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.

TMDS64GPEVM: CCS V11 with TI Clang compiler Floating point Error

Part Number: TMDS64GPEVM

Hello,

My current project is using the following:

- Code Composer Studio v11.2.0.00007
- TI Clang v1.3.1 LTS compiler

gpio_led_blink_am64x-evm_r5fss0-0_nortos_ti-arm-clang

I modified the LED toggle example for the TMDS64GPEVM board to add my own code. The current compiling error I am getting is when using 2 floating point instruction set (FLDMIAD/ FSTMDBD )and the error are as follow:

* FLDMIAD : "invalid instruction, did you mean: fldmiax? “

* A similar error happens for FSTMDBD.

The suggestion that the compiler gives for using the X condition at the end of the instruction does not work for my code, and I am looking for a work around that you could suggest for using another instruction set that does the same functionality. My code uses those instructions as follow:

FLDMIAD SP!, {D0-D15}

FSTMDBD SP!, {D0-D15}

Updating to CCSv12 is also an option if those floating point instructions are supported on the latest release.

Regards,

Fernando

  • Are these floating point instructions generated by the compiler tiarmclang when building a C source file?  If so, for that source file, please follow the directions in the article How to Submit a Compiler Test Case.  If these instructions appear in hand-coded assembly, please submit a short assembly source file, which includes no other files, that uses those instructions the same way your code does.  Note the code in this short file only has to build, and not execute.  In addition, show the exact options used with tiarmclang to build this file and see those same diagnostics.

    Thanks and regards,

    -George