[PATCH net 4/7] netfilter: nft_set_pipapo: walk over current view on netlink dump

From: Pablo Neira Ayuso <pablo@netfilter.org>
To: netfilter-devel@vger.kernel.org
Cc: davem@davemloft.net, netdev@vger.kernel.org, kuba@kernel.org,
	pabeni@redhat.com, edumazet@google.com, fw@strlen.de
Subject: [PATCH net 4/7] netfilter: nft_set_pipapo: walk over current view on netlink dump
Date: Thu, 11 Apr 2024 13:28:57 +0200	[thread overview]
Message-ID: <20240411112900.129414-5-pablo@netfilter.org> (raw)
In-Reply-To: <20240411112900.129414-1-pablo@netfilter.org>

The generation mask can be updated while netlink dump is in progress.
The pipapo set backend walk iterator cannot rely on it to infer what
view of the datastructure is to be used. Add notation to specify if user
wants to read/update the set.

Based on patch from Florian Westphal.

Fixes: 2b84e215f874 ("netfilter: nft_set_pipapo: .walk does not deal with generations")
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 include/net/netfilter/nf_tables.h | 14 ++++++++++++++
 net/netfilter/nf_tables_api.c     |  6 ++++++
 net/netfilter/nft_set_pipapo.c    |  5 +++--
 3 files changed, 23 insertions(+), 2 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index e27c28b612e4..3f1ed467f951 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -307,9 +307,23 @@ static inline void *nft_elem_priv_cast(const struct nft_elem_priv *priv)
 	return (void *)priv;
 }
 
+
+/**
+ * enum nft_iter_type - nftables set iterator type
+ *
+ * @NFT_ITER_READ: read-only iteration over set elements
+ * @NFT_ITER_UPDATE: iteration under mutex to update set element state
+ */
+enum nft_iter_type {
+	NFT_ITER_UNSPEC,
+	NFT_ITER_READ,
+	NFT_ITER_UPDATE,
+};
+
 struct nft_set;
 struct nft_set_iter {
 	u8		genmask;
+	enum nft_iter_type type:8;
 	unsigned int	count;
 	unsigned int	skip;
 	int		err;
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index f11d0c0a2c73..a7a34db62ea9 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -626,6 +626,7 @@ static void nft_map_deactivate(const struct nft_ctx *ctx, struct nft_set *set)
 {
 	struct nft_set_iter iter = {
 		.genmask	= nft_genmask_next(ctx->net),
+		.type		= NFT_ITER_UPDATE,
 		.fn		= nft_mapelem_deactivate,
 	};
 
@@ -5445,6 +5446,7 @@ int nf_tables_bind_set(const struct nft_ctx *ctx, struct nft_set *set,
 		}
 
 		iter.genmask	= nft_genmask_next(ctx->net);
+		iter.type	= NFT_ITER_UPDATE;
 		iter.skip 	= 0;
 		iter.count	= 0;
 		iter.err	= 0;
@@ -5518,6 +5520,7 @@ static void nft_map_activate(const struct nft_ctx *ctx, struct nft_set *set)
 {
 	struct nft_set_iter iter = {
 		.genmask	= nft_genmask_next(ctx->net),
+		.type		= NFT_ITER_UPDATE,
 		.fn		= nft_mapelem_activate,
 	};
 
@@ -5892,6 +5895,7 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb)
 	args.skb		= skb;
 	args.reset		= dump_ctx->reset;
 	args.iter.genmask	= nft_genmask_cur(net);
+	args.iter.type		= NFT_ITER_READ;
 	args.iter.skip		= cb->args[0];
 	args.iter.count		= 0;
 	args.iter.err		= 0;
@@ -7376,6 +7380,7 @@ static int nft_set_flush(struct nft_ctx *ctx, struct nft_set *set, u8 genmask)
 {
 	struct nft_set_iter iter = {
 		.genmask	= genmask,
+		.type		= NFT_ITER_UPDATE,
 		.fn		= nft_setelem_flush,
 	};
 
@@ -10879,6 +10884,7 @@ static int nf_tables_check_loops(const struct nft_ctx *ctx,
 				continue;
 
 			iter.genmask	= nft_genmask_next(ctx->net);
+			iter.type	= NFT_ITER_UPDATE;
 			iter.skip 	= 0;
 			iter.count	= 0;
 			iter.err	= 0;
diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c
index df8de5090246..11e44e4dfb1f 100644
--- a/net/netfilter/nft_set_pipapo.c
+++ b/net/netfilter/nft_set_pipapo.c
@@ -2115,13 +2115,14 @@ static void nft_pipapo_walk(const struct nft_ctx *ctx, struct nft_set *set,
 			    struct nft_set_iter *iter)
 {
 	struct nft_pipapo *priv = nft_set_priv(set);
-	struct net *net = read_pnet(&set->net);
 	const struct nft_pipapo_match *m;
 	const struct nft_pipapo_field *f;
 	unsigned int i, r;
 
+	WARN_ON_ONCE(iter->type == NFT_ITER_UNSPEC);
+
 	rcu_read_lock();
-	if (iter->genmask == nft_genmask_cur(net))
+	if (iter->type == NFT_ITER_READ)
 		m = rcu_dereference(priv->match);
 	else
 		m = priv->clone;
-- 
2.30.2

