There is no need to disable transmitter/receiver after each loop
iteration and re-enable it for next loop iteration. Enable the
transmitter/receiver before xfer loop starts and disable it when
the whole transfer is done.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Currently the driver only uses one internal chip select.
Add support for gpio chip selects configured by cs-gpios
DT binding.

Signed-off-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

speed_hz is a write only member, so we can safely remove it and its
generation. Also fixes the missing clk_put after getting the periph
clock.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Neither tegra20_spi_cdata nor tegra30_spi_cdata are used outside this
file so they can, and should, be static.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Reviewed-by: Thierry Reding <thierry.reding@avionic-design.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Use the newly introduced devm_ioremap_resource() instead of
devm_request_and_ioremap() which provides more consistent error handling.

devm_ioremap_resource() provides its own error messages; so all explicit
error messages can be removed from the failure code paths.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Acked-by: Eric Miao <eric.y.miao@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Add SPI driver for NVIDIA's Tegra114 SPI controller. This controller
is different than the older SoCs SPI controller in internal design as
well as register interface.

This driver supports the:
- non DMA based transfer for smaller transfer i.e. less than FIFO depth.
- APB DMA based transfer for larger transfer i.e. more than FIFO depth.
- Clock gating through runtime PM callbacks.
- registration through DT only.

Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Reviewed-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

The platform data header is no longer used. Delete it.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

Tegra only supports, and always enables, device tree. Remove all ifdefs
and runtime checks for DT support from the driver. Platform data is
therefore no longer required. Rework the driver to parse the device tree
directly into struct tegra_slink_data.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

Tegra only supports, and always enables, device tree. Remove all ifdefs
and runtime checks for DT support from the driver. Platform data is
therefore no longer required. Rework the driver to parse the device tree
directly into struct tegra_sflash_data.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

There is no code to set spi->controller_data, and hence the HW CS logic
can never trigger. Remove the unused code.

Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

This relies upon of_spi_register_master to find out which gpios to use.

Acked-by: Anton Vorontsov <anton@enomsg.org>
Signed-off-by: Andreas Larsson <andreas@gaisler.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

This adds support for the mostly register-compatible SPICTRL cores from the
GRLIB VHDL IP core library from Aeroflex Gaisler. They are normally running on
SPARC. A different entry in of_fsl_spi_match matches this core and indicates a
different hardware type that is used to set up different function pointers and
special cases.

The GRLIB core operates in cpu mode. The number of bits per word might be
limited. There might be native chipselects selected via a slave select
register. These differences to the FSL type cores, if present, are indicated by
a capabilities register. Other register and function differences exists but are
not relevant to the driver.

Acked-by: Anton Vorontsov <anton@enomsg.org>
Signed-off-by: Andreas Larsson <andreas@gaisler.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

Acked-by: Anton Vorontsov <anton@enomsg.org>
Signed-off-by: Andreas Larsson <andreas@gaisler.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

For being able to distinguishing between the regular type of cores and others
with different entries in of_fsl_spi_match.

Acked-by: Anton Vorontsov <anton@enomsg.org>
Signed-off-by: Andreas Larsson <andreas@gaisler.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

spi/spi-fsl-spi: Move setting non-zero tx and rx shifts to a function accessed by a function pointer

Acked-by: Anton Vorontsov <anton@enomsg.org>
Signed-off-by: Andreas Larsson <andreas@gaisler.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

This is needed for a device in SPI_CS_HIGH mode that otherwise could start out
active for the first transaction.

Acked-by: Anton Vorontsov <anton@enomsg.org>
Signed-off-by: Andreas Larsson <andreas@gaisler.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

This makes the spi-fsl-spi driver usable in CPU mode outside of an FSL_SOC and
even an powerpc environment by moving CPM mode functionality to a separate file
that is only compiled and linked in an FSL_SOC environment and adding some
ifdefs to hide types and functions or provide alternatives.

For devicetree probing a "clock-frequency" property is used for clock frequency
instead of calls to FSL_SOC-specific functions.

Acked-by: Anton Vorontsov <anton@enomsg.org>
Signed-off-by: Andreas Larsson <andreas@gaisler.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>

Fix to return a negative error code from the error handling
case instead of 0, as returned elsewhere in this function.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

This reverts commit faa98f7e which was
applied in error due to discussion ending up in the wrong thread.

Add CONFIG_PM_SLEEP to suspend/resume functions to fix the following
build warning when CONFIG_PM_SLEEP is not selected. This is because
sleep PM callbacks defined by SET_SYSTEM_SLEEP_PM_OPS are only used
when the CONFIG_PM_SLEEP is enabled.

drivers/spi/spi-s3c64xx.c:1362:12: warning: 's3c64xx_spi_suspend' defined but not used [-Wunused-function]
drivers/spi/spi-s3c64xx.c:1381:12: warning: 's3c64xx_spi_resume' defined but not used [-Wunused-function]

Signed-off-by: Jingoo Han <jg1.han@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

The status of transfer is stored in controller data structure
so that it can be used not only by atmel_spi_msg_done() function.
This will be useful for upcoming dmaengine enabled driver.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

The "has_dma_support" needed for future use with dmaengine driver.

[Fixed some unneded ternery operators -- broonie]

Signed-off-by: Wenyou Yang <wenyou.yang@atmel.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Spi starts transfer using dma with DMA_CTRL_ACK which is not require
becasue spi driver does not use completed dma_desc after transfer
done and so it does not ack the dma descriptor. Removing the
DMA_CTRL_ACK flag to avoid memory leak in dma driver.

Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Replaced calls to IS_ERR and PTR_ERR with PTR_RET function.
Patch found using coccinelle.

Signed-off-by: Alexandru Gheorghiu <gheorghiuandru@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Replaced calls to IS_ERR and PTR_ERR with PTR_RET function.
Patch found using coccinelle.

Signed-off-by: Alexandru Gheorghiu <gheorghiuandru@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Some SPI slave devices require asserted chip select signal across
multiple transfer segments of an SPI message. Currently the driver
always de-asserts the internal SS signal for every single transfer
segment of the message and ignores the 'cs_change' flag of the
transfer description. Disable the internal chip select (SS) only
if this is needed and indicated by the 'cs_change' flag.

Without this change, each partial transfer of a surrounding
multi-part SPI transaction might erroneously change the SS
signal, which might prevent slaves from answering the request
that was sent in a previous transfer segment because the
transaction could be considered aborted (SS was de-asserted
before reading the response).

Reported-by: Gerhard Sittig <gerhard.sittig@ifm.com>
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

The variable bits_per_word is initialized but never used
otherwise, so remove the unused variable.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Acked-By: Laxman Dewangan <ldewangan@nvidia.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

This eliminates having an #ifdef returning NULL for the case
when OF is disabled.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

This eliminates having an #ifdef returning NULL for the case
when OF is disabled.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Acked-by: Thomas Chou <thomas@wytron.com.tw>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

spi_pump_messages() calls into a controller driver with
unprepare_transfer_hardware() which is documented as "This may sleep".
As in the prepare_transfer_hardware() call below, we should release the
queue_lock spinlock before making the call.
Rework the logic a bit to hold queue_lock to protect the 'busy' flag,
then release it to call unprepare_transfer_hardware().

Signed-off-by: Bryan Freed <bfreed@chromium.org>
Reviewed-by: Doug Anderson <dianders@chromium.org>
Signed-off-by: Doug Anderson <dianders@chromium.org>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

It is possible that the handler gets interrupted after checking the
status. After it resumes the time out is due but the condition it was
waiting for might be true as well. Therefore it is necessary to check
the condition in case of an time out to be sure that the condition is
not true after the time passed by.

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

The core can do the validation for us.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

This driver only supports bits_per_word==8, so inform the SPI core of
this. Remove all the open-coded validation that's no longer needed.

Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Allow SPI masters to define the set of bits_per_word values they support.
If they do this, then the SPI core will reject transfers that attempt to
use an unsupported bits_per_word value. This eliminates the need for each
SPI driver to implement this checking in most cases.

Signed-off-by: Stephen Warren <swarren@wwwdotorg.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

The status of the interrupt is available in the status register,
so reading the clear pending register and writing back the same
value will not actually clear the pending interrupts. This patch
modifies the interrupt handler to read the status register and
clear the corresponding pending bit in the clear pending register.

Modified the hwInit function to clear all the pending interrupts.

Signed-off-by: Girish K S <ks.giri@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

The pointer to the driver data is never used to get the slave
controller data. We can delete the unused argument from the function.

Signed-off-by: Matthias Brugger <matthias.bgg@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Use devm_ioremap_resource() which provides its own error messages.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Acked-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>