spi_master vs spi_master_5w?

[Mike Wirth]

Folks,

In the latest SDK (6.1) for the nRF51822 I see two versions of SPI master code in the Source folder and examples: with and without "5W" appended to the filename.

Looks like the 5W version has more features (and complexity) and is designed for "serialization", I.e., the ability to put to sleep and awaken a running device state. Is that the correct interpretation? Is the 5W version ready for use (or just an intermediate development version)?

In either case, the SDK 6.1 online docs seem confused as to version. I would appreciate some advice on which version to use and documentation on same.

Also, looks like the (non-5W version of the) SPI Master driver assumes chip select is active low. I have a device that's active high and I can modify your driver code, but would prefer not to do that. How about adding chip select high vs. low to the configuration params for the SPI open call?

Thanks,

Mike

[Asbjørn]

The 5W example is a SPI serialized version of the HRS example. The latest update about this can be found here, but it's pretty much the same as in SDK 6.1

The serialization examples are made to just send data to the nRF51 through UART/SPI and the nRF51 will handle all the BLE traffic and your co-processor doesn't need to handle anything. If you're looking for just a SPI example I would have a look at the non-serialized examples, but rather the spi_master_example.

Regarding your request, I can't say whether or not we will or when it then would be implemented, nevertheless, I've reported it internally to our SDK team.

[Mike Wirth]

Thanks, Asbjorn and Maki. Answers my question. BTW, in the Doxygen-generated documentation for functions, it would be very useful if in describing parameters, the acceptable range of values is included, as well as an example. For enums I can trace back the definition and find the acceptable values, but for other types it's often not clear (e.g. when a return is specified as unint32_t and it's an error code). Mike