OCLSocketError when debugging Kernel

[Edgardo_Doerner]

Well, I think that this is my first post on the forum so hi! to everyone.

I have an issue when debugging an OpenCl kernel, when I try to enter a function declared outside it and invoqued inside the kernel I have the following error:

First-chance exception at 0x763cc41f in OpenCL_EGS.exe: Microsoft C++ exception: OclSocketError at memory location 0x02dff9f0..
The thread 'Win32 Thread' (0x11b4) has exited with code -1 (0xffffffff).

In general this error is quite random, sometimes I am able to enter and execute some lines inside the function before the error appears. But what is more weird is that if I do not enter the function and just execute the line where it is invoqued it executes without problem and behaves correctly.

I really need to be able to use this ability inside the function and it is quite frustrating to have this error time by time...

Thanks for your help!

[Uri_L_Intel]

Hello, Thank you for using the forum! In order to help us with further investigation can you please provide us some more information on your environment: - Which version of the SDK our you using (2012 or 2013 beta) - Which version of OS are you working on (32-bit or 64-bit) - Is it possible to send a code snippet of the kernel code that you’re trying to debug (it will help us to reproduce it on our environment) Thanks, Uri

[Edgardo_Doerner]

I have the error with both versions (2012 and 2013 beta). I had the beta first but then made a downgrade in order to test the platform. The error is less recurrent, but it still exist. The OS version is Windows 7 Professional 64-bit, but I compile the program with the 32-bit libraries. The program is an implementation of a Monte Carlo simulation of photon transport. There is a lot of input data that is transferred to the device (for the moment only my CPU, a Core i7 on a Lenovo T420) that is read from inputs files, therefore it is better to send the entire project. I attach a file with the code. thanks for your help!. Thanks for your help.

[Uri_L_Intel]

Thanks for the information!

We were able to reproduce the failure and find the root cause.

We hope to have the fix available on our next SDK release.

Thanks,

Uri

[Edgardo_Doerner]

Well, it is great to have contributed a bit with the developing process of the SDK...