[PATCH nf-next 0/9] netfilter updates for -next

[PATCH nf-next 0/9] netfilter updates for -next

[Florian Westphal]

Hello,

This batch contains updates for your *next* tree.

First three changes, from Phil Sutter, allow userspace to define
a table that is exclusively owned by a daemon (via netlink socket
aliveness) without auto-removing this table when the userspace program
exits.  Such table gets marked as orphaned and a restarting management
daemon may re-attach/reassume ownership.

Next patch, from Pablo, passes already-validated flags variable around
rather than having called code re-fetch it from netlnik message.

Patches 5 and 6 update ipvs and nf_conncount to use the recently
introduced KMEM_CACHE() macro.

Last three patches, from myself, tweak kconfig logic a little to
permit a kernel configuration that can run iptables-over-nftables
but not classic (setsockopt) iptables.

Such builds lack the builtin-filter/mangle/raw/nat/security tables,
the set/getsockopt interface and the "old blob format"
interpreter/traverser.  For now, this is 'oldconfig friendly', users
need to manually deselect existing config options for this.

The following changes since commit 723de3ebef03bc14bd72531f00f9094337654009:

  net: free altname using an RCU callback (2024-01-29 14:40:38 +0000)

are available in the Git repository at:

  https://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next.git tags/nf-next-24-01-29

for you to fetch changes up to 7ad269787b6615ca56bb161063331991fce51abf:

  netfilter: ebtables: allow xtables-nft only builds (2024-01-29 15:43:21 +0100)

----------------------------------------------------------------
nf-next pr 2024-01-29

----------------------------------------------------------------
Florian Westphal (3):
      netfilter: arptables: allow xtables-nft only builds
      netfilter: xtables: allow xtables-nft only builds
      netfilter: ebtables: allow xtables-nft only builds

Kunwu Chan (2):
      netfilter: nf_conncount: Use KMEM_CACHE instead of kmem_cache_create()
      ipvs: Simplify the allocation of ip_vs_conn slab caches

Pablo Neira Ayuso (1):
      netfilter: nf_tables: pass flags to set backend selection routine

Phil Sutter (3):
      netfilter: uapi: Document NFT_TABLE_F_OWNER flag
      netfilter: nf_tables: Introduce NFT_TABLE_F_PERSIST
      netfilter: nf_tables: Implement table adoption support

 include/net/netfilter/nf_tables.h        |  6 +++++
 include/uapi/linux/netfilter/nf_tables.h |  6 ++++-
 net/bridge/netfilter/Kconfig             |  7 ++++++
 net/bridge/netfilter/Makefile            |  2 +-
 net/ipv4/netfilter/Kconfig               | 43 +++++++++++++++++++-------------
 net/ipv4/netfilter/Makefile              |  2 +-
 net/ipv6/netfilter/Kconfig               | 20 ++++++++++-----
 net/ipv6/netfilter/Makefile              |  2 +-
 net/netfilter/Kconfig                    | 12 ++++-----
 net/netfilter/ipvs/ip_vs_conn.c          |  4 +--
 net/netfilter/nf_conncount.c             |  8 ++----
 net/netfilter/nf_tables_api.c            | 35 ++++++++++++++++++--------
 12 files changed, 94 insertions(+), 53 deletions(-)

[PATCH nf-next 1/9] netfilter: uapi: Document NFT_TABLE_F_OWNER flag

[Florian Westphal]

From: Phil Sutter <phil@nwl.cc>

Add at least this one-liner describing the obvious.

Fixes: 6001a930ce03 ("netfilter: nftables: introduce table ownership")
Signed-off-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 include/uapi/linux/netfilter/nf_tables.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index ca30232b7bc8..fbce238abdc1 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -179,6 +179,7 @@ enum nft_hook_attributes {
  * enum nft_table_flags - nf_tables table flags
  *
  * @NFT_TABLE_F_DORMANT: this table is not active
+ * @NFT_TABLE_F_OWNER:   this table is owned by a process
  */
 enum nft_table_flags {
 	NFT_TABLE_F_DORMANT	= 0x1,
-- 
2.43.0

Re: [PATCH nf-next 1/9] netfilter: uapi: Document NFT_TABLE_F_OWNER flag

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (main)
by Florian Westphal <fw@strlen.de>:

On Mon, 29 Jan 2024 15:57:51 +0100 you wrote:
> From: Phil Sutter <phil@nwl.cc>
> 
> Add at least this one-liner describing the obvious.
> 
> Fixes: 6001a930ce03 ("netfilter: nftables: introduce table ownership")
> Signed-off-by: Phil Sutter <phil@nwl.cc>
> Signed-off-by: Florian Westphal <fw@strlen.de>
> 
> [...]

Here is the summary with links:
  - [nf-next,1/9] netfilter: uapi: Document NFT_TABLE_F_OWNER flag
    https://git.kernel.org/netdev/net-next/c/941988af5724
  - [nf-next,2/9] netfilter: nf_tables: Introduce NFT_TABLE_F_PERSIST
    https://git.kernel.org/netdev/net-next/c/da5141bbe0c2
  - [nf-next,3/9] netfilter: nf_tables: Implement table adoption support
    https://git.kernel.org/netdev/net-next/c/31bf508be656
  - [nf-next,4/9] netfilter: nf_tables: pass flags to set backend selection routine
    https://git.kernel.org/netdev/net-next/c/a128885ace60
  - [nf-next,5/9] netfilter: nf_conncount: Use KMEM_CACHE instead of kmem_cache_create()
    https://git.kernel.org/netdev/net-next/c/2ae6e9a03dad
  - [nf-next,6/9] ipvs: Simplify the allocation of ip_vs_conn slab caches
    https://git.kernel.org/netdev/net-next/c/d5f9142fb96d
  - [nf-next,7/9] netfilter: arptables: allow xtables-nft only builds
    https://git.kernel.org/netdev/net-next/c/4654467dc7e1
  - [nf-next,8/9] netfilter: xtables: allow xtables-nft only builds
    https://git.kernel.org/netdev/net-next/c/a9525c7f6219
  - [nf-next,9/9] netfilter: ebtables: allow xtables-nft only builds
    https://git.kernel.org/netdev/net-next/c/7ad269787b66

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

[PATCH nf-next 2/9] netfilter: nf_tables: Introduce NFT_TABLE_F_PERSIST

[Florian Westphal]

From: Phil Sutter <phil@nwl.cc>

This companion flag to NFT_TABLE_F_OWNER requests the kernel to keep the
table around after the process has exited. It marks such table as
orphaned (by dropping OWNER flag but keeping PERSIST flag in place),
which opens it for other processes to manipulate. For the sake of
simplicity, PERSIST flag may not be altered though.

Signed-off-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 include/uapi/linux/netfilter/nf_tables.h | 5 ++++-
 net/netfilter/nf_tables_api.c            | 7 +++++++
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index fbce238abdc1..3fee994721cd 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -180,13 +180,16 @@ enum nft_hook_attributes {
  *
  * @NFT_TABLE_F_DORMANT: this table is not active
  * @NFT_TABLE_F_OWNER:   this table is owned by a process
+ * @NFT_TABLE_F_PERSIST: this table shall outlive its owner
  */
 enum nft_table_flags {
 	NFT_TABLE_F_DORMANT	= 0x1,
 	NFT_TABLE_F_OWNER	= 0x2,
+	NFT_TABLE_F_PERSIST	= 0x4,
 };
 #define NFT_TABLE_F_MASK	(NFT_TABLE_F_DORMANT | \
-				 NFT_TABLE_F_OWNER)
+				 NFT_TABLE_F_OWNER | \
+				 NFT_TABLE_F_PERSIST)
 
 /**
  * enum nft_table_attributes - nf_tables table netlink attributes
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index c537104411e7..6a96f0003faa 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -1219,6 +1219,9 @@ static int nf_tables_updtable(struct nft_ctx *ctx)
 	     flags & NFT_TABLE_F_OWNER))
 		return -EOPNOTSUPP;
 
+	if ((flags ^ ctx->table->flags) & NFT_TABLE_F_PERSIST)
+		return -EOPNOTSUPP;
+
 	/* No dormant off/on/off/on games in single transaction */
 	if (ctx->table->flags & __NFT_TABLE_F_UPDATE)
 		return -EINVAL;
@@ -11345,6 +11348,10 @@ static int nft_rcv_nl_event(struct notifier_block *this, unsigned long event,
 	list_for_each_entry(table, &nft_net->tables, list) {
 		if (nft_table_has_owner(table) &&
 		    n->portid == table->nlpid) {
+			if (table->flags & NFT_TABLE_F_PERSIST) {
+				table->flags &= ~NFT_TABLE_F_OWNER;
+				continue;
+			}
 			__nft_release_hook(net, table);
 			list_del_rcu(&table->list);
 			to_delete[deleted++] = table;
-- 
2.43.0

[PATCH nf-next 3/9] netfilter: nf_tables: Implement table adoption support

[Florian Westphal]

From: Phil Sutter <phil@nwl.cc>

Allow a new process to take ownership of a previously owned table,
useful mostly for firewall management services restarting or suspending
when idle.

By extending __NFT_TABLE_F_UPDATE, the on/off/on check in
nf_tables_updtable() also covers table adoption, although it is actually
not needed: Table adoption is irreversible because nf_tables_updtable()
rejects attempts to drop NFT_TABLE_F_OWNER so table->nlpid setting can
happen just once within the transaction.

If the transaction commences, table's nlpid and flags fields are already
set and no further action is required. If it aborts, the table returns
to orphaned state.

Signed-off-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 include/net/netfilter/nf_tables.h |  6 ++++++
 net/netfilter/nf_tables_api.c     | 19 ++++++++++++++++---
 2 files changed, 22 insertions(+), 3 deletions(-)

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 4e1ea18eb5f0..ac7c94d3648e 100644
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1271,6 +1271,12 @@ static inline bool nft_table_has_owner(const struct nft_table *table)
 	return table->flags & NFT_TABLE_F_OWNER;
 }
 
+static inline bool nft_table_is_orphan(const struct nft_table *table)
+{
+	return (table->flags & (NFT_TABLE_F_OWNER | NFT_TABLE_F_PERSIST)) ==
+			NFT_TABLE_F_PERSIST;
+}
+
 static inline bool nft_base_chain_netdev(int family, u32 hooknum)
 {
 	return family == NFPROTO_NETDEV ||
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 6a96f0003faa..b0e0d039897e 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -1194,8 +1194,10 @@ static void nf_tables_table_disable(struct net *net, struct nft_table *table)
 #define __NFT_TABLE_F_INTERNAL		(NFT_TABLE_F_MASK + 1)
 #define __NFT_TABLE_F_WAS_DORMANT	(__NFT_TABLE_F_INTERNAL << 0)
 #define __NFT_TABLE_F_WAS_AWAKEN	(__NFT_TABLE_F_INTERNAL << 1)
+#define __NFT_TABLE_F_WAS_ORPHAN	(__NFT_TABLE_F_INTERNAL << 2)
 #define __NFT_TABLE_F_UPDATE		(__NFT_TABLE_F_WAS_DORMANT | \
-					 __NFT_TABLE_F_WAS_AWAKEN)
+					 __NFT_TABLE_F_WAS_AWAKEN | \
+					 __NFT_TABLE_F_WAS_ORPHAN)
 
 static int nf_tables_updtable(struct nft_ctx *ctx)
 {
@@ -1215,8 +1217,8 @@ static int nf_tables_updtable(struct nft_ctx *ctx)
 
 	if ((nft_table_has_owner(ctx->table) &&
 	     !(flags & NFT_TABLE_F_OWNER)) ||
-	    (!nft_table_has_owner(ctx->table) &&
-	     flags & NFT_TABLE_F_OWNER))
+	    (flags & NFT_TABLE_F_OWNER &&
+	     !nft_table_is_orphan(ctx->table)))
 		return -EOPNOTSUPP;
 
 	if ((flags ^ ctx->table->flags) & NFT_TABLE_F_PERSIST)
@@ -1248,6 +1250,13 @@ static int nf_tables_updtable(struct nft_ctx *ctx)
 		}
 	}
 
+	if ((flags & NFT_TABLE_F_OWNER) &&
+	    !nft_table_has_owner(ctx->table)) {
+		ctx->table->nlpid = ctx->portid;
+		ctx->table->flags |= NFT_TABLE_F_OWNER |
+				     __NFT_TABLE_F_WAS_ORPHAN;
+	}
+
 	nft_trans_table_update(trans) = true;
 	nft_trans_commit_list_add_tail(ctx->net, trans);
 
@@ -10423,6 +10432,10 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 				} else if (trans->ctx.table->flags & __NFT_TABLE_F_WAS_AWAKEN) {
 					trans->ctx.table->flags &= ~NFT_TABLE_F_DORMANT;
 				}
+				if (trans->ctx.table->flags & __NFT_TABLE_F_WAS_ORPHAN) {
+					trans->ctx.table->flags &= ~NFT_TABLE_F_OWNER;
+					trans->ctx.table->nlpid = 0;
+				}
 				trans->ctx.table->flags &= ~__NFT_TABLE_F_UPDATE;
 				nft_trans_destroy(trans);
 			} else {
-- 
2.43.0

[PATCH nf-next 4/9] netfilter: nf_tables: pass flags to set backend selection routine

[Florian Westphal]

From: Pablo Neira Ayuso <pablo@netfilter.org>

No need to refetch the flag from the netlink attribute, pass the
existing flags variable which already provide validated flags.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/netfilter/nf_tables_api.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index b0e0d039897e..7f25a04e4b81 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -4247,23 +4247,18 @@ static bool nft_set_ops_candidate(const struct nft_set_type *type, u32 flags)
  * given, in that case the amount of memory per element is used.
  */
 static const struct nft_set_ops *
-nft_select_set_ops(const struct nft_ctx *ctx,
-		   const struct nlattr * const nla[],
+nft_select_set_ops(const struct nft_ctx *ctx, u32 flags,
 		   const struct nft_set_desc *desc)
 {
 	struct nftables_pernet *nft_net = nft_pernet(ctx->net);
 	const struct nft_set_ops *ops, *bops;
 	struct nft_set_estimate est, best;
 	const struct nft_set_type *type;
-	u32 flags = 0;
 	int i;
 
 	lockdep_assert_held(&nft_net->commit_mutex);
 	lockdep_nfnl_nft_mutex_not_held();
 
-	if (nla[NFTA_SET_FLAGS] != NULL)
-		flags = ntohl(nla_get_be32(nla[NFTA_SET_FLAGS]));
-
 	bops	    = NULL;
 	best.size   = ~0;
 	best.lookup = ~0;
@@ -5149,7 +5144,7 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info,
 	if (!(info->nlh->nlmsg_flags & NLM_F_CREATE))
 		return -ENOENT;
 
-	ops = nft_select_set_ops(&ctx, nla, &desc);
+	ops = nft_select_set_ops(&ctx, flags, &desc);
 	if (IS_ERR(ops))
 		return PTR_ERR(ops);
 
-- 
2.43.0

[PATCH nf-next 5/9] netfilter: nf_conncount: Use KMEM_CACHE instead of kmem_cache_create()

[Florian Westphal]

From: Kunwu Chan <chentao@kylinos.cn>

Use the new KMEM_CACHE() macro instead of direct kmem_cache_create
to simplify the creation of SLAB caches.

Signed-off-by: Kunwu Chan <chentao@kylinos.cn>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/netfilter/nf_conncount.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/net/netfilter/nf_conncount.c b/net/netfilter/nf_conncount.c
index 5d8ed6c90b7e..8715617b02fe 100644
--- a/net/netfilter/nf_conncount.c
+++ b/net/netfilter/nf_conncount.c
@@ -605,15 +605,11 @@ static int __init nf_conncount_modinit(void)
 	for (i = 0; i < CONNCOUNT_SLOTS; ++i)
 		spin_lock_init(&nf_conncount_locks[i]);
 
-	conncount_conn_cachep = kmem_cache_create("nf_conncount_tuple",
-					   sizeof(struct nf_conncount_tuple),
-					   0, 0, NULL);
+	conncount_conn_cachep = KMEM_CACHE(nf_conncount_tuple, 0);
 	if (!conncount_conn_cachep)
 		return -ENOMEM;
 
-	conncount_rb_cachep = kmem_cache_create("nf_conncount_rb",
-					   sizeof(struct nf_conncount_rb),
-					   0, 0, NULL);
+	conncount_rb_cachep = KMEM_CACHE(nf_conncount_rb, 0);
 	if (!conncount_rb_cachep) {
 		kmem_cache_destroy(conncount_conn_cachep);
 		return -ENOMEM;
-- 
2.43.0

[PATCH nf-next 6/9] ipvs: Simplify the allocation of ip_vs_conn slab caches

[Florian Westphal]

From: Kunwu Chan <chentao@kylinos.cn>

Use the new KMEM_CACHE() macro instead of direct kmem_cache_create
to simplify the creation of SLAB caches.

Signed-off-by: Kunwu Chan <chentao@kylinos.cn>
Acked-by: Simon Horman <horms@kernel.org>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/netfilter/ipvs/ip_vs_conn.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c
index a743db073887..98d7dbe3d787 100644
--- a/net/netfilter/ipvs/ip_vs_conn.c
+++ b/net/netfilter/ipvs/ip_vs_conn.c
@@ -1511,9 +1511,7 @@ int __init ip_vs_conn_init(void)
 		return -ENOMEM;
 
 	/* Allocate ip_vs_conn slab cache */
-	ip_vs_conn_cachep = kmem_cache_create("ip_vs_conn",
-					      sizeof(struct ip_vs_conn), 0,
-					      SLAB_HWCACHE_ALIGN, NULL);
+	ip_vs_conn_cachep = KMEM_CACHE(ip_vs_conn, SLAB_HWCACHE_ALIGN);
 	if (!ip_vs_conn_cachep) {
 		kvfree(ip_vs_conn_tab);
 		return -ENOMEM;
-- 
2.43.0

[PATCH nf-next 7/9] netfilter: arptables: allow xtables-nft only builds

[Florian Westphal]

Allows to build kernel that supports the arptables mangle target
via nftables' compat infra but without the arptables get/setsockopt
interface or the old arptables filter interpreter.

IOW, setting IP_NF_ARPFILTER=n will break arptables-legacy, but
arptables-nft will continue to work as long as nftables compat
support is enabled.

Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Phil Sutter <phil@nwl.cc>
---
 net/ipv4/netfilter/Kconfig | 28 +++++++++++++---------------
 1 file changed, 13 insertions(+), 15 deletions(-)

diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index f71a7e9a7de6..070475392236 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -309,36 +309,34 @@ endif # IP_NF_IPTABLES
 
 # ARP tables
 config IP_NF_ARPTABLES
-	tristate "ARP tables support"
-	select NETFILTER_XTABLES
-	select NETFILTER_FAMILY_ARP
-	depends on NETFILTER_ADVANCED
-	help
-	  arptables is a general, extensible packet identification framework.
-	  The ARP packet filtering and mangling (manipulation)subsystems
-	  use this: say Y or M here if you want to use either of those.
-
-	  To compile it as a module, choose M here.  If unsure, say N.
+	tristate
 
-if IP_NF_ARPTABLES
+config NFT_COMPAT_ARP
+	tristate
+	depends on NF_TABLES_ARP && NFT_COMPAT
+	default m if NFT_COMPAT=m
+	default y if NFT_COMPAT=y
 
 config IP_NF_ARPFILTER
-	tristate "ARP packet filtering"
+	tristate "arptables-legacy packet filtering support"
+	select IP_NF_ARPTABLES
 	help
 	  ARP packet filtering defines a table `filter', which has a series of
 	  rules for simple ARP packet filtering at local input and
-	  local output.  On a bridge, you can also specify filtering rules
-	  for forwarded ARP packets. See the man page for arptables(8).
+	  local output.  This is only needed for arptables-legacy(8).
+	  Neither arptables-nft nor nftables need this to work.
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
 config IP_NF_ARP_MANGLE
 	tristate "ARP payload mangling"
+	depends on IP_NF_ARPTABLES || NFT_COMPAT_ARP
 	help
 	  Allows altering the ARP packet payload: source and destination
 	  hardware and network addresses.
 
-endif # IP_NF_ARPTABLES
+	  This option is needed by both arptables-legacy and arptables-nft.
+	  It is not used by nftables.
 
 endmenu
 
-- 
2.43.0

[PATCH nf-next 8/9] netfilter: xtables: allow xtables-nft only builds

[Florian Westphal]

Add hidden IP(6)_NF_IPTABLES_LEGACY symbol.

When any of the "old" builtin tables are enabled the "old" iptables
interface will be supported.

To disable the old set/getsockopt interface the existing options
for the builtin tables need to be turned off:

CONFIG_IP_NF_IPTABLES=m
CONFIG_IP_NF_FILTER is not set
CONFIG_IP_NF_NAT is not set
CONFIG_IP_NF_MANGLE is not set
CONFIG_IP_NF_RAW is not set
CONFIG_IP_NF_SECURITY is not set

Same for CONFIG_IP6_NF_ variants.

This allows to build a kernel that only supports ip(6)tables-nft
(iptables-over-nftables api).

In the future the _LEGACY symbol will become visible and the select
statements will be turned into 'depends on', but for now be on safe side
so "make oldconfig" won't break things.

Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/ipv4/netfilter/Kconfig  | 15 ++++++++++++---
 net/ipv4/netfilter/Makefile |  2 +-
 net/ipv6/netfilter/Kconfig  | 20 ++++++++++++++------
 net/ipv6/netfilter/Makefile |  2 +-
 net/netfilter/Kconfig       | 12 ++++++------
 5 files changed, 34 insertions(+), 17 deletions(-)

diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index 070475392236..783523087281 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -10,6 +10,10 @@ config NF_DEFRAG_IPV4
 	tristate
 	default n
 
+# old sockopt interface and eval loop
+config IP_NF_IPTABLES_LEGACY
+	tristate
+
 config NF_SOCKET_IPV4
 	tristate "IPv4 socket lookup support"
 	help
@@ -152,7 +156,7 @@ config IP_NF_MATCH_ECN
 config IP_NF_MATCH_RPFILTER
 	tristate '"rpfilter" reverse path filter match support'
 	depends on NETFILTER_ADVANCED
-	depends on IP_NF_MANGLE || IP_NF_RAW
+	depends on IP_NF_MANGLE || IP_NF_RAW || NFT_COMPAT
 	help
 	  This option allows you to match packets whose replies would
 	  go out via the interface the packet came in.
@@ -173,6 +177,7 @@ config IP_NF_MATCH_TTL
 config IP_NF_FILTER
 	tristate "Packet filtering"
 	default m if NETFILTER_ADVANCED=n
+	select IP_NF_IPTABLES_LEGACY
 	help
 	  Packet filtering defines a table `filter', which has a series of
 	  rules for simple packet filtering at local input, forwarding and
@@ -182,7 +187,7 @@ config IP_NF_FILTER
 
 config IP_NF_TARGET_REJECT
 	tristate "REJECT target support"
-	depends on IP_NF_FILTER
+	depends on IP_NF_FILTER || NFT_COMPAT
 	select NF_REJECT_IPV4
 	default m if NETFILTER_ADVANCED=n
 	help
@@ -212,6 +217,7 @@ config IP_NF_NAT
 	default m if NETFILTER_ADVANCED=n
 	select NF_NAT
 	select NETFILTER_XT_NAT
+	select IP6_NF_IPTABLES_LEGACY
 	help
 	  This enables the `nat' table in iptables. This allows masquerading,
 	  port forwarding and other forms of full Network Address Port
@@ -252,6 +258,7 @@ endif # IP_NF_NAT
 config IP_NF_MANGLE
 	tristate "Packet mangling"
 	default m if NETFILTER_ADVANCED=n
+	select IP_NF_IPTABLES_LEGACY
 	help
 	  This option adds a `mangle' table to iptables: see the man page for
 	  iptables(8).  This table is used for various packet alterations
@@ -261,7 +268,7 @@ config IP_NF_MANGLE
 
 config IP_NF_TARGET_ECN
 	tristate "ECN target support"
-	depends on IP_NF_MANGLE
+	depends on IP_NF_MANGLE || NFT_COMPAT
 	depends on NETFILTER_ADVANCED
 	help
 	  This option adds a `ECN' target, which can be used in the iptables mangle
@@ -286,6 +293,7 @@ config IP_NF_TARGET_TTL
 # raw + specific targets
 config IP_NF_RAW
 	tristate  'raw table support (required for NOTRACK/TRACE)'
+	select IP_NF_IPTABLES_LEGACY
 	help
 	  This option adds a `raw' table to iptables. This table is the very
 	  first in the netfilter framework and hooks in at the PREROUTING
@@ -299,6 +307,7 @@ config IP_NF_SECURITY
 	tristate "Security table"
 	depends on SECURITY
 	depends on NETFILTER_ADVANCED
+	select IP_NF_IPTABLES_LEGACY
 	help
 	  This option adds a `security' table to iptables, for use
 	  with Mandatory Access Control (MAC) policy.
diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile
index 5a26f9de1ab9..85502d4dfbb4 100644
--- a/net/ipv4/netfilter/Makefile
+++ b/net/ipv4/netfilter/Makefile
@@ -25,7 +25,7 @@ obj-$(CONFIG_NFT_FIB_IPV4) += nft_fib_ipv4.o
 obj-$(CONFIG_NFT_DUP_IPV4) += nft_dup_ipv4.o
 
 # generic IP tables
-obj-$(CONFIG_IP_NF_IPTABLES) += ip_tables.o
+obj-$(CONFIG_IP_NF_IPTABLES_LEGACY) += ip_tables.o
 
 # the three instances of ip_tables
 obj-$(CONFIG_IP_NF_FILTER) += iptable_filter.o
diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig
index 0ba62f4868f9..f3c8e2d918e1 100644
--- a/net/ipv6/netfilter/Kconfig
+++ b/net/ipv6/netfilter/Kconfig
@@ -6,6 +6,10 @@
 menu "IPv6: Netfilter Configuration"
 	depends on INET && IPV6 && NETFILTER
 
+# old sockopt interface and eval loop
+config IP6_NF_IPTABLES_LEGACY
+	tristate
+
 config NF_SOCKET_IPV6
 	tristate "IPv6 socket lookup support"
 	help
@@ -147,7 +151,7 @@ config IP6_NF_MATCH_MH
 config IP6_NF_MATCH_RPFILTER
 	tristate '"rpfilter" reverse path filter match support'
 	depends on NETFILTER_ADVANCED
-	depends on IP6_NF_MANGLE || IP6_NF_RAW
+	depends on IP6_NF_MANGLE || IP6_NF_RAW || NFT_COMPAT
 	help
 	  This option allows you to match packets whose replies would
 	  go out via the interface the packet came in.
@@ -186,6 +190,8 @@ config IP6_NF_TARGET_HL
 config IP6_NF_FILTER
 	tristate "Packet filtering"
 	default m if NETFILTER_ADVANCED=n
+	select IP6_NF_IPTABLES_LEGACY
+	tristate
 	help
 	  Packet filtering defines a table `filter', which has a series of
 	  rules for simple packet filtering at local input, forwarding and
@@ -195,7 +201,7 @@ config IP6_NF_FILTER
 
 config IP6_NF_TARGET_REJECT
 	tristate "REJECT target support"
-	depends on IP6_NF_FILTER
+	depends on IP6_NF_FILTER || NFT_COMPAT
 	select NF_REJECT_IPV6
 	default m if NETFILTER_ADVANCED=n
 	help
@@ -221,6 +227,7 @@ config IP6_NF_TARGET_SYNPROXY
 config IP6_NF_MANGLE
 	tristate "Packet mangling"
 	default m if NETFILTER_ADVANCED=n
+	select IP6_NF_IPTABLES_LEGACY
 	help
 	  This option adds a `mangle' table to iptables: see the man page for
 	  iptables(8).  This table is used for various packet alterations
@@ -230,6 +237,7 @@ config IP6_NF_MANGLE
 
 config IP6_NF_RAW
 	tristate  'raw table support (required for TRACE)'
+	select IP6_NF_IPTABLES_LEGACY
 	help
 	  This option adds a `raw' table to ip6tables. This table is the very
 	  first in the netfilter framework and hooks in at the PREROUTING
@@ -243,6 +251,7 @@ config IP6_NF_SECURITY
 	tristate "Security table"
 	depends on SECURITY
 	depends on NETFILTER_ADVANCED
+	select IP6_NF_IPTABLES_LEGACY
 	help
 	  This option adds a `security' table to iptables, for use
 	  with Mandatory Access Control (MAC) policy.
@@ -254,6 +263,7 @@ config IP6_NF_NAT
 	depends on NF_CONNTRACK
 	depends on NETFILTER_ADVANCED
 	select NF_NAT
+	select IP6_NF_IPTABLES_LEGACY
 	select NETFILTER_XT_NAT
 	help
 	  This enables the `nat' table in ip6tables. This allows masquerading,
@@ -262,25 +272,23 @@ config IP6_NF_NAT
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
-if IP6_NF_NAT
-
 config IP6_NF_TARGET_MASQUERADE
 	tristate "MASQUERADE target support"
 	select NETFILTER_XT_TARGET_MASQUERADE
+	depends on IP6_NF_NAT
 	help
 	  This is a backwards-compat option for the user's convenience
 	  (e.g. when running oldconfig). It selects NETFILTER_XT_TARGET_MASQUERADE.
 
 config IP6_NF_TARGET_NPT
 	tristate "NPT (Network Prefix translation) target support"
+	depends on IP6_NF_NAT || NFT_COMPAT
 	help
 	  This option adds the `SNPT' and `DNPT' target, which perform
 	  stateless IPv6-to-IPv6 Network Prefix Translation per RFC 6296.
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
-endif # IP6_NF_NAT
-
 endif # IP6_NF_IPTABLES
 endmenu
 
diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile
index b8d6dc9aeeb6..66ce6fa5b2f5 100644
--- a/net/ipv6/netfilter/Makefile
+++ b/net/ipv6/netfilter/Makefile
@@ -4,7 +4,7 @@
 #
 
 # Link order matters here.
-obj-$(CONFIG_IP6_NF_IPTABLES) += ip6_tables.o
+obj-$(CONFIG_IP6_NF_IPTABLES_LEGACY) += ip6_tables.o
 obj-$(CONFIG_IP6_NF_FILTER) += ip6table_filter.o
 obj-$(CONFIG_IP6_NF_MANGLE) += ip6table_mangle.o
 obj-$(CONFIG_IP6_NF_RAW) += ip6table_raw.o
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 441d1f134110..df2dc21304ef 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -818,7 +818,7 @@ config NETFILTER_XT_TARGET_AUDIT
 
 config NETFILTER_XT_TARGET_CHECKSUM
 	tristate "CHECKSUM target support"
-	depends on IP_NF_MANGLE || IP6_NF_MANGLE
+	depends on IP_NF_MANGLE || IP6_NF_MANGLE || NFT_COMPAT
 	depends on NETFILTER_ADVANCED
 	help
 	  This option adds a `CHECKSUM' target, which can be used in the iptables mangle
@@ -869,7 +869,7 @@ config NETFILTER_XT_TARGET_CONNSECMARK
 config NETFILTER_XT_TARGET_CT
 	tristate '"CT" target support'
 	depends on NF_CONNTRACK
-	depends on IP_NF_RAW || IP6_NF_RAW
+	depends on IP_NF_RAW || IP6_NF_RAW || NFT_COMPAT
 	depends on NETFILTER_ADVANCED
 	help
 	  This options adds a `CT' target, which allows to specify initial
@@ -880,7 +880,7 @@ config NETFILTER_XT_TARGET_CT
 
 config NETFILTER_XT_TARGET_DSCP
 	tristate '"DSCP" and "TOS" target support'
-	depends on IP_NF_MANGLE || IP6_NF_MANGLE
+	depends on IP_NF_MANGLE || IP6_NF_MANGLE || NFT_COMPAT
 	depends on NETFILTER_ADVANCED
 	help
 	  This option adds a `DSCP' target, which allows you to manipulate
@@ -896,7 +896,7 @@ config NETFILTER_XT_TARGET_DSCP
 
 config NETFILTER_XT_TARGET_HL
 	tristate '"HL" hoplimit target support'
-	depends on IP_NF_MANGLE || IP6_NF_MANGLE
+	depends on IP_NF_MANGLE || IP6_NF_MANGLE || NFT_COMPAT
 	depends on NETFILTER_ADVANCED
 	help
 	This option adds the "HL" (for IPv6) and "TTL" (for IPv4)
@@ -1080,7 +1080,7 @@ config NETFILTER_XT_TARGET_TPROXY
 	depends on NETFILTER_ADVANCED
 	depends on IPV6 || IPV6=n
 	depends on IP6_NF_IPTABLES || IP6_NF_IPTABLES=n
-	depends on IP_NF_MANGLE
+	depends on IP_NF_MANGLE || NFT_COMPAT
 	select NF_DEFRAG_IPV4
 	select NF_DEFRAG_IPV6 if IP6_NF_IPTABLES != n
 	select NF_TPROXY_IPV4
@@ -1147,7 +1147,7 @@ config NETFILTER_XT_TARGET_TCPMSS
 
 config NETFILTER_XT_TARGET_TCPOPTSTRIP
 	tristate '"TCPOPTSTRIP" target support'
-	depends on IP_NF_MANGLE || IP6_NF_MANGLE
+	depends on IP_NF_MANGLE || IP6_NF_MANGLE || NFT_COMPAT
 	depends on NETFILTER_ADVANCED
 	help
 	  This option adds a "TCPOPTSTRIP" target, which allows you to strip
-- 
2.43.0

[PATCH nf-next 9/9] netfilter: ebtables: allow xtables-nft only builds

[Florian Westphal]

Same patch as previous one, but for ebtables.

To build a kernel that only supports ebtables-nft, the builtin tables
need to be disabled, i.e.:

CONFIG_BRIDGE_EBT_BROUTE=n
CONFIG_BRIDGE_EBT_T_FILTER=n
CONFIG_BRIDGE_EBT_T_NAT=n

The ebtables specific extensions can then be used nftables'
NFT_COMPAT interface.

Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/bridge/netfilter/Kconfig  | 7 +++++++
 net/bridge/netfilter/Makefile | 2 +-
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/net/bridge/netfilter/Kconfig b/net/bridge/netfilter/Kconfig
index 7f304a19ac1b..104c0125e32e 100644
--- a/net/bridge/netfilter/Kconfig
+++ b/net/bridge/netfilter/Kconfig
@@ -39,6 +39,10 @@ config NF_CONNTRACK_BRIDGE
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
+# old sockopt interface and eval loop
+config BRIDGE_NF_EBTABLES_LEGACY
+	tristate
+
 menuconfig BRIDGE_NF_EBTABLES
 	tristate "Ethernet Bridge tables (ebtables) support"
 	depends on BRIDGE && NETFILTER && NETFILTER_XTABLES
@@ -55,6 +59,7 @@ if BRIDGE_NF_EBTABLES
 #
 config BRIDGE_EBT_BROUTE
 	tristate "ebt: broute table support"
+	select BRIDGE_NF_EBTABLES_LEGACY
 	help
 	  The ebtables broute table is used to define rules that decide between
 	  bridging and routing frames, giving Linux the functionality of a
@@ -65,6 +70,7 @@ config BRIDGE_EBT_BROUTE
 
 config BRIDGE_EBT_T_FILTER
 	tristate "ebt: filter table support"
+	select BRIDGE_NF_EBTABLES_LEGACY
 	help
 	  The ebtables filter table is used to define frame filtering rules at
 	  local input, forwarding and local output. See the man page for
@@ -74,6 +80,7 @@ config BRIDGE_EBT_T_FILTER
 
 config BRIDGE_EBT_T_NAT
 	tristate "ebt: nat table support"
+	select BRIDGE_NF_EBTABLES_LEGACY
 	help
 	  The ebtables nat table is used to define rules that alter the MAC
 	  source address (MAC SNAT) or the MAC destination address (MAC DNAT).
diff --git a/net/bridge/netfilter/Makefile b/net/bridge/netfilter/Makefile
index 1c9ce49ab651..b9a1303da977 100644
--- a/net/bridge/netfilter/Makefile
+++ b/net/bridge/netfilter/Makefile
@@ -9,7 +9,7 @@ obj-$(CONFIG_NFT_BRIDGE_REJECT)  += nft_reject_bridge.o
 # connection tracking
 obj-$(CONFIG_NF_CONNTRACK_BRIDGE) += nf_conntrack_bridge.o
 
-obj-$(CONFIG_BRIDGE_NF_EBTABLES) += ebtables.o
+obj-$(CONFIG_BRIDGE_NF_EBTABLES_LEGACY) += ebtables.o
 
 # tables
 obj-$(CONFIG_BRIDGE_EBT_BROUTE) += ebtable_broute.o
-- 
2.43.0

Re: [PATCH nf-next 0/9] netfilter updates for -next

[Jakub Kicinski]

On Mon, 29 Jan 2024 15:57:50 +0100 Florian Westphal wrote:
> Hello,
> 
> This batch contains updates for your *next* tree.

The nf-next in the subject is a typo, right? It's for net-next?
Looks like it but better safe than sorry :)

Re: [PATCH nf-next 0/9] netfilter updates for -next

[Pablo Neira Ayuso]

On Tue, Jan 30, 2024 at 06:37:29PM -0800, Jakub Kicinski wrote:
> On Mon, 29 Jan 2024 15:57:50 +0100 Florian Westphal wrote:
> > Hello,
> > 
> > This batch contains updates for your *next* tree.
> 
> The nf-next in the subject is a typo, right? It's for net-next?
> Looks like it but better safe than sorry :)

Yes, I confirm this is net-next material.

Re: [PATCH nf-next 0/9] netfilter updates for -next

[Florian Westphal]

Jakub Kicinski <kuba@kernel.org> wrote:
> On Mon, 29 Jan 2024 15:57:50 +0100 Florian Westphal wrote:
> > Hello,
> > 
> > This batch contains updates for your *next* tree.
> 
> The nf-next in the subject is a typo, right? It's for net-next?
> Looks like it but better safe than sorry :)

Yes, should've been *net-next*,  used to typing 'nf-next'...

I've updated local plumbing to refuse sending if cover letter
is absent or lacks PATCH net(-next) in subject, so this should
not happen again.