On Tue, Aug 05, 2025 at 08:23:12PM +0800, Tao Chen wrote:
> bpf program should run under migration disabled, kprobe_multi_link_prog_run
> called the way from graph tracer, which disables preemption in
> function_graph_enter_regs, as Jiri and Yonghong suggested, there is no
> need to use migrate_disable. As a result, some overhead maybe will be
> reduced.
>
> Fixes: 0dcac2725406 ("bpf: Add multi kprobe link")
> Signed-off-by: Tao Chen <chen.dylane@xxxxxxxxx>
Acked-by: Jiri Olsa <jolsa@xxxxxxxxxx>
thanks,
jirka
> ---
> kernel/trace/bpf_trace.c | 9 +++++++--
> 1 file changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
> index 3ae52978cae..1993fc62539 100644
> --- a/kernel/trace/bpf_trace.c
> +++ b/kernel/trace/bpf_trace.c
> @@ -2734,14 +2734,19 @@ kprobe_multi_link_prog_run(struct bpf_kprobe_multi_link *link,
> goto out;
> }
>
> - migrate_disable();
> + /*
> + * bpf program should run under migration disabled, kprobe_multi_link_prog_run
> + * called the way from graph tracer, which disables preemption in
nit, s/called the way/called all the way/
> + * function_graph_enter_regs, so there is no need to use migrate_disable.
> + * Accessing the above percpu data bpf_prog_active is also safe for the same
> + * reason.
> + */
> rcu_read_lock();
> regs = ftrace_partial_regs(fregs, bpf_kprobe_multi_pt_regs_ptr());
> old_run_ctx = bpf_set_run_ctx(&run_ctx.session_ctx.run_ctx);
> err = bpf_prog_run(link->link.prog, regs);
> bpf_reset_run_ctx(old_run_ctx);
> rcu_read_unlock();
> - migrate_enable();
>
> out:
> __this_cpu_dec(bpf_prog_active);
> --
> 2.48.1
>
