netdev - Re: Queue with wait-free enqueue, blocking dequeue, splice

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Tue, 21 Oct 2014 12:02:38 +0000 (UTC)
From:	Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
To:	Jesper Dangaard Brouer <brouer@...hat.com>
Cc:	"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
	netdev@...r.kernel.org, Jamal Hadi Salim <jhs@...atatu.com>
Subject: Re: Queue with wait-free enqueue, blocking dequeue, splice



----- Original Message -----
> From: "Mathieu Desnoyers" <mathieu.desnoyers@...icios.com>
> To: "Jesper Dangaard Brouer" <brouer@...hat.com>
> Cc: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>, netdev@...r.kernel.org, "Jamal Hadi Salim" <jhs@...atatu.com>
> Sent: Monday, October 20, 2014 8:04:10 PM
> Subject: Re: Queue with wait-free enqueue, blocking dequeue, splice
> 
> ----- Original Message -----
> > From: "Jesper Dangaard Brouer" <brouer@...hat.com>
> > To: "Mathieu Desnoyers" <mathieu.desnoyers@...icios.com>
> > Cc: brouer@...hat.com, "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
> > netdev@...r.kernel.org, "Jamal Hadi Salim"
> > <jhs@...atatu.com>
> > Sent: Monday, October 20, 2014 10:02:37 AM
> > Subject: Re: Queue with wait-free enqueue, blocking dequeue, splice
> > 

[...]

> > 
> > AFAIK your queue implementation is a CAS-based, Wait-Free on enqueue,
> > but Lock-Free on dequeue with the potential for waiting/blocking on
> > a enqueue processes.
> >  I'm not 100% sure, that we want this behavior for the qdisc system.
> 
> It's actually xchg-based (not CAS-based). It is indeed wait-free
> in the strictest sense of the term on enqueue (at least on x86,
> some other arch implement xchg using ll/sc, which is not strictly
> wait-free).
> 
> On dequeue, it can busy-wait for a short while that the enqueue
> completes. Note that in kernel, since we disable preemption during
> enqueue, the dequeue does not have to ever block, just busy-looping
> is OK, since the longest thing that could nest over the enqueue
> is possibly an interrupt and softirq. So yes, I guess the dequeue
> would qualify as lock-free.

Scratch this last part about dequeue lock-freedom: dequeue can
busy-wait endlessly long if an enqueue is, for instance, interrupted
by a dequeue operation that would sit within an interrupt handler.
Dequeue is therefore not "lock-free". It is not even obstruction-free.

This just means that you need to be aware of this if you use dequeue
in an execution context that can interrupt enqueue.

Thanks,

Mathieu

-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html