Merge tag 'net-6.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Pull networking fixes from Jakub Kicinski:
 "Including fixes from bpf and netfilter.

  Previous releases - regressions:

   - Revert "net: rtnetlink: Enslave device before bringing it up",
     breaks the case inverse to the one it was trying to fix

   - net: dsa: fix oob access in DSA's netdevice event handler
     dereference netdev_priv() before check its a DSA port

   - sched: track device in tcf_block_get/put_ext() only for clsact
     binder types

   - net: tls, fix WARNING in __sk_msg_free when record becomes full
     during splice and MORE hint set

   - sfp-bus: fix SFP mode detect from bitrate

   - drv: stmmac: prevent DSA tags from breaking COE

  Previous releases - always broken:

   - bpf: fix no forward progress in in bpf_iter_udp if output buffer is
     too small

   - bpf: reject variable offset alu on registers with a type of
     PTR_TO_FLOW_KEYS to prevent oob access

   - netfilter: tighten input validation

   - net: add more sanity check in virtio_net_hdr_to_skb()

   - rxrpc: fix use of Don't Fragment flag on RESPONSE packets, avoid
     infinite loop

   - amt: do not use the portion of skb->cb area which may get clobbered

   - mptcp: improve validation of the MPTCPOPT_MP_JOIN MCTCP option

  Misc:

   - spring cleanup of inactive maintainers"

* tag 'net-6.8-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (88 commits)
  i40e: Include types.h to some headers
  ipv6: mcast: fix data-race in ipv6_mc_down / mld_ifc_work
  selftests: mlxsw: qos_pfc: Adjust the test to support 8 lanes
  selftests: mlxsw: qos_pfc: Remove wrong description
  mlxsw: spectrum_router: Register netdevice notifier before nexthop
  mlxsw: spectrum_acl_tcam: Fix stack corruption
  mlxsw: spectrum_acl_tcam: Fix NULL pointer dereference in error path
  mlxsw: spectrum_acl_erp: Fix error flow of pool allocation failure
  ethtool: netlink: Add missing ethnl_ops_begin/complete
  selftests: bonding: Add more missing config options
  selftests: netdevsim: add a config file
  libbpf: warn on unexpected __arg_ctx type when rewriting BTF
  selftests/bpf: add tests confirming type logic in kernel for __arg_ctx
  bpf: enforce types for __arg_ctx-tagged arguments in global subprogs
  bpf: extract bpf_ctx_convert_map logic and make it more reusable
  libbpf: feature-detect arg:ctx tag support in kernel
  ipvs: avoid stat macros calls from preemptible context
  netfilter: nf_tables: reject NFT_SET_CONCAT with not field length description
  netfilter: nf_tables: skip dead set elements in netlink dump
  netfilter: nf_tables: do not allow mismatch field size and set key length
  ...

1 files changed, 135 insertions, 0 deletions

diff --git a/tools/testing/selftests/bpf/prog_tests/sock_iter_batch.c b/tools/testing/selftests/bpf/prog_tests/sock_iter_batch.c
new file mode 100644
index 0000000000000..0c365f36c73b5
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/sock_iter_batch.c
@@ -0,0 +1,135 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2024 Meta
+
+#include <test_progs.h>
+#include "network_helpers.h"
+#include "sock_iter_batch.skel.h"
+
+#define TEST_NS "sock_iter_batch_netns"
+
+static const int nr_soreuse = 4;
+
+static void do_test(int sock_type, bool onebyone)
+{
+	int err, i, nread, to_read, total_read, iter_fd = -1;
+	int first_idx, second_idx, indices[nr_soreuse];
+	struct bpf_link *link = NULL;
+	struct sock_iter_batch *skel;
+	int *fds[2] = {};
+
+	skel = sock_iter_batch__open();
+	if (!ASSERT_OK_PTR(skel, "sock_iter_batch__open"))
+		return;
+
+	/* Prepare 2 buckets of sockets in the kernel hashtable */
+	for (i = 0; i < ARRAY_SIZE(fds); i++) {
+		int local_port;
+
+		fds[i] = start_reuseport_server(AF_INET6, sock_type, "::1", 0, 0,
+						nr_soreuse);
+		if (!ASSERT_OK_PTR(fds[i], "start_reuseport_server"))
+			goto done;
+		local_port = get_socket_local_port(*fds[i]);
+		if (!ASSERT_GE(local_port, 0, "get_socket_local_port"))
+			goto done;
+		skel->rodata->ports[i] = ntohs(local_port);
+	}
+
+	err = sock_iter_batch__load(skel);
+	if (!ASSERT_OK(err, "sock_iter_batch__load"))
+		goto done;
+
+	link = bpf_program__attach_iter(sock_type == SOCK_STREAM ?
+					skel->progs.iter_tcp_soreuse :
+					skel->progs.iter_udp_soreuse,
+					NULL);
+	if (!ASSERT_OK_PTR(link, "bpf_program__attach_iter"))
+		goto done;
+
+	iter_fd = bpf_iter_create(bpf_link__fd(link));
+	if (!ASSERT_GE(iter_fd, 0, "bpf_iter_create"))
+		goto done;
+
+	/* Test reading a bucket (either from fds[0] or fds[1]).
+	 * Only read "nr_soreuse - 1" number of sockets
+	 * from a bucket and leave one socket out from
+	 * that bucket on purpose.
+	 */
+	to_read = (nr_soreuse - 1) * sizeof(*indices);
+	total_read = 0;
+	first_idx = -1;
+	do {
+		nread = read(iter_fd, indices, onebyone ? sizeof(*indices) : to_read);
+		if (nread <= 0 || nread % sizeof(*indices))
+			break;
+		total_read += nread;
+
+		if (first_idx == -1)
+			first_idx = indices[0];
+		for (i = 0; i < nread / sizeof(*indices); i++)
+			ASSERT_EQ(indices[i], first_idx, "first_idx");
+	} while (total_read < to_read);
+	ASSERT_EQ(nread, onebyone ? sizeof(*indices) : to_read, "nread");
+	ASSERT_EQ(total_read, to_read, "total_read");
+
+	free_fds(fds[first_idx], nr_soreuse);
+	fds[first_idx] = NULL;
+
+	/* Read the "whole" second bucket */
+	to_read = nr_soreuse * sizeof(*indices);
+	total_read = 0;
+	second_idx = !first_idx;
+	do {
+		nread = read(iter_fd, indices, onebyone ? sizeof(*indices) : to_read);
+		if (nread <= 0 || nread % sizeof(*indices))
+			break;
+		total_read += nread;
+
+		for (i = 0; i < nread / sizeof(*indices); i++)
+			ASSERT_EQ(indices[i], second_idx, "second_idx");
+	} while (total_read <= to_read);
+	ASSERT_EQ(nread, 0, "nread");
+	/* Both so_reuseport ports should be in different buckets, so
+	 * total_read must equal to the expected to_read.
+	 *
+	 * For a very unlikely case, both ports collide at the same bucket,
+	 * the bucket offset (i.e. 3) will be skipped and it cannot
+	 * expect the to_read number of bytes.
+	 */
+	if (skel->bss->bucket[0] != skel->bss->bucket[1])
+		ASSERT_EQ(total_read, to_read, "total_read");
+
+done:
+	for (i = 0; i < ARRAY_SIZE(fds); i++)
+		free_fds(fds[i], nr_soreuse);
+	if (iter_fd < 0)
+		close(iter_fd);
+	bpf_link__destroy(link);
+	sock_iter_batch__destroy(skel);
+}
+
+void test_sock_iter_batch(void)
+{
+	struct nstoken *nstoken = NULL;
+
+	SYS_NOFAIL("ip netns del " TEST_NS " &> /dev/null");
+	SYS(done, "ip netns add %s", TEST_NS);
+	SYS(done, "ip -net %s link set dev lo up", TEST_NS);
+
+	nstoken = open_netns(TEST_NS);
+	if (!ASSERT_OK_PTR(nstoken, "open_netns"))
+		goto done;
+
+	if (test__start_subtest("tcp")) {
+		do_test(SOCK_STREAM, true);
+		do_test(SOCK_STREAM, false);
+	}
+	if (test__start_subtest("udp")) {
+		do_test(SOCK_DGRAM, true);
+		do_test(SOCK_DGRAM, false);
+	}
+	close_netns(nstoken);
+
+done:
+	SYS_NOFAIL("ip netns del " TEST_NS " &> /dev/null");
+}