2023-09-19 23:10:38

by John Ogness

Subject: [PATCH printk v2 11/11] lockdep: Add atomic write enforcement for lockdep splats

Invoke the atomic write enforcement functions for lockdep
splats to ensure that the information gets out to the consoles.

It is important to note that if there are any legacy consoles
registered, they will be attempting to directly print from the
printk-caller context, which may jeopardize the reliability of
the atomic consoles. Optimally there should be no legacy
consoles registered.

Signed-off-by: John Ogness <[email protected]>
---
kernel/locking/lockdep.c | 7 +++++++
1 file changed, 7 insertions(+)

diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index e85b5ad3e206..5310a94e3efd 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -56,6 +56,7 @@
#include <linux/kprobes.h>
#include <linux/lockdep.h>
#include <linux/context_tracking.h>
+#include <linux/console.h>

#include <asm/sections.h>

@@ -3967,9 +3968,13 @@ static void
print_usage_bug(struct task_struct *curr, struct held_lock *this,
enum lock_usage_bit prev_bit, enum lock_usage_bit new_bit)
{
+ enum nbcon_prio prev_prio;
+
if (!debug_locks_off() || debug_locks_silent)
return;

+ prev_prio = nbcon_atomic_enter(NBCON_PRIO_EMERGENCY);
+
pr_warn("\n");
pr_warn("================================\n");
pr_warn("WARNING: inconsistent lock state\n");
@@ -3998,6 +4003,8 @@ print_usage_bug(struct task_struct *curr, struct held_lock *this,

pr_warn("\nstack backtrace:\n");
dump_stack();
+
+ nbcon_atomic_exit(NBCON_PRIO_EMERGENCY, prev_prio);
}

/*
--
2.39.2

2023-09-29 09:53:31

by Petr Mladek

Subject: Re: [PATCH printk v2 11/11] lockdep: Add atomic write enforcement for lockdep splats

On Wed 2023-09-20 01:14:56, John Ogness wrote:
> Invoke the atomic write enforcement functions for lockdep
> splats to ensure that the information gets out to the consoles.
>
> --- a/kernel/locking/lockdep.c
> +++ b/kernel/locking/lockdep.c
> @@ -3967,9 +3968,13 @@ static void
> print_usage_bug(struct task_struct *curr, struct held_lock *this,
> enum lock_usage_bit prev_bit, enum lock_usage_bit new_bit)
> {
> + enum nbcon_prio prev_prio;
> +
> if (!debug_locks_off() || debug_locks_silent)
> return;
>
> + prev_prio = nbcon_atomic_enter(NBCON_PRIO_EMERGENCY);
> +
> pr_warn("\n");
> pr_warn("================================\n");
> pr_warn("WARNING: inconsistent lock state\n");
> @@ -3998,6 +4003,8 @@ print_usage_bug(struct task_struct *curr, struct held_lock *this,
>
> pr_warn("\nstack backtrace:\n");
> dump_stack();
> +
> + nbcon_atomic_exit(NBCON_PRIO_EMERGENCY, prev_prio);
> }

The location of the emergency context looks good. I have just proposed
another way for tracking the emergency context. It would allow to
call nbcon_emergency_enter()/exit() without any parameter,
see https://lore.kernel.org/r/ZRLBxsXPCym2NC5Q@alley

Best Regards,
Petr