[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <da189b9d-760f-0591-2acf-1a6b7910e227@gmail.com>
Date: Tue, 6 Jun 2017 18:34:47 -0600
From: David Ahern <dsahern@...il.com>
To: Eric Dumazet <eric.dumazet@...il.com>
Cc: Ben Greear <greearb@...delatech.com>,
netdev <netdev@...r.kernel.org>
Subject: Re: Repeatable inet6_dump_fib crash in stock 4.12.0-rc4+
On 6/6/17 6:27 PM, Eric Dumazet wrote:
> Good catch, but it looks like similar fix is needed a few lines before.
>
> diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
> index deea901746c8570c5e801e40592c91e3b62812e0..b214443dc8346cef3690df7f27cc48a864028865 100644
> --- a/net/ipv6/ip6_fib.c
> +++ b/net/ipv6/ip6_fib.c
> @@ -372,12 +372,13 @@ static int fib6_dump_table(struct fib6_table *table, struct sk_buff *skb,
>
> read_lock_bh(&table->tb6_lock);
> res = fib6_walk(net, w);
> - read_unlock_bh(&table->tb6_lock);
> if (res > 0) {
> cb->args[4] = 1;
> cb->args[5] = w->root->fn_sernum;
> }
> + read_unlock_bh(&table->tb6_lock);
indeed. tunnel vision on Ben's problem
Powered by blists - more mailing lists