linux-kernel - Re: [next-20240925] possible circular locking: uart vs kmemleak

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]

Message-ID: <20240926144159.GF11458@google.com>
Date: Thu, 26 Sep 2024 23:41:59 +0900
From: Sergey Senozhatsky <senozhatsky@...omium.org>
To: Petr Mladek <pmladek@...e.com>
Cc: Sergey Senozhatsky <senozhatsky@...omium.org>,
	John Ogness <john.ogness@...utronix.de>,
	Catalin Marinas <catalin.marinas@....com>,
	Steven Rostedt <rostedt@...dmis.org>,
	Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org,
	linux-kernel@...r.kernel.org, linux-serial@...r.kernel.org
Subject: Re: [next-20240925] possible circular locking: uart vs kmemleak

On (24/09/26 15:32), Petr Mladek wrote:
> On Thu 2024-09-26 13:07:15, Sergey Senozhatsky wrote:
> > Greetings,
> > 
> > Ran into the following issue today.  It's sort of interesting, not sure
> > what even to do about it.  The
> > 
> > 	uart -> tty -> mm /* kmalloc -> kmemleak */
> > 
> > chain looks problematic, it certainly overlaps with
> > 
> > 	mm -> printk -> uart  /* which can kmalloc and re-enter mm -> kmemleak? */
> 
> I believe that it will get solved by the uart console driver
> conversion to nbcon. I should remove the path:
> 
> 	printk -> uart
> 
> or more precisely, it should remove the path:
> 
> 	console_owner --> &port_lock_key
> 
> The patchset with the uart 8250 console driver conversion is still
> pending a review, see the last version at
> https://lore.kernel.org/r/20240913140538.221708-1-john.ogness@linutronix.de

It does fix the issue for me, you are right.  Thanks for the pointers.