On Thu, Jan 18, 2024 at 11:50 AM Eduard Zingerman <eddyz87@xxxxxxxxx> wrote: > > On Wed, 2024-01-17 at 14:33 -0800, Andrii Nakryiko wrote: > [...] > > > diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c > > index c5a42ac309fd..61db92189517 100644 > > --- a/tools/lib/bpf/libbpf.c > > +++ b/tools/lib/bpf/libbpf.c > > @@ -6757,6 +6757,69 @@ static int clone_func_btf_info(struct btf *btf, int orig_fn_id, struct bpf_progr > > return fn_id; > > } > > > > +static int probe_kern_arg_ctx_tag(void) > > [...] > > > + btf_fd = libbpf__load_raw_btf((char *)types, sizeof(types), strs, sizeof(strs)); > > + if (btf_fd < 0) > > + return 0; > > Question: > suppose this is an old kernel and decl tags are not supported, > should negative result be cached as 0 in such case? yep, it should, and it will be eventually when probe_kern_arg_ctx_tag() will be plugged back into kernel_supports() framework. For now this small inefficiency seems fine, given it's temporary. > > [...] > >
