This post is older than 2 years and might not be relevant anymore
More Info: Consider searching for newer posts

Math support not found when linking with third party library

Hello all,

I am attempting to develop an application which utilizes a third party library for proprietary sensor calculations.  This library requires the use of floating point math library calls, such as hardfp_pow, hardfp_exp, hardfp_sqrtf, hardfp_sin, etc.

Is there a specific library configuration I need to include in my project settings so that the linker can include the appropriate math library supporting these calls?

Thanks in advance!

Mark

Parents
  • Hi,

    Is the third party library compiled for the same compiler you are using with your project? Which compiler/IDE are you using for your project?

    This sounds more like a general compiler/linker question, than related to the Nordic platform.

    Best regards,
    Jørgen

Reply
  • Hi,

    Is the third party library compiled for the same compiler you are using with your project? Which compiler/IDE are you using for your project?

    This sounds more like a general compiler/linker question, than related to the Nordic platform.

    Best regards,
    Jørgen

Children
  • Hi Jorgen,

    My apologies, I neglected to specify I was working under Segger Embedded Studio and using the GCC compiler.  Yes, the manufacturer has told me the library was specifically built to target this environment and compiler.

    Regards,

    Mark

  • Which SDK version do you use, and which example is your project based on?

    What library is this, and is it developed for embedded applications?

    Can you provide the library and/or project where you use the library, for reproducing the problem?

  • Hi Jorgen,

    I am using SDK v15.2.  The library was received from Bosch for one of their Air Quality sensors and was intended to target the nRF52832.  I have been in communication with Bosch about this issue as well, and I just discovered that the build they provided was not done in a way that was compatible.  I have since been able to successfully build with their floating point library that was designed for the Cortex M4 using the GCC compiler.

    My apologies for the false alarm... this was all apparently due to a problem with the library.

    Best Regards,

    Mark