some error information

Jogging Song said:

The first one is "The CIO command cmd:(11) in the CIO buffer at address (0x808b0d00) was not recognized.
Please check the device and program memory maps." In the map file,
I find there is the section and the section begins with the starting
address 808b0d00. Is there some specific content at the address?
My application changes it, so it can't be recognized.

The CIO command error is related to C i/o operations such as printf, fopen, etc, essentially it contains information related to communication with CCS on the host PC. If this is getting corrupted it could mean that you have a lose pointer writing bad data into the CIO buffer, or your stack could be overflowing into it depending on your memory map.

Jogging Song said:

The second one is that the PC points to UTL_halt when the program
hangs at some place and don't move on. In the document of DSP/BIOS API
it says that it belongs to SYS module. If PC points to UTL_halt, what
can I get from this?

Ending up in UTL_halt could be caused by many conditions, it is where BIOS ends up at if you have some bad event, such as making a blocking call when the scheduler is disabled (for example blocking in an ISR).

Thanks, Bernie
     It seems that these information just indicates that something is wrong in program.
No other information is available. To solve the problem, experience is very important.
     Why not provide more information? I guess some debug functional module is removed
in order to reduce response time.

Best Regards
Jogging

In my program, rand from RTS is called for generating data.
If I replace rand with a constant such as 1, CIO error will disappear.
However, I need rand.

Another solution is that I disable Real-time Analysis, RTDX and TSK
Manager when adding DSP/BIOS.
But I still know the exact reason that causes the CIO error.

Best Regards

Jogging