Re: [PATCH] mm/kmemleak: avoid deadlock by moving pr_warn() outside kmemleak_lock

[Catalin Marinas <catalin.marinas@arm.com>]

On Thu, Jul 31, 2025 at 02:57:18AM -0700, Breno Leitao wrote:
> When netpoll is enabled, calling pr_warn_once() while holding
> kmemleak_lock in mem_pool_alloc() can cause a deadlock due to lock
> inversion with the netconsole subsystem. This occurs because
> pr_warn_once() may trigger netpoll, which eventually leads to
> __alloc_skb() and back into kmemleak code, attempting to reacquire
> kmemleak_lock.
> 
> This is the path for the deadlock.
> 
> mem_pool_alloc()
>   -> raw_spin_lock_irqsave(&kmemleak_lock, flags);
>       -> pr_warn_once()
>           -> netconsole subsystem
> 	     -> netpoll
> 	         -> __alloc_skb
> 		   -> __create_object
> 		     -> raw_spin_lock_irqsave(&kmemleak_lock, flags);
> 
> Fix this by setting a flag and issuing the pr_warn_once() after
> kmemleak_lock is released.
> 
> Reported-by: Jakub Kicinski <kuba@kernel.org>
> Fixes: c5665868183fec ("mm: kmemleak: use the memory pool for early allocations")
> Signed-off-by: Breno Leitao <leitao@debian.org>

I think Andrew already added this to mm-stable but, for the record:

Acked-by: Catalin Marinas <catalin.marinas@arm.com>