On Thu, Aug 11, 2005 at 11:26:35AM -0400, Alan Stern wrote:
> On Wed, 10 Aug 2005, Greg KH wrote:
> 
> > On Mon, Aug 08, 2005 at 03:59:56PM -0400, Alan Stern wrote:
> > > On Mon, 8 Aug 2005, Patrick Mochel wrote:
> > > > On Sat, 6 Aug 2005, Alan Stern wrote:
> > > > > There probably are situations where this doesn't matter.  But in 
> > > > > general,
> > > > > don't you think the driver core should automatically remove all 
> > > > > children
> > > > > below a device that has just lost its driver?  Or is it good enough to
> > > > > rely on all the individual drivers to make sure the problem can never
> > > > > arise?
> > > > 
> > > > The driver core shouldn't do anything special in this situation for
> > > > devices that have children. It doesn't now - think of PCI bridges 
> > > > (either
> > > > Host Bridges, or PCI-PCI Bridges). They have children but no drivers.
> > > > 
> > > > It is up to the ->remove() method in the driver to handle children. Note
> > > > that it can now safely remove them without deadlocking (because of the
> > > > klist patches), which was one impetus for proceeding with the 
> > > > development,
> > > > AFAIK.
> > > 
> > > Okay.  I'd still like to hear from Greg about the special cases involved 
> > > with USB hubs and root hubs.
> > 
> > I see that your patch takes care of this now, so there isn't anything
> > else to discuss :)
> 
> Well, there's still a little bit to discuss.  Pat mentioned that in
> general, the remove method is supposed to handle children.  Presumably
> this applies to the usb_generic driver.  Should it try to unconfigure a
> USB device (thus deleting the interfaces) when it is unbound?  Or am I 
> going overboard here, considering that usb_generic is pretty much a 
> do-nothing placeholder?

It is a "placeholder" but an important one (it wouldn't work without
it...)  I hadn't thought about unbinding that "driver" before, I guess
if it's not too much trouble we should also support that kind of
unbinding.  That would allow people to test "disconnect" without ever
removing a device, much like the "fakephp" driver allows you to do this
for pci devices.

thanks,

greg k-h


-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to