This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Hello,
I'm working on an embedded system based around the LM3S6965, using CCS4.2.3.00004 with a Spectrum Digital XDS510USB+ emulator. Most things work: I can load code, use breakpoints, debug most of my code, etc., but one block of code always raises the same CCS error.
First off, here's the code:
*((unsigned long *) 0x400FD000) = addr; // FMA register, set address
*((unsigned long *) 0x400FD008) = 0xA442002; // FMC register, start erase
while (*((unsigned long *) 0x400FD008) & 0x02) // Poll erase status until done.
;
All this does is clear the flash memory page given by addr. And it works. However, every time it runs, CCS complains with this error message:
Spectrum Digital XDS510USB Emulator With SdOpts_0/CORTEX_M3_0
Error:
Error 0x20000020/-1311
Severe Error during: Execution,
CPU stall due to LOAD/STORE detected...attempting recovery...
Stall recovery and target halt were applied successfully!
Reccommendation: Normal use of debugger can resume.
[Cancel] [Disconnect]
If I click Cancel, the program execution breaks on the third line (while ...). Then I can click the continue button to continue debugging. The specified page is in fact cleared, but my code has to clear sometimes up to 128 pages in memory, and at that point it's unreasonable to click cancel, continue, cancel, continue, ..., 128 times over.
Running that code on the processor without the debugger works fine.
I noticed a similar error on e2e here, but I didn't find any suggestions to help my case. It doesn't even seem like anything is even going wrong, but is it possible to either a) fix it or b) mask the error message and make the debugger automatically continue?
Thanks,
Chris
Hello Chris,
I can't seem to reproduce this with my LM3S6965 + SD XDS510USB+ configuration. I tried out your snippet of code and it was fine. I am using 4.2.4 but I'm not sure if that is the reason. You can try and update to 4.2.4 and see if that helps.
Would it be possible to get a full reproducible test case (project + source + executable)? Also provide your target ccxml file (and GEL file if you are using a custom GEL).
Thanks
ki