Register spilling problem
========== Build: started ========== OpenCL Compile Error: Compiling for device: Turks Warning: expand kernel has register spilling. Lower performance is expected. -------- ========== Build: 1 of 1...
View Articleopencl on Radeon 6970
Hello, I'm trying to get opencl up and running on my new Radeon HD 6970. I installed the newest version of catalyst (13.4) and AMD APP SDK (2.8.1.0) on Windows 7 64bit. When I try to run various...
View ArticleAMD APP 2.8.1 and VS2008
Hello, The release notes for AMD APP 2.8.1 say (quoting): 'Support for Microsoft Visual Studio 2008 is deprecated.' Could you please elaborate on what that deprecation refers to practically speaking?...
View ArticleClik here to view.
SDK 8.2.1 - Compiler crashes/Undeclared function index 1036 (with...
After upgrading drivers/catalyst control center (which allowed me to debug some kernels) I now get some compiler crashes.... The attached kernel code crashes in kernel analyzer (and crashes in code),...
View ArticleGlobal synchronization inside the kernel
Hi, I'm facing with the following problem: I have to use LDS for a relatively long time and also I need to gather/scatter data across all the LDS memory.Scheduling more than one kernel is not an option...
View ArticleOpenCL usage as Window's service
Is it possible to use OpenCL in a Windows's service without being logged?
View ArticleClik here to view.
How to implement cl_khr_icd?
Hi, I want to run my application on both intel CPU and nvidia GPU. As I read some other posts, I clearly understood that I need 2 SDK for this configuration so I chose : AMD SDK and NVIDIA SDK. A part...
View ArticleOpenCL + 4870?
Please could someone help me get OpenCL to work on my 4870? I have installed the latest catalyst Driver 13.1 for my card :...
View ArticleReduced cache hit when I put a piece of code under loop!!
Hi, Card: 7970 Catalyst: 13.4APP : 2.8OS : Kubuntu 12.04 x64 Code snippet://for (i =0 ; i < 25 ; i++) encrypt(); When I comment the the loop the cache hit (tested using...
View ArticleApple's FFT on AMD cards with C++ OpenCL API
Hi, I'm trying to run Apple's clFFT on AMD cards using the OpenCL C++ API, tocreate the platform, context and devices, queues. It seems that it works onlyfor the C API, but not for the C++. On Nvidia...
View ArticleTrinity and Richland: cl_khr_fp64 confusion again
I think the situation about Trinity and Richland supporting cl_khr_fp64 remains unclear. On Trinity's GPU, the support for cl_khr_fp64 has certainly been there on at least some driver versions that I...
View ArticleSemaphore handle leak when host threads are destroyed on Windows
If one creates a context in the application main thread and then calls OpenCL functions utilizing this context from other threads (such as memory allocation/deallocation), OpenCL runtime creates 3...
View ArticleDS_WRITE GCN instruction
Hello! I've been trying to write GCN ISA assembly code by hand and I just can't get the "DS_" instructions to work.The docs said that the address shouldn't be the same in all threads, because it causes...
View ArticleClik here to view.
Ubuntu 12.04/12.10 - Enable OpenCL with headless/ssh with few easy steps...
I thought it might be useful for everyone to put this on a thread. X must be running obviously, but you dont have to login to run OpenCL programs and any user who can ssh into the box can access the...
View ArticleKernel performance variability with the same code and same data
Platform:OS: win7 64bitSDK: APP SDK 2.8.1CodeXL: latestedDriver: Catelyst 13.4 Problems:100 times kernel execution with same input data, there are 2-3 Kernel (besides the first one) execution time is...
View ArticleHSA and it's collateral benefits
Hi! There was a session by Greg Stoner of AMD, and in the news report by Phoronix it can be read that soon there will be a release of LLVM >>HSAIL backend, as well as a C++AMP compiler front-end...
View ArticleFastest device to host transfer
My question is this: how to achieve the fastest device to host transfer speed. The short answer is pinned memory, however my problem is a bit more complex. I have a piece of device memory which I have...
View ArticleWhy does increasing the number of kernel arguments impacts performance ?
GPU: 7970OS: Kubuntu 12.04 x64Driver: 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...
View ArticleI have bought a Amd / Ati Brecht van Lommel (creator of cycles to Blender).
Ihavebought aAmd/AtiBrechtvan Lommel(creator ofcyclestoBlender).I'd thinkitwould be goodifamdwould givegraphicsprogrammersalguanleastlow-endso they canprovide solutions toproblems suchcycles.Andthe big...
View ArticleHelloWorld Error: access violation
Hi,I'm new to opencl. I'm using vs 2008 on windows 7. I installed AMD APP SDK 2.8 and started to run the demo helloworld in debug mode .But it breaks down with an access violation and the error happens...
View Article