RUN_VOLK_TESTS: volk_64u_popcntpuppet_64u(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 1285.69 ms | 3241.2 MB/s | - | - | rvv | 934.97 ms | 4457.0 MB/s | 1.38x | 0.0e+00 | * Best aligned arch | rvv (1.38x) Best unaligned arch | rvv (1.38x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16u_byteswappuppet_16u(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 68.91 ms | 15118.8 MB/s | - | - | rvv | 59.66 ms | 17463.1 MB/s | 1.16x | 0.0e+00 | * Best aligned arch | rvv (1.16x) Best unaligned arch | rvv (1.16x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32u_byteswappuppet_32u(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 375.12 ms | 5554.5 MB/s | - | - | rvv | 231.43 ms | 9003.3 MB/s | 1.62x | 0.0e+00 | * Best aligned arch | rvv (1.62x) Best unaligned arch | rvv (1.62x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32u_popcntpuppet_32u(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 512.02 ms | 4069.3 MB/s | - | - | * rvv | 898.97 ms | 2317.8 MB/s | 0.57x | 0.0e+00 | Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_64u_byteswappuppet_64u(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 1203.50 ms | 3462.5 MB/s | - | - | rvv | 751.81 ms | 5542.8 MB/s | 1.60x | 0.0e+00 | * Best aligned arch | rvv (1.60x) Best unaligned arch | rvv (1.60x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_s32fc_rotator2puppet_32fc(vlen=131071, iter=1987, tol=1e-03) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 2114.33 ms | 1970.9 MB/s | - | - | rvv | 482.31 ms | 8640.0 MB/s | 4.38x | 9.8e-05 | rvvseg | 479.19 ms | 8696.3 MB/s | 4.41x | 9.8e-05 | * Best aligned arch | rvvseg (4.41x) Best unaligned arch | rvvseg (4.41x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8u_conv_k7_r2puppet_8u(vlen=131071, iter=1987, tol=0e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 40690.85 ms | 12.8 MB/s | - | - | rvv | 4187.28 ms | 124.4 MB/s | 9.72x | 0.0e+00 | * Best aligned arch | rvv (9.72x) Best unaligned arch | rvv (9.72x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_fm_detectpuppet_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 268.66 ms | 11633.0 MB/s | - | - | rvv | 155.79 ms | 20062.0 MB/s | 1.72x | 0.0e+00 | * Best aligned arch | rvv (1.72x) Best unaligned arch | rvv (1.72x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16ic_s32f_deinterleave_real_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 126.85 ms | 16425.1 MB/s | - | - | rvv | 87.47 ms | 23820.1 MB/s | 1.45x | 0.0e+00 | * Best aligned arch | rvv (1.45x) Best unaligned arch | rvv (1.45x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16ic_deinterleave_real_8i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 64.26 ms | 20266.1 MB/s | - | - | rvv | 50.13 ms | 25978.6 MB/s | 1.28x | 0.0e+00 | * Best aligned arch | rvv (1.28x) Best unaligned arch | rvv (1.28x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16ic_deinterleave_16i_x2(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 92.66 ms | 22486.9 MB/s | - | - | rvv | 83.81 ms | 24860.1 MB/s | 1.11x | 0.0e+00 | * rvvseg | 124.17 ms | 16779.7 MB/s | 0.75x | 0.0e+00 | Best aligned arch | rvv (1.11x) Best unaligned arch | rvv (1.11x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16ic_s32f_deinterleave_32f_x2(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 1632.53 ms | 1914.4 MB/s | - | - | rvv | 141.62 ms | 22068.9 MB/s | 11.53x | 1.0e-07 | * rvvseg | 146.48 ms | 21337.1 MB/s | 11.15x | 1.0e-07 | Best aligned arch | rvv (11.53x) Best unaligned arch | rvv (11.53x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16ic_deinterleave_real_16i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 61.89 ms | 25249.4 MB/s | - | - | rvv | 61.60 ms | 25370.3 MB/s | 1.00x | 0.0e+00 | * Best aligned arch | rvv (1.00x) Best unaligned arch | rvv (1.00x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16ic_magnitude_16i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 2432.40 ms | 642.4 MB/s | - | - | rvv | 804.71 ms | 1941.9 MB/s | 3.02x | 0.0e+00 | rvvseg | 800.74 ms | 1951.5 MB/s | 3.04x | 0.0e+00 | * Best aligned arch | rvvseg (3.04x) Best unaligned arch | rvvseg (3.04x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16ic_s32f_magnitude_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 808.36 ms | 2577.6 MB/s | - | - | rvv | 784.30 ms | 2656.6 MB/s | 1.03x | 0.0e+00 | * rvvseg | 808.69 ms | 2576.5 MB/s | 1.00x | 0.0e+00 | Best aligned arch | rvv (1.03x) Best unaligned arch | rvv (1.03x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16ic_convert_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 175.31 ms | 17827.9 MB/s | - | - | rvv | 127.40 ms | 24532.4 MB/s | 1.38x | 0.0e+00 | * Best aligned arch | rvv (1.38x) Best unaligned arch | rvv (1.38x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16ic_x2_multiply_16ic(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 171.20 ms | 18255.9 MB/s | - | - | * rvv | 179.76 ms | 17386.1 MB/s | 0.95x | 0.0e+00 | rvvseg | 311.08 ms | 10046.9 MB/s | 0.55x | 0.0e+00 | Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16ic_x2_dot_prod_16ic(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 2511.98 ms | 1244.2 MB/s | - | - | rvv | 166.19 ms | 18805.8 MB/s | 15.11x | 0.0e+00 | * rvvseg | 273.97 ms | 11407.8 MB/s | 9.17x | 0.0e+00 | Best aligned arch | rvv (15.11x) Best unaligned arch | rvv (15.11x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16i_s32f_convert_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 816.61 ms | 1913.6 MB/s | - | - | rvv | 54.56 ms | 28641.1 MB/s | 14.97x | 1.0e-07 | * Best aligned arch | rvv (14.97x) Best unaligned arch | rvv (14.97x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16i_convert_8i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 16.70 ms | 46789.6 MB/s | - | - | * rvv | 20.19 ms | 38700.5 MB/s | 0.83x | 0.0e+00 | Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16i_32fc_dot_prod_32fc(vlen=131071, iter=1987, tol=1e-01) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 661.75 ms | 7084.3 MB/s | - | - | rvv | 196.96 ms | 23802.5 MB/s | 3.36x | 5.2e-03 | * rvvseg | 521.56 ms | 8988.5 MB/s | 1.27x | 5.2e-03 | Best aligned arch | rvv (3.36x) Best unaligned arch | rvv (3.36x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_accumulator_s32f(vlen=131071, iter=1987, tol=2e-02) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 326.48 ms | 6382.1 MB/s | - | - | rvv | 44.98 ms | 46327.1 MB/s | 7.26x | 1.5e-03 | * Best aligned arch | rvv (7.26x) Best unaligned arch | rvv (7.26x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_add_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 157.43 ms | 19852.2 MB/s | - | - | rvv | 153.35 ms | 20380.6 MB/s | 1.03x | 0.0e+00 | * Best aligned arch | rvv (1.03x) Best unaligned arch | rvv (1.03x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_index_max_16u(vlen=131071, iter=1987, tol=0e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 261.53 ms | 5975.2 MB/s | - | - | rvv | 32.67 ms | 47827.4 MB/s | 8.00x | 0.0e+00 | * Best aligned arch | rvv (8.00x) Best unaligned arch | rvv (8.00x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_index_max_32u(vlen=131071, iter=1987, tol=0e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 528.00 ms | 3946.2 MB/s | - | - | rvv | 96.36 ms | 21622.2 MB/s | 5.48x | 0.0e+00 | * Best aligned arch | rvv (5.48x) Best unaligned arch | rvv (5.48x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_index_min_16u(vlen=131071, iter=1987, tol=0e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 260.97 ms | 5988.0 MB/s | - | - | rvv | 32.67 ms | 47834.5 MB/s | 7.99x | 0.0e+00 | * Best aligned arch | rvv (7.99x) Best unaligned arch | rvv (7.99x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_index_min_32u(vlen=131071, iter=1987, tol=0e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 527.34 ms | 3951.1 MB/s | - | - | rvv | 99.41 ms | 20959.0 MB/s | 5.30x | 0.0e+00 | * Best aligned arch | rvv (5.30x) Best unaligned arch | rvv (5.30x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_32f_multiply_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 279.06 ms | 18665.9 MB/s | - | - | rvv | 223.89 ms | 23265.9 MB/s | 1.25x | 0.0e+00 | * Best aligned arch | rvv (1.25x) Best unaligned arch | rvv (1.25x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_32f_add_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 316.16 ms | 16475.6 MB/s | - | - | rvv | 226.49 ms | 22998.4 MB/s | 1.40x | 0.0e+00 | * Best aligned arch | rvv (1.40x) Best unaligned arch | rvv (1.40x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_log2_32f(vlen=131071, iter=1987, tol=5e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 4589.33 ms | 454.0 MB/s | - | - | rvv | 845.81 ms | 2463.4 MB/s | 5.43x | 4.2e-06 | * Best aligned arch | rvv (5.43x) Best unaligned arch | rvv (5.43x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_expfast_32f(vlen=131071, iter=1987, tol=1e-01) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 1885.50 ms | 1105.1 MB/s | - | - | rvv | 85.87 ms | 24265.7 MB/s | 21.96x | 5.6e-02 | * Best aligned arch | rvv (21.96x) Best unaligned arch | rvv (21.96x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_sin_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 3054.33 ms | 682.2 MB/s | - | - | polynomial | 879.77 ms | 2368.3 MB/s | 3.47x | 1.7e-07 | * rvv | 1476.06 ms | 1411.6 MB/s | 2.07x | 5.4e-07 | Best aligned arch | polynomial (3.47x) Best unaligned arch | polynomial (3.47x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_cos_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 2875.10 ms | 724.7 MB/s | - | - | polynomial | 903.95 ms | 2305.0 MB/s | 3.18x | 1.7e-07 | * rvv | 1449.12 ms | 1437.8 MB/s | 1.98x | 2.2e-07 | Best aligned arch | polynomial (3.18x) Best unaligned arch | polynomial (3.18x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_sincos_32f_x2(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 4405.76 ms | 709.4 MB/s | - | - | polynomial | 1103.89 ms | 2831.3 MB/s | 3.99x | 1.7e-07 | * Best aligned arch | polynomial (3.99x) Best unaligned arch | polynomial (3.99x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_tan_32f(vlen=131071, iter=1987, tol=1e-02) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 5980.76 ms | 348.4 MB/s | - | - | rvv | 1889.71 ms | 1102.6 MB/s | 3.16x | 6.2e-07 | * Best aligned arch | rvv (3.16x) Best unaligned arch | rvv (3.16x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_atan_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 5486.33 ms | 379.8 MB/s | - | - | polynomial | 1012.57 ms | 2057.7 MB/s | 5.42x | 7.7e-07 | rvv | 544.04 ms | 3829.8 MB/s | 10.08x | 7.7e-07 | * Best aligned arch | rvv (10.08x) Best unaligned arch | rvv (10.08x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_asin_32f(vlen=131071, iter=1987, tol=1e-05) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 2458.98 ms | 847.3 MB/s | - | - | * Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_acos_32f(vlen=131071, iter=1987, tol=1e-05) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 2625.09 ms | 793.7 MB/s | - | - | * Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_s32f_power_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 19858.58 ms | 209.8 MB/s | - | - | * Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_calc_spectral_noise_floor_32f(vlen=131071, iter=1987, tol=1e-04) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 1434.60 ms | 1452.4 MB/s | - | - | rvv | 115.05 ms | 18109.5 MB/s | 12.47x | 9.7e-07 | * Best aligned arch | rvv (12.47x) Best unaligned arch | rvv (12.47x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_s32f_atan2_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 11752.97 ms | 265.9 MB/s | - | - | polynomial | 2663.72 ms | 1173.3 MB/s | 4.41x | 8.3e-07 | rvv | 2018.05 ms | 1548.7 MB/s | 5.82x | 8.3e-07 | rvvseg | 2000.22 ms | 1562.5 MB/s | 5.88x | 8.3e-07 | * Best aligned arch | rvvseg (5.88x) Best unaligned arch | rvvseg (5.88x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_x2_conjugate_dot_prod_32fc(vlen=131071, iter=1987, tol=2e-02) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 1936.65 ms | 3227.6 MB/s | - | - | block | 795.30 ms | 7859.6 MB/s | 2.44x | 7.9e-04 | rvv | 435.83 ms | 14342.3 MB/s | 4.44x | 2.2e-04 | rvvseg | 374.85 ms | 16675.3 MB/s | 5.17x | 2.2e-04 | * Best aligned arch | rvvseg (5.17x) Best unaligned arch | rvvseg (5.17x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_deinterleave_32f_x2(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 226.59 ms | 18390.6 MB/s | - | - | rvv | 183.28 ms | 22736.2 MB/s | 1.24x | 0.0e+00 | * rvvseg | 255.88 ms | 16285.7 MB/s | 0.89x | 0.0e+00 | Best aligned arch | rvv (1.24x) Best unaligned arch | rvv (1.24x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_accumulator_s32fc(vlen=131071, iter=1987, tol=3e-02) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 653.62 ms | 6375.6 MB/s | - | - | rvv | 112.62 ms | 37003.2 MB/s | 5.80x | 1.0e-03 | * Best aligned arch | rvv (5.80x) Best unaligned arch | rvv (5.80x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_deinterleave_64f_x2(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 341.09 ms | 18325.9 MB/s | - | - | rvv | 290.96 ms | 21483.4 MB/s | 1.17x | 0.0e+00 | * rvvseg | 345.44 ms | 18095.0 MB/s | 0.99x | 0.0e+00 | Best aligned arch | rvv (1.17x) Best unaligned arch | rvv (1.17x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_s32f_deinterleave_real_16i(vlen=131071, iter=1987, tol=1e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 165.97 ms | 15692.6 MB/s | - | - | rvv | 150.23 ms | 17337.0 MB/s | 1.10x | 0.0e+00 | * Best aligned arch | rvv (1.10x) Best unaligned arch | rvv (1.10x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_deinterleave_imag_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 167.11 ms | 18702.9 MB/s | - | - | rvv | 146.33 ms | 21358.1 MB/s | 1.14x | 0.0e+00 | * Best aligned arch | rvv (1.14x) Best unaligned arch | rvv (1.14x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_deinterleave_real_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 145.49 ms | 21481.3 MB/s | - | - | * rvv | 146.29 ms | 21364.7 MB/s | 0.99x | 0.0e+00 | Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_deinterleave_real_64f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 230.22 ms | 18100.5 MB/s | - | - | rvv | 183.40 ms | 22721.2 MB/s | 1.26x | 0.0e+00 | * Best aligned arch | rvv (1.26x) Best unaligned arch | rvv (1.26x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_x2_dot_prod_32fc(vlen=131071, iter=1987, tol=2e-02) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 782.08 ms | 7992.5 MB/s | - | - | sifive_u74 | 948.05 ms | 6593.3 MB/s | 0.82x | 5.2e-04 | rvv | 408.98 ms | 15283.7 MB/s | 1.91x | 1.2e-04 | * rvvseg | 555.33 ms | 11255.9 MB/s | 1.41x | 4.7e-04 | Best aligned arch | rvv (1.91x) Best unaligned arch | rvv (1.91x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_32f_dot_prod_32fc(vlen=131071, iter=1987, tol=1e-02) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 654.02 ms | 7964.6 MB/s | - | - | rvv | 207.62 ms | 25088.7 MB/s | 3.15x | 1.3e-03 | * rvvseg | 707.75 ms | 7359.9 MB/s | 0.92x | 1.3e-03 | Best aligned arch | rvv (3.15x) Best unaligned arch | rvv (3.15x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_index_max_16u(vlen=131071, iter=1987, tol=0e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 309.26 ms | 8421.6 MB/s | - | - | rvv | 87.17 ms | 29877.6 MB/s | 3.55x | 0.0e+00 | rvvseg | 75.23 ms | 34622.5 MB/s | 4.11x | 0.0e+00 | * Best aligned arch | rvvseg (4.11x) Best unaligned arch | rvvseg (4.11x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_index_max_32u(vlen=131071, iter=1987, tol=0e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 882.54 ms | 3541.3 MB/s | - | - | rvv | 219.32 ms | 14250.5 MB/s | 4.02x | 0.0e+00 | * rvvseg | 243.49 ms | 12835.8 MB/s | 3.62x | 0.0e+00 | Best aligned arch | rvv (4.02x) Best unaligned arch | rvv (4.02x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_index_min_16u(vlen=131071, iter=1987, tol=0e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 299.45 ms | 8697.6 MB/s | - | - | rvv | 81.20 ms | 32074.5 MB/s | 3.69x | 0.0e+00 | rvvseg | 72.31 ms | 36018.9 MB/s | 4.14x | 0.0e+00 | * Best aligned arch | rvvseg (4.14x) Best unaligned arch | rvvseg (4.14x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_index_min_32u(vlen=131071, iter=1987, tol=0e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 891.92 ms | 3504.1 MB/s | - | - | rvv | 221.23 ms | 14127.1 MB/s | 4.03x | 0.0e+00 | * rvvseg | 242.16 ms | 12906.3 MB/s | 3.68x | 0.0e+00 | Best aligned arch | rvv (4.03x) Best unaligned arch | rvv (4.03x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_s32f_magnitude_16i(vlen=131071, iter=1987, tol=1e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 800.54 ms | 3253.4 MB/s | - | - | rvv | 800.94 ms | 3251.8 MB/s | 1.00x | 0.0e+00 | rvvseg | 792.82 ms | 3285.1 MB/s | 1.01x | 0.0e+00 | * Best aligned arch | rvvseg (1.01x) Best unaligned arch | rvvseg (1.01x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_magnitude_32f(vlen=131071, iter=1987, tol=1e-01) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 784.19 ms | 3985.5 MB/s | - | - | * rvv | 784.38 ms | 3984.5 MB/s | 1.00x | 0.0e+00 | rvvseg | 784.27 ms | 3985.1 MB/s | 1.00x | 0.0e+00 | Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_magnitude_squared_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 173.74 ms | 17988.4 MB/s | - | - | rvv | 167.15 ms | 18697.7 MB/s | 1.04x | 0.0e+00 | * rvvseg | 255.19 ms | 12247.1 MB/s | 0.68x | 0.0e+00 | Best aligned arch | rvv (1.04x) Best unaligned arch | rvv (1.04x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_x2_add_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 383.85 ms | 16284.4 MB/s | - | - | rvv | 311.07 ms | 20094.3 MB/s | 1.23x | 0.0e+00 | * Best aligned arch | rvv (1.23x) Best unaligned arch | rvv (1.23x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_x2_multiply_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 1781.30 ms | 3509.1 MB/s | - | - | rvv | 361.35 ms | 17298.5 MB/s | 4.93x | 1.2e-07 | * rvvseg | 628.83 ms | 9940.2 MB/s | 2.83x | 1.2e-07 | Best aligned arch | rvv (4.93x) Best unaligned arch | rvv (4.93x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_x2_multiply_conjugate_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 1703.51 ms | 3669.3 MB/s | - | - | rvv | 360.16 ms | 17355.3 MB/s | 4.73x | 1.2e-07 | * rvvseg | 627.35 ms | 9963.7 MB/s | 2.72x | 1.2e-07 | Best aligned arch | rvv (4.73x) Best unaligned arch | rvv (4.73x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_x2_divide_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 2982.27 ms | 2096.0 MB/s | - | - | rvv | 921.34 ms | 6784.4 MB/s | 3.24x | 2.3e-07 | rvvseg | 842.86 ms | 7416.2 MB/s | 3.54x | 2.3e-07 | * Best aligned arch | rvvseg (3.54x) Best unaligned arch | rvvseg (3.54x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_conjugate_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 188.35 ms | 22124.2 MB/s | - | - | rvv | 185.62 ms | 22450.0 MB/s | 1.01x | 0.0e+00 | * Best aligned arch | rvv (1.01x) Best unaligned arch | rvv (1.01x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_convert_16i(vlen=131071, iter=1987, tol=1e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 262.80 ms | 5946.2 MB/s | - | - | rvv | 61.96 ms | 25222.2 MB/s | 4.24x | 0.0e+00 | * Best aligned arch | rvv (4.24x) Best unaligned arch | rvv (4.24x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_convert_32i(vlen=131071, iter=1987, tol=1e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 300.01 ms | 6945.1 MB/s | - | - | rvv | 86.17 ms | 24181.2 MB/s | 3.48x | 0.0e+00 | * Best aligned arch | rvv (3.48x) Best unaligned arch | rvv (3.48x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_convert_64f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 165.19 ms | 18919.8 MB/s | - | - | rvv | 127.58 ms | 24497.2 MB/s | 1.29x | 0.0e+00 | * Best aligned arch | rvv (1.29x) Best unaligned arch | rvv (1.29x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_convert_8i(vlen=131071, iter=1987, tol=1e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 389.84 ms | 3340.4 MB/s | - | - | rvv | 61.56 ms | 21153.3 MB/s | 6.33x | 0.0e+00 | * Best aligned arch | rvv (6.33x) Best unaligned arch | rvv (6.33x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_convert_16ic(vlen=131071, iter=1987, tol=1e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 565.71 ms | 5524.7 MB/s | - | - | rvv | 146.30 ms | 21362.6 MB/s | 3.87x | 0.0e+00 | * Best aligned arch | rvv (3.87x) Best unaligned arch | rvv (3.87x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_s32f_power_spectrum_32f(vlen=131071, iter=1987, tol=2e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 994.06 ms | 3144.1 MB/s | - | - | rvv | 748.17 ms | 4177.4 MB/s | 1.33x | 7.9e-07 | rvvseg | 728.38 ms | 4290.8 MB/s | 1.36x | 7.9e-07 | * Best aligned arch | rvvseg (1.36x) Best unaligned arch | rvvseg (1.36x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_x2_square_dist_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 230.47 ms | 22601.3 MB/s | - | - | rvv | 199.36 ms | 26128.2 MB/s | 1.16x | 0.0e+00 | * rvvseg | 243.93 ms | 21354.2 MB/s | 0.94x | 0.0e+00 | Best aligned arch | rvv (1.16x) Best unaligned arch | rvv (1.16x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_x2_s32f_square_dist_scalar_mult_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 234.12 ms | 22248.8 MB/s | - | - | rvv | 218.05 ms | 23888.7 MB/s | 1.07x | 0.0e+00 | * rvvseg | 243.18 ms | 21420.3 MB/s | 0.96x | 0.0e+00 | Best aligned arch | rvv (1.07x) Best unaligned arch | rvv (1.07x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_divide_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 816.52 ms | 3827.7 MB/s | - | - | * rvv | 855.26 ms | 3654.3 MB/s | 0.95x | 0.0e+00 | Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_dot_prod_32f(vlen=131071, iter=1987, tol=1e-02) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 343.67 ms | 9094.1 MB/s | - | - | rvv | 141.55 ms | 22080.3 MB/s | 2.43x | 3.2e-03 | * Best aligned arch | rvv (2.43x) Best unaligned arch | rvv (2.43x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_s32f_interleave_16ic(vlen=131071, iter=1987, tol=1e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 190.95 ms | 16367.1 MB/s | - | - | rvv | 167.89 ms | 18615.3 MB/s | 1.14x | 0.0e+00 | rvvseg | 153.03 ms | 20423.5 MB/s | 1.25x | 0.0e+00 | * Best aligned arch | rvvseg (1.25x) Best unaligned arch | rvvseg (1.25x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_interleave_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 181.07 ms | 23013.6 MB/s | - | - | rvv | 182.07 ms | 22888.3 MB/s | 0.99x | 0.0e+00 | rvvseg | 177.23 ms | 23512.3 MB/s | 1.02x | 0.0e+00 | * Best aligned arch | rvvseg (1.02x) Best unaligned arch | rvvseg (1.02x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_max_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 171.64 ms | 18208.8 MB/s | - | - | rvv | 153.03 ms | 20422.9 MB/s | 1.12x | 0.0e+00 | * Best aligned arch | rvv (1.12x) Best unaligned arch | rvv (1.12x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_min_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 171.96 ms | 18174.8 MB/s | - | - | rvv | 152.84 ms | 20448.5 MB/s | 1.13x | 0.0e+00 | * Best aligned arch | rvv (1.13x) Best unaligned arch | rvv (1.13x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_multiply_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 158.12 ms | 19765.4 MB/s | - | - | rvv | 154.00 ms | 20295.0 MB/s | 1.03x | 0.0e+00 | * Best aligned arch | rvv (1.03x) Best unaligned arch | rvv (1.03x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_64f_multiply_64f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 292.49 ms | 17809.0 MB/s | - | - | rvv | 232.04 ms | 22448.9 MB/s | 1.26x | 0.0e+00 | * Best aligned arch | rvv (1.26x) Best unaligned arch | rvv (1.26x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_64f_add_64f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 268.01 ms | 19436.0 MB/s | - | - | rvv | 231.73 ms | 22478.2 MB/s | 1.16x | 0.0e+00 | * Best aligned arch | rvv (1.16x) Best unaligned arch | rvv (1.16x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_normalize(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 63.61 ms | 16376.5 MB/s | - | - | rvv | 60.13 ms | 17324.2 MB/s | 1.06x | 0.0e+00 | * Best aligned arch | rvv (1.06x) Best unaligned arch | rvv (1.06x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_power_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 7354.45 ms | 283.3 MB/s | - | - | * Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_reciprocal_32f(vlen=131071, iter=1987, tol=6e-05) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 816.94 ms | 2550.5 MB/s | - | - | * rvv | 817.38 ms | 2549.1 MB/s | 1.00x | 0.0e+00 | Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_sqrt_32f(vlen=131071, iter=1987, tol=1e-02) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 3872.83 ms | 538.0 MB/s | - | - | rvv | 759.36 ms | 2743.9 MB/s | 5.10x | 0.0e+00 | * Best aligned arch | rvv (5.10x) Best unaligned arch | rvv (5.10x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_invsqrt_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 4345.56 ms | 479.5 MB/s | - | - | recip_sqrt | 4146.62 ms | 502.5 MB/s | 1.05x | 1.3e-07 | Q_rsqrt | 779.61 ms | 2672.6 MB/s | 5.57x | 1.7e-07 | rvv | 341.60 ms | 6099.5 MB/s | 12.72x | 1.7e-07 | * Best aligned arch | rvv (12.72x) Best unaligned arch | rvv (12.72x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_stddev_32f(vlen=131071, iter=1987, tol=1e-02) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 343.43 ms | 6066.9 MB/s | - | - | rvv | 47.11 ms | 44227.5 MB/s | 7.29x | 0.0e+00 | * Best aligned arch | rvv (7.29x) Best unaligned arch | rvv (7.29x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_stddev_and_mean_32f_x2(vlen=131071, iter=1987, tol=1e-05) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 569.16 ms | 5491.2 MB/s | - | - | rvv | 110.44 ms | 28298.6 MB/s | 5.15x | 2.7e-06 | * Best aligned arch | rvv (5.15x) Best unaligned arch | rvv (5.15x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_subtract_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 159.16 ms | 19636.7 MB/s | - | - | rvv | 156.24 ms | 20003.2 MB/s | 1.02x | 0.0e+00 | * Best aligned arch | rvv (1.02x) Best unaligned arch | rvv (1.02x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x3_sum_of_poly_32f(vlen=131071, iter=1987, tol=1e+03) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 566.55 ms | 7355.3 MB/s | - | - | rvv | 230.12 ms | 18108.9 MB/s | 2.46x | 1.3e+00 | * Best aligned arch | rvv (2.46x) Best unaligned arch | rvv (2.46x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32i_x2_and_32i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 159.98 ms | 19536.3 MB/s | - | - | rvv | 155.33 ms | 20121.1 MB/s | 1.03x | 0.0e+00 | * Best aligned arch | rvv (1.03x) Best unaligned arch | rvv (1.03x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32i_s32f_convert_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 110.05 ms | 18933.6 MB/s | - | - | rvv | 87.64 ms | 23774.9 MB/s | 1.26x | 0.0e+00 | * Best aligned arch | rvv (1.26x) Best unaligned arch | rvv (1.26x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32i_x2_or_32i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 159.83 ms | 19554.7 MB/s | - | - | rvv | 155.21 ms | 20136.9 MB/s | 1.03x | 0.0e+00 | * Best aligned arch | rvv (1.03x) Best unaligned arch | rvv (1.03x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_dot_prod_16i(vlen=131071, iter=1987, tol=1e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 344.38 ms | 7562.9 MB/s | - | - | rvv | 139.05 ms | 18730.5 MB/s | 2.48x | 0.0e+00 | * Best aligned arch | rvv (2.48x) Best unaligned arch | rvv (2.48x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_64f_convert_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 188.83 ms | 16551.5 MB/s | - | - | rvv | 146.88 ms | 21278.3 MB/s | 1.29x | 0.0e+00 | * Best aligned arch | rvv (1.29x) Best unaligned arch | rvv (1.29x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_64f_x2_max_64f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 342.00 ms | 18277.1 MB/s | - | - | rvv | 314.77 ms | 19858.2 MB/s | 1.09x | 0.0e+00 | * Best aligned arch | rvv (1.09x) Best unaligned arch | rvv (1.09x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_64f_x2_min_64f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 343.15 ms | 18216.1 MB/s | - | - | rvv | 316.41 ms | 19755.5 MB/s | 1.08x | 0.0e+00 | * Best aligned arch | rvv (1.08x) Best unaligned arch | rvv (1.08x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_64f_x2_multiply_64f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 342.00 ms | 18277.3 MB/s | - | - | rvv | 339.55 ms | 18408.9 MB/s | 1.01x | 0.0e+00 | * Best aligned arch | rvv (1.01x) Best unaligned arch | rvv (1.01x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_64f_x2_add_64f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 343.11 ms | 18218.0 MB/s | - | - | rvv | 340.31 ms | 18367.6 MB/s | 1.01x | 0.0e+00 | * Best aligned arch | rvv (1.01x) Best unaligned arch | rvv (1.01x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_64f_x2_dot_prod_64f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 400.84 ms | 15594.0 MB/s | - | - | rvv | 307.53 ms | 20325.6 MB/s | 1.30x | 2.7e-13 | * Best aligned arch | rvv (1.30x) Best unaligned arch | rvv (1.30x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8ic_deinterleave_16i_x2(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 74.38 ms | 21008.7 MB/s | - | - | rvv | 60.03 ms | 26032.2 MB/s | 1.24x | 0.0e+00 | * Best aligned arch | rvv (1.24x) Best unaligned arch | rvv (1.24x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8ic_s32f_deinterleave_32f_x2(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 200.60 ms | 12983.3 MB/s | - | - | rvv | 136.51 ms | 19079.3 MB/s | 1.47x | 0.0e+00 | * Best aligned arch | rvv (1.47x) Best unaligned arch | rvv (1.47x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8ic_deinterleave_real_16i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 35.52 ms | 29327.4 MB/s | - | - | rvv | 30.01 ms | 34715.5 MB/s | 1.18x | 0.0e+00 | * Best aligned arch | rvv (1.18x) Best unaligned arch | rvv (1.18x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8ic_s32f_deinterleave_real_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 121.07 ms | 12907.5 MB/s | - | - | rvv | 68.94 ms | 22667.1 MB/s | 1.76x | 0.0e+00 | * Best aligned arch | rvv (1.76x) Best unaligned arch | rvv (1.76x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8ic_deinterleave_real_8i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 17.51 ms | 44628.6 MB/s | - | - | * rvv | 19.38 ms | 40307.0 MB/s | 0.90x | 0.0e+00 | Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8ic_x2_multiply_conjugate_16ic(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 1946.92 ms | 1070.2 MB/s | - | - | rvv | 161.84 ms | 12874.1 MB/s | 12.03x | 0.0e+00 | rvvseg | 109.15 ms | 19089.8 MB/s | 17.84x | 0.0e+00 | * Best aligned arch | rvvseg (17.84x) Best unaligned arch | rvvseg (17.84x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8ic_x2_s32f_multiply_conjugate_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 1769.59 ms | 1766.2 MB/s | - | - | rvv | 323.89 ms | 9649.4 MB/s | 5.46x | 0.0e+00 | rvvseg | 229.37 ms | 13625.8 MB/s | 7.71x | 0.0e+00 | * Best aligned arch | rvvseg (7.71x) Best unaligned arch | rvvseg (7.71x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8i_convert_16i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 21.12 ms | 36998.2 MB/s | - | - | rvv | 19.62 ms | 39823.4 MB/s | 1.08x | 0.0e+00 | * Best aligned arch | rvv (1.08x) Best unaligned arch | rvv (1.08x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8i_s32f_convert_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 49.66 ms | 26223.0 MB/s | - | - | * rvv | 55.33 ms | 23534.5 MB/s | 0.90x | 0.0e+00 | Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8i_x2_add_saturated_8i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 44.83 ms | 17429.6 MB/s | - | - | rvv | 19.27 ms | 40542.8 MB/s | 2.33x | 0.0e+00 | * Best aligned arch | rvv (2.33x) Best unaligned arch | rvv (2.33x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8u_x2_add_saturated_8u(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 19.75 ms | 39569.7 MB/s | - | - | rvv | 19.32 ms | 40446.4 MB/s | 1.02x | 0.0e+00 | * Best aligned arch | rvv (1.02x) Best unaligned arch | rvv (1.02x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16i_x2_add_saturated_16i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 179.01 ms | 8729.4 MB/s | - | - | rvv | 72.19 ms | 21645.7 MB/s | 2.48x | 0.0e+00 | * Best aligned arch | rvv (2.48x) Best unaligned arch | rvv (2.48x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_16u_x2_add_saturated_16u(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 79.59 ms | 19634.3 MB/s | - | - | rvv | 72.47 ms | 21563.4 MB/s | 1.10x | 0.0e+00 | * Best aligned arch | rvv (1.10x) Best unaligned arch | rvv (1.10x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_s32fc_multiply2_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 1203.39 ms | 3462.9 MB/s | - | - | * Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_multiply_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 110.24 ms | 18900.9 MB/s | - | - | sifive_u74 | 500.81 ms | 4160.4 MB/s | 0.22x | 0.0e+00 | rvv | 97.52 ms | 21365.0 MB/s | 1.13x | 0.0e+00 | * Best aligned arch | rvv (1.13x) Best unaligned arch | rvv (1.13x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_add_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 109.39 ms | 19047.5 MB/s | - | - | rvv | 97.85 ms | 21292.9 MB/s | 1.12x | 0.0e+00 | * Best aligned arch | rvv (1.12x) Best unaligned arch | rvv (1.12x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_binary_slicer_32i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 147.80 ms | 14097.4 MB/s | - | - | generic_branchless | 146.18 ms | 14253.5 MB/s | 1.01x | 0.0e+00 | rvv | 94.24 ms | 22110.2 MB/s | 1.57x | 0.0e+00 | * Best aligned arch | rvv (1.57x) Best unaligned arch | rvv (1.57x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_binary_slicer_8i(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 113.93 ms | 11430.5 MB/s | - | - | generic_branchless | 120.75 ms | 10784.9 MB/s | 0.94x | 0.0e+00 | rvv | 57.59 ms | 22613.3 MB/s | 1.98x | 0.0e+00 | * Best aligned arch | rvv (1.98x) Best unaligned arch | rvv (1.98x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32u_reverse_32u(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 778.17 ms | 2677.6 MB/s | - | - | byte_shuffle | 1105.08 ms | 1885.5 MB/s | 0.70x | 0.0e+00 | lut | 645.73 ms | 3226.7 MB/s | 1.21x | 0.0e+00 | 2001magic | 919.45 ms | 2266.1 MB/s | 0.85x | 0.0e+00 | 1972magic | 1223.43 ms | 1703.1 MB/s | 0.64x | 0.0e+00 | bintree_permute_top_down | 737.42 ms | 2825.5 MB/s | 1.06x | 0.0e+00 | bintree_permute_bottom_up | 775.93 ms | 2685.3 MB/s | 1.00x | 0.0e+00 | rvv | 285.85 ms | 7289.1 MB/s | 2.72x | 0.0e+00 | * Best aligned arch | rvv (2.72x) Best unaligned arch | rvv (2.72x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_tanh_32f(vlen=131071, iter=1987, tol=1e-02) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 7741.41 ms | 269.1 MB/s | - | - | series | 1639.67 ms | 1270.7 MB/s | 4.72x | 2.9e-07 | rvv | 819.17 ms | 2543.5 MB/s | 9.45x | 2.9e-07 | * Best aligned arch | rvv (9.45x) Best unaligned arch | rvv (9.45x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_x2_s32fc_multiply_conjugate_add2_32fc(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 2037.49 ms | 3067.9 MB/s | - | - | rvv | 548.42 ms | 11397.8 MB/s | 3.72x | 0.0e+00 | * rvvseg | 611.84 ms | 10216.3 MB/s | 3.33x | 0.0e+00 | Best aligned arch | rvv (3.72x) Best unaligned arch | rvv (3.72x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_exp_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 1967.93 ms | 1058.8 MB/s | - | - | rvv | 824.71 ms | 2526.4 MB/s | 2.39x | 1.2e-07 | * Best aligned arch | rvv (2.39x) Best unaligned arch | rvv (2.39x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_x2_powpuppet_32f(vlen=131071, iter=1987, tol=1e-02) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 5090.73 ms | 613.9 MB/s | - | - | rvv | 1953.16 ms | 1600.2 MB/s | 2.61x | 1.9e-03 | * Best aligned arch | rvv (2.61x) Best unaligned arch | rvv (2.61x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_mod_rangepuppet_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 1657.30 ms | 1257.2 MB/s | - | - | rvv | 828.63 ms | 2514.5 MB/s | 2.00x | 0.0e+00 | * Best aligned arch | rvv (2.00x) Best unaligned arch | rvv (2.00x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_8u_x3_encodepolarpuppet_8u(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 1733.87 ms | 600.8 MB/s | - | - | rvv | 1204.30 ms | 865.1 MB/s | 1.44x | 0.0e+00 | * rvvseg | 1292.69 ms | 805.9 MB/s | 1.34x | 0.0e+00 | Best aligned arch | rvv (1.44x) Best unaligned arch | rvv (1.44x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_8u_polarbutterflypuppet_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 61291.22 ms | 38.2 MB/s | - | - | rvv | 7948.01 ms | 294.9 MB/s | 7.71x | 0.0e+00 | * rvvseg | 8133.28 ms | 288.2 MB/s | 7.54x | 0.0e+00 | Best aligned arch | rvv (7.71x) Best unaligned arch | rvv (7.71x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32fc_s32f_power_spectral_densitypuppet_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 782.28 ms | 3995.2 MB/s | - | - | * Best aligned arch | generic Best unaligned arch | generic ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_clamppuppet_32f(vlen=131071, iter=1987, tol=1e-06) arch | time | throughput | speedup | max_rel | ---------------------------+----------------+----------------+----------+------------+ generic | 260.97 ms | 7984.1 MB/s | - | - | rvv | 92.33 ms | 22566.6 MB/s | 2.83x | 0.0e+00 | * Best aligned arch | rvv (2.83x) Best unaligned arch | rvv (2.83x) ---------------------------------------------------------------------------------------- RUN_VOLK_TESTS: volk_32f_s32f_convertpuppet_8u(vlen=131071, iter=1987, tol=1e+00) arch | time | throughput | speedup | max_abs | ---------------------------+----------------+----------------+----------+------------+ generic | 408.30 ms | 3189.4 MB/s | - | - | rvv | 65.22 ms | 19968.3 MB/s | 6.26x | 0.0e+00 | * Best aligned arch | rvv (6.26x) Best unaligned arch | rvv (6.26x) ---------------------------------------------------------------------------------------- Session summary (128 kernels): Average speedup vs generic: 3.29x Max speedup vs generic: 21.96x (volk_32f_expfast_32f) Creating "/home/evb2-test/.volk"... Writing /home/evb2-test/.volk/volk_config...