[PATCH 3.13 150/162] mm, thp: fix collapsing of hugepages on madvise

From: Kamal Mostafa
Date: Thu Nov 06 2014 - 17:40:07 EST


3.13.11.11 -stable review patch. If anyone has any objections, please let me know.

------------------

From: David Rientjes <rientjes@xxxxxxxxxx>

commit 6d50e60cd2edb5a57154db5a6f64eef5aa59b751 upstream.

If an anonymous mapping is not allowed to fault thp memory and then
madvise(MADV_HUGEPAGE) is used after fault, khugepaged will never
collapse this memory into thp memory.

This occurs because the madvise(2) handler for thp, hugepage_madvise(),
clears VM_NOHUGEPAGE on the stack and it isn't stored in vma->vm_flags
until the final action of madvise_behavior(). This causes the
khugepaged_enter_vma_merge() to be a no-op in hugepage_madvise() when
the vma had previously had VM_NOHUGEPAGE set.

Fix this by passing the correct vma flags to the khugepaged mm slot
handler. There's no chance khugepaged can run on this vma until after
madvise_behavior() returns since we hold mm->mmap_sem.

It would be possible to clear VM_NOHUGEPAGE directly from vma->vm_flags
in hugepage_advise(), but I didn't want to introduce special case
behavior into madvise_behavior(). I think it's best to just let it
always set vma->vm_flags itself.

Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx>
Reported-by: Suleiman Souhlal <suleiman@xxxxxxxxxx>
Cc: "Kirill A. Shutemov" <kirill.shutemov@xxxxxxxxxxxxxxx>
Cc: Andrea Arcangeli <aarcange@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
[ kamal: backport to 3.13: no VM_BUG_ON_VMA ]
Signed-off-by: Kamal Mostafa <kamal@xxxxxxxxxxxxx>
---
include/linux/khugepaged.h | 17 ++++++++++-------
mm/huge_memory.c | 11 ++++++-----
mm/mmap.c | 8 ++++----
3 files changed, 20 insertions(+), 16 deletions(-)

diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h
index 6b394f0..eeb3079 100644
--- a/include/linux/khugepaged.h
+++ b/include/linux/khugepaged.h
@@ -6,7 +6,8 @@
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
extern int __khugepaged_enter(struct mm_struct *mm);
extern void __khugepaged_exit(struct mm_struct *mm);
-extern int khugepaged_enter_vma_merge(struct vm_area_struct *vma);
+extern int khugepaged_enter_vma_merge(struct vm_area_struct *vma,
+ unsigned long vm_flags);

#define khugepaged_enabled() \
(transparent_hugepage_flags & \
@@ -35,13 +36,13 @@ static inline void khugepaged_exit(struct mm_struct *mm)
__khugepaged_exit(mm);
}

-static inline int khugepaged_enter(struct vm_area_struct *vma)
+static inline int khugepaged_enter(struct vm_area_struct *vma,
+ unsigned long vm_flags)
{
if (!test_bit(MMF_VM_HUGEPAGE, &vma->vm_mm->flags))
if ((khugepaged_always() ||
- (khugepaged_req_madv() &&
- vma->vm_flags & VM_HUGEPAGE)) &&
- !(vma->vm_flags & VM_NOHUGEPAGE))
+ (khugepaged_req_madv() && (vm_flags & VM_HUGEPAGE))) &&
+ !(vm_flags & VM_NOHUGEPAGE))
if (__khugepaged_enter(vma->vm_mm))
return -ENOMEM;
return 0;
@@ -54,11 +55,13 @@ static inline int khugepaged_fork(struct mm_struct *mm, struct mm_struct *oldmm)
static inline void khugepaged_exit(struct mm_struct *mm)
{
}
-static inline int khugepaged_enter(struct vm_area_struct *vma)
+static inline int khugepaged_enter(struct vm_area_struct *vma,
+ unsigned long vm_flags)
{
return 0;
}
-static inline int khugepaged_enter_vma_merge(struct vm_area_struct *vma)
+static inline int khugepaged_enter_vma_merge(struct vm_area_struct *vma,
+ unsigned long vm_flags)
{
return 0;
}
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 310f27e..65aa131 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -788,7 +788,7 @@ int do_huge_pmd_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
return VM_FAULT_FALLBACK;
if (unlikely(anon_vma_prepare(vma)))
return VM_FAULT_OOM;
- if (unlikely(khugepaged_enter(vma)))
+ if (unlikely(khugepaged_enter(vma, vma->vm_flags)))
return VM_FAULT_OOM;
if (!(flags & FAULT_FLAG_WRITE) &&
transparent_hugepage_use_zero_page()) {
@@ -1993,7 +1993,7 @@ int hugepage_madvise(struct vm_area_struct *vma,
* register it here without waiting a page fault that
* may not happen any time soon.
*/
- if (unlikely(khugepaged_enter_vma_merge(vma)))
+ if (unlikely(khugepaged_enter_vma_merge(vma, *vm_flags)))
return -ENOMEM;
break;
case MADV_NOHUGEPAGE:
@@ -2094,7 +2094,8 @@ int __khugepaged_enter(struct mm_struct *mm)
return 0;
}

-int khugepaged_enter_vma_merge(struct vm_area_struct *vma)
+int khugepaged_enter_vma_merge(struct vm_area_struct *vma,
+ unsigned long vm_flags)
{
unsigned long hstart, hend;
if (!vma->anon_vma)
@@ -2106,11 +2107,11 @@ int khugepaged_enter_vma_merge(struct vm_area_struct *vma)
if (vma->vm_ops)
/* khugepaged not yet working on file or special mappings */
return 0;
- VM_BUG_ON(vma->vm_flags & VM_NO_THP);
+ VM_BUG_ON(vm_flags & VM_NO_THP);
hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK;
hend = vma->vm_end & HPAGE_PMD_MASK;
if (hstart < hend)
- return khugepaged_enter(vma);
+ return khugepaged_enter(vma, vm_flags);
return 0;
}

diff --git a/mm/mmap.c b/mm/mmap.c
index 546db74..32dbf6d 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1046,7 +1046,7 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm,
end, prev->vm_pgoff, NULL);
if (err)
return NULL;
- khugepaged_enter_vma_merge(prev);
+ khugepaged_enter_vma_merge(prev, vm_flags);
return prev;
}

@@ -1065,7 +1065,7 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm,
next->vm_pgoff - pglen, NULL);
if (err)
return NULL;
- khugepaged_enter_vma_merge(area);
+ khugepaged_enter_vma_merge(area, vm_flags);
return area;
}

@@ -2155,7 +2155,7 @@ int expand_upwards(struct vm_area_struct *vma, unsigned long address)
}
}
vma_unlock_anon_vma(vma);
- khugepaged_enter_vma_merge(vma);
+ khugepaged_enter_vma_merge(vma, vma->vm_flags);
validate_mm(vma->vm_mm);
return error;
}
@@ -2224,7 +2224,7 @@ int expand_downwards(struct vm_area_struct *vma,
}
}
vma_unlock_anon_vma(vma);
- khugepaged_enter_vma_merge(vma);
+ khugepaged_enter_vma_merge(vma, vma->vm_flags);
validate_mm(vma->vm_mm);
return error;
}
--
1.9.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/