On Thu, 25 Jan 2007, Oleg Verych wrote:
> On Thu, Jan 25, 2007 at 03:53:58PM +0100, Oliver Neukum wrote:
> []
> > Yes, there is something wrong with the design of this device.
> > For these cases there's
> > int usb_driver_set_configuration(struct usb_device *udev, int config)
> >
> > You check the current configuration probe() is called with, call
> > this helper function and exit with an error from probe.
>
> It is relatively new. As comment states, it doesn't guarantee
> anything, caller hangs in air...
Actually that routine has only two significant failure modes:
1. Unable to allocate a small block of memory. If memory is that
tight, you probably don't want your driver running at all.
2. The Set-Config request itself fails. If that happens you're
completely stuck -- even if a userspace program had requested
the change it wouldn't work.
So on the whole usb_driver_set_configuration() should be very reliable.
And the only times it can fail are when it wouldn't do any good anyway.
As for being relatively new... That's true, and it's a fault all code
suffers from at one time or another. There's not much we can do about it
except wait until the routine gets older. :-)
Alan Stern
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
