Search:

Type: Posts; User: lava

Search: Search took 0.00 seconds.

  1. Replies
    2
    Views
    1,938

    Re: clCreateBuffer() need lots of time

    i have looked at these functions. but strangely the clEnqueueMapBuffer used more time than clEnqueueReadBuffer .
  2. Replies
    2
    Views
    1,938

    clCreateBuffer() need lots of time

    given a kernel and repeat running it, the corresponding cl_mem object is constant, but values of a array binded with the cl_mem object is changing, how to synchronize data between host and device ?...
  3. Replies
    6
    Views
    3,430

    Re: max local array size?

    u can check it by CL_DEVICE_LOCAL_MEM_SIZE argument :wink:
  4. Replies
    0
    Views
    1,694

    sorting big array in GPU

    Given a big array(about 8 digits for length), only containing 0 and 1 value, how to classify them into two classes efficiently and quickly using parallel algorithm? if the type of elements of the...
  5. Replies
    5
    Views
    2,317

    Re: About writing to Global Memory

    To Alex:
    yes, in each cycle of RunKernel() function, k1 or k2 will be called one time, and at each cycle of k2(where tag[tid] = 1) the time consumed is equal. but not that of k1(that depends on...
  6. Replies
    5
    Views
    2,317

    Re: About writing to Global Memory

    I mean``` the host code invokes the kernel code(f1 or f2).In RunKernel() function it will call
    clEnqueueNDRangeKernel() and at the last cycle it read data tag from device. The 3rd one will run...
  7. Replies
    5
    Views
    2,317

    Re: About writing to Global Memory

    PS: the actual global memory int *gtag and local memory ltag has its different name respectively.
  8. Replies
    5
    Views
    2,317

    About writing to Global Memory

    here are two kernel functions:



    __kernel void f1(__global int *tag,```)
    {
    uint tid = get_global_id(0);
    if(tag[tid] == 0)
    {
    ```;
  9. problem of communication between global memory and private's

    When i run a program, which contains a big recycle of data communication between private variables and global memory, i got a very bad performance result, which compared to the situation when i...
Results 1 to 9 of 10