Search:

Type: Posts; User: codedivine

Search: Search took 0.00 seconds.

  1. Sticky: Going through the specs slowly. Very high level...

    Going through the specs slowly. Very high level feedback is that we need more examples. I already mentioned this to Andrew on twitter.
    Also, compilation workflow is really unclear.
    For example:
    ...
  2. Thread: Bolt

    by codedivine
    Replies
    3
    Views
    1,686

    Re: Bolt

    I don't think it has been released yet.
  3. Replies
    2
    Views
    3,210

    Re: Timeouts for GPU kernels

    As far as I can tell, some big ISVs have faced the GPU timeout issue as well.
    For example, you can check Adobe's presentation at the SIGGRAPH 2012 OpenCL BOF presentation. Link:...
  4. Re: Proposal: More detailed device ID detection

    I have created a topic in the suggestion section. Please do not reply here.
    viewtopic.php?f=41&t=5402
  5. Replies
    1
    Views
    2,513

    Well defined ways of detecting product ID

    Problem: Often optimized kernels are written for a particular chip (say very optimized kernel specifically for AMD Tahiti GPUs) or for a particular family (say very optimized kernels for Nvidia Fermi...
  6. Replies
    2
    Views
    3,210

    Timeouts for GPU kernels

    When you run a kernel on the GPU, sometimes the kernel ends up running for a very long amount of time. Depending upon the vendor and OS combination, it can lead to anything from application crashes...
  7. Proposal: More detailed device ID detection

    Problem: Often optimized kernels are written for a particular chip (say very optimized kernel specifically for AMD Tahiti GPUs) or for a particular family (say very optimized kernels for Nvidia Fermi...
  8. Suggestion: Querying memory object alignment

    I am writing a library. I have multiple versions of the kernel depending on the alignment. The library API receives a buffer (as a cl_mem) and depending on whether or not it is aligned to (say)...
  9. Re: Driver changes, clCreateProgramWithBinary, clGetProgramI

    After driver change, the generated binary can definitely change as optimizations, binary object format and code generation strategies can change with driver revisions. And it is not necessary that...
  10. Profiling, performance counters on Nvidia OpenCL

    Is there anything similar to GPUPerfAPI on Nvidia OpenCL implementation? What libraries and tools do you use on nvidia hardware to understand performance of OpenCL kernels?
Results 1 to 10 of 20