• Possible bug in read_imagef

    Strange behavior - kernel works fine with this: __constant sampler_t imageSampler = CLK_NORMALIZED_COORDS_FALSE | CLK_ADDRESS_NONE | CLK_FILTER_NEAREST, using this - read_imagef(image, imageSampler, coords); but!.. ...
    last modified by sandyandr
  • How GPU handles its code?

    I mean how kernel binary passed to GPU (on context creation, before kernel launch and so forth)? Where it stored? (CPU uses usual system memory for code storage but has separare L1 instruction cache to speed up instr...
    last modified by Raistmer
  • URGENT: switchable graphics and resources sharing

    We have some troubles on AMD Switchable Graphics system with Direct3D surfaces sharing technology. To be clear, Direct3D can find only single adapter and I will mean High Performance mode by "AMD GPU" and Power saving...
    last modified by vladimir
  • Having Inspiron 15R with gnome ubuntu 14.04 graphics issue

    I am having ubuntu 14.04 gnome version with dual boot windows 7. No ATI/AMD graphic driver is getting installed on my system. I tried everything and it crashes my complete system. I am having AMD 8850 graphic card. Pl...
    last modified by rohi222
  • ACML and iGPU of Kaveri

    Will (in the future) ACML use iGPU in Kaveri, for example, to execute instructions BLAS?. I use ACML instead of  BLAS (package "libblas3") under Ubuntu just linked by  "update-alternatives" (as described, f...
    last modified by huber
  • CodeXL fails to find AMD OpenCL runtime (among other things)

    Hi!   CodeXL hasn't been of much help since it got released, since I haven't seen the debugging part of it work a single time. I reported bugs in the installer, at least those went away, but the bigger problem, ...
    last modified by Meteorhead
  • Why does increasing the number of kernel arguments impacts performance ?

    GPU: 7970 OS: Kubuntu 12.04 x64 Driver: Catlyst 13.4   Hi,   I know this sounds crazy but it is actually true. Increasing the number of kernel arguments beyond a certain number causes the performance to ...
    Sayantan Datta
    last modified by Sayantan Datta
  • Values on Radeon HD 7520G running less than optimum performance

    Hi, my video card is not being utilize completely. the same thing happened to the HD 7870 on update 13.1 /, and this has been a problem for around two months. I don't think too many people updated, but ...
    last modified by vampiredragon
  • Why only 256 workitems per workgroup for ATi GPU?

    For CPU AMD supports 1024 workitems. NV supports 1024 for GPUs from very beginning... AMD has same or larger amount of shared memory, same or larger register file... so why such limitation? Why only 4 waves per wor...
    last modified by Raistmer
  • Scratch registers - how to prevent their usage?

    I use fixed-size array in registers to reduce fetch size required by kernel.   At some size (11 elements) kernel performance dropped considerably (3 times slowdown) and 22 scratch registers were used.   Ke...
    last modified by Raistmer
  • Bug in clEnqueueTask?

    I was using CodeXL to do some profiling and it kept saying:   Opencl Memory leak detected [Ref =1] Object created by clEnqueueNDRangeKernel   Eventually it was narrowed down to this line:   status = ...
    last modified by jasonwhite
  • x264 demo not working on AMD platform

    Hi,   I have downloaded a OpenCL enabled x264 encoder from Universität Heidelberg , compilation was fine on AMD platform but it failed to run on Ubuntu_13.10 + HD7970, segmentation fault. Sharing the compi...
    last modified by scharupa
  • clBuildProgram crash

    This kernel runs fine on my computer on a CPU device and the CodeXL Kernel Analyzer verifies it should work on a wide variety of cards - everything HD5k or newer for image support. I'm unfortunately currently stuck wi...
    last modified by msqrt
  • Bad performance FX on ACML-multithread (Linux)

    Hi! I use GetDP with MUMPS (4.10) and ACML (5.3.1) instead BLAS (ATLAS, OpenBLAS). I use "gfortran64_fma4_mp" version of ACML. When I set environment variable OMP_NUM_THREADS=4 (or more) on system with FX-6300 or OMP_...
    last modified by huber
  • Unable to compile ACML in AMD processor terminal

    When ever I am trying to compile the program written in C with gcc compiler in Linux system to generate Gaussian Random Number in Parallel threading (using skip ahead method) in AMD processor using ACML library then t...
    last modified by sam215
  • ACML 6 Beta released

    Hi all,   ACML 6 Beta is now available to download at http://developer.amd.com/tools-and-sdks/cpu-development/amd-core-math-library-acml/acml-downloads-resources/ For more details of features please refer to a...
    Timmy Liu
    last modified by Timmy Liu
  • Why do I not get more Opengl handles with additional cards.

    My questions are: What determines the number of contexts that OpenGL can create and is there any way to boost this value. Also, why doesn't having 3 cards help.   Here is the situation: I am working on a system ...
    last modified by skelley
  • Blender Cycles(Opencl on AMD GPUS)

    Dear Opencl Developer Why doesn't the AMD opencl compiler work with blender cycles? whenever i compile the blender cycles kernel, the system either crashes due to lack of memory, or takes too long to compile the ble...
    nazim mer
    last modified by nazim mer
  • Driver graphics card ati Radeon X1900 pour W8.1?

    Hello,   I'd like your help to find a driver for my graphics card Ati Radeon X1900 for Windows 8.1 X64. Windows installed the generic microsoft driver for this card but the resolution is very low. Thank y...
    last modified by egval
  • AMD Radeon 2400 HD Windows 8.1 x64 Driver

    Hi, let me ask you a question,   Are you developing the amd radeon 2400 hd driver for windows 8.1 ? I would like to upgrade my pc and everyday i come to this site to see if it has the driver, please let me kn...
    last modified by pedroxaz