[PATCH wayland] protocol: Clarify semantics of sub-surface placement requests

[Bill Spitzak]

Jonas Ådahl wrote:

> For example one could queue the operations until commit, having a
> subsequent request replace a previous one, instead of executing
> them immediately relying on commit to take a snapshot.
> 
> It could also be read as a subsequent request that now replaces a
> previous request to be invalid. This should also be clarified in
> set_position so I'll send a patch for that too.

I think the confusion is that the requests are not queued and then 
played back during a commit.

Instead the requests are applied immediately, but to a hidden copy of 
the window stacking. Also creation and destruction of surfaces are 
applied to this hidden copy. The commit then makes the visible stacking 
equal to this hidden copy. Think of it as a back buffer for the window 
stacking.