01-15-2016 06:49 AM
Hi everyone,
I'm trying to implement a parallel fourier transformation of my 2D data using the GPU Analysis Toolkit. Since I never used this tool I tried first to implement a simple fourier transform of a simple real signal to a complex output vector. For this I found an example on the internet and adapted it a little. Unfortunatly I don't get any output, but I don't have an idea why. I attached the VI so maybe someone could have a look? I would very much appreciate if some could help me with this issue.
Thanks to anyone who can help me
Simon
Solved! Go to Solution.
01-19-2016 02:14 AM
Hi,
do you get any errors?
At which point exactly is no output there?
Did you already use this example for reference: labview\examples\lvgpu\Multi-channel FFT?
Best,
Anna
01-19-2016 03:19 AM
Hi Anna,
this is the strange thing. I don't get any errors. I don*'t know how I can check if the FFT in the CUDA part of the VI is working, but I don't get any entries in the array after downloading the data from the CUDA environment. I also tried the example von NI as you mentioned (see attached file). I'm not sure if the output is correct but I don't get any errors. Because I wasn't sure if the CUDA Toolkit is working correctly I did some tests with example code (C++) provided by nVidia and it worked well. So now I am thinking the problem could be a misstake in the posted VI or the communication between Labview and the graphics card. But I have no idea how to test this.
Simon
01-19-2016 09:28 AM
After an additional trial I got an error message. But I still can't figure out the problem. Here is the error message:
Error code: -359631
Message:
call to cudaMalloc in cudart32_75.dll.
<ERR>NVIDIA provides the following information on this error condition:
<b>code:</b>
cudaErrorUnknown = 30
<b>comments:</b>
This indicates that an unknown internal error has occurred.
<b>library version supplying error info:</b>
4.1
The following are details specific to LabVIEW execution.
<b>library path:</b>
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v7.5\bin
<b>call chain:</b>
-> lvcuda.lvlib:CUDA SGL Device Ptr.lvclass:Allocate Memory.vi:6140001
-> r2c_out-o-f-place_fft_ example.vi
<b>IMPORTANT NOTE:</b>
Most NVIDIA functions execute asynchronously. This means the function that generated this error information may not be the function responsible for the error condition.
If the functions are from different NVIDIA libraries, the detailed information here is for a unrelated error potentially.
Simon
01-20-2016 03:45 AM
A short information. I was able to figure out the problem. The problem was caused by a to new Version of the CUDA Toolkit. I installed Version 4.1 and now everything is working fine.
Simon
01-20-2016 03:54 AM
Great it works now.
Thanks for reporting that back.
Anna