linux-kernel - Re: [PATCHv3 2/6] tty/ldsem: Update waiter->task before waking up reader

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]

Message-ID: <20180911050449.GB4065@jagdpanzerIV>
Date:   Tue, 11 Sep 2018 14:04:49 +0900
From:   Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>
To:     Dmitry Safonov <dima@...sta.com>
Cc:     linux-kernel@...r.kernel.org,
        Dmitry Safonov <0x7f454c46@...il.com>,
        Daniel Axtens <dja@...ens.net>,
        Dmitry Vyukov <dvyukov@...gle.com>,
        Michael Neuling <mikey@...ling.org>,
        Mikulas Patocka <mpatocka@...hat.com>,
        Nathan March <nathan@...net>,
        Pasi Kärkkäinen <pasik@....fi>,
        Peter Hurley <peter@...leysoftware.com>,
        "Rong, Chen" <rong.a.chen@...el.com>,
        Sergey Senozhatsky <sergey.senozhatsky.work@...il.com>,
        Tan Xiaojun <tanxiaojun@...wei.com>,
        Tetsuo Handa <penguin-kernel@...ove.SAKURA.ne.jp>,
        stable@...r.kernel.org,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Jiri Slaby <jslaby@...e.com>,
        Peter Zijlstra <peterz@...radead.org>,
        "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
Subject: Re: [PATCHv3 2/6] tty/ldsem: Update waiter->task before waking up
 reader

On (09/11/18 02:48), Dmitry Safonov wrote:
> There is a couple of reports about lockup in ldsem_down_read() without
> anyone holding write end of ldisc semaphore:
> lkml.kernel.org/r/<20171121132855.ajdv4k6swzhvktl6@...-t540p.sh.intel.com>
> lkml.kernel.org/r/<20180907045041.GF1110@...o2-debian>
> 
> They all looked like a missed wake up.
> I wasn't lucky enough to reproduce it, but it seems like reader on
> another CPU can miss waiter->task update and schedule again, resulting
> in indefinite (MAX_SCHEDULE_TIMEOUT) sleep.

Certainly, something suspicious is going on.

> @@ -118,6 +118,8 @@ static void __ldsem_wake_readers(struct ld_semaphore *sem)
>  		tsk = waiter->task;
>  		smp_mb();
>  		waiter->task = NULL;
> +		/* Make sure down_read_failed() will see !waiter->task update */
> +		smp_wmb();
>  		wake_up_process(tsk);

Hmm. I think wake_up_process() executes a full memory barrier, because
it accesses task state.

>  		put_task_struct(tsk);
>  	}
> @@ -217,7 +219,7 @@ down_read_failed(struct ld_semaphore *sem, long count, long timeout)
>  	for (;;) {
>  		set_current_state(TASK_UNINTERRUPTIBLE);

I think that set_current_state() also executes memory barrier. Just
because it accesses task state.

> -		if (!waiter.task)
> +		if (!READ_ONCE(waiter.task))
>  			break;
>  		if (!timeout)
>  			break;

	-ss