• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 99 subscribers
  • Views 237 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.

PROCESSOR-SDK-J784S4: TIOVX: user kernels vs MPU target kernels

Jakub Olesinski
Prodigy 40 points

Part Number: PROCESSOR-SDK-J784S4

Tool/software:

Hello TI experts,

In TIOVX, when adding a custom kernel code that is meant to be executed on the MPU, we have two methods:

  • OpenVX standard user kernels
  • TIOVX extension target kernels

In PSDK I see that all of MPU kernels are implemented as TIOVX target kernels. Is there any disadvantage of using user kernels? Do you have any guidelines when each method should be preferred?

Regards,

Jakub

  • +1
    TI__Mastermind 22600 points

    Hi Jakub,

    In openvx standard user kernels you have initialize, validate, delete and run callback function. It is easy to implement.

    tiovx extension has initialize, validate in host side and create, delete, process, control callbacks so you need to have to sets of kernel creation, one for host and one for target.

    this gives flexibility to extend the target kernels to other cores apart from MPU, and it provides control callback command which can be used by tivxNodeSendCommand() api to

    send non-real time data to a node and/or to query for information from the node.

    Regards,
    Gokul