kernel/linux.git/include/net/nexthop.h, branch v6.12.80

net: nexthop: Increase weight to u16

2024-08-13T00:50:34+00:00

In CLOS networks, as link failures occur at various points in the network, ECMP weights of the involved nodes are adjusted to compensate. With high fan-out of the involved nodes, and overall high number of nodes, a (non-)ECMP weight ratio that we would like to configure does not fit into 8 bits. Instead of, say, 255:254, we might like to configure something like 1000:999. For these deployments, the 8-bit weight may not be enough. To that end, in this patch increase the next hop weight from u8 to u16. Increasing the width of an integral type can be tricky, because while the code still compiles, the types may not check out anymore, and numerical errors come up. To prevent this, the conversion was done in two steps. First the type was changed from u8 to a single-member structure, which invalidated all uses of the field. This allowed going through them one by one and audit for type correctness. Then the structure was replaced with a vanilla u16 again. This should ensure that no place was missed. The UAPI for configuring nexthop group members is that an attribute NHA_GROUP carries an array of struct nexthop_grp entries: struct nexthop_grp { __u32 id; /* nexthop id - must exist */ __u8 weight; /* weight of this nexthop */ __u8 resvd1; __u16 resvd2; }; The field resvd1 is currently validated and required to be zero. We can lift this requirement and carry high-order bits of the weight in the reserved field: struct nexthop_grp { __u32 id; /* nexthop id - must exist */ __u8 weight; /* weight of this nexthop */ __u8 weight_high; __u16 resvd2; }; Keeping the fields split this way was chosen in case an existing userspace makes assumptions about the width of the weight field, and to sidestep any endianness issues. The weight field is currently encoded as the weight value minus one, because weight of 0 is invalid. This same trick is impossible for the new weight_high field, because zero must mean actual zero. With this in place: - Old userspace is guaranteed to carry weight_high of 0, therefore configuring 8-bit weights as appropriate. When dumping nexthops with 16-bit weight, it would only show the lower 8 bits. But configuring such nexthops implies existence of userspace aware of the extension in the first place. - New userspace talking to an old kernel will work as long as it only attempts to configure 8-bit weights, where the high-order bits are zero. Old kernel will bounce attempts at configuring >8-bit weights. Renaming reserved fields as they are allocated for some purpose is commonly done in Linux. Whoever touches a reserved field is doing so at their own risk. nexthop_grp::resvd1 in particular is currently used by at least strace, however they carry an own copy of UAPI headers, and the conversion should be trivial. A helper is provided for decoding the weight out of the two fields. Forcing a conversion seems preferable to bending backwards and introducing anonymous unions or whatever. Signed-off-by: Petr Machata Reviewed-by: Ido Schimmel Reviewed-by: David Ahern Reviewed-by: Przemek Kitszel Link: https://patch.msgid.link/483e2fcf4beb0d9135d62e7d27b46fa2685479d4.1723036486.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski

net: add two more call_rcu_hurry()

2024-04-25T22:24:23+00:00

I had failures with pmtu.sh selftests lately, with netns dismantles firing ref_tracking alerts [1]. After much debugging, I found that some queued rcu callbacks were delayed by minutes, because of CONFIG_RCU_LAZY=y option. Joel Fernandes had a similar issue in the past, fixed with commit 483c26ff63f4 ("net: Use call_rcu_hurry() for dst_release()") In this commit, I make sure nexthop_free_rcu() and free_fib_info_rcu() are not delayed too much because they both can release device references. tools/testing/selftests/net/pmtu.sh no longer fails. Traces were: [ 968.179860] ref_tracker: veth_A-R1@00000000d0ff3fe2 has 3/5 users at dst_alloc+0x76/0x160 ip6_dst_alloc+0x25/0x80 ip6_pol_route+0x2a8/0x450 ip6_pol_route_output+0x1f/0x30 fib6_rule_lookup+0x163/0x270 ip6_route_output_flags+0xda/0x190 ip6_dst_lookup_tail.constprop.0+0x1d0/0x260 ip6_dst_lookup_flow+0x47/0xa0 udp_tunnel6_dst_lookup+0x158/0x210 vxlan_xmit_one+0x4c2/0x1550 [vxlan] vxlan_xmit+0x52d/0x14f0 [vxlan] dev_hard_start_xmit+0x7b/0x1e0 __dev_queue_xmit+0x20b/0xe40 ip6_finish_output2+0x2ea/0x6e0 ip6_finish_output+0x143/0x320 ip6_output+0x74/0x140 [ 968.179860] ref_tracker: veth_A-R1@00000000d0ff3fe2 has 1/5 users at netdev_get_by_index+0xc0/0xe0 fib6_nh_init+0x1a9/0xa90 rtm_new_nexthop+0x6fa/0x1580 rtnetlink_rcv_msg+0x155/0x3e0 netlink_rcv_skb+0x61/0x110 rtnetlink_rcv+0x19/0x20 netlink_unicast+0x23f/0x380 netlink_sendmsg+0x1fc/0x430 ____sys_sendmsg+0x2ef/0x320 ___sys_sendmsg+0x86/0xd0 __sys_sendmsg+0x67/0xc0 __x64_sys_sendmsg+0x21/0x30 x64_sys_call+0x252/0x2030 do_syscall_64+0x6c/0x190 entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 968.179860] ref_tracker: veth_A-R1@00000000d0ff3fe2 has 1/5 users at ipv6_add_dev+0x136/0x530 addrconf_notify+0x19d/0x770 notifier_call_chain+0x65/0xd0 raw_notifier_call_chain+0x1a/0x20 call_netdevice_notifiers_info+0x54/0x90 register_netdevice+0x61e/0x790 veth_newlink+0x230/0x440 __rtnl_newlink+0x7d2/0xaa0 rtnl_newlink+0x4c/0x70 rtnetlink_rcv_msg+0x155/0x3e0 netlink_rcv_skb+0x61/0x110 rtnetlink_rcv+0x19/0x20 netlink_unicast+0x23f/0x380 netlink_sendmsg+0x1fc/0x430 ____sys_sendmsg+0x2ef/0x320 ___sys_sendmsg+0x86/0xd0 .... [ 1079.316024] ? show_regs+0x68/0x80 [ 1079.316087] ? __warn+0x8c/0x140 [ 1079.316103] ? ref_tracker_free+0x1a0/0x270 [ 1079.316117] ? report_bug+0x196/0x1c0 [ 1079.316135] ? handle_bug+0x42/0x80 [ 1079.316149] ? exc_invalid_op+0x1c/0x70 [ 1079.316162] ? asm_exc_invalid_op+0x1f/0x30 [ 1079.316193] ? ref_tracker_free+0x1a0/0x270 [ 1079.316208] ? _raw_spin_unlock+0x1a/0x40 [ 1079.316222] ? free_unref_page+0x126/0x1a0 [ 1079.316239] ? destroy_large_folio+0x69/0x90 [ 1079.316251] ? __folio_put+0x99/0xd0 [ 1079.316276] dst_dev_put+0x69/0xd0 [ 1079.316308] fib6_nh_release_dsts.part.0+0x3d/0x80 [ 1079.316327] fib6_nh_release+0x45/0x70 [ 1079.316340] nexthop_free_rcu+0x131/0x170 [ 1079.316356] rcu_do_batch+0x1ee/0x820 [ 1079.316370] ? rcu_do_batch+0x179/0x820 [ 1079.316388] rcu_core+0x1aa/0x4d0 [ 1079.316405] rcu_core_si+0x12/0x20 [ 1079.316417] __do_softirq+0x13a/0x3dc [ 1079.316435] __irq_exit_rcu+0xa3/0x110 [ 1079.316449] irq_exit_rcu+0x12/0x30 [ 1079.316462] sysvec_apic_timer_interrupt+0x5b/0xe0 [ 1079.316474] asm_sysvec_apic_timer_interrupt+0x1f/0x30 [ 1079.316569] RIP: 0033:0x7f06b65c63f0 Signed-off-by: Eric Dumazet Cc: Joel Fernandes (Google) Cc: Paul E. McKenney Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20240423205408.39632-1-edumazet@google.com Signed-off-by: Jakub Kicinski

net: nexthop: Have all NH notifiers carry NH ID

2024-03-11T21:14:07+00:00

When sending the notifications to collect NH statistics for resilient groups, the driver will need to know the nexthop IDs in individual buckets to look up the right counter. To that end, move the nexthop ID from struct nh_notifier_grp_entry_info to nh_notifier_single_info. Signed-off-by: Petr Machata Reviewed-by: Ido Schimmel Link: https://lore.kernel.org/r/8f964cd50b1a56d3606ce7ab4c50354ae019c43b.1709901020.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski

net: nexthop: Expose nexthop group HW stats to user space

2024-03-08T10:35:47+00:00

Add netlink support for reading NH group hardware stats. Stats collection is done through a new notifier, NEXTHOP_EVENT_HW_STATS_REPORT_DELTA. Drivers that implement HW counters for a given NH group are thereby asked to collect the stats and report back to core by calling nh_grp_hw_stats_report_delta(). This is similar to what netdevice L3 stats do. Besides exposing number of packets that passed in the HW datapath, also include information on whether any driver actually realizes the counters. The core can tell based on whether it got any _report_delta() reports from the drivers. This allows enabling the statistics at the group at any time, with drivers opting into supporting them. This is also in line with what netdevice L3 stats are doing. So as not to waste time and space, tie the collection and reporting of HW stats with a new op flag, NHA_OP_FLAG_DUMP_HW_STATS. Co-developed-by: Petr Machata Signed-off-by: Petr Machata Signed-off-by: Ido Schimmel Reviewed-by: Kees Cook # For the __counted_by bits Reviewed-by: David Ahern Signed-off-by: David S. Miller

net: nexthop: Add ability to enable / disable hardware statistics

2024-03-08T10:35:47+00:00

Add netlink support for enabling collection of HW statistics on nexthop groups. Signed-off-by: Ido Schimmel Reviewed-by: David Ahern Signed-off-by: Petr Machata Signed-off-by: David S. Miller

net: nexthop: Add hardware statistics notifications

2024-03-08T10:35:47+00:00

Add hw_stats field to several notifier structures to communicate to the drivers that HW statistics should be configured for nexthops within a given group. Signed-off-by: Ido Schimmel Reviewed-by: David Ahern Signed-off-by: Petr Machata Signed-off-by: David S. Miller

net: nexthop: Add nexthop group entry stats

2024-03-08T10:35:46+00:00

Add nexthop group entry stats to count the number of packets forwarded via each nexthop in the group. The stats will be exposed to user space for better data path observability in the next patch. The per-CPU stats pointer is placed at the beginning of 'struct nh_grp_entry', so that all the fields accessed for the data path reside on the same cache line: struct nh_grp_entry { struct nexthop * nh; /* 0 8 */ struct nh_grp_entry_stats * stats; /* 8 8 */ u8 weight; /* 16 1 */ /* XXX 7 bytes hole, try to pack */ union { struct { atomic_t upper_bound; /* 24 4 */ } hthr; /* 24 4 */ struct { struct list_head uw_nh_entry; /* 24 16 */ u16 count_buckets; /* 40 2 */ u16 wants_buckets; /* 42 2 */ } res; /* 24 24 */ }; /* 24 24 */ struct list_head nh_list; /* 48 16 */ /* --- cacheline 1 boundary (64 bytes) --- */ struct nexthop * nh_parent; /* 64 8 */ /* size: 72, cachelines: 2, members: 6 */ /* sum members: 65, holes: 1, sum holes: 7 */ /* last cacheline: 8 bytes */ }; Co-developed-by: Petr Machata Signed-off-by: Petr Machata Signed-off-by: Ido Schimmel Signed-off-by: David S. Miller

nexthop: allow nexthop_mpath_fill_node() to be called without RTNL

2024-02-26T11:46:12+00:00

nexthop_mpath_fill_node() will be potentially called from contexts holding rcu_lock instead of RTNL. Suggested-by: Ido Schimmel Link: https://lore.kernel.org/all/ZdZDWVdjMaQkXBgW@shredder/ Signed-off-by: Eric Dumazet Reviewed-by: Donald Hunter Signed-off-by: David S. Miller

ipv4: add __unregister_nexthop_notifier()

2024-02-08T02:55:11+00:00

unregister_nexthop_notifier() assumes the caller does not hold rtnl. We need in the following patch to use it from a context already holding rtnl. Add __unregister_nexthop_notifier(). unregister_nexthop_notifier() becomes a wrapper. Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Reviewed-by: Antoine Tenart Link: https://lore.kernel.org/r/20240206144313.2050392-9-edumazet@google.com Signed-off-by: Jakub Kicinski

nexthop: Annotate struct nh_group with __counted_by

2023-10-06T09:50:34+00:00

Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). As found with Coccinelle[1], add __counted_by for struct nh_group. Cc: David Ahern Cc: "David S. Miller" Cc: Eric Dumazet Cc: Jakub Kicinski Cc: Paolo Abeni Cc: netdev@vger.kernel.org Link: https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci [1] Signed-off-by: Kees Cook Reviewed-by: David Ahern Signed-off-by: David S. Miller