Re: [PATCH linux next] delayacct: add delay max to record delay peak

[Andrew Morton]

On Tue, 3 Dec 2024 16:48:48 +0800 (CST) <jiang.kun2@xxxxxxxxxx> wrote:

> From: Wang Yaxin <wang.yaxin@xxxxxxxxxx>
> 
> Introduce the use cases of delay max, which can help quickly detect
> potential abnormal delays in the system and record the types and
> specific details of delay spikes.
>   
> Problem
> ========
> Delay accounting can track the average delay of processes to show
> system workload. However, when a process experiences a significant
> delay, maybe a delay spike, which adversely affects performance,
> getdelays can only display the average system delay over a period
> of time. Yet, average delay is unhelpful for diagnosing delay peak.
> It is not even possible to determine which type of delay has spiked,
> as this information might be masked by the average delay.
> 
> Solution
> =========
> the 'delay max' can display delay peak since the system's startup,
> which can record potential abnormal delays over time, including
> the type of delay and the maximum delay. This is helpful for
> quickly identifying crash caused by delay.

Seems sensible, and the implementation is straightforward.  I'll queue
it for testing and review.

>  include/linux/delayacct.h      |  7 ++++
>  include/linux/sched.h          |  3 ++
>  include/uapi/linux/taskstats.h |  9 ++++++
>  kernel/delayacct.c             | 35 ++++++++++++++------
>  kernel/sched/stats.h           |  5 ++-
>  tools/accounting/getdelays.c   | 59 ++++++++++++++++++++--------------

Please prepare a v2 which includes an update to
Documentation/accounting/delay-accounting.rst.

> 
> ...
>
> --- a/kernel/delayacct.c
> +++ b/kernel/delayacct.c
> @@ -95,7 +95,7 @@ void __delayacct_tsk_init(struct task_struct *tsk)
>   * Finish delay accounting for a statistic using its timestamps (@start),
>   * accumalator (@total) and @count

And I wouldn't be offended if v2 fixed that spelling error!  "accumulator".