IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
Discussion of the series:
https://lore.kernel.org/all/20220405135758.774016-1-catalin.marinas@arm.com/
mm, arm64: Reduce ARCH_KMALLOC_MINALIGN brought to my attention that
our current IIO usage of L1CACHE_ALIGN is insufficient as their are Arm
platforms out their with non coherent DMA and larger cache lines at
at higher levels of their cache hierarchy.
Rename the define to make it's purpose more explicit. It will be used
much more widely going forwards (to replace incorrect ____cacheline_aligned
markings.
Note this patch will greatly reduce the padding on some architectures
that have smaller requirements for DMA safe buffers.
The history of changing values of ARCH_KMALLOC_MINALIGN via
ARCH_DMA_MINALIGN on arm64 is rather complex. I'm not tagging this
as fixing a particular patch from that route as it's not clear what to tag.
Most recently a change to bring them back inline was reverted because
of some Qualcomm Kryo cores with an L2 cache with 128-byte lines
sitting above the point of coherency.
c1132702c7 Revert "arm64: cache: Lower ARCH_DMA_MINALIGN to 64 (L1_CACHE_BYTES)"
That reverts:
65688d2a05 arm64: cache: Lower ARCH_DMA_MINALIGN to 64 (L1_CACHE_BYTES) which
refers to the change originally being motivated by Thunder x1 performance
rather than correctness.
Fixes: 6f7c8ee585 ("staging:iio: Add ability to allocate private data space to iio_allocate_device")
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Acked-by: Nuno Sá <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20220508175712.647246-2-jic23@kernel.org
This function was introduced with the ability to pick a clock.
There are no upstream users so presumably it isn't as obviously useful
as it seemed at the time. Hence drop it.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20220220163327.424696-1-jic23@kernel.org
Switch the IIO core to use firmware node handle instead of OF node.
This will allow to get label from firmware on non-OF systems.
Note, this doesn't change of_iio_*() APIs for now.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20220413180202.19220-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This entry should, under no situation, be modified by device
drivers. Now that we have limited its read access to device drivers
really needing it and did so through a dedicated helper, we can
easily move this variable to the opaque structure in order to
prevent any further modification from non-authorized code (out of the
core, basically).
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Link: https://lore.kernel.org/r/20220207143840.707510-12-miquel.raynal@bootlin.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
In order to later move this variable within the opaque structure, let's
create a helper for accessing it in read-only mode. This helper will be
exposed to device drivers and kept accessible for the few that could need
it. The write access to this variable however should be fully reserved to
the core so in a second step we will hide this variable into the opaque
structure.
Cc: Eugen Hristev <eugen.hristev@microchip.com>
Cc: Nicolas Ferre <nicolas.ferre@microchip.com>
Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
Cc: Ludovic Desroches <ludovic.desroches@microchip.com>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/r/20220207143840.707510-11-miquel.raynal@bootlin.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
As we are going to hide the currentmode inside the opaque structure,
this helper would soon need to call a non-inline function which would
simply drop the benefit of having the helper defined inline in a header.
One alternative is to move this helper in the core as there is no more
interest in defining it inline in a header. We will pay the minor cost
either way.
Let's do like the iio_device_id() helper which also refers to the opaque
structure and gets defined in the core.
Suggested-by: Jonathan Cameron <jic23@kernel.org>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/r/20220207143840.707510-10-miquel.raynal@bootlin.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
sysfs_emit() is preferred over raw s*printf() for sysfs attributes since it
knows about the sysfs buffer specifics and has some built-in checks for
size and alignment.
This patch converts the places in the IIO core that follow the pattern of
return s*printf(...)
to
return sysfs_emit(...)
This covers the new places that have been introduced where sprintf() is
used for formatting sysfs output since the last time this was done in
commit 83ca56b663 ("iio: core: Use sysfs_emit() (trivial bits)").
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20211216185217.1054495-2-lars@metafoo.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Drvdata is typically used by drivers to attach driver specific data to a
device. It is used to retrieve driver specific information when only the
device to which the data is attached is available.
In the IIO core in the `iio_device_alloc()` function we call
`iio_device_set_drvdata(indio_dev, indio_dev)`. This sets the drvdata of
the IIO device to itself.
This is rather unnecessary since if we have a pointer to the IIO device to
call `iio_device_get_drvdata()` on it we don't need to call the function
since we already have the pointer. If we only have a pointer to the `struct
device` we can use `dev_to_iio_dev()` to get the IIO device from it.
Furthermore the drvdata is supposed to be reserved for drivers, so it
should not be used by the IIO core in the first place.
The `set_drvdata()` has been around from the very beginning of the IIO
framework and back then it was used in the IIO device sysfs attribute
handling code. But that was subsequently replaced with a `dev_to_iio_dev()`
in commit e53f5ac52e ("iio: Use dev_to_iio_dev()") and other cleanups.
The self `set_drvdata()` is now no longer needed and can be removed.
Verified that there no longer any users by checking for potential users
using the following two coccinelle scripts and reviewing that none of the
matches are problematic code.
<smpl>
@@
struct iio_dev *iio_dev;
expression dev;
identifier fn !~ "(remove|resume|suspend)";
@@
fn(...)
{
...
*iio_dev = dev_get_drvdata(dev)
...
}
</smpl>
<smpl>
@r1@
position p;
struct iio_dev *indio_dev;
identifier dev_fn =~ "^dev_";
identifier devm_fn =~ "^devm_";
@@
(
dev_fn
|
devm_fn
)
(&indio_dev@p->dev, ...)
@@
struct iio_dev *indio_dev;
position p != r1.p;
@@
*&indio_dev@p->dev</smpl>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The iio_device_type struct is never modified, mark it as const. This allows
it to be placed in a read-only memory section, which will protect against
accidental or deliberate modification.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20211031080421.2086-1-lars@metafoo.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
As these are very late in the 5.15 cycle and non are particularly urgent,
they can wait for the merge window.
Key element in this set is Yang Yingliang has identified a number of
issues in error paths introduced recently when we added multiple
buffer support.
Other fixes:
* adi,ad5662
- Fix handling of i2c_master_send() return value.
* adi,ad5766
- Fix a wrong dt-property name that indicated wrong units and
did not mach the bindings.
- Associated 'fix' of the bindings example to have a possible scale.
* st,pressure-spi
- Add some missing entries to the spi_device_id table to ensure
auto-loading works.
* ti,tsc2046
- Fix a backwards comparison leading to a false dev_warn
-----BEGIN PGP SIGNATURE-----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=jYCH
-----END PGP SIGNATURE-----
Merge tag 'iio-fixes-for-5.16a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into char-misc-next
Jonathan writes:
First set of IIO fixes for the 5.16 cycle
As these are very late in the 5.15 cycle and non are particularly urgent,
they can wait for the merge window.
Key element in this set is Yang Yingliang has identified a number of
issues in error paths introduced recently when we added multiple
buffer support.
Other fixes:
* adi,ad5662
- Fix handling of i2c_master_send() return value.
* adi,ad5766
- Fix a wrong dt-property name that indicated wrong units and
did not mach the bindings.
- Associated 'fix' of the bindings example to have a possible scale.
* st,pressure-spi
- Add some missing entries to the spi_device_id table to ensure
auto-loading works.
* ti,tsc2046
- Fix a backwards comparison leading to a false dev_warn
* tag 'iio-fixes-for-5.16a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio:
iio: buffer: Fix memory leak in iio_buffers_alloc_sysfs_and_mask()
iio: adc: tsc2046: fix scan interval warning
iio: core: fix double free in iio_device_unregister_sysfs()
iio: core: check return value when calling dev_set_name()
iio: buffer: Fix memory leak in iio_buffer_register_legacy_sysfs_groups()
iio: buffer: Fix double-free in iio_buffers_alloc_sysfs_and_mask()
iio: buffer: Fix memory leak in __iio_buffer_alloc_sysfs_and_mask()
iio: buffer: check return value of kstrdup_const()
iio: dac: ad5446: Fix ad5622_write() return value
Documentation:devicetree:bindings:iio:dac: Fix val
drivers: iio: dac: ad5766: Fix dt property name
iio: st_pressure_spi: Add missing entries SPI to device ID table
Currently IIO only supports buffer mode for capture devices like ADCs. Add
support for buffered mode for output devices like DACs.
The output buffer implementation is analogous to the input buffer
implementation. Instead of using read() to get data from the buffer write()
is used to copy data into the buffer.
poll() with POLLOUT will wakeup if there is space available.
Drivers can remove data from a buffer using iio_pop_from_buffer(), the
function can e.g. called from a trigger handler to write the data to
hardware.
A buffer can only be either a output buffer or an input, but not both. So,
for a device that has an ADC and DAC path, this will mean 2 IIO buffers
(one for each direction).
The direction of the buffer is decided by the new direction field of the
iio_buffer struct and should be set after allocating and before registering
it.
Co-developed-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Co-developed-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Signed-off-by: Mihail Chindris <mihail.chindris@analog.com>
Link: https://lore.kernel.org/r/20211007080035.2531-2-mihail.chindris@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
I got the double free report:
BUG: KASAN: double-free or invalid-free in kfree+0xce/0x390
iio_device_unregister_sysfs+0x108/0x13b [industrialio]
iio_dev_release+0x9e/0x10e [industrialio]
device_release+0xa5/0x240
If __iio_device_register() fails, iio_dev_opaque->groups will be freed
in error path in iio_device_unregister_sysfs(), then iio_dev_release()
will call iio_device_unregister_sysfs() again, it causes double free.
Set iio_dev_opaque->groups to NULL when it's freed to fix this double free.
Not this is a local work around for a more general mess around life time
management that will get cleaned up and should make this handling
unnecesarry.
Fixes: 32f171724e ("iio: core: rework iio device group creation")
Reported-by: Hulk Robot <hulkci@huawei.com>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20211013030532.956133-1-yangyingliang@huawei.com
Cc: <Stable@vger.kernel.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The point of this new change is to make the IIO tree actually parsable.
Before, given this attribute as a filename:
in_voltage0_aux_sample_rate
Userspace had no way to know if the attribute name was
"aux_sample_rate" with no extended name, or "sample_rate" with "aux" as
the extended name, or just "rate" with "aux_sample" as the extended
name.
This was somewhat possible to deduce when there was more than one
attribute present for a given channel, e.g:
in_voltage0_aux_sample_rate
in_voltage0_aux_frequency
There, it was possible to deduce that "aux" was the extended name. But
even with more than one attribute, this wasn't very robust, as two
attributes starting with the same prefix (e.g. "sample_rate" and
"sample_size") would result in the first part of the prefix being
interpreted as being part of the extended name.
To address the issue, knowing that channels will never have both a label
and an extended name, set the channel's label to the extended name.
In this case, the label's attribute will also have the extended name in
its filename, but we can live with that - userspace can open
in_voltage0_<prefix>_label and verify that it returns <prefix> to obtain
the extended name.
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Link: https://lore.kernel.org/r/20210618123005.49867-3-paul@crapouillou.net
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Extended names are a problem for user-space as they make the filenames
in sysfs sometimes not parsable. They are now deprecated in favor of
labels.
This change makes sure that a device driver won't provide both labels
and extended names for its channels. It has never been the case and we
don't want it to happen.
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Link: https://lore.kernel.org/r/20210618123005.49867-2-paul@crapouillou.net
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
All of the users of iio_read_mount_matrix() are using the very same
property name. Moreover, the property name is hard coded in the API
documentation.
Make this clear and avoid duplication now and in the future.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Sean Nyekjaer <sean@geanix.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20210518112546.44592-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The devm_iio_device_alloc() function is supposed to return NULL and not
error pointers. Returning an error pointer will lead to a crash in the
callers.
Fixes: cf5724e915 ("iio: core: simplify some devm functions")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Link: https://lore.kernel.org/r/YJ+a1yaMu2QNATgt@mwanda
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
There is already an acessor function used to access it, making this
move straight forward.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Link: https://lore.kernel.org/r/20210426174911.397061-10-jic23@kernel.org
No reason for this to be exposed to the drivers, so lets move it to the
opaque structure.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Link: https://lore.kernel.org/r/20210426174911.397061-8-jic23@kernel.org
This lock is only of interest to the IIO core, so make it only
visible there.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Link: https://lore.kernel.org/r/20210426174911.397061-7-jic23@kernel.org
Continuing move to hide internal elements from drivers, move this structure
element over. It's only accessed from iio core files so this one was
straight forward and no accessor functions are needed.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Link: https://lore.kernel.org/r/20210426174911.397061-4-jic23@kernel.org
Continuing from Alexandru Ardelean's introduction of the split between
driver modifiable fields and those that should only be set by the core.
This could have been done in two steps to make the actual move after
introducing iio_device_id() but there seemed limited point to that
given how mechanical the majority of the patch is.
Includes fixup from Alex for missing mxs-lradc-adc conversion.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Link: https://lore.kernel.org/r/20210426174911.397061-2-jic23@kernel.org
Use devm_add_action_or_reset() instead of devres_alloc() and
devres_add(), which works the same. This will simplify the
code. There is no functional changes.
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Reviewed-by: Lars-Peter Clausen <lars@metafoo.de>
Reviewed-by: Nuno Sa <nuno.sa@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/1617881896-3164-6-git-send-email-yangyicong@hisilicon.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
When the ioctl() mechanism was introduced in IIO core to centralize the
registration of all ioctls in one place via commit 8dedcc3eee ("iio:
core: centralize ioctl() calls to the main chardev"), the return code was
changed from ENODEV to EINVAL, when the ioctl code isn't known.
This was done by accident.
This change reverts back to the old behavior, where if the ioctl() code
isn't known, ENODEV is returned (vs EINVAL).
This was brought into perspective by this patch:
https://lore.kernel.org/linux-iio/20210428150815.136150-1-paul@crapouillou.net/
Fixes: 8dedcc3eee ("iio: core: centralize ioctl() calls to the main chardev")
Signed-off-by: Alexandru Ardelean <aardelean@deviqon.com>
Reviewed-by: Nuno Sá <nuno.sa@analog.com>
Tested-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Currently ioctl handlers are removed twice. For the first time during
iio_device_unregister() then later on inside
iio_device_unregister_eventset() and iio_buffers_free_sysfs_and_mask().
Double free leads to kernel panic.
Fix this by not touching ioctl handlers list directly but rather
letting code responsible for registration call the matching cleanup
routine itself.
Fixes: 8dedcc3eee ("iio: core: centralize ioctl() calls to the main chardev")
Signed-off-by: Tomasz Duszynski <tomasz.duszynski@octakon.com>
Acked-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Cc: <Stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20210423080244.2790-1-tomasz.duszynski@octakon.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
sysfs_emit() is preferred over raw s*printf() for sysfs attributes since it
knows about the sysfs buffer specifics and has some built-in sanity checks.
Convert __iio_format_value() and related functions to use this new
interface.
This conversion involves changing the signature of __iio_format_value() so
that it similar to sysfs_emit_at() and takes the buffers start address and
an offset where to write within the buffer.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20210320071405.9347-4-lars@metafoo.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
sysfs_emit() is preferred over raw s*printf() for sysfs attributes since it
knows about the sysfs buffer specifics and has some built-in sanity checks.
Convert the iio_enum_available_read() function to use sysfs_emit_at()
instead of scnprintf().
The conversion is straight forward, the only difference is that
sysfs_emit_at() takes the buffers start address and an offset as parameters
and already knows about the buffer's size limit.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20210320071405.9347-3-lars@metafoo.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
sysfs_emit() is preferred over raw s*printf() for sysfs attributes since it
knows about the sysfs buffer specifics and has some built-in sanity checks.
This patch converts the places in the iio core that follow the pattern of
return s*printf(...)
to
return sysfs_emit(...)
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20210320071405.9347-2-lars@metafoo.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Thanks to Lars for finding this.
The free of the 'attached_buffers' array should be done as late as
possible. This change moves it to iio_buffers_put(), which looks like
the best place for it, since it takes place right before the IIO device
data is free'd.
The free of this array will be handled by calling iio_device_free().
The iio_buffers_put() function is renamed to iio_device_detach_buffers()
since the role of this function changes a bit.
It looks like this issue was ocurring on the error path of
iio_buffers_alloc_sysfs_and_mask() and in
iio_buffers_free_sysfs_and_mask()
Added a comment in the doc-header of iio_device_attach_buffer() to
mention how this will be free'd in case anyone is reading the code
and becoming confused about it.
Fixes: ee708e6baa ("iio: buffer: introduce support for attaching more IIO buffers")
Reported-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Link: https://lore.kernel.org/r/20210307185444.32924-1-ardeleanalex@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Some hid sensors may use relative sensitivity such as als sensor.
This patch adds relative sensitivity checking for all hid sensors.
Signed-off-by: Ye Xiang <xiang.ye@intel.com>
Acked-by: Jiri Kosina <jkosina@suse.cz>
Link: https://lore.kernel.org/r/20210207070048.23935-2-xiang.ye@intel.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The 'dev' variable name usually refers to 'struct device' types. However in
iio_device_alloc() this was used for the 'struct iio_dev' type, which was
sometimes causing minor confusions.
This change renames the variable to 'indio_dev', which is the usual name
used around IIO for 'struct iio_dev' type objects.
It makes grepping a bit easier as well.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20210215104043.91251-22-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
With this change, calling iio_device_attach_buffer() will actually attach
more buffers.
Right now this doesn't do any validation of whether a buffer is attached
twice; maybe that can be added later (if needed). Attaching a buffer more
than once should yield noticeably bad results.
The first buffer is the legacy buffer, so a reference is kept to it.
At this point, accessing the data for the extra buffers (that are added
after the first one) isn't possible yet.
The iio_device_attach_buffer() is also changed to return an error code,
which for now is -ENOMEM if the array could not be realloc-ed for more
buffers.
To adapt to this new change iio_device_attach_buffer() is called last in
all place where it's called. The realloc failure is a bit difficult to
handle during un-managed calls when unwinding, so it's better to have this
as the last error in the setup_buffer calls.
At this point, no driver should call iio_device_attach_buffer() directly,
it should call one of the {devm_}iio_triggered_buffer_setup() or
devm_iio_kfifo_buffer_setup() or devm_iio_dmaengine_buffer_setup()
functions. This makes iio_device_attach_buffer() a bit easier to handle.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20210215104043.91251-20-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
In order to keep backwards compatibility with the current chardev
mechanism, and in order to add support for multiple buffers per IIO device,
we need to pass both the IIO device & IIO buffer to the chardev.
This is particularly needed for the iio_buffer_read_outer() function, where
we need to pass another buffer object than 'indio_dev->buffer'.
Since we'll also open some chardevs via anon inodes, we can pass extra
buffers in that function by assigning another object to the
iio_dev_buffer_pair object.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20210215104043.91251-17-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This change wraps all buffer attributes into iio_dev_attr objects, and
assigns a reference to the IIO buffer they belong to.
With the addition of multiple IIO buffers per one IIO device, we need a way
to know which IIO buffer is being enabled/disabled/controlled.
We know that all buffer attributes are device_attributes. So we can wrap
them with a iio_dev_attr types. In the iio_dev_attr type, we can also hold
a reference to an IIO buffer.
So, we end up being able to allocate wrapped attributes for all buffer
attributes (even the one from other drivers).
The neat part with this mechanism, is that we don't need to add any extra
cleanup, because these attributes are being added to a dynamic list that
will get cleaned up via iio_free_chan_devattr_list().
With this change, the 'buffer->scan_el_dev_attr_list' list is being renamed
to 'buffer->buffer_attr_list', effectively merging (or finalizing the
merge) of the buffer/ & scan_elements/ attributes internally.
Accessing these new buffer attributes can now be done via
'to_iio_dev_attr(attr)->buffer' inside the show/store handlers.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20210215104043.91251-15-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This change adds a reference to a 'struct iio_buffer' object on the
iio_dev_attr object. This way, we can use the created iio_dev_attr objects
on per-buffer basis (since they're allocated anyway).
A minor downside of this change is that the number of parameters on
__iio_add_chan_devattr() grows by 1. This looks like it could do with a bit
of a re-think.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20210215104043.91251-14-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Up until now, the device groups that an IIO device had were limited to 6.
Two of these groups would account for buffer attributes (the buffer/ and
scan_elements/ directories).
Since we want to add multiple buffers per IIO device, this number may not
be enough, when adding a second buffer. So, this change reallocates the
groups array whenever an IIO device group is added, via a
iio_device_register_sysfs_group() helper.
This also means that the groups array should be assigned to
'indio_dev.dev.groups' really late, right before {cdev_}device_add() is
called to do the entire setup.
And we also must take care to free this array when the sysfs resources are
being cleaned up.
With this change we can also move the 'groups' & 'groupcounter' fields to
the iio_dev_opaque object. Up until now, this didn't make a whole lot of
sense (especially since we weren't sure how multibuffer support would look
like in the end).
But doing it now kills one birds with one stone.
An alternative, would be to add a configurable Kconfig symbol
CONFIG_IIO_MAX_BUFFERS_PER_DEVICE (or something like that) and compute a
static maximum of the groups we can support per IIO device. But that would
probably annoy a few people since that would make the system less
configurable.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20210215104043.91251-11-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
We only need a chardev if we need to support buffers and/or events.
With this change, a chardev will be created only if an IIO buffer is
attached OR an event_interface is configured.
Otherwise, no chardev will be created, and the IIO device will get
registered with the 'device_add()' call.
Quite a lot of IIO devices don't really need a chardev, so this is a minor
improvement to the IIO core, as the IIO device will take up (slightly)
fewer resources.
In order to not create a chardev, we mostly just need to not initialize the
indio_dev->dev.devt field. If that is un-initialized, cdev_device_add()
behaves like device_add().
This change has a small chance of breaking some userspace ABI, because it
removes un-needed chardevs. While these chardevs (that are being removed)
have always been unusable, it is likely that some scripts may check their
existence (for whatever logic).
And we also hope that before opening these chardevs, userspace would have
already checked for some pre-conditions to make sure that opening these
chardevs makes sense.
For the most part, there is also the hope that it would be easier to change
userspace code than revert this. But in the case that reverting this is
required, it should be easy enough to do it.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20210215104043.91251-9-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
When formatting a value using IIO_VAL_FRACTIONAL_LOG2 and the values is
between -1 and 0 the sign is omitted.
We need the same trick as for IIO_VAL_FRACTIONAL to make sure this gets
formatted correctly.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20201215191743.2725-2-lars@metafoo.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
IIO_VAL_FRACTIONAL_LOG2 works with signed values, yet the temporary we use
is unsigned. This works at the moment because the variable is implicitly
cast to signed everywhere where it is used.
But it will certainly be cleaner to use a signed variable in the first
place.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20201215191743.2725-1-lars@metafoo.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Only set indio_dev->label from of/dt if there actually is a label
specified in of.
This allows drivers to set a label without this being overwritten with
NULL when there is no label specified in of. This is esp. useful on
devices where of is not used at all, such as your typical x86/ACPI device.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20210207160901.110643-2-hdegoede@redhat.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Some enums might have gaps or reserved values in the middle of their value
range. E.g. consider a 2-bit enum where the values 0, 1 and 3 have a
meaning, but 2 is a reserved value and can not be used.
Add support for such enums to the IIO enum helper functions. A reserved
values is marked by setting its entry in the items array to NULL rather
than the normal descriptive string value.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20210107112049.10815-1-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The iio-core extends the attr_group provided by the driver with its
own attributes. To be able to do this it:
1. Has its own (non const) io_dev_opaque.chan_attr_group attr_group struct
2. It allocates a new attrs array with room for both the drivers and its
own attributes
3. It copies over the driver provided attributes into the newly allocated
attrs array.
But the drivers attr_group may contain more then just the attrs array, it
may also contain an is_visible callback and at least the adi-axi-adc.c
is currently defining such a callback.
Change the attr_group copying code to also copy over the is_visible
callback, so that drivers can define one and have it workins as is
normal for attr_group-s all over the kernel.
Note that the is_visible callback takes an index into the array as
argument, so that indices of the driver's attributes must not change,
this is not a problem as the driver's own attributes are added first
to the newly allocated attrs array and the attributes handled by the
core are appended after the driver's attributes.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Cc: Michael Hennerich <michael.hennerich@analog.com>
Link: https://lore.kernel.org/r/20201125084606.11404-2-hdegoede@redhat.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
iio_format_list() has two branches in a switch statement that are almost
identical. They only differ in the stride that is used to iterate through
the item list.
Consolidate this into a common code path to simplify the code.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20201114120000.6533-2-lars@metafoo.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The iio_format_avail_list() and iio_format_avail_range() functions are
almost identical. The only differences are that iio_format_avail_range()
expects a fixed amount of items and adds brackets "[ ]" around the output.
Refactor them into a common helper function. This improves the
maintainability of the code as it makes it easier to modify the
implementation of these functions.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20201114120000.6533-1-lars@metafoo.de
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
It seems that when this was tested the happy case was more tested. A few of
the userspace apps rely on this returning negative error codes in case an
ioctl() is not available.
When running multiple ioctl() handlers or when calling an ioctl() that
doesn't exist, IIO_IOCTL_UNHANDLED is returned. In that case -EINVAL should
be returned.
Fixes: 8dedcc3eee ("iio: core: centralize ioctl() calls to the main chardev")
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20201117095154.7189-1-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
