Page 2 of 2 FirstFirst 12
Results 11 to 12 of 12

Thread: clEnqueueNDRangeKernel max global_work_size

  1. #11
    Senior Member
    Join Date
    Nov 2009
    Posts
    118

    Re: clEnqueueNDRangeKernel max global_work_size

    In NVIDIA OpenCL guide we can read :
    The maximum size of each dimension of a grid of thread blocks is 65535;
    So there is a limitation, fixed by nVidia, which is'nt define by specification and that we can't query with API... This constraint limits a bit the portability

  2. #12
    Senior Member
    Join Date
    Jul 2009
    Location
    Northern Europe
    Posts
    311

    Re: clEnqueueNDRangeKernel max global_work_size

    It is strange that you can use arbitrary sizes on Nvidia cards on MacOS X but not with Nvidia's own drivers. All they would have to do is breakup the runtime kernel execution into 65k sized chunks since each execution is necessarily independent.

Page 2 of 2 FirstFirst 12

Similar Threads

  1. About Open Max AL
    By Rekha.sm in forum OpenMAX AL
    Replies: 3
    Last Post: 03-21-2013, 08:26 AM
  2. Replies: 1
    Last Post: 08-13-2010, 09:19 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •