This CL attempts to implement npcx's flash driver instead of the
original one (npcx spi driver plus spi_nor flash driver).
Signed-off-by: Mulin Chao <mlchao@nuvoton.com>
The MCUX DMA controller only supports a single data_size
for a DMA transfer, not separate ones for source and
dest. An older version of the DMA driver used
dest_data_size as the DMA transfer size, but the
current one uses MIN(dest/source) as the trasnfer
size, which breaks case when SPI wants to do 2-byte
transfers.
Signed-off-by: Mike J. Chen <mjchen@google.com>
The `spi_mcux_transceive` had 2 return calls when the
`CONFIG_SPI_MCUX_LPSPI_DMA` flag was active. The first return would be
called and the later was unreachable. With the fix, now the return calls
are mutually exclusive. Also, the `transceive` call is not compiled with
the `CONFIG_SPI_MCUX_LPSPI_DMA` flag is active.
Fixes#59533
Signed-off-by: Dipak Shetty <dipak.shetty@zeiss.com>
The eSPI, I2C, and SPI emulators use devicetree macros to build an array
of devices on the virtual bus. Currently, they will add device nodes that
are not status-okay. This leads to linker errors because the respective
device drivers would not have instantiated device structs for these
nodes --assuming the driver was even compiled. This can be frustrating
if nodes need to be disabled for debugging or configuration purposes.
Update the bus emulators to only consider status-okay nodes by changing
the macros used to iterate over bus devices.
Signed-off-by: Tristan Honscheid <honscheid@google.com>
Fix for bug:
https://github.com/zephyrproject-rtos/zephyr/issues/59575
The dma version of the version of the driver can
invoke multiple intermediate dma transfers, like
when the spi_buf_set count is greater than one.
However, there is a bug where chip select is not kept
asserted for all intermediate dma transfers required
to process the entire spi_buf_set.
Signed-off-by: Mike J. Chen <mjchen@google.com>
Replace the timeout for a SPI transceive in slave mode for STM32 DMA
operations with a K_FOREVER. Being an SPI slave means we do not
know when the transaction will start, hence it does not make sense
to have a timeout in such a case.
This will resolve issue #60000.
Signed-off-by: Benedikt Schmidt <benedikt.schmidt@embedded-solutions.at>
The `SPI_EMUL_INIT(n)` macro used to instantiate SPI bus emulators
builds an array of SPI devices attached to the bus, but always checks
instance #0 of the bus and not `n`. This makes multiple instances of an
emulated SPI bus unusable and popualtes them all with bus 0's devices.
Change the `0` to `n`, which matches how the I2C bus emulator does it.
Signed-off-by: Tristan Honscheid <honscheid@google.com>
This adds a few line use zephyr_syscall_header() to include
headers containing syscall function prototypes.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Found EMSDP board SPI-FLASH sample broke after adding DFSS
into spi_dw. Found wrong interrput mask resulting in false
interrupt enabled. Now fixed it to fit both DFSS and DW.
Signed-off-by: Siyuan Cheng <siyuanc@synopsys.com>
Introduce DesignWare ARC Data Fusion IP Subsystem(DFSS) SPI
driver for ARC boards, i.e. EMSDP, which uses DW SPI to controll
SPI-Flash and DFSS SPI to connect external devices. Both drivers
share most source code, but DFSS uses ARC auxiliary registers.
Move FIFO depth setting to device tree.
Signed-off-by: Siyuan Cheng <siyuanc@synopsys.com>
Commit 246393e830
("drivers: spi: spi_nrfx_spim: Remove nrf_frequency_t handling")'
introduced two changes, one of them is removing the function
get_nrf_spim_frequency with a strange justification.
This change breaks support for peripherals written in a common way,
where the maximum frequency is set to the maximum supported
by the peripheral, not the controller, see shields for example.
On the occasion of bringing it back, the original function was
refactored to be easier to read and understand.
Signed-off-by: Johann Fischer <johann.fischer@nordicsemi.no>
Signed-off-by: Andrzej Głąbek <andrzej.glabek@nordicsemi.no>
Add driver for OpenTitan SPI host.
Tested and verified single SPI full duplex on opentitan_earlgrey in
Verilator.
Signed-off-by: Shawn Nematbakhsh <shawn@rivosinc.com>
Rather than the rings, which weren't shared between userspace and kernel
space in Zephyr like they are in Linux with io_uring, use atomic mpsc
queues for submission and completion queues.
Most importantly this removes a potential head of line blocker in the
submission queue as the sqe would be held until a task is completed.
As additional bonuses this avoids some additional locks and restrictions
about what can be submitted and where. It also removes the need for
two executors as all chains/transactions are done concurrently.
Lastly this opens up the possibility for a common pool of sqe's to
allocate from potentially saving lots of memory.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Added device tree bindings and implementaion for setting the
spi controllers chip select setup and hold time settings.
Signed-off-by: Dean Sellers <dsellers@evos.com.au>
The nrfx SPI driver depends on semaphores, which require multithreading
support to be enabled.
Signed-off-by: Eivind Jølsgard <eivind.jolsgard@nordicsemi.no>
Make use of positive status values in the DMA callback to pass
info to the DMA client after a successful DMA operation.
A completed DMA transfer uses the status 0 while a reached
water mark uses the status 1.
Signed-off-by: Cyril Fougeray <cyril.fougeray@worldcoin.org>
Word size setting limited any SPI transaction to the frame
size. In addition to making the SPI inefficient this
broke drivers that set the word size. It appears that most
drivers use a one byte (8) size for this setting.
This change respects what I think is the intended use of
the word size setting. That is to set the length of
each element in a tx/rx buffer struct.
Signed-off-by: Dean Sellers <dsellers@evos.com.au>
With new nrfx release SPIM driver stores frequency as a `uint32_t`
type representing frequency in `Hz` in its configuration structure.
Additionally `NRFX_SPIM_PIN_NOT_USED` has been removed,
`NRF_SPIM_PIN_NOT_CONNECTED` symbol is used instead.
Signed-off-by: Adam Wojasinski <adam.wojasinski@nordicsemi.no>
When dma is enabled, spi_xmc4xxx_transcieve_dma() needs to disable the
isr used by spi_xmc4xxx_transceive(). Renable the isr at the
end of spi_xmc4xxx_transceive_dma() instead of in spi_xmc4xxx_transceive().
Doing this in latter function will fail compilation when interrupt
support is enabled but dma is disabled.
This is regression from 8494b6413a
but was only caught when xmc47_relax_kit was added which tested this
scenario in tests/drivers/spi/spi_loopback.
Fixes#57494
Signed-off-by: Andriy Gelman <andriy.gelman@gmail.com>
When using gpio chip select the clock line seems to get stuck low after
some transactions. When attempting to use other SPI_CSR registers
the peripheral fails to work as expected.
Always using SPI_CSR[0] when using gpio chip selects resolves the issue.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
spi_context_get_next_buf must not divide `len` by `dfs` because, in SPI,
buffer lengths are given in units of data (in this case, 16 bits), not in
bytes.
Signed-off-by: Daniel Gaston Ochoa <dgastonochoa@gmail.com>
Following #56576, the `cs` field in `struct spi_config` is of type
`struct spi_cs_control` instead of a pointer to the same type.
`spi_xec_qmspi_ldma.c:qmspi_xec_init` tries assigning `NULL` to the
`.cs` field through a designated initializer, which causes a compilation
error.
This PR simply removes the `.cs = NULL` line. The designated initializer
will automatically zeroize the underlying GPIO pin info, which should
have the same effect that setting the pointer to NULL did previously.
Signed-off-by: Tristan Honscheid <honscheid@google.com>
Add spi_cs_is_gpio(_dt) helpers to check whether SPI CS is controlled by
GPIO or not. This both improves code readability and isolates SPI
internals.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
As of today it is not possible to use SPI dt-spec macros in C++,
something known and documented. The main reason is because `cs` property
is initialized using a compound literal, something not supported in C++.
This PR takes another approach, that is to not make `cs` a pointer but a
struct member. This way, we can perform a regular initialization, at the
cost of using extra memory for unused delay/pin/flags if `cs` is not
used.
Fixes#56572
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The rvm32m1 platform always uses pinctrl, there's no need to keep
extra macrology around pinctrl. Also updated driver's Kconfig to `select
PINCTRL`.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The MCUX platform always uses pinctrl, there's no need to keep extra
macrology around pinctrl. Also updated driver's Kconfig options to
`select PINCTRL` (note that some already did).
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
When using a dummy buffer on the RX side, do not
increment the destination memory buffer address.
This issue was uncovered when running the SPI
loopback test.
Signed-off-by: Mahesh Mahadevan <mahesh.mahadevan@nxp.com>
The spi-sam0 driver does not initialize DMA parameters when the MCLK
peripheral is defined in the microcontroller header file. Fix it by
copying the initialization code from the non-MCLK case.
Signed-off-by: Brandon Del Bel <delbel@umn.edu>
This adds the SPI driver for the Renesas SmartBond(tm) DA1469x MCU family.
The driver only supports controller mode. All four SPI modes are supported.
Note that the lowest supported speed is 2285714Hz.
Requesting speeds higher than 16MHz, will result in a 16MHz SCLK.
Co-authored-by: Stan Geitel <stan@geitel.nl>
Signed-off-by: Ben Lauret <ben.lauret.wm@renesas.com>
Remove all init functions that do nothing, and provide a `NULL` to
*DEVICE*DEFINE* macros.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
Zephyr SPI driver model for full-duplex operation assumes
data will be transmitted and received during each clock period.
The QMSPI driver for the XEC family also supported dual and
quad I/O use cases which are inherently half-duplex. To
support dual/quad the driver incorrectly processed spi buffers
as all transmit buffers first then all receive buffers. This
worked if only the SPI driver was used. It did not work with
the Zephyr flash SPI NOR driver which assumes SPI drivers
follow the SPI driver model. This commit implements a QMSPI
driver that follows the Zephyr SPI driver model resulting in
a slightly smaller driver. Dual/quad SPI transactions are
supported if the experimental SPI extended mode Zephyr
configuration flag is enabled. We also remove the QMSPI full duplex
driver added previously to support the flash SPI NOR driver.
Added board to spi loop-back test and spi_flash sample.
Signed-off-by: Scott Worley <scott.worley@microchip.com>
There have an extral TBE interrupt event though we have cleared the
SPI_CTL1_TBEIE bit. To cover this situation, add a on_going check
before enter exchange function.
Signed-off-by: HaiLong Yang <hailong.yang@brainco.cn>
Implements the SPI RTIO API. Refactors many of the internal transfer
functions to work on simple types to allow for both the RTIO and existing
SPI API functions to co-exist.
Removes the unneeded shift_master specialization. Either a polling or DMA
transfer flow is used when used with RTIO as all operations are normalized.
When SPI_RTIO is enabled the spi_transceive blocking call translates
the request into an RTIO transaction placed in the queue of transactions
the device is meant to do.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
Provides a macro and submit API for SPI drivers to support RTIO.
A copy function enables compatibility with the existing blocking API
and very easily the existing async API as well.
Signed-off-by: Tom Burdick <thomas.burdick@intel.com>
This commit fixes the word size configuration issue
described in #54746 by considering the data frame size
when trasmitting in case the configuration is applied.
It also fixes an heap corruption problem when using
SPI DMA with a buffer that is not multiple of 32 bits
in lenght and GDMA instance in initialization.
Signed-off-by: Lucas Tamborrino <lucas.tamborrino@espressif.com>
Unify the drivers/*/Kconfig menuconfig title strings to the format
"<class> [(acronym)] [bus] drivers".
Including both the full name of the driver class and an acronym makes
menuconfig more user friendly as some of the acronyms are less well-known
than others. It also improves Kconfig search, both via menuconfig and via
the generated Kconfig documentation.
Signed-off-by: Henrik Brix Andersen <hebad@vestas.com>
The number of available EasyDMA MAXCNT bits is now defined per-instance
in Devicetree.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
A MACRO argument naming mismatch causes a bug when trying to use DMA.
Fix the MACRO argument and conditional DMA configuration.
Signed-off-by: Pieter De Gendt <pieter.degendt@basalte.be>
Uninitialized sam spinlocks were causing an error if the stack
happens to have 'thread_cpu' memory set to something other than 0.
Signed-off-by: Yuval Peress <peress@google.com>
