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.

The speed of access L2

Other Parts Discussed in Thread: TMS320C6678

Can some one tell me is there some difference of  speed for access L2 as below:

1. Core access MSMC(configured as a shared  L2,) .

2. Core access Local L2RAM (0x00800000), e.g. core0 acess 0x10800000. 

3. Core access the L2 of other core , e.g. core0 acess 0x12800000. 

which is the most slowly  ? 

  • Xi Su,

    It may help for us to know which device you are using. But assuming you are using the C6678, please go to the TMS320C6678 Product Folder and search for "throughput" or "performance" to find an Application Note that has some details on various performance measurements. Honestly, there is a lot of detail in there that I do not fully understand, but you can apply your knowledge of processors to try to figure it out better than I can.

    In general, I would expect Local L2RAM access to be the fastest of these, followed by MSMC SRAM L2 (or L3) access, and the slowest would be accessing the internal L2RAM of another CorePac.

    Please look at the System Interconnect of the C6678 to see the paths through which the data would pass. The more components like SCRs and Bridges in the path, the slower that path will be.

    Regards,
    RandyP

  • Thank you !