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]
Date:   Tue, 29 May 2018 10:09:57 +0300
From:   Oleksandr Andrushchenko <andr2000@...il.com>
To:     "Oleksandr_Andrushchenko@...m.com" <Oleksandr_Andrushchenko@...m.com>,
        linux-kernel@...r.kernel.org, dri-devel@...ts.freedesktop.org,
        airlied@...ux.ie, daniel.vetter@...el.com, seanpaul@...omium.org,
        gustavo@...ovan.org,
        Noralf Trønnes <noralf@...nnes.org>
Subject: Re: [PATCH] drm: Fix possible race conditions while unplugging DRM
 device

On 05/29/2018 10:02 AM, Daniel Vetter wrote:
> On Tue, May 22, 2018 at 05:13:04PM +0300, Oleksandr Andrushchenko wrote:
>> From: Oleksandr Andrushchenko <oleksandr_andrushchenko@...m.com>
>>
>> When unplugging a hotpluggable DRM device we first unregister it
>> with drm_dev_unregister and then set drm_device.unplugged flag which
>> is used to mark device critical sections with drm_dev_enter()/
>> drm_dev_exit() preventing access to device resources that are not
>> available after the device is gone.
>> But drm_dev_unregister may lead to hotplug uevent(s) fired to
>> user-space on card and/or connector removal, thus making it possible
>> for user-space to try accessing a disconnected device.
>>
>> Fix this by first making sure device is properly marked as
>> disconnected and only then unregister it.
>>
>> Fixes: bee330f3d672 ("drm: Use srcu to protect drm_device.unplugged")
>>
>> Signed-off-by: Oleksandr Andrushchenko <oleksandr_andrushchenko@...m.com>
>> Reported-by: Andrii Chepurnyi <andrii_chepurnyi@...m.com>
>> Cc: "Noralf Trønnes" <noralf@...nnes.org>
> Nice catch.
>
> Reviewed-by: Daniel Vetter <daniel.vetter@...ll.ch>
>
> I think you need to push this to drm-misc-next-fixes to make sure it's on
> the 4.17 train.
Sure, after I have r-b from Noralf
> -Daniel
>
>> ---
>>   drivers/gpu/drm/drm_drv.c | 14 +++++++-------
>>   1 file changed, 7 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
>> index b553a6f2ff0e..7af748ed1c58 100644
>> --- a/drivers/gpu/drm/drm_drv.c
>> +++ b/drivers/gpu/drm/drm_drv.c
>> @@ -369,13 +369,6 @@ EXPORT_SYMBOL(drm_dev_exit);
>>    */
>>   void drm_dev_unplug(struct drm_device *dev)
>>   {
>> -	drm_dev_unregister(dev);
>> -
>> -	mutex_lock(&drm_global_mutex);
>> -	if (dev->open_count == 0)
>> -		drm_dev_put(dev);
>> -	mutex_unlock(&drm_global_mutex);
>> -
>>   	/*
>>   	 * After synchronizing any critical read section is guaranteed to see
>>   	 * the new value of ->unplugged, and any critical section which might
>> @@ -384,6 +377,13 @@ void drm_dev_unplug(struct drm_device *dev)
>>   	 */
>>   	dev->unplugged = true;
>>   	synchronize_srcu(&drm_unplug_srcu);
>> +
>> +	drm_dev_unregister(dev);
>> +
>> +	mutex_lock(&drm_global_mutex);
>> +	if (dev->open_count == 0)
>> +		drm_dev_put(dev);
>> +	mutex_unlock(&drm_global_mutex);
>>   }
>>   EXPORT_SYMBOL(drm_dev_unplug);
>>   
>> -- 
>> 2.17.0
>>
>> _______________________________________________
>> dri-devel mailing list
>> dri-devel@...ts.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/dri-devel

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ