The MMAL and V4L2 buffers had been disassociated, and linked on
demand.  Seeing as both are finite and low in number, and we now have
the same number of each, link them for the duration.  This removes the
complexity of maintaining lists as the struct mmal_buffer context
comes back from the VPU, so we can directly link back to the relevant
V4L2 buffer.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

For historical reasons, the number of buffers passed to the VPU over
MMAL did not match that passed from V4L2.  That is a silly situation
as the driver has to duplicate serialisation and other functions that
have already been implemented in V4L2/videobuf2.

As we had more V4L2 buffers than MMAL ones, the MMAL buffer headers
were returned to the VPU immediately on being filled, which is now
invalid.

Match the number of buffers notified in queue_setup with that used in
MMAL.  Return buffers only when we get them from V4L2.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

There is no requirement to serialise bulk transfers as that is all
done in VCHI, and if a second MMAL_MSG_TYPE_BUFFER_TO_HOST happened
before the VCHI_CALLBACK_BULK_RECEIVED, then the service_callback
thread is deadlocked.

Remove the bulk_mutex so that multiple receives can be scheduled at a
time.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The struct mmal_msg_context was being allocated for every message
being sent to the VPU, and freed when it came back.  Whilst that is
required behaviour for some messages (mainly the synchronous ones), it
is wasteful for the video buffers that make up the majority of the
traffic.

Add to the buffer_init/cleanup hooks that it allocates/frees the
msg_context required.

v2: changes by anholt from the downstream tree: clean up indentation,
    pass an error value through, forward-declare the struct so we have
    less void *

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Interleaved RGB and single plane YUV formats can be delivered by the
GPU without the secondary step of removing padding, as the
bytesperline field can be set appropriately.

Planar YUV needs the GPU to still remove padding, as there is no way
to report that there is padding between the planes (ie on the height).
The multi-planar formats are NOT applicable, as there is no easy way
to make them contiguous in memory (ie one large allocation that gets
broken up). The whole task is passed across to videobuf2 which has no
notion of that requirement.

v2: Changes by anholt from the downstream driver: Flag two more planar
    formats as needing padding removal, and remove broken userspace
    workaround.

Signed-off-by: Dave Stevenson <dave.stevenson@raspberrypi.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This allows bcm2835-camera to automatically probe after VCHI has
loaded, rather than only successfully probing if the arbitrary probe
order chooses us after VCHI.

Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

We had the camera driver set up in a module_init function, but that
meant that the camera driver would fail to load if it was initialized
before VCHI.  By attaching to this platform_device, it can get a
defined load order.

Signed-off-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Rx handler (assigned with netdev_rx_handler_register,
called from __netif_receive_skb()) uses value of dev->rx_handler_data.
The driver has no rx handler and does not need it, so remove rx_handler_data read.

Signed-off-by: Ivan Safonov <insafonov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Add a static prefix to the declaration for libcfs_dev. This would fix
the following sparse warning:

drivers/staging/lustre/lnet/libcfs/module.c:317:19: warning: symbol
'libcfs_dev' was not declared. Should it be static?

Signed-off-by: Justin Skists <justin.skists@juzza.co.uk>
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Trivial fix to spelling mistake in ODM_RT_TRACE message text

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch removes debugging code in video.c that causes the following
checkpatch warning:

WARNING: Prefer using '\"%s...\", __func__' to using function's name in a string

Signed-off-by: Abdun Nihaal <abdun.nihaal@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This satisfies a checkpatch.pl warning and is the preferred method for
notating the license due to its lack of ambiguity.

Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge tag 'iio-for-4.18a' of git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next

Jonathan writes:

1st round of IIO new device support, features and cleanup for the 4.18 cycle

A nice mix this time of excellent cleanups (many to send drivers
speeding toward staging graduations) and new drivers / device support.
A good part of this is Brian Masney's never ending task on the tsl2x7x
driver.  The end is in sight so hopefully we'll get that one out of
staging very soon!

New device support
* AD5686
  - Support AD5685R (was wrongly present as AD5685)
  - Support AD5672R, AD5676, AD5676, AD5684R and AD5686R 4 and 8 channel
    SPI DACs with various precisions.
  - Support AD5671R, AD5675R, AD5694, AD5694R, AD5695R, AD5696 and AD5696R
    I2C DACs with various percisions and numbers of channels.
* Analog front end rescale driver - New driver.
  - Support current sensing usings a shunt resistor.
  - Support simple voltage dividers.
  - support simple current sense amplifiers.
* TI dac5571
  - New driver and device bindings supporting:
    dac5571, dac6571, dac7571, dac5574, dac6574, dac7574,
    dac5573, dac6573 and dac7573
* Meson-adc
  - Support for Meson AXG with DT bindings.
* mpu6050
  - Support the mpu9255 which only requires additional WHOAMI entry and
    compatible string.
* st_lsm6dsx
  - Support for lsm330dlc combinded accelerometer and gyro sensors with
    DT bindings.
* stm32_adc
  - Add support for STM32MP1 with bindings.

Staging graduations
* adis16201 after some excelent cleanup by Himanshu Jha.
* adis16029 after some excelent cleanup by Shreeya Patel.

New features:
* ABI docs
  - Add core ABI docs for angle channels.
* inv_mpu6050
  - Provide support for the full range of interrupts the device
    supports.
* st_accel
  - Add SMO8840 ACPI ID seen in the wild on some Lenovo machines.
* stx104
  - Provide a multiple gpio get function.

Cleanups / Minor fixes
* core
  - Use new nested structure support to improve kernel-doc.
* ad2s1200
  - Use be16_to_cpup instead of opencoding.
* ad5686
  - Indentation tidy up.
  - Switch to SPDX
  - Refactor to allow various numbers of channels.
  - Refactor to separate core and SPI specific support, prior to
    addition of i2c equivalent devices.
* ad7606
  - Use drvdata directly from device rather than boucing via the
    platform_device structure.
* ad7746
  - Replace opencoded byte swapped i2c calls with _swapped variants.
  - White space and line break readability improvements.
  - Reorder includes and variable declarations where appropriate.
* ad7791
  - Changes to the AD ADC library used by this driver took in the
    sampling frequency.  This lead to be the wrong path being the one
    tied to the resulting attribute, so it didn't work, and a warning
    to be printed.
* ad7780
  - Remove apparent support for sampling frequency control on devices
    that don't support changing the sampling attributes.
* ade7854
  - Fix a read of the wrong number of bits.
  - Improve error handling on i2c read/write errors.
  - Rework i2c and spi code to reduce duplication.
* adis16201 (staging)
  - Improve meaning inherent in some macro names by adding units etc
    where relevant.
  - Adjust comments to improve detail and drop the irrelevant.
  - Rename register address definitions definitions to add a _REG
    postfix, clearly separating them from field definitions. Reorganize
    the definitions to group register address and fields.
  - Use sign_extend32 rather than open coding.
  - Reverse Xmas tree ordering where appropriate and align function args.
  - Remove unused headers.
  - Use GENMASK where appropriate instead of open coding.
* adis16209 (staging)
  - Indent field definitions to visually separate them from
    register address definitions.
  - Use reverse xmas tree ordering where appropriate.
  - Add some whitespace where it will help readability.
  - Drop some unused headers.
  - Use GENMASK where appropriate.
* ad2s1200
  - Drop unnecessary includes and reorder alphabetically.
  - Reverse xmas tree and blank line cleanups.
* atlas-ph-sensor
  - Use msleep instead of usleep_range where the precise value doesn't
    matter and the delays are long.
* bcm150
  - Drop transaction splitting as core now handles it.
* cros_ec
  - Move the shared header to the include/iio/common directory.
    This brings it inline with the other multiple type devices.
  - Use drvdata directly from device rather than boucing via the
    platform_device structure.
* hid-sensors
  - Use drvdata directly from device rather than boucing via the
    platform_device structure.
* inv_mpu6050
  - Clear out a second function definition for the same function.
  - Don't flush fifo when the iio buffer is full but just drop excess
    data.
  - Tidy up set_power_itg and ensure it is used in the right places.
  - Use set_power_itg rather than opencoding it again in the i2c mux
    control.
  - Make sure error paths disable the power if undoing power on.
  - Used managed devm_ functions during probe. Delete remove function.
  - Refactor to pull raw data read out of read_raw function.
  - Simplify data reading error paths.
  - Only enable the i2c mux for chips with the i2c aux bus (not icm20608)
  - Fix a potential deadlock due to varying lock ordering.
  - Fix an issue where first sample from gyro after enabling is unstable
    by dropping the first sample.
  - Fix an issue where the user_ctrl register is incorrectly overwritten.
  - Tidy up some grammar and spelling minor issus.
* mcp320x
  - Use vendor compatible strings.
* mcp4018
  - Switch to using i2c .probe_new.
* mcp4351
  - switch to using i2c .probe_new.
* meson-adc
  - rework handing on common ADC platform data so it can be shared
    across multiple families of SoCs.
* sca3000
  - Fix an error handling path if the ring configure fails.
* st_lsm6dsx
  - Fix a wrong fifo threshold mask (no actual effect)
* stm32-dfsdm
  - Style fixes and cleanups.
  - Check filter ID is in range and check spi-max-frequency.
* tsl2x7x (staging)
  - Drop some unnecessary function calls, unused variables and
    unnecessary local variables.
  - Fix wrong interrupt type.
  - Avoid unnecessary double clear of interrupt.
  - Simplify proximity calibration call which did various things
    unrelated to actually calibrating.
  - Separate control of the proximity and ALS interrupts.
  - Improve consistency of logging.
  - Separate ALS and proximity persistence settings as they have
    separate hardware controls.
  - Tidy up variable ordering.
  - Add Brian to copyright notice given consider work on this driver.
  - Take advantage of hardware support for I2C address auto increment.
  - Combine individuaal enable and period attributes for the two
    directions on the threshold events into a single value as the
    hardware doesn't separate them.
  - Move integration_time* attributes from light channel to
    intensity value as they effect the intensity readings directly
    and the light reading only indirectly.  Hence this better
    reflects reality. Also move the calibscale_available.
  - Avoid returning an error in the IRQ handler.
  - Hard code the reg value in _clear_interrupts as it only takes
    one value in the code.   Result is the function has little
    purpose so opencode the two remaining i2c_smbus_write_byte
    calls.
  - Drop some unnecessary checking of the chip status register.
  - Tidy up return path in _write_interrupt_config.
  - Tidy up the ID verification code.
  - Move the power and diode settings defines into the header as these
    are needed for platform data configuration.
  - Various renames and comment cleanups for consistency and clarity.
  - Use actual device defaults for default startup settings.
  - SPDX
  - Add some range sanity checking to sysfs attribute writes.
  - Don't provide event interfaces if the interrupt line isn't available.
  - Use IIO_CONST_ATTR macro for calibscale_available as it's a constant
    string.
  - Fix the integration time and lux equations.
  - Make device IDs explicit index values in the device_channel_config array.

Fix alignment issues reported by checkpatch.pl.

Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Signed-off-by: NeilBrown <neil@brown.name>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Add missing blank lines after declarations to solve checkpatch.pl errors.

Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Reviewed-by: NeilBrown <neil@brown.name>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Remove un-necessary blank lines to solve errors found by checkpatch.pl.

Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Reviewed-by: NeilBrown <neil@brown.name>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Fix the following comilation error by making sure that "phy_ring_head" is
defined as "dma_addr_t" and change "phy_ring_tail" to match it as both
should be "dma_addr_t".

error: passing argument 3 of ‘dma_alloc_coherent’ from incompatible pointer type

Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Reviewed-by: NeilBrown <neil@brown.name>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

OPA driver optimizations are based on the MPI model where it is
expected to have multiple endpoints between two given nodes. To
enable this optimization for Lustre, we need to make it possible,
via an LND-specific tuneable, to create multiple endpoints and to
balance the traffic over them.

Both sides of a connection must have this patch for it to work.
Only the active side of the connection (usually the client)
needs to have the new tuneable set > 1.

Signed-off-by: Doug Oucharek <doug.s.oucharek@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8943
Reviewed-on: https://review.whamcloud.com/25168


Reviewed-by: Amir Shehata <amir.shehata@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Doug Oucharek <dougso@me.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

uwrq is an unsigned 32-bit integer, it cannot be less than zero.

Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch removes the local variable dev that is used to store the pointer
to the usb_device whenever it is used only once.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The functions usb_disconnect and usb_sndbulkpipe are racing for the struct
usb_device, which might cause a null pointer dereference exception. This
patch fixes this race condition by protecting the critical section inside
the function hdm_enque with the io_mutex.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch avoids setting the URB_ZERO_PACKET transfer flag for synchronous
data. This is needed to prevent the host from sending an empty packet when
data is aligned to an endpoint packet boundary.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

On several architectures the allocation of coherent memory needs a device
that has the dma_ops structure properly initialized. This patch enables
the DIM2 platform to be used to allocate this type of memory.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch implemets reading of the clock speed from DT.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch removes the dependency to platform specific source files
that do platform specific initialization and supply the IRQ number.
Instead DT code is added

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch fixes a race condition between the functions disconnect and poll.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Platform specific initialization (data->init) has to be done before
calling dim_startup to start the DIM2 IP.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The function ch_get_mbo declares its return value as type bool,
but returns a pointer to mbo.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch adds the endpoint number of the USB pipe that reports to be
broken into the log message. It is needed to make debugging for
applications more comfortable.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The function unregister_chrdev_region is called with a different counter
as the alloc_chrdev_region. To fix this, this patch introduces the
constant CHRDEV_REGION_SIZE that is used in both functions.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch avoids the warning that the pointer mbo might be used
uninitialized that some environmens throw.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch is needed as function snd_card_new needs a valid
parent device. Passing a NULL pointer leads to kernel Ooops.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

On arm64/aarch64 architectures the allocation of coherent memory needs a
device that has the dma_ops properly set. That's why the core module of
the MOST driver is no longer able to allocate this type or memory. This
patch moves the allocation process down to the interface drivers where
the proper devices exist (e.g. platform device or USB system software).

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Currently, there are two module parameters for the i2c driver:
  - polling_req: boolean irq/polling mode;
  - scan_rate: polling rate, that is used in the case where the polling
    mode is active

This model is misconfiguration-prone.  For example, it is possible to
select polling mode with the zero polling rate or configure non-zero
polling rate in a combination with the IRQ mode.

This patch replaces the 'polling_req' and 'scan_rate' by the
'polling_rate', where the value zero means the interrupt driven mode and
other values are used as the polling rate in the polling mode.

Signed-off-by: Andrey Shvetsov <andrey.shvetsov@k2l.de>
Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The elements of the dev->rx.list are consumed in the pending_rx_work and
populated in the function enqueue() that cancels the pending_rx_work.

The function enqueue() and poison_channel() do not race anyway.

Signed-off-by: Andrey Shvetsov <andrey.shvetsov@k2l.de>
Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The variable is_open is checked only in the work function
pending_rx_work() that is only active between the calls
configure_channel() and poison_channel().

Signed-off-by: Andrey Shvetsov <andrey.shvetsov@k2l.de>
Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

The real polling rate depends on the CONFIG_HZ and may differ from the
required polling rate.

Signed-off-by: Andrey Shvetsov <andrey.shvetsov@k2l.de>
Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch avoids that a configured scan_rate of more than MSEC_PER_SEC
might result in a polling delay of zero.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch prevents the driver from falling back to polling mode
in case of IRQ misconfiguration.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch removes the function wait_event_interruptible from the
work function to avoid waiting.

Signed-off-by: Christian Gromm <christian.gromm@microchip.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>