This could very easily be the bug where UHCI garbages the USB data toggle when control queueing is in use ... it shouldn't be patching the toggles after they're set up, that's only for queues of bulk or interrupt transfers. I'm not sure what happened to the patch which fixed that, it was available last summer.
Here it is:
http://marc.theaimsgroup.com/?l=linux-usb-devel&m=106061932810463&w=2
By now parts of it may need a little adjusting before the patch can be applied.
Yes that's the patch I was thinking about. It still applies, if you strip out the sf.net blurb at the end and ignore a small offset.
I ran it through basic "usbtest" control traffic testing, and it behaved just fine. (After reloading "modprobe uhci_hcd debug=0", to eliminate many Kbyte/sec of kernel log messages!)
Johannes, Greg -- any reason not to merge that to the 2.6.2 kernel?
- 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
