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

Groups > linux.kernel > #1661039 > unrolled thread

Re: [PATCH 1/3] mm: numa: avoid waiting on freed migrated pages

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

  Re: [PATCH 1/3] mm: numa: avoid waiting on freed migrated pages "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> - 2017-06-08 12:30 +0200

#1661039 — Re: [PATCH 1/3] mm: numa: avoid waiting on freed migrated pages

On Tue, Jun 06, 2017 at 06:58:34PM +0100, Will Deacon wrote:
> From: Mark Rutland <mark.rutland@arm.com>
> 
> In do_huge_pmd_numa_page(), we attempt to handle a migrating thp pmd by
> waiting until the pmd is unlocked before we return and retry. However,
> we can race with migrate_misplaced_transhuge_page():
> 
> // do_huge_pmd_numa_page                // migrate_misplaced_transhuge_page()
> // Holds 0 refs on page                 // Holds 2 refs on page
> 
> vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd);
> /* ... */
> if (pmd_trans_migrating(*vmf->pmd)) {
>         page = pmd_page(*vmf->pmd);
>         spin_unlock(vmf->ptl);
>                                         ptl = pmd_lock(mm, pmd);
>                                         if (page_count(page) != 2)) {
>                                                 /* roll back */
>                                         }
>                                         /* ... */
>                                         mlock_migrate_page(new_page, page);
>                                         /* ... */
>                                         spin_unlock(ptl);
>                                         put_page(page);
>                                         put_page(page); // page freed here
>         wait_on_page_locked(page);
>         goto out;
> }
> 
> This can result in the freed page having its waiters flag set
> unexpectedly, which trips the PAGE_FLAGS_CHECK_AT_PREP checks in the
> page alloc/free functions. This has been observed on arm64 KVM guests.
> 
> We can avoid this by having do_huge_pmd_numa_page() take a reference on
> the page before dropping the pmd lock, mirroring what we do in
> __migration_entry_wait().
> 
> When we hit the race, migrate_misplaced_transhuge_page() will see the
> reference and abort the migration, as it may do today in other cases.
> 
> Acked-by: Steve Capper <steve.capper@arm.com>
> Signed-off-by: Mark Rutland <mark.rutland@arm.com>
> Signed-off-by: Will Deacon <will.deacon@arm.com>

Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

-- 
 Kirill A. Shutemov

[toc] | [standalone]

Back to top | Article view | linux.kernel

csiph-web