On Wed, 21 Jul 2004, Oliver Neukum wrote:

> > It's unclear to me to what extent all these things really need to be
> > allocated beforehand, however.  Certainly for things like usb-storage it
> > makes sense to do this with buffers involved in actual data transfers, to
> > minimize latency if nothing else.  But for subsidiary usages, like the
> > URBs and buffers used in doing a reset?  These are pretty rare and not
> > especially time-critical, although they do need to use GFP_NOIO.
> 
> We want only one method of URB allocation, don't we?

Of course.  But that method could be used either to pre-allocate URBs and
buffers like Matt is suggesting, or to allocate them on demand like we do
now.

> > I'm puzzled how pre-allocation would help usb_kill_urb().
> 
> Reentrancy. The URB can be resubmitted from the completion handler, but
> it cannot have completed.

I don't see what this has to do with pre-allocating URBs.  (Also I don't 
understand what you mean "it cannot have completed", but maybe that's not 
important for this discussion.)

Alan Stern



-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to