On Wed, 9 Jun 2004, David Brownell wrote: > So Greg, the quick fix should just be reverting the last patch > that touches hub.c ... that's what looks to have moved the > initialization of the TT state after its first use! :)
Clearly David has hit the nail on the head. Patch as305 (that's the one David identified) moved the Transaction Translator initialization to a later point. The idea was that devices being reset already have their TT set up so they don't need to execute the initialization code. What I forgot was that new devices being set up for the first time need to have their TT initialized quite early on! Unfortunately I'm not in a position to send in a fix for this right now. Maybe the best thing will be to revert as305, as308b, as310, and as311 for the time being (and obviously don't apply as312). When I get a chance I will redo as305, leaving the TT code where it belongs. Alternatively, if someone feels up to it they can create a patch that will undo the part of as305 that moves the TT initialization. (Or you can just wait until early next week, when I will have some free time.) Unfortunately not having a high-speed hub of my own to test with, TT's are kind of a blind spot for me. I apologize to everyone for the inconvenience. Alan Stern ------------------------------------------------------- This SF.Net email is sponsored by: GNOME Foundation Hackers Unite! GUADEC: The world's #1 Open Source Desktop Event. GNOME Users and Developers European Conference, 28-30th June in Norway http://2004/guadec.org _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
