rtel wrote on Monday, December 02, 2013:
I have not seen this before - it is a giggle ;O)
“No mechanisms to counter priority inversion are implemented.”
Mutexes include a priority inheritance mechanism. So that is not true - maybe it was when it was written.
“Overall, FreeRTOS was determined to be slightly too feature-rich for limited resource embedded devices”
That contradicts the authors other assertions that it could do with including both a priority inversion and a deadlock prevention scheme - which would of course make it more feature rich. It is in fact massively more feature rich now than when the paper was written. Especially with the introduction of event groups/flags and a centralised deferred interrupt mechanism in the soon to be released V8.
“FreeRTOS implements co-routines. These are not considered in this document as they
are duplicative of the existing functionality”
This also contradicts other statements, as co-routines are (were) designed specifically for the tiny platforms.
I’m afraid I didn’t read the rest, just a few snippets, but wonder why you are reading a really old document like that rather than the wealth of up to date information that is freely available to you?