From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262349AbULONwi (ORCPT ); Wed, 15 Dec 2004 08:52:38 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262347AbULONwi (ORCPT ); Wed, 15 Dec 2004 08:52:38 -0500 Received: from e3.ny.us.ibm.com ([32.97.182.143]:48593 "EHLO e3.ny.us.ibm.com") by vger.kernel.org with ESMTP id S262349AbULONwf (ORCPT ); Wed, 15 Dec 2004 08:52:35 -0500 From: Arnd Bergmann To: "Michael S. Tsirkin" Subject: Re: unregister_ioctl32_conversion and modules. ioctl32 revisited. Date: Wed, 15 Dec 2004 14:46:01 +0100 User-Agent: KMail/1.6.2 Cc: Andi Kleen , linux-kernel@vger.kernel.org, pavel@suse.cz, discuss@x86-64.org, gordon.jin@intel.com References: <20041215065650.GM27225@wotan.suse.de> <20041215082917.GW27225@wotan.suse.de> <20041215114246.GB12187@mellanox.co.il> In-Reply-To: <20041215114246.GB12187@mellanox.co.il> MIME-Version: 1.0 Message-Id: <200412151446.01913.arnd@arndb.de> Content-Type: multipart/signed; protocol="application/pgp-signature"; micalg=pgp-sha1; boundary="Boundary-02=_ZAEwBtvB5wiZqAM"; charset="iso-8859-15" Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org --Boundary-02=_ZAEwBtvB5wiZqAM Content-Type: text/plain; charset="iso-8859-15" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline On Middeweken 15 Dezember 2004 12:42, Michael S. Tsirkin wrote: > +=A0=A0=A0=A0=A0=A0=A0/* The two calls ioctl_native and ioctl_compat desc= ribed below > +=A0=A0=A0=A0=A0=A0=A0 * can be used as a replacement for the ioctl call = above. They > +=A0=A0=A0=A0=A0=A0=A0 * take precedence over ioctl: thus if they are set= , ioctl is > +=A0=A0=A0=A0=A0=A0=A0 * not used. =A0Unlike ioctl, BKL is not taken: dri= vers manage > +=A0=A0=A0=A0=A0=A0=A0 * their own locking. */ I don't really understand how this should deal with drivers that have both generic and private ioctl conversion handlers. E.g.=20 drivers/s390/block/dasd*.c want to go through the handlers in=20 fs/compat_ioctl.c for stuff like BLKGETSIZE, while it also allows third party modules to dynamically register additional ioctls. See drivers/s390/block/dasd_cmb.c for an example, I think EMC are doing something similar to support their drives. If the dasd driver now implements an ioctl_compat() method, who will call the standard conversion handlers? Arnd <>< --Boundary-02=_ZAEwBtvB5wiZqAM Content-Type: application/pgp-signature Content-Description: signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQBBwEAZ5t5GS2LDRf4RAqWvAJ9090BRVthjX9JDHGFr1ZEwI+AFTgCeJIPX gj0X2dKVisAsM0dRJ12zxy8= =YUqC -----END PGP SIGNATURE----- --Boundary-02=_ZAEwBtvB5wiZqAM--