On 13/10/2016 12:48, Halil Pasic wrote: >> > > I'm fine with this. I just think, it would be nice if the contract between > the vmstate-core and the client code implementing VMStateInfo callbacks > could be documented, including when are certain parameters valid, what > they stand for, and how are they supposed to be used for the next version of > the patch. Just to improve readability. Would this be OK with everybody? > > By the way the flag VMS_SINGLE is documented as ignored. Should we drop > it?
Yes, I think so. Paolo