• OpenCL Kernel produces wrong output in HSA mode

    Hi all, I have a severe problem when using one of my OpenCL kernels in HSA mode. When HSA is disabled, everything computes just fine, but by enabling HSA, a certain kernel produces a wrong output!   For demons...
    last modified by ced123
  • Failed to Run OpenCL on CPU of Microsoft Azure

    Hi, Recently I want to build a OpenCL training platfrom on MicroSoft Azure. To get the customer familiar with OpenCL, beginning with CPU as OpenCL device. However, I failed to get clinfo run on Micrsoft Azure. Help n...
    last modified by digbug
  • ACML 6 example segfaults

    Hello, Many of the example programs for ACML 6 are compiling fine but then terminating with segfaults. Specifically, the following are not working for me: cgels_c_example.exe cgesv_c_example.exe dgesdd_c_exampl...
    last modified by vigo
  • why clinfo shows 3GB global memory for 4GB on W7000 card

    Hi,   I'm using W7000 on Windows 7-64. Why clinfo shows 3GB global memory for 4GB W7000 card? What could be the reason? Should not clinfo show hardware specs?   Thanks.
    last modified by gouse
  • CodeXL 1.4 integration with HSA OpenCL 1.2 Beta Preview Driver

    I was able to install the AMD HSA driver (Beta preview for windows 8.1) and successfully compile and run the sample programs. Digging deeper however, the CodeXL (1.4.5724.0) on my workstation is unable to profile th...
    last modified by zhongliangong
  • Downside of using image vs global memory?

    I  have a kernel that processes RGB images. Currently, I take each channel one by one, and run the same kernel on that channel The kernel input is a global memory buffer: data is moved in chunks from the global b...
    last modified by boxerab
  • Status of OpenCL 2.0 SDK and reference manual?

    Initially the updated APP SDK was expected this summer. Is this still the case? Current APP programming guide is too outdated! No information is included about Bonaire and later GPUs and Sea islands or Volcanic Island...
    last modified by ekondis
  • Compiling OpenCL kernel on build server without a GPU

    Is there a way to compile OpenCL kernels for a specific GPU (Pitcairn) on a computer without a GPU installed?   We would like to compile the OpenCL kernels on our build server. It is virtual and doesn't have a G...
    last modified by jakob
  • Check if a recent driver is installed

    Hi,   I've had one user of my application report an inexplicable OpenCL error (CL_INVALID_MEM_OBJ in clSetKernelArg). I suspected might be due to a driver bug, and indeed after updating the driver all is working...
    last modified by joostn
  • OpenGL Texture in OpenCL not showing

    Hey Guys,   I am working on my thesis and i am programming a ray tracer in OpenCL. Well it is already finished, but i have to analyse it, how it is working on different GPUs. The Problem is now that on my AMD/A...
    last modified by kjo
  • Second graphics adapter not detected

    Hi, I have two 6970 in my system - but only one is detected. * Strange thing: Windows Device Manager shows under "Display adapters" two entries like this: "AMD Radeon HD 6900 Series"   * Catalyst Control Cen...
    last modified by Tristan23
  • OpenCL program crash

    Hello to everyone,   I am trying to figure out why my opencl program crash and it seem very strange I made these test   #define TTPROGRAMDATA  __global struct TProgramData  ...   // WORK OK...
    last modified by denis
  • About HSA Linux kernel driver

    Hi all, I have several questions related to the support of HSA linux driver.   1. Does Linux kfd v0.6.1 driver support HSA runtime API defined in the spec? 2. Is there any tutorial on how to use the kfd driver...
    last modified by grey9305
  • warning,version 410 is not yet complete

    I had code a vertexshader. Then I use it on an AMD card,but it didn't work. I can't find what's wrong with the vertexshader. So I run it on a NVIDIA card and it work well.   Then I compile it through the referen...
    last modified by yan_qingsong
  • False "The thread tried to access an array element that is out of bounds [...]" error

    Trying to use the CodeXL (from Visual Studio) OpenCL debugging using breakpoints, I invariably get the message:   The thread tried to access an array element that is out of bounds and the underlying hardware supp...
    last modified by maxbarraclough
  • DirectX IDXGIAdapter::GetDesc and AMD Switchable Graphics

    I have a Windows 7 Dell Latitude laptop with Intel HD Graphics 4600 and AMD Radeon HD 8790M GPUs. Our application reports the active GPU during startup using the DirectX IDXGIAdapter::GetDesc function. This informatio...
    last modified by joger.hansegard@ge.com
  • Incorrect results when multiple processes use the same GPU

    Hello. I've noticed strange behaviour of our scientific application when launching several instances on the same GPU - sometimes it gives incorrect results. I've reproduced this behaviour on the small program (see at...
    last modified by ajk
  • clmath installation issue (Ubuntu 12.04)

    Hi All,   This happens after running $sudo make install:   CMake Error at library/cmake_install.cmake:67 (FILE):   file INSTALL cannot copy file   "/home/vishal/clBLAS/src/package/lib64/src/packa...
    last modified by vish108
  • OpenCL 2.0 producer/consumer pattern example on Kaveri HSA Beta

    Hi, The Kaveri OpenCL Programmer Guide.docx suggests that producer/consumer patterns are allowed, is it possible to have an example please? I'm running on the HSA Beta. What can be used as a semaphore between the ...
    last modified by nickpenny
  • Help for compiling modules for Fedora 19,20 and futures

    Hi everybody,   It seems that the Catalyst driver is not supported fof Fedora recent versions by AMD neither by rpmfusion.org   Is there a way to workaround this issue and get Catalyst running again on Fed...
    last modified by nouvo09