[PATCH v2 00/12] mm: make userland page table freeing RCU-safe

[Lorenzo Stoakes (ARM)]

On Tue, Sep 08, 2026 at 02:15:43PM -0700, Andrew Morton wrote:
> On Tue, 08 Sep 2026 13:32:09 +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.
> >
> > Doing so allows for reduced lock contention, avoids lock ordering concerns
> > and enables fast, efficient and correct page table walking as a result.
> >
> > Additionally it removes a bunch of code and architecture-specific behaviour
> > which is always a beneficial thing to do.
>
> Thanks.  It's early and unreviewed, but this is mainly an
> if-it-compiles-its-good thing, so I'll add it for exposure to the
> build/checker universe.

Yeah largely, and I have done additional testing for cases which need it.

>
> > However some arches required extra attention - sh-X2, m68k-motorola and
> > sparc32.
>
> OK.  Please lmk later if you think additional reviewer/tester attention
> is desirable in these areas.

Only really the fancier sparc32 setup which can't be emulated [there's a call
out in the cover for that one!] :) though sashiko found an issue with that that
is now fixed.

The other two I boot tested locally under emulation with no issues :)

>

--
Cheers, Lorenzo