This message is from the T13 list server.
> > Wanna walk thru some concrete real world use cases? > > > > 1) I remember seeing an Apple host combine with a mouse to only > > occasionally zero the trailing bits of a command on its way to a Usb > > disk drive. The drive saw a request to read/write zero bytes and > > completed that request without error. The host only knew trouble > > was at hand because the Usb1/Pio4 bridge was able to report > > precisely that zero bytes had moved. > I don't recall the particulars of this case you mention, > but I do remember there being a problem > with a certain usb hub or device > which corrupted an otherwise well-formed > command from the host. Sorry I didn't more clearly state: I meant only to name names of things that just plain work. The bridge in question was clearly out of Usb spec, the Apple host was fine. Any bus analyser you can name showed the bytes of the command transferred with Crc-checked integrity, NOT zeroed. The academic fact that Apple was right didn't change the interoperability reality of "but it works with Windows". This particular device only also worked on the Mac because the generic UsbMass protocol was by design robust enough to tell the host what happened. The only host-side Usb corruption I remember was low-speed traffic mixed into full-speed traffic scrambling the full-speed data. But I thought that was not Apple but rather ... except just now I promised to name only things that just plain work. My personal level of involvement varies. Some Atapi hardware ships whose design I influenced much. More Atapi hardware ships whose firmware I wrote. No FireWire/Usb hardware ships whose firmware I wrote (yet?). The further I am way from the metal, the less I can fix. I'm into having the standard be clear and capable and broadly understood because so often I can fix things no other way. Pat LaVarre Subscribe/Unsubscribe instructions can be found at www.t13.org.
