Re: [PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another memory barrier

From	Manfred Spraul <manfred@colorfullife.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another memory barrier
Date	2016-09-02 08:50 +0200
Message-ID	<scQsx-4p5-1@gated-at.bofh.it> (permalink)
References	<scC6e-3Bv-23@gated-at.bofh.it> <scCfT-3EJ-25@gated-at.bofh.it> <scDlE-4lf-45@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On 09/01/2016 06:41 PM, Peter Zijlstra wrote:
> On Thu, Sep 01, 2016 at 04:30:39PM +0100, Will Deacon wrote:
>> On Thu, Sep 01, 2016 at 05:27:52PM +0200, Manfred Spraul wrote:
>>> Since spin_unlock_wait() is defined as equivalent to spin_lock();
>>> spin_unlock(), the memory barrier before spin_unlock_wait() is
>>> also not required.
> Note that ACQUIRE+RELEASE isn't a barrier.
>
> Both are semi-permeable and things can cross in the middle, like:
>
>
> 	x = 1;
> 	LOCK
> 	UNLOCK
> 	r = y;
>
> can (validly) get re-ordered like:
>
> 	LOCK
> 	r = y;
> 	x = 1;
> 	UNLOCK
>
> So if you want things ordered, as I think you do, I think the smp_mb()
> is still needed.
CPU1:
x=1; /* without WRITE_ONCE */
LOCK(l);
UNLOCK(l);
<do_semop>
smp_store_release(x,0)


CPU2;
LOCK(l)
if (smp_load_acquire(x)==1) goto slow_path
<do_semop>
UNLOCK(l)

Ordering is enforced because both CPUs access the same lock.

x=1 can't be reordered past the UNLOCK(l), I don't see that further 
guarantees are necessary.

Correct?

--
     Manfred

Thread

[PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another memory barrier Manfred Spraul <manfred@colorfullife.com> - 2016-09-01 17:30 +0200
  Re: [PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another  memory barrier Will Deacon <will.deacon@arm.com> - 2016-09-01 17:40 +0200
    Re: [PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another  memory barrier Peter Zijlstra <peterz@infradead.org> - 2016-09-01 18:50 +0200
      Re: [PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another  memory barrier Boqun Feng <boqun.feng@gmail.com> - 2016-09-02 08:20 +0200
      Re: [PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another  memory barrier Manfred Spraul <manfred@colorfullife.com> - 2016-09-02 08:50 +0200
        Re: [PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another  memory barrier Peter Zijlstra <peterz@infradead.org> - 2016-09-02 21:30 +0200
          Re: [PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another  memory barrier Manfred Spraul <manfred@colorfullife.com> - 2016-09-03 07:40 +0200
          Re: [PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another  memory barrier Manfred Spraul <manfred@colorfullife.com> - 2016-09-05 21:00 +0200
            Re: [PATCH 8/7] net/netfilter/nf_conntrack_core: Remove another  memory barrier Will Deacon <will.deacon@arm.com> - 2016-09-06 20:00 +0200

csiph-web