Re: [PATCH bpf-next] bpf,x86: do RSB balance for trampoline

[Alexei Starovoitov]

On Mon, Nov 10, 2025 at 5:28 PM Menglong Dong <menglong.dong@xxxxxxxxx> wrote:
>
>
> Some kind. According to my testing, the performance of bpf
> trampoline is much better than ftrace trampoline, so if we
> can implement it with bpf trampoline, the performance can be
> improved. Of course, the bpf trampoline need to offer a API
> to the livepatch for this propose.

Sure, then improve ftrace trampoline by doing the same tricks
as bpf trampoline.

> Any way, let me finish the work in this patch first. After that,
> I can send a RFC of the proposal.

Don't. livepathcing is not a job of bpf trampoline.
Song recently fixed interaction between livepatch and
fexit. We will not be adding another dimension
of complexity here where bpf trampoline is used for
livepatching and for bpf progs.