linux-kernel - Re: [RFC 4/5] BKL: use no BKL in llseek

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: <20100603070827.GB12123@infradead.org>
Date:	Thu, 3 Jun 2010 03:08:27 -0400
From:	Christoph Hellwig <hch@...radead.org>
To:	Frederic Weisbecker <fweisbec@...il.com>
Cc:	Arnd Bergmann <arnd@...db.de>, linux-kernel@...r.kernel.org,
	linux-usb@...r.kernel.org, John Kacur <jkacur@...hat.com>,
	Ingo Molnar <mingo@...e.hu>, Jan Blunck <jblunck@...e.de>
Subject: Re: [RFC 4/5] BKL: use no BKL in llseek

On Thu, Jun 03, 2010 at 02:38:27AM +0200, Frederic Weisbecker wrote:
> This should be simply dropped and we can make use of generic_file_llseek
> instead, as the fallback llseek.

generic_file_llseek contains the sb->s_maxbytes check, so it needs at
least a careful audit first.  But as mentioned about a dozend times
before it's an just as bad default.  The correct default to go ahead
it to not allow seeking and pread/pwrite unless a ->lseek method is set,
and I've already explained a step by step guide how to get there in
easily bisectable steps.  And the addition of nonseekable_open/no_llseek
by you BKL guys already got us much closer to this, so please follow up
on that work and do it properly.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/