[PATCH net v2] netpoll: Fix deadlock in memory allocation under spinlock

[PATCH net v2] netpoll: Fix deadlock in memory allocation under spinlock

[Breno Leitao]

Fix a AA deadlock in refill_skbs() where memory allocation while holding
skb_pool->lock can trigger a recursive lock acquisition attempt.

The deadlock scenario occurs when the system is under severe memory
pressure:

1. refill_skbs() acquires skb_pool->lock (spinlock)
2. alloc_skb() is called while holding the lock
3. Memory allocator fails and calls slab_out_of_memory()
4. This triggers printk() for the OOM warning
5. The console output path calls netpoll_send_udp()
6. netpoll_send_udp() attempts to acquire the same skb_pool->lock
7. Deadlock: the lock is already held by the same CPU

Call stack:
  refill_skbs()
    spin_lock_irqsave(&skb_pool->lock)    <- lock acquired
    __alloc_skb()
      kmem_cache_alloc_node_noprof()
        slab_out_of_memory()
          printk()
            console_flush_all()
              netpoll_send_udp()
                skb_dequeue()
                  spin_lock_irqsave(&skb_pool->lock)     <- deadlock attempt

This bug was exposed by commit 248f6571fd4c51 ("netpoll: Optimize skb
refilling on critical path") which removed refill_skbs() from the
critical path (where nested printk was being deferred), letting nested
printk being calld form inside refill_skbs()

Refactor refill_skbs() to never allocate memory while holding
the spinlock.

Another possible solution to fix this problem is protecting the
refill_skbs() from nested printks, basically calling
printk_deferred_{enter,exit}() in refill_skbs(), then, any nested
pr_warn() would be deferred.

I prefer tthis approach, given I _think_ it might be a good idea to move
the alloc_skb() from GFP_ATOMIC to GFP_KERNEL in the future, so, having
the alloc_skb() outside of the lock will be necessary step.

Signed-off-by: Breno Leitao <leitao@debian.org>
Fixes: 248f6571fd4c51 ("netpoll: Optimize skb refilling on critical path")
---
Changes in v2:
- Added a return after the successful path (Rik van Riel)
- Changed the Fixes tag to point to the commit that exposed the problem.
- Link to v1: https://lore.kernel.org/r/20251013-fix_netpoll_aa-v1-1-94a1091f92f0@debian.org
---
 net/core/netpoll.c | 20 +++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)

diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 60a05d3b7c249..c19dada9283ce 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -232,14 +232,28 @@ static void refill_skbs(struct netpoll *np)
 
 	skb_pool = &np->skb_pool;
 
-	spin_lock_irqsave(&skb_pool->lock, flags);
-	while (skb_pool->qlen < MAX_SKBS) {
+	while (1) {
+		spin_lock_irqsave(&skb_pool->lock, flags);
+		if (skb_pool->qlen >= MAX_SKBS)
+			goto unlock;
+		spin_unlock_irqrestore(&skb_pool->lock, flags);
+
 		skb = alloc_skb(MAX_SKB_SIZE, GFP_ATOMIC);
 		if (!skb)
-			break;
+			return;
 
+		spin_lock_irqsave(&skb_pool->lock, flags);
+		if (skb_pool->qlen >= MAX_SKBS)
+			/* Discard if len got increased (TOCTOU) */
+			goto discard;
 		__skb_queue_tail(skb_pool, skb);
+		spin_unlock_irqrestore(&skb_pool->lock, flags);
 	}
+
+	return;
+discard:
+	dev_kfree_skb_any(skb);
+unlock:
 	spin_unlock_irqrestore(&skb_pool->lock, flags);
 }
 

---
base-commit: c5705a2a4aa35350e504b72a94b5c71c3754833c
change-id: 20251013-fix_netpoll_aa-c991ac5f2138

Best regards,
--  
Breno Leitao <leitao@debian.org>

Re: [PATCH net v2] netpoll: Fix deadlock in memory allocation under spinlock

[Simon Horman]

On Tue, Oct 14, 2025 at 03:10:51AM -0700, Breno Leitao wrote:
> Fix a AA deadlock in refill_skbs() where memory allocation while holding
> skb_pool->lock can trigger a recursive lock acquisition attempt.
> 
> The deadlock scenario occurs when the system is under severe memory
> pressure:
> 
> 1. refill_skbs() acquires skb_pool->lock (spinlock)
> 2. alloc_skb() is called while holding the lock
> 3. Memory allocator fails and calls slab_out_of_memory()
> 4. This triggers printk() for the OOM warning
> 5. The console output path calls netpoll_send_udp()
> 6. netpoll_send_udp() attempts to acquire the same skb_pool->lock
> 7. Deadlock: the lock is already held by the same CPU
> 
> Call stack:
>   refill_skbs()
>     spin_lock_irqsave(&skb_pool->lock)    <- lock acquired
>     __alloc_skb()
>       kmem_cache_alloc_node_noprof()
>         slab_out_of_memory()
>           printk()
>             console_flush_all()
>               netpoll_send_udp()
>                 skb_dequeue()
>                   spin_lock_irqsave(&skb_pool->lock)     <- deadlock attempt
> 
> This bug was exposed by commit 248f6571fd4c51 ("netpoll: Optimize skb
> refilling on critical path") which removed refill_skbs() from the
> critical path (where nested printk was being deferred), letting nested
> printk being calld form inside refill_skbs()
> 
> Refactor refill_skbs() to never allocate memory while holding
> the spinlock.
> 
> Another possible solution to fix this problem is protecting the
> refill_skbs() from nested printks, basically calling
> printk_deferred_{enter,exit}() in refill_skbs(), then, any nested
> pr_warn() would be deferred.
> 
> I prefer tthis approach, given I _think_ it might be a good idea to move
> the alloc_skb() from GFP_ATOMIC to GFP_KERNEL in the future, so, having
> the alloc_skb() outside of the lock will be necessary step.
> 
> Signed-off-by: Breno Leitao <leitao@debian.org>
> Fixes: 248f6571fd4c51 ("netpoll: Optimize skb refilling on critical path")
> ---
> Changes in v2:
> - Added a return after the successful path (Rik van Riel)
> - Changed the Fixes tag to point to the commit that exposed the problem.
> - Link to v1: https://lore.kernel.org/r/20251013-fix_netpoll_aa-v1-1-94a1091f92f0@debian.org
> ---
>  net/core/netpoll.c | 20 +++++++++++++++++---
>  1 file changed, 17 insertions(+), 3 deletions(-)
> 
> diff --git a/net/core/netpoll.c b/net/core/netpoll.c
> index 60a05d3b7c249..c19dada9283ce 100644
> --- a/net/core/netpoll.c
> +++ b/net/core/netpoll.c
> @@ -232,14 +232,28 @@ static void refill_skbs(struct netpoll *np)
>  
>  	skb_pool = &np->skb_pool;
>  
> -	spin_lock_irqsave(&skb_pool->lock, flags);
> -	while (skb_pool->qlen < MAX_SKBS) {
> +	while (1) {
> +		spin_lock_irqsave(&skb_pool->lock, flags);
> +		if (skb_pool->qlen >= MAX_SKBS)
> +			goto unlock;
> +		spin_unlock_irqrestore(&skb_pool->lock, flags);
> +
>  		skb = alloc_skb(MAX_SKB_SIZE, GFP_ATOMIC);
>  		if (!skb)
> -			break;
> +			return;
>  
> +		spin_lock_irqsave(&skb_pool->lock, flags);
> +		if (skb_pool->qlen >= MAX_SKBS)
> +			/* Discard if len got increased (TOCTOU) */
> +			goto discard;
>  		__skb_queue_tail(skb_pool, skb);
> +		spin_unlock_irqrestore(&skb_pool->lock, flags);
>  	}
> +
> +	return;

Maybe it is worth leaving alone for clarity.
And certainly it does no harm.
But the line above is never reached.

Flagged by Smatch.

> +discard:
> +	dev_kfree_skb_any(skb);
> +unlock:
>  	spin_unlock_irqrestore(&skb_pool->lock, flags);
>  }
>  
> 
> ---
> base-commit: c5705a2a4aa35350e504b72a94b5c71c3754833c
> change-id: 20251013-fix_netpoll_aa-c991ac5f2138
> 
> Best regards,
> --  
> Breno Leitao <leitao@debian.org>
> 

Re: [PATCH net v2] netpoll: Fix deadlock in memory allocation under spinlock

[Breno Leitao]

On Tue, Oct 14, 2025 at 03:11:51PM +0100, Simon Horman wrote:
> On Tue, Oct 14, 2025 at 03:10:51AM -0700, Breno Leitao wrote:
> > Fix a AA deadlock in refill_skbs() where memory allocation while holding
> > skb_pool->lock can trigger a recursive lock acquisition attempt.
> > 
> > The deadlock scenario occurs when the system is under severe memory
> > pressure:
> > 
> > 1. refill_skbs() acquires skb_pool->lock (spinlock)
> > 2. alloc_skb() is called while holding the lock
> > 3. Memory allocator fails and calls slab_out_of_memory()
> > 4. This triggers printk() for the OOM warning
> > 5. The console output path calls netpoll_send_udp()
> > 6. netpoll_send_udp() attempts to acquire the same skb_pool->lock
> > 7. Deadlock: the lock is already held by the same CPU
> > 
> > Call stack:
> >   refill_skbs()
> >     spin_lock_irqsave(&skb_pool->lock)    <- lock acquired
> >     __alloc_skb()
> >       kmem_cache_alloc_node_noprof()
> >         slab_out_of_memory()
> >           printk()
> >             console_flush_all()
> >               netpoll_send_udp()
> >                 skb_dequeue()
> >                   spin_lock_irqsave(&skb_pool->lock)     <- deadlock attempt
> > 
> > This bug was exposed by commit 248f6571fd4c51 ("netpoll: Optimize skb
> > refilling on critical path") which removed refill_skbs() from the
> > critical path (where nested printk was being deferred), letting nested
> > printk being calld form inside refill_skbs()
> > 
> > Refactor refill_skbs() to never allocate memory while holding
> > the spinlock.
> > 
> > Another possible solution to fix this problem is protecting the
> > refill_skbs() from nested printks, basically calling
> > printk_deferred_{enter,exit}() in refill_skbs(), then, any nested
> > pr_warn() would be deferred.
> > 
> > I prefer tthis approach, given I _think_ it might be a good idea to move
> > the alloc_skb() from GFP_ATOMIC to GFP_KERNEL in the future, so, having
> > the alloc_skb() outside of the lock will be necessary step.
> > 
> > Signed-off-by: Breno Leitao <leitao@debian.org>
> > Fixes: 248f6571fd4c51 ("netpoll: Optimize skb refilling on critical path")
> > ---
> > Changes in v2:
> > - Added a return after the successful path (Rik van Riel)
> > - Changed the Fixes tag to point to the commit that exposed the problem.
> > - Link to v1: https://lore.kernel.org/r/20251013-fix_netpoll_aa-v1-1-94a1091f92f0@debian.org
> > ---
> >  net/core/netpoll.c | 20 +++++++++++++++++---
> >  1 file changed, 17 insertions(+), 3 deletions(-)
> > 
> > diff --git a/net/core/netpoll.c b/net/core/netpoll.c
> > index 60a05d3b7c249..c19dada9283ce 100644
> > --- a/net/core/netpoll.c
> > +++ b/net/core/netpoll.c
> > @@ -232,14 +232,28 @@ static void refill_skbs(struct netpoll *np)
> >  
> >  	skb_pool = &np->skb_pool;
> >  
> > -	spin_lock_irqsave(&skb_pool->lock, flags);
> > -	while (skb_pool->qlen < MAX_SKBS) {
> > +	while (1) {
> > +		spin_lock_irqsave(&skb_pool->lock, flags);
> > +		if (skb_pool->qlen >= MAX_SKBS)
> > +			goto unlock;
> > +		spin_unlock_irqrestore(&skb_pool->lock, flags);
> > +
> >  		skb = alloc_skb(MAX_SKB_SIZE, GFP_ATOMIC);
> >  		if (!skb)
> > -			break;
> > +			return;
> >  
> > +		spin_lock_irqsave(&skb_pool->lock, flags);
> > +		if (skb_pool->qlen >= MAX_SKBS)
> > +			/* Discard if len got increased (TOCTOU) */
> > +			goto discard;
> >  		__skb_queue_tail(skb_pool, skb);
> > +		spin_unlock_irqrestore(&skb_pool->lock, flags);
> >  	}
> > +
> > +	return;
> 
> Maybe it is worth leaving alone for clarity.
> And certainly it does no harm.
> But the line above is never reached.

Thanks for the catch. I will remove it since it is useless and respin.