Duplication of dirent names in JFFS2 summary

[David Woodhouse]

On Fri, 2006-05-19 at 19:07 +0400, Artem B. Bityutskiy wrote:
> > First, there's the case where we've garbage-collected a dirent node but
> > haven't _yet_ deleted the original. But deleting the original is the
> > whole _point_ in GC, so that's not a common case.
> Sorry, -ENOPARSE on "But deleting the original is the whole _point_ in
> GC, so that's not a common case." 

Q: Why do we write new nodes out during GC?
A: So that we can delete the original.

Q: What is the point in GC?
A: The point in GC is that we can delete the original nodes. 

Better now?

My point is that when we've written out a colliding node during garbage
collection, that is because we intend to delete the original node real
soon now -- that deletion is what the GC is _for_. That's the whole
_point_ in the GC.

So we needn't worry too much about these collisions -- they'll be
short-lived and hence uncommon.

-- 
dwmw2