• utnapishtim's Avatar
    Today, 03:07 AM
    The host buffer is not necessarily up-to-date when your kernel ends because its content can be cached in device memory. You have to use...
    1 replies | 28 view(s)
  • utnapishtim's Avatar
    04-11-2014, 10:04 AM
    Check whether the extension is present in the string returned by clGetDeviceInfo() with CL_DEVICE_EXTENSIONS.
    3 replies | 131 view(s)
  • utnapishtim's Avatar
    04-11-2014, 06:36 AM
    Are you sure that your device has support for the cl_khr_3d_image_writes extension? Also use clGetProgramBuildInfo() with CL_PROGRAM_BUILD_LOG to...
    3 replies | 131 view(s)
  • utnapishtim's Avatar
    04-10-2014, 03:50 AM
    Your kernels could be optimized, but the most important parameter when using a GPU is the local work size. NVIDIA GPUs for instance are optimized...
    3 replies | 238 view(s)
  • utnapishtim's Avatar
    03-27-2014, 04:13 AM
    utnapishtim replied to a thread Stream compaction in OpenCL
    CL_MEM_READ_WRITE flag will create a buffer in device memory. CL_MEM_HOST_NO_ACCESS is just an optional hint.
    7 replies | 232 view(s)
  • utnapishtim's Avatar
    03-26-2014, 09:37 AM
    utnapishtim replied to a thread Stream compaction in OpenCL
    Just use clCreateBuffer() with CL_MEM_READ_WRITE flag. You can also add the hint flag CL_MEM_HOST_NO_ACCESS if your device has support for OpenCL 1.2.
    7 replies | 232 view(s)
  • utnapishtim's Avatar
    03-24-2014, 04:27 AM
    Note that buffers use the endianness of the device, so a buffer should be read or written taking this into account. You can change this behavior...
    5 replies | 213 view(s)
  • utnapishtim's Avatar
    03-24-2014, 03:51 AM
    The example at the bottom of page 363 in appendix B uses illegal casts: float4 v = vload4( 0, x ); uint4 y = (uint4) v; // legal, portable...
    30 replies | 10437 view(s)
  • utnapishtim's Avatar
    03-21-2014, 06:21 AM
    You have to install an OpenCL driver for a supported device. Since you have an Intel CPU but no GPU, you should install the Intel OpenCL driver...
    2 replies | 212 view(s)
  • utnapishtim's Avatar
    03-21-2014, 04:03 AM
    One of the job of the OpenCL runtime is to marshal data between host and device transparently. Alignment and packing are defined in the OpenCL...
    5 replies | 213 view(s)
  • utnapishtim's Avatar
    03-20-2014, 03:37 AM
    utnapishtim replied to a thread Stream compaction in OpenCL
    "Recent GPU" probably means less than 10-year old here... Gather means that the GPU can do random-access loads, while scatter means that the GPU...
    7 replies | 232 view(s)
No More Results
About utnapishtim

Basic Information

Statistics


Total Posts
Total Posts
103
Posts Per Day
0.19
General Information
Last Activity
Today 08:51 AM
Join Date
10-23-2012
Referrals
0