kernel/linux.git/drivers/net/wireguard, branch linux-5.11.y

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

2021-03-04T11:15:44+00:00

commit 8b5553ace83cced775eefd0f3f18b5c6214ccf7a upstream. 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 Signed-off-by: Greg Kroah-Hartman

wireguard: device: do not generate ICMP for non-IP packets

2021-03-04T11:15:13+00:00

[ Upstream commit 99fff5264e7ab06f45b0ad60243475be0a8d0559 ] If skb->protocol doesn't match the actual skb->data header, it's probably not a good idea to pass it off to icmp{,v6}_ndo_send, which is expecting to reply to a valid IP packet. So this commit has that early mismatch case jump to a later error label. Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin

Merge tag 'selinux-pr-20201214' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux

2020-12-16T19:01:04+00:00

Pull selinux updates from Paul Moore: "While we have a small number of SELinux patches for v5.11, there are a few changes worth highlighting: - Change the LSM network hooks to pass flowi_common structs instead of the parent flowi struct as the LSMs do not currently need the full flowi struct and they do not have enough information to use it safely (missing information on the address family). This patch was discussed both with Herbert Xu (representing team netdev) and James Morris (representing team LSMs-other-than-SELinux). - Fix how we handle errors in inode_doinit_with_dentry() so that we attempt to properly label the inode on following lookups instead of continuing to treat it as unlabeled. - Tweak the kernel logic around allowx, auditallowx, and dontauditx SELinux policy statements such that the auditx/dontauditx are effective even without the allowx statement. Everything passes our test suite" * tag 'selinux-pr-20201214' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux: lsm,selinux: pass flowi_common instead of flowi to the LSM hooks selinux: Fix fall-through warnings for Clang selinux: drop super_block backpointer from superblock_security_struct selinux: fix inode_doinit_with_dentry() LABEL_INVALID error handling selinux: allow dontauditx and auditallowx rules to take effect without allowx selinux: fix error initialization in inode_doinit_with_dentry()

lsm,selinux: pass flowi_common instead of flowi to the LSM hooks

2020-11-23T23:36:21+00:00

As pointed out by Herbert in a recent related patch, the LSM hooks do not have the necessary address family information to use the flowi struct safely. As none of the LSMs currently use any of the protocol specific flowi information, replace the flowi pointers with pointers to the address family independent flowi_common struct. Reported-by: Herbert Xu Acked-by: James Morris Signed-off-by: Paul Moore

wireguard: switch to dev_get_tstats64

2020-11-10T01:50:28+00:00

Replace ip_tunnel_get_stats64() with the new identical core function dev_get_tstats64(). Reviewed-by: Jason A. Donenfeld Signed-off-by: Heiner Kallweit Signed-off-by: Jakub Kicinski

Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2020-09-22T23:45:34+00:00

Two minor conflicts: 1) net/ipv4/route.c, adding a new local variable while moving another local variable and removing it's initial assignment. 2) drivers/net/dsa/microchip/ksz9477.c, overlapping changes. One pretty prints the port mode differently, whilst another changes the driver to try and obtain the port mode from the port node rather than the switch node. Signed-off-by: David S. Miller

wireguard: peerlookup: take lock before checking hash in replace operation

2020-09-09T18:31:38+00:00

Eric's suggested fix for the previous commit's mentioned race condition was to simply take the table->lock in wg_index_hashtable_replace(). The table->lock of the hash table is supposed to protect the bucket heads, not the entires, but actually, since all the mutator functions are already taking it, it makes sense to take it too for the test to hlist_unhashed, as a defense in depth measure, so that it no longer races with deletions, regardless of what other locks are protecting individual entries. This is sensible from a performance perspective because, as Eric pointed out, the case of being unhashed is already the unlikely case, so this won't add common contention. And comparing instructions, this basically doesn't make much of a difference other than pushing and popping %r13, used by the new `bool ret`. More generally, I like the idea of locking consistency across table mutator functions, and this might let me rest slightly easier at night. Suggested-by: Eric Dumazet Link: https://lore.kernel.org/wireguard/20200908145911.4090480-1-edumazet@google.com/ Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld Signed-off-by: David S. Miller

wireguard: noise: take lock when removing handshake entry from table

2020-09-09T18:31:37+00:00

Eric reported that syzkaller found a race of this variety: CPU 1 CPU 2 -------------------------------------------|--------------------------------------- wg_index_hashtable_replace(old, ...) | if (hlist_unhashed(&old->index_hash)) | | wg_index_hashtable_remove(old) | hlist_del_init_rcu(&old->index_hash) | old->index_hash.pprev = NULL hlist_replace_rcu(&old->index_hash, ...) | *old->index_hash.pprev | Syzbot wasn't actually able to reproduce this more than once or create a reproducer, because the race window between checking "hlist_unhashed" and calling "hlist_replace_rcu" is just so small. Adding an mdelay(5) or similar there helps make this demonstrable using this simple script: #!/bin/bash set -ex trap 'kill $pid1; kill $pid2; ip link del wg0; ip link del wg1' EXIT ip link add wg0 type wireguard ip link add wg1 type wireguard wg set wg0 private-key <(wg genkey) listen-port 9999 wg set wg1 private-key <(wg genkey) peer $(wg show wg0 public-key) endpoint 127.0.0.1:9999 persistent-keepalive 1 wg set wg0 peer $(wg show wg1 public-key) ip link set wg0 up yes link set wg1 up | ip -force -batch - & pid1=$! yes link set wg1 down | ip -force -batch - & pid2=$! wait The fundumental underlying problem is that we permit calls to wg_index_ hashtable_remove(handshake.entry) without requiring the caller to take the handshake mutex that is intended to protect members of handshake during mutations. This is consistently the case with calls to wg_index_ hashtable_insert(handshake.entry) and wg_index_hashtable_replace( handshake.entry), but it's missing from a pertinent callsite of wg_ index_hashtable_remove(handshake.entry). So, this patch makes sure that mutex is taken. The original code was a little bit funky though, in the form of: remove(handshake.entry) lock(), memzero(handshake.some_members), unlock() remove(handshake.entry) The original intention of that double removal pattern outside the lock appears to be some attempt to prevent insertions that might happen while locks are dropped during expensive crypto operations, but actually, all callers of wg_index_hashtable_insert(handshake.entry) take the write lock and then explicitly check handshake.state, as they should, which the aforementioned memzero clears, which means an insertion should already be impossible. And regardless, the original intention was necessarily racy, since it wasn't guaranteed that something else would run after the unlock() instead of after the remove(). So, from a soundness perspective, it seems positive to remove what looks like a hack at best. The crash from both syzbot and from the script above is as follows: general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] CPU: 0 PID: 7395 Comm: kworker/0:3 Not tainted 5.9.0-rc4-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Workqueue: wg-kex-wg1 wg_packet_handshake_receive_worker RIP: 0010:hlist_replace_rcu include/linux/rculist.h:505 [inline] RIP: 0010:wg_index_hashtable_replace+0x176/0x330 drivers/net/wireguard/peerlookup.c:174 Code: 00 fc ff df 48 89 f9 48 c1 e9 03 80 3c 01 00 0f 85 44 01 00 00 48 b9 00 00 00 00 00 fc ff df 48 8b 45 10 48 89 c6 48 c1 ee 03 <80> 3c 0e 00 0f 85 06 01 00 00 48 85 d2 4c 89 28 74 47 e8 a3 4f b5 RSP: 0018:ffffc90006a97bf8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff888050ffc4f8 RCX: dffffc0000000000 RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff88808e04e010 RBP: ffff88808e04e000 R08: 0000000000000001 R09: ffff8880543d0000 R10: ffffed100a87a000 R11: 000000000000016e R12: ffff8880543d0000 R13: ffff88808e04e008 R14: ffff888050ffc508 R15: ffff888050ffc500 FS: 0000000000000000(0000) GS:ffff8880ae600000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000f5505db0 CR3: 0000000097cf7000 CR4: 00000000001526f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: wg_noise_handshake_begin_session+0x752/0xc9a drivers/net/wireguard/noise.c:820 wg_receive_handshake_packet drivers/net/wireguard/receive.c:183 [inline] wg_packet_handshake_receive_worker+0x33b/0x730 drivers/net/wireguard/receive.c:220 process_one_work+0x94c/0x1670 kernel/workqueue.c:2269 worker_thread+0x64c/0x1120 kernel/workqueue.c:2415 kthread+0x3b5/0x4a0 kernel/kthread.c:292 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294 Reported-by: syzbot Reported-by: Eric Dumazet Link: https://lore.kernel.org/wireguard/20200908145911.4090480-1-edumazet@google.com/ Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld Signed-off-by: David S. Miller

netlink: consistently use NLA_POLICY_MIN_LEN()

2020-08-18T19:28:45+00:00

Change places that open-code NLA_POLICY_MIN_LEN() to use the macro instead, giving us flexibility in how we handle the details of the macro. Signed-off-by: Johannes Berg Signed-off-by: David S. Miller

netlink: consistently use NLA_POLICY_EXACT_LEN()

2020-08-18T19:28:45+00:00

Change places that open-code NLA_POLICY_EXACT_LEN() to use the macro instead, giving us flexibility in how we handle the details of the macro. Signed-off-by: Johannes Berg Acked-by: Matthieu Baerts Signed-off-by: David S. Miller