deprecated functionality

See also:
 - https://www.opengl.org/registry/specs/ARB/glx_create_context.txt

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:05 +00:00
Emmanuele Bassi
22e6f37c9c GL: Split GL context creation in two phases
One of the major requests by OpenGL users has been the ability to
specify settings when creating a GL context, like the version to use
or whether the debug support should be enabled.

We have a couple of requirements in terms of API:

 • avoid, if at all possible, the "C arrays of integers with
   attribute, value pairs", which are hard to write and hard
   to bind in non-C languages.
 • allow failing in a recoverable way.
 • do not make the GL context creation API a mess of arguments.

Looking at prior art, it seems that a common pattern is to split the
construction phase in two:

 • a first phase that creates a GL context wrapper object and
   does preliminary checks on the environment.
 • a second phase that creates the backend-specific GL object.

We adopted a similar pattern:

 • gdk_window_create_gl_context() creates a GdkGLContext
 • gdk_gl_context_realize() creates the underlying resources

Calling gdk_gl_context_make_current() also realizes the context, so
simple GL users do not need to care. Advanced users will want to
call gdk_window_create_gl_context(), set up the optional requirements,
and then call gdk_gl_context_realize(). If either of these two steps
fails, it's possible to recover by changing the requirements, or simply
creating a new GdkGLContext instance.

https://bugzilla.gnome.org/show_bug.cgi?id=741946
2015-02-09 19:10:04 +00:00
Chun-wei Fan
0b06b1e3f7 gdkglcontext-win32.h: Clean Up a Bit
We don't have a Win32-specific ->upload_texture() anymore, so don't
indicate so.  Left out this part from my previous commit.
2015-02-06 13:00:59 +08:00
Chun-wei Fan
24e6bda264 GDK-Win32: Drop Win32-specific ->upload_texture()
The default ->upload_texture() works also for Windows since commit 27cf0fa,
as some of the problems described in 742953 also applied for GL core
contexts on Windows as well before 27cf0fa.  Clean up the GDK-Win32 code a
little bit as a result.
2015-02-06 12:41:38 +08:00
William Hua
0a64a4f3c0 mir: release surface before rendering parent
We should remove the mir and cairo surface before rendering the
transient_for, which will regenerate the cairo surface anyways.
Otherwise, we end up releasing both, when we only really want to get rid
of the mir surface.
2015-02-05 17:26:19 +01:00
William Hua
85c5f6058a mir: don't always request mir surfaces
Some child window behaviour is faked by painting in the parent, so we
don't bother creating the Mir surface in that case.
2015-02-05 17:26:19 +01:00
William Hua
bee39d2251 mir: reconstruct surface if window is moved 2015-02-05 17:26:19 +01:00
William Hua
16b250f9bb mir: stop using deprecated surface api 2015-02-05 17:26:19 +01:00
William Hua
5eacc92ba7 mir: fix issues with mouse over parent menu
Mouse over a parent menu[bar] didn't work while the menu was open.
The fix was to correct the behaviour of pointer crossing events so that
the pointer appears to be only inside one window at a time.

See: http://tronche.com/gui/x/xlib/events/window-entry-exit/normal.html
2015-02-05 17:26:19 +01:00
William Hua
6a3904cf0c mir: add back window hierarchy functions 2015-02-05 17:26:19 +01:00
William Hua
c2afa674e2 mir: stop forwarding motion events to subwindows 2015-02-05 17:26:19 +01:00
William Hua
901bc2957d mir: generate leave events for transient children 2015-02-05 17:26:19 +01:00
Marco Trevisan (Treviño)
737e598a17 mir: use monitor plug names more similar to the ones used in Xmir 2015-02-05 17:26:18 +01:00
Marco Trevisan (Treviño)
9a84c3e0f5 mir: only paint background if window has one 2015-02-05 17:26:18 +01:00
Marco Trevisan (Treviño)
38d46aacae mir: use old scrolling system if we get simple integer values 2015-02-05 17:26:18 +01:00
William Hua
584cfff90a mir: end the current grab on ungrab 2015-02-05 17:26:18 +01:00
William Hua
e843d94405 mir: only render non-top-level transients. 2015-02-05 17:26:18 +01:00
Marco Trevisan (Treviño)
9abf990656 mir: handle new MirCloseSurfaceEvent 2015-02-05 17:26:18 +01:00
William Hua
5ac5f6f81a mir: reset cursor_inside on hide/withdraw
Without this, we generate incorrect crossing events for transient
windows.
2015-02-05 17:26:18 +01:00
William Hua
8e0abb4bac mir: disable some debug output 2015-02-05 17:26:18 +01:00
William Hua
4f19be1c82 mir: skip windows that aren't visible 2015-02-05 17:26:18 +01:00
Marco Trevisan (Treviño)
05de5b02cb mir: fix IsModifierKey definition 2015-02-05 17:26:18 +01:00
Marco Trevisan (Treviño)
345140008e mir: check the log events env variable just once 2015-02-05 17:26:18 +01:00
Marco Trevisan (Treviño)
7e5b216511 mir: set event source device as well 2015-02-05 17:26:18 +01:00
William Hua
7c2e3a812a mir: fix time conversion macro 2015-02-05 17:26:17 +01:00
Marco Trevisan (Treviño)
54f9263f19 mir: implement stub _gdk_mir_keymap_key_is_modifier
Using IsModifierKey for now
2015-02-05 17:26:17 +01:00
Marco Trevisan (Treviño)
a8b1e81816 mir: check (using simple keycode checking) if we're handling a modifier key 2015-02-05 17:26:17 +01:00
Marco Trevisan (Treviño)
9e2eb5d2f6 mir: compute the modifiers using the saved button_state on key events 2015-02-05 17:26:17 +01:00
Marco Trevisan (Treviño)
7d4066de41 mir: add a NANO_TO_MILLI conversion macro 2015-02-05 17:26:17 +01:00
Marco Trevisan (Treviño)
6b6af59b9b mir: properly converts mir nanoseconds timestamps to milliseconds 2015-02-05 17:26:17 +01:00
Marco Trevisan (Treviño)
cb19fd50cb mir: use actual events time 2015-02-05 17:26:17 +01:00
William Hua
948afc68f6 mir: add timestamps to generated GdkEvents
We need this because it fixes menu activation. The menu activation code
looks at the time between events to determine if mouse clicks happen too
quickly.
2015-02-05 17:26:17 +01:00
William Hua
f1cfd23cd2 mir: pass motion events down to transient children 2015-02-05 17:26:17 +01:00
William Hua
e2320678ae mir: use parent_window's impl for GDK_INPUT_ONLY windows
Fixes a crash that occurs when the user selects text.
2015-02-05 17:26:17 +01:00
Carlos Garnacho
6f07d5e761 x11: Detect libinput touchpads
These aren't reported as XIDependentTouch devices, so make it poke a
property that's specific to touchpads managed by the libinput driver.
2015-02-02 13:55:38 +01:00
Víctor Manuel Jáquez Leal
64f26db038 gtkx: add missing header
The header file gdk/x11/gdkx-autocleanups.h is not installed, but
it is required by gdkx.h

https://bugzilla.gnome.org/show_bug.cgi?id=743640
2015-02-02 13:47:06 +01:00
Alexander Larsson
a71ff33266 gdk: Add support for g_autoptr()
https://bugzilla.gnome.org/show_bug.cgi?id=743640
2015-02-02 09:28:22 +01:00
Matthias Clasen
31d3be96e4 Be safe against stray booleans
Since gboolean is a typedef for int, the compiler won't complain
about gdk_window_set_event_compression (w, 2). So, make it work.

https://bugzilla.gnome.org/show_bug.cgi?id=742566
2015-02-01 17:30:10 -05:00
Matthias Clasen
561ff51abb Fix an oversight
Commit ff256956b2 introduced a frame_clock_events_paused
flag, but only ever set it to TRUE, instead of unsetting it when
events are resumed. This was leading to assertion failures in
_gdk_display_unpause_events().
2015-01-30 18:42:06 -05:00
Tom Hughes
ff256956b2 Resume events when disconnecting from a frame clock
If we are disconnecting from a frame clock that has paused event
processing and hasn't issued a resume yet make sure we resume the
events or they will stay blocked forever.

https://bugzilla.gnome.org/show_bug.cgi?id=742636
2015-01-30 17:32:58 -05:00
Carlos Garnacho
ea983404ec win32: Don't check the position of a NULL device
This function is given a barely setup GdkEvent, so the GdkDevice field
is still unset, causing warnings and misbehaviors when the position
is queried for it.

Given that the wintab GTK+ code seems to rely somewhat hard on the wintab
device managing the pointer cursor, query the pointer position from the
pointer itself.

https://bugzilla.gnome.org/show_bug.cgi?id=743330
2015-01-28 16:59:29 +00:00
Carlos Garnacho
1543890569 win32: Ensure we can create a window for wintab
The window used NULL as a parent window, which defaults internally to
using the root window of the default screen. But at the time wintab is
initialized, there is no default display/screen yet.

Fix this by retrieving this information from the given GdkDeviceManager,
so we don't have to wait for the display to be in place before
initialization.

https://bugzilla.gnome.org/show_bug.cgi?id=743330
2015-01-28 16:59:29 +00:00
Rui Matos
2dd07f346a gdkscreen-wayland: Don't leak wl_output objects
https://bugzilla.gnome.org/show_bug.cgi?id=743633
2015-01-28 17:17:09 +01:00
Niels Nesse
00834786be GDK GL context's should associate with "window->impl_window" not "window"
In some layouts this inconsistency results in crashes in
gdk_gl_texture_from_surface() since it uses gdk_gl_context_get_window() but
the returned window is not the same as the one that is being painted so
"window->current_paint.surface" is NULL. I saw this problem when packing a
GdkGLArea into a GtkPaned.

https://bugzilla.gnome.org/show_bug.cgi?id=743146
2015-01-28 12:00:30 +00:00
Niels Nesse
27cf0fa34c Fix core context creation in GdkGLContext
- Specifically request GL version when creating context. Just specifying core
profile bit results in the requested version defaulting to 1.0 which causes
the core profile bit to be ignored and an arbitrary compatability context to be
returned.

- Fix GL painting by removing GL calls that have been depricated by the 3.2 core
profile.

- Additionally remove glInvalidateFramebuffer() call, it is not supported by 3.2
core.

https://bugzilla.gnome.org/show_bug.cgi?id=742953
2015-01-28 12:00:30 +00:00
Matthias Clasen
5fc8cf4331 x11: Handle obsolete selection requests without crashing
The ICCCM says:

If the specified property is None, the requestor is an obsolete client.
Owners are encouraged to support these clients by using the specified
target atom as the property name to be used for the reply.

Lets do that, instead of crashing.

https://bugzilla.gnome.org/show_bug.cgi?id=740613

The previous fix for this issue in 732af31424 was incomplete.
2015-01-24 19:32:52 -05:00
Matthias Clasen
95fad690d2 wayland: Don't claim to support selection notification
We don't support selection notification, so don't claim
we do, only to fail later.

https://bugzilla.gnome.org/show_bug.cgi?id=743229
2015-01-24 17:25:43 -05:00
Matthias Clasen
b6551c84c9 Initialize variables
No need to risk valgrind complaints about initialized
values.

https://bugzilla.gnome.org/show_bug.cgi?id=743422
2015-01-24 16:45:04 -05:00
Carlos Garnacho
121fddc4d2 x11: Fetch vendor/product identifiers for input devices
These are retrieved from XInput device properties.

https://bugzilla.gnome.org/show_bug.cgi?id=740758
2015-01-19 15:09:19 +01:00
Carlos Garnacho
5e53676b46 device: Add property/getters for vendor/product identifiers
These are a construct only properties, expected to be filled in from
platform specific code.

https://bugzilla.gnome.org/show_bug.cgi?id=740758
2015-01-19 15:09:19 +01:00
Matthias Clasen
47e5e1c72d Fix a typo in the docs
https://bugzilla.gnome.org/show_bug.cgi?id=743037
2015-01-18 15:26:49 -05:00
Matthias Clasen
9c1063f602 broadway: Only use abstract sockets when they are supported
GIO has a function to check whether abstract sockets are
supported, so use it. This should make broadway work on
Freebsd again.

https://bugzilla.gnome.org/show_bug.cgi?id=742980
2015-01-18 15:24:30 -05:00
Matthias Clasen
811d602842 Make Broadway work on freebsd
It seems that posix_fallocate gives an ENODEV error when
called on an fd opened with shm_open on freebsd. Fix up
the error check to only trigger if we get ENOSPC.

https://bugzilla.gnome.org/show_bug.cgi?id=742980
2015-01-18 15:15:37 -05:00
Matthias Clasen
cf277fabb5 Remove some outdated references
We shouldn't recommend gtk_widget_modify_bg() or
gtk_style_set_background() anymore.
2015-01-18 12:41:07 -05:00
Maks Naumov
6113504495 Fix event->button.y_root when using mouse buttons and dnd canceled
Signed-off-by: Maks Naumov <maksqwe1@ukr.net>

https://bugzilla.gnome.org/show_bug.cgi?id=742785
2015-01-17 16:29:17 -05:00
Matthias Clasen
48865fb9d9 Prefer the Wayland backend
This does not affect X11 sessions, but has the advantage that
we get much more testing of the Wayland backend in Wayland
sessions.
2015-01-17 01:03:10 -05:00
Maks Naumov
430ea2fff6 Properly ignore BadWindow and BadDrawable in get_child_info_handler()
Signed-off-by: Maks Naumov <maksqwe1@ukr.net>

https://bugzilla.gnome.org/show_bug.cgi?id=742771
2015-01-11 14:33:10 -05:00
Emmanuele Bassi
a834078804 x11/gl: Do not specify the GLX context version
If we use GDK_GL_PROFILE_3_2_CORE we are asking for a core profile
according to the GLX_ARB_create_context_profile extension. For that,
we pass the GLX_CONTEXT_CORE_PROFILE_BIT_ARB value for the
GLX_CONTEXT_PROFILE_MASK_ARB attribute.

The specification for the extension says that:

  If the requested OpenGL version is less than 3.2,
  GLX_CONTEXT_PROFILE_MASK_ARB is ignored and the functionality
  of the context is determined solely by the requested version.

Since we're asking for a core profile, we assume a GL version greater
than or equal to 3.2; thus, we don't need to specify the
GLX_CONTEXT_MAJOR_VERSION_ARB or the GLX_CONTEXT_MINOR_VERSION_ARB
attributes, and instead just rely on whatever version GLX gives us.

This seems to work around a strange issue in Mesa; if we ask for a core
profile and any version > 3.0, we get broken rendering on any shared
context we create.
2015-01-10 00:07:51 +00:00
Jasper St. Pierre
ecc64f63e4 gdkeventsource-wayland: Fix style 2014-12-28 18:13:34 -08:00
Rico Tzschichholz
07efae7288 mir: Add missing reference to gdkmir-private.h 2014-12-28 13:34:16 +01:00
Emmanuele Bassi
5d42c55f28 x11/gl: Set the correct minor version
The major version is set twice.

https://bugzilla.gnome.org/show_bug.cgi?id=741954
2014-12-25 11:49:58 +00:00
Philip Chimento
ec37fbe2e2 quartz: NSWindow NULL check in get_scale_factor
Sending backingScaleFactor to a NULL NSWindow will silently give the
value 0 for the scale factor, causing insidious divide-by-zero bugs down
the line. This checks if the NSWindow is NULL first, as seems to happen
throughout the rest of the file.

Note that I don't have a hi-DPI OS X machine to test this on, though.

https://bugzilla.gnome.org/show_bug.cgi?id=738338
2014-12-24 14:57:40 -08:00
Chun-wei Fan
36245b7058 GDK-Win32: Drop Some Unused Items
There were a bit of unused items from the GDK-Win32 work on GL support, so
drop them.
2014-12-22 11:25:49 +08:00
Matthias Clasen
1dbc6d91f0 Don't initialize GL if it has been disabled
We've observed hangs of mutter when it initializes GTK+, which
are caused by initializing GL, which in turn makes xwayland
call back into mutter. With this change, mutter should just
disable GL support in GDK, and things will work.
2014-12-19 12:27:31 -05:00
Matthias Clasen
57d034ce9f Add an XSetting for the session bus ID
This will allow us to identify situations where we are not on
the same session bus as the xsettings provider.

https://bugzilla.gnome.org/show_bug.cgi?id=671802
2014-12-18 23:26:11 -05:00
Chun-wei Fan
536fa88cd8 Add OpenGL Support for the GDK Windows Backend
This adds support for OpenGL to the GDK Windows backend using the WGL API
calls, which enables programs that uses the GTK+ GLArea widgets to work on
Windows as well.

This also adds a simple utility function to query for the version of OpenGL
that is supported by the Windows system, like the one provided by the X11
backend.

Many thanks to Alex (and Emmanuele, who started the OpenGL integration in
GTK+) who offered advice and help along the way, as well as the X11 and
Wayland backend for this work to refer to and to model upon.

https://bugzilla.gnome.org/show_bug.cgi?id=740795
2014-12-17 16:07:02 +08:00
Chun-wei Fan
9fd9f61b00 gdkgl: Use vfunc For Uploading Textures
As the alignments, strides and image formats may be different across
platforms, make the texture upload a vfunc to allow backends to override
the GL commands for uploading textures for the software implementation for
gdk_gl_texture_from_surface(), if necessary.

Suggested by Alex to avoid copying non-trivial portions of code which would
then add maintainenace burden.

https://bugzilla.gnome.org/show_bug.cgi?id=740795
2014-12-17 16:06:25 +08:00
Alexander Larsson
a1a01983f7 gdk: Fix scissoring with scaled windows
Make sure the window scale is always taken into acount when setting up
the scissor regions.
2014-12-15 22:31:48 +01:00
Alexander Larsson
2693496b56 gdk_gl_texture_from_surface - fix software fallback
We can't combine multiple draws into one for the software fallback,
because each quad has a different texture. And we generally don't
want to make a larger single texture because then we would have
to upload more data.
2014-12-15 21:57:02 +01:00
Alexander Larsson
4a4125bcfc gdk_cairo_draw_from_gl - fix drawing with alpha on window scale > 1
The scissoring needs to take into account the window scale.
2014-12-15 21:57:02 +01:00
Michael Natterer
c00cc269c5 broadway: Initial SSL support
Use the new --cert and --key parameters to broadwayd to pass paths to
cert and key files.

https://bugzilla.gnome.org/show_bug.cgi?id=730364
2014-12-15 11:25:43 +00:00
Jasper St. Pierre
4c0b2087c6 gdk: Emit stamp-gc-h in the correct directory 2014-12-12 17:32:02 -08:00
Matthias Clasen
4f96353964 Fix some oversights with header guards
Pointed out in https://bugzilla.gnome.org/show_bug.cgi?id=741254
2014-12-08 19:31:56 -05:00
Marco Trevisan (Treviño)
af2ac29faf gdkgl: pass the correct number of quads to gdk_gl_texture_quads
And call it only if that's the case
2014-12-02 14:59:47 +01:00
Marco Trevisan (Treviño)
cf64f8a76a mir: don't initialize the display if the required pixel formats aren't available 2014-12-01 18:05:30 +01:00
Matthias Clasen
bae4450698 Quiet some make rules 2014-11-30 17:18:54 -05:00
Matthias Clasen
c304f2ea68 docs: Add a missing word 2014-11-28 14:25:42 -05:00
Marco Trevisan (Treviño)
b8925253b6 mir: clear screen, keymap and event source when disposing dpy 2014-11-27 19:25:55 +01:00
Matthias Clasen
732af31424 x11: Handle obsolete selection requests without crashing
The ICCCM says:

If the specified property is None , the requestor is an obsolete client.
Owners are encouraged to support these clients by using the specified
target atom as the property name to be used for the reply.

Lets do that, instead of crashing.

https://bugzilla.gnome.org/show_bug.cgi?id=740613
2014-11-26 22:27:49 -05:00
Jasper St. Pierre
fe8f6f60dc gdkdisplay-x11: Don't do extra work when simply moving the window
We shouldn't bother recalculating all the visible regions and wiping
away the old updating areas if we're simply dragging the window around.
2014-11-26 17:14:53 -08:00
Matthias Clasen
c1ca7986ee Make scale=2 work again
There was a leftover HAVE_CAIRO_SURFACE_SET_DEVICE_SCALE ifdef
that broke things, now that we don't use this define anymore.
2014-11-24 10:02:54 -05:00
Chun-wei Fan
16e863ce65 gdkgl.c: Avoid GCCism in Pointer Arithmetic
NULL-plus-something could be seen by the compiler to attempt to do
arithmetic with void *, which is a GCCism.  Instead, do the math normally
and cast the results as a void *.

https://bugzilla.gnome.org/show_bug.cgi?id=740605
2014-11-24 22:57:25 +08:00
Carlos Garnacho
44c412ecc7 wayland: Pass shadow width to the compositor
Use xdg_surface_set_window_geometry() to tell the compositor about the
shadow widths, this makes some gnome-shell/mutter features (edge resistance,
frames around windows in the overview, side maximization, ...) work alright
with GTK+.

In order to add this, some other places in gdkwindow-wayland had to gain
some knowledge about margins:

- xdg_surface_configure() now syncs the shadow after applying the state,
  and gdk_wayland_window_set_shadow_width() possibly reconfigures the
  window in order to preserve window geometry. This is necessary to keep
  shadows in sync with state/geometry changes, as this does not happen
  all at once.
- xdg_popups relative to an xdg_surface are shown relative to buffer
  coordinates, so the left/top margins must be added there.

https://bugzilla.gnome.org/show_bug.cgi?id=736742
2014-11-23 13:11:08 +01:00
Jasper St. Pierre
10a7a89311 gdkinternals: Document our drawing and painting process better 2014-11-22 11:46:53 -08:00
Jasper St. Pierre
c01e37a9a5 gdkgl: Texture quads in one giant draw call
This requires us to use GL_TRIANGLES and six verts per quad instead
of four, which makes me think it might not be worth it on
well-optimized GL drivers. However, from talking to some driver
developers about it, the GL_TRIANGLES should be faster, since this
means that there's one giant contiguous buffer instead of many small
buffers.

If we were really rendering a lot of quads, I'd use an element buffer
and GL_PRIMITIVE_RESTART, but we're really not ever rendering that
many quads, and the setup cost for that would just be too annoying.
2014-11-22 10:34:47 -08:00
Jasper St. Pierre
6ffba866a6 gdkgl: Use one VBO for both position and uv attributes 2014-11-22 10:17:43 -08:00
Jasper St. Pierre
37ad6e1147 gdkgl: Texture many quads at once for performance reasons
This isn't fully performant yet. To be fully performant, we'd need to
do everything in one giant buffer.
2014-11-22 10:13:58 -08:00
Jasper St. Pierre
7312c01f62 gdkglcontext-x11: Make sure to enable the scissor test 2014-11-22 10:01:51 -08:00
Jasper St. Pierre
37697f1817 gdkgl: Use a GdkTexturedQuad struct to paint quads
We'll soon have a new function that paints multiple quads.
2014-11-22 09:40:10 -08:00
Jasper St. Pierre
28846536cb gdkgl: Clean up program-manipulating code
Abstract the program ID and locations into a "program struct" which
we then use.
2014-11-22 09:29:27 -08:00
Jasper St. Pierre
3f0c84f7ad gdkwindow: Clean up whitespace slightly
This has been bothering me for a while.
2014-11-22 08:12:22 -08:00
Jasper St. Pierre
8976e84ded gdkwindow: Don't pass the region to the impl when beginning a paint
It's unused. At the same time, rename "begin_paint_region" to
"begin_paint". This will help us clean up how GDK painting works
in the future to allow more creative use of double-buffering.
2014-11-22 08:08:34 -08:00
Jasper St. Pierre
ec2925a5bf quartz: Don't bother storing the paint_clip_region
It's unused.
2014-11-22 08:07:28 -08:00
Marco Trevisan (Treviño)
1730bb0e81 mir: initialize surface name with program name 2014-11-21 04:46:47 +01:00
Alexander Larsson
442adf6359 GL: Fix typo in gdk_gl_texture_from_surface
Got a sign wrong in commit 800c712738
2014-11-20 15:37:40 +01:00
Alexander Larsson
cf94da2ca1 GdkGLContext: Remove unused update vfunc
The update virtual function for GdkGLContext is unused and is a
leftover from a previous GL approach. Just remove it.

https://bugzilla.gnome.org/show_bug.cgi?id=739750
2014-11-20 12:38:04 +01:00
Alexander Larsson
800c712738 GL: Fix GL Y coordinate flipping to use unscaled window height
This is needed in the edge case where the X11 backend rounded the actual
size, and the GL flipping really needs the correct window height to
do proper Y coordinate flipping.

https://bugzilla.gnome.org/show_bug.cgi?id=739750
2014-11-20 12:38:04 +01:00
Alexander Larsson
1eb3b34bce Add gdk_window_get_unscaled_size
This is required for the X backend GL integration. If the
window has a height that is not a multiple of the window scale
we can't properly do the y coordinate flipping that GL needs.
Other backends can ignore this and use the default implementation.

https://bugzilla.gnome.org/show_bug.cgi?id=739750
2014-11-20 12:38:04 +01:00
Alexander Larsson
788478dff9 x11: Return the exact pixel coverage in get_frame_extents
Rather than just rounding down the position *and* the size separately
we correctly calculate a rectangle in scaled window coords that fully
covers the real window size. This really only makes a difference
when the window size/position isn't a multiple of the window scale.

https://bugzilla.gnome.org/show_bug.cgi?id=739750
2014-11-20 12:38:04 +01:00
Owen W. Taylor
bd643e0e1a x11: Keep track of the exact size in X pixels of windows
Keep track of the exact size of X windows in underlying pixels; we
generally use the scaled size instead, but to properly handle the GL
viewport for windows that aren't a multiple of window_scale,
we need to know the real size.

https://bugzilla.gnome.org/show_bug.cgi?id=739750
2014-11-20 12:38:04 +01:00
Owen W. Taylor
608c25453b x11: round the scaled size *up* when we get a ConfigureNotify
Although we specify a resize increment to try and get a size that is
a multiple of the window scale, maximization typically wins
over the resize increment, so the window might be odd sized.

Round *up* in this case, rather than down, since it's better to
truncate a line or two at the bottom and right of the window rather
than have a line or two that we don't know what to do with.

https://bugzilla.gnome.org/show_bug.cgi?id=739750
2014-11-20 10:19:27 +01:00
Marco Trevisan (Treviño)
f5d96f7110 mir: use proper naming for mir window type checking macro 2014-11-20 05:51:12 +01:00
Marco Trevisan (Treviño)
99d849412b mir: add public getter for the internal MirSurface 2014-11-20 04:41:45 +01:00
Marco Trevisan (Treviño)
60e185e496 mir: don't use struct MirConnection*, just the latter 2014-11-20 04:41:06 +01:00
Marco Trevisan (Treviño)
6e9db05601 mir: add OpenGL support
Implement GdkGLContext using EGL and use hardware mir surfaces
when a GdkWindow uses gl rendering.

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=740346
2014-11-20 04:33:38 +01:00
Marco Trevisan (Treviño)
e0e46a4f6b mir: update surface type based on events 2014-11-17 17:23:36 +01:00
Marco Trevisan (Treviño)
da8fdab87b mir: fix typo in GDK_IS_WINDOW_MIR definition 2014-11-17 14:02:22 +01:00
Marco Trevisan (Treviño)
acfc4f8910 mir: invert scrolling delta 2014-11-17 14:02:22 +01:00
Marco Trevisan (Treviño)
f5c5e9385e mir: handle more surface types 2014-11-14 23:19:15 +01:00
Marco Trevisan (Treviño)
8bc523ef59 mir: fix compilation in mirscreen 2014-11-14 23:19:15 +01:00
Chun-wei Fan
61286d5927 gdk/gdkgl.c: Fix Typo
The extra ';' actually broke builds on C89 compilers...
2014-11-14 17:26:29 +08:00
Marco Trevisan (Treviño)
c42146d88c mir: get/set window surface type 2014-11-13 04:42:48 +01:00
Marco Trevisan (Treviño)
c7e7e35083 mir: don't request more surface formats than the supported ones 2014-11-13 03:59:29 +01:00
Marco Trevisan (Treviño)
4e4ccb76cd mir: implement screen get_plug_name 2014-11-13 02:26:38 +01:00
Marco Trevisan (Treviño)
a77571c42e mir: clear screen objects correctly 2014-11-13 02:19:29 +01:00
Marco Trevisan (Treviño)
0d024c2df8 mir: don't crash when querying infos for an invalid monitor 2014-11-13 02:07:46 +01:00
Marco Trevisan (Treviño)
c8826ba8c7 mir: clear screen data 2014-11-13 01:59:21 +01:00
Marco Trevisan (Treviño)
98a8c66970 mir: correctly get screen mm sizes 2014-11-13 01:47:56 +01:00
Marco Trevisan (Treviño)
c6de20496c mir: unset config change callback on finalize 2014-11-13 01:34:10 +01:00
Marco Trevisan (Treviño)
92c2e99dda mir: actually pass the GdkMirScreen to display config change cb 2014-11-13 01:30:10 +01:00
Matthias Clasen
d9b02ea4a3 gdk: Make debug-updates a per-display setting
Add private API to set this per-display, and make the existing
gdk_window_set_debug_update function set a global default.
2014-11-10 15:23:14 -05:00
Matthias Clasen
5536d9bcde gdk: Add private API for rendering mode
This will be used to expose rendering mode in the inspector.
2014-11-10 14:39:57 -05:00
Matthias Clasen
8338b7e4cd Move rendering mode into GdkDisplay
This is in preparation for making it runtime-settable in the
inspector.
2014-11-10 14:28:43 -05:00
Javier Jardón
710f332082 configure.ac: Depend on cairo 1.14.0
This is needed for cairo_set_device_scale()
2014-11-10 15:12:17 +00:00
Alexander Larsson
90555e06be gdk_window_end_paint: Flush destination surface after paint
It seems in cairo 1.14 we need this after having painted an image surface
to a X11 window surface (i.e. with GDK_RENDERING=image).
2014-11-10 15:25:31 +01:00
Emmanuele Bassi
cdcd3ca359 Hide GdkWindow libgtk_only API in the private vtable
These are the last two global GDK symbols that have a libgtk_only
suffix.

https://bugzilla.gnome.org/show_bug.cgi?id=739781
2014-11-08 00:20:15 -05:00
Emmanuele Bassi
585af218a3 Move GDK GL flags accessors to the private vtable
This allows us to use the GDK_PRIVATE_CALL macro inside gtk.

https://bugzilla.gnome.org/show_bug.cgi?id=739781
2014-11-08 00:20:15 -05:00
Emmanuele Bassi
eedbec2066 Make global GDK libgtk_only functions more private
The current way of exposing GDK API that should be considered internal
to GTK+ is to append a 'libgtk_only' suffix to the function name; this
is not really safe.

GLib has been using a slightly different approach: a private table of
function pointers, and a macro that allows accessing the desired symbol
inside that vtable.

We can copy the approach, and deprecate the 'libgtk_only' symbols in
lieu of outright removal.

https://bugzilla.gnome.org/show_bug.cgi?id=739781
2014-11-08 00:20:15 -05:00
Jasper St. Pierre
433120ae5e gdkwindow-x11: Fix indentation 2014-11-07 17:39:43 -08:00
Carlos Garnacho
af6eab850e wayland: Schedule a clock tick instead of wl_surface_commit()ting directly
Instead of possibly calling wl_surface_commit() out of
GdkFrameClock::after-paint, tick the transient parent clock so ::after-paint
can be eventually run.

This ensures that the subsurface coordinates (considered part of the state
of the parent) aren't committed untimely, and guaranteed to be orderly with
the wl_subsurface-relative state.

This is a gtk-side fix for https://bugzilla.gnome.org/show_bug.cgi?id=738887
2014-11-07 17:24:10 +01:00
Carlos Garnacho
964abf7423 wayland: Don't translate a NULL region into an empty one
cairo_region_copy(NULL) will effectively return an empty region, as this
function is always meant to return valid memory. This however inverts the
meaning of the NULL region and results in entirely non-clickable windows.
2014-11-07 17:23:54 +01:00
Matthias Clasen
7e405c286f Don't make the gl flags API
We need to export the symbols so they can be used in the
inspector, but we don't really want to make this supported
public API, so keep them out of installed headers.
2014-11-07 07:10:49 -05:00
Matthias Clasen
5d0d1e524e Export API to set GL flags
This will be used in the inspector.
2014-11-06 23:29:16 -05:00
Jasper St. Pierre
22ad7b7782 gdkgl: Make some helper functions static 2014-11-06 13:48:20 -08:00
Ryan Lortie
ec729d0649 mir: rework cursors a bit
Store the cursor name on the cursor (rather than always using its type).

Use this when setting a cursor on a surface.

The mir server will fallback to using standard cursors from the cursor
theme if the name used is not one of those defined by mir, which is more
or less what we want to happen here in case of creating a cursor by
name.
2014-11-06 16:18:53 -05:00
William Hua
2fdb266c0d mir: set cursor 2014-11-06 16:18:53 -05:00
Alexander Larsson
41bf293efd GdkWindow: Drop gl matrix setup which is not needed anymore. 2014-11-06 20:37:24 +01:00
Alexander Larsson
61eabc4778 gdk_gl_texture_from_surface: Enable scissor test 2014-11-06 20:37:10 +01:00
Alexander Larsson
56631d02bd gdk_gl_texture_quad: Fix handling of window scales 2014-11-06 19:33:58 +01:00
Alexander Larsson
1f3f933edf GdkX11GLContext: Use bitblit for swap if no buffer age support
If buffer age is undefined and the updated area is not the whole
window then we use bit-blits instead of swap-buffers to end the
frame.

This allows us to not repaint the entire window unnecessarily if
buffer_age is not supported, like e.g. with DRI2.
2014-11-06 12:24:43 +01:00
Alexander Larsson
21189b9f7e Add gdk_gl_context_has_framebuffer_blit()
This checks if the context supports GL_EXT_framebuffer_blit
2014-11-06 12:24:43 +01:00
Alexander Larsson
5f9e6ec2dc GdkGL: Fix up GL_ARB_texture_rectangle support
This broke when gdk_gl_texture_quad moved to shaders. We need
a specialized shader for the rectangle case.
2014-11-06 12:24:43 +01:00
Alexander Larsson
650c9f3b7e Add support for GDK_GL=texture-rectangles
This allows us to test our support for GL_ARB_texture_rectangle
2014-11-06 12:24:43 +01:00
Alexander Larsson
90a5fa80af Add GDK_GL env var and GdkGLFlags
This moves the GDK_ALWAYS_USE_GL env var to GDK_GL=always.
It also changes GDK_DEBUG=nogl to GDK_GL=disable, as GDK_DEBUG
is really only about debug loggin.

It also adds some completely new flags:

 software-draw-gl:
   Always use software fallback for drawing gl content to a cairo_t.
   This disables the fastpaths that exist for drawing directly to
   a window and instead reads back the pixels into a cairo image
   surface.

 software-draw-surface:
   Always use software fallback for drawing cairo surfaces onto a
   gl-using window.  This disables e.g. texture-from-pixmap on X11.

 software-draw:
   Enables both the above.
2014-11-06 12:24:43 +01:00
Alexander Larsson
7fde5213b0 gdk_cairo_draw_from_gl: Remove useless call
The glFramebufferTexture2DEXT call makes no sense here, we're
not using the texture as a framebuffer, just as a normal
texture source.
2014-11-06 12:24:43 +01:00
Alexander Larsson
9372cbdbf8 GdkWindow: Try to use a Core 3.2 GL profile for the paint context
If this is supported we avoid a lot of legacy baggage which we don't
need.
2014-11-06 12:24:43 +01:00
Alexander Larsson
e417b18373 gdk_gl_texture_quad: Use shaders to texture things
This is the modern way OpenGL works, and using it will let us
switch to a core context for the paint context, and work on
OpenGL ES 2.0.
2014-11-06 12:24:43 +01:00
Alexander Larsson
c78e7e7f17 GdkGL: Rename "context" to "paint_context" to clarify things a bit 2014-11-06 12:24:43 +01:00
Alexander Larsson
aaa335b9e8 gdk_cairo_draw_from_gl: Document that this changes the current context 2014-11-06 12:24:43 +01:00
Alexander Larsson
8c6395d8a1 GdkGLContextX11: Use passed in context, not get_current() 2014-11-06 12:24:43 +01:00
Alexander Larsson
10a44e8f7a Add gdk_gl_texture_quad() helper
Right now this just centralizes the glBegin/glEnd code, but
this will be replaced with buffer objects later.
2014-11-06 12:24:43 +01:00
Alexander Larsson
3c34ca3405 gdkgl: Don't constantly re-create the tmp framebuffer 2014-11-06 12:24:43 +01:00
Robert Ancell
053b65a21a mir: Handle pointer up/down events without a button mask set (i.e. touch events)
The Mir backend was checking for button mask changes to generate the appropriate
GDK event. When Mir generates a touch event it has no button mask. In this case
we'll just generate a primary button event.
2014-11-06 11:15:52 +13:00
Alexander Larsson
dd721acb84 GdkGLContextX11: Only set swap interval when it changed 2014-11-05 12:02:18 +01:00
Alexander Larsson
e4bf60224b GdkGL: Fix and clean up gdk_cairo_draw_from_gl()
This was unnecessarily creating a framebuffer in the texture case,
and it was not properly setting up a framebuffer with the texture
as source in the software fallback w/ texture source case.
2014-11-05 09:54:05 +01:00
Robert Ancell
c41231cc0f mir: Don't assert on unknown Mir events.
If we get a Mir event we don't understand generate a warning instead of an assert.
2014-11-05 15:50:55 +13:00
Matthias Clasen
113e1d1dc0 Make window scale changes work again
Commit afd9709aff made us keep impl window
cairo surfaces around across changes of window scale. But the
window scale setter forgot to update the size and scale of the
surface. The effect of this was that toggling the window scale
from 1 to 2 in the inspector was not causing the window to draw
at twice the size, although the X window was made twice as big,
and input was scaled too. Fix this by updating the surface when
the window scale changes.
2014-11-03 22:35:11 -05:00
Alexander Larsson
6fbc439fd7 GdkGLContext: Add display property
We need to use this in the code path where we make the context
non-current during destroy, because at that point the window
could be destroyed and gdk_window_get_display() would return
NULL.
2014-11-03 13:20:55 +01:00
Matthias Clasen
1a3bdefb00 Fix an ordering issue in event destruction
gdk_event_get_display accesses event->any.window, so don't
free that window first thing.
2014-11-01 14:44:03 -04:00
Alexander Larsson
869f300f0a gdk_x11_display_make_gl_context_current: move some code
This moves the code related to the frame sync code into
the is_attached check, which means we don't have to ever
run this when making non-window-paint contexts current.

This is a minior speed thing, but the main advantage
is that it makes making a non-paint context current
threadsafe.
2014-10-30 21:04:29 +01:00
Alexander Larsson
0eb2fb01fd Add GdkGLContext::profile 2014-10-30 12:43:03 +01:00
Alexander Larsson
398697eb2f Clarify docs for gdk_cairo_draw_from_gl 2014-10-30 12:43:03 +01:00
Alexander Larsson
f0ae8da993 Remove GdkGLContext::visual
This is not really needed. The gl context is totally tied to the
window it is created from by virtue of sharing the context with the
paint context of that window and that context always has the visual
of the window (which we already can get).

Also, all user visible contexts are essentially offscreen contexts, so
a visual doesn't make sense for them. They only use FBOs which have
whatever format that the users sets up.
2014-10-30 12:43:03 +01:00
Alexander Larsson
5f14630b76 GdkWindow: destroy paint gl context on destroy 2014-10-30 12:43:03 +01:00
Alexander Larsson
dbbab0791d GdkGLContext: Track shared context
This allows us to read it back, but primarily it ensures
the shared context wrapper stays alive as long as the context.
2014-10-30 12:43:03 +01:00
Alexander Larsson
fb50015519 GdkGLContext: Change the way we track the current context
To properly support multithreaded use we use a global GPrivate
to track the current context. Since we also don't need to track
the current context on the display we move gdk_display_destroy_gl_context
to GdkGLContext::discard.
2014-10-30 12:43:03 +01:00
Alexander Larsson
afd9709aff GdkWindowX11: Keep the impl window cairo surface around until destroy
We used to have a weak ref to the cairo surface and it was keep
alive by the references in the normal windows, but that reference
was removed by d48adf9cee, causing
us to constantly create and destroy the surface.

https://bugzilla.gnome.org/show_bug.cgi?id=738648
2014-10-30 10:32:47 +01:00
Carlos Garnacho
b0313c7bcf wayland: Emit GDK_OWNER_CHANGE events around wl_data_offer changes
This way the upper GtkClipboard code emits correctly ::owner-change when
receiving these events.
2014-10-29 18:33:48 +01:00
Alexander Larsson
c7451bbf3b GdkX11GLContext: Fix leak 2014-10-29 13:52:04 +01:00
Alexander Larsson
8c7623de84 Cache default gdk visuals in the GDK_VISUALS property on the root window
This means we don't have to try to initialize opengl in every gtk
instance that is stated. It will only happen for the first one.

https://bugzilla.gnome.org/show_bug.cgi?id=738670
2014-10-29 13:35:17 +01:00
Alexander Larsson
dae447728d X11: Pick better system and rgba visuals for GL
We want to create windows with the default visuals such that we then
have the right visual for GLX when we want to create the paint GL
context for the window.

For instance, (in bug 738670) the default rgba visual we picked for the
NVidia driver had an alpha size of 0 which gave us a BadMatch when later
trying to initialize a gl context on it with a alpha FBConfig.

Instead of just picking what the Xserver likes for the default, and just
picking the first rgba visual we now actually call into GLX to pick
an appropriate visual.
2014-10-29 13:35:17 +01:00
Alexander Larsson
d5cd05a0f4 GdkGLContextX11: Don't free NULL visual info 2014-10-29 11:26:00 +01:00
Matthias Clasen
3f1dc87188 Plug another memory leak
Also found by valgrind.
2014-10-28 16:07:57 -04:00
Matthias Clasen
c8d4eefa71 Plug some memory leaks in GL support
valgrind pointed these out.
2014-10-28 16:07:57 -04:00
Alexander Larsson
0814cd9f8b GdkGlContextWayland: Don't require specific bit depths
This just means we may fail to get a gl context, we just require
*some* color depth.
2014-10-28 18:30:09 +01:00
Alexander Larsson
3406bbd128 Pick the first rgba visual, not the last
The visuals are typically sorted by some sort of "most useful first"
order. And picking the last one is likely to give us the weirdest
matching glx visual.
2014-10-28 18:30:09 +01:00
Alexander Larsson
8d6d6dcca3 GdkGLContextX11: Ensure we get the fbconfig with the exact same visual
We really want a gl context with exactly the same visual, or we will
get a badmatch later, which hits us on nvidia as per:

https://bugzilla.gnome.org/show_bug.cgi?id=738670
2014-10-28 18:30:09 +01:00
Matthias Clasen
b673490a1b Document Wayland opacity handling 2014-10-28 10:48:49 -04:00
Jasper St. Pierre
c40ba85ace wayland: Don't use the default display
Pass a display in so that multi-display technology works.
2014-10-27 22:29:41 -07:00
Jasper St. Pierre
bbe9ef20da gdkevents: Use the correct display to copy event data 2014-10-27 22:29:41 -07:00
Jasper St. Pierre
887b7356c3 gdkdisplay: Provide a default event_data_copy / event_data_free
Every single implementation but Quartz is a no-op for this, so just
provide it once rather than in every backend.
2014-10-27 22:13:23 -07:00
Matthias Clasen
abe7187cfb Add another multi-device invariant
The device and window passed to gdk_device_grab() must be
on the same display, or things are unlikely to work.
Document and assert this.
2014-10-28 00:54:39 -04:00
Matthias Clasen
4c69dabb6e Don't return GDK_GRAB_SUCCESS on failure
That seems ... counterproductive.
2014-10-28 00:51:32 -04:00
Matthias Clasen
d1dcac85f3 Mention the 'same display' requirement in the docs too 2014-10-28 00:44:33 -04:00
Matthias Clasen
eedff39413 Deprecate gdk_cursor_new()
Too easy to introduce an unintended assumption about the display
being the default this way.
2014-10-28 00:38:59 -04:00
Jasper St. Pierre
52e3b88382 gdkwindow: Specify some invariants for cursors
Cursors should not be on a different display than their window /
device, as that would break Wayland.
2014-10-27 21:22:31 -07:00
Jasper St. Pierre
0f79cb3d11 x11: Scale the frame extents as well 2014-10-27 16:54:33 -07:00
Alexander Larsson
f7b1ba0943 opengl: Use GL_TEXTURE_2D rather than GL_TEXTURE_RECTANGLE_ARB when possible
This is more standard, and most driver support non-power-of-2 TEXTURE_2D
these days. We fall back for ancient drivers.
2014-10-27 21:17:08 +01:00
Alexander Larsson
08d44648ea opengl: Use proper blend function for premultiplied alpha 2014-10-27 21:17:08 +01:00
Alexander Larsson
72a6459d73 gdkglcontext: Track whether to use GL_TEXTURE_2D or GL_TEXTURE_RECTANGL_ARB 2014-10-27 21:17:08 +01:00
Alexander Larsson
3013997e23 Rename gdk_gl_context_flush_buffer to gdk_gl_context_end_frame
This makes a lot more sense.
2014-10-27 16:33:37 +01:00
Matthias Clasen
3a2bf9d878 quartz: Move the FullscreenSavedGeometry definition earlier
Move the FullscreenSavedGeometry before the first (conditional)
use, to fix the build on OS X 10.6.

https://bugzilla.gnome.org/show_bug.cgi?id=737561
2014-10-27 06:49:46 -04:00
Marcus Karlsson
a703437796 quartz: fix build on 10.6 and below
It is not possible to successfully build GTK+ on OS X 10.6 and below
since NSFullScreenWindowMask is only available starting with 10.7. Add
ifdef guards around setStyleMask: in order to allow it to build on
earlier OS X releases.

https://bugzilla.gnome.org/show_bug.cgi?id=737561
2014-10-27 06:46:12 -04:00
Matthias Clasen
e75310a445 Fix an out-of-bounds access in MyEnhancedXkbTranslateKeyCode
Commits 314b6abbe8 and eb9223c008 were ignoring
the fact that the code where found is set to 1 was modifying
col - which was an ok thing to do when that part of the code
was still breaking out of the loop, but it is no longer doing
that (since 2003 !). Fix things up by storing the final col
value in a separate variable and using that after the loop.

https://bugzilla.gnome.org/show_bug.cgi?id=738886
2014-10-26 21:48:59 -04:00