KCSAN: data-race in __dev_change_flags / is_upper_ndev_bond_master_filter (9)
Subsystems (2)
syzbot
syzbotHello,
syzbot found the following issue on:
HEAD commit: 8a9c6c40432e Merge tag 'io_uring-6.10-20240703' of git://g.. git tree: upstream console output: https://syzkaller.appspot.com/x/log.txt?x=1558eeb9980000 kernel config: https://syzkaller.appspot.com/x/.config?x=5b9537cd00be479e dashboard link: https://syzkaller.appspot.com/bug?extid=113b65786d8662e21ff7 compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
Unfortunately, I don't have any reproducer for this issue yet.
Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/f01153f3ec8e/disk-8a9c6c40.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/04b88663a824/vmlinux-8a9c6c40.xz
kernel image: https://storage.googleapis.com/syzbot-assets/63c518cc63c5/bzImage-8a9c6c40.xz
IMPORTANT: if you fix the issue, please add the following tag to the commit:
==================================================================
BUG: KCSAN: data-race in __dev_change_flags / is_upper_ndev_bond_master_filter
read-write to 0xffff888112fca0b0 of 4 bytes by task 9257 on cpu 1:
__dev_change_flags+0x9a/0x410 net/core/dev.c:8755 rtnl_configure_link net/core/rtnetlink.c:3321 [inline] rtnl_newlink_create net/core/rtnetlink.c:3518 [inline] __rtnl_newlink net/core/rtnetlink.c:3730 [inline] rtnl_newlink+0x121e/0x1690 net/core/rtnetlink.c:3743 rtnetlink_rcv_msg+0x85e/0x910 net/core/rtnetlink.c:6635 netlink_rcv_skb+0x12c/0x230 net/netlink/af_netlink.c:2564 rtnetlink_rcv+0x1c/0x30 net/core/rtnetlink.c:6653 netlink_unicast_kernel net/netlink/af_netlink.c:1335 [inline] netlink_unicast+0x58d/0x660 net/netlink/af_netlink.c:1361 netlink_sendmsg+0x5ca/0x6e0 net/netlink/af_netlink.c:1905 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg+0x140/0x180 net/socket.c:745 ____sys_sendmsg+0x312/0x410 net/socket.c:2585 ___sys_sendmsg net/socket.c:2639 [inline] __sys_sendmsg+0x1e9/0x280 net/socket.c:2668 __do_sys_sendmsg net/socket.c:2677 [inline] __se_sys_sendmsg net/socket.c:2675 [inline] __x64_sys_sendmsg+0x46/0x50 net/socket.c:2675 x64_sys_call+0xb25/0x2d70 arch/x86/include/generated/asm/syscalls_64.h:47 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xc9/0x1c0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f
read to 0xffff888112fca0b0 of 4 bytes by task 3104 on cpu 0:
netif_is_bond_master include/linux/netdevice.h:5020 [inline] is_upper_ndev_bond_master_filter+0x2b/0xb0 drivers/infiniband/core/roce_gid_mgmt.c:275 ib_enum_roce_netdev+0x124/0x1d0 drivers/infiniband/core/device.c:2310 ib_enum_all_roce_netdevs+0x8a/0x100 drivers/infiniband/core/device.c:2337 netdevice_event_work_handler+0x15b/0x3c0 drivers/infiniband/core/roce_gid_mgmt.c:626 process_one_work kernel/workqueue.c:3248 [inline] process_scheduled_works+0x483/0x9a0 kernel/workqueue.c:3329 worker_thread+0x526/0x720 kernel/workqueue.c:3409 kthread+0x1d1/0x210 kernel/kthread.c:389 ret_from_fork+0x4b/0x60 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244
value changed: 0x00001002 -> 0x00008002
Reported by Kernel Concurrency Sanitizer on:
CPU: 0 PID: 3104 Comm: kworker/u8:6 Not tainted 6.10.0-rc6-syzkaller-00067-g8a9c6c40432e #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 06/07/2024 Workqueue: gid-cache-wq netdevice_event_work_handler ==================================================================
netdevsim netdevsim3 netdevsim3 (unregistering): unset [1, 0] type 2 family 0 port 6081 - 0 netdevsim netdevsim3 netdevsim2 (unregistering): unset [1, 0] type 2 family 0 port 6081 - 0 netdevsim netdevsim3 netdevsim1 (unregistering): unset [1, 0] type 2 family 0 port 6081 - 0 netdevsim netdevsim3 netdevsim0 (unregistering): unset [1, 0] type 2 family 0 port 6081 - 0bridge_slave_1: left allmulticast modebridge_slave_1: left promiscuous mode
bridge0: port 2(bridge_slave_1) entered disabled statebridge_slave_0: left allmulticast modebridge_slave_0: left promiscuous mode
bridge0: port 1(bridge_slave_0) entered disabled state
bond0 (unregistering): (slave bond_slave_0): Releasing backup interface bond0 (unregistering): (slave bond_slave_1): Releasing backup interface bond0 (unregistering): Released all slaveshsr_slave_0: left promiscuous modehsr_slave_1: left promiscuous modebatman_adv: batadv0: Interface deactivated: batadv_slave_0batman_adv: batadv0: Removing interface: batadv_slave_0batman_adv: batadv0: Interface deactivated: batadv_slave_1batman_adv: batadv0: Removing interface: batadv_slave_1veth1_macvtap: left promiscuous modeveth0_macvtap: left promiscuous modeveth1_vlan: left promiscuous modeveth0_vlan: left promiscuous mode
team0 (unregistering): Port device team_slave_1 removed
team0 (unregistering): Port device team_slave_0 removed
---
This report is generated by a bot. It may contain errors. See https://goo.gl/tpsmEJ for more information about syzbot. syzbot engineers can be reached at [email protected].
syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title
If you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)
If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-report
If you want to undo deduplication, reply with:
#syz undup
Quoted 19 lines · syzbot
Hello,
syzbot found the following issue on:
HEAD commit: 8a9c6c40432e Merge tag 'io_uring-6.10-20240703' of git://g.. git tree: upstream console output: https://syzkaller.appspot.com/x/log.txt?x=1558eeb9980000 kernel config: https://syzkaller.appspot.com/x/.config?x=5b9537cd00be479e dashboard link: https://syzkaller.appspot.com/bug?extid=113b65786d8662e21ff7 compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40Unfortunately, I don't have any reproducer for this issue yet.
Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/f01153f3ec8e/disk-8a9c6c40.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/04b88663a824/vmlinux-8a9c6c40.xz
kernel image: https://storage.googleapis.com/syzbot-assets/63c518cc63c5/bzImage-8a9c6c40.xz
Quoted 8 lines
IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by [email protected]==================================================================
BUG: KCSAN: data-race in __dev_change_flags / is_upper_ndev_bond_master_filterread-write to
0xffff888112fca0b0of 4 bytes by task 9257 on cpu 1:
__dev_change_flags+0x9a/0x410net/core/dev.c:8755
This:
dev->flags = (flags & (IFF_DEBUG | IFF_NOTRAILERS | IFF_NOARP |
IFF_DYNAMIC | IFF_MULTICAST | IFF_PORTSEL | IFF_AUTOMEDIA)) |
(dev->flags & (IFF_UP | IFF_VOLATILE | IFF_PROMISC |
IFF_ALLMULTI));
lots of potential to be compiled in a way that temporarily exposes dev->flags as not having the first or the second part of flags, even though these flags are supposed to be set before and after the operation.
I guess randomly missing flags can lead to arbitrary bugs elsewhere.
Quoted 95 lines
rtnl_configure_link net/core/rtnetlink.c:3321 [inline] rtnl_newlink_create net/core/rtnetlink.c:3518 [inline] __rtnl_newlink net/core/rtnetlink.c:3730 [inline] rtnl_newlink+0x121e/0x1690 net/core/rtnetlink.c:3743 rtnetlink_rcv_msg+0x85e/0x910 net/core/rtnetlink.c:6635 netlink_rcv_skb+0x12c/0x230 net/netlink/af_netlink.c:2564 rtnetlink_rcv+0x1c/0x30 net/core/rtnetlink.c:6653 netlink_unicast_kernel net/netlink/af_netlink.c:1335 [inline] netlink_unicast+0x58d/0x660 net/netlink/af_netlink.c:1361 netlink_sendmsg+0x5ca/0x6e0 net/netlink/af_netlink.c:1905 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg+0x140/0x180 net/socket.c:745 ____sys_sendmsg+0x312/0x410 net/socket.c:2585 ___sys_sendmsg net/socket.c:2639 [inline] __sys_sendmsg+0x1e9/0x280 net/socket.c:2668 __do_sys_sendmsg net/socket.c:2677 [inline] __se_sys_sendmsg net/socket.c:2675 [inline] __x64_sys_sendmsg+0x46/0x50 net/socket.c:2675 x64_sys_call+0xb25/0x2d70 arch/x86/include/generated/asm/syscalls_64.h:47 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xc9/0x1c0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7fread to
0xffff888112fca0b0of 4 bytes by task 3104 on cpu 0:netif_is_bond_master include/linux/netdevice.h:5020 [inline] is_upper_ndev_bond_master_filter+0x2b/0xb0 drivers/infiniband/core/roce_gid_mgmt.c:275 ib_enum_roce_netdev+0x124/0x1d0 drivers/infiniband/core/device.c:2310 ib_enum_all_roce_netdevs+0x8a/0x100 drivers/infiniband/core/device.c:2337 netdevice_event_work_handler+0x15b/0x3c0 drivers/infiniband/core/roce_gid_mgmt.c:626 process_one_work kernel/workqueue.c:3248 [inline] process_scheduled_works+0x483/0x9a0 kernel/workqueue.c:3329 worker_thread+0x526/0x720 kernel/workqueue.c:3409 kthread+0x1d1/0x210 kernel/kthread.c:389 ret_from_fork+0x4b/0x60 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244value changed:
0x00001002->0x00008002Reported by Kernel Concurrency Sanitizer on:
CPU: 0 PID: 3104 Comm: kworker/u8:6 Not tainted 6.10.0-rc6-syzkaller-00067-g8a9c6c40432e #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 06/07/2024 Workqueue: gid-cache-wq netdevice_event_work_handler ==================================================================netdevsim netdevsim3 netdevsim3 (unregistering): unset [1, 0] type 2 family 0 port 6081 - 0 netdevsim netdevsim3 netdevsim2 (unregistering): unset [1, 0] type 2 family 0 port 6081 - 0 netdevsim netdevsim3 netdevsim1 (unregistering): unset [1, 0] type 2 family 0 port 6081 - 0 netdevsim netdevsim3 netdevsim0 (unregistering): unset [1, 0] type 2 family 0 port 6081 - 0
bridge_slave_1: left allmulticast modebridge_slave_1: left promiscuous mode
bridge0: port 2(bridge_slave_1) entered disabled statebridge_slave_0: left allmulticast modebridge_slave_0: left promiscuous mode
bridge0: port 1(bridge_slave_0) entered disabled state
bond0 (unregistering): (slavebond_slave_0): Releasing backup interface bond0 (unregistering): (slavebond_slave_1): Releasing backup interface bond0 (unregistering): Released all slaveshsr_slave_0: left promiscuous modehsr_slave_1: left promiscuous modebatman_adv: batadv0: Interface deactivated:batadv_slave_0batman_adv: batadv0: Removing interface:batadv_slave_0batman_adv: batadv0: Interface deactivated:batadv_slave_1batman_adv: batadv0: Removing interface:batadv_slave_1veth1_macvtap: left promiscuous modeveth0_macvtap: left promiscuous modeveth1_vlan: left promiscuous modeveth0_vlan: left promiscuous mode
team0 (unregistering): Port deviceteam_slave_1removed
team0 (unregistering): Port deviceteam_slave_0removed---
This report is generated by a bot. It may contain errors. See https://goo.gl/tpsmEJ for more information about syzbot. syzbot engineers can be reached at [email protected].syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-titleIf you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-reportIf you want to undo deduplication, reply with:
#syz undup--
You received this message because you are subscribed to the Google Groups "syzkaller-bugs" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/syzkaller-bugs/000000000000e9f499061c6d4d7a%40google.com.
PATCH 1/1 net: Fix data race around dev->flags in netif_is_bond_master
In reply to syzbotBUG: KCSAN: data-race in __dev_change_flags / is_upper_ndev_bond_master_filter
read-write to 0xffff888112d970b0 of 4 bytes by task 4888 on cpu 0:
__dev_change_flags+0x9a/0x410 net/core/dev.c:8755 rtnl_configure_link net/core/rtnetlink.c:3321 [inline] rtnl_newlink_create net/core/rtnetlink.c:3518 [inline] __rtnl_newlink net/core/rtnetlink.c:3730 [inline] rtnl_newlink+0x121e/0x1690 net/core/rtnetlink.c:3743 rtnetlink_rcv_msg+0x85e/0x910 net/core/rtnetlink.c:6635 netlink_rcv_skb+0x12c/0x230 net/netlink/af_netlink.c:2564 rtnetlink_rcv+0x1c/0x30 net/core/rtnetlink.c:6653 netlink_unicast_kernel net/netlink/af_netlink.c:1335 [inline] netlink_unicast+0x58d/0x660 net/netlink/af_netlink.c:1361 netlink_sendmsg+0x5ca/0x6e0 net/netlink/af_netlink.c:1905 sock_sendmsg_nosec net/socket.c:730 [inline] __sock_sendmsg+0x140/0x180 net/socket.c:745 ____sys_sendmsg+0x312/0x410 net/socket.c:2585 ___sys_sendmsg net/socket.c:2639 [inline] __sys_sendmsg+0x1e9/0x280 net/socket.c:2668 __do_sys_sendmsg net/socket.c:2677 [inline] __se_sys_sendmsg net/socket.c:2675 [inline] __x64_sys_sendmsg+0x46/0x50 net/socket.c:2675 x64_sys_call+0xb25/0x2d70 arch/x86/include/generated/asm/syscalls_64.h:47 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xc9/0x1c0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f
read to 0xffff888112d970b0 of 4 bytes by task 11 on cpu 1:
netif_is_bond_master include/linux/netdevice.h:5020 [inline] is_upper_ndev_bond_master_filter+0x2b/0xb0 drivers/infiniband/core/roce_gid_mgmt.c:275 ib_enum_roce_netdev+0x124/0x1d0 drivers/infiniband/core/device.c:2310 ib_enum_all_roce_netdevs+0x8a/0x100 drivers/infiniband/core/device.c:2337 netdevice_event_work_handler+0x15b/0x3c0 drivers/infiniband/core/roce_gid_mgmt.c:626 process_one_work kernel/workqueue.c:3248 [inline] process_scheduled_works+0x483/0x9a0 kernel/workqueue.c:3329 worker_thread+0x526/0x720 kernel/workqueue.c:3409 kthread+0x1d1/0x210 kernel/kthread.c:389 ret_from_fork+0x4b/0x60 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244
value changed: 0x00001002 -> 0x00000202
According to KCSAN report, there is a read/write race between __dev_change_flags and netif_is_bond_master for dev->flags. Therefore, should change to use READ_ONCE() when reading dev->flags.
--- include/linux/netdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/netdevice.h +1 -1
@@ -5043,7 +5043,7 @@ static inline bool netif_is_macvlan_port(const struct net_device *dev)static inline bool netif_is_bond_master (const struct net_device *dev) { -return dev->flags &IFF_MASTER && dev->priv_flags &IFF_BONDING ; +return READ_ONCE (dev->flags) &IFF_MASTER && dev->priv_flags &IFF_BONDING ; }static inline bool netif_is_bond_slave (const struct net_device *dev)
--
Jakub Kicinski
maintainerRe: [PATCH net] net: Fix data race around dev->flags in netif_is_bond_master
Quoted 3 lines · On Sun, 28 Jul 2024 19:54:29 +0900 Jeongjun Park wrote:
According to KCSAN report, there is a read/write race between
__dev_change_flagsandnetif_is_bond_masterfordev->flags. Therefore, should change to useREAD_ONCE()when readingdev->flags.
That will certainly silence the warning, but doesn't prove the change is correct.
Signature · 1 line
pw-bot: cr