From: Steven Rostedt <rostedt@goodmis.org>
To: Peter Zijlstra <peterz@infradead.org>
Cc: mingo@kernel.org, tglx@linutronix.de, juri.lelli@arm.com,
xlpang@redhat.com, bigeasy@linutronix.de,
linux-kernel@vger.kernel.org, mathieu.desnoyers@efficios.com,
jdesfossez@efficios.com, bristot@redhat.com
Subject: Re: [RFC][PATCH 4/4] futex: Rewrite FUTEX_UNLOCK_PI
Date: Mon, 3 Oct 2016 11:36:24 -0400 [thread overview]
Message-ID: <20161003113624.04f1f9f2@gandalf.local.home> (raw)
In-Reply-To: <20161003091847.704255067@infradead.org>
On Mon, 03 Oct 2016 11:12:38 +0200
Peter Zijlstra <peterz@infradead.org> wrote:
> There's a number of 'interesting' problems with FUTEX_UNLOCK_PI, all
> caused by holding hb->lock while doing the rt_mutex_unlock()
> equivalient.
>
> This patch doesn't attempt to fix any of the actual problems, but
> instead reworks the code to not hold hb->lock across the unlock,
> paving the way to actually fix the problems later.
>
> The current reason we hold hb->lock over unlock is that it serializes
> against FUTEX_LOCK_PI and avoids new waiters from coming in, this then
> ensures the rt_mutex_next_owner() value is stable and can be written
> into the user-space futex value before doing the unlock. Such that the
> unlock will indeed end up at new_owner.
>
> This patch recognises that holding rt_mutex::wait_lock results in the
> very same guarantee, no new waiters can come in while we hold that
> lock -- after all, waiters would need this lock to queue themselves.
>
> It therefore restructures the code to keep rt_mutex::wait_lock held.
>
> This (of course) is not entirely straight forward either, see the
> comment in rt_mutex_slowunlock(), doing the unlock itself might drop
> wait_lock, letting new waiters in. To cure this
> rt_mutex_futex_unlock() becomes a variant of rt_mutex_slowunlock()
> that return -EAGAIN instead. This ensures the FUTEX_UNLOCK_PI code
> aborts and restarts the entire operation.
>
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> ---
> kernel/futex.c | 63 +++++++++++++++++--------------
> kernel/locking/rtmutex.c | 81 ++++++++++++++++++++++++++++++++++++----
> kernel/locking/rtmutex_common.h | 4 -
> 3 files changed, 110 insertions(+), 38 deletions(-)
>
> --- a/kernel/futex.c
> +++ b/kernel/futex.c
> @@ -1294,24 +1294,21 @@ static void mark_wake_futex(struct wake_
> static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *top_waiter,
> struct futex_hash_bucket *hb)
> {
> - struct task_struct *new_owner;
> struct futex_pi_state *pi_state = top_waiter->pi_state;
> u32 uninitialized_var(curval), newval;
> + struct task_struct *new_owner;
> WAKE_Q(wake_q);
> - bool deboost;
> int ret = 0;
>
> - if (!pi_state)
> - return -EINVAL;
> + raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock);
>
> + WARN_ON_ONCE(!atomic_inc_not_zero(&pi_state->refcount));
Don't we have a rule where WARN_ON() and BUG_ON() should never have
"side effects"? That is, they should only check values, but their
contents should not update values.
hence have:
ret = atomic_inc_not_zero(&pi_state->refcount);
WARN_ON_ONCE(!ret);
> /*
> - * If current does not own the pi_state then the futex is
> - * inconsistent and user space fiddled with the futex value.
> + * Now that we hold wait_lock, no new waiters can happen on the
> + * rt_mutex and new owner is stable. Drop hb->lock.
> */
> - if (pi_state->owner != current)
> - return -EINVAL;
> + spin_unlock(&hb->lock);
>
Also, as Sebastian has said before, I believe this breaks rt's migrate
disable code. As migrate disable and migrate_enable are a nop if
preemption is disabled, thus if you hold a raw_spin_lock across a
spin_unlock() when the migrate enable will be a nop, and the
migrate_disable() will never stop.
-- Steve
> - raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock);
> new_owner = rt_mutex_next_owner(&pi_state->pi_mutex);
>
> /*
next prev parent reply other threads:[~2016-10-03 15:36 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-10-03 9:12 [RFC][PATCH 0/4] FUTEX_UNLOCK_PI wobbles Peter Zijlstra
2016-10-03 9:12 ` [RFC][PATCH 1/4] futex: Cleanup variable names for futex_top_waiter() Peter Zijlstra
2016-10-03 14:15 ` Steven Rostedt
2016-10-05 3:58 ` Davidlohr Bueso
2016-10-03 9:12 ` [RFC][PATCH 2/4] futex: Use smp_store_release() in mark_wake_futex() Peter Zijlstra
2016-10-03 14:19 ` Steven Rostedt
2016-10-05 3:57 ` Davidlohr Bueso
2016-10-05 6:20 ` Peter Zijlstra
2016-10-03 9:12 ` [RFC][PATCH 3/4] futex: Remove rt_mutex_deadlock_account_*() Peter Zijlstra
2016-10-03 9:34 ` Peter Zijlstra
2016-10-03 14:25 ` Steven Rostedt
2016-10-05 1:08 ` Davidlohr Bueso
2016-10-05 7:29 ` Sebastian Andrzej Siewior
2016-10-03 9:12 ` [RFC][PATCH 4/4] futex: Rewrite FUTEX_UNLOCK_PI Peter Zijlstra
2016-10-03 15:36 ` Steven Rostedt [this message]
2016-10-03 15:44 ` Peter Zijlstra
2016-10-03 15:45 ` Peter Zijlstra
2016-10-03 16:23 ` Steven Rostedt
2016-10-05 7:41 ` Sebastian Andrzej Siewior
2016-10-05 8:09 ` Peter Zijlstra
2016-10-05 8:21 ` Sebastian Andrzej Siewior
2016-10-05 8:32 ` Peter Zijlstra
2016-10-06 10:29 ` Peter Zijlstra
2016-10-07 11:21 ` Peter Zijlstra
2016-10-08 15:53 ` Thomas Gleixner
2016-10-08 16:55 ` Peter Zijlstra
2016-10-08 17:06 ` Thomas Gleixner
2016-10-10 10:17 ` Thomas Gleixner
2016-10-10 11:40 ` Peter Zijlstra
2016-10-21 12:27 ` Peter Zijlstra
2016-10-27 20:36 ` Thomas Gleixner
2016-11-23 19:20 ` Peter Zijlstra
2016-11-24 16:52 ` Peter Zijlstra
2016-11-24 17:56 ` Thomas Gleixner
2016-11-24 18:58 ` Peter Zijlstra
2016-11-25 9:23 ` Peter Zijlstra
2016-11-25 10:03 ` Peter Zijlstra
2016-11-25 19:13 ` Thomas Gleixner
2016-11-25 14:09 ` Peter Zijlstra
2016-10-08 18:22 ` Thomas Gleixner
2016-10-09 11:17 ` Thomas Gleixner
2016-10-10 14:06 ` Peter Zijlstra
2016-10-05 1:02 ` [RFC][PATCH 0/4] FUTEX_UNLOCK_PI wobbles Davidlohr Bueso
2016-10-05 6:20 ` Peter Zijlstra
2016-10-05 7:26 ` Sebastian Andrzej Siewior
2016-10-05 16:04 ` Davidlohr Bueso
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=20161003113624.04f1f9f2@gandalf.local.home \
--to=rostedt@goodmis.org \
--cc=bigeasy@linutronix.de \
--cc=bristot@redhat.com \
--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=peterz@infradead.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
all inboxes | Powered by JetHome®