Minchan Kim [Mon, 28 Jan 2013 01:00:08 +0000 (10:00 +0900)]
staging: zsmalloc: Fix TLB coherency and build problem
Recently, Matt Sealey reported he fail to build zsmalloc caused by
using of local_flush_tlb_kernel_range which are architecture dependent
function so !CONFIG_SMP in ARM couldn't implement it so it ends up
build error following as.
The reason we used that function is copy method by [1]
was really slow in ARM but at that time.
More severe problem is ARM can prefetch speculatively on other CPUs
so under us, other TLBs can have an entry only if we do flush local
CPU. Russell King pointed that. Thanks!
We don't have many choices except using flush_tlb_kernel_range.
My experiment in ARMv7 processor 4 core didn't make any difference with
zsmapbench[2] between local_flush_tlb_kernel_range and flush_tlb_kernel_range
but still page-table based is much better than copy-based.
This patch replace local_flush_tlb_kernel_range with
flush_tlb_kernel_range which are avaialbe in all architectures
because we already have used it in vmalloc allocator which are
generic one so build problem should go away and performane loss
shoud be void.
Seth Jennings [Fri, 25 Jan 2013 17:46:18 +0000 (11:46 -0600)]
staging: zsmalloc: make CLASS_DELTA relative to PAGE_SIZE
Right now ZS_SIZE_CLASS_DELTA is hardcoded to be 16. This
creates 254 classes for systems with 4k pages. However, on
PPC64 with 64k pages, it creates 4095 classes which is far
too many.
This patch makes ZS_SIZE_CLASS_DELTA relative to PAGE_SIZE
so that regardless of the page size, there will be the same
number of classes.
Acked-by: Nitin Gupta <ngupta@vflare.org> Acked-by: Minchan Kim <minchan@kernel.org> Signed-off-by: Seth Jennings <sjenning@linux.vnet.ibm.com> Acked-by: Dan Magenheimer <dan.magenheimer@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Peter Huewe [Mon, 28 Jan 2013 23:36:39 +0000 (00:36 +0100)]
staging/comedi/adl_pci8164: Don't assign string
Assigning a string is really bad,
and since we only have 1 char strings here we
can simply use a char to store the value and change the format specifier.
Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Ian Abbott [Tue, 29 Jan 2013 16:20:17 +0000 (16:20 +0000)]
staging: comedi: simplify comedi_set_hw_dev()
Since `get_device()` and `put_device()` can take a NULL device
parameter, `comedi_set_hw_dev()` can be simplified to always call
`get_device()` for the new, possibly NULL hardware device, and
`put_device()` for the old, possibly NULL hardware device. As long as
we do it in that order, there shouldn't be any problem with
`kref_release()` getting called unexpectedly when the new hardware
device is the old hardware device.
Simplify `comedi_set_hw_dev()` and update the comment because the
function is used for additional purposes since the old comment was
written.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Ian Abbott [Tue, 29 Jan 2013 14:05:58 +0000 (14:05 +0000)]
staging: comedi: don't override read/write subdevice if not supported
For comedi devices that support asynchronous commands on some of their
subdevices, the comedi core creates extra device files for each of those
subdevices of the form "/dev/comedi%i_subd%i". These use the same
comedi device as the corresponding "/dev/comedi%i" but override the
default "read" and "write" subdevice for the device. Currently it
overrides both the read and write subdevice, but it only makes sense to
override the "read" subdevice if the subdevice supports "read" commands,
and to override the "write" subdevice if the subdevice supports "write"
commands.
In `comedi_alloc_subdevice_minor()`, only set `info->read_subdevice`
non-NULL if the subdevice has the `SDF_CMD_READ` flag set, and only set
`info->write_subdevice` non-NULL if the subdevice has the
`SDF_CMD_WRITE` flag set. (`comedi_read_subdevice(info)` will use the
device's default read subdevice if `info->read_subdevice` is NULL.
`comedi_write_subdevice(info)` will use the device's default write
subdevice if `info->write_subdevice` is NULL.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: ni_labpc_cs: cleanup the boardinfo declaration
For aesthetic reasons, reformat the boardinfo declaration and
add some whitespace. Remove all the information that is set to
'0' as this is the default.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For aesthetic reasons, move all the pcmcia_driver functions so they
are near the pcmcia_driver declaration. This also removes the need
for some of the forward declarations.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: ni_labpc_cs: remove 'stop' from private pcmcia data
The pcmcia_driver suspend and remove functions set the 'stop' variable
and the resume function clears it. Nothing in the comedi_driver code
uses the 'stop' variable.
Just remove it so we can get rid of the suspend/resume.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: ni_mio_cs: cleanup the boardinfo declaration
For aesthetic reasons, reformat the boardinfo declaration and
add some whitespace. Remove all the information that is set to
'0' as this is the default.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove the DPRINTK() function trace message as well as the #if 0'ed
out debug code that dumps the board "fingerprint".
Remove the need for the local variable that holds the link->irq
passed to request_irq(). Also, return the error code from that
function instead of assuming -EINVAL. Use the dev->board_name for
the resource string passed to request_irq() instead of the open
coded string "ni_mio_cs".
For aesthetic reasons, add some whitespace to the initializatio
of the devpriv values.
Just return the result of ni_E_init() instead of checking it for
an error, returning the error, or returning "0".
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For aesthetic reasons, move all the pcmcia_driver functions so they
are near the pcmcia_driver declaration. This also removes the need
for a couple forward declarations.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For aesthetic reasons, add some whitespace to the subdevice init.
Remove the init of the s->len_chanlist for the subdevices that do
not support commands. It's not used by them and the comedi core
will handle initializing it properly in the postconfig.
Change the return after a successful attach to "0". The comedi
core expects a < 0 value to indicate an error and "0" is the
typical value returned to indicate success.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: quatech_daqp_cs: fix possible memory dereference issue
In daqp_attach(), the first options value passed in the comedi_devconfig
is used as an index to the private dev_table[] in this driver. This table
is used to pass the pcmcia_device to the comedi_driver.
Fix the code so that the index is checked before the table is accessed
so that we don't get a possible memory dereference BUG.
Change the error returned to the comedi core from -EIO to -ENODEV.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Merge tag 'iio-for-3.9b-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next
Jonathan writes:
Second set of IIO new drivers, cleanups and fixes for the 3.9 cycle.
This second version is due to a little fixup for an include
path being added to the tsl2563 move out of staging after
a report from Fengguang Wu and the 0-day kernel build testing
backend.
Minor bits:
1) A Kconfig dependency fix for the max1363 driver that
has been causing some autobuilder fails in Linux Next.
2) Removal of a stale makefile entry
3) Fix an incorrect arguement for a sizeof call
4) Duplicate code removal in tsl2x7x driver
5) A missing spin lock init in hid-sensor-time
New features:
1) mxs adc driver gains support for touchscreen special functions
2) mxs driver gainst supprot for the MX23 and dt entries added
3) adis16400 gains adis16448 support and some additional bells and whistles
Moves out of staging.
1) adis16400 - a venerable driver gets a make over and moves
out of staging.
2) Kxsd9 moved out fo staging
3) adis16080 gets cleaned up and moved out of staging
4) tsl2563 gets a little cleaned up and move out of staging
Removals
1) sw_ring is killed off with all remaining drivers converted to kfifo.
This has been scheduled for a long time since we switched to kfifo.
There is demand for a high performance alternative, but this was
never it and I'm glad to see this vestage of IIOs youth gone once
and for all!
Jonathan Cameron [Sat, 12 Jan 2013 10:35:00 +0000 (10:35 +0000)]
iio:light:tsl2563 move out of staging
This driver is simple, uses the latest interfaces and contains few if
any controversial elements. All of its interfaces have been in place
for a long time now. Hence let's move it out of staging.
Signed-off-by: Jonathan Cameron <jic23@kernel.org> Acked-by: Peter Meerwald <pmeerw@pmeerw.net>
The driver is rather simple and in a good shape. It follows the IIO ABI and the
standard codechecker tools do not report any issues, so move it out of staging.
While moving it also remove one outdated 'fixme' comment.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
All sections in which the transfer buffer is accessed are already protected by
the IIO device's mlock. So we do not need the extra mutex protecting the buffer.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
The sample buffer contains big endian 16bit words. So use the be16 datatype for
the buffer and use the proper helper functions for endianness conversion instead
of openconding it.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
During sampling the driver currently does a spi_read followed by a spi_write.
This is not a problem per se, since CS needs to be deasserted between the two
transfers. So even if another device claims the bus between the two transfers we
should still get a result. But the code is actually spread out over multiple
functions. E.g. the spi_read happens in one function the spi_write in another,
this makes the code harder to follow. This patch re-factors the code to just use
a single spi transaction to do both the read and the write transfer.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
staging:iio:adis16080: Add device id table entry for the adis16100
The adis16100 is very similar to the adis16080. The driver description already
states that the driver supports the adis16100 as-well. But so far the there is
no device id table for the adis16100 and the drivers does not bind to a device
named adis16100.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Marek Vasut [Mon, 21 Jan 2013 20:05:00 +0000 (20:05 +0000)]
iio: mxs: Add MX23 support into the IIO driver
This patch adds support for i.MX23 into the LRADC driver. The LRADC
block on MX23 is not much different from the one on MX28, thus this
is only a few changes fixing the parts that are specific to MX23.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Fabio Estevam <fabio.estevam@freescale.com> Cc: Jonathan Cameron <jic23@kernel.org> Cc: Shawn Guo <shawn.guo@linaro.org> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Jonathan Cameron [Fri, 30 Nov 2012 14:22:00 +0000 (14:22 +0000)]
staging:iio: drop sw_ring buffer implementation.
Whilst this is IIO's oldest buffer implementation it is messy, poorly
implemented and whilst it works, no one is entirely sure it always will.
New IIO drivers have not been using this for some time and now all remaining
old users have been converted to use the kfifo based alternative.
Clearly a fifo isn't the same as a ring buffer but in many use cases it
really doesn't matter. We also loose the watershed based poll implementation.
However having poll effectively report data only when the buffer was half
full was at best an 'unusual' use of the interface.
At somepoint in the future we may bring watersheds back on a different
buffer implementation, but then we will think a lot more about how to do
the interface first.
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Jonathan Cameron [Fri, 30 Nov 2012 14:22:00 +0000 (14:22 +0000)]
staging:iio:meter:ade7758 switch from sw_ring to kfifo
sw ring is going away so switch over to kfifo based buffer implementation.
The only real change is that poll will return on some data there rather than
buffer 50% full.
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
The adis16448 is more or less from the same family of devices as supported by
this driver. It features three acceleration channels, three angular velocity
channels, three magnetometer channels, one temperature channels and one
barometric pressure channel.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
iio:adis16400: Expose some debug information in debugfs
Expose some information useful for debugging a device in debugfs. This includes
for now the flash count, the product id and the serial number and raw register
access.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
iio:adis16400: Add support for the 52.85 Hz base sampling rate
The adis16400 and similar have two different base sampling rate available, from
which the actual sampling rate is derived. 1638 Hz and 52.85 Hz, switching to
the lower base sampling rate allows to support lower sampling rates.
This patch adds support for switching to the lower base sampling rate if the
requested sampling frequency is outside of the range which can be supported by
the higher base sampling rate.
The function which is used to read the current sampling rate already has support
for the lower sampling rate, so no changes are required there.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
The devices supported by this drivers support sample rates with less than one
sample per second. To support this increase the samplerate precession to allow
setting (and reading) the samplerate with a milli-HZ precession.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Do a set of minor miscellaneous code style cleanups for the adis16400 before
moving it out of staging. Delete outdated comments, removed excess whitespace,
add missing whitespace, replace u{8,16} with uint{8,16}_t.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
While the samplerate supported by the devices which are supported by this driver
is not continuous it supports a wide range, much more than currently listed in
the samplerate_available attribute. Also it accepts all values written to the
samplerate attribute and will round-up them to the nearest supported sample
rate. So remove the samplerate_available attribute.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
staging:iio:adis16400: Preallocate transfer message
Similar to like we already did for the generic adis library preallocate and
pre-construct the SPI transfer message for the adis16400. For devices which do
not support burst mode sampling does not differ from other adis devices and so
we use the generic functions of the adis library in this case. In burst mode we
can only sample all channels at once, so use the IIO cores demux facility
instead of doing this manually.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
staging:iio:adis16400: Add helper macros for channel declaration
Most of the channels declared in the adis16400 driver look quite similar. This
patch adds a bunch of helper macros to initialize the channel spec for this
driver. This allows us to drastically reduce the number of lines of code needed
for the channel spec declaration.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
staging:iio:adis16400: Use triggered buffer setup helper function
Use the triggered buffer helper functions to setup and tear down the buffer for
the adis16400 instead of doing this manually. This also means that we switch
away from the deprecated sw_ring buffer and use the kfifo buffer now instead.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Use the new adis library for the adis16400 driver. This allows us to completely
scrap the adis16400 trigger code and more than half of the core driver code. For
now we can not make use of the generic adis buffer implementation since the
adis16400 driver has special requirements due to its burst mode support. But
we will eventually get to this.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
staging:iio:adis16400: Fix and cleanup 3db filter setting
The 3db divisors table is partially wrong and incomplete. Also the code rounds
up to the next higher frequency if the requested frequency would matches one of
the available frequencies. These two issues are fixed by this patch. The patch
also changes the driver to round down the filter frequency if it is larger than
the largest supported frequency instead of rejecting it as an invalid value.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
ilog2 is not defined for 0, so we need to handle the case where the requested
frequency is larger than the base sampling rate. In this case we'll round down
and set the sampling rate to the base sampling rate.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
staging:iio:tsl2x7x: Use iio_str_to_fixedpoint instead of open-coding it
The tsl2x7x driver has a copy'n'pasted version of the iio_str_to_fixedpoint()
function from the IIO core. Replace this custom copy and use
iio_str_to_fixedpoint instead.
The patch also introduces a slight functional change in that it makes sure that
in case of a parsing error the error is reported back to userspace instead of
silently ignoring it.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Acked-by: Jon Brenner <jon.brenner@ams.com> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Marek Vasut [Fri, 11 Jan 2013 23:35:00 +0000 (23:35 +0000)]
iio: mxs: Implement support for touchscreen
This patch implements support for sampling of a touchscreen into
the MXS LRADC driver. The LRADC block allows configuring some of
it's channels into special mode where they either output the drive
voltage or sample it, allowing it to operate a 4-wire or 5-wire
resistive touchscreen.
In case the touchscreen mode is enabled, the LRADC slot #7 is
reserved for touchscreen only, therefore it is not possible to
sample 8 LRADC channels at time, but only 7 channels.
The touchscreen controller is configured such that the PENDOWN event
disables touchscreen interrupts and triggers execution of worker
thread, which then polls the touchscreen controller for X, Y and
Pressure values. This reduces the overhead of interrupt-driven
operation. Upon the PENUP event, the worker thread re-enables the
PENDOWN detection interrupt and exits.
Signed-off-by: Marek Vasut <marex@denx.de> Acked-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Commit a301d425e ("staging:iio:gyro remove adis16251 driver as now supported by
adis16260 driver") removed the adis16251, but left its Makefile entry intact.
This patch removes the unused Makefile entry.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Jonathan Cameron <jic23@kernel.org>
staging: comedi: addi_watchdog: all i/o registers are 32-bit
All the i/o registers used by the watchdog device on the addi-data
boards are 32-bit. Make sure all the i/o commands use outl/inl to
access the registers.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Peter Huewe [Tue, 22 Jan 2013 22:40:03 +0000 (23:40 +0100)]
staging/comedi: Use comedi_pci_auto_unconfig directly for pci_driver.remove
(Almost) all comedi pci drivers have some wrapper for their
pci_driver.remove function which simply calls comedi_pci_auto_unconfig
which has the same function prototype as the wrapper.
-> we can remove these wrappers and call comedi_pci_auto_unconfig
directly. This removes a lot some boilerplate code and saves some bytes.
Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Peter Huewe [Tue, 22 Jan 2013 23:03:30 +0000 (00:03 +0100)]
staging/comedi: Move comedi_pci_auto_unconfig to drivers.c
Since comedi_pci_auto_unconfig cannot be inlined anymore after
staging/comedi: Use comedi_pci_auto_unconfig directly for
pci_driver.remove
is applied, it makes sense to move it drivers.c
Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: addi_apci_3501: cleanup comments in hwdrv_apci3501.c
The watchdog/timer subdevice in this driver is basically broke. The
subdevice functions abuse the comedi API and the (*insn_config)
simply doesn't work due to it's treating data[0] as a parameter and
not as the config "instruction".
For now, cleanup the comments for the functions so they are at least
readable. Then we can figure out how to fix the subdevice.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: addi_apci_3501: change the MODULE_DESCRIPTION
Change the MODULE_DESCRIPTION to something more useful than the
generic "Comedi low-level driver" so that modinfo provides a
better description of the driver.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: comedi: addi_apci_3501: rewrite the analog output support
Currently the analog output subdevice has two support functions:
(*insn_config) - i_APCI3501_ConfigAnalogOutput()
(*insn_write) - i_APCI3501_WriteAnalogOutput()
The (*insn_config) function is used to configure the analog outputs
in either bipolar or unipolar mode. This function abuses the comedi
API since it treats the data[0] value as a parameter instead of as
the config "instruction".
The (*insn_write) function then writes a single value to the desired
analog output channel after doing some sanity checking on the channel
number. The sanity checking is not required since the comedi core has
already done it. Also, the (*insn_write) functions are supposed to
write all the data, indicated by insn->n, to the channel not just a
single value.
Rewrite the support code so it works properly with the comedi API.
The bipolar/unipolar configuration can be determine in the (*insn_write)
by checking the passed insn->chanspec.
Since the unipolar configuration only has 13-bit resolution, we need
to check that the data is in range because the subdevice 'maxdata' is
set to 14-bits for the bipolar mode. If the data is out of range,
output a dev_warn() and return -EINVAL.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Refactor the code that waits for the DAC to be ready into a helper
function.
A timeout of some sort should be added to this helper so code the
users to expect the error condition. In i_APCI3501_WriteAnalogOutput()
just return the error and don't actually write the new value to the
DAC. In apci3501_reset() output a dev_warn() that the DAC was not
ready.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Cc: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / mirror_ubuntu-hirsute-kernel.git / log