FAILED: patch "[PATCH] printk: fix deadlock when kernel panic" failed to apply to 4.9-stable tree

The patch below does not apply to the 4.9-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 8a8109f303e25a27f92c1d8edd67d7cbbc60a4eb Mon Sep 17 00:00:00 2001
From: Muchun Song <[email protected]>
Date: Wed, 10 Feb 2021 11:48:23 +0800
Subject: [PATCH] printk: fix deadlock when kernel panic

printk_safe_flush_on_panic() caused the following deadlock on our server:

CPU0: CPU1:
panic rcu_dump_cpu_stacks
kdump_nmi_shootdown_cpus nmi_trigger_cpumask_backtrace

    register_nmi_handler(crash_nmi_callback)      printk_safe_flush
                                                    __printk_safe_flush
                                                      raw_spin_lock_irqsave(&read_lock)
    // send NMI to other processors
    apic_send_IPI_allbutself(NMI_VECTOR)
                                                        // NMI interrupt, dead loop
                                                        crash_nmi_callback
  printk_safe_flush_on_panic
    printk_safe_flush
      __printk_safe_flush
        // deadlock
        raw_spin_lock_irqsave(&read_lock)

DEADLOCK: read_lock is taken on CPU1 and will never get released.

It happens when panic() stops a CPU by NMI while it has been in the middle of printk_safe_flush().

Handle the lock the same way as logbuf_lock. The printk_safe buffers are flushed only when both locks can be safely taken. It can avoid the deadlock _in this particular case_ at expense of losing contents of printk_safe buffers.

Note: It would actually be safe to re-init the locks when all CPUs were
stopped by NMI. But it would require passing this information from arch-specific code. It is not worth the complexity. Especially because logbuf_lock and printk_safe buffers have been obsoleted by the lockless ring buffer.

index a0e6f746de6c..2e9e3ed7d63e 100644
@@ -45,6 +45,8 @@ struct printk_safe_seq_buf {
 static DEFINE_PER_CPU(struct printk_safe_seq_buf, safe_print_seq);
 static DEFINE_PER_CPU(int, printk_context);
 
+static DEFINE_RAW_SPINLOCK(safe_read_lock);
+
 #ifdef CONFIG_PRINTK_NMI
 static DEFINE_PER_CPU(struct printk_safe_seq_buf, nmi_print_seq);
 #endif

@@ -180,8 +182,6 @@ static void report_message_lost(struct printk_safe_seq_buf *s)
  */
 static void __printk_safe_flush(struct irq_work *work)
 {
-	static raw_spinlock_t read_lock =
-		__RAW_SPIN_LOCK_INITIALIZER(read_lock);
 	struct printk_safe_seq_buf *s =
 		container_of(work, struct printk_safe_seq_buf, work);
 	unsigned long flags;

@@ -195,7 +195,7 @@ static void __printk_safe_flush(struct irq_work *work)
 	 * different CPUs. This is especially important when printing
 	 * a backtrace.
 	 */
-	raw_spin_lock_irqsave(&read_lock, flags);
+	raw_spin_lock_irqsave(&safe_read_lock, flags);
 
 	i = 0;
 more:

@@ -232,7 +232,7 @@ static void __printk_safe_flush(struct irq_work *work)
 
 out:
 	report_message_lost(s);
-	raw_spin_unlock_irqrestore(&read_lock, flags);
+	raw_spin_unlock_irqrestore(&safe_read_lock, flags);
 }
 
 /**

@@ -278,6 +278,14 @@ void printk_safe_flush_on_panic(void)
 		raw_spin_lock_init(&logbuf_lock);
 	}
 
+	if (raw_spin_is_locked(&safe_read_lock)) {
+		if (num_online_cpus() > 1)
+			return;
+
+		debug_locks_off();
+		raw_spin_lock_init(&safe_read_lock);
+	}
+
 	printk_safe_flush();
 }
 

Please, find below the backport for stable-4.9:

From bfc246e5ed94300431bc8d74b67c74b6f66e75ff Mon Sep 17 00:00:00 2001
From: Muchun Song <[email protected]>
Date: Wed, 10 Feb 2021 11:48:23 +0800
Subject: [PATCH] printk: fix deadlock when kernel panic

printk_nmi_flush_on_panic() caused the following deadlock on our server:

CPU0: CPU1:
panic rcu_dump_cpu_stacks
kdump_nmi_shootdown_cpus nmi_trigger_cpumask_backtrace

    register_nmi_handler(crash_nmi_callback)      printk_nmi_flush
                                                    __printk_nmi_flush
                                                      raw_spin_lock_irqsave(&read_lock)
    // send NMI to other processors
    apic_send_IPI_allbutself(NMI_VECTOR)
                                                        // NMI interrupt, dead loop
                                                        crash_nmi_callback
  printk_nmi_flush_on_panic
    printk_nmi_flush
      __printk_nmi_flush
        // deadlock
        raw_spin_lock_irqsave(&read_lock)

DEADLOCK: read_lock is taken on CPU1 and will never get released.

It happens when panic() stops a CPU by NMI while it has been in the middle of printk_nmi_flush().

Handle the lock the same way as logbuf_lock. The printk_nmi buffers are flushed only when both locks can be safely taken. It can avoid the deadlock _in this particular case_ at expense of losing contents of printk_nmi buffers.

Note: It would actually be safe to re-init the locks when all CPUs were
stopped by NMI. But it would require passing this information from arch-specific code. It is not worth the complexity. Especially because logbuf_lock and printk_nmi buffers have been obsoleted by the lockless ring buffer.

This is a backport of the commit 8a8109f303e25a27f92c ("printk: fix deadlock when kernel panic") for the stable 4.9 kernel.

---
 kernel/printk/nmi.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

index 2c3e7f024c15..7a50b405ad28 100644
@@ -52,6 +52,8 @@ struct nmi_seq_buf {
 };
 static DEFINE_PER_CPU(struct nmi_seq_buf, nmi_print_seq);
 
+static DEFINE_RAW_SPINLOCK(nmi_read_lock);
+
 /*
  * Safe printk() for NMI context. It uses a per-CPU buffer to
  * store the message. NMIs are not nested, so there is always only

@@ -134,8 +136,6 @@ static void printk_nmi_flush_seq_line(struct nmi_seq_buf *s,
  */
 static void __printk_nmi_flush(struct irq_work *work)
 {
-	static raw_spinlock_t read_lock =
-		__RAW_SPIN_LOCK_INITIALIZER(read_lock);
 	struct nmi_seq_buf *s = container_of(work, struct nmi_seq_buf, work);
 	unsigned long flags;
 	size_t len, size;

@@ -148,7 +148,7 @@ static void __printk_nmi_flush(struct irq_work *work)
 	 * different CPUs. This is especially important when printing
 	 * a backtrace.
 	 */
-	raw_spin_lock_irqsave(&read_lock, flags);
+	raw_spin_lock_irqsave(&nmi_read_lock, flags);
 
 	i = 0;
 more:

@@ -197,7 +197,7 @@ static void __printk_nmi_flush(struct irq_work *work)
 		goto more;
 
 out:
-	raw_spin_unlock_irqrestore(&read_lock, flags);
+	raw_spin_unlock_irqrestore(&nmi_read_lock, flags);
 }
 
 /**

@@ -239,6 +239,14 @@ void printk_nmi_flush_on_panic(void)
 		raw_spin_lock_init(&logbuf_lock);
 	}
 
+	if (in_nmi() && raw_spin_is_locked(&nmi_read_lock)) {
+		if (num_online_cpus() > 1)
+			return;
+
+		debug_locks_off();
+		raw_spin_lock_init(&nmi_read_lock);
+	}
+
 	printk_nmi_flush();
 }
 

2.16.4

Now queued up, thanks.

greg k-h