Add support for PMP8074 PMIC which is a companion PMIC for the Qualcomm
IPQ8074 SoC-s.
It shares the same subtype identifier as PM8901.
Signed-off-by: Robert Marko <robimarko@gmail.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220818221815.346233-2-robimarko@gmail.com
The mx25_tsadc_remove() function assumes all non-zero returns are success
but the platform_get_irq() function returns negative on error and
positive non-zero values on success. It never returns zero, but if it
did then treat that as a success.
Fixes: 18f7739379 ("mfd: fsl-imx25: Clean up irq settings during removal")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Martin Kaiser <martin@kaiser.cx>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/YvTfkbVQWYKMKS/t@kili
Document existing (MSM8996, SC7280) and new compatibles for TCSR syscon
registers (QCS404, SC7180, SDM630, SDM845, SM8150, MSM8998).
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220819083209.50844-2-krzysztof.kozlowski@linaro.org
There are two bindings for Qualcomm SPMI PMIC Thermal Monitoring ADC:
one for ADC HC and one for ADC TM5 and TM7. PM8998 uses the former one,
so fix matching of child schema:
qcom/msm8998-asus-novago-tp370ql.dtb: pmic@0: adc-tm@3400:compatible:0: 'qcom,spmi-adc-tm-hc' is not one of ['qcom,spmi-adc-tm5', 'qcom,spmi-adc-tm5-gen2', 'qcom,adc-tm7']
Fixes: 3f5117be95 ("dt-bindings: mfd: convert to yaml Qualcomm SPMI PMIC")
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220828084341.112146-15-krzysztof.kozlowski@linaro.org
The regulators node of Qualcomm SPMI PMIC represents sub-device, so it
has its own compatible, multiple regulators and uses dedicated bindings.
Fixes: 3f5117be95 ("dt-bindings: mfd: convert to yaml Qualcomm SPMI PMIC")
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220828084341.112146-14-krzysztof.kozlowski@linaro.org
Add a more complete example with PM6150 to provide better validation of
the bindings.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Rob Herring <robh@kernel.org>
Reviewed-by: Stephen Boyd <sboyd@kernel.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220828130113.5845-1-krzysztof.kozlowski@linaro.org
The syscon bindings require a device specific compatible, beside the
"syscon". However schema counts "simple-mfd" as such, which allows
simple-mfd+syscon to sneak in.
Adjust the match to be sure simple-mfd also comes with a device specific
compatible.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220817142246.828762-5-krzysztof.kozlowski@linaro.org
Conversion from TXT to DT schema lost several compatibles.
Fixes: 3f5117be95 ("dt-bindings: mfd: convert to yaml Qualcomm SPMI PMIC")
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Stephen Boyd <sboyd@kernel.org>
Reviewed-by: David Heidelberg <david@ixit.cz>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220828065123.39734-1-krzysztof.kozlowski@linaro.org
The System Control Processor System (SCPSYS) has several power
management related tasks in the system. Add the bindings for it.
Signed-off-by: Tinghan Shen <tinghan.shen@mediatek.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220811025813.21492-7-tinghan.shen@mediatek.com
Convert the aspeed,ast2[456]00-scu binding to DT schema format.
The original binding was missing '#address-cells', '#size-cells',
'ranges', and child nodes, so add them.
Signed-off-by: Rob Herring <robh@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220810161635.73936-3-robh@kernel.org
When we consider MFD which implements hotplug (e.g. USB hotplug
driver based on product and vendor IDs) functionality it turns out
that its sub-devices are correctly matched with corresponding device
tree nodes only at the first time. When physical device reboots
or is replugged (and MFD driver is disconnected and probed back
again) all sub-devices fails in mfd_add_device() with error
'Failed to locate of_node'.
The reason of that behavior is that when any MFD sub-device is
created for the first time (and matched with device tree node) it
is added to the mfd_of_node_list. It looks like this list is never
cleaned even if devices added there are intentionally removed from
the system. So when MFD device is replugged and all sub-devices
are matched with their device tree nodes again they fail as matched
nodes already exist in mfd_of_node_list. In other words current
implementation does not support MFD with hotplug feature.
This commit extends MFD core for hotplugging support by removing
appropriate OF node entry from mfd_of_node_list when corresponding
device is removed from the system. Thanks to that when device is
added once again it can be matched with its device tree node
successfully.
Signed-off-by: Michal Oleszczyk <oleszczyk.m@gmail.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220809060336.31892-1-m.oleszczyk@grinn-global.com
In lp8788_irq_init(), if an error occurs after a successful
irq_domain_add_linear() call, it must be undone by a corresponding
irq_domain_remove() call.
irq_domain_remove() should also be called in lp8788_irq_exit() for the same
reason.
Fixes: eea6b7cc53 ("mfd: Add lp8788 mfd driver")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/bcd5a72c9c1c383dd6324680116426e32737655a.1659261275.git.christophe.jaillet@wanadoo.fr
Should an error occurs in mfd_add_devices(), some resources need to be
released, as already done in the .remove() function.
Add an error handling path and a lp8788_irq_exit() call to undo a previous
lp8788_irq_init().
Fixes: eea6b7cc53 ("mfd: Add lp8788 mfd driver")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/18398722da9df9490722d853e4797350189ae79b.1659261275.git.christophe.jaillet@wanadoo.fr
If devm_of_platform_populate() fails, some resources need to be
released.
Introduce a mx25_tsadc_unset_irq() function that undoes
mx25_tsadc_setup_irq() and call it both from the new error handling path
of the probe and in the remove function.
Fixes: a55196eff6 ("mfd: fsl-imx25: Use devm_of_platform_populate()")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/d404e04828fc06bcfddf81f9f3e9b4babbe35415.1659269156.git.christophe.jaillet@wanadoo.fr
AXP228 is a PMIC used on boards such as the Clockwork ClockworkPi and
DevTerm. Its register map appears to be identical to the AXP221 variant.
The only known difference is in the default values for regulator on/off
states and voltages.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220810013430.27061-1-samuel@sholland.org
Add a specific MFD_SY7636A config option.
As part of this change we can use MFD_SY7636A as a dependency for all
SY7636a components and also remove the name from MFD_SIMPLE_MFD_I2C as
it no longer needs to be selectable.
Signed-off-by: Alistair Francis <alistair@alistair23.me>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Acked-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220525115554.430971-2-alistair@alistair23.me
This adds support for the MediaTek MT6370 SubPMIC. MediaTek MT6370 is a
SubPMIC consisting of a single cell battery charger with ADC monitoring,
RGB LEDs, dual channel flashlight, WLED backlight driver, display bias
voltage supply, one general purpose LDO, and the USB Type-C & PD controller
complies with the latest USB Type-C and PD standards.
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: ChiYuan Huang <cy_huang@richtek.com>
Signed-off-by: ChiaEn Wu <chiaen_wu@richtek.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220805070610.3516-8-peterwu.pub@gmail.com
To reflect the point that this driver is only for one type of the PMICs,
replace intel_soc_pmic with crystal_cove (avoid using crc for possible
namespace collisions with CRC library APIs).
Note, also rename the driver name since we don't expect any user
that enumerates by it, only ACPI known so far.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220801114211.36267-9-andriy.shevchenko@linux.intel.com
Use the ->probe_new() callback.
The driver does not use const struct i2c_device_id * argument,
so convert it to utilise the simplified I²C driver registration.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220801114211.36267-8-andriy.shevchenko@linux.intel.com
The driver depends on ACPI, ACPI_PTR() resolution is always the same.
Otherwise a compiler may produce a warning.
That said, the rule of thumb either ugly ifdeffery with ACPI_PTR or
none should be used in a driver.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220801114211.36267-7-andriy.shevchenko@linux.intel.com
Letting the compiler remove these functions when the kernel is built
without CONFIG_PM_SLEEP support is simpler and less error prone than the
use of #ifdef based kernel configuration guards.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220801114211.36267-6-andriy.shevchenko@linux.intel.com
We have the specific helpers for I2C device to set and get its driver data.
Convert driver to use them instead of open coded variants.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220801114211.36267-5-andriy.shevchenko@linux.intel.com
Use devm_regmap_add_irq_chip() to simplify the code.
While at it, replace -1 magic parameter by PLATFORM_DEVID_NONE when
calling mfd_add_devices().
Note, the mfd_add_devices() left in non-devm variant here due to
potentially increased churn while wrapping pwm_remove_table().
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220801114211.36267-4-andriy.shevchenko@linux.intel.com
It looks like a random position for couple of Makefile entries that are
disrupting Intel PMIC group. Move them to their own group.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220801114211.36267-3-andriy.shevchenko@linux.intel.com
The core part is misleading since its only purpose to serve Crystal Cove PMIC,
although for couple of different platforms. Merge core part into crc one.
Advantages among others are:
- speed up a compilation and build
- decreasing the code base
- reducing noise in the namespace by making some data static and const
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220801114211.36267-2-andriy.shevchenko@linux.intel.com
The commit in Fixes: has added a pwm_add_table() call in the probe() and
a pwm_remove_table() call in the remove(), but forget to update the error
handling path of the probe.
Add the missing pwm_remove_table() call.
Fixes: a3aa9a93df ("mfd: intel_soc_pmic_core: ADD PWM lookup table for CRC PMIC based PWM")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220801114211.36267-1-andriy.shevchenko@linux.intel.com
Use sub-function of_compatible during probe, instead of using the node
name. The code should not rely on the node names during probe, in
addition to that the previously hard-coded node names are not compliant
to the latest naming convention (they are not generic and they use
underscores), and it was broken by mistake already once [1].
[1] commit 56086b5e80 ("ARM: dts: imx6qdl-apalis: Avoid underscore in node name")
Suggested-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220712163345.445811-3-francesco.dolcini@toradex.com
Remove rotator block from probe, it is not used in any device tree file,
there is no related cell defined, it's just dead non-working code with no
of_compatible for it.
This is a preliminary change to allow probing by of_compatible and not
by a fixed name.
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220712163345.445811-2-francesco.dolcini@toradex.com
The SPI driver wants to know the exact type of the controller.
Provide this information to it. This is a complementary part to
the previously updated intel-lpss-acpi.c.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220702211903.9093-1-andriy.shevchenko@linux.intel.com
The VSC7512 is a networking chip that contains several peripherals. Many of
these peripherals are currently supported by the VSC7513 and VSC7514 chips,
but those run on an internal CPU. The VSC7512 lacks this CPU, and must be
controlled externally.
Utilize the existing drivers by referencing the chip as an MFD. Add support
for the two MDIO buses, the internal phys, pinctrl, and serial GPIO.
Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220905162132.2943088-9-colin.foster@in-advantage.com
Add devicetree bindings for SPI-controlled Ocelot chips, specifically the
VSC7512.
Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220905162132.2943088-8-colin.foster@in-advantage.com
DEFINE_RES_ macros have been created for the commonly used resource types,
but not IORESOURCE_REG. Add the macro so it can be used in a similar manner
to all other resource types.
Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220905162132.2943088-7-colin.foster@in-advantage.com
There are a few Ocelot chips that can contain SGPIO logic, but can be
controlled externally. Specifically the VSC7511, 7512, 7513, and 7514. In
the externally controlled configurations these registers are not
memory-mapped.
Add support for these non-memory-mapped configurations.
Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220905162132.2943088-6-colin.foster@in-advantage.com
As the commit message suggests, this simply adds the ability to select
SGPIO pinctrl as a module. This becomes more practical when the SGPIO
hardware exists on an external chip, controlled indirectly by I2C or SPI.
This commit enables that level of control.
Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220905162132.2943088-5-colin.foster@in-advantage.com
There are a few Ocelot chips that contain pinctrl logic, but can be
controlled externally. Specifically the VSC7511, 7512, 7513 and 7514. In
the externally controlled configurations these registers are not
memory-mapped.
Add support for these non-memory-mapped configurations.
Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220905162132.2943088-4-colin.foster@in-advantage.com
There are a few Ocelot chips that contain the logic for this bus, but are
controlled externally. Specifically the VSC7511, 7512, 7513, and 7514. In
the externally controlled configurations these registers are not
memory-mapped.
Add support for these non-memory-mapped configurations.
Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220905162132.2943088-3-colin.foster@in-advantage.com
Several ocelot-related modules are designed for MMIO / regmaps. As such,
they often use a combination of devm_platform_get_and_ioremap_resource()
and devm_regmap_init_mmio().
Operating in an MFD might be different, in that it could be memory mapped,
or it could be SPI, I2C... In these cases a fallback to use IORESOURCE_REG
instead of IORESOURCE_MEM becomes necessary.
When this happens, there's redundant logic that needs to be implemented in
every driver. In order to avoid this redundancy, utilize a single function
that, if the MFD scenario is enabled, will perform this fallback logic.
Signed-off-by: Colin Foster <colin.foster@in-advantage.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20220905162132.2943088-2-colin.foster@in-advantage.com
Radix tree header includes gfp.h for __GFP_BITS_SHIFT only. Now we
have gfp_types.h for this.
Fixes powerpc allmodconfig build:
In file included from include/linux/nodemask.h:97,
from include/linux/mmzone.h:17,
from include/linux/gfp.h:7,
from include/linux/radix-tree.h:12,
from include/linux/idr.h:15,
from include/linux/kernfs.h:12,
from include/linux/sysfs.h:16,
from include/linux/kobject.h:20,
from include/linux/pci.h:35,
from arch/powerpc/kernel/prom_init.c:24:
include/linux/random.h: In function 'add_latent_entropy':
>> include/linux/random.h:25:46: error: 'latent_entropy' undeclared (first use in this function); did you mean 'add_latent_entropy'?
25 | add_device_randomness((const void *)&latent_entropy, sizeof(latent_entropy));
| ^~~~~~~~~~~~~~
| add_latent_entropy
include/linux/random.h:25:46: note: each undeclared identifier is reported only once for each function it appears in
Reported-by: kernel test robot <lkp@intel.com>
CC: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
somebody left NULL ->proc_lseek, became a practical issue now.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQQqUNBr3gm4hGXdBJlZ7Krx/gZQ6wUCYvlK4QAKCRBZ7Krx/gZQ
68EUAQDFUFhyvVLnhh1tAkPt3mxHT2OHASN8z6gRBjxGmqj53QD8D4EZAMy3XHnB
/6ajdD7hl5pxZQnECPqbBjzPYHa22QE=
=BNbv
-----END PGP SIGNATURE-----
Merge tag 'pull-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull vfs lseek fix from Al Viro:
"Fix proc_reg_llseek() breakage. Always had been possible if somebody
left NULL ->proc_lseek, became a practical issue now"
* tag 'pull-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
take care to handle NULL ->proc_lseek()
Easily done now, just by clearing FMODE_LSEEK in ->f_mode
during proc_reg_open() for such entries.
Fixes: 868941b144 "fs: remove no_llseek"
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Robert Marko