Hi. I'm developing an application for a custom board using MKL26Z64. The application involves an FFT. To do so, I included the CMSIS DSP Library.
However, when I use the arm_rfft_init_q15 and arm_rfft_q15 functions to perform a fixed-point real FFT, the code (which was 8384 bytes before) overflows by 23124 bytes. When I remove those functions, there's no overflow.
I'm using CodeWarrior 10.6 and Processor Expert 10.3.
Any tips on this? Do I have to use another FFT implementation?
Thanks.
Solved! Go to Solution.
Hello Gustavo Leal:
Check the next thread and especially the Correct Answer by colleague Martyn Hunt. Maybe your project is linking the whole DSP library and you have to use the -ffunction-sections -fdata-sections compiler options:
Has anybody implemented CMSIS FFT library Successfully?
Regards!,
Jorge Gonzalez
-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------
Hello Gustavo Leal:
Check the next thread and especially the Correct Answer by colleague Martyn Hunt. Maybe your project is linking the whole DSP library and you have to use the -ffunction-sections -fdata-sections compiler options:
Has anybody implemented CMSIS FFT library Successfully?
Regards!,
Jorge Gonzalez
-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------