[x265] [arm64] port scale1D_128to64 and scale2D_64to32

Pop, Sebastian spop at amazon.com
Sat Jul 31 16:02:25 UTC 2021


> move SUB follow by LD1 will hidden memory operator latency,

Thanks, that helped a little bit:

Before:
        scale2D_64to32  86.83x  158.42        13756.12

After:
        scale2D_64to32  87.00x  158.20          13764.38

Added to the patch.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20210731/92e6b02c/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-arm64-port-scale1D_128to64-and-scale2D_64to32.patch
Type: application/octet-stream
Size: 3040 bytes
Desc: 0001-arm64-port-scale1D_128to64-and-scale2D_64to32.patch
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20210731/92e6b02c/attachment.obj>


More information about the x265-devel mailing list