February 2024 - Kernel - mailweb.openeuler.org

[PATCH openEuler-1.0-LTS v3 0/2] Save and restore msg_namelen in sock_sendmsg
by Liu Jian 02 Feb '24

02 Feb '24

v2: add cover letter v3: fix one compile error Jordan Rife (1): net: prevent rewrite of msg_name in sock_sendmsg() Marc Dionne (1): net: Save and restore msg_namelen in sock_sendmsg net/socket.c | 27 +++++++++++++++++++++++---- 1 file changed, 23 insertions(+), 4 deletions(-) -- 2.34.1

2 3

[PATCH openEuler-1.0-LTS v2] drm/atomic: Fix potential use-after-free in nonblocking commits
by Guo Mengqi 02 Feb '24

02 Feb '24

From: Daniel Vetter <daniel.vetter(a)ffwll.ch> stable inclusion from stable-v4.19.291 commit 73a82b22963defa87204f0f9f44a534adf7f831a category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I8YD5O CVE: CVE-2023-51043 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 4e076c73e4f6e90816b30fcd4a0d7ab365087255 upstream. This requires a bit of background. Properly done a modeset driver's unload/remove sequence should be drm_dev_unplug(); drm_atomic_helper_shutdown(); drm_dev_put(); The trouble is that the drm_dev_unplugged() checks are by design racy, they do not synchronize against all outstanding ioctl. This is because those ioctl could block forever (both for modeset and for driver specific ioctls), leading to deadlocks in hotunplug. Instead the code sections that touch the hardware need to be annotated with drm_dev_enter/exit, to avoid accessing hardware resources after the unload/remove has finished. To avoid use-after-free issues all the involved userspace visible objects are supposed to hold a reference on the underlying drm_device, like drm_file does. The issue now is that we missed one, the atomic modeset ioctl can be run in a nonblocking fashion, and in that case it cannot rely on the implied drm_device reference provided by the ioctl calling context. This can result in a use-after-free if an nonblocking atomic commit is carefully raced against a driver unload. Fix this by unconditionally grabbing a drm_device reference for any drm_atomic_state structures. Strictly speaking this isn't required for blocking commits and TEST_ONLY calls, but it's the simpler approach. Thanks to shanzhulig for the initial idea of grabbing an unconditional reference, I just added comments, a condensed commit message and fixed a minor potential issue in where exactly we drop the final reference. Reported-by: shanzhulig <shanzhulig(a)gmail.com> Suggested-by: shanzhulig <shanzhulig(a)gmail.com> Reviewed-by: Maxime Ripard <mripard(a)kernel.org> Cc: Maarten Lankhorst <maarten.lankhorst(a)linux.intel.com> Cc: Thomas Zimmermann <tzimmermann(a)suse.de> Cc: David Airlie <airlied(a)gmail.com> Cc: stable(a)kernel.org Signed-off-by: Daniel Vetter <daniel.vetter(a)intel.com> Signed-off-by: Daniel Vetter <daniel.vetter(a)ffwll.ch> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Guo Mengqi <guomengqi3(a)huawei.com> --- drivers/gpu/drm/drm_atomic.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index 1a4b44923aec..e703341e4cb2 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -91,6 +91,12 @@ drm_atomic_state_init(struct drm_device *dev, struct drm_atomic_state *state) if (!state->planes) goto fail; + /* + * Because drm_atomic_state can be committed asynchronously we need our + * own reference and cannot rely on the on implied by drm_file in the + * ioctl call. + */ + drm_dev_get(dev); state->dev = dev; DRM_DEBUG_ATOMIC("Allocated atomic state %p\n", state); @@ -250,7 +256,8 @@ EXPORT_SYMBOL(drm_atomic_state_clear); void __drm_atomic_state_free(struct kref *ref) { struct drm_atomic_state *state = container_of(ref, typeof(*state), ref); - struct drm_mode_config *config = &state->dev->mode_config; + struct drm_device *dev = state->dev; + struct drm_mode_config *config = &dev->mode_config; drm_atomic_state_clear(state); @@ -262,6 +269,8 @@ void __drm_atomic_state_free(struct kref *ref) drm_atomic_state_default_release(state); kfree(state); } + + drm_dev_put(dev); } EXPORT_SYMBOL(__drm_atomic_state_free); -- 2.17.1

2 1

[PATCH OLK-6.6 0/2] config: Update openeuler_defconfig base on current
by Zheng Zengkai 01 Feb '24

01 Feb '24

while building arm64 & x86 kernel, ci robot complains new kconfig were introduced, but the corresponding openeuler_defconfig is not updated. Run 'make update_oedefconfig' to fix it. Zheng Zengkai (2): config: arm64: Update openeuler_defconfig base on current source code config: x86: Update openeuler_defconfig base on current source code arch/arm64/configs/openeuler_defconfig | 21 +++++++++++++++++++++ arch/x86/configs/openeuler_defconfig | 19 +++++++++++++++++++ 2 files changed, 40 insertions(+) -- 2.20.1

2 3

[PATCH openEuler-1.0-LTS v2 0/2] Save and restore msg_namelen in sock_sendmsg
by Liu Jian 01 Feb '24

01 Feb '24

v1->v2: add cover letter Jordan Rife (1): net: prevent rewrite of msg_name in sock_sendmsg() Marc Dionne (1): net: Save and restore msg_namelen in sock_sendmsg net/socket.c | 31 +++++++++++++++++++++++++------ 1 file changed, 25 insertions(+), 6 deletions(-) -- 2.34.1

2 3

[PATCH OLK-6.6 v2] kabi: reserve space for struct acpi_device and acpi_scan_handler
by Xiongfeng Wang 01 Feb '24

01 Feb '24

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8ZCY9 CVE: NA ------------------------------- reserve space for struct acpi_device, struct acpi_device_power and struct acpi_scan_handler Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> --- include/acpi/acpi_bus.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index 0b7eab0ef7d7..6adf312c70dc 100644 --- a/include/acpi/acpi_bus.h +++ b/include/acpi/acpi_bus.h @@ -11,6 +11,7 @@ #include <linux/device.h> #include <linux/property.h> +#include <linux/kabi.h> /* TBD: Make dynamic */ #define ACPI_MAX_HANDLES 10 @@ -130,6 +131,8 @@ struct acpi_scan_handler { void (*bind)(struct device *phys_dev); void (*unbind)(struct device *phys_dev); struct acpi_hotplug_profile hotplug; + KABI_RESERVE(1) + KABI_RESERVE(2) }; /* @@ -282,6 +285,8 @@ struct acpi_device_power { struct acpi_device_power_flags flags; struct acpi_device_power_state states[ACPI_D_STATE_COUNT]; /* Power states (D0-D3Cold) */ u8 state_for_enumeration; /* Deepest power state for enumeration */ + KABI_RESERVE(1) + KABI_RESERVE(2) }; struct acpi_dep_data { @@ -389,6 +394,8 @@ struct acpi_device { struct list_head physical_node_list; struct mutex physical_node_lock; void (*remove)(struct acpi_device *); + KABI_RESERVE(1) + KABI_RESERVE(2) }; /* Non-device subnode */ -- 2.20.1

2 1

[PATCH OLK-6.6 V1] sched/topology: Fix cpus hotplug deadlock in check_node_limit()
by Cheng Yu 01 Feb '24

01 Feb '24

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8ZBLR CVE: NA ----------------- In some cases, when performing the CPU continuous offline operation, hungtask will be reported. The error stack is as follows: Call track: __switch_to+0xc4/0x160 __schedule+0x8a4/0xcc0 schedule+0x58/0xf0 percpu_rwsem_wait+0xac/0x1c0 __percpu_down_read+0x44/0x100 cpus_read_lock+0x5c/0x70 static_key_disable+0x20/0x48 sched_init_numa+0x588/0x5e8 sched_update_numa+0x88/0xa0 sched_cpu_deactivate+0x138/0x2b8 cpuhp_invoke_callback+0x13c/0x5a0 cpuhp_thread_fun+0x108/0x1b0 smpboot_thread_fn+0x14c/0x198 kthread+0xf0/0x108 ret_from_fork+0x10/0x20 Before calling check_node_limit(), cpus hotplug write lock has been locked, and static_branch_disable() will try to get cpus hotplug read lock, it cause a deadlock issue. To fix it, use static_branch_disable_cpuslocked to replace static_branch_disable in check_node_limit to avoid the deadlock issue. Fixes: c0830508683b ("sched/fair: disable stealing if too many NUMA nodes") Signed-off-by: Cheng Yu <serein.chengyu(a)huawei.com> --- kernel/sched/topology.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index 6e53175382ce..5dca59dfea85 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -1880,7 +1880,7 @@ static void check_node_limit(void) if (sched_steal_node_limit == 0) sched_steal_node_limit = SCHED_STEAL_NODE_LIMIT_DEFAULT; if (n > sched_steal_node_limit) { - static_branch_disable(&sched_steal_allow); + static_branch_disable_cpuslocked(&sched_steal_allow); pr_debug("Suppressing sched STEAL. To enable, reboot with sched_steal_node_limit=%d", n); } } -- 2.25.1

2 1

[PATCH v2 OLK-6.6] net/kabi: reserve space for net related structures
by Wang Yufen 01 Feb '24

01 Feb '24

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I903T2 -------------------------------- Signed-off-by: Wang Yufen <wangyufen(a)huawei.com> --- include/linux/inetdevice.h | 2 ++ include/net/if_inet6.h | 8 ++++++++ include/net/netns/sctp.h | 2 ++ include/net/sock_reuseport.h | 2 ++ include/rdma/ib_verbs.h | 12 ++++++++++++ 5 files changed, 26 insertions(+) diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index 5181993..aa816f5 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -163,6 +163,8 @@ struct in_ifaddr { __u32 ifa_preferred_lft; unsigned long ifa_cstamp; /* created timestamp */ unsigned long ifa_tstamp; /* updated timestamp */ + KABI_RESERVE(1) + KABI_RESERVE(2) }; struct in_validator_info { diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h index c849072..1199310 100644 --- a/include/net/if_inet6.h +++ b/include/net/if_inet6.h @@ -111,6 +111,8 @@ struct ip6_sf_list { unsigned char sf_oldin; /* change state */ unsigned char sf_crcount; /* retrans. left to send */ struct rcu_head rcu; + KABI_RESERVE(1) + KABI_RESERVE(2) }; #define MAF_TIMER_RUNNING 0x01 @@ -135,6 +137,10 @@ struct ifmcaddr6 { unsigned long mca_cstamp; unsigned long mca_tstamp; struct rcu_head rcu; + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; /* Anycast stuff */ @@ -223,6 +229,8 @@ struct inet6_dev { struct rcu_head rcu; unsigned int ra_mtu; + KABI_RESERVE(1) + KABI_RESERVE(2) }; static inline void ipv6_eth_mc_map(const struct in6_addr *addr, char *buf) diff --git a/include/net/netns/sctp.h b/include/net/netns/sctp.h index 7eff3d9..b812c1b 100644 --- a/include/net/netns/sctp.h +++ b/include/net/netns/sctp.h @@ -179,6 +179,8 @@ struct netns_sctp { #ifdef CONFIG_NET_L3_MASTER_DEV int l3mdev_accept; #endif + KABI_RESERVE(1) + KABI_RESERVE(2) }; #endif /* __NETNS_SCTP_H__ */ diff --git a/include/net/sock_reuseport.h b/include/net/sock_reuseport.h index 6ec140b..9458cc5 100644 --- a/include/net/sock_reuseport.h +++ b/include/net/sock_reuseport.h @@ -26,6 +26,8 @@ struct sock_reuseport { unsigned int bind_inany:1; unsigned int has_conns:1; struct bpf_prog __rcu *prog; /* optional BPF sock selector */ + KABI_RESERVE(1) + KABI_RESERVE(2) struct sock *socks[]; /* array of sock pointers */ }; diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 62f9d12..f0d7d2e 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -450,6 +450,8 @@ struct ib_device_attr { u64 max_dm_size; /* Max entries for sgl for optimized performance per READ */ u32 max_sgl_rd; + KABI_RESERVE(1) + KABI_RESERVE(2) }; enum ib_mtu { @@ -607,6 +609,8 @@ struct rdma_hw_stats { const struct rdma_stat_desc *descs; unsigned long *is_disabled; int num_counters; + KABI_RESERVE(1) + KABI_RESERVE(2) u64 value[]; }; @@ -1642,6 +1646,8 @@ struct ib_srq { * Implementation details of the RDMA core, don't use in drivers: */ struct rdma_restrack_entry res; + KABI_RESERVE(1) + KABI_RESERVE(2) }; enum ib_raw_packet_caps { @@ -2217,6 +2223,8 @@ struct ib_port_data { struct hlist_node ndev_hash_link; struct rdma_port_counter port_counter; struct ib_port *sysfs; + KABI_RESERVE(1) + KABI_RESERVE(2) }; /* rdma netdev type - specifies protocol type */ @@ -2684,6 +2692,8 @@ struct ib_device_ops { DECLARE_RDMA_OBJ_SIZE(ib_srq); DECLARE_RDMA_OBJ_SIZE(ib_ucontext); DECLARE_RDMA_OBJ_SIZE(ib_xrcd); + KABI_RESERVE(1) + KABI_RESERVE(2) }; struct ib_core_device { @@ -2784,6 +2794,8 @@ struct ib_device { char iw_ifname[IFNAMSIZ]; u32 iw_driver_flags; u32 lag_flags; + KABI_RESERVE(1) + KABI_RESERVE(2) }; static inline void *rdma_zalloc_obj(struct ib_device *dev, size_t size, -- 1.8.3.1

2 1

[PATCH OLK-6.6 0/7] mm: improve performance of accounted kernel memory allocations
by Liu Shixin 01 Feb '24

01 Feb '24

This patchset improves the performance of accounted kernel memory allocations by ~30% as measured by a micro-benchmark [1]. The benchmark is very straightforward: 1M of 64 bytes-large kmalloc() allocations. Below are results with the disabled kernel memory accounting, the original state and with this patchset applied. | | Kmem disabled | Original | Patched | Delta | |-------------+---------------+----------+---------+--------| | User cgroup | 29764 | 84548 | 59078 | -30.0% | | Root cgroup | 29742 | 48342 | 31501 | -34.8% | As we can see, the patchset removes the majority of the overhead when there is no actual accounting (a task belongs to the root memory cgroup) and almost halves the accounting overhead otherwise. The main idea is to get rid of unnecessary memcg to objcg conversions and switch to a scope-based protection of objcgs, which eliminates extra operations with objcg reference counters under a rcu read lock. More details are provided in individual commit descriptions. Roman Gushchin (7): mm: kmem: optimize get_obj_cgroup_from_current() mm: kmem: add direct objcg pointer to task_struct mm: kmem: make memcg keep a reference to the original objcg mm: kmem: scoped objcg protection percpu: scoped objcg protection mm: kmem: reimplement get_obj_cgroup_from_current() mm: kmem: properly initialize local objcg variable in current_obj_cgroup() include/linux/memcontrol.h | 28 +++++- include/linux/sched.h | 4 + include/linux/sched/mm.h | 4 + mm/memcontrol.c | 187 +++++++++++++++++++++++++++++++------ mm/percpu.c | 8 +- mm/slab.h | 15 +-- 6 files changed, 204 insertions(+), 42 deletions(-) -- 2.25.1

2 8

[PATCH OLK-6.6 v3 0/7] kabi reserve
by Yu Liao 01 Feb '24

01 Feb '24

Yu Liao (7): kabi: reserve space for kobject related structures kabi: reserve space for hrtimer related structures kabi: reserve space for posix clock related structure kabi: reserve space for workqueue subsystem related structure kabi: reserve space for struct ptp_clock_info kabi: reserve space for struct timer_list kabi: reserve space for struct clocksource include/linux/clocksource.h | 6 ++++++ include/linux/hrtimer.h | 10 ++++++++++ include/linux/kobject.h | 16 ++++++++++++++++ include/linux/posix-clock.h | 11 +++++++++++ include/linux/ptp_clock_kernel.h | 6 ++++++ include/linux/timer.h | 4 ++++ include/linux/workqueue.h | 11 +++++++++++ 7 files changed, 64 insertions(+) -- 2.33.0

2 8

[PATCH OLK-6.6 v2] kabi: net: reserve space for net related structure
by Zhengchao Shao 01 Feb '24

01 Feb '24

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8ZMS7 ---------------------------------------------------- Reserve some fields beforehand for net related structures prone to change. Signed-off-by: Zhengchao Shao <shaozhengchao(a)huawei.com> --- include/linux/netlink.h | 5 +++++ include/net/cfg802154.h | 5 +++++ include/net/inet_frag.h | 4 ++++ include/net/ip_fib.h | 8 ++++++++ include/net/ipv6_stubs.h | 4 ++++ include/net/netns/conntrack.h | 4 ++++ include/net/netns/core.h | 6 ++++++ include/net/xfrm.h | 8 ++++++++ 8 files changed, 44 insertions(+) diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 75d7de34c908..4111d53f4656 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -8,6 +8,7 @@ #include <linux/export.h> #include <net/scm.h> #include <uapi/linux/netlink.h> +#include <linux/kabi.h> struct net; @@ -88,6 +89,10 @@ struct netlink_ext_ack { u8 cookie[NETLINK_MAX_COOKIE_LEN]; u8 cookie_len; char _msg_buf[NETLINK_MAX_FMTMSG_LEN]; + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; /* Always use this macro, this allows later putting the diff --git a/include/net/cfg802154.h b/include/net/cfg802154.h index f79ce133e51a..61aabd82b789 100644 --- a/include/net/cfg802154.h +++ b/include/net/cfg802154.h @@ -16,6 +16,8 @@ #include <net/nl802154.h> +#include <linux/kabi.h> + struct wpan_phy; struct wpan_phy_cca; struct cfg802154_scan_request; @@ -242,6 +244,9 @@ struct wpan_phy { */ enum ieee802154_filtering_level filtering; + KABI_RESERVE(1) + KABI_RESERVE(2) + char priv[] __aligned(NETDEV_ALIGN); }; diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h index 153960663ce4..7d9354c3c2ed 100644 --- a/include/net/inet_frag.h +++ b/include/net/inet_frag.h @@ -8,6 +8,7 @@ #include <linux/rbtree_types.h> #include <linux/refcount.h> #include <net/dropreason-core.h> +#include <linux/kabi.h> /* Per netns frag queues directory */ struct fqdir { @@ -102,6 +103,9 @@ struct inet_frag_queue { u16 max_size; struct fqdir *fqdir; struct rcu_head rcu; + + KABI_RESERVE(1) + KABI_RESERVE(2) }; struct inet_frags { diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index 15de07d36540..7f3e0a92f857 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -22,6 +22,7 @@ #include <linux/percpu.h> #include <linux/notifier.h> #include <linux/refcount.h> +#include <linux/kabi.h> struct fib_config { u8 fc_dst_len; @@ -100,6 +101,9 @@ struct fib_nh_common { struct rtable __rcu * __percpu *nhc_pcpu_rth_output; struct rtable __rcu *nhc_rth_input; struct fnhe_hash_bucket __rcu *nhc_exceptions; + + KABI_RESERVE(1) + KABI_RESERVE(2) }; struct fib_nh { @@ -157,6 +161,10 @@ struct fib_info { bool pfsrc_removed; struct nexthop *nh; struct rcu_head rcu; + + KABI_RESERVE(1) + KABI_RESERVE(2) + struct fib_nh fib_nh[]; }; diff --git a/include/net/ipv6_stubs.h b/include/net/ipv6_stubs.h index c48186bf4737..c2f2cfbd1d03 100644 --- a/include/net/ipv6_stubs.h +++ b/include/net/ipv6_stubs.h @@ -9,6 +9,7 @@ #include <net/flow.h> #include <net/neighbour.h> #include <net/sock.h> +#include <linux/kabi.h> /* structs from net/ip6_fib.h */ struct fib6_info; @@ -69,6 +70,9 @@ struct ipv6_stub { int (*output)(struct net *, struct sock *, struct sk_buff *)); struct net_device *(*ipv6_dev_find)(struct net *net, const struct in6_addr *addr, struct net_device *dev); + + KABI_RESERVE(1) + KABI_RESERVE(2) }; extern const struct ipv6_stub *ipv6_stub __read_mostly; diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h index 1f463b3957c7..308c54643b3a 100644 --- a/include/net/netns/conntrack.h +++ b/include/net/netns/conntrack.h @@ -14,6 +14,7 @@ #include <linux/netfilter/nf_conntrack_sctp.h> #endif #include <linux/seqlock.h> +#include <linux/kabi.h> struct ctl_table_header; struct nf_conntrack_ecache; @@ -31,6 +32,9 @@ struct nf_tcp_net { #if IS_ENABLED(CONFIG_NF_FLOW_TABLE) unsigned int offload_timeout; #endif + + KABI_RESERVE(1) + KABI_RESERVE(2) }; enum udp_conntrack { diff --git a/include/net/netns/core.h b/include/net/netns/core.h index a91ef9f8de60..dfb54998c497 100644 --- a/include/net/netns/core.h +++ b/include/net/netns/core.h @@ -3,6 +3,7 @@ #define __NETNS_CORE_H__ #include <linux/types.h> +#include <linux/kabi.h> struct ctl_table_header; struct prot_inuse; @@ -22,6 +23,11 @@ struct netns_core { #if IS_ENABLED(CONFIG_RPS) && IS_ENABLED(CONFIG_SYSCTL) struct cpumask *rps_default_mask; #endif + + KABI_RESERVE(1) + KABI_RESERVE(2) + KABI_RESERVE(3) + KABI_RESERVE(4) }; #endif diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 363c7d510554..9eff85ef369c 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -32,6 +32,8 @@ #include <net/snmp.h> #endif +#include <linux/kabi.h> + #define XFRM_PROTO_ESP 50 #define XFRM_PROTO_AH 51 #define XFRM_PROTO_COMP 108 @@ -289,6 +291,9 @@ struct xfrm_state { /* Private data of this transformer, format is opaque, * interpreted by xfrm_type methods. */ void *data; + + KABI_RESERVE(1) + KABI_RESERVE(2) }; static inline struct net *xs_net(struct xfrm_state *x) @@ -549,6 +554,9 @@ struct xfrm_policy { struct rcu_head rcu; struct xfrm_dev_offload xdo; + + KABI_RESERVE(1) + KABI_RESERVE(2) }; static inline struct net *xp_net(const struct xfrm_policy *xp) -- 2.34.1

2 1