[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20091123102806.GD25978@dhcp-lab-161.englab.brq.redhat.com>
Date: Mon, 23 Nov 2009 11:28:07 +0100
From: Stanislaw Gruszka <sgruszka@...hat.com>
To: Hidetoshi Seto <seto.hidetoshi@...fujitsu.com>
Cc: linux-kernel@...r.kernel.org, Ingo Molnar <mingo@...e.hu>,
Peter Zijlstra <peterz@...radead.org>,
Spencer Candland <spencer@...ehost.com>,
Oleg Nesterov <oleg@...hat.com>,
Balbir Singh <balbir@...ibm.com>,
Américo Wang <xiyou.wangcong@...il.com>
Subject: Re: [PATCH tip/sched/core] introduce task_times() to replace
task_[us]time() pair
On Fri, Nov 20, 2009 at 01:44:10PM +0900, Hidetoshi Seto wrote:
> Function task_[us]times() are called consecutively in almost all
> cases. However task_stime() is implemented to call task_utime()
> from its inside, so such paired calls run task_utime() twice.
>
> It means we do heavy divisions (div_u64 + do_div) twice to get
> stime and utime which can be obtained at same time by one set
> of divisions.
>
> This patch introduces task_times(*tsk, *utime, *stime) to get
> stime and utime at once, in better, optimized way.
>
> Signed-off-by: Hidetoshi Seto <seto.hidetoshi@...fujitsu.com>
[snip]
> @@ -5155,6 +5155,14 @@ cputime_t task_stime(struct task_struct *p)
> {
> return p->stime;
> }
> +
> +void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st)
> +{
> + if (ut)
> + *ut = task_utime(p);
> + if (st)
> + *st = task_stime(p);
> +}
> #else
I think task_{u,s}time are not needed anymore. Can we just fully get
rid of them and only use task_times() ?
> #ifndef nsecs_to_cputime
> @@ -5162,41 +5170,48 @@ cputime_t task_stime(struct task_struct *p)
> msecs_to_cputime(div_u64((__nsecs), NSEC_PER_MSEC))
> #endif
Could we furhter optimize this? Perhaps we can use below code
(taken from timespec_to_jiffies()):
cputime = (nsec * NSEC_CONVERSION) >>
(NSEC_JIFFIE_SC - SEC_JIFFIE_SC))) >> SEC_JIFFIE_SC;
Stanislaw
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists