Received-SPF: pass (google.com: domain of linux-kernel+bounces-197872-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
From: Mateusz Guzik <mjguzik@gmail.com>
To: ast@kernel.org
Cc: linux-kernel@vger.kernel.org,
	bpf@vger.kernel.org,
	Mateusz Guzik <mjguzik@gmail.com>
Subject: [PATCH] bpf: plug a warn about bpf_session_cookie without CONFIG_FPROBE
Date: Sat,  1 Jun 2024 15:32:23 +0200
Message-ID: <20240601133224.674784-1-mjguzik@gmail.com>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

Building a kernel without said option results in:
WARN: resolve_btfids: unresolved symbol bpf_session_cookie

This is a bare-minimum patch to sort it out.

There are other uses of the bpf_session_cookie thing spread out
thorought the file, they don't seem to break anything though.

Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
---

I don't care how this gets addressed, I just want the warning gone.
So I am not going to fight any ideas how to do it, as long as it gets
done.

 kernel/bpf/verifier.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 48f3a9acdef3..b081bdd6f477 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -11128,7 +11128,9 @@ BTF_ID(func, bpf_iter_css_task_new)
 #else
 BTF_ID_UNUSED
 #endif
+#ifdef CONFIG_FPROBES
 BTF_ID(func, bpf_session_cookie)
+#endif
 
 static bool is_kfunc_ret_null(struct bpf_kfunc_call_arg_meta *meta)
 {
-- 
2.39.2