linux-kernel - Re: [PATCH] sched: remove resetting exec_start in put_prev_task

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: <BANLkTing2boYvDQqcxaqveYhbG2H1Zyvrw@mail.gmail.com>
Date:	Fri, 10 Jun 2011 22:48:46 +0800
From:	Hillf Danton <dhillf@...il.com>
To:	Steven Rostedt <rostedt@...dmis.org>
Cc:	Yong Zhang <yong.zhang0@...il.com>,
	LKML <linux-kernel@...r.kernel.org>,
	Mike Galbraith <efault@....de>,
	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...e.hu>
Subject: Re: [PATCH] sched: remove resetting exec_start in put_prev_task_rt()

On Thu, Jun 9, 2011 at 8:34 PM, Steven Rostedt <rostedt@...dmis.org> wrote:
> On Thu, 2011-06-02 at 16:04 +0800, Yong Zhang wrote:
>> On Wed, Jun 1, 2011 at 10:03 PM, Hillf Danton <dhillf@...il.com> wrote:
>> > Resetting exec_start, after updated in update_curr_rt(), could open window for
>> > messing up the subsequent computations of delta_exec of the given task.
>>
>> I can't see how could this happen. what kind of 'subsequent computations'
>> do you mean?
>
> I still don't see a race.
>
> Hilf, if you still believe there's a race here, can you explain it in
> detail. Do something like:
>
>        CPU0                    CPU1
>        ----                    ----
>   do_something;
>                                does_something_not_expected;
>   continue_something;
>
> Obviously changing what those "somethings" are. That way we can visually
> see what you are trying to say.
>
>>
>> But because exec_start will be reset by _pick_next_task_rt()/set_curr_task_rt(),
>> you patch is ok. IMHO it is not critical, it's just cleanup instead.
>
> I disagree. Yes the exec_start is reset there, but I like the fact that
> it's 0 when not running.
>

Hi Steve

Thank you a lot, and Peter as well, for your lessons on mutex_spin_on_owner:)

Resetting exec_start to zero has no negative functionality in RT scheduling,
as shown by Yong.

After put_prev_task() is called in schedule(),

	put_prev_task(rq, prev);
	next = pick_next_task(rq);
	clear_tsk_need_resched(prev);

next is picked. Lets assume that next is not prev, and prev is still on RQ,
then prev's sched_class is changed to CFS while it is waiting on RQ.
After sched_class switch, prev is under CFS charge, and the exec_start field
could be taken into other games.

In task_hot(), called when migrating task, zeroed exec_start is trapped as
the following.

btw, I could not locate where proc_sched_show_task() is called.

Again thanks all a lot, /Hillf

---
 kernel/sched.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index fd18f39..195bd4a 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2184,6 +2184,7 @@ task_hot(struct task_struct *p, u64 now, struct
sched_domain *sd)
 	if (sysctl_sched_migration_cost == 0)
 		return 0;

+	BUG_ON(p->se.exec_start == 0);
 	delta = now - p->se.exec_start;

 	return delta < (s64)sysctl_sched_migration_cost;
--
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/