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

How about Off-line Shader Compiler to analyze OpenCL?

Hi,

I'm studying OpenCL and try to kernel performance analyze with Mali Developer Guide

(Mali-T600 Series GPU OpenCL Developer Guide - Mali Developer Center)

But I can't understand below:  In chapter 8.1.3

  "Use the Off-line Shader Compiler to check the balancing between the different pipelines."

As far as my understanding, Off-line Shader Compiler cannot compile OpenCL kernel.

How to use the compiler?

And what can I get beneficial information with Off-line Shader Compiler?

Parents
  • Hi Saitoh,

    that part of the document refers to a private version of the Offline Compiler that we give to our partners, under NDA.

    I will let you know as soon as I have more news on this.

    In the meantime, I suggest you to have a look at GPUVerify, supported by ARM and in the Mali Graphics Debugger.

    GPUVerify is a tool for formal analysis of kernels written in OpenCL and was partly funded by the EU FP7 CARP project http://carpproject.eu/. The tool can prove that kernels don’t suffer from the following three issues:

    • Intra-group data races: This is when there is a data race between work items in the same work group.
    • Inter-group data races: This is when there is a data race between work items in different work groups.
    • Barrier divergence: This is when a kernel breaks the rules for barrier synchronization in conditional code defined in the OpenCL documentation.

    The tool was created by Imperial College London and more information about the tool and the issues it can diagnose can be found by visiting http://multicore.doc.ic.ac.uk/GPUVerify.

    Thanks,

    Lorenzo

Reply
  • Hi Saitoh,

    that part of the document refers to a private version of the Offline Compiler that we give to our partners, under NDA.

    I will let you know as soon as I have more news on this.

    In the meantime, I suggest you to have a look at GPUVerify, supported by ARM and in the Mali Graphics Debugger.

    GPUVerify is a tool for formal analysis of kernels written in OpenCL and was partly funded by the EU FP7 CARP project http://carpproject.eu/. The tool can prove that kernels don’t suffer from the following three issues:

    • Intra-group data races: This is when there is a data race between work items in the same work group.
    • Inter-group data races: This is when there is a data race between work items in different work groups.
    • Barrier divergence: This is when a kernel breaks the rules for barrier synchronization in conditional code defined in the OpenCL documentation.

    The tool was created by Imperial College London and more information about the tool and the issues it can diagnose can be found by visiting http://multicore.doc.ic.ac.uk/GPUVerify.

    Thanks,

    Lorenzo

Children