Re: [PATCH 4/4] locking: Introduce smp_cond_acquire()

From	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH 4/4] locking: Introduce smp_cond_acquire()
Date	2015-11-13 00:50 +0100
Message-ID	<qu9MS-79j-21@gated-at.bofh.it> (permalink)
References	(5 earlier) <qu4at-3v0-11@gated-at.bofh.it> <qu4tQ-3RE-29@gated-at.bofh.it> <qu4WT-42L-43@gated-at.bofh.it> <qu5ge-49Z-17@gated-at.bofh.it> <qu7L3-5SX-13@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Thu, Nov 12, 2015 at 09:33:39PM +0000, Will Deacon wrote:
> On Thu, Nov 12, 2015 at 10:59:06AM -0800, Paul E. McKenney wrote:
> > On Thu, Nov 12, 2015 at 08:33:02PM +0100, Oleg Nesterov wrote:
> > > On 11/12, Peter Zijlstra wrote:
> > > >
> > > > On Thu, Nov 12, 2015 at 07:38:07PM +0100, Oleg Nesterov wrote:
> > > > > It seems that PPC needs to define smp_mb__before_spinlock() as full mb(),
> > > > > no?
> > > >
> > > > It does:
> > > >
> > > > arch/powerpc/include/asm/barrier.h:#define smp_mb__before_spinlock() smp_mb()
> > > 
> > > Ah, indeed, thanks.
> > > 
> > > And given that it also defines smp_mb__after_unlock_lock() as smp_mb(),
> > > I am starting to understand how it can help to avoid the races with
> > > spin_unlock_wait() in (for example) do_exit().
> > > 
> > > But as Boqun has already mentioned, this means that mb__after_unlock_lock()
> > > has the new meaning which should be documented.
> > > 
> > > Hmm. And 12d560f4 "Privatize smp_mb__after_unlock_lock()" should be reverted
> > > then ;)
> > 
> > Surprisingly, this reverts cleanly against today's mainline, please see
> > the patch below.  Against my -rcu stack, not so much, but so it goes.  ;-)
> 
> I think we ended up concluding that smp_mb__after_unlock_lock is indeed
> required, but I don't think we should just resurrect the old definition,
> which doesn't keep UNLOCK -> LOCK distinct from RELEASE -> ACQUIRE. I'm
> still working on documenting the different types of transitivity that we
> identified in that thread, but it's slow going.
> 
> Also, as far as spin_unlock_wait is concerned, it is neither a LOCK or
> an UNLOCK and this barrier doesn't offer us anything. Sure, it might
> work because PPC defines it as smp_mb(), but it doesn't help on arm64
> and defining the macro is overkill for us in most places (i.e. RCU).
> 
> If we decide that the current usage of spin_unlock_wait is valid, then I
> would much rather implement a version of it in the arm64 backend that
> does something like:
> 
>  1:  ldrex r1, [&lock]
>      if r1 indicates that lock is taken, branch back to 1b
>      strex r1, [&lock]
>      if store failed, branch back to 1b
> 
> i.e. we don't just test the lock, but we also write it back atomically
> if we discover that it's free. That would then clear the exclusive monitor
> on any cores in the process of taking the lock and restore the ordering
> that we need.

We could clearly do something similar in PowerPC, but I suspect that this
would hurt really badly on large systems, given that there are PowerPC
systems with more than a thousand hardware threads.  So one approach
is ARM makes spin_unlock_wait() do the write, similar to spin_lock();
spin_lock(), but PowerPC relies on smp_mb__after_unlock_lock().

Or does someone have a better proposal?

							Thanx, Paul

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Thread

[PATCH 0/4] scheduler ordering bits Peter Zijlstra <peterz@infradead.org> - 2015-11-02 15:00 +0100
  [PATCH 2/4] sched: Document Program-Order guarantees Peter Zijlstra <peterz@infradead.org> - 2015-11-02 15:00 +0100
    Re: [PATCH 2/4] sched: Document Program-Order guarantees Paul Turner <pjt@google.com> - 2015-11-02 21:30 +0100
      Re: [PATCH 2/4] sched: Document Program-Order guarantees Peter Zijlstra <peterz@infradead.org> - 2015-11-02 21:40 +0100
        Re: [PATCH 2/4] sched: Document Program-Order guarantees Paul Turner <pjt@google.com> - 2015-11-02 23:10 +0100
          Re: [PATCH 2/4] sched: Document Program-Order guarantees Peter Zijlstra <peterz@infradead.org> - 2015-11-02 23:20 +0100
  [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-02 15:00 +0100
    Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-02 15:00 +0100
      Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Will Deacon <will.deacon@arm.com> - 2015-11-02 18:50 +0100
        Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-03 02:20 +0100
          Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Linus Torvalds <torvalds@linux-foundation.org> - 2015-11-03 02:30 +0100
    Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Will Deacon <will.deacon@arm.com> - 2015-11-02 18:50 +0100
    Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Linus Torvalds <torvalds@linux-foundation.org> - 2015-11-02 19:10 +0100
      Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Will Deacon <will.deacon@arm.com> - 2015-11-02 19:40 +0100
        Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Linus Torvalds <torvalds@linux-foundation.org> - 2015-11-02 20:20 +0100
          Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Will Deacon <will.deacon@arm.com> - 2015-11-02 21:00 +0100
            Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-02 21:30 +0100
          Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-02 23:00 +0100
          Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-03 03:00 +0100
            Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Linus Torvalds <torvalds@linux-foundation.org> - 2015-11-03 20:50 +0100
              Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-04 05:00 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Linus Torvalds <torvalds@linux-foundation.org> - 2015-11-04 05:50 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-04 14:10 +0100
      Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() David Howells <dhowells@redhat.com> - 2015-11-02 21:40 +0100
        Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-02 21:50 +0100
        Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Linus Torvalds <torvalds@linux-foundation.org> - 2015-11-02 22:20 +0100
    Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Oleg Nesterov <oleg@redhat.com> - 2015-11-03 18:10 +0100
      Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-03 19:30 +0100
      Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Boqun Feng <boqun.feng@gmail.com> - 2015-11-11 10:50 +0100
        Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Boqun Feng <boqun.feng@gmail.com> - 2015-11-11 11:40 +0100
          Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Oleg Nesterov <oleg@redhat.com> - 2015-11-11 20:00 +0100
          Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-12 15:00 +0100
        Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-11 13:20 +0100
          Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Oleg Nesterov <oleg@redhat.com> - 2015-11-11 19:50 +0100
            Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Linus Torvalds <torvalds@linux-foundation.org> - 2015-11-11 22:30 +0100
            Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Boqun Feng <boqun.feng@gmail.com> - 2015-11-12 08:20 +0100
              Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-12 11:30 +0100
              Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Oleg Nesterov <oleg@redhat.com> - 2015-11-12 15:10 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Boqun Feng <boqun.feng@gmail.com> - 2015-11-12 15:50 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-12 16:50 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Will Deacon <will.deacon@arm.com> - 2015-11-12 23:00 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-12 15:50 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-12 16:00 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-12 16:10 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-12 16:10 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-12 16:30 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Will Deacon <will.deacon@arm.com> - 2015-11-12 22:30 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Boqun Feng <boqun.feng@gmail.com> - 2015-11-12 16:20 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Oleg Nesterov <oleg@redhat.com> - 2015-11-12 18:50 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Peter Zijlstra <peterz@infradead.org> - 2015-11-12 19:10 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Oleg Nesterov <oleg@redhat.com> - 2015-11-12 19:40 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-12 20:00 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Will Deacon <will.deacon@arm.com> - 2015-11-12 22:40 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2015-11-13 00:50 +0100
              Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Linus Torvalds <torvalds@linux-foundation.org> - 2015-11-12 19:30 +0100
                Re: [PATCH 4/4] locking: Introduce smp_cond_acquire() Will Deacon <will.deacon@arm.com> - 2015-11-12 23:10 +0100
  [PATCH 1/4] sched: Better document the try_to_wake_up() barriers Peter Zijlstra <peterz@infradead.org> - 2015-11-02 15:00 +0100
  [PATCH 3/4] sched: Fix a race in try_to_wake_up() vs schedule() Peter Zijlstra <peterz@infradead.org> - 2015-11-02 15:00 +0100

csiph-web