Merge branch 'net-visibility-of-memory-limits-in-netns'

Matteo Croce says:

====================
net: visibility of memory limits in netns

Some programs need to know the size of the network buffers to operate
correctly, export the following sysctls read-only in network namespaces:

- net.core.rmem_default
- net.core.rmem_max
- net.core.wmem_default
- net.core.wmem_max
====================

Link: https://lore.kernel.org/r/20240530232722.45255-1-technoboy85@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/core/sysctl_net_core.c

@@ -382,38 +382,6 @@ proc_dolongvec_minmax_bpf_restricted(struct ctl_table *table, int write,
 #endif
 
 static struct ctl_table net_core_table[] = {
-	{
-		.procname	= "wmem_max",
-		.data		= &sysctl_wmem_max,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &min_sndbuf,
-	},
-	{
-		.procname	= "rmem_max",
-		.data		= &sysctl_rmem_max,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &min_rcvbuf,
-	},
-	{
-		.procname	= "wmem_default",
-		.data		= &sysctl_wmem_default,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &min_sndbuf,
-	},
-	{
-		.procname	= "rmem_default",
-		.data		= &sysctl_rmem_default,
-		.maxlen		= sizeof(int),
-		.mode		= 0644,
-		.proc_handler	= proc_dointvec_minmax,
-		.extra1		= &min_rcvbuf,
-	},
 	{
 		.procname	= "mem_pcpu_rsv",
 		.data		= &net_hotdata.sysctl_mem_pcpu_rsv,
@@ -697,6 +665,41 @@ static struct ctl_table netns_core_table[] = {
 		.extra2		= SYSCTL_ONE,
 		.proc_handler	= proc_dou8vec_minmax,
 	},
+	/* sysctl_core_net_init() will set the values after this
+	 * to readonly in network namespaces
+	 */
+	{
+		.procname	= "wmem_max",
+		.data		= &sysctl_wmem_max,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_sndbuf,
+	},
+	{
+		.procname	= "rmem_max",
+		.data		= &sysctl_rmem_max,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_rcvbuf,
+	},
+	{
+		.procname	= "wmem_default",
+		.data		= &sysctl_wmem_default,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_sndbuf,
+	},
+	{
+		.procname	= "rmem_default",
+		.data		= &sysctl_rmem_default,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_rcvbuf,
+	},
 };
 
 static int __init fb_tunnels_only_for_init_net_sysctl_setup(char *str)
@@ -724,8 +727,14 @@ static __net_init int sysctl_core_net_init(struct net *net)
 		if (tbl == NULL)
 			goto err_dup;
 
-		for (i = 0; i < table_size; ++i)
+		for (i = 0; i < table_size; ++i) {
+			if (tbl[i].data == &sysctl_wmem_max)
+				break;
+
 			tbl[i].data += (char *)net - (char *)&init_net;
+		}
+		for (; i < table_size; ++i)
+			tbl[i].mode &= ~0222;
 	}
 
 	net->core.sysctl_hdr = register_net_sysctl_sz(net, "net/core", tbl, table_size);

tools/testing/selftests/net/Makefile

@@ -53,6 +53,7 @@ TEST_PROGS += bind_bhash.sh
 TEST_PROGS += ip_local_port_range.sh
 TEST_PROGS += rps_default_mask.sh
 TEST_PROGS += big_tcp.sh
+TEST_PROGS += netns-sysctl.sh
 TEST_PROGS_EXTENDED := toeplitz_client.sh toeplitz.sh
 TEST_GEN_FILES =  socket nettest
 TEST_GEN_FILES += psock_fanout psock_tpacket msg_zerocopy reuseport_addr_any

tools/testing/selftests/net/netns-sysctl.sh

@@ -0,0 +1,40 @@
+#!/bin/bash -e
+# SPDX-License-Identifier: GPL-2.0
+#
+# This test checks that the network buffer sysctls are present
+# in a network namespaces, and that they are readonly.
+
+source lib.sh
+
+cleanup() {
+    cleanup_ns $test_ns
+}
+
+trap cleanup EXIT
+
+fail() {
+	echo "ERROR: $*" >&2
+	exit 1
+}
+
+setup_ns test_ns
+
+for sc in {r,w}mem_{default,max}; do
+	# check that this is writable in a netns
+	[ -w "/proc/sys/net/core/$sc" ] ||
+		fail "$sc isn't writable in the init netns!"
+
+	# change the value in the host netns
+	sysctl -qw "net.core.$sc=300000" ||
+		fail "Can't write $sc in init netns!"
+
+	# check that the value is read from the init netns
+	[ "$(ip netns exec $test_ns sysctl -n "net.core.$sc")" -eq 300000 ] ||
+		fail "Value for $sc mismatch!"
+
+	# check that this isn't writable in a netns
+	ip netns exec $test_ns [ -w "/proc/sys/net/core/$sc" ] &&
+		fail "$sc is writable in a netns!"
+done
+
+echo 'Test passed OK'