Re: [PATCH] drivers-core: nullify private pointer on device-release

From: Guennadi Liakhovetski
Date: Mon Oct 05 2009 - 12:16:01 EST


On Mon, 5 Oct 2009, Greg KH wrote:

> On Mon, Oct 05, 2009 at 10:11:50AM +0200, Guennadi Liakhovetski wrote:
> > On Thu, 1 Oct 2009, Greg KH wrote:
> >
> > > On Thu, Oct 01, 2009 at 03:45:56PM +0200, Guennadi Liakhovetski wrote:
> > > > On Thu, 1 Oct 2009, Greg KH wrote:
> > > >
> > > > > On Thu, Oct 01, 2009 at 10:02:08AM +0200, Guennadi Liakhovetski wrote:
> > > > > > Device structures can be reused over multiple device_add / device_release
> > > > > > cycles.
> > > > >
> > > > > They shouldn't be as they should be dynamic, not static.
> > > >
> > > > Should they? I'm pretty sure this is not the first time this comes up -
> > > > there are several drivers and / or subsystems, that re-use driver objects.
> > >
> > > Then those drivers and subsystems should be fixed, as that is incorrect.
> > >
> > > > But finding in mail archives wouldn't be very easy. And it worked until
> > > > now - why should we break it?
> > >
> > > I would argue that this code was always broken.
> > > When did this problem show up for you?
> >
> > Since commit b4028437876866aba4747a655ede00f892089e14
>
> Again, which driver/devices are having this problem?

Quoting my previous reply in this thread:

> > What device is having this problem?
>
> My problem case is the soc-camera framework. There device struct is
> embedded into the video client object, which are kept as long as the
> driver is loaded.

</quote>

> The patch
> referenced above had been in linux-next for almost 6 months with no
> reported problems, so this is news to me :)

Hm, really 6 months in linux-next? That surprises me too, that I didn't
notice it until now. But in any case, the aforementioned commit introduced
a regression. If you disagree, that it has to be fixed centrally as per
proposed patch, no problem, I'm pushing a patch tonight, that will fix
this for soc-camera. No idea about other drivers.

Thanks
Guennadi
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/