From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: [PATCH (resend)]: SCSI: fix /proc memory leak in the SCSI core Date: Sun, 22 Feb 2009 13:46:03 -0600 Message-ID: <1235331963.4531.61.camel@localhost.localdomain> References: Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from accolon.hansenpartnership.com ([76.243.235.52]:51311 "EHLO accolon.hansenpartnership.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752263AbZBVTqI (ORCPT ); Sun, 22 Feb 2009 14:46:08 -0500 In-Reply-To: Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Alan Stern Cc: Andrew Morton , SCSI development list On Sun, 2009-02-22 at 14:34 -0500, Alan Stern wrote: > On Sun, 22 Feb 2009, James Bottomley wrote: > > > OK, but resending a patch you expressed reservations about putting in > > without testing doesn't really help me. I need a way to get comfortable > > with its safety. > > > > So, what about this alternative fix instead: if the removal were moved > > to scsi_host_put(), that would address all the problems and have the > > advantage that everyone will test it ... > > I thought of doing it that way too. It has the disadvantage of > exposing part of the proc interface to userspace before the host is > registered. Now, since all we're adding is the host's directory, maybe > this doesn't matter. But it didn't seem like a good idea. It's current behaviour (and has been so for all of git history) with no reported bugs. All it's doing is creating a proc dir ... it's not exposing any interfaces within, so from a theoretical standpoint it's perfectly OK. The necessity for finding a legacy system to test was precisely because you moved it. James