Hans de Goede [Sun, 24 Dec 2023 18:34:02 +0000 (19:34 +0100)]
misc: lis3lv02d_i2c: Add missing setting of the reg_ctrl callback
The lis3lv02d_i2c driver was missing a line to set the lis3_dev's
reg_ctrl callback.
lis3_reg_ctrl(on) is called from the init callback, but due to
the missing reg_ctrl callback the regulators where never turned off
again leading to the following oops/backtrace when detaching the driver:
platform/surface: aggregator: make ssam_bus_type constant and static
Now that the driver core can properly handle constant struct bus_type,
move the ssam_bus_type variable to be a constant structure as well,
placing it into read-only memory which can not be modified at runtime.
It's also never used outside of
drivers/platform/surface/aggregator/bus.c so make it static and don't
export it as no one is using it.
Cc: Maximilian Luz <luzmaximilian@gmail.com> Cc: Hans de Goede <hdegoede@redhat.com> Cc: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Cc: <platform-driver-x86@vger.kernel.org> Reviewed-by: Maximilian Luz <luzmaximilian@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Link: https://lore.kernel.org/r/2023121957-tapered-upswing-8326@gregkh Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Heiner Kallweit [Wed, 20 Dec 2023 12:55:58 +0000 (13:55 +0100)]
eeprom: at24: Probe for DDR3 thermal sensor in the SPD case
The DDR3 SPD data structure advertises the presence of a thermal
sensor on a DDR3 module in byte 32, bit 7. Let's use this information
to explicitly instantiate the thermal sensor I2C client instead of
having to rely on class-based I2C probing.
The temp sensor i2c address can be derived from the SPD i2c address,
so we can directly instantiate the device and don't have to probe
for it. If the temp sensor has been instantiated already by other
means (e.g. class-based auto-detection), then the busy-check in
i2c_new_client_device will detect this.
Note: Thermal sensors on DDR4 DIMM's are instantiated from the
ee1004 driver.
Heiner Kallweit [Tue, 19 Dec 2023 22:09:48 +0000 (23:09 +0100)]
eeprom: ee1004: add support for temperature sensor
The EE1004 SPD data structure advertises the presence of a thermal
sensor on a DDR4 module in byte 14, bit 7. Let's use this information
to explicitly instantiate the thermal sensor I2C client instead of
having to rely on class-based I2C probing.
The temp sensor i2c address can be derived from the SPD i2c address,
so we can directly instantiate the device and don't have to probe
for it. If the temp sensor has been instantiated already by other
means (e.g. class-based auto-detection), then the busy-check in
i2c_new_client_device will detect this.
Patch was successfully tested with a Corsair Vengeance RGB PRO
DDR4 module which comes with a thermal sensor.
Now that the driver core can properly handle constant struct bus_type,
move the moxtet_bus_type to be a constant structure as well, placing it
into read-only memory which can not be modified at runtime.
For some reason, moxtet_type was defined in moxtet.h, but never actually
used. Looks like a left-over from the original commit that was
exporting the moxtet bus type, but that wasn't needed, and it was a
different variable name, so no one noticed this one dangling around.
Dan Carpenter [Tue, 2 Jan 2024 13:10:21 +0000 (16:10 +0300)]
cdx: call of_node_put() on error path
Add a missing call to of_node_put(np) on error.
There was a second error path where "np" was NULL, but that situation is
impossible. The for_each_compatible_node() loop iterator is always
non-NULL. Just deleted that error path.
Fixes: 54b406e10f03 ("cdx: Remove cdx controller list from cdx bus system") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Acked-by: Abhijit Gangurde <abhijit.gangurde@amd.com> Link: https://lore.kernel.org/r/2e66efc4-a13a-4774-8c9d-763455fe4834@moroto.mountain Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Abhijit Gangurde [Fri, 22 Dec 2023 06:46:27 +0000 (12:16 +0530)]
cdx: Create resource debugfs file for cdx device
resource debugfs file contains host addresses of CDX device resources.
Each line of the resource file describe type of resource, a region
with start-end and flag fields.
Rafał Miłecki [Thu, 21 Dec 2023 17:34:20 +0000 (18:34 +0100)]
nvmem: u-boot-env: improve coding style
1. Prefer kzalloc() over kcalloc()
See memory-allocation.rst which says: "to be on the safe side it's
best to use routines that set memory to zero, like kzalloc()"
2. Drop dev_err() for u_boot_env_add_cells() fail
It can fail only on -ENOMEM. We don't want to print error then.
3. Add extra "crc32_addr" variable
It makes code reading header's crc32 easier to understand / review.
Rafał Miłecki [Thu, 21 Dec 2023 17:34:17 +0000 (18:34 +0100)]
nvmem: core: add nvmem_dev_size() helper
This is required by layouts that need to read whole NVMEM content. It's
especially useful for NVMEM devices without hardcoded layout (like
U-Boot environment data block).
Simply pass whole "struct nvmem_layout" instead of single variables.
There is nothing in "struct nvmem_layout" that we have to hide from
layout drivers. They also access it during .probe() and .remove().
Thanks to this change:
1. API gets more consistent
All layouts drivers callbacks get the same argument
2. Layouts get correct device
Before this change NVMEM core code was passing NVMEM device instead
of layout device. That resulted in:
* Confusing prints
* Calling devm_*() helpers on wrong device
* Helpers like of_device_get_match_data() dereferencing NULLs
3. It gets possible to get match data
First of all nvmem_layout_get_match_data() requires passing "struct
nvmem_layout" which .add_cells() callback didn't have before this. It
doesn't matter much as it's rather useless now anyway (and will be
dropped).
What's more important however is that of_device_get_match_data() can
be used now thanks to owning a proper device pointer.
asm-generic/errno-base.h can be replaced by linux/errno.h and the file
will still build correctly. It is an asm-generic file which should be
avoided if possible.
Now that the driver core can properly handle constant struct bus_type,
move the locomo_bus_type variable to be a constant structure as well,
placing it into read-only memory which can not be modified at runtime.
It's also never used outside of arch/arm/common/locomo.c so make it
static and don't export it as no one is using it.
dma-debug: make dma_debug_add_bus take a const pointer
The driver core now can handle a const struct bus_type pointer, and the
dma_debug_add_bus() call just passes on the pointer give to it to the
driver core, so make this pointer const as well to allow everyone to use
read-only struct bus_type pointers going forward.
Cc: Christoph Hellwig <hch@lst.de> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Robin Murphy <robin.murphy@arm.com> Cc: <iommu@lists.linux.dev> Reviewed-by: Robin Murphy <robin.murphy@arm.com> Link: https://lore.kernel.org/r/2023121941-dejected-nugget-681e@gregkh Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is no need to export maple_bus_type as no one uses it outside of
maple.c, so make it static, AND make it const as it can be read-only as
no one modifies it.
Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Cc: Rich Felker <dalias@libc.org> Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> Cc: <linux-sh@vger.kernel.org> Link: https://lore.kernel.org/r/2023121918-rejoicing-frostlike-d976@gregkh Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Merge tag 'iio-for-6.8b' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into char-misc-next
Jonathan writes:
IIO: 2nd set of new device support, features and cleanup for 6.8
A late/optimistic second pull request. The bots have been poking them
since Wednesday without any issues. There are a few fixes in the
ad7091r5 driver as part of rework to enable the ad7091r8 parts that
are included at start of that series.
Includes pre-work for major changes to the DMA buffers that should
land in 6.9 and will greatly improve performance and flexibility for
high performance devices by enabling DMABUF based zero copy transfers
when we don't need to bounce the data via user space.
New device support
------------------
adi,ad7091r8
- Major refactor of existing adi,ad7091r5 driver to separate out useful
shared library code that can be used by I2C and SPI parts.
- Use that library from a new driver supporting the AD7091R-2, AD7091R-4
and AD7091R-8 12-Bit SPI ADCs.
- Series includes some late breaking fixes for the ad7091r5.
microchip,mcp4821
- New driver for MCP4801, MCP4802, MCP4811, MCP4812, MCP4821 and MCP4822
I2C single / dual channel DACs.
Cleanup
-------
buffers:
- Use IIO_SEPARATE in place of some hard-coded 0 values.
dma-buffers:
- Simplify things to not use an outgoing queue given it only ever has
up to two elements and we only need to track which is first.
- Split the iio_dma_buffer_fileio_free() function up to make it easier
to read and enable reuse in a series lining up for 6.9
iio.h
- Drop some stale documentation of struct fields that don't exist.
* tag 'iio-for-6.8b' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio:
iio: linux/iio.h: fix Excess kernel-doc description warning
MAINTAINERS: Add MAINTAINERS entry for AD7091R
iio: adc: Add support for AD7091R-8
dt-bindings: iio: Add AD7091R-8
iio: adc: Split AD7091R-5 config symbol
iio: adc: ad7091r: Add chip_info callback to get conversion result channel
iio: adc: ad7091r: Set device mode through chip_info callback
iio: adc: ad7091r: Remove unneeded probe parameters
iio: adc: ad7091r: Move chip init data to container struct
iio: adc: ad7091r: Move generic AD7091R code to base driver and header file
iio: adc: ad7091r: Enable internal vref if external vref is not supplied
iio: adc: ad7091r: Allow users to configure device events
iio: dac: driver for MCP4821
dt-bindings: iio: dac: add MCP4821
iio: buffer-dma: split iio_dma_buffer_fileio_free() function
iio: buffer-dma: Get rid of outgoing queue
iio: buffer: Use IIO_SEPARATE instead of a hard-coded 0
Marcelo Schmitt [Tue, 19 Dec 2023 20:32:59 +0000 (17:32 -0300)]
MAINTAINERS: Add MAINTAINERS entry for AD7091R
The driver for AD7091R was added in ca693001: iio: adc: Add support for AD7091R5 ADC
but no MAINTAINERS file entry was added for it since then.
Add a proper MAINTAINERS file entry for the AD7091R driver.
Marcelo Schmitt [Tue, 19 Dec 2023 20:32:36 +0000 (17:32 -0300)]
iio: adc: Add support for AD7091R-8
Add support for Analog Devices AD7091R-2, AD7091R-4, and AD7091R-8
low power 12-Bit SAR ADCs with SPI interface.
Extend ad7091r-base driver so it can be used by the AD7091R-8 driver.
Marcelo Schmitt [Tue, 19 Dec 2023 20:29:30 +0000 (17:29 -0300)]
iio: adc: ad7091r: Add chip_info callback to get conversion result channel
AD7091R-5 and AD7091R-2/-4/-8 have slightly different register field
layout and due to that require different masks for getting the index of
the channel associated with each read.
Add a callback function so the base driver can get correct channel ID
for each chip variant.
Marcelo Schmitt [Tue, 19 Dec 2023 20:28:45 +0000 (17:28 -0300)]
iio: adc: ad7091r: Set device mode through chip_info callback
AD7091R-5 devices have a few modes of operation (sample, command,
autocycle) which are set by writing to configuration register fields.
Follow up patches will add support for AD7091R-2/-4/-8 which don't have
those operation modes nor the register fields for setting them.
Make ad7091r_set_mode() a callback function of AD7091R chip_info struct
so the base driver can appropriately handle each design without having
to check which actual chip is connected.
With the grouping of ad7091r initialization data and callbacks into the
init_info struct, there is no more need to pass the device name and
register map through probe function parameters as those will be available
in the init_info object.
Remove probe parameters that are not needed anymore.
Marcelo Schmitt [Tue, 19 Dec 2023 20:27:36 +0000 (17:27 -0300)]
iio: adc: ad7091r: Move chip init data to container struct
AD7091R designs may differ on their communication protocol and resources
required for proper setup. Extract what is design specific into a
init_info struct so the base driver can use data and callback functions
from that struct rather than checking which specific chip is connected
during device initialization.
Marcelo Schmitt [Tue, 19 Dec 2023 20:27:04 +0000 (17:27 -0300)]
iio: adc: ad7091r: Move generic AD7091R code to base driver and header file
Some code generic to AD7091R devices such as channel definitions were in
the AD7091R-5 driver. There was also some generic register definitions
declared in the base driver which would make more sense to be in the
header file.
The device state struct will be needed for the ad7091r8 driver in a
follow up patch so that ought to be moved to the header file as well.
Lastly, a couple of regmap callback functions are also capable of
abstracting characteristics of different AD7091R devices and those are
now being exported to IIO_AD7091R name space.
Move AD7091R generic code either to the base driver or to the header
file so both the ad7091r5 and the ad7091r8 driver can use those
declaration in follow up patches.
Marcelo Schmitt [Tue, 19 Dec 2023 20:26:27 +0000 (17:26 -0300)]
iio: adc: ad7091r: Enable internal vref if external vref is not supplied
The ADC needs a voltage reference to work correctly.
Users can provide an external voltage reference or use the chip internal
reference to operate the ADC.
The availability of an in chip reference for the ADC saves the user from
having to supply an external voltage reference, which makes the external
reference an optional property as described in the device tree
documentation.
Though, to use the internal reference, it must be enabled by writing to
the configuration register.
Enable AD7091R internal voltage reference if no external vref is supplied.
Marcelo Schmitt [Tue, 19 Dec 2023 20:26:01 +0000 (17:26 -0300)]
iio: adc: ad7091r: Allow users to configure device events
AD7091R-5 devices are supported by the ad7091r-5 driver together with
the ad7091r-base driver. Those drivers declared iio events for notifying
user space when ADC readings fall bellow the thresholds of low limit
registers or above the values set in high limit registers.
However, to configure iio events and their thresholds, a set of callback
functions must be implemented and those were not present until now.
The consequence of trying to configure ad7091r-5 events without the
proper callback functions was a null pointer dereference in the kernel
because the pointers to the callback functions were not set.
Implement event configuration callbacks allowing users to read/write
event thresholds and enable/disable event generation.
Since the event spec structs are generic to AD7091R devices, also move
those from the ad7091r-5 driver the base driver so they can be reused
when support for ad7091r-2/-4/-8 be added.
Randy Dunlap [Fri, 22 Dec 2023 05:25:20 +0000 (21:25 -0800)]
virt: vbox: linux: fix all kernel-doc warnings
Use kernel-doc format for functions that are almost complete in their
kernel-doc comments.
For other functions, just change the comment to a common C comment.
This prevents 7 kernel-doc warnings.
Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Cc: Hans de Goede <hdegoede@redhat.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Link: https://lore.kernel.org/r/20231222052521.14333-2-rdunlap@infradead.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Merge tag 'icc-6.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/djakov/icc into char-misc-next
Georgi writes:
interconnect changes for 6.8
This pull request contains the interconnect changes for the 6.8-rc1 merge
window. These are just driver changes with the following highlights:
Driver changes:
- New interconnect driver for the SM8650 platform.
- New interconnect driver for the SM6115 platform.
- New interconnect driver for the X1E80100 (Snapdragon X Elite) platform.
- Add compatible string for the BWMONv4 instance on the QCM2290 platform.
- Complete the platform drivers conversion to the .remove_new callback
returning void (mostly iMX, Exynos and the rest of Qcom drivers).
Signed-off-by: Georgi Djakov <djakov@kernel.org>
* tag 'icc-6.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/djakov/icc:
interconnect: qcom: sm6115: Fix up includes
dt-bindings: interconnect: qcom,msm8998-bwmon: Add QCM2290 bwmon instance
dt-bindings: interconnect: qcom,msm8998-bwmon: Add SM6115 bwmon instance
interconnect: qcom: Add SM6115 interconnect provider driver
dt-bindings: interconnect: Add Qualcomm SM6115 NoC
interconnect: qcom: Add X1E80100 interconnect provider driver
dt-bindings: interconnect: Add Qualcomm X1E80100 SoC
dt-bindings: interconnect: qcom-bwmon: document SM8650 BWMONs
interconnect: qcom: introduce RPMh Network-On-Chip Interconnect on SM8650 SoC
dt-bindings: interconnect: document the RPMh Network-On-Chip Interconnect in Qualcomm SM8650 SoC
interconnect: exynos: Convert to platform remove callback returning void
interconnect: qcom/smd-rpm: Convert to platform remove callback returning void
interconnect: qcom/osm-l3: Convert to platform remove callback returning void
interconnect: qcom/msm8974: Convert to platform remove callback returning void
interconnect: imx8mq: Convert to platform remove callback returning void
interconnect: imx8mp: Convert to platform remove callback returning void
interconnect: imx8mn: Convert to platform remove callback returning void
interconnect: imx8mm: Convert to platform remove callback returning void
interconnect: qcom: Make qnoc_remove return void
Anshul Dalal [Wed, 20 Dec 2023 15:19:53 +0000 (20:49 +0530)]
iio: dac: driver for MCP4821
Adds driver for the MCP48xx series of DACs.
Device uses a simplex SPI channel. To set the value of an output channel,
a 16-bit data of following format must be written:
Bit field | Description
15 [MSB] | Channel selection bit
0 -> Channel A
1 -> Channel B
13 | Output Gain Selection bit
0 -> 2x Gain (Vref = 4.096V)
1 -> 1x Gain (Vref = 2.048V)
12 | Output Shutdown Control bit
0 -> Shutdown the selected channel
1 -> Active mode operation
11-0 [LSB]| DAC Input Data bits
Value's big endian representation is taken as input for the
selected DAC channel. For devices with a resolution of less
than 12-bits, only the x most significant bits are considered
where x is the resolution of the device.
Reference: Page#22 [MCP48x2 Datasheet]
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:11 +0000 (18:32 +0100)]
fpga: zynq-fpga: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:10 +0000 (18:32 +0100)]
fpga: xilinx-pr-decoupler: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:09 +0000 (18:32 +0100)]
fpga: stratix10-soc: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:08 +0000 (18:32 +0100)]
fpga: socfpga-a10: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:07 +0000 (18:32 +0100)]
fpga: of-fpga-region: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:06 +0000 (18:32 +0100)]
fpga: intel-m10-bmc-sec-update: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:05 +0000 (18:32 +0100)]
fpga: dfl-fme-region: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:04 +0000 (18:32 +0100)]
fpga: dfl-fme-main: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:03 +0000 (18:32 +0100)]
fpga: dfl-fme-br: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:02 +0000 (18:32 +0100)]
fpga: dfl-afu-main: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:01 +0000 (18:32 +0100)]
fpga: altera-hps2fpga: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:32:00 +0000 (18:32 +0100)]
fpga: altera-freeze-bridge: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Uwe Kleine-König [Tue, 19 Dec 2023 17:31:59 +0000 (18:31 +0100)]
fpga: altera-fpga2sdram: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
iio: buffer-dma: split iio_dma_buffer_fileio_free() function
This change splits the logic into a separate function, which will be
re-used later.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com> Cc: Alexandru Ardelean <ardeleanalex@gmail.com> Signed-off-by: Paul Cercueil <paul@crapouillou.net> Link: https://lore.kernel.org/r/20231219175009.65482-3-paul@crapouillou.net Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Paul Cercueil [Tue, 19 Dec 2023 17:50:02 +0000 (18:50 +0100)]
iio: buffer-dma: Get rid of outgoing queue
The buffer-dma code was using two queues, incoming and outgoing, to
manage the state of the blocks in use.
While this totally works, it adds some complexity to the code,
especially since the code only manages 2 blocks. It is much easier to
just check each block's state manually, and keep a counter for the next
block to dequeue.
Since the new DMABUF based API wouldn't use the outgoing queue anyway,
getting rid of it now makes the upcoming changes simpler.
With this change, the IIO_BLOCK_STATE_DEQUEUED is now useless, and can
be removed.
iio: buffer: Use IIO_SEPARATE instead of a hard-coded 0
Use an explicit IIO_SEPARATE instead of 0 for the 'shared_by' parameter
when calling __iio_add_chan_devattr().
For some reason, commit 3704432fb1fd ("iio: refactor info mask and ext_info
attribute creation.") updated only 1 place out of 4.
Update the remaining ones now.
Merge tag 'w1-drv-6.8' of https://git.kernel.org/pub/scm/linux/kernel/git/krzk/linux-w1 into char-misc-next
Krzysztof writes:
1-Wire bus drivers for v6.8
1. Add new AMD AXI 1-wire host driver for AMD programmable logic IP
core.
2. Add support for Analog Devices DS28EC20 EEPROM to existing DS2433
driver.
3. Few cleanups in W1 GPIO driver.
* tag 'w1-drv-6.8' of https://git.kernel.org/pub/scm/linux/kernel/git/krzk/linux-w1:
w1: ds2433: add support for ds28ec20 eeprom
w1: ds2433: use the kernel bitmap implementation
w1: ds2433: introduce a configuration structure
w1: ds2433: remove unused definitions
w1: ds2490: support block sizes larger than 128 bytes in ds_read_block
w1: amd_axi_w1: Explicitly include correct DT includes
w1: gpio: rename pointer to driver data from pdata to ddata
w1: gpio: Drop unused enable_external_pullup from driver data
w1: gpio: Don't use platform data for driver data
w1: Add AXI 1-wire host driver for AMD programmable logic IP core
dt-bindings: w1: Add AMD AXI w1 host and MAINTAINERS entry
Merge tag 'iio-for-6.8a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into char-misc-next
Jonathan writes:
1st set of IIO new device support, features and cleanup for 6.8
New device support
------------------
adi,hmc425a
* Add support for ADRF5740 attenuators. Minor changes to driver needed
alongside new IDs.
aosong,ags02ma
* New driver for this volatile organic compounds sensor.
bosch,bmp280
* Add BMP390 (small amount of refactoring + ID)
bosch,bmi323
* New driver to support the BMI323 6-axis IMU.
honeywell,hsc030pa
* New driver supporting a huge number of SSC and HSC series pressure and
temperature sensors.
isil,isl76682
* New driver for this simple Ambient Light sensor.
liteon,ltr390
* New driver for this ambient and ultraviolet light sensor.
maxim,max34408
* New driver to support the MAX34408 and MAX34409 current monitoring ADCs.
melexis,mlx90635
* New driver for this Infrared contactless temperature sensor.
mirochip,mcp9600
* New driver for this thermocouple EMF convertor.
ti,hdc3020
* New driver for this integrated relative humidity and temperature
sensor.
vishay,veml6075
* New driver for this UVA and UVB light sensor.
General features
----------------
Device properties
* Add fwnode_property_match_property_string() helper to allow matching
single value property against an array of predefined strings.
* Use fwnode_property_string_array_count() inside
fwnode_property_match_string() instead of open coding the same.
checkpatch.pl
* Add exclusion of __aligned() from a warning reducing false positives
on IIO drivers (and hopefully beyond)
IIO Features
------------
core
* New light channel modifiers for UVA and UVB.
* Add IIO_CHAN_INFO_TROUGH as counterpart to IIO_CHAN_INFO_PEAK so that
we can support device that keep running track of the lowest value they
have seen in similar fashion to the existing peak tracking.
adi,adis library
* Use spi cs inactive delay even when a burst reading is performed.
As it's now used every time, can centralize the handling in the SPI
setup code in the driver.
adi,ad2s1210
* Support for fixed-mode to this resolver driver where the A0 and A1
pins are hard wired to config mode in which case position and config
must be read from appropriate config registers.
* Support reset GPIO if present.
adi,ad5791
* Allow configuration of presence of external amplifier in DT binding.
adi,adis16400
* Add spi-cs-inactive-delay-ns to bindings to allow it to be tweaked
if default delays are not quite enough for a specific board.
adi,adis16475
* Add spi-cs-inactive-delay-ns to bindings to allow it to be tweaked
if default delays are not quite enough for a specific board.
bosch,bmp280
* Enable multiple chip IDs per family of devices.
rohm,bu27008
* Add an illuminance channel calculated from RGB and IR data.
Cleanup
-------
Minor white space, typos and tidy up not explicitly called out.
Core
* Check that the available_scan_masks array passed to the IIO core
by a driver is sensible by ensuring the entries are ordered so the
minimum number of channels is enabled in the earlier entries (as they
will be selected if sufficient for the requested channels).
* Document that the available_scan_masks infrastructure doesn't currently
handle masks that don't fit in a long int.
* Improve intensity documentation to reflect that there is no expectation
of sensible units (it's dependent on a frequency sensitivity curve)
Various
* Use new device_property_match_property_string() to replace open coded
versions of the same thing.
* Fix a few MAINTAINERS filenames.
* i2c_get_match_data() and spi_get_device_match_data() pushed into
more drivers reducing boilerplate handling.
* Some unnecessary headers removed.
* ACPI_PTR() removals. It's rarely worth using this.
adi,ad7091r (early part of a series adding device support - useful in
their own right)
* Pass iio_dev directly an event handler rather than relying
on broken use of dev_get_drvdata() as drvdata is never set in this driver.
* Make sure alert is turned on.
adi,ad9467 (general driver fixing up as precursor to iio-backend proposal
which is under review for 6.9)
* Fix reset gpio handling to match expected polarity.
* Always handle error codes from spi_writes.
* Add a driver instance local mutex to avoid some races.
* Fix scale setting to align with available scale values.
* Split array of chip_info structures up into named individual elements.
* Convert to regmap.
honeywell,mprls0025pa
* Drop now unnecessary type references in DT binding for properties in
pascals.
invensense,mpu6050
* Don't eat a potentially useful return value from regmap_bulk_write()
invensense,icm42600
* Use max macro to improve code readability and save a few lines.
liteon,ltrf216a
* Improve prevision of light intensity.
microchip,mcp3911
* Use cleanup.h magic.
qcom,spmi*
* Fix wrong descriptions of SPMI reg fields in bindings.
Other
----
mailmap
* Update for Matt Ranostay
* tag 'iio-for-6.8a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio: (83 commits)
iio: adc: ad7091r: Align arguments to function call parenthesis
iio: adc: ad7091r: Set alert bit in config register
iio: adc: ad7091r: Pass iio_dev to event handler
scripts: checkpatch: Add __aligned to the list of attribute notes
iio: chemical: add support for Aosong AGS02MA
dt-bindings: iio: chemical: add aosong,ags02ma
dt-bindings: vendor-prefixes: add aosong
iio: accel: bmi088: update comments and Kconfig
dt-bindings: iio: humidity: Add TI HDC302x support
iio: humidity: Add driver for ti HDC302x humidity sensors
iio: ABI: document temperature and humidity peak/trough raw attributes
iio: core: introduce trough info element for minimum values
iio: light: driver for Lite-On ltr390
dt-bindings: iio: light: add ltr390
iio: light: isl76682: remove unreachable code
iio: pressure: driver for Honeywell HSC/SSC series
dt-bindings: iio: pressure: add honeywell,hsc030
doc: iio: Document intensity scale as poorly defined
dt-bindings: iio: temperature: add MLX90635 device
iio: temperature: mlx90635 MLX90635 IR Temperature sensor
...
Marc Ferland [Mon, 18 Dec 2023 15:02:30 +0000 (10:02 -0500)]
w1: ds2433: add support for ds28ec20 eeprom
The ds28ec20 eeprom is (almost) backward compatible with the
ds2433. The only differences are:
- the eeprom size is now 2560 bytes instead of 512;
- the number of pages is now 80 (same page size as the ds2433: 256 bits);
- the programming time has increased from 5ms to 10ms;
This patch adds support for the ds28ec20 to the ds2433 driver. From
the datasheet: The DS28EC20 provides a high degree of backward
compatibility with the DS2433. Besides the different family codes, the
only protocol change that is required on an existing DS2433
implementation is a lengthening of the programming duration (tPROG)
from 5ms to 10ms.
dmesg now returns:
w1_master_driver w1_bus_master1: Attaching one wire slave 43.000000478756 crc e0
instead of:
w1_master_driver w1_bus_master1: Attaching one wire slave 43.000000478756 crc e0
w1_master_driver w1_bus_master1: Family 43 for 43.000000478756.e0 is not registered.
Test script writing/reading random data (CONFIG_W1_SLAVE_DS2433_CRC is
not set):
Marc Ferland [Mon, 18 Dec 2023 15:02:29 +0000 (10:02 -0500)]
w1: ds2433: use the kernel bitmap implementation
The ds2433 driver uses the 'validcrc' variable to mark out which pages
have been successfully (crc is valid) retrieved from the eeprom and
placed in the internal 'memory' buffer (see CONFIG_W1_SLAVE_DS2433_CRC).
The current implementation assumes that the number of pages will never
go beyond 32 pages (bit field is a u32). This is fine for the ds2433
since it only has 16 pages.
On the ds28ec20 though, the number of pages increases to 80 which will
not fit on a single u32.
As a solution, I replaced the u32 variable with a standard bitmap and
set the number of bits to 32 which is the same size we had before.
Marc Ferland [Mon, 18 Dec 2023 15:02:28 +0000 (10:02 -0500)]
w1: ds2433: introduce a configuration structure
Add a ds2433_config structure for parameters that are different
between the ds2433 and the ds28ec20. The goal is to reuse the same
code for both chips.
A pointer to this config structure is added to w1_f23_data and the
CONFIG_W1_SLAVE_DS2433_CRC ifdefs are adjusted since now both driver
configurations (with or without crc support) will make use of
w1_f23_data.
Also, the 'memory' buffer is now dynamically allocated based on the
size specififed in the config structure to help support memories of
different sizes.
usb 5-1: Failed to write 1-wire data to ep0x2: err=-110.
I discovered this issue while implementing support for the ds28ec20
eeprom in the w1-2433 driver. This driver accepts reading blocks of
sizes up to the size of the entire memory (2560 bytes in the case of
the ds28ec20). Note that this issue _does not_ arise when the kernel
is configured with CONFIG_W1_SLAVE_DS2433_CRC enabled since in this
mode the driver reads one 32 byte block at a time (a single memory
page).
Also, from the ds2490 datasheet (2995.pdf, page 22, BLOCK I/O
command):
For a block write sequence the EP2 FIFO must be pre-filled with
data before command execution. Additionally, for block sizes
greater then the FIFO size, the FIFO content status must be
monitored by host SW so that additional data can be sent to the
FIFO when necessary. A similar EP3 FIFO content monitoring
requirement exists for block read sequences. During a block read
the number of bytes loaded into the EP3 FIFO must be monitored so
that the data can be read before the FIFO overflows.
Breaking the block in smaller 128 bytes chunks and simply calling the
original code sequence has solved the issue for me.
Tested with a DS1490F usb<->one-wire adapter and both the DS28EC20 and
DS2433 eeprom memories.
Merge tag 'extcon-next-for-6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/extcon into char-misc-next
Chanwoo writes:
Update extcon next for v6.8
Detailed description for this pull request:
1. Fix possible memory leak of device name in extcon_dev_register()
: Fix memory leak on error path of extcon_dev_register().
2. Set interrupt polarity based on device-tree for extcon-usbc-tusb320.c
:Remove 'IRQF_TRIGGER_FALLING' request which is not allowed on
every interrupt controller (i.e. arm64 GIC). Replace flag by a
request that depends on the actual device-tree setting.
3. Fix the comment style according to guide on extcon-qcom-spmi-misc.c.
* tag 'extcon-next-for-6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/extcon:
extcon: qcom-spmi-misc: don't use kernel-doc marker for comment
extcon: usbc-tusb320: Set interrupt polarity based on device-tree
extcon: fix possible name leak in extcon_dev_register()
Merge tag 'mhi-for-v6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/mani/mhi into char-misc-next
Manivannan writes:
MHI Host
========
- Added alignment check for event ring read pointer to avoid the potential
buffer corruption issue.
- Added support for SDX75 modem which takes longer time to enter READY state.
- Added spinlock to protect concurrent access while queuing transfer ring
elements.
- Dropped the read channel lock before invoking the client callback as the
client can potentially queue buffers thus ending up wtih soft lockup.
MHI Endpoint
============
- Used kzalloc() to allocate event ring elements instead of allocating the
elements on the stack, as the endpoint controller trying to queue them may not
be able to use vmalloc memory (using DMA).
- Used slab allocator for allocting the memory for objects used frequently and
are of fixed size.
- Added support for interrupt moderation timer feature which is used by the host
to limit the number of interrupts raised by the device for an event ring.
- Added async read/write DMA support for transferring data between host and the
endpoint. So far MHI EP stack assumed that the data will be transferred
synchronously (i.e., it sends completion once the transfer APIs are returned).
But this impacts the throughput if the controller is using DMA to do the
transfer.
So to add async suport, existing sync transfer APIs are renamed to
{read/write}_sync and also introduced two new APIs {read/write}_async for
carrying out the async transfer.
Controllers implementing the async APIs should queue the buffers and return
immediately without waiting for transfer completion. Once the transfer
completion happens later, they should invoke the completion callback so that
the MHI EP stack can send the completion event to the host.
The controller driver patches (PCI EPF) for this async support are also merged
to the MHI tree with Acks from PCI maintainers.
- Fixed the DMA channel direction in error path of the PCI EPF driver.
* tag 'mhi-for-v6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/mani/mhi:
bus: mhi: host: Drop chan lock before queuing buffers
bus: mhi: host: Add spinlock to protect WP access when queueing TREs
PCI: epf-mhi: Fix the DMA data direction of dma_unmap_single()
bus: mhi: ep: Add checks for read/write callbacks while registering controllers
bus: mhi: ep: Add support for async DMA read operation
bus: mhi: ep: Add support for async DMA write operation
PCI: epf-mhi: Enable MHI async read/write support
PCI: epf-mhi: Add support for DMA async read/write operation
PCI: epf-mhi: Simulate async read/write using iATU
bus: mhi: ep: Introduce async read/write callbacks
bus: mhi: ep: Rename read_from_host() and write_to_host() APIs
bus: mhi: ep: Pass mhi_ep_buf_info struct to read/write APIs
bus: mhi: ep: Add support for interrupt moderation timer
bus: mhi: ep: Use slab allocator where applicable
bus: mhi: host: Add alignment check for event ring read pointer
bus: mhi: host: pci_generic: Add SDX75 based modem support
bus: mhi: host: Add a separate timeout parameter for waiting ready
bus: mhi: ep: Do not allocate event ring element on stack
Fabrice Gasnier [Wed, 13 Dec 2023 17:31:16 +0000 (18:31 +0100)]
tools/counter: add a flexible watch events tool
This adds a new counter tool to be able to test various watch events.
A flexible watch array can be populated from command line, each field
may be tuned with a dedicated command line sub-option in "--watch" string.
Several watch events can be defined, each can have specific watch options,
by using "--watch <watch 1 options> --watch <watch 2 options>".
Watch options is a comma separated list.
It also comes with a simple default watch (to monitor overflow/underflow
events), used when no watch parameters are provided. It's equivalent to:
counter_watch_events -w comp_count,scope_count,evt_ovf_udf
The print_usage() routine proposes another example, from the command line,
which generates a 2 elements watch array, to monitor:
- overflow underflow events
- capture events, on channel 3, that reads read captured data by
specifying the component id (capture3_component_id being 7 here).
Merge tag 'fpga-for-6.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/fpga/linux-fpga into char-misc-next
Xu writes:
FPGA Manager changes for 6.8-rc1
DFL:
- Philipp's change uses memdup_array_user() to do better overflow check
All patches have been reviewed on the mailing list, and have been in the
last linux-next releases (as part of our for-next branch).
Signed-off-by: Xu Yilun <yilun.xu@intel.com>
* tag 'fpga-for-6.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/fpga/linux-fpga:
drivers/fpga: use standard array-copy function
Marcelo Schmitt [Sat, 16 Dec 2023 17:46:37 +0000 (14:46 -0300)]
iio: adc: ad7091r: Set alert bit in config register
The ad7091r-base driver sets up an interrupt handler for firing events
when inputs are either above or below a certain threshold.
However, for the interrupt signal to come from the device it must be
configured to enable the ALERT/BUSY/GPO pin to be used as ALERT, which
was not being done until now.
Enable interrupt signals on the ALERT/BUSY/GPO pin by setting the proper
bit in the configuration register.
Marcelo Schmitt [Sat, 16 Dec 2023 17:46:11 +0000 (14:46 -0300)]
iio: adc: ad7091r: Pass iio_dev to event handler
Previous version of ad7091r event handler received the ADC state pointer
and retrieved the iio device from driver data field with dev_get_drvdata().
However, no driver data have ever been set, which led to null pointer
dereference when running the event handler.
Pass the iio device to the event handler and retrieve the ADC state struct
from it so we avoid the null pointer dereference and save the driver from
filling the driver data field.
Marcelo Schmitt [Sat, 16 Dec 2023 17:45:27 +0000 (14:45 -0300)]
scripts: checkpatch: Add __aligned to the list of attribute notes
Checkpatch presumes attributes marked with __aligned(alignment) are part
of a function declaration and throws a warning stating that those
compiler attributes should have an identifier name which is not correct.
Add __aligned compiler attributes to the list of attribute notes
so they don't cause warnings anymore.
Anshul Dalal [Fri, 15 Dec 2023 16:23:11 +0000 (21:53 +0530)]
iio: chemical: add support for Aosong AGS02MA
A simple driver for the TVOC (Total Volatile Organic Compounds)
sensor from Aosong: AGS02MA
Steps in reading the VOC sensor value over i2c:
1. Read 5 bytes from the register `AGS02MA_TVOC_READ_REG` [0x00]
2. The first 4 bytes are taken as the big endian sensor data with final
byte being the CRC
3. The CRC is verified and the value is returned over an
`IIO_CHAN_INFO_RAW` channel as percents
Qiang Yu [Mon, 11 Dec 2023 06:42:52 +0000 (14:42 +0800)]
bus: mhi: host: Drop chan lock before queuing buffers
Ensure read and write locks for the channel are not taken in succession by
dropping the read lock from parse_xfer_event() such that a callback given
to client can potentially queue buffers and acquire the write lock in that
process. Any queueing of buffers should be done without channel read lock
acquired as it can result in multiple locks and a soft lockup.
Cc: <stable@vger.kernel.org> # 5.7 Fixes: 1d3173a3bae7 ("bus: mhi: core: Add support for processing events from client device") Signed-off-by: Qiang Yu <quic_qianyu@quicinc.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Tested-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Link: https://lore.kernel.org/r/1702276972-41296-3-git-send-email-quic_qianyu@quicinc.com
[mani: added fixes tag and cc'ed stable] Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Bhaumik Bhatt [Mon, 11 Dec 2023 06:42:51 +0000 (14:42 +0800)]
bus: mhi: host: Add spinlock to protect WP access when queueing TREs
Protect WP accesses such that multiple threads queueing buffers for
incoming data do not race.
Meanwhile, if CONFIG_TRACE_IRQFLAGS is enabled, irq will be enabled once
__local_bh_enable_ip is called as part of write_unlock_bh. Hence, let's
take irqsave lock after TRE is generated to avoid running write_unlock_bh
when irqsave lock is held.
Cc: stable@vger.kernel.org Fixes: 189ff97cca53 ("bus: mhi: core: Add support for data transfer") Signed-off-by: Bhaumik Bhatt <bbhatt@codeaurora.org> Signed-off-by: Qiang Yu <quic_qianyu@quicinc.com> Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Tested-by: Jeffrey Hugo <quic_jhugo@quicinc.com> Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Link: https://lore.kernel.org/r/1702276972-41296-2-git-send-email-quic_qianyu@quicinc.com Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
parport: parport_serial: Add Brainboxes device IDs and geometry
Add device IDs for the Brainboxes UC-203, UC-257, UC-414, UC-475,
IS-300/IS-500 and PX-263/PX-295 and define the relevant "geometry"
for the cards.
This patch requires part 1 of this series.
Michal Simek [Thu, 14 Dec 2023 14:53:47 +0000 (15:53 +0100)]
firmware: xilinx: Remove clock_setrate and clock_getrate api
As per the current code base, PM_CLOCK_SETRATE and PM_CLOCK_GETRATE
APIs are not supported for the runtime operations. In the case of
ZynqMP returning an error from TF-A when there is any request to
access these APIs and for Versal also it is returning an error like
NO_ACCESS from the firmware. So, just removing the unused code to
avoid the confusion around these APIs.
Also, there is no issue with the backward compatibility as these APIs
were never used since implemented. Hence no need to bump up the
version of the feature check API as well.
Use struct_size() instead of handle_arr_calc_size().
This is much more conventional.
While at it, use size_add() when computing the needed size in
vmci_handle_arr_append_entry(). This prevents from (unlikely) overflow
when computing the new size to reallocate.
Fei Shao [Wed, 6 Dec 2023 23:17:28 +0000 (15:17 -0800)]
spmi: mtk-pmif: Reorder driver remove sequence
This driver enables clocks and then adds SPMI controller in probing, so
we expect the reversed sequence in removal.
Fix the order in the remove callback.
Yu-Che Cheng [Wed, 6 Dec 2023 23:17:25 +0000 (15:17 -0800)]
spmi: mediatek: Fix UAF on device remove
The pmif driver data that contains the clocks is allocated along with
spmi_controller.
On device remove, spmi_controller will be freed first, and then devres
, including the clocks, will be cleanup.
This leads to UAF because putting the clocks will access the clocks in
the pmif driver data, which is already freed along with spmi_controller.
This can be reproduced by enabling DEBUG_TEST_DRIVER_REMOVE and
building the kernel with KASAN.
Fix the UAF issue by using unmanaged clk_bulk_get() and putting the
clocks before freeing spmi_controller.
spmi: mtk-pmif: Serialize PMIF status check and command submission
Before writing the read or write command to the SPMI arbiter through the
PMIF interface, the current status of the channel is checked to ensure
it is idle. However, since the status only changes from idle when the
command is written, it is possible for two concurrent calls to determine
that the channel is idle and simultaneously send their commands. At this
point the PMIF interface hangs, with the status register no longer being
updated, and thus causing all subsequent operations to time out.
This was observed on the mt8195-cherry-tomato-r2 machine, particularly
after commit 46600ab142f8 ("regulator: Set PROBE_PREFER_ASYNCHRONOUS for
drivers between 5.10 and 5.15") was applied, since then the two MT6315
devices present on the SPMI bus would probe assynchronously and
sometimes (during probe or at a later point) read the bus
simultaneously, breaking the PMIF interface and consequently slowing
down the whole system.
To fix the issue at its root cause, introduce locking around the channel
status check and the command write, so that both become an atomic
operation, preventing race conditions between two (or more) SPMI bus
read/write operations. A spinlock is used since this is a fast bus, as
indicated by the usage of the atomic variant of readl_poll, and
'.fast_io = true' being used in the mt6315 driver, so spinlocks are
already used for the regmap access.
projects / thirdparty / kernel / linux.git / log
