Replace ACPI specific calls to get device property with the agnostic one.
The code looses the direct dependency to the ACPI APIs and get cleaner.
This doesn't change functionality.
While at it, drop now unneeded acpi.h.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://patch.msgid.link/20251129151739.3998668-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Replace CR_TCIE and CR_TEIE irq usage by a read_poll_timeout_atomic() in
stm32_qspi_wait_cmd(). It will reduce the time waiting for TCF or TEF flags
to optimize throughput.
before after
average time spent in stm32_qspi_wait_cmd: 2615 ns 712 ns
Signed-off-by: Patrice Chotard <patrice.chotard@foss.st.com>
Link: https://patch.msgid.link/20251208-upstream_qspi_ospi_updates-v2-7-62526c9467dc@foss.st.com
Signed-off-by: Mark Brown <broonie@kernel.org>
FIFO accesses uses u8 only for read/write.
In order to optimize throughput, add u16 or u32 read/write
accesses when possible.
Running mtd_speedtest on a 4MB sNOR partition using a
stm32mp257f-ev1 board gives the following results:
before after gain
Read : 5773 KiB/s 22170 KiB/s 384%
Write: 796 KiB/s 890 KiB/s 12%
Signed-off-by: Patrice Chotard <patrice.chotard@foss.st.com>
Link: https://patch.msgid.link/20251208-upstream_qspi_ospi_updates-v2-6-62526c9467dc@foss.st.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Replace CR_TCIE and CR_TEIE irq usage by a read_poll_timeout_atomic() in
stm32_ospi_wait_cmd(). It will reduce the time waiting for TCF or TEF flags
to optimize throughput.
before after
average time spent in stm32_omi_wait_cmd: 5685 ns 923 ns
Signed-off-by: Patrice Chotard <patrice.chotard@foss.st.com>
Link: https://patch.msgid.link/20251208-upstream_qspi_ospi_updates-v2-3-62526c9467dc@foss.st.com
Signed-off-by: Mark Brown <broonie@kernel.org>
FIFO accesses uses u8 only for read/write.
In order to optimize throughput, add u16 or u32 read/write
accesses when possible.
Running mtd_speedtest on a 4MB sNOR partition using a
stm32mp257f-ev1 board gives the following results:
before after gain
Read : 5693 KiB/s 21139 KiB/s 371%
Write: 765 KiB/s 910 KiB/s 19%
Signed-off-by: Patrice Chotard <patrice.chotard@foss.st.com>
Link: https://patch.msgid.link/20251208-upstream_qspi_ospi_updates-v2-2-62526c9467dc@foss.st.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Enable DMA mode for SPI IMX in target mode. Disable the word delay feature
for target mode, because target mode should always keep high performance
to make sure it can follow the master. Target mode continues to operate in
dynamic burst mode.
Signed-off-by: Carlos Song <carlos.song@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20251203085949.2922166-7-carlos.song@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
ECSPI transfers only one word per frame in DMA mode, causing SCLK stalls
between words due to BURST_LENGTH updates, which significantly impacts
performance.
To improve throughput, configure BURST_LENGTH as large as possible (up to
512 bytes per frame) instead of word length. This avoids delays between
words. When transfer length is not 4-byte aligned, use bounce buffers to
align data for DMA. TX uses aligned words for TXFIFO, while RX trims DMA
buffer data after transfer completion.
Introduce a new dma_package structure to store:
1. BURST_LENGTH values for each DMA request
2. Variables for DMA submission
3. DMA transmission length and actual data length
Handle three cases:
- len <= 512 bytes: one package, BURST_LENGTH = len * 8 - 1
- len > 512 and aligned: one package, BURST_LENGTH = max (512 bytes)
- len > 512 and unaligned: two packages, second for tail data
Performance test (spidev_test @10MHz, 4KB):
Before: tx/rx ~6651.9 kbps
After: tx/rx ~9922.2 kbps (~50% improvement)
For compatibility with slow SPI devices, add configurable word delay in
DMA mode. When word delay is set, dynamic burst is disabled and
BURST_LENGTH equals word length.
Signed-off-by: Carlos Song <carlos.song@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20251203085949.2922166-6-carlos.song@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Add error handling for DMA request submission by checking the return
cookie with dma_submit_error(). This prevents propagating submission
failures through the DMA transfer process, which could lead to
unexpected behavior.
Signed-off-by: Carlos Song <carlos.song@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20251203085949.2922166-5-carlos.song@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
If dmaengine_prep_slave_sg() fails, no descriptor is submitted to the TX
channel and DMA is never started. Therefore, calling
dmaengine_terminate_all() for the TX DMA channel is unnecessary.
Signed-off-by: Carlos Song <carlos.song@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20251203085949.2922166-4-carlos.song@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Add a helper function to handle clearing DMA mode, including getting the
maximum watermark length and submitting the DMA request. This refactoring
makes the code more concise and improves readability.
No functional changes.
Signed-off-by: Carlos Song <carlos.song@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20251203085949.2922166-3-carlos.song@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Relocate spi_imx_dma_configure() next to spi_imx_dma_transfer() so that
all DMA-related functions are grouped together for better readability.
No functional changes.
Signed-off-by: Carlos Song <carlos.song@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20251203085949.2922166-2-carlos.song@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The DMA controller can be used to transfer data to and from the SPI
controller without involving the CPU for each word of a SPI transfer.
Add support for DMA mode.
Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com>
Link: https://patch.msgid.link/20251201134229.600817-12-cosmin-gabriel.tanislav.xa@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
In preparation for implementing DMA support, enable the transmit buffer
empty interrupt, which is necessary for DMA to write more data to the
FIFO.
This does not affect the PIO mode as we do not even request the TX
interrupt line.
Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com>
Link: https://patch.msgid.link/20251201134229.600817-9-cosmin-gabriel.tanislav.xa@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
In PIO mode we send data word-by-word, and wait for the received data
to be available after each sent word, making no use of the TX interrupt.
In DMA mode, we need to set the RX and TX FIFO thresholds to 0, as
described in the User Manual.
In preparation for implementing DMA support, set TX FIFO threshold to 0,
as RX FIFO threshold is already 0.
Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com>
Link: https://patch.msgid.link/20251201134229.600817-8-cosmin-gabriel.tanislav.xa@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
In preparation for implementing DMA support, set MUST_RX and MUST_TX
flags on the controller so that we always receive non-NULL buffers.
The PIO mode already handles this manually by checking if rx_buf/tx_buf
are set on the transfer, and doing a dummy read/write if not.
DMA will not be able to implement this special handling, and although
the SPI controller advertises support for transmit-only or receive-only
transfers via SPCR's register TXMD bitfield, it does not seem to work.
Remove the special handling for PIO and let the SPI controller core
handle it.
Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com>
Link: https://patch.msgid.link/20251201134229.600817-7-cosmin-gabriel.tanislav.xa@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Non-device-managed APIs were initially used here to avoid the buggy
interaction between PM domains and device-managed actions.
Commit f99508074e ("PM: domains: Detach on device_unbind_cleanup()")
fixed the interaction between PM domains and device-managed actions.
Simplify the code by using device-managed actions to unregister the SPI
controller and to assert and release the resets.
Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com>
Link: https://patch.msgid.link/20251201134229.600817-5-cosmin-gabriel.tanislav.xa@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Setting SPI_TRANS_FAIL_IO has no effect if the transfer completes in the
current context, as it is only handled when .transfer_one() returns > 0,
when the SPI core must wait for the SPI transfer to complete.
Do not set SPI_TRANS_FAIL_IO as we either return an error or 0, since we
do our own waiting.
Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com>
Link: https://patch.msgid.link/20251201134229.600817-4-cosmin-gabriel.tanislav.xa@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
A call to spi_finalize_current_transfer() is only needed when the SPI
transfer is completed outside of the current context, when the
.transfer_one() implementation returns > 0.
Since the SPI transfer is completed in the current context, and we
return 0 from .transfer_one(), the SPI core assumes that the transfer
has completed and it does not wait for the completion variable that
would be set by a call to spi_finalize_current_transfer().
Remove the call to spi_finalize_current_transfer().
Signed-off-by: Cosmin Tanislav <cosmin-gabriel.tanislav.xa@renesas.com>
Link: https://patch.msgid.link/20251201134229.600817-3-cosmin-gabriel.tanislav.xa@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Add support for the Maxim Integrated MAX77675 PMIC regulator.
The MAX77675 is a compact, highly efficient SIMO (Single Inductor Multiple Output)
power management IC that provides four programmable buck-boost switching regulators
with only one inductor. It supports up to 700mA total output current and operates
from a single-cell Li-ion battery.
An integrated power-up sequencer and I2C interface allow flexible startup
configuration and runtime control.
Signed-off-by: Joan Na <joan.na@analog.com>
Link: https://patch.msgid.link/20251207032907.4850-3-joan.na@analog.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Add device tree binding YAML schema for the Maxim MAX77675 PMIC regulator.
Signed-off-by: Joan Na <joan.na@analog.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://patch.msgid.link/20251207032907.4850-2-joan.na@analog.com
Signed-off-by: Mark Brown <broonie@kernel.org>
A remove callback is optional and having no such function has the same
semantic as one returning zero (and other return values are effectively
ignored).
This allows to remove the remove function without replacement.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@baylibre.com>
Link: https://patch.msgid.link/20251212073555.1065284-2-u.kleine-koenig@baylibre.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Replace the __free(kfree) in cs_dsp_load() and cs_dsp_load_coeff() with
a kfree(buf) at the end of the function.
The use of __free() can create new cleanup bugs that are difficult to spot
because the defective code is idiomatically correct regular C. In these two
functions the __free() was mixed with gotos, and also used the suspect
declaration __free(kfree) = NULL;.
The __free() did not do anything to simplify the code. There aren't any
early returns after the pointer is set, and the __free() can be replaced by
a kfree() at the end of the function.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Fixes: 900baa6e7b ("firmware: cs_dsp: Remove redundant download buffer allocator")
Link: https://patch.msgid.link/20251201160729.231867-1-rf@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>
A recent change switched to using guards for the port list lock but only
modified two out of three functions where the lock is held.
Convert also the third function for consistency while switching to a
scoped guard in q6afe_port_get_from_id() for clarity.
Signed-off-by: Johan Hovold <johan@kernel.org>
Link: https://patch.msgid.link/20251203105542.24765-3-johan@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>
The enablement of the CONFIG_OF_GPIO switch has nothing to do with the
"gpio-controller" property which may as well come from software nodes
and GPIOLIB can still be enabled separately.
This driver does not call any symbols from gpiolib-of.h so has no need
to check this option at all. Just use the generic device property
accessor.
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
Link: https://patch.msgid.link/20251205090534.27845-1-bartosz.golaszewski@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Replace snprintf() with scnprintf() when constructing the property
and remove negative return error handling as scnprintf() returns the
actual number of bytes written to buffer.
snprintf() as defined by the C99 standard,returns the number of
characters that *would have been* written if enough space were
available.Use scnprintf() that returns the actual number of
characters written.
Link: https://github.com/KSPP/linux/issues/105
Signed-off-by: HariKrishna Sagala <hariconscious@gmail.com>
Link: https://patch.msgid.link/20251212044408.1286-2-hariconscious@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Don't use __free(kfree) in cs_dsp_debugfs_string_read. Instead use
normal kfree() to cleanup.
The use of __free() can create new cleanup bugs that are difficult to spot
because the defective code is idiomatically correct regular C. This
function used the suspect declaration __free(kfree) = NULL;.
The __free(kfree) didn't really do anything here. The function can be
rearranged to avoid any need to return or goto within the code.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Fixes: 3045e29d24 ("firmware: cs_dsp: Append \n to debugfs string during read")
Link: https://patch.msgid.link/20251202113425.413700-1-rf@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Replace manual "request_threaded_irq()" with the device managed
"devm_request_threaded_irq" to manage the IRQ lifetime and also
it removes the smatch reported warning.
Remove the manual "free_irq()" in the "remove" function as free_irq
is tied to device teardown.
Signed-off-by: HariKrishna Sagala <hariconscious@gmail.com>
Link: https://patch.msgid.link/20251121140940.40678-4-hariconscious@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
State files to load explicitly in the device descriptor instead of
hiding the details within a loading function. Apart from readability,
this also reduces the catpt module size slightly.
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://patch.msgid.link/20251212103858.110701-5-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Make it easier for functions that call IPC handlers to deal with their
results by accounting for '0' (success) code. Rename the macro to
reflect this behaviour change.
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://patch.msgid.link/20251212103858.110701-3-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Code size can be reduced if catpt_dsp_do_send_msg() takes responsibility
for dumping logs in case of an IPC message failure.
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://patch.msgid.link/20251212103858.110701-2-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Initializing automatic __free variables to NULL without need (e.g.
branches with different allocations), followed by actual allocation is
in contrary to explicit coding rules guiding cleanup.h:
"Given that the "__free(...) = NULL" pattern for variables defined at
the top of the function poses this potential interdependency problem the
recommendation is to always define and assign variables in one statement
and not group variable definitions at the top of the function when
__free() is used."
Code does not have a bug, but is less readable and uses discouraged
coding practice, so fix that by moving declaration to the place of
assignment.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://patch.msgid.link/20251203-asoc-wrong-cleanup-h-continued-v1-3-5142be4874fb@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Initializing automatic __free variables to NULL without need (e.g.
branches with different allocations), followed by actual allocation is
in contrary to explicit coding rules guiding cleanup.h:
"Given that the "__free(...) = NULL" pattern for variables defined at
the top of the function poses this potential interdependency problem the
recommendation is to always define and assign variables in one statement
and not group variable definitions at the top of the function when
__free() is used."
Code does not have a bug, but is less readable and uses discouraged
coding practice, so fix that by moving declaration to the place of
assignment.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Link: https://patch.msgid.link/20251203-asoc-wrong-cleanup-h-continued-v1-2-5142be4874fb@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Initializing automatic __free variables to NULL without need (e.g.
branches with different allocations), followed by actual allocation is
in contrary to explicit coding rules guiding cleanup.h:
"Given that the "__free(...) = NULL" pattern for variables defined at
the top of the function poses this potential interdependency problem the
recommendation is to always define and assign variables in one statement
and not group variable definitions at the top of the function when
__free() is used."
Code does not have a bug, but is less readable and uses discouraged
coding practice, so fix that by moving declaration to the place of
assignment.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Link: https://patch.msgid.link/20251203-asoc-wrong-cleanup-h-continued-v1-1-5142be4874fb@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Variables with automatic cleanup are special because they do not follow
standard rules of declaration at top of function (see cleanup.h), but on
the other hand we always expect line break between top-function
declarations and first instructions.
Don't pretend automatic cleanup variables are part of top-level
declaration to improve readability when variable is followed by nun-NULL
check. No functional impact, only style.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
Link: https://patch.msgid.link/20251129-asoc-wrong-cleanup-h-can-people-stop-sending-this-without-reading-docs-v1-6-c38b06884e39@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Commit 88a5f8e628 ("ASoc: qcom: audioreach: Use automatic cleanup of
kfree()") did not make the code simpler but more complicated. Already
simple code of allocation and free, without any error paths, got now
declaration with one constructor followed by another allocation, which
is in contrary to explicit coding rules guiding cleanup.h:
"Given that the "__free(...) = NULL" pattern for variables defined at
the top of the function poses this potential interdependency problem the
recommendation is to always define and assign variables in one statement
and not group variable definitions at the top of the function when
__free() is used."
Code does not have a bug, but is less readable and uses discouraged
coding practice, so fix that by moving declaration to the place of
assignment.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
Link: https://patch.msgid.link/20251129-asoc-wrong-cleanup-h-can-people-stop-sending-this-without-reading-docs-v1-5-c38b06884e39@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Commit 55094e55ae ("ASoc: qcom: q6afe: Use automatic cleanup of
kfree()") did not make the code simpler but more complicated. Already
simple code of allocation and free, without any error paths, got now
declaration with one constructor followed by another allocation, which
is in contrary to explicit coding rules guiding cleanup.h:
"Given that the "__free(...) = NULL" pattern for variables defined at
the top of the function poses this potential interdependency problem the
recommendation is to always define and assign variables in one statement
and not group variable definitions at the top of the function when
__free() is used."
Code does not have a bug, but is less readable and uses discouraged
coding practice, so fix that by moving declaration to the place of
assignment.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
Link: https://patch.msgid.link/20251129-asoc-wrong-cleanup-h-can-people-stop-sending-this-without-reading-docs-v1-4-c38b06884e39@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
