Scheduler doubt

nobody wrote on Thursday, January 11, 2007:


When in the Test results you have:

Scheduler tick function
66.2μs with optimization. 68.4μs with no optimization.

That is the time the secheduler takes to make the context switch between two threads running?


rtel wrote on Thursday, January 11, 2007:

As described on the WEB site: "The saving and restoration of a task context is not included within the measurement. Each microcontroller requires a different set of registers to be saved meaning a comparison is not directly possible. The PUSH and POP test detailed in the next section can be used as a guide to how efficient the context switch itself may be."  Therefore it is the time taken to run the function that selects the task but excluded the push and pop of the context.


imajeff wrote on Monday, January 15, 2007:

For these vague speed questions, which port are you even talking about? It sounds like you haven’t even picked one, so you’ve got no context, and no wonder you don’t know how fast it is.

I hope this is not a test question because you’re going to miss that one for sure.