I've found, in the EHCI driver, that queue heads set up for an interrupt pipe (such as the one set up to check for connection changes on the USB2.0 hub) are set up correctly, but the "halt" (QTD_STS_HALT) bit in the hw_token in the overlay area is never cleared after the QH and QTDs are set up and
On 2.4, that doesn't happen since about 20-June-2003; and on 2.6, just a few days earlier (BK describes it as a "micro-patch").
scheduled. With this halt bit set, the HC will ignore the QH, so it doesn't get executed. If the hw_token in the overlay is cleared to 0, the HC will advance the queue (copy the first QTD into the overlay area) and do the transaction.
Sounds to me like your base kernel is missing EHCI patches I'd call essential. More current kernels should be fine.
- Dave
------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
