RTOS/TM4C1294NCPDT: Locating embedded application to specific address

Hi Harshal,

There is an example that covers much of what you will need:
processors.wiki.ti.com/.../TI-RTOS_USB_DFU

Theexample is about Device Firmware Upgrades over USB but it covers bootloaders and specifying application boundaries in CCS. You can simply ignore the USB parts in the USB_Serial_DFU.pdf.

As for jumping between applications, you will need to define the c_int00 addresses of each of your apps and either use an inline branch instruction or a function call on a function pointer that points to one of these addresses. I hope that covers most of your questions, let me know if you need further clarifications.

Thanks,

Sean

No problem Harshal. Your RTOS applications will be configured to place their vector tables in their own flash space per the example documentation. So there is no need to modify their startups. If you did want to modify start up for TI RTOS apps, there is an XDC runtime module called Startup which allows users run their own code during startup before Main(). The documentation for startup (xdc->runtime->startup) can be found in the SYS/BIOS documentation:

http://software-dl.ti.com/dsps/dsps_public_sw/sdo_sb/targetcontent/sysbios/6_75_01_05/exports/bios_6_75_01_05/docs/cdoc/index.html

Note you should go to the docs for the version of SYS BIOS you are using. This version number should be in a manifest in your SDK.

For your scenario #1, the first app is placed at 0x0 and runs until it jumps to your next app (let's say you placed it at 0x4000) which then runs as usual. For jumping to your next application, you can do what the bootloader does with inline assembly:

    //
    // Load the stack pointer from the application's vector table.
    //
#if (APP_START_ADDRESS != VTABLE_START_ADDRESS)
    asm(" movw    r0, #(APP_START_ADDRESS & 0xffff)");
#if (APP_START_ADDRESS > 0xffff)
    asm(" movt    r0, #(APP_START_ADDRESS >> 16)");
#endif
#endif
    asm(" ldr     sp, [r0]");

    //
    // Load the initial PC from the application's vector table and branch to
    // the application's entry point.
    //
    asm(" ldr     r0, [r0, #4]");
    asm(" bx      r0");

You only have to provide to the application (at compile time) what the address of the next application is (0x4000 for example). Since you will configure your apps to have their vector tables at their start address, the apps will run as if they were placed normally.

Thanks,

Sean