Re: [PATCH bpf-next] libbpf: Improve debug message when the base BTF cannot be found

[Olson, Matthew]

On Thu, Nov 21, 2024 at 03:55:15PM -0800, Andrii Nakryiko wrote:
> On Thu, Nov 21, 2024 at 2:08 PM Ben Olson <matthew.olson@xxxxxxxxx> wrote:
> >
> > When running `bpftool` on a kernel module installed in `/lib/modules...`,
> > this error is encountered if the user does not specify `--base-btf` to
> > point to a valid base BTF (e.g. usually in `/sys/kernel/btf/vmlinux`).
> > However, looking at the debug output to determine the cause of the error
> > simply says `Invalid BTF string section`, which does not point to the
> > actual source of the error. This just improves that debug message to tell
> > users what happened.
> >
> > Signed-off-by: Ben Olson <matthew.olson@xxxxxxxxx>
> > ---
> >  tools/lib/bpf/btf.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
> > index 12468ae0d573..1a17de9d99e6 100644
> > --- a/tools/lib/bpf/btf.c
> > +++ b/tools/lib/bpf/btf.c
> > @@ -283,7 +283,7 @@ static int btf_parse_str_sec(struct btf *btf)
> >                 return -EINVAL;
> >         }
> >         if (!btf->base_btf && start[0]) {
> > -               pr_debug("Invalid BTF string section\n");
> > +               pr_debug("Cannot find base BTF\n");
> 
> Well, the check indeed checks the well-formedness of the BTF string
> section. It is specified that the first byte has to be zero ("empty
> string"), unless it's a split BTF.
> 
> Base BTF being missing is just one possible reason for this condition,
> so I'm not sure if it's completely accurate to specialize this error
> message so much. Perhaps maybe emitting "Malformed BTF string section,
> did you forget to provide base BTF?" would be a bit better.

That sounds much better; as long as it hints that the user should
check if they specified a base BTF or not, it's good with me! Thanks.

> 
> pw-bot: cr
> 
> >                 return -EINVAL;
> >         }
> >         return 0;
> > --
> > 2.47.0
>