linux-kernel - Re: [RFC PATCH] sched/wait: Make interruptible exclusive waitqueue wakeups reliable

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20191209130005.GB5388@redhat.com>
Date:   Mon, 9 Dec 2019 14:00:06 +0100
From:   Oleg Nesterov <oleg@...hat.com>
To:     Ingo Molnar <mingo@...nel.org>
Cc:     Linus Torvalds <torvalds@...ux-foundation.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Juri Lelli <juri.lelli@...hat.com>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Miklos Szeredi <miklos@...redi.hu>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Felipe Balbi <balbi@...nel.org>
Subject: Re: [RFC PATCH] sched/wait: Make interruptible exclusive waitqueue
 wakeups reliable

On 12/09, Ingo Molnar wrote:
>
> Any consumed exclusive event is handled in finish_wait_exclusive() now:
>
> +               } else {
> +                       /* We got removed from the waitqueue already, wake up the next exclusive waiter (if any): */
> +                       if (interrupted && waitqueue_active(wq_head))
> +                               __wake_up_locked_key(wq_head, TASK_NORMAL, NULL);

See my previous email, I don't think we need this...

But if we do this, then __wake_up_locked_key(key => NULL) doesn't look right.
It should use the same "key" which was passed to __wake_up(key) which removed
us from list.

Currently this doesn't really matter, the only user of prepare_to_wait_event()
which relies on the "keyed" wakeup is ___wait_var_event() and it doesn't have
"exclusive" waiters, but still.

Hmm. and it seems that init_wait_var_entry() is buggy? Again, currently this
doesn't matter, but don't we need the trivial fix below?

Oleg.

--- x/kernel/sched/wait_bit.c
+++ x/kernel/sched/wait_bit.c
@@ -179,6 +179,7 @@ void init_wait_var_entry(struct wait_bit
 			.bit_nr = -1,
 		},
 		.wq_entry = {
+			.flags	 = flags,
 			.private = current,
 			.func	 = var_wake_function,
 			.entry	 = LIST_HEAD_INIT(wbq_entry->wq_entry.entry),