Re: [PATCH] driver core: fix shutdown races with probe/remove

[Ming Lei]

Hi,

On Tue, Jun 5, 2012 at 5:18 PM, Greg Kroah-Hartman
<gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> On Tue, Jun 05, 2012 at 04:59:07PM +0800, Ming Lei wrote:
>> Firstly, .shutdown callback may touch a uninitialized hardware
>> if dev->driver is set and .probe is not completed.
>
> When would that ever happen?

It may happen when someone plugs a device, then run 'shutdown' immediately.

>
>> Secondly, device_shutdown() may dereference a null pointer to cause
>> oops when dev->driver is cleared after it is checked in device_shutdown().
>
> Again, when would that happen?  A new device/driver being added to the
> system as it is being shutdown?

It is usually that someone presses 'power off' button and unplugs
its usb devices or pci/e devices, maybe together with a new plug.

In such case, the oops may be triggered.

>
> Has anyone ever hit that before?

Looks someone has reported it, I also can trigger it
with some delay added in device_shutdown.

>
> Is this really needed?

In case that it is triggered, oops will prevent system from being
shutdown, and it should be a big problem.

IMO, we can avoid it without much change and side effect, so
looks worthy the change.

Thanks,
--
Ming Lei
--
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/