[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Thu, 6 Jun 2019 06:51:49 -0700
From: "Paul E. McKenney" <paulmck@...ux.ibm.com>
To: Herbert Xu <herbert@...dor.apana.org.au>
Cc: rcu@...r.kernel.org, linux-kernel@...r.kernel.org,
mingo@...nel.org, jiangshanlai@...il.com, dipankar@...ibm.com,
akpm@...ux-foundation.org, mathieu.desnoyers@...icios.com,
josh@...htriplett.org, tglx@...utronix.de, peterz@...radead.org,
rostedt@...dmis.org, dhowells@...hat.com, edumazet@...gle.com,
fweisbec@...il.com, oleg@...hat.com, joel@...lfernandes.org,
torvalds@...ux-foundation.org
Subject: Re: [PATCH RFC tip/core/rcu] Restore barrier() to rcu_read_lock()
and rcu_read_unlock()
On Thu, Jun 06, 2019 at 09:42:33PM +0800, Herbert Xu wrote:
> On Thu, Jun 06, 2019 at 06:19:33AM -0700, Paul E. McKenney wrote:
> > Commit bb73c52bad36 ("rcu: Don't disable preemption for Tiny and Tree
> > RCU readers") removed the barrier() calls from rcu_read_lock() and
> > rcu_write_lock() in CONFIG_PREEMPT=n&&CONFIG_PREEMPT_COUNT=n kernels.
> > Within RCU, this commit was OK, but it failed to account for things like
> > get_user() that can pagefault and that can be reordered by the compiler.
> > Lack of the barrier() calls in rcu_read_lock() and rcu_read_unlock()
> > can cause these page faults to migrate into RCU read-side critical
> > sections, which in CONFIG_PREEMPT=n kernels could result in too-short
> > grace periods and arbitrary misbehavior. Please see commit 386afc91144b
> > ("spinlocks and preemption points need to be at least compiler barriers")
> > for more details.
> >
> > This commit therefore restores the barrier() call to both rcu_read_lock()
> > and rcu_read_unlock(). It also removes them from places in the RCU update
> > machinery that used to need compensatory barrier() calls, effectively
> > reverting commit bb73c52bad36 ("rcu: Don't disable preemption for Tiny
> > and Tree RCU readers").
> >
> > Reported-by: Herbert Xu <herbert@...dor.apana.org.au>
> > Reported-by: Linus Torvalds <torvalds@...ux-foundation.org>
> > Signed-off-by: Paul E. McKenney <paulmck@...ux.ibm.com>
>
> Paul, Linus has already commited his patch:
>
> commit 66be4e66a7f422128748e3c3ef6ee72b20a6197b
> Author: Linus Torvalds <torvalds@...ux-foundation.org>
> Date: Mon Jun 3 13:26:20 2019 -0700
>
> rcu: locking and unlocking need to always be at least barriers
>
> So you'll need to rebase this.
Thank you for letting me know. Easy enough to do, just remove those
two hunks from my patch.
Thanx, Paul
Powered by blists - more mailing lists