linux-kernel - Re: 2.6.25-rc6: CONFIG_USB

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: <20080609205916.GC21429@elf.ucw.cz>
Date:	Mon, 9 Jun 2008 22:59:16 +0200
From:	Pavel Machek <pavel@...e.cz>
To:	Greg KH <gregkh@...e.de>
Cc:	Alan Stern <stern@...land.harvard.edu>,
	Oliver Neukum <oliver@...kum.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Linus Torvalds <torvalds@...uxfoundation.org>,
	Andrew Morton <akpm@...uxfoundation.org>,
	kernel list <linux-kernel@...r.kernel.org>,
	"Rafael J. Wysocki" <rjw@...k.pl>
Subject: Re: 2.6.25-rc6: CONFIG_USB_PERSIST forced on

On Mon 2008-06-09 12:50:15, Greg KH wrote:
> On Mon, Jun 09, 2008 at 11:44:21AM -0400, Alan Stern wrote:
> > On Mon, 9 Jun 2008, Oliver Neukum wrote:
> > 
> > > Am Montag 09 Juni 2008 17:03:10 schrieb Alan Stern:
> > > > On Mon, 9 Jun 2008, Pavel Machek wrote:
> > > > 
> > > > > Besides, it seems to break suspend/resume of printers, and probably
> > > > > all the drivers that do not have reset_resume() method. That's
> > > > > actually a regression.
> > > > > 
> > > > > https://bugzilla.novell.com/show_bug.cgi?id=394820
> > > > 
> > > > The right way to fix this is to add reset_resume to the printer driver.  
> > > 
> > > reset_resume() is supposed to restore all state. The printer driver does
> > > not know which state a printer is in, except for the trivial case of the
> > > printer not being in use, as it doesn't know the meaning of the data
> > > going to the printer.
> > > 
> > > You might argue that you deserve what you get when you hibernate
> > > while printing, but then it makes no sense to implement it anyhow,
> > > disconnection and reconnection work just as well and are cleaner.
> > > The same is true for many devices.
> > 
> > In which case the correct approach is the second one I mentioned (which 
> > you omitted in your reply): Make usbcore unbind drivers that don't 
> > support reset_resume.
> 
> That sounds reasonable to me.  Oliver or Pavel, care to try this out?

I'm not an USB hacker, and 2.6.26 release is pretty near. I do have
USB printer somewhere around, maybe it still works.

But problem is _not_ limited to usblp. usblp is just a part of
problem, the one we caught in testing.

reset_resume() is only implemented in storage/usb.c... while resume()
is pretty widespread in the usb drivers.

I believe we should just revert the "CONFIG_USB_PERSIST force on"
patch, and solve this properly in 2.6.27.
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
--
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/