On 04/03/2012 10:38 PM, Anthony Liguori wrote: > I would hope that the distributions would also adopt a similar policy of > avoiding introducing a large number of machine types too. I think this is the > only practical way to achieve long term migration compatibility.
This is most likely correct, but what's the impact of establishing a migration baseline every 2 years? - users have to choose between features and migration compatibility - a mismatch between qemu major release cycles and downstream major release cycles means that downstreams have to create their own machine types (already true, but I was hoping to reduce the differences rather than increase them) - reduced test coverage from casual users as the default has many features disabled So I'm worried about how practical this is. There is continued pressure both to add features and to maintain migration compatibility. I don't see RHEL not adding a new machine type every minor RHEL release. -- error compiling committee.c: too many arguments to function
