• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 30 subscribers
  • Views 192 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TM4C129XNCZAD: How to switch from using IQmathLib to using hardware floating point capability

Mark Sheats
Intellectual 500 points
Part Number: TM4C129XNCZAD

Tool/software:

As I have mentioned in other posts, I have taken over a project from another programmer and I'm working on figuring everything out.  It was a big deal in the beginning to get the program to build properly, because he was using functions in the IQmathLib.  I had to copy the IQmath folder from his old laptop in order to compile successfully.

Was there a time that it was required to use this library?  Or did he choose the wrong library to start with?

I am now informed that the processor has hardware floating point capability.  I assume this would provide a big increase in performance.  How do I switch from software floating point functions to using hardware floating point functions?

  • +1
    TI__Guru**** 184346 points

    Hi Mark,

    Mark Sheats said:
    I am now informed that the processor has hardware floating point capability.  I assume this would provide a big increase in performance.  How do I switch from software floating point functions to using hardware floating point functions?

     Your understanding is correct.  IQmath a collection of optimized mathematical functions for C programmers to port a floating-point algorithm into fixed-point code on devices without built-in floating point hardware. TM4C129 is based off of Cortex-M4F which has the built-in hardware floating point. The floating point is enabled in the CCS setting during build time. 

  • 0 in reply to Charles Tsai
    Intellectual 500 points

    I checked, and this option is specified in my project already.  So if this option is specified as you have directed, the compiler will do what it needs to do and prefer hardware floating point support over the IQmath that is also listed in the include options?

  • 0 in reply to Mark Sheats
    TI__Guru**** 184346 points

    I think it depends what you are trying to achieve. If you are simply doing simple floating point operations like multiplying some floating numbers then I believe the integrated hardware floating unit will do the job fine. If you have some routines that are used in computationally intensive
    real-time applications where optimal execution speed and high accuracy is critical then you can still use IQMath library, Refer to the IQMath user's guide for details. 

    SW-TM4C-IQMATH-UG-2.1.4.178.pdf

  • 0 in reply to Charles Tsai
    Intellectual 500 points

    Okay, unless you tell me otherwise, I'm not going to change anything.  The program is just doing multiplication and division.  I thought maybe I had to remove any reference to IQMath and use a new include file or something like that.

  • +1 in reply to Mark Sheats
    TI__Guru**** 184346 points

    You can keep the IQMath library in your include path and it won't increase your memory footprint if no IQMath function is used. Keep it there incase you do want to use it in the future.