• Is there an elegant way to force recalculation (of values or addresses)

    Well the question in the title already hits it. I got a rather simple kernel, which uses 20 vgpr and the complete 32 kByte of shared memory. So all fine for running 2x 1024 threads per work group. So fine so far. Bu...
    last modified by lolliedieb
  • Why did AMD decide to remove SPIR 1.2 support without adding SPIR-V?

    We have found, like many others, that the AMD GPU may report OpenCL 1.2 or even 2.0 support, with SPIR. However, that may not actually be the case. We're aware that you're looking into this, and will remove reported s...
    last modified by torbsorb
  • OpenCL & DirectX12 Interoperability

    AMD supports OpenCL extentions such as clGetDeviceIDsFromD3D11KHR, clCreateFromD3D11BufferKHR & clCreateFromD3D11Texture2DKHR to support OpenCL & D3D11 resource sharing and synchronization. ...
    last modified by elad
  • OpenCL single memory allocation limit

    AMD OpenCL limits a single buffer allocation (CL_DEVICE_MAX_MEM_ALLOC_SIZE) to 50% of total memory.   So on a 8Gb card there is only 4Gb available for allocation in a single chunk.   This is very sad for ...
    last modified by octoboar
  • Feature request: expose newer AMD GCN / RDNA features as CL extension

    Back in the early days of OpenCL AMD added the famous cl_amd_media_ops (2) to expose hardware features to the programmers. Sadly with some of there more recent or more hidden hardware features like GDS or the cross la...
    last modified by lolliedieb
  • Heterogeneous toolchain for Windows?

    Good day,   I am currently running windows with OpenCL kernels across CPU(2990WX) and AMD GPU with C++17.   As AMD stopped support for OpenCL on CPU how can I adapt my tool-chain to still leverage the CPU ...
    last modified by genestoltz
  • Please add support for image atomics in OpenCL

    Hi,   I have a kernel where I accumulate a lot of values with atomics. The values to accumulate are in a 2D neighborhood, and neighboring threads treat similar regions, but with a small random (x,y) shift. and t...
    last modified by mannerov
  • Optimize LC0 - Leela Chess Zero - for AMD GPUs

    Heyho AMD community,   we are all aware about the neural network hype on gpus, and most have noticed that Nvidia has simply the forehand with their cuDNN framework.   Personally I am convinced that AMD mak...
    created by smato2018
  • What the error "HSAIL doesn't support OpenCL extension spir" means?

    I am trying to use SPIR in Windows with the latest AMD Adrenalin 19.5.2, but since I upgraded to the latest version I'm getting this error in the call to clBuildProgramWithBinary   Error: HSAIL doesn't support Op...
    last modified by tluisrs
  • Current status of OpenCL for a SI (R7 370) in Linux

    Hi,   As I asked in Linux OpenCL not working AMDGPU-PRO (max global size 0, CL_OUT_OF_HOST_MEMORY), the problem persist.   I tried both in Arch and in Ubuntu 18.04. If I go for mesa + amdgpu open source dr...
    last modified by userxx
  • SPIR support in new drivers lost

    I already ask this question in Drivers & Software section but nobody answer.   --------------------------------------------------------------------------------------------------------------------------------...
    last modified by ipse
  • line-by-line profiling

    I am wondering if there is a profiler for OpenCL on the AMD devices that supports line-by-line profiling? For CUDA, nvprof already has the PC sampling profiling option that gives per-line run time info; for OpenCL, ri...
    last modified by FangQ
  • OpenCL CPU runtime

    Hi all,   Since AMD dropped support for OpenCL SDK, is there alternative option for OpenCL runtime on CPU? I'm well aware of GPU options, but seems like removal of SDK means also that there is no more CPU runtim...
    last modified by expro
  • CL-GL Interop fastest way to synchronize?

    We are using OpenCL on Windows as part of a proprietary game-engine where we use the CL-GL interop functionality to communicate between the simulation and the rendering engine. Our core loop currently executes the fol...
    last modified by george72
  • Is there any way to split a compute device (emulate two, or more, GPUs on the a single device)?

    For example I have a RX Vega 64 with 64 CUs and 8 GB VRAM, would it be somehow possible to make it appear like two GPU compute devices with 32 CUs and 4 GB VRAM each? Maybe some undocumented environment variable for d...
    last modified by bomby
  • Any instruction level or line-by-line profiler?

    It will be very helper if we can analyze the cost of each instruction or each OpenCL line. Either ROCm or AMDGPU driver is fine. Thanks in advance.
    last modified by fancyix
  • List of neural network/machine learning/GPU computing apps that support OpenCL acceleration on AMD Fx HW?

    Hi, I have a few questions. I hope you can help me.   I am trying to learn neural nets/ML on my older, Fx based hardware.   I very much prefer the openCL development model. As discussed elsewhere, people ...
    last modified by devlista
  • Please add new extension for refined reduce in wavefront

    Hi,   According to https://gpuopen.com/amd-gcn-assembly-cross-lane-operations/  the hardware is able to do refined reduce operations.   By 'refined', I have in mind doing an add/min/max among neighbor...
    last modified by mannerov
  • OpenCL with SVM extensions on Linux for modern APUs?

    Hi,   I'm evaluating OpenCL-accelerated OpenCV on V1807B (Raven Ridge APU) and am wondering what options I have to get SVM support on Linux on APU.   It seems there are multiple approaches: - a fully open...
    last modified by epvbergen
  • Missing OpenCL CPU support under Windows

    On my system the (i think) most recent version of the AMD drivers (18.8.1, Windows 10 x64) no longer returns the CPU (FX-8350) as a valid OpenCL device.  Is this intended behavior or just a bug in my specific ins...
    last modified by pangea