https://github.com/ttarter3/CallbackFFT.git
to build and run the code on a ASA-X type
module load git
git submodulate update --init
make pgen
make run_p
to build and run the code on Voyager1 type
# Make sure to run:
. /data001/heterogene_mw/spack/share/spack/setup-env.sh -> ~/.bashrc
git submodulate update --init
make sgen
make run_s
assuming python3 is installed, you can run the following to see the plot of the generated data in terms of SNR:
cd ./python_scripts
./runPlotFFT.sh
The code compares the calculated values to cuFFT. If the average error is greater than the threshold, then we write out a error message to the screen. Either "Error: Invalid Vector Size" or "Error: Measurment Error Greater than expected(###)". There are also data checks to make sure that the algorithm you are requesting to use can support the size of the data being used. This error message looks something like "Error: Radix # must satisfy 0 == log2(M)"