On Thu, 4 Jan 2007, Oliver Neukum wrote: > Am Donnerstag, 4. Januar 2007 16:38 schrieb Alan Stern: > > > I don't know what you mean by "RESPONSE_AVAILABLE is reliable in one > > direction only". The message should be treated as a hint that data may be > > available so the host should attempt to start a transfer -- but it's still > > possible that no data bytes are available. > > I do read the spec in the way that there may be no pending response without > RESPONSE_AVAILABLE.
All it says is "using any appropriate algorithm", which is not very clear. Somebody might think their algorithm was appropriate even though it doesn't always send RESPONSE_AVAILABLE the instant a response becomes pending. For instance, let's say the RESPONSE_AVAILABLE message is sent over an interrupt endpoint with an interval of 4 ms. This means there could be a window up to 4 ms long between the time some response data becomes pending and the time the RESPONSE_AVAILABLE message is sent to the host. Alan Stern ------------------------------------------------------------------------- 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
