Freezer retry loops during suspend are often triggered by tasks entering
D-state (TASK_UNINTERRUPTIBLE), which cannot be frozen.  This patch adds
a simple retry counter to freeze_processes() to help quantify how many
attempts were required before all tasks entered the frozen state. This
is useful for performance tuning and debugging unpredictable suspend
delays.

A new dmesg log is added for visibility:

freeze round: xx, tasks to freeze: xx

This message allows users to correlate freeze instability with system
state.

Signed-off-by: Zihuan Zhang <zhangzihuan@kylinos.cn>
---
 kernel/power/process.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/kernel/power/process.c b/kernel/power/process.c
index dc0dfc349f22..87616ca710ac 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -32,6 +32,7 @@ static int try_to_freeze_tasks(bool user_only)
 	struct task_struct *g, *p;
 	unsigned long end_time;
 	unsigned int todo;
+	unsigned int retry = 0;
 	bool wq_busy = false;
 	ktime_t start, end, elapsed;
 	unsigned int elapsed_msecs;
@@ -63,6 +64,8 @@ static int try_to_freeze_tasks(bool user_only)
 			todo += wq_busy;
 		}
 
+		pm_pr_dbg("freeze round: %d, task to freeze: %d\n", retry, todo);
+
 		if (!todo || time_after(jiffies, end_time))
 			break;
 
@@ -79,6 +82,7 @@ static int try_to_freeze_tasks(bool user_only)
 		usleep_range(sleep_usecs / 2, sleep_usecs);
 		if (sleep_usecs < 8 * USEC_PER_MSEC)
 			sleep_usecs *= 2;
+		retry++;
 	}
 
 	end = ktime_get_boottime();
-- 
2.25.1

On Thu, Jun 19, 2025 at 5:54 AM Zihuan Zhang <zhangzihuan@kylinos.cn> wrote:
>
> Freezer retry loops during suspend are often triggered by tasks entering
> D-state (TASK_UNINTERRUPTIBLE), which cannot be frozen.  This patch adds
> a simple retry counter to freeze_processes() to help quantify how many
> attempts were required before all tasks entered the frozen state. This
> is useful for performance tuning and debugging unpredictable suspend
> delays.
>
> A new dmesg log is added for visibility:
>
> freeze round: xx, tasks to freeze: xx
>
> This message allows users to correlate freeze instability with system
> state.
>
> Signed-off-by: Zihuan Zhang <zhangzihuan@kylinos.cn>
> ---
>  kernel/power/process.c | 4 ++++
>  1 file changed, 4 insertions(+)
>
> diff --git a/kernel/power/process.c b/kernel/power/process.c
> index dc0dfc349f22..87616ca710ac 100644
> --- a/kernel/power/process.c
> +++ b/kernel/power/process.c
> @@ -32,6 +32,7 @@ static int try_to_freeze_tasks(bool user_only)
>         struct task_struct *g, *p;
>         unsigned long end_time;
>         unsigned int todo;
> +       unsigned int retry = 0;
>         bool wq_busy = false;
>         ktime_t start, end, elapsed;
>         unsigned int elapsed_msecs;
> @@ -63,6 +64,8 @@ static int try_to_freeze_tasks(bool user_only)
>                         todo += wq_busy;
>                 }
>
> +               pm_pr_dbg("freeze round: %d, task to freeze: %d\n", retry, todo);
> +

This is kind of fine if it is really needed, but it needs to have a
separate on/off switch as far as I'm concerned.

Maybe use pr_debug() for it in which case it would be controlled via
dynamic debug?

>                 if (!todo || time_after(jiffies, end_time))
>                         break;
>
> @@ -79,6 +82,7 @@ static int try_to_freeze_tasks(bool user_only)
>                 usleep_range(sleep_usecs / 2, sleep_usecs);
>                 if (sleep_usecs < 8 * USEC_PER_MSEC)
>                         sleep_usecs *= 2;
> +               retry++;
>         }
>
>         end = ktime_get_boottime();
> --
> 2.25.1
>

Hi Rafael,
Thanks for the feedback.

在 2025/7/3 22:38, Rafael J. Wysocki 写道:
> On Thu, Jun 19, 2025 at 5:54 AM Zihuan Zhang <zhangzihuan@kylinos.cn> wrote:
>> Freezer retry loops during suspend are often triggered by tasks entering
>> D-state (TASK_UNINTERRUPTIBLE), which cannot be frozen.  This patch adds
>> a simple retry counter to freeze_processes() to help quantify how many
>> attempts were required before all tasks entered the frozen state. This
>> is useful for performance tuning and debugging unpredictable suspend
>> delays.
>>
>> A new dmesg log is added for visibility:
>>
>> freeze round: xx, tasks to freeze: xx
>>
>> This message allows users to correlate freeze instability with system
>> state.
>>
>> Signed-off-by: Zihuan Zhang <zhangzihuan@kylinos.cn>
>> ---
>>   kernel/power/process.c | 4 ++++
>>   1 file changed, 4 insertions(+)
>>
>> diff --git a/kernel/power/process.c b/kernel/power/process.c
>> index dc0dfc349f22..87616ca710ac 100644
>> --- a/kernel/power/process.c
>> +++ b/kernel/power/process.c
>> @@ -32,6 +32,7 @@ static int try_to_freeze_tasks(bool user_only)
>>          struct task_struct *g, *p;
>>          unsigned long end_time;
>>          unsigned int todo;
>> +       unsigned int retry = 0;
>>          bool wq_busy = false;
>>          ktime_t start, end, elapsed;
>>          unsigned int elapsed_msecs;
>> @@ -63,6 +64,8 @@ static int try_to_freeze_tasks(bool user_only)
>>                          todo += wq_busy;
>>                  }
>>
>> +               pm_pr_dbg("freeze round: %d, task to freeze: %d\n", retry, todo);
>> +
> This is kind of fine if it is really needed, but it needs to have a
> separate on/off switch as far as I'm concerned.
>
> Maybe use pr_debug() for it in which case it would be controlled via
> dynamic debug?
I agree with your suggestion — I’ll switch to pr_debug() in the next 
version so that the message can be controlled via dynamic debug.
>>                  if (!todo || time_after(jiffies, end_time))
>>                          break;
>>
>> @@ -79,6 +82,7 @@ static int try_to_freeze_tasks(bool user_only)
>>                  usleep_range(sleep_usecs / 2, sleep_usecs);
>>                  if (sleep_usecs < 8 * USEC_PER_MSEC)
>>                          sleep_usecs *= 2;
>> +               retry++;
>>          }
>>
>>          end = ktime_get_boottime();
>> --
>> 2.25.1
>>
Best regards,
Zihuan Zhang