Add a new gpiolib mechanism: gpio_chip instances can provide mappings
between their (input) GPIOs and any associated IRQs.  This makes it easier
for platforms to support IRQs that are provided by board-specific external
chips instead of as part of their core (such as SOC-integrated GPIOs).

Also update the irq_to_gpio() description, saying to avoid it because it's
not always supported.
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Make the I2C external GPIO expander drivers register themselves at
subsys_initcall() time when they're statically linked.

SOC-integrated GPIOs are available starting very early -- early in
arch_initcall() at latest, but often even before initcalls start to run --
so this improves consistency, so more subsystems can rely on GPIOs in
their own subsys_initcall() code.

(This isn't a theoretical problem.  This is one of several patches needed
to resolve oopsing observed when statically linking kernels on a DaVinci
EVM.  Its pcf857x GPIOs needed to be available well before some other
drivers initialized.)
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
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>

According to the documentation gpio_free should only be called from task
context only.  To make this more explicit add a might sleep to all
implementations.

This is the generic part which changes gpiolib and the fallback
implementation only.
Signed-off-by: NUwe Kleine-König <ukleinek@informatik.uni-freiburg.de>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

There is a small race and code ugliness in max7301: pins are reconfigured
after the chip is registered.  Swap these calls so that the device is
registered in correct state.

Also this fixes the comile-time warning about unchecked gpiochip_remove.
Signed-off-by: NDmitry Baryshkov <dbaryshkov@gmail.com>
Cc: Juergen Beisert <j.beisert@pengutronix.de>
Cc: Guennadi Liakhovetski <g.liakhovetski@pengutronix.de>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

New style conversion and reformatting as per indent --linux-style
Signed-off-by: NAlessandro Zummo <a.zummo@towertech.it>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

New style conversion and reformatting as per indent --linux-style
Signed-off-by: NAlessandro Zummo <a.zummo@towertech.it>
Cc: Dale Farnsworth <dale@farnsworth.org>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The non-functional periodic IRQ support was previously removed from the
AT91RM9200 RTC driver.  Remove the remaining AT91_RTC_FREQ definition.
Signed-off-by: NAndrew Victor <linux@maxim.org.za>
Cc: David Brownell: <david-b@pacbell.net>
Cc: Alessandro Zummo: <a.zummo@towertech.it>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add support for M41T65 Real Time Clock chip.

The main differences I see between the M41T65 and M41T80 are that:

1) The M41T65 watchdog timer has three bits controlling resolution
   (versus two for the M41T80).

2) There is no register 0x13 for controlling square-wave output.
Signed-off-by: NSteven A. Falco <sfalco@harris.com>
Acked-by: NAlessandro Zummo <a.zummo@towertech.it>
Cc: "Maciej W. Rozycki" <macro@linux-mips.org>
Acked-by: NAtsushi Nemoto <anemo@mba.ocn.ne.jp>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Now that arch/ppc is dead CONFIG_PPC_MERGE is always defined for all
powerpc platforms and we want to get rid of it use CONFIG_PPC instead.
Signed-off-by: NKumar Gala <galak@kernel.crashing.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: David Brownell <david-b@pacbell.net>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Make the rtc framework consistent about disabling 1/second update IRQs
that may have been activated through the /dev interface, when that /dev
file is closed.  (It may have closed because of coredump, etc.) This was
previously done only for emulated update IRQs ...  now, do it always.

Also comment the current policy: repeating IRQs (periodic, update) that
userspace enabled will be cleanly disabled, but alarms are left alone.
Such repeating IRQs are a constant and pointless system load.

Update some RTC drivers to remove now-needless release() methods.  Most
such methods just enforce that policy.  The others all seem to be buggy,
and mistreat in-kernel clients of periodic or alarm IRQs.
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Acked-by: NAndrew Sharp <andy.sharp@onstor.com>
Cc: Angelo Castello <angelo.castello@st.com>
Acked-by: NAtsushi Nemoto <anemo@mba.ocn.ne.jp>
Acked-by: NPaul Mundt <lethal@linux-sh.org>
Cc: Thomas Hommel <thomas.hommel@gefanuc.com>
Acked-by: NAlessandro Zummo <a.zummo@towertech.it>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This adds support for the Ricoh R2025S/D series of I2C RTCs, produced by
Ricoh Japan and described at:

	http://www.ricoh.co.jp/LSI/product_rtc/2wire/r2025x/

This series has very minor deviations from the rest of the RS5C chips,
most of which have to do with the oscillator, which was abstracted away in
an earlier patch.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>
Acked-by: NDavid Brownell <david-b@pacbell.net>
Tested-by: NRiku Voipio <riku.voipio@movial.fi>
Acked-by: NAlessandro Zummo <a.zummo@towertech.it>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

rtc-rs5c372 presently depends on I2C master mode transfers, despite the
fact that these RTCs frequently find themselves on SMBus-only adapters.

Given that the only capabilities that were checked were for I2C_FUNC_I2C,
it's assumed that most of the adapters that are currently using this
driver are fairly sane, and are able to handle SMBus emulation (though we
adjust the default capabilities to check for I2C_FUNC_SMBUS_EMUL anyways,
which is the vast majority of them.  The adapters that don't have their
own ->smbus_xfer() fall back on the ->master_xfer() through the emulated
transfer).

The special case is iop3xx, which has more than its fair share of hacks
within this driver, it remains untested -- though also claims to support
emulated SMBus accesses.  The corner case there is rs5c_get_regs() which
uses access mode #3 for transferring the register state, while we use mode
#1 for SMBus.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>
Acked-by: NDavid Brownell <david-b@pacbell.net>
Tested-by: NRiku Voipio <riku.voipio@movial.fi>
Acked-by: NAlessandro Zummo <a.zummo@towertech.it>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

HPET_RTC_IRQ is no longer needed; HPET_EMULATE_RTC suffices and is more
correct.  (http://bugzilla.kernel.org/show_bug.cgi?id=11111)

Note that when using the legacy RTC driver, platforms don't really do a
dynamic switch between HPET and non-HPET modes based on whether HPET
hardware actually exists ...  only rtc-cmos (using the new RTC framework)
currently switches that way.

So this reflects bitrot in that legacy code, for x86/ia64: kernels with
HPET support configured (e.g.  for a clocksource) can't get IRQs from the
legacy RTC driver unless they really have HPET hardware.  (The obvious
workaround is to not use the legacy RTC driver on those platforms when you
configure HPET ...  unless you know the target really has a HPET.)
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Remove NOP methods from rtc-pl030 and rtc-pl031 drivers;
this is pure wasted code.
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Update the ds1307 driver with alarm support for ds1337/ds1339.  This uses
the first alarm (there are two), and matches on seconds, minutes, hours,
and day-of-month.  Tested on ds1339.

[dbrownell@users.sourceforge.net: add comments; fixup style, valid irq
checks, debug dumps; lock; more careful IRQ shutdown; switch BCD2BIN to
bcd2bin (and vice versa); ENOTTY not EINVAL.]
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NRodolfo Giometti <giometti@linux.it>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add support for the Dallas DS3234 chip - extremely accurate SPI bus RTC
with integrated crystal and SRAM.

[akpm@linux-foundation.org: don't use BIN2BCD/BCD2BIN]
Signed-off-by: NDennis Aberilla <denzzzhome@yahoo.com>
Signed-off-by: NAlessandro Zummo <a.zummo@towertech.it>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Wakeup support implementation.
Signed-off-by: NMarc Pignat <marc.pignat@hevs.ch>
Cc: David Brownell <david-b@pacbell.net>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Validating clients with black magic register checks doesn't make much
sense for new-style i2c driver and has been known to fail on valid NXP
pcf8563 chips.  This patch removes the client validation code.
Signed-off-by: NLaurent Pinchart <laurentp@cse-semaphore.com>
Signed-off-by: NAlessandro Zummo <a.zummo@towertech.it>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The I2O ioctls assume 32bits.  In itself that is fine as they are old
cards and nobody uses 64bit.  However on LKML it was noted this
assumption is also made for allocated memory and is unsafe on 64bit
systems.

Fixing this is a mess.  It turns out there is tons of crap buried in a
header file that does racy 32/64bit filtering on the masks.

So we:
- Verify all callers of the racy code can sleep (i2o_dma_[re]alloc)
- Move the code into a new i2o/memory.c file
- Remove the gfp_mask argument so nobody can try and misuse the function
- Wrap a mutex around the problem area (a single mutex is easy to do and
  none of this is performance relevant)
- Switch the remaining problem kmalloc holdout to use i2o_dma_alloc

Cc: Markus Lidel <Markus.Lidel@shadowconnect.com>
Cc: Vasily Averin <vvs@sw.ru>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Make the SPI external GPIO expander drivers register themselves at
subsys_initcall() time when they're statically linked, and make the SPI
core do its driver model initialization earlier so that's safe.

SOC-integrated GPIOs are available starting very early -- often before
initcalls start to run, or earily in arch_initcall() at latest -- so this
improves consistency, letting more subsystems rely on GPIOs being usable
by their own subsys_initcall() code.
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add support to orion_spi for the 88F6183 ARM SoC by adding code to work
around a 6183-specific erratum.
Signed-off-by: NLennert Buytenhek <buytenh@marvell.com>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Make the chip info structure data optional by providing reasonable
defaults.  Improve corresponding documentation, and highlight the drawback
of not providing explicit chipselect control.

DMA can determine appropriate dma_burst_size and thresholds automatically
so use DMA even if dma_burst_size is not specified.
Signed-off-by: NVernon Sauder <VernonInHand@gmail.com>
Reviewed-by: NNed Forrester <nforrester@whoi.edu>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Minor fixes: remove redundant local variable initialization, fix "can not"
to what I _think_ is a preferred spelling, output IRQ number if requesting
it failed.
Signed-off-by: NGuennadi Liakhovetski <g.liakhovetski@pengutronix.de>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add a pin configuration callback for the s3c24xx SPI driver, as there are
several options depending on the channel and the chip in use.

This is needed as the controller may not have been setup by the initial
bootloader and the fact that the SPI controller gets reset over
suspend/resume into slave mode but the GPIO function registers do not.
Signed-off-by: NBen Dooks <ben-linux@fluff.org>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Modify spi_write_then_read() to use one transfer.  This speeds up all
callers, and is a minor code shrink.
Signed-off-by: NVernon Sauder <Vernon.Sauder@gmail.com>
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Now that arch/ppc is gone we don't need CONFIG_PPC_MERGE anymore
remove the dead code associated with !CONFIG_PPC_MERGE.
Signed-off-by: NKumar Gala <galak@kernel.crashing.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: David Brownell <david-b@pacbell.net>
Cc: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

People can use the real name an an index into MAINTAINERS to find the
current email address.
Signed-off-by: NFrancois Cami <francois.cami@free.fr>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Only three of Atmel's AT91 processors (SAM9263, SAM9RL and CAP9) include a
PWM controller.

It should therefore only be possible to enable the misc/atmel_pwm.c driver
on those processors (and not all AT91 processors).
Signed-off-by: NAndrew Victor <linux@maxim.org.za>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAlberto Bertogli <albertito@blitiri.com.ar>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It is written in the Documentation/sysrq.txt that oom-killer is enabled
when we set "64" in /proc/sys/kernel/sysrq:

<Documentation/sysrq.txt>

    Here is the list of possible values in /proc/sys/kernel/sysrq:
         64 - enable signalling of processes (term, kill, oom-kill)
                                                          ^^^^^^^^

but enable_mask is not set in sysrq_moom_op.
Signed-off-by: NNaohiro Ooiwa <nooiwa@miraclelinux.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Using "def_bool n" is pointless, simply using bool here appears more
appropriate.

Further, retaining such options that don't have a prompt and aren't
selected by anything seems also at least questionable.
Signed-off-by: NJan Beulich <jbeulich@novell.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
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 the '%pF' format to get rid of an "#ifdef DEBUG" and make some printks
atomic.

This removes the last in-tree uses of print_fn_descriptor_symbol().  I
marked print_fn_descriptor_symbol() deprecated and scheduled it for
removal next year to give time for out-of-tree modules to be updated.

parisc's print_fn_descriptor_symbol() is currently broken there (it needs
to dereference the function pointer similar to ia64 and power).  This
patch shouldn't make anything worse, but it means we need to fix
dereference_function_descriptor() instead of print_fn_descriptor_symbol()
to get meaningful initcall_debug output.
Signed-off-by: NBjorn Helgaas <bjorn.helgaas@hp.com>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It seems this is the right way around because otherwise the len usage in
the outer loop would be pretty pointless.
Signed-off-by: NIlpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

EEEPC_LAPTOP uses RFKILL, so the former should depend on RFKILL.
Build errors happen when EEEPC_LAPTOP=y and RFKILL=m.

eeepc-laptop.c:(.text+0xd5a7b): undefined reference to `rfkill_allocate'
eeepc-laptop.c:(.text+0xd5b04): undefined reference to `rfkill_register'
eeepc-laptop.c:(.text+0xd5b48): undefined reference to `rfkill_allocate'
eeepc-laptop.c:(.text+0xd5bd4): undefined reference to `rfkill_register'
eeepc-laptop.c:(.text+0xd5ece): undefined reference to `rfkill_unregister'
eeepc-laptop.c:(.text+0xd5ef6): undefined reference to `rfkill_unregister'
make[1]: *** [.tmp_vmlinux1] Error 1
Signed-off-by: NRandy Dunlap <randy.dunlap@oracle.com>
Cc: Corentin Chary <corentincj@iksaif.net>
Cc: Matthew Garrett <mjg@redhat.com>
Cc: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Cc: Karol Kozimor <sziwan@users.sourceforge.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The kernel.h macro DIV_ROUND_UP performs the computation (((n) + (d) - 1) /
(d)) but is perhaps more readable.

An extract of the semantic patch that makes this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@haskernel@
@@

#include <linux/kernel.h>

@depends on haskernel@
expression n,d;
@@

(
- (n + d - 1) / d
+ DIV_ROUND_UP(n,d)
|
- (n + (d - 1)) / d
+ DIV_ROUND_UP(n,d)
)

@depends on haskernel@
expression n,d;
@@

- DIV_ROUND_UP((n),d)
+ DIV_ROUND_UP(n,d)

@depends on haskernel@
expression n,d;
@@

- DIV_ROUND_UP(n,(d))
+ DIV_ROUND_UP(n,d)
// </smpl>
Signed-off-by: NJulia Lawall <julia@diku.dk>
Cc: Jack Steiner <steiner@sgi.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

__FUNCTION__ is gcc-specific, use __func__
Signed-off-by: NHarvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NHarvey Harrison <harvey.harrison@gmail.com>
Acked-by: NAndres Salomon <dilinger@queued.net>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

proc_clear_tty() gets called with interrupts off (while holding the task list
lock) from sys_setid. This means that it needs the _irqsave version of the
locking primitives.
Reported-by: NMarcin Slusarz <marcin.slusarz@gmail.com>
Tested-by: NMarcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The metronome driver produces warnings when built on x86-64 as it assumes that
size_t is an int. Use %Zd instead.
Signed-off-by: NAlan Cox <alan@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This driver replaces the broken ip27-rtc driver in drivers/char and
gives back RTC support for SGI IP27 machines.
Signed-off-by: NThomas Bogendoerfer <tsbogend@alpha.franken.de>
Acked-by: NAlessandro Zummo <alessandro.zummo@towertech.it>
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>