Back in October, I asked: > Has anyone ever tried to couple a device controller (gadget) and a host > controller > on the same processor to make a transparent feed-through USB channel? If it > were > possible to do this then you could make a [sort of] USB analyser. I > understand that > both sets of code use the same data structures, so how hard would it be?
I have been thinking about this again recently. I now know a bit more about the gadget API, and I think I can see the fundamental problem. Can anyone confirm whether I am on the right track? Consider a BULK IN data flow. The gadget API requires that the driver supplies it with data in advance via usb_ep_queue(). The device controller hardware buffers this data and sends it to the host when it asks for it. For my proposed coupling to work it would need to function differently: when the host asked to read something a request would be sent to the driver via a registered callback of some sort, and the driver would reply with the data to return to the host. To make it into an analyser, the driver would forward the request to the host controller to get the data, logging everything on the way through. Any thoughts? Phil. ------------------------------------------------------------------------- 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-users@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-users
