Add support for I2C Slave Drivers #5224

superna9999 · 2017-12-01T11:22:37Z

This PR adds support for I2C Slave by:

Adding proper I2C slave calls to I2C Controllers
Adding I2C Slave Drivers infrastructure and sample
Adding test with physical loopback
Adding support to STM32 V2 Driver

The Slave support is based on Linux I2C Slave support by Wolfram Sang and documented at https://www.kernel.org/doc/Documentation/i2c/slave-interface

The I2C Slave support is pretty simple and straightforward and it's desired.

No code has been copied, this PR only contains new code written from scratch.

This PR should solve #2956

superna9999 · 2017-12-01T11:23:32Z

I would like @tbursztyka @erwango @ldts to look at this !

tbursztyka

I quickly went through, it looks quite nice to me at least. Let's see what others have to say.

Beware the necessary {} around one liners on if/else/while/for. That's basically the only difference we enforce in zephyr compared to Linux code style.

tbursztyka · 2017-12-01T12:21:36Z

include/i2c.h

+ */
+__syscall bool i2c_slave_is_supported(struct device *dev);
+
+static inline bool _impl_i2c_slave_is_supported(struct device *dev)


I don't think you need this.
user/dev should know what the controller he's working against should support.

And, if he is really trying to mess with the wrong controller: i2c_slave_attach() should return -ENOTSUP (or -EBUSY if already used by another slave driver) and that's it.

You are right, this call is quite useless indeed

tbursztyka · 2017-12-01T12:31:28Z

drivers/i2c/i2c_ll_stm32_v2.c

+				LL_I2C_AcknowledgeNextData(i2c, LL_I2C_NACK);
+		}
+
+		return;


Could you move the logic from line 67 to here, into a dedicated function? just to get a clearer isr handler.

It was my plan, will do

tbursztyka · 2017-12-01T12:34:06Z

include/i2c.h

+
+
+/**
+ * @brief Get the I2C Slave device specific funcs


Will this be really needed?

The controller, once the driver slave is attached knows the functions, and the slave driver obviously too. What will need this? (besides the test app which uses it to initialize some data into the "eeprom").

In my mind it's useful, but there must be a simpler way to achieve the same goal

tbursztyka · 2017-12-01T12:36:13Z

drivers/i2c/slave/Kconfig

+#
+menuconfig I2C_SLAVE
+	bool
+	prompt "I2C Slave Drivers"


Either use the prompt or not. Don't mix both in the same Kconfig.

Actually, my personal preference would go for just "message to prompt" and not using prompt keyword.

I copied it from i2c verbatim, i will fix this

tbursztyka · 2017-12-01T12:37:19Z

drivers/i2c/slave/Kconfig

+	  - 3 INFO, write SYS_LOG_INF in addition to previous levels
+	  - 4 DEBUG, write SYS_LOG_DBG in addition to previous levels
+
+config I2C_SLAVE_EEPROM


I think you can move i²c slave eeprom stuff into its own Kconfig.eeprom.
If your api gets in - and it will - there will be a lot more i²c slaves I hope here.

tbursztyka · 2017-12-01T12:49:03Z

tests/drivers/i2c/i2c_slave_eeprom/src/main.c

+	}
+
+	/* Program dummy bytes */
+	ret = funcs->program(eeprom, eeprom_data, TEST_DATA_SIZE);


There should be a way to do this differently.

By exposing a way in the eeprom driver to initialize it with a memory address and the size it's pointing. (so the driver would not own the memory buffer by itself, but an address, etc...). You could have such function in include/drivers/i2c/slave/eeprom.h for instance.

Hmm I think I see what you mean, a eeprom_slave function only defined in the eeprom.h header. Indeed, it will simplify a lot.

SebastianBoe · 2017-12-01T12:56:56Z

include/drivers/i2c/slave_eeprom.h

+typedef int (*eeprom_slave_read_t)(struct device *dev, u8_t *data,
+				   unsigned int offset);
+
+struct eeprom_slave_api {


Is it necessary to be inconsistent with

slave_eeprom
eeprom_slave

?

You are right, it was in my todo list

SebastianBoe · 2017-12-01T13:00:39Z

tests/drivers/i2c/i2c_slave_eeprom/CMakeLists.txt

+    set(CONF_FILE "prj.conf ${APPLICATION_SOURCE_DIR}/boards/${BOARD}.conf")
+  else()
+    set(CONF_FILE "prj.conf")
+  endif()


There exists a convention that board-specific .conf files are named

${APPLICATION_SOURCE_DIR}/prj_${BOARD}.conf

The build system is aware of this convention and will use it automatically to detect board-specific .conf files.

So if you

mv boards/nucleo_f091rc.conf prj_nucleo_f091rc.conf

then you can get rid of this macro and it will just-work.

It's the same issue I want to solve as in #5200, where I want a commpon .conf and a board specific .conf fragment as @mbolivar suggested.

mnkp

Thanks @superna9999 for working on I2C slave support. I have some initial comments regarding your implementation.

mnkp · 2017-12-05T01:23:49Z

include/i2c.h

+struct i2c_slave_api {
+	i2c_slave_write_request_t write_request;
+	i2c_slave_read_request_t read_request;
+	i2c_slave_write_done_t write_done;


I find names used in the document you refer Linux I2C Slave support by Wolfram Sang more meaningful. What about re-using them. This function can be called write_received.

Indeed, seems simpler to reuse them.

mnkp · 2017-12-05T01:24:13Z

include/i2c.h

+	i2c_slave_write_request_t write_request;
+	i2c_slave_read_request_t read_request;
+	i2c_slave_write_done_t write_done;
+	i2c_slave_read_done_t read_done;


This one read_processed.

mnkp · 2017-12-05T01:52:07Z

include/i2c.h

+typedef int (*i2c_slave_api_attach_t)(struct device *dev);
+typedef int (*i2c_slave_api_detach_t)(struct device *dev);
+
+struct i2c_slave_driver_api {


I believe we should not do this. This struct with all the related content belongs to the API of device drivers built on top of the new I2C slave functionality (e.g. eeprom), not here. By partitioning API like this we break the Zephyr convention. API of the eeprom driver shouldn't really be different (in design) than API of any other Zephyr driver.

I was thinking of moving this in a separate i2c-slave.h header. What were thinking ?

There are many other device drivers which relay on I2C API. Sensor drivers would be a good example. Yet, we do not place any sensor related functions in the main I2C API. The drivers which relay on the slave part of I2C API, like eeprom, are no different. We should keep APIs of different drivers clean and do not mix them.

Also, the next driver using I2C slave interface may choose to attach itself to the I2C bus driver during initialization phase and do not support dedicated attach/detach functions. I don't think we should enforce any specific API design on the device drivers using I2C slave interface. It's best to simply remove these functions from I2C API.

Ok, I got your point, you think I should move these attach/detach to eeprom.h header and make them specific to the slave driver.

Yes I was thinking about auto-attach drivers.

@tbursztyka what do you think about that ?

I would keep all in i2c.h (slave or master API, it's still i²c APIs anyway)

I don't see how you will register/unregister (these names are nicer than attach/detach actually), if there is no such functions available on the i²c controller side. It's not a slave API here, actually name could be:

i2c_api_register_slave

i2c_api_unregister_slave

that's semantically better.

auto-attach, why not yes.

Question: it's not going to be possible for an i²c controller to serve 2+ i2c slave drivers, here it's one. Would that make sense to actually handle that use case?

If it does, you will have to take a look at how - for instance - I did that in gpio. The callbacks could be in a dedicated, with a sys_node_t in the beginning.
And the address it would serve as well.

So for the controller it would be a matter of going through that list and find a match on the address, to call the right slave callbacks. What do you think?

Also, it would save the need of passing a void *priv (or context) pointer to the callbacks. Just pass the callback sturct pointer itself, with CONTAINER_OF(), up to the slave driver to get the context back, as it's done in gpio.

Yes you are right, the controllers should be able to handle both master and slave at the same time.

Today, the API permits adding any number of slaves, it's a matter of the I2C controller driver to handle that, and it's proven in my i2c_virtual driver implementation.

The STM32 handle at least a maximum of 2 addresses, but I think there is some modes to ack a mask of adresses...

As I've already stated I believe we need to remove this structure and i2c_slave_driver_register(), i2c_slave_driver_unregister() functions from the I2C API. Only i2c_slave_register(), i2c_slave_unregister() should stay.

Ok, but maybe you should decide with @tbursztyka !

In either case, would you want them to go in a separate ic2_slave.h header ?

@tbursztyka what's your opinion on presence of struct i2c_slave_driver_api in this file after my clarifying comments?

@superna9999 I personally wouldn't create i2c_slave.h header file but keep everything here. That said I'm not an authoritative voice here.

mnkp · 2017-12-05T02:48:18Z

include/drivers/i2c/slave/eeprom.h

+#ifndef I2C_EEPROM_SLAVE_H
+#define I2C_EEPROM_SLAVE_H
+
+int eeprom_slave_program(struct device *dev, u8_t *eeprom_data,


This file should resemble a typical device driver API/header file. E.g. it should contain doxygen documentation for the new EEPROM I2C Slave driver.

superna9999 · 2017-12-05T09:39:51Z

Thanks for the initial review, I just pushed an updated version with:

cleaned up all the blocking issues
changed call names aligned with Linux names (read_processed, write_received)
added a new I2C Virtual driver that will communicate only with the attached I2C slaves (up to 128 !)
added Kconfig in the test to add configurations for device name for each target

The I2C slave permit to test the I2C Slave on any boards, but I failed to make it work on qemu_x86 since they use userspace threads by default.

@tbursztyka @mnkp about the I2C Slave Driver API and syscalls, I think it should go in a separate i2c-slave.h header, what do you think ?

mnkp

I have some further comments. However, I haven't been able to complete this review yet.

mnkp · 2017-12-05T22:48:40Z

drivers/i2c/i2c_virtual.c

@@ -0,0 +1,214 @@
+/*


This driver is not really useful for anything else but testing I2C slave EEPROM driver. I think we should move it to the tests/ directory.

OK, but where ?

tests/drivers/i2c/i2c_slave_api ?
Nothing prevents you from creating an i2c driver elsewhere than drivers/i2c

We actually do that in tests/net, creating dummy net device drivers to validate some use cases.

Ok, seems fine for me, I'll move it

mnkp · 2017-12-05T22:58:21Z

tests/drivers/i2c/i2c_slave_eeprom/prj.conf

@@ -0,0 +1,6 @@
+CONFIG_SYS_LOG=y
+CONFIG_SYS_LOG_I2C_LEVEL=4


With level 4 - Debug drivers may not be fully functional, e.g. printing stuff from within ISR. Level 4 is only really useful for debugging. We should change it to 3.

Inded, but for the Virtual I2C, it's can be usefull to have them

mnkp · 2017-12-05T22:59:51Z

tests/drivers/i2c/i2c_slave_eeprom/prj.conf

+CONFIG_BOOT_BANNER=y
+CONFIG_I2C=y
+CONFIG_I2C_SLAVE=y
+CONFIG_I2C_EEPROM_SLAVE=y


We should add here

CONFIG_ZTEST=y

and use Zephyr test framework features within the testcase.

I'll try to use this

mnkp · 2017-12-05T23:03:39Z

tests/drivers/i2c/i2c_slave_eeprom/src/main.c

@@ -0,0 +1,207 @@
+/*


This file needs to be re-written as a testcase, i.e. it should use zassert_* macros such as zassert_true, zassert_equal, etc.

mnkp · 2017-12-05T23:12:41Z

include/i2c.h

+typedef int (*i2c_slave_read_processed_t)(void *priv, u8_t *val);
+typedef int (*i2c_slave_stop_t)(void *priv);
+
+struct i2c_slave_api {


This is not really API of the I2C slave driver. This structure contains a set of callbacks. I think it will be better to call it i2c_slave_callbacks. Current name is a bit confusing.

OK, but "callbacks" does not seem very used in the zephyr codebase, maybe _funcs ?

callback is ok, gpio use it.

I would add _cb in above typedefs (before the _t, so it would be _cb_t )

mnkp · 2017-12-05T23:50:38Z

include/i2c.h

+ * Slave driver implementation.
+ *
+ * @param dev Pointer to the device structure for the driver instance.
+ * @param address Address of the I2C Slave to match on the bus


Let's shorten it to "Address of the I2C slave".

mnkp · 2017-12-05T23:54:00Z

include/i2c.h

+ *
+ * @param dev Pointer to the device structure for the driver instance.
+ * @param address Address of the I2C Slave to match on the bus
+ * @param funcs The Slave API funcs used by the controller to send bus events


What about changing it to "Callback functions used by the I2C driver to send bus events"?

ok, but I'll keep the "by the I2C controller driver" because it's sent by the controller, through the controller

What about "by the I2C bus driver"? We usually talk about controller in a different context.

mnkp · 2017-12-06T00:04:22Z

include/i2c.h

+ * @retval -EINVAL If parameters are invalid
+ * @retval -EIO General input / output error.
+ */
+__syscall int i2c_slave_attach(struct device *dev, u8_t address,


We should use u16_t for the address parameter. Also, let's shorten address to addr to be consistent with existing I2C API.

Yes, I must add the 10bit address support, and a flag is missing here somewhere, maybe in a dedicated struct containing the context, address and flags.

Yes, actually we do need a flag to set the addressing mode.

mnkp · 2017-12-06T00:17:13Z

include/i2c.h

+/**
+ * @brief Attaches the provided funcs and address as Slave device
+ *
+ * This routine starts the I2C Controller in Slave mode and uses


I would write here:

Enable I2C slave mode for the 'dev' I2C bus driver. The I2C slave will be registered at the address provided as 'address' parameter. Addressing mode - 7 or 10 bit - depends on the current configuration of the I2C bus driver. Any I2C bus events related to the slave mode will be passed onto I2C slave device driver via a set of callback functions provided as 'funcs' parameter.

Most of the existing hardware allows simultaneous support for master and slave mode. This is however not guaranteed.

mnkp · 2017-12-06T00:43:30Z

include/i2c.h

+ * @retval -EINVAL If parameters are invalid
+ * @retval -EIO General input / output error.
+ */
+__syscall int i2c_slave_attach(struct device *dev, u8_t address,


We also need to provide a proper _SYSCALL_HANDLER in drivers/i2c/i2c_handlers.c file.

ok, even for the i2c driver ?

For all the functions that are preceded by __syscall.

tbursztyka · 2017-12-06T09:53:31Z

include/i2c.h

+typedef int (*i2c_slave_read_processed_t)(void *priv, u8_t *val);
+typedef int (*i2c_slave_stop_t)(void *priv);
+
+struct i2c_slave_api {


callback is ok, gpio use it.

I would add _cb in above typedefs (before the _t, so it would be _cb_t )

tbursztyka · 2017-12-06T10:00:25Z

include/i2c.h

+typedef int (*i2c_slave_api_attach_t)(struct device *dev);
+typedef int (*i2c_slave_api_detach_t)(struct device *dev);
+
+struct i2c_slave_driver_api {


I would keep all in i2c.h (slave or master API, it's still i²c APIs anyway)

I don't see how you will register/unregister (these names are nicer than attach/detach actually), if there is no such functions available on the i²c controller side. It's not a slave API here, actually name could be:

i2c_api_register_slave

i2c_api_unregister_slave

that's semantically better.

auto-attach, why not yes.

Question: it's not going to be possible for an i²c controller to serve 2+ i2c slave drivers, here it's one. Would that make sense to actually handle that use case?

If it does, you will have to take a look at how - for instance - I did that in gpio. The callbacks could be in a dedicated, with a sys_node_t in the beginning.
And the address it would serve as well.

So for the controller it would be a matter of going through that list and find a match on the address, to call the right slave callbacks. What do you think?

Also, it would save the need of passing a void *priv (or context) pointer to the callbacks. Just pass the callback sturct pointer itself, with CONTAINER_OF(), up to the slave driver to get the context back, as it's done in gpio.

tbursztyka · 2017-12-06T10:03:54Z

include/i2c.h

+	const struct i2c_driver_api *api = dev->driver_api;
+
+	if (!api->slave_detach)
+		return -ENOTSUP;


don't forget the {} ;) (check, but afaik there are many places that require this)

damn, I must really look better for these

@tbursztyka do you mean here ? I see the {}

now I do as well. nevermind

tbursztyka · 2017-12-06T10:23:45Z

include/i2c.h

@@ -175,6 +204,105 @@ static inline int _impl_i2c_transfer(struct device *dev,
 	return api->transfer(dev, msgs, num_msgs, addr);
 }

+/**
+ * @brief Attaches the provided funcs and address as Slave device


register/unregister sounds nicer actually yes.

superna9999 · 2017-12-13T10:41:37Z

Hi @mnkp @tbursztyka

I made a big rework to match all your comments:

renamed to register/unregister
use a single callbacks struct with a slist node like for gpio
renamed callbacks to *_requested
move i2c virtual to test dir
uses ztest now
implemented syscalls handlers (now runs into qemu_x86)

dbkinder · 2017-12-13T18:41:50Z

include/drivers/i2c/slave/eeprom.h

+ */
+#ifndef I2C_EEPROM_SLAVE_H
+#define I2C_EEPROM_SLAVE_H
+


There should be a doxygen defgroup to contain these APIs (and for them to be displayed in the generated API docs)

mbolivar · 2017-12-14T16:01:49Z

tests/drivers/i2c/i2c_slave_api/Kconfig

+	string
+	option env="ZEPHYR_BASE"
+
+config APPLICATION_BASE


I just glanced over this PR without paying much attention to the details, but this APPLICATION_BASE config option should be removed. It's my fault, I added it to the documentation by mistake. I'm trying to fix it here:

#5399

Otherwise I have no issues with this PR

please re-review

deepcube · 2018-07-03T17:34:57Z

@dwagenk Thank you for your help. Your are correct, there is an error condition. It turns out we send a reset command to the peripheral, and the data sheet says that it will not ACK. If I don't reset the device everything works as it should.

However if I do reset the device, ignore the error, then try the next write, it hangs waiting for the semaphore in stm32_i2c_msg_write(). I see that the semaphore is given in the stm32_i2c_event() and stm32_i2c_error() which are called from the combined ISR. What I can't figure out is why this gets stuck. Here is a screenshot of the reset message in a protocol analyzer:

https://i.imgur.com/NDRAH2K.png

You can see the rising edge labeled as stop which is the error, then one after that which should be the stop. Do you recognize what would cause the problem in the restructured code after this?

Again thank you for your help and explanations.

EDIT: noticed I'm using the combined ISR
EDIT2: I'm stepping through the code now and realized there is much that I didn't understand. Hopefully I can figure it out, but I'd still greatly appreciate any direction.
EDIT3: I also realize this is not the proper place for this discussion. I'll take it to the Zephyr mailing list

carlescufi · 2018-07-04T08:41:40Z

@galak @nashif I have just had a meeting with @anangl and @Mierunski. Mieszko will add DTS support for I2C slave once this PR is merged in the next weeks, as part of adding an implementation for our ICs.

carlescufi · 2018-07-04T08:41:52Z

@superna9999 if you rebase this I believe we can merge it now.

In preparation to implementing slave and multi-master capabilities for STM32 I2C V2 driver move the checks, whether all messages of a transfer are valid, before starting the transaction. This prevents having to abort a transmission, that is already partly transmitted. Signed-off-by: Daniel Wagenknecht <[email protected]>

This patch restructures stm32_i2c_v2 drivers Interrupts: - NACK failures trigger an I2C event interrupt, so move handling of NACK failure from error isr to event isr. - Extract logic of interrupt handling to static functions. Use isr functions (event & error OR combined) to call these. This reduces duplication between error isr and combined isr. - Restructure the error interrupt handling, so that it has no effect when no errors occured and thus can be called by combined isr. - Change interrupt logic from if/else to pure ifs for each flag. This reduces code paths and leads to one call of the isr handling multiple interrupt conditions, if there's more than one I2C interrupt flags set. This is the way it's done e.g. in linux kernels (since 4.14) drivers/i2c/busses/i2c-stm32f7.c Signed-off-by: Daniel Wagenknecht <[email protected]>

This patchs adds new I2C Syscalls for : - I2C controllers slave support - I2C Slave drivers support I2C Controllers slave support adds 2 new ops to : - register new slave driver - unregister slave driver Slave funcs consists on all the I2C phases : - read or write request once address matches - read or write done once the byte has been received/sent - stop when the trasmit stops I2C Slave drivers syscall are also added to make new "I2C Slave" drivers to : - register them to their I2C controller - unregister them to their I2C controller Signed-off-by: Neil Armstrong <[email protected]>

Add a new EEPROM simulated driver with all it's build infrastructure. The EEPROM can be loaded/poked from applications by getting its funcs. It is multi-instance capable, right now 2 instances are supported by enabling them in KConfig. Signed-off-by: Neil Armstrong <[email protected]> Signed-off-by: Daniel Wagenknecht <[email protected]>

Add a new test to make I2C transfers between an I2C controller and an I2C slave driver, either by: - Using a virtual I2C driver - Using a physical link and 2 physical controllers The Physical test uses : - Two EEPROM Slave drivers each connected to one I2C controller - Both I2C controllers also act as master and take turns writing to the opposite EEPROM Slave. All of this connected physically on the board. Signed-off-by: Neil Armstrong <[email protected]> Signed-off-by: Daniel Wagenknecht <[email protected]>

This patch adds I2C Slave support conforming to the syscalls and funcs, only for the STM32 V2 I2C Driver for the moment. It is capable of handling multi-master bus setups. Signed-off-by: Neil Armstrong <[email protected]> Signed-off-by: Daniel Wagenknecht <[email protected]>

superna9999 · 2018-07-04T08:46:02Z

@carlescufi Just did, thanks @Mierunski for taking over, I'll have some time to review it anyway, so you can ping me and add me to the PR

carlescufi · 2018-07-04T09:57:29Z

Merging after TSC meeting yesterday and update with @Mierunski today.

The existing sam0 i2c driver does not support slave mode. This commit enables slave support for sam0 devices. The Slave support is based on Linux I2C Slave support by Wolfram Sang and documented at https://www.kernel.org/doc/Documentation/i2c/slave-interface The code for this commit is largely based on code taken from PR zephyrproject-rtos#5224 and the stm mcu implementation. This was modified to work with the sam0 SERCOM perhiperal and integrated into the sam0 driver. Fix whitespace and comments Signed-off-by: Steven Slupsky <[email protected]>

The existing sam0 i2c driver does not support slave mode. This commit enables slave support for sam0 devices. The Slave support is based on Linux I2C Slave support by Wolfram Sang and documented at https://www.kernel.org/doc/Documentation/i2c/slave-interface The code for this commit is largely based on code taken from PR zephyrproject-rtos#5224 and the stm mcu implementation. This was modified to work with the sam0 SERCOM perhiperal and integrated into the sam0 driver. A slave must wait for data ready when there is an address match to ensure the data is valid. Disable SCLSM (SCL stetch mode) when operating in slave mode to ensure that i2c ACK's are handled properly. Fix k_timeout_t. Signed-off-by: Steven Slupsky <[email protected]>

superna9999 requested review from nashif and SebastianBoe as code owners December 1, 2017 11:22

tbursztyka requested changes Dec 1, 2017

View reviewed changes

SebastianBoe reviewed Dec 1, 2017

View reviewed changes

superna9999 mentioned this pull request Dec 1, 2017

I2C Slave Driver #2956

Closed

galak requested review from MaureenHelm and mnkp December 1, 2017 15:35

superna9999 force-pushed the for-pr/stm32_i2c_slave branch from f10c340 to a7b83d7 Compare December 4, 2017 14:36

mnkp requested changes Dec 5, 2017

View reviewed changes

superna9999 force-pushed the for-pr/stm32_i2c_slave branch from a7b83d7 to 45fbf74 Compare December 5, 2017 09:35

superna9999 force-pushed the for-pr/stm32_i2c_slave branch from 45fbf74 to b5b0614 Compare December 5, 2017 10:19

mnkp requested changes Dec 6, 2017

View reviewed changes

tbursztyka previously requested changes Dec 6, 2017

View reviewed changes

galak added the area: I2C label Dec 6, 2017

superna9999 force-pushed the for-pr/stm32_i2c_slave branch from b5b0614 to 5d7f3eb Compare December 13, 2017 10:39

superna9999 requested review from dbkinder, erwango and mbolivar as code owners December 13, 2017 10:39

superna9999 changed the base branch from master to arm December 13, 2017 10:39

superna9999 force-pushed the for-pr/stm32_i2c_slave branch 3 times, most recently from 0bb7082 to 5c913bf Compare December 13, 2017 12:13

dbkinder suggested changes Dec 13, 2017

View reviewed changes

superna9999 force-pushed the for-pr/stm32_i2c_slave branch from 5c913bf to a80b72f Compare December 14, 2017 10:29

mbolivar reviewed Dec 14, 2017

View reviewed changes

superna9999 force-pushed the for-pr/stm32_i2c_slave branch from a80b72f to b0eb6ff Compare December 14, 2017 18:57

nashif removed request for sjanc, Vudentz and andyross July 2, 2018 13:03

Daniel Wagenknecht and others added 6 commits July 4, 2018 10:43

superna9999 force-pushed the for-pr/stm32_i2c_slave branch from c784a97 to 12ee47a Compare July 4, 2018 08:44

carlescufi merged commit c7875b7 into zephyrproject-rtos:master Jul 4, 2018

This was referenced Jul 4, 2018

i2c: i2c_ll_stm32_v2: fix NACK handling for STM32L4 #7626

Closed

Add support for I2C on STM32F7 family and on STM32F723E-DISCO board #8716

Merged

sslupsky mentioned this pull request Apr 2, 2020

drivers: i2c_sam0: Add i2c slave support #24047

Closed

sslupsky mentioned this pull request Apr 2, 2020

drivers: i2c_sam0: Optimize and add i2c slave #24050

Closed

		@@ -0,0 +1,6 @@
		CONFIG_SYS_LOG=y
		CONFIG_SYS_LOG_I2C_LEVEL=4

Add support for I2C Slave Drivers #5224

Add support for I2C Slave Drivers #5224

Conversation

superna9999 commented Dec 1, 2017

superna9999 commented Dec 1, 2017

tbursztyka left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mnkp left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

superna9999 commented Dec 5, 2017

mnkp left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment