linux-kernel - [PATCH AUTOSEL 6.10 08/27] rtnetlink: move rtnl_lock handling out of af

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: <20240728005329.1723272-8-sashal@kernel.org>
Date: Sat, 27 Jul 2024 20:52:51 -0400
From: Sasha Levin <sashal@...nel.org>
To: linux-kernel@...r.kernel.org,
	stable@...r.kernel.org
Cc: Jakub Kicinski <kuba@...nel.org>,
	Kuniyuki Iwashima <kuniyu@...zon.com>,
	Eric Dumazet <edumazet@...gle.com>,
	"David S . Miller" <davem@...emloft.net>,
	Sasha Levin <sashal@...nel.org>,
	pabeni@...hat.com,
	idosch@...dia.com,
	jiri@...nulli.us,
	amcohen@...dia.com,
	horms@...nel.org,
	lirongqing@...du.com,
	juntong.deng@...look.com,
	netdev@...r.kernel.org
Subject: [PATCH AUTOSEL 6.10 08/27] rtnetlink: move rtnl_lock handling out of af_netlink

From: Jakub Kicinski <kuba@...nel.org>

[ Upstream commit 5380d64f8d766576ac5c0f627418b2d0e1d2641f ]

Now that we have an intermediate layer of code for handling
rtnl-level netlink dump quirks, we can move the rtnl_lock
taking there.

For dump handlers with RTNL_FLAG_DUMP_SPLIT_NLM_DONE we can
avoid taking rtnl_lock just to generate NLM_DONE, once again.

Signed-off-by: Jakub Kicinski <kuba@...nel.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@...zon.com>
Reviewed-by: Eric Dumazet <edumazet@...gle.com>
Signed-off-by: David S. Miller <davem@...emloft.net>
Signed-off-by: Sasha Levin <sashal@...nel.org>
---
 net/core/rtnetlink.c     | 9 +++++++--
 net/netlink/af_netlink.c | 2 --
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 4668d67180407..eabfc8290f5e2 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -6486,6 +6486,7 @@ static int rtnl_mdb_del(struct sk_buff *skb, struct nlmsghdr *nlh,
 
 static int rtnl_dumpit(struct sk_buff *skb, struct netlink_callback *cb)
 {
+	const bool needs_lock = !(cb->flags & RTNL_FLAG_DUMP_UNLOCKED);
 	rtnl_dumpit_func dumpit = cb->data;
 	int err;
 
@@ -6495,7 +6496,11 @@ static int rtnl_dumpit(struct sk_buff *skb, struct netlink_callback *cb)
 	if (!dumpit)
 		return 0;
 
+	if (needs_lock)
+		rtnl_lock();
 	err = dumpit(skb, cb);
+	if (needs_lock)
+		rtnl_unlock();
 
 	/* Old dump handlers used to send NLM_DONE as in a separate recvmsg().
 	 * Some applications which parse netlink manually depend on this.
@@ -6515,7 +6520,8 @@ static int rtnetlink_dump_start(struct sock *ssk, struct sk_buff *skb,
 				const struct nlmsghdr *nlh,
 				struct netlink_dump_control *control)
 {
-	if (control->flags & RTNL_FLAG_DUMP_SPLIT_NLM_DONE) {
+	if (control->flags & RTNL_FLAG_DUMP_SPLIT_NLM_DONE ||
+	    !(control->flags & RTNL_FLAG_DUMP_UNLOCKED)) {
 		WARN_ON(control->data);
 		control->data = control->dump;
 		control->dump = rtnl_dumpit;
@@ -6703,7 +6709,6 @@ static int __net_init rtnetlink_net_init(struct net *net)
 	struct netlink_kernel_cfg cfg = {
 		.groups		= RTNLGRP_MAX,
 		.input		= rtnetlink_rcv,
-		.cb_mutex	= &rtnl_mutex,
 		.flags		= NL_CFG_F_NONROOT_RECV,
 		.bind		= rtnetlink_bind,
 	};
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index fa9c090cf629e..8bbbe75e75dbe 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -2330,8 +2330,6 @@ static int netlink_dump(struct sock *sk, bool lock_taken)
 
 		cb->extack = &extack;
 
-		if (cb->flags & RTNL_FLAG_DUMP_UNLOCKED)
-			extra_mutex = NULL;
 		if (extra_mutex)
 			mutex_lock(extra_mutex);
 		nlk->dump_done_errno = cb->dump(skb, cb);
-- 
2.43.0