On Sat, Nov 5, 2016 at 5:06 AM, Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote: > On Fri, Nov 4, 2016 at 2:30 AM, Miklos Szeredi <miklos@xxxxxxxxxx> wrote: >> >> Also introduce the concept of feature flags to allow backward incompatible >> changes to the overlay format. This should have been there from day one; the >> best we can do now is backport to stable kernels. Add the check for features >> without adding any actual features yet. > > No. I pulled the three other commits, but not that last one. > > That feature just seems to actively *encourage* backwards incompatible > features. It's a bad idea. Don't do it. If we've been able to do > without it so far, then why should we suddenly start doing things like > this? > > So I don't agree that it should have been there since day one, it just > shouldn't exist at all. > Linus, Can you please clarify your objection? I suppose you do not object to the concept of on-disk format version nor on-disk format compatible/incompatible features sets. Just to fact that overlayfs didn't have those form day one, so it should find a way to cope with that situation without patching stable kernels? Thanks, Amir. -- To unsubscribe from this list: send the line "unsubscribe linux-unionfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
