I want to force a reboot if a serious error occurs. I have tried the _trap_(0x00); function which inserts the SRST instruction but nothing happens. In debug mode i get an error message in the output window - Error occured during device access.
What does the mean and how can i force a reboot?
Any help appreciated.
Err - No - The debugger was not surprised, even if the hardware runs without the debugger, still it does not reset. The only clue i have is (as i said before) when debugging and the reset command is executed, i get 'device access error' in the output window.
Maybe a little more help rather than picking faults in my grammer might be better.
So just to clarify, when running standalone and the _trap_(0x00) command is executed, the chip does not appear to reset. I expected the chip to reboot the code and start again. This is not the case.
Any helpful, non sarcastic advice welcome.
What about not servicing your watchdog?
Hi, Im not using the Watchdog.
1) Do you have an way to monitor the /RSTOUT pin? 2) Could you verify that the _trap_(0) generates the SRST instruction (it should but just to be sure it is not just a jump to 0)?
From the manual: The Reset Output RSTOUT provides a special reset signal for external circuitry. RSTOUT is activated at the beginning of the reset sequence, triggered via RSTIN, a watchdog timer overflow or by the SRST instruction. For internal resets the activation of RSTOUT can be disabled. RSTOUT remains active (low) until the end of the reset sequence, until disabled by user software, or latest until the EINIT instruction is executed.
This should give you an indication if the device going through a reset.
Im not using the Watchdog.
How about using it? That is, the implementation of your soft reset function would enable the watchdog, program it for the shortest interval you can, and then spin, waiting for the WDT to reset the chip.
Thanks guy's ...