On Fri, 26 Feb 2016 03:31:19 +0900
Taeung Song <treeze.tae...@gmail.com> wrote:
>
> OK, I will test it.
> But IMHO, I think the bottom change has a problem.
> Because sys_enter_io_getevent() has a argument 'long nr'.
> So this if statement must not have strcmp(sc->args->name, "nr") == 0.
>
> + if (sc->args && strcmp(sc->args->name, "syscall_nr") == 0) {
>
> I think the above instance seem better than the bottom.
>
> + if (sc->args && (strcmp(sc->args->name, "syscall_nr") ||
> strcmp(sc->args->name, "nr")) == 0) {
>
> But I'll test again with perf-trace.
> And then will say the result.
But doesn't this break new perf running on older kernels? We can't have
that either.
>
> >
> > diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
> > index 20916dd77aac..b31eed102a83 100644
> > --- a/tools/perf/builtin-trace.c
> > +++ b/tools/perf/builtin-trace.c
> > @@ -1724,8 +1724,8 @@ static int trace__read_syscall_info(struct trace
> > *trace, int id)
> >
> > sc->args = sc->tp_format->format.fields;
> > sc->nr_args = sc->tp_format->format.nr_fields;
> > - /* drop nr field - not relevant here; does not exist on older kernels */
> > - if (sc->args && strcmp(sc->args->name, "nr") == 0) {
> > + /* drop (syscall_)?nr field - not relevant here; does not exist on
> > older kernels */
> > + if (sc->args && (strcmp(sc->args->name, "syscall_nr") ||
> > strcmp(sc->args->name, "nr")) == 0) {
> > sc->args = sc->args->next;
> > --sc->nr_args;
> > }
> >
> >
> > ----------------------
> >
> > But then I wonder if it wouldn't be better to prefix this with double
> > underscores, making it "__syscall_nr" :-\
> >
>
> I so agree. Low probability but the name 'syscall_nr' may also
> have similar problems.
I honestly doubt it.
-- Steve
