Commit 1ad9f620 authored by Mel Gorman's avatar Mel Gorman Committed by Linus Torvalds

mm: numa: recheck for transhuge pages under lock during protection changes

Sasha reported the following bug using trinity

  kernel BUG at mm/mprotect.c:149!
  invalid opcode: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
  Dumping ftrace buffer:
     (ftrace buffer empty)
  Modules linked in:
  CPU: 20 PID: 26219 Comm: trinity-c216 Tainted: G        W    3.14.0-rc5-next-20140305-sasha-00011-ge06f5f3-dirty #105
  task: ffff8800b6c80000 ti: ffff880228436000 task.ti: ffff880228436000
  RIP: change_protection_range+0x3b3/0x500
  Call Trace:

The VM_BUG_ON was added in -mm by the patch "mm,numa: reorganize
change_pmd_range".  The race existed without the patch but was just
harder to hit.

The problem is that a transhuge check is made without holding the PTL.
It's possible at the time of the check that a parallel fault clears the
pmd and inserts a new one which then triggers the VM_BUG_ON check.  This
patch removes the VM_BUG_ON but fixes the race by rechecking transhuge
under the PTL when marking page tables for NUMA hinting and bailing if a
race occurred.  It is not a problem for calls to mprotect() as they hold
mmap_sem for write.
Signed-off-by: default avatarMel Gorman <>
Reported-by: default avatarSasha Levin <>
Reviewed-by: default avatarRik van Riel <>
Signed-off-by: default avatarAndrew Morton <>
Signed-off-by: default avatarLinus Torvalds <>
parent 88a9ab6e
......@@ -36,6 +36,34 @@ static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot)
* For a prot_numa update we only hold mmap_sem for read so there is a
* potential race with faulting where a pmd was temporarily none. This
* function checks for a transhuge pmd under the appropriate lock. It
* returns a pte if it was successfully locked or NULL if it raced with
* a transhuge insertion.
static pte_t *lock_pte_protection(struct vm_area_struct *vma, pmd_t *pmd,
unsigned long addr, int prot_numa, spinlock_t **ptl)
pte_t *pte;
spinlock_t *pmdl;
/* !prot_numa is protected by mmap_sem held for write */
if (!prot_numa)
return pte_offset_map_lock(vma->vm_mm, pmd, addr, ptl);
pmdl = pmd_lock(vma->vm_mm, pmd);
if (unlikely(pmd_trans_huge(*pmd) || pmd_none(*pmd))) {
return NULL;
pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, ptl);
return pte;
static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
unsigned long addr, unsigned long end, pgprot_t newprot,
int dirty_accountable, int prot_numa)
......@@ -45,7 +73,10 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
spinlock_t *ptl;
unsigned long pages = 0;
pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
pte = lock_pte_protection(vma, pmd, addr, prot_numa, &ptl);
if (!pte)
return 0;
do {
oldpte = *pte;
......@@ -132,12 +163,13 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
pages += HPAGE_PMD_NR;
/* huge pmd was handled */
/* fall through, the trans huge pmd just split */
this_pages = change_pte_range(vma, pmd, addr, next, newprot,
dirty_accountable, prot_numa);
pages += this_pages;
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment