Re: [PATCH v5 1/2] mm/mmu_notifier: make interval notifier updates safe

[Jason Gunthorpe]

On Wed, Jan 15, 2020 at 02:04:47PM -0800, Ralph Campbell wrote:

> But I see your point if this sequence is done outside of the invalidate
> callback. In that case, if the driver shrank the interval, an invalidate
> callback for the right hand side could be missed before the insertion of
> the new interval for the right hand side.
> I'll explain this in the comments for nouveau_svmm_do_unmap() and
> dmirror_do_unmap().

Yes, this is why I'm not sure this is a good API for the core to
expose.

Batch manipulations is a resonable thing, but it should be forced to
work under safe conditions, ie while holding the required 'inv_begin'
on the interval tree, and the batching APIs should assert this
requirement.

Jason