On Thu, 22 Dec 2011, Sarah Sharp wrote: > On Mon, Dec 19, 2011 at 10:08:55AM -0800, Sarah Sharp wrote: > > On Mon, Dec 19, 2011 at 11:57:46AM -0500, Alan Stern wrote: > > > Sarah: > > > > > > Do you have any experts on EHCI on tap at Intel? I've got some > > > questions about the Interrupt on Async Advance Doorbell (IAAD) bit in > > > the USBCMD register. The documentation is incomplete, inconsistent, > > > and confusing. > > > > Sure, I'll forward your email to John Howard and Karthi, who were on the > > list of authors of the EHCI spec. > > John says: > > This is a 'doorbell', the intent is a transition of 0b to 1b would 'ring' the > doorbell (think edge-triggered event). The interrupt associated with this event > occurs at the next interrupt threshold, at which time hardware will set the > bit to 0b. > > The implications after the 0-1 transition and to the interrupt threshold are: > > The bit will persist as a 1b value. > > Software cannot cancel the event by writing a zero (0b) to the bit. > > Software can write the bit with any value it wants and it will have no effect. > > After the interrupt threshold time, the process resets. > > Does this help? Yes, it does help considerably. Thanks, and thanks to John. It would have been nice if these points had been explained in the spec... But I really can't complain; the EHCI spec is a marvel of clarity and completeness compared to many others. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
