netdev - Re: [PATCH net-next] net: sched: Avoid using yield() in a busy waiting loop

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

Date:   Sat, 12 Oct 2019 22:14:17 +0300
From:   Sergei Shtylyov <sergei.shtylyov@...entembedded.com>
To:     Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        netdev@...r.kernel.org
Cc:     Jamal Hadi Salim <jhs@...atatu.com>,
        Cong Wang <xiyou.wangcong@...il.com>,
        Jiri Pirko <jiri@...nulli.us>,
        Eric Dumazet <edumazet@...gle.com>, tglx@...utronix.de,
        Marc Kleine-Budde <mkl@...gutronix.de>,
        Peter Zijlstra <peterz@...radead.org>,
        "David S. Miller" <davem@...emloft.net>
Subject: Re: [PATCH net-next] net: sched: Avoid using yield() in a busy
 waiting loop

Hello!

On 10/11/2019 08:15 PM, Sebastian Andrzej Siewior wrote:

> From: Marc Kleine-Budde <mkl@...gutronix.de>
> 
> With threaded interrupts enabled, the interrupt thread runs as SCHED_RR
> with priority 50. If a user application with a higher priority preempts
> the interrupt thread and tries to shutdown the network interface then it
> will loop forever. The kernel will spin in the loop waiting for the
> device to become idle and the scheduler will never consider the
> interrupt thread because its priority is lower.
> 
> Avoid the problem by using by sleeping for a jiffy giving other tasks,

   So "using" or "sleeping"? :-)

> including the interrupt thread, a chance to run and make progress.
> 
> In the original thread it has been suggested to use wait_event() and
> properly waiting for the state to occur. DaveM explained that this would
> require to add expensive checks in the fast paths of packet processing.
> 
> Link: https://lkml.kernel.org/r/1393976987-23555-1-git-send-email-mkl@pengutronix.de
> Signed-off-by: Marc Kleine-Budde <mkl@...gutronix.de>
> [bigeasy: Rewrite commit message, add comment, use
>           schedule_timeout_uninterruptible()]
> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
> ---
> 
> The old thread also pointed anoth yield() loop which was resolved by
> commit
>    845704a535e9b ("tcp: avoid looping in tcp_send_fin()")
> 
>  net/sched/sch_generic.c | 9 +++++++--
>  1 file changed, 7 insertions(+), 2 deletions(-)
> 
> diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
> index 17bd8f539bc7f..b27574f2c6b47 100644
> --- a/net/sched/sch_generic.c
> +++ b/net/sched/sch_generic.c
> @@ -1217,8 +1217,13 @@ void dev_deactivate_many(struct list_head *head)
>  
>  	/* Wait for outstanding qdisc_run calls. */
>  	list_for_each_entry(dev, head, close_list) {
> -		while (some_qdisc_is_busy(dev))
> -			yield();
> +		while (some_qdisc_is_busy(dev)) {
> +			/* wait_event() would avoid this sleep-loop but would
> +			 * require expesive checks in the fast paths of packet

   Expensive?

> +			 * processing which isn't worth it.
> +			 */
> +			schedule_timeout_uninterruptible(1);
> +		}
>  		/* The new qdisc is assigned at this point so we can safely
>  		 * unwind stale skb lists and qdisc statistics
>  		 */

MBR, Sergei