On Tuesday 03 October 2006 11:04 am, Alan Stern wrote: > On Mon, 2 Oct 2006, David Brownell wrote: > > > > > > In principle it's possible to change ehci-hcd so that the SETUP > > > transaction of a control transfer is sent in a different microframe from > > > the following IN/OUT transactions. In practice it might be a painfully > > > difficult change to write. > > > > Probably just mark only one TD active at a time, and have the qh > > completion logic activate the others ... except that runs smack > > into some silicon errata, some controllers don't handle inactive > > TDs correctly. > > If these controllers don't handle inactive TDs correctly, what do they do > when they reach the dummy TD at the end of a queue?
Misbehave: it goes into the overlay. There's code to explicitly handle that class of silicon bug. (Observed on EHCI 0.95 chips, so one could charatibly blame it on "things" not being final.) - Dave ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
