Segfault on clBuildProgram for simple CPU Kernel
Hi all, I have run into a simple, reproducible segfault that occurs during compilation of a kernel, i.e. in clBuildProgram(). I am using the intel opencl-1.2-6.4.0.24 runtime on an Intel(R) Xeon(R)...
View Articledevice fission on gen9 GPU
Is device fission is supported on Gen 9 GPU?I want to run multiple OpenCL kernel concurrently on GPU. Is there any way i can run?Thanks,Biren Doshi
View ArticleAdvanced Motion Estimation Extension and GPU
Hi Can we use Intel advanced motion estimation accelerator and GPU computing power concurrently? Thanks,Bieren Doshi
View Articleioc64 installed on Fedora 25 throwing instance of 'std::string', aborting
I am working on getting the Intel OpenCL SDK up and running on a Fedora 25 Skylake machine (Core i7-6700). So far I have:Installed the xorg-x11-drv-intel package from the Fedora (or maybe RPMFusion)...
View ArticleHD Graphics 5500 is unavailable with CentOS7.2 + intel-opencl-r3.1 driver
Hi,I am trying to enable the HD Graphics 5500 of i7-5600U on CentOS7.2. I posted a similar inquiry to the following Intel's community and had a suggestion to ask at this forum:No "CL_DEVICE_TYPE_GPU"...
View ArticleSame instruction on all 8 EU?
To get peak performance, all EU in single sub-slice should issue same instruction or in single EU only we need same instruction? At what granularity i should avoid branching ? Thanks and regards,Biren...
View ArticleUninstall Intel OpenCL Platform?
Machine Details:Windows 10CPU: Intel Core i5 4670K Quad CoreIntegrated Graphics: HD 4600I've been developing an AMD OpenCL application using their SDK. I use VS2015 to compile, then I test on a...
View ArticleUbuntu: application detects 1 device while clinfo detects 2
Hi I've installed both CPU and GPU Intel OpenCL drivers on a x86_64 machine running Ubuntu 16.10 and I've encountered an unexplainable (yet) behaviour on my OpenCL project. My application only detects...
View ArticleOpenCL lists two Intel (R) OpenCL platforms
On a laptop running Windows 10 64 bits with an Intel HD GPU and NVIDIA GPU, my OpenCL program and one of the Intel SDK OpenCL sample program both lists two "Intel (R) OpenCL" platform, with same...
View ArticleclCreateContext fails on CPU but not on GPU
On a laptop running under Windows 10 64 bits and dual Intel GPU / NVIDIA GPU configuration, my OpenCL program returns a CL_OUT_OF_HOST_MEMORY when creating an OpenCL context with the "Intel (R) OpenCL"...
View Articlesub_group_broadcast() broken on GEN9 (21.20.16.4552)
I have a kernel with a "required subgroup size" of 8.My test is launching a grid of 24 global work items and 8 local work items (only for testing purposes).After much debugging, the...
View Articlegdbserver-igfx "failed to initialize"
Hello,I'm trying to use OpenCL™ Kernel Debugger as explained here: https://software.intel.com/en-us/node/671873 I have a target as follow:Ubuntu 16.04LTSIntel Celeron J3160Intel® SDK for OpenCL™...
View ArticleIGIL vs GEN
Hi,I see that on OS X, Apple CL compiler has an IGIL target. Is IGIL code the same as GEN code?Is there a way to inspect IGIL/GEN programs resulting from building a CL kernel on OS X? This has been...
View ArticleClik here to view.
ioc32 build failure
ioc32.exe (6.3.0.1904) exited with code -1073741819 (0xc0000005) and a stack dump when compiling surf.cl from OpenCV 2.4.x. See the attached log file. Platform is VS 2013 on Windows 10.I narrowed the...
View Article__private memory, spills and loop unrolling on HD Graphics
I have a few questions:If IOC reports that private memory is being used then does that always imply that it's being spilled or can it reside in the remaining registers?How do I detect or analyze my...
View Articleperformance of half2 vector vs. half scalars per SIMD8 lane?
Basic question for GEN8+ experts:In a SIMD8 kernel, does the GEN8+ EU achieve maximum fp16 throughput with half2 vectors per SIMD lane or are independent half scalars going to be better/worse/same? I...
View ArticleClik here to view.
Eclipse Code Builder - kernel arguments and configuration locked
Hello,I 'm currently trying to deploy Intel OpenCL on the GPU of my Intel(R) Xeon(R) CPU E3-1505M v5 @ 2.80GHz on Centos 7.1. My configuration is :- intel_sdk_for_opencl_2016_6.3.0.1904_x64.tgz- driver...
View ArticleCL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT on Intel HD Graphics 400
Hi,I have a target as follow:Ubuntu 16.04LTSIntel Celeron J3160 (GPU Intel HD Graphics 400)Intel® SDK for OpenCL™ Applications 2016 R3Intel driver for OpenCL - intel-opencl-xxx-r3.0-57406.x86_64intel...
View ArticleCL_KERNEL_SPILL_MEM_SIZE_INTEL interpretation?
I'm trying to hunt down spills in my kernel.Is there a more detailed description of the CL_KERNEL_SPILL_MEM_SIZE_INTEL query (link)?Is this byte value per work item, sub_group or work_group?I ask...
View ArticleIntel GEN SLM allocation granularity still 4KB per workgroup?
Given a kernel that uses no barriers, does this recommendation still hold for GEN8 and beyond?https://software.intel.com/en-us/node/540442 NOTEA bare minimum SLM allocation size is 4k per workgroup, so...
View Article