Oliver Neukum wrote:
Right. Previously resubmit errors could happen, but you'd have hadam I right in assuming that cutting out automatic resubmission means that resubmitting an interrupt urb may now end in ENOMEM ?
no way to know, since all resubmit errors were dropped on the floor.
ENOMEM isn't very likely in practice since the HCDs use dedicated
pools of memory that don't usually need to grow, and they just put
some memory back into the pool (e.g. for TDs).
I think EPIPE would be a smidgeon more likely. Both of them have
a common recovery technique involving schedule_work().
- Dave
-------------------------------------------------------
This sf.net email is sponsored by: To learn the basics of securing your web site with SSL, click here to get a FREE TRIAL of a Thawte Server Certificate: http://www.gothawte.com/rd524.html
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
