我用fft_sp_1d_r2c这个例子,想评估FFTLIB库中fft_sp_1d_r2c这个fft函数中ecpy fft模式和direct fft模式 性能同时也能和DSPLIB库中DSPF_sp_fftSPxSP函数性能做对比,我该怎么加记录时间?
如图
下面是我自己添加的 请大神帮忙确认是否正确?
Shine:
为什么还要减t_overhead?
user5329195:
回复 Shine:
我是参考了ecpy 模式和 direct模式 计算时间时都减了t_overhead
其实我对这个t_overhead还有些不明白,有没有想象介绍的?
user5329195:
回复 Shine:
[TMS320C66x_0] N = 16384 DSPF_sp_fftSPxSP Cycle: 151780
FFT memory buffers: Buffer Size(bytes) Alignment 0 131072 3 1 131072 3 2 1032 3 3 0 3 4 4096 3 5 4096 3 6 1032 3 7 0 3 8 0 3 9 2048 3 fft_sp_1d_c2c_ecpy size= 16384max_diff = 0.017578 N = 16384 Cycle: 1609576
FFT memory buffers: Buffer Size(bytes) Alignment 0 131072 3 1 131072 3 2 8 3 3 0 3 4 0 3 5 0 3 6 0 3 7 0 3 8 0 3 9 0 3 fft_sp_1d_c2c_direct size= 16384max_diff = 6.367188 N = 16384 Cycle: 152897
N = 32768 DSPF_sp_fftSPxSP Cycle: 348407
FFT memory buffers: Buffer Size(bytes) Alignment 0 262144 3 1 262144 3 2 1032 3 3 0 3 4 8192 3 5 8192 3 6 1032 3 7 0 3 8 0 3 9 4096 3 fft_sp_1d_c2c_ecpy size= 32768max_diff = 0.032227 N = 32768 Cycle: 3061928
FFT memory buffers: Buffer Size(bytes) Alignment 0 262144 3 1 262144 3 2 8 3 3 0 3 4 0 3 5 0 3 6 0 3 7 0 3 8 0 3 9 0 3 fft_sp_1d_c2c_direct size= 32768max_diff = 14.457031 N = 32768 Cycle: 338085
N = 65536 DSPF_sp_fftSPxSP Cycle: 696576
FFT memory buffers: Buffer Size(bytes) Alignment 0 524288 3 1 524288 3 2 2056 3 3 0 3 4 8192 3 5 8192 3 6 2056 3 7 0 3 8 0 3 9 4096 3 fft_sp_1d_c2c_ecpy size= 65536max_diff = 0.164063 N = 65536 Cycle: 5912488
FFT memory buffers: Buffer Size(bytes) Alignment 0 524288 3 1 524288 3 2 8 3 3 0 3 4 0 3 5 0 3 6 0 3 7 0 3 8 0 3 9 0 3 fft_sp_1d_c2c_direct size= 65536max_diff = 32.437500 N = 65536 Cycle: 741153
为什么同样size的数据
DSPF_sp_fftSPxSP函数执行时间比
fft_sp_1d_r2c还要少???
user5329195:
回复 Shine:
下面是我的代码