linux-kernel - Re: [PATCH] locking/ww_mutex: Initialize waiter.ww

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: <YR6hbiQLYYpadhkT@hirez.programming.kicks-ass.net>
Date:   Thu, 19 Aug 2021 20:22:38 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Sebastian Andrzej Siewior <bigeasy@...utronix.de>
Cc:     Guenter Roeck <linux@...ck-us.net>,
        Thomas Gleixner <tglx@...utronix.de>,
        LKML <linux-kernel@...r.kernel.org>,
        Ingo Molnar <mingo@...nel.org>,
        Juri Lelli <juri.lelli@...hat.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        Will Deacon <will@...nel.org>,
        Waiman Long <longman@...hat.com>,
        Boqun Feng <boqun.feng@...il.com>,
        Davidlohr Bueso <dave@...olabs.net>,
        Mike Galbraith <efault@....de>
Subject: Re: [PATCH] locking/ww_mutex: Initialize waiter.ww_ctx properly

On Thu, Aug 19, 2021 at 08:08:50PM +0200, Sebastian Andrzej Siewior wrote:
> The gathering of the debug code for the ww-mutex initialized moved the
> POISON initialiation into one spot and only set waiter.ww_ctx if the
> ww_ctx was non-NULL thus keeping the POISON value in ww-mutex case.
> 
> For ww-mutex without a context it is expected to set the context to
> NULL, the poison value was intended only for the regular mutex.
> 
> Move the waiter.ww_ctx where it was previously so it is initialized to
> NULL if no ww_ctx was passed.
> 
> Fixes: c0afb0ffc06e6 ("locking/ww_mutex: Gather mutex_waiter initialization")
> Reported-by: Guenter Roeck <linux@...ck-us.net>
> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
> ---
>  kernel/locking/mutex.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/kernel/locking/mutex.c b/kernel/locking/mutex.c
> index 05b68931622d1..d1068b0b163af 100644
> --- a/kernel/locking/mutex.c
> +++ b/kernel/locking/mutex.c
> @@ -614,8 +614,6 @@ __mutex_lock_common(struct mutex *lock, unsigned int state, unsigned int subclas
>  
>  	debug_mutex_lock_common(lock, &waiter);
>  	waiter.task = current;
> -	if (ww_ctx)
> -		waiter.ww_ctx = ww_ctx;
>  
>  	lock_contended(&lock->dep_map, ip);
>  
> @@ -630,6 +628,7 @@ __mutex_lock_common(struct mutex *lock, unsigned int state, unsigned int subclas
>  		ret = __ww_mutex_add_waiter(&waiter, lock, ww_ctx);
>  		if (ret)
>  			goto err_early_kill;
> +		waiter.ww_ctx = ww_ctx;

Yeah, I suppose this works too, but I wanted to keep the waiter
initialization together. Note how you've already called
__ww_mutex_add_waiter() on it before you actually set part of the waiter
state.