On Monday 14 January 2008, Vijay Ramamurthi wrote: > Thanks for the response David, > but somehow I want to recover, I use the asynchronous, usb_submit_urb call and > the call back never returns... > > any idea why that would happen, I have no visibility beyond my layer...! > > when would the host, think the device is dead?
I don't seem to recall such a state in the USB spec... There are disconnect states; and there are situations where the device stops behaving according to the spec. But there's no pushing-up-daisies state there, and hence none in the Linux-USB stack. > no of PINGS? what is > the condition under which the host would think device ceases to > exist... > > but I do see the IN tokens happening in my IN endpoints.. If the IN tokens are happening, then and your device is NAKing them, it's hardly dead. > > > On 1/13/08, David Brownell <[EMAIL PROTECTED]> wrote: > > > I am trying to send 64 byte packets from a linux host into a device at > > > the rate of 4000 packets/second > > > > If that rate is critical, then it should be a periodic transfer ... > > likely an "interrupt" transfer. > > > > > > > it is a bulk endpoint > > > i submit one URB per packet > > > and the device is a modem, packet deliantion is critical > > > > Sounds like a badly designed modem. Notice how it won't work right > > at full speed, since it can't possibly get that same rate. > > > > Also, consider that there are many millions of working USB modems > > that have shipped *without* that oddball 4000 packet/sec requirement. > > > > > - To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
