stm32_spi_remove() accesses the driver's private data after calling
spi_unregister_master() even though that function releases the last
reference on the spi_master and thereby frees the private data.

Fix by switching over to the new devm_spi_alloc_master() helper which
keeps the private data accessible until the driver has unbound.

Fixes: 8d559a64 ("spi: stm32: drop devres version of spi_register_master")
Reported-by: NLukas Wunner <lukas@wunner.de>
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1616052290-10887-1-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

Avoid CONFIG_PM preprocessor check for pm suspend/resume
callbacks and identify the functions with __maybe_unused.
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1615545329-5496-1-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

A call to spi_unregister_master() triggers calling remove()
for all the spi devices binded to the spi master.

Some spi device driver requires to "talk" with the spi device
during the remove(), e.g.:
- a LCD panel like drivers/gpu/drm/panel/panel-lg-lg4573.c
  will turn off the backlighting sending a command over spi.
This implies that the spi master must be fully functional when
spi_unregister_master() is called, either if it is called
explicitly in the master's remove() code or implicitly by the
devres framework.

Devres calls devres_release_all() to release all the resources
"after" the remove() of the spi master driver (check code of
__device_release_driver() in drivers/base/dd.c).
If the spi master driver has an empty remove() then there would
be no issue; the devres_release_all() will release everything
in reverse order w.r.t. probe().
But if code in spi master driver remove() disables the spi or
makes it not functional (like in this spi-stm32), then devres
cannot be used safely for unregistering the spi master and the
binded spi devices.

Replace devm_spi_register_master() with spi_register_master()
and add spi_unregister_master() as first action in remove().

Fixes: dcbe0d84 ("spi: add driver for STM32 SPI controller")
Signed-off-by: NAntonio Borneo <antonio.borneo@foss.st.com>
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1615545286-5395-1-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

We do not expect to receive spurious interrupts so rise a warning
if it happens.

RX overrun is an error condition that signals a corrupted RX
stream both in dma and in irq modes. Report the error and
abort the transfer in either cases.
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1612551572-495-9-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

Replace SPI_1HZ_NS private constant with NSEC_PER_SEC, which is easier
to read and understand.
Signed-off-by: NAmelie Delaunay <amelie.delaunay@foss.st.com>
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1612551572-495-8-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

Defer the probe operation when a reset controller device is expected
but have not yet been probed.

This change replaces use of devm_reset_control_get_exclusive() with
devm_reset_control_get_optional_exclusive() as reset controller is
optional which is now explicitly stated.
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1612551572-495-7-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

Remove reset controller device reference from the device private
structure since it is used only at probe time and can be discarded
once used to reset the SPI device.
Signed-off-by: NEtienne Carriere <etienne.carriere@st.com>
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1612551572-495-6-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

Ensure that messages given to transfer_one handler can actually be
handled by it. For that purpose rely on the SPI framework
spi_split_transfers_maxsize function to split messages whenever necessary.
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1612551572-495-5-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

To avoid defining shift and mask separately and hand-coding the bit
manipulation, use the bitfield macros.
Signed-off-by: NAmelie Delaunay <amelie.delaunay@foss.st.com>
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1612551572-495-4-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

CS gpios is not mandatory, the driver should allow working
even when CS are not given.
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1612551572-495-3-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

On 0 byte transfer request, return straight from the
xfer function after finalizing the transfer.

Fixes: dcbe0d84 ("spi: add driver for STM32 SPI controller")
Signed-off-by: NAlain Volmat <alain.volmat@foss.st.com>
Link: https://lore.kernel.org/r/1612551572-495-2-git-send-email-alain.volmat@foss.st.comSigned-off-by: NMark Brown <broonie@kernel.org>

Simplify stm32h7_spi_prepare_fthlv() function implementation,
no functional change intended.
Signed-off-by: NMarek Vasut <marex@denx.de>
Cc: Alain Volmat <alain.volmat@st.com>
Cc: Alexandre Torgue <alexandre.torgue@st.com>
Cc: Amelie Delaunay <amelie.delaunay@st.com>
Cc: Antonio Borneo <antonio.borneo@st.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com>
Cc: Roman Guskov <rguskov@dh-electronics.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-stm32@st-md-mailman.stormreply.com
To: linux-spi@vger.kernel.org
Link: https://lore.kernel.org/r/20210104123114.261596-1-marex@denx.deSigned-off-by: NMark Brown <broonie@kernel.org>

With the introduction of the 'include/uapi/linux/spi/spi.h' header, the
type of the macros are enforced to 'unsigned long int' via the _BITUL()
macro.

This causes some -Wformat warnings in the spi-stm32 driver.
This patch adds a double-negation operator to the bit-masks. Essentially,
the important values for debugging are 0 or 1, while masking them directly
would show 0 or BIT(x) values.
This way, the type of the arguments are automatically re-cast.

Fixes: f7005142 ("spi: uapi: unify SPI modes into a single spi.h header")
Reported-by: Nkernel test robot <lkp@intel.com>
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Cc: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NAlexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20210104143103.56510-1-alexandru.ardelean@analog.comSigned-off-by: NMark Brown <broonie@kernel.org>

if cur_bpw <= 8 and xfer_len < 4 then the value of fthlv will be 1 and
SPI registers content may have been lost.

* If SPI data register is accessed as a 16-bit register and DSIZE <= 8bit,
  better to select FTHLV = 2, 4, 6 etc

* If SPI data register is accessed as a 32-bit register and DSIZE > 8bit,
  better to select FTHLV = 2, 4, 6 etc, while if DSIZE <= 8bit,
  better to select FTHLV = 4, 8, 12 etc
Signed-off-by: NRoman Guskov <rguskov@dh-electronics.com>
Fixes: dcbe0d84 ("spi: add driver for STM32 SPI controller")
Reviewed-by: NMarek Vasut <marex@denx.de>
Link: https://lore.kernel.org/r/20201221123532.27272-1-rguskov@dh-electronics.comSigned-off-by: NMark Brown <broonie@kernel.org>

pm_runtime_get_sync will increment pm usage counter even it
failed. Forgetting to pm_runtime_put_noidle will result in
reference leak in stm32_spi_resume, so we should fix it.

Fixes: db96bf97 ("spi: stm32: fixes suspend/resume management")
Signed-off-by: NZhang Qilong <zhangqilong3@huawei.com>
Reviewed-by: NAlain Volmat <alain.volmat@st.com>
Link: https://lore.kernel.org/r/20201106015217.140476-1-zhangqilong3@huawei.comSigned-off-by: NMark Brown <broonie@kernel.org>

Running in hardIRQ, disabling IRQ is redundant.
Signed-off-by: NBarry Song <song.bao.hua@hisilicon.com>
Link: https://lore.kernel.org/r/20200926001616.21292-2-song.bao.hua@hisilicon.comSigned-off-by: NMark Brown <broonie@kernel.org>

The pm_runtime_get_sync() can return either 0 or 1 on success but this
code treats 1 as a failure.

Fixes: db96bf97 ("spi: stm32: fixes suspend/resume management")
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: NAlain Volmat <alain.volmat@st.com>
Link: https://lore.kernel.org/r/20200909094304.GA420136@mwandaSigned-off-by: NMark Brown <broonie@kernel.org>

Common pattern of handling deferred probe can be simplified with
dev_err_probe().  Less code and the error value gets printed.
Signed-off-by: NKrzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20200901152713.18629-8-krzk@kernel.orgSigned-off-by: NMark Brown <broonie@kernel.org>

The 'spi_stm32 44004000.spi: Communication suspended' message means that
when using PIO, the kernel did not read the FIFO fast enough and so the
SPI controller paused the transfer. Currently, this is printed on every
single such event, so if the kernel is busy and the controller is pausing
the transfers often, the kernel will be all the more busy scrolling this
message into the log buffer every few milliseconds. That is not helpful.

Instead, rate-limit the message and print it every once in a while. It is
not possible to use the default dev_warn_ratelimited(), because that is
still too verbose, as it prints 10 lines (DEFAULT_RATELIMIT_BURST) every
5 seconds (DEFAULT_RATELIMIT_INTERVAL). The policy here is to print 1 line
every 50 seconds (DEFAULT_RATELIMIT_INTERVAL * 10), because 1 line is more
than enough and the cycles saved on printing are better left to the CPU to
handle the SPI. However, dev_warn_once() is also not useful, as the user
should be aware that this condition is possibly recurring or ongoing. Thus
the custom rate-limit policy.

Finally, turn the message from dev_warn() to dev_dbg(), since the system
does not suffer any sort of malfunction if this message appears, it is
just slowing down. This further reduces the printing into the log buffer
and frees the CPU to do useful work.

Fixes: dcbe0d84 ("spi: add driver for STM32 SPI controller")
Signed-off-by: NMarek Vasut <marex@denx.de>
Cc: Alexandre Torgue <alexandre.torgue@st.com>
Cc: Amelie Delaunay <amelie.delaunay@st.com>
Cc: Antonio Borneo <borneo.antonio@gmail.com>
Cc: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20200905151913.117775-1-marex@denx.deSigned-off-by: NMark Brown <broonie@kernel.org>

SPI registers content may have been lost upon suspend/resume sequence.
So, always compute and apply the necessary configuration in
stm32_spi_transfer_one_setup routine.
Signed-off-by: NAlain Volmat <alain.volmat@st.com>
Link: https://lore.kernel.org/r/1597043558-29668-6-git-send-email-alain.volmat@st.comSigned-off-by: NMark Brown <broonie@kernel.org>

This patch adds pinctrl power management, and reconfigure spi controller
in case of resume.

Fixes: 038ac869 ("spi: stm32: add runtime PM support")
Signed-off-by: NAmelie Delaunay <amelie.delaunay@st.com>
Signed-off-by: NAlain Volmat <alain.volmat@st.com>
Link: https://lore.kernel.org/r/1597043558-29668-5-git-send-email-alain.volmat@st.comSigned-off-by: NMark Brown <broonie@kernel.org>

Fix spi->clk_rate when it is odd to the nearest lowest even value because
minimum SPI divider is 2.
Signed-off-by: NAmelie Delaunay <amelie.delaunay@st.com>
Signed-off-by: NAlain Volmat <alain.volmat@st.com>
Link: https://lore.kernel.org/r/1597043558-29668-4-git-send-email-alain.volmat@st.comSigned-off-by: NMark Brown <broonie@kernel.org>

When transfer is shorter than half of the fifo, set the data packet size
up to transfer size instead of up to half of the fifo.
Check also that threshold is set at least to 1 data frame.
Signed-off-by: NAmelie Delaunay <amelie.delaunay@st.com>
Signed-off-by: NAlain Volmat <alain.volmat@st.com>
Link: https://lore.kernel.org/r/1597043558-29668-3-git-send-email-alain.volmat@st.comSigned-off-by: NMark Brown <broonie@kernel.org>

The caller of stm32_spi_transfer_one(), spi_transfer_one_message(),
is waiting for us to call spi_finalize_current_transfer() and will
eventually schedule a new transfer, if available.
We should guarantee that the spi controller is really available
before calling spi_finalize_current_transfer().

Move the call to spi_finalize_current_transfer() _after_ the call
to stm32_spi_disable().
Signed-off-by: NAntonio Borneo <antonio.borneo@st.com>
Signed-off-by: NAlain Volmat <alain.volmat@st.com>
Link: https://lore.kernel.org/r/1597043558-29668-2-git-send-email-alain.volmat@st.comSigned-off-by: NMark Brown <broonie@kernel.org>

Previously the stm32h7 interrupt thread cleared all non-masked interrupts.
If an interrupt was to occur during the handling of another interrupt its
flag would be unset, resulting in a lost interrupt.
This patches fixes the issue by clearing only the currently set interrupt
flags.
Signed-off-by: NTobias Schramm <t.schramm@manjaro.org>
Link: https://lore.kernel.org/r/20200804195136.1485392-1-t.schramm@manjaro.orgSigned-off-by: NMark Brown <broonie@kernel.org>

in l3gd20 driver startup, there is a setup failed error return from
stm32 spi driver

     "
     [    2.687630] st-gyro-spi spi0.0: supply vdd not found, using dummy
     regulator
     [    2.696869] st-gyro-spi spi0.0: supply vddio not found, using dummy
     regulator
     [    2.706707] spi_stm32 40015000.spi: SPI transfer setup failed
     [    2.713741] st-gyro-spi spi0.0: SPI transfer failed: -22
     [    2.721096] spi_master spi0: failed to transfer one message from queue
     [    2.729268] iio iio:device0: failed to read Who-Am-I register.
     [    2.737504] st-gyro-spi: probe of spi0.0 failed with error -22
     "

after debug into spi-stm32 driver, st-gyro-spi split two steps to read
l3gd20 id

first: send command to l3gd20 with read id command in tx_buf, rx_buf
is null.
second: read id with tx_buf is null, rx_buf not null.

so, for second step, stm32 driver recongise this process as 'SPI_SIMPLE_RX'
from stm32_spi_communication_type(), but there is no related process for this
type in stm32f4_spi_set_mode(), then we get error from
stm32_spi_transfer_one_setup().

we can use two method to fix this bug.
1, use stm32 spi's "In unidirectional receive-only mode (BIDIMODE=0 and
RXONLY=1)". but as our code running in sdram, the read latency is too large
to get so many receive overrun error in interrupts handler.

2, use stm32 spi's "In full-duplex (BIDIMODE=0 and RXONLY=0)", as tx_buf is
null, so add flag 'SPI_MASTER_MUST_TX' to spi master.

Change since V4:
1 remove dummy data sent out by stm32 spi driver
2 add flag 'SPI_MASTER_MUST_TX' to spi master
Signed-off-by: Ndillon min <dillon.minfei@gmail.com>
Link: https://lore.kernel.org/r/1590378348-8115-8-git-send-email-dillon.minfei@gmail.comSigned-off-by: NMark Brown <broonie@kernel.org>

Fix all functions and structure descriptions to have the driver
warning free when built with W=1.
Signed-off-by: NAlain Volmat <alain.volmat@st.com>
Reviewed-by: NAmelie Delaunay <amelie.delaunay@st.com>
Link: https://lore.kernel.org/r/1584711857-9162-1-git-send-email-alain.volmat@st.comSigned-off-by: NMark Brown <broonie@kernel.org>

dma_request_slave_channel() is a wrapper on top of dma_request_chan()
eating up the error code.

By using dma_request_chan() directly the driver can support deferred
probing against DMA.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20191212135550.4634-10-peter.ujfalusi@ti.comSigned-off-by: NMark Brown <broonie@kernel.org>

This switches the STM32 SPI driver over to using GPIO
descriptors for chip select. Instead of the callbacks for
picking the GPIO lines using the legacy API we just let
the core handle it all using descriptors.

Cc: Fabien Dessenne <fabien.dessenne@st.com>
Cc: Amelie Delaunay <amelie.delaunay@st.com>
Cc: Cezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20191205083401.27077-1-linus.walleij@linaro.orgSigned-off-by: NMark Brown <broonie@kernel.org>

During probe, return the "get_irq" error value instead of -ENOENT. This
allows the driver to be defer probed if needed.
Signed-off-by: NFabien Dessenne <fabien.dessenne@st.com>
Acked-by: NAmelie Delaunay <amelie.delaunay@st.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Add routines, registers & bitfield definition. Also baud rate divisor
definitions for STM32F4 SPI. This version supports full-duplex,
simplex TX and half-duplex TX communication with 8 or 16-bit per word.
DMA capability is optionally supported for transfer longer than 16 bytes.
For transfer less than 16 bytes frames can be send in discontinuous mode.
Signed-off-by: NCezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Prepare support for STM32F4 spi variant by introducing compatible
configuration data.
Move STM32H7 specific stuff to compatible data structure:
 - registers & bit fields
 - routines to control driver
 - baud rate divisor definitions
 - fifo availability
 - split IRQ functions to parts to be called when the IRQ occurs
   and for threaded interrupt what helps to provide less discontinuous
   mode for drivers without FIFO.
Signed-off-by: NCezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Add transfer_one_dma_start function to be more generic for other
stm32 SPI family drivers.
Signed-off-by: NCezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Split stm32_spi_transfer_one_setup function into smaller chunks
to be more generic for other stm32 SPI family drivers.
Signed-off-by: NCezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Interrupt function is used as a thread so rename it to express
meaning directly by more clear function name.
Signed-off-by: NCezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Rename STM32 SPI registers to be related to STM32H7 SPI driver
and not STM32 generally.
Signed-off-by: NCezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

This driver does not support SPI LOOP mode by configuration
in registers.
Signed-off-by: NCezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Driver supports SPI mode 0 to 3 not only the mode 3.
Use SPI_CPOL and SPI_CPHA indicates that these bits
can be changed to obtain modes 0 - 3.
Signed-off-by: NCezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

When SPI driver is configured to work only with TX or RX DMA channel
then dmaengine functions can dereferene NULL pointer.

Running full-duplex mode when when only RX or TX DMA channel is
available can cause overrun condition or incorrect writing to transmit
buffer so disable this types of DMA configuration and go back to
interrupt mode.
Signed-off-by: NCezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Patch fixes sparse warning: Using plain integer as NULL pointer. Replaces
second argument of function devm_clk_get from 0 to NULL.
Signed-off-by: NCezary Gapinski <cezary.gapinski@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>