netdev - Re: [PATCH net-next 1/4] udp: annotate data-race in __udp_enqueue_schedule

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <660602dbd4e85_2c31072943f@willemb.c.googlers.com.notmuch>
Date: Thu, 28 Mar 2024 19:52:59 -0400
From: Willem de Bruijn <willemdebruijn.kernel@...il.com>
To: Eric Dumazet <edumazet@...gle.com>, 
 "David S . Miller" <davem@...emloft.net>, 
 Jakub Kicinski <kuba@...nel.org>, 
 Paolo Abeni <pabeni@...hat.com>
Cc: Willem de Bruijn <willemb@...gle.com>, 
 netdev@...r.kernel.org, 
 eric.dumazet@...il.com, 
 Eric Dumazet <edumazet@...gle.com>
Subject: Re: [PATCH net-next 1/4] udp: annotate data-race in
 __udp_enqueue_schedule_skb()

Eric Dumazet wrote:
> sk->sk_rcvbuf is read locklessly twice, while other threads
> could change its value.
> 
> Use a READ_ONCE() to annotate the race.
> 
> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
> ---
>  net/ipv4/udp.c | 11 ++++++-----
>  1 file changed, 6 insertions(+), 5 deletions(-)
> 
> diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
> index 661d0e0d273f616ad82746b69b2c76d056633017..f2736e8958187e132ef45d8e25ab2b4ea7bcbc3d 100644
> --- a/net/ipv4/udp.c
> +++ b/net/ipv4/udp.c
> @@ -1492,13 +1492,14 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
>  	struct sk_buff_head *list = &sk->sk_receive_queue;
>  	int rmem, err = -ENOMEM;
>  	spinlock_t *busy = NULL;
> -	int size;
> +	int size, rcvbuf;
>  
> -	/* try to avoid the costly atomic add/sub pair when the receive
> -	 * queue is full; always allow at least a packet
> +	/* Immediately drop when the receive queue is full.
> +	 * Always allow at least one packet.
>  	 */
>  	rmem = atomic_read(&sk->sk_rmem_alloc);
> -	if (rmem > sk->sk_rcvbuf)
> +	rcvbuf = READ_ONCE(sk->sk_rcvbuf);
> +	if (rmem > rcvbuf)
>  		goto drop;
>  
>  	/* Under mem pressure, it might be helpful to help udp_recvmsg()
> @@ -1507,7 +1508,7 @@ int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
>  	 * - Less cache line misses at copyout() time
>  	 * - Less work at consume_skb() (less alien page frag freeing)
>  	 */
> -	if (rmem > (sk->sk_rcvbuf >> 1)) {
> +	if (rmem > (rcvbuf >> 1)) {
>  		skb_condense(skb);
>  
>  		busy = busylock_acquire(sk);

There's a third read in this function:

        /* we drop only if the receive buf is full and the receive
         * queue contains some other skb
         */
        rmem = atomic_add_return(size, &sk->sk_rmem_alloc);
        if (rmem > (size + (unsigned int)sk->sk_rcvbuf))
                goto uncharge_drop;

Another READ_ONCE if intent is to not use the locally cached copy?