| View previous topic :: View next topic |
| Author |
Message |
a3ka2000
Joined: 26 Feb 2010
Posts: 11
|
| dsPIC30F6014A restart? |
 Posted: Wed Mar 17, 2010 12:27 am |
 |
|
Hi, I have a problem with dsPIC30F6014A used for TCP/IP communication with ENC28J60.
I'm implementing the code based on CCS Ethernet Development Tutorial, ex-7b. The board seems to restart by itself when coming to access this following struct:
| Code: | | __mac_tx_buffers[NICCurrentTxBuffer].len=len; |
The declaration for that struct is as follows:
| Code: | struct {
int8 isReserved;
int8 isActive;
int8 isFlushing;
int16 len;
} __mac_tx_buffers[MAC_TX_BUFFER_COUNT]; |
How do I know it was restarting is because it keeps going back to the start without executing the remaining portion after the abovementioned part. This is funny because when I used MPLAB debug mode to animate the code, the process ran smoothly.
Is there anyway I can prevent restart? I already disabled watchdog (NOWDT) by using #fuses.
thank you very much
best wishes |
|
 |
FvM
Joined: 27 Aug 2008
Posts: 2337
Location: Germany
|
|
| Posted: Wed Mar 17, 2010 12:52 am |
|
|
You may want to check the restart cause by reading the RCON register.
Generally, address error trap and stack overflow trap are popular reset causes with PIC24/dsPIC. See the respective forum discussion. |
|
|
a3ka2000
Joined: 26 Feb 2010
Posts: 11
|
|
| Posted: Wed Mar 17, 2010 1:24 am |
|
|
| FvM wrote: | You may want to check the restart cause by reading the RCON register.
Generally, address error trap and stack overflow trap are popular reset causes with PIC24/dsPIC. See the respective forum discussion. |
Hi, thanks for the suggestion. Let me try to take a look. Most prob it's not caused by voltage drop, unregulated PSU, or watchdog timer.
If I replace NICCurrentTxBuffer with a fixed number, say, '0', the the program runs okay without reset occuring (but still incorrect TCP/IP response).
best rgds |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
