Use FIR accelerator to perform convolution

Hi,

I want to use FIR accelerator to perform convolution filter.

I picked the example FIR_LowPassFilter from Board Support Package.

I executed the example and it worked fine.

Now my data size is different (Input Data Length is 2048 and filter coefficient length is 21).

So I changed the TAP_LENGTH to 21 and WINDOW_SIZE to 2048. And corresponding data was also changed in .dat files.

Now when the program is run, it fails with error message "Test Failed adi_fir_GetInputBufferIndex".

Please help in letting me know the reason for this failure.

Also,

How does FIR accelerator copies data from memory buffer? Does it use DMA to copy data?

Thanks

Edit Notes
Question elaborated

[edited by: yogeshkverma at 10:30 AM (GMT -4) on 25 Oct 2018]

sc589

sharc

fir

evaluation_board