Gerd Hoffmann wrote: > On 04/06/10 13:05, Amit Shah wrote: > >On (Tue) Apr 06 2010 [12:21:52], Gerd Hoffmann wrote: > >>On 04/06/10 11:58, Amit Shah wrote: > >>>>>It would certainly be beneficial for consumers of virtio-serial to be > >>>>>notified of -EAGAIN so that the guest can be throttled till the chardev > >>>>>catches up with the data being sent. > >>>> > >>>>EAGAIN should only ever occur if no bytes are written. > >>> > >>>Right. That, or just return 0 and let the caller handle the situation? > >> > >>Go with the usual unix semantics instead of creating something new. > >> > >>When something was written -- return the number of bytes. Caller has to > >>compare with the length passed in to figure whenever it was a partial > >>write or not. > >> > >>When nothing was written -- return the error. > > > >Unless Paul wants to never return recoverable error messages as he > >mentioned in his first mail. > > Return value "0" usually means end-of-file, I would not use that for > something else too. We have to agree on something though ...
Which is why EAGAIN was invented. -- Jamie