> > Two GTrees are used to give access to the hashes : one indexed by hash and > > one other indexed by physical offset. > > What is the GTree indexed by physical offset used for?
I think I can get rid of the second GTree for ram based deduplication. It need to: -Start qcow2 with the deduplication operation turned off while loading hashes. So no need to check for new confliction hashes. Dedup will be resumed after complete loading. -For cluster removal and overflow the hash can be read for free from the deduplication table using the physical cluster index. Then the gtree can be manipulated with the hash. I'll try to implement this tomorow. Best regards Benoît
