linux-toradex.git/net/openvswitch, branch master Linux kernel for Apalis and Colibri modules openvswitch: vport: fix self-deadlock on release of tunnel ports 2026-05-05T13:19:37+00:00 Ilya Maximets i.maximets@ovn.org 2026-04-30T23:38:37+00:00 aa69918bd418e700309fdd08509dba324fb24296 vports are used concurrently and protected by RCU, so netdev_put() must happen after the RCU grace period. So, either in an RCU call or after the synchronize_net(). The rtnl_delete_link() must happen under RTNL and so can't be executed in RCU context. Calling synchronize_net() while holding RTNL is not a good idea for performance and system stability under load in general, so calling netdev_put() in RCU call is the right solution here. However, when the device is deleted, rtnl_unlock() will call netdev_run_todo() and block until all the references are gone. In the current code this means that we never reach the call_rcu() and the vport is never freed and the reference is never released, causing a self-deadlock on device removal. Fix that by moving the rcu_call() before the rtnl_unlock(), so the scheduled RCU callback will be executed when synchronize_net() is called from the rtnl_unlock()->netdev_run_todo() while the RTNL itself is already released. Fixes: 6931d21f87bc ("openvswitch: defer tunnel netdev_put to RCU release") Cc: stable@vger.kernel.org Acked-by: Eelco Chaudron <echaudro@redhat.com> Signed-off-by: Ilya Maximets <i.maximets@ovn.org> Acked-by: Aaron Conole <aconole@redhat.com> Link: https://patch.msgid.link/20260430233848.440994-2-i.maximets@ovn.org Signed-off-by: Paolo Abeni <pabeni@redhat.com> 
vports are used concurrently and protected by RCU, so netdev_put()
must happen after the RCU grace period.  So, either in an RCU call or
after the synchronize_net().  The rtnl_delete_link() must happen under
RTNL and so can't be executed in RCU context.  Calling synchronize_net()
while holding RTNL is not a good idea for performance and system
stability under load in general, so calling netdev_put() in RCU call
is the right solution here.

However,
when the device is deleted, rtnl_unlock() will call netdev_run_todo()
and block until all the references are gone.  In the current code this
means that we never reach the call_rcu() and the vport is never freed
and the reference is never released, causing a self-deadlock on device
removal.

Fix that by moving the rcu_call() before the rtnl_unlock(), so the
scheduled RCU callback will be executed when synchronize_net() is
called from the rtnl_unlock()->netdev_run_todo() while the RTNL itself
is already released.

Fixes: 6931d21f87bc ("openvswitch: defer tunnel netdev_put to RCU release")
Cc: stable@vger.kernel.org
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Acked-by: Aaron Conole <aconole@redhat.com>
Link: https://patch.msgid.link/20260430233848.440994-2-i.maximets@ovn.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
openvswitch: vport: fix race between tunnel creation and linking 2026-05-05T13:14:33+00:00 Ilya Maximets i.maximets@ovn.org 2026-04-30T21:32:50+00:00 83861c48ba122f85cc8384780764b3a791341678 When a tunnel vport is created it first creates the tunnel device, e.g., with geneve_dev_create_fb(), then it calls ovs_netdev_link() to take a reference and link it to the device that represents openvswitch datapath. The creation of the device is happening under RTNL, but then RTNL is released and re-acquired to find the device by name. It is technically possible for the tunnel device to be re-named or deleted within that window while RTNL is not held, and some other device created in its place. This will cause a non-tunnel device to be referenced in the vport and tunnel-specific functions used on it, e.g. vxlan_get_options() that directly casts the private netdev data into a struct vxlan_dev causing an invalid memory access: BUG: KASAN: slab-use-after-free in vxlan_get_options+0x323/0x3a0 vxlan_get_options+0x323/0x3a0 ovs_vport_cmd_new+0x6e3/0xd30 Fix that by taking a reference to the just created device before releasing RTNL. This ensures that the device in the vport is always the one that was just created. The search by name is only needed for a standard vport-netdev that links pre-existing devices, so that functionality and device type checks are moved to netdev_create(). It is also awkward that ovs_netdev_link() takes ownership of the vport and destroys it on failure. It doesn't know the type of the port it is dealing with, so we need to pass down the indicator that it's a tunnel, so the link can be properly deleted on failure. It's possible to refactor the logic to make the ovs_netdev_link() do only the linking part and let the callers perform a proper destruction, but it will be much more code for each legacy tunnel port type, so it is not worth it for the bug fix. Fixes: 614732eaa12d ("openvswitch: Use regular VXLAN net_device device") Reported-by: Yuan Tan <tanyuan98@outlook.com> Reported-by: Yifan Wu <yifanwucs@gmail.com> Reported-by: Juefei Pu <tomapufckgml@gmail.com> Reported-by: Xin Liu <bird@lzu.edu.cn> Reported-by: Yang Yang <n05ec@lzu.edu.cn> Signed-off-by: Ilya Maximets <i.maximets@ovn.org> Acked-by: Eelco Chaudron <echaudro@redhat.com> Link: https://patch.msgid.link/20260430213349.407991-1-i.maximets@ovn.org Signed-off-by: Paolo Abeni <pabeni@redhat.com> 
When a tunnel vport is created it first creates the tunnel device, e.g.,
with geneve_dev_create_fb(), then it calls ovs_netdev_link() to take a
reference and link it to the device that represents openvswitch datapath.

The creation of the device is happening under RTNL, but then RTNL is
released and re-acquired to find the device by name.  It is technically
possible for the tunnel device to be re-named or deleted within that
window while RTNL is not held, and some other device created in its
place.  This will cause a non-tunnel device to be referenced in the
vport and tunnel-specific functions used on it, e.g. vxlan_get_options()
that directly casts the private netdev data into a struct vxlan_dev
causing an invalid memory access:

 BUG: KASAN: slab-use-after-free in vxlan_get_options+0x323/0x3a0
  vxlan_get_options+0x323/0x3a0
  ovs_vport_cmd_new+0x6e3/0xd30

Fix that by taking a reference to the just created device before
releasing RTNL.  This ensures that the device in the vport is always
the one that was just created.  The search by name is only needed
for a standard vport-netdev that links pre-existing devices, so that
functionality and device type checks are moved to netdev_create().

It is also awkward that ovs_netdev_link() takes ownership of the vport
and destroys it on failure.  It doesn't know the type of the port it is
dealing with, so we need to pass down the indicator that it's a tunnel,
so the link can be properly deleted on failure.

It's possible to refactor the logic to make the ovs_netdev_link() do
only the linking part and let the callers perform a proper destruction,
but it will be much more code for each legacy tunnel port type, so it
is not worth it for the bug fix.

Fixes: 614732eaa12d ("openvswitch: Use regular VXLAN net_device device")
Reported-by: Yuan Tan <tanyuan98@outlook.com>
Reported-by: Yifan Wu <yifanwucs@gmail.com>
Reported-by: Juefei Pu <tomapufckgml@gmail.com>
Reported-by: Xin Liu <bird@lzu.edu.cn>
Reported-by: Yang Yang <n05ec@lzu.edu.cn>
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Link: https://patch.msgid.link/20260430213349.407991-1-i.maximets@ovn.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
openvswitch: cap upcall PID array size and pre-size vport replies 2026-04-20T18:43:04+00:00 Weiming Shi bestswngs@gmail.com 2026-04-16T02:46:54+00:00 2091c6aa0df6aba47deb5c8ab232b1cb60af3519 The vport netlink reply helpers allocate a fixed-size skb with nlmsg_new(NLMSG_DEFAULT_SIZE, ...) but serialize the full upcall PID array via ovs_vport_get_upcall_portids(). Since ovs_vport_set_upcall_portids() accepts any non-zero multiple of sizeof(u32) with no upper bound, a CAP_NET_ADMIN user can install a PID array large enough to overflow the reply buffer, causing nla_put() to fail with -EMSGSIZE and hitting BUG_ON(err < 0). On systems with unprivileged user namespaces enabled (e.g., Ubuntu default), this is reachable via unshare -Urn since OVS vport mutation operations use GENL_UNS_ADMIN_PERM. kernel BUG at net/openvswitch/datapath.c:2414! Oops: invalid opcode: 0000 [#1] SMP KASAN NOPTI CPU: 1 UID: 0 PID: 65 Comm: poc Not tainted 7.0.0-rc7-00195-geb216e422044 #1 RIP: 0010:ovs_vport_cmd_set+0x34c/0x400 Call Trace: <TASK> genl_family_rcv_msg_doit (net/netlink/genetlink.c:1116) genl_rcv_msg (net/netlink/genetlink.c:1194) netlink_rcv_skb (net/netlink/af_netlink.c:2550) genl_rcv (net/netlink/genetlink.c:1219) netlink_unicast (net/netlink/af_netlink.c:1344) netlink_sendmsg (net/netlink/af_netlink.c:1894) __sys_sendto (net/socket.c:2206) __x64_sys_sendto (net/socket.c:2209) do_syscall_64 (arch/x86/entry/syscall_64.c:63) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) </TASK> Kernel panic - not syncing: Fatal exception Reject attempts to set more PIDs than nr_cpu_ids in ovs_vport_set_upcall_portids(), and pre-compute the worst-case reply size in ovs_vport_cmd_msg_size() based on that bound, similar to the existing ovs_dp_cmd_msg_size(). nr_cpu_ids matches the cap already used by the per-CPU dispatch configuration on the datapath side (ovs_dp_cmd_fill_info() serialises at most nr_cpu_ids PIDs), so the two sides stay consistent. Fixes: 5cd667b0a456 ("openvswitch: Allow each vport to have an array of 'port_id's.") Reported-by: Xiang Mei <xmei5@asu.edu> Assisted-by: Claude:claude-opus-4-6 Signed-off-by: Weiming Shi <bestswngs@gmail.com> Reviewed-by: Ilya Maximets <i.maximets@ovn.org> Link: https://patch.msgid.link/20260416024653.153456-2-bestswngs@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> 
The vport netlink reply helpers allocate a fixed-size skb with
nlmsg_new(NLMSG_DEFAULT_SIZE, ...) but serialize the full upcall PID
array via ovs_vport_get_upcall_portids().  Since
ovs_vport_set_upcall_portids() accepts any non-zero multiple of
sizeof(u32) with no upper bound, a CAP_NET_ADMIN user can install a PID
array large enough to overflow the reply buffer, causing nla_put() to
fail with -EMSGSIZE and hitting BUG_ON(err < 0).  On systems with
unprivileged user namespaces enabled (e.g., Ubuntu default), this is
reachable via unshare -Urn since OVS vport mutation operations use
GENL_UNS_ADMIN_PERM.

 kernel BUG at net/openvswitch/datapath.c:2414!
 Oops: invalid opcode: 0000 [#1] SMP KASAN NOPTI
 CPU: 1 UID: 0 PID: 65 Comm: poc Not tainted 7.0.0-rc7-00195-geb216e422044 #1
 RIP: 0010:ovs_vport_cmd_set+0x34c/0x400
 Call Trace:
  <TASK>
  genl_family_rcv_msg_doit (net/netlink/genetlink.c:1116)
  genl_rcv_msg (net/netlink/genetlink.c:1194)
  netlink_rcv_skb (net/netlink/af_netlink.c:2550)
  genl_rcv (net/netlink/genetlink.c:1219)
  netlink_unicast (net/netlink/af_netlink.c:1344)
  netlink_sendmsg (net/netlink/af_netlink.c:1894)
  __sys_sendto (net/socket.c:2206)
  __x64_sys_sendto (net/socket.c:2209)
  do_syscall_64 (arch/x86/entry/syscall_64.c:63)
  entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
  </TASK>
 Kernel panic - not syncing: Fatal exception

Reject attempts to set more PIDs than nr_cpu_ids in
ovs_vport_set_upcall_portids(), and pre-compute the worst-case reply
size in ovs_vport_cmd_msg_size() based on that bound, similar to the
existing ovs_dp_cmd_msg_size().  nr_cpu_ids matches the cap already
used by the per-CPU dispatch configuration on the datapath side
(ovs_dp_cmd_fill_info() serialises at most nr_cpu_ids PIDs), so the
two sides stay consistent.

Fixes: 5cd667b0a456 ("openvswitch: Allow each vport to have an array of 'port_id's.")
Reported-by: Xiang Mei <xmei5@asu.edu>
Assisted-by: Claude:claude-opus-4-6
Signed-off-by: Weiming Shi <bestswngs@gmail.com>
Reviewed-by: Ilya Maximets <i.maximets@ovn.org>
Link: https://patch.msgid.link/20260416024653.153456-2-bestswngs@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net: use get_random_u{16,32,64}() where appropriate 2026-04-10T02:27:43+00:00 David Carlier devnexen@gmail.com 2026-04-07T15:07:58+00:00 9addea5d44b69d377ba97a36f7a19e1097969e18 Use the typed random integer helpers instead of get_random_bytes() when filling a single integer variable. The helpers return the value directly, require no pointer or size argument, and better express intent. Skipped sites writing into __be16 (netdevsim) and __le64 (ceph) fields where a direct assignment would trigger sparse endianness warnings. Signed-off-by: David Carlier <devnexen@gmail.com> Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/20260407150758.5889-1-devnexen@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> 
Use the typed random integer helpers instead of
get_random_bytes() when filling a single integer variable.
The helpers return the value directly, require no pointer
or size argument, and better express intent.

Skipped sites writing into __be16 (netdevsim) and __le64
(ceph) fields where a direct assignment would trigger
sparse endianness warnings.

Signed-off-by: David Carlier <devnexen@gmail.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260407150758.5889-1-devnexen@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net: convert remaining ipv6_stub users to direct function calls 2026-03-29T18:21:23+00:00 Fernando Fernandez Mancera fmancera@suse.de 2026-03-25T12:08:49+00:00 d76f6b170a10414a9c674b6db4bd0473b1b30050 As IPv6 is built-in only, the ipv6_stub infrastructure is no longer necessary. Convert remaining ipv6_stub users to make direct function calls. The fallback functions introduced previously will prevent linkage errors when CONFIG_IPV6 is disabled. Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de> Tested-by: Ricardo B. Marlière <rbm@suse.com> Link: https://patch.msgid.link/20260325120928.15848-9-fmancera@suse.de Signed-off-by: Jakub Kicinski <kuba@kernel.org> 
As IPv6 is built-in only, the ipv6_stub infrastructure is no longer
necessary.

Convert remaining ipv6_stub users to make direct function calls. The
fallback functions introduced previously will prevent linkage errors
when CONFIG_IPV6 is disabled.

Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
Tested-by: Ricardo B. Marlière <rbm@suse.com>
Link: https://patch.msgid.link/20260325120928.15848-9-fmancera@suse.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
openvswitch: validate MPLS set/set_masked payload length 2026-03-21T01:37:31+00:00 Yang Yang n05ec@lzu.edu.cn 2026-03-19T08:02:27+00:00 546b68ac893595877ffbd7751e5c55fd1c43ede6 validate_set() accepted OVS_KEY_ATTR_MPLS as variable-sized payload for SET/SET_MASKED actions. In action handling, OVS expects fixed-size MPLS key data (struct ovs_key_mpls). Use the already normalized key_len (masked case included) and reject non-matching MPLS action key sizes. Reject invalid MPLS action payload lengths early. Fixes: fbdcdd78da7c ("Change in Openvswitch to support MPLS label depth of 3 in ingress direction") Reported-by: Yifan Wu <yifanwucs@gmail.com> Reported-by: Juefei Pu <tomapufckgml@gmail.com> Tested-by: Ao Zhou <n05ec@lzu.edu.cn> Co-developed-by: Yuan Tan <tanyuan98@outlook.com> Signed-off-by: Yuan Tan <tanyuan98@outlook.com> Suggested-by: Xin Liu <bird@lzu.edu.cn> Signed-off-by: Yang Yang <n05ec@lzu.edu.cn> Reviewed-by: Ilya Maximets <i.maximets@ovn.org> Link: https://patch.msgid.link/20260319080228.3423307-1-n05ec@lzu.edu.cn Signed-off-by: Jakub Kicinski <kuba@kernel.org> 
validate_set() accepted OVS_KEY_ATTR_MPLS as variable-sized payload for
SET/SET_MASKED actions. In action handling, OVS expects fixed-size
MPLS key data (struct ovs_key_mpls).

Use the already normalized key_len (masked case included) and reject
non-matching MPLS action key sizes.

Reject invalid MPLS action payload lengths early.

Fixes: fbdcdd78da7c ("Change in Openvswitch to support MPLS label depth of 3 in ingress direction")
Reported-by: Yifan Wu <yifanwucs@gmail.com>
Reported-by: Juefei Pu <tomapufckgml@gmail.com>
Tested-by: Ao Zhou <n05ec@lzu.edu.cn>
Co-developed-by: Yuan Tan <tanyuan98@outlook.com>
Signed-off-by: Yuan Tan <tanyuan98@outlook.com>
Suggested-by: Xin Liu <bird@lzu.edu.cn>
Signed-off-by: Yang Yang <n05ec@lzu.edu.cn>
Reviewed-by: Ilya Maximets <i.maximets@ovn.org>
Link: https://patch.msgid.link/20260319080228.3423307-1-n05ec@lzu.edu.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
openvswitch: defer tunnel netdev_put to RCU release 2026-03-21T01:36:29+00:00 Yang Yang n05ec@lzu.edu.cn 2026-03-19T07:42:41+00:00 6931d21f87bc6d657f145798fad0bf077b82486c ovs_netdev_tunnel_destroy() may run after NETDEV_UNREGISTER already detached the device. Dropping the netdev reference in destroy can race with concurrent readers that still observe vport->dev. Do not release vport->dev in ovs_netdev_tunnel_destroy(). Instead, let vport_netdev_free() drop the reference from the RCU callback, matching the non-tunnel destroy path and avoiding additional synchronization under RTNL. Fixes: a9020fde67a6 ("openvswitch: Move tunnel destroy function to oppenvswitch module.") Reported-by: Yifan Wu <yifanwucs@gmail.com> Reported-by: Juefei Pu <tomapufckgml@gmail.com> Tested-by: Ao Zhou <n05ec@lzu.edu.cn> Co-developed-by: Yuan Tan <tanyuan98@outlook.com> Signed-off-by: Yuan Tan <tanyuan98@outlook.com> Suggested-by: Xin Liu <bird@lzu.edu.cn> Signed-off-by: Yang Yang <n05ec@lzu.edu.cn> Reviewed-by: Ilya Maximets <i.maximets@ovn.org> Link: https://patch.msgid.link/20260319074241.3405262-1-n05ec@lzu.edu.cn Signed-off-by: Jakub Kicinski <kuba@kernel.org> 
ovs_netdev_tunnel_destroy() may run after NETDEV_UNREGISTER already
detached the device. Dropping the netdev reference in destroy can race
with concurrent readers that still observe vport->dev.

Do not release vport->dev in ovs_netdev_tunnel_destroy(). Instead, let
vport_netdev_free() drop the reference from the RCU callback, matching
the non-tunnel destroy path and avoiding additional synchronization
under RTNL.

Fixes: a9020fde67a6 ("openvswitch: Move tunnel destroy function to oppenvswitch module.")
Reported-by: Yifan Wu <yifanwucs@gmail.com>
Reported-by: Juefei Pu <tomapufckgml@gmail.com>
Tested-by: Ao Zhou <n05ec@lzu.edu.cn>
Co-developed-by: Yuan Tan <tanyuan98@outlook.com>
Signed-off-by: Yuan Tan <tanyuan98@outlook.com>
Suggested-by: Xin Liu <bird@lzu.edu.cn>
Signed-off-by: Yang Yang <n05ec@lzu.edu.cn>
Reviewed-by: Ilya Maximets <i.maximets@ovn.org>
Link: https://patch.msgid.link/20260319074241.3405262-1-n05ec@lzu.edu.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
net: openvswitch: Avoid releasing netdev before teardown completes 2026-03-20T00:15:37+00:00 Toke Høiland-Jørgensen toke@redhat.com 2026-03-18T15:55:51+00:00 7c770dadfda5cbbde6aa3c4363ed513f1d212bf8 The patch cited in the Fixes tag below changed the teardown code for OVS ports to no longer unconditionally take the RTNL. After this change, the netdev_destroy() callback can proceed immediately to the call_rcu() invocation if the IFF_OVS_DATAPATH flag is already cleared on the netdev. The ovs_netdev_detach_dev() function clears the flag before completing the unregistration, and if it gets preempted after clearing the flag (as can happen on an -rt kernel), netdev_destroy() can complete and the device can be freed before the unregistration completes. This leads to a splat like: [ 998.393867] Oops: general protection fault, probably for non-canonical address 0xff00000001000239: 0000 [#1] SMP PTI [ 998.393877] CPU: 42 UID: 0 PID: 55177 Comm: ip Kdump: loaded Not tainted 6.12.0-211.1.1.el10_2.x86_64+rt #1 PREEMPT_RT [ 998.393886] Hardware name: Dell Inc. PowerEdge R740/0JMK61, BIOS 2.24.0 03/27/2025 [ 998.393889] RIP: 0010:dev_set_promiscuity+0x8d/0xa0 [ 998.393901] Code: 00 00 75 d8 48 8b 53 08 48 83 ba b0 02 00 00 00 75 ca 48 83 c4 08 5b c3 cc cc cc cc 48 83 bf 48 09 00 00 00 75 91 48 8b 47 08 <48> 83 b8 b0 02 00 00 00 74 97 eb 81 0f 1f 80 00 00 00 00 90 90 90 [ 998.393906] RSP: 0018:ffffce5864a5f6a0 EFLAGS: 00010246 [ 998.393912] RAX: ff00000000ffff89 RBX: ffff894d0adf5a05 RCX: 0000000000000000 [ 998.393917] RDX: 0000000000000000 RSI: 00000000ffffffff RDI: ffff894d0adf5a05 [ 998.393921] RBP: ffff894d19252000 R08: ffff894d19252000 R09: 0000000000000000 [ 998.393924] R10: ffff894d19252000 R11: ffff894d192521b8 R12: 0000000000000006 [ 998.393927] R13: ffffce5864a5f738 R14: 00000000ffffffe2 R15: 0000000000000000 [ 998.393931] FS: 00007fad61971800(0000) GS:ffff894cc0140000(0000) knlGS:0000000000000000 [ 998.393936] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 998.393940] CR2: 000055df0a2a6e40 CR3: 000000011c7fe003 CR4: 00000000007726f0 [ 998.393944] PKRU: 55555554 [ 998.393946] Call Trace: [ 998.393949] <TASK> [ 998.393952] ? show_trace_log_lvl+0x1b0/0x2f0 [ 998.393961] ? show_trace_log_lvl+0x1b0/0x2f0 [ 998.393975] ? dp_device_event+0x41/0x80 [openvswitch] [ 998.394009] ? __die_body.cold+0x8/0x12 [ 998.394016] ? die_addr+0x3c/0x60 [ 998.394027] ? exc_general_protection+0x16d/0x390 [ 998.394042] ? asm_exc_general_protection+0x26/0x30 [ 998.394058] ? dev_set_promiscuity+0x8d/0xa0 [ 998.394066] ? ovs_netdev_detach_dev+0x3a/0x80 [openvswitch] [ 998.394092] dp_device_event+0x41/0x80 [openvswitch] [ 998.394102] notifier_call_chain+0x5a/0xd0 [ 998.394106] unregister_netdevice_many_notify+0x51b/0xa60 [ 998.394110] rtnl_dellink+0x169/0x3e0 [ 998.394121] ? rt_mutex_slowlock.constprop.0+0x95/0xd0 [ 998.394125] rtnetlink_rcv_msg+0x142/0x3f0 [ 998.394128] ? avc_has_perm_noaudit+0x69/0xf0 [ 998.394130] ? __pfx_rtnetlink_rcv_msg+0x10/0x10 [ 998.394132] netlink_rcv_skb+0x50/0x100 [ 998.394138] netlink_unicast+0x292/0x3f0 [ 998.394141] netlink_sendmsg+0x21b/0x470 [ 998.394145] ____sys_sendmsg+0x39d/0x3d0 [ 998.394149] ___sys_sendmsg+0x9a/0xe0 [ 998.394156] __sys_sendmsg+0x7a/0xd0 [ 998.394160] do_syscall_64+0x7f/0x170 [ 998.394162] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 998.394165] RIP: 0033:0x7fad61bf4724 [ 998.394188] Code: 89 02 b8 ff ff ff ff eb bb 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3 0f 1e fa 80 3d c5 e9 0c 00 00 74 13 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 54 c3 0f 1f 00 48 83 ec 28 89 54 24 1c 48 89 [ 998.394189] RSP: 002b:00007ffd7e2f7cb8 EFLAGS: 00000202 ORIG_RAX: 000000000000002e [ 998.394191] RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007fad61bf4724 [ 998.394193] RDX: 0000000000000000 RSI: 00007ffd7e2f7d20 RDI: 0000000000000003 [ 998.394194] RBP: 00007ffd7e2f7d90 R08: 0000000000000010 R09: 000000000000003f [ 998.394195] R10: 000055df11558010 R11: 0000000000000202 R12: 00007ffd7e2f8380 [ 998.394196] R13: 0000000069b233d7 R14: 000055df0a256040 R15: 0000000000000000 [ 998.394200] </TASK> To fix this, reorder the operations in ovs_netdev_detach_dev() to only clear the flag after completing the other operations, and introduce an smp_wmb() to make the ordering requirement explicit. The smp_wmb() is paired with a full smp_mb() in netdev_destroy() to make sure the call_rcu() invocation does not happen before the unregister operations are visible. Reported-by: Minxi Hou <mhou@redhat.com> Tested-by: Minxi Hou <mhou@redhat.com> Fixes: 549822767630 ("net: openvswitch: Avoid needlessly taking the RTNL on vport destroy") Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://patch.msgid.link/20260318155554.1133405-1-toke@redhat.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> 
The patch cited in the Fixes tag below changed the teardown code for
OVS ports to no longer unconditionally take the RTNL. After this change,
the netdev_destroy() callback can proceed immediately to the call_rcu()
invocation if the IFF_OVS_DATAPATH flag is already cleared on the
netdev.

The ovs_netdev_detach_dev() function clears the flag before completing
the unregistration, and if it gets preempted after clearing the flag (as
can happen on an -rt kernel), netdev_destroy() can complete and the
device can be freed before the unregistration completes. This leads to a
splat like:

[  998.393867] Oops: general protection fault, probably for non-canonical address 0xff00000001000239: 0000 [#1] SMP PTI
[  998.393877] CPU: 42 UID: 0 PID: 55177 Comm: ip Kdump: loaded Not tainted 6.12.0-211.1.1.el10_2.x86_64+rt #1 PREEMPT_RT
[  998.393886] Hardware name: Dell Inc. PowerEdge R740/0JMK61, BIOS 2.24.0 03/27/2025
[  998.393889] RIP: 0010:dev_set_promiscuity+0x8d/0xa0
[  998.393901] Code: 00 00 75 d8 48 8b 53 08 48 83 ba b0 02 00 00 00 75 ca 48 83 c4 08 5b c3 cc cc cc cc 48 83 bf 48 09 00 00 00 75 91 48 8b 47 08 <48> 83 b8 b0 02 00 00 00 74 97 eb 81 0f 1f 80 00 00 00 00 90 90 90
[  998.393906] RSP: 0018:ffffce5864a5f6a0 EFLAGS: 00010246
[  998.393912] RAX: ff00000000ffff89 RBX: ffff894d0adf5a05 RCX: 0000000000000000
[  998.393917] RDX: 0000000000000000 RSI: 00000000ffffffff RDI: ffff894d0adf5a05
[  998.393921] RBP: ffff894d19252000 R08: ffff894d19252000 R09: 0000000000000000
[  998.393924] R10: ffff894d19252000 R11: ffff894d192521b8 R12: 0000000000000006
[  998.393927] R13: ffffce5864a5f738 R14: 00000000ffffffe2 R15: 0000000000000000
[  998.393931] FS:  00007fad61971800(0000) GS:ffff894cc0140000(0000) knlGS:0000000000000000
[  998.393936] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  998.393940] CR2: 000055df0a2a6e40 CR3: 000000011c7fe003 CR4: 00000000007726f0
[  998.393944] PKRU: 55555554
[  998.393946] Call Trace:
[  998.393949]  <TASK>
[  998.393952]  ? show_trace_log_lvl+0x1b0/0x2f0
[  998.393961]  ? show_trace_log_lvl+0x1b0/0x2f0
[  998.393975]  ? dp_device_event+0x41/0x80 [openvswitch]
[  998.394009]  ? __die_body.cold+0x8/0x12
[  998.394016]  ? die_addr+0x3c/0x60
[  998.394027]  ? exc_general_protection+0x16d/0x390
[  998.394042]  ? asm_exc_general_protection+0x26/0x30
[  998.394058]  ? dev_set_promiscuity+0x8d/0xa0
[  998.394066]  ? ovs_netdev_detach_dev+0x3a/0x80 [openvswitch]
[  998.394092]  dp_device_event+0x41/0x80 [openvswitch]
[  998.394102]  notifier_call_chain+0x5a/0xd0
[  998.394106]  unregister_netdevice_many_notify+0x51b/0xa60
[  998.394110]  rtnl_dellink+0x169/0x3e0
[  998.394121]  ? rt_mutex_slowlock.constprop.0+0x95/0xd0
[  998.394125]  rtnetlink_rcv_msg+0x142/0x3f0
[  998.394128]  ? avc_has_perm_noaudit+0x69/0xf0
[  998.394130]  ? __pfx_rtnetlink_rcv_msg+0x10/0x10
[  998.394132]  netlink_rcv_skb+0x50/0x100
[  998.394138]  netlink_unicast+0x292/0x3f0
[  998.394141]  netlink_sendmsg+0x21b/0x470
[  998.394145]  ____sys_sendmsg+0x39d/0x3d0
[  998.394149]  ___sys_sendmsg+0x9a/0xe0
[  998.394156]  __sys_sendmsg+0x7a/0xd0
[  998.394160]  do_syscall_64+0x7f/0x170
[  998.394162]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  998.394165] RIP: 0033:0x7fad61bf4724
[  998.394188] Code: 89 02 b8 ff ff ff ff eb bb 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3 0f 1e fa 80 3d c5 e9 0c 00 00 74 13 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 54 c3 0f 1f 00 48 83 ec 28 89 54 24 1c 48 89
[  998.394189] RSP: 002b:00007ffd7e2f7cb8 EFLAGS: 00000202 ORIG_RAX: 000000000000002e
[  998.394191] RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007fad61bf4724
[  998.394193] RDX: 0000000000000000 RSI: 00007ffd7e2f7d20 RDI: 0000000000000003
[  998.394194] RBP: 00007ffd7e2f7d90 R08: 0000000000000010 R09: 000000000000003f
[  998.394195] R10: 000055df11558010 R11: 0000000000000202 R12: 00007ffd7e2f8380
[  998.394196] R13: 0000000069b233d7 R14: 000055df0a256040 R15: 0000000000000000
[  998.394200]  </TASK>

To fix this, reorder the operations in ovs_netdev_detach_dev() to only
clear the flag after completing the other operations, and introduce an
smp_wmb() to make the ordering requirement explicit. The smp_wmb() is
paired with a full smp_mb() in netdev_destroy() to make sure the
call_rcu() invocation does not happen before the unregister operations
are visible.

Reported-by: Minxi Hou <mhou@redhat.com>
Tested-by: Minxi Hou <mhou@redhat.com>
Fixes: 549822767630 ("net: openvswitch: Avoid needlessly taking the RTNL on vport destroy")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://patch.msgid.link/20260318155554.1133405-1-toke@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Convert more 'alloc_obj' cases to default GFP_KERNEL arguments 2026-02-22T04:03:00+00:00 Linus Torvalds torvalds@linux-foundation.org 2026-02-22T04:03:00+00:00 32a92f8c89326985e05dce8b22d3f0aa07a3e1bd This converts some of the visually simpler cases that have been split over multiple lines. I only did the ones that are easy to verify the resulting diff by having just that final GFP_KERNEL argument on the next line. Somebody should probably do a proper coccinelle script for this, but for me the trivial script actually resulted in an assertion failure in the middle of the script. I probably had made it a bit _too_ trivial. So after fighting that far a while I decided to just do some of the syntactically simpler cases with variations of the previous 'sed' scripts. The more syntactically complex multi-line cases would mostly really want whitespace cleanup anyway. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> 
This converts some of the visually simpler cases that have been split
over multiple lines.  I only did the ones that are easy to verify the
resulting diff by having just that final GFP_KERNEL argument on the next
line.

Somebody should probably do a proper coccinelle script for this, but for
me the trivial script actually resulted in an assertion failure in the
middle of the script.  I probably had made it a bit _too_ trivial.

So after fighting that far a while I decided to just do some of the
syntactically simpler cases with variations of the previous 'sed'
scripts.

The more syntactically complex multi-line cases would mostly really want
whitespace cleanup anyway.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Convert 'alloc_flex' family to use the new default GFP_KERNEL argument 2026-02-22T01:09:51+00:00 Linus Torvalds torvalds@linux-foundation.org 2026-02-22T01:06:51+00:00 323bbfcf1ef8836d0d2ad9e2c1f1c684f0e3b5b3 This is the exact same thing as the 'alloc_obj()' version, only much smaller because there are a lot fewer users of the *alloc_flex() interface. As with alloc_obj() version, this was done entirely with mindless brute force, using the same script, except using 'flex' in the pattern rather than 'objs*'. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> 
This is the exact same thing as the 'alloc_obj()' version, only much
smaller because there are a lot fewer users of the *alloc_flex()
interface.

As with alloc_obj() version, this was done entirely with mindless brute
force, using the same script, except using 'flex' in the pattern rather
than 'objs*'.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>