Re: [PATCH v2] sched: Warn on long periods of pending need_resched

[Peter Zijlstra]

On Mon, Mar 22, 2021 at 08:57:06PM -0700, Josh Don wrote:
> From: Paul Turner <pjt@xxxxxxxxxx>
> 
> CPU scheduler marks need_resched flag to signal a schedule() on a
> particular CPU. But, schedule() may not happen immediately in cases
> where the current task is executing in the kernel mode (no
> preemption state) for extended periods of time.
> 
> This patch adds a warn_on if need_resched is pending for more than the
> time specified in sysctl resched_latency_warn_ms. If it goes off, it is
> likely that there is a missing cond_resched() somewhere. Monitoring is
> done via the tick and the accuracy is hence limited to jiffy scale. This
> also means that we won't trigger the warning if the tick is disabled.
> 
> This feature is default disabled. It can be toggled on using sysctl
> resched_latency_warn_enabled.
> 
> Signed-off-by: Paul Turner <pjt@xxxxxxxxxx>
> Signed-off-by: Josh Don <joshdon@xxxxxxxxxx>
> ---
> Delta from v1:
> - separate sysctl for enabling/disabling and triggering warn_once
>   behavior
> - add documentation
> - static branch for the enable
>  Documentation/admin-guide/sysctl/kernel.rst | 23 ++++++
>  include/linux/sched/sysctl.h                |  4 ++
>  kernel/sched/core.c                         | 78 ++++++++++++++++++++-
>  kernel/sched/debug.c                        | 10 +++
>  kernel/sched/sched.h                        | 10 +++
>  kernel/sysctl.c                             | 24 +++++++
>  6 files changed, 148 insertions(+), 1 deletion(-)
> 
> diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst
> index 1d56a6b73a4e..2d4a21d3b79f 100644
> --- a/Documentation/admin-guide/sysctl/kernel.rst
> +++ b/Documentation/admin-guide/sysctl/kernel.rst
> @@ -1077,6 +1077,29 @@ ROM/Flash boot loader. Maybe to tell it what to do after
>  rebooting. ???
>  
>  
> +resched_latency_warn_enabled
> +============================
> +
> +Enables/disables a warning that will trigger if need_resched is set for
> +longer than sysctl ``resched_latency_warn_ms``. This warning likely
> +indicates a kernel bug, such as a failure to call cond_resched().
> +
> +Requires ``CONFIG_SCHED_DEBUG``.
> +
> +
> +resched_latency_warn_ms
> +=======================
> +
> +See ``resched_latency_warn_enabled``.
> +
> +
> +resched_latency_warn_once
> +=========================
> +
> +If set, ``resched_latency_warn_enabled`` will only trigger one warning
> +per boot.
> +
> +
>  sched_energy_aware
>  ==================

Should we perhaps take out all SCHED_DEBUG sysctls and move them to
/debug/sched/ ? (along with the existing /debug/sched_{debug,features,preemp}
files)

Having all that in sysctl and documented gives them far too much sheen
of ABI.

Not saying this patch should do that, just as a general observation.