[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20160603133350.GW5231@linux.vnet.ibm.com>
Date: Fri, 3 Jun 2016 06:33:50 -0700
From: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Vineet Gupta <Vineet.Gupta1@...opsys.com>,
Waiman Long <waiman.long@....com>,
linux-kernel@...r.kernel.org, torvalds@...ux-foundation.org,
manfred@...orfullife.com, dave@...olabs.net, will.deacon@....com,
boqun.feng@...il.com, tj@...nel.org, pablo@...filter.org,
kaber@...sh.net, davem@...emloft.net, oleg@...hat.com,
netfilter-devel@...r.kernel.org, sasha.levin@...cle.com,
hofrat@...dl.org
Subject: Re: [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep
On Fri, Jun 03, 2016 at 02:27:52PM +0200, Peter Zijlstra wrote:
> On Fri, Jun 03, 2016 at 02:23:10PM +0200, Peter Zijlstra wrote:
> > On Fri, Jun 03, 2016 at 05:08:27AM -0700, Paul E. McKenney wrote:
> > > On Fri, Jun 03, 2016 at 11:38:34AM +0200, Peter Zijlstra wrote:
> > > > On Fri, Jun 03, 2016 at 02:48:38PM +0530, Vineet Gupta wrote:
> > > > > On Wednesday 25 May 2016 09:27 PM, Paul E. McKenney wrote:
> > > > > > For your example, but keeping the compiler in check:
> > > > > >
> > > > > > if (READ_ONCE(a))
> > > > > > WRITE_ONCE(b, 1);
> > > > > > smp_rmb();
> > > > > > WRITE_ONCE(c, 2);
> > > >
> > > > So I think it example is broken. The store to @c is not in fact
> > > > dependent on the condition of @a.
> > >
> > > At first glance, the compiler could pull the write to "c" above the
> > > conditional, but the "memory" constraint in smp_rmb() prevents this.
> > > From a hardware viewpoint, the write to "c" does depend on the "if",
> > > as the conditional branch does precede that write in execution order.
> > >
> > > But yes, this is using smp_rmb() in a very strange way, if that is
> > > what you are getting at.
> >
> > Well, the CPU could decide that the store to C happens either way around
> > the branch. I'm not sure I'd rely on CPUs not being _that_ clever.
> >
> > test %a, $0
> > jnz 1f
> > mov $1, %b
> > 1: mov $2, %c
> >
> > Its not too much to ask the CPU to look ahead 2 instructions to figure
> > out the store into c is going to happen unconditionally.
> >
> > I would really only rely on stores immediately dependent on the
> > conditional.
>
> Ah, interrupts could observe the difference, which is your execution
> order constraint. So yes, maybe.
>
> I'm still rather hesitant on this.
Any info on Itanium? Again, I am reasonably sure that Power and ARM
architectures guarantee to avoid excess cleverness on the part of the
CPU in this case.
Thanx, Paul
Powered by blists - more mailing lists