[PATCH v6 5/5] tracing: Convert sys_enter/exit to faultable tracepoints

[Mathieu Desnoyers]

Convert the definition of the system call enter/exit tracepoints to
faultable tracepoints now that all upstream tracers handle it.

This allows tracers to fault-in userspace system call arguments such as
path strings within their probe callbacks.

Link: https://lore.kernel.org/lkml/20231002202531.3160-1-mathieu.desnoyers@xxxxxxxxxxxx/
Co-developed-by: Michael Jeanson <mjeanson@xxxxxxxxxxxx>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx>
Signed-off-by: Michael Jeanson <mjeanson@xxxxxxxxxxxx>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@xxxxxxxxxx>
Cc: Steven Rostedt <rostedt@xxxxxxxxxxx>
Cc: Masami Hiramatsu <mhiramat@xxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Alexei Starovoitov <ast@xxxxxxxxxx>
Cc: Yonghong Song <yhs@xxxxxx>
Cc: Paul E. McKenney <paulmck@xxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
Cc: Mark Rutland <mark.rutland@xxxxxxx>
Cc: Alexander Shishkin <alexander.shishkin@xxxxxxxxxxxxxxx>
Cc: Namhyung Kim <namhyung@xxxxxxxxxx>
Cc: bpf@xxxxxxxxxxxxxxx
Cc: Joel Fernandes <joel@xxxxxxxxxxxxxxxxx>
---
Since v4:
- Use 'guard(preempt_notrace)'.
- Add brackets to multiline 'if' statements.
---
 include/trace/events/syscalls.h |  4 +--
 kernel/trace/trace_syscalls.c   | 52 ++++++++++++++++++++++++++++-----
 2 files changed, 46 insertions(+), 10 deletions(-)

diff --git a/include/trace/events/syscalls.h b/include/trace/events/syscalls.h
index b6e0cbc2c71f..dc30e3004818 100644
--- a/include/trace/events/syscalls.h
+++ b/include/trace/events/syscalls.h
@@ -15,7 +15,7 @@
 
 #ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS
 
-TRACE_EVENT_FN(sys_enter,
+TRACE_EVENT_FN_MAY_FAULT(sys_enter,
 
 	TP_PROTO(struct pt_regs *regs, long id),
 
@@ -41,7 +41,7 @@ TRACE_EVENT_FN(sys_enter,
 
 TRACE_EVENT_FLAGS(sys_enter, TRACE_EVENT_FL_CAP_ANY)
 
-TRACE_EVENT_FN(sys_exit,
+TRACE_EVENT_FN_MAY_FAULT(sys_exit,
 
 	TP_PROTO(struct pt_regs *regs, long ret),
 
diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c
index 9c581d6da843..314666d663b6 100644
--- a/kernel/trace/trace_syscalls.c
+++ b/kernel/trace/trace_syscalls.c
@@ -299,6 +299,12 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id)
 	int syscall_nr;
 	int size;
 
+	/*
+	 * Probe called with preemption enabled (may_fault), but ring buffer and
+	 * per-cpu data require preemption to be disabled.
+	 */
+	guard(preempt_notrace)();
+
 	syscall_nr = trace_get_syscall_nr(current, regs);
 	if (syscall_nr < 0 || syscall_nr >= NR_syscalls)
 		return;
@@ -338,6 +344,12 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret)
 	struct trace_event_buffer fbuffer;
 	int syscall_nr;
 
+	/*
+	 * Probe called with preemption enabled (may_fault), but ring buffer and
+	 * per-cpu data require preemption to be disabled.
+	 */
+	guard(preempt_notrace)();
+
 	syscall_nr = trace_get_syscall_nr(current, regs);
 	if (syscall_nr < 0 || syscall_nr >= NR_syscalls)
 		return;
@@ -376,8 +388,11 @@ static int reg_event_syscall_enter(struct trace_event_file *file,
 	if (WARN_ON_ONCE(num < 0 || num >= NR_syscalls))
 		return -ENOSYS;
 	mutex_lock(&syscall_trace_lock);
-	if (!tr->sys_refcount_enter)
-		ret = register_trace_sys_enter(ftrace_syscall_enter, tr);
+	if (!tr->sys_refcount_enter) {
+		ret = register_trace_prio_flags_sys_enter(ftrace_syscall_enter, tr,
+							  TRACEPOINT_DEFAULT_PRIO,
+							  TRACEPOINT_MAY_FAULT);
+	}
 	if (!ret) {
 		rcu_assign_pointer(tr->enter_syscall_files[num], file);
 		tr->sys_refcount_enter++;
@@ -414,8 +429,11 @@ static int reg_event_syscall_exit(struct trace_event_file *file,
 	if (WARN_ON_ONCE(num < 0 || num >= NR_syscalls))
 		return -ENOSYS;
 	mutex_lock(&syscall_trace_lock);
-	if (!tr->sys_refcount_exit)
-		ret = register_trace_sys_exit(ftrace_syscall_exit, tr);
+	if (!tr->sys_refcount_exit) {
+		ret = register_trace_prio_flags_sys_exit(ftrace_syscall_exit, tr,
+							 TRACEPOINT_DEFAULT_PRIO,
+							 TRACEPOINT_MAY_FAULT);
+	}
 	if (!ret) {
 		rcu_assign_pointer(tr->exit_syscall_files[num], file);
 		tr->sys_refcount_exit++;
@@ -582,6 +600,12 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id)
 	int rctx;
 	int size;
 
+	/*
+	 * Probe called with preemption enabled (may_fault), but ring buffer and
+	 * per-cpu data require preemption to be disabled.
+	 */
+	guard(preempt_notrace)();
+
 	syscall_nr = trace_get_syscall_nr(current, regs);
 	if (syscall_nr < 0 || syscall_nr >= NR_syscalls)
 		return;
@@ -630,8 +654,11 @@ static int perf_sysenter_enable(struct trace_event_call *call)
 	num = ((struct syscall_metadata *)call->data)->syscall_nr;
 
 	mutex_lock(&syscall_trace_lock);
-	if (!sys_perf_refcount_enter)
-		ret = register_trace_sys_enter(perf_syscall_enter, NULL);
+	if (!sys_perf_refcount_enter) {
+		ret = register_trace_prio_flags_sys_enter(perf_syscall_enter, NULL,
+							  TRACEPOINT_DEFAULT_PRIO,
+							  TRACEPOINT_MAY_FAULT);
+	}
 	if (ret) {
 		pr_info("event trace: Could not activate syscall entry trace point");
 	} else {
@@ -682,6 +709,12 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret)
 	int rctx;
 	int size;
 
+	/*
+	 * Probe called with preemption enabled (may_fault), but ring buffer and
+	 * per-cpu data require preemption to be disabled.
+	 */
+	guard(preempt_notrace)();
+
 	syscall_nr = trace_get_syscall_nr(current, regs);
 	if (syscall_nr < 0 || syscall_nr >= NR_syscalls)
 		return;
@@ -727,8 +760,11 @@ static int perf_sysexit_enable(struct trace_event_call *call)
 	num = ((struct syscall_metadata *)call->data)->syscall_nr;
 
 	mutex_lock(&syscall_trace_lock);
-	if (!sys_perf_refcount_exit)
-		ret = register_trace_sys_exit(perf_syscall_exit, NULL);
+	if (!sys_perf_refcount_exit) {
+		ret = register_trace_prio_flags_sys_exit(perf_syscall_exit, NULL,
+							 TRACEPOINT_DEFAULT_PRIO,
+							 TRACEPOINT_MAY_FAULT);
+	}
 	if (ret) {
 		pr_info("event trace: Could not activate syscall exit trace point");
 	} else {
-- 
2.39.2