OpenCL GPU access on Windows over RDP/Remote Desktop

I would like to be able to detect and access the Intel GPUs using OpenCL even over RDP/Remote Desktop sessions so that I can deploy applications that can utilize the full system capabilities even without direct console access. Right now, it appears that I would have to switch to use an alternate remote access client like VNC to keep the Intel video drivers loaded and keep the GPU accessible.

Events not getting their ref count incremented

When I pass an array of 2 wait events into clEnqueueReadBuffer, with one event created manually, the ref counts are not incremented.

As I understand, all events passed into cl methods should get their ref counts incremented.

This works correctly with my GPU, but not for CPU (WIndows 7 64 bit,  core i7 3770, latest Intel OpenCL driver for CPU)

clSetEventCallback Oddity

I suspect there's an aspect of clSetEventCallback that I'm misunderstanding. If you have a look at this simplified code below, you can see a kernel call, followed by a clSetEventCallback. The program then sits and waits on a windows event (inside the callback) to be set.
If the clSetEventCallback is immediately followed by a clFlush(), then the callback is called and the wait released as expected. However without the clFlush, the kernel is never called, nor its callback, and the wait is eternal. What am I missing here? 

Error when I run occupancy for kernels

I am using Visual Studio Community 2013 Version 12 Update 4 and OpenCL Code Builder on 64-bit laptop (Windows 8.1, Intel Core i5-5200U and GPU HD Graphics 5500).

When I try to run occupancy for kernels, I get a script error. On line 2863, character 98,  ":" was expected.

How can I fix the problem?

Thanks in advance



OpenCL drivers releases and bug report contact


My company is in the process of validating our software against Intel OpenCL drivers (Xeon only, not Xeon phi)

We are currently relying on the version for Linux and for Windows.

Windows driver works like a charm, but we are struggling with instabilities using Linux (Like kernels that work on a lot of other platforms (Including Intel's windows crashing in a non deterministic fashion... Sometimes they work fine, sometimes not so good...)

Bug in OpenCL Code Builder's Debugger

To start this post off, I would like to say that I do not take the term "bug" lightly - it gets thrown around on the internet often, but usually when someone does not understand the tool set. I hope this post is not one of those, but my research tells me otherwise. TL;DR clause: there appears to be a problem with get_global_size(dim) from the debugger.

Lets start by defining a simple kernel (emphasis on simple) - set the value of the array element in an in/out buffer to its associated linear index:

Memcpy performance using opencl kernel


I have written a simple memcpy kernel as written below:

I am analyzing its performance on GPU using vtune.

__kernel void deinterlace_Y(__read_only image2d_t YIn, __write_only image2d_t YOut)

/* Doing operation of Memcpy */

int2 coord_src = (int2)(get_global_id(0), get_global_id(1));


uint4 pixel4 = read_imageui(YIn, smp, coord_src);

write_imageui(YOut, coord_src, pixel4);


I observe the below stats for Execution units:

Visual Studio Community 2013 crashes when building program from Code-builder

I am using Visual Studio Community 2013 Version 12 Update 4 and OpenCL Code Builder on 64-bit laptop (Windows 8.1, Intel Core i5-5200U and GPU HD Graphics 5500).

I have created a Code Builder project for Windows in C++ and I can build (Configuration: Debug Win32) and run it without any problem.

I have created a session, and Visual Studio always crashes when I try to use following options

订阅 OpenCL*