[PATCH v4 00/12] mm: make userland page table freeing RCU-safe

[Lorenzo Stoakes (ARM)]

On Tue, Sep 22, 2026 at 04:00:21PM -0700, Andrew Morton wrote:
> On Tue, 22 Sep 2026 16:35:31 +0100 "Lorenzo Stoakes (ARM)" <ljs at kernel.org> wrote:
>
> > The majority of architectures in the kernel defer page table freeing until
> > an RCU grace period has elapsed, this series converts all remaining
> > architectures to do so too and eliminates CONFIG_MMU_GATHER_RCU_TABLE_FREE
> > altogether.
> >
> > This is important because it enables safe lockless page table walking under
> > RCU alone.
>
> Do we have reason to believe that anyone is seeing any effects from
> their arch's failure to rcu-free the page tables?

We have reason to believe that it completely blocks other work like
scalable CoW :)

>
> This sounds like a rather long-standing wart, yes?

Not really, if you read the cover letter you can see the history of people
working on this, this is the final piece of the puzzle and despite the
rather muted reception this change is a big deal (TM).

>
> The patchset is thinly reviewed and mm.git is bursting at the seams
> again.  I'll play wait-and-see on this series.  Feel free to cluebat me
> if that's a bad call.

Err, Kiryl reviewed the entire series, I sent it in -rc1 of this cycle, and
it's been in -next forever?

Lance is reviewing now, presumably maybe David will review based on the
fact I updated on his feedback to _expand_ some functionality here.

So I'd argue this is contender for one of the top priority series for this
cycle, actually! [and is CRITICAL for my scalable CoW work, easily the most
important this cycle]

--
Cheers, Lorenzo