I am creating a bootloader project and an application (firmware) project to run on our STM8s, which communicate over a CAN network. I got some very helpful advice earlier from Bruno, and I ended up needing to share my interrupt between the Bootloader and the Application, but the actual interrupt code is in the Bootloader project. The code looks something like #define CAN_GetDataFW (*((void(*)(void))0xB500)) void CAN_RX_IRQHandler (void) interrupt 8 { if(appStart) { CAN_GetDataFW(); } else { CAN_GetData(); } } So if in the bootloader portion of the code, CAN messages are handled with CAN_GetData in the Bootloader project. If the application has started, CAN messages are handled with CAN_GetDataFW in the Firmware project, which has an absolute address to keep the two projects separate. Now, CAN_GetDataFW looks something like: at 0xB500 void CAN_GetDataFW(void) { .....process CAN commands.... main_fw2(); } where main_fw2 is just the main waiting loop that the chip starts in after initialized, and waits in until it receives the next CAN command. I have to include this call because there is a problem where the code will never return back to the Interrupt CAN_RX_IRQHandler. If I placed a piece of code directly after CAN_GetDataFW(); in if(appStart), the program will never reach that line, it will just freeze. I had thought I had bypassed the problem by going directly back to the waiting loop before reaching this problem point. This solution, while messy, had seemed to work, although now I realize that the chips also freeze after about 20-40 iterations of CAN_GetDataFW(). I now believe THIS issue might be caused by never actually finishing the CAN_GetDataFW function, so the function and all the variables within it are never undeclared. The RAM is slowly filling up with these copies of CAN_GetDataFW and everything in it, until after 30 repetitions or so, the RAM is full and the chip freezes. I can't confirm this because, with the Bootloader and Application in 2 different projects, I cannot use Ride7's debug feature with the SWIM cable to check registry values. So, I was wondering if anyone had a suggested fix for this, which I believe could be solved in one of three ways: 1) Is there a fix to the problem about the program not able to return to "CAN_RX_IRQHandler (void) interrupt 8" after finishing "CAN_GetDataFW"? I am sure this is because CAN_GetDataFW is at an absolute address, but I don't know if there is a known problem and if there is a workaround 2) Is there a way to declare CAN_GetDataFW as a "static" function, but not the normal "function static" in C, which just makes a function local, but the "variable static", which causes a variable not to allocate memory when called multiple times. 3) Is there a way to just clear everything on the RAM, or something similar, that doesn't solve the problem of CAN_GetDataFW from being re-initialized everytime, but somehow prevents the memory from filling up and the base from freezing after 30 iterations. It doesn't need to be pretty, it just needs to work indefinitely. Thanks for any help you can suggest, -Dan