lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20151115230255.GZ17128@valkosipuli.retiisi.org.uk>
Date:	Mon, 16 Nov 2015 01:02:56 +0200
From:	Sakari Ailus <sakari.ailus@....fi>
To:	Shuah Khan <shuahkh@....samsung.com>
Cc:	mchehab@....samsung.com, linux-media@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] media: fix kernel hang in media_device_unregister()
 during device removal

Hi Shuah,

On Thu, Nov 12, 2015 at 07:41:47AM -0700, Shuah Khan wrote:
> Media core drivers (dvb, v4l2, bridge driver) unregister
> their entities calling media_device_unregister_entity()
> during device removal from their unregister paths. In
> addition media_device_unregister() tries to unregister
> entity calling media_device_unregister_entity() for each
> one of them. This adds lot of contention on mdev->lock in
> device removal sequence. Fix to not unregister entities
> from media_device_unregister(), and let drivers take care
> of it. Drivers need to unregister to cover the case of
> module removal. This patch fixes the problem by deleting
> the entity list walk to call media_device_unregister_entity()
> for each entity. With this fix there is no kernel hang after
> a sequence of device insertions followed by removal.
> 
> Signed-off-by: Shuah Khan <shuahkh@....samsung.com>
> ---
>  drivers/media/media-device.c | 5 -----
>  1 file changed, 5 deletions(-)
> 
> diff --git a/drivers/media/media-device.c b/drivers/media/media-device.c
> index 1312e93..c7ab7c9 100644
> --- a/drivers/media/media-device.c
> +++ b/drivers/media/media-device.c
> @@ -577,8 +577,6 @@ EXPORT_SYMBOL_GPL(__media_device_register);
>   */
>  void media_device_unregister(struct media_device *mdev)
>  {
> -	struct media_entity *entity;
> -	struct media_entity *next;
>  	struct media_link *link, *tmp_link;
>  	struct media_interface *intf, *tmp_intf;
>  
> @@ -596,9 +594,6 @@ void media_device_unregister(struct media_device *mdev)
>  		kfree(intf);
>  	}
>  
> -	list_for_each_entry_safe(entity, next, &mdev->entities, graph_obj.list)
> -		media_device_unregister_entity(entity);
> -
>  	device_remove_file(&mdev->devnode.dev, &dev_attr_model);
>  	media_devnode_unregister(&mdev->devnode);
>  

media_device_unregister() is expected to clean up all the entities still
registered with it (as it does to links and interfaces). Could you share
details of the problems you saw in case you haven't found the actual
underlying issue causing them?

-- 
Kind regards,

Sakari Ailus
e-mail: sakari.ailus@....fi	XMPP: sailus@...iisi.org.uk
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ