• Evaluating workgroup reductions in OpenCL 2.0

    I was comparing shared memory reductions with reductions based on the new workgroup reduction functions available in OpenCL 2.0.  Though I believed the workgroup function would lead to even more optimized code th...
    ekondis
    last modified by ekondis
  • OpenCL in games on amd dual-graphics devices

    Hello, what about opencl calculation of physics in games like Nvidia Physix on AMD, by OpenCL? To use discrete GPU in games for graphics, and integrated in APU GPU for physics? It will by userful for dual-gpu machine...
    maestrodan
    last modified by maestrodan
  • fm2+ dual socket apu-support

    hi all out there,   i would like to know, if it's a good suggestion of the following thinkings.   i thought about a dual fm2+-socket-board to raise gaming-performance with apu's. pros:   - raise 2x4...
    th3r0ck03
    last modified by th3r0ck03
  • Problem when install sdk 3.0

    Hello guys   recently I'm install the sdk3.0 but some problem occurs...   when I tried to install the app sdk sample vit script in /opt/AMDAPPSDK-3.0-Beta/install.sh   it shows the following error me...
    davies1828
    last modified by davies1828
  • CPU instructions not accurately profiled

    Hello everyone,   I've been using codeXL for a while, and it works pretty good, except for source code viewing, when most of the time I see that just one single instruction takes most of the execution time. Se...
    aleksazr
    last modified by aleksazr
  • Completely wrong results of sgemm & dgemm with ACML 6.0.6 on Intel CPUs?

    Hello everybody,   while trying to check the performance of my computer's GPU (Pitcairn) I noticed that even when submitting the calculation to host cpu as default results in completely wrong results of sgemm ...
    bugchucker
    last modified by bugchucker
  • Most cache friendly tiling of OpenCL image ?

    I have an OpenCL image that is broken into tiles of 64x64 pixels. I am designing a kernel to run through all tiles and process the pixels. Target is AMD GCN. Currently, I process the tiles in raster order: left to ri...
    boxerab
    last modified by boxerab
  • Poor workgroup reduction function performance (OpenCL 2.0)

    I had previously reported this but here I provide a test case for the examination of workgroup reduction function. Kernels perform a workgroup reduction in 3 ways: 1) The classical one with shared memory (OpenCL 1.2)...
    ekondis
    last modified by ekondis
  • Execution issues of CodeXL 1.6 on 64bit Linux

    The CodeXL analyser has a problem when is executed either as standalone or through CodeXL GUI. It seems it tries to load the 32bit GL library. For instance using it to get a list of supporting GPUs I get the following...
    ekondis
    last modified by ekondis
  • NEW - GPU PerfStudio 3.1 has been released

    NEW - GPU PerfStudio 3.1 has been released and can be downloaded from http://developer.amd.com/tools-and-sdks/graphics-development/gpu-perfstudio/ It includes: * GPU ShaderAnalyzer for GCN - Optimize your shader cod...
    Lonesled
    last modified by Lonesled
  • Few problems with new release 1.5.6571.0

    - Debugger still does not work for me, it locks up Visual Studio, CodeXL crashes inside clBuildProgram. The same program runs fine normally. I read somewhere that the previous version had problems with the latest rade...
    gabest
    last modified by gabest
  • Reuse of Kernel in aparapi / memoryleak ?

    Hi,   have I missed something how I can tell aparapi / openCL that I will use a given Kernel more than once, with other data?   The thing is, I have written a Java Prog which uses aparapi to calculate matr...
    kulan
    last modified by kulan
  • glCopyImageSubData for small compressed mips

    glCopyImageSubData thows a GL_INVALID_VALUE when copying mips of a block compressed texture that are smaller than the block size in any dimension.   e.g. GLuint tex0 = ...; // 1x1 compressed texture GLuint tex1...
    baldurk
    last modified by baldurk
  • Can't Debug Kernel in Teapot Example

    I have an HP Z820 configured width an Intel Xeon E5-2650, 32GB RAM, and an msi Radeon R9 280X Gaming 6G card. I am running Windows 8.1 Pro, and have Visual Studio Premium 2013 Update 3, AMD Catalyst Version 14.9(Drive...
    jsnyder
    last modified by jsnyder
  • power profiling in latest CodeXL 1.6.7247.0

    Hi there. Recently I installed the latest CodeXL (v1.6.7247.0) and found that it can do power profiling, which is a good function for AMD APUs because APUs highlight energy efficiency. However, I have a little confusi...
    acekiller
    last modified by acekiller
  • Ocl debuggers

    In the User's Guide there is a whole section on CodeXL, with very interesting points. There is another debugger, opencl-emu, for which I know very little. Is it any good? How does it compare to CodeXL? Which one would...
    nibal
    last modified by nibal
  • clEnqueueWriteBufferRect does not work when region width is not equal to src pitch: broken again in Catalyst 14.12

    I have already reported this problem in an earlier post: http://devgurus.amd.com/thread/160312 and the error has been fixed in Catalyst 13.4. However, the error seems to be reintroduced in Catalyst 14.12. The error ...
    timchist
    last modified by timchist
  • Can I keep a kernel running all the time on Kaveri ?

    On Kaveri, is it possible to launch a kernel once and keep it running all the time?   Any help will be great appreciated~
    humasama
    last modified by humasama
  • Missing libraries from omega catalyst?

    When I try to run an ocl-c file, I get:   libGL error: dlopen /usr/lib/fglrx/dri/swrast_dri.so failed (/usr/lib/fglrx/dri/swrast_dri.so: cannot open shared object file: No such file or directory)   There i...
    nibal
    last modified by nibal
  • Do you plan to keep AMD Core Math Library (ACML) for Linux built with Intel® FORTRAN?

    Hello all,   I'm very interesting to use the latest ACML version (6.1.x) for Linux with Intel FORTRAN support. Can I expect such release or Intel FORTRAN support is dropped on Linux?   Other question is a...
    oasanov
    last modified by oasanov