[RFC PATCH 4/6] mm/memory: Add tree limit to free_pgtables()

[Lorenzo Stoakes]

On Wed, Sep 03, 2025 at 04:19:04PM -0400, Liam R. Howlett wrote:
> * Lorenzo Stoakes <lorenzo.stoakes at oracle.com> [250819 15:14]:
> > On Fri, Aug 15, 2025 at 03:10:29PM -0400, Liam R. Howlett wrote:
> > > The ceiling and tree search limit need to be different arguments for the
> > > future change in the failed fork attempt.
> > >
> > > No functional changes intended.
> > >
> > > Signed-off-by: Liam R. Howlett <Liam.Howlett at oracle.com>
> >
> > (Obv. in addition to comment about broken VMA tests :P)
> >
> > I guess intent is that if we discover any page tables beyond tree_max then
> > we ought to just wipe them all out so, in effect, we don't consider
> > mappings at or past tree_max to be valid?
>
> Actually... there are some archs that map outside the vma and they are
> valid.. I think mips? and I think lower, but yeah.. it's needed.  This
> is why prev->vm_end and next->vm_start are used as page table limits,
> afaik.  This is a serious annoyance because it frequently adds walks
> that are infrequently necessary to the vma tree.

ugh god. I was vaguely aware of this but that's gross. Very gross.

I need to document all the VMA weirdnesses smoewhere.

What do they do this for? Guard pages or something?

>
> >
> > I feel like we need a comment to this effect as this is confusing as it is.
> >
> > Could we add a kerneldoc comment for free_pgtables() spelling this out?
>
> I'll add a note here, but confusion will probably increase.  I'll add a
> note about the tree max as well.

Thanks!

>
> >
> > > ---
> > >  mm/internal.h | 4 +++-
> > >  mm/memory.c   | 7 ++++---
> > >  mm/mmap.c     | 2 +-
> > >  mm/vma.c      | 3 ++-
> > >  4 files changed, 10 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/mm/internal.h b/mm/internal.h
> > > index 45b725c3dc030..f9a278ac76d83 100644
> > > --- a/mm/internal.h
> > > +++ b/mm/internal.h
> > > @@ -444,7 +444,9 @@ void folio_activate(struct folio *folio);
> > >
> > >  void free_pgtables(struct mmu_gather *tlb, struct ma_state *mas,
> > >  		   struct vm_area_struct *start_vma, unsigned long floor,
> > > -		   unsigned long ceiling, bool mm_wr_locked);
> > > +		   unsigned long ceiling, unsigned long tree_max,
> > > +		   bool mm_wr_locked);
> > > +
> > >  void pmd_install(struct mm_struct *mm, pmd_t *pmd, pgtable_t *pte);
> > >
> > >  struct zap_details;
> > > diff --git a/mm/memory.c b/mm/memory.c
> > > index 0ba4f6b718471..3346514562bba 100644
> > > --- a/mm/memory.c
> > > +++ b/mm/memory.c
> > > @@ -371,7 +371,8 @@ void free_pgd_range(struct mmu_gather *tlb,
> > >
> > >  void free_pgtables(struct mmu_gather *tlb, struct ma_state *mas,
> > >  		   struct vm_area_struct *vma, unsigned long floor,
> > > -		   unsigned long ceiling, bool mm_wr_locked)
> > > +		   unsigned long ceiling, unsigned long tree_max,
> > > +		   bool mm_wr_locked)
> > >  {
> > >  	struct unlink_vma_file_batch vb;
> > >
> > > @@ -385,7 +386,7 @@ void free_pgtables(struct mmu_gather *tlb, struct ma_state *mas,
> > >  		 * Note: USER_PGTABLES_CEILING may be passed as ceiling and may
> > >  		 * be 0.  This will underflow and is okay.
> > >  		 */
> > > -		next = mas_find(mas, ceiling - 1);
> > > +		next = mas_find(mas, tree_max - 1);
> >
> > Do we need to put some sort of sanity checks in to make sure tree_max <= ceiling
> > (though this 0 case is a pain... so I guess tree_max - 1 <= ceiling - 1?)
>
> Sure!

Thanks!