From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dmitry Torokhov Subject: Re: [PATCH 2/3] input: use input_mutex instead of BKL when opening input device Date: Tue, 9 Mar 2010 22:56:38 -0800 Message-ID: <201003092256.39223.dmitry.torokhov@gmail.com> References: <1268162386-2745-1-git-send-email-cascardo@holoscopio.com> <20100310062015.GA2445@core.coreip.homeip.net> <201003100749.20723.oliver@neukum.org> Mime-Version: 1.0 Content-Type: Text/Plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <201003100749.20723.oliver@neukum.org> Sender: linux-kernel-owner@vger.kernel.org To: Oliver Neukum Cc: Thadeu Lima de Souza Cascardo , Arnd Bergmann , linux-input@vger.kernel.org, linux-kernel@vger.kernel.org, John Kacur , jkosina@suse.cz List-Id: linux-input@vger.kernel.org On Tuesday 09 March 2010 10:49:20 pm Oliver Neukum wrote: > Am Mittwoch, 10. M=E4rz 2010 07:20:15 schrieb Dmitry Torokhov: > > > I was willing to let the open call go unprotected. But ended up > > > checking that the three callees were fine (they do not call any o= f the > > > other functions that take the mutex). > > >=20 > > > Since the fops_put/fops_get do protect that section from the hand= ler > > > removal and I can't think of any other race right now, I think yo= ur > > > version is really better. > > >=20 > > > Acked-by: Thadeu Lima de Souza Cascardo > >=20 > > OK, applied 2 Thadeu's patches and one Arnd's. >=20 > Jiri, >=20 > it is possible that this requires a change in usbhid as it uses BKL. > Do you remember? > No, different layer altogether. --=20 Dmitry