From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755612AbZEJCvT (ORCPT ); Sat, 9 May 2009 22:51:19 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753883AbZEJCvE (ORCPT ); Sat, 9 May 2009 22:51:04 -0400 Received: from gate.crashing.org ([63.228.1.57]:35530 "EHLO gate.crashing.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753236AbZEJCvE (ORCPT ); Sat, 9 May 2009 22:51:04 -0400 Subject: Re: dma-debug: add a check dma memory leaks From: Benjamin Herrenschmidt To: David Woodhouse Cc: Joerg Roedel , Linux Kernel Mailing List , reinette.chatre@intel.com, greg@kroah.com In-Reply-To: <1241913552.24436.113.camel@macbook.infradead.org> References: <200903302101.n2UL1O1n011970@hera.kernel.org> <1240005065.19059.19.camel@macbook.infradead.org> <20090421092016.GH17438@amd.com> <1241913552.24436.113.camel@macbook.infradead.org> Content-Type: text/plain; charset="UTF-8" Date: Sun, 10 May 2009 11:46:24 +1000 Message-Id: <1241919984.19955.9.camel@pasglop> Mime-Version: 1.0 X-Mailer: Evolution 2.26.1 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 2009-05-10 at 00:59 +0100, David Woodhouse wrote: > Subject: Add BUS_NOTIFY_UNBOUND_DRIVER callback after driver removal > > This adds a notifier callback which happens _after_ the driver has been > unbound from the device, needed for things like the DMA debugging API > which want to check that all DMA mappings have been correctly torn down. > > Signed-off-by: David Woodhouse Acked-by: Benjamin Herrenschmidt > --- > On Tue, 2009-04-21 at 11:20 +0200, Joerg Roedel wrote: > > On Fri, Apr 17, 2009 at 10:51:05PM +0100, David Woodhouse wrote: > > > Hm, cute... but not quite functioning as you intended. If you look at > > > __device_release_driver() in drivers/base/dd.c you'll see it actually > > > calls the notifier _before_ calling into the driver's ->remove() method. > > > So it's hardly surprising that not everything has been freed yet... > > > > > > Reported by Reinette when it bit iwlwifi. > > > > > > Ben, can we get away with changing the order so that the ->remove() is > > > called before the notifier, in this case? > > > > Ben? I would like to keep this check. If its not possible to move this > > one behind the drivers ->remove function it may be an option to add > > another notifier? > > It doesn't look like it's possible to move the UNBIND callback. Let's > add an UNBOUND callback instead... I've tested this and it fixes the > false positives. > > Jörg, should you be using UNBOUND instead of UNBIND in the amd_iommu > notifier callback too? > > diff --git a/drivers/base/dd.c b/drivers/base/dd.c > index 742cbe6..efd00de 100644 > --- a/drivers/base/dd.c > +++ b/drivers/base/dd.c > @@ -320,6 +320,10 @@ static void __device_release_driver(struct device *dev) > devres_release_all(dev); > dev->driver = NULL; > klist_remove(&dev->p->knode_driver); > + if (dev->bus) > + blocking_notifier_call_chain(&dev->bus->p->bus_notifier, > + BUS_NOTIFY_UNBOUND_DRIVER, > + dev); > } > } > > diff --git a/include/linux/device.h b/include/linux/device.h > index 6a69caa..4ded2ae 100644 > --- a/include/linux/device.h > +++ b/include/linux/device.h > @@ -116,6 +116,7 @@ extern int bus_unregister_notifier(struct bus_type *bus, > #define BUS_NOTIFY_BOUND_DRIVER 0x00000003 /* driver bound to device */ > #define BUS_NOTIFY_UNBIND_DRIVER 0x00000004 /* driver about to be > unbound */ > +#define BUS_NOTIFY_UNBOUND_DRIVER 0x00000005 /* driver has been unbound */ > > extern struct kset *bus_get_kset(struct bus_type *bus); > extern struct klist *bus_get_device_klist(struct bus_type *bus); > >