kernel/linux.git/drivers/net/wireguard/send.c, branch v6.18.21

wireguard: send: annotate intentional data race in checking empty queue

2024-07-06T00:21:10+00:00

KCSAN reports a race in wg_packet_send_keepalive, which is intentional: BUG: KCSAN: data-race in wg_packet_send_keepalive / wg_packet_send_staged_packets write to 0xffff88814cd91280 of 8 bytes by task 3194 on cpu 0: __skb_queue_head_init include/linux/skbuff.h:2162 [inline] skb_queue_splice_init include/linux/skbuff.h:2248 [inline] wg_packet_send_staged_packets+0xe5/0xad0 drivers/net/wireguard/send.c:351 wg_xmit+0x5b8/0x660 drivers/net/wireguard/device.c:218 __netdev_start_xmit include/linux/netdevice.h:4940 [inline] netdev_start_xmit include/linux/netdevice.h:4954 [inline] xmit_one net/core/dev.c:3548 [inline] dev_hard_start_xmit+0x11b/0x3f0 net/core/dev.c:3564 __dev_queue_xmit+0xeff/0x1d80 net/core/dev.c:4349 dev_queue_xmit include/linux/netdevice.h:3134 [inline] neigh_connected_output+0x231/0x2a0 net/core/neighbour.c:1592 neigh_output include/net/neighbour.h:542 [inline] ip6_finish_output2+0xa66/0xce0 net/ipv6/ip6_output.c:137 ip6_finish_output+0x1a5/0x490 net/ipv6/ip6_output.c:222 NF_HOOK_COND include/linux/netfilter.h:303 [inline] ip6_output+0xeb/0x220 net/ipv6/ip6_output.c:243 dst_output include/net/dst.h:451 [inline] NF_HOOK include/linux/netfilter.h:314 [inline] ndisc_send_skb+0x4a2/0x670 net/ipv6/ndisc.c:509 ndisc_send_rs+0x3ab/0x3e0 net/ipv6/ndisc.c:719 addrconf_dad_completed+0x640/0x8e0 net/ipv6/addrconf.c:4295 addrconf_dad_work+0x891/0xbc0 process_one_work kernel/workqueue.c:2633 [inline] process_scheduled_works+0x5b8/0xa30 kernel/workqueue.c:2706 worker_thread+0x525/0x730 kernel/workqueue.c:2787 kthread+0x1d7/0x210 kernel/kthread.c:388 ret_from_fork+0x48/0x60 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:242 read to 0xffff88814cd91280 of 8 bytes by task 3202 on cpu 1: skb_queue_empty include/linux/skbuff.h:1798 [inline] wg_packet_send_keepalive+0x20/0x100 drivers/net/wireguard/send.c:225 wg_receive_handshake_packet drivers/net/wireguard/receive.c:186 [inline] wg_packet_handshake_receive_worker+0x445/0x5e0 drivers/net/wireguard/receive.c:213 process_one_work kernel/workqueue.c:2633 [inline] process_scheduled_works+0x5b8/0xa30 kernel/workqueue.c:2706 worker_thread+0x525/0x730 kernel/workqueue.c:2787 kthread+0x1d7/0x210 kernel/kthread.c:388 ret_from_fork+0x48/0x60 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:242 value changed: 0xffff888148fef200 -> 0xffff88814cd91280 Mark this race as intentional by using the skb_queue_empty_lockless() function rather than skb_queue_empty(), which uses READ_ONCE() internally to annotate the race. Cc: stable@vger.kernel.org Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld Link: https://patch.msgid.link/20240704154517.1572127-5-Jason@zx2c4.com Signed-off-by: Jakub Kicinski

wireguard: use DEV_STATS_INC()

2023-11-19T19:48:25+00:00

wg_xmit() can be called concurrently, KCSAN reported [1] some device stats updates can be lost. Use DEV_STATS_INC() for this unlikely case. [1] BUG: KCSAN: data-race in wg_xmit / wg_xmit read-write to 0xffff888104239160 of 8 bytes by task 1375 on cpu 0: wg_xmit+0x60f/0x680 drivers/net/wireguard/device.c:231 __netdev_start_xmit include/linux/netdevice.h:4918 [inline] netdev_start_xmit include/linux/netdevice.h:4932 [inline] xmit_one net/core/dev.c:3543 [inline] dev_hard_start_xmit+0x11b/0x3f0 net/core/dev.c:3559 ... read-write to 0xffff888104239160 of 8 bytes by task 1378 on cpu 1: wg_xmit+0x60f/0x680 drivers/net/wireguard/device.c:231 __netdev_start_xmit include/linux/netdevice.h:4918 [inline] netdev_start_xmit include/linux/netdevice.h:4932 [inline] xmit_one net/core/dev.c:3543 [inline] dev_hard_start_xmit+0x11b/0x3f0 net/core/dev.c:3559 ... v2: also change wg_packet_consume_data_done() (Hangbin Liu) and wg_packet_purge_staged_packets() Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Reported-by: syzbot Signed-off-by: Eric Dumazet Cc: Jason A. Donenfeld Cc: Hangbin Liu Signed-off-by: Jason A. Donenfeld Reviewed-by: Hangbin Liu Signed-off-by: David S. Miller

wireguard: queueing: use saner cpu selection wrapping

2023-07-03T08:17:52+00:00

Using `% nr_cpumask_bits` is slow and complicated, and not totally robust toward dynamic changes to CPU topologies. Rather than storing the next CPU in the round-robin, just store the last one, and also return that value. This simplifies the loop drastically into a much more common pattern. Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Cc: stable@vger.kernel.org Reported-by: Linus Torvalds Tested-by: Manuel Leiner Signed-off-by: Jason A. Donenfeld Signed-off-by: David S. Miller

wireguard: queueing: get rid of per-peer ring buffers

2021-02-23T23:59:34+00:00

Having two ring buffers per-peer means that every peer results in two massive ring allocations. On an 8-core x86_64 machine, this commit reduces the per-peer allocation from 18,688 bytes to 1,856 bytes, which is an 90% reduction. Ninety percent! With some single-machine deployments approaching 500,000 peers, we're talking about a reduction from 7 gigs of memory down to 700 megs of memory. In order to get rid of these per-peer allocations, this commit switches to using a list-based queueing approach. Currently GSO fragments are chained together using the skb->next pointer (the skb_list_* singly linked list approach), so we form the per-peer queue around the unused skb->prev pointer (which sort of makes sense because the links are pointing backwards). Use of skb_queue_* is not possible here, because that is based on doubly linked lists and spinlocks. Multiple cores can write into the queue at any given time, because its writes occur in the start_xmit path or in the udp_recv path. But reads happen in a single workqueue item per-peer, amounting to a multi-producer, single-consumer paradigm. The MPSC queue is implemented locklessly and never blocks. However, it is not linearizable (though it is serializable), with a very tight and unlikely race on writes, which, when hit (some tiny fraction of the 0.15% of partial adds on a fully loaded 16-core x86_64 system), causes the queue reader to terminate early. However, because every packet sent queues up the same workqueue item after it is fully added, the worker resumes again, and stopping early isn't actually a problem, since at that point the packet wouldn't have yet been added to the encryption queue. These properties allow us to avoid disabling interrupts or spinning. The design is based on Dmitry Vyukov's algorithm [1]. Performance-wise, ordinarily list-based queues aren't preferable to ringbuffers, because of cache misses when following pointers around. However, we *already* have to follow the adjacent pointers when working through fragments, so there shouldn't actually be any change there. A potential downside is that dequeueing is a bit more complicated, but the ptr_ring structure used prior had a spinlock when dequeueing, so all and all the difference appears to be a wash. Actually, from profiling, the biggest performance hit, by far, of this commit winds up being atomic_add_unless(count, 1, max) and atomic_ dec(count), which account for the majority of CPU time, according to perf. In that sense, the previous ring buffer was superior in that it could check if it was full by head==tail, which the list-based approach cannot do. But all and all, this enables us to get massive memory savings, allowing WireGuard to scale for real world deployments, without taking much of a performance hit. [1] http://www.1024cores.net/home/lock-free-algorithms/queues/intrusive-mpsc-node-based-queue Reviewed-by: Dmitry Vyukov Reviewed-by: Toke Høiland-Jørgensen Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld Signed-off-by: Jakub Kicinski

wireguard: noise: separate receive counter from send counter

2020-05-21T03:55:09+00:00

In "wireguard: queueing: preserve flow hash across packet scrubbing", we were required to slightly increase the size of the receive replay counter to something still fairly small, but an increase nonetheless. It turns out that we can recoup some of the additional memory overhead by splitting up the prior union type into two distinct types. Before, we used the same "noise_counter" union for both sending and receiving, with sending just using a simple atomic64_t, while receiving used the full replay counter checker. This meant that most of the memory being allocated for the sending counter was being wasted. Since the old "noise_counter" type increased in size in the prior commit, now is a good time to split up that union type into a distinct "noise_replay_ counter" for receiving and a boring atomic64_t for sending, each using neither more nor less memory than required. Also, since sometimes the replay counter is accessed without necessitating additional accesses to the bitmap, we can reduce cache misses by hoisting the always-necessary lock above the bitmap in the struct layout. We also change a "noise_replay_counter" stack allocation to kmalloc in a -DDEBUG selftest so that KASAN doesn't trigger a stack frame warning. All and all, removing a bit of abstraction in this commit makes the code simpler and smaller, in addition to the motivating memory usage recuperation. For example, passing around raw "noise_symmetric_key" structs is something that really only makes sense within noise.c, in the one place where the sending and receiving keys can safely be thought of as the same type of object; subsequent to that, it's important that we uniformly access these through keypair->{sending,receiving}, where their distinct roles are always made explicit. So this patch allows us to draw that distinction clearly as well. Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld Signed-off-by: David S. Miller

wireguard: queueing: preserve flow hash across packet scrubbing

2020-05-21T03:55:09+00:00

It's important that we clear most header fields during encapsulation and decapsulation, because the packet is substantially changed, and we don't want any info leak or logic bug due to an accidental correlation. But, for encapsulation, it's wrong to clear skb->hash, since it's used by fq_codel and flow dissection in general. Without it, classification does not proceed as usual. This change might make it easier to estimate the number of innerflows by examining clustering of out of order packets, but this shouldn't open up anything that can't already be inferred otherwise (e.g. syn packet size inference), and fq_codel can be disabled anyway. Furthermore, it might be the case that the hash isn't used or queried at all until after wireguard transmits the encrypted UDP packet, which means skb->hash might still be zero at this point, and thus no hash taken over the inner packet data. In order to address this situation, we force a calculation of skb->hash before encrypting packet data. Of course this means that fq_codel might transmit packets slightly more out of order than usual. Toke did some testing on beefy machines with high quantities of parallel flows and found that increasing the reply-attack counter to 8192 takes care of the most pathological cases pretty well. Reported-by: Dave Taht Reviewed-and-tested-by: Toke Høiland-Jørgensen Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld Signed-off-by: David S. Miller

wireguard: send/receive: use explicit unlikely branch instead of implicit coalescing

2020-05-07T03:03:47+00:00

It's very unlikely that send will become true. It's nearly always false between 0 and 120 seconds of a session, and in most cases becomes true only between 120 and 121 seconds before becoming false again. So, unlikely(send) is clearly the right option here. What happened before was that we had this complex boolean expression with multiple likely and unlikely clauses nested. Since this is evaluated left-to-right anyway, the whole thing got converted to unlikely. So, we can clean this up to better represent what's going on. The generated code is the same. Suggested-by: Sultan Alsawaf Signed-off-by: Jason A. Donenfeld Signed-off-by: David S. Miller

wireguard: send/receive: cond_resched() when processing worker ringbuffers

2020-05-07T03:03:47+00:00

Users with pathological hardware reported CPU stalls on CONFIG_ PREEMPT_VOLUNTARY=y, because the ringbuffers would stay full, meaning these workers would never terminate. That turned out not to be okay on systems without forced preemption, which Sultan observed. This commit adds a cond_resched() to the bottom of each loop iteration, so that these workers don't hog the core. Note that we don't need this on the napi poll worker, since that terminates after its budget is expended. Suggested-by: Sultan Alsawaf Reported-by: Wang Jian Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld Signed-off-by: David S. Miller

wireguard: send: remove errant newline from packet_encrypt_worker

2020-04-29T21:23:05+00:00

This commit removes a useless newline at the end of a scope, which doesn't add anything in the way of organization or readability. Signed-off-by: Sultan Alsawaf Signed-off-by: Jason A. Donenfeld Signed-off-by: David S. Miller

wireguard: send: account for mtu=0 devices

2020-02-17T03:21:56+00:00

It turns out there's an easy way to get packets queued up while still having an MTU of zero, and that's via persistent keep alive. This commit makes sure that in whatever condition, we don't wind up dividing by zero. Note that an MTU of zero for a wireguard interface is something quasi-valid, so I don't think the correct fix is to limit it via min_mtu. This can be reproduced easily with: ip link add wg0 type wireguard ip link add wg1 type wireguard ip link set wg0 up mtu 0 ip link set wg1 up wg set wg0 private-key <(wg genkey) wg set wg1 listen-port 1 private-key <(wg genkey) peer $(wg show wg0 public-key) wg set wg0 peer $(wg show wg1 public-key) persistent-keepalive 1 endpoint 127.0.0.1:1 However, while min_mtu=0 seems fine, it makes sense to restrict the max_mtu. This commit also restricts the maximum MTU to the greatest number for which rounding up to the padding multiple won't overflow a signed integer. Packets this large were always rejected anyway eventually, due to checks deeper in, but it seems more sound not to even let the administrator configure something that won't work anyway. We use this opportunity to clean up this function a bit so that it's clear which paths we're expecting. Signed-off-by: Jason A. Donenfeld Cc: Eric Dumazet Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller