From mboxrd@z Thu Jan 1 00:00:00 1970 From: Lan Tianyu Subject: Re: [PATCH 2/2] ACPI / bind: Use (put|get)_device() on ACPI device objects too Date: Thu, 14 Nov 2013 15:20:09 +0800 Message-ID: <528479A9.9020802@intel.com> References: <3268437.YsusHvklcv@vostro.rjw.lan> <1803808.ghCmmaO2Cj@vostro.rjw.lan> <9712243.It8gvaRCs0@vostro.rjw.lan> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mga09.intel.com ([134.134.136.24]:41047 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751087Ab3KNHbW (ORCPT ); Thu, 14 Nov 2013 02:31:22 -0500 In-Reply-To: <9712243.It8gvaRCs0@vostro.rjw.lan> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: "Rafael J. Wysocki" Cc: ACPI Devel Maling List , LKML , Linux PCI , Greg Kroah-Hartman , Bjorn Helgaas , Aaron Lu , Jarkko Nikula , Mika Westerberg , "Luck, Tony" On 2013=E5=B9=B411=E6=9C=8814=E6=97=A5 07:26, Rafael J. Wysocki wrote: > From: Rafael J. Wysocki >=20 > When associating a "physical" device with an ACPI device object > acpi_bind_one() only uses get_device() to increment the reference > counter of the former, but there is no reason not to do that with > the latter too. Among other things, that may help to avoid > use-after-free when an ACPI device object is freed without calling > acpi_unbind_one() for all "physical" devices associated with it > (that only can happen in buggy code, but then it's better if the > kernel doesn't crash as a result of a bug). >=20 > For this reason, modify acpi_bind_one() to apply get_device() to > the ACPI device object too and update acpi_unbind_one() to drop > that reference using put_device() as appropriate. >=20 > Signed-off-by: Rafael J. Wysocki > --- > drivers/acpi/glue.c | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) >=20 > Index: linux-pm/drivers/acpi/glue.c > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > --- linux-pm.orig/drivers/acpi/glue.c > +++ linux-pm/drivers/acpi/glue.c > @@ -217,6 +217,7 @@ int acpi_bind_one(struct device *dev, ac > if (!acpi_dev) > return -EINVAL; > =20 > + get_device(&acpi_dev->dev); > get_device(dev); > physical_node =3D kzalloc(sizeof(*physical_node), GFP_KERNEL); > if (!physical_node) { > @@ -243,6 +244,7 @@ int acpi_bind_one(struct device *dev, ac > goto err; > =20 > put_device(dev); > + put_device(&acpi_dev->dev); > return 0; > } > if (pn->node_id =3D=3D node_id) { > @@ -282,6 +284,7 @@ int acpi_bind_one(struct device *dev, ac > err: > ACPI_COMPANION_SET(dev, NULL); > put_device(dev); > + put_device(&acpi_dev->dev); > return retval; > } > EXPORT_SYMBOL_GPL(acpi_bind_one); > @@ -307,8 +310,9 @@ int acpi_unbind_one(struct device *dev) > sysfs_remove_link(&acpi_dev->dev.kobj, physnode_name); > sysfs_remove_link(&dev->kobj, "firmware_node"); > ACPI_COMPANION_SET(dev, NULL); > - /* acpi_bind_one() increase refcnt by one. */ > + /* Drop references taken by acpi_bind_one(). */ > put_device(dev); > + put_device(&acpi_dev->dev); > kfree(entry); > break; > } >=20 Reviewed-by: Lan Tianyu --=20 Best regards Tianyu Lan -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Message-ID: <528479A9.9020802@intel.com> Date: Thu, 14 Nov 2013 15:20:09 +0800 From: Lan Tianyu MIME-Version: 1.0 To: "Rafael J. Wysocki" CC: ACPI Devel Maling List , LKML , Linux PCI , Greg Kroah-Hartman , Bjorn Helgaas , Aaron Lu , Jarkko Nikula , Mika Westerberg , "Luck, Tony" Subject: Re: [PATCH 2/2] ACPI / bind: Use (put|get)_device() on ACPI device objects too References: <3268437.YsusHvklcv@vostro.rjw.lan> <1803808.ghCmmaO2Cj@vostro.rjw.lan> <9712243.It8gvaRCs0@vostro.rjw.lan> In-Reply-To: <9712243.It8gvaRCs0@vostro.rjw.lan> Content-Type: text/plain; charset=UTF-8 Sender: linux-acpi-owner@vger.kernel.org List-ID: On 2013年11月14日 07:26, Rafael J. Wysocki wrote: > From: Rafael J. Wysocki > > When associating a "physical" device with an ACPI device object > acpi_bind_one() only uses get_device() to increment the reference > counter of the former, but there is no reason not to do that with > the latter too. Among other things, that may help to avoid > use-after-free when an ACPI device object is freed without calling > acpi_unbind_one() for all "physical" devices associated with it > (that only can happen in buggy code, but then it's better if the > kernel doesn't crash as a result of a bug). > > For this reason, modify acpi_bind_one() to apply get_device() to > the ACPI device object too and update acpi_unbind_one() to drop > that reference using put_device() as appropriate. > > Signed-off-by: Rafael J. Wysocki > --- > drivers/acpi/glue.c | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > Index: linux-pm/drivers/acpi/glue.c > =================================================================== > --- linux-pm.orig/drivers/acpi/glue.c > +++ linux-pm/drivers/acpi/glue.c > @@ -217,6 +217,7 @@ int acpi_bind_one(struct device *dev, ac > if (!acpi_dev) > return -EINVAL; > > + get_device(&acpi_dev->dev); > get_device(dev); > physical_node = kzalloc(sizeof(*physical_node), GFP_KERNEL); > if (!physical_node) { > @@ -243,6 +244,7 @@ int acpi_bind_one(struct device *dev, ac > goto err; > > put_device(dev); > + put_device(&acpi_dev->dev); > return 0; > } > if (pn->node_id == node_id) { > @@ -282,6 +284,7 @@ int acpi_bind_one(struct device *dev, ac > err: > ACPI_COMPANION_SET(dev, NULL); > put_device(dev); > + put_device(&acpi_dev->dev); > return retval; > } > EXPORT_SYMBOL_GPL(acpi_bind_one); > @@ -307,8 +310,9 @@ int acpi_unbind_one(struct device *dev) > sysfs_remove_link(&acpi_dev->dev.kobj, physnode_name); > sysfs_remove_link(&dev->kobj, "firmware_node"); > ACPI_COMPANION_SET(dev, NULL); > - /* acpi_bind_one() increase refcnt by one. */ > + /* Drop references taken by acpi_bind_one(). */ > put_device(dev); > + put_device(&acpi_dev->dev); > kfree(entry); > break; > } > Reviewed-by: Lan Tianyu -- Best regards Tianyu Lan