bpf.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jiri Olsa <jolsa@kernel.org>
To: Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Andrii Nakryiko <andriin@fb.com>
Cc: "kernel test robot" <lkp@intel.com>,
	"Julia Lawall" <julia.lawall@lip6.fr>,
	"Toke Høiland-Jørgensen" <toke@redhat.com>,
	netdev@vger.kernel.org, bpf@vger.kernel.org,
	"Martin KaFai Lau" <kafai@fb.com>,
	"Song Liu" <songliubraving@fb.com>, "Yonghong Song" <yhs@fb.com>,
	"John Fastabend" <john.fastabend@gmail.com>,
	"KP Singh" <kpsingh@chromium.org>,
	"Julia Lawall" <julia.lawall@inria.fr>
Subject: [PATCHv5 bpf-next 1/7] bpf: Allow trampoline re-attach for tracing and lsm programs
Date: Wed, 14 Apr 2021 21:51:41 +0200	[thread overview]
Message-ID: <20210414195147.1624932-2-jolsa@kernel.org> (raw)
In-Reply-To: <20210414195147.1624932-1-jolsa@kernel.org>

Currently we don't allow re-attaching of trampolines. Once
it's detached, it can't be re-attach even when the program
is still loaded.

Adding the possibility to re-attach the loaded tracing and
lsm programs.

Fixing missing unlock with proper cleanup goto jump reported
by Julia.

Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Julia Lawall <julia.lawall@lip6.fr>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
---
 kernel/bpf/syscall.c    | 23 +++++++++++++++++------
 kernel/bpf/trampoline.c |  4 ++--
 2 files changed, 19 insertions(+), 8 deletions(-)

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index fd495190115e..941ca06d9dfa 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -2648,14 +2648,25 @@ static int bpf_tracing_prog_attach(struct bpf_prog *prog,
 	 *   target_btf_id using the link_create API.
 	 *
 	 * - if tgt_prog == NULL when this function was called using the old
-         *   raw_tracepoint_open API, and we need a target from prog->aux
-         *
-         * The combination of no saved target in prog->aux, and no target
-         * specified on load is illegal, and we reject that here.
+	 *   raw_tracepoint_open API, and we need a target from prog->aux
+	 *
+	 * - if prog->aux->dst_trampoline and tgt_prog is NULL, the program
+	 *   was detached and is going for re-attachment.
 	 */
 	if (!prog->aux->dst_trampoline && !tgt_prog) {
-		err = -ENOENT;
-		goto out_unlock;
+		/*
+		 * Allow re-attach for TRACING and LSM programs. If it's
+		 * currently linked, bpf_trampoline_link_prog will fail.
+		 * EXT programs need to specify tgt_prog_fd, so they
+		 * re-attach in separate code path.
+		 */
+		if (prog->type != BPF_PROG_TYPE_TRACING &&
+		    prog->type != BPF_PROG_TYPE_LSM) {
+			err = -EINVAL;
+			goto out_unlock;
+		}
+		btf_id = prog->aux->attach_btf_id;
+		key = bpf_trampoline_compute_key(NULL, prog->aux->attach_btf, btf_id);
 	}
 
 	if (!prog->aux->dst_trampoline ||
diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index 1f3a4be4b175..205c2cc36ad7 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -414,7 +414,7 @@ int bpf_trampoline_link_prog(struct bpf_prog *prog, struct bpf_trampoline *tr)
 	tr->progs_cnt[kind]++;
 	err = bpf_trampoline_update(tr);
 	if (err) {
-		hlist_del(&prog->aux->tramp_hlist);
+		hlist_del_init(&prog->aux->tramp_hlist);
 		tr->progs_cnt[kind]--;
 	}
 out:
@@ -437,7 +437,7 @@ int bpf_trampoline_unlink_prog(struct bpf_prog *prog, struct bpf_trampoline *tr)
 		tr->extension_prog = NULL;
 		goto out;
 	}
-	hlist_del(&prog->aux->tramp_hlist);
+	hlist_del_init(&prog->aux->tramp_hlist);
 	tr->progs_cnt[kind]--;
 	err = bpf_trampoline_update(tr);
 out:
-- 
2.30.2


  reply	other threads:[~2021-04-14 19:52 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-14 19:51 [PATCHv5 bpf-next 0/7] bpf: Tracing and lsm programs re-attach Jiri Olsa
2021-04-14 19:51 ` Jiri Olsa [this message]
2021-04-15 23:22   ` [PATCHv5 bpf-next 1/7] bpf: Allow trampoline re-attach for tracing and lsm programs Andrii Nakryiko
2021-04-19 22:47     ` KP Singh
2021-04-14 19:51 ` [PATCHv5 bpf-next 2/7] selftests/bpf: Add missing semicolon Jiri Olsa
2021-04-15 23:23   ` Andrii Nakryiko
2021-04-14 19:51 ` [PATCHv5 bpf-next 3/7] selftests/bpf: Add re-attach test to fentry_test Jiri Olsa
2021-04-14 19:51 ` [PATCHv5 bpf-next 4/7] selftests/bpf: Add re-attach test to fexit_test Jiri Olsa
2021-04-14 19:51 ` [PATCHv5 bpf-next 5/7] selftests/bpf: Add re-attach test to lsm test Jiri Olsa
2021-04-14 19:51 ` [PATCHv5 bpf-next 6/7] selftests/bpf: Test that module can't be unloaded with attached trampoline Jiri Olsa
2021-04-14 19:51 ` [PATCHv5 bpf-next 7/7] selftests/bpf: Use ASSERT macros in lsm test Jiri Olsa
2021-04-15 23:24   ` Andrii Nakryiko
2021-04-15 23:45 ` [PATCHv5 bpf-next 0/7] bpf: Tracing and lsm programs re-attach Alexei Starovoitov
2021-04-16  6:55   ` Jiri Olsa
2021-04-26  4:14     ` Alexei Starovoitov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210414195147.1624932-2-jolsa@kernel.org \
    --to=jolsa@kernel.org \
    --cc=andriin@fb.com \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=john.fastabend@gmail.com \
    --cc=julia.lawall@inria.fr \
    --cc=julia.lawall@lip6.fr \
    --cc=kafai@fb.com \
    --cc=kpsingh@chromium.org \
    --cc=lkp@intel.com \
    --cc=netdev@vger.kernel.org \
    --cc=songliubraving@fb.com \
    --cc=toke@redhat.com \
    --cc=yhs@fb.com \
    --subject='Re: [PATCHv5 bpf-next 1/7] bpf: Allow trampoline re-attach for tracing and lsm programs' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).