[foreign-memaccess] on shared segments

[Maurizio Cimadamore]

On 27/09/2019 11:46, Jorn Vernee wrote:
> 2) seems fine to me. We can phrase the doc to be vague about whether a 
> newly created segment will be returned, and just say we will return 'a 
> segment with properties XYZ', then returning the same segment in some 
> cases will just be an implementation detail.

I think the problem here is slightly different... yes, we can make the 
doc vague about the identity of the returned object - but we have to 
describe the side-effect of invalidating existing segment (because that 
effect will be visible to clients!). And that side-effect only happens 
if a 'new' segment is returned.

So, compared to (2), (3) applies the invalidation logic always.

Maurizio