Reverts llvm/llvm-project#108939
When `AVX` is available but `-mprefer-vector-width=128` some of the
`mov` instructions turn into the x86 `rep;movsb` instruction leading to
poor performance on "old" architectures (sandybridge, haswell). The
possible solutions are : get rid of the `-mprefer-vector-width` option
or use smaller static copy sizes in
`inline_memcpy_x86_sse2_ge64_sw_prefetching`. Right now a copy size of 3
cache lines (192B) relying exclusively on xmm registers gets turned into
`rep;movsb`.
dda107b8cb