Re: [PATCH] mm: swap: check for xa_zero_entry() on vma in swapoff path

[Lorenzo Stoakes <lorenzo.stoakes@oracle.com>]

On Mon, Aug 11, 2025 at 03:19:53PM +0200, David Hildenbrand wrote:
> > > >
> > > > > When registering vmas for uprobe, skip the vmas in an mm that is marked
> > > > > unstable.  Modifying a vma in an unstable mm may cause issues if the mm
> > > > > isn't fully initialised.__
> > > > >
> > > > > > Is there anything preventing us from just leaving a proper tree that
> > > > > > reflects reality in place before we drop the write lock?
> > > > >
> > > > > When you mean proper tree, is this about the your previous question? --
> > > > > Shouldn't we just remove anything from the tree here that was not copied
> > > > > immediately?
> > > >
> > > > Commit d24062914837 (" fork: use __mt_dup() to duplicate maple tree in
> > > > dup_mmap()") did this for efficiency, so it'd be a regression to do this.
> > >
> > > We're talking about the case where fork *fails*. That cannot possibly be
> > > relevant for performance, can it? :)
> >
> > I think it optimises the overall operation, but as a product of that, has to
> > handle this edge case, and that necessitated this rather horrble stuff.
> >
> > Obviously we don't need to optimise a 'we are about to die' case :)
>
> Right, so my original question was whether we could just drop all stale
> stuff from the tree before we lift the mmap write lock, leaving only the
> VMAs in there that we actually processed successfully.

That'd be better answered by Liam who's more familiar with it.

I think it may actually be difficult to do on some level or there was some
reason we couldn't, but I may be mistaken.

>
> --
> Cheers,
>
> David / dhildenb
>