platform/kernel/linux-starfive.git
4 years agostaging: vt6656: correct packet types for CTS protect, mode.
Malcolm Priestley [Wed, 8 Jan 2020 21:40:58 +0000 (21:40 +0000)]
staging: vt6656: correct packet types for CTS protect, mode.

It appears that the driver still transmits in CTS protect mode even
though it is not enabled in mac80211.

That is both packet types PK_TYPE_11GA and PK_TYPE_11GB both use CTS protect.
The only difference between them GA does not use B rates.

Find if only B rate in GB or GA in protect mode otherwise transmit packets
as PK_TYPE_11A.

Cc: stable <stable@vger.kernel.org>
Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Link: https://lore.kernel.org/r/9c1323ff-dbb3-0eaa-43e1-9453f7390dc0@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: mt7621-dts: fix register range of memc node in mt7621.dtsi
Chuanhong Guo [Thu, 9 Jan 2020 08:00:03 +0000 (16:00 +0800)]
staging: mt7621-dts: fix register range of memc node in mt7621.dtsi

The memc node from mt7621.dtsi has incorrect register resource.
Fix it according to the programming guide.

Signed-off-by: Weijie Gao <hackpascal@gmail.com>
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
Link: https://lore.kernel.org/r/20200109080120.362110-1-gch981213@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: rtl8192u: remove unneeded compiler flags
Masahiro Yamada [Sat, 4 Jan 2020 16:21:36 +0000 (01:21 +0900)]
staging: rtl8192u: remove unneeded compiler flags

-std=gnu89 is specified by the top Makefile. Adding it in a driver
Makefile is redundant.

A driver should avoid specifying the optimization flag.
-O2, -O3, or -Os is passed by the top Makefile based on the
CONFIG_CC_OPTIMIZE_FOR_* option.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Link: https://lore.kernel.org/r/20200104162136.19170-3-masahiroy@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: rtl8192u: remove header include path to ieee80211/
Masahiro Yamada [Sat, 4 Jan 2020 16:21:35 +0000 (01:21 +0900)]
staging: rtl8192u: remove header include path to ieee80211/

There is no need to add "ccflags-y += -I $(srctree)/$(src)/ieee80211"
just for including "dot11d.h".

Use the correct relative path for the #include "..." directive.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Link: https://lore.kernel.org/r/20200104162136.19170-2-masahiroy@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: rtl8192u: remove unused Makefile
Masahiro Yamada [Sat, 4 Jan 2020 16:21:34 +0000 (01:21 +0900)]
staging: rtl8192u: remove unused Makefile

drivers/staging/rtl8192u/ieee80211/Makefile is not used at all.
All the build rules are described in drivers/staging/rtl8192u/Makefile.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Link: https://lore.kernel.org/r/20200104162136.19170-1-masahiroy@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: rtl8188eu: refactor rtl88eu_dm_update_rx_idle_ant()
Michael Straube [Sun, 5 Jan 2020 19:49:36 +0000 (20:49 +0100)]
staging: rtl8188eu: refactor rtl88eu_dm_update_rx_idle_ant()

Refactor rtl88eu_dm_update_rx_idle_ant() to reduce indentation level
and clear line over 80 characters checkpatch warnings.

Signed-off-by: Michael Straube <straube.linux@gmail.com>
Link: https://lore.kernel.org/r/20200105194936.5477-2-straube.linux@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: rtl8188eu: remove else after return
Michael Straube [Sun, 5 Jan 2020 19:49:35 +0000 (20:49 +0100)]
staging: rtl8188eu: remove else after return

Remove else after return in rtl88eu_dm_antenna_diversity() to improve
readability and clear a checkpatch warning.

WARNING: else is not generally useful after a break or return

Signed-off-by: Michael Straube <straube.linux@gmail.com>
Link: https://lore.kernel.org/r/20200105194936.5477-1-straube.linux@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: rtl8188: avoid excessive stack usage
Arnd Bergmann [Sat, 4 Jan 2020 21:48:21 +0000 (22:48 +0100)]
staging: rtl8188: avoid excessive stack usage

The rtl8188 copy of the os_dep support code causes a
warning about a very significant stack usage in the translate_scan()
function:

drivers/staging/rtl8188eu/os_dep/ioctl_linux.c: In function 'translate_scan':
drivers/staging/rtl8188eu/os_dep/ioctl_linux.c:306:1: error: the frame size of 1560 bytes is larger than 1400 bytes [-Werror=frame-larger-than=]

Use the same trick as in the rtl8723bs copy of the same function, and
allocate it dynamically.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20200104214832.558198-1-arnd@arndb.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: rts5208: remove unneeded header include path
Masahiro Yamada [Sat, 4 Jan 2020 16:37:10 +0000 (01:37 +0900)]
staging: rts5208: remove unneeded header include path

A header include path without $(srctree)/ is suspicous because it does
not work with O= builds.

I can build drivers/staging/rts5208/ without this include path.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Link: https://lore.kernel.org/r/20200104163710.21582-1-masahiroy@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agoMerge tag 'iio-for-5.6a' of git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio...
Greg Kroah-Hartman [Fri, 10 Jan 2020 09:44:00 +0000 (10:44 +0100)]
Merge tag 'iio-for-5.6a' of git://git./linux/kernel/git/jic23/iio into staging-next

Jonathan writes:

First set of new device support, features and cleanups for IIO in the 5.6 cycle

New device support

* ad7091r5 ADC
  - New driver with follow up patch adding scale and vref support.
  - DT bindings
* ad7923
  - Support for ad7908, ad7918 and ad7928 added to driver.
* bma180
  - Support the BMA254 accelerometer. Required fairly substantial rework
    to allow for small differences between this an existing parts.
* bma400 accelerometer
  - New driver with follow up patch for regulator support.
  - DT bindings.
* asc dlhl60d
  - New driver support this range of pressure and temperature sensors.
  - DT bindings.
* ltc2496 ADC
  - New driver to support this ADC.
  - Split the existing LTC2497 driver generic component out and reuse.
  - DT bindings.
* parallax ping
  - New driver supporting ultrasonic and laser tof distance sensors.
  - Bindings for these sensors.

New features

* core
  - New char type for read_raw returns, used for thermocouple types.
  - Rename read_first_n callback to read.   The reasons behind the original
    naming are lost to the mists of time.
* ad799x
  - Allow pm_ops to disable device completely allowing regulator power down.
* bma180
  - Enable basic regulator support.
* dmaengine buffer
  - Report platform data alignment requirements via new ABI.
* max31856
  - Add option to set mains filter rejection frequency and document
    new in_temp_filter_notch_center_frequency ABI.
  - Add support for configuring HW averaging (oversampling ratio)
  - Add runtime configuration of thermocouple type and document new ABI.
* maxim-thermocouple
  - Add read only access to thermocouple type using new ABI, includes
    adding more specific compatibles to reflect which variant of the
    chip is being used.
* mpu6050
  - Provide option to support the PMU9150 in package magnetometer directly
    rather than via auxiliary bus.
* stm32_adc
  - Add overrun interrupt checks to detect if this happens.
* st_lsm6dsx
  - Enable the sensor-hub support for lsm6dsm. Includes various reworks to
    allow this.

Cleanups and minor fixes

* Subsystem wide
  - Tidy up indentation in Kconfig and fix alphabetical order of AD7091R5.
  - Drop linux/gpio.h and linux/of_gpio.h from drivers that don't use them.
* ad7266
  - Convert to GPIO descriptors.
* ad7303
  - Avoid a dance with checking if the regulator is supplied by just
    using the optional request interface.
* ad7887
  - Simplify channel specification assignment to enable adding more devices.
* ad7923
  - Drop some unused and largely pointless defines of BOB_N==N variety.
  - Tidy up checkpatch warnings.
  - Add missing of_device_id table.
* adf4350
  - Convert to GPIO descriptors.
* ak8975
  - Convert to GPIO descriptors.
* ADIS library and drivers
  - Expand scope of txrx_lock to cover all state and rename as state_lock
  - Add unlocked read / write to allow grouping of consecutive calls under
    single lock / unlock.
  - Add unlocked check_status, reset to allow grouping under single
    lock / unlock.
  - Remove remaining uses of core mlock for local state protection.
    mlock should never be used directly as it protects tightly defined
    core IIO device management state.
* adis16240
  - Enforce only supported SPI mode on driver load + add DT binding doc.
* atlas-ph-sensor
  - Rename to atlas-sensor given it now covers things beyond ph sensors.
* bma180
  - Use local dev variable to tidy up code.
  - Use c99 style explicity .member assignment to make driver more readable.
* bmp280
  - Drop ACPI support. No evidence this was used and appropriate ID is not
    registered.
  - Allow ACPI to bind device via PRP0001
* dmaengine buffer
  - Use dma_request_chan instead of dma_request_slave_channel_reason as that
    ABI is going away.
  - Add module info to avoid tainting the kernel.
* hts221
  - Avoid magic number defines when only used to fill structure elements
    that are self describing.
* lm3533
  - Drop a stray semicolon.
* max9611
  - Cleanup enum handling to be more resilient to future changes.
* mpu6050
  - Delete MPU9150 from supported SPI devices as doesn't provide SPI.
  - Select I2C_MUX again after kbuild issue fixed elsewhere.
* stm32-timer
  - Drop an unnecessary register update.
* ssp_sensors
  - Convert to GPIO descriptors.
* st_sensors
  - drop !CONFIG_ACPI defines as ACPI_PTR() will stop them being used
    anyway.
  - Make default platform data structures __maybe_unsued.
  - Fill in some missing kernel-doc function parameters.
* st_lsm6dsx
  - white space fixes.
  - Mark some constants that aren't always used as __maybe_unused.
  - Drop of ID table guards as they just pervent use under ACPI.
  - Switch to device properties to allow ACPI usage.
* st_uvis25
  - Drop acpi.h include as no ACPI APIs used.
* ti-ads1015
  - Drop legacy platform data as no one seems to be using it.
  - Use the device property API instead of OF specific.
* ti-ads7950
  - typo fix in error message.

* tag 'iio-for-5.6a' of git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio: (99 commits)
  iio: accel: bma180: BMA254 support
  iio: pressure: bmp280: Allow device to be enumerated from ACPI
  iio: pressure: bmp280: Drop ACPI support
  dt-bindings: iio: adc: convert sd modulator to json-schema
  iio: buffer: rename 'read_first_n' callback to 'read'
  iio: buffer-dmaengine: Report buffer length requirements
  bindings: iio: pressure: Add documentation for dlh driver
  dt-bindings: Add asc vendor
  iio: pressure: Add driver for DLH pressure sensors
  iio: buffer-dmaengine: Add module information
  iio: accel: bma180: Use explicit member assignment
  iio: accel: bma180: Basic regulator support
  iio: accel: bma180: Add dev helper variable
  iio: imu: st_lsm6dsx: enable sensor-hub support for lsm6dsm
  iio: imu: st_lsm6dsx: rename st_lsm6dsx_shub_read_reg in st_lsm6dsx_shub_read_output
  iio: imu: st_lsm6dsx: check if shub_output reg is located in primary page
  iio: imu: st_lsm6dsx: check if pull_up is located in primary page
  iio: imu: st_lsm6dsx: check if master_enable is located in primary page
  iio: imu: st_lsm6dsx: export max num of slave devices in st_lsm6dsx_shub_settings
  iio: light: remove unneeded semicolon
  ...

4 years agostaging: wfx: check for memory allocation failures from wfx_alloc_hif
Colin Ian King [Sat, 21 Dec 2019 00:15:43 +0000 (00:15 +0000)]
staging: wfx: check for memory allocation failures from wfx_alloc_hif

Currently calls to wfx_alloc_hif are not checking for a null return
when a memory allocation fails and this leads to null pointer
dereferencing issues.  Fix this by adding null pointer checks and
returning passing down -ENOMEM errors where necessary. The error
checking in the current driver is a bit sparse, so this may need
some extra attention later if required.

Fixes: f95a29d40782 ("staging: wfx: add HIF commands helpers")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Link: https://lore.kernel.org/r/20191221001543.15255-1-colin.king@canonical.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: exfat: add STAGING prefix to config names
Namjae Jeon [Fri, 3 Jan 2020 01:13:45 +0000 (09:13 +0800)]
staging: exfat: add STAGING prefix to config names

Add STAGING prefix to config names to avoid collsion with fs/exfat config.

Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com>
Link: https://lore.kernel.org/r/20200103011345.25245-1-namjae.jeon@samsung.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: vt6656: remove unnecessary parenthesis
Amir Mahdi Ghorbanian [Mon, 30 Dec 2019 15:55:20 +0000 (10:55 -0500)]
staging: vt6656: remove unnecessary parenthesis

Remove unnecessary parenthesis to abide by kernel
coding-style.

Signed-off-by: Amir Mahdi Ghorbanian <indigoomega021@gmail.com>
Link: https://lore.kernel.org/r/20191230155520.GA27072@user-ThinkPad-X230
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: rtl8188eu: use break to exit while loop
Michael Straube [Fri, 20 Dec 2019 17:44:13 +0000 (18:44 +0100)]
staging: rtl8188eu: use break to exit while loop

The variable bContinual in Efuse_PgPacketRead() is only used to break
out of a while loop. Remove the variable and use break instead.

Signed-off-by: Michael Straube <straube.linux@gmail.com>
Link: https://lore.kernel.org/r/20191220174413.13913-1-straube.linux@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: kpc2000: remove unnecessary assertions in kpc_dma_transfer
Aditya Pakki [Thu, 19 Dec 2019 17:21:11 +0000 (11:21 -0600)]
staging: kpc2000: remove unnecessary assertions in kpc_dma_transfer

In kpc_dma_transfer(), the assertion that priv is NULL and priv->ldev
is NULL, are never satisfied. The two callers of the function,
dereference the fields before the function is called. This patch
removes the two BUG_ON calls.

Signed-off-by: Aditya Pakki <pakki001@umn.edu>
Link: https://lore.kernel.org/r/20191219172118.17456-1-pakki001@umn.edu
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agoiio: accel: bma180: BMA254 support
Linus Walleij [Wed, 11 Dec 2019 21:38:19 +0000 (22:38 +0100)]
iio: accel: bma180: BMA254 support

This adds support for the BMA254 variant of this
accelerometer. The only difference for the simple IIO
driver is that values are 12 bit and the temperature
offset differs by 1 degree.

Whilst wildcards in naming are normally frowned upon:

The cases where I have labeled variables "25x" is where the
models are identical, so as to make things easier for people
that want to add support for BMA253 and BMA255.

Cc: Peter Meerwald <pmeerw@pmeerw.net>
Cc: Oleksandr Kravchenko <o.v.kravchenko@globallogic.com>
Cc: devicetree@vger.kernel.org
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: pressure: bmp280: Allow device to be enumerated from ACPI
Andy Shevchenko [Mon, 9 Dec 2019 18:28:05 +0000 (20:28 +0200)]
iio: pressure: bmp280: Allow device to be enumerated from ACPI

There is no need to limit the driver use by OF/platform code.
In this case we simple remove redundant OF parts from the code.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: pressure: bmp280: Drop ACPI support
Andy Shevchenko [Mon, 9 Dec 2019 18:28:04 +0000 (20:28 +0200)]
iio: pressure: bmp280: Drop ACPI support

There is no evidence of officially registered ACPI IDs for these devices.
Thus, revert ACPI support from the driver. All authors of the respective
changes are being informed here:

  d5c94568cc1d ("iio: add bmp280 pressure and temperature driver")
  6dba72eca7fb ("iio: pressure: bmp280: add support for BMP180")
  14beaa8f5ab1 ("iio: pressure: bmp280: add humidity support")

Above seems a cargo cult without paying attention to how ACPI IDs
are being allocated.

Cc: Vlad Dogaru <ddvlad@gmail.com>
Cc: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Matt Ranostay <mranostay@gmail.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agodt-bindings: iio: adc: convert sd modulator to json-schema
Olivier Moysan [Fri, 6 Dec 2019 10:00:58 +0000 (11:00 +0100)]
dt-bindings: iio: adc: convert sd modulator to json-schema

Convert the sigma delta modulator bindings
to DT schema format using json-schema.

Signed-off-by: Olivier Moysan <olivier.moysan@st.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: buffer: rename 'read_first_n' callback to 'read'
Lars-Peter Clausen [Wed, 11 Dec 2019 10:43:00 +0000 (12:43 +0200)]
iio: buffer: rename 'read_first_n' callback to 'read'

It is implied that 'read' will read the first n bytes and not e.g. bytes
only from offsets within the buffer that are a prime number.

This change is non-functional, mostly just a rename.
A secondary intent with this patch is to make room later to add a write
callback.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: buffer-dmaengine: Report buffer length requirements
Lars-Peter Clausen [Wed, 11 Dec 2019 11:56:15 +0000 (13:56 +0200)]
iio: buffer-dmaengine: Report buffer length requirements

The dmaengine buffer has some length alignment requirements that can differ
from platform to platform. If the length alignment requirements are not met
unexpected behavior like dropping of samples can occur.

Currently these requirements are not reported and applications need to know
the requirements of the platform by some out-of-band means.

Add a new buffer attribute that reports the length alignment requirements
called `length_align_bytes`. The reported length alignment is in bytes that
means the buffer length alignment in sample sets depends on the number of
enabled channels and the bytes per channel. Applications using this
attribute to determine the buffer size requirements need to consider this.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agobindings: iio: pressure: Add documentation for dlh driver
Tomislav Denis [Wed, 11 Dec 2019 09:59:46 +0000 (10:59 +0100)]
bindings: iio: pressure: Add documentation for dlh driver

Add a device tree binding documentation for DLH series pressure
sensors.

Signed-off-by: Tomislav Denis <tomislav.denis@avl.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agodt-bindings: Add asc vendor
Tomislav Denis [Wed, 11 Dec 2019 09:59:45 +0000 (10:59 +0100)]
dt-bindings: Add asc vendor

All Sensors Corporation is a manufacturer of MEMS piezoresitive
ultra low pressure sensors and pressure transducers.

Signed-off-by: Tomislav Denis <tomislav.denis@avl.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: pressure: Add driver for DLH pressure sensors
Tomislav Denis [Wed, 11 Dec 2019 09:59:44 +0000 (10:59 +0100)]
iio: pressure: Add driver for DLH pressure sensors

All Sensors DLH is series of low voltage digital pressure sensors.
Additionally to pressure value sensors deliver a temperature value.
Sensors can be accessed over I2C and SPI, this driver supports
only I2C access.

Signed-off-by: Tomislav Denis <tomislav.denis@avl.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: buffer-dmaengine: Add module information
Lars-Peter Clausen [Wed, 11 Dec 2019 10:41:47 +0000 (12:41 +0200)]
iio: buffer-dmaengine: Add module information

Make sure that the industrialio-buffer-dmaengine has proper license
information so it can be build as a module and loaded without tainting the
kernel.

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: accel: bma180: Use explicit member assignment
Linus Walleij [Wed, 11 Dec 2019 21:38:18 +0000 (22:38 +0100)]
iio: accel: bma180: Use explicit member assignment

This uses the C99 explicit .member assignment for the
variant data in struct bma180_part_info. This makes it
easier to understand and add new variants.

Cc: Peter Meerwald <pmeerw@pmeerw.net>
Cc: Oleksandr Kravchenko <o.v.kravchenko@globallogic.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: accel: bma180: Basic regulator support
Linus Walleij [Wed, 11 Dec 2019 21:38:17 +0000 (22:38 +0100)]
iio: accel: bma180: Basic regulator support

This brings up the VDD and VDDIO regulators using the
regulator framework. Platforms that do not use regulators
will provide stubs or dummy regulators.

Cc: Peter Meerwald <pmeerw@pmeerw.net>
Cc: Oleksandr Kravchenko <o.v.kravchenko@globallogic.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: accel: bma180: Add dev helper variable
Linus Walleij [Wed, 11 Dec 2019 21:38:16 +0000 (22:38 +0100)]
iio: accel: bma180: Add dev helper variable

Having a shorthand "dev" instead of &client->dev everywhere
makes the code easier to read (more compact).

Cc: Peter Meerwald <pmeerw@pmeerw.net>
Cc: Oleksandr Kravchenko <o.v.kravchenko@globallogic.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: imu: st_lsm6dsx: enable sensor-hub support for lsm6dsm
Lorenzo Bianconi [Sat, 14 Dec 2019 16:52:59 +0000 (17:52 +0100)]
iio: imu: st_lsm6dsx: enable sensor-hub support for lsm6dsm

Enabled i2c master controller support for LSM6DSM sensor. Enable
ext_sensor0 for lsm6dsm. This series has been tested using LIS2MDL as
slave device connected to the i2c controller of the LSM6DSM

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: imu: st_lsm6dsx: rename st_lsm6dsx_shub_read_reg in st_lsm6dsx_shub_read_output
Lorenzo Bianconi [Sat, 14 Dec 2019 16:52:58 +0000 (17:52 +0100)]
iio: imu: st_lsm6dsx: rename st_lsm6dsx_shub_read_reg in st_lsm6dsx_shub_read_output

Rename st_lsm6dsx_shub_read_reg routine in st_lsm6dsx_shub_read_output
since it is used to read from sensorhub channel0 output register

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: imu: st_lsm6dsx: check if shub_output reg is located in primary page
Lorenzo Bianconi [Sat, 14 Dec 2019 16:52:57 +0000 (17:52 +0100)]
iio: imu: st_lsm6dsx: check if shub_output reg is located in primary page

Check if the sensor hub output register is located in the primary or in the
secondary memory page. This is a preliminary patch to support i2c master
controller on lsm6dsm devices

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: imu: st_lsm6dsx: check if pull_up is located in primary page
Lorenzo Bianconi [Sat, 14 Dec 2019 16:52:56 +0000 (17:52 +0100)]
iio: imu: st_lsm6dsx: check if pull_up is located in primary page

Check if the pull up register is located in the primary or in the
secondary memory page. This is a preliminary patch to support i2c master
controller on lsm6dsm devices

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: imu: st_lsm6dsx: check if master_enable is located in primary page
Lorenzo Bianconi [Sat, 14 Dec 2019 16:52:55 +0000 (17:52 +0100)]
iio: imu: st_lsm6dsx: check if master_enable is located in primary page

Check if the master enable register is located in the primary or in the
secondary memory page. This is a preliminary patch to support i2c master
controller on lsm6dsm devices

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: imu: st_lsm6dsx: export max num of slave devices in st_lsm6dsx_shub_settings
Lorenzo Bianconi [Sat, 14 Dec 2019 16:52:54 +0000 (17:52 +0100)]
iio: imu: st_lsm6dsx: export max num of slave devices in st_lsm6dsx_shub_settings

Export max number of slave devices supported by the i2c master
controller in st_lsm6dsx_shub_settings data structure.
This is a preliminary patch to support i2c master controller
on lsm6dsm sensors

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: light: remove unneeded semicolon
Chen Zhou [Mon, 16 Dec 2019 10:51:01 +0000 (18:51 +0800)]
iio: light: remove unneeded semicolon

Fixes coccicheck warning:

./drivers/iio/light/lm3533-als.c:745:2-3: Unneeded semicolon

Signed-off-by: Chen Zhou <chenzhou10@huawei.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: chemical: atlas-ph-sensor: rename atlas-ph-sensor to atlas-sensor
Matt Ranostay [Mon, 16 Dec 2019 00:00:45 +0000 (16:00 -0800)]
iio: chemical: atlas-ph-sensor: rename atlas-ph-sensor to atlas-sensor

Since the orginal scope of the driver was to only support
the pH product from Atlas it has evolved to other sensors.

Rename the file, driver name, and regmap to atlas-sensor which
reflects this, although keep CONFIG_ATLAS_PH_SENSOR to not cause
regressions with current configurations.

Signed-off-by: Matt Ranostay <matt.ranostay@konsulko.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: st_sensors: Describe function parameters in kernel-doc
Andy Shevchenko [Mon, 16 Dec 2019 17:38:51 +0000 (19:38 +0200)]
iio: st_sensors: Describe function parameters in kernel-doc

Kernel documentation script complains that some of the function parameters
are not described:

.../common/st_sensors/st_sensors_trigger.c:29: warning: Function parameter or member 'indio_dev' not described in 'st_sensors_new_samples_available'
.../common/st_sensors/st_sensors_trigger.c:29: warning: Function parameter or member 'sdata' not described in 'st_sensors_new_samples_available'

Describe function parameters where it's appropriate.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: pressure: st_press: Mark default_press_pdata with __maybe_unused
Andy Shevchenko [Mon, 16 Dec 2019 17:38:50 +0000 (19:38 +0200)]
iio: pressure: st_press: Mark default_press_pdata with __maybe_unused

Since we put static variable to a header file it's copied to each module
that includes the header. But not all of them are actually used it.

Mark default_press_pdata with __maybe_unused to calm a compiler down:

In file included from drivers/iio/pressure/st_pressure_spi.c:18:
drivers/iio/pressure/st_pressure.h:40:46: warning: ‘default_press_pdata’ defined but not used [-Wunused-const-variable=]
   40 | static const struct st_sensors_platform_data default_press_pdata = {
      |                                              ^~~~~~~~~~~~~~~~~~~
...

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: accel: st_accel: Mark default_accel_pdata with __maybe_unused
Andy Shevchenko [Mon, 16 Dec 2019 17:38:49 +0000 (19:38 +0200)]
iio: accel: st_accel: Mark default_accel_pdata with __maybe_unused

Since we put static variable to a header file it's copied to each module
that includes the header. But not all of them are actually used it.

Mark default_accel_pdata with __maybe_unused to calm a compiler down:

In file included from drivers/iio/accel/st_accel_i2c.c:19:
drivers/iio/accel/st_accel.h:67:46: warning: ‘default_accel_pdata’ defined but not used [-Wunused-const-variable=]
   67 | static const struct st_sensors_platform_data default_accel_pdata = {
      |                                              ^~~~~~~~~~~~~~~~~~~
...

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: gyro: st_gyro: Mark gyro_pdata with __maybe_unused
Andy Shevchenko [Mon, 16 Dec 2019 17:38:48 +0000 (19:38 +0200)]
iio: gyro: st_gyro: Mark gyro_pdata with __maybe_unused

Since we put static variable to a header file it's copied to each module
that includes the header. But not all of them are actually used it.

Mark gyro_pdata with __maybe_unused to calm a compiler down:

In file included from drivers/iio/gyro/st_gyro_i2c.c:18:
drivers/iio/gyro/st_gyro.h:31:46: warning: ‘gyro_pdata’ defined but not used [-Wunused-const-variable=]
   31 | static const struct st_sensors_platform_data gyro_pdata = {
      |                                              ^~~~~~~~~~
...

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: pressure: st_press: Drop unnecessary #else branch for ACPI
Andy Shevchenko [Mon, 16 Dec 2019 17:38:47 +0000 (19:38 +0200)]
iio: pressure: st_press: Drop unnecessary #else branch for ACPI

ACPI_PTR() takes care of the argument in case of !CONFIG_ACPI.
Remove unnecessary #else branch.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: accel: st_accel: Drop unnecessary #else branch for ACPI
Andy Shevchenko [Mon, 16 Dec 2019 17:38:46 +0000 (19:38 +0200)]
iio: accel: st_accel: Drop unnecessary #else branch for ACPI

ACPI_PTR() takes care of the argument in case of !CONFIG_ACPI.
Remove unnecessary #else branch.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: light: st_uvis25: Drop unneeded header inclusion
Andy Shevchenko [Mon, 16 Dec 2019 17:38:45 +0000 (19:38 +0200)]
iio: light: st_uvis25: Drop unneeded header inclusion

There is no evidence that the driver supports or needs ACPI APIs.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: st_lsm6dsx: Make use of device properties
Andy Shevchenko [Mon, 16 Dec 2019 18:19:25 +0000 (20:19 +0200)]
iio: st_lsm6dsx: Make use of device properties

Device property API allows to gather device resources from different sources,
such as ACPI. Convert the drivers to unleash the power of device property API.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: st_lsm6dsx: Drop unneeded OF code
Andy Shevchenko [Mon, 16 Dec 2019 18:19:24 +0000 (20:19 +0200)]
iio: st_lsm6dsx: Drop unneeded OF code

There is no need to have OF guard against ID table.
Drop it for good.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: st_lsm6dsx: Mark predefined constants with __maybe_unused
Andy Shevchenko [Mon, 16 Dec 2019 18:19:23 +0000 (20:19 +0200)]
iio: st_lsm6dsx: Mark predefined constants with __maybe_unused

Since we put static variable to a header file it's copied to each module
that includes the header. But not all of them are actually used it.

Mark predefined constants with __maybe_unused to calm a compiler down:

In file included from drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_i2c.c:17:
.../st_lsm6dsx/st_lsm6dsx.h:399:28: warning: ‘st_lsm6dsx_available_scan_masks’ defined but not used [-Wunused-const-variable=]
  399 | static const unsigned long st_lsm6dsx_available_scan_masks[] = {0x7, 0x0};
      |                            ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.../st_lsm6dsx/st_lsm6dsx.h:392:36: warning: ‘st_lsm6dsx_event’ defined but not used [-Wunused-const-variable=]
  392 | static const struct iio_event_spec st_lsm6dsx_event = {
      |                                    ^~~~~~~~~~~~~~~~
...

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: adc: ad7923: Add support for the ad7908/ad7918/ad7928
Daniel Junho [Tue, 17 Dec 2019 11:11:58 +0000 (08:11 -0300)]
iio: adc: ad7923: Add support for the ad7908/ad7918/ad7928

The ad7928 is software compatible with the ad7923.
The ad7908 and ad7918 are the 8 and 10-bit versions of the ad7928.

Signed-off-by: Daniel Junho <djunho@gmail.com>
Reviewed-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: adc: ad7923: Add of_device_id table
Daniel Junho [Tue, 17 Dec 2019 11:11:57 +0000 (08:11 -0300)]
iio: adc: ad7923: Add of_device_id table

Accomplish device tree compatibility to driver AD7923
by adding of_device_id table and making a subsequent call to
MODULE_DEVICE_TABLE.

Signed-off-by: Daniel Junho <djunho@gmail.com>
Reviewed-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: adc: ad7923: Fix checkpatch warning
Daniel Junho [Tue, 17 Dec 2019 11:11:56 +0000 (08:11 -0300)]
iio: adc: ad7923: Fix checkpatch warning

Fix checkpatch warning:
WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
+static int ad7923_scan_direct(struct ad7923_state *st, unsigned ch)

Signed-off-by: Daniel Junho <djunho@gmail.com>
Reviewed-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: adc: ad7923: Remove the unused defines
Daniel Junho [Tue, 17 Dec 2019 11:11:55 +0000 (08:11 -0300)]
iio: adc: ad7923: Remove the unused defines

Removes the unused define AD7923_CHANNEL_x from the code.

Signed-off-by: Daniel Junho <djunho@gmail.com>
Reviewed-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: accel: bma400: basic regulator support
Dan Robertson [Fri, 20 Dec 2019 16:00:51 +0000 (16:00 +0000)]
iio: accel: bma400: basic regulator support

Add support for the VDD and VDDIO regulators using the regulator
framework.

Signed-off-by: Dan Robertson <dan@dlrobertson.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoiio: accel: Add driver for the BMA400
Dan Robertson [Fri, 20 Dec 2019 16:00:50 +0000 (16:00 +0000)]
iio: accel: Add driver for the BMA400

Add a IIO driver for the Bosch BMA400 3-axes ultra-low power accelerometer.
The driver supports reading from the acceleration and temperature
registers. The driver also supports reading and configuring the output data
rate, oversampling ratio, and scale.

Signed-off-by: Dan Robertson <dan@dlrobertson.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agodt-bindings: iio: accel: bma400: add bindings
Dan Robertson [Fri, 20 Dec 2019 16:00:49 +0000 (16:00 +0000)]
dt-bindings: iio: accel: bma400: add bindings

Add devicetree binding for the Bosch BMA400 3-axes ultra-low power
accelerometer sensor.

Signed-off-by: Dan Robertson <dan@dlrobertson.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
4 years agoMerge 5.5-rc3 into staging-next
Greg Kroah-Hartman [Mon, 23 Dec 2019 12:00:09 +0000 (07:00 -0500)]
Merge 5.5-rc3 into staging-next

We need the staging fixes in here as well.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agoLinux 5.5-rc3
Linus Torvalds [Mon, 23 Dec 2019 01:02:23 +0000 (17:02 -0800)]
Linux 5.5-rc3

4 years agoMerge branch 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Linus Torvalds [Mon, 23 Dec 2019 01:00:04 +0000 (17:00 -0800)]
Merge branch 'fixes' of git://git./linux/kernel/git/viro/vfs

Pull vfs fixes from Al Viro:
 "Eric's s_inodes softlockup fixes + Jan's fix for recent regression
  from pipe rework"

* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  fs: call fsnotify_sb_delete after evict_inodes
  fs: avoid softlockups in s_inodes iterators
  pipe: Fix bogus dereference in iov_iter_alignment()

4 years agoMerge tag 'xfs-5.5-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Linus Torvalds [Sun, 22 Dec 2019 18:59:06 +0000 (10:59 -0800)]
Merge tag 'xfs-5.5-fixes-2' of git://git./fs/xfs/xfs-linux

Pull xfs fixes from Darrick Wong:
 "Fix a few bugs that could lead to corrupt files, fsck complaints, and
  filesystem crashes:

   - Minor documentation fixes

   - Fix a file corruption due to read racing with an insert range
     operation.

   - Fix log reservation overflows when allocating large rt extents

   - Fix a buffer log item flags check

   - Don't allow administrators to mount with sunit= options that will
     cause later xfs_repair complaints about the root directory being
     suspicious because the fs geometry appeared inconsistent

   - Fix a non-static helper that should have been static"

* tag 'xfs-5.5-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: Make the symbol 'xfs_rtalloc_log_count' static
  xfs: don't commit sunit/swidth updates to disk if that would cause repair failures
  xfs: split the sunit parameter update into two parts
  xfs: refactor agfl length computation function
  libxfs: resync with the userspace libxfs
  xfs: use bitops interface for buf log item AIL flag check
  xfs: fix log reservation overflows when allocating large rt extents
  xfs: stabilize insert range start boundary to avoid COW writeback race
  xfs: fix Sphinx documentation warning

4 years agoMerge tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 22 Dec 2019 18:41:48 +0000 (10:41 -0800)]
Merge tag 'ext4_for_linus_stable' of git://git./linux/kernel/git/tytso/ext4

Pull ext4 bug fixes from Ted Ts'o:
 "Ext4 bug fixes, including a regression fix"

* tag 'ext4_for_linus_stable' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: clarify impact of 'commit' mount option
  ext4: fix unused-but-set-variable warning in ext4_add_entry()
  jbd2: fix kernel-doc notation warning
  ext4: use RCU API in debug_print_tree
  ext4: validate the debug_want_extra_isize mount option at parse time
  ext4: reserve revoke credits in __ext4_new_inode
  ext4: unlock on error in ext4_expand_extra_isize()
  ext4: optimize __ext4_check_dir_entry()
  ext4: check for directory entries too close to block end
  ext4: fix ext4_empty_dir() for directories with holes

4 years agoMerge tag 'block-5.5-20191221' of git://git.kernel.dk/linux-block
Linus Torvalds [Sun, 22 Dec 2019 18:36:55 +0000 (10:36 -0800)]
Merge tag 'block-5.5-20191221' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "Let's try this one again, this time without the compat_ioctl changes.
  We've got those fixed up, but that can go out next week.

  This contains:

   - block queue flush lockdep annotation (Bart)

   - Type fix for bsg_queue_rq() (Bart)

   - Three dasd fixes (Stefan, Jan)

   - nbd deadlock fix (Mike)

   - Error handling bio user map fix (Yang)

   - iocost fix (Tejun)

   - sbitmap waitqueue addition fix that affects the kyber IO scheduler
     (David)"

* tag 'block-5.5-20191221' of git://git.kernel.dk/linux-block:
  sbitmap: only queue kyber's wait callback if not already active
  block: fix memleak when __blk_rq_map_user_iov() is failed
  s390/dasd: fix typo in copyright statement
  s390/dasd: fix memleak in path handling error case
  s390/dasd/cio: Interpret ccw_device_get_mdc return value correctly
  block: Fix a lockdep complaint triggered by request queue flushing
  block: Fix the type of 'sts' in bsg_queue_rq()
  block: end bio with BLK_STS_AGAIN in case of non-mq devs and REQ_NOWAIT
  nbd: fix shutdown and recv work deadlock v2
  iocost: over-budget forced IOs should schedule async delay

4 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Linus Torvalds [Sun, 22 Dec 2019 18:26:59 +0000 (10:26 -0800)]
Merge tag 'for-linus' of git://git./virt/kvm/kvm

Pull KVM fixes from Paolo Bonzini:
 "PPC:
   - Fix a bug where we try to do an ultracall on a system without an
     ultravisor

  KVM:
   - Fix uninitialised sysreg accessor
   - Fix handling of demand-paged device mappings
   - Stop spamming the console on IMPDEF sysregs
   - Relax mappings of writable memslots
   - Assorted cleanups

  MIPS:
   - Now orphan, James Hogan is stepping down

  x86:
   - MAINTAINERS change, so long Radim and thanks for all the fish
   - supported CPUID fixes for AMD machines without SPEC_CTRL"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  MAINTAINERS: remove Radim from KVM maintainers
  MAINTAINERS: Orphan KVM for MIPS
  kvm: x86: Host feature SSBD doesn't imply guest feature AMD_SSBD
  kvm: x86: Host feature SSBD doesn't imply guest feature SPEC_CTRL_SSBD
  KVM: PPC: Book3S HV: Don't do ultravisor calls on systems without ultravisor
  KVM: arm/arm64: Properly handle faulting of device mappings
  KVM: arm64: Ensure 'params' is initialised when looking up sys register
  KVM: arm/arm64: Remove excessive permission check in kvm_arch_prepare_memory_region
  KVM: arm64: Don't log IMP DEF sysreg traps
  KVM: arm64: Sanely ratelimit sysreg messages
  KVM: arm/arm64: vgic: Use wrapper function to lock/unlock all vcpus in kvm_vgic_create()
  KVM: arm/arm64: vgic: Fix potential double free dist->spis in __kvm_vgic_destroy()
  KVM: arm/arm64: Get rid of unused arg in cpu_init_hyp_mode()

4 years agoMerge tag 'riscv/for-v5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv...
Linus Torvalds [Sun, 22 Dec 2019 18:22:47 +0000 (10:22 -0800)]
Merge tag 'riscv/for-v5.5-rc3' of git://git./linux/kernel/git/riscv/linux

Pull RISC-V fixes from Paul Walmsley:
 "Several fixes, and one cleanup, for RISC-V.

  Fixes:

   - Fix an error in a Kconfig file that resulted in an undefined
     Kconfig option "CONFIG_CONFIG_MMU"

   - Fix undefined Kconfig option "CONFIG_CONFIG_MMU"

   - Fix scratch register clearing in M-mode (affects nommu users)

   - Fix a mismerge on my part that broke the build for
     CONFIG_SPARSEMEM_VMEMMAP users

  Cleanup:

   - Move SiFive L2 cache-related code to drivers/soc, per request"

* tag 'riscv/for-v5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
  riscv: move sifive_l2_cache.c to drivers/soc
  riscv: define vmemmap before pfn_to_page calls
  riscv: fix scratch register clearing in M-mode.
  riscv: Fix use of undefined config option CONFIG_CONFIG_MMU

4 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Linus Torvalds [Sun, 22 Dec 2019 17:54:33 +0000 (09:54 -0800)]
Merge git://git./linux/kernel/git/netdev/net

Pull networking fixes from David Miller:

 1) Several nf_flow_table_offload fixes from Pablo Neira Ayuso,
    including adding a missing ipv6 match description.

 2) Several heap overflow fixes in mwifiex from qize wang and Ganapathi
    Bhat.

 3) Fix uninit value in bond_neigh_init(), from Eric Dumazet.

 4) Fix non-ACPI probing of nxp-nci, from Stephan Gerhold.

 5) Fix use after free in tipc_disc_rcv(), from Tuong Lien.

 6) Enforce limit of 33 tail calls in mips and riscv JIT, from Paul
    Chaignon.

 7) Multicast MAC limit test is off by one in qede, from Manish Chopra.

 8) Fix established socket lookup race when socket goes from
    TCP_ESTABLISHED to TCP_LISTEN, because there lacks an intervening
    RCU grace period. From Eric Dumazet.

 9) Don't send empty SKBs from tcp_write_xmit(), also from Eric Dumazet.

10) Fix active backup transition after link failure in bonding, from
    Mahesh Bandewar.

11) Avoid zero sized hash table in gtp driver, from Taehee Yoo.

12) Fix wrong interface passed to ->mac_link_up(), from Russell King.

13) Fix DSA egress flooding settings in b53, from Florian Fainelli.

14) Memory leak in gmac_setup_txqs(), from Navid Emamdoost.

15) Fix double free in dpaa2-ptp code, from Ioana Ciornei.

16) Reject invalid MTU values in stmmac, from Jose Abreu.

17) Fix refcount leak in error path of u32 classifier, from Davide
    Caratti.

18) Fix regression causing iwlwifi firmware crashes on boot, from Anders
    Kaseorg.

19) Fix inverted return value logic in llc2 code, from Chan Shu Tak.

20) Disable hardware GRO when XDP is attached to qede, frm Manish
    Chopra.

21) Since we encode state in the low pointer bits, dst metrics must be
    at least 4 byte aligned, which is not necessarily true on m68k. Add
    annotations to fix this, from Geert Uytterhoeven.

* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (160 commits)
  sfc: Include XDP packet headroom in buffer step size.
  sfc: fix channel allocation with brute force
  net: dst: Force 4-byte alignment of dst_metrics
  selftests: pmtu: fix init mtu value in description
  hv_netvsc: Fix unwanted rx_table reset
  net: phy: ensure that phy IDs are correctly typed
  mod_devicetable: fix PHY module format
  qede: Disable hardware gro when xdp prog is installed
  net: ena: fix issues in setting interrupt moderation params in ethtool
  net: ena: fix default tx interrupt moderation interval
  net/smc: unregister ib devices in reboot_event
  net: stmmac: platform: Fix MDIO init for platforms without PHY
  llc2: Fix return statement of llc_stat_ev_rx_null_dsap_xid_c (and _test_c)
  net: hisilicon: Fix a BUG trigered by wrong bytes_compl
  net: dsa: ksz: use common define for tag len
  s390/qeth: don't return -ENOTSUPP to userspace
  s390/qeth: fix promiscuous mode after reset
  s390/qeth: handle error due to unsupported transport mode
  cxgb4: fix refcount init for TC-MQPRIO offload
  tc-testing: initial tdc selftests for cls_u32
  ...

4 years agopipe: fix empty pipe check in pipe_write()
Jan Stancek [Sun, 22 Dec 2019 12:33:24 +0000 (13:33 +0100)]
pipe: fix empty pipe check in pipe_write()

LTP pipeio_1 test is hanging with v5.5-rc2-385-gb8e382a185eb,
with read side observing empty pipe and sleeping and write
side running out of space and then sleeping as well. In this
scenario there are 5 writers and 1 reader.

Problem is that after pipe_write() reacquires pipe lock, it
re-checks for empty pipe with potentially stale 'head' and
doesn't wake up read side anymore. pipe->tail can advance
beyond 'head', because there are multiple writers.

Use pipe->head for empty pipe check after reacquiring lock
to observe current state.

Testing: With patch, LTP pipeio_1 ran successfully in loop for 1 hour.
         Without patch it hanged within a minute.

Fixes: 1b6b26ae7053 ("pipe: fix and clarify pipe write wakeup logic")
Reported-by: Rachel Sibley <rasibley@redhat.com>
Signed-off-by: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
4 years agoMerge tag 'kvm-ppc-fixes-5.5-1' of git://git.kernel.org/pub/scm/linux/kernel/git...
Paolo Bonzini [Sun, 22 Dec 2019 12:18:15 +0000 (13:18 +0100)]
Merge tag 'kvm-ppc-fixes-5.5-1' of git://git./linux/kernel/git/paulus/powerpc into kvm-master

PPC KVM fix for 5.5

- Fix a bug where we try to do an ultracall on a system without an
  ultravisor.

4 years agoMAINTAINERS: remove Radim from KVM maintainers
Paolo Bonzini [Wed, 4 Dec 2019 14:33:35 +0000 (15:33 +0100)]
MAINTAINERS: remove Radim from KVM maintainers

Radim's kernel.org email is bouncing, which I take as a signal that
he is not really able to deal with KVM at this time.  Make MAINTAINERS
match the effective value of KVM's bus factor.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
4 years agoMAINTAINERS: Orphan KVM for MIPS
James Hogan [Sat, 21 Dec 2019 15:50:13 +0000 (15:50 +0000)]
MAINTAINERS: Orphan KVM for MIPS

I haven't been active for 18 months, and don't have the hardware set up
to test KVM for MIPS, so mark it as orphaned and remove myself as
maintainer. Hopefully somebody from MIPS can pick this up.

Signed-off-by: James Hogan <jhogan@kernel.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Radim Krčmář" <rkrcmar@redhat.com>
Cc: Paul Burton <paulburton@kernel.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: kvm@vger.kernel.org
Cc: linux-mips@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
4 years agoext4: clarify impact of 'commit' mount option
Jan Kara [Wed, 18 Dec 2019 11:12:10 +0000 (12:12 +0100)]
ext4: clarify impact of 'commit' mount option

The description of 'commit' mount option dates back to ext3 times.
Update the description to match current meaning for ext4.

Reported-by: Paul Richards <paul.richards@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20191218111210.14161-1-jack@suse.cz
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
4 years agoext4: fix unused-but-set-variable warning in ext4_add_entry()
Yunfeng Ye [Tue, 17 Dec 2019 14:46:49 +0000 (22:46 +0800)]
ext4: fix unused-but-set-variable warning in ext4_add_entry()

Warning is found when compile with "-Wunused-but-set-variable":

fs/ext4/namei.c: In function ‘ext4_add_entry’:
fs/ext4/namei.c:2167:23: warning: variable ‘sbi’ set but not used
[-Wunused-but-set-variable]
  struct ext4_sb_info *sbi;
                       ^~~
Fix this by moving the variable @sbi under CONFIG_UNICODE.

Signed-off-by: Yunfeng Ye <yeyunfeng@huawei.com>
Reviewed-by: Ritesh Harjani <riteshh@linux.ibm.com>
Link: https://lore.kernel.org/r/cb5eb904-224a-9701-c38f-cb23514b1fff@huawei.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
4 years agoMerge tag 'trace-v5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt...
Linus Torvalds [Sat, 21 Dec 2019 23:16:56 +0000 (15:16 -0800)]
Merge tag 'trace-v5.5-rc2' of git://git./linux/kernel/git/rostedt/linux-trace

Pull tracing fixes from Steven Rostedt:

 - Fix memory leak on error path of process_system_preds()

 - Lock inversion fix with updating tgid recording option

 - Fix histogram compare function on big endian machines

 - Fix histogram trigger function on big endian machines

 - Make trace_printk() irq sync on init for kprobe selftest correctness

* tag 'trace-v5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing: Fix endianness bug in histogram trigger
  samples/trace_printk: Wait for IRQ work to finish
  tracing: Fix lock inversion in trace_event_enable_tgid_record()
  tracing: Have the histogram compare functions convert to u64 first
  tracing: Avoid memory leak in process_system_preds()

4 years agoMerge tag 'libnvdimm-fix-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 21 Dec 2019 23:12:26 +0000 (15:12 -0800)]
Merge tag 'libnvdimm-fix-5.5-rc3' of git://git./linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm fix from Dan Williams:
 "A minor regression fix.

  The libnvdimm unit tests were expecting to mock calls to
  ioremap_nocache() which disappeared in v5.5-rc1. This fix has appeared
  in -next and collided with some cleanups that Christoph has planned
  for v5.6, but he will fix up his branch once this goes in.

  Summary:

   - Restore the operation of the libnvdimm unit tests after the removal
     of ioremap_nocache()"

* tag 'libnvdimm-fix-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  tools/testing/nvdimm: Fix mock support for ioremap

4 years agotracing: Fix endianness bug in histogram trigger
Sven Schnelle [Wed, 18 Dec 2019 07:44:27 +0000 (08:44 +0100)]
tracing: Fix endianness bug in histogram trigger

At least on PA-RISC and s390 synthetic histogram triggers are failing
selftests because trace_event_raw_event_synth() always writes a 64 bit
values, but the reader expects a field->size sized value. On little endian
machines this doesn't hurt, but on big endian this makes the reader always
read zero values.

Link: http://lore.kernel.org/linux-trace-devel/20191218074427.96184-4-svens@linux.ibm.com
Cc: stable@vger.kernel.org
Fixes: 4b147936fa509 ("tracing: Add support for 'synthetic' events")
Acked-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
4 years agosamples/trace_printk: Wait for IRQ work to finish
Sven Schnelle [Wed, 18 Dec 2019 07:44:26 +0000 (08:44 +0100)]
samples/trace_printk: Wait for IRQ work to finish

trace_printk schedules work via irq_work_queue(), but doesn't
wait until it was processed. The kprobe_module.tc testcase does:

:;: "Load module again, which means the event1 should be recorded";:
modprobe trace-printk
grep "event1:" trace

so the grep which checks the trace file might run before the irq work
was processed. Fix this by adding a irq_work_sync().

Link: http://lore.kernel.org/linux-trace-devel/20191218074427.96184-3-svens@linux.ibm.com
Cc: stable@vger.kernel.org
Fixes: af2a0750f3749 ("selftests/ftrace: Improve kprobe on module testcase to load/unload module")
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
4 years agotracing: Fix lock inversion in trace_event_enable_tgid_record()
Prateek Sood [Tue, 10 Dec 2019 09:15:16 +0000 (09:15 +0000)]
tracing: Fix lock inversion in trace_event_enable_tgid_record()

       Task T2                             Task T3
trace_options_core_write()            subsystem_open()

 mutex_lock(trace_types_lock)           mutex_lock(event_mutex)

 set_tracer_flag()

   trace_event_enable_tgid_record()       mutex_lock(trace_types_lock)

    mutex_lock(event_mutex)

This gives a circular dependency deadlock between trace_types_lock and
event_mutex. To fix this invert the usage of trace_types_lock and
event_mutex in trace_options_core_write(). This keeps the sequence of
lock usage consistent.

Link: http://lkml.kernel.org/r/0101016eef175e38-8ca71caf-a4eb-480d-a1e6-6f0bbc015495-000000@us-west-2.amazonses.com
Cc: stable@vger.kernel.org
Fixes: d914ba37d7145 ("tracing: Add support for recording tgid of tasks")
Signed-off-by: Prateek Sood <prsood@codeaurora.org>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
4 years agoMerge tag 's390-5.5-4' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Linus Torvalds [Sat, 21 Dec 2019 20:17:14 +0000 (12:17 -0800)]
Merge tag 's390-5.5-4' of git://git./linux/kernel/git/s390/linux

Pull s390 fixes from Vasily Gorbik:

 - Fix unwinding from irq context of interrupted user process.

 - Add purgatory build missing symbols check. That helped to uncover and
   fix missing symbols when built with kasan support enabled.

 - Couple of ftrace fixes. Avoid broken stack trace and fix recursion
   loop in function_graph tracer.

* tag 's390-5.5-4' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
  s390/ftrace: save traced function caller
  s390/unwind: stop gracefully at user mode pt_regs in irq stack
  s390/purgatory: do not build purgatory with kcov, kasan and friends
  s390/purgatory: Make sure we fail the build if purgatory has missing symbols
  s390/ftrace: fix endless recursion in function_graph tracer

4 years agoMerge branch 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 21 Dec 2019 18:52:10 +0000 (10:52 -0800)]
Merge branch 'sched-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull scheduler fixes from Ingo Molnar:
 "Misc fixes: a (rare) PSI crash fix, a CPU affinity related balancing
  fix, and a toning down of active migration attempts"

* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/cfs: fix spurious active migration
  sched/fair: Fix find_idlest_group() to handle CPU affinity
  psi: Fix a division error in psi poll()
  sched/psi: Fix sampling error and rare div0 crashes with cgroups and high uptime

4 years agoMerge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 21 Dec 2019 18:51:00 +0000 (10:51 -0800)]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull perf fixes from Ingo Molnar:
 "Misc fixes: a BTS fix, a PT NMI handling fix, a PMU sysfs fix and an
  SRCU annotation"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/core: Add SRCU annotation for pmus list walk
  perf/x86/intel: Fix PT PMI handling
  perf/x86/intel/bts: Fix the use of page_private()
  perf/x86: Fix potential out-of-bounds access

4 years agoMerge tag 'kbuild-fixes-v5.5' of git://git.kernel.org/pub/scm/linux/kernel/git/masahi...
Linus Torvalds [Sat, 21 Dec 2019 18:49:47 +0000 (10:49 -0800)]
Merge tag 'kbuild-fixes-v5.5' of git://git./linux/kernel/git/masahiroy/linux-kbuild

Pull Kbuild fixes from Masahiro Yamada:

 - fix warning in out-of-tree 'make clean'

 - add READELF variable to the top Makefile

 - fix broken builds when LINUX_COMPILE_BY contains a backslash

 - fix build warning in kallsyms

 - fix NULL pointer access in expr_eq() in Kconfig

 - fix missing dependency on rsync in deb-pkg build

 - remove ---help--- from documentation

 - fix misleading documentation about directory descending

* tag 'kbuild-fixes-v5.5' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
  kbuild: clarify the difference between obj-y and obj-m w.r.t. descending
  kconfig: remove ---help--- from documentation
  scripts: package: mkdebian: add missing rsync dependency
  kconfig: don't crash on NULL expressions in expr_eq()
  scripts/kallsyms: fix offset overflow of kallsyms_relative_base
  mkcompile_h: use printf for LINUX_COMPILE_BY
  mkcompile_h: git rid of UTS_TRUNCATE from LINUX_COMPILE_{BY,HOST}
  x86/boot: kbuild: allow readelf executable to be specified
  kbuild: fix 'No such file or directory' warning when cleaning

4 years agokbuild: clarify the difference between obj-y and obj-m w.r.t. descending
Masahiro Yamada [Thu, 19 Dec 2019 11:51:00 +0000 (20:51 +0900)]
kbuild: clarify the difference between obj-y and obj-m w.r.t. descending

Kbuild descends into a directory by either 'y' or 'm', but there is an
important difference.

Kbuild combines the built-in objects into built-in.a in each directory.
The built-in.a in the directory visited by obj-y is merged into the
built-in.a in the parent directory. This merge happens recursively
when Kbuild is ascending back towards the top directory, then built-in
objects are linked into vmlinux eventually. This works properly only
when the Makefile specifying obj-y is reachable by the chain of obj-y.

On the other hand, Kbuild does not take built-in.a from the directory
visited by obj-m. This it, all the objects in that directory are
supposed to be modular. If Kbuild descends into a directory by obj-m,
but the Makefile in the sub-directory specifies obj-y, those objects
are just left orphan.

The current statement "Kbuild only uses this information to decide that
it needs to visit the directory" is misleading. Clarify the difference.

Reported-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Johan Hovold <johan@kernel.org>
4 years agoMerge branch 'parisc-5.5-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller...
Linus Torvalds [Sat, 21 Dec 2019 14:49:41 +0000 (06:49 -0800)]
Merge branch 'parisc-5.5-2' of git://git./linux/kernel/git/deller/parisc-linux

Pul parisc fixes from Helge Deller:
 "Two build error fixes, one for the soft_offline_page() parameter
  change and one for a specific KEXEC/KEXEC_FILE configuration, as well
  as a compiler and a linker warning fix"

* 'parisc-5.5-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: Fix compiler warnings in debug_core.c
  parisc: soft_offline_page() now takes the pfn
  parisc: add missing __init annotation
  parisc: fix compilation when KEXEC=n and KEXEC_FILE=y

4 years agoMerge tag 'for-linus-5.5b-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 21 Dec 2019 14:24:56 +0000 (06:24 -0800)]
Merge tag 'for-linus-5.5b-rc3-tag' of git://git./linux/kernel/git/xen/tip

Pull xen fixes from Juergen Gross:
 "This contains two cleanup patches and a small series for supporting
  reloading the Xen block backend driver"

* tag 'for-linus-5.5b-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/grant-table: remove multiple BUG_ON on gnttab_interface
  xen-blkback: support dynamic unbind/bind
  xen/interface: re-define FRONT/BACK_RING_ATTACH()
  xenbus: limit when state is forced to closed
  xenbus: move xenbus_dev_shutdown() into frontend code...
  xen/blkfront: Adjust indentation in xlvbd_alloc_gendisk

4 years agoMerge tag 'powerpc-5.5-4' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc...
Linus Torvalds [Sat, 21 Dec 2019 14:17:05 +0000 (06:17 -0800)]
Merge tag 'powerpc-5.5-4' of git://git./linux/kernel/git/powerpc/linux

Pull powerpc fixes from Michael Ellerman:
 "Two weeks worth of accumulated fixes:

   - A fix for a performance regression seen on PowerVM LPARs using
     dedicated CPUs, caused by our vcpu_is_preempted() returning true
     even for idle CPUs.

   - One of the ultravisor support patches broke KVM on big endian hosts
     in v5.4.

   - Our KUAP (Kernel User Access Prevention) code missed allowing
     access in __clear_user(), which could lead to an oops or erroneous
     SEGV when triggered via PTRACE_GETREGSET.

   - Two fixes for the ocxl driver, an open/remove race, and a memory
     leak in an error path.

   - A handful of other small fixes.

  Thanks to: Andrew Donnellan, Christian Zigotzky, Christophe Leroy,
  Christoph Hellwig, Daniel Axtens, David Hildenbrand, Frederic Barrat,
  Gautham R. Shenoy, Greg Kurz, Ihor Pasichnyk, Juri Lelli, Marcus
  Comstedt, Mike Rapoport, Parth Shah, Srikar Dronamraju, Vaidyanathan
  Srinivasan"

* tag 'powerpc-5.5-4' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  KVM: PPC: Book3S HV: Fix regression on big endian hosts
  powerpc: Fix __clear_user() with KUAP enabled
  powerpc/pseries/cmm: fix managed page counts when migrating pages between zones
  powerpc/8xx: fix bogus __init on mmu_mapin_ram_chunk()
  ocxl: Fix potential memory leak on context creation
  powerpc/irq: fix stack overflow verification
  powerpc: Ensure that swiotlb buffer is allocated from low memory
  powerpc/shared: Use static key to detect shared processor
  powerpc/vcpu: Assume dedicated processors as non-preempt
  ocxl: Fix concurrent AFU open and device removal

4 years agoMerge branch 'ras-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 21 Dec 2019 14:04:12 +0000 (06:04 -0800)]
Merge branch 'ras-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull x86 RAS fixes from Borislav Petkov:
 "Three urgent RAS fixes for the AMD side of things:

   - initialize struct mce.bank so that calculated error severity on AMD
     SMCA machines is correct

   - do not send IPIs early during bank initialization, when interrupts
     are disabled

   - a fix for when only a subset of MCA banks are enabled, which led to
     boot hangs on some new AMD CPUs"

* 'ras-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mce: Fix possibly incorrect severity calculation on AMD
  x86/MCE/AMD: Allow Reserved types to be overwritten in smca_banks[]
  x86/MCE/AMD: Do not use rdmsr_safe_on_cpu() in smca_configure()

4 years agoMerge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 21 Dec 2019 13:55:35 +0000 (05:55 -0800)]
Merge tag 'clk-fixes-for-linus' of git://git./linux/kernel/git/clk/linux

Pull clk fixes from Stephen Boyd:
 "One core framework fix to walk the orphan list and match up clks to
  parents when clk providers register the DT provider after registering
  all their clks (as they should).

  Then a handful of driver fixes for the qcom, imx, and at91 drivers.

  The driver fixes are relatively small fixes for incorrect register
  settings or missing locks causing race conditions"

* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
  clk: qcom: Avoid SMMU/cx gdsc corner cases
  clk: qcom: gcc-sc7180: Fix setting flag for votable GDSCs
  clk: Move clk_core_reparent_orphans() under CONFIG_OF
  clk: at91: fix possible deadlock
  clk: walk orphan list on clock provider registration
  clk: imx: pll14xx: fix clk_pll14xx_wait_lock
  clk: imx: clk-imx7ulp: Add missing sentinel of ulp_div_table
  clk: imx: clk-composite-8m: add lock to gate/mux

4 years agoMerge branch 'sfc-fix-bugs-introduced-by-XDP-patches'
David S. Miller [Sat, 21 Dec 2019 05:56:48 +0000 (21:56 -0800)]
Merge branch 'sfc-fix-bugs-introduced-by-XDP-patches'

Edward Cree says:

====================
sfc: fix bugs introduced by XDP patches

Two fixes for bugs introduced by the XDP support in the sfc driver.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agosfc: Include XDP packet headroom in buffer step size.
Charles McLachlan [Fri, 20 Dec 2019 16:27:10 +0000 (16:27 +0000)]
sfc: Include XDP packet headroom in buffer step size.

Correct a mismatch between rx_page_buf_step and the actual step size
used when filling buffer pages.

This patch fixes the page overrun that occured when the MTU was set to
anything bigger than 1692.

Fixes: 3990a8fffbda ("sfc: allocate channels for XDP tx queues")
Signed-off-by: Charles McLachlan <cmclachlan@solarflare.com>
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agosfc: fix channel allocation with brute force
Edward Cree [Fri, 20 Dec 2019 16:26:40 +0000 (16:26 +0000)]
sfc: fix channel allocation with brute force

It was possible for channel allocation logic to get confused between what
 it had and what it wanted, and end up trying to use the same channel for
 both PTP and regular TX.  This led to a kernel panic:
    BUG: unable to handle page fault for address: 0000000000047635
    #PF: supervisor write access in kernel mode
    #PF: error_code(0x0002) - not-present page
    PGD 0 P4D 0
    Oops: 0002 [#1] SMP PTI
    CPU: 0 PID: 0 Comm: swapper/0 Tainted: G        W         5.4.0-rc3-ehc14+ #900
    Hardware name: Dell Inc. PowerEdge R710/0M233H, BIOS 6.4.0 07/23/2013
    RIP: 0010:native_queued_spin_lock_slowpath+0x188/0x1e0
    Code: f3 90 48 8b 32 48 85 f6 74 f6 eb e8 c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 98 02 00 48 03 04 f5 a0 c6 ed 81 <48> 89 10 8b 42 08 85 c0 75 09 f3 90 8b 42 08 85 c0 74 f7 48 8b 32
    RSP: 0018:ffffc90000003d28 EFLAGS: 00010006
    RAX: 0000000000047635 RBX: 0000000000000246 RCX: 0000000000040000
    RDX: ffff888627a298c0 RSI: 0000000000003ffe RDI: ffff88861f6b8dd4
    RBP: ffff8886225c6e00 R08: 0000000000040000 R09: 0000000000000000
    R10: 0000000616f080c6 R11: 00000000000000c0 R12: ffff88861f6b8dd4
    R13: ffffc90000003dc8 R14: ffff88861942bf00 R15: ffff8886150f2000
    FS:  0000000000000000(0000) GS:ffff888627a00000(0000) knlGS:0000000000000000
    CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
    CR2: 0000000000047635 CR3: 000000000200a000 CR4: 00000000000006f0
    Call Trace:
     <IRQ>
     _raw_spin_lock_irqsave+0x22/0x30
     skb_queue_tail+0x1b/0x50
     sock_queue_err_skb+0x9d/0xf0
     __skb_complete_tx_timestamp+0x9d/0xc0
     efx_dequeue_buffer+0x126/0x180 [sfc]
     efx_xmit_done+0x73/0x1c0 [sfc]
     efx_ef10_ev_process+0x56a/0xfe0 [sfc]
     ? tick_sched_do_timer+0x60/0x60
     ? timerqueue_add+0x5d/0x70
     ? enqueue_hrtimer+0x39/0x90
     efx_poll+0x111/0x380 [sfc]
     ? rcu_accelerate_cbs+0x50/0x160
     net_rx_action+0x14a/0x400
     __do_softirq+0xdd/0x2d0
     irq_exit+0xa0/0xb0
     do_IRQ+0x53/0xe0
     common_interrupt+0xf/0xf
     </IRQ>

In the long run we intend to rewrite the channel allocation code, but for
 'net' fix this by allocating extra_channels, and giving them TX queues,
 even if we do not in fact need them (e.g. on NICs without MAC TX
 timestamping), and thereby using simpler logic to assign the channels
 once they're allocated.

Fixes: 3990a8fffbda ("sfc: allocate channels for XDP tx queues")
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agonet: dst: Force 4-byte alignment of dst_metrics
Geert Uytterhoeven [Fri, 20 Dec 2019 13:31:40 +0000 (14:31 +0100)]
net: dst: Force 4-byte alignment of dst_metrics

When storing a pointer to a dst_metrics structure in dst_entry._metrics,
two flags are added in the least significant bits of the pointer value.
Hence this assumes all pointers to dst_metrics structures have at least
4-byte alignment.

However, on m68k, the minimum alignment of 32-bit values is 2 bytes, not
4 bytes.  Hence in some kernel builds, dst_default_metrics may be only
2-byte aligned, leading to obscure boot warnings like:

    WARNING: CPU: 0 PID: 7 at lib/refcount.c:28 refcount_warn_saturate+0x44/0x9a
    refcount_t: underflow; use-after-free.
    Modules linked in:
    CPU: 0 PID: 7 Comm: ksoftirqd/0 Tainted: G        W         5.5.0-rc2-atari-01448-g114a1a1038af891d-dirty #261
    Stack from 10835e6c:
    10835e6c 0038134f 00023fa6 00394b0f 0000001c 00000009 00321560 00023fea
    00394b0f 0000001c 001a70f8 00000009 00000000 10835eb4 00000001 00000000
    04208040 0000000a 00394b4a 10835ed4 00043aa8 001a70f8 00394b0f 0000001c
    00000009 00394b4a 0026aba8 003215a4 00000003 00000000 0026d5a8 00000001
    003215a4 003a4361 003238d6 000001f0 00000000 003215a4 10aa3b00 00025e84
    003ddb00 10834000 002416a8 10aa3b00 00000000 00000080 000aa038 0004854a
    Call Trace: [<00023fa6>] __warn+0xb2/0xb4
     [<00023fea>] warn_slowpath_fmt+0x42/0x64
     [<001a70f8>] refcount_warn_saturate+0x44/0x9a
     [<00043aa8>] printk+0x0/0x18
     [<001a70f8>] refcount_warn_saturate+0x44/0x9a
     [<0026aba8>] refcount_sub_and_test.constprop.73+0x38/0x3e
     [<0026d5a8>] ipv4_dst_destroy+0x5e/0x7e
     [<00025e84>] __local_bh_enable_ip+0x0/0x8e
     [<002416a8>] dst_destroy+0x40/0xae

Fix this by forcing 4-byte alignment of all dst_metrics structures.

Fixes: e5fd387ad5b30ca3 ("ipv6: do not overwrite inetpeer metrics prematurely")
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agoselftests: pmtu: fix init mtu value in description
Hangbin Liu [Fri, 20 Dec 2019 07:08:06 +0000 (15:08 +0800)]
selftests: pmtu: fix init mtu value in description

There is no a_r3, a_r4 in the testing topology.
It should be b_r1, b_r2. Also b_r1 mtu is 1400 and b_r2 mtu is 1500.

Fixes: e44e428f59e4 ("selftests: pmtu: add basic IPv4 and IPv6 PMTU tests")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Acked-by: Stefano Brivio <sbrivio@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agohv_netvsc: Fix unwanted rx_table reset
Haiyang Zhang [Fri, 20 Dec 2019 02:28:10 +0000 (18:28 -0800)]
hv_netvsc: Fix unwanted rx_table reset

In existing code, the receive indirection table, rx_table, is in
struct rndis_device, which will be reset when changing MTU, ringparam,
etc. User configured receive indirection table values will be lost.

To fix this, move rx_table to struct net_device_context, and check
netif_is_rxfh_configured(), so rx_table will be set to default only
if no user configured value.

Fixes: ff4a44199012 ("netvsc: allow get/set of RSS indirection table")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agonet: phy: ensure that phy IDs are correctly typed
Russell King [Thu, 19 Dec 2019 23:24:52 +0000 (23:24 +0000)]
net: phy: ensure that phy IDs are correctly typed

PHY IDs are 32-bit unsigned quantities. Ensure that they are always
treated as such, and not passed around as "int"s.

Fixes: 13d0ab6750b2 ("net: phy: check return code when requesting PHY driver module")
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agomod_devicetable: fix PHY module format
Russell King [Thu, 19 Dec 2019 23:24:47 +0000 (23:24 +0000)]
mod_devicetable: fix PHY module format

When a PHY is probed, if the top bit is set, we end up requesting a
module with the string "mdio:-10101110000000100101000101010001" -
the top bit is printed to a signed -1 value. This leads to the module
not being loaded.

Fix the module format string and the macro generating the values for
it to ensure that we only print unsigned types and the top bit is
always 0/1. We correctly end up with
"mdio:10101110000000100101000101010001".

Fixes: 8626d3b43280 ("phylib: Support phy module autoloading")
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agoqede: Disable hardware gro when xdp prog is installed
Manish Chopra [Thu, 19 Dec 2019 18:35:16 +0000 (10:35 -0800)]
qede: Disable hardware gro when xdp prog is installed

commit 18c602dee472 ("qede: Use NETIF_F_GRO_HW.") introduced
a regression in driver that when xdp program is installed on
qede device, device's aggregation feature (hardware GRO) is not
getting disabled, which is unexpected with xdp.

Fixes: 18c602dee472 ("qede: Use NETIF_F_GRO_HW.")
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Ariel Elior <aelior@marvell.com>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agoMerge branch 'ena-fixes-of-interrupt-moderation-bugs'
David S. Miller [Sat, 21 Dec 2019 05:43:09 +0000 (21:43 -0800)]
Merge branch 'ena-fixes-of-interrupt-moderation-bugs'

Arthur Kiyanovski says:

====================
ena: fixes of interrupt moderation bugs

Differences from V1:
1. Updated default tx interrupt moderation to 64us
2. Added "Fixes:" tags.
3. Removed cosmetic changes that are not relevant for these bug fixes

This patchset includes a couple of fixes of bugs in the implemenation of
interrupt moderation.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agonet: ena: fix issues in setting interrupt moderation params in ethtool
Arthur Kiyanovski [Thu, 19 Dec 2019 15:40:56 +0000 (17:40 +0200)]
net: ena: fix issues in setting interrupt moderation params in ethtool

Issue 1:
--------
Reproduction steps:
1. sudo ethtool -C eth0 rx-usecs 128
2. sudo ethtool -C eth0 adaptive-rx on
3. sudo ethtool -C eth0 adaptive-rx off
4. ethtool -c eth0

expected output: rx-usecs 128
actual output: rx-usecs 0

Reason for issue:
In stage 3, ethtool userspace calls first the ena_get_coalesce() handler
to get the current value of all properties, and then the ena_set_coalesce()
handler. When ena_get_coalesce() is called the adaptive interrupt
moderation is still on. There is an if in the code that returns the
rx_coalesce_usecs only if the adaptive interrupt moderation is off.
And since it is still on, rx_coalesce_usecs is not set, meaning it
stays 0.

Solution to issue:
Remove this if static interrupt moderation intervals have nothing to do
with dynamic ones.

Issue 2:
--------
Reproduction steps:
1. sudo ethtool -C eth0 adaptive-rx on
2. sudo ethtool -C eth0 rx-usecs 128
3. ethtool -c eth0

expected output: rx-usecs 128
actual output: rx-usecs 0

Reason for issue:
In stage 2, when ena_set_coalesce() is called, the handler tests if
rx adaptive interrupt moderation is on, and if it is, it returns before
getting to the part in the function that sets the rx non-adaptive
interrupt moderation interval.

Solution to issue:
Remove the return from the function when rx adaptive interrupt moderation
is on.

Also cleaned up the fixed code in ena_set_coalesce by grouping together
adaptive interrupt moderation toggling, and using && instead of nested
ifs.

Fixes: b3db86dc4b82 ("net: ena: reimplement set/get_coalesce()")
Fixes: 0eda847953d8 ("net: ena: fix retrieval of nonadaptive interrupt moderation intervals")
Fixes: 1738cd3ed342 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)")
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agonet: ena: fix default tx interrupt moderation interval
Arthur Kiyanovski [Thu, 19 Dec 2019 15:40:55 +0000 (17:40 +0200)]
net: ena: fix default tx interrupt moderation interval

Current default non-adaptive tx interrupt moderation interval is 196 us.
This value is too high and might cause the tx queue to fill up.

In this commit we set the default non-adaptive tx interrupt moderation
interval to 64 us in order to:
1. Reduce the probability of the queue filling-up (when compared to the
   current default value of 196 us).
2. Reduce unnecessary tx interrupt overhead (which happens if we set the
   default tx interval to 0).
   We determined experimentally that 64 us is an optimal value that
   reduces interrupt rate by more than 20% without affecting performance.

Fixes: 1738cd3ed342 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)")
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agonet/smc: unregister ib devices in reboot_event
Karsten Graul [Thu, 19 Dec 2019 11:51:13 +0000 (12:51 +0100)]
net/smc: unregister ib devices in reboot_event

In the reboot_event handler, unregister the ib devices and enable
the IB layer to release the devices before the reboot.

Fixes: a33a803cfe64 ("net/smc: guarantee removal of link groups in reboot")
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Reviewed-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agonet: stmmac: platform: Fix MDIO init for platforms without PHY
Padmanabhan Rajanbabu [Thu, 19 Dec 2019 10:17:01 +0000 (15:47 +0530)]
net: stmmac: platform: Fix MDIO init for platforms without PHY

The current implementation of "stmmac_dt_phy" function initializes
the MDIO platform bus data, even in the absence of PHY. This fix
will skip MDIO initialization if there is no PHY present.

Fixes: 7437127 ("net: stmmac: Convert to phylink and remove phylib logic")
Acked-by: Jayati Sahu <jayati.sahu@samsung.com>
Signed-off-by: Sriram Dash <sriram.dash@samsung.com>
Signed-off-by: Padmanabhan Rajanbabu <p.rajanbabu@samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agollc2: Fix return statement of llc_stat_ev_rx_null_dsap_xid_c (and _test_c)
Chan Shu Tak, Alex [Thu, 19 Dec 2019 06:16:18 +0000 (14:16 +0800)]
llc2: Fix return statement of llc_stat_ev_rx_null_dsap_xid_c (and _test_c)

When a frame with NULL DSAP is received, llc_station_rcv is called.
In turn, llc_stat_ev_rx_null_dsap_xid_c is called to check if it is a NULL
XID frame. The return statement of llc_stat_ev_rx_null_dsap_xid_c returns 1
when the incoming frame is not a NULL XID frame and 0 otherwise. Hence, a
NULL XID response is returned unexpectedly, e.g. when the incoming frame is
a NULL TEST command.

To fix the error, simply remove the conditional operator.

A similar error in llc_stat_ev_rx_null_dsap_test_c is also fixed.

Signed-off-by: Chan Shu Tak, Alex <alexchan@task.com.hk>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agonet: hisilicon: Fix a BUG trigered by wrong bytes_compl
Jiangfeng Xiao [Thu, 19 Dec 2019 02:08:07 +0000 (10:08 +0800)]
net: hisilicon: Fix a BUG trigered by wrong bytes_compl

When doing stress test, we get the following trace:
kernel BUG at lib/dynamic_queue_limits.c:26!
Internal error: Oops - BUG: 0 [#1] SMP ARM
Modules linked in: hip04_eth
CPU: 0 PID: 2003 Comm: tDblStackPcap0 Tainted: G           O L  4.4.197 #1
Hardware name: Hisilicon A15
task: c3637668 task.stack: de3bc000
PC is at dql_completed+0x18/0x154
LR is at hip04_tx_reclaim+0x110/0x174 [hip04_eth]
pc : [<c041abfc>]    lr : [<bf0003a8>]    psr: 800f0313
sp : de3bdc2c  ip : 00000000  fp : c020fb10
r10: 00000000  r9 : c39b4224  r8 : 00000001
r7 : 00000046  r6 : c39b4000  r5 : 0078f392  r4 : 0078f392
r3 : 00000047  r2 : 00000000  r1 : 00000046  r0 : df5d5c80
Flags: Nzcv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
Control: 32c5387d  Table: 1e189b80  DAC: 55555555
Process tDblStackPcap0 (pid: 2003, stack limit = 0xde3bc190)
Stack: (0xde3bdc2c to 0xde3be000)
[<c041abfc>] (dql_completed) from [<bf0003a8>] (hip04_tx_reclaim+0x110/0x174 [hip04_eth])
[<bf0003a8>] (hip04_tx_reclaim [hip04_eth]) from [<bf0012c0>] (hip04_rx_poll+0x20/0x388 [hip04_eth])
[<bf0012c0>] (hip04_rx_poll [hip04_eth]) from [<c04c8d9c>] (net_rx_action+0x120/0x374)
[<c04c8d9c>] (net_rx_action) from [<c021eaf4>] (__do_softirq+0x218/0x318)
[<c021eaf4>] (__do_softirq) from [<c021eea0>] (irq_exit+0x88/0xac)
[<c021eea0>] (irq_exit) from [<c0240130>] (msa_irq_exit+0x11c/0x1d4)
[<c0240130>] (msa_irq_exit) from [<c0267ba8>] (__handle_domain_irq+0x110/0x148)
[<c0267ba8>] (__handle_domain_irq) from [<c0201588>] (gic_handle_irq+0xd4/0x118)
[<c0201588>] (gic_handle_irq) from [<c0558360>] (__irq_svc+0x40/0x58)
Exception stack(0xde3bdde0 to 0xde3bde28)
dde0: 00000000 00008001 c3637668 00000000 00000000 a00f0213 dd3627a0 c0af6380
de00: c086d380 a00f0213 c0a22a50 de3bde6c 00000002 de3bde30 c0558138 c055813c
de20: 600f0213 ffffffff
[<c0558360>] (__irq_svc) from [<c055813c>] (_raw_spin_unlock_irqrestore+0x44/0x54)
Kernel panic - not syncing: Fatal exception in interrupt

Pre-modification code:
int hip04_mac_start_xmit(struct sk_buff *skb, struct net_device *ndev)
{
[...]
[1] priv->tx_head = TX_NEXT(tx_head);
[2] count++;
[3] netdev_sent_queue(ndev, skb->len);
[...]
}
An rx interrupt occurs if hip04_mac_start_xmit just executes to the line 2,
tx_head has been updated, but corresponding 'skb->len' has not been
added to dql_queue.

And then
hip04_mac_interrupt->__napi_schedule->hip04_rx_poll->hip04_tx_reclaim

In hip04_tx_reclaim, because tx_head has been updated,
bytes_compl will plus an additional "skb-> len"
which has not been added to dql_queue. And then
trigger the BUG_ON(bytes_compl > num_queued - dql->num_completed).

To solve the problem described above, we put
"netdev_sent_queue(ndev, skb->len);"
before
"priv->tx_head = TX_NEXT(tx_head);"

Fixes: a41ea46a9a12 ("net: hisilicon: new hip04 ethernet driver")
Signed-off-by: Jiangfeng Xiao <xiaojiangfeng@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
4 years agonet: dsa: ksz: use common define for tag len
Michael Grzeschik [Wed, 18 Dec 2019 16:01:39 +0000 (17:01 +0100)]
net: dsa: ksz: use common define for tag len

Remove special taglen define KSZ8795_INGRESS_TAG_LEN
and use generic KSZ_INGRESS_TAG_LEN instead.

Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>