mainnika
/
weston
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
5961 Commits
1 Branch
0 Tags
13 MiB
Tag: Branch: Tree: 0c30fa5503
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '0c30fa5503'
${ noResults }
weston/libweston/launcher-logind.c

851 lines
19 KiB
Raw Normal View History Unescape

launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
/*
* Copyright © 2013 David Herrmann
*
src: Update boilerplate from MIT X11 license to MIT Expat license
10 years ago
* 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:
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
*
src: Update boilerplate from MIT X11 license to MIT Expat license
10 years ago
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial
* portions of the Software.
*
* 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.
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
*/
#include "config.h"
#include <errno.h>
#include <fcntl.h>
#include <signal.h>
#include <stdarg.h>
#include <stdbool.h>
include stdint.h for int32_t/uint32_t Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com> Reviewed-by: Yong Bakos <ybakos@humanoriented.com>
8 years ago
#include <stdint.h>
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/stat.h>
launcher: Add sysmacros.h include for major() glibc 2.25 produces a warning when sysmacros.h is not directly included but major() is used, as it is intended to be moved to sysmacros.h and only there. Include it to keep the build happy. Signed-off-by: Daniel Stone <daniels@collabora.com> Reviewed-by: Emilio Pozuelo Monfort <emilio.pozuelo@collabora.co.uk>
8 years ago
#include <sys/sysmacros.h>
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
#include <systemd/sd-login.h>
#include <unistd.h>
#include "compositor.h"
#include "dbus.h"
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
#include "launcher-impl.h"
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
logind: delay wakeup until DRM-device is resumed The logind API was designed to allow any kind of devices and any number of devices. It has no idea of "main DRM device" or similar. However, the weston DRM backend was designed with a single DRM device as master. Therefore, we wake it up unconditionally on session-wakeup. But this may fail with logind as a session may be awake, but not all devices have been resumed, yet. Therefore, we change the weston-logind backend to deal with this case correctly. Instead of waking up the compositor on session-wakeup, we wait for the main DRM device to wake up. Once we get the event, we notify the compositor. For sleep, we reverse this logic. On *any* of the following events we tell the compositor to go to sleep: - Session gets inactive - DRM device gets inactive - DRM device is removed This guarantees, that weston is only active if *both*, the session and the main DRM device are awake/active. Note that we could actually rely solely on the DRM-device Pause/Resume events from logind and drop all the Active-Prop-Changed handling. logind guarantees proper ordering of both. However, in case we ever change weston to support multiple GPUs, we need the per-device notification. Thus, keep the code. This also makes weston more fail-safe in case logind fails to send the PauseDevice event (for whatever reason..).
11 years ago
#define DRM_MAJOR 226
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
struct launcher_logind {
struct weston_launcher base;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
struct weston_compositor *compositor;
logind: forward Active=true changes for non-DRM backends Logind sends us a notification whenever the Active attribute of our session changes. However, due to the way compositor-drm.c relies on the master DRM device to be synced with the session, we used to delay Active=true handling until the DRM device was up, too. See: commit aedc7732ebd9bc7b4f51ee247ea857ffec6260a7 Author: David Herrmann <dh.herrmann@gmail.com> Date: Sat Nov 30 11:25:45 2013 +0100 logind: delay wakeup until DRM-device is resumed However, the other compositor backends do not use DRM, so logind-util will never get notified about any DRM device. Therefore, we have to forward the Active=true change immediately. This commit fixes logind-util to take sync_drm as argument. If it is true, we do DRM-device synchronisation, otherwise we don't. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=86889 Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
bool sync_drm;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
char *seat;
char *sid;
unsigned int vtnr;
int vt;
int kb_mode;
DBusConnection *dbus;
struct wl_event_source *dbus_ctx;
char *spath;
DBusPendingCall *pending_active;
};
static int
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_take_device(struct launcher_logind *wl, uint32_t major,
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
uint32_t minor, bool *paused_out)
{
DBusMessage *m, *reply;
logind: Use dbus_bool_t for bool types in dbus calls The gcc built-in 'bool' type is not the same size as dbus_bool_t, which is an uint32_t. Passing a pointer to bool where dbus expects a uint32_t * doesn't work. Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
11 years ago
bool b;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
int r, fd;
logind: Use dbus_bool_t for bool types in dbus calls The gcc built-in 'bool' type is not the same size as dbus_bool_t, which is an uint32_t. Passing a pointer to bool where dbus expects a uint32_t * doesn't work. Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
11 years ago
dbus_bool_t paused;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
m = dbus_message_new_method_call("org.freedesktop.login1",
wl->spath,
"org.freedesktop.login1.Session",
"TakeDevice");
if (!m)
return -ENOMEM;
b = dbus_message_append_args(m,
DBUS_TYPE_UINT32, &major,
DBUS_TYPE_UINT32, &minor,
DBUS_TYPE_INVALID);
if (!b) {
r = -ENOMEM;
goto err_unref;
}
reply = dbus_connection_send_with_reply_and_block(wl->dbus, m,
-1, NULL);
if (!reply) {
r = -ENODEV;
goto err_unref;
}
b = dbus_message_get_args(reply, NULL,
DBUS_TYPE_UNIX_FD, &fd,
DBUS_TYPE_BOOLEAN, &paused,
DBUS_TYPE_INVALID);
if (!b) {
r = -ENODEV;
goto err_reply;
}
r = fd;
if (paused_out)
*paused_out = paused;
err_reply:
dbus_message_unref(reply);
err_unref:
dbus_message_unref(m);
return r;
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_release_device(struct launcher_logind *wl, uint32_t major,
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
uint32_t minor)
{
DBusMessage *m;
bool b;
m = dbus_message_new_method_call("org.freedesktop.login1",
wl->spath,
"org.freedesktop.login1.Session",
"ReleaseDevice");
if (m) {
b = dbus_message_append_args(m,
DBUS_TYPE_UINT32, &major,
DBUS_TYPE_UINT32, &minor,
DBUS_TYPE_INVALID);
if (b)
dbus_connection_send(wl->dbus, m, NULL);
dbus_message_unref(m);
}
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_pause_device_complete(struct launcher_logind *wl, uint32_t major,
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
uint32_t minor)
{
DBusMessage *m;
bool b;
m = dbus_message_new_method_call("org.freedesktop.login1",
wl->spath,
"org.freedesktop.login1.Session",
"PauseDeviceComplete");
if (m) {
b = dbus_message_append_args(m,
DBUS_TYPE_UINT32, &major,
DBUS_TYPE_UINT32, &minor,
DBUS_TYPE_INVALID);
if (b)
dbus_connection_send(wl->dbus, m, NULL);
dbus_message_unref(m);
}
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
static int
launcher_logind_open(struct weston_launcher *launcher, const char *path, int flags)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
struct launcher_logind *wl = wl_container_of(launcher, wl, base);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
struct stat st;
int fl, r, fd;
r = stat(path, &st);
if (r < 0)
logind: change to -1+errno Set errno and return -1 in public API calls like all other weston code does. Most systemd+dbus calls return negative error-codes instead of -1 and setting errno. Thus, we need to explicitly set errno before returning. Also note that we must set errno _after_ the cleanup path. Calling functions like close() in the cleanup path might overwrite errno (which is not what we want). So protect errno until the final return -1;
11 years ago
return -1;
if (!S_ISCHR(st.st_mode)) {
errno = ENODEV;
return -1;
}
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
fd = launcher_logind_take_device(wl, major(st.st_rdev),
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
minor(st.st_rdev), NULL);
if (fd < 0)
return fd;
/* Compared to weston_launcher_open() we cannot specify the open-mode
* directly. Instead, logind passes us an fd with sane default modes.
* For DRM and evdev this means O_RDWR | O_CLOEXEC. If we want
* something else, we need to change it afterwards. We currently
launcher-util: Force all weston_launcher_open()s to use O_CLOEXEC Really, there's pretty much no time we'd ever want O_CLOEXEC unset, as it will likely result in leaking fds to processes that aren't interested in them or shouldn't have them. This also removes the (now unused) code from weston_logind_open() that could drop O_CLOEXEC. Signed-off-by: Derek Foreman <derekf@osg.samsung.com> Reviewed-by: Daniel Stone <daniels@collabora.com> Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
10 years ago
* only support setting O_NONBLOCK. Changing access-modes is not
* possible so accept whatever logind passes us. */
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
fl = fcntl(fd, F_GETFL);
if (fl < 0) {
r = -errno;
goto err_close;
}
if (flags & O_NONBLOCK)
fl |= O_NONBLOCK;
r = fcntl(fd, F_SETFL, fl);
if (r < 0) {
r = -errno;
goto err_close;
}
return fd;
err_close:
close(fd);
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_release_device(wl, major(st.st_rdev),
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
minor(st.st_rdev));
logind: change to -1+errno Set errno and return -1 in public API calls like all other weston code does. Most systemd+dbus calls return negative error-codes instead of -1 and setting errno. Thus, we need to explicitly set errno before returning. Also note that we must set errno _after_ the cleanup path. Calling functions like close() in the cleanup path might overwrite errno (which is not what we want). So protect errno until the final return -1;
11 years ago
errno = -r;
return -1;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
static void
launcher_logind_close(struct weston_launcher *launcher, int fd)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
struct launcher_logind *wl = wl_container_of(launcher, wl, base);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
struct stat st;
int r;
r = fstat(fd, &st);
logind: actually close fd in launcher_logind_close We still need to close fds passed to us - or we leak quite a few fds on VC switch. Regression, originally fixed in 8f5acc2f3a29c3831af4ddd6bed57f703c98dc77 and re-broken in commit 72dea06d7952e3ce8dd8057f7106186da4fa2678 but only for the logind launcher. Signed-off-by: Derek Foreman <derekf@osg.samsung.com> Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
7 years ago
close(fd);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
if (r < 0) {
weston_log("logind: cannot fstat fd: %m\n");
return;
}
if (!S_ISCHR(st.st_mode)) {
weston_log("logind: invalid device passed\n");
return;
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_release_device(wl, major(st.st_rdev),
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
minor(st.st_rdev));
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
static void
launcher_logind_restore(struct weston_launcher *launcher)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
static int
launcher_logind_activate_vt(struct weston_launcher *launcher, int vt)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
struct launcher_logind *wl = wl_container_of(launcher, wl, base);
launcher-logind: Remove old VT switching code, move to SwitchTo/Activate In the time since this code was written, logind has gained new APIs to handle VT switching automatically and activate sessions. Switch to that. Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
9 years ago
DBusMessage *m;
bool b;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
int r;
launcher-logind: Remove old VT switching code, move to SwitchTo/Activate In the time since this code was written, logind has gained new APIs to handle VT switching automatically and activate sessions. Switch to that. Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
9 years ago
m = dbus_message_new_method_call("org.freedesktop.login1",
"/org/freedesktop/login1/seat/self",
"org.freedesktop.login1.Seat",
"SwitchTo");
if (!m)
return -ENOMEM;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
launcher-logind: Remove old VT switching code, move to SwitchTo/Activate In the time since this code was written, logind has gained new APIs to handle VT switching automatically and activate sessions. Switch to that. Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
9 years ago
b = dbus_message_append_args(m,
DBUS_TYPE_UINT32, &vt,
DBUS_TYPE_INVALID);
if (!b) {
r = -ENOMEM;
goto err_unref;
}
dbus_connection_send(wl->dbus, m, NULL);
r = 0;
err_unref:
dbus_message_unref(m);
return r;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_set_active(struct launcher_logind *wl, bool active)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
logind: delay wakeup until DRM-device is resumed The logind API was designed to allow any kind of devices and any number of devices. It has no idea of "main DRM device" or similar. However, the weston DRM backend was designed with a single DRM device as master. Therefore, we wake it up unconditionally on session-wakeup. But this may fail with logind as a session may be awake, but not all devices have been resumed, yet. Therefore, we change the weston-logind backend to deal with this case correctly. Instead of waking up the compositor on session-wakeup, we wait for the main DRM device to wake up. Once we get the event, we notify the compositor. For sleep, we reverse this logic. On *any* of the following events we tell the compositor to go to sleep: - Session gets inactive - DRM device gets inactive - DRM device is removed This guarantees, that weston is only active if *both*, the session and the main DRM device are awake/active. Note that we could actually rely solely on the DRM-device Pause/Resume events from logind and drop all the Active-Prop-Changed handling. logind guarantees proper ordering of both. However, in case we ever change weston to support multiple GPUs, we need the per-device notification. Thus, keep the code. This also makes weston more fail-safe in case logind fails to send the PauseDevice event (for whatever reason..).
11 years ago
if (!wl->compositor->session_active == !active)
return;
wl->compositor->session_active = active;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
wl_signal_emit(&wl->compositor->session_signal,
wl->compositor);
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
parse_active(struct launcher_logind *wl, DBusMessage *m, DBusMessageIter *iter)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
logind: fix PropertiesChanged parser The current parser directly reads a BOOLEAN on the PropertiesChanged signal for 'Active' properties. However, all property-values are packed in a VARIANT, otherwise, we wouldn't know the type. Fix the parser to recurse into the variant before reading the boolean. To avoid such bugs in the future, we extract the 'Active' parser into a helper function parse_active(), which is then shared between the PropertiesChanged and Get handlers. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
DBusMessageIter sub;
logind: Use dbus_bool_t for bool types in dbus calls The gcc built-in 'bool' type is not the same size as dbus_bool_t, which is an uint32_t. Passing a pointer to bool where dbus expects a uint32_t * doesn't work. Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
11 years ago
dbus_bool_t b;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
logind: fix PropertiesChanged parser The current parser directly reads a BOOLEAN on the PropertiesChanged signal for 'Active' properties. However, all property-values are packed in a VARIANT, otherwise, we wouldn't know the type. Fix the parser to recurse into the variant before reading the boolean. To avoid such bugs in the future, we extract the 'Active' parser into a helper function parse_active(), which is then shared between the PropertiesChanged and Get handlers. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
if (dbus_message_iter_get_arg_type(iter) != DBUS_TYPE_VARIANT)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
return;
logind: fix PropertiesChanged parser The current parser directly reads a BOOLEAN on the PropertiesChanged signal for 'Active' properties. However, all property-values are packed in a VARIANT, otherwise, we wouldn't know the type. Fix the parser to recurse into the variant before reading the boolean. To avoid such bugs in the future, we extract the 'Active' parser into a helper function parse_active(), which is then shared between the PropertiesChanged and Get handlers. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
dbus_message_iter_recurse(iter, &sub);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
if (dbus_message_iter_get_arg_type(&sub) != DBUS_TYPE_BOOLEAN)
logind: fix PropertiesChanged parser The current parser directly reads a BOOLEAN on the PropertiesChanged signal for 'Active' properties. However, all property-values are packed in a VARIANT, otherwise, we wouldn't know the type. Fix the parser to recurse into the variant before reading the boolean. To avoid such bugs in the future, we extract the 'Active' parser into a helper function parse_active(), which is then shared between the PropertiesChanged and Get handlers. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
return;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
dbus_message_iter_get_basic(&sub, &b);
logind: forward Active=true changes for non-DRM backends Logind sends us a notification whenever the Active attribute of our session changes. However, due to the way compositor-drm.c relies on the master DRM device to be synced with the session, we used to delay Active=true handling until the DRM device was up, too. See: commit aedc7732ebd9bc7b4f51ee247ea857ffec6260a7 Author: David Herrmann <dh.herrmann@gmail.com> Date: Sat Nov 30 11:25:45 2013 +0100 logind: delay wakeup until DRM-device is resumed However, the other compositor backends do not use DRM, so logind-util will never get notified about any DRM device. Therefore, we have to forward the Active=true change immediately. This commit fixes logind-util to take sync_drm as argument. If it is true, we do DRM-device synchronisation, otherwise we don't. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=86889 Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
/* If the backend requested DRM master-device synchronization, we only
* wake-up the compositor once the master-device is up and running. For
* other backends, we immediately forward the Active-change event. */
if (!wl->sync_drm || !b)
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_set_active(wl, b);
logind: fix PropertiesChanged parser The current parser directly reads a BOOLEAN on the PropertiesChanged signal for 'Active' properties. However, all property-values are packed in a VARIANT, otherwise, we wouldn't know the type. Fix the parser to recurse into the variant before reading the boolean. To avoid such bugs in the future, we extract the 'Active' parser into a helper function parse_active(), which is then shared between the PropertiesChanged and Get handlers. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
}
static void
get_active_cb(DBusPendingCall *pending, void *data)
{
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
struct launcher_logind *wl = data;
logind: fix PropertiesChanged parser The current parser directly reads a BOOLEAN on the PropertiesChanged signal for 'Active' properties. However, all property-values are packed in a VARIANT, otherwise, we wouldn't know the type. Fix the parser to recurse into the variant before reading the boolean. To avoid such bugs in the future, we extract the 'Active' parser into a helper function parse_active(), which is then shared between the PropertiesChanged and Get handlers. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
DBusMessageIter iter;
DBusMessage *m;
int type;
dbus_pending_call_unref(wl->pending_active);
wl->pending_active = NULL;
m = dbus_pending_call_steal_reply(pending);
if (!m)
return;
type = dbus_message_get_type(m);
if (type == DBUS_MESSAGE_TYPE_METHOD_RETURN &&
dbus_message_iter_init(m, &iter))
parse_active(wl, m, &iter);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
dbus_message_unref(m);
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_get_active(struct launcher_logind *wl)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
DBusPendingCall *pending;
DBusMessage *m;
bool b;
const char *iface, *name;
m = dbus_message_new_method_call("org.freedesktop.login1",
wl->spath,
"org.freedesktop.DBus.Properties",
"Get");
if (!m)
return;
iface = "org.freedesktop.login1.Session";
name = "Active";
b = dbus_message_append_args(m,
DBUS_TYPE_STRING, &iface,
DBUS_TYPE_STRING, &name,
DBUS_TYPE_INVALID);
if (!b)
goto err_unref;
b = dbus_connection_send_with_reply(wl->dbus, m, &pending, -1);
if (!b)
goto err_unref;
b = dbus_pending_call_set_notify(pending, get_active_cb, wl, NULL);
if (!b) {
dbus_pending_call_cancel(pending);
dbus_pending_call_unref(pending);
goto err_unref;
}
if (wl->pending_active) {
dbus_pending_call_cancel(wl->pending_active);
dbus_pending_call_unref(wl->pending_active);
}
wl->pending_active = pending;
return;
err_unref:
dbus_message_unref(m);
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
disconnected_dbus(struct launcher_logind *wl)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
weston_log("logind: dbus connection lost, exiting..\n");
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_restore(&wl->base);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
exit(-1);
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
session_removed(struct launcher_logind *wl, DBusMessage *m)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
const char *name, *obj;
bool r;
r = dbus_message_get_args(m, NULL,
DBUS_TYPE_STRING, &name,
DBUS_TYPE_OBJECT_PATH, &obj,
DBUS_TYPE_INVALID);
if (!r) {
weston_log("logind: cannot parse SessionRemoved dbus signal\n");
return;
}
if (!strcmp(name, wl->sid)) {
weston_log("logind: our session got closed, exiting..\n");
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_restore(&wl->base);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
exit(-1);
}
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
property_changed(struct launcher_logind *wl, DBusMessage *m)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
DBusMessageIter iter, sub, entry;
const char *interface, *name;
if (!dbus_message_iter_init(m, &iter) ||
dbus_message_iter_get_arg_type(&iter) != DBUS_TYPE_STRING)
goto error;
dbus_message_iter_get_basic(&iter, &interface);
if (!dbus_message_iter_next(&iter) ||
dbus_message_iter_get_arg_type(&iter) != DBUS_TYPE_ARRAY)
goto error;
dbus_message_iter_recurse(&iter, &sub);
while (dbus_message_iter_get_arg_type(&sub) == DBUS_TYPE_DICT_ENTRY) {
dbus_message_iter_recurse(&sub, &entry);
if (dbus_message_iter_get_arg_type(&entry) != DBUS_TYPE_STRING)
goto error;
dbus_message_iter_get_basic(&entry, &name);
if (!dbus_message_iter_next(&entry))
goto error;
if (!strcmp(name, "Active")) {
logind: fix PropertiesChanged parser The current parser directly reads a BOOLEAN on the PropertiesChanged signal for 'Active' properties. However, all property-values are packed in a VARIANT, otherwise, we wouldn't know the type. Fix the parser to recurse into the variant before reading the boolean. To avoid such bugs in the future, we extract the 'Active' parser into a helper function parse_active(), which is then shared between the PropertiesChanged and Get handlers. Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
parse_active(wl, m, &entry);
return;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
}
dbus_message_iter_next(&sub);
}
if (!dbus_message_iter_next(&iter) ||
dbus_message_iter_get_arg_type(&iter) != DBUS_TYPE_ARRAY)
goto error;
dbus_message_iter_recurse(&iter, &sub);
while (dbus_message_iter_get_arg_type(&sub) == DBUS_TYPE_STRING) {
dbus_message_iter_get_basic(&sub, &name);
if (!strcmp(name, "Active")) {
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_get_active(wl);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
return;
}
dbus_message_iter_next(&sub);
}
return;
error:
weston_log("logind: cannot parse PropertiesChanged dbus signal\n");
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
device_paused(struct launcher_logind *wl, DBusMessage *m)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
bool r;
const char *type;
uint32_t major, minor;
r = dbus_message_get_args(m, NULL,
DBUS_TYPE_UINT32, &major,
DBUS_TYPE_UINT32, &minor,
DBUS_TYPE_STRING, &type,
DBUS_TYPE_INVALID);
if (!r) {
weston_log("logind: cannot parse PauseDevice dbus signal\n");
return;
}
/* "pause" means synchronous pausing. Acknowledge it unconditionally
* as we support asynchronous device shutdowns, anyway.
logind: delay wakeup until DRM-device is resumed The logind API was designed to allow any kind of devices and any number of devices. It has no idea of "main DRM device" or similar. However, the weston DRM backend was designed with a single DRM device as master. Therefore, we wake it up unconditionally on session-wakeup. But this may fail with logind as a session may be awake, but not all devices have been resumed, yet. Therefore, we change the weston-logind backend to deal with this case correctly. Instead of waking up the compositor on session-wakeup, we wait for the main DRM device to wake up. Once we get the event, we notify the compositor. For sleep, we reverse this logic. On *any* of the following events we tell the compositor to go to sleep: - Session gets inactive - DRM device gets inactive - DRM device is removed This guarantees, that weston is only active if *both*, the session and the main DRM device are awake/active. Note that we could actually rely solely on the DRM-device Pause/Resume events from logind and drop all the Active-Prop-Changed handling. logind guarantees proper ordering of both. However, in case we ever change weston to support multiple GPUs, we need the per-device notification. Thus, keep the code. This also makes weston more fail-safe in case logind fails to send the PauseDevice event (for whatever reason..).
11 years ago
* "force" means asynchronous pausing.
* "gone" means the device is gone. We handle it the same as "force" as
* a following udev event will be caught, too.
*
* If it's our main DRM device, tell the compositor to go asleep. */
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
if (!strcmp(type, "pause"))
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_pause_device_complete(wl, major, minor);
logind: delay wakeup until DRM-device is resumed The logind API was designed to allow any kind of devices and any number of devices. It has no idea of "main DRM device" or similar. However, the weston DRM backend was designed with a single DRM device as master. Therefore, we wake it up unconditionally on session-wakeup. But this may fail with logind as a session may be awake, but not all devices have been resumed, yet. Therefore, we change the weston-logind backend to deal with this case correctly. Instead of waking up the compositor on session-wakeup, we wait for the main DRM device to wake up. Once we get the event, we notify the compositor. For sleep, we reverse this logic. On *any* of the following events we tell the compositor to go to sleep: - Session gets inactive - DRM device gets inactive - DRM device is removed This guarantees, that weston is only active if *both*, the session and the main DRM device are awake/active. Note that we could actually rely solely on the DRM-device Pause/Resume events from logind and drop all the Active-Prop-Changed handling. logind guarantees proper ordering of both. However, in case we ever change weston to support multiple GPUs, we need the per-device notification. Thus, keep the code. This also makes weston more fail-safe in case logind fails to send the PauseDevice event (for whatever reason..).
11 years ago
logind: forward Active=true changes for non-DRM backends Logind sends us a notification whenever the Active attribute of our session changes. However, due to the way compositor-drm.c relies on the master DRM device to be synced with the session, we used to delay Active=true handling until the DRM device was up, too. See: commit aedc7732ebd9bc7b4f51ee247ea857ffec6260a7 Author: David Herrmann <dh.herrmann@gmail.com> Date: Sat Nov 30 11:25:45 2013 +0100 logind: delay wakeup until DRM-device is resumed However, the other compositor backends do not use DRM, so logind-util will never get notified about any DRM device. Therefore, we have to forward the Active=true change immediately. This commit fixes logind-util to take sync_drm as argument. If it is true, we do DRM-device synchronisation, otherwise we don't. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=86889 Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
if (wl->sync_drm && major == DRM_MAJOR)
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_set_active(wl, false);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
device_resumed(struct launcher_logind *wl, DBusMessage *m)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
logind: delay wakeup until DRM-device is resumed The logind API was designed to allow any kind of devices and any number of devices. It has no idea of "main DRM device" or similar. However, the weston DRM backend was designed with a single DRM device as master. Therefore, we wake it up unconditionally on session-wakeup. But this may fail with logind as a session may be awake, but not all devices have been resumed, yet. Therefore, we change the weston-logind backend to deal with this case correctly. Instead of waking up the compositor on session-wakeup, we wait for the main DRM device to wake up. Once we get the event, we notify the compositor. For sleep, we reverse this logic. On *any* of the following events we tell the compositor to go to sleep: - Session gets inactive - DRM device gets inactive - DRM device is removed This guarantees, that weston is only active if *both*, the session and the main DRM device are awake/active. Note that we could actually rely solely on the DRM-device Pause/Resume events from logind and drop all the Active-Prop-Changed handling. logind guarantees proper ordering of both. However, in case we ever change weston to support multiple GPUs, we need the per-device notification. Thus, keep the code. This also makes weston more fail-safe in case logind fails to send the PauseDevice event (for whatever reason..).
11 years ago
bool r;
uint32_t major;
r = dbus_message_get_args(m, NULL,
DBUS_TYPE_UINT32, &major,
/*DBUS_TYPE_UINT32, &minor,
DBUS_TYPE_UNIX_FD, &fd,*/
DBUS_TYPE_INVALID);
if (!r) {
weston_log("logind: cannot parse ResumeDevice dbus signal\n");
return;
}
/* DeviceResumed messages provide us a new file-descriptor for
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
* resumed devices. For DRM devices it's the same as before, for evdev
* devices it's a new open-file. As we reopen evdev devices, anyway,
logind: delay wakeup until DRM-device is resumed The logind API was designed to allow any kind of devices and any number of devices. It has no idea of "main DRM device" or similar. However, the weston DRM backend was designed with a single DRM device as master. Therefore, we wake it up unconditionally on session-wakeup. But this may fail with logind as a session may be awake, but not all devices have been resumed, yet. Therefore, we change the weston-logind backend to deal with this case correctly. Instead of waking up the compositor on session-wakeup, we wait for the main DRM device to wake up. Once we get the event, we notify the compositor. For sleep, we reverse this logic. On *any* of the following events we tell the compositor to go to sleep: - Session gets inactive - DRM device gets inactive - DRM device is removed This guarantees, that weston is only active if *both*, the session and the main DRM device are awake/active. Note that we could actually rely solely on the DRM-device Pause/Resume events from logind and drop all the Active-Prop-Changed handling. logind guarantees proper ordering of both. However, in case we ever change weston to support multiple GPUs, we need the per-device notification. Thus, keep the code. This also makes weston more fail-safe in case logind fails to send the PauseDevice event (for whatever reason..).
11 years ago
* there is no need for us to handle this event for evdev. For DRM, we
* notify the compositor to wake up. */
logind: forward Active=true changes for non-DRM backends Logind sends us a notification whenever the Active attribute of our session changes. However, due to the way compositor-drm.c relies on the master DRM device to be synced with the session, we used to delay Active=true handling until the DRM device was up, too. See: commit aedc7732ebd9bc7b4f51ee247ea857ffec6260a7 Author: David Herrmann <dh.herrmann@gmail.com> Date: Sat Nov 30 11:25:45 2013 +0100 logind: delay wakeup until DRM-device is resumed However, the other compositor backends do not use DRM, so logind-util will never get notified about any DRM device. Therefore, we have to forward the Active=true change immediately. This commit fixes logind-util to take sync_drm as argument. If it is true, we do DRM-device synchronisation, otherwise we don't. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=86889 Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
if (wl->sync_drm && major == DRM_MAJOR)
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_set_active(wl, true);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
}
static DBusHandlerResult
filter_dbus(DBusConnection *c, DBusMessage *m, void *data)
{
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
struct launcher_logind *wl = data;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
if (dbus_message_is_signal(m, DBUS_INTERFACE_LOCAL, "Disconnected")) {
disconnected_dbus(wl);
} else if (dbus_message_is_signal(m, "org.freedesktop.login1.Manager",
"SessionRemoved")) {
session_removed(wl, m);
} else if (dbus_message_is_signal(m, "org.freedesktop.DBus.Properties",
"PropertiesChanged")) {
property_changed(wl, m);
} else if (dbus_message_is_signal(m, "org.freedesktop.login1.Session",
"PauseDevice")) {
device_paused(wl, m);
} else if (dbus_message_is_signal(m, "org.freedesktop.login1.Session",
"ResumeDevice")) {
device_resumed(wl, m);
}
return DBUS_HANDLER_RESULT_NOT_YET_HANDLED;
}
static int
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_setup_dbus(struct launcher_logind *wl)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
bool b;
int r;
r = asprintf(&wl->spath, "/org/freedesktop/login1/session/%s",
wl->sid);
if (r < 0)
return -ENOMEM;
b = dbus_connection_add_filter(wl->dbus, filter_dbus, wl, NULL);
if (!b) {
weston_log("logind: cannot add dbus filter\n");
r = -ENOMEM;
goto err_spath;
}
r = weston_dbus_add_match_signal(wl->dbus,
"org.freedesktop.login1",
"org.freedesktop.login1.Manager",
"SessionRemoved",
"/org/freedesktop/login1");
if (r < 0) {
weston_log("logind: cannot add dbus match\n");
goto err_spath;
}
r = weston_dbus_add_match_signal(wl->dbus,
"org.freedesktop.login1",
"org.freedesktop.login1.Session",
"PauseDevice",
wl->spath);
if (r < 0) {
weston_log("logind: cannot add dbus match\n");
goto err_spath;
}
r = weston_dbus_add_match_signal(wl->dbus,
"org.freedesktop.login1",
"org.freedesktop.login1.Session",
"ResumeDevice",
wl->spath);
if (r < 0) {
weston_log("logind: cannot add dbus match\n");
goto err_spath;
}
r = weston_dbus_add_match_signal(wl->dbus,
"org.freedesktop.login1",
"org.freedesktop.DBus.Properties",
"PropertiesChanged",
wl->spath);
if (r < 0) {
weston_log("logind: cannot add dbus match\n");
goto err_spath;
}
return 0;
err_spath:
/* don't remove any dbus-match as the connection is closed, anyway */
free(wl->spath);
return r;
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_destroy_dbus(struct launcher_logind *wl)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
/* don't remove any dbus-match as the connection is closed, anyway */
free(wl->spath);
}
static int
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_take_control(struct launcher_logind *wl)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
DBusError err;
DBusMessage *m, *reply;
dbus_bool_t force;
bool b;
int r;
dbus_error_init(&err);
m = dbus_message_new_method_call("org.freedesktop.login1",
wl->spath,
"org.freedesktop.login1.Session",
"TakeControl");
if (!m)
return -ENOMEM;
force = false;
b = dbus_message_append_args(m,
DBUS_TYPE_BOOLEAN, &force,
DBUS_TYPE_INVALID);
if (!b) {
r = -ENOMEM;
goto err_unref;
}
reply = dbus_connection_send_with_reply_and_block(wl->dbus,
m, -1, &err);
if (!reply) {
if (dbus_error_has_name(&err, DBUS_ERROR_UNKNOWN_METHOD))
weston_log("logind: old systemd version detected\n");
else
weston_log("logind: cannot take control over session %s\n", wl->sid);
dbus_error_free(&err);
r = -EIO;
goto err_unref;
}
dbus_message_unref(reply);
dbus_message_unref(m);
return 0;
err_unref:
dbus_message_unref(m);
return r;
}
static void
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_release_control(struct launcher_logind *wl)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
DBusMessage *m;
m = dbus_message_new_method_call("org.freedesktop.login1",
wl->spath,
"org.freedesktop.login1.Session",
"ReleaseControl");
if (m) {
dbus_connection_send(wl->dbus, m, NULL);
dbus_message_unref(m);
}
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
static int
weston_sd_session_get_vt(const char *sid, unsigned int *out)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
#ifdef HAVE_SYSTEMD_LOGIN_209
return sd_session_get_vt(sid, out);
#else
int r;
char *tty;
r = sd_session_get_tty(sid, &tty);
if (r < 0)
return r;
r = sscanf(tty, "tty%u", out);
free(tty);
if (r != 1)
return -EINVAL;
return 0;
#endif
}
launcher-logind: Remove old VT switching code, move to SwitchTo/Activate In the time since this code was written, logind has gained new APIs to handle VT switching automatically and activate sessions. Switch to that. Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
9 years ago
static int
launcher_logind_activate(struct launcher_logind *wl)
{
DBusMessage *m;
m = dbus_message_new_method_call("org.freedesktop.login1",
wl->spath,
"org.freedesktop.login1.Session",
"Activate");
if (!m)
return -ENOMEM;
dbus_connection_send(wl->dbus, m, NULL);
return 0;
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
static int
launcher_logind_connect(struct weston_launcher **out, struct weston_compositor *compositor,
int tty, const char *seat_id, bool sync_drm)
{
struct launcher_logind *wl;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
struct wl_event_loop *loop;
char *t;
int r;
Use zalloc instead of calloc(1, ...) Signed-off-by: Bryce Harrington <bryce@osg.samsung.com> Reviewed-by: Marek Chalupa <mchqwerty@gmail.com>
10 years ago
wl = zalloc(sizeof(*wl));
if (wl == NULL) {
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
r = -ENOMEM;
goto err_out;
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
wl->base.iface = &launcher_logind_iface;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
wl->compositor = compositor;
logind: forward Active=true changes for non-DRM backends Logind sends us a notification whenever the Active attribute of our session changes. However, due to the way compositor-drm.c relies on the master DRM device to be synced with the session, we used to delay Active=true handling until the DRM device was up, too. See: commit aedc7732ebd9bc7b4f51ee247ea857ffec6260a7 Author: David Herrmann <dh.herrmann@gmail.com> Date: Sat Nov 30 11:25:45 2013 +0100 logind: delay wakeup until DRM-device is resumed However, the other compositor backends do not use DRM, so logind-util will never get notified about any DRM device. Therefore, we have to forward the Active=true change immediately. This commit fixes logind-util to take sync_drm as argument. If it is true, we do DRM-device synchronisation, otherwise we don't. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=86889 Signed-off-by: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Bryce Harrington <bryce@osg.samsung.com> Tested-by: nerdopolis <bluescreen_avenger@verizon.net>
10 years ago
wl->sync_drm = sync_drm;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
wl->seat = strdup(seat_id);
if (!wl->seat) {
r = -ENOMEM;
goto err_wl;
}
r = sd_pid_get_session(getpid(), &wl->sid);
if (r < 0) {
weston_log("logind: not running in a systemd session\n");
goto err_seat;
}
t = NULL;
r = sd_session_get_seat(wl->sid, &t);
logind: Return proper error value when tty or seat doesn't match If the seat or tty doesn't match we return with r == 0, which looks like success to weston_launcher_connect(), which then fails to fall back to the legacy path. https://bugs.freedesktop.org/show_bug.cgi?id=70876
11 years ago
if (r < 0) {
weston_log("logind: failed to get session seat\n");
free(t);
goto err_session;
} else if (strcmp(seat_id, t)) {
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
weston_log("logind: weston's seat '%s' differs from session-seat '%s'\n",
seat_id, t);
logind: Return proper error value when tty or seat doesn't match If the seat or tty doesn't match we return with r == 0, which looks like success to weston_launcher_connect(), which then fails to fall back to the legacy path. https://bugs.freedesktop.org/show_bug.cgi?id=70876
11 years ago
r = -EINVAL;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
free(t);
goto err_session;
}
free(t);
r = weston_sd_session_get_vt(wl->sid, &wl->vtnr);
if (r < 0) {
weston_log("logind: session not running on a VT\n");
goto err_session;
} else if (tty > 0 && wl->vtnr != (unsigned int )tty) {
weston_log("logind: requested VT --tty=%d differs from real session VT %u\n",
tty, wl->vtnr);
logind: Return proper error value when tty or seat doesn't match If the seat or tty doesn't match we return with r == 0, which looks like success to weston_launcher_connect(), which then fails to fall back to the legacy path. https://bugs.freedesktop.org/show_bug.cgi?id=70876
11 years ago
r = -EINVAL;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
goto err_session;
}
loop = wl_display_get_event_loop(compositor->wl_display);
r = weston_dbus_open(loop, DBUS_BUS_SYSTEM, &wl->dbus, &wl->dbus_ctx);
if (r < 0) {
weston_log("logind: cannot connect to system dbus\n");
goto err_session;
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
r = launcher_logind_setup_dbus(wl);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
if (r < 0)
goto err_dbus;
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
r = launcher_logind_take_control(wl);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
if (r < 0)
goto err_dbus_cleanup;
launcher-logind: Remove old VT switching code, move to SwitchTo/Activate In the time since this code was written, logind has gained new APIs to handle VT switching automatically and activate sessions. Switch to that. Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
9 years ago
r = launcher_logind_activate(wl);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
if (r < 0)
launcher-logind: Remove old VT switching code, move to SwitchTo/Activate In the time since this code was written, logind has gained new APIs to handle VT switching automatically and activate sessions. Switch to that. Reviewed-by: Derek Foreman <derekf@osg.samsung.com>
9 years ago
goto err_dbus_cleanup;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
weston_log("logind: session control granted\n");
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
* (struct launcher_logind **) out = wl;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
return 0;
err_dbus_cleanup:
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_destroy_dbus(wl);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
err_dbus:
weston_dbus_close(wl->dbus, wl->dbus_ctx);
err_session:
free(wl->sid);
err_seat:
free(wl->seat);
err_wl:
free(wl);
err_out:
weston_log("logind: cannot setup systemd-logind helper (%d), using legacy fallback\n", r);
logind: change to -1+errno Set errno and return -1 in public API calls like all other weston code does. Most systemd+dbus calls return negative error-codes instead of -1 and setting errno. Thus, we need to explicitly set errno before returning. Also note that we must set errno _after_ the cleanup path. Calling functions like close() in the cleanup path might overwrite errno (which is not what we want). So protect errno until the final return -1;
11 years ago
errno = -r;
return -1;
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
static void
launcher_logind_destroy(struct weston_launcher *launcher)
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
{
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
struct launcher_logind *wl = wl_container_of(launcher, wl, base);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
if (wl->pending_active) {
dbus_pending_call_cancel(wl->pending_active);
dbus_pending_call_unref(wl->pending_active);
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher_logind_release_control(wl);
launcher_logind_destroy_dbus(wl);
launcher: add logind backend Instead of connecting to weston-launch from launcher-util, we now try to connect to logind first. If logind provides session-devices, we use them. If not, we fall back to the old weston-launch facility.
11 years ago
weston_dbus_close(wl->dbus, wl->dbus_ctx);
free(wl->sid);
free(wl->seat);
free(wl);
}
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
launcher: don't try to switch to weston's vt Signed-off-by: Giulio Camuffo <giuliocamuffo@gmail.com> Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
8 years ago
static int
launcher_logind_get_vt(struct weston_launcher *launcher)
{
struct launcher_logind *wl = wl_container_of(launcher, wl, base);
return wl->vtnr;
}
libweston/launcher: annotate iface(s) as constant data Already considered and handled as such. Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> Reviewed-by: Daniel Stone <daniels@collabora.com>
8 years ago
const struct launcher_interface launcher_logind_iface = {
libweston/launcher: use C99 initializers for the iface(s) Makes the code easier to read and browse through. Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> Reviewed-by: Daniel Stone <daniels@collabora.com>
8 years ago
.connect = launcher_logind_connect,
.destroy = launcher_logind_destroy,
.open = launcher_logind_open,
.close = launcher_logind_close,
.activate_vt = launcher_logind_activate_vt,
.restore = launcher_logind_restore,
.get_vt = launcher_logind_get_vt,
launcher: Split out launcher implementations into three distinct ones 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()
9 years ago
};