Re: [PATCH] mm/huge_memory: bypass THP tuneables for huge pfnmap mappings

["Lorenzo Stoakes (ARM)" <ljs@kernel.org>]

On Mon, Sep 07, 2026 at 06:14:14PM +0200, David Hildenbrand (Arm) wrote:
> On 8/27/26 21:55, Lorenzo Stoakes (ARM) wrote:
> > The sysfs THP tuneables at /sys/kernel/mm/transparent_huge_pages/ rather
> > confusingly only control the behaviour of THP in some instances.
> >
> > They are not applicable to MADV_COLLAPSE operations, nor to DAX mappings.
> >
> > Long-term, THP is predicated upon compaction being able to obtain large
> > folios to populate THP ranges.
> >
> > However, vm_normal_folio() returns NULL for PFN map mappings, thus their
> > reference count is maintained by the driver, not core mm.
> >
> > As a consequence, the folios are not subject to reclaim nor compaction, so
> > are not truly part of the THP mechanism at all.
> >
> > However, since commit 5dd40721f147 ("mm: allow THP orders for PFNMAPs")
> > introduced the ability to establish huge PFN maps, they have been subject
> > to THP tuneables.
> >
> > This is incorrect - if a huge PFN map is available (defined by
> > vma->vm_ops->huge_fault being non-NULL for a VMA_PFNMAP_BIT VMA), then it
> > should be mapped huge upon fault-in.
> >
> > Correct this by explicitly checking for this while ensuring that smaps
> > continues to accurately report THPeligible statistics.
> >
> > While here, abstract the entire file-backed THP check in
> > vma_can_map_huge_file(), with sensible separation of logic into helper
> > functions.
> >
> > Note that drm_gem_shmem_mmap() and panthor_gem_mmap() establish huge PFN
> > maps of shmem folios, however they are marked unevictable in
> > drm_gem_get_pages(), and in any case would fail the reference check in
> > __remove_mapping() even if they weren't.
> >
> > Failing to map huge PFN maps has resulted in significant real-world
> > performance degradation, see links for details.
> >
> > Reported-by: Cedric Le Goater <clg@redhat.com>
> > Closes: https://lore.kernel.org/linux-mm/20260805055544.1568534-1-clg@redhat.com/
> > Reported-by: Saravanan D <saravanand@crusoe.ai>
> > Closes: https://lore.kernel.org/linux-mm/20260821070520.25759-1-saravanand@crusoe.ai/
> > Fixes: 5dd40721f147 ("mm: allow THP orders for PFNMAPs")
> > Cc: stable@vger.kernel.org
> > Signed-off-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
> > ---
> >  mm/huge_memory.c | 86 +++++++++++++++++++++++++++++++++++++++++---------------
> >  1 file changed, 64 insertions(+), 22 deletions(-)
> >
> > diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> > index afbb5974bd22..4bf7b670586d 100644
> > --- a/mm/huge_memory.c
> > +++ b/mm/huge_memory.c
> > @@ -92,7 +92,7 @@ unsigned long huge_anon_orders_madvise __read_mostly;
> >  unsigned long huge_anon_orders_inherit __read_mostly;
> >  static bool anon_orders_configured __initdata;
> >
> > -static inline bool file_thp_enabled(struct vm_area_struct *vma)
> > +static inline bool file_thp_enabled(const struct vm_area_struct *vma)
> >  {
> >  	struct inode *inode;
> >
> > @@ -118,6 +118,67 @@ static bool vma_is_special_huge(const struct vm_area_struct *vma)
> >  	return vma_test_any(vma, VMA_PFNMAP_BIT, VMA_MIXEDMAP_BIT);
> >  }
> >
> > +static bool vma_bypass_thp_tuneables_file(const struct vm_area_struct *vma,
> > +		enum tva_type type)
> > +{
> > +	const bool has_huge_fault = vma->vm_ops->huge_fault;
> > +
> > +	/* MADV_COLLAPSE ignores tuneables. */
> > +	if (type == TVA_FORCED_COLLAPSE)
> > +		return true;
> > +	/* Huge PFN mappings are uncompactable so the policy doesn't apply. */
> > +	if (vma_test(vma, VMA_PFNMAP_BIT) && has_huge_fault)
> > +		return true;
> > +	return false;
> > +}
> > +
> > +static bool vma_thp_tuneables_allow_file(vm_flags_t vm_flags)
> > +{
> > +	/* THP=always? */
> > +	if (hugepage_global_always())
> > +		return true;
> > +	/* THP=madvise and marked MADV_HUGEPAGE? */
> > +	if (hugepage_global_enabled() && (vm_flags & VM_HUGEPAGE))
> > +		return true;
> > +	return false;
> > +}
> > +
> > +static bool vma_check_thp_tuneables_file(const struct vm_area_struct *vma,
> > +		vm_flags_t vm_flags, enum tva_type type)
> > +{
> > +	return vma_bypass_thp_tuneables_file(vma, type) ||
> > +		vma_thp_tuneables_allow_file(vm_flags);
> > +}
> > +
> > +static bool vma_can_map_huge_file(const struct vm_area_struct *vma,
> > +		vm_flags_t vm_flags, enum tva_type type)
> > +{
> > +	const bool has_huge_fault = vma->vm_ops->huge_fault;
> > +
> > +	/*
> > +	 * Enforce THP collapse requirements as necessary. Anonymous vmas
> > +	 * were already handled in thp_vma_allowable_orders().
> > +	 */
> > +	if (!vma_check_thp_tuneables_file(vma, vm_flags, type))
> > +		return false;
> > +
> > +	switch (type) {
> > +	case TVA_PAGEFAULT:
> > +		/*
> > +		 * Trust that ->huge_fault() handlers know what they are doing
> > +		 * in fault path.
> > +		 */
> > +		return has_huge_fault;
> > +	case TVA_SMAPS:
> > +		if (has_huge_fault)
> > +			return true;
> > +		fallthrough;
> > +	default:
> > +		/* Only regular file is valid in collapse path. */
> > +		return file_thp_enabled(vma);
> > +	}
> > +}
> > +
> >  unsigned long __thp_vma_allowable_orders(struct vm_area_struct *vma,
> >  					 vm_flags_t vm_flags,
> >  					 enum tva_type type,
> > @@ -190,27 +251,8 @@ unsigned long __thp_vma_allowable_orders(struct vm_area_struct *vma,
> >  						   vma, vma_start_pgoff(vma), 0,
> >  						   forced_collapse);
> >
> > -	if (!vma_is_anonymous(vma)) {
> > -		/*
> > -		 * Enforce THP collapse requirements as necessary. Anonymous vmas
> > -		 * were already handled in thp_vma_allowable_orders().
> > -		 */
> > -		if (!forced_collapse &&
> > -		    (!hugepage_global_enabled() || (!(vm_flags & VM_HUGEPAGE) &&
> > -						    !hugepage_global_always())))
> > -			return 0;
> > -
> > -		/*
> > -		 * Trust that ->huge_fault() handlers know what they are doing
> > -		 * in fault path.
> > -		 */
> > -		if (((in_pf || smaps)) && vma->vm_ops->huge_fault)
> > -			return orders;
> > -		/* Only regular file is valid in collapse path */
> > -		if (((!in_pf || smaps)) && file_thp_enabled(vma))
> > -			return orders;
> > -		return 0;
> > -	}
> > +	if (!vma_is_anonymous(vma))
> > +		return vma_can_map_huge_file(vma, vm_flags, type) ? orders : 0;
>
> I don't have the time to find out how these functions ended up being named, I'll
> have to trust that it will be alright.

:-O

>
> Adding to what Zi say, vma_can_map_huge_file is also confusing, no?

Yeah I agree, I had a model in my head as to why it was an OK name then when Zi
raised it I realised that my naming scheme was, in fact, terrible :)

>
> are we mapping a huge file? :)

Lol... hugetlb maybe? HW poision too? Perhaps? OK you've only been back for a
day I should stop using these swear words :P

>
> --
> Cheers,
>
> David

--
Cheers, Lorenzo