Hi all,
we would like to split our processor in 2 cores: one for functional safety and one for QM software. Therefore we're using the S32K358
Regarding the multi-core capabilities of the RTD, a couple of questions arose.
Referenced document:
- Application Note "Multi-Core Implementation in RTD of S32K3" (Rev. 1, 07/2022)
1) General questions regarding RTD multi-core terminology
In this AN, a difference is made between the "service API" and the "module kernel". Especially for Type II and Type III modules I am not sure, what exactly is meant.
An example is given with the module kernel function being stated as "init", "deinit". But is there an actual list what functions are allocated to which part?
2) "module kernel" displayed as running on both cores
When looking at Type II and Type III, the "module kernel" is displayed as one unit over both cores. What does this mean looking at integration/code level?
Is this a reference to the scope ("global" vs. "local") at least of the code/constants part being shared between the cores, in addition for Type III even the variables?
3) "codes/constants" definition
For both Type II and Type III, "codes/constants" are shared globally. Is it correct to asume, that the config structure (also somewhat constant) passed to the init function is NOT part of this but basically the code from the RTD directory?
4) Global sharing
This means both cores can use the same code in the flash to run the RTD, i.e. they execute from the same flash address (i.e. linker section)?
So when providing the same code twice (in each flash section of the corresponding core), this should also be ok?
Thank you
Andreas
