Please note: We are aware of an issue affecting replies on the Arm Community forums, which may not be loading as expected.

We apologize for any inconvenience and appreciate your patience while we investigate and work to resolve the issue.

Thank you for your understanding.


This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Streamline performance counter G715: $MaliCoreInstructionsNarrowInstructions

Hi, I recently saw the new instruction counter, and wanted to know which unit handles Narrow instructions : FMA, CVT or SFU?

Parents
  • All of them =)

    This counter is intended to give developers feedback about how well their shaders are using mediump/relaxed precision. It increments for every 8-bit and 16-bit instruction executed, no matter which pipe it executes in. 

    You can compute the percentage of narrow instructions as "narrow / (FMA + CVT + SFU)".

    HTH, 
    Pete

Reply
  • All of them =)

    This counter is intended to give developers feedback about how well their shaders are using mediump/relaxed precision. It increments for every 8-bit and 16-bit instruction executed, no matter which pipe it executes in. 

    You can compute the percentage of narrow instructions as "narrow / (FMA + CVT + SFU)".

    HTH, 
    Pete

Children