mm: mempolicy: Skip inaccessible VMAs when setting MPOL_MF_LAZY

PATCH In review Mel Gorman 3 participants 3 messages Oct 2, 2014

Subsystems (2)

Memory management MM NUMA placement

lore

Share & tools

lore

Mel Gorman

author

2014-10-02 19:17 UTC #0 lore raw

PATCH 1/1 mm: mempolicy: Skip inaccessible VMAs when setting MPOL_MF_LAZY

PROT_NUMA VMAs are skipped to avoid problems distinguishing between present, prot_none and special entries. MPOL_MF_LAZY is not visible from userspace since commit a720094ded8c ("mm: mempolicy: Hide MPOL_NOOP and MPOL_MF_LAZY from userspace for now") but it should still skip VMAs the same way task_numa_work does.

Signed-off-by Mel Gorman <[email protected]>

Acked-by Rik van Riel <[email protected]>

---
 mm/mempolicy.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

mm/mempolicy.c +3 -1

index 8f5330d..a5877ce 100644
@@ -683,7 +683,9 @@ queue_pages_range(struct mm_struct *mm, unsigned long start, unsigned long end,
 		}
 
 		if (flags & MPOL_MF_LAZY) {
-			change_prot_numa(vma, start, endvma);
+			/* Similar to task_numa_work, skip inaccessible VMAs */
+			if (vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE))
+				change_prot_numa(vma, start, endvma);
 			goto next;
 		}

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Hugh Dickins

Acked-by on v1 1/1

2014-10-02 19:43 UTC #1 lore raw

Quoted 8 lines · Mel Gorman

PROT_NUMA VMAs are skipped to avoid problems distinguishing between present, prot_none and special entries. MPOL_MF_LAZY is not visible from userspace since commit a720094ded8c ("mm: mempolicy: Hide MPOL_NOOP and MPOL_MF_LAZY from userspace for now") but it should still skip VMAs the same way task_numa_work does.
Signed-off-by Mel Gorman <[email protected]>
Acked-by Rik van Riel <[email protected]>

Acked-by Hugh Dickins <[email protected]>

Yes, this is much the same as the patch I wrote for Linus two days ago, then discovered that we don't need until MPOL_MF_LAZY gets brought back into MPOL_MF_VALID. (As a bonus, my patch did also remove the currently bogus paragraph of comment above change_prot_numa(); and I would prefer a code comment to make clear that we never exercise this path at present.)

Quoted 9 lines

---
 mm/mempolicy.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

mm/mempolicy.c +0 -0

index 8f5330d..a5877ce 100644

mm/mempolicy.c queue_pages_range(struct mm_struct *mm, unsigned long start, unsigned long end, · line 691

Quoted 10 lines

mm/mempolicy.c +3 -1

@@ -683,7 +683,9 @@ queue_pages_range(struct mm_struct *mm, unsigned long start, unsigned long end,
 		}
 
 		if (flags & MPOL_MF_LAZY) {
-			change_prot_numa(vma, start, endvma);
+			/* Similar to task_numa_work, skip inaccessible VMAs */
+			if (vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE))
+				change_prot_numa(vma, start, endvma);
 			goto next;
 		}

Peter Zijlstra

Acked-by on v1 1/1

2014-10-02 19:59 UTC #2 lore raw

In reply to Mel Gorman

Quoted 8 lines · Mel Gorman

PROT_NUMA VMAs are skipped to avoid problems distinguishing between present, prot_none and special entries. MPOL_MF_LAZY is not visible from userspace since commit a720094ded8c ("mm: mempolicy: Hide MPOL_NOOP and MPOL_MF_LAZY from userspace for now") but it should still skip VMAs the same way task_numa_work does.
Signed-off-by Mel Gorman <[email protected]>
Acked-by Rik van Riel <[email protected]>

Acked-by Peter Zijlstra <[email protected]>

Quoted 19 lines

---
 mm/mempolicy.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

mm/mempolicy.c +3 -1

index 8f5330d..a5877ce 100644
@@ -683,7 +683,9 @@ queue_pages_range(struct mm_struct *mm, unsigned long start, unsigned long end,
 		}
 
 		if (flags & MPOL_MF_LAZY) {
-			change_prot_numa(vma, start, endvma);
+			/* Similar to task_numa_work, skip inaccessible VMAs */
+			if (vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE))
+				change_prot_numa(vma, start, endvma);
 			goto next;
 		}