We are working on a T1042D4RDB; we want to use the i2c bus.

[bkoball]

Our h/w FAE confirms T1042D4RDB i2c port 3 is unused; we've tapped it at the pullups R100 (IIC3_SCL) & R102 (IIC3_SDA), but can't find any doc's on kernel drivers or i2c function calls. We are writing small test routines in c and compiling on the target using gcc, Any advice on where to find docs on i2c drivers and function calls?

[bkoball · Accepted solution]

Following up on the chance anyone else encounters this problem:

Pavel's mention of the device tree config (.dts) pointed us in the right direction.

It turned out that the Reset Configuration Word (set in u-boot) was assigning the unused i2c port to GPIO.

With some help from both our h/w and s/w FAE's we're now happily talking to external i2c devices with our T1042D4RDB.

thanks NXP!

[Pavel]

This port is enabled in .dts file. See attachment.

Look at the following pages for I2C using under Linux:

https://www.kernel.org/doc/Documentation/i2c/dev-interface

http://www.frank-buss.de/io-expander/linux-i2c-test.c

http://lxr.free-electrons.com/source/Documentation/i2c/dev-interface

Find also "Thermal Monitor User Manual" on the QorIQ SDK 1.9 Documentation page:

https://freescale.sdlproducts.com/LiveContent/content/en-US/QorIQ_SDK_1.9/GUID-52C420E4-E67D-4D0D-84...

Source I2C tools is available using the following path:

/QorIQ-SDK-V1.9-20151210-yocto/build_t1042d4rdb/tmp/work/ppce5500-fsl-linux/i2c-tools

Have a great day,
Pavel Chubakov

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[bkoball]

thanks for the pointers.

This port is enabled in .dts file. See attachment.

I'm not seeing any attachment to this email.

We've tapped the SCL and SDA lines of the unused i2c port (i2c-2)

at the pull-ups on the RDB, monitoring them on a scope, and scanning

the port with "i2cdetect -r 2", but we don't see any activity on the

lines.

I think we should at least see the processor transmit a slave

address before the transaction times out, even if there's no other

i2c device on the bus, yes?

any other suggestions welcome.

best,

-brk-

Bruce R. Koball Motion West (direct) 510 845-1350

bkoball@well.com 2210 Sixth St (studio) 510 845-3946

bkoball@motionwest.com Berkeley, CA 94710