Re: [PATCHv2 bpf-next 4/9] libbpf: Add support for uprobe multi session attach

[Jiri Olsa]

On Tue, Jul 02, 2024 at 02:34:34PM -0700, Andrii Nakryiko wrote:
> On Mon, Jul 1, 2024 at 9:42 AM Jiri Olsa <jolsa@xxxxxxxxxx> wrote:
> >
> > Adding support to attach program in uprobe session mode
> > with bpf_program__attach_uprobe_multi function.
> >
> > Adding session bool to bpf_uprobe_multi_opts struct that allows
> > to load and attach the bpf program via uprobe session.
> > the attachment to create uprobe multi session.
> >
> > Also adding new program loader section that allows:
> >   SEC("uprobe.session/bpf_fentry_test*")
> >
> > and loads/attaches uprobe program as uprobe session.
> >
> > Signed-off-by: Jiri Olsa <jolsa@xxxxxxxxxx>
> > ---
> >  tools/lib/bpf/bpf.c    |  1 +
> >  tools/lib/bpf/libbpf.c | 50 ++++++++++++++++++++++++++++++++++++++++--
> >  tools/lib/bpf/libbpf.h |  4 +++-
> >  3 files changed, 52 insertions(+), 3 deletions(-)
> >
> 
> [...]
> 
> > @@ -9362,6 +9363,7 @@ static const struct bpf_sec_def section_defs[] = {
> >         SEC_DEF("kprobe.session+",      KPROBE, BPF_TRACE_KPROBE_SESSION, SEC_NONE, attach_kprobe_session),
> >         SEC_DEF("uprobe.multi+",        KPROBE, BPF_TRACE_UPROBE_MULTI, SEC_NONE, attach_uprobe_multi),
> >         SEC_DEF("uretprobe.multi+",     KPROBE, BPF_TRACE_UPROBE_MULTI, SEC_NONE, attach_uprobe_multi),
> > +       SEC_DEF("uprobe.session+",      KPROBE, BPF_TRACE_UPROBE_SESSION, SEC_NONE, attach_uprobe_session),
> 
> sleepable ones as well?

ah right, forgot.. will add

> 
> >         SEC_DEF("uprobe.multi.s+",      KPROBE, BPF_TRACE_UPROBE_MULTI, SEC_SLEEPABLE, attach_uprobe_multi),
> >         SEC_DEF("uretprobe.multi.s+",   KPROBE, BPF_TRACE_UPROBE_MULTI, SEC_SLEEPABLE, attach_uprobe_multi),
> >         SEC_DEF("ksyscall+",            KPROBE, 0, SEC_NONE, attach_ksyscall),
> > @@ -11698,6 +11700,40 @@ static int attach_uprobe_multi(const struct bpf_program *prog, long cookie, stru
> >         return ret;
> >  }
> >
> > +static int attach_uprobe_session(const struct bpf_program *prog, long cookie, struct bpf_link **link)
> > +{
> > +       char *binary_path = NULL, *func_name = NULL;
> > +       LIBBPF_OPTS(bpf_uprobe_multi_opts, opts,
> > +               .session = true,
> > +       );
> 
> nit: keep a single line?

ok

> 
> > +       int n, ret = -EINVAL;
> > +       const char *spec;
> > +
> > +       *link = NULL;
> > +
> > +       spec = prog->sec_name + sizeof("uprobe.session/") - 1;
> > +       n = sscanf(spec, "%m[^:]:%m[^\n]",
> > +                  &binary_path, &func_name);
> 
> single line, wrapping lines is a necessary evil, please

ok

thanks,
jirka