>>Since some point in the 2.3 development cycle, device drivers use the >>common usb_submit_urb() API to request bandwidth. And that bandwidth >>is released when that URB's endpoint is no longer active. It's a >>simple model that works ... and should continue to work. > > > It does *not* work. It's too simple. Bandwidth reservation and message > transmission are different. And because they are different, a good design > should separate them.
When you say it does not work, do you mean you've been sitting on a bug report all this while? Because I don't recall *EVER* hearing about any problems in the bandwidth reservation code. Not one. That's despite the fact that I knew of several implementation bugs there. And I never said they weren't separate. I just said that only one of them was visible in the driver API. (And I showed how it's done easily, safely, and correctly.) There are quite a lot of things that are appropriate to be hidden from API users. I haven't heard any reason why this one should be different. (Other than the implicit "we haven't had a flamewar in a long time" ;) > And how do you release bandwidth? Never? You should read either my earlier message where I showed the full lifecycle of a bandwidth reservation, or the code in ohci which actually implements that. It's released when the device driver stops keeping a request active on that endpoint. - Dave ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
