Hi Bjorn,

On Mon, Mar 31, 2008 at 12:42:03AM +0200, Bj?rn Steinbrink wrote:
quoted text
> On 2008.03.30 15:22:28 -0700, Greg KH wrote:
> > On Sun, Mar 30, 2008 at 02:51:02PM -0700, Linus Torvalds wrote:
> > > On Sun, 30 Mar 2008, Bj?rn Steinbrink wrote:
> > > > I can't reproduce the bug on my UP box and currently can't afford
> > > > crashing my SMP box (all the oopses seem to come from SMP kernels, so I
> > > > guess it needs SMP to crash), so while this doesn't show any new
> > > > problems, I can't tell whether it actually fixes anything. Testers
> > > > welcome!
> > > 
> > > Ok, I applied this because I will do an -rc8 today or tomorrow, but I 
> > > really really hope somebody can figure out what made this all start to 
> > > trigger. It does smell like some core device layer change, because we do 
> > > not seem to have a lot of changes since 2.6.24 in evdev.c and input.c that 
> > > seem relevant.
> > > 
> > > Greg, are there any refcounting changes that would cause the input devices 
> > > to be free'd earlier or something?
> > 
> > Earlier?  No, not that I know of at all, as long as the reference
> > counting logic was correct originally.  All of the problems we have been
> > fixing were ones where we accidentally were grabbing too many references
> > and then wondering why things were not getting cleaned up properly as
> > the kobject rework exposed these problems making them more obvious.
> 
> Not freeing the input device at all would of course also hide any
> access-after-free problems :-) So if that's the case, that might explain
> the sudden exposure of the problem. IMHO, my patch is the right thing to
> do anyway, because releasing a grab on the underlying input device from
> within evdev clearly needs to happen before we release that device. So
> AFAICT we're really just looking for "why do we see that bug now?" and
> "is there another bug?"
> 

If device is being disconnected (rdestroyed) then we dont really need to
release grab since there won't be any input events coming through anyway,
so there is no "another bug". I am considering removing the call to
release device once we sort out the issue with lifetime rules change,
since it is not needed.

-- 
Dmitry
--
unsubscribe notice
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/