lkml.org 
[lkml]   [2009]   [Sep]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[PATCH 6/8] mm: fix anonymous dirtying
do_anonymous_page() has been wrong to dirty the pte regardless.
If it's not going to mark the pte writable, then it won't help
to mark it dirty here, and clogs up memory with pages which will
need swap instead of being thrown away. Especially wrong if no
overcommit is chosen, and this vma is not yet VM_ACCOUNTed -
we could exceed the limit and OOM despite no overcommit.

Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: stable@kernel.org
---
mm/memory.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
--- mm5/mm/memory.c 2009-09-07 13:16:46.000000000 +0100
+++ mm6/mm/memory.c 2009-09-07 13:16:53.000000000 +0100
@@ -2608,7 +2608,8 @@ static int do_anonymous_page(struct mm_s
goto oom_free_page;

entry = mk_pte(page, vma->vm_page_prot);
- entry = maybe_mkwrite(pte_mkdirty(entry), vma);
+ if (vma->vm_flags & VM_WRITE)
+ entry = pte_mkwrite(pte_mkdirty(entry));

page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
if (!pte_none(*page_table))

\
 
 \ /
  Last update: 2009-09-07 23:41    [from the cache]
©2003-2011 Jasper Spaans