I have a built-in device ( Technologic TS-7800 ) that advertises real-time capabilities, but doesn't say anything about “hard” or “complex”, soft '. While I was waiting for a response from the manufacturer, I thought that it would not hurt to test the system myself.
What are some established procedures for determining the “rigidity” of a particular device with respect to real-time / deterministic behavior (latency and jitter)?
While in college, I have access to fairly tidy hardware (good oscilloscopes and signal generators), so I don’t think that I will have problems testing equipment, just experience.
I have the same board here at work. This is a slightly modified 2.6 Kernel, I believe ... not in real time.
I do not know that I read anything in the documents, but this indicates that it is designed to work with RTOS.
With the help of such equipment, it is easy enough to synchronize the o-frame with a stable watch, to burst every time the real-time system exits, and see how much this spike changes from the center. The smaller the variation, the greater the hardness.
, :
. .
( ), .
RTOS , .
RTOS- 2 .
/. A, B., .
workd, . . , . . ., SD SOFTNESS. ( , , .) , RTOS . . , . RTOS , 1/f. , . , , - . , , . .
, 1uS , 0.5us, .
, (, , , , , ).
, .
, , . , - , ( , RTOS ). , .
, , (, ), . /-, / .
, , , .
, , . , , . , , .
, . RTOS , , , .
Edit after comment, ( , ), - . , RTOS-, . . 50%, - 20% , . . . .
Edit after comment
- . reset . , . , , , , , .
, . , , , .
,
, , RTOS.
, , Ethernet/usb/ () .
.
( ) - , - . .
, , . , .
, - Linux Win CE, , .
, , , , . , . , , , . .
PS: The conclusion is that even for critical systems you really do not need a hard real-time mode if you have hardware.