Use the AVX512 flavors of VMOVUPD to do a simple array copy.
Note: To execute this instruction CPU needs avx512 support (AVX512F instructions support).
If your cpu does not support avx512 , then consider using 256 bit flavor of vmovupd (requires AVX support) or 128 bit flavors (requires SSE4.2 support).
There are eight double values (512 bits) being copied. So when using the 256/128 bit flavors, remember to repeat the instruction pair appropriate number of times to process the entire 512 bits.
Source code:
https://github.com/vishmohan/inline_assembly_examples/tree/master/array_copy
