linux-kernel - Re: [JANITOR PROPOSAL] Switch ioctl functions to ->unlocked

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: <200801100952.55039.eike-kernel@sf-tec.de>
Date:	Thu, 10 Jan 2008 09:52:49 +0100
From:	Rolf Eike Beer <eike-kernel@...tec.de>
To:	Andi Kleen <andi@...stfloor.org>
Cc:	linux-kernel@...r.kernel.org, kernel-janitors@...r.kernel.org,
	paolo.ciarrocchi@...il.com, gorcunov@...il.com
Subject: Re: [JANITOR PROPOSAL] Switch ioctl functions to ->unlocked_ioctl

Andi Kleen wrote:
> Here's a proposal for some useful code transformations the kernel janitors
> could do as opposed to running checkpatch.pl.
>
> Most ioctl handlers still running implicitely under the big kernel
> lock (BKL). But long term Linux is trying to get away from that. There is a
> new ->unlocked_ioctl entry point that allows ioctls without BKL, but the
> code needs to be explicitely converted to use this.
>
> The first step of getting rid of the BKL is typically to make it visible
> in the source. Once it is visible people will have incentive to eliminate
> it. That is how the BKL conversion project for Linux long ago started too.
> On 2.0 all system calls were still implicitely BKL and in 2.1 the
> lock/unlock_kernel()s were moved into the various syscall functions and
> then step by step eliminated.

Can you explain the rationale behind that running on the BKL? What type of 
things needs to be protected so that this huge hammer is needed? What would 
be an earlier point to release the BKL?

Greetings,

Eike

Download attachment "signature.asc " of type "application/pgp-signature" (195 bytes)