linux/drivers/media/video/pvrusb2/pvrusb2-context.c

347 lines
8.2 KiB
C
Raw Normal View History

/*
* $Id$
*
* Copyright (C) 2005 Mike Isely <isely@pobox.com>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
*/
#include "pvrusb2-context.h"
#include "pvrusb2-io.h"
#include "pvrusb2-ioread.h"
#include "pvrusb2-hdw.h"
#include "pvrusb2-debug.h"
#include <linux/wait.h>
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
#include <linux/kthread.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/slab.h>
static struct pvr2_context *pvr2_context_exist_first;
static struct pvr2_context *pvr2_context_exist_last;
static struct pvr2_context *pvr2_context_notify_first;
static struct pvr2_context *pvr2_context_notify_last;
static DEFINE_MUTEX(pvr2_context_mutex);
static DECLARE_WAIT_QUEUE_HEAD(pvr2_context_sync_data);
static struct task_struct *pvr2_context_thread_ptr;
static void pvr2_context_set_notify(struct pvr2_context *mp, int fl)
{
int signal_flag = 0;
mutex_lock(&pvr2_context_mutex);
if (fl) {
if (!mp->notify_flag) {
signal_flag = (pvr2_context_notify_first == NULL);
mp->notify_prev = pvr2_context_notify_last;
mp->notify_next = NULL;
pvr2_context_notify_last = mp;
if (mp->notify_prev) {
mp->notify_prev->notify_next = mp;
} else {
pvr2_context_notify_first = mp;
}
mp->notify_flag = !0;
}
} else {
if (mp->notify_flag) {
mp->notify_flag = 0;
if (mp->notify_next) {
mp->notify_next->notify_prev = mp->notify_prev;
} else {
pvr2_context_notify_last = mp->notify_prev;
}
if (mp->notify_prev) {
mp->notify_prev->notify_next = mp->notify_next;
} else {
pvr2_context_notify_first = mp->notify_next;
}
}
}
mutex_unlock(&pvr2_context_mutex);
if (signal_flag) wake_up(&pvr2_context_sync_data);
}
static void pvr2_context_destroy(struct pvr2_context *mp)
{
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
pvr2_trace(PVR2_TRACE_CTXT,"pvr2_context %p (destroy)",mp);
if (mp->hdw) pvr2_hdw_destroy(mp->hdw);
pvr2_context_set_notify(mp, 0);
mutex_lock(&pvr2_context_mutex);
if (mp->exist_next) {
mp->exist_next->exist_prev = mp->exist_prev;
} else {
pvr2_context_exist_last = mp->exist_prev;
}
if (mp->exist_prev) {
mp->exist_prev->exist_next = mp->exist_next;
} else {
pvr2_context_exist_first = mp->exist_next;
}
if (!pvr2_context_exist_first) {
/* Trigger wakeup on control thread in case it is waiting
for an exit condition. */
wake_up(&pvr2_context_sync_data);
}
mutex_unlock(&pvr2_context_mutex);
kfree(mp);
}
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
static void pvr2_context_notify(struct pvr2_context *mp)
{
pvr2_context_set_notify(mp,!0);
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
}
static void pvr2_context_check(struct pvr2_context *mp)
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
{
struct pvr2_channel *ch1, *ch2;
pvr2_trace(PVR2_TRACE_CTXT,
"pvr2_context %p (notify)", mp);
if (!mp->initialized_flag && !mp->disconnect_flag) {
mp->initialized_flag = !0;
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
pvr2_trace(PVR2_TRACE_CTXT,
"pvr2_context %p (initialize)", mp);
/* Finish hardware initialization */
if (pvr2_hdw_initialize(mp->hdw,
(void (*)(void *))pvr2_context_notify,
mp)) {
mp->video_stream.stream =
pvr2_hdw_get_video_stream(mp->hdw);
/* Trigger interface initialization. By doing this
here initialization runs in our own safe and
cozy thread context. */
if (mp->setup_func) mp->setup_func(mp);
} else {
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
pvr2_trace(PVR2_TRACE_CTXT,
"pvr2_context %p (thread skipping setup)",
mp);
/* Even though initialization did not succeed,
we're still going to continue anyway. We need
to do this in order to await the expected
disconnect (which we will detect in the normal
course of operation). */
}
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
}
for (ch1 = mp->mc_first; ch1; ch1 = ch2) {
ch2 = ch1->mc_next;
if (ch1->check_func) ch1->check_func(ch1);
}
if (mp->disconnect_flag && !mp->mc_first) {
/* Go away... */
pvr2_context_destroy(mp);
return;
}
}
static int pvr2_context_shutok(void)
{
return kthread_should_stop() && (pvr2_context_exist_first == NULL);
}
static int pvr2_context_thread_func(void *foo)
{
struct pvr2_context *mp;
pvr2_trace(PVR2_TRACE_CTXT,"pvr2_context thread start");
do {
while ((mp = pvr2_context_notify_first) != NULL) {
pvr2_context_set_notify(mp, 0);
pvr2_context_check(mp);
}
wait_event_interruptible(
pvr2_context_sync_data,
((pvr2_context_notify_first != NULL) ||
pvr2_context_shutok()));
} while (!pvr2_context_shutok());
pvr2_trace(PVR2_TRACE_CTXT,"pvr2_context thread end");
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
return 0;
}
int pvr2_context_global_init(void)
{
pvr2_context_thread_ptr = kthread_run(pvr2_context_thread_func,
0,
"pvrusb2-context");
return (pvr2_context_thread_ptr ? 0 : -ENOMEM);
}
void pvr2_context_global_done(void)
{
kthread_stop(pvr2_context_thread_ptr);
}
struct pvr2_context *pvr2_context_create(
struct usb_interface *intf,
const struct usb_device_id *devid,
void (*setup_func)(struct pvr2_context *))
{
struct pvr2_context *mp = NULL;
mp = kzalloc(sizeof(*mp),GFP_KERNEL);
if (!mp) goto done;
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
pvr2_trace(PVR2_TRACE_CTXT,"pvr2_context %p (create)",mp);
mp->setup_func = setup_func;
mutex_init(&mp->mutex);
mutex_lock(&pvr2_context_mutex);
mp->exist_prev = pvr2_context_exist_last;
mp->exist_next = NULL;
pvr2_context_exist_last = mp;
if (mp->exist_prev) {
mp->exist_prev->exist_next = mp;
} else {
pvr2_context_exist_first = mp;
}
mutex_unlock(&pvr2_context_mutex);
mp->hdw = pvr2_hdw_create(intf,devid);
if (!mp->hdw) {
pvr2_context_destroy(mp);
mp = NULL;
goto done;
}
pvr2_context_set_notify(mp, !0);
done:
return mp;
}
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
static void pvr2_context_enter(struct pvr2_context *mp)
{
mutex_lock(&mp->mutex);
}
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
static void pvr2_context_exit(struct pvr2_context *mp)
{
int destroy_flag = 0;
if (!(mp->mc_first || !mp->disconnect_flag)) {
destroy_flag = !0;
}
mutex_unlock(&mp->mutex);
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
if (destroy_flag) pvr2_context_notify(mp);
}
void pvr2_context_disconnect(struct pvr2_context *mp)
{
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
pvr2_hdw_disconnect(mp->hdw);
mp->disconnect_flag = !0;
pvr2_context_notify(mp);
}
void pvr2_channel_init(struct pvr2_channel *cp,struct pvr2_context *mp)
{
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
pvr2_context_enter(mp);
cp->hdw = mp->hdw;
cp->mc_head = mp;
cp->mc_next = NULL;
cp->mc_prev = mp->mc_last;
if (mp->mc_last) {
mp->mc_last->mc_next = cp;
} else {
mp->mc_first = cp;
}
mp->mc_last = cp;
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
pvr2_context_exit(mp);
}
static void pvr2_channel_disclaim_stream(struct pvr2_channel *cp)
{
if (!cp->stream) return;
pvr2_stream_kill(cp->stream->stream);
cp->stream->user = NULL;
cp->stream = NULL;
}
void pvr2_channel_done(struct pvr2_channel *cp)
{
struct pvr2_context *mp = cp->mc_head;
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
pvr2_context_enter(mp);
pvr2_channel_disclaim_stream(cp);
if (cp->mc_next) {
cp->mc_next->mc_prev = cp->mc_prev;
} else {
mp->mc_last = cp->mc_prev;
}
if (cp->mc_prev) {
cp->mc_prev->mc_next = cp->mc_next;
} else {
mp->mc_first = cp->mc_next;
}
cp->hdw = NULL;
V4L/DVB (7321): pvrusb2: Rework context handling and initialization This change significantly rearranges pvr2_context level initialization and operation: 1. A new kernel thread is set up for management of the context. 2. Destruction of the pvr2_context instance is moved into the kernel thread. No other context is able to remove the instance; doing this simplifies lock handling. 3. The callback into pvrusb2-main, which is used to trigger initialization of each interface, is now issued from this kernel thread. Previously it had been indirectly issued out of the work queue thread in pvr2_hdw, which led to deadlock issues if the interface needed to change a control setting (which in turn requires dispatch of another work queue entry). 4. Callbacks into the interfaces (via the pvr2_channel structure) are now issued strictly from this thread. The net result of this is that such callback functions can now also safely operate driver controls without deadlocking the work queue. (At the moment this is not actually a problem, but I'm anticipating issues with this in the future). 5. There is no longer any need for anyone to enter / exit the pvr2_context structure. Implementation of the kernel thread here allows this all to be internal now, simplifying other logic. 6. A very very longstanding issue involving a mutex deadlock between the pvrusb2 driver and v4l should now be solved. The deadlock involved the pvr2_context mutex and a globals-protecting mutex in v4l. During initialization the driver would take the pvr2_context mutex first then the v4l2 interface would register with v4l and implicitly take the v4l mutex. Later when v4l would call back into the driver, the two mutexes could possibly be taken in the opposite order, a situation that can lead to deadlock. In practice this really wasn't an issue unless a v4l app tried to start VERY early after the driver appeared. However it still needed to be solved, and with the use of the kernel thread relieving need for pvr2_context mutex, the problem should be finally solved. Signed-off-by: Mike Isely <isely@pobox.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2008-04-22 17:45:45 +00:00
pvr2_context_exit(mp);
}
int pvr2_channel_claim_stream(struct pvr2_channel *cp,
struct pvr2_context_stream *sp)
{
int code = 0;
pvr2_context_enter(cp->mc_head); do {
if (sp == cp->stream) break;
if (sp && sp->user) {
code = -EBUSY;
break;
}
pvr2_channel_disclaim_stream(cp);
if (!sp) break;
sp->user = cp;
cp->stream = sp;
} while (0); pvr2_context_exit(cp->mc_head);
return code;
}
// This is the marker for the real beginning of a legitimate mpeg2 stream.
static char stream_sync_key[] = {
0x00, 0x00, 0x01, 0xba,
};
struct pvr2_ioread *pvr2_channel_create_mpeg_stream(
struct pvr2_context_stream *sp)
{
struct pvr2_ioread *cp;
cp = pvr2_ioread_create();
if (!cp) return NULL;
pvr2_ioread_setup(cp,sp->stream);
pvr2_ioread_set_sync_key(cp,stream_sync_key,sizeof(stream_sync_key));
return cp;
}
/*
Stuff for Emacs to see, in order to encourage consistent editing style:
*** Local Variables: ***
*** mode: c ***
*** fill-column: 75 ***
*** tab-width: 8 ***
*** c-basic-offset: 8 ***
*** End: ***
*/