Enabling append status breaks projects that also use SPI engine offload
(cn0363). Append status is only necessary when using CPU for reading
samples.

Check if SPI engine offload is used before enabling append status.

Fixes: 27b588f ("Merge tag 'xilinx-v2023.1' of https://github.com/Xilinx/linux-xlnx.git")
Signed-off-by: Dumitru Ceclan <dumitru.ceclan@analog.com>

ad7768_scale() is dependent on the st->vref voltage. Hence, simplify the
code and get the voltage only during probe (as it should not change at
runtime) with devm_regulator_get_enable_read_voltage()

[nsa: devm_regulator_get_enable_read_voltage() is not available in
2023_R2, so do it by hand to get the voltage during probe]

Fixes: 92cc452 ("iio: adc: ad7768: simplify probe")
Co-developed-by: Dragos Bogdan <dragos.bogdan@analog.com>
Signed-off-by: Dragos Bogdan <dragos.bogdan@analog.com>
Signed-off-by: Nuno Sa <nuno.sa@analog.com>
(cherry picked from commit 44e7ac1)

Similar to

commit 7ebe866 ("iio: adc: adrv9002: fixup tx2 power level")

we need to apply the same fixup even in rx2tx2. The reasoning for not
doing it at the time was that if the channels are tuned together and
have the same SSI delays, it should not be needed. Well...

While at it (and as advised by HW folks), we're now doing the fixup for
all TX channels.

Signed-off-by: Nuno Sa <nuno.sa@analog.com>
(cherry picked from commit 13a58d5)

spi_split_transfers_maxsize() was recently changed to not take a gfp
parameter. Fix this caller accordingly to make the spi-bcm2835 driver
compile again.

Fixes: 98847e6 ("spi: drop gpf arg from __spi_split_transfer_maxsize()")
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@baylibre.com>
(cherry picked from commit d9dc785)

The patch posted at
https://lore.kernel.org/r/20240206200648.1782234-1-dlechner@baylibre.com
was based on next-20240207 (and was applied as c0c0293 ("spi: drop
gpf arg from __spi_split_transfer_maxsize()") for v6.9-rc1).

However when applied to a v6.6 based tree, the conflict with commit
f6cd662 ("spi: stm32: add st,stm32mp25-spi compatible supporting
STM32MP25 soc") wasn't resolved correctly.

Fix that to make the driver build again.

Fixes: 98847e6 ("spi: drop gpf arg from __spi_split_transfer_maxsize()")
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@baylibre.com>
(cherry picked from commit 455592e)

This reverts commit f6e3323.

The commit is clearly bogus and does not even compile with clang 18.1.8.
Hence, let's revert it and wait until AMD fixes it (unless we need it
first). Here are the compiler complains (warnings included):

././include/linux/compiler_types.h:357:45: error: call to '__compiletime_assert_273' declared with attribute error: FIELD_GET: mask is zero
  357 |         _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)

  CALL    scripts/checksyscalls.sh
  CC      drivers/iio/adc/xilinx-ams.o
drivers/iio/adc/xilinx-ams.c: In function 'ams_update_alarm':
drivers/iio/adc/xilinx-ams.c:416:51: warning: left shift count >= width of type [-Wshift-count-overflow]
  416 |                              (AMS_ISR1_ALARM_MASK << AMS_ISR1_ALARM_SHIFT)),
      |                                                   ^~
In file included from ./include/linux/bits.h:6,
                 from drivers/iio/adc/xilinx-ams.c:11:
drivers/iio/adc/xilinx-ams.c: In function 'ams_get_alarm_mask':
./include/vdso/bits.h:7:40: warning: left shift count >= width of type [-Wshift-count-overflow]
    7 | #define BIT(nr)                 (UL(1) << (nr))
      |                                        ^~
drivers/iio/adc/xilinx-ams.c:921:24: note: in expansion of macro 'BIT'
  921 |                 return BIT(AMS_ALARM_BIT_TEMP_OT_REMOTE);
      |                        ^~~
drivers/iio/adc/xilinx-ams.c: In function 'ams_irq':
drivers/iio/adc/xilinx-ams.c:1129:42: warning: left shift count >= width of type [-Wshift-count-overflow]
 1129 |                     (AMS_ISR1_ALARM_MASK << AMS_ISR1_ALARM_SHIFT)) |

Signed-off-by: Nuno Sa <nuno.sa@analog.com>

The update was not updated during the stable merge. Hence, do it know so
we do not have a broken driver.

Fixes: 3a22b09 ("Merge tag 'v6.1.70' of https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git")
Signed-off-by: Nuno Sa <nuno.sa@analog.com>

Make sure to include slab.h so that the driver compiles.

Signed-off-by: Nuno Sa <nuno.sa@analog.com>

User space types should be '__uXX' instead of 'uXX'. Furthermore,
types.h needs to be included. Otherwise we get a failure doing

HDRTEST usr/include/misc/xilinx_puf.h

Signed-off-by: Nuno Sa <nuno.sa@analog.com>

Add device tree for using adrv9025 with vcu118

Signed-off-by: AndrDragomir <andrei.dragomir@analog.com>

The simpleImage for microblaze didn't include adrv9025 by default.
This is needed for the adrv9025 + vcu118 support.

Signed-off-by: AndrDragomir <andrei.dragomir@analog.com>

The SDK is taken as is which means that compilation will likely fail.

The folloing changes were squashed in the new SDK (need to be sent to
the BU):

* 9d9cde6 ("iio: adc: adrv9002: radio: turn radio functions private")
* 67b6d36 ("iio: adc: adrv9002: radio: fix pin mode")

Signed-off-by: Nuno Sa <nuno.sa@analog.com>
(cherry picked from commit 3c53527)

Arm firmware updated to version 0.22.39.
Default stream also changed.
Warmboot coefficients.

Signed-off-by: Nuno Sa <nuno.sa@analog.com>
(cherry picked from commit d769f82)

The adi_adrv9001_Temperature_Get() API and now get's an extra argument.
Fix that call and use SPI for reading the temperature (which means we
can some outdated values sometimes but that should not be a big issue).

Signed-off-by: Nuno Sa <nuno.sa@analog.com>
(cherry picked from commit 1a5ae65)

Fix 'hdl_project' tag from file header to match HDL project name:
adrv9026/vcu118.

Signed-off-by: Stefan Raus <Stefan.Raus@analog.com>

The SDK is taken as is which means that compilation will likely fail.

The folloing changes were squashed in the new SDK (need to be sent to
the BU):

 * 9d9cde6 ("iio: adc: adrv9002: radio: turn radio functions private")
 * 67b6d36 ("iio: adc: adrv9002: radio: fix pin mode")

Signed-off-by: Nuno Sá <nuno.sa@analog.com>

Removing fan_ctl gpio from zynqmp-jupiter-sdr devicetree
because it will be used in userspace and boot scripts.

Signed-off-by: Ramona Alexandra Nechita <ramona.nechita@analog.com>

Make use of pix to install dtschema. This is the preferred way of
installing apps from python packages (one cannot use pip anymore in
distros like ubuntu 24.04 for example). This creates an isolated
virtual environment and ensures the command are accessible in $PATH.
It also makes sure all dependencies are met and do not interfere
with OS's python packages.

Signed-off-by: Nuno Sa <nuno.sa@analog.com>
(cherry picked from commit daf69ea)

This reverts commit 71f0532.

As it turns out, API 68.14.13 has no meaningful fixes and this release only
affects users using hdl that we do not use in the linux driver. Hence
this would only raise unnecessary questions without any added value.

Signed-off-by: Nuno Sa <nuno.sa@analog.com>

Specify the external shunt resistor value for properly display the current
and the power.

Signed-off-by: Dragos Bogdan <dragos.bogdan@analog.com>