I don't know if btrfs does this already; the idea is, defer reference count decrements until the spindle is available, instead of keeping up with them as writes occur. Risk: a system crash could occur between change and decrementing, causing space that should get recycled to get stuck Engineering complexity: more memory would be required to maintain a queue of deferred decrements. Logic would need to be worked out about when to include a decrement within the current critical section and when to defer it. History of concept: I was thinking about reference counting extents and wondering if mark and sweep instead would make any sense, and realized that the benefit (don't have to do another write right away) could also be gained by deferring decrements until the disk is idle instead of deferring them to a GC sweep. -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
