• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 30 subscribers
  • Views 1093 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.

TMS570LS1224: Library for doing floating point calculations in software

Eirik Nobuki Kosaka
Prodigy 140 points
Part Number: TMS570LS1224
Other Parts Discussed in Thread: HALCOGEN, , HERCULES-DSPLIB

Hello,

I might be asking this the wrong place, but I am not sure where I should be asking if this is the case so I'll ask here. 

So I am using the TMS570ls1224 and I am testing some software that does some floating point calculations. What I want to do is test the code with the floating point unit in the chip, and also without the FPU, simulating it like a processor without FPU. Does TI have some library for their chips without FPU that I can use for the TMS570ls1224 when the FPU/VFP is disabled in halcogen?

Eirik N. Kosaka

  • 0
    TI__Guru 59540 points
    Hello Elrik,

    The only math library we offer is the CMSIS library (HERCULES-DSPLIB) which offers a variety of advance math algorithms/functions. I have not used it so I don't know if it includes floating point algorithms or not but you are welcome to take a look.

    For our devices that do not support floating point, our customers usually have their own fixed point math algorithms already developed based on years of use of non-floating point capable MCUs since floating point support has only recently (relatively speaking) become the norm in the industry.
  • 0 in reply to Chuck Davenport
    Prodigy 140 points
    Hey Chuck,

    So I found out that if i turn off the VFP in HalCoGen, and I go to the compiler option in Code Composer I can switch from VFPv3D16 to vfplib in the "Specify floating point support" dropdown menu and I think this is what I am looking for.

    However I seem to run into a new problem with this, I get 24 new error messages similar to this one:

    file "C:/ti/ccsv7/tools/compiler/ti-cgt-arm_16.9.1.LTS/lib/rtsv7R4_T_be_v3D16_eabi.lib<_printfi.obj>" has a Tag_ABI_VFP_args attribute value of "1" that is different than one previously seen ("0"); combining incompatible files file "C:/ti/ccsv7/tools/compiler/ti-cgt-arm_16.9.1.LTS/lib/rtsv7R4_T_be_v3D16_eabi.lib<_printfi.obj>" has a Tag_ABI_VFP_args attribute value of "1" that is different than one previously seen ("0"); combining incompatible files

    I can't seem to find a solution for this?

    Regards,
    Eirik N. Kosaka
  • 0 in reply to Eirik Nobuki Kosaka
    TI__Guru 59540 points

    Hello Elrik,

    First, my apologies for the delayed response.

    vfplib is the wrong selection to eliminate floating point. You should select fpalib.

    In addition, you want to make sure you select the run time support library that also avoids the floating point support. For this go to the same properties window and select General. At the bottom of the entry fields on the left side of the window you will see runtime support library. Select rtsv7R4_T_be_eabi.lib which is the non-floating point RT support library.

    If you make these changes it should eliminate the error messages you are seeing.