On Fri, 27 Sep 2024 11:13:30 -0700
Andrii Nakryiko <andrii.nakryiko@xxxxxxxxx> wrote:
> Hm... sounds like two versions of my patch were applied to two
> different trees or something? FWIW, 10cdb82aa77f is the right one cto
> pick (I didn't check which one is in Linus' tree), but the differences
> are tiny.
>
> diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c
> index 87b468d93f6a..c3df411a2684 100644
> --- a/kernel/trace/trace_uprobe.c
> +++ b/kernel/trace/trace_uprobe.c
> @@ -834,7 +834,7 @@ static int probes_profile_seq_show(struct seq_file
> *m, void *v)
>
> nhits = 0;
> for_each_possible_cpu(cpu) {
> - nhits += READ_ONCE(*per_cpu_ptr(tu->nhits, cpu));
> + nhits += per_cpu(*tu->nhits, cpu);
> }
>
> seq_printf(m, " %s %-44s %15lu\n", tu->filename,
>
> >
It looks like Masami rebased his tree and I didn't do the update yet.
I updated the latest for-next in the tracing repo, so everything should be
good again.
-- Steve
