Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1666776 > unrolled thread

[PATCHv2 2/3] mm: Do not loose dirty and access bits in pmdp_invalidate()

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [PATCHv2 2/3] mm: Do not loose dirty and access bits in pmdp_invalidate() "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> - 2017-06-15 17:00 +0200
    Re: [PATCHv2 2/3] mm: Do not loose dirty and access bits in  pmdp_invalidate() Andrea Arcangeli <aarcange@redhat.com> - 2017-06-16 15:50 +0200
      Re: [PATCHv2 2/3] mm: Do not loose dirty and access bits in  pmdp_invalidate() "Kirill A. Shutemov" <kirill@shutemov.name> - 2017-06-19 15:30 +0200

#1666776 — [PATCHv2 2/3] mm: Do not loose dirty and access bits in pmdp_invalidate()

Vlastimil noted that pmdp_invalidate() is not atomic and we can loose
dirty and access bits if CPU sets them after pmdp dereference, but
before set_pmd_at().

The bug doesn't lead to user-visible misbehaviour in current kernel.

Loosing access bit can lead to sub-optimal reclaim behaviour for THP,
but nothing destructive.

Loosing dirty bit is not a big deal too: we would make page dirty
unconditionally on splitting huge page.

The fix is critical for future work on THP: both huge-ext4 and THP swap
out rely on proper dirty tracking.

The patch change pmdp_invalidate() to make the entry non-present atomically and
return previous value of the entry. This value can be used to check if
CPU set dirty/accessed bits under us.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reported-by: Vlastimil Babka <vbabka@suse.cz>
---
 include/asm-generic/pgtable.h | 2 +-
 mm/pgtable-generic.c          | 9 +++++----
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 7dfa767dc680..ece5e399567a 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -309,7 +309,7 @@ extern pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp);
 #endif
 
 #ifndef __HAVE_ARCH_PMDP_INVALIDATE
-extern void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
+extern pmd_t pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
 			    pmd_t *pmdp);
 #endif
 
diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c
index c99d9512a45b..148fe36f61a7 100644
--- a/mm/pgtable-generic.c
+++ b/mm/pgtable-generic.c
@@ -179,12 +179,13 @@ pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp)
 #endif
 
 #ifndef __HAVE_ARCH_PMDP_INVALIDATE
-void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
+pmd_t pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
 		     pmd_t *pmdp)
 {
-	pmd_t entry = *pmdp;
-	set_pmd_at(vma->vm_mm, address, pmdp, pmd_mknotpresent(entry));
-	flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+	pmd_t old = pmdp_establish(pmdp, pmd_mknotpresent(*pmdp));
+	if (pmd_present(old))
+		flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+	return old;
 }
 #endif
 
-- 
2.11.0

[toc] | [next] | [standalone]

#1667755 — Re: [PATCHv2 2/3] mm: Do not loose dirty and access bits in pmdp_invalidate()

On Thu, Jun 15, 2017 at 05:52:23PM +0300, Kirill A. Shutemov wrote:
> -void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
> +pmd_t pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
>  		     pmd_t *pmdp)
>  {
> -	pmd_t entry = *pmdp;
> -	set_pmd_at(vma->vm_mm, address, pmdp, pmd_mknotpresent(entry));
> -	flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
> +	pmd_t old = pmdp_establish(pmdp, pmd_mknotpresent(*pmdp));
> +	if (pmd_present(old))
> +		flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
> +	return old;
>  }
>  #endif

The pmd_present() check added above is superflous because there's no
point to call pmdp_invalidate if the pmd is not present (present as in
pmd_present) already. pmd_present returns true if _PAGE_PSE is set
and it was always set before calling pmdp_invalidate.

It looks like we could skip the flush if _PAGE_PRESENT is not set
(i.e. for example if the pmd is PROTNONE) but that's not what the above
pmd_present will do.

[toc] | [prev] | [next] | [standalone]

#1669067 — Re: [PATCHv2 2/3] mm: Do not loose dirty and access bits in pmdp_invalidate()

On Fri, Jun 16, 2017 at 03:40:41PM +0200, Andrea Arcangeli wrote:
> On Thu, Jun 15, 2017 at 05:52:23PM +0300, Kirill A. Shutemov wrote:
> > -void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
> > +pmd_t pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
> >  		     pmd_t *pmdp)
> >  {
> > -	pmd_t entry = *pmdp;
> > -	set_pmd_at(vma->vm_mm, address, pmdp, pmd_mknotpresent(entry));
> > -	flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
> > +	pmd_t old = pmdp_establish(pmdp, pmd_mknotpresent(*pmdp));
> > +	if (pmd_present(old))
> > +		flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
> > +	return old;
> >  }
> >  #endif
> 
> The pmd_present() check added above is superflous because there's no
> point to call pmdp_invalidate if the pmd is not present (present as in
> pmd_present) already. pmd_present returns true if _PAGE_PSE is set
> and it was always set before calling pmdp_invalidate.
> 
> It looks like we could skip the flush if _PAGE_PRESENT is not set
> (i.e. for example if the pmd is PROTNONE) but that's not what the above
> pmd_present will do.

You are right. We seems don't have a generic way to check the entry is
present to CPU.

I guess I'll drop the check then.

-- 
 Kirill A. Shutemov

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web