History

mm: thp: fix SMP race condition between THP page fault and MADV_DONTNEED

pmd_trans_unstable()/pmd_none_or_trans_huge_or_clear_bad() were introduced to locklessy (but atomically) detect when a pmd is a regular (stable) pmd or when the pmd is unstable and can infinitely transition from pmd_none() and pmd_trans_huge() from under us, while only holding the mmap_sem for reading (for writing not). While holding the mmap_sem only for reading, MADV_DONTNEED can run from under us and so before we can assume the pmd to be a regular stable pmd we need to compare it against pmd_none() and pmd_trans_huge() in an atomic way, with pmd_trans_unstable(). The old pmd_trans_huge() left a tiny window for a race. Useful applications are unlikely to notice the difference as doing MADV_DONTNEED concurrently with a page fault would lead to undefined behavior. [akpm@linux-foundation.org: tidy up comment grammar/layout] Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Reported-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Andrea Arcangeli committed 10 years ago

Commit ad33bb01 parent 29a9faa1 file · +12 −2Browse files

Unified · Split

mm/memory.c+12−2
@@ -3404,8 +3404,18 @@ static int __handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, · 3403 unmodified lines
34043404 if (unlikely(pmd_none(*pmd)) &&
34053405 unlikely(__pte_alloc(mm, vma, pmd, address)))
34063406 return VM_FAULT_OOM;
3407− /* if an huge pmd materialized from under us just retry later */
3408− if (unlikely(pmd_trans_huge(*pmd) || pmd_devmap(*pmd)))
3407+ /*
3408+ * If a huge pmd materialized under us just retry later. Use
3409+ * pmd_trans_unstable() instead of pmd_trans_huge() to ensure the pmd
3410+ * didn't become pmd_trans_huge under us and then back to pmd_none, as
3411+ * a result of MADV_DONTNEED running immediately after a huge pmd fault
3412+ * in a different thread of this mm, in turn leading to a misleading
3413+ * pmd_trans_huge() retval. All we have to ensure is that it is a
3414+ * regular pmd that we can walk with pte_offset_map() and we can do that
3415+ * through an atomic read in C, which is what pmd_trans_unstable()
3416+ * provides.
3417+ */
3418+ if (unlikely(pmd_trans_unstable(pmd) || pmd_devmap(*pmd)))
34093419 return 0;
34103420 /*
34113421 * A regular pmd is established and it can't morph into a huge pmd

Keyboard shortcuts

General

Show keyboard shortcuts
?
Focus the current search
/

Issues and pull requests

Open filters
f

Issues, pull requests and Inbox

Select next item
j↓
Select previous item
k↑
Open selected item
Enter
Dismiss filters or clear selection
Esc

Repository code

Go to file
t