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?...
View Article1 bits count in a variable
Hi everybody,I was wondering if there was an integer function to count the number of 1 bits contained in a variable.I saw the "clz" function allows to count the number of "leading 0 bits". If it was...
View ArticleWavefronts sharing same local memory allocation
I have a 32 x 4 work group, and each row of 32 of work items uses a block of 33 bytes of local memory.So, I have allocated 33*4 == 132 bytes of local memory for each work group. Is it possible to...
View ArticleArrayFire benchmarks - Kaveri vs Haswell
ArrayFire published some GPU benchmarks on a collection of ArrayFire functions.See the results here: ArrayFire Benchmarks: AMD Kaveri vs Intel Haswell Part-1 | ArrayFire (Hint: The AMD processor did...
View ArticleQuality of FirePRO OpenCL drivers on linux
Can anyone comment on the quality of FirePRO OpenCL drivers on Linux?How does it compare to windows, performance-wise? Stability?
View ArticleComprehensive instruction latency table
Hi! I am creating slides for a university course and I was looking to compare various instruction latencies on CPUs and GPUs. Inside the OpenCL Optimization Guide, there is a very short table for VLIW...
View ArticleBufferBandwidth results on Kaveri
Hello,I was wondering why the CPU read/writes are so slow on the BufferBandwidth example irrelevant of if the memory is allocated in host or not? Also why the GPU writes are slow if the kernel is...
View ArticleWhy not a clWaitAnyEvent?
The behavior of select(...) is to wake up when at least one watched descriptor is "ready".Pthreads take it easy with a single condition variable to pthread_cond_wait. Windows has...
View ArticleGuide to run OpenCL headless, without X server and as normal user
EDIT: this guide is no more relevant as from catalyst 14.12 omega it is supported to run OpenCL applications without Xserver. read discussion below. You may know that you can run OpenCL program as root...
View ArticleLinking clFFT
Hello, I am new to ocl. I am running an HD 7950 and trying out various tests to see ifI can speed up my fluid flow simulations which are in fp64. Basic ocl tests are ok, but I got stuck with the...
View Articleproblems with OpenCL constants and OpenCL compiler on Linux
I am having a strange problem when building a kernel on Linux targeting the CPU device in the AMD OpenCL platform - if the kernel code contains a reference to one of the predefined OpenCL constants,...
View ArticleClik here to view.
Crash on clGetPlatformIDs when AMD runtime can't create temp file
When calling clGetPlatformIDs AMD OpenCL runtime tries to create temporary file (version 14.12 tries to do this in TMP folder, earlier versions used current(!) folder). Now suppose it failed to do so,...
View ArticleCannot make OpenCL runtime expose more than 3 GB of RAM
On GPUs having more than 3GB memory aboard it does not seem possible to make OpenCL runtime expose more than 3 GB of RAM. Changing the GPU_MAX_HEAP_SIZE variable as described here only allows to...
View ArticleHow to use mangled name of templated kernel, specified in custom namespace?
The problem is: when I use Static C++ OpenCL kernel extension I give mangled names to variants of templated kernel, e.g:template<typename T>__kernel void vec_add(__global T *A, __global T *B,...
View ArticleDocker image to support OpenCL env on AMD APU
Hi, dev I had created a stuff to try to put OpenCL env into Docker support, this can be used to combine such as Apache Aurora or Docker swarm to run our opencl application across you cluster node that...
View ArticlePerformance of work-item built-in functions
Questions on performance of work-item built-in functions. I'm talking about these:get_work_dimNumber of dimensions in useget_global_sizeNumber of global work itemsget_global_idGlobal work item ID...
View Articlerelative importance of bank conflicts in local vs global per optimization guide
Hello, something is not clear to me from going through the optimization guide. My question is whether bank & channel conflicts are primarily a concern using OpenCL local memory, or if they are also...
View ArticleInstalling APP SDK 3.0 on Xubuntu 14.10
Hi peeps, I'm trying to get OpenCL to work on Xubuntu 14.10. I installed the AMD APP SDK 3.0 as per the instructions here. I can see the environment variables have been set correctly and I did log out...
View ArticleGPU Compute Cloud Services
Hello everyone, I wanted to open a conversation and get some feedback on the seeming unavailability of a compute cloud featuring AMD GPUs. I have been working on an OpenCL project for over a year now...
View ArticleFirePro W5000 error OpenCL Premiere render
Recently thatbought aFireProW5000, after watchingseveral videosof hiscompanysaying that itworksfasterway, accelerating therenderingvideoswithOpenCLin PremiereProCC, including the sitesays itwill...
View Article