From: Peter Zijlstra <peterz@infradead.org>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: mingo@kernel.org, juri.lelli@arm.com, rostedt@goodmis.org,
xlpang@redhat.com, bigeasy@linutronix.de,
linux-kernel@vger.kernel.org, mathieu.desnoyers@efficios.com,
jdesfossez@efficios.com, bristot@redhat.com,
dvhart@infradead.org
Subject: Re: [PATCH -v5 14/14] futex: futex_unlock_pi() determinism
Date: Tue, 7 Mar 2017 18:59:23 +0100 [thread overview]
Message-ID: <20170307175923.GE3312@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <alpine.DEB.2.20.1703071526110.3584@nanos>
On Tue, Mar 07, 2017 at 03:31:50PM +0100, Thomas Gleixner wrote:
> On Sat, 4 Mar 2017, Peter Zijlstra wrote:
>
> > The problem with returning -EAGAIN when the waiter state mismatches is
> > that it becomes very hard to proof a bounded execution time on the
> > operation. And seeing that this is a RT operation, this is somewhat
> > important.
> >
> > While in practise it will be very unlikely to ever really take more
> > than one or two rounds, proving so becomes rather hard.
>
> Oh no. Assume the following:
>
> T1 and T2 are both pinned to CPU0. prio(T2) > prio(T1)
>
> CPU0
>
> T1
> lock_pi()
> queue_me() <- Waiter is visible
>
> preemption
>
> T2
> unlock_pi()
> loops with -EAGAIN forever
Ah! indeed.
> > Now that modifying wait_list is done while holding both hb->lock and
> > wait_lock, we can avoid the scenario entirely if we acquire wait_lock
> > while still holding hb-lock. Doing a hand-over, without leaving a
> > hole.
>
> > Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> > ---
> > kernel/futex.c | 26 ++++++++++++--------------
> > 1 file changed, 12 insertions(+), 14 deletions(-)
> >
> > --- a/kernel/futex.c
> > +++ b/kernel/futex.c
> > @@ -1391,16 +1391,11 @@ static int wake_futex_pi(u32 __user *uad
> > DEFINE_WAKE_Q(wake_q);
> > int ret = 0;
> >
> > new_owner = rt_mutex_next_owner(&pi_state->pi_mutex);
> > + if (WARN_ON_ONCE(!new_owner)) {
> > /*
> > + * Should be impossible now... but if weirdness happens,
>
> 'now...' is not very useful 6 month from NOW :)
I'll put in a reference to the below comment in, that explains why this
should now be impossible.
> > + * returning -EAGAIN is safe and correct.
> > */
> > ret = -EAGAIN;
> > goto out_unlock;
> > @@ -2770,15 +2765,18 @@ static int futex_unlock_pi(u32 __user *u
> > if (pi_state->owner != current)
> > goto out_unlock;
> >
> > + get_pi_state(pi_state);
> > /*
> > + * Since modifying the wait_list is done while holding both
> > + * hb->lock and wait_lock, holding either is sufficient to
> > + * observe it.
> > *
> > + * By taking wait_lock while still holding hb->lock, we ensure
> > + * there is no point where we hold neither; and therefore
> > + * wake_futex_pi() must observe a state consistent with what we
> > + * observed.
> > */
^^ that one.
> > + raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock);
> > spin_unlock(&hb->lock);
next prev parent reply other threads:[~2017-03-07 18:44 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-04 9:27 [PATCH -v5 00/14] the saga of FUTEX_UNLOCK_PI wobbles continues Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 01/14] futex: Fix potential use-after-free in FUTEX_REQUEUE_PI Peter Zijlstra
2017-03-14 20:48 ` [tip:locking/urgent] " tip-bot for Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 02/14] futex: Add missing error handling to FUTEX_REQUEUE_PI Peter Zijlstra
2017-03-14 20:49 ` [tip:locking/urgent] " tip-bot for Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 03/14] futex: Cleanup variable names for futex_top_waiter() Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 04/14] futex: Use smp_store_release() in mark_wake_futex() Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 05/14] futex: Remove rt_mutex_deadlock_account_*() Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 06/14] futex,rt_mutex: Provide futex specific rt_mutex API Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 07/14] futex: Change locking rules Peter Zijlstra
2017-03-07 13:22 ` Thomas Gleixner
2017-03-07 16:47 ` Sebastian Andrzej Siewior
2017-03-07 18:01 ` Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 08/14] futex: Cleanup refcounting Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 09/14] futex: Rework inconsistent rt_mutex/futex_q state Peter Zijlstra
2017-03-07 13:26 ` Thomas Gleixner
2017-03-04 9:27 ` [PATCH -v5 10/14] futex: Pull rt_mutex_futex_unlock() out from under hb->lock Peter Zijlstra
2017-03-07 14:08 ` Thomas Gleixner
2017-03-07 18:01 ` Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 11/14] futex,rt_mutex: Introduce rt_mutex_init_waiter() Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 12/14] futex,rt_mutex: Restructure rt_mutex_finish_proxy_lock() Peter Zijlstra
2017-03-07 14:18 ` Thomas Gleixner
2017-03-07 17:57 ` Peter Zijlstra
2017-03-07 17:59 ` Thomas Gleixner
2017-03-08 15:29 ` [PATCH] futex: move debug_rt_mutex_free_waiter() further down Sebastian Andrzej Siewior
2017-03-08 15:37 ` Sebastian Andrzej Siewior
2017-03-08 16:21 ` Steven Rostedt
2017-03-08 16:20 ` Steven Rostedt
2017-03-13 9:16 ` Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 13/14] futex: Rework futex_lock_pi() to use rt_mutex_*_proxy_lock() Peter Zijlstra
2017-03-04 9:27 ` [PATCH -v5 14/14] futex: futex_unlock_pi() determinism Peter Zijlstra
2017-03-07 14:31 ` Thomas Gleixner
2017-03-07 17:59 ` Peter Zijlstra [this message]
2017-03-13 9:25 ` Peter Zijlstra
2017-03-13 14:25 ` Thomas Gleixner
2017-03-13 15:11 ` Peter Zijlstra
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=20170307175923.GE3312@twins.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=bigeasy@linutronix.de \
--cc=bristot@redhat.com \
--cc=dvhart@infradead.org \
--cc=jdesfossez@efficios.com \
--cc=juri.lelli@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@kernel.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).