linux-kernel - Re: [PATCH V3 4/4] Sched: Add tracepoint for sched time slice extension

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20250502091434.GV4198@noisy.programming.kicks-ass.net>
Date: Fri, 2 May 2025 11:14:34 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Prakash Sangappa <prakash.sangappa@...cle.com>
Cc: linux-kernel@...r.kernel.org, rostedt@...dmis.org,
	mathieu.desnoyers@...icios.com, tglx@...utronix.de,
	bigeasy@...utronix.de, kprateek.nayak@....com
Subject: Re: [PATCH V3 4/4] Sched: Add tracepoint for sched time slice
 extension

On Fri, May 02, 2025 at 01:59:55AM +0000, Prakash Sangappa wrote:
> Trace task's preemption due to a wakeups, getting delayed. Which
> can occurs when the running task requested extra time on cpu.
> Also, indicate the NEED_RESCHED flag set on the task getting
> cleared.
> 
> Signed-off-by: Prakash Sangappa <prakash.sangappa@...cle.com>
> ---
>  include/trace/events/sched.h | 28 ++++++++++++++++++++++++++++
>  kernel/entry/common.c        |  8 ++++++--
>  2 files changed, 34 insertions(+), 2 deletions(-)
> 
> diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h
> index 8994e97d86c1..4aa04044b14a 100644
> --- a/include/trace/events/sched.h
> +++ b/include/trace/events/sched.h
> @@ -296,6 +296,34 @@ TRACE_EVENT(sched_migrate_task,
>  		  __entry->orig_cpu, __entry->dest_cpu)
>  );
>  
> +/*
> + * Tracepoint for delayed resched requested by task:
> + */
> +TRACE_EVENT(sched_delay_resched,
> +
> +	TP_PROTO(struct task_struct *p, unsigned int resched_flg),
> +
> +	TP_ARGS(p, resched_flg),
> +
> +	TP_STRUCT__entry(
> +		__array( char, comm, TASK_COMM_LEN	)
> +		__field( pid_t, pid			)
> +		__field( int, cpu			)
> +		__field( int, flg			)
> +	),
> +
> +	TP_fast_assign(
> +		memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
> +		__entry->pid		= p->pid;
> +		__entry->cpu 		= task_cpu(p);
> +		__entry->flg		= resched_flg;
> +	),
> +
> +	TP_printk("comm=%s pid=%d cpu=%d resched_flg_cleared=0x%x",
> +		__entry->comm, __entry->pid, __entry->cpu, __entry->flg)
> +
> +);
> +
>  DECLARE_EVENT_CLASS(sched_process_template,
>  
>  	TP_PROTO(struct task_struct *p),
> diff --git a/kernel/entry/common.c b/kernel/entry/common.c
> index b26adccb32df..1951e6a4e9bc 100644
> --- a/kernel/entry/common.c
> +++ b/kernel/entry/common.c
> @@ -12,6 +12,7 @@
>  
>  #include "common.h"
>  
> +#include <trace/events/sched.h>
>  #define CREATE_TRACE_POINTS
>  #include <trace/events/syscalls.h>
>  
> @@ -100,9 +101,12 @@ __always_inline unsigned long exit_to_user_mode_loop(struct pt_regs *regs,
>  		local_irq_enable_exit_to_user(ti_work);
>  
>  		if (ti_work & (_TIF_NEED_RESCHED | _TIF_NEED_RESCHED_LAZY)) {
> -		       if (irq && rseq_delay_resched())
> +		       if (irq && rseq_delay_resched()) {
>  			       clear_tsk_need_resched(current);
> -		       else
> +			       trace_sched_delay_resched(current,
> +					       ti_work & (_TIF_NEED_RESCHED |
> +						       _TIF_NEED_RESCHED_LAZY));
> +		       } else
>  			       schedule();
>  		}

This is horrific coding style. But really why do we need this? I'm not,
in general, a fan of tracepoints.