linux/arch
Mel Gorman 32b154c0b0 x86: ignore VM_LOCKED when determining if hugetlb-backed page tables can be shared or not
Addresses http://bugzilla.kernel.org/show_bug.cgi?id=13302

On x86 and x86-64, it is possible that page tables are shared beween
shared mappings backed by hugetlbfs.  As part of this,
page_table_shareable() checks a pair of vma->vm_flags and they must match
if they are to be shared.  All VMA flags are taken into account, including
VM_LOCKED.

The problem is that VM_LOCKED is cleared on fork().  When a process with a
shared memory segment forks() to exec() a helper, there will be shared
VMAs with different flags.  The impact is that the shared segment is
sometimes considered shareable and other times not, depending on what
process is checking.

What happens is that the segment page tables are being shared but the
count is inaccurate depending on the ordering of events.  As the page
tables are freed with put_page(), bad pmd's are found when some of the
children exit.  The hugepage counters also get corrupted and the Total and
Free count will no longer match even when all the hugepage-backed regions
are freed.  This requires a reboot of the machine to "fix".

This patch addresses the problem by comparing all flags except VM_LOCKED
when deciding if pagetables should be shared or not for hugetlbfs-backed
mapping.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: <stable@kernel.org>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: <starlight@binnacle.cx>
Cc: Eric B Munson <ebmunson@us.ibm.com>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Andy Whitcroft <apw@canonical.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-05-29 08:40:03 -07:00
..
alpha alpha: unbreak percpu again 2009-05-12 14:11:35 -07:00
arm flat: fix data sections alignment 2009-05-29 08:40:02 -07:00
avr32 avr32: drop unused CLEAN_FILES 2009-05-01 10:54:00 +02:00
blackfin flat: fix data sections alignment 2009-05-29 08:40:02 -07:00
cris CRISv32: Fix typo compile error in ARTPEC-3 gpio driver. 2009-04-28 19:08:10 +02:00
frv FRV: Use __INIT macro instead of .text.init. 2009-04-27 19:46:30 -07:00
h8300 flat: fix data sections alignment 2009-05-29 08:40:02 -07:00
ia64 [IA64] xen_domu_defconfig: fix build issues/warnings 2009-05-05 11:43:13 -07:00
m32r flat: fix data sections alignment 2009-05-29 08:40:02 -07:00
m68k flat: fix data sections alignment 2009-05-29 08:40:02 -07:00
m68knommu Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gerg/m68knommu 2009-04-24 08:45:53 -07:00
microblaze microblaze: Fix kind-of-intr checking against number of interrupts 2009-05-18 14:47:42 +02:00
mips MIPS: IP32: Remove unnecessary if not even harmful volatile keywords. 2009-05-22 13:52:06 +01:00
mn10300 mn10300: convert to use __HEAD and HEAD_TEXT macros. 2009-04-26 09:20:38 -07:00
parisc Merge git://git.kernel.org/pub/scm/linux/kernel/git/kyle/parisc-2.6 2009-04-03 09:52:04 -07:00
powerpc powerpc: Fix up dma_alloc_coherent() on platforms without cache coherency. 2009-05-27 16:33:59 +10:00
s390 s390: convert to use __HEAD and HEAD_TEXT macros. 2009-04-26 09:20:39 -07:00
sh flat: fix data sections alignment 2009-05-29 08:40:02 -07:00
sparc sparc: cleanup references to deprecated .text.init* sections. 2009-04-27 19:51:58 -07:00
um uml: kill a kconfig warning 2009-04-21 13:41:50 -07:00
x86 x86: ignore VM_LOCKED when determining if hugetlb-backed page tables can be shared or not 2009-05-29 08:40:03 -07:00
xtensa xtensa: Fix linker script to include .literal sections 2009-05-11 23:40:33 -07:00
.gitignore
Kconfig mutex: have non-spinning mutexes on s390 by default 2009-04-09 19:28:24 +02:00