linux-kernel - Re: lockdep warning in urb.c:363 usb_submit

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: <1584980280.27949.20.camel@suse.de>
Date:   Mon, 23 Mar 2020 17:18:00 +0100
From:   Oliver Neukum <oneukum@...e.de>
To:     Alan Stern <stern@...land.harvard.edu>
Cc:     Qais Yousef <qais.yousef@....com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        linux-usb@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: lockdep warning in urb.c:363 usb_submit_urb

Am Montag, den 23.03.2020, 11:54 -0400 schrieb Alan Stern:
> On Mon, 23 Mar 2020, Oliver Neukum wrote:
> 
> > Am Montag, den 23.03.2020, 14:38 +0000 schrieb Qais Yousef:
> > > Hi
> > > 
> > > I've hit the following lockdep warning when I trigger hibernate on arm64
> > > platform (Juno-r2)
> > > 
> > > 
> > > 	echo suspend > /sys/power/disk
> > > 	echo disk > /sys/power/state
> > > 
> > > I only had a usb flash drive attached to it. Let me know if you need more info.
> > 
> > Hi,
> > 
> > that is not a lockdep issue, but the hub driver is not properly killing
> > its URB presumably. Yet, the driver looks correct to me. Please use
> > the additional patch and activate dynamic debugging for usbcore.
> 
> Was the USB flash drive being used as a swap device for holding the 
> hibernation image?  That's not likely to work very well.  At least, I 
> doubt that it has been tested very much.

Right, but this is good. We are getting a test for something that needs
work. It does not really matetr why STD fails.

> This diagnostic was suggested by the runtime PM error that occurred 
> when the system was trying to store the hibernation image.  That's 
> probably when the hub driver's URB got restarted.

AFAICT hub_quiesce() unconditionally calls usb_kill_urb(). So I'd like
to verify that case is really triggered.

	Regards
		Oliver