crash when setting kernel argument for dynamic local memory
Hello List,I am trying to pass dynamically allocated local memory to my kernel.So, I am using the line clSetKernelArg(myKernel, 2, localMemSize, NULL);where argument two in the kernel is prefixed by...
View ArticleSubdevice Support on Intel i5-3427U Ivy Bridge?
Hi all,I am trying to divide my CPU into subdevices to solve a task parallel problem. Unfortunately I keep getting those CL_DEVICE_PARTITION_FAILED errors.Here is what I tried: cl_platform_id...
View ArticleOpenCL related question
Hello! I have restarted some of my experiments on the Intel Haswell processor and some of them stopped working, namely the ones related to examples meant to be executed for the GPU.My main question is...
View ArticleOpenCL related question
Hello! I have restarted some of my experiments on the Intel Haswell processor and some of them stopped working, namely the ones related to examples meant to be executed for the GPU.My main question is...
View ArticleClik here to view.
OpenCL Kernel Fails only on 4th Generation Processors
Hello,I am working on creating a N-Body gravitational simulation using Intel OpenCL SDK. I have attached our Kernel. This kernel fails to execute only on the systems with Haswell (4th Generation) Intel...
View ArticleWould you provide sample(Tutorial) for OpenCL 2.0 which include SVM and...
Would you provide sample(Tutorial) for OpenCL 2.0 which include SVM and Pipes, you know the product of Intel® SDK for OpenCL™ Applications 2014 can support the OpenCL2.0, and we begin to use it on...
View Articlecl_command_queue reference count does not track cl_event instances and leads...
I have discovered what I consider to be a bug in the reference tracking for cl_command_queue instances that leads to a segmentation fault under a reasonable usage scenario.The cl_command_queue...
View ArticleClik here to view.
HD4600/5200 doesn't show support GPU for OpenCL1.2
Hello, On I7-4770 (HD4600) and I7-4860EQ (HD5200) we install Win7 x64 with driver 15.33.18.3496, then install Intel SDK for OpenCL Application 2013 r3 x64. In SDK Kernel Builder Menu 'Device info', it...
View ArticleClik here to view.
clEnqueueNDRangeKernel may fail when using 2D local arrays
clEnqueueNDRangeKernel may fail on kernel with 2D local arrays but succeed with 1D local arrays and manual index computing.For example, the following matrix multiplication kernel fails with...
View Articlework group with 1 work item using ~100 float8 vectors?
Will the Intel HD Graphics OpenCL compiler support "1 work item" work groups that are float8 vectors?Example:__kernel __attribute__((vec_type_hint(float8),reqd_work_group_size(1,1,1))) void...
View ArticleCL_DEVICE_PREFERRED_VECTOR_WIDTH for Intel devices
Hello.I have tried to make some OpenCL-related performance optimization for Intel devices. I want to use vectorization and vector data type with optimal lenght for specified device. I called...
View ArticleOpenCL 2.0 Support
Hello!Is there a time frame for release of OpenCL 2.0 support? Also, will this require new silicon, or will I be able to run 2.0 on my existing 3rd gen CPU when it gets released.Thanks!Aaron
View ArticleXeon Phi: HW Exception: Segmentation Fault in all examples
Hey,I just updated my Phi to the latest MPSS version (3.2.1) and also the OpenCL Runtime (14.1) as well as the SDK (2014 4.4.0).Since then, every OCL example and code will crash when I let it run on...
View ArticleKernel Builder "Deep Analysis" function?
Are there hardware prerequisites in order to use the Kernel Builder's "Deep Analysis" function?It's throwing a "-34" error on my Ivy Bridge HD Graphics (6 EU) test machine.
View ArticleMeasuring occupancy and bank conflicts
Does the Intel SDK provide any tools to measure occupancy and bank conflicts?Thanks,Aaron
View ArticleHD4000: Can it load 4 floats at one clock ?
Hello,I'm running OpenCL on Intel's HD4000 GPU.It seems that this GPU has no vector calculations.According to the query:CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOATThe vector width is 1. Which means that...
View ArticleAddress bits: CPU = 32, GPU = 64?
I was playing around with the new 2014 Intel OpenCL SDK, and I noticed that when compiling for Win32, CL_DEVICE_ADDRESS_BITS query returned 32 for the CPU, but 64 for the GPU. That seems quite odd to...
View ArticleAddress bits: CPU = 32, GPU = 64?
I was playing around with the new 2014 Intel OpenCL SDK, and I noticed that when compiling for Win32, CL_DEVICE_ADDRESS_BITS query returned 32 for the CPU, but 64 for the GPU. That seems quite odd to...
View ArticleRegister Spilling
Hi, I am currently working on one OpenCL kernel with exhaustive register usage. I suspect the register spilling hinders the performance of my kernel on ivybridge. Using vtune amplifer I only get 10% EU...
View ArticleAccess to older CPU-only OpenCL runtime packages for compatibility issues
Would it be possible to provide access to the previous releases of the Intel OpenCL CPU-only runtime packages (e.g., 2012 and early 2013 versions)?The problem is that since some point in 2013, the...
View Article