From: Juri Lelli <juri.lelli@arm.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: mingo@kernel.org, tglx@linutronix.de, rostedt@goodmis.org,
xlpang@redhat.com, linux-kernel@vger.kernel.org,
mathieu.desnoyers@efficios.com, jdesfossez@efficios.com,
bristot@redhat.com, Ingo Molnar <mingo@redhat.com>
Subject: Re: [RFC][PATCH 1/8] rtmutex: Deboost before waking up the top waiter
Date: Tue, 14 Jun 2016 10:09:34 +0100 [thread overview]
Message-ID: <20160614090934.GE5981@e106622-lin> (raw)
In-Reply-To: <20160607200215.637804442@infradead.org>
Hi,
I've got only nitpicks for the changelog. Otherwise the patch looks good
to me (and yes, without it bw inheritance would be a problem).
On 07/06/16 21:56, Peter Zijlstra wrote:
> From: Xunlei Pang <xlpang@redhat.com>
>
> We should deboost before waking the high-prio task, such that
> we don't run two tasks with the same "state"(priority, deadline,
^
space
> sched_class, etc) during the period between the end of wake_up_q()
> and the end of rt_mutex_adjust_prio().
>
> As "Peter Zijlstra" said:
> Its semantically icky to have the two tasks running off the same
s/Its/It's/
> state and practically icky when you consider bandwidth inheritance --
> where the boosted task wants to explicitly modify the state of the
> booster. In that latter case you really want to unboost before you
> let the booster run again.
>
> But this however can lead to prio-inversion if current would get
> preempted after the deboost but before waking our high-prio task,
> hence we disable preemption before doing deboost, and enabling it
s/enabling/re-enable/
> after the wake up is over.
>
> The patch fixed the logic, and introduced rt_mutex_postunlock()
s/The/This/
s/fixed/fixes/
s/introduced/introduces/
> to do some code refactor.
>
> Most importantly however; this change ensures pointer stability for
> the next patch, where we have rt_mutex_setprio() cache a pointer to
> the top-most waiter task. If we, as before this change, do the wakeup
> first and then deboost, this pointer might point into thin air.
>
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Juri Lelli <juri.lelli@arm.com>
> Suggested-by: Peter Zijlstra <peterz@infradead.org>
> [peterz: Changelog]
> Signed-off-by: Xunlei Pang <xlpang@redhat.com>
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> Link: http://lkml.kernel.org/r/1461659449-19497-1-git-send-email-xlpang@redhat.com
Do we have any specific tests for this set? I'm running mine.
Best,
- Juri
> ---
>
> kernel/futex.c | 5 ++---
> kernel/locking/rtmutex.c | 28 ++++++++++++++++++++++++----
> kernel/locking/rtmutex_common.h | 1 +
> 3 files changed, 27 insertions(+), 7 deletions(-)
>
> --- a/kernel/futex.c
> +++ b/kernel/futex.c
> @@ -1336,9 +1336,8 @@ static int wake_futex_pi(u32 __user *uad
> * scheduled away before the wake up can take place.
> */
> spin_unlock(&hb->lock);
> - wake_up_q(&wake_q);
> - if (deboost)
> - rt_mutex_adjust_prio(current);
> +
> + rt_mutex_postunlock(&wake_q, deboost);
>
> return 0;
> }
> --- a/kernel/locking/rtmutex.c
> +++ b/kernel/locking/rtmutex.c
> @@ -1390,12 +1390,32 @@ rt_mutex_fastunlock(struct rt_mutex *loc
> } else {
> bool deboost = slowfn(lock, &wake_q);
>
> - wake_up_q(&wake_q);
> + rt_mutex_postunlock(&wake_q, deboost);
> + }
> +}
> +
>
> - /* Undo pi boosting if necessary: */
> - if (deboost)
> - rt_mutex_adjust_prio(current);
> +/*
> + * Undo pi boosting (if necessary) and wake top waiter.
> + */
> +void rt_mutex_postunlock(struct wake_q_head *wake_q, bool deboost)
> +{
> + /*
> + * We should deboost before waking the top waiter task such that
> + * we don't run two tasks with the 'same' priority. This however
> + * can lead to prio-inversion if we would get preempted after
> + * the deboost but before waking our high-prio task, hence the
> + * preempt_disable.
> + */
> + if (deboost) {
> + preempt_disable();
> + rt_mutex_adjust_prio(current);
> }
> +
> + wake_up_q(wake_q);
> +
> + if (deboost)
> + preempt_enable();
> }
>
> /**
> --- a/kernel/locking/rtmutex_common.h
> +++ b/kernel/locking/rtmutex_common.h
> @@ -111,6 +111,7 @@ extern int rt_mutex_finish_proxy_lock(st
> extern int rt_mutex_timed_futex_lock(struct rt_mutex *l, struct hrtimer_sleeper *to);
> extern bool rt_mutex_futex_unlock(struct rt_mutex *lock,
> struct wake_q_head *wqh);
> +extern void rt_mutex_postunlock(struct wake_q_head *wake_q, bool deboost);
> extern void rt_mutex_adjust_prio(struct task_struct *task);
>
> #ifdef CONFIG_DEBUG_RT_MUTEXES
>
>
next prev parent reply other threads:[~2016-06-14 9:09 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-07 19:56 [RFC][PATCH 0/8] PI and assorted failings Peter Zijlstra
2016-06-07 19:56 ` [RFC][PATCH 1/8] rtmutex: Deboost before waking up the top waiter Peter Zijlstra
2016-06-14 9:09 ` Juri Lelli [this message]
2016-06-14 12:54 ` Peter Zijlstra
2016-06-14 13:20 ` Juri Lelli
2016-06-14 13:59 ` Peter Zijlstra
2016-06-14 16:36 ` Davidlohr Bueso
2016-06-14 17:01 ` Juri Lelli
2016-06-14 18:22 ` Steven Rostedt
2016-06-07 19:56 ` [RFC][PATCH 2/8] sched/rtmutex/deadline: Fix a PI crash for deadline tasks Peter Zijlstra
2016-06-14 10:21 ` Juri Lelli
2016-06-14 12:30 ` Peter Zijlstra
2016-06-14 12:53 ` Xunlei Pang
2016-06-14 13:07 ` Juri Lelli
2016-06-14 16:39 ` Juri Lelli
2016-06-14 18:42 ` Steven Rostedt
2016-06-14 20:28 ` Peter Zijlstra
2016-06-15 16:14 ` Steven Rostedt
2016-06-07 19:56 ` [RFC][PATCH 3/8] sched/deadline/rtmutex: Dont miss the dl_runtime/dl_period update Peter Zijlstra
2016-06-14 10:43 ` Juri Lelli
2016-06-14 12:09 ` Peter Zijlstra
2016-06-15 16:30 ` Steven Rostedt
2016-06-15 17:55 ` Peter Zijlstra
2016-06-07 19:56 ` [RFC][PATCH 4/8] rtmutex: Remove rt_mutex_fastunlock() Peter Zijlstra
2016-06-15 16:43 ` Steven Rostedt
2016-06-07 19:56 ` [RFC][PATCH 5/8] rtmutex: Clean up Peter Zijlstra
2016-06-14 12:08 ` Juri Lelli
2016-06-14 12:32 ` Peter Zijlstra
2016-06-14 12:41 ` Juri Lelli
2016-06-07 19:56 ` [RFC][PATCH 6/8] sched/rtmutex: Refactor rt_mutex_setprio() Peter Zijlstra
2016-06-14 13:14 ` Juri Lelli
2016-06-14 14:08 ` Peter Zijlstra
2016-06-07 19:56 ` [RFC][PATCH 7/8] sched,tracing: Update trace_sched_pi_setprio() Peter Zijlstra
2016-06-07 19:56 ` [RFC][PATCH 8/8] rtmutex: Fix PI chain order integrity Peter Zijlstra
2016-06-14 17:39 ` Juri Lelli
2016-06-14 19:44 ` Peter Zijlstra
2016-06-15 7:25 ` Juri Lelli
2016-06-27 12:23 ` Peter Zijlstra
2016-06-27 12:40 ` Thomas Gleixner
2016-06-28 9:05 ` Juri Lelli
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160614090934.GE5981@e106622-lin \
--to=juri.lelli@arm.com \
--cc=bristot@redhat.com \
--cc=jdesfossez@efficios.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=xlpang@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).