| Age | Commit message (Collapse) | Author |
|---|
|
Use 2-factor argument form kvcalloc() instead of kvzalloc().
Link: https://github.com/KSPP/linux/issues/162
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
[Jason: Gustavo's link above is for KSPP, but this isn't actually a
security fix, as table_size is bounded to 8192 anyway, and gcc realizes
this, so the codegen comes out to be about the same.]
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
If we're being delivered packets from multiple CPUs so quickly that the
ring lock is contended for CPU tries, then it's safe to assume that the
queue is near capacity anyway, so just drop the packet rather than
spinning. This helps deal with multicore DoS that can interfere with
data path performance. It _still_ does not completely fix the issue, but
it again chips away at it.
Reported-by: Streun Fabio <fstreun@student.ethz.ch>
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Apparently the spinlock on incoming_handshake's skb_queue is highly
contended, and a torrent of handshake or cookie packets can bring the
data plane to its knees, simply by virtue of enqueueing the handshake
packets to be processed asynchronously. So, we try switching this to a
ring buffer to hopefully have less lock contention. This alleviates the
problem somewhat, though it still isn't perfect, so future patches will
have to improve this further. However, it at least doesn't completely
diminish the data plane.
Reported-by: Streun Fabio <fstreun@student.ethz.ch>
Reported-by: Joel Wanner <joel.wanner@inf.ethz.ch>
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Each peer's endpoint contains a dst_cache entry that takes a reference
to another netdev. When the containing namespace exits, we take down the
socket and prevent future sockets from being created (by setting
creating_net to NULL), which removes that potential reference on the
netns. However, it doesn't release references to the netns that a netdev
cached in dst_cache might be taking, so the netns still might fail to
exit. Since the socket is gimped anyway, we can simply clear all the
dst_caches (by way of clearing the endpoint src), which will release all
references.
However, the current dst_cache_reset function only releases those
references lazily. But it turns out that all of our usages of
wg_socket_clear_peer_endpoint_src are called from contexts that are not
exactly high-speed or bottle-necked. For example, when there's
connection difficulty, or when userspace is reconfiguring the interface.
And in particular for this patch, when the netns is exiting. So for
those cases, it makes more sense to call dst_release immediately. For
that, we add a small helper function to dst_cache.
This patch also adds a test to netns.sh from Hangbin Liu to ensure this
doesn't regress.
Tested-by: Hangbin Liu <liuhangbin@gmail.com>
Reported-by: Xiumei Mu <xmu@redhat.com>
Cc: Toke Høiland-Jørgensen <toke@redhat.com>
Cc: Paolo Abeni <pabeni@redhat.com>
Fixes: 900575aa33a3 ("wireguard: device: avoid circular netns references")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Rename module_init & module_exit functions that are named
"mod_init" and "mod_exit" so that they are unique in both the
System.map file and in initcall_debug output instead of showing
up as almost anonymous "mod_init".
This is helpful for debugging and in determining how long certain
module_init calls take to execute.
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
A __rcu annotation got lost during refactoring, which caused sparse to
become enraged.
Fixes: bf7b042dc62a ("wireguard: allowedips: free empty intermediate nodes when removing single node")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
When removing single nodes, it's possible that that node's parent is an
empty intermediate node, in which case, it too should be removed.
Otherwise the trie fills up and never is fully emptied, leading to
gradual memory leaks over time for tries that are modified often. There
was originally code to do this, but was removed during refactoring in
2016 and never reworked. Now that we have proper parent pointers from
the previous commits, we can implement this properly.
In order to reduce branching and expensive comparisons, we want to keep
the double pointer for parent assignment (which lets us easily chain up
to the root), but we still need to actually get the parent's base
address. So encode the bit number into the last two bits of the pointer,
and pack and unpack it as needed. This is a little bit clumsy but is the
fastest and less memory wasteful of the compromises. Note that we align
the root struct here to a minimum of 4, because it's embedded into a
larger struct, and we're relying on having the bottom two bits for our
flag, which would only be 16-bit aligned on m68k.
The existing macro-based helpers were a bit unwieldy for adding the bit
packing to, so this commit replaces them with safer and clearer ordinary
functions.
We add a test to the randomized/fuzzer part of the selftests, to free
the randomized tries by-peer, refuzz it, and repeat, until it's supposed
to be empty, and then then see if that actually resulted in the whole
thing being emptied. That combined with kmemcheck should hopefully make
sure this commit is doing what it should. Along the way this resulted in
various other cleanups of the tests and fixes for recent graphviz.
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Cc: stable@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The previous commit moved from O(n) to O(1) for removal, but in the
process introduced an additional pointer member to a struct that
increased the size from 60 to 68 bytes, putting nodes in the 128-byte
slab. With deployed systems having as many as 2 million nodes, this
represents a significant doubling in memory usage (128 MiB -> 256 MiB).
Fix this by using our own kmem_cache, that's sized exactly right. This
also makes wireguard's memory usage more transparent in tools like
slabtop and /proc/slabinfo.
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Suggested-by: Matthew Wilcox <willy@infradead.org>
Cc: stable@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Previously, deleting peers would require traversing the entire trie in
order to rebalance nodes and safely free them. This meant that removing
1000 peers from a trie with a half million nodes would take an extremely
long time, during which we're holding the rtnl lock. Large-scale users
were reporting 200ms latencies added to the networking stack as a whole
every time their userspace software would queue up significant removals.
That's a serious situation.
This commit fixes that by maintaining a double pointer to the parent's
bit pointer for each node, and then using the already existing node list
belonging to each peer to go directly to the node, fix up its pointers,
and free it with RCU. This means removal is O(1) instead of O(n), and we
don't use gobs of stack.
The removal algorithm has the same downside as the code that it fixes:
it won't collapse needlessly long runs of fillers. We can enhance that
in the future if it ever becomes a problem. This commit documents that
limitation with a TODO comment in code, a small but meaningful
improvement over the prior situation.
Currently the biggest flaw, which the next commit addresses, is that
because this increases the node size on 64-bit machines from 60 bytes to
68 bytes. 60 rounds up to 64, but 68 rounds up to 128. So we wind up
using twice as much memory per node, because of power-of-two
allocations, which is a big bummer. We'll need to figure something out
there.
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Cc: stable@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The randomized trie tests weren't initializing the dummy peer list head,
resulting in a NULL pointer dereference when used. Fix this by
initializing it in the randomized trie test, just like we do for the
static unit test.
While we're at it, all of the other strings like this have the word
"self-test", so add it to the missing place here.
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Cc: stable@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
With deployments having upwards of 600k peers now, this somewhat heavy
structure could benefit from more fine-grained allocations.
Specifically, instead of using a 2048-byte slab for a 1544-byte object,
we can now use 1544-byte objects directly, thus saving almost 25%
per-peer, or with 600k peers, that's a savings of 303 MiB. This also
makes wireguard's memory usage more transparent in tools like slabtop
and /proc/slabinfo.
Fixes: 8b5553ace83c ("wireguard: queueing: get rid of per-peer ring buffers")
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Suggested-by: Matthew Wilcox <willy@infradead.org>
Cc: stable@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Many of the synchronization points are sometimes called under the rtnl
lock, which means we should use synchronize_net rather than
synchronize_rcu. Under the hood, this expands to using the expedited
flavor of function in the event that rtnl is held, in order to not stall
other concurrent changes.
This fixes some very, very long delays when removing multiple peers at
once, which would cause some operations to take several minutes.
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Cc: stable@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Apparently, various versions of gcc have O3-related miscompiles. Looking
at the difference between -O2 and -O3 for gcc 11 doesn't indicate
miscompiles, but the difference also doesn't seem so significant for
performance that it's worth risking.
Link: https://lore.kernel.org/lkml/CAHk-=wjuoGyxDhAF8SsrTkN0-YfCx7E6jUN3ikC_tn2AKWTTsA@mail.gmail.com/
Link: https://lore.kernel.org/lkml/CAHmME9otB5Wwxp7H8bR_i2uH2esEMvoBMC8uEXBMH9p0q1s6Bw@mail.gmail.com/
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Cc: stable@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
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 <dvyukov@google.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
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 <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The is_dead boolean is checked for every single packet, while the
internal_id member is used basically only for pr_debug messages. So it
makes sense to hoist up is_dead into some space formerly unused by a
struct hole, while demoting internal_api to below the lowest struct
cache line.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The endpoint->src_if4 has nothing to do with fixed-endian numbers; remove
the bogus annotation.
This was introduced in
https://git.zx2c4.com/wireguard-monolithic-historical/commit?id=14e7d0a499a676ec55176c0de2f9fcbd34074a82
in the historical WireGuard repo because the old code used to
zero-initialize multiple members as follows:
endpoint->src4.s_addr = endpoint->src_if4 = fl.saddr = 0;
Because fl.saddr is fixed-endian and an assignment returns a value with the
type of its left operand, this meant that sparse detected an assignment
between values of different endianness.
Since then, this assignment was already split up into separate statements;
just the cast survived.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The definition of IS_ERR() already applies the unlikely() notation
when checking the error status of the passed pointer. For this
reason there is no need to have the same notation outside of
IS_ERR() itself.
Clean up code by removing redundant notation.
Signed-off-by: Antonio Quartulli <a@unstable.cc>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux
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()
|
|
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 <herbert@gondor.apana.org.au>
Acked-by: James Morris <jamorris@linux.microsoft.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
|
|
Replace ip_tunnel_get_stats64() with the new identical core function
dev_get_tstats64().
Reviewed-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
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 <davem@davemloft.net>
|
|
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 <edumazet@google.com>
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 <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
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 <syzkaller@googlegroups.com>
Reported-by: Eric Dumazet <edumazet@google.com>
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 <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
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 <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
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 <johannes.berg@intel.com>
Acked-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
As said by Linus:
A symmetric naming is only helpful if it implies symmetries in use.
Otherwise it's actively misleading.
In "kzalloc()", the z is meaningful and an important part of what the
caller wants.
In "kzfree()", the z is actively detrimental, because maybe in the
future we really _might_ want to use that "memfill(0xdeadbeef)" or
something. The "zero" part of the interface isn't even _relevant_.
The main reason that kzfree() exists is to clear sensitive information
that should not be leaked to other future users of the same memory
objects.
Rename kzfree() to kfree_sensitive() to follow the example of the recently
added kvfree_sensitive() and make the intention of the API more explicit.
In addition, memzero_explicit() is used to clear the memory to make sure
that it won't get optimized away by the compiler.
The renaming is done by using the command sequence:
git grep -w --name-only kzfree |\
xargs sed -i 's/kzfree/kfree_sensitive/'
followed by some editing of the kfree_sensitive() kerneldoc and adding
a kzfree backward compatibility macro in slab.h.
[akpm@linux-foundation.org: fs/crypto/inline_crypt.c needs linux/slab.h]
[akpm@linux-foundation.org: fix fs/crypto/inline_crypt.c some more]
Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: David Howells <dhowells@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Cc: James Morris <jmorris@namei.org>
Cc: "Serge E. Hallyn" <serge@hallyn.com>
Cc: Joe Perches <joe@perches.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Cc: "Jason A . Donenfeld" <Jason@zx2c4.com>
Link: http://lkml.kernel.org/r/20200616154311.12314-3-longman@redhat.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Now that wg_examine_packet_protocol has been added for general
consumption as ip_tunnel_parse_protocol, it's possible to remove
wg_examine_packet_protocol and simply use the new
ip_tunnel_parse_protocol function directly.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
WireGuard uses skb->protocol to determine packet type, and bails out if
it's not set or set to something it's not expecting. For AF_PACKET
injection, we need to support its call chain of:
packet_sendmsg -> packet_snd -> packet_parse_headers ->
dev_parse_header_protocol -> parse_protocol
Without a valid parse_protocol, this returns zero, and wireguard then
rejects the skb. So, this wires up the ip_tunnel handler for layer 3
packets for that case.
Reported-by: Hans Wippel <ndev@hwipl.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The napi_gro_receive function no longer returns GRO_DROP ever, making
handling GRO_DROP dead code. This commit removes that dead code.
Further, it's not even clear that device drivers have any business in
taking action after passing off received packets; that's arguably out of
their hands.
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Fixes: 6570bc79c0df ("net: core: use listified Rx for GRO_NORMAL in napi_gro_receive()")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Before, we took a reference to the creating netns if the new netns was
different. This caused issues with circular references, with two
wireguard interfaces swapping namespaces. The solution is to rather not
take any extra references at all, but instead simply invalidate the
creating netns pointer when that netns is deleted.
In order to prevent this from happening again, this commit improves the
rough object leak tracking by allowing it to account for created and
destroyed interfaces, aside from just peers and keys. That then makes it
possible to check for the object leak when having two interfaces take a
reference to each others' namespaces.
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Fixes an error condition reported by checkpatch.pl which caused by
assigning a variable in an if condition in wg_noise_handshake_consume_
initiation().
Signed-off-by: Frank Werner-Krippendorf <mail@hb9fxq.ch>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
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 <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
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 <dave.taht@gmail.com>
Reviewed-and-tested-by: Toke Høiland-Jørgensen <toke@toke.dk>
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Prior we read the preshared key after dropping the handshake lock, which
isn't an actual crypto issue if it races, but it's still not quite
correct. So copy that part of the state into a temporary like we do with
the rest of the handshake state variables. Then we can release the lock,
operate on the temporary, and zero it out at the end of the function. In
performance tests, the impact of this was entirely unnoticable, probably
because those bytes are coming from the same cacheline as other things
that are being copied out in the same manner.
Reported-by: Matt Dunwoodie <ncon@noconroy.net>
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
coalescing
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 <sultan@kerneltoast.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Without setting these to NULL, clang complains in certain
configurations that have CONFIG_IPV6=n:
In file included from drivers/net/wireguard/ratelimiter.c:223:
drivers/net/wireguard/selftest/ratelimiter.c:173:34: error: variable 'skb6' is uninitialized when used here [-Werror,-Wuninitialized]
ret = timings_test(skb4, hdr4, skb6, hdr6, &test_count);
^~~~
drivers/net/wireguard/selftest/ratelimiter.c:123:29: note: initialize the variable 'skb6' to silence this warning
struct sk_buff *skb4, *skb6;
^
= NULL
drivers/net/wireguard/selftest/ratelimiter.c:173:40: error: variable 'hdr6' is uninitialized when used here [-Werror,-Wuninitialized]
ret = timings_test(skb4, hdr4, skb6, hdr6, &test_count);
^~~~
drivers/net/wireguard/selftest/ratelimiter.c:125:22: note: initialize the variable 'hdr6' to silence this warning
struct ipv6hdr *hdr6;
^
We silence this warning by setting the variables to NULL as the warning
suggests.
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
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 <sultan@kerneltoast.com>
Reported-by: Wang Jian <larkwang@gmail.com>
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
It's already possible to create two different interfaces and loop
packets between them. This has always been possible with tunnels in the
kernel, and isn't specific to wireguard. Therefore, the networking stack
already needs to deal with that. At the very least, the packet winds up
exceeding the MTU and is discarded at that point. So, since this is
already something that happens, there's no need to forbid the not very
exceptional case of routing a packet back to the same interface; this
loop is no different than others, and we shouldn't special case it, but
rather rely on generic handling of loops in general. This also makes it
easier to do interesting things with wireguard such as onion routing.
At the same time, we add a selftest for this, ensuring that both onion
routing works and infinite routing loops do not crash the kernel. We
also add a test case for wireguard interfaces nesting packets and
sending traffic between each other, as well as the loop in this case
too. We make sure to send some throughput-heavy traffic for this use
case, to stress out any possible recursion issues with the locks around
workqueues.
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
WireGuard currently only propagates ECN markings on tunnel decap according
to the old RFC3168 specification. However, the spec has since been updated
in RFC6040 to recommend slightly different decapsulation semantics. This
was implemented in the kernel as a set of common helpers for ECN
decapsulation, so let's just switch over WireGuard to using those, so it
can benefit from this enhancement and any future tweaks. We do not drop
packets with invalid ECN marking combinations, because WireGuard is
frequently used to work around broken ISPs, which could be doing that.
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Reported-by: Olivier Tilmans <olivier.tilmans@nokia-bell-labs.com>
Cc: Dave Taht <dave.taht@gmail.com>
Cc: Rodney W. Grimes <ietf@gndrsh.dnsmgr.net>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Prior, if the alloc_percpu of packet_percpu_multicore_worker_alloc
failed, the previously allocated ptr_ring wouldn't be freed. This commit
adds the missing call to ptr_ring_cleanup in the error case.
Reported-by: Sultan Alsawaf <sultan@kerneltoast.com>
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
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 <sultan@kerneltoast.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
net/netfilter/nft_fwd_netdev.c: In function ‘nft_fwd_netdev_eval’:
net/netfilter/nft_fwd_netdev.c:32:10: error: ‘struct sk_buff’ has no member named ‘tc_redirected’
pkt->skb->tc_redirected = 1;
^~
net/netfilter/nft_fwd_netdev.c:33:10: error: ‘struct sk_buff’ has no member named ‘tc_from_ingress’
pkt->skb->tc_from_ingress = 1;
^~
To avoid a direct dependency with tc actions from netfilter, wrap the
redirect bits around CONFIG_NET_REDIRECT and move helpers to
include/linux/skbuff.h. Turn on this toggle from the ifb driver, the
only existing client of these bits in the tree.
This patch adds skb_set_redirected() that sets on the redirected bit
on the skbuff, it specifies if the packet was redirect from ingress
and resets the timestamp (timestamp reset was originally missing in the
netfilter bugfix).
Fixes: bcfabee1afd99484 ("netfilter: nft_fwd_netdev: allow to redirect to ifb via ingress")
Reported-by: noreply@ellerman.id.au
Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
We precompute the static-static ECDH during configuration time, in order
to save an expensive computation later when receiving network packets.
However, not all ECDH computations yield a contributory result. Prior,
we were just not letting those peers be added to the interface. However,
this creates a strange inconsistency, since it was still possible to add
other weird points, like a valid public key plus a low-order point, and,
like points that result in zeros, a handshake would not complete. In
order to make the behavior more uniform and less surprising, simply
allow all peers to be added. Then, we'll error out later when doing the
crypto if there's an issue. This also adds more separation between the
crypto layer and the configuration layer.
Discussed-with: Mathias Hall-Andersen <mathias@hall-andersen.dk>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The situation in which we wind up hitting the default case here
indicates a major bug in earlier parsing code. It is not a usual thing
that should ever happen, which means a "friendly" message for it doesn't
make sense. Rather, replace this with a WARN_ON, just like we do earlier
in the file for a similar situation, so that somebody sends us a bug
report and we can fix it.
Reported-by: Fabian Freyer <fabianfreyer@radicallyopensecurity.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
We carry out checks to the effect of:
if (skb->protocol != wg_examine_packet_protocol(skb))
goto err;
By having wg_skb_examine_untrusted_ip_hdr return 0 on failure, this
means that the check above still passes in the case where skb->protocol
is zero, which is possible to hit with AF_PACKET:
struct sockaddr_pkt saddr = { .spkt_device = "wg0" };
unsigned char buffer[5] = { 0 };
sendto(socket(AF_PACKET, SOCK_PACKET, /* skb->protocol = */ 0),
buffer, sizeof(buffer), 0, (const struct sockaddr *)&saddr, sizeof(saddr));
Additional checks mean that this isn't actually a problem in the code
base, but I could imagine it becoming a problem later if the function is
used more liberally.
I would prefer to fix this by having wg_examine_packet_protocol return a
32-bit ~0 value on failure, which will never match any value of
skb->protocol, which would simply change the generated code from a mov
to a movzx. However, sparse complains, and adding __force casts doesn't
seem like a good idea, so instead we just add a simple helper function
to check for the zero return value. Since wg_examine_packet_protocol
itself gets inlined, this winds up not adding an additional branch to
the generated code, since the 0 return value already happens in a
mergable branch.
Reported-by: Fabian Freyer <fabianfreyer@radicallyopensecurity.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
synchronize_net() is a wrapper around synchronize_rcu(), so there's no
point in having synchronize_net and synchronize_rcu back to back,
despite the documentation comment suggesting maybe it's somewhat useful,
"Wait for packets currently being received to be done." This commit
removes the extra call.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Suggested-by: Eric Dumazet <eric.dumazet@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
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 <Jason@zx2c4.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This is a small optimization that prevents more expensive comparisons
from happening when they are no longer necessary, by clearing the
last_under_load variable whenever we wind up in a state where we were
under load but we no longer are.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Suggested-by: Matt Dunwoodie <ncon@noconroy.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Because wireguard is calling icmp from network device context, it should
use the ndo helper so that the rate limiting applies correctly. This
commit adds a small test to the wireguard test suite to ensure that the
new functions continue doing the right thing in the context of
wireguard. It does this by setting up a condition that will definately
evoke an icmp error message from the driver, but along a nat'd path.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
