https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81673
--- Comment #3 from Martin Jambor <jamborm at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #1)
> What happens if you use -march=intel.
With -mtune=intel, the lower half of the vector is moved directly
whereas the upper one is still done through the stack:
.cfi_startproc
leaq -56(%rsp), %rsp
.cfi_def_cfa_offset 64
movq %rdx, %xmm0
movq %rcx, (%rsp)
leaq 16(%rsp), %rdi
movq %r9, 8(%rsp)
movhps (%rsp), %xmm0
movdqa %xmm0, 32(%rsp)
movq %r8, %xmm0
movhps 8(%rsp), %xmm0
movdqa %xmm0, 16(%rsp)
call bar
leaq 56(%rsp), %rsp
.cfi_def_cfa_offset 8
ret
.cfi_endproc
...so I guess this would still incur some penalty on the benchmark,
but I am not sure.
