5.15-stable review patch. If anyone has any objections, please let me know. ------------------ From: Xiang Mei [ Upstream commit 2884bf72fb8f03409e423397319205de48adca16 ] bond_xmit_broadcast() reuses the original skb for the last slave (determined by bond_is_last_slave()) and clones it for others. Concurrent slave enslave/release can mutate the slave list during RCU-protected iteration, changing which slave is "last" mid-loop. This causes the original skb to be double-consumed (double-freed). Replace the racy bond_is_last_slave() check with a simple index comparison (i + 1 == slaves_count) against the pre-snapshot slave count taken via READ_ONCE() before the loop. This preserves the zero-copy optimization for the last slave while making the "last" determination stable against concurrent list mutations. The UAF can trigger the following crash: ================================================================== BUG: KASAN: slab-use-after-free in skb_clone Read of size 8 at addr ffff888100ef8d40 by task exploit/147 CPU: 1 UID: 0 PID: 147 Comm: exploit Not tainted 7.0.0-rc3+ #4 PREEMPTLAZY Call Trace: dump_stack_lvl (lib/dump_stack.c:123) print_report (mm/kasan/report.c:379 mm/kasan/report.c:482) kasan_report (mm/kasan/report.c:597) skb_clone (include/linux/skbuff.h:1724 include/linux/skbuff.h:1792 include/linux/skbuff.h:3396 net/core/skbuff.c:2108) bond_xmit_broadcast (drivers/net/bonding/bond_main.c:5334) bond_start_xmit (drivers/net/bonding/bond_main.c:5567 drivers/net/bonding/bond_main.c:5593) dev_hard_start_xmit (include/linux/netdevice.h:5325 include/linux/netdevice.h:5334 net/core/dev.c:3871 net/core/dev.c:3887) __dev_queue_xmit (include/linux/netdevice.h:3601 net/core/dev.c:4838) ip6_finish_output2 (include/net/neighbour.h:540 include/net/neighbour.h:554 net/ipv6/ip6_output.c:136) ip6_finish_output (net/ipv6/ip6_output.c:208 net/ipv6/ip6_output.c:219) ip6_output (net/ipv6/ip6_output.c:250) ip6_send_skb (net/ipv6/ip6_output.c:1985) udp_v6_send_skb (net/ipv6/udp.c:1442) udpv6_sendmsg (net/ipv6/udp.c:1733) __sys_sendto (net/socket.c:730 net/socket.c:742 net/socket.c:2206) __x64_sys_sendto (net/socket.c:2209) do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) Allocated by task 147: Freed by task 147: The buggy address belongs to the object at ffff888100ef8c80 which belongs to the cache skbuff_head_cache of size 224 The buggy address is located 192 bytes inside of freed 224-byte region [ffff888100ef8c80, ffff888100ef8d60) Memory state around the buggy address: ffff888100ef8c00: fb fb fb fb fc fc fc fc fc fc fc fc fc fc fc fc ffff888100ef8c80: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb >ffff888100ef8d00: fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc ^ ffff888100ef8d80: fc fc fc fc fc fc fc fc fa fb fb fb fb fb fb fb ffff888100ef8e00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ================================================================== Fixes: 4e5bd03ae346 ("net: bonding: fix bond_xmit_broadcast return value error bug") Reported-by: Weiming Shi Signed-off-by: Xiang Mei Link: https://patch.msgid.link/20260326075553.3960562-1-xmei5@asu.edu Signed-off-by: Paolo Abeni [ Backport to 5.15.y: 5.15's bond_xmit_broadcast() walks the live slave list, not the all_slaves array upstream indexes into (that walk came with ce7a381697cb "net: bonding: add broadcast_neighbor option for 802.3ad"), so there is no index/count pair that stays consistent while the list changes. Instead of choosing the "last" slave up front, defer it: clone the skb for each up slave as the next one is found, and transmit the original skb to the last up slave seen only after the walk has finished. The original skb is consumed exactly once and not touched afterwards, however the list changes during the RCU walk; slave pointers stay valid under rcu_read_lock() since __bond_release_one() does synchronize_rcu() before freeing. ] Assisted-by: Claude:claude-opus-5-5 Signed-off-by: Steffen Pfendtner Signed-off-by: Sasha Levin --- drivers/net/bonding/bond_main.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index b74732d2924b1..12dc1d3c6e2b6 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4915,35 +4915,36 @@ static netdev_tx_t bond_xmit_broadcast(struct sk_buff *skb, struct net_device *bond_dev) { struct bonding *bond = netdev_priv(bond_dev); - struct slave *slave = NULL; + struct slave *slave = NULL, *last = NULL; struct list_head *iter; bool xmit_suc = false; - bool skb_used = false; + /* The slave list can change under us during the RCU walk, so which + * slave is "last" is only known once the walk is over: clone for + * every up slave but the last one seen, then hand the original skb + * to that one after the loop. + */ bond_for_each_slave_rcu(bond, slave, iter) { struct sk_buff *skb2; if (!(bond_slave_is_up(slave) && slave->link == BOND_LINK_UP)) continue; - if (bond_is_last_slave(bond, slave)) { - skb2 = skb; - skb_used = true; - } else { + if (last) { skb2 = skb_clone(skb, GFP_ATOMIC); - if (!skb2) { + if (!skb2) net_err_ratelimited("%s: Error: %s: skb_clone() failed\n", bond_dev->name, __func__); - continue; - } + else if (bond_dev_queue_xmit(bond, skb2, last->dev) == NETDEV_TX_OK) + xmit_suc = true; } - - if (bond_dev_queue_xmit(bond, skb2, slave->dev) == NETDEV_TX_OK) - xmit_suc = true; + last = slave; } - if (!skb_used) + if (!last) dev_kfree_skb_any(skb); + else if (bond_dev_queue_xmit(bond, skb, last->dev) == NETDEV_TX_OK) + xmit_suc = true; if (xmit_suc) return NETDEV_TX_OK; -- 2.53.0