linux-kernel - Re: [PATCH] libbpf: fix multi-uprobe attach not working with dynamic symbols

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CAEf4BzbGbfhdanY0yZtRoRTZaiMG4ML1PYUz1m4QbG-Kw2tNtA@mail.gmail.com>
Date: Fri, 28 Mar 2025 11:14:54 -0700
From: Andrii Nakryiko <andrii.nakryiko@...il.com>
To: Mateusz Bieganski <bieganski.gm@...il.com>
Cc: Andrii Nakryiko <andrii@...nel.org>, Eduard Zingerman <eddyz87@...il.com>, 
	Alexei Starovoitov <ast@...nel.org>, Daniel Borkmann <daniel@...earbox.net>, 
	Martin KaFai Lau <martin.lau@...ux.dev>, Song Liu <song@...nel.org>, 
	Yonghong Song <yonghong.song@...ux.dev>, John Fastabend <john.fastabend@...il.com>, 
	KP Singh <kpsingh@...nel.org>, Stanislav Fomichev <sdf@...ichev.me>, Hao Luo <haoluo@...gle.com>, 
	Jiri Olsa <jolsa@...nel.org>, "open list:BPF [LIBRARY] (libbpf)" <bpf@...r.kernel.org>, 
	open list <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] libbpf: fix multi-uprobe attach not working with dynamic symbols

On Thu, Mar 27, 2025 at 3:08 AM Mateusz Bieganski
<bieganski.gm@...il.com> wrote:
>
> ENOENT is incorrectly propagated to caller, if requested symbol is
> present in dynamic linker symbol table and not present in symbol table.
>
> Signed-off-by: Mateusz Bieganski <bieganski.gm@...il.com>
> ---
>  tools/lib/bpf/elf.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/tools/lib/bpf/elf.c b/tools/lib/bpf/elf.c
> index 823f83ad819c..41839ef5bc97 100644
> --- a/tools/lib/bpf/elf.c
> +++ b/tools/lib/bpf/elf.c
> @@ -439,8 +439,10 @@ int elf_resolve_syms_offsets(const char *binary_path, int cnt,
>                 struct elf_sym *sym;
>
>                 err = elf_sym_iter_new(&iter, elf_fd.elf, binary_path, sh_types[i], st_type);
> -               if (err == -ENOENT)
> +               if (err == -ENOENT) {
> +                       err = 0;
>                         continue;
> +               }

Don't we have the same problem in elf_resolve_pattern_offsets() as
well? Can you please fix both issues in one go?

It seems it's only elf_find_func_offset() that do want to preserve
that -ENOENT, all others are just error-prone implementations.

pw-bot: cr

>                 if (err)
>                         goto out;
>
> --
> 2.39.5
>