Re: 2.6.30-rc6: Reported regressions from 2.6.29

From: Greg KH
Date: Sun May 17 2009 - 23:49:23 EST


On Sat, May 16, 2009 at 07:13:59PM -0700, Linus Torvalds wrote:
> On Sun, 17 May 2009, Kay Sievers wrote:
> >
> > This makes the oops in the driver-core, caused by the rtc driver
> > unregister, go away. The original issue is also fixed in the rtc driver
> > itself.
>
> I don't think this is sufficient.
>
> > --- a/drivers/base/driver.c
> > +++ b/drivers/base/driver.c
> > @@ -257,6 +257,8 @@ EXPORT_SYMBOL_GPL(driver_register);
> > */
> > void driver_unregister(struct device_driver *drv)
> > {
> > + if (!drv || !drv->p)
> > + return;
> > driver_remove_groups(drv, drv->groups);
> > bus_remove_driver(drv);
> > }
>
> Ok, fine so far, but look at "driver_register()".
>
> It will set drv->p, but then not unset it if it fails! (For a certain
> class of failures)
>
> So for a certain failure pattern, drv->p will point to some stale value.
> Should we not clear drv->p in the "out_unregister" patch?
>
> To confuse the thing more, there are actually "half-way failures" that
> _succeed_ in driver registration, but then return an error code. See that
> whole
>
>
> kobject_uevent(&priv->kobj, KOBJ_ADD);
> return error;
>
> case in the "success" path driver_register(). We may return an error
> despite the fact that we actually attached the driver to bus, but
> "add_bind_files()" failed. A caller would be understandable very unhappy.
>
> So I suspect we should do something like the appended (in addition to your
> patch). Comments?

Yes, that's needed. We also don't clean up properly when we do the same
thing for devices (error in the initialization of them).

So, here's a patch that combines everyone's intentions in this thread.
Any objections to it?

thanks,

greg k-h

------------------------
From: Kay Sievers <kay.sievers@xxxxxxxx>
Subject: Driver Core: do not oops when driver_unregister() is called for unregistered drivers

We also fix a problem with cleaning up properly when initializing
drivers and devices, so checks like this will work successfully.

Portions of the patch by Linus and Greg and Ingo.

Reported-by: Ozan ÃaÄlayan <ozan@xxxxxxxxxxxxx>
Signed-off-by: Kay Sievers <kay.sievers@xxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxx>

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index dc030f1..c659961 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -700,8 +700,10 @@ int bus_add_driver(struct device_driver *drv)
}

kobject_uevent(&priv->kobj, KOBJ_ADD);
- return error;
+ return 0;
out_unregister:
+ kfree(drv->p);
+ drv->p = NULL;
kobject_put(&priv->kobj);
out_put_bus:
bus_put(bus);
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 4aa527b..1977d4b 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -879,7 +879,7 @@ int device_add(struct device *dev)
}

if (!dev_name(dev))
- goto done;
+ goto name_error;

pr_debug("device: '%s': %s\n", dev_name(dev), __func__);

@@ -978,6 +978,9 @@ done:
cleanup_device_parent(dev);
if (parent)
put_device(parent);
+name_error:
+ kfree(dev->p);
+ dev->p = NULL;
goto done;
}

diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index c51f11b..8ae0f63 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -257,6 +257,10 @@ EXPORT_SYMBOL_GPL(driver_register);
*/
void driver_unregister(struct device_driver *drv)
{
+ if (!drv || !drv->p) {
+ WARN(1, "Unexpected driver unregister!\n");
+ return;
+ }
driver_remove_groups(drv, drv->groups);
bus_remove_driver(drv);
}
--
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/