Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Interrupt latency estimation

Status
Not open for further replies.

alzomor

Advanced Member level 2
Joined
Jun 9, 2005
Messages
678
Helped
39
Reputation
78
Reaction score
8
Trophy points
1,298
Location
Cairo
Activity points
6,417
Hi

How to calculate the exact interrupt latency for ARM926EJ-s?
I have found Some info in ARM Documentation for FIQ , Vectored IRQ and none Vectored IRQ but I think it's for the worst case Interrupt latency?
My Question is: Is it possible to know the exact interrupt latency?

Salam
Hossam Alzomor
www(.)i-g(.)org
 

vaddavalliseshu

Newbie level 6
Joined
Aug 28, 2007
Messages
14
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,381
LATENCY= intrrupt event occurred time - actual event time
 

alzomor

Advanced Member level 2
Joined
Jun 9, 2005
Messages
678
Helped
39
Reputation
78
Reaction score
8
Trophy points
1,298
Location
Cairo
Activity points
6,417
Hi

I was asking about exact value in ARM926EJ-s
I was not asking about the definition, even that your definition is not clear.
a good definition could be ,
time of the execution of 1st instruction in interrupt service routine - the time of the interrupt itself.

thanks any way

Salam
Hossam Alzomor
 

Enlightenment

Full Member level 3
Joined
Jul 26, 2006
Messages
174
Helped
29
Reputation
58
Reaction score
23
Trophy points
1,298
Location
3rd Rock From Sun
Activity points
2,671
Not an ARM calculation, but an old-school approach that works on any processor: at the beginning of your interrupt function, pulse a output pin high then low.

Hook up a logic analyzer to the interrupt line and the output line you are pulsing and measure it. If you can capture a lot of results and have abilty to dump timings as text to a file, then examine all to calculate the min, max, avg. This will be helpful to determine jitter.

You should be able to look at the interrupt source code, if in C, then subtract the instruction timing from where your instruction pulses the line to the beginning of the interrupt code from the above time measurements to come up with the answer.

People want nice pretty answers from a book...but it is hard to beat real-world measurements.
 

Status
Not open for further replies.
Toggle Sidebar

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top