From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 272473C1D62 for ; Fri, 9 Oct 2026 07:49:12 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1791532154; cv=none; b=ajYq+NUwi9T/EwQ/BXlgA4JYPgYNQrSO8ncogUgeMsVf/wzPrmPKzo1wN3y06Dwd0JAbOP3F3okTkXVVP2yomfpQ20Tbyby22eSZe2z6IgNEsuX5kra1r5CTckbJykPPTMAEgyjT5jLCKn2KdrA/qibpUSBtaVtYD98W137rVDA= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1791532154; c=relaxed/simple; bh=LC+iT6cz6gt1xJGfnqi7RV6noX5MBVINAfFkxDUTe5A=; h=From:To:Cc:Subject:Date:Message-ID:MIME-Version; b=odKsJOl1k4+YoINZ8yz5GC8iJrtby8UcFNsQDUyAqdQwRWIPxNDDvu/Kst4SwJhDBk85w9g79r3cqX0zu1gGmFAMFtjeFqlHrQ1qi+iTGRVBtAWpr5JKQZZUAcJvGo0ebBDSRmPTzNkoTrFzmMKEOnU3d2hN1+6Q243+cQgnzZc= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=STM8SBJ2; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="STM8SBJ2" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 635A81F000FF; Fri, 9 Oct 2026 07:49:09 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1791532152; bh=OjPt+wkbBrxQpcTb2uuLv60lWItNEsjFUGl75FoNi2Q=; h=From:To:Cc:Subject:Date; b=STM8SBJ2of9UVOrpXQy07E1G2iwc0zAB+xMhMSsx8yDzz9nzJyaTa/TkwtfeQhYbf NOzNJM7wt4qalP5QhnwxgFeEr0ViveoDFzsy8kSPJvpJ5wINozsJ5KxPA+tILS7Cfs ozTExEdPqPWOkNuWRsp1j1Uj8s2q3olqgbjtS8a/eU3jKwBaPHv8HfEAkteDSOTDQg CvfmOh2SRz6zLCaab+6M2bDqP9YUty/1eoem8kYT895JHmwQSOVuSsEIUsoRPWFA9u swSLsfYGCm/XKi1lf72rZSXMFpnNdbCnBJ1QgzstczKhzLe85TwHoklxfWImRlnOa8 Acu23pad80Ifw== From: Eric Dumazet To: "David S . Miller" , Jakub Kicinski , Paolo Abeni Cc: Simon Horman , David Ahern , Ido Schimmel , Kuniyuki Iwashima , netdev@vger.kernel.org, Eric Dumazet , Baul Lee Subject: [PATCH net] ipv4: use RCU protection in inet_netconf_get_devconf() Date: Fri, 9 Oct 2026 09:49:05 +0200 Message-ID: <20261009074906.729877-1-edumazet@kernel.org> X-Mailer: git-send-email 2.53.0 Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit inet_netconf_get_devconf() runs without RTNL. It uses dev_get_by_index() and in_dev_get() to pin the device and its in_device while the reply is allocated and filled. in_dev_get() is not safe for a lockless caller. Since commit 9d40c84cf5bc ("net: devinet: Reduce refcount before grace period"), inetdev_destroy() drops the final reference before the RCU grace period. A reader can load dev->ip_ptr, then call refcount_inc() on a zero refcount after in_dev_free_rcu() has been queued, and keep using the in_device once it has been freed. None of these references are needed. An in_device is freed after an RCU grace period, ipv4_devconf is embedded in it, and inet_netconf_fill_devconf() does not sleep. Allocate the skb first, then perform the lookup and the fill under rcu_read_lock(), using dev_get_by_index_rcu() and __in_dev_get_rcu(), like inet_netconf_dump_devconf() already does. Fixes: bbcf91053bb6 ("inet: do not use RTNL in inet_netconf_get_devconf()") Reported-by: Baul Lee Closes: https://lore.kernel.org/netdev/20260815172032.79740-1-baul.lee@xbow.com/ Signed-off-by: Eric Dumazet --- Sorry for a resend, I did a mistake, using edumazet@google.com instead of edumazet@kernel.org, not enough sleep I guess. net/ipv4/devinet.c | 35 +++++++++++++++++------------------ 1 file changed, 17 insertions(+), 18 deletions(-) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 5b6b11c943e453742b4893f1c9bd9e70d3d37a6f..f9a361621f74b420276006b13de94926e13af18e 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -2379,8 +2379,8 @@ static int inet_netconf_get_devconf(struct sk_buff *in_skb, struct net *net = sock_net(in_skb->sk); struct nlattr *tb[NETCONFA_MAX + 1]; const struct ipv4_devconf *devconf; - struct in_device *in_dev = NULL; - struct net_device *dev = NULL; + struct in_device *in_dev; + struct net_device *dev; struct sk_buff *skb; int ifindex; int err; @@ -2392,7 +2392,13 @@ static int inet_netconf_get_devconf(struct sk_buff *in_skb, if (!tb[NETCONFA_IFINDEX]) return -EINVAL; + skb = nlmsg_new(inet_netconf_msgsize_devconf(NETCONFA_ALL), GFP_KERNEL); + if (!skb) + return -ENOBUFS; + ifindex = nla_get_s32(tb[NETCONFA_IFINDEX]); + + rcu_read_lock(); switch (ifindex) { case NETCONFA_IFINDEX_ALL: devconf = net->ipv4.devconf_all; @@ -2402,36 +2408,29 @@ static int inet_netconf_get_devconf(struct sk_buff *in_skb, break; default: err = -ENODEV; - dev = dev_get_by_index(net, ifindex); - if (dev) - in_dev = in_dev_get(dev); + dev = dev_get_by_index_rcu(net, ifindex); + if (!dev) + goto out_unlock; + in_dev = __in_dev_get_rcu(dev); if (!in_dev) - goto errout; + goto out_unlock; devconf = &in_dev->cnf; break; } - err = -ENOBUFS; - skb = nlmsg_new(inet_netconf_msgsize_devconf(NETCONFA_ALL), GFP_KERNEL); - if (!skb) - goto errout; - err = inet_netconf_fill_devconf(skb, ifindex, devconf, NETLINK_CB(in_skb).portid, nlh->nlmsg_seq, RTM_NEWNETCONF, 0, NETCONFA_ALL); +out_unlock: + rcu_read_unlock(); if (err < 0) { /* -EMSGSIZE implies BUG in inet_netconf_msgsize_devconf() */ WARN_ON(err == -EMSGSIZE); kfree_skb(skb); - goto errout; + return err; } - err = rtnl_unicast(skb, net, NETLINK_CB(in_skb).portid); -errout: - if (in_dev) - in_dev_put(in_dev); - dev_put(dev); - return err; + return rtnl_unicast(skb, net, NETLINK_CB(in_skb).portid); } static int inet_netconf_dump_devconf(struct sk_buff *skb, -- 2.53.0