* [PATCH] fprobe: Use guard(rcu_sched_notrace) and check rcu_is_watching()
@ 2026-09-29 0:19 Masami Hiramatsu (Google)
0 siblings, 0 replies; only message in thread
From: Masami Hiramatsu (Google) @ 2026-09-29 0:19 UTC (permalink / raw)
To: Steven Rostedt, Paul E . McKenney, Frederic Weisbecker, Neeraj Upadhyay
Cc: Mathieu Desnoyers, Josef Bacik, Masami Hiramatsu, linux-kernel,
linux-trace-kernel, rcu
From: Masami Hiramatsu (Google) <mhiramat@kernel.org>
unregister_fprobe() and unregister_fprobe_async() (used by BPF
kprobe-multi) rely on standard RCU grace periods (synchronize_rcu()
and call_rcu()) to wait until in-flight fprobe handlers complete before
freeing the fprobe.
However, if an fprobe handler executes while RCU is not watching (such
as in the idle loop or nohz_full extended quiescent states), standard
RCU does not track preemption-disabled sections. Consequently,
synchronize_rcu() does not wait for those executions, which can lead
to a use-after-free if the fprobe is freed immediately after
unregistration. Ensure handlers exit early when !rcu_is_watching().
Furthermore, fprobe_fgraph_entry() and fprobe_ftrace_entry() previously
used guard(rcu)() and rcu_read_lock(), which invoke lockdep on every
hit under CONFIG_PROVE_LOCKING. This adds overhead and can cause lockdep
recursion if probed functions interact with lockdep.
Since rhltable_lookup() and rhl_for_each_entry_rcu() use
rcu_dereference_all_check() (which checks rcu_read_lock_any_held()),
holding preemption disabled via rcu_read_lock_sched_notrace() is fully
valid and sufficient so long as rcu_is_watching() is true.
Define and use guard(rcu_sched_notrace)() across fprobe_ftrace_entry(),
fprobe_fgraph_entry(), and fprobe_return(). This eliminates fast-path
rcu_read_lock() and lockdep overhead while guaranteeing safe grace
period synchronization.
Reported-by: Sashiko <sashiko-bot@kernel.org>
Closes: https://sashiko.dev/#/bug/linux-e46bcd68-4a56-4f19-a255-e3772980e5e3
Fixes: 657b594b2084 ("fprobe: Fix unregister_fprobe() to wait for RCU grace period")
Cc: stable@vger.kernel.org
Assisted-by: LLM
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
---
kernel/trace/fprobe.c | 26 ++++++++++++++++----------
1 file changed, 16 insertions(+), 10 deletions(-)
diff --git a/kernel/trace/fprobe.c b/kernel/trace/fprobe.c
index 9f2d98181779..da286619c5d8 100644
--- a/kernel/trace/fprobe.c
+++ b/kernel/trace/fprobe.c
@@ -47,6 +47,10 @@ static struct rhltable fprobe_ip_table;
static DEFINE_MUTEX(fprobe_mutex);
static struct fgraph_ops fprobe_graph_ops;
+DEFINE_LOCK_GUARD_0(rcu_sched_notrace,
+ rcu_read_lock_sched_notrace(),
+ rcu_read_unlock_sched_notrace())
+
static u32 fprobe_node_hashfn(const void *data, u32 len, u32 seed)
{
return hash_ptr(*(unsigned long **)data, 32);
@@ -329,16 +333,14 @@ static void fprobe_ftrace_entry(unsigned long ip, unsigned long parent_ip,
struct fprobe *fp;
int bit;
+ if (!rcu_is_watching())
+ return;
+
bit = ftrace_test_recursion_trylock(ip, parent_ip);
if (bit < 0)
return;
- /*
- * ftrace_test_recursion_trylock() disables preemption, but
- * rhltable_lookup() checks whether rcu_read_lcok is held.
- * So we take rcu_read_lock() here.
- */
- rcu_read_lock();
+ guard(rcu_sched_notrace)();
head = rhltable_lookup(&fprobe_ip_table, &ip, fprobe_rht_params);
rhl_for_each_entry_rcu(node, pos, head, hlist) {
@@ -353,7 +355,6 @@ static void fprobe_ftrace_entry(unsigned long ip, unsigned long parent_ip,
else
__fprobe_handler(ip, parent_ip, fp, fregs, NULL);
}
- rcu_read_unlock();
ftrace_test_recursion_unlock(bit);
}
NOKPROBE_SYMBOL(fprobe_ftrace_entry);
@@ -567,10 +568,13 @@ static int fprobe_fgraph_entry(struct ftrace_graph_ent *trace, struct fgraph_ops
struct fprobe *fp;
int used, ret;
+ if (!rcu_is_watching())
+ return 0;
+
if (WARN_ON_ONCE(!fregs))
return 0;
- guard(rcu)();
+ guard(rcu_sched_notrace)();
head = rhltable_lookup(&fprobe_ip_table, &func, fprobe_rht_params);
reserved_words = 0;
rhl_for_each_entry_rcu(node, pos, head, hlist) {
@@ -665,13 +669,16 @@ static void fprobe_return(struct ftrace_graph_ret *trace,
int size, curr;
int size_words;
+ if (!rcu_is_watching())
+ return;
+
fgraph_data = (unsigned long *)fgraph_retrieve_data(gops->idx, &size);
if (WARN_ON_ONCE(!fgraph_data))
return;
size_words = SIZE_IN_LONG(size);
ret_ip = ftrace_regs_get_instruction_pointer(fregs);
- preempt_disable_notrace();
+ guard(rcu_sched_notrace)();
curr = 0;
while (size_words > curr) {
@@ -687,7 +694,6 @@ static void fprobe_return(struct ftrace_graph_ret *trace,
}
curr += size;
}
- preempt_enable_notrace();
}
NOKPROBE_SYMBOL(fprobe_return);
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2026-09-29 0:19 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-09-29 0:19 [PATCH] fprobe: Use guard(rcu_sched_notrace) and check rcu_is_watching() Masami Hiramatsu (Google)
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®