after 12 Hours, 5223X demo baord hangs, no response from CPU

キャンセル
次の結果を表示 
表示  限定  | 次の代わりに検索 
もしかして: 

after 12 Hours, 5223X demo baord hangs, no response from CPU

702件の閲覧回数
TVNAIDU
Contributor III


I am using MCF5223X (Coldfire) based demo baord with tiny web application (keep refresh for some readings from GPIO), after 12 Hours of running, board locksup, any idea how to find why board hangs?. no response from console also, CPU clogs, appreciated for any inputs.

ラベル(1)
0 件の賞賛
返信
3 返答(返信)

567件の閲覧回数
FridgeFreezer
Senior Contributor I

Is there any sort of watchdog timer or any other timer running? Is there anything storing variables/data in RAM that might run out of space?

 

You don't really give enough information about the system/software, what you've done already to try and find the fault, etc. Have you got a debugger you can pause the system and look at what's going on (maybe after ~11 hours?), make it print out status messages on the console at regular intervals, etc.?

0 件の賞賛
返信

567件の閲覧回数
J2MEJediMaster
Specialist I

Are you using an interrupt handler? If so, is it cleaning up the stack properly when it exits? If it does not, this creates a "leak" where the stack slowly expands in memory until it stomps on something critical. Easiest way to confirm this is to increase the stack size (if possible) and if the board runs longer, that's your culprit.

 

Or, as was suggested earlier, halt the board at 11 hours and snoop around with the debugger to check out where the top of the stack is, among other things.

 

---Tom

0 件の賞賛
返信

567件の閲覧回数
FridgeFreezer
Senior Contributor I

You could also try halving the clock frequency (if possible) and see if it takes twice as long to lock up.

0 件の賞賛
返信