With the generic tracing recursion prevention mechanism applied, it is
safe to trace migrate_{disable,enable} and preempt_count_{sub,add}. So
we can remove them from the deny list.
However we can't remove rcu_read_unlock_strict and
__rcu_read_{lock,unlock}, because they are used in rcu_read_unlock() or
rcu_read_lock().
Signed-off-by: Yafang Shao <laoar.shao@xxxxxxxxx>
---
kernel/bpf/verifier.c | 8 --------
1 file changed, 8 deletions(-)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 83fb94f..40f6b2c 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -18634,17 +18634,9 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
BTF_SET_START(btf_id_deny)
BTF_ID_UNUSED
-#ifdef CONFIG_SMP
-BTF_ID(func, migrate_disable)
-BTF_ID(func, migrate_enable)
-#endif
#if !defined CONFIG_PREEMPT_RCU && !defined CONFIG_TINY_RCU
BTF_ID(func, rcu_read_unlock_strict)
#endif
-#if defined(CONFIG_DEBUG_PREEMPT) || defined(CONFIG_TRACE_PREEMPT_TOGGLE)
-BTF_ID(func, preempt_count_add)
-BTF_ID(func, preempt_count_sub)
-#endif
#ifdef CONFIG_PREEMPT_RCU
BTF_ID(func, __rcu_read_lock)
BTF_ID(func, __rcu_read_unlock)
--
1.8.3.1
