From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Graf Subject: Re: [PATCH] PKT_SCHED: Initialize list field in dummy qdiscs Date: Sun, 7 Nov 2004 20:08:41 +0100 Message-ID: <20041107190841.GD31969@postel.suug.ch> References: <418C2D40.9020300@trash.net> <20041106015931.GA28715@postel.suug.ch> <20041106145036.GB28715@postel.suug.ch> <418DE37E.2050504@trash.net> <20041107140015.GA31969@postel.suug.ch> <418E4B2E.1070407@trash.net> <20041107163330.GB31969@postel.suug.ch> <418E553C.2070006@trash.net> <20041107174909.GC31969@postel.suug.ch> <418E67D0.8010301@trash.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: davem@davemloft.net, netdev@oss.sgi.com, spam@crocom.com.pl, kuznet@ms2.inr.ac.ru, jmorris@redhat.com Return-path: To: Patrick McHardy Content-Disposition: inline In-Reply-To: <418E67D0.8010301@trash.net> Sender: netdev-bounce@oss.sgi.com Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org > >Anyways, I do think we should force the task to be completed, or > >at least all the list unlinking, before the rtnl semaphore is given > >back. I'm fine with postponing the deletion of the object but not > >to postpone list manipulations even if we cannot reproduce it now. > > This is what I'm doing now, your patch to set qdisc->parent makes > this very easy :) Simply remove all qdiscs with > (TC_H_MAJ(q->parent) == TC_H_MAJ(qdisc->handle)) from > dev->qdisc_list when a classful qdisc is destroyed. Great, that's even better, speeds up deletion of big class trees. Thanks.