Commit graph

934 commits

Author SHA1 Message Date
snow flurry 5657c3c1e8 tun_netbsd: update to follow tun_openbsd changes
Most of tun_netbsd is from tun_openbsd with some small changes, so we
should follow the changes made in tun_openbsd.
2022-02-18 19:15:16 -08:00
snow flurry ff49dc80a3 tun_netbsd: support unpatched kernels for IFHEAD
This previously assumed a user's kernel was patched a la
http://gnats.netbsd.org/cgi-bin/query-pr-single.pl?number=55463, making
it effectively useless for most people. Since we know IFHEAD will always
be one of 0 (false) or something that isn't 0 (true), we can generalize
things a bit more.

Suggestion provided by artemis
https://artemis.sh/2022/02/16/tailscale-on-netbsd-proof-of-concept.html
2022-02-18 19:14:23 -08:00
snow c644c61a8e Add support for NetBSD tun(4) interface 2022-02-18 18:40:27 -08:00
Jason A. Donenfeld 3b95c81cc1 tun/netstack: simplify read timeout on ping socket
I'm not 100% sure this is correct, but it certainly is a lot simpler.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2022-02-02 23:30:31 +01:00
Thomas H. Ptacek b9669b734e tun/netstack: implement ICMP ping
Provide a PacketConn interface for netstack's ICMP endpoint; netstack
currently only provides EchoRequest/EchoResponse ICMP support, so this
code exposes only an interface for doing ping.

Signed-off-by: Thomas Ptacek <thomas@sockpuppet.org>
[Jason: rework structure, match std go interfaces, add example code]
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2022-02-02 23:09:37 +01:00
Jason A. Donenfeld e0b8f11489 version: bump snapshot
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2022-01-17 17:37:42 +01:00
Jason A. Donenfeld 114a3db918 ipc: bsd: try again if kqueue returns EINTR
Reported-by: J. Michael McAtee <mmcatee@jumptrading.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2022-01-14 16:10:43 +01:00
Jason A. Donenfeld 9c9e7e2724 global: apply gofumpt
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-12-09 23:15:55 +01:00
Jason A. Donenfeld 2dd424e2d8 device: handle peer post config on blank line
We missed a function exit point. This was exacerbated by e3134bf
("device: defer state machine transitions until configuration is
complete"), but the bug existed prior. Minus provided the following
useful reproducer script:

    #!/usr/bin/env bash

    set -eux

    make wireguard-go || exit 125

    ip netns del test-ns || true
    ip netns add test-ns
    ip link add test-kernel type wireguard
    wg set test-kernel listen-port 0 private-key <(echo "QMCfZcp1KU27kEkpcMCgASEjDnDZDYsfMLHPed7+538=") peer "eDPZJMdfnb8ZcA/VSUnLZvLB2k8HVH12ufCGa7Z7rHI=" allowed-ips 10.51.234.10/32
    ip link set test-kernel netns test-ns up
    ip -n test-ns addr add 10.51.234.1/24 dev test-kernel
    port=$(ip netns exec test-ns wg show test-kernel listen-port)

    ip link del test-go || true
    ./wireguard-go test-go
    wg set test-go private-key <(echo "WBM7qimR3vFk1QtWNfH+F4ggy/hmO+5hfIHKxxI4nF4=") peer "+nj9Dkqpl4phsHo2dQliGm5aEiWJJgBtYKbh7XjeNjg=" allowed-ips 0.0.0.0/0 endpoint 127.0.0.1:$port
    ip addr add 10.51.234.10/24 dev test-go
    ip link set test-go up

    ping -c2 -W1 10.51.234.1

Reported-by: minus <minus@mnus.de>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-29 12:31:54 -05:00
Josh Bleecher Snyder 387f7c461a device: reduce peer lock critical section in UAPI
The deferred RUnlock calls weren't executing until all peers
had been processed. Add an anonymous function so that each
peer may be unlocked as soon as it is completed.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-23 22:03:15 +01:00
Josh Bleecher Snyder 4d87c9e824 device: remove code using unsafe
There is no performance impact.

name                             old time/op  new time/op  delta
TrieIPv4Peers100Addresses1000-8  78.6ns ± 1%  79.4ns ± 3%    ~     (p=0.604 n=10+9)
TrieIPv4Peers10Addresses10-8     29.1ns ± 2%  28.8ns ± 1%  -1.12%  (p=0.014 n=10+9)
TrieIPv6Peers100Addresses1000-8  78.9ns ± 1%  78.6ns ± 1%    ~     (p=0.492 n=10+10)
TrieIPv6Peers10Addresses10-8     29.3ns ± 2%  28.6ns ± 2%  -2.16%  (p=0.000 n=10+10)

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-23 22:03:15 +01:00
Jason A. Donenfeld ef8d6804d7 global: use netip where possible now
There are more places where we'll need to add it later, when Go 1.18
comes out with support for it in the "net" package. Also, allowedips
still uses slices internally, which might be suboptimal.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-23 22:03:15 +01:00
Jason A. Donenfeld de7c702ace device: only propagate roaming value before peer is referenced elsewhere
A peer.endpoint never becomes nil after being not-nil, so creation is
the only time we actually need to set this. This prevents a race from
when the variable is actually used elsewhere, and allows us to avoid an
expensive atomic.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-16 21:16:04 +01:00
Jason A. Donenfeld fc4f975a4d device: align 64-bit atomic member in Device
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-16 21:07:31 +01:00
Jason A. Donenfeld 9d699ba730 device: start peers before running handshake test
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-16 21:07:31 +01:00
Jason A. Donenfeld 425f7c726b Makefile: don't use test -v because it hides failures in scrollback
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-16 21:07:31 +01:00
David Anderson 3cae233d69 device: fix nil pointer dereference in uapi read
Signed-off-by: David Anderson <danderson@tailscale.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-16 20:43:26 +01:00
Jason A. Donenfeld 111e0566dc device: make new peers inherit broken mobile semantics
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-15 23:40:47 +01:00
Jason A. Donenfeld e3134bf665 device: defer state machine transitions until configuration is complete
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-15 23:40:47 +01:00
Jason A. Donenfeld 63abb5537b device: do not consume handshake messages if not running
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-15 23:40:47 +01:00
Jason A. Donenfeld 851efb1bb6 tun: move wintun to its own repo
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-04 12:53:55 +01:00
Jason A. Donenfeld c07dd60cdb namedpipe: rename from winpipe to keep in sync with CL299009
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-11-04 12:53:52 +01:00
Jason A. Donenfeld eb6302c7eb device: timers: use pre-seeded per-thread unlocked fastrandn for jitter
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-28 13:47:50 +02:00
Jason A. Donenfeld 60683d7361 device: timers: seed unsafe rng before use for jitter
Forgetting to seed the unsafe rng, the jitter before followed a fixed
pattern, which didn't help when a fleet of computers all boot at once.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-28 13:34:21 +02:00
Jason A. Donenfeld e42c6c4bc2 wintun: align 64-bit argument on ARM32
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-26 14:53:40 +02:00
Jason A. Donenfeld 828a885a71 README: raise minimum Go to 1.17
Suggested-by: Adam Bliss <abliss@gmail.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-25 17:53:11 +02:00
Mikael Magnusson f1f626090e tun/netstack: update gvisor
Update gvisor to v0.0.0-20211020211948-f76a604701b6, which requires some
changes to tun.go:

WriteRawPacket: Add function with not implemented error.

CreateNetTUN: Replace stack.AddAddress with stack.AddProtocolAddress, and
fix IPv6 address in error message.

Signed-off-by: Mikael Magnusson <mikma@users.sourceforge.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-22 13:22:29 -06:00
Brad Fitzpatrick 82e0b734e5 ipc, rwcancel: compile on js/wasm
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2021-10-20 14:50:05 -06:00
Jason A. Donenfeld fdf57a1fa4 wintun: allow retrieving DLL version
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-20 12:13:44 -06:00
Jason A. Donenfeld f87e87af0d version: bump snapshot
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-16 23:27:13 -06:00
Jason A. Donenfeld ba9e364dab wintun: remove memmod option for dll loading
Only wireguard-windows used this, and it's moving to wgnt exclusively.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-16 22:49:38 -06:00
Jason A. Donenfeld dfd688b6aa global: remove old-style build tags
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-12 12:02:10 -06:00
Jason A. Donenfeld c01d52b66a global: add newer-style build tags
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-12 11:46:53 -06:00
Jason A. Donenfeld 82d2aa87aa wintun: use new swdevice-based API for upcoming Wintun 0.14
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-12 00:26:46 -06:00
Jason A. Donenfeld 982d5d2e84 conn,wintun: use unsafe.Slice instead of unsafeSlice
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-11 14:57:53 -06:00
Jason A. Donenfeld 642a56e165 memmod: import from wireguard-windows
We'll eventually be getting rid of it here, but keep it sync'd up for
now.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-10-11 14:53:36 -06:00
Jason A. Donenfeld bb745b2ea3 rwcancel: use unix.Poll again but bump x/sys so it uses ppoll under the hood
This reverts commit fcc601dbf0 but then
bumps go.mod.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-09-27 14:19:15 -06:00
Jason A. Donenfeld fcc601dbf0 rwcancel: use ppoll on Linux for Android
This is a temporary measure while we wait for
https://go-review.googlesource.com/c/sys/+/352310 to land.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-09-26 17:16:38 -06:00
Tobias Klauser 217ac1016b tun: make operateonfd.go build tags more specific
(*NativeTun).operateOnFd is only used on darwin and freebsd. Adjust the
build tags accordingly.

Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-09-23 09:54:01 -06:00
Tobias Klauser eae5e0f3a3 tun: avoid leaking sock fd in CreateTUN error cases
At these points, the socket file descriptor is not yet wrapped in an
*os.File, so it needs to be closed explicitly on error.

Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-09-23 09:53:49 -06:00
Jason A. Donenfeld 2ef39d4754 global: add new go 1.17 build comments
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-09-05 16:00:43 +02:00
Jason A. Donenfeld 3957e9b9dd memmod: register exception handler tables
Otherwise recent WDK binaries fail on ARM64, where an exception handler
is used for trapping an illegal instruction when ARMv8.1 atomics are
being tested for functionality.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-08-05 14:56:48 +02:00
Jason A. Donenfeld bad6caeb82 memmod: fix protected delayed load the right way
The reason this was failing before is that dloadsup.h's
DloadObtainSection was doing a linear search of sections to find which
header corresponds with the IMAGE_DELAYLOAD_DESCRIPTOR section, and we
were stupidly overwriting the VirtualSize field, so the linear search
wound up matching the .text section, which then it found to not be
marked writable and failed with FAST_FAIL_DLOAD_PROTECTION_FAILURE.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-07-29 01:27:40 +02:00
Jason A. Donenfeld c89f5ca665 memmod: disable protected delayed load for now
Probably a bad idea, but we don't currently support it, and those huge
windows.NewCallback trampolines make juicer targets anyway.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-07-29 01:13:03 +02:00
Jason A. Donenfeld 15b24b6179 ipc: allow admins but require high integrity label
Might be more reasonable.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-06-24 17:01:02 +02:00
Jason A. Donenfeld f9b48a961c device: zero out allowedip node pointers when removing
This should make it a bit easier for the garbage collector.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-06-04 16:33:28 +02:00
Jason A. Donenfeld d0cf96114f device: limit allowedip fuzzer a to 4 times through
Trying this for every peer winds up being very slow and precludes it
from acceptable runtime in the CI, so reduce this to 4.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-06-03 18:22:50 +02:00
Jason A. Donenfeld 841756e328 device: simplify allowedips lookup signature
The inliner should handle this for us.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-06-03 16:29:43 +02:00
Jason A. Donenfeld c382222eab device: remove nodes by peer in O(1) instead of O(n)
Now that we have parent pointers hooked up, we can simply go right to
the node and remove it in place, rather than having to recursively walk
the entire trie.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-06-03 16:29:43 +02:00
Jason A. Donenfeld b41f4cc768 device: remove recursion from insertion and connect parent pointers
This makes the insertion algorithm a bit more efficient, while also now
taking on the additional task of connecting up parent pointers. This
will be handy in the following commit.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2021-06-03 15:08:42 +02:00