kernel panic when loading vmlinux?

This topic contains 2 replies, has 2 voices, and was last updated by  ZubairLK 1 year, 7 months ago.

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #53098

    Chun-Kai
    Member

    Hi all,

    I changed the cache size in m14k_config.vh of the MIPSfpga core in the MIPSfpgaSOC package and I managed to generate a bitstream and download it to Nexys4 DDR. However when I loaded the pre-compiled vmlinux into it, kernel panic occurred. This does not happen when I load the vmlinux using unmodified project_linux.bit. What modification should I do during the compilation of linux kernel?

    #53221

    ZubairLK
    Member

    Hi Chun-Kai,

    Apologies for the delay.

    Two questions. When you update m14k_config.vh, did you just change the #define.

    1)
    With MIPSfpga SoC, you have to update the version number of the packaged IP and then update in the project_linux design.

    This ensures that the new source code is used in the next synthesis run.

    2) Can you please share the linux kernel boot log and panic message?

    Thanks
    ZubairLK

    #53234

    ZubairLK
    Member

    Hi Chun-Kai,

    I also spoke to some hardware engineers.

    Commercial customers do get a utility to configure the core and change these parameters.
    Unfortunately, that utility is not part of MIPSfpga under the Imagination University Program.

    The hardware engineers mentioned that simply changing a #define in one place won’t change the cache size.
    You are more than welcome to try changing the cache size manually in the code.

    However, I would recommend trying to test the larger/smaller cache with bare metal code and the see how it goes. Trying directly using MIPSfpga SoC might make it difficult.

    Regards,
    ZubairLK

Viewing 3 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic.