This patch adds waiting until transmit buffer and shifter will be empty
before clock disabling.
Without this fix it's possible to have clock disabled while data was
not transmited yet, which causes unproper state of TX line and problems
in following data transfers.
Cc: <stable@vger.kernel.org> # v2.6.26+
Signed-off-by: Robert Baldyga <r.baldyga@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Return a negative error code on failure.
A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)
// <smpl>
@@
identifier ret; expression e1,e2;
@@
(
if (\(ret < 0\|ret != 0\))
{ ... return ret; }
|
ret = 0
)
... when != ret = e1
when != &ret
*if(...)
{
... when != ret = e2
when forall
return ret;
}
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The tty break and error flags are not bit masks so do not to use bitwise
OR when assigning them.
Note that there is no functional change due to the if-else construct and
flag having been initialised to zero (TTY_NORMAL).
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The original driver fixed the io address with 0xe000+idx*8,
but real io address assigned from BIOS is dynamically from
read PCI configure space 0x24, 0x20, 0x1c.
The Fintek F81504/F81508/F81512 maybe malfunction without
this patch and malfunction surely when more then 1 PCI card.
Signed-off-by: Peter Hung <hpeter+linux_kernel@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
tx_empty() should test whether both the transmitter fifo and shifter
for the port is empty, ie. the whole last char was transmitted.
The shift register is empty if AUART_STAT_BUSY is cleared.
The patch fixes the function against the shift register.
According to i.MX23 and i.MX28 Reference Manual:
AUART_STAT_TXFE:
TX FIFO or transmit holding register is empty.
AUART_STAT_BUSY:
AUART still transmits bits.
The BUSY signal goes HIGH as soon as the data is written to the
transmit FIFO (that is, the FIFO is non-empty) and remains asserted
HIGH while data is being transmitted.
BUSY is negated only when the transmit FIFO is empty, and the last
character has been transmitted from the shift register, including
the stop bits.
Signed-off-by: Janusz Uzycki <j.uzycki@elproma.com.pl>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
mxs_auart_modem_status() did't detect gpio's state change
because s->mctrl_prev was modified before by mctrl_gpio_get().
The patch introduces mctrl_temp variable to fix the bug.
Signed-off-by: Janusz Uzycki <j.uzycki@elproma.com.pl>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 732a84a037 ("serial: core: Pass termios to set_ldisc() notifications")
changed the set_ldisc prototype.
At the time of this commit the mxs_auart driver did not implement set_ldisc,
so that's why it has not been converted.
Adapt also mxs_auart_set_ldisc() so that the following build warning can be
fixed:
drivers/tty/serial/mxs-auart.c:962:2: warning: initialization from
incompatible pointer type
.set_ldisc = mxs_auart_set_ldisc,
^
drivers/tty/serial/mxs-auart.c:962:2: warning: (near initialization
for 'mxs_auart_ops.set_ldisc')
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Tested-by: Janusz Uzycki <j.uzycki@elproma.com.pl>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Although the existing code appears to work on most hardware, the
hardware designers tell us that 8-bit access to the registers is not
guaranteed to be reliable. Also the OCTEON simulation environments
prohibit 8-bit accesses.
For these reasons, we use __raw_readq/__raw_writeq for OCTEON. This
code is protected with #ifdef CONFIG_64BIT so it still builds under
configurations lacking readq/writeq.
We can get rid of the #ifdef __BIG_ENDIAN, as under 64-bit accesses,
OCTEON is byte order invariant.
Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: Aleksey Makarov <aleksey.makarov@auriga.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When CONFIG_SERIAL_IMX_CONSOLE=n and CONFIG_CONSOLE_POLL=y we get the following
build warning:
drivers/tty/serial/imx.c:306:13: warning: 'imx_port_ucrs_save' defined but not used [-Wunused-function]
drivers/tty/serial/imx.c:315:13: warning: 'imx_port_ucrs_restore' defined but not used [-Wunused-function]
imx_port_ucrs_save/restore are only used under CONFIG_SERIAL_IMX_CONSOLE, so
their definitions should be also be protected only by CONFIG_SERIAL_IMX_CONSOLE.
This was detected when building 'allmodconfig'.
Reported-by: Olof's autobuilder <build@lixom.net>
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
struct bfin_serial_port contains structs circ_buf, timer_list, and
work_struct directly (not pointers). This means that these structures
have to be defined completely to be inlined in struct
bfin_serial_port.
So instead of struct declarations, define the structures by including
proper headers. They were pulled in by linux/serial_core.h or others
until now. But experimenting with circ_buf removal from serial_core,
struct circ_buf becomes undefined in bfin_serial:
In file included from arch/blackfin/kernel/debug-mmrs.c:21:0:
arch/blackfin/include/asm/bfin_serial.h:44:18: error: field 'rx_dma_buf' has incomplete type
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Dan Carpenter reported:
|drivers/tty/serial/8250/8250_omap.c:1025 omap8250_probe()
|warn: unsigned 'up.port.line' is never less than zero.
|1025 if (up.port.line < 0) {
Since of_alias_get_id() and pdev->id can get negative I check for the
error via ret variable.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Dan Carpenter reported:
|drivers/tty/serial/8250/8250_omap.c:1025 omap8250_probe()
|warn: unsigned 'up.port.line' is never less than zero.
|1025 if (up.port.line < 0) {
I (wrongly) assumed that line is an int and compiler didn't complain nor
did sparse. Since of_alias_get_id() and pdev->id can get negative I
check for the error via ret variable.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The mcb_device_id table is supposed to be zero-terminated.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
in CSR A7DA SoC, uart6 located at BT module and it need multiple clock
sources, so for "sirf,marco-bt-uart" compatible uarts, drivers take 3
clock sources and enable them.
this patch also replaces clk_get by devm_clk_get function and fix DT
binding document in which we missed to fix when we added marco platform
in commit 909102db44 "serial: sirf: add support for Marco chip".
Signed-off-by: Qipan Li <Qipan.Li@csr.com>
Signed-off-by: Barry Song <Baohua.Song@csr.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert the pr_warning to the more common pr_warn.
Other miscellanea:
o Convert unusual PR_FMT define and uses to pr_fmt
o Remove unnecessary OOM message
o Fix grammar in an error message
o Convert a pr_warning with a KERN_ERR to pr_err
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit deals with the checkpatch warning "Unnecessary space before
function pointer arguments".
Signed-off-by: Konrad Zapalowicz <bergo.torino@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The changed function flushes the tx UART and the '4' corresponds to the
UART_FCR_CLEAR_XMIT value. This commit replaces the magic number with
this define.
Signed-off-by: Konrad Zapalowicz <bergo.torino@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit fixes the alignment of the 'case's i the switch statement.
Signed-off-by: Konrad Zapalowicz <bergo.torino@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The flow of {neo,cls}_param() shows that at this stage the baud rate
has a non-zero value. This fact makes the if clausule obsolete and
acknowledges it's removal.
Signed-off-by: Konrad Zapalowicz <bergo.torino@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The 'force' parameter to the {cls,neo}_send_break() function has been
removed because it has not been used. The client to this API (the tty
code) always called this function with only one value.
Signed-off-by: Konrad Zapalowicz <bergo.torino@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This changes negative values of error rate to be checked, because these
values are valid as error rate. And this changes in the process of adopting
a value close to 0.
Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Simply document the new compat string (and keep the list sorted by SoC).
There appears to be no need for a driver update.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Simply document the new compat string.
There appears to be no need for a driver update.
Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
[geert: Reworded to match previous commits]
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The iput() function tests whether its argument is NULL and then
returns immediately. Thus the test around the call is not needed.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
According with Malcolm, the missing breaks are intentional.
So, let's revert commit d442b15fb4,
add some comments to document it and fix the two smatch warnings:
drivers/media/usb/dvb-usb-v2/lmedm04.c:828 lme_firmware_switch() warn: missing break? reassigning 'st->dvb_usb_lme2510_firmware'
drivers/media/usb/dvb-usb-v2/lmedm04.c:850 lme_firmware_switch() warn: missing break? reassigning 'st->dvb_usb_lme2510_firmware'
using a different strategy to avoid reassign values to
st->dvb_usb_lme2510_firmware.
Acked-by: Malcolm Priestley <tvboxspy@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
The drm_prime_pages_to_sg() function never returns NULL pointers, only
error pointers and valid pointers.
Signed-off-by: Haixia Shi <hshi@chromium.org>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Fixes dmabuf export failure with -E_NOMEM when the page is not mapped.
Signed-off-by: Haixia Shi <hshi@chromium.org>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Add myself as the maintainer of the i.MX DRM driver.
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Acked-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The imx-drm driver was put into staging mostly for the following reasons,
all of which have been addressed or superseded:
- convert the irq driver to use linear irq domains
- work out the device tree bindings, this lead to the common of_graph
bindings being used
- factor out common helper functions, this mostly resulted in the
component framework and drm of_graph helpers.
Before adding new fixes, and certainly before adding new features,
move it into its proper place below drivers/gpu/drm.
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
* Enable DU using DT on marzen/r8a7779, lager/r8a7790 and koelsch/r8a7791
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=m0Ov
-----END PGP SIGNATURE-----
Merge tag 'tags/renesas-dt-du-for-v3.19' into drm/next/adv7511-base
Renesas ARM Based SoC DT DU Updates for v3.19
* Enable DU using DT on marzen/r8a7779, lager/r8a7790 and koelsch/r8a7791
Add Exynos4415 SoC support, some fixups and cleanups.
Summary:
- Resolve kernel lockup issue incurred by probe request in probe context.
. For this, it moves all register codes of sub drivers into init function
and adds component binding support for vidi driver.
- Add Exynos4415 SoC support.
- Make each manager and display object to be embedded
in each driver context.
- Fix and clean up FIMD and MIPI-DSI drivers.
- Clean up unnecesary or wrong descriptions.
- And trivial cleanups.
* 'exynos-drm-next' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos: (58 commits)
drm/exynos: avoid leak if exynos_dpi_probe() fails
drm/exynos: Fix exynos_dpi_remove() parameter
drm/exynos: vidi: add component support
drm/exynos: fix exynos_drm_component_del
drm/exynos/ipp: fix error return code
drm/exynos: clean up machine compatible string check
drm/exynos: move Exynos platform drivers registration to init
Revert "drm/exynos: fix null pointer dereference issue"
drm/exynos/dpi: stop using display->ctx pointer
drm/exynos/dpi: embed display into private context
drm/exynos/dp: stop using display->ctx pointer
drm/exynos/dp: embed display into private context
drm/exynos/vidi: stop using display->ctx pointer
drm/exynos/vidi: embed display into private context
drm/exynos/hdmi: stop using display->ctx pointer
drm/exynos/hdmi: embed display into private context
drm/exynos/fimd: stop using manager->ctx pointer
drm/exynos/fimd: embed manager into private context
drm/exynos/vidi: stop using manager->ctx pointer
drm/exynos/vidi: embed manager into private context
...
When a Greybus message is sent, the host driver supplies a cookie
for Greybus to use to identify the sent message in the event it
needs to be canceled. The cookie will be non-null while the message
is in flight, and a null pointer otherwise.
There are two problems with this, which arise out of the fact that a
message can be canceled at any time--even concurrent with it getting
sent (such as when Greybus is getting shut down).
First, the host driver's buffer_send method can return an error
value, which is non-null but not a valid cookie. So we need to
ensure such a bogus cookie is never used to cancel a message.
Second, we can't resolve that problem by assigning message->cookie
only after we've determined it's not an error. The instant
buffer_send() returns, the message may well be in flight and *should*
be canceled at shutdown, so we need the cookie value to reflect
that.
In order to avoid these problems, protect access to a message's
cookie value with a mutex. A spin lock can't be used because the
window that needs protecting covers code that can block. We
reset the cookie value to NULL as soon as the host driver has
notified us it has been sent (or failed to).
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>
It's possible for an in-flight buffer to be recorded as sent *after*
a thread has begin the process of canceling it. In that case the
Greybus message cookie will be set to NULL, and that value can end
up getting passed to buffer_cancel(). Change buffer_cancel() so
it properly handles (ignores) a null cookie pointer.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>
An operation result can be set both in and out of interrupt context.
For example, a response message could be arriving at the same time a
timeout of the operation is getting processed. We therefore need to
ensure the result is accessed atomically.
Protect updates to the errno field using the operations spinlock.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>
It's incorrect to assign a signed value to an unsigned acpi_status
variable. This patch fixes it by using a signed integer to return
error values.
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Few elements in the acpi_processor_power structure are unused. It could
be remnant in the header missed while the code got removed from the
corresponding driver file.
This patch removes those unused variables in the structure declaration.
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
As Russell King's explained it, there should not be pointers to
struct device_node:
"struct device_node is a ref-counted structure. That means if you
store a reference to it, you should "get" it, and you should "put"
it once you've done. The act of "put"ing the pointed-to structure
involves writing to that structure, so it is totally unappropriate
to store a device_node structure as a const pointer. It forces you
to have to cast it back to a non-const pointer at various points
in time to use various OF function calls."
[This isn't quite the application here, we're not geting or putting the
pointer though we did add some other users who call non-const OF
functions -- broonie]
Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mark Brown <broonie@kernel.org>
The vfree() function performs also input parameter validation.
Thus the test around the call is not needed.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Do it before it's assigned to cpufreq_cpu_data, otherwise when a driver
tries to get the cpu frequency during initialization the policy kobj is
referenced and we get this warning:
------------[ cut here ]------------
WARNING: CPU: 1 PID: 64 at include/linux/kref.h:47 kobject_get+0x64/0x70()
Modules linked in:
CPU: 1 PID: 64 Comm: irq/77-tegra-ac Not tainted 3.18.0-rc4-next-20141114ccu-00050-g3eff942 #326
[<c0016fac>] (unwind_backtrace) from [<c001272c>] (show_stack+0x10/0x14)
[<c001272c>] (show_stack) from [<c06085d8>] (dump_stack+0x98/0xd8)
[<c06085d8>] (dump_stack) from [<c002892c>] (warn_slowpath_common+0x84/0xb4)
[<c002892c>] (warn_slowpath_common) from [<c00289f8>] (warn_slowpath_null+0x1c/0x24)
[<c00289f8>] (warn_slowpath_null) from [<c0220290>] (kobject_get+0x64/0x70)
[<c0220290>] (kobject_get) from [<c03e944c>] (cpufreq_cpu_get+0x88/0xc8)
[<c03e944c>] (cpufreq_cpu_get) from [<c03e9500>] (cpufreq_get+0xc/0x64)
[<c03e9500>] (cpufreq_get) from [<c0285288>] (actmon_thread_isr+0x134/0x198)
[<c0285288>] (actmon_thread_isr) from [<c0069008>] (irq_thread_fn+0x1c/0x40)
[<c0069008>] (irq_thread_fn) from [<c0069324>] (irq_thread+0x134/0x174)
[<c0069324>] (irq_thread) from [<c0040290>] (kthread+0xdc/0xf4)
[<c0040290>] (kthread) from [<c000f4b8>] (ret_from_fork+0x14/0x3c)
---[ end trace b7bd64a81b340c59 ]---
Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
This patch adds support for using the NFS v4.2 operation DEALLOCATE to
punch holes in a file.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
This patch adds support for using the NFS v4.2 operation ALLOCATE to
preallocate data in a file.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Setting retval to zero is not needed in ext4_unlink.
Remove unneeded code.
Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Signed-off-by: Ashish Sangwan <a.sangwan@samsung.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
This was fixed for ext3 with:
e6d8fb3 ext3: Count internal journal as bsddf overhead in ext3_statfs
but was never fixed for ext4.
With a large external journal and no used disk blocks, df comes
out negative without this, as journal blocks are added to the
overhead & subtracted from used blocks unconditionally.
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
path[depth].p_hdr can never be NULL for a path passed to us (and even if
it could, EXT_LAST_EXTENT() would make something != NULL from it). So
just remove the branch.
Coverity-id: 1196498
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
nfs4_init_callback() is never invoked for NFS versions other than 4.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Use the correct calculation of the maximum size of a clientaddr4
when encoding and decoding SETCLIENTID operations. clientaddr4 is
defined in section 2.2.10 of RFC3530bis-31.
The usage in encode_setclientid_maxsz is missing the 4-byte length
in both strings, but is otherwise correct. decode_setclientid_maxsz
simply asks for a page of receive buffer space, which is
unnecessarily large (more than 4KB).
Note that a SETCLIENTID reply is either clientid+verifier, or
clientaddr4, depending on the returned NFS status. It doesn't
hurt to allocate enough space for both.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Occasionally mountstats reports a negative retransmission rate.
Ensure that two RPCs completing concurrently don't confuse the sums
in the transport's op_metrics array.
Since pNFS filelayout can invoke rpc_count_iostats() on another
transport from xprt_release(), we can't rely on simply holding the
transport_lock in xprt_release(). There's nothing for it but hard
serialization. One spin lock per RPC operation should make this as
painless as it can be.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
When an operation is created its receive buffer size is specified.
In all current cases, the size supplied for the receive buffer is
exactly the size that should be returned. In other words, if
any fewer than that many bytes arrived in a response, it would be
an error.
So tighten the check on the number of bytes arriving for a response
message, ensuring that the number of bytes received is *exactly the
same* as the number of bytes available (rather than just less than).
We'll expand our interpretation of of -EMSGSIZE to mean "wrong
message size" rather than just "message too long."
If we someday encounter an actual case where we want to be able to
successfully receive something less than the full receive buffer we
can adjust the code to handle that (and give it a way to tell the
receiver how many bytes are present).
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>