RAPP Compute Tuning Report


Best generic

Best SWAR

Best SIMD
Linux gcc116 3.13.0-57-generic #95-Ubuntu SMP Fri Jun 19 09:31:00 UTC 2015 aarch64 aarch64 aarch64 GNU/Linux
Function
0%Relative Execution Speed100%
Gain
rc_bitblt_wa_copy_bin

1.07
rc_pixop_set_u8

1.04
rc_pixop_not_u8

5.90
rc_pixop_flip_u8

5.77
rc_pixop_abs_u8

2.12
rc_pixop_addc_u8

2.50
rc_pixop_lerpc_u8

5.86
rc_pixop_lerpnc_u8

5.22
rc_pixop_add_u8

2.55
rc_pixop_avg_u8

3.79
rc_pixop_sub_u8

1.78
rc_pixop_subh_u8

3.42
rc_pixop_suba_u8

3.23
rc_pixop_lerp_u8

2.46
rc_pixop_lerpn_u8

3.48
rc_pixop_lerpi_u8

2.81
rc_pixop_norm_u8

1.97
rc_type_u8_to_bin

3.37
rc_type_bin_to_u8

2.53
rc_thresh_gt_u8

2.24
rc_thresh_lt_u8

2.05
rc_thresh_gtlt_u8

2.44
rc_thresh_ltgt_u8

2.26
rc_thresh_gt_pixel_u8

2.21
rc_thresh_lt_pixel_u8

2.18
rc_thresh_gtlt_pixel_u8

2.32
rc_thresh_ltgt_pixel_u8

1.99
rc_stat_sum_bin

1.76
rc_stat_sum_u8

2.34
rc_stat_min_bin

1.01
rc_stat_max_bin

1.02
rc_stat_min_u8

2.10
rc_stat_max_u8

2.09
rc_reduce_1x2_u8

1.45
rc_reduce_2x1_u8

4.02
rc_reduce_2x2_u8

1.60
rc_filter_diff_1x2_horz_u8

2.63
rc_filter_diff_1x2_horz_abs_u8

2.56
rc_filter_diff_2x1_vert_u8

3.58
rc_filter_diff_2x1_vert_abs_u8

3.04
rc_filter_diff_2x2_magn_u8

3.13
rc_filter_sobel_3x3_horz_u8

2.51
rc_filter_sobel_3x3_horz_abs_u8

3.24
rc_filter_sobel_3x3_vert_u8

2.06
rc_filter_sobel_3x3_vert_abs_u8

2.71
rc_filter_sobel_3x3_magn_u8

2.35
rc_filter_gauss_3x3_u8

2.69
rc_filter_laplace_3x3_u8

2.40
rc_filter_laplace_3x3_abs_u8

2.68
rc_filter_highpass_3x3_u8

1.81
rc_filter_highpass_3x3_abs_u8

1.17
rc_margin_horz_bin

1.00
rc_cond_set_u8

3.17
rc_cond_addc_u8

1.63
rc_cond_subc_u8

1.49
rc_cond_copy_u8

2.93
rc_cond_add_u8

2.25