This reverts commit 7c86d44c.
Stephen says:

IIRC, I proposed it before solely to solve some suspend/resume ordering
issues, and Colin Cross NAKd it. These days, deferred probe should make
this change unnecessary.
Reported-by: NStephen Warren <swarren@wwwdotorg.org>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Extends the driver to get properties from device tree. Rather than
pass the N & M factors in DT, use the more standard clock-frequency
property. Calculate N & M at run time. In order to do this, we need to
know tclk. So the driver uses clk_get() etc in order to get the clock
and clk_get_rate() to determine the tclk rate. Not all platforms
however have CLK, so some #ifdefery is needed to ensure the driver
still compiles when CLK is not available.
Signed-off-by: NAndrew Lunn <andrew@lunn.ch>

[wsa: converted some ints to u32 to match signedness]
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Allocate memory for device state using devm_kzalloc(), get the
clock using devm_clk_get(), get the IRQ using devm_request_irq(),
request and remap memory using devm_request_and_ioremap().
All to simplify accounting and letting the kernel do the
garbage-collection.
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Some architectures supports only 16-bit or 32-bit read/write access to
their IO space. Add a 'reg-io-width' platform and OF parameter which
specifies the IO width to support these platforms.

reg-io-width can be specified as 1, 2 or 4, and has a default value
of 1 if it is unspecified.
Signed-off-by: NGanesan Ramalingam <ganesanr@broadcom.com>
Signed-off-by: NJayachandran C <jayachandranc@netlogicmicro.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Update for change in i2c-ocores.h which uses reg_shift to
specify the register offset shifts instead of regstep.
Signed-off-by: NJayachandran C <jayachandranc@netlogicmicro.com>
Acked-by: NRichard Rojfors <richard.rojfors@pelagicore.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Deprecate 'regstep' property and use the standard 'reg-shift' property
for register offset shifts. 'regstep' will still be supported as an
optional property, but will give a warning when used.
Signed-off-by: NGanesan Ramalingam <ganesanr@broadcom.com>
Signed-off-by: NJayachandran C <jayachandranc@netlogicmicro.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Cleanups to i2c-cores, no change in logic, changes are:
* Move i2c-ocores device tree documentation from source file to
  Documentation/devicetree/bindings/i2c/i2c-ocores.txt.
* Add \n to dev_warn and dev_err messages where missing
* Minor updates to the text and formatting fixes.
Signed-off-by: NJayachandran C <jayachandranc@netlogicmicro.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

As git history indicates, the driver predates the git era and is heavily
used and worked on since. Not EXPERIMENTAL anymore.
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>
Cc: Rodolfo Giometti <giometti@linux.it>
Cc: Florian Fainelli <ffainelli@freebox.fr>

There's no point in using _sync() as we don't really care if the suspend
has completed immediately.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Reviewed-by: NShubhrajyoti D <shubhrajyoti@ti.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

This commit fixes warning introduced in 27452498 ("i2c-s3c2410: Rework
device type handling"):

  drivers/i2c/busses/i2c-s3c2410.c: In function 's3c24xx_get_device_quirks':
  drivers/i2c/busses/i2c-s3c2410.c:125: warning: passing argument 1 of 'of_match_node' from incompatible pointer type
  include/linux/of.h:245: note: expected 'const struct of_device_id *' but argument is of type 'const struct of_device_id (*)[4]'
Signed-off-by: NKarol Lewandowski <k.lewandowsk@samsung.com>
Signed-off-by: NKyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

This patch configures the I2C bus timing registers according
to information passed via DT. Currently, 100kHz and 400kHz
modes are supported.

The TIMING2 register value is wrong in the documentation for
i.MX28! This was found and fixed by:
  Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMarek Vasut <marex@denx.de>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Depend on 1e92bf6dSigned-off-by: NSonic Zhang <sonic.zhang@analog.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Depend on commit cf93feb3Signed-off-by: NSonic Zhang <sonic.zhang@analog.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

TWI bit mask macros are moved to twi head file.
Depend on commit 61c16b5cSigned-off-by: NSonic Zhang <sonic.zhang@analog.com>
Signed-off-by: NBob Liu <lliubbo@gmail.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Current driver was developed based on BF537 0.2 HRM. In high system load, BUFRDERR error
interrupt may be raised if XMTSERV interrupt of last TX byte is not served in time
(set RSTART bit), which breaks restart tranfer as expected.

"Buffer Read Error (BUFRDERR)" description in Blackfin HRM only applys to BF537
rev. < 0.3. In later rev. and later announced Blackfin chips, such as BF527 and
BF548, a new TWI master feature "Clock Stretching" is added into the TWI controller,
BUFRDERR interrupt is not triggered after TX FIFO is empty.

This patch sets RSTART bit at the beginning of the first transfer. The SCL and SDA
is hold till XMTSERV interrupt of last TX byte is served. Restart transfer is not broken
in high system load.
Signed-off-by: NSonic Zhang <sonic.zhang@analog.com>

[wsa: fixed spaces around operators]
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

In order to mark I2C transfer fail when MEN bit in I2C controller is
reset unexpectedly in MCOMP interrupt, interrupt status bits XMTSERV or
RCVSERV should be checked.

Master Transfer Complete (MCOMP).
[1] The initiated master transfer has completed. In the absence of a
repeat start, the bus has been released.
[0] The completion of a transfer has not been detected.
Signed-off-by: NSonic Zhang <sonic.zhang@analog.com>

[wsa: fixed spaces around operators and typo in commit message]
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Some fault i2c device may hold the sda/scl line and cause i2c driver
wait in the BUS busy loop. The I2C framework already retry the
transfer loop before timeout. Return -EAGAIN instead of pull BUSBUSY
in the other loop.
Signed-off-by: NSonic Zhang <sonic.zhang@analog.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

i2c: i2c-bfin-twi: Improve the patch for bug "Illegal i2c bus lock upon certain transfer scenarios".

For transfer counts > 255 bytes i2c-bfin-twi sets the data
transfer counter DCNT to 0xFF indicating unlimited transfers.
It then uses a flag iface->manual_stop to manually issue the STOP
condition, once the required amount of bytes are received.

We found that on I2C receive operation issuing the STOP condition
together with a FULL RCV FIFO (2bytes) will cause SDA and SCL be
constantly driven low.

This patch stops receiving operation immediately in last rx interrupt.
This patch also wakes up waiting process when transfer completes.
Signed-off-by: NSonic Zhang <sonic.zhang@analog.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

For transfer counts > 255 bytes i2c-bfin-twi sets the data
transfer counter DCNT to 0xFF indicating unlimited transfers.
It then uses a flag iface->manual_stop to manually issue the STOP
condition, once the required amount of bytes are received.

We found that on I2C receive operation issuing the STOP condition
together with a FULL RCV FIFO (2bytes) will cause SDA and SCL be
constantly driven low.

Temporary workaround until further investigation:
Discard the RCV FIFO before issuing the STOP condition.
Signed-off-by: NMichael Hennerich <michael.hennerich@analog.com>
Signed-off-by: NSonic Zhang <sonic.zhang@analog.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

The driver currently checks the platform device id and rejects platform
device id different from 0. This prevents the registration of a second
i2c controller on systems where a second one might be available (such as
Kirkwood 88F6282).

CC: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: NNicolas Schichan <nschichan@freebox.fr>
Signed-off-by: NFlorian Fainelli <ffainelli@freebox.fr>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

The OMAP3530 is based upon the same silicon as the OMAP3430 and so the I2C
revision is the same for 3430 and 3530. However, the OMAP3630 device has the
same I2C revision as OMAP4. Correct the revision definition to reflect this.

This patch is based on work done by Jon Hunter <jon-hunter@ti.com>
Changes from his patch
- Update OMAP_I2C_REV_ON_3430 also to reflect that it is same as 3530
Reviewed-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NJon Hunter <jon-hunter@ti.com>
Signed-off-by: NShubhrajyoti D <shubhrajyoti@ti.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Use INIT_COMPLETION instead of init_completion in transfer.
Reviewed-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NShubhrajyoti D <shubhrajyoti@ti.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Use SET_RUNTIME_PM_OPS macro to set runtime functions.
Acked-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NShubhrajyoti D <shubhrajyoti@ti.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

The omap_i2c_remove function may not be needed after
device exit so the memory could be freed.
Signed-off-by: NShubhrajyoti D <shubhrajyoti@ti.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

To provide the late suspend and early resume for i2c
driver, convert the suspend/resume as
	suspend-> suspend_noirq
	resume -> resume_noirq
Signed-off-by: NLaxman Dewangan <ldewangan@nvidia.com>

[wsa: fixed up to match previous pm_ops change]
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Make the Tegra I2C controller driver define its PM callbacks through
a struct dev_pm_ops object rather than by using legacy PM hooks
in struct platform_driver.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Acked-by: NLaxman Dewangan <ldewangan@nvidia.com>

[wsa] adapt to of_match_ptr change
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Make the ST-Ericsson U300 I2C controller driver define its PM
callbacks through a struct dev_pm_ops object rather than by using
legacy PM hooks in struct platform_driver.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Make the PKUnity-v3 SoC I2C controller driver define its suspend
callback through a struct dev_pm_ops object rather than by using
a legacy PM hook in struct platform_driver.  The empty resume
callback is not necessary, so remove it.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Acked-by: NGuan Xuetao <gxt@mprc.pku.edu.cn>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Make the PNX I2C controller driver define its PM callbacks through
a struct dev_pm_ops object rather than by using legacy PM hooks
in struct platform_driver.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Make the OpenCores I2C controller driver define its PM callbacks
through a struct dev_pm_ops object rather than by using legacy PM
hooks in struct platform_driver.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Make the Blackfin On-Chip Two Wire Interface driver define its PM
callbacks through a struct dev_pm_ops object rather than by using
legacy PM hooks in struct platform_driver.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Acked-by: NSonic Zhang <sonic.zhang@analog.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Make the AT91 Two-Wire Interface driver define its PM callbacks
through a struct dev_pm_ops object rather than by using legacy PM
hooks in struct platform_driver.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

In place of defining match_table for non-DT based as NULL,
use of_match_ptr() for initialzing the of_match_table.
Signed-off-by: NLaxman Dewangan <ldewangan@nvidia.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Use clk_disable_unprepare() inplace of clk_disable().
This was missed as part of moving clock enable/disable to
prepare/unprepare for using the common clock framework.
Signed-off-by: NLaxman Dewangan <ldewangan@nvidia.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Remove unused member variable "iomem" of the
i2c device structure.

This variable becomes unused when converted all allocation
to devm_* in following change:
    i2c: tegra: make all resource allocation through devm_*
Signed-off-by: NLaxman Dewangan <ldewangan@nvidia.com>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

sparse found this assignment of u32 to an int. Fix it:

drivers/i2c/busses/i2c-imx.c:540:56: warning: incorrect type in argument 3 (different signedness)

and also fix the type in platform_data. All current users use values
which fit into the old and new type, so it is a safe change.
Signed-off-by: NWolfram Sang <wolfram@the-dreams.de>
Reviewed-by: NRichard Zhao <richard.zhao@freescale.com>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>

With the new i.mx clock framework the i2c clock is registered as:

clk_register_clkdev(clk[i2c1_ipg_gate], NULL, "imx-i2c.0")

So we do not need to pass "i2c_clk" string and can use NULL instead.
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>

[wsa: rebased on top of the devm-conversion]
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Make sure we prepare/unprepare the clock for the ST U300
I2C driver as is required by the clk API especially if you
use common clock.
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

This patch mutes the false positive compiler warning:

drivers/i2c/busses/i2c-mxs.c: In function 'mxs_i2c_xfer_msg':
drivers/i2c/busses/i2c-mxs.c:206:8: warning: 'data' may be used uninitialized in this function [-Wuninitialized]
drivers/i2c/busses/i2c-mxs.c:196:6: note: 'data' was declared here
Signed-off-by: NMarc Kleine-Budde <mkl@pengutronix.de>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>

Signed-off-by: NVirupax Sadashivpetimath <virupax.sadashivpetimath@stericsson.com>
Signed-off-by: NSrinidhi KASAGAR <srinidhi.kasagar@stericsson.com>
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>