[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAM_iQpUEvWrMPWQussA31kEAiutLR4WUVOtsohHm5hg4wt59Hw@mail.gmail.com>
Date: Wed, 23 May 2018 21:40:47 -0700
From: Cong Wang <xiyou.wangcong@...il.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Ivan Babrou <ibobrik@...il.com>,
LKML <linux-kernel@...r.kernel.org>,
Paul Turner <pjt@...gle.com>, Mike Galbraith <efault@....de>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...nel.org>
Subject: Re: [PATCH 3/3] sched/fair: add tracepoints for cfs throttle
On Wed, May 23, 2018 at 2:09 AM, Peter Zijlstra <peterz@...radead.org> wrote:
> On Mon, May 21, 2018 at 11:35:38PM -0700, Ivan Babrou wrote:
>> > + TP_printk("path=%s cpu=%d runtime_remaining=%lld",
>> > __get_str(cfs_path),
>> > + __entry->cpu, __entry->runtime_remaining)
>> >
>>
>> Can you add "[ns]" as the unit to runtime_remaining? We have it in
>> "sched:sched_stat_runtime".
>
> Yeah, don't worry. I hate tracepoints, I regret the existing ones, no
> new ones will happen.
No matter you hate them or not, they are useful:
https://github.com/iovisor/bcc/blob/master/tools/runqlat.py
Unlike in sched, tracepoints are welcome in networking:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/log/include/trace/events/tcp.h
I don't want to change your mind, just want to show the facts.
Powered by blists - more mailing lists