]> pilppa.com Git - linux-2.6-omap-h63xx.git/commit
[PATCH] hugetlb: preserve hugetlb pte dirty state
authorKen Chen <kenchen@google.com>
Thu, 8 Feb 2007 22:20:27 +0000 (14:20 -0800)
committerLinus Torvalds <torvalds@woody.linux-foundation.org>
Fri, 9 Feb 2007 17:25:46 +0000 (09:25 -0800)
commit6649a3863232eb2e2f15ea6c622bd8ceacf96d76
treec3b77d20afd1e7215186244375f6cdcaad94d4b2
parentf336953bfdee8d5e7f69cb8e080704546541f04b
[PATCH] hugetlb: preserve hugetlb pte dirty state

__unmap_hugepage_range() is buggy that it does not preserve dirty state of
huge_pte when unmapping hugepage range.  It causes data corruption in the
event of dop_caches being used by sys admin.  For example, an application
creates a hugetlb file, modify pages, then unmap it.  While leaving the
hugetlb file alive, comes along sys admin doing a "echo 3 >
/proc/sys/vm/drop_caches".

drop_pagecache_sb() will happily free all pages that aren't marked dirty if
there are no active mapping.  Later when application remaps the hugetlb
file back and all data are gone, triggering catastrophic flip over on
application.

Not only that, the internal resv_huge_pages count will also get all messed
up.  Fix it up by marking page dirty appropriately.

Signed-off-by: Ken Chen <kenchen@google.com>
Cc: "Nish Aravamudan" <nish.aravamudan@gmail.com>
Cc: Adam Litke <agl@us.ibm.com>
Cc: David Gibson <david@gibson.dropbear.id.au>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: <stable@kernel.org>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
fs/hugetlbfs/inode.c
mm/hugetlb.c