Intel® Makers
Intel® Edison, Intel® Joule™, Intel® Curie™, Intel® Galileo
Welcome - This is a Peer-to-Peer Forum only. Intel has discontinued these products but you may find support from other customers on this Forum
9873 Discussions

Test and Verification of Intel Joule, Ostro, and mraa

New Contributor II


is there a list,and or document which states what mraa API operations work successfully with which hardware peripherals? And list which ones do not. Thanks in advance.

Is there an Errata sheet?

As an embedded system, the community expects the hardware features for the Joule, and software feature to operate.

It is concerning, that portions of the hardware and software are not complete, and are detrimental to simple embedded system product design/implementation.

Please list all of the hardware features for the device, and the associated current status.

Also provide test scenarios, results, and test code, such that your results can be verified by others. Thanks in Advance.

If these have not been tested, a simple statement would be adequate.


GPIO x successfully tested as an input and an output.

UART x successfully tested with the following setting, and associated code.

SPI x successfully tested with the following setting, and associated code.

WatchDog successfully tested with the following setting, and associated code.

Real-time clock

Non Volatile







3 Replies

I second this. Some things work, some don't, and and I'm battling minutiae instead of writing product code.

The Intel page for MRAA ( mraa: Main Page ) claims support for the Intel Joule under GNU/Linux . It would be incredibly useful to have a table of what works and what doesn't.

I have a design that needs i2c, GPIO, PWM, and am detouring into writing buckets of test code to determine how much of it is or is not functional. This activity is not conducive towards meeting project milestones. It also gets awkward when the hardcore HW design engineer says, "It sez right here (points at Intel docs) this is available." and I have to point at community FORUMS to say, "It's not."

Forums are more like... "tribal knowledge".

I don't have sufficient equipment to check all the devices for which support is claimed for the Joule 570x and under which OS. Wish I did. And the TIME to look.

Could there be a table with function name on the left, OS along the top, and symbols indicating functionality? In the process of creating that table, test code would need to be written. Please, don't put it into a dark corner. Link it to the table so we poor souls have a guidepost.



Community Manager

Hi guys,



Thank you for contacting us. I think what you're asking is very important to the community. As you know, the Joule is a new product and we're working to have more documentation and software releases that will really make the customer experience easier. We always take the community users input and ideas on how we can improve and pass all the valuable feedback to the team that would be able to make it happen. At the moment, we don't have a document such as the one you're requesting, but, we'll make sure to pass your recommendation so hopefully something like this can be published. The Joule documentation available as of yet can be found here , as well as links to downloads and sample codes.





-Sergio A


New Contributor II


thanks for the update. But this response provides little or no benefit for the issues stated.

I think what surprises most me regarding themultitude of questions and responses on the forum, is how many Intel individuals appear to be involved with the discussion. Yet how few of them appear to have actually used the device to build anything, or test anything.

Any enthusiastic "Hacker Lab" builder would have already connected up to all of your GPIOs, I2C, SPI, Uarts, PWMS, WiFi, BLE, ..., just like they currently develop with when on an Arduino, Raspberry Pi, Microchip, NXP, ....

They typically would up and running building their custom application in a short time frame. And spending quality time with "what makes their device unique", and not the issues associated with the fundamental foundation of the hardware, documentation, APIs, and tools.

We just expect Intel to at least be at par with Arduino.

I'd recommend management, and the Intel engineers actual spend some time learning and "using" the system, and or getting out in the real world, and listening, to the voice of the client for the clients needs.

Best regards.