From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [PATCH] proper replacements for ->proc_info Date: Fri, 25 Apr 2003 17:17:01 +0200 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <20030425171701.B8222@lst.de> References: <20030423212157.A18850@lst.de> <1051126145.1970.7.camel@mulgrave> <20030425122754.A5030@lst.de> <20030425104137.GE10374@parcelfarce.linux.theplanet.co.uk> <20030425125918.A5330@lst.de> <20030425165252.A7951@lst.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from verein.lst.de ([212.34.181.86]:53775 "EHLO verein.lst.de") by vger.kernel.org with ESMTP id S263301AbTDYPEy (ORCPT ); Fri, 25 Apr 2003 11:04:54 -0400 Content-Disposition: inline In-Reply-To: <20030425165252.A7951@lst.de>; from hch@lst.de on Fri, Apr 25, 2003 at 04:52:52PM +0200 List-Id: linux-scsi@vger.kernel.org To: viro@parcelfarce.linux.theplanet.co.uk Cc: James Bottomley , SCSI Mailing List On Fri, Apr 25, 2003 at 04:52:52PM +0200, Christoph Hellwig wrote: > On Fri, Apr 25, 2003 at 12:59:18PM +0200, Christoph Hellwig wrote: > > > And that loop is protected by ...? > > > > The right lock would be sdev->list_lock. > > Actually it would probably be host->host_lock, not a per-sdev lock > of course. Locking down the my_device traversal is on my TODO > list but a separate thing, though. Oh, well. Digging through all the (ab)users of ->my_devices I start to develop a really strong favour of just nuking /proc/scsi//* stuff entirely. There's really not much in there that shouldn't better be done in sysfs.