Subject: Re: [PATCH 32/36] driver core: Introduce device_move(): move a
	device to a new parent.
From: Marcel Holtmann <marcel@holtmann.org>
To: Greg KH <greg@kroah.com>
Cc: linux-kernel@vger.kernel.org, Cornelia Huck <cornelia.huck@de.ibm.com>,
       Greg Kroah-Hartman <gregkh@suse.de>
In-Reply-To: <20061204230511.GA9382@kroah.com>
References: <11650154123942-git-send-email-greg@kroah.com>
	 <1165015415131-git-send-email-greg@kroah.com>
	 <11650154181661-git-send-email-greg@kroah.com>
	 <11650154221716-git-send-email-greg@kroah.com>
	 <11650154251022-git-send-email-greg@kroah.com>
	 <11650154282911-git-send-email-greg@kroah.com>
	 <11650154311175-git-send-email-greg@kroah.com>
	 <1165163163.19590.62.camel@localhost> <20061204195859.GB29637@kroah.com>
	 <1165266903.12640.35.camel@localhost>  <20061204230511.GA9382@kroah.com>
Content-Type: text/plain
Date: Tue, 05 Dec 2006 16:26:11 +0100
Message-Id: <1165332371.2756.23.camel@localhost>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2580
Lines: 65

Hi Greg,

> > > > > Provide a function device_move() to move a device to a new parent device. Add
> > > > > auxilliary functions kobject_move() and sysfs_move_dir().
> > > > > kobject_move() generates a new uevent of type KOBJ_MOVE, containing the
> > > > > previous path (DEVPATH_OLD) in addition to the usual values. For this, a new
> > > > > interface kobject_uevent_env() is created that allows to add further
> > > > > environmental data to the uevent at the kobject layer.
> > > > 
> > > > has this one been tested? I don't get it working. I always get an EINVAL
> > > > when trying to move the TTY device of a Bluetooth RFCOMM link around.
> > > 
> > > I relied on Cornelia to test this.  I think some s390 patches depend on
> > > this change, right?
> > 
> > my pre-condition is that the TTY device has no parent and then we move
> > it to a Bluetooth ACL link as child. This however is not working or the
> > TTY change to use device instead of class_device has broken something.
> 
> Hm, I don't think the class_device stuff has broken anything, but if you
> think so, please let me know.

I was checking why device_move() fails and it seems that the check for
is_registered is the problem here.

        if (!device_is_registered(dev)) {
                error = -EINVAL;
                goto out;
        }

The ACL link has been attached to the Bluetooth bus, but for some reason
it still thinks that it is unregistered. Is this check really needed. I
think it should be possible to also move devices that are not part of a
bus, yet. And removing that check makes it work for me.

And btw. I can't see any s390 patches that are using device_move() at
the moment.

> > > > And shouldn't device_move(dev, NULL) re-attach it to the virtual device
> > > > tree instead of failing?
> > > 
> > > Yes, that would be good to have.
> > 
> > Cornelia, please fix this, because otherwise we can't detach a device
> > from its parent. Storing the current virtual parent looks racy to me.
> 
> You can always restore the previous "virtual" parent from the
> information given to you in the device itself.  That is what the code
> does when it first registers the device.
> 
> And yes, I too think it should be fixed.

My knowledge of the driver model is still limited. Can you fix that
quickly. This is really needed.

Regards

Marcel


-
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/