Search:

Type: Posts; User: Gamingdrake

Page 1 of 2 1 2

Search: Search took 0.00 seconds.

  1. Re: out of resources when clEnqueueReadBuffer

    Going on what Andrew said, I was just having this error on a convolution algorithm.makes sense if you were to have an index_out_of_bounds exception that is caught by the GPU. You would be accessing a...
  2. Replies
    5
    Views
    2,413

    Re: Barrier and Array copying

    Thank you for your help Centershock. I have been looking for answers to this problem for awhile. I was able to unroll the loops, by cutting the data up into another grid. Turns out copying it to...
  3. Replies
    5
    Views
    2,413

    Re: Barrier and Array copying

    Again thank you for your answer. The reason I am keeping the code small is because I am technically not allowed to show the full code, due to a pending patent. So to keep things simple, I have a...
  4. Replies
    4
    Views
    1,570

    Re: local declaration inside the kernel

    Thank you for your answer.
    When I say:

    it means that I have output coming back from the GPU. In the first case, before it breaks, I tell it to output 32 to every value in "global float* output"....
  5. Replies
    4
    Views
    1,570

    Re: local declaration inside the kernel

    I even tried changing the output array to a local variable, and it spit out random values, or the values that had been last run, but not the ones from the current run
  6. Replies
    4
    Views
    1,570

    local declaration inside the kernel

    What happens when you declare a variable local inside the kernel itself. I have some code that works, but i want to make it faster. So, I want to copy the global input array into a local tempArray. I...
  7. Replies
    2
    Views
    1,310

    Re: Efficient use of memory in GPU

    EDIT:

    creating the tempBuffer inside the GPU works, rather than passing it in, but I cannot pass the size a variable, it has to be hardcoded. Could there be a workaround to this?
  8. Replies
    2
    Views
    1,310

    Efficient use of memory in GPU

    Hey guys,

    I have been looking for answers now for about a week and cant find anything useful, so here goes.

    I have a kernel that takes a global float* as an input parameter, and another as an...
  9. Replies
    5
    Views
    2,413

    Re: Barrier and Array copying

    I tried passing the array (float*) into the kernel as an empty array that had been initialized. It is not even giving me answers that make sense. Im telling it to print out 15, but it just prints out...
  10. Replies
    5
    Views
    2,413

    Barrier and Array copying

    Hello,

    I am having a problem with the efficiency of my kernel due to too many global reads. Therefore, I would like to copy the global array into a "shared" array inside my kernel. The code that...
  11. Replies
    2
    Views
    1,406

    Re: Best way to implement this?

    maybe prefetch or async_copy?
  12. Replies
    2
    Views
    1,406

    Re: Best way to implement this?

    To specify (maybe it will help), I am running a grid of weights over a larger grid of values. For each value, I am recalculating it based on the (values around it) * (weights that correspond), but it...
  13. Re: Is it possible to batch a 2d portion of a grid?

    Or maybe my problem is here. This is the call that I am executing in c++.



    clEnqueueNDRangeKernel(queue, kernel, 1, NULL, &sizeIn, NULL, 0, NULL, NULL);

    My problem is that I have a Quadro...
  14. Is it possible to batch a 2d portion of a grid?

    I am trying to overlay a grid (grid1) onto another grid (grid2) in parallel, and edit grid2 based on the values in grid1. Here is my kernel in pseudo



    _kernel void simple(
    global const float*...
  15. Replies
    2
    Views
    1,406

    Best way to implement this?

    In advance, I do not expect you to do my work for me, I would just like some thoughts.

    I have a kernel that needs to scan every item in an array of data. (pseudocode)



    kernal void myKernel(...
  16. Re: how to implement serial calculation in kernel code?

    You also have to watch out that your workload is not too big and a thread doesnt "hang" too long. In my experience, if I have a kernel hang two long, then too things happen:

    The OS stops drawing
    ...
  17. Re: regarding usage of barrier in openCL kernel

    This may not quite be the answer you are looking for, but try to find an example that uses barriers. Seeing as OpenCL is still new, you could also just try your own code, throw stuff together, and do...
  18. Replies
    4
    Views
    1,574

    Re: NVidia implementation works on ATI?

    David,

    I got it. If I get the library from NVidia (It comes prepackaged with the latest driver), then I link to that and use the OpenCL libraries stored in a folder (opencl/CL), then my program...
  19. Replies
    1
    Views
    1,540

    FULL OPENCL EXAMPLE

    Hey Guys,
    I have used the forums multiple times to get answers and would like to submit my code to the forums. Please do not replicate it and call it your own, but I would not be opposed to using...
  20. Re: How to properly create Read and Write Buffers:

    Thank you for your help. I went back through the code and the changes that you provided, and apparently I changed too much last night. I recommitted and the changes work. Thank you so much for your...
  21. Re: Setup OpenCL for completely heterogeneous implementation

    I do not want the SDK...that is why I posted this. You link to the library given to you by the driver...not the SDK
  22. Replies
    5
    Views
    2,270

    Re: OpenCL without Nvidia or AMD?

    Solved...The way is to just include the headers in a opencl/CL folder and link to the opencl folder when building. You also have to make sure that the latest dll is in your system, usually in system32
  23. Re: How to properly create Read and Write Buffers:

    OK so I fixed the code samples that you gave me. Thank you for your help btw. But there are still problems.
    I changed the kernel to use const instead of read_only and nothing for the write_only...
  24. Re: How to properly create Read and Write Buffers:

    The error that I get is Access Violation reading location blahblahblah and when I trace the callstack it sends me to nvcuda.dll.

    I am thinking that I have gone wrong somewhere in my buffers, but...
  25. How to properly create Read and Write Buffers:

    i am currently doing a straight implementation of openCL in c++, meaning that I am not using the SDKs. I am experiencing a problem with the writebuffer and readbuffer that I cannot debug, and it has...
Results 1 to 25 of 31
Page 1 of 2 1 2