On 12/04/2018 13:53, Kevin Wolf wrote:
>> The problem I have is that there is a direction through which I/O flows
>> (parent-to-child), so why can't draining follow that natural direction.
>> Having to check for the parents' I/O, while draining the child, seems
>> wrong.  Perhaps we can't help it, but I cannot understand the reason.
> I'm not sure what's there that could be not understood. You already
> confirmed that we need to drain the parents, too, when we drain a node.
> Drain really must propagate in the opposite direction of I/O, because
> part of its job is to quiesce the origin of any I/O to the node that
> should be drained. Opposite of I/O _is_ the natural direction for drain.

Opposite of I/O is the natural direction for drain to propagate, yes.

However, I/O direction is the natural direction for requests to stop.
After quiescing X and calling X->drv->bdrv_drain(X), there can be
pending requests only in X's children.  So I don't understand why you
need to keep checking in_flight over the whole subgraph, when there are
roots that will conclude their request first, and then their children,
and so on so forth.



> We also have subtree drains, but that's not because that's the natural
> direction for drain, but just as a convenience function because some
> operations (e.g. reopen) affect a whole subtree, so they need everything
> in that subtree drained rather than just a single node.

Reply via email to