Received-SPF: pass (google.com: domain of linux-kernel+bounces-201315-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
From: Jiri Olsa <jolsa@kernel.org>
To: Oleg Nesterov <oleg@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Andrii Nakryiko <andrii@kernel.org>
Cc: 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>,
	Stanislav Fomichev <sdf@google.com>,
	Hao Luo <haoluo@google.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Masami Hiramatsu <mhiramat@kernel.org>,
	linux-kernel@vger.kernel.org,
	linux-trace-kernel@vger.kernel.org
Subject: [RFC bpf-next 02/10] bpf: Add support for uprobe multi session attach
Date: Tue,  4 Jun 2024 22:02:13 +0200
Message-ID: <20240604200221.377848-3-jolsa@kernel.org>
In-Reply-To: <20240604200221.377848-1-jolsa@kernel.org>
References: <20240604200221.377848-1-jolsa@kernel.org>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

Adding support to attach bpf program for entry and return probe
of the same function. This is common use case which at the moment
requires to create two uprobe multi links.

Adding new BPF_TRACE_UPROBE_SESSION attach type that instructs
kernel to attach single link program to both entry and exit probe.

It's possible to control execution of the bpf program on return
probe simply by returning zero or non zero from the entry bpf
program execution to execute or not the bpf program on return
probe respectively.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
---
 include/uapi/linux/bpf.h       |  1 +
 kernel/bpf/syscall.c           |  9 ++++--
 kernel/trace/bpf_trace.c       | 50 +++++++++++++++++++++++++++++-----
 tools/include/uapi/linux/bpf.h |  1 +
 4 files changed, 52 insertions(+), 9 deletions(-)

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 25ea393cf084..b400f50e2c3c 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -1116,6 +1116,7 @@ enum bpf_attach_type {
 	BPF_NETKIT_PRIMARY,
 	BPF_NETKIT_PEER,
 	BPF_TRACE_KPROBE_SESSION,
+	BPF_TRACE_UPROBE_SESSION,
 	__MAX_BPF_ATTACH_TYPE
 };
 
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 5070fa20d05c..71d279907a0c 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -4048,10 +4048,14 @@ static int bpf_prog_attach_check_attach_type(const struct bpf_prog *prog,
 		if (prog->expected_attach_type == BPF_TRACE_UPROBE_MULTI &&
 		    attach_type != BPF_TRACE_UPROBE_MULTI)
 			return -EINVAL;
+		if (prog->expected_attach_type == BPF_TRACE_UPROBE_SESSION &&
+		    attach_type != BPF_TRACE_UPROBE_SESSION)
+			return -EINVAL;
 		if (attach_type != BPF_PERF_EVENT &&
 		    attach_type != BPF_TRACE_KPROBE_MULTI &&
 		    attach_type != BPF_TRACE_KPROBE_SESSION &&
-		    attach_type != BPF_TRACE_UPROBE_MULTI)
+		    attach_type != BPF_TRACE_UPROBE_MULTI &&
+		    attach_type != BPF_TRACE_UPROBE_SESSION)
 			return -EINVAL;
 		return 0;
 	case BPF_PROG_TYPE_SCHED_CLS:
@@ -5314,7 +5318,8 @@ static int link_create(union bpf_attr *attr, bpfptr_t uattr)
 		else if (attr->link_create.attach_type == BPF_TRACE_KPROBE_MULTI ||
 			 attr->link_create.attach_type == BPF_TRACE_KPROBE_SESSION)
 			ret = bpf_kprobe_multi_link_attach(attr, prog);
-		else if (attr->link_create.attach_type == BPF_TRACE_UPROBE_MULTI)
+		else if (attr->link_create.attach_type == BPF_TRACE_UPROBE_MULTI ||
+			 attr->link_create.attach_type == BPF_TRACE_UPROBE_SESSION)
 			ret = bpf_uprobe_multi_link_attach(attr, prog);
 		break;
 	default:
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index f5154c051d2c..53b111c8e887 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -1642,6 +1642,17 @@ static inline bool is_kprobe_session(const struct bpf_prog *prog)
 	return prog->expected_attach_type == BPF_TRACE_KPROBE_SESSION;
 }
 
+static inline bool is_uprobe_multi(const struct bpf_prog *prog)
+{
+	return prog->expected_attach_type == BPF_TRACE_UPROBE_MULTI ||
+	       prog->expected_attach_type == BPF_TRACE_UPROBE_SESSION;
+}
+
+static inline bool is_uprobe_session(const struct bpf_prog *prog)
+{
+	return prog->expected_attach_type == BPF_TRACE_UPROBE_SESSION;
+}
+
 static const struct bpf_func_proto *
 kprobe_prog_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 {
@@ -1659,13 +1670,13 @@ kprobe_prog_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 	case BPF_FUNC_get_func_ip:
 		if (is_kprobe_multi(prog))
 			return &bpf_get_func_ip_proto_kprobe_multi;
-		if (prog->expected_attach_type == BPF_TRACE_UPROBE_MULTI)
+		if (is_uprobe_multi(prog))
 			return &bpf_get_func_ip_proto_uprobe_multi;
 		return &bpf_get_func_ip_proto_kprobe;
 	case BPF_FUNC_get_attach_cookie:
 		if (is_kprobe_multi(prog))
 			return &bpf_get_attach_cookie_proto_kmulti;
-		if (prog->expected_attach_type == BPF_TRACE_UPROBE_MULTI)
+		if (is_uprobe_multi(prog))
 			return &bpf_get_attach_cookie_proto_umulti;
 		return &bpf_get_attach_cookie_proto_trace;
 	default:
@@ -3346,6 +3357,26 @@ uprobe_multi_link_ret_handler(struct uprobe_consumer *con, unsigned long func, s
 	return uprobe_prog_run(uprobe, func, regs);
 }
 
+static int
+uprobe_multi_link_handler_session(struct uprobe_consumer *con, struct pt_regs *regs,
+				  unsigned long *data)
+{
+	struct bpf_uprobe *uprobe;
+
+	uprobe = container_of(con, struct bpf_uprobe, consumer);
+	return uprobe_prog_run(uprobe, instruction_pointer(regs), regs);
+}
+
+static int
+uprobe_multi_link_ret_handler_session(struct uprobe_consumer *con, unsigned long func,
+				      struct pt_regs *regs, unsigned long *data)
+{
+	struct bpf_uprobe *uprobe;
+
+	uprobe = container_of(con, struct bpf_uprobe, consumer);
+	return uprobe_prog_run(uprobe, func, regs);
+}
+
 static u64 bpf_uprobe_multi_entry_ip(struct bpf_run_ctx *ctx)
 {
 	struct bpf_uprobe_multi_run_ctx *run_ctx;
@@ -3382,7 +3413,7 @@ int bpf_uprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr
 	if (sizeof(u64) != sizeof(void *))
 		return -EOPNOTSUPP;
 
-	if (prog->expected_attach_type != BPF_TRACE_UPROBE_MULTI)
+	if (!is_uprobe_multi(prog))
 		return -EINVAL;
 
 	flags = attr->link_create.uprobe_multi.flags;
@@ -3460,10 +3491,15 @@ int bpf_uprobe_multi_link_attach(const union bpf_attr *attr, struct bpf_prog *pr
 
 		uprobes[i].link = link;
 
-		if (flags & BPF_F_UPROBE_MULTI_RETURN)
-			uprobes[i].consumer.ret_handler = uprobe_multi_link_ret_handler;
-		else
-			uprobes[i].consumer.handler = uprobe_multi_link_handler;
+		if (is_uprobe_session(prog)) {
+			uprobes[i].consumer.handler_session = uprobe_multi_link_handler_session;
+			uprobes[i].consumer.ret_handler_session = uprobe_multi_link_ret_handler_session;
+		} else {
+			if (flags & BPF_F_UPROBE_MULTI_RETURN)
+				uprobes[i].consumer.ret_handler = uprobe_multi_link_ret_handler;
+			else
+				uprobes[i].consumer.handler = uprobe_multi_link_handler;
+		}
 
 		if (pid)
 			uprobes[i].consumer.filter = uprobe_multi_link_filter;
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 25ea393cf084..b400f50e2c3c 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -1116,6 +1116,7 @@ enum bpf_attach_type {
 	BPF_NETKIT_PRIMARY,
 	BPF_NETKIT_PEER,
 	BPF_TRACE_KPROBE_SESSION,
+	BPF_TRACE_UPROBE_SESSION,
 	__MAX_BPF_ATTACH_TYPE
 };
 
-- 
2.45.1