Batch sequential write transfers up to the max TX size (40 bytes).
This controller must specify a max transfer size of only 8 bytes for
RX operations.
Signed-off-by: Eddie James <eajames@linux.ibm.com>
Link: https://patch.msgid.link/20250131200158.732898-1-eajames@linux.ibm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Change suggested from the AXI HDL team, modify the function
axi_dac_data_stream_enable() to check for interface busy, to avoid
possible issues when starting the stream.
Fixes: e61d717842 ("iio: dac: adi-axi-dac: extend features")
Reviewed-by: Nuno Sa <nuno.sa@analog.com>
Signed-off-by: Angelo Dureghello <adureghello@baylibre.com>
Link: https://patch.msgid.link/20250114-wip-bl-ad3552r-axi-v0-iio-testing-carlos-v4-3-979402e33545@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Make available scale building more clear. This hurts the performance
quite a bit by looping throgh the scales many times instead of doing
everything in one loop. It however simplifies logic by:
- decoupling the gain and scale allocations & computations
- keeping the temporary 'per_time_gains' table inside the
per_time_scales computation function.
- separating building the 'all scales' table in own function and doing
it based on the already computed per-time scales.
Signed-off-by: Matti Vaittinen <mazziesaccount@gmail.com>
Tested-by: subhajit.ghosh@tweaklogic.com
Link: https://patch.msgid.link/Z1_rRXqdhxhL6wBw@mva-rohm
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
According to Documentation/timers/delay_sleep_functions.rst,
fsleep() is the preferred delay function to be used in non-atomic
context, so switch to it accordingly.
Suggested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Fabio Estevam <festevam@denx.de>
Link: https://patch.msgid.link/20241209181624.1260868-1-festevam@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Add a custom implementation of regmap read/write callbacks using the SPI
bus. This allows them to be performed at a lower SCLK rate than data
reads. Previously, all SPI transfers were being performed at a lower
speed, but with this change sample data is read at the max bus speed
while the register reads/writes remain at the lower rate.
Also remove .can_multi_write from the AD4695 driver's regmap_configs, as
this isn't implemented or needed.
For some background context, see:
https://lore.kernel.org/linux-iio/20241028163907.00007e12@Huawei.com/
Suggested-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Trevor Gamblin <tgamblin@baylibre.com>
Reviewed-by: David Lechner <dlechner@baylibre.com>
Tested-by: David Lechner <dlechner@baylibre.com>
Link: https://patch.msgid.link/20241113-tgamblin-ad4695_improvements-v2-3-b6bb7c758fc4@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Ensure that conversion mode is successfully exited when the command is
issued by adding an extra transfer beforehand, matching the minimum CNV
high and low times from the AD4695 datasheet. The AD4695 has a quirk
where the exit command only works during a conversion, so guarantee this
happens by triggering a conversion in ad4695_exit_conversion_mode().
Then make this even more robust by ensuring that the exit command is run
at AD4695_REG_ACCESS_SCLK_HZ rather than the bus maximum.
Signed-off-by: Trevor Gamblin <tgamblin@baylibre.com>
Reviewed-by: David Lechner <dlechner@baylibre.com>
Tested-by: David Lechner <dlechner@baylibre.com>
Link: https://patch.msgid.link/20241113-tgamblin-ad4695_improvements-v2-2-b6bb7c758fc4@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This change integrates mc3510c support into the mc3230 driver.
MC3510C uses the same registers as MC3230, but a different value scale.
Tested on Huawei MediaPad T3 10 (huawei-agassi)
Signed-off-by: Vasiliy Doylov <nekodevelopper@gmail.com>
Link: https://patch.msgid.link/20250116-mainlining-mc3510c-v4-5-a41308b85ec2@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Refactor code to support multiple generations of MCUBE devices by
defining name, chip id and product code in mc3230_chip_info struct.
Signed-off-by: Vasiliy Doylov <nekodevelopper@gmail.com>
Link: https://patch.msgid.link/20250116-mainlining-mc3510c-v4-4-a41308b85ec2@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The MC3510C is a 3 asix digital accelerometer. It handled by the same
driver as MC3230. Document it as a trivial device.
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Vasiliy Doylov <nekodevelopper@gmail.com>
Link: https://patch.msgid.link/20250116-mainlining-mc3510c-v4-1-a41308b85ec2@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Replace ternary (condition ? "enable" : "disable") syntax with helpers
from string_choices.h because:
1. Simple function call with one argument is easier to read. Ternary
operator has three arguments and with wrapping might lead to quite
long code.
2. Is slightly shorter thus also easier to read.
3. It brings uniformity in the text - same string.
4. Allows deduping by the linker, which results in a smaller binary
file.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patch.msgid.link/20250114192716.912476-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The ad7625 driver was submitted under a dual BSD/GPL license, but this
isn't a requirement for the code, and adds extra complexity. To make it
consistent with similar drivers, drop the BSD tag and leave it as
GPL-2.0-only.
Suggested-by: Nuno Sa <nuno.sa@analog.com>
Signed-off-by: Trevor Gamblin <tgamblin@baylibre.com>
Link: https://patch.msgid.link/20250114-ad7625_license-v1-1-6555b7be05ab@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Use two separate static const struct ad_sigma_delta_info instances
instead of making a copy for each driver instance.
Typically in the IIO subsystem, we use multiple static const instances
of the same struct when there are different variants of the same family
of devices as opposed to making a copy for each driver instance and
modifying it.
Signed-off-by: David Lechner <dlechner@baylibre.com>
Link: https://patch.msgid.link/20250113-iio-adc-ad7313-fix-non-const-info-struct-v4-2-b63be3ecac4a@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Move the call to fwnode_irq_get_byname() from the driver-specific
ad7173_fw_parse_device_config() to the shared ad_sd_init() function.
The main reason for this is that we want struct ad_sigma_delta_info to
be static const data that describes the actual ADC chip, not the
application-specific configuration or any runtime state.
Previously, this struct was being used to pass the IRQ number to the
shared ad_sd_init() function. Now, this is replaced by a boolean flag
that is set at compile time and the ad_sd_init() function handles
looking up the IRQ number instead. This also has the added benefit that
if any other drivers need to make use of this in the future, they just
have to set the flag and the shared code will take care of the rest
rather than duplicating the code in each driver.
Signed-off-by: David Lechner <dlechner@baylibre.com>
Link: https://patch.msgid.link/20250113-iio-adc-ad7313-fix-non-const-info-struct-v4-1-b63be3ecac4a@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
There's no reason to check for regulator supply property presence before
calling devm_regulator_get_optional() as that will return -ENODEV if
the supply is not present.
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Acked-by: Fabrice Gasnier <fabrice.gasnier@foss.st.com>
Tested-by: Fabrice Gasnier <fabrice.gasnier@foss.st.com>
Link: https://patch.msgid.link/20250109182325.3973684-2-robh@kernel.org
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
DTS example in the bindings should be indented with 2- or 4-spaces and
aligned with opening '- |', so correct any differences like 3-spaces or
mixtures 2- and 4-spaces in one binding.
No functional changes here, but saves some comments during reviews of
new patches built on existing code.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Nuno Sa <nuno.sa@analog.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://patch.msgid.link/20250107125848.226899-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Add a section for alert support, explaining how user can use iio events
attributes to enable alert and set thresholds.
Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20250108-ad7380-add-alert-support-v4-5-1751802471ba@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The alert functionality is an out of range indicator and can be used as
an early indicator of an out of bounds conversion result.
ALERT_LOW_THRESHOLD and ALERT_HIGH_THRESHOLD registers are common to all
channels.
When using 1 SDO line (only mode supported by the driver right now), i.e
data outputs only on SDOA, SDOB (or SDOD for 4 channels variants) is
used as an alert pin. The alert pin is updated at the end of the
conversion (set to low if an alert occurs) and is cleared on a falling
edge of CS.
The ALERT register contains information about the exact alert status:
channel and direction. ALERT register can be accessed using debugfs if
enabled.
User can set high/low thresholds and enable alert detection using the
regular iio events attributes:
events/in_thresh_falling_value events/in_thresh_rising_value
events/thresh_either_en
In most use cases, user will hardwire the alert pin to trigger a shutdown.
In theory, we could generate userspace IIO events for alerts, but this
is not implemented yet for several reasons [1]. This can be implemented
later if a real use case actually requires it.
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
[1] https://lore.kernel.org/all/4be16272-5197-4fa1-918c-c4cdfcaee02e@baylibre.com/
Reviewed-by: David Lechner <dlechner@baylibre.com>
Link: https://patch.msgid.link/20250108-ad7380-add-alert-support-v4-4-1751802471ba@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Since regmap cache is now enabled, we don't need to store the
oversampling ratio in the private data structure.
Reviewed-by: David Lechner <dlechner@baylibre.com>
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20250108-ad7380-add-alert-support-v4-3-1751802471ba@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Conditionnal scoped handlers are turning out to be a real pain:
readability issues, compiler and linker handling issues among others so
rollback and remove the scoped version of iio_dvice_claim_direct_mode.
To impove code readability factorize code to set oversampling ratio.
Signed-off-by: Julien Stephan <jstephan@baylibre.com>
Link: https://patch.msgid.link/20250108-ad7380-add-alert-support-v4-1-1751802471ba@baylibre.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
current-speed description has been added to uartlite description by commit
3de536a8c3 ("dt-bindings: serial: uartlite: Add properties for
synthesis-time parameters") that's why no reason to have in xilinx.txt too.
Fixes: 3de536a8c3 ("dt-bindings: serial: uartlite: Add properties for synthesis-time parameters")
Signed-off-by: Michal Simek <michal.simek@amd.com>
Reviewed-by: Sean Anderson <sean.anderson@seco.com>
Link: https://lore.kernel.org/r/aa0b7f9a851c6b8d11f37050f84e0ec69cfa72a2.1738600116.git.michal.simek@amd.com
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Fix incorrect format of compatible string (comma instead of hyphen) for
TI's AM62A7 SoC.
s/ti,am62a7,dss/ti,am62a7-dss
Fixes: 7959ceb767 ("dt-bindings: display: ti: Add support for am62a7 dss")
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250203155431.2174170-1-devarsht@ti.com
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Fully migrate parisc to the IO functions from lib/iomem_copy.c. In a
recent patch the functions memset_io and memcpy_toio were removed, but
the memcpy_fromio was kept, because for very short sequences it does
half word accesses, whereas the functions in lib/iomem_copy.c do byte
accesses until the memory is naturally aligned and then do machine word
accesses. But I don't think the single half-word access merits keeping
the arch specific implementation, so, remove it as well.
Signed-off-by: Julian Vetter <julian@outer-limits.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Helge Deller <deller@gmx.de>
Mutliple lines in the file contain tabs in lines where there is no
code. Just remove them.
Signed-off-by: Julian Vetter <julian@outer-limits.org>
Signed-off-by: Helge Deller <deller@gmx.de>
Commit 70fb86a85d ("drm/xe: Revert some changes that break a mesa
debug tool") partially reverted some changes to workaround breakage
caused to mesa tools. However, in doing so it also broke fetching the
GuC log via debugfs since xe_print_blob_ascii85() simply bails out.
The fix is to avoid the extra newlines: the devcoredump interface is
line-oriented and adding random newlines in the middle breaks it. If a
tool is able to parse it by looking at the data and checking for chars
that are out of the ascii85 space, it can still do so. A format change
that breaks the line-oriented output on devcoredump however needs better
coordination with existing tools.
v2: Add suffix description comment
v3: Reword explanation of xe_print_blob_ascii85() calling drm_puts()
in a loop
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Julia Filipchuk <julia.filipchuk@intel.com>
Cc: José Roberto de Souza <jose.souza@intel.com>
Cc: stable@vger.kernel.org
Fixes: 70fb86a85d ("drm/xe: Revert some changes that break a mesa debug tool")
Fixes: ec1455ce7e ("drm/xe/devcoredump: Add ASCII85 dump helper function")
Link: https://patchwork.freedesktop.org/patch/msgid/20250123202307.95103-2-jose.souza@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
(cherry picked from commit 2c95bbf500)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Having the exec queue snapshot inside a "GuC CT" section was always
wrong. Commit c28fd6c358 ("drm/xe/devcoredump: Improve section
headings and add tile info") tried to fix that bug, but with that also
broke the mesa tool that parses the devcoredump, hence it was reverted
in commit a53da2fb25 ("drm/xe: Revert some changes that break a mesa
debug tool").
With the mesa tool also fixed, this can propagate as a fix on both
kernel and userspace side to avoid unnecessary headache for a debug
feature.
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Julia Filipchuk <julia.filipchuk@intel.com>
Cc: José Roberto de Souza <jose.souza@intel.com>
Cc: stable@vger.kernel.org
Fixes: a53da2fb25 ("drm/xe: Revert some changes that break a mesa debug tool")
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250123051112.1938193-2-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
(cherry picked from commit a37934ea75)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
We rely on stream->pollin to decide whether or not to block during
poll/read calls. However, currently there are blocking read code paths
which don't even set stream->pollin. The best place to consistently set
stream->pollin for all code paths is therefore to set it in
xe_oa_buffer_check_unlocked.
Fixes: e936f885f1 ("drm/xe/oa/uapi: Expose OA stream fd")
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250115222029.3002103-1-ashutosh.dixit@intel.com
(cherry picked from commit d3fedff828)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
The migration support only needs to be initialized once, but it
was incorrectly called from the xe_gt_sriov_pf_init_hw(), which
is part of the reset flow and may be called multiple times.
Fixes: d86e3737c7 ("drm/xe/pf: Add functions to save and restore VF GuC state")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250120232443.544-1-michal.wajdeczko@intel.com
(cherry picked from commit 9ebb5846e1)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
UMD's have interest in setting unused bits of the oa_ctrl register "out of
band" for certain experiments. To facilitate this, don't clobber previous
oa_ctrl unused bits, i.e. rmw the values rather than simply write them.
Fixes: e936f885f1 ("drm/xe/oa/uapi: Expose OA stream fd")
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250117032155.3048063-1-ashutosh.dixit@intel.com
(cherry picked from commit cfa9d40db8)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
[WHY]
When the system powers up eDP with external monitors in seamless boot
sequence, stutter get enabled before TTU and HUBP registers being
programmed, which resulting in underflow.
[HOW]
Enable TTU in hubp_init.
Change the sequence that do not perpare_bandwidth and optimize_bandwidth
while having seamless boot streams.
Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Lo-an Chen <lo-an.chen@amd.com>
Signed-off-by: Paul Hsieh <paul.hsieh@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[WHAT & HOW]
hpo_stream_to_link_encoder_mapping has size MAX_HPO_DP2_ENCODERS(=4),
but location can have size up to 6. As a result, it is necessary to
check location against MAX_HPO_DP2_ENCODERS.
Similiarly, disp_cfg_stream_location can be used as an array index which
should be 0..5, so the ASSERT's conditions should be less without equal.
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3904
Reviewed-by: Austin Zheng <Austin.Zheng@amd.com>
Reviewed-by: Rodrigo Siqueira <rodrigo.siqueira@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Vulkan can't support DCC and Z/S compression on GFX12 without
WRITE_COMPRESS_DISABLE in this commit or a completely different DCC
interface.
AMDGPU_TILING_GFX12_SCANOUT is added because it's already used by userspace.
Cc: stable@vger.kernel.org # 6.12.x
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
- Properly cast the input to secs_to_jiffies() to unsigned long as
otherwise the result uses the data type of the input variable, which
causes result range checks to fail if the input data type is signed and
smaller than unsigned long.
- Handle late armed hrtimers gracefully on CPU hotplug
There are legitimate cases where a hrtimer is (re)armed on an outgoing
CPU after the timers have been migrated away. This triggers warnings and
caused people to implement horrible workarounds in RCU. But those work
arounds are incomplete and do not cover e.g. the scheduler hrtimers.
Stop this by force moving timer which are enqueued on the current CPU
after timer migration to be queued on a remote online CPU.
This allows to undo the workarounds in a seperate step.
- Demote a warning level printk() to info level in the clocksource
watchdog code as there is no point to emit a warning level message for a
purely informational message.
- Mark a helper function __always_inline and move it into the existing
#ifdef block to avoid 'unused function' warnings from CLANG
-----BEGIN PGP SIGNATURE-----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=lUGt
-----END PGP SIGNATURE-----
Merge tag 'timers-urgent-2025-02-03' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull timer fixes from Thomas Gleixner:
- Properly cast the input to secs_to_jiffies() to unsigned long as
otherwise the result uses the data type of the input variable, which
causes result range checks to fail if the input data type is signed
and smaller than unsigned long.
- Handle late armed hrtimers gracefully on CPU hotplug
There are legitimate cases where a hrtimer is (re)armed on an
outgoing CPU after the timers have been migrated away. This triggers
warnings and caused people to implement horrible workarounds in RCU.
But those workarounds are incomplete and do not cover e.g. the
scheduler hrtimers.
Stop this by force moving timer which are enqueued on the current CPU
after timer migration to be queued on a remote online CPU.
This allows to undo the workarounds in a seperate step.
- Demote a warning level printk() to info level in the clocksource
watchdog code as there is no point to emit a warning level message
for a purely informational message.
- Mark a helper function __always_inline and move it into the existing
#ifdef block to avoid 'unused function' warnings from CLANG
* tag 'timers-urgent-2025-02-03' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
jiffies: Cast to unsigned long in secs_to_jiffies() conversion
clocksource: Use pr_info() for "Checking clocksource synchronization" message
hrtimers: Force migrate away hrtimers queued after CPUHP_AP_HRTIMERS_DYING
hrtimers: Mark is_migration_base() with __always_inline
- Ensure ordering of memory and device I/O for IPIs on RISCV
The RISCV interrupt controllers use writel_relaxed() for generating an
IPI. That's a device I/O write which is not guaranteed to be ordered
against preceding memory writes. As a consequence a IPI receiving CPU
might not be able to observe the actual IPI data which is required to
handle it. Switch to writel() which contains the necessary memory
barriers to enforce ordering.
- Fix up the fallout of the MSI conversion in the MVEVBU ICU driver.
The conversion failed to handle the change of the data storage and kept
the original code which uses the domain::host_data pointer
unchanged. After the conversion domain::host_data points to the new
msi_domain_info structure and not longer to the MVEBU specific MSI data,
which is now stored in a member of msi_domain_info. This leads to
malfunction of the transalate() callback.
- Only handle the PMC in FIQ mode when it is configured that way.
The original check was incorrect as it did not explicitely check for the
proper conditions, which led to malfunctions of the PMU interrupt.
- Improve Kconfig dependencies for the LAN966x Outband Interrupt
controller to avoid pointless pronmpts.
-----BEGIN PGP SIGNATURE-----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=9x00
-----END PGP SIGNATURE-----
Merge tag 'irq-urgent-2025-02-03' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull irq fixes from Thomas Gleixner:
- Ensure ordering of memory and device I/O for IPIs on RISCV
The RISCV interrupt controllers use writel_relaxed() for generating
an IPI. That's a device I/O write which is not guaranteed to be
ordered against preceding memory writes. As a consequence a IPI
receiving CPU might not be able to observe the actual IPI data which
is required to handle it. Switch to writel() which contains the
necessary memory barriers to enforce ordering.
- Fix up the fallout of the MSI conversion in the MVEVBU ICU driver.
The conversion failed to handle the change of the data storage and
kept the original code which uses the domain::host_data pointer
unchanged. After the conversion domain::host_data points to the new
msi_domain_info structure and not longer to the MVEBU specific MSI
data, which is now stored in a member of msi_domain_info. This leads
to malfunction of the transalate() callback.
- Only handle the PMC in FIQ mode when it is configured that way.
The original check was incorrect as it did not explicitely check for
the proper conditions, which led to malfunctions of the PMU
interrupt.
- Improve Kconfig dependencies for the LAN966x Outband Interrupt
controller to avoid pointless pronmpts.
* tag 'irq-urgent-2025-02-03' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
irqchip/apple-aic: Only handle PMC interrupt as FIQ when configured so
irqchip/irq-mvebu-icu: Fix access to msi_data from irq_domain::host_data
irqchip/riscv: Ensure ordering of memory writes and IPI writes
irqchip/lan966x-oic: Make CONFIG_LAN966X_OIC depend on CONFIG_MCHP_LAN966X_PCI
dt-bindings: interrupt-controller: microchip,lan966x-oic: Clarify endpoint use
Signed-off-by: Carlos Maiolino <cem@kernel.org>
-----BEGIN PGP SIGNATURE-----
iJUEABMJAB0WIQSmtYVZ/MfVMGUq1GNcsMJ8RxYuYwUCZ6C1iQAKCRBcsMJ8RxYu
Y1k+AX9qlNEvvzfcJRP7wcqjX64B0IIpuBNd86flWIYcRUug9d/vIExKvRF2A+t7
x1kc5bsBgONUBNY+xDTMAXU58MnONDIWoy9oeP/WXGaXWelN5TKlVDhHU30gXNT1
CoDNOQ/B1w==
=pduU
-----END PGP SIGNATURE-----
Merge tag 'xfs-fixes-6.14-rc2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Pull xfs bug fixes from Carlos Maiolino:
"A few fixes for XFS, but the most notable one is:
- xfs: remove xfs_buf_cache.bc_lock
which has been hit by different persons including syzbot"
* tag 'xfs-fixes-6.14-rc2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: remove xfs_buf_cache.bc_lock
xfs: Add error handling for xfs_reflink_cancel_cow_range
xfs: Propagate errors from xfs_reflink_cancel_cow_range in xfs_dax_write_iomap_end
xfs: don't call remap_verify_area with sb write protection held
xfs: remove an out of data comment in _xfs_buf_alloc
xfs: fix the entry condition of exact EOF block allocation optimization
- Connection fixes for fibre channel transport (Daniel)
- Endian fixes (Keith, Christoph)
- Cleanup fix for host memory buffer (Francis)
- Platform specific power quirks (Georg)
- Target memory leak (Sagi)
- Use appropriate controller state accessor (Daniel)
-----BEGIN PGP SIGNATURE-----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=uYd9
-----END PGP SIGNATURE-----
Merge tag 'nvme-6.14-2025-01-31' of git://git.infradead.org/nvme into block-6.14
Pull NVMe fixes from Keith:
"nvme fixes for Linux 6.14
- Connection fixes for fibre channel transport (Daniel)
- Endian fixes (Keith, Christoph)
- Cleanup fix for host memory buffer (Francis)
- Platform specific power quirks (Georg)
- Target memory leak (Sagi)
- Use appropriate controller state accessor (Daniel)"
* tag 'nvme-6.14-2025-01-31' of git://git.infradead.org/nvme:
nvme-fc: use ctrl state getter
nvme: make nvme_tls_attrs_group static
nvmet: add a missing endianess conversion in nvmet_execute_admin_connect
nvmet: the result field in nvmet_alloc_ctrl_args is little endian
nvmet: fix a memory leak in controller identify
nvme-fc: do not ignore connectivity loss during connecting
nvme: handle connectivity loss in nvme_set_queue_count
nvme-fc: go straight to connecting state when initializing
nvme-pci: Add TUXEDO IBP Gen9 to Samsung sleep quirk
nvme-pci: Add TUXEDO InfinityFlex to Samsung sleep quirk
nvme-pci: remove redundant dma frees in hmb
nvmet: fix rw control endian access
Add clock-output-names property to clock nodes, so that the resulting
clock name do not change when clock node name is changed.
Also, replace underscores with hyphens in the clock node names as per
dt-schema rule.
Signed-off-by: Naman Trivedi <naman.trivedimanojbhai@amd.com>
Acked-by: Senthil Nathan Thangaraj <senthilnathan.thangaraj@amd.com>
Link: https://lore.kernel.org/r/20241122095712.1166883-1-naman.trivedimanojbhai@amd.com
Signed-off-by: Michal Simek <michal.simek@amd.com>
