Messages in this thread | | | Date | Fri, 9 Jul 2021 15:06:23 +0200 | From | Peter Zijlstra <> | Subject | Re: [PATCH v2] kernel/sched/core: Print the time when the process entered the D state |
| |
On Fri, Jul 09, 2021 at 04:54:49AM -0700, zhouchuangao wrote: > Sometimes we need to trigger /proc/sysrq-trigger to view the information > of the D state process in the system. > e.g: > echo w > /proc/sysrq-trigger > > Here we print the scheduling time of the process, which is convenient > for us to judge the state of the system and for debugging. > > Signed-off-by: zhouchuangao <zhouchuangao@vivo.com> > --- > kernel/sched/core.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/kernel/sched/core.c b/kernel/sched/core.c > index cf16f8f..ab422ab 100644 > --- a/kernel/sched/core.c > +++ b/kernel/sched/core.c > @@ -8154,10 +8154,12 @@ void sched_show_task(struct task_struct *p) > if (pid_alive(p)) > ppid = task_pid_nr(rcu_dereference(p->real_parent)); > rcu_read_unlock(); > - pr_cont(" stack:%5lu pid:%5d ppid:%6d flags:0x%08lx\n", > + pr_cont(" stack:%5lu pid:%5d ppid:%6d flags:0x%08lx", > free, task_pid_nr(p), ppid, > (unsigned long)task_thread_info(p)->flags); > - > +#ifdef CONFIG_SCHED_INFO > + pr_cont(" sched:%llu\n", p->sched_info.last_arrival);
"sched:" isn't really descriptive here, is it?
> +#endif
Also, you lost the "\n" for SCHED_INFO=n.
Please, take more than 5 minutes to fix the issues and test your patch before sending it again.
> print_worker_info(KERN_INFO, p); > print_stop_info(KERN_INFO, p); > show_stack(p, NULL, KERN_INFO); > -- > 2.7.4 >
| |