- 21 Sep, 2020 40 commits
-
-
Jonathan Cameron authored
This case is a bit different to the rest of the series. The driver was doing a regmap_bulk_read into a buffer that wasn't dma safe as it was on the stack with no guarantee of it being in a cacheline on it's own. Fixing that also dealt with the data leak and alignment issues that Lars-Peter pointed out. Also removed some unaligned handling as we are now aligned. Fixes tag is for the dma safe buffer issue. Potentially we would need to backport timestamp alignment futher but that is a totally different patch. Fixes: fd64df16 ("iio: imu: inv_mpu6050: Add SPI support for MPU6000") Reported-by:
Lars-Peter Clausen <lars@metafoo.de> Signed-off-by:
Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by:
Andy Shevchenko <andy.shevchenko@gmail.com> Reviewed-by:
Jean-Baptiste Maneyrol <jmaneyrol@invensense.com> Cc: <Stable@vger.kernel.org> Link: https://lore.kernel.org/r/20200722155103.979802-18-jic23@kernel.org
-
Jonathan Cameron authored
One of a class of bugs pointed out by Lars in a recent review. iio_push_to_buffers_with_timestamp assumes the buffer used is aligned to the size of the timestamp (8 bytes). This is not guaranteed in this driver which uses an array of smaller elements on the stack. We move to a suitable structure in the iio_priv() data with alignment explicitly requested. This data is allocated with kzalloc so no data can leak apart from previous readings. Note that previously no leak at all could occur, but previous readings should never be a problem. In this case the timestamp location depends on what other channels are enabled. As such we can't use a structure without misleading by suggesting only one possible timestamp location. Fixes: 50a6edb1 ("iio: adc: add ADC12130/ADC12132/ADC12138 ADC driver") Reported-by:
-
Jonathan Cameron authored
One of a class of bugs pointed out by Lars in a recent review. iio_push_to_buffers_with_timestamp assumes the buffer used is aligned to the size of the timestamp (8 bytes). This is not guaranteed in this driver which uses an array of smaller elements on the stack. We fix this issues by moving to a suitable structure in the iio_priv() data with alignment explicitly requested. This data is allocated with kzalloc so no data can leak apart from previous readings. Note that previously no data could leak 'including' previous readings but I don't think it is an issue to potentially leak them like this now does. In this case the postioning of the timestamp is depends on what other channels are enabled. As such we cannot use a structure to make the alignment explicit as it would be missleading by suggesting only one possible location for the timestamp. Fixes: 815bbc87 ("iio: ti-adc0832: add triggered buffer support") Reported-by:
-
Jonathan Cameron authored
One of a class of bugs pointed out by Lars in a recent review. iio_push_to_buffers_with_timestamp assumes the buffer used is aligned to the size of the timestamp (8 bytes). This is not guaranteed in this driver which uses an array of smaller elements on the stack. As Lars also noted this anti pattern can involve a leak of data to userspace and that indeed can happen here. We close both issues by moving to an array of suitable structures in the iio_priv() data. This data is allocated with kzalloc so no data can leak apart from previous readings. For the tagged path the data is aligned by using __aligned(8) for the buffer on the stack. There has been a lot of churn in this driver, so likely backports may be needed for stable. Fixes: 290a6ce1 ("iio: imu: add support to lsm6dsx driver") Reported-by:
-
Jonathan Cameron authored
One of a class of bugs pointed out by Lars in a recent review. iio_push_to_buffers_with_timestamp assumes the buffer used is aligned to the size of the timestamp (8 bytes). This is not guaranteed in this driver which uses a 24 byte array of smaller elements on the stack. As Lars also noted this anti pattern can involve a leak of data to userspace and that indeed can happen here. We close both issues by moving to a suitable array in the iio_priv() data with alignment explicitly requested. This data is allocated with kzalloc so no data can leak appart from previous readings. Depending on the enabled channels, the location of the timestamp can be at various aligned offsets through the buffer. As such we any use of a structure to enforce this alignment would incorrectly suggest a single location for the timestamp. Comments adjusted to express this clearly in the code. Fixes: ac45e57f ("iio: light: Add driver for Silabs si1132, si1141/2/3 and si1145/6/7 ambient light, uv index and proximity sensors") Reported-by:
-
Jonathan Cameron authored
One of a class of bugs pointed out by Lars in a recent review. iio_push_to_buffers_with_timestamp assumes the buffer used is aligned to the size of the timestamp (8 bytes). This is not guaranteed in this driver which uses a 16 byte array of smaller elements on the stack. This is fixed by using an explicit c structure. As there are no holes in the structure, there is no possiblity of data leakage in this case. The explicit alignment of ts is not strictly necessary but potentially makes the code slightly less fragile. It also removes the possibility of this being cut and paste into another driver where the alignment isn't already true. Fixes: 36e0371e ("iio:itg3200: Use iio_push_to_buffers_with_timestamp()") Reported-by:
-
Tom Rix authored
Potential error return is not checked. This can lead to use of undefined data. Detected by clang static analysis. st_lsm6dsx_shub.c:540:8: warning: Assigned value is garbage or undefined *val = (s16)le16_to_cpu(*((__le16 *)data)); ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Fixes: c91c1c84 ("iio: imu: st_lsm6dsx: add i2c embedded controller support") Signed-off-by: Tom Rix <trix@redhat.com Cc: <Stable@vger.kernel.org> Reviewed-by: -
Sergiu Cuciurean authored
As part of the general cleanup of indio_dev->mlock, this change replaces it with a local lock, to protect potential concurrent access to the completion callback during a conversion. This is part of a bigger cleanup. Link: https://lore.kernel.org/linux-iio/CA+U=Dsoo6YABe5ODLp+eFNPGFDjk5ZeQEceGkqjxXcVEhLWubw@mail.gmail.com/Signed-off-by:
Sergiu Cuciurean <sergiu.cuciurean@analog.com> Signed-off-by:
Alexandru Ardelean <alexandru.ardelean@analog.com> Reviewed-by:
Krzysztof Kozlowski <krzk@kernel.org> Link: https://lore.kernel.org/r/20200916093123.78954-1-alexandru.ardelean@analog.comSigned-off-by:
-
Jonathan Cameron authored
Very simple binding. I've changed the example to use the node name threshold-detector@0 as sensor@0 seemed too generic. Signed-off-by:
Rob Herring <robh@kernel.org> Cc: Vladimir Barinov <vladimir.barinov@cogentembedded.com> Link: https://lore.kernel.org/r/20200909175946.395313-18-jic23@kernel.org
-
Jonathan Cameron authored
Fairly straight conversion. The one oddity in the original binding is that spi-cpha and spi-cpol were not marked as required, but were in the example. Looking at the datasheet, there isn't any documented flexibility in the possible SPI modes, so I have moved these to requires. For spi-max-frequency I have gone the other way. I absolutely agree that it is good to specify this in the dt-binding, but it's not strictly required. As Stefan's email is bouncing I have gone with Michael as maintainer of this one as it falls under the ADI catch all entry in MAINTAINERS. Signed-off-by:
-
Jonathan Cameron authored
Simple SPI driver. I've added the #io-channel-cells as an optional parameter to allow use of this device as a provider of ADC capabilities to other devices. Signed-off-by:
-
Jonathan Cameron authored
This is a small part of an MFD so perhaps ultimately it makes more sense to document it with that MFD binding rather than separately. In the meantime it's a straightforward conversion from txt to yaml. Signed-off-by:
Adam Thomson <Adam.Thomson.Opensource@diasemi.com> Link: https://lore.kernel.org/r/20200909175946.395313-13-jic23@kernel.org
-
Jonathan Cameron authored
Conversion from txt to yaml. Slightly expanded example to give a bit more context. Description lifted from the original driver commit. Signed-off-by:
-
Jonathan Cameron authored
Renamed to remove the wild cards. These go wrong far too often so in general preferred to use the name of a specific part. As this binding only provides one compatible, I went with that. Signed-off-by:
-
Jonathan Cameron authored
Most of the description in the original doc is effectively boilerplate and does not bring much value so I have not carried it over into the yaml. Added #io-channel-cells to simplify use of channels on this ADC by consumer drivers. Signed-off-by:
-
Jonathan Cameron authored
This binding has a few corners that would have been done different today but hopefully the yaml schema captures the constraints correctly. The child node names are not constrained hence the fairly open regexp. I've also documented the defaults for the two references that the driver seems to use and copied the value descriptions from the header because I think they should be in the dt-binding itself. This is part of a general effort to convert all the IIO bindings over to yaml Unfortunately I don't have a current address for Markus, so have put myself as the maintainer for this binding until someone else steps up! Signed-off-by:
-
Jonathan Cameron authored
A nice simple binding. Only real different from txt is that I dropped some descriptions where the naming of the parameter was self explanatory Signed-off-by:
-
Jonathan Cameron authored
This binding was moved over from hwmon some time ago so is a bit unusual in terms of IIO bindings. However, conversion was fairly straight forwards. I've listed both Dirk (who think wrote original binding) and Daniel who added the IIO driver for this device. Signed-off-by:
-
Jonathan Cameron authored
I don't really know much about this one, hence the binding is a simple conversion of what was in the txt file. Note that I have taken on maintenance of this binding as I don't have a recent address for Phani Movva. Signed-off-by:
-
Jonathan Cameron authored
Renamed the file to reflect the only compatible. Added #io-channel-cells to make it easier to support consumers of the ADC channels this device provides. Signed-off-by:
-
Jonathan Cameron authored
A simple conversion from txt file to yaml. I added the #io-channel-cells property as optional to allow the channels of this ADCs to be used to provide services to other drivers, for example if an analog accelerometer is connected. Signed-off-by:
-
Jonathan Cameron authored
I changed the name to reflect a specific part in line with normal naming conventions. If there is a particularly strong reason to keep the wild cards let me know. Otherwise this was a fairly simple conversion as part of converting all the IIO bindings to yaml. Signed-off-by:
Orson Zhai <orsonzhai@gmail.com> Cc: Baolin Wang <baolin.wang7@gmail.com> Cc: Chunyan Zhang <zhang.lyra@gmail.com> Link: https://lore.kernel.org/r/20200909175946.395313-7-jic23@kernel.org
-
Jonathan Cameron authored
Part of a general move of IIO bindings over to yaml. Signed-off-by:
-
Jonathan Cameron authored
Simple txt to yaml conversion. Part of a general move to convert all the IIO bindings over to yaml. Signed-off-by:
Joel Stanley <joel@jms.id.au> Cc: Rick Altherr <raltherr@google.com> Cc: Joel Stanley <joel@jms.id.au> Cc: Andrew Jeffery <andrew@aj.id.au> Link: https://lore.kernel.org/r/20200909175946.395313-5-jic23@kernel.org
-
Jonathan Cameron authored
Conversion from freeform text to yaml. One oddity in this binding is that, for historical reasons it requires the node name to be stmpe_adc. I've put that in the decription field but I'm not sure if there is a better way to specify this? Signed-off-by:
-
Jonathan Cameron authored
Conversion from txt to yaml as part of a general move of IIO bindings to the new format. Signed-off-by:
Sebastian Reichel <sre@kernel.org> Link: https://lore.kernel.org/r/20200909175946.395313-3-jic23@kernel.org
-
Jonathan Cameron authored
A simple conversion of this freescale ADC binding from txt to yaml. For maintainer I went with Fugang Duan as the original author of the binding. Would be great to have confirmation of this. Signed-off-by:
Fugang Duan <fugang.duan@nxp.com> Reviewed-by:
-
Mario Tesi authored
Scaling factor values for Acc lead to an unacceptable rounding of the full scale (FS) calculated by some SensorHAL on Android devices. For examples setting FS to 4g the in_accel_x_scale, in_accel_y_scale and in_accel_z_scale are 0.001196 on 6 decimal digits and the FS is 0.001196 × ((2^15) − 1) ~= 39.1893 m/s^2. Android CTS R10 SensorParameterRangeTest test expects a value greater than 39.20 m/s^2 so this test fails (ACCELEROMETER_MAX_RANGE = 4 * 9.80). Using 9 decimal digits the new scale factor is 0.001196411 and the FS now is 0.001196411 × ((2^15)−1) ~= 39.2028 m/s^2. This patch extends to IIO_VAL_INT_PLUS_NANO type the scaling factor to all IMU devices where SensorParameterRangeTest CTS test fails. Signed-off-by:
Mario Tesi <mario.tesi@st.com> Acked-by:
Lorenzo Bianconi <lorenzo@kernel.org> Signed-off-by:
-
Nuno Sá authored
As there are no users anymore of this structure, it can be safely removed. Signed-off-by:
Nuno Sá <nuno.sa@analog.com> Signed-off-by:
-
Nuno Sá authored
Burst mode variables are now part of the `adis_data` struct. The driver also has now to explicitly define the length of the burst buffer. Signed-off-by:
-
Nuno Sá authored
Burst mode variables are now part of the `adis_data` struct. The driver also has now to explicitly define the length of the burst buffer. Signed-off-by:
-
Nuno Sá authored
Add burst mode variables in the per device specific data structure. As some drivers support multiple devices with different burst sizes it makes sense this data to be in `adis_data`. While moving the variables, there are two main differences: 1. The `en`variable is dropped. If a device supports burst mode, it will just use it as it will has better performance for almost all real use cases. 2. Replace `extra_len` by `burst_len`. Users should now explicitly define the length of the burst buffer as it is typically constant. This also allows to remove the following line from the library: ``` /* All but the timestamp channel */ burst_length = (indio_dev->num_channels - 1) * sizeof(u16); ``` The library should not assume that a timestamp channel is defined. Moreover, most parts also include some diagnostic data, crc, etc.. in the burst buffer that needed to be included in an `extra_len` variable which is not that nice. On top of this, some devices already start to have some 32bit size channels ... This patch is also a move to completely drop the `struct adis_burst` from the library. Signed-off-by:
-
Jonathan Cameron authored
Given a value of 1 corresponds to IIO_SHARE_BY_TYPE I have replaced it with that. Should cause no functional change. Fixes: fdadbce0 ("iio: add Bosch BMA180 acceleration sensor driver") Signed-off-by:
Matt Ranostay <matt.ranostay@konsulko.com> Reviewed-by:
-
Jonathan Cameron authored
So it's obvious that the code is wrong in passing true, but I'm assuming that will actually evaluate to 1 and hence IIO_SHARED_BY_TYPE. The documentation however has this attribute as IIO_SHARED_BY_ALL. My current assumption is the documentation is wrong. If anyone knows otherwise please shout out! Fixes: 7247645f ("iio: hmc5843: Move hmc5843 out of staging") Signed-off-by:
H. Nikolaus Schaller <hns@goldelico.com> Link: https://lore.kernel.org/r/20200913112546.715624-1-jic23@kernel.org
-
Jonathan Cameron authored
A partial set of these was added to IIO a long time back. This fills in some gaps in coverage highlighted by building with W=1 Signed-off-by:
-
Jonathan Cameron authored
One comment isn't kernel-doc at all, but starts with /** and another is simply missing a parameter that was introduced recently. Signed-off-by:
-
Alexandru Ardelean authored
Since there was a recently discovered issue with these locks, it probably makes sense to cleanup the code a bit, to prevent it from being used as an example/reference. This change moves the lock only where it is explicitly needed to protect resources from potential concurrent accesses. It also reworks the switch statements to do direct returns vs caching the return value on a variable. Signed-off-by:
-
Alexandru Ardelean authored
The next rework may require an unindentation of a code block in ad5592r_read_raw(), which would make review a bit more difficult. This change unindents the code block for reading the scale of the non-temperature channels. Signed-off-by:
-
Jonathan Cameron authored
struct iio_chan_spec_ext_info shared element is of type enum iio_shared_by, not boolean. It's like the enum value will for IIO_SHARED_BY_TYPE == 1 == true, hence no actual problem has been observed. CC [M] drivers/iio/dac/ad5592r-base.o 491 | .shared = true, | Fixes: 56ca9db8 ("iio: dac: Add support for the AD5592R/AD5593R ADCs/DACs") Signed-off-by: -
Lee Jones authored
Fixes the following W=1 kernel build warning(s): drivers/iio/chemical/sgp30.c:236: warning: Function parameter or member 'duration_us' not described in 'sgp_read_cmd' Signed-off-by:
Lee Jones <lee.jones@linaro.org> Cc: Andreas Brauchli <a.brauchli@elementarea.net> Cc: Pascal Sachs <pascal.sachs@sensirion.com> Link: https://lore.kernel.org/r/20200716135928.1456727-6-lee.jones@linaro.orgSigned-off-by:
-
