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.
Hi team,
I am looking at the TMU cycles in F280049 datasheet. Here is a quick questions:
In the 1st row, it is noted that a = b*2pi would take 2/3 cycles.
1. Why this instruction would take 2 or 3 cycles but not a fixed number?
2. What data type is PI in this instruction? According to my knowledge,C28 is able to hanlde 32 bit float multiply in a single cycle? Why the instruction including PI would take more?
Thanks!
Regards,
Brian
Brian,
Its' 2 cycles if the 3rd doesn't conflict. Please refer to the corresponding instruction in the Extended Instruction Sets TRM (SPRUHS1) and it will explain the details.
It's a 32-bit float. Yes, that's an interesting question - single-precision float multiply is 2p cycles (1 cycle + 2nd if it doesn't conflict with the first). I'll try to find out more details.
Thanks,
Sira
Brian, it appears there is a bug. The MPY2PIF32 instruction should also take 2p instructions, like the MPYF32. I'll file a JIRA to have this fixed.