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.

C66x and 64 bit fixed point arithmetic

From the TMS320C66x DSP CPU and Instruction Set document, I see that the MPY32 instruction performs a 32 x 32 multiply to give a 64 bit fixed point result.

Are there any instructions that can perform computations with the 64 bit results, such as ADDs, SHIFTs, CONVERTs between different number formats? Or must these sorts of computations be synthesized from combinations of 32 bit operations?

  • J A Valina,

    There are no instructions that use a 64-bit Integer as the src (denoted as sIlong) for the instructions.  To process the results, you would need to break up the computations into smaller data sizes to process as you've surmised.

    Best Regards,

    Chad