Am Dienstag, 21. Januar 2003 20:47 schrieb Alan Stern:
> Oliver et al.:
>
> In response to one of your previous messages, I tried reducing the scope
> of the changes required for properly tracking urb states and synchronous
> unlinking. It turns out that in addition to the extra spinlock, I needed
> 2 bits: one to indicate when the urb is in use (owned by the core) and
> one to indicate that it has been unlinked.
>
> Just for kicks I created a patch (included below) that allocates those
> bits from urb->transfer_flags. That's not such a great place, because
> it's liable to be overwritten by device drivers. But the only alternative
> is to add a whole extra field (like internal_state).
I believe that you should do that.
> Functionally this does pretty much what we have been discussing. Overall
> the changes are pretty minimal, just what is necessary to make things
> work.
It would be great if we could eleminate the need for handling unlinking
during completion at all. Is there a reason you cannot make them mutually
exclusive? IMHO it should be possible to do that if you make it a requirement
to call giveback with handler_lock held.
Regards
Oliver
-------------------------------------------------------
This SF.net email is sponsored by: Scholarships for Techies!
Can't afford IT training? All 2003 ictp students receive scholarships.
Get hands-on training in Microsoft, Cisco, Sun, Linux/UNIX, and more.
www.ictp.com/training/sourceforge.asp
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
