Hello,

 

I'm a newbie both to uClinux and embedded programming. We have an embedded
device that runs uClinux (2.4.x) with Coldfire 5272 processor without MMU.
The firmware is stable for a few years. Recently, we are porting the
firmware to Coldifre 5275 processor. We've got the kernel and firmware
running. During a stress test we find that firmware crashes. I did some
research on this and found some suggested fix for this is to increase the
stack size, we increased the stack size upto 64k and the problem persists,
the default stack size for our firmware was set to 8k and that is stable for
5272 processor. I did some experiments and finally found that if I add a
routine (to print some debug info when the exception occurs) to traps.c of
kernel and call that routine from trap_c function then crash goes away and
we ran the stress test for ovenight. The weird thing is that trap_c is never
called since the firmware doesn't crash. I think this sort of problem common
to embedded programming. Could you provide some hints and tips to understand
the reason of this kind of problem and possible fixes?

 

Thanks,

Hasan

_______________________________________________
uClinux-dev mailing list
[email protected]
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by [email protected]
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

Reply via email to