• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 98 subscribers
  • Views 764 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TDA4VM: Memory map, adding new memory region

Marko Milosevic
Intellectual 412 points
Part Number: TDA4VM


Hello TI,

We have a question about the memory map for the J721E (TDA4).
We want to add new memory region(start address: 0xFC000000, end address:0xFC7FFFFF, size: 8.00 MB) for hosts MCU2_0, MCU2_1 and MPU1_0(QNX). We will use this memory for sharing data between CPUs.
We will follow this manual link.
Is linker.cmd (vision_apps\apps\basic_demos\app_tirtos\tirtos_qnx\mcu2_0\linker.cmd) file generated or we need to update manualy this file?
Do we need to change anything else file manually (we know for file r5_mpu.xs )?

Best Regards,

Marko.

  • 0
    TI__Guru**** 479865 points

    Hi Marko,

    Yes, you would require to change r5_mpu.xs file manually, depending on how you want to map this new region on R5F. Rest all is taken care by script.

    Regards,

    Brijesh

  • 0 in reply to Brijesh Jadav
    Intellectual 412 points

    Hi Brijesh,

    Thank you for the answer. We added a new memory region on address (0xA910000, size 8MB). Created memory map for MCU2_0 and MCU2_1 ( mcu2_0/1_mmap.addMemSection( example_shared_mem ); ). Execute ./gen_linker_mem_map.py python file to generate the files.

    We manually changed files r5_mpu.xs and linker.cmd for MCU2_0 and MCU2_1. Build works after some modification (build fail for MCU1_0, we manually added R5F_TCMB0_VECS in file mcu1_0/linker_mem_map.cmd).

    After generation we noticed that we changed the file /basic_demos/app_tirtos/tirtos_qnx/mcu1_0/linker_mem_map.cmd  (whether this is expected behavior )

    /* R5F_TCMA [ size 32.00 KB ] */
    R5F_TCMA                 (    X ) : ORIGIN = 0x00000000 , LENGTH = 0x00008000
    /* R5F_TCMB0 [ size 32.00 KB ] */
    R5F_TCMB0_VECS           ( RWIX ) : ORIGIN = 0x41010000 , LENGTH = 0x00000100
    /* R5F_TCMB0 [ size 32.00 KB ] */
    R5F_TCMB0                ( RWIX ) : ORIGIN = 0x41010100 , LENGTH = 0x00007F00

    /* R5F_TCMA [ size 32.00 KB ] */
    R5F_TCMA                 (    X ) : ORIGIN = 0x00000000 , LENGTH = 0x00008000
    /* R5F_TCMB0 [ size 32.00 KB ] */
    R5F_TCMB0                ( RWIX ) : ORIGIN = 0x41010000 , LENGTH = 0x00008000

    Also, this happens without our changes (just execute ./gen_linker_mem_map.py). 

    Best Regards, 

    Marko.

  • +1 in reply to Marko Milosevic
    TI__Guru**** 479865 points

    Hi Marko,

    Is this happening only for mcu1_0 linker command file? 

    Typically mcu1_0 is not included in the build, so you could ignore these changes. But in case, if you want to include mcu1_0 and run vision apps on this core, we would have to revert these changes.

    I will raise this to the team and will try to get it fixed in the tool. 

    Regards,

    Brijesh

  • 0 in reply to Brijesh Jadav
    Intellectual 412 points

    Hi Brijesh,

    Thank you. In the future, we don't need vision apps on mcu1_0, so we will exclude this core from the build. Everything is fine.

    Best regards,

    Marko.

  • 0 in reply to Marko Milosevic
    TI__Guru**** 479865 points

    Thanks Marko, closing this thread.