[for-next][PATCH] kprobes: Fix to delay the kprobes jump optimization

From: Steven Rostedt
Date: Sat Feb 20 2021 - 12:00:59 EST



Masami Hiramatsu (1):
kprobes: Fix to delay the kprobes jump optimization

----
kernel/kprobes.c | 31 +++++++++++++++++++++----------
1 file changed, 21 insertions(+), 10 deletions(-)
---------------------------
commit c85c9a2c6e368dc94907e63babb18a9788e5c9b6
Author: Masami Hiramatsu <mhiramat@xxxxxxxxxx>
Date: Thu Feb 18 23:29:23 2021 +0900

kprobes: Fix to delay the kprobes jump optimization

Commit 36dadef23fcc ("kprobes: Init kprobes in early_initcall")
moved the kprobe setup in early_initcall(), which includes kprobe
jump optimization.
The kprobes jump optimizer involves synchronize_rcu_tasks() which
depends on the ksoftirqd and rcu_spawn_tasks_*(). However, since
those are setup in core_initcall(), kprobes jump optimizer can not
run at the early_initcall().

To avoid this issue, make the kprobe optimization disabled in the
early_initcall() and enables it in subsys_initcall().

Note that non-optimized kprobes is still available after
early_initcall(). Only jump optimization is delayed.

Link: https://lkml.kernel.org/r/161365856280.719838.12423085451287256713.stgit@devnote2

Fixes: 36dadef23fcc ("kprobes: Init kprobes in early_initcall")
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: RCU <rcu@xxxxxxxxxxxxxxx>
Cc: Michael Ellerman <mpe@xxxxxxxxxxxxxx>
Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Cc: Daniel Axtens <dja@xxxxxxxxxx>
Cc: Frederic Weisbecker <frederic@xxxxxxxxxx>
Cc: Neeraj Upadhyay <neeraju@xxxxxxxxxxxxxx>
Cc: Joel Fernandes <joel@xxxxxxxxxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxx>
Cc: "Theodore Y . Ts'o" <tytso@xxxxxxx>
Cc: Oleksiy Avramchenko <oleksiy.avramchenko@xxxxxxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
Reported-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
Reported-by: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
Reported-by: Uladzislau Rezki <urezki@xxxxxxxxx>
Acked-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
Signed-off-by: Masami Hiramatsu <mhiramat@xxxxxxxxxx>
Signed-off-by: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx>

diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index dd1d027455c4..745f08fdd7a6 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -861,7 +861,6 @@ static void try_to_optimize_kprobe(struct kprobe *p)
cpus_read_unlock();
}

-#ifdef CONFIG_SYSCTL
static void optimize_all_kprobes(void)
{
struct hlist_head *head;
@@ -887,6 +886,7 @@ static void optimize_all_kprobes(void)
mutex_unlock(&kprobe_mutex);
}

+#ifdef CONFIG_SYSCTL
static void unoptimize_all_kprobes(void)
{
struct hlist_head *head;
@@ -2500,18 +2500,14 @@ static int __init init_kprobes(void)
}
}

-#if defined(CONFIG_OPTPROBES)
-#if defined(__ARCH_WANT_KPROBES_INSN_SLOT)
- /* Init kprobe_optinsn_slots */
- kprobe_optinsn_slots.insn_size = MAX_OPTINSN_SIZE;
-#endif
- /* By default, kprobes can be optimized */
- kprobes_allow_optimization = true;
-#endif
-
/* By default, kprobes are armed */
kprobes_all_disarmed = false;

+#if defined(CONFIG_OPTPROBES) && defined(__ARCH_WANT_KPROBES_INSN_SLOT)
+ /* Init kprobe_optinsn_slots for allocation */
+ kprobe_optinsn_slots.insn_size = MAX_OPTINSN_SIZE;
+#endif
+
err = arch_init_kprobes();
if (!err)
err = register_die_notifier(&kprobe_exceptions_nb);
@@ -2526,6 +2522,21 @@ static int __init init_kprobes(void)
}
early_initcall(init_kprobes);

+#if defined(CONFIG_OPTPROBES)
+static int __init init_optprobes(void)
+{
+ /*
+ * Enable kprobe optimization - this kicks the optimizer which
+ * depends on synchronize_rcu_tasks() and ksoftirqd, that is
+ * not spawned in early initcall. So delay the optimization.
+ */
+ optimize_all_kprobes();
+
+ return 0;
+}
+subsys_initcall(init_optprobes);
+#endif
+
#ifdef CONFIG_DEBUG_FS
static void report_probe(struct seq_file *pi, struct kprobe *p,
const char *sym, int offset, char *modname, struct kprobe *pp)