Re: [PATCH] x86: handle the tail in rep_movs_alternative() with an overlapping store

[Linus Torvalds]

On Tue, 25 Mar 2025 at 15:42, Herton Krzesinski <hkrzesin@xxxxxxxxxx> wrote:
>
> I have been trying to also measure the impact of changes like above, however,
> it seems I don't get improvement or it's limited due impact of
> profiling,

In my experience, profiling overhead - assuming you have half-way
modern hardware that has support for it - is simply not an issue.

So I think that if we don't have performance numbers on some hardware
where it can be shown to matter, let's just leave things be.

            Linus