[PATCH] aarch64:Modify the copy_long function in the SVE memcpy implementation for 32-byte aligned access

[Wilco Dijkstra]

Hi longwei,

Does Kunpeng support SVE? If not, it seems odd to modify the SVE memcpy...
Did you run the GLIBC benchtests to verify it shows a speedup on your system?
Especially bench-malloc-random and bench-malloc-large.

Btw is this related to https://sourceware.org/bugzilla/show_bug.cgi?id=27437#c10 ?
As suggested, I would first benchmark the existing memcpy/memset implementations
and select the one that works best for you.

Cheers,
Wilco