Some IP cores have an internal 'bus free' logic which may be more
advanced than just checking if SDA is high. Add a separate callback to
get this status. Filling it is optional.
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

After exiting the while loop, we checked if recovery was successful and
sent a STOP to the clients. Meanwhile however, we send a STOP after
every pulse, so it is not needed after the loop. If we move the check
for a free bus to the end of the while loop, we can shorten and simplify
the logic. It is still ensured that at least one STOP will be sent to
the wire even if SDA was not stuck low.
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: NPeter Rosin <peda@axentia.se>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

For bus recovery, we either need to bail out early if we can read SDA or
we need to send STOP after every pulse. Otherwise recovery might be
misinterpreted as an unwanted write. So, require one of those SDA
handling functions to avoid this problem.
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Acked-by: NPeter Rosin <peda@axentia.se>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

I2C clients may misunderstand recovery pulses if they can't read SDA to
bail out early. In the worst case, as a write operation. To avoid that
and if we can write SDA, try to send STOP to avoid the
misinterpretation.
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: NPeter Rosin <peda@axentia.se>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>
Cc: stable@kernel.org

Currently, of_i2c_register_devices() is responsible for retaining
info->of_node, but we're about to expose a function to parse I2C board
info without registering the I2C device.

We could possibly let this function retain ->of_node, but this approach
is prone to reference leak since people will have to remember to call
of_node_put() if something goes wrong between the OF node parsing and
the registration step.
Let's just retain the ->of_node in i2c_new_register() instead.
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

The only user of i2c_board_info->archdata is the OF parsing code and it
just pass a zero-initialized object which has the same effect as leaving
->archdata to NULL since the client object is allocated with kzalloc().

Get rid of this useless field.
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

The limitation of being able to check only for -EPROBE_DEFER from
dev_pm_domain_attach() has been removed. Hence let's respect all error
codes and bail out accordingly.
Signed-off-by: NUlf Hansson <ulf.hansson@linaro.org>
Acked-by: NWolfram Sang <wsa@the-dreams.de>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The API docs describe i2c_transfer() as taking a pointer to an array
of i2c_msg containing at least 1 entry, but leaves it to the individual
drivers to sanity check the msgs and num parameters. Let's do this in
core code instead.
Signed-off-by: NArd Biesheuvel <ard.biesheuvel@linaro.org>
[wsa: changed '<= 0' to '< 1']
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

No changes in refcount semantics -- key init is false; replace

static_key_slow_inc|dec   with   static_branch_inc|dec
static_key_false          with   static_branch_unlikely

Added a '_key' suffix to i2c_trace_msg, for better self
documentation.
Signed-off-by: NDavidlohr Bueso <dbueso@suse.de>
Reviewed-by: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

According to the event i2c_result defined in include/trace/events/i2c.h,
the second parameter should be the number of messages instead of the
ended loop index. The value of ended loop index is the same as ret.
Signed-off-by: NAhbong Chang <cwahbong@google.com>
Reviewed-by: NTodd Poynor <toddpoynor@google.com>
Reviewed-by: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

After previous refactoring, there is only one user in the same file
left. Make the function static now.

[wsa: added 'int' to bare 'unsigned']
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Can be used during probe to double check that the probed device is
what is expected.

Loosely based on code from Adrian Fiergolski <adrian.fiergolski@cern.ch>.
Tested-by: NAdrian Fiergolski <adrian.fiergolski@cern.ch>
Reviewed-by: NWolfram Sang <wsa@the-dreams.de>
Signed-off-by: NPeter Rosin <peda@axentia.se>

The buses should honor the firmware interface used to register the device,
but the I2C core reports a MODALIAS of the form i2c:<device> even for I2C
devices registered via OF.

This means that user-space will never get an OF stype uevent MODALIAS even
when the drivers modules contain aliases exported from both the I2C and OF
device ID tables. For example, an Atmel maXTouch Touchscreen registered by
a DT node with compatible "atmel,maxtouch" has the following module alias:

$ cat /sys/class/i2c-adapter/i2c-8/8-004b/modalias
i2c:maxtouch

So udev won't be able to auto-load a module for an OF-only device driver.
Many OF-only drivers duplicate the OF device ID table entries in an I2C ID
table only has a workaround for how the I2C core reports the module alias.

This patch changes the I2C core to report an OF related MODALIAS uevent if
the device was registered via OF. So for the previous example, after this
patch, the reported MODALIAS for the Atmel maXTouch will be the following:

$ cat /sys/class/i2c-adapter/i2c-8/8-004b/modalias
of:NtrackpadT<NULL>Catmel,maxtouch

NOTE: This patch may break out-of-tree drivers that were relying on this
      behavior, and only had an I2C device ID table even when the device
      was registered via OF. There are no remaining drivers in mainline
      that do this, but out-of-tree drivers have to be fixed and define
      a proper OF device ID table to have module auto-loading working.
Signed-off-by: NJavier Martinez Canillas <javierm@redhat.com>
Tested-by: NDmitry Mastykin <mastichi@gmail.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Reference count of device node was increased in of_i2c_register_device,
but without decreasing it in i2c_unregister_device. Then the added
device node will never be released. Fix this by adding the of_node_put.
Signed-off-by: NLixin Wang <alan.1.wang@nokia-sbell.com>
Tested-by: NWolfram Sang <wsa@the-dreams.de>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>
Cc: stable@kernel.org

If we managed to get a client release SDA again, send a STOP afterwards
to make sure we have a consistent state on the bus again.
Tested-by: NPhil Reid <preid@electromag.com.au>
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

If we have a function to control SDA, we should ensure that SDA is not
held down by us. So, release the GPIO in this case.
Tested-by: NPhil Reid <preid@electromag.com.au>
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

This will be needed when we want to create STOP conditions, too, later.
Create the needed fields and populate them for the GPIO case if the GPIO
is set to output.
Tested-by: NPhil Reid <preid@electromag.com.au>
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Those two functions are very similar, the only differences are that one
needs the I2C_M_RD flag for its message while the other one needs the
buffer casted to drop the const. Introduce a generic helper which allows
to specify the flags (also needed later for DMA safe variants of these
calls) and let the casting be done in the inlining functions which are
now calling the new helper function.
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: NJonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

One helper checks if DMA is suitable and optionally creates a bounce
buffer, if not. The other function returns the bounce buffer and makes
sure the data is properly copied back to the message.
Reviewed-by: NJonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

The correct header to include for the gpiod interface is
<linux/gpio/consumer.h>.

Fixes: 3991c5c8 ("i2c: Switch to using gpiod interface for gpio bus recovery")
Signed-off-by: NPhil Reid <preid@electromag.com.au>
Reviewed-by: NJarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Remove all reference to code related to using integer based ids for
scl/sda gpio for bus recovery. All in tree drivers are now using the
gpio descriptors to specific the required gpios.
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NPhil Reid <preid@electromag.com.au>
Reviewed-by: NJarkko Nikula <jarkko.nikula@linux.intel.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Currently the i2c gpio recovery code uses gpio integer interface
instead of the gpiod. This change switch the core code to use
the gpiod while still retaining compatibility with the gpio integer
interface. This will allow individual driver to be updated and tested
individual to switch to using the gpiod interface.
Reviewed-by: NJarkko Nikula <jarkko.nikula@linux.intel.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NPhil Reid <preid@electromag.com.au>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

It's a common pattern to be NULL-aware when freeing resources.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Add a call to of_i2c_setup_smbus_alert when a i2c adapter is registered
so the the smbalert driver can be registered.
Signed-off-by: NPhil Reid <preid@electromag.com.au>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Move the check for a stuck SCL before the check for a high SDA.
This prevent false positives in the specific case that SDA is fine
and SCL is stuck, which previously returned 0.

Also check SDA again after the loop, if we can.
Together, these changes should lead to a lot more failed
recoveries being caught and returning error codes.
Signed-off-by: NClaudio Foellmi <claudio.foellmi@ergon.ch>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

For devices not instantiated through ACPI the i2c-client's device-name
gets set to <busnr>-<addr> by default, e.g. "0-0022" this means that
the device-name is dependent on the order in which the i2c-busses are
enumerated.

In some cases having a predictable constant device-name is desirable,
for example on non device-tree platforms the link between a regulator
and its consumers is specified by the platform code by setting
regulator_init_data.consumers. This array identifies the regulator's
consumers by dev_name and supply(-name). Which requires a constant
dev_name.

This commit adds a dev_name field to i2c_board_info allowing
platform code to set a contstant dev_name so that the device can
be identified by its dev_name in other platform code.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Acked-by: Mark Brown <broonie@kernel.org> (live at ELCE17)
Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> (live at ELCE17)
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

I2C drivers were required to have an I2C device ID table even if were for
devices that would only be registered using a specific firmware interface
(e.g: OF or ACPI).

But commit da10c06a ("i2c: Make I2C ID tables non-mandatory for DT'ed
devices") changed the I2C core to relax the requirement and allow drivers
to avoid defining this table.

Unfortunately it only took into account drivers for OF-only devices and
forgot about ACPI-only ones, and this was fixed by commit c64ffff7
("i2c: core: Allow empty id_table in ACPI case as well").

But the latter didn't update the original comment, so it doesn't reflect
what the code does now.
Signed-off-by: NJavier Martinez Canillas <javierm@redhat.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

For now empty ID table is not allowed with ACPI and prevents driver to
be probed.

Add a check to allow empty ID table.

This introduces a helper i2c_acpi_match_device().

Note, we rename some static function in i2c-core-acpi.c to distinguish
with public API.

Fixes: da10c06a ("i2c: Make I2C ID tables non-mandatory for DT'ed devices")
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: NRajmohan Mani <rajmohan.mani@intel.com>
Acked-by: NMika Westerberg <mika.westerberg@linux.intel.com>
[wsa: needed to get some drivers probed again]
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Finally, apply modern comment rules to the file header. The old style
looked very non-Linuxish and challenged my eyes for some time now. I
also added my own copyright for the period of me being the maintainer.
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

They seem like cruft to me. I couldn't find any evidence that something
included from there is actually used.
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Removes some ifdeffery. Also add the new file to the relevant
MAINTAINERS section.
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Also removes some ifdeffery.
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Break out the exported SMBus functions and the emulation layer into a
separate file. This also involved splitting up the tracing header into
an I2C and an SMBus part.
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Also removes some ifdeffery.
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

The I2C core became quite huge and its monolithic structure makes
maintenance hard. So, prepare to break out some functionality into
separate files by renaming the source file. Note that we keep the
resulting object name constant to avoid regressions.
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

By default the i2c-core will try to get an irq with index 0 on ACPI / of
instantiated devices. This is troublesome on some ACPI systems where the
irq info at index 0 in the CRS table may contain nonsense and/or point
to an irqchip for which there is no Linux driver.

If this happens then before this commit the driver's probe method would
never get called because i2c_device_probe will try to get an irq by
calling acpi_dev_gpio_irq_get which will always return -EPROBE in this
case, as it waits for a matching irqchip driver to load. Thus causing
the driver to not get a chance to bind.

This commit adds a new disable_i2c_core_irq_mapping flag to struct
i2c_driver which a driver can set to tell the core to skip irq mapping.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

By default the i2c subsys creates an i2c-client for the first I2cSerialBus
resource of an acpi_device, but some acpi_devices have multiple
I2cSerialBus resources and we may want to instantiate i2c-clients for
the others.

This commit adds a new i2c_acpi_new_device function which can be used to
create an i2c-client for any I2cSerialBus resource of an acpi_device.

Note that the other resources may even be on a different i2c bus, so just
retrieving the client address is not enough.

Here is an example DSDT excerpt from such a device:

Device (WIDR)
{
    Name (_HID, "INT33FE" /* XPOWER Battery Device */)
    Name (_CID, "INT33FE" /* XPOWER Battery Device */)
    Name (_DDN, "WC PMIC Battery Device")
<snip>
    Name (RBUF, ResourceTemplate ()
    {
        I2cSerialBusV2 (0x005E, ControllerInitiated, 0x000186A0,
            AddressingMode7Bit, "\\_SB.PCI0.I2C7",
            0x00, ResourceConsumer, , Exclusive,
            )
        I2cSerialBusV2 (0x0036, ControllerInitiated, 0x000186A0,
            AddressingMode7Bit, "\\_SB.PCI0.I2C1",
            0x00, ResourceConsumer, , Exclusive,
            )
        I2cSerialBusV2 (0x0022, ControllerInitiated, 0x00061A80,
            AddressingMode7Bit, "\\_SB.PCI0.I2C1",
            0x00, ResourceConsumer, , Exclusive,
            )
        I2cSerialBusV2 (0x0054, ControllerInitiated, 0x00061A80,
            AddressingMode7Bit, "\\_SB.PCI0.I2C1",
            0x00, ResourceConsumer, , Exclusive,
            )
        GpioInt (Level, ActiveLow, Exclusive, PullNone, 0x0000,
            "\\_SB.PCI0.I2C7.PMI5", 0x00, ResourceConsumer, ,
            )
            {   // Pin list
        0x0012
            }
        GpioInt (Edge, ActiveLow, ExclusiveAndWake, PullNone, 0x0000,
            "\\_SB.GPO1", 0x00, ResourceConsumer, ,
            )
            {   // Pin list
        0x0005
            }
        GpioInt (Level, ActiveLow, Exclusive, PullNone, 0x0000,
            "\\_SB.PCI0.I2C7.PMI5", 0x00, ResourceConsumer, ,
            )
            {   // Pin list
        0x0013
            }
    })
    Method (_CRS, 0, NotSerialized)  // _CRS: Current Resource Settings
    {
        Return (RBUF) /* \_SB_.PCI0.I2C7.WIDR.RBUF */
    }
<snip>
}
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Modify struct i2c_acpi_lookup and i2c_acpi_fill_info() to allow
using them to get the info from a certain index in the ACPI-resource
list rather then taking the first I2cSerialBus resource.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NWolfram Sang <wsa@the-dreams.de>

Simple integer for interrupt number is not expressive enough, as it does
not convey interrupt trigger type that should be used. Let's allow
attaching array of resources to the board info and have i2c core parse
first IRQ resource and set up interrupt trigger as needed.
Reviewed-by: NWolfram Sang <wsa@the-dreams.de>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>

i2c bus has 2 different types of device belonging to the same bus and
bus notifiers use device type to distinguish between adapters and clients.
Previously we only had i2c_adapter_type exported, which made code wanting
to work with i2c_client devices test for type not equal to adapter type.
This unfortunately is not safe if we ever add another type to the bus,
so let's export i2c_client_type as well.
Reviewed-by: NJean Delvare <jdelvare@suse.de>
Acked-by: NBenjamin Tissoires <benjamin.tissoires@redhat.com>
Reviewed-by: NWolfram Sang <wsa@the-dreams.de>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>