The assertion that checks for sge context overflow is
incorrectly hard-coded to 32. This causes a kernel bug
check when using big-data mounts. Changed the BUG_ON to
use the computed value RPCSVC_MAXPAGES.
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

RDMA connection shutdown on an SMP machine can cause a kernel crash due
to the transport close path racing with the I/O tasklet.

Additional transport references were added as follows:
- A reference when on the DTO Q to avoid having the transport
  deleted while queued for I/O.
- A reference while there is a QP able to generate events.
- A reference until the DISCONNECTED event is received on the CM ID
Signed-off-by: NTom Tucker <tom@opengridcomputing.com>
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fixes:

	block/genhd.c:361: warning: ignoring return value of ‘class_register’, declared with attribute warn_unused_result
Signed-off-by: NRoland McGrath <roland@redhat.com>
Acked-by: NJeff Garzik <jeff@garzik.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It was all wrapped in '#ifdef CONFIG_BLOCK' anyway, so userspace was
getting nothing useful out of it. And the special #ifndef __KERNEL__
version of 'struct partition' makes me inclined to promote an attitude
of violence...

Stick some comments on some of the #endifs too, while we're at it.
Signed-off-by: NDavid Woodhouse <dwmw2@infradead.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Introduced in commit-id 9e2779fa and
ifdef'ed out for nommu in 8ca3ed87, both
approaches end up breaking the nommu build in different ways. An
impressive feat for a 2-liner.

Current is_vmalloc_addr() users fall in to two camps:

	- Determining whether to use vfree()/kfree()
	- Whether to do vmlist traversal (only /proc/kcore).

Since we don't support /proc/kcore on nommu, that leaves the
vfree()/kfree() determination use cases. nommu vfree() happens to be a
wrapper to kfree() anyways, so is_vmalloc_addr() can always return 0
and end up with the right behaviour.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86:
  x86: remove quicklists
  x86: ia32 syscall restart fix
  x86: ioremap, remove WARN_ON()

Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched-devel

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched-devel:
  keep rd->online and cpu_online_map in sync
  Revert "cpu hotplug: adjust root-domain->online span in response to hotplug event"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  RDMA/iwcm: Don't access a cm_id after dropping reference
  IB/iser: Handle iser_device allocation error gracefully
  IB/iser: Fix list iteration bug
  RDMA/cxgb3: Fix iwch_create_cq() off-by-one error
  RDMA/cxgb3: Return correct max_inline_data when creating a QP
  IB/fmr_pool: Flush all dirty FMRs from ib_fmr_pool_flush()
  Revert "IB/fmr_pool: ib_fmr_pool_flush() should flush all dirty FMRs"
  IB/cm: Flush workqueue when removing device
  MAINTAINERS: update ipath owner

quicklists cause a serious memory leak on 32-bit x86,
as documented at:

  http://bugzilla.kernel.org/show_bug.cgi?id=9991

the reason is that the quicklist pool is a special-purpose
cache that grows out of proportion. It is not accounted for
anywhere and users have no way to even realize that it's
the quicklists that are causing RAM usage spikes. It was
supposed to be a relatively small pool, but as demonstrated
by KOSAKI Motohiro, they can grow as large as:

  Quicklists:    1194304 kB

given how much trouble this code has caused historically,
and given that Andrew objected to its introduction on x86
(years ago), the best option at this point is to remove them.

[ any performance benefits of caching constructed pgds should
  be implemented in a more generic way (possibly within the page
  allocator), while still allowing constructed pages to be
  allocated by other workloads. ]
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The code to restart syscalls after signals depends on checking for a
negative orig_ax, and for particular negative -ERESTART* values in ax.
These fields are 64 bits and for a 32-bit task they get zero-extended.
The syscall restart behavior is lost, a regression from a native 32-bit
kernel and from 64-bit tasks' behavior.

This patch fixes the problem by doing sign-extension where it matters.

For orig_ax, the only time the value should be -1 but winds up as
0x0ffffffff is via a 32-bit ptrace call. So the patch changes ptrace to
sign-extend the 32-bit orig_eax value when it's stored; it doesn't
change the checks on orig_ax, though it uses the new current_syscall()
inline to better document the subtle importance of the used of
signedness there.

The ax value is stored a lot of ways and it seems hard to get them all
sign-extended at their origins. So for that, we use the
current_syscall_ret() to sign-extend it only for 32-bit tasks at the
time of the -ERESTART* comparisons.
Signed-off-by: NRoland McGrath <roland@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Signed-off-by: NIngo Molnar <mingo@elte.hu>

It is possible to allow the root-domain cache of online cpus to
become out of sync with the global cpu_online_map.  This is because we
currently trigger removal of cpus too early in the notifier chain.
Other DOWN_PREPARE handlers may in fact run and reconfigure the
root-domain topology, thereby stomping on our own offline handling.

The end result is that rd->online may become out of sync with
cpu_online_map, which results in potential task misrouting.

So change the offline handling to be more tightly coupled with the
global offline process by triggering on CPU_DYING intead of
CPU_DOWN_PREPARE.
Signed-off-by: NGregory Haskins <ghaskins@novell.com>
Cc: Gautham R Shenoy <ego@in.ibm.com>
Cc: "Siddha, Suresh B" <suresh.b.siddha@intel.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This reverts commit 393d94d9.

Lets fix this right.
Signed-off-by: NGregory Haskins <ghaskins@novell.com>
Cc: Gautham R Shenoy <ego@in.ibm.com>
Cc: "Siddha, Suresh B" <suresh.b.siddha@intel.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

cm_work_handler() can access cm_id_priv after it drops its reference
by calling iwch_deref_id(), which might cause it to be freed.  The fix
is to look at whether IWCM_F_CALLBACK_DESTROY is set _before_ dropping
the reference.  Then if it was set, free the cm_id on this thread.
Signed-off-by: NSteve Wise <swise@opengridcomputing.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

"iser_device" allocation failure is "handled" with a BUG_ON() right
before dereferencing the NULL-pointer - fix this!
Signed-off-by: NArne Redlich <arne.redlich@xiranet.com>
Signed-off-by: NErez Zilber <erezz@voltaire.com>

The iteration through the list of "iser_device"s during device
lookup/creation is broken -- it might result in an infinite loop if
more than one HCA is used with iSER.  Fix this by using
list_for_each_entry() instead of the open-coded flawed list iteration
code.
Signed-off-by: NArne Redlich <arne.redlich@xiranet.com>
Signed-off-by: NErez Zilber <erezz@voltaire.com>
Signed-off-by: NRoland Dreier <rolandd@cisco.com>

* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev:
  [libata] Add support for the RB500 PATA CompactFlash
  ahci: logical-bitwise and confusion in ahci_save_initial_config()
  libata: don't allow sysfs read access to force param
  ahci: add the Device IDs for nvidia MCP7B AHCI
  libata-sff: handle controllers w/o ctl register
  libata: allow LLDs w/o any reset method
  ata: replace remaining __FUNCTION__ occurrences

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6:
  USB:Update mailing list information in documentation
  USB: fix ehci unlink regressions
  USB: new ftdi_sio device id
  USB: Remove __KERNEL__ check from non-exported gadget.h.
  USB: g_printer.h does not need to be "unifdef"ed.
  USB: fsl_usb2_udc: fix broken Kconfig
  USB: option: add novatel device ids
  USB: usbaudio: handle kcalloc failure
  USB: cypress_m8: add UPS Powercom (0d9f:0002)
  USB: drivers/usb/storage/sddr55.c: fix uninitialized var warnings
  USB: fix usb-serial generic recursive lock

I figured out another ACPI related regression today.

randconfig testing triggered an early boot-time hang on a laptop of mine
(32-bit x86, config attached) - the screen was scrolling ACPI AML
exceptions [with no serial port and no early debugging available].

v2.6.24 works fine on that laptop with the same .config, so after a few
hours of bisection (had to restart it 3 times - other regressions
interacted), it honed in on this commit:

| 10270d48 is first bad commit
|
| Author: Linus Torvalds <torvalds@woody.linux-foundation.org>
| Date:   Wed Feb 13 09:56:14 2008 -0800
|
|     acpi: fix acpi_os_read_pci_configuration() misuse of raw_pci_read()

reverting this commit ontop of -rc5 gave a correctly booting kernel.

But this commit fixes a real bug so the real question is, why did it
break the bootup?

After quite some head-scratching, the following change stood out:

-                               pci_id->bus = tu8;
+                               pci_id->bus = val;

pci_id->bus is defined as u16:

   struct acpi_pci_id {
           u16 segment;
           u16 bus;
   ...

and 'tu8' changed from u8 to u32. So previously we'd unconditionally
mask the return value of acpi_os_read_pci_configuration()
(raw_pci_read()) to 8 bits, but now we just trust whatever comes back
from the PCI access routines and only crop it to 16 bits.

But if the high 8 bits of that result contains any noise then we'll
write that into ACPI's PCI ID descriptor and confuse the heck out of the
rest of ACPI.

So lets check the PCI-BIOS code on that theory. We have this codepath
for 8-bit accesses (arch/x86/pci/pcbios.c:pci_bios_read()):

        switch (len) {
        case 1:
                __asm__("lcall *(%%esi); cld\n\t"
                        "jc 1f\n\t"
                        "xor %%ah, %%ah\n"
                        "1:"
                        : "=c" (*value),
                          "=a" (result)
                        : "1" (PCIBIOS_READ_CONFIG_BYTE),
                          "b" (bx),
                          "D" ((long)reg),
                          "S" (&pci_indirect));

Aha! The "=a" output constraint puts the full 32 bits of EAX into
*value. But if the BIOS's routines set any of the high bits to nonzero,
we'll return a value with more set in it than intended.

The other, more common PCI access methods (v1 and v2 PCI reads) clear
out the high bits already, for example pci_conf1_read() does:

        switch (len) {
        case 1:
                *value = inb(0xCFC + (reg & 3));

which explicitly converts the return byte up to 32 bits and zero-extends
it.

So zero-extending the result in the PCI-BIOS read routine fixes the
regression on my laptop. ( It might fix some other long-standing issues
we had with PCI-BIOS during the past decade ... ) Both 8-bit and 16-bit
accesses were buggy.
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/pci-2.6:
  PCI Hotplug: Fix small mem leak in IBM Hot Plug Controller Driver
  PCI: rename DECLARE_PCI_DEVICE_TABLE to DEFINE_PCI_DEVICE_TABLE

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-2.6:
  drivers: fix dma_get_required_mask
  firmware: provide stubs for the FW_LOADER=n case
  nozomi: fix initialization and early flow control access
  sysdev: fix problem with sysdev_class being re-registered

* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
  lguest: Do not append space to guests kernel command line
  lguest: Revert 1ce70c4f, fix real problem.
  lguest: Sanitize the lguest clock.
  lguest: fix __get_vm_area usage.
  lguest: make sure cpu is initialized before accessing it

* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2:
  ocfs2: Fix NULL pointer dereferences in o2net
  ocfs2/dlm: dlm_thread should not sleep while holding the dlm_spinlock
  ocfs2/dlm: Print message showing the recovery master
  ocfs2/dlm: Add missing dlm_lockres_put()s
  ocfs2/dlm: Add missing dlm_lockres_put()s in migration path
  ocfs2/dlm: Add missing dlm_lock_put()s
  ocfs2: Fix an endian bug in online resize.
  [PATCH] [OCFS2]: constify function pointer tables
  ocfs2: Fix endian bug in o2dlm protocol negotiation.
  ocfs2: Use dlm_print_one_lock_resource for lock resource print
  [PATCH] fs/ocfs2/dlm/dlmdomain.c: fix printk warning

* git://git.kernel.org/pub/scm/linux/kernel/git/wim/linux-2.6-watchdog:
  [WATCHDOG] make watchdog/hpwdt.c:asminline_call() static
  [WATCHDOG] Remove volatiles from watchdog device structures
  [WATCHDOG] replace remaining __FUNCTION__ occurrences
  [WATCHDOG] hpwdt: Use dmi_walk() instead of own copy
  [WATCHDOG] Fix return value warning in hpwdt
  [WATCHDOG] Fix declaration of struct smbios_entry_point in hpwdt
  [WATCHDOG] it8712f_wdt support for 16-bit timeout values, WDIOC_GETSTATUS

Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NBryan Wu <bryan.wu@analog.com>
Cc: Antonino Daplas <adaplas@pol.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NBryan Wu <bryan.wu@analog.com>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NBryan Wu <bryan.wu@analog.com>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NBryan Wu <bryan.wu@analog.com>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix kernel crash when stifb driver is used with a A1439A CRX (Rattler)
graphics card.  (Reference:
http://thread.gmane.org/gmane.linux.ports.hppa/1834)
Signed-off-by: NHelge Deller <deller@gmx.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix wrong pointer type passed into the dev_dbg() function.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Acked-by: NMike Rapoport <mike@compulab.co.il>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

iov_iter_advance() skips over zero-length iovecs, however it does not properly
terminate at the end of the iovec array.  Fix this by checking against
i->count before we skip a zero-length iov.

The bug was reproduced with a test program that continually randomly creates
iovs to writev.  The fix was also verified with the same program and also it
could verify that the correct data was contained in the file after each
writev.
Signed-off-by: NNick Piggin <npiggin@suse.de>
Tested-by: N"Kevin Coffman" <kwc@citi.umich.edu>
Cc: "Alexey Dobriyan" <adobriyan@gmail.com>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The original preemptible-RCU patch put the choice between classic and
preemptible RCU into kernel/Kconfig.preempt, which resulted in build failures
on machines not supporting CONFIG_PREEMPT.  This choice was therefore moved to
init/Kconfig, which worked, but placed the choice between classic and
preemptible RCU at the top level, a very obtuse choice indeed.

This patch changes from the Kconfig "choice" mechanism to a pair of booleans,
only one of which (CONFIG_PREEMPT_RCU) is user-visible, and is located in
kernel/Kconfig.preempt, where one would expect it to be.  The other
(CONFIG_CLASSIC_RCU) is in init/Kconfig so that it is available to all
architectures, hopefully avoiding build breakage.  Thanks to Roman Zippel for
suggesting this approach.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Acked-by: NSteven Rostedt <rostedt@goodmis.org>
Cc: Dipankar Sarma <dipankar@in.ibm.com>
Cc: Josh Triplett <josh@freedesktop.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Use SGI_HAS_I8042 to select SGI i8042 handling
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Return value convention of module's init functions is 0/-E.  Sometimes,
e.g.  during forward-porting mistakes happen and buggy module created,
where result of comparison "workqueue != NULL" is propagated all the way up
to sys_init_module.  What happens is that some other module created
workqueue in question, our module created it again and module was
successfully loaded.

Or it could be some other bug.

Let's make such mistakes much more visible.  In retrospective, such
messages would noticeably shorten some of my head-scratching sessions.

Note, that dump_stack() is just a way to get attention from user.  Sample
message:

sys_init_module: 'foo'->init suspiciously returned 1, it should follow 0/-E convention
sys_init_module: loading module anyway...
Pid: 4223, comm: modprobe Not tainted 2.6.24-25f66630 #5

Call Trace:
 [<ffffffff80254b05>] sys_init_module+0xe5/0x1d0
 [<ffffffff8020b39b>] system_call_after_swapgs+0x7b/0x80
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit c9a3ba55 (module: wait for dependent modules doing init.) didn't quite
work because the waiter holds the module lock, meaning that the state of the
module it's waiting for cannot change.

Fortunately, it's fairly simple to update the state outside the lock and do
the wakeup.

Thanks to Jan Glauber for tracking this down and testing (qdio and qeth).
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Cc: Jan Glauber <jang@linux.vnet.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Free pages in the hugetlb pool are free and as such have a reference count of
zero.  Regular allocations into the pool from the buddy are "freed" into the
pool which results in their page_count dropping to zero.  However, surplus
pages can be directly utilized by the caller without first being freed to the
pool.  Therefore, a call to put_page_testzero() is in order so that such a
page will be handed to the caller with a correct count.

This has not affected end users because the bad page count is reset before the
page is handed off.  However, under CONFIG_DEBUG_VM this triggers a BUG when
the page count is validated.

Thanks go to Mel for first spotting this issue and providing an initial fix.
Signed-off-by: NAdam Litke <agl@us.ibm.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The pca953x driver is an I2C driver so gpio_chip->can_sleep should be set.
This lets upper layers know they should use the gpio_*_cansleep() calls to
access values, and may not access them from nonsleeping contexts.
Signed-off-by: NArnaud Patard <arnaud.patard@rtp-net.org>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Acked-by: N"eric miao" <eric.y.miao@gmail.com>
Cc: Jean Delvare <khali@linux-fr.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Recent patch titled
  Reduce CPU wastage on idle md array with a write-intent bitmap.

would sometimes leave the array with dirty bitmap bits that stay dirty.  A
subsequent write would sort things out so it isn't a big problem, but should
be fixed nonetheless.

We need to make sure that when the bitmap becomes not "allclean", the
daemon_sleep really does get set to a sensible value.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

If an md array is "auto-read-only", then this appears in /proc/mdstat as

   /dev/md0: active(auto-read-only)

whereas if it is truely readonly, it appears as

   /dev/md0: active (read-only)

The difference being a space.

One program known to parse this file expects the space and gets badly
confused.  It will be fixed, but it would be best if what the kernel generates
is more consistent too.
Signed-off-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Address 3 known bugs in the current memory policy reference counting method.
I have a series of patches to rework the reference counting to reduce overhead
in the allocation path.  However, that series will require testing in -mm once
I repost it.

1) alloc_page_vma() does not release the extra reference taken for
   vma/shared mempolicy when the mode == MPOL_INTERLEAVE.  This can result in
   leaking mempolicy structures.  This is probably occurring, but not being
   noticed.

   Fix:  add the conditional release of the reference.

2) hugezonelist unconditionally releases a reference on the mempolicy when
   mode == MPOL_INTERLEAVE.  This can result in decrementing the reference
   count for system default policy [should have no ill effect] or premature
   freeing of task policy.  If this occurred, the next allocation using task
   mempolicy would use the freed structure and probably BUG out.

   Fix:  add the necessary check to the release.

3) The current reference counting method assumes that vma 'get_policy()'
   methods automatically add an extra reference a non-NULL returned mempolicy.
    This is true for shmem_get_policy() used by tmpfs mappings, including
   regular page shm segments.  However, SHM_HUGETLB shm's, backed by
   hugetlbfs, just use the vma policy without the extra reference.  This
   results in freeing of the vma policy on the first allocation, with reuse of
   the freed mempolicy structure on subsequent allocations.

   Fix: Rather than add another condition to the conditional reference
   release, which occur in the allocation path, just add a reference when
   returning the vma policy in shm_get_policy() to match the assumptions.
Signed-off-by: NLee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Greg KH <greg@kroah.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: David Rientjes <rientjes@google.com>
Cc: <eric.whitney@hp.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>