blackbird-op-linux/drivers/net/bonding, branch master

bonding/alb: properly access headers in bond_alb_xmit()

2020-02-05T13:28:09+00:00

syzbot managed to send an IPX packet through bond_alb_xmit() and af_packet and triggered a use-after-free. First, bond_alb_xmit() was using ipx_hdr() helper to reach the IPX header, but ipx_hdr() was using the transport offset instead of the network offset. In the particular syzbot report transport offset was 0xFFFF This patch removes ipx_hdr() since it was only (mis)used from bonding. Then we need to make sure IPv4/IPv6/IPX headers are pulled in skb->head before dereferencing anything. BUG: KASAN: use-after-free in bond_alb_xmit+0x153a/0x1590 drivers/net/bonding/bond_alb.c:1452 Read of size 2 at addr ffff8801ce56dfff by task syz-executor.2/18108 (if (ipx_hdr(skb)->ipx_checksum != IPX_NO_CHECKSUM) ...) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: [] __dump_stack lib/dump_stack.c:17 [inline] [] dump_stack+0x14d/0x20b lib/dump_stack.c:53 [] print_address_description+0x6f/0x20b mm/kasan/report.c:282 [] kasan_report_error mm/kasan/report.c:380 [inline] [] kasan_report mm/kasan/report.c:438 [inline] [] kasan_report.cold+0x8c/0x2a0 mm/kasan/report.c:422 [] __asan_report_load_n_noabort+0xf/0x20 mm/kasan/report.c:469 [] bond_alb_xmit+0x153a/0x1590 drivers/net/bonding/bond_alb.c:1452 [] __bond_start_xmit drivers/net/bonding/bond_main.c:4199 [inline] [] bond_start_xmit+0x4f4/0x1570 drivers/net/bonding/bond_main.c:4224 [] __netdev_start_xmit include/linux/netdevice.h:4525 [inline] [] netdev_start_xmit include/linux/netdevice.h:4539 [inline] [] xmit_one net/core/dev.c:3611 [inline] [] dev_hard_start_xmit+0x168/0x910 net/core/dev.c:3627 [] __dev_queue_xmit+0x1f55/0x33b0 net/core/dev.c:4238 [] dev_queue_xmit+0x18/0x20 net/core/dev.c:4278 [] packet_snd net/packet/af_packet.c:3226 [inline] [] packet_sendmsg+0x4919/0x70b0 net/packet/af_packet.c:3252 [] sock_sendmsg_nosec net/socket.c:673 [inline] [] sock_sendmsg+0x12c/0x160 net/socket.c:684 [] __sys_sendto+0x262/0x380 net/socket.c:1996 [] SYSC_sendto net/socket.c:2008 [inline] [] SyS_sendto+0x40/0x60 net/socket.c:2004 Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Dumazet Reported-by: syzbot Cc: Jay Vosburgh Cc: Veaceslav Falico Cc: Andy Gospodarek Signed-off-by: David S. Miller

bonding: rename AD_STATE_* to LACP_STATE_*

2019-12-26T21:09:37+00:00

As the LACP actor/partner state is now part of the uapi, rename the 3ad state defines with LACP prefix. The LACP prefix is preferred over BOND_3AD as the LACP standard moved to 802.1AX. Fixes: 826f66b30c2e3 ("bonding: move 802.3ad port state flags to uapi") Signed-off-by: Andy Roulin Signed-off-by: David S. Miller

Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2019-12-22T23:15:05+00:00

Mere overlapping changes in the conflicts here. Signed-off-by: David S. Miller

bonding: fix active-backup transition after link failure

2019-12-15T00:22:34+00:00

After the recent fix in commit 1899bb325149 ("bonding: fix state transition issue in link monitoring"), the active-backup mode with miimon initially come-up fine but after a link-failure, both members transition into backup state. Following steps to reproduce the scenario (eth1 and eth2 are the slaves of the bond): ip link set eth1 up ip link set eth2 down sleep 1 ip link set eth2 up ip link set eth1 down cat /sys/class/net/eth1/bonding_slave/state cat /sys/class/net/eth2/bonding_slave/state Fixes: 1899bb325149 ("bonding: fix state transition issue in link monitoring") CC: Jay Vosburgh Signed-off-by: Mahesh Bandewar Acked-by: Jay Vosburgh Signed-off-by: Jakub Kicinski

bonding: move 802.3ad port state flags to uapi

2019-12-14T21:07:44+00:00

The bond slave actor/partner operating state is exported as bitfield to userspace, which lacks a way to interpret it, e.g., iproute2 only prints the state as a number: ad_actor_oper_port_state 15 For userspace to interpret the bitfield, the bitfield definitions should be part of the uapi. The bitfield itself is defined in the 802.3ad standard. This commit moves the 802.3ad bitfield definitions to uapi. Related iproute2 patches, soon to be posted upstream, use the new uapi headers to pretty-print bond slave state, e.g., with ip -d link show ad_actor_oper_port_state_str Signed-off-by: Andy Roulin Acked-by: Roopa Prabhu Acked-by: Jay Vosburgh Signed-off-by: Jakub Kicinski

bonding: fix bond_neigh_init()

2019-12-09T17:49:07+00:00

1) syzbot reported an uninit-value in bond_neigh_setup() [1] bond_neigh_setup() uses a temporary on-stack 'struct neigh_parms parms', but only clears parms.neigh_setup field. A stacked bonding device would then enter bond_neigh_setup() and read garbage from parms->dev. If we get really unlucky and garbage is matching @dev, then we could recurse and eventually crash. Let's make sure the whole structure is cleared to avoid surprises. 2) bond_neigh_setup() can be called while another cpu manipulates the master device, removing or adding a slave. We need at least rcu protection to prevent use-after-free. Note: Prior code does not support a stack of bonding devices, this patch does not attempt to fix this, and leave a comment instead. [1] BUG: KMSAN: uninit-value in bond_neigh_setup+0xa4/0x110 drivers/net/bonding/bond_main.c:3655 CPU: 0 PID: 11256 Comm: syz-executor.0 Not tainted 5.4.0-rc8-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x1c9/0x220 lib/dump_stack.c:118 kmsan_report+0x128/0x220 mm/kmsan/kmsan_report.c:108 __msan_warning+0x57/0xa0 mm/kmsan/kmsan_instr.c:245 bond_neigh_setup+0xa4/0x110 drivers/net/bonding/bond_main.c:3655 bond_neigh_init+0x216/0x4b0 drivers/net/bonding/bond_main.c:3626 ___neigh_create+0x169e/0x2c40 net/core/neighbour.c:613 __neigh_create+0xbd/0xd0 net/core/neighbour.c:674 ip6_finish_output2+0x149a/0x2670 net/ipv6/ip6_output.c:113 __ip6_finish_output+0x83d/0x8f0 net/ipv6/ip6_output.c:142 ip6_finish_output+0x2db/0x420 net/ipv6/ip6_output.c:152 NF_HOOK_COND include/linux/netfilter.h:294 [inline] ip6_output+0x5d3/0x720 net/ipv6/ip6_output.c:175 dst_output include/net/dst.h:436 [inline] NF_HOOK include/linux/netfilter.h:305 [inline] mld_sendpack+0xebd/0x13d0 net/ipv6/mcast.c:1682 mld_send_cr net/ipv6/mcast.c:1978 [inline] mld_ifc_timer_expire+0x116b/0x1680 net/ipv6/mcast.c:2477 call_timer_fn+0x232/0x530 kernel/time/timer.c:1404 expire_timers kernel/time/timer.c:1449 [inline] __run_timers+0xd60/0x1270 kernel/time/timer.c:1773 run_timer_softirq+0x2d/0x50 kernel/time/timer.c:1786 __do_softirq+0x4a1/0x83a kernel/softirq.c:293 invoke_softirq kernel/softirq.c:375 [inline] irq_exit+0x230/0x280 kernel/softirq.c:416 exiting_irq+0xe/0x10 arch/x86/include/asm/apic.h:536 smp_apic_timer_interrupt+0x48/0x70 arch/x86/kernel/apic/apic.c:1138 apic_timer_interrupt+0x2e/0x40 arch/x86/entry/entry_64.S:835 RIP: 0010:kmsan_free_page+0x18d/0x1c0 mm/kmsan/kmsan_shadow.c:439 Code: 4c 89 ff 44 89 f6 e8 82 0d ee ff 65 ff 0d 9f 26 3b 60 65 8b 05 98 26 3b 60 85 c0 75 24 e8 5b f6 35 ff 4c 89 6d d0 ff 75 d0 9d <48> 83 c4 10 5b 41 5c 41 5d 41 5e 41 5f 5d c3 0f 0b 0f 0b 0f 0b 0f RSP: 0018:ffffb328034af818 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff13 RAX: 0000000000000000 RBX: ffffe2d7471f8360 RCX: 0000000000000000 RDX: ffffffffadea7000 RSI: 0000000000000004 RDI: ffff93496fcda104 RBP: ffffb328034af850 R08: ffff934a47e86d00 R09: ffff93496fc41900 R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000001 R13: 0000000000000246 R14: 0000000000000000 R15: ffffe2d7472225c0 free_pages_prepare mm/page_alloc.c:1138 [inline] free_pcp_prepare mm/page_alloc.c:1230 [inline] free_unref_page_prepare+0x1d9/0x770 mm/page_alloc.c:3025 free_unref_page mm/page_alloc.c:3074 [inline] free_the_page mm/page_alloc.c:4832 [inline] __free_pages+0x154/0x230 mm/page_alloc.c:4840 __vunmap+0xdac/0xf20 mm/vmalloc.c:2277 __vfree mm/vmalloc.c:2325 [inline] vfree+0x7c/0x170 mm/vmalloc.c:2355 copy_entries_to_user net/ipv6/netfilter/ip6_tables.c:883 [inline] get_entries net/ipv6/netfilter/ip6_tables.c:1041 [inline] do_ip6t_get_ctl+0xfa4/0x1030 net/ipv6/netfilter/ip6_tables.c:1709 nf_sockopt net/netfilter/nf_sockopt.c:104 [inline] nf_getsockopt+0x481/0x4e0 net/netfilter/nf_sockopt.c:122 ipv6_getsockopt+0x264/0x510 net/ipv6/ipv6_sockglue.c:1400 tcp_getsockopt+0x1c6/0x1f0 net/ipv4/tcp.c:3688 sock_common_getsockopt+0x13f/0x180 net/core/sock.c:3110 __sys_getsockopt+0x533/0x7b0 net/socket.c:2129 __do_sys_getsockopt net/socket.c:2144 [inline] __se_sys_getsockopt+0xe1/0x100 net/socket.c:2141 __x64_sys_getsockopt+0x62/0x80 net/socket.c:2141 do_syscall_64+0xb6/0x160 arch/x86/entry/common.c:291 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x45d20a Code: b8 34 01 00 00 0f 05 48 3d 01 f0 ff ff 0f 83 8d 8b fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 49 89 ca b8 37 00 00 00 0f 05 <48> 3d 01 f0 ff ff 0f 83 6a 8b fb ff c3 66 0f 1f 84 00 00 00 00 00 RSP: 002b:0000000000a6f618 EFLAGS: 00000212 ORIG_RAX: 0000000000000037 RAX: ffffffffffffffda RBX: 0000000000a6f640 RCX: 000000000045d20a RDX: 0000000000000041 RSI: 0000000000000029 RDI: 0000000000000003 RBP: 0000000000717cc0 R08: 0000000000a6f63c R09: 0000000000004000 R10: 0000000000a6f740 R11: 0000000000000212 R12: 0000000000000003 R13: 0000000000000000 R14: 0000000000000029 R15: 0000000000715b00 Local variable description: ----parms@bond_neigh_init Variable was created at: bond_neigh_init+0x8c/0x4b0 drivers/net/bonding/bond_main.c:3617 bond_neigh_init+0x8c/0x4b0 drivers/net/bonding/bond_main.c:3617 Fixes: 9918d5bf329d ("bonding: modify only neigh_parms owned by us") Fixes: 234bcf8a499e ("net/bonding: correctly proxy slave neigh param setup ndo function") Signed-off-by: Eric Dumazet Reported-by: syzbot Cc: Jay Vosburgh Cc: Veaceslav Falico Cc: Andy Gospodarek Signed-off-by: David S. Miller

neighbour: remove neigh_cleanup() method

2019-12-09T17:48:47+00:00

neigh_cleanup() has not been used for seven years, and was a wrong design. Messing with shared pointer in bond_neigh_init() without proper memory barriers would at least trigger syzbot complains eventually. It is time to remove this stuff. Fixes: b63b70d87741 ("IPoIB: Use a private hash table for path lookup in xmit path") Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller

bonding: symmetric ICMP transmit

2019-11-16T21:02:53+00:00

A bonding with layer2+3 or layer3+4 hashing uses the IP addresses and the ports to balance packets between slaves. With some network errors, we receive an ICMP error packet by the remote host or a router. If sent by a router, the source IP can differ from the remote host one. Additionally the ICMP protocol has no port numbers, so a layer3+4 bonding will get a different hash than the previous one. These two conditions could let the packet go through a different interface than the other packets of the same flow: # tcpdump -qltnni veth0 |sed 's/^/0: /' & # tcpdump -qltnni veth1 |sed 's/^/1: /' & # hping3 -2 192.168.0.2 -p 9 0: IP 192.168.0.1.2251 > 192.168.0.2.9: UDP, length 0 1: IP 192.168.0.2 > 192.168.0.1: ICMP 192.168.0.2 udp port 9 unreachable, length 36 1: IP 192.168.0.1.2252 > 192.168.0.2.9: UDP, length 0 1: IP 192.168.0.2 > 192.168.0.1: ICMP 192.168.0.2 udp port 9 unreachable, length 36 1: IP 192.168.0.1.2253 > 192.168.0.2.9: UDP, length 0 1: IP 192.168.0.2 > 192.168.0.1: ICMP 192.168.0.2 udp port 9 unreachable, length 36 0: IP 192.168.0.1.2254 > 192.168.0.2.9: UDP, length 0 1: IP 192.168.0.2 > 192.168.0.1: ICMP 192.168.0.2 udp port 9 unreachable, length 36 An ICMP error packet contains the header of the packet which caused the network error, so inspect it and match the flow against it, so we can send the ICMP via the same interface of the previous packet in the flow. Move the IP and port dissect code into a generic function bond_flow_ip() and if we are dissecting an ICMP error packet, call it again with the adjusted offset. # hping3 -2 192.168.0.2 -p 9 1: IP 192.168.0.1.1224 > 192.168.0.2.9: UDP, length 0 1: IP 192.168.0.2 > 192.168.0.1: ICMP 192.168.0.2 udp port 9 unreachable, length 36 1: IP 192.168.0.1.1225 > 192.168.0.2.9: UDP, length 0 1: IP 192.168.0.2 > 192.168.0.1: ICMP 192.168.0.2 udp port 9 unreachable, length 36 0: IP 192.168.0.1.1226 > 192.168.0.2.9: UDP, length 0 0: IP 192.168.0.2 > 192.168.0.1: ICMP 192.168.0.2 udp port 9 unreachable, length 36 0: IP 192.168.0.1.1227 > 192.168.0.2.9: UDP, length 0 0: IP 192.168.0.2 > 192.168.0.1: ICMP 192.168.0.2 udp port 9 unreachable, length 36 Signed-off-by: Matteo Croce Signed-off-by: David S. Miller

Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2019-11-09T19:04:37+00:00

One conflict in the BPF samples Makefile, some fixes in 'net' whilst we were converting over to Makefile.target rules in 'net-next'. Signed-off-by: David S. Miller

bonding: fix state transition issue in link monitoring

2019-11-06T01:40:16+00:00

Since de77ecd4ef02 ("bonding: improve link-status update in mii-monitoring"), the bonding driver has utilized two separate variables to indicate the next link state a particular slave should transition to. Each is used to communicate to a different portion of the link state change commit logic; one to the bond_miimon_commit function itself, and another to the state transition logic. Unfortunately, the two variables can become unsynchronized, resulting in incorrect link state transitions within bonding. This can cause slaves to become stuck in an incorrect link state until a subsequent carrier state transition. The issue occurs when a special case in bond_slave_netdev_event sets slave->link directly to BOND_LINK_FAIL. On the next pass through bond_miimon_inspect after the slave goes carrier up, the BOND_LINK_FAIL case will set the proposed next state (link_new_state) to BOND_LINK_UP, but the new_link to BOND_LINK_DOWN. The setting of the final link state from new_link comes after that from link_new_state, and so the slave will end up incorrectly in _DOWN state. Resolve this by combining the two variables into one. Reported-by: Aleksei Zakharov Reported-by: Sha Zhang Cc: Mahesh Bandewar Fixes: de77ecd4ef02 ("bonding: improve link-status update in mii-monitoring") Signed-off-by: Jay Vosburgh Signed-off-by: David S. Miller