The backends are now installed in lib/libweston-0, and the include
files that will be used by libweston in include/libweston-0. The other
modules and weston-specific include files are kept in the old paths.
A new wet_load_module() is added to load plugins in the old path,
which is not part of libweston, but weston only and defined in main.c.
To allow that to be used by out of tree weston plugins, the function
is declared in a new weston.h, installed in include/weston.
The -0 in the paths is the abi version of libweston, and it will also
be used by the libweston .so. If the abi changes the number will need
to be increased.
Signed-off-by: Giulio Camuffo <giuliocamuffo@gmail.com>
Reviewed-by: Quentin Glidic <sardemff7+git@sardemff7.net>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Instead add callbacks to the drm and fbdev backends and pass that to
the input backens so that when a new device needs to be configured
that is called and the compositor can configure it.
Signed-off-by: Giulio Camuffo <giuliocamuffo@gmail.com>
Reviewed-by: Quentin Glidic <sardemff7+git@sardemff7.net>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
This patch splits screensooter.c so that the code implementing
the private screenshooter protocol and launching the client is
moved to a weston specific file, leaving only the code that can
be shared between compositors in screenshooter.c.
Two exported functions are added in screenshooter.c to start and
stop the recorder.
Signed-off-by: Giulio Camuffo <giuliocamuffo@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
This patch completely removes the Raspberry Pi backend and the renderer.
The backend and the renderer were written to use the proprietary
DispmanX API available only on the Raspberry Pi, to demonstrate what the
tiny computer is capable of graphics wise. They were also used to
demonstrate how Wayland and Weston in particular could leverage hardware
compositing capabilities that are not OpenGL. The backend was first
added in e8de35c922, in 2012.
Since then, the major point has been proven. Over time, support for the
rpi-backend diminished, it started to deteriorate and hinder Weston
development. On May 11, I tried to ask if anyone actually cared about
the rpi-backend, but did not get any votes for keeping it:
https://lists.freedesktop.org/archives/wayland-devel/2016-May/028764.html
The rpi-backend is a good example of how using an API that is only
available for specific hardware, even more so as it is only available
with a proprietary driver stack, is not maintainable in the long run.
Most developers working on Weston either just cannot, or cannot bother
to test things also on the RPi. Breakage creeps in without anyone
noticing. If someone actually notices it, fixing it will require a very
specific environment to be able to test. Also the quality of the
proprietary implementation fluctuated. There are reports that RPi
firmware updates randomly broke Weston, and that nowadays it is very
hard to find a RPi firmware version that you could expect to work with
Weston if Weston itself was not broken. We are not even sure what is
broken nowadays.
This removal does not leave Raspberry Pi users cold (for long), though.
There is serious work going on in implementing a FOSS driver stack for
Raspberry Pi, including modern kernel DRM drivers and Mesa drivers. It
might not be fully there yet, but the plan is to be able to use the
standard DRM-backend of Weston on the RPis. See:
http://dri.freedesktop.org/wiki/VC4/
The rpi-backend had its moments. Now, it needs to go. Good riddance!
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
Acked-by: Bryce Harrington <bryce@osg.samsung.com>
Acked-by: Jonas Ådahl <jadahl@gmail.com>
Acked-by: Daniel Stone <daniels@collabora.com>
This adds an alternate weston terminal icon and icons for the flower and
editor clients. The original Inkscape SVG file is included.
Example screenshot:
http://www.bryceharrington.org/Files/weston-icons.png
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Yong Bakos <ybakos@humanoriented.com>
Preparing for libweston and for the separation of the code base into
libweston vs. weston the compositor, we must remove all uses
weston_config structures from the backends. We have decided that all
option and config input happens in the compositor (main.c), and
configuration is passed in for the backends as structs.
Most other backends have already converted, and this patch converts the
DRM-backend to the libweston-style init API.
The libweston-style init API includes a header for each backend (here
compositor-drm.h) defining the configuration interface. The compositor
(main.c) prepares a configuration struct to be passed through libweston
core to the backend during initialization.
A complication with the DRM-backend is that outputs can be hotplugged,
and their configuration needs to be fetched from the compositor
(main.c). For this, the config struct contains a callback member. The
output configuration API is subject to change later, this is just a
temporary API to get libweston forward.
As weston_compositor's user_data was not previously used for anything,
and the output configuration callback needs data, the user_data is set
to the 'config' pointer. This pointer is only used in
drm_configure_output() in main.c.
[Bryce: lots of stuff and rebasing]
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Quentin Glidic <sardemff7+git@sardemff7.net>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Tested-by: Benoit Gschwind <gschwind@gnu-log.net>
[Pekka: write commit message]
[Pekka: squash in "drm: Don't hang onto the backend config object
post-backend_init" from Bryce Harrington]
[Pekka: drop the compositor.h hunk]
[Pekka: do not #include inside extern "C"]
[Pekka: remove incorrect comment about weston_drm_backend_config
ownership.]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Benoit Gschwind <gschwind@gnu-log.net>
[Pekka: moved #include out of extern "C".]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Use a "well" defined structure to configure x11-backend and move configuration
file parsing inside the weston compositor code.
Enforce destruction of all backend config objects after initialization.
Since the backend config struct versioning implies that there we expect
potential future descrepancy between main's definition of the config
object and the backend's, don't allow the backend to hang onto the
config object outside the initialization scope.
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Giulio Camuffo <giuliocamuffo@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Benoit Gschwind <gschwind@gnu-log.net>
Tested-by: Benoit Gschwind <gschwind@gnu-log.net>
Implement a "well" defined API to configure the fbdev backend.
Following and according to discussion about libweston API
The output transform configuration is moved into weston and added to the
fbdev configuration structure.
Signed-off-by: Benoit Gschwind <gschwind@gnu-log.net>
[Pekka: squashed two patches and rebased.]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Implement a "well" defined API to configure the rdp backend.
Following according to discution about libweston API.
Signed-off-by: Benoit Gschwind <gschwind@gnu-log.net>
Reviewed-by: David FORT <rdp.effort@gmail.com>
[Pekka: added missing headers to Makefile.am]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
The fix/check in 34d59859 is incomplete.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
zuctest is another clock_gettime() user that fails to link against librt when
necessary.
Instead of adding another -lrt LDADD entry i've opted for the saner way and
converted the check to a configure test that will set CLOCK_GETTIME_LIBS
appropiately and replaced all instances of -lrt with it.
Built-tested against old and new glibc.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
refactor configuration API of headless-backend
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
v6:
- Define version number in the header
- Don't use leading underscores in header guards
- Add stub config_init_to_defaults()
- Allocate config on stack
- Drop unused display_name parameter
- Add error message when config is invalid
- Install compositor-headless.h and list it in headless-backend sources
v5:
- Update to current trunk
- Fixed typo 'struct weston_wayland_backend_config'
- Dropped unused variables
- Dropped weston_headless_backend_config_create() in favor of
directly zalloc'ing the object
- Dropped weston_headless_backend_load() in favor of the more
generalized load_backend_new().
- Dropped typedef from header
- Restored use of 'backend_init' entry point
- Backend_init() takes a base weston_backend_config object
- Renamed 'param' to 'config' in a few places for consistency
- Renamed 'headless_options' variable to 'options for consistency
- Version the base struct
- Free config on error
- Don't free config during backend_init normal operations
- Adjust header ordering
- Make header guard naming consistent with other headers
- Light reformatting for code style and consistency with other
backend config patches
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Giulio Camuffo <giuliocamuffo@gmail.com>
[Pekka: rebased to apply before drm and x11 backends]
[Pekka: squashed in the headless part of "Enforce destruction of all
backend config objects after initialization"]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
SYSTEMD_DAEMON_LIBS does not belong in CFLAGS, but SYSTEMD_DAEMON_CFLAGS
does. Fix that.
Add the missing COMPOSITOR_CFLAGS. Otherwise compiling the plugin will
use the system wayland-server.h when it should be using the one pointed
to by pkg-config.
The latter fixes the build for me, as my system libwayland-server is
older than what Weston and this plugin require, and the correct version
is only found in my install $prefix.
Cc: Egor Starkov <egor.starkov@ge.com>
Cc: Eugen Friedrich <efriedrich@de.adit-jv.com>
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Direct fail_on_null calls now produce output like:
[weston-info] clients/weston-info.c:714: out of memory
xmalloc, et al produce output on failure like:
[weston-info] out of memory (-1)
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
This new header encapsulates the API that ivi-layout offers to
ivi-shell.c to call.
ivi-shell.c no longer uses ivi-layout-private.h. This limits the
ivi-layout internal structures to just ivi-layout code.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Emre Ucan <eucan@de.adit-jv.com>
Remove the unstable presentation_timing.xml file, and use
presentation-time.xml from wayland-protocols instead to generate all the
Presentation extension bindings.
The following renames are done according to the XML changes:
- generated header includes
- enum constants and macros prefixed with WP_
- interface symbol names prefixed with wp_
- protocol API calls prefixed with wp_
Clients use wp_presentation_interface.name rather than hardcoding the
global interface name: presentation-shm, weston-info, presentation-test.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>
[Pekka: updated wayland-protocols dependency to 1.2]
Fix the protostability function to handle stable protocol files
correctly. Stable protocol XML file names do not have 'stable' in their
name, nor do we want to write that in the prerequisite lists in the
Makefile.
Function 'protoname' does not need fixing, because for stable protocol
prerequisites, the sed pattern will not match, and it passes stem
through as is, which is correct.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Toytoolkit sources don't actually use the presentation_timing client
protocol bindings for anything. Apparently they were there only because
that's how they end up in BUILT_SOURCES.
Move them from toytoolkit sources to BUILT_SOURCES where also other such
things are.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
This client opens a V4L2 device, usually exposed as /dev/videoN, and
retrieves its frames as dmabuf for later import into the compositor.
It supports both single- and multi-planar devices, and any format
exposed by the V4L2 device the Wayland compositor accepts.
This client never changes the v4l2 settings, use `v4l2-ctl -c` if you
want to change those.
Signed-off-by: Emmanuel Gil Peyrot <emmanuel.peyrot@collabora.com>
Acked-by: Derek Foreman <derekf@osg.samsung.com>
Tested-by: Derek Foreman <derekf@osg.samsung.com>
Maniphest Tasks: T90
Differential Revision: https://phabricator.freedesktop.org/D339
This client was using an Intel-specific way to allocate a dmabuf, so it
makes sense to have that in its name.
Signed-off-by: Emmanuel Gil Peyrot <emmanuel.peyrot@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
Differential Revision: https://phabricator.freedesktop.org/D342
It doesn't fill a useful function and is not intended to be continued.
If there is need for workspace manipulation from clients a protocol
based on those future needs need to be properly designed.
workspaces.xml is probably not very relevant since it did the bare
minimum.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mariusz Ceier <mceier+wayland@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Due to the effort of moving a way from non-prefixed protocols, rename
the weston specific screenshooter protocol to weston_screenshooter.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mariusz Ceier <mceier+wayland@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
In the effort of going away from generic names of protocols only
relevant for weston, rename the weston desktop shell
weston_desktop_shell.
This also resets the version to 1, as there will be no prior versions
to weston_desktop_shell.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mariusz Ceier <mceier+wayland@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mariusz Ceier <mceier+wayland@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Jasper St. Pierre <jstpierre@mecheye.net>
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mariusz Ceier <mceier+wayland@gmail.com>
Reviewed-by: Jan Arne Petersen <janarne@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mariusz Ceier <mceier+wayland@gmail.com>
Reviewed-by: Jan Arne Petersen <janarne@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Use the fullscreen-shell protocol XML from the wayland-protocols
installation, and remove the one we provide ourself.
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Mariusz Ceier <mceier+wayland@gmail.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
glibc requires this prior to 2.17, and we already do it in a few other
places.
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Giulio Camuffo <giuliocamuffo@gmail.com>
IVI-Shell is designed to be used with other controller modules
than hmi-controller.These controller modules require
the ivi-layout-export header file to properly integrate
with the ivi-shell. The header file should be installed
when ivi-shell is enabled, because these controller modules
are not a part of the weston repository.
Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Systemd notifications support was converted into loadable
module, so systemd-notify.h header is not needed.
Signed-off-by: Egor Starkov <egor.starkov@ge.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Add systemd status and watchdog notification support.
Feature is not compiled by default and can be enabled by
"--enable-systemd-notify" configuration flag. It compiles
into module "systemd-notify.so" and can be loaded by
adding it in weston.ini like any other module, i.e.
"modules=systemd-notify.so". Watchdog timeout equals to
half of timeout defined by "WATCHDOG_USEC" environment
variable, which is set by "WatchdogSec=" setting in
service file.
Signed-off-by: Egor Starkov <egor.starkov@ge.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
These routines provide test cases an ability to capture screen images
for rendering verification.
This commit is a no-change refactoring, except for making the routines
non-static. Makefile rules are also updated; most notably, this links
test clients against the cairo libraries now.
v2: Fix pointer code styling, suggested in review
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
We were missing launcher-impl.h in Makefile.am
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
We now have a launcher interface and distinct implementations for
logind, weston-launch, and direct DRM, each in their own files.
This helps up clean up the spaghetti code into something that's
hopefully a bit more understood. There should be no functional
changes here.
Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
update: Dropped redundant free() in weston_launcher_destroy()
v2:
- adapted to protocol changes
- added TODO comments
- minor clean-up
- change y-invert from per-plane boolean to per-buffer flag
v3:
- fix a typo: 1 -> i (noticed by Carlos Olmedo Escobar)
Signed-off-by: George Kiagiadakis <george.kiagiadakis@collabora.com>
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Import dmabuf as an EGLImage, and hold on to the EGLImage until we are
signalled a content change. On content change, destroy the EGLImage and
re-import to trigger GPU cache flushes.
We hold on to the EGLImage as long as possible just in case the client
does other imports that might later make re-importing fail.
As dmabuf protocol uses drm_fourcc codes, we need libdrm for
drm_fourcc.h. However, we are not doing any libdrm function calls, so
there is no new need to link to libdrm.
RFCv1 changes:
- fix error if dmabuf exposed unsupported
- always use GL_TEXTURE_EXTERNAL_OES with dmabuf
v2 changes:
- improve support check and error handling
- hold on to the imported EGLImage to avoid the dmabuf becoming
unimportable in the future
- send internal errors with linux_dmabuf_buffer_send_server_error()
- import EGL_EXT_image_dma_buf_import extension headers
- use heuristics to decide between GL_TEXTURE_2D and
GL_TEXTURE_EXTERNAL_OES
- add comment about Mesa requirements
- change y-invert from per-plane boolean to per-buffer flag
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
v2 changes:
- implement the revised protocol
- add basic sanity checks when creating buffer and check for support
- add way to attach user data to the dmabuf for renderer use
- bump max number of planes to 4 to follow DRM AddFb2 ioctl
- improve errors handling
- use separate linux_dmabuf_buffer fields for the different wl_resource
types
- as SERVER_ERROR code is no more, use a wl_display "generic" error for
emergency-disconneting a client we fail to process
- more documentation
- change y-invert from per-plane boolean to per-buffer flag
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
An experimental (hence the 'z' prefix) linux_dmabuf Wayland protocol
extension for creating dmabuf-based wl_buffers in a generic manner.
This does not include proper dmabuf metadata negotiation because
there is no way to communicate all dmabuf constraints from the
compositor to a client before-hand. The client has to create a
wl_buffer wrapping one or more dmabuf buffers and then listen at
the feedback object returned to know if the operation was successful.
RFCv1 changes (after a first draft without code):
- some renames of interfaces and argument, no semantic changes
- added destructor protocol to dmabuf_batch
- added feedback interface for buffer creation
v2 changes:
- use drm_fourcc.h as authoritative source for format codes
- add support for the 64-bit layout qualifier and y-inverted dmabufs
- simplify the 'add' request (no need to preserve fd numerical id)
- add explicit plane index in the 'add' request
- integrate the 'feedback' object events to the batch interface
- rename 'create_buffer' to 'create' and move it into the batch interface
- add requirements needed from the graphics stack and clients
- improve existing errors and add batch error codes
- removed error codes from the global interface
- improve documentation for arguments, enums, etc.
- rename dmabuf_batch to zlinux_buffer_params
- The y-inverted property makes more sense as a whole buffer property.
Y-flipping individual planes of the same buffer object is hardly useful.
The y-invert is also converted into a flag, so we may add more flags
later.
- add flags for interlaced buffer content
v3 changes:
- Apply Daniel Vetter's comments about wording on coherency.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Louis-Francis Ratté-Boulianne <lfrb@collabora.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Acked-by: Benjamin Gaignard <benjamin.gaignard@linaro.org>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
In parallel out-of-tree builds it is possible for e.g. ivi-shell/weston.ini to
be written before ivi-shell/ exists. Solve this by creating the target
directory first.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Reviewed-by: Quentin Glidic <sardemff7+git@sardemff7.net>
Adds basic support for optionally outputting in the XML format
commonly used by JUnit compatible tools.
This format is supported by default by many tools, including
the Jenkins build system. It also is more detailed and
captures more information than the more simplistic TAP
format.
Signed-off-by: Jon A. Cruz <jonc@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
dist_wayland_session_DATA will distribute and install src/weston.desktop, so the
definition of wayland_session_DATA which also installs src/weston.desktop will
result in the file being installed twice and (rarely) cause install to fail.
Spotted and fix by Roy Li <rongqing.li@windriver.com>.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Reviewed-By: Derek Foreman <derekf@osg.samsung.com>
drm_output_start_repaint_loop() incurred a delay of
one refresh cycle by using a no-op page-flip to get
an accurate vblank timestamp as reference. This causes
unwanted lag whenever Weston exited its repaint loop, e.g.,
whenever an application wants to repaint with less than
full video refresh rate but still minimum lag.
Try to use the drmWaitVblank ioctl to get a proper
timestamp instantaneously without lag. If that does
not work, fall back to the old method of idle page-flip.
This optimization will work on any drm/kms driver
which supports high precision vblank timestamping.
As of Linux 4.0 these would be intel, radeon and
nouveau on all their supported gpu's.
On kms drivers without instant high precision timestamping
support, the kernel is supposed to return a timestamp
of zero when calling drmWaitVblank() to query the current
vblank count and time iff vblank irqs are currently
disabled, because the only way to get a valid timestamp
on such kms drivers is to enable vblank interrupts and
then wait a bit for the next vblank irq to take a new valid
timestamp. The caller is supposed to poll until at next
vblank irq it gets a valid non-zero timestamp if it needs
a timestamp.
This zero-timestamp signalling works up to Linux 3.17, but
got broken due to a regression in Linux 3.18 and later. On
Linux 3.18+ with kms drivers that don't have high precision
timestamping, the kernel erroneously returns a stale timestamp
from an earlier vblank, ie. the vblank count and timestamp are
mismatched. A patch is under way to fix this, but to deal with
broken kernels, we also check non-zero timestamps if they are
more than one refresh duration in the past, as this indicates
a stale/invalid timestamp, so we need to take the page-flip
fallback for restarting the repaint loop.
v2: Implement review suggestions by Pekka Paalanen, especially
extend the commit message to describe when and why the
instant restart won't work due to missing Linux kernel
functionality or a Linux kernel regression.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
v3: Fix timespec_to_nsec() which was computing picoseconds,
use the new timespec-util.h helpers.
v4: Rebased to master, split long lines.
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Giulio Camuffo
Pekka Paalanen
Bryce Harrington
Benoit Gschwind
Gustavo Zacarias
Emmanuel Gil Peyrot
Jonas Ådahl
Derek Foreman
Ucan, Emre \(ADITG/SW1\)
Egor Starkov
Jasper St. Pierre
George Kiagiadakis
Ross Burton
Jon A. Cruz
Mario Kleiner
