Re: [PATCH v4 bpf-next 00/11] bpf: support resilient split BTF

[Eduard Zingerman <eddyz87@gmail.com>]

On Tue, 2024-05-21 at 11:54 -0700, Andrii Nakryiko wrote:

[...]

> I'm probably leaning towards not doing automatic relocations in
> btf__parse(), tbh. Distilled BTF is a rather special kernel-specific
> feature, if we need to teach resolve_btfids and bpftool to do
> something extra for that case (i.e., call another API for relocation,
> if necessary), then it's fine, doesn't seems like a problem.

My point is that with current implementation it does not even make
sense to call btf__parse() for an ELF with distilled base,
because it would fail.

And selecting BTF encoding based on a few retries seems like a kludge
if there is a simple way to check if distilled base has to be used
(presence of the .BTF.base section).

> Much worse is having to do some workarounds to prevent an API from
> doing some undesired extra steps (like in resolve_btfids not wanting a
> relocation). Orthogonality FTW, IMO.

For resolve_btfids it is a bit different, imo.
It does want some base: for in-tree modules it wants vmlinux,
for out-of-tree it wants distilled base.
So it has to be adjusted either way.