Currently show_workqueue_state shows the state of all workqueues and of
all worker pools. In certain cases we may need to dump state of only a
specific workqueue or worker pool. For example in destroy_workqueue we
only need to show state of the workqueue which is getting destroyed.
So rename show_workqueue_state to show_all_workqueues(to signify it
dumps state of all busy workqueues) and divide it into more granular
functions (show_one_workqueue and show_one_worker_pool), that would show
states of individual workqueues and worker pools and can be used in
cases such as the one mentioned above.
Also, as mentioned earlier, make destroy_workqueue dump data pertaining
to only the workqueue that is being destroyed and make user(s) of
earlier interface(show_workqueue_state), use new interface
(show_all_workqueues).
Signed-off-by: Imran Khan <[email protected]>
---
Changes in v4:
- Rebase on git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq.git
for-5.16
drivers/tty/sysrq.c | 2 +-
include/linux/workqueue.h | 3 +-
kernel/power/process.c | 2 +-
kernel/workqueue.c | 172 +++++++++++++++++++++-----------------
4 files changed, 100 insertions(+), 79 deletions(-)
diff --git a/drivers/tty/sysrq.c b/drivers/tty/sysrq.c
index c911196ac893..8d0f07509ca7 100644
--- a/drivers/tty/sysrq.c
+++ b/drivers/tty/sysrq.c
@@ -296,7 +296,7 @@ static const struct sysrq_key_op sysrq_showregs_op = {
static void sysrq_handle_showstate(int key)
{
show_state();
- show_workqueue_state();
+ show_all_workqueues();
}
static const struct sysrq_key_op sysrq_showstate_op = {
.handler = sysrq_handle_showstate,
diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index 74d3c1efd9bb..7fee9b6cfede 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -469,7 +469,8 @@ extern bool workqueue_congested(int cpu, struct workqueue_struct *wq);
extern unsigned int work_busy(struct work_struct *work);
extern __printf(1, 2) void set_worker_desc(const char *fmt, ...);
extern void print_worker_info(const char *log_lvl, struct task_struct *task);
-extern void show_workqueue_state(void);
+extern void show_all_workqueues(void);
+extern void show_one_workqueue(struct workqueue_struct *wq);
extern void wq_worker_comm(char *buf, size_t size, struct task_struct *task);
/**
diff --git a/kernel/power/process.c b/kernel/power/process.c
index 37401c99b7d7..b7e7798637b8 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -94,7 +94,7 @@ static int try_to_freeze_tasks(bool user_only)
todo - wq_busy, wq_busy);
if (wq_busy)
- show_workqueue_state();
+ show_all_workqueues();
if (!wakeup || pm_debug_messages_on) {
read_lock(&tasklist_lock);
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 76988f39ed5a..1a7df882f55e 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -375,6 +375,7 @@ EXPORT_SYMBOL_GPL(system_freezable_power_efficient_wq);
static int worker_thread(void *__worker);
static void workqueue_sysfs_unregister(struct workqueue_struct *wq);
static void show_pwq(struct pool_workqueue *pwq);
+static void show_one_worker_pool(struct worker_pool *pool);
#define CREATE_TRACE_POINTS
#include <trace/events/workqueue.h>
@@ -4447,7 +4448,7 @@ void destroy_workqueue(struct workqueue_struct *wq)
raw_spin_unlock_irq(&pwq->pool->lock);
mutex_unlock(&wq->mutex);
mutex_unlock(&wq_pool_mutex);
- show_workqueue_state();
+ show_one_workqueue(wq);
return;
}
raw_spin_unlock_irq(&pwq->pool->lock);
@@ -4797,97 +4798,116 @@ static void show_pwq(struct pool_workqueue *pwq)
}
/**
- * show_workqueue_state - dump workqueue state
- *
- * Called from a sysrq handler or try_to_freeze_tasks() and prints out
- * all busy workqueues and pools.
+ * show_one_workqueue - dump state of specified workqueue
+ * @wq: workqueue whose state will be printed
*/
-void show_workqueue_state(void)
+void show_one_workqueue(struct workqueue_struct *wq)
{
- struct workqueue_struct *wq;
- struct worker_pool *pool;
+ struct pool_workqueue *pwq;
+ bool idle = true;
unsigned long flags;
- int pi;
-
- rcu_read_lock();
- pr_info("Showing busy workqueues and worker pools:\n");
-
- list_for_each_entry_rcu(wq, &workqueues, list) {
- struct pool_workqueue *pwq;
- bool idle = true;
-
- for_each_pwq(pwq, wq) {
- if (pwq->nr_active || !list_empty(&pwq->inactive_works)) {
- idle = false;
- break;
- }
+ for_each_pwq(pwq, wq) {
+ if (pwq->nr_active || !list_empty(&pwq->inactive_works)) {
+ idle = false;
+ break;
}
- if (idle)
- continue;
+ }
+ if (idle) /* Nothing to print for idle workqueue */
+ return;
- pr_info("workqueue %s: flags=0x%x\n", wq->name, wq->flags);
+ pr_info("workqueue %s: flags=0x%x\n", wq->name, wq->flags);
- for_each_pwq(pwq, wq) {
- raw_spin_lock_irqsave(&pwq->pool->lock, flags);
- if (pwq->nr_active || !list_empty(&pwq->inactive_works)) {
- /*
- * Defer printing to avoid deadlocks in console
- * drivers that queue work while holding locks
- * also taken in their write paths.
- */
- printk_deferred_enter();
- show_pwq(pwq);
- printk_deferred_exit();
- }
- raw_spin_unlock_irqrestore(&pwq->pool->lock, flags);
+ for_each_pwq(pwq, wq) {
+ raw_spin_lock_irqsave(&pwq->pool->lock, flags);
+ if (pwq->nr_active || !list_empty(&pwq->inactive_works)) {
/*
- * We could be printing a lot from atomic context, e.g.
- * sysrq-t -> show_workqueue_state(). Avoid triggering
- * hard lockup.
+ * Defer printing to avoid deadlocks in console
+ * drivers that queue work while holding locks
+ * also taken in their write paths.
*/
- touch_nmi_watchdog();
- }
- }
-
- for_each_pool(pool, pi) {
- struct worker *worker;
- bool first = true;
-
- raw_spin_lock_irqsave(&pool->lock, flags);
- if (pool->nr_workers == pool->nr_idle)
- goto next_pool;
- /*
- * Defer printing to avoid deadlocks in console drivers that
- * queue work while holding locks also taken in their write
- * paths.
- */
- printk_deferred_enter();
- pr_info("pool %d:", pool->id);
- pr_cont_pool_info(pool);
- pr_cont(" hung=%us workers=%d",
- jiffies_to_msecs(jiffies - pool->watchdog_ts) / 1000,
- pool->nr_workers);
- if (pool->manager)
- pr_cont(" manager: %d",
- task_pid_nr(pool->manager->task));
- list_for_each_entry(worker, &pool->idle_list, entry) {
- pr_cont(" %s%d", first ? "idle: " : "",
- task_pid_nr(worker->task));
- first = false;
+ printk_deferred_enter();
+ show_pwq(pwq);
+ printk_deferred_exit();
}
- pr_cont("\n");
- printk_deferred_exit();
- next_pool:
- raw_spin_unlock_irqrestore(&pool->lock, flags);
+ raw_spin_unlock_irqrestore(&pwq->pool->lock, flags);
/*
* We could be printing a lot from atomic context, e.g.
- * sysrq-t -> show_workqueue_state(). Avoid triggering
+ * sysrq-t -> show_all_workqueues(). Avoid triggering
* hard lockup.
*/
touch_nmi_watchdog();
}
+}
+
+/**
+ * show_one_worker_pool - dump state of specified worker pool
+ * @pool: worker pool whose state will be printed
+ */
+static void show_one_worker_pool(struct worker_pool *pool)
+{
+ struct worker *worker;
+ bool first = true;
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&pool->lock, flags);
+ if (pool->nr_workers == pool->nr_idle)
+ goto next_pool;
+ /*
+ * Defer printing to avoid deadlocks in console drivers that
+ * queue work while holding locks also taken in their write
+ * paths.
+ */
+ printk_deferred_enter();
+ pr_info("pool %d:", pool->id);
+ pr_cont_pool_info(pool);
+ pr_cont(" hung=%us workers=%d",
+ jiffies_to_msecs(jiffies - pool->watchdog_ts) / 1000,
+ pool->nr_workers);
+ if (pool->manager)
+ pr_cont(" manager: %d",
+ task_pid_nr(pool->manager->task));
+ list_for_each_entry(worker, &pool->idle_list, entry) {
+ pr_cont(" %s%d", first ? "idle: " : "",
+ task_pid_nr(worker->task));
+ first = false;
+ }
+ pr_cont("\n");
+ printk_deferred_exit();
+next_pool:
+ raw_spin_unlock_irqrestore(&pool->lock, flags);
+ /*
+ * We could be printing a lot from atomic context, e.g.
+ * sysrq-t -> show_all_workqueues(). Avoid triggering
+ * hard lockup.
+ */
+ touch_nmi_watchdog();
+
+}
+
+/**
+ * show_all_workqueues - dump workqueue state
+ *
+ * Called from a sysrq handler or try_to_freeze_tasks() and prints out
+ * all busy workqueues and pools.
+ */
+void show_all_workqueues(void)
+{
+ struct workqueue_struct *wq;
+ struct worker_pool *pool;
+ int pi;
+
+ rcu_read_lock();
+
+ pr_info("Showing busy workqueues and worker pools:\n");
+
+ list_for_each_entry_rcu(wq, &workqueues, list)
+ show_one_workqueue(wq);
+
+ for_each_pool(pool, pi)
+ show_one_worker_pool(pool);
+
rcu_read_unlock();
}
@@ -5876,7 +5896,7 @@ static void wq_watchdog_timer_fn(struct timer_list *unused)
rcu_read_unlock();
if (lockup_detected)
- show_workqueue_state();
+ show_all_workqueues();
wq_watchdog_reset_touched();
mod_timer(&wq_watchdog_timer, jiffies + thresh);
base-commit: d25302e46592c97d29f70ccb1be558df31a9a360
--
2.30.2
On Wed, Oct 20, 2021 at 02:09:00PM +1100, Imran Khan wrote:
> Currently show_workqueue_state shows the state of all workqueues and of
> all worker pools. In certain cases we may need to dump state of only a
> specific workqueue or worker pool. For example in destroy_workqueue we
> only need to show state of the workqueue which is getting destroyed.
>
> So rename show_workqueue_state to show_all_workqueues(to signify it
> dumps state of all busy workqueues) and divide it into more granular
> functions (show_one_workqueue and show_one_worker_pool), that would show
> states of individual workqueues and worker pools and can be used in
> cases such as the one mentioned above.
>
> Also, as mentioned earlier, make destroy_workqueue dump data pertaining
> to only the workqueue that is being destroyed and make user(s) of
> earlier interface(show_workqueue_state), use new interface
> (show_all_workqueues).
>
> Signed-off-by: Imran Khan <[email protected]>
> ---
> Changes in v4:
> - Rebase on git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq.git
> for-5.16
"RFC" means "I am not sure about this patch, please do not apply it."
Yet a v4 is very odd, why not be sure about this?
thanks,
greg k-h
Hi Greg,
On 20/10/21 5:00 pm, Greg KH wrote:
> On Wed, Oct 20, 2021 at 02:09:00PM +1100, Imran Khan wrote:
[...]
>>
>> Signed-off-by: Imran Khan <[email protected]>
>> ---
>> Changes in v4:
>> - Rebase on git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq.git
>> for-5.16
>
> "RFC" means "I am not sure about this patch, please do not apply it."
>
Thanks for clarifying this.
> Yet a v4 is very odd, why not be sure about this?
I have not got any negative feedback about the change so far, so I am
fairly sure about it now. There was a mistake in my understanding where
I thought that something that starts as an RFC remains like that in
subsequent revisions. I will definitely be mindful about it in future.
Could you please let me know if I should send a "PATCH v5", removing RFC
from the subject.
Thanks,
Imran
>
> thanks,
>
> greg k-h
>
On Wed, Oct 20, 2021 at 02:09:00PM +1100, Imran Khan wrote:
> Currently show_workqueue_state shows the state of all workqueues and of
> all worker pools. In certain cases we may need to dump state of only a
> specific workqueue or worker pool. For example in destroy_workqueue we
> only need to show state of the workqueue which is getting destroyed.
>
> So rename show_workqueue_state to show_all_workqueues(to signify it
> dumps state of all busy workqueues) and divide it into more granular
> functions (show_one_workqueue and show_one_worker_pool), that would show
> states of individual workqueues and worker pools and can be used in
> cases such as the one mentioned above.
>
> Also, as mentioned earlier, make destroy_workqueue dump data pertaining
> to only the workqueue that is being destroyed and make user(s) of
> earlier interface(show_workqueue_state), use new interface
> (show_all_workqueues).
>
> Signed-off-by: Imran Khan <[email protected]>
Applied to wq/for-5.16.
Thank you.
--
tejun