# Re: \[PATCH v1 06/11\] mm: support GUP-triggered unsharing via FAULT\_FLAG\_UNSHARE (!hugetlb)

- From: Linus Torvalds \<torvalds@linux-foundation.org\>
- Date: 2021-12-19 21:53:57 
- List: linux-doc
- Page: https://ratatoskr.run/linux-doc/2021/12/1613513
- Thread: https://ratatoskr.run/linux-mm/2021/12/1612167/t/2.md

```text
On Sun, Dec 19, 2021 at 1:48 PM Matthew Wilcox [ref] wrote:
>
> Yes, agreed, I was thinking that we could use "not mapped at all"
> as an optimisation to avoid doing rmap walks.  eg __unmap_and_move().

So the thing is, it's a very dodgy optimization for a rather simple
reason: what if somebody pages the page in?

So even "not mapped at all" is questionable.

You have to check that it's also not a swapcache page, and hold the
page lock for that check, at the very least.

And by then, you're really in a very unusual situation - and my gut
feel says not one worth optimizing for (because anon pages are
_usually_ mapped at least once).

But I dunno - it might depend on your load. Maybe you have some very
special load that happens to trigger this case a lot?

              Linus
```
