[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <BANLkTim=di8pbt1CzHjHYSC0SbtgJzbmkw@mail.gmail.com>
Date: Tue, 17 May 2011 10:28:35 +0800
From: Yong Zhang <yong.zhang0@...il.com>
To: Hillf Danton <dhillf@...il.com>
Cc: LKML <linux-kernel@...r.kernel.org>, Ingo Molnar <mingo@...e.hu>,
Peter Zijlstra <peterz@...radead.org>,
Mike Galbraith <efault@....de>
Subject: Re: [PATCH] sched: fix priority leakage in pick_next_highest_task_rt()
On Mon, May 16, 2011 at 8:55 PM, Hillf Danton <dhillf@...il.com> wrote:
> When picking the second highest RT task for a given runqueue, if no
> task found after scanning the queue of priority == idx, the next idx
> should also be checked even in case that next is already existing, or
> the window of priority leakage could be opened.
I don't see what kind of problem you patch will fix.
And mind explaining how priority leakage could happen?
Thanks,
Yong
>
> Signed-off-by: Hillf Danton <dhillf@...il.com>
> ---
>
> --- a/kernel/sched_rt.c 2011-04-27 11:48:50.000000000 +0800
> +++ b/kernel/sched_rt.c 2011-05-16 19:58:42.000000000 +0800
> @@ -1166,6 +1166,8 @@ static struct task_struct *pick_next_hig
> int idx;
>
> for_each_leaf_rt_rq(rt_rq, rq) {
> + struct task_struct *this;
> +
> array = &rt_rq->active;
> idx = sched_find_first_bit(array->bitmap);
> next_idx:
> @@ -1173,6 +1175,7 @@ next_idx:
> continue;
> if (next && next->prio < idx)
> continue;
> + this = NULL;
> list_for_each_entry(rt_se, array->queue + idx, run_list) {
> struct task_struct *p;
>
> @@ -1181,11 +1184,15 @@ next_idx:
>
> p = rt_task_of(rt_se);
> if (pick_rt_task(rq, p, cpu)) {
> - next = p;
> + this = p;
> break;
> }
> }
> - if (!next) {
> + if (this != NULL)
> + next = this;
> + else { /*
> + * we have to check next idx even if next != NULL
> + */
> idx = find_next_bit(array->bitmap, MAX_RT_PRIO, idx+1);
> goto next_idx;
> }
>
--
Only stand for myself
Powered by blists - more mailing lists