On Thu, Sep 17, 2020 at 02:14:38PM -0700, Alexei Starovoitov wrote: > On Thu, Sep 17, 2020 at 1:25 AM Jiri Olsa <jolsa@xxxxxxxxxx> wrote: > > > > > Ideally resolve_btfids would parse dwarf info and check > > > whether any of the funcs in allowlist were inlined. > > > That would be more reliable, but not pretty to drag libdw > > > dependency into resolve_btfids. > > > > hm, we could add some check to perf|bpftrace that would > > show you all the places where function is called from and > > if it was inlined or is a regular call.. so user is aware > > what probe calls to expect > > The check like this belongs in some library, > but making libbpf depend on dwarf is not great. > I think we're at the point where we need to break libbpf > into many libraries. This one could be called libbpftrace. > It would potentially include symbolizer and other dwarf > related operations. ok > Such inlining check would be good to do not only for d_path > allowlist, but for any kprobe/fentry function. yes, that's what I meant jirka
