The name of brd block device is "ramdisk", it's not "brd".
(The block device is registered by register_blkdev(RAMDISK_MAJOR, "ramdisk")
So it should be unregistered by unregister_blkdev(RAMDISK_MAJOR, "ramdisk")
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Acked-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We leak the memory allocated for the nbd_dev array at multiple places.
Fix them by either adding a kfree() or by rearranging code to return
before we allocate the memory.
Signed-off-by: Sven Wegener <sven.wegener@stealer.net>
Cc: Paul Clements <paul.clements@steeleye.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mminit_loglevel is now used from mminit_verify_zonelist <- build_all_zonelists <-
1. online_pages <- memory_block_action <- memory_block_change_state <- store_mem_state (sys handler)
2. numa_zonelist_order_handler (proc handler)
so it cannot be annotated __meminit - drop it
fixes following section mismatch warning:
WARNING: vmlinux.o(.text+0x71628): Section mismatch in reference from the function mminit_verify_zonelist() to the variable .meminit.data:mminit_loglevel
The function mminit_verify_zonelist() references
the variable __meminitdata mminit_loglevel.
This is often because mminit_verify_zonelist lacks a __meminitdata
annotation or the annotation of mminit_loglevel is wrong.
Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Adjust <Alt><SysRq>m show_swap_cache_info() to show "Free swap" as a
signed long: the signed format is preferable, because during swapoff
nr_swap_pages can legitimately go negative, so makes more sense thus
(it used to be shown redundantly, once as signed and once as unsigned).
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add a comment to s390's page_test_dirty/page_clear_dirty/page_set_dirty
dance in page_remove_rmap(): I was wrong to think the PageSwapCache test
could be avoided, and would like a comment in there to remind me. And
mention s390, to help us remember that this block is not really common.
Also move down the "It would be tidy to reset PageAnon" comment: it does
not belong to s390's block, and it would be unwise to reset PageAnon
before we're done with testing it.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
"bootmem_debug" is not mentioned in kernel-parameters.txt. Recently I
had to use that kernel option and I think it should be documented.
Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
eeepc-laptop uses the hwmon struct after unregistering the device, causing
an oops on module unload. Flip the ordering to fix.
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Cc: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Cc: Corentin Chary <corentincj@iksaif.net>
Cc: Karol Kozimor <sziwan@users.sourceforge.net>
Cc: <stable@kernel.org> [2.6.26.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix
arch/arm/mach-pxa/pxa300.c:94: error: 'CKEN_MMC3' undeclared here (not in a function)
when building for PXA300.
Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Acked-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Purely cosmetic for now, but we might as well get it merged ASAP.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
I won't say 'fix', because they still look broken, although this will at
least allow 'make ARCH=CRIS headers_install' to _complete_.
For headers which are exported, we should probably choose between
asm/arch-v10 and asm/arch-v32 by something that GCC defines -- we can't
rely on a generated symlink. And we certainly can't export an arch/
directory which doesn't even exist.
And the only thing that we seem to include from the arch/ directory is
<asm/arch/ptrace.h> from <asm/ptrace.h> ... and that isn't exported in
either arch-v10 or arch-v32 _anyway_.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc:
powerpc: Fix vio_bus_probe oops on probe error
powerpc/ibmebus: Restore "name" sysfs attribute on ibmebus devices
powerpc: Fix /dev/oldmem interface for kdump
powerpc/spufs: Remove invalid semicolon after if statement
powerpc/spufs: reference context while dropping state mutex in scheduler
powerpc/spufs: fix npc setting for NOSCHED contexts
* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6: (22 commits)
[SCSI] ibmvfc: Driver version 1.0.2
[SCSI] ibmvfc: Add details to async event log
[SCSI] ibmvfc: Sanitize response lengths
[SCSI] ibmvfc: Fix for lost async events
[SCSI] ibmvfc: Fixup host state during reinit
[SCSI] ibmvfc: Fix another hang on module removal
[SCSI] ibmvscsi: Fixup desired DMA value for shared memory partitions
[SCSI] megaraid_sas: remove sysfs dbg_lvl world writeable permissions
[SCSI] qla2xxx: Update version number to 8.02.01-k7.
[SCSI] qla2xxx: Explicitly tear-down vports during PCI remove_one().
[SCSI] qla2xxx: Reference proper ha during SBR handling.
[SCSI] qla2xxx: Set npiv_supported flag for FCoE HBAs.
[SCSI] qla2xxx: Don't leak SG-DMA mappings while aborting commands.
[SCSI] qla2xxx: Correct vport-state management issues during ISP-ABORT.
[SCSI] qla2xxx: Correct synchronization of software/firmware fcport states.
[SCSI] scsi_dh: Initialize lun_state in check_ownership()
[SCSI] scsi_dh: Do not use scsilun in rdac hardware handler
[SCSI] megaraid_sas: version and Documentation Update
[SCSI] megaraid_sas: add new controllers (0x78 0x79)
[SCSI] megaraid_sas: add the shutdown DCMD cmd to driver shutdown routine
...
There was another FAT BKL conversion deadlock reported by Bart
Trojanowski due to the BKL being used as a recursive lock by FAT, which
was missed because it only triggers with 'sync' (or 'dirsync') mounts.
The recursion worked for the BKL, but after the conversion to lock_super
(which uses a mutex), it just deadlocks.
Thanks to Bart for debugging this and testing the fix. The lock
debugging information from the original report:
=============================================
[ INFO: possible recursive locking detected ]
2.6.27-rc3-bisect-00448-ga7f5aaf #16
---------------------------------------------
mv/4020 is trying to acquire lock:
(&type->s_lock_key#9){--..}, at: [<c01a90fe>] lock_super+0x1e/0x20
but task is already holding lock:
(&type->s_lock_key#9){--..}, at: [<c01a90fe>] lock_super+0x1e/0x20
other info that might help us debug this:
3 locks held by mv/4020:
#0: (&sb->s_type->i_mutex_key#9/1){--..}, at: [<c01b2336>] do_unlinkat+0x66/0x140
#1: (&sb->s_type->i_mutex_key#9){--..}, at: [<c01b0954>] vfs_unlink+0x84/0x110
#2: (&type->s_lock_key#9){--..}, at: [<c01a90fe>] lock_super+0x1e/0x20
stack backtrace:
Pid: 4020, comm: mv Not tainted 2.6.27-rc3-bisect-00448-ga7f5aaf #16
[<c014e694>] validate_chain+0x984/0xea0
[<c0108d70>] ? native_sched_clock+0x0/0xf0
[<c014ee9c>] __lock_acquire+0x2ec/0x9b0
[<c014f5cf>] lock_acquire+0x6f/0x90
[<c01a90fe>] ? lock_super+0x1e/0x20
[<c044e5fd>] mutex_lock_nested+0xad/0x300
[<c01a90fe>] ? lock_super+0x1e/0x20
[<c01a90fe>] ? lock_super+0x1e/0x20
[<c01a90fe>] lock_super+0x1e/0x20
[<f8b3a700>] fat_write_inode+0x60/0x2b0 [fat]
[<c0450878>] ? _spin_unlock_irqrestore+0x48/0x80
[<f8b3a953>] ? fat_sync_inode+0x3/0x20 [fat]
[<f8b3a962>] fat_sync_inode+0x12/0x20 [fat]
[<f8b37c7e>] fat_remove_entries+0xbe/0x120 [fat]
[<f8b422ef>] vfat_unlink+0x5f/0x90 [vfat]
[<f8b42290>] ? vfat_unlink+0x0/0x90 [vfat]
[<c01b0968>] vfs_unlink+0x98/0x110
[<c01b2400>] do_unlinkat+0x130/0x140
[<c016a8f5>] ? audit_syscall_entry+0x105/0x150
[<c01b253b>] sys_unlinkat+0x3b/0x40
[<c01040d3>] sysenter_do_call+0x12/0x3f
=======================
where the deadlock is due to the nesting of lock_super from vfat_unlink
to fat_write_inode:
- do_unlinkat
- vfs_unlink
- vfat_unlink
* lock_super
- fat_remove_entries
- fat_sync_inode
- fat_write_inode
* lock_super
and the fix is to simply remove the use of lock_super() in fat_write_inode.
The lock_super() there had been just an automatic conversion of the
kernel lock to the superblock lock, but no locking was actually needed
there, since the code in fat_write_inode already protected all relevant
accesses with a spinlock (sbi->inode_hash_lock to be exact). The only
code inside the BKL (and thus the superblock lock) was accesses tp local
variables or calls to functions that have long been SMP-safe (i.e.
sb_bread, mark_buffe_dirty and brlese).
Bart reports:
"Looks good. I ran 10 parallel processes creating 1M files truncating
them, writing to them again and then deleting them. This patch fixes
the issue I ran into.
Signed-off-by: Bart Trojanowski <bart@jukie.net>"
Reported-and-tested-by: Bart Trojanowski <bart@jukie.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Vegard Nossum reported oprofile + hibernation problems:
> Now some warnings:
>
> ------------[ cut here ]------------
> WARNING: at /uio/arkimedes/s29/vegardno/git-working/linux-2.6/kernel/smp.c:328 s
> mp_call_function_mask+0x194/0x1a0()
The usual problem: the suspend function when interrupts are
already disabled calls smp_call_function which is not allowed with
interrupt off. But at this point all the other CPUs should be already
down anyways, so it should be enough to just drop that.
This patch should fix that problem at least by fixing cpu hotplug&
suspend support.
[ mingo@elte.hu: fixed 5 coding style errors. ]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Tested-by: Vegard Nossum <vegard.nossum@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The UV TLB shootdown mechanism needs a system interrupt vector.
Its vector had been hardcoded as 200, but needs to moved to the reserved
system vector range so that it does not collide with some device vector.
This is still temporary until dynamic system IRQ allocation is provided.
But it will be needed when real UV hardware becomes available and runs 2.6.27.
Signed-off-by: Cliff Wickman <cpw@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Rene Herman reported significant Xorg startup/shutdown slowdown due
to PAT. It turns out that the memtype list has thousands of entries.
Add cached_entry to list add routine, in order to speed up the
lookup for sequential reserve_memtype calls.
Reported-by: Rene Herman <rene.herman@keyaccess.nl>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cyrix MediaGXm/Cx5530 Unicorn Revision 1.19.3B has stopped
booting starting at v2.6.22.
The reason is this commit:
> commit f25f64ed5b
> Author: Juergen Beisert <juergen@kreuzholzen.de>
> Date: Sun Jul 22 11:12:38 2007 +0200
>
> x86: Replace NSC/Cyrix specific chipset access macros by inlined functions.
this commit activated a macro which was dormant before due to (buggy)
macro side-effects.
I've looked through various datasheets and found that the GXm and GXLV
Geode processors don't have an incrementor.
Remove the incrementor setup entirely. As the incrementor value
differs according to clock speed and we would hope that the BIOS
configures it correctly, it is probably the right solution.
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
I outwitted myself again in commit 2b2a1ff64a,
and broke the SA_NOCLDWAIT behavior so it leaks zombies. This fixes it.
Reported-by: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Roland McGrath <roland@redhat.com>
When CMO is enabled and booted on a non CMO system and the VIO
device's probe function fails, an oops can result since
vio_cmo_bus_remove is called when it should not. This fixes it by
avoiding the vio_cmo_bus_remove call on platforms that don't implement
CMO.
cpu 0x0: Vector: 300 (Data Access) at [c00000000e13b3d0]
pc: c000000000020d34: .vio_cmo_bus_remove+0xc0/0x1f4
lr: c000000000020ca4: .vio_cmo_bus_remove+0x30/0x1f4
sp: c00000000e13b650
msr: 8000000000009032
dar: 0
dsisr: 40000000
current = 0xc00000000e0566c0
paca = 0xc0000000006f9b80
pid = 2428, comm = modprobe
enter ? for help
[c00000000e13b6e0] c000000000021d94 .vio_bus_probe+0x2f8/0x33c
[c00000000e13b7a0] c00000000029fc88 .driver_probe_device+0x13c/0x200
[c00000000e13b830] c00000000029fdac .__driver_attach+0x60/0xa4
[c00000000e13b8c0] c00000000029f050 .bus_for_each_dev+0x80/0xd8
[c00000000e13b980] c00000000029f9ec .driver_attach+0x28/0x40
[c00000000e13ba00] c00000000029f630 .bus_add_driver+0xd4/0x284
[c00000000e13baa0] c0000000002a01bc .driver_register+0xc4/0x198
[c00000000e13bb50] c00000000002168c .vio_register_driver+0x40/0x5c
[c00000000e13bbe0] d0000000003b3f1c .ibmvfc_module_init+0x70/0x109c [ibmvfc]
[c00000000e13bc70] c0000000000acf08 .sys_init_module+0x184c/0x1a10
[c00000000e13be30] c000000000008748 syscall_exit+0x0/0x40
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Recent of_platform changes made of_bus_type_init() overwrite the bus
type's .dev_attrs list, meaning that the "name" attribute that ibmebus
devices previously had is no longer present. This is a user-visible
regression which breaks the userspace eHCA support, since the eHCA
userspace driver relies on the name attribute to check for valid
adapters.
This fixes it by providing the "name" attribute in the generic OF
device code instead. Tested on POWER.
Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
A change to __ioremap() broke reading /dev/oldmem because we're no
longer able to ioremap pfn 0 (d177c207, "[PATCH] powerpc: IOMMU: don't
ioremap null addresses").
We actually don't need to ioremap for anything that's part of the linear
mapping, so just read it directly.
Also make sure we're only reading one page or less at a time.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Sachin Sant <sachinp@in.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Commit c8c2afe3 ("IPoIB: Use rtnl lock/unlock when changing device
flags") added a call to rtnl_lock() in ipoib_mcast_join_task(), which
is run from the ipoib_workqueue. However, ipoib_stop() (which is run
inside rtnl_lock()) flushes this workqueue, which leads to a deadlock
if the join task is pending.
Fix this by simply not flushing the workqueue from ipoib_stop(). It
turns out that we really don't care about workqueue tasks running
during or after ipoib_stop(), as long as we make sure to flush the
workqueue before unregistering a netdev.
This fixes <https://bugs.openfabrics.org/show_bug.cgi?id=1114>.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6:
firewire: Kconfig help update
ieee1394: sbp2: let nodemgr retry node updates during bus reset series
ieee1394: don't drop nodes during bus reset series
ieee1394: regression in 2.6.25: updates should happen before probes
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6:
PCI: add acpi_find_root_bridge_handle
PCI: acpi_pcihp: run _OSC on a root bridge
x86/PCI: irq and pci_ids patch for Intel Ibex Peak PCHs
x86/PCI: allow scanning of 255 PCI busses
x86, pci: detect end_bus_number according to acpi/e820 reserved, v2
pci: debug extra pci bus resources
pci: debug extra pci resources range
This reverts commit 34ae7f35a2, which has
been reported to cause a number of problems. During suspend and resume,
it apparently causes a crash in a CPU hotplug notifier to happen,
although the exact details are sketchy because of the inability to get
good traces during the suspend sequence.
See buzilla entries
http://bugzilla.kernel.org/show_bug.cgi?id=11296http://bugzilla.kernel.org/show_bug.cgi?id=11339
for more examples and details.
[ Mark: "Revert the patch for now. I'm still looking into getting a
reliable reproduction and I do not have a fix at this time." ]
Requested-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Mark Langsdorf <mark.langsdorf@amd.com>
Acked-by: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@inux-foundation.org>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
Input: evdev - fix printf() format for sizeof
Input: remove version.h from drivers that don't need it
Input: cobalt_btns - add missing MODULE_LICENSE
Properly handle MSKRB5 by passing sec=mskrb5 to the upcall so that the
spengo blob can be generated appropriately. Also, make
decode_negTokenInit prefer whichever mechanism is first in the list.
Needed for some NetApp servers, and possibly some older
versions of Windows which treat the two KRB5 mechanisms differently.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
cifs_setup_session references pSesInfo->server several times. That
pointer shouldn't change during the life of the function so grab it
once and store it in a local var. This makes the code look a little
cleaner too.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (94 commits)
pkt_sched: Prevent livelock in TX queue running.
Revert "pkt_sched: Add BH protection for qdisc_stab_lock."
Revert "pkt_sched: Protect gen estimators under est_lock."
pkt_sched: remove bogus block (cleanup)
nf_nat: use secure_ipv4_port_ephemeral() for NAT port randomization
netfilter: ctnetlink: sleepable allocation with spin lock bh
netfilter: ctnetlink: fix sleep in read-side lock section
netfilter: ctnetlink: fix double helper assignation for NAT'ed conntracks
netfilter: ipt_addrtype: Fix matching of inverted destination address type
dccp: Fix panic caused by too early termination of retransmission mechanism
pkt_sched: Don't hold qdisc lock over qdisc_destroy().
pkt_sched: Add lockdep annotation for qdisc locks
pkt_sched: Never schedule non-root qdiscs.
removed unused #include <version.h>
rt2x00: Fix txdone_entry_desc_flags
b43: Fix for another Bluetooth Coexistence SPROM Programming error for BCM4306
mac80211: remove kdoc references to IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE
p54u: reset skb's data/tail pointer on requeue
p54: move p54_vdcf_init to the right place.
iwlwifi: fix printk newlines
...
sbp2 was too quick to report .update() to the ieee1394 core as failed.
(Logged as "Failed to reconnect to sbp2 device!".) The core would then
unbind sbp2 from the device.
This is not justified if the .update() failed because another bus reset
happened. We check this and tell the ieee1394 that .update() succeeded,
and the core will call sbp2's .update() for the new bus reset as well.
This improves reconnection/re-login especially on buses with several
disks as they may issue bus resets in close succession when they come
online.
Tested by Damien Benoist.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
nodemgr_node_probe checked for generation increments too late and
therefore prematurely reported nodes as "suspended".
Fixes http://bugzilla.kernel.org/show_bug.cgi?id=11349. Reported and
tested by Damien Benoist.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Regression since commit 73cf60232e,
"ieee1394: use class iteration api": The two loops for (1.) driver
updates and (2.) driver probes were replaced by a single loop with
bogus needs_probe checks. Hence updates and probes were now intermixed,
and especially sbp2 updates (reconnects) held up longer than necessary.
While we fix it, change the needs_probe flag to bool type for clarity.
Tested by Damien Benoist.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
This patch against 2.6.27-rc3 does the following changes to FSG-3 setup code:
1) Enable RTC on FSG-3 (proper name is needed). This change is needed due to a recent change in i2c.
2) i variable is only used when compiling for big endian.
So move its declaration to ARMEB ifdef to silence the warning when compiling for LE.
Mailing list link:
http://lists.arm.linux.org.uk/lurker/message/20080813.091556.cae2917e.en.html
Signed-off-by: Tomasz Chmielewski <mangoo@wpkg.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
commit f2afa7711f ("Input: paper over a bug in
Synaptics X driver") introduced a compiler warning on 64-bit platforms, as
sizeof() returns a size_t, not an (unsigned) int:
| drivers/input/evdev.c: In function 'handle_eviocgbit':
| drivers/input/evdev.c:684: warning: format '%d' expects type 'int', but argument 3 has type 'long unsigned int'
Use the proper `z' modifier for size_t, and make the printf() formats for the
sizes unsigned while we're at it.
Signed-off-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
If dev_deactivate() is trying to quiesce the queue, it
is theoretically possible for another cpu to livelock
trying to process that queue. This happens because
dev_deactivate() grabs the queue spinlock as it checks
the queue state, whereas net_tx_action() does a trylock
and reschedules the qdisc if it hits the lock.
This breaks the livelock by adding a check on
__QDISC_STATE_DEACTIVATED to net_tx_action() when
the trylock fails.
Based upon feedback from Herbert Xu and Jarek Poplawski.
Signed-off-by: David S. Miller <davem@davemloft.net>
