Search:

Type: Posts; User: jprice

Page 1 of 2 1 2

Search: Search took 0.00 seconds.

  1. The function you've defined is called...

    The function you've defined is called cltocppfloat4, but the function you are trying to call is called cltocppfloat.
  2. I've just run your code and it produces the...

    I've just run your code and it produces the desired results (30-79). Which OpenCL platform/device are you running this on?
  3. Replies
    8
    Views
    6,793

    Sticky: It's not clear to me what the problem is. There...

    It's not clear to me what the problem is. There is no requirement that all pending operations using a buffer complete before you can release it - the buffer will only be destroyed when the reference...
  4. Replies
    1
    Views
    1,020

    Yes, as long as both queues were created within...

    Yes, as long as both queues were created within the same context.
  5. Replies
    2
    Views
    1,307

    I'm afraid you've made the same mistake as in...

    I'm afraid you've made the same mistake as in your previous topic, in that you are misunderstanding how the sizeof() operator works. In this case, the datatype of 'a' is a 'pointer to unsigned...
  6. Replies
    4
    Views
    935

    This issue is that you are incorrectly using...

    This issue is that you are incorrectly using 'sizeof(a)' to determine the size of your array. The sizeof operator computes the size of the datatype you give it, not the size of the array you...
  7. Replies
    2
    Views
    1,993

    Hi, For NVIDIA (assuming proprietary driver),...

    Hi,

    For NVIDIA (assuming proprietary driver), you can use the nvidia-smi command-line tool to gauge approximate GPU load:

    $ nvidia-smi
    Thu Dec 5 15:36:15 2013 ...
  8. Hi Zvika, The preferred vector width is just a...

    Hi Zvika,

    The preferred vector width is just a recommendation for improving performance. In this case, NVIDIA's OpenCL implementation is telling you that it would prefer vectors of size 1 (i.e....
  9. Hi Nikki, If you just want to be able to...

    Hi Nikki,

    If you just want to be able to develop/run OpenCL code (and aren't too concerned about performance), then almost anything will do since OpenCL will run perfectly well on the CPU. If your...
  10. Hi Nikki, At present, the intermediate...

    Hi Nikki,

    At present, the intermediate representations used by these vendors are not compatible. AMD's compiler will generate AMD IL, where as NVIDIA's implementation generates and consumes PTX....
  11. Hi Zvika, 1. The OpenCL implementation for...

    Hi Zvika,

    1. The OpenCL implementation for NVIDIA's GPUs is packaged with their driver. Therefore, if you have the driver installed (which I assume you do), then you will be able to run OpenCL...
  12. Replies
    1
    Views
    1,064

    Hi, If the clCreateContext() function...

    Hi,

    If the clCreateContext() function succeeds, then status will be set to CL_SUCCESS, and context will be non-NULL. So yes, your test should be sufficient to determine if the context is valid....
  13. Replies
    3
    Views
    1,750

    Hi Tim, There is indeed a gap in the current...

    Hi Tim,

    There is indeed a gap in the current specification regarding implicit conversions from scalar to vector types, which results in undefined compiler behaviour for builtin functions. This has...
  14. Replies
    4
    Views
    1,212

    No, the function get_global_id() just returns the...

    No, the function get_global_id() just returns the index of the thread (work-item) that is executing. When you call clEnqueueNDRangeKernel(), you specify how many threads (work-items) you want to...
  15. Replies
    4
    Views
    2,040

    Re: First NVIDIA OpenCL Driver Version?

    295.40 definitely has support for OpenCL on some NVIDIA GPUs, but not all GPUs are supported (for example, a Tesla K20c did not work with this driver version on one of our boxes). Which GPU are you...
  16. Replies
    3
    Views
    1,446

    Re: opencl sdk problem

    I believe it's the same SDK for both CUDA and OpenCL, but I could be mistaken.
  17. Replies
    3
    Views
    1,446

    Re: opencl sdk problem

    You can find a full archive of all the CUDA Toolkit versions here:
    https://developer.nvidia.com/cuda-toolkit-archive
  18. Replies
    2
    Views
    1,601

    Re: Is it OpenCL properly installed in my PC?

    I believe NVIDIA only added OpenCL/CUDA support from 8-series cards onwards, so your GPU would not support OpenCL.
  19. Replies
    2
    Views
    2,227

    Re: Timing with clGetEventProfilingInfo

    Ah, that's what I was missing. I guess I read over that a little too fast, making the statement about CL_DEVICE_PROFILING_TIMER_RESOLUTION a little confusing.

    That makes everything much clearer,...
  20. Replies
    2
    Views
    2,227

    Timing with clGetEventProfilingInfo

    Hi,

    I'm having trouble getting the correct timings from the OpenCL profiling functions. I'm using the CL_DEVICE_PROFILING_TIMER_RESOLUTION property combined with the clGetEventProfilingInfo...
  21. Replies
    4
    Views
    2,172

    Re: Detecting same device on different platforms

    The AMD OpenCL SDK does indeed work for all x86 CPUs, including Intels. Perhaps AMD's driver works differently to Intel's and doesn't require SSE4. My guess is the latter of your hunches - the ATI...
  22. Replies
    4
    Views
    2,172

    Re: Detecting same device on different platforms

    Fair enough, I guess that's more or less the workaround I have at the moment.

    Thanks very much.
  23. Replies
    4
    Views
    2,172

    Detecting same device on different platforms

    Hi,

    I would like to be able to automatically use all available devices on any given system. The problem I'm having is that I can't find any way of telling if two devices on different platforms are...
Results 1 to 23 of 28
Page 1 of 2 1 2