Changes between Version 15 and Version 16 of GPUApp


Ignore:
Timestamp:
Jul 15, 2010, 3:53:56 PM (14 years ago)
Author:
TuanLe
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • GPUApp

    v15 v16  
    2626
    2727The project file, example_app_nvopencl.vcproj, is in "boinc/win_build/".
    28 OpenCL doesn't have emulation mode,
     28OpenCL doesn't have emulation mode, 
    2929but it could run on CPU alone by specifying CL_DEVICE_TYPE_CPU or CL_DEVICE_TYPE_ALL when creating an OpenCL context.
     30
     31The executable file can be found at "boinc/win_build/Build/Win32/build_mode/"
     32(build_mode could be release or debug).
    3033 
    3134=== ATI Stream OpenCL ===
     
    7679/usr/bin/ld: cannot find -lcutil_i386
    7780}}}
    78 
    7981If your machine doesn't have an NVIDIA Cuda-enabled GPU,
    8082you can compile and run your app in emulation mode.
     
    9294./example_app_nvcuda: error while loading shared libraries: libcudart.so.3: cannot open shared object file: No such file or directory
    9395}}}
     96The executable file is created in "/boinc/samples/nvcuda/linux/build_mode/" (build_mode could be release or emurelease).
    9497
    9598=== NVIDIA OpenCL ===
     
    105108is installed at the default location "$(ROOT)/home/USER_NAME/NVIDIA_GPU_Computing_SDK/".
    106109On my Linux machine, for example, it would be "$(ROOT)/home/tuanle/NVIDIA_GPU_Computing_SDK/".
    107 Since the absolute path is used in Makefile, you will probably need to edit file "common.mk" before compiling the sample app.
    108 Do so by looking for "/home/tuanle/NVIDIA_GPU_Computing_SDK" in file "common.mk"
    109 and replace this path with an appropriate path on your machine.
     110Since the absolute path is used in Makefile, you will probably need to edit file "common_opencl.mk" before
     111compiling the sample app. Do so by looking for "/home/tuanle/NVIDIA_GPU_Computing_SDK" in file
     112"common_opencl.mk" and replace this path with an appropriate path on your machine.
     113
     114You will also need to build the necessary Cuda SDK libraries by running makefile in
     115"~/NVIDIA_GPU_Computing_SDK/OpenCL/". The resulting libraries such as "liboclUtil_i386" are
     116created in "~/NVIDIA_GPU_Computing_SDK/OpenCL/common/lib/". Notice that if these libraries are not built,
     117running nvopencl-sample-app makefile could result in the following error:
     118{{{
     119/usr/bin/ld: cannot find -loclUtil_i386
     120}}}
     121
     122The executable file is created in "/boinc/samples/nvopencl/linux/release/".
    110123
    111124=== ATI Stream OpenCL ===