Search:

Type: Posts; User: opencl_beginner

Page 1 of 2 1 2

Search: Search took 0.01 seconds.

  1. Re: which format supported in clCreateFromD3D10Texture2DKHR?

    clCreateFromD3D10Texture2DKHR will increase the reference count.
    but clEnqueueReleaseD3D10ObjectsKHR can not decrease the reference count!
    ......what is wrong? how to use...
  2. Re: which format supported in clCreateFromD3D10Texture2DKHR?

    If I call clCreateFromD3D10Texture2DKHR using DXGI_FORMAT_B8G8R8A8_UNORM texture, I will get CL_INVALID_D3D10_RESOURCE_KHR.
    clEnqueueAcquireD3D10ObjectsKHR will not increase texture reference count....
  3. Re: which format supported in clCreateFromD3D10Texture2DKHR?

    Actually, I have a very very urgent question. Whether clEnqueueReleaseD3D10ObjectsKHR cause GPU memory leak?
    code]err =clEnqueueReleaseD3D10ObjectsKHR(m_Command_queue, 1, &m_Src_mem, NULL, NULL,...
  4. Re: which format supported in clCreateFromD3D10Texture2DKHR?

    I mean if you create a texture(its format is DXGI_FORMAT_B8G8R8A8_UNORM). Then use this texture in clCreateFromD3D10Texture2DKHR. I got an error!
    maybe you can try this...
  5. Re: which format supported in clCreateFromD3D10Texture2DKHR?

    very strange
    clGetSupportedImageFormats displays my card support CL_BGRA
    But I use DXGI_FORMAT_B8G8R8A8_UNORM texture to clCreateFromD3D10Texture2DKHR, I got an error!
    why?
  6. Replies
    1
    Views
    1,107

    strange question

    I want to do gauss blur to a lot of textures, so I use use a loop to handle every texture,
    I used clCreateFromD3D10Texture2DKHR to contact one texture with one cl_mem, then copy every texture
    to...
  7. Re: which format supported in clCreateFromD3D10Texture2DKHR?

    Thanks chai!
    I used Nvidia sample and found that even clEnqueueReleaseD3D10ObjectsKHR return 0, the reference count of the surface had not been reduced. maybe I should check ATI sample.
  8. Re: which format supported in clCreateFromD3D10Texture2DKHR?

    Hi, chai
    You may have used clEnqueueReleaseD3D10ObjectsKHR. Does this function cause a GPU memory leak? thank you!
  9. Will ATI support interoperability between OpenCL and D3d9?

    Hi
    Will ATI support interoperability between OpenCL and D3d9?
    thanks.
  10. which format supported in clCreateFromD3D10Texture2DKHR?

    Hi, All

    cl_mem clCreateFromD3D10Texture2DKHR( cl_context context,
    cl_mem_flags flags,
    ID3D10Texture2D *resource,
    uint subresource,
    cl_int *errcode_ret )

    which type of...
  11. Replies
    0
    Views
    1,654

    reduce the number of registers used

    Hi
    I wrote code like this:

    float4 f4Sum;
    for (int i=0; i<length; ++i)
    f4Sum += convert_float4(uc4Data[i])*pCo[i];

    the type of uc4Data is uchar4
    the type of pCo is float*
    I use compute...
  12. Replies
    1
    Views
    1,151

    Re: clEnqueueReleaseD3D10ObjectsKHR bug?

    I just used some test codes and found clEnqueueReleaseD3D10ObjectsKHR has the same problem. Is this a bug?
  13. Replies
    1
    Views
    1,151

    clEnqueueReleaseD3D10ObjectsKHR bug?

    Hi
    Sorry to tell you that I have not used clEnqueueReleaseD3D10ObjectsKHR , instead I used clEnqueueReleaseD3D9ObjectsNV_fn when I use d3d9ex intergration with OpenCL. I find that...
  14. Re: render cl_mem(image object) to window

    Thank you! I know how to convert texture to cl_mem. But I dont know how to convert cl_mem to texture. I dont want to use host memory copy. Actually the problem is that I have a cl_mem(image object),...
  15. render cl_mem(image object) to window

    Hi,
    clCreateFromD3D10Texture2DKHR can convert texture to cl_mem, but I want to transfer cl_mem to texture,then I can render it to window. but there is no such functions. so how could I render...
  16. Re: performance comparison between OpenCL & DirectCompute

    thank you david. the same program runs as fast as DirectCompute version on Geforce 240. so I think 8600gt may dont support OpenCL well.
  17. Re: performance comparison between OpenCL & DirectCompute

    "g_ImageHeight*2+1" is the template length of gauss blur. for every pixel I should use "g_ImageHeight*2+1" pixels to calculate its RGBA, g_ImageHeight before this pixel and g_ImageHeight after it. ...
  18. Re: performance comparison between OpenCL & DirectCompute

    I tried best to optimize OpenCL code but no use. OpenCL version is almost 2.5 times slower than DirectCompute version. I posted the OpenCL code here, and ask your help to optimize it.
    The algorithm...
  19. Re: performance comparison between OpenCL & DirectCompute

    Theoretically the same algorithm implemented by OpenCL or DirectCompute will have approximate performance? Is it right?
  20. Re: performance comparison between OpenCL & DirectCompute

    take an image 1000*1000 for example,
    I set the global size 1000*64, set local size 64,
    both for OpenCL and DirectCompute
    I got the result DirectCompute is 2 times faster than OpenCL on Nvidia...
  21. performance comparison between OpenCL & DirectCompute

    Hi, all
    I implemented the same algorithm using OpenCL and DirectCompute. I used the same totoal thread number, same group number. I think they will have approximate performance. but i am wrong....
  22. Replies
    1
    Views
    1,218

    Re: opencl interop with directx9

    problem solved according to nvidia sample.
    thanks to nvidia.
  23. Replies
    1
    Views
    1,157

    Re: where can i find cl_d3d9.h?

    I solved the problem, nvidia provide a sample for d3d9 intero with opencl.
  24. Replies
    1
    Views
    1,218

    opencl interop with directx9

    Hi, all

    IDirect3D9* d3d =NULL;
    d3d = Direct3DCreate9(D3D_SDK_VERSION);
    if (d3d == NULL)
    {
    //error handle
    }
    //set Direct3D presentation parameters
    D3DPRESENT_PARAMETERS...
  25. Replies
    1
    Views
    1,157

    where can i find cl_d3d9.h?

    where can I find cl_d3d9.h? I noly find cl_d3d9_ext.h, but no cl_d3d9.h, why? thanks!
Results 1 to 25 of 37
Page 1 of 2 1 2