Mystery Startup Failure

I'm using a Sharp LH79524 ARM7 microcontoller along with the IAR compiler 
and Segger's J-link. I was developing my program yesterday and it downloaded 
to the target and ran fine but this morning I have downloaded it and it gets 
stuck on the data abort exception. This happened with the project I did on 
Wednesday and I never managed to get it to load again. Both projects are 
based on a smaller working project and grew in size (with the addition of 
several bitmaps) during the day but still downloaded fine until I tried them 
the following day.

Another thing I did notice about these programs as they grew was that they 
suddenly began to kick up a prefetch exception when reset using the 
debugger. Intitally the program would correctly reset back to main().

The code and data sizes are as follows:
46 640 bytes of CODE  memory
1 451 401 bytes of DATA  memory
176 124 bytes of CONST memory

But I am using a LogicPD development board with 32Mb or RAM and (I think) I 
have the linker set up to use the upper 16 for data and lower 16 for code 
and constants so there should be plenty of room. I've put the interrupt 
vectors in the lower 16 as well, rather than using the MCU's RAM but I'm not 
sure if that makes a difference.

Can anybody help me? 

Additionally, I've found that if I hold the reset switch on the board until 
that last second before the download starts (preventing the LogicPD 
bootloader from starting) then it works. However, using the debug reset now 
hangs the system and the disassembly ends up near the top of the ARM address 
space in the E0000000's.


"Tom" <tlucasremoveall@thisextragubbinstoreplyautoflame.co.uk> wrote in 
message news:433d2459$0$15077$ed2619ec@ptn-nntp-reader02.plus.net...
> I'm using a Sharp LH79524 ARM7 microcontoller along with the IAR compiler 
> and Segger's J-link. I was developing my program yesterday and it 
> downloaded to the target and ran fine but this morning I have downloaded 
> it and it gets stuck on the data abort exception. This happened with the 
> project I did on Wednesday and I never managed to get it to load again. 
> Both projects are based on a smaller working project and grew in size 
> (with the addition of several bitmaps) during the day but still downloaded 
> fine until I tried them the following day.
>
> Another thing I did notice about these programs as they grew was that they 
> suddenly began to kick up a prefetch exception when reset using the 
> debugger. Intitally the program would correctly reset back to main().
>
> The code and data sizes are as follows:
> 46 640 bytes of CODE  memory
> 1 451 401 bytes of DATA  memory
> 176 124 bytes of CONST memory
>
> But I am using a LogicPD development board with 32Mb or RAM and (I think) 
> I have the linker set up to use the upper 16 for data and lower 16 for 
> code and constants so there should be plenty of room. I've put the 
> interrupt vectors in the lower 16 as well, rather than using the MCU's RAM 
> but I'm not sure if that makes a difference.
>
> Can anybody help me?
> 

Tom wrote:

> Additionally, I've found that if I hold the reset switch on the board until 
> that last second before the download starts (preventing the LogicPD 
> bootloader from starting) then it works. However, using the debug reset now 
> hangs the system and the disassembly ends up near the top of the ARM address 
> space in the E0000000's.

starting to sound like power.
volts within specs, and well filtered?
A nice big electro cap may make a difference.


Don...

-- 
Don McKenzie
E-Mail Contact Page:       http://www.e-dotcom.com/ecp.php?un=Dontronics

USB,RS232 or TTL to VGA Monitor http://www.dontronics.com/micro-vga.html
World's smallest USB 2 TTL Conv http://www.dontronics.com/micro-usb.html
USB to RS232 Converter that works http://www.dontronics.com/usb_232.html