[x265] [arm64] port satd and sa8d

[Pop, Sebastian]

Hi,
the attached patch ports to arm64 the following kernels:

               satd[  4x4]  8.52x    15.86           135.07
               satd[  8x8]  3.73x    20.82           77.73
        [i420] satd[  4x4]  8.59x    15.72           135.02
        [i422] satd[  4x8]  2.39x    18.66           44.55
               satd[  8x4]  6.22x    31.69           197.09
        [i422] satd[  4x4]  8.58x    15.72           134.98
               satd[  4x8]  2.55x    17.44           44.50
               satd[16x16]  4.10x    67.97           278.72
        [i420] satd[  8x8]  3.66x    21.16           77.56
        [i422] satd[ 8x16]  3.96x    37.50           148.56
               satd[ 16x8]  3.97x    35.74           141.92
        [i420] satd[  8x4]  6.24x    31.66           197.40
        [i422] satd[  8x8]  3.64x    21.33           77.57
               satd[ 8x16]  3.88x    38.29           148.69
        [i420] satd[  4x8]  2.57x    17.49           45.01
        [i422] satd[ 4x16]  2.37x    34.97           82.73
               satd[ 16x4]  3.71x    19.99           74.16
        [i422] satd[  8x4]  6.22x    31.64           196.89
        [i422] satd[ 8x12]  1.25x    95.51           119.75
               satd[ 4x16]  2.39x    34.70           83.07
               satd[12x16]  2.30x    99.35           228.32
               satd[32x32]  4.19x    260.66          1093.26
        [i420] satd[16x16]  4.15x    67.12           278.49
        [i422] satd[16x32]  4.25x    131.44          559.04
               satd[32x16]  4.12x    132.00          543.40
        [i420] satd[ 16x8]  4.04x    35.50           143.52
        [i422] satd[16x16]  4.15x    67.11           278.60
               satd[16x32]  4.26x    131.44          559.93
        [i420] satd[ 8x16]  4.02x    36.97           148.73
        [i422] satd[ 8x32]  4.09x    70.71           289.36
               satd[ 32x8]  4.00x    68.54           273.82
        [i420] satd[ 16x4]  3.92x    19.02           74.65
        [i422] satd[ 16x8]  4.04x    35.49           143.40
               satd[32x24]  4.17x    196.93          822.04
        [i420] satd[16x12]  4.13x    50.94           210.62
        [i422] satd[16x24]  4.21x    98.89           416.66
               satd[ 8x32]  4.10x    70.39           288.67
        [i420] satd[ 4x16]  2.35x    35.05           82.48
        [i422] satd[ 4x32]  2.40x    66.54           159.82
               satd[24x32]  3.91x    211.47          826.42
        [i420] satd[12x16]  2.29x    99.65           227.78
        [i422] satd[12x32]  2.36x    196.13          462.74
               satd[64x64]  4.12x    1038.41         4275.62
        [i420] satd[32x32]  4.17x    261.28          1089.51
        [i422] satd[32x64]  4.17x    518.04          2161.90
               satd[64x32]  4.13x    520.88          2149.74
        [i420] satd[32x16]  4.10x    132.40          543.38
        [i422] satd[32x32]  4.20x    260.24          1092.69
               satd[32x64]  4.17x    518.06          2162.26
        [i420] satd[16x32]  4.27x    131.34          560.19
        [i422] satd[16x64]  4.28x    257.35          1102.24
               satd[64x16]  4.14x    262.07          1084.33
        [i420] satd[ 32x8]  4.05x    68.09           275.70
        [i422] satd[32x16]  4.11x    132.29          543.94
               satd[64x48]  4.13x    777.01          3206.25
        [i420] satd[32x24]  4.14x    198.18          821.15
        [i422] satd[32x48]  4.16x    390.86          1626.74
               satd[16x64]  4.30x    256.50          1103.04
        [i420] satd[ 8x32]  4.14x    69.95           289.74
        [i422] satd[ 8x64]  4.16x    140.01          582.32
               satd[48x64]  4.16x    772.92          3216.96
        [i420] satd[24x32]  3.89x    212.13          824.99
        [i422] satd[24x64]  3.89x    420.72          1636.46
                 sa8d[4x4]  8.22x    16.08           132.18
                 sa8d[8x8]  18.18x   27.91           507.32
               sa8d[16x16]  20.07x   101.90          2045.08
               sa8d[32x32]  20.95x   393.03          8233.36
               sa8d[64x64]  20.79x   1564.42         32520.55

Ok to commit?

Thanks,
Sebastian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20210805/21968193/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-arm64-port-satd-and-sa8d.patch
Type: application/octet-stream
Size: 43101 bytes
Desc: 0001-arm64-port-satd-and-sa8d.patch
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20210805/21968193/attachment-0001.obj>